Northern and Central California Biogeographic Assessment - Marine Fishes, Birds and Mammals
US Dept of Commerce, NOAA, Center for Coastal Monitoring and Assessment - Biogeography Team
National Centers for Coastal Ocean Science
December 2003
Prepared for Cordell Bank, Gulf of the Farallones and Monterey Bay National Marine Sanctuaries
__________________________________________________________________________________________________________________________________________
A Biogeographic Assessment off North/Central California:
To Support the Joint Management Plan Review for
Cordell Bank, Gulf of the Farallones, and Monterey Bay
National Marine Sanctuaries: Phase I -
Marine Fishes, Birds and Mammals
Prepared by NOAA's Center for Coastal Monitoring and Assessment - Biogeography Team
National Centers for Coastal Ocean Science
December 2003
ABOUT THIS DOCUMENT
In the spring of 2001, NOAA’s National Marine Sanctuary Program (NMSP) and National Centers for Coastal Ocean Science (NCCOS), launched a 24-month effort to assess biogeographic patterns of selected marine species found within and adjacent to
the boundaries of three west coast National Marine Sanctuaries. These sanctuaries, Monterey Bay, Gulf of the Farallones, and Cordell Bank, are conducting a joint review process to update sanctuary management plans. To support this review, NCCOS’s
Biogeography Program is leading a partnership effort to conduct a robust analytical assessment to define important biological areas and time periods within, and adjacent to, current sanctuary boundaries. The assessment was based on a synthesis of
many data sets that were provided by project partners. This document represents the results of the first of two phases of the assessment. Phase I provides data, analytical results, a description of ecosystems and their linkages, identifies data gaps, and
suggests future activities to be addressed in Phase II.
Phase I of the biogeographic assessment was formulated around three integrated study components: 1) an Ecological Linkages Report, 2) biogeographic analyses, and 3) development of Geographical Information System (GIS) data for incorporation into
NMSP’s Marine Information System (MarIS). The majority of the results from the assessment are presented as a suite of GIS maps to visually display species biogeographic patterns across the study area. The body of this document provides examples
of the entire suite of digital map products found on the CD-ROM located on the back cover of this document. The spatial data and additional information, such as digital species distribution maps, and additional details on analytical methodologies are also
presented on the CD-ROM. An HTML version of the CD-ROM can be found on the Biogeography Program website: http://biogeo.nos.noaa.gov/products/canms_cd/.
Results of the assessment are being used to assist the NMSP in addressing issues such as evaluating potential modification of sanctuary boundaries, and changes in management strategies or administration, based on the principles of biogeography. The
progress of the biogeographic assessment for Central and Northern California National Marine Sanctuaries can be followed by consulting NCCOS’s Biogeography Program web site: http://biogeo.nos.noaa.gov/projects/assess/ca_nms/.
For questions and comments, please contact:
Project Team
David Ainley, H.T. Harvey and Associates
Dr. Mark E. Monaco, Biogeography Team Leader
Satie Airamé, National Marine Sanctuary Program
NOAA/NCCOS/CCMA
Charles Alexander, National Marine Sanctuary Program
1305 East West Highway, N/SCI1
Lisa Ballance, Southwest Fisheries Science Center
Silver Spring, MD 20910
Maria Brown, National Marine Sanctuary Program
p. 301-713-3028 x 160
Erica Burton, National Marine Sanctuary Program
f. 301-713-4384
Kenneth Buja, National Centers for Coastal Ocean Science
mark.monaco@noaa.gov
Chris Caldow, National Centers for Coastal Ocean Science
Janet Casey, R.G. Ford Consulting Company
or
John Christensen, National Centers for Coastal Ocean Science
Lawrence Claflin, National Centers for Coastal Ocean Science
Mr. Charles E. Alexander, National Programs Branch Chief
Randall Clark, National Centers for Coastal Ocean Science
NOAA/NMSP
Michael Coyne, National Centers for Coastal Ocean Science
1305 East West Highway,
Andrew DeVogelaere, National Marine Sanctuary Program
Silver Spring, MD 20910
William Douros, National Marine Sanctuary Program
p. 301-713-3125 x 147
R. Glenn Ford, R.G. Ford Consulting Company
f. 301-713-0404
Steve Gaines, University of California Santa Barbara
charles.alexander@noaa.gov
Tracy Gill, National Centers for Coastal Ocean Science
Jamison Higgins, National Centers for Coastal Ocean Science
Dan Howard, National Marine Sanctuary Program
Olaf Jensen, National Centers for Coastal Ocean Science
Carol Keiper, Oikonos
Matthew Kendall, National Centers for Coastal Ocean Science
Kevin McMahon, National Centers for Coastal Ocean Science
Mark Monaco, National Centers for Coastal Ocean Science
Wendy Morrison, National Centers for Coastal Ocean Science
Sean Morton, National Marine Sanctuary Program
Jan Roletto, National Marine Sanctuary Program
Larry Spear, H.T. Harvey and Associates
Lynn Takata, National Centers for Coastal Ocean Science
Mitchell Tartt, National Marine Sanctuary Program
Christine Taylor, National Marine Sanctuary Program
Citation: Ed Ueber, National Marine Sanctuary Program
NOAA National Centers for Coastal Ocean Science (NCCOS) 2003. A Biogeographic Assessment off North/Central Jeannette Waddell, National Centers for Coastal Ocean Science
California: To Support the Joint Management Plan Review for Cordell Bank, Gulf of the Farallones, and Monterey Anne Walton, National Marine Sanctuary Program
Bay National Marine Sanctuaries: Phase I - Marine Fishes, Birds and Mammals. Prepared by NCCOS’s Biogeog- Wendy Williams, R.G. Ford Consulting Co.
raphy Team in cooperation with the National Marine Sanctuary Program. Silver Spring, MD 145 pp. Cover Photo by Kip Evans
i
TABLE OF CONTENTS
Table 8. Average frequency of occurrence of fish species assemblages (percent
About This Document ................................................................................................. i Section 3 Integration of Analyses................................................................................ 129
occurrence calculated for each species and then averaged for each fish
Table of Contents ....................................................................................................... ii Introduction.............................................................................................................. 129
assemblage) for each 1999 midwater site group. Number of trawls in each
Introduction................................................................................................................. 1 Data and Analyses .................................................................................................. 129
site group is provided in the first row. Bold numbers represent influential
The Study Area..................................................................................................... 1 Analytical Map Products .......................................................................................... 130
species assemblages within that site group.................................................... 32
Study Objectives .................................................................................................. 1 References .............................................................................................................. 136
Table 9. Example data matrix for calculating bathymetry SI values for subadult
Section 1 Synopsis of Ecological Linkages Report ................................................ 6 Section 4 Data Sources and Gaps ............................................................................... 138
bocaccio taken in NMFS trawl samples (Rubec et al., 1999). ........................ 36
Introduction........................................................................................................... 6 Introduction.............................................................................................................. 138
Table 10. Example presence/absence information and SI calculation from scientific
National Marine Sanctuaries of Central and Northern California ........................ 6 Conclusions and Recommendations for Future Activities ....................................... 138
literature. ........................................................................................................ 36
Geographic Setting of the Study Area .................................................................. 6 References .............................................................................................................. 139
Table 11. Marine bird species used in this analysis........................................................ 46
Ocean Currents .................................................................................................... 6 Section 5 Summary of Biogeographic Assessment .................................................. 140
Table 12. Summary of at-sea survey data sets used in the analyses............................. 47
Oceanographic Seasons ...................................................................................... 7 Background ............................................................................................................. 140
Table 13. Summary of combined data set effort by ocean season. .............................. . 48
Natural Perturbations ........................................................................................... 7 Biogeographic Assessment off North/Central California.......................................... 140
Table 14. Assignment of warm, cold and neutral periods, based on surface water
Ecosystems .......................................................................................................... 7 Ecological Linkages Report ..................................................................................... 140
temperatures off Cental California .................................................................. 50
Biogeography ...................................................................................................... 8 Biogeographic Analyses .......................................................................................... 141
Table 15. Life history and management information for selected marine birds off
Conclusions .......................................................................................................... 9 Integration of Analyses ............................................................................................ 142
north/central California. ................................................................................... 84
References ........................................................................................................... 9 Data Sources and Gaps .......................................................................................... 143
Table 16. A summary of temporal and spatial patterns in the at-sea survey data
Section 2 Biogeographic Analyses ........................................................................ 10 Phase II Biogeographic Assessment ....................................................................... 143
(1980-2001) of selected marine birds off north/central California. .................. 85
Introduction......................................................................................................... 10 CD-ROM.................................................................................................................. 143
Table 17. Important at-sea areas for marine birds off north/central California,
Assessment Process ......................................................................................... 10 Concluding Comments ............................................................................................ 143
References ......................................................................................................... 10 References .............................................................................................................. 143 based on biomass, density and diversity.. ...................................................... 86
Section 2.1 Biogeography of Fishes....................................................................... 12 Section 6 Phase II Biogeographic Assessment.......................................................... 144 Table 18. Major marine bird colonies along the central California coast . ...................... 87
Introduction......................................................................................................... 12 Acknowledgments......................................................................................................... 145 Table 19. Three most important variables (of nine investigated) having independent
Subsection 2.1.1 Assemblage Analyses ........................................................ 13 effects in explaining the variance in density of 25 selected marine bird
Introduction......................................................................................................... 13 species. .......................................................................................................... 88
Review of Relevant Literature ............................................................................ 13 Table 20. Effects of ocean season and ENSO events on the abundance of 26
Introduction to Clustering ................................................................................... 14 marine bird species off central California between 1985 and 2002, as
List of Tables
Section Summary ............................................................................................... 33 determined through multiple regression analyses.. ........................................ 88
Table 1. Site group results for recreational data. The numbers of trip/location
References ......................................................................................................... 34 Table 21. A summary of changes in marine bird occurrence patterns, as a response
combinations associated with each group as well as average depth, ± stan-
Subsection 2.1.2 Habitat Suitability Modeling ............................................... 35 to warm and cold ocean anomalies, as determined by visual comparison of
dard deviation, are provided. Different letters signify a significant difference
Introduction......................................................................................................... 35 species’ maps during the 1997-1998 El Niño event and the
using Tukey’s pairwise comparison on log adjusted depth with overall
Data and Analyses ............................................................................................. 35 1999-2000 La Niña event................................................................................ 89
alpha set at 0.001. .......................................................................................... 24
Analytical Map Products ..................................................................................... 36 Table 22. Marine mammal species included in this assessment and map types
Table 2. Average frequency of occurrence of fish species assemblages (percent
Section Summary ............................................................................................... 44 developed for them (Phase I and Phase II) .................................................... 92
occurrence calculated for each species and then averaged for each fish
Reviews .............................................................................................................. 44 Table 23. Summary of at-sea data sets used in the preliminary marine mammal
assemblage) for each recreational site group. Numbers in bold represent
Reviewers........................................................................................................... 44 analyses. ......................................................................................................... 93
influential species assemblages within that site group. .................................. 24
References ......................................................................................................... 45 Table 24. Summary of combined data set effort for mammals, by ocean season.. ........ 93
Table 3. Site group results for shelf trawl data. The numbers of trawls associated with
Section 2.2 Biogeography of Marine Birds ............................................................ 46 Table 25. Preliminary life history and management information for selected marine
each group as well as average depth ± standard deviation are provided.
Introduction......................................................................................................... 46 mammals off north/central California.. .......................................................... 123
Different letters signify a significant difference using Tukey’s pairwise
Data and Analyses ............................................................................................ 46 Table 26. Summary statistics and parameter estimates for spatial models. .................130
comparison on log adjusted depth with overall alpha set at 0.001. ................ 26
Analytical Map Products ..................................................................................... 50 Table 27. Matrix of data sets and their associated characteristics that were used or
Table 4. Average frequency of occurrence of fish species assemblages (percent
Section Summary ............................................................................................... 84 referenced in the biogeographic assessment. ............................................. 138
occurrence calculated for each species and then averaged for each fish
Major Section Contributors ................................................................................. 89
assemblage) for each shelf site group. Numbers in bold represent
Reviewers........................................................................................................... 89 List of Figures
influential species assemblages within that site group. .................................. 27
Personal Communications ................................................................................. 89 Figure 1. Project flow diagram showing steps to complete Biogeographic
Table 5. Site group results for slope trawl data. The numbers of trawls associated with
References ......................................................................................................... 90 Assessment Phase 1........................................................................................ 1
each group as well as average depth ± standard deviation are provided.
Section 2.3 Biogeography of Marine Mammals ..................................................... 92 Figure 2. Locator map of entire study area from Point Arena to Point Sal. National
Different letters signify a significant difference using Tukey’s pairwise
Introduction......................................................................................................... 92 Marine Sanctuary boundaries shown in red. .................................................... 2
comparison on log adjusted depth with overall alpha set at 0.001. ................ 28
Data and Analyses ............................................................................................. 92 Figure 3. Detailed locator map of northern study area from Bodega Head to
Table 6. Average frequency of occurrence of fish species assemblages (percent
Analytical Map Products ..................................................................................... 94 Pescadero Point. National Marine Sanctuary boundaries shown in
occurrence calculated for each species and then averaged for each fish
Section Summary ............................................................................................. 123 red............. ....................................................................................................... 3
assemblage) for each slope site group. Bold numbers represent influential
Discussion ........................................................................................................ 125 Figure 4. Detailed locator map of central study area from Pescadero Point to Pfeiffer
species assemblages within that site group.................................................... 29
Major Section Contributors ............................................................................... 125 Point. National Marine Sanctuary boundaries shown in red. ........................... 4
Table 7. Average frequency of occurrence of fish species assemblages (percent
Reviewers......................................................................................................... 125 Figure 5. Detailed locator map of southern study area from Pfeiffer Point to
occurrence calculated for each species and then averaged for each fish
Personal Communications ................................................................................126 Point Sal. National Marine Sanctuary boundaries shown in red. ..................... 5
assemblage) for each 1998 midwater site group. Number of trawls in each
References ........................................................................................................126 Figure 6. Biogeographic assessment process ............................................................... 10
site group is provided in the first row. Bold numbers represent influential
Figure 7. 3-D image of bathymetric relief within and adjacent to the sanctuaries .......... 11
species assemblages within that site group.................................................... 31
ii
TABLE OF CONTENTS
Figure 23. Location of site groups for NMFS slope trawls. Lines showing the 50, Figure 51. Rhinoceros auklet, seasonal density, high use areas and breeding
Figure 8. Biogeographic approach to fish analysis ........................................................ 12
100, 200 and 2,000 depth contours are provided. .......................................... 29 colonies. .......................................................................................................... 69
Figure 9. Hypothetical example of the methods used to determine species assemblag-
Figure 24. Species assemblage results for the midwater trawls utilizing all data from Figure 52. Marine bird density, by season and for all seasons........................................ 71
es, site groups, and the interaction between species assemblages and
1986 to 2001. Assemblages are named for the most influential species in Figure 53. Marine bird biomass, by season and for all seasons. .................................... 72
site groups...................................................................................................... 15
each group. Non-italicized species were consistently placed into the same Figure 54. Marine bird diversity, by season and for all seasons...................................... 73
Figure 10. Standard deviation of bathymetry calculated for a 1 km radius around each
species assemblage >80% of the time; italicized species tended to roam Figure 55. Major marine bird breeding colonies. . ........................................................... 75
cell. Results are presented in standard deviations above or below
into other assemblages with random sampling.............................................. 30 Figure 56. Density in warm, cold, and neutral periods: 1980-2001. ............................... 76
the mean. ....................................................................................................... 16
Figure 25. Species assemblage results for the midwater trawls conducted in 1998. As- Figure 57. Biomass in warm, cold and neutral periods: 1980-2001. ............................... 78
Figure 11. Species richness of individual NMFS shelf and slope trawls. ...................... . 17
semblages are named for the most influential species in each group. Non- Figure 58. Diversity in warm, cold and neutral periods: 1980-2001. ............................... 80
Figure 12. Mean species richness of NMFS shelf and slope trawls for 5’ grid cells. The
italicized species were consistently placed into the same species assemblage Figure 59. Density during El Niño and La Niña events, 1997-2000. ............................... 82
deviation is shown as an overlay to provide an indication of the variability
>80% of the time; italicized species tended to roam into other Figure 60. El Niño/La Niña Event changes, as an example of regime shift effects... ...... 83
in results for each grid cell. ............................................................................. 18
assemblages with random sampling.. ............................................................. 31 Figure 61. Total at-sea survey effort for marine mammal analysis. ................................. 94
Figure 13. Species diversity of individual NMFS shelf and slope trawls. . ...................... 19
Figure 26. Location of site groups for NMFS 1998 midwater trawls. Lines showing Figure 62. Maps for southern sea otter: rangewide count and linear density,
Figure 14. Mean species diversity of NMFS shelf and slope trawls for 5’ grid cells. The
the 50, 100, 200, and 2,000 depth contours are provided. ............................ 31 fall 2001 and spring 2002. .............................................................................. 95
deviation is shown as an overlay to provide an indication of the
Figure 27. Species assemblage results for the midwater trawls conducted in 1999. As- Figure 63. Map for California sea lion: haulouts and rookeries. ...................................... 96
variability in results for each grid cell. ............................................................. 20
semblages are named for the most influential species in each group. Non- Figure 64. Maps for California sea lion: seasonal at-sea densities, high use areas
Figure 15. Species richness of rockfish from individual NMFS shelf and slope trawls.... 21
italicized species were consistently placed into the same species assemblage and rookeries. ................................................................................................. 97
Figure 16.The relationship between depth and rockfish richness showing mean rockfish
>80% of the time; italicized species tended to roam into other Figure 65. Map for Steller sea lion: at-sea sightings and survey effort,
richness (for 10 meter depth intervals between 50-1,300 meters).
assemblages with random sampling... ............................................................ 32 rookeries and haulouts.................................................................................... 99
The relationship was fit with a smoothing spline, lambda = 1,000,000. .......... 21
Figure 28. Location of site groups for NMFS 1999 midwater trawls. Lines showing Figure 66. Maps for northern fur seal: seasonal at-sea densities, high use areas
Figure 17. NMFS shelf and slope trawls with the highest species diversity, species rich-
the 50, 100, 200, and 2,000 depth contours are provided. ............................ 32 and rookery. .................................................................................................. 101
ness, and rockfish richness are mapped. The underlying map illustrates the
Figure 29. Overlap between the three data sets that analyzed demersal fish: CDF&G Figure 67. Map for harbor seal: at-sea sightings, survey effort and haulouts................ 103
bathymetric complexity of the study area and can be used to identify
recreational (yellow), NMFS shelf (green), and NMFS slope (orange). .......... 33 Figure 68. Map for northern elephant seal: at-sea sightings and survey effort,
the shelf break. ............................................................................................... 22
Figure 30. Species habitat suitability modeling approach. .............................................. 35 rookeries and haulouts.................................................................................. 105
Figure 18.Species assemblage results for the recreational data. Assemblages are named
Figure 31. Polynomial regression curve fit with mean log abundance by categorical Figure 69. Maps for Dall’s porpoise: seasonal at-sea densities and high use areas..... 107
for the most influential species in each group. Assemblages are arranged
bathymetric class for subadult bocaccio. ........................................................ 35 Figure 70. Maps for Pacific white-sided dolphin: seasonal at-sea densities and
from shallow to deep, unless they are influential at all or none of the depths.
Figure 32. Bathymetric map for the north/central California study area. Red lines high use areas. ............................................................................................. 109
The assemblages that were not influential at any depth were composed of
indicate National Marine Sanctuary boundaries. . .......................................... 37 Figure 71. Maps for Risso’s dolphin: seasonal at-sea densities and high use areas. ....111
relatively rare species, making depth associations indiscernible given the
Figure 33. Substrate types for the north/central California marine region. ...................... 38 Figure 72. Maps for northern right-whale dolphin: seasonal at-sea densities and
methodology for defining “influential” assemblages. Non-italicized species
Figure 34. Potential distribution of habitat suitability for adult and subadult bocaccio. high use areas. ............................................................................................ 113
were consistently placed into the same species assemblage >80% of the
Map inset contains validation statistics. SI values for bathymetry and Figure 73. Map for blue whale: at-sea sightings and survey effort... ............................. 115
time; italicized species tended to roam into other assemblages with
substrate are graphically displayed below the map. ....................................... 39 Figure 74. Maps for humpback whale: seasonal at-sea densities and high use areas. 117
random sampling. ........................................................................................... 23
Figure 35. Potential distribution of habitat suitability for adult and subadult Dover sole. Figure 75. Maps for gray whale: seasonal at-sea densities and high use areas........... 119
Figure 19. Location of CDF&G recreational fishing data in 2.5 minute grids which are
Map inset contains validation statistics. SI values for bathymetry Figure 76. Maps for Dall’s porpoise: SWFSC stock assessment data: average
color coded according to the average depth of the fishing trips within the grid
and substrate are displayed below the maps. ................................................ 40 group size of sightings and survey effort....................................................... 121
cell. Lines showing the 50, 100, 200, and 2,000 depth contours
Figure 36. Potential distribution of habitat suitability for adult dungeness crab. Map inset Figure 77. Maps for blue whale: SWFSC stock assessment data: average
are provided. ................................................................................................... 25
contains validation statistics. SI values for bathymetry and substrate group size of sightings and survey effort....................................................... 122
Figure 20. Species assemblage results for the shelf trawls. Assemblages are named
are graphically displayed below the map. ....................................................... 41 Figure 78. Maps for humpback whale: SWFSC stock assessment data: average
for the most influential species in each group. Assemblages are arranged
Figure 37. Areas of groundfish potential hot spots based on mean fish species HSI group size of sightings and survey effort....................................................... 122
from shallow to deep, unless they are influential at all or none of the depths.
models and overlap of predicted highly suitable habitats. .............................. 42 Figure 79. Pictogram of species diversity. ..................................................................... 129
The assemblages that were not influential at any depth were composed of
Figure 38. Areas of potential habitat importance based on mean HSI models for Figure 80. Estimated diversity (a), density (b), and hot spots (top 20%)
relatively rare species, making depth associations indiscernible given the
selected species assemblages. ...................................................................... 43 (c) for marine birds. ....................................................................................... 131
methodology for defining “influential” assemblages. Non-italicized species
Figure 39. Spatial extent of data sets used in the marine bird analysis: Figure 81. Relationships between bird diversity and bathymetric variance and
were consistently placed into the same species assemblage >80% of the time;
ship-based surveys. ....................................................................................... 47 between bird density and depth. ................................................................... 131
italicized species tended to roam into other assemblages with random
Figure 40. Spatial extent of data sets used in the analysis: aerial surveys. .................... 48 Figure 82. Estimated diversity (a), density (b), and hot spots (top 20%) (c) for fish...... 132
sampling. ....................................................................................................... 26
Figure 41. Total survey effort for marine bird analyses.................................................... 49 Figure 83. Integration option 1, diversity hot spots (top 20%) for fish and
Figure 21.Location of site groups for NMFS shelf trawls. Lines showing the 50, 100,
Figure 42. Western and Clark’s grebes, seasonal density and high use areas............... 51 marine birds. Coastal kelp bed areas are also shown. ................................. 133
200, and 2,000 depth contours are provided. ................................................. 27
Figure 43. Northern fulmar, seasonal density and high use areas. ................................. 53 Figure 84. Integration option 2, density hot spots (top 20%) for marine birds and fish.
Figure 22. Species assemblage results for the slope trawls. Assemblages are named
Figure 44. Sooty shearwater, seasonal density and high use areas. .............................. 55 Coastal kelp bed areas are also shown. ....................................................... 134
for the most influential species in each group. Assemblages are arranged
Figure 45. Ashy storm-petrel, seasonal density, high use areas, and breeding colonies.57 Figure 85. Integration option 3, diversity and density, hot spots (top 20%) for fish
from shallow to deep, unless they are influential at all or none of the depths.
Figure 46. Leach’s storm-petrel, seasonal density, high use areas, and and marine birds. Coastal kelp bed areas are also shown. .......................... 135
The assemblages that were not influential at any depth were composed of
breeding colonies.. .......................................................................................... 59
relatively rare species, making depth associations indiscernible given the
Figure 47. Scoters, seasonal density and high use areas............................................... 61
methodology for defining “influential” assemblages. Non-italicized species
Figure 48. Brown pelican, seasonal density and high use areas. ................................... 63
were consistently placed into the same species assemblage >80% of the time;
Figure 49. Black-legged kittiwake, seasonal density and high use areas. ...................... 65
italicized species tended to roam into other assemblages with random
Figure 50. Common murre, seasonal density, high use areas and breeding colonies. . . 67
sampling.......................................................................................................... 28
iii
INTRODUCTION
In the spring of 2001, NOAA’s National Marine Sanctuary Northern Santa Barbara county (Figure 2). Based on the Cali-
Program (NMSP) and National Centers for Coastal Ocean fornia Department of Fish and Game’s 200-meter resolution
Ecological Linkages Report
Science (NCCOS), in consultation with the National Marine bathymetry data, this map displays the locations of prominent
PIs PIs develop
PIs deliver
Develop Statement of Ecolinkage
Fisheries Service (NMFS), launched a 24-month effort to bathymetric features occurring off the coast, including the con-
deliver presentation
iterative draft
Work and select PIs to Report
define and assess biogeographic patterns of selected marine tinental shelf/slope interface. In support of this assessment,
final of results
reports for
write the report
species found within and adjacent to the boundaries of three the National Geophysical Data Center, Monterey Bay Aquarium
report
NOAA review
west coast National Marine Sanctuaries. These sanctuaries, Research Institute, and the NMSP jointly developed the first
Biogeographic Analyses
Monterey Bay, Gulf of the Farallones, and Cordell Bank are high-resolution 70-meter bathymetry maps for the region. The
Evaluate data Revise Determine Phase 1
Conduct
conducting a joint review process to update sanctuary man- highly resolved bathymetry is shown on the three regional maps
Conduct
Identify
DATA GIS
sets and species and optimal analyses review and
species and
agement plans. The management plans for these sanctuaries to highlight the complexity of the seafloor and to begin outlining
Project habitat list analytical Analytical
conduct and develop revise
habitats of
have not been updated for over ten years and the status of the the multitude of distinct ecosystems occurring in this region
COLLECTION
preliminary based on data approach Results
draft
‘Kickoff analytical
interest with
natural resources and their management issues in and around (Figures 3-5). The Northern Region map focuses on Cordell
analyses quality and products for products
sanctuaries
Meetings’
the sanctuaries may have changed. In addition, significant ac- Bank and Gulf of the Farallones National Marine Sanctuar-
availability review
staff
with Team
complishments in research and resource assessments have ies. Cordell Bank, the Farallon Islands, Bodega and Pioneer
Members
been made within the region. Thus, it is important to incorporate Canyons, and Gumdrop and Pioneer Seamounts appear as
GIS Data Development Digital
new and expanding knowledge into the revised management very clear features along the seafloor. The focus of the Central
Products
Organize data and Integrate into NMSP’s
Data collection and Metadata
plans for these Sanctuaries. Region map is Monterey Bay. The most significant feature in
MarIS system
standardization creation transfer to CD-ROM
this regional view is the Monterey Canyon, although the nearby
As part of the review process, the NMSP requires an integrated canyons, Ascension, Año Nuevo, Cabrillo, Soquel, and Carmel,
Biogeographic
Interim Products
biogeographic assessment of the spatial and temporal distribu- are also labeled. Other significant bathymetric features in this
Assessment
tions of marine resources off north/central California. The NMSP region include the Guide Seamount, Sur Ridge, and Shepard
Draft
Interim Product:
Create web page for Phase I
Biogeographic
Biogeographic
headquarters and sanctuary field personnel have partnered Meander. The bathymetry for most of the Southern Region
project information and
Assessment
Assessment Atlas
product review
with NCCOS’s Biogeography Team to conduct this assessment. map is less resolved than the other two, as the frequency of
Biogeographic
The biogeographic assessment includes the identification and sampling was significantly less. In this area, the Sur and Lucia
Assessment
characterization of important biological areas and time periods Canyons are found, as well as Santa Lucia Bank and one of
Phase II
off the coast and addresses existing and emerging issues con- the most prominent features in this region, the Davidson Sea-
cerning management of biotic resources in the area. Results of mount. Descriptions of the features observed in these maps,
this assessment aid the NMSP in addressing issues, such as Figure 1. Project flow diagram showing steps to complete Biogeographic Assessment Phase 1. along with the linkages and processes operating to influence
potential modification of sanctuary boundaries and changes in the distribution of associated biota, are found in the Ecological
management approaches based on the principles of biogeog- CD-ROM on the back cover of this document. To enable devel- understanding the potential implications to changes in sanctu- Linkages Report.
raphy. The publication of this document completes Phase I of opment and integration of these components and to support ary boundaries or management strategies relative to marine
the biogeographic assessment for the North/Central California project management, the overall process used to conduct the biogeography within and adjacent to the sanctuaries. STUDY OBJECTIVES
National Marine Sanctuaries. The assessment and additional biogeographic assessment is shown in Figure 1. Based on consultations with NMSP field and headquarters staff
ecosystem characterization of habitats and species (e.g., es- and requirements to update the sanctuary management plans,
To enable NMSP and others to make maximum use of the spa-
tuaries) will continue over the next few years. The initial plans The Ecological Linkages Report is a comprehensive synthesis tial data generated from this study and other activities that are the following study objectives were addressed:
of existing information on ecological relationships between ma- supporting the joint management plan revision process (e.g.,
for Phase II are discussed in Section 6 of this document.
rine biota and the habitats they utilize along the West Coast. economic assessments), the NMSP is developing a GIS tool. 1. Identify and compile available priority biological and en-
The Phase I assessment is based on biogeographic patterns The report is much broader in geographic scope than the This GIS tool, the Marine Assessment and Resource Informa- vironmental data sets in the study area in order to conduct
of fishes, macroinvertebrates, marine mammals, and marine project study area and provides the context to understand tion System (MarIS), has been designed to facilitate the orga- biogeographic analyses.
birds and the distribution of their habitats. The study did not overall assessment results relative to the biogeography of the nization and analysis of spatial data to support NMSP manage-
attempt to define biogeographic patterns along the entire US West Coast. In addition, the report addresses near-shore and ment questions and issues within and across the sanctuaries. 2. Conduct marine biogeographic analyses of available data
west coast nor in very near shore environments (e.g, estuaries). estuarine ecosystems while quantitative data analyses were All GIS compatible data, the Ecological Linkages Report, and to define significant biological areas (i.e., “hot spots”) and time
Rather, the study area was restricted to the marine area from conducted only for the marine waters of the National Marine products from the biogeographic analyses component are found periods, based on species distributions, abundance, habitats,
Point Arena (in the north) to Point Sal, California (in the south). Sanctuaries. life stage function, and community metrics (e.g., species rich-
on the companion CD-ROM. The contents of the CD-ROM are
The Assessment was based on a synthesis of data provided ness, diversity).
also found on the web at http://biogeo.nos.noaa.gov/products/
by project partners (e.g., NMFS fishery independent surveys). The biogeographic analyses component includes a suite of canms_cd/. All of the applicable digital data will be integrated
The biogeographic assessment was formulated around three quantitative spatial and statistical analyses based on the dis- into MarIS. 3. Produce a report that describes the ecological components
integrated study components: 1) an Ecological Linkages Re- tribution and abundance of fishes, marine birds, and marine and linkages between the estuarine, coastal, and marine eco-
port, 2) biogeographic analyses, and 3) development of spatial mammals found in the Point Arena to Point Sal study area. The THE STUDY AREA systems of north/central California.
data for incorporation into NMSP’s Marine Information System analytical results contributed to defining important biological The study area, shown on the locator maps, extends from Point
(MarIS). GIS-based data and additional information, such as areas throughout the region, based on visualization of species Arena, in the southern portion of Mendocino county, to Point 4. Develop GIS compatible data for integration into the
the complete Ecological Linkages Report, can be found on the distribution patterns and community metrics. The results aid in Sal, just south of Pismo Beach and the Nipomo Dunes area in NMSP’s Marine Information System (MarIS) to support
1
INTRODUCTION
sanctuary staff in developing and evaluating resource manage- 124°W 123°W 122°W 121°W
ment scenarios.
39°N
39°N
Northern/Central California
Point Arena
5. Support sanctuary staff in the integration of biogeographic 3
na
assessment products into revisions of the sanctuary manage-
ut
ica
Region of Interest
ment plans.
lm Fort
ile Ross n
The publication of this product completes Phase I efforts to lin sia r
us ive
e
meet objectives 1-4. The data and analytical results from these RR
objectives will be used to address objective 5 over the next year. 0 10 20 40 60 80 100
This investigation synthesized many databases and information
Kilometers
sources for the study area. The data and information originated Cordell Bank
from a wide variety of government, academic, and private in- NMS
38°N
38°N
stitution studies that had different objectives, study areas, and
Gulf of the
methodologies. Thus, several criteria were used in selecting
Farallones
appropriate data sets for biogeographic analyses. For example, San
NMS
the selection process favored databases that addressed the en- Francisco
Gumdrop
tire study area and were conducted relatively consistently over
Seamount
time. Thus, small databases that were limited in both content
and spatial coverage were generally not useful in developing Pioneer Monterey Bay
Seamount
the assessment. When appropriate, these types of databases NMS
were used to aid in the interpretation of results and to develop
Santa
and validate species habitat suitability models. Cruz
37°N
37°N
Guide
Seamount
The following sections of this document provide information
on the data compiled, analytical approaches, and Phase I as-
Cany on
rey
sessment results. For the three main study components, (the
te
Ecological Linkages Report, biogeographic analyses, and the
n
Monterey
Mo
CD-ROM contents), the primary information found within each
component is introduced, methods described, and represen-
tative or example results provided. Many of the results are
presented as map products to easily convey the biogeographic
distribution of species and associated habitats. In addition, a
36°N
36°N
summary of the biogeographic assessment is found in section
Davidson
5. For more complete information (e.g., complete suite of digital Seamount
species maps), please review and use the digital contents of
the CD-ROM or the web version at http://biogeo.nos.noaa.gov/
products/canms_cd/. Cambria
35°N
35°N
Area
Point Sal
Enlarged
400
200 m
50 m
0 m
20 1000
m
00
Point Arguello
m
30
00
Point Conception
m
124°W 123°W 122°W 121°W
Figure 2. Locator map of entire study area from Point Arena to Point Sal. National Marine Sanctuary boundaries shown in red.
2
124°W 123°30'W 123°W 122°30'W 122°W
Bodega Head
Northern Region
10
20
00
0
m
Bodega
m
Bay
20
n y on Tomales Point
Bode g a Ca
00
m
30
0 5 10 20 30 40 50
00
50 m
m
Tomales Kilometers
Bay
mil utical
ne
Cordell Bank
e li
Drakes
a
3n
National Estero
San Pablo Bay
Cordell
Marine Sanctuary Limantour Estero
Bank
Drakes
38°N
38°N
Bay
Point
Fa
Reyes Bolinas
Double Point Lagoon
ra
ll
Gulf of the Farallones
n
o
Duxberry
National Reef
R
id
Marine Sanctuary
g Fanny
e North
Shoal
San Francisco Bay
Farallones
Fa
(esti mated sum
ra Middle
llo San
S
Farallon
n
a
Francisco
Es
n
ca
F
rp
ra
Southeast
m
mer
en Farallones
n
T id e nt)
c
t
e xt
al
is
Pacifica o
Plu
c
m
B
e
a
y
37°30'N
37°30'N
Gumdrop Pillar Point Half
Seamount
Moon
Bay
Monterey Bay
Pioneer
Area
Seamount
Enlarged National
Marine Sanctuary
er Canyon
ne
50
m
0m
o
100
Pi
20
m
m
0
3000
m
Pescadero Point
2000
124°W 123°30'W 123°W 122°30'W 122°W
Figure 3. Detailed locator map of northern study area from Bodega Head to Pescadero Point. National Marine Sanctuary boundaries shown in red.
3
123°30'W 123°W 122°30'W 122°W
Pescadero Point
Central Region
50
m
20
m
00
0
Pigeon Point
30
m
200
0 m 10
00
m
Point Año Nuevo
37°N
37°N
Santa
Guide
n
yo
Cruz
Seamount
an
nC
n
yo
an
si o
Elkhorn
C
en
o Slough
c
ev
Mo
As
yon
Nu
on
o
nte
an
ny
Añ
Ca
oC
Moss
rey Bay
el
il l
Landing
qu
br
3 n a line
Ca
So
mile
Monterey Bay
utica
National
Sa
Marine Sanctuary
l
lin
C ar m
C any o
a sR
Point Pinos ive
r
el
rey Ca
Monte
n
ny
on
Monterey
Cypress Point
Carmel Bay Carmel
Point Lobos
36°30'N
36°30'N
e g
Su r Rid
Point Sur
Big Sur
Area
Enlarged
Pfeiffer Point
20
50
10 m
00
00
Shepard m
30
m
200 m
00
Meander
m
123°30'W 123°W 122°30'W 122°W
Figure 4. Detailed locator map of central study area from Pescadero Point to Pfeiffer Point. National Marine Sanctuary boundaries shown in red.
4
123°W 122°30'W 122°W 121°30'W 121°W 120°30'W
Pfeiffer Point
100
20
30
200 m 50 m
0 m
00
00
Southern Region
m
m
on
ny
Monterey Bay
a
rC
Su National
Marine Sanctuary
Lopez Point 0 5 10 20 30 40 50
36°N
36°N
on
y
a C an Kilometers
ci yo n
Lu Mill Creek Ca n
Cape San Martin
a
Vill on
m
ny
00
Ca z
m 20
00 u
Cr n
30
La ny o
Ca
Davidson
Seamount
Point Piedras Blancas
Cambria
35°30'N
35°30'N
Point Estero
Estero
Bay
Morro
Bay
Point San Luis San
Sa
Luis Obispo
Bay
nt
a
Lu
50 m
ci
200
a
mile tical
35°N
35°N
Ba
line
m
Area
u
nk
3 na
Enlarged
30
4000 m
1000
Point Sal
00
20
m
00
m
m
123°W 122°30'W 122°W 121°30'W 121°W 120°30'W
Figure 5. Detailed locator map of southern study area from Pfeiffer Point to Point Sal. National Marine Sanctuary boundaries shown in red.
5
Section 1: SYNOPSIS OF ECOLOGICAL LINKAGES REPORT
INTRODUCTION a variety of coastal and marine habitats, such as rugged rocky
The Ecological Linkages Report complements the biogeo- shores, lush kelp forests, and several underwater canyons, the
graphic analyses conducted by The National Centers for largest of which is the Monterey Submarine Canyon. North of
Coastal Ocean Science (NCCOS) by providing an overview Partington Point and within the Gulf of the Farallones, the con-
of the physical and biological characteristics of the region. Key tinental shelf is relatively wide and shallow. South of Partington
ecosystems and species occurring in estuarine and marine Point, the Sanctuary generally protects deep ocean, owing to
waters off northern and central California are highlighted and the consistently narrow continental shelf that extends south to
linkages between them discussed. In addition, this report de- Point Conception. The diverse array of habitats in the Sanctu-
scribes biogeographic processes operating to affect species’ ary is home to 33 marine mammals, 94 species of seabirds,
distributional patterns. The biogeographic analyses build upon at least 345 species of fishes, and numerous invertebrates,
this background to further understanding of the biogeography and plants.
of this region. The following material is a synopsis of the report
either excerpted or directly summarized from the completed GEOGRAPHIC SETTING OF THE STUDY AREA
document (Airamé et al., 2003) found on the companion CD- The study region extends from Point Arena, a small peninsula
ROM. on an elevated coastal plain in the southern portion of Men-
docino County, to Point Sal, just south of Pismo Beach and
NATIONAL MARINE SANCTUARIES OFF CENTRAL AND the Nipomo Dunes area. The region consists of a multitude
NORTHERN CALIFORNIA of diverse and important ecosystems that are very unique in
The study area, from Point Arena to Point Sal, includes three their assemblages of marine organisms. Beginning near-shore,
national marine sanctuaries (Cordell Bank, Gulf of the Faral- the coast of California, especially north of Point Reyes and
lones, and Monterey Bay) encompassing marine and estuarine south of Point Pinos, is renowned for its strikingly beautiful,
habitats along the central and northern coast of California. dramatic rocky cliffs. Pocket beaches occur along the coast
Together, these contiguous national marine sanctuaries include where streams and rivers deposit sediment along the shore.
more than 650 km of coastline, from Bodega Bay, north of San Rivers that flow over broad, flat expanses of soft sediments
Francisco, to Cambria, near San Luis Obispo, and a total area into the ocean may be strongly influenced by tides and are
of approximately 18,000 km2. frequently associated with upland and salt marshes, sandy
beaches, intertidal flats, and estuaries. Estuaries and lagoons
The Gulf of the Farallones National Marine Sanctuary, estab- commonly form where rivers enter the ocean, mixing fresh and
lished in 1981, includes an area of 3,250 km2 off the northern salt water. Rocky shores, which are more resistant to erosion
and central California coast. The Gulf of the Farallones extends than the sandy beaches, support a complex intertidal com-
beyond the Sanctuary’s boundaries and is one of the broadest munity, influenced primarily by the semidiurnal movements of
sections of the continental shelf off the U.S. West Coast. Be- tides. Moving offshore, subtidal communities are strongly influ-
sides the broad shelf, the major oceanographic feature that af- ocean’s surface. The base of the Bank is over 120 m deep. The enced by sediment type, nutrient input, and depth. The majority
at all levels of the marine food web. In periods when upwelling
fects this coastal region is the San Francisco Bay Plume, which, combination of oceanographic conditions and undersea topog- of the continental shelf is sandy, but rocky outcrops cover a
is reduced, the nutrient input from the San Francisco Plume
under certain conditions, extends outwards to all areas of the raphy of Cordell Bank supports a diverse and productive marine portion of it, forming submerged reefs, seamounts, and other
becomes important. The Farallon Islands, which are protected
Gulf. The Golden Gate, from which the plume emanates, lies ecosystem. A persistent upwelling plume projects southward features. Marine algae, unable to attach to the shifting sandy
as a National Wildlife Refuge, are home to the largest concen-
midway along this section of coast. The Gulf of the Farallones and offshore from Point Arena and Point Reyes, transporting sediments, find more secure substrate on rocky reefs. At the
tration of breeding seabirds in the contiguous United States (12
National Marine Sanctuary itself, however, extends along the nutrients and organisms suspended in the water column into the shelf break, the continental slope drops precipitously to depths
species), as well as one of the richest assemblages of pinnipeds
coast only as far south as Rocky Point, Marin County (where bank’s relatively shallow waters. Insolation fuels primary pro- of over 3000 m. Sediments, transported down the continental
(5 species). About 163 species of marine, coastal, and estuarine
the Gulf of the Farallones National Marine Sanctuary abuts the ductivity and eventually influences the entire food web through slope and submarine canyons, collect in broad fans at the
birds and 36 species of marine mammals use the Sanctuary
Monterey Bay National Marine Sanctuary). Offshore, the Gulf direct and indirect trophic linkages. This high local productivity base of the slope. Below the rise, the abyssal plain is relatively
during breeding or migration. Further, great white sharks are
of the Farallones National Marine Sanctuary extends farther supports abundant resident populations of invertebrates, fishes flat, broken occasionally by such features as seamounts and
attracted to marine mammal colonies on the Farallon Islands,
south to the waters west of San Mateo County. Habitats within (240 species), seabirds (69 species), and marine mammals (28 small depressions. It is this array of ecosystems, combined
Point Año Nuevo, and Año Nuevo Island.
the Gulf of the Farallones National Marine Sanctuary include species), and attracts many migratory species. with the oceanographic processes affecting the composition
rocky shores, sandy beaches, estuaries, lagoons and bays, and abundance of marine organisms in them, that make this
The Cordell Bank National Marine Sanctuary, designated in
as well as the Farallon Islands and the subsurface Farallon The Monterey Bay National Marine Sanctuary, established in such a unique area.
May 1989, includes an area of 1,362 km2 off the coast of central
Ridge. The entire stretch of the broad shelf within the physi- 1992, is the largest of 13 marine sanctuaries administered by
California (Figure 2: Northern Region). Cordell Bank is located
cal features described above is strongly influenced by coastal the National Marine Sanctuary Program. The Sanctuary ex- OCEAN CURRENTS
at the edge of the continental shelf, about 80 km northwest of
upwelling and the San Francisco Bay Plume. The upwelled tends from Rocky Point to Cambria Rock, encompassing nearly The cold water California Current and comparatively warm-
the Golden Gate Bridge and 33 km west of Point Reyes. The
waters, which support tremendous phytoplankton production, 450 km of shoreline and 13,780 km2 of ocean, extending an water Davidson Currents are major forces shaping the eco-
main feature of the Sanctuary is an offshore granite bank, 7 km
are advected offshore into the California Current as eddies and average distance of 32 km from shore. At its deepest point, the systems in and around the study region. They affect upwelling
wide and 15 km long. The rocky bank emerges from the soft
jets. These productive waters stimulate growth of organisms Sanctuary reaches a depth of 3,250 m. The Sanctuary includes and downwelling and, consequently, the amount of productivity
sediments of the continental shelf, reaching within 37 m of the
6
Section 1: SYNOPSIS OF ECOLOGICAL LINKAGES REPORT
fishes decrease with plankton abundance. Marine mammals ing spring and fall migrations and the winter months. Migratory
and seabirds, which depend on these organisms for food, species require consistent sources of food and shelter along
suffer food shortages, leading to widespread starvation and their migration route. If the distances between wetlands are too
decreased reproductive success. large, migrating birds may become exhausted and disoriented,
increasing mortality.
Every 20-30 years, the surface waters of the central and
northern Pacific Ocean shift several degrees from the mean Numerous marine species use embayments, lagoons, and
temperature. Such shifts in mean surface water temperature, estuaries as spawning and nursery grounds. Bat rays, leop-
known as the Pacific Decadal Oscillation, have been detected ard and smoothhound sharks, midshipman, Pacific herring,
5 times during the past century, with the most recent shift in starry flounder, staghorn sculpin, surf perch of several species,
1998. The Pacific Decadal Oscillation impacts production in jacksmelt, topsmelt, and pile perch mate and bear their young
the eastern Pacific Ocean and, consequently, affects organism in estuarine habitats. Shallow, coastal waters of central and
abundance and distribution throughout the food chain. northern California also are critical habitat for chinook and,
especially, coho salmon as they travel en route to spawning
Ocean waters off the coast of California have warmed consid- grounds in autumn and winter.
erably over the last 40 years. It is not yet clear if this warming
is a consequence of an interdecadal climate shift or global Geology and other physical forces influence the structure of
warming. the coastline. Energetic forces of water and wind erode the
rocky coastline, creating the dramatic rocky intertidal habitat
In response to these three phenomena, some species have characteristic of the northern and central California coast. Small
shifted their geographic ranges northward, altering the com- beaches form along the northern California coast where wind
position of local assemblages. and waves erode granite and basalt cliffs. Further sourth, ero-
sion of soft shale and sandstone bluffs creates the broad sandy
ECOSYSTEMS beaches typical of southern California.
The Land-Sea Interface
Rivers carry freshwater and sediments to bays, estuaries, and Sandy beach and rocky intertidal habitats are divided into dis-
the ocean. Thirteen major watersheds are located along the tinct biological zones relative to height above mean high tide.
central California coast. Historically, these supported large In part, species’ distributions are affected by their physiologi-
numbers of coho and chinook salmon, steelhead trout, and cal tolerance to temperature, moisture, and salt. In response
sturgeon. Today, many native anadromous fish stocks through- to these and other physical factors, the number of species of
out California are in danger of extinction. General degradation marine algae, gastropods, and fishes increases with depth in
of upland watershed and freshwater ecosystems is a major the intertidal zone.
factor in the decline.
seen along the coast. Additionally, where the two converge the study region. Over a 12 month time-frame, the study area
at Point Conception, a barrier is created that many species is exposed to three distinct oceanographic periods that vary Nutrients processed in marine systems are essential to commu-
Two major estuaries in northern and central California are
do not cross. The species north of Point Conception, encom- with respect to prominence and location of ocean currents. nities using sandy beaches and rocky intertidal habitat. Waves
San Francisco Bay and Tomales Bay. Several smaller estu-
passing the entire study region right up through Washington These periods, described by upwelling (March to August), wind carry and deposit plankton, macroalgae, as well as occasional
aries and lagoons within the region, from north to south, are
State, are a part of the Oregonian Province, while just south relaxation (August to November), and winter storms (November corpses of fishes, birds, and marine mammals in the intertidal
Estero Americano, Estero de San Antonio, Bolinas Lagoon,
of Point Conception, they belong to the Californian Province. to March), are associated with different degrees of upwelling zone, which provide an unpredictable and patchy source of
Drakes/Limantour Estero, and Elkhorn Slough (National Es-
Although many species have ranges that end at the borders or downwelling. The amount of production in surface waters food. Beach wrack attracts numerous mobile organisms, in-
tuarine Reserve). Estuaries and bays are vulnerable to coastal
of these biogeographic zones under normal conditions, spe- and the ability of organisms to disperse is directly impacted cluding amphipods, isopods, flies, beetles, and shorebirds. The
development, pollution, introduction of invasive species, and
cies of the subtropical Californian Province may occasionally by these processes. In response to these periods, the abun- sporadic deposition of food from the ocean sustains intertidal
commercial and recreational fishing for species that live in
extend their ranges to central and northern California during dance and types of organisms present in a given region change communities in habitats that are subjected to strong physical
near-shore waters. Humans have modified and transformed
unusually warm oceanographic events, such as El Niño and throughout the year. forces and relatively low local production.
about 90% of the wetlands in California. The existence and
the Pacific Decadal Oscillation. Other more localized oceano-
health of these coastal wetlands is critical to the survival of
graphic features, such as eddies, internal waves and bores, NATURAL PERTURBATIONS Marine Ecosystems
organisms that depend on these habitats for survival. One of
are also important factors influencing the distribution and Longer term climatic phenomena influencing the region include: Production in subtidal habitats depends on levels of light and
California’s wetland sites, Bolinas Lagoon, was designated as
abundance of marine species. El Niño, Pacific Decadal Oscillation, and global warming. Off nutrients, and exposure to physical forces. Sufficient light to
internationally important in this role under the Convention on
the coast of California, El Niño events are characterized by support highly productive photosynthetic communities pen-
Wetlands, signed in Ramsar, Iran in 1971.
OCEANOGRAPHIC SEASONS increases in ocean temperature and sea level, enhanced on- etrates surface waters to approximately 30 m. Kelp, which
While certain geological and biological features are evident shore and northward flow, and reduced coastal upwelling of can grow up to 10 cm per day, is among the most productive
Wetlands along the central California coast are sparse, but
along particular regions of the coast, the same oceanographic deep, cold, nutrient-rich water. During this period, survivorship of marine plants. Primary productivity in kelp forests has been
those present support millions of shorebirds and waterfowl dur-
processes and climatic phenomena are operating throughout and reproductive success of planktivorous invertebrates and estimated at 350 to 2,800 grams of carbon per square meter.
7
Section 1: SYNOPSIS OF ECOLOGICAL LINKAGES REPORT
the northeastern Pacific Ocean where production is particularly
(Cottidae) occur in shallow water and tidepools, as well as in
high, approximately 5-15% of the surface production eventually
deeper water around kelp forests, rocky reefs, and sand or mud
reaches the deep sea.
bottoms on the continental shelf. Lingcod (Hexigrammidae),
commonly associated with shallow rocky reefs, also occur in
In a few places, extinct volcanoes or seamounts disrupt the
waters as deep as 300 m.
monotony of the abyssal plain. Off central California, several
seamounts (Gumdrop, Pioneer, Guide, and Davidson) are
Soft bottom habitats on the continental shelf lack the physi-
located near the bottom of the continental slope. Seamounts
cal structure and high biological production of kelp forests
provide physical structures which support complex deep-sea
and rocky reefs. Species that live in soft sediments on the
communities of benthic invertebrates and some fishes.
continental shelf are subjected to shifting sediments through
wave action and near bottom currents. Some species that live
Offshore Islands
in these habitats, such as crustaceans and mollusks, secure
The California coast is a tectonic subduction zone, inhibiting
themselves in tubes and burrows. Other species, such as flat-
the formation of offshore islands. The few that do exist are
fishes, are camouflaged on sandy sediments of the seafloor
extremely important sites for breeding by seabirds and pin-
by their color and shape.
nipeds. The largest offshore islands in the study region are
the Farallon Islands, west of San Francisco. The Farallon
Deep submarine canyons, such as Monterey, Ascension, Pio-
Islands support some of the largest colonies of breeding sea-
neer and Bodega, are remnants of riverbeds that deeply incised
birds south of Alaska. Numerous marine mammals, including
the continental shelf and slope during glacial periods. Because
northern elephant seals, Steller sea lions, harbor seals, and
they cut into the continental shelf, submarine canyons support
fur seals, haul out and breed on the Farallon Islands as well.
deep-sea communities relatively close to shore. Canyon walls
Other important but much smaller breeding populations occur
are often steep and rocky, providing shelter for species, such
on rocks off Point Reyes, Año Nuevo Island, and on rocks off
as rockfishes and thornyheads, which are associated with
Big Sur. Subsurface features, e.g., the Farallon Ridge, Cordell
complex physical structures. Canyon bottoms tend to slope
Bank, and various seamounts, provide substrate and protec-
gently and accumulate finer sediments, such as silt and mud,
tion for diverse communities of benthic invertebrates and some
providing habitat for species such as flatfishes. In addition,
fishes, as well.
the structure of submarine canyons affects the circulation of
near-shore waters and the concentration of organisms in the
BIOGEOGRAPHY
water column.
An understanding of biogeographic patterns and how they are
influenced by ecological linkages enables management deci-
Submarine canyons, submerged volcanoes, and other physical
sions to be placed in a spatial context relative to the distribu-
features under high pressure often concentrate gases and fluids
tion of marine resources. Distributions of marine species are
beneath the sea floor. In some areas, where the sea floor is
determined by oceanographic phenomena, physical tolerances,
weak, these gases and fluids may be forced through the sedi-
and biological interactions. Each species responds to these
ments, creating features known as cold seeps. Most cold seeps
Kelp provides substrate for numerous benthic and epibenthic Productivity from seaweeds can also have indirect effects on
factors in slightly different ways. Despite the physiological and
are found in the deep sea (600-3000 m) under conditions of
invertebrates, as well as food and shelter for many fishes, coastal food webs. Particulate and dissolved organic carbon
ecological differences between species’ response, there are
low light, temperature, and oxygen, and high pressure. In spite
seabirds, and marine mammals. Colonies of bryozoans grow that results from fragmentation and decomposition of kelps
many similarities in species’ distributions, which can be used
of these difficult conditions, numerous organisms are adapted
on kelp fronds. Several species of snails, including purple ring and other seaweeds can be consumed by suspension-feeding
to define biogeographic regions. The transitions between bio-
to life around cold seeps. Vesicomyid clams are the dominant
top snail and blue top snail, feed on kelp, while kelp crabs zooplankton or benthic invertebrates, providing a trophic link
geographic regions are more distinct outside the study area
species at cold seeps off central and northern California. These
cling to the underside of kelp fronds. During periods of low between kelps and higher-level pelagic consumers, such as
than within.
clams support chemoautotrophic bacteria in a symbiotic re-
productivity, sea urchins may emerge from protective crevices fish. A small portion of the drift algae may be transported off
lationship. The bacteria use inorganic chemical compounds
in rocky reefs to graze on kelp. At the surface, floating kelp the reef, where it can contribute to production in submarine
The geographic distributions of numerous marine organisms
released by the cold seeps to produce organic compounds,
masses are important habitats for juvenile fishes, particularly canyons and the deep sea.
of the northeastern Pacific Ocean coincide with major oceano-
which are used by their vesicomyid clam hosts.
rockfishes and kelp surfperch. Schools of blue, black, and kelp
graphic shifts. The biogeographic boundary at the Gulf of Alaska
rockfishes and bocaccio are generally recorded in the midwater Productivity is reduced on rocky reefs below 30 m, where light
occurs at the transition between sea and land along the south
Deep-sea communities depend on the distribution and quantity
kelp canopy. Gopher, copper, black, and yellow rockfishes, levels are low and kelp is unable to flourish. However, the physi-
coast of Alaska. The biogeographic transition at Vancouver
of primary production in surface waters, the rate of movement
lingcod, cabezon, and greenlings tend to associate with the cal structure of rocky reefs does provide shelter for numerous
Island corresponds to the eastern portion of the North Pacific
of organic material to the bottom, and the conditions of deposi-
bottom of the kelp fronds. In addition, the sea otter has been benthic invertebrates and fishes. Shortbelly rockfish (Sebastes
Drift, which bifurcates in this region with part diverted north into
tion and transformation of the organic matter in the sediment.
described as a “keystone species” for its role in structuring kelp jordani), the most abundant rockfish species on the continental
the Gulf of Alaska and part diverted south along the western
A portion of dead organic matter produced in surface waters
forest communities through predation of herbivores, particularly shelf and upper slope off California, are often associated with
coast of North America as the California Current. The biogeo-
is transported to the sea floor either through passive sinking,
sea urchins, resulting in increased kelp growth. rocky reefs between 30-80 m depth. Various seabirds and
graphic transition at Point Conception corresponds to a shift in
or by active transport during vertical migration of plankton. In
marine mammals rely on shortbelly rockfish for food. Sculpin
8
Section 1: SYNOPSIS OF ECOLOGICAL LINKAGES REPORT
the oceanographic regime. At Point Conception, the California results from the biogeographic analyses are interpreted within
coastline turns abruptly east and the cool water moving south in the context of the Ecological Linkages Report as it demon-
the California Current is diverted offshore. The most significant strates the understanding that the patterns presented in the
biogeographic boundary in the study region occurs at Monterey Geographical Information System are dynamic in nature due
Bay; however, other minor boundaries occur around points and to the multitude of factors operating to shape them. Changes in
bays along the coast. any of these factors can result in changes to the biogeography
of the region.
In addition to the changes in latitudinal distributions, the diver-
REFERENCE
sity of species changes with depth. The changes in species
Airamé, S., S. Gaines, and C. Caldow. 2003. Ecological Link-
composition and abundance are associated with physiological
ages: Marine and estuarine ecosystems of central and northern
tolerances for temperature, exposure, light and nutrient input, as
California. NOAA, National Ocean Service. Silver Spring, MD.
well as a wide range of biological interactions among species. At
163 p.
all latitudes, the average number of species of algae and marine
gastropods increased with depth from high to low intertidal and
subtidal zones. In addition, species that occur across several
depth zones are likely to have broader latitudinal distributions
than species that occupy a single depth zone. In contrast to the
patterns observed for marine algae and gastropods, the aver-
age number of fish species declined with latitude and depth.
The greatest numbers of fish species occurred south of 50oN
latitude and shallower than 200 m.
For some species, the range of single individuals spans nearly
the entire geographical distribution of the species. These spe-
cies use local resources during long-distance migration, but no
individual resource supports a resident population. Examples
of these species include baleen whales that feed at highly
productive sites along their migration route, and seabirds that
use estuaries along the coast as resting and feeding sites
during their annual migrations. For other species, the entire
geographical range far exceeds the range of an individual.
Many intertidal invertebrates and fishes have dispersal and
sedentary phases during their life cycles. Examples of these
species include barnacles, mussels, and clams that settle into
intertidal habitats, and rockfish that settle into kelp forests or
rocky reefs after a pelagic larval stage.
CONCLUSIONS
Within the study region there are many distinct ecosystems
each hosting a unique assemblage of organisms. In addition
to describing these key ecosystems and species in the region,
the Ecological Linkages Report provides information on link-
ages within and between these systems. By understanding
the climatic, oceanographic, physical and biological influences
operating together to shape the regional biogeography, the
background exists for the biogeographic analyses to be inter-
preted. The "Biogeographic Analyses" section complements the
synthesis of literature in the Ecological Linkages Report with a
data driven look into the biogeographic patterns evident around
the sanctuaries. The analyses provide a spatially explicit view
of marine resources within the study area from which manage-
Annie Crawley
ment decisions can be better enacted. It is important that the
9
Section 2: BIOGEOGRAPHIC ANALYSES
INTRODUCTION species within a group, such as by family (e.g. rockfish), and
The biogeographic analyses component is the cornerstone of comparison of spatial patterns between groups. Analysis of
Integrated Analyses
Analyses
the overall assessment to support the joint management plan spatial patterns resulted in information on the relationships
Data Layers
revision process. The data, analyses, and supporting informa-
and Products to between individual species, between assemblages of species,
and Products
tion are linked using statistical and GIS tools to visualize the and of the relationship of species to specific environmental
Aid Sanctuary
location of significant biological areas or “hot spots.” There were and habitat parameters. Furthermore, the compilation and
many different ways to analyze and organize the biological integration of individual species maps were used to calculate
Management
data compiled for this assessment. To efficiently support the community metrics, such as total richness or diversity of fish
management plan revision process, only a limited number of and marine bird species, at a specific location.
Study Area
analytical options were selected based on reviewer’s comments
on the Interim Product, mission of the NMSP, technical review To define species assemblages, multivariate techniques were
tem
meetings, and peer review workshops. These key analyses are applied to various data sets to group organisms found at spe-
s
Catch Data Species Distributions
ion Sy
presented in this document and on the CD-ROM. In addition to cific sampling sites. The assemblage analyses defined species
these results, spatial data and information on the companion groups across the study area. By visualizing the assemblages
CD-ROM enable NMSP staff, advisory councils, and research geographically, areas of overlap became apparent and group
t
forma
Sightings
partners to conduct additional analyses not specifically ad- habitat affinities, such as depth range, were delineated.
dressed in this product.
n
Species habitat suitability index modeling (HSI) studies were
phic I
A critical step in the biogeographic analyses component was the undertaken for 20 fish and invertebrate species in an attempt
Community Distributions
Bathymetry
extensive effort to have data, analytical approaches, and results to characterize areas within the study region that suffered a
eogra
peer reviewed. Initial results from the suite of analyses were lack of sampling data, particularly in near-shore habitats. The
presented to experts on marine ecosystems of north/central integration of HSI models into a GIS provides a spatial depic-
ated G
Substrate
California, as well as to the originators of the data sources in tion of species habitat suitability models for individual species
an attempt to improve the analyses. The role of expert review by integrating information on species habitat affinities and the
and input has been considerable, and the contributions made distribution of those habitats in space and time (Brown et al.,
Integr
Temperature
by experts have significantly enhanced the analyses. 2002; Monaco and Christensen 1997). The modeling compo-
Modeled Distributions nent of the biogeographic analyses is a necessary step due to
ASSESSMENT PROCESS the incomplete distribution of sampling data across the entire
Life History
To aid in focusing on the most important analyses, the biogeo- study area. Thus, species that were representative of the as-
graphic assessment process displayed in Figure 6 was utilized. semblages described above and/or other key species were
Data Significant Biological
This process is currently being implemented through a joint selected for modeling their potential distribution. The composite
Areas
NMSP and NCCOS Biogeography Team effort to initiate bio- set of species habitat suitability models contributed to defining
geographic assessments across all sanctuaries within the next significant biological areas within the region.
Ecological Linkages Report
five years. The process is organized around development of
biogeographical data layers, integrated analyses, and specific Measures of community structure for fishes and marine birds
Figure 6. Biogeographic assessment process.
products to aid in sanctuary management plans (Kendall and were calculated independently by species group, compared,
Monaco 2003). Thus, the integration of partner’s comments and, where applicable, integrated. Convergence of overlapping
and use of the biogeographic assessment process resulted in spatial patterns defined significant biological areas based on
To develop this capability, a suite of analyses were conducted
species richness statistics), to be directly linked to specific
the analyses and results presented in this document. a number of criteria (e.g., high species abundance, high spe-
that were most appropriate in addressing NMSP natural re-
areas or habitats they correspond to across the study area.
cies diversity).
source management issues. The biogeographic assessment
The GIS also facilitated integration of multiple data types and
Biogeographic data assembled for this project were derived framework (Figure 6) aided in targeting the suite of analytical
sources into a common spatial and temporal framework (Gill
from many sources (see section 4), including NOAA Fisher- Thus, the biogeographic analyses component was a result of
approaches to define biologically significant areas in support
et al., 2001). The following suite of map products quantitatively
ies, academia, state government, and data housed within the interpreting or visualizing the analytical results from statistical
of the sanctuary management plan reviews. Categories of
defined significant biological areas that are within or adjacent
NMS and Biogeography Programs. The biogeographic data analyses, ecological modeling, and integration of results across
analysis include: temporal and spatial analysis of individual
to existing boundaries of Cordell Bank, Gulf of the Farallones,
cut across various themes, such as species distributions and biota and habitats. The cumulative results aided in assess-
species’ distributions, species assemblage analyses, habitat
and Monterey Bay National Marine Sanctuaries. The GIS-based
habitats, and are integrated using a common spatial framework ing the biogeographic patterns in the study with regard to the
suitability modeling, and community metrics within and across
products are intended to aid in evaluating current sanctuary
in a GIS. The GIS enables a user to select particular data layers distribution of individual species, species assemblages, and
species groups. Important individual species maps were de-
boundaries relative to biological resources and habitats (Fig-
to be displayed, combined, and manipulated in a wide variety species habitat utilization patterns.
veloped from a number of data sets to visualize species pres-
ure 7), explore options for environmental protection of existing
of ways to achieve specific analytical objectives (Figure 6). ence and/or abundance data within the study area by season.
NMS areas, identify additional biologically important areas, and
REFERENCES
Where possible, well-established breeding colonies, rookeries,
evaluate alternative management strategies.
The use of the GIS enabled species-specific data, such as Brown, S.K., K.R. Buja, S.H. Jury, M.E. Monaco, and A. Banner.
and high concentrations of species are displayed on the digital
distribution and abundance data or community metrics (e.g., 2000. Habitat suitability index models for eight fish and inver-
maps. The single species maps enabled various groupings of
10
Section 2: BIOGEOGRAPHIC ANALYSES
tebrate species in Casco and Sheepscot Bays, Maine. North
American Journal of Fisheries Management 20:408-435.
Gill, T.A., M.E. Monaco, S.K. Brown, and S.P. Orlando. 2001.
Three GIS tools for assessing or predicting distributions of
species, habitats, and impacts: CORA, HSM, and CA&DS. In:
Proceedings of the first international symposium on geographic
information systems (GIS) in fishery science. Nishida, T., Kai-
lola, P. and Hollingworth, C.E. (Eds.). pp 404-415.
Kendall, M.S. and M.E. Monaco. 2003. Biogeography of the
National Marine Sanctuaries: A partnership between the NOS
Biogeography Program and the National Marine Sanctuary
Program. January 2003. 8pp.
Monaco, M.E. and J.D. Christensen. 1997. Biogeography Pro-
gram: Coupling species distributions and habitat. Pages 133-
139 in G.W. Boehlert and J.D. Schumacher, editors. Changing
oceans and changing fisheries: Environmental data for fisheries
research and management. National Marine Fisheries Service
Technical Memorandum NOAA-TM-NMFS-SWRSC-239, Pa-
cific Grove, California.
Figure 7. 3-D image of bathymetric relief within and adjacent to the sanctuaries.
11
Section 2.1: BIOGEOGRAPHY OF Approach to Fish Analyses
Biogeographic FISHES
INTRODUCTION trawls on the continental shelf and slope or de-
The biogeography of fishes section is a robust rived from species habitat affinities described in
statistical analysis of fish and a few economically the literature. Model results were validated with
1. Assemblage Analyses
important macro-invertebrates. A two-pronged ap- NMFS trawls on the shelf and slope (a different
proach was conducted to examine both fisheries- subset than that used to create the affinities) or
dependent and independent catch data and model CDF&G recreational catch data. Model results
Results:
Data Input: Analytical Approach:
the potential distribution and relative abundance of for 3 species are included in the main body of
Define and Interpret
NMFS shelf Clustering of species
selected species (Figure 8). Analysis of fisheries this document, with the models for the rest of
data can be a slow process which often requires the species included in the CD-ROM.
Biological Hot spots.
NMFS slope assemblages, site groups,
extensive exploratory statistical techniques to in-
CDF&G recreational species diversity, and (e.g. assemblages of
crease our understanding of the data before pre- The integration of the fish and invertebrate
NMFS midwater richness. station/species,
senting reliable and salient results. Many data sets analyses is shown in Section 3. For example,
diversity, richness)
were evaluated, however, only four data sets that a comparison of important areas derived from
were spatially comprehensive within the study area the diversity of NMFS trawls and those derived
were analyzed: 1) the California Department of Fish from overlays of the HSM’s was conducted to
Overlay, compare, and
and Game fishery dependent recreational fishing define significant biological areas. In addition,
combine with marine
trips targeting rockfish (CDF&G recreational); 2) the aspects of the Ecological Linkages Report were
bird results.
National Marine Fisheries Service fishery indepen- qualitatively incorporated into the assemblage
dent benthic trawls on the continental shelf (NMFS and HSM discussions to aid in interpretation of
Integrate into final
shelf trawls); 3) NMFS fishery independent benthic these analyses.
report.
trawls on the slope (NMFS slope trawls); and 4)
the NMFS fishery independent trawls in midwater
2. Habitat Suitability Modeling
(NMFS midwater trawls). Detailed information on
these trawls is given in each respective section.
Analytical Approach: Results:
Data Input:
The NMFS trawls on the continental shelf and slope
Literature review
provide information on the diverse demersal fish as- Distribution maps for Define and Interpret
NMFS trawl data
semblages found on trawlable habitats between 50
representative species Biological Hot spots.
GIS habitat layers
and 1280 meters depth throughout the study area.
based on habitat (e.g. overlay maps of
Key species used:
Pelagic fish encountered either as the trawl de-
suitability indices. modeled species to
scended or ascended are also included with these Representatives of local
determine locations of
assemblages, or species
analyses. The CDF&G recreational hook and line Statistical validation with
co-occurrence)
of economical/ecological
data complements the NMFS data sets by provid- catch data.
importance.
ing information on midwater as well as demersal
species. The recreational data was collected over
soft bottom and hard bottom habitats between
5 and 200 meters depth. The NMFS midwater
trawl data targets juvenile rockfish and provides Figure 8. Biogeographic approach to fish analysis.
information on fish and invertebrates found in the
neritic environment during the upwelling season.
Even though the spatial extent of this data set does not cover all four data sets, 119 species were analyzed. A complete list of fish and invertebrate species across the study area (Brown
the entire study area, it provides a source of information on of species included in the assemblage analyses can be found et al., 2000). Thirty two species of fish and invertebrates were
the neritic environment, which is important as juvenile habitat, on the CD-ROM. In addition, community metrics, including initially investigated through literature searches to determine if
and as the base of the food web for marine birds, fish, and species richness, species diversity, and rockfish richness sufficient information was available to model potential distribu-
marine mammals. were calculated for the NMFS shelf and slope trawls and tions. The process of determining which species to include in
the results presented spatially using GIS. However, due to the modeling procedure included consultation by the sanctu-
Due to time constraints, analysis of all species individually spatial and temperal limitations of available data, (i.e. NMFS ary staff, and integration of information on the economic and
was not feasible. Instead, all four data sets were analyzed demersal trawls have no information on rocky or shallow areas ecological importance of the species as well as initial results on
using multivariate statistics to identify species assemblages, (<50 meter) and contain trawls only for the months of June species assemblages. Of the 32 species initially investigated,
site groups, and the location of the species assemblages in through November) areas such as Cordell Bank and the Far- there was sufficient information available to conduct HSM on
space using a Geographic Information System (GIS). For the allon Islands are not adequately sampled. Therefore, habitat the adult and subadult stages of 14 fish species, and adult stage
multivariate statistics, species were included in an analysis if suitability modeling (HSM) was conducted to supplement the of 4 fish and 2 invertebrate species. Habitat suitability models
they were captured in at least 5% of the collections. Through analysis of catch data, and to model the potential distribution were either derived from an analysis of a portion of the NMFS
12
Subsection 2.1.1: ASSEMBLAGE ANALYSES
in content and spatial coverage was that the rockfish management groups could be defined,
INTRODUCTION at the effect these trawls had on the
Mutivariate Analyses of Fisheries Dependent
were not utilized in this analysis, and that both depth and latitude were important.
No species exists in isolation from other species or their envi- estimates of biomass of selected
and Independent Data
but were used to help interpret
ronment. Monitoring species individually may cause managers species through time. Based on
the results. Even though the neritic zone is ecologically important, little re-
to miss important interactions (Chavez et al., 2003; Worm and Zimmerman’s results, we excluded
search has addressed the midwater environment. Cailliet et al.
Myers, 2003; Baraff and Loughlin, 2000; Estes et al., 1998). these abnormal “water hauls” from
Five specific objectives of the (1979) described fish and invertebrate species co-occurrences
In addition, individual species' abundance may be considered our analyses. Williams and Ralston
assemblage analysis (all of in anchovy purse seines, and midwater trawls. An extremely
sustainable in the context of a fishery, but still be low enough (2002) analyzed data from NMFS
which aim to increase our under- thorough report by Larson et al. (1994) looked at the NMFS
to influence ecosystem dynamics and health (NMFS 2001). shelf trawls to determine rockfish
standing of the biogeography of midwater trawl results in conjunction with local environmental
There has been a growing recognition that effort needs to be species assemblages. The same
fishes and macro-invertebrates conditions to determine juvenile rockfish assemblages. Their
extended toward understanding the entire ecosystem. The data were used in this analysis;
in relationship to their environ- results emphasize the ephemeral nature of the pelagic environ-
National Marine Sanctuary Program is tasked with ensur- however, the multivariate statisti-
ment, as well as identify impor- ment, but they were able to document two consistent spatial
ing the continued health of the ecosystems contained in the cal method utilized, the spatial
tant areas or habitats within the trends: 1) the rockfish are larger inshore than offshore, and 2)
sanctuaries. However, important species-species interactions coverage employed, and the spe-
study area), were as follows: there was a north/south gradient in species composition and
and species-habitat interactions are still not well understood. cies examined were different. The
1. Identifying spatial patterns abundance. Moser et al (2000) described changes in rockfish
Abiotic (e.g., habitat preferences toward depth or sediment) overall conclusion from Williams and
and hot spots in community larvae abundance in CalCOFI plankton tows from 1951 to 1998
and biotic (e.g., presence or absence of prey, predators) fac- Ralston was that rockfish richness
metrics (diversity and richness); in response to adult biomass and environmental conditions.
tors can impact the importance of an area to fish. Elucidating was highest at a depth of 200-250 meters, where the shelf
2. Determining which species tended to be caught together He concluded that over-fishing as well as decadal shifts in
habitat characteristics that are most important to animals, and and slope meet, and that depth and latitude were the main
(species assemblages); environmental conditions were affecting the stocks.
understanding the co-occurrence of species, is a first step in determinants of rockfish assemblages. Jay (1996) analyzed
3. Analyzing fishing locations to determine which locations
determining areas that should be managed as “essential” habi- the 1977-1992 NMFS shelf trawls to determine site groups
contained similar catches (site groups); Underwater submersibles have been used to describe fish as-
tats. This study aids in clarifying the interactions among species that contained similar catches. Using 33 species of fish, he
4. Resolving where the species assemblages were being semblages and their interaction with habitat at spatial scales
and between broad scale habitat characteristics and species identified 23 site groups, many of which contained the same
caught by combining results from objectives 2 and 3 and relevant to the fish themselves. Yoklavich et al. (2000 and
on the scale of the commercial and recreational fisheries. Even species, but with different relative abundance. Even though
then utilizing GIS to map the results; and 2002) surveyed Soquel Canyon and Big Creek Ecological
though these data sets were originally deployed to collect infor- depth and latitude showed some influence on site groups,
5. Identifying significant relationships between site groups Reserve on the Big Sur coast, Field et al. (2002) looked at
mation necessary for setting fishing limits, these data sets can overall he found little association between the site groups and
identified in objective 3 and broad scale habitat character- Big Creek Ecological Reserve, while Hixon et al. (1991) and
provide preliminary information on multi-species interactions. a suite of environmental parameters.
istics (bathymetry, bathymetric complexity, and large-scale Hixon and Tissot (1992) researched Haceta, Coquille, Daisy,
Recreational hook and line drifts covering approximately one
habitat classification). and Stonewall Banks off the Pacific northwest. These results
kilometer, demersal trawls on the continental shelf and slope Gabriel and Tyler (1980) used data from the Oregon Depart-
are very important to managers because they show fish and
covering one kilometer, and fifteen minute trawls in midwater, ment of Fish and Wildlife Trawl Survey and the West Coast Joint
Community metrics can be used to determine which areas are habitat interactions on very small scales. However, many of the
were analyzed to determine species assemblages, site group- Agency Rockfish Survey to look for site groups from California
important to multiple species. Experts in California have ac- results from these studies are not comparable with the current
ings, and the interaction between species and locations. In to Alaska. They differentiated three large site groups: “inter-
knowledged a management need to increase our understanding studies due to large differences in scale. Hixon et al. (1991)
addition, analyses were completed to determine larger scale mediate” at less than 145 meters, “deep” between 145 and
of fish species interactions (objective 2) (Starr, 1998) and the documented that the species composition observed from the
environmental variables that were significantly different among 200 meters, and “slope” greater than 200 meters deep. They
interactions between fish assemblages and habitat (objective submersibles was different than that seen in trawls. The results
identified groups. Due to limitations of the data sets (section found that site groups were “strongly associated with depth
5) (Starr, 1998; Yoklavich et al., 2000, 2002). Studies exist that from these studies reveal the importance of habitat, especially
4), and the lack of results on individual species’ distributions, contours”. Matthews and Richards (1991) compared gill net
identify either species assemblages or site groups (see below), rugosity, to fish species composition.
habitat suitability models (section 2.1.2) for selected species catches from trawlable and untrawlable areas to determine if
but so far none have integrated multiple data sets, provided
were completed to complement this analysis. untrawlable areas could be considered de-facto fish reserves.
the interaction between species assemblages and site groups, Substantial declines in the standing stock biomass of eco-
Even though some species overlapped, they concluded that the
nor presented spatially explicit results. The results of these nomically important rockfish species (Ralston, 1998) prompted
The primary objective of the assemblage analysis is to define species assemblages were significantly different; suggesting
analyses aid in defining the region's biogeography based on NMFS to organize a symposium to discuss the implications of
spatial biogeographic patterns of fishes and macro-inverte- that species assemblages determined from trawls cannot be
the spatial pattern of fishes and macroinvertebrates. no-take areas for rockfish in September of 1997. Eleven plenary
brates within the study area from Point Arena to Point Sal in extrapolated to non-trawlable habitats.
papers and six case studies are available online, and cover a
California. The study is based on a synthesis of four primary
REVIEW OF RELEVANT LITERATURE range of topics. Starr (1998) provided a thorough evaluation
databases of fish and invertebrates that were spatially compre- Only a few studies have analyzed recreational hook and line
Due to the economic importance of recreational and commercial of the potential of rockfish no-take reserves. He expressed
hensive throughout the study area including: 1) the California data. For a general analysis of a species’ specific decline
fisheries in California, several studies have been completed a management need for the identification of species assem-
Department of Fish and Game fishery dependent recreational in recreational catch see Love et al. (1998), Mason (1998),
that look at species co-occurrences or species interactions blages. Once assemblages are identified, management can
fishing trips targeting rockfish (CDF&G recreational); 2) the or Wilson-Vandenberg et al. (1996). Mason (1995) analyzed
with their environment. NMFS publishes yearly reports on the address actions for adequate protection of each species as-
National Marine Fisheries Service fishery independent benthic various CDF&G recreational fishing surveys and documented
status of demersal fish species by analyzing results from their semblage. Starr also suggested protecting rectangular areas
trawls on the continental shelf (NMFS shelf trawls); 3) NMFS trends in effort, fishing location, and species catch. She docu-
shelf and slope trawls (Turk et al., 2001; Weinberg et al., 2002; that cover 20-50 km of the coast and extend west to the edge
fishery independent benthic trawls on the slope (NMFS slope mented two principal rockfish assemblages and distinguished
Lauth, 2001; Shaw et al, 2000). Zimmerman et al. (2001) looked of the continental shelf.
trawls); and 4) the NMFS fishery independent trawls in midwater them by depth (less than 70 meters and greater than 70 me-
at the biomass of demersal species to determine NMFS shelf
(NMFS midwater trawls). Detailed information on these surveys ters). Sullivan (1995) used the CDF&G recreational fishing data
trawls that did not fish the bottom as intended. He then looked
is given in each respective section. Databases that were limited (1987-1992) to determine site groups. His overall conclusion
13
Subsection 2.1.1: ASSEMBLAGE ANALYSES
Within hierarchical clustering, the choice of resemblance met- the reasons behind those decisions are provided within the Only the Ward’s minimum variance (it calculates an internal
INTRODUCTION TO CLUSTERING
ric (a formula that determines how similar two things are) and following detailed methodology section. dissimilarity matrix) and the 1-Pearson correlation with aver-
The use of multivariate analyses is gaining popularity in
clustering model (a method that groups things based on their age linkage consistently produced results without chaining,
ecology and fisheries management (McGarigal et al., 2000;
resemblance metrics) can also have an influence by adjusting DETAILED METHODOLOGY and were therefore interpretable. The Pearson method was
Paukert and Wittig, 2002), but can be confusing due to the
the importance of abundant and rare species and changing To reduce the amount of material that is repeated with each sec- chosen over Ward’s primarily because the results were very
availability of many statistical techniques. Therefore, this sec-
the importance of zero values (Gauch, 1982; Boesch, 1977; tion, a detailed methodology for all four data sets is presented similar to the results from principal component analysis (both
tion provides a basic introduction to the principles of clustering,
McGarigal et al., 2000). Common resemblance metrics used in here, with only a brief overview provided in each section. Figure methods are based on a correlation matrix), but also because
one form of multivariate analysis. Interested readers should
ecology are the Bray-Curtis, Euclidean distance, Jaccard, and 9 is an example of the clustering methods using hypothetical the results showed spatial patterns when mapped.
reference Boesch (1977), Gauch (1982), or McGarigal et al.,
Pearson correlation coefficients. Common clustering models data. The first objective of this study was to look for patterns in
(2000) for more detailed information on clustering than will be
include Ward’s minimum variance, average linkage, centroid community metrics. For these analyses, all unknown species, To determine which fish species tended to be caught together
provided in this brief introduction. Clustering is “a technique
linkage, and single linkage. Because of chaining, not all outputs and those without an abundance estimate, were eliminated. (species assemblages; objective 2), an index of dissimilarity
for optimal grouping of entities according to the resemblance
from cluster analyses can be utilized. When chaining occurs, The Shannon index (H’) was calculated for each NMFS shelf between species was calculated as 1-Pearson correlation coef-
of their attributes as expressed by given criteria” (Boesch,
entities fuse to a few nuclear groups one at a time rather than and slope trawl based on the following equation: ficient, with the resulting matrix of species dissimilarities clus-
1977) or, in short, a method that places things (sites, species,
forming new groups, and make it impossible to divide the data tered by using average means as discussed above. Changes in
etc.) into groups. Clustering uses statistics to determine these
S
n n
into meaningful smaller groups (Boesch, 1977). fishing depth or the abundance of target species through time
groups, but the method also possesses aspects that are sub-
H ′ = − ∑ i ln i
could influence the results. Therefore, a second analysis was
jective and require an understanding of the ecosystem being i =1 n n
Hierarchical clustering results in a tree diagram, called a den- run using only current data (1993+ for recreational, 1989+ for
analyzed. There are five steps to clustering, which all have
drogram, which shows the linkages between all of the entities Where S is the number of species, ni is the number of individu- shelf trawls), so that a comparison of species associations from
some aspect of subjectivity (adapted from Sullivan, 1995): 1)
and groups at different levels of similarity. Various objective als found of species i, and n is the total number of individuals all the data and current data could be completed. The determi-
the choice of data, including what data to include or remove,
methods, such as scree plots, are used to determine what level (Shannon and Weaver, 1949). Richness (total number of fish nation of what should be considered “current” was based on the
and how to transform and standardize that data; 2) the choice
of similarity is important or how many groups are created. A species caught) was also enumerated for each NMFS shelf and expert opinion of the scientists who collected the data (pers.
of a resemblance metric (can be based on similarity or dis-
scree plot shows the dissimilarity values plotted against the slope trawl. In addition, the rockfish richness was calculated by comm. Deb Wilson-Vandenberg and Mark Wilkins) according
similarity); 3) the choice of clustering model; 4) the choice of
number of clusters such that breaks in the level of dissimilarity counting the number of Sebastes and Sebastolobus species to known shifts in effort or species abundance. Differences in
number of groups (or level of similarity) and; 5) the choice of
are revealed through the shape of the curve (McGarigal et al., caught in each trawl. assemblage groups between the entire data set and current
whether or not to reassign objects to more appropriate groups.
2000). Experts recommend combining this objectivity with eco- data are noted.
The following is a more detailed description of each of these
logical knowledge to determine clustering results that are eco- Due to the overlap in trawls between years (boats often returned
five steps.
logically meaningful (Boesch, 1977). The final decision when to approximately the same area each year), it was sometimes To determine which locations contained similar fish catches
clustering is whether to reclassify entities into more appropriate difficult to determine spatial trends in diversity. Therefore, mean (site groups), the 1-Pearson correlation clustering method, as
The choice of data to include in the analyses can influence
groups based on some identified criterion. This step has created diversity and mean richness were calculated for 5’ grid cells explained above, was used again, but this time to cluster sites
results in many ways. Rare species are usually removed from
controversy because of the subjectivity introduced. (approximately 5 km by 9 km cells, although actual dimensions with similar catches. No secondary analyses with a random
analyses because they can have a disproportionate influence
vary with latitude) throughout the study area. The 5’ grid cell subset or current data were run because the influence of such
on the resulting clusters. What is considered “rare” can vary
For comparison, a brief introduction to another widely used form size was utilized because it was employed in other analyses parameters could be inferred by looking at the date of each
for analyses, but typically species found in less than 5% of the
multivariate analysis, the Principal Component Analysis (PCA), (bird, mammal, and environmental maps) and facilitates the trawl in each group.
trawls/catches are removed (Gauch, 1982). Additionally, the
is provided. PCA reduces the dimensionality of data (Gauch, comparison of results among these analyses.
transformation and standardization of the data may affect the
1982, McGarigal et al., 2000). One difference between the In order to decide how many groups to keep, statistical methods
influence of rare and abundant species. Binary data (presence/
results from PCA and hierarchical clustering is that in cluster- Both clustering analyses (objectives 2 and 3) began with either (scree plots) were employed to determine where breaks in the
absence) weights all species the same, and reduces the influ-
ing each species is ultimately assigned to one and only one a site by species or species by site matrix of presence/absence similarity level occurred, then group composition was analyzed
ence of abundant species while increasing the influence of rare
group. In PCA, only species with a certain loading (i.e. level or log adjusted abundance. All species that were present in to determine the best ecological groupings (i.e. if smaller or
species (Boesch, 1977). Transformations are needed because
of influence) are included in a group. This means that in PCA at least 5% of the trawls were included in this analysis. This larger groups would provide a better ecological explanation).
most ecological data do not conform to the assumptions of
some species are never assigned to a component, and others number was chosen because it is a commonly used method in Expert opinion on ecologically relevant groups was solicited
a normal distribution and homogeneity of variances required
are assigned to more than one. This can be a drawback be- fisheries management (Gauch, 1982), and because it reduced at review meetings held in Seattle, San Francisco, and Mon-
for parametric analyses (Boesch, 1977). Transformations
cause often very important species that are found everywhere the number of zero values in the data set while keeping an terey in October, 2001. No reclassifications were completed
decrease the variation between abundant and rare species,
are never attributed to any group. In clustering methods, since adequate number of species for analysis. Since the raw abun- in this study. Instead, a modified bootstrapping procedure was
thereby reducing the influence of abundant species.
every fish must be placed into a group, species which are found dance data did not conform to assumptions of a normal distribu- implemented. Fifty random samples of one-half or three-quar-
everywhere can be grouped together. At the same time, spe- tion and homogeneity of variances, either log transformations ters of the data were extracted and run through the clustering
The clustering method dictates the way that groups are formed.
cies which are only marginally associated with a group can be (if effort was available) or presence/absence (if no effort was process. The amount of data included in the random analyses
The two classes of clustering are: 1) hierarchical, which show
added to a group. provided) were used. No standardization was completed be- depended on the size of the original data set; if the data set
the relationship between groups in dendrograms, and 2) non-
cause correlation coefficients were used. Exploratory analyses was too small, samples consisting of half of the data often had
heirarchical, which do not. Hierarchical clustering may be bi-
For this study, an extensive exploration was conducted on the were run to investigate multiple resemblance metrics (including zero catch for some species, creating error messages when
ased because the researcher can choose the results that best
data using multiple data transformation, similarity metrics, and Euclidean Distance, Jaccard, and Pearson correlation matrices) running the analysis. The results from the random samples were
match expectations (Williams and Ralston, 2002); however,
clustering models. Ultimately, one method was chosen, and and multiple clustering models (including Ward’s minimum vari- used to determine the stability of the species assemblages in
it is advantageous because it clearly shows the relationship
ance, average linkage, centroid linkage, and single linkage). the given data set.
between the resulting groups.
14
Subsection 2.1.1: ASSEMBLAGE ANALYSES
In order to combine these two analyses and resolve where the
Determining species assemblages, site groups, and their interaction
fish assemblages were being caught (objective 4), the average
frequency of occurrence for species assemblages was calcu-
(example with hypothetical data)
lated for each site group to determine the overlap between
the site and species groups. By looking at the frequency of
spiny
occurrence of species assemblages in each site group, it was Site1 Site2 Site3 Site4 Site5
bocaccio chilipepper widow rf Rex sole Dover sole
dogfish
possible to determine which species groups were influential Start with site by species or bocaccio 23 66 0 2 1
Site1 23 11 15 1 0 0 chilipepper 11 5 7 0 0
in forming the site groups. Species groups were considered
species by site matrices of
Site2 66 5 47 0 0 4 widow rf 15 47 0 0 0
influential if, on average, species were present in 25% of the
abundances
Site3 0 7 0 55 43 0 Rex sole 1 0 55 0 23
trawls. Since rare species had low frequency of occurrence for
Site4 2 0 0 0 0 44 Dover sole 0 0 43 0 21
all site groups, 25% is a reasonable number when rare and spiny dogfish 0 4 0 44 0
Site5 1 0 0 23 21 0
abundant species are averaged. Spatial distribution of the site
groups was determined by mapping the site groups in GIS.
bocaccio chilipepper widow rf Rex sole Dover sole spiny dogfish
For management purposes, it is important to understand which Site1 Site2 Site3 Site4 Site5
bocaccio 1.00
environmental characteristics influence species distributions. Site1 1.00
Calculate Pearson
chilipepper 0.30 1.00
Analyses of variance (ANOVA) were conducted to determine Site2 0.92 1.00
widow rf 1.00 0.27 1.00
if there were significant differences in bathymetry, bathymetric correlation coefficients Site3 -0.66 -0.61 1.00
Rex sole -0.52 0.07 -0.53 1.00
complexity, and gross sediment type between the site groups
between species or sites Site4 -0.38 -0.22 -0.35 1.00
Dover sole -0.53 0.02 -0.54 1.00 1.00
at the scale of this analysis. Bottom depth was measured in the
Site5 -0.65 -0.56 0.99 -0.30 1.00
spiny dogfish -0.29 -0.57 -0.25 -0.35 -0.35 1.00
field for each of the four data sets. Using ArcView, individual
trawl locations were overlaid on the sediment (pp. 38) and
bathymetric complexity (pp. 16) maps and the underlying pa-
rameters extracted. For the midwater trawl, other environmental
conditions measured in the field, such as water temperature, Run average means cluster Site 1
spiny dogfish
salinity, and density, were also tested.
analyses on correlations to
widow rf
Site 2
determine species
bocaccio
Bathymetry appeared to be the overriding influence in de- Site 4
assemblages and site
termining fish assemblages. Attempts were made to statisti- chilipepper
Site 3
groups. Determine
cally remove the influence of bathymetry from the data and Dover sole
then re-analyze the data for assemblage patterns caused by appropriate number of
Rex sole Site 5
secondary influences. However, two general problems were
groups.
encountered. First, the standard statistical procedure to re-
move the influence of bathymetry required a linear relationship
between species abundance and bathymetry. Unfortunately,
the relationship between species abundance and bathymetry Calculate frequency of occurrence of species assemblages in site groups.
was non-linear even after various transformations were tried.
Determine which species groups were influential in forming the site groups
Experiments with spline-fitting were also unsuccessful. A major
site 1, 2 site 4 site 3,5
problem was the presence of zero species abundance values
for those depths the species assemblages were not present.
Rockfish 100 33 33
Secondly, the species abundance data were collected over
Sp. Dogfish 50 100 0
narrow ranges of other influences, such as bathymetric com-
plexity and substrate/sediment size. Again, the problems of Sole 25 0 100
non-linearity and zero species abundance prevented further
conventional statistical analyses. Figure 9. Hypothetical example of the methods used to determine species assemblages, site groups, and the interaction between species assemblages and site groups.
15
Subsection 2.1.1: ASSEMBLAGE ANALYSES
ABOUT THIS MAP To determine the importance of bathymetric complexity to the
124°W 123°W 122°W 121°W
Figure 10 displays bathymetric complexity derived from high formation of fish abundance, an analysis of variance was run
39°N
39°N
resolution bathymetry. Bathymetric complexity is calculated for to test for significant differences in bathymetric complexity
Bathymetric Complexity each cell as the standard deviation of depth for all grid cells between site groups for CDF&G recreational hook and line,
within a 1 kilometer radius. The range of resulting bathymetric NMFS shelf trawl, and NMFS slope trawl catches (see individual
complexity is large, and the majority of cells show a low vari- sections for results).
Calculated for a 1 km radius ance. Therefore, in order to visualize differences, results have
to be displayed as standard deviations above and below the
around each grid cell mean. The areas in blue are relatively flat with little slope, and
the darker red shows the highest variance. Results highlight the
edge between the shelf and slope areas, and create a dramatic
visual for the canyons and seamounts.
Departure from Mean
38°N
38°N
DATA SOURCES
-1 - 0 Std. Dev.
Results were calculated from 3 arc second (nominally 70 x
0 - 1 Std. Dev. 70 meters) bathymetry derived from NGDC and MBARI data
sources. All available multibeam points were used in the area.
1 - 2 Std. Dev.
Hydrographic survey data (echo sounder data) was eliminated
2 - 3 Std. Dev. from the interpolation if it overlapped with multibeam data. Verti-
cal and horizontal correction was performed on all data prior to
>3 Std. Dev.
incorporating it into the data set. All data were triangulated and
gridded using "The Vertical Mapper" extension with MapInfo
0 10 20 40 60 80
6.5. Cell size varies depending on the available data for each
37°N
37°N
Kilometers
area, with a minimum cell size of 70 x 70 meters.
METHODS
Bathymetric complexity was calculated using the "neighborhood
statistics" option in Arcview 3.2. Arcview computes a standard
deviation from all grid cells within a 1 kilometer radius around
each cell. The results are displayed as standard deviations
from the mean as this scale provides the best resolution for
visualizing the location of high slope areas.
36°N
36°N
RESULTS AND DISCUSSION
Fish species, especially some rockfish species, have a very
strong affinity to areas with a high relief (Yoklavich et al., 2000,
2002; Hixon et al., 1991; Hixon and Tissot, 1992; Field et al.,
2002; Starr, 1998; and Williams and Ralston, 2002). Calculating
the variability in bathymetry for a given area can provide a rough
estimate of bottom rugosity on a scale of km. Smaller pinnacles
may not be distinguished at this scale, but the large physical
characteristics, such as the edge between the continental shelf
and slope, canyons, and seamounts, will be displayed. The
35°N
35°N
variable depth of the continental shelf break can be estimated
using these maps. North of Cordell Bank NMS, the break occurs
around 300 meters, within Cordell Bank and Gulf of the Faral-
50
50 m
20 lones NMS, it is around 200 meters depth, north of Monterey
100
00
20
0 m
m
0
0
Bay it becomes shallower at 150 meters, and inside Monterey
m
m
Bay and to the south, the break is as shallow as 100 meters.
124°W 123°W 122°W 121°W
Figure 10. Standard deviation of bathymetry calculated for a 1 km radius around each cell. Results are presented in standard
deviations above or below the mean.
16
Subsection 2.1.1: ASSEMBLAGE ANALYSES
in measurements between cells and facilitates comparisons
ABOUT THESE MAPS
124°W 123°W 122°W 121°W
between maps. Since the placement of the grids is arbitrary, the
Species richness of demersal fish was calculated for NMFS
Species Richness of
results will in-part depend on where the grid falls. An analysis
39°N
39°N
fish trawls (shelf and slope) conducted at depths between 50
comparing three different grid placements was conducted. It
and 1280 meters. The mean number of fish species recorded
was determined that the placement of the grids had minimal
for trawls (± standard deviation) was 16±5. Species richness
Demersal Fish
influence on the results.
results are displayed for individual trawls (Figure 11), as well
as mean richness for 5’ grid cells (Figure 12). There appear to
RESULTS AND DISCUSSION
be three trawl areas with consistently high species richness:
Richness by Individual Trawl
The mean (± standard deviation) number of fish species re-
NW of Point Año Nuevo, SW of Morro Bay, and near the center
corded for a demersal trawl was 16±5, and ranged from 1 to
of Cordell Bank NMS. In addition, there are smaller clusters
33. There are large areas with high species richness directly
within all three national marine sanctuaries. This map can be
west and north of Point Año Nuevo, between the 50 and 100 m
used to identify hot spots of demersal fish biodiversity.
contour lines as well as west and south of Morro Bay between
38°N
38°N
Richness 50 and 200 M depth. There are smaller hot spots within Cordell
DATA SOURCES
Bank NMS, between 100 and 200 meters, as well as along the
Richness estimates were derived from 1,336 NMFS (AKFSC
19 - 33
200 meter contour in four locations: north of Cordell Bank NMS,
and NWFSC) shelf (pp. 26) and slope (pp. 28) trawls conducted
15 - 18
just north of the southern Gulf of the Farallones NMS boundary,
between 50-1280 meters depth during June-November every
3 - 14 north of Monterey Bay and in southern Monterey Bay (Figure
third year from 1977-2001. For details on the trawl methods
11). For all trawls, there was a significant negative relationship
see Lauth (2001), Shaw et al. (2000), Turk et al. (2001), and
0 10 20 40 60
between richness and depth, and a significant positive relation-
Williams and Ralston (2002). All fish identified to the species
ship between richness and latitude. However, neither of these
level were included (230 species).
Kilometers
relationships explained much of the variance (r2=0.04, p<0.0001
for depth; and r2=0.005, p<0.004 for latitude, N=1336).
METHODS
37°N
37°N
Richness is defined as the number of fish species present at
Many fish species are associated with near-shore areas, and
a given location. To calculate richness, data were tabulated
were not included in this analysis due to the absence of NMFS
to determine the number of species caught in each trawl. Al-
trawls in shallow water areas. Other analyses in shallow wa-
though there was a significant positive relationship between
ter can provide a comparison to these results. Laidig (Pers
effort (calculated as distance fished x net width) and species
Comm NMFS) has completed underwater scuba surveys to
richness (p<0.0002), this accounted for a very small percentage
determine the presence of fish on kelp beds near Sonoma and
of the variability in the data set (adjusted r2=.01). Therefore,
Monterey. Average richness recorded on 43 dives in Sonoma
raw values of species richness for each trawl were used for this
was 5±3 (range of 1 to 15), and 15±4 on 9 dives in Monterey
analysis. Trawls are only possible along relatively flat bottom
(range of 9 to 21). California Department of Fish and Game
areas with a minor incline, and no data were available for rocky,
36°N
36°N
recreational fishing trips targeting rockfish (pp. 23) can also
highly sloped areas. In addition, the NMFS data did not include
be used to determine approximate fish richness. Without effort
trawls conducted in water less than 50 meters deep, therefore,
information on angler hours, the utility of mapping richness is
shallow water sites are not represented with these results.
questionable. However, the mean richness recorded was 7±4
(range of 1 to 21). The estimate of richness for near-shore
Figure 11 is useful for identifying actual trends in space, as
areas from CDF&G trawls is lower than those measured with
well as identifying where the trawls occurred. Species richness
the NMFS shelf and slope trawl data, but the difference could
results were organized into three equally sized groups repre-
be due to fishing method (hook and line vs. trawl) and is only
senting the lowest, middle, and highest third of richness values.
mentioned as anecdotal validation. There was a large difference
It is also useful to consider the mean richness for a small area.
in the number of species observed in Sonoma and Monterey
Therefore, mean richness and its deviation (how variable it is)
35°N
35°N
by Laidig, providing an example of the variability that can be
was calculated for 5’ grid cells throughout the study area (Figure
experienced in kelp areas. Managers interested in protecting
12). Cells with no deviation contained only one trawl. Cells that
biodiversity of demersal fish could use this information in com-
document high species richness, with low deviation, represent
20
100
50 m
bination with the other assemblage analysis to address various
an area with consistently high species richness. Cell size was
00
20
m management strategies. Cells with high species richness and
determined by minimizing the number of cells containing only
m
0
m
low deviation could be used to identify potentially important
one measurement yet retaining a reasonable spatial resolution
areas which deserve further investigation.
of the cells. This also was the cell size used for integration with
124°W 123°W 122°W 121°W
marine bird results. Species richness in cells is also presented
in three equal sized groups as this best represents differences
Figure 11. Species richness of individual NMFS shelf and slope trawls.
17
Subsection 2.1.1: ASSEMBLAGE ANALYSES
124°W 123°W 122°W 121°W
39°N
39°N
Species Richness of
Demersal Fish
Mean Richness in 5' Cells
38°N
38°N
Richness
18 - 25
15 - 17
8 - 14
Deviation of Richness
No Deviation
Less than Mean Deviation
Greater than Mean Deviation
37°N
37°N
0 10 20 40 60 80 100
Kilometers
36°N
36°N
35°N
35°N
20
100
50 m
00
20
m
m
0m
124°W 123°W 122°W 121°W
Figure 12. Mean species richness of NMFS shelf and slope trawls for 5’ grid cells. The deviation is shown as an overlay to provide an
indication of the variability in results for each grid cell.
18
Subsection 2.1.1: ASSEMBLAGE ANALYSES
ABOUT THESE MAPS divided into 3 equal sized groups representing the lowest,
124°W 123°W 122°W 121°W
Species diversity of demersal fish was calculated for NMFS middle, and highest third of diversity values.
trawls on the shelf and slope at depths between 50 and 1280
Species Diversity of
39°N
39°N
meters. The mean diversity recorded for trawls (± standard It is also useful to consider the mean diversity for a small area.
deviation) was 1.5±0.5. Species diversity results are displayed Therefore, mean diversity and its deviation (how variable it is)
Demersal Fish
for individual trawls (Figure 13), as well as mean diversity for was calculated for all trawls within 5’ grid cells throughout the
5’ grid cells (Figure 14). The largest cluster of high species study area (Figure 14). Cells with no deviation contained only
diversity trawls is found 20 km north and south of the border one trawl. Cells that document high species diversity, with low
between Monterey Bay NMS and Gulf of the Farallones NMS.
Diversity by Individual Trawl
deviation, represent an area with consistently high species
Smaller clusters of high diversity values are present in the diversity. Cell size was determined by minimizing the number
northwest corner of Cordell Bank NMS and to the north and of cells containing only one measurement, yet retaining a rea-
south of the NMS boundaries in waters slightly deeper than sonable spatial resolution of the cells. In addition, this was also
the 200 meter contour line. the cell size used for integration with marine bird analyses (sec-
38°N
38°N
Diversity tion 3). Species diversity in cells is also presented in 3 equally
DATA SOURCES sized groups. Since the placement of the grids is arbitrary, the
1.78 - 2.54
Diversity estimates were derived from 1,336 NMFS (AKFSC results will in-part depend on where the grid falls. An analysis
1.37 - 1.77
and NWFSC) shelf (pp. 26) and slope (pp. 28) trawls conducted comparing three different grid placements was conducted, and
0.02 - 1.36 between 50-1280 meters depth during June-November every it was determined that the placement of the grids had minimal
third year from 1977-2001. For details on the trawl methods influence on the results.
0 10 20 40 60 see Lauth (2001), Shaw et al. (2000), Turk et al. (2001), and
Williams and Ralston (2002). All fish identified to the species RESULTS AND DISCUSSION
Kilometers
level were included (230 species). The mean (± standard deviation) diversity recorded for a de-
mersal trawl was 1.5±0.5, with range from 0.02 to 2.54. The
37°N
37°N
METHODS largest cluster of high species diversity straddles the boundary
Diversity reflects the distribution of species’ abundance within between Monterey Bay NMS and Gulf of the Farallones NMS.
a trawl. For example, a trawl dominated by one species would Fifty-eight (13%) of the high diversity trawls are located within
have a low diversity, while a trawl with an even number of all 20 kilometers of this boundary. The western edge of this area
species would have a high diversity (see Figure 78, pp. 124). contains consistently high diversity trawls (low deviation). A
Since diversity is dependent on the abundance of species, fish smaller cluster of high diversity trawls is present in the north-
caught in a trawl which were not given an estimate of abun- west corner of Cordell Bank NMS, extending approximately 6
dance, or were not identified to species, were eliminated from kilometers north of the current boundary. Within this cluster,
the analysis. The Shannon index (H’) was calculated for each 95% of the trawls are classified as either medium or high
NMFS shelf and slope trawl based on the following equation: diversity. In addition, there are two lines of trawls with high
36°N
36°N
species diversity located slightly deeper than the 200 meter
n n
S
contour line: one north of Cordell Bank NMS to the northern
H ′ = − ∑ i ln i
edge of the study area, and the other from Lopez Point south
i =1 n n
to the southern edge of the study area. A large portion of these
Where S is the number of species, ni is the number of indi- trawls are outside sanctuary boundaries. For all trawls, there
viduals found of species i, and n is the total number of indi- was no significant relationship between diversity and latitude
viduals (Shannon and Weaver, 1949). Although there was a (r2=0.0, p=0.57, N=1336). There was a significant relation-
significant positive relationship between effort (calculated as ship between diversity and depth; however, it did not explain
distance fished x net width) and species diversity (p<0.0001), much of the variance in the data (r2=0.04, p<0.0001, N=1336).
this accounted for a very small percentage of the variability in Many fish species associated with near-shore, or high relief
35°N
35°N
the data set (adjusted r2=.06). Therefore, raw values of species areas, were not included in this analysis due to the absence of
diversity for each trawl were used for this analysis. Trawls are NMFS trawls in these areas. California Department of Fish and
only possible along relatively flat bottom areas with a minor in- Game recreational fishing trips (pp. 23) were often located in
20
100
cline, and no data were available for rocky, highly sloped areas. near-shore or high relief areas, and can be used to determine
50 m
00
20
m In addition, the NMFS data did not include trawls conducted approximate fish diversity over these habitats. The mean fish
m
0
m
in water less than 50 m deep, therefore, shallow water sites diversity recorded for recreational hook and line locations was
are not represented with these results. Figure 13 is useful for 1.3 ± 0.6 (range of 0 to 2.5). This estimate of diversity is similar
124°W 123°W 122°W 121°W
identifying actual trends in space, as well as identifying where to those measured with the NMFS shelf and slope trawl data;
effort occurred. For this figure, species diversity results were however, since the collection method was different, no statisti-
Figure 13. Species diversity of individual NMFS shelf and slope trawls.
19
Subsection 2.1.1: ASSEMBLAGE ANALYSES
cal comparisons can be completed and these results are only
124°W 123°W 122°W 121°W
intended as anecdotal validation.
39°N
39°N
Species Diversity of Trawls with high species diversity are not necessarily trawls with
high richness. Since diversity takes into account the number of
Demersal Fish fish of each species found, areas with one or two abundant spe-
cies have a lower diversity than areas with less fish species, but
an even distribution. High richness trawls are slightly shallower
Mean Diversity in 5' Cells than high diversity trawls suggesting that the trawls deeper than
200 meters have fewer species, but a more even distribution.
The presence of a high diversity area along the boundary be-
tween Gulf of the Farallones and Monterey Bay Sanctuaries
38°N
38°N
defines an area of biological significance for demersal fish. In
Diversity addition, there are lines of high species diversity north and south
1.63 - 2.27 of the current boundaries deeper than the 200 meter contour.
The trawls located on top of Santa Lucia Bank had medium to
1.39 - 1.62
high species diversity, and represent a large expanse of deep
0.18 - 1.38
habitat not within sanctuary boundaries. This area, combined
Deviation of Diversity with existing NMS shelf and slope areas, appears important to
No Deviation groundfish as indicated by high diversity patterns. Managers
interested in protecting biodiversity of demersal fish can use this
Less than Mean Deviation
information in combination with the other assemblage analysis
Greater than Mean Deviation
results to address various management strategies.
37°N
37°N
0 10 20 40 60 80 100
Kilometers
36°N
36°N
35°N
35°N
20
100
50 m
00
20
m
m
0m
124°W 123°W 122°W 121°W
Figure 14. Mean species diversity of NMFS shelf and slope trawls for 5’ grid cells. The deviation is shown as an overlay to provide an
indication of the variability in results for each grid cell.
20
Subsection 2.1.1: ASSEMBLAGE ANALYSES
ABOUT THIS MAP or high relief areas which were not included in this analysis.
124°W 123°W 122°W 121°W
Species richness of demersal rockfishes was calculated from Other analyses conducted in these habitats can provide a com-
Species Richness of
NMFS shelf and slope trawls (Figure 15). The information on parison to these results. Tom Laidig (pers. comm. NMFS) has
39°N
39°N
this map identifies areas with a high number of rockfish spe- conducted scuba surveys to determine the presence of fish on
cies. Values were not influenced by latitude, but were highly kelp beds near Sonoma and Monterey in California. Average
Demersal Rockfish
influenced by depth. The highest rockfish richness values rockfish richness recorded on 43 dives in Sonoma, between
were observed along the edge between the shelf and slope, 1983 and 1995, was 5±2 (range of 0 to 9), and 8±2 on 9 dives
emphasizing the importance of these areas to rockfish. Sanctu- in Monterey (range of 5 to 12). California Department of Fish
Richness by Individual Trawl
ary boundaries now include more than 500 square kilometers and Game recreational fishing trips targeting rockfish (pp. 23)
of this edge area (between 200 and 300 meters depth), with can also be used to determine approximate rockfish richness.
75% of this area within Monterey Bay NMS. The area south However, without fishing effort information, the utility of mapping
of Monterey Bay NMS to the edge of the study area contains the richness from recreational fishing data is questionable. Av-
another 500 square kilometers of habitat between 200 and erage rockfish richness recorded per location/trip combination
38°N
38°N
Richness 300 meters. was 6±2 (range of 0 to 12). The estimate of rockfish richness
for near-shore areas from CDF&G trawls is similar to those
7 - 14
DATA SOURCES measured with the NMFS shelf and slope trawl data, but since
5-6 Data were derived from 1,336 NMFS (AKFSC and NWFSC) the capture method was different, these results should only be
shelf (pp. 26) and slope (pp. 28) trawls conducted between used as an anecdotal validation.
3-4
50-1280 meters depth during June-November from 1977-2001.
2
For details on the trawl methods see Lauth (2001), Shaw et al. The results of this analysis illustrate the importance of the edge
0-1 (2000), Turk et al. (2001), and Williams and Ralston (2002). between shelf and slope areas. This result supports that of Wil-
All rockfish identified to the species level were included (48 liams and Ralston (2002), who found highest rockfish richness
species). between 200 to 250 meters depth using NMFS shelf data for
0 10 20 40 60
37°N
37°N
California and Oregon.
Kilometers
METHODS
Species richness is defined as the number of fish species
present at a given location. To calculate rockfish richness, data
were tabulated to determine the number of rockfish species
8
Sebastes or Sebastolobus present in each trawl. There was no
significant relationship between trawl effort (distance fished x
7
net width) and species richness (N=1336, F=1.3, p=0.26), so
Mean rockfish richness
raw values of species richness for each trawl were used for this
6
analysis. Trawls are only possible along relatively flat bottom
36°N
36°N
areas. No trawls were conducted over rocky, high relief areas or
5
areas in water less than 50 m deep, therefore, some potentially
important sites were not considered in these analyses.
4
RESULTS AND DISCUSSION
3
The mean (± standard deviation) number of rockfish species
recorded for a demersal trawl was 4±3, with a range from 0 to
2
14. The results show that bathymetry has a strong influence
on rockfish species richness. The lowest rockfish richness is
1
found in the shallower (but still >50 m) and deeper waters.
35°N
35°N
0 100200 400 600 800 1000 1200
A band of high rockfish richness is located around 200-300
meters depth and parallels the edge between the continental Depth (10 meter intervals)
shelf and slope. For all trawls, there was a significant non-linear
20
Figure 16. The relationship between depth and rockfish richness
00
100
50 m
relationship between richness and depth (Figure 16, F=166,
m showing mean rockfish richness (for 10 meter depth intervals be-
20
p<0.0001), and no significant relationship between richness
m
0
tween 50-1300 meters). The relationship was fit with a smoothing
m
and latitude (F=2.1, p=0.15). Almost all trawls on the deep slope spline, lambda = 1,000,000.
(deeper than 600 meters) contain the same two rockfish spe-
124°W 123°W 122°W 121°W
cies (shortspine and longspine thornyheads). It is important to
note that many rockfish species are associated with kelp beds
Figure 15. Species richness of rockfish from individual NMFS shelf and slope trawls.
21
Subsection 2.1.1: ASSEMBLAGE ANALYSES
ABOUT THIS MAP A good example of this split by depth can be found south of
124°W 123°W 122°W 121°W
The last three sections showing species diversity, species rich- the Monterey Bay NMS. This suggests that trawls with high
39°N
39°N
ness, and rockfish richness have provided results relevant to species richness found just east of the 200 meter contour are
Integration of Community managing resources. Figure 17 illustrates the overlay of the dominated by a few influential species. Conversely, the areas
top 17-20% of trawls for high species diversity, species rich- of high diversity just west of the 200 meter contour might have
ness, or rockfish richness. The background of the map shows one or two fewer species, but overall the species are evenly
Metrics for Fish the bathymetric complexity from page 16. The overlay of the distributed.
points provides visual representation of the results.
Results from the assemblage analyses were significantly tied to
DATA SOURCES depth; therefore, maps show bands of similar sites along depth
Legend Diversity, richness, and rockfish richness estimates were de- contours and do not delineate areas important to demersal fish.
rived from 1,336 NMFS (AKFSC and NWFSC) shelf (pp. 26) Conversely, the results from the community metrics do delin-
Top 20% of
38°N
38°N
and slope (pp. 28) trawls conducted between 50-1280 meters eate hot spots. Results are limited by collection method since
Diversity and Richness
depth during June-November from 1977-2001. For details on rocky, highly sloped, or shallow (less than 50 meters depth)
Top 20% of Diversity the trawl methods see Lauth (2001), Shaw et al. (2000), Turk areas were not sampled. Managers could use the interaction
et al. (2001), and Williams and Ralston (2002). of the community metrics to decide on proper management
Top 17% of Richness
strategies. For example, management is often tasked with pro-
Top 17% of
METHODS tecting biodiversity, and is therefore interested in delineating
Rockfish Richness
Methods for calculating diversity, richness, and rockfish rich- areas that contain the highest number of species. However, if
Bathymetric Complexity ness are detailed in each section. The top 20% of trawls for an area is high in richness, but is dominated by one economi-
Departure from Mean diversity were extracted and mapped. Ideally, the top 20% cally important species, protecting this area could contribute
of trawls for overall species richness and rockfish richness to resource use conflicts. The interplay between diversity and
-1 - 0 Std. Dev.
would be provided; however, since richness is discrete and not richness should be carefully evaluated.
37°N
37°N
0 - 1 Std. Dev.
continuous, either 17% (21+ species) or 23% (20+ species)
1 - 2 Std. Dev. could be mapped. The trawls which were within the top 20%
2 - 3 Std. Dev. for both richness and diversity are distinguished.
>3 Std. Dev.
RESULTS AND DISCUSSION
0 10 20 40 60 80 100
Richness calculates the number of fish species present in
each trawl, while diversity takes into account the abundance
Kilometers
of fish species as well. Diversity and richness are correlated
(r2=0.06), but trawls with high diversity are not necessarily
trawls with high richness. Trawls which were high in overall
36°N
36°N
richness and diversity were distinguished, and show areas
important to demersal fish. A cluster of trawls with high diver-
sity and high richness straddle the boundary between Gulf of
the Farallones and Monterey Bay NMS, as well as along the
200 meter contour north of Cordell Bank NMS. Small clusters
of high diversity and high richness trawls are present within
each sanctuary. Depth varied between the three community
metrics, with high richness, rockfish richness, and diversity
progressing from shallow to deep. The mean depth for trawls
with the top 17% of rockfish richness was 221±87, with 43%
35°N
35°N
of the trawls between 200 and 300 meters depth. Showing
these trawls reemphasizes the interaction between rockfish
richness and the edge of the continental shelf. The trawls with
50 m
50 m
20 high species richness show much more variability with depth
100
00
20
0
00
m
m (mean depth 212±225 meters), but 64% of them are in water
m
less than 200 meters deep. The trawls with high species diver-
sity were deeper (mean depth 372±289 meters), with 52% of
124°W 123°W 122°W 121°W
them greater than 300 meters depth. Overall, the trawls with
Figure 17. NMFS shelf and slope trawls with the highest species diversity, species richness, and rockfish richness are mapped. The high diversity were deeper than the trawls with high richness.
underlying map illustrates the bathymetric complexity of the study area and can be used to identify the shelf break.
22
Subsection 2.1.1: ASSEMBLAGE ANALYSES
ABOUT THESE ANALYSES the results compared for persistence and precision. Additionally, the data
Managers have recently begun to understand the importance of studying from 1993 to 1998 was analyzed separately to determine if current condi-
CDFG Recreational Hook and Line
entire ecosystems rather than looking at each species individually. This study tions have changed enough to affect the resultant species assemblages.
Species Assemblages
took a first step in clarifying multi-species interactions by determining which Conditions that could have changed through time include: abiotic shifts, such
Gopher Rockfish
species tended to be caught together, and where. Multivariate statistics were as decadal shifts in water temperature; biotic shifts, such as depletion of
Black Rockfish
used to analyze fish species assemblages on the scale of the recreational key species; and effort shifts, such as fishing farther offshore. To determine
Brown Rockfish
Cabezon
fishery over marine habitats off central California. This data set, while fishery which species groups were influential in forming the site groups, the aver-
China Rockfish
dependent, includes demersal, as well as midwater species captured on age frequency of occurrence for species assemblages in each site group
Kelp Greenling
variable habitats, including rock, mud, and sand. Some species and habitats was calculated. Species assemblages were considered influential if, on
Blue Rockfish
in this analysis are not covered with the other data sets in this study, and average, species were present in 25% of the trip/location combinations in
Olive Rockfish
therefore provide complimentary information. Twenty-seven fish species were a site group. The mean depth associated with each site group is provided
Yellowtail Rockfish
Canary Rockfish
grouped into seven species assemblages (Figure 18), and 4,357 trip/location in conjunction with a map showing the fishing locations in 2.5 minute grids,
Copper Rockfish
combinations were grouped into eight site groups (Table 1). Unfortunately, and color coded according to the average depth of the fishing trips within the
Lingcod
due to the nature of the data set (see methods), exact fishing locations could grid cell. Two-way analyses of variance (ANOVA) were conducted with depth
Rosy Rockfish
not be mapped. Therefore, the mean depth associated with each site group (pp. 37 Figure 32) and latitude, sediment (pp. 38 Figure 33), and bathymetric
Starry Rockfish
is provided in conjunction with a map showing the fishing locations in 2.5 complexity (pp. 16 Figure 10) to determine if any of these factors have an
Vermilion Rockfish
Pacific Sanddab
minute grids, which were color coded according to the average depth of the Bocaccio influence on the site group results at the scale of this analysis.
Flag Rockfish
Sh
fishing trips within the grid cell. The two analyses mentioned above provide
all Speckled Rockfish
ow
information on species which were caught together, and locations with similar RESULTS AND DISCUSSION
Widow Rockfish
to
catch. Combining the two results was the challenge. The average frequency of Species Assemblages (Objective A)
D Yelloweye Rockfish
ee
Quillback Rockfish
p
occurrence of species assemblages (percent occurrence calculated for each Seven species assemblages were differentiated from the recreational data,
species and then averaged for each fish assemblage) within each site group and named according to the most influential species (Figure 18). When
Greenspotted Rockfish
was calculated to analyze the interaction between the species assemblages the data from 1993-1998 were analyzed separately, there were two minor
Not Greenstriped Rockfish
Pacific Mackerel
in
at an fluentia
and site groups (Table 2). As with all data sets in this assessment, the most changes: the yellowtail rockfish assemblage split into two assemblages, and
Chilipepper
Squarespot Rockfish
l
y de
significant result was the effect of depth. This supports previous work done squarespot rockfish moved from the Pacific mackerel assemblage to the
pth
by Williams and Ralston (2002), Sullivan (1995), Field et al. (2002), Gabriel bocaccio assemblage. Overall, the species assemblages delineated were
and Tyler (1980), and Matthews and Richards (1991), who found bathymetry surprisingly robust; almost all fish were consistently placed in the same as-
Figure 18. Species assemblage results for the recreational data. Assemblages are named for from FishBase and NMFS
Pictures
the most influ-
to be an important factor in defining fish assemblages. All attempts to isolate semblages for more than 80% of the random runs, providing confidence in
ential species in each group. Assemblages are arranged from shallow to deep, unless they are influential at
and remove the effects of depth in order to determine secondary effects were the stability of the assemblages. Running the modified bootstrap technique
all or none of the depths. The assemblages that were not influential at any depth were composed of relatively
unsuccessful. Certainly, secondary effects exist, but at the scale of this study can provide an estimate of the precision of results, but verifying the accuracy
rare species, making depth associations indiscernible given the methodology for defining “influential” assem-
they were not discernible. Through this analysis, a large amount of informa- blages. Non-italicized species were consistently placed into the same species assemblage >80% of the time; of the results is more difficult. Comparisons of the results with past studies
italicized species tended to roam into other assemblages with random sampling.
tion has been condensed to assemblages of co-occurring species, as well as can give feedback on the accuracy of the results. Assemblages are not
groups of similar locations. A map is provided to visually portray the spatial static, and may modify in response to environmental conditions, such as
arrangement of the results. warm or cold conditions (see CD-ROM for changes in species assemblages
METHODS in response to water temperature or season).
DATA SOURCES The aim of the entire assemblage analysis was to increase our understanding of the biogeography
Data from 2167 commercial passenger fishing vessels, fishing for rockfish or of fishes and macro-invertebrates in relationship to their environment, and identify important areas Love et al. (2002) provides a summary of rockfish habitat requirements and
lingcod, using hook and line, were collected during all months between 1987 or habitats. Four of the five man objectives were addressed in this recreational analysis: species co-occurrences. The gopher rockfish and blue rockfish assemblages
and 1998 at depths between 2-360 meters. Each trip visited between 1 and A. Determine which species tended to be caught together (species assemblages); are supported by Love et al. (2002) as the species in each assemblage are
8 locations, with each trip/location combination considered a unique site. For B. Analyze fishing locations to determine which locations contained similar catches (site described as having the same habitat or co-occurring. In addition, Mason
this data set, effort was not provided, and therefore only presence/absence groups); (1995) looked at the recreational logbook and described a shallow rockfish
was analyzed at each trip/location combination. The data set contained in- C. Resolve where the species assemblages were being caught by combining results from assemblage composed of blue, black, brown, gopher, and olive rockfishes,
formation on 103 fish species, but after removal of rare species, the data objectives A and B and then utilizing GIS to map the results; and all found within the gopher and blue assemblages described in this study.
matrix used for classification contained information on 27 fish species at 4357 D. Identify significant relationships between the site groups identified in objective B and The greenspotted assemblage from this analysis is not necessarily intui-
trip/location combinations. A list of common and scientific names of the spe- broad scale habitat characteristics (bathymetry, bathymetric complexity, and large-scale tive. Greenspotted and greenstriped can both be found on mud near rocks
cies included in the analysis is available on the accompanying CD-ROM. To habitat classification). (Love et al., 2002), but this is also a characteristic of some of bocaccio and
protect individual fishing locations as requested by the CDF&G, results are yellowtail assemblage species (Love et al., 2002). Mason (1995) desig-
presented in 2.5 minute grids. For more information on the data collection Clustering is a technique used to summarize information into similar groups. The 1-Pearson cor- nated a deepwater red rockfish assemblage that included greenspotted,
process see Wilson-Vandenberg et al. (1996). relation coefficients, with the average means clustering method (see "Introduction to Clustering" pp. greenstriped, chilipepper, and bocaccio, which provides some support to
14) was used to summarize fish species into assemblages and catch locations into site groups. In the greenspotted assemblage of this study. Flag rockfish is an example of a
order to determine how variable the species assemblage results could be within the data, a modified species co-occurrence mentioned in Love et al. (2002) that is not supported
bootstrapping procedure was employed on 50 random samples composed of 50% of the data and here. Flag rockfish was placed in the bocaccio assemblage in this study, but
23
Subsection 2.1.1: ASSEMBLAGE ANALYSES
due to overlap of more than one group within the same cell. For
according to Love et al. (2002), flag is often found with species from the yellowtail
Group 26 Group 40 Group 44 Group 59 Group 64 Group 77 Group 98 Group 125
example, within one grid cell on the southern side of Monterey Bay,
rockfish assemblage. Within the modified bootstrapping procedure, flag rockfish was
meters meters meters meters meters meters meters meters
the maximum depth fished ranged between 37 and 660 meters,
placed with the Bocaccio assemblage 78% of the time, and with the yellowtail rockfish
Gopher
and contained sites from all 8 cluster groups. Therefore, the mean
assemblage only 28% of the time, supporting its placement in this analysis. 0.23 0.14 0.09 0.09 0.01 0.01 0.00
0.36
Assemblage
depths fished ±SD are presented, which can be used in conjunction
with Figure 19 to determine the approximate location of the site
The results comply with the large scale assemblages designated by NMFS: near-
Blue Assemblage 0.07 0.19 0.20 0.07 0.00
0.72 0.74 0.69
groups. Depth was the primary determinant of site groupings. All but
shore, shelf, and slope species groups (NMFS). All of the rockfish in each species
two (groups 40 and 44) of the eight site clusters were significantly
assemblage from this study come from the same NMFS group, except for the yellow-
Yellowtail
different in depth (see Table 1), suggesting that depth is highly in-
tail assemblage, which contains four species designated as “shelf” and one species 0.22 0.08 0.08
0.42 0.31 0.74 0.31 0.57
Assemblage
fluential in determining species distributions within the study area.
designated as “near-shore”. Williams and Ralston (2002) grouped rockfish from the
Bocaccio
The site groups we identified were similar to results of Sullivan
NMFS shelf trawl data into eight groups. While their assemblages differ from this
0.01 0.05 0.00 0.05 0.23 0.22
0.25 0.43
Assemblage
(1995), who analyzed a subset of this same data to differentiate
study's results, of the eleven species analyzed in both data sets, species from the
areas based on species composition. A direct comparison between
bocaccio and greenspotted rockfish assemblages are placed together, and species Greenspotted
0.00 0.00 0.02 0.02 0.07 0.10 0.50 0.59
Sullivan and this study is difficult because Sullivan describes his
from the gopher rockfish and yellowtail rockfish assemblages are placed together. Assemblage
locations verbally using land identifiers, while this study describes
Comparison of the results from this study with results based on trawl (NMFS; Williams
Pacific Mackerel
locations by depth. The importance of depth in this ecosystem is
and Ralston, 2002; Gabriel and Tyler, 1980; Jay, 1996), or results from submersibles 0.02 0.07 0.13 0.07 0.12 0.17 0.09 0.06
Assemblage
not a new idea; many researchers have already commented on
(Yoklavich et al., 2000, 2002; Hixon et al., 1991; Hixon and Tissot, 1992; Field et al.,
its influence (Williams and Ralston, 2002; Sullivan, 1995; Gabriel
2002), is difficult due to the species analyzed, the different habitats targeted, and the
Quillback 0.01 0.06 0.05 0.04 0.09 0.01 0.02 0.00
and Tyler, 1980; Field et al., 2002; Matthews and Richards, 1991).
variable scale of the results. Matthews and Richards (1991) found different species
Latitude has also been described as having an influence on Cali-
assemblages over trawlable and untrawlable habitats, showing the effect targeting Table 2. Average frequency of occurrence of fish species assemblages (percent occurrence calculated for each
fornia fish species composition (Williams and Ralston, 2002; Horn
different environments can have on species assemblages. Scale is important since species and then averaged for each fish assemblage) for each recreational site group. Numbers in bold represent
and Allen, 1978; Sullivan, 1995), but for the area of this study, no
the recreational boat drifts over multiple habitats during a set, and fish from multiple influential species assemblages within that site group.
latitudinal results were evident.
habitats can be present in one trip/location combination. In addition, species as-
semblage results could also be confounded by ontogenetic habitat shifts because
Interaction of Species and Sites (Objective C)
the sizes of the fish captured were not considered.
In conclusion, this analysis provides results showing species assemblages, site as-
The interaction between site groups and species assemblages (i.e. the location of
semblages, and the location of species assemblages for the important near-shore,
species assemblages) reemphasized the relationship between species and depth.
Site Groups (Objective B)
rocky environment. Understanding species assemblages and mapping their loca-
Species assemblages which were influential in forming each site group are identified
Eight site groups were identified from the 4,357 trip/location combinations (Table 1).
tion provides important information for managers. For example, to include the most
(Table 2). Site group 44 did not seem to be associated with any fish assemblages
To make interpretation easier, the site groups are named according to mean depth.
species assemblages, protecting an area that covers a large variation in depth may
(none with a frequency of occurrence greater than 25). At this point, it is uncertain
Maps with the location of the site groupings in the 2.5 minute grid are hard to interpret
be more important than protecting an area that covers a large variation in latitude.
what factor caused the clustering of this group. For all of the trip/location combina-
The results of this analysis in conjunction with similar analyses on the three other
tions, on average 68 fish were caught. The average number of fish caught for group
data sets provides a fairly comprehensive overview of fish and macro-invertebrate
44 was 12, suggesting that some outside factor, such as poor weather, was influenc-
species within the study area.
ing catches at these sights.
Site Group Depth±SD
(Names based on depth) (meters) Habitat Correlations (Objective D)
N
Other factors besides depth can have an impact on species assemblages. Examples
26 ± 13a
Group 26 meters 581 include latitude (Horn and Allen, 1978; Sullivan, 1995), sediment type (Yoklavich et
40 ± 16b al., 2000, 2002; Field et al., 2002; Hixon et al., 1991; Hixon and Tissot, 1992), and
Group 40 meters 688
substrate relief (see bathymetric complexity section pp. 16) (Yoklavich et al., 2000,
44 ± 27b
Group 44 meters 183 2002; Field et al., 2002). Unfortunately, since there were significant interactions
59 ± 26c present between all of these variables and depth, it could not be determined if the
Group 59 meters 235
significance detected for these factors was due to this interaction with depth. Even
64 ± 18d
Group 64 meters 1,501 though bathymetric complexity increases as depth increases, groups 44, 59, and
77 ± 22e 125 meters have a higher bathymetric complexity than the groups around them with
Group 77 meters 207
similar depth. All attempts to remove depth and determine secondary influences on
98 ± 21f
Group 98 meters 683 group designation were unsuccessful. The non-linear relationship between the fish
125 ± 32g
Group 125 meters 279 species and depth made removal of depth impossible. While these other factors
appear to have a decreased significance when compared to depth, more complex
analyses exploring ways to remove the effects of depth and determine the relative
Table 1. Site group results for recreational data. The numbers of trip/location combinations
significance of these factors may be completed in Phase II.
associated with each group as well as average depth, ± standard deviation, are provided.
Different letters signify a significant difference using Tukey’s pairwise comparison on log
adjusted depth with overall alpha set at 0.001.
24
Subsection 2.1.1: ASSEMBLAGE ANALYSES
124°W 123°W 122°W 121°W
CDF&G Recreational Data
39°N
39°N
in 2.5' Grids
Mean Depth
0 - 15 meters
15 - 25 meters
38°N
38°N
25 - 35 meters
35 - 45 meters
45 - 55 meters
55 - 65 meters
65 - 75 meters
75 - 85 meters
85 - 100 meters
37°N
37°N
>100 meters
0 10 20 40 60 80
Kilometers
36°N
36°N
35°N
35°N
100
50 m
20
00
20
m
m
0
m
124°W 123°W 122°W 121°W
Figure 19. Location of CDF&G recreational fishing data in 2.5 minute grids which are color coded according to the average depth of the
fishing trips within the grid cell. Lines showing the 50, 100, 200, and 2,000 depth contours are provided.
25
Subsection 2.1.1: ASSEMBLAGE ANALYSES
ABOUT THIS MAP were mapped using GIS. To determine which species groups were influential
Not
Recently managers and scientists have begun to understand the importance in forming the site groups, the average frequency of occurrence for species
in
at a fluentia
NMFS Shelf Trawls
of studying communities of species rather than just managing by individual assemblages in each site group was calculated. Species assemblages were
ny d
epth l
Sharpchin Rockfish
Pacific Herring
Species Assemblages
species. This study was an initial assessment aimed at determining which considered influential if, on average, species were present in 25% of the trawls
Spot Shrimp
American Shad
Threadfin Sculpin
California Market Squid
species tended to be caught together, and where. Multivariate statistics in a site group. Two-way analyses of variance (ANOVA) were conducted with
Chinook Salmon
were used to analyze species assemblages over trawlable habitats of the depth (pp. 37) and latitude, sediment (pp. 38), and bathymetric complexity
Curlfin Sole
Canary Rockfish
Dungeness Crab
continental shelf between 55 and 500 meters depth off California. For an (pp. 16) to determine if any of these factors have an influence on the site
Longspine Combfish Yellowtail Rockfish
Northern Anchovy Widow Rockfish
introduction to the continental shelf ecosystem, see the Ecological Linkages group results at the scale of this analysis.
Pacific Pompano Rock Sole Sp
Report. Sixty-one species were grouped into thirteen species assemblages Halfbanded Rockfish
White Croaker
Pacific Electric Ray Pacific Mackerel
RESULTS AND DISCUSSION
Arrowtooth Flounder
(Figure 20), and 883 trawls grouped into eight site groups (Table 3, Figure Jack Mackerel
Species Assemblages (Objective A)
Pacific Argentine
21). The average frequency of occurrence of species assemblages (percent Pacific Sanddab
English Sole
Thirteen species assemblages were determined in the NMFS trawl data
occurrence calculated for each species and then averaged for each fish as- Petrale Sole
set and named according to the most influential species (Figure 20). There
semblage) within each site group was calculated to analyze the interaction Pink Seaperch
Plainfin Midshipman
were no differences in the results when data from 1989-2001 were analyzed
between the species assemblages and site groups (Table 4). As with all data Lingcod
Big Skate
separately. Overall, the species assemblages delineated were robust; eight
sets, the most significant result was the effect of depth on species assem- California Skate
of the thirteen groups were consistently placed together for more than 80%
Spiny Dogfish
blages. All attempts to isolate and remove the effects of depth in order to Chilipepper
Bocaccio
of the random samples. This provides confidence that the results do not rep-
determine secondary effects were unsuccessful. Certainly secondary effects Cowcod
Rex Sole
resent just random groupings. Running the modified bootstrap technique can
Sh
exist, but at the scale of this study, they were not discernible. Our results Greenspotted Rockfish
Slender Sole
all Greenstriped Rockfish
Spotted Cusk-eel
provide an estimate of the precision of results, but verifying the accuracy of
ow
support previous results by Williams and Ralston (2002), Sullivan (1995), Shortbelly Rockfish
Pacific Hake
to Stripetail Rockfish
the results is more difficult. Comparisons of the results with past studies can
Field et al. (2002), Gabriel and Tyler (1980), and Matthews and Richards D
Shortspine
Longnose Skate
ee Darkblotched Rockfish
give feedback on the accuracy. Assemblages are not static and may modify
Spotted Ratfish Thornyhead
(1991), who found bathymetry to be an important factor in defining west p Bank Rockfish
Bering Skate
in response to environmental parameters, such as warm or cold conditions
coast demersal fish assemblages. Through this analysis, a large amount of Redbanded Rockfish
Bigfin Eelpout
Blackgill Rockfish
Splitnose Rockfish
Dover Sole
(see CD-ROM for changes in species assemblages in response to water
information has been condensed to assemblages of co-occurring species, Aurora Rockfish
Sablefish
Black Eelpout
temperature).
as well as groups of similar locations. A map is provided to visually portray Blacktail Snailfish
Al l d
epth
the spatial arrangement of the results. Brown Cat Shark
s Filetail Cat Shark
The species cluster results from the shelf trawls make intuitive sense in many
Lanternfish
ways. For example, most of the pelagic species were clustered together
DATA SOURCES Pictures from FishBase and NMFS
(Pacific herring assemblage), and the soft bottom and hard bottom species
Data from 883 fisheries independent research trawls (55-500 meters depth) Figure 20. Species assemblage results for the shelf trawls. Assemblages are named for the most influential
are separated for most of the groups. Only a few assemblages contain both
were collected every third year, between 1977 and 2001, during the months species in each group. Assemblages are arranged from shallow to deep, unless they are influential at all or
soft bottom and hard bottom species, for example the inclusion of the soft
of June-August. Gear included a nor’eastern trawl (127 mm stretched-mesh none of the depths. The assemblages that were not influential at any depth were composed of relatively rare
bottom- associated stripetail rockfish and rock sole with hard bottom assem-
body; 89 mm stretched-mesh codend; and 32 mm stretched-mesh codend species, making depth associations indiscernible given the methodology for defining “influential” assemblages.
blages (bocaccio and canary rockfish, respectively), and the placement of
liner) with a rubber bobbin roller which was trawled for 15-30 minutes on Non-italicized species were consistently placed into the same species assemblage >80% of the time; italicized
stripetail rockfish (hard bottom) in the darkblotched rockfish assemblage (soft
species tended to roam into other assemblages with random sampling.
the bottom. Data was adjusted for effort and to meet statistical assumptions
bottom) (Love et al., 2002). The chilipepper group contains fish species that
by dividing number of fish caught by the area covered and then log trans-
A. Determine which species tended to be caught together (species assemblages); are benthic as well as midwater schoolers, suggesting that even though these
forming. The data set contained information on 167 fish species, but after
B. Analyze fishing locations to determine which locations contained similar catches (site species behave differently they are responding to similar habitat characteris-
removal of rare species, the data matrix used for clustering contained only
groups);
58 fish and 3 invertebrate species. A list of common and scientific names
C. Resolve where the species assemblages were being caught by combining results from
of the species included in the analysis is available on the accompanying
Site Group Depth±SD
objectives A and B and then utilizing GIS to map the results; and
CD-ROM. Since each NMFS cruise hosted scientists with varying levels of
(Names based on depth) (meters)
N
D. Identify significant relationships between site groups identified in objective B and broad
expertise in invertebrate identification, NMFS scientists recommended that
± 16a
125 78
Group 78 meters
scale habitat characteristics (bathymetry, bathymetric complexity, and large-scale habitat
only well known/common invertebrate species be included in the analyses.
± 19b
103
classification). 93
Group 93 meters
Fish species assemblages were identical with and without the inclusion of
± 25b
invertebrates in the analysis. For more information on how the data were 136 96
Group 96 meters
Clustering is a technique used to summarize information into similar groups. The 1-Pearson correla-
collected, including the site selection process and how it changed through ± 37c
72 119
Group 119 meters
tion coefficients with the average means clustering method (see "Introduction to Clustering" pp. 14)
time, see Shaw et al. (2000), Wilkins et al. (1998), and Zimmermann et al.
± 41d
171 153
Group 153 meters
was used to first summarize fish species into assemblages, and to then summarize catch locations
(2001).
± 52e
116 268
Group 268 meters
into site groupings. In order to determine how variable the species cluster results could be within
± 51f
37 328
Group 328 meters
the data, a modified bootstrapping procedure was employed on 50 random samples composed of
METHODS
± 48g
415
123
50% of the data and the results compared for persistence and precision. Additionally, the data from Group 415 meters
The aim of the entire assemblage analysis was to increase our understand-
1989 to 2001 were analyzed separately to determine if current conditions have changed enough
ing of the biogeography of fishes and macro-invertebrates in relationship to Table 3. Site group results for shelf trawl data. The numbers of trawls associated
to affect the resultant species assemblages. Conditions that could have changed through time
their environment, and identify important areas or habitats. Four of the five with each group as well as average depth ± standard deviation are provided. Differ-
include: abiotic shifts, such as decadal shift in water temperature; biotic shifts, such as depletion
main objectives were addressed in this analysis: ent letters signify a significant difference using Tukey’s pairwise comparison on log
of key species; or effort shifts, such as fishing farther offshore. The location of the eight site groups adjusted depth with overall alpha set at 0.001.
26
Subsection 2.1.1: ASSEMBLAGE ANALYSES
tics. Love et al. (2002) mention species co-occurrences, such as the association of mented on its influence (Williams and Ralston, 2002; Sullivan, 123°W 122°W 121°W
cowcod (immature) with bocaccio and greenstripe rockfish, which are corroborated 1995; Gabriel and Tyler, 1980; Field et al., 2002; Matthews and
39°N
39°N
by the results of this analysis. However, this study, unlike Love, found that stripetail Richards, 1991). Species assemblages which were influential in
NMFS Shelf Trawls
rockfish and a splitnose rockfish did not occur together. The placement of yellowtail forming each site group are identified (Table 4). The interaction
rockfish, canary rockfish, and widow rockfish together in this analysis is supported between site groups and species assemblages (i.e. the location
by multiple studies (Tagart and Wallace, Leet, 2001, Star et al., 2002). of species assemblages) reemphasized the relationship between
species and depth. Three species groups (rex sole, Pacific hake,
All rockfish groups concurred with the broad characterization by NMFS (into near- and shortspine thornyhead assemblages) had a high frequency
Site Groups
shore, shelf, and slope species groups), which were based on an assemblage analysis of occurrence in all trawl groups. The rest of the species groups
completed by Gabriel and Tyler (1980). Williams and Ralston (2002) used the same were arranged from shallow to deep. In all cases, the assem- Group 78 Meters
data set as this study to examine rockfish species assemblages. Their “southern blages with a low frequency of occurrence for all site groups were Group 93 Meters
shelf group” contained species clustered together in this report's chilipepper and composed of relatively rare species. Depth associations were
Group 96 Meters
canary groups, while their “deep-water slope” group was split among three groups
38°N
38°N
present, just not discernible given the methodology for defining
in this report. The results from submersibles (Yoklavich et al., 2000, 2002; Hixon et “influential” assemblages. The location of the trawls designated Group 119 Meters
al., 1991; Hixon and Tissot, 1992; Field et al., 2002) provide relevant species/habitat to each site group were mapped using GIS (Figure 21). Group 153 Meters
interactions at a scale meaningful to fish, however, many of the results from these
Group 268 Meters
studies are not comparable with the current studies due to the large difference in Habitat Correlations (Objective D)
scale. Hixon et al. (1991) documented that the species composition observed from Other factors besides depth can have an impact on species Group 328 Meters
the submersibles was different than species captured in trawls. assemblages. Examples of these factors include latitude (Horn Group 415 Meters
and Allen, 1978; Sullivan, 1995), sediment type (Yoklavich et
Site Groups and Interaction of Species and Sites (Objectives B and C) al., 2000, 2002; Field et al., 2002; Hixon et al., 1991; Hixon and 0 10 20 40 60 80 100
Eight site groups were identified from the 883 shelf trawls (Table 3). To make inter- Tissot, 1992), and bathymetric relief (see bathymetric complexity
Kilometers
pretation easier, the site groups are named according to mean depth. All but two pp. 16) (Yoklavich et al., 2000, 2002; Field et al., 2002). Unfortu-
37°N
37°N
groups are significantly different in depth using an ANOVA (Table 3). The importance nately, since there were significant interactions present between
of depth in this ecosystem is not a new idea; many researchers have already com- all of these variables and depth, it could not be determined if the
significance detected for these factors was due to this interaction
with depth. Even though bathymetric complexity increases as
Group 78 Group 93 Group 96 Group 119 Group 153 Group 268 Group 328 Group 415
meters meters meters meters meters meters meters meters
depth increases, group 268 has a higher bathymetric complexity
than the groups around them with similar depth. It is interesting
0.64 0.59 0.82 0.74 0.80 0.78 0.93 0.63
Rex Sole Assemblage
to note that for this data set, 89% of the trawls occurred over
Pacific Hake
0.41 0.37 0.62 0.39 0.58 0.61 0.87 0.68
areas delineated as mud, and 8% over areas delineated as sand
Assemblage
(pp. 38) making it impossible to examine the effects of sediment.
Shortspine Thornyhead
0.28 0.32 0.39 0.64 0.96 0.83
0.20 0.23
Assemblage
However, all 15 trawls that occurred over habitat designated as
36°N
36°N
Pacific Herring
“mud-rock mix” were clustered together into the "415 meters
0.63 0.31 0.34 0.21 0.06 0.01 0.01 0.01
Assemblage
group". All attempts to remove depth and determine secondary
0.27
Halfbanded Assemblage 0.22 0.11 0.24 0.07 0.02 0.01 0.01 influences on group designation were unsuccessful. The non-
linear relationship between the fish species and depth made
Pacific Sanddab
0.90 0.91 0.88 0.82 0.55 0.13 0.10 0.03
Assemblage removal of depth impossible. While these other factors appear
to have a decreased significance when compared to depth, more
0.27 0.26 0.40 0.25 0.25
Big Skate Assemblage 0.16 0.24 0.12
complex analyses exploring ways to remove the effects of depth
and determine the relative significance of these factors may be
0.57 0.61 0.42
Chilipepper Assemblage 0.12 0.26 0.22 0.12 0.03
completed in Phase II.
Darkblotched
0.41 0.41
0.01 0.00 0.03 0.03 0.10 0.17
Assemblage
35°N
35°N
In conclusion, this analysis provides results showing species
Blackgill Rockfish
0.30 0.56
0.01 0.01 0.00 0.02 0.01 0.05
Assemblage
assemblages, site assemblages, and the location of species
assemblages for the important shelf environment. The species
Canary Assemblage 0.02 0.10 0.06 0.21 0.15 0.04 0.01 0.00
20
50 m
100
00
assemblages are relevant to the scale of a commercial trawl. The
20
m
0
Sharpchin Assemblage 0.01 0.01 0.02 0.05 0.13 0.23 0.10 0.01
larger species assemblages that were reported in the literature
m
m
were confirmed (NMFS near-shore, shelf and slope groups). For
Arrowtooth Flounder
0.00 0.05 0.01 0.01 0.11 0.14 0.14 0.02
Assemblage the most part, pelagic, soft bottom, and hard bottom species as- 123°W 122°W 121°W
Table 4. Average frequency of occurrence of fish species assemblages (percent occurrence semblages were distinguished, providing initial feedback to the
calculated for each species and then averaged for each fish assemblage) for each shelf site Figure 21. Location of site groups for NMFS shelf trawls. Lines showing the 50, 100, 200, and 2,000 depth
accuracy of the species assemblages.
group. Numbers in bold represent influential species assemblages within that site group. contours are provided.
27
Subsection 2.1.1: ASSEMBLAGE ANALYSES
rence for species assemblages in each site group was calculated. Species
ABOUT THIS MAP Not
in
NMFS Slope Trawls at a fluentia assemblages were considered influential if, on average, species were present
Little is known about the deep slope species, especially information on which
ny d
epth l in 25% of the trawls in a site group. Two-way analyses of variance (ANOVA)
Species Assemblages
species are found together on what habitats. Multivariate statistics were used
were conducted with depth and latitude, sediment, and bathymetric complexity,
to analyze species assemblages over trawlable habitats between 190 and Stripetail Rockfish California Market Squid
to determine if any of these factors have an influence on the site group results
1280 meters depth off California. This is the first attempt to exclusively define Bocaccio Blackbelly Eelpout
Chilipepper at the scale of this analysis.
species assemblages on the deep slope community. For an introduction to Robust Clubhook Squid
Darkblotched Rockfish
the continental slope ecosystem, see the Ecological Linkages Report. Eight English Sole
RESULTS AND DISCUSSION
Greenstriped Rockfish
species assemblages (Figure 22) and seven site groups (Table 5, Figure 23) Splitnose Rockfish
Lingcod
Species Assemblages (Objective A)
were identified. The average frequency of occurrence of species assemblages Bering Skate
Petrale Sole
Bigfin Eelpout Eight species assemblages were determined for the NMFS slope trawls, and
(percent occurrence calculated for each species and then averaged for each Sharpchin Rockfish
Longnose Skate
Shortbelly Rockfish named according to the most influential species (Figure 22). Overall, the spe-
fish assemblage) within each site group was calculated to analyze the inter- Pacific Hake
Spot Shrimp
cies assemblages delineated were robust; seven of eight assemblages were
Rex Sole
action between the species assemblages and site groups (Table 6). As with Slender Sole
Spotted Ratfish Filetail Catshark
consistently placed together for more than 80% of the random samples. This
Redbanded Rockfish
all data sets, the most significant result was the effect of depth on species Black Eelpout
Rosethorn Rockfish
provides confidence that the results do not represent just random groupings.
assemblages. All attempts to isolate and remove the effects of depth in order California Grenadier
Pacific Electric Ray
Flapjack Devilfish Running the modified bootstrap technique can provide an estimate of the preci-
to determine secondary effects were unsuccessful. Certainly secondary ef-
Pacific Glass Shrimp Longspine Thornyhead sion of results, but verifying the accuracy of the results is more difficult.
fects exist, but at the scale of this study, they were not discernible. Our results
Black Skate
support previous results by Williams and Ralston (2002), Sullivan (1995), Califonia Slickhead
Sh Pacific Viperfish
The species cluster results from these NMFS slope trawls seem much less
all
Field et al. (2002), Gabriel and Tyler (1980), and Matthews and Richards Crimson Pasiphaeid Black Hagfish
ow Deepsea Sole Deepsea Skate intuitive than those from the NMFS shelf trawls. This is partly due to a lack of
Aurora Rockfish
(1991), who found bathymetry to be an important factor in defining demersal to Giant Grenadier Fangtooth
Blackgill Rockfish
D research and subsequent decreased understanding of the behavior of slope
fish assemblages on the West Coast. Through this analysis a large amount ee Grooved Tanner Crab Longfin Dragonfish
Dover Sole
p species. Many of the species from the NMFS slope trawls understandably
Pacific Flatnose
of information has been condensed down to assemblages of co-occurring Pacific Blackdragon
Spiny Dogfish
Pacific Grenadier Rhomboid Squid
Bank Rockfish overlap with species from the NMFS shelf trawls. Surprisingly, some of the
species, as well as groups of similar locations. A map is provided to visually Snakehead Eelpout Sawtooth Eel
species interactions noted with the shelf trawls are not upheld with the slope
portray the spatial arrangement of the results. Twoline Eelpout Smooth Grenadier
All Sablefish Threadfin Slickhead data. The stripetail rockfish slope assemblage is composed of all of the shal-
d ep Blacktail Snailfish Vampire Squid
lower species (soft and hard bottom) that were distributed among 6 shelf as-
DATA SOURCES th Brown Catshark Magistrate Armhook Squid
s semblages. This does not imply that species co-occurrences changed between
Shortspine Thornyhead
Data from 454 fisheries independent research trawls between depths of 190-
Pictures from FishBase and NMFS
the shelf and slope trawls, just that cluster results were sensitive to the depth
1280 meters were collected in 1991, 1997, 1999, 2000, and 2001, during the
range covered by the data set.
months of July-November. For 1999, 2000, and 2001 (NWFSC), gear included Figure 22. Species assemblage results for the slope trawls. Assemblages are named for the most influential
species in each group. Assemblages are arranged from shallow to deep, unless they are influential at all
an aberdeen net with a small mesh liner (2 inches stretched) at the codend
or none of the depths. The assemblages that were not influential at any depth were composed of relatively Love et al. (2002) provides a summary of rockfish habitat requirements and
which was trawled along the bottom along east-west transects for 15 min-
rare species, making depth associations indiscernible given the methodology for defining “influential” as- species co-occurrences. However, since only 15 (24%) species are rockfish,
utes. For 1991, 1997, 1999, and 2000 (AKFSC), gear included a nor’eastern
semblages. Non-italicized species were consistently placed into the same species assemblage >80% of this information cannot be used to assess all results. None of the assemblages
(127 mm stretched-mesh body, 89 mm stretched-mesh codend, and 32 mm
the time; italicized species tended to roam into other assemblages with random sampling.
in this study completely agree with species co-occurrences listed in Love et
stretched-mesh codend liner) with a rubber bobbin roller which was trawled
al. (2002). For example, Love et al. stated that stripetail rockfish and splitnose
on the bottom for 15-30 minutes. Although different gears were utilized in this
rockfish are found together. However, within both the shelf and slope data sets
data set, preliminary analyses found no significant difference between years, objectives were addressed in this analysis:
of this analysis the stripetail and splitnose rockfish were placed into different
allowing the data sets to be combined (pers com Tonya Builder, NMFS). Data A. Determine which species tended to be caught together (species assemblages);
groups. The results are consistent with the large scale assemblages desig-
was adjusted for effort and statistical assumptions as in the NMFS Shelf Trawls. B. Analyze fishing locations to determine which locations contained similar catches
The data set contained information on 161 fish species, but after removal (site groups);
of rare species, the matrix used for classification contained 52 fish and 10 C. Resolve where the species assemblages were being caught by combining results
Site Group Depth±SD
invertebrate species. A list of common and scientific names of species in- from objectives A and B and then utilizing GIS to map the results; and
(Names based on depth) (meters)
N
cluded in this analysis is available on the accompanying CD-ROM. Since each D. Identify significant relationships between site groups identified in objective B and
a
NMFS cruise hosted scientists with varying levels of expertise in invertebrate Group 263 meters 84
broad scale habitat characteristics (bathymetry, bathymetric complexity, and large- 263 ± 49
b
identification, NMFS scientists recommended that only well known/common scale habitat classification). Group 410 meters 86 410 ± 46
invertebrate species be included in the analyses. Fish species assemblages c
Group 530 meters 43 530 ± 42
were identical with and without the inclusion of invertebrates in the analysis. Clustering is a technique used to summarize information into similar groups. The 1-Pearson d
Group 622 meters 29 622 ± 27
For more information on how the data were collected, including site selection correlation coefficients with the average means clustering method (see introduction to cluster- e
Group 733 meters 48 733 ± 71
procedures for each data set, see Turk et al. (2001) and Lauth (2001). ing pp. 14) was used to first summarize the fish species into assemblages, and to then sum- f
Group 931 meters 90 931 ± 132
marize the catch locations into site groupings. In order to determine how variable the species
g
Group 1112 meters 74
METHODS cluster results could be within the data, a modified bootstrapping procedure was employed on 1112 ± 95
The aim of the entire assemblage analysis was to increase our understanding random samples composed of 50% of the data and the results compared for persistence and Table 5. Site group results for slope trawl data. The numbers of trawls associated with
of the biogeography of fishes and macro-invertebrates in relationship to their precision. The location of the seven site groups were mapped using GIS. To determine which each group as well as average depth ± standard deviation are provided. Different let-
environment, and identify important areas or habitats. Four of the five main ters signify a significant difference using Tukey’s pairwise comparison on log adjusted
species groups were influential in forming the site groups, the average frequency of occur-
depth with overall alpha set at 0.001.
28
Subsection 2.1.1: ASSEMBLAGE ANALYSES
nated by NMFS: near-shore, shelf, and slope species groups 1978; Sullivan, 1995), but for the area of this study, no latitudinal
124°W 123°W 122°W 121°W
(NMFS, 2002). All of the rockfish in each species assemblage results were evident.
from this study come from the same NMFS group. The rock-
39°N
39°N
fish groups identified by Williams and Ralston (2002) were not The interaction between site groups and species assemblages (i.e.
NMFS Slope Trawls
completely corroborated by this study. The species in their the location of species assemblages), reemphasized the relation-
“southern shelf group” were placed into the stripetail rockfish ship between species and depth (Table 6). The species groups
group which included all shallow-water species. However, their were arranged such that they went from shallow to deep. In all
“deep water slope” group was split among four of this report's cases, the assemblages with a low frequency of occurrence for all
groups. Comparisons of the results from this study with other site groups were composed of relatively rare species. Depth asso- Site Groups
assemblage studies are difficult due to the variability in species ciations were present, just not discernible, given the methodology
Group 263 Meters
analyzed between studies, the different habitats targeted, and for defining “influential” assemblages. The location of the trawls
the discrepancy in scale. For example, results from submers- Group 410 Meters
contained in each site group were mapped (Figure 23).
ibles (Yoklavich et al., 2000, 2002; Hixon et al., 1991; Hixon Group 530 Meters
38°N
38°N
and Tissot, 1992; Field et al., 2002) record interactions at a Habitat Correlations (Objective D)
Group 622 Meters
much smaller scale compared to trawls, which can fish multiple Other factors besides depth, such as latitude (Williams and
Group 733 Meters
habitats during a 1 km tow. Ralston, 2002; Horn and Allen, 1978; Sullivan, 1995), bottom
composition (Yoklavich et al., 2000, 2002; Field et al., 2002; Hixon Group 931 Meters
Site Groups and Interaction of Species and Sites (Objec- et al., 1991; Hixon and Tissot, 1992), or bathymetric complexity
Group 1112 Meters
tives B and C) (pp. 16) (Yoklavich et al., 2000, 2002; Field et al., 2002) can have
Seven site clusters were identified from the 454 slope trawls an impact on species assemblages. For this data set, 95% of the 0 10 20 40 60 80
(Table 5). To make interpretation easier, the site groups are trawls occurred over areas delineated as mud (pp. 38), making
Kilometers
named according to mean depth. Species assemblages which it impossible to examine the effects of bottom composition. It is
were influential in forming each site group are identified. The interesting to note that of all the 16 trawls completed over the
37°N
37°N
average frequency of occurrence of fish species assemblages bottom type designated as a combination of mud and rock, 33
for each site group (Table 6) was used to determine where percent occurred in the "410 meters" site group, and 56 percent in
species assemblages were found. An ANOVA determined that the "deepest" group. For the slope trawls, there was no interaction
all groups were significantly different in depth (Table 5). The present between depth and bathymetric complexity or latitude, so
importance of depth in this ecosystem is not a new idea; many the effects of these parameters could be tested. Neither bathy-
researchers have already commented on its influence (Williams metric complexity (pp 16), nor latitude, had a significant impact
and Ralston, 2002; Sullivan, 1995; Gabriel and Tyler, 1980; Field on site grouping when the effect for depth was accounted for
et al., 2002; Matthews and Richards, 1991). Latitude has also (bathymetric complexity: df=1, F=0.94, P=0.33; latitude: df=1,
been described as having an influence on California fish spe- F=0.11, P=0.74).
cies composition (Williams and Ralston, 2002; Horn and Allen,
36°N
36°N
In conclusion, this analysis provides results
showing species assemblages, site assem-
Group 263 Group 410 Group 530 Group 622 Group 733 Group 931 Group 1112
meters meters meters meters meters meters meters
blages, and the location of species assem-
blages for the deep slope environment that
0.41 0.78 0.87 0.97 0.95 0.84 0.68
Sablefish Assemblage
are relevant to the scale of a commercial
Aurora Rockfish
trawl. The larger species assemblages
0.40 0.75 0.55 0.31 0.27 0.25 0.17
Assemblage
that were reported in the literature were
Stripetail Rockfish
0.51 0.08 0.00 0.00 0.00 0.00 0.00
confirmed (NMFS, 2002) (near-shore,
Assemblage
shelf and slope groups), but some of the
Splitnose Rockfish
0.88 0.92 0.64 0.36 0.18 0.04 0.06
Assemblage
smaller groups were not corroborated. Half
Filetail Catshark
of the sites designated as "1,112 meters"
35°N
35°N
0.33 0.51 0.43
0.03 0.23 0.08 0.05
Assemblage
are located outside sanctuary boundaries.
Longspine Thornyhead
This is mainly due to the large number of
0.65 0.87 0.84
0.01 0.05 0.22 0.41
Assemblage
deep sites located to the south and west of
20
Pacific Viperfish
100
50 m
00
0.27
0.01 0.02 0.06 0.09 0.14 0.16
20
Sanctuary boundaries. The results of this
Assemblage
m
0
m
analysis in conjunction with similar analyses
m
Market Squid
0.14 0.15 0.13 0.07 0.07 0.03 0.05
Assemblage on the three other data sets provides a fairly
comprehensive overview of fish and macro-
Table 6. Average frequency of occurrence of fish species assemblages (percent occurrence 124°W 123°W 122°W 121°W
calculated for each species and then averaged for each fish assemblage) for each slope site invertebrate species within the study area. Figure 23. Location of site groups for NMFS slope trawls. Lines showing the 50, 100, 200, and 2,000 depth contours are provided.
group. Bold numbers represent influential species assemblages within that site group.
29
Subsection 2.1.1: ASSEMBLAGE ANALYSES
if, on average, species were present in 25% of the trawls in a site group.
ABOUT THESE MAPS
Interactions between environmental variables (salinity, temperature, density,
The pelagic environment is home to many marine species at some stage in
NMFS Midwater Trawls bottom depth, and bathymetric complexity) were investigated by conducting
their life. Therefore, it is important to document what species interact in this
step-wise discriminant analyses.
Species Assemblages
environment, and determine environmental influences on species abundance.
Multivariate statistics were used to analyze fish and invertebrate species
(All Years) RESULTS AND DISCUSSION
assemblages caught in trawls conducted at 7 and 30 meters depth between
Pacific Hake, juv. Species Assemblages (Objective A)
Cordell Bank NMS and Monterey Bay. For more information on the neritic
Calif. Smoothtongue The neritic environment is an important ecosystem in central California. Most
environment see the Ecological Linkages Report. When determining species
Deep-sea Smelt
benthic species have a larval stage dependent on the neritic environment.
assemblages, three separate analyses were completed: all data (1986-2001), Canary Rockfish, juv. Medusafish
Euphausiid
In addition, neritic species are an important base for the food web for fish,
only 1998 (warm year), and only 1999 (cold year). There were differences Black Rockfish, juv. King-of-the-salmon
Myctophid
Slender Barracudina Blue Rockfish, juv. birds, and mammals. Due to the removed rare species, different species were
between years in the species present, as well as in the organization of Rex Sole, juv.
Bocaccio, juv. included in the analyses depending on the year (1998, 1999, or all years).
species assemblages. The site groups identified were significantly related Slender Sole, juv.
Chilipepper, juv. Ten species were present in 1999 that were absent in 1998, including six
to environmental conditions, but the influential conditions varied between Pacific Tomcod, juv.
Pygmy Rockfish,juv.
species of juvenile rockfish. For 1998 and 1999, there were five and six spe-
years. To investigate assemblages persistent over more than one year, all Sand Sole, juv.
Shortbelly Rockfish, juv.
Squarespot Rockfish, juv. Slender Sole, adult cies assemblages identified, respectively, and seven species assemblages
of the data collected was analyzed for species assemblages, but not for site
were differentiated in the entire data set. Species assemblages were named
assemblages, due to the difficulty in mapping assemblages through time as Stripetail Rockfish, juv.
Spiny Dogfish according to the most influential species (Figures 24, 25, 27). Overall, the
the same location may house any number of assemblages depending on Widow Rockfish, juv.
Pacific Hake, adult
Yellowtail Rockfish, juv. species assemblages were much less robust than those from the other data
environmental conditions. See Larson et al (1994) for a detailed analysis of
sets. Average persistence (percentage of time species were grouped to-
rockfish assemblages in response to short-term environmental variability.
Pacific Sanddab, juv.
gether) through random runs varied from 36% to 94% for all assemblages.
Through this analysis a large amount of information has been condensed Market Squid
Brown Rockfish, juv.
This variability in results reflects two things: 1) the ephemeral nature of the
down to assemblages of co-occurring species, as well as groups of similar Lingcod, juv.
Copper Rockfish complex
neritic ecosystem and its expression through the species assemblages, and
locations. Maps are provided to portray the spatial arrangement of the site Dover Sole, juv. Northern Anchovy
2) the higher variability in results from the random runs due to smaller sample
Northern Anchovy, larval
groups for 1998 and 1999. Pacific Butterfish
Pacific Argentine, juv. size. Some of the persistent groups (persistent through random runs as well
Pacific Electric Ray
Speckled Sanddab, juv. as persistent through the three analyses) were: 1) market squid, northern
DATA SOURCES Pacific Sanddab, adult
anchovy, Pacific electric ray, and Pacific sardine; 2) euphasid, Pacific hake,
Data from 1543 fisheries independent research trawls were collected from Pacific Sardine
and deep sea smelt; and 3) myctophid and slender barracuda. Different
1986-2001, during May and June. The purpose of the trawl was to determine Plainfin Midshipman
juvenile rockfish were present in 1998 and 1999. In 1998, a warm year, only
the number and quantity of juvenile rockfish present during the upwelling
Pictures from FishBase and NMFS
shortbelly and stripetail rockfish were in greater than 5% of the trawls, and
season at night. The midwater trawl net was a “Cobb trawl” constructed of
nylon webbing. A fine mesh (1.25 cm) liner was inserted in the codend to Figure 24. Species assemblage results for the midwater trawls utilizing all data from 1986 to 2001. Assemblages were grouped together. In 1999, a cold year, 8 species of juvenile rockfish
are named for the most influential species in each group. Non-italicized species were consistently placed into
were identified, and all grouped together except for blue rockfish and stripetail
retain small midwater organisms. To open the net vertically, 42 8-inch floats the same species assemblage >80% of the time; italicized species tended to roam into other assemblages with
rockfish. This supports the observation by Loeb et al. (1994), Yoklavich et al.
were attached to the headrope and 145 pounds of chain was lashed to the random sampling.
(1996), and Moser et al. (2000), that warm (El Niño) years are not good for
footrope. To spread the net horizontally, 850 lb steel ‘V’ doors 5 x 7 feet in
rockfish recruitment. For the entire data set there were thirteen species of
A. Determine which species tended to be caught together (species as assemblages);
size were used. The net was deployed while the ship was underway at a
rockfish, which were all grouped together except for copper rockfish complex
B. Analyze fishing locations to determine which locations contained similar catches (site
speed of approximately 2.7 knots. Seventy-five meters of trawl cable were
and brown rockfish. The consistent grouping of juvenile rockfish together
groups);
payed-out which nominally puts the headrope at a depth of 30 meters (100
suggests that the rockfish species are responding to similar environmental
C. Resolve where the species assemblages were being caught by combining results from
feet), the depth at which most species of juvenile rockfishes are most abun-
conditions. The copper and brown rockfishes are not well known; however,
objectives A and B and then utilizing GIS to map the results; and
dant (Lenarz et. al., 1991). The spatial effort was consistent between years
Larson et al. (1994) looking at the midwater trawls in 1987 and 1988, deter-
D. Identify significant relationships between site groups identified in objective B and broad
with 1-4 hauls completed at 32 stations (see Figure 26 for spatial extent of
mined that some species in the copper complex arrive later in the season
scale habitat characteristics (bathymetry, bathymetric complexity, and large-scale habitat
data). Analyses were completed for the entire data set, as well as 1998 and
than most other rockfish species, which could affect their association with
classification).
1999 individually (representing warm and cold water years, respectively). Fish
assemblages.
numbers were adjusted for effort by NMFS then log transformed for analyses.
The data matrices used for classification contained data for 16, 24, and 41 Clustering is a technique used to summarize information into similar groups. The 1-Pearson correla-
Running the modified bootstrap technique can provide an estimate of the
species in 1998, 1999, and all years, respectively. A complete list of common tion coefficients with the average means clustering method (see "Introduction to Clustering" pp. 14)
precision of results, but verifying the accuracy of the results is more difficult.
and scientific names of the species included in these analyses is available on was used to first summarize the fish species into assemblages, and to then summarize the catch
Only two studies were identified which have investigated species assem-
locations into site groupings. Species assemblages were determined for 1998, 1999, and all data,
the accompanying CD-ROM.
blages in the neritic environment (Larson et al., 1994; Calliet et al., 1979).
but site groups were only determined for 1998 and 1999. In order to determine how variable the
Larson et al. (1994) analyzed this data set (1987-1988) for juvenile rockfish
species cluster results could be within the data, a modified bootstrapping procedure was employed
METHODS
assemblages, looking at each sweep individually, and described the short
The aim of the entire assemblage analysis was to increase our understanding on random samples composed of 75% of the data and the results compared for persistence and
term variation in assemblages and environmental conditions. Longer term
of the biogeography of fishes and macro-invertebrates in relationship to their precision. The location of the site groups were mapped using GIS. To determine which species
trends were not analyzed. Since the species groups changed with each
environment, and identify important areas or habitats. Four of the five main groups were influential in forming the site groups, the average frequency of occurrence for species
sweep, and most rockfish were grouped together for this study, comparing
assemblages in each site group was calculated. Species assemblages were considered influential
objectives were addressed in this analysis:
30
Subsection 2.1.1: ASSEMBLAGE ANALYSES
results from the two studies is difficult. Calliet et al. 123°W 122°W
(1979) analyzed midwater trawls, and purse seines,
NMFS Midwater Trawls
NMFS Midwater Trawls
to determine species assemblages associated with
Species Assemblages
Myctophid market squid in shallow and deep environments. In
(1998)
Deep-sea Smelt
the results from anchovy hauls, only three species
1998 Trawls
Slender Barracudina
showed high affinity with market squid in both shallow
Pacific Hake, juv.
and deep hauls: northern anchovy, pacific electric ray,
Calif. Smoothtongue
and Pacific herring. This study's results are similar
Euphausiid
since all species assemblages placed market squid,
Stripetail Rockfish, juv.
Site Groups
northern anchovy, and Pacific electric ray together.
Shortbelly Rockfish, juv.
Plainfin Midshipman Pacific herring were not present in this analysis. Site Group A
Market Squid Site Group B
Site Groups and Interaction of Species and Sites
Northern Anchovy
38°N
38°N
Site Group C
Pacific Electric Ray (Objectives B and C)
Pacific Sardine
For 1998 and 1999, six site groups were identified for Site Group D
each year. No analysis was run to group sites from the
Northern Anchovy, larval Site Group E
Pacific Sanddab, juv. entire data set, since preliminary results suggested
Site Group F
Pacific Sanddab, adult no spatial trends in results. Because groups are from
Speckled Sanddab, juv.
Not Slender Sole, adult
in
at a fluentia midwater environments they are not named by depth, 0 10 20
ny d
epth l but distinguished as site group A, site group B, etc. Kilometers
Pictures from FishBase and NMFS
Species assemblages which were influential in form-
Figure 25. Species assemblage results for the midwater trawls conducted in 1998. ing each site group are identified. The average fre-
Assemblages are named for the most influential species in each group. Non-itali-
quency of occurrence of fish species assemblages for
cized species were consistently placed into the same species assemblage >80%
each site group (Tables 7 and 8) is used to determine
of the time; italicized species were more ephemeral and tended to roam into other
where species assemblages were found. Maps with
assemblages with random sampling.
the location of the site groups are provided (Figures
26 and 28). Multiple assemblages can be found at
the same site as 3-4 sweeps were made per year
and environmental conditions could change between
sweeps (see Larson et al, 1995).
50
m
10
Habitat Correlations (Objective D)
Site group A Site group B Site group C Site group D Site group E Site group F 0m
20
Depending on the data set, various environmental
37°N
37°N
0
m
N trawls 15 20 14 5 11 26 conditions had a significant influence on site groups.
In 1998, a warm year, bottom depth (log transformed:
Myctophid
0.00 0.02 0.02 0.00 0.18 0.49 N=91, F=18.94, P=<0.0001) was significant. In 1999,
Assemblage
a colder year, bottom depth (log transformed: N=91,
Pacific Hake, juv. F=23.87, P=<0.0001), latitude (N=91, F=8.76,
0.17
0.40 0.31 0.40 0.76 0.62
Assemblage P=<0.0001), and water density (N=91, F=5.93,
Stripetail Rockfish, P<0.0001) were significant. The significance of bot-
0.13 0.08 0.00 0.18 0.15
0.55
juv. Assemblage tom depth to each analysis highlights the importance
m
of location on species assemblages even in the neritic
00
Market Squid
0.00 0.07
0.33 0.58 0.29 0.30
20
environment. Groups were labeled according to their
Assemblage
mean depth (i.e. the shallowest group was group A
Pacific Sanddab and the deepest was group F). The Pacific sanddab
0.10 0.07 0.07 0.12 0.01
0.27
Assemblage assemblage in 1998 and market squid assemblage in
1999 were only influential in the shallowest site group.
Table 7. Average frequency of occurrence of fish species assemblages (percent 10
20
0 m 0 m 50 m
In 1999 there was a north/south split where group F
occurrence calculated for each species and then averaged for each fish assem-
20
00
was more southern and groups B and C more northern
blage) for each 1998 midwater site group. Number of trawls in each site group is
m
provided in the first row. Bold numbers represent influential species assemblages (except for one point each south of Point Año Nuevo). 123°W 122°W
within that site group. In 1998, there were only two rockfish species captured
in greater than 5% of the trawls (stripetail rockfish as- Figure 26. Location of site groups for NMFS 1998 midwater trawls. Lines showing the 50, 100, 200, and 2,000 depth contours are
provided.
31
Subsection 2.1.1: ASSEMBLAGE ANALYSES
123°W 122°W
semblage) which were only influential in site group
NMFS Midwater Trawls C. In 1999, there were six rockfish species grouped
NMFS Midwater Trawls
Species Assemblages together (canary rockfish assemblage), which were
Pacific Hake, juv.
(1999) influential for site groups C and E. The 1999 rock-
Calif. Smoothtongue
1999 Trawls
fish were caught in trawls further offshore than the
Deep-sea Smelt
Euphausiid trawls that contained rockfish in 1998.
Myctophid Canary Rockfish, juv.
Black Rockfish, juv.
These analyses provide trends in species as-
Medusafish
Bocaccio, juv.
Chilipepper, juv. semblages in the midwater environment during Site Groups
Shortbelly Rockfish, juv. Pacific Hake, adult the upwelling season. The grouping of juvenile
Site Group A
Widow Rockfish, juv.
rockfish together suggests that when conditions
Pacific Sanddab, juv.
Slender Sole, juv.
Site Group B
are suitable for one species, they are also suit-
Blue Rockfish, juv.
38°N
38°N
Dover Sole, juv. able for the other species. Within the entire data Site Group C
Lingcod, juv. set, three species assemblages were consistently
Site Group D
Pacific Argentine, juv.
grouped together greater than 80% of the time:
Pacific Sanddab, adult
Site Group E
Pacific hake, juv. assemblage; canary rockfish, juv.
Speckled Sanddab, juv.
Stripetail Rockfish, juv. Market Squid
assemblage; and spiny dogfish assemblage. Two Site Group F
Northern Anchovy
groups were slightly less stable and were grouped
Not Pacific Electric Ray
influ 0 10 20
together greater than 70% of the time: medusafish
entia
at an Pacific Sardine
l
y de
pth assemblage and market squid assemblage. Even Kilometers
Pictures from FishBase and NMFS
though results were not as stable as with the other
Figure 27. Species assemblage results for the midwater trawls conducted in 1999.
data sets, this analysis identifies assemblages that
Assemblages are named for the most influential species in each group. Non-italicized
were consistent through time. The reduced number
species were consistently placed into the same species assemblage >80% of the time;
of species present in 1998 highlights the effects of
italicized species were more ephemeral and tended to roam into other assemblages
water temperature on species assemblages in the
with random sampling.
neritic environment. For this data set, both bottom
depth, latitude, and water density were found to
have a significant influence on site groups.
50
m
10
0
20
m
37°N
37°N
0
Site group A Site group B Site group C Site group D Site group E Site group F
m
N trawls 23 7 9 21 16 15
Pacific Hake, juv.
0.21 0.37 0.40 0.33 0.69 0.77
Assemblage
Canary Rockfish, juv
0.08 0.02 0.07 0.15
0.25 0.53
Assemblage
Medusafish 0.04 0.00 0.11 0.05 0.13 0.00
m
00
20
Pacific Hake, adult 0.04 0.00 0.00 0.06 0.00
0.38
Pacific Saddab, juv
0.14 0.19
0.40 0.30 0.34 0.27
Assemblage
100
20 50
Market Squid 0 m m
m
0.11 0.06 0.12 0.05 0.03
0.55
20
Assemblage
00
m
Table 8. Average frequency of occurrence of fish species assemblages (percent occurrence calculated for each 123°W 122°W
species and then averaged for each fish assemblage) for each 1999 midwater site group. Number of trawls in
Figure 28. Location of site groups for NMFS 1999 midwater trawls. Lines showing the 50, 100, 200, and 2,000 depth contours are
each site group is provided in the first row. Bold numbers represent influential species assemblages within that
provided.
site group.
32
Subsection 2.1.1: ASSEMBLAGE ANALYSES
SECTION SUMMARY tal conditions, especially seasonal water
0
Twenty-eight species assem- temperature (1998 warm year vs. 1999 cold
Depth Range of Species Assemblages
blages were identified from the year) was obvious in its influence on what
Gopher Rockfish
Black Rockfish
25
CDF&G recreational, NMFS species were present in the neritic environ-
Brown Rockfish
Yellowtail Rockfish
Cabezon
shelf, and NMFS slope data sets. ment during upwelling season. Using data
Canary Rockfish
CDF&G Recreational
China Rockfish
Copper Rockfish
Figure 29 illustrates the overlap Kelp Greenling
from all years, species assemblages could
Hook and Line
Lingcod
Pacific Sanddab
Rosy Rockfish
between the three data sets. The be delineated, but these assemblages were
50 English Sole
Starry Rockfish
NMFS Shelf Trawls
Petrale Sole
length of the vertical line depicts more sensitive with regards to random
Vermilion Rockfish Chilipepper Stripetail Rockfish
Pink Seaperch
Pacific Sanddab Bocaccio Bocaccio
the depth interval where species samples. This emphasizes the ephemeral
Plainfin Midshipman
Greenspotted Cowcod Chilipepper
Lingcod
Rockfish
Blue Rockfish
NMFS Slope Trawls
Greenspotted Rockfish Darkblotched Rockfish
assemblages were "influential" nature of the neritic environment, and the
75 Greenstriped Halfbanded Rockfish
Olive Rockfish Greenstriped Rockfish English Sole
Depth (meters) Rockfish Pacific Mackerel Pacific Hake
Bocaccio Shortbelly Rockfish Greenstriped Rockfish
(see Tables 2, 4, 6-8). Shading resulting transient nature of the species as-
Flag Rockfish Chilipepper Jack Mackerel Longnose Skate Stripetail Rockfish Lingcod
Pacific Argentine
was included to give the impres- semblages. There were three species as-
Spotted Ratfish
Speckled Rockfish Petrale Sole
Pacific Herring
Widow Rockfish Sharpchin Rockfish
Shortspine
sion of where the continental shelf semblages that occurred together in all data
American Shad
Yelloweye Rockfish Shortbelly Rockfish
100 Thornyhead
Big Skate
California Market Squid Spot Shrimp
Bering Skate
California Skate
end and continental slope begins. sets: 1) Market squid, Northern anchovy,
Shelf break Filetail Catshark
Chinook Salmon Slender Sole
Darkblotched Rockfish
Bigfin Eelpout
Spiny Dogfish Black Eelpout
Curlfin Sole Redbanded Rockfish
The edge between the shelf and Bank Rockfish
Pacific electric ray, and Pacific sardine;
Dover Sole Longspine Thornyhead
California
Dungeness Crab Rosethorn Rockfish
Redbanded Rockfish
Sablefish Black Skate
Grenadier
Longspine Combfish
slope, although variable within 2) Euphausiids, Pacific hake, and deep
Pacific Electric Ray
Splitnose Rockfish
Rex Sole Califonia Slickhead
Flapjack Devilfish
325 Northern Anchovy Crimson Pasiphaeid
Slender Sole Pacific Glass
the study area, was presented at sea smelt; and 3) Myctophid and slender
Pacific Pompano Deepsea Sole
Spotted Cusk-eel Blackgill Rockfish Splitnose Shrimp
White Croaker Giant Grenadier
200 meters to be consistent with barracuda. In addition, most larval rockfish
Aurora Rockfish Grooved Tanner Crab
Pacific Electric Ray Rockfish
Black Eelpout Pacific Flatnose
Bering Skate
Williams and Ralston (2002). In species co-occur. More in-depth analyses,
Aurora Rockfish Pacific Grenadier
Blacktail Snailfish Bigfin Eelpout
550 Blackgill Rockfish Snakehead Eelpout
Brown Cat Shark Longnose Skate
all cases, the assemblages with a taking advantage of the available informa-
Twoline Eelpout
Dover Sole
Filetail Cat Shark Pacific Hake
Spiny Dogfish Pacific Viperfish
Lanternfish
low frequency of occurrence at all tion on environmental conditions, could be
Rex Sole
Bank Rockfish Black Hagfish
Spotted Ratfish Deepsea Skate
depths were composed of species conducted (see Larson et al., 1994).
775
Fangtooth
Sablefish Longfin Dragonfish
present in less than 20% of the Blacktail Snailfish
Assemblages with Low Frequency of Occurrence at All Depths Pacific Blackdragon
Brown Catshark
In conclusion, species assemblages and Rhomboid Squid
trawls. For these assemblages, Shortspine Sawtooth Eel
Canary Rockfish Smooth Grenadier
site groups were delineated and mapped Thornyhead
depth associations may have Sharpchin Rockfish California Market Squid
1000 Pacific Mackerel Yellowtail Rockfish Threadfin Slickhead
Spot Shrimp
Quillback Rockfish Arrowtooth Flounder Blackbelly Eelpout Vampire Squid
for four separate data sets. Depth had a
Widow Rockfish
Squarespot Rockfish
been present, just not discernible, Threadfin Sculpin Robust Clubhook Squid
Magistrate Armhook Squid
Rock Sole Sp.
significant influence on all four data sets.
given the methodology for defining Figure 29. Overlap between the three data sets that analyzed demersal fish: CDF&G recreational (yellow), NMFS shelf (green), and NMFS slope (orange). The length of the vertical line
The influence of depth is not a new con-
“influential” assemblages. depicts the depth interval where species assemblages had an average frequency of occurrence in at least 25% of the trawls (see Tables 2,4,6-8). Non-italicized species were consistently
cept (Williams and Ralston, 2002; Sullivan,
placed into the same species assemblage; italicized species tended to roam into other assemblages with random sampling.
Data Sources 1995; Gabriel and Tyler, 1980; Field et al.,
See associated sections for information and spatial extent may have overlapped for the 200-500 meters depth range, more difficult to compare due to the different fishing methods 2002; Matthews and Richards, 1991); however, this is the
of CDF&G recreational hook and line data (pp. 23), NMFS the species included in the analyses differed. It is interesting employed. Only nine species overlap between the recreational first time a study has demonstrated its significance on three
demersal trawls on the continental shelf (pp. 26), and NMFS to note that the shallow water species included in the NMFS data and the shelf trawls. For both data sets, chilipepper, green- separate data sets. All attempts to remove depth and look for
demersal trawls on the continental slope (pp. 28). slope trawl analysis were all placed in one assemblage, the striped rockfish, and greenspotted rockfish were grouped to- secondary influences on group designation were unsuccessful.
stripetail rockfish assemblage, and that this assemblage was gether (chilipepper and greenspotted assemblages) and found For the neritic environment, depth, latitude, and water density
Methods only present in the shallowest slope trawl site group. The same at similar depths. For the most part, fish species were associ- had a significant impact on site groups in 1999. Starr (1998)
Results from the overlap between species assemblages and species included in this stripetail rockfish assemblage were ated with a shallower depth with the recreational hook and line addressed the implementation of rockfish no-take areas and
site groups from each analysis (see Tables 2, 4, 6-8) were used found in five different NMFS shelf assemblages. Conversely, analysis than with the shelf trawl analysis. This difference could made two important recommendations. First, in order to prop-
to determine the depths at which species assemblages were the blackgill assemblage on the NMFS shelf trawls contains be due to the nature of the assemblage analysis or due to the erly manage marine ecosystems, there is a need for a better
present. For each assemblage, the shallowest site group (mean the deeper species caught in the shelf trawls and found within variable size selectivity of the fishing methods and habitats. understanding of fish assemblages. Once these assemblages
depth minus the standard deviation) from which their average three different NMFS slope species assemblages. This does Many of the deepwater rockfish species settle as juveniles in are delineated, managers can take steps to ensure each as-
frequency of occurrence was >25% was used to determine not imply that species co-occurrences changed between the shallow water, and slowly shift to deeper water as they mature semblage receives proper management. The results from this
the minimum depth. Similarly, the mean plus the standard de- shelf and slope trawls, just that cluster results were sensitive to (Love et al., 2002). Future analyses could include information study provide information on these assemblages for near-
viation for the deepest site group was used to determine the the depth range covered by the data set. For example, bocac- on fish total length to determine if ontogenetic shifts occur and if shore, shelf, slope, and midwater ecosystems. The second
maximum depth. cio and English sole do not co-occur, as bocaccio is attracted they generate the differences in species’ depth range between recommendation by Starr (1998) was to delineate rectangular
to rocky ledges and English sole to soft bottom areas with low data sets noted above. The effect water temperature has on the no-take areas that cover 20-50 km of the coast and extend west
Results and Discussion relief (Love et al., 2002). These species are grouped together species present, and the composition of species assemblages, to the edge of the continental shelf. From a biogeographic view-
The shallow assemblages had more limited depth ranges, 92% of the time when included in an analysis with deep slope was investigated for the recreational and shelf data sets and point, the spatial analyses coincide with that recommendation
which is not obvious given the log scale of depth in Figure 29. species, but never grouped together when included in an provides preliminary results on which assemblages are persis- and also determined that deep slope communities contribute
Species included with the three data sets differed, especially analysis with shelf species. significantly to ground fish biogeographic patterns. Because
tent through environmental change (see CD-ROM).
since only species caught in at least 5% of the trawls were assemblages follow bathymetry at the scale of this analysis,
analyzed. Therefore, while the NMFS shelf and slope trawls The overlap between the shelf and recreational trawls was For the midwater trawl data set, the importance of environmen- this approach could protect all demersal species assemblages
identified in this study.
33
Subsection 2.1.1: ASSEMBLAGE ANALYSES
REFERENCES Horn, M.H. and L.G. Allen. 1978. A distributional analysis of California National Marine Fisheries Service. 2001. Endangered Species Act. Williams, E.H. and S. Ralston. 2002. Distribution and co-occurrence
Baraff L.S. and T.R. Loughlin. 2000. Trends and potential interactions coastal marine fishes. J. Biogeog., Vol. 5, pp. 23-42. Section 7 Consultation - Biological opinion on authorization of Gulf of of rockfishes over trawlable shelf and slope habitats of California and
between pinnipeds and fisheries of New England and the U.S. West Alaska and Bering Sea/ Aleutian Islands groundfish fisheries based southern Oregon. Fish. Bull., Vol. 100, pp. 836-855.
Coast. Mar. Fish. Rev., Vol62, pp.1-39. Jay, C.V. 1996. Distribution of bottom-trawl fish assemblages over the on the Fishery Management Plan for the Gulf of Alaska and Bering
continental shelf and upper slope of the U.S. west coast, 1977-1992. Sea/ Aleutian Islands groundfish as modified by amendments 61 Wilkins, M.E., M. Zimmerman, and K.L. Weinberg. 1998. The 1995
Boesch, D.F. 1977. Application of numerical classification in ecologi- Can. J. Fish. Aquat. Sci., Vol. 53, pp.1203-1225. and 70. NMFS-Alaska Region, Protected Resources Division, Silver Pacific west coast bottom trawl survey of groundfish resources: Es-
cal investigations of water pollution. Special Scientific Report No. 77, Spring, MD. timates of distribution, abundance, and length and age composition.
EPA-600/3-77-033. Virginia Institute of Marine Science, Williamsburg, Larson, R.J., W.H. Lenarz, and S. Ralston. 1994. The distribution of NOAA Tech. Memo NMFS-AFSC-89. Seattle, WA. 138 pp.
VA. 114 pp. pelagic juvenile rockfish of the genus Sebastes in the upwelling region National Marine Fisheries Service. 2002. Magnuson-Stevens Act
off central California. CalCOFI Rep., Vol. 35, pp. 175-221. Provisions. http://www.nwr.noaa.gov/1sustfsh/groundfish/proposed_ Wilson-Vandenberg, D., P.N. Reilly, and C.E. Wilson. 1996. Onboard
Brown, S.K., K.R. Buja, S.H. Jury, and M.E. Monaco.2000. Habitat rule.pdf. Federal Register, Vol. 67(8), 1582 pp sampling of the rockfish and lingcod commercial passenger fishing
suitability index models for eight fish and invertebrate species in Lauth, R.R. 2001. The 2000 Pacific west coast upper continental slope vessel industry in northern and central California, January through
Casco and Sheepscot Bays, Maine. N. Amer.J. Fish. Mgmt., Vol. 20, trawl survey of groundfish resources off Washington, Oregon and Cali- Paukert, C.P. and T.A. Wittig. 2002. Applications of multivariate statisti- December 1994. Marine Resources Division Administrative Report
pp. 408-435. fornia: estimates of distribution, abundance, and length composition. cal methods in fisheries. Fisheries, Vol. 27(9), pp. 16-22. No. 96-6. San Francisco, CA. 96 pp.
NOAA Tech. Memo. NMFS-AFSC-120. Silver Spring, MD. 284 pp.
Cailliet, G.M., K.A. Karpov, and D.A. Ambrose. 1979. Pelagic as- Ralston, S. 1998. The status of federally managed rockfish on the Worm, B. and R. A. Myers. 2003. Meta-analysis of cod-shrimp
semblages as determined from purse seine and large midwater trawl Lenarz, W.H., R.J. Larson, and S. Ralston. 1991. Depth distributions U.S. West Coast in Marine harvest refugia for West Coast rockfish: A interactions reveal top-down control in oceanic food webs. Ecology,
catches in Monterey Bay and their affinities with the market squid, of late larvae and pelagic juveniles of some fishes of the California workshop. NOAA-TM-NMFS-SWFSC-255. Pacific Grove, California. 84(1), pp. 162-173.
Loligo opalescens. CalCOFI Rep., Vol. 20, pp. 21-30. Current. Calif. Coop. Oceanic Fish. Invest. Rep., Vol. 32, pp. 41-46. 11 pp.
Yoklavich, M.M., H.G. Greene, G.M. Cailliet, D.E. Sullivan, R.N. Lea,
Carr, M.H. 1991. Habitat selection and recruitment of an assemblage Loeb, V., M. Yoklavich, and G. Cailliet. 1994. California Sea Grant Shannon, C.E., and W. Weaver. 1949. The mathematical theory of and M.S. Love. 2000. Habitat associations of deep-water rockfishes
of temperate zone reef fishes. J. Exp. Mar. Biol. Ecol., Vol. 146, pp. biennial report of completed projects, 1992-1994. Moss Marine Labo- communication. University of Illinois Press, Urbana, IL. 125pp. in a submarine canyon: An example of a natural refuge. Fish. Bull.,
113-137. ratories: La Jolla, CA. Pp. 79-95. Vol. 98, pp. 625-641.
Shaw, F.R., Wilkins, M.E., Weinberg, K.L., Zimmermann, M., and
Chavez, F. P, J. Ryan, S. E. Lluch-Cota, and M. Ñiquen. 2002. From Love, M.S., J.E. Caselle, and W.V. Buskirk. 1998. A severe decline R.R. Lauth. 2000. The 1998 Pacific west coast bottom trawl survey of Yoklavich, M.M., V.J. Loeb, M. Nishimoto, and B. Daly. 1996. Near-
anchovies to sardines and back: Multidecadal change in the Pacific in the commercial passenger fishing vessel rockfish (Sebastes spp.) groundfish resources: estimates of distribution, abundance, and length shore assemblages of larval rockfishes and their physical environment
Ocean. Science, Vol 299, pp. 217-221. catch in the southern California Bight, 1980-1996. CalCOFI Rep., and age composition. U.S. Department of Commerce, NOAA Technical off central California during an extended El Niño event, 1991-1993.
Vol. 39, pp. 180-195. Memorandum NMFS-AFSC-114. Silver Spring, MD. 138 pp. Fish. Bull., Vol. 94, pp. 766-782.
Estes, J.A., M.T. Tinker, T.M. Williams, and D.F. Doak. 1998. Killer
whale predation on sea otters linking oceanic and nearshore ecosys- Love, M. S. 1996. Probably more than you want to know about the Starr, R.M. 1998. Marine harvest refugia for West Coast rock- Yoklavich, M.M, G.M. Cailliet, R.N. Lea, H.G. Greene, R.M. Starr, J.
tems. Science, Vol 282, pp.473-476. fishes of the Pacific Coast. Santa Barbara, Calif.: Really Big Press. fish: A workshop. NOAA-TM-NMFS-SWFSC-255. Pacific Grove, de Marignac, and J. Field. 2002. Deepwater habitat and fish resources
88 pp. California.14pp. associated with a marine reserve: implications for fisheries. Marine
Field, J.M., M.M. Yoklavich, J. deMarignac, G.M. Cailliet, R. N. Lea, Ecological Reserves Research Program Research Results 1996-2001.
and S.M. Bros. 2002. Small-scale analysis of subtidal fish assem- Love, M.S., M.M. Yoklavich, and L. Thorsteinson. 2002. The rockfishes Starr, R.M, J.M. Cope, and L. A. Kerr. 2002. Trends in fisheries and California College Sea Grant Program CD-Rom. www.csgc.ucsd.edu.
blages and associated habitat characteristics off Central California of the northeast Pacific. University of California Press, Berkeley and fishery resources associated with the Monterey Bay National Marine La Jolla, CA. 63 pp.
in Marine Ecological Reserves Research Program: research results Los Angeles, CA. 404 pp. Sanctuary from 1981-2000. California Sea Grant College Program
1996-2001. California College Sea Grant Program CD-Rom, La Jolla, Publication No. T-046. Monterey, CA. 156 pp. Zar, J.H. 1999. Biostatistical Analysis, fourth edition. Prentice-Hall,
CA, www.csgc.ucsd.edu. 16 pp. Mason, J.E. 1995. Species trends in sport fisheries, Monterey Bay, Inc. Upper Saddle River, New Jersey. 80 pp.
California, 1959-86. Mar. Fish. Rev., Vol. 57(1), pp. 1-16. Sullivan, C.M. 1995. Grouping of fishing locations using similarities
Gabriel, W.L. and A.V. Tyler. 1980. Preliminary analysis of Pacific in species composition for the Monterey Bay area Commercial Pas- Zimmerman, M., M.E. Wilkins, K.L. Weinberg, R.R. Lauth, and F.R.
Coast demersal fish assemblages. Mar. Fish. Rev., Vol. 42(3-4), pp. Mason, J.E. 1998. Declining rockfish lengths in the Monterey Bay, senger Fishing Vessel Fishery, 1987-1992. California Department of Shaw. 2001. Retrospective analysis of suspiciously small catches in
83-88. California, recreational fishery, 1959-94. Mar. Fish. Rev., Vol. 60(3), Fish and Game, Marine Resources Technical Report No. 59. Monterey, the National Marine Fisheries Service West Coast Triennial Bottom
pp. 15-28. CA. 37 pp. Trawl Survey. Alaska Fisheries Science Center Processed Report,
Gauch, H.G. Jr. 1982. Multivariate analysis in community ecology. 2001-3. Anchorage, AK. 135 pp.
Cambridge Univ. Press, New York. Matthews, K.R. and L.J. Richards. 1991. Rockfish (Scorpaenidae) as- Turk, T. A., T. Builder, C. W. West, D. J. Kamikawa, J. R. Wallace,
semblages of trawlable and untrawlable habitats off Vancouver Island, R. D. Methot, A. R. Bailey, K. L. Bosley, A. J. Cook, E. L. Fruh, B. H.
Hallacher, L.E. and D.A. Roberts. 1985. Differential utilization of British Columbia. N. Amer. J. Fish. Mgmt., Vol. 11, pp. 312-318. Hormess, K. Piner, H. R. Sanborn, and W. W. Wakefield. 2001. The
space and food by the inshore rockfishes (Sebastes) of Carmel Bay, 1998 Northwest Fisheries Science Center Pacific west coast upper
California. Env. Biol. Fish., Vol. 12(2), pp. 91-110. McGarigal, K., S. Cushman, and S. Stafford. 2000. Multivariate sta- continental slope trawl survey of groundfish resources off Washington,
tistics for wildlife and ecology research. Springer-Verlag, New York. Oregon, and California: Estimates of distribution, abundance, and
Hixon, M.A., B.N. Tissot, and W.G. Pearcy. 1991. Fish assemblages 130 pp. length composition. U.S. Department of Commerce, NOAA Technical
of rocky banks of the Pacific Northwest: final report. Minerals Manage- Memorandum NMFS-NWFSC-50. Seattle, WA. 122 pp.
ment Service, MMS 91-0052. Camarillo, California. 410 pp. Moser, H.G. R. L. Charter, W. Watson, D. A. Ambrose, J. L. Butler, S.
R. Charter, and E. M. Sandknop. 1999. Abundance and distribution of Weinberg, K.L., M.E. Wilkins, F.R. Shaw, and M. Zimmermann.
Hixon, M.A., and B.N. Tissot. 1992. Fish assemblages of rocky banks rockfish (Sebastes) larvae in the Southern California Bight in relation 2002. The 2001 Pacific west coast bottom trawl survey of groundfish
of the Pacific Northwest: supplement. Minerals Management Service to environmental conditions and fishery exploitation. CalCOFI Rep., resources: Estimates of distribution, abundance, and length and age
MMS 92-0025. Camarillo, California. 128 pp. Vol. 41, pp. 132-147. composition. U. S. Dep. Commer., NOAA Tech. Memo. NMFS-AFSC-
128. Seattle, WA. 140pp.
34
Subsection 2.1.2: HABITAT SUITABILITY MODELING
maximum observed across the bathymetric gradient (Table 9)
INTRODUCTION
(Rubec et al., 1999). Resultant values were multiplied by 10
Habitat suitability modeling (HSM) is a tool for predicting the
to scale SI’s by whole integers (0-10), as reclassification of
suitability of habitat for a given species based on known affini-
Map Layers Habitat Suitability Maps
Species Habitat Affinities environmental grids is done using ArcView which does not rec-
ties with environmental parameters. This technique was chosen
ognize decimals. For species that had limited or no trawl data,
for this project to provide a synoptic view of habitat suitability
SI values were developed from bathymetric ranges reported
for specific species as well as assess habitat suitability for 1.0
in the literature (Christensen et al., 1997; Brown et al., 2000).
species assemblages. One HSM technique is termed “habitat S1
0.5 Table 10 displays a sample data matrix generated from literature
suitability index (HSI) modeling”. HSI models are simple math-
=
+
SI value
Bathymetry
0 sources, where presence (1) or absence (0) is coded within the
ematical expressions for calculating a unitless index of habitat
bathymetric classes for each particular species and life stage.
quality as a function of one or more environmental variables. depth category
In this technique, the total number of references that denote
Using GIS, these index values can be mapped and analyzed
presence of the species are summed within each depth class
to portray areas of potential distribution for a species (Brown Substrate
1.0
and then divided by the total number of references examined
S2
et al., 2000) (Figure 30). High-quality habitat may provide high
0.5
to obtain the final SI value. Literature review provided only
carrying capacity and support higher rates of growth, survival,
0 general ranges of species occurrence in relation to bathymetry,
or reproduction for a given species, whereas low-quality or un-
substrate category therefore, classes of 50 m were chosen to confidently develop
suitable habitat may have little or no carrying capacity (Brown et
SI values rather than the 20 m classes used above. Differen-
al., 2000). The HSI methods were adapted from the U.S. Fish
tiating depth ranges for adults and juveniles from literature
and Wildlife Service (USFWS) Habitat Evaluation Procedures 1
n sources was difficult due to lack of data, therefore, only adult
program (USFWS, 1980a, 1980b, 1981) to provide spatially
( ∏ Si )
n
HSI = SI values could be developed using this technique. SI’s for af-
explicit estimates of suitability across the entire study area. It
finities with substrate were also created using this technique.
is important to note that the model results depict potentially
i =1 SI’s for juveniles based on bathymetry were developed using
suitable habitat for a given species and not actual distribution.
NMFS trawl data, when available, or were simply not modeled
This section provides the methodology, results, validation, and Figure 30. Species habitat suitability modeling approach.
where trawl data was limited or absent. Contrasting evidence
interpretation of HSI models developed for selected adult and
exists within the literature that bathymetric preferences can shift
subadult stages of commercially and recreationally important
for many groundfish species based on latitude (PFMC, 1999;
groundfish and invertebrate species. The models are based on etry. Bathymetry was rasterized with 70 m cell size for most HSI Data/SI Development: Initially, suitability index (SI) values
Williams and Ralston, 2002). For the present study, it was as-
species’ affinities to substrate types and bathymetric ranges of the study area for depths to 4810 m. Benthic substrate was for bathymetry and substrate type were developed through
sumed that depth preference was similar regardless of latitude,
(Monaco et al., 1998). mapped from Point Arena in the north to Point Sal in the south literature review and modeled in GIS. During October 2002,
although further exploration into this reported trend is currently
to conform to the latitudinal limits of the study area. Substrates the methodological approach and results were peer-reviewed
underway. Similarly, preference for substrate was assumed to
DATA AND ANALYSES were characterized using 5 classifications: sand, mud, rock, by NMSP and NMFS staff who suggested that, where suf-
be the same throughout each species' range.
Environmental Data: Initially bathymetry, benthic substrate pebble/cobble/gravel, and mud/rock mix. ficient data were available, bathymetry SI values should be
type, and bottom temperature were chosen as the environ- developed using NMFS trawl data. In addition, the panel
requested separate models for adults and juveniles. As a The NMFS trawls were conducted in depths of 50 – 1300 m;
mental data to be included in the models. Although water tem- Species selected for HSM: The primary criteria to select spe-
result, a subset of NMFS trawl data on the shelf (1977-1995) therefore bathymetric SI’s outside this range could not be
perature is an influential factor that affects species distributions cies for which HSMs were developed was their commercial
and slope (1984-1999) for the entire west coast were used to calculated. Depth information within the literature exists for
and movement, several factors led to the exclusion of bottom and ecological importance. In addition, several species were
temperature from final model development: 1) information re- included based on recommendations by staff members from develop SI values for adults and juveniles
garding species associations with bottom temperature were too the NMSP. Overall, 20 species were modeled, 14 of which for most species. SI values for bathym-
general or absent from scientific literature; 2) statistical analyses included models for adult and subadult distribution. Species etry were developed from NMFS trawl
revealed collinearity between bottom temperatures collected with two life stage models include: bocaccio, canary rockfish, data by fitting a polynomial regression 0.12
with NMFS trawl samples and bathymetry; and 3) since most chilipepper rockfish, darkblotched rockfish, longspine thorny- to bathymetric classes and mean spe- 2
mean log abundance = -0.0329 + 0.00122(depth) - 0.0000036(depth)
0.1
of the species modeled are benthic organisms where bottom
Mean log abundance
head, shortspine thornyhead, lingcod, sablefish, Pacific whiting cies abundance (log transformed) (Fig-
0.08
temperature is not highly variable, numerous authors state that (hake), dover sole, english sole, petrale sole, rex sole, and ure 31). Since trawl samples were not
depth is the most significant factor regulating species distribu- Pacific sanddab. Potential adult distributions were modeled for collected in waters less than 50 m, the 0.06
tions (Gabriel and Tyler, 1980; Matthews and Richards, 1991; Dungeness crab, California market squid, blue rockfish, widow bathymetric classes begin at 50 m with 0.04
Yoklavich et al., 2000; Williams and Ralston, 2002). As a result, rockfish, yelloweye rockfish, and yellowtail rockfish. Some of a range of 20 m between classes. The
0.02
water temperature was eliminated as a modeling variable. This these species were chosen to represent species assemblages fitted curve was weighted by sampling
does not preclude using water temperature as a variable for as determined in Section 2.1.1, and mapped to display the effort to account for disproportionate 0
modeling pelagic species, however, more information will have potential distribution of suitable habitats for the assemblage. sample sizes within bathymetric class- 50 70 90 110 130 150 170 190 210 230 250 270 290 310 330 350 370
to be collected to explore their affinities for this variable. Based For example, cluster analyses determined that Dover sole, es. Predicted mean abundance along Depth (m)
on these considerations, bathymetry and substrate data were sablefish, and shortspine thornyhead were commonly captured the curve was then used to calculate
used to map HSI model results. Numerous data sources were in NMFS trawl surveys, indicating a deep water shelf assem- SI’s for each bathymetric class by divid- Figure 31. Polynomial regression curve fit with mean log abundance by categorical bathymetric
ing each mean abundance value by the class for subadult bocaccio.
combined to produce a digital, high resolution map of bathym- blage for these species.
35
Subsection 2.1.2: HABITAT SUITABILITY MODELING
The resulting maps display the potential suitability of cells for each species
Table 9. Example data matrix for calculating bathymetry SI values for subadult bocaccio
taken in NMFS trawl samples (Rubec et al., 1999). based on the strength of their affinities to depth and substrate type in each
cell. The map displays habitat suitability in a unitless index from 0 (unsuitable)
Depth Effort Mean log Predicted mean log HSI
to 10 (highly suitable).
Class (m) (# of samples) abundance abundance (x) (x/xmax)*10
50-69 219 .014 .019 3
Validation of HSI Model: The remaining subsets (i.e. independent data) of NMFS
70-89 361 .029 .035 5
trawl data from the shelf (1998-2001) and slope (2000-2001) were used to as-
90-109 447 .049 .048 7
sess model performance. Mean abundance was calculated for each species
110-129 489 .060 .058 8
130-149 398 .056 .065 9 from these data and superimposed over the predicted HSI values and compared
150-169 252 .100 .069 10 by regressing observed catch data on the predicted HSI values. The statistical
170-189 200 .094 .070 10 results are not intended to be definitive tests of the model, but provide sup-
190-209 213 .065 .069 10 porting evidence for the existence and strength of the relationship between the
210-229 182 .037 .064 9
model predictions and the catch data. It is important to note that these models
230-249 98 .059 .057 8
are based on two independent parameters and are not the definitive predictors
250-269 92 .019 .047 7
of habitat utilization for these species. Fishery-dependent data from CDF&G
270-289 89 .003 .034 5
recreational surveys were also used to validate models for species that had
290-309 74 .008 .018 3
limited trawl information, i.e. species that display affinities for rocky substrates
310-329 98 .003 0 0
330-349 52 0 0 0 (rockfishes) and had poor representation in trawl data. If the model performs
correctly, this validation procedure should demonstrate increasing mean abun-
dance with increasing habitat suitability.
Table 10. Example presence/absence information and SI calculation from scientific
literature.
Integrative Maps: Management plans are often developed for a group of spe-
Depth Class (m)
cies that exhibit similar life history strategies. Selected species assemblages,
100- 150- 200- 250- 300- 350- 400-
as defined in Section 2.1.1, were analyzed and mapped to identify the spatial
Source 50-99 149 199 249 299 349 399 449
A 1 1 0 0 0 0 0 0 distribution of their important habitats. In addition, two analyses were conducted
B 1 1 1 0 0 0 0 0 (page 42) which examine the overlap of highly suitable habitat based on all spe-
C 1 1 1 1 1 0 0 0 cies for which HSI maps were developed. Areas with the most overlap of high
D 1 1 1 1 1 0 0 0
suitability could be considered important habitats for selected groundfish.
Sum 4 4 3 2 2 0 0 0
SI=Sum/Total
ANALYTICAL MAP PRODUCTS
References*10 10 10 7 5 5 0 0 0
As part of the biogeographic assessment, digital data were developed as prod-
ucts from the study. Digital bathymetry and substrate maps were created as
most species outside of this range, but was omitted from modeling and mapping to
ArcView shape and raster files. Maps of these environmental data can be seen
match the depth range associated with the NMFS trawls. Trawls were conducted
on pages 36-37, while digital files are located on the accompanying CD-ROM.
during June to November, thus models for different “seasons” could not be created.
Three representative HSI models are presented: bocaccio (adult and subadult),
Therefore, modeled map surfaces represent species potential distributions for water
Dover sole (adult and subadult), and Dungeness crab (adult). The remaining 31
depths from 50-1300 meters and for the summer and late fall time period. Also,
species' HSI maps are on the CD-ROM. Representative maps displaying habitat
many of the species modeled exhibit inshore/offshore migrations based on habitat
importance based on all HSI models and select species assemblages are also
shifts associated with life history requirements and/or spawning activity. Additional
included. Additional integrative maps for shelf assemblages, all rockfish, and
data will have to be collected to reflect these shifts in abundance and distribution;
all flatfish, are also included on the CD-ROM.
thusly, no attempt was made to model them here.
HSI Results-Mapping: Once SI values were determined for bathymetry and sub-
strate type, these values were inserted into the environmental grids. Once each
species’ suitability indices were derived (either through regression or through the
literature), the values were combined with the bathymetry and substrate map layers
to calculate an index of habitat suitability. The habitat suitability was calculated as
the geometric mean of suitability indices (SI) for the two (n) environmental factors
(Rubec et al. (1999) (Figure 30):
1
n
HSI = (∏ S i ) n
i =1
36
Subsection 2.1.2: HABITAT SUITABILITY MODELING
ABOUT THIS MAP
124°W 123°W 122°W 121°W
Figure 32 displays a bathymetric model of the north/central California study area. Prominent bottom features, such as canyons,
High Resolution
39°N
39°N
seamounts, banks, and other large scale geological formations, are evident at the scale presented.
300
20
50
DATA SOURCES
00
m
0
10
20
Bathymetry
m
m
0
00
NOAA/NOS hydrographic survey data available from the National Geodetic Data Center (NGDC) and Monterey Bay Research
m
m
Institute (MBARI) – multibeam data.
METHODS
Results were calculated from 3 arc second bathymetry (nominally 70 m) derived from NGDC and MBARI data sources. All avail-
able multibeam data were used. Hydrographic survey data (echo soundings) were eliminated from the calculation if it occurred
coincidentally with multibeam information. Vertical and horizontal correction was performed on all data prior to modeling. All
0 10 20 40 60 80 100
data were triangulated and rasterized using Vertical Mapper extension of MapInfo 6.5. Cell size varied throughout the study
38°N
38°N
Kilometers
area, but significant portions were mapped at 70 m2 grid cell resolution.
RESULTS AND DISCUSSION
The study area contains two distinct bathymetric regions. The northern portion of the study area, from Monterey Canyon north-
ward, is characterized by having a broad continental shelf (15-50 km wide), while the southern region has a very narrow shelf
with rapidly increasing depth close to shore. This pattern results in significantly shallower mean depth for Cordell Bank (394.9
m) and Gulf of the Farallones (265.3 m) sanctuaries compared to the mean depth within Monterey’s sanctuary (876.9 m). For
a more detailed description of bathymetric features see the explanation of the study area on page 1.
This map is not intended for navigational purposes. Some areas on this map are created from old or sparse data, and are not
37°N
37°N
necessarily representative of the actual seafloor characteristics.
36°N
36°N
10
20
30
00
00
00
m
200 m
m
50 m
m
35°N
35°N
124°W 123°W 122°W 121°W
Figure 32. Bathymetric map for the north/central California study area. Red lines indicate National Marine Sanctuary
boundaries.
37
Subsection 2.1.2: HABITAT SUITABILITY MODELING
ABOUT THIS MAP mud/rock mix. Within sanctuary boundaries, rocky substrates
124°W 123°W 122°W 121°W
Figure 33 displays distribution of substrate types throughout are distributed predominantly on the shelf, occurring in the ar-
39°N
39°N
the study area, from Point Arena to Point Sal California. The eas near Cordell Bank, Farallone Islands, in many near-shore
Substrate Types substrate is classified into 5 categories: mud, sand, pebbles/ areas, and scattered within Monterey Bay. Outside sanctuary
cobbles/gravel (pcg), rock, and mud/rock mix. boundaries, several large areas of rock are found on the slope
in depths greater than 1200 m. Mixed rock/mud substrate is
DATA SOURCES scarce within the study area, with most occurring southwest of
California Continental Margin Geologic Map Series (Maps 4- Monterey canyon. One large area of mixed rock/mud is present
Legend 6) (Greene and Kennedy, 1989). These maps were originally southwest of the southern Monterey Sanctuary boundary. Areas
created with a 1:250,000 resolution and were used as the containing pebble, cobble, and gravel are found exclusively in
Mud
basemaps for recent revisions and incorporation of new high Monterey’s sanctuary and are generally found within depths of
Sand resolution multibeam data in small portions of the study area. 100-200 m. The majority of sand substrate is found near-shore
Pebbles/Cobble/Gravel in the northernmost and southernmost portions of the study
38°N
38°N
METHODS area, with significant coverage also occurring around Cordell
Rock
Initially, seven maps were developed that displayed substrate Bank. Figure 33 displays the percent coverage of bottom types
Rock/Mud
type and geologic formations throughout California’s coastal within each sanctuary. Other important substrate types exist
and marine environments. The original data were compiled by within the study area, such as near-shore kelp beds, but were
the California Division of Mines and Geology, USGS, and Cali- not included in the map based on their ephemeral distribution
0 10 20 40 60 80 100
fornia Coastal Commission to produce paper maps. Geologists and the limitations associated with development of bathymetric
Kilometers
from California State University-Monterey Bay digitized these SI’s in near-shore areas.
maps in 1999 and further interpreted these data to develop
boundaries of substrate types (Greene et al., 1999). Three of Although the substrate map is a probabilistic map of substrate
the seven maps provide data for the study area and together types, it reflects the most complete and current knowledge of
37°N
37°N
provide the most comprehensive map of substrate type for the benthic substrates for the north/central California region. The
north/central California marine and near-shore region. For a map alone can be used to support investigations that require
detailed description of the development of the original maps advanced knowledge of sea floor type. In addition, the maps
and classification scheme, refer to Greene et al., 1999 and are useful identifications of critical or important habitats for par-
Greene et al., 2002. Eight substrate types were classified, how- ticular species or for determining essential fish habitat that can
ever, some were grouped together when the digital substrate aid conservation and management plans for fisheries species.
shapefile was rasterized into 1 km2 grid cells to facilitate their In this study, the map was primarily used as an environmental
use in the HSI model analyses (See Section 2.1.2). “Boulders” layer in GIS (in addition to bathymetry) to determine habitat
and “Hard/Anthropogenic” polygons were grouped within the suitability for groundfish species. This approach assumes that
“Rock” substrate type, and “Gravel" was grouped with “Cobbles/ the underlying environmental GIS layers are an accurate rep-
36°N
36°N
Pebbles” because boulder and gravel polygons were limited in resentation of that particular variable, thus the model results
number and affinities to these were considered to be similar to are only as good as the underlying digital information. The
the group it was reclassified with. substrate map is conservative, based on its original scale (1:
250,000), and may have fine scale inaccuracies throughout
RESULTS AND DISCUSSION the study area. The majority of this map has not been field
The substrate map covers an area of approximately 44,000 tested; therefore, inaccuracies in classification may exist. For
Gulf of the Farallones
Cordell Bank Monterey Bay km2, of which mud accounts for 86.4% (38,023 km2) of the example, several small polygons classified as rock near Point
total bottom area. Substrate containing pebble/cobble/gravel Reyes have been questioned. HSM results presented herein
were the least abundant substrate types, only encompass- do not contain these polygons due to depth limitations with the
6% 1% 4%
4% 7%
15% 19%
ing 100 km2 of the study area. Rock substrates were mostly fish and invertebrate catch data (50-1300 m). Small localized
35°N
35°N
patchy throughout the region, encompassing 1,561 km2. Mud areas of high resolution information (on the scale of 10’s of
and rock mixed substrate (1,706 km2) was almost exclusively meters) have been included in this map, however, these areas
81% 89%
74% distributed in the southern portion of the study area, with small comprise a small percentage of the overall study area. More
localized areas near Monterey canyon. Sand (2,611 km2) is information is required to test the accuracy of the map; hence,
predominantly located near-shore with a large area located thematic accuracy of substrate types is unknown.
rock pcg mud sand
near and around Cordell Bank. Within sanctuary boundaries,
Abbreviation: pcg= pebble/cobble/gravel
soft sediments (mud, sand) account for almost 95% of the
124°W 123°W 122°W 121°W
substrate. Rocky areas (including pebbles, cobbles, gravel)
Figure 33. Substrate types for the north/central California marine region. account for approximately 5%, while less than 1% consists of
38
Subsection 2.1.2: HABITAT SUITABILITY MODELING
ABOUT THESE MAPS
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Figure 34 displays HSI model results for adult (left) and subadult (right) bocaccio during June-
November. The maps exhibit the potential distribution of the species based on affinities to ba-
Bocaccio Bocaccio
39°N
39°N
39°N
thymetry and substrate. Predicted HSI values range in scale from 10 (highest) to 0 (unsuitable)
and were grouped into five classes: highest suitability (10-8), moderate (7-5), low (4-2), lowest
Subadult
Adult
(1), and unsuitable (0). SI values for bathymetry and substrate type are shown in the graphics
below the mapped HSI results. Model performance graphics and statistical details are displayed
in the map insets.
DATA SOURCES
HSI Results HSI Results
38°N
38°N
38°N
Bathymetry SI: Alverson et al., 1964; Feder et al., 1974; Dark et al., 1983; Gunderson and
Highest
Highest Sample, 1980; Tagart and Kimura, 1982; Eschmeyer et al., 1983; Allen and Smith, 1988; Love
Moderate Moderate et al., 1990; Wolotira et al., 1993; Wilkins et al., 1998; Yoklavich et al., 2000; Lauth, 2001; and
Love et al., 2002.
Low Low
Substrate SI: Feder et al., 1974; Eldridge, 1994; Yoklavich et al., 2000; and Love et al., 2002.
Lowest Lowest
Validation: Wilson-Vandenberg et al., 1996; Wilkins et al., 1998; and Turk et al., 2001.
Unsuitable Unsuitable
Life stage information: Love et al., 2002.
37°N
37°N
37°N
0 25 50 Km 0 25 50 Km
METHODS
Bathymetry SI values for adult bocaccio were developed using the literature review method,
whereas subadult SI values were assigned based on the regression fitting technique using NMFS
trawl data.
RESULTS AND DISCUSSION
36°N
36°N
36°N
Length at maturity information (Love et al., 2002) was used to determine life stage for bocaccio.
Adults were defined as: females >360 mm and males >350 mm total length. Depth suitability for
Validation - CDFG Rec. Data Validation - CDFG Rec. Data subadults was highest from 90-270 m, while highest suitability for adults was similar, ranging from
1
1.2
50-299 m (Figure 34). Literature sources indicate that adult bocaccio are almost exclusively found
r2=0.90
2
r =0.70
p=0.0001
p=0.0046
around rocky substrates, while subadults exhibit broader affinity among substrate types (Figure
1 0.8
Mean Abundance
Mean Abundance
34). Comparison of the two HSM maps show that the marked difference in substrate preference
0.8
0.6
for adults yields a more limited spatial distribution than subadults. Less than 5% of the available
0.6
35°N
35°N
35°N
habitat within each sanctuary was predicted highly suitable (HSI values >8) for adult bocaccio
0.4
0.4
(Cordell Bank – 4.6%, Gulf of Farallones – 2.9%, Monterey – 1.7%) Within the study area, habitat
0.2
of high suitability occurs exclusively inside sanctuary boundaries. High suitability covers more
0.2
area for subadults than adults and extends well beyond sanctuary boundaries. Nearly 10% of
0
0
0 2 4 6 8 10
Cordell Bank’s sanctuary was considered highly suitable for subadults. This percentage drops to
0 2 4 6 8 10
HSI Value HSI Value
1.6% for Gulf of Farallones, and 2.6% for Monterey. Approximately 556 km2 of potential high suit-
able habitat was located within the three sanctuaries, while an additional 355 km2 were predicted
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
outside sanctuary boundaries. Although the proportion of highly suitable habitats were similar for
Sediment
Bathymetry
Bathymetry Sediment
adults and subadults, large areas of potentially moderate suitability for subadults were observed
10 10
throughout the study area; whereas no areas were predicted moderate for adults.
8
8
Generally, subadult bocaccio are more commonly found in shallower waters than adults (Love
6 6
SI Values
et al., 2002). Current scientific literature does not provide enough information to develop depth
SI Values
SI values for subadults; therefore, limited trawl information was used to develop SI values for
4 4
bathymetry. Despite this, model performance for subadults yielded a strong positive correlation
2 2
between observed abundance estimates from CDFG recreational catch data and predicted suit-
ability (see map inset). Model performance for adult bocaccio also exhibited a strong positive
0 0
Sand Mud Rock
0 - 49
0-49
50-69
70-89
90-109
Pebble
50 - 299
300 - 399
400 - 599
600 - 1300
110-129
130-149
150-209
210-229
230-249
250-269
270-289
290-309
310-1300
Sand Mud Rock
Pebble
correlation between predicted suitability and CDFG catch data. More information regarding bocac-
Cobble
Cobble
Gravel
Gravel
cio life history requirements are necessary to strengthen the HSI models; however, the mapped
results and validation based on currently available information provide an adequate delineation
Figure 34. Potential distribution of habitat suitability for adult and subadult bocaccio. Map inset contains validation statistics. SI values for bathymetry and substrate are of potential habitat suitability for adult and subadult bocaccio.
graphically displayed below the map.
39
Subsection 2.1.2: HABITAT SUITABILITY MODELING
ABOUT THESE MAPS
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Figure 35 displays the HSI model results for adult (left) and subadult (right) Dover sole during
June-November. The maps exhibit the potential distribution based on affinities to bathymetry
39°N
39°N
39°N
39°N
Dover sole Dover sole and substrate. Predicted HSI values range in scale from 10 (highest) to 0 (unsuitable) and
were grouped into five classes: highest suitability (10-8), moderate (7-5), low (4-2), lowest
(1), and unsuitable (0). SI values for bathymetry and substrate type are shown in the graph-
Adult Subadult ics below the mapped HSI results. Model performance graphics and statistical details are
displayed in the map insets.
DATA SOURCES
HSI Results HSI Results
38°N
38°N
38°N
Bathymetry SI: Wilkins et al., 1998 and Lauth, 2001.
Highest Highest Substrate SI: Demory, 1975; Demory et al., 1976; Barss et al., 1977; Pearcy, 1978; NOAA,
Moderate Moderate 1990; Stein et al., 1992; and CDFG, 2002.
Validation: Wilkins et al., 1998 and Turk et al., 2001.
Low Low
Life stage information: PFMC, 1999.
Lowest Lowest
Unsuitable Unsuitable
METHODS
37°N
37°N
37°N
Bathymetry SI values for adults and subadults were developed from the regression fitting
0 25 50 Km 0 25 50 Km technique. Substrate SI values were developed through literature review.
RESULTS AND DISCUSSION
Adult Dover sole are reported to be >300 mm total length for male and female individuals
(PFMC, 1999). Both adult and subadult Dover sole inhabit deep water slope habitats; sub-
adults exhibited a shallower range of depth preference (130-650 m) than adults (290-1070
m) (Figure 35). Adults and subadults prefer soft sediments (sand and mud) throughout their
36°N
36°N
36°N
range. Highest habitat suitability for subadults was predicted to occur along the shallower
portions of the continental slope (200-550 m). A large area of moderate suitability was also
Validation - NMFS Trawl Data Validation - NMFS Trawl Data
1.5
predicted for an area that extends throughout the majority of the continental shelf. The most
2
r2=0.79
2
r =0.90
suitable habitats for adults consisted of deeper slope waters, with only moderate suitability
p=0.0005
Mean Abundance
Mean Abundance
1.2
p=0.0001
1.5
extending onto the shelf region. Within Cordell Bank sanctuary, high subadult suitability (val-
0.9
ues 8-10) was calculated for 22% of the available habitat, 6.4% within Gulf of Farallones, and
1
19% within Monterey sanctuaries. Cordell Bank and Gulf of the Farallone sanctuaries are
35°N
35°N
35°N
0.6
comprised of shallower (50-300m) shelf waters, thus the percentage of highly suitable habitat
0.5 0.3
for adults is lower (based on their calculated affinity for deeper waters) than that observed
for subadults (21% and 12%, respectively). However, Monterey’s sanctuary is considerably
0
0
deeper and a larger proportion of available habitats (30%) were predicted to be highly suitable
0 2 4 6 8 10
0 2 4 6 8 10
HSI Value HSI Value for adults. Approximately 50% of areas that were predicted to be potentially suitable habitats
for both adults and subadults occurred outside of sanctuary boundaries. These areas are
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
most prominent south of Monterey’s sanctuary.
Sediment
Bathymetry
Bathymetry Sediment
10 10
Model performance was assessed by regressing predicted HSI values on mean log abundance
values from NMFS trawl samples (1998-2001). Significant positive correlations were observed
8
8
for both adult and subadult models, however, these are based on limited trawl samples (N =
6 6
311). Discrepancies in model performance, such as small peaks of mean abundance within
SI Values
SI Values
low suitability areas, are a result of limited observations within that category. Additional trawl
4 4
information would strengthen model development and performance.
2 2
0 0
0-49
50-89
90-129
0-49
Pebble
50-69
70-109
130-169
170-209
210-239
290-489
490-549
550-609
610-649
650-669
670-709
710-729
730-769
770-789
790-809
810-1300
110-149
150-189
190-249
250-289
290-349
350-429
430-529
530-829
830-929
930-1009
1010-1069
1070-1109
1110-1149
1150-1209
1210-1249
1250-1300
Sand Mud Rock
Pebble Sand Mud Rock
Cobble
Cobble
Gravel
Gravel
Figure 35. Potential distribution of habitat suitability for adult and subadult Dover sole. Map inset contains validation statistics. SI values for bathymetry and
substrate are displayed below the maps.
40
Subsection 2.1.2: HABITAT SUITABILITY MODELING
ABOUT THIS MAP
124°W 123°W 122°W 121°W
This map displays HSI model results for adult Dungeness crab during June-November (Figure 36). The map displays the potential distribution based on affinities to ba-
thymetry and substrate. Predicted HSI values range from 10 (highest) to 0 (unsuitable) and were grouped into five classes: highest suitability (10-8), moderate (7-5), low
Dungeness crab
39°N
39°N
(4-2), lowest (1), and unsuitable (0). SI values for bathymetry and substrate type are shown in the graphics below the mapped HSI results. Model performance graphics
and statistical details are displayed in the map insets.
Adult DATA SOURCES
Bathymetry SI: Wilkins et al., 1998 and Lauth, 2001.
Substrate SI: Pauley et al., 1989; Emmett et al., 1991; Leet et al., 2001; and CDFG, 2002.
Validation: Wilkins et al., 1998 and Turk et al., 2001.
HSI Results
38°N
38°N
Highest METHODS
Moderate Bathymetry SI values for adult Dungeness crab were developed using the regression fitting technique. Substrate SI values were developed through literature review.
Low
RESULTS AND DISCUSSION
Lowest
Only adults were modeled within the study area because size information was lacking for crabs in the NMFS trawl data and scientific literature was not detailed enough
Unsuitable
to develop SI values for subadults. Dungeness crabs are an estuarine dependent species (Pauley et al., 1989), with adults exhibiting a shallow distribution (to 90 m) in
37°N
37°N
coastal marine waters. Depth SI values derived from NMFS trawls confirmed this trend by exhibiting high SI values within 50-90 m. Suitability is probably high in the
0 25 50 Km
shallower near-shore environment (Emmett et al., 1991); however, trawl information was not available for this area. Literature sources described crab substrate prefer-
ence to be soft sediments, with occasional utilization of rocky substrate. Habitat suitability based on these data resulted in a broad area of high suitability throughout the
shallower waters of the Gulf of Farallones sanctuary (38% of available habitat), and much smaller proportions within Cordell (8.7%) and Monterey (10.4%) sanctuaries.
Overall, this amounts to 2,809 km2 of highly suitable habitat within the three sanctuaries. Moderate suitability, encompassing approximately 2,477.8 km2, extends further
offshore to approximately 130 m. The potential suitability of habitats rapidly declines to unsuitable beyond 130 m in depth. The model performed well with NMFS valida-
tion data and exhibited a strong positive correlation with predicted suitability values.
36°N
36°N
Validation - NMFS Trawl Data
0.5
r2=0.89
p=0.0014
Mean Abundance
0.4
0.3
35°N
35°N
0.2
0.1
0
0 2 4 6 8 10
HSI Value
124°W 123°W 122°W 121°W
Bathymetry Sediment
10
8
6
SI Values
4
2
0
0-49
50-69
70-89
90-109
110-129
130-149
150-1300
Sand Mud Rock
Pebble
Cobble
Gravel
Figure 36. Potential distribution of habitat suitability for adult dungeness crab. Map inset contains vali-
dation statistics. SI values for bathymetry and substrate are graphically displayed below the map.
41
Subsection 2.1.2: HABITAT SUITABILITY MODELING
or other (0). All maps were overlain and summed to create a map of suitability overlap
within the study area. These areas represent potential groundfish hot spots.
124°W 123°W 122°W 121°W
124°W 123°W 122°W 121°W
RESULTS AND DISCUSSION
HSI Model
Mean HSI Model
39°N
39°N
39°N
The techniques described above are two possible approaches to estimate potential
hot spots or areas of habitat importance. Composite maps displaying these areas were
Species Overlap developed using all fish HSI model results to simulate the groundfish management
strategy employed by NMFS, where all groundfish (83 species) are managed under
HSI Results one Fishery Management Plan. Mean HSI values across all 32 fish species and
life stages yield no areas ranked as highly suitable (HSI values 10-8). Moderate
Highest
# of species suitability (7-5) occurs over the majority of the shelf region (to approximately 200
Moderate
38°N
38°N
38°N
m) throughout the study area, most notably in the northern portion, where the shelf
15 - 20
Low extends significantly farther offshore than in the southern portion. The majority of the
12 - 14 area north of Monterey canyon consists of moderate suitability (to approximately 200
Lowest
8 - 11 m), with low suitability extending through the deeper slope habitat. Smaller localized
Unsuitable
areas of low suitability exist within the shelf and represent areas of hard substrate.
1-7
South of the Monterey canyon, low suitability comprises most of the study area, with
0
0 25 50 Km
a narrow zone of moderate suitability along the shallower shelf waters. Suitability
drops from moderate to low just beyond the shelf edge throughout the study area.
37°N
37°N
37°N
0 25 50 Km
Throughout the study area, maximum overlap of cumulative high suitability occurs
on the shelf edge over soft sediments, which closely contour the 100 m isobath.
Approximately half of the models overlap in this zone. The top two quintals encompass
most of the shelf region and the zones of overlap are much broader in the northern
portion of the study area compared to the southern.
36°N
36°N
36°N
These analyses reveal patterns of suitability related to depth and substrate. Highest
suitability occurs on the continental shelf, over soft sediments, based on the two
analytical approaches using the 33 HSI maps. These areas could be considered
as habitats of importance that support fish abundance and diversity. Both methods
portray highest suitability over the shelf that decline beyond the shelf edge. This
pattern conforms to literature sources which state that the shelf, and more importantly
the shelf break, are important areas for fish abundance and diversity (Yoklavich et
35°N
35°N
35°N
al., 2000; Williams and Ralston, 2002). In addition, soft sediments are potentially
more suitable than hard bottom throughout the study area. It is important to note that
the results of these analyses are based on 19 species and are only a subset of the
many groundfish species that occur within the study area. These results are clearly
biased based on the species modeled and may not provide adequate representation
of groundfish as a whole within the study area. Most of the species modeled have
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
substrate affinities for soft sediments, and most exhibit depth preferences that fall
Figure 37. Areas of groundfish potential hot spots based on mean fish species HSI models and overlap of predicted highly suitable habitats. within the shelf region. Ideally, many more models should be developed for additional
species and analyzed to provide a more representative depiction of groundfish
distribution within the study area.
ABOUT THESE MAPS METHODS
These maps display the results of two approaches that provide a synoptic view of Mean HSI: HSI maps for all fish species and life stages were overlain and averaged
overall habitat importance based on all the HSI models developed for the study area. by grid cell to evaluate overall suitability. Results were scaled in the same manner
Areas of potential habitat importance were first defined as an average view of habitat as individual HSI model results: Highest suitability (10-8), moderate (7-5), low (4-2),
suitability across species and life stages. Secondly, individual maps of highly suitable lowest (1) and unsuitable (0).
habitats were overlain and areas or regions with the most overlap were considered
important habitat or hot spots (Figure 37). Cumulative Suitability: Frequency of occurrence of predicted HSI values for each
fish and invertebrate species life stage were calculated and values greater than one
DATA SOURCES standard deviation above the mean were chosen to represent highest suitability. New
See Individual HSI model results – CD-ROM. individual maps were created and grid cells were reclassified as highest suitability (1)
42
Subsection 2.1.2: HABITAT SUITABILITY MODELING
ABOUT THESE MAPS
124°W 123°W 122°W 121°W
124°W 123°W 122°W 121°W
The maps provide one approach to assess habitat suitability based on HSI results for multiple
species (Figure 38). HSI model results were averaged to assess the potential distribution of
All Rockfish Slope Assemblage
suitable habitats for 8 species of adult rockfish (left) and 3 adult slope species (right). Predicted
39°N
39°N
39°N
HSI values range in scale from 10 (highest) to 0 (unsuitable). HSI results were grouped into five
classes: highest suitability (10-8), moderate (7-5), low (4-2), lowest (1), and unsuitable (0).
Adult Adult DATA SOURCES
Adult rockfish map – HSI maps for adult bocaccio, chilipepper, darkblotched, canary, yellowtail,
yelloweye, and widow rockfishes (CD-ROM).
HSI Results HSI Results Slope assemblage adults – HSI maps for adult Dover sole, sablefish, and shortspine thorny-
38°N
38°N
38°N
Highest
Highest head (CD-ROM).
Moderate
Moderate
METHODS
Low
Low
The slope assemblage was determined through cluster analysis of NMFS benthic slope trawl
Lowest
Lowest
data (see Section 2.1.1 for methodology). All models for adult rockfish were combined to evalu-
Unsuitable Unsuitable ate habitat suitability for these species as an assemblage. Both assemblages of fishes were
analyzed by overlaying each individual HSI map and calculating the arithmetic mean across
37°N
37°N
37°N
grid cells.
0 25 50 Km 0 25 50 Km
RESULTS AND DISCUSSION
Typically, management plans are not based on single species, but rather groups of species that
exhibit similar life histories (Williams and Ralston, 2002). Estimating potential distributions for
species assemblages from HSI models could be a valuable tool for resource managers to aid
in the development of fishery management plans and conservation strategies. This approach
36°N
36°N
36°N
provides a spatial view of important habitats for a given assemblage and generates a baseline
set of data which can be used for a variety of management needs.
Individual HSI results for the 8 species of rockfishes displayed similar patterns of habitat suit-
ability within the study area and, not surprisingly, the map of mean habitat suitability for these
species is nearly identical to the individual maps. Hard substrates (pebble, cobble, gravel, rocky)
within the shelf region promote highest suitability areas for these species. Moderate suitability
35°N
35°N
35°N
was predicted for areas with mixed mud/rock substrate and mud areas in waters with depths
between 200-450 m. These areas are emphasized based on HSI results from darkblotched
rockfish, which exhibited strong affinity for hard and soft substrates, rather than only rocky
substrate preference exhibited by the other rockfish species. Also, darkblotched rockfish distri-
bution occurs in deeper waters compared to the other species of rockfish and may necessitate
their omission from this assemblage. Regardless, suitable habitat for this group of species is
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
limited, based on the distribution of rocky substrate within the study area. Overall, highly suit-
Species: Species: able habitat comprises 364 km2 within the three sanctuary boundaries or 2% of the available
Species: Species:
Blue rockfish habitat. Moderate suitability comprises even less area, 247 km2, or 1.3% of available habitat.
Dover sole
Blue rockfish Dover sole
Bocaccio
Sablefish
Bocaccio Sablefish Cluster analysis of NMFS trawl data revealed many assemblages of species that tend to occur
Chilipepper rockfish
Shortspine thornyhead together (see Section 2.1.1). Dover sole, sablefish, and shortspine thornyhead were identified
Chilipepper rockfish Shortspine thornyhead
Darkblotched rockfish as members of a strong species assemblage that occurs over soft sediments in deep waters of
Canary rockfish rockfish
Darkblotched
the continental shelf and slope. Mean HSI calculations resulted in a broad range of highly suit-
Yellowtail rockfish
Canary rockfish able habitats throughout the study area. Overall, 23% (4,257 km2) of the available habitat within
Yellowtail rockfish
Yelloweye rockfish the sanctuaries was predicted to be highly suitable habitat for this assemblage. An additional
Yelloweye rockfish
Widow rockfish 10% (1,907 km2) was predicted moderately suitable habitat. Because of its greater depths and
larger area, Monterey Bay sanctuary contained significantly more highly suitable habitat (3,773
Widow rockfish
km2) than Cordell Bank (170 km2) and Gulf of the Farallones (312 km2) sanctuaries.
Figure 38. Areas of potential habitat importance based on mean HSI models for selected species assemblages.
43
Subsection 2.1.2: HABITAT SUITABILITY MODELING
distribution), and others can be developed and
SECTION SUMMARY
incorporated into the model as needed.
HSI modeling and mapping were considered to be a compo-
nent of the biogeographic assessment because this approach
Decision making processes are typically not ad-
provides spatial species- and lifestage-specific information for
dressed at the species level but rather at a multi-
the north/central California marine region. This approach is in-
species assemblage level. Thirty-four HSI models
tended to serve as an analytical tool for resource managers that
were created for 18 fish and 2 macro-invertebrate
can address a variety of needs: 1) developing maps in poorly
species to support multi-species analyses or as-
sampled areas, 2) evaluating impact scenarios, 3) identifying
sessments. Several techniques were conducted
habitats or areas for conservation or protection, and 4) assess-
to assess habitat quality within the entire study
ing impacts of environmental change. The approach used here
area. One result indicated that the most potentially
is similar to previous efforts that mapped near-shore rockfish
suitable habitat occurred on the shelf over mud
distributions (Wright et al., 2000). The maps displayed near-
substrates within depths of 100-120 m. As previ-
shore rockfish distributions in relation to latitude, and maximum
ously mentioned, these results were biased based
and common bathymetric ranges, based on information from
on the selection of species modeled; however,
peer-reviewed literature. The products generated from this
the technique provides one method to identify
study expand on this approach by including an additional pa-
areas of potential high habitat quality. Additional
rameter (substrate type). Also, models were developed which
analyses identified important habitats for select
predicted the potential spatial distribution (based on affinities
species assemblages. Habitat suitability models
for bathymetric ranges and substrate type) for a select group
for an assemblage of rockfish were developed and
of groundfish species. The maps provide a unique spatial view
indicated that rocky habitats located on the shelf
of potential groundfish habitats within and outside central Cali-
were identified as potential hot spots for adults;
fornia sanctuary boundaries.
whereas, mud and sand substrates on the shelf
were delineated as potentially important habitats
It is important to note that the model results previously de-
for subadult rockfish.
scribed are not actual, but potential distributions based on
species affinities to the environmental variables used in the
In conclusion, the HSI maps can be used in a
models. Interpretation of these results should be conducted
broad range of assessments which require in-
carefully due to the variety of limitations associated with the
formation on habitat distribution and suitability.
biological and environmental data. Both bathymetry and sub-
Individual species maps can be used to identify
strate-type maps were created from the most current informa-
areas of varying habitat suitability and can be used
tion available; however, the scale of information may cause
to assess sensitivity to environmental or anthro-
inaccuracies in the interpretation of the model results. The
pogenic impacts. Lastly, it is recommended to
bathymetry map, created with 78 million data points, provides
continue developing HSI models for remaining
a high quality, high resolution image of depth throughout the
groundfish species, as those presented here are
study area. The digital substrate map is a probabilistic inter-
REVIEWERS Chris Harvey, Northwest Fisheries Science Center, NMFS
only a small subset of the available resources
pretation of imagery data that has yet to be field validated.
Tara Anderson, University of California, Santa Cruz Ruth Howell, Gulf of the Farallones National Marine Sanctuary
within the study area.
Given its original resolution (1:250,000), the map may under
Carol Bernthal, Olympic Coast National Marine Sanctuary Program
or overestimate substrate distribution within the study area;
Program Roxanne Jordan, Alliance of Communities for Sustainable
REVIEWS
however, localized areas have more accurate information.
Tonya Builder, Northwest Fisheries Science Center, NMFS Fisheries
Two reviews were completed for the fish assemblage and
For example, predicted areas of potential high suitability were
Gregor Cailliet, Moss Landing Marine Laboratory Chad King, Monterey Bay National Marine Sanctuary Pro-
habitat suitability analyses. During May and June 2002 in-
extremely limited for species that exhibit strong affinities for
Mark Carr, University of California, Santa Cruz gram
formal meetings were held in Monterey, San Francisco, and
rock substrate (rockfish, lingcod). Generally, less than 1% of
Josh Churchman, Fisherman Howatt King, California Department of Fish and Game
Seattle to receive feedback on the approach and verify from
the study area was considered optimal habitat for these spe-
Elizabeth Clarke, Northwest Fisheries Science Center, Bob Lauth, Alaska Fisheries Science Center, NMFS
the scientists that collected the data that the analyses were
cies and may underestimate actual habitat distribution. These
NMFS Phil Levin, Northwest Fisheries Science Center, NMFS
valid. Formal review workshops were held in October, 2002 in
results are reflective of the low percentage of rock substrate
Steve Copps, Northwest Region, NMFS Steve Lonhart, Monterey Bay National Marine Sanctuary
San Francisco, Seattle, and Monterey Bay and hosted local
included in the substrate map, which could be a result of the
Brad Damitz, Monterey Bay National Marine Sanctuary Pro- Program
scientists, fishermen, and National Marine Sanctuary Program
scale in which the original data were collected. Nevertheless,
gram David Lott, Monterey Bay National Marine Sanctuary Pro-
staff. Review comments were either incorporated or addressed
the map provides the most comprehensive substrate inventory
Kathey Fosmark, Alliance of Communities for Sustainable gram
in this product. We appreciate all the reviewers’ time and effort
for this region and it is recommended that additional substrate
Fisheries Huff McGonigal, Monterey Bay National Marine Sanctuary
when providing us this important feedback.
information be collected to further refine the maps. Additional
Jim Glock, Sustainable Fisheries Division, NMFS Program
digital data are available (e.g. sea surface temperature, kelp
Gary Greene, Moss Landing Marine Laboratory Nazila Merati, Pacific Marine Environmental Laboratory,
NOAA
44
Subsection 2.1.2: HABITAT SUITABILITY MODELING
Richard Methot, Northwest Fisheries Science Center, NMFS Feder, H.M., C.H. Turner, and C. Limbaugh. 1974. Observations on fishes associ- NOAA/NOS. 1990. West Coast of North America strategic assessment: Data atlas.
Jim Nybakken, Moss Landing Marine Laboratory ated with kelp beds in southern California. Cal. Dept. Fish and Game, Fish. Bull. Invertebrate and fish volume. Pre-publication edition. Strategic Assessment Branch,
John Pearse, University of California, Santa Cruz Vol. 160, pp. 2-135. NOAA/NOS, Rockville, MD. 112 pp.
Stephen Ralston, Southwest Fisheries Science Center, NMFS
Fishbase. 2002. Species summaries (online). http://www.fishbase.org. Phillips, J.B. 1964. Life history studies on ten species of rockfish (genus Sebastes).
Paul Reilly, California Department of Fish and Game
Cal. Dept. Fish and Game, Fish Bull., Vol. 126, p. 70.
Dale Roberts, Cordell Bank National Marine Sanctuary Program
Fitch, J.E., and R.J. Lavenberg. 1971. Marine Food and Game Fishes of California.
Teresa Turk, Northwest Fisheries Science Center, NMFS
California Natural History Guides, Vol. 28. Berkeley: Univ. California Press, Berkeley, Rubec, P.J., J.C. Bexley, H. Norris, M.S. Coyne, M.E. Monaco, S.G. Smith, and J.S.
Tiffany Vance, Alaska Fisheries Science Center, NMFS
CA. 98 pp. Ault. 1999. Suitability modeling to delineate habitat essential to sustainable fisheries.
Mark Wilkins, Alaska Fisheries Science Center, NMFS
Am. Fish. Soc. Symp., Vol. 22: pp. 108-133.
Deb Wilson-Vandenberg, California Department of Fish and Game
Gauch, H.G., Jr. 1982. Multivariate analysis in community ecology. Cambridge Univ.
Lisa Wooninck, Southwest Fisheries Science Center, NMFS
Press, New York, NY. 145 pp. Shaw, W.N., and T.J. Hassler. 1989. Species profiles: life histories and environmental
Nancy Wright, California Department of Fish and Game
requirements of coastal fishes and invertebrates (Pacific Northwest) – Lingcod. U.S.
Levon Yengoyan, TerraLogic GIS Lassuy, D.R. 1989. Species profiles: life histories and environmental requirements of Fish Wildl. Serv. Biol. Rep. 82(11.119), U.S. Army Corps Eng. TR EL-82-4, 10 p.
Mary Yoklavich, Southwest Fisheries Science Center, NMFS coastal fishes and invertebrates (Pacific Northwest) – English sole. USFWS - Biol. Starr, R.M. 1998. Design principles for rockfish reserves on the U. S. West Coast.
Mark Zimmermann, Alaska Fisheries Science Center, NMFS Rep., Vol. 82(11.101), U.S. Army Corps Eng. TR EL-82-4. 17 pp. Marine harvest refugia for West Coast rockfish: A workshop., Pacific Grove, Cali-
Lauth, R.R. 2001. The 2000 Pacific West Coast Upper Continental Slope trawl survey fornia.
REFERENCES of groundfish resources off Washington, Oregon, and California: Estimates of distri-
Allen, M.J., and G.B. Smith. 1988. Atlas and zoogeography of common fishes in bution, abundance, and length composition. NOAA Tech. Memo.NMFS-AFSC-120. U.S. Department of Commerce. Magnuson-Stevens Act Provisions. Federal Regis-
the Bering Sea and Northeastern Pacific. NOAA Tech. Rep. NMFS 66. Seattle, WA. Seattle, WA. 284 pp. ter: January 11, 2002 (Vol. 67, No. 8). Washington, DC: U.S. Govt. Printing Office.
151 pp. p. 1582.
Leet, W.S., C.M. DeWees, and C.W. Haugen. 1992. California’s Living Marine Re-
Brown, S.K., K.R. Buja, S.H. Jury, M.E. Monaco, and A. Banner. 2000. Habitat suit- sources: A Status Report. California Dept. Fish and Game. Sacramento, CA. 247 USFWS. 1980a. Habitat as a basis for environmental assessment. Report ESM 101.
ability models for eight fish and invertebrate species in Casco and Sheepscot Bays, pp. U.S. Fish Wildl.. Serv., Washington, DC.
Maine. N. Am. J. Fish. Manag., Vol. 20, pp. 408-435.
Leet, W.S., C.M. DeWees, R. Klingbeil, and E.J. Larson. 2001. California’s Living USFWS. 1980b. Habitat evaluation procedures (HEP). Report ESM 102. U.S. Fish
Cailliet, G.M., E.J. Burton, J.M. Cope, L.A. Kerr, R.J. Larson, D. VenTresca, R.N. Marine Resources: A Status Report. California Dept. Fish and Game. Sacramento, Wildl. Serv., Washington, DC.
Lea, and E. Knaggs. 2000. Final Report – Biological characteristics of near-shore CA. 592 pp.
fishes of California: A review of existing knowledge and proposed additional studies. USFWS. 1981. Standards for the development of habitat suitability index models
Love, M. 1996. Probably More Than You Want to Know About the Fishes of the
Submitted to Pacific States Marine Fisheries Commission. for use with the habitat evaluation procedures. Report ESM 103. U.S. Fish Wildl.
Pacific Coast. Really Big Press, Santa Barbara, CA. 45 pp. Serv., Washington, DC.
Cass, A.J., R.J. Beamish, and G.A. McFarlane. 1990. Lingcod (Ophiodon elongatus).
Love, M.S., L. Thorsteinson, C.W. Mecklenburg, and T.A. Mecklenberg. 1996. A
Can. Spec. Pub. Fish. Aquat. Sci., Vol. 109, 40 pp. Wilkins, M.E., M. Zimmermann, and K.L. Weinberg. 1998. The 1995 Pacific West
checklist of marine and estuarine fishes of the Northeast Pacific, from Alaska to Baja Coast Bottom Trawl Survey of Groundfish Resources: Estimates of Distribution, Abun-
Coyne, M.S. and J.D. Christensen. 1997. Habitat Suitability Index Modeling: Techni- California. National Biological Service. http://id-www.ucsb.edu/lovelab/home.html dance, and Length and Age Composition. NOAA Tech. Memo NMFS-AFSC-89.
cal Guidelines. NOAA/NOS Strategic Environmental Assessments Division, Silver University of California, Santa Barbara, CA. 24 pp.
Spring, MD. 19 pp. Williams, E.H., and S. Ralston. In press. Distribution and co-occurrence of rock-
Love, M.S., P. Morris, M. McCrae, and R. Collins. 1990. Life History Aspects of 19 fishes (family Scorpaenidae) over trawlable shelf and slope habitats of California
Dark, T.A., and M.E. Wilkins. 1994. Distribution, abundance, and biological character- Rockfish Species (Scorpaenidae: Sebastes) from the Southern California Bight. and Southern Oregon. Fishery Bulletin 000:000-000.
istics of groundfish off the coast of Washington, Oregon, and California, 1977-1986. NOAA Tech. Memo. NMFS 87. Seattle, WA. 113 pp.
NOAA Tech. Memo. NMFS 117. Seattle, WA. 114 pp. Wolotira, R.J., T.M. Sample, S.F. Noel, and C.R. Iten. 1993. Geographic and bathy-
Mahon, R., S.K. Brown, K.C.T. Zwanenburg, D. Atkinson, K.R. Buja, L. Claflin, G.D. metric distributions for many commercially important fishes and shellfishes off the
Dark, T.A., M.E. Wilkins, and K. Edwards. 1983. Bottom Trawl Survey of Canary Howell, M.E. Monaco, R.N. O’Boyle, and M. Sinclair. 1998. Assemblage and bioge- West Coast of North America, based on research survey and commercial catch data,
Rockfish (Sebastes pinninger), Yellowtail Rockfish (S. flavidus), Bocaccio (S. pau- ography of demersal fishes of the east coast of North America. Can. J. Fish. Aquat. 1912-84. NOAA Tech. Memo. NMFS-AFSC-6. 184 p.
cispinis) and Chilipepper (S. goodei) off Washington—California, 1980. NOAA Tech. Sci., Vol. 55: pp. 1704-1738.
Memo. F/NWC, Seattle, WA. 48 pp. Wright, N., J. Kum, C. King, E. Knaggs, B. Leos, and C. Perez. 2000. Marine fish-
Miller, D.J., and R.N. Lea. 1972. Guide to the Coastal Marine Fishes of California. ery profiles. Volume 1: Nearshore-Version 1. California Department of Fish and
Emmett, R.L., S.L. Stone, S.A. Hinton, and M.E. Monaco. 1991. Distribution and Cal. Dept. Fish and Game, Fish Bull. Vol. 157, pp. 23-29. Game, Marine Region. 150 pp.
abundance of fishes and invertebrates in West Coast estuaries, Volume II: Species
NOAA/NMFS. 1998. Essential Fish Habitat – West Coast Groundfish Appendix
life history summaries. ELMR Rep. No. 8. NOAA/NOS SEA Division, Rockville, MD. Yoklavich, M.M., H.G. Greene, G.M. Cailliet, D.E. Sullivan, R.N. Lea, and M.S. Love.
(online). http://www.nwr.noaa.gov/1sustfsh/efhappendix. NOAA/NMFS, EFH Core
329 pp. 2000. Habitat associations of deep-water rockfishes in a submarine canyon: an
Team for West Coast Groundfish, Seattle, WA. example of a natural refuge. Fish Bulletin, U.S. 98:625-641.
Eschmeyer, W.N., E.S. Herald, H. Hammann, and K.P. Smith. 1983. A field guide to
Pacific Coast Fishes. Houghton Mifflin Company, Boston, MA. 89 pp.
45
Cassin's Auklet Ptychoramphus aleuticus Charadriiformes/Alcidae
Parkinson's Petrel Procellaria parkinsoni Procellariiformes/Procellariidae
Rhinoceros Auklet Cerorhinca monocerata Charadriiformes/Alcidae
Flesh-footed Shearwater Puffinus carneipes Procellariiformes/Procellariidae
Tufted Puffin Fratercula cirrhata Charadriiformes/Alcidae
Short-tailed Shearwater Puffinus tenuirostris Procellariiformes/Procellariidae
Species mapped with related species &puffinus the summary diversity and density analyses (n=9, 4 maps)
Manx Shearwater Puffinus used in Procellariiformes/Procellariidae
Western Grebe Aechmophorus occidentalis Podicipediformes/Podicipedidae
Townsend's Shearwater Puffinus auricularis Procellariiformes/Procellariidae
Clark's Grebe Aechmophorus clarkii Podicipediformes/Podicipedidae
Wilson's Storm-petrel Oceanites oceanicus Procellariiformes/Hydrobatidae
Black Scoter Melanitta nigra tethys Anseriformes/Anatidae
Wedge-rumped Storm-petrel Oceanodroma Procellariiformes/Hydrobatidae
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Surf Scoter Storm-petrel Melanitta perspicillata Anseriformes/Anatidae
Markham's Oceanodroma markhami Procellariiformes/Hydrobatidae
White-winged Scoter Melanitta fuscamicrosoma Anseriformes/Anatidae
Least Storm-petrel Oceanodroma Procellariiformes/Hydrobatidae
Caspian Tern Sterna caspia Charadriiformes/Laridae/Sterninae
Red-billed Tropicbird Phaethon aethereus Pelecaniformes/Phaethonidae
Elegant Tern Sterna elegans Charadriiformes/Laridae/Sterninae
Brown Booby Sula leucogaster Pelecaniformes/Phaethonidae
Table 11. Marine bird species used in this analysis. TablePelican
Xantus's Murrelet Marine bird species usedhypoleucus
White 11 cont. Synthliboramphus in this analysis.
Charadriiformes/Alcidae
Pelecanus erythrorynchos Pelecaniformes/Pelecanidae
Common Name Scientific Name Order/Family/SubFamily Common Name Scientific Namecraveri Order/Family/SubFamily
Craveri's Murrelet Synthliboramphus Charadriiformes/Alcidae
South Polar Skua Stercorarius maccormicki Charadriiformes/Laridae/Stercorariinae
Species that were mapped separately and used in the summary bird diversity and density analyses (n=31) Pomarine Jaeger
in the data set used only in the summary the diversity Charadriiformes/Laridae/Stercorariinae(n=31)
Stercorarius pomarinus
Species that were mapped separately and used in bird summary and density analyses (n=37)
bird diversity and density analyses
Pacific Loon Gavia pacifica Gaviiformes/Gaviiadae Red-throated
Pacific Loon Loon stellata Gaviiformes/Gaviidae
Parasitic Jaeger Stercorarius parasiticus Charadriiformes/Laridae/Stercorariinae
Gavia pacifica Gaviiformes/Gaviiadae
Laysan Albatross Phoebastria immutabilis Procellariiformes/Diomedeidae Common Loon Gavia immerlongicaudus Gaviiformes/Gaviidae
Long-tailed Jaeger Stercorarius Charadriiformes/Laridae/Stercorariinae
Laysan Albatross Phoebastria immutabilis Procellariiformes/Diomedeidae
Black-footed Albatross Phoebastria nigripes Procellariiformes/Diomedeidae Horned Grebe Podiceps auritus Podicipediformes/Podicipedidae
Black-footed Gull
Bonaparte's Albatross Larus philadelphia Charadriiformes/Laridae/Larinae
Phoebastria nigripes Procellariiformes/Diomedeidae
Northern Fulmar Fulmarus glacialis Procellariiformes/Procellariidae Red-necked Grebe Podiceps glacialis
Fulmarus grisegena Podicipediformes/Podicipedidae
Mew Gull Larus canus Charadriiformes/Laridae/Larinae
Northern Fulmar Procellariiformes/Procellariidae
Pink-footed Shearwater Puffinus creatopus Procellariiformes/Procellariidae Eared Grebe Podiceps nigricollis Podicipediformes/Podicipedidae
Pink-footedGull
Ring-billed Shearwater Larus delawarensis Charadriiformes/Laridae/Larinae
Puffinus creatopus Procellariiformes/Procellariidae
Buller's Shearwater Puffinus bulleri Procellariiformes/Procellariidae Murphy's Petrel Pterodroma ultima
California Gull Larus californicus Charadriiformes/Laridae/Larinae
Buller's Shearwater Puffinus bulleri Procellariiformes/Procellariidae
Sooty Shearwater Puffinus griseus Procellariiformes/Procellariidae Cook's Petrel Pterodroma cookii
Thayer's Gull Larus thayeri Charadriiformes/Laridae/Larinae
Sooty Shearwater Puffinus griseus Procellariiformes/Procellariidae
Black-vented Shearwater Puffinus opisthomelas Procellariiformes/Procellariidae Parkinson's Petrel Procellaria parkinsoni
Glaucous Gull Larus hyperboreus Charadriiformes/Laridae/Larinae
Black-vented Shearwater Puffinus opisthomelas Procellariiformes/Procellariidae
Fork-tailed Storm-petrel Oceanodroma furcata Procellariiformes/Hydrobatidae Flesh-footedKittiwake
Red-legged Shearwater Puffinus carneipes Procellariiformes/Procellariidae
Rissa brevirostris Charadriiformes/Laridae/Larinae
Fork-tailed Storm-petrel Oceanodroma furcata Procellariiformes/Hydrobatidae
Leach's Storm-petrel Oceanodroma leucorhoa Procellariiformes/Hydrobatidae Short-tailed Shearwater Puffinus tenuirostris Procellariiformes/Procellariidae
Royal Tern Sterna maxima Charadriiformes/Laridae/Sterninae
Leach's Storm-petrel Oceanodroma leucorhoa Procellariiformes/Hydrobatidae
Ashy Storm-petrel Oceanodroma homochroa Procellariiformes/Hydrobatidae Manx Shearwater Puffinus puffinus Procellariiformes/Procellariidae
Common Tern Sterna hirundo Charadriiformes/Laridae/Sterninae
Ashy Storm-petrel Oceanodroma homochroa Procellariiformes/Hydrobatidae
Black Storm-petrel Oceanodroma melania Procellariiformes/Hydrobatidae Townsend's Shearwater Puffinus auricularis Procellariiformes/Procellariidae
Forster's Tern Sterna forsteri Charadriiformes/Laridae/Sterninae
Black Storm-petrel Oceanodroma melania Procellariiformes/Hydrobatidae
Brown Pelican Pelecanus occidentalis Pelecaniformes/Pelecanidae Wilson's Storm-petrel Oceanites oceanicus Procellariiformes/Hydrobatidae
Thick-billed Murre Uria lomvia Charadriiformes/Alcidae
Brown Pelican Pelecanus occidentalis Pelecaniformes/Pelecanidae
Brandt's Cormorant Phalacrocorax penicillatus Pelecaniformes/Phalacrocoracidae Wedge-rumped
Brandt'sMurreletStorm-petrel Oceanodroma penicillatus
Phalacrocorax tethys Procellariiformes/Hydrobatidae
Ancient Cormorant Synthliboramphus antiquus Charadriiformes/Alcidae
Pelecaniformes/Phalacrocoracidae
INTRODUCTION Double-crested Cormorant Phalacrocorax auritus Pelecaniformes/Phalacrocoracidae Markham's Storm-petrel Oceanodroma auritus
Phalacrocorax markhami Procellariiformes/Hydrobatidae
Parakeet Auklet Aethia psittacula Charadriiformes/Alcidae
Double-crested Cormorant Pelecaniformes/Phalacrocoracidae
The California Current system runs south through the north/central California Pelagic Cormorant Phalacrocorax pelagicus Pelecaniformes/Phalacrocoracidae Least Storm-petrel Oceanodroma pelagicus
Phalacrocorax microsoma Procellariiformes/Hydrobatidae
Horned Puffin Fratercula corniculata Charadriiformes/Alcidae
Pelagic Cormorant Pelecaniformes/Phalacrocoracidae
study area; it is one of the most productive ocean systems in the world (Glantz Red-necked Phalarope Phalaropus lobatus Charadriiformes/Scolopacidae Red-billed Tropicbird Phaethon aethereus Pelecaniformes/Phaethonidae
Red-necked Phalarope Phalaropus lobatus Charadriiformes/Scolopacidae
Red Phalarope Phalaropus fulicarius Charadriiformes/Scolopacidae Brown Booby Sula leucogaster Pelecaniformes/Phaethonidae
Table 11 contains the marine birdfulicarius that were selected for this analysis; data
Red Phalarope Phalaropus Charadriiformes/Scolopacidae
and Thompson, 1981). Hence, the study area contains a rich fauna of marine species
Heermann's Gull Larus heermanni Charadriiformes/Laridae/Larinae White Pelican Pelecanus erythrorynchos Pelecaniformes/Pelecanidae
Heermann's Gull Larus heermanni Charadriiformes/Laridae/Larinae
birds, as evidenced in species abundance and richness. In addition to a populous for 40 Polar Skua were either mapped separately, or Charadriiformes/Laridae/Larinae
South species together for small species groups
Western Gull Larus occidentalis Charadriiformes/Laridae/Larinae Stercorarius maccormicki Charadriiformes/Laridae/Stercorariinae
Western Gull Larus occidentalis
breeding community, the community of seasonal residents and migrants is even that generally co-occur (e.g., scoters). Ten of these species maps are included in this
Glaucous-winged Gull Larus glaucescens Charadriiformes/Laridae/Larinae Pomarine Jaeger Gull Stercorarius pomarinus Charadriiformes/Laridae/Stercorariinae
Glaucous-winged Larus glaucescens Charadriiformes/Laridae/Larinae
more robust, as central California is the destination for many marine bird species section; Gull remaining maps are on the accompanying CD-ROM. The remaining 37
Parasitic the
Sabine's Gull Xema sabini Charadriiformes/Laridae/Larinae Sabine's Jaeger Stercorarius
Xema sabini parasiticus Charadriiformes/Laridae/Stercorariinae
Charadriiformes/Laridae/Larinae
California Gull Larus californicus Charadriiformes/Laridae/Larinae
species that occurred in the study longicaudus data set were used to develop summary
Long-tailed Jaeger Stercorarius Charadriiformes/Laridae/Stercorariinae
California Gull Larus californicus Charadriiformes/Laridae/Larinae
seeking productive feeding areas and acceptable habitat in which to spend their non- area and
Black-legged Kittiwake Rissa tridactyla Charadriiformes/Laridae/Larinae Bonaparte's Kittiwake
Black-leggedGull Larus philadelphia
Rissa tridactyla Charadriiformes/Laridae/Larinae
breeding periods. Unlike many marine organisms, marine birds have a tremendous marine bird maps on marine canus diversity and density.
Sternabird
Arctic Tern Sterna paradisaea Charadriiformes/Laridae/Sterninae Mew Tern
ArcticGull Larus paradisaea Charadriiformes/Laridae/Larinae
Charadriiformes/Laridae/Sterninae
mobility and the fact that many seek this region to find food bespeaks the region’s Common Murre Uria aalge Charadriiformes/Alcidae Ring-billed Gull Larus delawarensis Charadriiformes/Laridae/Larinae
Common Murre Uria aalge Charadriiformes/Alcidae
Pigeon Guillemot Cepphus columba Charadriiformes/Alcidae California Gull Larus californicus Charadriiformes/Laridae/Larinae
trophic richness. Fortunately for the purpose of management of the central California About Guillemot
Pigeon the Survey Data and Literature Used in this Assessment. The survey data
Cepphus columba Charadriiformes/Alcidae
Marbled Murrelet Brachyramphus marmoratus Charadriiformes/Alcidae Thayer's
Marbled Gull Larus thayeri Charadriiformes/Laridae/Larinae
used in Murrelet Brachyramphus marmoratus Charadriiformes/Alcidae
National Marine Sanctuaries, the marine avifauna of the study area have been one this summary were not designed with sanctuary resource management in
Cassin's Auklet Ptychoramphus aleuticus Charadriiformes/Alcidae GlaucousAuklet
Cassin's Gull Larus hyperboreus Charadriiformes/Laridae/Larinae
Ptychoramphus aleuticus Charadriiformes/Alcidae
of the most thoroughly surveyed. mind, but include the interests of individual researchers to study spatial and temporal
Rhinoceros Auklet Cerorhinca monocerata Charadriiformes/Alcidae Red-legged Kittiwake Rissa brevirostris Charadriiformes/Laridae/Larinae
Rhinoceros Auklet Cerorhinca monocerata Charadriiformes/Alcidae
patterns of marine birds,Fratercula cirrhata
federal government efforts to assess potential biological
Tufted Puffin Fratercula cirrhata Charadriiformes/Alcidae Royal Puffin
TuftedTern Sterna maxima Charadriiformes/Laridae/Sterninae
Charadriiformes/Alcidae
Species mapped with related species & used in the summary diversity and density analyses (n=9, 4 maps) Species Tern
Common of oil with related species hirundo in government efforts density analyses oil 4 maps)
Sterna & used Charadriiformes/Laridae/Sterninae
DATA AND ANALYSES impacts mappeddevelopment, and statethe summary diversity and to respond to(n=9,spills, of
Western Grebe Aechmophorus occidentalis Podicipediformes/Podicipedidae Forster's Tern Sterna forsteri Charadriiformes/Laridae/Sterninae
Western Grebe Aechmophorus occidentalis Podicipediformes/Podicipedidae
Overview of Map Development and Analysis Process. The methods used in each which there have been several major ones in the Charadriiformes/Alcidae
study area.
Clark's Grebe Aechmophorus clarkii Podicipediformes/Podicipedidae Thick-billed Murre Uria lomvia
Clark's Grebe Aechmophorus clarkii Podicipediformes/Podicipedidae
survey were different, and because of this, careful consideration and correction are Black Scoter Melanitta nigra Anseriformes/Anatidae Ancient Murrelet Synthliboramphus antiquus Charadriiformes/Alcidae
Black Scoter Melanitta nigra Anseriformes/Anatidae
required to merge the data sets in a meaningful and scientifically acceptable way. The The Literature. Several reports,perspicillata from these surveys, provided background
Melanitta resulting
Surf Scoter Melanitta perspicillata Anseriformes/Anatidae Parakeet Auklet Aethia psittacula Charadriiformes/Alcidae
Surf Scoter Anseriformes/Anatidae
White-winged Scoter Melanitta fusca Anseriformes/Anatidae Horned Puffin Fraterculafusca
Melanitta corniculata Charadriiformes/Alcidae
information Scoter occurrence patterns of marine birds in the region. The general
White-winged on the Anseriformes/Anatidae
major steps of the data development for the bird analyses were as follows: species
Caspian Tern Sterna caspia Charadriiformes/Laridae/Sterninae Caspian Tern Sterna caspia Charadriiformes/Laridae/Sterninae
and study area selection; data set identification and collection; data corrections; composition and distributionelegans marine avifauna was described by Ainley
of the
Elegant Tern Sterna elegans Charadriiformes/Laridae/Sterninae Elegant Tern Sterna Charadriiformes/Laridae/Sterninae
data conversion into common comparable units; organizing the data into 5’ latitude (1976) and Briggs et al.Synthliboramphus hypoleucus
(1983, 1987a, b). Ainley and DeSante (1980) and Pyle
Xantus's Murrelet Synthliboramphus hypoleucus Charadriiformes/Alcidae Xantus's Murrelet Charadriiformes/Alcidae
by 5’ longitude cells; and calculating effort and density for each marine bird species. and Henderson (1991) provide a fine-scale look at species’ seasonal presence and
Craveri's Murrelet Synthliboramphus craveri Charadriiformes/Alcidae Craveri's Murrelet Synthliboramphus craveri Charadriiformes/Alcidae
Species in the data set used only in the summary bird diversity and density analyses (n=37)
migratory the data set as viewedthe summaryFarallon Islands; Ainley et al. (1995a, c), Veit
Species in periods, used only in from the bird diversity and density analyses (n=37)
Seasonal density maps were then created for 40 species. Overall density, biomass Red-throated Loon Gavia stellata Gaviiformes/Gaviidae Red-throated Loon Gavia stellata Gaviiformes/Gaviidae
density, and diversity maps were also created using distribution and abundance data etCommon Loon and Oedekoven immer (2001) provide an interannual view of variability
al. (1997) Gavia et al.
Common Loon Gavia immer Gaviiformes/Gaviidae Gaviiformes/Gaviidae
for 76 bird species combined. These maps were reviewed at an expert workshop in inHorned Grebe
spatial occurrence. The last four references, as Podicipediformes/Podicipedidae
well as Ainley et al. (1994), Spear
Horned Grebe Podiceps auritus Podicipediformes/Podicipedidae Podiceps auritus
October 2002. The draft bird report was also sent out for expert review in November and AinleyGrebe
Red-necked (1999) and Ainley and Divoky (2001), investigated long-term temporal
Red-necked Grebe Podiceps grisegena Podicipediformes/Podicipedidae Podiceps grisegena Podicipediformes/Podicipedidae
Eared Grebe Podiceps nigricollis Podicipediformes/Podicipedidae
trendsGrebe
Eared Podiceps nigricollis Podicipediformes/Podicipedidae
(see list of reviewers at end of this section). Revisions were made to the maps and in populations. Information on habitat preferences of marine birds and how
Murphy's Petrel Pterodroma ultima Procellariiformes/Procellariidae Murphy's Petrel Pterodroma ultima Procellariiformes/Procellariidae
text based on that review. these are affected by ocean climate variability are provided for selected species
Cook's Petrel Pterodroma cookii Procellariiformes/Procellariidae Cook's Petrel Pterodroma cookii Procellariiformes/Procellariidae
inParkinson'sand Boekelheide (1990), Oedekoven etProcellariiformes/Procellariidae
Ainley Petrel al. (2001), and in a GIS analysis
Parkinson's Petrel Procellaria parkinsoni Procellariiformes/Procellariidae Procellaria parkinsoni
Flesh-footed Shearwater Puffinus carneipes Procellariiformes/Procellariidae
Species Selected for Analysis. Selection criteria for bird species included in this by Allen (1994). The food-webcarneipes
Puffinus relationships of marine birds in this region are also
Flesh-footed Shearwater Procellariiformes/Procellariidae
Short-tailed Shearwater Puffinus tenuirostris Procellariiformes/Procellariidae Short-tailed Shearwater Puffinus tenuirostris Procellariiformes/Procellariidae
assessment were: 1) the species must have a mostly marine distribution in the remarkably well known (Balz and Morejohn 1977; Ainley and Sanger 1979; Briggs et
Manx Shearwater Puffinus puffinus Procellariiformes/Procellariidae Manx Shearwater Puffinus puffinus Procellariiformes/Procellariidae
study area; and 2) adequate ocean survey data for the species is available and in al. 1984, Briggs and Chu Puffinus auricularis
1987; Chu 1984; Ainley and Boekelheide 1990; Ainley et al.
Townsend's Shearwater Puffinus auricularis Procellariiformes/Procellariidae Townsend's Shearwater Procellariiformes/Procellariidae
a useable format. Species that are abundant, endangered, threatened, or a state 1996a, b; and Sydeman et al. 1997); and the breeding biology, including interannual
Wilson's Storm-petrel Oceanites oceanicus Procellariiformes/Hydrobatidae Wilson's Storm-petrel Oceanites oceanicus Procellariiformes/Hydrobatidae
Wedge-rumped Storm-petrel Oceanodroma tethys Procellariiformes/Hydrobatidae
species of concern were also a priority. The study area for the GIS assessment was variability in productivity Oceanodroma tethys to food-web variation, is very well known
and relationship
Wedge-rumped Storm-petrel Procellariiformes/Hydrobatidae
Markham's Storm-petrel Oceanodroma markhami Procellariiformes/Hydrobatidae Markham's Storm-petrel Oceanodroma markhami Procellariiformes/Hydrobatidae
seaward of the beach and did not include estuaries, so few shorebirds and waterfowl (Ainley and BoekelheideOceanodroma microsoma 1995b). See the end of this section for
1990, Ainley et al.
Least Storm-petrel Oceanodroma microsoma Procellariiformes/Hydrobatidae Least Storm-petrel Procellariiformes/Hydrobatidae
were included. Because marine distributions of birds are affected by where they completeTropicbirdreferences used.
Red-billed list of
Red-billed Tropicbird Phaethon aethereus Pelecaniformes/Phaethonidae Phaethon aethereus Pelecaniformes/Phaethonidae
breed and roost, we included information on the location and size of breeding and Brown Booby Sula leucogaster Pelecaniformes/Phaethonidae Brown Booby Sula leucogaster Pelecaniformes/Phaethonidae
White Pelican Pelecanus erythrorynchos Pelecaniformes/Pelecanidae
The Data Sets. See Table 12, aerythrorynchos of data sets used in the analyses, and
White Pelican Pelecanus summary Pelecaniformes/Pelecanidae
roosting sites, where available.
South Polar Skua Stercorarius maccormicki Charadriiformes/Laridae/Stercorariinae South Polar Skua Stercorarius maccormicki Charadriiformes/Laridae/Stercorariinae
Figures 39 and 40, which show the spatial extent of the individual data sets used
Pomarine Jaeger Stercorarius pomarinus Charadriiformes/Laridae/Stercorariinae Pomarine Jaeger Stercorarius pomarinus Charadriiformes/Laridae/Stercorariinae
Parasitic Jaeger Stercorarius parasiticus Charadriiformes/Laridae/Stercorariinae Parasitic Jaeger Stercorarius parasiticus Charadriiformes/Laridae/Stercorariinae
46
Long-tailed Jaeger Stercorarius longicaudus Charadriiformes/Laridae/Stercorariinae Long-tailed Jaeger Stercorarius longicaudus Charadriiformes/Laridae/Stercorariinae
Bonaparte's Gull Larus philadelphia Charadriiformes/Laridae/Larinae Bonaparte's Gull Larus philadelphia Charadriiformes/Laridae/Larinae
Mew Gull Larus canus Charadriiformes/Laridae/Larinae Mew Gull Larus canus Charadriiformes/Laridae/Larinae
Ring-billed Gull Larus delawarensis Charadriiformes/Laridae/Larinae Ring-billed Gull Larus delawarensis Charadriiformes/Laridae/Larinae
California Gull Larus californicus Charadriiformes/Laridae/Larinae California Gull Larus californicus Charadriiformes/Laridae/Larinae
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
in the bird analyses. The ship and aerial strip transect data used in the GIS assessment were
collected from 1980-2001 and occurred from Point Arena south to Point Sal, and offshore to the
extent of data availability. However, the species maps do not generally include the full extent of
Spatial Extent of Data Sets: Ship-based Surveys available data, primarily because the assessment was focused on the national marine sanctuaries
off central California. Also, estuaries were not part of the study area, but coastal colonies in
estuaries were mapped to provide a more complete view of important areas for breeding species.
129°W 128°W 127°W 126°W 125°W 124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
See a more detailed description of data sets on the accompaning CD-ROM.
200 m
200 m
20
20
SF-DODS Cruises
Midwater Trawls for Juvenile Rockfish
0
0
0m
0m
1985-2001 1996-2000
39°N
39°N
Data Synthesis.
Mainly upwelling period Year round Summarizing Transect Data into Grid Cells. The above data sets required a significant amount
of processing and correction in order to synthesize them. Because wind speed affects detection
of marine birds, data collected when wind speed exceeded 25 knots were excluded. Data were
38°N
allocated into 5’ latitude by 5’ longitude cells. All aerial data were continuous; each ship-based
38°N
data set was converted separately into a continuous transect format to the extent possible. The
continuous aerial data were binned into the appropriate cell. For the SF-DODS and EPOCS
Table 12. Summary of at-sea survey data sets used in the analyses.
37°N
37°N
Ocean Total
Principal Platform Habitat Seasons Transect
0 25 50 Km Data Set Investigator Height Covered2 Years Sampled Width
Surface survey
MMS Low of the shelf,
36°N
36°N
Altitude Aerial slope & deep All three
Surveys Briggs Pembroke, 62m ocean beyond 1980-1983 seasons 50m
Surveyor, 12m,
EPOCS Discoverer, Surface survey
35°N
35°N
Shipboard Oceano- of the deep All three
a b Surveys Ainley grapher, 15m ocean 1984-1994 seasons 300-600m
CA Seabird
Ecology Low- Surface survey
200 m
200 m
Altitude Aerial Partenavia, of shelf and Mainly
20
20
EPOCS Cruises
NMFS ORCAWALE Cruises
0
0
0m
0m
Surveys Briggs 62m slope 1985 Upwelling 50m
1984-1994
2001
39°N
39°N
NMFS
Year round
Mainly oceanic period Midwater Trawl
Juv. Rockfish Surface survey
Assessment: David Starr of shelf and Mainly
Ship Surveys Ainley Jordan, 10m slope to 3000 m 1985-2001 Upwelling 300m
38°N
38°N
OSPR Low Surface survey
Altitude Aerial Partenavia, of shelf and 1994-1998, All three
Surveys Bonnell, Tyler 62m slope 2001 seasons 50m
MMS Santa
Barbara
37°N
37°N
Channel Low Surface survey
Altitude Aerial Partenavia, of shelf and All three
Surveys Bonnell 62m slope 1995-1997 seasons 50m
Surface survey
SF-DODS Ship of shelf and All three
36°N
36°N
Surveys Ainley Point Sur, 8m slope to 3000 m 1996-2000 seasons 300m
Surface survey 200-300m,
NMFS/SWFSC of the shelf, Mainly depending on
ORCAWALE slope & deep Oceanic species &
Ship Survey Ballance MacArthur, 11m ocean beyond 2001 (Aug-Nov) conditions
35°N
35°N
c d Note
See description of data sets on the CD for more information on the data sets.
129°W 128°W 127°W 126°W 125°W 124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Source Data: See text. studies, and the Rockfish Assessment cruises prior to 1997, the beginning position, ship heading,
Figure 39. Spatial extent of data sets used in the marine bird analysis: ship-based surveys. and speed were used to compute the end position of each 2-4 km continuous transect. From
this, a midpoint of the transect was determined. As times of observations were not available, the
position of the midpoint was used to select the cell to which the survey effort was assigned. If
this midpoint fell on a cell boundary, it was assigned to the cell to the north or west. To maintain
47
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Table 13. Summary of combined data set effort by ocean season.
Dates Used for Number Kilometers of Number of
Spatial Extent of Data Sets: Aerial Surveys Ocean Each Ocean of Years Trackline Number 5' Cells
Season Season Months Included Surveyed of Visits Sampled
1980-1982,
Upwelling 1985-2001
15 Mar-14 Aug 5 64177 11050 1335
129°W 128°W 127°W 126°W 125°W 124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
200 m
200 m
MMS Aerial Surveys OSPR Aerial Surveys 1980-1982,
20
20
0
0
0m
0m
1991, 1994-
1980-1983 1994-1998, 2001
39°N
39°N
Oceanic 2001
15 Aug-14 Nov 3 29263 4171 1130
Year round Year round
Davidson 1980-1986,
Current 1991-2001
15 Nov-14 Mar 4 40265 5878 1593
1980-2001
TOTAL 1 Jan – 31 Dec 12 133705 21099 2294
38°N
38°N
Note. The total number of cells sampled is not a straight sum; it refers to the number of unique
cells surveyed.
the correspondence between effort and bird observations, observations were also assigned
37°N
37°N
to the transect midpoints. For the Rockfish Assessment Cruises from 1997 onward, effort was
assigned to the cells through which the vessel passed based on the proportion of trackline that
0 25 50 Km
fell within each cell, and observations were interpolated along the cruise track according to
the time of each observation. The marine bird survey data from the ORCAWALE cruise were
36°N
36°N
recorded continuously using automatic recording software and were processed like the aerial
survey data.
Data Analysis.
35°N
35°N
e f Effort. The combined at-sea survey effort for birds included 133,705 kilometers of trackline, as
well as 128,886 observations of 973,318 birds in the analyzed data set. Survey effort by ocean
season is summarized in Figure 41 and Table 13.
200 m
200 m
Seabird Ecology Aerial Surveys
20
20
MMS Santa Barbara Channel
0
0
0m
0m
1985 Aerial Surveys
39°N
39°N
Calculating Density. From the digitized survey data, we mapped the distribution of effort and of
1995-1997
Mainly upwelling period
species observations into a grid of 5-minute latitude by 5-minute longitude cells, using MMS-
Year round
CDAS (Marine Mammal and Seabird Computer Database Analysis System, MMS 2001). The
species data were first transformed into densities on the basis of strip widths (which varied by
38°N
38°N
platform, depending on speed and height above water; see Table 12). The number of birds of
each species seen was then divided by area surveyed to estimate density in each cell for that
data set. For construction of density plots, if a cell was censused in other years or the same
year by another survey, densities in cells were averaged and weighted according to effort.
37°N
37°N
Organizing Data into Ocean Seasons. Effort and species data were organized and mapped into
three distinct ocean seasons (Bolin and Abott 1963): Upwelling, Oceanic, and Davidson Current,
because ocean conditions differ distinctly among them and are known to affect the biota of the
36°N
36°N
California Current (e.g. Ainley 1976, Briggs et al. 1987). As there is significant interannual variation
in the actual initiation and termination of these seasons, the following dates were defined for each
season for purposes of analysis: Upwelling Season is 15 March-14 August; Oceanic Season is
15 August-14 November; and Davidson Current Season is 15 November-14 March.
35°N
35°N
g h
Seasonal Density Maps for Individual Species. Seasonal density maps were generated for 40 bird
129°W 128°W 127°W 126°W 125°W 124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
species. These maps were then reviewed to characterize the spatial and seasonal occurrence
Source Data: See text.
pattern of each species in the study area.
Figure 40. Spatial extent of data sets used in the analysis: aerial surveys.
Seasonal High Use Areas for Individual Species. In order to provide a summary map of space
use, seasonal density data were binned into 10-minute latitude by 10-minute longitude cells
for each species or species group. The purpose of the seasonal high use maps is to provide
48
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
one overall map for each species (or group of species) that standardize for variable effort among cells and variable strip
Combined At-Sea Effort for Marine Bird Analysis describes the spatial and temporal use patterns, as clearly as width for species, density was used for each species in each
possible. The seasonal high use index is based on the top 20% cell as the basis for calculating the diversity index value.
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
of sampled cells within a given season. The index is therefore
Upwelling Season Oceanic Season
200 m
200 m
20
20
sensitive to cells which were not sampled in any one of the The Shannon Index was selected as the diversity metric
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) three seasons, causing a downward bias in the index. because it is widely used and accepted in community ecology.
39°N
39°N
It has three desirable properties for a diversity index, noted
Effort Use of a 10-minute block size greatly reduces the magnitude below. Most diversity indices do not take these three qualities
(Km of trackline)
of this bias. Non-zero cells were then ranked and those in the into account. For more information on diversity indices, see
>1000.00
top 20 percent were selected and defined as seasonal high use Ecological Diversity, E.C. Pielou, pp 7-18.
500.01 - 1000.00
38°N
38°N
areas. Cells were then mapped with colors corresponding to
250.01 - 500.00
100.01 - 250.00
the number of seasons of high use. Cells in which there was 1. The diversity index is greatest when all species in the
50.01 - 100.00
effort but birds were not observed, and cells where sightings community are equally represented in numbers (e.g., evenness.
25.01 - 50.00
occurred but were never high use areas, were also mapped in a community). Or, for a given number of species (e.g.,
5.01 - 25.00
0.01 - 5.00
with two additional colors. richness value), the diversity index should have it’s greatest
37°N
37°N
value when the proportion of each species is the same.
0 25 50 Km
Major Breeding Colonies. Best available breeding colony data
(number of breeding birds, mostly from Carter, et al. 1992, 2. Given two completely diverse or even communities, the
with some updates) were mapped for each species for which one with the higher number of species has a greater diversity
colony information was available, on the same map as the value.
36°N
36°N
"seasonal high use" information. A map (pp. 81) and table (p.
53) of the top 40 breeding colonies is included in Section 2.2; 3. The last property is difficult to summarize: This property takes
the complete colony table, based on best available data, will into account the hierarchical nature, or "representativeness" in
be included on the CD-ROM. the biological classification of each species when estimating
diversity.
35°N
35°N
a b Spatial and Temporal Patterns Summary Table. Density
maps for 44 species were inspected to identify which cells Evaluating Variation in Species Abundance. In order to
200 m
Davidson Current Season
200 m
All Seasons
20
20
exhibited the highest density each season. Using the two evaluate factors that affect the abundance of marine birds
0
0
0m
0m
(Nov. 15 - Mar. 14) highest density categories for each species, relatively high in the study area, a regression model was developed (Seber
39°N
39°N
density areas associated with large bathymetric areas (inner 1977, Kleinbaum et al. 1988), with marine bird density as
shelf, outer shelf, upper slope) were identified, as well as with the dependent variable. Independent variables that could be
several smaller discrete habitat features (e.g., Monterey Bay addressed in the limited time frame included: ocean season,
Canyon) (p. 51). year, ocean depth, distance to nearest breeding colony,
distance to shelf break (estimated to the 200 meter isobath),
38°N
38°N
Summary of Overall Density, Biomass and Diversity Maps distance to deep ocean (estimated to the 2000 meter isobath),
for 76 Marine Bird Species. Overall marine bird densities latitude, periods of short-term ocean climate anomalies (e.g.,
were mapped for each season and for all seasons combined. El Niño or La Niña events), and latitude. The data used for
Densities of all species in a cell were converted to biomass by the multiple regression analyses was a subset of the mapping
37°N
37°N
multiplying density for each species by its average body mass data set; the regression data set included cell-based density
(from Dunning 1993), then summing for all species detected data from 1985 - 2002 (6,641 cell samples, all with effort ≥
in that cell. Biomass was then mapped in a fashion similar to 0.24km2 per cell).
the individual species’ density maps.
Response to Variation in Marine Climate. Short-term ocean
36°N
36°N
The Shannon Index (Shannon and Weaver 1949) climate anomaly in this report is often referred to as ENSO
(El Niño/Southern Oscillation), and generally refers to the
n n
S
H ′ = − ∑ i ln i climatic events that cause significant interannual changes in
i =1 n n
thermocline depth and water temperature in the study area,
was used to quantify species diversity. This index measures resulting in warm-water periods (often known as El Niño
35°N
35°N
c d the degree to which the species assemblage is dominated by events), cold-water periods (often known as La Niña events),
a single species. If species A dominates all the species seen or neutral periods, when the water is neither unusually warm
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
within a cell, then diversity is low, and vice versa. Diversity nor cold (Ainley et al. 1995b and references therein).
Source Data: See text.
was calculated for each season and all seasons combined. To
Figure 41. Total survey effort for marine bird analyses.
49
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
sets ranged from 1975-2001. Table 14 indicates the periods
The official ENSO events and time periods tracked by NOAA
of unusual weather (warm water, cold water, and neutral) as
are relevant for regions to the far south and well outside of the
determined from these data.
study area; the official NOAA ENSO periods do not accurately
reflect the timing of the ENSO-related periods that occur off
Also affecting marine climate are decadal-scale factors involved
central California. In part, this is because the marine climate of
in the Pacific Decadal Oscillation (Mantua and Hare 2002). A
the central and northern California Current region is affected
regime shift occurred in 1976, from cold to warm, and may
as well by variation in atmospheric pressure centers in the
have occurred again in the winter of 1998/1999, from warm to
Gulf of Alaska.
cold. This means that the overall average state of the system
could be characterized as warm or cold, with other shorter-
Table 14. Assignment of warm, cold and neutral periods,
based on surface water temperatures off Cental California. term variation embedded (e.g., ENSO). The effect of regime
shifts on marine bird occurrence is addressed near the end
Davidson
Current Upwelling Oceanic of this report.
Year Season Season Season
1975 Cold Cold Cold ANALYTICAL MAP PRODUCTS
The analytical map products (Figures 42-60) include seasonal
1976 Cold Cold Warm
density maps for 13 species and nine summary analyses maps
1977 Warm Cold Neutral
for marine birds. These maps are a subset of the total mapped
1978 Warm Warm Cold
results for this analysis. Additional maps and text products
1979 Cold Neutral Neutral
are included on the CD-ROM. Of the 35 species maps, these
1980 Warm Neutral Cold
ten were chosen for inclusion in the document because they
1981 Warm Cold Cold
represent a variety of spatial and temporal patterns in and
1982 Neutral Neutral Neutral
around the sanctuaries.
1983 Warm Warm Warm
1984 Warm Neutral Neutral
1985 Cold Warm Cold
1986 Neutral Neutral Neutral
1987 Warm Warm Warm
1988 Neutral Neutral Cold
1989 Cold Neutral Neutral
1990 Cold Cold Neutral
1991 Cold Cold Neutral
1992 Warm Warm Warm
1993 Warm Warm Warm
1994 Warm Neutral Cold
1995 Neutral Warm Neutral
1996 Warm Neutral Cold
1997 Neutral Neutral Warm
1998 Warm Warm Cold
1999 Cold Cold Cold
2000 Cold Cold Cold
-
2001 Cold Cold
To determine the time periods and effects of interannual climate
anomalies of marine birds as evidenced in the study area (i.e.,
warm-water, cold-water and neutral periods), two sea-surface
temperature data sets for central California were analyzed:
daily temperatures taken as part of a Scripps Institution of
Oceanography program at Southeast Farallon Island and
the NOAA CoastWatch data off central California. Both data
50
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS gram of the Southwest Fisheries Science Center, NMFS, NOAA
Figures 42a, b, and c show the density (birds/km2) of western
Western and Clark's Grebes (unpublished data).
Aechmophorus occidentalis, A. clarkii and Clark’s grebes (combined) in the Upwelling, Oceanic, and
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Davidson Current seasons, displayed in five minute latitude by Although the at-sea data span the years from 1980 to 2001,
Oceanic Season
Upwelling Season
200 m
200 m
20
20
five minute longitude cells. Densities are based on the combined data are not available for all seasons in all years. For the
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) data sets of several studies (see “Methods” and “Data Sources” Upwelling Season, data are from 1980-1982 and 1985-2001.
39°N
39°N
below). The color and mapping intervals were customized to For the Oceanic Season, data are from 1980-1982, 1991, and
Density show the most structure and to highlight significant areas, while 1994-2001. For the Davidson Current Season, data are from
(Animals/km²)
allowing comparisons among marine bird species. Cells that 1980-1986 and 1991-2001.
> 100.00
were surveyed but in which no grebes were observed have a
50.01 - 100.00
density of zero. Areas not surveyed appear white; no informa-
38°N
38°N
METHODS
10.01 - 50.00
tion is available for these areas. Blue lines indicate the boundar-
5.01 - 10.00 At-sea densities are the result of a synthesis of data from eight
1.01 - 5.00
ies of the National Marine Sanctuaries in the study area: Cordell shipboard and aerial survey programs conducted in the study
0.51 - 1.00
Bank, Gulf of the Farallones, and Monterey Bay. area in the years 1980-2001 (see “Data Sources” below). Bird
0.11 - 0.50
observation data and trackline data from these studies were
0.06 - 0.10
37°N
37°N
In order to provide one map for the species that integrates the
0.01 - 0.05 converted to a common format. All aerial data were continuous;
0.00
patterns of its spatial and temporal occurrence and abundance ship-based data were converted separately into a continuous
0 25 50 Km
in the study area, map d shows seasonal high-use areas, dis- transect to the extent possible. From the digitized survey data,
played in 10 minute latitude by 10 minute longitude cells. The the distributions of effort and of species were mapped into five
seasonal high use map provides a further synthesis of densities minute latitude by five minute longitude cells using CDAS, a
36°N
36°N
presented in maps a, b and c, and portrays the relative impor- custom geographic information system for analyzing marine
tance of various areas to the species. Areas with consistently bird and mammal surveys (MMS, 2001). The length and width
high use are highlighted on this map. To provide a relative refer- of the survey trackline in a given cell (estimated trackline width
ence for the “high use” areas, cells are also shown where the varied by platform, depending on speed and height above wa-
species were absent (i.e., the cell was sampled but the species ter) were used to estimate the area sampled. The number of
35°N
35°N
a b was not recorded there), or present but at lesser concentrations birds of each species seen in a cell was then divided by the area
in any particular season. See the "Methods" section below for sampled in the cell to estimate density. If a cell was surveyed
Davidson Current Season Seasonal High Use Areas and
200 m
200 m
20
further explanation of seasonal high-use areas.
20
more than once, densities were averaged, with an adjustment
0
0
0m
0m
Breeding Colonies
(Nov. 15 - Mar. 14) made for effort.
39°N
39°N
DATA SOURCES
Persistence of
High Use Densities for marine birds at sea are based on data from eight The seasonal high-use areas on map d were developed using
3 Seasons
survey programs conducted between 1980 and 2001, which a similar approach as for Maps a, b and c, but the data were
2 Seasons
were combined into a new MMS-CDAS data set (MMS, 2001) binned into 10’ x 10’ cells. For each season, the cells with den-
1 Season
Birds present
using software (CDAS) developed for the Minerals Management sities in the top 20% of non-zero values were designated “high
38°N
38°N
Birds absent
Service. Of the data sets on the original MMS-CDAS CD-ROM, use” for that season. Cells were scored for “high use” in one,
four aerial survey data sets contained data in the study area two, or three seasons and are depicted by color. To provide a
from Point Arena to Point Sal. Of these, the OSPR survey relative reference for the “high use” areas, cells are also shown
program is ongoing and data from recent years were added where the species were absent (i.e., the cell was sampled but
37°N
37°N
to this data set. In addition, data from four ship-based survey the species was not recorded there) or present (but densities
programs were converted to a compatible format for analysis were never in the top 20% for any season).
(see section overview for details on individual data sets).
RESULTS AND DISCUSSION
Data sources for aerial, at-sea data include MMS-CDAS (MMS Individuals of this closely-related species pair (separable by
36°N
36°N
2001), and California Department of Fish and Game, Office of plumage, but sharing the same ecological niche) are abundant
Spill Prevention and Response (CDF&G-OSPR, unpublished in the near-shore waters of the study area. Surveys tallied
data). Early data were collected using methods described by 2,511 sightings of 13,525 individuals. During most oil spills in
Briggs et al. (1983, 1987b); more recent data were collected this region, these species have been near the top of the list, by
using updated technology but using the same general method. number, of oiled birds. These birds breed inland at freshwater
35°N
35°N
These species do not
c d Data sources for ship-based survey data include: David Ainley lakes and marshes.
breed within the study area.
of H. T. Harvey and Associates and Carol Keiper of Oikonos
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
(unpublished data; see Oedekoven et al., 2001 for details on A multiple regression model of nine independent variables ex-
Source Data: See text.
survey methods); and Lisa T. Ballance, from the Ecology Pro- plained 15.5% of variation in cell density; most important vari-
Figure 42. Western and Clark’s grebes, seasonal density and high use areas.
51
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ables were season, and an inverse relationship with distance
to land and to depth; see Table 19. These results reflect the
large number of grebes found in shallow waters (mean depth
was 131 ± 37 m) within a few kilometers of shore (mean dis-
tance to land was 7.4 ± 1 km), and primarily during the Oceanic
Season. Moderate numbers are present during the Upwelling
and Davidson Current seasons. During the latter, these grebes
expanded farther offshore to the middle continental shelf (mean
depth of occurrence 260 ± 80 m).
Inshore waters of the Gulf of the Farallones (San Francisco Bay
tidal plume), Monterey Bay, and Estero/San Luis Obispo bays
had particularly high concentrations of these birds. North and
south of marine sanctuary boundaries in the study area, these
species were found only at isolated river mouths. Therefore, the
sanctuary boundaries encompass the majority of the species
habitat in the study area, except for the ‘sanctuary exclusion
area’ off San Francisco and Pacifica, which contained many
grebes. The broad continental shelf off central California is ideal
for these grebes, which capture prey by diving; it is likely they
are capable of exploiting most of the water column lying over
the shelf, in spite of their inshore occurrence. Abundance of this
species-pair remained stable between 1985 and 2002.
These grebes feed mainly on long-bodied, fusiform fish, such
as herring and anchovy. See Tables 15 and 16 for related sum-
mary information.
52
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
data). Early data were collected using methods described by
ABOUT THESE MAPS
Northern Fulmar Briggs et al. (1983, 1987b); more recent data were collected
Figures 43 a, b, and c show the density (birds/km2) of northern
Fulmarus glacialis using updated technology but using the same general method.
fulmar in the Upwelling, Oceanic, and Davidson Current
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Data sources for ship-based survey data include: David Ainley
seasons, displayed in five minute latitude by five minute
Upwelling Season Oceanic Season
200 m
200 m
20
20
of H. T. Harvey and Associates and Carol Keiper of Oikonos
longitude cells. Densities are based on the combined data sets
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) (unpublished data; see Oedekoven et al., 2001 for details
of several studies (see “Methods” and “Data Sources” below).
39°N
39°N
on survey methods); and Lisa T. Ballance, from the Ecology
The color and mapping intervals were customized to show the
Density Program of the Southwest Fisheries Science Center, NMFS,
most structure and to highlight significant areas, while allowing
(Animals/km²)
NOAA (unpublished data).
comparisons among marine bird species. Cells that were
> 100.00
surveyed but in which no fulmars were observed have a density
50.01 - 100.00
38°N
38°N
Although the at-sea data span the years from 1980 to 2001,
of zero. Areas not surveyed appear white; no information is
10.01 - 50.00
data are not available for all seasons in all years. For the
5.01 - 10.00
available for these areas. Blue lines indicate the boundaries
1.01 - 5.00
Upwelling Season, data are from 1980-1982 and 1985-2001.
of the National Marine Sanctuaries in the study area: Cordell
0.51 - 1.00
For the Oceanic Season, data are from 1980-1982, 1991, and
Bank, Gulf of the Farallones, and Monterey Bay.
0.11 - 0.50
1994-2001. For the Davidson Current Season, data are from
0.06 - 0.10
37°N
37°N
0.01 - 0.05 1980-1986 and 1991-2001.
In order to provide one map for the species that integrates the
0.00
patterns of its spatial and temporal occurrence and abundance
0 25 50 Km
METHODS
in the study area, map d shows seasonal high-use areas,
At-sea densities are the result of a synthesis of data from eight
displayed in 10 minute latitude by 10 minute longitude cells.
shipboard and aerial survey programs conducted in the study
The seasonal high use map provides a further synthesis of
36°N
36°N
area in the years 1980-2001 (see “Data Sources” below). Bird
densities presented in Maps a, b and c, and portrays the
observation data and trackline data from these studies were
relative importance of various areas to the species. Areas
converted to a common format. All aerial data were continuous;
with consistently high use are highlighted on this map. To
ship-based data were converted separately into a continuous
provide a relative reference for the “high use” areas, cells are
transect to the extent possible. From the digitized survey data,
also shown where the species were absent (i.e., the cell was
35°N
35°N
a b the distributions of effort and of species were mapped into five
sampled but the species was not recorded there), or present
minute latitude by five minute longitude cells using CDAS, a
but at lesser concentrations in any particular season. See the
Seasonal High Use Areas and
200 m
Davidson Current Season
200 m
20
custom geographic information system for analyzing marine
20
"Methods" section below for further explanation of seasonal
0
0
0m
0m
Breeding Colonies bird and mammal surveys (MMS, 2001). The length and width
(Nov. 15 - Mar. 14) high-use areas.
39°N
39°N
of the survey trackline in a given cell (estimated trackline width
Persistence of
High Use varied by platform, depending on speed and height above
Because the sighting data for this species extends beyond
3 Seasons
water) were used to estimate the area sampled. The number
the western extent of the standard map frame shown here,
2 Seasons
of birds of each species seen in a cell was then divided by
1 Season
additional maps were made that include a greater western
Birds present
the area sampled in the cell to estimate density. If a cell was
extent. These maps (with the word "pelagic" in the filename)
38°N
38°N
Birds absent
surveyed more than once, densities were averaged, with an
are included on the CDROM.
adjustment made for effort.
DATA SOURCES
The seasonal high-use areas on map d were developed using
Densities for marine birds at sea are based on data from eight
37°N
37°N
a similar approach as for Maps a, b and c, but the data were
survey programs conducted between 1980 and 2001, which
binned into 10’ x 10’ cells. For each season, the cells with
were combined into a new MMS-CDAS data set (MMS, 2001)
densities in the top 20% of non-zero values were designated
using software (CDAS) developed for the Minerals Management
“high use” for that season. Cells were scored for “high use”
Service. Of the data sets on the original MMS-CDAS CD-ROM,
in one, two, or three seasons and are depicted by color. To
four aerial survey data sets contained data in the study area
36°N
36°N
provide a relative reference for the “high use” areas, cells are
from Point Arena to Point Sal. Of these, the OSPR survey
also shown where the species were absent (i.e., the cell was
program is ongoing and data from recent years were added
sampled but the species was not recorded there) or present
to this data set. In addition, data from four ship-based survey
(but densities were never in the top 20% for any season).
programs were converted to a compatible format for analysis
(see section overview for details on individual data sets).
35°N
35°N
This species does not
c d RESULTS AND DISCUSSION
breed within the study area.
Northern fulmar, which nests on islands in the Aleutian Island
Data sources for aerial, at-sea data include MMS-CDAS
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
chain and Bering Sea, is common in waters of the continental
(2001), and California Department of Fish and Game, Office
Source Data: See text.
slope as well as the outer waters of the continental shelf off
of Spill Prevention and Response (CDF&G-OSPR, unpublished
Figure 43. Northern fulmar, seasonal density and high use areas.
53
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
north/central California. Surveys recorded 4,486 sightings of
6,345 individuals. In some winters, fulmars were particularly
abundant off this coast, such as in 1986, 1991, 1996, and
1999.
A multiple regression analysis of nine independent variables
explained 21.3% of the variability of this species’ cell density;
important explanatory variables were season, ENSO period
(periods of unusually warm or cold sea temperatures), and year;
see Table 19. The species’ occurrence is confined principally to
the Davidson Current Season, especially prevalent during La
Niña. For a subarctic species, surprisingly high densities are
present during the Upwelling Season; many of these individuals
exhibit heavy molt indicating that they might be juveniles.
Based on the data available, the species’ population trajectory
during the study period exhibited a curvilinear pattern: a slight
decline between 1985 and 1989, followed by an increase from
1990 to 2002. Numbers rose particularly in the last few years,
perhaps indicating a response to the shift in 1999 from a warm
to a cold ocean regime (see subsection on response to climate
change).
Like the albatrosses, this species is attracted to trawlers, where
the species scavenges offal. Therefore, areas of concentration
for northern fulmars during the study period were (and may still
be) important areas of traditionally higher fishing activity such as
Cordell Bank, Fanny Shoal, and nearby canyons. This pattern
is better illustrated during the Upwelling Season, when the
species is much less abundant. In the latter season, the species
spreads far more widely, occurring farther offshore and over
deeper depths. Although fulmars are widespread off central
California, the boundaries of the National Marine Sanctuaries
encompass an important area for this species.
Northern fulmars are generalists that feed on live and dead prey
found at the surface. They are one of the few marine species
that feed extensively on gelatinous zooplankton, e.g. jellyfish.
See Tables 15 and 16 for related summary information.
54
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Data sources for aerial, at-sea data include MMS-CDAS (MMS,
ABOUT THESE MAPS
Sooty Shearwater 2001), and California Department of Fish and Game, Office of
Figures 44a, b, and c show the density (birds/km2) of sooty
Puffinus griseus Spill Prevention and Response (CDF&G-OSPR, unpublished
shearwater in the Upwelling, Oceanic, and Davidson Current
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
data). Early data were collected using methods described by
seasons, displayed in five minute latitude by five minute
Upwelling Season Oceanic Season
200 m
200 m
20
20
Briggs et al. (1983, 1987b); more recent data were collected
longitude cells. Densities are based on the combined data
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) using updated technology but using the same general method.
sets of several studies (see “Methods” and “Data Sources”
39°N
39°N
Data sources for ship-based survey data include: David Ainley
below). The color and mapping intervals were customized to
Density
of H. T. Harvey and Associates and Carol Keiper of Oikonos
show the most structure and to highlight significant areas,
(Animals/km²)
(unpublished data; see Oedekoven et al., 2001 for details
while allowing comparisons among marine bird species. Cells
> 100.00
on survey methods); and Lisa T. Ballance, from the Ecology
that were surveyed but in which no sooty shearwaters were
50.01 - 100.00
38°N
38°N
Program of the Southwest Fisheries Science Center, NMFS,
observed have a density of zero. Areas not surveyed appear
10.01 - 50.00
5.01 - 10.00
NOAA (unpublished data).
white; no information is available for these areas. Blue lines
1.01 - 5.00
indicate the boundaries of the National Marine Sanctuaries
0.51 - 1.00
Although the at-sea data span the years from 1980 to 2001,
in the study area: Cordell Bank, Gulf of the Farallones, and
0.11 - 0.50
data are not available for all seasons in all years. For the
Monterey Bay.
0.06 - 0.10
37°N
37°N
0.01 - 0.05
Upwelling Season, data are from 1980-1982 and 1985-2001.
0.00
For the Oceanic Season, data are from 1980-1982, 1991, and
In order to provide one map for the species that integrates the
0 25 50 Km
1994-2001. For the Davidson Current Season, data are from
patterns of its spatial and temporal occurrence and abundance
1980-1986 and 1991-2001.
in the study area, map d shows seasonal high-use areas,
displayed in 10 minute latitude by 10 minute longitude cells.
36°N
36°N
METHODS
The seasonal high use map provides a further synthesis of
At-sea densities are the result of a synthesis of data from eight
densities presented in Maps a, b and c, and portrays the
shipboard and aerial survey programs conducted in the study
relative importance of various areas to the species. Areas
area in the years 1980-2001 (see “Data Sources” below). Bird
with consistently high use are highlighted on this map. To
observation data and trackline data from these studies were
provide a relative reference for the “high use” areas, cells are
35°N
35°N
a b converted to a common format. All aerial data were continuous;
also shown where the species were absent (i.e., the cell was
ship-based data were converted separately into a continuous
sampled but the species was not recorded there), or present
Seasonal High Use Areas and
200 m
Davidson Current Season
200 m
20
20
transect to the extent possible. From the digitized survey data,
but at lesser concentrations in any particular season. See the
0
0
0m
0m
Breeding Colonies
(Nov. 15 - Mar. 14) the distributions of effort and of species were mapped into five
"Methods" section below for further explanation of seasonal
39°N
39°N
minute latitude by five minute longitude cells using CDAS, a
high-use areas.
Persistence of
High Use custom geographic information system for analyzing marine
3 Seasons
bird and mammal surveys (MMS, 2001). The length and width
Because the sighting data for this species extends beyond
2 Seasons
1 Season
of the survey trackline in a given cell (estimated trackline width
the western extent of the standard map frame shown here,
Birds present
varied by platform, depending on speed and height above
38°N
38°N
additional maps were made that include a greater western
Birds absent
water) were used to estimate the area sampled. The number
extent. These maps (with the word "pelagic" in the file name)
of birds of each species seen in a cell was then divided by
are included on the CDROM.
the area sampled in the cell to estimate density. If a cell was
surveyed more than once, densities were averaged, with an
DATA SOURCES
37°N
37°N
adjustment made for effort.
Densities for marine birds at sea are based on data from eight
survey programs conducted between 1980 and 2001, which
The seasonal high-use areas on map d were developed using
were combined into a new MMS-CDAS data set (MMS, 2001)
a similar approach as for Maps a, b and c, but the data were
using software (CDAS) developed for the Minerals Management
binned into 10’ x 10’ cells. For each season, the cells with
Service. Of the data sets on the original MMS-CDAS CD-ROM,
36°N
36°N
densities in the top 20% of non-zero values were designated
four aerial survey data sets contained data in the study area
“high use” for that season. Cells were scored for “high use”
from Point Arena to Point Sal. Of these, the OSPR survey
in one, two, or three seasons and are depicted by color. To
program is ongoing and data from recent years were added
provide a relative reference for the “high use” areas, cells are
to this data set. In addition, data from four ship-based survey
also shown where the species were absent (i.e., the cell was
programs were converted to a compatible format for analysis
35°N
35°N
This species does not
c d breed within the study area. sampled but the species was not recorded there) or present
(see section overview for details on individual data sets).
(but densities were never in the top 20% for any season).
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Source Data: See text.
Figure 44. Sooty shearwater, seasonal density and high use areas.
55
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
RESULTS AND DISCUSSION
Sooty shearwaters nest in the sub-Antarctic, particularly on
island of Tierra del Fuego and New Zealand, and winters in
the Peru and California current regions. During the Upwelling
Season, the sooty shearwater is the most abundant marine
bird off California, and this is the case, by far, for waters within
the boundaries of the north/central California national marine
sanctuaries. Surveys tallied 20,750 sightings of 323,176
individuals, indicating that the species usually occurs in large
concentrations.
A multiple regression analysis of nine independent variables
explained 43.3% of the variation in cell density, with season,
an inverse relationship to year, and ENSO period (periods of
unusually warm or cold sea temperatures) being the most
important variables; see Table 19. These results further reflect
the restriction of this species’ occurrence off California largely to
the Upwelling Season, and to greater abundance when ocean
climate is unaffected by short-term climate anomaly. In other
words, sooty shearwaters were less abundant in the study area
during El Niño and La Niña. From a decadal perspective they
declined over the years, although this effect was curvilinear:
a slight increase between 1985 and 1991, a steep decline to
1998, and a moderate increase subsequently. Whether or not
the latter increase is a response to the shift to a cold ocean
regime in 1999 remains to be seen. The continental shelf and
upper slope are the main habitats frequented by this species
(mean ocean depth where sooty shearwaters occurred was
380 ± 10m).
The sooty shearwater was present in greatest densities in
Monterey Bay. Throughout the California current (Veit et al,
1997), this species has declined severely in abundance during
the recent warm regime (1976-1999), as noted above. Even
now, though, it is still very abundant in Monterey Bay, probably
because of the large anchovy source there. Other important
areas (but not comparable to Monterey Bay), include Pioneer
and Ascension canyons, Farallon Escarpment and Fanny
Shoal, as well as the ocean area off Pacifica and Estero/San
Luis Obispo bays. National marine sanctuary waters become
even more important to this species during the Oceanic Season,
as remnants of the population, just before their long southward
migration, fatten on the oil-rich anchovies.
Sooty shearwaters feed on fish, squid, and invertebrates that
they acquire by pursuit, plunging to a depth of 10-15 m. During
the early Upwelling Season the main prey are euphausiids and
squid, a diet that shifts more to oily fish, such as anchovy, in
the late Upwelling Season. See Tables 15 and 16 for related
summary information.
56
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS Data sources for aerial, at-sea data include MMS-CDAS (MMS,
Ashy Storm-Petrel Figures 45a, b, and c show the density (birds/km2) of ashy 2001), California Department of Fish and Game, Office of Spill
Oceanodroma homochroa 32
storm-petrel in the Upwelling, Oceanic, and Davidson Current Prevention and Response (CDF&G-OSPR, unpublished data).
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
seasons, displayed in five minute latitude by five minute Early data were collected using methods described by Briggs
Upwelling Season Oceanic Season
200 m
200 m
20
20
longitude cells. Densities are based on the combined data et al. (1983, 1987b); more recent data were collected using
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) sets of several studies (see “Methods” and “Data Sources” updated technology but using the same general method. Data
39°N
39°N
below). The color and mapping intervals were customized sources for ship-based survey data include: David Ainley of
Density to show the most structure and to highlight significant areas, H. T. Harvey and Associates and Carol Keiper of Oikonos
(Animals/km²)
while allowing comparisons among marine bird species. Cells (unpublished data; see Oedekoven et al., 2001 for details
> 100.00
that were surveyed but in which no ashy storm-petrels were on survey methods); and Lisa T. Ballance, from the Ecology
50.01 - 100.00
38°N
38°N
observed have a density of zero. Areas not surveyed appear Program of the Southwest Fisheries Science Center, NMFS,
10.01 - 50.00
white; no information is available for these areas. Blue lines NOAA (unpublished data). Data on ashy storm-petrel colonies
5.01 - 10.00
1.01 - 5.00
indicate the boundaries of the National Marine Sanctuaries were obtained from Carter et al. (1992) supplemented by
0.51 - 1.00
in the study area: Cordell Bank, Gulf of the Farallones, and Sydeman et al. (1998), Whitworth et al. (2002).
0.11 - 0.50
Monterey Bay.
0.06 - 0.10
37°N
37°N
Although the at-sea data span the years from 1980 to 2001,
0.01 - 0.05
0.00
In order to provide one map for the species that integrates the data are not available for all seasons in all years. For the
0 25 50 Km
patterns of its spatial and temporal occurrence and abundance Upwelling Season, data are from 1980-1982 and 1985-2001.
in the study area, map d shows seasonal high-use areas, For the Oceanic Season, data are from 1980-1982, 1991 and
displayed in 10 minute latitude by 10 minute longitude cells, 1994-2001. For the Davidson Current Season, data are from
36°N
36°N
and breeding colonies. The seasonal high use map provides 1980-1986 and 1991-2001.
a further synthesis of densities presented in Maps a, b and c,
and portrays the relative importance of various areas to the METHODS
species. Areas with consistently high use are highlighted on this At-sea densities are the result of a synthesis of data from eight
map. To provide a relative reference for the “high use” areas, shipboard and aerial survey programs conducted in the study
35°N
35°N
a b cells are also shown where the species were absent (i.e., the area in the years 1980-2001 (see “Data Sources” below). Bird
cell was sampled but the species was not recorded there), or observation data and trackline data from these studies were
Seasonal High Use Areas and
200 m
Davidson Current Season
200 m
20
present but at lesser concentrations in any particular season. converted to a common format. All aerial data were continuous;
20
0
0
0m
0m
Breeding Colonies See the "Methods" section below for further explanation of ship-based data were converted separately into a continuous
(Nov. 15 - Mar. 14)
39°N
39°N
seasonal high-use areas. Breeding colonies are also shown; transect to the extent possible. From the digitized survey data,
Persistence of
High Use the relative size of the symbols indicates the colony size. the distributions of effort and of species were mapped into five
3 Seasons
minute latitude by five minute longitude cells using CDAS, a
2 Seasons
1 Season
Because the sighting data for this species extends beyond custom geographic information system for analyzing marine
Birds present
the western extent of the standard map frame shown here, bird and mammal surveys (MMS, 2001). The length and width
38°N
38°N
Birds absent
additional maps were made that include a greater western of the survey trackline in a given cell (estimated trackline width
Colony Size
(Breeding birds)
extent. These maps (with the word "pelagic" in the filename) varied by platform, depending on speed and height above
10,000 - 50,000
are included on the CDROM. water) were used to estimate the area sampled. The number
of birds of each species seen in a cell was then divided by
5001 - 10,000
37°N
37°N
DATA SOURCES the area sampled in the cell to estimate density. If a cell was
1001 - 5000
501 - 1000
Densities for marine birds at sea are based on data from eight surveyed more than once, densities were averaged, with an
101 - 500
survey programs conducted between 1980 and 2001, which adjustment made for effort.
51 - 100
2 - 50
were combined into a new MMS-CDAS data set (MMS, 2001)
Historical
using software (CDAS) developed for the Minerals Management The seasonal high-use areas on map d were developed using
36°N
36°N
Service. Of the data sets on the original MMS-CDAS CD-ROM, a similar approach as for Maps a, b and c, but the data were
four aerial survey data sets contained data in the study area binned into 10’x10’ cells. For each season, the cells with
from Point Arena to Point Sal. Of these, the OSPR survey densities in the top 20% of non-zero values were designated
program is ongoing and data from recent years were added “high use” for that season. Cells were scored for “high use”
to this data set. In addition, data from four ship-based survey in one, two, or three seasons and are depicted by color. To
35°N
35°N
c d programs were converted to a compatible format for analysis provide a relative reference for the “high use” areas, cells are
(see section overview for details on individual data sets). also shown where the species were absent (i.e., the cell was
124°W 123°W 122°W 121°W
124°W 123°W 122°W 121°W
sampled but the species was not recorded there) or present
Source Data: See text.
(but densities were never in the top 20% for any season).
Figure 45. Ashy storm-petrel, seasonal density, high use areas, and breeding colonies.
57
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
RESULTS AND DISCUSSION
Ashy storm-petrel is endemic to the California Current and is
considered by the State to be a “Species of Special Concern”;
a major colony is at the Farallon Islands. It is common in the
study area and the most abundant storm-petrel in waters of
the central California national marine sanctuaries. Surveys
recorded 1,472 sightings of 4,339 individuals.
A multiple regression model of nine variables explained 17.3%
of variation in cell density, with important explanatory variables
being ENSO period (i.e., periods of unusually warm or cold ocean
temperature), season, and year; see Table 19. The species was
more abundant during the Oceanic Season and during years of
La Niña, indicating that when ocean temperatures were cold,
Ashy storm-petrels were concentrated closer to the Farallon
breeding colony, which they visit only at night. During nesting
(Upwelling Season), this species occupies waters mainly over
the outer slope (mean depth of occurrence 1,615 ± 52 m),
mostly outside of National Marine Sanctuary boundaries. During
the period of molt (Oceanic Season), ashy storm-petrels move
inshore to frequent shallower slope waters (mean depth of
occurrence 1,144 ± 61 m) and a large concentration occurred
over the Monterey Bay canyon as shown in maps on upwelling
and seasonal high use areas.
In recent years, however, this post-breeding concentration
has shifted to the area around Cordell Bank (not shown on
the maps). As the species begins its seasonal return to the
Farallon nesting colony (Davidson Current Season), they again
shift north to deeper waters of the outer slope (mean depth
of occurrence then was 2,579 ± 121 m). The species seems
to be most dispersed during the Davidson Current Season,
but in all seasons the Farallon Escarpment is by far its most
important area.
Overall, ashy storm-petrel numbers increased from 1985 to
2002 in a curvilinear fashion: steeper increase in numbers
between 1985 and 1992, followed by a less steep increase
to 2002.
This species feeds on invertebrates and larval fish found
at the surface. See Tables 15 and 16 for related summary
information.
58
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS programs were converted to a compatible format for analysis
Figures 46a, b, and c show the density (birds/km2) of Leach’s (see section overview for details on individual data sets).
Leach's Storm-Petrel Oceanodroma leucorhoa storm-petrel in the Upwelling, Oceanic, and Davidson Current
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
seasons, displayed in five minute latitude by five minute Data sources for aerial, at-sea data include MMS-CDAS (MMS,
Upwelling Season Oceanic Season
200 m
longitude cells. Densities are based on the combined data 2001), California Department of Fish and Game, Office of Spill
200 m
20
20
0
0
0m
0m
sets of several studies (see “Methods” and “Data Sources” Prevention and Response (CDF&G-OSPR, unpublished data).
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14)
39°N
39°N
below). The color and mapping intervals were customized to Early data were collected using methods described by Briggs
Density show the most structure and to highlight significant areas, while et al. (1983, 1987b); more recent data were collected using
(Animals/km²)
allowing comparisons among marine bird species. Cells that updated technology but using the same general method.
> 100.00
were surveyed but in which no Leach’s storm-petrels were Data sources for ship-based survey data include: David
50.01 - 100.00
observed have a density of zero. Areas not surveyed appear Ainley of H. T. Harvey and Associates and Carol Keiper of
38°N
38°N
10.01 - 50.00
white; no information is available for these areas. Blue lines Oikonos (unpublished data; see Oedekoven et al., 2001 for
5.01 - 10.00
1.01 - 5.00
indicate the boundaries of the National Marine Sanctuaries details on survey methods); and Lisa T. Ballance, from the
0.51 - 1.00
in the study area: Cordell Bank, Gulf of the Farallones, and Ecology Program of the Southwest Fisheries Science Center,
0.11 - 0.50
Monterey Bay. An additional set of maps was done for this NMFS, NOAA (unpublished data). Data on Leach’s storm-
0.06 - 0.10
37°N
37°N
species to show the offshore extent of its distribution; these petrel colonies were obtained from Carter et al. (1992, and
0.01 - 0.05
0.00 maps are included on the CD-ROM. supplements).
0 25 50 Km
In order to provide one map for the species that integrates the Although the at-sea data span the years from 1980 to 2001,
patterns of its spatial and temporal occurrence and abundance data are not available for all seasons in all years. For the
36°N
36°N
in the study area, map d shows seasonal high-use areas, Upwelling Season, data are from 1980-1982 and 1985-2001.
displayed in 10 minute latitude by 10 minute longitude cells, For the Oceanic Season, data are from 1980-1982, 1991, and
and breeding colonies. The seasonal high use map provides a 1994-2001. For the Davidson Current Season, data are from
further synthesis of densities presented in Maps a, b and c, and 1980-1986 and 1991-2001.
portrays the relative importance of various areas to the species.
35°N
35°N
a b Areas with consistently high use are highlighted on this map. To METHODS
provide a relative reference for the “high use” areas, cells are At-sea densities are the result of a synthesis of data from eight
Seasonal High Use Areas and also shown where the species were absent (i.e., the cell was shipboard and aerial survey programs conducted in the study
200 m
Davidson Current Season
200 m
20
20
0
0
0m
0m
sampled but the species was not recorded there), or present area in the years 1980-2001 (see “Data Sources” below). Bird
Breeding Colonies
(Nov. 15 - Mar. 14)
39°N
39°N
but at lesser concentrations in any particular season. See the observation data and trackline data from these studies were
Persistence of
"Methods" section below for further explanation of seasonal converted to a common format. All aerial data were continuous;
High Use
3 Seasons
high-use areas. Breeding colonies are also shown; the relative ship-based data were converted separately into a continuous
2 Seasons
size of the symbols indicates the colony size. An additional set transect to the extent possible. From the digitized survey data,
1 Season
Birds present
of maps was developed for this species to include the offshore the distributions of effort and of species were mapped into five
38°N
38°N
Birds absent
extent of its distribution. These maps are on the CD-ROM. minute latitude by five minute longitude cells using CDAS, a
Colony Size
(Breeding birds) custom geographic information system for analyzing marine
10,000 - 50,000
Because the sighting data for this species extends beyond bird and mammal surveys (MMS, 2001). The length and width
the western extent of the standard map frame shown here, of the survey trackline in a given cell (estimated trackline width
5001 - 10,000
37°N
37°N
additional maps were made that include a greater western varied by platform, depending on speed and height above
1001 - 5000
501 - 1000
extent. These maps (with the word "pelagic" in the filename) water) were used to estimate the area sampled. The number
101 - 500
are included on the CDROM. of birds of each species seen in a cell was then divided by
51 - 100
2 - 50
the area sampled in the cell to estimate density. If a cell was
Historical
DATA SOURCES surveyed more than once, densities were averaged, with an
36°N
36°N
Densities for marine birds at sea are based on data from eight adjustment made for effort.
survey programs conducted between 1980 and 2001, which
were combined into a new MMS-CDAS data set (MMS, 2001) The seasonal high-use areas on map d were developed using
using software (CDAS) developed for the Minerals Management a similar approach as for Maps a, b and c, but the data were
Service. Of the data sets on the original MMS-CDAS CD-ROM, binned into 10’x10’ cells. For each season, the cells with
35°N
35°N
c d four aerial survey data sets contained data in the study area densities in the top 20% of non-zero values were designated
from Point Arena to Point Sal. Of these, the OSPR survey “high use” for that season. Cells were scored for “high use”
124°W 123°W 122°W 121°W
124°W 123°W 122°W 121°W
program is ongoing and data from recent years were added in one, two, or three seasons and are depicted by color. To
Source Data: See text.
to this data set. In addition, data from four ship-based survey provide a relative reference for the “high use” areas, cells are
Figure 46. Leach’s storm-petrel, seasonal density, high use areas, and breeding colonies.
59
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
also shown where the species were absent (i.e., the cell was
sampled but the species was not recorded there) or present
(but densities were never in the top 20% for any season).
RESULTS AND DISCUSSION
The Leach’s storm-petrel, which has a breeding population
numbering in the millions in Alaska, is represented south to
Baja California by smaller colonies as latitude decreases. In
comparison, the estimated 12,551 birds breeding along the
California coast is miniscule (Carter et al, 1992). This, and
the fact that this species is highly migratory, suggests that
many of the birds seen in the study areas are migrants from
the north. This was also indicated by the lack of importance in
a multiple regression model of distance to colony as a factor
explaining this species’ variation in cell density; see Table 19.
Surveys recorded 1,118 sightings of 1,576 individuals, although
survey effort was sparse in the offshore waters this species
frequents.
This common species frequents waters much farther offshore
than the other storm-petrels, i.e. well beyond the continental
slope. Thus, the National Marine Sanctuary boundaries (and
most of the data sets in this study) do not encompass much of
this species’ preferred habitat. The species was most abundant
during the Upwelling Season (breeding) and occurred in greater
numbers closer to the coast. They visit the Farallon colony only
at night, but are at-sea during the day. During the Oceanic
and Davidson Current seasons few occurred near the shelf.
The birds present during the latter two seasons likely were
migrants from more northern populations. Given the huge North
Pacific population, the number recorded during surveys in the
study area was relatively small, because they were mostly far
offshore and not observed as often in the surveys available
for this assessment.
Yet, a multiple regression model of nine independent variables
explained 28.4% of variation in cell density, indicating that this
species responded consistently to the variables examined.
Most important of the nine variables were season, distance to
the 2000 m isobath, and ENSO period (periods of unusually
warm or cold ocean temperature); see Table 19. Abundance
of this species in the study area has increased between 1985
and 2002, and it was more abundant during periods of warm-
water conditions.
This species feeds on invertebrates captured at the surface.
See Tables 15 and 16 for related summary information.
60
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS Lisa T. Ballance, from the Ecology Program of the Southwest
Figures 47a, b, and c show the combined density (birds/km2) Fisheries Science Center, NMFS, NOAA (unpublished data).
Black, Surf and White-winged Scoters Melanitta nigra, M. perspicillata, M. fusca of three scoter species (white-winged, surf, and black) in the
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Upwelling, Oceanic, and Davidson Current seasons, displayed Although the at-sea data span the years from 1980 to 2001,
Upwelling Season Oceanic Season
200 m
200 m
20
20
in five minute latitude by five minute longitude cells. Densities data are not available for all seasons in all years. For the
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) are based on the combined data sets of several studies (see Upwelling Season, data are from 1980-1982 and 1985-2001.
39°N
39°N
“Methods” and “Data Sources” below). The color and mapping For the Oceanic Season, data are from 1980-1982, 1991, and
Density
intervals were customized to show the most structure and to 1994-2001. For the Davidson Current Season, data are from
(Animals/km²)
highlight significant areas, while allowing comparisons among 1980-1986 and 1991-2001.
> 100.00
marine bird species. Cells that were surveyed but in which
50.01 - 100.00
38°N
38°N
no scoters were observed have a density of zero. Areas not METHODS
10.01 - 50.00
5.01 - 10.00
surveyed appear white; no information is available for these At-sea densities are the result of a synthesis of data from eight
1.01 - 5.00
areas. Blue lines indicate the boundaries of the National Marine shipboard and aerial survey programs conducted in the study
0.51 - 1.00
Sanctuaries in the study area: Cordell Bank, Gulf of the Faral- area in the years 1980-2001 (see “Data Sources” below). Bird
0.11 - 0.50
lones, and Monterey Bay. observation data and trackline data from these studies were
0.06 - 0.10
37°N
37°N
0.01 - 0.05
converted to a common format. All aerial data were continuous;
0.00
In order to provide one map for the species that integrates the ship-based data were converted separately into a continuous
0 25 50 Km
patterns of its spatial and temporal occurrence and abundance transect to the extent possible. From the digitized survey data,
in the study area, map d shows seasonal high-use areas, dis- the distributions of effort and of species were mapped into five
played in 10 minute latitude by 10 minute longitude cells. The minute latitude by five minute longitude cells using CDAS, a
36°N
36°N
seasonal high use map provides a further synthesis of densities custom geographic information system for analyzing marine
presented in Maps a, b and c, and portrays the relative impor- bird and mammal surveys (MMS, 2001). The length and width
tance of various areas to the species. Areas with consistently of the survey trackline in a given cell (estimated trackline width
high use are highlighted on this map. To provide a relative refer- varied by platform, depending on speed and height above wa-
ence for the “high use” areas, cells are also shown where the ter) were used to estimate the area sampled. The number of
35°N
35°N
a b species were absent (i.e., the cell was sampled but the species birds of each species seen in a cell was then divided by the area
was not recorded there), or present but at lesser concentrations sampled in the cell to estimate density. If a cell was surveyed
Seasonal High Use Areas and
200 m
Davidson Current Season
200 m
20
20
in any particular season. See the "Methods" section below for more than once, densities were averaged, with an adjustment
0
0
0m
0m
Breeding Colonies
(Nov. 15 - Mar. 14) further explanation of seasonal high-use areas. made for effort.
39°N
39°N
Persistence of
High Use DATA SOURCES The seasonal high-use areas on map d were developed using
3 Seasons
Densities for marine birds at sea are based on data from eight a similar approach as for Maps a, b and c, but the data were
2 Seasons
1 Season
survey programs conducted between 1980 and 2001, which binned into 10’x10’ cells. For each season, the cells with densi-
Birds present
were combined into a new MMS-CDAS data set (MMS, 2001) ties in the top 20% of non-zero values were designated “high
38°N
38°N
Birds absent
using software (CDAS) developed for the Minerals Manage- use” for that season. Cells were scored for “high use” in one,
ment Service. Of the data sets on the original MMS-CDAS CD- two, or three seasons and are depicted by color. To provide a
ROM, four aerial survey data sets contained data in the study relative reference for the “high use” areas, cells are also shown
area from Point Arena to Point Sal.. Of these, the OSPR survey where the species were absent (i.e., the cell was sampled but
37°N
37°N
program is ongoing and data from recent years were added the species was not recorded there) or present (but densities
to this data set. In addition, data from four ship-based survey were never in the top 20% for any season).
programs were converted to a compatible format for analysis
(see section overview for details on individual data sets). RESULTS AND DISCUSSION
The distribution of white-winged, surf, and black scoters in
36°N
36°N
Data sources for aerial, at-sea data include MMS-CDAS (MMS, the north/central California study area is very similar to that
2001), California Department of Fish and Game, Office of Spill of the grebes (see above), although they are somewhat less
Prevention and (CDF&G-OSPR, unpublished data). Early data abundant and found closer to shore. There they forage mostly
were collected using methods described by Briggs et al. (1983, just outside the surf break. On the outer coast, the abundant
1987b); more recent data were collected using updated tech- surf scoter dominates over the other two scoters, and black
35°N
35°N
These species do not
c d nology but using the same general method. Data sources for scoters, which occur in more protected waters, are rare. Sur-
breed within the study area.
ship-based survey data include: David Ainley of H. T. Harvey veys recorded 1,787 sightings of scoters that included 42,691
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
and Associates and Carol Keiper of Oikonos (unpublished data; individuals; more than half were identified as surf scoter. The
Source Data: See text.
see Oedekoven et al., 2001 for details on survey methods); and most important areas for surf scoters within the study area is
Figure 47. Black, surf and white-winged scoters, seasonal density and high use areas.
61
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
the San Francisco Bay tidal plume, especially southward along
the Pacifica shore to Half Moon Bay, and the shallow parts of
Bodega, Monterey, Estero, and San Luis Obispo bays. These
birds nest in the arctic tundra along the north slope of North
America; specific nesting areas of birds found wintering in the
marine sanctuary boundaries have not been identified.
The apparent movement of these sea ducks’ offshore, i.e. to
the outer parts of the Gulf of the Farallones, in the Upwelling
Season is an artifact of their migration north or south, to or from
Alaskan breeding areas. That portion of the population wintering
south of central California takes the shortest distance across
the Gulf of the Farallones; the offshore density cells highlighted
in the maps is a record of flying scoters.
These scoters do not forage far from the mainland beach, where
they eat invertebrates; several dozen usually winter around the
Farallon Islands. The inshore distribution of these ducks, like
the grebes, makes them vulnerable to coastal oil spills. See
Tables 15 and 16 for related summary information.
62
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS of H. T. Harvey and Associates and Carol Keiper of Oikonos
Brown Pelican Figures 48a, b, and c show the density (birds/km2) of brown (unpublished data; see Oedekoven et al., 2001 for details
Pelecanus occidentalis pelicans in the Upwelling, Oceanic, and Davidson Current on survey methods); and Lisa T. Ballance, from the Ecology
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
seasons, displayed in five minute latitude by five minute Program of the Southwest Fisheries Science Center, NMFS,
Oceanic Season
Upwelling Season
200 m
200 m
20
20
longitude cells. Densities are based on the combined data NOAA (unpublished data). Data on brown pelican colonies were
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) sets of several studies (see “Methods” and “Data Sources” obtained from Carter et al. (1992, and supplements).
39°N
39°N
below). The color and mapping intervals were customized to
Density
show the most structure and to highlight significant areas, while Although the at-sea data span the years from 1980 to 2001,
(Animals/km²)
allowing comparisons among marine bird species. Cells that data are not available for all seasons in all years. For the
> 100.00
were surveyed but in which no brown pelicans were observed Upwelling Season, data are from 1980-1982 and 1985-2001.
50.01 - 100.00
38°N
38°N
have a density of zero; areas not surveyed are white. Blue lines For the Oceanic Season, data are from 1980-1982, 1991 and
10.01 - 50.00
5.01 - 10.00 indicate the boundaries of the National Marine Sanctuaries 1994-2001. For the Davidson Current Season, data are from
1.01 - 5.00
in the study area: Cordell Bank, Gulf of the Farallones, and 1980-1986 and 1991-2001.
0.51 - 1.00
Monterey Bay.
0.11 - 0.50
METHODS
0.06 - 0.10
37°N
37°N
0.01 - 0.05
In order to provide one map for the species that integrates the At-sea densities are the result of a synthesis of data from eight
0.00
patterns of its spatial and temporal occurrence and abundance shipboard and aerial survey programs conducted in the study
0 25 50 Km
in the study area, map d shows seasonal high-use areas, area in the years 1980-2001 (see “Data Sources” below). Bird
displayed in 10 minute latitude by 10 minute longitude cells, observation data and trackline data from these studies were
and breeding colonies (in this species’ case, a site where it converted to a common format. All aerial data were continuous;
36°N
36°N
bred in the past). The seasonal high use map provides a ship-based data were converted separately into a continuous
further synthesis of densities presented in Maps a, b and c, transect to the extent possible. From the digitized survey data,
and portrays the relative importance of various areas to the the distributions of effort and of species were mapped into five
species. Areas with consistently high use are highlighted on this minute latitude by five minute longitude cells using CDAS, a
map. To provide a relative reference for the “high use” areas, custom geographic information system for analyzing marine
35°N
35°N
a b cells are also shown where the species were absent (i.e., the bird and mammal surveys (MMS, 2001). The length and width
cell was sampled but the species was not recorded there), or of the survey trackline in a given cell (estimated trackline width
Seasonal High Use Areas and
Davidson Current Season
200 m
200 m
20
20
present but at lesser concentrations in any particular season. varied by platform, depending on speed and height above
0
0
0m
0m
Breeding Colonies
(Nov. 15 - Mar. 14) See the "Methods" section below for further explanation of water) were used to estimate the area sampled. The number
39°N
39°N
seasonal high-use areas. Breeding colonies are also shown; of birds of each species seen in a cell was then divided by
Persistence of
High Use the relative size of the symbols indicates the colony size. the area sampled in the cell to estimate density. If a cell was
3 Seasons
surveyed more than once, densities were averaged, with an
2 Seasons
1 Season
DATA SOURCES adjustment made for effort.
Birds present
38°N
38°N
Densities for marine birds at sea are based on data from eight
Birds absent
survey programs conducted between 1980 and 2001, which The seasonal high-use areas on map d were developed using
were combined into a new MMS-CDAS data set (MMS, 2001) a similar approach as for Maps a, b and c, but the data were
using software (CDAS) developed for the Minerals Management binned into 10’x10’ cells. For each season, the cells with
Service. Of the data sets on the original MMS-CDAS CD-ROM, densities in the top 20% of non-zero values were designated
37°N
37°N
four aerial survey data sets contained data in the study area “high use” for that season. Cells were scored for “high use”
from Point Arena to Point Sal. Of these, the OSPR survey in one, two, or three seasons and are depicted by color. To
program is ongoing and data from recent years were added provide a relative reference for the “high use” areas, cells are
to this data set. In addition, data from four ship-based survey also shown where the species were absent (i.e., the cell was
programs were converted to a compatible format for analysis sampled but the species was not recorded there) or present
36°N
36°N
(see section overview for details on individual data sets). (but densities were never in the top 20% for any season).
Data sources for aerial, at-sea data include MMS-CDAS (MMS, RESULTS AND DISCUSSION
2001), and California Department of Fish and Game, Office of Brown pelicans are included in the State and Federal
Spill Prevention and Response (CDF&G-OSPR, unpublished endangered species lists, and are common year-round in
35°N
35°N
This species does not
c d breed within the study area. data). Early data were collected using methods described by Monterey Bay and to the south. Surveys recorded 1,447
Briggs et al. (1983, 1987b); more recent data were collected sightings of 3,003 individuals.
124°W 123°W 122°W 121°W
124°W 123°W 122°W 121°W
Source Data: See text. using updated technology but using the same general method.
Data sources for ship-based survey data include: David Ainley
Figure 48. Brown pelican, seasonal density and high use areas.
63
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
This population breeds on selected islands off Baja Mexico
and southern California, with small colonies extending north
to the Channel Islands. Brown pelicans once bred on rocks off
Monterey, but are now concentrated in the central California
study area at roosts such as Morro Rock, Monterey Breakwater,
Año Nuevo Island, Southeast Farallon Island, Bird Rock in
Monterey county, and Bodega Rock. Nesting occurs in southern
California and Baja Mexico and begins in November and can
extend through June, when the species is most sparse in
the central California study area. The brown pelican is most
abundant in the study area during the Oceanic Season; the
species’ presence then constitutes a post-breeding increase
from southern breeding grounds.
North of Monterey and Estero/San Luis Obispo bays, this
species’ presence is much more seasonal and dependent on
ocean climate. Most sightings in the Gulf of the Farallones
during the Davidson Current and Upwelling seasons occurred
during warm-water years, often associated with the species
choosing to forego breeding at southern colonies. Thus,
wintering birds may remain in central California waters, while
others may move farther north than usual at that time. In most
cool-or coldwater years, adults are not abundant north of
Monterey Bay during these two seasons. This could change,
however, as sardines, an important prey item, continue to
increase in California waters.
The species frequents waters within several miles of shore
(mean distance to land was 10.3 ± 0.4 km) and rarely occurs
in waters deeper than the shelf break (mean depth was
266 ± 21 m). Consistent with these patterns are results of
a multiple regression model of nine independent variables,
which explained 15.2% of the variation; important variables
were season, and inverse relationships to distance to land and
latitude; see Table 19. Therefore, the broad shelf of central
California is well suited to this species; its occurrence becomes
sporadic north of Point Reyes. Inshore Monterey, Estero, and
San Luis Obispo bays are especially important, where this
species is common year round; the San Francisco Bay tidal
plume is also important. Abundance of this species in the study
area has increased between 1985 and 2002.
This species preys exclusively on fish, especially anchovies,
mackerel, and sardines, that it catches by plunging to just
below the surface. See Tables 15 and 16 for related summary
information.
64
Section 2.3: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS on survey methods); and Lisa T. Ballance, from the Ecology
Figures 49a, b, and c show the density (birds/km2) of black- Program of the Southwest Fisheries Science Center, NMFS,
Black-legged Kittiwake Rissa tridactyla legged kittiwakes in the Upwelling, Oceanic, and Davidson NOAA (unpublished data).
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Current seasons, displayed in five minute latitude by five minute
Oceanic Season
Upwelling Season
200 m
longitude cells. Densities are based on the combined data sets Although the at-sea data span the years from 1980 to 2001,
200 m
20
20
0
0
0m
0m
of several studies (see “Methods” and “Data Sources” below). data are not available for all seasons in all years. For the
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14)
39°N
39°N
The color and mapping intervals were customized to show Upwelling Season, data are from 1980-1982 and 1985-2001.
Density the most structure and to highlight significant areas, while For the Oceanic Season, data are from 1980-1982, 1991, and
(Animals/km²)
allowing comparisons among marine bird species. Cells that 1994-2001. For the Davidson Current Season, data are from
> 100.00
were surveyed but in which no black-legged kittiwakes were 1980-1986 and 1991-2001.
50.01 - 100.00
observed have a density of zero. Areas not surveyed appear
38°N
38°N
10.01 - 50.00
white; no information is available for these areas. Blue lines METHODS
5.01 - 10.00
1.01 - 5.00 indicate the boundaries of the National Marine Sanctuaries At-sea densities are the result of a synthesis of data from eight
0.51 - 1.00
in the study area: Cordell Bank, Gulf of the Farallones, and shipboard and aerial survey programs conducted in the study
0.11 - 0.50
Monterey Bay. area in the years 1980-2001 (see “Data Sources” below). Bird
0.06 - 0.10
37°N
37°N
observation data and trackline data from these studies were
0.01 - 0.05
0.00 In order to provide one map for the species that integrates the converted to a common format. All aerial data were continuous;
0 25 50 Km
patterns of its spatial and temporal occurrence and abundance ship-based data were converted separately into a continuous
in the study area, map d shows seasonal high-use areas, dis- transect to the extent possible. From the digitized survey data,
played in 10 minute latitude by 10 minute longitude cells. The the distributions of effort and of species were mapped into five
36°N
36°N
seasonal high use map provides a further synthesis of densities minute latitude by five minute longitude cells using CDAS, a
presented in Maps a, b and c, and portrays the relative impor- custom geographic information system for analyzing marine
tance of various areas to the species. Areas with consistently bird and mammal surveys (MMS, 2001). The length and width
high use are highlighted on this map. To provide a relative refer- of the survey trackline in a given cell (estimated trackline width
ence for the “high use” areas, cells are also shown where the varied by platform, depending on speed and height above wa-
35°N
35°N
a b species were absent (i.e., the cell was sampled but the species ter) were used to estimate the area sampled. The number of
was not recorded there), or present but at lesser concentrations birds of each species seen in a cell was then divided by the area
in any particular season. See the "Methods" section below for sampled in the cell to estimate density. If a cell was surveyed
Davidson Current Season Seasonal High Use Areas and
200 m
200 m
20
20
0
0
0m
0m
further explanation of seasonal high-use areas. more than once, densities were averaged, with an adjustment
Breeding Colonies
(Nov. 15 - Mar. 14)
39°N
39°N
made for effort.
Persistence of
DATA SOURCES
High Use
3 Seasons
Densities for marine birds at sea are based on data from eight The seasonal high-use areas on map d were developed using
2 Seasons
survey programs conducted between 1980 and 2001, which a similar approach as for Maps a, b and c, but the data were
1 Season
Birds present
were combined into a new MMS-CDAS data set (MMS, 2001) binned into 10’x10’ cells. For each season, the cells with densi-
38°N
38°N
Birds absent
using software (CDAS) developed for the Minerals Manage- ties in the top 20% of non-zero values were designated “high
ment Service. Of the data sets on the original MMS-CDAS CD- use” for that season. Cells were scored for “high use” in one,
ROM, four aerial survey data sets contained data in the study two, or three seasons and are depicted by color. To provide a
area from Point Arena to Point Sal. Of these, the OSPR survey relative reference for the “high use” areas, cells are also shown
37°N
37°N
program is ongoing and data from recent years were added where the species were absent (i.e., the cell was sampled but
to this data set. In addition, data from four ship-based survey the species was not recorded there) or present (but densities
programs were converted to a compatible format for analysis were never in the top 20% for any season).
(see section overview for details on individual data sets).
RESULTS AND DISCUSSION
36°N
36°N
Data sources for aerial, at-sea data include MMS-CDAS (MMS, The black-legged kittiwake, like the northern fulmar, breeds on
2001), and California Department of Fish and Game, Office of islands along the northern coast of North America and Asia, but
Spill Prevention and Response (CDF&G-OSPR, unpublished large numbers ‘winter’ in the study area off central California. It
data). Early data were collected using methods described by is a common species in the study area; surveys recorded 2,079
Briggs et al. (1983, 1987b); more recent data were collected sightings of 5,003 individuals. A multiple-regression model of
35°N
35°N
This species does not
c d using updated technology but using the same general method. eight independent variables explained 28.9% of variation in
breed within the study area.
Data sources for ship-based survey data include: David Ainley cell density; important variables were season, ENSO period
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
of H. T. Harvey and Associates and Carol Keiper of Oikonos (i.e., periods of climatic variation), and year (increasing abun-
Source Data: See text.
(unpublished data; see Oedekoven et al., 2001 for details dance). The black-legged kittiwake was most abundant in the
Figure 49. Black-legged kittiwake, seasonal density and high use areas.
65
Section 2.3: BIOGEOGRAPHY OF MARINE BIRDS
study area during the Davidson Current Season and less so
during the early Upwelling Season; it was largely absent during
the late-Upwelling Season and Oceanic Season (which cor-
responds to the breeding season at northern-latitude nesting
sites). Abundance was highest during periods of La Niña. Most
kittiwakes occurred in waters overlying the continental slope,
and deeper waters seaward of National Marine Sanctuary
boundaries (mean depth of occurrence was 1,408 m; mean
distance from shore was 29.0 km). A minority of kittiwakes oc-
curred over the shelf, mainly where the shelf is narrow. There
was an “invasion” of kittiwakes in 1999, coincident with the
beginning of the cold-water regime shift (see below).
This species feeds on fish and pelagic invertebrates that they
catch by dipping and plunging to the surface. No studies of
kittiwake diet at sea were available. See Tables 15 and 16 for
related summary information.
66
Section 2.3: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS Data sources for ship-based survey data include: David Ainley
Common Murre Figures 50a, b, and c show the density (birds/km2) of common of H. T. Harvey and Associates and Carol Keiper of Oikonos
Uria aalge murre in the Upwelling, Oceanic, and Davidson Current (unpublished data; see Oedekoven et al., 2001 for details
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
seasons, displayed in five minute latitude by five minute on survey methods); and Lisa T. Ballance, from the Ecology
Upwelling Season Oceanic Season
200 m
200 m
20
20
longitude cells. Densities are based on the combined data Program of the Southwest Fisheries Science Center, NMFS,
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) sets of several studies (see “Methods” and “Data Sources” NOAA (unpublished data). Data on common murre colonies
39°N
39°N
below). The color and mapping intervals were customized to were obtained from Carter et al. (1992), with updates for Devil’s
Density
show the most structure and to highlight significant areas, Slide (Gerry McChesney, USFWS, pers. comm) and South
(Animals/km²)
while allowing comparisons among marine bird species. Cells Farallon Island (Warzybok et al. 2002).
> 100.00
that were surveyed but in which no common murres were
50.01 - 100.00
38°N
38°N
observed have a density of zero. Areas not surveyed appear Although the at-sea data span the years from 1980 to 2001,
10.01 - 50.00
5.01 - 10.00 white; no information is available for these areas. Blue lines data are not available for all seasons in all years. For the
1.01 - 5.00
indicate the boundaries of the National Marine Sanctuaries Upwelling Season, data are from 1980-1982 and 1985-2001.
0.51 - 1.00
in the study area: Cordell Bank, Gulf of the Farallones, and For the Oceanic Season, data are from 1980-1982, 1991, and
0.11 - 0.50
Monterey Bay. 1994-2001. For the Davidson Current Season, data are from
0.06 - 0.10
37°N
37°N
0.01 - 0.05
1980-1986 and 1991-2001.
0.00
In order to provide one map for the species that integrates the
0 25 50 Km
patterns of its spatial and temporal occurrence and abundance METHODS
in the study area, map d shows seasonal high-use areas, At-sea densities are the result of a synthesis of data from eight
displayed in 10 minute latitude by 10 minute longitude cells, shipboard and aerial survey programs conducted in the study
36°N
36°N
and breeding colonies. The seasonal high use map provides a area in the years 1980-2001 (see “Data Sources” below). Bird
further synthesis of densities presented in Maps a, b and c, and observation data and trackline data from these studies were
portrays the relative importance of various areas to the species. converted to a common format. All aerial data were continuous;
Areas with consistently high use are highlighted on this map. To ship-based data were converted separately into a continuous
provide a relative reference for the “high use” areas, cells are transect to the extent possible. From the digitized survey data,
35°N
35°N
a b also shown where the species were absent (i.e., the cell was the distributions of effort and of species were mapped into five
sampled but the species was not recorded there), or present minute latitude by five minute longitude cells using CDAS, a
Seasonal High Use Areas and
200 m
Davidson Current Season
200 m
20
20
but at lesser concentrations in any particular season. See the custom geographic information system for analyzing marine
0
0
0m
0m
Breeding Colonies
(Nov. 15 - Mar. 14) "Methods" section below for further explanation of seasonal bird and mammal surveys (MMS, 2001). The length and width
39°N
39°N
high-use areas. Breeding colonies are also shown; the relative of the survey trackline in a given cell (estimated trackline width
Persistence of
High Use size of the symbols indicates the colony size. varied by platform, depending on speed and height above
3 Seasons
water) were used to estimate the area sampled. The number
2 Seasons
1 Season
DATA SOURCES of birds of each species seen in a cell was then divided by
Birds present
38°N
Densities for marine birds at sea are based on data from the area sampled in the cell to estimate density. If a cell was
38°N
Birds absent
eight survey programs conducted between 1980 and 2001, surveyed more than once, densities were averaged, with an
Colony Size
(Breeding birds)
which were combined into a new MMS-CDAS data set (MMS, adjustment made for effort.
50,001 - 104,000
2001) using software (CDAS) developed for the Minerals
10,001 - 50,000
Management Service. Of the data sets on the original MMS- The seasonal high-use areas on map d were developed using
37°N
37°N
5001 - 10,000
CDAS CD-ROM, four aerial survey data sets contained data a similar approach as for Maps a, b and c, but the data were
1001 - 5000
in the study area from Point Arena to Point Sal. Of these, the binned into 10’x10’ cells. For each season, the cells with
501 - 1000
OSPR survey program is ongoing and data from recent years densities in the top 20% of non-zero values were designated
101 - 500
50 - 100
were added to this data set. In addition, data from four ship- “high use” for that season. Cells were scored for “high use”
Historical
based survey programs were converted to a compatible format in one, two, or three seasons and are depicted by color. To
36°N
36°N
for analysis (see section overview for details on individual provide a relative reference for the “high use” areas, cells are
data sets). also shown where the species were absent (i.e., the cell was
sampled but the species was not recorded there) or present
Data sources for aerial, at-sea data include MMS-CDAS (MMS, (but densities were never in the top 20% for any season).
2001), and California Department of Fish and Game, Office of
35°N
35°N
c d Spill Prevention and Response (CDF&G-OSPR, unpublished RESULTS AND DISCUSSION
data). Early data were collected using methods described by The common murre is very abundant in the study area, being
124°W 123°W 122°W 121°W
124°W 123°W 122°W 121°W
Briggs et al. (1983, 1987b); more recent data were collected the second most numerous marine bird in Central California.
Source Data: See text.
using updated technology but using the same general method. There have been 21,893 sightings of 141,964 individuals,
Figure 50. Common murre, seasonal density and high use areas and breeding colonies.
67
Section 2.3: BIOGEOGRAPHY OF MARINE BIRDS
with the ratio between these numbers indicating that murres
usually occur in flocks. The species nests at a complex of
related and densely occupied colonies including the Farallon
Islands, Point Reyes, Double Point (including Point Resistance
and Millers Point Rocks), and a small colony at Devils Slide.
This complex constitutes one of the largest, if not the largest,
breeding population of this species south of Alaska. Two small,
disjunct breeding colonies, the southernmost for this species,
occur off the Big Sur coast.
Based on analysis of the data, common murres reside in the
study area year-round, being particularly abundant in waters
overlying the shelf and upper slope (mean depth of 110 ± 5
m), with little seasonal change in distribution. Murre densities,
however, were, in general, significantly higher during the
Upwelling Season, probably because the entire population is
present at that time. During the other seasons, some breeding
individuals disperse outside of the study area. A multiple
regression model of nine independent variables explained
52.3% of variation in density; especially through inverse
relationships with distance to colony, ocean depth, and distance
to land; see Table 19. No significant trend in common murre
abundance existed between 1985 and 2002, and abundance
was not affected by short-term climate fluctuations (e.g., periods
of unusually warm or cold sea temperatures).
Near the large Farallon Islands colony during nesting, many
murres range seaward beyond the continental slope (and
outside sanctuary boundaries), perhaps as a response to
increased intraspecific competition for prey at that time. As
a result, the Farallon Escarpment became an area of high
concentration as well as the Farallon Ridge and shelf waters
inshore of it. Murres occur in Monterey Bay after nesting and
mainly during the Oceanic Season. During years of unusually
warm waters (and depleted prey), murres occur more frequently
inshore, especially along the coast from Point Reyes south to
Año Nuevo Island, the usual area of concentration during the
relatively warm Oceanic Season.
This species is a deep diver (to 180m depth, Ainley et al,
2002) that feeds on fish and invertebrates. During winter and
early spring, major prey include herring, market squid and
euphausiids; this diet then shifts mostly to juvenile rockfish
and anchovies in mid-summer. See Tables 15 and 16 for related
summary information.
68
Section 2.3: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS of H. T. Harvey and Associates and Carol Keiper of Oikonos
Rhinoceros Auklet Figures 51a, b, and c show the density (birds/km 2) of (unpublished data; see Oedekoven et al., 2001 for details
Cerorhinca monocerata Rhinoceros Auklet in the Upwelling, Oceanic, and Davidson on survey methods); and Lisa T. Ballance, from the Ecology
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Current seasons, displayed in five minute latitude by five minute Program of the Southwest Fisheries Science Center, NMFS,
Upwelling Season Oceanic Season
200 m
200 m
20
20
longitude cells. Densities are based on the combined data NOAA (unpublished data). Data on breeding colonies in the
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) sets of several studies (see “Methods” and “Data Sources” study area were obtained from Carter et al. (1992), with most
39°N
39°N
below). The color and mapping intervals were customized recent estimates for Año Nuevo from Thayer and Sydeman
Density
to show the most structure and to highlight significant areas, (2002).
(Animals/km²)
while allowing comparisons among marine bird species. Cells
> 100.00
that were surveyed but in which no Rhinoceros Auklets were Although the at-sea data span the years from 1980 to 2001,
50.01 - 100.00
38°N
38°N
observed have a density of zero. Areas not surveyed appear data are not available for all seasons in all years. For the
10.01 - 50.00
5.01 - 10.00
white; no information is available for these areas. Blue lines Upwelling Season, data are from 1980-1982 and 1985-2001.
1.01 - 5.00
indicate the boundaries of the National Marine Sanctuaries For the Oceanic Season, data are from 1980-1982, 1991, and
0.51 - 1.00
in the study area: Cordell Bank, Gulf of the Farallones, and 1994-2001. For the Davidson Current Season, data are from
0.11 - 0.50
Monterey Bay. 1980-1986 and 1991-2001.
0.06 - 0.10
37°N
37°N
0.01 - 0.05
0.00
In order to provide one map for the species that integrates the METHODS
0 25 50 Km
patterns of its spatial and temporal occurrence and abundance At-sea densities are the result of a synthesis of data from eight
in the study area, map d shows seasonal high-use areas, shipboard and aerial survey programs conducted in the study
displayed in 10 minute latitude by 10 minute longitude cells, area in the years 1980-2001 (see “Data Sources” below). Bird
36°N
36°N
and breeding colonies. The seasonal high use map provides observation data and trackline data from these studies were
a further synthesis of densities presented in Maps a, b and c, converted to a common format. All aerial data were continuous;
and portrays the relative importance of various areas to the ship-based data were converted separately into a continuous
species. Areas with consistently high use are highlighted on this transect to the extent possible. From the digitized survey data,
map. To provide a relative reference for the “high use” areas, the distributions of effort and of species were mapped into five
35°N
35°N
a b cells are also shown where the species were absent (i.e., the minute latitude by five minute longitude cells using CDAS, a
cell was sampled but the species was not recorded there), or custom geographic information system for analyzing marine
Seasonal High Use Areas and
200 m
Davidson Current Season
200 m
20
20
present but at lesser concentrations in any particular season. bird and mammal surveys (MMS, 2001). The length and width
0
0
0m
0m
Breeding Colonies
(Nov. 15 - Mar. 14) See the "Methods" section below for further explanation of of the survey trackline in a given cell (estimated trackline width
39°N
39°N
seasonal high-use areas. Breeding colonies are also shown; varied by platform, depending on speed and height above
Persistence of
High Use the relative size of the symbols indicates the colony size. water) were used to estimate the area sampled. The number
3 Seasons
of birds of each species seen in a cell was then divided by
2 Seasons
1 Season
DATA SOURCES the area sampled in the cell to estimate density. If a cell was
Birds present
38°N
Densities for marine birds at sea are based on data from eight surveyed more than once, densities were averaged, with an
38°N
Birds absent
survey programs conducted between 1980 and 2001, which adjustment made for effort.
Colony Size
(Breeding birds)
were combined into a new MMS-CDAS data set (MMS, 2001)
10,000 - 50,000
using software (CDAS) developed for the Minerals Management The seasonal high-use areas on map d were developed using
5001 - 10,000
Service. Of the data sets on the original MMS-CDAS CD-ROM, a similar approach as for Maps a, b and c, but the data were
37°N
37°N
1001 - 5000
four aerial survey data sets contained data in the study area binned into 10’x10’ cells. For each season, the cells with
501 - 1000
from Point Arena to Point Sal. Of these, the OSPR survey densities in the top 20% of non-zero values were designated
101 - 500
program is ongoing and data from recent years were added “high use” for that season. Cells were scored for “high use”
51 - 100
2 - 50
to this data set. In addition, data from four ship-based survey in one, two, or three seasons and are depicted by color. To
Historical
programs were converted to a compatible format for analysis provide a relative reference for the “high use” areas, cells are
36°N
36°N
(see section overview for details on individual data sets). also shown where the species were absent (i.e., the cell was
sampled but the species was not recorded there) or present
Data sources for aerial, at-sea data include MMS-CDAS (MMS, (but densities were never in the top 20% for any season).
2001), and California Department of Fish and Game, Office of
Spill Prevention and Response (CDF&G-OSPR, unpublished RESULTS AND DISCUSSION
35°N
35°N
c d data). Early data were collected using methods described by In the study area, this common species nests principally at
Briggs et al. (1983, 1987b); more recent data were collected the Farallon Islands; a smaller nesting population occurs at
124°W 123°W 122°W 121°W
124°W 123°W 122°W 121°W
using updated technology but using the same general method. Año Nuevo. The Farallones constitute the southernmost large
Source Data: See text.
Data sources for ship-based survey data include: David Ainley nesting colony. At-sea surveys recorded 5,415 sightings of
Figure 51. Rhinoceros auklet, seasonal density and high use areas and breeding colonies.
69
Section 2.3: BIOGEOGRAPHY OF MARINE BIRDS
15,454 individuals. Based on the analysis of the combined data
sets described in this section, the abundance of Rhinoceros
Auklets has increased significantly since the 1970s (Ainley et
al. 1994). Based on patterns apparent in the maps, the current
at-sea population probably far exceeds the estimates of nesting
populations in central California (Michelle Hester, pers.comm.).
Therefore, if there was more nesting habitat (e.g., burrows,
holes, crevices on offshore islands), the nesting population
would probably be much larger.
Rhinoceros auklets, which mainly visit colonies at night,
occurred principally in waters overlying the slope (mean depth
of occurrence was 762 ± 22 m), particularly the shelf break,
and, including the Farallon Escarpment. A sizeable portion of
the population occurs outside of the National Marine Sanctuary
boundaries. This is especially true in the vicinity of the Gulf
of the Farallones during the Upwelling (nesting) and Oceanic
seasons, when these auklets occur farther offshore (mean
depths were 791 m and 1,370 m, respectively). This expansion
of habitat, causing a ‘halo’ of increased density around the
islands, may be a response to the large numbers nesting at the
Farallones, a pattern typical of the Western Gull and Common
Murre (see those accounts). The species’ concentration,
especially along the shelf break and upper continental slope,
is particularly evident during the Oceanic Season, when the
nesting populations are no longer associated with colonies.
A multiple-regression model of nine independent variables
explained 19.8% of variation in cell density; important variables
were a negative relationship to distance to land, and positive
ones to season and ocean depth; see Table 19. The relationship
with season reflected a dramatic increase in abundance during
the Davidson Current Season (mean density of 161 birds per
100km2) compared to the Upwelling and Oceanic seasons
(mean densities of 48 and 62 birds per 100 km2, respectively).
This increase during the Davidson Current Season was likely
due to an influx of birds from the north where much larger
populations breed, compared to those of the study area.
This species feeds by diving, probably to relatively deep depths
(100 m, Ainley and Boekelheide, 1990), capturing mostly fish
but also euphuasiids. Important prey are juvenile rockfish,
anchovy and saury. See Tables 15 and 16 for related summary
information.
70
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS the distributions of effort and of species were mapped into five
Marine Bird Density Figures 52a, b, and c show the combined density (birds/km2) minute latitude by five minute longitude cells. The length and
of 76 species of marine birds in the Upwelling, Oceanic, and width of the survey trackline in a given cell (estimated trackline
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Davidson Current seasons, displayed in five minute latitude by width varied by platform, depending on speed and height above
Upwelling Season Oceanic Season
200 m
200 m
20
20
five minute longitude cells. Map d shows density for all seasons water) were used to estimate the area sampled. The number
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) and years combined. Densities are based on combined data of marine birds seen in a cell was then divided by the area
39°N
39°N
of several studies (see “Methods” and “Data Sources” above). sampled in the cell to estimate density. If a cell was censused
Density
The color and mapping intervals were customized to show the more than once, densities were averaged, with adjustment
(Animals/km²)
most structure and highlight significant areas. Cells that were made for effort.
> 100.00
surveyed but in which no birds were observed have a density
50.01 - 100.00
38°N
38°N
of zero; unsurveyed areas are white. Blue lines indicate the RESULTS AND DISCUSSION
10.01 - 50.00
5.01 - 10.00 National Marine Sanctuary boundaries of Cordell Bank, Gulf of Overall density is dominated by two abundant marine bird
1.01 - 5.00
the Farallones, and Monterey Bay; bathymetric contours for the species: common murre and sooty shearwater.
0.51 - 1.00
200 meter and 2,000 meter isobaths are also shown in blue.
0.11 - 0.50
0.06 - 0.10
Based on visual inspection of the maps, density was highest,
37°N
37°N
0.01 - 0.05
DATA SOURCES during the Upwelling Season with cells of highest density
0.00
At-sea densities are based on data from eight survey programs most widespread as well. Except for a few highest-density
0 25 50 Km
conducted in 1980-2001, which were combined using software ‘hot spots,’(see Table 17) marine birds were distributed evenly
developed for MMS-CDAS (2001) and expanded for this at high density (>10 individuals per km2) over the shelf and
project. Of the data sets on the original CD-ROM, four aerial slope from north to south in the study area. Particular hot spots
36°N
36°N
survey data sets provided data in the study area from Point were inshore Monterey Bay, Farallon Ridge and Cordell Bank.
Arena to Point Sal. Of these, one program was still ongoing and The pattern during this season generally matched the pattern
data from recent years were added to this data set. In addition, apparent when all seasons were combined.
data from four ship-based survey programs were converted to
a compatible format for analysis. See section introduction for
35°N
35°N
During the Oceanic Season, highest density areas increased
a b details on individual data sets. in prevalence inshore. At that time, hot spots were the San
Francisco Bay tidal plume, inshore near Año Nuevo, innermost
200 m
Davidson Current Season
200 m
All Seasons
20
20
Data sources for aerial at-sea data include MMS-CDAS (MMS,
0
0
Monterey Bay and San Luis Obispo Bay.
0m
0m
(Nov. 15 - Mar. 14) 2001), and California Department of Fish and Game Office of
39°N
39°N
Spill Prevention and Response (CDF&G-OSPR, unpublished During the Davidson Current Season, birds shifted more to
data). Early data were collected using methods described by the mid-shelf.
Briggs et al. (1987b); more recent data were collected using
updated technology but the same general method. Data
38°N
38°N
sources for ship-based survey data include: David Ainley and
Carol Keiper (unpublished data; see Oedekoven et al., 2001
for details on survey methods).
Although the at-sea data span the years 1980 to 2001, data
37°N
37°N
are not available for all seasons in all years. For the Upwelling
Season, data are from 1980-1982 and 1985-2001. For the
Oceanic Season, data are from 1980-1982, 1991, and 1994-
2001. For the Davidson Current Season, data are from 1980-
1986 and 1991-2001.
36°N
36°N
METHODS
At-sea densities are the result of a synthesis of data from
eight shipboard and aerial survey programs conducted in
the study area in the years 1980-2001 (see “Data Sources”
35°N
35°N
c d above). Observation and trackline data from these studies were
converted to a common format. All aerial data were continuous;
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Source Data: See text. ship-based data were converted separately into a continuous
transect to the extent possible. From the digitized survey data,
Figure 52. Marine bird density, by season and for all seasons.
71
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS above). Observation and trackline data from these studies were
Figure 53a, b, and c shows total marine bird biomass (kg/ converted to a common format. All aerial data were continuous;
Marine Bird Biomass km2) in each five minute latitude by five minute longitude cell ship-based data were converted separately into a continuous
124°W 123°W 122°W 121°W
for each oceanographic season and for all seasons combined
124°W 123°W 122°W 121°W
transect to the extent possible. From the digitized survey data,
Upwelling Season Oceanic Season 53d. Density for each of 76 species was multiplied by average
200 m
the distributions of effort and of species were mapped into five
200 m
20
20
0
0
0m
0m
body mass for that species. These products were summed minute latitude by five minute longitude cells. The length and
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14)
39°N
39°N
for all species in a cell. The color and mapping intervals were width of the survey trackline in a given cell (estimated trackline
Biomass customized to show the most structure and highlight significant width varied by platform, depending on speed and height above
(kg/km²)
areas. Cells that were surveyed but in which no birds were water) were used to estimate the area sampled. The number
> 100
observed have a biomass density of zero; unsurveyed areas of marine birds seen in a cell was then divided by the area
51 - 100
are white. Blue lines indicate the National Marine Sanctuary sampled in the cell to estimate density. If a cell was censused
38°N
38°N
21 - 50
boundaries of Cordell Bank, Gulf of the Farallones, and more than once, densities were averaged, with adjustment
11 - 20
6 - 10 Monterey Bay; bathymetric contours for the 200 meter and made for effort.
4-5
2,000 meter isobaths are also shown in blue.
3
Once the weighted densities had been determined for each
2
37°N
37°N
DATA SOURCES
1 species in each cell, densities of each species were multiplied
0
At-sea biomass densities are based on data from eight survey by the average body mass of that species. These ‘biomass
programs conducted in 1980-2001, which were combined using
0 25 50 Km
densities’ were then summed for each cell and the results
software developed for MMS-CDAS (2001) and expanded for plotted.
this project. Of the data sets on the original CD-ROM, four aerial
36°N
36°N
survey data sets provided data in the study area from Point RESULTS AND DISCUSSION
Arena to Point Sal. Of these, one program was still ongoing and In general, the biomass maps are dominated by two, rela-
data from recent years were added to this data set. In addition, tively heavy-bodied, numerically dominant species: com-
data from four ship-based survey programs were converted to mon murre and sooty shearwater. These maps are also
a compatible format for analysis. See section introduction for influenced, to a lesser degree, by the species identified as
35°N
35°N
a b details on individual data sets. abundant in the study area (see Table 15).
Data sources for aerial at-sea data include MMS-CDAS (MMS,
200 m
Davidson Current Season
200 m
All Seasons
20
20
Looking first at a summary of all seasons, high biomass densi-
0
0
0m
0m
2001) and California Department of Fish and Game Office of
(Nov. 15 - Mar. 14) ties occurred in the Gulf of the Farallones, especially around
39°N
39°N
Spill Prevention and Response (CDF&G-OSPR), unpublished the Farallon Islands, the San Francisco Bay tidal plume, off
data. Early data were collected using methods described by Half-moon Bay, just south of Point Año Nuevo and in inner
Briggs et al. (1987b); more recent data were collected using Monterey Bay.
updated technology but the same general method. Data
sources for ship-based survey data include David Ainley and During the Upwelling season, high biomass densities occurred
38°N
38°N
Carol Keiper (unpublished data; see Oedekoven et al., 2001 over the shelf and upper slope with highest density areas oc-
for details on survey methods). Although the at-sea data span curring at Monterey Bay, Farallon Ridge, and Cordell Bank.
the years 1980-2001, data are not available for all seasons in The distribution of high biomass during the Upwelling Season
all years. For the Upwelling Season, data are from 1980-1982 mimicked that described in the all seasons map (map d).
and 1985-2001. For the Oceanic Season, data are from 1980-
37°N
37°N
1982, 1991 and 1994-2001. For the Davidson Current Season, During the Oceanic Season high biomass was concentrated
data are from 1980-1986 and 1991-2001. more over the inner shelf than in the Upwelling Season, par-
ticularly evident from Point Reyes to Monterey, as well as San
Data on average biomass for each species were derived Luis Obispo Bay.
36°N
36°N
from Body Weights of 686 Species of North American Birds
(Dunning 1993). In a few instances, a species was not listed in During the Davidson Current Season (DCS), virtually the entire
this reference; in these cases, the biomass of a closely related continental shelf from Point Reyes to Point Sur exhibited high
bird of a similar size was used. marine bird biomass.
35°N
35°N
c d METHODS
At-sea densities are the result of a synthesis of data from
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
eight shipboard and aerial survey programs conducted in
Source Data: See text.
the study area in the years 1980-2001 (see “Data Sources”
Figure 53. Marine bird biomass, by season and for all seasons.
72
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Upwelling Season, data are from 1980-1982 and 1985-2001.
ABOUT THESE MAPS
For the Oceanic Season, data are from 1980-1982, 1991 and
Species diversity was calculated for each five minute latitude
Marine Bird Species Diversity 1994-2001. For the Davidson Current Season, data are from
by five minute longitude cell using density as the variable in the
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
1980-1986 and 1991-2001.
Shannon [Diversity] Index (Shannon and Weaver 1949). This
Upwelling Season Oceanic Season
200 m
200 m
20
20
index measures the degree to which a species assemblage is
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) METHODS
dominated by a few species. If a cell contains high densities
39°N
39°N
At-sea densities are the result of a synthesis of data from
of a few species and low densities of all others, the value of
Diversity eight shipboard and aerial survey programs conducted in
diversity (H’) will be low, indicating low diversity. Alternatively,
(H')
the study area in the years 1980-2001 (see “Data Sources”
if many species are present at similar densities, the value will
2.001 - 3.000
above). Observation and trackline data from these studies were
be high, indicating high diversity. Figures 54a, b, and c show
1.801 - 2.000
converted to a common format. All aerial data were continuous;
the diversity index H’ in three oceanographic seasons; map d
38°N
38°N
1.501 - 1.800
ship-based data were converted separately into a continuous
shows diversity for all seasons and years combined. Although
0.801 - 1.500
0.000 - 0.800 transect to the extent possible. From the digitized survey data,
a density-based calculation of the Shannon Index is less
the distributions of effort and of species were mapped into five
influenced by differences in effort as compared with the index
0 25 50 Km
minute latitude by five minute longitude cells. The length and
calculated using species counts, a significant correlation (p<
37°N
37°N
width of the survey trackline in a given cell (estimated trackline
0.001) remained between diversity and effort.
width varied by platform, depending on speed and height above
water) were used to estimate the area sampled. The number
To standardize for variable effort among cells and variable
of marine birds seen in a cell was then divided by the area
strip width among species, density was used for each species
sampled in the cell to estimate density. If a cell was censused
in each cell as the basis for calculating the diversity index
36°N
36°N
more than once, densities were averaged, with adjustment
value. All 76 marine bird species that had been recorded in
made for effort.
the data set were included. Cells are colored based on the
value of H’ computed for a particular season. Red indicates
The Shannon Index (Shannon and Weaver 1949) was used to
high diversity, blue indicates low diversity. Unsurveyed areas
quantify species diversity. For each cell, diversity was calculated
are colored white. Blue lines indicate the National Marine
35°N
35°N
a b using the formula
Sanctuary boundaries of Cordell Bank, Gulf of the Farallones,
and Monterey Bay; bathymetric contours for the 200 meter and n n
S
H ′ = − ∑ i ln i
Davidson Current Season
200 m
200 m
All Seasons
20
20
2,000 meter isobaths are also shown in blue.
0
0
0m
0m
i =1 n n
(Nov. 15 - Mar. 14)
39°N
39°N
DATA SOURCES
where ni is the density of species in that cell. Density was
At-sea densities are based on data from eight survey programs
used for calculating the index value in order to compensate for
conducted in 1980-2001, which were combined using CDAS
variable effort among cells and species. We looked at three
software into an MMS-CDAS data set (MMS, 2001) developed
38°N
38°N
oceanographic seasons and at all seasons combined.
for Minerals Management Service and expanded for this project.
Of the data sets on the original CD-ROM, four aerial survey
The Shannon Index was selected as the diversity metric
data sets provided data in the study area from Point Arena to
because it is widely used and accepted in community ecology.
Point Sal. Of these, one program was still ongoing and data
It has three desirable properties for a diversity index, noted
from recent years were added to this data set. In addition,
37°N
37°N
below. Most diversity indices do not take these three qualities
data from four ship-based survey programs were converted to
into account. For more information on diversity indices, see
a compatible format for analysis. See section introduction for
Ecological Diversity, E.C. Pielou, pp 7-18.
details on individual data sets.
1. The diversity index is greatest when all species in the
Data sources for aerial at-sea data include MMS-CDAS (2001)
36°N
36°N
community are equally represented in numbers (e.g., evenness
and California Department of Fish and Game Office of Spill
in a community). Or, for a given number of species (e.g.,
Prevention and Response (CDF&G-OSPR), unpublished data.
richness value), the diversity index should have it’s greatest
Early data were collected using methods described by Briggs
value when the proportion of each species is the same.
et al. (1987b); more recent data were collected using updated
technology but the same general method. Data sources for
35°N
35°N
c d 2. Given two completely diverse or similiar communities, the
ship-based survey data include David Ainley and Caol Keiper
one with the higher number of species has a greater diversity
(unpublished data; see Oedekoven et al., 2001 for details on
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Source Data: See text. value.
methods). Although the at-sea data span the years 1980-2001,
data are not available for all seasons in all years. For the
Figure 54. Marine bird diversity, by season and for all seasons.
73
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
3. The last property is difficult to summarize but is something
like this: This property takes into account the hierarchical
nature, or representativeness in the biological classification of
each species, when estimating diversity.
RESULTS AND DISCUSSION
Looking first at a summary of all seasons, the marine avifauna
was most diverse in areas largely outside of National Marine
Sanctuary boundaries, especially in areas of the continental
slope and particularly the Farallon Escarpment. Localized areas
of high diversity occurring within sanctuary boundaries include:
Pioneer, Ascension/Cabrillo, and Carmel canyons, as well as
the continental slope off Point Sur.
During the Upwelling Season, the avifauna was the least
diverse; areas of highest diversity in this season included the
Farallon Escarpment, and Pioneer, Ascension, and Carmel
canyons.
During the Oceanic Season, diversity was comparable to that
of the Upwelling Season in general. Areas of high diversity
continued to include the Farallon Escarpment area, Pioneer
Canyon, and inner Monterey Bay Canyon.
During the Davidson Current Season, marine bird diversity, in
general, was the highest of the year. Areas of high diversity
were all localized, and most occurred over the continental slope
(e.g., Farallon Escarpment, and Pioneer, Ascension, Monterey
Bay and Carmel canyons) but some also occurred over the
shelf (e.g., the inner San Francisco Bay tidal plume and inner
portions of Monterey Bay).
However, because of the significant correlation between diver-
sity and effort, some of the diversity patterns may be influenced
by differences in effort across the study area. See the additional
analysis and discussion of diversity in the Integration section.
74
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THIS MAP
124°W 123°W 122°W 121°W
The 40 largest marine bird breeding colonies in the study
200 m Major Marine Bird Breeding Colonies
20
39°N
39°N
0
area were chosen for this map (Figure 55). The total number
0m
of breeding marine bird species is indicated by the size of
Fish Rocks
the circle, and the number of species using a particular
Gualala Point Island
colony is indicated by the circle color. The large symbol at
the San Francisco Bay entrance represents a summary of
Russian River Rocks all the colonies in San Francisco Bay. See Table 18 for more
Colony Populations
Arched Rock
information on these colonies.
Bodega Rock Number of Species
9 - 12
DATA SOURCES
7-8 Data on marine bird colonies were derived primarily from
Point Reyes Point Resistance
38°N
38°N
6 Breeding Populations of Seabirds in California, 1989-1991
Millers Point Rocks
(Carter et al., 1992, unpublished data). Colony data were
5
San Francisco Bay
updated where more current information was available.
4
and Alcatraz Island*
North Farallon Islands
Updated information is presented for some species on South
3
South Farallon Island Farallon Island (Sydeman et al., 1998, Warzybok et al., 2002),
1-2
Devil's Slide Rock
Año Nuevo Island (Thayer and Sydeman 2002 a, b), Bird Rock,
Number of Birds Point Reyes, and Double Point Rocks (Whitworth et al., 2002),
Big Basin State Park and vicinity (Laird Henkel, pers. comm)
50,001 - 153,000
and Devil’s Slide Rock (Gerry McChesney, USFWS, pers.
Big Basin State Park and vicinity
10,001 - 50,000
Vicinity of Año Nuevo
comm).
Island and Point El Jarro Point to Davenport
5001 - 10,000
37°N
37°N
Davenport to Sand Hill Bluff
1001 - 5000
METHODS
501 - 1000
Colony locations were plotted using latitude and longitude
250 - 500
coordinates from Breeding Populations of Seabirds in California,
1989-1991 (Carter et al., 1992, unpublished data).
Bird Rock 0 25 50 Km
Bird Island
RESULTS AND DISCUSSION
Castle Rocks and Mainland
The study area is in a geologic subduction zone of the eastern
Pacific and adjacent continental margin. Therefore, as with
Anderson Canyon Rocks
analogous regions elsewhere on the globe (e.g., west coasts
of South America and Africa), islands are not common. In
36°N
36°N
Plaskett Rock
somewhat of a departure from this pattern, the Gulf of the
Cape San Martin
Farallones contains far more coastal rocks and offshore islands
La Cruz Rock than anywhere else in the study area and, in fact, this is the
Piedras Blancas Island
case for 400 miles to the north and south. Obvious in this
map is the importance to breeding marine birds of the Gulf of
the Farallones, defined as the broad shelf from Point Reyes/
Fairbank Point Tomales Point to Año Nuevo and out to the Farallon Islands. A
disproportionate number of breeding colonies and, certainly,
Pup Rock and Pecho Rock
Adjacent Mainland
most of various species’ regional breeding populations, occur
here. These colonies are large and diverse owing to the high
35°N
35°N
productivity of surrounding waters and the complexity of
habitats in the region. See Table 18 for a numerical summary
of each colony’s contribution to the breeding marine avifauna
of the study area, composed of 16 species, 12 of which breed
within the Gulf of the Farallones.
* Summary of all San Francisco Bay colonies.
124°W 123°W 122°W 121°W
Source Data: See text.
Figure 55. Major marine bird breeding colonies.
75
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
METHODS
ABOUT THESE MAPS
Density in Warm, Cold and Neutral Periods: 1980-2001 At-sea densities are the result of a synthesis of data from
A comparison of the abundance and distribution of 76 marine
eight shipboard and aerial survey programs conducted in
birds during warm-water periods (including El Niño), cold-water
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
the study area in the years 1980-2001 (see “Data Sources”
periods (La Niña) and normal (neutral) periods is provided here
Warm-water Conditions Neutral Conditions
200 m
200 m
20
20
above). Observation and trackline data from these studies were
as an example of how marine birds may respond to short-term
0
0
0m
0m
converted to a common format. All aerial data were continuous;
variation in marine climate. In this synthesis, what is shown is
39°N
39°N
ship-based data were converted separately into a continuous
density, which treats all species equally regardless of body size.
Density
transect to the extent possible. From the digitized survey data,
Therefore, the patterns demonstrated by tiny, more abundant
(Animals/km²)
the distributions of effort and of species were mapped into five
species, such as storm-petrels and phalaropes, are more
> 100.00
minute latitude by five minute longitude cells. The length and
greatly expressed. For a description of how these periods were
50.01 - 100.00
38°N
38°N
width of the survey trackline in a given cell (estimated trackline
chosen, see the following topic in the bird section: "Response
10.01 - 50.00
5.01 - 10.00 width varied by platform, depending on speed and height above
to Variation in Marine Climate" (pages 49-50).
1.01 - 5.00
water) were used to estimate the area sampled. The number
0.51 - 1.00
of marine birds seen in a cell was then divided by the area
Figures 56a, b and c show the combined density (birds/km2)
0.11 - 0.50
sampled in the cell to estimate density. If a cell was censused
of 76 species of marine birds in cold-water, neutral, and
0.06 - 0.10
37°N
37°N
0.01 - 0.05 more than once, densities were averaged, with adjustment
warm-water periods, displayed in five minute latitude by
0.00
made for effort.
five minute longitude cells. Map d shows overall patterns of
0 25 50 Km
density. Densities are based on combined data of several
Marine bird density data was then organized into periods where
studies (see “Methods” and “Data Sources” below). The
surface ocean conditions were warm (including El Niños),
color and mapping intervals were customized to show the
36°N
36°N
cold (including La Niñas) or neither (neutral). The density of
most structure and highlight significant areas. Cells that were
all species seen within respective cells was summed for that
surveyed but in which no birds were observed have a density
cell.
of zero; unsurveyed areas are white. Blue lines indicate the
National Marine Sanctuary boundaries of Cordell Bank, Gulf of
To illustrate these temperature conditions, a comparison of
the Farallones, and Monterey Bay; bathymetric contours for the
35°N
35°N
a b marine bird densities was made by making maps that use
200 meter and 2,000 meter isobaths are also shown in blue.
selected season/year periods that represented these cold,
200 m
Cold-water Conditions
200 m
20
Overall Patterns
20
warm and neutral periods. The data for each "condition" map
DATA SOURCES
0
0
0m
0m
was grouped as shown below; these groupings were based
At-sea densities are based on data from eight survey programs
39°N
39°N
on the assignments made in Table 14. Once the selection of
conducted in 1980-2001; these data sets were combined using
data were made for each analysis period (i.e., warm, neutral or
CDAS software into an MMS-CDAS data set (MMS, 2001) and
cold), the density of all birds seen within each cell was summed
expanded for this project. Of the data sets on the original CD-
for that cell.
ROM, four aerial survey data sets provided data in the study
38°N
38°N
area from Point Arena to Point Sal. Of these, one program was
For the warm-water conditions (including El Niños) map, the
still ongoing and data from recent years were added to this data
following seasons and years were used: Davidson Current
set. In addition, data from four ship-based survey programs
Season: 1981, 1983, 1984, 1992, 1993, 1994, 1996, 1998.
were converted to a compatible format for analysis. See section
Upwelling Season: 1985, 1987, 1992, 1993, 1995, 1998.
introduction for details on individual data sets.
37°N
37°N
Oceanic Season: 1983, 1997.
Data sources for aerial at-sea data include MMS-CDAS (MMS,
For the neutral conditions map, the following seasons and years
2001) and California Department of Fish and Game Office of
were used: Davidson Current Season: 1982, 1986, 1995, 1997.
Spill Prevention and Response (CDF&G-OSPR), unpublished
Upwelling Season: 1980, 1982, 1986, 1988, 1989, 1994, 1996,
data. Early data were collected using methods described by
36°N
36°N
1997. Oceanic Season: 1982, 1991, 1995.
Briggs et al. (1983); more recent data were collected using
updated technology but the same general method. Data
For the cold-water conditions (including La Niñas) map, the
sources for ship-based survey data include David Ainley and
following seasons and years were used: Davidson Current
Carol Keiper (unpublished data; see Oedekoven et al., 2001
Season: 1985, 1991, 1999, 2000, 2001, 2002. Upwelling
for details on methods). Although the at-sea data span the
35°N
35°N
c d Season: 1981, 1990, 1991, 1999, 2000, 2001. Oceanic Season:
years 1980-2001, data are not available for all seasons in all
1980, 1981, 1994, 1996, 1998, 1999, 2000, 2001.
years.
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Source Data: See text.
Figure 56. Density in warm, cold, and neutral periods: 1980-2001.
76
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
RESULTS AND DISCUSSION
There was not a great deal of difference in density apparent in
the exhibited patterns for the different periods. Nevertheless,
during warm-water conditions (e.g., El Niño events) marine
bird populations appear to contract more into the area defined
by the boundaries of the central California National Marine
Sanctuaries, from Tomales Point south to Monterey. Generally,
this area contains most of the shelf habitat of the study area,
which tends to have a greater complexity of microhabitats than
deeper waters. The reason there was not much of an apparent
pattern or major difference seen in these maps, is that individual
species respond differently to the three different temperature
conditions shown. For instance, some may move out of an
area but others may move in, and therefore, when species are
combined, these individual responses are homogenized.
During both warm and cold excursion from ‘normal’/neutral
marine climate, populations seemed to be slightly more
widespread, with major concentrations in Monterey Bay.
During cold-water conditions (e.g., La Niña events), densities
appeared to be the highest, especially in waters close to the
coast (e.g., see contiguous high-density, red and orange cells
along the coast). During warm-water events, the concentrations
are further offshore in the mid to outer shelf and there are fewer
highest density (red) cells.
As noted earlier, overall marine bird density in this analysis
is generally dominated by two numerically dominant spe-
cies, common murre and sooty shearwater, and to a lesser
degree, by the species identified as abundant in the study
area (see Table 15).
77
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS reference; in these cases, the mass of a closely related bird
Biomass in Warm, Cold and Neutral Periods: 1980-2001 A comparison of the abundance and distribution of 76 marine of a similar size was used.
birds during warm-water (El Niño) compared to cold-water (La
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Niña) and normal (neutral) periods provides an example of how METHODS
Warm-water Periods Neutral Periods
200 m
200 m
20
20
marine birds respond to short-term excursions from the usual At-sea densities are the result of a synthesis of data from
0
0
0m
0
m
marine climate. In this comparison, densities were converted to eight shipboard and aerial survey programs conducted in
39°N
39°N
biomass by multiplying density by body mass of each species. the study area in the years 1980-2001 (see “Data Sources”
Biomass
This comparison, thus, emphasizes more the larger-bodied above). Observation and trackline data from these studies were
(kg/km²)
species, such as Sooty Shearwater and Common Murre. converted to a common format. All aerial data were continuous;
> 100
ship-based data were converted separately into a continuous
51 - 100
21 - 50
38°N
38°N
Figures 57a, b, and c show the combined biomass density transect to the extent possible. From the digitized survey data,
11 - 20
(kg/km2) of 76 species of marine birds in cold-water, neutral, the distributions of effort and of species were mapped into five
6 - 10
and warm-water periods, displayed in five minute latitude by minute latitude by five minute longitude cells. The length and
4-5
five minute longitude cells. Map d shows overall patterns of width of the survey trackline in a given cell (estimated trackline
3
2 biomass density. Densities are based on combined data of width varied by platform, depending on speed and height above
37°N
37°N
1
several studies (see “Methods” and “Data Sources” below). water) were used to estimate the area sampled. The number
0
The color and mapping intervals were customized to show the of marine birds seen in a cell was then divided by the area
0 25 50 Km
most structure and highlight significant areas. Cells that were sampled in the cell to estimate density. If a cell was censused
surveyed but in which no birds were observed have a density more than once, densities were averaged, with adjustment
of zero; unsurveyed areas are white. Blue lines indicate the made for effort.
36°N
36°N
National Marine Sanctuary boundaries of Cordell Bank, Gulf of
the Farallones, and Monterey Bay; bathymetric contours for the For each species that occurred in a cell, the average density
200 meter and 2,000 meter isobaths are also shown in blue. was then multiplied by a species’ body mass (from Dunning,
1993). This resulted in an estimate of biomass for that species.
DATA SOURCES The biomass of all species in each cell was summed to give
35°N
35°N
a b At-sea densities are based on data from eight survey programs the cell biomass.
conducted in 1980-2001. These data were combined using
200 m
Cold-water Periods
200 m
Overall Patterns
20
20
CDAS software into an MMS-CDAS data system (MMS, 2001) Marine bird density data was then organized into periods where
0
0
0m
0m
for the Minerals Management Service and expanded for this surface ocean conditions were warm (including El Niños),
39°N
39°N
project. Of the data sets on the original MMS-CDAS CD-ROM, cold (including La Niñas) or neither (neutral). The density of
four aerial survey data sets contained data in the study area all species seen within respective cells was summed for that
from Point Arena to Point Sal. Of these, one program was still cell.
ongoing and data from recent years were added to this data
set. In addition, data from four ship-based survey programs To illustrate these temperature conditions, a comparison of
38°N
38°N
were converted to a compatible format for analysis. See section marine bird densities was made by making maps that use
introduction for details on individual data sets. selected season/year periods that represented these cold,
warm and neutral periods. The data for each condition map
Data sources for aerial at-sea data include MMS-CDAS (MMS, was grouped as shown below; these groupings were based
37°N
37°N
2001) and California Department of Fish and Game Office of on the assignments made in Table 14. Once the selection of
Spill Prevention and Response (CDF&G-OSPR), unpublished data were made for each analysis period (i.e., warm, neutral or
data. Early data were collected using methods described by cold), the density of all birds seen within each cell was summed
Briggs et al. (1987b); more recent data were collected using for that cell.
updated technology but the same general method. Data
36°N
36°N
sources for ship-based survey data include David Ainley and For the warm-water conditions (including El Niños) map, the
Carol Kieper (unpublished data; see Oedekoven et al., 2001 following seasons and years were used: Davidson Current
for details on methods). Although the at-sea data span the Season: 1981, 1983, 1984, 1992, 1993, 1994, 1996, 1998.
years 1980-2001, data are not available for all seasons in all Upwelling Season: 1985, 1987, 1992, 1993, 1995, 1998.
years. Oceanic Season: 1983, 1997.
35°N
35°N
c d
Data on average mass for each species were derived from For the neutral conditions map, the following seasons and years
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Body Weights of 686 Species of North American Birds (Dunning were used: Davidson Current Season: 1982, 1986, 1995, 1997.
Source Data: See text.
1993). In a few instances, a species was not listed in this
Figure 57. Biomass in warm, cold and neutral periods: 1980-2001.
78
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Upwelling Season: 1980, 1982, 1986, 1988, 1989, 1994, 1996,
1997. Oceanic Season: 1982, 1991, 1995.
For the cold-water conditions (including La Niñas) map, the
following seasons and years were used: Davidson Current
Season: 1985, 1991, 1999, 2000, 2001, 2002. Upwelling
Season: 1981, 1990, 1991, 1999, 2000, 2001. Oceanic Season:
1980, 1981, 1994, 1996, 1998, 1999, 2000, 2001.
RESULTS AND DISCUSSION
There was slightly more difference in biomass than was observed
for the analogous comparison of density. Biomass was generally
more concentrated during warm and cold conditions than during
neutral conditions, especially cold-water periods, which were
mimicked by the overall all-conditions summary. Many inner
shelf habitat areas exhibited high marine bird biomass during
cold-water periods. The Farallon Ridge and Monterey Bay had
relatively high biomass under all conditions.
As noted earlier, marine bird biomass in this analysis is gen-
erally dominated by two, relatively heavy-bodied, numerically
dominant species: common murre and sooty shearwater, and
to a lesser degree, by the species identified as abundant in
the study area (see Table 15).
79
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS details on methods). Although the at-sea data span the years
A comparison of the abundance and distribution of marine birds
Diversity in Warm, Cold and Neutral Periods: 1980-2001 1980-2001, data are not available for all seasons in all years.
during warm-water periods (e.g., El Niño events), cold-water
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
periods (e.g., La Niña events) and normal (neutral) periods METHODS
Neutral Conditions
Warm-water Conditions
200 m
200 m
20
20
provides an example of how marine birds may respond to At-sea densities are the result of a synthesis of data from
0
0
0m
0m
short-term excursions from the usual marine climate. These eight shipboard and aerial survey programs conducted in
39°N
39°N
maps (Figure 58) show species diversity, calculated for each the study area in the years 1980-2001 (see “Data Sources”
Diversity five minute latitude by five minute longitude cell using density above). Observation and trackline data from these studies were
(H')
as the variable in the Shannon [Diversity] Index (Shannon converted to a common format. All aerial data were continuous;
2.001 - 3.000
and Weaver 1949); all 76 marine bird species that had been ship-based data were converted separately into a continuous
1.801 - 2.000
recorded in the data set were included. transect to the extent possible. From the digitized survey data,
38°N
38°N
1.501 - 1.800
the distributions of effort and of species were mapped into five
0.801 - 1.500
0.000 - 0.800 minute latitude by five minute longitude cells. The length and
The Shannon Index measures the degree to which a species
width of the survey trackline in a given cell (estimated trackline
assemblage is dominated by a few species. If a cell contains
width varied by platform, depending on speed and height above
0 25 50 Km
high densities of a few species and low densities of all others,
37°N
37°N
water) were used to estimate the area sampled. The number
the value of H’ will be low, indicating low diversity. Alternatively,
of marine birds seen in a cell was then divided by the area
if many species are present at similar densities, the value will
sampled in the cell to estimate density. If a cell was censused
be high, indicating high diversity. Maps a, b and c show the
more than once, densities were averaged, with adjustment
diversity index H’ in cold-water, neutral, and warm-water
made for effort.
periods; map d shows overall patterns. Cells are colored
36°N
36°N
based on the value of H’ computed for a particular season. Red
The Shannon Index (Shannon and Weaver 1949) was used to
indicates high diversity, blue indicates low diversity. Although
quantify species diversity.
there was a significant correlation between diversity and effort,
the observed patterns of bird diversity are robust and were n n
S
H ′ = − ∑ i ln i
largely unchanged by methods designed to correct for effort.
35°N
35°N
i =1 n n
a b
This index measures the degree to which the species
Unsurveyed areas are white. Blue lines indicate the National
Cold-water Conditions
200 m
200 m
Overall Patterns
20
assemblage is dominated by a single species. If species A
20
Marine Sanctuary boundaries of Cordell Bank, Gulf of the
0
0
0m
0m
dominates all the species seen within a cell, then diversity is
Farallones, and Monterey Bay; bathymetric contours for the
39°N
39°N
low; and vice versa. To standardize for variable effort among
200 meter and 2,000 meter isobaths are also shown in blue.
cells and variable strip width among species, we used the
density for each species in each cell as the basis for calculating
DATA SOURCES
the index value.
At-sea densities are based on data from eight survey programs
38°N
38°N
conducted in 1980-2001, which were combined using software
Marine bird density data was then organized into periods where
developed for MMS-CDAS (MMS, 2001) and expanded for this
surface ocean conditions were warm (including El Niños), cold
project. Of the data sets on the original MMS-CDAS CD-ROM,
(Including La Niñas) or neither (neutral). The diversity of all
four aerial survey data sets provided data in the study area
species seen within respective cells was determined for that
from Point Arena to Point Sal. Of these, one program was still
37°N
37°N
cell.
ongoing and data from recent years were added to this data
set. In addition, data from four ship-based survey programs
were converted to a compatible format for analysis. See section To illustrate these temperature conditions, a comparison of
introduction for details on individual data sets. marine bird densities was made by making maps that use
selected season/year periods that represented these cold,
36°N
36°N
Data sources for aerial at-sea data include MMS-CDAS (MMS, warm and neutral periods. The data for each condition map
2001) and California Department of Fish and Game Office of was grouped as shown below; these groupings were based
Spill Prevention and Response (CDF&G-OSPR), unpublished on the assignments made in Table 14. Once the selection of
data. Early data were collected using methods described by data were made for each analysis period (i.e., warm, neutral or
Briggs et al. (1983); more recent data were collected using cold), the density of all birds seen within each cell was summed
35°N
35°N
c d updated technology but the same general method. Data for that cell.
sources for ship-based survey data include David Ainley and
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Source Data: See text. Carol Keiper (unpublished data; see Oedekoven et al., 2001 for
Figure 58. Diversity in warm, cold and neutral periods: 1980-2001.
80
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
For the warm-water conditions (Including El Niños) map, the
following seasons and years were used: Davidson Current
Season: 1981, 1983, 1984, 1992, 1993, 1994, 1996, 1998.
Upwelling Season: 1985, 1987, 1992, 1993, 1995, 1998.
Oceanic Season: 1983, 1997.
For the neutral conditions map, the following seasons and years
were used: Davidson Current Season: 1982, 1986, 1995, 1997.
Upwelling Season: 1980, 1982, 1986, 1988, 1989, 1994, 1996,
1997. Oceanic Season: 1982, 1991, 1995.
For the cold-water conditions (Including La Niñas) map, the
following seasons and years were used: Davidson Current
Season: 1985, 1991, 1999, 2000, 2001, 2002. Upwelling
Season: 1981, 1990, 1991, 1999, 2000, 2001. Oceanic Season:
1980, 1981, 1994, 1996, 1998, 1999, 2000, 2001.
RESULTS AND DISCUSSION
Under all variations of climate, marine bird diversity was highest
over the continental slope, with the Farallon Escarpment
and Pioneer Canyon, in particular, standing out. Of lesser
importance was outer Monterey Bay Canyon and Point Sur
slope. Areas of high diversity were more spread out along
the slope when ocean temperatures were warm. Adding to
the latter hot spots was the area around Ascension Canyon.
During neutral conditions, diversity everywhere was relatively
low, when compared with higher diversities during the warm-
water and cold-water periods.
Although there was a significant correlation between diversity
and effort, the observed patterns of bird diversity are robust
and were largely unchanged by methods designed to correct
for effort.
81
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Observation and trackline data from these studies were con-
ABOUT THESE MAPS
Density During El Niño and La Niña Events, 1997-2000 verted to a common format. All aerial data were continuous;
A comparison of the density and distribution for two species
ship-based data were converted separately into a continuous
during an intense El Niño (1997-98) and an adjacent and in-
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
transect to the extent possible. From the digitized survey data,
tense La Niña (1999-00) provides an example of how marine
200 m
Brown Pelican Brown Pelican
200 m
20
20
the distributions of effort and of species were mapped into five
birds respond to short-term anomalies of marine climate (Figure
0
0
0m
0m
La Niña ('99 - '00)
El Niño ('97 - '98) minute latitude by five minute longitude cells using CDAS, a
59). In this comparison, the responses of individual species
39°N
39°N
custom geographic information system for analyzing marine
do not cancel out the effects of another, as was the case in
Density bird and mammal surveys (MMS, 2001). The length and width
comparisons when measures of overall abundance were used
(Animals/km²)
of the survey trackline in a given cell (estimated trackline width
(Figures 52, 53, 56 and 57).
> 100.00
varied by platform, depending on speed and height above
50.01 - 100.00
10.01 - 50.00
38°N
38°N
water) were used to estimate the area sampled. The number
Densities are based on combined data of several studies (see
5.01 - 10.00
of marine birds seen in a cell was then divided by the area
“Methods” and “Data Sources” below). The color and mapping
1.01 - 5.00
sampled in the cell to estimate density. If a cell was censused
intervals were customized to show the most structure and high-
0.51 - 1.00
more than once, densities were averaged, with adjustment
light significant areas. Cells that were surveyed but in which no
0.11 - 0.50
0.06 - 0.10 made for effort.
birds were observed have a density of zero; unsurveyed areas
0.01 - 0.05
37°N
37°N
are white. Blue lines indicate the National Marine Sanctuary
0.00
The most intense events were selected for this comparison,
boundaries of Cordell Bank, Gulf of the Farallones, and Mon-
0 25 50 Km
as well as events that occurred very close in time. In that way,
terey Bay; bathymetric contours for the 200 meter and 2,000
long-term changes in populations were not involved in the spe-
meter isobaths are also shown in blue.
cies’ occurrence patterns. For El Niño, data were used from
36°N
36°N
the Oceanic Season 1997 through Upwelling Season 1998; for
DATA SOURCES
La Niña the data were from the Oceanic Season 1998 through
At-sea densities are based on data from eight survey programs
Oceanic Season 1999; see Table 14.
conducted in 1980-2001. These data were combined using
CDAS software into an MMS-CDAS data system (MMS, 2001)
RESULTS AND DISCUSSION
for the Minerals Management Service and expanded for this
35°N
35°N
a b During intense warm periods, species such as brown pelican,
project. Of the data sets on the original MMS-CDAS CD-ROM,
black storm-petrel and black-vented shearwater, which zoo-
four aerial survey data sets contained data in the study area
200 m
Black-vented Shearwater Black-vented Shearwater
200 m
20
20
geographically are centered to the south of central California
from Point Arena to Point Sal. Of these, one program was still
0
0
0m
0m
(where waters are normally warmer and food availability rela-
ongoing and data from recent years were added to this data
El Niño ('97 - '98) La Niña ('99 - '00)
39°N
39°N
tively lower), move into central California waters when warmer
set. In addition, data from four ship-based survey programs
ocean temperatures expand northward. Many of these individu-
were converted to a compatible format for analysis. See section
als have foregone breeding owing to depleted food availability,
introduction for details on individual data sets.
which is often more extreme in areas to the south where these
species breed. Shown here are comparisons for brown peli-
Data sources for aerial at-sea data include MMS-CDAS (2001)
38°N
38°N
can and black-vented shearwater. In both cases, densities are
and California Department of Fish and Game Office of Spill
much higher in central California during warm-water periods.
Prevention and Response (CDF&G-OSPR), unpublished data.
In fact, during these conditions brown pelicans expand as far
Early data were collected using methods described by Briggs
north as the Columbia River and even farther; black-vented
et al. (1983); more recent data were collected using updated
37°N
37°N
shearwaters, however, don’t go much farther than central
technology but the same general method. Data sources for
California waters.
ship-based survey data include David Ainley and Carol Keiper
(unpublished data; see Oedekoven et al., 2001 for details on
The response of species to short-term cold-water conditions (La
methods). Although the at-sea data span the years 1980-2001,
Niña) is far less dramatic and, in fact, no examples could be found
data are not available for all seasons in all years. For the Up-
36°N
36°N
to clearly illustrate this. This is due to many factors, perhaps the
welling Season, data are from 1980-1982 and 1985-2001.
most important being that the geographic affinity of central California
For the Oceanic Season, data are from 1980-1982, 1991 and
marine birds is largely ‘subarctic’ and therefore, the central California
1994-2001. For the Davidson Current Season, data are from
avifauna is at the southern extreme of its range. As a result, there is
1980-1986 and 1991-2001.
little reason for northern species to shift into the area when the latter
35°N
35°N
c d becomes slightly colder. The other main reason for lack of examples
METHODS
illustrating response to cold conditions is that a major regime shift
At-sea densities are the result of a synthesis of data from eight
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
coincided with the best example, i.e. intense La Niña conditions
shipboard and aerial survey programs conducted in the study
Source Data: See text.
in 1999-00 (see Figure 60). Therefore, it is difficult to separate the
area in the years 1980-2001 (see “Data Sources” above).
Figure 59. Density during El Niño and La Niña events, 1997-2000.
factors responsible in the avifaunal shifts observed.
82
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
(estimated trackline width varied by platform, depending on
ABOUT THESE MAPS
Density During El Niño and La Niña Events: 1997-2000 speed and height above water) were used to estimate the area
A comparison of the abundance and distribution of two species
sampled. The number of marine birds seen in a cell was then
during intense El Niño events compared to La Niña events
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
divided by the area sampled in the cell to estimate density. If a
provides an example of how marine birds may respond to
200 m
Fork-tailed Storm-Petrel Fork-tailed Storm-Petrel
200 m
20
20
cell was censused more than once, densities were averaged,
short-term anomalous marine climate. In the case of regime
0
0
0m
0m
with adjustment made for effort.
shifts, which involve climate change on a decadal time scale, a
La Niña ('99 - '00)
El Niño ('97 - '98)
39°N
39°N
shift may have occurred during the study period corresponding
Density Species densities were mapped by grid cells on either side of
also to the switch from intense El Niño (Oceanic Season 1997
(Animals/km²)
the regime shift node, Oceanic Season 1997 through Upwelling
- Upwelling Season 1998) to intense La Niña (Oceanic Season
> 100.00
Season 1998 versus Oceanic Season 1998 through Oceanic
98 - Oceanic Season 2000). Therefore, at this time, it is difficult
50.01 - 100.00
Season 2000.
38°N
38°N
to perceive whether the changed bird distributional patterns
10.01 - 50.00
5.01 - 10.00
were short-term or long-term. Subsequently the cold conditions
1.01 - 5.00
RESULTS AND DISCUSSION
continued, thus indicating a longer-term regime shift (Bogard et
0.51 - 1.00
In 1999-2000, the mean state of the California Current System
al., 2000, Schwing and Moore 2000, Schwing et al. 2002).
0.11 - 0.50
may have moved from a “warm regime”, present since 1976,
0.06 - 0.10
0.01 - 0.05
37°N
37°N
to a “cold” regime (Schwing et al. 2002). On the other hand,
Densities are based on combined data of several studies (see
0.00
subsequent years of observation may indicate that we only
Methods and Data Sources below). The color and mapping
0 25 50 Km
witnessed the transition from one of the strongest El Niños to
intervals were customized to show the most structure and
one of the strongest La Niñas seen in the past 100 years.
highlight significant areas. Cells that were surveyed but in which
no birds were observed have a density of zero; unsurveyed
36°N
36°N
Regardless, in response, more northern species such as
areas are white. Blue lines indicate the National Marine
fork-tailed storm-petrel and black-legged kittiwake, which are
Sanctuary boundaries of Cordell Bank, Gulf of the Farallones,
present mostly during the Davidson current season, found
and Monterey Bay; bathymetric contours for the 200 meter and
the cooler, central California waters more to their liking and,
2,000 meter isobaths are also shown in blue.
rather than avoiding the area as in the 20 previous years of the
35°N
35°N
a b warm regime, arrived or remained longer to winter in very large
DATA SOURCES
numbers. Unfortunately, data during the Oceanic and Davidson
At-sea densities for this analysis are based on a subset of data
200 m
Black-legged Kittiwake Black-legged Kittiwake
200 m
current seasons of years since 2000 were not collected.
20
from the eight survey programs. These data were combined
20
0
0
0m
0m
using CDAS software into an MMS-CDAS data system (MMS,
La Niña ('99 - '00)
El Niño ('97 - '98)
39°N
39°N
2001) for the Minerals Management Service and expanded for
this project. Of the data sets on the original MMS-CDAS CD-
ROM, four aerial survey data sets contained data in the study
area from Point Arena to Point Sal. Of these, one program was
still ongoing and data from recent years were added to this data
38°N
38°N
set. In addition, data from four ship-based survey programs
were converted to a compatible format for analysis. See section
introduction for details on individual data sets. Data collected
since 1996 was used for this comparison.
37°N
37°N
METHODS
At-sea densities for this analysis are the result of a synthesis of
subsetted data from eight shipboard and aerial survey programs
conducted in the study area in the years 1980-2001 (see Data
36°N
36°N
Sources above). Observation and trackline data from these
studies were converted to a common format. All aerial data
were continuous; ship-based data were converted separately
into a continuous transect to the extent possible. From the
digitized survey data, the distributions of effort and of species
35°N
35°N
c d were mapped into five minute latitude by five minute longitude
cells using CDAS, a custom geographic information system
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Source Data: See text. for analyzing marine bird and mammal surveys (MMS, 2001).
The length and width of the survey trackline in a given cell
Figure 60. El Niño/La Niña Event changes, as an example of regime shift effects.
83
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
changes greatly due to the presence of southern
SECTION SUMMARY Table 15. Life history and management information for selected marine birds off north/central California.
Major Food Items3
Temporal Occurrence in Study Area hemisphere-breeding species that are ‘wintering’ in
The following section provides a summary discussion of Population Status in Study Area
large numbers in the study area during the Upwelling
the marine bird analyses, relative to the study area and
(squid, octopus)
Cephalopods
Invertebrates
Season and subarctic-breeding species ‘wintering’ in
the three national marine sanctuaries off north/central
Euphausiids
Gemeralist
large numbers during the Davidson Current Season.
California.
Plankton
Relative Primary Months
Benthic
Feeder
Additionally, many migrants pass through the region,
Protection Population Abundance Seasonal Months of When
(krill)
Fish
but foraging as they go, during the Oceanic and early
Life History and Management Characteristics Status1 Trend6 at Sea2 Occurrence5 Presence5 Breeding5
Common Name Scientific Name
Upwelling seasons.
The marine avifauna off north/central California, as Loons/Grebes
Pacific loon Gavia pacifica Unknown Common Seasonal Nov-Apr x
represented in the summary data set, are composed of
Common loon Gavia immer Unknown Uncommon Seasonal Nov-Apr x
In general, the highest concentrations and the greatest
76 marine bird species, with 39 occurring regularly enough Western & Clark's grebes Aechmorphorus occidentalis, A. clarksii Unknown Abundant Year-round Nov-Sept Apr-Sept x
variety of marine birds are found over the continental
to assess and map patterns of their occurrence. Table 15 Albatrosses/Petrels
shelf and slope, where there are more microhabitats
is a summary of selected life history and management Black-footed albatross Phoebastria nigripes Decreasing Common Year-round Mar-Aug x x
defined by ocean complexity (depth, currents, tide rips,
information for 39 of the marine bird species. Laysan's albatross Phoebastria immutabilis Unknown Rare Seasonal Nov-Mar x x
Northern fulmar Fulmarus glacialis Increasing Common Seasonal Nov-Mar x
etc.). A number of smaller, more discrete areas attract
Sooty shearwater Puffinus griseus Declining V.Abundant Seasonal Apr-Aug x x x x
marine birds because more food is available; the most
Species Relative Abundance in the Study Area. Based on Pink-footed shearwater Puffinus creatopus Stable? Common Seasonal Apr-Aug x x
important of these are Farallon Escarpment, Farallon
the analysis of the at-sea data and as indicated in Table Buller's shearwater Puffinus bulleri Unknown Common Seasonal Aug-Nov x
Ridge, San Francisco Bay tidal plume, inner Monterey
15, among the more regularly occurring species, two are Black-vented shearwater Puffinus opisthomelas Stable? Uncommon Seasonal Aug-Nov x?
Leach's storm-petrel Oceanodroma leucorhoa Unknown Common Seasonal March-Sept April-Sept x x x Bay, and Estero/San Luis Obispo bays.
very abundant, eight are abundant, 16 are common, 12
Ashy storm-petrel Oceanodroma homochroa SSC Increasing Common Year-round All Feb-Oct x x x
are uncommon, and two are rare. Relative abundance was Fork-tailed storm-petrel Oceanodroma furcata Decreasing? Uncommon Seasonal Nov-Mar x x x
Patterns Observed in Density, Biomass Density and
estimated on a logarithmic scale of number of individuals Black storm-petrel Oceanodroma melania SSC Unknown Uncommon Seasonal Nov-Mar x x
Diversity Across Species
seen within the study area on surveys during the study Sea Ducks (Scoters)
Another way to summarize occurrence patterns of
period. The majority of the species (26) are present only Surf scoter Mellanita perspicillata Stable Abundant Seasonal Nov-Apr x
Pelican/Cormorants
marine birds in the study area is to combine species
seasonally, but of the 14 species that breed in the study
California brown pelican Pelecanus occidentalis californicus FE, SE Increasing Common Seasonal Aug-Nov x
distribution and abundance data and analyze for the
area, 10 are present year round. Pelagic cormorant Phalacrocorax pelagicus Stable? Uncommon Year-round All Apr-Sept x x x x
biological metrics of species diversity, biomass and
Brant's cormorant Phalacrocorax pennicilatus Stable? Abundant Year-round All Apr-Sept x
density. Analyses for overall density, biomass and
Food Types. With regard to trophic relationships, the Double-crested cormorant Phalacrocorax auritus Increasing? Uncommon Year-round All Mar-Aug x
diversity were done with respect to ocean season
majority of marine bird species are either zooplanktivores Phalaropes
Red phalarope Phalaropus fulicaria Stable? Common Seasonal Mar-Aug x x x and to periods of unusual ocean climate (i.e., warm-
(generally smaller-bodied) and/or piscivores. Major prey
Red-necked phalorope Phalaropus lobatus Stable? Common Seasonal Mar-Aug x x x
water, cold-water and neutral periods). For these
items are: euphausiids (Thysanoessa spinifera and Gulls/Terns
summary analyses across species, we used the data
Euphausia pacifica), market squid (Loligo opalescens), Western gull Larus occidentalis Declining Abundant Year-round All Apr-Aug x x
for 76 marine bird species that were contained in the
juvenile rockfish (Sebastes spp, especially Sebastes California gull Larus californicus Increasing Abundant Seasonal Nov-Mar x
combined data set.
jordani), anchovy (Engraulis mordax), herring (Clupea Glaucous-winged gull Larus glaucescens Stable Uncommon Seasonal Nov-Mar x
Heermann's gull Larus heermanni Stable? Common Year-round Aug-Nov x
harengus), smelt (Atherinops californiensis and Spirinchus
Sabine's gull Xerna sabini Stable Common Seasonal Mar-Aug x x
Overall Density and Biomass. The seasonal and
starksi), Pacific saury (Cololabis saira), sardine (Sardinops Black-legged kittiwake Rissa tridactyla Increasing Common Seasonal Nov-Mar x
‘combined-season” densities of 76 marine birds were
sagax), midshipman (Porichthys notatus), surfperch Caspian tern Sterna caspia SSC Stable Uncommon Seasonal Mar-Nov Apr-Aug x
calculated for each five-minute by five-minute cell as
(several species) and myctophids (several species), with Elegant tern Sterna elegans Stable Uncommon Seasonal July?-Nov x
Arctic tern Sterna paradisaea Stable? Common Seasonal Mar-Nov x the number of individuals per km2. Biomass (kg/km2)
importance varying by the habitat and time of year in which
Alcids
was then calculated as the product of density and the
a particular bird species was foraging (Briggs and Chu
Common murre Uria aalge Increasing V.Abundant Year-round All Apr-Aug x x x x
mean body mass for each species, taken from Dunning
1987, Ainley and Boekelheide 1990). Pigeon guillemot Cepphus columba Stable? Uncommon Seasonal Mar-Aug Mar-Aug x x x
(1993). If a species was not listed in this reference,
Cassin's auklet Ptychoramphus aleuticus Decreasing? Abundant Year-round All Mar-July x x
the body weight of a related species of a similar size
Summary of Spatial and Temporal Patterns in Large Marbled murrelet Synthliboramphus marmoratus FT, ST Stable Uncommon Year-round All Apr-Aug x x
Xantus’ murrelet Synthliboramphus hypoleucus ST Unknown Uncommon Seasonal May-July x
was used.
and Relatively Smaller Areas
Craveri’s murrelet Synthliboramphus craveri Unknown Rare Seasonal May-July x
Table 16 is a summary of the temporal and spatial patterns Tufted puffin Fratercula cirrhata Decreasing Uncommon Seasonal Mar-Aug Apr-Aug x x
The distribution of marine birds across all taxa is similar
observed for the regularly occurring marine birds of the Rhinoceros auklet Cerorhinca monocerata Stable Abundant Year-round Nov-Aug Apr-Aug x x x x
for density (Figure 52) and biomass density (Figure 53).
study area. This summary was developed by visual Notes
This is because the avifauna is dominated (in terms of
inspection of the species seasonal density maps and is 1. Management status categories are as follows: FE–federally endangered; FT–federally threatened; SE–state endangered; ST–state threatened; SSC–state species of special concern.
2. Relative abundance estimates are based on the number of individuals tallied in the at-sea survey data, and the categories are defined as follows:
both number of individuals and their body mass) by the
provided as a simple summary of species distributions
Rare – up to 100 birds; Uncommon – up to 1,000; Common – up to 10,000; Abundant – up to 100,000; and Very Abundant – up to 1,000,000.
Common Murre and Sooty Shearwater. Therefore, the
and abundance by season and for selected habitat and 3. Information on food items are from Ainley & Sanger 1979, Briggs & Chu 1987, and Ainley & Boekelheide 1990.
patterns in sum are close to what is evident individually
management features. 4. Entries with question marks are the principal investigators best estimate.
for these particular dominant species. Accordingly, the
5. Timing information is from Cogswell, 1977 and Ainley and Boekelheide, 1990, except for Caspian tern breeding time, which came from Joelle Buffa, FWS, pers. comm.
major biomass and density areas i.e., the inner and
6. Information on population status was based on analysis of the shipboard data sets, from 1985-2001.
It is obvious that large numbers of marine birds occur in the
7. Months of presence and breeding in the study area are approximations, because timing is strongly influenced by the interannual variability of environmental conditions in the study area.
outer shelf, are biased toward these two species.
study area year round. The species composition, however, 8. Information on population status and temporal occurrence refers only to birds and their activities in the study area.
84
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Phalaropes can also be very abundant but don’t contribute
much biomass and they are also most abundant over
Table 16. A summary of temporal and spatial patterns in the at-sea survey data (1980-2001) of selected marine birds off north/central California.
the shelf waters. Smaller-scale biomass and density hot
spots are also the same, e.g. inner Monterey Bay, San
Species
Francisco Bay tidal plume, the area around the Farallon
Occurs in
Associations with Large Study
Islands, Pioneer and Ascension Canyons, and Cordell
Bathymetrically-Defined Areas
Seasonal Occurrence Associations with Discrete Physiographic/Oceanic Features
Area, But
Bank. Moreover, as will be noted later (see also below)
Mostly
Davidson Ascen- Estero
Outside
density and biomass are more spread out during warm-
Ocean Current Coast & Outer Upper Lower Deep San Fran- cion, Mon- David- Bay/San
Sanc-
Upwelling Season Season Inner Shelf Slope Slope Ocean Farallon cisco Pioneer Cabrillo, Pt. Mon- terey Pt. son Luis
water than during cold-water or neutral periods.
Family Name/ tuary
Season (8/15- (11/15- Shelf (~0 (~100- (~200- (~1000- (beyond Cordell Bodega Fanny Escarp- Farallon Bay Sea Pioneer & Año Ano terey Bay Carmel Pt. Sur Sur Sea- Obispo
Species Common Name Bounds Bank Canyon Shoal ment
(3/15-8/14) 11/14) 3/14) 100m) 200m) 1000m) 2000m) 2000m) Ridge Plume mount Canyon Canyons Nuevo Canyon Inshore Canyon Slope Shelf mount Bay
Podicipedidae
Seemingly, highest biomass occurred during the Upwelling
Western and Clark's grebes x x **X X X x X X
and Oceanic seasons, as was the case for density. Any
Gaviidae
Pacific loon X ***X x x x X X
seasonal difference was least clear in regard to density.
Diomedeidae
Black-footed albatross ***X x x X X X x X X X x x X x x
Laysan albatross x o ***X X X X x X X
Diversity. To assess species diversity, the Shannon Index
Procellariidae
(Shannon and Weaver 1949) and species density data
Northern fulmar x x ***X X X X x x x x X x x x x x x
Sooty shearwater ***X x o X X X x x X x x x x X X X X X x X
were used (see Figure 54). This index measures the
Short-tailed shearwater o o ***X X X x x X x
degree to which the species assemblage is dominated
Pink-footed shearwater ***X x o X X X x x x x X x x X x
Buller's shearwater o ***X o X X X x X x X x x x x x x
by a single species. For example, if “Species A” dominates
Black-vented shearwater o **X x X X x X x x x
all the species seen within a cell, then diversity is low;
Hydrobatidae
Leach's storm-petrel **X **X o X X X x x x x
and if all species are “evenly” represented, then diversity
Fork-tailed storm-petrel o o X x X X x X X x x x
is high. Diversity was calculated using all bird species
Ashy storm-petrel **X ***X **X x X X x X x X X x x X x
Black storm-petrel o **X **X x X X x X x x X x
(n=76) in the data set, for each ocean season and for all
Anatidae
ocean seasons combined.
White-winged scoter o x **X X X X X
Surf scoter o x **X X X X X
Pelecanidae
Highest diversity indices are about the same in all
Brown pelican **X **X x X x x x X X X x X
three seasons. In all cases, at the smaller spatial scale
Phalacrocoracidae
Brandt's cormorant ***X ***X ***X X X x x x X X X X x x
(less-detailed), species diversity was greatest along
Pelagic cormorant **X **X **X X x X x x x
the continental slope. This is expected given that the
Double-crested cormorant *X o o X x X X x x
Scolopacidae
slope habitat constitutes the boundary as well as the
Red phalarope ***X x o X X X x X x X x x x x x
overlap between the shelf and oceanic habitats. At a
Red-necked phalarope ***X x o X X x x x x x x x x x x x
Laridae
larger scale (more detailed), in all seasons there was an
Glaucous-winged gull o x ***X X X x x x x x X x x X X
area of notable diversity seaward of the Farallon Islands
Western gull ***X ***X ***X X X X x x x X X X X X X x X
California gull o x ***X X X x x X x X x X
(Farallon Escarpment) and to some degree outside of the
Ring-billed gull o x **X X x X X x X
sanctuary boundaries. Likely the diversity here resulted
Mew gull o x **X X x X X
Heermann's gull o x **X X x X X
from a coincidence of occurrence of: 1) oceanic species;
Bonaparte's gull ***X x o x X X x X x X
2) shelf species; 3) Farallon breeding species that would
Sabines gull ***X x o X X X x X x x X
Black-legged kittiwake x o ***X X X X x X x x X x x x x x x x
not occur offshore were it not for the Farallones; and 4)
Caspian/Elegant terns **X x o X X X x
the location of a persistent boundary there of a coastal
Arctic tern ***X ***X o x x X X x X x x x x
Forster's tern **X x o X X X
upwelling front that extends southwestward from Point
Alcidae
Arena. Accordingly, during La Niña, when upwelling
Common murre ***X ***X ***X X X x x x x x X X x x x x
features are well developed, this area exhibits much
Pigeon guillemot **X x o X X X x X X
Tufted puffin *X x o x X X X X X x
greater diversity than is apparent during El Niño.
Rhinoceros auklet ***X ***X x x X X x x x x x x X x x x
Cassin's auklet ***X ***X x X X x X x X X X x x
Marbled murrelet *X *X *X X x X X
Although there was a significant correlation between bird
Xantus'/Craveri's murrelets ***X x x X X X X X
diversity and effort, the observed patterns of bird diversity
All Xs 29 35 27 21 28 24 22 16 11 13 14 14 22 17 20 6 18 19 11 15 18 6 8 8 6 16
Large X's 25 14 21 20 19 19 13 5 11 3 1 4 16 9 17 0 2 1 5 7 16 0 8 2 0 10
are robust and were largely unchanged by methods
Notes
1. A summary of temporal and spatial patterns in the occurrence of 44 marine birds off north/central California, based on visual inspection of the species' seasonal density maps by the principal investigators.
designed to correct for effort. See an additional discussion
2. The spatial and temporal patterns summarized here may be affected by variation in the sampling effort of the combined data sets; that said, this table is included because it provides a summary of the relative use of the various habitat features, as viewed in the species maps.
of bird diversity in the Integration section.
3. In the "seasonal occurrence" columns, the number of asterisks indicate the number of sanctuaries in the study area that are used by the species during the season. A large "X" means a relatively major occurrence in the data sets, a small "x" means a minor occurrence,
and a "o" means the species was mostly absent.
4. Under the heading for "large, bathymetrically-defined areas", a large "X" indicates where the species was most abundant, a small "x" means a minor occurrence, and a "o" means the species was mostly absent.
5. Under the heading for "discrete physiographic/oceanic features", large X’s refer to ‘hot spots’; small x's indicate areas of secondary importance for that species.
6. A blank in the table means a species was not present at the location indicated in the maps/data reviewed.
85
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Important Marine Areas for Birds: Considering Overall Shearwater is one of these and becomes the most abundant Biomass remains high but shifts closer to shore than during The next three variables of importance were “ENSO” (eight
Biomass, Density and Diversity of Marine Birds. All species in the study area. the Upwelling Period, in large part due to an inshore shift of species), “Year” (seven species), and “Distance to Colony”. In
marine habitat off central California, especially that of the murres and shearwaters. some respects, for species having many small colonies (e.g.
continental shelf and slope, is fully used by marine birds. Owing to the addition of Sooty Shearwaters to the avifauna and pelagic cormorant), Distance to Land and Distance to Colony
Based on the analyses of maps for overall density, biomass the continued abundance of Common Murres, overall density Davidson Current Season (~Winter). During this season, may have co-varied. The variable “Year” indicated whether
density, and diversity, the following at-sea areas were identified and biomass of marine birds is highest during the Upwelling when ocean temperatures are relatively warm and there is there was an increasing or decreasing trend in the species’
as important (Table 17). Areas with more and bigger Xs may and Oceanic Seasons and is widely spread from the coast no upwelling (but frequent downwelling owing to southerly abundance. An effect of ENSO would indicate an especially
be more important, as they show more expression of density, to beyond the shelfbreak. Diversity over the shelf, where the storms) the area is inundated by such species as black-legged complex relationship, possibly meaning either an effect of prey
biomass and diversity. shearwaters and murres mostly reside, is relatively low. kittiwake, northern fulmar and several larger gulls. All these availability or ocean climate.
species nest outside of the region. Also present are nesting
Important Breeding Colonies for Marine Birds. Although Oceanic Season (~Autumn). In this season, when upwelling species that reside year-round in the region, such as Brandt’s Please note that while the three most important variables of
breeding colonies and roosts are on land and technically not winds have noticeably relaxed, allowing offshore, warmer cormorant, western gull, common murre, rhinoceros auklet those evaluated are indicated in Table 19, for most species
part of the study area, a table and map of the major colonies oceanic water to flow shoreward, the avifauna begins to and Cassin’s auklet. In fact, many of the latter species begin there are likely other variables of greater importance (e.g.,
are included in this section because they provide a context for diversify. However, as more sooty shearwaters and other to occupy nesting colonies during this season, well before the prey availability, depth of thermocline). On average, only
understanding the distributions of species that breed or roost southern hemisphere seasonal residents depart the avifauna nesting period. about 20% of the variance was explained with the top three
in the study area based on size and species composition (most variables presented. Additional data and time would be required
data was from Carter et al., 1992, with updates, as available). During this season, to evaluate other variables that might be of greater importance
Table 17. Important at-sea areas and ocean seasons for marine birds off north/central California, based on maps
Table 18 shows the top 40 marine bird breeding colonies in the the species diversity in explaining the variation in a species’ distributions.
of biomass, density and diversity.
study area; see Figure 55 for a map of these locations; see the of shelf waters
Biomass/Density Diversity
CD-ROM X for a full listing of breeding sites. increases. In fact, Species Use of the Water Column. Several marine bird
Davidson Davidson
areas of high species are capable of exploiting the entire water column of the
Current Upwelling Oceanic Current Upwelling Oceanic
Marine birds in this area breed mainly during the Upwelling diversity are more shelf, e.g. Pacific loon, western/Clark’s grebes, and common
Season Season Season Season Season Season
Area
Season, anticipating that food availability will be greatest widespread in this murre, but for unknown reasons (possibly prey selection,
x X
Cordell Bank
from July-October, toward the end of this season and into the season than in the perhaps interference competition from murres and shearwaters)
Farallon Escarpment X x x X X X
Oceanic Season. During this period ample supplies of prey will others. the grebes mainly frequent the inner most portion of the shelf.
(slope)
be needed to feed growing chicks and recently fledged young. The very abundant common murre is found everywhere on the
Farallon Ridge (includes X X x
Farallon Island area)
Egg laying occurs in March-May, depending on species, and Analysis of shelf especially during the breeding/Upwelling season. Other
different species require different amounts of time to complete Variation diving species, such as scoters or marbled murrelet, frequent
San Francisco Bay Tidal x X X
Plume
the breeding task (petrels longest, gulls shortest). in Species only shallow waters of the inner shelf, while other species,
Pioneer Canyon X x x Abundance such as tufted puffin, rhinoceros auklet and Cassin’s auklet
Año Nuevo Shelf X X
The greatest concentration of colonies occurs in the Gulf of Patterns. Many frequent waters much deeper (continental slope) than their
the Farallones, in the broad shelf area from Point Reyes south factors influence diving capabilities allow. Species such as the very abundant
Ascension, Año and X x x X x
Cabrillo Canyons
to Año Nuevo and out to the Farallon Islands. The breeding the distribution sooty shearwater (a shallow diver to 20 meters) are found
Monterey Bay Inshore X x X X
avifauna is dominated by alcids, with six species. Fifteen and abundance everywhere from outer slope to inner shelf.
Monterey Bay Canyon X x
species of marine birds breed at sites within or immediately of marine birds;
adjacent to the National Marine Sanctuaries in the study area; Carmel Canyon X in this study, the These differences in patterns of habitat use are likely related
several others breed inland or in San Francisco Bay and to a effects of nine to factors such as the occurrence patterns of different
Point Sur Shelf x
lesser degree use marine sanctuary waters. independent prey (species/sizes), interspecific competition, or temporal
Point Sur Slope X x
variables on occurrence of certain prey (species/sizes). The latter would
Estero Bay & San Luis X x
Importance of Ocean Seasons to Marine Birds. As seen in species density account for why some year-round resident species feed over
Obispo Bay
the maps for individual species, temporal differences in spe- were investigated waters of different depths during one season compared to
Note: Large, bold Xs refer to most important areas, and smaller xs refer to other important areas.
cies occurrence patterns are strong for many species in the for 26 species. another. Species such as the sooty shearwater, which use a
study area (see Tables 15 and 16). Below is a brief summary The data used for wide range of ocean depths and habitats, are likely to be more
of marine bird activity in the three ocean seasons. the regression analyses were a subset of the mapped generalized in prey selection, possibly due to their fast, efficient
becomes more sparse. At this time, resident breeding species
data, and included data from the Davidson Current Period flight allowing them to forage over much larger areas than many
are also dispersing more widely as they finish breeding
Upwelling Season (~Spring/Summer). With the onset of upwell- from 1985 through the same for 2002; also, cells with area other marine birds, particularly the alcids and cormorants.
duties.
ing, when cold, nutrient-rich water is brought to the surface by surveyed less than 0.25km2 were excluded.
persistent northwest wind and the Coriolis effect, most of the Response to “Short-Term” Changes in Ocean Climate. The
Additionally, several species (e.g., phalaropes, jaegers, Arctic
seasonal winter residents depart and several other species Among the nine variables investigated, the three most study area is subjected frequently to shifts in marine climate of
terns, Sabine’s gulls) are migrating through the region, the local
migrate through the region (e.g. Sabine’s Gull, Arctic Tern). Ar- important variables that explained variation in species density different scales and periodicity and this makes management
nesting species are all present, and several species that nest
riving are several species that nest in the Southern Hemisphere, were “Distance to Land” (16 of 25 species), “Ocean Season” a challenge, because populations are affected by natural
elsewhere are abundant as well (e.g. several shearwaters,
thus spending their ‘wintering’ period in the region. The Sooty (13 species), and “Ocean Depth” (11 species, see Table 19). environmental factors that cannot be addressed proactively
brown pelican, and Heermann’s gull).
86
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Table 18. Major marine bird colonies along the central California coast
Double- Other Site Total
USFWS Leach's Ashy Brandt's crested Pelagic Pigeon Rhino- Species (No. of
CA Colony Colony Storm- Storm- Cor- Cormor Cor- Western Caspian Common Guille- Marbled Cassin's ceros Tufted (see Breeding
Colony/Composite Name Number Number Latitude Longitude Petrel Petrel morant ant morant Gull Tern Murre mot Murrelet Auklet Auklet Puffin notes) Birds)
Fish Rocks ME-384-10 404-003 38°47'59" N 123°35'31" W 100 211 123 170 119 P 4 15 6 748
Gualala Point Island SO-384-01 404-004 38°45'3" N 123°31'42" W 521 4 26 29 1 581
Russian Gulch SO-382-08 404-033 38°28'0" N 123°9'35" W 227 42 20 7 296
Russian River Rocks SO-382-09 404-005 38°27'14" N 123°8'34" W 51 422 125 44 5 2 649
Arched Rock SO-382-11 404-006 38°25'53" N 123°7'31" W 717 9 34 2 762
Bodega Rock SO-380-02 404-008 38°17'48" N 123°2'49" W 1,228 24 30 2 1,284
Bird Rock MA-380-04 404-010 38°13'49" N 122°59'35" W 15 55 37 168 115 3 H 6 399
Point Resistance MA-374-03 429-024 37°59'54" N 122°49'40" W P 46 H H 8 3,518 50 3,622
Point Reyes MA-374-01 429-001 37°59'30" N 122°58'59" W 15 1,522 266 178 15,155 616 4 6 17,762
Millers Point Rocks MA-374-04 429-002 37°58'52" N 122°48'34" W 114 59 30 358 55 1 617
Double Point Rocks MA-374-05 429-003 37°56'50" N 122°47'8" W 30 180 9 8 4,464 22 4,713
San Francisco Bay & Alcatraz Island Composite Composite 37°49'33" N 122°25'19" W 9 2,789 4 4,174 2,818 10 8498 18,302
North Farallon Islands SF-FAI-01 429-051 37°46'4" N 123°5'56" W 161 62 32 27,308 42 27,605
South Farallon Island SF-FAI-02 429-052 37°42'0" N 123°0'0" W 1,400 1,990 9,466 486 442 15,095 103,588 499 18,807 516 128 30 152,447
Devil's Slide Rock SF-372-03 429-014 37°34'28" N 122°31'39" W 7 46 16 246 30 345
Big Basin State Park and vicinity None None 37°8'48" N 122°17'58" W 600 600
Vicinity of Año Nuevo Island and Point Composite Composite 37°6'30" N 122°20'8" W 4 117 1,382 219 24 224 27 1,997
Greyhound Rock to El Jarro Point SC-370-01 429-049 37°3'31" N 122°15'0" W 66 2 321 9 398
El Jarro Point to Davenport SC-370-02 429-050 37°1'5" N 122°12'23" W 308 22 313 1 644
Davenport to Sand Hill Bluff SC-364-01 454-038 36°59'44" N 122°10'32" W 13 495 1 509
Cannery Row MO-362-02 454-044 36°36'46" N 121°53'47" W 198 86 88 372
Bird Rock MO-362-03 454-006 36°35'30" N 121°57'59" W 2,651 16 2 2,669
Guillemot Island Area MO-362-06 454-023 36°31'45" N 121°56'47" W 554 20 30 18 10 632
Bird Island MO-362-09 454-009 36°30'24" N 121°56'32" W 6,151 4 90 5 2 6,252
Castle Rocks and Mainland MO-362-19 454-010 36°22'35" N 121°54'25" W P 750 46 12 1,050 19 1,877
Hurricane Point Rocks MO-362-20 454-011 36°21'40" N 121°54'25" W P 29 H 14 613 20 H 676
Anderson Canyon Rocks MO-360-13 454-016 36°9'7" N 121°39'53" W 321 5 26 2 32 386
Burns Creek Rocks MO-360-14 454-017 36°8'29" N 121°39'28" W 323 2 12 337
Plaskett Rock MO-354-07 477-002 35°55'13" N 121°28'41" W 849 H 5 H 1 855
Cape San Martin MO-354-08 477-003 35°53'16" N 121°27'54" W 750 2 18 349 H 1,119
Redwood Gulch Rock MO-354-12 477-005 35°50'19" N 121°24'3" W 372 2 H 374
La Cruz Rock SL-354-04 477-006 35°42'22" N 121°18'45" W 678 18 696
Piedras Blancas Island SL-352-01 477-007 35°39'51" N 121°17'17" W 2,627 34 29 3 H 1 2,694
Morro Rock and Pillar Rock SL-352-07 477-026 35°22'13" N 120°52'8" W 117 24 53 114 24 332
Fairbank Point SL-352-08 477-044 35°21'5" N 120°50'37" W 331 331
Unnamed Rocks SL-350-03 477-010 35°14'40" N 120°53'38" W 174 49 242 6 471
Lion Rock SL-350-05 477-011 35°13'3" N 120°52'16" W 277 H 24 18 1 320
Pup Rock and Adjacent Mainland SL-350-04 477-028 35°13'17" N 120°52'13" W 1,309 44 2 1,355
Pecho Rock SL-350-09 477-032 35°10'45" N 120°48'59" W 321 14 335
Table Notes
1. This table contains numbers of breeding birds at specific colonies or composite sites, for the species indicated. The table shows the best available data for approximately 40 of the largest colonies and colony composites for selected marine birds that occur in the study area.
All colonies shown have 296 or more breeding birds, and sites are listed from north to south.
2. The primary source for these data is Carter et al. 1992 (unpublished data); most estimates from this source were made from 1989-1991. Older data older data are indicated by italics (e.g., data for Leach's storm-petrel), and more recent or updated data (from various sources,
identified below) are indicated in bold type.
3. Key to symbols in table: H=historically nesting species; P = present and probably breeding. A blank in the table for a species/colony cell means the species was not present in the available data.
4. The column titled "Other Species" contains available estimates for all other breeding bird species. For most sites, this includes Black Oystercatcher. For the San Francisco Bay/Alcatraz composite site, the "Others" estimate includes California Gull, Forster's Tern, and Least Tern,
which breed at sites in the Bay.
5. For Ashy Storm-petrel, the updates at Bird Rock, Point Reyes and Double Point are from 2001 (Whitworth et al. 2002). The update at South Farallon Island is from 1992 (Sydeman et al. 1998).
6. The estimate of 600 breeding Marbled Murrelets at Big Basin State Park and Vicinity was provided by Laird Henkel, pers. comm.
7. For Cassin's Auklet and Rhinoceros Auklet, the updates in the vicinity of Año Nuevo are from 2002 and do not include the small breeding area within the Brandt's Cormorant colony. Sources: Thayer and Sydeman 2002a,b.
8. The estimates of breeding birds for Leach's Storm-petrels are in italics and from Ainley and Lewis (1974); these older estimates are likely much higher than the current colonies' status. The number of breeding birds at Fish Rock has likely signficantly decreased; in August 1989, no Leach's
Storm-petrels were captured at Fish Rocks, but this may be due to low sample effort (Harry Carter, pers. comm). And based on annual mark-recapture efforts since 1992, the number of breeding birds at S. Farallon Island has also likely significantly decreased (Bill Sydeman, pers. comm.).
9. Updates from 2002 (in bold) are included for the following eight species at South Farallon Island: Cassin's Auklet, Common Murre, Tufted Puffin, Pigeon Guillemot, Double-crested Cormorant; Pelagic and Brandt's cormorants, and Western Gull. Source: Warzybok et al. 2002.
10. The 2002 update for 246 breeding Common Murres at Devil's Slide Rock was provided by Gerry McChesney, U.S. Fish and Wildlife Service.
11. Fork-tail Storm-petrels have been noted as present and probably breeding at South Farallon Island, but no estimate is available.
87
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Table 19. Three most important variables (of nine investigated) having independent effects in explaining the variance in density of 25 selected
the marine sanctuaries; these species include pink- both were greater during these periods than during the Neutral
marine bird species.
footed shearwater, Leach’s storm-petrel, and western, period. On the other hand, the brown pelican responded strongly
Number of Percent glaucous-winged and Heermann’s gulls; see Table 20 to the disparity of conditions between the very strong 1997-98
Birds Variance
below. Also increasing in the study area during warm- El Niño and the very strong 1999-00 La Niña (Figure 59). The
Recorded Explained by
water periods were less abundant species including black-vented shearwater, for which there were insufficient data
During Top Three Three Most Important Variables of Those Investigated,
black and least storm-petrels and black-vented for long-term trend analysis, responded much more dramatically,
Surveys Variables in Order of Importance
Species
shearwater. with large numbers invading central California waters during the
Pacific loon 3,802 10.6 Ocean season, distance to land (-), latitude (-)
1997-98 El Niño. Finally, there were no significant differences
Western grebe 7,080 15.5 Ocean season, distance to land (-), ocean depth (-)
Cold-water Periods (e.g., La Niña events). During
Black-footed albatross 3,149 22.2 ocean depth (+), distance to land (-), year (-)
coldwater periods, areas
Laysan albatross 96 6.9 Ocean season, ocean depth (+), distance to land (+)
of relatively high avifaunal
Northern fulmar 5,882 21.3 Ocean season, ENSO period, year (+) Table 20. Effects of ocean season and ENSO events on the abundance of 26 marine bird
biomass and density
Sooty shearwater 296,065 43.4 Ocean season, year (-), ENSO period species off central California between 1985 and 2002, as determined through multiple regres-
expanded to cover broader
Pink-footed shearwater 4,145 13.1 Ocean depth (-), distance to land (-), ESNO period sion analyses.
portions of the study area;
Leach’s storm-petrel 1,414 28.4 Ocean season, distance to 2000m isobath (+), ENSO period Ocean Season(s)
11 of the predominant
Ashy storm-petrel 4,201 17.3 ENSO period, season, year (+) of Highest ENSO Event of
species became more
Fork-tailed storm-petrel 393 9.2 ENSO period, season, ocean depth (+) Species Abundance* Highest Abundance*
abundant. Species whose
Brown pelican 2,333 15.2 Distance to land (-), latitude (-), Ocean season
Pacific Loon DC LA
abundance showed the
Brandt’s cormorant 9,482 28.7 Distance from colony (-), dist. to 200 m isobath, dist. to land (-)
Western Grebe DC LA
greatest increases were the
Pelagic cormorant 396 6.1 Distance to land (-), ocean depth (-), dist. to 200 m isobath (-)
Black-footed Albatross UP NE
western grebe, northern
Double-crested cormorant 300 9.7 Dist. from colony (-), dist. To 200m isobath (-), dist. to land (-)
Laysan Albatross DC LA
fulmar, ashy storm-petrel, red
Red & Red-necked phalaropes 49,195 9.6 ENSO period, distance to land (-), ocean depth (-)
Northern Fulmar DC LA
and red-necked phalaropes
Western gull 29,545 44.2 Distance from colony (-), distance to land (-), ENSO period
(grouped), and black-legged
Glaucous-winged gull 767 17.1 Ocean season, ocean depth (-), latitude (+) Sooty Shearwater UP NE
kittiwake (Table 20). Others,
Heermann’s gull 1,121 6.5 Distance to land (-), ENSO period, latitude (-) Pink-footed Shearwater OC EL
whose abundance were also
California gull 13,721 24 Ocean season, year (+), latitude (+) Leach’s Storm-Petrel DC/UP EL
Black-legged kittiwake 4,565 28.9 Ocean season, ENSO period, year (+) significantly greater during La Ashy Storm-Petrel OC LA
Common murre 131,675 52.3 Distance to colony (-), ocean depth (-), distance to land (-) Niña periods were Pacific loon, Fork-tailed Storm-Petrel DC LA/EL
Rhinoceros auklet 14,679 19.8 Distance to land (-), season, ocean depth (+) Laysan albatross, California
Brown Pelican OC LA/EL
Tufted puffin 235 10.1 Distance from colony (-), year (-), distance to 200 m iso (+) gull, rhinoceros auklet, and
Brandt’s Cormorant UP ns
Cassin’s auklet 63,465 25.8 Distance to land (-), year (-), ocean depth (-) marbled murrelet.
Pelagic Cormorant UP ns
Marbled murrelet 273 4.7 Distance to land (-), latitude (+), ENSO period
Double-crested Cormorant UP ns
Notes Neutral Periods (i.e., neither
Red & Red-necked Phalaropes OC LA
1. For “continuous” variables, a positive (+) included with a variable indicates that density increased with an increase in unusually warm nor cold
the magnitude of that variable; (-) denotes the opposite. Western Gull UP EL
water). Four species, black-
2. The nine independent variables used in the regression analysis were: distance to nearest land; ocean season; ocean depth; Glaucous-winged Gull DC EL
footed albatross, sooty
ENSO period; year; latitude; distance to colony; distance to 200m isobath; and distance to 2,000m isobath. shearwater, tufted puffin Heermann’s Gull ns EL
3. Species that breed in the study area are shown in bold. and Cassin’s auklet were California Gull DC LA
significantly more abundant Black-legged Kittiwake DC LA
by management. The individual species text (accompanying two periods were chosen because climate differences were during the neutral period than Common Murre UP ns
each map) details where selected species become more or less extreme, i.e. among the strongest ENSO events of the past 100 during the warm or cold periods. Rhinoceros Auklet DC LA
abundant in north/central California during periods of warmer years, and they occurred adjacent to one another. Therefore,
Tufted Puffin UP/DC NE
or colder than average ocean temperatures. Actually, the shift a comparison of population response was not confounded by Other Responses to Short-Term
Cassin’s Auklet UP NE
in temperature is a proxy for many other changes, all of which long-term trajectories in base population size. Tables 20 and 21, Climatic Change. In regression
Marbled Murrelet DC/UP LA
ultimately affect the food web. as well as Figure 58, provide some examples of these changes analyses, the densities of the brown
* Notes. Ocean seasons are: Davidson Current (DC), Upwelling (UP), and Oceanic (OC);
due to interannual climate events (see Table 14). pelican and fork-tailed storm-petrel
ENSO periods are El Niño (EL), La Niña (LA), and neutral (NE). For species having
To illustrate the short-term ocean climate effects using species were not significantly different during
significant differences in abundance during respective seasons/periods (Sidak tests,
maps, a comparison was done using selected species density Warm-water Periods (e.g., El Niño events). During warm-water warm-water and cold-water periods.
P < 0.01), the season/period in which they were most abundant is given. If
maps for a specific El Niño and La Niña period. For the El events (including El Niño), many marine bird species tended Basically, this was because such
Niño period, the most recent, and very intense El Niño event, to occur closer to shore than during other years (Ainley and periods varied greatly in intensity, densities did not differ between the two seasons/periods in which they were most
Oceanic Season 1997 through Upwelling Season 1998, was Boekelheide 1990, Oedekoven et al. 2001). During warm- thus reducing effects especially if abundant, then the two are listed (eg. DC/UP, where densities were slightly higher
used; for the La Niña event, the period covering the Oceanic water periods, five of the predominant species became more other factors (not studied) were more in DC season that the UP season). Species for which there was no significant effect
Season 1998 through Oceanic Season 1999 was used. These abundant in central California waters in general, and inside important. However, the densities of of season or period are denoted with “ns”.
88
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
MAJOR SECTION CONTRIBUTORS
Table 21. A summary of changes in marine bird occurrence patterns, as a response to breeding season during warm-water periods national marine sanctuaries off north/central California
warm and cold ocean anomalies, as determined by visual comparison of species’ maps David Ainley, H.T. Harvey and Associates
(Table 21). This spreading is often affected generally encompass the areas of high concentrations and
during the 1997-1998 El Niño event and the 1999-2000 La Niña event. R. Glenn Ford, R.G. Ford Consulting Co.
most strongly by Farallon breeding species, diversity for marine birds, except for the western edge of the
Janet Casey R.G. Ford Consulting Co.
which usually concentrate near the Gulf of the Gulf of Farallones area and the "sanctuary exclusion area" off
Effect on Distribution
Larry Spear, H.T. Harvey and Associates
Farallones (outer shelf), and which move more San Francisco and Pacifica.
Species No Change El Niño La Niña
Carol Keiper, Oikonos
X to coastal waters.
Western grebe
Lisa Ballance, Southwest Fisheries Science Center, NOAA
X Owing perhaps to a response to competition for food by the
Pacific loon
X Tracy Gill, NCCOS, NOAA
Population Trends. Owing mainly to longer- large numbers of marine birds nesting on the Farallon Islands
Black-footed albatross
X Wendy Willams, R.G. Ford Consulting Co.
term, decadal shifts in marine climate (Hare and Point Reyes Headlands, during the breeding (Upwelling)
Laysan albatross
More spread Ken Buja, NCCOS, NOAA
Northern fulmar and Mantua, 2000, Mantua and Hare 2002), a season high concentrations of several breeding species extend
To Monterey number of species exhibited gradual changes seaward of the western boundary of the Gulf of the Farallones
REVIEWERS
Bay
Sooty shearwater in population size within the study area, from National Marine Sanctuary, over the Farallon Escarpment and
Two reviews were done for the marine bird analyses. The first
X
Pink-footed shearwater 1985 to 2002. These patterns were revealed beyond. This is especially true of Ashy Storm-petrel, Western
was a workshop to review draft maps in October 2002, and the
X
Buller’s shearwater using regression analyses, especially in cases Gull, Common Murre, and Rhinoceros Auklet. For the gull and
To Gulf of second review focused on the overall draft bird report (which
where Year was an important explanatory murre these deeper waters are not their preferred foraging
Farallones
Black-vented shearwater contains the maps) and was conducted via email November
variable to species occurrence (Table 19). habitat, but they choose to forage there during breeding
X
Leach’s storm-petrel and December, 2002. Most of the review comments were ad-
because more suitable continental shelf habitat to the north
More spread
Ashy storm-petrel dressed or incorporated for this product, which is a summary
Ashy storm-petrel and California gull and south is too far out of range.
X
Fork-tailed storm-petrel of a full bird report to be released later this year.
exhibited a gradual increase in population
To Gulf of (from 1985-2002); Tufted puffin showed a To a lesser degree, a smaller high density area existed seaward
Farallones Workshop Participants/Reviewers:
Black storm-petrel gradual decrease. The Black-legged kittiwake of Año Nuevo Island, where there is a smaller, but important
More spread David Ainley, H.T. Harvey and Associates
Brown pelican increased gradually, too, but was especially colony. These three colonies (Farallon Islands, Point Reyes
More spread
Brandt’s cormorant Sarah Allen, Point Reyes National Seashore, National Park
abundant after 1998 (Figure 60). Headlands and Año Nuevo Island), and the waters between
More spread
Pelagic cormorant Service
them which comprise the Gulf of the Farallones, possess
X
Red phalarope Lisa Ballance, Southwest Fisheries Science Center, NOAA
For other species the pattern was more populations that interact regularly in the shallow waters that
X
Red-necked phalarope Janet Casey, R.G. Ford Consulting Co.
complex. Black-footed albatross, sooty lie between them; many individuals marked at one have been
More confined
Glaucous-winged gull Glenn Ford, R.G. Ford Consulting Co.
shearwater, fork-tailed storm-petrel and seen at the other two sites. Therefore, in terms of marine birds
X
Western gull Carol Keiper, Oikonos
Cassin’s auklet showed a gradual decrease the waters of the Gulf of the Farallones, as defined above,
More confined
California gull Nora Rojek, formerly of California Department of Fish and
from 1985 until about 1999, when they began constitute a natural management unit.
Heermann’s gull Game
to increase.
To Monterey Jan Roletto, Gulf of the Farallones National Marine Sanctuary
In addition, it was apparent from visual inspection of the maps,
Bay
Bonaparte’s gull Program, NOAA
The year 1999 is when the system likely that the "sanctuary exclusion area" (i.e., the ocean area off San
X
Sabine’s gull Ed Ueber, NOAA, Gulf of the Farallones National Marine
shifted from a ‘warm’ to a ‘cold’ ocean regime Francisco and Pacifica that is excluded from the Monterey Bay
Confined to Sanctuary Program
(Bogard, 2000; Schwing and Moore, 2000). National Marine Sanctuary) represents a very important area
slope More spread
Black-legged kittiwake And several other members of the NOAA project team and
The northern fulmar exhibited a variable but for marine birds, especially those that breed at localities within
X
Caspian tern sanctuary programs.
‘steady’ population size during the 1980s and the Gulf of the Farallones National Marine Sanctuary (e.g. Point
X
Elegant tern
early 1990s but then began to increase with Reyes, Farallon Islands) (David Ainley, pers. comm.). This area
X
Arctic tern Reviewers of the Draft Bird Report:
arrival of the ‘cold’ regime. The population of is influenced strongly by the San Francisco Bay tidal plume,
More spread
Common murre Sarah Allen, Point Reyes National Seashore, National Park
the common murre remained stable through which provides habitat for many forage fish. This "sanctuary
More spread
Pigeon guillemot Service
most of the study period following a dramatic exclusion area" is also one of the main foraging areas of the
More spread
Tufted puffin Scott Benson, Southwest Fisheries Science Center, NOAA
X decline in 1982 (Ainley and Divoky 2001, Devil’s Slide murre colony, which is in the process of being
Rhinoceros auklet Karin Forney, Southwest Fisheries Science Center, NOAA
More spread Manuwal and Carter 2001), but recently it has restored (David Ainley, pers. comm.).
Cassin’s auklet Doug Forsell, Chesapeake Bay Program, U.S. Fish and Wildlife
More offshore begun to increase (H. Carter, pers. comm.).
Marbled murrelet Service
X These responses to decadal regime shifts With regard to the offshore bounds of the sanctuaries, among
Xantus/Craveri murrelets Laird Henkle, H.T. Harvey and Associates
present challenges to the researchers and the species of marine birds mapped, only 11 had significant David Hyrenbach, Duke University Marine Lab
managers even greater than those offered by concentrations seaward of sanctuary boundaries (see Table
in the abundance of common murre and the three cormorant Hannah Nevins, Moss Landing Marine Laboratories
short-term climate shifts (e.g., ENSO events). It takes several 16). In terms of management, it is important to consider Ashy
species among the three climate-related periods. Jan Roletto, Gulf of the Farallones National Marine Sanctuary
years of monitoring to detect long-term shifts in population storm-petrel and its habitat, because it is listed as a "State Program, NOAA
Breeding species (e.g. common murre, Cassin’s auklet) whose size. Species of Concern". The Xantus’ murrelet, a recently listed Franklin Schwing, Southwest Fisheries Science Center,
species, also deserves consideration, although the proportion
populations are not increased by an influx of visitors from NOAA
colonies outside the study area during warm-water periods, Relevance of Marine Sanctuary Boundaries to Marine of this species’ population that visits the central California And several other members of the NOAA project team and
and the cormorant species, are more dispersed during the Birds. Based on the available data, the boundaries of the National Marine Sanctuaries is relatively low. sanctuary programs.
89
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Ainley, D.G., L.B. Spear and S.G. Allen. 1996b. Temporal and Briggs, K.T., W.B. Tyler, D.B. Lewis, and K.F. Dettman. 1983. Cogswell, H.L.1977. Water birds of California. California natural
PERSONAL COMMUNICATIONS
spatial variation in the diet of the Common Murre in California. Seabirds of Central and Northern California, 1980-1983: Status, History Guide 40, Univ. California Press, Berkeley and Los
Laurence Breaker, Moss Landing Marine Laboratory
Condor, Vol. 98, pp. 691-705. Abundance, and Distribution. Part of Investigator’s Final Report, Angeles, 399 pp.
Harry R. Carter, U.S. Geological Survey, Dixon CA
Marine Mammal and Seabird Study, Central and Northern
Laird Henkel, H.T. Harvey and Associates, San Jose CA
Ainley, D.G., W.J. Sydeman, S.A. Hatch and U.W. Wilson. California, Contract No. 14-12-0001-29090. Prepared by Center Dunning, J.B. 1993. Body Weights of 686 Species of North
Michelle Hester, Oikonos, Bolinas, CA
1994. Seabird population trends along the west coast of for Marine Sciences, University of California, Santa Cruz, for American Birds. Internatl. Wildl. Rehab. Council, Univ. Georgia,
Gerry McChesney, U.S. Fish and Wildlife Service
North America: causes and the extent of regional concordance. the Pacific OCS Region, Minerals Management Service. OCS Athens, GA. 59 pp.
Studies Avian Biol., Vol. 15, pp. 119-133. Study MMS 84-0043. 246 pp.
REFERENCES
Ford, R.G. and J. L. Casey. 2003. CDAS Density Maps for
Ainley, D.G. 1976. The occurrence of seabirds in the coastal
Ainley, D.G., W.J. Sydeman and J. Norton. 1995b. Upper- Briggs, K.T., K.F. Dettman, D.B. Lewis, and W.B. Tyler. 1984. Marine Birds off North/Central California, developed for NOAA’s
region of California. West. Birds, Vol 7, pp. 33-68.
trophic level predators indicate interannual negative and Phalarope feeding in relation to autumn upwelling off California. National Centers for Coastal Ocean Science. Portland, Oregon.
positive anomalies in the California Current food web. Mar. Conservation of Seabirds in Western North America (J.C. 49 pp.
Ainley, D.G. Unpublished data. Shipboard surveys for seabirds
Ecol. Progr. Ser., Vol. 118, pp. 69-79. Bartonek and D.N. Nettleship, Eds.). U.S. Fish Wildl. Serv.,
at sea off northern and central California, 1985-2001. 105 Head-
Wildl. Res. Rept., Vol. 11, pp. 51-62. Glantz, M.R., and J.D. Thompson. 1981. Resource Manage-
lands Court, Sausalito, CA, 94965.
Ainley, D.G., R.R. Veit, S.G. Allen, L.B. Spear and P. Pyle. ment and Environmental Uncertainty. Wiley, New York. 63
1995c. Variations in seabird numbers in the California Current, Briggs, K.T., W.B. Tyler, D.B. Lewis, D.R. Carlson. 1987a. Bird pp.
Ainley, D.G., S.G. Allen and L.B. Spear. 1995a. Offshore
1986-1994. Calif. Coop. Ocean. Fish. Investig., Repts. Vol. 36, communities at sea off California: 1975 to 1983. Studies Avian
occurrence patterns of Marbled Murrelets in central California.
pp. 72-77. Biol., Vol. 11, pp. 1-74. Hare, S.R., and N.J. Mantua. 2000. Empirical evidence for North
Marbled Murrelet Conservation Assessment. Ecology and
Pacific regime shifts in 1977 and 1989. Progr. Oceanogr., Vol.
Conservation of the Marbled Murrelet (C. J. Ralph et al.,
Allen, S.G. 1994. The distribution and abundance of marine Briggs, K. T., D. G. Ainley, D. R. Carlson, D. B. Lewis, W. 47, pp. 103-145.
eds.), Gen. Tech. Rept. PSW-GTR-152, U. S. Dept. Agriculture,
birds and mammals in the Gulf of the Farallones and adjacent B. Tyler, L. B. Spear, and L. A. Ferris. 1987b. Final Report:
Forest Service, Pacific Southwest Res. Station, Albany, CA.
waters, 1985-1992. PhD thesis Univ. California, Berkeley. 222 California Seabird Ecology Study. Prepared by the Institute of Kleinbaum D.G., L.L. Kupper, and K.E. Muller. 1988. Applied
Pp. 361-370.
pp. Marine Sciences, University of California, Santa Cruz, for the regression analysis and other multivariable methods. PWS-
Pacific OCS Region, Minerals Manage. Serv., Contract No. KENT Publishing Co., Boston, MA. 191 pp.
Ainley, D.G. and R.J. Boekelheide (eds.). 1990. Seabirds of
Ballance, L.T. 2001. Unpublished data. Ecology Program, 14-12-001-30183. vii and 153 pp
the Farallon Islands: Ecology, Structure and Dynamics of an
Southwest Fisheries Science Center, 8604 La Jolla Shores Manuwal, D. A., and H. R. Carter. 2001. Natural history of the
Upwelling System Community. Stanford Univ. Press, Palo
Drive, La Jolla, CA 92037. Lisa.Ballance@noaa.gov; http: California Department of Fish and Game, Office of Spill Common Murre (Uria aalge californica). Biology and conserva-
Alto, CA. 425 pp.
//swfsc.nmfs.noaa.gov/PRD/ecology/ecology.html Prevention and Response (CDF&G-OSPR) aerial surveys off tion of the Common Murre in California, Oregon, Washington,
Metadata available from Philbrick et al. 2003. California, 1994-2001. Sacramento, CA. and British Columbia. Vol.1: Natural history and population
Ainley, D.G. and D.F. DeSante. 1980. The Avifauna of the
trends (D.A. Manuwal, H.R. Carter, and T.S. Zimmerman, Eds.)
Farallon Islands, California. Studies Avian Biol., Vol. 4, pp.
Baltz, D.M., and G.V. Morejohn. 1977. Food habits and niche Carter, H. R., G.J.McChesney, D.L. Jaques, C.S. Strong, M.W. U.S. Geol. Surv., Info. Tech. Rep. USGS/BRD/ITR-2000-0012.
1-104.
overlap of seabirds wintering on Monterey Bay. Auk, Vol. 94, Parker, J.E. Takekawa, D.L. Jory, and D.L. -Whitworth. 1992. Washington, D.C. Pp. 1-33.
pp. 526-543. Breeding Populations of Seabirds in California, 1989-1991.
Ainley, D.G. and G.J. Divoky. 2001. Seabirds: Effects of
U.S.Fish Wildl. Serv. Unpubl. Report, Pacific OCS Region, Mantua, N. J. and S.R. Hare. 2002. The Pacific Decadal Oscil-
climate change. Encyclopedia of Ocean Sciences (J. Steele,
Bograd, S.J., P.M. Digiacomo, R.Durazo, T.L.Hayward, Minerals Manage. Serv., Los Angeles CA. lation. Progr Oceanogr Vol. 58, pp. 35-44.
S. Thorpe and K. Tarekian, eds.). Academic Press, London.
K.D.Hyrenbach, J.J.Lynn, A.W.Mantyla, F.B. Schwing,
pp. 2669-2677.
W.J.Sydeman, T.Baumgartner, JB.Lavaniegos, CS.Moore. Carter, H.R., G.J. McChesney, D.L. Jaques, D.L. Jory, McChesney, G.J., H.R. Carter, and M.W. Parker. 2000. Nest-
2000. The State of the California Current, 1999-2000: Forward C.S. Strong, M.W. Parker, J.E. Takekawa, D.L. Jory, and ing of Ashy Storm-Petrels and Cassin’s Auklets in Monterey
D.G. Ainley, D.N. Nettleship, H.C. Carter & A. Storey. 2002.
to a new regime? Calif. Coop. Ocean. Fish. Investig., Repts., D.L. Whitworth. 1992. Breeding Populations of Seabirds in County, California. Western Birds 31: 178-183.
Common Murre (Uria aalge). In "The Birds of North America"
Vol. 41, pp. 26-52. California, 1989-1991. Draft Final Report prepared by U.S.Fish
(A. Poole & F. Gill, Eds.). The American Ornithologists’
and Wildlife Service, Northern Prairie Wildlife Research Center MMS (Minerals Management Service). 2001. Marine Mammal
Union, Washington, D.C. 54 pp.
Bolin, R. L. and D. P. Abbott. 1963. Studies on the marine and San Francisco Bay National Wildlife Refuge Complex, and Seabird Computer Database Analysis System
climate and phytoplankton of the central coastal area of for the Pacific OCS Region, Minerals Management Service, Washington, Oregon and California, 1975-1997 (MMS-CDAS,
Ainley, D.G. and G. A. Sanger. 1979. Trophic relationships of
California, 1954-60. Calif. Coop. Ocean. Fish. Investig., Repts., under Inter-agency Agreement No. 14-12-001-30456 with the version 2.1). Prepared by Ecological Consulting, Inc. (now
seabirds in the northeastern Pacific Ocean and Bering Sea.
Vol. 9: pp. 23-45. USFWS. 155 pp. R.G. Ford and Consulting Co.), Portland, Oregon for the
, Conservation of Seabirds in Western North America, (J.C.
Minerals Management Service, Pacific OCS Region, Order
Bartonek and D. N. Nettleship Eds.), U.S. Fish Wildl. Serv.,
Briggs, K.T., and E.W. Chu. 1987. Trophic relationships and Chu, E.W. 1984. Sooty Shearwaters off California: diet No.1435-01-97-PO-4206.
Wildl. Res. Rept. 11, pp. 95-122.
food requirements of California seabirds: updating models and energy gain. Marine Birds: Their Feeding Ecology and
of trophic impact. Seabirds: Feeding Ecology and Role in Commercial Fisheries Relationships (D.N. Nettleship, G. A. Oedekoven, C.S. D.G. Ainley and L.B. Spear. 2001. Variable
Ainley, D.G., L.B. Spear and S.G. Allen. 1996a. Seasonal and
Marine Ecosystems (J.P. Croxall, Ed.). Cambridge Univ. Press, Sanger and P.F. Springer, Eds.). Canad. Wildl. Serv., Ottawa. responses of seabirds to change in marine climate : California
spatial variation in the diet of Cassin’s Auklet reveals occurrence
Cambridge, UK. Pp. 279-304. pp. 64-71. Current, 1985-1994. Mar. Ecol. Progr. Ser. Vol. 212, pp. 265-
patterns of coastal euphausiids off California. Mar. Ecol. Progr.
281.
Ser., Vol. 137, pp. 1-10.
90
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Philbrick, V.A., P.C. Fiedler, L.T. Ballance, and D.A. Demer. Thayer, J.A., and W.J. Sydeman. 2002. Breeding biology and
2003. Report of ecosystem studies conducted during the 2001 ecology of Rhinoceros Auklets, Cerorhinca monocerata, on
Oregon, California, and Washington (ORCAWALE) marine Año Nuevo Island, California, 2002. Unpublished PRBO report,
mammal survey on the research vessels David Starr Jordan Stinson Beach, CA.
and McArthur. U.S. Department of Commerce. NOAA Technical
Memorandum NOAA-TM-NMFS-SWFSC-349. Available from Thayer, J.A. and W.J. Sydeman. 2002. Population and
http://swfsc.nmfs.noaa.gov/PRD/ecology/ecology.html productivity studies of Año Nuevo seabirds. Unpublished PRBO
report, Stinson Beach, CA.
Pyle, P. and R.P. Henderson. 1991. The birds of Southeast
Farallon Island: Occurrrence and seasonal distribution of mi- Trenberth, K. E. 1997. The definition of El Niño. Bull. Amer.
gratory species. West. Birds Vol. 22, pp. 41-84. Meteorol. Soc. 78: 2771-2777.
Veit, R.R., J.A. McGowan, D.G. Ainley T.R. Wahl and P. Pyle.
Schwing, F. B., Murphree, T., deWitt, L., Green, P. M., 2002. 1997. Apex marine predator declines 90% in association with
The evolution of oceanic and atmospheric anomalies in the changing oceanic climate. Global Change Biol., Vol. 3, pp.
northeast Pacific during the El NiÑo and La NiÑa events of 23-28.
1995-2001. Progress in Oceanography 54, 459-491.
Warzybok, P.M., R.W. Bradley, and W.J. Sydeman. 2002.
Schwing, F.B. and C.S. Moore. 2000. A year without summer Population size and reproductive success of seabirds on
for California, or a harbinger of a climate shift? Trans., Amer. Southeast Farallon Island, 2002. Unpublished Report. PRBO
Geophys. U., Vol. 81(27), p. 301. Conservation Science, Stinson Beach, CA.
Schwing, F.B.,T.L. Hayward, K.M. Sakuma, T. Murphree, A.S. Whitworth D.L., H.R. Carter, R.J. Young, G.J. McChesney,
Mascarenas Jr., S.I. Larios Castillo, A.W. Mantyla, S.L. Cum- M. Hester and S. Allen. 2002. Status and distribution of the
mings, F.P. Chavez, D.G. Ainley and K. Baltz. 1997. The state Ashy Storm-petrel (Oceanodroma homochroa) at Point Reyes
of the California Current, 1996-1997: mixed signals from the National Seashore, California, in 2001. Unpubl. Rpt, Humboldt
tropics. Calif. Coop. Ocean. Fish. Investig., Repts., Vol. 38, State University, Dept Wildlife, Arcata. 15 pp
pp. 22-48.
Seber, G. A. F. 1977. Linear regression analysis. John Wiley
and Sons, New York. 171 pp.
Shannon, C.E. and W.W. Weaver. 1949. The Mathematical
Theory of Communication. Univ. Illinois Press, Urbana. 90
pp.
Spear, L.B. and D.G. Ainley. 1999. Migration routes of Sooty
Shearwaters in the Pacific Ocean. Condor, Vol. 101, pp. 205-
218.
Sydeman, W.J., N. Nur, E.B. McLaren, and G.J. McChesney.
1998. Status and trends of the Ashy Storm-petrel on Southeast
Farallon Island, California, based on capture-recapture
analyses. Condor, Vol. 100, pp. 438-447.
Sydeman, W.J., K.A. Hobson, P. Pyle, and E.B. McLaren. 1997.
Trophic relationships among seabirds in central California:
combined stable isotope and conventional dietary approach.
Condor, Vol. 99, pp. 327-336.
91
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
INTRODUCTION the difficulty in obtaining adequate distributional data sets and in a useable format. Some species of marine mammals are NOAA’s Southwest Fisheries Science Center (SWFSC). These
The California Current passes south through the study area the complexity of combining available marine mammal data infrequently sighted and their distributions are therefore difficult maps are included to illustrate one of the data sets that will
off north/central California, which, along with areas of strong sets, maps for only 13 species were completed in Phase 1, to map. In addition, some data sets were unavailable at the likely be incorporated into the mammal map analysis in Phase
coastal upwelling, makes this area one of the most productive and are included in this document. These maps are referred time of the analysis. Over the past few months, additional data II. SWFSC maps for an additional 13 species are included
ocean systems in the world (Glantz and Thompson, 1981). to as CDAS maps, and represent a compilation of data sets sets have been made available, and these will be added in on the CD-ROM. Plans for Phase II include the following: 1)
Because of this productive environment, the study area from 1980 through 2001, as described below. Also included in Phase II. Improvements and updates are planned for the at-sea acquire the additional available data sets; 2) correct the data
contains a rich fauna of marine mammals, as evidenced in this document and on the CDROM are sighting and effort maps maps, as well as for the haulout and rookery maps, if funding sets for species’ sightability, detectability and other factors;
marine mammal abundance and richness. for 16 marine mammals from a single data set (the marine is made available. and 3) develop composite maps for about 23 species, singly
mammal stock assessment surveys from NOAA's Southwest or in combination.
In addition to many marine mammal species that live here year- Fisheries Science Center (SWFSC)). These data will likely be Table 22 is a list of marine mammal species that were
round and use the region’s coasts and islands for breeding and incorporated into Phase II of the analysis. considered for this analysis; density maps were developed About the Literature and Survey Data Used in this
hauling out, the community of seasonal residents and migrants for eight species in MMS-CDAS (MMS, 2001) a data display Assessment. This assessment is based on the efforts of
is even more robust. Central California is the destination for Additional data compilation, data analysis and expert review system developed by R.G. Ford Consulting Co. for the Minerals individual researchers to study marine mammal spatial and
many marine mammal species seeking productive feeding are needed to complete the final analyses of marine mammals Management Service. Sightings maps were developed in CDAS temporal patterns, federal and state government efforts to
areas and acceptable habitat in which to spend their occurrences in the study area. More complete results will be for an additional four species. Maps of sea otter counts and assess stock size and the potential biological impacts of oil
nonbreeding periods, providing evidence of the region’s presented in a final report in Phase II of this project. Additional pinniped haulouts and rookeries were also developed. Also development, and state government efforts to respond to oil
trophic richness. Over 29 species of marine mammals occur products planned for Phase II are listed at the end of this included in this document are maps for three species from spills.
in the study area off north/central California; over 22 cetaceans section.
(whales, dolphins, and porpoises), six pinnipeds (seals and
Table 22. Marine mammal species included in this assessment and map types developed for them (Phase I and Phase II)
sea lions), and one fissiped species (the sea otter). DATA AND ANALYSES
Overview of Map Development and No. of Draft No. of No. of Maps
CDAS Maps, SWFSC Planned for
The objectives of this assessment were to: 1) identify spatial Analysis Process To Date. The methods
Common Name Scientific Name Order/Suborder Family Phase l Maps, Phase l Phase II
and temporal distributions and patterns for marine mammals used in each survey were different, and
Fissiped
that occur in ocean waters off north/central California between because of this, careful consideration and
Southern sea otter Enhydra lutris nereis Carnivora/(none) Mustelidae 1 1
Point Arena (38.91ºN) and Point Sal (34.90ºN); 2) identify correction are required to merge the data
Pinnipeds
important areas and time periods associated with higher sets in a meaningful and scientifically
California sea lion Zalophus californianus Carnivora/PinnipediaOtariidae 2 1
concentrations of these species; and 3) identify important data acceptable way. Data preparation for the
Steller sea lion Eumetopias jubatus Carnivora/PinnipediaOtariidae 1 1
and information gaps observed as a result of this analysis. mammal analyses included the following
Northern fur seal Callorhinus ursinus Carnivora/PinnipediaOtariidae 1 1
In this analysis, ‘important’ season or area refers to those steps: species and study area selection,
Harbor seal Phoca vitulina richardsi Carnivora/PinnipediaPhocidae 1 1
having relatively higher concentrations of a particular species; data set identification and collection, data
Northern elephant seal Mirounga angustirostris Carnivora/PinnipediaPhocidae 1 1
in Phase II diversity may also be considered. corrections, data conversion into common
Cetaceans
units, organizing the data into 10’x10’ cells
Dall's porpoise Phocoenoides dalli Cetacea/Odontoceti Phocoenidae 1 1 1
Preliminary Results. Summarized below are the spatial or leaving them as sightings and effort, and
Harbor porpoise (stocks: Northern CA, San
and temporal occurrence patterns of data for 13 species that map development. For species present in
Francisco/Russian River, Monterey Bay) Phocoena phocoena Cetacea/Odontoceti Phocoenidae 1 1?
regularly occur in marine waters off north/central California. The sufficient numbers, seasonal density maps
Lagenorhynchus obliquidens
Pacific white-sided dolphin Cetacea/Odontoceti Delphinidae 1 1 1
results of this marine mammal assessment are preliminary were developed, and for infrequently sighted
Risso's dolphin Grampus griseus Cetacea/Odontoceti Delphinidae 1 1 1
(Phase I) and feature highlights of work in progress. Due to species, sighting and effort maps were
Bottlenose dolphin (California coastal stock) Tursiops truncatus Cetacea/Odontoceti Delphinidae 1 1?
developed. CDAS maps were created for
Short-beaked common dolphin Delphinus delphis Cetacea/Odontoceti Delphinidae 1 1
13 species. The original draft maps were
Northern right whale dolphin Lissodelphis borealis Cetacea/Odontoceti Delphinidae 1 1 1
reviewed at an expert workshop in October
Killer whale Orcinus orca Cetacea/Odontoceti Delphinidae 1 1?
2002; there it was determined that additional
Baird's beaked whale Berardius bairdii Cetacea/Odontoceti Ziphiidae 1 1?
data, corrections and analyses were required
Cuvier's beaked whale Ziphius cavirostris Cetacea/Odontoceti Ziphiidae 1 1?
to improve the mammal maps; this work will
Beaked whales (Mesoplodonts) Mesoplodon spp. Cetacea/Odontoceti Ziphiidae 1 1?
be done in Phase II of this project. Some
Sperm whale Physeter macrocephalus Cetacea/Odontoceti Physeteridae 1 1?
revisions have been made to the maps and
Blue whale Balaenoptera musculus Cetacea/Mysticeti Balaenopteridae 1 1 1
text of this document since its draft release
in April 2003. Humpback whale Megaptera novaeangliae Cetacea/Mysticeti Balaenopteridae 1 1 1
Fin whale Balaenoptera physalus Cetacea/Mysticeti Balaenopteridae 1 1?
Species Selected for Analysis. Selection Minke whale Balaenoptera acutorostrata Cetacea/Mysticeti Balaenopteridae 1 1?
criteria for marine mammal species included Gray whale Eschrichtius robustus Cetacea/Mysticeti Eschrichtiidae 1 1
in this assessment were: 1) the species MAP TOTALS 14 16 14-23
distribution includes the study area, and 2)
survey data for the species was available
92
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Table 23. A Summary of At-Sea Data Sets Used in the Preliminary Marine Mammal Assessment
Because wind speed affects detection of marine mammals, data
Table 23. Summary of at-sea data sets used in the preliminary marine mammal analyses.
collected when wind speed exceeded 25 kt were excluded. Data
Vessel Name Ocean Total Transect Total Transect
Habitat were allocated into 10’ x 10’ cells (i.e., 10-minute latitude by
Principal & Platform Seasons Width: Width:
10-minute longitude cells). All aerial data were continuous; each
Covered2
Data Set Investigator Height Years Covered Pinnipeds Cetaceans
ship-based data set was converted separately into a continuous
Surface survey harbor porpoise,
transect format to the extent possible. The continuous aerial
MMS High- of the shelf, 254m; great
data were binned into the appropriate cell. For the SF-DODS
Altitude Aerial Pembroke, slope & deep All three whales, 1130m;
Surveys Dohl 270m ocean beyond 1980-1983 seasons N/A all others, 885m and EPOCS studies, and the Rockfish Assessment cruises prior
Surface survey to 1997, the beginning position, ship heading, and speed were
MMS Low- of the shelf, used to compute the end position of each 2-4 km continuous
Altitude Aerial slope & deep All three
transect. From this, a midpoint of the transect was determined.
Surveys Bonnell Pembroke, 62m ocean beyond 1980-1983 seasons 109m 109m
As times of observations were not available, the position of
Surveyor, 12m, the midpoint was used to select the cell to which the survey
EPOCS Discoverer, Surface survey
effort was assigned. If this midpoint fell on a cell boundary, it
Shipboard Oceano- of the deep All three
was assigned to the cell to the north or west. To maintain the
Surveys Ainley grapher, 15m ocean 1984-1994 seasons 300-600m 800m
correspondence between effort and mammal observations,
CA Seabird
Davidson Current, because ocean conditions differ distinctly
observations were also assigned to the transect midpoints.
Ecology Low- Surface survey
among them and are known to affect the biota of the California
For the Rockfish Assessment Cruises from 1997 onward, effort
Altitude Aerial Partenavia, of shelf and Mainly
Surveys Briggs 62m slope 1985 Upwelling 50m 50m Current (e.g. Ainley 1976, Briggs et al. 1987). As there is
was assigned to the cells through which the vessel passed
NMFS significant interannual variation in the actual duration of these
based on the proportion of trackline that fell within each cell, and
Midwater seasons, the following dates were ‘defined’ for each season for
observations were interpolated along the cruise track according
Trawls for Juv. Surface survey
purposes of analysis: Upwelling Season is 15 March-14 August,
to the time of each observation.
Rockfish: David Starr of shelf and Mainly
Oceanic Season is 15 August-14 November, and Davidson
Ship Surveys Ainley Jordan,10m slope to 3000 m 1985-2001 Upwelling 300m 800m
Current Season is 15 November-14 March.
Data Analysis.
OSPR Low Surface survey
Effort Summary. For all surveys, 132,521 kilometers of trackline
Altitude Aerial Partenavia, of shelf and 1994-1998, All three
As evident in Table 24, the Upwelling Season had the greatest
(pinnipeds and cetaceans) and 78,486 kilometers of additional
Surveys Bonnell 62m slope 2001 seasons 50m 50m
MMS Santa amount of survey effort, followed by the Davidson Current
trackline (cetaceans only) were analyzed (Table 24). A total of
Barbara Season. The Oceanic Season had the lowest effort. Unlike
3,459 observations of 7,039 pinnipeds and 2,313 observations
Channel Low Surface survey the other seasons, the Oceanic Season had no data from
of 69,286 cetaceans were included in analyzed data. Survey
Altitude Aerial Partenavia, of shelf and All three
the 1980s. Because of the variation in effort coverage across
effort used in this assessment for pinnipeds and cetaceans are
Surveys Bonnell 62m slope 1995-1997 seasons 50m 50m
space and time (and methods, as well as many other factors),
summarized as maps in Figure 61.
SF-DODS Surface survey interpretation of the data requires careful consideration.
Shipboard of shelf and All three
Organizing Data into Ocean Seasons. Effort and species
Surveys Ainley Point Sur, 8m slope to 3000 m 1996-2000 seasons 300m 800m
data were organized and mapped into three distinct ocean
Notes
seasons (Bolin and Abott 1963): Upwelling, Oceanic, and
See additional description of these data sets on the CD for more information on the CDAS data sets.
Data from the marine mammal stock assessment of NOAA's SWFSC were not included in the preliminary CDAS asssessment.
Table 24. Summary of combined data set effort at sea for mammals, by ocean season.
Number
The Data Sets. The ship and aerial strip transect data used in
Related Literature. The at-sea distribution and abundance of
Dates Used for Number Kilometers of 10'
this assessment were collected from 1980-2001 and ranged
marine mammals within the study area has been described
Ocean Each Ocean of Years of Trackline Number Cells
from Point Arena south to Point Sal, and offshore to the extent
in many publications, some of which include the following:
Season Season Months Included Taxa Surveyed of Visits Sampled
of data availability. Estuaries were not part of the GIS study
Bonnell et al. (1983), Dohl et al. (1983), Calambokidis et
Pinnipeds: 63,262 10,902 283
1980-1982,
area, but coastal haulouts and rookeries, when available, were
al (1988, 1990, 1996), and Allen (1994). Numerous marine
Upwelling 15 Mar-14 Aug 5 1985-2001 Cetaceans:
mapped to provide a more complete view of important areas 96,978 15,280 317
mammal stock assessment studies have been conducted by
for pinniped species. See Table 23 for additional information
the Southwest Fisheries Science Center, LaJolla, CA (NMFS/ Pinnipeds: 30,443 4270 263
1980-1982,
on the data.
SWFSC ship surveys): Barlow (1988, 1995), Barlow and Forney 1991, 1994-
(1994), Barlow and Gerrodette (1996), and Forney and Barlow
Oceanic 15 Aug-14 Nov 3 2001 Cetaceans: 49,981 6821 322
Data Synthesis.
(1998). A few ecosystem studies of marine mammals in this
Pinnipeds:
Davidson 1980-1986, 38,816 5594 360
Summarizing Transect Data into Grid Cells for CDAS Maps. The
region have also been conducted by Schoenherr (1991), Black
Current 15 Nov-14 Mar 4 1991-2001 Cetaceans: 64,048 8897 383
above data sets were processed to compensate and correct
(1994), Kieckhefer (1992), Croll et al. (1998), Forney and
Pinnipeds: 132,521 20,766 395
for differences in survey methodology, including platform type
Barlow 1998, Forney 2000, Benson et al. (2002) and Keiper
TOTAL 1 Jan-31 Dec 12 1980-2001 Cetaceans:
(ship or aerial) and transect width, among the various studies.
et al. (In Review). 211,007 30,998 416
93
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Calculating Density and Developing Seasonal Density CDAS will be added and an overall rookery and haulout map for the
Combined At-Sea Effort for Marine Mammal Analysis Maps. From the digitized survey data, we mapped the distribution pinnipeds will be developed.
of effort and of species observations into a grid of 10’ by 10’
cells, using the MMS-CDAS mapping system (MMS, 2001). ANALYTICAL MAP PRODUCTS
129°W 128°W 127°W 126°W 125°W 124°W 123°W 122°W 121°W
The larger cell size was determined to be more meaningful by A series of over 50 preliminary maps (41 CDAS maps and 12
200 m
Pinniped Effort
20
0
experts at a preliminary map/data review session. SWFSC maps) and related results are presented in this section;
0
m
39°N
39°N
additional SWFSC maps for 13 mammal species/species
Effort
The species data were first transformed into densities on the groups are included on the CD-ROM. These preliminary maps
(Km of trackline)
basis of strip widths (which varied by ship or aerial platform, will be finalized and included in a Phase II report.
> 1500
depending on speed and height above water; see Table 23).
1000.01 - 1500.00
500.01 - 1000.00
The number of individuals of each species seen was then
250.01 - 500.00
38°N
38°N
divided by area surveyed to estimate density in each cell for
100.01 - 250.00
50.01 - 100.00
that data set. For construction of density plots, if a cell was
25.01 - 50.00
5.01 - 25.00
censused in other years or the same year by another survey,
0.01 - 5.00
densities in cells were averaged and weighted according to
effort. These maps display observed densities; in Phase II these
0 25 50 Km
37°N
37°N
densities will be corrected to account for additional factors such
as sightability.
Seasonal High Use Areas for Individual Species. The purpose
36°N
36°N
of the seasonal high use maps is to provide a summary map
for a species' spatial patterns. These maps were developed for
mammal species with density data, with the seasonal density
data binned into 10’ x 10’ cells for each species or species
group. Non-zero cells were then ranked and those in the top
a
35°N
35°N
20 percent were selected and defined as seasonal high use
areas. Cells were then mapped with color corresponding to the
number of ocean seasons of high use. The index is therefore
200 m
20
Cetacean Effort
0
sensitive to cells that were not sampled in any one of the three
0m
39°N
39°N
seasons, causing a downward bias in the index. Use of a 10'x10'
Effort cell size greatly reduces the magnitude of this bias.
(Km of trackline)
> 1500
Cells in which there was effort but animals were not observed,
1000.01 - 1500.00
and cells where sightings occurred but were never high use
500.01 - 1000.00
38°N
38°N
250.01 - 500.00
areas, were also provided.
100.01 - 250.00
50.01 - 100.00
25.01 - 50.00
Developing Sighting and Effort CDAS Maps for Infrequently
5.01 - 25.00
0.01 - 5.00
Sighted Species. Where sightings were too few to warrant
seasonal density maps, observations were mapped as point
0 25 50 Km
37°N
37°N
locations. For context, overall survey effort is also presented.
This display method was chosen in response to comments by
expert reviewers at the October 2002 workshop and in view of
the low numbers of sightings of certain species.
36°N
36°N
Preliminary Rookeries and Haulouts by Species. Pinniped
rookeries and haulouts are monitored and surveyed by a variety
of institutions and individuals. Recent data (varying by species,
but generally from 1998-2002) were used to represent locations
b
35°N
35°N
of rookeries and haulout sites for five pinniped species. In
Phase II, additional information on harbor seal pupping sites
129°W 128°W 127°W 126°W 125°W 124°W 123°W 122°W 121°W
Source Data: See text.
Figure 61. Total at-sea survey effort for marine mammal analyses.
94
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THESE MAPS from Half Moon Bay to Goleta, just south of Point Conception
DRAFT Maps 62a and b display the locations of groups of southern sea with uncommon sightings of animals beyond these areas (pers.
Southern sea otter Enhydra lutris nereis otters during the Fall 2001 and Spring 2002 rangewide counts. comm. B. Hatfield); the distribution of otters along the south
123°W 122°W 121°W
123°W 122°W 121°W
Maps 62c and d summarize these rangewide count data into end has been highly variable since the expansion of the sea
coastal strips approximately 10km in length, in order to display
Rangewide Count Rangewide Count otter range south of Point Conception (pers.comm. M.Harris).
20 20
0 0
m m
20 20
linear densities along the shore. The northern extent of the data In the study area, sightings have occurred as far north as Point
0
Fall 2001 Spring 2002
0
0 0
m m
is south of Half Moon Bay; sea otters are also present to the Reyes (Point Reyes Headlands, Double Point, Duxbury Reef;
south of the mapped area. not shown on map; pers.comm. S. Allen). Sea otters occur
Number of
along rocky shorelines with kelp beds (but also in open water
Sea Otters
DATA SOURCES habitats, sandy/soft bottom areas, and tidal estuaries) and in
16 - 28
Data were collected by wildlife biologists from the U.S. Depart-
37°N
37°N
depths of water about 20-40 m (some to 60 m, and rarely to
11 - 15
ment of the Interior (currently the USGS Biological Resources 100 m; M. Kenner pers. comm).
6 - 10
Division), California Department of Fish and Game, the Mon-
2-5
terey Bay Aquarium, and trained volunteers during semi-an- Overall, numbers of otters per segment were greater in the
1
nual rangewide counts in Fall 2001 and Spring 2002. The Fall southern portion of the Monterey Bay National Marine Sanc-
0 25 50 Km
2001 count was conducted during the period 4-20 November tuary. In the census of Fall 2001 (map a), greater numbers of
2001, the Spring 2002 count was conducted during the period otters per segment occurred along the Carmel coast and from
5-22 May 2002. The data set was provided by Mike Kenner, Piedras Blancas south to Point San Luis. Seasonal changes
36°N
36°N
UCSC but is sourced to USGS; contact Brian Hatfield for more in abundance and distribution of sea otters are believed to
information. be affected by male movements during the period when most
breeding occurs (June/July through October/November) when
METHODS they move from the periphery of the range toward the center of
The original data were entered from hand marked maps into a the range in search of estrous females (Bonnell et al., 1983).
custom designed digitizing program which assigned coordinates From December to April, many males migrate to the range pe-
to each observed sea otter group. Positions of animals toward
35°N
35°N
ripheries, perhaps in search of more abundant prey (M.Harris
the ends of the range and in Elkhorn Slough were not assigned
a b pers. comm.). However, this is not evident in the maps. Sea-
coordinates by this program. Each group was also assigned to sonal changes also are affected by factors such as weather,
38°N
an ATOS (As The Otter Swims) number, which are numbers sea conditions and abundance of kelp canopy (see Reidman
Linear Density Linear Density
20 20
approximately 0.5 km apart along a smoothed 5 fathom contour
0 0
m m
20
and Estes, 1990).
20
0
Rangewide Count Rangewide Count
0
0 0
line along the coast from Golden Gate to approximately Santa
m m
Fall 2001 Spring 2002 Barbara. These numbers were used to get approximate posi- From 1983 until the mid 1990’s, trends in spring southern sea
tions for otters without assigned coordinates. otter counts indicated sea otters increased steadily; in the mid-
Otters per to late 1990’s, sea otter numbers declined (USFWS, 2000)
A series of coastal segments approximately 2 km in width was
Segment and have since remained relatively constant (pers.comm. B.
37°N
37°N
created for display purposes. Each segment was approximately
151 - 200 Hatfield). Sea otter count data is used as an index to assess
10 km in length; divisions were based on the ATOS numbers
101 - 150 trends in the population dynamics, not as a population estimate
described above. Twenty ATOS numbers approximately 500m
51 - 100
(pers.comm. M.Harris). The 2002 spring count was 1% below
apart were included in each segment. The coordinates of each
26 - 50
the 2001 count, from 2161 otters in 2001 to 2139 in 2002. The
otter group were used to place it within a particular segment,
11 - 25
2001 count was 6.7% below counts from the previous year
1 - 10 and the otters in each segment were summed. This provides
(USGS 2002). Due to its small population size, the southern sea
0 an estimate of linear density (otters per segment or otters
otter population is especially vulnerable to human disturbance,
36°N
36°N
per 10km) since the segments were approximately 10km in
competition with fisheries, and pollution, including the threat
length.
of a major oil spill. The lack of population growth and recent
decline coincides with an increase in mortality (e.g., infectious
RESULTS AND DISCUSSION
diseases, white shark attacks) as indicated by the number of
The southern sea otter (Enhydra lutris nereis) is one of three
beach-cast sea otter carcasses (Estes et al., 2003). Otters near
subspecies: southern (E.l.nereis), northern (E.l.kenyoni), and
heavy freshwater flows are three times more likely to have been
Russian (E.l.lutris). The southern sea otter is listed as threat-
infected by Toxoplasma gondii, a protozoan parasite caused by
35°N
35°N
ened under the Federal Endangered Species Act, and depleted
c d parasite eggs in cat droppings (see Miller et al., 2002).
under the Marine Mammal Protection Act (MMPA). Under Cali-
fornia Fish and Game Code, the southern sea otter is listed as
123°W 122°W 121°W 123°W 122°W 121°W
Southern sea otters are key predators of benthic species (e.g.
a “fully protected” species. The southern sea otter generally
Source Data: See text.
sea urchins, sea stars, mussels, clams, abalone, crabs) and
inhabits the near-shore waters of the central California coast,
Figure 62. Maps for southern sea otter: rangewide count and linear density, fall 2001 and spring 2002. octopus (see Riedman and Estes, 1990).
95
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
124°W 123°W 122°W 121°W
ABOUT THIS MAP California sea lions feed on a diversity of fish (e.g., Pacific hake,
200 m
Figure 63 summarizes information on California sea lion
20
northern anchovy, Pacific sardine, herring, rockfish, salmon,
California sea lion Zalophus californianus
39°N
39°N
0
0
haulouts and rookeries in the study area based on number of steelhead) and invertebrates (e.g., squid and octopus) (Weise,
m
animals, frequency of use, and rookery status. 2000; see also Riedman 1990).
Fish Rocks
Haulout Sites DATA
Haulout data and rookery information are from aerial counts
Haulout Occupancy
(July, 1998-2002) provided by Mark Lowry of the NMFS’
1998 - 2001
Southwest Fisheries Science Center.
Frequent (4 years)
Bodega Rock
Occasional (3 years)
METHODS
Infrequent (1-2 years)
Haulout locations were mapped using coordinates included
Point Reyes Minor Rookery,
38°N
38°N
in the files from Mark Lowry, SWFSC. Data from July counts
1998-2001
Sea Lion Cove
Occasional Minor in four years, 1998-2002, were used to calculate frequency of
Rookery
Pier 39 use for each haulout location and mean number of animals
North Farallon Islands (e.g. El Niño years)
using each location when that location was occupied. Rookery
Number of Sea Lions
Southeast Farallon Island
status was determined by the inclusion of pups in the counts.
Mean of July Counts*: Pups were observed in all years at two sites, while three sites
1998-2001
had pups only in 1998, an El Niño year.
2501 - 6706
RESULTS AND DISCUSSION
501 - 2500
Año Nuevo Island
Haulout sites for the California sea lion are located along the
37°N
37°N
101 - 500
coast from Fish Rocks (just south of Point Arena) to the south,
at Point Sal Rock, and inside San Francisco Bay (Pier 39).
51 - 100
Minor rookeries are located on the Southeast Farallon Island
26 - 50
Monterey Breakwater 1 - 25
and Año Nuevo Island.
0 25 50 Km
Sea Lion Rocks
Periods of unusually warm ocean waters associated with El
Lobos Rocks
Nino oceanographic conditions affect pup production (i.e., fewer
pups are born) and result in higher mortality rates for pups and
Unnamed; N of Partington Pt juveniles. During the El Niño periods of 1983, 1992, and 1998,
pup production decreased by 35, 27, and 64%, respectively at
36°N
36°N
rookeries in southern California (SWFSC 2001).
San Martin Rocks
Similar to at-sea occurrence patterns, haulout patterns and
rookery locations also change during warmer water periods. For
Point Piedras Blancas
example, rookery locations during the strong El Niño of 1998
White Rock
(shown on the map) included the rookeries at the Farallones
and Año Nuevo as well as additional rookeries located near
Partington Point, and Lion, Pecho, and Pup Rocks located to
Lion Rock
Pecho Rock the south of the Monterey Bay National Marine Sanctuary. Haul-
out patterns at the Farallon Islands and Point Reyes National
35°N
35°N
Seashore also changed, indicated by an influx of immatures
(Sydeman and Allen, 1999; S. Allen pers.comm.).
Point Sal Rock
DRAFT Greater numbers of California sea lions in the study area during
El Niño events likely reflected a greater than usual northward
migration in response to a reduction of food resources near
* When occupied.
southern breeding grounds.
124°W 123°W 122°W 121°W
Source Data: See text.
Figure 63. Map for California sea lion: haulouts and rookeries.
96
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THESE MAPS For the Oceanic Season, data are from 1980-1982, 1991, and
DRAFT
California sea lion Figures 64a, b, and c show the at-sea density (animals/km2) of 1994-2001. For the Davidson Current Season, data are from
Zalophus californianus California sea lions in the Upwelling, Oceanic, and Davidson 1980-1986 and 1991-2001. The rookery and haulout counts
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Current seasons, displayed in 10’x10’ cells. Densities are are shown as a general range, based on counts of all animals
Upwelling Season Oceanic Season
200 m
200 m
20
20
based on combined data of several studies; see “Methods” (pups and adults) in three years, 1999-2001.
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) and “Data Sources” sections. The color and mapping intervals
39°N
39°N
were customized to show the most structure and to highlight METHODS
Density significant areas, while allowing comparisons among marine At-sea densities are the result of a synthesis of data from
(Animals/km²)
mammal species. Cells that were surveyed but which had seven shipboard and aerial survey programs conducted in
10.01 - 50.00
no California sea lions have a density of zero; unsurveyed the study area in the years 1980-2001 (see “Data Sources”
5.01 - 10.00
38°N
38°N
areas appear white. Blue lines indicate the National Marine section). Pinniped observation data and trackline data from
1.01 - 5.00
Sanctuary boundaries of Cordell Bank, Gulf of the Farallones, these studies were converted to a common format. All aerial
0.51 - 1.00
and Monterey Bay; bathymetric contours for the 200 m and data were continuous; ship-based data were converted
0.11 - 0.50
2,000 m isobaths are also shown in blue. separately into a continuous transect to the extent possible.
0.06 - 0.10
0.01 - 0.05 From the digitized survey data, the distributions of effort and of
37°N
37°N
0.00 In order to provide one map for the species that integrates the species were mapped into 10’x10’ cells using CDAS, a custom
patterns of its spatial and temporal occurrence in the study geographic information system for analyzing marine bird and
0 25 50 Km
area, map d shows seasonal high use areas, displayed in mammal surveys (MMS, 2001). The length and width of the
10’x10’ cells. This map provides a further synthesis of densities survey trackline in a given cell (estimated trackline width varied
presented in maps a, b and c (see “Methods” section for details), by platform, depending on speed and height above water) were
36°N
36°N
and portrays the relative importance of various areas to the used to estimate the area sampled. The number of cetaceans
species. Areas with consistent high use are highlighted on this of each species seen in a cell was then divided by the area
map. To provide a relative reference for the “high use” areas, sampled in the cell to estimate density. If a cell was censused
cells are also shown where the species were absent (i.e., the more than once, densities were averaged, with adjustment
cell was sampled but the species was not recorded there), or made for effort.
35°N
35°N
a b present but at lesser concentrations in any particular season.
Note that these maps represent either sighting locations or
200 m
Davidson Current Season
200 m
Seasonal High Use Areas
20
20
DATA SOURCES densities that used survey strip widths relative to each survey
0
0
0m
0m
(Nov. 15 - Mar. 14) and Rookeries At-sea densities for the California sea lion are based on data platform (e.g., plane, ship); density was calculated on the basis
39°N
39°N
from seven survey programs conducted in 1980-2001. These of the number of animals sighted and area surveyed. The data
Persistence of data were combined using CDAS software into the MMS-CDAS have only been corrected to normalize for survey effort and to
High Use
data system (MMS, 2001), developed for Minerals Management exclude observations with winds greater than 25 knots (smaller
3 Seasons
Service and expanded for this project. Of the data sets on the or less obvious species are often less detectable even at wind
2 Seasons
1 Season
original CD-ROM, four aerial survey data sets contained data speeds of less than 25 knots). Additional corrections are
38°N
38°N
Sea lions present
in the study area from Point Arena to Point Sal. Of these, the planned for Phase 2 of this project and are briefly discussed
Sea lions absent
OSPR survey program was still ongoing and data from recent below.
years were added to this data set. In addition, data from three
ship-based survey programs were converted to a compatible For example, no adjustments or corrections have been made to
37°N
37°N
format for analysis. See section text for details on individual account for differences in marine mammal detectability among
data sets. species and differential probability of detecting animals from
aerial and shipboard platforms. Individual body size, group
Data sources for aerial at-sea data include MMS-CDAS (MMS, size, and species-specific behaviors, such as proportion of time
2001) and California Department of Fish and Game Office of spent submerged, are all factors known to affect detection and
36°N
36°N
Spill Prevention and Response (CDF&G-OSPR), unpublished hence, observed distribution and density estimates as well.
data. Early data were collected using methods described by Because of the very different attributes of aerial and shipboard
Bonnell et al. (1983); more recent data were collected using platforms, these factors, and the associated adjustments for
updated technology but with the same general method. Data observations, vary among the studies.
sources for ship-based survey data include David Ainley,
35°N
35°N
See additional California sea lion map
c d unpublished data (see Oedekoven et al., 2001 for details Map d was developed using the same approach as for maps a,
for haulout and rookery locations.
on methods). Although the at-sea data span the years 1980- b and c. For each season, the cells with densities in the top 20%
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
2001, data are not available for all seasons in all years. For the of non-zero values were designated “high use” for that season.
Source Data: See text.
Figure 64. Maps for California sea lion: seasonal at-sea densities, high use areas, and rookeries. Upwelling Season, data are from 1980-1982 and 1985-2001.
97
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Cells were scored for “high use” in one, two, or three seasons
and are depicted by color. To provide a relative reference for
the “high use” areas, cells are also shown where the species
were absent (i.e., the cell was sampled but the species was not
recorded there) or present (but densities were never in the top
20% for any season). Further detail on methods is provided in
the Data and Analysis section.
RESULTS AND DISCUSSION
The California sea lion (Z.c.californianus) is subdivided into
three stocks (U.S., Western Baja California, and Gulf of
California); the United States stock begins at the U.S. Mexico
border and extends northward into Canada (Carretta et al.
2001). The breeding areas are on islands located in southern
California, western Baja California, and the Gulf of California.
In the study area, a small number of pups are born on Año
Nuevo Island and Southeast Farallon Island; otherwise the
central California population is composed of non-breeders.
Adult females and immatures remain near the rookeries year-
round, whereas adult males (along with most immatures)
migrate northward to feeding areas from central California to
British Columbia.
In the study area, this species was the most abundant of the
pinnipeds (at-sea sightings: n=1,497 individuals: n=4,411) and
was widely distributed throughout the shelf and upper slope
regions of the three national marine sanctuaries. The seasonal
abundance of California sea lions off central California is linked
to spring and fall pre- and post-breeding migrations. Densities
were greatest during the Oceanic Season (just after breeding)
and Davidson Current Season (before the next breeding period)
and somewhat lower during the Upwelling Season (breeding
period).
Periods of unusually warm ocean waters associated with El
Niño oceanographic conditions affect pup production (i.e., fewer
pups are born) and result in higher mortality rates for pups and
juveniles. During the 1983, 1992, and 1998 El Niño events,
pup production decreased by 35, 27, and 64%, respectively,
at rookeries in southern California (SWFCS 2001). At-sea
distribution patterns were also altered; greater numbers of sea
lions were sighted off central California during these warmer
periods (see also Bonnell & Ford 1987, Trillmich & Ono 1991,
Allen 1994, Keiper 2001, and Keiper et al. In Review.).
98
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THIS MAP from eight shipboard and aerial survey programs conducted in
124°W 123°W 122°W 121°W
Figure 65 shows individual sightings of Steller sea lions at sea the study area in the years 1980-2001 (see “Data Sources”).
200 m
20
39°N
39°N
Steller sea lion Eumetopias jubatus
0
along with the locations of haulouts, rookeries, and at-sea effort Cetacean observation data and trackline data from these stud-
0m
in the study area. At-sea observations are based on combined ies were converted to a common format. All aerial data were
Fish Rocks
data of several studies (see “Methods” and “Data Sources” continuous; ship-based data were converted separately into a
sections). For context, the amount of combined survey effort continuous transect to the extent possible. From the digitized
(km of trackline) is also shown, summarized in 10’x10’ cells. survey data, effort was mapped into 10’x10’ cells using CDAS,
Northwest Cape Rocks
Haulout locations are based on counts conducted in July 2000. a custom geographic information system for analyzing marine
Number of Sea Lions
Blue lines indicate the National Marine Sanctuary boundaries bird and mammal surveys (MMS, 2001). The length and width
At Sea
Bodega Rock
of Cordell Bank, Gulf of the Farallones, and Monterey Bay; of the survey trackline in a given cell (estimated trackline width
Haulouts
Sightings
bathymetric contours for the 200 m and 2,000 m isobaths are varied by platform, depending on speed and height above wa-
3 16 - 36
2 also shown in blue.
3 - 15 ter) were used to estimate the area sampled.
38°N
38°N
1 1-2
Point Reyes
Rookeries DATA SOURCES Note that the these maps represent either sighting locations or
North Farallon Islands 450 - 502 At-sea sightings for the Steller sea lion are based on data from densities that used survey strip widths relative to each survey
seven survey programs conducted in 1980-2001. These data platform (e.g., plane, ship); density was calculated on the basis
50-215
were combined using CDAS software into the MMS-CDAS data of the number of animals sighted and area surveyed. The data
South Farallon Island
Survey Effort
system (MMS, 2001), developed for Minerals Management have only been corrected to normalize for survey effort and
(km of trackline)
Service and expanded for this project. Of the data sets on the to exclude observations with winds greater than 25 knots;
> 3000.00
original CD-ROM, four aerial survey data sets contained data additional corrections are planned for Phase 2 of this project
1500.01 - 3000.00
1000.01 - 1500.00 in the study area from Point Arena to Point Sal. Of these, the and are briefly discussed below.
Año Nuevo Island
500.01 - 1000.00 OSPR survey program was still ongoing and data from recent
100.01 - 500.00
37°N
37°N
years were added to this data set. In addition, data from three For example, no adjustments/corrections have been made to
0.01 - 100.00 ship-based survey programs were converted to a compatible account for differences in marine mammal detectability among
format for analysis. See section overview for details on indi- species and differential probability of detecting animals from
0 25 50 Km
vidual data sets. aerial and shipboard platforms. Individual body size, group
size, and species-specific behaviors, such as proportion of time
Sea Lion Rocks (Pt. Lobos)
Data sources for aerial at-sea data include MMS-CDAS (MMS, spent submerged, are all factors known to affect detection and
Lobos Rocks
2001) and California Department of Fish and Game Office of hence, observed distribution and density estimates as well.
Spill Prevention and Response (CDFG-OSPR), unpublished Because of the very different attributes of aerial and shipboard
data. Early data were collected using methods described by platforms, these factors,and the associated adjustments for
Bonnell et al. (1983); more recent data were collected using observations, vary among the studies.
updated technology but with the same general method. Data
36°N
36°N
sources for ship-based survey data include David Ainley, unpub-
Cape San Martin The data in these maps include wind conditions of up to 25
lished data (see Oedekoven et al., 2001 for details on methods). knots; smaller or less obvious species are often less detectable
Although the at-sea data span the years 1980-2001, data are even at wind speeds of less than 25 knots. The seasonal maps
not available for all seasons in all years. For the Upwelling Sea- contain different combinations of shipboard and aerial data;
son, data are from 1980-1982 and 1985-2001. For the Oceanic therefore the seasonal densities from these platforms may not
Season, data are from 1980-1982, 1991 and 1994-2001. For be directly comparable. A full consideration of these factors, and
the Davidson Current Season, data are from 1980-1986 and revised maps, are planned for Phase 2 of this project.
1991-2001. Rookery and haulout data are from Mark Lowry
Pecho Rock
of NMFS’ Southwest Fisheries Science Center. Rookery and RESULTS AND DISCUSSION
haulout data are from Mark Lowry of NMFS’ Southwest Fisher- Steller sea lions range from northern Japan, the Aleutian Islands
35°N
35°N
ies Science Center. The rookery numbers represent a general and Gulf of Alaska, south to Año Nuevo Island, California (the
range based on counts of all animals (pups and adults) in three southernmost rookery). Steller sea lion females and pups are
years, 1999-2001. The haulout data are from July 2000.
DRAFT
found at the rookeries year-round, but adult bulls are only at
the rookery during the breeding season (mid-May to mid-July).
METHODS In the study area, the Steller sea lion occurred over the shelf
The latitude and longitude coordinates of Steller sea lions at and slope, and, although there were few at-sea sightings in
sea were used to plot the individual sightings; the coordinates the data set, most occurred in the area between Cordell Bank
124°W 123°W 122°W 121°W
from Mark Lowry were used to plot the haulouts and rookeries. and Año Nuevo Island.
Source Data: See text.
At-sea sightings and effort are the result of a synthesis of data
Figure 65. Map for Steller sea lion: at-sea sightings and survey effort, rookeries and haulouts.
99
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
The Steller sea lion population has declined approximately
64% throughout its range (NMFS Biological Opinion, 2000;
see also NMFS 1992). Based on distributional data, Steller
sea lions are classified into two separate stocks within U.S.
waters: 1) the western stock, that includes animals at, and
west of, Cape Suckling, Alaska, (classified as endangered);
and 2) the eastern stock (including the California population)
that includes animals east of Cape Suckling (classified as
federally threatened). Greatest concentrations of Steller sea
lions occur north of central California, hence, relatively few
sightings (n=45 sightings; n=50 individuals) occurred in the
study area. Insufficient data precluded mapping the Steller
sea lion by seasons.
The breeding season is from mid-May to mid-July. In the study
area, rookeries are located at the Farallon Islands (where they
breed in small numbers and haul-out in slightly larger numbers
throughout the year, USFWS 2000) and at Point Año Nuevo
Island (see LeBoeuf et al. 1991). From 1977 to 1996 on the
Farallon Islands, adult females present during the breeding
season declined by 5.9% and maximum number of pups
counted declined significantly (see Hastings and Sydeman
2002). Until the early 1970’s, Steller sea lions used to breed
at the Point Reyes Headlands but in recent years numbers
have been low (fewer than 50; S. Allen pers. comm., 2003).
Haulout sites north of San Francisco are located at Fish Rocks,
Northwest Cape Rocks, Bodega Rocks, Point Reyes and the
Farallon Islands. Another haulout site not on the map is located
north of Fort Ross at “Sea Lion Rocks”; maximum counts at this
site occur in June (approx. 50) and consist mostly of females
with pups of the year (J. Mortenson pers.comm., 2003). Adult
males disperse widely during the non-breeding season.
Numbers of Steller sea lions off southern and central California
have declined significantly, from 5,000-7,000 non-pups in
1927-1947, to 1,500-2,000 non-pups between 1980-1998
(NMFS Biological Opinion, 2000). Threats to Steller sea
lions include incidental take by commercial fisheries, getting
shot, entanglement in marine debris, declining trends in prey
availability, disease, and contaminants (e.g. premature births
accounted for 20-60% of pup mortality in the South Farallon
Islands between 1973-83). Organochlorine and trace metal
contaminant levels are still elevated in central California Steller
sea lions (NMFS Biological Opinion 2000).
Steller sea lions feed on walleye pollock, capelin, mackerel,
rockfish, herring, salmon, octopus and squid (see Riedman,
1990).
100
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THESE MAPS Upwelling Season, data are from 1980-1982 and 1985-2001.
DRAFT
Northern fur seal Figures 66a, b and c show the density (animals/km2) of northern Oceanic Season, data are from 1980-1982, 1991, and 1994-
Callorhinus ursinus fur seals in the Upwelling, Oceanic, and Davidson Current 2001. Davidson Current Season, data are from 1980-1986 and
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
seasons, displayed in 10’x10’ cells. Densities are based on 1991-2001.
Upwelling Season Oceanic Season
200 m
200 m
20
20
combined data of several studies (see “Methods” and "Data
0
0
0m
0
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14)
m
Sources" sections). The color and mapping intervals were Information on the northern fur seal rookery was provided in
39°N
39°N
customized to show the most structure and highlight significant 2003, courtesy of William Sydeman of PRBO Conservation
Density areas, while allowing comparisons among marine mammal Science, and Joelle Buffa of the Farallon Islands National
(Animals/km²)
species. Cells that were surveyed but which had no northern fur Wildlife Refuge.
10.01 - 50.00 seal’s have a density of zero; unsurveyed areas appear white.
5.01 - 10.00
Blue lines indicate the National Marine Sanctuary boundaries METHODS
38°N
38°N
1.01 - 5.00
of Cordell Bank, Gulf of the Farallones, and Monterey Bay; At-sea densities are the result of a synthesis of data from seven
0.51 - 1.00
bathymetric contours for the 200 m and 2,000 m isobaths are shipboard and aerial survey programs conducted in the study
0.11 - 0.50
also shown in blue. area in the years 1980-2001 (see “Data Sources”). Pinniped
0.06 - 0.10
observation data and trackline data from these studies were
0.01 - 0.05
37°N
37°N
In order to provide one map for the species that integrates the converted to a common format. All aerial data were continuous;
0.00
patterns of its spatial and temporal occurrence in the study ship-based data were converted separately into a continuous
0 25 50 Km
area, map d shows seasonal high use areas, displayed in transect to the extent possible. From the digitized survey data,
10’x10’ cells. This map provides a further synthesis of densities the distributions of effort and of species were mapped into
presented in maps a, b and c (see “Methods” section for details), 10’x10’ cells using CDAS, a custom geographic information
36°N
36°N
and portrays the relative importance of various areas to the system for analyzing marine bird and mammal surveys (MMS,
species. Areas with consistent high use are highlighted on this 2001). The length and width of the survey trackline in a given
map. To provide a relative reference for the “high use” areas, cell (estimated trackline width varied by platform, depending on
cells are also shown where the species were absent (i.e., the speed and height above water) were used to estimate the area
cell was sampled but the species was not recorded there), or sampled. The number of each species seen in a cell was then
35°N
35°N
a b present but at lesser concentrations in any particular season. divided by the area sampled in the cell to estimate density. If a
The single rookery location is also shown. cell was censused more than once, densities were averaged,
200 m
Davidson Current Season
200 m
Seasonal High Use Areas
20
with adjustment made for effort.
20
0
0
0m
0m
DATA SOURCES
(Nov. 15 - Mar. 14) and Rookery
39°N
39°N
At-sea densities for the northern fur seal are based on data Note that these maps represent either sighting locations or
Persistence of
from seven survey programs conducted in 1980-2001. These densities that used survey strip widths relative to each survey
High Use
data were combined using CDAS software into the MMS-CDAS platform (e.g., plane, ship); density was calculated on the basis
3 Seasons
data system (MMS, 2001), developed for Minerals Management of the number of animals sighted and area surveyed. The data
2 Seasons
1 Season
Service and expanded for this project. Of the data sets on the have only been corrected to normalize for survey effort and to
38°N
38°N
Seals present
original CD-ROM, four aerial survey data sets contained data exclude observations with winds greater than 25 knots (smaller
Seals absent
in the study area from Point Arena to Point Sal. Of these, the or less obvious species are often less detectable even at wind
Rookery
OSPR survey program was still ongoing and data from recent speeds of less than 25 knots). Additional corrections are
years were added to this data set. In addition, data from three planned for Phase 2 of this project and are briefly discussed
37°N
37°N
ship-based survey programs were converted to a compatible below.
format for analysis. See "Data and Analyses" subsection in 2.3
for details on individual data sets. For example, no adjustments or corrections have been made to
account for differences in marine mammal detectability among
Data sources for aerial at-sea data include MMS-CDAS (MMS, species and differential probability of detecting animals from
36°N
36°N
2001) and California Department of Fish and Game Office of aerial and shipboard platforms. Individual body size, group
Spill Prevention and Response (CDF&G-OSPR), unpublished size, and species-specific behaviors, such as proportion of
data. Early data were collected using methods described by time spent submerged, are all factors known to affect detection
Bonnell et al. (1983); more recent data were collected using and hence, observed distribution and density estimates as well.
updated technology but with the same general method. Data Because of the very different attributes of aerial and shipboard
35°N
35°N
c d sources for ship-based survey data include David Ainley, platforms, these factors, and the associated adjustments for
unpublished data (see Oedekoven et al. 2001 for details on observations, vary among the studies.
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
methods). Although the at-sea data span the years 1980-2001,
Source Data: See text.
data are not available for all seasons in all years. For the
Figure 66. Maps for northern fur seal: seasonal at-sea densities, high use areas, and rookery.
101
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Northern fur seals feed on a great diversity of seasonally
Map d was developed using the same approach as for maps
abundant prey, and, off California, primary prey species include
a, b and c. For each season, the cells with densities in the top
Pacific hake, northern anchovy, mesopelagic fishes, and market
20% of non-zero values were designated “high use” for that
squid (Kajimura, 1984; see also Riedman, 1990).
season. Cells were scored for “high use” in one, two, or three
seasons and are depicted by color. To provide a relative refer-
ence for the “high use” areas, cells are also shown where the
species were absent (i.e., the cell was sampled but the species
was not recorded there) or present (but densities were never
in the top 20% for any season). Further detail on methods is
provided in the "Data and Analysis" section.
RESULTS AND DISCUSSION
The northern fur seal, one of the most pelagic of the pinnipeds,
is most abundant in continental shelf-slope waters of mid-
latitudes off western North America during winter and early
spring. Except for a small, recently re-established rookery
(south Farallon Island, see below), rookeries occur primarily
outside of the study area. The breeding and pupping season
is June-July, and suckling can continue for three additional
months. During autumn, adult females and juveniles migrate
from rookeries on San Miguel Island in the southern California
Bight (the San Miguel Island stock) and from the Eastern Pacific
stock of the Pribilof Islands in the Bering Sea (Kajimura, 1980;
Kenyon and Wilke, 1953; Pyle et al., 2001). Adult females and
pups from the Pribilof Islands migrate into the North Pacific
Ocean and to waters off Oregon and California.
In data used for this assessment (1980-2001), the northern fur
seal was the second most abundant pinniped observed, with
a total of 1,459 sightings and 2,070 individuals. In the study
area, greatest densities occurred seaward of National Marine
Sanctuary boundaries in the shelf-break, slope, and deep ocean
habitats. The distinctly seasonal presence of this species is
clearly evident in the study area, with greater numbers from
February to May (Kajimura 1984). Greatest densities occur
during the Upwelling and Davidson Current seasons (non-
breeding period) and lesser densities during the Oceanic
Season (breeding period).
Severe declines associated with periods of unusually warm
ocean conditions affect pup production, mortality rates on San
Miguel Island and the Pribilof Islands, and at-sea presence of
this species (see DeLong and Antonelis, 1991; Allen, 1994;
DeLong and Melin, 1999; Melin and DeLong, 2000; Keiper,
2001; and Keiper et al., In Review) In the early 19th century,
American, British, and Russian sealers removed the breeding
population from the South Farallon Islands (Pyle et al., 2001).
Beginning in 1996, however, the species has re-established a
breeding population on the South Farallon Islands, with fewer
than 10 pups produced each year, 1997-2001 (Pyle et al.,
2001). Seasonal high use areas occurred mostly to the west
of National Marine Sanctuary boundaries.
102
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THESE MAPS a common format. All aerial data were continuous; ship-based
124°W 123°W 122°W 121°W
Figure 67 shows individual sightings of harbor seals at sea data were converted separately into a continuous transect to
200 m
20
39°N
39°N
Harbor seal Phoca vitulina
0
0m
along with the locations of haulouts and at-sea survey effort in the extent possible. From the digitized survey data, effort was
the study area. At-sea observations are based on combined mapped into 10’x10’ cells using CDAS, a custom geographic
data of several studies (see “Methods” and “Data Sources” information system for analyzing marine bird and mammal sur-
sections). For context, the amount of combined survey effort veys (MMS-CDAS, 2001). The length and width of the survey
(km of trackline) is also shown, summarized in 10’x10’ cells. trackline in a given cell (estimated trackline width varied by
Haulout locations are based on aerial counts from 2002. Blue platform, depending on speed and height above water) were
Number of Seals lines indicate the National Marine Sanctuary boundaries of used to estimate the area sampled.
At Sea
Cordell Bank, Gulf of the Farallones, and Monterey Bay; bathy-
Haulout Sites
Sightings
metric contours for the 200 m and 2,000 m isobaths are also Note that these maps represent either sighting locations or
7 - 15 401 - 835
shown in blue. densities that used survey strip widths relative to each survey
38°N
38°N
4-6 201 - 400
platform (e.g., plane, ship); density was calculated on the basis
2-3 51 - 200
DATA SOURCES of the number of animals sighted and area surveyed. The data
1 - 50
1
At-sea sightings for the harbor seal are based on data from have only been corrected to normalize for survey effort and
seven survey programs conducted in 1980-2001. These data to exclude observations with winds greater than 25 knots;
Survey Effort
were combined using CDAS software into the MMS-CDAS data additional corrections are planned for Phase 2 of this project
(km of trackline)
system (MMS, 2001), developed for Minerals Management and are briefly discussed below.
> 3000.00
1500.01 - 3000.00 Service and expanded for this project. Of the data sets on the
1000.01 - 1500.00 original CD-ROM, four aerial survey data sets contained data For example, no adjustments/corrections have been made to
500.01 - 1000.00 in the study area from Point Arena to Point Sal. Of these, the account for differences in marine mammal detectability among
100.01 - 500.00 OSPR survey program was still ongoing and data from recent species and differential probability of detecting animals from
0.01 - 100.00
37°N
37°N
years were added to this data set. In addition, data from three aerial and shipboard platforms. Individual body size, group
ship-based survey programs were converted to a compatible size, and species-specific behaviors, such as proportion of
format for analysis. See "Data and Analyses" subsection of this time spent submerged, are all factors known to affect detection
0 25 50 Km
mammal section (2.3). and hence, observed distribution and density estimates as well.
Because of the very different attributes of aerial and shipboard
Data sources for aerial at-sea data include MMS-CDAS (2001) platforms, these factors, and the associated adjustments for
and California Department of Fish and Game Office of Spill observations, vary among the studies.
Prevention and Response (CDFG-OSPR), unpublished data.
Early data were collected using methods described by Bonnell The data in these maps include wind conditions of up to 25
et al. (1983); more recent data were collected using updated knots; smaller or less obvious species are often less detectable
36°N
36°N
technology but with the same general method. Data sources for even at wind speeds of less than 25 knots. The seasonal maps
ship-based survey data include David Ainley, unpublished data contain different combinations of shipboard and aerial data;
(see Oedekoven et al., 2001 for details on methods). Although therefore the seasonal densities from these platforms may not
the overall at-sea data span the years 1980-2001, data are not be directly comparable. A full consideration of these factors, and
available for all seasons in all years. For the Upwelling Season, revised maps, are planned for Phase 2 of this project.
data are from 1980-1982 and 1985-2001. For the Oceanic
Season, data are from 1980-1982, 1991 and 1994-2001. For RESULTS AND DISCUSSION
the Davidson Current Season, data are from 1980-1986 and The harbor seal is distributed from the eastern Aleutian Islands
1991-2001. Haulout information is from 2002 aerial survey data to Baja California and inhabits near-shore estuarine, coastal
(6/12/2002-7/1/2002), from Mark Lowry of NMFS’ Southwest and shelf areas. When at sea, harbor seals were distributed
Fisheries Science Center. in shelf habitats in relatively low densities in all three national
35°N
35°N
marine sanctuaries; therefore, insufficient data precluded
METHODS generating seasonal maps. Harbor seals forage throughout
DRAFT
The latitude/longitude coordinates of harbor seals at sea were the coastal waters. Because the at-sea locations in this map
used to plot the individual sightings. Haulouts were mapped are influenced by survey effort (where survey effort was unequal
coordinates provided by Mark Lowry. At-sea sightings and ef- and coverage was less along the coast), the map may not
fort are the result of a synthesis of data from eight shipboard accurately represent the foraging distribution of harbor seals.
and aerial survey programs conducted in the study area in the Although not evident in the maps, densities are higher in the
124°W 123°W 122°W 121°W
Source Data: See text. years 1980-2001 (see “Data Sources”). Cetacean observation Gulf of the Farallones because there are more and larger
data and trackline data from these studies were converted to haul-out sites in this area (Allen et al., 2002). Harbor seals do
Figure 67. Map for harbor seal: at-sea sightings, survey effort and haulouts.
103
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
not make extensive migrations, and tend to remain relatively
close to their haul-out sites throughout the year. Harbor seals
are inconspicuous at sea and may explain the relatively
low numbers of animals surveyed at sea (sightings: n=192;
individuals: n=235). A long-term monitoring project at Bolinas
Lagoon (Gulf of the Farallones Sanctuary Education Awareness
and Long-term Stewardship Program) protects the seals from
human disturbance. During breeding and molting, relative
abundance increases at Drakes Estero, whereas during winter
(and during herring spawns) relative abundance increases in
Tomales Bay. The Point Reyes region represents ~20% (6000
seals) of the breeding population of the state of California (S.
Allen pers. comm.). Results of recent (2002) tagging studies
have indicated individuals from San Francisco Bay travel to
Duxbury Reef and out to the Farallon Islands to forage (S. Allen
pers.comm.). Harbor seals feed on seasonally abundant prey
that includes topsmelt, night smelt, white croaker, English sole
(Harvey et al., 1995), salmonids (Weise, 2001), and squid and
octopus (see also Riedman, 1990).
In the study area, the species is present year-round, and on
land it is found on sandy beach, mudflat and rocky habitats.
Haulout sites (identified by Lowry 2002) are located along the
coast from Point Arena south to Point Conception, within San
Francisco Bay, and at the Southeast Farallon Islands; habitat
use at these sites, however, varies seasonally throughout the
year (S. Allen, pers. comm.).
Breeding and pupping occurs March-July, and many pupping
sites occur in the study area. Along the Point Reyes National
Seashore, major pupping sites occur at the following locations
(S. Allen pers.comm.): Bodega Rock, Bodega Point, Tomales
Bay (four sites), Tomales Point (five sites), Drakes Estero
(five sites), Limantour Spit, Double Point (two sites), Abalone
Point, Bolinas Point, Duxbury Reef, Bolinas Lagoon (3 sites),
Slide Ranch, and Point Bonita. Sites along the coast south of
San Francisco may exist at Pescadero and Bean Hollow, but
these sites are poorly documented (D.Greig, pers.comm.).
Pupping sites within San Francisco Bay are located at Mowry
Slough and Castro Rocks. Farther south, pupping sites also
occur at Año Nuevo Island, Elkhorn Slough, Hopkins Marine
Station, Cypress Point, Fanshell Beach and Cypress Point,
San Lorenzo River and Point Lobos (D.Greig pers.comm.). A
few pups (less than five) were also produced on South Farallon
Island (USFWS, 2000).
104
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THIS MAP
124°W 123°W 122°W 121°W
continuous; ship-based data were converted separately into a
Figure 68 shows individual at-sea sightings of northern elephant continuous transect to the extent possible. From the digitized
200 m
20
39°N
39°N
Northern elephant seal Mirounga angustirostris
0
0m
seals at sea, along with the locations of rookeries and at-sea survey data, effort was mapped into 10’x10’ cells using CDAS,
survey effort in the study area. At-sea observations are based a custom geographic information system for analyzing marine
on combined data of several studies (see “Methods” and “Data bird and mammal surveys (MMS, 2001). The length and width
Sources” sections). For context, the amount of combined survey of the survey trackline in a given cell (estimated trackline width
effort (km of trackline) is also shown, summarized in 10’x10’ varied by platform, depending on speed and height above wa-
cells. Blue lines indicate the National Marine Sanctuary bound- ter) were used to estimate the area sampled.
Number of Seals
aries of Cordell Bank, Gulf of the Farallones, and Monterey Bay;
At Sea Rookery and
bathymetric contours for the 200 m and 2,000 m isobaths are Note that the these maps represent either sighting locations or
Sightings Haulout Sites
also shown in blue. densities that used survey strip widths relative to each survey
5
platform (e.g., plane, ship); density was calculated on the basis
Point Reyes
38°N
38°N
2
DATA SOURCES of the number of animals sighted and area surveyed. The data
1 At-sea sightings for the northern elephant seal are based on have only been corrected to normalize for survey effort and
data from seven survey programs conducted in 1980-2001. to exclude observations with winds greater than 25 knots;
South Farallon Island
These data were combined using CDAS software into the additional corrections are planned for Phase 2 of this project
Survey Effort
MMS-CDAS data system (MMS, 2001), developed for Miner- and are briefly discussed below.
(km of trackline)
als Management Service and expanded for this project. Of the
> 3000.00
data sets on the original CD-ROM, four aerial survey data sets For example, no adjustments/corrections have been made to
1500.01 - 3000.00
contained data in the study area from Point Arena to Point Sal. account for differences in marine mammal detectability among
1000.01 - 1500.00
500.01 - 1000.00 Of these, the OSPR survey program was still ongoing and data species and differential probability of detecting animals from
Año Nuevo Mainland/Island
100.01 - 500.00 from recent years were added to this data set. In addition, data aerial and shipboard platforms. Individual body size, group
37°N
37°N
0.01 - 100.00 from three ship-based survey programs were converted to a size, and species-specific behaviors, such as proportion of time
compatible format for analysis; see Data and Analysis subsec- spent submerged, are all factors known to affect detection and
tion for more information on individual data sets. hence, observed distribution and density estimates as well.
0 25 50 Km
Because of the very different attributes of aerial and shipboard
Data sources for aerial at-sea data include MMS-CDAS (MMS, platforms, these factors, and the associated adjustments for
2001) and California Department of Fish and Game Office of observations, vary among the studies.
Spill Prevention and Response (CDFG-OSPR), unpublished
data. Early data were collected using methods described by The data in these maps include wind conditions of up to 25
Bonnell et al. (1983); more recent data were collected using knots; smaller or less obvious species are often less detectable
updated technology but with the same general method. Data even at wind speeds of less than 25 knots. The seasonal maps
36°N
36°N
sources for ship-based survey data include David Ainley, contain different combinations of shipboard and aerial data;
Cape San Martin
unpublished data (see Oedekoven et al., 2001 for details therefore the seasonal densities from these platforms may not
Pt. Piedras Blancas on methods). Although the at-sea data span the years 1980- be directly comparable. A full consideration of these factors, and
2001, data are not available for all seasons in all years. For the revised maps, are planned for Phase 2 of this project.
Upwelling Season, data are from 1980-1982 and 1985-2001.
For the Oceanic Season, data are from 1980-1982, 1991, and RESULTS AND DISCUSSION
1994-2001. For the Davidson Current Season, data are from The northern elephant seal is present year-round in the
1980-1986 and 1991-2001. Information on rookery locations study area; however, because they spend very little time
was obtained from Pat Morris, USCS; Brian Hatfield, USGS; at the surface, at-sea sightings are rare, as evidenced by
and Joelle Buffa, FWS. the relatively few sightings during surveys in the study area
(n=266 sightings; n=273 individuals). Therfore, insufficient data
35°N
35°N
METHODS precluded mapping the northern elephant seal by seasons.
The latitude/longitude coordinates of northern elephant seals at
DRAFT
sea were used to plot the individual sightings; the coordinates Northern elephant seals were widely distributed in shelf, shelf-
for rookeries and haulouts were used to plot their locations. break, and slope habitats within the three national marine
At-sea sightings and effort are the result of a synthesis of data sanctuaries, and also occurred in deep ocean habitats seaward
from eight shipboard and aerial survey programs conducted in of the 2000 m isobath. They also occurred well to the north,
the study area in the years 1980-2001 (see “Data Sources”). west, and south of sanctuary boundaries. In these data sets,
124°W 123°W 122°W 121°W
Source Data: See text. Cetacean observation data and trackline data from these stud- age classes of at-sea sightings of seals are unknown.
ies were converted to a common format. All aerial data were
Figure 68. Map for northern elephant seal: at-sea sightings and survey effort, rookeries and haulouts.
105
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
The northern elephant seal breeds, gives birth, and molts on Northern elephant seals are prolonged deep divers that feed on
islands and coastal regions in California, as well as offshore deepwater fishes and invertebrates, including Pacific hagfish
islands of Baja California. The breeding period in the study (Eptatretus stouti), ratfish (Hydrolagus colliei), Pacific hake,
area is generally December through March (Stewart and rockfish, sharks, rays, squid and octopus (Antonelis et al., 1987;
Huber,1993). Northern elephant seals migrate between Condit and LeBoeuf, 1984).
rookeries located within sanctuary boundaries, Farallon Islands,
Point Reyes, Año Nuevo Island and the mainland, Piedras
Blancas, Cape San Martin, and San Simeon, and waters to
the north, where they spend eight to ten months of the year
feeding. Adult males feed in the eastern Aleutian Islands and
the Gulf of Alaska; adult females feed to the west and south of
45º N in deep, oceanic water (Le Boeuf et al., 1993; Stewart
and Huber, 1993; Stewart et al., 1994).
On land, there are three peaks in abundance: 1) during the
breeding/pupping season December to March, with peaks
the last week of January; 2) during the molting season when
female and immatures are on shore April to July with peaks in
May, and adult males are on shore June to early August; and
3) during September to October when immatures haul-out (S.
Allen, pers. comm). Pups depart the pupping sites during the
Upwelling Season. Recent tagging studies indicate that pups
from this region travel as far as Alaska (S.Allen, unpublished
data, National Park Service).
Each year at Año Nuevo Island and mainland, there are
approximately 2,400 females and 300-400 males present,
and approximately 2,200 pups are produced (P. Morris, pers.
comm, 2003). Based on pup counts, the population there
steadily increased through the mid 1990s, but now appears to
be stable (P. Morris, pers. comm., credited to B.J. Le Boeuf).
In contrast, the colony at Piedras Blancas has continued to
rise (in general) over the past five years (B. Hatfield, pers.
comm.) Productivity has declined at two major breeding sites
on Southeast Farallon Island (Sydeman and Allen, 1999;
Nusbaum, 2002), with erosion playing a major role in limiting
the species’ population (USFWS 2000). In California, the net
productivity rate for northern elephant seals also appears to
have declined in recent years (Carretta et al., 2002). However,
the colony at Point Reyes Headlands has continued to increase
by 5-10% per year (Sydeman and Allen, 1999; S. Allen, pers.
comm. 2003). Due to the high surf during the strong El Niño
of 1998, extensive pup mortality occurred at the Point Reyes
colony (Pettee, 1999), but also forced the relocation of the
breeding area; some moved from the main colony at Point
Reyes Headlands to South Beach and North Drakes Bay Beach
(Pettee, 1999).
106
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THESE MAPS the years 1980-2001, data are not available for all seasons in
DRAFT
Dall's porpoise Figures 69a, b and c show the density (animals/km2) of Dall’s all years. For the Upwelling Season, data are from 1980-1982
Phocoenoides dalli porpoise in the Upwelling, Oceanic, and Davidson Current and 1985-2001. For the Oceanic Season, data are from 1980-
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
seasons, displayed in 10’x10’ cells. Densities are based on 1982, 1991, and 1994-2001. For the Davidson Current Season,
Upwelling Season Oceanic Season
200 m
200 m
20
20
combined data of several studies (see “Methods” and “Data data are from 1980-1986 and 1991-2001.
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) Sources” sections). The color and mapping intervals were
39°N
39°N
customized to highlight significant areas and the structure of METHODS
Density species spatial occurrence, while allowing comparisons among At-sea densities are the result of a synthesis of data from eight
(Animals/km²)
marine mammal species. Cells that were surveyed but which shipboard or aerial survey programs conducted in the study
10.01 - 50.00
had no Dall’s porpoise have a density of zero; unsurveyed area in the years 1980-2001 (see “Data Sources”). Cetacean
5.01 - 10.00
38°N
38°N
areas appear white. Blue lines indicate the National Marine observation data and trackline data from these studies were
1.01 - 5.00
Sanctuary boundaries of Cordell Bank, Gulf of the Farallones, converted to a common format. All aerial data were continuous;
0.51 - 1.00
and Monterey Bay; bathymetric contours for the 200 m and ship-based data were converted separately into a continuous
0.11 - 0.50
2,000 m isobaths are also shown in blue. transect to the extent possible. From the digitized survey data,
0.06 - 0.10
the distributions of effort and of species were mapped into
0.01 - 0.05
37°N
37°N
In order to provide one map for the species that integrates the 10’x10’ cells using CDAS, a custom geographic information
0.00
patterns of its spatial and temporal occurrence year-round in the system for analyzing marine bird and mammal surveys (MMS,
0 25 50 Km
study area, map d shows seasonal high use areas, displayed in 2001). The length and width of the survey trackline in a given
10’x10’ cells. This map provides a further synthesis of densities cell (estimated trackline width varied by platform, depending on
presented in maps a, b and c (see the “Methods” section for speed and height above water) were used to estimate the area
36°N
36°N
details), and portrays the relative importance of various areas sampled. The number of cetaceans of each species seen in a
to the species. Areas of consistent high use (colored according cell was then divided by the area sampled in the cell to estimate
to the number of high seasonal use) are highlighted on this density. If a cell was censused more than once, densities were
map. To provide a relative reference for the “high use” areas, averaged, with adjustment made for effort.
cells are also shown where the species were absent (i.e., the
35°N
35°N
a b cell was sampled but the species was not recorded there) or Note that these maps represent either sighting locations or
present but at lower concentrations (densities in a cell were densities that used survey strip widths relative to each survey
200 m
Davidson Current Season
200 m
Seasonal High Use Areas
20
never in the top 20%) in any particular season.
20
platform (e.g., plane, ship); density was calculated on the basis
0
0
0m
0m
(Nov. 15 - Mar. 14) of the number of animals sighted and area surveyed. The data
39°N
39°N
have only been corrected to normalize for survey effort and to
DATA SOURCES
Persistence of
High Use exclude observations with winds greater than 25 knots (smaller
At-sea densities for cetaceans are based on data from eight
3 Seasons
or less obvious species are often less detectable even at
survey programs conducted in 1980-2001. These data were
2 Seasons
wind speeds of less than 25 knots). Additional corrections are
1 Season combined using CDAS software into the MMS-CDAS data
Porpoises present
planned for Phase 2 of this project and are briefly discussed
38°N
38°N
system (MMS, 2001), developed for Minerals Management
Porpoises absent
below.
Service and expanded for this project. Of the data sets on the
original CD-ROM, five aerial survey data sets contained data
For example, no adjustments or corrections have been made to
in the study area from Point Arena to Point Sal. Of these, the
account for differences in marine mammal detectability among
OSPR survey program was still ongoing and data from recent
37°N
37°N
species and differential probability of detecting animals from
years were added to this data set. In addition, data from three
aerial and shipboard platforms. Individual body size, group
ship-based survey programs were converted to a compatible
size, and species-specific behaviors, such as proportion of time
format for analysis. See "Data and Analyses" subsection in 2.3
spent submerged, are all factors known to affect detection and
for details on individual data sets.
hence, observed distribution and density estimates as well.
36°N
36°N
Because of the very different attributes of aerial and shipboard
Data sources for aerial at-sea data include MMS-CDAS (MMS,
platforms, these factors, and the associated adjustments for
2001) and California Department of Fish and Game Office of
observations, vary among the studies.
Spill Prevention and Response (CDF&G-OSPR), unpublished
data. Early data were collected using methods described by
Map d was developed using the same approach as for maps a,
Bonnell et al. (1983) and Dohl et al. (1983); more recent data
35°N
35°N
c d b and c. For each season, the cells with densities in the top 20%
were collected using updated technology but with the same
of non-zero values were designated “high use” for that season.
general method. Data sources for ship-based survey data
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Cells were scored for “high use” in one, two, or three seasons
Source Data: See text. include David Ainley, unpublished data (see Oedekoven et al.
and are depicted by color. To provide a relative reference for
2001 for details on methods). Although the at-sea data span
Figure 69. Maps for Dall’s porpoise: seasonal at-sea densities and high use areas.
107
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
the “high use” areas, cells are also shown where the species
were absent (i.e., the cell was sampled but the species was not
recorded there) or present (but densities were never in the top
20% for any season). Further detail on methods is provided in
the "Data and Analysis" section.
RESULTS AND DISCUSSION
Dall’s porpoise is widely distributed in temperate North Pacific
waters. In the study area, this species was the fourth most
numerous small cetacean. Dall’s porpoise was present during
all seasons in shelf, upper/lower slope, canyon, and deep ocean
habitats seaward of the 2000 m isobath.
During the Upwelling Season densities were somewhat greater
in the Cordell Bank National Marine Sanctuary (NMS) and
northern regions of the Gulf of the Farallones NMS. During
the Oceanic Season, densities were somewhat greater within
and to the north of Cordell Bank and the northern portion of
the Monterey Bay National Marine Sanctuary (MBNMS). The
widespread and deep ocean distribution of the Dall’s porpoise
(well to the west of the National Marine Sanctuary boundaries)
was most evident during the Davidson Current Season (when
effort was greater offshore). No clear seasonal pattern was
evident.
The distribution of Dall’s porpoise is highly variable between
years and appears to be affected by oceanographic conditions
(Forney and Barlow 1998). North-south movements of this
species occur as oceanographic conditions change on seasonal
and interannual time scales (see Green et al., 1992; Barlow,
1995; Forney et al., 1995).
High use areas (based on the CDAS maps) occurred along the
200 m isobath in the Cordell Bank and Gulf of the Farallones
NMS. Given the highly variable distribution of Dall’s porpoise,
the apparent higher relative density in these regions may not
be a seasonal pattern.
See map of SWFSC survey data for Dall’s porpoise (Figure 76)
in this section for the greater geographic extent of the range
and interannual variations for this species.
Dall’s porpoise feeds mostly on Pacific hake (Merluccius
productus), northern anchovy (Engraulis mordax), Pacific
saury (Cololabis saira), juvenile rockfish (Sebastes spp), and
cephalopods (Koskii et al., 1998; Morejohn, 1979).
108
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THESE MAPS and 1985-2001. For the Oceanic Season, data are from 1980-
Lagenorhynchus obliquidens DRAFT
Pacific white-sided dolphin Figures 70a, b and c show the density (animals/km2) of Pacific 1982, 1991 and 1994-2001. For the Davidson Current Season,
white-sided dolphin in the Upwelling, Oceanic, and Davidson data are from 1980-1986 and 1991-2001.
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Current seasons, displayed in 10’x10’ cells. Densities are
Upwelling Season Oceanic Season
200 m
200 m
20
based on combined data of several studies (see “Methods”
20
METHODS
0
0
0m
0m
and “Data Sources” sections). The color and mapping intervals
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) At-sea densities are the result of a synthesis of data from eight
39°N
39°N
were customized to show the most structure and highlight shipboard and aerial survey programs conducted in the study
Density significant areas, while allowing comparisons among marine area in the years 1980-2001 (see “Data Sources” section).
(Animals/km²)
mammal species. Cells that were surveyed but which had no Cetacean observation data and trackline data from these
10.01 - 50.00
Pacific white-sided dolphins have a density of zero; unsurveyed studies were converted to a common format. All aerial data were
5.01 - 10.00
areas appear white. Blue lines indicate the National Marine continuous; ship-based data were converted separately into a
38°N
38°N
1.01 - 5.00
Sanctuary boundaries of Cordell Bank, Gulf of the Farallones, continuous transect to the extent possible. From the digitized
0.51 - 1.00
and Monterey Bay; bathymetric contours for the 200m and survey data, the distributions of effort and of species were
0.11 - 0.50
2,000m isobaths are also shown in blue. mapped into 10’x10’ cells using CDAS, a custom geographic
0.06 - 0.10
information system for analyzing marine bird and mammal
0.01 - 0.05
37°N
37°N
In order to provide one map for the species that integrates the surveys (MMS, 2001). The length and width of the survey
0.00
patterns of its spatial and temporal occurrence in the study area, trackline in a given cell (estimated trackline width varied by
0 25 50 Km
map d shows seasonal high use areas, displayed in 10’x10’ cells. platform, depending on speed and height above water) were
This map provides a further synthesis of densities presented used to estimate the area sampled. The number of cetaceans
in maps a, b and c (see the “Methods” section for details), and of each species seen in a cell was then divided by the area
36°N
36°N
portrays the relative importance of various areas to the species. sampled in the cell to estimate density. If a cell was censused
Areas with consistent high use are highlighted on this map. To more than once, densities were averaged, with adjustment
provide a relative reference for the “high use” areas, cells are made for effort.
also shown where the species were absent (i.e., the cell was
sampled but the species was not recorded there), or present Note that these maps represent either sighting locations or
35°N
35°N
a b but at lesser concentrations in any particular season. densities that used survey strip widths relative to each survey
platform (e.g., plane, ship); density was calculated on the basis
200 m
Davidson Current Season
200 m
DATA SOURCES
Seasonal High Use Areas of the number of animals sighted and area surveyed. The data
20
20
0
0
0m
0m
At-sea densities for cetaceans are based on data from eight have only been corrected to normalize for survey effort and to
(Nov. 15 - Mar. 14)
39°N
39°N
survey programs conducted in 1980-2001. These data were exclude observations with winds greater than 25 knots (smaller
Persistence of
combined using CDAS software into the MMS-CDAS data
High Use or less obvious species are often less detectable even at wind
3 Seasons
system (MMS, 2001), developed for Minerals Management speeds of less than 25 knots). Additional corrections are
2 Seasons
Service and expanded for this project. Of the data sets on the planned for Phase 2 of this project and are briefly discussed
1 Season
Dolphins present original CD-ROM, five aerial survey data sets contained data below.
38°N
38°N
Dolphins absent
in the study area from Point Arena to Point Sal. Of these, the
OSPR survey program was still ongoing and data from recent For example, no adjustments or corrections have been made to
years were added to this data set. In addition, data from three account for differences in marine mammal detectability among
ship-based survey programs were converted to a compatible species and differential probability of detecting animals from
37°N
37°N
format for analysis. See "Data and Analyses" subsection in 2.3 aerial and shipboard platforms. Individual body size, group
for details on individual data sets. size, and species-specific behaviors, such as proportion of time
spent submerged, are all factors known to affect detection and
Data sources for aerial at-sea data include MMS-CDAS (MMS, hence, observed distribution and density estimates as well.
2001) and California Department of Fish and Game Office of Because of the very different attributes of aerial and shipboard
36°N
36°N
Spill Prevention and Response (CDF&G-OSPR), unpublished platforms, these factors, and the associated adjustments for
data. Early data were collected using methods described by observations, vary among the studies.
Bonnell et al. (1983) and Dohl et al. (1983); more recent data
were collected using updated technology but with the same Map d was developed using the same approach as for Maps a,
general method. Data sources for ship-based survey data b and c. For each season, the cells with densities in the top 20%
35°N
35°N
c d include David Ainley, unpublished data (see Oedekoven et al., of non-zero values were designated “high use” for that season.
2001 for details on methods). Although the at-sea data span Cells were scored for “high use” in one, two, or three seasons
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
the years 1980-2001, data are not available for all seasons in and are depicted by color. To provide a relative reference for
Source Data: See text.
all years. For the Upwelling Season, data are from 1980-1982 the “high use” areas, cells are also shown where the species
Figure 70. Maps for Pacific white-sided dolphin: seasonal at-sea densities and high use areas.
109
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
were absent (i.e., the cell was sampled but the species was not
recorded there) or present (but densities were never in the top
20% for any season). Further detail on methods is provided in
the "Data and Analysis" section.
RESULTS AND DISCUSSION
The Pacific white-sided dolphin is one of the most abundant
dolphin species of the temperate eastern North Pacific. In the
present study, it was the most abundant of the small cetaceans
(sightings: n=456; numbers of individuals: n=28,809). Pacific
white-sided dolphins occurred throughout the study area during
all oceanographic seasons in outer shelf, upper/lower slope
and canyon habitats.
Some seasonal shifts in the occurrence of Pacific white-sided
dolphins were observed in the data; densities were relatively
greater during the Oceanic Season, with concentrations near
Pioneer Canyon and Pioneer Seamount and regions over
Monterey Canyon. Because the occurrence of Pacific white-
sided dolphins is highly variable and this species responds to
oceanographic conditions on both seasonal and interannual
time scales (see Forney and Barlow, 1998), the apparent
seasonal shifts observed in these data may not be a seasonal
pattern.
However, in a study in Monterey Bay (Black, 1994), group size
and relative abundance of the Pacific white-sided dolphin varied
seasonally and was greater during the Oceanic and Davidson
Current Seasons than during the Upwelling Season, when
relative individual and group abundance was low and group
sizes were small (not shown in maps; Black, 1994).
Furthermore, in habitats over and near shelf-breaks and greater
bottom relief, feeding behavior was observed more than other
behaviors (Black, 1994). Based on available information, high
use areas mostly occurred over the slope.
Prey of the Pacific white-sided dolphin includes: Pacific whiting,
northern anchovy, rockfish, Pacific saury, and market squid
(Loligo opalescens) (Stroud et al., 1981; Black, 1994).
110
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THESE MAPS and 1985-2001. For the Oceanic Season, data are from 1980-
DRAFT
Risso's dolphin Figures 71a, b and c show the density (animals/km2) of Risso’s 1982, 1991 and 1994-2001. For the Davidson Current Season,
Grampus griseus dolphin in the Upwelling, Oceanic, and Davidson Current data are from 1980-1986 and 1991-2001.
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
seasons, displayed in 10’x10’ cells. Densities are based on
Upwelling Season Oceanic Season
200 m
200 m
20
20
combined data of several studies (see “Methods” and "Data METHODS
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) Sources" sections). The color and mapping intervals were At-sea densities are the result of a synthesis of data from eight
39°N
39°N
customized to show the most structure and highlight significant shipboard and aerial survey programs conducted in the study
Density areas, while allowing comparisons among marine mammal area in the years 1980-2001 (see “Data Sources” section).
(Animals/km²)
species. Cells that were surveyed but which had no Risso’s Cetacean observation data and trackline data from these
10.01 - 50.00 dolphins have a density of zero; unsurveyed areas appear white. studies were converted to a common format. All aerial data were
5.01 - 10.00
Blue lines indicate the National Marine Sanctuary boundaries continuous; ship-based data were converted separately into a
38°N
38°N
1.01 - 5.00
of Cordell Bank, Gulf of the Farallones, and Monterey Bay; continuous transect to the extent possible. From the digitized
0.51 - 1.00
bathymetric contours for the 200m and 2,000m isobaths are survey data, the distributions of effort and of species were
0.11 - 0.50
also shown in blue. mapped into 10’x10’ cells using CDAS, a custom geographic
0.06 - 0.10
information system for analyzing marine bird and mammal
0.01 - 0.05
37°N
37°N
In order to provide one map for the species that integrates the surveys (MMS, 2001). The length and width of the survey
0.00
patterns of its spatial and temporal occurrence in the study area, trackline in a given cell (estimated trackline width varied by
0 25 50 Km
map d shows seasonal high use areas, displayed in 10’x10 cells. platform, depending on speed and height above water) were
This map provides a further synthesis of densities presented used to estimate the area sampled. The number of cetaceans
in maps a, b and c (see “Methods” section for details), and of each species seen in a cell was then divided by the area
36°N
36°N
portrays the relative importance of various areas to the species. sampled in the cell to estimate density. If a cell was censused
Areas with consistent high use are highlighted on this map. To more than once, densities were averaged, with adjustment
provide a relative reference for the “high use” areas, cells are made for effort.
also shown where the species were absent (i.e., the cell was
sampled but the species was not recorded there) or present but Note that these maps represent either sighting locations or
35°N
35°N
a b at lesser concentrations in any particular season. densities that used survey strip widths relative to each survey
platform (e.g., plane, ship); density was calculated on the basis
200 m
Davidson Current Season
200 m
Seasonal High Use Areas DATA SOURCES of the number of animals sighted and area surveyed. The data
20
20
0
0
0m
0m
At-sea densities for cetaceans are based on data from eight have only been corrected to normalize for survey effort and to
(Nov. 15 - Mar. 14)
39°N
39°N
survey programs conducted in 1980-2001. These data were exclude observations with winds greater than 25 knots (smaller
Persistence of
High Use combined using CDAS software into the MMS-CDAS data or less obvious species are often less detectable even at wind
3 Seasons
system (MMS, 2001), developed for Minerals Management speeds of less than 25 knots). Additional corrections are
2 Seasons
Service and expanded for this project. Of the data sets on the planned for Phase 2 of this project and are briefly discussed
1 Season
Dolphins present original CD-ROM, five aerial survey data sets contained data below.
38°N
38°N
Dolphins absent
in the study area from Point Arena to Point Sal. Of these, the
OSPR survey program was still ongoing and data from recent For example, no adjustments or corrections have been made to
years were added to this data set. In addition, data from three account for differences in marine mammal detectability among
ship-based survey programs were converted to a compatible species and differential probability of detecting animals from
37°N
37°N
format for analysis. See "Data and Analyses" subsection in 2.3 aerial and shipboard platforms. Individual body size, group
for details on individual data sets. size, and species-specific behaviors, such as proportion of
time spent submerged, are all factors known to affect detection
Data sources for aerial at-sea data include MMS-CDAS (MMS,, and hence, observed distribution and density estimates as well.
2001) and California Department of Fish and Game Office of Because of the very different attributes of aerial and shipboard
36°N
36°N
Spill Prevention and Response (CDF&G-OSPR), unpublished platforms, these factors, and the associated adjustments for
data. Early data were collected using methods described by observations, vary among the studies.
Bonnell et al. (1983) and Dohl et al. (1983); more recent data
were collected using updated technology but with the same Map d was developed using the same approach as for maps a,
general method. Data sources for ship-based survey data b and c. For each season, the cells with densities in the top 20%
35°N
35°N
c d include David Ainley, unpublished data (see Oedekoven et al. of non-zero values were designated “high use” for that season.
2001 for details on methods). Although the at-sea data span Cells were scored for “high use” in one, two, or three seasons
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
the years 1980-2001, data are not available for all seasons in and are depicted by color. To provide a relative reference for
Source Data: See text.
all years. For the Upwelling Season, data are from 1980-1982 the “high use” areas, cells are also shown where the species
Figure 71. Maps for Risso’s dolphin: seasonal at-sea densities and high use areas.
111
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
were absent (i.e., the cell was sampled but the species was not
recorded there) or present (but densities were never in the top
20% for any season). Further detail on methods is provided in
the "Data and Analysis" section.
RESULTS AND DISCUSSION
The Risso’s dolphin is widely distributed in warm-temperate
waters from southern California north to Washington, and in
the study area, occurred over outer shelf, upper and lower
slope, and canyon habitats, and in offshore waters seaward
of the 2000 m isobath. Risso’s dolphin was the third most
abundant dolphin in the study area, with 250 sightings of
2,248 individuals.
During the Upwelling Season, Risso’s dolphins were distributed
throughout the study area over the outer shelf, slope and deep
ocean, with greatest densities in (and to the south and west
of) the Monterey Bay National Marine Sanctuary (MBNMS).
During the Oceanic Season, greatest densities occurred within
and south and west of the southern portion of MBNMS. During
the Davidson Current Season, overall densities were mostly
in the southern portion of the MBNMS and areas to the south
and west of the MBNMS boundary.
Distribution of Risso’s dolphin off California, Oregon, and
Washington is highly variable, apparently in response to
seasonal and interannual oceanographic changes (Forney
and Barlow, 1998). Dolphins found off California during colder
water months are thought to shift northward into Oregon and
Washington as water temperatures increase in late spring and
summer (Carretta et al., 2001; Green et al., 1992). Given the
highly variable distribution of Risso’s dolphin, the apparent
relative decrease in relative density observed in this study
during the Davidson Current Season may not be a seasonal
pattern. Based on this data set, most high use areas occurred
in the Monterey Bay national marine sanctuary and adjacent
areas to the south (see map).
Risso’s dolphin feed almost exclusively on squid (Koski et al.,
1998; Orr, 1966).
112
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
DRAFT
ABOUT THESE MAPS 1985-2001. Oceanic Season, data are from 1980-1982, 1991
Northern right-whale dolphin Figures 72a, b and c show the density (animals/km2) of northern
Lissodelphis borealis and 1994-2001. Davidson Current Season, data are from 1980-
right whale dolphins in the Upwelling, Oceanic, and Davidson 1986 and 1991-2001.
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Current seasons, displayed in 10’x10’ cells. Densities are based
Upwelling Season Oceanic Season
200 m
200 m
20
20
on combined data of several studies (see “Methods” and “Data METHODS
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14)
39°N
39°N
Sources”). The color and mapping intervals were customized to At-sea densities are the result of a synthesis of data from eight
show the most structure and highlight significant areas, while shipboard and aerial survey programs conducted in the study
Density
allowing comparisons among marine mammal species. Cells area in the years 1980-2001 (see “Data Sources” section).
(Animals/km²)
that were surveyed but which had no northern right whale Cetacean observation data and trackline data from these
10.01 - 50.00
dolphins have a density of zero; unsurveyed areas appear white. studies were converted to a common format. All aerial data were
5.01 - 10.00
38°N
38°N
Blue lines indicate the National Marine Sanctuary boundaries continuous; ship-based data were converted separately into a
1.01 - 5.00
of Cordell Bank, Gulf of the Farallones, and Monterey Bay; continuous transect to the extent possible. From the digitized
0.51 - 1.00
bathymetric contours for the 200m and 2,000m isobaths are survey data, the distributions of effort and of species were
0.11 - 0.50
also shown in blue. mapped into 10’x10’ cells using CDAS, a custom geographic
0.06 - 0.10
information system for analyzing marine bird and mammal
0.01 - 0.05
37°N
37°N
In order to provide one map for the species that integrates the surveys (MMS, 2001). The length and width of the survey
0.00
patterns of its spatial and temporal occurrence in the study trackline in a given cell (estimated trackline width varied by
0 25 50 Km
area, map d shows seasonal high use areas, displayed in platform, depending on speed and height above water) were
10’x10’ cells. This map provides a further synthesis of densities used to estimate the area sampled. The number of cetaceans
presented in maps a, b and c (see “Methods” section for details), of each species seen in a cell was then divided by the area
36°N
36°N
and portrays the relative importance of various areas to the sampled in the cell to estimate density. If a cell was censused
species. Areas with consistent high use are highlighted on this more than once, densities were averaged, with adjustment
map. To provide a relative reference for the “high use” areas, made for effort.
cells are also shown where the species were absent (i.e., the
35°N
35°N
cell was sampled but the species was not recorded there), or Note that these maps represent either sighting locations or
a b present but at lesser concentrations in any particular season. densities that used survey strip widths relative to each survey
platform (e.g., plane, ship); density was calculated on the basis
200 m
Davidson Current Season
200 m
Seasonal High Use Areas
20
20
DATA SOURCES of the number of animals sighted and area surveyed. The data
0
0
0m
0m
(Nov. 15 - Mar. 14) At-sea densities for cetaceans are based on data from eight have only been corrected to normalize for survey effort and to
39°N
39°N
Persistence of survey programs conducted in 1980-2001. These data were exclude observations with winds greater than 25 knots (smaller
High Use
combined using CDAS software into the MMS-CDAS data or less obvious species are often less detectable even at wind
3 Seasons
system (MMS, 2001), developed for Minerals Management speeds of less than 25 knots). Additional corrections are
2 Seasons
1 Season
Service and expanded for this project. Of the data sets on the planned for Phase 2 of this project and are briefly discussed
Dolphins present
38°N
38°N
original CD-ROM, five aerial survey data sets contained data below.
Dolphins absent
in the study area from Point Arena to Point Sal. Of these, the
OSPR survey program was still ongoing and data from recent For example, no adjustments or corrections have been made to
years were added to this data set. In addition, data from three account for differences in marine mammal detectability among
ship-based survey programs were converted to a compatible species and differential probability of detecting animals from
37°N
37°N
format for analysis. See section introduction for details on aerial and shipboard platforms. Individual body size, group
individual data sets. size, and species-specific behaviors, such as proportion of time
spent submerged, are all factors known to affect detection and
Data sources for aerial at-sea data include MMS-CDAS (MMS, hence, observed distribution and density estimates as well.
2001) and California Department of Fish and Game Office of Because of the very different attributes of aerial and shipboard
36°N
36°N
Spill Prevention and Response (CDF&G-OSPR), unpublished platforms, these factors, and the associated adjustments for
data. Early data were collected using methods described by observations, vary among the studies.
Bonnell et al. (1983) and Dohl et al. (1983); more recent data
were collected using updated technology but with the same Map d was developed using the same approach as for maps a,
general method. Data sources for ship-based survey data
35°N
35°N
b and c. For each season, the cells with densities in the top 20%
c d include David Ainley, unpublished data (see Oedekoven et al., of non-zero values were designated “high use” for that season.
2001 for details on methods). Although the at-sea data span Cells were scored for “high use” in one, two, or three seasons
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Source Data: See text. the years 1980-2001, data are not available for all seasons and are depicted by color. To provide a relative reference for
in all years. Upwelling Season, data are from 1980-1982 and the “high use” areas, cells are also shown where the species
Figure 72. Maps for northern right-whale dolphin: seasonal at-sea densities and high use areas.
113
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
were absent (i.e., the cell was sampled but the species was not
recorded there) or present (but densities were never in the top
20% for any season). Further detail on methods is provided in
the "Data and Analysis" section.
RESULTS AND DISCUSSION
The northern right whale dolphin occurs in the temperate
North Pacific, primarily in shelf, slope, and to some degree,
deep ocean waters. In the study area, this species occurred
in outer shelf, slope and canyon habitats. The northern right
whale dolphin was the second most abundant small cetacean
in the study area, with 135 sightings of 22,578 individuals.
Distribution of northern right whale dolphins is highly
variable, apparently in response to seasonal and interannual
oceanographic changes (Forney and Barlow, 1998). Northern
right whale dolphins are found primarily off California during
colder-water months and shift northward into Oregon and
Washington as water temperatures increase in late spring
and summer (Carretta et al., 2001; Forney and Barlow,
1998). Patterns of seasonal abundance have been observed
throughout their range, but there is no information to indicate
that large numbers move between California, Oregon, and
Washington waters (Green et al., 1992). In this study, the
apparent increase in relative densities in the southern portion
of MBNMS during the Davidson Current Season may not
be a seasonal pattern, given the highly variable distribution
of northern right whale dolphins, apparently in response to
seasonal and interannual oceanographic changes (Forney
and Barlow, 1998).
Northern right whale dolphins feed on mesopelagic fishes (e.g.
lanternfish) and squid (Leatherwood and Reeves, 1983).
114
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THIS MAP the extent possible. From the digitized survey data, effort was
124°W 123°W 122°W 121°W
Figure 73 shows the individual sightings of blue whales at sea, mapped into 10’x10’ cells using CDAS, a custom geographic
200 m
20
39°N
39°N
Blue whale Balaenoptera musculus
0
0m
along with at-sea survey effort. Due to insufficient sightings in information system for analyzing marine bird and mammal
the data set (49 sightings of 77 individuals) for the study area, surveys (MMS, 2001). The length and width of the survey
seasonal maps of blue whale density were not generated. trackline in a given cell (estimated trackline width varied by
At-sea sightings for cetaceans are from several studies (see platform, depending on speed and height above water) were
“Methods” and “Data Sources” sections). For context, the com- used to estimate the area sampled.
bined survey effort is also shown, summarized in 10’x10’ cells.
Sightings Blue lines indicate the National Marine Sanctuary boundaries Note that the these maps represent either sighting locations or
(Number of whales) of Cordell Bank, Gulf of the Farallones, and Monterey Bay; densities that used survey strip widths relative to each survey
5 bathymetric contours for the 200m and 2,000m isobaths are platform (e.g., plane, ship); density was calculated on the basis
also shown in blue. Additional data to be added in Phase II may of the number of animals sighted and area surveyed. The data
3
38°N
38°N
make it possible to develop seasonal maps. have only been corrected to normalize for survey effort and
2
to exclude observations with winds greater than 25 knots;
1 DATA SOURCES additional corrections are planned for Phase 2 of this project
At-sea sightings for cetaceans are based on data from eight and are briefly discussed below.
Survey Effort
survey programs conducted in 1980-2001. These data were
(km of trackline)
combined using CDAS software into the MMS-CDAS data For example, no adjustments/corrections have been made to
> 3000.00
system (MMS, 2001), developed for Minerals Management account for differences in marine mammal detectability among
1500.01 - 3000.00
Service and expanded for this project. Of the data sets on the species and differential probability of detecting animals from
1000.01 - 1500.00
original CD-ROM, five aerial survey data sets contained data aerial and shipboard platforms. Individual body size, group
500.01 - 1000.00
in the study area from Point Arena to Point Sal. Of these, the size, and species-specific behaviors, such as proportion of time
100.01 - 500.00
37°N
37°N
OSPR survey program was still ongoing and data from recent spent submerged, are all factors known to affect detection and
0.01 - 100.00
years were added to this data set. In addition, data from three hence, observed distribution and density estimates as well.
ship-based survey programs were converted to a compatible Because of the very different attributes of aerial and shipboard
0 25 50 Km
format for analysis. See section introduction for details on platforms, these factors, and the associated adjustments for
individual data sets. observations, vary among the studies.
Data sources for aerial at-sea data include MMS-CDAS (MMS, The data in these maps include wind conditions of up to 25
2001) and California Department of Fish and Game Office of knots; smaller or less obvious species are often less detectable
Spill Prevention and Response (CDFG-OSPR), unpublished even at wind speeds of less than 25 knots. The seasonal maps
data. Early data were collected using methods described by contain different combinations of shipboard and aerial data;
36°N
36°N
Bonnell et al. (1983) and Dohl et al. (1983); more recent data therefore the seasonal densities from these platforms may not
were collected using updated technology but with the same be directly comparable. A full consideration of these factors, and
general method. Data sources for ship-based survey data in- revised maps, are planned for Phase 2 of this project.
clude David Ainley, unpublished data (see Oedekoven et al.,
2001 for details on methods). Although the at-sea data span RESULTS AND DISCUSSION
the years 1980-2001, data are not available for all seasons in The blue whale is federally listed as endangered under the
all years. For the Upwelling Season, data are from 1980-1982 Endangered Species Act. One population of blue whale (there
and 1985-2001. For the Oceanic Season, data are from 1980- may be as many as five (Carretta et al., 2001; Reeves et al.,
1982, 1991 and 1994-2001. For the Davidson Current Season, 1998)) is present in California waters, generally from June
data are from 1980-1986 and 1991-2001. through November. Arrival and departure times in the study
35°N
35°N
area are highly variable both seasonally and inter-annually (see
METHODS Benson et al., 2002; Calambokidis et al., 1998).
The latitiude and longitude coordinates for blue whales at sea
DRAFT
were used to plot the individual sightings. At-sea sightings and Movement patterns, distribution, and occurrence of blue whales
effort are the result of a synthesis of data from eight shipboard off California are related to their annual migration between
and aerial survey programs conducted in the study area in the foraging areas predominately off central California (but some
years 1980-2001 (see “Data Sources”). Cetacean observation north to British Columbia and south to Baja Mexico and the
data and trackline data from these studies were converted to
124°W 123°W 122°W 121°W
Costa Rican Dome), and the following breeding areas: 1) off
Source Data: See text. a common format. All aerial data were continuous; ship-based the west coast of Baja California (September-December), 2)
data were converted separately into a continuous transect to
Figure 73. Map for blue whale: at-sea sightings and survey effort. the Gulf of California (January-April), and 3) the Costa Rica
115
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Dome (Mate et al., 1999). And although blue whales are often
present in parts of the National Marine Sanctuary waters from
June through November, their occurrence and distribution
during this feeding period is highly variable. Due to insufficient
sightings in the data set (49 sightings of 77 individuals) for
the study area, seasonal maps of blue whale density were
not generated. Additional sighting data for blue whale will be
integrated in Phase 2, and seasonal maps may be generated
at that time.
Blue whales feed on seasonally abundant and dense euphausiid
(krill) schools in discrete depths in the water column (Benson
et al., 2002), concentrated in the deep scattering layer along
canyon and shelf-break edges, and in daytime surface swarms
of krill (Schoenherr, 1991; Croll et al., 1998; Forney and Barlow,
1998). Spatially, they were widely distributed in shelf-break
and slope habitats, as well as seaward of National Marine
Sanctuary boundaries, and to a lesser extent, over the shelf.
Although not directly shown on this map, blue whales also occur
in the Cordell Bank National Marine Sanctuary and off Bodega
Bay (Calambokidis et al., 1990b; Calambokidis et al., 1998),
as well as in waters around the Farallon Islands (C.Keiper,
pers.comm).
There is considerable interchange and interregional movements
between Blue whales that occur off southern California (from
the Santa Barbara Channel and Southern California Bight) to
areas in the Monterey Bay, Gulf of the Farallones, Bodega Bay,
and northern California (Calambokidis et al., 1998). In a study
of the Monterey Bay area (Benson et al., 2002), occurrence
of Blue whales in Monterey Bay was related to seasonal
upwelling patterns that affect seasonally abundant, dense (and
ephemeral) patches of euphausiids that occur during summer
and fall (Benson et al., 2002). See map of SWFSC survey data
for blue whale (Figure 77) for greater geographic extent of the
range and interannual variations for this species.
116
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THESE MAPS and 1985-2001. For the Oceanic Season, data are from 1980-
DRAFT
Humpback whale Figures 74a, b and c show the density (animals/km2) of 1982, 1991 and 1994-2001. For the Davidson Current Season,
Megaptera novaeangliae humpback whales in the Upwelling, Oceanic, and Davidson data are from 1980-1986 and 1991-2001.
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Current seasons, displayed in 10’x10’ cells. Densities are based
Upwelling Season Oceanic Season
200 m
200 m
20
20
on combined data of several studies (see “Methods” and “Data METHODS
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) Sources”). The color and mapping intervals were customized to At-sea densities are the result of a synthesis of data from eight
39°N
39°N
show the most structure and highlight significant areas, while shipboard and aerial survey programs conducted in the study
Density
allowing comparisons among marine mammal species. Cells area in the years 1980-2001 (see “Data Sources”). Cetacean
(Animals/km²)
that were surveyed but which had no humpback whales have observation data and trackline data from these studies were
10.01 - 50.00
a density of zero; unsurveyed areas appear white. Blue lines converted to a common format. All aerial data were continuous;
5.01 - 10.00
38°N
38°N
indicate the National Marine Sanctuary boundaries of Cordell ship-based data were converted separately into a continuous
1.01 - 5.00
Bank, Gulf of the Farallones, and Monterey Bay; bathymetric transect to the extent possible. From the digitized survey data,
0.51 - 1.00
contours for the 200m and 2,000m isobaths are also shown the distributions of effort and of species were mapped into
0.11 - 0.50
in blue. 10’x10’ cells using CDAS, a custom geographic information
0.06 - 0.10
system for analyzing marine bird and mammal surveys (MMS,
0.01 - 0.05
37°N
37°N
In order to provide one map for the species that integrates the 2001). The length and width of the survey trackline in a given
0.00
patterns of its spatial and temporal occurrence in the study cell (estimated trackline width varied by platform, depending on
0 25 50 Km
area, map d shows seasonal high use areas, displayed in speed and height above water) were used to estimate the area
10’x10' cells. This map provides a further synthesis of densities sampled. The number of cetaceans of each species seen in a
presented in maps a, b and c (see “Methods” for details), and cell was then divided by the area sampled in the cell to estimate
36°N
36°N
portrays the relative importance of various areas to the species. density. If a cell was censused more than once, densities were
Areas with consistent high use are highlighted on this map. To averaged, with adjustment made for effort.
provide a relative reference for the “high use” areas, cells are
also shown where the species were absent (i.e., the cell was Note that these maps represent either sighting locations or
sampled but the species was not recorded there), or present densities that used survey strip widths relative to each survey
35°N
35°N
a b but at lesser concentrations in any particular season. platform (e.g., plane, ship); density was calculated on the basis
of the number of animals sighted and area surveyed. The data
200 m
Davidson Current Season
200 m
Seasonal High Use Areas
20
20
DATA SOURCES have only been corrected to normalize for survey effort and to
0
0
0m
0m
(Nov. 15 - Mar. 14) At-sea densities for cetaceans are based on data from eight exclude observations with winds greater than 25 knots (smaller
39°N
39°N
survey programs conducted in 1980-2001. These data were or less obvious species are often less detectable even at wind
Persistence of
High Use combined using CDAS software into the MMS-CDAS data speeds of less than 25 knots). Additional corrections are
3 Seasons
system (MMS, 2001), developed for Minerals Management planned for Phase 2 of this project and are briefly discussed
2 Seasons
Service and expanded for this project. Of the data sets on the below.
1 Season
Whales present
original CD-ROM, five aerial survey data sets contained data
38°N
38°N
Whales absent
in the study area from Point Arena to Point Sal. Of these, the For example, no adjustments or corrections have been made to
OSPR survey program was still ongoing and data from recent account for differences in marine mammal detectability among
years were added to this data set. In addition, data from three species and differential probability of detecting animals from
ship-based survey programs were converted to a compatible aerial and shipboard platforms. Individual body size, group
37°N
37°N
format for analysis. See section introduction for details on size, and species-specific behaviors, such as proportion of time
individual data sets. spent submerged, are all factors known to affect detection and
hence, observed distribution and density estimates as well.
Data sources for aerial at-sea data include MMS-CDAS (MMS, Because of the very different attributes of aerial and shipboard
2001) and California Department of Fish and Game Office of platforms, these factors, and the associated adjustments for
36°N
36°N
Spill Prevention and Response (CDF&G-OSPR), unpublished observations, vary among the studies. Additional data, mapping
data. Early data were collected using methods described by and analysis in Phase 2 may provide more definitive spatial
Bonnell et al. (1983) and Dohl et al. (1983); more recent data patterns for this species.
were collected using updated technology but with the same
general method. Data sources for ship-based survey data Map d was developed using the same approach as for maps a,
35°N
35°N
c d include David Ainley, unpublished data (see Oedekoven et al., b and c . For each season, the cells with densities in the top 20%
2001 for details on methods). Although the at-sea data span of non-zero values were designated “high use” for that season.
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
the years 1980-2001, data are not available for all seasons in Cells were scored for “high use” in one, two, or three seasons
Source Data: See text.
all years. For the Upwelling Season, data are from 1980-1982 and are depicted by color. To provide a relative reference for
Figure 74. Maps for humpback whale: seasonal at-sea densities and high use areas.
117
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
the “high use” areas, cells are also shown where the species Season, most humpback whales are in breeding/calving areas,
were absent (i.e., the cell was sampled but the species was not hence the relatively few sightings in the study area (1980, 1982,
recorded there) or present (but densities were never in the top and 1993) during this season.
20% for any season). Further detail on methods is provided in
the "Data and Analysis" section. The NOAA/SWFSC stock assessment sightings maps
(Figure 78) indicate humpback whales occurred off northern
RESULTS AND DISCUSSION California, and south to Point Conception, with sightings in
The humpback whale is federally listed as endangered under the CBNMS, GFNMS, and MBNMS during 1993, 1996, and
the Endangered Species Act. The eastern North Pacific 2001. During the 1996 and 2001 surveys (when effort extended
stock of the humpback whale that occurs in the study area, north to Washington), humpback whales were also sighted
feeds off California, Oregon, and Washington and migrates off Washington and Oregon. Based on CDAS data shown in
from its breeding and calving areas off coastal Mexico and these maps, most high use areas occurred over the shelf and
Central America (Calambokidis et al., 2000). In this study, the slope.
humpback whale was the most numerous pelagic baleen whale
sighted and was primarily distributed over the shelf, upper slope Humpback whales feed on seasonally abundant, small
and some lower slope habitats. Humpback whales are sighted schooling fishes (e.g. northern anchovy, Pacific sardine,
from the Farallon Islands in all months (Pyle and Gilbert, 1996), Pacific herring) and euphausiids (primarily T. spinifera and E.
though they are more frequently sighted off central California Pacifica; Kieckhefer, 1992). See map of SWFSC survey data
from March through November, with peaks in the summer for humpback whale (Figure 78) for additional geographic extent
and fall (Calambokidis et al., 1996), a pattern reflected in the of the humpback whale range and interannual variations for
seasonal distribution maps. this species.
During the Upwelling Season, humpback whales mostly
occurred in the shelf and slope areas of, and adjacent to,
the Gulf of the Farallones (GFNMS) and the northern part
of Monterey Bay National Marine Sanctuary (MBNMS); see
map for other areas. During the Oceanic Season, the CDAS
map shows the Humpback whales more concentrated in
the areas of the GFNMS, the Cordell Bank National Marine
sanctuary (CBNMS), the northwest corner of the MBNMS, and
the adjacent slope area; the SWFSC Humpback whale map
(Figure 77) shows concentrations over the shelf and slope
throughout the study area extent. Densities and sightings for
the Davidson season were lowest, but like the other seasons,
most occurrences were over the shelf and slope.
A major food type for humpback whales is euphausiids (krill).
The Upwelling Season and beginning of the Oceanic Season
is characterized by a seasonal peak in euphausiid density
that occurs in July/August but can extend into the Oceanic
Season (8/15-11/14). Krill abundance increases one to four
months after seasonal peaks in primary production (Croll et
al. 1998). One of the dominant species of krill (Thysanoessa
spinifera) forms dense shoals in the shelf region from Fort
Ross south to the Channel Islands (Kieckhefer 1992). Primary
feeding sites of humpback whales are located at Monterey
Bay (Benson et al., 2002), Bodega Canyon, Cordell Bank, and
the Farallon Islands (Kieckhefer, 1992). There is considerable
interchange and inter-regional movement of humpback whales
within a feeding season between the Santa Barbara Channel,
Monterey Bay, and to the north off Eureka (Calambokidis et al.,
1996, Calambokidis et al., 1998). During the Davidson Current
118
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
For the Oceanic Season, data are from 1980-1982, 1991 and
ABOUT THESE MAPS
DRAFT
Gray whale 1994-2001. For the Davidson Current Season, data are from
Figures 75a, b and c show the density (animals/km2) of gray
Eschrichtius robustus 1980-1986 and 1991-2001.
whales in the Upwelling, Oceanic, and Davidson Current
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
seasons, displayed in 10’x10’ cells. Densities are based on
Upwelling Season Oceanic Season
200 m
200 m
20
20
METHODS
combined data of several studies (see “Methods” and “Data
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) At-sea densities are the result of a synthesis of data from eight
Sources”). The color and mapping intervals were customized
39°N
39°N
shipboard and aerial survey programs conducted in the study
to show the most structure and highlight significant areas,
Density area in the years 1980-2001 (see “Data Sources”). Cetacean
while allowing comparisons among marine mammal species.
(Animals/km²)
observation data and trackline data from these studies were
Cells that were surveyed but which had no gray whales have
10.01 - 50.00
converted to a common format. All aerial data were continuous;
a density of zero; unsurveyed areas appear white. Blue lines
5.01 - 10.00
ship-based data were converted separately into a continuous
38°N
38°N
indicate the National Marine Sanctuary boundaries of Cordell
1.01 - 5.00
transect to the extent possible. From the digitized survey data,
Bank, Gulf of the Farallones, and Monterey Bay; bathymetric
0.51 - 1.00
the distributions of effort and of species were mapped into
contours for the 200m and 2,000m isobaths are also shown
0.11 - 0.50
10’x10’ cells using CDAS, a custom geographic information
in blue.
0.06 - 0.10
system for analyzing marine bird and mammal surveys (MMS,
0.01 - 0.05
37°N
37°N
2001). The length and width of the survey trackline in a given
In order to provide one map for the species that integrates the
0.00
cell (estimated trackline width varied by platform, depending on
patterns of its spatial and temporal occurrence in the study
0 25 50 Km
speed and height above water) were used to estimate the area
area, map d shows seasonal high use areas, displayed in
sampled. The number of cetaceans of each species seen in a
10’x10’ cells. This map provides a further synthesis of densities
cell was then divided by the area sampled in the cell to estimate
presented in maps a, b and c (see “Methods” section for details),
36°N
36°N
density. If a cell was censused more than once, densities were
and portrays the relative importance of various areas to the
averaged, with adjustment made for effort.
species. Areas with consistent high use are highlighted on this
map. To provide a relative reference for the “high use” areas,
Note that these maps represent either sighting locations or
cells are also shown where the species were absent (i.e., the
densities that used survey strip widths relative to each survey
cell was sampled but the species was not recorded there), or
35°N
35°N
a b platform (e.g., plane, ship); density was calculated on the basis
present but at lesser concentrations in any particular season.
of the number of animals sighted and area surveyed. The data
200 m
Davidson Current Season
200 m
Seasonal High Use Areas
20
have only been corrected to normalize for survey effort and to
DATA SOURCES
20
0
0
0m
0m
exclude observations with winds greater than 25 knots (smaller
At-sea densities for cetaceans are based on data from eight
(Nov. 15 - Mar. 14)
39°N
39°N
or less obvious species are often less detectable even at wind
survey programs conducted in 1980-2001. These data were
Persistence of
High Use speeds of less than 25 knots). Additional corrections are
combined using CDAS software into the MMS-CDAS data
3 Seasons
planned for Phase 2 of this project and are briefly discussed
system (MMS, 2001), developed for Minerals Management
2 Seasons
below.
Service and expanded for this project. Of the data sets on the
1 Season
Whales present original CD-ROM, five aerial survey data sets contained data
38°N
38°N
Whales absent
For example, no adjustments or corrections have been made to
in the study area from Point Arena to Point Sal. Of these, the
account for differences in marine mammal detectability among
OSPR survey program was still ongoing and data from recent
species and differential probability of detecting animals from
years were added to this data set. In addition, data from three
aerial and shipboard platforms. Individual body size, group
ship-based survey programs were converted to a compatible
37°N
37°N
size, and species-specific behaviors, such as proportion of time
format for analysis. See "Data and Analyses" subsection in 2.3
spent submerged, are all factors known to affect detection and
for details on individual data sets.
hence, observed distribution and density estimates as well.
Because of the very different attributes of aerial and shipboard
Data sources for aerial at-sea data include MMS-CDAS (2001)
platforms, these factors, and the associated adjustments for
and California Department of Fish and Game Office of Spill
36°N
36°N
observations, vary among the studies.
Prevention and Response (CDF&G-OSPR), unpublished data.
Early data were collected using methods described by Bonnell
Map d was developed using the same approach as for maps a,
et al. (1983) and Dohl et al. (1983); more recent data were
b and c. For each season, the cells with densities in the top 20%
collected using updated technology but with the same general
of non-zero values were designated “high use” for that season.
method. Data sources for ship-based survey data include David
35°N
35°N
c d Cells were scored for “high use” in one, two, or three seasons
Ainley, unpublished data (see Oedekoven et al., 2001 for details
and are depicted by color. To provide a relative reference for
on methods). Although the at-sea data span the years 1980-
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
the “high use” areas, cells are also shown where the species
2001, data are not available for all seasons in all years. For the
Source Data: See text.
was absent (i.e., the cell was sampled but the species was not
Upwelling Season, data are from 1980-1982 and 1985-2001.
Figure 75. Maps for gray whale: seasonal at-sea densities and high use areas.
119
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Significant declines in calf counts also occurred during this
recorded there) or present (but densities were never in the top
same period (Perryman et al., 2002).
20% for any season). Further detail on methods is provided in
the "Data and Analysis" section.
The apparent seasonal high-use areas noted in map d (the
Farallon Islands, off Año Nuevo, and north of Cordell Bank near
RESULTS AND DISCUSSION
Point Arena and Fort Ross) are likely related to the timing of the
The eastern population of the gray whale migrates from
migration and may not represent a discrete spatial pattern.
summer feeding grounds in the Bering, Chukchi, and western
Beaufort Seas, south along the west coast of North America
Gray whales feed on benthic invertebrates (e.g. gammarid
to its winter breeding and calving areas off the coast of Baja
amphipods; Leatherwood and Reeves, 1983), mysid shrimp,
California. The southward migration includes (in the order of sex
herring eggs/larvae, crab larvae, ghost shrimp (Darling et al.,
and age-class) females in late pregnancy, females that have
1998), and surface swarms of euphausiids (Benson et al.,
recently ovulated, adult males, immature females, and lastly,
2002). Although most individuals of gray whales in the study
immature males. In the study area, this southward migration
area were non-feeding migrants, some individuals do feed on
generally occurs from December through February and peaks
a regular basis near the South Farallon Islands, at the mouth
in January. The northward migration generally occurs from
of Tomales Bay and Drakes Bay (S. Allen, 2002, pers. comm.),
February through May and peaks in March and includes (in
during their northward migration. Gray whales also feed in San
the order of reproductive condition, sex, age-class,) newly
Francisco Bay (in some years: 1999, 2000,-2001; Oliver et al.,
pregnant females, adult males, immature females, and last
2001)) and Monterey Bay (Benson et al., 2002). Gray whales
in this migration, the females with calves. The latter migrate
also have been seen regularly off Point Reyes, Tomales Bay,
northward through the study area during April and May, and
Drakes Bay, and the Farallon Islands during non-migratory
sometimes June. The northward migration is reflected in the
periods (S. Allen, pers. comm.).
distribution patterns during the Upwelling Season, when gray
whales are distributed in the coastal and inner/outer shelf
habitats throughout the study area, en route to their northern
feeding grounds, a pattern reflected in their virtual absence
(according to the data set) in the study area during the Oceanic
Season.
In the study area and data sets analyszed, the gray whale was
the second most numerous baleen whale. Concentrations of
this species were greatest during the Davidson Current Season,
a period that encompasses both the southward and northward
migration, with greatest concentrations observed along the
coast near Cypress Point and south of Point Sur to Lopez
Point. Relative densities were somewhat greater to the north
of CBNMS and to the south of MBNMS (likely related to the
timing of individuals moving north or south). Recent preliminary
documentation of the southbound migration during 2000 and
2001 indicated population estimates of 17,414 (CV=10%), well
below previous (1997/98) estimates of 26,635 (CV=10%; Rugh
et al., 2002). These low estimates may have been caused by
an unusual number of whales that did not migrate as far south
as Granite Canyon (the survey location), or abundance may
have declined following the high mortality rates observed in
1999 and 2000 (Rugh et al., 2002).
Strandings along the coast of North America were six times
more prevalent than during 1995-1998 (Gulland et al., 2001).
Factors that may have contributed to the high number of
strandings include: starvation, anthropogenic and natural
toxins, infectious diseases, ship strikes, detection effort and
reporting, and wind and current effects (Gulland et al., 2001).
120
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Introduction to the SWFSC Data Set, to be used in Phase in the CDAS maps). A visual comparison of the SWFSC maps
Dall's porpoise Phocoenoides dalli
II of this Analysis. The following marine mammal maps are among years (1991, 1993, 1996, 2001) indicates occurrence
based on data from NOAA’s marine mammal stock assessment patterns of blue whales varied; relatively greater concentrations 120°W 118°W 118°W
132°W 130°W 128°W 126°W 124°W 122°W 132°W 130°W 128°W 126°W 124°W 122°W 120°W
program, conducted by NOAA’s Southwest Fisheries Science of blue whales off southern California were evident in 1991,
48°N
48°N
1993
1991
Center. Maps for three species are included below (and 13 more 1993, and 1996, however, this species was virtually absent
are on the CD-ROM) to provide the reader with an idea of some (except for a few sightings) in this region during the survey of Average Average
of the additional data that will be incorporated into the overall 2001. Sightings occurred within NMS boundaries (when effort
46°N
46°N
Group Size Group Size
mammal data set and analysis in Phase II (Figures 76-78). extended into these areas) off Point Reyes in 1991 and 1996,
5.1 - 27.0 5.1 - 27.0
within the Gulf of the Farallones National Marine Sanctuary in
44°N
44°N
3.1 - 5.0 3.1 - 5.0
These maps show sightings (species group size), and effort 1993, and within Monterey Bay in 1996. See "Map Text" and
2.1 - 3.0
locations generally for the late summer/fall season (data ranged "Discussion" in CDAS map section for additional information 2.1 - 3.0
42°N
42°N
1.6 - 2.0
1.6 - 2.0
from July-December) for four years: 1991, 1993, 1996 and on blue whales.
1.0 - 1.5
1.0 - 1.5
2001, off the coasts of California, Oregon and Washington.
40°N
40°N
These maps are the results of broad-scale, ocean ship About the SWFSC Humpback Whale Maps (Figure 78). These Effort
Effort
surveys (aerial surveys are not included), and are used in the SWFSC maps of surveys conducted in July through early
38°N
38°N
development of stock estimates and trend analyses for most December 1991, 1993, 1996, and 2001, (late upwelling and
marine mammals that occur off the coasts of California, Oregon Oceanic season) encompass a much larger geographic extent
36°N
36°N
and Washington. These SWFSC maps do not represent the and indicate concentrations of humpback whales in central
distribution of the species, but they do provide an indication California relatively closer to shore. (See SWFSC blue whale
of the broader spatial extent of the species during the late maps for comparison) and distributed off northern California
34°N
34°N
summer/fall season. and south to Point Conception. A visual comparison of the
SWFSC maps among years (1991, 1993, 1996, 2001) indicates
32°N
32°N
For more information on the marine mammal stock assessment occurrence patterns of humpback whales varied; relatively
survey data, visit: http://swfsc.nmfs.noaa.gov/PRD/CMMP/ or greater concentrations occurred off central California in the
30°N
30°N
contact Dr. Jay Barlow at Jay.Barlow@noaa,gov. survey of 1996, compared to the survey of 1991 (when survey
effort was similar off central California). Sightings within the
About the SWFSC Dall’s Porpoise Maps (Figure 76). These CBNMS, GFNMS, and MBNMS occurred during 1993, 1996,
48°N
48°N
2001
1996
SWFSC maps of surveys conducted in July through early and 2001 (when effort extended into these areas). During
Average Average
December 1991, 1993, 1996, and 2001, (late upwelling and the surveys of 1996 and 2001 (when effort extended north to
46°N
46°N
Group Size Group Size
Oceanic season) encompass a much larger geographic Washington and Oregon), humpback whales also were sighted
extent and provide an example of the off-shore and northern off Washington and Oregon. See "Map Text" and "Discussion" 5.1 - 27.0 5.1 - 27.0
44°N
44°N
geographic extent of the Dall’s porpoise. A visual comparison in CDAS map section for additional information on Humpback 3.1 - 5.0
3.1 - 5.0
of the SWFSC maps among years (1991, 1993, 1996, 2001) whales. 2.1 - 3.0
2.1 - 3.0
42°N
42°N
indicates occurrence patterns of Dall’s porpoise varied; number 1.6 - 2.0
1.6 - 2.0
of sightings was relatively greater off northern California than 1.0 - 1.5
1.0 - 1.5
40°N
40°N
off central California in 1991, (when survey effort was only off
Effort
Effort
California). In the survey of 1996, (when survey effort extended
north to Oregon and Washington), number of sightings and
38°N
38°N
average group size was relatively greater off Oregon and
northern California than off central California. Sightings that
36°N
36°N
occurred within NMS boundaries occurred in Monterey Bay
and off Point Reyes (when effort extended into these areas).
34°N
34°N
See "Map Text" and "Discussion" in CDAS map section for
additional information on Dall’s porpoise.
32°N
32°N
About the SWFSC Blue Whale Maps (Figure 77). These
30°N
30°N
SWFSC maps of surveys conducted in July through early
December 1991, 1993, 1996, and 2001, (late upwelling and 132°W 130°W 128°W 126°W 124°W 122°W 120°W 118°W 132°W 130°W 128°W 126°W 124°W 122°W 120°W 118°W
Oceanic season) encompass a much larger geographic extent These maps contain data from one source: the NMFS/SWFSC cetacean stock assessment
National Marine Fisheries Service,
shipboard surveys, generally conducted during the late summer and fall, mostly from July through
Southwest Fisheries Science Center
December. These maps do not represent the species complete spatial and temporal distribution.
than the study area covered with the CDAS maps and indicate National Ocean Service,
Group size was estimated independently by all observers on each survey vessel who obtained a
good look at that group. These independent estimates of group size were averaged to give the National Centers for Coastal Ocean Science
concentrations of Blue whales off southern California and average group sized estimate for each sighting.
Figure 76. Maps for Dall’s porpoise: SWFSC stock assessment data: average group size of sightings and survey
further off-shore in pelagic, deep ocean habitats (not shown
effort.
121
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Blue Whale Balaenoptera musculus Humpback Whale Megaptera novaeangliae
132°W 130°W 128°W 126°W 132°W
124°W 130°W
122°W 128°W
120°W 126°W
118°W 124°W 122°W 120°W 118°W
132°W 130°W 128°W 126°W 132°W
124°W 130°W
122°W 128°W
120°W 126°W
118°W 124°W 122°W 120°W 118°W
48°N
48°N
48°N
48°N
1991 1993 1991 1993
Average Average Average Average
46°N
46°N
46°N
46°N
Group Size Group Size Group Size Group Size
3.1 - 5.0 3.1 - 14.0
3.1 - 5.0 3.1 - 14.0
44°N
44°N
44°N
44°N
2.1 - 3.0 2.1 - 3.0
2.1 - 3.0 2.1 - 3.0
2.0 2.0 1.9 - 2.0
1.9 - 2.0
42°N
42°N
42°N
42°N
1.1 - 1.9
1.1 - 1.9 1.1 - 1.8
1.1 - 1.8
1.0 1.0
1.0 1.0
40°N
40°N
40°N
40°N
Effort Effort
Effort Effort
38°N
38°N
38°N
38°N
36°N
36°N
36°N
36°N
34°N
34°N
34°N
34°N
32°N
32°N
32°N
32°N
30°N
30°N
30°N
30°N
48°N
48°N
48°N
48°N
1996 2001 1996 2001
Average Average Average Average
46°N
46°N
46°N
46°N
Group Size Group Size Group Size Group Size
3.1 - 14.0
3.1 - 5.0 3.1 - 14.0
3.1 - 5.0
44°N
44°N
44°N
44°N
2.1 - 3.0 2.1 - 3.0 2.1 - 3.0
2.1 - 3.0
2.0 1.9 - 2.0
1.9 - 2.0
2.0
42°N
42°N
42°N
42°N
1.1 - 1.9 1.1 - 1.8
1.1 - 1.9 1.1 - 1.8
1.0 1.0
1.0 1.0
40°N
40°N
40°N
40°N
Effort Effort
Effort Effort
38°N
38°N
38°N
38°N
36°N
36°N
36°N
36°N
34°N
34°N
34°N
34°N
32°N
32°N
32°N
32°N
30°N
30°N
30°N
30°N
132°W 130°W 128°W 126°W 124°W 122°W 120°W 118°W 132°W 130°W 128°W 126°W 124°W 122°W 120°W 118°W 132°W 130°W 128°W 126°W 124°W 122°W 120°W 118°W 132°W 130°W 128°W 126°W 124°W 122°W 120°W 118°W
These maps contain data from one source: the NMFS/SWFSC cetacean stock assessment These maps contain data from one source: the NMFS/SWFSC cetacean stock assessment
National Marine Fisheries Service, National Marine Fisheries Service,
shipboard surveys, generally conducted during the late summer and fall, mostly from July through shipboard surveys, generally conducted during the late summer and fall, mostly from July through
Southwest Fisheries Science Center Southwest Fisheries Science Center
December. These maps do not represent the species complete spatial and temporal distribution. December. These maps do not represent the species complete spatial and temporal distribution.
Group size was estimated independently by all observers on each survey vessel who obtained a National Ocean Service, National Ocean Service,
Group size was estimated independently by all observers on each survey vessel who obtained a
good look at that group. These independent estimates of group size were averaged to give the good look at that group. These independent estimates of group size were averaged to give the
National Centers for Coastal Ocean Science National Centers for Coastal Ocean Science
average group sized estimate for each sighting. average group sized estimate for each sighting.
Figure 78. Maps for humpback whale: SWFSC stock assessment data: average group size of sightings and
Figure 77. Maps for blue whale: SWFSC stock assessment data: average group size of sightings and survey
survey effort.
effort.
122
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Table 25. Preliminary life history and management information for selected marine mammals off north/central California.
SECTION SUMMARY Associations with Bathymetric Protection & Population Status Occurrence & Breeding in Study Area Prey Types
The marine mammal fauna of the study area include species Areas
Other Pelagic Invert's
Benthic Invertebrates
(e.g. squid, octopus)
with a variety of spatial and temporal patterns and can be Marine mammal distributions can be
Euphausiids (krill)
Other Vertebrates
generally characterized as: associated with bathymetrically-defined
• resident, breeding species that occur year-round Protection Time Period of
areas and results include: Status in Population Trend of Population in Primary
Plankton
(e.g. harbor seal, southern sea otter, Steller sea lion); Study Area the Study Area (Increasing, Occurrence in Breeding
Fishes
(FE, FT, SE, Decreasing, Relatively Stable, Study Area Time Period
• species that breed, pup, and molt in the study area and • Widely distributed (found throughout Common Name Scientific Name ST) Unknown) Temporal Occurrence (months) (months)
then as adults, feed elsewhere (e.g. northern elephant the study area): Dall’s porpoise and Federally June/July -
Southern sea otter Enhydra lutris nereis Year-round All months X X
Declining1
Threatened Oct/Nov
seals); northern fur seal (but mostly occurs Present year-round;
California sea lion Zalophus californianus Aug-Mar May-July X X
Increasing2
• species that are seasonally abundant during their migration over the slope and in the deep ocean); seasonally abundant
(e.g. gray whale); California sea lion, (but mostly occurs Año Nuevo: possibly stable last 3 yrs3;
Federally mid-May to
Steller sea lion Eumetopias jubatus Year-round All months X X
Threatened mid-July
Farallones: declining4
• seasonally abundant species that have either migrated to along the coast and over the inner
Stable (California net production may
these waters to forage during summer and fall shelf); and northern elephant seal Harbor seal Phoca vitulina richardsi Year-round All months Mar-June X X
be slowing)2
(e.g. humpback and blue whales) or to forage during and Steller sea lion.
Año Nuevo: Stable, last 5 yrs3;
winter (e.g. northern fur seal and California sea lions); and • Coastal: Southern sea otter, gray At-sea - unknown
Farallones: declining5 ; Pt. Reyes
• species which, though present year-round, exhibit highly whale (but this species also occurs Present year-round; at this time; mid-Dec thru
Northern elephant seal Mirounga angustirostris X X
Headlands: increasing6; California: net seasonally abundant at rookeries - mid-Mar
variable seasonal shifts in distribution (e.g. several species throughout the broad shallow shelf 2a
Nov-Mar
productivity rate declining ; number
of dolphins and porpoises). of the Gulf of the Farallones and in pups appears to be leveling off2a
proximity of the Farallon Islands). San Miguel Island stock: increasing2; Present year-round;
Northern fur seal Callorhinus ursinus Feb-May June-July X X
seasonally abundant
Preliminary CDAS maps for 13 species were developed for • Inner Shelf: Southern sea otter, Pribilof Is: rate of increase 8.12%7
No trend apparent
this document; this is fewer than half of the mammal species California sea lion, Steller sea lion, Dall's porpoise Phocoenoides dalli Year-round Unknown X X
Trends Unknown2 in data
in the study area and the maps are draft. No summary harbor seal, humpback whale and N. California stock: No Trends2a; San
Harbor porpoise (Northern CA,
analyses across mammal species were done, as they would gray whale. Francisco/Russian River and Monterey Unknown at this
San Francisco/Russian River, Phocoena phocoena Year-round Unknown X X
stock: trends in relative abundance not time
be inconclusive, and biased by the limited number and type of • Outer Shelf: California sea lion, Monterey stocks)
statistially significant.2a
species mapped (e.g., coastal, offshore). Steller sea lion, harbor seal, northern No trend apparent
Pacific white-sided dolphin Lagenorhynchus obliquidens Year-round Unknown X X
No Trends2
elephant seal, northern fur seal, in data
No trend apparent
However, preliminary data products do show that marine Risso’s dolphin, Dall’s porpoise, Risso's dolphin Grampus griseus Year-round Unknown X X
Trends Unknown2 in data
mammals of the study area are widely distributed from the Pacific white-sided dolphin, blue Bottlenose dolphin (California No trend apparent
Tursiops truncatus Year-round N/A X X
Stable2
shore to deep ocean, and while some species are found mostly whale, humpback whale and gray coastal stock) in data
Stock status unknown; likely
over the shelf, or deep offshore, most species occur over a whale. Not detected in CDAS No trend apparent
distributional shifts rather than
Short-beaked common dolphin Delphinus delphis Unknown X X
data in data
variety of bathymetric zones. Given that the data and maps • Slope: California sea lion, northern population increase2
are preliminary and most likely incomplete, it is not possible at elephant seal, Northern fur seal, Dall’s No trend apparent
Northern right whale dolphin Lissodelphis borealis Year-round Unknown X X
Trends Unknown2 in data
this time to evaluate the importance of smaller, discrete areas porpoise, Pacific white-sided dolphin, No trend apparent
Killer whale Orcinus orca Year-round Unknown X X X
Trends Unknown2
for the mammal species listed. Risso’s dolphin, northern right whale in data
Baird's beaked whale Berardius bairdii Unknown Insufficient data N/A ? ?
dolphin, blue whale and humpback Trends Unknown2
Beaked Whales
The broad-scale spatial coverage of the 16 maps for cetaceans whale. Mesoplodond spp. Unknown Insufficient data N/A ? ?
Trends Unknown2
(Mesoplodonts)
from the NMFS/SWFSC marine mammal stock assessment • Deep Ocean: California sea lion, Cuvier's beaked whale Ziphius cavirostris Unknown Insufficient data N/A ? ?
2
Trends Unknown
program (Barlow, unpublished data), provided additional northern fur seal, northern elephant Federally No trend apparent
Sperm whale Physeter macrocephalus Seasonal N/A X
2
Trends Unknown
Endangered in data
information for 13 species that were distributed in deep ocean seal, Dall’s porpoise, Pacific white- Federally
Blue whale Balaenoptera musculus Seasonal Aug-Nov N/A X X?
Increasing?8
habitats, and well beyond the range of the current CDAS data sided dolphin, Risso’s dolphin, Endangered
Federally Present year round;
set. These data will likely be incorporated into the CDAS data northern right-whale dolphin, blue Humpback whale Megaptera novaeangliae June-Nov N/A X X
Increasing 6-7%/yr9
Endangered seasonally abundant
set and mammal analysis planned for Phase II. whale, and humpback whale. Federally
Fin whale Balaenoptera physalus Seasonal Aug-Nov N/A X X
Trends Unknown2
Endangered
Stock status unknown; no data on No trend apparent
The marine mammal life history information and analytical Occurrence by Oceanographic Minke whale Balaenoptera acutorostrata Year-round N/A X X
in data
trends2
map products were used to develop the summary spatial and Season Dec-Jan
Increasing to late 1990's; Decreasing
Delisted Present year round;
Gray whale Eschrichtius robustus Dec-Apr (breeds off X X X
temporal distributions described below. Federal 1994 seasonally abundant
The seasonal occurrence patterns of (2002)10 Baja)
marine mammals in waters off north/ Notes
1. This table is preliminary; in Phase II more information will be added and the table will be reviewed by experts.
Life History Characteristics central California were clearly evident 2. A question mark (?) in the table indicates the entry is a probable entry (e.g., prey type); these items may be further evaluated in Phase II.
Table 25 is an initial summary of life history and management for migrating species of large cetaceans 3. Superscripts indicate sources as follows: 1-USGS, 2002; 2-Carretta et al.; 2001; 2a-Carretta et al. 2002; 3-P.Morris pers.comm., credited to B. Le Boeuf; 4-Hastings and Sydeman, 2001; and 5-USFWS, 2000.
6-Sydeman and Allen, 1999; 7-Gerrodette et al.,1985; 8-Calambokidis pers.comm.; 9-Forney et al., 2000; 10-Rugh et al., 2002
information that was identified in the marine mammal mapping (gray, blue and humpback whales) and 4. All marine mammal species have legal protection under the Marine Mammal Protection Act of 1972; species identified as Federally Endangered (FE), Federally Threatened (FT) are identified.
analyses. This table will be expanded in Phase II. for the non-breeding pinnipeds that No marine mammal species have designation as state endangered (SE) or state threatened (ST) in the California at this time.
123
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
breed mostly outside the study area (northern fur seal and seal and California sea lion also were relatively abundant (see Pinnipeds. Pinnipeds were found in the coast, shelf, slope and
• The Pacific white-sided dolphin was the most numerous of
California sea lion). The species occurrences in the three comments on the Upwelling and Oceanic seasons above). deep ocean habitats of the study area.
the small cetaceans and appeared to be more abundant
oceanographic seasons are described below. during the Oceanic Season in the Gulf of the Farallones and
Additional Observations. No clear seasonal patterns could Monterey Bay National Marine Sanctuaries; seasonal high • The California sea lion was the most numerous pinniped
Upwelling Season (~Spring/Summer). This season is be determined in this preliminary visual assessment for use areas were upper/lower slope regions in the Monterey seen in the study area and occurred throughout the region
characterized by an increase in cold, nutrient-rich water the smaller cetaceans: northern right-whale dolphin, Dall’s Bay and Cordell Bank National Marine Sanctuary. Given in coastal, shelf, and upper slope habitats. This species was
brought to the surface by persistent northwest wind and the porpoise, Pacific white-sided dolphin and Risso’s dolphin. the highly variable distribution of the Pacific white-sided most abundant during the Oceanic (just after its breeding
Coriolis effect, followed by intermittent relaxation of upwelling. For the latter two species, however, a shift in distribution was dolphin, the observed spatial and temporal distribution may period) and Davidson Current Season (before its next
Within-season variability during the upwelling process affects evident during the Oceanic season; see Figures 69 and 70. not indicate a general spatial/temporal pattern. breeding period) Seasons. Seasonal high use areas were
food web development and the availability of prey to marine Given the highly variable distribution of the smaller cetaceans, in proximity to major haulout sites near Año Nuevo and the
mammals. shifts in distribution (as indicated on the maps) may not indicate • The northern right whale dolphin was the second most Farallon Islands. Seasonal trends in relative abundance and
a seasonal pattern. numerous of the small cetaceans; concentrations appeared attendance at haulout sites were associated with warm-water
The Upwelling Season is also characterized by variations and to be greater within the Monterey Bay National Marine periods (El Niño events); sea lions were more numerous both
fluctuations in seasonal peaks in abundance of small schooling Elephant seals, Steller sea lions, and harbor seals were present Sanctuary, as well as outside National Marine Sanctuary at-sea and on land during these warm-water periods.
fish and relative densities of euphausiids. The humpback whale in National Marine Sanctuary waters year-round. And although boundaries. Seasonal high use areas were upper/lower slope
was present in greater abundance during the Upwelling and at-sea sightings are relatively infrequent, these species are regions in the Monterey Bay National Marine Sanctuary. No • The northern fur seal was the second most numerous
Oceanic seasons. Humpback whales migrate to waters off frequently sighted at haulouts and rookeries at specific times seasonal pattern in relative abundance was visually detected pinniped seen in the study area and occurred in outer shelf,
north/central California to feed on seasonally-abundant prey. of the year, as noted in Table 25. Important at-sea time periods in the maps; however, a seasonal shift in the distribution upper/lower slope and deep ocean habitats. Although seen
for these infrequently-sighted species is inconclusive due to of this species in the study area was apparent during the in all seasons, this species was most abundant during the
The northern fur seal also was relatively abundant during the insuffcient at-sea data and differences in behavior that affect Oceanic Season (concentrations were greater outside than Upwelling and Davidson Current Seasons (non-breeding
Upwelling and Davidson Seasons. After the breeding/pupping sighting frequency and otherwise low abundance. For example, within National Marine Sanctuary boundaries); during the period), a pattern that coincided with their migration to north/
season (June-July), adult females and juveniles migrate from some of these sighting issues include: at-sea sightings typically Davidson Season the greatest concentrations occurred in central California from San Miguel Island and the Pribilof
rookeries on San Miguel Island in the southern California Bight consist of single individuals or small groups of two or three; the southern regions of the Monterey Bay National Marine Islands. Seasonal high use areas were outside (to the west
(the San Miguel Island stock) and from the Eastern Pacific stock elephant seals are rarely at the surface; and Steller sea lions Sanctuary. Given the highly variable distribution of the and north) of National Marine Sanctuary boundaries.
of the Pribilof Islands and are therefore relatively abundant in are a threatened species and thus occur in small numbers. northern right whale dolphin, the observed occurrences
the study area during winter and early spring. may not indicate a general spatial/temporal pattern. • The northern elephant seal was the third most numerous
Overview of Occurrence Patterns pinniped seen in the study area, however, sightings were too
Oceanic Season (~Autumn). During the Oceanic Season, the • Risso’s dolphin was the third most numerous of the small
Cetaceans. Cetaceans were found throughout the study area; infrequent to determine seasonal trends in at-sea distribution.
northwest winds subside, warmer offshore water is advected cetaceans and occurred in shelf, and upper/lower slope
in coast, shelf, upper/lower slope and deep ocean habitats. Sightings occurred throughout the study region in shelf,
onshore, thermoclines strengthen, ocean conditions become habitats. This species was more widespread during the upper/lower slope and deep ocean habitats.
more stratified and marine mammal prey become more Upwelling Season and more concentrated in the southern
• The humpback whale was the most numerous pelagic baleen
stabilized. The following four species were relatively more portion of the study area (within and outside National Marine
whale seen in the study area and was seen more frequently • The harbor seal was the fourth most numerous pinniped seen
abundant during the Oceanic season (evaluated by the visual Sanctuary boundaries) during the Oceanic and Davidson
during the Upwelling and Oceanic Seasons. Seasonal high in the study area, however sightings were too infrequent to
inspection of the maps): Pacific white-sided dolphin, blue whale, Current Seasons. No clear seasonal shift in relative
use areas within the Gulf of the Farallones National Marine determine seasonal trends in at-sea distribution. Sightings
humpback whale, and California sea lion. Although the Pacific abundance in the study area was detected in a visual
Sanctuary were regions around the Farallon Islands and to occurred in coastal and shelf habitats.
white-sided dolphin occurred during all seasons, it appeared to inspection of maps. Seasonal high use areas were in the
the west of the islands, on the outer shelf and upper slope,
be more numerous during the Oceanic Season. The blue whale Monterey Bay National Marine Sanctuary over slope/canyon
regions over Pioneer, Ascension and Monterey canyons. • The Steller sea lion was sighted rarely, therefore no seasonal
(like the humpback whale) migrates to north/central California habitats. Given the highly variable distribution of the Risso’s
Given the highly variable distribution of humpback whales trends in at-sea distribution could be determined. Sightings
to forage on seasonally-abundant euhausiids during summer dolphin, the observed spatial and temporal distribution may
in the study area during the feeding season, observed spatial of this species occurred in coastal, shelf and upper slope
and fall. The California sea lion, the most abundant pinniped not indicate a general spatial/temporal pattern.
distribution may not indicate a general spatial pattern. habitats.
in the study area was present year-round, however, greater
numbers of sea lions were present during the Oceanic season • The Dall’s porpoise was the fourth most numerous small
• The gray whale was the second most abundant baleen A Fissiped. The southern sea otter is the only fissiped included
(just after the breeding season), but also during the Davidson cetacean and distribution was widespread on shelf, upper/
whale and was found in coastal and shelf regions; relative in the analysis. This species occurs year-round mostly along
Season (before the next breeding season). lower slope, and deep ocean habitats. No clear seasonal
abundance of the gray whale in the study area was greater the coast and inner shelf. Due to insufficient data no spatial/
pattern in relative abundance in the study area was visually
during the Upwelling and Davidson Current Seasons that temporal trends could be determined.
Davidson Current Season (~Winter). The Davidson Current detected in the maps. Seasonal high use areas were upper
coincided with the north and south migration of this species.
Season is characterized by frequent winter storms, downwelling, slope in the Cordell Bank and Gulf of the Farallones National
Seasonal high use areas were to the north of Cordell Bank Preliminary Observations of Species Distributions Relative
relatively warm uniform temperature to considerable depths and Marine Sanctuaries. Given the highly variable distribution
National Marine Sanctuary near Point Arena. Given the to National Marine Sanctuary Boundaries
a deep mixed layer. During this season, the gray whale was of the Dall’s porpoise, the observed occurrences may not
variable distribution of the gray whale, relative to the timing • Eight of the 13 marine mammals evaluated in this assessment
relatively abundant because it migrates through the study area indicate a general spatial/temporal pattern.
of the migration, this observed spatial distribution may not are relatively pelagic, far-ranging marine mammals that are
on its way south (or north) during this period. The northern fur indicate a general spatial pattern. widely distributed, and are either species that occur mostly
in deep ocean habitats (northern fur seal, northern elephant
124
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
mammals (e.g. harbor seals) feed on locally available and
seal, the endangered humpback and blue whales, Dall’s • Rookeries for the northern elephant seal, Steller sea lion and analysis will address the following factors:
seasonally abundant invertebrates or fish in relative proximity
porpoise, Risso’s dolphin), or over upper/lower slope habitats California sea lion occurred within National Marine Sanctuary • differences in survey methodology (e.g. line transect
to their breeding/pupping/haulout sites, whereas the seasonal
(northern right-whale dolphin, Pacific white-sided dolphin). All boundaries of the study area, however, during the El Niño of vs. strip transect);
migrants (e.g. humpback and blue whales) forage on seasonally
occur both in and outside of the National Marine Sanctuary 1998, California sea lion rookeries were located at Lion rock • differences in the detectability of pinnipeds, small and
available krill or fish, a pattern reflected in their relative
boundaries of the study area. and Point Sal Rock, to the south of National Marine Sanctuary large cetaceans, and effects of group size;
abundance during the Upwelling and Oceanic seasons.
boundaries. Haulout sites (n=3) for the threatened Steller sea • differences in time spent underwater; and
• The gray whale also occurs outside National Marine Sanctuary lion are located along the coast to the north of Cordell Bank • differences in environmental conditions (e.g. sea state
Response to Short-Term Changes in Climate. Although it
boundaries, but migrates through sanctuary waters along the National Marine Sanctuary. and other weather conditions).
is likely that short periods of unusually warm or cold waters
coast and over the continental shelf.
affect migratory species and shorter-ranging more resident
• Southern sea otters occurred in coastal shelf waters, almost Major tasks for Phase II are as follows:
species, it was not possible to determine effects of these events
• To the north of Cordell Bank (within relatively close proximity exclusively in Monterey Bay National Marine Sanctuary and 1. Complete the acquisition of data sets for the marine mam-
during this preliminary assessment. Distributional responses
to that National Marine Sanctuary boundary), the Pacific to the south, outside of the study area to San Nicholas Island mals from institutions already contacted (see partial list in
to the extremes of climate (e.g. El Niño vs La Niña) may be
white-sided dolphin, humpback whale, Dall’s porpoise, and in the southern California Bight. No. 2 below).
confounded by: 1) issues associated with comparing different
northern fur seal were present. A relatively high seasonal use
data sets without application of correction factors, 2) small
area for the gray whale also occurred north of Cordell Bank DISCUSSION 2. Continue working with marine mammal experts, and deter-
at-sea populations and therefore small sample size for some
National Marine Sanctuary. Given the general variable nature Differences in habitat use relative to large bathymetric features mine appropriate methods required to analyze additional
species (e.g. elephant seals, harbor seals), 3) far-ranging,
of cetacean distributions, these observations are preliminary are likely related to factors such as the distribution, abundance, data sets and apply appropriate correction factors. At a
migratory species being affected outside of the study area, 4)
and may not indicate a spatial pattern. and availability of various prey (species/sizes). Therefore, the minimum, these data sets will include: sighting data from
demographic lags to species’ responses, 5) variable effects on
importance of the study area must be considered in the context John Calambokidis at Cascadia Research, and the marine
different marine mammal prey, and 6) behavioral differences
• Relatively high seasonal use areas of the Dall’s porpoise, of the variability of ocean climate and oceanography, which mammal stock assessment program data from NOAA’s
among species.
humpback whale and blue whale were located seaward or strongly affects prey availability. Southwest Fisheries Science Center.
west of the Gulf of the Farallones National Marine Sanctuary
Nevertheless, the California sea lion provides an example
over the lower slope. Given the highly variable distribution of Cordell Bank, the Gulf of the Farallones, and Monterey Bay 3. Develop a composite marine mammal data set and maps
of a species shift in distribution in response to changing
these species, these observations are preliminary and may National Marine Sanctuaries encompass some of the most of occurrence patterns for additional mammal species,
oceanographic conditions. The relative increase in at-sea
not indicate a spatial pattern. productive waters along the California coast. Presence of marine as well as summary maps and analyses across species,
abundance during El Niño 1986-87, 1992-93, and 1997-98
mammals in these waters is affected not only by bathymetric for seasons and other selected time periods. Asemblage
(not presented in maps; see studies below), likely reflected a
• Seasonal high use areas of the northern fur seal, northern features, but also changing oceanographic conditions that analyses may be done to identify spatial/temporal species
greater than usual influx of individuals in response to a reduction
right whale dolphin, and Risso’s dolphin occurred seaward result in fluctuations in abundance and distribution of patchily groups.
in food off southern California (see Trillmich and Ono, 1991;
of the Monterey Bay National Marine Sanctuary (western distributed prey. The unique bathymetric features, coupled with
Allen, 1994; Keiper, 2001; Keiper et al., In Review). The greater
areas of the Monterey Canyon and the Shepard’s Meander). the complex physical oceanography off central California play 4. Complete a report on the mammal analyses that will ad-
numbers of sea lions at sea coincided with greater numbers
Given the highly variable distribution of these species, these an important role in the distribution of marine mammal prey dress survey data for 14-23 marine mammal species and
that occurred at haulout sites in the study area. For example,
observations are preliminary and may not indicate a spatial and, in turn, the distribution of mammals themselves. related summary mammal maps (e.g., a composite rookery
an influx of immature sea lions hauled out at Double Point
pattern. and haulout map, spatial and temporal summaries of at-
and at Point Reyes Headlands (per. comm S. Allen) during El
This unique combination of both wide and narrow continental sea occurrence data across selected mammal groups, and
Niño, as was also true at the Farallon Islands (Sydeman and
• Seasonal high use areas of Risso’s dolphin, northern fur seal, shelf, areas of high topographical relief (canyon edges, steep assemblage analyses).
Allen, 1999).
and Pacific white-sided dolphin also were located seaward slopes, ridges, banks, shelf breaks, seamounts), and the
of the southern regions of the Monterey Bay National distinctive oceanographic features associated with seasonal 5. Conduct an expert review of the maps and report and
In summary, seasonal and interannual processes in the ocean
Marine Sanctuary (near Lucia Canyon and to the south). upwelling (e.g., upwelling plumes, fronts, temporal and spatial incorporate necessary revisions.
climate affect variability in ocean conditions and food web
Given the highly variable distribution of these species, these variation in thermocline depth, surface and subsurface currents
development, and thus, the spatial and temporal occurrence
observations are preliminary and may not indicate a spatial and eddies) affect the distribution patterns of organisms at
patterns of marine mammals are strongly linked to the physical MAJOR SECTION CONTRIBUTORS
pattern. many trophic levels. For example, large concentrations of small
and biological processes that affect their prey. Glenn Ford, Carol Keiper, Janet Casey, David Ainley, Sarah
schooling fishes and euphausiids (krill) that are maintained by
Allen, Mark Lowry, Tracy Gill, Ken Buja and Wendy Williams.
• Within the study area, haulout sites for the northern the seasonally high primary productivity (supported by seasonal
Phase II Marine Mammal Assessment. This section provides
elephant seal all occurred within National Marine Sanctuary coastal upwelling), often occur along canyons, shelf-breaks,
preliminary results of the mammal analyses. The maps REVIEWERS
boundaries. Haulout sites for the harbor seal, California sea seamounts, and downstream of upwelling centers located at
presented here provide a preliminary estimate of the mammal The following institutes and people participated in the initial
lion and Steller sea lion also occurred within National Marine Point Arena, Point Reyes, Point Año Nuevo, and Point Sur,
species spatial and temporal use of the study area. In Phase map review in October 2002:
Sanctuary boundaries, but also both north (Steller sea lion, features that also are important areas for both large and small
II, additional data and analysis will likely yield revised maps Sarah G. Allen, Point Reyes National Seashore, Nat'l Park
California sea lion, harbor seal) and south (California sea lion, cetaceans.
for the existing species and additional maps for other species. Service
harbor seal) of the boundaries. Harbor seal haulouts occurred
Some of the data sets for marine mammals have only recently Scott Benson, NMFS/Southwest Fisheries Science Center
along the coast from Point Arena to Point Sal. Marine mammals are highly mobile marine predators that
been received and require further processing before species Jay Barlow, NMFS/Southwest Fisheries Science Center
feed on a great diversity of prey and are attracted to regions
distribution maps can be developed in the GIS. Phase II of this Nancy Black, Monterey Bay Whale Watch Institute
of seasonally abundant high prey densities. Resident marine
125
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Barlow, J. 1988. Harbor porpoise, Phocoena phocoena, abun- Calambokidis, J., J.C. Cubbage, H.H. Steiger, K.C. Balcomb, Croll, D.A., Tershy, B.R., Hewitt, R., Demer,D., Hayes, S.,
Don Croll, University of California, Santa Cruz
dance estimation for California, Oregon and Washington: I. Ship P.Bloedel. 1988. Humpback whale (Megaptera novaeangliae) Fiedler, Pl, Popp, J., and Lopez, V.L. 1998. An integrated ap-
Karin Forney, NMFS/Southwest Fisheries Science Center
surveys. Fish. Bull., US, Vol. 86(3), pp. 417-32. distribution and abundance in the Gulf of the Farallones, 1987 proach to the foraging ecology of marine birds and mammals.,
Mark S. Lowry, NMFS/Southwest Fisheries Science Center
Annual Report to Point Reyes National Seashore NPS and Gulf Deep-Sea Research II, Vol. 45, pp. 1353-1371.
Michelle Staedler, Monterey Bay Aquarium
Barlow, J., Oliver, C.W., Jackson, T.D. and Taylor, B.L. 1988. of the Farallones National Marine Sanctuary NOAA. NFMS. La
Jan Roletto Research Coordinator, GFNMS/CBNMS
Harbor porpoise, Phocoena phocoena, abundance estimation Jolla, CA. 74 pp. Darling, J.D., K.E. Keogh, T.E. Steeves. 1998. Gray whale
And several other members of the NOAA project team
for California, Oregon, and Washingtonf: II. Aeroal surveys. (Eschrichtius robustus) habitat utilization and prey species
and sanctuary programs.
Fish. Bull., US, Vol. 86(3), pp. 433-44. Calambokidis, J., J.C. Cubbage, G.H. Steiger, K.C. Balcomb, off Vancouver Island, B.C. J. Marine Mammal Science, Vol.
and P. Bloedel. 1990a. Population estimates of humpback 14(4), pp. 692-720.
PERSONAL COMMUNICATIONS
Barlow, J. and D. Hanan. 1995. An assessment of the status of whales in the Gulf of the Farallones, California, Report to the
Sarah Allen, Point Reyes National Seashore,
harbor porpoise in central California. Rept. Int. Whal., Special International Whaling Commission, Vol. 12, pp. 325-333 DeLong, R.L., and G.A. Antonelis. 1991. Impacts of the 1982-
U.S. National Park Service
Issue Vol. 16, pp. 123-140. 1983 El Niño on the northern fur seal population at at San
Jay Barlow, Southwest Fisheries Science Center, NOAA
Calambokidis J. G.H. Steiger, J.C. Cubbage, K.C.Balcomb, Miguel Island, California. Pinnipeds and El Niño: Responses to
Joelle Buffa, U.S. Fish and Wildlife Service
Barlow, J. and T. Gerrodette. 1996. Abundance of cetaceans C.Ewald, S. Kruse, R. Wells, and R. Sears. 1990b. Sightings Environmental Stress. (Trillmich, F., and K. Ono, Eds.) Springer-
John Calambokidis, Cascadia Research
in California waters based on 1991 and 1993 ship surveys. and movements of blue whales off central California 1986-88 Verlag, New York. Pp. 75-83.
Karin Forney, Southwest Fisheries Science Center, NOAA
NOAA Tech. Memo NOAA-TM-NMFS-SWFSC-205. La Jolla, from photo-identification of individuals. Report to the Interna-
Denise Greig, The Marine Mammal Center
CA. 68 pp. tional Whaling Commission, Vol. 12, pp. 343-348. DeLong, R.L., S.R. Melin. 1999. Population monitoring stud-
Mike Harris, California Department of Fish and Game
ies of northern fur seals at San Miguel Island, California. Fur
Brian Hatfield, U.S. Geological Survey
Benson S.R., D.A. Croll, B.B. Marinovic, F.P.Chavez, J.T. Har- Calambokidis, J., T. Chandler, K. Rasmussen, G.H. Steiger, seal investigations, 1997. (E.H. Sinclair, BW Robson, Eds) US
Mike Kenner, California Department of Fish and Game
vey. 2002. Changes in the cetacean assemblage of a coastal L. Schlender. 1998. Humpback and blue whale photographic Dep. Commer. NOAA Tech. Memo NMFS-AFSC-69 Kodiak,
Mark Lowry, Southwest Fisheries Science Center, NOAA
upwelling ecosystem during El Niño 1997-98 and La Niña 1999. identification: report on research in 1997. Report to Southwest AK. Pp. 73-80.
Pat Morris, University of California, Santa Cruz
Prog. in Ocgrphy, Vol. 54, pp. 279-291. Fisheries Science Center, NMFS, La Jolla, California. 36 pp.
Joe Mortenson, Gulf of the Farallones National Marine
Dohl, T.P., R.C. Guess, M. Duman and R.C. Helm. 1983. Ce-
Sanctuary, Friends of Marine Sanctuaries Association
Byrd, B. Abundance, distribution, food habits, and prey avail- Calambokidis, J. G.H. Steiger, K. Rasmussen, J.H. Urban R, taceans of central and northern California: status, abundance,
Bob Read, California Department of Fish and Game
ability of the harbor porpoise (Phocoena phocoena) in northern K.C. Balcomb, P. Ladron de Guevara P, M. Salinas Z, J.K. Ja- and distribution Final Rep, Minerals management Service
Michelle Staedler, Monterey Bay Aquarium
Monterey Bay, CA. M.S. Thesis, Moss Landing Marine Labora- cobsen, C.S. Baker, L.M. Herman, S. Cerchio and J.D. Darling. Contract 14-12-0001-29090 OCS Study MMS 84-0044. Los
William Sydeman, PRBO Conservation Science
tories, Stanislaus California State University. 159 pp. 2000. Migratory destinations of humpback whales that feed off Angeles, CA. 143 pp.
California, Oregon and Washington. Mar. Ecol. Prog. Ser., Vol.
REFERENCES
Black, N.A. 1994. Behavior and ecology of Pacific white-sided 192, pp. 295-304. Dohl, T.P., R.C. Guess, M.L. Duman, and R.C. Helm. 1983. Ce-
Allen, S.G. 1994. The distribution and abundance of marine
dolphins (Lagenorhynchus obliquidens) in Monterey Bay, taceans of Central and Northern California, 1980-1983: Status,
birds and mammals in the Gulf of the Farallones and adjacent
California. MS thesis, Moss Landing Marine Laboratories, Calambokidis, J. G.H. Steiger, J.R.Evenson, K.R .Flynn, K.C. Abundance, and Distribution. Part of Investigator’s Final Re-
waters. PhD. Dissertation UC Berkeley.
San Francisco State University CA. 111 pp. Balcomb, D.E. Claridge, P. Bloedel, J.M. Straley, C. S. Baker, O. port, Marine Mammal and Seabird Study, Central and Northern
von Ziegesar, M. E. Dahlheim J.M. Waite, J.D. Darling, G. Ellis California, Contract No.14-12-0001-29090. Prepared by Center
Allen, S.G. , D. Press, S. Waber. 2002. Long term monitoring
Bonnell, M.L., M.O. Pierson and G.D. Farrens. 1983. Pinnipeds and G.A. Green. 1996. Interchange and isolation of humpback for Marine Sciences, University of California, Santa Cruz, for
of harbor seals at Point Reyes National Seashore, Point Reyes
and sea otters of central and northern California, 1980-1983: whales off California and other north Pacific feeding grounds. the Pacific OCS Region, Minerals Management Service. OCS
California 5 year Annual Report. National Park Service Techni-
Status, abundance and distribution. Final report prepared by Mar. Mam. Sci., Vol. 12(2), pp. 215-226. Study MMS 84-0045. Santa Cruz, CA. 284 pp.
cal Report. Washington, D.C. 41 pp.
Center for Marine Studies, University of California, Santa Cruz,
for the Minerals Management Service, Contract 14-12-00. OCS Carretta, J.V., J. Barlow, K.A. Forney, M.M. Muto and J. Baker. Estes, J.A., B.B. Hatfield, K. Ralls, J. Ames. 2003. Causes of
Antonelis, G.A., Jr., M.S. Lowry, D.P. DeMaster, and C.H., Fis-
Study MMS 84-0044. Santa Cruz, CA. 82 pp. 2001. U.S. Pacific marine mammal stock assessments: 2001, mortality in California sea otters during periods of population
cus. 1987. Assessing northern elephant seal feeding habits by
U.S. Dept Commerce, NOAA-NMFS-SWFSC, NOAA Tech growth and decline. J. Marine Mammal Science, Vol. 19(1),
stomach lavage. Mar. Mam. Sci., Vol. 3, pp. 308-322.
Bonell, M. L., M.O. Pierson, and G.D. Farrens. 1983. Pinnipeds Memo NOAA-TM-NMFS-SWFSC-317. La Jolla, CA. 209 pp. pp. 198-216.
and Sea Otters of Central and Northern California, 1980-1983:
Barlow, J. Unpublished data. Cetacean Stock Assessment
Status, Abundance, and Distribution. Part of Investigator’s Fi- Carretta, J.V., M.M. Muto, J. Barlow, J. Baker. Draft U.S. Pacific Forney, K.A., J. Barlow, J.V. Carretta. 1995. The abundance of
Ship Survey Data from the Summer and Fall of 1991, 1993,
nal Report, Marine Mammal and Seabird Study, Central and marine mammal stock assessments 2002, U.S. Dept Com- cetaceans in California waters. Part II: Aerial surveys in winter
1996 and 2001. Coastal Marine Mammal Program, Southwest
Northern California, Contract No. 14-12-0001-29090. Prepared merce, NOAA-TM-NMFS-SWFSC Technical Memorandum. and spring of 1991 and 1992. Fish Bull. Vol. 93, pp. 15-26.
Fisheries Science Center, 8604 La Jolla Shores Dr., La Jolla,
by Center for Marine Sciences, University of California, Santa La Jolla, CA. 199 pp.
CA 92037. Jay.Barlow@noaa.gov http://swfsc.nmfs.noaa.gov/
Cruz, for the Pacific OCS Region, Minerals Management Ser- Forney, K.A., J. Barlow, and J.V. Carretta. 1995. The abundance
prd/cmmp/
vice. OCS Study MMS 84-0044. Santa Cruz, CA. 220 pp. Condit, R.S., and B.J. LeBoeuf. 1984. Feeding habits and of cetaceans in California waters. Part I: Aerial surveys in winter
feeding grounds of the northern elephant seal. J. of Mam., and spring of 1991 and 1992. Fish. Bull., Vol. 93, pp. 15-26.
Barlow, J. 1995. The abundance of cetaceans in California
Bonnell, M.L. and R.G. Ford. 1987. California sea lion distri- Vol. 65, pp. 281-290.
waters. Part I: Ship surveys in summer and fall of 1991. Fish.
butions: A statistical analysis of aerial transect data. J Wildl.
Bull. Vol. 93, pp. 1-14.
Manage. Vol. 51(1), pp. 13-20.
126
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
LeBoeuf, B.J., K. Ono and J. Reiter. 1991. History of the Steller
Hastings, K.K. and W.J. Sydeman. 2002. Population status,
Forney, K.A. and J. Barlow. 1998. Seasonal patterns in the Miller, M.A., L.A. Gardner, C. Krueder, D.M. Paradies, K.R.
sea lion population at Año Nuevo Island, 1961-1991. National
seasonal variation in abundance, and long-term population
abundance and distribution of California cetaceans, 1991-1992. Worcester, D.A. Jessup, E. Dodd, M.D. Harris, J.A. Ames, A.E.
Marine Fisheries Service, Southwest Fisheries Science Center,
trends of Steller sea lions (Eumetopias jubatus) at the South
Mar. Mam. Sci., Vol. 14, pp. 460-489. Packham and P.A. Conrad. 2002. Coastal freshwater runoff is
Administrative Report LJ-91-45C. La Jolla, CA. pp. 24.
Farallon Islands, California. Fish. Bulletin, Vol. 100, pp. 51- a risk factor for Toxoplasma gondii infection of southern sea
62.
Forney, K.A. 1999a. The abundance of California harbor por- otters (Enhydra lutris nereis). Intl. J. of Parasitology, Vol. 32,
Lowry, Mark S. Unpublished data. Survey data for haulouts
poise estimated from 1993-97 aerial line-transect surveys. pp. 997-1006.
and rookeries of California sea lion (Zalophus californianus)
Kajimura, H. 1980. Distribtuion and migration of northern fur
NOAA/SWFSC Admin.Rep. LJ-9-02. La Jolla, CA. 16 pp.
and Steller sea lion (Eumetopias jubatus) in central and north-
seals (Callorhinus ursinus) in the Eastern Pacific. Further analy- National Marine Fisheries Service. 1992. Final recovery plan for
ern California, 1998-2001. National Marine Fisheries Service,
sis of pelagic fur seal data collected by the United States and
Forney K.A. 1999b. Trends in harbor porpoise abundance off Steller sea lions, Eumetopias jubatus. Prepared by the Steller
Southwest Fisheries Science Center. La Jolla, CA.
Canada during 1958-74. Kajimura, H. R.A. Lander, M.A. Perez,
central California, 1986-95: Evidence for interannual changes sea lion recovery team for the National Marine Fisheries Ser-
A.E. Yor (eds). North Pacific Fur Seal Commission 23rd annual
in distribution? J. Cetacean Res. Manage., Vol. 1, pp. 73-80. vice. Silver Spring, Maryland. 290 pp.
Lowry, M.S., C.W. Oliver, C. Macky, and J.B. Wexler. 1990.
meeting Standing Scientific Committee 7-12, April 1980, USSR:
Food habits of California sea lions Zalophus californianus at
Moscow. pp. 19-48.
Forney, K.A., and M.M. Muto, and J. Baker. 2000. Draft U.S. National Marine Fisheries Service. 2000. Biological opinion
San Clemente Island, California, 1981-1986. Fish. Bull. U.S.,
Pacific marine mammal stock assessments: 1999. U.S. Depart- on the California/Oregon Drift Gillnet Fishery. Silver Spring,
Vol. 88, pp. 509-521.
Kajimura, H. 1984. Opportunistic feeding of the northern fur
ment of Commerce, NOAA Technical Memorandum. NMFS- MD. 64 pp.
seal, Callorhinus ursinus, in the eastern North Pacific Ocean
SWFSC-282. La Jolla, CA. 62 pp.
Lowry, M.S., B.S. Stewart, C.B. Hyeath, P.K. Yochem, and J.M.
and eastern Bering Sea. U.S. Dept.of Commerce, NOAA Tech- Nusbaum, J. 2002. Population size and breeding success of
Francis. 1991. Seasonal and annual variability in the diet of Cali-
nical Report NMFS SSFR-779. Washington, D.C. 79 pp.
Gaskin, D.E. 1984. The harbour porpoise (Phocoena pho- the Northern elephant seal on Southeast Farallon Island, PRBO
fornia sea lions Zalophus californianus at San Nicolas Island,
coena): regional populations, status, and information on Marine Science Division; Report to USFWS, San Francisco Bay
California, 1981-1986. Fish. Bull. U.S., Vol. 89, pp. 331-336.
Keiper, C.A. 2001. Marine mammals off central California rela-
direct and indirect catches. Rep. Int. Whal. Comm., Vol. 34, National Wildlife Refuge. San Francisco, CA. 59 pp.
tive to hydrography; 1986-94, 1997-99. MS Thesis, Moss Land-
pp. 569-586.
Lowry, Mark S. and J. V. Carretta (In review). Pacific harbor
ing Marine Laboratories, San Jose State University, 98 pp. Oliver, G.W., J. Gilardi, C.Toropoa, P. Folkens, K. Cronin, N.
seal, Phoca vitulina richardsi, census in California during May-
Glantz, M.R., and J.D. Thompson. 1981. Resource Manage- Bodorff, K. Sanchez, W. Damon, K.A. Zagzebski, B. Winning.
July 2002. To be published as: NOAA Technical Memorandum
Keiper, C.A., D.G. Ainley, S.G. Allen, J.T. Harvey. In Review.
ment and Environmental Uncertainty. Wiley, New York. 63 2001. Gray whales in San Francisco Bay. Poster Abstract
NMFS, NOAA-TM-NMFS-SWFSC. La Jolla, CA.
Inter-annual variability in ocean climate and marine mammal
pp. submitted at the 14th Biennial Conference on the Biology of
occurrence patterns off central California, 1986-1994, 1997- Marine Mammals.
Lowry, M.S. and K.A. Forney. In Review. Abundance and dis-
1999. Marine Ecology Progress Series.
Green, G.A., J.J. Brueggeman, R.A. Grotefendt, D.E. Bowlby,
tributon of California sea lions in central and northern California
M.L. Bonnell and K.C. Balcomb, III. 1992. Cetacean distribu- Orr, R.T. 1966. Risso’s Dolphin on the Pacific Coast of North
during 1998 and summer 1999. Fish. Bull. US.
Kenyon, K.W. and F. Wilke. 1953. Migration of the northern fur
tion and abundance off Oregon and Washington, 1989-1990. America. J. of Mamm., Vol. 47, pp. 341-343.
seal, Callorhinus ursinus. J. of Mamm., Vol. 34(1), pp. 86-98.
Oregon and Washing ton Marine Mammal and Seabird Surveys.
MMS (Minerals Management Service). 2001. Marine Mam-
(J.J Brueggeman, Ed.). Minerals Management Service Contract Perryman, W.L., M.A.Dohahue, P.C.Perkins, S.B.Reilly. 2002.
mal and Seabird Computer Database Analysis System for
Kieckhefer, T.R. 1994. Feeding ecology of humpback whales
Report 14-12-0001-30426. Anchorage, AK. 121 pp. Gray whae calf production 1994-2000: Are observed fluctua-
Washington, Oregon and California, 1975-1997 (MMS-
in continental shelf waters near Cordell Bank, California. MS tions related to changes in seasonal ice cover? J. Mar. Mam.
CDAS, version 2.1). Prepared by Ecological Consulting,
thesis, Moss Landing Marine Laboratories, San Jose State
Gulland, F.M.D., H. Perez-Cortes, J. Urban R., L. Rojas-Bracho, Sci., Vol. 18(1), pp. 121-144.
Inc. (now R.G. Ford and Consulting Co.), Portland, Oregon
University, San Jose, CA. 170 pp.
G. Ylitalo, C. Kreuder, and T. Rowles. 2002. Eastern north Pa-
for the Minerals Management Service, Pacific OCS Region,
cific gray whale (Eschrichtius robustus) unusual mortality event, Pettee, J. 1989. Female northern elephant seal reproductive
Order No.1435-01-97-PO-4206. Anchorage, AK.
Koski, W.R., J.W. Lawson, D.H,Thomson, W.J. Richardson.
1999-2000: A compilation. International Whaling Commission success at Point Reyes National Seashore and micro-habitat
1998. Point Mugu sea range marine mammal technical report.
Rpt SC/54/BRG23. Cambridge, UK. 200 pp. features. MS Thesis, San Francisco State University, San
Mate, B.R., B.A. Lagerquist, and J.Calambokidis. 1999. Move-
Naval Air Warfare Center Weapons Division. San Diego, CA. Francisco, CA.
ments of North Pacific blue whales during their feeding season
280 pp.
Harris, M. Unpublished data. Survey data of southern sea ot-
off Southern California and their southern fall migration. Mar.
ters in California: Fall 2001 and Spring 2002 counts. Marine Pyle, P. and L Gilbert. 1996. Occurrence patterns and trends of
Mamm. Sci., Vol. 15 (4), pp. 1246-1257.
Leatherwood, S., and R.R. Reeves. 1983. The Sierra Club
Region, California Department of Fish and Game. San Diego, cetaceans recorded from southeast Farallon Island, California,
Handbook of Whales and Dolphins. Sierra Club, San Francisco,
California. 1973 to 1994. Northwestern Naturalist 77:1-8.
Melin, S.R., and R.L. Delong 2000. Population and monitoring
CA. 302 pp.
studies of northern fur seals at San Miguel Island, California.
Harvey, J.T., R.C. Helm, and G.V. Morejohn. 1995. Food Habits Pyle P., D.L. Long, J. Schonewald. 2001. Historical and re-
Fur seal investigations, 1998. B.W. Robson (ed.). NOAA Tech.
Le Boeuf, B.J., D. Crocker, S. Blackwell, P. Morris. 1993. Sex
of Harbor Seals Inhabiting Elkhorn Slough, California. California cent colonization of the South Farallon Islands, California, by
Memo. NMFS-AFSC-113. Silver Spring, MD. pp. 41-5.
differences in diving and foraging behaviour of northern el-
Fish and Game, Vol. 81, pp. 1-9. Northern fur seals (Callorhinus ursinus). J. Mar. Mam. Sci.,
ephant seals. Marine Mammal: Advances in Behavioural and Vol. 17(2), pp. 397-402.
Morejohn, G.V. 1979. The Natural History of Dall’s Porpoise in
Population Biology. ( I. Boyd, ed.). Oxford Univ. Press. pp.
the North Pacific Ocean. Behavior of Marine Animals, Vol. 3.
23-45.
Cetaceans. H.E. Winn, and B.L. Olla, eds. Plenum, New York,
NY. pp. 45-83.
127
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Ream, R.R., J.D. Baker and R.G. Towell. 1999. Bogoslof Is- Weise, M.J. 2000. Abundance, food habits, and annual fish
land studies, 1998. Fur seal investigations, 1997. E.H. Sinclair consumption of California sea lion (Zalophus californianus) and
and B.W. Robson (eds). US Dep Commer, NOAA Tech Memo its impact on salmonid fisheries in Monterey Bay, CA M.S.
NMFS-AFSC-69. Kodiak, Ak. 268 pp. Thesis, Moss Landing Marine Laboratories, San Jose State
University. 87 pp.
Reidman, M.L. and J.A. Estes. 1990. The sea other (Enhydra
lutris): Behavior, ecology, and natural history. USFWS. Biologi- Weise, M.J., J.T. Harvey 2001. Nighttime foraging of harbor
cal Report, Vol. 90(14). seals on winter-run steelhead in the San Lorenzo River, Cali-
fornia. Paper presented at the 14th Biennial Conference on the
Rugh, D.J., J.M. Breiwick, R.C. Hobbs, and J.A. Lerczak. Biology of Marine Mammals. 27 pp.
2002. A preliminary estimate of abundance of the eastern
North Pacific stock of gray whales in 2001 and 2002. Paper
SC/545/BRG6 presented to the IWC Scientific Committee,
April 2002. 14 pp.
Schoenherr J.R. 1991. Blue whales feeding on high concen-
trations of euphausiids around Monterey Submarine Canyon.
Can. J. Zool., Vol. 69, pp. 583-594.
Stewart, B.S. and H.R. Huber. 1993. Mirounga angustirostris.
Mammalian Species 449:1-10.
Stewart, B.S., B.J. Le Boeuf, P.K. Yochem, H.R. Huber, R.L.
DeLong, R.J. Jameson, W.,Sydeman, and S.G. Allen. 1994.
History and present status of the northern elephant seal popu-
lation. Elephant Seals. B.J. Le Boeuf and R. M. Laws (eds.)
Univ. Calif. Press, Los Angeles. pp. 101-135.
Stroud, R.K., C.H.Fiscus, and H.Kajimura. 1981. Food of the
Pacific White-sided Dolphin , Lagenorhynchus obliquidens,
Dall’s Porpoise, Phocoenoides dalli, and Northern Fur Seal,
Callorhinus ursinus, off California and Washington. Fishery
Bulletin, Vol. 78, pp. 951-959.
Sydeman, W.J., S.G. Allen. 1999. Pinniped population dynamics
in central California: Correlations with sea surface temperature
and upwelling indices. Mar. Mam. Sci., Vol. 15(2), pp. 446-461.
Trillmich F. and K.A. Ono (eds). 1991. Pinnipeds and El Niño:
Responses to environmental stress. Springer-Verlag, Berlin.
97 pp.
U.S. Geological Survey 2002. (www.werc.usgs.gov/otters/ca-
surveydata.html).
USFWS. 2000. Draft Revised Recovery Plan for the Southern
sea other (Enhydra lutris nereis). Portland, Oregon. U.S. Fish
and Wildlife Service, Ecological Services, Ventura Field Office.
Ventura, California. 64 pp.
128
Section 3: INTEGRATION OF ANALYSES
where ni is the number of individuals belonging to the species
• Option 2: A co-occurrence analysis of marine bird density a continuous modeled surface rather than estimates per 5
(S) in the sample (5 minute grid), and n is the total number of
and fish density minute grid. This approach takes into consideration the spa-
individuals in the sample (Ludwig and Reynolds 1988). Diver-
tial structure in the data to model the gradient of the metric
sity was calculated independently for birds and fishes using all
• Option 3: A co-occurrence analysis of density and diversity between any given pair of sampling points. This results in
species observed within a grid cell.
(options 1 and 2 combined) for both fish and marine birds smoothed surfaces that permit easier visualization of biologi-
cally significant areas. Resulting large-scale patterns have
In the first of these approaches, only patterns of species been described in the context of sanctuary boundaries to
diversity were analyzed. This index was relatively simple to provide insights that may enhance management efficacy in
calculate using the data available for birds and fishes, and these protected areas.
represents a common metric for integration. The second op-
tion focuses on spatial patterns of density. Density is a more DATA AND ANALYSES
intuitive measure than diversity, and it highlights regions of Integration Metrics. There are a number of ways by which
highest marine bird concentrations (abundance). An added ecologists measure diversity. The simplest metric is a count of
attraction of density is that it is only weakly influenced by the total number of unique species in a community, also called
effort. The third approach incorporates the two metrics for species richness (S). This is a straightforward, though poten-
marine birds and fishes simultaneously by combining results tially misleading, measure of diversity. Sampling must be con-
of options 1 and 2. ducted at all locations with the same amount of effort for this
estimate to be comparable across a study region or between
Metrics used in these three options were chosen to best de- data sets. Unfortunately, this was not the case with any of the
INTRODUCTION fine the biogeography for each taxon based on the available source data available for integration. For example, marine
The greatest challenge in developing a large-scale bio- data. Once each integration parameter was mapped, patterns bird observation transects were far more numerous (more ef-
geographic assessment is the synthesis and subsequent of community structure were superimposed and interpreted fort) near shore, and declined dramatically with distance from
analysis of spatial data collected at different scales for varied in the context of various biological and physical covariates. shore. Because this is often the case with biological sampling,
objectives (Gotway and Young 2002). This is particularly true Figure 79. Pictogram of species diversity. Both fish communi-
These spatial covariates were used to better understand gen- a number of diversity indexes have been developed that are,
when attempting to describe meso-scale (tens to hundreds of ties are comprised of 5 species and 14 individuals. In the com-
eral biogeographic patterns, and through interpretation, sug- in theory, more independent of sample size. These are based
kilometers) spatial patterns using data for a range of taxa that munity on the left side, there are 9 individuals of species 1, 1
gest reasons for the observed spatial trends. For example, on the relationship between species richness and the total
were each collected using different sampling techniques. The of species 2, 2 of species 3, 1 of species 4, and 1 of species
results indicated that a portion of highest observed bird and number of individuals observed (n), both of which increase as
taxon-specific sections of this document describe spatial pat- 5. Using the distribution of abundance within this community,
fish diversity occurred adjacent to the shelf/slope interface. It a function of effort, and, ideally, cancel out the effect of effort
terns of community structure for marine birds, mammals, and Shannon’s Index of diversity is 1.12. The community on the
is well documented that strong upwelling of deep ocean wa- on the resulting index (Ludwig and Reynolds, 1988). Here the
fishes. The intent of this section is to coalesce these results right also consists of 5 species with 14 individuals; however, the
ters consistently occurs in areas along the slope. Nutrients in Shannon index of diversity (Shannon and Weaver, 1949) was
and construct a unified and biologically relevant assessment distribution of abundance is more even (2, 3, 3, 4, and 2 indi-
these waters support high phytoplankton productivity, which chosen, as this index is the most widely used in community
of the biogeographic patterns observed. viduals), and consequently Shannon’s Index is higher (1.57).
stimulates a cascade of productivity at all levels of the marine ecology and has relatively small statistical bias when sample
food web (Bolin and Abbott, 1963; Ryther, 1969; Malone, sizes are large (as is the case with this source data).
There are a number of ways to address the challenge of inte- Once diversity was calculated for each taxon in each sample,
1971; Barber and Chavez, 1983; Chavez 1995, 1996; Bakun,
grating results for multiple taxa, and this section contains re- a continuous map surface was interpolated to predict diversity
1996). Diversity may be thought of as being composed of two distinct
sults for three (of many) reasonable options. This integration patterns throughout the study area. The same process was
components: 1) species richness, and 2) species evenness.
effort has been tailored to the NMSP mission of “...enhancing used to model density (see below for detailed methods).
Furthermore, by combining multiple parameters across taxa Evenness is defined as how the number of individuals is dis-
biodiversity, ecological integrity, and cultural heritage”, and (option 3), it was possible to link results presented in earlier tributed among the species. For example, for a community
specifically focuses on the notion of biodiversity in describing Spatial Modeling. This section details the procedure used to
sections to an integrated composite. This approach provides comprised of five species with 70% of the individuals belong-
the overall biogeography of the region. process input data for the integration analyses. While techni-
a clear and tractable interpretation that the reader can follow ing to one species and 30% distributed among the remaining
cal in nature, it provides the information necessary for NMSP
as a logical end point to the preceding series of analyses. four species, the evenness component would be lower than if
After a thorough assessment of the spatial data for each and others to generate results identical to those presented
The combination of diversity and density presents an inclu- there were a more even distribution of individuals among the
taxon, it was concluded that the marine mammal data were here using data provided in the appendix to this document
sive view of important areas across taxa, and is less likely to five species (Ludwig and Reynolds. 1988) (Figure 79). Maxi-
not robust enough in present form to include in the integra- (CD-ROM), and to explore results of alternate modeling op-
overlook regions of potential importance when compared to mum diversity for a given number of species and individuals is
tion process. As such, only birds and fish were considered tions. The observed patterns in diversity and density were
maps depicting a single estimate (e.g., options 1 and 2). This achieved where equal numbers are found for each species in
here. Additional efforts to reconcile outstanding issues in the found to be robust to changes in model parameters; however,
is a critical point as the most diverse patch in a seascape is a community. For consistency, data for all taxa included in this
marine mammal data are ongoing. A final integrated analysis, calculations of the aerial extent of persistent patterns may be
not necessarily the most productive. In addition to the general section were summarized by five minute grids (see sections
including mammal data, will be completed during Phase II of more sensitive. For example, the location of areas of high bird
patterns observed for each metric, the spatial coincidence of 2.1, 2.2, 2.3). Total diversity was estimated within each grid cell
this assessment. The integration alternatives provided in this diversity tends to be relatively constant, regardless of model
hot spots among taxa is emphasized to provide a view into using the Shannon index (H’);
section include: parameters. The quantity (e.g., square kilometers) of these
the integrated ecosystem. The metrics used in this section
n n
S
high areas that fall inside sanctuary boundaries, however,
H ′ = − ∑ i ln i
are similar to those described in sections 2.1 (fish) and 2.2
• Option 1: A co-occurrence analysis of diversity hot spots for may change.
(marine birds); however, data were interpolated to produce i =1 n n
marine birds and marine fishes
129
Section 3: INTEGRATION OF ANALYSES
on the pattern of the empirical variograms and the lack of data
For interpolation and calculation of spatial autocorrelation sta-
Table 26. Summary statistics and parameter estimates for spatial models.
at short lag distances (due to the five minute minimum sepa-
tistics, data for each 5 minute grid cell were assigned to the
Geary's C (significance)
Moran's I (significance)
ration between points), which are necessary to differentiate
cell centroid. All data were analyzed in the Universal Trans-
(total,minimum per
anisotropic model)
Cross Validation r 2r2
Cross-validation
between spherical and Gaussian models.
verse Mercator (UTM) projection. Projection is necessary to
(minor range for
Number of lags
Lag size (km)
ensure that the value of x and y units is equivalent and con-
Sample Size
Detrending?
Range (km)
Lag size (km)
Neighbors
4) Surface Interpolation: The interpolation method used
stant across the study region. The spatial modeling process
Partial Sill
Geary's C
Sample size
Moran's I
Range (km)
Detrending
Neighbors
Partial sill
Nugget
is termed ‘ordinary kriging’. Kriging is a linear interpolation
to generate an interpolated surface consisted of the following
sector)
Nugget
method that allows predictions of unknown values of a ran-
sequence of operations:
Lags
dom function from observations at known locations (Kaluzny
Bird diversity 1,163 0.141** 0.830** Yes 20.451 12 99.851 0.25 0.117 0.541 8,2
et al., 1998). Ordinary kriging is the kriging method gener-
1) Checking for Spatial Autocorrelation: Prior to interpola-
Bird diversity residual 1,163 0.067** 0.893** YesYes 9.966 12 88.062 0.199 0.125 0.407 8,8,2
Bird diversity 1163 0.141** 0.830** 20.451 12 99.851 0.25 0.117 0.541 2
ally used for interpolation of a single continuous variable of
tion, all data were tested for the presence of spatial autocorre-
Bird density 1,403 0.058** 0.891** Yes 5 30 149.54 1,189.30 1,189.50 0.253 20,5
unknown mean. Kriging is preferred over other interpolation
lation. Positive autocorrelation (where values for neighboring Bird diversity residual 1163 0.067** 0.893** Yes 9.966 12 88.062 0.199 0.125 0.407 8, 2
methods because: 1) weights are based on an empirical as-
pairs of points are more similar to one another than are distant Fish diversity 301 0.018**
1403 0.058** 0.973* No 5 20 30 149.54 0.148
29.595 0.046 0.025 20,5
Bird density 0.891** Yes 5000 1189.3 1189.5 0.253 20, 5
sessment of the data’s spatial structure (the variogram), 2)
ones) is important for accurate interpolation. Moran’s I and Fish diversity residual 301 3010.013*
0.018** 0.975* No no 55 10 18.394 0.11 0.061 0.076 20,5
Fish diversity 0.973* 20 29.595 0.148 0.046 0.025 20, 5
kriging is an unbiased predictor, and 3) for many variables,
Geary’s C statistics were calculated for each interpolated vari- Fish diversity residual 301 0.013* 0.975* no 5 10 18.394 0.11 0.061 0.076 20, 5
37.929
kriging has been shown to outperform other interpolation
able to test for the presence of significant spatial autocorrela- (16.802)
Fish density 301 0.020** 0.969* No 5 8 0.0046 0.001 0.074 20,5
methods, such as inverse distance weighting (IDW) and trian-
tion using CrimeStat (Levine, 2002). Moran’s I is the standard ** indicates significance at p = 0.001, * indicates significance at p = 0.05
gulated irregular networking (TIN) (Guan et al., 1999). Before
autocorrelation statistic and provides a global (i.e. across the
kriging can be applied, two assumptions must be checked.
study area) test of spatial autocorrelation. Geary’s C is more
The first is stationarity; the mean (and ideally the variance)
sensitive to autocorrelation within small neighborhoods. Con-
a correlation exists, maps of diversity may simply reflect the birds. Areas of high marine bird diversity that overlap with low
must be constant across the spatial extent of the data. That
firmation of statistically significant spatial autocorrelation sug-
distribution of effort. In order to correct for differences in effort residuals should be interpreted with caution, as these hot
is, any large scale trend must be removed (see #2 above).
gests that point data are suitable for interpolation. As such,
across the study region, the following technique was applied: spots may simply reflect areas of unusually high effort. Since
The second assumption is isotropy of the variogram. The co-
interpolation was performed only where this was true for both
A second order polynomial regression of diversity on effort bird and fish density were only weakly correlated with effort,
variance between any two points is assumed to be a function
autocorrelation statistics.
was conducted and the residuals were interpolated as de- no attempt was made to correct the density maps.
only of the distance between the points, not of their location
scribed above. The interpolated map of residuals depicts ar-
or angle. This assumption can be examined and, if necessary,
2) Detrending: Detrending is done to ‘standardize’ the es-
eas of higher or lower diversity relative to that expected given ANALYTICAL MAP PRODUCTS
corrected for during the variogram modeling stage (see #3
timate across the analysis extent, and is a prerequisite for
the amount of local effort. This map was overlayed on the Spatial Statistics. Table 26 summarizes the results of
above). Trend analysis was conducted using JMP statistical
the interpolation procedure used here. After interpolation, the
interpolated map of diversity to visualize the impact of effort spatial autocorrelation tests, variogram fitting, and kriging
software (SAS Institute), while detrending, variogram mod-
removed trend is added back into the model results. Each in-
on the observed patterns in diversity. Although significantly cross validation. All variables were found to be significantly
eling, and kriging were conducted using the ArcView (GIS)
terpolated variable was plotted against Northing and Easting,
correlated with effort, fish diversity showed nearly identical positively spatially autocorrelated (p < 0.05) by both the
Geostatistical Analyst Extension (ESRI Inc.).
and a linear trend was fit to each plot. When significant trend
patterns as the map of fish diversity residuals. Fish diversity is Moran’s I and Geary’s C statistics, for fish and marine birds.
(p < 0.05) was present for either Northing or Easting, the data
therefore not shown with the overlay of residuals. Patterns of Spatial autocorrelation was more pronounced in the marine
The kriging neighborhood was set to the twenty nearest
were detrended (first order) before variogram modeling and
marine bird diversity, however, differed substantially from pat- bird data than in the fish data resulting in better model fit and a
neighbors with a minimum of five neighbors for each 90 de-
kriging.
terns of the bird diversity residuals, indicating that differences higher cross-validation r-squared value for the bird data sets.
gree angular sector for the fish data, and reduced to eight and
in effort are responsible for some of the observed pattern. Ma-
five for birds in order to capture small scale variability. Cross
3) Variogram Modeling: Empirical variograms show the de-
rine bird diversity hot spots, as represented by the top 20% of
validation was conducted to assess model accuracy by re-
crease in relatedness between pairs of points as a function of
diversity cell values, are therefore presented, along with an
gressing observed versus predicted values. Maps of the krig-
distance. In order to calculate the empirical variogram, pairs
overlay of the lower third of the diversity residuals for marine
ing standard error were also generated and used to restrict
of points must be binned by distance, and an average value
the analysis extent. In order to avoid unsupported extrapola-
(diversity, density) calculated for all pairs within a given bin.
tion into poorly sampled areas, the interpolated maps were
The size of the bin is referred to as the lag size. A variogram
clipped to remove areas of higher standard error. Interpolated
model is fit to the empirical variogram and its parameters
maps were clipped so that only grid cells for which the stan-
are later used in interpolation. Empirical variograms were
dard error was in the lowest 20% were used for subsequent
calculated using the default lag size and number, as well
display and analysis.
as for 1km, 5km, and 10km lag sizes. The appropriate lag
size and number of lags were chosen to optimize variogram
5) Correcting for Effort: Total effort was calculated as the to-
coherence. Directional variograms were then plotted to inves-
tal length of trawls falling within a grid cell for the NMFS trawl
tigate possible anisotropy not removed by detrending. Strong
data and as the total area surveyed within a grid cell for the
anisotropy was found only for the fish density data, and ac-
marine bird survey data. Although diversity is less related to
cordingly a geometrically anisotropic variogram model was fit
effort than other metrics, some significant correlation (p<0.05)
to this data set. Spherical variogram models were fit to the
between the two was found for both fish and birds. When such
empirical variograms. A spherical model was chosen based
Gulf of the Farallons National Marine Sanctuary
130
Section 3: INTEGRATION OF ANALYSES
feature acting as a biogeographic boundary, where oceanic and shelf species of
ABOUT THESE MAPS
birds show maximum overlap. The region seaward of the Farallon Islands displays
Figure 80a depicts interpolated marine bird diversity throughout the study region.
Density
Diversity high diversity not only because of its proximity to the shelf break, but also because
The top 20% of predicted diversity is bounded by a thin black line. Because bird
many species of birds breed on these islands and would not otherwise be found so
diversity was significantly correlated with survey effort, we have also provided a
High
High far offshore.
mask (cross hatched area) indicating where residual estimates provided evidence
that diversity was lower than expected given the amount of effort spent there (re-
Density. A large region of high (top 20th percentile) marine bird density exists ad-
siduals were among the lowest third). Interpret with caution in this area, as the
Low
Low
jacent to and shoreward of the marine bird diversity hot spot. This density hot spot
expression of high diversity under the mask may actually be an artifact of high
Top 20%
Top 20%
covers most of the shelf waters of all three sanctuaries, from Point Sur in the south
sampling effort. Figure 80b depicts interpolated bird density. Again, the top 20% of
Residual to midway between Bodega Head and Point Arena in the north. The density hot
this estimate is bounded by a thin black line. No statistical relationship was found
0 50 100
Mask spot extends into Monterey Bay. Major regions of overlap between marine bird di-
between density and effort; therefore, no residual mask is provided for this model.
Kilometers
versity and density occur along the shelf break. An additional density hot spot exists
Figures 80a and b have both been clipped using the standard error estimate for
off of Morro Bay to the south of the Monterey Bay NMS. There is some indication of
the interpolated surfaces (access these data on the CD-ROM). This was done to
high marine bird diversity in this region as well.
avoid unsupported extrapolation into poorly sampled areas. Figure 80c depicts the
top 20th percentile for diversity and density and the overlap between them.
A total of 60,000 square kilometers were modeled for bird density, with approxi-
a b mately 10,000 square kilometers classified in this analysis as a hot spot. Approxi-
DATA SOURCES
mately 28% of the entire modeled surface fell inside the boundaries of the three
R.G. Ford and J.L. Casey. 2003. CDAS Density Maps for Marine Birds off North/
National Marine Sanctuaries; however, 84% (8,962 km2) of the high marine bird
Central California, developed for NOAA’s National Centers for Coastal Ocean Sci-
Hot Spots density hot spot was found in the sanctuaries. The proportion of high density inside
ence. Portland, Oregon.
sanctuaries suggests that the boundaries include most areas of high density.
Top 20% Diversity METHODS
Summary. Patterns of bird diversity and density exhibited distinct spatial patterns,
See "Data and Analysis" section.
Top 20% Density
with diversity concentrated from the slope seaward, and density from the slope
shoreward. The overlap of these estimates mainly occurs along the shelf break; an
RESULTS AND DISCUSSION
Overlap of both
Diversity and Density area of high meso-scale bathymetric complexity. It is interesting to note that marine
Species Diversity. The interpolated maps of marine bird diversity show one con-
bird diversity exhibited a statistically significant positive correlation (r=0.33, p<0.0001)
tinuous area of high diversity along the continental slope, and, to a lesser extent,
0 25 50 100
with bathymetric variance (Figure 81a)(see section 2.1.1 for details on this estimate).
along the shelf between Point Arena and Point Sur. Within this area, diversity ap-
Density, on the other hand, exhibited a strong negative correlation (r=0.65, p<0.0001)
pears highest on, and seaward of, the Farallon Escarpment in the northwestern
Kilometers
with depth rather than bathymetric variance (Figure 81b).
corner of the Monterey Bay NMS (Pioneer Canyon), and off of the region between
Point Lobos and Point Sur (refer to locator map). Since marine bird diversity was
correlated with survey effort, much of the hot spot region coincides with areas of
high survey effort. The Farallon Escarpment, in particular, received a dispropor-
tionate amount of survey effort. When the map of interpolated residuals was over-
layed on marine bird diversity, some parts of the diversity hot spot (top 20%) fell in
a region of low (bottom third) residual diversity (the masked portion of Figure 79a).
This indicates that the high estimated diversity in the Farallon Escarpment is due,
at least in part, to high sampling effort. The portion of the marine bird diversity hot
spot between Point Lobos and Point Sur coincides with a region of high residual
diversity. This indicates that diversity in this region was both high and higher than
expected given relatively moderate sampling effort.
Overall, a total of 62,000 square kilometers were modeled for bird diversity. Of a b
that, roughly 12,000 square kilometers were classified as a hot spot (top 20% of
400
0 m estimated diversity). Approximately 28% of the entire modeled surface, and 58%
(7,158 km2) of the hot spot, fell inside the boundaries of the three National Marine
Sanctuaries. This disproportionate allocation of high diversity inside sanctuaries
50 m
Figure 81. The left graphic (a) shows the strong positive relationship observed be-
indicates that current boundaries generally incorporate areas of high regional di-
20 1000 m tween bird diversity and bathymetric variance, while the right graphic (b) shows a
c
20
30
00 versity. A considerable area of high diversity can be found seaward of the northern
0
00
strong negative relationship between bird density and depth. In both cases, the esti-
m
m
m
Monterey Bay NMS, and seaward of the entire Gulf of the Farallones and Cordell mate (diversity and density) have been classified into 20th percentiles.
Bank NMS boundaries. As mentioned in section 2.2, the persistence of high spe-
cies diversity along the shelf break may be attributed to this natural physiographic
Figure 80. Estimated diversity (a), density (b), and hot spots (top 20%) (c) for
marine birds.
131
Section 3: INTEGRATION OF ANALYSES
ABOUT THESE MAPS Density. Interpretation of the fish density maps suffers from the same problems
Figure 82a depicts estimated demersal fish diversity throughout the study region. (i.e. lack of data to the west of sanctuary boundaries and high spatial variability) as
Diversity Density Unlike the mean diversity mapped in section 2.1.1, this surface was generated us- those encountered for diversity. In addition, densities tend to emphasize the distri-
ing estimates of total diversity for each 5 minute grid cell. The top 20% of predicted bution of common numerically dominant species. High density areas of the map
High High diversity is bounded by a thin black line. Though fish diversity was significantly can be divided into four major hot spots (top 20%). One hot spot occurs on and to
correlated with survey effort in this model, high residual values overlapped areas the southeast of Cordell Bank. A second hot spot is found off of Point Reyes. The
of highest (top 20%) estimated diversity. This indicates that interpolated areas of largest density hot spot covers a large portion of the shelf to the north of Monterey
Low Low
highest diversity showed little effect of effort. As such, no residual mask is pro- Canyon, the entire area of Monterey Bay, and near shore waters south to Point Sur.
Top 20% Top 20% vided. Figure 82b depicts fish density, and, like the diversity map, is based on an Although portions of this hot spot are found over Monterey Canyon, this fact should
interpolation of total density (individuals per area swept (km2)) within each 5 min- be incorporated with caution since the deep canyon waters themselves were not
0 50 100
ute grid cell. The top 20% of this estimate is bounded by a thin black line. Figures sampled. The fourth hot spot is found to the south of Monterey Bay NMS and cov-
Kilometers
82a and 82b were clipped using the standard error estimates for the respective ers a substantial area of the shelf from Point Estero to Point Sal. This final hot spot
interpolated surfaces (access these data on the CD-ROM). This was done to avoid is the largest region of high fish density within the mapped area that falls outside of
unsupported extrapolation into poorly sampled areas. Figure 82c depicts the top Sanctuary boundaries and overlaps with a much smaller fish diversity hot spot to
20% for diversity and density, and the overlap between the two. the north.
DATA SOURCES A total of 27,000 square kilometers were modeled for fish density, with approximate-
Species diversity was calculated using NMFS shelf and slope trawl data collected ly 5,200 square kilometers classified in this analysis as a hot spot. Approximately
at depths between 50-1280 meters, between June and November, every third 54% of the entire modeled surface fell inside the boundaries of the three National
Marine Sanctuaries; however, 76% (4,041 km2) of the hot spot was contained
Hot Spots year from 1977-2001. For details on trawl methods see Lauth (2001), Shaw et al.
(2000), Turk et al. (2001), and Williams and Ralston (2002). within the sanctuary boundaries.
Top 20% of Diversity
METHODS Summary. Patterns in both fish diversity and density appear in many cases to be
Top 20% of Density See "Data and Analysis" section. linked to known oceanographic features already mentioned in previous sections. For
example, the northernmost diversity hot spot, and some parts of the density hot spot,
Overlap of both
RESULTS AND DISCUSSION straddle the shelf break, an area known to concentrate a variety of marine fauna (Kim,
Diversity and Density
Diversity. Interpretation of the interpolated maps of fish diversity is hindered by 2000; Adams et al., 1995; Yoklavich et al., 2000). The quickly changing depths of the
the lack of available data west of the sanctuary boundaries and the high spatial shelf break and slope may also increase diversity by allowing fish with overlapping
0 25 50 100
variability within the data. Despite these limitations, three hot spots (top 20%) of bathymetric preferences to coexist. Both diversity and density also appear high near
Kilometers
fish diversity are apparent: The northernmost hot spot is centered on Cordell Bank well known upwelling regions, including Point Sur, near Point Año Nuevo, and near
within the northwestern corner of the Cordell Bank NMS, and extends northward Cordell Bank. Although the majority of the fish diversity and density hot spots fall
along the continental slope outside of sanctuary boundaries to Point Arena. Its within sanctuary boundaries, this fact should be interpreted with caution since the
northern and western extent cannot be determined with the available trawl data sanctuary area represents approximately half of the mapped region for both of these
as sampling stopped along the edge of high predicted diversity. Extrapolation to variables. Areas of high diversity and density outside of the sanctuary boundaries
the north and west of this area indicates that high diversity may continue beyond exist to the north and south. Diversity and density to the west of sanctuary boundar-
the available data. A second area of high diversity is centered at the boundary ies cannot be adequately assessed with the available data.
between the Gulf of the Farallones NMS and the Monterey Bay NMS. The area
extends in a southeasterly direction past Point Año Nuevo and ends off northern
Monterey Bay. The southernmost hot spot is located between Point Sur and Lopez
Point and covers the inshore portions of Sur and Lucia Canyons. Portions of this
last hot spot, however, were poorly sampled. There is some evidence of an addi-
tional hot spot in the shallow waters (<200m) straddling the southern boundary of
the Monterey Bay NMS and extending into Morro Bay.
400
0 Overall, a total of 27,000 square kilometers were modeled for fish diversity. Of
m
that, roughly 5,400 square kilometers were classified as a hot spot (top 20th per-
centile of estimated diversity). Approximately 53% of the entire modeled surface
50 m
fell inside the boundaries of the 3 National Marine Sanctuaries, with 67% (3,675
20 100 0 m
20
30
km2) of the hot spot contained within the sanctuaries. Much of the remaining 33%
00
0m
00
m
m
of high diversity extends along the shelf break north of the Cordell Bank NMS to
Point Arena.
Figure 82. Estimated diversity (a), density (b), and hot spots (top 20%) (c) for fish.
132
Section 3: INTEGRATION OF ANALYSES
ABOUT THIS MAP tion of the Northern overlap of diversity that continues along
124°W 123°W 122°W 121°W
Figure 83 shows the overlap of diversity hot spots for birds the slope for 40-50 km beyond the northern boundary of
and fishes. As described previously, hot spots were defined
39°N
39°N
Cordell Bank National Marine Sanctuary.
Integration: Option 1 as the top 20% of diversity estimated through the spatial
modeling process (kriging). Also shown is the most recent Overlap in diversity hot spots occurs in both slope and shelf
estimated distribution of kelp beds within the study area. waters. The northernmost hot spot is clearly associated with
Species Diversity Although no specific analysis of biodiversity was done for the slope. The southernmost hot spot is also found in an area
kelp communities, it is well documented that these habitats of rapidly changing bathymetry off of Point Sur. A large portion
support a rich and diverse faunal assemblage (Abbott and of the central diversity hot spot, however, occurs over primar-
Hollenberg, 1976; VanWagenen, 2001; McLean, 1962; Fos- ily soft bottom shelf regions. The ecological linkages report
Legend ter and Schiel, 1985; Harrold et al., 1988; Thorson 1950; (see CD-ROM) cites a considerable volume of literature that
Randall 1965; Dayton 1984; Dean et al., 1984; Ebeling et al., describes slope communities as diverse, with well document-
Fish - Top 20% Diversity 1985; Harrold and Reed, 1985; Miller and Geibel 1973; King ed trophic interactions between birds and fishes. The authors
38°N
38°N
and DeVogelaere, 2000; Van Blaricom and Estes, 1988). report that spatial and temporal distribution of plankton is
Birds - Top 20% Diversity
Because of this, we have chosen to include kelp distributions thought to affect the distributions of many fishes and marine
in all of the integrated hot spot maps. The kelp distributions birds. In particular, marine birds aggregate in regions with ex-
Overlap of both
depicted here represent only a "snapshot" view of a highly tremely high plankton density, such as Cordell Bank, the Gulf
Fish and Birds
dynamic feature. of the Farallones, and parts of Monterey Submarine Canyon
Kelp Beds (1999) (Croll et al., in press). Each of these areas were identified in
DATA SOURCES this analysis as being biodiverse. Furthermore, squid, a pri-
0 10 20 40 60 80
Species diversity for fishes was estimated using NMFS shelf mary food source for numerous fishes and birds, concentrate
and slope trawls data collected at depths between 50-1280 in areas of high plankton productivity (Mais, 1972; Roper and
Kilometers
meters, between June and November, every third year from Young, 1975; Anderson, 1977; Pearcy et al., 1977; Anderson
37°N
37°N
1977-2001. For details on trawl methods see Lauth (2001), and Morel, 1978; Cailliet et al., 1979), where they consume
Shaw et al. (2000), Turk et al. (2001), and Williams and euphausiids and copepods (Karpov and Cailliet, 1979; Chen
Ralston (2002). Species diversity for birds was estimated us- et al., 1996). This provides further evidence that trophic set-
ing data provided by R.G. Ford Consulting and H.T. Harvey ting might be partially responsible for the expression of high
and Associates. 1999 kelp distribution data were provided by diversity in areas of upwelling.
California Department of Fish and Game.
Summary
METHODS 1) Diversity overlap between birds and fishes appear to be
See "Data and Analysis" section. correlated to known centers of coastal upwelling.
2) Overlap occurs in slope and shelf waters.
36°N
36°N
RESULTS AND DISCUSSION 3) Much of the expression of high diversity may be related to
All three regions of high fish diversity show some overlap the trophic setting in these areas rather than directly to the
with the regions of high bird diversity. An interesting result physical factors that characterize these areas.
of this analysis is that all regions of overlap occur near well
known upwelling centers (Huyer and Kosro, 1987; Brink
and Cowles, 1991; Kelly, 1985; Breaker and Mooers, 1986;
Breaker and Gilliland, 1981; Tracy, 1990; Schwing et al.,
1991; Breaker and Broenkow, 1994; Rosenfeld et al., 1994);
including the area surrounding Cordell Bank, the area south
of the Farallones (off point Año Nuevo), and directly adjacent
35°N
35°N
to point Sur. The northernmost fish diversity hot spot overlaps
the marine bird diversity hot spot from Cordell Bank north
400
to approximately midway between Bodega Head and Point
0 m
20 Arena. The seaward half of the central fish diversity hot spot
50 m
00
20
1000
m overlaps with the area of high marine bird diversity within the
m
0
30
m
00
Gulf of the Farallones NMS and the Monterey Bay NMS. The
m
northern half of the southernmost fish hot spot overlaps the
124°W 123°W 122°W 121°W
southern tip of the marine bird hot spot. There is a small por-
Figure 83. Integration option 1, diversity hot spots (top 20%) for fish and marine birds. Coastal kelp bed areas are also shown.
133
Section 3: INTEGRATION OF ANALYSES
Summary
ABOUT THIS MAP
124°W 123°W 122°W 121°W
1) There is considerable overlap between areas of high bird
Figure 84 shows the overlap of density hot spots for fish and
and fish density.
birds. As described previously, hot spots were defined as the
39°N
39°N
Integration: Option 2
2) Density maps should be interpreted with caution due to
top 20% of density estimated through the spatial modeling
their inherent biases toward numerically dominant species.
process. Also shown is the most recent estimated distribu-
tion of Kelp beds within the study area. Although no specific
Species Density analysis of density was done for kelp communities, it is well
documented that these habitats support a productive faunal
assemblage (Abbott and Hollenberg, 1976; VanWagenen,
2001; McLean, 1962; Foster and Schiel, 1985; Harrold et al.,
Legend 1988; Thorson, 1950; Randall, 1965; Dayton, 1984; Dean et
al., 1984; Ebeling et al., 1985; Harrold and Reed, 1985; Miller
and Geibel, 1973; King and DeVogelaere, 2000; Van Blaricom
Fish - Top 20% Density
38°N
38°N
and Estes, 1988). Because of this, we have chosen to include
Birds - Top 20% Density kelp distributions in all of the integrated hot spot maps. The
kelp distributions depicted here represent only a "snapshot"
Overlap of both
view of a highly dynamic feature.
Fish and Birds
DATA SOURCES
Kelp Beds (1999)
Species density for birds was estimated using data provided
by R.G. Ford Consulting and H.T. Harvey and Associates.
0 10 20 40 60 80
1999 Kelp distribution data were provided by California De-
Kilometers
partment of Fish and Game.
37°N
37°N
METHODS
See "Data and Analysis" section.
RESULTS AND DISCUSSION
Nearly all of the fish density hot spot is coincident with the two
areas of high bird density. The distributions for both metrics
are generally confined to the shelf (<200m) with the notable
exception of Monterey Canyon which appears as a density
hot spot for both groups. Although the majority of the hot spots
36°N
36°N
for fish and bird density fall within sanctuary boundaries, it is
notable that overlapping hot spots for both groups exist to the
south of Monterey Bay NMS. The pattern of marine bird den-
sity is dominated by the distributions of the Common Murre
(Uria aalge) and Sooty Shearwater (Puffinus griseus) be-
cause they are so abundant. Fish density reflects a somewhat
more balanced species composition. Among the most numeri-
cally dominant fish species are shortbelly rockfish (Sebastes
jordani) and Pacific hake (Merluccius productus).
35°N
35°N
Because the modeled distribution of bird density is dominated
by two species and all density maps emphasize common spe-
cies, these maps should be interpreted with caution. While
400
0 m
the density interpolation for birds closely approximates what
20
50 m
00
20
1000 is generally observed in the wild, it is heavily biased towards a
m m
0
30
m
few numerically dominant species. This fact may tend to over-
00
m
shadow the density distribution for rare and/or endangered
124°W 123°W 122°W 121°W
species.
Figure 84. Integration option 2, density hot spots (top 20%) for marine birds and fish. Coastal kelp bed areas are also shown.
134
Section 3: INTEGRATION OF ANALYSES
ABOUT THIS MAP north and south. The westward extent of important areas for
124°W 123°W 122°W 121°W
Figure 85 shows the overlap of options one and two. The top fish cannot be determined from the available trawl data, and
20% for bird diversity and density were combined, as were the may extend beyond the pictured hot spots. Since Option 3 is
39°N
39°N
Integration: Option 3 top 20% of fish diversity and density. This is the most inclusive simply a combination of Options 1 and 2, all of the concerns
view of marine bird and fish hot spots and the areas they over- and results for those two sections apply here as well.
lap. Also shown is the most recent estimated distribution of
Diversity and Density kelp beds within the study area. Although no specific analysis Summary
of biodiversity was done for kelp communities, it is well docu- 1) The sanctuary boundaries incorporate much of the highest
mented that these habitats support a rich and diverse faunal diversity and highest density areas within the region.
assemblage (Abbott and Hollenberg, 1976; VanWagenen, 2) Many of these biologically important regions coincide with
Legend 2001; McLean, 1962; Foster and Schiel, 1985; Harrold et al., known oceanographic and bathymetric features, such as up-
1988; Thorson, 1950; Randall, 1965; Dayton, 1984; Dean et welling regions, areas of high bathymetric variance, and the
Fish - Top 20% al., 1984; Ebeling et al., 1985; Harrold and Reed; 1985, Miller continental shelf break.
38°N
38°N
Diversity and Density and Geibel; 1973, King and DeVogelaere, 2000; Van Blaricom 3) Regions of high diversity and high density outside of the
and Estes, 1988). Because of this, we have chosen to include current sanctuary boundaries exist to the north, across much
Birds - Top 20%
kelp distributions in all of the integrated hot spot maps. The of the shelf and slope, and to the south, in near-shore wa-
Diversity and Density
kelp distributions depicted here represent only a "snapshot" ters.
Overlap of both view of a highly dynamic feature. 4) Uneven sampling effort across the study region and lack
Fish and Birds of trawl samples to the west of the sanctuary boundaries limit
DATA SOURCES the scope of any integrated biogeographic assessment.
Kelp Beds (1999)
Species diversity for fishes was estimated using NMFS shelf 5) Known limitations and biases of the two metrics (diversity and
and slope trawls data collected at depths between 50-1280 density) exist and are discussed elsewhere within this section
meters, between June and November, during every third (Section 3 – Integration).
0 10 20 40 60 80
37°N
37°N
year from 1977-2001. For details on trawl methods see Lauth
Kilometers
(2001), Shaw et al. (2000), Turk et al. (2001), and Williams
and Ralston (2002). Species diversity and density for birds
was estimated using data provided by R.G. Ford Consulting
and H.T. Harvey and Associates. Kelp distribution data were
provided by California Department of Fish and Game.
METHODS
See "Data and Analysis" section.
36°N
36°N
RESULTS AND DISCUSSION
The majority (71%) of the fish hot spot is coincident with the
much larger bird hot spot. The greater area of the bird hot spot
(~19,000 km2 for birds compared to ~10,000 km2 for fish) is
due to the greater spatial extent of the bird survey data. Major
areas of overlap occur in the following regions:
1) from Cordell Bank and the northwest corner of the Gulf of
the Farallones NMS north to approximately midway between
Bodega Head and Point Arena,
35°N
35°N
2) off Point Reyes,
3) shelf waters from the southern boundary of the Gulf of the
400
Farallones NMS south to Point Sur, including Monterey Bay,
0 m
and
20
50 m
00
20
1000
4) near shore waters off of Point Buchon.
m m
0
30
m
00
m
Although the majority of the regions that were identified as
124°W 123°W 122°W 121°W
hot spots for fish and birds occur within Sanctuary waters,
Figure 85. Integration option 3, diversity and density, hot spots (top 20%) for fish and marine birds. Coastal kelp bed areas are also there are hot spots beyond Sanctuary boundaries to the
shown.
135
Section 3: INTEGRATION OF ANALYSES
REFERENCES Chen, D. S., G. Van Dykhuizen, J. Hodge, and W. F. Gilly. Karpov, K. A., and G. M. Cailliet. 1979. Prey composition of Pearcy, W. G., E. E. Krygier, R. Mesecar, and F. Ramsey.
Abbott, I. A., and G. J. Hollenberg. 1976. Marine algae of Cali- 1996. Ontogeny of copepod predation in juvenile squid (Lo- the market squid, Loligo opalescens Berry, in relation to depth 1977. Vertical distribution and migration of oceanic micronek-
fornia. Stanford University Press. Stanford, California. ligo opalescens). Biological Bulletin 190(1): 69-81 and location of capture, size of squid, and sex of spawning ton off Oregon. Deep-Sea Research 24: 223-245.
squid. California Cooperative Oceanic Fisheries Investiga-
Adams, P. B., J. L. Butler, C. H. Baxter, T. E. Laidig, K. A. Croll, D. A., B. Marinovic, S. Benson, F. P. Chavez, N. Black, tions Reports 20: 51-57. Randall, J. E. 1965. Grazing effects on sea grasses by her-
Dahlin, and W. W. Wakefield. 1995. Population estimates of R. Ternullo, and B. R. Tershy. In press. From wind to whales: bivorous reef fishes in the West Indies. Ecology 46: 255-260.
Pacific coast groundfishes from video transects and swept trophic links in a coastal upwelling system. Kelly, K. A. 1985. The influence of wind and topography on the
area trawls. Fishery Bulletin 93: 446-455. sea surface temperature patterns over the northern California Roper, C. F. E., and R. E. Young. 1975. Vertical Distribution of
Dayton, P. K., V. Curries, T. Gerrodette, B. Keller, R. Rosen- slope. Journal of Geophysical Research 90: 11783-11798. Pelagic Cephalopods. Smithsonian Contributions to Zoology
Bakun, A. 1996. Patterns in the Ocean: Ocean processes thal, and D. Van Tresca. 1984. Patch dynamics and stability 209. 51 pages.
and marine population dynamics. California Sea Grant Col- of some southern California kelp communities. Ecological Kim, S. 2000. Patterns in Benthic Communities Within the
lege System and the National Oceanic and Atmospheric Monographs 54: 253-289. Sanctuary. Presented at the Monterey Bay National Marine Rosenfeld, L. F. Schwing, N. Garfield, and D. E. Tracy. 1994.
Administration. Sanctuary Symposium: Sanctuary Currents 2000: Causes of Bifurcated flow from an upwelling center: a cold water source
Dean, T. A., S. C. Schroeter, and J. Dixon. 1984. Effects of Ecosystem Change: Natural or Human? March 18, 2000. Co- for Monterey Bay. Continental Shelf Research 14: 931-964.
Barber, R. T., and F. P. Chavez. 1983. Biological consequenc- grazing by two species of sea urchins (Strongylocentrotus conut Grove, Santa Cruz, CA
es of El Niño. Science 222: 1203-1210. franciscanus and Lytechinus anamesus) on recruitment and Ryther, J. H. 1969. Photosynthesis and fish production in the
survival of two species of kelp (Macrocystis pyrifera and King, A., and A. DeVogelaere. 2000. Monterey Bay National sea. Science 166: 72-76.
Bolin, R. I., and D. P. Abbott. 1963. Studies on the marine Pterygophora californica). Marine Biology 78: 301-313. Marine Sanctuary final Kelp Management Report. Back-
climate and phytoplankton of the central coast area of Cali- ground, Environmental Setting, and Recommendations. Oc- Schwing, F. B., D. M. Husby, N. Garfield, and D. E. Tracy.
fornia, 1954-1960. California Cooperative Oceanic Fishery Ebeling, A. W., D. R. Laur, and R. J. Rowley. 1985. Severe tober 3, 2000. 1991. Mesoscale oceanic response to wind events off central
Investigations Report 9: 23-45. storm disturbances and reversal of community structure in a California in spring 1989: CTD surveys and AVHRR imagery.
southern California kelp forest. Marine Biology 84: 287-294. Lauth, R. R. 2001. The 2000 Pacific west coast upper California Cooperative Oceanic Fisheries Investigations Re-
Breaker, L. C., and R. P. Gilliland. 1981. A satellite sequence continental slope trawl survey of groundfish resources off ports 32: 47-62.
of upwelling along the California coast. In Richards, F. A. Foster, M. S., and D. R. Schiel. 1985. The ecology of giant Washington, Oregon and California: estimates of distribution,
[Ed.]. Coastal Upwelling. American Geophysical Union. kelp forests in California: a community profile. U.S. Fish and abundance, and length composition. NOAA Tech. Memo. Shannon, C.E., and W. Weaver. 1949. The mathematical
Washington, D. C. Pp. 97-94. Wildlife Service Biological Report 85(7.2): 152 pages. NMFS-AFSC-120. 284 pp. theory of communication. University of Illinois Press, Urbana,
IL 125p.
Breaker, L. C., and C. N. K. Mooers. 1986. Oceanic variability Gotway, C.A, and L. J. Young. 2002. Combining incompatible Levine, Ned. CrimeStat II: A Spatial Statistics Program for
off the central California coast. Progress in Oceanography spatial data. Journal of the American Statistical Association the Analysis of Crime Incident Locations (version 2.0). Ned Shaw, F.R., Wilkins, M.E., Weinberg, K.L., Zimmermann, M.,
17: 61-135. 97(458): 632-648. Levine and Associates, Houston, TX, and the National Insti- and R. R. Lauth. 2000. The 1998 Pacific west coast bottom
tute of Justice, Washington, DC. May 2002. trawl survey of groundfish resources: estimates of distribu-
Breaker, L. C., and W. W. Broenkow. 1994. The circulation Guan, W., Chamberlain, R.H., Sabol, B.M. and P.H. Doering. tion, abundance, and length and age composition. U.S. De-
of Monterey Bay and related processes. Oceanography and 1999. Mapping submerged aquatic vegetation with GIS in the Ludwig, J.A., and J.F. Reynolds. 1988. Statistical ecology: partment of Commerce, NOAA Technical Memorandum
Marine Biology 32: 1-64. Caloosahatchee Estuary: Evaluation of different interpolation A primer on methods and computing. John Wiley and Sons. NMFS-AFSC-114, 138 p.
methods. Marine Geodesy 22(2). 69-92. New York, NY.
Brink, K. H., and T. J. Cowles. 1991. The coastal transition Thorson, G. 1950. Reproductive and larval ecology of marine
zone program. Journal of Geophysical Research 96: 14637- Harrold, C., and D. C. Reed. 1985. Food availability, sea Mais, K. F. 1972. California Department of Fish and Game bottom invertebrates. Biological Review 25: 1-45.
14647. urchin grazing and kelp forest community structure. Ecology Sea Survey Cruises 1971. California Cooperative Oceanic
66(4): 1160-1169. Fisheries Investigations Data Reports 21: 1-132. Tracy, D. E. 1990. Source of cold water in Monterey Bay ob-
Cailliet, G. M., K. A. Karpov, and D. A. Ambrose. 1979. Pe- served by AVHRR satellite imagery. Masters Thesis. Naval
lagic assemblages as determined from purse seine and large Harrold, C., J. Watanabe, and S. Lisin. 1988. Spatial variation Malone, T. C. 1971. The relative importance of nannoplankton Postgraduate School. Annapolis, Maryland. 125 pages.
midwater trawl catches in Monterey Bay and their affinities in the structure of kelp forest communities along a wave expo- and netplankton as primary producers in the California Cur-
with the market squid, Loligo opalescens. California Coopera- sure gradient. Marine Ecology 9(2): 131-156. rent system. Fisheries Bulletin 69: 799-820. Turk, T. A., T. Builder, C. W. West, D. J. Kamikawa, J. R. Wal-
tive Oceanic Fisheries Investigations Reports 20: 21-30. lace, R. D. Methot, A. R. Bailey, K. L. Bosley, A. J. Cook, E.
Huyer, A., and P. M. Kosro. 1987. Mesoscale surveys over the McLean, J. H. 1962. Sublittoral ecology of kelp beds of the L. Fruh, B. H. Hormess, K. Piner, H. R. Sanborn, and W. W.
Chavez, F. P. 1995. A comparison of ship and satellite chlo- shelf and slope in the upwelling region near Point Arena, Cali- open coast area near Carmel, California. Biological Bulletin Wakefield. 2001. The 1998 Northwest Fisheries Science Cen-
rophyll from California and Peru. Journal of Geophysical Re- fornia. Journal of Geophysical Research 92: 1655-1681. 122: 95-114. ter Pacific west coast upper continental slope trawl survey of
search 100: 24855-24862. groundfish resources off Washington, Oregon, and California:
Kaluzny, S.P., Vega, S.C., Cardoso, T.P., and A.A. Shelly. Miller, R. J., and J. J. Geibel. 1973. Summary of blue rockfish Estimates of distribution, abundance, and length composition.
Chavez, F. P. 1996. Forcing and biological impact of onset of 1998. S+SpatialStats: Users Manual for Windows and UNIX. and lingcod life histories: a reed ecology study; and giant kelp U. S. Department of Commerce, NOAA Technical Memoran-
the 1992 El Niño in central California. Geophysical Research MathSoft Inc. Macrocystis pyrifera experiments in Monterey Bay, California. dum NMFS-NWFSC-50, 122 p.
Letters 23: 265-268. Fish and Game Fishery Bulletin 158: 1-137.
136
Section 3: INTEGRATION OF ANALYSES
Van Blaricom, G. R., and J. A. Estes [Eds.]. 1988. The Com-
munity Ecology of Sea Otters. Springer-Verlag. Berlin, Ger-
many.
Van Wagenen, R. F. 2001. California Coastal Kelp Resources,
Summer 2000. Final Report to the Monterey Bay National Ma-
rine Sanctuary.
Williams, E.H. and S. Ralston, Distribution and co-occur-
rence of rockfishes over trawlable shelf and slope habitats
of California and southern Oregon. Fisheries Bulletin, 2002.
100: p. 836-855.
137
Section 4: DATA SOURCES AND GAPS
INTRODUCTION Within this library of data, some sets emerged as
Table 27. Matrix of data sets and their associated characteristics that were used or referenced in the biogeographic assessment.
This section addresses a secondary objective primary data sources, while others contributed
Source Org Strengths of Data for Biogeographic Constraints of Data for Biogeographic
Data Set Target Info. Dates Samples Depth Range
of the project: the acquisition and assessment to the project in terms of providing contextual
and/or PI Assessment Assessment
of available comprehensive data for the study information, reference or validation.
NMFS Long time-series, wide spatial extent, Sampling only in summer season, some
Triennial Shelf Fish, (Alaska & 1977-1998 every 3 fairly good depth range, abandoned trawls suspected of being off-bottom,
area and the identification of data gaps in such n=994 55-500m
Trawl Data Invertebrates Northwest years, June-Aug only trawl areas suggest areas of high rocky areas undersampled due to threat
information. In addition, suggestions are made The data sets that ultimately proved most use-
FSC) rugosity of gear damage.
for prioritizing future research efforts to generate ful in this undertaking are summarized in Table
1991, 1997, 1999,
Slope Trawl Fish, 10+ years of data, wide spatial Sampling for only 5 months (July-Nov.);
data that would be especially valuable for future 27 below, which provides information about the
NMFS 2000, 2001, July-Nov n=454 190-1280m
Data Invertebrates extent, good depth range identification of common inverts only
only
biogeographic analyses. source of the data, target information, dates of
NMFS
Rockfish NMFS / 1986-2001, May-June 6-32m Sampling only in May & June, along collection, number of samples, and depth range
n=1548 1500+ tows
Midwater Trawl
juveniles SWFSC only mean=26m transects, targets juv. rockfish
Throughout the project, members of the Bio- when available. Additionally, comments about
Data
provides information about
geography Program contacted numerous the general strengths and constraints of the data
Recreational 1987-1998, 2-360 fm (3-650 No effort data (presence/absence only),
Rockfish CDF&G n=4357 nearshore areas, wide range of
Fish Data continuous m) effort targets Rockfish only
academics, scientists, and agency personnel sets in the context of this analysis are noted in
depths.
who were likely to have knowledge of data separate columns. Full citation information for
1983-1995 (Sonoma)
provides year-round look at kelp in
relevant to the study area, compiling a contact each data set is provided in the list of references
Kelp- year-round but to max extent of
Tom Laidig/ Sonoma 1983-95, quick look at Sampling is sparse, presence/absence
Laidig Data Associated sparse, 1984, 1997, n=43 surveys scuba or kelp
list of over 160 people. Additionally, staff con- that appears at the end of each section utilizing
NMFS Monterey in 84, 97, 2001; data only, very limited spatial extent
Species 2001 (Monterey) May- (<130 ft.)
differentiates juveniles and adults.
sulted the impressive compendium of studies the data.
Oct only
compiled by Monterey Bay Sanctuary staff for n=117,176 Variable reliability re: locations of fishing,
Commercially
their Sanctuary Integrated Monitoring Network CONCLUSIONS AND RECOMMENDATIONS
Commercial 1988-2000, year- (records grouped shoreline to All gear types, data can be sorted by summarized to 10-minute grids (large
Valuable CDF&G
Fishing Data round by species, not trip 4810m gear, long time series. scale), fisheries-dependent, can't sort by
(SIMoN) program, which provides a “blueprint for FOR FUTURE ACTIVITIES
Fishes
or boat) boat or trip/effort
a comprehensive, integrated monitoring network During the course of this project, Biogeography
5 categories, provides the most
to detect natural and human induced changes Program staff gained a unique familiarity with
comprehensive view of benthic
to the Monterey Bay National Marine Sanctuary the data available and the data necessary to
substrates for the study area; in
some places very detailed, high
and its resources” (http://montereybay.noaa.gov/ undertake the analysis. This position enables
resolution; Original data consisted of
Greene et. al.,
). Through extensive consultation with contacts, project team members to make observations
sampling dates seismic-reflection profiles and Based on surface extrapolation of point
Substrate National Sea shoreline to
Sediment unknown, received unknown number sediment/rock sample data collected data, most of the map is low resolution
approximately 62 data sets were investigated about the types of data that, if acquired, could
Composition Grant College ~3500m
12/2002 by California Division of Mines and 1:250,000.
for incorporation into this study. Data sets were improve the biogeographic assessment in the
Project
Geology, USGS, and California
considered in terms of sampling objective, the future. Some of the data sets may exist, but in
Coastal Commission. New data
include multibeam data from MBARI,
extent of their spatial and temporal coverage, the a format that could not, in their current state,
and Center for Habitat Studies at
existing format and ability to be converted into a be incorporated into the project. For example,
Moss Landing Marine Laboratories
GIS layer, the utility of the data compared to the historical benthic infauna data exists, but has not
Bathymetry- Depth, shoreline to best available at the start of the
CDF&G unknown unknown Medium resolution
work involved in its incorporation into the project, yet been converted to digital format or updated
200m Topography 4810m project
provides higher resolution which
and whether or not the Biogeography Program to reflect current taxonomy. Several important
NOAA /
Bathymetry- Depth, shoreline to Better resolution, late availability for this
increases ability to identify smaller
was granted access to the data. In general, team data sets do not, to our knowledge, exist at all
NGDC & unknown unknown
30m Topography 4810m project
areas of high bathymetric variance,
MBARI
members acquired only accessible data sets that or at the necessary spatial scale, but should be
ie. pinnacles and drop-offs
Polygons generated from aerial
had broad spatial extents covering a significant considered priorities for future analytical efforts.
many, polygon Changes since 1999, missing sections,
Kelp Data Kelp Location CDF&G 1989, 1999 surface photos provide literal 'snapshot' of
data doesn't differentiate species
portion of the study area, a large number of They appear in the following list.
kelp along long stretch of coast.
samples that could be georeferenced, and high Monthly composites will smooth out
Sea Surface Sea Surface NOAA / Monthly composites available for
important temporal fluctuations, ie. short
confidence in data quality. •Resolution. Finer thematic and spatial resolu-
monthly composites satellite data surface
Temperature Temperature Coastwatch years 1992 - current. upwelling and relaxation events; corrected
tion on the substrate and bathymetric maps will
for best surface.
Data sets that met the above criteria were be highly advantageous to the analyses based
Dohl, 1,057 cells visited; Data from early 1980s may not represent
surface survey Relatively large spatial coverage;
MMS High Minerals 76,888km of current status and distribution of species;
requested and obtained if possible. Once in- on Habitat Suitability Modeling. Improving the
Cetaceans & 1980-1983, in all three of the shelf, cost-effective, year-round synoptic
Altitude Aerial Management trackline; 10,014 high altitude surveys may not provide
turtles ocean seasons slope and deep surveys for cetaceans over the shelf
house, data were further evaluated in how well resolution for the bathymetry map will help
Surveys Service cell-study-day good characterization of smaller, less
ocean beyond and slope.
(MMS) visits visible species.
they served the objectives of the study, and the identify more small areas of high relief, such as
most useful data were synthesized into a working rocky pinnacles, that are known to be important
Bonnell-PI for 870 cells visited; surface survey Relatively large spatial coverage; Data from early 1980s may not represent
MMS Low
Marine Birds mammals, 1980-1983, in all three 70,114km of of the shelf, cost-effective, year-round synoptic current status and distribution of species;
GIS library. This involved conversion into GIS habitat for some species.
Altitude Aerial
and Mammals Briggs-PI for ocean seasons trackline; 9,306 cell-slope and deep surveys for species over the shelf low altitude surveys may not provide good
Surveys
format, standardization of geographic projection, birds; MMS study-day visits ocean beyond and slope. characterization of rare cetacean species.
and when possible, the aggregation of smaller •Spatial Data for Additional Species. Because
76 cells visited;
EPOCS surface survey Includes outer Calif Current surveys; Spatial coverage is not as robust as with
data sets into a master data layer. of sampling limitations (i.e. mesh or hook size)
Marine Birds 1984-1994, in all three 1,033km of
Shipboard Ainley of the deep better species sightability on a ship the aerial surveys, but sightability for
and Mammals ocean seasons trackline; 77 cell-
or a lack of published life history information,
Surveys ocean than an airplane. some species is better.
study-day visits
some ecologically important species are not
represented in this study. More information
138
Section 4: DATA SOURCES AND GAPS
•Location Verification for Fisheries Data. Verification of the spa- the resource could be used to couple estuarine, coastal and Yoklavich, M.M., G.M. Caillet, R.N. Lea, H.G. Greene, R.M.
should be collected on these species through other means.
tial reporting from commercial fishing logbooks would enable marine ecosystems. Starr, J. de Maringnac, and J. Field. 2002. Deepwater habitat
One example is the pygmy rockfish (Sebastes wilsoni), a small
the full incorporation of this valuable data source. and fish resources associated with a marine reserve: impli-
but abundant species, that does not show up in any of the
•Wider Regional Context. Expansion of the scope of the study cations for marine fisheries in marine ecological reserves
fisheries-dependent data sets. S. wilsoni can be surveyed via
•Abandoned Trawls. Incorporation of NMFS’s study of aban- to include the biogeography of the entire west coast of North research program research results. 1996-2001. California
submersible, as documented in various studies (Yoklavich et. al,
doned trawl locations from the Triennial Surveys. Such infor- America to better understand how the north/central California College Sea Grant Program CD-ROM. La Jolla, CA. 63 pp.
2002; Yoklavich et. al., 2000; Hixon et al., 1991). Other species
mation indicates areas that are difficult to trawl or ‘untrawlable’ region fits into the wider biogeographic context. A precedent
of interest include white shark, pelagic fishes, intertidal species,
due to the fact that the nets repeatedly became caught or torn exists in the West Coast Atlas produced by NOAA’s SEA Divi- Yoklavich, M.M., H.G. Greene, G.M. Caillet, D.E. Sullivan,
krill, marine birds, marine mammals and sea turtles.
during trawl attempts. These locations may indicate areas with sion. Such a document could serve as a blueprint for defining R.N. Lea, and M.S. Love. 2000. Habitat associations of deep-
rocky substrates and high rugosity, which, though still targeted species distributions along the west coast of the continental water rockfishes in a submarine canyon: an example of a
•Survey Methods. Sampling fish communities using a consistent
by hook and line and recreational fishers, are generally pro- U.S. (SAB/NWAFS 1988, SAB 1990). natural refuge. Fish Bulletin, U.S., Vol. 98, pp. 625-641.
sampling method over all substrate/habitat types, based on
tected from trawl fishing methods.
stratified random sampling. Multiple survey methods should be
REFERENCES
employed to ensure representation of important fish species
•Oceanographic Influences. Incorporation of more oceano- Emmett, R.L., S.L. Stone, S.A. Hinton, and M.E. Monaco.
that are not susceptible to current sampling methods.
graphic features and parameters into the analyses, especially 1991. Distribution and abundance of fishes and invertebrates
for birds and mammals. Ephemeral features such as currents, in west coast estuaries, Volume II: Species life history
•Sampling Strategies. Sampling strategies should be tailored
the San Francisco plume, and sources of upwelling could be summaries. ELMR Rep. No. 8. NOAA/NOS/Strategic
to include more life history stages of fish, especially larval
represented in probability maps or by aggregating empirical Environmental Assessments Division. Rockville, MD. 329 pp.
stages.
data by various temporal categories (e.g. by week, event,
month, season, warm/cold period, etc.). Hixon, M.A., B.N. Tissot, and W.G. Pearcy. 1991. Fish As-
•Sampling Strategy. Sampling should also be better spread
semblages of Rocky Banks of the Pacific Northwest [Coquille,
spatially and temporally to:
•Kelp Surveys. Increased frequency of surveys to better moni- Daisy, and Heceta Banks]. A final report by the Department
tor changes in kelp distribution; differentiation between kelp of Zoology and College of Oceanography of Oregon State
•Reduce Effort Disparity. Increased sampling in certain ar-
species. University for the U.S. Department of the Interior, Minerals
eas would help equalize the distribution of sampling effort
Management Service Pacific OCS Office. Contract No. 14-12-
across the study area. Some analyses were confounded
•Life History/Trophic Linkages. Expanded knowledge of life 0001-30445. Camarillo, CA. 410 pp.
by the wide range of sampling effort.
history characteristics, habitat affinities, distribution and
abundance of pelagic prey species, and links between preda- Monaco, M.E., R.L. Emmett, D.M. Nelson, and S.A. Hinton.
•Target Important and Under-Sampled Areas. Increas-
tors and prey species (i.e. hake, krill, and plankton) will help 1990. Distribution and abundance of fishes and invertebrates
ing sampling in important areas that are currently under
describe distributional changes based on trophic linkages and in west coast estuaries, Volume I: Data summaries.
sampled (e.g. the entire near-shore region and the slope
foraging behavior. ELMR Rep. No. 4. NOAA/NOS/Strategic Environmental
area west of Cordell Bank) or in areas of particular man-
Assessments Division. Silver Spring, MD. 332 pp.
agement interest (e.g. boundary regions) would help to
•Life History/Spawning Areas. Incorporation of known spawning
better characterize these areas. The techniques used in
areas will help identify important areas and seasonal changes Nelson, D.M, and M.E. Monaco. 2000. National overview
the integration section can accommodate preferential (i.e.
in distribution for fish. and evolution of NOAA’s Estuarine Living Marine Resources
non-random) sampling in areas of interest.
(ELMR) Program. NOAA Tech. Memo. NOS NCCOS CCMA
•Data QA/QC. Data quality assurance would allow the incor- 144. Silver Spring, MD. 60 pp.
•Describing Effort. Increase consistency in recording effort-re-
poration of some existing data sets that were discarded due to
lated parameters for fisheries trawls (e.g. recording start and
inconsistencies in species coding (e.g. the fisheries data set SAB (Strategic Assessment Branch). 1990. NOAA’s west
end coordinates of trawls) and naturalists’ surveys.
targeting salmon), taxonomic changes (e.g. benthic infauna coast of North America data atlas: Invertebrate and fish
data) or other reasons. volume. National Oceanic and Atmospheric Administration.
•Expert Knowledge. Incorporate additional expert knowledge
Rockville, MD. 111pp.
and data from the fishing community, naturalists (e.g. Mon-
•Expansion of Scope. The scope of this analysis could be
terey Bay Whale Watch cruise data), and recognized experts,
broadened to include adjacent habitats. For example, the in- SABNAFC (Strategic Assessment Branch, and Northwest
especially for areas and time periods where there is little or
teraction between marine and estuarine habitats could possibly and Alaska Fisheries Center). 1988. West coast of North
no data.
be addressed using network analysis of energy flows between America strategic assessment: data atlas: Marine mammal
ecosystems. As part of the Estuarine Living Marine Resources volume. National Oceanic and Atmospheric Administration.
•Data Compatibility. Achieve consensus on the best way to
(ELMR) series, a 2-volume comprehensive data base on the Rockville, MD. 33 pp.
merge aerial and ship-based survey data for birds and mam-
distribution of estuarine fishes and invertebrates in West Coast
mals.
estuaries was completed in 1990-91 (Emmett, et al., 1991;
Monaco, et al., 1990; Nelson and Monaco, 2000). If updated,
139
Section 5: SUMMARY OF BIOGEOGRAPHIC ASSESSMENT
BACKGROUND Ideally, biogeographic assessments utilize significant amounts
The mission of NOAA National Ocean Service’s (NOS) National of data that have been collected over the entire spatial extent of
Marine Sanctuary Program (NMSP) is to serve as the trustee the study area over a long time period. However, such a wealth
for a system of marine protected areas, to conserve, protect, of data is rarely available. In many instances, little information
and enhance their biodiversity, ecological integrity, and exists to accurately characterize the study area or associated
cultural legacy. To assist in accomplishing this mission, the living marine resources. This paucity of comprehensive data
NMSP has developed a partnership with NOAA’s National can limit the efficacy of biogeographic assessments, but
Centers for Coastal Ocean Science (NCCOS) to conduct additional analytical methods can be employed to complement
biogeographic assessments of living marine resources in all the assessment. In addition to analysis of databases, two
National Marine Sanctuaries to characterize and assess the additional tasks were used to conduct the assessment.
distribution of marine resources that occur within and adjacent First, a synthesis of existing information was compiled and
to the sanctuaries. The NMSP and NCCOS’s Biogeography presented in the Ecological Linkages Report to incorporate
Program have developed a five-year plan to implement the qualitative information about species, habitats and ecological
assessments across the system of National Marine Sanctuaries characterization of marine ecosystems and linkages within
(Kendall and Monaco 2003). The biogeographic assessment the study area. Second, species habitat suitability modeling
process as defined in the plan is used to conduct studies that efforts were conducted for fishes to define potential species’
are designed to address research needs and support a wide distributions based on known habitat affinities and physiological
array of sanctuary management decisions. In general, the limitations. The potential species distribution maps are displayed
priority to implement the biogeographic assessments is based as a series of digital maps found on the CD-ROM.
on the need to update sanctuary management plans. Thus, the
joint efforts are systematically proceeding to work with each In addition, a critical component of the assessment process was
sanctuary to provide assessments of species’ distributions and the extensive effort to have the data, analytical approaches,
their associated habitats in a region. and results peer reviewed. Initial results from the suite of
biogeographic analyses were presented to experts familiar
Since establishment, many of the sanctuaries have witnessed with the marine ecosystem off north/central California, as
increased pressure on marine resources from natural and well as to the originators of the data sources, in an attempt
anthropogenic phenomena, including climatic variation and to improve the analyses. The role of expert review and input
degradation of habitats. In order for the NMSP to increase was considerable, and the contributions made by experts have
management capabilities, it is imperative that the spatial and to the boundaries of three contiguous West Coast National beyond the limits of current sanctuary boundaries to place significantly enhanced the assessment. In June 2002, project
temporal distributions of biota and habitats within sanctuaries Marine Sanctuaries. These sanctuaries, Monterey Bay, Gulf study results in the context of north/cental California Coast team members traveled to Seattle, WA and Santa Cruz, CA to
be delineated. Biogeography provides a framework to integrate of the Farallones, and Cordell Bank, are conducting a joint biogeographic patterns. The biogeographic analyses are based discuss and present the results of the Interim Product to West
species distributions and life history data with information on review to update sanctuary management plans. To support on a synthesis of many data sources that were provided by Coast experts (NOAA, 2002). Suggestions were incorporated
the habitats of the region to characterize marine resources the management plan review process, the Biogeography project partners and contributors. Results of this assessment and a Web site was created to further disseminate analytical
in a sanctuary. When the biogeographic data are integrated Program is leading a partnership effort to conduct a robust are being used to assist the NMSP in addressing issues such as products prior to an additional series of meetings. The final
into a Geographic Information System (GIS), it enables users analytical assessment to define important biological areas and evaluating potential modification of sanctuary boundaries and suite of review meetings was held in October 2002 in San
to visualize species’ spatial and temporal distributions and time periods within the region. This document represents the changes in management strategies or administration, based Francisco and Monterey, CA and in Seattle, WA. At that time,
conduct ecological forecasts to assess potential changes in results of the first of two assessment phases. Phase I provides on the principles of biogeography. NOS staff invited members of the scientific community to
species distributions that may result from a variety of natural data, analytical results, and descriptions of ecosystems and review the preliminary results of the biogeographic analyses.
and anthropogenic perturbations. In addition, based on specific their linkages; it also identifies data gaps, and suggests future The biogeographic assessment was formulated around three Comments from the October meetings were compiled and
ecological metrics (e.g., diversity), biologically significant activities to be addressed in Phase II. closely integrated study components: (1) an Ecological Linkages reviewed by project personnel, who either incorporated the
areas can be delineated. This document provides the results Report, (2) biogeographic analyses, and (3) development of GIS experts' suggestions or provided explanations as to why they
of the GIS-based assessment conducted for the National Phase I of this effort was a biogeographic assessment of existing data for incorporation into NMSP’s Marine Information System did not. Thus, the integration of the synthesis of ecological
Marine Sanctuaries off north/central California to initiate data on the distribution and abundance of marine fishes, marine (MarIS). The majority of the results from the assessment are linkage information, statistical analyses of existing databases,
development of a biogeographic assessment capability for birds, marine mammals and their associated habitats. The study presented as a suite of GIS maps to visually display species’ species habitat suitability modeling, and peer review, resulted
the sanctuaries. did not attempt to define biogeographic patterns along the entire biogeographic patterns across the study area. The body of the in this biogeographic assessment product.
U.S. West Coast nor in very near-shore environments (e.g., document provides examples of the entire suite of digital map
BIOGEOGRAPHIC ASSESSMENT OFF NORTH/CENTRAL estuaries). Rather, the study area was restricted to the marine products found on the companion CD-ROM and located on ECOLOGICAL LINKAGES REPORT
CALIFORNIA area from Point Arena in Mendocino County (38˚54’32” N, the the Web at http://biogeo.nos.noaa.gov/products/canms_cd/. Section 1 of the document presents a synopsis of the Ecological
The initial biogeographic assessment outlined in the five-year northern bound) to Point Sal in northern Santa Barbara County The spatial data and additional information, such as digital Linkages Report and provides the context to understand overall
NCCOS/NMSP plan was implemented in the spring of 2001 (34˚54’05” N, the southern bound). The entire study area and species distribution maps and additional details on analytical biogeographic product results, relative to the ecosystems
to conduct a 24-month investigation to assess biogeographic regional maps of the area are depicted in Figures 2-5. This methodologies, are also presented on the CD-ROM. along the California Coast. The bulk of the report describes
patterns of selected marine species found within and adjacent relatively large study area enabled the assessment to extend ecosystems in the region, key species associated with these
140
Section 5: SUMMARY OF BIOGEOGRAPHIC ASSESSMENT
• Diversity and richness can be used to delineate fish hot
ecosystems, and linkages between and among them. In on reviewers' comments on the Interim Product (NOAA 2002), species resulted in information on species’ habitat affinities
spots. While little variation in diversity and richness were
addition, the report presents latitudinal range distributions of feedback from technical review meetings, and peer review that were converted into quantifiable habitat suitability index
explained by depth (r2 =0.04 between both richness and depth
species groups, including algae, invertebrates, fish, marine workshops. Thus, a very difficult step in the project was to values (Monaco et al., 1997). The life history information and
and diversity and depth), trawls with high diversity tended to
birds and marine mammals. These maps provide an overview select and rely on the most appropriate data and analyses to associated species habitat suitability index values are found on
be deeper than trawls with high richness. Trawls with high
of marine species’ distributions and biogeographic transitions characterize the various components of the marine ecosystem the CD-ROM. These derived values were input into an equation
species richness of rockfish (Sebastes and Sebastolobos)
along the entire west coast of North America. The report also that exist in the study area. The inclusion of the GIS-based and used to predict potential species’ distributions based on
followed the 200-meter contour, which approximates the
includes important information on ecosystems not easily products on the companion CD-ROM will enable NOAA staff, their affinity for the mosaic of bathymetry and bottom habitats
break between the shelf and slope.
studied at this large scale via GIS, particularly near-shore advisory councils, and research partners to query data and found throughout the region. The species habitat suitability
communities. The complete report (163 pp.) is on the CD-ROM information relevant for questions and issues that are not models were validated through statistical and spatial analyses,
• Even though richness and diversity are correlated, the
that accompanies this document (Airamé et al., 2003). specifically addressed in this product. using fishery-independent survey data.
maps showed different results. Hot spots in either richness
or diversity were identified in all three sanctuaries. In Cordell
Key West Coast Biogeographic Transitions The first analyses focused on a suite of assemblages analyses • Bottom substrate and water depth were statistically significant
Bank NMS, there was a group of trawls with high richness
• Benthic algae exhibit three major biogeographic transitions to assess the biogeography of fishes and a few macro- variables used to predict the potential distribution of species
near the center of the sanctuary, and another group of
at Point Conception, Puget Sound, and the Gulf of Alaska. At invertebrates. Primary data included fisheries-independent based on their habitat affinities.
trawls with high diversity in the region around the northwest
all latitudes, the average number of algal species increased data, such as those collected by researchers from the National
boundary. There was also a large collection of trawls with
with depth from high to low intertidal and subtidal zones. Marine Fisheries Service (NMFS), and fisheries-dependent • Habitat suitability models for an assemblage of rockfish
either high richness or diversity straddling the boundary
data, such as those collected by the California Department of were developed and results indicated that rocky habitats
between Gulf of the Farallones and Monterey Bay NMS.
• Five major transitions occur in distributions of marine Fish and Game (CDF&G) for recreational fisheries. These data located on the shelf were identified as potential hot spots
There were lines of high diversity along the 200-meter depth
invertebrates found in California waters: at Point Conception, sets, although not spatially or temporally comprehensive, are for adults; whereas mud and sand substrates on the shelf
contour north of CBNMS boundary and from Lopez Point
Monterey Bay, Puget Sound, and off the coasts of British the most robust data sets that exist for the entire region, and were delineated as potentially important habitats for subadult
south to the southern edge of the study area.
Columbia and southeastern Alaska. At all latitudes, greater provide considerable information on the distribution of several rockfish.
numbers of gastropod species occur in the euphotic zone and hundred fish and invertebrate species.
• Starr (1998) addressed the implementation of rockfish no-take
on the continental shelf than on the continental slope. • Map overlays of all species’ HSI models resulted in the
areas with two important recommendations. First, in order
Key Assemblage Analysis Results for Fishes delineation of a broad range of important areas that cover
to properly manage marine ecosystems, fish assemblages
• Pacific coast fishes exhibit two major biogeographic transitions. • Species assemblages and site groups were distinguished the majority of the continental shelf within and adjacent to
must be better understood. Starr stated that once these
A biogeographic transition at the Bering Sea is relatively through 1-Pearson correlation coefficients with average the three sanctuary boundaries.
assemblages are delineated, steps can be taken to ensure
abrupt, corresponding to the northern limit of distributions means clustering technique. Species assemblages from
that each assemblage receives proper management.
of over 100 fish species. A broader biogeographic transition CDF&G recreational, NMFS shelf, and NMFS slope data Key Marine Bird Analytical Results
This study defined assemblages of fishes for near-shore,
occurs along the southern coast of California between Baja sets were more resilient than assemblages from the NMFS The Biogeography Team contracted principal investigators
shelf, slope and midwater ecosystems. The results of the
California and Point Conception. A few minor shifts in fish midwater data set, emphasizing the ephemeral nature David Ainley and Glenn Ford (of H.T. Harvey and Associates
community metrics and species assemblages are displayed
species composition occur between Point Conception and of the midwater environment and the smaller midwater and R.G. Ford Consulting Co.) to work with the NOAA project
in this document as a series of maps and tables.
the Bering Sea, particularly at Monterey Bay. data set. The site groups were displayed spatially in a team to define and assess biogeographic patterns and
GIS and the average frequency of occurrence of species important areas for marine birds (and mammals) found within
• The second recommendation by Starr (1998) was to delineate
BIOGEOGRAPHIC ANALYSES assemblages was calculated to show the interaction between the study area. These experts used regression analysis, GIS
rectangular no-take areas that cover 20-50 km of the coast
Section 2 introduces the methods used to conduct the species assemblages and site groups (i.e., where species and over eight spatial data sets to develop over 50 maps that
and extend west to the edge of the continental shelf. From a
assessment and the results of the biogeographic analyses assemblages were caught). display marine spatial and temporal patterns, and estimated
biogeographic viewpoint, the results of the spatial analyses
of selected marine biota off the north/central California coast. densities and diversity for selected marine birds in the study
coincided with that recommendation, and also identified that
This component of the assessment is the cornerstone of the • The interaction of the site groups with environmental area. The resulting maps and discussion summarize important
deep-slope communities significantly contribute to ground
overall biogeographic product to support the NMSP joint parameters that were not used to create the groups can locations, time periods, and life history information for marine
fish biogeographic patterns. Because assemblages follow
management plan review process. The data, analyses, and be informative about what conditions are affecting species birds in the study area. Phase II of the assessment may include
bathymetry at the scale of this analysis, setting aside an area
supporting information are linked using statistical and GIS tools distribution. Depth was highly significant between site a technical report on the methods and results summarized in
from the coast through the continental slope could protect all
to portray in space and time significant biological areas or “hot groups in all data sets, emphasizing the importance of the Phase I map and tabular products.
demersal species assemblages identified in this study.
spots.” The term “hot spot” is defined based on specific criteria depth in structuring marine biological communities. Analyses
or metrics (e.g., species diversity, high species abundance). comparing the site groups to other environmental parameters • In general, the marine birds of the three sanctuaries are
Key Species Habitat Suitability Model Results
The vast majority of the analytical results are displayed as a (latitude, sediment size, and bathymetric complexity) were dominated in number and biomass by seasonally resident,
Due to limitations in the spatial and temporal extent of data and
series of maps to identify biologically significant areas in the inconclusive, as these parameters often had significant nonbreeding species, such as sooty shearwater, pink-footed
to complement the assemblage analyses of fishes, species
study area. interactions with depth. Latitude was found to have a shearwater, northern fulmar and black-legged kittiwake. The
habitat suitability index (HSI) models were developed (Brown
significant effect only on the midwater assemblages in 1999; richness of the food web is the primary factor that attracts
et al., 2000). This was done primarily to accommodate the
There are many different ways to analyze and organize there were no discernible latitudinal breaks within the other these species to the region.
paucity of empirical data in near-shore areas and to target
biogeographic information; however, to efficiently support the four assemblages.
species of special significance to the sanctuaries. An extensive
management plan process, only a limited number of analytical • Seasonal, interannual and decadal variation of the regional
literature review of the life history characteristics of individual
options were invoked. These analyses were selected based biogeography of marine birds is influenced by the vagaries
141
Section 5: SUMMARY OF BIOGEOGRAPHIC ASSESSMENT
work is also planned.
of marine climate, which is driven by the California Current abundance of the humpback and blue whales during the these concerns is presented in Section 3.
System and local upwelling centers. Therefore, the Upwelling and Oceanic seasons.
Spatial Patterns. The spatial occurrence of marine mammals
biogeographic patterns of marine birds are not static and Key Findings of Integration of Analyses
relative to large bathymetric features (e.g., shelf, upper
exhibit a dramatic spatial and temporal variation, both in Small Cetaceans. An important time period for the Pacific Fish Diversity (Trawl data). Three major areas of relatively
slope, lower slope) and discrete physiographic features (e.g.,
species composition and species abundance. white-sided dolphin (the most abundant small cetacean in high fish diversity (i.e., hot spots) were delineated, as noted
seamounts, banks, canyons, points and islands) varied by this study) was the Oceanic season. Important time periods below.
species and ocean condition. The occurrence patterns of
• The Gulf of the Farallones, the area lying inside a triangle for the other relatively abundant smaller cetaceans (northern • The northernmost hot spot is centered on Cordell Bank,
most marine mammals are strongly linked to the highly variable
defined by Point Reyes, the Farallon Islands and Año right-whale dolphin, Risso’s dolphin, Dall’s porpoise) could not within the northwestern corner of the Cordell Bank NMS,
ocean conditions of the study area, which significantly affect the
Nuevo Island, is the most important area for marine birds be determined in this preliminary assessment. and extends northward along the continental slope to Point
distribution of prey availability. In Phase II of this work, when the
in California. The reasons are: (1) large and taxonomically Arena.
data sets are more spatially and temporally robust, summary
diverse, demographically related populations breed at the Pinnipeds. The seasonal occurrence of pinnipeds was
analyses will be conducted to identify important areas and time
three afore-mentioned sites; and (2) an unparalleled diversity associated with the breeding cycles of the species. Important • The central hot spot is centered at the boundary between
periods across marine mammal groups.
of habitat (e.g., San Francisco Bay tidal plume, shallow time periods for the relatively abundant northern fur seal were the Gulf of the Farallones NMS and the Monterey Bay NMS.
sandy shelf, rocky reefs, submarine peaks, and the upper winter and early spring (Davidson Current and early Upwelling The area extends in a southeasterly direction past Point Año
Large Cetaceans. Important areas for the large cetaceans
continental slope) attracts a variety of migrant and seasonally seasons), which reflected the pelagic offshore distribution along Nuevo and ends offshore, north of Monterey Bay.
varied by species: the coast and inner shelf were important
resident species. the West Coast during the nonbreeding season. The relatively
for the gray whale; the outer shelf, slope, and deep ocean abundant California sea lion was present year-round in the • The southernmost hot spot is located between Point Sur and
were important for the humpback and blue whales; and many
• A "halo" of individuals was apparent around important study area, with densities greater during the Oceanic season Lopez Point and covers the inshore portions of Sur and Lucia
important areas for large cetaceans were identified seaward
breeding sites, such as the Farallon Islands and Año Nuevo (just after breeding) and Davidson Current season (before the Canyons. Portions of this last hot spot, however, were poorly
of the sanctuary boundaries.
Island. This pattern is the result of breeding individuals breeding season). Elephant seals, Steller sea lions, and harbor sampled. There is evidence of an additional hot spot strad-
searching for food, but going only as far as necessary to seals were present in sanctuary waters year-round in relatively dling the southern boundary of the Monterey Bay NMS.
Small Cetaceans. Review of the maps indicated that important
provide for their young. The Farallon "halo" for ashy storm- low numbers; and important time periods for these infrequently
areas for the relatively abundant small cetaceans were the
petrel, western gull, common murre, rhinoceros auklet and sighted species were inconclusive due to differences in Marine Bird Diversity
outer shelf and upper slope, Monterey Canyon, Sur and Lucia
Cassin’s auklet, extends substantially west of the Gulf of the behavior and low abundance (e.g., at-sea sightings typically • The interpolated maps of marine bird diversity show one con-
Canyons (west and south of the Monterey Bay NMS), Pioneer
Farallones National Marine Sanctuary. consist of single individuals or small groups of two or three, tinuous area of high diversity along the continental slope, and,
Canyon (west of the Monterey Bay NMS), Ascension, Cabrillo, elephant seals are rarely at the surface, and Steller sea lions to a lesser extent, along the shelf between Point Arena and
Año and Carmel canyons, Cordell Bank (and to the north of
• The marine birds of the Gulf of the Farallones/Cordell Bank are a threatened species). Point Sur. Within this area, diversity appears to be highest on,
the Cordell Bank NMS boundary), and the San Francisco Bay
NMS (as defined above) and the birds of the Monterey Bay and seaward of, the Farallon Escarpment, in the northwestern
tidal plume area (e.g., harbor porpoise). Smaller cetaceans
NMS are associated with different habitat features. The Gulf of INTEGRATION OF ANALYSES corner of the Monterey Bay NMS (Pioneer Canyon), and in
were also relatively abundant in areas that include canyons,
the Farallones has islands and a relatively broad shelf, while The integration of analyses across taxa occurs in Section 3. the marine region between Point Lobos and Point Sur.
and in locations beyond sanctuary boundaries, but within the
Monterey Bay has a relatively narrow but sheltered shelf, cut Many possible combinations of the data layers could be inte-
study area.
by an immense, deep submarine canyon. The greater oceanic grated for the biogeographic assessment. Because of differ- • The Farallon Escarpment in particular received a dispropor-
influence and lack of breeding islands in the Monterey Bay ences in sampling design, it was not appropriate to combine tionate amount of survey effort. The high estimated marine
Pinnipeds. Important areas for resident breeders (e.g., harbor
NMS drive the marine bird species groups there. data from different taxa (e.g. birds and fish) in order to calculate bird diversity for the Farallon Escarpment is, in part, due to
seal, Steller sea lion) were inner and outer shelf habitats, and for community metrics. Therefore, to minimize confounding results high sampling effort.
northern elephant seal, pelagic deep ocean habitats seaward
Preliminary Marine Mammal Analytical Results and to focus on the “protection of biodiversity” component of
of sanctuary boundaries. Seasonal visitors (e.g., northern fur
The Biogeography Team contracted principal investigators the NMSP mission, diversity and density were calculated sepa- • A marine bird diversity hot spot was found in the region be-
seals) occurred mostly in slope and deep ocean habitats,
David Ainley and Glenn Ford (of H.T. Harvey and Associates rately for each taxon and the resulting patterns were overlayed tween Point Lobos and Point Sur. The high residual diversity
seaward of sanctuary boundaries.
and R.G. Ford Consulting Co.) to work with the NOAA to identify biologically important areas across species groups. in this area supports the interpretation that this is a real hot
project team and local marine mammal experts to identify Spatial interpolation methods were applied to survey data to spot and not an artifact of survey effort.
Temporal Patterns. The patterns of seasonal occurrence for
biogeographic patterns and important areas and time periods provide a clearer picture of the distribution of diversity and den-
marine mammals varied by species. In Phase II of this work,
for marine mammals occurring in the study area. NOAA/NMFS sity within the study area. Hot spots were defined as regions • Marine bird diversity was correlated with survey effort, so
when the data sets are more complete, summary spatial and
scientists provided additional marine mammal sightings data in which diversity or density were estimated to be in the top some of the “hot spot” diversity areas coinciding with areas
temporal analyses across marine mammal groups will be
along the entire West Coast to aid in analyzing marine mammal 20% for a particular taxon. These hot spots were mapped for of high survey effort may, in part, be influenced by high lev-
conducted.
biogeographic patterns relative to the study area. The "bird and fish and birds individually and then combined to show areas els of effort. However, the general patterns of marine bird
mammal team" used a GIS to develop a preliminary series of of overlap. These areas of significant biological importance diversity are robust, and were largely unchanged by methods
Large Cetaceans. The seasonal occurrence of the larger
maps that show occurrence patterns and important areas and contributed to defining and assessing biogeographic patterns designed to correct for effort.
cetaceans in the study area reflected their migrations. The
time periods for 13 marine mammals in the study area. Phase within the study area and are discussed in the context of known
Davidson Current season was important for the gray whale,
II of this assessment will: incorporate additional data, develop oceanographic features and Sanctuary boundaries. All of the Overlap of Marine Bird and Fish (Trawl) Diversity
a period when this species is migrating either south or north.
additional marine mammal species and species group maps, conclusions listed below should be considered with an under- • Fish diversity shows overlap with the areas of high bird di-
Several species of the large cetaceans migrate to forage
and attempt to develop selected community metrics analyses standing of the inherent limitations of the available data and versity. The northernmost fish hot spot overlaps the marine
seasonally in the study area, a pattern reflected in the relative
across species and time periods. A technical report on this the approaches used to analyze it. A detailed discussion of bird hot spot from Cordell Bank north to approximately mid-
142
Section 5: SUMMARY OF BIOGEOGRAPHIC ASSESSMENT
support the Cordell Bank, Gulf of the Farallones, and Monterey
as that component of the study was not completed in Phase
• Although the majority of the hot spots for fish and bird density
way between Bodega Head and Point Arena. The seaward
Bay National Marine Sanctuaries.
I. The marine mammal analyses are one of the first efforts to
fall within sanctuary boundaries, it is notable that overlap-
half of the central fish hot spot overlaps with the area of
assess biogeographic patterns of marine mammals in the study
ping hot spots for both groups exist to the south of Monterey
high marine bird diversity within the Gulf of the Farallones
area; thus, additional analyses and peer review are required
Bay NMS.
NMS and the Monterey Bay NMS. The northern half of the
to complete this component of the study. REFERENCES
southernmost fish hot spot overlaps the southern tip of the
Overall Integration Summary Airamé, S., S. Gaines, and C. Caldow. 2003. Ecological Link-
marine bird hot spot.
Phase II activities may include publishing technical reports
•The current Sanctuary boundaries incorporate much of ages: Marine and estuarine ecosystems of central and northern
and peer-reviewed articles that complement the results of the
the highest diversity and highest density areas within the California. NOAA, National Ocean Service. Silver Spring, MD.
Fish Density (Trawl Data)
Phase I assessment, as well as additional analyses to further
region. 163 pp.
Four major hot spots of fish density were found:
define biological areas and time periods important to marine
• The northernmost hot spot is found on and to the southeast
fishes, birds, and mammals found throughout the study area.
• Many of these biologically important regions coincide with Brown, S.K., K.R. Buja, S.H. Jury, M.E. Monaco, and A. Ban-
of Cordell Bank.
known oceanographic and bathymetric features, such as ner. 2000. Habitat suitability index models for eight fish and
CD-ROM
upwelling regions, areas of high bathymetric variance, and invertebrate species in Casco and Sheepscot Bays, Maine.
• A small hot spot is found off of Point Reyes.
A digital version of this document, the Ecological Linkages
the continental shelf break. North American Journal of Fisheries Management, Vol. 20,
Report, all GIS-compatible files used to conduct the pp. 408-435. 28 pp.
• The largest fish density hot spot covers a large portion of
biogeographic analyses, metadata for GIS files, and a complete
• Regions of high diversity and high density outside of the
the shelf to the north of Monterey Canyon, the entire area of
suite of digital species maps, are found on the CD-ROM located
current sanctuary boundaries exist to the north, across Kendall, M.S. and M.E. Monaco. 2003. Biogeography of the
Monterey Bay, and the near shore waters south to Point Sur.
on the back cover of this document.
much of the shelf and slope, and to the south, in near- National Marine Sanctuaries: A partnership between the NOS
Although portions of this hot spot are found over Monterey
shore waters. Biogeography Program and the National Marine Sanctuary
Canyon, this fact should be interpreted with caution since the
All appropriate digital data and analytical products are found on Program. NOAA. Silver Spring, MD. 8 pp.
deep canyon waters themselves were not sampled.
the CD-ROM. The products come in several formats, including
• Uneven sampling effort across the study region and a lack
this document (in .pdf format), map products in a browsable
of trawl samples to the west of the Sanctuary boundaries Monaco, M.E. and J.D. Christensen. 1997. Biogeography Pro-
• The fourth hot spot is found to the south of Monterey Bay
web format (HTML), GIS shapefiles and grids for use with
limit the scope of any integrated biogeographic assess- gram: Coupling species distributions and habitat. In: Changing
NMS and covers a substantial area of the shelf from Point
MarIS or ArcView (GIS) software, tables in Excel format (.xls),
ment. oceans and changing fisheries: Environmental data for fisheries
Estero to Point Sal. This final hot spot is the largest region
and descriptive text files. Metadata for each shapefile or grid research and management. G.W. Boehlert and J.D. Schum-
of high fish density within the mapped area that falls outside
accompanies each file and appears in .xml format.
• Known limitations and biases of the two metrics (diversity acher (Eds.). National Marine Fisheries Service Technical
of Sanctuary boundaries.
and density) exist and are discussed in greater detail within Memorandum NOAA-TM-NMFS-SWRSC-239, Pacific Grove,
To support the NMSP and others in making maximum use of
Section 3. California. 7 pp.
Marine Bird Density
the spatial data generated from this study, along with other
• Marine bird density patterns should be interpreted with
products (e.g., economic assessments) that support the
DATA SOURCES AND GAPS NOAA's National Centers for Coastal Ocean Science and
caution since they largely reflect the distribution of the two
joint management plan review, the NMSP is developing a
Recognizing that any analysis is only as good as the data National Marine Sanctuary Program. 2002. Interim Product:
numerically dominant species.
GIS tool, the Marine Resource Information System (MarIS).
upon which it is based, the project team undertook a qualitative A Biogeographic Assessment off North/Central California: To
MarIS has been designed to facilitate the organization,
evaluation of the data used in this project and identified relevant Support the Joint Management Plan Review for Cordell Bank,
• A large region of high (top 20th percentile) marine bird density
analysis and display of spatial data to support analysis of
data gaps. This information is presented in Section 4: Data Gulf of the Farallones and Monterey Bay National Marine
exists adjacent to and shoreward of the marine bird diversity
NMSP management questions and issues within and across
Content and Gaps. This section describes the process used Sanctuaries. Silver Spring, MD. 38 pp.
hot spot. This density hot spot covers most of the shelf wa-
sanctuaries. All applicable spatial data will be integrated into
to select key databases for analyses and briefly addresses
ters of all three sanctuaries, from Point Sur in the south to
MarIS to enable NMSP staff and partners to conduct additional
strengths and limitations of each database. This information was Starr, R.M. 1998. Marine harvest refugia for West Coast rock-
midway between Bodega Head and Point Arena in the north.
biogeographic analyses in Phase II.
used to aid in the interpretation of the biogeographic analyses to fish: A workshop. Pacific Grove, California. NOAA-TM-NMFS-
The density hot spot extends into Monterey Bay.
minimize confounding of results due to information gaps. Also SWFSC-255. La Jolla, CA. 14 pp.
CONCLUDING COMMENTS
provided are recommendations for future research activities
• An additional density hot spot exists off of Morro Bay to the
This spatially explicit assessment provides a robust set of
that would enhance biogeographic assessment products.
south of the Monterey Bay NMS.
analytical results and GIS data to strengthen the sustainable
management of marine resources within and adjacent to the
PHASE II BIOGEOGRAPHIC ASSESSMENT
Overlap of Marine Bird and Fish (Trawl) Density
sanctuaries. A primary use of the biogeographic assessment
Section 6 suggests potential next steps to augment the
• Nearly all of the fish density hot spots are coincident with
will be to support the NMSP as it continues to conduct the joint
Phase I analyses. Phase II, however, will not be completely
the two areas of high bird density.
management plan review for the three sanctuaries. In addition,
designed until a review of Phase I products has occurred. The
the Biogeography Program will assist the NMSP in further
NMSP and NCCOS project team members will meet to define
• The hot spots for both metrics are generally confined to the
analyses and presentations of the data and analytical results
the additional suite of activities that will comprise Phase II.
shelf (<200m) with the notable exception of Monterey Canyon
to address specific research and management questions.
Nevertheless, a few priority activities are expected to occur in
which appears as a density hot spot for both groups. The
This Phase I product provides the foundation to continue the
Phase II, including expanding the analytical products for fishes,
deep Canyon, however, was not sampled in the fish trawl
development of a biogeographic assessment capability to
marine birds, and marine mammals. Special emphasis will be
surveys.
placed on the biogeographic analyses of marine mammal data,
143
Section 6: PHASE II BIOGEOGRAPHIC ASSESSMENT
Phase II of the biogeographic assessment of north/central California to support the research and management needs of Cordell Bank, Gulf of the Farallones, and Monterey Bay National Marine Sanctuaries will build on the information and analytical results
presented in Phase I. Phase II, however, will not be completely designed until a review of Phase I products has occurred. Most important, the NMSP and NCCOS project team members will meet to define an additional suite of activities that may comprise
Phase II. Nevertheless, a few priority activities are expected to occur in Phase II, including expanding the analytical products for fishes, marine birds, and marine mammals. A special emphasis will be placed on the biogeographic analyses of marine
mammal data, as that component of the study was not completed in Phase I. The marine mammal analyses are one of the first efforts to assess biogeographic patterns of mammals in the study area, thus additional analyses and peer review are required
to complete this section of the study.
Phase II activities may include publishing technical reports and peer reviewed articles that complement the results of the Phase I assessment and further define areas and time periods important to fishes, marine birds, and marine mammals found
throughout the study area. Also, discussions will be held between project partners to determine if additional assessments should be implemented to study near-shore and estuarine ecosystems and associated key species groups such as marine and coastal
invertebrates. These ecosystems and species were only qualitatively addressed in the Ecological Linkages Report due to data limitations, and time and resource constraints to complete the first phase of the project. In addition, to continue to implement
the 5-year Biogeography Program plan developed by NCCOS in consultation with the NMSP, additional habitat and environmental maps under various temporal climate regimes could possibly be addressed to support future biogeographic analyses. For
example, climatic regime shifts and associated influences on the distribution of living marine resources may provide additional insight into natural or anthropogenic perturbations on regional biogeographic patterns.
For now, project partners and colleagues are encouraged to provide comments on the information and analytical results provided in this document and on the CD-ROM. Also, please provide suggestions on how best to address Phase II proposed activities
and new biogeographic assessment studies that may complement or improve Phase I analyses. For further information or to provide comments on the Phase I product and Phase II activities, please contact:
Dr. Mark E. Monaco, Biogeography Team Leader
NOAA/NCCOS/CCMA
1305 East West Highway, N/SCI1
Silver Spring, MD 20910
p. 301-713-3028 x 160
f. 301-713-4384
mark.monaco@noaa.gov
or
Mr. Charles E. Alexander, National Programs Branch Chief
NOAA/NMSP
1305 East West Highway,
Silver Spring, MD 20910
p. 301-713-3125 x 147
f. 301-713-0404
charles.alexander@noaa.gov
144
ACKNOWLEDGEMENTS
A project of this magnitude would not have been possible without the cooperation, time, data and effort of numerous people and institutions. We would like to thank everyone who participated in this significant undertaking. The following is a list of those
who have contributed to this work.
Jamie Kum
Sarah Allen
Federal Government
Tom Laidig
Tara Anderson
National Centers for Coastal Ocean Science, National Ocean Service (NOS), NOAA
Mark Lampinen
Jay Barlow
National Marine Sanctuaries National Program Office, NOS, NOAA
Bob Lauth
Lillian Becker
Cordell Bank National Marine Sanctuary Office, NOS, NOAA
Bob Leeworthy
Heather Beeler
Gulf of the Farallones National Marine Sanctuary Office, NOS, NOAA
Phil Levin
Scott Benson
Monterey Bay National Marine Sanctuary Office, NOS, NOAA
Steve Lonhart
Carol Bernthal
Channel Islands National Marine Sanctuary Office, NOS, NOAA
David Lott
Nancy Black
Special Projects Office, NOS, NOAA
Mark Lowry
Laurence Breaker
Coastal Services Center, NOS, NOAA
Gerry McChesney
Joelle Buffa
Southwest Fisheries Science Center, National Marine Fisheries Service (NMFS), NOAA
Huff McGonigal
Tonya Builder
Northwest Fisheries Science Center, NMFS, NOAA
Nazila Merati
Gregor Cailliet
Alaska Fisheries Science Center, NMFS, NOAA
Richard Methot
John Calambokdis
U.S. Geological Survey, Department of Interior
Pat Morris
James Caretta
Minerals Management Service, Department of Interior
Joe Mortenson
Mark Carr
U.S. Fish and Wildlife Service, Department of Interior
Hannah Nevins
Harry Carter
Point Reyes National Seashore, National Park Service, Department of Interior
Kelly Newton
Josh Churchman
Jim Nybakken
Elizabeth Clarke
State/Local Government
Jim Oakden
Steve Copps
California Department of Fish and Game, Marine Region
Mike Parker
Natalie Cosentino
University of California, Santa Barbara
John Pearse
Don Croll
California State University, Monterey Bay
Brady Phillips
Brad Damitz
California State University, Moss Landing Marine Laboratories
Holly Price
Chris Essert
San Francisco Public Utilities Commission
Stephen Ralston
Megan Ferguson
Bob Read
Doug Forsell
Non-Governmental Organizations
Paul Reilly
Fathey Fosmark
Alliance of Communities for Sustainable Fisheries
Dale Roberts
Michael Gallagher
H.T. Harvey and Associates
Nora Rojek
Jim Glock
R.G. Ford Consulting Company
Kaustuv Roy
Gary Greene
Monterey Bay Aquarium Research Institute
Mary Jane Schramm
Denise Greig
PRBO Conservation Science
Michelle Staedler
Deirdre Hall
Diablo Canyon Power Plant
Rick Starr
Rick Hanks
William Sydeman
Michael Harris
Mario Tamburri
Chris Harvey
Christine Taylor
Jim Harvey
Julie Thayer
Brian Hatfield
Teresa Turk
Laird Henkel
W. Breck Tyler
Michelle Hester
Tiffany Vance
Tom Hourigan
Kerstin Wasson
Ruth Howell
Mark Wilkins
David Hyrenbach
Deb Wilson-Vandenberg
Todd Jacobs
Lisa Wooninck
Brian Jarvis
Nancy Wright
Roxanne Jordan
Levon Yengoyan
Mike Kenner
Mary Yoklavich
Stacy Kim
Mark Zimmermann
Chad King
Howatt King
145
A Biogeographic Assessment off North/Central California:
To Support the Joint Management Plan Review for
Cordell Bank, Gulf of the Farallones, and Monterey Bay
National Marine Sanctuaries: Phase I -
Marine Fishes, Birds and Mammals
Prepared by NOAA's Center for Coastal Monitoring and Assessment - Biogeography Team
National Centers for Coastal Ocean Science
December 2003
ABOUT THIS DOCUMENT
In the spring of 2001, NOAA’s National Marine Sanctuary Program (NMSP) and National Centers for Coastal Ocean Science (NCCOS), launched a 24-month effort to assess biogeographic patterns of selected marine species found within and adjacent to
the boundaries of three west coast National Marine Sanctuaries. These sanctuaries, Monterey Bay, Gulf of the Farallones, and Cordell Bank, are conducting a joint review process to update sanctuary management plans. To support this review, NCCOS’s
Biogeography Program is leading a partnership effort to conduct a robust analytical assessment to define important biological areas and time periods within, and adjacent to, current sanctuary boundaries. The assessment was based on a synthesis of
many data sets that were provided by project partners. This document represents the results of the first of two phases of the assessment. Phase I provides data, analytical results, a description of ecosystems and their linkages, identifies data gaps, and
suggests future activities to be addressed in Phase II.
Phase I of the biogeographic assessment was formulated around three integrated study components: 1) an Ecological Linkages Report, 2) biogeographic analyses, and 3) development of Geographical Information System (GIS) data for incorporation into
NMSP’s Marine Information System (MarIS). The majority of the results from the assessment are presented as a suite of GIS maps to visually display species biogeographic patterns across the study area. The body of this document provides examples
of the entire suite of digital map products found on the CD-ROM located on the back cover of this document. The spatial data and additional information, such as digital species distribution maps, and additional details on analytical methodologies are also
presented on the CD-ROM. An HTML version of the CD-ROM can be found on the Biogeography Program website: http://biogeo.nos.noaa.gov/products/canms_cd/.
Results of the assessment are being used to assist the NMSP in addressing issues such as evaluating potential modification of sanctuary boundaries, and changes in management strategies or administration, based on the principles of biogeography. The
progress of the biogeographic assessment for Central and Northern California National Marine Sanctuaries can be followed by consulting NCCOS’s Biogeography Program web site: http://biogeo.nos.noaa.gov/projects/assess/ca_nms/.
For questions and comments, please contact:
Project Team
David Ainley, H.T. Harvey and Associates
Dr. Mark E. Monaco, Biogeography Team Leader
Satie Airamé, National Marine Sanctuary Program
NOAA/NCCOS/CCMA
Charles Alexander, National Marine Sanctuary Program
1305 East West Highway, N/SCI1
Lisa Ballance, Southwest Fisheries Science Center
Silver Spring, MD 20910
Maria Brown, National Marine Sanctuary Program
p. 301-713-3028 x 160
Erica Burton, National Marine Sanctuary Program
f. 301-713-4384
Kenneth Buja, National Centers for Coastal Ocean Science
mark.monaco@noaa.gov
Chris Caldow, National Centers for Coastal Ocean Science
Janet Casey, R.G. Ford Consulting Company
or
John Christensen, National Centers for Coastal Ocean Science
Lawrence Claflin, National Centers for Coastal Ocean Science
Mr. Charles E. Alexander, National Programs Branch Chief
Randall Clark, National Centers for Coastal Ocean Science
NOAA/NMSP
Michael Coyne, National Centers for Coastal Ocean Science
1305 East West Highway,
Andrew DeVogelaere, National Marine Sanctuary Program
Silver Spring, MD 20910
William Douros, National Marine Sanctuary Program
p. 301-713-3125 x 147
R. Glenn Ford, R.G. Ford Consulting Company
f. 301-713-0404
Steve Gaines, University of California Santa Barbara
charles.alexander@noaa.gov
Tracy Gill, National Centers for Coastal Ocean Science
Jamison Higgins, National Centers for Coastal Ocean Science
Dan Howard, National Marine Sanctuary Program
Olaf Jensen, National Centers for Coastal Ocean Science
Carol Keiper, Oikonos
Matthew Kendall, National Centers for Coastal Ocean Science
Kevin McMahon, National Centers for Coastal Ocean Science
Mark Monaco, National Centers for Coastal Ocean Science
Wendy Morrison, National Centers for Coastal Ocean Science
Sean Morton, National Marine Sanctuary Program
Jan Roletto, National Marine Sanctuary Program
Larry Spear, H.T. Harvey and Associates
Lynn Takata, National Centers for Coastal Ocean Science
Mitchell Tartt, National Marine Sanctuary Program
Christine Taylor, National Marine Sanctuary Program
Citation: Ed Ueber, National Marine Sanctuary Program
NOAA National Centers for Coastal Ocean Science (NCCOS) 2003. A Biogeographic Assessment off North/Central Jeannette Waddell, National Centers for Coastal Ocean Science
California: To Support the Joint Management Plan Review for Cordell Bank, Gulf of the Farallones, and Monterey Anne Walton, National Marine Sanctuary Program
Bay National Marine Sanctuaries: Phase I - Marine Fishes, Birds and Mammals. Prepared by NCCOS’s Biogeog- Wendy Williams, R.G. Ford Consulting Co.
raphy Team in cooperation with the National Marine Sanctuary Program. Silver Spring, MD 145 pp. Cover Photo by Kip Evans
i
TABLE OF CONTENTS
Table 8. Average frequency of occurrence of fish species assemblages (percent
About This Document ................................................................................................. i Section 3 Integration of Analyses................................................................................ 129
occurrence calculated for each species and then averaged for each fish
Table of Contents ....................................................................................................... ii Introduction.............................................................................................................. 129
assemblage) for each 1999 midwater site group. Number of trawls in each
Introduction................................................................................................................. 1 Data and Analyses .................................................................................................. 129
site group is provided in the first row. Bold numbers represent influential
The Study Area..................................................................................................... 1 Analytical Map Products .......................................................................................... 130
species assemblages within that site group.................................................... 32
Study Objectives .................................................................................................. 1 References .............................................................................................................. 136
Table 9. Example data matrix for calculating bathymetry SI values for subadult
Section 1 Synopsis of Ecological Linkages Report ................................................ 6 Section 4 Data Sources and Gaps ............................................................................... 138
bocaccio taken in NMFS trawl samples (Rubec et al., 1999). ........................ 36
Introduction........................................................................................................... 6 Introduction.............................................................................................................. 138
Table 10. Example presence/absence information and SI calculation from scientific
National Marine Sanctuaries of Central and Northern California ........................ 6 Conclusions and Recommendations for Future Activities ....................................... 138
literature. ........................................................................................................ 36
Geographic Setting of the Study Area .................................................................. 6 References .............................................................................................................. 139
Table 11. Marine bird species used in this analysis........................................................ 46
Ocean Currents .................................................................................................... 6 Section 5 Summary of Biogeographic Assessment .................................................. 140
Table 12. Summary of at-sea survey data sets used in the analyses............................. 47
Oceanographic Seasons ...................................................................................... 7 Background ............................................................................................................. 140
Table 13. Summary of combined data set effort by ocean season. .............................. . 48
Natural Perturbations ........................................................................................... 7 Biogeographic Assessment off North/Central California.......................................... 140
Table 14. Assignment of warm, cold and neutral periods, based on surface water
Ecosystems .......................................................................................................... 7 Ecological Linkages Report ..................................................................................... 140
temperatures off Cental California .................................................................. 50
Biogeography ...................................................................................................... 8 Biogeographic Analyses .......................................................................................... 141
Table 15. Life history and management information for selected marine birds off
Conclusions .......................................................................................................... 9 Integration of Analyses ............................................................................................ 142
north/central California. ................................................................................... 84
References ........................................................................................................... 9 Data Sources and Gaps .......................................................................................... 143
Table 16. A summary of temporal and spatial patterns in the at-sea survey data
Section 2 Biogeographic Analyses ........................................................................ 10 Phase II Biogeographic Assessment ....................................................................... 143
(1980-2001) of selected marine birds off north/central California. .................. 85
Introduction......................................................................................................... 10 CD-ROM.................................................................................................................. 143
Table 17. Important at-sea areas for marine birds off north/central California,
Assessment Process ......................................................................................... 10 Concluding Comments ............................................................................................ 143
References ......................................................................................................... 10 References .............................................................................................................. 143 based on biomass, density and diversity.. ...................................................... 86
Section 2.1 Biogeography of Fishes....................................................................... 12 Section 6 Phase II Biogeographic Assessment.......................................................... 144 Table 18. Major marine bird colonies along the central California coast . ...................... 87
Introduction......................................................................................................... 12 Acknowledgments......................................................................................................... 145 Table 19. Three most important variables (of nine investigated) having independent
Subsection 2.1.1 Assemblage Analyses ........................................................ 13 effects in explaining the variance in density of 25 selected marine bird
Introduction......................................................................................................... 13 species. .......................................................................................................... 88
Review of Relevant Literature ............................................................................ 13 Table 20. Effects of ocean season and ENSO events on the abundance of 26
Introduction to Clustering ................................................................................... 14 marine bird species off central California between 1985 and 2002, as
List of Tables
Section Summary ............................................................................................... 33 determined through multiple regression analyses.. ........................................ 88
Table 1. Site group results for recreational data. The numbers of trip/location
References ......................................................................................................... 34 Table 21. A summary of changes in marine bird occurrence patterns, as a response
combinations associated with each group as well as average depth, ± stan-
Subsection 2.1.2 Habitat Suitability Modeling ............................................... 35 to warm and cold ocean anomalies, as determined by visual comparison of
dard deviation, are provided. Different letters signify a significant difference
Introduction......................................................................................................... 35 species’ maps during the 1997-1998 El Niño event and the
using Tukey’s pairwise comparison on log adjusted depth with overall
Data and Analyses ............................................................................................. 35 1999-2000 La Niña event................................................................................ 89
alpha set at 0.001. .......................................................................................... 24
Analytical Map Products ..................................................................................... 36 Table 22. Marine mammal species included in this assessment and map types
Table 2. Average frequency of occurrence of fish species assemblages (percent
Section Summary ............................................................................................... 44 developed for them (Phase I and Phase II) .................................................... 92
occurrence calculated for each species and then averaged for each fish
Reviews .............................................................................................................. 44 Table 23. Summary of at-sea data sets used in the preliminary marine mammal
assemblage) for each recreational site group. Numbers in bold represent
Reviewers........................................................................................................... 44 analyses. ......................................................................................................... 93
influential species assemblages within that site group. .................................. 24
References ......................................................................................................... 45 Table 24. Summary of combined data set effort for mammals, by ocean season.. ........ 93
Table 3. Site group results for shelf trawl data. The numbers of trawls associated with
Section 2.2 Biogeography of Marine Birds ............................................................ 46 Table 25. Preliminary life history and management information for selected marine
each group as well as average depth ± standard deviation are provided.
Introduction......................................................................................................... 46 mammals off north/central California.. .......................................................... 123
Different letters signify a significant difference using Tukey’s pairwise
Data and Analyses ............................................................................................ 46 Table 26. Summary statistics and parameter estimates for spatial models. .................130
comparison on log adjusted depth with overall alpha set at 0.001. ................ 26
Analytical Map Products ..................................................................................... 50 Table 27. Matrix of data sets and their associated characteristics that were used or
Table 4. Average frequency of occurrence of fish species assemblages (percent
Section Summary ............................................................................................... 84 referenced in the biogeographic assessment. ............................................. 138
occurrence calculated for each species and then averaged for each fish
Major Section Contributors ................................................................................. 89
assemblage) for each shelf site group. Numbers in bold represent
Reviewers........................................................................................................... 89 List of Figures
influential species assemblages within that site group. .................................. 27
Personal Communications ................................................................................. 89 Figure 1. Project flow diagram showing steps to complete Biogeographic
Table 5. Site group results for slope trawl data. The numbers of trawls associated with
References ......................................................................................................... 90 Assessment Phase 1........................................................................................ 1
each group as well as average depth ± standard deviation are provided.
Section 2.3 Biogeography of Marine Mammals ..................................................... 92 Figure 2. Locator map of entire study area from Point Arena to Point Sal. National
Different letters signify a significant difference using Tukey’s pairwise
Introduction......................................................................................................... 92 Marine Sanctuary boundaries shown in red. .................................................... 2
comparison on log adjusted depth with overall alpha set at 0.001. ................ 28
Data and Analyses ............................................................................................. 92 Figure 3. Detailed locator map of northern study area from Bodega Head to
Table 6. Average frequency of occurrence of fish species assemblages (percent
Analytical Map Products ..................................................................................... 94 Pescadero Point. National Marine Sanctuary boundaries shown in
occurrence calculated for each species and then averaged for each fish
Section Summary ............................................................................................. 123 red............. ....................................................................................................... 3
assemblage) for each slope site group. Bold numbers represent influential
Discussion ........................................................................................................ 125 Figure 4. Detailed locator map of central study area from Pescadero Point to Pfeiffer
species assemblages within that site group.................................................... 29
Major Section Contributors ............................................................................... 125 Point. National Marine Sanctuary boundaries shown in red. ........................... 4
Table 7. Average frequency of occurrence of fish species assemblages (percent
Reviewers......................................................................................................... 125 Figure 5. Detailed locator map of southern study area from Pfeiffer Point to
occurrence calculated for each species and then averaged for each fish
Personal Communications ................................................................................126 Point Sal. National Marine Sanctuary boundaries shown in red. ..................... 5
assemblage) for each 1998 midwater site group. Number of trawls in each
References ........................................................................................................126 Figure 6. Biogeographic assessment process ............................................................... 10
site group is provided in the first row. Bold numbers represent influential
Figure 7. 3-D image of bathymetric relief within and adjacent to the sanctuaries .......... 11
species assemblages within that site group.................................................... 31
ii
TABLE OF CONTENTS
Figure 23. Location of site groups for NMFS slope trawls. Lines showing the 50, Figure 51. Rhinoceros auklet, seasonal density, high use areas and breeding
Figure 8. Biogeographic approach to fish analysis ........................................................ 12
100, 200 and 2,000 depth contours are provided. .......................................... 29 colonies. .......................................................................................................... 69
Figure 9. Hypothetical example of the methods used to determine species assemblag-
Figure 24. Species assemblage results for the midwater trawls utilizing all data from Figure 52. Marine bird density, by season and for all seasons........................................ 71
es, site groups, and the interaction between species assemblages and
1986 to 2001. Assemblages are named for the most influential species in Figure 53. Marine bird biomass, by season and for all seasons. .................................... 72
site groups...................................................................................................... 15
each group. Non-italicized species were consistently placed into the same Figure 54. Marine bird diversity, by season and for all seasons...................................... 73
Figure 10. Standard deviation of bathymetry calculated for a 1 km radius around each
species assemblage >80% of the time; italicized species tended to roam Figure 55. Major marine bird breeding colonies. . ........................................................... 75
cell. Results are presented in standard deviations above or below
into other assemblages with random sampling.............................................. 30 Figure 56. Density in warm, cold, and neutral periods: 1980-2001. ............................... 76
the mean. ....................................................................................................... 16
Figure 25. Species assemblage results for the midwater trawls conducted in 1998. As- Figure 57. Biomass in warm, cold and neutral periods: 1980-2001. ............................... 78
Figure 11. Species richness of individual NMFS shelf and slope trawls. ...................... . 17
semblages are named for the most influential species in each group. Non- Figure 58. Diversity in warm, cold and neutral periods: 1980-2001. ............................... 80
Figure 12. Mean species richness of NMFS shelf and slope trawls for 5’ grid cells. The
italicized species were consistently placed into the same species assemblage Figure 59. Density during El Niño and La Niña events, 1997-2000. ............................... 82
deviation is shown as an overlay to provide an indication of the variability
>80% of the time; italicized species tended to roam into other Figure 60. El Niño/La Niña Event changes, as an example of regime shift effects... ...... 83
in results for each grid cell. ............................................................................. 18
assemblages with random sampling.. ............................................................. 31 Figure 61. Total at-sea survey effort for marine mammal analysis. ................................. 94
Figure 13. Species diversity of individual NMFS shelf and slope trawls. . ...................... 19
Figure 26. Location of site groups for NMFS 1998 midwater trawls. Lines showing Figure 62. Maps for southern sea otter: rangewide count and linear density,
Figure 14. Mean species diversity of NMFS shelf and slope trawls for 5’ grid cells. The
the 50, 100, 200, and 2,000 depth contours are provided. ............................ 31 fall 2001 and spring 2002. .............................................................................. 95
deviation is shown as an overlay to provide an indication of the
Figure 27. Species assemblage results for the midwater trawls conducted in 1999. As- Figure 63. Map for California sea lion: haulouts and rookeries. ...................................... 96
variability in results for each grid cell. ............................................................. 20
semblages are named for the most influential species in each group. Non- Figure 64. Maps for California sea lion: seasonal at-sea densities, high use areas
Figure 15. Species richness of rockfish from individual NMFS shelf and slope trawls.... 21
italicized species were consistently placed into the same species assemblage and rookeries. ................................................................................................. 97
Figure 16.The relationship between depth and rockfish richness showing mean rockfish
>80% of the time; italicized species tended to roam into other Figure 65. Map for Steller sea lion: at-sea sightings and survey effort,
richness (for 10 meter depth intervals between 50-1,300 meters).
assemblages with random sampling... ............................................................ 32 rookeries and haulouts.................................................................................... 99
The relationship was fit with a smoothing spline, lambda = 1,000,000. .......... 21
Figure 28. Location of site groups for NMFS 1999 midwater trawls. Lines showing Figure 66. Maps for northern fur seal: seasonal at-sea densities, high use areas
Figure 17. NMFS shelf and slope trawls with the highest species diversity, species rich-
the 50, 100, 200, and 2,000 depth contours are provided. ............................ 32 and rookery. .................................................................................................. 101
ness, and rockfish richness are mapped. The underlying map illustrates the
Figure 29. Overlap between the three data sets that analyzed demersal fish: CDF&G Figure 67. Map for harbor seal: at-sea sightings, survey effort and haulouts................ 103
bathymetric complexity of the study area and can be used to identify
recreational (yellow), NMFS shelf (green), and NMFS slope (orange). .......... 33 Figure 68. Map for northern elephant seal: at-sea sightings and survey effort,
the shelf break. ............................................................................................... 22
Figure 30. Species habitat suitability modeling approach. .............................................. 35 rookeries and haulouts.................................................................................. 105
Figure 18.Species assemblage results for the recreational data. Assemblages are named
Figure 31. Polynomial regression curve fit with mean log abundance by categorical Figure 69. Maps for Dall’s porpoise: seasonal at-sea densities and high use areas..... 107
for the most influential species in each group. Assemblages are arranged
bathymetric class for subadult bocaccio. ........................................................ 35 Figure 70. Maps for Pacific white-sided dolphin: seasonal at-sea densities and
from shallow to deep, unless they are influential at all or none of the depths.
Figure 32. Bathymetric map for the north/central California study area. Red lines high use areas. ............................................................................................. 109
The assemblages that were not influential at any depth were composed of
indicate National Marine Sanctuary boundaries. . .......................................... 37 Figure 71. Maps for Risso’s dolphin: seasonal at-sea densities and high use areas. ....111
relatively rare species, making depth associations indiscernible given the
Figure 33. Substrate types for the north/central California marine region. ...................... 38 Figure 72. Maps for northern right-whale dolphin: seasonal at-sea densities and
methodology for defining “influential” assemblages. Non-italicized species
Figure 34. Potential distribution of habitat suitability for adult and subadult bocaccio. high use areas. ............................................................................................ 113
were consistently placed into the same species assemblage >80% of the
Map inset contains validation statistics. SI values for bathymetry and Figure 73. Map for blue whale: at-sea sightings and survey effort... ............................. 115
time; italicized species tended to roam into other assemblages with
substrate are graphically displayed below the map. ....................................... 39 Figure 74. Maps for humpback whale: seasonal at-sea densities and high use areas. 117
random sampling. ........................................................................................... 23
Figure 35. Potential distribution of habitat suitability for adult and subadult Dover sole. Figure 75. Maps for gray whale: seasonal at-sea densities and high use areas........... 119
Figure 19. Location of CDF&G recreational fishing data in 2.5 minute grids which are
Map inset contains validation statistics. SI values for bathymetry Figure 76. Maps for Dall’s porpoise: SWFSC stock assessment data: average
color coded according to the average depth of the fishing trips within the grid
and substrate are displayed below the maps. ................................................ 40 group size of sightings and survey effort....................................................... 121
cell. Lines showing the 50, 100, 200, and 2,000 depth contours
Figure 36. Potential distribution of habitat suitability for adult dungeness crab. Map inset Figure 77. Maps for blue whale: SWFSC stock assessment data: average
are provided. ................................................................................................... 25
contains validation statistics. SI values for bathymetry and substrate group size of sightings and survey effort....................................................... 122
Figure 20. Species assemblage results for the shelf trawls. Assemblages are named
are graphically displayed below the map. ....................................................... 41 Figure 78. Maps for humpback whale: SWFSC stock assessment data: average
for the most influential species in each group. Assemblages are arranged
Figure 37. Areas of groundfish potential hot spots based on mean fish species HSI group size of sightings and survey effort....................................................... 122
from shallow to deep, unless they are influential at all or none of the depths.
models and overlap of predicted highly suitable habitats. .............................. 42 Figure 79. Pictogram of species diversity. ..................................................................... 129
The assemblages that were not influential at any depth were composed of
Figure 38. Areas of potential habitat importance based on mean HSI models for Figure 80. Estimated diversity (a), density (b), and hot spots (top 20%)
relatively rare species, making depth associations indiscernible given the
selected species assemblages. ...................................................................... 43 (c) for marine birds. ....................................................................................... 131
methodology for defining “influential” assemblages. Non-italicized species
Figure 39. Spatial extent of data sets used in the marine bird analysis: Figure 81. Relationships between bird diversity and bathymetric variance and
were consistently placed into the same species assemblage >80% of the time;
ship-based surveys. ....................................................................................... 47 between bird density and depth. ................................................................... 131
italicized species tended to roam into other assemblages with random
Figure 40. Spatial extent of data sets used in the analysis: aerial surveys. .................... 48 Figure 82. Estimated diversity (a), density (b), and hot spots (top 20%) (c) for fish...... 132
sampling. ....................................................................................................... 26
Figure 41. Total survey effort for marine bird analyses.................................................... 49 Figure 83. Integration option 1, diversity hot spots (top 20%) for fish and
Figure 21.Location of site groups for NMFS shelf trawls. Lines showing the 50, 100,
Figure 42. Western and Clark’s grebes, seasonal density and high use areas............... 51 marine birds. Coastal kelp bed areas are also shown. ................................. 133
200, and 2,000 depth contours are provided. ................................................. 27
Figure 43. Northern fulmar, seasonal density and high use areas. ................................. 53 Figure 84. Integration option 2, density hot spots (top 20%) for marine birds and fish.
Figure 22. Species assemblage results for the slope trawls. Assemblages are named
Figure 44. Sooty shearwater, seasonal density and high use areas. .............................. 55 Coastal kelp bed areas are also shown. ....................................................... 134
for the most influential species in each group. Assemblages are arranged
Figure 45. Ashy storm-petrel, seasonal density, high use areas, and breeding colonies.57 Figure 85. Integration option 3, diversity and density, hot spots (top 20%) for fish
from shallow to deep, unless they are influential at all or none of the depths.
Figure 46. Leach’s storm-petrel, seasonal density, high use areas, and and marine birds. Coastal kelp bed areas are also shown. .......................... 135
The assemblages that were not influential at any depth were composed of
breeding colonies.. .......................................................................................... 59
relatively rare species, making depth associations indiscernible given the
Figure 47. Scoters, seasonal density and high use areas............................................... 61
methodology for defining “influential” assemblages. Non-italicized species
Figure 48. Brown pelican, seasonal density and high use areas. ................................... 63
were consistently placed into the same species assemblage >80% of the time;
Figure 49. Black-legged kittiwake, seasonal density and high use areas. ...................... 65
italicized species tended to roam into other assemblages with random
Figure 50. Common murre, seasonal density, high use areas and breeding colonies. . . 67
sampling.......................................................................................................... 28
iii
INTRODUCTION
In the spring of 2001, NOAA’s National Marine Sanctuary Northern Santa Barbara county (Figure 2). Based on the Cali-
Program (NMSP) and National Centers for Coastal Ocean fornia Department of Fish and Game’s 200-meter resolution
Ecological Linkages Report
Science (NCCOS), in consultation with the National Marine bathymetry data, this map displays the locations of prominent
PIs PIs develop
PIs deliver
Develop Statement of Ecolinkage
Fisheries Service (NMFS), launched a 24-month effort to bathymetric features occurring off the coast, including the con-
deliver presentation
iterative draft
Work and select PIs to Report
define and assess biogeographic patterns of selected marine tinental shelf/slope interface. In support of this assessment,
final of results
reports for
write the report
species found within and adjacent to the boundaries of three the National Geophysical Data Center, Monterey Bay Aquarium
report
NOAA review
west coast National Marine Sanctuaries. These sanctuaries, Research Institute, and the NMSP jointly developed the first
Biogeographic Analyses
Monterey Bay, Gulf of the Farallones, and Cordell Bank are high-resolution 70-meter bathymetry maps for the region. The
Evaluate data Revise Determine Phase 1
Conduct
conducting a joint review process to update sanctuary man- highly resolved bathymetry is shown on the three regional maps
Conduct
Identify
DATA GIS
sets and species and optimal analyses review and
species and
agement plans. The management plans for these sanctuaries to highlight the complexity of the seafloor and to begin outlining
Project habitat list analytical Analytical
conduct and develop revise
habitats of
have not been updated for over ten years and the status of the the multitude of distinct ecosystems occurring in this region
COLLECTION
preliminary based on data approach Results
draft
‘Kickoff analytical
interest with
natural resources and their management issues in and around (Figures 3-5). The Northern Region map focuses on Cordell
analyses quality and products for products
sanctuaries
Meetings’
the sanctuaries may have changed. In addition, significant ac- Bank and Gulf of the Farallones National Marine Sanctuar-
availability review
staff
with Team
complishments in research and resource assessments have ies. Cordell Bank, the Farallon Islands, Bodega and Pioneer
Members
been made within the region. Thus, it is important to incorporate Canyons, and Gumdrop and Pioneer Seamounts appear as
GIS Data Development Digital
new and expanding knowledge into the revised management very clear features along the seafloor. The focus of the Central
Products
Organize data and Integrate into NMSP’s
Data collection and Metadata
plans for these Sanctuaries. Region map is Monterey Bay. The most significant feature in
MarIS system
standardization creation transfer to CD-ROM
this regional view is the Monterey Canyon, although the nearby
As part of the review process, the NMSP requires an integrated canyons, Ascension, Año Nuevo, Cabrillo, Soquel, and Carmel,
Biogeographic
Interim Products
biogeographic assessment of the spatial and temporal distribu- are also labeled. Other significant bathymetric features in this
Assessment
tions of marine resources off north/central California. The NMSP region include the Guide Seamount, Sur Ridge, and Shepard
Draft
Interim Product:
Create web page for Phase I
Biogeographic
Biogeographic
headquarters and sanctuary field personnel have partnered Meander. The bathymetry for most of the Southern Region
project information and
Assessment
Assessment Atlas
product review
with NCCOS’s Biogeography Team to conduct this assessment. map is less resolved than the other two, as the frequency of
Biogeographic
The biogeographic assessment includes the identification and sampling was significantly less. In this area, the Sur and Lucia
Assessment
characterization of important biological areas and time periods Canyons are found, as well as Santa Lucia Bank and one of
Phase II
off the coast and addresses existing and emerging issues con- the most prominent features in this region, the Davidson Sea-
cerning management of biotic resources in the area. Results of mount. Descriptions of the features observed in these maps,
this assessment aid the NMSP in addressing issues, such as Figure 1. Project flow diagram showing steps to complete Biogeographic Assessment Phase 1. along with the linkages and processes operating to influence
potential modification of sanctuary boundaries and changes in the distribution of associated biota, are found in the Ecological
management approaches based on the principles of biogeog- CD-ROM on the back cover of this document. To enable devel- understanding the potential implications to changes in sanctu- Linkages Report.
raphy. The publication of this document completes Phase I of opment and integration of these components and to support ary boundaries or management strategies relative to marine
the biogeographic assessment for the North/Central California project management, the overall process used to conduct the biogeography within and adjacent to the sanctuaries. STUDY OBJECTIVES
National Marine Sanctuaries. The assessment and additional biogeographic assessment is shown in Figure 1. Based on consultations with NMSP field and headquarters staff
ecosystem characterization of habitats and species (e.g., es- and requirements to update the sanctuary management plans,
To enable NMSP and others to make maximum use of the spa-
tuaries) will continue over the next few years. The initial plans The Ecological Linkages Report is a comprehensive synthesis tial data generated from this study and other activities that are the following study objectives were addressed:
of existing information on ecological relationships between ma- supporting the joint management plan revision process (e.g.,
for Phase II are discussed in Section 6 of this document.
rine biota and the habitats they utilize along the West Coast. economic assessments), the NMSP is developing a GIS tool. 1. Identify and compile available priority biological and en-
The Phase I assessment is based on biogeographic patterns The report is much broader in geographic scope than the This GIS tool, the Marine Assessment and Resource Informa- vironmental data sets in the study area in order to conduct
of fishes, macroinvertebrates, marine mammals, and marine project study area and provides the context to understand tion System (MarIS), has been designed to facilitate the orga- biogeographic analyses.
birds and the distribution of their habitats. The study did not overall assessment results relative to the biogeography of the nization and analysis of spatial data to support NMSP manage-
attempt to define biogeographic patterns along the entire US West Coast. In addition, the report addresses near-shore and ment questions and issues within and across the sanctuaries. 2. Conduct marine biogeographic analyses of available data
west coast nor in very near shore environments (e.g, estuaries). estuarine ecosystems while quantitative data analyses were All GIS compatible data, the Ecological Linkages Report, and to define significant biological areas (i.e., “hot spots”) and time
Rather, the study area was restricted to the marine area from conducted only for the marine waters of the National Marine products from the biogeographic analyses component are found periods, based on species distributions, abundance, habitats,
Point Arena (in the north) to Point Sal, California (in the south). Sanctuaries. life stage function, and community metrics (e.g., species rich-
on the companion CD-ROM. The contents of the CD-ROM are
The Assessment was based on a synthesis of data provided ness, diversity).
also found on the web at http://biogeo.nos.noaa.gov/products/
by project partners (e.g., NMFS fishery independent surveys). The biogeographic analyses component includes a suite of canms_cd/. All of the applicable digital data will be integrated
The biogeographic assessment was formulated around three quantitative spatial and statistical analyses based on the dis- into MarIS. 3. Produce a report that describes the ecological components
integrated study components: 1) an Ecological Linkages Re- tribution and abundance of fishes, marine birds, and marine and linkages between the estuarine, coastal, and marine eco-
port, 2) biogeographic analyses, and 3) development of spatial mammals found in the Point Arena to Point Sal study area. The THE STUDY AREA systems of north/central California.
data for incorporation into NMSP’s Marine Information System analytical results contributed to defining important biological The study area, shown on the locator maps, extends from Point
(MarIS). GIS-based data and additional information, such as areas throughout the region, based on visualization of species Arena, in the southern portion of Mendocino county, to Point 4. Develop GIS compatible data for integration into the
the complete Ecological Linkages Report, can be found on the distribution patterns and community metrics. The results aid in Sal, just south of Pismo Beach and the Nipomo Dunes area in NMSP’s Marine Information System (MarIS) to support
1
INTRODUCTION
sanctuary staff in developing and evaluating resource manage- 124°W 123°W 122°W 121°W
ment scenarios.
39°N
39°N
Northern/Central California
Point Arena
5. Support sanctuary staff in the integration of biogeographic 3
na
assessment products into revisions of the sanctuary manage-
ut
ica
Region of Interest
ment plans.
lm Fort
ile Ross n
The publication of this product completes Phase I efforts to lin sia r
us ive
e
meet objectives 1-4. The data and analytical results from these RR
objectives will be used to address objective 5 over the next year. 0 10 20 40 60 80 100
This investigation synthesized many databases and information
Kilometers
sources for the study area. The data and information originated Cordell Bank
from a wide variety of government, academic, and private in- NMS
38°N
38°N
stitution studies that had different objectives, study areas, and
Gulf of the
methodologies. Thus, several criteria were used in selecting
Farallones
appropriate data sets for biogeographic analyses. For example, San
NMS
the selection process favored databases that addressed the en- Francisco
Gumdrop
tire study area and were conducted relatively consistently over
Seamount
time. Thus, small databases that were limited in both content
and spatial coverage were generally not useful in developing Pioneer Monterey Bay
Seamount
the assessment. When appropriate, these types of databases NMS
were used to aid in the interpretation of results and to develop
Santa
and validate species habitat suitability models. Cruz
37°N
37°N
Guide
Seamount
The following sections of this document provide information
on the data compiled, analytical approaches, and Phase I as-
Cany on
rey
sessment results. For the three main study components, (the
te
Ecological Linkages Report, biogeographic analyses, and the
n
Monterey
Mo
CD-ROM contents), the primary information found within each
component is introduced, methods described, and represen-
tative or example results provided. Many of the results are
presented as map products to easily convey the biogeographic
distribution of species and associated habitats. In addition, a
36°N
36°N
summary of the biogeographic assessment is found in section
Davidson
5. For more complete information (e.g., complete suite of digital Seamount
species maps), please review and use the digital contents of
the CD-ROM or the web version at http://biogeo.nos.noaa.gov/
products/canms_cd/. Cambria
35°N
35°N
Area
Point Sal
Enlarged
400
200 m
50 m
0 m
20 1000
m
00
Point Arguello
m
30
00
Point Conception
m
124°W 123°W 122°W 121°W
Figure 2. Locator map of entire study area from Point Arena to Point Sal. National Marine Sanctuary boundaries shown in red.
2
124°W 123°30'W 123°W 122°30'W 122°W
Bodega Head
Northern Region
10
20
00
0
m
Bodega
m
Bay
20
n y on Tomales Point
Bode g a Ca
00
m
30
0 5 10 20 30 40 50
00
50 m
m
Tomales Kilometers
Bay
mil utical
ne
Cordell Bank
e li
Drakes
a
3n
National Estero
San Pablo Bay
Cordell
Marine Sanctuary Limantour Estero
Bank
Drakes
38°N
38°N
Bay
Point
Fa
Reyes Bolinas
Double Point Lagoon
ra
ll
Gulf of the Farallones
n
o
Duxberry
National Reef
R
id
Marine Sanctuary
g Fanny
e North
Shoal
San Francisco Bay
Farallones
Fa
(esti mated sum
ra Middle
llo San
S
Farallon
n
a
Francisco
Es
n
ca
F
rp
ra
Southeast
m
mer
en Farallones
n
T id e nt)
c
t
e xt
al
is
Pacifica o
Plu
c
m
B
e
a
y
37°30'N
37°30'N
Gumdrop Pillar Point Half
Seamount
Moon
Bay
Monterey Bay
Pioneer
Area
Seamount
Enlarged National
Marine Sanctuary
er Canyon
ne
50
m
0m
o
100
Pi
20
m
m
0
3000
m
Pescadero Point
2000
124°W 123°30'W 123°W 122°30'W 122°W
Figure 3. Detailed locator map of northern study area from Bodega Head to Pescadero Point. National Marine Sanctuary boundaries shown in red.
3
123°30'W 123°W 122°30'W 122°W
Pescadero Point
Central Region
50
m
20
m
00
0
Pigeon Point
30
m
200
0 m 10
00
m
Point Año Nuevo
37°N
37°N
Santa
Guide
n
yo
Cruz
Seamount
an
nC
n
yo
an
si o
Elkhorn
C
en
o Slough
c
ev
Mo
As
yon
Nu
on
o
nte
an
ny
Añ
Ca
oC
Moss
rey Bay
el
il l
Landing
qu
br
3 n a line
Ca
So
mile
Monterey Bay
utica
National
Sa
Marine Sanctuary
l
lin
C ar m
C any o
a sR
Point Pinos ive
r
el
rey Ca
Monte
n
ny
on
Monterey
Cypress Point
Carmel Bay Carmel
Point Lobos
36°30'N
36°30'N
e g
Su r Rid
Point Sur
Big Sur
Area
Enlarged
Pfeiffer Point
20
50
10 m
00
00
Shepard m
30
m
200 m
00
Meander
m
123°30'W 123°W 122°30'W 122°W
Figure 4. Detailed locator map of central study area from Pescadero Point to Pfeiffer Point. National Marine Sanctuary boundaries shown in red.
4
123°W 122°30'W 122°W 121°30'W 121°W 120°30'W
Pfeiffer Point
100
20
30
200 m 50 m
0 m
00
00
Southern Region
m
m
on
ny
Monterey Bay
a
rC
Su National
Marine Sanctuary
Lopez Point 0 5 10 20 30 40 50
36°N
36°N
on
y
a C an Kilometers
ci yo n
Lu Mill Creek Ca n
Cape San Martin
a
Vill on
m
ny
00
Ca z
m 20
00 u
Cr n
30
La ny o
Ca
Davidson
Seamount
Point Piedras Blancas
Cambria
35°30'N
35°30'N
Point Estero
Estero
Bay
Morro
Bay
Point San Luis San
Sa
Luis Obispo
Bay
nt
a
Lu
50 m
ci
200
a
mile tical
35°N
35°N
Ba
line
m
Area
u
nk
3 na
Enlarged
30
4000 m
1000
Point Sal
00
20
m
00
m
m
123°W 122°30'W 122°W 121°30'W 121°W 120°30'W
Figure 5. Detailed locator map of southern study area from Pfeiffer Point to Point Sal. National Marine Sanctuary boundaries shown in red.
5
Section 1: SYNOPSIS OF ECOLOGICAL LINKAGES REPORT
INTRODUCTION a variety of coastal and marine habitats, such as rugged rocky
The Ecological Linkages Report complements the biogeo- shores, lush kelp forests, and several underwater canyons, the
graphic analyses conducted by The National Centers for largest of which is the Monterey Submarine Canyon. North of
Coastal Ocean Science (NCCOS) by providing an overview Partington Point and within the Gulf of the Farallones, the con-
of the physical and biological characteristics of the region. Key tinental shelf is relatively wide and shallow. South of Partington
ecosystems and species occurring in estuarine and marine Point, the Sanctuary generally protects deep ocean, owing to
waters off northern and central California are highlighted and the consistently narrow continental shelf that extends south to
linkages between them discussed. In addition, this report de- Point Conception. The diverse array of habitats in the Sanctu-
scribes biogeographic processes operating to affect species’ ary is home to 33 marine mammals, 94 species of seabirds,
distributional patterns. The biogeographic analyses build upon at least 345 species of fishes, and numerous invertebrates,
this background to further understanding of the biogeography and plants.
of this region. The following material is a synopsis of the report
either excerpted or directly summarized from the completed GEOGRAPHIC SETTING OF THE STUDY AREA
document (Airamé et al., 2003) found on the companion CD- The study region extends from Point Arena, a small peninsula
ROM. on an elevated coastal plain in the southern portion of Men-
docino County, to Point Sal, just south of Pismo Beach and
NATIONAL MARINE SANCTUARIES OFF CENTRAL AND the Nipomo Dunes area. The region consists of a multitude
NORTHERN CALIFORNIA of diverse and important ecosystems that are very unique in
The study area, from Point Arena to Point Sal, includes three their assemblages of marine organisms. Beginning near-shore,
national marine sanctuaries (Cordell Bank, Gulf of the Faral- the coast of California, especially north of Point Reyes and
lones, and Monterey Bay) encompassing marine and estuarine south of Point Pinos, is renowned for its strikingly beautiful,
habitats along the central and northern coast of California. dramatic rocky cliffs. Pocket beaches occur along the coast
Together, these contiguous national marine sanctuaries include where streams and rivers deposit sediment along the shore.
more than 650 km of coastline, from Bodega Bay, north of San Rivers that flow over broad, flat expanses of soft sediments
Francisco, to Cambria, near San Luis Obispo, and a total area into the ocean may be strongly influenced by tides and are
of approximately 18,000 km2. frequently associated with upland and salt marshes, sandy
beaches, intertidal flats, and estuaries. Estuaries and lagoons
The Gulf of the Farallones National Marine Sanctuary, estab- commonly form where rivers enter the ocean, mixing fresh and
lished in 1981, includes an area of 3,250 km2 off the northern salt water. Rocky shores, which are more resistant to erosion
and central California coast. The Gulf of the Farallones extends than the sandy beaches, support a complex intertidal com-
beyond the Sanctuary’s boundaries and is one of the broadest munity, influenced primarily by the semidiurnal movements of
sections of the continental shelf off the U.S. West Coast. Be- tides. Moving offshore, subtidal communities are strongly influ-
sides the broad shelf, the major oceanographic feature that af- ocean’s surface. The base of the Bank is over 120 m deep. The enced by sediment type, nutrient input, and depth. The majority
at all levels of the marine food web. In periods when upwelling
fects this coastal region is the San Francisco Bay Plume, which, combination of oceanographic conditions and undersea topog- of the continental shelf is sandy, but rocky outcrops cover a
is reduced, the nutrient input from the San Francisco Plume
under certain conditions, extends outwards to all areas of the raphy of Cordell Bank supports a diverse and productive marine portion of it, forming submerged reefs, seamounts, and other
becomes important. The Farallon Islands, which are protected
Gulf. The Golden Gate, from which the plume emanates, lies ecosystem. A persistent upwelling plume projects southward features. Marine algae, unable to attach to the shifting sandy
as a National Wildlife Refuge, are home to the largest concen-
midway along this section of coast. The Gulf of the Farallones and offshore from Point Arena and Point Reyes, transporting sediments, find more secure substrate on rocky reefs. At the
tration of breeding seabirds in the contiguous United States (12
National Marine Sanctuary itself, however, extends along the nutrients and organisms suspended in the water column into the shelf break, the continental slope drops precipitously to depths
species), as well as one of the richest assemblages of pinnipeds
coast only as far south as Rocky Point, Marin County (where bank’s relatively shallow waters. Insolation fuels primary pro- of over 3000 m. Sediments, transported down the continental
(5 species). About 163 species of marine, coastal, and estuarine
the Gulf of the Farallones National Marine Sanctuary abuts the ductivity and eventually influences the entire food web through slope and submarine canyons, collect in broad fans at the
birds and 36 species of marine mammals use the Sanctuary
Monterey Bay National Marine Sanctuary). Offshore, the Gulf direct and indirect trophic linkages. This high local productivity base of the slope. Below the rise, the abyssal plain is relatively
during breeding or migration. Further, great white sharks are
of the Farallones National Marine Sanctuary extends farther supports abundant resident populations of invertebrates, fishes flat, broken occasionally by such features as seamounts and
attracted to marine mammal colonies on the Farallon Islands,
south to the waters west of San Mateo County. Habitats within (240 species), seabirds (69 species), and marine mammals (28 small depressions. It is this array of ecosystems, combined
Point Año Nuevo, and Año Nuevo Island.
the Gulf of the Farallones National Marine Sanctuary include species), and attracts many migratory species. with the oceanographic processes affecting the composition
rocky shores, sandy beaches, estuaries, lagoons and bays, and abundance of marine organisms in them, that make this
The Cordell Bank National Marine Sanctuary, designated in
as well as the Farallon Islands and the subsurface Farallon The Monterey Bay National Marine Sanctuary, established in such a unique area.
May 1989, includes an area of 1,362 km2 off the coast of central
Ridge. The entire stretch of the broad shelf within the physi- 1992, is the largest of 13 marine sanctuaries administered by
California (Figure 2: Northern Region). Cordell Bank is located
cal features described above is strongly influenced by coastal the National Marine Sanctuary Program. The Sanctuary ex- OCEAN CURRENTS
at the edge of the continental shelf, about 80 km northwest of
upwelling and the San Francisco Bay Plume. The upwelled tends from Rocky Point to Cambria Rock, encompassing nearly The cold water California Current and comparatively warm-
the Golden Gate Bridge and 33 km west of Point Reyes. The
waters, which support tremendous phytoplankton production, 450 km of shoreline and 13,780 km2 of ocean, extending an water Davidson Currents are major forces shaping the eco-
main feature of the Sanctuary is an offshore granite bank, 7 km
are advected offshore into the California Current as eddies and average distance of 32 km from shore. At its deepest point, the systems in and around the study region. They affect upwelling
wide and 15 km long. The rocky bank emerges from the soft
jets. These productive waters stimulate growth of organisms Sanctuary reaches a depth of 3,250 m. The Sanctuary includes and downwelling and, consequently, the amount of productivity
sediments of the continental shelf, reaching within 37 m of the
6
Section 1: SYNOPSIS OF ECOLOGICAL LINKAGES REPORT
fishes decrease with plankton abundance. Marine mammals ing spring and fall migrations and the winter months. Migratory
and seabirds, which depend on these organisms for food, species require consistent sources of food and shelter along
suffer food shortages, leading to widespread starvation and their migration route. If the distances between wetlands are too
decreased reproductive success. large, migrating birds may become exhausted and disoriented,
increasing mortality.
Every 20-30 years, the surface waters of the central and
northern Pacific Ocean shift several degrees from the mean Numerous marine species use embayments, lagoons, and
temperature. Such shifts in mean surface water temperature, estuaries as spawning and nursery grounds. Bat rays, leop-
known as the Pacific Decadal Oscillation, have been detected ard and smoothhound sharks, midshipman, Pacific herring,
5 times during the past century, with the most recent shift in starry flounder, staghorn sculpin, surf perch of several species,
1998. The Pacific Decadal Oscillation impacts production in jacksmelt, topsmelt, and pile perch mate and bear their young
the eastern Pacific Ocean and, consequently, affects organism in estuarine habitats. Shallow, coastal waters of central and
abundance and distribution throughout the food chain. northern California also are critical habitat for chinook and,
especially, coho salmon as they travel en route to spawning
Ocean waters off the coast of California have warmed consid- grounds in autumn and winter.
erably over the last 40 years. It is not yet clear if this warming
is a consequence of an interdecadal climate shift or global Geology and other physical forces influence the structure of
warming. the coastline. Energetic forces of water and wind erode the
rocky coastline, creating the dramatic rocky intertidal habitat
In response to these three phenomena, some species have characteristic of the northern and central California coast. Small
shifted their geographic ranges northward, altering the com- beaches form along the northern California coast where wind
position of local assemblages. and waves erode granite and basalt cliffs. Further sourth, ero-
sion of soft shale and sandstone bluffs creates the broad sandy
ECOSYSTEMS beaches typical of southern California.
The Land-Sea Interface
Rivers carry freshwater and sediments to bays, estuaries, and Sandy beach and rocky intertidal habitats are divided into dis-
the ocean. Thirteen major watersheds are located along the tinct biological zones relative to height above mean high tide.
central California coast. Historically, these supported large In part, species’ distributions are affected by their physiologi-
numbers of coho and chinook salmon, steelhead trout, and cal tolerance to temperature, moisture, and salt. In response
sturgeon. Today, many native anadromous fish stocks through- to these and other physical factors, the number of species of
out California are in danger of extinction. General degradation marine algae, gastropods, and fishes increases with depth in
of upland watershed and freshwater ecosystems is a major the intertidal zone.
factor in the decline.
seen along the coast. Additionally, where the two converge the study region. Over a 12 month time-frame, the study area
at Point Conception, a barrier is created that many species is exposed to three distinct oceanographic periods that vary Nutrients processed in marine systems are essential to commu-
Two major estuaries in northern and central California are
do not cross. The species north of Point Conception, encom- with respect to prominence and location of ocean currents. nities using sandy beaches and rocky intertidal habitat. Waves
San Francisco Bay and Tomales Bay. Several smaller estu-
passing the entire study region right up through Washington These periods, described by upwelling (March to August), wind carry and deposit plankton, macroalgae, as well as occasional
aries and lagoons within the region, from north to south, are
State, are a part of the Oregonian Province, while just south relaxation (August to November), and winter storms (November corpses of fishes, birds, and marine mammals in the intertidal
Estero Americano, Estero de San Antonio, Bolinas Lagoon,
of Point Conception, they belong to the Californian Province. to March), are associated with different degrees of upwelling zone, which provide an unpredictable and patchy source of
Drakes/Limantour Estero, and Elkhorn Slough (National Es-
Although many species have ranges that end at the borders or downwelling. The amount of production in surface waters food. Beach wrack attracts numerous mobile organisms, in-
tuarine Reserve). Estuaries and bays are vulnerable to coastal
of these biogeographic zones under normal conditions, spe- and the ability of organisms to disperse is directly impacted cluding amphipods, isopods, flies, beetles, and shorebirds. The
development, pollution, introduction of invasive species, and
cies of the subtropical Californian Province may occasionally by these processes. In response to these periods, the abun- sporadic deposition of food from the ocean sustains intertidal
commercial and recreational fishing for species that live in
extend their ranges to central and northern California during dance and types of organisms present in a given region change communities in habitats that are subjected to strong physical
near-shore waters. Humans have modified and transformed
unusually warm oceanographic events, such as El Niño and throughout the year. forces and relatively low local production.
about 90% of the wetlands in California. The existence and
the Pacific Decadal Oscillation. Other more localized oceano-
health of these coastal wetlands is critical to the survival of
graphic features, such as eddies, internal waves and bores, NATURAL PERTURBATIONS Marine Ecosystems
organisms that depend on these habitats for survival. One of
are also important factors influencing the distribution and Longer term climatic phenomena influencing the region include: Production in subtidal habitats depends on levels of light and
California’s wetland sites, Bolinas Lagoon, was designated as
abundance of marine species. El Niño, Pacific Decadal Oscillation, and global warming. Off nutrients, and exposure to physical forces. Sufficient light to
internationally important in this role under the Convention on
the coast of California, El Niño events are characterized by support highly productive photosynthetic communities pen-
Wetlands, signed in Ramsar, Iran in 1971.
OCEANOGRAPHIC SEASONS increases in ocean temperature and sea level, enhanced on- etrates surface waters to approximately 30 m. Kelp, which
While certain geological and biological features are evident shore and northward flow, and reduced coastal upwelling of can grow up to 10 cm per day, is among the most productive
Wetlands along the central California coast are sparse, but
along particular regions of the coast, the same oceanographic deep, cold, nutrient-rich water. During this period, survivorship of marine plants. Primary productivity in kelp forests has been
those present support millions of shorebirds and waterfowl dur-
processes and climatic phenomena are operating throughout and reproductive success of planktivorous invertebrates and estimated at 350 to 2,800 grams of carbon per square meter.
7
Section 1: SYNOPSIS OF ECOLOGICAL LINKAGES REPORT
the northeastern Pacific Ocean where production is particularly
(Cottidae) occur in shallow water and tidepools, as well as in
high, approximately 5-15% of the surface production eventually
deeper water around kelp forests, rocky reefs, and sand or mud
reaches the deep sea.
bottoms on the continental shelf. Lingcod (Hexigrammidae),
commonly associated with shallow rocky reefs, also occur in
In a few places, extinct volcanoes or seamounts disrupt the
waters as deep as 300 m.
monotony of the abyssal plain. Off central California, several
seamounts (Gumdrop, Pioneer, Guide, and Davidson) are
Soft bottom habitats on the continental shelf lack the physi-
located near the bottom of the continental slope. Seamounts
cal structure and high biological production of kelp forests
provide physical structures which support complex deep-sea
and rocky reefs. Species that live in soft sediments on the
communities of benthic invertebrates and some fishes.
continental shelf are subjected to shifting sediments through
wave action and near bottom currents. Some species that live
Offshore Islands
in these habitats, such as crustaceans and mollusks, secure
The California coast is a tectonic subduction zone, inhibiting
themselves in tubes and burrows. Other species, such as flat-
the formation of offshore islands. The few that do exist are
fishes, are camouflaged on sandy sediments of the seafloor
extremely important sites for breeding by seabirds and pin-
by their color and shape.
nipeds. The largest offshore islands in the study region are
the Farallon Islands, west of San Francisco. The Farallon
Deep submarine canyons, such as Monterey, Ascension, Pio-
Islands support some of the largest colonies of breeding sea-
neer and Bodega, are remnants of riverbeds that deeply incised
birds south of Alaska. Numerous marine mammals, including
the continental shelf and slope during glacial periods. Because
northern elephant seals, Steller sea lions, harbor seals, and
they cut into the continental shelf, submarine canyons support
fur seals, haul out and breed on the Farallon Islands as well.
deep-sea communities relatively close to shore. Canyon walls
Other important but much smaller breeding populations occur
are often steep and rocky, providing shelter for species, such
on rocks off Point Reyes, Año Nuevo Island, and on rocks off
as rockfishes and thornyheads, which are associated with
Big Sur. Subsurface features, e.g., the Farallon Ridge, Cordell
complex physical structures. Canyon bottoms tend to slope
Bank, and various seamounts, provide substrate and protec-
gently and accumulate finer sediments, such as silt and mud,
tion for diverse communities of benthic invertebrates and some
providing habitat for species such as flatfishes. In addition,
fishes, as well.
the structure of submarine canyons affects the circulation of
near-shore waters and the concentration of organisms in the
BIOGEOGRAPHY
water column.
An understanding of biogeographic patterns and how they are
influenced by ecological linkages enables management deci-
Submarine canyons, submerged volcanoes, and other physical
sions to be placed in a spatial context relative to the distribu-
features under high pressure often concentrate gases and fluids
tion of marine resources. Distributions of marine species are
beneath the sea floor. In some areas, where the sea floor is
determined by oceanographic phenomena, physical tolerances,
weak, these gases and fluids may be forced through the sedi-
and biological interactions. Each species responds to these
ments, creating features known as cold seeps. Most cold seeps
Kelp provides substrate for numerous benthic and epibenthic Productivity from seaweeds can also have indirect effects on
factors in slightly different ways. Despite the physiological and
are found in the deep sea (600-3000 m) under conditions of
invertebrates, as well as food and shelter for many fishes, coastal food webs. Particulate and dissolved organic carbon
ecological differences between species’ response, there are
low light, temperature, and oxygen, and high pressure. In spite
seabirds, and marine mammals. Colonies of bryozoans grow that results from fragmentation and decomposition of kelps
many similarities in species’ distributions, which can be used
of these difficult conditions, numerous organisms are adapted
on kelp fronds. Several species of snails, including purple ring and other seaweeds can be consumed by suspension-feeding
to define biogeographic regions. The transitions between bio-
to life around cold seeps. Vesicomyid clams are the dominant
top snail and blue top snail, feed on kelp, while kelp crabs zooplankton or benthic invertebrates, providing a trophic link
geographic regions are more distinct outside the study area
species at cold seeps off central and northern California. These
cling to the underside of kelp fronds. During periods of low between kelps and higher-level pelagic consumers, such as
than within.
clams support chemoautotrophic bacteria in a symbiotic re-
productivity, sea urchins may emerge from protective crevices fish. A small portion of the drift algae may be transported off
lationship. The bacteria use inorganic chemical compounds
in rocky reefs to graze on kelp. At the surface, floating kelp the reef, where it can contribute to production in submarine
The geographic distributions of numerous marine organisms
released by the cold seeps to produce organic compounds,
masses are important habitats for juvenile fishes, particularly canyons and the deep sea.
of the northeastern Pacific Ocean coincide with major oceano-
which are used by their vesicomyid clam hosts.
rockfishes and kelp surfperch. Schools of blue, black, and kelp
graphic shifts. The biogeographic boundary at the Gulf of Alaska
rockfishes and bocaccio are generally recorded in the midwater Productivity is reduced on rocky reefs below 30 m, where light
occurs at the transition between sea and land along the south
Deep-sea communities depend on the distribution and quantity
kelp canopy. Gopher, copper, black, and yellow rockfishes, levels are low and kelp is unable to flourish. However, the physi-
coast of Alaska. The biogeographic transition at Vancouver
of primary production in surface waters, the rate of movement
lingcod, cabezon, and greenlings tend to associate with the cal structure of rocky reefs does provide shelter for numerous
Island corresponds to the eastern portion of the North Pacific
of organic material to the bottom, and the conditions of deposi-
bottom of the kelp fronds. In addition, the sea otter has been benthic invertebrates and fishes. Shortbelly rockfish (Sebastes
Drift, which bifurcates in this region with part diverted north into
tion and transformation of the organic matter in the sediment.
described as a “keystone species” for its role in structuring kelp jordani), the most abundant rockfish species on the continental
the Gulf of Alaska and part diverted south along the western
A portion of dead organic matter produced in surface waters
forest communities through predation of herbivores, particularly shelf and upper slope off California, are often associated with
coast of North America as the California Current. The biogeo-
is transported to the sea floor either through passive sinking,
sea urchins, resulting in increased kelp growth. rocky reefs between 30-80 m depth. Various seabirds and
graphic transition at Point Conception corresponds to a shift in
or by active transport during vertical migration of plankton. In
marine mammals rely on shortbelly rockfish for food. Sculpin
8
Section 1: SYNOPSIS OF ECOLOGICAL LINKAGES REPORT
the oceanographic regime. At Point Conception, the California results from the biogeographic analyses are interpreted within
coastline turns abruptly east and the cool water moving south in the context of the Ecological Linkages Report as it demon-
the California Current is diverted offshore. The most significant strates the understanding that the patterns presented in the
biogeographic boundary in the study region occurs at Monterey Geographical Information System are dynamic in nature due
Bay; however, other minor boundaries occur around points and to the multitude of factors operating to shape them. Changes in
bays along the coast. any of these factors can result in changes to the biogeography
of the region.
In addition to the changes in latitudinal distributions, the diver-
REFERENCE
sity of species changes with depth. The changes in species
Airamé, S., S. Gaines, and C. Caldow. 2003. Ecological Link-
composition and abundance are associated with physiological
ages: Marine and estuarine ecosystems of central and northern
tolerances for temperature, exposure, light and nutrient input, as
California. NOAA, National Ocean Service. Silver Spring, MD.
well as a wide range of biological interactions among species. At
163 p.
all latitudes, the average number of species of algae and marine
gastropods increased with depth from high to low intertidal and
subtidal zones. In addition, species that occur across several
depth zones are likely to have broader latitudinal distributions
than species that occupy a single depth zone. In contrast to the
patterns observed for marine algae and gastropods, the aver-
age number of fish species declined with latitude and depth.
The greatest numbers of fish species occurred south of 50oN
latitude and shallower than 200 m.
For some species, the range of single individuals spans nearly
the entire geographical distribution of the species. These spe-
cies use local resources during long-distance migration, but no
individual resource supports a resident population. Examples
of these species include baleen whales that feed at highly
productive sites along their migration route, and seabirds that
use estuaries along the coast as resting and feeding sites
during their annual migrations. For other species, the entire
geographical range far exceeds the range of an individual.
Many intertidal invertebrates and fishes have dispersal and
sedentary phases during their life cycles. Examples of these
species include barnacles, mussels, and clams that settle into
intertidal habitats, and rockfish that settle into kelp forests or
rocky reefs after a pelagic larval stage.
CONCLUSIONS
Within the study region there are many distinct ecosystems
each hosting a unique assemblage of organisms. In addition
to describing these key ecosystems and species in the region,
the Ecological Linkages Report provides information on link-
ages within and between these systems. By understanding
the climatic, oceanographic, physical and biological influences
operating together to shape the regional biogeography, the
background exists for the biogeographic analyses to be inter-
preted. The "Biogeographic Analyses" section complements the
synthesis of literature in the Ecological Linkages Report with a
data driven look into the biogeographic patterns evident around
the sanctuaries. The analyses provide a spatially explicit view
of marine resources within the study area from which manage-
Annie Crawley
ment decisions can be better enacted. It is important that the
9
Section 2: BIOGEOGRAPHIC ANALYSES
INTRODUCTION species within a group, such as by family (e.g. rockfish), and
The biogeographic analyses component is the cornerstone of comparison of spatial patterns between groups. Analysis of
Integrated Analyses
Analyses
the overall assessment to support the joint management plan spatial patterns resulted in information on the relationships
Data Layers
revision process. The data, analyses, and supporting informa-
and Products to between individual species, between assemblages of species,
and Products
tion are linked using statistical and GIS tools to visualize the and of the relationship of species to specific environmental
Aid Sanctuary
location of significant biological areas or “hot spots.” There were and habitat parameters. Furthermore, the compilation and
many different ways to analyze and organize the biological integration of individual species maps were used to calculate
Management
data compiled for this assessment. To efficiently support the community metrics, such as total richness or diversity of fish
management plan revision process, only a limited number of and marine bird species, at a specific location.
Study Area
analytical options were selected based on reviewer’s comments
on the Interim Product, mission of the NMSP, technical review To define species assemblages, multivariate techniques were
tem
meetings, and peer review workshops. These key analyses are applied to various data sets to group organisms found at spe-
s
Catch Data Species Distributions
ion Sy
presented in this document and on the CD-ROM. In addition to cific sampling sites. The assemblage analyses defined species
these results, spatial data and information on the companion groups across the study area. By visualizing the assemblages
CD-ROM enable NMSP staff, advisory councils, and research geographically, areas of overlap became apparent and group
t
forma
Sightings
partners to conduct additional analyses not specifically ad- habitat affinities, such as depth range, were delineated.
dressed in this product.
n
Species habitat suitability index modeling (HSI) studies were
phic I
A critical step in the biogeographic analyses component was the undertaken for 20 fish and invertebrate species in an attempt
Community Distributions
Bathymetry
extensive effort to have data, analytical approaches, and results to characterize areas within the study region that suffered a
eogra
peer reviewed. Initial results from the suite of analyses were lack of sampling data, particularly in near-shore habitats. The
presented to experts on marine ecosystems of north/central integration of HSI models into a GIS provides a spatial depic-
ated G
Substrate
California, as well as to the originators of the data sources in tion of species habitat suitability models for individual species
an attempt to improve the analyses. The role of expert review by integrating information on species habitat affinities and the
and input has been considerable, and the contributions made distribution of those habitats in space and time (Brown et al.,
Integr
Temperature
by experts have significantly enhanced the analyses. 2002; Monaco and Christensen 1997). The modeling compo-
Modeled Distributions nent of the biogeographic analyses is a necessary step due to
ASSESSMENT PROCESS the incomplete distribution of sampling data across the entire
Life History
To aid in focusing on the most important analyses, the biogeo- study area. Thus, species that were representative of the as-
graphic assessment process displayed in Figure 6 was utilized. semblages described above and/or other key species were
Data Significant Biological
This process is currently being implemented through a joint selected for modeling their potential distribution. The composite
Areas
NMSP and NCCOS Biogeography Team effort to initiate bio- set of species habitat suitability models contributed to defining
geographic assessments across all sanctuaries within the next significant biological areas within the region.
Ecological Linkages Report
five years. The process is organized around development of
biogeographical data layers, integrated analyses, and specific Measures of community structure for fishes and marine birds
Figure 6. Biogeographic assessment process.
products to aid in sanctuary management plans (Kendall and were calculated independently by species group, compared,
Monaco 2003). Thus, the integration of partner’s comments and, where applicable, integrated. Convergence of overlapping
and use of the biogeographic assessment process resulted in spatial patterns defined significant biological areas based on
To develop this capability, a suite of analyses were conducted
species richness statistics), to be directly linked to specific
the analyses and results presented in this document. a number of criteria (e.g., high species abundance, high spe-
that were most appropriate in addressing NMSP natural re-
areas or habitats they correspond to across the study area.
cies diversity).
source management issues. The biogeographic assessment
The GIS also facilitated integration of multiple data types and
Biogeographic data assembled for this project were derived framework (Figure 6) aided in targeting the suite of analytical
sources into a common spatial and temporal framework (Gill
from many sources (see section 4), including NOAA Fisher- Thus, the biogeographic analyses component was a result of
approaches to define biologically significant areas in support
et al., 2001). The following suite of map products quantitatively
ies, academia, state government, and data housed within the interpreting or visualizing the analytical results from statistical
of the sanctuary management plan reviews. Categories of
defined significant biological areas that are within or adjacent
NMS and Biogeography Programs. The biogeographic data analyses, ecological modeling, and integration of results across
analysis include: temporal and spatial analysis of individual
to existing boundaries of Cordell Bank, Gulf of the Farallones,
cut across various themes, such as species distributions and biota and habitats. The cumulative results aided in assess-
species’ distributions, species assemblage analyses, habitat
and Monterey Bay National Marine Sanctuaries. The GIS-based
habitats, and are integrated using a common spatial framework ing the biogeographic patterns in the study with regard to the
suitability modeling, and community metrics within and across
products are intended to aid in evaluating current sanctuary
in a GIS. The GIS enables a user to select particular data layers distribution of individual species, species assemblages, and
species groups. Important individual species maps were de-
boundaries relative to biological resources and habitats (Fig-
to be displayed, combined, and manipulated in a wide variety species habitat utilization patterns.
veloped from a number of data sets to visualize species pres-
ure 7), explore options for environmental protection of existing
of ways to achieve specific analytical objectives (Figure 6). ence and/or abundance data within the study area by season.
NMS areas, identify additional biologically important areas, and
REFERENCES
Where possible, well-established breeding colonies, rookeries,
evaluate alternative management strategies.
The use of the GIS enabled species-specific data, such as Brown, S.K., K.R. Buja, S.H. Jury, M.E. Monaco, and A. Banner.
and high concentrations of species are displayed on the digital
distribution and abundance data or community metrics (e.g., 2000. Habitat suitability index models for eight fish and inver-
maps. The single species maps enabled various groupings of
10
Section 2: BIOGEOGRAPHIC ANALYSES
tebrate species in Casco and Sheepscot Bays, Maine. North
American Journal of Fisheries Management 20:408-435.
Gill, T.A., M.E. Monaco, S.K. Brown, and S.P. Orlando. 2001.
Three GIS tools for assessing or predicting distributions of
species, habitats, and impacts: CORA, HSM, and CA&DS. In:
Proceedings of the first international symposium on geographic
information systems (GIS) in fishery science. Nishida, T., Kai-
lola, P. and Hollingworth, C.E. (Eds.). pp 404-415.
Kendall, M.S. and M.E. Monaco. 2003. Biogeography of the
National Marine Sanctuaries: A partnership between the NOS
Biogeography Program and the National Marine Sanctuary
Program. January 2003. 8pp.
Monaco, M.E. and J.D. Christensen. 1997. Biogeography Pro-
gram: Coupling species distributions and habitat. Pages 133-
139 in G.W. Boehlert and J.D. Schumacher, editors. Changing
oceans and changing fisheries: Environmental data for fisheries
research and management. National Marine Fisheries Service
Technical Memorandum NOAA-TM-NMFS-SWRSC-239, Pa-
cific Grove, California.
Figure 7. 3-D image of bathymetric relief within and adjacent to the sanctuaries.
11
Section 2.1: BIOGEOGRAPHY OF Approach to Fish Analyses
Biogeographic FISHES
INTRODUCTION trawls on the continental shelf and slope or de-
The biogeography of fishes section is a robust rived from species habitat affinities described in
statistical analysis of fish and a few economically the literature. Model results were validated with
1. Assemblage Analyses
important macro-invertebrates. A two-pronged ap- NMFS trawls on the shelf and slope (a different
proach was conducted to examine both fisheries- subset than that used to create the affinities) or
dependent and independent catch data and model CDF&G recreational catch data. Model results
Results:
Data Input: Analytical Approach:
the potential distribution and relative abundance of for 3 species are included in the main body of
Define and Interpret
NMFS shelf Clustering of species
selected species (Figure 8). Analysis of fisheries this document, with the models for the rest of
data can be a slow process which often requires the species included in the CD-ROM.
Biological Hot spots.
NMFS slope assemblages, site groups,
extensive exploratory statistical techniques to in-
CDF&G recreational species diversity, and (e.g. assemblages of
crease our understanding of the data before pre- The integration of the fish and invertebrate
NMFS midwater richness. station/species,
senting reliable and salient results. Many data sets analyses is shown in Section 3. For example,
diversity, richness)
were evaluated, however, only four data sets that a comparison of important areas derived from
were spatially comprehensive within the study area the diversity of NMFS trawls and those derived
were analyzed: 1) the California Department of Fish from overlays of the HSM’s was conducted to
Overlay, compare, and
and Game fishery dependent recreational fishing define significant biological areas. In addition,
combine with marine
trips targeting rockfish (CDF&G recreational); 2) the aspects of the Ecological Linkages Report were
bird results.
National Marine Fisheries Service fishery indepen- qualitatively incorporated into the assemblage
dent benthic trawls on the continental shelf (NMFS and HSM discussions to aid in interpretation of
Integrate into final
shelf trawls); 3) NMFS fishery independent benthic these analyses.
report.
trawls on the slope (NMFS slope trawls); and 4)
the NMFS fishery independent trawls in midwater
2. Habitat Suitability Modeling
(NMFS midwater trawls). Detailed information on
these trawls is given in each respective section.
Analytical Approach: Results:
Data Input:
The NMFS trawls on the continental shelf and slope
Literature review
provide information on the diverse demersal fish as- Distribution maps for Define and Interpret
NMFS trawl data
semblages found on trawlable habitats between 50
representative species Biological Hot spots.
GIS habitat layers
and 1280 meters depth throughout the study area.
based on habitat (e.g. overlay maps of
Key species used:
Pelagic fish encountered either as the trawl de-
suitability indices. modeled species to
scended or ascended are also included with these Representatives of local
determine locations of
assemblages, or species
analyses. The CDF&G recreational hook and line Statistical validation with
co-occurrence)
of economical/ecological
data complements the NMFS data sets by provid- catch data.
importance.
ing information on midwater as well as demersal
species. The recreational data was collected over
soft bottom and hard bottom habitats between
5 and 200 meters depth. The NMFS midwater
trawl data targets juvenile rockfish and provides Figure 8. Biogeographic approach to fish analysis.
information on fish and invertebrates found in the
neritic environment during the upwelling season.
Even though the spatial extent of this data set does not cover all four data sets, 119 species were analyzed. A complete list of fish and invertebrate species across the study area (Brown
the entire study area, it provides a source of information on of species included in the assemblage analyses can be found et al., 2000). Thirty two species of fish and invertebrates were
the neritic environment, which is important as juvenile habitat, on the CD-ROM. In addition, community metrics, including initially investigated through literature searches to determine if
and as the base of the food web for marine birds, fish, and species richness, species diversity, and rockfish richness sufficient information was available to model potential distribu-
marine mammals. were calculated for the NMFS shelf and slope trawls and tions. The process of determining which species to include in
the results presented spatially using GIS. However, due to the modeling procedure included consultation by the sanctu-
Due to time constraints, analysis of all species individually spatial and temperal limitations of available data, (i.e. NMFS ary staff, and integration of information on the economic and
was not feasible. Instead, all four data sets were analyzed demersal trawls have no information on rocky or shallow areas ecological importance of the species as well as initial results on
using multivariate statistics to identify species assemblages, (<50 meter) and contain trawls only for the months of June species assemblages. Of the 32 species initially investigated,
site groups, and the location of the species assemblages in through November) areas such as Cordell Bank and the Far- there was sufficient information available to conduct HSM on
space using a Geographic Information System (GIS). For the allon Islands are not adequately sampled. Therefore, habitat the adult and subadult stages of 14 fish species, and adult stage
multivariate statistics, species were included in an analysis if suitability modeling (HSM) was conducted to supplement the of 4 fish and 2 invertebrate species. Habitat suitability models
they were captured in at least 5% of the collections. Through analysis of catch data, and to model the potential distribution were either derived from an analysis of a portion of the NMFS
12
Subsection 2.1.1: ASSEMBLAGE ANALYSES
in content and spatial coverage was that the rockfish management groups could be defined,
INTRODUCTION at the effect these trawls had on the
Mutivariate Analyses of Fisheries Dependent
were not utilized in this analysis, and that both depth and latitude were important.
No species exists in isolation from other species or their envi- estimates of biomass of selected
and Independent Data
but were used to help interpret
ronment. Monitoring species individually may cause managers species through time. Based on
the results. Even though the neritic zone is ecologically important, little re-
to miss important interactions (Chavez et al., 2003; Worm and Zimmerman’s results, we excluded
search has addressed the midwater environment. Cailliet et al.
Myers, 2003; Baraff and Loughlin, 2000; Estes et al., 1998). these abnormal “water hauls” from
Five specific objectives of the (1979) described fish and invertebrate species co-occurrences
In addition, individual species' abundance may be considered our analyses. Williams and Ralston
assemblage analysis (all of in anchovy purse seines, and midwater trawls. An extremely
sustainable in the context of a fishery, but still be low enough (2002) analyzed data from NMFS
which aim to increase our under- thorough report by Larson et al. (1994) looked at the NMFS
to influence ecosystem dynamics and health (NMFS 2001). shelf trawls to determine rockfish
standing of the biogeography of midwater trawl results in conjunction with local environmental
There has been a growing recognition that effort needs to be species assemblages. The same
fishes and macro-invertebrates conditions to determine juvenile rockfish assemblages. Their
extended toward understanding the entire ecosystem. The data were used in this analysis;
in relationship to their environ- results emphasize the ephemeral nature of the pelagic environ-
National Marine Sanctuary Program is tasked with ensur- however, the multivariate statisti-
ment, as well as identify impor- ment, but they were able to document two consistent spatial
ing the continued health of the ecosystems contained in the cal method utilized, the spatial
tant areas or habitats within the trends: 1) the rockfish are larger inshore than offshore, and 2)
sanctuaries. However, important species-species interactions coverage employed, and the spe-
study area), were as follows: there was a north/south gradient in species composition and
and species-habitat interactions are still not well understood. cies examined were different. The
1. Identifying spatial patterns abundance. Moser et al (2000) described changes in rockfish
Abiotic (e.g., habitat preferences toward depth or sediment) overall conclusion from Williams and
and hot spots in community larvae abundance in CalCOFI plankton tows from 1951 to 1998
and biotic (e.g., presence or absence of prey, predators) fac- Ralston was that rockfish richness
metrics (diversity and richness); in response to adult biomass and environmental conditions.
tors can impact the importance of an area to fish. Elucidating was highest at a depth of 200-250 meters, where the shelf
2. Determining which species tended to be caught together He concluded that over-fishing as well as decadal shifts in
habitat characteristics that are most important to animals, and and slope meet, and that depth and latitude were the main
(species assemblages); environmental conditions were affecting the stocks.
understanding the co-occurrence of species, is a first step in determinants of rockfish assemblages. Jay (1996) analyzed
3. Analyzing fishing locations to determine which locations
determining areas that should be managed as “essential” habi- the 1977-1992 NMFS shelf trawls to determine site groups
contained similar catches (site groups); Underwater submersibles have been used to describe fish as-
tats. This study aids in clarifying the interactions among species that contained similar catches. Using 33 species of fish, he
4. Resolving where the species assemblages were being semblages and their interaction with habitat at spatial scales
and between broad scale habitat characteristics and species identified 23 site groups, many of which contained the same
caught by combining results from objectives 2 and 3 and relevant to the fish themselves. Yoklavich et al. (2000 and
on the scale of the commercial and recreational fisheries. Even species, but with different relative abundance. Even though
then utilizing GIS to map the results; and 2002) surveyed Soquel Canyon and Big Creek Ecological
though these data sets were originally deployed to collect infor- depth and latitude showed some influence on site groups,
5. Identifying significant relationships between site groups Reserve on the Big Sur coast, Field et al. (2002) looked at
mation necessary for setting fishing limits, these data sets can overall he found little association between the site groups and
identified in objective 3 and broad scale habitat character- Big Creek Ecological Reserve, while Hixon et al. (1991) and
provide preliminary information on multi-species interactions. a suite of environmental parameters.
istics (bathymetry, bathymetric complexity, and large-scale Hixon and Tissot (1992) researched Haceta, Coquille, Daisy,
Recreational hook and line drifts covering approximately one
habitat classification). and Stonewall Banks off the Pacific northwest. These results
kilometer, demersal trawls on the continental shelf and slope Gabriel and Tyler (1980) used data from the Oregon Depart-
are very important to managers because they show fish and
covering one kilometer, and fifteen minute trawls in midwater, ment of Fish and Wildlife Trawl Survey and the West Coast Joint
Community metrics can be used to determine which areas are habitat interactions on very small scales. However, many of the
were analyzed to determine species assemblages, site group- Agency Rockfish Survey to look for site groups from California
important to multiple species. Experts in California have ac- results from these studies are not comparable with the current
ings, and the interaction between species and locations. In to Alaska. They differentiated three large site groups: “inter-
knowledged a management need to increase our understanding studies due to large differences in scale. Hixon et al. (1991)
addition, analyses were completed to determine larger scale mediate” at less than 145 meters, “deep” between 145 and
of fish species interactions (objective 2) (Starr, 1998) and the documented that the species composition observed from the
environmental variables that were significantly different among 200 meters, and “slope” greater than 200 meters deep. They
interactions between fish assemblages and habitat (objective submersibles was different than that seen in trawls. The results
identified groups. Due to limitations of the data sets (section found that site groups were “strongly associated with depth
5) (Starr, 1998; Yoklavich et al., 2000, 2002). Studies exist that from these studies reveal the importance of habitat, especially
4), and the lack of results on individual species’ distributions, contours”. Matthews and Richards (1991) compared gill net
identify either species assemblages or site groups (see below), rugosity, to fish species composition.
habitat suitability models (section 2.1.2) for selected species catches from trawlable and untrawlable areas to determine if
but so far none have integrated multiple data sets, provided
were completed to complement this analysis. untrawlable areas could be considered de-facto fish reserves.
the interaction between species assemblages and site groups, Substantial declines in the standing stock biomass of eco-
Even though some species overlapped, they concluded that the
nor presented spatially explicit results. The results of these nomically important rockfish species (Ralston, 1998) prompted
The primary objective of the assemblage analysis is to define species assemblages were significantly different; suggesting
analyses aid in defining the region's biogeography based on NMFS to organize a symposium to discuss the implications of
spatial biogeographic patterns of fishes and macro-inverte- that species assemblages determined from trawls cannot be
the spatial pattern of fishes and macroinvertebrates. no-take areas for rockfish in September of 1997. Eleven plenary
brates within the study area from Point Arena to Point Sal in extrapolated to non-trawlable habitats.
papers and six case studies are available online, and cover a
California. The study is based on a synthesis of four primary
REVIEW OF RELEVANT LITERATURE range of topics. Starr (1998) provided a thorough evaluation
databases of fish and invertebrates that were spatially compre- Only a few studies have analyzed recreational hook and line
Due to the economic importance of recreational and commercial of the potential of rockfish no-take reserves. He expressed
hensive throughout the study area including: 1) the California data. For a general analysis of a species’ specific decline
fisheries in California, several studies have been completed a management need for the identification of species assem-
Department of Fish and Game fishery dependent recreational in recreational catch see Love et al. (1998), Mason (1998),
that look at species co-occurrences or species interactions blages. Once assemblages are identified, management can
fishing trips targeting rockfish (CDF&G recreational); 2) the or Wilson-Vandenberg et al. (1996). Mason (1995) analyzed
with their environment. NMFS publishes yearly reports on the address actions for adequate protection of each species as-
National Marine Fisheries Service fishery independent benthic various CDF&G recreational fishing surveys and documented
status of demersal fish species by analyzing results from their semblage. Starr also suggested protecting rectangular areas
trawls on the continental shelf (NMFS shelf trawls); 3) NMFS trends in effort, fishing location, and species catch. She docu-
shelf and slope trawls (Turk et al., 2001; Weinberg et al., 2002; that cover 20-50 km of the coast and extend west to the edge
fishery independent benthic trawls on the slope (NMFS slope mented two principal rockfish assemblages and distinguished
Lauth, 2001; Shaw et al, 2000). Zimmerman et al. (2001) looked of the continental shelf.
trawls); and 4) the NMFS fishery independent trawls in midwater them by depth (less than 70 meters and greater than 70 me-
at the biomass of demersal species to determine NMFS shelf
(NMFS midwater trawls). Detailed information on these surveys ters). Sullivan (1995) used the CDF&G recreational fishing data
trawls that did not fish the bottom as intended. He then looked
is given in each respective section. Databases that were limited (1987-1992) to determine site groups. His overall conclusion
13
Subsection 2.1.1: ASSEMBLAGE ANALYSES
Within hierarchical clustering, the choice of resemblance met- the reasons behind those decisions are provided within the Only the Ward’s minimum variance (it calculates an internal
INTRODUCTION TO CLUSTERING
ric (a formula that determines how similar two things are) and following detailed methodology section. dissimilarity matrix) and the 1-Pearson correlation with aver-
The use of multivariate analyses is gaining popularity in
clustering model (a method that groups things based on their age linkage consistently produced results without chaining,
ecology and fisheries management (McGarigal et al., 2000;
resemblance metrics) can also have an influence by adjusting DETAILED METHODOLOGY and were therefore interpretable. The Pearson method was
Paukert and Wittig, 2002), but can be confusing due to the
the importance of abundant and rare species and changing To reduce the amount of material that is repeated with each sec- chosen over Ward’s primarily because the results were very
availability of many statistical techniques. Therefore, this sec-
the importance of zero values (Gauch, 1982; Boesch, 1977; tion, a detailed methodology for all four data sets is presented similar to the results from principal component analysis (both
tion provides a basic introduction to the principles of clustering,
McGarigal et al., 2000). Common resemblance metrics used in here, with only a brief overview provided in each section. Figure methods are based on a correlation matrix), but also because
one form of multivariate analysis. Interested readers should
ecology are the Bray-Curtis, Euclidean distance, Jaccard, and 9 is an example of the clustering methods using hypothetical the results showed spatial patterns when mapped.
reference Boesch (1977), Gauch (1982), or McGarigal et al.,
Pearson correlation coefficients. Common clustering models data. The first objective of this study was to look for patterns in
(2000) for more detailed information on clustering than will be
include Ward’s minimum variance, average linkage, centroid community metrics. For these analyses, all unknown species, To determine which fish species tended to be caught together
provided in this brief introduction. Clustering is “a technique
linkage, and single linkage. Because of chaining, not all outputs and those without an abundance estimate, were eliminated. (species assemblages; objective 2), an index of dissimilarity
for optimal grouping of entities according to the resemblance
from cluster analyses can be utilized. When chaining occurs, The Shannon index (H’) was calculated for each NMFS shelf between species was calculated as 1-Pearson correlation coef-
of their attributes as expressed by given criteria” (Boesch,
entities fuse to a few nuclear groups one at a time rather than and slope trawl based on the following equation: ficient, with the resulting matrix of species dissimilarities clus-
1977) or, in short, a method that places things (sites, species,
forming new groups, and make it impossible to divide the data tered by using average means as discussed above. Changes in
etc.) into groups. Clustering uses statistics to determine these
S
n n
into meaningful smaller groups (Boesch, 1977). fishing depth or the abundance of target species through time
groups, but the method also possesses aspects that are sub-
H ′ = − ∑ i ln i
could influence the results. Therefore, a second analysis was
jective and require an understanding of the ecosystem being i =1 n n
Hierarchical clustering results in a tree diagram, called a den- run using only current data (1993+ for recreational, 1989+ for
analyzed. There are five steps to clustering, which all have
drogram, which shows the linkages between all of the entities Where S is the number of species, ni is the number of individu- shelf trawls), so that a comparison of species associations from
some aspect of subjectivity (adapted from Sullivan, 1995): 1)
and groups at different levels of similarity. Various objective als found of species i, and n is the total number of individuals all the data and current data could be completed. The determi-
the choice of data, including what data to include or remove,
methods, such as scree plots, are used to determine what level (Shannon and Weaver, 1949). Richness (total number of fish nation of what should be considered “current” was based on the
and how to transform and standardize that data; 2) the choice
of similarity is important or how many groups are created. A species caught) was also enumerated for each NMFS shelf and expert opinion of the scientists who collected the data (pers.
of a resemblance metric (can be based on similarity or dis-
scree plot shows the dissimilarity values plotted against the slope trawl. In addition, the rockfish richness was calculated by comm. Deb Wilson-Vandenberg and Mark Wilkins) according
similarity); 3) the choice of clustering model; 4) the choice of
number of clusters such that breaks in the level of dissimilarity counting the number of Sebastes and Sebastolobus species to known shifts in effort or species abundance. Differences in
number of groups (or level of similarity) and; 5) the choice of
are revealed through the shape of the curve (McGarigal et al., caught in each trawl. assemblage groups between the entire data set and current
whether or not to reassign objects to more appropriate groups.
2000). Experts recommend combining this objectivity with eco- data are noted.
The following is a more detailed description of each of these
logical knowledge to determine clustering results that are eco- Due to the overlap in trawls between years (boats often returned
five steps.
logically meaningful (Boesch, 1977). The final decision when to approximately the same area each year), it was sometimes To determine which locations contained similar fish catches
clustering is whether to reclassify entities into more appropriate difficult to determine spatial trends in diversity. Therefore, mean (site groups), the 1-Pearson correlation clustering method, as
The choice of data to include in the analyses can influence
groups based on some identified criterion. This step has created diversity and mean richness were calculated for 5’ grid cells explained above, was used again, but this time to cluster sites
results in many ways. Rare species are usually removed from
controversy because of the subjectivity introduced. (approximately 5 km by 9 km cells, although actual dimensions with similar catches. No secondary analyses with a random
analyses because they can have a disproportionate influence
vary with latitude) throughout the study area. The 5’ grid cell subset or current data were run because the influence of such
on the resulting clusters. What is considered “rare” can vary
For comparison, a brief introduction to another widely used form size was utilized because it was employed in other analyses parameters could be inferred by looking at the date of each
for analyses, but typically species found in less than 5% of the
multivariate analysis, the Principal Component Analysis (PCA), (bird, mammal, and environmental maps) and facilitates the trawl in each group.
trawls/catches are removed (Gauch, 1982). Additionally, the
is provided. PCA reduces the dimensionality of data (Gauch, comparison of results among these analyses.
transformation and standardization of the data may affect the
1982, McGarigal et al., 2000). One difference between the In order to decide how many groups to keep, statistical methods
influence of rare and abundant species. Binary data (presence/
results from PCA and hierarchical clustering is that in cluster- Both clustering analyses (objectives 2 and 3) began with either (scree plots) were employed to determine where breaks in the
absence) weights all species the same, and reduces the influ-
ing each species is ultimately assigned to one and only one a site by species or species by site matrix of presence/absence similarity level occurred, then group composition was analyzed
ence of abundant species while increasing the influence of rare
group. In PCA, only species with a certain loading (i.e. level or log adjusted abundance. All species that were present in to determine the best ecological groupings (i.e. if smaller or
species (Boesch, 1977). Transformations are needed because
of influence) are included in a group. This means that in PCA at least 5% of the trawls were included in this analysis. This larger groups would provide a better ecological explanation).
most ecological data do not conform to the assumptions of
some species are never assigned to a component, and others number was chosen because it is a commonly used method in Expert opinion on ecologically relevant groups was solicited
a normal distribution and homogeneity of variances required
are assigned to more than one. This can be a drawback be- fisheries management (Gauch, 1982), and because it reduced at review meetings held in Seattle, San Francisco, and Mon-
for parametric analyses (Boesch, 1977). Transformations
cause often very important species that are found everywhere the number of zero values in the data set while keeping an terey in October, 2001. No reclassifications were completed
decrease the variation between abundant and rare species,
are never attributed to any group. In clustering methods, since adequate number of species for analysis. Since the raw abun- in this study. Instead, a modified bootstrapping procedure was
thereby reducing the influence of abundant species.
every fish must be placed into a group, species which are found dance data did not conform to assumptions of a normal distribu- implemented. Fifty random samples of one-half or three-quar-
everywhere can be grouped together. At the same time, spe- tion and homogeneity of variances, either log transformations ters of the data were extracted and run through the clustering
The clustering method dictates the way that groups are formed.
cies which are only marginally associated with a group can be (if effort was available) or presence/absence (if no effort was process. The amount of data included in the random analyses
The two classes of clustering are: 1) hierarchical, which show
added to a group. provided) were used. No standardization was completed be- depended on the size of the original data set; if the data set
the relationship between groups in dendrograms, and 2) non-
cause correlation coefficients were used. Exploratory analyses was too small, samples consisting of half of the data often had
heirarchical, which do not. Hierarchical clustering may be bi-
For this study, an extensive exploration was conducted on the were run to investigate multiple resemblance metrics (including zero catch for some species, creating error messages when
ased because the researcher can choose the results that best
data using multiple data transformation, similarity metrics, and Euclidean Distance, Jaccard, and Pearson correlation matrices) running the analysis. The results from the random samples were
match expectations (Williams and Ralston, 2002); however,
clustering models. Ultimately, one method was chosen, and and multiple clustering models (including Ward’s minimum vari- used to determine the stability of the species assemblages in
it is advantageous because it clearly shows the relationship
ance, average linkage, centroid linkage, and single linkage). the given data set.
between the resulting groups.
14
Subsection 2.1.1: ASSEMBLAGE ANALYSES
In order to combine these two analyses and resolve where the
Determining species assemblages, site groups, and their interaction
fish assemblages were being caught (objective 4), the average
frequency of occurrence for species assemblages was calcu-
(example with hypothetical data)
lated for each site group to determine the overlap between
the site and species groups. By looking at the frequency of
spiny
occurrence of species assemblages in each site group, it was Site1 Site2 Site3 Site4 Site5
bocaccio chilipepper widow rf Rex sole Dover sole
dogfish
possible to determine which species groups were influential Start with site by species or bocaccio 23 66 0 2 1
Site1 23 11 15 1 0 0 chilipepper 11 5 7 0 0
in forming the site groups. Species groups were considered
species by site matrices of
Site2 66 5 47 0 0 4 widow rf 15 47 0 0 0
influential if, on average, species were present in 25% of the
abundances
Site3 0 7 0 55 43 0 Rex sole 1 0 55 0 23
trawls. Since rare species had low frequency of occurrence for
Site4 2 0 0 0 0 44 Dover sole 0 0 43 0 21
all site groups, 25% is a reasonable number when rare and spiny dogfish 0 4 0 44 0
Site5 1 0 0 23 21 0
abundant species are averaged. Spatial distribution of the site
groups was determined by mapping the site groups in GIS.
bocaccio chilipepper widow rf Rex sole Dover sole spiny dogfish
For management purposes, it is important to understand which Site1 Site2 Site3 Site4 Site5
bocaccio 1.00
environmental characteristics influence species distributions. Site1 1.00
Calculate Pearson
chilipepper 0.30 1.00
Analyses of variance (ANOVA) were conducted to determine Site2 0.92 1.00
widow rf 1.00 0.27 1.00
if there were significant differences in bathymetry, bathymetric correlation coefficients Site3 -0.66 -0.61 1.00
Rex sole -0.52 0.07 -0.53 1.00
complexity, and gross sediment type between the site groups
between species or sites Site4 -0.38 -0.22 -0.35 1.00
Dover sole -0.53 0.02 -0.54 1.00 1.00
at the scale of this analysis. Bottom depth was measured in the
Site5 -0.65 -0.56 0.99 -0.30 1.00
spiny dogfish -0.29 -0.57 -0.25 -0.35 -0.35 1.00
field for each of the four data sets. Using ArcView, individual
trawl locations were overlaid on the sediment (pp. 38) and
bathymetric complexity (pp. 16) maps and the underlying pa-
rameters extracted. For the midwater trawl, other environmental
conditions measured in the field, such as water temperature, Run average means cluster Site 1
spiny dogfish
salinity, and density, were also tested.
analyses on correlations to
widow rf
Site 2
determine species
bocaccio
Bathymetry appeared to be the overriding influence in de- Site 4
assemblages and site
termining fish assemblages. Attempts were made to statisti- chilipepper
Site 3
groups. Determine
cally remove the influence of bathymetry from the data and Dover sole
then re-analyze the data for assemblage patterns caused by appropriate number of
Rex sole Site 5
secondary influences. However, two general problems were
groups.
encountered. First, the standard statistical procedure to re-
move the influence of bathymetry required a linear relationship
between species abundance and bathymetry. Unfortunately,
the relationship between species abundance and bathymetry Calculate frequency of occurrence of species assemblages in site groups.
was non-linear even after various transformations were tried.
Determine which species groups were influential in forming the site groups
Experiments with spline-fitting were also unsuccessful. A major
site 1, 2 site 4 site 3,5
problem was the presence of zero species abundance values
for those depths the species assemblages were not present.
Rockfish 100 33 33
Secondly, the species abundance data were collected over
Sp. Dogfish 50 100 0
narrow ranges of other influences, such as bathymetric com-
plexity and substrate/sediment size. Again, the problems of Sole 25 0 100
non-linearity and zero species abundance prevented further
conventional statistical analyses. Figure 9. Hypothetical example of the methods used to determine species assemblages, site groups, and the interaction between species assemblages and site groups.
15
Subsection 2.1.1: ASSEMBLAGE ANALYSES
ABOUT THIS MAP To determine the importance of bathymetric complexity to the
124°W 123°W 122°W 121°W
Figure 10 displays bathymetric complexity derived from high formation of fish abundance, an analysis of variance was run
39°N
39°N
resolution bathymetry. Bathymetric complexity is calculated for to test for significant differences in bathymetric complexity
Bathymetric Complexity each cell as the standard deviation of depth for all grid cells between site groups for CDF&G recreational hook and line,
within a 1 kilometer radius. The range of resulting bathymetric NMFS shelf trawl, and NMFS slope trawl catches (see individual
complexity is large, and the majority of cells show a low vari- sections for results).
Calculated for a 1 km radius ance. Therefore, in order to visualize differences, results have
to be displayed as standard deviations above and below the
around each grid cell mean. The areas in blue are relatively flat with little slope, and
the darker red shows the highest variance. Results highlight the
edge between the shelf and slope areas, and create a dramatic
visual for the canyons and seamounts.
Departure from Mean
38°N
38°N
DATA SOURCES
-1 - 0 Std. Dev.
Results were calculated from 3 arc second (nominally 70 x
0 - 1 Std. Dev. 70 meters) bathymetry derived from NGDC and MBARI data
sources. All available multibeam points were used in the area.
1 - 2 Std. Dev.
Hydrographic survey data (echo sounder data) was eliminated
2 - 3 Std. Dev. from the interpolation if it overlapped with multibeam data. Verti-
cal and horizontal correction was performed on all data prior to
>3 Std. Dev.
incorporating it into the data set. All data were triangulated and
gridded using "The Vertical Mapper" extension with MapInfo
0 10 20 40 60 80
6.5. Cell size varies depending on the available data for each
37°N
37°N
Kilometers
area, with a minimum cell size of 70 x 70 meters.
METHODS
Bathymetric complexity was calculated using the "neighborhood
statistics" option in Arcview 3.2. Arcview computes a standard
deviation from all grid cells within a 1 kilometer radius around
each cell. The results are displayed as standard deviations
from the mean as this scale provides the best resolution for
visualizing the location of high slope areas.
36°N
36°N
RESULTS AND DISCUSSION
Fish species, especially some rockfish species, have a very
strong affinity to areas with a high relief (Yoklavich et al., 2000,
2002; Hixon et al., 1991; Hixon and Tissot, 1992; Field et al.,
2002; Starr, 1998; and Williams and Ralston, 2002). Calculating
the variability in bathymetry for a given area can provide a rough
estimate of bottom rugosity on a scale of km. Smaller pinnacles
may not be distinguished at this scale, but the large physical
characteristics, such as the edge between the continental shelf
and slope, canyons, and seamounts, will be displayed. The
35°N
35°N
variable depth of the continental shelf break can be estimated
using these maps. North of Cordell Bank NMS, the break occurs
around 300 meters, within Cordell Bank and Gulf of the Faral-
50
50 m
20 lones NMS, it is around 200 meters depth, north of Monterey
100
00
20
0 m
m
0
0
Bay it becomes shallower at 150 meters, and inside Monterey
m
m
Bay and to the south, the break is as shallow as 100 meters.
124°W 123°W 122°W 121°W
Figure 10. Standard deviation of bathymetry calculated for a 1 km radius around each cell. Results are presented in standard
deviations above or below the mean.
16
Subsection 2.1.1: ASSEMBLAGE ANALYSES
in measurements between cells and facilitates comparisons
ABOUT THESE MAPS
124°W 123°W 122°W 121°W
between maps. Since the placement of the grids is arbitrary, the
Species richness of demersal fish was calculated for NMFS
Species Richness of
results will in-part depend on where the grid falls. An analysis
39°N
39°N
fish trawls (shelf and slope) conducted at depths between 50
comparing three different grid placements was conducted. It
and 1280 meters. The mean number of fish species recorded
was determined that the placement of the grids had minimal
for trawls (± standard deviation) was 16±5. Species richness
Demersal Fish
influence on the results.
results are displayed for individual trawls (Figure 11), as well
as mean richness for 5’ grid cells (Figure 12). There appear to
RESULTS AND DISCUSSION
be three trawl areas with consistently high species richness:
Richness by Individual Trawl
The mean (± standard deviation) number of fish species re-
NW of Point Año Nuevo, SW of Morro Bay, and near the center
corded for a demersal trawl was 16±5, and ranged from 1 to
of Cordell Bank NMS. In addition, there are smaller clusters
33. There are large areas with high species richness directly
within all three national marine sanctuaries. This map can be
west and north of Point Año Nuevo, between the 50 and 100 m
used to identify hot spots of demersal fish biodiversity.
contour lines as well as west and south of Morro Bay between
38°N
38°N
Richness 50 and 200 M depth. There are smaller hot spots within Cordell
DATA SOURCES
Bank NMS, between 100 and 200 meters, as well as along the
Richness estimates were derived from 1,336 NMFS (AKFSC
19 - 33
200 meter contour in four locations: north of Cordell Bank NMS,
and NWFSC) shelf (pp. 26) and slope (pp. 28) trawls conducted
15 - 18
just north of the southern Gulf of the Farallones NMS boundary,
between 50-1280 meters depth during June-November every
3 - 14 north of Monterey Bay and in southern Monterey Bay (Figure
third year from 1977-2001. For details on the trawl methods
11). For all trawls, there was a significant negative relationship
see Lauth (2001), Shaw et al. (2000), Turk et al. (2001), and
0 10 20 40 60
between richness and depth, and a significant positive relation-
Williams and Ralston (2002). All fish identified to the species
ship between richness and latitude. However, neither of these
level were included (230 species).
Kilometers
relationships explained much of the variance (r2=0.04, p<0.0001
for depth; and r2=0.005, p<0.004 for latitude, N=1336).
METHODS
37°N
37°N
Richness is defined as the number of fish species present at
Many fish species are associated with near-shore areas, and
a given location. To calculate richness, data were tabulated
were not included in this analysis due to the absence of NMFS
to determine the number of species caught in each trawl. Al-
trawls in shallow water areas. Other analyses in shallow wa-
though there was a significant positive relationship between
ter can provide a comparison to these results. Laidig (Pers
effort (calculated as distance fished x net width) and species
Comm NMFS) has completed underwater scuba surveys to
richness (p<0.0002), this accounted for a very small percentage
determine the presence of fish on kelp beds near Sonoma and
of the variability in the data set (adjusted r2=.01). Therefore,
Monterey. Average richness recorded on 43 dives in Sonoma
raw values of species richness for each trawl were used for this
was 5±3 (range of 1 to 15), and 15±4 on 9 dives in Monterey
analysis. Trawls are only possible along relatively flat bottom
(range of 9 to 21). California Department of Fish and Game
areas with a minor incline, and no data were available for rocky,
36°N
36°N
recreational fishing trips targeting rockfish (pp. 23) can also
highly sloped areas. In addition, the NMFS data did not include
be used to determine approximate fish richness. Without effort
trawls conducted in water less than 50 meters deep, therefore,
information on angler hours, the utility of mapping richness is
shallow water sites are not represented with these results.
questionable. However, the mean richness recorded was 7±4
(range of 1 to 21). The estimate of richness for near-shore
Figure 11 is useful for identifying actual trends in space, as
areas from CDF&G trawls is lower than those measured with
well as identifying where the trawls occurred. Species richness
the NMFS shelf and slope trawl data, but the difference could
results were organized into three equally sized groups repre-
be due to fishing method (hook and line vs. trawl) and is only
senting the lowest, middle, and highest third of richness values.
mentioned as anecdotal validation. There was a large difference
It is also useful to consider the mean richness for a small area.
in the number of species observed in Sonoma and Monterey
Therefore, mean richness and its deviation (how variable it is)
35°N
35°N
by Laidig, providing an example of the variability that can be
was calculated for 5’ grid cells throughout the study area (Figure
experienced in kelp areas. Managers interested in protecting
12). Cells with no deviation contained only one trawl. Cells that
biodiversity of demersal fish could use this information in com-
document high species richness, with low deviation, represent
20
100
50 m
bination with the other assemblage analysis to address various
an area with consistently high species richness. Cell size was
00
20
m management strategies. Cells with high species richness and
determined by minimizing the number of cells containing only
m
0
m
low deviation could be used to identify potentially important
one measurement yet retaining a reasonable spatial resolution
areas which deserve further investigation.
of the cells. This also was the cell size used for integration with
124°W 123°W 122°W 121°W
marine bird results. Species richness in cells is also presented
in three equal sized groups as this best represents differences
Figure 11. Species richness of individual NMFS shelf and slope trawls.
17
Subsection 2.1.1: ASSEMBLAGE ANALYSES
124°W 123°W 122°W 121°W
39°N
39°N
Species Richness of
Demersal Fish
Mean Richness in 5' Cells
38°N
38°N
Richness
18 - 25
15 - 17
8 - 14
Deviation of Richness
No Deviation
Less than Mean Deviation
Greater than Mean Deviation
37°N
37°N
0 10 20 40 60 80 100
Kilometers
36°N
36°N
35°N
35°N
20
100
50 m
00
20
m
m
0m
124°W 123°W 122°W 121°W
Figure 12. Mean species richness of NMFS shelf and slope trawls for 5’ grid cells. The deviation is shown as an overlay to provide an
indication of the variability in results for each grid cell.
18
Subsection 2.1.1: ASSEMBLAGE ANALYSES
ABOUT THESE MAPS divided into 3 equal sized groups representing the lowest,
124°W 123°W 122°W 121°W
Species diversity of demersal fish was calculated for NMFS middle, and highest third of diversity values.
trawls on the shelf and slope at depths between 50 and 1280
Species Diversity of
39°N
39°N
meters. The mean diversity recorded for trawls (± standard It is also useful to consider the mean diversity for a small area.
deviation) was 1.5±0.5. Species diversity results are displayed Therefore, mean diversity and its deviation (how variable it is)
Demersal Fish
for individual trawls (Figure 13), as well as mean diversity for was calculated for all trawls within 5’ grid cells throughout the
5’ grid cells (Figure 14). The largest cluster of high species study area (Figure 14). Cells with no deviation contained only
diversity trawls is found 20 km north and south of the border one trawl. Cells that document high species diversity, with low
between Monterey Bay NMS and Gulf of the Farallones NMS.
Diversity by Individual Trawl
deviation, represent an area with consistently high species
Smaller clusters of high diversity values are present in the diversity. Cell size was determined by minimizing the number
northwest corner of Cordell Bank NMS and to the north and of cells containing only one measurement, yet retaining a rea-
south of the NMS boundaries in waters slightly deeper than sonable spatial resolution of the cells. In addition, this was also
the 200 meter contour line. the cell size used for integration with marine bird analyses (sec-
38°N
38°N
Diversity tion 3). Species diversity in cells is also presented in 3 equally
DATA SOURCES sized groups. Since the placement of the grids is arbitrary, the
1.78 - 2.54
Diversity estimates were derived from 1,336 NMFS (AKFSC results will in-part depend on where the grid falls. An analysis
1.37 - 1.77
and NWFSC) shelf (pp. 26) and slope (pp. 28) trawls conducted comparing three different grid placements was conducted, and
0.02 - 1.36 between 50-1280 meters depth during June-November every it was determined that the placement of the grids had minimal
third year from 1977-2001. For details on the trawl methods influence on the results.
0 10 20 40 60 see Lauth (2001), Shaw et al. (2000), Turk et al. (2001), and
Williams and Ralston (2002). All fish identified to the species RESULTS AND DISCUSSION
Kilometers
level were included (230 species). The mean (± standard deviation) diversity recorded for a de-
mersal trawl was 1.5±0.5, with range from 0.02 to 2.54. The
37°N
37°N
METHODS largest cluster of high species diversity straddles the boundary
Diversity reflects the distribution of species’ abundance within between Monterey Bay NMS and Gulf of the Farallones NMS.
a trawl. For example, a trawl dominated by one species would Fifty-eight (13%) of the high diversity trawls are located within
have a low diversity, while a trawl with an even number of all 20 kilometers of this boundary. The western edge of this area
species would have a high diversity (see Figure 78, pp. 124). contains consistently high diversity trawls (low deviation). A
Since diversity is dependent on the abundance of species, fish smaller cluster of high diversity trawls is present in the north-
caught in a trawl which were not given an estimate of abun- west corner of Cordell Bank NMS, extending approximately 6
dance, or were not identified to species, were eliminated from kilometers north of the current boundary. Within this cluster,
the analysis. The Shannon index (H’) was calculated for each 95% of the trawls are classified as either medium or high
NMFS shelf and slope trawl based on the following equation: diversity. In addition, there are two lines of trawls with high
36°N
36°N
species diversity located slightly deeper than the 200 meter
n n
S
contour line: one north of Cordell Bank NMS to the northern
H ′ = − ∑ i ln i
edge of the study area, and the other from Lopez Point south
i =1 n n
to the southern edge of the study area. A large portion of these
Where S is the number of species, ni is the number of indi- trawls are outside sanctuary boundaries. For all trawls, there
viduals found of species i, and n is the total number of indi- was no significant relationship between diversity and latitude
viduals (Shannon and Weaver, 1949). Although there was a (r2=0.0, p=0.57, N=1336). There was a significant relation-
significant positive relationship between effort (calculated as ship between diversity and depth; however, it did not explain
distance fished x net width) and species diversity (p<0.0001), much of the variance in the data (r2=0.04, p<0.0001, N=1336).
this accounted for a very small percentage of the variability in Many fish species associated with near-shore, or high relief
35°N
35°N
the data set (adjusted r2=.06). Therefore, raw values of species areas, were not included in this analysis due to the absence of
diversity for each trawl were used for this analysis. Trawls are NMFS trawls in these areas. California Department of Fish and
only possible along relatively flat bottom areas with a minor in- Game recreational fishing trips (pp. 23) were often located in
20
100
cline, and no data were available for rocky, highly sloped areas. near-shore or high relief areas, and can be used to determine
50 m
00
20
m In addition, the NMFS data did not include trawls conducted approximate fish diversity over these habitats. The mean fish
m
0
m
in water less than 50 m deep, therefore, shallow water sites diversity recorded for recreational hook and line locations was
are not represented with these results. Figure 13 is useful for 1.3 ± 0.6 (range of 0 to 2.5). This estimate of diversity is similar
124°W 123°W 122°W 121°W
identifying actual trends in space, as well as identifying where to those measured with the NMFS shelf and slope trawl data;
effort occurred. For this figure, species diversity results were however, since the collection method was different, no statisti-
Figure 13. Species diversity of individual NMFS shelf and slope trawls.
19
Subsection 2.1.1: ASSEMBLAGE ANALYSES
cal comparisons can be completed and these results are only
124°W 123°W 122°W 121°W
intended as anecdotal validation.
39°N
39°N
Species Diversity of Trawls with high species diversity are not necessarily trawls with
high richness. Since diversity takes into account the number of
Demersal Fish fish of each species found, areas with one or two abundant spe-
cies have a lower diversity than areas with less fish species, but
an even distribution. High richness trawls are slightly shallower
Mean Diversity in 5' Cells than high diversity trawls suggesting that the trawls deeper than
200 meters have fewer species, but a more even distribution.
The presence of a high diversity area along the boundary be-
tween Gulf of the Farallones and Monterey Bay Sanctuaries
38°N
38°N
defines an area of biological significance for demersal fish. In
Diversity addition, there are lines of high species diversity north and south
1.63 - 2.27 of the current boundaries deeper than the 200 meter contour.
The trawls located on top of Santa Lucia Bank had medium to
1.39 - 1.62
high species diversity, and represent a large expanse of deep
0.18 - 1.38
habitat not within sanctuary boundaries. This area, combined
Deviation of Diversity with existing NMS shelf and slope areas, appears important to
No Deviation groundfish as indicated by high diversity patterns. Managers
interested in protecting biodiversity of demersal fish can use this
Less than Mean Deviation
information in combination with the other assemblage analysis
Greater than Mean Deviation
results to address various management strategies.
37°N
37°N
0 10 20 40 60 80 100
Kilometers
36°N
36°N
35°N
35°N
20
100
50 m
00
20
m
m
0m
124°W 123°W 122°W 121°W
Figure 14. Mean species diversity of NMFS shelf and slope trawls for 5’ grid cells. The deviation is shown as an overlay to provide an
indication of the variability in results for each grid cell.
20
Subsection 2.1.1: ASSEMBLAGE ANALYSES
ABOUT THIS MAP or high relief areas which were not included in this analysis.
124°W 123°W 122°W 121°W
Species richness of demersal rockfishes was calculated from Other analyses conducted in these habitats can provide a com-
Species Richness of
NMFS shelf and slope trawls (Figure 15). The information on parison to these results. Tom Laidig (pers. comm. NMFS) has
39°N
39°N
this map identifies areas with a high number of rockfish spe- conducted scuba surveys to determine the presence of fish on
cies. Values were not influenced by latitude, but were highly kelp beds near Sonoma and Monterey in California. Average
Demersal Rockfish
influenced by depth. The highest rockfish richness values rockfish richness recorded on 43 dives in Sonoma, between
were observed along the edge between the shelf and slope, 1983 and 1995, was 5±2 (range of 0 to 9), and 8±2 on 9 dives
emphasizing the importance of these areas to rockfish. Sanctu- in Monterey (range of 5 to 12). California Department of Fish
Richness by Individual Trawl
ary boundaries now include more than 500 square kilometers and Game recreational fishing trips targeting rockfish (pp. 23)
of this edge area (between 200 and 300 meters depth), with can also be used to determine approximate rockfish richness.
75% of this area within Monterey Bay NMS. The area south However, without fishing effort information, the utility of mapping
of Monterey Bay NMS to the edge of the study area contains the richness from recreational fishing data is questionable. Av-
another 500 square kilometers of habitat between 200 and erage rockfish richness recorded per location/trip combination
38°N
38°N
Richness 300 meters. was 6±2 (range of 0 to 12). The estimate of rockfish richness
for near-shore areas from CDF&G trawls is similar to those
7 - 14
DATA SOURCES measured with the NMFS shelf and slope trawl data, but since
5-6 Data were derived from 1,336 NMFS (AKFSC and NWFSC) the capture method was different, these results should only be
shelf (pp. 26) and slope (pp. 28) trawls conducted between used as an anecdotal validation.
3-4
50-1280 meters depth during June-November from 1977-2001.
2
For details on the trawl methods see Lauth (2001), Shaw et al. The results of this analysis illustrate the importance of the edge
0-1 (2000), Turk et al. (2001), and Williams and Ralston (2002). between shelf and slope areas. This result supports that of Wil-
All rockfish identified to the species level were included (48 liams and Ralston (2002), who found highest rockfish richness
species). between 200 to 250 meters depth using NMFS shelf data for
0 10 20 40 60
37°N
37°N
California and Oregon.
Kilometers
METHODS
Species richness is defined as the number of fish species
present at a given location. To calculate rockfish richness, data
were tabulated to determine the number of rockfish species
8
Sebastes or Sebastolobus present in each trawl. There was no
significant relationship between trawl effort (distance fished x
7
net width) and species richness (N=1336, F=1.3, p=0.26), so
Mean rockfish richness
raw values of species richness for each trawl were used for this
6
analysis. Trawls are only possible along relatively flat bottom
36°N
36°N
areas. No trawls were conducted over rocky, high relief areas or
5
areas in water less than 50 m deep, therefore, some potentially
important sites were not considered in these analyses.
4
RESULTS AND DISCUSSION
3
The mean (± standard deviation) number of rockfish species
recorded for a demersal trawl was 4±3, with a range from 0 to
2
14. The results show that bathymetry has a strong influence
on rockfish species richness. The lowest rockfish richness is
1
found in the shallower (but still >50 m) and deeper waters.
35°N
35°N
0 100200 400 600 800 1000 1200
A band of high rockfish richness is located around 200-300
meters depth and parallels the edge between the continental Depth (10 meter intervals)
shelf and slope. For all trawls, there was a significant non-linear
20
Figure 16. The relationship between depth and rockfish richness
00
100
50 m
relationship between richness and depth (Figure 16, F=166,
m showing mean rockfish richness (for 10 meter depth intervals be-
20
p<0.0001), and no significant relationship between richness
m
0
tween 50-1300 meters). The relationship was fit with a smoothing
m
and latitude (F=2.1, p=0.15). Almost all trawls on the deep slope spline, lambda = 1,000,000.
(deeper than 600 meters) contain the same two rockfish spe-
124°W 123°W 122°W 121°W
cies (shortspine and longspine thornyheads). It is important to
note that many rockfish species are associated with kelp beds
Figure 15. Species richness of rockfish from individual NMFS shelf and slope trawls.
21
Subsection 2.1.1: ASSEMBLAGE ANALYSES
ABOUT THIS MAP A good example of this split by depth can be found south of
124°W 123°W 122°W 121°W
The last three sections showing species diversity, species rich- the Monterey Bay NMS. This suggests that trawls with high
39°N
39°N
ness, and rockfish richness have provided results relevant to species richness found just east of the 200 meter contour are
Integration of Community managing resources. Figure 17 illustrates the overlay of the dominated by a few influential species. Conversely, the areas
top 17-20% of trawls for high species diversity, species rich- of high diversity just west of the 200 meter contour might have
ness, or rockfish richness. The background of the map shows one or two fewer species, but overall the species are evenly
Metrics for Fish the bathymetric complexity from page 16. The overlay of the distributed.
points provides visual representation of the results.
Results from the assemblage analyses were significantly tied to
DATA SOURCES depth; therefore, maps show bands of similar sites along depth
Legend Diversity, richness, and rockfish richness estimates were de- contours and do not delineate areas important to demersal fish.
rived from 1,336 NMFS (AKFSC and NWFSC) shelf (pp. 26) Conversely, the results from the community metrics do delin-
Top 20% of
38°N
38°N
and slope (pp. 28) trawls conducted between 50-1280 meters eate hot spots. Results are limited by collection method since
Diversity and Richness
depth during June-November from 1977-2001. For details on rocky, highly sloped, or shallow (less than 50 meters depth)
Top 20% of Diversity the trawl methods see Lauth (2001), Shaw et al. (2000), Turk areas were not sampled. Managers could use the interaction
et al. (2001), and Williams and Ralston (2002). of the community metrics to decide on proper management
Top 17% of Richness
strategies. For example, management is often tasked with pro-
Top 17% of
METHODS tecting biodiversity, and is therefore interested in delineating
Rockfish Richness
Methods for calculating diversity, richness, and rockfish rich- areas that contain the highest number of species. However, if
Bathymetric Complexity ness are detailed in each section. The top 20% of trawls for an area is high in richness, but is dominated by one economi-
Departure from Mean diversity were extracted and mapped. Ideally, the top 20% cally important species, protecting this area could contribute
of trawls for overall species richness and rockfish richness to resource use conflicts. The interplay between diversity and
-1 - 0 Std. Dev.
would be provided; however, since richness is discrete and not richness should be carefully evaluated.
37°N
37°N
0 - 1 Std. Dev.
continuous, either 17% (21+ species) or 23% (20+ species)
1 - 2 Std. Dev. could be mapped. The trawls which were within the top 20%
2 - 3 Std. Dev. for both richness and diversity are distinguished.
>3 Std. Dev.
RESULTS AND DISCUSSION
0 10 20 40 60 80 100
Richness calculates the number of fish species present in
each trawl, while diversity takes into account the abundance
Kilometers
of fish species as well. Diversity and richness are correlated
(r2=0.06), but trawls with high diversity are not necessarily
trawls with high richness. Trawls which were high in overall
36°N
36°N
richness and diversity were distinguished, and show areas
important to demersal fish. A cluster of trawls with high diver-
sity and high richness straddle the boundary between Gulf of
the Farallones and Monterey Bay NMS, as well as along the
200 meter contour north of Cordell Bank NMS. Small clusters
of high diversity and high richness trawls are present within
each sanctuary. Depth varied between the three community
metrics, with high richness, rockfish richness, and diversity
progressing from shallow to deep. The mean depth for trawls
with the top 17% of rockfish richness was 221±87, with 43%
35°N
35°N
of the trawls between 200 and 300 meters depth. Showing
these trawls reemphasizes the interaction between rockfish
richness and the edge of the continental shelf. The trawls with
50 m
50 m
20 high species richness show much more variability with depth
100
00
20
0
00
m
m (mean depth 212±225 meters), but 64% of them are in water
m
less than 200 meters deep. The trawls with high species diver-
sity were deeper (mean depth 372±289 meters), with 52% of
124°W 123°W 122°W 121°W
them greater than 300 meters depth. Overall, the trawls with
Figure 17. NMFS shelf and slope trawls with the highest species diversity, species richness, and rockfish richness are mapped. The high diversity were deeper than the trawls with high richness.
underlying map illustrates the bathymetric complexity of the study area and can be used to identify the shelf break.
22
Subsection 2.1.1: ASSEMBLAGE ANALYSES
ABOUT THESE ANALYSES the results compared for persistence and precision. Additionally, the data
Managers have recently begun to understand the importance of studying from 1993 to 1998 was analyzed separately to determine if current condi-
CDFG Recreational Hook and Line
entire ecosystems rather than looking at each species individually. This study tions have changed enough to affect the resultant species assemblages.
Species Assemblages
took a first step in clarifying multi-species interactions by determining which Conditions that could have changed through time include: abiotic shifts, such
Gopher Rockfish
species tended to be caught together, and where. Multivariate statistics were as decadal shifts in water temperature; biotic shifts, such as depletion of
Black Rockfish
used to analyze fish species assemblages on the scale of the recreational key species; and effort shifts, such as fishing farther offshore. To determine
Brown Rockfish
Cabezon
fishery over marine habitats off central California. This data set, while fishery which species groups were influential in forming the site groups, the aver-
China Rockfish
dependent, includes demersal, as well as midwater species captured on age frequency of occurrence for species assemblages in each site group
Kelp Greenling
variable habitats, including rock, mud, and sand. Some species and habitats was calculated. Species assemblages were considered influential if, on
Blue Rockfish
in this analysis are not covered with the other data sets in this study, and average, species were present in 25% of the trip/location combinations in
Olive Rockfish
therefore provide complimentary information. Twenty-seven fish species were a site group. The mean depth associated with each site group is provided
Yellowtail Rockfish
Canary Rockfish
grouped into seven species assemblages (Figure 18), and 4,357 trip/location in conjunction with a map showing the fishing locations in 2.5 minute grids,
Copper Rockfish
combinations were grouped into eight site groups (Table 1). Unfortunately, and color coded according to the average depth of the fishing trips within the
Lingcod
due to the nature of the data set (see methods), exact fishing locations could grid cell. Two-way analyses of variance (ANOVA) were conducted with depth
Rosy Rockfish
not be mapped. Therefore, the mean depth associated with each site group (pp. 37 Figure 32) and latitude, sediment (pp. 38 Figure 33), and bathymetric
Starry Rockfish
is provided in conjunction with a map showing the fishing locations in 2.5 complexity (pp. 16 Figure 10) to determine if any of these factors have an
Vermilion Rockfish
Pacific Sanddab
minute grids, which were color coded according to the average depth of the Bocaccio influence on the site group results at the scale of this analysis.
Flag Rockfish
Sh
fishing trips within the grid cell. The two analyses mentioned above provide
all Speckled Rockfish
ow
information on species which were caught together, and locations with similar RESULTS AND DISCUSSION
Widow Rockfish
to
catch. Combining the two results was the challenge. The average frequency of Species Assemblages (Objective A)
D Yelloweye Rockfish
ee
Quillback Rockfish
p
occurrence of species assemblages (percent occurrence calculated for each Seven species assemblages were differentiated from the recreational data,
species and then averaged for each fish assemblage) within each site group and named according to the most influential species (Figure 18). When
Greenspotted Rockfish
was calculated to analyze the interaction between the species assemblages the data from 1993-1998 were analyzed separately, there were two minor
Not Greenstriped Rockfish
Pacific Mackerel
in
at an fluentia
and site groups (Table 2). As with all data sets in this assessment, the most changes: the yellowtail rockfish assemblage split into two assemblages, and
Chilipepper
Squarespot Rockfish
l
y de
significant result was the effect of depth. This supports previous work done squarespot rockfish moved from the Pacific mackerel assemblage to the
pth
by Williams and Ralston (2002), Sullivan (1995), Field et al. (2002), Gabriel bocaccio assemblage. Overall, the species assemblages delineated were
and Tyler (1980), and Matthews and Richards (1991), who found bathymetry surprisingly robust; almost all fish were consistently placed in the same as-
Figure 18. Species assemblage results for the recreational data. Assemblages are named for from FishBase and NMFS
Pictures
the most influ-
to be an important factor in defining fish assemblages. All attempts to isolate semblages for more than 80% of the random runs, providing confidence in
ential species in each group. Assemblages are arranged from shallow to deep, unless they are influential at
and remove the effects of depth in order to determine secondary effects were the stability of the assemblages. Running the modified bootstrap technique
all or none of the depths. The assemblages that were not influential at any depth were composed of relatively
unsuccessful. Certainly, secondary effects exist, but at the scale of this study can provide an estimate of the precision of results, but verifying the accuracy
rare species, making depth associations indiscernible given the methodology for defining “influential” assem-
they were not discernible. Through this analysis, a large amount of informa- blages. Non-italicized species were consistently placed into the same species assemblage >80% of the time; of the results is more difficult. Comparisons of the results with past studies
italicized species tended to roam into other assemblages with random sampling.
tion has been condensed to assemblages of co-occurring species, as well as can give feedback on the accuracy of the results. Assemblages are not
groups of similar locations. A map is provided to visually portray the spatial static, and may modify in response to environmental conditions, such as
arrangement of the results. warm or cold conditions (see CD-ROM for changes in species assemblages
METHODS in response to water temperature or season).
DATA SOURCES The aim of the entire assemblage analysis was to increase our understanding of the biogeography
Data from 2167 commercial passenger fishing vessels, fishing for rockfish or of fishes and macro-invertebrates in relationship to their environment, and identify important areas Love et al. (2002) provides a summary of rockfish habitat requirements and
lingcod, using hook and line, were collected during all months between 1987 or habitats. Four of the five man objectives were addressed in this recreational analysis: species co-occurrences. The gopher rockfish and blue rockfish assemblages
and 1998 at depths between 2-360 meters. Each trip visited between 1 and A. Determine which species tended to be caught together (species assemblages); are supported by Love et al. (2002) as the species in each assemblage are
8 locations, with each trip/location combination considered a unique site. For B. Analyze fishing locations to determine which locations contained similar catches (site described as having the same habitat or co-occurring. In addition, Mason
this data set, effort was not provided, and therefore only presence/absence groups); (1995) looked at the recreational logbook and described a shallow rockfish
was analyzed at each trip/location combination. The data set contained in- C. Resolve where the species assemblages were being caught by combining results from assemblage composed of blue, black, brown, gopher, and olive rockfishes,
formation on 103 fish species, but after removal of rare species, the data objectives A and B and then utilizing GIS to map the results; and all found within the gopher and blue assemblages described in this study.
matrix used for classification contained information on 27 fish species at 4357 D. Identify significant relationships between the site groups identified in objective B and The greenspotted assemblage from this analysis is not necessarily intui-
trip/location combinations. A list of common and scientific names of the spe- broad scale habitat characteristics (bathymetry, bathymetric complexity, and large-scale tive. Greenspotted and greenstriped can both be found on mud near rocks
cies included in the analysis is available on the accompanying CD-ROM. To habitat classification). (Love et al., 2002), but this is also a characteristic of some of bocaccio and
protect individual fishing locations as requested by the CDF&G, results are yellowtail assemblage species (Love et al., 2002). Mason (1995) desig-
presented in 2.5 minute grids. For more information on the data collection Clustering is a technique used to summarize information into similar groups. The 1-Pearson cor- nated a deepwater red rockfish assemblage that included greenspotted,
process see Wilson-Vandenberg et al. (1996). relation coefficients, with the average means clustering method (see "Introduction to Clustering" pp. greenstriped, chilipepper, and bocaccio, which provides some support to
14) was used to summarize fish species into assemblages and catch locations into site groups. In the greenspotted assemblage of this study. Flag rockfish is an example of a
order to determine how variable the species assemblage results could be within the data, a modified species co-occurrence mentioned in Love et al. (2002) that is not supported
bootstrapping procedure was employed on 50 random samples composed of 50% of the data and here. Flag rockfish was placed in the bocaccio assemblage in this study, but
23
Subsection 2.1.1: ASSEMBLAGE ANALYSES
due to overlap of more than one group within the same cell. For
according to Love et al. (2002), flag is often found with species from the yellowtail
Group 26 Group 40 Group 44 Group 59 Group 64 Group 77 Group 98 Group 125
example, within one grid cell on the southern side of Monterey Bay,
rockfish assemblage. Within the modified bootstrapping procedure, flag rockfish was
meters meters meters meters meters meters meters meters
the maximum depth fished ranged between 37 and 660 meters,
placed with the Bocaccio assemblage 78% of the time, and with the yellowtail rockfish
Gopher
and contained sites from all 8 cluster groups. Therefore, the mean
assemblage only 28% of the time, supporting its placement in this analysis. 0.23 0.14 0.09 0.09 0.01 0.01 0.00
0.36
Assemblage
depths fished ±SD are presented, which can be used in conjunction
with Figure 19 to determine the approximate location of the site
The results comply with the large scale assemblages designated by NMFS: near-
Blue Assemblage 0.07 0.19 0.20 0.07 0.00
0.72 0.74 0.69
groups. Depth was the primary determinant of site groupings. All but
shore, shelf, and slope species groups (NMFS). All of the rockfish in each species
two (groups 40 and 44) of the eight site clusters were significantly
assemblage from this study come from the same NMFS group, except for the yellow-
Yellowtail
different in depth (see Table 1), suggesting that depth is highly in-
tail assemblage, which contains four species designated as “shelf” and one species 0.22 0.08 0.08
0.42 0.31 0.74 0.31 0.57
Assemblage
fluential in determining species distributions within the study area.
designated as “near-shore”. Williams and Ralston (2002) grouped rockfish from the
Bocaccio
The site groups we identified were similar to results of Sullivan
NMFS shelf trawl data into eight groups. While their assemblages differ from this
0.01 0.05 0.00 0.05 0.23 0.22
0.25 0.43
Assemblage
(1995), who analyzed a subset of this same data to differentiate
study's results, of the eleven species analyzed in both data sets, species from the
areas based on species composition. A direct comparison between
bocaccio and greenspotted rockfish assemblages are placed together, and species Greenspotted
0.00 0.00 0.02 0.02 0.07 0.10 0.50 0.59
Sullivan and this study is difficult because Sullivan describes his
from the gopher rockfish and yellowtail rockfish assemblages are placed together. Assemblage
locations verbally using land identifiers, while this study describes
Comparison of the results from this study with results based on trawl (NMFS; Williams
Pacific Mackerel
locations by depth. The importance of depth in this ecosystem is
and Ralston, 2002; Gabriel and Tyler, 1980; Jay, 1996), or results from submersibles 0.02 0.07 0.13 0.07 0.12 0.17 0.09 0.06
Assemblage
not a new idea; many researchers have already commented on
(Yoklavich et al., 2000, 2002; Hixon et al., 1991; Hixon and Tissot, 1992; Field et al.,
its influence (Williams and Ralston, 2002; Sullivan, 1995; Gabriel
2002), is difficult due to the species analyzed, the different habitats targeted, and the
Quillback 0.01 0.06 0.05 0.04 0.09 0.01 0.02 0.00
and Tyler, 1980; Field et al., 2002; Matthews and Richards, 1991).
variable scale of the results. Matthews and Richards (1991) found different species
Latitude has also been described as having an influence on Cali-
assemblages over trawlable and untrawlable habitats, showing the effect targeting Table 2. Average frequency of occurrence of fish species assemblages (percent occurrence calculated for each
fornia fish species composition (Williams and Ralston, 2002; Horn
different environments can have on species assemblages. Scale is important since species and then averaged for each fish assemblage) for each recreational site group. Numbers in bold represent
and Allen, 1978; Sullivan, 1995), but for the area of this study, no
the recreational boat drifts over multiple habitats during a set, and fish from multiple influential species assemblages within that site group.
latitudinal results were evident.
habitats can be present in one trip/location combination. In addition, species as-
semblage results could also be confounded by ontogenetic habitat shifts because
Interaction of Species and Sites (Objective C)
the sizes of the fish captured were not considered.
In conclusion, this analysis provides results showing species assemblages, site as-
The interaction between site groups and species assemblages (i.e. the location of
semblages, and the location of species assemblages for the important near-shore,
species assemblages) reemphasized the relationship between species and depth.
Site Groups (Objective B)
rocky environment. Understanding species assemblages and mapping their loca-
Species assemblages which were influential in forming each site group are identified
Eight site groups were identified from the 4,357 trip/location combinations (Table 1).
tion provides important information for managers. For example, to include the most
(Table 2). Site group 44 did not seem to be associated with any fish assemblages
To make interpretation easier, the site groups are named according to mean depth.
species assemblages, protecting an area that covers a large variation in depth may
(none with a frequency of occurrence greater than 25). At this point, it is uncertain
Maps with the location of the site groupings in the 2.5 minute grid are hard to interpret
be more important than protecting an area that covers a large variation in latitude.
what factor caused the clustering of this group. For all of the trip/location combina-
The results of this analysis in conjunction with similar analyses on the three other
tions, on average 68 fish were caught. The average number of fish caught for group
data sets provides a fairly comprehensive overview of fish and macro-invertebrate
44 was 12, suggesting that some outside factor, such as poor weather, was influenc-
species within the study area.
ing catches at these sights.
Site Group Depth±SD
(Names based on depth) (meters) Habitat Correlations (Objective D)
N
Other factors besides depth can have an impact on species assemblages. Examples
26 ± 13a
Group 26 meters 581 include latitude (Horn and Allen, 1978; Sullivan, 1995), sediment type (Yoklavich et
40 ± 16b al., 2000, 2002; Field et al., 2002; Hixon et al., 1991; Hixon and Tissot, 1992), and
Group 40 meters 688
substrate relief (see bathymetric complexity section pp. 16) (Yoklavich et al., 2000,
44 ± 27b
Group 44 meters 183 2002; Field et al., 2002). Unfortunately, since there were significant interactions
59 ± 26c present between all of these variables and depth, it could not be determined if the
Group 59 meters 235
significance detected for these factors was due to this interaction with depth. Even
64 ± 18d
Group 64 meters 1,501 though bathymetric complexity increases as depth increases, groups 44, 59, and
77 ± 22e 125 meters have a higher bathymetric complexity than the groups around them with
Group 77 meters 207
similar depth. All attempts to remove depth and determine secondary influences on
98 ± 21f
Group 98 meters 683 group designation were unsuccessful. The non-linear relationship between the fish
125 ± 32g
Group 125 meters 279 species and depth made removal of depth impossible. While these other factors
appear to have a decreased significance when compared to depth, more complex
analyses exploring ways to remove the effects of depth and determine the relative
Table 1. Site group results for recreational data. The numbers of trip/location combinations
significance of these factors may be completed in Phase II.
associated with each group as well as average depth, ± standard deviation, are provided.
Different letters signify a significant difference using Tukey’s pairwise comparison on log
adjusted depth with overall alpha set at 0.001.
24
Subsection 2.1.1: ASSEMBLAGE ANALYSES
124°W 123°W 122°W 121°W
CDF&G Recreational Data
39°N
39°N
in 2.5' Grids
Mean Depth
0 - 15 meters
15 - 25 meters
38°N
38°N
25 - 35 meters
35 - 45 meters
45 - 55 meters
55 - 65 meters
65 - 75 meters
75 - 85 meters
85 - 100 meters
37°N
37°N
>100 meters
0 10 20 40 60 80
Kilometers
36°N
36°N
35°N
35°N
100
50 m
20
00
20
m
m
0
m
124°W 123°W 122°W 121°W
Figure 19. Location of CDF&G recreational fishing data in 2.5 minute grids which are color coded according to the average depth of the
fishing trips within the grid cell. Lines showing the 50, 100, 200, and 2,000 depth contours are provided.
25
Subsection 2.1.1: ASSEMBLAGE ANALYSES
ABOUT THIS MAP were mapped using GIS. To determine which species groups were influential
Not
Recently managers and scientists have begun to understand the importance in forming the site groups, the average frequency of occurrence for species
in
at a fluentia
NMFS Shelf Trawls
of studying communities of species rather than just managing by individual assemblages in each site group was calculated. Species assemblages were
ny d
epth l
Sharpchin Rockfish
Pacific Herring
Species Assemblages
species. This study was an initial assessment aimed at determining which considered influential if, on average, species were present in 25% of the trawls
Spot Shrimp
American Shad
Threadfin Sculpin
California Market Squid
species tended to be caught together, and where. Multivariate statistics in a site group. Two-way analyses of variance (ANOVA) were conducted with
Chinook Salmon
were used to analyze species assemblages over trawlable habitats of the depth (pp. 37) and latitude, sediment (pp. 38), and bathymetric complexity
Curlfin Sole
Canary Rockfish
Dungeness Crab
continental shelf between 55 and 500 meters depth off California. For an (pp. 16) to determine if any of these factors have an influence on the site
Longspine Combfish Yellowtail Rockfish
Northern Anchovy Widow Rockfish
introduction to the continental shelf ecosystem, see the Ecological Linkages group results at the scale of this analysis.
Pacific Pompano Rock Sole Sp
Report. Sixty-one species were grouped into thirteen species assemblages Halfbanded Rockfish
White Croaker
Pacific Electric Ray Pacific Mackerel
RESULTS AND DISCUSSION
Arrowtooth Flounder
(Figure 20), and 883 trawls grouped into eight site groups (Table 3, Figure Jack Mackerel
Species Assemblages (Objective A)
Pacific Argentine
21). The average frequency of occurrence of species assemblages (percent Pacific Sanddab
English Sole
Thirteen species assemblages were determined in the NMFS trawl data
occurrence calculated for each species and then averaged for each fish as- Petrale Sole
set and named according to the most influential species (Figure 20). There
semblage) within each site group was calculated to analyze the interaction Pink Seaperch
Plainfin Midshipman
were no differences in the results when data from 1989-2001 were analyzed
between the species assemblages and site groups (Table 4). As with all data Lingcod
Big Skate
separately. Overall, the species assemblages delineated were robust; eight
sets, the most significant result was the effect of depth on species assem- California Skate
of the thirteen groups were consistently placed together for more than 80%
Spiny Dogfish
blages. All attempts to isolate and remove the effects of depth in order to Chilipepper
Bocaccio
of the random samples. This provides confidence that the results do not rep-
determine secondary effects were unsuccessful. Certainly secondary effects Cowcod
Rex Sole
resent just random groupings. Running the modified bootstrap technique can
Sh
exist, but at the scale of this study, they were not discernible. Our results Greenspotted Rockfish
Slender Sole
all Greenstriped Rockfish
Spotted Cusk-eel
provide an estimate of the precision of results, but verifying the accuracy of
ow
support previous results by Williams and Ralston (2002), Sullivan (1995), Shortbelly Rockfish
Pacific Hake
to Stripetail Rockfish
the results is more difficult. Comparisons of the results with past studies can
Field et al. (2002), Gabriel and Tyler (1980), and Matthews and Richards D
Shortspine
Longnose Skate
ee Darkblotched Rockfish
give feedback on the accuracy. Assemblages are not static and may modify
Spotted Ratfish Thornyhead
(1991), who found bathymetry to be an important factor in defining west p Bank Rockfish
Bering Skate
in response to environmental parameters, such as warm or cold conditions
coast demersal fish assemblages. Through this analysis, a large amount of Redbanded Rockfish
Bigfin Eelpout
Blackgill Rockfish
Splitnose Rockfish
Dover Sole
(see CD-ROM for changes in species assemblages in response to water
information has been condensed to assemblages of co-occurring species, Aurora Rockfish
Sablefish
Black Eelpout
temperature).
as well as groups of similar locations. A map is provided to visually portray Blacktail Snailfish
Al l d
epth
the spatial arrangement of the results. Brown Cat Shark
s Filetail Cat Shark
The species cluster results from the shelf trawls make intuitive sense in many
Lanternfish
ways. For example, most of the pelagic species were clustered together
DATA SOURCES Pictures from FishBase and NMFS
(Pacific herring assemblage), and the soft bottom and hard bottom species
Data from 883 fisheries independent research trawls (55-500 meters depth) Figure 20. Species assemblage results for the shelf trawls. Assemblages are named for the most influential
are separated for most of the groups. Only a few assemblages contain both
were collected every third year, between 1977 and 2001, during the months species in each group. Assemblages are arranged from shallow to deep, unless they are influential at all or
soft bottom and hard bottom species, for example the inclusion of the soft
of June-August. Gear included a nor’eastern trawl (127 mm stretched-mesh none of the depths. The assemblages that were not influential at any depth were composed of relatively rare
bottom- associated stripetail rockfish and rock sole with hard bottom assem-
body; 89 mm stretched-mesh codend; and 32 mm stretched-mesh codend species, making depth associations indiscernible given the methodology for defining “influential” assemblages.
blages (bocaccio and canary rockfish, respectively), and the placement of
liner) with a rubber bobbin roller which was trawled for 15-30 minutes on Non-italicized species were consistently placed into the same species assemblage >80% of the time; italicized
stripetail rockfish (hard bottom) in the darkblotched rockfish assemblage (soft
species tended to roam into other assemblages with random sampling.
the bottom. Data was adjusted for effort and to meet statistical assumptions
bottom) (Love et al., 2002). The chilipepper group contains fish species that
by dividing number of fish caught by the area covered and then log trans-
A. Determine which species tended to be caught together (species assemblages); are benthic as well as midwater schoolers, suggesting that even though these
forming. The data set contained information on 167 fish species, but after
B. Analyze fishing locations to determine which locations contained similar catches (site species behave differently they are responding to similar habitat characteris-
removal of rare species, the data matrix used for clustering contained only
groups);
58 fish and 3 invertebrate species. A list of common and scientific names
C. Resolve where the species assemblages were being caught by combining results from
of the species included in the analysis is available on the accompanying
Site Group Depth±SD
objectives A and B and then utilizing GIS to map the results; and
CD-ROM. Since each NMFS cruise hosted scientists with varying levels of
(Names based on depth) (meters)
N
D. Identify significant relationships between site groups identified in objective B and broad
expertise in invertebrate identification, NMFS scientists recommended that
± 16a
125 78
Group 78 meters
scale habitat characteristics (bathymetry, bathymetric complexity, and large-scale habitat
only well known/common invertebrate species be included in the analyses.
± 19b
103
classification). 93
Group 93 meters
Fish species assemblages were identical with and without the inclusion of
± 25b
invertebrates in the analysis. For more information on how the data were 136 96
Group 96 meters
Clustering is a technique used to summarize information into similar groups. The 1-Pearson correla-
collected, including the site selection process and how it changed through ± 37c
72 119
Group 119 meters
tion coefficients with the average means clustering method (see "Introduction to Clustering" pp. 14)
time, see Shaw et al. (2000), Wilkins et al. (1998), and Zimmermann et al.
± 41d
171 153
Group 153 meters
was used to first summarize fish species into assemblages, and to then summarize catch locations
(2001).
± 52e
116 268
Group 268 meters
into site groupings. In order to determine how variable the species cluster results could be within
± 51f
37 328
Group 328 meters
the data, a modified bootstrapping procedure was employed on 50 random samples composed of
METHODS
± 48g
415
123
50% of the data and the results compared for persistence and precision. Additionally, the data from Group 415 meters
The aim of the entire assemblage analysis was to increase our understand-
1989 to 2001 were analyzed separately to determine if current conditions have changed enough
ing of the biogeography of fishes and macro-invertebrates in relationship to Table 3. Site group results for shelf trawl data. The numbers of trawls associated
to affect the resultant species assemblages. Conditions that could have changed through time
their environment, and identify important areas or habitats. Four of the five with each group as well as average depth ± standard deviation are provided. Differ-
include: abiotic shifts, such as decadal shift in water temperature; biotic shifts, such as depletion
main objectives were addressed in this analysis: ent letters signify a significant difference using Tukey’s pairwise comparison on log
of key species; or effort shifts, such as fishing farther offshore. The location of the eight site groups adjusted depth with overall alpha set at 0.001.
26
Subsection 2.1.1: ASSEMBLAGE ANALYSES
tics. Love et al. (2002) mention species co-occurrences, such as the association of mented on its influence (Williams and Ralston, 2002; Sullivan, 123°W 122°W 121°W
cowcod (immature) with bocaccio and greenstripe rockfish, which are corroborated 1995; Gabriel and Tyler, 1980; Field et al., 2002; Matthews and
39°N
39°N
by the results of this analysis. However, this study, unlike Love, found that stripetail Richards, 1991). Species assemblages which were influential in
NMFS Shelf Trawls
rockfish and a splitnose rockfish did not occur together. The placement of yellowtail forming each site group are identified (Table 4). The interaction
rockfish, canary rockfish, and widow rockfish together in this analysis is supported between site groups and species assemblages (i.e. the location
by multiple studies (Tagart and Wallace, Leet, 2001, Star et al., 2002). of species assemblages) reemphasized the relationship between
species and depth. Three species groups (rex sole, Pacific hake,
All rockfish groups concurred with the broad characterization by NMFS (into near- and shortspine thornyhead assemblages) had a high frequency
Site Groups
shore, shelf, and slope species groups), which were based on an assemblage analysis of occurrence in all trawl groups. The rest of the species groups
completed by Gabriel and Tyler (1980). Williams and Ralston (2002) used the same were arranged from shallow to deep. In all cases, the assem- Group 78 Meters
data set as this study to examine rockfish species assemblages. Their “southern blages with a low frequency of occurrence for all site groups were Group 93 Meters
shelf group” contained species clustered together in this report's chilipepper and composed of relatively rare species. Depth associations were
Group 96 Meters
canary groups, while their “deep-water slope” group was split among three groups
38°N
38°N
present, just not discernible given the methodology for defining
in this report. The results from submersibles (Yoklavich et al., 2000, 2002; Hixon et “influential” assemblages. The location of the trawls designated Group 119 Meters
al., 1991; Hixon and Tissot, 1992; Field et al., 2002) provide relevant species/habitat to each site group were mapped using GIS (Figure 21). Group 153 Meters
interactions at a scale meaningful to fish, however, many of the results from these
Group 268 Meters
studies are not comparable with the current studies due to the large difference in Habitat Correlations (Objective D)
scale. Hixon et al. (1991) documented that the species composition observed from Other factors besides depth can have an impact on species Group 328 Meters
the submersibles was different than species captured in trawls. assemblages. Examples of these factors include latitude (Horn Group 415 Meters
and Allen, 1978; Sullivan, 1995), sediment type (Yoklavich et
Site Groups and Interaction of Species and Sites (Objectives B and C) al., 2000, 2002; Field et al., 2002; Hixon et al., 1991; Hixon and 0 10 20 40 60 80 100
Eight site groups were identified from the 883 shelf trawls (Table 3). To make inter- Tissot, 1992), and bathymetric relief (see bathymetric complexity
Kilometers
pretation easier, the site groups are named according to mean depth. All but two pp. 16) (Yoklavich et al., 2000, 2002; Field et al., 2002). Unfortu-
37°N
37°N
groups are significantly different in depth using an ANOVA (Table 3). The importance nately, since there were significant interactions present between
of depth in this ecosystem is not a new idea; many researchers have already com- all of these variables and depth, it could not be determined if the
significance detected for these factors was due to this interaction
with depth. Even though bathymetric complexity increases as
Group 78 Group 93 Group 96 Group 119 Group 153 Group 268 Group 328 Group 415
meters meters meters meters meters meters meters meters
depth increases, group 268 has a higher bathymetric complexity
than the groups around them with similar depth. It is interesting
0.64 0.59 0.82 0.74 0.80 0.78 0.93 0.63
Rex Sole Assemblage
to note that for this data set, 89% of the trawls occurred over
Pacific Hake
0.41 0.37 0.62 0.39 0.58 0.61 0.87 0.68
areas delineated as mud, and 8% over areas delineated as sand
Assemblage
(pp. 38) making it impossible to examine the effects of sediment.
Shortspine Thornyhead
0.28 0.32 0.39 0.64 0.96 0.83
0.20 0.23
Assemblage
However, all 15 trawls that occurred over habitat designated as
36°N
36°N
Pacific Herring
“mud-rock mix” were clustered together into the "415 meters
0.63 0.31 0.34 0.21 0.06 0.01 0.01 0.01
Assemblage
group". All attempts to remove depth and determine secondary
0.27
Halfbanded Assemblage 0.22 0.11 0.24 0.07 0.02 0.01 0.01 influences on group designation were unsuccessful. The non-
linear relationship between the fish species and depth made
Pacific Sanddab
0.90 0.91 0.88 0.82 0.55 0.13 0.10 0.03
Assemblage removal of depth impossible. While these other factors appear
to have a decreased significance when compared to depth, more
0.27 0.26 0.40 0.25 0.25
Big Skate Assemblage 0.16 0.24 0.12
complex analyses exploring ways to remove the effects of depth
and determine the relative significance of these factors may be
0.57 0.61 0.42
Chilipepper Assemblage 0.12 0.26 0.22 0.12 0.03
completed in Phase II.
Darkblotched
0.41 0.41
0.01 0.00 0.03 0.03 0.10 0.17
Assemblage
35°N
35°N
In conclusion, this analysis provides results showing species
Blackgill Rockfish
0.30 0.56
0.01 0.01 0.00 0.02 0.01 0.05
Assemblage
assemblages, site assemblages, and the location of species
assemblages for the important shelf environment. The species
Canary Assemblage 0.02 0.10 0.06 0.21 0.15 0.04 0.01 0.00
20
50 m
100
00
assemblages are relevant to the scale of a commercial trawl. The
20
m
0
Sharpchin Assemblage 0.01 0.01 0.02 0.05 0.13 0.23 0.10 0.01
larger species assemblages that were reported in the literature
m
m
were confirmed (NMFS near-shore, shelf and slope groups). For
Arrowtooth Flounder
0.00 0.05 0.01 0.01 0.11 0.14 0.14 0.02
Assemblage the most part, pelagic, soft bottom, and hard bottom species as- 123°W 122°W 121°W
Table 4. Average frequency of occurrence of fish species assemblages (percent occurrence semblages were distinguished, providing initial feedback to the
calculated for each species and then averaged for each fish assemblage) for each shelf site Figure 21. Location of site groups for NMFS shelf trawls. Lines showing the 50, 100, 200, and 2,000 depth
accuracy of the species assemblages.
group. Numbers in bold represent influential species assemblages within that site group. contours are provided.
27
Subsection 2.1.1: ASSEMBLAGE ANALYSES
rence for species assemblages in each site group was calculated. Species
ABOUT THIS MAP Not
in
NMFS Slope Trawls at a fluentia assemblages were considered influential if, on average, species were present
Little is known about the deep slope species, especially information on which
ny d
epth l in 25% of the trawls in a site group. Two-way analyses of variance (ANOVA)
Species Assemblages
species are found together on what habitats. Multivariate statistics were used
were conducted with depth and latitude, sediment, and bathymetric complexity,
to analyze species assemblages over trawlable habitats between 190 and Stripetail Rockfish California Market Squid
to determine if any of these factors have an influence on the site group results
1280 meters depth off California. This is the first attempt to exclusively define Bocaccio Blackbelly Eelpout
Chilipepper at the scale of this analysis.
species assemblages on the deep slope community. For an introduction to Robust Clubhook Squid
Darkblotched Rockfish
the continental slope ecosystem, see the Ecological Linkages Report. Eight English Sole
RESULTS AND DISCUSSION
Greenstriped Rockfish
species assemblages (Figure 22) and seven site groups (Table 5, Figure 23) Splitnose Rockfish
Lingcod
Species Assemblages (Objective A)
were identified. The average frequency of occurrence of species assemblages Bering Skate
Petrale Sole
Bigfin Eelpout Eight species assemblages were determined for the NMFS slope trawls, and
(percent occurrence calculated for each species and then averaged for each Sharpchin Rockfish
Longnose Skate
Shortbelly Rockfish named according to the most influential species (Figure 22). Overall, the spe-
fish assemblage) within each site group was calculated to analyze the inter- Pacific Hake
Spot Shrimp
cies assemblages delineated were robust; seven of eight assemblages were
Rex Sole
action between the species assemblages and site groups (Table 6). As with Slender Sole
Spotted Ratfish Filetail Catshark
consistently placed together for more than 80% of the random samples. This
Redbanded Rockfish
all data sets, the most significant result was the effect of depth on species Black Eelpout
Rosethorn Rockfish
provides confidence that the results do not represent just random groupings.
assemblages. All attempts to isolate and remove the effects of depth in order California Grenadier
Pacific Electric Ray
Flapjack Devilfish Running the modified bootstrap technique can provide an estimate of the preci-
to determine secondary effects were unsuccessful. Certainly secondary ef-
Pacific Glass Shrimp Longspine Thornyhead sion of results, but verifying the accuracy of the results is more difficult.
fects exist, but at the scale of this study, they were not discernible. Our results
Black Skate
support previous results by Williams and Ralston (2002), Sullivan (1995), Califonia Slickhead
Sh Pacific Viperfish
The species cluster results from these NMFS slope trawls seem much less
all
Field et al. (2002), Gabriel and Tyler (1980), and Matthews and Richards Crimson Pasiphaeid Black Hagfish
ow Deepsea Sole Deepsea Skate intuitive than those from the NMFS shelf trawls. This is partly due to a lack of
Aurora Rockfish
(1991), who found bathymetry to be an important factor in defining demersal to Giant Grenadier Fangtooth
Blackgill Rockfish
D research and subsequent decreased understanding of the behavior of slope
fish assemblages on the West Coast. Through this analysis a large amount ee Grooved Tanner Crab Longfin Dragonfish
Dover Sole
p species. Many of the species from the NMFS slope trawls understandably
Pacific Flatnose
of information has been condensed down to assemblages of co-occurring Pacific Blackdragon
Spiny Dogfish
Pacific Grenadier Rhomboid Squid
Bank Rockfish overlap with species from the NMFS shelf trawls. Surprisingly, some of the
species, as well as groups of similar locations. A map is provided to visually Snakehead Eelpout Sawtooth Eel
species interactions noted with the shelf trawls are not upheld with the slope
portray the spatial arrangement of the results. Twoline Eelpout Smooth Grenadier
All Sablefish Threadfin Slickhead data. The stripetail rockfish slope assemblage is composed of all of the shal-
d ep Blacktail Snailfish Vampire Squid
lower species (soft and hard bottom) that were distributed among 6 shelf as-
DATA SOURCES th Brown Catshark Magistrate Armhook Squid
s semblages. This does not imply that species co-occurrences changed between
Shortspine Thornyhead
Data from 454 fisheries independent research trawls between depths of 190-
Pictures from FishBase and NMFS
the shelf and slope trawls, just that cluster results were sensitive to the depth
1280 meters were collected in 1991, 1997, 1999, 2000, and 2001, during the
range covered by the data set.
months of July-November. For 1999, 2000, and 2001 (NWFSC), gear included Figure 22. Species assemblage results for the slope trawls. Assemblages are named for the most influential
species in each group. Assemblages are arranged from shallow to deep, unless they are influential at all
an aberdeen net with a small mesh liner (2 inches stretched) at the codend
or none of the depths. The assemblages that were not influential at any depth were composed of relatively Love et al. (2002) provides a summary of rockfish habitat requirements and
which was trawled along the bottom along east-west transects for 15 min-
rare species, making depth associations indiscernible given the methodology for defining “influential” as- species co-occurrences. However, since only 15 (24%) species are rockfish,
utes. For 1991, 1997, 1999, and 2000 (AKFSC), gear included a nor’eastern
semblages. Non-italicized species were consistently placed into the same species assemblage >80% of this information cannot be used to assess all results. None of the assemblages
(127 mm stretched-mesh body, 89 mm stretched-mesh codend, and 32 mm
the time; italicized species tended to roam into other assemblages with random sampling.
in this study completely agree with species co-occurrences listed in Love et
stretched-mesh codend liner) with a rubber bobbin roller which was trawled
al. (2002). For example, Love et al. stated that stripetail rockfish and splitnose
on the bottom for 15-30 minutes. Although different gears were utilized in this
rockfish are found together. However, within both the shelf and slope data sets
data set, preliminary analyses found no significant difference between years, objectives were addressed in this analysis:
of this analysis the stripetail and splitnose rockfish were placed into different
allowing the data sets to be combined (pers com Tonya Builder, NMFS). Data A. Determine which species tended to be caught together (species assemblages);
groups. The results are consistent with the large scale assemblages desig-
was adjusted for effort and statistical assumptions as in the NMFS Shelf Trawls. B. Analyze fishing locations to determine which locations contained similar catches
The data set contained information on 161 fish species, but after removal (site groups);
of rare species, the matrix used for classification contained 52 fish and 10 C. Resolve where the species assemblages were being caught by combining results
Site Group Depth±SD
invertebrate species. A list of common and scientific names of species in- from objectives A and B and then utilizing GIS to map the results; and
(Names based on depth) (meters)
N
cluded in this analysis is available on the accompanying CD-ROM. Since each D. Identify significant relationships between site groups identified in objective B and
a
NMFS cruise hosted scientists with varying levels of expertise in invertebrate Group 263 meters 84
broad scale habitat characteristics (bathymetry, bathymetric complexity, and large- 263 ± 49
b
identification, NMFS scientists recommended that only well known/common scale habitat classification). Group 410 meters 86 410 ± 46
invertebrate species be included in the analyses. Fish species assemblages c
Group 530 meters 43 530 ± 42
were identical with and without the inclusion of invertebrates in the analysis. Clustering is a technique used to summarize information into similar groups. The 1-Pearson d
Group 622 meters 29 622 ± 27
For more information on how the data were collected, including site selection correlation coefficients with the average means clustering method (see introduction to cluster- e
Group 733 meters 48 733 ± 71
procedures for each data set, see Turk et al. (2001) and Lauth (2001). ing pp. 14) was used to first summarize the fish species into assemblages, and to then sum- f
Group 931 meters 90 931 ± 132
marize the catch locations into site groupings. In order to determine how variable the species
g
Group 1112 meters 74
METHODS cluster results could be within the data, a modified bootstrapping procedure was employed on 1112 ± 95
The aim of the entire assemblage analysis was to increase our understanding random samples composed of 50% of the data and the results compared for persistence and Table 5. Site group results for slope trawl data. The numbers of trawls associated with
of the biogeography of fishes and macro-invertebrates in relationship to their precision. The location of the seven site groups were mapped using GIS. To determine which each group as well as average depth ± standard deviation are provided. Different let-
environment, and identify important areas or habitats. Four of the five main ters signify a significant difference using Tukey’s pairwise comparison on log adjusted
species groups were influential in forming the site groups, the average frequency of occur-
depth with overall alpha set at 0.001.
28
Subsection 2.1.1: ASSEMBLAGE ANALYSES
nated by NMFS: near-shore, shelf, and slope species groups 1978; Sullivan, 1995), but for the area of this study, no latitudinal
124°W 123°W 122°W 121°W
(NMFS, 2002). All of the rockfish in each species assemblage results were evident.
from this study come from the same NMFS group. The rock-
39°N
39°N
fish groups identified by Williams and Ralston (2002) were not The interaction between site groups and species assemblages (i.e.
NMFS Slope Trawls
completely corroborated by this study. The species in their the location of species assemblages), reemphasized the relation-
“southern shelf group” were placed into the stripetail rockfish ship between species and depth (Table 6). The species groups
group which included all shallow-water species. However, their were arranged such that they went from shallow to deep. In all
“deep water slope” group was split among four of this report's cases, the assemblages with a low frequency of occurrence for all
groups. Comparisons of the results from this study with other site groups were composed of relatively rare species. Depth asso- Site Groups
assemblage studies are difficult due to the variability in species ciations were present, just not discernible, given the methodology
Group 263 Meters
analyzed between studies, the different habitats targeted, and for defining “influential” assemblages. The location of the trawls
the discrepancy in scale. For example, results from submers- Group 410 Meters
contained in each site group were mapped (Figure 23).
ibles (Yoklavich et al., 2000, 2002; Hixon et al., 1991; Hixon Group 530 Meters
38°N
38°N
and Tissot, 1992; Field et al., 2002) record interactions at a Habitat Correlations (Objective D)
Group 622 Meters
much smaller scale compared to trawls, which can fish multiple Other factors besides depth, such as latitude (Williams and
Group 733 Meters
habitats during a 1 km tow. Ralston, 2002; Horn and Allen, 1978; Sullivan, 1995), bottom
composition (Yoklavich et al., 2000, 2002; Field et al., 2002; Hixon Group 931 Meters
Site Groups and Interaction of Species and Sites (Objec- et al., 1991; Hixon and Tissot, 1992), or bathymetric complexity
Group 1112 Meters
tives B and C) (pp. 16) (Yoklavich et al., 2000, 2002; Field et al., 2002) can have
Seven site clusters were identified from the 454 slope trawls an impact on species assemblages. For this data set, 95% of the 0 10 20 40 60 80
(Table 5). To make interpretation easier, the site groups are trawls occurred over areas delineated as mud (pp. 38), making
Kilometers
named according to mean depth. Species assemblages which it impossible to examine the effects of bottom composition. It is
were influential in forming each site group are identified. The interesting to note that of all the 16 trawls completed over the
37°N
37°N
average frequency of occurrence of fish species assemblages bottom type designated as a combination of mud and rock, 33
for each site group (Table 6) was used to determine where percent occurred in the "410 meters" site group, and 56 percent in
species assemblages were found. An ANOVA determined that the "deepest" group. For the slope trawls, there was no interaction
all groups were significantly different in depth (Table 5). The present between depth and bathymetric complexity or latitude, so
importance of depth in this ecosystem is not a new idea; many the effects of these parameters could be tested. Neither bathy-
researchers have already commented on its influence (Williams metric complexity (pp 16), nor latitude, had a significant impact
and Ralston, 2002; Sullivan, 1995; Gabriel and Tyler, 1980; Field on site grouping when the effect for depth was accounted for
et al., 2002; Matthews and Richards, 1991). Latitude has also (bathymetric complexity: df=1, F=0.94, P=0.33; latitude: df=1,
been described as having an influence on California fish spe- F=0.11, P=0.74).
cies composition (Williams and Ralston, 2002; Horn and Allen,
36°N
36°N
In conclusion, this analysis provides results
showing species assemblages, site assem-
Group 263 Group 410 Group 530 Group 622 Group 733 Group 931 Group 1112
meters meters meters meters meters meters meters
blages, and the location of species assem-
blages for the deep slope environment that
0.41 0.78 0.87 0.97 0.95 0.84 0.68
Sablefish Assemblage
are relevant to the scale of a commercial
Aurora Rockfish
trawl. The larger species assemblages
0.40 0.75 0.55 0.31 0.27 0.25 0.17
Assemblage
that were reported in the literature were
Stripetail Rockfish
0.51 0.08 0.00 0.00 0.00 0.00 0.00
confirmed (NMFS, 2002) (near-shore,
Assemblage
shelf and slope groups), but some of the
Splitnose Rockfish
0.88 0.92 0.64 0.36 0.18 0.04 0.06
Assemblage
smaller groups were not corroborated. Half
Filetail Catshark
of the sites designated as "1,112 meters"
35°N
35°N
0.33 0.51 0.43
0.03 0.23 0.08 0.05
Assemblage
are located outside sanctuary boundaries.
Longspine Thornyhead
This is mainly due to the large number of
0.65 0.87 0.84
0.01 0.05 0.22 0.41
Assemblage
deep sites located to the south and west of
20
Pacific Viperfish
100
50 m
00
0.27
0.01 0.02 0.06 0.09 0.14 0.16
20
Sanctuary boundaries. The results of this
Assemblage
m
0
m
analysis in conjunction with similar analyses
m
Market Squid
0.14 0.15 0.13 0.07 0.07 0.03 0.05
Assemblage on the three other data sets provides a fairly
comprehensive overview of fish and macro-
Table 6. Average frequency of occurrence of fish species assemblages (percent occurrence 124°W 123°W 122°W 121°W
calculated for each species and then averaged for each fish assemblage) for each slope site invertebrate species within the study area. Figure 23. Location of site groups for NMFS slope trawls. Lines showing the 50, 100, 200, and 2,000 depth contours are provided.
group. Bold numbers represent influential species assemblages within that site group.
29
Subsection 2.1.1: ASSEMBLAGE ANALYSES
if, on average, species were present in 25% of the trawls in a site group.
ABOUT THESE MAPS
Interactions between environmental variables (salinity, temperature, density,
The pelagic environment is home to many marine species at some stage in
NMFS Midwater Trawls bottom depth, and bathymetric complexity) were investigated by conducting
their life. Therefore, it is important to document what species interact in this
step-wise discriminant analyses.
Species Assemblages
environment, and determine environmental influences on species abundance.
Multivariate statistics were used to analyze fish and invertebrate species
(All Years) RESULTS AND DISCUSSION
assemblages caught in trawls conducted at 7 and 30 meters depth between
Pacific Hake, juv. Species Assemblages (Objective A)
Cordell Bank NMS and Monterey Bay. For more information on the neritic
Calif. Smoothtongue The neritic environment is an important ecosystem in central California. Most
environment see the Ecological Linkages Report. When determining species
Deep-sea Smelt
benthic species have a larval stage dependent on the neritic environment.
assemblages, three separate analyses were completed: all data (1986-2001), Canary Rockfish, juv. Medusafish
Euphausiid
In addition, neritic species are an important base for the food web for fish,
only 1998 (warm year), and only 1999 (cold year). There were differences Black Rockfish, juv. King-of-the-salmon
Myctophid
Slender Barracudina Blue Rockfish, juv. birds, and mammals. Due to the removed rare species, different species were
between years in the species present, as well as in the organization of Rex Sole, juv.
Bocaccio, juv. included in the analyses depending on the year (1998, 1999, or all years).
species assemblages. The site groups identified were significantly related Slender Sole, juv.
Chilipepper, juv. Ten species were present in 1999 that were absent in 1998, including six
to environmental conditions, but the influential conditions varied between Pacific Tomcod, juv.
Pygmy Rockfish,juv.
species of juvenile rockfish. For 1998 and 1999, there were five and six spe-
years. To investigate assemblages persistent over more than one year, all Sand Sole, juv.
Shortbelly Rockfish, juv.
Squarespot Rockfish, juv. Slender Sole, adult cies assemblages identified, respectively, and seven species assemblages
of the data collected was analyzed for species assemblages, but not for site
were differentiated in the entire data set. Species assemblages were named
assemblages, due to the difficulty in mapping assemblages through time as Stripetail Rockfish, juv.
Spiny Dogfish according to the most influential species (Figures 24, 25, 27). Overall, the
the same location may house any number of assemblages depending on Widow Rockfish, juv.
Pacific Hake, adult
Yellowtail Rockfish, juv. species assemblages were much less robust than those from the other data
environmental conditions. See Larson et al (1994) for a detailed analysis of
sets. Average persistence (percentage of time species were grouped to-
rockfish assemblages in response to short-term environmental variability.
Pacific Sanddab, juv.
gether) through random runs varied from 36% to 94% for all assemblages.
Through this analysis a large amount of information has been condensed Market Squid
Brown Rockfish, juv.
This variability in results reflects two things: 1) the ephemeral nature of the
down to assemblages of co-occurring species, as well as groups of similar Lingcod, juv.
Copper Rockfish complex
neritic ecosystem and its expression through the species assemblages, and
locations. Maps are provided to portray the spatial arrangement of the site Dover Sole, juv. Northern Anchovy
2) the higher variability in results from the random runs due to smaller sample
Northern Anchovy, larval
groups for 1998 and 1999. Pacific Butterfish
Pacific Argentine, juv. size. Some of the persistent groups (persistent through random runs as well
Pacific Electric Ray
Speckled Sanddab, juv. as persistent through the three analyses) were: 1) market squid, northern
DATA SOURCES Pacific Sanddab, adult
anchovy, Pacific electric ray, and Pacific sardine; 2) euphasid, Pacific hake,
Data from 1543 fisheries independent research trawls were collected from Pacific Sardine
and deep sea smelt; and 3) myctophid and slender barracuda. Different
1986-2001, during May and June. The purpose of the trawl was to determine Plainfin Midshipman
juvenile rockfish were present in 1998 and 1999. In 1998, a warm year, only
the number and quantity of juvenile rockfish present during the upwelling
Pictures from FishBase and NMFS
shortbelly and stripetail rockfish were in greater than 5% of the trawls, and
season at night. The midwater trawl net was a “Cobb trawl” constructed of
nylon webbing. A fine mesh (1.25 cm) liner was inserted in the codend to Figure 24. Species assemblage results for the midwater trawls utilizing all data from 1986 to 2001. Assemblages were grouped together. In 1999, a cold year, 8 species of juvenile rockfish
are named for the most influential species in each group. Non-italicized species were consistently placed into
were identified, and all grouped together except for blue rockfish and stripetail
retain small midwater organisms. To open the net vertically, 42 8-inch floats the same species assemblage >80% of the time; italicized species tended to roam into other assemblages with
rockfish. This supports the observation by Loeb et al. (1994), Yoklavich et al.
were attached to the headrope and 145 pounds of chain was lashed to the random sampling.
(1996), and Moser et al. (2000), that warm (El Niño) years are not good for
footrope. To spread the net horizontally, 850 lb steel ‘V’ doors 5 x 7 feet in
rockfish recruitment. For the entire data set there were thirteen species of
A. Determine which species tended to be caught together (species as assemblages);
size were used. The net was deployed while the ship was underway at a
rockfish, which were all grouped together except for copper rockfish complex
B. Analyze fishing locations to determine which locations contained similar catches (site
speed of approximately 2.7 knots. Seventy-five meters of trawl cable were
and brown rockfish. The consistent grouping of juvenile rockfish together
groups);
payed-out which nominally puts the headrope at a depth of 30 meters (100
suggests that the rockfish species are responding to similar environmental
C. Resolve where the species assemblages were being caught by combining results from
feet), the depth at which most species of juvenile rockfishes are most abun-
conditions. The copper and brown rockfishes are not well known; however,
objectives A and B and then utilizing GIS to map the results; and
dant (Lenarz et. al., 1991). The spatial effort was consistent between years
Larson et al. (1994) looking at the midwater trawls in 1987 and 1988, deter-
D. Identify significant relationships between site groups identified in objective B and broad
with 1-4 hauls completed at 32 stations (see Figure 26 for spatial extent of
mined that some species in the copper complex arrive later in the season
scale habitat characteristics (bathymetry, bathymetric complexity, and large-scale habitat
data). Analyses were completed for the entire data set, as well as 1998 and
than most other rockfish species, which could affect their association with
classification).
1999 individually (representing warm and cold water years, respectively). Fish
assemblages.
numbers were adjusted for effort by NMFS then log transformed for analyses.
The data matrices used for classification contained data for 16, 24, and 41 Clustering is a technique used to summarize information into similar groups. The 1-Pearson correla-
Running the modified bootstrap technique can provide an estimate of the
species in 1998, 1999, and all years, respectively. A complete list of common tion coefficients with the average means clustering method (see "Introduction to Clustering" pp. 14)
precision of results, but verifying the accuracy of the results is more difficult.
and scientific names of the species included in these analyses is available on was used to first summarize the fish species into assemblages, and to then summarize the catch
Only two studies were identified which have investigated species assem-
locations into site groupings. Species assemblages were determined for 1998, 1999, and all data,
the accompanying CD-ROM.
blages in the neritic environment (Larson et al., 1994; Calliet et al., 1979).
but site groups were only determined for 1998 and 1999. In order to determine how variable the
Larson et al. (1994) analyzed this data set (1987-1988) for juvenile rockfish
species cluster results could be within the data, a modified bootstrapping procedure was employed
METHODS
assemblages, looking at each sweep individually, and described the short
The aim of the entire assemblage analysis was to increase our understanding on random samples composed of 75% of the data and the results compared for persistence and
term variation in assemblages and environmental conditions. Longer term
of the biogeography of fishes and macro-invertebrates in relationship to their precision. The location of the site groups were mapped using GIS. To determine which species
trends were not analyzed. Since the species groups changed with each
environment, and identify important areas or habitats. Four of the five main groups were influential in forming the site groups, the average frequency of occurrence for species
sweep, and most rockfish were grouped together for this study, comparing
assemblages in each site group was calculated. Species assemblages were considered influential
objectives were addressed in this analysis:
30
Subsection 2.1.1: ASSEMBLAGE ANALYSES
results from the two studies is difficult. Calliet et al. 123°W 122°W
(1979) analyzed midwater trawls, and purse seines,
NMFS Midwater Trawls
NMFS Midwater Trawls
to determine species assemblages associated with
Species Assemblages
Myctophid market squid in shallow and deep environments. In
(1998)
Deep-sea Smelt
the results from anchovy hauls, only three species
1998 Trawls
Slender Barracudina
showed high affinity with market squid in both shallow
Pacific Hake, juv.
and deep hauls: northern anchovy, pacific electric ray,
Calif. Smoothtongue
and Pacific herring. This study's results are similar
Euphausiid
since all species assemblages placed market squid,
Stripetail Rockfish, juv.
Site Groups
northern anchovy, and Pacific electric ray together.
Shortbelly Rockfish, juv.
Plainfin Midshipman Pacific herring were not present in this analysis. Site Group A
Market Squid Site Group B
Site Groups and Interaction of Species and Sites
Northern Anchovy
38°N
38°N
Site Group C
Pacific Electric Ray (Objectives B and C)
Pacific Sardine
For 1998 and 1999, six site groups were identified for Site Group D
each year. No analysis was run to group sites from the
Northern Anchovy, larval Site Group E
Pacific Sanddab, juv. entire data set, since preliminary results suggested
Site Group F
Pacific Sanddab, adult no spatial trends in results. Because groups are from
Speckled Sanddab, juv.
Not Slender Sole, adult
in
at a fluentia midwater environments they are not named by depth, 0 10 20
ny d
epth l but distinguished as site group A, site group B, etc. Kilometers
Pictures from FishBase and NMFS
Species assemblages which were influential in form-
Figure 25. Species assemblage results for the midwater trawls conducted in 1998. ing each site group are identified. The average fre-
Assemblages are named for the most influential species in each group. Non-itali-
quency of occurrence of fish species assemblages for
cized species were consistently placed into the same species assemblage >80%
each site group (Tables 7 and 8) is used to determine
of the time; italicized species were more ephemeral and tended to roam into other
where species assemblages were found. Maps with
assemblages with random sampling.
the location of the site groups are provided (Figures
26 and 28). Multiple assemblages can be found at
the same site as 3-4 sweeps were made per year
and environmental conditions could change between
sweeps (see Larson et al, 1995).
50
m
10
Habitat Correlations (Objective D)
Site group A Site group B Site group C Site group D Site group E Site group F 0m
20
Depending on the data set, various environmental
37°N
37°N
0
m
N trawls 15 20 14 5 11 26 conditions had a significant influence on site groups.
In 1998, a warm year, bottom depth (log transformed:
Myctophid
0.00 0.02 0.02 0.00 0.18 0.49 N=91, F=18.94, P=<0.0001) was significant. In 1999,
Assemblage
a colder year, bottom depth (log transformed: N=91,
Pacific Hake, juv. F=23.87, P=<0.0001), latitude (N=91, F=8.76,
0.17
0.40 0.31 0.40 0.76 0.62
Assemblage P=<0.0001), and water density (N=91, F=5.93,
Stripetail Rockfish, P<0.0001) were significant. The significance of bot-
0.13 0.08 0.00 0.18 0.15
0.55
juv. Assemblage tom depth to each analysis highlights the importance
m
of location on species assemblages even in the neritic
00
Market Squid
0.00 0.07
0.33 0.58 0.29 0.30
20
environment. Groups were labeled according to their
Assemblage
mean depth (i.e. the shallowest group was group A
Pacific Sanddab and the deepest was group F). The Pacific sanddab
0.10 0.07 0.07 0.12 0.01
0.27
Assemblage assemblage in 1998 and market squid assemblage in
1999 were only influential in the shallowest site group.
Table 7. Average frequency of occurrence of fish species assemblages (percent 10
20
0 m 0 m 50 m
In 1999 there was a north/south split where group F
occurrence calculated for each species and then averaged for each fish assem-
20
00
was more southern and groups B and C more northern
blage) for each 1998 midwater site group. Number of trawls in each site group is
m
provided in the first row. Bold numbers represent influential species assemblages (except for one point each south of Point Año Nuevo). 123°W 122°W
within that site group. In 1998, there were only two rockfish species captured
in greater than 5% of the trawls (stripetail rockfish as- Figure 26. Location of site groups for NMFS 1998 midwater trawls. Lines showing the 50, 100, 200, and 2,000 depth contours are
provided.
31
Subsection 2.1.1: ASSEMBLAGE ANALYSES
123°W 122°W
semblage) which were only influential in site group
NMFS Midwater Trawls C. In 1999, there were six rockfish species grouped
NMFS Midwater Trawls
Species Assemblages together (canary rockfish assemblage), which were
Pacific Hake, juv.
(1999) influential for site groups C and E. The 1999 rock-
Calif. Smoothtongue
1999 Trawls
fish were caught in trawls further offshore than the
Deep-sea Smelt
Euphausiid trawls that contained rockfish in 1998.
Myctophid Canary Rockfish, juv.
Black Rockfish, juv.
These analyses provide trends in species as-
Medusafish
Bocaccio, juv.
Chilipepper, juv. semblages in the midwater environment during Site Groups
Shortbelly Rockfish, juv. Pacific Hake, adult the upwelling season. The grouping of juvenile
Site Group A
Widow Rockfish, juv.
rockfish together suggests that when conditions
Pacific Sanddab, juv.
Slender Sole, juv.
Site Group B
are suitable for one species, they are also suit-
Blue Rockfish, juv.
38°N
38°N
Dover Sole, juv. able for the other species. Within the entire data Site Group C
Lingcod, juv. set, three species assemblages were consistently
Site Group D
Pacific Argentine, juv.
grouped together greater than 80% of the time:
Pacific Sanddab, adult
Site Group E
Pacific hake, juv. assemblage; canary rockfish, juv.
Speckled Sanddab, juv.
Stripetail Rockfish, juv. Market Squid
assemblage; and spiny dogfish assemblage. Two Site Group F
Northern Anchovy
groups were slightly less stable and were grouped
Not Pacific Electric Ray
influ 0 10 20
together greater than 70% of the time: medusafish
entia
at an Pacific Sardine
l
y de
pth assemblage and market squid assemblage. Even Kilometers
Pictures from FishBase and NMFS
though results were not as stable as with the other
Figure 27. Species assemblage results for the midwater trawls conducted in 1999.
data sets, this analysis identifies assemblages that
Assemblages are named for the most influential species in each group. Non-italicized
were consistent through time. The reduced number
species were consistently placed into the same species assemblage >80% of the time;
of species present in 1998 highlights the effects of
italicized species were more ephemeral and tended to roam into other assemblages
water temperature on species assemblages in the
with random sampling.
neritic environment. For this data set, both bottom
depth, latitude, and water density were found to
have a significant influence on site groups.
50
m
10
0
20
m
37°N
37°N
0
Site group A Site group B Site group C Site group D Site group E Site group F
m
N trawls 23 7 9 21 16 15
Pacific Hake, juv.
0.21 0.37 0.40 0.33 0.69 0.77
Assemblage
Canary Rockfish, juv
0.08 0.02 0.07 0.15
0.25 0.53
Assemblage
Medusafish 0.04 0.00 0.11 0.05 0.13 0.00
m
00
20
Pacific Hake, adult 0.04 0.00 0.00 0.06 0.00
0.38
Pacific Saddab, juv
0.14 0.19
0.40 0.30 0.34 0.27
Assemblage
100
20 50
Market Squid 0 m m
m
0.11 0.06 0.12 0.05 0.03
0.55
20
Assemblage
00
m
Table 8. Average frequency of occurrence of fish species assemblages (percent occurrence calculated for each 123°W 122°W
species and then averaged for each fish assemblage) for each 1999 midwater site group. Number of trawls in
Figure 28. Location of site groups for NMFS 1999 midwater trawls. Lines showing the 50, 100, 200, and 2,000 depth contours are
each site group is provided in the first row. Bold numbers represent influential species assemblages within that
provided.
site group.
32
Subsection 2.1.1: ASSEMBLAGE ANALYSES
SECTION SUMMARY tal conditions, especially seasonal water
0
Twenty-eight species assem- temperature (1998 warm year vs. 1999 cold
Depth Range of Species Assemblages
blages were identified from the year) was obvious in its influence on what
Gopher Rockfish
Black Rockfish
25
CDF&G recreational, NMFS species were present in the neritic environ-
Brown Rockfish
Yellowtail Rockfish
Cabezon
shelf, and NMFS slope data sets. ment during upwelling season. Using data
Canary Rockfish
CDF&G Recreational
China Rockfish
Copper Rockfish
Figure 29 illustrates the overlap Kelp Greenling
from all years, species assemblages could
Hook and Line
Lingcod
Pacific Sanddab
Rosy Rockfish
between the three data sets. The be delineated, but these assemblages were
50 English Sole
Starry Rockfish
NMFS Shelf Trawls
Petrale Sole
length of the vertical line depicts more sensitive with regards to random
Vermilion Rockfish Chilipepper Stripetail Rockfish
Pink Seaperch
Pacific Sanddab Bocaccio Bocaccio
the depth interval where species samples. This emphasizes the ephemeral
Plainfin Midshipman
Greenspotted Cowcod Chilipepper
Lingcod
Rockfish
Blue Rockfish
NMFS Slope Trawls
Greenspotted Rockfish Darkblotched Rockfish
assemblages were "influential" nature of the neritic environment, and the
75 Greenstriped Halfbanded Rockfish
Olive Rockfish Greenstriped Rockfish English Sole
Depth (meters) Rockfish Pacific Mackerel Pacific Hake
Bocaccio Shortbelly Rockfish Greenstriped Rockfish
(see Tables 2, 4, 6-8). Shading resulting transient nature of the species as-
Flag Rockfish Chilipepper Jack Mackerel Longnose Skate Stripetail Rockfish Lingcod
Pacific Argentine
was included to give the impres- semblages. There were three species as-
Spotted Ratfish
Speckled Rockfish Petrale Sole
Pacific Herring
Widow Rockfish Sharpchin Rockfish
Shortspine
sion of where the continental shelf semblages that occurred together in all data
American Shad
Yelloweye Rockfish Shortbelly Rockfish
100 Thornyhead
Big Skate
California Market Squid Spot Shrimp
Bering Skate
California Skate
end and continental slope begins. sets: 1) Market squid, Northern anchovy,
Shelf break Filetail Catshark
Chinook Salmon Slender Sole
Darkblotched Rockfish
Bigfin Eelpout
Spiny Dogfish Black Eelpout
Curlfin Sole Redbanded Rockfish
The edge between the shelf and Bank Rockfish
Pacific electric ray, and Pacific sardine;
Dover Sole Longspine Thornyhead
California
Dungeness Crab Rosethorn Rockfish
Redbanded Rockfish
Sablefish Black Skate
Grenadier
Longspine Combfish
slope, although variable within 2) Euphausiids, Pacific hake, and deep
Pacific Electric Ray
Splitnose Rockfish
Rex Sole Califonia Slickhead
Flapjack Devilfish
325 Northern Anchovy Crimson Pasiphaeid
Slender Sole Pacific Glass
the study area, was presented at sea smelt; and 3) Myctophid and slender
Pacific Pompano Deepsea Sole
Spotted Cusk-eel Blackgill Rockfish Splitnose Shrimp
White Croaker Giant Grenadier
200 meters to be consistent with barracuda. In addition, most larval rockfish
Aurora Rockfish Grooved Tanner Crab
Pacific Electric Ray Rockfish
Black Eelpout Pacific Flatnose
Bering Skate
Williams and Ralston (2002). In species co-occur. More in-depth analyses,
Aurora Rockfish Pacific Grenadier
Blacktail Snailfish Bigfin Eelpout
550 Blackgill Rockfish Snakehead Eelpout
Brown Cat Shark Longnose Skate
all cases, the assemblages with a taking advantage of the available informa-
Twoline Eelpout
Dover Sole
Filetail Cat Shark Pacific Hake
Spiny Dogfish Pacific Viperfish
Lanternfish
low frequency of occurrence at all tion on environmental conditions, could be
Rex Sole
Bank Rockfish Black Hagfish
Spotted Ratfish Deepsea Skate
depths were composed of species conducted (see Larson et al., 1994).
775
Fangtooth
Sablefish Longfin Dragonfish
present in less than 20% of the Blacktail Snailfish
Assemblages with Low Frequency of Occurrence at All Depths Pacific Blackdragon
Brown Catshark
In conclusion, species assemblages and Rhomboid Squid
trawls. For these assemblages, Shortspine Sawtooth Eel
Canary Rockfish Smooth Grenadier
site groups were delineated and mapped Thornyhead
depth associations may have Sharpchin Rockfish California Market Squid
1000 Pacific Mackerel Yellowtail Rockfish Threadfin Slickhead
Spot Shrimp
Quillback Rockfish Arrowtooth Flounder Blackbelly Eelpout Vampire Squid
for four separate data sets. Depth had a
Widow Rockfish
Squarespot Rockfish
been present, just not discernible, Threadfin Sculpin Robust Clubhook Squid
Magistrate Armhook Squid
Rock Sole Sp.
significant influence on all four data sets.
given the methodology for defining Figure 29. Overlap between the three data sets that analyzed demersal fish: CDF&G recreational (yellow), NMFS shelf (green), and NMFS slope (orange). The length of the vertical line
The influence of depth is not a new con-
“influential” assemblages. depicts the depth interval where species assemblages had an average frequency of occurrence in at least 25% of the trawls (see Tables 2,4,6-8). Non-italicized species were consistently
cept (Williams and Ralston, 2002; Sullivan,
placed into the same species assemblage; italicized species tended to roam into other assemblages with random sampling.
Data Sources 1995; Gabriel and Tyler, 1980; Field et al.,
See associated sections for information and spatial extent may have overlapped for the 200-500 meters depth range, more difficult to compare due to the different fishing methods 2002; Matthews and Richards, 1991); however, this is the
of CDF&G recreational hook and line data (pp. 23), NMFS the species included in the analyses differed. It is interesting employed. Only nine species overlap between the recreational first time a study has demonstrated its significance on three
demersal trawls on the continental shelf (pp. 26), and NMFS to note that the shallow water species included in the NMFS data and the shelf trawls. For both data sets, chilipepper, green- separate data sets. All attempts to remove depth and look for
demersal trawls on the continental slope (pp. 28). slope trawl analysis were all placed in one assemblage, the striped rockfish, and greenspotted rockfish were grouped to- secondary influences on group designation were unsuccessful.
stripetail rockfish assemblage, and that this assemblage was gether (chilipepper and greenspotted assemblages) and found For the neritic environment, depth, latitude, and water density
Methods only present in the shallowest slope trawl site group. The same at similar depths. For the most part, fish species were associ- had a significant impact on site groups in 1999. Starr (1998)
Results from the overlap between species assemblages and species included in this stripetail rockfish assemblage were ated with a shallower depth with the recreational hook and line addressed the implementation of rockfish no-take areas and
site groups from each analysis (see Tables 2, 4, 6-8) were used found in five different NMFS shelf assemblages. Conversely, analysis than with the shelf trawl analysis. This difference could made two important recommendations. First, in order to prop-
to determine the depths at which species assemblages were the blackgill assemblage on the NMFS shelf trawls contains be due to the nature of the assemblage analysis or due to the erly manage marine ecosystems, there is a need for a better
present. For each assemblage, the shallowest site group (mean the deeper species caught in the shelf trawls and found within variable size selectivity of the fishing methods and habitats. understanding of fish assemblages. Once these assemblages
depth minus the standard deviation) from which their average three different NMFS slope species assemblages. This does Many of the deepwater rockfish species settle as juveniles in are delineated, managers can take steps to ensure each as-
frequency of occurrence was >25% was used to determine not imply that species co-occurrences changed between the shallow water, and slowly shift to deeper water as they mature semblage receives proper management. The results from this
the minimum depth. Similarly, the mean plus the standard de- shelf and slope trawls, just that cluster results were sensitive to (Love et al., 2002). Future analyses could include information study provide information on these assemblages for near-
viation for the deepest site group was used to determine the the depth range covered by the data set. For example, bocac- on fish total length to determine if ontogenetic shifts occur and if shore, shelf, slope, and midwater ecosystems. The second
maximum depth. cio and English sole do not co-occur, as bocaccio is attracted they generate the differences in species’ depth range between recommendation by Starr (1998) was to delineate rectangular
to rocky ledges and English sole to soft bottom areas with low data sets noted above. The effect water temperature has on the no-take areas that cover 20-50 km of the coast and extend west
Results and Discussion relief (Love et al., 2002). These species are grouped together species present, and the composition of species assemblages, to the edge of the continental shelf. From a biogeographic view-
The shallow assemblages had more limited depth ranges, 92% of the time when included in an analysis with deep slope was investigated for the recreational and shelf data sets and point, the spatial analyses coincide with that recommendation
which is not obvious given the log scale of depth in Figure 29. species, but never grouped together when included in an provides preliminary results on which assemblages are persis- and also determined that deep slope communities contribute
Species included with the three data sets differed, especially analysis with shelf species. significantly to ground fish biogeographic patterns. Because
tent through environmental change (see CD-ROM).
since only species caught in at least 5% of the trawls were assemblages follow bathymetry at the scale of this analysis,
analyzed. Therefore, while the NMFS shelf and slope trawls The overlap between the shelf and recreational trawls was For the midwater trawl data set, the importance of environmen- this approach could protect all demersal species assemblages
identified in this study.
33
Subsection 2.1.1: ASSEMBLAGE ANALYSES
REFERENCES Horn, M.H. and L.G. Allen. 1978. A distributional analysis of California National Marine Fisheries Service. 2001. Endangered Species Act. Williams, E.H. and S. Ralston. 2002. Distribution and co-occurrence
Baraff L.S. and T.R. Loughlin. 2000. Trends and potential interactions coastal marine fishes. J. Biogeog., Vol. 5, pp. 23-42. Section 7 Consultation - Biological opinion on authorization of Gulf of of rockfishes over trawlable shelf and slope habitats of California and
between pinnipeds and fisheries of New England and the U.S. West Alaska and Bering Sea/ Aleutian Islands groundfish fisheries based southern Oregon. Fish. Bull., Vol. 100, pp. 836-855.
Coast. Mar. Fish. Rev., Vol62, pp.1-39. Jay, C.V. 1996. Distribution of bottom-trawl fish assemblages over the on the Fishery Management Plan for the Gulf of Alaska and Bering
continental shelf and upper slope of the U.S. west coast, 1977-1992. Sea/ Aleutian Islands groundfish as modified by amendments 61 Wilkins, M.E., M. Zimmerman, and K.L. Weinberg. 1998. The 1995
Boesch, D.F. 1977. Application of numerical classification in ecologi- Can. J. Fish. Aquat. Sci., Vol. 53, pp.1203-1225. and 70. NMFS-Alaska Region, Protected Resources Division, Silver Pacific west coast bottom trawl survey of groundfish resources: Es-
cal investigations of water pollution. Special Scientific Report No. 77, Spring, MD. timates of distribution, abundance, and length and age composition.
EPA-600/3-77-033. Virginia Institute of Marine Science, Williamsburg, Larson, R.J., W.H. Lenarz, and S. Ralston. 1994. The distribution of NOAA Tech. Memo NMFS-AFSC-89. Seattle, WA. 138 pp.
VA. 114 pp. pelagic juvenile rockfish of the genus Sebastes in the upwelling region National Marine Fisheries Service. 2002. Magnuson-Stevens Act
off central California. CalCOFI Rep., Vol. 35, pp. 175-221. Provisions. http://www.nwr.noaa.gov/1sustfsh/groundfish/proposed_ Wilson-Vandenberg, D., P.N. Reilly, and C.E. Wilson. 1996. Onboard
Brown, S.K., K.R. Buja, S.H. Jury, and M.E. Monaco.2000. Habitat rule.pdf. Federal Register, Vol. 67(8), 1582 pp sampling of the rockfish and lingcod commercial passenger fishing
suitability index models for eight fish and invertebrate species in Lauth, R.R. 2001. The 2000 Pacific west coast upper continental slope vessel industry in northern and central California, January through
Casco and Sheepscot Bays, Maine. N. Amer.J. Fish. Mgmt., Vol. 20, trawl survey of groundfish resources off Washington, Oregon and Cali- Paukert, C.P. and T.A. Wittig. 2002. Applications of multivariate statisti- December 1994. Marine Resources Division Administrative Report
pp. 408-435. fornia: estimates of distribution, abundance, and length composition. cal methods in fisheries. Fisheries, Vol. 27(9), pp. 16-22. No. 96-6. San Francisco, CA. 96 pp.
NOAA Tech. Memo. NMFS-AFSC-120. Silver Spring, MD. 284 pp.
Cailliet, G.M., K.A. Karpov, and D.A. Ambrose. 1979. Pelagic as- Ralston, S. 1998. The status of federally managed rockfish on the Worm, B. and R. A. Myers. 2003. Meta-analysis of cod-shrimp
semblages as determined from purse seine and large midwater trawl Lenarz, W.H., R.J. Larson, and S. Ralston. 1991. Depth distributions U.S. West Coast in Marine harvest refugia for West Coast rockfish: A interactions reveal top-down control in oceanic food webs. Ecology,
catches in Monterey Bay and their affinities with the market squid, of late larvae and pelagic juveniles of some fishes of the California workshop. NOAA-TM-NMFS-SWFSC-255. Pacific Grove, California. 84(1), pp. 162-173.
Loligo opalescens. CalCOFI Rep., Vol. 20, pp. 21-30. Current. Calif. Coop. Oceanic Fish. Invest. Rep., Vol. 32, pp. 41-46. 11 pp.
Yoklavich, M.M., H.G. Greene, G.M. Cailliet, D.E. Sullivan, R.N. Lea,
Carr, M.H. 1991. Habitat selection and recruitment of an assemblage Loeb, V., M. Yoklavich, and G. Cailliet. 1994. California Sea Grant Shannon, C.E., and W. Weaver. 1949. The mathematical theory of and M.S. Love. 2000. Habitat associations of deep-water rockfishes
of temperate zone reef fishes. J. Exp. Mar. Biol. Ecol., Vol. 146, pp. biennial report of completed projects, 1992-1994. Moss Marine Labo- communication. University of Illinois Press, Urbana, IL. 125pp. in a submarine canyon: An example of a natural refuge. Fish. Bull.,
113-137. ratories: La Jolla, CA. Pp. 79-95. Vol. 98, pp. 625-641.
Shaw, F.R., Wilkins, M.E., Weinberg, K.L., Zimmermann, M., and
Chavez, F. P, J. Ryan, S. E. Lluch-Cota, and M. Ñiquen. 2002. From Love, M.S., J.E. Caselle, and W.V. Buskirk. 1998. A severe decline R.R. Lauth. 2000. The 1998 Pacific west coast bottom trawl survey of Yoklavich, M.M., V.J. Loeb, M. Nishimoto, and B. Daly. 1996. Near-
anchovies to sardines and back: Multidecadal change in the Pacific in the commercial passenger fishing vessel rockfish (Sebastes spp.) groundfish resources: estimates of distribution, abundance, and length shore assemblages of larval rockfishes and their physical environment
Ocean. Science, Vol 299, pp. 217-221. catch in the southern California Bight, 1980-1996. CalCOFI Rep., and age composition. U.S. Department of Commerce, NOAA Technical off central California during an extended El Niño event, 1991-1993.
Vol. 39, pp. 180-195. Memorandum NMFS-AFSC-114. Silver Spring, MD. 138 pp. Fish. Bull., Vol. 94, pp. 766-782.
Estes, J.A., M.T. Tinker, T.M. Williams, and D.F. Doak. 1998. Killer
whale predation on sea otters linking oceanic and nearshore ecosys- Love, M. S. 1996. Probably more than you want to know about the Starr, R.M. 1998. Marine harvest refugia for West Coast rock- Yoklavich, M.M, G.M. Cailliet, R.N. Lea, H.G. Greene, R.M. Starr, J.
tems. Science, Vol 282, pp.473-476. fishes of the Pacific Coast. Santa Barbara, Calif.: Really Big Press. fish: A workshop. NOAA-TM-NMFS-SWFSC-255. Pacific Grove, de Marignac, and J. Field. 2002. Deepwater habitat and fish resources
88 pp. California.14pp. associated with a marine reserve: implications for fisheries. Marine
Field, J.M., M.M. Yoklavich, J. deMarignac, G.M. Cailliet, R. N. Lea, Ecological Reserves Research Program Research Results 1996-2001.
and S.M. Bros. 2002. Small-scale analysis of subtidal fish assem- Love, M.S., M.M. Yoklavich, and L. Thorsteinson. 2002. The rockfishes Starr, R.M, J.M. Cope, and L. A. Kerr. 2002. Trends in fisheries and California College Sea Grant Program CD-Rom. www.csgc.ucsd.edu.
blages and associated habitat characteristics off Central California of the northeast Pacific. University of California Press, Berkeley and fishery resources associated with the Monterey Bay National Marine La Jolla, CA. 63 pp.
in Marine Ecological Reserves Research Program: research results Los Angeles, CA. 404 pp. Sanctuary from 1981-2000. California Sea Grant College Program
1996-2001. California College Sea Grant Program CD-Rom, La Jolla, Publication No. T-046. Monterey, CA. 156 pp. Zar, J.H. 1999. Biostatistical Analysis, fourth edition. Prentice-Hall,
CA, www.csgc.ucsd.edu. 16 pp. Mason, J.E. 1995. Species trends in sport fisheries, Monterey Bay, Inc. Upper Saddle River, New Jersey. 80 pp.
California, 1959-86. Mar. Fish. Rev., Vol. 57(1), pp. 1-16. Sullivan, C.M. 1995. Grouping of fishing locations using similarities
Gabriel, W.L. and A.V. Tyler. 1980. Preliminary analysis of Pacific in species composition for the Monterey Bay area Commercial Pas- Zimmerman, M., M.E. Wilkins, K.L. Weinberg, R.R. Lauth, and F.R.
Coast demersal fish assemblages. Mar. Fish. Rev., Vol. 42(3-4), pp. Mason, J.E. 1998. Declining rockfish lengths in the Monterey Bay, senger Fishing Vessel Fishery, 1987-1992. California Department of Shaw. 2001. Retrospective analysis of suspiciously small catches in
83-88. California, recreational fishery, 1959-94. Mar. Fish. Rev., Vol. 60(3), Fish and Game, Marine Resources Technical Report No. 59. Monterey, the National Marine Fisheries Service West Coast Triennial Bottom
pp. 15-28. CA. 37 pp. Trawl Survey. Alaska Fisheries Science Center Processed Report,
Gauch, H.G. Jr. 1982. Multivariate analysis in community ecology. 2001-3. Anchorage, AK. 135 pp.
Cambridge Univ. Press, New York. Matthews, K.R. and L.J. Richards. 1991. Rockfish (Scorpaenidae) as- Turk, T. A., T. Builder, C. W. West, D. J. Kamikawa, J. R. Wallace,
semblages of trawlable and untrawlable habitats off Vancouver Island, R. D. Methot, A. R. Bailey, K. L. Bosley, A. J. Cook, E. L. Fruh, B. H.
Hallacher, L.E. and D.A. Roberts. 1985. Differential utilization of British Columbia. N. Amer. J. Fish. Mgmt., Vol. 11, pp. 312-318. Hormess, K. Piner, H. R. Sanborn, and W. W. Wakefield. 2001. The
space and food by the inshore rockfishes (Sebastes) of Carmel Bay, 1998 Northwest Fisheries Science Center Pacific west coast upper
California. Env. Biol. Fish., Vol. 12(2), pp. 91-110. McGarigal, K., S. Cushman, and S. Stafford. 2000. Multivariate sta- continental slope trawl survey of groundfish resources off Washington,
tistics for wildlife and ecology research. Springer-Verlag, New York. Oregon, and California: Estimates of distribution, abundance, and
Hixon, M.A., B.N. Tissot, and W.G. Pearcy. 1991. Fish assemblages 130 pp. length composition. U.S. Department of Commerce, NOAA Technical
of rocky banks of the Pacific Northwest: final report. Minerals Manage- Memorandum NMFS-NWFSC-50. Seattle, WA. 122 pp.
ment Service, MMS 91-0052. Camarillo, California. 410 pp. Moser, H.G. R. L. Charter, W. Watson, D. A. Ambrose, J. L. Butler, S.
R. Charter, and E. M. Sandknop. 1999. Abundance and distribution of Weinberg, K.L., M.E. Wilkins, F.R. Shaw, and M. Zimmermann.
Hixon, M.A., and B.N. Tissot. 1992. Fish assemblages of rocky banks rockfish (Sebastes) larvae in the Southern California Bight in relation 2002. The 2001 Pacific west coast bottom trawl survey of groundfish
of the Pacific Northwest: supplement. Minerals Management Service to environmental conditions and fishery exploitation. CalCOFI Rep., resources: Estimates of distribution, abundance, and length and age
MMS 92-0025. Camarillo, California. 128 pp. Vol. 41, pp. 132-147. composition. U. S. Dep. Commer., NOAA Tech. Memo. NMFS-AFSC-
128. Seattle, WA. 140pp.
34
Subsection 2.1.2: HABITAT SUITABILITY MODELING
maximum observed across the bathymetric gradient (Table 9)
INTRODUCTION
(Rubec et al., 1999). Resultant values were multiplied by 10
Habitat suitability modeling (HSM) is a tool for predicting the
to scale SI’s by whole integers (0-10), as reclassification of
suitability of habitat for a given species based on known affini-
Map Layers Habitat Suitability Maps
Species Habitat Affinities environmental grids is done using ArcView which does not rec-
ties with environmental parameters. This technique was chosen
ognize decimals. For species that had limited or no trawl data,
for this project to provide a synoptic view of habitat suitability
SI values were developed from bathymetric ranges reported
for specific species as well as assess habitat suitability for 1.0
in the literature (Christensen et al., 1997; Brown et al., 2000).
species assemblages. One HSM technique is termed “habitat S1
0.5 Table 10 displays a sample data matrix generated from literature
suitability index (HSI) modeling”. HSI models are simple math-
=
+
SI value
Bathymetry
0 sources, where presence (1) or absence (0) is coded within the
ematical expressions for calculating a unitless index of habitat
bathymetric classes for each particular species and life stage.
quality as a function of one or more environmental variables. depth category
In this technique, the total number of references that denote
Using GIS, these index values can be mapped and analyzed
presence of the species are summed within each depth class
to portray areas of potential distribution for a species (Brown Substrate
1.0
and then divided by the total number of references examined
S2
et al., 2000) (Figure 30). High-quality habitat may provide high
0.5
to obtain the final SI value. Literature review provided only
carrying capacity and support higher rates of growth, survival,
0 general ranges of species occurrence in relation to bathymetry,
or reproduction for a given species, whereas low-quality or un-
substrate category therefore, classes of 50 m were chosen to confidently develop
suitable habitat may have little or no carrying capacity (Brown et
SI values rather than the 20 m classes used above. Differen-
al., 2000). The HSI methods were adapted from the U.S. Fish
tiating depth ranges for adults and juveniles from literature
and Wildlife Service (USFWS) Habitat Evaluation Procedures 1
n sources was difficult due to lack of data, therefore, only adult
program (USFWS, 1980a, 1980b, 1981) to provide spatially
( ∏ Si )
n
HSI = SI values could be developed using this technique. SI’s for af-
explicit estimates of suitability across the entire study area. It
finities with substrate were also created using this technique.
is important to note that the model results depict potentially
i =1 SI’s for juveniles based on bathymetry were developed using
suitable habitat for a given species and not actual distribution.
NMFS trawl data, when available, or were simply not modeled
This section provides the methodology, results, validation, and Figure 30. Species habitat suitability modeling approach.
where trawl data was limited or absent. Contrasting evidence
interpretation of HSI models developed for selected adult and
exists within the literature that bathymetric preferences can shift
subadult stages of commercially and recreationally important
for many groundfish species based on latitude (PFMC, 1999;
groundfish and invertebrate species. The models are based on etry. Bathymetry was rasterized with 70 m cell size for most HSI Data/SI Development: Initially, suitability index (SI) values
Williams and Ralston, 2002). For the present study, it was as-
species’ affinities to substrate types and bathymetric ranges of the study area for depths to 4810 m. Benthic substrate was for bathymetry and substrate type were developed through
sumed that depth preference was similar regardless of latitude,
(Monaco et al., 1998). mapped from Point Arena in the north to Point Sal in the south literature review and modeled in GIS. During October 2002,
although further exploration into this reported trend is currently
to conform to the latitudinal limits of the study area. Substrates the methodological approach and results were peer-reviewed
underway. Similarly, preference for substrate was assumed to
DATA AND ANALYSES were characterized using 5 classifications: sand, mud, rock, by NMSP and NMFS staff who suggested that, where suf-
be the same throughout each species' range.
Environmental Data: Initially bathymetry, benthic substrate pebble/cobble/gravel, and mud/rock mix. ficient data were available, bathymetry SI values should be
type, and bottom temperature were chosen as the environ- developed using NMFS trawl data. In addition, the panel
requested separate models for adults and juveniles. As a The NMFS trawls were conducted in depths of 50 – 1300 m;
mental data to be included in the models. Although water tem- Species selected for HSM: The primary criteria to select spe-
result, a subset of NMFS trawl data on the shelf (1977-1995) therefore bathymetric SI’s outside this range could not be
perature is an influential factor that affects species distributions cies for which HSMs were developed was their commercial
and slope (1984-1999) for the entire west coast were used to calculated. Depth information within the literature exists for
and movement, several factors led to the exclusion of bottom and ecological importance. In addition, several species were
temperature from final model development: 1) information re- included based on recommendations by staff members from develop SI values for adults and juveniles
garding species associations with bottom temperature were too the NMSP. Overall, 20 species were modeled, 14 of which for most species. SI values for bathym-
general or absent from scientific literature; 2) statistical analyses included models for adult and subadult distribution. Species etry were developed from NMFS trawl
revealed collinearity between bottom temperatures collected with two life stage models include: bocaccio, canary rockfish, data by fitting a polynomial regression 0.12
with NMFS trawl samples and bathymetry; and 3) since most chilipepper rockfish, darkblotched rockfish, longspine thorny- to bathymetric classes and mean spe- 2
mean log abundance = -0.0329 + 0.00122(depth) - 0.0000036(depth)
0.1
of the species modeled are benthic organisms where bottom
Mean log abundance
head, shortspine thornyhead, lingcod, sablefish, Pacific whiting cies abundance (log transformed) (Fig-
0.08
temperature is not highly variable, numerous authors state that (hake), dover sole, english sole, petrale sole, rex sole, and ure 31). Since trawl samples were not
depth is the most significant factor regulating species distribu- Pacific sanddab. Potential adult distributions were modeled for collected in waters less than 50 m, the 0.06
tions (Gabriel and Tyler, 1980; Matthews and Richards, 1991; Dungeness crab, California market squid, blue rockfish, widow bathymetric classes begin at 50 m with 0.04
Yoklavich et al., 2000; Williams and Ralston, 2002). As a result, rockfish, yelloweye rockfish, and yellowtail rockfish. Some of a range of 20 m between classes. The
0.02
water temperature was eliminated as a modeling variable. This these species were chosen to represent species assemblages fitted curve was weighted by sampling
does not preclude using water temperature as a variable for as determined in Section 2.1.1, and mapped to display the effort to account for disproportionate 0
modeling pelagic species, however, more information will have potential distribution of suitable habitats for the assemblage. sample sizes within bathymetric class- 50 70 90 110 130 150 170 190 210 230 250 270 290 310 330 350 370
to be collected to explore their affinities for this variable. Based For example, cluster analyses determined that Dover sole, es. Predicted mean abundance along Depth (m)
on these considerations, bathymetry and substrate data were sablefish, and shortspine thornyhead were commonly captured the curve was then used to calculate
used to map HSI model results. Numerous data sources were in NMFS trawl surveys, indicating a deep water shelf assem- SI’s for each bathymetric class by divid- Figure 31. Polynomial regression curve fit with mean log abundance by categorical bathymetric
ing each mean abundance value by the class for subadult bocaccio.
combined to produce a digital, high resolution map of bathym- blage for these species.
35
Subsection 2.1.2: HABITAT SUITABILITY MODELING
The resulting maps display the potential suitability of cells for each species
Table 9. Example data matrix for calculating bathymetry SI values for subadult bocaccio
taken in NMFS trawl samples (Rubec et al., 1999). based on the strength of their affinities to depth and substrate type in each
cell. The map displays habitat suitability in a unitless index from 0 (unsuitable)
Depth Effort Mean log Predicted mean log HSI
to 10 (highly suitable).
Class (m) (# of samples) abundance abundance (x) (x/xmax)*10
50-69 219 .014 .019 3
Validation of HSI Model: The remaining subsets (i.e. independent data) of NMFS
70-89 361 .029 .035 5
trawl data from the shelf (1998-2001) and slope (2000-2001) were used to as-
90-109 447 .049 .048 7
sess model performance. Mean abundance was calculated for each species
110-129 489 .060 .058 8
130-149 398 .056 .065 9 from these data and superimposed over the predicted HSI values and compared
150-169 252 .100 .069 10 by regressing observed catch data on the predicted HSI values. The statistical
170-189 200 .094 .070 10 results are not intended to be definitive tests of the model, but provide sup-
190-209 213 .065 .069 10 porting evidence for the existence and strength of the relationship between the
210-229 182 .037 .064 9
model predictions and the catch data. It is important to note that these models
230-249 98 .059 .057 8
are based on two independent parameters and are not the definitive predictors
250-269 92 .019 .047 7
of habitat utilization for these species. Fishery-dependent data from CDF&G
270-289 89 .003 .034 5
recreational surveys were also used to validate models for species that had
290-309 74 .008 .018 3
limited trawl information, i.e. species that display affinities for rocky substrates
310-329 98 .003 0 0
330-349 52 0 0 0 (rockfishes) and had poor representation in trawl data. If the model performs
correctly, this validation procedure should demonstrate increasing mean abun-
dance with increasing habitat suitability.
Table 10. Example presence/absence information and SI calculation from scientific
literature.
Integrative Maps: Management plans are often developed for a group of spe-
Depth Class (m)
cies that exhibit similar life history strategies. Selected species assemblages,
100- 150- 200- 250- 300- 350- 400-
as defined in Section 2.1.1, were analyzed and mapped to identify the spatial
Source 50-99 149 199 249 299 349 399 449
A 1 1 0 0 0 0 0 0 distribution of their important habitats. In addition, two analyses were conducted
B 1 1 1 0 0 0 0 0 (page 42) which examine the overlap of highly suitable habitat based on all spe-
C 1 1 1 1 1 0 0 0 cies for which HSI maps were developed. Areas with the most overlap of high
D 1 1 1 1 1 0 0 0
suitability could be considered important habitats for selected groundfish.
Sum 4 4 3 2 2 0 0 0
SI=Sum/Total
ANALYTICAL MAP PRODUCTS
References*10 10 10 7 5 5 0 0 0
As part of the biogeographic assessment, digital data were developed as prod-
ucts from the study. Digital bathymetry and substrate maps were created as
most species outside of this range, but was omitted from modeling and mapping to
ArcView shape and raster files. Maps of these environmental data can be seen
match the depth range associated with the NMFS trawls. Trawls were conducted
on pages 36-37, while digital files are located on the accompanying CD-ROM.
during June to November, thus models for different “seasons” could not be created.
Three representative HSI models are presented: bocaccio (adult and subadult),
Therefore, modeled map surfaces represent species potential distributions for water
Dover sole (adult and subadult), and Dungeness crab (adult). The remaining 31
depths from 50-1300 meters and for the summer and late fall time period. Also,
species' HSI maps are on the CD-ROM. Representative maps displaying habitat
many of the species modeled exhibit inshore/offshore migrations based on habitat
importance based on all HSI models and select species assemblages are also
shifts associated with life history requirements and/or spawning activity. Additional
included. Additional integrative maps for shelf assemblages, all rockfish, and
data will have to be collected to reflect these shifts in abundance and distribution;
all flatfish, are also included on the CD-ROM.
thusly, no attempt was made to model them here.
HSI Results-Mapping: Once SI values were determined for bathymetry and sub-
strate type, these values were inserted into the environmental grids. Once each
species’ suitability indices were derived (either through regression or through the
literature), the values were combined with the bathymetry and substrate map layers
to calculate an index of habitat suitability. The habitat suitability was calculated as
the geometric mean of suitability indices (SI) for the two (n) environmental factors
(Rubec et al. (1999) (Figure 30):
1
n
HSI = (∏ S i ) n
i =1
36
Subsection 2.1.2: HABITAT SUITABILITY MODELING
ABOUT THIS MAP
124°W 123°W 122°W 121°W
Figure 32 displays a bathymetric model of the north/central California study area. Prominent bottom features, such as canyons,
High Resolution
39°N
39°N
seamounts, banks, and other large scale geological formations, are evident at the scale presented.
300
20
50
DATA SOURCES
00
m
0
10
20
Bathymetry
m
m
0
00
NOAA/NOS hydrographic survey data available from the National Geodetic Data Center (NGDC) and Monterey Bay Research
m
m
Institute (MBARI) – multibeam data.
METHODS
Results were calculated from 3 arc second bathymetry (nominally 70 m) derived from NGDC and MBARI data sources. All avail-
able multibeam data were used. Hydrographic survey data (echo soundings) were eliminated from the calculation if it occurred
coincidentally with multibeam information. Vertical and horizontal correction was performed on all data prior to modeling. All
0 10 20 40 60 80 100
data were triangulated and rasterized using Vertical Mapper extension of MapInfo 6.5. Cell size varied throughout the study
38°N
38°N
Kilometers
area, but significant portions were mapped at 70 m2 grid cell resolution.
RESULTS AND DISCUSSION
The study area contains two distinct bathymetric regions. The northern portion of the study area, from Monterey Canyon north-
ward, is characterized by having a broad continental shelf (15-50 km wide), while the southern region has a very narrow shelf
with rapidly increasing depth close to shore. This pattern results in significantly shallower mean depth for Cordell Bank (394.9
m) and Gulf of the Farallones (265.3 m) sanctuaries compared to the mean depth within Monterey’s sanctuary (876.9 m). For
a more detailed description of bathymetric features see the explanation of the study area on page 1.
This map is not intended for navigational purposes. Some areas on this map are created from old or sparse data, and are not
37°N
37°N
necessarily representative of the actual seafloor characteristics.
36°N
36°N
10
20
30
00
00
00
m
200 m
m
50 m
m
35°N
35°N
124°W 123°W 122°W 121°W
Figure 32. Bathymetric map for the north/central California study area. Red lines indicate National Marine Sanctuary
boundaries.
37
Subsection 2.1.2: HABITAT SUITABILITY MODELING
ABOUT THIS MAP mud/rock mix. Within sanctuary boundaries, rocky substrates
124°W 123°W 122°W 121°W
Figure 33 displays distribution of substrate types throughout are distributed predominantly on the shelf, occurring in the ar-
39°N
39°N
the study area, from Point Arena to Point Sal California. The eas near Cordell Bank, Farallone Islands, in many near-shore
Substrate Types substrate is classified into 5 categories: mud, sand, pebbles/ areas, and scattered within Monterey Bay. Outside sanctuary
cobbles/gravel (pcg), rock, and mud/rock mix. boundaries, several large areas of rock are found on the slope
in depths greater than 1200 m. Mixed rock/mud substrate is
DATA SOURCES scarce within the study area, with most occurring southwest of
California Continental Margin Geologic Map Series (Maps 4- Monterey canyon. One large area of mixed rock/mud is present
Legend 6) (Greene and Kennedy, 1989). These maps were originally southwest of the southern Monterey Sanctuary boundary. Areas
created with a 1:250,000 resolution and were used as the containing pebble, cobble, and gravel are found exclusively in
Mud
basemaps for recent revisions and incorporation of new high Monterey’s sanctuary and are generally found within depths of
Sand resolution multibeam data in small portions of the study area. 100-200 m. The majority of sand substrate is found near-shore
Pebbles/Cobble/Gravel in the northernmost and southernmost portions of the study
38°N
38°N
METHODS area, with significant coverage also occurring around Cordell
Rock
Initially, seven maps were developed that displayed substrate Bank. Figure 33 displays the percent coverage of bottom types
Rock/Mud
type and geologic formations throughout California’s coastal within each sanctuary. Other important substrate types exist
and marine environments. The original data were compiled by within the study area, such as near-shore kelp beds, but were
the California Division of Mines and Geology, USGS, and Cali- not included in the map based on their ephemeral distribution
0 10 20 40 60 80 100
fornia Coastal Commission to produce paper maps. Geologists and the limitations associated with development of bathymetric
Kilometers
from California State University-Monterey Bay digitized these SI’s in near-shore areas.
maps in 1999 and further interpreted these data to develop
boundaries of substrate types (Greene et al., 1999). Three of Although the substrate map is a probabilistic map of substrate
the seven maps provide data for the study area and together types, it reflects the most complete and current knowledge of
37°N
37°N
provide the most comprehensive map of substrate type for the benthic substrates for the north/central California region. The
north/central California marine and near-shore region. For a map alone can be used to support investigations that require
detailed description of the development of the original maps advanced knowledge of sea floor type. In addition, the maps
and classification scheme, refer to Greene et al., 1999 and are useful identifications of critical or important habitats for par-
Greene et al., 2002. Eight substrate types were classified, how- ticular species or for determining essential fish habitat that can
ever, some were grouped together when the digital substrate aid conservation and management plans for fisheries species.
shapefile was rasterized into 1 km2 grid cells to facilitate their In this study, the map was primarily used as an environmental
use in the HSI model analyses (See Section 2.1.2). “Boulders” layer in GIS (in addition to bathymetry) to determine habitat
and “Hard/Anthropogenic” polygons were grouped within the suitability for groundfish species. This approach assumes that
“Rock” substrate type, and “Gravel" was grouped with “Cobbles/ the underlying environmental GIS layers are an accurate rep-
36°N
36°N
Pebbles” because boulder and gravel polygons were limited in resentation of that particular variable, thus the model results
number and affinities to these were considered to be similar to are only as good as the underlying digital information. The
the group it was reclassified with. substrate map is conservative, based on its original scale (1:
250,000), and may have fine scale inaccuracies throughout
RESULTS AND DISCUSSION the study area. The majority of this map has not been field
The substrate map covers an area of approximately 44,000 tested; therefore, inaccuracies in classification may exist. For
Gulf of the Farallones
Cordell Bank Monterey Bay km2, of which mud accounts for 86.4% (38,023 km2) of the example, several small polygons classified as rock near Point
total bottom area. Substrate containing pebble/cobble/gravel Reyes have been questioned. HSM results presented herein
were the least abundant substrate types, only encompass- do not contain these polygons due to depth limitations with the
6% 1% 4%
4% 7%
15% 19%
ing 100 km2 of the study area. Rock substrates were mostly fish and invertebrate catch data (50-1300 m). Small localized
35°N
35°N
patchy throughout the region, encompassing 1,561 km2. Mud areas of high resolution information (on the scale of 10’s of
and rock mixed substrate (1,706 km2) was almost exclusively meters) have been included in this map, however, these areas
81% 89%
74% distributed in the southern portion of the study area, with small comprise a small percentage of the overall study area. More
localized areas near Monterey canyon. Sand (2,611 km2) is information is required to test the accuracy of the map; hence,
predominantly located near-shore with a large area located thematic accuracy of substrate types is unknown.
rock pcg mud sand
near and around Cordell Bank. Within sanctuary boundaries,
Abbreviation: pcg= pebble/cobble/gravel
soft sediments (mud, sand) account for almost 95% of the
124°W 123°W 122°W 121°W
substrate. Rocky areas (including pebbles, cobbles, gravel)
Figure 33. Substrate types for the north/central California marine region. account for approximately 5%, while less than 1% consists of
38
Subsection 2.1.2: HABITAT SUITABILITY MODELING
ABOUT THESE MAPS
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Figure 34 displays HSI model results for adult (left) and subadult (right) bocaccio during June-
November. The maps exhibit the potential distribution of the species based on affinities to ba-
Bocaccio Bocaccio
39°N
39°N
39°N
thymetry and substrate. Predicted HSI values range in scale from 10 (highest) to 0 (unsuitable)
and were grouped into five classes: highest suitability (10-8), moderate (7-5), low (4-2), lowest
Subadult
Adult
(1), and unsuitable (0). SI values for bathymetry and substrate type are shown in the graphics
below the mapped HSI results. Model performance graphics and statistical details are displayed
in the map insets.
DATA SOURCES
HSI Results HSI Results
38°N
38°N
38°N
Bathymetry SI: Alverson et al., 1964; Feder et al., 1974; Dark et al., 1983; Gunderson and
Highest
Highest Sample, 1980; Tagart and Kimura, 1982; Eschmeyer et al., 1983; Allen and Smith, 1988; Love
Moderate Moderate et al., 1990; Wolotira et al., 1993; Wilkins et al., 1998; Yoklavich et al., 2000; Lauth, 2001; and
Love et al., 2002.
Low Low
Substrate SI: Feder et al., 1974; Eldridge, 1994; Yoklavich et al., 2000; and Love et al., 2002.
Lowest Lowest
Validation: Wilson-Vandenberg et al., 1996; Wilkins et al., 1998; and Turk et al., 2001.
Unsuitable Unsuitable
Life stage information: Love et al., 2002.
37°N
37°N
37°N
0 25 50 Km 0 25 50 Km
METHODS
Bathymetry SI values for adult bocaccio were developed using the literature review method,
whereas subadult SI values were assigned based on the regression fitting technique using NMFS
trawl data.
RESULTS AND DISCUSSION
36°N
36°N
36°N
Length at maturity information (Love et al., 2002) was used to determine life stage for bocaccio.
Adults were defined as: females >360 mm and males >350 mm total length. Depth suitability for
Validation - CDFG Rec. Data Validation - CDFG Rec. Data subadults was highest from 90-270 m, while highest suitability for adults was similar, ranging from
1
1.2
50-299 m (Figure 34). Literature sources indicate that adult bocaccio are almost exclusively found
r2=0.90
2
r =0.70
p=0.0001
p=0.0046
around rocky substrates, while subadults exhibit broader affinity among substrate types (Figure
1 0.8
Mean Abundance
Mean Abundance
34). Comparison of the two HSM maps show that the marked difference in substrate preference
0.8
0.6
for adults yields a more limited spatial distribution than subadults. Less than 5% of the available
0.6
35°N
35°N
35°N
habitat within each sanctuary was predicted highly suitable (HSI values >8) for adult bocaccio
0.4
0.4
(Cordell Bank – 4.6%, Gulf of Farallones – 2.9%, Monterey – 1.7%) Within the study area, habitat
0.2
of high suitability occurs exclusively inside sanctuary boundaries. High suitability covers more
0.2
area for subadults than adults and extends well beyond sanctuary boundaries. Nearly 10% of
0
0
0 2 4 6 8 10
Cordell Bank’s sanctuary was considered highly suitable for subadults. This percentage drops to
0 2 4 6 8 10
HSI Value HSI Value
1.6% for Gulf of Farallones, and 2.6% for Monterey. Approximately 556 km2 of potential high suit-
able habitat was located within the three sanctuaries, while an additional 355 km2 were predicted
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
outside sanctuary boundaries. Although the proportion of highly suitable habitats were similar for
Sediment
Bathymetry
Bathymetry Sediment
adults and subadults, large areas of potentially moderate suitability for subadults were observed
10 10
throughout the study area; whereas no areas were predicted moderate for adults.
8
8
Generally, subadult bocaccio are more commonly found in shallower waters than adults (Love
6 6
SI Values
et al., 2002). Current scientific literature does not provide enough information to develop depth
SI Values
SI values for subadults; therefore, limited trawl information was used to develop SI values for
4 4
bathymetry. Despite this, model performance for subadults yielded a strong positive correlation
2 2
between observed abundance estimates from CDFG recreational catch data and predicted suit-
ability (see map inset). Model performance for adult bocaccio also exhibited a strong positive
0 0
Sand Mud Rock
0 - 49
0-49
50-69
70-89
90-109
Pebble
50 - 299
300 - 399
400 - 599
600 - 1300
110-129
130-149
150-209
210-229
230-249
250-269
270-289
290-309
310-1300
Sand Mud Rock
Pebble
correlation between predicted suitability and CDFG catch data. More information regarding bocac-
Cobble
Cobble
Gravel
Gravel
cio life history requirements are necessary to strengthen the HSI models; however, the mapped
results and validation based on currently available information provide an adequate delineation
Figure 34. Potential distribution of habitat suitability for adult and subadult bocaccio. Map inset contains validation statistics. SI values for bathymetry and substrate are of potential habitat suitability for adult and subadult bocaccio.
graphically displayed below the map.
39
Subsection 2.1.2: HABITAT SUITABILITY MODELING
ABOUT THESE MAPS
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Figure 35 displays the HSI model results for adult (left) and subadult (right) Dover sole during
June-November. The maps exhibit the potential distribution based on affinities to bathymetry
39°N
39°N
39°N
39°N
Dover sole Dover sole and substrate. Predicted HSI values range in scale from 10 (highest) to 0 (unsuitable) and
were grouped into five classes: highest suitability (10-8), moderate (7-5), low (4-2), lowest
(1), and unsuitable (0). SI values for bathymetry and substrate type are shown in the graph-
Adult Subadult ics below the mapped HSI results. Model performance graphics and statistical details are
displayed in the map insets.
DATA SOURCES
HSI Results HSI Results
38°N
38°N
38°N
Bathymetry SI: Wilkins et al., 1998 and Lauth, 2001.
Highest Highest Substrate SI: Demory, 1975; Demory et al., 1976; Barss et al., 1977; Pearcy, 1978; NOAA,
Moderate Moderate 1990; Stein et al., 1992; and CDFG, 2002.
Validation: Wilkins et al., 1998 and Turk et al., 2001.
Low Low
Life stage information: PFMC, 1999.
Lowest Lowest
Unsuitable Unsuitable
METHODS
37°N
37°N
37°N
Bathymetry SI values for adults and subadults were developed from the regression fitting
0 25 50 Km 0 25 50 Km technique. Substrate SI values were developed through literature review.
RESULTS AND DISCUSSION
Adult Dover sole are reported to be >300 mm total length for male and female individuals
(PFMC, 1999). Both adult and subadult Dover sole inhabit deep water slope habitats; sub-
adults exhibited a shallower range of depth preference (130-650 m) than adults (290-1070
m) (Figure 35). Adults and subadults prefer soft sediments (sand and mud) throughout their
36°N
36°N
36°N
range. Highest habitat suitability for subadults was predicted to occur along the shallower
portions of the continental slope (200-550 m). A large area of moderate suitability was also
Validation - NMFS Trawl Data Validation - NMFS Trawl Data
1.5
predicted for an area that extends throughout the majority of the continental shelf. The most
2
r2=0.79
2
r =0.90
suitable habitats for adults consisted of deeper slope waters, with only moderate suitability
p=0.0005
Mean Abundance
Mean Abundance
1.2
p=0.0001
1.5
extending onto the shelf region. Within Cordell Bank sanctuary, high subadult suitability (val-
0.9
ues 8-10) was calculated for 22% of the available habitat, 6.4% within Gulf of Farallones, and
1
19% within Monterey sanctuaries. Cordell Bank and Gulf of the Farallone sanctuaries are
35°N
35°N
35°N
0.6
comprised of shallower (50-300m) shelf waters, thus the percentage of highly suitable habitat
0.5 0.3
for adults is lower (based on their calculated affinity for deeper waters) than that observed
for subadults (21% and 12%, respectively). However, Monterey’s sanctuary is considerably
0
0
deeper and a larger proportion of available habitats (30%) were predicted to be highly suitable
0 2 4 6 8 10
0 2 4 6 8 10
HSI Value HSI Value for adults. Approximately 50% of areas that were predicted to be potentially suitable habitats
for both adults and subadults occurred outside of sanctuary boundaries. These areas are
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
most prominent south of Monterey’s sanctuary.
Sediment
Bathymetry
Bathymetry Sediment
10 10
Model performance was assessed by regressing predicted HSI values on mean log abundance
values from NMFS trawl samples (1998-2001). Significant positive correlations were observed
8
8
for both adult and subadult models, however, these are based on limited trawl samples (N =
6 6
311). Discrepancies in model performance, such as small peaks of mean abundance within
SI Values
SI Values
low suitability areas, are a result of limited observations within that category. Additional trawl
4 4
information would strengthen model development and performance.
2 2
0 0
0-49
50-89
90-129
0-49
Pebble
50-69
70-109
130-169
170-209
210-239
290-489
490-549
550-609
610-649
650-669
670-709
710-729
730-769
770-789
790-809
810-1300
110-149
150-189
190-249
250-289
290-349
350-429
430-529
530-829
830-929
930-1009
1010-1069
1070-1109
1110-1149
1150-1209
1210-1249
1250-1300
Sand Mud Rock
Pebble Sand Mud Rock
Cobble
Cobble
Gravel
Gravel
Figure 35. Potential distribution of habitat suitability for adult and subadult Dover sole. Map inset contains validation statistics. SI values for bathymetry and
substrate are displayed below the maps.
40
Subsection 2.1.2: HABITAT SUITABILITY MODELING
ABOUT THIS MAP
124°W 123°W 122°W 121°W
This map displays HSI model results for adult Dungeness crab during June-November (Figure 36). The map displays the potential distribution based on affinities to ba-
thymetry and substrate. Predicted HSI values range from 10 (highest) to 0 (unsuitable) and were grouped into five classes: highest suitability (10-8), moderate (7-5), low
Dungeness crab
39°N
39°N
(4-2), lowest (1), and unsuitable (0). SI values for bathymetry and substrate type are shown in the graphics below the mapped HSI results. Model performance graphics
and statistical details are displayed in the map insets.
Adult DATA SOURCES
Bathymetry SI: Wilkins et al., 1998 and Lauth, 2001.
Substrate SI: Pauley et al., 1989; Emmett et al., 1991; Leet et al., 2001; and CDFG, 2002.
Validation: Wilkins et al., 1998 and Turk et al., 2001.
HSI Results
38°N
38°N
Highest METHODS
Moderate Bathymetry SI values for adult Dungeness crab were developed using the regression fitting technique. Substrate SI values were developed through literature review.
Low
RESULTS AND DISCUSSION
Lowest
Only adults were modeled within the study area because size information was lacking for crabs in the NMFS trawl data and scientific literature was not detailed enough
Unsuitable
to develop SI values for subadults. Dungeness crabs are an estuarine dependent species (Pauley et al., 1989), with adults exhibiting a shallow distribution (to 90 m) in
37°N
37°N
coastal marine waters. Depth SI values derived from NMFS trawls confirmed this trend by exhibiting high SI values within 50-90 m. Suitability is probably high in the
0 25 50 Km
shallower near-shore environment (Emmett et al., 1991); however, trawl information was not available for this area. Literature sources described crab substrate prefer-
ence to be soft sediments, with occasional utilization of rocky substrate. Habitat suitability based on these data resulted in a broad area of high suitability throughout the
shallower waters of the Gulf of Farallones sanctuary (38% of available habitat), and much smaller proportions within Cordell (8.7%) and Monterey (10.4%) sanctuaries.
Overall, this amounts to 2,809 km2 of highly suitable habitat within the three sanctuaries. Moderate suitability, encompassing approximately 2,477.8 km2, extends further
offshore to approximately 130 m. The potential suitability of habitats rapidly declines to unsuitable beyond 130 m in depth. The model performed well with NMFS valida-
tion data and exhibited a strong positive correlation with predicted suitability values.
36°N
36°N
Validation - NMFS Trawl Data
0.5
r2=0.89
p=0.0014
Mean Abundance
0.4
0.3
35°N
35°N
0.2
0.1
0
0 2 4 6 8 10
HSI Value
124°W 123°W 122°W 121°W
Bathymetry Sediment
10
8
6
SI Values
4
2
0
0-49
50-69
70-89
90-109
110-129
130-149
150-1300
Sand Mud Rock
Pebble
Cobble
Gravel
Figure 36. Potential distribution of habitat suitability for adult dungeness crab. Map inset contains vali-
dation statistics. SI values for bathymetry and substrate are graphically displayed below the map.
41
Subsection 2.1.2: HABITAT SUITABILITY MODELING
or other (0). All maps were overlain and summed to create a map of suitability overlap
within the study area. These areas represent potential groundfish hot spots.
124°W 123°W 122°W 121°W
124°W 123°W 122°W 121°W
RESULTS AND DISCUSSION
HSI Model
Mean HSI Model
39°N
39°N
39°N
The techniques described above are two possible approaches to estimate potential
hot spots or areas of habitat importance. Composite maps displaying these areas were
Species Overlap developed using all fish HSI model results to simulate the groundfish management
strategy employed by NMFS, where all groundfish (83 species) are managed under
HSI Results one Fishery Management Plan. Mean HSI values across all 32 fish species and
life stages yield no areas ranked as highly suitable (HSI values 10-8). Moderate
Highest
# of species suitability (7-5) occurs over the majority of the shelf region (to approximately 200
Moderate
38°N
38°N
38°N
m) throughout the study area, most notably in the northern portion, where the shelf
15 - 20
Low extends significantly farther offshore than in the southern portion. The majority of the
12 - 14 area north of Monterey canyon consists of moderate suitability (to approximately 200
Lowest
8 - 11 m), with low suitability extending through the deeper slope habitat. Smaller localized
Unsuitable
areas of low suitability exist within the shelf and represent areas of hard substrate.
1-7
South of the Monterey canyon, low suitability comprises most of the study area, with
0
0 25 50 Km
a narrow zone of moderate suitability along the shallower shelf waters. Suitability
drops from moderate to low just beyond the shelf edge throughout the study area.
37°N
37°N
37°N
0 25 50 Km
Throughout the study area, maximum overlap of cumulative high suitability occurs
on the shelf edge over soft sediments, which closely contour the 100 m isobath.
Approximately half of the models overlap in this zone. The top two quintals encompass
most of the shelf region and the zones of overlap are much broader in the northern
portion of the study area compared to the southern.
36°N
36°N
36°N
These analyses reveal patterns of suitability related to depth and substrate. Highest
suitability occurs on the continental shelf, over soft sediments, based on the two
analytical approaches using the 33 HSI maps. These areas could be considered
as habitats of importance that support fish abundance and diversity. Both methods
portray highest suitability over the shelf that decline beyond the shelf edge. This
pattern conforms to literature sources which state that the shelf, and more importantly
the shelf break, are important areas for fish abundance and diversity (Yoklavich et
35°N
35°N
35°N
al., 2000; Williams and Ralston, 2002). In addition, soft sediments are potentially
more suitable than hard bottom throughout the study area. It is important to note that
the results of these analyses are based on 19 species and are only a subset of the
many groundfish species that occur within the study area. These results are clearly
biased based on the species modeled and may not provide adequate representation
of groundfish as a whole within the study area. Most of the species modeled have
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
substrate affinities for soft sediments, and most exhibit depth preferences that fall
Figure 37. Areas of groundfish potential hot spots based on mean fish species HSI models and overlap of predicted highly suitable habitats. within the shelf region. Ideally, many more models should be developed for additional
species and analyzed to provide a more representative depiction of groundfish
distribution within the study area.
ABOUT THESE MAPS METHODS
These maps display the results of two approaches that provide a synoptic view of Mean HSI: HSI maps for all fish species and life stages were overlain and averaged
overall habitat importance based on all the HSI models developed for the study area. by grid cell to evaluate overall suitability. Results were scaled in the same manner
Areas of potential habitat importance were first defined as an average view of habitat as individual HSI model results: Highest suitability (10-8), moderate (7-5), low (4-2),
suitability across species and life stages. Secondly, individual maps of highly suitable lowest (1) and unsuitable (0).
habitats were overlain and areas or regions with the most overlap were considered
important habitat or hot spots (Figure 37). Cumulative Suitability: Frequency of occurrence of predicted HSI values for each
fish and invertebrate species life stage were calculated and values greater than one
DATA SOURCES standard deviation above the mean were chosen to represent highest suitability. New
See Individual HSI model results – CD-ROM. individual maps were created and grid cells were reclassified as highest suitability (1)
42
Subsection 2.1.2: HABITAT SUITABILITY MODELING
ABOUT THESE MAPS
124°W 123°W 122°W 121°W
124°W 123°W 122°W 121°W
The maps provide one approach to assess habitat suitability based on HSI results for multiple
species (Figure 38). HSI model results were averaged to assess the potential distribution of
All Rockfish Slope Assemblage
suitable habitats for 8 species of adult rockfish (left) and 3 adult slope species (right). Predicted
39°N
39°N
39°N
HSI values range in scale from 10 (highest) to 0 (unsuitable). HSI results were grouped into five
classes: highest suitability (10-8), moderate (7-5), low (4-2), lowest (1), and unsuitable (0).
Adult Adult DATA SOURCES
Adult rockfish map – HSI maps for adult bocaccio, chilipepper, darkblotched, canary, yellowtail,
yelloweye, and widow rockfishes (CD-ROM).
HSI Results HSI Results Slope assemblage adults – HSI maps for adult Dover sole, sablefish, and shortspine thorny-
38°N
38°N
38°N
Highest
Highest head (CD-ROM).
Moderate
Moderate
METHODS
Low
Low
The slope assemblage was determined through cluster analysis of NMFS benthic slope trawl
Lowest
Lowest
data (see Section 2.1.1 for methodology). All models for adult rockfish were combined to evalu-
Unsuitable Unsuitable ate habitat suitability for these species as an assemblage. Both assemblages of fishes were
analyzed by overlaying each individual HSI map and calculating the arithmetic mean across
37°N
37°N
37°N
grid cells.
0 25 50 Km 0 25 50 Km
RESULTS AND DISCUSSION
Typically, management plans are not based on single species, but rather groups of species that
exhibit similar life histories (Williams and Ralston, 2002). Estimating potential distributions for
species assemblages from HSI models could be a valuable tool for resource managers to aid
in the development of fishery management plans and conservation strategies. This approach
36°N
36°N
36°N
provides a spatial view of important habitats for a given assemblage and generates a baseline
set of data which can be used for a variety of management needs.
Individual HSI results for the 8 species of rockfishes displayed similar patterns of habitat suit-
ability within the study area and, not surprisingly, the map of mean habitat suitability for these
species is nearly identical to the individual maps. Hard substrates (pebble, cobble, gravel, rocky)
within the shelf region promote highest suitability areas for these species. Moderate suitability
35°N
35°N
35°N
was predicted for areas with mixed mud/rock substrate and mud areas in waters with depths
between 200-450 m. These areas are emphasized based on HSI results from darkblotched
rockfish, which exhibited strong affinity for hard and soft substrates, rather than only rocky
substrate preference exhibited by the other rockfish species. Also, darkblotched rockfish distri-
bution occurs in deeper waters compared to the other species of rockfish and may necessitate
their omission from this assemblage. Regardless, suitable habitat for this group of species is
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
limited, based on the distribution of rocky substrate within the study area. Overall, highly suit-
Species: Species: able habitat comprises 364 km2 within the three sanctuary boundaries or 2% of the available
Species: Species:
Blue rockfish habitat. Moderate suitability comprises even less area, 247 km2, or 1.3% of available habitat.
Dover sole
Blue rockfish Dover sole
Bocaccio
Sablefish
Bocaccio Sablefish Cluster analysis of NMFS trawl data revealed many assemblages of species that tend to occur
Chilipepper rockfish
Shortspine thornyhead together (see Section 2.1.1). Dover sole, sablefish, and shortspine thornyhead were identified
Chilipepper rockfish Shortspine thornyhead
Darkblotched rockfish as members of a strong species assemblage that occurs over soft sediments in deep waters of
Canary rockfish rockfish
Darkblotched
the continental shelf and slope. Mean HSI calculations resulted in a broad range of highly suit-
Yellowtail rockfish
Canary rockfish able habitats throughout the study area. Overall, 23% (4,257 km2) of the available habitat within
Yellowtail rockfish
Yelloweye rockfish the sanctuaries was predicted to be highly suitable habitat for this assemblage. An additional
Yelloweye rockfish
Widow rockfish 10% (1,907 km2) was predicted moderately suitable habitat. Because of its greater depths and
larger area, Monterey Bay sanctuary contained significantly more highly suitable habitat (3,773
Widow rockfish
km2) than Cordell Bank (170 km2) and Gulf of the Farallones (312 km2) sanctuaries.
Figure 38. Areas of potential habitat importance based on mean HSI models for selected species assemblages.
43
Subsection 2.1.2: HABITAT SUITABILITY MODELING
distribution), and others can be developed and
SECTION SUMMARY
incorporated into the model as needed.
HSI modeling and mapping were considered to be a compo-
nent of the biogeographic assessment because this approach
Decision making processes are typically not ad-
provides spatial species- and lifestage-specific information for
dressed at the species level but rather at a multi-
the north/central California marine region. This approach is in-
species assemblage level. Thirty-four HSI models
tended to serve as an analytical tool for resource managers that
were created for 18 fish and 2 macro-invertebrate
can address a variety of needs: 1) developing maps in poorly
species to support multi-species analyses or as-
sampled areas, 2) evaluating impact scenarios, 3) identifying
sessments. Several techniques were conducted
habitats or areas for conservation or protection, and 4) assess-
to assess habitat quality within the entire study
ing impacts of environmental change. The approach used here
area. One result indicated that the most potentially
is similar to previous efforts that mapped near-shore rockfish
suitable habitat occurred on the shelf over mud
distributions (Wright et al., 2000). The maps displayed near-
substrates within depths of 100-120 m. As previ-
shore rockfish distributions in relation to latitude, and maximum
ously mentioned, these results were biased based
and common bathymetric ranges, based on information from
on the selection of species modeled; however,
peer-reviewed literature. The products generated from this
the technique provides one method to identify
study expand on this approach by including an additional pa-
areas of potential high habitat quality. Additional
rameter (substrate type). Also, models were developed which
analyses identified important habitats for select
predicted the potential spatial distribution (based on affinities
species assemblages. Habitat suitability models
for bathymetric ranges and substrate type) for a select group
for an assemblage of rockfish were developed and
of groundfish species. The maps provide a unique spatial view
indicated that rocky habitats located on the shelf
of potential groundfish habitats within and outside central Cali-
were identified as potential hot spots for adults;
fornia sanctuary boundaries.
whereas, mud and sand substrates on the shelf
were delineated as potentially important habitats
It is important to note that the model results previously de-
for subadult rockfish.
scribed are not actual, but potential distributions based on
species affinities to the environmental variables used in the
In conclusion, the HSI maps can be used in a
models. Interpretation of these results should be conducted
broad range of assessments which require in-
carefully due to the variety of limitations associated with the
formation on habitat distribution and suitability.
biological and environmental data. Both bathymetry and sub-
Individual species maps can be used to identify
strate-type maps were created from the most current informa-
areas of varying habitat suitability and can be used
tion available; however, the scale of information may cause
to assess sensitivity to environmental or anthro-
inaccuracies in the interpretation of the model results. The
pogenic impacts. Lastly, it is recommended to
bathymetry map, created with 78 million data points, provides
continue developing HSI models for remaining
a high quality, high resolution image of depth throughout the
groundfish species, as those presented here are
study area. The digital substrate map is a probabilistic inter-
REVIEWERS Chris Harvey, Northwest Fisheries Science Center, NMFS
only a small subset of the available resources
pretation of imagery data that has yet to be field validated.
Tara Anderson, University of California, Santa Cruz Ruth Howell, Gulf of the Farallones National Marine Sanctuary
within the study area.
Given its original resolution (1:250,000), the map may under
Carol Bernthal, Olympic Coast National Marine Sanctuary Program
or overestimate substrate distribution within the study area;
Program Roxanne Jordan, Alliance of Communities for Sustainable
REVIEWS
however, localized areas have more accurate information.
Tonya Builder, Northwest Fisheries Science Center, NMFS Fisheries
Two reviews were completed for the fish assemblage and
For example, predicted areas of potential high suitability were
Gregor Cailliet, Moss Landing Marine Laboratory Chad King, Monterey Bay National Marine Sanctuary Pro-
habitat suitability analyses. During May and June 2002 in-
extremely limited for species that exhibit strong affinities for
Mark Carr, University of California, Santa Cruz gram
formal meetings were held in Monterey, San Francisco, and
rock substrate (rockfish, lingcod). Generally, less than 1% of
Josh Churchman, Fisherman Howatt King, California Department of Fish and Game
Seattle to receive feedback on the approach and verify from
the study area was considered optimal habitat for these spe-
Elizabeth Clarke, Northwest Fisheries Science Center, Bob Lauth, Alaska Fisheries Science Center, NMFS
the scientists that collected the data that the analyses were
cies and may underestimate actual habitat distribution. These
NMFS Phil Levin, Northwest Fisheries Science Center, NMFS
valid. Formal review workshops were held in October, 2002 in
results are reflective of the low percentage of rock substrate
Steve Copps, Northwest Region, NMFS Steve Lonhart, Monterey Bay National Marine Sanctuary
San Francisco, Seattle, and Monterey Bay and hosted local
included in the substrate map, which could be a result of the
Brad Damitz, Monterey Bay National Marine Sanctuary Pro- Program
scientists, fishermen, and National Marine Sanctuary Program
scale in which the original data were collected. Nevertheless,
gram David Lott, Monterey Bay National Marine Sanctuary Pro-
staff. Review comments were either incorporated or addressed
the map provides the most comprehensive substrate inventory
Kathey Fosmark, Alliance of Communities for Sustainable gram
in this product. We appreciate all the reviewers’ time and effort
for this region and it is recommended that additional substrate
Fisheries Huff McGonigal, Monterey Bay National Marine Sanctuary
when providing us this important feedback.
information be collected to further refine the maps. Additional
Jim Glock, Sustainable Fisheries Division, NMFS Program
digital data are available (e.g. sea surface temperature, kelp
Gary Greene, Moss Landing Marine Laboratory Nazila Merati, Pacific Marine Environmental Laboratory,
NOAA
44
Subsection 2.1.2: HABITAT SUITABILITY MODELING
Richard Methot, Northwest Fisheries Science Center, NMFS Feder, H.M., C.H. Turner, and C. Limbaugh. 1974. Observations on fishes associ- NOAA/NOS. 1990. West Coast of North America strategic assessment: Data atlas.
Jim Nybakken, Moss Landing Marine Laboratory ated with kelp beds in southern California. Cal. Dept. Fish and Game, Fish. Bull. Invertebrate and fish volume. Pre-publication edition. Strategic Assessment Branch,
John Pearse, University of California, Santa Cruz Vol. 160, pp. 2-135. NOAA/NOS, Rockville, MD. 112 pp.
Stephen Ralston, Southwest Fisheries Science Center, NMFS
Fishbase. 2002. Species summaries (online). http://www.fishbase.org. Phillips, J.B. 1964. Life history studies on ten species of rockfish (genus Sebastes).
Paul Reilly, California Department of Fish and Game
Cal. Dept. Fish and Game, Fish Bull., Vol. 126, p. 70.
Dale Roberts, Cordell Bank National Marine Sanctuary Program
Fitch, J.E., and R.J. Lavenberg. 1971. Marine Food and Game Fishes of California.
Teresa Turk, Northwest Fisheries Science Center, NMFS
California Natural History Guides, Vol. 28. Berkeley: Univ. California Press, Berkeley, Rubec, P.J., J.C. Bexley, H. Norris, M.S. Coyne, M.E. Monaco, S.G. Smith, and J.S.
Tiffany Vance, Alaska Fisheries Science Center, NMFS
CA. 98 pp. Ault. 1999. Suitability modeling to delineate habitat essential to sustainable fisheries.
Mark Wilkins, Alaska Fisheries Science Center, NMFS
Am. Fish. Soc. Symp., Vol. 22: pp. 108-133.
Deb Wilson-Vandenberg, California Department of Fish and Game
Gauch, H.G., Jr. 1982. Multivariate analysis in community ecology. Cambridge Univ.
Lisa Wooninck, Southwest Fisheries Science Center, NMFS
Press, New York, NY. 145 pp. Shaw, W.N., and T.J. Hassler. 1989. Species profiles: life histories and environmental
Nancy Wright, California Department of Fish and Game
requirements of coastal fishes and invertebrates (Pacific Northwest) – Lingcod. U.S.
Levon Yengoyan, TerraLogic GIS Lassuy, D.R. 1989. Species profiles: life histories and environmental requirements of Fish Wildl. Serv. Biol. Rep. 82(11.119), U.S. Army Corps Eng. TR EL-82-4, 10 p.
Mary Yoklavich, Southwest Fisheries Science Center, NMFS coastal fishes and invertebrates (Pacific Northwest) – English sole. USFWS - Biol. Starr, R.M. 1998. Design principles for rockfish reserves on the U. S. West Coast.
Mark Zimmermann, Alaska Fisheries Science Center, NMFS Rep., Vol. 82(11.101), U.S. Army Corps Eng. TR EL-82-4. 17 pp. Marine harvest refugia for West Coast rockfish: A workshop., Pacific Grove, Cali-
Lauth, R.R. 2001. The 2000 Pacific West Coast Upper Continental Slope trawl survey fornia.
REFERENCES of groundfish resources off Washington, Oregon, and California: Estimates of distri-
Allen, M.J., and G.B. Smith. 1988. Atlas and zoogeography of common fishes in bution, abundance, and length composition. NOAA Tech. Memo.NMFS-AFSC-120. U.S. Department of Commerce. Magnuson-Stevens Act Provisions. Federal Regis-
the Bering Sea and Northeastern Pacific. NOAA Tech. Rep. NMFS 66. Seattle, WA. Seattle, WA. 284 pp. ter: January 11, 2002 (Vol. 67, No. 8). Washington, DC: U.S. Govt. Printing Office.
151 pp. p. 1582.
Leet, W.S., C.M. DeWees, and C.W. Haugen. 1992. California’s Living Marine Re-
Brown, S.K., K.R. Buja, S.H. Jury, M.E. Monaco, and A. Banner. 2000. Habitat suit- sources: A Status Report. California Dept. Fish and Game. Sacramento, CA. 247 USFWS. 1980a. Habitat as a basis for environmental assessment. Report ESM 101.
ability models for eight fish and invertebrate species in Casco and Sheepscot Bays, pp. U.S. Fish Wildl.. Serv., Washington, DC.
Maine. N. Am. J. Fish. Manag., Vol. 20, pp. 408-435.
Leet, W.S., C.M. DeWees, R. Klingbeil, and E.J. Larson. 2001. California’s Living USFWS. 1980b. Habitat evaluation procedures (HEP). Report ESM 102. U.S. Fish
Cailliet, G.M., E.J. Burton, J.M. Cope, L.A. Kerr, R.J. Larson, D. VenTresca, R.N. Marine Resources: A Status Report. California Dept. Fish and Game. Sacramento, Wildl. Serv., Washington, DC.
Lea, and E. Knaggs. 2000. Final Report – Biological characteristics of near-shore CA. 592 pp.
fishes of California: A review of existing knowledge and proposed additional studies. USFWS. 1981. Standards for the development of habitat suitability index models
Love, M. 1996. Probably More Than You Want to Know About the Fishes of the
Submitted to Pacific States Marine Fisheries Commission. for use with the habitat evaluation procedures. Report ESM 103. U.S. Fish Wildl.
Pacific Coast. Really Big Press, Santa Barbara, CA. 45 pp. Serv., Washington, DC.
Cass, A.J., R.J. Beamish, and G.A. McFarlane. 1990. Lingcod (Ophiodon elongatus).
Love, M.S., L. Thorsteinson, C.W. Mecklenburg, and T.A. Mecklenberg. 1996. A
Can. Spec. Pub. Fish. Aquat. Sci., Vol. 109, 40 pp. Wilkins, M.E., M. Zimmermann, and K.L. Weinberg. 1998. The 1995 Pacific West
checklist of marine and estuarine fishes of the Northeast Pacific, from Alaska to Baja Coast Bottom Trawl Survey of Groundfish Resources: Estimates of Distribution, Abun-
Coyne, M.S. and J.D. Christensen. 1997. Habitat Suitability Index Modeling: Techni- California. National Biological Service. http://id-www.ucsb.edu/lovelab/home.html dance, and Length and Age Composition. NOAA Tech. Memo NMFS-AFSC-89.
cal Guidelines. NOAA/NOS Strategic Environmental Assessments Division, Silver University of California, Santa Barbara, CA. 24 pp.
Spring, MD. 19 pp. Williams, E.H., and S. Ralston. In press. Distribution and co-occurrence of rock-
Love, M.S., P. Morris, M. McCrae, and R. Collins. 1990. Life History Aspects of 19 fishes (family Scorpaenidae) over trawlable shelf and slope habitats of California
Dark, T.A., and M.E. Wilkins. 1994. Distribution, abundance, and biological character- Rockfish Species (Scorpaenidae: Sebastes) from the Southern California Bight. and Southern Oregon. Fishery Bulletin 000:000-000.
istics of groundfish off the coast of Washington, Oregon, and California, 1977-1986. NOAA Tech. Memo. NMFS 87. Seattle, WA. 113 pp.
NOAA Tech. Memo. NMFS 117. Seattle, WA. 114 pp. Wolotira, R.J., T.M. Sample, S.F. Noel, and C.R. Iten. 1993. Geographic and bathy-
Mahon, R., S.K. Brown, K.C.T. Zwanenburg, D. Atkinson, K.R. Buja, L. Claflin, G.D. metric distributions for many commercially important fishes and shellfishes off the
Dark, T.A., M.E. Wilkins, and K. Edwards. 1983. Bottom Trawl Survey of Canary Howell, M.E. Monaco, R.N. O’Boyle, and M. Sinclair. 1998. Assemblage and bioge- West Coast of North America, based on research survey and commercial catch data,
Rockfish (Sebastes pinninger), Yellowtail Rockfish (S. flavidus), Bocaccio (S. pau- ography of demersal fishes of the east coast of North America. Can. J. Fish. Aquat. 1912-84. NOAA Tech. Memo. NMFS-AFSC-6. 184 p.
cispinis) and Chilipepper (S. goodei) off Washington—California, 1980. NOAA Tech. Sci., Vol. 55: pp. 1704-1738.
Memo. F/NWC, Seattle, WA. 48 pp. Wright, N., J. Kum, C. King, E. Knaggs, B. Leos, and C. Perez. 2000. Marine fish-
Miller, D.J., and R.N. Lea. 1972. Guide to the Coastal Marine Fishes of California. ery profiles. Volume 1: Nearshore-Version 1. California Department of Fish and
Emmett, R.L., S.L. Stone, S.A. Hinton, and M.E. Monaco. 1991. Distribution and Cal. Dept. Fish and Game, Fish Bull. Vol. 157, pp. 23-29. Game, Marine Region. 150 pp.
abundance of fishes and invertebrates in West Coast estuaries, Volume II: Species
NOAA/NMFS. 1998. Essential Fish Habitat – West Coast Groundfish Appendix
life history summaries. ELMR Rep. No. 8. NOAA/NOS SEA Division, Rockville, MD. Yoklavich, M.M., H.G. Greene, G.M. Cailliet, D.E. Sullivan, R.N. Lea, and M.S. Love.
(online). http://www.nwr.noaa.gov/1sustfsh/efhappendix. NOAA/NMFS, EFH Core
329 pp. 2000. Habitat associations of deep-water rockfishes in a submarine canyon: an
Team for West Coast Groundfish, Seattle, WA. example of a natural refuge. Fish Bulletin, U.S. 98:625-641.
Eschmeyer, W.N., E.S. Herald, H. Hammann, and K.P. Smith. 1983. A field guide to
Pacific Coast Fishes. Houghton Mifflin Company, Boston, MA. 89 pp.
45
Cassin's Auklet Ptychoramphus aleuticus Charadriiformes/Alcidae
Parkinson's Petrel Procellaria parkinsoni Procellariiformes/Procellariidae
Rhinoceros Auklet Cerorhinca monocerata Charadriiformes/Alcidae
Flesh-footed Shearwater Puffinus carneipes Procellariiformes/Procellariidae
Tufted Puffin Fratercula cirrhata Charadriiformes/Alcidae
Short-tailed Shearwater Puffinus tenuirostris Procellariiformes/Procellariidae
Species mapped with related species &puffinus the summary diversity and density analyses (n=9, 4 maps)
Manx Shearwater Puffinus used in Procellariiformes/Procellariidae
Western Grebe Aechmophorus occidentalis Podicipediformes/Podicipedidae
Townsend's Shearwater Puffinus auricularis Procellariiformes/Procellariidae
Clark's Grebe Aechmophorus clarkii Podicipediformes/Podicipedidae
Wilson's Storm-petrel Oceanites oceanicus Procellariiformes/Hydrobatidae
Black Scoter Melanitta nigra tethys Anseriformes/Anatidae
Wedge-rumped Storm-petrel Oceanodroma Procellariiformes/Hydrobatidae
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Surf Scoter Storm-petrel Melanitta perspicillata Anseriformes/Anatidae
Markham's Oceanodroma markhami Procellariiformes/Hydrobatidae
White-winged Scoter Melanitta fuscamicrosoma Anseriformes/Anatidae
Least Storm-petrel Oceanodroma Procellariiformes/Hydrobatidae
Caspian Tern Sterna caspia Charadriiformes/Laridae/Sterninae
Red-billed Tropicbird Phaethon aethereus Pelecaniformes/Phaethonidae
Elegant Tern Sterna elegans Charadriiformes/Laridae/Sterninae
Brown Booby Sula leucogaster Pelecaniformes/Phaethonidae
Table 11. Marine bird species used in this analysis. TablePelican
Xantus's Murrelet Marine bird species usedhypoleucus
White 11 cont. Synthliboramphus in this analysis.
Charadriiformes/Alcidae
Pelecanus erythrorynchos Pelecaniformes/Pelecanidae
Common Name Scientific Name Order/Family/SubFamily Common Name Scientific Namecraveri Order/Family/SubFamily
Craveri's Murrelet Synthliboramphus Charadriiformes/Alcidae
South Polar Skua Stercorarius maccormicki Charadriiformes/Laridae/Stercorariinae
Species that were mapped separately and used in the summary bird diversity and density analyses (n=31) Pomarine Jaeger
in the data set used only in the summary the diversity Charadriiformes/Laridae/Stercorariinae(n=31)
Stercorarius pomarinus
Species that were mapped separately and used in bird summary and density analyses (n=37)
bird diversity and density analyses
Pacific Loon Gavia pacifica Gaviiformes/Gaviiadae Red-throated
Pacific Loon Loon stellata Gaviiformes/Gaviidae
Parasitic Jaeger Stercorarius parasiticus Charadriiformes/Laridae/Stercorariinae
Gavia pacifica Gaviiformes/Gaviiadae
Laysan Albatross Phoebastria immutabilis Procellariiformes/Diomedeidae Common Loon Gavia immerlongicaudus Gaviiformes/Gaviidae
Long-tailed Jaeger Stercorarius Charadriiformes/Laridae/Stercorariinae
Laysan Albatross Phoebastria immutabilis Procellariiformes/Diomedeidae
Black-footed Albatross Phoebastria nigripes Procellariiformes/Diomedeidae Horned Grebe Podiceps auritus Podicipediformes/Podicipedidae
Black-footed Gull
Bonaparte's Albatross Larus philadelphia Charadriiformes/Laridae/Larinae
Phoebastria nigripes Procellariiformes/Diomedeidae
Northern Fulmar Fulmarus glacialis Procellariiformes/Procellariidae Red-necked Grebe Podiceps glacialis
Fulmarus grisegena Podicipediformes/Podicipedidae
Mew Gull Larus canus Charadriiformes/Laridae/Larinae
Northern Fulmar Procellariiformes/Procellariidae
Pink-footed Shearwater Puffinus creatopus Procellariiformes/Procellariidae Eared Grebe Podiceps nigricollis Podicipediformes/Podicipedidae
Pink-footedGull
Ring-billed Shearwater Larus delawarensis Charadriiformes/Laridae/Larinae
Puffinus creatopus Procellariiformes/Procellariidae
Buller's Shearwater Puffinus bulleri Procellariiformes/Procellariidae Murphy's Petrel Pterodroma ultima
California Gull Larus californicus Charadriiformes/Laridae/Larinae
Buller's Shearwater Puffinus bulleri Procellariiformes/Procellariidae
Sooty Shearwater Puffinus griseus Procellariiformes/Procellariidae Cook's Petrel Pterodroma cookii
Thayer's Gull Larus thayeri Charadriiformes/Laridae/Larinae
Sooty Shearwater Puffinus griseus Procellariiformes/Procellariidae
Black-vented Shearwater Puffinus opisthomelas Procellariiformes/Procellariidae Parkinson's Petrel Procellaria parkinsoni
Glaucous Gull Larus hyperboreus Charadriiformes/Laridae/Larinae
Black-vented Shearwater Puffinus opisthomelas Procellariiformes/Procellariidae
Fork-tailed Storm-petrel Oceanodroma furcata Procellariiformes/Hydrobatidae Flesh-footedKittiwake
Red-legged Shearwater Puffinus carneipes Procellariiformes/Procellariidae
Rissa brevirostris Charadriiformes/Laridae/Larinae
Fork-tailed Storm-petrel Oceanodroma furcata Procellariiformes/Hydrobatidae
Leach's Storm-petrel Oceanodroma leucorhoa Procellariiformes/Hydrobatidae Short-tailed Shearwater Puffinus tenuirostris Procellariiformes/Procellariidae
Royal Tern Sterna maxima Charadriiformes/Laridae/Sterninae
Leach's Storm-petrel Oceanodroma leucorhoa Procellariiformes/Hydrobatidae
Ashy Storm-petrel Oceanodroma homochroa Procellariiformes/Hydrobatidae Manx Shearwater Puffinus puffinus Procellariiformes/Procellariidae
Common Tern Sterna hirundo Charadriiformes/Laridae/Sterninae
Ashy Storm-petrel Oceanodroma homochroa Procellariiformes/Hydrobatidae
Black Storm-petrel Oceanodroma melania Procellariiformes/Hydrobatidae Townsend's Shearwater Puffinus auricularis Procellariiformes/Procellariidae
Forster's Tern Sterna forsteri Charadriiformes/Laridae/Sterninae
Black Storm-petrel Oceanodroma melania Procellariiformes/Hydrobatidae
Brown Pelican Pelecanus occidentalis Pelecaniformes/Pelecanidae Wilson's Storm-petrel Oceanites oceanicus Procellariiformes/Hydrobatidae
Thick-billed Murre Uria lomvia Charadriiformes/Alcidae
Brown Pelican Pelecanus occidentalis Pelecaniformes/Pelecanidae
Brandt's Cormorant Phalacrocorax penicillatus Pelecaniformes/Phalacrocoracidae Wedge-rumped
Brandt'sMurreletStorm-petrel Oceanodroma penicillatus
Phalacrocorax tethys Procellariiformes/Hydrobatidae
Ancient Cormorant Synthliboramphus antiquus Charadriiformes/Alcidae
Pelecaniformes/Phalacrocoracidae
INTRODUCTION Double-crested Cormorant Phalacrocorax auritus Pelecaniformes/Phalacrocoracidae Markham's Storm-petrel Oceanodroma auritus
Phalacrocorax markhami Procellariiformes/Hydrobatidae
Parakeet Auklet Aethia psittacula Charadriiformes/Alcidae
Double-crested Cormorant Pelecaniformes/Phalacrocoracidae
The California Current system runs south through the north/central California Pelagic Cormorant Phalacrocorax pelagicus Pelecaniformes/Phalacrocoracidae Least Storm-petrel Oceanodroma pelagicus
Phalacrocorax microsoma Procellariiformes/Hydrobatidae
Horned Puffin Fratercula corniculata Charadriiformes/Alcidae
Pelagic Cormorant Pelecaniformes/Phalacrocoracidae
study area; it is one of the most productive ocean systems in the world (Glantz Red-necked Phalarope Phalaropus lobatus Charadriiformes/Scolopacidae Red-billed Tropicbird Phaethon aethereus Pelecaniformes/Phaethonidae
Red-necked Phalarope Phalaropus lobatus Charadriiformes/Scolopacidae
Red Phalarope Phalaropus fulicarius Charadriiformes/Scolopacidae Brown Booby Sula leucogaster Pelecaniformes/Phaethonidae
Table 11 contains the marine birdfulicarius that were selected for this analysis; data
Red Phalarope Phalaropus Charadriiformes/Scolopacidae
and Thompson, 1981). Hence, the study area contains a rich fauna of marine species
Heermann's Gull Larus heermanni Charadriiformes/Laridae/Larinae White Pelican Pelecanus erythrorynchos Pelecaniformes/Pelecanidae
Heermann's Gull Larus heermanni Charadriiformes/Laridae/Larinae
birds, as evidenced in species abundance and richness. In addition to a populous for 40 Polar Skua were either mapped separately, or Charadriiformes/Laridae/Larinae
South species together for small species groups
Western Gull Larus occidentalis Charadriiformes/Laridae/Larinae Stercorarius maccormicki Charadriiformes/Laridae/Stercorariinae
Western Gull Larus occidentalis
breeding community, the community of seasonal residents and migrants is even that generally co-occur (e.g., scoters). Ten of these species maps are included in this
Glaucous-winged Gull Larus glaucescens Charadriiformes/Laridae/Larinae Pomarine Jaeger Gull Stercorarius pomarinus Charadriiformes/Laridae/Stercorariinae
Glaucous-winged Larus glaucescens Charadriiformes/Laridae/Larinae
more robust, as central California is the destination for many marine bird species section; Gull remaining maps are on the accompanying CD-ROM. The remaining 37
Parasitic the
Sabine's Gull Xema sabini Charadriiformes/Laridae/Larinae Sabine's Jaeger Stercorarius
Xema sabini parasiticus Charadriiformes/Laridae/Stercorariinae
Charadriiformes/Laridae/Larinae
California Gull Larus californicus Charadriiformes/Laridae/Larinae
species that occurred in the study longicaudus data set were used to develop summary
Long-tailed Jaeger Stercorarius Charadriiformes/Laridae/Stercorariinae
California Gull Larus californicus Charadriiformes/Laridae/Larinae
seeking productive feeding areas and acceptable habitat in which to spend their non- area and
Black-legged Kittiwake Rissa tridactyla Charadriiformes/Laridae/Larinae Bonaparte's Kittiwake
Black-leggedGull Larus philadelphia
Rissa tridactyla Charadriiformes/Laridae/Larinae
breeding periods. Unlike many marine organisms, marine birds have a tremendous marine bird maps on marine canus diversity and density.
Sternabird
Arctic Tern Sterna paradisaea Charadriiformes/Laridae/Sterninae Mew Tern
ArcticGull Larus paradisaea Charadriiformes/Laridae/Larinae
Charadriiformes/Laridae/Sterninae
mobility and the fact that many seek this region to find food bespeaks the region’s Common Murre Uria aalge Charadriiformes/Alcidae Ring-billed Gull Larus delawarensis Charadriiformes/Laridae/Larinae
Common Murre Uria aalge Charadriiformes/Alcidae
Pigeon Guillemot Cepphus columba Charadriiformes/Alcidae California Gull Larus californicus Charadriiformes/Laridae/Larinae
trophic richness. Fortunately for the purpose of management of the central California About Guillemot
Pigeon the Survey Data and Literature Used in this Assessment. The survey data
Cepphus columba Charadriiformes/Alcidae
Marbled Murrelet Brachyramphus marmoratus Charadriiformes/Alcidae Thayer's
Marbled Gull Larus thayeri Charadriiformes/Laridae/Larinae
used in Murrelet Brachyramphus marmoratus Charadriiformes/Alcidae
National Marine Sanctuaries, the marine avifauna of the study area have been one this summary were not designed with sanctuary resource management in
Cassin's Auklet Ptychoramphus aleuticus Charadriiformes/Alcidae GlaucousAuklet
Cassin's Gull Larus hyperboreus Charadriiformes/Laridae/Larinae
Ptychoramphus aleuticus Charadriiformes/Alcidae
of the most thoroughly surveyed. mind, but include the interests of individual researchers to study spatial and temporal
Rhinoceros Auklet Cerorhinca monocerata Charadriiformes/Alcidae Red-legged Kittiwake Rissa brevirostris Charadriiformes/Laridae/Larinae
Rhinoceros Auklet Cerorhinca monocerata Charadriiformes/Alcidae
patterns of marine birds,Fratercula cirrhata
federal government efforts to assess potential biological
Tufted Puffin Fratercula cirrhata Charadriiformes/Alcidae Royal Puffin
TuftedTern Sterna maxima Charadriiformes/Laridae/Sterninae
Charadriiformes/Alcidae
Species mapped with related species & used in the summary diversity and density analyses (n=9, 4 maps) Species Tern
Common of oil with related species hirundo in government efforts density analyses oil 4 maps)
Sterna & used Charadriiformes/Laridae/Sterninae
DATA AND ANALYSES impacts mappeddevelopment, and statethe summary diversity and to respond to(n=9,spills, of
Western Grebe Aechmophorus occidentalis Podicipediformes/Podicipedidae Forster's Tern Sterna forsteri Charadriiformes/Laridae/Sterninae
Western Grebe Aechmophorus occidentalis Podicipediformes/Podicipedidae
Overview of Map Development and Analysis Process. The methods used in each which there have been several major ones in the Charadriiformes/Alcidae
study area.
Clark's Grebe Aechmophorus clarkii Podicipediformes/Podicipedidae Thick-billed Murre Uria lomvia
Clark's Grebe Aechmophorus clarkii Podicipediformes/Podicipedidae
survey were different, and because of this, careful consideration and correction are Black Scoter Melanitta nigra Anseriformes/Anatidae Ancient Murrelet Synthliboramphus antiquus Charadriiformes/Alcidae
Black Scoter Melanitta nigra Anseriformes/Anatidae
required to merge the data sets in a meaningful and scientifically acceptable way. The The Literature. Several reports,perspicillata from these surveys, provided background
Melanitta resulting
Surf Scoter Melanitta perspicillata Anseriformes/Anatidae Parakeet Auklet Aethia psittacula Charadriiformes/Alcidae
Surf Scoter Anseriformes/Anatidae
White-winged Scoter Melanitta fusca Anseriformes/Anatidae Horned Puffin Fraterculafusca
Melanitta corniculata Charadriiformes/Alcidae
information Scoter occurrence patterns of marine birds in the region. The general
White-winged on the Anseriformes/Anatidae
major steps of the data development for the bird analyses were as follows: species
Caspian Tern Sterna caspia Charadriiformes/Laridae/Sterninae Caspian Tern Sterna caspia Charadriiformes/Laridae/Sterninae
and study area selection; data set identification and collection; data corrections; composition and distributionelegans marine avifauna was described by Ainley
of the
Elegant Tern Sterna elegans Charadriiformes/Laridae/Sterninae Elegant Tern Sterna Charadriiformes/Laridae/Sterninae
data conversion into common comparable units; organizing the data into 5’ latitude (1976) and Briggs et al.Synthliboramphus hypoleucus
(1983, 1987a, b). Ainley and DeSante (1980) and Pyle
Xantus's Murrelet Synthliboramphus hypoleucus Charadriiformes/Alcidae Xantus's Murrelet Charadriiformes/Alcidae
by 5’ longitude cells; and calculating effort and density for each marine bird species. and Henderson (1991) provide a fine-scale look at species’ seasonal presence and
Craveri's Murrelet Synthliboramphus craveri Charadriiformes/Alcidae Craveri's Murrelet Synthliboramphus craveri Charadriiformes/Alcidae
Species in the data set used only in the summary bird diversity and density analyses (n=37)
migratory the data set as viewedthe summaryFarallon Islands; Ainley et al. (1995a, c), Veit
Species in periods, used only in from the bird diversity and density analyses (n=37)
Seasonal density maps were then created for 40 species. Overall density, biomass Red-throated Loon Gavia stellata Gaviiformes/Gaviidae Red-throated Loon Gavia stellata Gaviiformes/Gaviidae
density, and diversity maps were also created using distribution and abundance data etCommon Loon and Oedekoven immer (2001) provide an interannual view of variability
al. (1997) Gavia et al.
Common Loon Gavia immer Gaviiformes/Gaviidae Gaviiformes/Gaviidae
for 76 bird species combined. These maps were reviewed at an expert workshop in inHorned Grebe
spatial occurrence. The last four references, as Podicipediformes/Podicipedidae
well as Ainley et al. (1994), Spear
Horned Grebe Podiceps auritus Podicipediformes/Podicipedidae Podiceps auritus
October 2002. The draft bird report was also sent out for expert review in November and AinleyGrebe
Red-necked (1999) and Ainley and Divoky (2001), investigated long-term temporal
Red-necked Grebe Podiceps grisegena Podicipediformes/Podicipedidae Podiceps grisegena Podicipediformes/Podicipedidae
Eared Grebe Podiceps nigricollis Podicipediformes/Podicipedidae
trendsGrebe
Eared Podiceps nigricollis Podicipediformes/Podicipedidae
(see list of reviewers at end of this section). Revisions were made to the maps and in populations. Information on habitat preferences of marine birds and how
Murphy's Petrel Pterodroma ultima Procellariiformes/Procellariidae Murphy's Petrel Pterodroma ultima Procellariiformes/Procellariidae
text based on that review. these are affected by ocean climate variability are provided for selected species
Cook's Petrel Pterodroma cookii Procellariiformes/Procellariidae Cook's Petrel Pterodroma cookii Procellariiformes/Procellariidae
inParkinson'sand Boekelheide (1990), Oedekoven etProcellariiformes/Procellariidae
Ainley Petrel al. (2001), and in a GIS analysis
Parkinson's Petrel Procellaria parkinsoni Procellariiformes/Procellariidae Procellaria parkinsoni
Flesh-footed Shearwater Puffinus carneipes Procellariiformes/Procellariidae
Species Selected for Analysis. Selection criteria for bird species included in this by Allen (1994). The food-webcarneipes
Puffinus relationships of marine birds in this region are also
Flesh-footed Shearwater Procellariiformes/Procellariidae
Short-tailed Shearwater Puffinus tenuirostris Procellariiformes/Procellariidae Short-tailed Shearwater Puffinus tenuirostris Procellariiformes/Procellariidae
assessment were: 1) the species must have a mostly marine distribution in the remarkably well known (Balz and Morejohn 1977; Ainley and Sanger 1979; Briggs et
Manx Shearwater Puffinus puffinus Procellariiformes/Procellariidae Manx Shearwater Puffinus puffinus Procellariiformes/Procellariidae
study area; and 2) adequate ocean survey data for the species is available and in al. 1984, Briggs and Chu Puffinus auricularis
1987; Chu 1984; Ainley and Boekelheide 1990; Ainley et al.
Townsend's Shearwater Puffinus auricularis Procellariiformes/Procellariidae Townsend's Shearwater Procellariiformes/Procellariidae
a useable format. Species that are abundant, endangered, threatened, or a state 1996a, b; and Sydeman et al. 1997); and the breeding biology, including interannual
Wilson's Storm-petrel Oceanites oceanicus Procellariiformes/Hydrobatidae Wilson's Storm-petrel Oceanites oceanicus Procellariiformes/Hydrobatidae
Wedge-rumped Storm-petrel Oceanodroma tethys Procellariiformes/Hydrobatidae
species of concern were also a priority. The study area for the GIS assessment was variability in productivity Oceanodroma tethys to food-web variation, is very well known
and relationship
Wedge-rumped Storm-petrel Procellariiformes/Hydrobatidae
Markham's Storm-petrel Oceanodroma markhami Procellariiformes/Hydrobatidae Markham's Storm-petrel Oceanodroma markhami Procellariiformes/Hydrobatidae
seaward of the beach and did not include estuaries, so few shorebirds and waterfowl (Ainley and BoekelheideOceanodroma microsoma 1995b). See the end of this section for
1990, Ainley et al.
Least Storm-petrel Oceanodroma microsoma Procellariiformes/Hydrobatidae Least Storm-petrel Procellariiformes/Hydrobatidae
were included. Because marine distributions of birds are affected by where they completeTropicbirdreferences used.
Red-billed list of
Red-billed Tropicbird Phaethon aethereus Pelecaniformes/Phaethonidae Phaethon aethereus Pelecaniformes/Phaethonidae
breed and roost, we included information on the location and size of breeding and Brown Booby Sula leucogaster Pelecaniformes/Phaethonidae Brown Booby Sula leucogaster Pelecaniformes/Phaethonidae
White Pelican Pelecanus erythrorynchos Pelecaniformes/Pelecanidae
The Data Sets. See Table 12, aerythrorynchos of data sets used in the analyses, and
White Pelican Pelecanus summary Pelecaniformes/Pelecanidae
roosting sites, where available.
South Polar Skua Stercorarius maccormicki Charadriiformes/Laridae/Stercorariinae South Polar Skua Stercorarius maccormicki Charadriiformes/Laridae/Stercorariinae
Figures 39 and 40, which show the spatial extent of the individual data sets used
Pomarine Jaeger Stercorarius pomarinus Charadriiformes/Laridae/Stercorariinae Pomarine Jaeger Stercorarius pomarinus Charadriiformes/Laridae/Stercorariinae
Parasitic Jaeger Stercorarius parasiticus Charadriiformes/Laridae/Stercorariinae Parasitic Jaeger Stercorarius parasiticus Charadriiformes/Laridae/Stercorariinae
46
Long-tailed Jaeger Stercorarius longicaudus Charadriiformes/Laridae/Stercorariinae Long-tailed Jaeger Stercorarius longicaudus Charadriiformes/Laridae/Stercorariinae
Bonaparte's Gull Larus philadelphia Charadriiformes/Laridae/Larinae Bonaparte's Gull Larus philadelphia Charadriiformes/Laridae/Larinae
Mew Gull Larus canus Charadriiformes/Laridae/Larinae Mew Gull Larus canus Charadriiformes/Laridae/Larinae
Ring-billed Gull Larus delawarensis Charadriiformes/Laridae/Larinae Ring-billed Gull Larus delawarensis Charadriiformes/Laridae/Larinae
California Gull Larus californicus Charadriiformes/Laridae/Larinae California Gull Larus californicus Charadriiformes/Laridae/Larinae
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
in the bird analyses. The ship and aerial strip transect data used in the GIS assessment were
collected from 1980-2001 and occurred from Point Arena south to Point Sal, and offshore to the
extent of data availability. However, the species maps do not generally include the full extent of
Spatial Extent of Data Sets: Ship-based Surveys available data, primarily because the assessment was focused on the national marine sanctuaries
off central California. Also, estuaries were not part of the study area, but coastal colonies in
estuaries were mapped to provide a more complete view of important areas for breeding species.
129°W 128°W 127°W 126°W 125°W 124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
See a more detailed description of data sets on the accompaning CD-ROM.
200 m
200 m
20
20
SF-DODS Cruises
Midwater Trawls for Juvenile Rockfish
0
0
0m
0m
1985-2001 1996-2000
39°N
39°N
Data Synthesis.
Mainly upwelling period Year round Summarizing Transect Data into Grid Cells. The above data sets required a significant amount
of processing and correction in order to synthesize them. Because wind speed affects detection
of marine birds, data collected when wind speed exceeded 25 knots were excluded. Data were
38°N
allocated into 5’ latitude by 5’ longitude cells. All aerial data were continuous; each ship-based
38°N
data set was converted separately into a continuous transect format to the extent possible. The
continuous aerial data were binned into the appropriate cell. For the SF-DODS and EPOCS
Table 12. Summary of at-sea survey data sets used in the analyses.
37°N
37°N
Ocean Total
Principal Platform Habitat Seasons Transect
0 25 50 Km Data Set Investigator Height Covered2 Years Sampled Width
Surface survey
MMS Low of the shelf,
36°N
36°N
Altitude Aerial slope & deep All three
Surveys Briggs Pembroke, 62m ocean beyond 1980-1983 seasons 50m
Surveyor, 12m,
EPOCS Discoverer, Surface survey
35°N
35°N
Shipboard Oceano- of the deep All three
a b Surveys Ainley grapher, 15m ocean 1984-1994 seasons 300-600m
CA Seabird
Ecology Low- Surface survey
200 m
200 m
Altitude Aerial Partenavia, of shelf and Mainly
20
20
EPOCS Cruises
NMFS ORCAWALE Cruises
0
0
0m
0m
Surveys Briggs 62m slope 1985 Upwelling 50m
1984-1994
2001
39°N
39°N
NMFS
Year round
Mainly oceanic period Midwater Trawl
Juv. Rockfish Surface survey
Assessment: David Starr of shelf and Mainly
Ship Surveys Ainley Jordan, 10m slope to 3000 m 1985-2001 Upwelling 300m
38°N
38°N
OSPR Low Surface survey
Altitude Aerial Partenavia, of shelf and 1994-1998, All three
Surveys Bonnell, Tyler 62m slope 2001 seasons 50m
MMS Santa
Barbara
37°N
37°N
Channel Low Surface survey
Altitude Aerial Partenavia, of shelf and All three
Surveys Bonnell 62m slope 1995-1997 seasons 50m
Surface survey
SF-DODS Ship of shelf and All three
36°N
36°N
Surveys Ainley Point Sur, 8m slope to 3000 m 1996-2000 seasons 300m
Surface survey 200-300m,
NMFS/SWFSC of the shelf, Mainly depending on
ORCAWALE slope & deep Oceanic species &
Ship Survey Ballance MacArthur, 11m ocean beyond 2001 (Aug-Nov) conditions
35°N
35°N
c d Note
See description of data sets on the CD for more information on the data sets.
129°W 128°W 127°W 126°W 125°W 124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Source Data: See text. studies, and the Rockfish Assessment cruises prior to 1997, the beginning position, ship heading,
Figure 39. Spatial extent of data sets used in the marine bird analysis: ship-based surveys. and speed were used to compute the end position of each 2-4 km continuous transect. From
this, a midpoint of the transect was determined. As times of observations were not available, the
position of the midpoint was used to select the cell to which the survey effort was assigned. If
this midpoint fell on a cell boundary, it was assigned to the cell to the north or west. To maintain
47
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Table 13. Summary of combined data set effort by ocean season.
Dates Used for Number Kilometers of Number of
Spatial Extent of Data Sets: Aerial Surveys Ocean Each Ocean of Years Trackline Number 5' Cells
Season Season Months Included Surveyed of Visits Sampled
1980-1982,
Upwelling 1985-2001
15 Mar-14 Aug 5 64177 11050 1335
129°W 128°W 127°W 126°W 125°W 124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
200 m
200 m
MMS Aerial Surveys OSPR Aerial Surveys 1980-1982,
20
20
0
0
0m
0m
1991, 1994-
1980-1983 1994-1998, 2001
39°N
39°N
Oceanic 2001
15 Aug-14 Nov 3 29263 4171 1130
Year round Year round
Davidson 1980-1986,
Current 1991-2001
15 Nov-14 Mar 4 40265 5878 1593
1980-2001
TOTAL 1 Jan – 31 Dec 12 133705 21099 2294
38°N
38°N
Note. The total number of cells sampled is not a straight sum; it refers to the number of unique
cells surveyed.
the correspondence between effort and bird observations, observations were also assigned
37°N
37°N
to the transect midpoints. For the Rockfish Assessment Cruises from 1997 onward, effort was
assigned to the cells through which the vessel passed based on the proportion of trackline that
0 25 50 Km
fell within each cell, and observations were interpolated along the cruise track according to
the time of each observation. The marine bird survey data from the ORCAWALE cruise were
36°N
36°N
recorded continuously using automatic recording software and were processed like the aerial
survey data.
Data Analysis.
35°N
35°N
e f Effort. The combined at-sea survey effort for birds included 133,705 kilometers of trackline, as
well as 128,886 observations of 973,318 birds in the analyzed data set. Survey effort by ocean
season is summarized in Figure 41 and Table 13.
200 m
200 m
Seabird Ecology Aerial Surveys
20
20
MMS Santa Barbara Channel
0
0
0m
0m
1985 Aerial Surveys
39°N
39°N
Calculating Density. From the digitized survey data, we mapped the distribution of effort and of
1995-1997
Mainly upwelling period
species observations into a grid of 5-minute latitude by 5-minute longitude cells, using MMS-
Year round
CDAS (Marine Mammal and Seabird Computer Database Analysis System, MMS 2001). The
species data were first transformed into densities on the basis of strip widths (which varied by
38°N
38°N
platform, depending on speed and height above water; see Table 12). The number of birds of
each species seen was then divided by area surveyed to estimate density in each cell for that
data set. For construction of density plots, if a cell was censused in other years or the same
year by another survey, densities in cells were averaged and weighted according to effort.
37°N
37°N
Organizing Data into Ocean Seasons. Effort and species data were organized and mapped into
three distinct ocean seasons (Bolin and Abott 1963): Upwelling, Oceanic, and Davidson Current,
because ocean conditions differ distinctly among them and are known to affect the biota of the
36°N
36°N
California Current (e.g. Ainley 1976, Briggs et al. 1987). As there is significant interannual variation
in the actual initiation and termination of these seasons, the following dates were defined for each
season for purposes of analysis: Upwelling Season is 15 March-14 August; Oceanic Season is
15 August-14 November; and Davidson Current Season is 15 November-14 March.
35°N
35°N
g h
Seasonal Density Maps for Individual Species. Seasonal density maps were generated for 40 bird
129°W 128°W 127°W 126°W 125°W 124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
species. These maps were then reviewed to characterize the spatial and seasonal occurrence
Source Data: See text.
pattern of each species in the study area.
Figure 40. Spatial extent of data sets used in the analysis: aerial surveys.
Seasonal High Use Areas for Individual Species. In order to provide a summary map of space
use, seasonal density data were binned into 10-minute latitude by 10-minute longitude cells
for each species or species group. The purpose of the seasonal high use maps is to provide
48
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
one overall map for each species (or group of species) that standardize for variable effort among cells and variable strip
Combined At-Sea Effort for Marine Bird Analysis describes the spatial and temporal use patterns, as clearly as width for species, density was used for each species in each
possible. The seasonal high use index is based on the top 20% cell as the basis for calculating the diversity index value.
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
of sampled cells within a given season. The index is therefore
Upwelling Season Oceanic Season
200 m
200 m
20
20
sensitive to cells which were not sampled in any one of the The Shannon Index was selected as the diversity metric
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) three seasons, causing a downward bias in the index. because it is widely used and accepted in community ecology.
39°N
39°N
It has three desirable properties for a diversity index, noted
Effort Use of a 10-minute block size greatly reduces the magnitude below. Most diversity indices do not take these three qualities
(Km of trackline)
of this bias. Non-zero cells were then ranked and those in the into account. For more information on diversity indices, see
>1000.00
top 20 percent were selected and defined as seasonal high use Ecological Diversity, E.C. Pielou, pp 7-18.
500.01 - 1000.00
38°N
38°N
areas. Cells were then mapped with colors corresponding to
250.01 - 500.00
100.01 - 250.00
the number of seasons of high use. Cells in which there was 1. The diversity index is greatest when all species in the
50.01 - 100.00
effort but birds were not observed, and cells where sightings community are equally represented in numbers (e.g., evenness.
25.01 - 50.00
occurred but were never high use areas, were also mapped in a community). Or, for a given number of species (e.g.,
5.01 - 25.00
0.01 - 5.00
with two additional colors. richness value), the diversity index should have it’s greatest
37°N
37°N
value when the proportion of each species is the same.
0 25 50 Km
Major Breeding Colonies. Best available breeding colony data
(number of breeding birds, mostly from Carter, et al. 1992, 2. Given two completely diverse or even communities, the
with some updates) were mapped for each species for which one with the higher number of species has a greater diversity
colony information was available, on the same map as the value.
36°N
36°N
"seasonal high use" information. A map (pp. 81) and table (p.
53) of the top 40 breeding colonies is included in Section 2.2; 3. The last property is difficult to summarize: This property takes
the complete colony table, based on best available data, will into account the hierarchical nature, or "representativeness" in
be included on the CD-ROM. the biological classification of each species when estimating
diversity.
35°N
35°N
a b Spatial and Temporal Patterns Summary Table. Density
maps for 44 species were inspected to identify which cells Evaluating Variation in Species Abundance. In order to
200 m
Davidson Current Season
200 m
All Seasons
20
20
exhibited the highest density each season. Using the two evaluate factors that affect the abundance of marine birds
0
0
0m
0m
(Nov. 15 - Mar. 14) highest density categories for each species, relatively high in the study area, a regression model was developed (Seber
39°N
39°N
density areas associated with large bathymetric areas (inner 1977, Kleinbaum et al. 1988), with marine bird density as
shelf, outer shelf, upper slope) were identified, as well as with the dependent variable. Independent variables that could be
several smaller discrete habitat features (e.g., Monterey Bay addressed in the limited time frame included: ocean season,
Canyon) (p. 51). year, ocean depth, distance to nearest breeding colony,
distance to shelf break (estimated to the 200 meter isobath),
38°N
38°N
Summary of Overall Density, Biomass and Diversity Maps distance to deep ocean (estimated to the 2000 meter isobath),
for 76 Marine Bird Species. Overall marine bird densities latitude, periods of short-term ocean climate anomalies (e.g.,
were mapped for each season and for all seasons combined. El Niño or La Niña events), and latitude. The data used for
Densities of all species in a cell were converted to biomass by the multiple regression analyses was a subset of the mapping
37°N
37°N
multiplying density for each species by its average body mass data set; the regression data set included cell-based density
(from Dunning 1993), then summing for all species detected data from 1985 - 2002 (6,641 cell samples, all with effort ≥
in that cell. Biomass was then mapped in a fashion similar to 0.24km2 per cell).
the individual species’ density maps.
Response to Variation in Marine Climate. Short-term ocean
36°N
36°N
The Shannon Index (Shannon and Weaver 1949) climate anomaly in this report is often referred to as ENSO
(El Niño/Southern Oscillation), and generally refers to the
n n
S
H ′ = − ∑ i ln i climatic events that cause significant interannual changes in
i =1 n n
thermocline depth and water temperature in the study area,
was used to quantify species diversity. This index measures resulting in warm-water periods (often known as El Niño
35°N
35°N
c d the degree to which the species assemblage is dominated by events), cold-water periods (often known as La Niña events),
a single species. If species A dominates all the species seen or neutral periods, when the water is neither unusually warm
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
within a cell, then diversity is low, and vice versa. Diversity nor cold (Ainley et al. 1995b and references therein).
Source Data: See text.
was calculated for each season and all seasons combined. To
Figure 41. Total survey effort for marine bird analyses.
49
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
sets ranged from 1975-2001. Table 14 indicates the periods
The official ENSO events and time periods tracked by NOAA
of unusual weather (warm water, cold water, and neutral) as
are relevant for regions to the far south and well outside of the
determined from these data.
study area; the official NOAA ENSO periods do not accurately
reflect the timing of the ENSO-related periods that occur off
Also affecting marine climate are decadal-scale factors involved
central California. In part, this is because the marine climate of
in the Pacific Decadal Oscillation (Mantua and Hare 2002). A
the central and northern California Current region is affected
regime shift occurred in 1976, from cold to warm, and may
as well by variation in atmospheric pressure centers in the
have occurred again in the winter of 1998/1999, from warm to
Gulf of Alaska.
cold. This means that the overall average state of the system
could be characterized as warm or cold, with other shorter-
Table 14. Assignment of warm, cold and neutral periods,
based on surface water temperatures off Cental California. term variation embedded (e.g., ENSO). The effect of regime
shifts on marine bird occurrence is addressed near the end
Davidson
Current Upwelling Oceanic of this report.
Year Season Season Season
1975 Cold Cold Cold ANALYTICAL MAP PRODUCTS
The analytical map products (Figures 42-60) include seasonal
1976 Cold Cold Warm
density maps for 13 species and nine summary analyses maps
1977 Warm Cold Neutral
for marine birds. These maps are a subset of the total mapped
1978 Warm Warm Cold
results for this analysis. Additional maps and text products
1979 Cold Neutral Neutral
are included on the CD-ROM. Of the 35 species maps, these
1980 Warm Neutral Cold
ten were chosen for inclusion in the document because they
1981 Warm Cold Cold
represent a variety of spatial and temporal patterns in and
1982 Neutral Neutral Neutral
around the sanctuaries.
1983 Warm Warm Warm
1984 Warm Neutral Neutral
1985 Cold Warm Cold
1986 Neutral Neutral Neutral
1987 Warm Warm Warm
1988 Neutral Neutral Cold
1989 Cold Neutral Neutral
1990 Cold Cold Neutral
1991 Cold Cold Neutral
1992 Warm Warm Warm
1993 Warm Warm Warm
1994 Warm Neutral Cold
1995 Neutral Warm Neutral
1996 Warm Neutral Cold
1997 Neutral Neutral Warm
1998 Warm Warm Cold
1999 Cold Cold Cold
2000 Cold Cold Cold
-
2001 Cold Cold
To determine the time periods and effects of interannual climate
anomalies of marine birds as evidenced in the study area (i.e.,
warm-water, cold-water and neutral periods), two sea-surface
temperature data sets for central California were analyzed:
daily temperatures taken as part of a Scripps Institution of
Oceanography program at Southeast Farallon Island and
the NOAA CoastWatch data off central California. Both data
50
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS gram of the Southwest Fisheries Science Center, NMFS, NOAA
Figures 42a, b, and c show the density (birds/km2) of western
Western and Clark's Grebes (unpublished data).
Aechmophorus occidentalis, A. clarkii and Clark’s grebes (combined) in the Upwelling, Oceanic, and
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Davidson Current seasons, displayed in five minute latitude by Although the at-sea data span the years from 1980 to 2001,
Oceanic Season
Upwelling Season
200 m
200 m
20
20
five minute longitude cells. Densities are based on the combined data are not available for all seasons in all years. For the
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) data sets of several studies (see “Methods” and “Data Sources” Upwelling Season, data are from 1980-1982 and 1985-2001.
39°N
39°N
below). The color and mapping intervals were customized to For the Oceanic Season, data are from 1980-1982, 1991, and
Density show the most structure and to highlight significant areas, while 1994-2001. For the Davidson Current Season, data are from
(Animals/km²)
allowing comparisons among marine bird species. Cells that 1980-1986 and 1991-2001.
> 100.00
were surveyed but in which no grebes were observed have a
50.01 - 100.00
density of zero. Areas not surveyed appear white; no informa-
38°N
38°N
METHODS
10.01 - 50.00
tion is available for these areas. Blue lines indicate the boundar-
5.01 - 10.00 At-sea densities are the result of a synthesis of data from eight
1.01 - 5.00
ies of the National Marine Sanctuaries in the study area: Cordell shipboard and aerial survey programs conducted in the study
0.51 - 1.00
Bank, Gulf of the Farallones, and Monterey Bay. area in the years 1980-2001 (see “Data Sources” below). Bird
0.11 - 0.50
observation data and trackline data from these studies were
0.06 - 0.10
37°N
37°N
In order to provide one map for the species that integrates the
0.01 - 0.05 converted to a common format. All aerial data were continuous;
0.00
patterns of its spatial and temporal occurrence and abundance ship-based data were converted separately into a continuous
0 25 50 Km
in the study area, map d shows seasonal high-use areas, dis- transect to the extent possible. From the digitized survey data,
played in 10 minute latitude by 10 minute longitude cells. The the distributions of effort and of species were mapped into five
seasonal high use map provides a further synthesis of densities minute latitude by five minute longitude cells using CDAS, a
36°N
36°N
presented in maps a, b and c, and portrays the relative impor- custom geographic information system for analyzing marine
tance of various areas to the species. Areas with consistently bird and mammal surveys (MMS, 2001). The length and width
high use are highlighted on this map. To provide a relative refer- of the survey trackline in a given cell (estimated trackline width
ence for the “high use” areas, cells are also shown where the varied by platform, depending on speed and height above wa-
species were absent (i.e., the cell was sampled but the species ter) were used to estimate the area sampled. The number of
35°N
35°N
a b was not recorded there), or present but at lesser concentrations birds of each species seen in a cell was then divided by the area
in any particular season. See the "Methods" section below for sampled in the cell to estimate density. If a cell was surveyed
Davidson Current Season Seasonal High Use Areas and
200 m
200 m
20
further explanation of seasonal high-use areas.
20
more than once, densities were averaged, with an adjustment
0
0
0m
0m
Breeding Colonies
(Nov. 15 - Mar. 14) made for effort.
39°N
39°N
DATA SOURCES
Persistence of
High Use Densities for marine birds at sea are based on data from eight The seasonal high-use areas on map d were developed using
3 Seasons
survey programs conducted between 1980 and 2001, which a similar approach as for Maps a, b and c, but the data were
2 Seasons
were combined into a new MMS-CDAS data set (MMS, 2001) binned into 10’ x 10’ cells. For each season, the cells with den-
1 Season
Birds present
using software (CDAS) developed for the Minerals Management sities in the top 20% of non-zero values were designated “high
38°N
38°N
Birds absent
Service. Of the data sets on the original MMS-CDAS CD-ROM, use” for that season. Cells were scored for “high use” in one,
four aerial survey data sets contained data in the study area two, or three seasons and are depicted by color. To provide a
from Point Arena to Point Sal. Of these, the OSPR survey relative reference for the “high use” areas, cells are also shown
program is ongoing and data from recent years were added where the species were absent (i.e., the cell was sampled but
37°N
37°N
to this data set. In addition, data from four ship-based survey the species was not recorded there) or present (but densities
programs were converted to a compatible format for analysis were never in the top 20% for any season).
(see section overview for details on individual data sets).
RESULTS AND DISCUSSION
Data sources for aerial, at-sea data include MMS-CDAS (MMS Individuals of this closely-related species pair (separable by
36°N
36°N
2001), and California Department of Fish and Game, Office of plumage, but sharing the same ecological niche) are abundant
Spill Prevention and Response (CDF&G-OSPR, unpublished in the near-shore waters of the study area. Surveys tallied
data). Early data were collected using methods described by 2,511 sightings of 13,525 individuals. During most oil spills in
Briggs et al. (1983, 1987b); more recent data were collected this region, these species have been near the top of the list, by
using updated technology but using the same general method. number, of oiled birds. These birds breed inland at freshwater
35°N
35°N
These species do not
c d Data sources for ship-based survey data include: David Ainley lakes and marshes.
breed within the study area.
of H. T. Harvey and Associates and Carol Keiper of Oikonos
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
(unpublished data; see Oedekoven et al., 2001 for details on A multiple regression model of nine independent variables ex-
Source Data: See text.
survey methods); and Lisa T. Ballance, from the Ecology Pro- plained 15.5% of variation in cell density; most important vari-
Figure 42. Western and Clark’s grebes, seasonal density and high use areas.
51
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ables were season, and an inverse relationship with distance
to land and to depth; see Table 19. These results reflect the
large number of grebes found in shallow waters (mean depth
was 131 ± 37 m) within a few kilometers of shore (mean dis-
tance to land was 7.4 ± 1 km), and primarily during the Oceanic
Season. Moderate numbers are present during the Upwelling
and Davidson Current seasons. During the latter, these grebes
expanded farther offshore to the middle continental shelf (mean
depth of occurrence 260 ± 80 m).
Inshore waters of the Gulf of the Farallones (San Francisco Bay
tidal plume), Monterey Bay, and Estero/San Luis Obispo bays
had particularly high concentrations of these birds. North and
south of marine sanctuary boundaries in the study area, these
species were found only at isolated river mouths. Therefore, the
sanctuary boundaries encompass the majority of the species
habitat in the study area, except for the ‘sanctuary exclusion
area’ off San Francisco and Pacifica, which contained many
grebes. The broad continental shelf off central California is ideal
for these grebes, which capture prey by diving; it is likely they
are capable of exploiting most of the water column lying over
the shelf, in spite of their inshore occurrence. Abundance of this
species-pair remained stable between 1985 and 2002.
These grebes feed mainly on long-bodied, fusiform fish, such
as herring and anchovy. See Tables 15 and 16 for related sum-
mary information.
52
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
data). Early data were collected using methods described by
ABOUT THESE MAPS
Northern Fulmar Briggs et al. (1983, 1987b); more recent data were collected
Figures 43 a, b, and c show the density (birds/km2) of northern
Fulmarus glacialis using updated technology but using the same general method.
fulmar in the Upwelling, Oceanic, and Davidson Current
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Data sources for ship-based survey data include: David Ainley
seasons, displayed in five minute latitude by five minute
Upwelling Season Oceanic Season
200 m
200 m
20
20
of H. T. Harvey and Associates and Carol Keiper of Oikonos
longitude cells. Densities are based on the combined data sets
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) (unpublished data; see Oedekoven et al., 2001 for details
of several studies (see “Methods” and “Data Sources” below).
39°N
39°N
on survey methods); and Lisa T. Ballance, from the Ecology
The color and mapping intervals were customized to show the
Density Program of the Southwest Fisheries Science Center, NMFS,
most structure and to highlight significant areas, while allowing
(Animals/km²)
NOAA (unpublished data).
comparisons among marine bird species. Cells that were
> 100.00
surveyed but in which no fulmars were observed have a density
50.01 - 100.00
38°N
38°N
Although the at-sea data span the years from 1980 to 2001,
of zero. Areas not surveyed appear white; no information is
10.01 - 50.00
data are not available for all seasons in all years. For the
5.01 - 10.00
available for these areas. Blue lines indicate the boundaries
1.01 - 5.00
Upwelling Season, data are from 1980-1982 and 1985-2001.
of the National Marine Sanctuaries in the study area: Cordell
0.51 - 1.00
For the Oceanic Season, data are from 1980-1982, 1991, and
Bank, Gulf of the Farallones, and Monterey Bay.
0.11 - 0.50
1994-2001. For the Davidson Current Season, data are from
0.06 - 0.10
37°N
37°N
0.01 - 0.05 1980-1986 and 1991-2001.
In order to provide one map for the species that integrates the
0.00
patterns of its spatial and temporal occurrence and abundance
0 25 50 Km
METHODS
in the study area, map d shows seasonal high-use areas,
At-sea densities are the result of a synthesis of data from eight
displayed in 10 minute latitude by 10 minute longitude cells.
shipboard and aerial survey programs conducted in the study
The seasonal high use map provides a further synthesis of
36°N
36°N
area in the years 1980-2001 (see “Data Sources” below). Bird
densities presented in Maps a, b and c, and portrays the
observation data and trackline data from these studies were
relative importance of various areas to the species. Areas
converted to a common format. All aerial data were continuous;
with consistently high use are highlighted on this map. To
ship-based data were converted separately into a continuous
provide a relative reference for the “high use” areas, cells are
transect to the extent possible. From the digitized survey data,
also shown where the species were absent (i.e., the cell was
35°N
35°N
a b the distributions of effort and of species were mapped into five
sampled but the species was not recorded there), or present
minute latitude by five minute longitude cells using CDAS, a
but at lesser concentrations in any particular season. See the
Seasonal High Use Areas and
200 m
Davidson Current Season
200 m
20
custom geographic information system for analyzing marine
20
"Methods" section below for further explanation of seasonal
0
0
0m
0m
Breeding Colonies bird and mammal surveys (MMS, 2001). The length and width
(Nov. 15 - Mar. 14) high-use areas.
39°N
39°N
of the survey trackline in a given cell (estimated trackline width
Persistence of
High Use varied by platform, depending on speed and height above
Because the sighting data for this species extends beyond
3 Seasons
water) were used to estimate the area sampled. The number
the western extent of the standard map frame shown here,
2 Seasons
of birds of each species seen in a cell was then divided by
1 Season
additional maps were made that include a greater western
Birds present
the area sampled in the cell to estimate density. If a cell was
extent. These maps (with the word "pelagic" in the filename)
38°N
38°N
Birds absent
surveyed more than once, densities were averaged, with an
are included on the CDROM.
adjustment made for effort.
DATA SOURCES
The seasonal high-use areas on map d were developed using
Densities for marine birds at sea are based on data from eight
37°N
37°N
a similar approach as for Maps a, b and c, but the data were
survey programs conducted between 1980 and 2001, which
binned into 10’ x 10’ cells. For each season, the cells with
were combined into a new MMS-CDAS data set (MMS, 2001)
densities in the top 20% of non-zero values were designated
using software (CDAS) developed for the Minerals Management
“high use” for that season. Cells were scored for “high use”
Service. Of the data sets on the original MMS-CDAS CD-ROM,
in one, two, or three seasons and are depicted by color. To
four aerial survey data sets contained data in the study area
36°N
36°N
provide a relative reference for the “high use” areas, cells are
from Point Arena to Point Sal. Of these, the OSPR survey
also shown where the species were absent (i.e., the cell was
program is ongoing and data from recent years were added
sampled but the species was not recorded there) or present
to this data set. In addition, data from four ship-based survey
(but densities were never in the top 20% for any season).
programs were converted to a compatible format for analysis
(see section overview for details on individual data sets).
35°N
35°N
This species does not
c d RESULTS AND DISCUSSION
breed within the study area.
Northern fulmar, which nests on islands in the Aleutian Island
Data sources for aerial, at-sea data include MMS-CDAS
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
chain and Bering Sea, is common in waters of the continental
(2001), and California Department of Fish and Game, Office
Source Data: See text.
slope as well as the outer waters of the continental shelf off
of Spill Prevention and Response (CDF&G-OSPR, unpublished
Figure 43. Northern fulmar, seasonal density and high use areas.
53
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
north/central California. Surveys recorded 4,486 sightings of
6,345 individuals. In some winters, fulmars were particularly
abundant off this coast, such as in 1986, 1991, 1996, and
1999.
A multiple regression analysis of nine independent variables
explained 21.3% of the variability of this species’ cell density;
important explanatory variables were season, ENSO period
(periods of unusually warm or cold sea temperatures), and year;
see Table 19. The species’ occurrence is confined principally to
the Davidson Current Season, especially prevalent during La
Niña. For a subarctic species, surprisingly high densities are
present during the Upwelling Season; many of these individuals
exhibit heavy molt indicating that they might be juveniles.
Based on the data available, the species’ population trajectory
during the study period exhibited a curvilinear pattern: a slight
decline between 1985 and 1989, followed by an increase from
1990 to 2002. Numbers rose particularly in the last few years,
perhaps indicating a response to the shift in 1999 from a warm
to a cold ocean regime (see subsection on response to climate
change).
Like the albatrosses, this species is attracted to trawlers, where
the species scavenges offal. Therefore, areas of concentration
for northern fulmars during the study period were (and may still
be) important areas of traditionally higher fishing activity such as
Cordell Bank, Fanny Shoal, and nearby canyons. This pattern
is better illustrated during the Upwelling Season, when the
species is much less abundant. In the latter season, the species
spreads far more widely, occurring farther offshore and over
deeper depths. Although fulmars are widespread off central
California, the boundaries of the National Marine Sanctuaries
encompass an important area for this species.
Northern fulmars are generalists that feed on live and dead prey
found at the surface. They are one of the few marine species
that feed extensively on gelatinous zooplankton, e.g. jellyfish.
See Tables 15 and 16 for related summary information.
54
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Data sources for aerial, at-sea data include MMS-CDAS (MMS,
ABOUT THESE MAPS
Sooty Shearwater 2001), and California Department of Fish and Game, Office of
Figures 44a, b, and c show the density (birds/km2) of sooty
Puffinus griseus Spill Prevention and Response (CDF&G-OSPR, unpublished
shearwater in the Upwelling, Oceanic, and Davidson Current
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
data). Early data were collected using methods described by
seasons, displayed in five minute latitude by five minute
Upwelling Season Oceanic Season
200 m
200 m
20
20
Briggs et al. (1983, 1987b); more recent data were collected
longitude cells. Densities are based on the combined data
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) using updated technology but using the same general method.
sets of several studies (see “Methods” and “Data Sources”
39°N
39°N
Data sources for ship-based survey data include: David Ainley
below). The color and mapping intervals were customized to
Density
of H. T. Harvey and Associates and Carol Keiper of Oikonos
show the most structure and to highlight significant areas,
(Animals/km²)
(unpublished data; see Oedekoven et al., 2001 for details
while allowing comparisons among marine bird species. Cells
> 100.00
on survey methods); and Lisa T. Ballance, from the Ecology
that were surveyed but in which no sooty shearwaters were
50.01 - 100.00
38°N
38°N
Program of the Southwest Fisheries Science Center, NMFS,
observed have a density of zero. Areas not surveyed appear
10.01 - 50.00
5.01 - 10.00
NOAA (unpublished data).
white; no information is available for these areas. Blue lines
1.01 - 5.00
indicate the boundaries of the National Marine Sanctuaries
0.51 - 1.00
Although the at-sea data span the years from 1980 to 2001,
in the study area: Cordell Bank, Gulf of the Farallones, and
0.11 - 0.50
data are not available for all seasons in all years. For the
Monterey Bay.
0.06 - 0.10
37°N
37°N
0.01 - 0.05
Upwelling Season, data are from 1980-1982 and 1985-2001.
0.00
For the Oceanic Season, data are from 1980-1982, 1991, and
In order to provide one map for the species that integrates the
0 25 50 Km
1994-2001. For the Davidson Current Season, data are from
patterns of its spatial and temporal occurrence and abundance
1980-1986 and 1991-2001.
in the study area, map d shows seasonal high-use areas,
displayed in 10 minute latitude by 10 minute longitude cells.
36°N
36°N
METHODS
The seasonal high use map provides a further synthesis of
At-sea densities are the result of a synthesis of data from eight
densities presented in Maps a, b and c, and portrays the
shipboard and aerial survey programs conducted in the study
relative importance of various areas to the species. Areas
area in the years 1980-2001 (see “Data Sources” below). Bird
with consistently high use are highlighted on this map. To
observation data and trackline data from these studies were
provide a relative reference for the “high use” areas, cells are
35°N
35°N
a b converted to a common format. All aerial data were continuous;
also shown where the species were absent (i.e., the cell was
ship-based data were converted separately into a continuous
sampled but the species was not recorded there), or present
Seasonal High Use Areas and
200 m
Davidson Current Season
200 m
20
20
transect to the extent possible. From the digitized survey data,
but at lesser concentrations in any particular season. See the
0
0
0m
0m
Breeding Colonies
(Nov. 15 - Mar. 14) the distributions of effort and of species were mapped into five
"Methods" section below for further explanation of seasonal
39°N
39°N
minute latitude by five minute longitude cells using CDAS, a
high-use areas.
Persistence of
High Use custom geographic information system for analyzing marine
3 Seasons
bird and mammal surveys (MMS, 2001). The length and width
Because the sighting data for this species extends beyond
2 Seasons
1 Season
of the survey trackline in a given cell (estimated trackline width
the western extent of the standard map frame shown here,
Birds present
varied by platform, depending on speed and height above
38°N
38°N
additional maps were made that include a greater western
Birds absent
water) were used to estimate the area sampled. The number
extent. These maps (with the word "pelagic" in the file name)
of birds of each species seen in a cell was then divided by
are included on the CDROM.
the area sampled in the cell to estimate density. If a cell was
surveyed more than once, densities were averaged, with an
DATA SOURCES
37°N
37°N
adjustment made for effort.
Densities for marine birds at sea are based on data from eight
survey programs conducted between 1980 and 2001, which
The seasonal high-use areas on map d were developed using
were combined into a new MMS-CDAS data set (MMS, 2001)
a similar approach as for Maps a, b and c, but the data were
using software (CDAS) developed for the Minerals Management
binned into 10’ x 10’ cells. For each season, the cells with
Service. Of the data sets on the original MMS-CDAS CD-ROM,
36°N
36°N
densities in the top 20% of non-zero values were designated
four aerial survey data sets contained data in the study area
“high use” for that season. Cells were scored for “high use”
from Point Arena to Point Sal. Of these, the OSPR survey
in one, two, or three seasons and are depicted by color. To
program is ongoing and data from recent years were added
provide a relative reference for the “high use” areas, cells are
to this data set. In addition, data from four ship-based survey
also shown where the species were absent (i.e., the cell was
programs were converted to a compatible format for analysis
35°N
35°N
This species does not
c d breed within the study area. sampled but the species was not recorded there) or present
(see section overview for details on individual data sets).
(but densities were never in the top 20% for any season).
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Source Data: See text.
Figure 44. Sooty shearwater, seasonal density and high use areas.
55
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
RESULTS AND DISCUSSION
Sooty shearwaters nest in the sub-Antarctic, particularly on
island of Tierra del Fuego and New Zealand, and winters in
the Peru and California current regions. During the Upwelling
Season, the sooty shearwater is the most abundant marine
bird off California, and this is the case, by far, for waters within
the boundaries of the north/central California national marine
sanctuaries. Surveys tallied 20,750 sightings of 323,176
individuals, indicating that the species usually occurs in large
concentrations.
A multiple regression analysis of nine independent variables
explained 43.3% of the variation in cell density, with season,
an inverse relationship to year, and ENSO period (periods of
unusually warm or cold sea temperatures) being the most
important variables; see Table 19. These results further reflect
the restriction of this species’ occurrence off California largely to
the Upwelling Season, and to greater abundance when ocean
climate is unaffected by short-term climate anomaly. In other
words, sooty shearwaters were less abundant in the study area
during El Niño and La Niña. From a decadal perspective they
declined over the years, although this effect was curvilinear:
a slight increase between 1985 and 1991, a steep decline to
1998, and a moderate increase subsequently. Whether or not
the latter increase is a response to the shift to a cold ocean
regime in 1999 remains to be seen. The continental shelf and
upper slope are the main habitats frequented by this species
(mean ocean depth where sooty shearwaters occurred was
380 ± 10m).
The sooty shearwater was present in greatest densities in
Monterey Bay. Throughout the California current (Veit et al,
1997), this species has declined severely in abundance during
the recent warm regime (1976-1999), as noted above. Even
now, though, it is still very abundant in Monterey Bay, probably
because of the large anchovy source there. Other important
areas (but not comparable to Monterey Bay), include Pioneer
and Ascension canyons, Farallon Escarpment and Fanny
Shoal, as well as the ocean area off Pacifica and Estero/San
Luis Obispo bays. National marine sanctuary waters become
even more important to this species during the Oceanic Season,
as remnants of the population, just before their long southward
migration, fatten on the oil-rich anchovies.
Sooty shearwaters feed on fish, squid, and invertebrates that
they acquire by pursuit, plunging to a depth of 10-15 m. During
the early Upwelling Season the main prey are euphausiids and
squid, a diet that shifts more to oily fish, such as anchovy, in
the late Upwelling Season. See Tables 15 and 16 for related
summary information.
56
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS Data sources for aerial, at-sea data include MMS-CDAS (MMS,
Ashy Storm-Petrel Figures 45a, b, and c show the density (birds/km2) of ashy 2001), California Department of Fish and Game, Office of Spill
Oceanodroma homochroa 32
storm-petrel in the Upwelling, Oceanic, and Davidson Current Prevention and Response (CDF&G-OSPR, unpublished data).
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
seasons, displayed in five minute latitude by five minute Early data were collected using methods described by Briggs
Upwelling Season Oceanic Season
200 m
200 m
20
20
longitude cells. Densities are based on the combined data et al. (1983, 1987b); more recent data were collected using
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) sets of several studies (see “Methods” and “Data Sources” updated technology but using the same general method. Data
39°N
39°N
below). The color and mapping intervals were customized sources for ship-based survey data include: David Ainley of
Density to show the most structure and to highlight significant areas, H. T. Harvey and Associates and Carol Keiper of Oikonos
(Animals/km²)
while allowing comparisons among marine bird species. Cells (unpublished data; see Oedekoven et al., 2001 for details
> 100.00
that were surveyed but in which no ashy storm-petrels were on survey methods); and Lisa T. Ballance, from the Ecology
50.01 - 100.00
38°N
38°N
observed have a density of zero. Areas not surveyed appear Program of the Southwest Fisheries Science Center, NMFS,
10.01 - 50.00
white; no information is available for these areas. Blue lines NOAA (unpublished data). Data on ashy storm-petrel colonies
5.01 - 10.00
1.01 - 5.00
indicate the boundaries of the National Marine Sanctuaries were obtained from Carter et al. (1992) supplemented by
0.51 - 1.00
in the study area: Cordell Bank, Gulf of the Farallones, and Sydeman et al. (1998), Whitworth et al. (2002).
0.11 - 0.50
Monterey Bay.
0.06 - 0.10
37°N
37°N
Although the at-sea data span the years from 1980 to 2001,
0.01 - 0.05
0.00
In order to provide one map for the species that integrates the data are not available for all seasons in all years. For the
0 25 50 Km
patterns of its spatial and temporal occurrence and abundance Upwelling Season, data are from 1980-1982 and 1985-2001.
in the study area, map d shows seasonal high-use areas, For the Oceanic Season, data are from 1980-1982, 1991 and
displayed in 10 minute latitude by 10 minute longitude cells, 1994-2001. For the Davidson Current Season, data are from
36°N
36°N
and breeding colonies. The seasonal high use map provides 1980-1986 and 1991-2001.
a further synthesis of densities presented in Maps a, b and c,
and portrays the relative importance of various areas to the METHODS
species. Areas with consistently high use are highlighted on this At-sea densities are the result of a synthesis of data from eight
map. To provide a relative reference for the “high use” areas, shipboard and aerial survey programs conducted in the study
35°N
35°N
a b cells are also shown where the species were absent (i.e., the area in the years 1980-2001 (see “Data Sources” below). Bird
cell was sampled but the species was not recorded there), or observation data and trackline data from these studies were
Seasonal High Use Areas and
200 m
Davidson Current Season
200 m
20
present but at lesser concentrations in any particular season. converted to a common format. All aerial data were continuous;
20
0
0
0m
0m
Breeding Colonies See the "Methods" section below for further explanation of ship-based data were converted separately into a continuous
(Nov. 15 - Mar. 14)
39°N
39°N
seasonal high-use areas. Breeding colonies are also shown; transect to the extent possible. From the digitized survey data,
Persistence of
High Use the relative size of the symbols indicates the colony size. the distributions of effort and of species were mapped into five
3 Seasons
minute latitude by five minute longitude cells using CDAS, a
2 Seasons
1 Season
Because the sighting data for this species extends beyond custom geographic information system for analyzing marine
Birds present
the western extent of the standard map frame shown here, bird and mammal surveys (MMS, 2001). The length and width
38°N
38°N
Birds absent
additional maps were made that include a greater western of the survey trackline in a given cell (estimated trackline width
Colony Size
(Breeding birds)
extent. These maps (with the word "pelagic" in the filename) varied by platform, depending on speed and height above
10,000 - 50,000
are included on the CDROM. water) were used to estimate the area sampled. The number
of birds of each species seen in a cell was then divided by
5001 - 10,000
37°N
37°N
DATA SOURCES the area sampled in the cell to estimate density. If a cell was
1001 - 5000
501 - 1000
Densities for marine birds at sea are based on data from eight surveyed more than once, densities were averaged, with an
101 - 500
survey programs conducted between 1980 and 2001, which adjustment made for effort.
51 - 100
2 - 50
were combined into a new MMS-CDAS data set (MMS, 2001)
Historical
using software (CDAS) developed for the Minerals Management The seasonal high-use areas on map d were developed using
36°N
36°N
Service. Of the data sets on the original MMS-CDAS CD-ROM, a similar approach as for Maps a, b and c, but the data were
four aerial survey data sets contained data in the study area binned into 10’x10’ cells. For each season, the cells with
from Point Arena to Point Sal. Of these, the OSPR survey densities in the top 20% of non-zero values were designated
program is ongoing and data from recent years were added “high use” for that season. Cells were scored for “high use”
to this data set. In addition, data from four ship-based survey in one, two, or three seasons and are depicted by color. To
35°N
35°N
c d programs were converted to a compatible format for analysis provide a relative reference for the “high use” areas, cells are
(see section overview for details on individual data sets). also shown where the species were absent (i.e., the cell was
124°W 123°W 122°W 121°W
124°W 123°W 122°W 121°W
sampled but the species was not recorded there) or present
Source Data: See text.
(but densities were never in the top 20% for any season).
Figure 45. Ashy storm-petrel, seasonal density, high use areas, and breeding colonies.
57
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
RESULTS AND DISCUSSION
Ashy storm-petrel is endemic to the California Current and is
considered by the State to be a “Species of Special Concern”;
a major colony is at the Farallon Islands. It is common in the
study area and the most abundant storm-petrel in waters of
the central California national marine sanctuaries. Surveys
recorded 1,472 sightings of 4,339 individuals.
A multiple regression model of nine variables explained 17.3%
of variation in cell density, with important explanatory variables
being ENSO period (i.e., periods of unusually warm or cold ocean
temperature), season, and year; see Table 19. The species was
more abundant during the Oceanic Season and during years of
La Niña, indicating that when ocean temperatures were cold,
Ashy storm-petrels were concentrated closer to the Farallon
breeding colony, which they visit only at night. During nesting
(Upwelling Season), this species occupies waters mainly over
the outer slope (mean depth of occurrence 1,615 ± 52 m),
mostly outside of National Marine Sanctuary boundaries. During
the period of molt (Oceanic Season), ashy storm-petrels move
inshore to frequent shallower slope waters (mean depth of
occurrence 1,144 ± 61 m) and a large concentration occurred
over the Monterey Bay canyon as shown in maps on upwelling
and seasonal high use areas.
In recent years, however, this post-breeding concentration
has shifted to the area around Cordell Bank (not shown on
the maps). As the species begins its seasonal return to the
Farallon nesting colony (Davidson Current Season), they again
shift north to deeper waters of the outer slope (mean depth
of occurrence then was 2,579 ± 121 m). The species seems
to be most dispersed during the Davidson Current Season,
but in all seasons the Farallon Escarpment is by far its most
important area.
Overall, ashy storm-petrel numbers increased from 1985 to
2002 in a curvilinear fashion: steeper increase in numbers
between 1985 and 1992, followed by a less steep increase
to 2002.
This species feeds on invertebrates and larval fish found
at the surface. See Tables 15 and 16 for related summary
information.
58
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS programs were converted to a compatible format for analysis
Figures 46a, b, and c show the density (birds/km2) of Leach’s (see section overview for details on individual data sets).
Leach's Storm-Petrel Oceanodroma leucorhoa storm-petrel in the Upwelling, Oceanic, and Davidson Current
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
seasons, displayed in five minute latitude by five minute Data sources for aerial, at-sea data include MMS-CDAS (MMS,
Upwelling Season Oceanic Season
200 m
longitude cells. Densities are based on the combined data 2001), California Department of Fish and Game, Office of Spill
200 m
20
20
0
0
0m
0m
sets of several studies (see “Methods” and “Data Sources” Prevention and Response (CDF&G-OSPR, unpublished data).
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14)
39°N
39°N
below). The color and mapping intervals were customized to Early data were collected using methods described by Briggs
Density show the most structure and to highlight significant areas, while et al. (1983, 1987b); more recent data were collected using
(Animals/km²)
allowing comparisons among marine bird species. Cells that updated technology but using the same general method.
> 100.00
were surveyed but in which no Leach’s storm-petrels were Data sources for ship-based survey data include: David
50.01 - 100.00
observed have a density of zero. Areas not surveyed appear Ainley of H. T. Harvey and Associates and Carol Keiper of
38°N
38°N
10.01 - 50.00
white; no information is available for these areas. Blue lines Oikonos (unpublished data; see Oedekoven et al., 2001 for
5.01 - 10.00
1.01 - 5.00
indicate the boundaries of the National Marine Sanctuaries details on survey methods); and Lisa T. Ballance, from the
0.51 - 1.00
in the study area: Cordell Bank, Gulf of the Farallones, and Ecology Program of the Southwest Fisheries Science Center,
0.11 - 0.50
Monterey Bay. An additional set of maps was done for this NMFS, NOAA (unpublished data). Data on Leach’s storm-
0.06 - 0.10
37°N
37°N
species to show the offshore extent of its distribution; these petrel colonies were obtained from Carter et al. (1992, and
0.01 - 0.05
0.00 maps are included on the CD-ROM. supplements).
0 25 50 Km
In order to provide one map for the species that integrates the Although the at-sea data span the years from 1980 to 2001,
patterns of its spatial and temporal occurrence and abundance data are not available for all seasons in all years. For the
36°N
36°N
in the study area, map d shows seasonal high-use areas, Upwelling Season, data are from 1980-1982 and 1985-2001.
displayed in 10 minute latitude by 10 minute longitude cells, For the Oceanic Season, data are from 1980-1982, 1991, and
and breeding colonies. The seasonal high use map provides a 1994-2001. For the Davidson Current Season, data are from
further synthesis of densities presented in Maps a, b and c, and 1980-1986 and 1991-2001.
portrays the relative importance of various areas to the species.
35°N
35°N
a b Areas with consistently high use are highlighted on this map. To METHODS
provide a relative reference for the “high use” areas, cells are At-sea densities are the result of a synthesis of data from eight
Seasonal High Use Areas and also shown where the species were absent (i.e., the cell was shipboard and aerial survey programs conducted in the study
200 m
Davidson Current Season
200 m
20
20
0
0
0m
0m
sampled but the species was not recorded there), or present area in the years 1980-2001 (see “Data Sources” below). Bird
Breeding Colonies
(Nov. 15 - Mar. 14)
39°N
39°N
but at lesser concentrations in any particular season. See the observation data and trackline data from these studies were
Persistence of
"Methods" section below for further explanation of seasonal converted to a common format. All aerial data were continuous;
High Use
3 Seasons
high-use areas. Breeding colonies are also shown; the relative ship-based data were converted separately into a continuous
2 Seasons
size of the symbols indicates the colony size. An additional set transect to the extent possible. From the digitized survey data,
1 Season
Birds present
of maps was developed for this species to include the offshore the distributions of effort and of species were mapped into five
38°N
38°N
Birds absent
extent of its distribution. These maps are on the CD-ROM. minute latitude by five minute longitude cells using CDAS, a
Colony Size
(Breeding birds) custom geographic information system for analyzing marine
10,000 - 50,000
Because the sighting data for this species extends beyond bird and mammal surveys (MMS, 2001). The length and width
the western extent of the standard map frame shown here, of the survey trackline in a given cell (estimated trackline width
5001 - 10,000
37°N
37°N
additional maps were made that include a greater western varied by platform, depending on speed and height above
1001 - 5000
501 - 1000
extent. These maps (with the word "pelagic" in the filename) water) were used to estimate the area sampled. The number
101 - 500
are included on the CDROM. of birds of each species seen in a cell was then divided by
51 - 100
2 - 50
the area sampled in the cell to estimate density. If a cell was
Historical
DATA SOURCES surveyed more than once, densities were averaged, with an
36°N
36°N
Densities for marine birds at sea are based on data from eight adjustment made for effort.
survey programs conducted between 1980 and 2001, which
were combined into a new MMS-CDAS data set (MMS, 2001) The seasonal high-use areas on map d were developed using
using software (CDAS) developed for the Minerals Management a similar approach as for Maps a, b and c, but the data were
Service. Of the data sets on the original MMS-CDAS CD-ROM, binned into 10’x10’ cells. For each season, the cells with
35°N
35°N
c d four aerial survey data sets contained data in the study area densities in the top 20% of non-zero values were designated
from Point Arena to Point Sal. Of these, the OSPR survey “high use” for that season. Cells were scored for “high use”
124°W 123°W 122°W 121°W
124°W 123°W 122°W 121°W
program is ongoing and data from recent years were added in one, two, or three seasons and are depicted by color. To
Source Data: See text.
to this data set. In addition, data from four ship-based survey provide a relative reference for the “high use” areas, cells are
Figure 46. Leach’s storm-petrel, seasonal density, high use areas, and breeding colonies.
59
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
also shown where the species were absent (i.e., the cell was
sampled but the species was not recorded there) or present
(but densities were never in the top 20% for any season).
RESULTS AND DISCUSSION
The Leach’s storm-petrel, which has a breeding population
numbering in the millions in Alaska, is represented south to
Baja California by smaller colonies as latitude decreases. In
comparison, the estimated 12,551 birds breeding along the
California coast is miniscule (Carter et al, 1992). This, and
the fact that this species is highly migratory, suggests that
many of the birds seen in the study areas are migrants from
the north. This was also indicated by the lack of importance in
a multiple regression model of distance to colony as a factor
explaining this species’ variation in cell density; see Table 19.
Surveys recorded 1,118 sightings of 1,576 individuals, although
survey effort was sparse in the offshore waters this species
frequents.
This common species frequents waters much farther offshore
than the other storm-petrels, i.e. well beyond the continental
slope. Thus, the National Marine Sanctuary boundaries (and
most of the data sets in this study) do not encompass much of
this species’ preferred habitat. The species was most abundant
during the Upwelling Season (breeding) and occurred in greater
numbers closer to the coast. They visit the Farallon colony only
at night, but are at-sea during the day. During the Oceanic
and Davidson Current seasons few occurred near the shelf.
The birds present during the latter two seasons likely were
migrants from more northern populations. Given the huge North
Pacific population, the number recorded during surveys in the
study area was relatively small, because they were mostly far
offshore and not observed as often in the surveys available
for this assessment.
Yet, a multiple regression model of nine independent variables
explained 28.4% of variation in cell density, indicating that this
species responded consistently to the variables examined.
Most important of the nine variables were season, distance to
the 2000 m isobath, and ENSO period (periods of unusually
warm or cold ocean temperature); see Table 19. Abundance
of this species in the study area has increased between 1985
and 2002, and it was more abundant during periods of warm-
water conditions.
This species feeds on invertebrates captured at the surface.
See Tables 15 and 16 for related summary information.
60
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS Lisa T. Ballance, from the Ecology Program of the Southwest
Figures 47a, b, and c show the combined density (birds/km2) Fisheries Science Center, NMFS, NOAA (unpublished data).
Black, Surf and White-winged Scoters Melanitta nigra, M. perspicillata, M. fusca of three scoter species (white-winged, surf, and black) in the
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Upwelling, Oceanic, and Davidson Current seasons, displayed Although the at-sea data span the years from 1980 to 2001,
Upwelling Season Oceanic Season
200 m
200 m
20
20
in five minute latitude by five minute longitude cells. Densities data are not available for all seasons in all years. For the
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) are based on the combined data sets of several studies (see Upwelling Season, data are from 1980-1982 and 1985-2001.
39°N
39°N
“Methods” and “Data Sources” below). The color and mapping For the Oceanic Season, data are from 1980-1982, 1991, and
Density
intervals were customized to show the most structure and to 1994-2001. For the Davidson Current Season, data are from
(Animals/km²)
highlight significant areas, while allowing comparisons among 1980-1986 and 1991-2001.
> 100.00
marine bird species. Cells that were surveyed but in which
50.01 - 100.00
38°N
38°N
no scoters were observed have a density of zero. Areas not METHODS
10.01 - 50.00
5.01 - 10.00
surveyed appear white; no information is available for these At-sea densities are the result of a synthesis of data from eight
1.01 - 5.00
areas. Blue lines indicate the boundaries of the National Marine shipboard and aerial survey programs conducted in the study
0.51 - 1.00
Sanctuaries in the study area: Cordell Bank, Gulf of the Faral- area in the years 1980-2001 (see “Data Sources” below). Bird
0.11 - 0.50
lones, and Monterey Bay. observation data and trackline data from these studies were
0.06 - 0.10
37°N
37°N
0.01 - 0.05
converted to a common format. All aerial data were continuous;
0.00
In order to provide one map for the species that integrates the ship-based data were converted separately into a continuous
0 25 50 Km
patterns of its spatial and temporal occurrence and abundance transect to the extent possible. From the digitized survey data,
in the study area, map d shows seasonal high-use areas, dis- the distributions of effort and of species were mapped into five
played in 10 minute latitude by 10 minute longitude cells. The minute latitude by five minute longitude cells using CDAS, a
36°N
36°N
seasonal high use map provides a further synthesis of densities custom geographic information system for analyzing marine
presented in Maps a, b and c, and portrays the relative impor- bird and mammal surveys (MMS, 2001). The length and width
tance of various areas to the species. Areas with consistently of the survey trackline in a given cell (estimated trackline width
high use are highlighted on this map. To provide a relative refer- varied by platform, depending on speed and height above wa-
ence for the “high use” areas, cells are also shown where the ter) were used to estimate the area sampled. The number of
35°N
35°N
a b species were absent (i.e., the cell was sampled but the species birds of each species seen in a cell was then divided by the area
was not recorded there), or present but at lesser concentrations sampled in the cell to estimate density. If a cell was surveyed
Seasonal High Use Areas and
200 m
Davidson Current Season
200 m
20
20
in any particular season. See the "Methods" section below for more than once, densities were averaged, with an adjustment
0
0
0m
0m
Breeding Colonies
(Nov. 15 - Mar. 14) further explanation of seasonal high-use areas. made for effort.
39°N
39°N
Persistence of
High Use DATA SOURCES The seasonal high-use areas on map d were developed using
3 Seasons
Densities for marine birds at sea are based on data from eight a similar approach as for Maps a, b and c, but the data were
2 Seasons
1 Season
survey programs conducted between 1980 and 2001, which binned into 10’x10’ cells. For each season, the cells with densi-
Birds present
were combined into a new MMS-CDAS data set (MMS, 2001) ties in the top 20% of non-zero values were designated “high
38°N
38°N
Birds absent
using software (CDAS) developed for the Minerals Manage- use” for that season. Cells were scored for “high use” in one,
ment Service. Of the data sets on the original MMS-CDAS CD- two, or three seasons and are depicted by color. To provide a
ROM, four aerial survey data sets contained data in the study relative reference for the “high use” areas, cells are also shown
area from Point Arena to Point Sal.. Of these, the OSPR survey where the species were absent (i.e., the cell was sampled but
37°N
37°N
program is ongoing and data from recent years were added the species was not recorded there) or present (but densities
to this data set. In addition, data from four ship-based survey were never in the top 20% for any season).
programs were converted to a compatible format for analysis
(see section overview for details on individual data sets). RESULTS AND DISCUSSION
The distribution of white-winged, surf, and black scoters in
36°N
36°N
Data sources for aerial, at-sea data include MMS-CDAS (MMS, the north/central California study area is very similar to that
2001), California Department of Fish and Game, Office of Spill of the grebes (see above), although they are somewhat less
Prevention and (CDF&G-OSPR, unpublished data). Early data abundant and found closer to shore. There they forage mostly
were collected using methods described by Briggs et al. (1983, just outside the surf break. On the outer coast, the abundant
1987b); more recent data were collected using updated tech- surf scoter dominates over the other two scoters, and black
35°N
35°N
These species do not
c d nology but using the same general method. Data sources for scoters, which occur in more protected waters, are rare. Sur-
breed within the study area.
ship-based survey data include: David Ainley of H. T. Harvey veys recorded 1,787 sightings of scoters that included 42,691
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
and Associates and Carol Keiper of Oikonos (unpublished data; individuals; more than half were identified as surf scoter. The
Source Data: See text.
see Oedekoven et al., 2001 for details on survey methods); and most important areas for surf scoters within the study area is
Figure 47. Black, surf and white-winged scoters, seasonal density and high use areas.
61
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
the San Francisco Bay tidal plume, especially southward along
the Pacifica shore to Half Moon Bay, and the shallow parts of
Bodega, Monterey, Estero, and San Luis Obispo bays. These
birds nest in the arctic tundra along the north slope of North
America; specific nesting areas of birds found wintering in the
marine sanctuary boundaries have not been identified.
The apparent movement of these sea ducks’ offshore, i.e. to
the outer parts of the Gulf of the Farallones, in the Upwelling
Season is an artifact of their migration north or south, to or from
Alaskan breeding areas. That portion of the population wintering
south of central California takes the shortest distance across
the Gulf of the Farallones; the offshore density cells highlighted
in the maps is a record of flying scoters.
These scoters do not forage far from the mainland beach, where
they eat invertebrates; several dozen usually winter around the
Farallon Islands. The inshore distribution of these ducks, like
the grebes, makes them vulnerable to coastal oil spills. See
Tables 15 and 16 for related summary information.
62
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS of H. T. Harvey and Associates and Carol Keiper of Oikonos
Brown Pelican Figures 48a, b, and c show the density (birds/km2) of brown (unpublished data; see Oedekoven et al., 2001 for details
Pelecanus occidentalis pelicans in the Upwelling, Oceanic, and Davidson Current on survey methods); and Lisa T. Ballance, from the Ecology
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
seasons, displayed in five minute latitude by five minute Program of the Southwest Fisheries Science Center, NMFS,
Oceanic Season
Upwelling Season
200 m
200 m
20
20
longitude cells. Densities are based on the combined data NOAA (unpublished data). Data on brown pelican colonies were
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) sets of several studies (see “Methods” and “Data Sources” obtained from Carter et al. (1992, and supplements).
39°N
39°N
below). The color and mapping intervals were customized to
Density
show the most structure and to highlight significant areas, while Although the at-sea data span the years from 1980 to 2001,
(Animals/km²)
allowing comparisons among marine bird species. Cells that data are not available for all seasons in all years. For the
> 100.00
were surveyed but in which no brown pelicans were observed Upwelling Season, data are from 1980-1982 and 1985-2001.
50.01 - 100.00
38°N
38°N
have a density of zero; areas not surveyed are white. Blue lines For the Oceanic Season, data are from 1980-1982, 1991 and
10.01 - 50.00
5.01 - 10.00 indicate the boundaries of the National Marine Sanctuaries 1994-2001. For the Davidson Current Season, data are from
1.01 - 5.00
in the study area: Cordell Bank, Gulf of the Farallones, and 1980-1986 and 1991-2001.
0.51 - 1.00
Monterey Bay.
0.11 - 0.50
METHODS
0.06 - 0.10
37°N
37°N
0.01 - 0.05
In order to provide one map for the species that integrates the At-sea densities are the result of a synthesis of data from eight
0.00
patterns of its spatial and temporal occurrence and abundance shipboard and aerial survey programs conducted in the study
0 25 50 Km
in the study area, map d shows seasonal high-use areas, area in the years 1980-2001 (see “Data Sources” below). Bird
displayed in 10 minute latitude by 10 minute longitude cells, observation data and trackline data from these studies were
and breeding colonies (in this species’ case, a site where it converted to a common format. All aerial data were continuous;
36°N
36°N
bred in the past). The seasonal high use map provides a ship-based data were converted separately into a continuous
further synthesis of densities presented in Maps a, b and c, transect to the extent possible. From the digitized survey data,
and portrays the relative importance of various areas to the the distributions of effort and of species were mapped into five
species. Areas with consistently high use are highlighted on this minute latitude by five minute longitude cells using CDAS, a
map. To provide a relative reference for the “high use” areas, custom geographic information system for analyzing marine
35°N
35°N
a b cells are also shown where the species were absent (i.e., the bird and mammal surveys (MMS, 2001). The length and width
cell was sampled but the species was not recorded there), or of the survey trackline in a given cell (estimated trackline width
Seasonal High Use Areas and
Davidson Current Season
200 m
200 m
20
20
present but at lesser concentrations in any particular season. varied by platform, depending on speed and height above
0
0
0m
0m
Breeding Colonies
(Nov. 15 - Mar. 14) See the "Methods" section below for further explanation of water) were used to estimate the area sampled. The number
39°N
39°N
seasonal high-use areas. Breeding colonies are also shown; of birds of each species seen in a cell was then divided by
Persistence of
High Use the relative size of the symbols indicates the colony size. the area sampled in the cell to estimate density. If a cell was
3 Seasons
surveyed more than once, densities were averaged, with an
2 Seasons
1 Season
DATA SOURCES adjustment made for effort.
Birds present
38°N
38°N
Densities for marine birds at sea are based on data from eight
Birds absent
survey programs conducted between 1980 and 2001, which The seasonal high-use areas on map d were developed using
were combined into a new MMS-CDAS data set (MMS, 2001) a similar approach as for Maps a, b and c, but the data were
using software (CDAS) developed for the Minerals Management binned into 10’x10’ cells. For each season, the cells with
Service. Of the data sets on the original MMS-CDAS CD-ROM, densities in the top 20% of non-zero values were designated
37°N
37°N
four aerial survey data sets contained data in the study area “high use” for that season. Cells were scored for “high use”
from Point Arena to Point Sal. Of these, the OSPR survey in one, two, or three seasons and are depicted by color. To
program is ongoing and data from recent years were added provide a relative reference for the “high use” areas, cells are
to this data set. In addition, data from four ship-based survey also shown where the species were absent (i.e., the cell was
programs were converted to a compatible format for analysis sampled but the species was not recorded there) or present
36°N
36°N
(see section overview for details on individual data sets). (but densities were never in the top 20% for any season).
Data sources for aerial, at-sea data include MMS-CDAS (MMS, RESULTS AND DISCUSSION
2001), and California Department of Fish and Game, Office of Brown pelicans are included in the State and Federal
Spill Prevention and Response (CDF&G-OSPR, unpublished endangered species lists, and are common year-round in
35°N
35°N
This species does not
c d breed within the study area. data). Early data were collected using methods described by Monterey Bay and to the south. Surveys recorded 1,447
Briggs et al. (1983, 1987b); more recent data were collected sightings of 3,003 individuals.
124°W 123°W 122°W 121°W
124°W 123°W 122°W 121°W
Source Data: See text. using updated technology but using the same general method.
Data sources for ship-based survey data include: David Ainley
Figure 48. Brown pelican, seasonal density and high use areas.
63
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
This population breeds on selected islands off Baja Mexico
and southern California, with small colonies extending north
to the Channel Islands. Brown pelicans once bred on rocks off
Monterey, but are now concentrated in the central California
study area at roosts such as Morro Rock, Monterey Breakwater,
Año Nuevo Island, Southeast Farallon Island, Bird Rock in
Monterey county, and Bodega Rock. Nesting occurs in southern
California and Baja Mexico and begins in November and can
extend through June, when the species is most sparse in
the central California study area. The brown pelican is most
abundant in the study area during the Oceanic Season; the
species’ presence then constitutes a post-breeding increase
from southern breeding grounds.
North of Monterey and Estero/San Luis Obispo bays, this
species’ presence is much more seasonal and dependent on
ocean climate. Most sightings in the Gulf of the Farallones
during the Davidson Current and Upwelling seasons occurred
during warm-water years, often associated with the species
choosing to forego breeding at southern colonies. Thus,
wintering birds may remain in central California waters, while
others may move farther north than usual at that time. In most
cool-or coldwater years, adults are not abundant north of
Monterey Bay during these two seasons. This could change,
however, as sardines, an important prey item, continue to
increase in California waters.
The species frequents waters within several miles of shore
(mean distance to land was 10.3 ± 0.4 km) and rarely occurs
in waters deeper than the shelf break (mean depth was
266 ± 21 m). Consistent with these patterns are results of
a multiple regression model of nine independent variables,
which explained 15.2% of the variation; important variables
were season, and inverse relationships to distance to land and
latitude; see Table 19. Therefore, the broad shelf of central
California is well suited to this species; its occurrence becomes
sporadic north of Point Reyes. Inshore Monterey, Estero, and
San Luis Obispo bays are especially important, where this
species is common year round; the San Francisco Bay tidal
plume is also important. Abundance of this species in the study
area has increased between 1985 and 2002.
This species preys exclusively on fish, especially anchovies,
mackerel, and sardines, that it catches by plunging to just
below the surface. See Tables 15 and 16 for related summary
information.
64
Section 2.3: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS on survey methods); and Lisa T. Ballance, from the Ecology
Figures 49a, b, and c show the density (birds/km2) of black- Program of the Southwest Fisheries Science Center, NMFS,
Black-legged Kittiwake Rissa tridactyla legged kittiwakes in the Upwelling, Oceanic, and Davidson NOAA (unpublished data).
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Current seasons, displayed in five minute latitude by five minute
Oceanic Season
Upwelling Season
200 m
longitude cells. Densities are based on the combined data sets Although the at-sea data span the years from 1980 to 2001,
200 m
20
20
0
0
0m
0m
of several studies (see “Methods” and “Data Sources” below). data are not available for all seasons in all years. For the
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14)
39°N
39°N
The color and mapping intervals were customized to show Upwelling Season, data are from 1980-1982 and 1985-2001.
Density the most structure and to highlight significant areas, while For the Oceanic Season, data are from 1980-1982, 1991, and
(Animals/km²)
allowing comparisons among marine bird species. Cells that 1994-2001. For the Davidson Current Season, data are from
> 100.00
were surveyed but in which no black-legged kittiwakes were 1980-1986 and 1991-2001.
50.01 - 100.00
observed have a density of zero. Areas not surveyed appear
38°N
38°N
10.01 - 50.00
white; no information is available for these areas. Blue lines METHODS
5.01 - 10.00
1.01 - 5.00 indicate the boundaries of the National Marine Sanctuaries At-sea densities are the result of a synthesis of data from eight
0.51 - 1.00
in the study area: Cordell Bank, Gulf of the Farallones, and shipboard and aerial survey programs conducted in the study
0.11 - 0.50
Monterey Bay. area in the years 1980-2001 (see “Data Sources” below). Bird
0.06 - 0.10
37°N
37°N
observation data and trackline data from these studies were
0.01 - 0.05
0.00 In order to provide one map for the species that integrates the converted to a common format. All aerial data were continuous;
0 25 50 Km
patterns of its spatial and temporal occurrence and abundance ship-based data were converted separately into a continuous
in the study area, map d shows seasonal high-use areas, dis- transect to the extent possible. From the digitized survey data,
played in 10 minute latitude by 10 minute longitude cells. The the distributions of effort and of species were mapped into five
36°N
36°N
seasonal high use map provides a further synthesis of densities minute latitude by five minute longitude cells using CDAS, a
presented in Maps a, b and c, and portrays the relative impor- custom geographic information system for analyzing marine
tance of various areas to the species. Areas with consistently bird and mammal surveys (MMS, 2001). The length and width
high use are highlighted on this map. To provide a relative refer- of the survey trackline in a given cell (estimated trackline width
ence for the “high use” areas, cells are also shown where the varied by platform, depending on speed and height above wa-
35°N
35°N
a b species were absent (i.e., the cell was sampled but the species ter) were used to estimate the area sampled. The number of
was not recorded there), or present but at lesser concentrations birds of each species seen in a cell was then divided by the area
in any particular season. See the "Methods" section below for sampled in the cell to estimate density. If a cell was surveyed
Davidson Current Season Seasonal High Use Areas and
200 m
200 m
20
20
0
0
0m
0m
further explanation of seasonal high-use areas. more than once, densities were averaged, with an adjustment
Breeding Colonies
(Nov. 15 - Mar. 14)
39°N
39°N
made for effort.
Persistence of
DATA SOURCES
High Use
3 Seasons
Densities for marine birds at sea are based on data from eight The seasonal high-use areas on map d were developed using
2 Seasons
survey programs conducted between 1980 and 2001, which a similar approach as for Maps a, b and c, but the data were
1 Season
Birds present
were combined into a new MMS-CDAS data set (MMS, 2001) binned into 10’x10’ cells. For each season, the cells with densi-
38°N
38°N
Birds absent
using software (CDAS) developed for the Minerals Manage- ties in the top 20% of non-zero values were designated “high
ment Service. Of the data sets on the original MMS-CDAS CD- use” for that season. Cells were scored for “high use” in one,
ROM, four aerial survey data sets contained data in the study two, or three seasons and are depicted by color. To provide a
area from Point Arena to Point Sal. Of these, the OSPR survey relative reference for the “high use” areas, cells are also shown
37°N
37°N
program is ongoing and data from recent years were added where the species were absent (i.e., the cell was sampled but
to this data set. In addition, data from four ship-based survey the species was not recorded there) or present (but densities
programs were converted to a compatible format for analysis were never in the top 20% for any season).
(see section overview for details on individual data sets).
RESULTS AND DISCUSSION
36°N
36°N
Data sources for aerial, at-sea data include MMS-CDAS (MMS, The black-legged kittiwake, like the northern fulmar, breeds on
2001), and California Department of Fish and Game, Office of islands along the northern coast of North America and Asia, but
Spill Prevention and Response (CDF&G-OSPR, unpublished large numbers ‘winter’ in the study area off central California. It
data). Early data were collected using methods described by is a common species in the study area; surveys recorded 2,079
Briggs et al. (1983, 1987b); more recent data were collected sightings of 5,003 individuals. A multiple-regression model of
35°N
35°N
This species does not
c d using updated technology but using the same general method. eight independent variables explained 28.9% of variation in
breed within the study area.
Data sources for ship-based survey data include: David Ainley cell density; important variables were season, ENSO period
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
of H. T. Harvey and Associates and Carol Keiper of Oikonos (i.e., periods of climatic variation), and year (increasing abun-
Source Data: See text.
(unpublished data; see Oedekoven et al., 2001 for details dance). The black-legged kittiwake was most abundant in the
Figure 49. Black-legged kittiwake, seasonal density and high use areas.
65
Section 2.3: BIOGEOGRAPHY OF MARINE BIRDS
study area during the Davidson Current Season and less so
during the early Upwelling Season; it was largely absent during
the late-Upwelling Season and Oceanic Season (which cor-
responds to the breeding season at northern-latitude nesting
sites). Abundance was highest during periods of La Niña. Most
kittiwakes occurred in waters overlying the continental slope,
and deeper waters seaward of National Marine Sanctuary
boundaries (mean depth of occurrence was 1,408 m; mean
distance from shore was 29.0 km). A minority of kittiwakes oc-
curred over the shelf, mainly where the shelf is narrow. There
was an “invasion” of kittiwakes in 1999, coincident with the
beginning of the cold-water regime shift (see below).
This species feeds on fish and pelagic invertebrates that they
catch by dipping and plunging to the surface. No studies of
kittiwake diet at sea were available. See Tables 15 and 16 for
related summary information.
66
Section 2.3: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS Data sources for ship-based survey data include: David Ainley
Common Murre Figures 50a, b, and c show the density (birds/km2) of common of H. T. Harvey and Associates and Carol Keiper of Oikonos
Uria aalge murre in the Upwelling, Oceanic, and Davidson Current (unpublished data; see Oedekoven et al., 2001 for details
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
seasons, displayed in five minute latitude by five minute on survey methods); and Lisa T. Ballance, from the Ecology
Upwelling Season Oceanic Season
200 m
200 m
20
20
longitude cells. Densities are based on the combined data Program of the Southwest Fisheries Science Center, NMFS,
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) sets of several studies (see “Methods” and “Data Sources” NOAA (unpublished data). Data on common murre colonies
39°N
39°N
below). The color and mapping intervals were customized to were obtained from Carter et al. (1992), with updates for Devil’s
Density
show the most structure and to highlight significant areas, Slide (Gerry McChesney, USFWS, pers. comm) and South
(Animals/km²)
while allowing comparisons among marine bird species. Cells Farallon Island (Warzybok et al. 2002).
> 100.00
that were surveyed but in which no common murres were
50.01 - 100.00
38°N
38°N
observed have a density of zero. Areas not surveyed appear Although the at-sea data span the years from 1980 to 2001,
10.01 - 50.00
5.01 - 10.00 white; no information is available for these areas. Blue lines data are not available for all seasons in all years. For the
1.01 - 5.00
indicate the boundaries of the National Marine Sanctuaries Upwelling Season, data are from 1980-1982 and 1985-2001.
0.51 - 1.00
in the study area: Cordell Bank, Gulf of the Farallones, and For the Oceanic Season, data are from 1980-1982, 1991, and
0.11 - 0.50
Monterey Bay. 1994-2001. For the Davidson Current Season, data are from
0.06 - 0.10
37°N
37°N
0.01 - 0.05
1980-1986 and 1991-2001.
0.00
In order to provide one map for the species that integrates the
0 25 50 Km
patterns of its spatial and temporal occurrence and abundance METHODS
in the study area, map d shows seasonal high-use areas, At-sea densities are the result of a synthesis of data from eight
displayed in 10 minute latitude by 10 minute longitude cells, shipboard and aerial survey programs conducted in the study
36°N
36°N
and breeding colonies. The seasonal high use map provides a area in the years 1980-2001 (see “Data Sources” below). Bird
further synthesis of densities presented in Maps a, b and c, and observation data and trackline data from these studies were
portrays the relative importance of various areas to the species. converted to a common format. All aerial data were continuous;
Areas with consistently high use are highlighted on this map. To ship-based data were converted separately into a continuous
provide a relative reference for the “high use” areas, cells are transect to the extent possible. From the digitized survey data,
35°N
35°N
a b also shown where the species were absent (i.e., the cell was the distributions of effort and of species were mapped into five
sampled but the species was not recorded there), or present minute latitude by five minute longitude cells using CDAS, a
Seasonal High Use Areas and
200 m
Davidson Current Season
200 m
20
20
but at lesser concentrations in any particular season. See the custom geographic information system for analyzing marine
0
0
0m
0m
Breeding Colonies
(Nov. 15 - Mar. 14) "Methods" section below for further explanation of seasonal bird and mammal surveys (MMS, 2001). The length and width
39°N
39°N
high-use areas. Breeding colonies are also shown; the relative of the survey trackline in a given cell (estimated trackline width
Persistence of
High Use size of the symbols indicates the colony size. varied by platform, depending on speed and height above
3 Seasons
water) were used to estimate the area sampled. The number
2 Seasons
1 Season
DATA SOURCES of birds of each species seen in a cell was then divided by
Birds present
38°N
Densities for marine birds at sea are based on data from the area sampled in the cell to estimate density. If a cell was
38°N
Birds absent
eight survey programs conducted between 1980 and 2001, surveyed more than once, densities were averaged, with an
Colony Size
(Breeding birds)
which were combined into a new MMS-CDAS data set (MMS, adjustment made for effort.
50,001 - 104,000
2001) using software (CDAS) developed for the Minerals
10,001 - 50,000
Management Service. Of the data sets on the original MMS- The seasonal high-use areas on map d were developed using
37°N
37°N
5001 - 10,000
CDAS CD-ROM, four aerial survey data sets contained data a similar approach as for Maps a, b and c, but the data were
1001 - 5000
in the study area from Point Arena to Point Sal. Of these, the binned into 10’x10’ cells. For each season, the cells with
501 - 1000
OSPR survey program is ongoing and data from recent years densities in the top 20% of non-zero values were designated
101 - 500
50 - 100
were added to this data set. In addition, data from four ship- “high use” for that season. Cells were scored for “high use”
Historical
based survey programs were converted to a compatible format in one, two, or three seasons and are depicted by color. To
36°N
36°N
for analysis (see section overview for details on individual provide a relative reference for the “high use” areas, cells are
data sets). also shown where the species were absent (i.e., the cell was
sampled but the species was not recorded there) or present
Data sources for aerial, at-sea data include MMS-CDAS (MMS, (but densities were never in the top 20% for any season).
2001), and California Department of Fish and Game, Office of
35°N
35°N
c d Spill Prevention and Response (CDF&G-OSPR, unpublished RESULTS AND DISCUSSION
data). Early data were collected using methods described by The common murre is very abundant in the study area, being
124°W 123°W 122°W 121°W
124°W 123°W 122°W 121°W
Briggs et al. (1983, 1987b); more recent data were collected the second most numerous marine bird in Central California.
Source Data: See text.
using updated technology but using the same general method. There have been 21,893 sightings of 141,964 individuals,
Figure 50. Common murre, seasonal density and high use areas and breeding colonies.
67
Section 2.3: BIOGEOGRAPHY OF MARINE BIRDS
with the ratio between these numbers indicating that murres
usually occur in flocks. The species nests at a complex of
related and densely occupied colonies including the Farallon
Islands, Point Reyes, Double Point (including Point Resistance
and Millers Point Rocks), and a small colony at Devils Slide.
This complex constitutes one of the largest, if not the largest,
breeding population of this species south of Alaska. Two small,
disjunct breeding colonies, the southernmost for this species,
occur off the Big Sur coast.
Based on analysis of the data, common murres reside in the
study area year-round, being particularly abundant in waters
overlying the shelf and upper slope (mean depth of 110 ± 5
m), with little seasonal change in distribution. Murre densities,
however, were, in general, significantly higher during the
Upwelling Season, probably because the entire population is
present at that time. During the other seasons, some breeding
individuals disperse outside of the study area. A multiple
regression model of nine independent variables explained
52.3% of variation in density; especially through inverse
relationships with distance to colony, ocean depth, and distance
to land; see Table 19. No significant trend in common murre
abundance existed between 1985 and 2002, and abundance
was not affected by short-term climate fluctuations (e.g., periods
of unusually warm or cold sea temperatures).
Near the large Farallon Islands colony during nesting, many
murres range seaward beyond the continental slope (and
outside sanctuary boundaries), perhaps as a response to
increased intraspecific competition for prey at that time. As
a result, the Farallon Escarpment became an area of high
concentration as well as the Farallon Ridge and shelf waters
inshore of it. Murres occur in Monterey Bay after nesting and
mainly during the Oceanic Season. During years of unusually
warm waters (and depleted prey), murres occur more frequently
inshore, especially along the coast from Point Reyes south to
Año Nuevo Island, the usual area of concentration during the
relatively warm Oceanic Season.
This species is a deep diver (to 180m depth, Ainley et al,
2002) that feeds on fish and invertebrates. During winter and
early spring, major prey include herring, market squid and
euphausiids; this diet then shifts mostly to juvenile rockfish
and anchovies in mid-summer. See Tables 15 and 16 for related
summary information.
68
Section 2.3: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS of H. T. Harvey and Associates and Carol Keiper of Oikonos
Rhinoceros Auklet Figures 51a, b, and c show the density (birds/km 2) of (unpublished data; see Oedekoven et al., 2001 for details
Cerorhinca monocerata Rhinoceros Auklet in the Upwelling, Oceanic, and Davidson on survey methods); and Lisa T. Ballance, from the Ecology
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Current seasons, displayed in five minute latitude by five minute Program of the Southwest Fisheries Science Center, NMFS,
Upwelling Season Oceanic Season
200 m
200 m
20
20
longitude cells. Densities are based on the combined data NOAA (unpublished data). Data on breeding colonies in the
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) sets of several studies (see “Methods” and “Data Sources” study area were obtained from Carter et al. (1992), with most
39°N
39°N
below). The color and mapping intervals were customized recent estimates for Año Nuevo from Thayer and Sydeman
Density
to show the most structure and to highlight significant areas, (2002).
(Animals/km²)
while allowing comparisons among marine bird species. Cells
> 100.00
that were surveyed but in which no Rhinoceros Auklets were Although the at-sea data span the years from 1980 to 2001,
50.01 - 100.00
38°N
38°N
observed have a density of zero. Areas not surveyed appear data are not available for all seasons in all years. For the
10.01 - 50.00
5.01 - 10.00
white; no information is available for these areas. Blue lines Upwelling Season, data are from 1980-1982 and 1985-2001.
1.01 - 5.00
indicate the boundaries of the National Marine Sanctuaries For the Oceanic Season, data are from 1980-1982, 1991, and
0.51 - 1.00
in the study area: Cordell Bank, Gulf of the Farallones, and 1994-2001. For the Davidson Current Season, data are from
0.11 - 0.50
Monterey Bay. 1980-1986 and 1991-2001.
0.06 - 0.10
37°N
37°N
0.01 - 0.05
0.00
In order to provide one map for the species that integrates the METHODS
0 25 50 Km
patterns of its spatial and temporal occurrence and abundance At-sea densities are the result of a synthesis of data from eight
in the study area, map d shows seasonal high-use areas, shipboard and aerial survey programs conducted in the study
displayed in 10 minute latitude by 10 minute longitude cells, area in the years 1980-2001 (see “Data Sources” below). Bird
36°N
36°N
and breeding colonies. The seasonal high use map provides observation data and trackline data from these studies were
a further synthesis of densities presented in Maps a, b and c, converted to a common format. All aerial data were continuous;
and portrays the relative importance of various areas to the ship-based data were converted separately into a continuous
species. Areas with consistently high use are highlighted on this transect to the extent possible. From the digitized survey data,
map. To provide a relative reference for the “high use” areas, the distributions of effort and of species were mapped into five
35°N
35°N
a b cells are also shown where the species were absent (i.e., the minute latitude by five minute longitude cells using CDAS, a
cell was sampled but the species was not recorded there), or custom geographic information system for analyzing marine
Seasonal High Use Areas and
200 m
Davidson Current Season
200 m
20
20
present but at lesser concentrations in any particular season. bird and mammal surveys (MMS, 2001). The length and width
0
0
0m
0m
Breeding Colonies
(Nov. 15 - Mar. 14) See the "Methods" section below for further explanation of of the survey trackline in a given cell (estimated trackline width
39°N
39°N
seasonal high-use areas. Breeding colonies are also shown; varied by platform, depending on speed and height above
Persistence of
High Use the relative size of the symbols indicates the colony size. water) were used to estimate the area sampled. The number
3 Seasons
of birds of each species seen in a cell was then divided by
2 Seasons
1 Season
DATA SOURCES the area sampled in the cell to estimate density. If a cell was
Birds present
38°N
Densities for marine birds at sea are based on data from eight surveyed more than once, densities were averaged, with an
38°N
Birds absent
survey programs conducted between 1980 and 2001, which adjustment made for effort.
Colony Size
(Breeding birds)
were combined into a new MMS-CDAS data set (MMS, 2001)
10,000 - 50,000
using software (CDAS) developed for the Minerals Management The seasonal high-use areas on map d were developed using
5001 - 10,000
Service. Of the data sets on the original MMS-CDAS CD-ROM, a similar approach as for Maps a, b and c, but the data were
37°N
37°N
1001 - 5000
four aerial survey data sets contained data in the study area binned into 10’x10’ cells. For each season, the cells with
501 - 1000
from Point Arena to Point Sal. Of these, the OSPR survey densities in the top 20% of non-zero values were designated
101 - 500
program is ongoing and data from recent years were added “high use” for that season. Cells were scored for “high use”
51 - 100
2 - 50
to this data set. In addition, data from four ship-based survey in one, two, or three seasons and are depicted by color. To
Historical
programs were converted to a compatible format for analysis provide a relative reference for the “high use” areas, cells are
36°N
36°N
(see section overview for details on individual data sets). also shown where the species were absent (i.e., the cell was
sampled but the species was not recorded there) or present
Data sources for aerial, at-sea data include MMS-CDAS (MMS, (but densities were never in the top 20% for any season).
2001), and California Department of Fish and Game, Office of
Spill Prevention and Response (CDF&G-OSPR, unpublished RESULTS AND DISCUSSION
35°N
35°N
c d data). Early data were collected using methods described by In the study area, this common species nests principally at
Briggs et al. (1983, 1987b); more recent data were collected the Farallon Islands; a smaller nesting population occurs at
124°W 123°W 122°W 121°W
124°W 123°W 122°W 121°W
using updated technology but using the same general method. Año Nuevo. The Farallones constitute the southernmost large
Source Data: See text.
Data sources for ship-based survey data include: David Ainley nesting colony. At-sea surveys recorded 5,415 sightings of
Figure 51. Rhinoceros auklet, seasonal density and high use areas and breeding colonies.
69
Section 2.3: BIOGEOGRAPHY OF MARINE BIRDS
15,454 individuals. Based on the analysis of the combined data
sets described in this section, the abundance of Rhinoceros
Auklets has increased significantly since the 1970s (Ainley et
al. 1994). Based on patterns apparent in the maps, the current
at-sea population probably far exceeds the estimates of nesting
populations in central California (Michelle Hester, pers.comm.).
Therefore, if there was more nesting habitat (e.g., burrows,
holes, crevices on offshore islands), the nesting population
would probably be much larger.
Rhinoceros auklets, which mainly visit colonies at night,
occurred principally in waters overlying the slope (mean depth
of occurrence was 762 ± 22 m), particularly the shelf break,
and, including the Farallon Escarpment. A sizeable portion of
the population occurs outside of the National Marine Sanctuary
boundaries. This is especially true in the vicinity of the Gulf
of the Farallones during the Upwelling (nesting) and Oceanic
seasons, when these auklets occur farther offshore (mean
depths were 791 m and 1,370 m, respectively). This expansion
of habitat, causing a ‘halo’ of increased density around the
islands, may be a response to the large numbers nesting at the
Farallones, a pattern typical of the Western Gull and Common
Murre (see those accounts). The species’ concentration,
especially along the shelf break and upper continental slope,
is particularly evident during the Oceanic Season, when the
nesting populations are no longer associated with colonies.
A multiple-regression model of nine independent variables
explained 19.8% of variation in cell density; important variables
were a negative relationship to distance to land, and positive
ones to season and ocean depth; see Table 19. The relationship
with season reflected a dramatic increase in abundance during
the Davidson Current Season (mean density of 161 birds per
100km2) compared to the Upwelling and Oceanic seasons
(mean densities of 48 and 62 birds per 100 km2, respectively).
This increase during the Davidson Current Season was likely
due to an influx of birds from the north where much larger
populations breed, compared to those of the study area.
This species feeds by diving, probably to relatively deep depths
(100 m, Ainley and Boekelheide, 1990), capturing mostly fish
but also euphuasiids. Important prey are juvenile rockfish,
anchovy and saury. See Tables 15 and 16 for related summary
information.
70
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS the distributions of effort and of species were mapped into five
Marine Bird Density Figures 52a, b, and c show the combined density (birds/km2) minute latitude by five minute longitude cells. The length and
of 76 species of marine birds in the Upwelling, Oceanic, and width of the survey trackline in a given cell (estimated trackline
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Davidson Current seasons, displayed in five minute latitude by width varied by platform, depending on speed and height above
Upwelling Season Oceanic Season
200 m
200 m
20
20
five minute longitude cells. Map d shows density for all seasons water) were used to estimate the area sampled. The number
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) and years combined. Densities are based on combined data of marine birds seen in a cell was then divided by the area
39°N
39°N
of several studies (see “Methods” and “Data Sources” above). sampled in the cell to estimate density. If a cell was censused
Density
The color and mapping intervals were customized to show the more than once, densities were averaged, with adjustment
(Animals/km²)
most structure and highlight significant areas. Cells that were made for effort.
> 100.00
surveyed but in which no birds were observed have a density
50.01 - 100.00
38°N
38°N
of zero; unsurveyed areas are white. Blue lines indicate the RESULTS AND DISCUSSION
10.01 - 50.00
5.01 - 10.00 National Marine Sanctuary boundaries of Cordell Bank, Gulf of Overall density is dominated by two abundant marine bird
1.01 - 5.00
the Farallones, and Monterey Bay; bathymetric contours for the species: common murre and sooty shearwater.
0.51 - 1.00
200 meter and 2,000 meter isobaths are also shown in blue.
0.11 - 0.50
0.06 - 0.10
Based on visual inspection of the maps, density was highest,
37°N
37°N
0.01 - 0.05
DATA SOURCES during the Upwelling Season with cells of highest density
0.00
At-sea densities are based on data from eight survey programs most widespread as well. Except for a few highest-density
0 25 50 Km
conducted in 1980-2001, which were combined using software ‘hot spots,’(see Table 17) marine birds were distributed evenly
developed for MMS-CDAS (2001) and expanded for this at high density (>10 individuals per km2) over the shelf and
project. Of the data sets on the original CD-ROM, four aerial slope from north to south in the study area. Particular hot spots
36°N
36°N
survey data sets provided data in the study area from Point were inshore Monterey Bay, Farallon Ridge and Cordell Bank.
Arena to Point Sal. Of these, one program was still ongoing and The pattern during this season generally matched the pattern
data from recent years were added to this data set. In addition, apparent when all seasons were combined.
data from four ship-based survey programs were converted to
a compatible format for analysis. See section introduction for
35°N
35°N
During the Oceanic Season, highest density areas increased
a b details on individual data sets. in prevalence inshore. At that time, hot spots were the San
Francisco Bay tidal plume, inshore near Año Nuevo, innermost
200 m
Davidson Current Season
200 m
All Seasons
20
20
Data sources for aerial at-sea data include MMS-CDAS (MMS,
0
0
Monterey Bay and San Luis Obispo Bay.
0m
0m
(Nov. 15 - Mar. 14) 2001), and California Department of Fish and Game Office of
39°N
39°N
Spill Prevention and Response (CDF&G-OSPR, unpublished During the Davidson Current Season, birds shifted more to
data). Early data were collected using methods described by the mid-shelf.
Briggs et al. (1987b); more recent data were collected using
updated technology but the same general method. Data
38°N
38°N
sources for ship-based survey data include: David Ainley and
Carol Keiper (unpublished data; see Oedekoven et al., 2001
for details on survey methods).
Although the at-sea data span the years 1980 to 2001, data
37°N
37°N
are not available for all seasons in all years. For the Upwelling
Season, data are from 1980-1982 and 1985-2001. For the
Oceanic Season, data are from 1980-1982, 1991, and 1994-
2001. For the Davidson Current Season, data are from 1980-
1986 and 1991-2001.
36°N
36°N
METHODS
At-sea densities are the result of a synthesis of data from
eight shipboard and aerial survey programs conducted in
the study area in the years 1980-2001 (see “Data Sources”
35°N
35°N
c d above). Observation and trackline data from these studies were
converted to a common format. All aerial data were continuous;
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Source Data: See text. ship-based data were converted separately into a continuous
transect to the extent possible. From the digitized survey data,
Figure 52. Marine bird density, by season and for all seasons.
71
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS above). Observation and trackline data from these studies were
Figure 53a, b, and c shows total marine bird biomass (kg/ converted to a common format. All aerial data were continuous;
Marine Bird Biomass km2) in each five minute latitude by five minute longitude cell ship-based data were converted separately into a continuous
124°W 123°W 122°W 121°W
for each oceanographic season and for all seasons combined
124°W 123°W 122°W 121°W
transect to the extent possible. From the digitized survey data,
Upwelling Season Oceanic Season 53d. Density for each of 76 species was multiplied by average
200 m
the distributions of effort and of species were mapped into five
200 m
20
20
0
0
0m
0m
body mass for that species. These products were summed minute latitude by five minute longitude cells. The length and
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14)
39°N
39°N
for all species in a cell. The color and mapping intervals were width of the survey trackline in a given cell (estimated trackline
Biomass customized to show the most structure and highlight significant width varied by platform, depending on speed and height above
(kg/km²)
areas. Cells that were surveyed but in which no birds were water) were used to estimate the area sampled. The number
> 100
observed have a biomass density of zero; unsurveyed areas of marine birds seen in a cell was then divided by the area
51 - 100
are white. Blue lines indicate the National Marine Sanctuary sampled in the cell to estimate density. If a cell was censused
38°N
38°N
21 - 50
boundaries of Cordell Bank, Gulf of the Farallones, and more than once, densities were averaged, with adjustment
11 - 20
6 - 10 Monterey Bay; bathymetric contours for the 200 meter and made for effort.
4-5
2,000 meter isobaths are also shown in blue.
3
Once the weighted densities had been determined for each
2
37°N
37°N
DATA SOURCES
1 species in each cell, densities of each species were multiplied
0
At-sea biomass densities are based on data from eight survey by the average body mass of that species. These ‘biomass
programs conducted in 1980-2001, which were combined using
0 25 50 Km
densities’ were then summed for each cell and the results
software developed for MMS-CDAS (2001) and expanded for plotted.
this project. Of the data sets on the original CD-ROM, four aerial
36°N
36°N
survey data sets provided data in the study area from Point RESULTS AND DISCUSSION
Arena to Point Sal. Of these, one program was still ongoing and In general, the biomass maps are dominated by two, rela-
data from recent years were added to this data set. In addition, tively heavy-bodied, numerically dominant species: com-
data from four ship-based survey programs were converted to mon murre and sooty shearwater. These maps are also
a compatible format for analysis. See section introduction for influenced, to a lesser degree, by the species identified as
35°N
35°N
a b details on individual data sets. abundant in the study area (see Table 15).
Data sources for aerial at-sea data include MMS-CDAS (MMS,
200 m
Davidson Current Season
200 m
All Seasons
20
20
Looking first at a summary of all seasons, high biomass densi-
0
0
0m
0m
2001) and California Department of Fish and Game Office of
(Nov. 15 - Mar. 14) ties occurred in the Gulf of the Farallones, especially around
39°N
39°N
Spill Prevention and Response (CDF&G-OSPR), unpublished the Farallon Islands, the San Francisco Bay tidal plume, off
data. Early data were collected using methods described by Half-moon Bay, just south of Point Año Nuevo and in inner
Briggs et al. (1987b); more recent data were collected using Monterey Bay.
updated technology but the same general method. Data
sources for ship-based survey data include David Ainley and During the Upwelling season, high biomass densities occurred
38°N
38°N
Carol Keiper (unpublished data; see Oedekoven et al., 2001 over the shelf and upper slope with highest density areas oc-
for details on survey methods). Although the at-sea data span curring at Monterey Bay, Farallon Ridge, and Cordell Bank.
the years 1980-2001, data are not available for all seasons in The distribution of high biomass during the Upwelling Season
all years. For the Upwelling Season, data are from 1980-1982 mimicked that described in the all seasons map (map d).
and 1985-2001. For the Oceanic Season, data are from 1980-
37°N
37°N
1982, 1991 and 1994-2001. For the Davidson Current Season, During the Oceanic Season high biomass was concentrated
data are from 1980-1986 and 1991-2001. more over the inner shelf than in the Upwelling Season, par-
ticularly evident from Point Reyes to Monterey, as well as San
Data on average biomass for each species were derived Luis Obispo Bay.
36°N
36°N
from Body Weights of 686 Species of North American Birds
(Dunning 1993). In a few instances, a species was not listed in During the Davidson Current Season (DCS), virtually the entire
this reference; in these cases, the biomass of a closely related continental shelf from Point Reyes to Point Sur exhibited high
bird of a similar size was used. marine bird biomass.
35°N
35°N
c d METHODS
At-sea densities are the result of a synthesis of data from
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
eight shipboard and aerial survey programs conducted in
Source Data: See text.
the study area in the years 1980-2001 (see “Data Sources”
Figure 53. Marine bird biomass, by season and for all seasons.
72
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Upwelling Season, data are from 1980-1982 and 1985-2001.
ABOUT THESE MAPS
For the Oceanic Season, data are from 1980-1982, 1991 and
Species diversity was calculated for each five minute latitude
Marine Bird Species Diversity 1994-2001. For the Davidson Current Season, data are from
by five minute longitude cell using density as the variable in the
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
1980-1986 and 1991-2001.
Shannon [Diversity] Index (Shannon and Weaver 1949). This
Upwelling Season Oceanic Season
200 m
200 m
20
20
index measures the degree to which a species assemblage is
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) METHODS
dominated by a few species. If a cell contains high densities
39°N
39°N
At-sea densities are the result of a synthesis of data from
of a few species and low densities of all others, the value of
Diversity eight shipboard and aerial survey programs conducted in
diversity (H’) will be low, indicating low diversity. Alternatively,
(H')
the study area in the years 1980-2001 (see “Data Sources”
if many species are present at similar densities, the value will
2.001 - 3.000
above). Observation and trackline data from these studies were
be high, indicating high diversity. Figures 54a, b, and c show
1.801 - 2.000
converted to a common format. All aerial data were continuous;
the diversity index H’ in three oceanographic seasons; map d
38°N
38°N
1.501 - 1.800
ship-based data were converted separately into a continuous
shows diversity for all seasons and years combined. Although
0.801 - 1.500
0.000 - 0.800 transect to the extent possible. From the digitized survey data,
a density-based calculation of the Shannon Index is less
the distributions of effort and of species were mapped into five
influenced by differences in effort as compared with the index
0 25 50 Km
minute latitude by five minute longitude cells. The length and
calculated using species counts, a significant correlation (p<
37°N
37°N
width of the survey trackline in a given cell (estimated trackline
0.001) remained between diversity and effort.
width varied by platform, depending on speed and height above
water) were used to estimate the area sampled. The number
To standardize for variable effort among cells and variable
of marine birds seen in a cell was then divided by the area
strip width among species, density was used for each species
sampled in the cell to estimate density. If a cell was censused
in each cell as the basis for calculating the diversity index
36°N
36°N
more than once, densities were averaged, with adjustment
value. All 76 marine bird species that had been recorded in
made for effort.
the data set were included. Cells are colored based on the
value of H’ computed for a particular season. Red indicates
The Shannon Index (Shannon and Weaver 1949) was used to
high diversity, blue indicates low diversity. Unsurveyed areas
quantify species diversity. For each cell, diversity was calculated
are colored white. Blue lines indicate the National Marine
35°N
35°N
a b using the formula
Sanctuary boundaries of Cordell Bank, Gulf of the Farallones,
and Monterey Bay; bathymetric contours for the 200 meter and n n
S
H ′ = − ∑ i ln i
Davidson Current Season
200 m
200 m
All Seasons
20
20
2,000 meter isobaths are also shown in blue.
0
0
0m
0m
i =1 n n
(Nov. 15 - Mar. 14)
39°N
39°N
DATA SOURCES
where ni is the density of species in that cell. Density was
At-sea densities are based on data from eight survey programs
used for calculating the index value in order to compensate for
conducted in 1980-2001, which were combined using CDAS
variable effort among cells and species. We looked at three
software into an MMS-CDAS data set (MMS, 2001) developed
38°N
38°N
oceanographic seasons and at all seasons combined.
for Minerals Management Service and expanded for this project.
Of the data sets on the original CD-ROM, four aerial survey
The Shannon Index was selected as the diversity metric
data sets provided data in the study area from Point Arena to
because it is widely used and accepted in community ecology.
Point Sal. Of these, one program was still ongoing and data
It has three desirable properties for a diversity index, noted
from recent years were added to this data set. In addition,
37°N
37°N
below. Most diversity indices do not take these three qualities
data from four ship-based survey programs were converted to
into account. For more information on diversity indices, see
a compatible format for analysis. See section introduction for
Ecological Diversity, E.C. Pielou, pp 7-18.
details on individual data sets.
1. The diversity index is greatest when all species in the
Data sources for aerial at-sea data include MMS-CDAS (2001)
36°N
36°N
community are equally represented in numbers (e.g., evenness
and California Department of Fish and Game Office of Spill
in a community). Or, for a given number of species (e.g.,
Prevention and Response (CDF&G-OSPR), unpublished data.
richness value), the diversity index should have it’s greatest
Early data were collected using methods described by Briggs
value when the proportion of each species is the same.
et al. (1987b); more recent data were collected using updated
technology but the same general method. Data sources for
35°N
35°N
c d 2. Given two completely diverse or similiar communities, the
ship-based survey data include David Ainley and Caol Keiper
one with the higher number of species has a greater diversity
(unpublished data; see Oedekoven et al., 2001 for details on
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Source Data: See text. value.
methods). Although the at-sea data span the years 1980-2001,
data are not available for all seasons in all years. For the
Figure 54. Marine bird diversity, by season and for all seasons.
73
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
3. The last property is difficult to summarize but is something
like this: This property takes into account the hierarchical
nature, or representativeness in the biological classification of
each species, when estimating diversity.
RESULTS AND DISCUSSION
Looking first at a summary of all seasons, the marine avifauna
was most diverse in areas largely outside of National Marine
Sanctuary boundaries, especially in areas of the continental
slope and particularly the Farallon Escarpment. Localized areas
of high diversity occurring within sanctuary boundaries include:
Pioneer, Ascension/Cabrillo, and Carmel canyons, as well as
the continental slope off Point Sur.
During the Upwelling Season, the avifauna was the least
diverse; areas of highest diversity in this season included the
Farallon Escarpment, and Pioneer, Ascension, and Carmel
canyons.
During the Oceanic Season, diversity was comparable to that
of the Upwelling Season in general. Areas of high diversity
continued to include the Farallon Escarpment area, Pioneer
Canyon, and inner Monterey Bay Canyon.
During the Davidson Current Season, marine bird diversity, in
general, was the highest of the year. Areas of high diversity
were all localized, and most occurred over the continental slope
(e.g., Farallon Escarpment, and Pioneer, Ascension, Monterey
Bay and Carmel canyons) but some also occurred over the
shelf (e.g., the inner San Francisco Bay tidal plume and inner
portions of Monterey Bay).
However, because of the significant correlation between diver-
sity and effort, some of the diversity patterns may be influenced
by differences in effort across the study area. See the additional
analysis and discussion of diversity in the Integration section.
74
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THIS MAP
124°W 123°W 122°W 121°W
The 40 largest marine bird breeding colonies in the study
200 m Major Marine Bird Breeding Colonies
20
39°N
39°N
0
area were chosen for this map (Figure 55). The total number
0m
of breeding marine bird species is indicated by the size of
Fish Rocks
the circle, and the number of species using a particular
Gualala Point Island
colony is indicated by the circle color. The large symbol at
the San Francisco Bay entrance represents a summary of
Russian River Rocks all the colonies in San Francisco Bay. See Table 18 for more
Colony Populations
Arched Rock
information on these colonies.
Bodega Rock Number of Species
9 - 12
DATA SOURCES
7-8 Data on marine bird colonies were derived primarily from
Point Reyes Point Resistance
38°N
38°N
6 Breeding Populations of Seabirds in California, 1989-1991
Millers Point Rocks
(Carter et al., 1992, unpublished data). Colony data were
5
San Francisco Bay
updated where more current information was available.
4
and Alcatraz Island*
North Farallon Islands
Updated information is presented for some species on South
3
South Farallon Island Farallon Island (Sydeman et al., 1998, Warzybok et al., 2002),
1-2
Devil's Slide Rock
Año Nuevo Island (Thayer and Sydeman 2002 a, b), Bird Rock,
Number of Birds Point Reyes, and Double Point Rocks (Whitworth et al., 2002),
Big Basin State Park and vicinity (Laird Henkel, pers. comm)
50,001 - 153,000
and Devil’s Slide Rock (Gerry McChesney, USFWS, pers.
Big Basin State Park and vicinity
10,001 - 50,000
Vicinity of Año Nuevo
comm).
Island and Point El Jarro Point to Davenport
5001 - 10,000
37°N
37°N
Davenport to Sand Hill Bluff
1001 - 5000
METHODS
501 - 1000
Colony locations were plotted using latitude and longitude
250 - 500
coordinates from Breeding Populations of Seabirds in California,
1989-1991 (Carter et al., 1992, unpublished data).
Bird Rock 0 25 50 Km
Bird Island
RESULTS AND DISCUSSION
Castle Rocks and Mainland
The study area is in a geologic subduction zone of the eastern
Pacific and adjacent continental margin. Therefore, as with
Anderson Canyon Rocks
analogous regions elsewhere on the globe (e.g., west coasts
of South America and Africa), islands are not common. In
36°N
36°N
Plaskett Rock
somewhat of a departure from this pattern, the Gulf of the
Cape San Martin
Farallones contains far more coastal rocks and offshore islands
La Cruz Rock than anywhere else in the study area and, in fact, this is the
Piedras Blancas Island
case for 400 miles to the north and south. Obvious in this
map is the importance to breeding marine birds of the Gulf of
the Farallones, defined as the broad shelf from Point Reyes/
Fairbank Point Tomales Point to Año Nuevo and out to the Farallon Islands. A
disproportionate number of breeding colonies and, certainly,
Pup Rock and Pecho Rock
Adjacent Mainland
most of various species’ regional breeding populations, occur
here. These colonies are large and diverse owing to the high
35°N
35°N
productivity of surrounding waters and the complexity of
habitats in the region. See Table 18 for a numerical summary
of each colony’s contribution to the breeding marine avifauna
of the study area, composed of 16 species, 12 of which breed
within the Gulf of the Farallones.
* Summary of all San Francisco Bay colonies.
124°W 123°W 122°W 121°W
Source Data: See text.
Figure 55. Major marine bird breeding colonies.
75
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
METHODS
ABOUT THESE MAPS
Density in Warm, Cold and Neutral Periods: 1980-2001 At-sea densities are the result of a synthesis of data from
A comparison of the abundance and distribution of 76 marine
eight shipboard and aerial survey programs conducted in
birds during warm-water periods (including El Niño), cold-water
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
the study area in the years 1980-2001 (see “Data Sources”
periods (La Niña) and normal (neutral) periods is provided here
Warm-water Conditions Neutral Conditions
200 m
200 m
20
20
above). Observation and trackline data from these studies were
as an example of how marine birds may respond to short-term
0
0
0m
0m
converted to a common format. All aerial data were continuous;
variation in marine climate. In this synthesis, what is shown is
39°N
39°N
ship-based data were converted separately into a continuous
density, which treats all species equally regardless of body size.
Density
transect to the extent possible. From the digitized survey data,
Therefore, the patterns demonstrated by tiny, more abundant
(Animals/km²)
the distributions of effort and of species were mapped into five
species, such as storm-petrels and phalaropes, are more
> 100.00
minute latitude by five minute longitude cells. The length and
greatly expressed. For a description of how these periods were
50.01 - 100.00
38°N
38°N
width of the survey trackline in a given cell (estimated trackline
chosen, see the following topic in the bird section: "Response
10.01 - 50.00
5.01 - 10.00 width varied by platform, depending on speed and height above
to Variation in Marine Climate" (pages 49-50).
1.01 - 5.00
water) were used to estimate the area sampled. The number
0.51 - 1.00
of marine birds seen in a cell was then divided by the area
Figures 56a, b and c show the combined density (birds/km2)
0.11 - 0.50
sampled in the cell to estimate density. If a cell was censused
of 76 species of marine birds in cold-water, neutral, and
0.06 - 0.10
37°N
37°N
0.01 - 0.05 more than once, densities were averaged, with adjustment
warm-water periods, displayed in five minute latitude by
0.00
made for effort.
five minute longitude cells. Map d shows overall patterns of
0 25 50 Km
density. Densities are based on combined data of several
Marine bird density data was then organized into periods where
studies (see “Methods” and “Data Sources” below). The
surface ocean conditions were warm (including El Niños),
color and mapping intervals were customized to show the
36°N
36°N
cold (including La Niñas) or neither (neutral). The density of
most structure and highlight significant areas. Cells that were
all species seen within respective cells was summed for that
surveyed but in which no birds were observed have a density
cell.
of zero; unsurveyed areas are white. Blue lines indicate the
National Marine Sanctuary boundaries of Cordell Bank, Gulf of
To illustrate these temperature conditions, a comparison of
the Farallones, and Monterey Bay; bathymetric contours for the
35°N
35°N
a b marine bird densities was made by making maps that use
200 meter and 2,000 meter isobaths are also shown in blue.
selected season/year periods that represented these cold,
200 m
Cold-water Conditions
200 m
20
Overall Patterns
20
warm and neutral periods. The data for each "condition" map
DATA SOURCES
0
0
0m
0m
was grouped as shown below; these groupings were based
At-sea densities are based on data from eight survey programs
39°N
39°N
on the assignments made in Table 14. Once the selection of
conducted in 1980-2001; these data sets were combined using
data were made for each analysis period (i.e., warm, neutral or
CDAS software into an MMS-CDAS data set (MMS, 2001) and
cold), the density of all birds seen within each cell was summed
expanded for this project. Of the data sets on the original CD-
for that cell.
ROM, four aerial survey data sets provided data in the study
38°N
38°N
area from Point Arena to Point Sal. Of these, one program was
For the warm-water conditions (including El Niños) map, the
still ongoing and data from recent years were added to this data
following seasons and years were used: Davidson Current
set. In addition, data from four ship-based survey programs
Season: 1981, 1983, 1984, 1992, 1993, 1994, 1996, 1998.
were converted to a compatible format for analysis. See section
Upwelling Season: 1985, 1987, 1992, 1993, 1995, 1998.
introduction for details on individual data sets.
37°N
37°N
Oceanic Season: 1983, 1997.
Data sources for aerial at-sea data include MMS-CDAS (MMS,
For the neutral conditions map, the following seasons and years
2001) and California Department of Fish and Game Office of
were used: Davidson Current Season: 1982, 1986, 1995, 1997.
Spill Prevention and Response (CDF&G-OSPR), unpublished
Upwelling Season: 1980, 1982, 1986, 1988, 1989, 1994, 1996,
data. Early data were collected using methods described by
36°N
36°N
1997. Oceanic Season: 1982, 1991, 1995.
Briggs et al. (1983); more recent data were collected using
updated technology but the same general method. Data
For the cold-water conditions (including La Niñas) map, the
sources for ship-based survey data include David Ainley and
following seasons and years were used: Davidson Current
Carol Keiper (unpublished data; see Oedekoven et al., 2001
Season: 1985, 1991, 1999, 2000, 2001, 2002. Upwelling
for details on methods). Although the at-sea data span the
35°N
35°N
c d Season: 1981, 1990, 1991, 1999, 2000, 2001. Oceanic Season:
years 1980-2001, data are not available for all seasons in all
1980, 1981, 1994, 1996, 1998, 1999, 2000, 2001.
years.
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Source Data: See text.
Figure 56. Density in warm, cold, and neutral periods: 1980-2001.
76
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
RESULTS AND DISCUSSION
There was not a great deal of difference in density apparent in
the exhibited patterns for the different periods. Nevertheless,
during warm-water conditions (e.g., El Niño events) marine
bird populations appear to contract more into the area defined
by the boundaries of the central California National Marine
Sanctuaries, from Tomales Point south to Monterey. Generally,
this area contains most of the shelf habitat of the study area,
which tends to have a greater complexity of microhabitats than
deeper waters. The reason there was not much of an apparent
pattern or major difference seen in these maps, is that individual
species respond differently to the three different temperature
conditions shown. For instance, some may move out of an
area but others may move in, and therefore, when species are
combined, these individual responses are homogenized.
During both warm and cold excursion from ‘normal’/neutral
marine climate, populations seemed to be slightly more
widespread, with major concentrations in Monterey Bay.
During cold-water conditions (e.g., La Niña events), densities
appeared to be the highest, especially in waters close to the
coast (e.g., see contiguous high-density, red and orange cells
along the coast). During warm-water events, the concentrations
are further offshore in the mid to outer shelf and there are fewer
highest density (red) cells.
As noted earlier, overall marine bird density in this analysis
is generally dominated by two numerically dominant spe-
cies, common murre and sooty shearwater, and to a lesser
degree, by the species identified as abundant in the study
area (see Table 15).
77
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS reference; in these cases, the mass of a closely related bird
Biomass in Warm, Cold and Neutral Periods: 1980-2001 A comparison of the abundance and distribution of 76 marine of a similar size was used.
birds during warm-water (El Niño) compared to cold-water (La
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Niña) and normal (neutral) periods provides an example of how METHODS
Warm-water Periods Neutral Periods
200 m
200 m
20
20
marine birds respond to short-term excursions from the usual At-sea densities are the result of a synthesis of data from
0
0
0m
0
m
marine climate. In this comparison, densities were converted to eight shipboard and aerial survey programs conducted in
39°N
39°N
biomass by multiplying density by body mass of each species. the study area in the years 1980-2001 (see “Data Sources”
Biomass
This comparison, thus, emphasizes more the larger-bodied above). Observation and trackline data from these studies were
(kg/km²)
species, such as Sooty Shearwater and Common Murre. converted to a common format. All aerial data were continuous;
> 100
ship-based data were converted separately into a continuous
51 - 100
21 - 50
38°N
38°N
Figures 57a, b, and c show the combined biomass density transect to the extent possible. From the digitized survey data,
11 - 20
(kg/km2) of 76 species of marine birds in cold-water, neutral, the distributions of effort and of species were mapped into five
6 - 10
and warm-water periods, displayed in five minute latitude by minute latitude by five minute longitude cells. The length and
4-5
five minute longitude cells. Map d shows overall patterns of width of the survey trackline in a given cell (estimated trackline
3
2 biomass density. Densities are based on combined data of width varied by platform, depending on speed and height above
37°N
37°N
1
several studies (see “Methods” and “Data Sources” below). water) were used to estimate the area sampled. The number
0
The color and mapping intervals were customized to show the of marine birds seen in a cell was then divided by the area
0 25 50 Km
most structure and highlight significant areas. Cells that were sampled in the cell to estimate density. If a cell was censused
surveyed but in which no birds were observed have a density more than once, densities were averaged, with adjustment
of zero; unsurveyed areas are white. Blue lines indicate the made for effort.
36°N
36°N
National Marine Sanctuary boundaries of Cordell Bank, Gulf of
the Farallones, and Monterey Bay; bathymetric contours for the For each species that occurred in a cell, the average density
200 meter and 2,000 meter isobaths are also shown in blue. was then multiplied by a species’ body mass (from Dunning,
1993). This resulted in an estimate of biomass for that species.
DATA SOURCES The biomass of all species in each cell was summed to give
35°N
35°N
a b At-sea densities are based on data from eight survey programs the cell biomass.
conducted in 1980-2001. These data were combined using
200 m
Cold-water Periods
200 m
Overall Patterns
20
20
CDAS software into an MMS-CDAS data system (MMS, 2001) Marine bird density data was then organized into periods where
0
0
0m
0m
for the Minerals Management Service and expanded for this surface ocean conditions were warm (including El Niños),
39°N
39°N
project. Of the data sets on the original MMS-CDAS CD-ROM, cold (including La Niñas) or neither (neutral). The density of
four aerial survey data sets contained data in the study area all species seen within respective cells was summed for that
from Point Arena to Point Sal. Of these, one program was still cell.
ongoing and data from recent years were added to this data
set. In addition, data from four ship-based survey programs To illustrate these temperature conditions, a comparison of
38°N
38°N
were converted to a compatible format for analysis. See section marine bird densities was made by making maps that use
introduction for details on individual data sets. selected season/year periods that represented these cold,
warm and neutral periods. The data for each condition map
Data sources for aerial at-sea data include MMS-CDAS (MMS, was grouped as shown below; these groupings were based
37°N
37°N
2001) and California Department of Fish and Game Office of on the assignments made in Table 14. Once the selection of
Spill Prevention and Response (CDF&G-OSPR), unpublished data were made for each analysis period (i.e., warm, neutral or
data. Early data were collected using methods described by cold), the density of all birds seen within each cell was summed
Briggs et al. (1987b); more recent data were collected using for that cell.
updated technology but the same general method. Data
36°N
36°N
sources for ship-based survey data include David Ainley and For the warm-water conditions (including El Niños) map, the
Carol Kieper (unpublished data; see Oedekoven et al., 2001 following seasons and years were used: Davidson Current
for details on methods). Although the at-sea data span the Season: 1981, 1983, 1984, 1992, 1993, 1994, 1996, 1998.
years 1980-2001, data are not available for all seasons in all Upwelling Season: 1985, 1987, 1992, 1993, 1995, 1998.
years. Oceanic Season: 1983, 1997.
35°N
35°N
c d
Data on average mass for each species were derived from For the neutral conditions map, the following seasons and years
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Body Weights of 686 Species of North American Birds (Dunning were used: Davidson Current Season: 1982, 1986, 1995, 1997.
Source Data: See text.
1993). In a few instances, a species was not listed in this
Figure 57. Biomass in warm, cold and neutral periods: 1980-2001.
78
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Upwelling Season: 1980, 1982, 1986, 1988, 1989, 1994, 1996,
1997. Oceanic Season: 1982, 1991, 1995.
For the cold-water conditions (including La Niñas) map, the
following seasons and years were used: Davidson Current
Season: 1985, 1991, 1999, 2000, 2001, 2002. Upwelling
Season: 1981, 1990, 1991, 1999, 2000, 2001. Oceanic Season:
1980, 1981, 1994, 1996, 1998, 1999, 2000, 2001.
RESULTS AND DISCUSSION
There was slightly more difference in biomass than was observed
for the analogous comparison of density. Biomass was generally
more concentrated during warm and cold conditions than during
neutral conditions, especially cold-water periods, which were
mimicked by the overall all-conditions summary. Many inner
shelf habitat areas exhibited high marine bird biomass during
cold-water periods. The Farallon Ridge and Monterey Bay had
relatively high biomass under all conditions.
As noted earlier, marine bird biomass in this analysis is gen-
erally dominated by two, relatively heavy-bodied, numerically
dominant species: common murre and sooty shearwater, and
to a lesser degree, by the species identified as abundant in
the study area (see Table 15).
79
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
ABOUT THESE MAPS details on methods). Although the at-sea data span the years
A comparison of the abundance and distribution of marine birds
Diversity in Warm, Cold and Neutral Periods: 1980-2001 1980-2001, data are not available for all seasons in all years.
during warm-water periods (e.g., El Niño events), cold-water
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
periods (e.g., La Niña events) and normal (neutral) periods METHODS
Neutral Conditions
Warm-water Conditions
200 m
200 m
20
20
provides an example of how marine birds may respond to At-sea densities are the result of a synthesis of data from
0
0
0m
0m
short-term excursions from the usual marine climate. These eight shipboard and aerial survey programs conducted in
39°N
39°N
maps (Figure 58) show species diversity, calculated for each the study area in the years 1980-2001 (see “Data Sources”
Diversity five minute latitude by five minute longitude cell using density above). Observation and trackline data from these studies were
(H')
as the variable in the Shannon [Diversity] Index (Shannon converted to a common format. All aerial data were continuous;
2.001 - 3.000
and Weaver 1949); all 76 marine bird species that had been ship-based data were converted separately into a continuous
1.801 - 2.000
recorded in the data set were included. transect to the extent possible. From the digitized survey data,
38°N
38°N
1.501 - 1.800
the distributions of effort and of species were mapped into five
0.801 - 1.500
0.000 - 0.800 minute latitude by five minute longitude cells. The length and
The Shannon Index measures the degree to which a species
width of the survey trackline in a given cell (estimated trackline
assemblage is dominated by a few species. If a cell contains
width varied by platform, depending on speed and height above
0 25 50 Km
high densities of a few species and low densities of all others,
37°N
37°N
water) were used to estimate the area sampled. The number
the value of H’ will be low, indicating low diversity. Alternatively,
of marine birds seen in a cell was then divided by the area
if many species are present at similar densities, the value will
sampled in the cell to estimate density. If a cell was censused
be high, indicating high diversity. Maps a, b and c show the
more than once, densities were averaged, with adjustment
diversity index H’ in cold-water, neutral, and warm-water
made for effort.
periods; map d shows overall patterns. Cells are colored
36°N
36°N
based on the value of H’ computed for a particular season. Red
The Shannon Index (Shannon and Weaver 1949) was used to
indicates high diversity, blue indicates low diversity. Although
quantify species diversity.
there was a significant correlation between diversity and effort,
the observed patterns of bird diversity are robust and were n n
S
H ′ = − ∑ i ln i
largely unchanged by methods designed to correct for effort.
35°N
35°N
i =1 n n
a b
This index measures the degree to which the species
Unsurveyed areas are white. Blue lines indicate the National
Cold-water Conditions
200 m
200 m
Overall Patterns
20
assemblage is dominated by a single species. If species A
20
Marine Sanctuary boundaries of Cordell Bank, Gulf of the
0
0
0m
0m
dominates all the species seen within a cell, then diversity is
Farallones, and Monterey Bay; bathymetric contours for the
39°N
39°N
low; and vice versa. To standardize for variable effort among
200 meter and 2,000 meter isobaths are also shown in blue.
cells and variable strip width among species, we used the
density for each species in each cell as the basis for calculating
DATA SOURCES
the index value.
At-sea densities are based on data from eight survey programs
38°N
38°N
conducted in 1980-2001, which were combined using software
Marine bird density data was then organized into periods where
developed for MMS-CDAS (MMS, 2001) and expanded for this
surface ocean conditions were warm (including El Niños), cold
project. Of the data sets on the original MMS-CDAS CD-ROM,
(Including La Niñas) or neither (neutral). The diversity of all
four aerial survey data sets provided data in the study area
species seen within respective cells was determined for that
from Point Arena to Point Sal. Of these, one program was still
37°N
37°N
cell.
ongoing and data from recent years were added to this data
set. In addition, data from four ship-based survey programs
were converted to a compatible format for analysis. See section To illustrate these temperature conditions, a comparison of
introduction for details on individual data sets. marine bird densities was made by making maps that use
selected season/year periods that represented these cold,
36°N
36°N
Data sources for aerial at-sea data include MMS-CDAS (MMS, warm and neutral periods. The data for each condition map
2001) and California Department of Fish and Game Office of was grouped as shown below; these groupings were based
Spill Prevention and Response (CDF&G-OSPR), unpublished on the assignments made in Table 14. Once the selection of
data. Early data were collected using methods described by data were made for each analysis period (i.e., warm, neutral or
Briggs et al. (1983); more recent data were collected using cold), the density of all birds seen within each cell was summed
35°N
35°N
c d updated technology but the same general method. Data for that cell.
sources for ship-based survey data include David Ainley and
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Source Data: See text. Carol Keiper (unpublished data; see Oedekoven et al., 2001 for
Figure 58. Diversity in warm, cold and neutral periods: 1980-2001.
80
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
For the warm-water conditions (Including El Niños) map, the
following seasons and years were used: Davidson Current
Season: 1981, 1983, 1984, 1992, 1993, 1994, 1996, 1998.
Upwelling Season: 1985, 1987, 1992, 1993, 1995, 1998.
Oceanic Season: 1983, 1997.
For the neutral conditions map, the following seasons and years
were used: Davidson Current Season: 1982, 1986, 1995, 1997.
Upwelling Season: 1980, 1982, 1986, 1988, 1989, 1994, 1996,
1997. Oceanic Season: 1982, 1991, 1995.
For the cold-water conditions (Including La Niñas) map, the
following seasons and years were used: Davidson Current
Season: 1985, 1991, 1999, 2000, 2001, 2002. Upwelling
Season: 1981, 1990, 1991, 1999, 2000, 2001. Oceanic Season:
1980, 1981, 1994, 1996, 1998, 1999, 2000, 2001.
RESULTS AND DISCUSSION
Under all variations of climate, marine bird diversity was highest
over the continental slope, with the Farallon Escarpment
and Pioneer Canyon, in particular, standing out. Of lesser
importance was outer Monterey Bay Canyon and Point Sur
slope. Areas of high diversity were more spread out along
the slope when ocean temperatures were warm. Adding to
the latter hot spots was the area around Ascension Canyon.
During neutral conditions, diversity everywhere was relatively
low, when compared with higher diversities during the warm-
water and cold-water periods.
Although there was a significant correlation between diversity
and effort, the observed patterns of bird diversity are robust
and were largely unchanged by methods designed to correct
for effort.
81
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Observation and trackline data from these studies were con-
ABOUT THESE MAPS
Density During El Niño and La Niña Events, 1997-2000 verted to a common format. All aerial data were continuous;
A comparison of the density and distribution for two species
ship-based data were converted separately into a continuous
during an intense El Niño (1997-98) and an adjacent and in-
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
transect to the extent possible. From the digitized survey data,
tense La Niña (1999-00) provides an example of how marine
200 m
Brown Pelican Brown Pelican
200 m
20
20
the distributions of effort and of species were mapped into five
birds respond to short-term anomalies of marine climate (Figure
0
0
0m
0m
La Niña ('99 - '00)
El Niño ('97 - '98) minute latitude by five minute longitude cells using CDAS, a
59). In this comparison, the responses of individual species
39°N
39°N
custom geographic information system for analyzing marine
do not cancel out the effects of another, as was the case in
Density bird and mammal surveys (MMS, 2001). The length and width
comparisons when measures of overall abundance were used
(Animals/km²)
of the survey trackline in a given cell (estimated trackline width
(Figures 52, 53, 56 and 57).
> 100.00
varied by platform, depending on speed and height above
50.01 - 100.00
10.01 - 50.00
38°N
38°N
water) were used to estimate the area sampled. The number
Densities are based on combined data of several studies (see
5.01 - 10.00
of marine birds seen in a cell was then divided by the area
“Methods” and “Data Sources” below). The color and mapping
1.01 - 5.00
sampled in the cell to estimate density. If a cell was censused
intervals were customized to show the most structure and high-
0.51 - 1.00
more than once, densities were averaged, with adjustment
light significant areas. Cells that were surveyed but in which no
0.11 - 0.50
0.06 - 0.10 made for effort.
birds were observed have a density of zero; unsurveyed areas
0.01 - 0.05
37°N
37°N
are white. Blue lines indicate the National Marine Sanctuary
0.00
The most intense events were selected for this comparison,
boundaries of Cordell Bank, Gulf of the Farallones, and Mon-
0 25 50 Km
as well as events that occurred very close in time. In that way,
terey Bay; bathymetric contours for the 200 meter and 2,000
long-term changes in populations were not involved in the spe-
meter isobaths are also shown in blue.
cies’ occurrence patterns. For El Niño, data were used from
36°N
36°N
the Oceanic Season 1997 through Upwelling Season 1998; for
DATA SOURCES
La Niña the data were from the Oceanic Season 1998 through
At-sea densities are based on data from eight survey programs
Oceanic Season 1999; see Table 14.
conducted in 1980-2001. These data were combined using
CDAS software into an MMS-CDAS data system (MMS, 2001)
RESULTS AND DISCUSSION
for the Minerals Management Service and expanded for this
35°N
35°N
a b During intense warm periods, species such as brown pelican,
project. Of the data sets on the original MMS-CDAS CD-ROM,
black storm-petrel and black-vented shearwater, which zoo-
four aerial survey data sets contained data in the study area
200 m
Black-vented Shearwater Black-vented Shearwater
200 m
20
20
geographically are centered to the south of central California
from Point Arena to Point Sal. Of these, one program was still
0
0
0m
0m
(where waters are normally warmer and food availability rela-
ongoing and data from recent years were added to this data
El Niño ('97 - '98) La Niña ('99 - '00)
39°N
39°N
tively lower), move into central California waters when warmer
set. In addition, data from four ship-based survey programs
ocean temperatures expand northward. Many of these individu-
were converted to a compatible format for analysis. See section
als have foregone breeding owing to depleted food availability,
introduction for details on individual data sets.
which is often more extreme in areas to the south where these
species breed. Shown here are comparisons for brown peli-
Data sources for aerial at-sea data include MMS-CDAS (2001)
38°N
38°N
can and black-vented shearwater. In both cases, densities are
and California Department of Fish and Game Office of Spill
much higher in central California during warm-water periods.
Prevention and Response (CDF&G-OSPR), unpublished data.
In fact, during these conditions brown pelicans expand as far
Early data were collected using methods described by Briggs
north as the Columbia River and even farther; black-vented
et al. (1983); more recent data were collected using updated
37°N
37°N
shearwaters, however, don’t go much farther than central
technology but the same general method. Data sources for
California waters.
ship-based survey data include David Ainley and Carol Keiper
(unpublished data; see Oedekoven et al., 2001 for details on
The response of species to short-term cold-water conditions (La
methods). Although the at-sea data span the years 1980-2001,
Niña) is far less dramatic and, in fact, no examples could be found
data are not available for all seasons in all years. For the Up-
36°N
36°N
to clearly illustrate this. This is due to many factors, perhaps the
welling Season, data are from 1980-1982 and 1985-2001.
most important being that the geographic affinity of central California
For the Oceanic Season, data are from 1980-1982, 1991 and
marine birds is largely ‘subarctic’ and therefore, the central California
1994-2001. For the Davidson Current Season, data are from
avifauna is at the southern extreme of its range. As a result, there is
1980-1986 and 1991-2001.
little reason for northern species to shift into the area when the latter
35°N
35°N
c d becomes slightly colder. The other main reason for lack of examples
METHODS
illustrating response to cold conditions is that a major regime shift
At-sea densities are the result of a synthesis of data from eight
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
coincided with the best example, i.e. intense La Niña conditions
shipboard and aerial survey programs conducted in the study
Source Data: See text.
in 1999-00 (see Figure 60). Therefore, it is difficult to separate the
area in the years 1980-2001 (see “Data Sources” above).
Figure 59. Density during El Niño and La Niña events, 1997-2000.
factors responsible in the avifaunal shifts observed.
82
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
(estimated trackline width varied by platform, depending on
ABOUT THESE MAPS
Density During El Niño and La Niña Events: 1997-2000 speed and height above water) were used to estimate the area
A comparison of the abundance and distribution of two species
sampled. The number of marine birds seen in a cell was then
during intense El Niño events compared to La Niña events
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
divided by the area sampled in the cell to estimate density. If a
provides an example of how marine birds may respond to
200 m
Fork-tailed Storm-Petrel Fork-tailed Storm-Petrel
200 m
20
20
cell was censused more than once, densities were averaged,
short-term anomalous marine climate. In the case of regime
0
0
0m
0m
with adjustment made for effort.
shifts, which involve climate change on a decadal time scale, a
La Niña ('99 - '00)
El Niño ('97 - '98)
39°N
39°N
shift may have occurred during the study period corresponding
Density Species densities were mapped by grid cells on either side of
also to the switch from intense El Niño (Oceanic Season 1997
(Animals/km²)
the regime shift node, Oceanic Season 1997 through Upwelling
- Upwelling Season 1998) to intense La Niña (Oceanic Season
> 100.00
Season 1998 versus Oceanic Season 1998 through Oceanic
98 - Oceanic Season 2000). Therefore, at this time, it is difficult
50.01 - 100.00
Season 2000.
38°N
38°N
to perceive whether the changed bird distributional patterns
10.01 - 50.00
5.01 - 10.00
were short-term or long-term. Subsequently the cold conditions
1.01 - 5.00
RESULTS AND DISCUSSION
continued, thus indicating a longer-term regime shift (Bogard et
0.51 - 1.00
In 1999-2000, the mean state of the California Current System
al., 2000, Schwing and Moore 2000, Schwing et al. 2002).
0.11 - 0.50
may have moved from a “warm regime”, present since 1976,
0.06 - 0.10
0.01 - 0.05
37°N
37°N
to a “cold” regime (Schwing et al. 2002). On the other hand,
Densities are based on combined data of several studies (see
0.00
subsequent years of observation may indicate that we only
Methods and Data Sources below). The color and mapping
0 25 50 Km
witnessed the transition from one of the strongest El Niños to
intervals were customized to show the most structure and
one of the strongest La Niñas seen in the past 100 years.
highlight significant areas. Cells that were surveyed but in which
no birds were observed have a density of zero; unsurveyed
36°N
36°N
Regardless, in response, more northern species such as
areas are white. Blue lines indicate the National Marine
fork-tailed storm-petrel and black-legged kittiwake, which are
Sanctuary boundaries of Cordell Bank, Gulf of the Farallones,
present mostly during the Davidson current season, found
and Monterey Bay; bathymetric contours for the 200 meter and
the cooler, central California waters more to their liking and,
2,000 meter isobaths are also shown in blue.
rather than avoiding the area as in the 20 previous years of the
35°N
35°N
a b warm regime, arrived or remained longer to winter in very large
DATA SOURCES
numbers. Unfortunately, data during the Oceanic and Davidson
At-sea densities for this analysis are based on a subset of data
200 m
Black-legged Kittiwake Black-legged Kittiwake
200 m
current seasons of years since 2000 were not collected.
20
from the eight survey programs. These data were combined
20
0
0
0m
0m
using CDAS software into an MMS-CDAS data system (MMS,
La Niña ('99 - '00)
El Niño ('97 - '98)
39°N
39°N
2001) for the Minerals Management Service and expanded for
this project. Of the data sets on the original MMS-CDAS CD-
ROM, four aerial survey data sets contained data in the study
area from Point Arena to Point Sal. Of these, one program was
still ongoing and data from recent years were added to this data
38°N
38°N
set. In addition, data from four ship-based survey programs
were converted to a compatible format for analysis. See section
introduction for details on individual data sets. Data collected
since 1996 was used for this comparison.
37°N
37°N
METHODS
At-sea densities for this analysis are the result of a synthesis of
subsetted data from eight shipboard and aerial survey programs
conducted in the study area in the years 1980-2001 (see Data
36°N
36°N
Sources above). Observation and trackline data from these
studies were converted to a common format. All aerial data
were continuous; ship-based data were converted separately
into a continuous transect to the extent possible. From the
digitized survey data, the distributions of effort and of species
35°N
35°N
c d were mapped into five minute latitude by five minute longitude
cells using CDAS, a custom geographic information system
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Source Data: See text. for analyzing marine bird and mammal surveys (MMS, 2001).
The length and width of the survey trackline in a given cell
Figure 60. El Niño/La Niña Event changes, as an example of regime shift effects.
83
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
changes greatly due to the presence of southern
SECTION SUMMARY Table 15. Life history and management information for selected marine birds off north/central California.
Major Food Items3
Temporal Occurrence in Study Area hemisphere-breeding species that are ‘wintering’ in
The following section provides a summary discussion of Population Status in Study Area
large numbers in the study area during the Upwelling
the marine bird analyses, relative to the study area and
(squid, octopus)
Cephalopods
Invertebrates
Season and subarctic-breeding species ‘wintering’ in
the three national marine sanctuaries off north/central
Euphausiids
Gemeralist
large numbers during the Davidson Current Season.
California.
Plankton
Relative Primary Months
Benthic
Feeder
Additionally, many migrants pass through the region,
Protection Population Abundance Seasonal Months of When
(krill)
Fish
but foraging as they go, during the Oceanic and early
Life History and Management Characteristics Status1 Trend6 at Sea2 Occurrence5 Presence5 Breeding5
Common Name Scientific Name
Upwelling seasons.
The marine avifauna off north/central California, as Loons/Grebes
Pacific loon Gavia pacifica Unknown Common Seasonal Nov-Apr x
represented in the summary data set, are composed of
Common loon Gavia immer Unknown Uncommon Seasonal Nov-Apr x
In general, the highest concentrations and the greatest
76 marine bird species, with 39 occurring regularly enough Western & Clark's grebes Aechmorphorus occidentalis, A. clarksii Unknown Abundant Year-round Nov-Sept Apr-Sept x
variety of marine birds are found over the continental
to assess and map patterns of their occurrence. Table 15 Albatrosses/Petrels
shelf and slope, where there are more microhabitats
is a summary of selected life history and management Black-footed albatross Phoebastria nigripes Decreasing Common Year-round Mar-Aug x x
defined by ocean complexity (depth, currents, tide rips,
information for 39 of the marine bird species. Laysan's albatross Phoebastria immutabilis Unknown Rare Seasonal Nov-Mar x x
Northern fulmar Fulmarus glacialis Increasing Common Seasonal Nov-Mar x
etc.). A number of smaller, more discrete areas attract
Sooty shearwater Puffinus griseus Declining V.Abundant Seasonal Apr-Aug x x x x
marine birds because more food is available; the most
Species Relative Abundance in the Study Area. Based on Pink-footed shearwater Puffinus creatopus Stable? Common Seasonal Apr-Aug x x
important of these are Farallon Escarpment, Farallon
the analysis of the at-sea data and as indicated in Table Buller's shearwater Puffinus bulleri Unknown Common Seasonal Aug-Nov x
Ridge, San Francisco Bay tidal plume, inner Monterey
15, among the more regularly occurring species, two are Black-vented shearwater Puffinus opisthomelas Stable? Uncommon Seasonal Aug-Nov x?
Leach's storm-petrel Oceanodroma leucorhoa Unknown Common Seasonal March-Sept April-Sept x x x Bay, and Estero/San Luis Obispo bays.
very abundant, eight are abundant, 16 are common, 12
Ashy storm-petrel Oceanodroma homochroa SSC Increasing Common Year-round All Feb-Oct x x x
are uncommon, and two are rare. Relative abundance was Fork-tailed storm-petrel Oceanodroma furcata Decreasing? Uncommon Seasonal Nov-Mar x x x
Patterns Observed in Density, Biomass Density and
estimated on a logarithmic scale of number of individuals Black storm-petrel Oceanodroma melania SSC Unknown Uncommon Seasonal Nov-Mar x x
Diversity Across Species
seen within the study area on surveys during the study Sea Ducks (Scoters)
Another way to summarize occurrence patterns of
period. The majority of the species (26) are present only Surf scoter Mellanita perspicillata Stable Abundant Seasonal Nov-Apr x
Pelican/Cormorants
marine birds in the study area is to combine species
seasonally, but of the 14 species that breed in the study
California brown pelican Pelecanus occidentalis californicus FE, SE Increasing Common Seasonal Aug-Nov x
distribution and abundance data and analyze for the
area, 10 are present year round. Pelagic cormorant Phalacrocorax pelagicus Stable? Uncommon Year-round All Apr-Sept x x x x
biological metrics of species diversity, biomass and
Brant's cormorant Phalacrocorax pennicilatus Stable? Abundant Year-round All Apr-Sept x
density. Analyses for overall density, biomass and
Food Types. With regard to trophic relationships, the Double-crested cormorant Phalacrocorax auritus Increasing? Uncommon Year-round All Mar-Aug x
diversity were done with respect to ocean season
majority of marine bird species are either zooplanktivores Phalaropes
Red phalarope Phalaropus fulicaria Stable? Common Seasonal Mar-Aug x x x and to periods of unusual ocean climate (i.e., warm-
(generally smaller-bodied) and/or piscivores. Major prey
Red-necked phalorope Phalaropus lobatus Stable? Common Seasonal Mar-Aug x x x
water, cold-water and neutral periods). For these
items are: euphausiids (Thysanoessa spinifera and Gulls/Terns
summary analyses across species, we used the data
Euphausia pacifica), market squid (Loligo opalescens), Western gull Larus occidentalis Declining Abundant Year-round All Apr-Aug x x
for 76 marine bird species that were contained in the
juvenile rockfish (Sebastes spp, especially Sebastes California gull Larus californicus Increasing Abundant Seasonal Nov-Mar x
combined data set.
jordani), anchovy (Engraulis mordax), herring (Clupea Glaucous-winged gull Larus glaucescens Stable Uncommon Seasonal Nov-Mar x
Heermann's gull Larus heermanni Stable? Common Year-round Aug-Nov x
harengus), smelt (Atherinops californiensis and Spirinchus
Sabine's gull Xerna sabini Stable Common Seasonal Mar-Aug x x
Overall Density and Biomass. The seasonal and
starksi), Pacific saury (Cololabis saira), sardine (Sardinops Black-legged kittiwake Rissa tridactyla Increasing Common Seasonal Nov-Mar x
‘combined-season” densities of 76 marine birds were
sagax), midshipman (Porichthys notatus), surfperch Caspian tern Sterna caspia SSC Stable Uncommon Seasonal Mar-Nov Apr-Aug x
calculated for each five-minute by five-minute cell as
(several species) and myctophids (several species), with Elegant tern Sterna elegans Stable Uncommon Seasonal July?-Nov x
Arctic tern Sterna paradisaea Stable? Common Seasonal Mar-Nov x the number of individuals per km2. Biomass (kg/km2)
importance varying by the habitat and time of year in which
Alcids
was then calculated as the product of density and the
a particular bird species was foraging (Briggs and Chu
Common murre Uria aalge Increasing V.Abundant Year-round All Apr-Aug x x x x
mean body mass for each species, taken from Dunning
1987, Ainley and Boekelheide 1990). Pigeon guillemot Cepphus columba Stable? Uncommon Seasonal Mar-Aug Mar-Aug x x x
(1993). If a species was not listed in this reference,
Cassin's auklet Ptychoramphus aleuticus Decreasing? Abundant Year-round All Mar-July x x
the body weight of a related species of a similar size
Summary of Spatial and Temporal Patterns in Large Marbled murrelet Synthliboramphus marmoratus FT, ST Stable Uncommon Year-round All Apr-Aug x x
Xantus’ murrelet Synthliboramphus hypoleucus ST Unknown Uncommon Seasonal May-July x
was used.
and Relatively Smaller Areas
Craveri’s murrelet Synthliboramphus craveri Unknown Rare Seasonal May-July x
Table 16 is a summary of the temporal and spatial patterns Tufted puffin Fratercula cirrhata Decreasing Uncommon Seasonal Mar-Aug Apr-Aug x x
The distribution of marine birds across all taxa is similar
observed for the regularly occurring marine birds of the Rhinoceros auklet Cerorhinca monocerata Stable Abundant Year-round Nov-Aug Apr-Aug x x x x
for density (Figure 52) and biomass density (Figure 53).
study area. This summary was developed by visual Notes
This is because the avifauna is dominated (in terms of
inspection of the species seasonal density maps and is 1. Management status categories are as follows: FE–federally endangered; FT–federally threatened; SE–state endangered; ST–state threatened; SSC–state species of special concern.
2. Relative abundance estimates are based on the number of individuals tallied in the at-sea survey data, and the categories are defined as follows:
both number of individuals and their body mass) by the
provided as a simple summary of species distributions
Rare – up to 100 birds; Uncommon – up to 1,000; Common – up to 10,000; Abundant – up to 100,000; and Very Abundant – up to 1,000,000.
Common Murre and Sooty Shearwater. Therefore, the
and abundance by season and for selected habitat and 3. Information on food items are from Ainley & Sanger 1979, Briggs & Chu 1987, and Ainley & Boekelheide 1990.
patterns in sum are close to what is evident individually
management features. 4. Entries with question marks are the principal investigators best estimate.
for these particular dominant species. Accordingly, the
5. Timing information is from Cogswell, 1977 and Ainley and Boekelheide, 1990, except for Caspian tern breeding time, which came from Joelle Buffa, FWS, pers. comm.
major biomass and density areas i.e., the inner and
6. Information on population status was based on analysis of the shipboard data sets, from 1985-2001.
It is obvious that large numbers of marine birds occur in the
7. Months of presence and breeding in the study area are approximations, because timing is strongly influenced by the interannual variability of environmental conditions in the study area.
outer shelf, are biased toward these two species.
study area year round. The species composition, however, 8. Information on population status and temporal occurrence refers only to birds and their activities in the study area.
84
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Phalaropes can also be very abundant but don’t contribute
much biomass and they are also most abundant over
Table 16. A summary of temporal and spatial patterns in the at-sea survey data (1980-2001) of selected marine birds off north/central California.
the shelf waters. Smaller-scale biomass and density hot
spots are also the same, e.g. inner Monterey Bay, San
Species
Francisco Bay tidal plume, the area around the Farallon
Occurs in
Associations with Large Study
Islands, Pioneer and Ascension Canyons, and Cordell
Bathymetrically-Defined Areas
Seasonal Occurrence Associations with Discrete Physiographic/Oceanic Features
Area, But
Bank. Moreover, as will be noted later (see also below)
Mostly
Davidson Ascen- Estero
Outside
density and biomass are more spread out during warm-
Ocean Current Coast & Outer Upper Lower Deep San Fran- cion, Mon- David- Bay/San
Sanc-
Upwelling Season Season Inner Shelf Slope Slope Ocean Farallon cisco Pioneer Cabrillo, Pt. Mon- terey Pt. son Luis
water than during cold-water or neutral periods.
Family Name/ tuary
Season (8/15- (11/15- Shelf (~0 (~100- (~200- (~1000- (beyond Cordell Bodega Fanny Escarp- Farallon Bay Sea Pioneer & Año Ano terey Bay Carmel Pt. Sur Sur Sea- Obispo
Species Common Name Bounds Bank Canyon Shoal ment
(3/15-8/14) 11/14) 3/14) 100m) 200m) 1000m) 2000m) 2000m) Ridge Plume mount Canyon Canyons Nuevo Canyon Inshore Canyon Slope Shelf mount Bay
Podicipedidae
Seemingly, highest biomass occurred during the Upwelling
Western and Clark's grebes x x **X X X x X X
and Oceanic seasons, as was the case for density. Any
Gaviidae
Pacific loon X ***X x x x X X
seasonal difference was least clear in regard to density.
Diomedeidae
Black-footed albatross ***X x x X X X x X X X x x X x x
Laysan albatross x o ***X X X X x X X
Diversity. To assess species diversity, the Shannon Index
Procellariidae
(Shannon and Weaver 1949) and species density data
Northern fulmar x x ***X X X X x x x x X x x x x x x
Sooty shearwater ***X x o X X X x x X x x x x X X X X X x X
were used (see Figure 54). This index measures the
Short-tailed shearwater o o ***X X X x x X x
degree to which the species assemblage is dominated
Pink-footed shearwater ***X x o X X X x x x x X x x X x
Buller's shearwater o ***X o X X X x X x X x x x x x x
by a single species. For example, if “Species A” dominates
Black-vented shearwater o **X x X X x X x x x
all the species seen within a cell, then diversity is low;
Hydrobatidae
Leach's storm-petrel **X **X o X X X x x x x
and if all species are “evenly” represented, then diversity
Fork-tailed storm-petrel o o X x X X x X X x x x
is high. Diversity was calculated using all bird species
Ashy storm-petrel **X ***X **X x X X x X x X X x x X x
Black storm-petrel o **X **X x X X x X x x X x
(n=76) in the data set, for each ocean season and for all
Anatidae
ocean seasons combined.
White-winged scoter o x **X X X X X
Surf scoter o x **X X X X X
Pelecanidae
Highest diversity indices are about the same in all
Brown pelican **X **X x X x x x X X X x X
three seasons. In all cases, at the smaller spatial scale
Phalacrocoracidae
Brandt's cormorant ***X ***X ***X X X x x x X X X X x x
(less-detailed), species diversity was greatest along
Pelagic cormorant **X **X **X X x X x x x
the continental slope. This is expected given that the
Double-crested cormorant *X o o X x X X x x
Scolopacidae
slope habitat constitutes the boundary as well as the
Red phalarope ***X x o X X X x X x X x x x x x
overlap between the shelf and oceanic habitats. At a
Red-necked phalarope ***X x o X X x x x x x x x x x x x
Laridae
larger scale (more detailed), in all seasons there was an
Glaucous-winged gull o x ***X X X x x x x x X x x X X
area of notable diversity seaward of the Farallon Islands
Western gull ***X ***X ***X X X X x x x X X X X X X x X
California gull o x ***X X X x x X x X x X
(Farallon Escarpment) and to some degree outside of the
Ring-billed gull o x **X X x X X x X
sanctuary boundaries. Likely the diversity here resulted
Mew gull o x **X X x X X
Heermann's gull o x **X X x X X
from a coincidence of occurrence of: 1) oceanic species;
Bonaparte's gull ***X x o x X X x X x X
2) shelf species; 3) Farallon breeding species that would
Sabines gull ***X x o X X X x X x x X
Black-legged kittiwake x o ***X X X X x X x x X x x x x x x x
not occur offshore were it not for the Farallones; and 4)
Caspian/Elegant terns **X x o X X X x
the location of a persistent boundary there of a coastal
Arctic tern ***X ***X o x x X X x X x x x x
Forster's tern **X x o X X X
upwelling front that extends southwestward from Point
Alcidae
Arena. Accordingly, during La Niña, when upwelling
Common murre ***X ***X ***X X X x x x x x X X x x x x
features are well developed, this area exhibits much
Pigeon guillemot **X x o X X X x X X
Tufted puffin *X x o x X X X X X x
greater diversity than is apparent during El Niño.
Rhinoceros auklet ***X ***X x x X X x x x x x x X x x x
Cassin's auklet ***X ***X x X X x X x X X X x x
Marbled murrelet *X *X *X X x X X
Although there was a significant correlation between bird
Xantus'/Craveri's murrelets ***X x x X X X X X
diversity and effort, the observed patterns of bird diversity
All Xs 29 35 27 21 28 24 22 16 11 13 14 14 22 17 20 6 18 19 11 15 18 6 8 8 6 16
Large X's 25 14 21 20 19 19 13 5 11 3 1 4 16 9 17 0 2 1 5 7 16 0 8 2 0 10
are robust and were largely unchanged by methods
Notes
1. A summary of temporal and spatial patterns in the occurrence of 44 marine birds off north/central California, based on visual inspection of the species' seasonal density maps by the principal investigators.
designed to correct for effort. See an additional discussion
2. The spatial and temporal patterns summarized here may be affected by variation in the sampling effort of the combined data sets; that said, this table is included because it provides a summary of the relative use of the various habitat features, as viewed in the species maps.
of bird diversity in the Integration section.
3. In the "seasonal occurrence" columns, the number of asterisks indicate the number of sanctuaries in the study area that are used by the species during the season. A large "X" means a relatively major occurrence in the data sets, a small "x" means a minor occurrence,
and a "o" means the species was mostly absent.
4. Under the heading for "large, bathymetrically-defined areas", a large "X" indicates where the species was most abundant, a small "x" means a minor occurrence, and a "o" means the species was mostly absent.
5. Under the heading for "discrete physiographic/oceanic features", large X’s refer to ‘hot spots’; small x's indicate areas of secondary importance for that species.
6. A blank in the table means a species was not present at the location indicated in the maps/data reviewed.
85
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Important Marine Areas for Birds: Considering Overall Shearwater is one of these and becomes the most abundant Biomass remains high but shifts closer to shore than during The next three variables of importance were “ENSO” (eight
Biomass, Density and Diversity of Marine Birds. All species in the study area. the Upwelling Period, in large part due to an inshore shift of species), “Year” (seven species), and “Distance to Colony”. In
marine habitat off central California, especially that of the murres and shearwaters. some respects, for species having many small colonies (e.g.
continental shelf and slope, is fully used by marine birds. Owing to the addition of Sooty Shearwaters to the avifauna and pelagic cormorant), Distance to Land and Distance to Colony
Based on the analyses of maps for overall density, biomass the continued abundance of Common Murres, overall density Davidson Current Season (~Winter). During this season, may have co-varied. The variable “Year” indicated whether
density, and diversity, the following at-sea areas were identified and biomass of marine birds is highest during the Upwelling when ocean temperatures are relatively warm and there is there was an increasing or decreasing trend in the species’
as important (Table 17). Areas with more and bigger Xs may and Oceanic Seasons and is widely spread from the coast no upwelling (but frequent downwelling owing to southerly abundance. An effect of ENSO would indicate an especially
be more important, as they show more expression of density, to beyond the shelfbreak. Diversity over the shelf, where the storms) the area is inundated by such species as black-legged complex relationship, possibly meaning either an effect of prey
biomass and diversity. shearwaters and murres mostly reside, is relatively low. kittiwake, northern fulmar and several larger gulls. All these availability or ocean climate.
species nest outside of the region. Also present are nesting
Important Breeding Colonies for Marine Birds. Although Oceanic Season (~Autumn). In this season, when upwelling species that reside year-round in the region, such as Brandt’s Please note that while the three most important variables of
breeding colonies and roosts are on land and technically not winds have noticeably relaxed, allowing offshore, warmer cormorant, western gull, common murre, rhinoceros auklet those evaluated are indicated in Table 19, for most species
part of the study area, a table and map of the major colonies oceanic water to flow shoreward, the avifauna begins to and Cassin’s auklet. In fact, many of the latter species begin there are likely other variables of greater importance (e.g.,
are included in this section because they provide a context for diversify. However, as more sooty shearwaters and other to occupy nesting colonies during this season, well before the prey availability, depth of thermocline). On average, only
understanding the distributions of species that breed or roost southern hemisphere seasonal residents depart the avifauna nesting period. about 20% of the variance was explained with the top three
in the study area based on size and species composition (most variables presented. Additional data and time would be required
data was from Carter et al., 1992, with updates, as available). During this season, to evaluate other variables that might be of greater importance
Table 17. Important at-sea areas and ocean seasons for marine birds off north/central California, based on maps
Table 18 shows the top 40 marine bird breeding colonies in the the species diversity in explaining the variation in a species’ distributions.
of biomass, density and diversity.
study area; see Figure 55 for a map of these locations; see the of shelf waters
Biomass/Density Diversity
CD-ROM X for a full listing of breeding sites. increases. In fact, Species Use of the Water Column. Several marine bird
Davidson Davidson
areas of high species are capable of exploiting the entire water column of the
Current Upwelling Oceanic Current Upwelling Oceanic
Marine birds in this area breed mainly during the Upwelling diversity are more shelf, e.g. Pacific loon, western/Clark’s grebes, and common
Season Season Season Season Season Season
Area
Season, anticipating that food availability will be greatest widespread in this murre, but for unknown reasons (possibly prey selection,
x X
Cordell Bank
from July-October, toward the end of this season and into the season than in the perhaps interference competition from murres and shearwaters)
Farallon Escarpment X x x X X X
Oceanic Season. During this period ample supplies of prey will others. the grebes mainly frequent the inner most portion of the shelf.
(slope)
be needed to feed growing chicks and recently fledged young. The very abundant common murre is found everywhere on the
Farallon Ridge (includes X X x
Farallon Island area)
Egg laying occurs in March-May, depending on species, and Analysis of shelf especially during the breeding/Upwelling season. Other
different species require different amounts of time to complete Variation diving species, such as scoters or marbled murrelet, frequent
San Francisco Bay Tidal x X X
Plume
the breeding task (petrels longest, gulls shortest). in Species only shallow waters of the inner shelf, while other species,
Pioneer Canyon X x x Abundance such as tufted puffin, rhinoceros auklet and Cassin’s auklet
Año Nuevo Shelf X X
The greatest concentration of colonies occurs in the Gulf of Patterns. Many frequent waters much deeper (continental slope) than their
the Farallones, in the broad shelf area from Point Reyes south factors influence diving capabilities allow. Species such as the very abundant
Ascension, Año and X x x X x
Cabrillo Canyons
to Año Nuevo and out to the Farallon Islands. The breeding the distribution sooty shearwater (a shallow diver to 20 meters) are found
Monterey Bay Inshore X x X X
avifauna is dominated by alcids, with six species. Fifteen and abundance everywhere from outer slope to inner shelf.
Monterey Bay Canyon X x
species of marine birds breed at sites within or immediately of marine birds;
adjacent to the National Marine Sanctuaries in the study area; Carmel Canyon X in this study, the These differences in patterns of habitat use are likely related
several others breed inland or in San Francisco Bay and to a effects of nine to factors such as the occurrence patterns of different
Point Sur Shelf x
lesser degree use marine sanctuary waters. independent prey (species/sizes), interspecific competition, or temporal
Point Sur Slope X x
variables on occurrence of certain prey (species/sizes). The latter would
Estero Bay & San Luis X x
Importance of Ocean Seasons to Marine Birds. As seen in species density account for why some year-round resident species feed over
Obispo Bay
the maps for individual species, temporal differences in spe- were investigated waters of different depths during one season compared to
Note: Large, bold Xs refer to most important areas, and smaller xs refer to other important areas.
cies occurrence patterns are strong for many species in the for 26 species. another. Species such as the sooty shearwater, which use a
study area (see Tables 15 and 16). Below is a brief summary The data used for wide range of ocean depths and habitats, are likely to be more
of marine bird activity in the three ocean seasons. the regression analyses were a subset of the mapped generalized in prey selection, possibly due to their fast, efficient
becomes more sparse. At this time, resident breeding species
data, and included data from the Davidson Current Period flight allowing them to forage over much larger areas than many
are also dispersing more widely as they finish breeding
Upwelling Season (~Spring/Summer). With the onset of upwell- from 1985 through the same for 2002; also, cells with area other marine birds, particularly the alcids and cormorants.
duties.
ing, when cold, nutrient-rich water is brought to the surface by surveyed less than 0.25km2 were excluded.
persistent northwest wind and the Coriolis effect, most of the Response to “Short-Term” Changes in Ocean Climate. The
Additionally, several species (e.g., phalaropes, jaegers, Arctic
seasonal winter residents depart and several other species Among the nine variables investigated, the three most study area is subjected frequently to shifts in marine climate of
terns, Sabine’s gulls) are migrating through the region, the local
migrate through the region (e.g. Sabine’s Gull, Arctic Tern). Ar- important variables that explained variation in species density different scales and periodicity and this makes management
nesting species are all present, and several species that nest
riving are several species that nest in the Southern Hemisphere, were “Distance to Land” (16 of 25 species), “Ocean Season” a challenge, because populations are affected by natural
elsewhere are abundant as well (e.g. several shearwaters,
thus spending their ‘wintering’ period in the region. The Sooty (13 species), and “Ocean Depth” (11 species, see Table 19). environmental factors that cannot be addressed proactively
brown pelican, and Heermann’s gull).
86
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Table 18. Major marine bird colonies along the central California coast
Double- Other Site Total
USFWS Leach's Ashy Brandt's crested Pelagic Pigeon Rhino- Species (No. of
CA Colony Colony Storm- Storm- Cor- Cormor Cor- Western Caspian Common Guille- Marbled Cassin's ceros Tufted (see Breeding
Colony/Composite Name Number Number Latitude Longitude Petrel Petrel morant ant morant Gull Tern Murre mot Murrelet Auklet Auklet Puffin notes) Birds)
Fish Rocks ME-384-10 404-003 38°47'59" N 123°35'31" W 100 211 123 170 119 P 4 15 6 748
Gualala Point Island SO-384-01 404-004 38°45'3" N 123°31'42" W 521 4 26 29 1 581
Russian Gulch SO-382-08 404-033 38°28'0" N 123°9'35" W 227 42 20 7 296
Russian River Rocks SO-382-09 404-005 38°27'14" N 123°8'34" W 51 422 125 44 5 2 649
Arched Rock SO-382-11 404-006 38°25'53" N 123°7'31" W 717 9 34 2 762
Bodega Rock SO-380-02 404-008 38°17'48" N 123°2'49" W 1,228 24 30 2 1,284
Bird Rock MA-380-04 404-010 38°13'49" N 122°59'35" W 15 55 37 168 115 3 H 6 399
Point Resistance MA-374-03 429-024 37°59'54" N 122°49'40" W P 46 H H 8 3,518 50 3,622
Point Reyes MA-374-01 429-001 37°59'30" N 122°58'59" W 15 1,522 266 178 15,155 616 4 6 17,762
Millers Point Rocks MA-374-04 429-002 37°58'52" N 122°48'34" W 114 59 30 358 55 1 617
Double Point Rocks MA-374-05 429-003 37°56'50" N 122°47'8" W 30 180 9 8 4,464 22 4,713
San Francisco Bay & Alcatraz Island Composite Composite 37°49'33" N 122°25'19" W 9 2,789 4 4,174 2,818 10 8498 18,302
North Farallon Islands SF-FAI-01 429-051 37°46'4" N 123°5'56" W 161 62 32 27,308 42 27,605
South Farallon Island SF-FAI-02 429-052 37°42'0" N 123°0'0" W 1,400 1,990 9,466 486 442 15,095 103,588 499 18,807 516 128 30 152,447
Devil's Slide Rock SF-372-03 429-014 37°34'28" N 122°31'39" W 7 46 16 246 30 345
Big Basin State Park and vicinity None None 37°8'48" N 122°17'58" W 600 600
Vicinity of Año Nuevo Island and Point Composite Composite 37°6'30" N 122°20'8" W 4 117 1,382 219 24 224 27 1,997
Greyhound Rock to El Jarro Point SC-370-01 429-049 37°3'31" N 122°15'0" W 66 2 321 9 398
El Jarro Point to Davenport SC-370-02 429-050 37°1'5" N 122°12'23" W 308 22 313 1 644
Davenport to Sand Hill Bluff SC-364-01 454-038 36°59'44" N 122°10'32" W 13 495 1 509
Cannery Row MO-362-02 454-044 36°36'46" N 121°53'47" W 198 86 88 372
Bird Rock MO-362-03 454-006 36°35'30" N 121°57'59" W 2,651 16 2 2,669
Guillemot Island Area MO-362-06 454-023 36°31'45" N 121°56'47" W 554 20 30 18 10 632
Bird Island MO-362-09 454-009 36°30'24" N 121°56'32" W 6,151 4 90 5 2 6,252
Castle Rocks and Mainland MO-362-19 454-010 36°22'35" N 121°54'25" W P 750 46 12 1,050 19 1,877
Hurricane Point Rocks MO-362-20 454-011 36°21'40" N 121°54'25" W P 29 H 14 613 20 H 676
Anderson Canyon Rocks MO-360-13 454-016 36°9'7" N 121°39'53" W 321 5 26 2 32 386
Burns Creek Rocks MO-360-14 454-017 36°8'29" N 121°39'28" W 323 2 12 337
Plaskett Rock MO-354-07 477-002 35°55'13" N 121°28'41" W 849 H 5 H 1 855
Cape San Martin MO-354-08 477-003 35°53'16" N 121°27'54" W 750 2 18 349 H 1,119
Redwood Gulch Rock MO-354-12 477-005 35°50'19" N 121°24'3" W 372 2 H 374
La Cruz Rock SL-354-04 477-006 35°42'22" N 121°18'45" W 678 18 696
Piedras Blancas Island SL-352-01 477-007 35°39'51" N 121°17'17" W 2,627 34 29 3 H 1 2,694
Morro Rock and Pillar Rock SL-352-07 477-026 35°22'13" N 120°52'8" W 117 24 53 114 24 332
Fairbank Point SL-352-08 477-044 35°21'5" N 120°50'37" W 331 331
Unnamed Rocks SL-350-03 477-010 35°14'40" N 120°53'38" W 174 49 242 6 471
Lion Rock SL-350-05 477-011 35°13'3" N 120°52'16" W 277 H 24 18 1 320
Pup Rock and Adjacent Mainland SL-350-04 477-028 35°13'17" N 120°52'13" W 1,309 44 2 1,355
Pecho Rock SL-350-09 477-032 35°10'45" N 120°48'59" W 321 14 335
Table Notes
1. This table contains numbers of breeding birds at specific colonies or composite sites, for the species indicated. The table shows the best available data for approximately 40 of the largest colonies and colony composites for selected marine birds that occur in the study area.
All colonies shown have 296 or more breeding birds, and sites are listed from north to south.
2. The primary source for these data is Carter et al. 1992 (unpublished data); most estimates from this source were made from 1989-1991. Older data older data are indicated by italics (e.g., data for Leach's storm-petrel), and more recent or updated data (from various sources,
identified below) are indicated in bold type.
3. Key to symbols in table: H=historically nesting species; P = present and probably breeding. A blank in the table for a species/colony cell means the species was not present in the available data.
4. The column titled "Other Species" contains available estimates for all other breeding bird species. For most sites, this includes Black Oystercatcher. For the San Francisco Bay/Alcatraz composite site, the "Others" estimate includes California Gull, Forster's Tern, and Least Tern,
which breed at sites in the Bay.
5. For Ashy Storm-petrel, the updates at Bird Rock, Point Reyes and Double Point are from 2001 (Whitworth et al. 2002). The update at South Farallon Island is from 1992 (Sydeman et al. 1998).
6. The estimate of 600 breeding Marbled Murrelets at Big Basin State Park and Vicinity was provided by Laird Henkel, pers. comm.
7. For Cassin's Auklet and Rhinoceros Auklet, the updates in the vicinity of Año Nuevo are from 2002 and do not include the small breeding area within the Brandt's Cormorant colony. Sources: Thayer and Sydeman 2002a,b.
8. The estimates of breeding birds for Leach's Storm-petrels are in italics and from Ainley and Lewis (1974); these older estimates are likely much higher than the current colonies' status. The number of breeding birds at Fish Rock has likely signficantly decreased; in August 1989, no Leach's
Storm-petrels were captured at Fish Rocks, but this may be due to low sample effort (Harry Carter, pers. comm). And based on annual mark-recapture efforts since 1992, the number of breeding birds at S. Farallon Island has also likely significantly decreased (Bill Sydeman, pers. comm.).
9. Updates from 2002 (in bold) are included for the following eight species at South Farallon Island: Cassin's Auklet, Common Murre, Tufted Puffin, Pigeon Guillemot, Double-crested Cormorant; Pelagic and Brandt's cormorants, and Western Gull. Source: Warzybok et al. 2002.
10. The 2002 update for 246 breeding Common Murres at Devil's Slide Rock was provided by Gerry McChesney, U.S. Fish and Wildlife Service.
11. Fork-tail Storm-petrels have been noted as present and probably breeding at South Farallon Island, but no estimate is available.
87
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Table 19. Three most important variables (of nine investigated) having independent effects in explaining the variance in density of 25 selected
the marine sanctuaries; these species include pink- both were greater during these periods than during the Neutral
marine bird species.
footed shearwater, Leach’s storm-petrel, and western, period. On the other hand, the brown pelican responded strongly
Number of Percent glaucous-winged and Heermann’s gulls; see Table 20 to the disparity of conditions between the very strong 1997-98
Birds Variance
below. Also increasing in the study area during warm- El Niño and the very strong 1999-00 La Niña (Figure 59). The
Recorded Explained by
water periods were less abundant species including black-vented shearwater, for which there were insufficient data
During Top Three Three Most Important Variables of Those Investigated,
black and least storm-petrels and black-vented for long-term trend analysis, responded much more dramatically,
Surveys Variables in Order of Importance
Species
shearwater. with large numbers invading central California waters during the
Pacific loon 3,802 10.6 Ocean season, distance to land (-), latitude (-)
1997-98 El Niño. Finally, there were no significant differences
Western grebe 7,080 15.5 Ocean season, distance to land (-), ocean depth (-)
Cold-water Periods (e.g., La Niña events). During
Black-footed albatross 3,149 22.2 ocean depth (+), distance to land (-), year (-)
coldwater periods, areas
Laysan albatross 96 6.9 Ocean season, ocean depth (+), distance to land (+)
of relatively high avifaunal
Northern fulmar 5,882 21.3 Ocean season, ENSO period, year (+) Table 20. Effects of ocean season and ENSO events on the abundance of 26 marine bird
biomass and density
Sooty shearwater 296,065 43.4 Ocean season, year (-), ENSO period species off central California between 1985 and 2002, as determined through multiple regres-
expanded to cover broader
Pink-footed shearwater 4,145 13.1 Ocean depth (-), distance to land (-), ESNO period sion analyses.
portions of the study area;
Leach’s storm-petrel 1,414 28.4 Ocean season, distance to 2000m isobath (+), ENSO period Ocean Season(s)
11 of the predominant
Ashy storm-petrel 4,201 17.3 ENSO period, season, year (+) of Highest ENSO Event of
species became more
Fork-tailed storm-petrel 393 9.2 ENSO period, season, ocean depth (+) Species Abundance* Highest Abundance*
abundant. Species whose
Brown pelican 2,333 15.2 Distance to land (-), latitude (-), Ocean season
Pacific Loon DC LA
abundance showed the
Brandt’s cormorant 9,482 28.7 Distance from colony (-), dist. to 200 m isobath, dist. to land (-)
Western Grebe DC LA
greatest increases were the
Pelagic cormorant 396 6.1 Distance to land (-), ocean depth (-), dist. to 200 m isobath (-)
Black-footed Albatross UP NE
western grebe, northern
Double-crested cormorant 300 9.7 Dist. from colony (-), dist. To 200m isobath (-), dist. to land (-)
Laysan Albatross DC LA
fulmar, ashy storm-petrel, red
Red & Red-necked phalaropes 49,195 9.6 ENSO period, distance to land (-), ocean depth (-)
Northern Fulmar DC LA
and red-necked phalaropes
Western gull 29,545 44.2 Distance from colony (-), distance to land (-), ENSO period
(grouped), and black-legged
Glaucous-winged gull 767 17.1 Ocean season, ocean depth (-), latitude (+) Sooty Shearwater UP NE
kittiwake (Table 20). Others,
Heermann’s gull 1,121 6.5 Distance to land (-), ENSO period, latitude (-) Pink-footed Shearwater OC EL
whose abundance were also
California gull 13,721 24 Ocean season, year (+), latitude (+) Leach’s Storm-Petrel DC/UP EL
Black-legged kittiwake 4,565 28.9 Ocean season, ENSO period, year (+) significantly greater during La Ashy Storm-Petrel OC LA
Common murre 131,675 52.3 Distance to colony (-), ocean depth (-), distance to land (-) Niña periods were Pacific loon, Fork-tailed Storm-Petrel DC LA/EL
Rhinoceros auklet 14,679 19.8 Distance to land (-), season, ocean depth (+) Laysan albatross, California
Brown Pelican OC LA/EL
Tufted puffin 235 10.1 Distance from colony (-), year (-), distance to 200 m iso (+) gull, rhinoceros auklet, and
Brandt’s Cormorant UP ns
Cassin’s auklet 63,465 25.8 Distance to land (-), year (-), ocean depth (-) marbled murrelet.
Pelagic Cormorant UP ns
Marbled murrelet 273 4.7 Distance to land (-), latitude (+), ENSO period
Double-crested Cormorant UP ns
Notes Neutral Periods (i.e., neither
Red & Red-necked Phalaropes OC LA
1. For “continuous” variables, a positive (+) included with a variable indicates that density increased with an increase in unusually warm nor cold
the magnitude of that variable; (-) denotes the opposite. Western Gull UP EL
water). Four species, black-
2. The nine independent variables used in the regression analysis were: distance to nearest land; ocean season; ocean depth; Glaucous-winged Gull DC EL
footed albatross, sooty
ENSO period; year; latitude; distance to colony; distance to 200m isobath; and distance to 2,000m isobath. shearwater, tufted puffin Heermann’s Gull ns EL
3. Species that breed in the study area are shown in bold. and Cassin’s auklet were California Gull DC LA
significantly more abundant Black-legged Kittiwake DC LA
by management. The individual species text (accompanying two periods were chosen because climate differences were during the neutral period than Common Murre UP ns
each map) details where selected species become more or less extreme, i.e. among the strongest ENSO events of the past 100 during the warm or cold periods. Rhinoceros Auklet DC LA
abundant in north/central California during periods of warmer years, and they occurred adjacent to one another. Therefore,
Tufted Puffin UP/DC NE
or colder than average ocean temperatures. Actually, the shift a comparison of population response was not confounded by Other Responses to Short-Term
Cassin’s Auklet UP NE
in temperature is a proxy for many other changes, all of which long-term trajectories in base population size. Tables 20 and 21, Climatic Change. In regression
Marbled Murrelet DC/UP LA
ultimately affect the food web. as well as Figure 58, provide some examples of these changes analyses, the densities of the brown
* Notes. Ocean seasons are: Davidson Current (DC), Upwelling (UP), and Oceanic (OC);
due to interannual climate events (see Table 14). pelican and fork-tailed storm-petrel
ENSO periods are El Niño (EL), La Niña (LA), and neutral (NE). For species having
To illustrate the short-term ocean climate effects using species were not significantly different during
significant differences in abundance during respective seasons/periods (Sidak tests,
maps, a comparison was done using selected species density Warm-water Periods (e.g., El Niño events). During warm-water warm-water and cold-water periods.
P < 0.01), the season/period in which they were most abundant is given. If
maps for a specific El Niño and La Niña period. For the El events (including El Niño), many marine bird species tended Basically, this was because such
Niño period, the most recent, and very intense El Niño event, to occur closer to shore than during other years (Ainley and periods varied greatly in intensity, densities did not differ between the two seasons/periods in which they were most
Oceanic Season 1997 through Upwelling Season 1998, was Boekelheide 1990, Oedekoven et al. 2001). During warm- thus reducing effects especially if abundant, then the two are listed (eg. DC/UP, where densities were slightly higher
used; for the La Niña event, the period covering the Oceanic water periods, five of the predominant species became more other factors (not studied) were more in DC season that the UP season). Species for which there was no significant effect
Season 1998 through Oceanic Season 1999 was used. These abundant in central California waters in general, and inside important. However, the densities of of season or period are denoted with “ns”.
88
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
MAJOR SECTION CONTRIBUTORS
Table 21. A summary of changes in marine bird occurrence patterns, as a response to breeding season during warm-water periods national marine sanctuaries off north/central California
warm and cold ocean anomalies, as determined by visual comparison of species’ maps David Ainley, H.T. Harvey and Associates
(Table 21). This spreading is often affected generally encompass the areas of high concentrations and
during the 1997-1998 El Niño event and the 1999-2000 La Niña event. R. Glenn Ford, R.G. Ford Consulting Co.
most strongly by Farallon breeding species, diversity for marine birds, except for the western edge of the
Janet Casey R.G. Ford Consulting Co.
which usually concentrate near the Gulf of the Gulf of Farallones area and the "sanctuary exclusion area" off
Effect on Distribution
Larry Spear, H.T. Harvey and Associates
Farallones (outer shelf), and which move more San Francisco and Pacifica.
Species No Change El Niño La Niña
Carol Keiper, Oikonos
X to coastal waters.
Western grebe
Lisa Ballance, Southwest Fisheries Science Center, NOAA
X Owing perhaps to a response to competition for food by the
Pacific loon
X Tracy Gill, NCCOS, NOAA
Population Trends. Owing mainly to longer- large numbers of marine birds nesting on the Farallon Islands
Black-footed albatross
X Wendy Willams, R.G. Ford Consulting Co.
term, decadal shifts in marine climate (Hare and Point Reyes Headlands, during the breeding (Upwelling)
Laysan albatross
More spread Ken Buja, NCCOS, NOAA
Northern fulmar and Mantua, 2000, Mantua and Hare 2002), a season high concentrations of several breeding species extend
To Monterey number of species exhibited gradual changes seaward of the western boundary of the Gulf of the Farallones
REVIEWERS
Bay
Sooty shearwater in population size within the study area, from National Marine Sanctuary, over the Farallon Escarpment and
Two reviews were done for the marine bird analyses. The first
X
Pink-footed shearwater 1985 to 2002. These patterns were revealed beyond. This is especially true of Ashy Storm-petrel, Western
was a workshop to review draft maps in October 2002, and the
X
Buller’s shearwater using regression analyses, especially in cases Gull, Common Murre, and Rhinoceros Auklet. For the gull and
To Gulf of second review focused on the overall draft bird report (which
where Year was an important explanatory murre these deeper waters are not their preferred foraging
Farallones
Black-vented shearwater contains the maps) and was conducted via email November
variable to species occurrence (Table 19). habitat, but they choose to forage there during breeding
X
Leach’s storm-petrel and December, 2002. Most of the review comments were ad-
because more suitable continental shelf habitat to the north
More spread
Ashy storm-petrel dressed or incorporated for this product, which is a summary
Ashy storm-petrel and California gull and south is too far out of range.
X
Fork-tailed storm-petrel of a full bird report to be released later this year.
exhibited a gradual increase in population
To Gulf of (from 1985-2002); Tufted puffin showed a To a lesser degree, a smaller high density area existed seaward
Farallones Workshop Participants/Reviewers:
Black storm-petrel gradual decrease. The Black-legged kittiwake of Año Nuevo Island, where there is a smaller, but important
More spread David Ainley, H.T. Harvey and Associates
Brown pelican increased gradually, too, but was especially colony. These three colonies (Farallon Islands, Point Reyes
More spread
Brandt’s cormorant Sarah Allen, Point Reyes National Seashore, National Park
abundant after 1998 (Figure 60). Headlands and Año Nuevo Island), and the waters between
More spread
Pelagic cormorant Service
them which comprise the Gulf of the Farallones, possess
X
Red phalarope Lisa Ballance, Southwest Fisheries Science Center, NOAA
For other species the pattern was more populations that interact regularly in the shallow waters that
X
Red-necked phalarope Janet Casey, R.G. Ford Consulting Co.
complex. Black-footed albatross, sooty lie between them; many individuals marked at one have been
More confined
Glaucous-winged gull Glenn Ford, R.G. Ford Consulting Co.
shearwater, fork-tailed storm-petrel and seen at the other two sites. Therefore, in terms of marine birds
X
Western gull Carol Keiper, Oikonos
Cassin’s auklet showed a gradual decrease the waters of the Gulf of the Farallones, as defined above,
More confined
California gull Nora Rojek, formerly of California Department of Fish and
from 1985 until about 1999, when they began constitute a natural management unit.
Heermann’s gull Game
to increase.
To Monterey Jan Roletto, Gulf of the Farallones National Marine Sanctuary
In addition, it was apparent from visual inspection of the maps,
Bay
Bonaparte’s gull Program, NOAA
The year 1999 is when the system likely that the "sanctuary exclusion area" (i.e., the ocean area off San
X
Sabine’s gull Ed Ueber, NOAA, Gulf of the Farallones National Marine
shifted from a ‘warm’ to a ‘cold’ ocean regime Francisco and Pacifica that is excluded from the Monterey Bay
Confined to Sanctuary Program
(Bogard, 2000; Schwing and Moore, 2000). National Marine Sanctuary) represents a very important area
slope More spread
Black-legged kittiwake And several other members of the NOAA project team and
The northern fulmar exhibited a variable but for marine birds, especially those that breed at localities within
X
Caspian tern sanctuary programs.
‘steady’ population size during the 1980s and the Gulf of the Farallones National Marine Sanctuary (e.g. Point
X
Elegant tern
early 1990s but then began to increase with Reyes, Farallon Islands) (David Ainley, pers. comm.). This area
X
Arctic tern Reviewers of the Draft Bird Report:
arrival of the ‘cold’ regime. The population of is influenced strongly by the San Francisco Bay tidal plume,
More spread
Common murre Sarah Allen, Point Reyes National Seashore, National Park
the common murre remained stable through which provides habitat for many forage fish. This "sanctuary
More spread
Pigeon guillemot Service
most of the study period following a dramatic exclusion area" is also one of the main foraging areas of the
More spread
Tufted puffin Scott Benson, Southwest Fisheries Science Center, NOAA
X decline in 1982 (Ainley and Divoky 2001, Devil’s Slide murre colony, which is in the process of being
Rhinoceros auklet Karin Forney, Southwest Fisheries Science Center, NOAA
More spread Manuwal and Carter 2001), but recently it has restored (David Ainley, pers. comm.).
Cassin’s auklet Doug Forsell, Chesapeake Bay Program, U.S. Fish and Wildlife
More offshore begun to increase (H. Carter, pers. comm.).
Marbled murrelet Service
X These responses to decadal regime shifts With regard to the offshore bounds of the sanctuaries, among
Xantus/Craveri murrelets Laird Henkle, H.T. Harvey and Associates
present challenges to the researchers and the species of marine birds mapped, only 11 had significant David Hyrenbach, Duke University Marine Lab
managers even greater than those offered by concentrations seaward of sanctuary boundaries (see Table
in the abundance of common murre and the three cormorant Hannah Nevins, Moss Landing Marine Laboratories
short-term climate shifts (e.g., ENSO events). It takes several 16). In terms of management, it is important to consider Ashy
species among the three climate-related periods. Jan Roletto, Gulf of the Farallones National Marine Sanctuary
years of monitoring to detect long-term shifts in population storm-petrel and its habitat, because it is listed as a "State Program, NOAA
Breeding species (e.g. common murre, Cassin’s auklet) whose size. Species of Concern". The Xantus’ murrelet, a recently listed Franklin Schwing, Southwest Fisheries Science Center,
species, also deserves consideration, although the proportion
populations are not increased by an influx of visitors from NOAA
colonies outside the study area during warm-water periods, Relevance of Marine Sanctuary Boundaries to Marine of this species’ population that visits the central California And several other members of the NOAA project team and
and the cormorant species, are more dispersed during the Birds. Based on the available data, the boundaries of the National Marine Sanctuaries is relatively low. sanctuary programs.
89
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Ainley, D.G., L.B. Spear and S.G. Allen. 1996b. Temporal and Briggs, K.T., W.B. Tyler, D.B. Lewis, and K.F. Dettman. 1983. Cogswell, H.L.1977. Water birds of California. California natural
PERSONAL COMMUNICATIONS
spatial variation in the diet of the Common Murre in California. Seabirds of Central and Northern California, 1980-1983: Status, History Guide 40, Univ. California Press, Berkeley and Los
Laurence Breaker, Moss Landing Marine Laboratory
Condor, Vol. 98, pp. 691-705. Abundance, and Distribution. Part of Investigator’s Final Report, Angeles, 399 pp.
Harry R. Carter, U.S. Geological Survey, Dixon CA
Marine Mammal and Seabird Study, Central and Northern
Laird Henkel, H.T. Harvey and Associates, San Jose CA
Ainley, D.G., W.J. Sydeman, S.A. Hatch and U.W. Wilson. California, Contract No. 14-12-0001-29090. Prepared by Center Dunning, J.B. 1993. Body Weights of 686 Species of North
Michelle Hester, Oikonos, Bolinas, CA
1994. Seabird population trends along the west coast of for Marine Sciences, University of California, Santa Cruz, for American Birds. Internatl. Wildl. Rehab. Council, Univ. Georgia,
Gerry McChesney, U.S. Fish and Wildlife Service
North America: causes and the extent of regional concordance. the Pacific OCS Region, Minerals Management Service. OCS Athens, GA. 59 pp.
Studies Avian Biol., Vol. 15, pp. 119-133. Study MMS 84-0043. 246 pp.
REFERENCES
Ford, R.G. and J. L. Casey. 2003. CDAS Density Maps for
Ainley, D.G. 1976. The occurrence of seabirds in the coastal
Ainley, D.G., W.J. Sydeman and J. Norton. 1995b. Upper- Briggs, K.T., K.F. Dettman, D.B. Lewis, and W.B. Tyler. 1984. Marine Birds off North/Central California, developed for NOAA’s
region of California. West. Birds, Vol 7, pp. 33-68.
trophic level predators indicate interannual negative and Phalarope feeding in relation to autumn upwelling off California. National Centers for Coastal Ocean Science. Portland, Oregon.
positive anomalies in the California Current food web. Mar. Conservation of Seabirds in Western North America (J.C. 49 pp.
Ainley, D.G. Unpublished data. Shipboard surveys for seabirds
Ecol. Progr. Ser., Vol. 118, pp. 69-79. Bartonek and D.N. Nettleship, Eds.). U.S. Fish Wildl. Serv.,
at sea off northern and central California, 1985-2001. 105 Head-
Wildl. Res. Rept., Vol. 11, pp. 51-62. Glantz, M.R., and J.D. Thompson. 1981. Resource Manage-
lands Court, Sausalito, CA, 94965.
Ainley, D.G., R.R. Veit, S.G. Allen, L.B. Spear and P. Pyle. ment and Environmental Uncertainty. Wiley, New York. 63
1995c. Variations in seabird numbers in the California Current, Briggs, K.T., W.B. Tyler, D.B. Lewis, D.R. Carlson. 1987a. Bird pp.
Ainley, D.G., S.G. Allen and L.B. Spear. 1995a. Offshore
1986-1994. Calif. Coop. Ocean. Fish. Investig., Repts. Vol. 36, communities at sea off California: 1975 to 1983. Studies Avian
occurrence patterns of Marbled Murrelets in central California.
pp. 72-77. Biol., Vol. 11, pp. 1-74. Hare, S.R., and N.J. Mantua. 2000. Empirical evidence for North
Marbled Murrelet Conservation Assessment. Ecology and
Pacific regime shifts in 1977 and 1989. Progr. Oceanogr., Vol.
Conservation of the Marbled Murrelet (C. J. Ralph et al.,
Allen, S.G. 1994. The distribution and abundance of marine Briggs, K. T., D. G. Ainley, D. R. Carlson, D. B. Lewis, W. 47, pp. 103-145.
eds.), Gen. Tech. Rept. PSW-GTR-152, U. S. Dept. Agriculture,
birds and mammals in the Gulf of the Farallones and adjacent B. Tyler, L. B. Spear, and L. A. Ferris. 1987b. Final Report:
Forest Service, Pacific Southwest Res. Station, Albany, CA.
waters, 1985-1992. PhD thesis Univ. California, Berkeley. 222 California Seabird Ecology Study. Prepared by the Institute of Kleinbaum D.G., L.L. Kupper, and K.E. Muller. 1988. Applied
Pp. 361-370.
pp. Marine Sciences, University of California, Santa Cruz, for the regression analysis and other multivariable methods. PWS-
Pacific OCS Region, Minerals Manage. Serv., Contract No. KENT Publishing Co., Boston, MA. 191 pp.
Ainley, D.G. and R.J. Boekelheide (eds.). 1990. Seabirds of
Ballance, L.T. 2001. Unpublished data. Ecology Program, 14-12-001-30183. vii and 153 pp
the Farallon Islands: Ecology, Structure and Dynamics of an
Southwest Fisheries Science Center, 8604 La Jolla Shores Manuwal, D. A., and H. R. Carter. 2001. Natural history of the
Upwelling System Community. Stanford Univ. Press, Palo
Drive, La Jolla, CA 92037. Lisa.Ballance@noaa.gov; http: California Department of Fish and Game, Office of Spill Common Murre (Uria aalge californica). Biology and conserva-
Alto, CA. 425 pp.
//swfsc.nmfs.noaa.gov/PRD/ecology/ecology.html Prevention and Response (CDF&G-OSPR) aerial surveys off tion of the Common Murre in California, Oregon, Washington,
Metadata available from Philbrick et al. 2003. California, 1994-2001. Sacramento, CA. and British Columbia. Vol.1: Natural history and population
Ainley, D.G. and D.F. DeSante. 1980. The Avifauna of the
trends (D.A. Manuwal, H.R. Carter, and T.S. Zimmerman, Eds.)
Farallon Islands, California. Studies Avian Biol., Vol. 4, pp.
Baltz, D.M., and G.V. Morejohn. 1977. Food habits and niche Carter, H. R., G.J.McChesney, D.L. Jaques, C.S. Strong, M.W. U.S. Geol. Surv., Info. Tech. Rep. USGS/BRD/ITR-2000-0012.
1-104.
overlap of seabirds wintering on Monterey Bay. Auk, Vol. 94, Parker, J.E. Takekawa, D.L. Jory, and D.L. -Whitworth. 1992. Washington, D.C. Pp. 1-33.
pp. 526-543. Breeding Populations of Seabirds in California, 1989-1991.
Ainley, D.G. and G.J. Divoky. 2001. Seabirds: Effects of
U.S.Fish Wildl. Serv. Unpubl. Report, Pacific OCS Region, Mantua, N. J. and S.R. Hare. 2002. The Pacific Decadal Oscil-
climate change. Encyclopedia of Ocean Sciences (J. Steele,
Bograd, S.J., P.M. Digiacomo, R.Durazo, T.L.Hayward, Minerals Manage. Serv., Los Angeles CA. lation. Progr Oceanogr Vol. 58, pp. 35-44.
S. Thorpe and K. Tarekian, eds.). Academic Press, London.
K.D.Hyrenbach, J.J.Lynn, A.W.Mantyla, F.B. Schwing,
pp. 2669-2677.
W.J.Sydeman, T.Baumgartner, JB.Lavaniegos, CS.Moore. Carter, H.R., G.J. McChesney, D.L. Jaques, D.L. Jory, McChesney, G.J., H.R. Carter, and M.W. Parker. 2000. Nest-
2000. The State of the California Current, 1999-2000: Forward C.S. Strong, M.W. Parker, J.E. Takekawa, D.L. Jory, and ing of Ashy Storm-Petrels and Cassin’s Auklets in Monterey
D.G. Ainley, D.N. Nettleship, H.C. Carter & A. Storey. 2002.
to a new regime? Calif. Coop. Ocean. Fish. Investig., Repts., D.L. Whitworth. 1992. Breeding Populations of Seabirds in County, California. Western Birds 31: 178-183.
Common Murre (Uria aalge). In "The Birds of North America"
Vol. 41, pp. 26-52. California, 1989-1991. Draft Final Report prepared by U.S.Fish
(A. Poole & F. Gill, Eds.). The American Ornithologists’
and Wildlife Service, Northern Prairie Wildlife Research Center MMS (Minerals Management Service). 2001. Marine Mammal
Union, Washington, D.C. 54 pp.
Bolin, R. L. and D. P. Abbott. 1963. Studies on the marine and San Francisco Bay National Wildlife Refuge Complex, and Seabird Computer Database Analysis System
climate and phytoplankton of the central coastal area of for the Pacific OCS Region, Minerals Management Service, Washington, Oregon and California, 1975-1997 (MMS-CDAS,
Ainley, D.G. and G. A. Sanger. 1979. Trophic relationships of
California, 1954-60. Calif. Coop. Ocean. Fish. Investig., Repts., under Inter-agency Agreement No. 14-12-001-30456 with the version 2.1). Prepared by Ecological Consulting, Inc. (now
seabirds in the northeastern Pacific Ocean and Bering Sea.
Vol. 9: pp. 23-45. USFWS. 155 pp. R.G. Ford and Consulting Co.), Portland, Oregon for the
, Conservation of Seabirds in Western North America, (J.C.
Minerals Management Service, Pacific OCS Region, Order
Bartonek and D. N. Nettleship Eds.), U.S. Fish Wildl. Serv.,
Briggs, K.T., and E.W. Chu. 1987. Trophic relationships and Chu, E.W. 1984. Sooty Shearwaters off California: diet No.1435-01-97-PO-4206.
Wildl. Res. Rept. 11, pp. 95-122.
food requirements of California seabirds: updating models and energy gain. Marine Birds: Their Feeding Ecology and
of trophic impact. Seabirds: Feeding Ecology and Role in Commercial Fisheries Relationships (D.N. Nettleship, G. A. Oedekoven, C.S. D.G. Ainley and L.B. Spear. 2001. Variable
Ainley, D.G., L.B. Spear and S.G. Allen. 1996a. Seasonal and
Marine Ecosystems (J.P. Croxall, Ed.). Cambridge Univ. Press, Sanger and P.F. Springer, Eds.). Canad. Wildl. Serv., Ottawa. responses of seabirds to change in marine climate : California
spatial variation in the diet of Cassin’s Auklet reveals occurrence
Cambridge, UK. Pp. 279-304. pp. 64-71. Current, 1985-1994. Mar. Ecol. Progr. Ser. Vol. 212, pp. 265-
patterns of coastal euphausiids off California. Mar. Ecol. Progr.
281.
Ser., Vol. 137, pp. 1-10.
90
Section 2.2: BIOGEOGRAPHY OF MARINE BIRDS
Philbrick, V.A., P.C. Fiedler, L.T. Ballance, and D.A. Demer. Thayer, J.A., and W.J. Sydeman. 2002. Breeding biology and
2003. Report of ecosystem studies conducted during the 2001 ecology of Rhinoceros Auklets, Cerorhinca monocerata, on
Oregon, California, and Washington (ORCAWALE) marine Año Nuevo Island, California, 2002. Unpublished PRBO report,
mammal survey on the research vessels David Starr Jordan Stinson Beach, CA.
and McArthur. U.S. Department of Commerce. NOAA Technical
Memorandum NOAA-TM-NMFS-SWFSC-349. Available from Thayer, J.A. and W.J. Sydeman. 2002. Population and
http://swfsc.nmfs.noaa.gov/PRD/ecology/ecology.html productivity studies of Año Nuevo seabirds. Unpublished PRBO
report, Stinson Beach, CA.
Pyle, P. and R.P. Henderson. 1991. The birds of Southeast
Farallon Island: Occurrrence and seasonal distribution of mi- Trenberth, K. E. 1997. The definition of El Niño. Bull. Amer.
gratory species. West. Birds Vol. 22, pp. 41-84. Meteorol. Soc. 78: 2771-2777.
Veit, R.R., J.A. McGowan, D.G. Ainley T.R. Wahl and P. Pyle.
Schwing, F. B., Murphree, T., deWitt, L., Green, P. M., 2002. 1997. Apex marine predator declines 90% in association with
The evolution of oceanic and atmospheric anomalies in the changing oceanic climate. Global Change Biol., Vol. 3, pp.
northeast Pacific during the El NiÑo and La NiÑa events of 23-28.
1995-2001. Progress in Oceanography 54, 459-491.
Warzybok, P.M., R.W. Bradley, and W.J. Sydeman. 2002.
Schwing, F.B. and C.S. Moore. 2000. A year without summer Population size and reproductive success of seabirds on
for California, or a harbinger of a climate shift? Trans., Amer. Southeast Farallon Island, 2002. Unpublished Report. PRBO
Geophys. U., Vol. 81(27), p. 301. Conservation Science, Stinson Beach, CA.
Schwing, F.B.,T.L. Hayward, K.M. Sakuma, T. Murphree, A.S. Whitworth D.L., H.R. Carter, R.J. Young, G.J. McChesney,
Mascarenas Jr., S.I. Larios Castillo, A.W. Mantyla, S.L. Cum- M. Hester and S. Allen. 2002. Status and distribution of the
mings, F.P. Chavez, D.G. Ainley and K. Baltz. 1997. The state Ashy Storm-petrel (Oceanodroma homochroa) at Point Reyes
of the California Current, 1996-1997: mixed signals from the National Seashore, California, in 2001. Unpubl. Rpt, Humboldt
tropics. Calif. Coop. Ocean. Fish. Investig., Repts., Vol. 38, State University, Dept Wildlife, Arcata. 15 pp
pp. 22-48.
Seber, G. A. F. 1977. Linear regression analysis. John Wiley
and Sons, New York. 171 pp.
Shannon, C.E. and W.W. Weaver. 1949. The Mathematical
Theory of Communication. Univ. Illinois Press, Urbana. 90
pp.
Spear, L.B. and D.G. Ainley. 1999. Migration routes of Sooty
Shearwaters in the Pacific Ocean. Condor, Vol. 101, pp. 205-
218.
Sydeman, W.J., N. Nur, E.B. McLaren, and G.J. McChesney.
1998. Status and trends of the Ashy Storm-petrel on Southeast
Farallon Island, California, based on capture-recapture
analyses. Condor, Vol. 100, pp. 438-447.
Sydeman, W.J., K.A. Hobson, P. Pyle, and E.B. McLaren. 1997.
Trophic relationships among seabirds in central California:
combined stable isotope and conventional dietary approach.
Condor, Vol. 99, pp. 327-336.
91
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
INTRODUCTION the difficulty in obtaining adequate distributional data sets and in a useable format. Some species of marine mammals are NOAA’s Southwest Fisheries Science Center (SWFSC). These
The California Current passes south through the study area the complexity of combining available marine mammal data infrequently sighted and their distributions are therefore difficult maps are included to illustrate one of the data sets that will
off north/central California, which, along with areas of strong sets, maps for only 13 species were completed in Phase 1, to map. In addition, some data sets were unavailable at the likely be incorporated into the mammal map analysis in Phase
coastal upwelling, makes this area one of the most productive and are included in this document. These maps are referred time of the analysis. Over the past few months, additional data II. SWFSC maps for an additional 13 species are included
ocean systems in the world (Glantz and Thompson, 1981). to as CDAS maps, and represent a compilation of data sets sets have been made available, and these will be added in on the CD-ROM. Plans for Phase II include the following: 1)
Because of this productive environment, the study area from 1980 through 2001, as described below. Also included in Phase II. Improvements and updates are planned for the at-sea acquire the additional available data sets; 2) correct the data
contains a rich fauna of marine mammals, as evidenced in this document and on the CDROM are sighting and effort maps maps, as well as for the haulout and rookery maps, if funding sets for species’ sightability, detectability and other factors;
marine mammal abundance and richness. for 16 marine mammals from a single data set (the marine is made available. and 3) develop composite maps for about 23 species, singly
mammal stock assessment surveys from NOAA's Southwest or in combination.
In addition to many marine mammal species that live here year- Fisheries Science Center (SWFSC)). These data will likely be Table 22 is a list of marine mammal species that were
round and use the region’s coasts and islands for breeding and incorporated into Phase II of the analysis. considered for this analysis; density maps were developed About the Literature and Survey Data Used in this
hauling out, the community of seasonal residents and migrants for eight species in MMS-CDAS (MMS, 2001) a data display Assessment. This assessment is based on the efforts of
is even more robust. Central California is the destination for Additional data compilation, data analysis and expert review system developed by R.G. Ford Consulting Co. for the Minerals individual researchers to study marine mammal spatial and
many marine mammal species seeking productive feeding are needed to complete the final analyses of marine mammals Management Service. Sightings maps were developed in CDAS temporal patterns, federal and state government efforts to
areas and acceptable habitat in which to spend their occurrences in the study area. More complete results will be for an additional four species. Maps of sea otter counts and assess stock size and the potential biological impacts of oil
nonbreeding periods, providing evidence of the region’s presented in a final report in Phase II of this project. Additional pinniped haulouts and rookeries were also developed. Also development, and state government efforts to respond to oil
trophic richness. Over 29 species of marine mammals occur products planned for Phase II are listed at the end of this included in this document are maps for three species from spills.
in the study area off north/central California; over 22 cetaceans section.
(whales, dolphins, and porpoises), six pinnipeds (seals and
Table 22. Marine mammal species included in this assessment and map types developed for them (Phase I and Phase II)
sea lions), and one fissiped species (the sea otter). DATA AND ANALYSES
Overview of Map Development and No. of Draft No. of No. of Maps
CDAS Maps, SWFSC Planned for
The objectives of this assessment were to: 1) identify spatial Analysis Process To Date. The methods
Common Name Scientific Name Order/Suborder Family Phase l Maps, Phase l Phase II
and temporal distributions and patterns for marine mammals used in each survey were different, and
Fissiped
that occur in ocean waters off north/central California between because of this, careful consideration and
Southern sea otter Enhydra lutris nereis Carnivora/(none) Mustelidae 1 1
Point Arena (38.91ºN) and Point Sal (34.90ºN); 2) identify correction are required to merge the data
Pinnipeds
important areas and time periods associated with higher sets in a meaningful and scientifically
California sea lion Zalophus californianus Carnivora/PinnipediaOtariidae 2 1
concentrations of these species; and 3) identify important data acceptable way. Data preparation for the
Steller sea lion Eumetopias jubatus Carnivora/PinnipediaOtariidae 1 1
and information gaps observed as a result of this analysis. mammal analyses included the following
Northern fur seal Callorhinus ursinus Carnivora/PinnipediaOtariidae 1 1
In this analysis, ‘important’ season or area refers to those steps: species and study area selection,
Harbor seal Phoca vitulina richardsi Carnivora/PinnipediaPhocidae 1 1
having relatively higher concentrations of a particular species; data set identification and collection, data
Northern elephant seal Mirounga angustirostris Carnivora/PinnipediaPhocidae 1 1
in Phase II diversity may also be considered. corrections, data conversion into common
Cetaceans
units, organizing the data into 10’x10’ cells
Dall's porpoise Phocoenoides dalli Cetacea/Odontoceti Phocoenidae 1 1 1
Preliminary Results. Summarized below are the spatial or leaving them as sightings and effort, and
Harbor porpoise (stocks: Northern CA, San
and temporal occurrence patterns of data for 13 species that map development. For species present in
Francisco/Russian River, Monterey Bay) Phocoena phocoena Cetacea/Odontoceti Phocoenidae 1 1?
regularly occur in marine waters off north/central California. The sufficient numbers, seasonal density maps
Lagenorhynchus obliquidens
Pacific white-sided dolphin Cetacea/Odontoceti Delphinidae 1 1 1
results of this marine mammal assessment are preliminary were developed, and for infrequently sighted
Risso's dolphin Grampus griseus Cetacea/Odontoceti Delphinidae 1 1 1
(Phase I) and feature highlights of work in progress. Due to species, sighting and effort maps were
Bottlenose dolphin (California coastal stock) Tursiops truncatus Cetacea/Odontoceti Delphinidae 1 1?
developed. CDAS maps were created for
Short-beaked common dolphin Delphinus delphis Cetacea/Odontoceti Delphinidae 1 1
13 species. The original draft maps were
Northern right whale dolphin Lissodelphis borealis Cetacea/Odontoceti Delphinidae 1 1 1
reviewed at an expert workshop in October
Killer whale Orcinus orca Cetacea/Odontoceti Delphinidae 1 1?
2002; there it was determined that additional
Baird's beaked whale Berardius bairdii Cetacea/Odontoceti Ziphiidae 1 1?
data, corrections and analyses were required
Cuvier's beaked whale Ziphius cavirostris Cetacea/Odontoceti Ziphiidae 1 1?
to improve the mammal maps; this work will
Beaked whales (Mesoplodonts) Mesoplodon spp. Cetacea/Odontoceti Ziphiidae 1 1?
be done in Phase II of this project. Some
Sperm whale Physeter macrocephalus Cetacea/Odontoceti Physeteridae 1 1?
revisions have been made to the maps and
Blue whale Balaenoptera musculus Cetacea/Mysticeti Balaenopteridae 1 1 1
text of this document since its draft release
in April 2003. Humpback whale Megaptera novaeangliae Cetacea/Mysticeti Balaenopteridae 1 1 1
Fin whale Balaenoptera physalus Cetacea/Mysticeti Balaenopteridae 1 1?
Species Selected for Analysis. Selection Minke whale Balaenoptera acutorostrata Cetacea/Mysticeti Balaenopteridae 1 1?
criteria for marine mammal species included Gray whale Eschrichtius robustus Cetacea/Mysticeti Eschrichtiidae 1 1
in this assessment were: 1) the species MAP TOTALS 14 16 14-23
distribution includes the study area, and 2)
survey data for the species was available
92
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Table 23. A Summary of At-Sea Data Sets Used in the Preliminary Marine Mammal Assessment
Because wind speed affects detection of marine mammals, data
Table 23. Summary of at-sea data sets used in the preliminary marine mammal analyses.
collected when wind speed exceeded 25 kt were excluded. Data
Vessel Name Ocean Total Transect Total Transect
Habitat were allocated into 10’ x 10’ cells (i.e., 10-minute latitude by
Principal & Platform Seasons Width: Width:
10-minute longitude cells). All aerial data were continuous; each
Covered2
Data Set Investigator Height Years Covered Pinnipeds Cetaceans
ship-based data set was converted separately into a continuous
Surface survey harbor porpoise,
transect format to the extent possible. The continuous aerial
MMS High- of the shelf, 254m; great
data were binned into the appropriate cell. For the SF-DODS
Altitude Aerial Pembroke, slope & deep All three whales, 1130m;
Surveys Dohl 270m ocean beyond 1980-1983 seasons N/A all others, 885m and EPOCS studies, and the Rockfish Assessment cruises prior
Surface survey to 1997, the beginning position, ship heading, and speed were
MMS Low- of the shelf, used to compute the end position of each 2-4 km continuous
Altitude Aerial slope & deep All three
transect. From this, a midpoint of the transect was determined.
Surveys Bonnell Pembroke, 62m ocean beyond 1980-1983 seasons 109m 109m
As times of observations were not available, the position of
Surveyor, 12m, the midpoint was used to select the cell to which the survey
EPOCS Discoverer, Surface survey
effort was assigned. If this midpoint fell on a cell boundary, it
Shipboard Oceano- of the deep All three
was assigned to the cell to the north or west. To maintain the
Surveys Ainley grapher, 15m ocean 1984-1994 seasons 300-600m 800m
correspondence between effort and mammal observations,
CA Seabird
Davidson Current, because ocean conditions differ distinctly
observations were also assigned to the transect midpoints.
Ecology Low- Surface survey
among them and are known to affect the biota of the California
For the Rockfish Assessment Cruises from 1997 onward, effort
Altitude Aerial Partenavia, of shelf and Mainly
Surveys Briggs 62m slope 1985 Upwelling 50m 50m Current (e.g. Ainley 1976, Briggs et al. 1987). As there is
was assigned to the cells through which the vessel passed
NMFS significant interannual variation in the actual duration of these
based on the proportion of trackline that fell within each cell, and
Midwater seasons, the following dates were ‘defined’ for each season for
observations were interpolated along the cruise track according
Trawls for Juv. Surface survey
purposes of analysis: Upwelling Season is 15 March-14 August,
to the time of each observation.
Rockfish: David Starr of shelf and Mainly
Oceanic Season is 15 August-14 November, and Davidson
Ship Surveys Ainley Jordan,10m slope to 3000 m 1985-2001 Upwelling 300m 800m
Current Season is 15 November-14 March.
Data Analysis.
OSPR Low Surface survey
Effort Summary. For all surveys, 132,521 kilometers of trackline
Altitude Aerial Partenavia, of shelf and 1994-1998, All three
As evident in Table 24, the Upwelling Season had the greatest
(pinnipeds and cetaceans) and 78,486 kilometers of additional
Surveys Bonnell 62m slope 2001 seasons 50m 50m
MMS Santa amount of survey effort, followed by the Davidson Current
trackline (cetaceans only) were analyzed (Table 24). A total of
Barbara Season. The Oceanic Season had the lowest effort. Unlike
3,459 observations of 7,039 pinnipeds and 2,313 observations
Channel Low Surface survey the other seasons, the Oceanic Season had no data from
of 69,286 cetaceans were included in analyzed data. Survey
Altitude Aerial Partenavia, of shelf and All three
the 1980s. Because of the variation in effort coverage across
effort used in this assessment for pinnipeds and cetaceans are
Surveys Bonnell 62m slope 1995-1997 seasons 50m 50m
space and time (and methods, as well as many other factors),
summarized as maps in Figure 61.
SF-DODS Surface survey interpretation of the data requires careful consideration.
Shipboard of shelf and All three
Organizing Data into Ocean Seasons. Effort and species
Surveys Ainley Point Sur, 8m slope to 3000 m 1996-2000 seasons 300m 800m
data were organized and mapped into three distinct ocean
Notes
seasons (Bolin and Abott 1963): Upwelling, Oceanic, and
See additional description of these data sets on the CD for more information on the CDAS data sets.
Data from the marine mammal stock assessment of NOAA's SWFSC were not included in the preliminary CDAS asssessment.
Table 24. Summary of combined data set effort at sea for mammals, by ocean season.
Number
The Data Sets. The ship and aerial strip transect data used in
Related Literature. The at-sea distribution and abundance of
Dates Used for Number Kilometers of 10'
this assessment were collected from 1980-2001 and ranged
marine mammals within the study area has been described
Ocean Each Ocean of Years of Trackline Number Cells
from Point Arena south to Point Sal, and offshore to the extent
in many publications, some of which include the following:
Season Season Months Included Taxa Surveyed of Visits Sampled
of data availability. Estuaries were not part of the GIS study
Bonnell et al. (1983), Dohl et al. (1983), Calambokidis et
Pinnipeds: 63,262 10,902 283
1980-1982,
area, but coastal haulouts and rookeries, when available, were
al (1988, 1990, 1996), and Allen (1994). Numerous marine
Upwelling 15 Mar-14 Aug 5 1985-2001 Cetaceans:
mapped to provide a more complete view of important areas 96,978 15,280 317
mammal stock assessment studies have been conducted by
for pinniped species. See Table 23 for additional information
the Southwest Fisheries Science Center, LaJolla, CA (NMFS/ Pinnipeds: 30,443 4270 263
1980-1982,
on the data.
SWFSC ship surveys): Barlow (1988, 1995), Barlow and Forney 1991, 1994-
(1994), Barlow and Gerrodette (1996), and Forney and Barlow
Oceanic 15 Aug-14 Nov 3 2001 Cetaceans: 49,981 6821 322
Data Synthesis.
(1998). A few ecosystem studies of marine mammals in this
Pinnipeds:
Davidson 1980-1986, 38,816 5594 360
Summarizing Transect Data into Grid Cells for CDAS Maps. The
region have also been conducted by Schoenherr (1991), Black
Current 15 Nov-14 Mar 4 1991-2001 Cetaceans: 64,048 8897 383
above data sets were processed to compensate and correct
(1994), Kieckhefer (1992), Croll et al. (1998), Forney and
Pinnipeds: 132,521 20,766 395
for differences in survey methodology, including platform type
Barlow 1998, Forney 2000, Benson et al. (2002) and Keiper
TOTAL 1 Jan-31 Dec 12 1980-2001 Cetaceans:
(ship or aerial) and transect width, among the various studies.
et al. (In Review). 211,007 30,998 416
93
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Calculating Density and Developing Seasonal Density CDAS will be added and an overall rookery and haulout map for the
Combined At-Sea Effort for Marine Mammal Analysis Maps. From the digitized survey data, we mapped the distribution pinnipeds will be developed.
of effort and of species observations into a grid of 10’ by 10’
cells, using the MMS-CDAS mapping system (MMS, 2001). ANALYTICAL MAP PRODUCTS
129°W 128°W 127°W 126°W 125°W 124°W 123°W 122°W 121°W
The larger cell size was determined to be more meaningful by A series of over 50 preliminary maps (41 CDAS maps and 12
200 m
Pinniped Effort
20
0
experts at a preliminary map/data review session. SWFSC maps) and related results are presented in this section;
0
m
39°N
39°N
additional SWFSC maps for 13 mammal species/species
Effort
The species data were first transformed into densities on the groups are included on the CD-ROM. These preliminary maps
(Km of trackline)
basis of strip widths (which varied by ship or aerial platform, will be finalized and included in a Phase II report.
> 1500
depending on speed and height above water; see Table 23).
1000.01 - 1500.00
500.01 - 1000.00
The number of individuals of each species seen was then
250.01 - 500.00
38°N
38°N
divided by area surveyed to estimate density in each cell for
100.01 - 250.00
50.01 - 100.00
that data set. For construction of density plots, if a cell was
25.01 - 50.00
5.01 - 25.00
censused in other years or the same year by another survey,
0.01 - 5.00
densities in cells were averaged and weighted according to
effort. These maps display observed densities; in Phase II these
0 25 50 Km
37°N
37°N
densities will be corrected to account for additional factors such
as sightability.
Seasonal High Use Areas for Individual Species. The purpose
36°N
36°N
of the seasonal high use maps is to provide a summary map
for a species' spatial patterns. These maps were developed for
mammal species with density data, with the seasonal density
data binned into 10’ x 10’ cells for each species or species
group. Non-zero cells were then ranked and those in the top
a
35°N
35°N
20 percent were selected and defined as seasonal high use
areas. Cells were then mapped with color corresponding to the
number of ocean seasons of high use. The index is therefore
200 m
20
Cetacean Effort
0
sensitive to cells that were not sampled in any one of the three
0m
39°N
39°N
seasons, causing a downward bias in the index. Use of a 10'x10'
Effort cell size greatly reduces the magnitude of this bias.
(Km of trackline)
> 1500
Cells in which there was effort but animals were not observed,
1000.01 - 1500.00
and cells where sightings occurred but were never high use
500.01 - 1000.00
38°N
38°N
250.01 - 500.00
areas, were also provided.
100.01 - 250.00
50.01 - 100.00
25.01 - 50.00
Developing Sighting and Effort CDAS Maps for Infrequently
5.01 - 25.00
0.01 - 5.00
Sighted Species. Where sightings were too few to warrant
seasonal density maps, observations were mapped as point
0 25 50 Km
37°N
37°N
locations. For context, overall survey effort is also presented.
This display method was chosen in response to comments by
expert reviewers at the October 2002 workshop and in view of
the low numbers of sightings of certain species.
36°N
36°N
Preliminary Rookeries and Haulouts by Species. Pinniped
rookeries and haulouts are monitored and surveyed by a variety
of institutions and individuals. Recent data (varying by species,
but generally from 1998-2002) were used to represent locations
b
35°N
35°N
of rookeries and haulout sites for five pinniped species. In
Phase II, additional information on harbor seal pupping sites
129°W 128°W 127°W 126°W 125°W 124°W 123°W 122°W 121°W
Source Data: See text.
Figure 61. Total at-sea survey effort for marine mammal analyses.
94
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THESE MAPS from Half Moon Bay to Goleta, just south of Point Conception
DRAFT Maps 62a and b display the locations of groups of southern sea with uncommon sightings of animals beyond these areas (pers.
Southern sea otter Enhydra lutris nereis otters during the Fall 2001 and Spring 2002 rangewide counts. comm. B. Hatfield); the distribution of otters along the south
123°W 122°W 121°W
123°W 122°W 121°W
Maps 62c and d summarize these rangewide count data into end has been highly variable since the expansion of the sea
coastal strips approximately 10km in length, in order to display
Rangewide Count Rangewide Count otter range south of Point Conception (pers.comm. M.Harris).
20 20
0 0
m m
20 20
linear densities along the shore. The northern extent of the data In the study area, sightings have occurred as far north as Point
0
Fall 2001 Spring 2002
0
0 0
m m
is south of Half Moon Bay; sea otters are also present to the Reyes (Point Reyes Headlands, Double Point, Duxbury Reef;
south of the mapped area. not shown on map; pers.comm. S. Allen). Sea otters occur
Number of
along rocky shorelines with kelp beds (but also in open water
Sea Otters
DATA SOURCES habitats, sandy/soft bottom areas, and tidal estuaries) and in
16 - 28
Data were collected by wildlife biologists from the U.S. Depart-
37°N
37°N
depths of water about 20-40 m (some to 60 m, and rarely to
11 - 15
ment of the Interior (currently the USGS Biological Resources 100 m; M. Kenner pers. comm).
6 - 10
Division), California Department of Fish and Game, the Mon-
2-5
terey Bay Aquarium, and trained volunteers during semi-an- Overall, numbers of otters per segment were greater in the
1
nual rangewide counts in Fall 2001 and Spring 2002. The Fall southern portion of the Monterey Bay National Marine Sanc-
0 25 50 Km
2001 count was conducted during the period 4-20 November tuary. In the census of Fall 2001 (map a), greater numbers of
2001, the Spring 2002 count was conducted during the period otters per segment occurred along the Carmel coast and from
5-22 May 2002. The data set was provided by Mike Kenner, Piedras Blancas south to Point San Luis. Seasonal changes
36°N
36°N
UCSC but is sourced to USGS; contact Brian Hatfield for more in abundance and distribution of sea otters are believed to
information. be affected by male movements during the period when most
breeding occurs (June/July through October/November) when
METHODS they move from the periphery of the range toward the center of
The original data were entered from hand marked maps into a the range in search of estrous females (Bonnell et al., 1983).
custom designed digitizing program which assigned coordinates From December to April, many males migrate to the range pe-
to each observed sea otter group. Positions of animals toward
35°N
35°N
ripheries, perhaps in search of more abundant prey (M.Harris
the ends of the range and in Elkhorn Slough were not assigned
a b pers. comm.). However, this is not evident in the maps. Sea-
coordinates by this program. Each group was also assigned to sonal changes also are affected by factors such as weather,
38°N
an ATOS (As The Otter Swims) number, which are numbers sea conditions and abundance of kelp canopy (see Reidman
Linear Density Linear Density
20 20
approximately 0.5 km apart along a smoothed 5 fathom contour
0 0
m m
20
and Estes, 1990).
20
0
Rangewide Count Rangewide Count
0
0 0
line along the coast from Golden Gate to approximately Santa
m m
Fall 2001 Spring 2002 Barbara. These numbers were used to get approximate posi- From 1983 until the mid 1990’s, trends in spring southern sea
tions for otters without assigned coordinates. otter counts indicated sea otters increased steadily; in the mid-
Otters per to late 1990’s, sea otter numbers declined (USFWS, 2000)
A series of coastal segments approximately 2 km in width was
Segment and have since remained relatively constant (pers.comm. B.
37°N
37°N
created for display purposes. Each segment was approximately
151 - 200 Hatfield). Sea otter count data is used as an index to assess
10 km in length; divisions were based on the ATOS numbers
101 - 150 trends in the population dynamics, not as a population estimate
described above. Twenty ATOS numbers approximately 500m
51 - 100
(pers.comm. M.Harris). The 2002 spring count was 1% below
apart were included in each segment. The coordinates of each
26 - 50
the 2001 count, from 2161 otters in 2001 to 2139 in 2002. The
otter group were used to place it within a particular segment,
11 - 25
2001 count was 6.7% below counts from the previous year
1 - 10 and the otters in each segment were summed. This provides
(USGS 2002). Due to its small population size, the southern sea
0 an estimate of linear density (otters per segment or otters
otter population is especially vulnerable to human disturbance,
36°N
36°N
per 10km) since the segments were approximately 10km in
competition with fisheries, and pollution, including the threat
length.
of a major oil spill. The lack of population growth and recent
decline coincides with an increase in mortality (e.g., infectious
RESULTS AND DISCUSSION
diseases, white shark attacks) as indicated by the number of
The southern sea otter (Enhydra lutris nereis) is one of three
beach-cast sea otter carcasses (Estes et al., 2003). Otters near
subspecies: southern (E.l.nereis), northern (E.l.kenyoni), and
heavy freshwater flows are three times more likely to have been
Russian (E.l.lutris). The southern sea otter is listed as threat-
infected by Toxoplasma gondii, a protozoan parasite caused by
35°N
35°N
ened under the Federal Endangered Species Act, and depleted
c d parasite eggs in cat droppings (see Miller et al., 2002).
under the Marine Mammal Protection Act (MMPA). Under Cali-
fornia Fish and Game Code, the southern sea otter is listed as
123°W 122°W 121°W 123°W 122°W 121°W
Southern sea otters are key predators of benthic species (e.g.
a “fully protected” species. The southern sea otter generally
Source Data: See text.
sea urchins, sea stars, mussels, clams, abalone, crabs) and
inhabits the near-shore waters of the central California coast,
Figure 62. Maps for southern sea otter: rangewide count and linear density, fall 2001 and spring 2002. octopus (see Riedman and Estes, 1990).
95
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
124°W 123°W 122°W 121°W
ABOUT THIS MAP California sea lions feed on a diversity of fish (e.g., Pacific hake,
200 m
Figure 63 summarizes information on California sea lion
20
northern anchovy, Pacific sardine, herring, rockfish, salmon,
California sea lion Zalophus californianus
39°N
39°N
0
0
haulouts and rookeries in the study area based on number of steelhead) and invertebrates (e.g., squid and octopus) (Weise,
m
animals, frequency of use, and rookery status. 2000; see also Riedman 1990).
Fish Rocks
Haulout Sites DATA
Haulout data and rookery information are from aerial counts
Haulout Occupancy
(July, 1998-2002) provided by Mark Lowry of the NMFS’
1998 - 2001
Southwest Fisheries Science Center.
Frequent (4 years)
Bodega Rock
Occasional (3 years)
METHODS
Infrequent (1-2 years)
Haulout locations were mapped using coordinates included
Point Reyes Minor Rookery,
38°N
38°N
in the files from Mark Lowry, SWFSC. Data from July counts
1998-2001
Sea Lion Cove
Occasional Minor in four years, 1998-2002, were used to calculate frequency of
Rookery
Pier 39 use for each haulout location and mean number of animals
North Farallon Islands (e.g. El Niño years)
using each location when that location was occupied. Rookery
Number of Sea Lions
Southeast Farallon Island
status was determined by the inclusion of pups in the counts.
Mean of July Counts*: Pups were observed in all years at two sites, while three sites
1998-2001
had pups only in 1998, an El Niño year.
2501 - 6706
RESULTS AND DISCUSSION
501 - 2500
Año Nuevo Island
Haulout sites for the California sea lion are located along the
37°N
37°N
101 - 500
coast from Fish Rocks (just south of Point Arena) to the south,
at Point Sal Rock, and inside San Francisco Bay (Pier 39).
51 - 100
Minor rookeries are located on the Southeast Farallon Island
26 - 50
Monterey Breakwater 1 - 25
and Año Nuevo Island.
0 25 50 Km
Sea Lion Rocks
Periods of unusually warm ocean waters associated with El
Lobos Rocks
Nino oceanographic conditions affect pup production (i.e., fewer
pups are born) and result in higher mortality rates for pups and
Unnamed; N of Partington Pt juveniles. During the El Niño periods of 1983, 1992, and 1998,
pup production decreased by 35, 27, and 64%, respectively at
36°N
36°N
rookeries in southern California (SWFSC 2001).
San Martin Rocks
Similar to at-sea occurrence patterns, haulout patterns and
rookery locations also change during warmer water periods. For
Point Piedras Blancas
example, rookery locations during the strong El Niño of 1998
White Rock
(shown on the map) included the rookeries at the Farallones
and Año Nuevo as well as additional rookeries located near
Partington Point, and Lion, Pecho, and Pup Rocks located to
Lion Rock
Pecho Rock the south of the Monterey Bay National Marine Sanctuary. Haul-
out patterns at the Farallon Islands and Point Reyes National
35°N
35°N
Seashore also changed, indicated by an influx of immatures
(Sydeman and Allen, 1999; S. Allen pers.comm.).
Point Sal Rock
DRAFT Greater numbers of California sea lions in the study area during
El Niño events likely reflected a greater than usual northward
migration in response to a reduction of food resources near
* When occupied.
southern breeding grounds.
124°W 123°W 122°W 121°W
Source Data: See text.
Figure 63. Map for California sea lion: haulouts and rookeries.
96
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THESE MAPS For the Oceanic Season, data are from 1980-1982, 1991, and
DRAFT
California sea lion Figures 64a, b, and c show the at-sea density (animals/km2) of 1994-2001. For the Davidson Current Season, data are from
Zalophus californianus California sea lions in the Upwelling, Oceanic, and Davidson 1980-1986 and 1991-2001. The rookery and haulout counts
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Current seasons, displayed in 10’x10’ cells. Densities are are shown as a general range, based on counts of all animals
Upwelling Season Oceanic Season
200 m
200 m
20
20
based on combined data of several studies; see “Methods” (pups and adults) in three years, 1999-2001.
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) and “Data Sources” sections. The color and mapping intervals
39°N
39°N
were customized to show the most structure and to highlight METHODS
Density significant areas, while allowing comparisons among marine At-sea densities are the result of a synthesis of data from
(Animals/km²)
mammal species. Cells that were surveyed but which had seven shipboard and aerial survey programs conducted in
10.01 - 50.00
no California sea lions have a density of zero; unsurveyed the study area in the years 1980-2001 (see “Data Sources”
5.01 - 10.00
38°N
38°N
areas appear white. Blue lines indicate the National Marine section). Pinniped observation data and trackline data from
1.01 - 5.00
Sanctuary boundaries of Cordell Bank, Gulf of the Farallones, these studies were converted to a common format. All aerial
0.51 - 1.00
and Monterey Bay; bathymetric contours for the 200 m and data were continuous; ship-based data were converted
0.11 - 0.50
2,000 m isobaths are also shown in blue. separately into a continuous transect to the extent possible.
0.06 - 0.10
0.01 - 0.05 From the digitized survey data, the distributions of effort and of
37°N
37°N
0.00 In order to provide one map for the species that integrates the species were mapped into 10’x10’ cells using CDAS, a custom
patterns of its spatial and temporal occurrence in the study geographic information system for analyzing marine bird and
0 25 50 Km
area, map d shows seasonal high use areas, displayed in mammal surveys (MMS, 2001). The length and width of the
10’x10’ cells. This map provides a further synthesis of densities survey trackline in a given cell (estimated trackline width varied
presented in maps a, b and c (see “Methods” section for details), by platform, depending on speed and height above water) were
36°N
36°N
and portrays the relative importance of various areas to the used to estimate the area sampled. The number of cetaceans
species. Areas with consistent high use are highlighted on this of each species seen in a cell was then divided by the area
map. To provide a relative reference for the “high use” areas, sampled in the cell to estimate density. If a cell was censused
cells are also shown where the species were absent (i.e., the more than once, densities were averaged, with adjustment
cell was sampled but the species was not recorded there), or made for effort.
35°N
35°N
a b present but at lesser concentrations in any particular season.
Note that these maps represent either sighting locations or
200 m
Davidson Current Season
200 m
Seasonal High Use Areas
20
20
DATA SOURCES densities that used survey strip widths relative to each survey
0
0
0m
0m
(Nov. 15 - Mar. 14) and Rookeries At-sea densities for the California sea lion are based on data platform (e.g., plane, ship); density was calculated on the basis
39°N
39°N
from seven survey programs conducted in 1980-2001. These of the number of animals sighted and area surveyed. The data
Persistence of data were combined using CDAS software into the MMS-CDAS have only been corrected to normalize for survey effort and to
High Use
data system (MMS, 2001), developed for Minerals Management exclude observations with winds greater than 25 knots (smaller
3 Seasons
Service and expanded for this project. Of the data sets on the or less obvious species are often less detectable even at wind
2 Seasons
1 Season
original CD-ROM, four aerial survey data sets contained data speeds of less than 25 knots). Additional corrections are
38°N
38°N
Sea lions present
in the study area from Point Arena to Point Sal. Of these, the planned for Phase 2 of this project and are briefly discussed
Sea lions absent
OSPR survey program was still ongoing and data from recent below.
years were added to this data set. In addition, data from three
ship-based survey programs were converted to a compatible For example, no adjustments or corrections have been made to
37°N
37°N
format for analysis. See section text for details on individual account for differences in marine mammal detectability among
data sets. species and differential probability of detecting animals from
aerial and shipboard platforms. Individual body size, group
Data sources for aerial at-sea data include MMS-CDAS (MMS, size, and species-specific behaviors, such as proportion of time
2001) and California Department of Fish and Game Office of spent submerged, are all factors known to affect detection and
36°N
36°N
Spill Prevention and Response (CDF&G-OSPR), unpublished hence, observed distribution and density estimates as well.
data. Early data were collected using methods described by Because of the very different attributes of aerial and shipboard
Bonnell et al. (1983); more recent data were collected using platforms, these factors, and the associated adjustments for
updated technology but with the same general method. Data observations, vary among the studies.
sources for ship-based survey data include David Ainley,
35°N
35°N
See additional California sea lion map
c d unpublished data (see Oedekoven et al., 2001 for details Map d was developed using the same approach as for maps a,
for haulout and rookery locations.
on methods). Although the at-sea data span the years 1980- b and c. For each season, the cells with densities in the top 20%
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
2001, data are not available for all seasons in all years. For the of non-zero values were designated “high use” for that season.
Source Data: See text.
Figure 64. Maps for California sea lion: seasonal at-sea densities, high use areas, and rookeries. Upwelling Season, data are from 1980-1982 and 1985-2001.
97
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Cells were scored for “high use” in one, two, or three seasons
and are depicted by color. To provide a relative reference for
the “high use” areas, cells are also shown where the species
were absent (i.e., the cell was sampled but the species was not
recorded there) or present (but densities were never in the top
20% for any season). Further detail on methods is provided in
the Data and Analysis section.
RESULTS AND DISCUSSION
The California sea lion (Z.c.californianus) is subdivided into
three stocks (U.S., Western Baja California, and Gulf of
California); the United States stock begins at the U.S. Mexico
border and extends northward into Canada (Carretta et al.
2001). The breeding areas are on islands located in southern
California, western Baja California, and the Gulf of California.
In the study area, a small number of pups are born on Año
Nuevo Island and Southeast Farallon Island; otherwise the
central California population is composed of non-breeders.
Adult females and immatures remain near the rookeries year-
round, whereas adult males (along with most immatures)
migrate northward to feeding areas from central California to
British Columbia.
In the study area, this species was the most abundant of the
pinnipeds (at-sea sightings: n=1,497 individuals: n=4,411) and
was widely distributed throughout the shelf and upper slope
regions of the three national marine sanctuaries. The seasonal
abundance of California sea lions off central California is linked
to spring and fall pre- and post-breeding migrations. Densities
were greatest during the Oceanic Season (just after breeding)
and Davidson Current Season (before the next breeding period)
and somewhat lower during the Upwelling Season (breeding
period).
Periods of unusually warm ocean waters associated with El
Niño oceanographic conditions affect pup production (i.e., fewer
pups are born) and result in higher mortality rates for pups and
juveniles. During the 1983, 1992, and 1998 El Niño events,
pup production decreased by 35, 27, and 64%, respectively,
at rookeries in southern California (SWFCS 2001). At-sea
distribution patterns were also altered; greater numbers of sea
lions were sighted off central California during these warmer
periods (see also Bonnell & Ford 1987, Trillmich & Ono 1991,
Allen 1994, Keiper 2001, and Keiper et al. In Review.).
98
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THIS MAP from eight shipboard and aerial survey programs conducted in
124°W 123°W 122°W 121°W
Figure 65 shows individual sightings of Steller sea lions at sea the study area in the years 1980-2001 (see “Data Sources”).
200 m
20
39°N
39°N
Steller sea lion Eumetopias jubatus
0
along with the locations of haulouts, rookeries, and at-sea effort Cetacean observation data and trackline data from these stud-
0m
in the study area. At-sea observations are based on combined ies were converted to a common format. All aerial data were
Fish Rocks
data of several studies (see “Methods” and “Data Sources” continuous; ship-based data were converted separately into a
sections). For context, the amount of combined survey effort continuous transect to the extent possible. From the digitized
(km of trackline) is also shown, summarized in 10’x10’ cells. survey data, effort was mapped into 10’x10’ cells using CDAS,
Northwest Cape Rocks
Haulout locations are based on counts conducted in July 2000. a custom geographic information system for analyzing marine
Number of Sea Lions
Blue lines indicate the National Marine Sanctuary boundaries bird and mammal surveys (MMS, 2001). The length and width
At Sea
Bodega Rock
of Cordell Bank, Gulf of the Farallones, and Monterey Bay; of the survey trackline in a given cell (estimated trackline width
Haulouts
Sightings
bathymetric contours for the 200 m and 2,000 m isobaths are varied by platform, depending on speed and height above wa-
3 16 - 36
2 also shown in blue.
3 - 15 ter) were used to estimate the area sampled.
38°N
38°N
1 1-2
Point Reyes
Rookeries DATA SOURCES Note that the these maps represent either sighting locations or
North Farallon Islands 450 - 502 At-sea sightings for the Steller sea lion are based on data from densities that used survey strip widths relative to each survey
seven survey programs conducted in 1980-2001. These data platform (e.g., plane, ship); density was calculated on the basis
50-215
were combined using CDAS software into the MMS-CDAS data of the number of animals sighted and area surveyed. The data
South Farallon Island
Survey Effort
system (MMS, 2001), developed for Minerals Management have only been corrected to normalize for survey effort and
(km of trackline)
Service and expanded for this project. Of the data sets on the to exclude observations with winds greater than 25 knots;
> 3000.00
original CD-ROM, four aerial survey data sets contained data additional corrections are planned for Phase 2 of this project
1500.01 - 3000.00
1000.01 - 1500.00 in the study area from Point Arena to Point Sal. Of these, the and are briefly discussed below.
Año Nuevo Island
500.01 - 1000.00 OSPR survey program was still ongoing and data from recent
100.01 - 500.00
37°N
37°N
years were added to this data set. In addition, data from three For example, no adjustments/corrections have been made to
0.01 - 100.00 ship-based survey programs were converted to a compatible account for differences in marine mammal detectability among
format for analysis. See section overview for details on indi- species and differential probability of detecting animals from
0 25 50 Km
vidual data sets. aerial and shipboard platforms. Individual body size, group
size, and species-specific behaviors, such as proportion of time
Sea Lion Rocks (Pt. Lobos)
Data sources for aerial at-sea data include MMS-CDAS (MMS, spent submerged, are all factors known to affect detection and
Lobos Rocks
2001) and California Department of Fish and Game Office of hence, observed distribution and density estimates as well.
Spill Prevention and Response (CDFG-OSPR), unpublished Because of the very different attributes of aerial and shipboard
data. Early data were collected using methods described by platforms, these factors,and the associated adjustments for
Bonnell et al. (1983); more recent data were collected using observations, vary among the studies.
updated technology but with the same general method. Data
36°N
36°N
sources for ship-based survey data include David Ainley, unpub-
Cape San Martin The data in these maps include wind conditions of up to 25
lished data (see Oedekoven et al., 2001 for details on methods). knots; smaller or less obvious species are often less detectable
Although the at-sea data span the years 1980-2001, data are even at wind speeds of less than 25 knots. The seasonal maps
not available for all seasons in all years. For the Upwelling Sea- contain different combinations of shipboard and aerial data;
son, data are from 1980-1982 and 1985-2001. For the Oceanic therefore the seasonal densities from these platforms may not
Season, data are from 1980-1982, 1991 and 1994-2001. For be directly comparable. A full consideration of these factors, and
the Davidson Current Season, data are from 1980-1986 and revised maps, are planned for Phase 2 of this project.
1991-2001. Rookery and haulout data are from Mark Lowry
Pecho Rock
of NMFS’ Southwest Fisheries Science Center. Rookery and RESULTS AND DISCUSSION
haulout data are from Mark Lowry of NMFS’ Southwest Fisher- Steller sea lions range from northern Japan, the Aleutian Islands
35°N
35°N
ies Science Center. The rookery numbers represent a general and Gulf of Alaska, south to Año Nuevo Island, California (the
range based on counts of all animals (pups and adults) in three southernmost rookery). Steller sea lion females and pups are
years, 1999-2001. The haulout data are from July 2000.
DRAFT
found at the rookeries year-round, but adult bulls are only at
the rookery during the breeding season (mid-May to mid-July).
METHODS In the study area, the Steller sea lion occurred over the shelf
The latitude and longitude coordinates of Steller sea lions at and slope, and, although there were few at-sea sightings in
sea were used to plot the individual sightings; the coordinates the data set, most occurred in the area between Cordell Bank
124°W 123°W 122°W 121°W
from Mark Lowry were used to plot the haulouts and rookeries. and Año Nuevo Island.
Source Data: See text.
At-sea sightings and effort are the result of a synthesis of data
Figure 65. Map for Steller sea lion: at-sea sightings and survey effort, rookeries and haulouts.
99
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
The Steller sea lion population has declined approximately
64% throughout its range (NMFS Biological Opinion, 2000;
see also NMFS 1992). Based on distributional data, Steller
sea lions are classified into two separate stocks within U.S.
waters: 1) the western stock, that includes animals at, and
west of, Cape Suckling, Alaska, (classified as endangered);
and 2) the eastern stock (including the California population)
that includes animals east of Cape Suckling (classified as
federally threatened). Greatest concentrations of Steller sea
lions occur north of central California, hence, relatively few
sightings (n=45 sightings; n=50 individuals) occurred in the
study area. Insufficient data precluded mapping the Steller
sea lion by seasons.
The breeding season is from mid-May to mid-July. In the study
area, rookeries are located at the Farallon Islands (where they
breed in small numbers and haul-out in slightly larger numbers
throughout the year, USFWS 2000) and at Point Año Nuevo
Island (see LeBoeuf et al. 1991). From 1977 to 1996 on the
Farallon Islands, adult females present during the breeding
season declined by 5.9% and maximum number of pups
counted declined significantly (see Hastings and Sydeman
2002). Until the early 1970’s, Steller sea lions used to breed
at the Point Reyes Headlands but in recent years numbers
have been low (fewer than 50; S. Allen pers. comm., 2003).
Haulout sites north of San Francisco are located at Fish Rocks,
Northwest Cape Rocks, Bodega Rocks, Point Reyes and the
Farallon Islands. Another haulout site not on the map is located
north of Fort Ross at “Sea Lion Rocks”; maximum counts at this
site occur in June (approx. 50) and consist mostly of females
with pups of the year (J. Mortenson pers.comm., 2003). Adult
males disperse widely during the non-breeding season.
Numbers of Steller sea lions off southern and central California
have declined significantly, from 5,000-7,000 non-pups in
1927-1947, to 1,500-2,000 non-pups between 1980-1998
(NMFS Biological Opinion, 2000). Threats to Steller sea
lions include incidental take by commercial fisheries, getting
shot, entanglement in marine debris, declining trends in prey
availability, disease, and contaminants (e.g. premature births
accounted for 20-60% of pup mortality in the South Farallon
Islands between 1973-83). Organochlorine and trace metal
contaminant levels are still elevated in central California Steller
sea lions (NMFS Biological Opinion 2000).
Steller sea lions feed on walleye pollock, capelin, mackerel,
rockfish, herring, salmon, octopus and squid (see Riedman,
1990).
100
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THESE MAPS Upwelling Season, data are from 1980-1982 and 1985-2001.
DRAFT
Northern fur seal Figures 66a, b and c show the density (animals/km2) of northern Oceanic Season, data are from 1980-1982, 1991, and 1994-
Callorhinus ursinus fur seals in the Upwelling, Oceanic, and Davidson Current 2001. Davidson Current Season, data are from 1980-1986 and
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
seasons, displayed in 10’x10’ cells. Densities are based on 1991-2001.
Upwelling Season Oceanic Season
200 m
200 m
20
20
combined data of several studies (see “Methods” and "Data
0
0
0m
0
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14)
m
Sources" sections). The color and mapping intervals were Information on the northern fur seal rookery was provided in
39°N
39°N
customized to show the most structure and highlight significant 2003, courtesy of William Sydeman of PRBO Conservation
Density areas, while allowing comparisons among marine mammal Science, and Joelle Buffa of the Farallon Islands National
(Animals/km²)
species. Cells that were surveyed but which had no northern fur Wildlife Refuge.
10.01 - 50.00 seal’s have a density of zero; unsurveyed areas appear white.
5.01 - 10.00
Blue lines indicate the National Marine Sanctuary boundaries METHODS
38°N
38°N
1.01 - 5.00
of Cordell Bank, Gulf of the Farallones, and Monterey Bay; At-sea densities are the result of a synthesis of data from seven
0.51 - 1.00
bathymetric contours for the 200 m and 2,000 m isobaths are shipboard and aerial survey programs conducted in the study
0.11 - 0.50
also shown in blue. area in the years 1980-2001 (see “Data Sources”). Pinniped
0.06 - 0.10
observation data and trackline data from these studies were
0.01 - 0.05
37°N
37°N
In order to provide one map for the species that integrates the converted to a common format. All aerial data were continuous;
0.00
patterns of its spatial and temporal occurrence in the study ship-based data were converted separately into a continuous
0 25 50 Km
area, map d shows seasonal high use areas, displayed in transect to the extent possible. From the digitized survey data,
10’x10’ cells. This map provides a further synthesis of densities the distributions of effort and of species were mapped into
presented in maps a, b and c (see “Methods” section for details), 10’x10’ cells using CDAS, a custom geographic information
36°N
36°N
and portrays the relative importance of various areas to the system for analyzing marine bird and mammal surveys (MMS,
species. Areas with consistent high use are highlighted on this 2001). The length and width of the survey trackline in a given
map. To provide a relative reference for the “high use” areas, cell (estimated trackline width varied by platform, depending on
cells are also shown where the species were absent (i.e., the speed and height above water) were used to estimate the area
cell was sampled but the species was not recorded there), or sampled. The number of each species seen in a cell was then
35°N
35°N
a b present but at lesser concentrations in any particular season. divided by the area sampled in the cell to estimate density. If a
The single rookery location is also shown. cell was censused more than once, densities were averaged,
200 m
Davidson Current Season
200 m
Seasonal High Use Areas
20
with adjustment made for effort.
20
0
0
0m
0m
DATA SOURCES
(Nov. 15 - Mar. 14) and Rookery
39°N
39°N
At-sea densities for the northern fur seal are based on data Note that these maps represent either sighting locations or
Persistence of
from seven survey programs conducted in 1980-2001. These densities that used survey strip widths relative to each survey
High Use
data were combined using CDAS software into the MMS-CDAS platform (e.g., plane, ship); density was calculated on the basis
3 Seasons
data system (MMS, 2001), developed for Minerals Management of the number of animals sighted and area surveyed. The data
2 Seasons
1 Season
Service and expanded for this project. Of the data sets on the have only been corrected to normalize for survey effort and to
38°N
38°N
Seals present
original CD-ROM, four aerial survey data sets contained data exclude observations with winds greater than 25 knots (smaller
Seals absent
in the study area from Point Arena to Point Sal. Of these, the or less obvious species are often less detectable even at wind
Rookery
OSPR survey program was still ongoing and data from recent speeds of less than 25 knots). Additional corrections are
years were added to this data set. In addition, data from three planned for Phase 2 of this project and are briefly discussed
37°N
37°N
ship-based survey programs were converted to a compatible below.
format for analysis. See "Data and Analyses" subsection in 2.3
for details on individual data sets. For example, no adjustments or corrections have been made to
account for differences in marine mammal detectability among
Data sources for aerial at-sea data include MMS-CDAS (MMS, species and differential probability of detecting animals from
36°N
36°N
2001) and California Department of Fish and Game Office of aerial and shipboard platforms. Individual body size, group
Spill Prevention and Response (CDF&G-OSPR), unpublished size, and species-specific behaviors, such as proportion of
data. Early data were collected using methods described by time spent submerged, are all factors known to affect detection
Bonnell et al. (1983); more recent data were collected using and hence, observed distribution and density estimates as well.
updated technology but with the same general method. Data Because of the very different attributes of aerial and shipboard
35°N
35°N
c d sources for ship-based survey data include David Ainley, platforms, these factors, and the associated adjustments for
unpublished data (see Oedekoven et al. 2001 for details on observations, vary among the studies.
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
methods). Although the at-sea data span the years 1980-2001,
Source Data: See text.
data are not available for all seasons in all years. For the
Figure 66. Maps for northern fur seal: seasonal at-sea densities, high use areas, and rookery.
101
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Northern fur seals feed on a great diversity of seasonally
Map d was developed using the same approach as for maps
abundant prey, and, off California, primary prey species include
a, b and c. For each season, the cells with densities in the top
Pacific hake, northern anchovy, mesopelagic fishes, and market
20% of non-zero values were designated “high use” for that
squid (Kajimura, 1984; see also Riedman, 1990).
season. Cells were scored for “high use” in one, two, or three
seasons and are depicted by color. To provide a relative refer-
ence for the “high use” areas, cells are also shown where the
species were absent (i.e., the cell was sampled but the species
was not recorded there) or present (but densities were never
in the top 20% for any season). Further detail on methods is
provided in the "Data and Analysis" section.
RESULTS AND DISCUSSION
The northern fur seal, one of the most pelagic of the pinnipeds,
is most abundant in continental shelf-slope waters of mid-
latitudes off western North America during winter and early
spring. Except for a small, recently re-established rookery
(south Farallon Island, see below), rookeries occur primarily
outside of the study area. The breeding and pupping season
is June-July, and suckling can continue for three additional
months. During autumn, adult females and juveniles migrate
from rookeries on San Miguel Island in the southern California
Bight (the San Miguel Island stock) and from the Eastern Pacific
stock of the Pribilof Islands in the Bering Sea (Kajimura, 1980;
Kenyon and Wilke, 1953; Pyle et al., 2001). Adult females and
pups from the Pribilof Islands migrate into the North Pacific
Ocean and to waters off Oregon and California.
In data used for this assessment (1980-2001), the northern fur
seal was the second most abundant pinniped observed, with
a total of 1,459 sightings and 2,070 individuals. In the study
area, greatest densities occurred seaward of National Marine
Sanctuary boundaries in the shelf-break, slope, and deep ocean
habitats. The distinctly seasonal presence of this species is
clearly evident in the study area, with greater numbers from
February to May (Kajimura 1984). Greatest densities occur
during the Upwelling and Davidson Current seasons (non-
breeding period) and lesser densities during the Oceanic
Season (breeding period).
Severe declines associated with periods of unusually warm
ocean conditions affect pup production, mortality rates on San
Miguel Island and the Pribilof Islands, and at-sea presence of
this species (see DeLong and Antonelis, 1991; Allen, 1994;
DeLong and Melin, 1999; Melin and DeLong, 2000; Keiper,
2001; and Keiper et al., In Review) In the early 19th century,
American, British, and Russian sealers removed the breeding
population from the South Farallon Islands (Pyle et al., 2001).
Beginning in 1996, however, the species has re-established a
breeding population on the South Farallon Islands, with fewer
than 10 pups produced each year, 1997-2001 (Pyle et al.,
2001). Seasonal high use areas occurred mostly to the west
of National Marine Sanctuary boundaries.
102
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THESE MAPS a common format. All aerial data were continuous; ship-based
124°W 123°W 122°W 121°W
Figure 67 shows individual sightings of harbor seals at sea data were converted separately into a continuous transect to
200 m
20
39°N
39°N
Harbor seal Phoca vitulina
0
0m
along with the locations of haulouts and at-sea survey effort in the extent possible. From the digitized survey data, effort was
the study area. At-sea observations are based on combined mapped into 10’x10’ cells using CDAS, a custom geographic
data of several studies (see “Methods” and “Data Sources” information system for analyzing marine bird and mammal sur-
sections). For context, the amount of combined survey effort veys (MMS-CDAS, 2001). The length and width of the survey
(km of trackline) is also shown, summarized in 10’x10’ cells. trackline in a given cell (estimated trackline width varied by
Haulout locations are based on aerial counts from 2002. Blue platform, depending on speed and height above water) were
Number of Seals lines indicate the National Marine Sanctuary boundaries of used to estimate the area sampled.
At Sea
Cordell Bank, Gulf of the Farallones, and Monterey Bay; bathy-
Haulout Sites
Sightings
metric contours for the 200 m and 2,000 m isobaths are also Note that these maps represent either sighting locations or
7 - 15 401 - 835
shown in blue. densities that used survey strip widths relative to each survey
38°N
38°N
4-6 201 - 400
platform (e.g., plane, ship); density was calculated on the basis
2-3 51 - 200
DATA SOURCES of the number of animals sighted and area surveyed. The data
1 - 50
1
At-sea sightings for the harbor seal are based on data from have only been corrected to normalize for survey effort and
seven survey programs conducted in 1980-2001. These data to exclude observations with winds greater than 25 knots;
Survey Effort
were combined using CDAS software into the MMS-CDAS data additional corrections are planned for Phase 2 of this project
(km of trackline)
system (MMS, 2001), developed for Minerals Management and are briefly discussed below.
> 3000.00
1500.01 - 3000.00 Service and expanded for this project. Of the data sets on the
1000.01 - 1500.00 original CD-ROM, four aerial survey data sets contained data For example, no adjustments/corrections have been made to
500.01 - 1000.00 in the study area from Point Arena to Point Sal. Of these, the account for differences in marine mammal detectability among
100.01 - 500.00 OSPR survey program was still ongoing and data from recent species and differential probability of detecting animals from
0.01 - 100.00
37°N
37°N
years were added to this data set. In addition, data from three aerial and shipboard platforms. Individual body size, group
ship-based survey programs were converted to a compatible size, and species-specific behaviors, such as proportion of
format for analysis. See "Data and Analyses" subsection of this time spent submerged, are all factors known to affect detection
0 25 50 Km
mammal section (2.3). and hence, observed distribution and density estimates as well.
Because of the very different attributes of aerial and shipboard
Data sources for aerial at-sea data include MMS-CDAS (2001) platforms, these factors, and the associated adjustments for
and California Department of Fish and Game Office of Spill observations, vary among the studies.
Prevention and Response (CDFG-OSPR), unpublished data.
Early data were collected using methods described by Bonnell The data in these maps include wind conditions of up to 25
et al. (1983); more recent data were collected using updated knots; smaller or less obvious species are often less detectable
36°N
36°N
technology but with the same general method. Data sources for even at wind speeds of less than 25 knots. The seasonal maps
ship-based survey data include David Ainley, unpublished data contain different combinations of shipboard and aerial data;
(see Oedekoven et al., 2001 for details on methods). Although therefore the seasonal densities from these platforms may not
the overall at-sea data span the years 1980-2001, data are not be directly comparable. A full consideration of these factors, and
available for all seasons in all years. For the Upwelling Season, revised maps, are planned for Phase 2 of this project.
data are from 1980-1982 and 1985-2001. For the Oceanic
Season, data are from 1980-1982, 1991 and 1994-2001. For RESULTS AND DISCUSSION
the Davidson Current Season, data are from 1980-1986 and The harbor seal is distributed from the eastern Aleutian Islands
1991-2001. Haulout information is from 2002 aerial survey data to Baja California and inhabits near-shore estuarine, coastal
(6/12/2002-7/1/2002), from Mark Lowry of NMFS’ Southwest and shelf areas. When at sea, harbor seals were distributed
Fisheries Science Center. in shelf habitats in relatively low densities in all three national
35°N
35°N
marine sanctuaries; therefore, insufficient data precluded
METHODS generating seasonal maps. Harbor seals forage throughout
DRAFT
The latitude/longitude coordinates of harbor seals at sea were the coastal waters. Because the at-sea locations in this map
used to plot the individual sightings. Haulouts were mapped are influenced by survey effort (where survey effort was unequal
coordinates provided by Mark Lowry. At-sea sightings and ef- and coverage was less along the coast), the map may not
fort are the result of a synthesis of data from eight shipboard accurately represent the foraging distribution of harbor seals.
and aerial survey programs conducted in the study area in the Although not evident in the maps, densities are higher in the
124°W 123°W 122°W 121°W
Source Data: See text. years 1980-2001 (see “Data Sources”). Cetacean observation Gulf of the Farallones because there are more and larger
data and trackline data from these studies were converted to haul-out sites in this area (Allen et al., 2002). Harbor seals do
Figure 67. Map for harbor seal: at-sea sightings, survey effort and haulouts.
103
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
not make extensive migrations, and tend to remain relatively
close to their haul-out sites throughout the year. Harbor seals
are inconspicuous at sea and may explain the relatively
low numbers of animals surveyed at sea (sightings: n=192;
individuals: n=235). A long-term monitoring project at Bolinas
Lagoon (Gulf of the Farallones Sanctuary Education Awareness
and Long-term Stewardship Program) protects the seals from
human disturbance. During breeding and molting, relative
abundance increases at Drakes Estero, whereas during winter
(and during herring spawns) relative abundance increases in
Tomales Bay. The Point Reyes region represents ~20% (6000
seals) of the breeding population of the state of California (S.
Allen pers. comm.). Results of recent (2002) tagging studies
have indicated individuals from San Francisco Bay travel to
Duxbury Reef and out to the Farallon Islands to forage (S. Allen
pers.comm.). Harbor seals feed on seasonally abundant prey
that includes topsmelt, night smelt, white croaker, English sole
(Harvey et al., 1995), salmonids (Weise, 2001), and squid and
octopus (see also Riedman, 1990).
In the study area, the species is present year-round, and on
land it is found on sandy beach, mudflat and rocky habitats.
Haulout sites (identified by Lowry 2002) are located along the
coast from Point Arena south to Point Conception, within San
Francisco Bay, and at the Southeast Farallon Islands; habitat
use at these sites, however, varies seasonally throughout the
year (S. Allen, pers. comm.).
Breeding and pupping occurs March-July, and many pupping
sites occur in the study area. Along the Point Reyes National
Seashore, major pupping sites occur at the following locations
(S. Allen pers.comm.): Bodega Rock, Bodega Point, Tomales
Bay (four sites), Tomales Point (five sites), Drakes Estero
(five sites), Limantour Spit, Double Point (two sites), Abalone
Point, Bolinas Point, Duxbury Reef, Bolinas Lagoon (3 sites),
Slide Ranch, and Point Bonita. Sites along the coast south of
San Francisco may exist at Pescadero and Bean Hollow, but
these sites are poorly documented (D.Greig, pers.comm.).
Pupping sites within San Francisco Bay are located at Mowry
Slough and Castro Rocks. Farther south, pupping sites also
occur at Año Nuevo Island, Elkhorn Slough, Hopkins Marine
Station, Cypress Point, Fanshell Beach and Cypress Point,
San Lorenzo River and Point Lobos (D.Greig pers.comm.). A
few pups (less than five) were also produced on South Farallon
Island (USFWS, 2000).
104
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THIS MAP
124°W 123°W 122°W 121°W
continuous; ship-based data were converted separately into a
Figure 68 shows individual at-sea sightings of northern elephant continuous transect to the extent possible. From the digitized
200 m
20
39°N
39°N
Northern elephant seal Mirounga angustirostris
0
0m
seals at sea, along with the locations of rookeries and at-sea survey data, effort was mapped into 10’x10’ cells using CDAS,
survey effort in the study area. At-sea observations are based a custom geographic information system for analyzing marine
on combined data of several studies (see “Methods” and “Data bird and mammal surveys (MMS, 2001). The length and width
Sources” sections). For context, the amount of combined survey of the survey trackline in a given cell (estimated trackline width
effort (km of trackline) is also shown, summarized in 10’x10’ varied by platform, depending on speed and height above wa-
cells. Blue lines indicate the National Marine Sanctuary bound- ter) were used to estimate the area sampled.
Number of Seals
aries of Cordell Bank, Gulf of the Farallones, and Monterey Bay;
At Sea Rookery and
bathymetric contours for the 200 m and 2,000 m isobaths are Note that the these maps represent either sighting locations or
Sightings Haulout Sites
also shown in blue. densities that used survey strip widths relative to each survey
5
platform (e.g., plane, ship); density was calculated on the basis
Point Reyes
38°N
38°N
2
DATA SOURCES of the number of animals sighted and area surveyed. The data
1 At-sea sightings for the northern elephant seal are based on have only been corrected to normalize for survey effort and
data from seven survey programs conducted in 1980-2001. to exclude observations with winds greater than 25 knots;
South Farallon Island
These data were combined using CDAS software into the additional corrections are planned for Phase 2 of this project
Survey Effort
MMS-CDAS data system (MMS, 2001), developed for Miner- and are briefly discussed below.
(km of trackline)
als Management Service and expanded for this project. Of the
> 3000.00
data sets on the original CD-ROM, four aerial survey data sets For example, no adjustments/corrections have been made to
1500.01 - 3000.00
contained data in the study area from Point Arena to Point Sal. account for differences in marine mammal detectability among
1000.01 - 1500.00
500.01 - 1000.00 Of these, the OSPR survey program was still ongoing and data species and differential probability of detecting animals from
Año Nuevo Mainland/Island
100.01 - 500.00 from recent years were added to this data set. In addition, data aerial and shipboard platforms. Individual body size, group
37°N
37°N
0.01 - 100.00 from three ship-based survey programs were converted to a size, and species-specific behaviors, such as proportion of time
compatible format for analysis; see Data and Analysis subsec- spent submerged, are all factors known to affect detection and
tion for more information on individual data sets. hence, observed distribution and density estimates as well.
0 25 50 Km
Because of the very different attributes of aerial and shipboard
Data sources for aerial at-sea data include MMS-CDAS (MMS, platforms, these factors, and the associated adjustments for
2001) and California Department of Fish and Game Office of observations, vary among the studies.
Spill Prevention and Response (CDFG-OSPR), unpublished
data. Early data were collected using methods described by The data in these maps include wind conditions of up to 25
Bonnell et al. (1983); more recent data were collected using knots; smaller or less obvious species are often less detectable
updated technology but with the same general method. Data even at wind speeds of less than 25 knots. The seasonal maps
36°N
36°N
sources for ship-based survey data include David Ainley, contain different combinations of shipboard and aerial data;
Cape San Martin
unpublished data (see Oedekoven et al., 2001 for details therefore the seasonal densities from these platforms may not
Pt. Piedras Blancas on methods). Although the at-sea data span the years 1980- be directly comparable. A full consideration of these factors, and
2001, data are not available for all seasons in all years. For the revised maps, are planned for Phase 2 of this project.
Upwelling Season, data are from 1980-1982 and 1985-2001.
For the Oceanic Season, data are from 1980-1982, 1991, and RESULTS AND DISCUSSION
1994-2001. For the Davidson Current Season, data are from The northern elephant seal is present year-round in the
1980-1986 and 1991-2001. Information on rookery locations study area; however, because they spend very little time
was obtained from Pat Morris, USCS; Brian Hatfield, USGS; at the surface, at-sea sightings are rare, as evidenced by
and Joelle Buffa, FWS. the relatively few sightings during surveys in the study area
(n=266 sightings; n=273 individuals). Therfore, insufficient data
35°N
35°N
METHODS precluded mapping the northern elephant seal by seasons.
The latitude/longitude coordinates of northern elephant seals at
DRAFT
sea were used to plot the individual sightings; the coordinates Northern elephant seals were widely distributed in shelf, shelf-
for rookeries and haulouts were used to plot their locations. break, and slope habitats within the three national marine
At-sea sightings and effort are the result of a synthesis of data sanctuaries, and also occurred in deep ocean habitats seaward
from eight shipboard and aerial survey programs conducted in of the 2000 m isobath. They also occurred well to the north,
the study area in the years 1980-2001 (see “Data Sources”). west, and south of sanctuary boundaries. In these data sets,
124°W 123°W 122°W 121°W
Source Data: See text. Cetacean observation data and trackline data from these stud- age classes of at-sea sightings of seals are unknown.
ies were converted to a common format. All aerial data were
Figure 68. Map for northern elephant seal: at-sea sightings and survey effort, rookeries and haulouts.
105
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
The northern elephant seal breeds, gives birth, and molts on Northern elephant seals are prolonged deep divers that feed on
islands and coastal regions in California, as well as offshore deepwater fishes and invertebrates, including Pacific hagfish
islands of Baja California. The breeding period in the study (Eptatretus stouti), ratfish (Hydrolagus colliei), Pacific hake,
area is generally December through March (Stewart and rockfish, sharks, rays, squid and octopus (Antonelis et al., 1987;
Huber,1993). Northern elephant seals migrate between Condit and LeBoeuf, 1984).
rookeries located within sanctuary boundaries, Farallon Islands,
Point Reyes, Año Nuevo Island and the mainland, Piedras
Blancas, Cape San Martin, and San Simeon, and waters to
the north, where they spend eight to ten months of the year
feeding. Adult males feed in the eastern Aleutian Islands and
the Gulf of Alaska; adult females feed to the west and south of
45º N in deep, oceanic water (Le Boeuf et al., 1993; Stewart
and Huber, 1993; Stewart et al., 1994).
On land, there are three peaks in abundance: 1) during the
breeding/pupping season December to March, with peaks
the last week of January; 2) during the molting season when
female and immatures are on shore April to July with peaks in
May, and adult males are on shore June to early August; and
3) during September to October when immatures haul-out (S.
Allen, pers. comm). Pups depart the pupping sites during the
Upwelling Season. Recent tagging studies indicate that pups
from this region travel as far as Alaska (S.Allen, unpublished
data, National Park Service).
Each year at Año Nuevo Island and mainland, there are
approximately 2,400 females and 300-400 males present,
and approximately 2,200 pups are produced (P. Morris, pers.
comm, 2003). Based on pup counts, the population there
steadily increased through the mid 1990s, but now appears to
be stable (P. Morris, pers. comm., credited to B.J. Le Boeuf).
In contrast, the colony at Piedras Blancas has continued to
rise (in general) over the past five years (B. Hatfield, pers.
comm.) Productivity has declined at two major breeding sites
on Southeast Farallon Island (Sydeman and Allen, 1999;
Nusbaum, 2002), with erosion playing a major role in limiting
the species’ population (USFWS 2000). In California, the net
productivity rate for northern elephant seals also appears to
have declined in recent years (Carretta et al., 2002). However,
the colony at Point Reyes Headlands has continued to increase
by 5-10% per year (Sydeman and Allen, 1999; S. Allen, pers.
comm. 2003). Due to the high surf during the strong El Niño
of 1998, extensive pup mortality occurred at the Point Reyes
colony (Pettee, 1999), but also forced the relocation of the
breeding area; some moved from the main colony at Point
Reyes Headlands to South Beach and North Drakes Bay Beach
(Pettee, 1999).
106
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THESE MAPS the years 1980-2001, data are not available for all seasons in
DRAFT
Dall's porpoise Figures 69a, b and c show the density (animals/km2) of Dall’s all years. For the Upwelling Season, data are from 1980-1982
Phocoenoides dalli porpoise in the Upwelling, Oceanic, and Davidson Current and 1985-2001. For the Oceanic Season, data are from 1980-
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
seasons, displayed in 10’x10’ cells. Densities are based on 1982, 1991, and 1994-2001. For the Davidson Current Season,
Upwelling Season Oceanic Season
200 m
200 m
20
20
combined data of several studies (see “Methods” and “Data data are from 1980-1986 and 1991-2001.
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) Sources” sections). The color and mapping intervals were
39°N
39°N
customized to highlight significant areas and the structure of METHODS
Density species spatial occurrence, while allowing comparisons among At-sea densities are the result of a synthesis of data from eight
(Animals/km²)
marine mammal species. Cells that were surveyed but which shipboard or aerial survey programs conducted in the study
10.01 - 50.00
had no Dall’s porpoise have a density of zero; unsurveyed area in the years 1980-2001 (see “Data Sources”). Cetacean
5.01 - 10.00
38°N
38°N
areas appear white. Blue lines indicate the National Marine observation data and trackline data from these studies were
1.01 - 5.00
Sanctuary boundaries of Cordell Bank, Gulf of the Farallones, converted to a common format. All aerial data were continuous;
0.51 - 1.00
and Monterey Bay; bathymetric contours for the 200 m and ship-based data were converted separately into a continuous
0.11 - 0.50
2,000 m isobaths are also shown in blue. transect to the extent possible. From the digitized survey data,
0.06 - 0.10
the distributions of effort and of species were mapped into
0.01 - 0.05
37°N
37°N
In order to provide one map for the species that integrates the 10’x10’ cells using CDAS, a custom geographic information
0.00
patterns of its spatial and temporal occurrence year-round in the system for analyzing marine bird and mammal surveys (MMS,
0 25 50 Km
study area, map d shows seasonal high use areas, displayed in 2001). The length and width of the survey trackline in a given
10’x10’ cells. This map provides a further synthesis of densities cell (estimated trackline width varied by platform, depending on
presented in maps a, b and c (see the “Methods” section for speed and height above water) were used to estimate the area
36°N
36°N
details), and portrays the relative importance of various areas sampled. The number of cetaceans of each species seen in a
to the species. Areas of consistent high use (colored according cell was then divided by the area sampled in the cell to estimate
to the number of high seasonal use) are highlighted on this density. If a cell was censused more than once, densities were
map. To provide a relative reference for the “high use” areas, averaged, with adjustment made for effort.
cells are also shown where the species were absent (i.e., the
35°N
35°N
a b cell was sampled but the species was not recorded there) or Note that these maps represent either sighting locations or
present but at lower concentrations (densities in a cell were densities that used survey strip widths relative to each survey
200 m
Davidson Current Season
200 m
Seasonal High Use Areas
20
never in the top 20%) in any particular season.
20
platform (e.g., plane, ship); density was calculated on the basis
0
0
0m
0m
(Nov. 15 - Mar. 14) of the number of animals sighted and area surveyed. The data
39°N
39°N
have only been corrected to normalize for survey effort and to
DATA SOURCES
Persistence of
High Use exclude observations with winds greater than 25 knots (smaller
At-sea densities for cetaceans are based on data from eight
3 Seasons
or less obvious species are often less detectable even at
survey programs conducted in 1980-2001. These data were
2 Seasons
wind speeds of less than 25 knots). Additional corrections are
1 Season combined using CDAS software into the MMS-CDAS data
Porpoises present
planned for Phase 2 of this project and are briefly discussed
38°N
38°N
system (MMS, 2001), developed for Minerals Management
Porpoises absent
below.
Service and expanded for this project. Of the data sets on the
original CD-ROM, five aerial survey data sets contained data
For example, no adjustments or corrections have been made to
in the study area from Point Arena to Point Sal. Of these, the
account for differences in marine mammal detectability among
OSPR survey program was still ongoing and data from recent
37°N
37°N
species and differential probability of detecting animals from
years were added to this data set. In addition, data from three
aerial and shipboard platforms. Individual body size, group
ship-based survey programs were converted to a compatible
size, and species-specific behaviors, such as proportion of time
format for analysis. See "Data and Analyses" subsection in 2.3
spent submerged, are all factors known to affect detection and
for details on individual data sets.
hence, observed distribution and density estimates as well.
36°N
36°N
Because of the very different attributes of aerial and shipboard
Data sources for aerial at-sea data include MMS-CDAS (MMS,
platforms, these factors, and the associated adjustments for
2001) and California Department of Fish and Game Office of
observations, vary among the studies.
Spill Prevention and Response (CDF&G-OSPR), unpublished
data. Early data were collected using methods described by
Map d was developed using the same approach as for maps a,
Bonnell et al. (1983) and Dohl et al. (1983); more recent data
35°N
35°N
c d b and c. For each season, the cells with densities in the top 20%
were collected using updated technology but with the same
of non-zero values were designated “high use” for that season.
general method. Data sources for ship-based survey data
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Cells were scored for “high use” in one, two, or three seasons
Source Data: See text. include David Ainley, unpublished data (see Oedekoven et al.
and are depicted by color. To provide a relative reference for
2001 for details on methods). Although the at-sea data span
Figure 69. Maps for Dall’s porpoise: seasonal at-sea densities and high use areas.
107
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
the “high use” areas, cells are also shown where the species
were absent (i.e., the cell was sampled but the species was not
recorded there) or present (but densities were never in the top
20% for any season). Further detail on methods is provided in
the "Data and Analysis" section.
RESULTS AND DISCUSSION
Dall’s porpoise is widely distributed in temperate North Pacific
waters. In the study area, this species was the fourth most
numerous small cetacean. Dall’s porpoise was present during
all seasons in shelf, upper/lower slope, canyon, and deep ocean
habitats seaward of the 2000 m isobath.
During the Upwelling Season densities were somewhat greater
in the Cordell Bank National Marine Sanctuary (NMS) and
northern regions of the Gulf of the Farallones NMS. During
the Oceanic Season, densities were somewhat greater within
and to the north of Cordell Bank and the northern portion of
the Monterey Bay National Marine Sanctuary (MBNMS). The
widespread and deep ocean distribution of the Dall’s porpoise
(well to the west of the National Marine Sanctuary boundaries)
was most evident during the Davidson Current Season (when
effort was greater offshore). No clear seasonal pattern was
evident.
The distribution of Dall’s porpoise is highly variable between
years and appears to be affected by oceanographic conditions
(Forney and Barlow 1998). North-south movements of this
species occur as oceanographic conditions change on seasonal
and interannual time scales (see Green et al., 1992; Barlow,
1995; Forney et al., 1995).
High use areas (based on the CDAS maps) occurred along the
200 m isobath in the Cordell Bank and Gulf of the Farallones
NMS. Given the highly variable distribution of Dall’s porpoise,
the apparent higher relative density in these regions may not
be a seasonal pattern.
See map of SWFSC survey data for Dall’s porpoise (Figure 76)
in this section for the greater geographic extent of the range
and interannual variations for this species.
Dall’s porpoise feeds mostly on Pacific hake (Merluccius
productus), northern anchovy (Engraulis mordax), Pacific
saury (Cololabis saira), juvenile rockfish (Sebastes spp), and
cephalopods (Koskii et al., 1998; Morejohn, 1979).
108
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THESE MAPS and 1985-2001. For the Oceanic Season, data are from 1980-
Lagenorhynchus obliquidens DRAFT
Pacific white-sided dolphin Figures 70a, b and c show the density (animals/km2) of Pacific 1982, 1991 and 1994-2001. For the Davidson Current Season,
white-sided dolphin in the Upwelling, Oceanic, and Davidson data are from 1980-1986 and 1991-2001.
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Current seasons, displayed in 10’x10’ cells. Densities are
Upwelling Season Oceanic Season
200 m
200 m
20
based on combined data of several studies (see “Methods”
20
METHODS
0
0
0m
0m
and “Data Sources” sections). The color and mapping intervals
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) At-sea densities are the result of a synthesis of data from eight
39°N
39°N
were customized to show the most structure and highlight shipboard and aerial survey programs conducted in the study
Density significant areas, while allowing comparisons among marine area in the years 1980-2001 (see “Data Sources” section).
(Animals/km²)
mammal species. Cells that were surveyed but which had no Cetacean observation data and trackline data from these
10.01 - 50.00
Pacific white-sided dolphins have a density of zero; unsurveyed studies were converted to a common format. All aerial data were
5.01 - 10.00
areas appear white. Blue lines indicate the National Marine continuous; ship-based data were converted separately into a
38°N
38°N
1.01 - 5.00
Sanctuary boundaries of Cordell Bank, Gulf of the Farallones, continuous transect to the extent possible. From the digitized
0.51 - 1.00
and Monterey Bay; bathymetric contours for the 200m and survey data, the distributions of effort and of species were
0.11 - 0.50
2,000m isobaths are also shown in blue. mapped into 10’x10’ cells using CDAS, a custom geographic
0.06 - 0.10
information system for analyzing marine bird and mammal
0.01 - 0.05
37°N
37°N
In order to provide one map for the species that integrates the surveys (MMS, 2001). The length and width of the survey
0.00
patterns of its spatial and temporal occurrence in the study area, trackline in a given cell (estimated trackline width varied by
0 25 50 Km
map d shows seasonal high use areas, displayed in 10’x10’ cells. platform, depending on speed and height above water) were
This map provides a further synthesis of densities presented used to estimate the area sampled. The number of cetaceans
in maps a, b and c (see the “Methods” section for details), and of each species seen in a cell was then divided by the area
36°N
36°N
portrays the relative importance of various areas to the species. sampled in the cell to estimate density. If a cell was censused
Areas with consistent high use are highlighted on this map. To more than once, densities were averaged, with adjustment
provide a relative reference for the “high use” areas, cells are made for effort.
also shown where the species were absent (i.e., the cell was
sampled but the species was not recorded there), or present Note that these maps represent either sighting locations or
35°N
35°N
a b but at lesser concentrations in any particular season. densities that used survey strip widths relative to each survey
platform (e.g., plane, ship); density was calculated on the basis
200 m
Davidson Current Season
200 m
DATA SOURCES
Seasonal High Use Areas of the number of animals sighted and area surveyed. The data
20
20
0
0
0m
0m
At-sea densities for cetaceans are based on data from eight have only been corrected to normalize for survey effort and to
(Nov. 15 - Mar. 14)
39°N
39°N
survey programs conducted in 1980-2001. These data were exclude observations with winds greater than 25 knots (smaller
Persistence of
combined using CDAS software into the MMS-CDAS data
High Use or less obvious species are often less detectable even at wind
3 Seasons
system (MMS, 2001), developed for Minerals Management speeds of less than 25 knots). Additional corrections are
2 Seasons
Service and expanded for this project. Of the data sets on the planned for Phase 2 of this project and are briefly discussed
1 Season
Dolphins present original CD-ROM, five aerial survey data sets contained data below.
38°N
38°N
Dolphins absent
in the study area from Point Arena to Point Sal. Of these, the
OSPR survey program was still ongoing and data from recent For example, no adjustments or corrections have been made to
years were added to this data set. In addition, data from three account for differences in marine mammal detectability among
ship-based survey programs were converted to a compatible species and differential probability of detecting animals from
37°N
37°N
format for analysis. See "Data and Analyses" subsection in 2.3 aerial and shipboard platforms. Individual body size, group
for details on individual data sets. size, and species-specific behaviors, such as proportion of time
spent submerged, are all factors known to affect detection and
Data sources for aerial at-sea data include MMS-CDAS (MMS, hence, observed distribution and density estimates as well.
2001) and California Department of Fish and Game Office of Because of the very different attributes of aerial and shipboard
36°N
36°N
Spill Prevention and Response (CDF&G-OSPR), unpublished platforms, these factors, and the associated adjustments for
data. Early data were collected using methods described by observations, vary among the studies.
Bonnell et al. (1983) and Dohl et al. (1983); more recent data
were collected using updated technology but with the same Map d was developed using the same approach as for Maps a,
general method. Data sources for ship-based survey data b and c. For each season, the cells with densities in the top 20%
35°N
35°N
c d include David Ainley, unpublished data (see Oedekoven et al., of non-zero values were designated “high use” for that season.
2001 for details on methods). Although the at-sea data span Cells were scored for “high use” in one, two, or three seasons
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
the years 1980-2001, data are not available for all seasons in and are depicted by color. To provide a relative reference for
Source Data: See text.
all years. For the Upwelling Season, data are from 1980-1982 the “high use” areas, cells are also shown where the species
Figure 70. Maps for Pacific white-sided dolphin: seasonal at-sea densities and high use areas.
109
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
were absent (i.e., the cell was sampled but the species was not
recorded there) or present (but densities were never in the top
20% for any season). Further detail on methods is provided in
the "Data and Analysis" section.
RESULTS AND DISCUSSION
The Pacific white-sided dolphin is one of the most abundant
dolphin species of the temperate eastern North Pacific. In the
present study, it was the most abundant of the small cetaceans
(sightings: n=456; numbers of individuals: n=28,809). Pacific
white-sided dolphins occurred throughout the study area during
all oceanographic seasons in outer shelf, upper/lower slope
and canyon habitats.
Some seasonal shifts in the occurrence of Pacific white-sided
dolphins were observed in the data; densities were relatively
greater during the Oceanic Season, with concentrations near
Pioneer Canyon and Pioneer Seamount and regions over
Monterey Canyon. Because the occurrence of Pacific white-
sided dolphins is highly variable and this species responds to
oceanographic conditions on both seasonal and interannual
time scales (see Forney and Barlow, 1998), the apparent
seasonal shifts observed in these data may not be a seasonal
pattern.
However, in a study in Monterey Bay (Black, 1994), group size
and relative abundance of the Pacific white-sided dolphin varied
seasonally and was greater during the Oceanic and Davidson
Current Seasons than during the Upwelling Season, when
relative individual and group abundance was low and group
sizes were small (not shown in maps; Black, 1994).
Furthermore, in habitats over and near shelf-breaks and greater
bottom relief, feeding behavior was observed more than other
behaviors (Black, 1994). Based on available information, high
use areas mostly occurred over the slope.
Prey of the Pacific white-sided dolphin includes: Pacific whiting,
northern anchovy, rockfish, Pacific saury, and market squid
(Loligo opalescens) (Stroud et al., 1981; Black, 1994).
110
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THESE MAPS and 1985-2001. For the Oceanic Season, data are from 1980-
DRAFT
Risso's dolphin Figures 71a, b and c show the density (animals/km2) of Risso’s 1982, 1991 and 1994-2001. For the Davidson Current Season,
Grampus griseus dolphin in the Upwelling, Oceanic, and Davidson Current data are from 1980-1986 and 1991-2001.
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
seasons, displayed in 10’x10’ cells. Densities are based on
Upwelling Season Oceanic Season
200 m
200 m
20
20
combined data of several studies (see “Methods” and "Data METHODS
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) Sources" sections). The color and mapping intervals were At-sea densities are the result of a synthesis of data from eight
39°N
39°N
customized to show the most structure and highlight significant shipboard and aerial survey programs conducted in the study
Density areas, while allowing comparisons among marine mammal area in the years 1980-2001 (see “Data Sources” section).
(Animals/km²)
species. Cells that were surveyed but which had no Risso’s Cetacean observation data and trackline data from these
10.01 - 50.00 dolphins have a density of zero; unsurveyed areas appear white. studies were converted to a common format. All aerial data were
5.01 - 10.00
Blue lines indicate the National Marine Sanctuary boundaries continuous; ship-based data were converted separately into a
38°N
38°N
1.01 - 5.00
of Cordell Bank, Gulf of the Farallones, and Monterey Bay; continuous transect to the extent possible. From the digitized
0.51 - 1.00
bathymetric contours for the 200m and 2,000m isobaths are survey data, the distributions of effort and of species were
0.11 - 0.50
also shown in blue. mapped into 10’x10’ cells using CDAS, a custom geographic
0.06 - 0.10
information system for analyzing marine bird and mammal
0.01 - 0.05
37°N
37°N
In order to provide one map for the species that integrates the surveys (MMS, 2001). The length and width of the survey
0.00
patterns of its spatial and temporal occurrence in the study area, trackline in a given cell (estimated trackline width varied by
0 25 50 Km
map d shows seasonal high use areas, displayed in 10’x10 cells. platform, depending on speed and height above water) were
This map provides a further synthesis of densities presented used to estimate the area sampled. The number of cetaceans
in maps a, b and c (see “Methods” section for details), and of each species seen in a cell was then divided by the area
36°N
36°N
portrays the relative importance of various areas to the species. sampled in the cell to estimate density. If a cell was censused
Areas with consistent high use are highlighted on this map. To more than once, densities were averaged, with adjustment
provide a relative reference for the “high use” areas, cells are made for effort.
also shown where the species were absent (i.e., the cell was
sampled but the species was not recorded there) or present but Note that these maps represent either sighting locations or
35°N
35°N
a b at lesser concentrations in any particular season. densities that used survey strip widths relative to each survey
platform (e.g., plane, ship); density was calculated on the basis
200 m
Davidson Current Season
200 m
Seasonal High Use Areas DATA SOURCES of the number of animals sighted and area surveyed. The data
20
20
0
0
0m
0m
At-sea densities for cetaceans are based on data from eight have only been corrected to normalize for survey effort and to
(Nov. 15 - Mar. 14)
39°N
39°N
survey programs conducted in 1980-2001. These data were exclude observations with winds greater than 25 knots (smaller
Persistence of
High Use combined using CDAS software into the MMS-CDAS data or less obvious species are often less detectable even at wind
3 Seasons
system (MMS, 2001), developed for Minerals Management speeds of less than 25 knots). Additional corrections are
2 Seasons
Service and expanded for this project. Of the data sets on the planned for Phase 2 of this project and are briefly discussed
1 Season
Dolphins present original CD-ROM, five aerial survey data sets contained data below.
38°N
38°N
Dolphins absent
in the study area from Point Arena to Point Sal. Of these, the
OSPR survey program was still ongoing and data from recent For example, no adjustments or corrections have been made to
years were added to this data set. In addition, data from three account for differences in marine mammal detectability among
ship-based survey programs were converted to a compatible species and differential probability of detecting animals from
37°N
37°N
format for analysis. See "Data and Analyses" subsection in 2.3 aerial and shipboard platforms. Individual body size, group
for details on individual data sets. size, and species-specific behaviors, such as proportion of
time spent submerged, are all factors known to affect detection
Data sources for aerial at-sea data include MMS-CDAS (MMS,, and hence, observed distribution and density estimates as well.
2001) and California Department of Fish and Game Office of Because of the very different attributes of aerial and shipboard
36°N
36°N
Spill Prevention and Response (CDF&G-OSPR), unpublished platforms, these factors, and the associated adjustments for
data. Early data were collected using methods described by observations, vary among the studies.
Bonnell et al. (1983) and Dohl et al. (1983); more recent data
were collected using updated technology but with the same Map d was developed using the same approach as for maps a,
general method. Data sources for ship-based survey data b and c. For each season, the cells with densities in the top 20%
35°N
35°N
c d include David Ainley, unpublished data (see Oedekoven et al. of non-zero values were designated “high use” for that season.
2001 for details on methods). Although the at-sea data span Cells were scored for “high use” in one, two, or three seasons
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
the years 1980-2001, data are not available for all seasons in and are depicted by color. To provide a relative reference for
Source Data: See text.
all years. For the Upwelling Season, data are from 1980-1982 the “high use” areas, cells are also shown where the species
Figure 71. Maps for Risso’s dolphin: seasonal at-sea densities and high use areas.
111
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
were absent (i.e., the cell was sampled but the species was not
recorded there) or present (but densities were never in the top
20% for any season). Further detail on methods is provided in
the "Data and Analysis" section.
RESULTS AND DISCUSSION
The Risso’s dolphin is widely distributed in warm-temperate
waters from southern California north to Washington, and in
the study area, occurred over outer shelf, upper and lower
slope, and canyon habitats, and in offshore waters seaward
of the 2000 m isobath. Risso’s dolphin was the third most
abundant dolphin in the study area, with 250 sightings of
2,248 individuals.
During the Upwelling Season, Risso’s dolphins were distributed
throughout the study area over the outer shelf, slope and deep
ocean, with greatest densities in (and to the south and west
of) the Monterey Bay National Marine Sanctuary (MBNMS).
During the Oceanic Season, greatest densities occurred within
and south and west of the southern portion of MBNMS. During
the Davidson Current Season, overall densities were mostly
in the southern portion of the MBNMS and areas to the south
and west of the MBNMS boundary.
Distribution of Risso’s dolphin off California, Oregon, and
Washington is highly variable, apparently in response to
seasonal and interannual oceanographic changes (Forney
and Barlow, 1998). Dolphins found off California during colder
water months are thought to shift northward into Oregon and
Washington as water temperatures increase in late spring and
summer (Carretta et al., 2001; Green et al., 1992). Given the
highly variable distribution of Risso’s dolphin, the apparent
relative decrease in relative density observed in this study
during the Davidson Current Season may not be a seasonal
pattern. Based on this data set, most high use areas occurred
in the Monterey Bay national marine sanctuary and adjacent
areas to the south (see map).
Risso’s dolphin feed almost exclusively on squid (Koski et al.,
1998; Orr, 1966).
112
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
DRAFT
ABOUT THESE MAPS 1985-2001. Oceanic Season, data are from 1980-1982, 1991
Northern right-whale dolphin Figures 72a, b and c show the density (animals/km2) of northern
Lissodelphis borealis and 1994-2001. Davidson Current Season, data are from 1980-
right whale dolphins in the Upwelling, Oceanic, and Davidson 1986 and 1991-2001.
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Current seasons, displayed in 10’x10’ cells. Densities are based
Upwelling Season Oceanic Season
200 m
200 m
20
20
on combined data of several studies (see “Methods” and “Data METHODS
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14)
39°N
39°N
Sources”). The color and mapping intervals were customized to At-sea densities are the result of a synthesis of data from eight
show the most structure and highlight significant areas, while shipboard and aerial survey programs conducted in the study
Density
allowing comparisons among marine mammal species. Cells area in the years 1980-2001 (see “Data Sources” section).
(Animals/km²)
that were surveyed but which had no northern right whale Cetacean observation data and trackline data from these
10.01 - 50.00
dolphins have a density of zero; unsurveyed areas appear white. studies were converted to a common format. All aerial data were
5.01 - 10.00
38°N
38°N
Blue lines indicate the National Marine Sanctuary boundaries continuous; ship-based data were converted separately into a
1.01 - 5.00
of Cordell Bank, Gulf of the Farallones, and Monterey Bay; continuous transect to the extent possible. From the digitized
0.51 - 1.00
bathymetric contours for the 200m and 2,000m isobaths are survey data, the distributions of effort and of species were
0.11 - 0.50
also shown in blue. mapped into 10’x10’ cells using CDAS, a custom geographic
0.06 - 0.10
information system for analyzing marine bird and mammal
0.01 - 0.05
37°N
37°N
In order to provide one map for the species that integrates the surveys (MMS, 2001). The length and width of the survey
0.00
patterns of its spatial and temporal occurrence in the study trackline in a given cell (estimated trackline width varied by
0 25 50 Km
area, map d shows seasonal high use areas, displayed in platform, depending on speed and height above water) were
10’x10’ cells. This map provides a further synthesis of densities used to estimate the area sampled. The number of cetaceans
presented in maps a, b and c (see “Methods” section for details), of each species seen in a cell was then divided by the area
36°N
36°N
and portrays the relative importance of various areas to the sampled in the cell to estimate density. If a cell was censused
species. Areas with consistent high use are highlighted on this more than once, densities were averaged, with adjustment
map. To provide a relative reference for the “high use” areas, made for effort.
cells are also shown where the species were absent (i.e., the
35°N
35°N
cell was sampled but the species was not recorded there), or Note that these maps represent either sighting locations or
a b present but at lesser concentrations in any particular season. densities that used survey strip widths relative to each survey
platform (e.g., plane, ship); density was calculated on the basis
200 m
Davidson Current Season
200 m
Seasonal High Use Areas
20
20
DATA SOURCES of the number of animals sighted and area surveyed. The data
0
0
0m
0m
(Nov. 15 - Mar. 14) At-sea densities for cetaceans are based on data from eight have only been corrected to normalize for survey effort and to
39°N
39°N
Persistence of survey programs conducted in 1980-2001. These data were exclude observations with winds greater than 25 knots (smaller
High Use
combined using CDAS software into the MMS-CDAS data or less obvious species are often less detectable even at wind
3 Seasons
system (MMS, 2001), developed for Minerals Management speeds of less than 25 knots). Additional corrections are
2 Seasons
1 Season
Service and expanded for this project. Of the data sets on the planned for Phase 2 of this project and are briefly discussed
Dolphins present
38°N
38°N
original CD-ROM, five aerial survey data sets contained data below.
Dolphins absent
in the study area from Point Arena to Point Sal. Of these, the
OSPR survey program was still ongoing and data from recent For example, no adjustments or corrections have been made to
years were added to this data set. In addition, data from three account for differences in marine mammal detectability among
ship-based survey programs were converted to a compatible species and differential probability of detecting animals from
37°N
37°N
format for analysis. See section introduction for details on aerial and shipboard platforms. Individual body size, group
individual data sets. size, and species-specific behaviors, such as proportion of time
spent submerged, are all factors known to affect detection and
Data sources for aerial at-sea data include MMS-CDAS (MMS, hence, observed distribution and density estimates as well.
2001) and California Department of Fish and Game Office of Because of the very different attributes of aerial and shipboard
36°N
36°N
Spill Prevention and Response (CDF&G-OSPR), unpublished platforms, these factors, and the associated adjustments for
data. Early data were collected using methods described by observations, vary among the studies.
Bonnell et al. (1983) and Dohl et al. (1983); more recent data
were collected using updated technology but with the same Map d was developed using the same approach as for maps a,
general method. Data sources for ship-based survey data
35°N
35°N
b and c. For each season, the cells with densities in the top 20%
c d include David Ainley, unpublished data (see Oedekoven et al., of non-zero values were designated “high use” for that season.
2001 for details on methods). Although the at-sea data span Cells were scored for “high use” in one, two, or three seasons
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Source Data: See text. the years 1980-2001, data are not available for all seasons and are depicted by color. To provide a relative reference for
in all years. Upwelling Season, data are from 1980-1982 and the “high use” areas, cells are also shown where the species
Figure 72. Maps for northern right-whale dolphin: seasonal at-sea densities and high use areas.
113
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
were absent (i.e., the cell was sampled but the species was not
recorded there) or present (but densities were never in the top
20% for any season). Further detail on methods is provided in
the "Data and Analysis" section.
RESULTS AND DISCUSSION
The northern right whale dolphin occurs in the temperate
North Pacific, primarily in shelf, slope, and to some degree,
deep ocean waters. In the study area, this species occurred
in outer shelf, slope and canyon habitats. The northern right
whale dolphin was the second most abundant small cetacean
in the study area, with 135 sightings of 22,578 individuals.
Distribution of northern right whale dolphins is highly
variable, apparently in response to seasonal and interannual
oceanographic changes (Forney and Barlow, 1998). Northern
right whale dolphins are found primarily off California during
colder-water months and shift northward into Oregon and
Washington as water temperatures increase in late spring
and summer (Carretta et al., 2001; Forney and Barlow,
1998). Patterns of seasonal abundance have been observed
throughout their range, but there is no information to indicate
that large numbers move between California, Oregon, and
Washington waters (Green et al., 1992). In this study, the
apparent increase in relative densities in the southern portion
of MBNMS during the Davidson Current Season may not
be a seasonal pattern, given the highly variable distribution
of northern right whale dolphins, apparently in response to
seasonal and interannual oceanographic changes (Forney
and Barlow, 1998).
Northern right whale dolphins feed on mesopelagic fishes (e.g.
lanternfish) and squid (Leatherwood and Reeves, 1983).
114
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THIS MAP the extent possible. From the digitized survey data, effort was
124°W 123°W 122°W 121°W
Figure 73 shows the individual sightings of blue whales at sea, mapped into 10’x10’ cells using CDAS, a custom geographic
200 m
20
39°N
39°N
Blue whale Balaenoptera musculus
0
0m
along with at-sea survey effort. Due to insufficient sightings in information system for analyzing marine bird and mammal
the data set (49 sightings of 77 individuals) for the study area, surveys (MMS, 2001). The length and width of the survey
seasonal maps of blue whale density were not generated. trackline in a given cell (estimated trackline width varied by
At-sea sightings for cetaceans are from several studies (see platform, depending on speed and height above water) were
“Methods” and “Data Sources” sections). For context, the com- used to estimate the area sampled.
bined survey effort is also shown, summarized in 10’x10’ cells.
Sightings Blue lines indicate the National Marine Sanctuary boundaries Note that the these maps represent either sighting locations or
(Number of whales) of Cordell Bank, Gulf of the Farallones, and Monterey Bay; densities that used survey strip widths relative to each survey
5 bathymetric contours for the 200m and 2,000m isobaths are platform (e.g., plane, ship); density was calculated on the basis
also shown in blue. Additional data to be added in Phase II may of the number of animals sighted and area surveyed. The data
3
38°N
38°N
make it possible to develop seasonal maps. have only been corrected to normalize for survey effort and
2
to exclude observations with winds greater than 25 knots;
1 DATA SOURCES additional corrections are planned for Phase 2 of this project
At-sea sightings for cetaceans are based on data from eight and are briefly discussed below.
Survey Effort
survey programs conducted in 1980-2001. These data were
(km of trackline)
combined using CDAS software into the MMS-CDAS data For example, no adjustments/corrections have been made to
> 3000.00
system (MMS, 2001), developed for Minerals Management account for differences in marine mammal detectability among
1500.01 - 3000.00
Service and expanded for this project. Of the data sets on the species and differential probability of detecting animals from
1000.01 - 1500.00
original CD-ROM, five aerial survey data sets contained data aerial and shipboard platforms. Individual body size, group
500.01 - 1000.00
in the study area from Point Arena to Point Sal. Of these, the size, and species-specific behaviors, such as proportion of time
100.01 - 500.00
37°N
37°N
OSPR survey program was still ongoing and data from recent spent submerged, are all factors known to affect detection and
0.01 - 100.00
years were added to this data set. In addition, data from three hence, observed distribution and density estimates as well.
ship-based survey programs were converted to a compatible Because of the very different attributes of aerial and shipboard
0 25 50 Km
format for analysis. See section introduction for details on platforms, these factors, and the associated adjustments for
individual data sets. observations, vary among the studies.
Data sources for aerial at-sea data include MMS-CDAS (MMS, The data in these maps include wind conditions of up to 25
2001) and California Department of Fish and Game Office of knots; smaller or less obvious species are often less detectable
Spill Prevention and Response (CDFG-OSPR), unpublished even at wind speeds of less than 25 knots. The seasonal maps
data. Early data were collected using methods described by contain different combinations of shipboard and aerial data;
36°N
36°N
Bonnell et al. (1983) and Dohl et al. (1983); more recent data therefore the seasonal densities from these platforms may not
were collected using updated technology but with the same be directly comparable. A full consideration of these factors, and
general method. Data sources for ship-based survey data in- revised maps, are planned for Phase 2 of this project.
clude David Ainley, unpublished data (see Oedekoven et al.,
2001 for details on methods). Although the at-sea data span RESULTS AND DISCUSSION
the years 1980-2001, data are not available for all seasons in The blue whale is federally listed as endangered under the
all years. For the Upwelling Season, data are from 1980-1982 Endangered Species Act. One population of blue whale (there
and 1985-2001. For the Oceanic Season, data are from 1980- may be as many as five (Carretta et al., 2001; Reeves et al.,
1982, 1991 and 1994-2001. For the Davidson Current Season, 1998)) is present in California waters, generally from June
data are from 1980-1986 and 1991-2001. through November. Arrival and departure times in the study
35°N
35°N
area are highly variable both seasonally and inter-annually (see
METHODS Benson et al., 2002; Calambokidis et al., 1998).
The latitiude and longitude coordinates for blue whales at sea
DRAFT
were used to plot the individual sightings. At-sea sightings and Movement patterns, distribution, and occurrence of blue whales
effort are the result of a synthesis of data from eight shipboard off California are related to their annual migration between
and aerial survey programs conducted in the study area in the foraging areas predominately off central California (but some
years 1980-2001 (see “Data Sources”). Cetacean observation north to British Columbia and south to Baja Mexico and the
data and trackline data from these studies were converted to
124°W 123°W 122°W 121°W
Costa Rican Dome), and the following breeding areas: 1) off
Source Data: See text. a common format. All aerial data were continuous; ship-based the west coast of Baja California (September-December), 2)
data were converted separately into a continuous transect to
Figure 73. Map for blue whale: at-sea sightings and survey effort. the Gulf of California (January-April), and 3) the Costa Rica
115
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Dome (Mate et al., 1999). And although blue whales are often
present in parts of the National Marine Sanctuary waters from
June through November, their occurrence and distribution
during this feeding period is highly variable. Due to insufficient
sightings in the data set (49 sightings of 77 individuals) for
the study area, seasonal maps of blue whale density were
not generated. Additional sighting data for blue whale will be
integrated in Phase 2, and seasonal maps may be generated
at that time.
Blue whales feed on seasonally abundant and dense euphausiid
(krill) schools in discrete depths in the water column (Benson
et al., 2002), concentrated in the deep scattering layer along
canyon and shelf-break edges, and in daytime surface swarms
of krill (Schoenherr, 1991; Croll et al., 1998; Forney and Barlow,
1998). Spatially, they were widely distributed in shelf-break
and slope habitats, as well as seaward of National Marine
Sanctuary boundaries, and to a lesser extent, over the shelf.
Although not directly shown on this map, blue whales also occur
in the Cordell Bank National Marine Sanctuary and off Bodega
Bay (Calambokidis et al., 1990b; Calambokidis et al., 1998),
as well as in waters around the Farallon Islands (C.Keiper,
pers.comm).
There is considerable interchange and interregional movements
between Blue whales that occur off southern California (from
the Santa Barbara Channel and Southern California Bight) to
areas in the Monterey Bay, Gulf of the Farallones, Bodega Bay,
and northern California (Calambokidis et al., 1998). In a study
of the Monterey Bay area (Benson et al., 2002), occurrence
of Blue whales in Monterey Bay was related to seasonal
upwelling patterns that affect seasonally abundant, dense (and
ephemeral) patches of euphausiids that occur during summer
and fall (Benson et al., 2002). See map of SWFSC survey data
for blue whale (Figure 77) for greater geographic extent of the
range and interannual variations for this species.
116
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
ABOUT THESE MAPS and 1985-2001. For the Oceanic Season, data are from 1980-
DRAFT
Humpback whale Figures 74a, b and c show the density (animals/km2) of 1982, 1991 and 1994-2001. For the Davidson Current Season,
Megaptera novaeangliae humpback whales in the Upwelling, Oceanic, and Davidson data are from 1980-1986 and 1991-2001.
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
Current seasons, displayed in 10’x10’ cells. Densities are based
Upwelling Season Oceanic Season
200 m
200 m
20
20
on combined data of several studies (see “Methods” and “Data METHODS
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) Sources”). The color and mapping intervals were customized to At-sea densities are the result of a synthesis of data from eight
39°N
39°N
show the most structure and highlight significant areas, while shipboard and aerial survey programs conducted in the study
Density
allowing comparisons among marine mammal species. Cells area in the years 1980-2001 (see “Data Sources”). Cetacean
(Animals/km²)
that were surveyed but which had no humpback whales have observation data and trackline data from these studies were
10.01 - 50.00
a density of zero; unsurveyed areas appear white. Blue lines converted to a common format. All aerial data were continuous;
5.01 - 10.00
38°N
38°N
indicate the National Marine Sanctuary boundaries of Cordell ship-based data were converted separately into a continuous
1.01 - 5.00
Bank, Gulf of the Farallones, and Monterey Bay; bathymetric transect to the extent possible. From the digitized survey data,
0.51 - 1.00
contours for the 200m and 2,000m isobaths are also shown the distributions of effort and of species were mapped into
0.11 - 0.50
in blue. 10’x10’ cells using CDAS, a custom geographic information
0.06 - 0.10
system for analyzing marine bird and mammal surveys (MMS,
0.01 - 0.05
37°N
37°N
In order to provide one map for the species that integrates the 2001). The length and width of the survey trackline in a given
0.00
patterns of its spatial and temporal occurrence in the study cell (estimated trackline width varied by platform, depending on
0 25 50 Km
area, map d shows seasonal high use areas, displayed in speed and height above water) were used to estimate the area
10’x10' cells. This map provides a further synthesis of densities sampled. The number of cetaceans of each species seen in a
presented in maps a, b and c (see “Methods” for details), and cell was then divided by the area sampled in the cell to estimate
36°N
36°N
portrays the relative importance of various areas to the species. density. If a cell was censused more than once, densities were
Areas with consistent high use are highlighted on this map. To averaged, with adjustment made for effort.
provide a relative reference for the “high use” areas, cells are
also shown where the species were absent (i.e., the cell was Note that these maps represent either sighting locations or
sampled but the species was not recorded there), or present densities that used survey strip widths relative to each survey
35°N
35°N
a b but at lesser concentrations in any particular season. platform (e.g., plane, ship); density was calculated on the basis
of the number of animals sighted and area surveyed. The data
200 m
Davidson Current Season
200 m
Seasonal High Use Areas
20
20
DATA SOURCES have only been corrected to normalize for survey effort and to
0
0
0m
0m
(Nov. 15 - Mar. 14) At-sea densities for cetaceans are based on data from eight exclude observations with winds greater than 25 knots (smaller
39°N
39°N
survey programs conducted in 1980-2001. These data were or less obvious species are often less detectable even at wind
Persistence of
High Use combined using CDAS software into the MMS-CDAS data speeds of less than 25 knots). Additional corrections are
3 Seasons
system (MMS, 2001), developed for Minerals Management planned for Phase 2 of this project and are briefly discussed
2 Seasons
Service and expanded for this project. Of the data sets on the below.
1 Season
Whales present
original CD-ROM, five aerial survey data sets contained data
38°N
38°N
Whales absent
in the study area from Point Arena to Point Sal. Of these, the For example, no adjustments or corrections have been made to
OSPR survey program was still ongoing and data from recent account for differences in marine mammal detectability among
years were added to this data set. In addition, data from three species and differential probability of detecting animals from
ship-based survey programs were converted to a compatible aerial and shipboard platforms. Individual body size, group
37°N
37°N
format for analysis. See section introduction for details on size, and species-specific behaviors, such as proportion of time
individual data sets. spent submerged, are all factors known to affect detection and
hence, observed distribution and density estimates as well.
Data sources for aerial at-sea data include MMS-CDAS (MMS, Because of the very different attributes of aerial and shipboard
2001) and California Department of Fish and Game Office of platforms, these factors, and the associated adjustments for
36°N
36°N
Spill Prevention and Response (CDF&G-OSPR), unpublished observations, vary among the studies. Additional data, mapping
data. Early data were collected using methods described by and analysis in Phase 2 may provide more definitive spatial
Bonnell et al. (1983) and Dohl et al. (1983); more recent data patterns for this species.
were collected using updated technology but with the same
general method. Data sources for ship-based survey data Map d was developed using the same approach as for maps a,
35°N
35°N
c d include David Ainley, unpublished data (see Oedekoven et al., b and c . For each season, the cells with densities in the top 20%
2001 for details on methods). Although the at-sea data span of non-zero values were designated “high use” for that season.
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
the years 1980-2001, data are not available for all seasons in Cells were scored for “high use” in one, two, or three seasons
Source Data: See text.
all years. For the Upwelling Season, data are from 1980-1982 and are depicted by color. To provide a relative reference for
Figure 74. Maps for humpback whale: seasonal at-sea densities and high use areas.
117
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
the “high use” areas, cells are also shown where the species Season, most humpback whales are in breeding/calving areas,
were absent (i.e., the cell was sampled but the species was not hence the relatively few sightings in the study area (1980, 1982,
recorded there) or present (but densities were never in the top and 1993) during this season.
20% for any season). Further detail on methods is provided in
the "Data and Analysis" section. The NOAA/SWFSC stock assessment sightings maps
(Figure 78) indicate humpback whales occurred off northern
RESULTS AND DISCUSSION California, and south to Point Conception, with sightings in
The humpback whale is federally listed as endangered under the CBNMS, GFNMS, and MBNMS during 1993, 1996, and
the Endangered Species Act. The eastern North Pacific 2001. During the 1996 and 2001 surveys (when effort extended
stock of the humpback whale that occurs in the study area, north to Washington), humpback whales were also sighted
feeds off California, Oregon, and Washington and migrates off Washington and Oregon. Based on CDAS data shown in
from its breeding and calving areas off coastal Mexico and these maps, most high use areas occurred over the shelf and
Central America (Calambokidis et al., 2000). In this study, the slope.
humpback whale was the most numerous pelagic baleen whale
sighted and was primarily distributed over the shelf, upper slope Humpback whales feed on seasonally abundant, small
and some lower slope habitats. Humpback whales are sighted schooling fishes (e.g. northern anchovy, Pacific sardine,
from the Farallon Islands in all months (Pyle and Gilbert, 1996), Pacific herring) and euphausiids (primarily T. spinifera and E.
though they are more frequently sighted off central California Pacifica; Kieckhefer, 1992). See map of SWFSC survey data
from March through November, with peaks in the summer for humpback whale (Figure 78) for additional geographic extent
and fall (Calambokidis et al., 1996), a pattern reflected in the of the humpback whale range and interannual variations for
seasonal distribution maps. this species.
During the Upwelling Season, humpback whales mostly
occurred in the shelf and slope areas of, and adjacent to,
the Gulf of the Farallones (GFNMS) and the northern part
of Monterey Bay National Marine Sanctuary (MBNMS); see
map for other areas. During the Oceanic Season, the CDAS
map shows the Humpback whales more concentrated in
the areas of the GFNMS, the Cordell Bank National Marine
sanctuary (CBNMS), the northwest corner of the MBNMS, and
the adjacent slope area; the SWFSC Humpback whale map
(Figure 77) shows concentrations over the shelf and slope
throughout the study area extent. Densities and sightings for
the Davidson season were lowest, but like the other seasons,
most occurrences were over the shelf and slope.
A major food type for humpback whales is euphausiids (krill).
The Upwelling Season and beginning of the Oceanic Season
is characterized by a seasonal peak in euphausiid density
that occurs in July/August but can extend into the Oceanic
Season (8/15-11/14). Krill abundance increases one to four
months after seasonal peaks in primary production (Croll et
al. 1998). One of the dominant species of krill (Thysanoessa
spinifera) forms dense shoals in the shelf region from Fort
Ross south to the Channel Islands (Kieckhefer 1992). Primary
feeding sites of humpback whales are located at Monterey
Bay (Benson et al., 2002), Bodega Canyon, Cordell Bank, and
the Farallon Islands (Kieckhefer, 1992). There is considerable
interchange and inter-regional movement of humpback whales
within a feeding season between the Santa Barbara Channel,
Monterey Bay, and to the north off Eureka (Calambokidis et al.,
1996, Calambokidis et al., 1998). During the Davidson Current
118
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
For the Oceanic Season, data are from 1980-1982, 1991 and
ABOUT THESE MAPS
DRAFT
Gray whale 1994-2001. For the Davidson Current Season, data are from
Figures 75a, b and c show the density (animals/km2) of gray
Eschrichtius robustus 1980-1986 and 1991-2001.
whales in the Upwelling, Oceanic, and Davidson Current
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
seasons, displayed in 10’x10’ cells. Densities are based on
Upwelling Season Oceanic Season
200 m
200 m
20
20
METHODS
combined data of several studies (see “Methods” and “Data
0
0
0m
0m
(Mar. 15 - Aug. 14) (Aug. 15 - Nov. 14) At-sea densities are the result of a synthesis of data from eight
Sources”). The color and mapping intervals were customized
39°N
39°N
shipboard and aerial survey programs conducted in the study
to show the most structure and highlight significant areas,
Density area in the years 1980-2001 (see “Data Sources”). Cetacean
while allowing comparisons among marine mammal species.
(Animals/km²)
observation data and trackline data from these studies were
Cells that were surveyed but which had no gray whales have
10.01 - 50.00
converted to a common format. All aerial data were continuous;
a density of zero; unsurveyed areas appear white. Blue lines
5.01 - 10.00
ship-based data were converted separately into a continuous
38°N
38°N
indicate the National Marine Sanctuary boundaries of Cordell
1.01 - 5.00
transect to the extent possible. From the digitized survey data,
Bank, Gulf of the Farallones, and Monterey Bay; bathymetric
0.51 - 1.00
the distributions of effort and of species were mapped into
contours for the 200m and 2,000m isobaths are also shown
0.11 - 0.50
10’x10’ cells using CDAS, a custom geographic information
in blue.
0.06 - 0.10
system for analyzing marine bird and mammal surveys (MMS,
0.01 - 0.05
37°N
37°N
2001). The length and width of the survey trackline in a given
In order to provide one map for the species that integrates the
0.00
cell (estimated trackline width varied by platform, depending on
patterns of its spatial and temporal occurrence in the study
0 25 50 Km
speed and height above water) were used to estimate the area
area, map d shows seasonal high use areas, displayed in
sampled. The number of cetaceans of each species seen in a
10’x10’ cells. This map provides a further synthesis of densities
cell was then divided by the area sampled in the cell to estimate
presented in maps a, b and c (see “Methods” section for details),
36°N
36°N
density. If a cell was censused more than once, densities were
and portrays the relative importance of various areas to the
averaged, with adjustment made for effort.
species. Areas with consistent high use are highlighted on this
map. To provide a relative reference for the “high use” areas,
Note that these maps represent either sighting locations or
cells are also shown where the species were absent (i.e., the
densities that used survey strip widths relative to each survey
cell was sampled but the species was not recorded there), or
35°N
35°N
a b platform (e.g., plane, ship); density was calculated on the basis
present but at lesser concentrations in any particular season.
of the number of animals sighted and area surveyed. The data
200 m
Davidson Current Season
200 m
Seasonal High Use Areas
20
have only been corrected to normalize for survey effort and to
DATA SOURCES
20
0
0
0m
0m
exclude observations with winds greater than 25 knots (smaller
At-sea densities for cetaceans are based on data from eight
(Nov. 15 - Mar. 14)
39°N
39°N
or less obvious species are often less detectable even at wind
survey programs conducted in 1980-2001. These data were
Persistence of
High Use speeds of less than 25 knots). Additional corrections are
combined using CDAS software into the MMS-CDAS data
3 Seasons
planned for Phase 2 of this project and are briefly discussed
system (MMS, 2001), developed for Minerals Management
2 Seasons
below.
Service and expanded for this project. Of the data sets on the
1 Season
Whales present original CD-ROM, five aerial survey data sets contained data
38°N
38°N
Whales absent
For example, no adjustments or corrections have been made to
in the study area from Point Arena to Point Sal. Of these, the
account for differences in marine mammal detectability among
OSPR survey program was still ongoing and data from recent
species and differential probability of detecting animals from
years were added to this data set. In addition, data from three
aerial and shipboard platforms. Individual body size, group
ship-based survey programs were converted to a compatible
37°N
37°N
size, and species-specific behaviors, such as proportion of time
format for analysis. See "Data and Analyses" subsection in 2.3
spent submerged, are all factors known to affect detection and
for details on individual data sets.
hence, observed distribution and density estimates as well.
Because of the very different attributes of aerial and shipboard
Data sources for aerial at-sea data include MMS-CDAS (2001)
platforms, these factors, and the associated adjustments for
and California Department of Fish and Game Office of Spill
36°N
36°N
observations, vary among the studies.
Prevention and Response (CDF&G-OSPR), unpublished data.
Early data were collected using methods described by Bonnell
Map d was developed using the same approach as for maps a,
et al. (1983) and Dohl et al. (1983); more recent data were
b and c. For each season, the cells with densities in the top 20%
collected using updated technology but with the same general
of non-zero values were designated “high use” for that season.
method. Data sources for ship-based survey data include David
35°N
35°N
c d Cells were scored for “high use” in one, two, or three seasons
Ainley, unpublished data (see Oedekoven et al., 2001 for details
and are depicted by color. To provide a relative reference for
on methods). Although the at-sea data span the years 1980-
124°W 123°W 122°W 121°W 124°W 123°W 122°W 121°W
the “high use” areas, cells are also shown where the species
2001, data are not available for all seasons in all years. For the
Source Data: See text.
was absent (i.e., the cell was sampled but the species was not
Upwelling Season, data are from 1980-1982 and 1985-2001.
Figure 75. Maps for gray whale: seasonal at-sea densities and high use areas.
119
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Significant declines in calf counts also occurred during this
recorded there) or present (but densities were never in the top
same period (Perryman et al., 2002).
20% for any season). Further detail on methods is provided in
the "Data and Analysis" section.
The apparent seasonal high-use areas noted in map d (the
Farallon Islands, off Año Nuevo, and north of Cordell Bank near
RESULTS AND DISCUSSION
Point Arena and Fort Ross) are likely related to the timing of the
The eastern population of the gray whale migrates from
migration and may not represent a discrete spatial pattern.
summer feeding grounds in the Bering, Chukchi, and western
Beaufort Seas, south along the west coast of North America
Gray whales feed on benthic invertebrates (e.g. gammarid
to its winter breeding and calving areas off the coast of Baja
amphipods; Leatherwood and Reeves, 1983), mysid shrimp,
California. The southward migration includes (in the order of sex
herring eggs/larvae, crab larvae, ghost shrimp (Darling et al.,
and age-class) females in late pregnancy, females that have
1998), and surface swarms of euphausiids (Benson et al.,
recently ovulated, adult males, immature females, and lastly,
2002). Although most individuals of gray whales in the study
immature males. In the study area, this southward migration
area were non-feeding migrants, some individuals do feed on
generally occurs from December through February and peaks
a regular basis near the South Farallon Islands, at the mouth
in January. The northward migration generally occurs from
of Tomales Bay and Drakes Bay (S. Allen, 2002, pers. comm.),
February through May and peaks in March and includes (in
during their northward migration. Gray whales also feed in San
the order of reproductive condition, sex, age-class,) newly
Francisco Bay (in some years: 1999, 2000,-2001; Oliver et al.,
pregnant females, adult males, immature females, and last
2001)) and Monterey Bay (Benson et al., 2002). Gray whales
in this migration, the females with calves. The latter migrate
also have been seen regularly off Point Reyes, Tomales Bay,
northward through the study area during April and May, and
Drakes Bay, and the Farallon Islands during non-migratory
sometimes June. The northward migration is reflected in the
periods (S. Allen, pers. comm.).
distribution patterns during the Upwelling Season, when gray
whales are distributed in the coastal and inner/outer shelf
habitats throughout the study area, en route to their northern
feeding grounds, a pattern reflected in their virtual absence
(according to the data set) in the study area during the Oceanic
Season.
In the study area and data sets analyszed, the gray whale was
the second most numerous baleen whale. Concentrations of
this species were greatest during the Davidson Current Season,
a period that encompasses both the southward and northward
migration, with greatest concentrations observed along the
coast near Cypress Point and south of Point Sur to Lopez
Point. Relative densities were somewhat greater to the north
of CBNMS and to the south of MBNMS (likely related to the
timing of individuals moving north or south). Recent preliminary
documentation of the southbound migration during 2000 and
2001 indicated population estimates of 17,414 (CV=10%), well
below previous (1997/98) estimates of 26,635 (CV=10%; Rugh
et al., 2002). These low estimates may have been caused by
an unusual number of whales that did not migrate as far south
as Granite Canyon (the survey location), or abundance may
have declined following the high mortality rates observed in
1999 and 2000 (Rugh et al., 2002).
Strandings along the coast of North America were six times
more prevalent than during 1995-1998 (Gulland et al., 2001).
Factors that may have contributed to the high number of
strandings include: starvation, anthropogenic and natural
toxins, infectious diseases, ship strikes, detection effort and
reporting, and wind and current effects (Gulland et al., 2001).
120
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Introduction to the SWFSC Data Set, to be used in Phase in the CDAS maps). A visual comparison of the SWFSC maps
Dall's porpoise Phocoenoides dalli
II of this Analysis. The following marine mammal maps are among years (1991, 1993, 1996, 2001) indicates occurrence
based on data from NOAA’s marine mammal stock assessment patterns of blue whales varied; relatively greater concentrations 120°W 118°W 118°W
132°W 130°W 128°W 126°W 124°W 122°W 132°W 130°W 128°W 126°W 124°W 122°W 120°W
program, conducted by NOAA’s Southwest Fisheries Science of blue whales off southern California were evident in 1991,
48°N
48°N
1993
1991
Center. Maps for three species are included below (and 13 more 1993, and 1996, however, this species was virtually absent
are on the CD-ROM) to provide the reader with an idea of some (except for a few sightings) in this region during the survey of Average Average
of the additional data that will be incorporated into the overall 2001. Sightings occurred within NMS boundaries (when effort
46°N
46°N
Group Size Group Size
mammal data set and analysis in Phase II (Figures 76-78). extended into these areas) off Point Reyes in 1991 and 1996,
5.1 - 27.0 5.1 - 27.0
within the Gulf of the Farallones National Marine Sanctuary in
44°N
44°N
3.1 - 5.0 3.1 - 5.0
These maps show sightings (species group size), and effort 1993, and within Monterey Bay in 1996. See "Map Text" and
2.1 - 3.0
locations generally for the late summer/fall season (data ranged "Discussion" in CDAS map section for additional information 2.1 - 3.0
42°N
42°N
1.6 - 2.0
1.6 - 2.0
from July-December) for four years: 1991, 1993, 1996 and on blue whales.
1.0 - 1.5
1.0 - 1.5
2001, off the coasts of California, Oregon and Washington.
40°N
40°N
These maps are the results of broad-scale, ocean ship About the SWFSC Humpback Whale Maps (Figure 78). These Effort
Effort
surveys (aerial surveys are not included), and are used in the SWFSC maps of surveys conducted in July through early
38°N
38°N
development of stock estimates and trend analyses for most December 1991, 1993, 1996, and 2001, (late upwelling and
marine mammals that occur off the coasts of California, Oregon Oceanic season) encompass a much larger geographic extent
36°N
36°N
and Washington. These SWFSC maps do not represent the and indicate concentrations of humpback whales in central
distribution of the species, but they do provide an indication California relatively closer to shore. (See SWFSC blue whale
of the broader spatial extent of the species during the late maps for comparison) and distributed off northern California
34°N
34°N
summer/fall season. and south to Point Conception. A visual comparison of the
SWFSC maps among years (1991, 1993, 1996, 2001) indicates
32°N
32°N
For more information on the marine mammal stock assessment occurrence patterns of humpback whales varied; relatively
survey data, visit: http://swfsc.nmfs.noaa.gov/PRD/CMMP/ or greater concentrations occurred off central California in the
30°N
30°N
contact Dr. Jay Barlow at Jay.Barlow@noaa,gov. survey of 1996, compared to the survey of 1991 (when survey
effort was similar off central California). Sightings within the
About the SWFSC Dall’s Porpoise Maps (Figure 76). These CBNMS, GFNMS, and MBNMS occurred during 1993, 1996,
48°N
48°N
2001
1996
SWFSC maps of surveys conducted in July through early and 2001 (when effort extended into these areas). During
Average Average
December 1991, 1993, 1996, and 2001, (late upwelling and the surveys of 1996 and 2001 (when effort extended north to
46°N
46°N
Group Size Group Size
Oceanic season) encompass a much larger geographic Washington and Oregon), humpback whales also were sighted
extent and provide an example of the off-shore and northern off Washington and Oregon. See "Map Text" and "Discussion" 5.1 - 27.0 5.1 - 27.0
44°N
44°N
geographic extent of the Dall’s porpoise. A visual comparison in CDAS map section for additional information on Humpback 3.1 - 5.0
3.1 - 5.0
of the SWFSC maps among years (1991, 1993, 1996, 2001) whales. 2.1 - 3.0
2.1 - 3.0
42°N
42°N
indicates occurrence patterns of Dall’s porpoise varied; number 1.6 - 2.0
1.6 - 2.0
of sightings was relatively greater off northern California than 1.0 - 1.5
1.0 - 1.5
40°N
40°N
off central California in 1991, (when survey effort was only off
Effort
Effort
California). In the survey of 1996, (when survey effort extended
north to Oregon and Washington), number of sightings and
38°N
38°N
average group size was relatively greater off Oregon and
northern California than off central California. Sightings that
36°N
36°N
occurred within NMS boundaries occurred in Monterey Bay
and off Point Reyes (when effort extended into these areas).
34°N
34°N
See "Map Text" and "Discussion" in CDAS map section for
additional information on Dall’s porpoise.
32°N
32°N
About the SWFSC Blue Whale Maps (Figure 77). These
30°N
30°N
SWFSC maps of surveys conducted in July through early
December 1991, 1993, 1996, and 2001, (late upwelling and 132°W 130°W 128°W 126°W 124°W 122°W 120°W 118°W 132°W 130°W 128°W 126°W 124°W 122°W 120°W 118°W
Oceanic season) encompass a much larger geographic extent These maps contain data from one source: the NMFS/SWFSC cetacean stock assessment
National Marine Fisheries Service,
shipboard surveys, generally conducted during the late summer and fall, mostly from July through
Southwest Fisheries Science Center
December. These maps do not represent the species complete spatial and temporal distribution.
than the study area covered with the CDAS maps and indicate National Ocean Service,
Group size was estimated independently by all observers on each survey vessel who obtained a
good look at that group. These independent estimates of group size were averaged to give the National Centers for Coastal Ocean Science
concentrations of Blue whales off southern California and average group sized estimate for each sighting.
Figure 76. Maps for Dall’s porpoise: SWFSC stock assessment data: average group size of sightings and survey
further off-shore in pelagic, deep ocean habitats (not shown
effort.
121
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Blue Whale Balaenoptera musculus Humpback Whale Megaptera novaeangliae
132°W 130°W 128°W 126°W 132°W
124°W 130°W
122°W 128°W
120°W 126°W
118°W 124°W 122°W 120°W 118°W
132°W 130°W 128°W 126°W 132°W
124°W 130°W
122°W 128°W
120°W 126°W
118°W 124°W 122°W 120°W 118°W
48°N
48°N
48°N
48°N
1991 1993 1991 1993
Average Average Average Average
46°N
46°N
46°N
46°N
Group Size Group Size Group Size Group Size
3.1 - 5.0 3.1 - 14.0
3.1 - 5.0 3.1 - 14.0
44°N
44°N
44°N
44°N
2.1 - 3.0 2.1 - 3.0
2.1 - 3.0 2.1 - 3.0
2.0 2.0 1.9 - 2.0
1.9 - 2.0
42°N
42°N
42°N
42°N
1.1 - 1.9
1.1 - 1.9 1.1 - 1.8
1.1 - 1.8
1.0 1.0
1.0 1.0
40°N
40°N
40°N
40°N
Effort Effort
Effort Effort
38°N
38°N
38°N
38°N
36°N
36°N
36°N
36°N
34°N
34°N
34°N
34°N
32°N
32°N
32°N
32°N
30°N
30°N
30°N
30°N
48°N
48°N
48°N
48°N
1996 2001 1996 2001
Average Average Average Average
46°N
46°N
46°N
46°N
Group Size Group Size Group Size Group Size
3.1 - 14.0
3.1 - 5.0 3.1 - 14.0
3.1 - 5.0
44°N
44°N
44°N
44°N
2.1 - 3.0 2.1 - 3.0 2.1 - 3.0
2.1 - 3.0
2.0 1.9 - 2.0
1.9 - 2.0
2.0
42°N
42°N
42°N
42°N
1.1 - 1.9 1.1 - 1.8
1.1 - 1.9 1.1 - 1.8
1.0 1.0
1.0 1.0
40°N
40°N
40°N
40°N
Effort Effort
Effort Effort
38°N
38°N
38°N
38°N
36°N
36°N
36°N
36°N
34°N
34°N
34°N
34°N
32°N
32°N
32°N
32°N
30°N
30°N
30°N
30°N
132°W 130°W 128°W 126°W 124°W 122°W 120°W 118°W 132°W 130°W 128°W 126°W 124°W 122°W 120°W 118°W 132°W 130°W 128°W 126°W 124°W 122°W 120°W 118°W 132°W 130°W 128°W 126°W 124°W 122°W 120°W 118°W
These maps contain data from one source: the NMFS/SWFSC cetacean stock assessment These maps contain data from one source: the NMFS/SWFSC cetacean stock assessment
National Marine Fisheries Service, National Marine Fisheries Service,
shipboard surveys, generally conducted during the late summer and fall, mostly from July through shipboard surveys, generally conducted during the late summer and fall, mostly from July through
Southwest Fisheries Science Center Southwest Fisheries Science Center
December. These maps do not represent the species complete spatial and temporal distribution. December. These maps do not represent the species complete spatial and temporal distribution.
Group size was estimated independently by all observers on each survey vessel who obtained a National Ocean Service, National Ocean Service,
Group size was estimated independently by all observers on each survey vessel who obtained a
good look at that group. These independent estimates of group size were averaged to give the good look at that group. These independent estimates of group size were averaged to give the
National Centers for Coastal Ocean Science National Centers for Coastal Ocean Science
average group sized estimate for each sighting. average group sized estimate for each sighting.
Figure 78. Maps for humpback whale: SWFSC stock assessment data: average group size of sightings and
Figure 77. Maps for blue whale: SWFSC stock assessment data: average group size of sightings and survey
survey effort.
effort.
122
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Table 25. Preliminary life history and management information for selected marine mammals off north/central California.
SECTION SUMMARY Associations with Bathymetric Protection & Population Status Occurrence & Breeding in Study Area Prey Types
The marine mammal fauna of the study area include species Areas
Other Pelagic Invert's
Benthic Invertebrates
(e.g. squid, octopus)
with a variety of spatial and temporal patterns and can be Marine mammal distributions can be
Euphausiids (krill)
Other Vertebrates
generally characterized as: associated with bathymetrically-defined
• resident, breeding species that occur year-round Protection Time Period of
areas and results include: Status in Population Trend of Population in Primary
Plankton
(e.g. harbor seal, southern sea otter, Steller sea lion); Study Area the Study Area (Increasing, Occurrence in Breeding
Fishes
(FE, FT, SE, Decreasing, Relatively Stable, Study Area Time Period
• species that breed, pup, and molt in the study area and • Widely distributed (found throughout Common Name Scientific Name ST) Unknown) Temporal Occurrence (months) (months)
then as adults, feed elsewhere (e.g. northern elephant the study area): Dall’s porpoise and Federally June/July -
Southern sea otter Enhydra lutris nereis Year-round All months X X
Declining1
Threatened Oct/Nov
seals); northern fur seal (but mostly occurs Present year-round;
California sea lion Zalophus californianus Aug-Mar May-July X X
Increasing2
• species that are seasonally abundant during their migration over the slope and in the deep ocean); seasonally abundant
(e.g. gray whale); California sea lion, (but mostly occurs Año Nuevo: possibly stable last 3 yrs3;
Federally mid-May to
Steller sea lion Eumetopias jubatus Year-round All months X X
Threatened mid-July
Farallones: declining4
• seasonally abundant species that have either migrated to along the coast and over the inner
Stable (California net production may
these waters to forage during summer and fall shelf); and northern elephant seal Harbor seal Phoca vitulina richardsi Year-round All months Mar-June X X
be slowing)2
(e.g. humpback and blue whales) or to forage during and Steller sea lion.
Año Nuevo: Stable, last 5 yrs3;
winter (e.g. northern fur seal and California sea lions); and • Coastal: Southern sea otter, gray At-sea - unknown
Farallones: declining5 ; Pt. Reyes
• species which, though present year-round, exhibit highly whale (but this species also occurs Present year-round; at this time; mid-Dec thru
Northern elephant seal Mirounga angustirostris X X
Headlands: increasing6; California: net seasonally abundant at rookeries - mid-Mar
variable seasonal shifts in distribution (e.g. several species throughout the broad shallow shelf 2a
Nov-Mar
productivity rate declining ; number
of dolphins and porpoises). of the Gulf of the Farallones and in pups appears to be leveling off2a
proximity of the Farallon Islands). San Miguel Island stock: increasing2; Present year-round;
Northern fur seal Callorhinus ursinus Feb-May June-July X X
seasonally abundant
Preliminary CDAS maps for 13 species were developed for • Inner Shelf: Southern sea otter, Pribilof Is: rate of increase 8.12%7
No trend apparent
this document; this is fewer than half of the mammal species California sea lion, Steller sea lion, Dall's porpoise Phocoenoides dalli Year-round Unknown X X
Trends Unknown2 in data
in the study area and the maps are draft. No summary harbor seal, humpback whale and N. California stock: No Trends2a; San
Harbor porpoise (Northern CA,
analyses across mammal species were done, as they would gray whale. Francisco/Russian River and Monterey Unknown at this
San Francisco/Russian River, Phocoena phocoena Year-round Unknown X X
stock: trends in relative abundance not time
be inconclusive, and biased by the limited number and type of • Outer Shelf: California sea lion, Monterey stocks)
statistially significant.2a
species mapped (e.g., coastal, offshore). Steller sea lion, harbor seal, northern No trend apparent
Pacific white-sided dolphin Lagenorhynchus obliquidens Year-round Unknown X X
No Trends2
elephant seal, northern fur seal, in data
No trend apparent
However, preliminary data products do show that marine Risso’s dolphin, Dall’s porpoise, Risso's dolphin Grampus griseus Year-round Unknown X X
Trends Unknown2 in data
mammals of the study area are widely distributed from the Pacific white-sided dolphin, blue Bottlenose dolphin (California No trend apparent
Tursiops truncatus Year-round N/A X X
Stable2
shore to deep ocean, and while some species are found mostly whale, humpback whale and gray coastal stock) in data
Stock status unknown; likely
over the shelf, or deep offshore, most species occur over a whale. Not detected in CDAS No trend apparent
distributional shifts rather than
Short-beaked common dolphin Delphinus delphis Unknown X X
data in data
variety of bathymetric zones. Given that the data and maps • Slope: California sea lion, northern population increase2
are preliminary and most likely incomplete, it is not possible at elephant seal, Northern fur seal, Dall’s No trend apparent
Northern right whale dolphin Lissodelphis borealis Year-round Unknown X X
Trends Unknown2 in data
this time to evaluate the importance of smaller, discrete areas porpoise, Pacific white-sided dolphin, No trend apparent
Killer whale Orcinus orca Year-round Unknown X X X
Trends Unknown2
for the mammal species listed. Risso’s dolphin, northern right whale in data
Baird's beaked whale Berardius bairdii Unknown Insufficient data N/A ? ?
dolphin, blue whale and humpback Trends Unknown2
Beaked Whales
The broad-scale spatial coverage of the 16 maps for cetaceans whale. Mesoplodond spp. Unknown Insufficient data N/A ? ?
Trends Unknown2
(Mesoplodonts)
from the NMFS/SWFSC marine mammal stock assessment • Deep Ocean: California sea lion, Cuvier's beaked whale Ziphius cavirostris Unknown Insufficient data N/A ? ?
2
Trends Unknown
program (Barlow, unpublished data), provided additional northern fur seal, northern elephant Federally No trend apparent
Sperm whale Physeter macrocephalus Seasonal N/A X
2
Trends Unknown
Endangered in data
information for 13 species that were distributed in deep ocean seal, Dall’s porpoise, Pacific white- Federally
Blue whale Balaenoptera musculus Seasonal Aug-Nov N/A X X?
Increasing?8
habitats, and well beyond the range of the current CDAS data sided dolphin, Risso’s dolphin, Endangered
Federally Present year round;
set. These data will likely be incorporated into the CDAS data northern right-whale dolphin, blue Humpback whale Megaptera novaeangliae June-Nov N/A X X
Increasing 6-7%/yr9
Endangered seasonally abundant
set and mammal analysis planned for Phase II. whale, and humpback whale. Federally
Fin whale Balaenoptera physalus Seasonal Aug-Nov N/A X X
Trends Unknown2
Endangered
Stock status unknown; no data on No trend apparent
The marine mammal life history information and analytical Occurrence by Oceanographic Minke whale Balaenoptera acutorostrata Year-round N/A X X
in data
trends2
map products were used to develop the summary spatial and Season Dec-Jan
Increasing to late 1990's; Decreasing
Delisted Present year round;
Gray whale Eschrichtius robustus Dec-Apr (breeds off X X X
temporal distributions described below. Federal 1994 seasonally abundant
The seasonal occurrence patterns of (2002)10 Baja)
marine mammals in waters off north/ Notes
1. This table is preliminary; in Phase II more information will be added and the table will be reviewed by experts.
Life History Characteristics central California were clearly evident 2. A question mark (?) in the table indicates the entry is a probable entry (e.g., prey type); these items may be further evaluated in Phase II.
Table 25 is an initial summary of life history and management for migrating species of large cetaceans 3. Superscripts indicate sources as follows: 1-USGS, 2002; 2-Carretta et al.; 2001; 2a-Carretta et al. 2002; 3-P.Morris pers.comm., credited to B. Le Boeuf; 4-Hastings and Sydeman, 2001; and 5-USFWS, 2000.
6-Sydeman and Allen, 1999; 7-Gerrodette et al.,1985; 8-Calambokidis pers.comm.; 9-Forney et al., 2000; 10-Rugh et al., 2002
information that was identified in the marine mammal mapping (gray, blue and humpback whales) and 4. All marine mammal species have legal protection under the Marine Mammal Protection Act of 1972; species identified as Federally Endangered (FE), Federally Threatened (FT) are identified.
analyses. This table will be expanded in Phase II. for the non-breeding pinnipeds that No marine mammal species have designation as state endangered (SE) or state threatened (ST) in the California at this time.
123
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
breed mostly outside the study area (northern fur seal and seal and California sea lion also were relatively abundant (see Pinnipeds. Pinnipeds were found in the coast, shelf, slope and
• The Pacific white-sided dolphin was the most numerous of
California sea lion). The species occurrences in the three comments on the Upwelling and Oceanic seasons above). deep ocean habitats of the study area.
the small cetaceans and appeared to be more abundant
oceanographic seasons are described below. during the Oceanic Season in the Gulf of the Farallones and
Additional Observations. No clear seasonal patterns could Monterey Bay National Marine Sanctuaries; seasonal high • The California sea lion was the most numerous pinniped
Upwelling Season (~Spring/Summer). This season is be determined in this preliminary visual assessment for use areas were upper/lower slope regions in the Monterey seen in the study area and occurred throughout the region
characterized by an increase in cold, nutrient-rich water the smaller cetaceans: northern right-whale dolphin, Dall’s Bay and Cordell Bank National Marine Sanctuary. Given in coastal, shelf, and upper slope habitats. This species was
brought to the surface by persistent northwest wind and the porpoise, Pacific white-sided dolphin and Risso’s dolphin. the highly variable distribution of the Pacific white-sided most abundant during the Oceanic (just after its breeding
Coriolis effect, followed by intermittent relaxation of upwelling. For the latter two species, however, a shift in distribution was dolphin, the observed spatial and temporal distribution may period) and Davidson Current Season (before its next
Within-season variability during the upwelling process affects evident during the Oceanic season; see Figures 69 and 70. not indicate a general spatial/temporal pattern. breeding period) Seasons. Seasonal high use areas were
food web development and the availability of prey to marine Given the highly variable distribution of the smaller cetaceans, in proximity to major haulout sites near Año Nuevo and the
mammals. shifts in distribution (as indicated on the maps) may not indicate • The northern right whale dolphin was the second most Farallon Islands. Seasonal trends in relative abundance and
a seasonal pattern. numerous of the small cetaceans; concentrations appeared attendance at haulout sites were associated with warm-water
The Upwelling Season is also characterized by variations and to be greater within the Monterey Bay National Marine periods (El Niño events); sea lions were more numerous both
fluctuations in seasonal peaks in abundance of small schooling Elephant seals, Steller sea lions, and harbor seals were present Sanctuary, as well as outside National Marine Sanctuary at-sea and on land during these warm-water periods.
fish and relative densities of euphausiids. The humpback whale in National Marine Sanctuary waters year-round. And although boundaries. Seasonal high use areas were upper/lower slope
was present in greater abundance during the Upwelling and at-sea sightings are relatively infrequent, these species are regions in the Monterey Bay National Marine Sanctuary. No • The northern fur seal was the second most numerous
Oceanic seasons. Humpback whales migrate to waters off frequently sighted at haulouts and rookeries at specific times seasonal pattern in relative abundance was visually detected pinniped seen in the study area and occurred in outer shelf,
north/central California to feed on seasonally-abundant prey. of the year, as noted in Table 25. Important at-sea time periods in the maps; however, a seasonal shift in the distribution upper/lower slope and deep ocean habitats. Although seen
for these infrequently-sighted species is inconclusive due to of this species in the study area was apparent during the in all seasons, this species was most abundant during the
The northern fur seal also was relatively abundant during the insuffcient at-sea data and differences in behavior that affect Oceanic Season (concentrations were greater outside than Upwelling and Davidson Current Seasons (non-breeding
Upwelling and Davidson Seasons. After the breeding/pupping sighting frequency and otherwise low abundance. For example, within National Marine Sanctuary boundaries); during the period), a pattern that coincided with their migration to north/
season (June-July), adult females and juveniles migrate from some of these sighting issues include: at-sea sightings typically Davidson Season the greatest concentrations occurred in central California from San Miguel Island and the Pribilof
rookeries on San Miguel Island in the southern California Bight consist of single individuals or small groups of two or three; the southern regions of the Monterey Bay National Marine Islands. Seasonal high use areas were outside (to the west
(the San Miguel Island stock) and from the Eastern Pacific stock elephant seals are rarely at the surface; and Steller sea lions Sanctuary. Given the highly variable distribution of the and north) of National Marine Sanctuary boundaries.
of the Pribilof Islands and are therefore relatively abundant in are a threatened species and thus occur in small numbers. northern right whale dolphin, the observed occurrences
the study area during winter and early spring. may not indicate a general spatial/temporal pattern. • The northern elephant seal was the third most numerous
Overview of Occurrence Patterns pinniped seen in the study area, however, sightings were too
Oceanic Season (~Autumn). During the Oceanic Season, the • Risso’s dolphin was the third most numerous of the small
Cetaceans. Cetaceans were found throughout the study area; infrequent to determine seasonal trends in at-sea distribution.
northwest winds subside, warmer offshore water is advected cetaceans and occurred in shelf, and upper/lower slope
in coast, shelf, upper/lower slope and deep ocean habitats. Sightings occurred throughout the study region in shelf,
onshore, thermoclines strengthen, ocean conditions become habitats. This species was more widespread during the upper/lower slope and deep ocean habitats.
more stratified and marine mammal prey become more Upwelling Season and more concentrated in the southern
• The humpback whale was the most numerous pelagic baleen
stabilized. The following four species were relatively more portion of the study area (within and outside National Marine
whale seen in the study area and was seen more frequently • The harbor seal was the fourth most numerous pinniped seen
abundant during the Oceanic season (evaluated by the visual Sanctuary boundaries) during the Oceanic and Davidson
during the Upwelling and Oceanic Seasons. Seasonal high in the study area, however sightings were too infrequent to
inspection of the maps): Pacific white-sided dolphin, blue whale, Current Seasons. No clear seasonal shift in relative
use areas within the Gulf of the Farallones National Marine determine seasonal trends in at-sea distribution. Sightings
humpback whale, and California sea lion. Although the Pacific abundance in the study area was detected in a visual
Sanctuary were regions around the Farallon Islands and to occurred in coastal and shelf habitats.
white-sided dolphin occurred during all seasons, it appeared to inspection of maps. Seasonal high use areas were in the
the west of the islands, on the outer shelf and upper slope,
be more numerous during the Oceanic Season. The blue whale Monterey Bay National Marine Sanctuary over slope/canyon
regions over Pioneer, Ascension and Monterey canyons. • The Steller sea lion was sighted rarely, therefore no seasonal
(like the humpback whale) migrates to north/central California habitats. Given the highly variable distribution of the Risso’s
Given the highly variable distribution of humpback whales trends in at-sea distribution could be determined. Sightings
to forage on seasonally-abundant euhausiids during summer dolphin, the observed spatial and temporal distribution may
in the study area during the feeding season, observed spatial of this species occurred in coastal, shelf and upper slope
and fall. The California sea lion, the most abundant pinniped not indicate a general spatial/temporal pattern.
distribution may not indicate a general spatial pattern. habitats.
in the study area was present year-round, however, greater
numbers of sea lions were present during the Oceanic season • The Dall’s porpoise was the fourth most numerous small
• The gray whale was the second most abundant baleen A Fissiped. The southern sea otter is the only fissiped included
(just after the breeding season), but also during the Davidson cetacean and distribution was widespread on shelf, upper/
whale and was found in coastal and shelf regions; relative in the analysis. This species occurs year-round mostly along
Season (before the next breeding season). lower slope, and deep ocean habitats. No clear seasonal
abundance of the gray whale in the study area was greater the coast and inner shelf. Due to insufficient data no spatial/
pattern in relative abundance in the study area was visually
during the Upwelling and Davidson Current Seasons that temporal trends could be determined.
Davidson Current Season (~Winter). The Davidson Current detected in the maps. Seasonal high use areas were upper
coincided with the north and south migration of this species.
Season is characterized by frequent winter storms, downwelling, slope in the Cordell Bank and Gulf of the Farallones National
Seasonal high use areas were to the north of Cordell Bank Preliminary Observations of Species Distributions Relative
relatively warm uniform temperature to considerable depths and Marine Sanctuaries. Given the highly variable distribution
National Marine Sanctuary near Point Arena. Given the to National Marine Sanctuary Boundaries
a deep mixed layer. During this season, the gray whale was of the Dall’s porpoise, the observed occurrences may not
variable distribution of the gray whale, relative to the timing • Eight of the 13 marine mammals evaluated in this assessment
relatively abundant because it migrates through the study area indicate a general spatial/temporal pattern.
of the migration, this observed spatial distribution may not are relatively pelagic, far-ranging marine mammals that are
on its way south (or north) during this period. The northern fur indicate a general spatial pattern. widely distributed, and are either species that occur mostly
in deep ocean habitats (northern fur seal, northern elephant
124
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
mammals (e.g. harbor seals) feed on locally available and
seal, the endangered humpback and blue whales, Dall’s • Rookeries for the northern elephant seal, Steller sea lion and analysis will address the following factors:
seasonally abundant invertebrates or fish in relative proximity
porpoise, Risso’s dolphin), or over upper/lower slope habitats California sea lion occurred within National Marine Sanctuary • differences in survey methodology (e.g. line transect
to their breeding/pupping/haulout sites, whereas the seasonal
(northern right-whale dolphin, Pacific white-sided dolphin). All boundaries of the study area, however, during the El Niño of vs. strip transect);
migrants (e.g. humpback and blue whales) forage on seasonally
occur both in and outside of the National Marine Sanctuary 1998, California sea lion rookeries were located at Lion rock • differences in the detectability of pinnipeds, small and
available krill or fish, a pattern reflected in their relative
boundaries of the study area. and Point Sal Rock, to the south of National Marine Sanctuary large cetaceans, and effects of group size;
abundance during the Upwelling and Oceanic seasons.
boundaries. Haulout sites (n=3) for the threatened Steller sea • differences in time spent underwater; and
• The gray whale also occurs outside National Marine Sanctuary lion are located along the coast to the north of Cordell Bank • differences in environmental conditions (e.g. sea state
Response to Short-Term Changes in Climate. Although it
boundaries, but migrates through sanctuary waters along the National Marine Sanctuary. and other weather conditions).
is likely that short periods of unusually warm or cold waters
coast and over the continental shelf.
affect migratory species and shorter-ranging more resident
• Southern sea otters occurred in coastal shelf waters, almost Major tasks for Phase II are as follows:
species, it was not possible to determine effects of these events
• To the north of Cordell Bank (within relatively close proximity exclusively in Monterey Bay National Marine Sanctuary and 1. Complete the acquisition of data sets for the marine mam-
during this preliminary assessment. Distributional responses
to that National Marine Sanctuary boundary), the Pacific to the south, outside of the study area to San Nicholas Island mals from institutions already contacted (see partial list in
to the extremes of climate (e.g. El Niño vs La Niña) may be
white-sided dolphin, humpback whale, Dall’s porpoise, and in the southern California Bight. No. 2 below).
confounded by: 1) issues associated with comparing different
northern fur seal were present. A relatively high seasonal use
data sets without application of correction factors, 2) small
area for the gray whale also occurred north of Cordell Bank DISCUSSION 2. Continue working with marine mammal experts, and deter-
at-sea populations and therefore small sample size for some
National Marine Sanctuary. Given the general variable nature Differences in habitat use relative to large bathymetric features mine appropriate methods required to analyze additional
species (e.g. elephant seals, harbor seals), 3) far-ranging,
of cetacean distributions, these observations are preliminary are likely related to factors such as the distribution, abundance, data sets and apply appropriate correction factors. At a
migratory species being affected outside of the study area, 4)
and may not indicate a spatial pattern. and availability of various prey (species/sizes). Therefore, the minimum, these data sets will include: sighting data from
demographic lags to species’ responses, 5) variable effects on
importance of the study area must be considered in the context John Calambokidis at Cascadia Research, and the marine
different marine mammal prey, and 6) behavioral differences
• Relatively high seasonal use areas of the Dall’s porpoise, of the variability of ocean climate and oceanography, which mammal stock assessment program data from NOAA’s
among species.
humpback whale and blue whale were located seaward or strongly affects prey availability. Southwest Fisheries Science Center.
west of the Gulf of the Farallones National Marine Sanctuary
Nevertheless, the California sea lion provides an example
over the lower slope. Given the highly variable distribution of Cordell Bank, the Gulf of the Farallones, and Monterey Bay 3. Develop a composite marine mammal data set and maps
of a species shift in distribution in response to changing
these species, these observations are preliminary and may National Marine Sanctuaries encompass some of the most of occurrence patterns for additional mammal species,
oceanographic conditions. The relative increase in at-sea
not indicate a spatial pattern. productive waters along the California coast. Presence of marine as well as summary maps and analyses across species,
abundance during El Niño 1986-87, 1992-93, and 1997-98
mammals in these waters is affected not only by bathymetric for seasons and other selected time periods. Asemblage
(not presented in maps; see studies below), likely reflected a
• Seasonal high use areas of the northern fur seal, northern features, but also changing oceanographic conditions that analyses may be done to identify spatial/temporal species
greater than usual influx of individuals in response to a reduction
right whale dolphin, and Risso’s dolphin occurred seaward result in fluctuations in abundance and distribution of patchily groups.
in food off southern California (see Trillmich and Ono, 1991;
of the Monterey Bay National Marine Sanctuary (western distributed prey. The unique bathymetric features, coupled with
Allen, 1994; Keiper, 2001; Keiper et al., In Review). The greater
areas of the Monterey Canyon and the Shepard’s Meander). the complex physical oceanography off central California play 4. Complete a report on the mammal analyses that will ad-
numbers of sea lions at sea coincided with greater numbers
Given the highly variable distribution of these species, these an important role in the distribution of marine mammal prey dress survey data for 14-23 marine mammal species and
that occurred at haulout sites in the study area. For example,
observations are preliminary and may not indicate a spatial and, in turn, the distribution of mammals themselves. related summary mammal maps (e.g., a composite rookery
an influx of immature sea lions hauled out at Double Point
pattern. and haulout map, spatial and temporal summaries of at-
and at Point Reyes Headlands (per. comm S. Allen) during El
This unique combination of both wide and narrow continental sea occurrence data across selected mammal groups, and
Niño, as was also true at the Farallon Islands (Sydeman and
• Seasonal high use areas of Risso’s dolphin, northern fur seal, shelf, areas of high topographical relief (canyon edges, steep assemblage analyses).
Allen, 1999).
and Pacific white-sided dolphin also were located seaward slopes, ridges, banks, shelf breaks, seamounts), and the
of the southern regions of the Monterey Bay National distinctive oceanographic features associated with seasonal 5. Conduct an expert review of the maps and report and
In summary, seasonal and interannual processes in the ocean
Marine Sanctuary (near Lucia Canyon and to the south). upwelling (e.g., upwelling plumes, fronts, temporal and spatial incorporate necessary revisions.
climate affect variability in ocean conditions and food web
Given the highly variable distribution of these species, these variation in thermocline depth, surface and subsurface currents
development, and thus, the spatial and temporal occurrence
observations are preliminary and may not indicate a spatial and eddies) affect the distribution patterns of organisms at
patterns of marine mammals are strongly linked to the physical MAJOR SECTION CONTRIBUTORS
pattern. many trophic levels. For example, large concentrations of small
and biological processes that affect their prey. Glenn Ford, Carol Keiper, Janet Casey, David Ainley, Sarah
schooling fishes and euphausiids (krill) that are maintained by
Allen, Mark Lowry, Tracy Gill, Ken Buja and Wendy Williams.
• Within the study area, haulout sites for the northern the seasonally high primary productivity (supported by seasonal
Phase II Marine Mammal Assessment. This section provides
elephant seal all occurred within National Marine Sanctuary coastal upwelling), often occur along canyons, shelf-breaks,
preliminary results of the mammal analyses. The maps REVIEWERS
boundaries. Haulout sites for the harbor seal, California sea seamounts, and downstream of upwelling centers located at
presented here provide a preliminary estimate of the mammal The following institutes and people participated in the initial
lion and Steller sea lion also occurred within National Marine Point Arena, Point Reyes, Point Año Nuevo, and Point Sur,
species spatial and temporal use of the study area. In Phase map review in October 2002:
Sanctuary boundaries, but also both north (Steller sea lion, features that also are important areas for both large and small
II, additional data and analysis will likely yield revised maps Sarah G. Allen, Point Reyes National Seashore, Nat'l Park
California sea lion, harbor seal) and south (California sea lion, cetaceans.
for the existing species and additional maps for other species. Service
harbor seal) of the boundaries. Harbor seal haulouts occurred
Some of the data sets for marine mammals have only recently Scott Benson, NMFS/Southwest Fisheries Science Center
along the coast from Point Arena to Point Sal. Marine mammals are highly mobile marine predators that
been received and require further processing before species Jay Barlow, NMFS/Southwest Fisheries Science Center
feed on a great diversity of prey and are attracted to regions
distribution maps can be developed in the GIS. Phase II of this Nancy Black, Monterey Bay Whale Watch Institute
of seasonally abundant high prey densities. Resident marine
125
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Barlow, J. 1988. Harbor porpoise, Phocoena phocoena, abun- Calambokidis, J., J.C. Cubbage, H.H. Steiger, K.C. Balcomb, Croll, D.A., Tershy, B.R., Hewitt, R., Demer,D., Hayes, S.,
Don Croll, University of California, Santa Cruz
dance estimation for California, Oregon and Washington: I. Ship P.Bloedel. 1988. Humpback whale (Megaptera novaeangliae) Fiedler, Pl, Popp, J., and Lopez, V.L. 1998. An integrated ap-
Karin Forney, NMFS/Southwest Fisheries Science Center
surveys. Fish. Bull., US, Vol. 86(3), pp. 417-32. distribution and abundance in the Gulf of the Farallones, 1987 proach to the foraging ecology of marine birds and mammals.,
Mark S. Lowry, NMFS/Southwest Fisheries Science Center
Annual Report to Point Reyes National Seashore NPS and Gulf Deep-Sea Research II, Vol. 45, pp. 1353-1371.
Michelle Staedler, Monterey Bay Aquarium
Barlow, J., Oliver, C.W., Jackson, T.D. and Taylor, B.L. 1988. of the Farallones National Marine Sanctuary NOAA. NFMS. La
Jan Roletto Research Coordinator, GFNMS/CBNMS
Harbor porpoise, Phocoena phocoena, abundance estimation Jolla, CA. 74 pp. Darling, J.D., K.E. Keogh, T.E. Steeves. 1998. Gray whale
And several other members of the NOAA project team
for California, Oregon, and Washingtonf: II. Aeroal surveys. (Eschrichtius robustus) habitat utilization and prey species
and sanctuary programs.
Fish. Bull., US, Vol. 86(3), pp. 433-44. Calambokidis, J., J.C. Cubbage, G.H. Steiger, K.C. Balcomb, off Vancouver Island, B.C. J. Marine Mammal Science, Vol.
and P. Bloedel. 1990a. Population estimates of humpback 14(4), pp. 692-720.
PERSONAL COMMUNICATIONS
Barlow, J. and D. Hanan. 1995. An assessment of the status of whales in the Gulf of the Farallones, California, Report to the
Sarah Allen, Point Reyes National Seashore,
harbor porpoise in central California. Rept. Int. Whal., Special International Whaling Commission, Vol. 12, pp. 325-333 DeLong, R.L., and G.A. Antonelis. 1991. Impacts of the 1982-
U.S. National Park Service
Issue Vol. 16, pp. 123-140. 1983 El Niño on the northern fur seal population at at San
Jay Barlow, Southwest Fisheries Science Center, NOAA
Calambokidis J. G.H. Steiger, J.C. Cubbage, K.C.Balcomb, Miguel Island, California. Pinnipeds and El Niño: Responses to
Joelle Buffa, U.S. Fish and Wildlife Service
Barlow, J. and T. Gerrodette. 1996. Abundance of cetaceans C.Ewald, S. Kruse, R. Wells, and R. Sears. 1990b. Sightings Environmental Stress. (Trillmich, F., and K. Ono, Eds.) Springer-
John Calambokidis, Cascadia Research
in California waters based on 1991 and 1993 ship surveys. and movements of blue whales off central California 1986-88 Verlag, New York. Pp. 75-83.
Karin Forney, Southwest Fisheries Science Center, NOAA
NOAA Tech. Memo NOAA-TM-NMFS-SWFSC-205. La Jolla, from photo-identification of individuals. Report to the Interna-
Denise Greig, The Marine Mammal Center
CA. 68 pp. tional Whaling Commission, Vol. 12, pp. 343-348. DeLong, R.L., S.R. Melin. 1999. Population monitoring stud-
Mike Harris, California Department of Fish and Game
ies of northern fur seals at San Miguel Island, California. Fur
Brian Hatfield, U.S. Geological Survey
Benson S.R., D.A. Croll, B.B. Marinovic, F.P.Chavez, J.T. Har- Calambokidis, J., T. Chandler, K. Rasmussen, G.H. Steiger, seal investigations, 1997. (E.H. Sinclair, BW Robson, Eds) US
Mike Kenner, California Department of Fish and Game
vey. 2002. Changes in the cetacean assemblage of a coastal L. Schlender. 1998. Humpback and blue whale photographic Dep. Commer. NOAA Tech. Memo NMFS-AFSC-69 Kodiak,
Mark Lowry, Southwest Fisheries Science Center, NOAA
upwelling ecosystem during El Niño 1997-98 and La Niña 1999. identification: report on research in 1997. Report to Southwest AK. Pp. 73-80.
Pat Morris, University of California, Santa Cruz
Prog. in Ocgrphy, Vol. 54, pp. 279-291. Fisheries Science Center, NMFS, La Jolla, California. 36 pp.
Joe Mortenson, Gulf of the Farallones National Marine
Dohl, T.P., R.C. Guess, M. Duman and R.C. Helm. 1983. Ce-
Sanctuary, Friends of Marine Sanctuaries Association
Byrd, B. Abundance, distribution, food habits, and prey avail- Calambokidis, J. G.H. Steiger, K. Rasmussen, J.H. Urban R, taceans of central and northern California: status, abundance,
Bob Read, California Department of Fish and Game
ability of the harbor porpoise (Phocoena phocoena) in northern K.C. Balcomb, P. Ladron de Guevara P, M. Salinas Z, J.K. Ja- and distribution Final Rep, Minerals management Service
Michelle Staedler, Monterey Bay Aquarium
Monterey Bay, CA. M.S. Thesis, Moss Landing Marine Labora- cobsen, C.S. Baker, L.M. Herman, S. Cerchio and J.D. Darling. Contract 14-12-0001-29090 OCS Study MMS 84-0044. Los
William Sydeman, PRBO Conservation Science
tories, Stanislaus California State University. 159 pp. 2000. Migratory destinations of humpback whales that feed off Angeles, CA. 143 pp.
California, Oregon and Washington. Mar. Ecol. Prog. Ser., Vol.
REFERENCES
Black, N.A. 1994. Behavior and ecology of Pacific white-sided 192, pp. 295-304. Dohl, T.P., R.C. Guess, M.L. Duman, and R.C. Helm. 1983. Ce-
Allen, S.G. 1994. The distribution and abundance of marine
dolphins (Lagenorhynchus obliquidens) in Monterey Bay, taceans of Central and Northern California, 1980-1983: Status,
birds and mammals in the Gulf of the Farallones and adjacent
California. MS thesis, Moss Landing Marine Laboratories, Calambokidis, J. G.H. Steiger, J.R.Evenson, K.R .Flynn, K.C. Abundance, and Distribution. Part of Investigator’s Final Re-
waters. PhD. Dissertation UC Berkeley.
San Francisco State University CA. 111 pp. Balcomb, D.E. Claridge, P. Bloedel, J.M. Straley, C. S. Baker, O. port, Marine Mammal and Seabird Study, Central and Northern
von Ziegesar, M. E. Dahlheim J.M. Waite, J.D. Darling, G. Ellis California, Contract No.14-12-0001-29090. Prepared by Center
Allen, S.G. , D. Press, S. Waber. 2002. Long term monitoring
Bonnell, M.L., M.O. Pierson and G.D. Farrens. 1983. Pinnipeds and G.A. Green. 1996. Interchange and isolation of humpback for Marine Sciences, University of California, Santa Cruz, for
of harbor seals at Point Reyes National Seashore, Point Reyes
and sea otters of central and northern California, 1980-1983: whales off California and other north Pacific feeding grounds. the Pacific OCS Region, Minerals Management Service. OCS
California 5 year Annual Report. National Park Service Techni-
Status, abundance and distribution. Final report prepared by Mar. Mam. Sci., Vol. 12(2), pp. 215-226. Study MMS 84-0045. Santa Cruz, CA. 284 pp.
cal Report. Washington, D.C. 41 pp.
Center for Marine Studies, University of California, Santa Cruz,
for the Minerals Management Service, Contract 14-12-00. OCS Carretta, J.V., J. Barlow, K.A. Forney, M.M. Muto and J. Baker. Estes, J.A., B.B. Hatfield, K. Ralls, J. Ames. 2003. Causes of
Antonelis, G.A., Jr., M.S. Lowry, D.P. DeMaster, and C.H., Fis-
Study MMS 84-0044. Santa Cruz, CA. 82 pp. 2001. U.S. Pacific marine mammal stock assessments: 2001, mortality in California sea otters during periods of population
cus. 1987. Assessing northern elephant seal feeding habits by
U.S. Dept Commerce, NOAA-NMFS-SWFSC, NOAA Tech growth and decline. J. Marine Mammal Science, Vol. 19(1),
stomach lavage. Mar. Mam. Sci., Vol. 3, pp. 308-322.
Bonell, M. L., M.O. Pierson, and G.D. Farrens. 1983. Pinnipeds Memo NOAA-TM-NMFS-SWFSC-317. La Jolla, CA. 209 pp. pp. 198-216.
and Sea Otters of Central and Northern California, 1980-1983:
Barlow, J. Unpublished data. Cetacean Stock Assessment
Status, Abundance, and Distribution. Part of Investigator’s Fi- Carretta, J.V., M.M. Muto, J. Barlow, J. Baker. Draft U.S. Pacific Forney, K.A., J. Barlow, J.V. Carretta. 1995. The abundance of
Ship Survey Data from the Summer and Fall of 1991, 1993,
nal Report, Marine Mammal and Seabird Study, Central and marine mammal stock assessments 2002, U.S. Dept Com- cetaceans in California waters. Part II: Aerial surveys in winter
1996 and 2001. Coastal Marine Mammal Program, Southwest
Northern California, Contract No. 14-12-0001-29090. Prepared merce, NOAA-TM-NMFS-SWFSC Technical Memorandum. and spring of 1991 and 1992. Fish Bull. Vol. 93, pp. 15-26.
Fisheries Science Center, 8604 La Jolla Shores Dr., La Jolla,
by Center for Marine Sciences, University of California, Santa La Jolla, CA. 199 pp.
CA 92037. Jay.Barlow@noaa.gov http://swfsc.nmfs.noaa.gov/
Cruz, for the Pacific OCS Region, Minerals Management Ser- Forney, K.A., J. Barlow, and J.V. Carretta. 1995. The abundance
prd/cmmp/
vice. OCS Study MMS 84-0044. Santa Cruz, CA. 220 pp. Condit, R.S., and B.J. LeBoeuf. 1984. Feeding habits and of cetaceans in California waters. Part I: Aerial surveys in winter
feeding grounds of the northern elephant seal. J. of Mam., and spring of 1991 and 1992. Fish. Bull., Vol. 93, pp. 15-26.
Barlow, J. 1995. The abundance of cetaceans in California
Bonnell, M.L. and R.G. Ford. 1987. California sea lion distri- Vol. 65, pp. 281-290.
waters. Part I: Ship surveys in summer and fall of 1991. Fish.
butions: A statistical analysis of aerial transect data. J Wildl.
Bull. Vol. 93, pp. 1-14.
Manage. Vol. 51(1), pp. 13-20.
126
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
LeBoeuf, B.J., K. Ono and J. Reiter. 1991. History of the Steller
Hastings, K.K. and W.J. Sydeman. 2002. Population status,
Forney, K.A. and J. Barlow. 1998. Seasonal patterns in the Miller, M.A., L.A. Gardner, C. Krueder, D.M. Paradies, K.R.
sea lion population at Año Nuevo Island, 1961-1991. National
seasonal variation in abundance, and long-term population
abundance and distribution of California cetaceans, 1991-1992. Worcester, D.A. Jessup, E. Dodd, M.D. Harris, J.A. Ames, A.E.
Marine Fisheries Service, Southwest Fisheries Science Center,
trends of Steller sea lions (Eumetopias jubatus) at the South
Mar. Mam. Sci., Vol. 14, pp. 460-489. Packham and P.A. Conrad. 2002. Coastal freshwater runoff is
Administrative Report LJ-91-45C. La Jolla, CA. pp. 24.
Farallon Islands, California. Fish. Bulletin, Vol. 100, pp. 51- a risk factor for Toxoplasma gondii infection of southern sea
62.
Forney, K.A. 1999a. The abundance of California harbor por- otters (Enhydra lutris nereis). Intl. J. of Parasitology, Vol. 32,
Lowry, Mark S. Unpublished data. Survey data for haulouts
poise estimated from 1993-97 aerial line-transect surveys. pp. 997-1006.
and rookeries of California sea lion (Zalophus californianus)
Kajimura, H. 1980. Distribtuion and migration of northern fur
NOAA/SWFSC Admin.Rep. LJ-9-02. La Jolla, CA. 16 pp.
and Steller sea lion (Eumetopias jubatus) in central and north-
seals (Callorhinus ursinus) in the Eastern Pacific. Further analy- National Marine Fisheries Service. 1992. Final recovery plan for
ern California, 1998-2001. National Marine Fisheries Service,
sis of pelagic fur seal data collected by the United States and
Forney K.A. 1999b. Trends in harbor porpoise abundance off Steller sea lions, Eumetopias jubatus. Prepared by the Steller
Southwest Fisheries Science Center. La Jolla, CA.
Canada during 1958-74. Kajimura, H. R.A. Lander, M.A. Perez,
central California, 1986-95: Evidence for interannual changes sea lion recovery team for the National Marine Fisheries Ser-
A.E. Yor (eds). North Pacific Fur Seal Commission 23rd annual
in distribution? J. Cetacean Res. Manage., Vol. 1, pp. 73-80. vice. Silver Spring, Maryland. 290 pp.
Lowry, M.S., C.W. Oliver, C. Macky, and J.B. Wexler. 1990.
meeting Standing Scientific Committee 7-12, April 1980, USSR:
Food habits of California sea lions Zalophus californianus at
Moscow. pp. 19-48.
Forney, K.A., and M.M. Muto, and J. Baker. 2000. Draft U.S. National Marine Fisheries Service. 2000. Biological opinion
San Clemente Island, California, 1981-1986. Fish. Bull. U.S.,
Pacific marine mammal stock assessments: 1999. U.S. Depart- on the California/Oregon Drift Gillnet Fishery. Silver Spring,
Vol. 88, pp. 509-521.
Kajimura, H. 1984. Opportunistic feeding of the northern fur
ment of Commerce, NOAA Technical Memorandum. NMFS- MD. 64 pp.
seal, Callorhinus ursinus, in the eastern North Pacific Ocean
SWFSC-282. La Jolla, CA. 62 pp.
Lowry, M.S., B.S. Stewart, C.B. Hyeath, P.K. Yochem, and J.M.
and eastern Bering Sea. U.S. Dept.of Commerce, NOAA Tech- Nusbaum, J. 2002. Population size and breeding success of
Francis. 1991. Seasonal and annual variability in the diet of Cali-
nical Report NMFS SSFR-779. Washington, D.C. 79 pp.
Gaskin, D.E. 1984. The harbour porpoise (Phocoena pho- the Northern elephant seal on Southeast Farallon Island, PRBO
fornia sea lions Zalophus californianus at San Nicolas Island,
coena): regional populations, status, and information on Marine Science Division; Report to USFWS, San Francisco Bay
California, 1981-1986. Fish. Bull. U.S., Vol. 89, pp. 331-336.
Keiper, C.A. 2001. Marine mammals off central California rela-
direct and indirect catches. Rep. Int. Whal. Comm., Vol. 34, National Wildlife Refuge. San Francisco, CA. 59 pp.
tive to hydrography; 1986-94, 1997-99. MS Thesis, Moss Land-
pp. 569-586.
Lowry, Mark S. and J. V. Carretta (In review). Pacific harbor
ing Marine Laboratories, San Jose State University, 98 pp. Oliver, G.W., J. Gilardi, C.Toropoa, P. Folkens, K. Cronin, N.
seal, Phoca vitulina richardsi, census in California during May-
Glantz, M.R., and J.D. Thompson. 1981. Resource Manage- Bodorff, K. Sanchez, W. Damon, K.A. Zagzebski, B. Winning.
July 2002. To be published as: NOAA Technical Memorandum
Keiper, C.A., D.G. Ainley, S.G. Allen, J.T. Harvey. In Review.
ment and Environmental Uncertainty. Wiley, New York. 63 2001. Gray whales in San Francisco Bay. Poster Abstract
NMFS, NOAA-TM-NMFS-SWFSC. La Jolla, CA.
Inter-annual variability in ocean climate and marine mammal
pp. submitted at the 14th Biennial Conference on the Biology of
occurrence patterns off central California, 1986-1994, 1997- Marine Mammals.
Lowry, M.S. and K.A. Forney. In Review. Abundance and dis-
1999. Marine Ecology Progress Series.
Green, G.A., J.J. Brueggeman, R.A. Grotefendt, D.E. Bowlby,
tributon of California sea lions in central and northern California
M.L. Bonnell and K.C. Balcomb, III. 1992. Cetacean distribu- Orr, R.T. 1966. Risso’s Dolphin on the Pacific Coast of North
during 1998 and summer 1999. Fish. Bull. US.
Kenyon, K.W. and F. Wilke. 1953. Migration of the northern fur
tion and abundance off Oregon and Washington, 1989-1990. America. J. of Mamm., Vol. 47, pp. 341-343.
seal, Callorhinus ursinus. J. of Mamm., Vol. 34(1), pp. 86-98.
Oregon and Washing ton Marine Mammal and Seabird Surveys.
MMS (Minerals Management Service). 2001. Marine Mam-
(J.J Brueggeman, Ed.). Minerals Management Service Contract Perryman, W.L., M.A.Dohahue, P.C.Perkins, S.B.Reilly. 2002.
mal and Seabird Computer Database Analysis System for
Kieckhefer, T.R. 1994. Feeding ecology of humpback whales
Report 14-12-0001-30426. Anchorage, AK. 121 pp. Gray whae calf production 1994-2000: Are observed fluctua-
Washington, Oregon and California, 1975-1997 (MMS-
in continental shelf waters near Cordell Bank, California. MS tions related to changes in seasonal ice cover? J. Mar. Mam.
CDAS, version 2.1). Prepared by Ecological Consulting,
thesis, Moss Landing Marine Laboratories, San Jose State
Gulland, F.M.D., H. Perez-Cortes, J. Urban R., L. Rojas-Bracho, Sci., Vol. 18(1), pp. 121-144.
Inc. (now R.G. Ford and Consulting Co.), Portland, Oregon
University, San Jose, CA. 170 pp.
G. Ylitalo, C. Kreuder, and T. Rowles. 2002. Eastern north Pa-
for the Minerals Management Service, Pacific OCS Region,
cific gray whale (Eschrichtius robustus) unusual mortality event, Pettee, J. 1989. Female northern elephant seal reproductive
Order No.1435-01-97-PO-4206. Anchorage, AK.
Koski, W.R., J.W. Lawson, D.H,Thomson, W.J. Richardson.
1999-2000: A compilation. International Whaling Commission success at Point Reyes National Seashore and micro-habitat
1998. Point Mugu sea range marine mammal technical report.
Rpt SC/54/BRG23. Cambridge, UK. 200 pp. features. MS Thesis, San Francisco State University, San
Mate, B.R., B.A. Lagerquist, and J.Calambokidis. 1999. Move-
Naval Air Warfare Center Weapons Division. San Diego, CA. Francisco, CA.
ments of North Pacific blue whales during their feeding season
280 pp.
Harris, M. Unpublished data. Survey data of southern sea ot-
off Southern California and their southern fall migration. Mar.
ters in California: Fall 2001 and Spring 2002 counts. Marine Pyle, P. and L Gilbert. 1996. Occurrence patterns and trends of
Mamm. Sci., Vol. 15 (4), pp. 1246-1257.
Leatherwood, S., and R.R. Reeves. 1983. The Sierra Club
Region, California Department of Fish and Game. San Diego, cetaceans recorded from southeast Farallon Island, California,
Handbook of Whales and Dolphins. Sierra Club, San Francisco,
California. 1973 to 1994. Northwestern Naturalist 77:1-8.
Melin, S.R., and R.L. Delong 2000. Population and monitoring
CA. 302 pp.
studies of northern fur seals at San Miguel Island, California.
Harvey, J.T., R.C. Helm, and G.V. Morejohn. 1995. Food Habits Pyle P., D.L. Long, J. Schonewald. 2001. Historical and re-
Fur seal investigations, 1998. B.W. Robson (ed.). NOAA Tech.
Le Boeuf, B.J., D. Crocker, S. Blackwell, P. Morris. 1993. Sex
of Harbor Seals Inhabiting Elkhorn Slough, California. California cent colonization of the South Farallon Islands, California, by
Memo. NMFS-AFSC-113. Silver Spring, MD. pp. 41-5.
differences in diving and foraging behaviour of northern el-
Fish and Game, Vol. 81, pp. 1-9. Northern fur seals (Callorhinus ursinus). J. Mar. Mam. Sci.,
ephant seals. Marine Mammal: Advances in Behavioural and Vol. 17(2), pp. 397-402.
Morejohn, G.V. 1979. The Natural History of Dall’s Porpoise in
Population Biology. ( I. Boyd, ed.). Oxford Univ. Press. pp.
the North Pacific Ocean. Behavior of Marine Animals, Vol. 3.
23-45.
Cetaceans. H.E. Winn, and B.L. Olla, eds. Plenum, New York,
NY. pp. 45-83.
127
Section 2.3: BIOGEOGRAPHY OF MARINE MAMMALS
Ream, R.R., J.D. Baker and R.G. Towell. 1999. Bogoslof Is- Weise, M.J. 2000. Abundance, food habits, and annual fish
land studies, 1998. Fur seal investigations, 1997. E.H. Sinclair consumption of California sea lion (Zalophus californianus) and
and B.W. Robson (eds). US Dep Commer, NOAA Tech Memo its impact on salmonid fisheries in Monterey Bay, CA M.S.
NMFS-AFSC-69. Kodiak, Ak. 268 pp. Thesis, Moss Landing Marine Laboratories, San Jose State
University. 87 pp.
Reidman, M.L. and J.A. Estes. 1990. The sea other (Enhydra
lutris): Behavior, ecology, and natural history. USFWS. Biologi- Weise, M.J., J.T. Harvey 2001. Nighttime foraging of harbor
cal Report, Vol. 90(14). seals on winter-run steelhead in the San Lorenzo River, Cali-
fornia. Paper presented at the 14th Biennial Conference on the
Rugh, D.J., J.M. Breiwick, R.C. Hobbs, and J.A. Lerczak. Biology of Marine Mammals. 27 pp.
2002. A preliminary estimate of abundance of the eastern
North Pacific stock of gray whales in 2001 and 2002. Paper
SC/545/BRG6 presented to the IWC Scientific Committee,
April 2002. 14 pp.
Schoenherr J.R. 1991. Blue whales feeding on high concen-
trations of euphausiids around Monterey Submarine Canyon.
Can. J. Zool., Vol. 69, pp. 583-594.
Stewart, B.S. and H.R. Huber. 1993. Mirounga angustirostris.
Mammalian Species 449:1-10.
Stewart, B.S., B.J. Le Boeuf, P.K. Yochem, H.R. Huber, R.L.
DeLong, R.J. Jameson, W.,Sydeman, and S.G. Allen. 1994.
History and present status of the northern elephant seal popu-
lation. Elephant Seals. B.J. Le Boeuf and R. M. Laws (eds.)
Univ. Calif. Press, Los Angeles. pp. 101-135.
Stroud, R.K., C.H.Fiscus, and H.Kajimura. 1981. Food of the
Pacific White-sided Dolphin , Lagenorhynchus obliquidens,
Dall’s Porpoise, Phocoenoides dalli, and Northern Fur Seal,
Callorhinus ursinus, off California and Washington. Fishery
Bulletin, Vol. 78, pp. 951-959.
Sydeman, W.J., S.G. Allen. 1999. Pinniped population dynamics
in central California: Correlations with sea surface temperature
and upwelling indices. Mar. Mam. Sci., Vol. 15(2), pp. 446-461.
Trillmich F. and K.A. Ono (eds). 1991. Pinnipeds and El Niño:
Responses to environmental stress. Springer-Verlag, Berlin.
97 pp.
U.S. Geological Survey 2002. (www.werc.usgs.gov/otters/ca-
surveydata.html).
USFWS. 2000. Draft Revised Recovery Plan for the Southern
sea other (Enhydra lutris nereis). Portland, Oregon. U.S. Fish
and Wildlife Service, Ecological Services, Ventura Field Office.
Ventura, California. 64 pp.
128
Section 3: INTEGRATION OF ANALYSES
where ni is the number of individuals belonging to the species
• Option 2: A co-occurrence analysis of marine bird density a continuous modeled surface rather than estimates per 5
(S) in the sample (5 minute grid), and n is the total number of
and fish density minute grid. This approach takes into consideration the spa-
individuals in the sample (Ludwig and Reynolds 1988). Diver-
tial structure in the data to model the gradient of the metric
sity was calculated independently for birds and fishes using all
• Option 3: A co-occurrence analysis of density and diversity between any given pair of sampling points. This results in
species observed within a grid cell.
(options 1 and 2 combined) for both fish and marine birds smoothed surfaces that permit easier visualization of biologi-
cally significant areas. Resulting large-scale patterns have
In the first of these approaches, only patterns of species been described in the context of sanctuary boundaries to
diversity were analyzed. This index was relatively simple to provide insights that may enhance management efficacy in
calculate using the data available for birds and fishes, and these protected areas.
represents a common metric for integration. The second op-
tion focuses on spatial patterns of density. Density is a more DATA AND ANALYSES
intuitive measure than diversity, and it highlights regions of Integration Metrics. There are a number of ways by which
highest marine bird concentrations (abundance). An added ecologists measure diversity. The simplest metric is a count of
attraction of density is that it is only weakly influenced by the total number of unique species in a community, also called
effort. The third approach incorporates the two metrics for species richness (S). This is a straightforward, though poten-
marine birds and fishes simultaneously by combining results tially misleading, measure of diversity. Sampling must be con-
of options 1 and 2. ducted at all locations with the same amount of effort for this
estimate to be comparable across a study region or between
Metrics used in these three options were chosen to best de- data sets. Unfortunately, this was not the case with any of the
INTRODUCTION fine the biogeography for each taxon based on the available source data available for integration. For example, marine
The greatest challenge in developing a large-scale bio- data. Once each integration parameter was mapped, patterns bird observation transects were far more numerous (more ef-
geographic assessment is the synthesis and subsequent of community structure were superimposed and interpreted fort) near shore, and declined dramatically with distance from
analysis of spatial data collected at different scales for varied in the context of various biological and physical covariates. shore. Because this is often the case with biological sampling,
objectives (Gotway and Young 2002). This is particularly true Figure 79. Pictogram of species diversity. Both fish communi-
These spatial covariates were used to better understand gen- a number of diversity indexes have been developed that are,
when attempting to describe meso-scale (tens to hundreds of ties are comprised of 5 species and 14 individuals. In the com-
eral biogeographic patterns, and through interpretation, sug- in theory, more independent of sample size. These are based
kilometers) spatial patterns using data for a range of taxa that munity on the left side, there are 9 individuals of species 1, 1
gest reasons for the observed spatial trends. For example, on the relationship between species richness and the total
were each collected using different sampling techniques. The of species 2, 2 of species 3, 1 of species 4, and 1 of species
results indicated that a portion of highest observed bird and number of individuals observed (n), both of which increase as
taxon-specific sections of this document describe spatial pat- 5. Using the distribution of abundance within this community,
fish diversity occurred adjacent to the shelf/slope interface. It a function of effort, and, ideally, cancel out the effect of effort
terns of community structure for marine birds, mammals, and Shannon’s Index of diversity is 1.12. The community on the
is well documented that strong upwelling of deep ocean wa- on the resulting index (Ludwig and Reynolds, 1988). Here the
fishes. The intent of this section is to coalesce these results right also consists of 5 species with 14 individuals; however, the
ters consistently occurs in areas along the slope. Nutrients in Shannon index of diversity (Shannon and Weaver, 1949) was
and construct a unified and biologically relevant assessment distribution of abundance is more even (2, 3, 3, 4, and 2 indi-
these waters support high phytoplankton productivity, which chosen, as this index is the most widely used in community
of the biogeographic patterns observed. viduals), and consequently Shannon’s Index is higher (1.57).
stimulates a cascade of productivity at all levels of the marine ecology and has relatively small statistical bias when sample
food web (Bolin and Abbott, 1963; Ryther, 1969; Malone, sizes are large (as is the case with this source data).
There are a number of ways to address the challenge of inte- Once diversity was calculated for each taxon in each sample,
1971; Barber and Chavez, 1983; Chavez 1995, 1996; Bakun,
grating results for multiple taxa, and this section contains re- a continuous map surface was interpolated to predict diversity
1996). Diversity may be thought of as being composed of two distinct
sults for three (of many) reasonable options. This integration patterns throughout the study area. The same process was
components: 1) species richness, and 2) species evenness.
effort has been tailored to the NMSP mission of “...enhancing used to model density (see below for detailed methods).
Furthermore, by combining multiple parameters across taxa Evenness is defined as how the number of individuals is dis-
biodiversity, ecological integrity, and cultural heritage”, and (option 3), it was possible to link results presented in earlier tributed among the species. For example, for a community
specifically focuses on the notion of biodiversity in describing Spatial Modeling. This section details the procedure used to
sections to an integrated composite. This approach provides comprised of five species with 70% of the individuals belong-
the overall biogeography of the region. process input data for the integration analyses. While techni-
a clear and tractable interpretation that the reader can follow ing to one species and 30% distributed among the remaining
cal in nature, it provides the information necessary for NMSP
as a logical end point to the preceding series of analyses. four species, the evenness component would be lower than if
After a thorough assessment of the spatial data for each and others to generate results identical to those presented
The combination of diversity and density presents an inclu- there were a more even distribution of individuals among the
taxon, it was concluded that the marine mammal data were here using data provided in the appendix to this document
sive view of important areas across taxa, and is less likely to five species (Ludwig and Reynolds. 1988) (Figure 79). Maxi-
not robust enough in present form to include in the integra- (CD-ROM), and to explore results of alternate modeling op-
overlook regions of potential importance when compared to mum diversity for a given number of species and individuals is
tion process. As such, only birds and fish were considered tions. The observed patterns in diversity and density were
maps depicting a single estimate (e.g., options 1 and 2). This achieved where equal numbers are found for each species in
here. Additional efforts to reconcile outstanding issues in the found to be robust to changes in model parameters; however,
is a critical point as the most diverse patch in a seascape is a community. For consistency, data for all taxa included in this
marine mammal data are ongoing. A final integrated analysis, calculations of the aerial extent of persistent patterns may be
not necessarily the most productive. In addition to the general section were summarized by five minute grids (see sections
including mammal data, will be completed during Phase II of more sensitive. For example, the location of areas of high bird
patterns observed for each metric, the spatial coincidence of 2.1, 2.2, 2.3). Total diversity was estimated within each grid cell
this assessment. The integration alternatives provided in this diversity tends to be relatively constant, regardless of model
hot spots among taxa is emphasized to provide a view into using the Shannon index (H’);
section include: parameters. The quantity (e.g., square kilometers) of these
the integrated ecosystem. The metrics used in this section
n n
S
high areas that fall inside sanctuary boundaries, however,
H ′ = − ∑ i ln i
are similar to those described in sections 2.1 (fish) and 2.2
• Option 1: A co-occurrence analysis of diversity hot spots for may change.
(marine birds); however, data were interpolated to produce i =1 n n
marine birds and marine fishes
129
Section 3: INTEGRATION OF ANALYSES
on the pattern of the empirical variograms and the lack of data
For interpolation and calculation of spatial autocorrelation sta-
Table 26. Summary statistics and parameter estimates for spatial models.
at short lag distances (due to the five minute minimum sepa-
tistics, data for each 5 minute grid cell were assigned to the
Geary's C (significance)
Moran's I (significance)
ration between points), which are necessary to differentiate
cell centroid. All data were analyzed in the Universal Trans-
(total,minimum per
anisotropic model)
Cross Validation r 2r2
Cross-validation
between spherical and Gaussian models.
verse Mercator (UTM) projection. Projection is necessary to
(minor range for
Number of lags
Lag size (km)
ensure that the value of x and y units is equivalent and con-
Sample Size
Detrending?
Range (km)
Lag size (km)
Neighbors
4) Surface Interpolation: The interpolation method used
stant across the study region. The spatial modeling process
Partial Sill
Geary's C
Sample size
Moran's I
Range (km)
Detrending
Neighbors
Partial sill
Nugget
is termed ‘ordinary kriging’. Kriging is a linear interpolation
to generate an interpolated surface consisted of the following
sector)
Nugget
method that allows predictions of unknown values of a ran-
sequence of operations:
Lags
dom function from observations at known locations (Kaluzny
Bird diversity 1,163 0.141** 0.830** Yes 20.451 12 99.851 0.25 0.117 0.541 8,2
et al., 1998). Ordinary kriging is the kriging method gener-
1) Checking for Spatial Autocorrelation: Prior to interpola-
Bird diversity residual 1,163 0.067** 0.893** YesYes 9.966 12 88.062 0.199 0.125 0.407 8,8,2
Bird diversity 1163 0.141** 0.830** 20.451 12 99.851 0.25 0.117 0.541 2
ally used for interpolation of a single continuous variable of
tion, all data were tested for the presence of spatial autocorre-
Bird density 1,403 0.058** 0.891** Yes 5 30 149.54 1,189.30 1,189.50 0.253 20,5
unknown mean. Kriging is preferred over other interpolation
lation. Positive autocorrelation (where values for neighboring Bird diversity residual 1163 0.067** 0.893** Yes 9.966 12 88.062 0.199 0.125 0.407 8, 2
methods because: 1) weights are based on an empirical as-
pairs of points are more similar to one another than are distant Fish diversity 301 0.018**
1403 0.058** 0.973* No 5 20 30 149.54 0.148
29.595 0.046 0.025 20,5
Bird density 0.891** Yes 5000 1189.3 1189.5 0.253 20, 5
sessment of the data’s spatial structure (the variogram), 2)
ones) is important for accurate interpolation. Moran’s I and Fish diversity residual 301 3010.013*
0.018** 0.975* No no 55 10 18.394 0.11 0.061 0.076 20,5
Fish diversity 0.973* 20 29.595 0.148 0.046 0.025 20, 5
kriging is an unbiased predictor, and 3) for many variables,
Geary’s C statistics were calculated for each interpolated vari- Fish diversity residual 301 0.013* 0.975* no 5 10 18.394 0.11 0.061 0.076 20, 5
37.929
kriging has been shown to outperform other interpolation
able to test for the presence of significant spatial autocorrela- (16.802)
Fish density 301 0.020** 0.969* No 5 8 0.0046 0.001 0.074 20,5
methods, such as inverse distance weighting (IDW) and trian-
tion using CrimeStat (Levine, 2002). Moran’s I is the standard ** indicates significance at p = 0.001, * indicates significance at p = 0.05
gulated irregular networking (TIN) (Guan et al., 1999). Before
autocorrelation statistic and provides a global (i.e. across the
kriging can be applied, two assumptions must be checked.
study area) test of spatial autocorrelation. Geary’s C is more
The first is stationarity; the mean (and ideally the variance)
sensitive to autocorrelation within small neighborhoods. Con-
a correlation exists, maps of diversity may simply reflect the birds. Areas of high marine bird diversity that overlap with low
must be constant across the spatial extent of the data. That
firmation of statistically significant spatial autocorrelation sug-
distribution of effort. In order to correct for differences in effort residuals should be interpreted with caution, as these hot
is, any large scale trend must be removed (see #2 above).
gests that point data are suitable for interpolation. As such,
across the study region, the following technique was applied: spots may simply reflect areas of unusually high effort. Since
The second assumption is isotropy of the variogram. The co-
interpolation was performed only where this was true for both
A second order polynomial regression of diversity on effort bird and fish density were only weakly correlated with effort,
variance between any two points is assumed to be a function
autocorrelation statistics.
was conducted and the residuals were interpolated as de- no attempt was made to correct the density maps.
only of the distance between the points, not of their location
scribed above. The interpolated map of residuals depicts ar-
or angle. This assumption can be examined and, if necessary,
2) Detrending: Detrending is done to ‘standardize’ the es-
eas of higher or lower diversity relative to that expected given ANALYTICAL MAP PRODUCTS
corrected for during the variogram modeling stage (see #3
timate across the analysis extent, and is a prerequisite for
the amount of local effort. This map was overlayed on the Spatial Statistics. Table 26 summarizes the results of
above). Trend analysis was conducted using JMP statistical
the interpolation procedure used here. After interpolation, the
interpolated map of diversity to visualize the impact of effort spatial autocorrelation tests, variogram fitting, and kriging
software (SAS Institute), while detrending, variogram mod-
removed trend is added back into the model results. Each in-
on the observed patterns in diversity. Although significantly cross validation. All variables were found to be significantly
eling, and kriging were conducted using the ArcView (GIS)
terpolated variable was plotted against Northing and Easting,
correlated with effort, fish diversity showed nearly identical positively spatially autocorrelated (p < 0.05) by both the
Geostatistical Analyst Extension (ESRI Inc.).
and a linear trend was fit to each plot. When significant trend
patterns as the map of fish diversity residuals. Fish diversity is Moran’s I and Geary’s C statistics, for fish and marine birds.
(p < 0.05) was present for either Northing or Easting, the data
therefore not shown with the overlay of residuals. Patterns of Spatial autocorrelation was more pronounced in the marine
The kriging neighborhood was set to the twenty nearest
were detrended (first order) before variogram modeling and
marine bird diversity, however, differed substantially from pat- bird data than in the fish data resulting in better model fit and a
neighbors with a minimum of five neighbors for each 90 de-
kriging.
terns of the bird diversity residuals, indicating that differences higher cross-validation r-squared value for the bird data sets.
gree angular sector for the fish data, and reduced to eight and
in effort are responsible for some of the observed pattern. Ma-
five for birds in order to capture small scale variability. Cross
3) Variogram Modeling: Empirical variograms show the de-
rine bird diversity hot spots, as represented by the top 20% of
validation was conducted to assess model accuracy by re-
crease in relatedness between pairs of points as a function of
diversity cell values, are therefore presented, along with an
gressing observed versus predicted values. Maps of the krig-
distance. In order to calculate the empirical variogram, pairs
overlay of the lower third of the diversity residuals for marine
ing standard error were also generated and used to restrict
of points must be binned by distance, and an average value
the analysis extent. In order to avoid unsupported extrapola-
(diversity, density) calculated for all pairs within a given bin.
tion into poorly sampled areas, the interpolated maps were
The size of the bin is referred to as the lag size. A variogram
clipped to remove areas of higher standard error. Interpolated
model is fit to the empirical variogram and its parameters
maps were clipped so that only grid cells for which the stan-
are later used in interpolation. Empirical variograms were
dard error was in the lowest 20% were used for subsequent
calculated using the default lag size and number, as well
display and analysis.
as for 1km, 5km, and 10km lag sizes. The appropriate lag
size and number of lags were chosen to optimize variogram
5) Correcting for Effort: Total effort was calculated as the to-
coherence. Directional variograms were then plotted to inves-
tal length of trawls falling within a grid cell for the NMFS trawl
tigate possible anisotropy not removed by detrending. Strong
data and as the total area surveyed within a grid cell for the
anisotropy was found only for the fish density data, and ac-
marine bird survey data. Although diversity is less related to
cordingly a geometrically anisotropic variogram model was fit
effort than other metrics, some significant correlation (p<0.05)
to this data set. Spherical variogram models were fit to the
between the two was found for both fish and birds. When such
empirical variograms. A spherical model was chosen based
Gulf of the Farallons National Marine Sanctuary
130
Section 3: INTEGRATION OF ANALYSES
feature acting as a biogeographic boundary, where oceanic and shelf species of
ABOUT THESE MAPS
birds show maximum overlap. The region seaward of the Farallon Islands displays
Figure 80a depicts interpolated marine bird diversity throughout the study region.
Density
Diversity high diversity not only because of its proximity to the shelf break, but also because
The top 20% of predicted diversity is bounded by a thin black line. Because bird
many species of birds breed on these islands and would not otherwise be found so
diversity was significantly correlated with survey effort, we have also provided a
High
High far offshore.
mask (cross hatched area) indicating where residual estimates provided evidence
that diversity was lower than expected given the amount of effort spent there (re-
Density. A large region of high (top 20th percentile) marine bird density exists ad-
siduals were among the lowest third). Interpret with caution in this area, as the
Low
Low
jacent to and shoreward of the marine bird diversity hot spot. This density hot spot
expression of high diversity under the mask may actually be an artifact of high
Top 20%
Top 20%
covers most of the shelf waters of all three sanctuaries, from Point Sur in the south
sampling effort. Figure 80b depicts interpolated bird density. Again, the top 20% of
Residual to midway between Bodega Head and Point Arena in the north. The density hot
this estimate is bounded by a thin black line. No statistical relationship was found
0 50 100
Mask spot extends into Monterey Bay. Major regions of overlap between marine bird di-
between density and effort; therefore, no residual mask is provided for this model.
Kilometers
versity and density occur along the shelf break. An additional density hot spot exists
Figures 80a and b have both been clipped using the standard error estimate for
off of Morro Bay to the south of the Monterey Bay NMS. There is some indication of
the interpolated surfaces (access these data on the CD-ROM). This was done to
high marine bird diversity in this region as well.
avoid unsupported extrapolation into poorly sampled areas. Figure 80c depicts the
top 20th percentile for diversity and density and the overlap between them.
A total of 60,000 square kilometers were modeled for bird density, with approxi-
a b mately 10,000 square kilometers classified in this analysis as a hot spot. Approxi-
DATA SOURCES
mately 28% of the entire modeled surface fell inside the boundaries of the three
R.G. Ford and J.L. Casey. 2003. CDAS Density Maps for Marine Birds off North/
National Marine Sanctuaries; however, 84% (8,962 km2) of the high marine bird
Central California, developed for NOAA’s National Centers for Coastal Ocean Sci-
Hot Spots density hot spot was found in the sanctuaries. The proportion of high density inside
ence. Portland, Oregon.
sanctuaries suggests that the boundaries include most areas of high density.
Top 20% Diversity METHODS
Summary. Patterns of bird diversity and density exhibited distinct spatial patterns,
See "Data and Analysis" section.
Top 20% Density
with diversity concentrated from the slope seaward, and density from the slope
shoreward. The overlap of these estimates mainly occurs along the shelf break; an
RESULTS AND DISCUSSION
Overlap of both
Diversity and Density area of high meso-scale bathymetric complexity. It is interesting to note that marine
Species Diversity. The interpolated maps of marine bird diversity show one con-
bird diversity exhibited a statistically significant positive correlation (r=0.33, p<0.0001)
tinuous area of high diversity along the continental slope, and, to a lesser extent,
0 25 50 100
with bathymetric variance (Figure 81a)(see section 2.1.1 for details on this estimate).
along the shelf between Point Arena and Point Sur. Within this area, diversity ap-
Density, on the other hand, exhibited a strong negative correlation (r=0.65, p<0.0001)
pears highest on, and seaward of, the Farallon Escarpment in the northwestern
Kilometers
with depth rather than bathymetric variance (Figure 81b).
corner of the Monterey Bay NMS (Pioneer Canyon), and off of the region between
Point Lobos and Point Sur (refer to locator map). Since marine bird diversity was
correlated with survey effort, much of the hot spot region coincides with areas of
high survey effort. The Farallon Escarpment, in particular, received a dispropor-
tionate amount of survey effort. When the map of interpolated residuals was over-
layed on marine bird diversity, some parts of the diversity hot spot (top 20%) fell in
a region of low (bottom third) residual diversity (the masked portion of Figure 79a).
This indicates that the high estimated diversity in the Farallon Escarpment is due,
at least in part, to high sampling effort. The portion of the marine bird diversity hot
spot between Point Lobos and Point Sur coincides with a region of high residual
diversity. This indicates that diversity in this region was both high and higher than
expected given relatively moderate sampling effort.
Overall, a total of 62,000 square kilometers were modeled for bird diversity. Of a b
that, roughly 12,000 square kilometers were classified as a hot spot (top 20% of
400
0 m estimated diversity). Approximately 28% of the entire modeled surface, and 58%
(7,158 km2) of the hot spot, fell inside the boundaries of the three National Marine
Sanctuaries. This disproportionate allocation of high diversity inside sanctuaries
50 m
Figure 81. The left graphic (a) shows the strong positive relationship observed be-
indicates that current boundaries generally incorporate areas of high regional di-
20 1000 m tween bird diversity and bathymetric variance, while the right graphic (b) shows a
c
20
30
00 versity. A considerable area of high diversity can be found seaward of the northern
0
00
strong negative relationship between bird density and depth. In both cases, the esti-
m
m
m
Monterey Bay NMS, and seaward of the entire Gulf of the Farallones and Cordell mate (diversity and density) have been classified into 20th percentiles.
Bank NMS boundaries. As mentioned in section 2.2, the persistence of high spe-
cies diversity along the shelf break may be attributed to this natural physiographic
Figure 80. Estimated diversity (a), density (b), and hot spots (top 20%) (c) for
marine birds.
131
Section 3: INTEGRATION OF ANALYSES
ABOUT THESE MAPS Density. Interpretation of the fish density maps suffers from the same problems
Figure 82a depicts estimated demersal fish diversity throughout the study region. (i.e. lack of data to the west of sanctuary boundaries and high spatial variability) as
Diversity Density Unlike the mean diversity mapped in section 2.1.1, this surface was generated us- those encountered for diversity. In addition, densities tend to emphasize the distri-
ing estimates of total diversity for each 5 minute grid cell. The top 20% of predicted bution of common numerically dominant species. High density areas of the map
High High diversity is bounded by a thin black line. Though fish diversity was significantly can be divided into four major hot spots (top 20%). One hot spot occurs on and to
correlated with survey effort in this model, high residual values overlapped areas the southeast of Cordell Bank. A second hot spot is found off of Point Reyes. The
of highest (top 20%) estimated diversity. This indicates that interpolated areas of largest density hot spot covers a large portion of the shelf to the north of Monterey
Low Low
highest diversity showed little effect of effort. As such, no residual mask is pro- Canyon, the entire area of Monterey Bay, and near shore waters south to Point Sur.
Top 20% Top 20% vided. Figure 82b depicts fish density, and, like the diversity map, is based on an Although portions of this hot spot are found over Monterey Canyon, this fact should
interpolation of total density (individuals per area swept (km2)) within each 5 min- be incorporated with caution since the deep canyon waters themselves were not
0 50 100
ute grid cell. The top 20% of this estimate is bounded by a thin black line. Figures sampled. The fourth hot spot is found to the south of Monterey Bay NMS and cov-
Kilometers
82a and 82b were clipped using the standard error estimates for the respective ers a substantial area of the shelf from Point Estero to Point Sal. This final hot spot
interpolated surfaces (access these data on the CD-ROM). This was done to avoid is the largest region of high fish density within the mapped area that falls outside of
unsupported extrapolation into poorly sampled areas. Figure 82c depicts the top Sanctuary boundaries and overlaps with a much smaller fish diversity hot spot to
20% for diversity and density, and the overlap between the two. the north.
DATA SOURCES A total of 27,000 square kilometers were modeled for fish density, with approximate-
Species diversity was calculated using NMFS shelf and slope trawl data collected ly 5,200 square kilometers classified in this analysis as a hot spot. Approximately
at depths between 50-1280 meters, between June and November, every third 54% of the entire modeled surface fell inside the boundaries of the three National
Marine Sanctuaries; however, 76% (4,041 km2) of the hot spot was contained
Hot Spots year from 1977-2001. For details on trawl methods see Lauth (2001), Shaw et al.
(2000), Turk et al. (2001), and Williams and Ralston (2002). within the sanctuary boundaries.
Top 20% of Diversity
METHODS Summary. Patterns in both fish diversity and density appear in many cases to be
Top 20% of Density See "Data and Analysis" section. linked to known oceanographic features already mentioned in previous sections. For
example, the northernmost diversity hot spot, and some parts of the density hot spot,
Overlap of both
RESULTS AND DISCUSSION straddle the shelf break, an area known to concentrate a variety of marine fauna (Kim,
Diversity and Density
Diversity. Interpretation of the interpolated maps of fish diversity is hindered by 2000; Adams et al., 1995; Yoklavich et al., 2000). The quickly changing depths of the
the lack of available data west of the sanctuary boundaries and the high spatial shelf break and slope may also increase diversity by allowing fish with overlapping
0 25 50 100
variability within the data. Despite these limitations, three hot spots (top 20%) of bathymetric preferences to coexist. Both diversity and density also appear high near
Kilometers
fish diversity are apparent: The northernmost hot spot is centered on Cordell Bank well known upwelling regions, including Point Sur, near Point Año Nuevo, and near
within the northwestern corner of the Cordell Bank NMS, and extends northward Cordell Bank. Although the majority of the fish diversity and density hot spots fall
along the continental slope outside of sanctuary boundaries to Point Arena. Its within sanctuary boundaries, this fact should be interpreted with caution since the
northern and western extent cannot be determined with the available trawl data sanctuary area represents approximately half of the mapped region for both of these
as sampling stopped along the edge of high predicted diversity. Extrapolation to variables. Areas of high diversity and density outside of the sanctuary boundaries
the north and west of this area indicates that high diversity may continue beyond exist to the north and south. Diversity and density to the west of sanctuary boundar-
the available data. A second area of high diversity is centered at the boundary ies cannot be adequately assessed with the available data.
between the Gulf of the Farallones NMS and the Monterey Bay NMS. The area
extends in a southeasterly direction past Point Año Nuevo and ends off northern
Monterey Bay. The southernmost hot spot is located between Point Sur and Lopez
Point and covers the inshore portions of Sur and Lucia Canyons. Portions of this
last hot spot, however, were poorly sampled. There is some evidence of an addi-
tional hot spot in the shallow waters (<200m) straddling the southern boundary of
the Monterey Bay NMS and extending into Morro Bay.
400
0 Overall, a total of 27,000 square kilometers were modeled for fish diversity. Of
m
that, roughly 5,400 square kilometers were classified as a hot spot (top 20th per-
centile of estimated diversity). Approximately 53% of the entire modeled surface
50 m
fell inside the boundaries of the 3 National Marine Sanctuaries, with 67% (3,675
20 100 0 m
20
30
km2) of the hot spot contained within the sanctuaries. Much of the remaining 33%
00
0m
00
m
m
of high diversity extends along the shelf break north of the Cordell Bank NMS to
Point Arena.
Figure 82. Estimated diversity (a), density (b), and hot spots (top 20%) (c) for fish.
132
Section 3: INTEGRATION OF ANALYSES
ABOUT THIS MAP tion of the Northern overlap of diversity that continues along
124°W 123°W 122°W 121°W
Figure 83 shows the overlap of diversity hot spots for birds the slope for 40-50 km beyond the northern boundary of
and fishes. As described previously, hot spots were defined
39°N
39°N
Cordell Bank National Marine Sanctuary.
Integration: Option 1 as the top 20% of diversity estimated through the spatial
modeling process (kriging). Also shown is the most recent Overlap in diversity hot spots occurs in both slope and shelf
estimated distribution of kelp beds within the study area. waters. The northernmost hot spot is clearly associated with
Species Diversity Although no specific analysis of biodiversity was done for the slope. The southernmost hot spot is also found in an area
kelp communities, it is well documented that these habitats of rapidly changing bathymetry off of Point Sur. A large portion
support a rich and diverse faunal assemblage (Abbott and of the central diversity hot spot, however, occurs over primar-
Hollenberg, 1976; VanWagenen, 2001; McLean, 1962; Fos- ily soft bottom shelf regions. The ecological linkages report
Legend ter and Schiel, 1985; Harrold et al., 1988; Thorson 1950; (see CD-ROM) cites a considerable volume of literature that
Randall 1965; Dayton 1984; Dean et al., 1984; Ebeling et al., describes slope communities as diverse, with well document-
Fish - Top 20% Diversity 1985; Harrold and Reed, 1985; Miller and Geibel 1973; King ed trophic interactions between birds and fishes. The authors
38°N
38°N
and DeVogelaere, 2000; Van Blaricom and Estes, 1988). report that spatial and temporal distribution of plankton is
Birds - Top 20% Diversity
Because of this, we have chosen to include kelp distributions thought to affect the distributions of many fishes and marine
in all of the integrated hot spot maps. The kelp distributions birds. In particular, marine birds aggregate in regions with ex-
Overlap of both
depicted here represent only a "snapshot" view of a highly tremely high plankton density, such as Cordell Bank, the Gulf
Fish and Birds
dynamic feature. of the Farallones, and parts of Monterey Submarine Canyon
Kelp Beds (1999) (Croll et al., in press). Each of these areas were identified in
DATA SOURCES this analysis as being biodiverse. Furthermore, squid, a pri-
0 10 20 40 60 80
Species diversity for fishes was estimated using NMFS shelf mary food source for numerous fishes and birds, concentrate
and slope trawls data collected at depths between 50-1280 in areas of high plankton productivity (Mais, 1972; Roper and
Kilometers
meters, between June and November, every third year from Young, 1975; Anderson, 1977; Pearcy et al., 1977; Anderson
37°N
37°N
1977-2001. For details on trawl methods see Lauth (2001), and Morel, 1978; Cailliet et al., 1979), where they consume
Shaw et al. (2000), Turk et al. (2001), and Williams and euphausiids and copepods (Karpov and Cailliet, 1979; Chen
Ralston (2002). Species diversity for birds was estimated us- et al., 1996). This provides further evidence that trophic set-
ing data provided by R.G. Ford Consulting and H.T. Harvey ting might be partially responsible for the expression of high
and Associates. 1999 kelp distribution data were provided by diversity in areas of upwelling.
California Department of Fish and Game.
Summary
METHODS 1) Diversity overlap between birds and fishes appear to be
See "Data and Analysis" section. correlated to known centers of coastal upwelling.
2) Overlap occurs in slope and shelf waters.
36°N
36°N
RESULTS AND DISCUSSION 3) Much of the expression of high diversity may be related to
All three regions of high fish diversity show some overlap the trophic setting in these areas rather than directly to the
with the regions of high bird diversity. An interesting result physical factors that characterize these areas.
of this analysis is that all regions of overlap occur near well
known upwelling centers (Huyer and Kosro, 1987; Brink
and Cowles, 1991; Kelly, 1985; Breaker and Mooers, 1986;
Breaker and Gilliland, 1981; Tracy, 1990; Schwing et al.,
1991; Breaker and Broenkow, 1994; Rosenfeld et al., 1994);
including the area surrounding Cordell Bank, the area south
of the Farallones (off point Año Nuevo), and directly adjacent
35°N
35°N
to point Sur. The northernmost fish diversity hot spot overlaps
the marine bird diversity hot spot from Cordell Bank north
400
to approximately midway between Bodega Head and Point
0 m
20 Arena. The seaward half of the central fish diversity hot spot
50 m
00
20
1000
m overlaps with the area of high marine bird diversity within the
m
0
30
m
00
Gulf of the Farallones NMS and the Monterey Bay NMS. The
m
northern half of the southernmost fish hot spot overlaps the
124°W 123°W 122°W 121°W
southern tip of the marine bird hot spot. There is a small por-
Figure 83. Integration option 1, diversity hot spots (top 20%) for fish and marine birds. Coastal kelp bed areas are also shown.
133
Section 3: INTEGRATION OF ANALYSES
Summary
ABOUT THIS MAP
124°W 123°W 122°W 121°W
1) There is considerable overlap between areas of high bird
Figure 84 shows the overlap of density hot spots for fish and
and fish density.
birds. As described previously, hot spots were defined as the
39°N
39°N
Integration: Option 2
2) Density maps should be interpreted with caution due to
top 20% of density estimated through the spatial modeling
their inherent biases toward numerically dominant species.
process. Also shown is the most recent estimated distribu-
tion of Kelp beds within the study area. Although no specific
Species Density analysis of density was done for kelp communities, it is well
documented that these habitats support a productive faunal
assemblage (Abbott and Hollenberg, 1976; VanWagenen,
2001; McLean, 1962; Foster and Schiel, 1985; Harrold et al.,
Legend 1988; Thorson, 1950; Randall, 1965; Dayton, 1984; Dean et
al., 1984; Ebeling et al., 1985; Harrold and Reed, 1985; Miller
and Geibel, 1973; King and DeVogelaere, 2000; Van Blaricom
Fish - Top 20% Density
38°N
38°N
and Estes, 1988). Because of this, we have chosen to include
Birds - Top 20% Density kelp distributions in all of the integrated hot spot maps. The
kelp distributions depicted here represent only a "snapshot"
Overlap of both
view of a highly dynamic feature.
Fish and Birds
DATA SOURCES
Kelp Beds (1999)
Species density for birds was estimated using data provided
by R.G. Ford Consulting and H.T. Harvey and Associates.
0 10 20 40 60 80
1999 Kelp distribution data were provided by California De-
Kilometers
partment of Fish and Game.
37°N
37°N
METHODS
See "Data and Analysis" section.
RESULTS AND DISCUSSION
Nearly all of the fish density hot spot is coincident with the two
areas of high bird density. The distributions for both metrics
are generally confined to the shelf (<200m) with the notable
exception of Monterey Canyon which appears as a density
hot spot for both groups. Although the majority of the hot spots
36°N
36°N
for fish and bird density fall within sanctuary boundaries, it is
notable that overlapping hot spots for both groups exist to the
south of Monterey Bay NMS. The pattern of marine bird den-
sity is dominated by the distributions of the Common Murre
(Uria aalge) and Sooty Shearwater (Puffinus griseus) be-
cause they are so abundant. Fish density reflects a somewhat
more balanced species composition. Among the most numeri-
cally dominant fish species are shortbelly rockfish (Sebastes
jordani) and Pacific hake (Merluccius productus).
35°N
35°N
Because the modeled distribution of bird density is dominated
by two species and all density maps emphasize common spe-
cies, these maps should be interpreted with caution. While
400
0 m
the density interpolation for birds closely approximates what
20
50 m
00
20
1000 is generally observed in the wild, it is heavily biased towards a
m m
0
30
m
few numerically dominant species. This fact may tend to over-
00
m
shadow the density distribution for rare and/or endangered
124°W 123°W 122°W 121°W
species.
Figure 84. Integration option 2, density hot spots (top 20%) for marine birds and fish. Coastal kelp bed areas are also shown.
134
Section 3: INTEGRATION OF ANALYSES
ABOUT THIS MAP north and south. The westward extent of important areas for
124°W 123°W 122°W 121°W
Figure 85 shows the overlap of options one and two. The top fish cannot be determined from the available trawl data, and
20% for bird diversity and density were combined, as were the may extend beyond the pictured hot spots. Since Option 3 is
39°N
39°N
Integration: Option 3 top 20% of fish diversity and density. This is the most inclusive simply a combination of Options 1 and 2, all of the concerns
view of marine bird and fish hot spots and the areas they over- and results for those two sections apply here as well.
lap. Also shown is the most recent estimated distribution of
Diversity and Density kelp beds within the study area. Although no specific analysis Summary
of biodiversity was done for kelp communities, it is well docu- 1) The sanctuary boundaries incorporate much of the highest
mented that these habitats support a rich and diverse faunal diversity and highest density areas within the region.
assemblage (Abbott and Hollenberg, 1976; VanWagenen, 2) Many of these biologically important regions coincide with
Legend 2001; McLean, 1962; Foster and Schiel, 1985; Harrold et al., known oceanographic and bathymetric features, such as up-
1988; Thorson, 1950; Randall, 1965; Dayton, 1984; Dean et welling regions, areas of high bathymetric variance, and the
Fish - Top 20% al., 1984; Ebeling et al., 1985; Harrold and Reed; 1985, Miller continental shelf break.
38°N
38°N
Diversity and Density and Geibel; 1973, King and DeVogelaere, 2000; Van Blaricom 3) Regions of high diversity and high density outside of the
and Estes, 1988). Because of this, we have chosen to include current sanctuary boundaries exist to the north, across much
Birds - Top 20%
kelp distributions in all of the integrated hot spot maps. The of the shelf and slope, and to the south, in near-shore wa-
Diversity and Density
kelp distributions depicted here represent only a "snapshot" ters.
Overlap of both view of a highly dynamic feature. 4) Uneven sampling effort across the study region and lack
Fish and Birds of trawl samples to the west of the sanctuary boundaries limit
DATA SOURCES the scope of any integrated biogeographic assessment.
Kelp Beds (1999)
Species diversity for fishes was estimated using NMFS shelf 5) Known limitations and biases of the two metrics (diversity and
and slope trawls data collected at depths between 50-1280 density) exist and are discussed elsewhere within this section
meters, between June and November, during every third (Section 3 – Integration).
0 10 20 40 60 80
37°N
37°N
year from 1977-2001. For details on trawl methods see Lauth
Kilometers
(2001), Shaw et al. (2000), Turk et al. (2001), and Williams
and Ralston (2002). Species diversity and density for birds
was estimated using data provided by R.G. Ford Consulting
and H.T. Harvey and Associates. Kelp distribution data were
provided by California Department of Fish and Game.
METHODS
See "Data and Analysis" section.
36°N
36°N
RESULTS AND DISCUSSION
The majority (71%) of the fish hot spot is coincident with the
much larger bird hot spot. The greater area of the bird hot spot
(~19,000 km2 for birds compared to ~10,000 km2 for fish) is
due to the greater spatial extent of the bird survey data. Major
areas of overlap occur in the following regions:
1) from Cordell Bank and the northwest corner of the Gulf of
the Farallones NMS north to approximately midway between
Bodega Head and Point Arena,
35°N
35°N
2) off Point Reyes,
3) shelf waters from the southern boundary of the Gulf of the
400
Farallones NMS south to Point Sur, including Monterey Bay,
0 m
and
20
50 m
00
20
1000
4) near shore waters off of Point Buchon.
m m
0
30
m
00
m
Although the majority of the regions that were identified as
124°W 123°W 122°W 121°W
hot spots for fish and birds occur within Sanctuary waters,
Figure 85. Integration option 3, diversity and density, hot spots (top 20%) for fish and marine birds. Coastal kelp bed areas are also there are hot spots beyond Sanctuary boundaries to the
shown.
135
Section 3: INTEGRATION OF ANALYSES
REFERENCES Chen, D. S., G. Van Dykhuizen, J. Hodge, and W. F. Gilly. Karpov, K. A., and G. M. Cailliet. 1979. Prey composition of Pearcy, W. G., E. E. Krygier, R. Mesecar, and F. Ramsey.
Abbott, I. A., and G. J. Hollenberg. 1976. Marine algae of Cali- 1996. Ontogeny of copepod predation in juvenile squid (Lo- the market squid, Loligo opalescens Berry, in relation to depth 1977. Vertical distribution and migration of oceanic micronek-
fornia. Stanford University Press. Stanford, California. ligo opalescens). Biological Bulletin 190(1): 69-81 and location of capture, size of squid, and sex of spawning ton off Oregon. Deep-Sea Research 24: 223-245.
squid. California Cooperative Oceanic Fisheries Investiga-
Adams, P. B., J. L. Butler, C. H. Baxter, T. E. Laidig, K. A. Croll, D. A., B. Marinovic, S. Benson, F. P. Chavez, N. Black, tions Reports 20: 51-57. Randall, J. E. 1965. Grazing effects on sea grasses by her-
Dahlin, and W. W. Wakefield. 1995. Population estimates of R. Ternullo, and B. R. Tershy. In press. From wind to whales: bivorous reef fishes in the West Indies. Ecology 46: 255-260.
Pacific coast groundfishes from video transects and swept trophic links in a coastal upwelling system. Kelly, K. A. 1985. The influence of wind and topography on the
area trawls. Fishery Bulletin 93: 446-455. sea surface temperature patterns over the northern California Roper, C. F. E., and R. E. Young. 1975. Vertical Distribution of
Dayton, P. K., V. Curries, T. Gerrodette, B. Keller, R. Rosen- slope. Journal of Geophysical Research 90: 11783-11798. Pelagic Cephalopods. Smithsonian Contributions to Zoology
Bakun, A. 1996. Patterns in the Ocean: Ocean processes thal, and D. Van Tresca. 1984. Patch dynamics and stability 209. 51 pages.
and marine population dynamics. California Sea Grant Col- of some southern California kelp communities. Ecological Kim, S. 2000. Patterns in Benthic Communities Within the
lege System and the National Oceanic and Atmospheric Monographs 54: 253-289. Sanctuary. Presented at the Monterey Bay National Marine Rosenfeld, L. F. Schwing, N. Garfield, and D. E. Tracy. 1994.
Administration. Sanctuary Symposium: Sanctuary Currents 2000: Causes of Bifurcated flow from an upwelling center: a cold water source
Dean, T. A., S. C. Schroeter, and J. Dixon. 1984. Effects of Ecosystem Change: Natural or Human? March 18, 2000. Co- for Monterey Bay. Continental Shelf Research 14: 931-964.
Barber, R. T., and F. P. Chavez. 1983. Biological consequenc- grazing by two species of sea urchins (Strongylocentrotus conut Grove, Santa Cruz, CA
es of El Niño. Science 222: 1203-1210. franciscanus and Lytechinus anamesus) on recruitment and Ryther, J. H. 1969. Photosynthesis and fish production in the
survival of two species of kelp (Macrocystis pyrifera and King, A., and A. DeVogelaere. 2000. Monterey Bay National sea. Science 166: 72-76.
Bolin, R. I., and D. P. Abbott. 1963. Studies on the marine Pterygophora californica). Marine Biology 78: 301-313. Marine Sanctuary final Kelp Management Report. Back-
climate and phytoplankton of the central coast area of Cali- ground, Environmental Setting, and Recommendations. Oc- Schwing, F. B., D. M. Husby, N. Garfield, and D. E. Tracy.
fornia, 1954-1960. California Cooperative Oceanic Fishery Ebeling, A. W., D. R. Laur, and R. J. Rowley. 1985. Severe tober 3, 2000. 1991. Mesoscale oceanic response to wind events off central
Investigations Report 9: 23-45. storm disturbances and reversal of community structure in a California in spring 1989: CTD surveys and AVHRR imagery.
southern California kelp forest. Marine Biology 84: 287-294. Lauth, R. R. 2001. The 2000 Pacific west coast upper California Cooperative Oceanic Fisheries Investigations Re-
Breaker, L. C., and R. P. Gilliland. 1981. A satellite sequence continental slope trawl survey of groundfish resources off ports 32: 47-62.
of upwelling along the California coast. In Richards, F. A. Foster, M. S., and D. R. Schiel. 1985. The ecology of giant Washington, Oregon and California: estimates of distribution,
[Ed.]. Coastal Upwelling. American Geophysical Union. kelp forests in California: a community profile. U.S. Fish and abundance, and length composition. NOAA Tech. Memo. Shannon, C.E., and W. Weaver. 1949. The mathematical
Washington, D. C. Pp. 97-94. Wildlife Service Biological Report 85(7.2): 152 pages. NMFS-AFSC-120. 284 pp. theory of communication. University of Illinois Press, Urbana,
IL 125p.
Breaker, L. C., and C. N. K. Mooers. 1986. Oceanic variability Gotway, C.A, and L. J. Young. 2002. Combining incompatible Levine, Ned. CrimeStat II: A Spatial Statistics Program for
off the central California coast. Progress in Oceanography spatial data. Journal of the American Statistical Association the Analysis of Crime Incident Locations (version 2.0). Ned Shaw, F.R., Wilkins, M.E., Weinberg, K.L., Zimmermann, M.,
17: 61-135. 97(458): 632-648. Levine and Associates, Houston, TX, and the National Insti- and R. R. Lauth. 2000. The 1998 Pacific west coast bottom
tute of Justice, Washington, DC. May 2002. trawl survey of groundfish resources: estimates of distribu-
Breaker, L. C., and W. W. Broenkow. 1994. The circulation Guan, W., Chamberlain, R.H., Sabol, B.M. and P.H. Doering. tion, abundance, and length and age composition. U.S. De-
of Monterey Bay and related processes. Oceanography and 1999. Mapping submerged aquatic vegetation with GIS in the Ludwig, J.A., and J.F. Reynolds. 1988. Statistical ecology: partment of Commerce, NOAA Technical Memorandum
Marine Biology 32: 1-64. Caloosahatchee Estuary: Evaluation of different interpolation A primer on methods and computing. John Wiley and Sons. NMFS-AFSC-114, 138 p.
methods. Marine Geodesy 22(2). 69-92. New York, NY.
Brink, K. H., and T. J. Cowles. 1991. The coastal transition Thorson, G. 1950. Reproductive and larval ecology of marine
zone program. Journal of Geophysical Research 96: 14637- Harrold, C., and D. C. Reed. 1985. Food availability, sea Mais, K. F. 1972. California Department of Fish and Game bottom invertebrates. Biological Review 25: 1-45.
14647. urchin grazing and kelp forest community structure. Ecology Sea Survey Cruises 1971. California Cooperative Oceanic
66(4): 1160-1169. Fisheries Investigations Data Reports 21: 1-132. Tracy, D. E. 1990. Source of cold water in Monterey Bay ob-
Cailliet, G. M., K. A. Karpov, and D. A. Ambrose. 1979. Pe- served by AVHRR satellite imagery. Masters Thesis. Naval
lagic assemblages as determined from purse seine and large Harrold, C., J. Watanabe, and S. Lisin. 1988. Spatial variation Malone, T. C. 1971. The relative importance of nannoplankton Postgraduate School. Annapolis, Maryland. 125 pages.
midwater trawl catches in Monterey Bay and their affinities in the structure of kelp forest communities along a wave expo- and netplankton as primary producers in the California Cur-
with the market squid, Loligo opalescens. California Coopera- sure gradient. Marine Ecology 9(2): 131-156. rent system. Fisheries Bulletin 69: 799-820. Turk, T. A., T. Builder, C. W. West, D. J. Kamikawa, J. R. Wal-
tive Oceanic Fisheries Investigations Reports 20: 21-30. lace, R. D. Methot, A. R. Bailey, K. L. Bosley, A. J. Cook, E.
Huyer, A., and P. M. Kosro. 1987. Mesoscale surveys over the McLean, J. H. 1962. Sublittoral ecology of kelp beds of the L. Fruh, B. H. Hormess, K. Piner, H. R. Sanborn, and W. W.
Chavez, F. P. 1995. A comparison of ship and satellite chlo- shelf and slope in the upwelling region near Point Arena, Cali- open coast area near Carmel, California. Biological Bulletin Wakefield. 2001. The 1998 Northwest Fisheries Science Cen-
rophyll from California and Peru. Journal of Geophysical Re- fornia. Journal of Geophysical Research 92: 1655-1681. 122: 95-114. ter Pacific west coast upper continental slope trawl survey of
search 100: 24855-24862. groundfish resources off Washington, Oregon, and California:
Kaluzny, S.P., Vega, S.C., Cardoso, T.P., and A.A. Shelly. Miller, R. J., and J. J. Geibel. 1973. Summary of blue rockfish Estimates of distribution, abundance, and length composition.
Chavez, F. P. 1996. Forcing and biological impact of onset of 1998. S+SpatialStats: Users Manual for Windows and UNIX. and lingcod life histories: a reed ecology study; and giant kelp U. S. Department of Commerce, NOAA Technical Memoran-
the 1992 El Niño in central California. Geophysical Research MathSoft Inc. Macrocystis pyrifera experiments in Monterey Bay, California. dum NMFS-NWFSC-50, 122 p.
Letters 23: 265-268. Fish and Game Fishery Bulletin 158: 1-137.
136
Section 3: INTEGRATION OF ANALYSES
Van Blaricom, G. R., and J. A. Estes [Eds.]. 1988. The Com-
munity Ecology of Sea Otters. Springer-Verlag. Berlin, Ger-
many.
Van Wagenen, R. F. 2001. California Coastal Kelp Resources,
Summer 2000. Final Report to the Monterey Bay National Ma-
rine Sanctuary.
Williams, E.H. and S. Ralston, Distribution and co-occur-
rence of rockfishes over trawlable shelf and slope habitats
of California and southern Oregon. Fisheries Bulletin, 2002.
100: p. 836-855.
137
Section 4: DATA SOURCES AND GAPS
INTRODUCTION Within this library of data, some sets emerged as
Table 27. Matrix of data sets and their associated characteristics that were used or referenced in the biogeographic assessment.
This section addresses a secondary objective primary data sources, while others contributed
Source Org Strengths of Data for Biogeographic Constraints of Data for Biogeographic
Data Set Target Info. Dates Samples Depth Range
of the project: the acquisition and assessment to the project in terms of providing contextual
and/or PI Assessment Assessment
of available comprehensive data for the study information, reference or validation.
NMFS Long time-series, wide spatial extent, Sampling only in summer season, some
Triennial Shelf Fish, (Alaska & 1977-1998 every 3 fairly good depth range, abandoned trawls suspected of being off-bottom,
area and the identification of data gaps in such n=994 55-500m
Trawl Data Invertebrates Northwest years, June-Aug only trawl areas suggest areas of high rocky areas undersampled due to threat
information. In addition, suggestions are made The data sets that ultimately proved most use-
FSC) rugosity of gear damage.
for prioritizing future research efforts to generate ful in this undertaking are summarized in Table
1991, 1997, 1999,
Slope Trawl Fish, 10+ years of data, wide spatial Sampling for only 5 months (July-Nov.);
data that would be especially valuable for future 27 below, which provides information about the
NMFS 2000, 2001, July-Nov n=454 190-1280m
Data Invertebrates extent, good depth range identification of common inverts only
only
biogeographic analyses. source of the data, target information, dates of
NMFS
Rockfish NMFS / 1986-2001, May-June 6-32m Sampling only in May & June, along collection, number of samples, and depth range
n=1548 1500+ tows
Midwater Trawl
juveniles SWFSC only mean=26m transects, targets juv. rockfish
Throughout the project, members of the Bio- when available. Additionally, comments about
Data
provides information about
geography Program contacted numerous the general strengths and constraints of the data
Recreational 1987-1998, 2-360 fm (3-650 No effort data (presence/absence only),
Rockfish CDF&G n=4357 nearshore areas, wide range of
Fish Data continuous m) effort targets Rockfish only
academics, scientists, and agency personnel sets in the context of this analysis are noted in
depths.
who were likely to have knowledge of data separate columns. Full citation information for
1983-1995 (Sonoma)
provides year-round look at kelp in
relevant to the study area, compiling a contact each data set is provided in the list of references
Kelp- year-round but to max extent of
Tom Laidig/ Sonoma 1983-95, quick look at Sampling is sparse, presence/absence
Laidig Data Associated sparse, 1984, 1997, n=43 surveys scuba or kelp
list of over 160 people. Additionally, staff con- that appears at the end of each section utilizing
NMFS Monterey in 84, 97, 2001; data only, very limited spatial extent
Species 2001 (Monterey) May- (<130 ft.)
differentiates juveniles and adults.
sulted the impressive compendium of studies the data.
Oct only
compiled by Monterey Bay Sanctuary staff for n=117,176 Variable reliability re: locations of fishing,
Commercially
their Sanctuary Integrated Monitoring Network CONCLUSIONS AND RECOMMENDATIONS
Commercial 1988-2000, year- (records grouped shoreline to All gear types, data can be sorted by summarized to 10-minute grids (large
Valuable CDF&G
Fishing Data round by species, not trip 4810m gear, long time series. scale), fisheries-dependent, can't sort by
(SIMoN) program, which provides a “blueprint for FOR FUTURE ACTIVITIES
Fishes
or boat) boat or trip/effort
a comprehensive, integrated monitoring network During the course of this project, Biogeography
5 categories, provides the most
to detect natural and human induced changes Program staff gained a unique familiarity with
comprehensive view of benthic
to the Monterey Bay National Marine Sanctuary the data available and the data necessary to
substrates for the study area; in
some places very detailed, high
and its resources” (http://montereybay.noaa.gov/ undertake the analysis. This position enables
resolution; Original data consisted of
Greene et. al.,
). Through extensive consultation with contacts, project team members to make observations
sampling dates seismic-reflection profiles and Based on surface extrapolation of point
Substrate National Sea shoreline to
Sediment unknown, received unknown number sediment/rock sample data collected data, most of the map is low resolution
approximately 62 data sets were investigated about the types of data that, if acquired, could
Composition Grant College ~3500m
12/2002 by California Division of Mines and 1:250,000.
for incorporation into this study. Data sets were improve the biogeographic assessment in the
Project
Geology, USGS, and California
considered in terms of sampling objective, the future. Some of the data sets may exist, but in
Coastal Commission. New data
include multibeam data from MBARI,
extent of their spatial and temporal coverage, the a format that could not, in their current state,
and Center for Habitat Studies at
existing format and ability to be converted into a be incorporated into the project. For example,
Moss Landing Marine Laboratories
GIS layer, the utility of the data compared to the historical benthic infauna data exists, but has not
Bathymetry- Depth, shoreline to best available at the start of the
CDF&G unknown unknown Medium resolution
work involved in its incorporation into the project, yet been converted to digital format or updated
200m Topography 4810m project
provides higher resolution which
and whether or not the Biogeography Program to reflect current taxonomy. Several important
NOAA /
Bathymetry- Depth, shoreline to Better resolution, late availability for this
increases ability to identify smaller
was granted access to the data. In general, team data sets do not, to our knowledge, exist at all
NGDC & unknown unknown
30m Topography 4810m project
areas of high bathymetric variance,
MBARI
members acquired only accessible data sets that or at the necessary spatial scale, but should be
ie. pinnacles and drop-offs
Polygons generated from aerial
had broad spatial extents covering a significant considered priorities for future analytical efforts.
many, polygon Changes since 1999, missing sections,
Kelp Data Kelp Location CDF&G 1989, 1999 surface photos provide literal 'snapshot' of
data doesn't differentiate species
portion of the study area, a large number of They appear in the following list.
kelp along long stretch of coast.
samples that could be georeferenced, and high Monthly composites will smooth out
Sea Surface Sea Surface NOAA / Monthly composites available for
important temporal fluctuations, ie. short
confidence in data quality. •Resolution. Finer thematic and spatial resolu-
monthly composites satellite data surface
Temperature Temperature Coastwatch years 1992 - current. upwelling and relaxation events; corrected
tion on the substrate and bathymetric maps will
for best surface.
Data sets that met the above criteria were be highly advantageous to the analyses based
Dohl, 1,057 cells visited; Data from early 1980s may not represent
surface survey Relatively large spatial coverage;
MMS High Minerals 76,888km of current status and distribution of species;
requested and obtained if possible. Once in- on Habitat Suitability Modeling. Improving the
Cetaceans & 1980-1983, in all three of the shelf, cost-effective, year-round synoptic
Altitude Aerial Management trackline; 10,014 high altitude surveys may not provide
turtles ocean seasons slope and deep surveys for cetaceans over the shelf
house, data were further evaluated in how well resolution for the bathymetry map will help
Surveys Service cell-study-day good characterization of smaller, less
ocean beyond and slope.
(MMS) visits visible species.
they served the objectives of the study, and the identify more small areas of high relief, such as
most useful data were synthesized into a working rocky pinnacles, that are known to be important
Bonnell-PI for 870 cells visited; surface survey Relatively large spatial coverage; Data from early 1980s may not represent
MMS Low
Marine Birds mammals, 1980-1983, in all three 70,114km of of the shelf, cost-effective, year-round synoptic current status and distribution of species;
GIS library. This involved conversion into GIS habitat for some species.
Altitude Aerial
and Mammals Briggs-PI for ocean seasons trackline; 9,306 cell-slope and deep surveys for species over the shelf low altitude surveys may not provide good
Surveys
format, standardization of geographic projection, birds; MMS study-day visits ocean beyond and slope. characterization of rare cetacean species.
and when possible, the aggregation of smaller •Spatial Data for Additional Species. Because
76 cells visited;
EPOCS surface survey Includes outer Calif Current surveys; Spatial coverage is not as robust as with
data sets into a master data layer. of sampling limitations (i.e. mesh or hook size)
Marine Birds 1984-1994, in all three 1,033km of
Shipboard Ainley of the deep better species sightability on a ship the aerial surveys, but sightability for
and Mammals ocean seasons trackline; 77 cell-
or a lack of published life history information,
Surveys ocean than an airplane. some species is better.
study-day visits
some ecologically important species are not
represented in this study. More information
138
Section 4: DATA SOURCES AND GAPS
•Location Verification for Fisheries Data. Verification of the spa- the resource could be used to couple estuarine, coastal and Yoklavich, M.M., G.M. Caillet, R.N. Lea, H.G. Greene, R.M.
should be collected on these species through other means.
tial reporting from commercial fishing logbooks would enable marine ecosystems. Starr, J. de Maringnac, and J. Field. 2002. Deepwater habitat
One example is the pygmy rockfish (Sebastes wilsoni), a small
the full incorporation of this valuable data source. and fish resources associated with a marine reserve: impli-
but abundant species, that does not show up in any of the
•Wider Regional Context. Expansion of the scope of the study cations for marine fisheries in marine ecological reserves
fisheries-dependent data sets. S. wilsoni can be surveyed via
•Abandoned Trawls. Incorporation of NMFS’s study of aban- to include the biogeography of the entire west coast of North research program research results. 1996-2001. California
submersible, as documented in various studies (Yoklavich et. al,
doned trawl locations from the Triennial Surveys. Such infor- America to better understand how the north/central California College Sea Grant Program CD-ROM. La Jolla, CA. 63 pp.
2002; Yoklavich et. al., 2000; Hixon et al., 1991). Other species
mation indicates areas that are difficult to trawl or ‘untrawlable’ region fits into the wider biogeographic context. A precedent
of interest include white shark, pelagic fishes, intertidal species,
due to the fact that the nets repeatedly became caught or torn exists in the West Coast Atlas produced by NOAA’s SEA Divi- Yoklavich, M.M., H.G. Greene, G.M. Caillet, D.E. Sullivan,
krill, marine birds, marine mammals and sea turtles.
during trawl attempts. These locations may indicate areas with sion. Such a document could serve as a blueprint for defining R.N. Lea, and M.S. Love. 2000. Habitat associations of deep-
rocky substrates and high rugosity, which, though still targeted species distributions along the west coast of the continental water rockfishes in a submarine canyon: an example of a
•Survey Methods. Sampling fish communities using a consistent
by hook and line and recreational fishers, are generally pro- U.S. (SAB/NWAFS 1988, SAB 1990). natural refuge. Fish Bulletin, U.S., Vol. 98, pp. 625-641.
sampling method over all substrate/habitat types, based on
tected from trawl fishing methods.
stratified random sampling. Multiple survey methods should be
REFERENCES
employed to ensure representation of important fish species
•Oceanographic Influences. Incorporation of more oceano- Emmett, R.L., S.L. Stone, S.A. Hinton, and M.E. Monaco.
that are not susceptible to current sampling methods.
graphic features and parameters into the analyses, especially 1991. Distribution and abundance of fishes and invertebrates
for birds and mammals. Ephemeral features such as currents, in west coast estuaries, Volume II: Species life history
•Sampling Strategies. Sampling strategies should be tailored
the San Francisco plume, and sources of upwelling could be summaries. ELMR Rep. No. 8. NOAA/NOS/Strategic
to include more life history stages of fish, especially larval
represented in probability maps or by aggregating empirical Environmental Assessments Division. Rockville, MD. 329 pp.
stages.
data by various temporal categories (e.g. by week, event,
month, season, warm/cold period, etc.). Hixon, M.A., B.N. Tissot, and W.G. Pearcy. 1991. Fish As-
•Sampling Strategy. Sampling should also be better spread
semblages of Rocky Banks of the Pacific Northwest [Coquille,
spatially and temporally to:
•Kelp Surveys. Increased frequency of surveys to better moni- Daisy, and Heceta Banks]. A final report by the Department
tor changes in kelp distribution; differentiation between kelp of Zoology and College of Oceanography of Oregon State
•Reduce Effort Disparity. Increased sampling in certain ar-
species. University for the U.S. Department of the Interior, Minerals
eas would help equalize the distribution of sampling effort
Management Service Pacific OCS Office. Contract No. 14-12-
across the study area. Some analyses were confounded
•Life History/Trophic Linkages. Expanded knowledge of life 0001-30445. Camarillo, CA. 410 pp.
by the wide range of sampling effort.
history characteristics, habitat affinities, distribution and
abundance of pelagic prey species, and links between preda- Monaco, M.E., R.L. Emmett, D.M. Nelson, and S.A. Hinton.
•Target Important and Under-Sampled Areas. Increas-
tors and prey species (i.e. hake, krill, and plankton) will help 1990. Distribution and abundance of fishes and invertebrates
ing sampling in important areas that are currently under
describe distributional changes based on trophic linkages and in west coast estuaries, Volume I: Data summaries.
sampled (e.g. the entire near-shore region and the slope
foraging behavior. ELMR Rep. No. 4. NOAA/NOS/Strategic Environmental
area west of Cordell Bank) or in areas of particular man-
Assessments Division. Silver Spring, MD. 332 pp.
agement interest (e.g. boundary regions) would help to
•Life History/Spawning Areas. Incorporation of known spawning
better characterize these areas. The techniques used in
areas will help identify important areas and seasonal changes Nelson, D.M, and M.E. Monaco. 2000. National overview
the integration section can accommodate preferential (i.e.
in distribution for fish. and evolution of NOAA’s Estuarine Living Marine Resources
non-random) sampling in areas of interest.
(ELMR) Program. NOAA Tech. Memo. NOS NCCOS CCMA
•Data QA/QC. Data quality assurance would allow the incor- 144. Silver Spring, MD. 60 pp.
•Describing Effort. Increase consistency in recording effort-re-
poration of some existing data sets that were discarded due to
lated parameters for fisheries trawls (e.g. recording start and
inconsistencies in species coding (e.g. the fisheries data set SAB (Strategic Assessment Branch). 1990. NOAA’s west
end coordinates of trawls) and naturalists’ surveys.
targeting salmon), taxonomic changes (e.g. benthic infauna coast of North America data atlas: Invertebrate and fish
data) or other reasons. volume. National Oceanic and Atmospheric Administration.
•Expert Knowledge. Incorporate additional expert knowledge
Rockville, MD. 111pp.
and data from the fishing community, naturalists (e.g. Mon-
•Expansion of Scope. The scope of this analysis could be
terey Bay Whale Watch cruise data), and recognized experts,
broadened to include adjacent habitats. For example, the in- SABNAFC (Strategic Assessment Branch, and Northwest
especially for areas and time periods where there is little or
teraction between marine and estuarine habitats could possibly and Alaska Fisheries Center). 1988. West coast of North
no data.
be addressed using network analysis of energy flows between America strategic assessment: data atlas: Marine mammal
ecosystems. As part of the Estuarine Living Marine Resources volume. National Oceanic and Atmospheric Administration.
•Data Compatibility. Achieve consensus on the best way to
(ELMR) series, a 2-volume comprehensive data base on the Rockville, MD. 33 pp.
merge aerial and ship-based survey data for birds and mam-
distribution of estuarine fishes and invertebrates in West Coast
mals.
estuaries was completed in 1990-91 (Emmett, et al., 1991;
Monaco, et al., 1990; Nelson and Monaco, 2000). If updated,
139
Section 5: SUMMARY OF BIOGEOGRAPHIC ASSESSMENT
BACKGROUND Ideally, biogeographic assessments utilize significant amounts
The mission of NOAA National Ocean Service’s (NOS) National of data that have been collected over the entire spatial extent of
Marine Sanctuary Program (NMSP) is to serve as the trustee the study area over a long time period. However, such a wealth
for a system of marine protected areas, to conserve, protect, of data is rarely available. In many instances, little information
and enhance their biodiversity, ecological integrity, and exists to accurately characterize the study area or associated
cultural legacy. To assist in accomplishing this mission, the living marine resources. This paucity of comprehensive data
NMSP has developed a partnership with NOAA’s National can limit the efficacy of biogeographic assessments, but
Centers for Coastal Ocean Science (NCCOS) to conduct additional analytical methods can be employed to complement
biogeographic assessments of living marine resources in all the assessment. In addition to analysis of databases, two
National Marine Sanctuaries to characterize and assess the additional tasks were used to conduct the assessment.
distribution of marine resources that occur within and adjacent First, a synthesis of existing information was compiled and
to the sanctuaries. The NMSP and NCCOS’s Biogeography presented in the Ecological Linkages Report to incorporate
Program have developed a five-year plan to implement the qualitative information about species, habitats and ecological
assessments across the system of National Marine Sanctuaries characterization of marine ecosystems and linkages within
(Kendall and Monaco 2003). The biogeographic assessment the study area. Second, species habitat suitability modeling
process as defined in the plan is used to conduct studies that efforts were conducted for fishes to define potential species’
are designed to address research needs and support a wide distributions based on known habitat affinities and physiological
array of sanctuary management decisions. In general, the limitations. The potential species distribution maps are displayed
priority to implement the biogeographic assessments is based as a series of digital maps found on the CD-ROM.
on the need to update sanctuary management plans. Thus, the
joint efforts are systematically proceeding to work with each In addition, a critical component of the assessment process was
sanctuary to provide assessments of species’ distributions and the extensive effort to have the data, analytical approaches,
their associated habitats in a region. and results peer reviewed. Initial results from the suite of
biogeographic analyses were presented to experts familiar
Since establishment, many of the sanctuaries have witnessed with the marine ecosystem off north/central California, as
increased pressure on marine resources from natural and well as to the originators of the data sources, in an attempt
anthropogenic phenomena, including climatic variation and to improve the analyses. The role of expert review and input
degradation of habitats. In order for the NMSP to increase was considerable, and the contributions made by experts have
management capabilities, it is imperative that the spatial and to the boundaries of three contiguous West Coast National beyond the limits of current sanctuary boundaries to place significantly enhanced the assessment. In June 2002, project
temporal distributions of biota and habitats within sanctuaries Marine Sanctuaries. These sanctuaries, Monterey Bay, Gulf study results in the context of north/cental California Coast team members traveled to Seattle, WA and Santa Cruz, CA to
be delineated. Biogeography provides a framework to integrate of the Farallones, and Cordell Bank, are conducting a joint biogeographic patterns. The biogeographic analyses are based discuss and present the results of the Interim Product to West
species distributions and life history data with information on review to update sanctuary management plans. To support on a synthesis of many data sources that were provided by Coast experts (NOAA, 2002). Suggestions were incorporated
the habitats of the region to characterize marine resources the management plan review process, the Biogeography project partners and contributors. Results of this assessment and a Web site was created to further disseminate analytical
in a sanctuary. When the biogeographic data are integrated Program is leading a partnership effort to conduct a robust are being used to assist the NMSP in addressing issues such as products prior to an additional series of meetings. The final
into a Geographic Information System (GIS), it enables users analytical assessment to define important biological areas and evaluating potential modification of sanctuary boundaries and suite of review meetings was held in October 2002 in San
to visualize species’ spatial and temporal distributions and time periods within the region. This document represents the changes in management strategies or administration, based Francisco and Monterey, CA and in Seattle, WA. At that time,
conduct ecological forecasts to assess potential changes in results of the first of two assessment phases. Phase I provides on the principles of biogeography. NOS staff invited members of the scientific community to
species distributions that may result from a variety of natural data, analytical results, and descriptions of ecosystems and review the preliminary results of the biogeographic analyses.
and anthropogenic perturbations. In addition, based on specific their linkages; it also identifies data gaps, and suggests future The biogeographic assessment was formulated around three Comments from the October meetings were compiled and
ecological metrics (e.g., diversity), biologically significant activities to be addressed in Phase II. closely integrated study components: (1) an Ecological Linkages reviewed by project personnel, who either incorporated the
areas can be delineated. This document provides the results Report, (2) biogeographic analyses, and (3) development of GIS experts' suggestions or provided explanations as to why they
of the GIS-based assessment conducted for the National Phase I of this effort was a biogeographic assessment of existing data for incorporation into NMSP’s Marine Information System did not. Thus, the integration of the synthesis of ecological
Marine Sanctuaries off north/central California to initiate data on the distribution and abundance of marine fishes, marine (MarIS). The majority of the results from the assessment are linkage information, statistical analyses of existing databases,
development of a biogeographic assessment capability for birds, marine mammals and their associated habitats. The study presented as a suite of GIS maps to visually display species’ species habitat suitability modeling, and peer review, resulted
the sanctuaries. did not attempt to define biogeographic patterns along the entire biogeographic patterns across the study area. The body of the in this biogeographic assessment product.
U.S. West Coast nor in very near-shore environments (e.g., document provides examples of the entire suite of digital map
BIOGEOGRAPHIC ASSESSMENT OFF NORTH/CENTRAL estuaries). Rather, the study area was restricted to the marine products found on the companion CD-ROM and located on ECOLOGICAL LINKAGES REPORT
CALIFORNIA area from Point Arena in Mendocino County (38˚54’32” N, the the Web at http://biogeo.nos.noaa.gov/products/canms_cd/. Section 1 of the document presents a synopsis of the Ecological
The initial biogeographic assessment outlined in the five-year northern bound) to Point Sal in northern Santa Barbara County The spatial data and additional information, such as digital Linkages Report and provides the context to understand overall
NCCOS/NMSP plan was implemented in the spring of 2001 (34˚54’05” N, the southern bound). The entire study area and species distribution maps and additional details on analytical biogeographic product results, relative to the ecosystems
to conduct a 24-month investigation to assess biogeographic regional maps of the area are depicted in Figures 2-5. This methodologies, are also presented on the CD-ROM. along the California Coast. The bulk of the report describes
patterns of selected marine species found within and adjacent relatively large study area enabled the assessment to extend ecosystems in the region, key species associated with these
140
Section 5: SUMMARY OF BIOGEOGRAPHIC ASSESSMENT
• Diversity and richness can be used to delineate fish hot
ecosystems, and linkages between and among them. In on reviewers' comments on the Interim Product (NOAA 2002), species resulted in information on species’ habitat affinities
spots. While little variation in diversity and richness were
addition, the report presents latitudinal range distributions of feedback from technical review meetings, and peer review that were converted into quantifiable habitat suitability index
explained by depth (r2 =0.04 between both richness and depth
species groups, including algae, invertebrates, fish, marine workshops. Thus, a very difficult step in the project was to values (Monaco et al., 1997). The life history information and
and diversity and depth), trawls with high diversity tended to
birds and marine mammals. These maps provide an overview select and rely on the most appropriate data and analyses to associated species habitat suitability index values are found on
be deeper than trawls with high richness. Trawls with high
of marine species’ distributions and biogeographic transitions characterize the various components of the marine ecosystem the CD-ROM. These derived values were input into an equation
species richness of rockfish (Sebastes and Sebastolobos)
along the entire west coast of North America. The report also that exist in the study area. The inclusion of the GIS-based and used to predict potential species’ distributions based on
followed the 200-meter contour, which approximates the
includes important information on ecosystems not easily products on the companion CD-ROM will enable NOAA staff, their affinity for the mosaic of bathymetry and bottom habitats
break between the shelf and slope.
studied at this large scale via GIS, particularly near-shore advisory councils, and research partners to query data and found throughout the region. The species habitat suitability
communities. The complete report (163 pp.) is on the CD-ROM information relevant for questions and issues that are not models were validated through statistical and spatial analyses,
• Even though richness and diversity are correlated, the
that accompanies this document (Airamé et al., 2003). specifically addressed in this product. using fishery-independent survey data.
maps showed different results. Hot spots in either richness
or diversity were identified in all three sanctuaries. In Cordell
Key West Coast Biogeographic Transitions The first analyses focused on a suite of assemblages analyses • Bottom substrate and water depth were statistically significant
Bank NMS, there was a group of trawls with high richness
• Benthic algae exhibit three major biogeographic transitions to assess the biogeography of fishes and a few macro- variables used to predict the potential distribution of species
near the center of the sanctuary, and another group of
at Point Conception, Puget Sound, and the Gulf of Alaska. At invertebrates. Primary data included fisheries-independent based on their habitat affinities.
trawls with high diversity in the region around the northwest
all latitudes, the average number of algal species increased data, such as those collected by researchers from the National
boundary. There was also a large collection of trawls with
with depth from high to low intertidal and subtidal zones. Marine Fisheries Service (NMFS), and fisheries-dependent • Habitat suitability models for an assemblage of rockfish
either high richness or diversity straddling the boundary
data, such as those collected by the California Department of were developed and results indicated that rocky habitats
between Gulf of the Farallones and Monterey Bay NMS.
• Five major transitions occur in distributions of marine Fish and Game (CDF&G) for recreational fisheries. These data located on the shelf were identified as potential hot spots
There were lines of high diversity along the 200-meter depth
invertebrates found in California waters: at Point Conception, sets, although not spatially or temporally comprehensive, are for adults; whereas mud and sand substrates on the shelf
contour north of CBNMS boundary and from Lopez Point
Monterey Bay, Puget Sound, and off the coasts of British the most robust data sets that exist for the entire region, and were delineated as potentially important habitats for subadult
south to the southern edge of the study area.
Columbia and southeastern Alaska. At all latitudes, greater provide considerable information on the distribution of several rockfish.
numbers of gastropod species occur in the euphotic zone and hundred fish and invertebrate species.
• Starr (1998) addressed the implementation of rockfish no-take
on the continental shelf than on the continental slope. • Map overlays of all species’ HSI models resulted in the
areas with two important recommendations. First, in order
Key Assemblage Analysis Results for Fishes delineation of a broad range of important areas that cover
to properly manage marine ecosystems, fish assemblages
• Pacific coast fishes exhibit two major biogeographic transitions. • Species assemblages and site groups were distinguished the majority of the continental shelf within and adjacent to
must be better understood. Starr stated that once these
A biogeographic transition at the Bering Sea is relatively through 1-Pearson correlation coefficients with average the three sanctuary boundaries.
assemblages are delineated, steps can be taken to ensure
abrupt, corresponding to the northern limit of distributions means clustering technique. Species assemblages from
that each assemblage receives proper management.
of over 100 fish species. A broader biogeographic transition CDF&G recreational, NMFS shelf, and NMFS slope data Key Marine Bird Analytical Results
This study defined assemblages of fishes for near-shore,
occurs along the southern coast of California between Baja sets were more resilient than assemblages from the NMFS The Biogeography Team contracted principal investigators
shelf, slope and midwater ecosystems. The results of the
California and Point Conception. A few minor shifts in fish midwater data set, emphasizing the ephemeral nature David Ainley and Glenn Ford (of H.T. Harvey and Associates
community metrics and species assemblages are displayed
species composition occur between Point Conception and of the midwater environment and the smaller midwater and R.G. Ford Consulting Co.) to work with the NOAA project
in this document as a series of maps and tables.
the Bering Sea, particularly at Monterey Bay. data set. The site groups were displayed spatially in a team to define and assess biogeographic patterns and
GIS and the average frequency of occurrence of species important areas for marine birds (and mammals) found within
• The second recommendation by Starr (1998) was to delineate
BIOGEOGRAPHIC ANALYSES assemblages was calculated to show the interaction between the study area. These experts used regression analysis, GIS
rectangular no-take areas that cover 20-50 km of the coast
Section 2 introduces the methods used to conduct the species assemblages and site groups (i.e., where species and over eight spatial data sets to develop over 50 maps that
and extend west to the edge of the continental shelf. From a
assessment and the results of the biogeographic analyses assemblages were caught). display marine spatial and temporal patterns, and estimated
biogeographic viewpoint, the results of the spatial analyses
of selected marine biota off the north/central California coast. densities and diversity for selected marine birds in the study
coincided with that recommendation, and also identified that
This component of the assessment is the cornerstone of the • The interaction of the site groups with environmental area. The resulting maps and discussion summarize important
deep-slope communities significantly contribute to ground
overall biogeographic product to support the NMSP joint parameters that were not used to create the groups can locations, time periods, and life history information for marine
fish biogeographic patterns. Because assemblages follow
management plan review process. The data, analyses, and be informative about what conditions are affecting species birds in the study area. Phase II of the assessment may include
bathymetry at the scale of this analysis, setting aside an area
supporting information are linked using statistical and GIS tools distribution. Depth was highly significant between site a technical report on the methods and results summarized in
from the coast through the continental slope could protect all
to portray in space and time significant biological areas or “hot groups in all data sets, emphasizing the importance of the Phase I map and tabular products.
demersal species assemblages identified in this study.
spots.” The term “hot spot” is defined based on specific criteria depth in structuring marine biological communities. Analyses
or metrics (e.g., species diversity, high species abundance). comparing the site groups to other environmental parameters • In general, the marine birds of the three sanctuaries are
Key Species Habitat Suitability Model Results
The vast majority of the analytical results are displayed as a (latitude, sediment size, and bathymetric complexity) were dominated in number and biomass by seasonally resident,
Due to limitations in the spatial and temporal extent of data and
series of maps to identify biologically significant areas in the inconclusive, as these parameters often had significant nonbreeding species, such as sooty shearwater, pink-footed
to complement the assemblage analyses of fishes, species
study area. interactions with depth. Latitude was found to have a shearwater, northern fulmar and black-legged kittiwake. The
habitat suitability index (HSI) models were developed (Brown
significant effect only on the midwater assemblages in 1999; richness of the food web is the primary factor that attracts
et al., 2000). This was done primarily to accommodate the
There are many different ways to analyze and organize there were no discernible latitudinal breaks within the other these species to the region.
paucity of empirical data in near-shore areas and to target
biogeographic information; however, to efficiently support the four assemblages.
species of special significance to the sanctuaries. An extensive
management plan process, only a limited number of analytical • Seasonal, interannual and decadal variation of the regional
literature review of the life history characteristics of individual
options were invoked. These analyses were selected based biogeography of marine birds is influenced by the vagaries
141
Section 5: SUMMARY OF BIOGEOGRAPHIC ASSESSMENT
work is also planned.
of marine climate, which is driven by the California Current abundance of the humpback and blue whales during the these concerns is presented in Section 3.
System and local upwelling centers. Therefore, the Upwelling and Oceanic seasons.
Spatial Patterns. The spatial occurrence of marine mammals
biogeographic patterns of marine birds are not static and Key Findings of Integration of Analyses
relative to large bathymetric features (e.g., shelf, upper
exhibit a dramatic spatial and temporal variation, both in Small Cetaceans. An important time period for the Pacific Fish Diversity (Trawl data). Three major areas of relatively
slope, lower slope) and discrete physiographic features (e.g.,
species composition and species abundance. white-sided dolphin (the most abundant small cetacean in high fish diversity (i.e., hot spots) were delineated, as noted
seamounts, banks, canyons, points and islands) varied by this study) was the Oceanic season. Important time periods below.
species and ocean condition. The occurrence patterns of
• The Gulf of the Farallones, the area lying inside a triangle for the other relatively abundant smaller cetaceans (northern • The northernmost hot spot is centered on Cordell Bank,
most marine mammals are strongly linked to the highly variable
defined by Point Reyes, the Farallon Islands and Año right-whale dolphin, Risso’s dolphin, Dall’s porpoise) could not within the northwestern corner of the Cordell Bank NMS,
ocean conditions of the study area, which significantly affect the
Nuevo Island, is the most important area for marine birds be determined in this preliminary assessment. and extends northward along the continental slope to Point
distribution of prey availability. In Phase II of this work, when the
in California. The reasons are: (1) large and taxonomically Arena.
data sets are more spatially and temporally robust, summary
diverse, demographically related populations breed at the Pinnipeds. The seasonal occurrence of pinnipeds was
analyses will be conducted to identify important areas and time
three afore-mentioned sites; and (2) an unparalleled diversity associated with the breeding cycles of the species. Important • The central hot spot is centered at the boundary between
periods across marine mammal groups.
of habitat (e.g., San Francisco Bay tidal plume, shallow time periods for the relatively abundant northern fur seal were the Gulf of the Farallones NMS and the Monterey Bay NMS.
sandy shelf, rocky reefs, submarine peaks, and the upper winter and early spring (Davidson Current and early Upwelling The area extends in a southeasterly direction past Point Año
Large Cetaceans. Important areas for the large cetaceans
continental slope) attracts a variety of migrant and seasonally seasons), which reflected the pelagic offshore distribution along Nuevo and ends offshore, north of Monterey Bay.
varied by species: the coast and inner shelf were important
resident species. the West Coast during the nonbreeding season. The relatively
for the gray whale; the outer shelf, slope, and deep ocean abundant California sea lion was present year-round in the • The southernmost hot spot is located between Point Sur and
were important for the humpback and blue whales; and many
• A "halo" of individuals was apparent around important study area, with densities greater during the Oceanic season Lopez Point and covers the inshore portions of Sur and Lucia
important areas for large cetaceans were identified seaward
breeding sites, such as the Farallon Islands and Año Nuevo (just after breeding) and Davidson Current season (before the Canyons. Portions of this last hot spot, however, were poorly
of the sanctuary boundaries.
Island. This pattern is the result of breeding individuals breeding season). Elephant seals, Steller sea lions, and harbor sampled. There is evidence of an additional hot spot strad-
searching for food, but going only as far as necessary to seals were present in sanctuary waters year-round in relatively dling the southern boundary of the Monterey Bay NMS.
Small Cetaceans. Review of the maps indicated that important
provide for their young. The Farallon "halo" for ashy storm- low numbers; and important time periods for these infrequently
areas for the relatively abundant small cetaceans were the
petrel, western gull, common murre, rhinoceros auklet and sighted species were inconclusive due to differences in Marine Bird Diversity
outer shelf and upper slope, Monterey Canyon, Sur and Lucia
Cassin’s auklet, extends substantially west of the Gulf of the behavior and low abundance (e.g., at-sea sightings typically • The interpolated maps of marine bird diversity show one con-
Canyons (west and south of the Monterey Bay NMS), Pioneer
Farallones National Marine Sanctuary. consist of single individuals or small groups of two or three, tinuous area of high diversity along the continental slope, and,
Canyon (west of the Monterey Bay NMS), Ascension, Cabrillo, elephant seals are rarely at the surface, and Steller sea lions to a lesser extent, along the shelf between Point Arena and
Año and Carmel canyons, Cordell Bank (and to the north of
• The marine birds of the Gulf of the Farallones/Cordell Bank are a threatened species). Point Sur. Within this area, diversity appears to be highest on,
the Cordell Bank NMS boundary), and the San Francisco Bay
NMS (as defined above) and the birds of the Monterey Bay and seaward of, the Farallon Escarpment, in the northwestern
tidal plume area (e.g., harbor porpoise). Smaller cetaceans
NMS are associated with different habitat features. The Gulf of INTEGRATION OF ANALYSES corner of the Monterey Bay NMS (Pioneer Canyon), and in
were also relatively abundant in areas that include canyons,
the Farallones has islands and a relatively broad shelf, while The integration of analyses across taxa occurs in Section 3. the marine region between Point Lobos and Point Sur.
and in locations beyond sanctuary boundaries, but within the
Monterey Bay has a relatively narrow but sheltered shelf, cut Many possible combinations of the data layers could be inte-
study area.
by an immense, deep submarine canyon. The greater oceanic grated for the biogeographic assessment. Because of differ- • The Farallon Escarpment in particular received a dispropor-
influence and lack of breeding islands in the Monterey Bay ences in sampling design, it was not appropriate to combine tionate amount of survey effort. The high estimated marine
Pinnipeds. Important areas for resident breeders (e.g., harbor
NMS drive the marine bird species groups there. data from different taxa (e.g. birds and fish) in order to calculate bird diversity for the Farallon Escarpment is, in part, due to
seal, Steller sea lion) were inner and outer shelf habitats, and for community metrics. Therefore, to minimize confounding results high sampling effort.
northern elephant seal, pelagic deep ocean habitats seaward
Preliminary Marine Mammal Analytical Results and to focus on the “protection of biodiversity” component of
of sanctuary boundaries. Seasonal visitors (e.g., northern fur
The Biogeography Team contracted principal investigators the NMSP mission, diversity and density were calculated sepa- • A marine bird diversity hot spot was found in the region be-
seals) occurred mostly in slope and deep ocean habitats,
David Ainley and Glenn Ford (of H.T. Harvey and Associates rately for each taxon and the resulting patterns were overlayed tween Point Lobos and Point Sur. The high residual diversity
seaward of sanctuary boundaries.
and R.G. Ford Consulting Co.) to work with the NOAA to identify biologically important areas across species groups. in this area supports the interpretation that this is a real hot
project team and local marine mammal experts to identify Spatial interpolation methods were applied to survey data to spot and not an artifact of survey effort.
Temporal Patterns. The patterns of seasonal occurrence for
biogeographic patterns and important areas and time periods provide a clearer picture of the distribution of diversity and den-
marine mammals varied by species. In Phase II of this work,
for marine mammals occurring in the study area. NOAA/NMFS sity within the study area. Hot spots were defined as regions • Marine bird diversity was correlated with survey effort, so
when the data sets are more complete, summary spatial and
scientists provided additional marine mammal sightings data in which diversity or density were estimated to be in the top some of the “hot spot” diversity areas coinciding with areas
temporal analyses across marine mammal groups will be
along the entire West Coast to aid in analyzing marine mammal 20% for a particular taxon. These hot spots were mapped for of high survey effort may, in part, be influenced by high lev-
conducted.
biogeographic patterns relative to the study area. The "bird and fish and birds individually and then combined to show areas els of effort. However, the general patterns of marine bird
mammal team" used a GIS to develop a preliminary series of of overlap. These areas of significant biological importance diversity are robust, and were largely unchanged by methods
Large Cetaceans. The seasonal occurrence of the larger
maps that show occurrence patterns and important areas and contributed to defining and assessing biogeographic patterns designed to correct for effort.
cetaceans in the study area reflected their migrations. The
time periods for 13 marine mammals in the study area. Phase within the study area and are discussed in the context of known
Davidson Current season was important for the gray whale,
II of this assessment will: incorporate additional data, develop oceanographic features and Sanctuary boundaries. All of the Overlap of Marine Bird and Fish (Trawl) Diversity
a period when this species is migrating either south or north.
additional marine mammal species and species group maps, conclusions listed below should be considered with an under- • Fish diversity shows overlap with the areas of high bird di-
Several species of the large cetaceans migrate to forage
and attempt to develop selected community metrics analyses standing of the inherent limitations of the available data and versity. The northernmost fish hot spot overlaps the marine
seasonally in the study area, a pattern reflected in the relative
across species and time periods. A technical report on this the approaches used to analyze it. A detailed discussion of bird hot spot from Cordell Bank north to approximately mid-
142
Section 5: SUMMARY OF BIOGEOGRAPHIC ASSESSMENT
support the Cordell Bank, Gulf of the Farallones, and Monterey
as that component of the study was not completed in Phase
• Although the majority of the hot spots for fish and bird density
way between Bodega Head and Point Arena. The seaward
Bay National Marine Sanctuaries.
I. The marine mammal analyses are one of the first efforts to
fall within sanctuary boundaries, it is notable that overlap-
half of the central fish hot spot overlaps with the area of
assess biogeographic patterns of marine mammals in the study
ping hot spots for both groups exist to the south of Monterey
high marine bird diversity within the Gulf of the Farallones
area; thus, additional analyses and peer review are required
Bay NMS.
NMS and the Monterey Bay NMS. The northern half of the
to complete this component of the study. REFERENCES
southernmost fish hot spot overlaps the southern tip of the
Overall Integration Summary Airamé, S., S. Gaines, and C. Caldow. 2003. Ecological Link-
marine bird hot spot.
Phase II activities may include publishing technical reports
•The current Sanctuary boundaries incorporate much of ages: Marine and estuarine ecosystems of central and northern
and peer-reviewed articles that complement the results of the
the highest diversity and highest density areas within the California. NOAA, National Ocean Service. Silver Spring, MD.
Fish Density (Trawl Data)
Phase I assessment, as well as additional analyses to further
region. 163 pp.
Four major hot spots of fish density were found:
define biological areas and time periods important to marine
• The northernmost hot spot is found on and to the southeast
fishes, birds, and mammals found throughout the study area.
• Many of these biologically important regions coincide with Brown, S.K., K.R. Buja, S.H. Jury, M.E. Monaco, and A. Ban-
of Cordell Bank.
known oceanographic and bathymetric features, such as ner. 2000. Habitat suitability index models for eight fish and
CD-ROM
upwelling regions, areas of high bathymetric variance, and invertebrate species in Casco and Sheepscot Bays, Maine.
• A small hot spot is found off of Point Reyes.
A digital version of this document, the Ecological Linkages
the continental shelf break. North American Journal of Fisheries Management, Vol. 20,
Report, all GIS-compatible files used to conduct the pp. 408-435. 28 pp.
• The largest fish density hot spot covers a large portion of
biogeographic analyses, metadata for GIS files, and a complete
• Regions of high diversity and high density outside of the
the shelf to the north of Monterey Canyon, the entire area of
suite of digital species maps, are found on the CD-ROM located
current sanctuary boundaries exist to the north, across Kendall, M.S. and M.E. Monaco. 2003. Biogeography of the
Monterey Bay, and the near shore waters south to Point Sur.
on the back cover of this document.
much of the shelf and slope, and to the south, in near- National Marine Sanctuaries: A partnership between the NOS
Although portions of this hot spot are found over Monterey
shore waters. Biogeography Program and the National Marine Sanctuary
Canyon, this fact should be interpreted with caution since the
All appropriate digital data and analytical products are found on Program. NOAA. Silver Spring, MD. 8 pp.
deep canyon waters themselves were not sampled.
the CD-ROM. The products come in several formats, including
• Uneven sampling effort across the study region and a lack
this document (in .pdf format), map products in a browsable
of trawl samples to the west of the Sanctuary boundaries Monaco, M.E. and J.D. Christensen. 1997. Biogeography Pro-
• The fourth hot spot is found to the south of Monterey Bay
web format (HTML), GIS shapefiles and grids for use with
limit the scope of any integrated biogeographic assess- gram: Coupling species distributions and habitat. In: Changing
NMS and covers a substantial area of the shelf from Point
MarIS or ArcView (GIS) software, tables in Excel format (.xls),
ment. oceans and changing fisheries: Environmental data for fisheries
Estero to Point Sal. This final hot spot is the largest region
and descriptive text files. Metadata for each shapefile or grid research and management. G.W. Boehlert and J.D. Schum-
of high fish density within the mapped area that falls outside
accompanies each file and appears in .xml format.
• Known limitations and biases of the two metrics (diversity acher (Eds.). National Marine Fisheries Service Technical
of Sanctuary boundaries.
and density) exist and are discussed in greater detail within Memorandum NOAA-TM-NMFS-SWRSC-239, Pacific Grove,
To support the NMSP and others in making maximum use of
Section 3. California. 7 pp.
Marine Bird Density
the spatial data generated from this study, along with other
• Marine bird density patterns should be interpreted with
products (e.g., economic assessments) that support the
DATA SOURCES AND GAPS NOAA's National Centers for Coastal Ocean Science and
caution since they largely reflect the distribution of the two
joint management plan review, the NMSP is developing a
Recognizing that any analysis is only as good as the data National Marine Sanctuary Program. 2002. Interim Product:
numerically dominant species.
GIS tool, the Marine Resource Information System (MarIS).
upon which it is based, the project team undertook a qualitative A Biogeographic Assessment off North/Central California: To
MarIS has been designed to facilitate the organization,
evaluation of the data used in this project and identified relevant Support the Joint Management Plan Review for Cordell Bank,
• A large region of high (top 20th percentile) marine bird density
analysis and display of spatial data to support analysis of
data gaps. This information is presented in Section 4: Data Gulf of the Farallones and Monterey Bay National Marine
exists adjacent to and shoreward of the marine bird diversity
NMSP management questions and issues within and across
Content and Gaps. This section describes the process used Sanctuaries. Silver Spring, MD. 38 pp.
hot spot. This density hot spot covers most of the shelf wa-
sanctuaries. All applicable spatial data will be integrated into
to select key databases for analyses and briefly addresses
ters of all three sanctuaries, from Point Sur in the south to
MarIS to enable NMSP staff and partners to conduct additional
strengths and limitations of each database. This information was Starr, R.M. 1998. Marine harvest refugia for West Coast rock-
midway between Bodega Head and Point Arena in the north.
biogeographic analyses in Phase II.
used to aid in the interpretation of the biogeographic analyses to fish: A workshop. Pacific Grove, California. NOAA-TM-NMFS-
The density hot spot extends into Monterey Bay.
minimize confounding of results due to information gaps. Also SWFSC-255. La Jolla, CA. 14 pp.
CONCLUDING COMMENTS
provided are recommendations for future research activities
• An additional density hot spot exists off of Morro Bay to the
This spatially explicit assessment provides a robust set of
that would enhance biogeographic assessment products.
south of the Monterey Bay NMS.
analytical results and GIS data to strengthen the sustainable
management of marine resources within and adjacent to the
PHASE II BIOGEOGRAPHIC ASSESSMENT
Overlap of Marine Bird and Fish (Trawl) Density
sanctuaries. A primary use of the biogeographic assessment
Section 6 suggests potential next steps to augment the
• Nearly all of the fish density hot spots are coincident with
will be to support the NMSP as it continues to conduct the joint
Phase I analyses. Phase II, however, will not be completely
the two areas of high bird density.
management plan review for the three sanctuaries. In addition,
designed until a review of Phase I products has occurred. The
the Biogeography Program will assist the NMSP in further
NMSP and NCCOS project team members will meet to define
• The hot spots for both metrics are generally confined to the
analyses and presentations of the data and analytical results
the additional suite of activities that will comprise Phase II.
shelf (<200m) with the notable exception of Monterey Canyon
to address specific research and management questions.
Nevertheless, a few priority activities are expected to occur in
which appears as a density hot spot for both groups. The
This Phase I product provides the foundation to continue the
Phase II, including expanding the analytical products for fishes,
deep Canyon, however, was not sampled in the fish trawl
development of a biogeographic assessment capability to
marine birds, and marine mammals. Special emphasis will be
surveys.
placed on the biogeographic analyses of marine mammal data,
143
Section 6: PHASE II BIOGEOGRAPHIC ASSESSMENT
Phase II of the biogeographic assessment of north/central California to support the research and management needs of Cordell Bank, Gulf of the Farallones, and Monterey Bay National Marine Sanctuaries will build on the information and analytical results
presented in Phase I. Phase II, however, will not be completely designed until a review of Phase I products has occurred. Most important, the NMSP and NCCOS project team members will meet to define an additional suite of activities that may comprise
Phase II. Nevertheless, a few priority activities are expected to occur in Phase II, including expanding the analytical products for fishes, marine birds, and marine mammals. A special emphasis will be placed on the biogeographic analyses of marine
mammal data, as that component of the study was not completed in Phase I. The marine mammal analyses are one of the first efforts to assess biogeographic patterns of mammals in the study area, thus additional analyses and peer review are required
to complete this section of the study.
Phase II activities may include publishing technical reports and peer reviewed articles that complement the results of the Phase I assessment and further define areas and time periods important to fishes, marine birds, and marine mammals found
throughout the study area. Also, discussions will be held between project partners to determine if additional assessments should be implemented to study near-shore and estuarine ecosystems and associated key species groups such as marine and coastal
invertebrates. These ecosystems and species were only qualitatively addressed in the Ecological Linkages Report due to data limitations, and time and resource constraints to complete the first phase of the project. In addition, to continue to implement
the 5-year Biogeography Program plan developed by NCCOS in consultation with the NMSP, additional habitat and environmental maps under various temporal climate regimes could possibly be addressed to support future biogeographic analyses. For
example, climatic regime shifts and associated influences on the distribution of living marine resources may provide additional insight into natural or anthropogenic perturbations on regional biogeographic patterns.
For now, project partners and colleagues are encouraged to provide comments on the information and analytical results provided in this document and on the CD-ROM. Also, please provide suggestions on how best to address Phase II proposed activities
and new biogeographic assessment studies that may complement or improve Phase I analyses. For further information or to provide comments on the Phase I product and Phase II activities, please contact:
Dr. Mark E. Monaco, Biogeography Team Leader
NOAA/NCCOS/CCMA
1305 East West Highway, N/SCI1
Silver Spring, MD 20910
p. 301-713-3028 x 160
f. 301-713-4384
mark.monaco@noaa.gov
or
Mr. Charles E. Alexander, National Programs Branch Chief
NOAA/NMSP
1305 East West Highway,
Silver Spring, MD 20910
p. 301-713-3125 x 147
f. 301-713-0404
charles.alexander@noaa.gov
144
ACKNOWLEDGEMENTS
A project of this magnitude would not have been possible without the cooperation, time, data and effort of numerous people and institutions. We would like to thank everyone who participated in this significant undertaking. The following is a list of those
who have contributed to this work.
Jamie Kum
Sarah Allen
Federal Government
Tom Laidig
Tara Anderson
National Centers for Coastal Ocean Science, National Ocean Service (NOS), NOAA
Mark Lampinen
Jay Barlow
National Marine Sanctuaries National Program Office, NOS, NOAA
Bob Lauth
Lillian Becker
Cordell Bank National Marine Sanctuary Office, NOS, NOAA
Bob Leeworthy
Heather Beeler
Gulf of the Farallones National Marine Sanctuary Office, NOS, NOAA
Phil Levin
Scott Benson
Monterey Bay National Marine Sanctuary Office, NOS, NOAA
Steve Lonhart
Carol Bernthal
Channel Islands National Marine Sanctuary Office, NOS, NOAA
David Lott
Nancy Black
Special Projects Office, NOS, NOAA
Mark Lowry
Laurence Breaker
Coastal Services Center, NOS, NOAA
Gerry McChesney
Joelle Buffa
Southwest Fisheries Science Center, National Marine Fisheries Service (NMFS), NOAA
Huff McGonigal
Tonya Builder
Northwest Fisheries Science Center, NMFS, NOAA
Nazila Merati
Gregor Cailliet
Alaska Fisheries Science Center, NMFS, NOAA
Richard Methot
John Calambokdis
U.S. Geological Survey, Department of Interior
Pat Morris
James Caretta
Minerals Management Service, Department of Interior
Joe Mortenson
Mark Carr
U.S. Fish and Wildlife Service, Department of Interior
Hannah Nevins
Harry Carter
Point Reyes National Seashore, National Park Service, Department of Interior
Kelly Newton
Josh Churchman
Jim Nybakken
Elizabeth Clarke
State/Local Government
Jim Oakden
Steve Copps
California Department of Fish and Game, Marine Region
Mike Parker
Natalie Cosentino
University of California, Santa Barbara
John Pearse
Don Croll
California State University, Monterey Bay
Brady Phillips
Brad Damitz
California State University, Moss Landing Marine Laboratories
Holly Price
Chris Essert
San Francisco Public Utilities Commission
Stephen Ralston
Megan Ferguson
Bob Read
Doug Forsell
Non-Governmental Organizations
Paul Reilly
Fathey Fosmark
Alliance of Communities for Sustainable Fisheries
Dale Roberts
Michael Gallagher
H.T. Harvey and Associates
Nora Rojek
Jim Glock
R.G. Ford Consulting Company
Kaustuv Roy
Gary Greene
Monterey Bay Aquarium Research Institute
Mary Jane Schramm
Denise Greig
PRBO Conservation Science
Michelle Staedler
Deirdre Hall
Diablo Canyon Power Plant
Rick Starr
Rick Hanks
William Sydeman
Michael Harris
Mario Tamburri
Chris Harvey
Christine Taylor
Jim Harvey
Julie Thayer
Brian Hatfield
Teresa Turk
Laird Henkel
W. Breck Tyler
Michelle Hester
Tiffany Vance
Tom Hourigan
Kerstin Wasson
Ruth Howell
Mark Wilkins
David Hyrenbach
Deb Wilson-Vandenberg
Todd Jacobs
Lisa Wooninck
Brian Jarvis
Nancy Wright
Roxanne Jordan
Levon Yengoyan
Mike Kenner
Mary Yoklavich
Stacy Kim
Mark Zimmermann
Chad King
Howatt King
145