MARINE ECOLOGY PROGRESS SERIES
   Vol. 327: 107–117, 2006                                           Published December 7
                            Mar Ecol Prog Ser




    Recruitment failure and shifts in community
   structure following mass mortality limit recovery
         prospects of black abalone
C. Melissa Miner1,*, Jessica M. Altstatt 2, Peter T. Raimondi3, Todd E. Minchinton3, 4
1
  Institute of Marine Science, University of California, Center for Ocean Health, Long Marine Laboratory, 100 Shaffer Road,
                       Santa Cruz, California 95060–5730, USA
           2
            Santa Barbara Channelkeeper, 714 Bond Street, Santa Barbara, California 93103, USA
   3
    Department of Ecology and Evolutionary Biology, University of California, Center for Ocean Health, Long Marine Lab,
                    100 Shaffer Road, Santa Cruz, California 95060, USA
    4
      Present address: Institute for Conservation Biology and School of Biological Sciences, University of Wollongong,
                         New South Wales 2522, Australia




      ABSTRACT: Mass mortalities of species can fundamentally alter the structure of natural communi-
      ties, which can in turn negatively impact species’ recovery. Beginning in 1994, some of the largest
      remaining populations of black abalone Haliotis cracherodii on the mainland coast of California,
      experienced mass mortalities due to the fatal disease called ‘withering syndrome’, which led to its
      listing as a species of concern by the USA National Marine Fisheries Service. We have been monitor-
      ing black abalone populations along the coast of southern and central California since 1992, and
      detection of withering syndrome at our southernmost site prompted us to investigate how the
      impending decline of this dominant grazer might correlate with changes in black abalone recruit-
      ment and the rocky intertidal community in which it lives. Quantitative surveys before and after mass
      mortalities revealed that, after black abalone declined, there was a shift in the composition of the
      intertidal species assemblage from one dominated by bare rock and crustose coralline algae (good
      quality abalone habitat) to one with increased cover of sessile invertebrates and sea urchins. Declines
      in abalone abundance were also correlated with a lack of recruitment to areas affected by withering
      syndrome, despite the presence of healthy adult populations only tens of kilometers away. This sug-
      gests that abalone recruitment might be limited by dispersal, a lack of quality habitat for settlement
      and early survival, or the continued presence of the disease agent. Recruitment failure and these dra-
      matic shifts in habitat quality indicate that the outlook for recovery of black abalone is poor.

      KEY WORDS: Withering syndrome · Haliotis cracherodii · Community structure · Recruitment limi-
      tation · Marine diseases · Mass mortality · Recovery · Rocky intertidal shores
                 Resale or republication not permitted without written consent of the publisher




            INTRODUCTION                    however, because the majority of taxa have planktonic
                                      propagules that may disperse great distances away
 Massive declines of dominant species in ecological            from their local population of reproductive adults (Kin-
communities can fundamentally alter the structure and           lan & Gaines 2003). Recovery is therefore contingent
composition of these natural assemblages. Disease and           upon successful recruitment of larvae into areas where
over-harvesting are 2 increasingly important factors            local populations have been impacted. However,
responsible for mass mortalities of marine species             recruitment can be highly variable in space and time
across the globe (Harvell et al. 1999, Jackson et al.           (e.g. Minchinton & Scheibling 1991), and requires a
2001, Harvell et al. 2002). Predicting the recovery of           convergence of favorable conditions, including an
marine species following mass mortalities is difficult,          available supply of larvae and specific conditions in


*Email: mwilson@biology.ucsc.edu                      © Inter-Research 2006 · www.int-res.com
108                    Mar Ecol Prog Ser 327: 107–117, 2006




the benthic habitat conducive to larval settlement and    tum and dislodging newly settled larvae or algal
early survival (e.g. Raimondi 1990, Minchinton &       spores through their movements and grazing
Scheibling 1993).                       (Leighton & Boolootian 1963). Indeed, researchers
  Ironically, successful recruitment into areas im-     monitoring black abalone on the Channel Islands in
pacted by mass mortality events is often dependent on     California, USA, noted increased cover of encrusting
the presence of healthy, conspecific adults. First, adults  sessile invertebrates following the decline of black
may be necessary to ensure a local supply of larvae if    abalone (Richards & Davis 1993). Although these
dispersal is limited by geographic or hydrological bar-    observations were correlative, they suggest that the
riers (e.g. headlands, ocean currents) or a species’ abil-  activities of adult abalone may ensure the availability
ity to disperse. Second, adults often attract larvae to an  of a suitable habitat for settlement, and thus facilitate
area simply by their presence or the release of chemi-    local recruitment and recovery of conspecific popula-
cal cues that induce settlement (e.g. Burke 1983, Rai-    tions.
mondi 1990, Minchinton 1997). Finally, adults may        In addition to these shifts in habitat structure due to
maintain suitable habitat for larval settlement and      the decline of adults, recruitment of black abalone may
juvenile survival (Douros 1985). Consequently, when      be further retarded by the decreased numbers of adults
adults decline in an area, there may no longer be an     in the local population. Abalone are dioecious broad-
adequate supply of larvae, suitable cues for settlement,   cast spawners and are thought to require densely ag-
or quality habitat for juvenile survival, which together   gregated adults to ensure fertilization success (Prince et
may limit recruitment necessary to ensure natural       al. 1987, Miller & Lawrenz-Miller 1993). If densities
recovery of impacted populations.               drop below a critical level, the probability of successful
  The once abundant black abalone Haliotis cracher-     fertilization is low because the distance between indi-
odii has virtually disappeared from rocky intertidal     viduals increases. The need for local fertilization is crit-
shores of southern California, USA. Declines are due     ical because there is evidence that black abalone larvae
primarily to a fatal disease called ‘withering syndrome’   have limited dispersal, localized recruitment and rela-
(Lafferty & Kuris 1993, Altstatt et al. 1996, Raimondi et   tively closed populations (Prince et al. 1988, Hamm &
al. 2002). Withering syndrome is caused by the bac-      Burton 2000, Chambers et al. 2006). Consequently,
terium ‘Candidatus Xenohaliotis californiensis’, which    black abalone recruitment and recovery is likely to be a
attacks the lining of the digestive track and results in   positive function of the local abundance of adults (Mc-
reduced body mass, weakness, and eventual withering      Shane 1992, Raimondi et al. 2002).
of the abalone’s foot until it can no longer cling to the    We hypothesize that recovery of black abalone pop-
substratum (Friedman et al. 2000). Declines have been     ulations will be hindered in areas where massive
so severe across all regions of southern California that   declines have occurred due to limitations placed on
the USA National Marine Fisheries Service now lists      recruitment, particularly a lack of larval supply and
Haliotis cracherodii as a species of concern.         shifts in community structure unfavorable to larval set-
  Black abalone are large, long-lived (> 30 yr), locally   tlement and early survival. No study, however, has
abundant space occupiers on rocky intertidal reefs and    investigated long-term changes in community struc-
may play a key role in maintaining favorable habitat     ture and patterns of abalone recruitment following
for conspecific recruitment and determining commu-      mass mortalities of black abalone due to withering syn-
nity structure (Cox 1962, Blecha et al. 1992, Richards &   drome. Here we present quantitative data from 2 long-
Davis 1993). Areas densely populated by black         term monitoring studies that extend from before to
abalone are typically characterized by bare rock or      after the mass mortalities of black abalone. In the first
crustose coralline algae (Douros 1985, authors’ pers.     study, we quantified how the structure and composi-
obs.), and crustose coralline algae release a chemical    tion of species assemblages in intertidal crevice habi-
cue that induces settlement and metamorphosis in sev-     tats changed over a 6 yr period at 3 sites before and
eral species of abalone larvae (Morse et al. 1979, Shep-   after abalone declines due to withering syndrome. In
herd & Turner 1985, McShane 1992), although this has     the second study, we quantified how recruitment of
not yet been tested directly for Haliotis cracheroidii.    abalone varies with adult abundance over a 13 yr
Other investigators have observed a substantial reduc-    period at 8 sites impacted and 4 sites not impacted by
tion or complete absence of abalone recruitment fol-     withering syndrome.
lowing mass mortality of black abalone due to wither-
ing syndrome (Richards & Davis 1993, Tissot 1995).
Adult abalone are primarily drift feeders, but they may          MATERIALS AND METHODS
maintain suitable conditions for recruitment of con-
specifics by preventing colonization of other inverte-     Shifts in community structure. Changes to intertidal
brates and algae by pre-empting space on the substra-     community structure coincident with black abalone
                     Miner et al.: Recovery of black abalone                    109




decline due to withering syndrome were monitored for       pled at these sites in June 1998 and January 2002 than
6 yr (1996 to 2002) at 3 sites (Boat House, Stairs, and      in January 1996.
Purisima Point) along the coast of southern California        At each sampling time, all black abalone within each
(Fig. 1). Sampling was done on 3 dates: January 1996,       crevice were counted and measured. Densities of
the initial sampling time; June 1998, when all 3 sites      abalone at each site were calculated by dividing the
had been severely impacted by withering syndrome;         sum of the number of abalone in all crevices by the
and January 2002, the final sampling date. Boat House       area of available crevice habitat in which abalone
and Stairs are gently sloping rocky platforms with        could potentially live, and density is expressed as
crevices, whereas Purisima Point is a long, jutting        ind. m–2. Individuals > 40 mm in length (maximal shell
boulder field interspersed by sections of rocky reef       length) were measured to the nearest 10 mm and those
with crevices. Crevices at Boat House and Purisima        ≤40 mm to the nearest 5 mm. Adult abalone were clas-
Point are wide and open, whereas those at Stairs are       sified as individuals ≥50 mm and juvenile recruits as
narrower and abalone were on average slightly           individuals < 50 mm, which is consistent with Leighton
smaller there, probably due to the restricted crevice       and Boolootian (1963) in that individuals ≥45 mm were
size. At all sites, black abalone were found almost        considered to be adults and juveniles were < 45 mm
exclusively lining the tops and bottoms of rock crevices     in length.
in the low to mid intertidal zone and, therefore, sam-        In each crevice, we also quantified the percent cover
pling of community structure (and recruitment, see        of species occupying the rock surface using a point
below) was restricted to these crevice habitats.         intercept method. Permanent bolts were installed to
 At each site we selected 10 crevices, each approxi-       delineate the ends of each crevice and a transect tape
mately 2 m long and of variable depth (ranging from 10      was then run parallel to the crevice opening and
to 75 cm deep), in which to sample community struc-        between the 2 bolts. Only the tops of crevices were
ture over time. Crevices within a site were selected for     sampled because the bottoms were occasionally cov-
similar shape, exposure to waves and sun, and black        ered with sand and shell grit, making sampling diffi-
abalone density. Abalone in 5 of the 10 crevices were       cult. Percent cover was estimated by placing a rod
not manipulated, whereas abalone were added to the        (with points marked every 3 or 6 cm along its length,
other 5 crevices as part of another experiment to con-      depending on crevice depth) perpendicular to and at
trol abalone densities in the face of declines due to       evenly spaced intervals along the transect tape. The
withering syndrome. Our attempt to maintain abalone        species occurring directly beneath each marked point
numbers failed and was discontinued after 2.5 yr.         on the rod was recorded. The interval at which the rod
There was no significant difference in abalone number       was placed along each transect was also adjusted
between crevices where abalone had been added and         depending on crevice length in order to sample a sim-
those where they had not, as determined by repeated        ilar number of points within each crevice (approxi-
measures analysis where the main effect and all inter-      mately 100 points).
actions involving treatment were not significant (for        Species were lumped into relatively broad cate-
each effect p > 0.6, see densities in Table 1). Thus, we     gories because we were mainly interested in general
treated all 10 crevices as ‘unmanipulated’ replicates.      changes in community structure (e.g. did free space
Several crevices at Boat House and Stairs were          get replaced by encrusting invertebrates?) and, more-
destroyed or filled by sediment and shell grit during       over, species identification was often difficult within
the 1997/98 El Niño storms and, consequently, fewer        crevices. Cover categories included: non-coralline
crevices (but at least n = 7 crevices per site) were sam-     algal crusts, coralline algal crusts, articulated coralline


Table 1. Haliotis cracherodii. Mean (± SE) densities (ind. m–2) in crevices used to determine change to community structure.
Abalone were added to ‘addition’ crevices only after withering syndrome caused populations to decline, in an attempt to
                         maintain original densities


Site       Treatment        Jan/Feb 1996           Jun 1998           Jan 2002
                    Adults   Juveniles      Adults   Juveniles      Adults  Juveniles

Boat House    Addition      15.8 ± 4.0   1.4 ± 0.5     3.6 ± 2.6  1.0 ± 1.0     0.7 ± 0.5  0.3 ± 0.3
          Control      11.0 ± 2.4   0.6 ± 0.3     4.3 ± 2.3    0       3.4 ± 1.4  0.7 ± 0.3
Stairs      Addition      38.4 ± 6.3   5.8 ± 3.3     3.9 ± 0.9  0.7 ± 0.7     0.8 ± 0.5    0
          Control      48.8 ± 8.5  12.9 ± 3.90    1.5 ± 1.5    0       1.1 ± 0.6    0
Purisima     Addition      19.0 ± 2.4   0.4 ± 0.4     5.4 ± 1.3  0.9 ± 0.4       0    0.5 ± 0.2
          Control      18.0 ± 5.5   0.6 ± 0.5     7.3 ± 3.6  0.1 ± 0.1     0.2 ± 0.2    0
110                    Mar Ecol Prog Ser 327: 107–117, 2006




algae, other algae (all algae not included in previous      and tube snails (Phragmatopoma californica, Serpula
algal groups, common spp. included Mastocarpus          vermicularis, Spirobranchus spinosus, Spirorbis spp.,
papillatus, Mazzaella flaccida, Endocladia muricata,       Salmacina tribranchiata, Serpulorbis squamigerus,
Porphyra spp.), barnacles (Tetraclita rubescens, Bal-      Petalochonchus montereyensis), anemones (Antho-
anus glandula, Chthamalus fissus/dalli), tube worms       plueura elegantissima, A. xanthogrammica), sea ur-




Fig. 1. Location of intertidal sites where community structure and recruitment were monitored, and densities of black abalone
        Haliotis cracherodii adults and juveniles at these sites over time. Note differences in scale on y-axes
                    Miner et al.: Recovery of black abalone                  111




chins (mainly Strongylocentrotus purpuratus, rarely S.    potentially live (which included areas with or without
franciscanus), encrusting invertebrates (sponges, bry-    abalone), and density is expressed as ind. m–2. Adults
ozoans, tunicates, hydroids), seastars (mainly Pisaster   and juvenile recruits were measured and classified
ochraceus and Asterina miniata), and bare rock. When     based on their lengths as described in the previous
a point fell on a black abalone, it was recorded as an    subsection. Additional details of sampling methods
abalone on top of whatever was the nearest species (or    have been previously described in Altstatt et al. (1996)
bare rock). For analysis, these points for this nearest   and Raimondi et al. (2002).
species were lumped with their respective species cat-     Using the data from all 12 sites, we examined the
egory so that a decline in percent cover of abalone     form of the relationship (particularly whether it was
would not drive the patterns of change over time. For    positive) between the abundance of adults and recruit-
example, if a point was recorded as ‘black abalone on    ment at a site, and how this relationship varied accord-
bare rock’ this point was lumped with the ‘bare rock’    ing to whether samples had been collected at sites that
category.                          had been impacted or not impacted by withering syn-
  Temporal changes in community structure among       drome. ‘Impacted’ samples were those collected from
sites were displayed using canonical discriminant      sites following abalone decline due to the disease. ‘Not
analysis (CDA) and were tested for statistically using    impacted’ samples were those collected from sites that
multivariate analysis of variance (MANOVA). We des-     had not been affected by withering syndrome at the
ignated species categories (see above) as the depen-     time of sampling, including those samples taken from
dent variables and site and time as the independent     impacted sites prior to the appearance of the disease
variables in the model. Percent cover data were exam-    and abalone decline at that site. Samples collected
ined for homogeneity of variances and normality using    from sites during periods when abalone were declining
box plots and normal distribution probability plots. To   due to withering syndrome were excluded. Examina-
meet these assumptions, data were square root trans-     tion of the adult-recruit relationship was done by plot-
formed. In addition, we examined the equality of the     ting the density of adults at a site in the autumn or
group variance-covariance matrices by running a prin-    spring of one year versus the density of juvenile
cipal components analysis on the covariance matrix of    recruits at the same site and time in the following year.
the transformed data. Only variables in the CDA with     The relationship was lagged by a year because repro-
loadings (correlations between each variable and the     duction by adults occurs from autumn to spring, with
discriminant function) ≥ 0.3, and standardized coeffi-    recruitment in the summer, and juveniles reach a size
cients (contribution of each variable to a canonical     at which they can be counted by the following autumn
axis) ≥ 0.3 were used to interpret community change     to spring. Therefore, the abundance of juveniles sam-
over time.                          pled in autumn (or spring) in one year would reflect the
  Recruitment of black abalone. In addition to docu-    abundance of reproductive adults from autumn (or
menting changes in community structure at Boat        spring) in the previous year. Note that our size cut-off
House, Stairs and Purisima Point, we quantified adult    for juvenile recruits (< 50 mm) may lead to individuals
abundance and recruitment of abalone at these and 9     being included in multiple samples since black
other sites over 5 to 13 yr (from spring 1992 to autumn   abalone take approximately 2 yr to reach sizes ≥50 mm
2004), 5 that had been impacted by withering syn-      (Leighton & Boolootian 1963). However, because we
drome (Government Point, Cayucos Point, Rancho        are simply showing the pattern of relationship be-
Marino, Piedras Blancas, and Point Sierra Nevada) and    tween adults and juveniles, and not performing sta-
4 not impacted by the disease (Mill Creek, Andrew      tistics on these data, this non-independence was not
Molera, Mal Paso, and Point Lobos) (Fig. 1). Note that    a concern.
due to the northward progression of the disease,
‘impacted’ and ‘non-impacted’ sites are spatially sepa-
rated. However, no formal statistics were performed on               RESULTS
the data. Numbers and sizes of black abalone occur-
ring within 3 large, permanent plots (plot areas ranged          Shifts in community structure
from 20 to 87 m2) at each of the 12 sites were recorded
in the spring (usually February to March) and autumn      Black abalone numbers began to decline in crevices
(usually October to November) of each year. Abalone     at Boat House just before the initiation of the study
occurred almost exclusively in a few crevices within     (Figs. 1 & 2). A preliminary count was done in No-
our much larger plots. Therefore, densities of abalone    vember 1995 when crevices were chosen, and num-
at each site were calculated by dividing the sum of the   bers had declined sharply by January 1996, when
number of abalone in all 3 plots by the area of the     treatments were initiated. A decline in abalone abun-
available crevice habitat in which abalone could       dance at Stairs soon followed, beginning shortly
112                     Mar Ecol Prog Ser 327: 107–117, 2006




Fig. 2. Haliotis cracherodii. Mean density of (A) adults and (B) juveniles in crevices at Boat House, Stairs and Purisima Point,
             where community structure was monitored. Note differences in scale on y-axes


after the initial sample in January 1996 (Figs. 1 & 2).      space (MANOVA: Pillai’s trace = 2.191, approximate
The black abalone population at Purisima Point           F = 2.297, df = 88, 536, p < 0.001) (Fig. 3). The change
appeared to be healthy until just after January 1998        in community structure displayed similar temporal tra-
(Figs. 1 & 2). At all sites, declines in abalone abun-       jectories along the first discriminant axis, with the
dance were accompanied by observations in the field        onset of change coincident with massive declines in
of individuals across all size classes with symptoms of      abalone numbers (Figs. 2 & 3). Changes were most
withering syndrome. The onset of withering syndrome        extreme along the first discriminant axis for Boat
and timing of abalone decline in crevices used to         House, which is likely due to abalone declining at this
monitor community structure at these sites was coin-        site first (Fig. 3). Community change along the second
cident with those observed in the permanent plots         discriminant axis was similar at Boat House and Stairs,
used to monitor recruitment at these sites since 1992       but not at Purisima.
(compare Figs. 1 & 2).                        Particular taxa explained a substantial part of the
 There were very high levels of discrimination in the       variation in the change in community structure as-
structure of the crevice communities over time and         sociated with abalone decline among sites over time




Fig. 3. Canonical discriminant analysis of species assemblages in crevices at Boat House, Stairs, and Purisima Point, showing
changes in community structure over time. Relative contribution of each variable to each discriminant function (the standardized
coefficient) is shown to the right of each graph. Only those variables with standardized coefficients and loadings ≥ 0.3 were used
for interpretation of community change. SE bars based on pooled SD across all samples. (A) CD1 explains 51.7% of total
                     variation and (B) CD2 explains an additional 23.8%
                      Miner et al.: Recovery of black abalone                    113




(Figs. 3 & 4). Over time at all sites, the decline of             Recruitment of black abalone
black abalone was coincident with a decrease in the
cover of bare rock and an increase in the cover of tube       Both before and after mass mortalities due to with-
worms and tube snails (particularly at Boat House), as      ering syndrome, the temporal patterns of abundance
well as other encrusting invertebrates (Figs. 3 &         of juvenile recruits generally paralleled those of the
4A,B,D). Similar to bare rock, the overall cover of        adults at each site (Fig. 1). The abundance of both
crustose coralline algae decreased over time (except       adult and juvenile abalone declined precipitously at
in the narrow crevices at Stairs where urchins were        the 5 southern sites impacted by withering syndrome
particularly abundant), dramatically reducing the         (Government Point, Boat House, Stairs, Purisima
availability of free space on the substratum at Boat       Point and Cayucos Point), and abalone numbers in
House and Purisima Point (Figs. 4E,F). Barnacle cover       both size classes were beginning to decrease at the 3
varied inconsistently among sites over time, parti-        sites more recently affected by the disease (Rancho
cularly from January 1996 to June 1998, but there was       Marino, Piedras Blancas and Point Sierra Nevada)
an overall decline at Boat House and Stairs, while        (Fig. 1). By contrast, densities of adult abalone and
cover remained relatively constant at Purisima Point       juvenile recruits remained relatively constant at the 4
(Figs. 3 & 4C). An increase in the number of sea         sites not yet affected by withering syndrome (Mill
urchins over time appears to be an important driver of      Creek, Andrew Molera, Mal Paso and Point Lobos).
community change at Boat House and Stairs as           Note that recruitment was patchy in time at all
abalone decline, but not at Purisima Point (Figs. 3 &       unaffected sites, as well as at affected sites before
4F). Articulated coralline algae covered on average        being impacted by withering syndrome, indicating
< 2% of the substratum, but great variability in their      that fluctuations between periods of low and high
cover among sites over time help to discriminate the       recruitment is typical for healthy abalone populations
communities (Fig. 3B).                      (Fig. 1). Recruitment of juvenile abalone to sites




Fig. 4. Mean percent cover over time of (A) tubeworms and tube snails, (B) encrusting invertebrates (includes sponges, bry-
ozoans, tunicates, and hydroids), (C) barnacles, (D) bare rock and (E) crustose coralline algae, and (F) mean number of urchins
Strongylocentrotus purpuratus, in crevices at Boat House, Stairs and Purisima Point, where community structure was monitored.
           SE bars based on pooled SD across all samples. Note differences in scale on y-axes
114                     Mar Ecol Prog Ser 327: 107–117, 2006




where there had been mass mortalities of adults            recruits following adult mortalities provide some of the
was rare, despite successful recruitment at sites only        best evidence to date suggesting that black abalone,
tens of kilometers away (Fig. 1).                   like other abalone species, may have localized disper-
 The relationship between the density of juvenile          sal and relatively closed populations. Abalone decline
recruits and adult abalone was striking, with almost         has been followed by unfavorable shifts in the species
no recruitment of juveniles where adult density was          assemblages occupying the crevices that provide
<1 ind. m–2 (Fig. 5). Although most samples with adult        important habitat for abalone recruitment and sur-
densities <1 ind. m–2 occurred at sites that had been         vival. Together, such changes may severely limit the
impacted by withering syndrome, this pattern also           ability of this threatened species to recover naturally
held for 2 non-impacted samples, suggesting that,           from this serious disease.
even in the absence of the disease, recruitment occurs          Several explanations can be put forth to account for
only when adult densities are > 1 ind. m–2. Densities of       the lack of local recruitment to sites experiencing mass
juvenile recruits varied considerably at adult densities       mortalities of black abalone including: (1) there is no
higher than this ‘threshold’, and did not appear to          local production of abalone larvae at these sites and,
show any positive or negative trend. Because there          due to localized dispersal, larvae produced at nearby
was no apparent relationship (linear or otherwise)          sites do not reach these sites, (2) abalone larvae dis-
between adult densities and juvenile recruits other          perse to sites but changes to the assemblage of species
than this ‘threshold’ in adult density associated with        following abalone mortality do not provide suitable
recruitment, we did not perform any statistical analy-        cues or habitats for successful settlement and early sur-
ses to test for a quantitative relationship.             vival of larvae, and (3) abalone larvae disperse and re-
                                   cruit to sites but persistence of the disease at these sites
                                   results in their early mortality before they are large
            DISCUSSION                  enough to be observed. While the data from our study
                                   cannot distinguish among these alternatives, evidence
  The long-term and large-scale set of monitoring data        suggests that some are more plausible than others
presented here demonstrate an almost complete fail-          and highlights important directions for research.
ure of recruitment to black abalone populations follow-         Results here suggest that a lack of local larval pro-
ing mass mortalities due to withering syndrome. While         duction and dispersal limitation due to extremely local-
this pattern has not been experimentally tested, the         ized dispersal of black abalone larvae may be the most
pattern of abalone declines from south to north, the         plausible explanation for the lack of abalone recruit-
close spacing of sites along the coast, and the lack of        ment to sites impacted by withering syndrome. First,
                                         the density and dispersion of adult black
                                         abalone at sites following disease was likely
                                         below that which is needed to ensure suc-
                                         cessful fertilization and, hence, a local supply
                                         of larvae. Abalone are dioecious broadcast
                                         spawners and require close proximity to other
                                         individuals for fertilization to occur (Prince et
                                         al. 1988, Miller & Lawrenz-Miller 1993). Con-
                                         sequently, if the density of black abalone
                                         drops below the level required for successful
                                         fertilization (or if abalone are not suitably
                                         aggregated), then a local supply of larvae
                                         may no longer be available to replenish local
                                         populations. Other researchers have ob-
                                         served a similar cessation of black abalone
                                         recruitment after local populations dropped
                                         below some critical level (Miller & Lawrenz-
                                         Miller 1993, Richards & Davis 1993). Second,
                                         despite an abundance of reproductive adults
Fig. 5. Haliotis cracherodii. Relationship (lagged by 1 yr) between density   at nearby northern sites, recruitment of juve-
of adults and juvenile recruits, for sites impacted or not impacted by with-   nile abalone was rare at any site where
ering syndrome (see ‘Materials and methods’ for details). To present
clearly the range of adult densities at which recruitment occurred, the x-
                                         abalone populations had been impacted by
axis has been fourth-root transformed and a dashed line placed on graph     withering syndrome. For example, no recruits
                 at 1 ind. m–2                  were observed at Government Point after
                     Miner et al.: Recovery of black abalone                   115




mass mortalities, despite the dense and healthy popu-      spaces between boulders, which are important micro-
lations of adults at Boat House, Stairs, and Purisima      habitats for abalone recruitment (Douros 1985, Shep-
Point only 15 to 30 km away. Evidence that black        herd & Turner 1985, Miller & Lawrenz-Miller 1993).
abalone populations are relatively closed is further        Shifts in the species assemblages from what might
supported by genetic data (Hamm & Burton 2000),         be considered favorable to unfavorable habitat for set-
which demonstrated localized population differentia-      tlement by abalone was not equivalent across sites. In
tion of black abalone, and the local retention of larvae    particular, changes in community structure at Stairs
has been shown to exist for another species of abalone     were largely driven by an increase in the number of
(Prince et al. 1988).                      urchins, which may compete with abalone for
  It is also possible that abalone larvae arrive at sites   resources (Davis et al. 1992, Day & Fleming 1992), but
impacted by the disease but that the species assem-       may also maintain appropriate habitat for abalone
blages of the crevice communities have changed,         recruitment through grazing of foliose algae and
potentially in a direction that is not suitable for the     increasing the cover of crustose coralline algae
recruitment of this once-dominant grazer. On average      (McShane 1992). Additionally, abalone recruits might
for all sites where we monitored community structure,      benefit from the presence of sea urchins as a source of
mass mortalities of black abalone were followed by       shelter from predation, as well as an enhanced food
declines in the cover of bare rock and crustose         supply (Rogers-Bennett & Pearse 1998, Day & Branch
coralline algae and increases in the cover of sessile      2002b, Day & Branch 2002c, Tomascik & Holmes 2003),
invertebrates. Because we did not directly manipulate      although this relationship has not been demonstrated
abalone abundance, we cannot rule out alternative        for Haliotis cracherodii and the relatively short-spined
explanations for the observed changes in community       Strongylocentrotus pupuratus, which predominates in
structure, such as physiological responses of organisms     this intertidal region. Therefore, in the narrow crevices
to climate change, or changes in the physical charac-      at Stairs it might be argued that habitat suitable for
teristics of the sites (e.g. sedimentation, water quality).   abalone recruitment persisted following the mass mor-
Nevertheless, the expectation is that these large-scale     talities, yet there was still no abalone recruitment, sup-
phenomena would lead to broad-scale changes to ben-       porting the contention that inadequate larval supply is
thic assemblages and habitat across entire sites        limiting black abalone recovery.
throughout the region. We have 15 yr of monitoring         It is also possible, however, that withering syndrome
data for a variety of species spread throughout each of     persists at sites long after it has devastated the local
the sites where crevice communities were studied, and      population so that black abalone recruits may not sur-
have not observed similar changes in areas outside of      vive once they arrive, but this is a less likely explana-
the crevices (Miner et al. 2005). Further support comes     tion. Although there has been little research on the
from Douros (1985), who found that the density of        potential for withering syndrome to persist at lethal
black abalone in intertidal areas was positively corre-     levels at a site after mass mortality, anecdotal evidence
lated with the cover of crustose coralline algae and      suggests that the amount of bacteria in the water col-
negatively correlated with the cover of tube worms       umn is reduced once black abalone populations have
and fleshy algae. Free space on the substratum is nec-     declined (C. Friedman pers. comm.). Therefore, it is
essary for abalone settlement, and crustose coralline      unlikely that the absence of recruits is due to mortality
algae (and their associated chemicals and biofilms) are     by disease, particularly as some sites such as Govern-
known to induce settlement of abalone larvae (Morse       ment Point have had no recruits for many years.
et al. 1979, Douros 1985, Day & Branch 2002a). The       Nevertheless, it is not known whether the bacterium
presence of conspecific individuals, particularly adults,    responsible for withering syndrome would multiply
may also be important for settlement as their grazing      rapidly to lethal levels were there a sudden increase in
activities might maintain the cover of crustose coralline    black abalone numbers, such as would occur with an
algae (Douros 1985) and their mucus on the substratum      influx of recruits.
may act as a positive settlement cue for abalone larvae      Finally, one could argue that abalone larvae are
(Bryan & Qian 1998), but Naylor & McShane (2001)        recruiting to areas in the intertidal zone outside of the
have reported that adult abalone can smother conspe-      crevice habitats studied here, and that we simply failed
cific recruits. The increased cover of tube worms and      to observe them. For example, black abalone have
tube snails documented in this study may be particu-      been observed to recruit to coralline encrusted cobble
larly detrimental to recolonization because tube worms     beds (P. Raimondi pers. obs.). However, many of our
have been shown to prey on abalone larvae (Naylor &       sites with high abalone densities are located on rocky
McShane 1997). Increases in the colonial tube worm       intertidal reefs that contain no cobble beds, and as
Phragmatopoma californica may have an additional        demonstrated here, crevices are also important areas
impact by filling in crevices and cementing together      for black abalone recruitment. Thus, if recruitment
116                     Mar Ecol Prog Ser 327: 107–117, 2006




was occurring at a site, at least some of the recruits     for funding from the Minerals Management Service, the
should have been recorded. Moreover, given the         Coastal Marine Institute at UC Santa Barbara, the County of
                                Santa Barbara, and the Partnership for Interdisciplinary Stud-
spatial and temporal scope of the present study (and
                                ies of Coastal Oceans (PISCO): a long-term ecological consor-
related monitoring programs sampling other species       tium funded by the David and Lucille Packard Foundation
on these and other shores; see Miner et al. 2005), it is    and the Gordon and Betty Moore Foundation. T.E.M. was par-
extremely unlikely that we would have failed to detect     tially supported by a Natural Sciences and Engineering
massive recruitment of black abalone. Ultimately, the      Research Council of Canada Postdoctoral Fellowship. We
                                acknowledge N. Read and Vandenberg Air Force Base, B.
rarity or absence of adults over more than a decade of     Lundberg and the Cojo-Bixby Ranch, the Hearst Corporation,
sampling provides the best proof of recruitment failure     the Bureau of Land Management, the California State Park
following abalone decline. Therefore, the recruitment      System, and the Kenneth S. Norris U.C. Reserve for access to
failure of black abalone may be attributable to a        sites. This is PISCO contribution number 217.
combination of the above explanations: an inadequate
larval supply, shifts in community structure unsuitable               LITERATURE CITED
for early settlement and survival, and the possible
persistence of withering syndrome. Studies designed       Altstatt JM, Ambrose RF, Engle JM, Haaker PL, Lafferty KD,
to differentiate among these alternatives should be        Raimondi PT (1996) Recent declines of black abalone Hali-
                                  otis cracherodii on the mainland coast of central Califor-
focal areas of research to understand key limitations       nia. Mar Ecol Prog Ser 142:185–192
to the natural recovery or restoration efforts of this     Blecha JB, Steinbeck JR, Sommerville DC (1992) Aspects of
threatened species.                        the biology of the black abalone (Haliotis cracherodii) near
  Thus far, management of black abalone populations        Diablo Canyon, central California. In: Shepherd SA, Teg-
                                  ner MJ, Guzmán del Próo SA (eds) Abalone of the world:
has focused solely on their biology and has been
                                  biology, fisheries, and culture. Proc 1st Int Symp Abalone.
geared towards successful spawning and rearing of         Blackwell Scientific Publications, Cambridge
black abalone larvae in the laboratory. The California     Bryan PJ, Qian PY (1998) Induction of larval attachment and
Department of Fish and Game’s Abalone Recovery and         metamorphosis in the abalone Haliotis diversicolor
Management Plan (CDF & G 2005) estimates that a          (Reeve). J Exp Mar Biol Ecol 223:39–51
                                Burke RD (1983) The induction of metamorphosis of marine
minimum viable population size for all species of         invertebrate larvae: stimulus and response. Can J Zool 61:
abalone, including black abalone, is 2000 ind. ha–1        1701–1719
(0.2 m–2) and that 6600 ind. ha–1 (0.66 m–2) will sustain    CDF & G (California Department of Fish & Game) (2005)
a fishery. These densities are far below the densities at     Abalone recovery and management plan. Resources
                                  Agency, Sacramento, CA. Available at: www.dfg.ca.gov/
which there was some natural recruitment in the pre-
                                  mrd/armp/index.html
sent study (where significant recruitment was only       Chambers MD, VanBlaricom GR, Hauser L, Utter F, Friedman
detected at densities >1 ind. m–2); furthermore, these       CS, (2006) Genetic structure of black abalone (Haliotis
estimates were at sites not impacted by withering syn-       cracherodii) populations in the California islands and cen-
drome. Moreover, given the reproductive mode of          tral California coast: impacts of larval dispersal and deci-
                                  mation from withering syndrome. J Exp Mar Biol Ecol 331:
abalone, which typically requires individuals in close       173–185
proximity for successful fertilization, and the patchy     Cox KW (1962) California abalones, family Haliotidae. Calif
nature of suitable abalone habitat in the intertidal        Dep Fish Game Fish Bull 118:1–133
area, management plans must consider not only the        Davis GE, Richards DV, Haaker PL, Parker DO (1992)
                                  Abalone population declines and fishery management in
density of individuals at a local site at an appropriate
                                  southern California. In: Shepherd SA, Tegner MJ,
spatial scale, but also the dispersion of larvae and the      Guzmán del Próo SA (eds) Abalone of the world: biology,
availability of suitable habitat. We advocate a shift to a     fisheries, and culture. Proc 1st Int Symp Abalone. Black-
whole-ecosystem based management approach and           well Scientific Publications, Cambridge, p 237–249
argue that knowledge about the entire intertidal spe-      Day E, Branch GM (2002a) Effects of benthic grazers on
                                  microalgal communities of morphologically different
cies assemblage, the population dynamics of black         encrusting corallines: implications for abalone recruits.
abalone (including dispersion, density, recruitment        Mar Ecol Prog Ser 244:95–103
trends, etc.), the habitat requirements for abalone set-    Day E, Branch GM (2002b) Effects of sea urchins (Parechinus
tlement and the availability of such habitat, and the       angulosus) on recruits and juveniles of abalone (Haliotis
                                  midae). Ecol Monogr 72:133–149
behavior of the withering syndrome pathogen would
                                Day EG, Branch GM (2002c) Influences of the sea urchin
be critical for management-based recovery of the          Parechinus angulosus (Leske) on the feeding behaviour
black abalone.                           and activity rhythms of juveniles of the South African
                                  abalone Haliotis midae Linn. J Exp Mar Biol Ecol 276:1–17
                                Day RW, Fleming AE (1992) The determinants and measure-
Acknowledgements. We thank L. MacDonald, G. Heistand,       ment of abalone growth. In: Shepherd SA, Tegner MJ,
D. Hubbard, D. Farrar, and many others for their help in the    Guzmán del Próo SA (eds) Abalone of the world: biology,
field. The manuscript was much improved by comments from      fisheries, and culture. Proc 1st Int Symp Abalone. Black-
Mike Graham and 2 anonymous reviewers. We are grateful       well Scientific Publications, Cambridge, p 141–168
                       Miner et al.: Recovery of black abalone                     117




Douros WJ (1985) Density, growth, reproduction and recruit-     Miner CM, Raimondi PT, Ambrose RF, Engle JM, Murray SN
  ment in an intertidal abalone: effects of intraspecific com-    (2005) Monitoring of rocky intertidal resources along the
  petition and prehistoric predation. MA, University of Cali-    central and southern California mainland. Comprehensive
  fornia Santa Barbara                        report (1992–2003) for San Luis Obispo, Santa Barbara,
Friedman CS, Andree KB, Beauchamp KA, Moore JD, Rob-          Ventura, Los Angeles, and Orange Counties. Report No.
  bins TT, Shields JD, Hedrick RP (2000) ‘Candidatus Xeno-      MMS-2005–071. US Minerals Management Service,
  haliotis californiensis’, a newly described pathogen of      Pacific OCS Region
  abalone, Haliotis spp., along the west coast of North      Morse DE, Hooker N, Duncan H, Jensen L (1979) Gamma-
  America. Int J Syst Evol Microbiol 50:847–855           aminobutyric acid, a neurotransmitter, induces planktonic
Hamm DE, Burton RS (2000) Population genetics of black         abalone larvae to settle and begin metamorphosis. Sci-
  abalone, Haliotis cracherodii, along the central California    ence 204:407–410
  coast. J Exp Mar Biol Ecol 254:235–247             Naylor JR, McShane PE (1997) Predation by polychaete
Harvell CD, Kim K, Burkholder JM, Colwell RR and 9 others       worms on larval and post-settlement abalone Haliotis iris
  (1999) Emerging marine diseases — climate links and        (Mollusca: Gastropoda). J Exp Mar Biol Ecol 214:283–290
  anthropogenic factors. Science 285:1505–1510          Naylor JR, McShane PE (2001) Mortality of post-settlement
Harvell CD, Mitchell CE, Ward JR, Altizer S, Dobson AP, Ost-      abalone Haliotis iris caused by conspecific adults and
  feld RS, Samuel MD (2002) Climate warming and disease       wave exposure. N Z J Mar Freshw Res 35:363–369
  risks for terrestrial and marine biota. Science 296:      Prince JD, Sellers TL, Ford WB, Talbot SR (1987) Experimen-
  2158–2162                             tal evidence for limited dispersal of haliotid larvae (genus
Jackson JBC, Kirby MX, Berger WH, Bjorndal KA and 15 oth-       Haliotis; Mollusca: Gastropoda). J Exp Mar Biol Ecol 106:
  ers (2001) Historical overfishing and the recent collapse of    243–263
  coastal ecosystems. Science 293:629–638             Prince JD, Sellers TL, Ford WB, Talbot SR (1988) Confirma-
Kinlan BP, Gaines SD (2003) Propagule dispersal in marine       tion of a relationship between the localized abundance of
  and terrestrial environments: a community perspective.       breeding stock and recruitment for Haliotis rubra Leach
  Ecology 84:2007–2020                        (Mollusca: Gastropoda). J Exp Mar Biol Ecol 122:91–104
Lafferty KD, Kuris AM (1993) Mass mortality of abalone       Raimondi PT (1990) Patterns, mechanisms, consequences of
  Haliotis cracherodii on the California Channel Islands:      variability in settlement and recruitment of an intertidal
  Tests of epidemiological hypotheses. Mar Ecol Prog Ser       barnacle. Ecol Monogr 60:283–309
  96:239–248                           Raimondi PT, Wilson CM, Ambrose RF, Engle JM, Minchin-
Leighton D, Boolootian RA (1963) Diet and growth in the        ton TE (2002) Continued declines of black abalone along
  black abalone, Haliotis cracherodii. Ecology 44:227–238      the coast of California: are mass mortalities related to El
McShane PE (1992) Early life history of abalone: a review. In:     Niño events? Mar Ecol Prog Ser 242:143–152
  Shepherd SA, Tegner MJ, Guzmán del Próo SA (eds)        Richards DV, Davis GE (1993) Early warnings of modern pop-
  Abalone of the world: biology, fisheries, and culture. Proc    ulation collapse in black abalone Haliotis cracherodii,
  1st Int Symp Abalone. Blackwell Scientific Publications,      Leach, 1814 at the California Channel Islands. J Shellfish
  Cambridge, p 120–137                        Res 12:189–194
Miller AC, Lawrenz-Miller SE (1993) Long-term trends in       Rogers-Bennett L, Pearse JS (1998) Experimental seeding of
  black abalone, Haliotis cracherodii Leach, 1814, popula-      hatchery-reared juvenile red abalone in northern Califor-
  tions along the Palos Verdes peninsula, California. J Shell-    nia. J Shellfish Res 17:877–880
  fish Res 12:195–200                       Shepherd SA, Turner JA (1985) Studies on southern Aus-
Minchinton TE (1997) Life on the edge: conspecific attraction     tralian abalone (genus Haliotis). VI. Habitat preference,
  and recruitment of populations to disturbed habitats.       abundance and predators of juveniles. J Exp Mar Biol Ecol
  Oecologia 111:45–52                        93:285–298
Minchinton TE, Scheibling RE (1991) The influence of larval     Tissot BN (1995) Recruitment, growth, and survivorship of
  supply and settlement on the population-structure of bar-     black abalone on Santa Cruz Island following mass mor-
  nacles. Ecology 72:1867–1879                    tality. Bull S California Acad Sci 94:179–189
Minchinton TE, Scheibling RE (1993) Free-space availability     Tomascik T, Holmes H (2003) Distribution and abundance of
  and larval substratum selection as determinants of barna-     Haliotis kamtschatkana in relation to habitat, competitors
  cle population-structure in a developing rocky intertidal     and predators in the Broken Group Islands, Pacific Rim Na-
  community. Mar Ecol Prog Ser 95:233–244              tional Park Reserve of Canada. J Shellfish Res 22:831–838

Editorial responsibility: Steven Morgan (Contributing        Submitted: October 28, 2005; Accepted: May 1, 2006
Editor), Bodega Bay, California, USA                Proofs received from author(s): November 13, 2006