Mangroves as fish habitat: 50 years of field studies
MARINE ECOLOGY PROGRESS SERIES
Vol. 318: 1–18, 2006 Published August 3
Mar Ecol Prog Ser
OPEN
PEN
ACCESS
CCESS
FEATURE ARTICLE: REVIEW
Mangroves as fish habitat: 50 years of field studies
Craig H. Faunce1, 3,*, Joseph E. Serafy1, 2
1
Division of Marine Biology and Fisheries, Rosenstiel School of Marine and Atmospheric Science, University of Miami,
4600 Rickenbacker Causeway, Miami, Florida 33149, USA
2
Southeast Fisheries Science Center, National Marine Fisheries Service, 75 Virginia Beach Drive, Miami, Florida 33149, USA
3
Present address: Florida Fish and Wildlife Conservation Commission, Fish and Wildlife Research Institute,
Tequesta Field Laboratory, 19100 SE Federal Highway, Tequesta, Florida 33469, USA
ABSTRACT: Mangroves dominate undisturbed nat-
ural shorelines of many sub-tropical and tropical
regions, yet their utilization by fishes is poorly
understood. To provide the first comprehensive list
of empirical field studies for comparative and ref-
erence purposes, we assembled and quantified
aspects of 111 mangrove-fish surveys published
between 1955 and 2005. Differences in the location,
purpose, methodology, data gathered, and analyses
performed among studies have resulted in a frag-
mented literature making cross-study comparisons
difficult, at best. Although the number of published
studies has increased over time, a geographical
bias in the literature has persisted towards studies
performed in the USA and Australia, and against
studies performed in Southeast Asia and West
Africa. The typical survey design has examined
<10 fixed locations on a monthly or bimonthly basis
for a period of less than 2 yr. Water temperature
and salinity measurements have been the most
reported habitat variables; others, such as structural
and landscape measures, continue to be rare.
Moreover, the focus to date has been on identifying Faunce & Serafy review mangrove fish studies, focusing on
sampling methodology, and on the types of fish and habitat
assemblage-level patterns of fish use, with very few
data reported. While most studies have addressed spatio-
studies providing species-specific estimates of
temporal patterns in fish assemblage structure, species-
abundance, growth, mortality, and secondary pro-
specific estimates of fish mortality, growth and secondary
duction. Unless future studies strive towards ob-
production are still required for appraisal of the importance
taining such estimates, gauging the importance of of mangroves as fish habitat.
mangroves as fish habitat and their broader contri- Photo: Jiangang Luo
bution to ecosystem diversity and production will
remain elusive.
INTRODUCTION
KEY WORDS: Fishes · Essential fish habitat · Man-
Mangrove wetlands are a dominant feature of undis-
groves · Nursery · Review
turbed tropical and subtropical shorelines around the
Resale or republication not permitted without
globe. Throughout their range, however, these habi-
written consent of the publisher
*Email: craig.faunce@myfwc.com © Inter-Research 2006 · www.int-res.com
2 Mar Ecol Prog Ser 318: 1–18, 2006
tats are in a state of decline. Approximately one-third words ‘mangrove(s)’ and/or ‘fish(es)’. Records were
of the world’s mangrove forests has been lost to coastal selected from the earliest available time period (1971)
development over the past 50 yr (Alongi 2002). While to January, 2005. The resulting list of over 500 publica-
there is general agreement that mangroves provide a tions was reduced to relevant works according to 2
buffer against storm surges, reduce shoreline erosion main criteria: (1) the study must have been published
and turbidity, absorb and transform nutrients, and are in readily available outlet (i.e. in the ‘primary litera-
inhabited by a variety of organisms, opinions vary as to ture’); and (2) each publication must have contained a
the importance of mangrove habitats to fishes and, by field-based survey of the ichthyofauna that was con-
extension, to offshore fisheries (Thollot & Kulbicki ducted within a natural mangrove system. Second, the
1988, Blaber et al. 1989, Thollot 1992, Nagelkerken et Science Citation Index (Web of Science; http://isi5.
al. 2001). For example, the sub-tidal prop-root habitats newisiknowledge.com) was used to identify articles
of mangroves are often cited as nurseries for fishes of that cited works from our reduced ASFA list. Again,
economic importance. Today, the protection of man- any additional publications were vetted according to
groves worldwide is based almost entirely on their the selection criteria above. Third, articles from
purported importance to fisheries and/or a number of the authors’ personal libraries and those introduced
rare and endangered species (Snedaker 1989, Baran & through peer review (of this paper) were added. The
Hambrey 1998). However, because the same man- references cited in each relevant article were exam-
grove species can often occur under marine, estuarine, ined for new items and this process was continued
and freshwater conditions, a wide variety of fish until no additional publications emerged.
assemblages can be found among their inundated Study locations were grouped into 5 geographic
prop-roots (hereafter termed ‘mangrove habitats’). As regions following the World Mangrove Atlas (Spalding
such, mangrove habitats likely play a variety of roles in et al. 1997): (1) South and Southeast Asia (Pakistan to
the lives of associated fishes; feeding areas for some the west, China and Japan to the northeast, including
species or life stages, daytime refugia for others, nurs- Indonesia), (2) Australasia (Australia, Papua New
ery and/or nesting areas for yet more. This situation Guinea, New Zealand, and the South Pacific islands),
suggests that questions regarding the contribution of a (3) the Americas (north, central, and south), (4) West
given mangrove habitat to the diversity, productivity Africa, and (5) East Africa and the Middle East (Iran to
and stability of broader fish communities (and their South Africa eastwards, including the islands in the
exploited components) must be carefully qualified, or, Indian Ocean). Using the selection (foraging) ratio of
in some cases, may be premature. Savage (1931), geographical bias in the literature was
The purpose of this paper is to address some of the expressed as the proportion of total studies realized
most basic questions regarding the body of literature per region relative to the area of mangrove coverage
on mangrove fishes that has been published over the within each region (Manly et al. 1993).
past 5 decades. These questions include: How many The study purpose, methodology, data gathered, and
field studies have been conducted, why and where analyses performed were extracted and tabulated
were they performed, and what techniques were used? using vote-counting procedures, where ‘present’ was
What types of measurements have been made of the given a value of 1 and ‘absent’ was given a value of
fish assemblages, their component species and their 0. Data were expressed as proportions of the total
habitats? Is there sufficient basis for comparing assem- number of votes per attribute. Study purposes included
blages of mangrove fishes with those associated with identifying spatial or temporal patterns, generating
other, structurally-complex habitats, such as seagrass species lists, identifying explanatory variables, biogeo-
beds and coral reefs? The answers to these questions graphic comparisons, restoration, water management,
are pertinent to researchers about to embark on new and gear evaluations. Methodologies included the
studies, as well as to those making efforts to balance sampling design (fixed, random, haphazard, or vari-
natural resource protection with pressing socio-eco- ous), sampling frequency (daily, weekly, fortnightly,
nomic considerations. monthly, bimonthly, quarterly, seasonally, semi-annu-
ally) sampling duration, and gear type. Gear types were
classified according to Rozas & Minello (1997) and
included entanglement gear (gill or trammel nets),
METHODS
towed nets (trawls, seines), passive samplers (fyke
Publications for this review were selected from 3 nets, flume nets, rotenone-used with or without nets,
databases. First, a search of the Aquatic Sciences fish traps, e.g. breder, plankton), and ‘enclosure sam-
and Fisheries Abstracts (ASFA) electronic database plers’ (block or drop nets, drop traps, and cast nets).
was conducted (Cambridge Scientific Abstracts; www. Visual surveys and angling were added as additional
csa.com) using keyword and title searches for the gears. The type of mangrove forest sampled was noted
3
Faunce & Serafy: Mangrove-fish literature review
using the classification scheme of Lugo & Snedaker mangrove coverage, it is encouraging that more litera-
(1974), which included fringing, riverine and/or basin ture is emerging from this region where coastal fish as-
forest. The data gathered in each study included biotic semblages are also relatively diverse (Blaber 2002).
and abiotic habitat metrics. Fish metrics included However, a literature void remains for the West Africa
groupings by family, maturation stage, residency sta- region — an area that is likely to continue to be under-
tus, trophic level, or diel habits as well as the type of represented unless specifically targeted for study.
fish data gathered and analyzed. We classified the Interestingly, the first 2 studies of mangrove fishes
types of fish metric data reported in each study accord- were conducted in the 2 regions that are least repre-
ing to the criteria recommended for determining sented today.
‘essential fish habitat’ (EFH) (USDOC 1996) which Although disproportionate, the spatial distribution of
included: presence/absence, frequency of occurrence, studies we examined covers much of the known global
percent composition, size, biomass (g), density (num- distribution of mangroves (Fig. 2). Specific areas that
ber/area), standing crop (g/area), growth and mortality have received the most thorough study include Florida
rates, and rates of secondary production. Finally, the (USA) and Moreton Bay (Australia). Similar dominance
focus of analyses (e.g. defining spatial and/or temporal of studies from the USA and Australia in the literature
patterns, examining fish–habitat correlations) was tab- has beset previous reviews of fishes occupying sea-
ulated and the type of statistical test(s) or data treat- grass beds (Heck et al. 2003), mangroves (Sheridan &
ment(s) performed (similarity measures, analysis of Hays 2003), and studies of ontogenetic fish movements
variance, ordination, or regression) were recorded. (Gillanders et al. 2003). Regions outside US and Aus-
tralian waters where data are particularly lacking
include: (1) Pacific Panama; (2) Colombia; (3) Central
Brazil; (4) the Red Sea; (5) Mozambique; (6) the Bay of
RESULTS AND DISCUSSION
Bengal and the Andaman Sea; and (7) Borneo (Fig. 2).
However, there are several important cases where
Chronology and geography
fishes inhabiting tropical estuaries from these areas
A total of 111 publications were examined from 104 have been reported. In such cases, data summaries
independent field surveys of mangrove fishes pub- were made from multiple surveys of fishes that rarely
lished between 1955 and 2005 (Table 1). The earliest mentioned study objectives, methods, gear, or habi-
records of mangrove-associated fishes were species tat(s) sampled, making their inclusion here difficult, at
lists compiled by Inger (1955) and Boeseman (1963) as best. In Florida (USA), species lists of mangrove fishes
part of broad ecological inventories of forests in Borneo were compiled from numerous studies by Odum et
(South and Southeast Asia region) and the Niger Delta al. (1982) and presented for tidal streams, estuarine
(West Africa region), respectively. Austin (1971) pro- bays, and oceanic bays. Off Columbia, Alvarez León
vided the first inventory of mangrove fishes from & Blanco Racedo (1985) reviewed aspects of 31 studies
Puerto Rico (American region), and Day (1974) from conducted in the Cartagena Bay system. They pro-
Mozambique (i.e. the East Africa and Middle East vided an overall species list for the system, and sum-
region). Blaber (1980), and Blaber & Blaber
(1980) published the first of many works on W Africa
S & SE Asia
100
assemblages of mangrove fishes from Aus- E Africa
Cumulative number of publications
tralia (Australasian region). Australasia
Americas
While the cumulative number of publica- 80
tions has grown steadily since the mid 1980s,
the sharpest increases occurred for the
regions of South and Southeast Asia and the 60
Americas (Fig. 1). Selection indices indicate
that the geographic distribution of studies
40
among regions has not been commensurate
with the proportion of the world’s mangrove
acreage that they contain (Table 2). Nearly 20
70% of studies have been conducted in either
the Americas or Australasia, and the South
0
and Southeast Asia and West Africa regions 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
Year
are clearly under-represented in the litera-
ture. Because South and Southeast Asia con- Fig. 1. Cumulative number of publications on mangrove fishes by each
tains the largest proportion of the world’s geographic region from 1955 to 2005 (n = 111)
4
Table 1. Chronological list of surveys of mangrove fishes conducted during 1955 to 2005. Design codes: FX: fixed; HZ: haphazard; RM: random; VAR: various. SU: sample
unit; Prop. mangrove: proportion of sites that were within mangrove habitats. Frequency codes: BM: bimonthly; D: daily; F: fortnightly; M: monthly; S: seasonal; Q: quar-
terly; Var: various. Gear codes: A: anglers; B: block net; C: cast net; F: fyke net; G: gill or trammel net; P: plankton net; R: rotenone; S: seine net; T: trap; TP: trap or flume net;
TW: trawl; V: visual surveys. Published studies with same letter [in brackets] are from the same fish survey. nr = not reported
Location Region Design SU No. No. Prop. Fre- Duration Gear Basin Fringe Riverine Source
SU sites mangrove quency (yr)
Dewhurst Bay, Borneo SA FX Site 11 11 0.36 1 nr A,C,R,S x Inger (1955)
Niger Delta WA nr Region 4 4 1.00 Var NR x Boseman (1963)
Western Puerto Rico AM FX Estuary 8 8 1.00 Var G,S x x Austin (1971)
Ponggol Estuary, Singapore SA FX Site 3 3 1.00 M 1.0 S x Thia-Eng (1973)
Morrumbene Estuary, Mozambique EA FX Site 6 6 0.50 3 A,S x x Day (1974)
Old Rhodes Key lagoon, USA AM FX Site 6 6 M 0.4 A,T,V Holm (1977)
Guadeloupe AM FX Site 12 12 0.33 M 0.1 F,TP x Lasserre & Toffart [a] (1977)
Marco Island Estuary, USA AM FX Site 11 11 0.00 M 4.1 TW Weinstein et al. (1977)
Huizache and Caimanero lagoons, AM FX Site 5 5 0.80 M 1.3 C,S x Warburton (1978)
Mexico
Mngazana Estuary, S Africa EA FX Site 7 7 1.00 3 G,P,S x x Branch & Grindley (1979)
Trinity Inlet System, Australia AU FX Site 18 18 0.39 nr B,G,P,S x Blaber (1980)
Moreton Bay, Australia AU FX Site 4 4 0.50 M 1.0 B,P,S x Blaber & Blaber (1980)
Serpentine Creek, Australia AU FX Site 10 10 0.00 F 1.2 TW Quinn (1980)
Tamil Nadu coast, India SA nr Region 1 nr P x Krishnamurthy &
Prince Jeyaseelan (1981)
Jiquilisco Bay, El Salvador AM FX Site 6 6 0.00 M 1.3 TW Phillips [b] (1981b)
Jiquilisco Bay, El Salvador AM FX Site 6 6 0.00 M 1.3 C,G,TW Phillips [b] (1981b)
Isle of Youth, Cuba AM FX Site 4 4 1.00 S 1.0 V x x Valdéz-Muñoz (1981)
Bahia de la paz, Mexico AM FX Site 3 7 0.00 M 0.5 TW Martínez et al. (1982)
Papua New Guinea AU FX Site 14 14 1.00 nr R x x x Collette (1983)
Botany Bay, SE Australia AU FX Site 1 1 1.00 BM 1.5 B,R x Bell et al. (1984)
Wairiki Creek, Fiji AU FX Site 6 6 1.00 M 1.0 G x Lal (1984)
Mar Ecol Prog Ser 318: 1–18, 2006
Dampier region, NW Australia AU FX Biotope 2 9 0.67 Q 1.0 G,R,S,V x x x Blaber et al. (1985)
Guadeloupe AM FX Bay 2 12 0.33 Var Var TP x Louis et al. [a] (1985)
Papua New Guinea AU FX Estuary 2 5 0.00 F 2.0 TW Quinn & Kojis (1985)
Phuket Island, Thailand SA FX Site 4 4 0.50 M 1.6 P,TW x x Janekarn & Boonruang (1986)
Laguna Joyuda, Puerto Rico AM FX Site 5 5 0.00 M 1.0 TW Stoner (1986)
Elichi Creek, Nigeria, W Africa WA FX Site 2 2 1.00 M 1.1 B x Wright (1986)
Pagbilao Bay, Phillipines SA FX Site 6 6 1.00 M 1.5 G,TW x Pinto (1987)
Alligator Creek, Australia AU nr Biotope 3.5 8 0.50 BM 1.2 S x Robertson & Duke [c] (1987)
Western Florida Bay, USA AM FX Site 8 8 1.00 Var Var B,R,TW x x Thayer et al. (1987)
Sinnamary river, French Guiana AM nr Region 3 M Var C,G,TP,R,TW x x Boujard & Beltran (1988)
Leanyer Swamp, N Australia AU FX Site 2 2 1.00 F 0.5 R,T x x Davis (1988)
Tudor Creek, Kenya, E Africa EA FX Site 4 4 0.75 F 0.6 P,S x Little et al. [d] (1988b)
Tudor Creek, Kenya, E Africa EA FX Site 5 5 0.60 nr nr P x Little et al. [d] (1988a)
Saint-Vincent Bay, New Caledonia AU nr Biotope 3 121 0.01 Var Var F,G,TW,R,V x x Thollot & Kulbicki (1988)
Teminos Lagoon, Mexico AM FX Biotope 2 2 0.00 M 1.0 TW Yáñez-Arancibia et al. [e]
(1988)
Embley Estuary, NE Australia AU FX Biotope 5 6 0.83 Q 2.0 B,G,R,S,TW x Blaber et al. (1989)
Table 1 (continued)
Location Region Design SU No. No. Prop. Fre- Duration Gear Basin Fringe Riverine Source
SU sites mangrove quency (yr)
Nigerian Coast, W Africa WA FX Estuary 4 nr P,S nr nr nr Amadi et al. (1990)
Solomon Islands, W Pacific AU FX Estuary 13 13 1.00 nr B,G,R,S x x Blaber & Milton (1990)
Selangor, Malaysia SA FX Biotope 4 41 0.22 Var Var F,G,TW x x Chong et al. [f] (1990)
Tecapan-Aqua Brava, Mexico AM FX Site 15 15 1.00 M 1.0 S,TW Flores-Verdugo et al. (1990)
Moreton Bay, E Australia AU FX Site 1 3 1.00 M 1.2 B,G,S x x Morton (1990)
Alligator Creek, Australia AU nr Estuary 1 BM 1.2 G,S,TP x x Robertson & Duke [c] (1990b)
Alligator Creek, Australia AU nr Estuary 4 BM 1.2 G,S,TP x x Robertson & Duke [c] (1990a)
Moreton Bay, Australia AU FX Biotope 5 5 0.60 M 0.8 TW x Weng (1990)
Dubreka and Tabunsu estuary,
Guinea WA FX Estuary 2 nr A,C,G,TW nr nr nr Boltachev (1991)
Cayo Collado, Puerto Rico AM FX Site 8 8 1.00 S 0.2 V x Rooker & Dennis (1991)
Northern US Virgin Islands AM FX Site 5 5 1.00 M 1.8 T,V x Boulon (1992)
La Parguera, Puerto Rico AM FX Biotope 4 4 0.50 M 1.1 P x Dennis (1992)
Fort-de-France, Martinique AM FX Site 8 8 1.00 Q 2.3 TP x Louis et al. [g] (1992)
Selangor, Malaysia SA FX Biotope 3 41 0.22 Var Var F,G,TW x x Sasekumar et al. [f] (1992)
Alligator Creek, Australia AU RM Region 3 6 0.50 M 0.4 A,S,T x Sheaves (1992)
Rookery Bay, USA AM VAR Biotope 3 30 0.33 Q 1.3 TP x Sheridan (1992)
Saint-Vincent Bay, New Caledonia AU VAR Biotope 2 577 0.03 M 1.0 G,R,TM,TP,V x Thollot (1992)
Tanshui River, Taiwan SA FX Site 3 3 0.00 M 2.8 B x Tzeng & Wang (1992)
Bonaire, Netherlands Antilles AM nr Biotope 6 nr V x van der Velde et al. [h] (1992)
Sabana-Camaguey, N Cuba AM RM Region 3 63 1.00 1 nr V x Claro & García-Arteaga (1993)
Phang-Nga Bay, Thailand SA FX Site 1 1 0.00 M 1.2 TW Janekarn (1993)
Ambergis Cay, Belize AM FX Biotope 3 3 0.33 BM 1.0 TW x Sedberry & Carter (1993)
Terminos Lagoon, Mexico AM FX Biotope 2 2 0.00 BM 1.2 TW Yáñez-Arancibia et al. [e]
(1993)
Gulf of Carpentaria, N Australia AU FX Biotope 4 S Var B,F,G,R,S,TW x x x Blabler et al. (1994)
Clarence River, Australia AU FX Site 18 18 0.28 Q 2.5 B,R x Pollard & Hannan (1994)
Gulf of Nicoya, Costa Rica AM FX Site 3 3 1.00 M 0.9 G,S x Rojas et al. [i] (1994a)
Gulf of Nicoya, Costa Rica AM FX Site 3 3 1.00 M 0.9 G,S x Rojas et al. [i] (1994b)
Faunce & Serafy: Mangrove-fish literature review
Celestum Lagoon, Mexico AM FX Site 1 1 1.00 S 1.5 B,R x x Vega-Cendejas et al. (1994)
Upper Tampa Bay, USA AM RM Bay 2 24 M 2.9 T x x x Vose & Bell (1994)
Raby Bay, E Austalia AU FX Biotope 2 7 0.43 M 1.1 S x Williamson et al. (1994)
Moreton Bay, E Australia AU FX Biotope 2 6 0.33 M 2.0 S,TP x x Laegdsgaard & Johnson (1995)
Fort-de-France, Martinique AM FX Site 8 8 1.00 Q 2.3 TP x Louis et al. [g] (1995)
Placido Bayou, USA AM FX Site 6 6 1.00 Var B,S x x Mullin (1995)
Lagos Lagoon, Nigeria WA FX Site 2 2 0.50 M S x Nwadukwe (1995)
Sao Luis, Brazil AM FX Site 4 4 1.00 M 0.8 TP x Batista & Rego (1996)
Tin Can Bay, NE Austalia AU FX Site 4 4 1.00 BM 2.2 B x Halliday & Young (1996)
Gazi Creek, Kenya, E Africa EA FX Site 3 3 0.67 M 0.8 S x Kimani et al. (1996)
Negombo Estuary, W Sri Lanka SA FX Site 6 4 1.00 M 2.1 S x Pinto & Punchihewa (1996)
Embley River, N Australia AU FX Site 4 4 1.00 D,S Var B x Vance et al. (1996)
La Parguera, Puerto Rico AM RM Biotope 3 43 0.21 nr G,V x Acosta (1997)
Karachi, Pakistan SA FX Site 2 2 0.50 F 0.9 C nr nr nr Atiqullah et al. (1997)
5
6
Table 1 (continued)
Location Region Design SU No. No. Prop Fre- Duration Gear Basin Fringe Riverine Source
SU sites mangrove quency (yr)
Tulear Lagoon, SW Madagascar EA FX Site 1 1 1.00 M 1.0 G Laroche et al. (1997)
Guaratuba, Brazil AM RM Region 1 1 1.00 nr C,S,TW Chaves & Correa (1998)
Sao Luis, Brazil AM FX Site 4 4 1.00 M 0.8 TP x Batista & Rego (1996)
Tin Can Bay, NE Australia AU FX Site 4 4 1.00 BM 2.2 B x Halliday & Young (1996)
Gazi Creek, Kenya, E Africa EA FX Site 3 3 0.67 M 0.8 S x Kimani et al. (1996)
Negombo Estuary, W Sri Lanka SA FX Site 6 4 1.00 M 2.1 S x Pinto & Punchihewa (1996)
Embley River, N Australia AU FX Site 4 4 1.00 D, S Var B x Vance et al. (1996)
La Parguera, Puerto Rico AM RM Biotope 3 43 0.21 nr G,V x Acosta (1997)
Karachi, Pakistan SA FX Site 2 2 0.50 F 0.9 C nr nr nr Atiqullah et al. (1997)
Tulear Lagoon, SW Madagascar EA FX Site 1 1 1.00 M 1.0 G Laroche et al. (1997)
Guaratuba, Brazil AM RM Region 1 1 1.00 nr C,S,TW Chaves & Correa (1998)
NE tropical Australia AU HZ Region 4 12 1.00 Q 2.3 T x Sheaves (1998)
Yingluo Bay, China SA FX Site 2 2 1.00 Q 1.0 B x x He et al. (2001)
Guaratuba, Brazil AM RM Region 1 1 1.00 M 2.9 TW Chaves & Bouchereau (1999)
Bay of La Paz, Mexico AM FX Site 1 1 1.00 nr F x Gonzalez-Acosta et al. (1999)
N & S Taiwan SA FX Region 6 6 1.00 BM 1.0 F x Kuo et al. (1999)
NE Florida Bay, USA AM FX Bay 6 42 1.00 M 1.1 B,R,V x x Ley et al. [j] (1999)
N Taiwan SA FX Site 1 1 1.00 M 1.0 F x Lin & Shao (1999)
NE Florida Bay, USA AM FX Site 4 24 1.00 BM 5.2 R,T x Lorenz (1999)
Pagbilao, Philippines SA FX Site 4 4 1.00 D 0.0 B x Rönnbäck et al. (1999)
Curacao, Netherlands Antilles AM FX Biotope 6 12 1.00 Var V x Nagelkerken et al. [k] (2000a)
Curacao, Netherlands Antilles AM FX Biotope 6 12 1.00 Var V x Nagelkerken et al. (2000b)
Bonaire, Netherlands Antilles AM FX Biotope 6 27 0.22 M 0.6 V x Nagelkerken et al. [h] (2000c)
Sine Saloum, Senegal WA FX Site 3 6 1.00 M 0.2 F Vidy (2000)
Guaratuba, Brazil AM RM Region 1 1 1.00 nr F,S,TW Chaves & Vendel (2001)
NE Florida Bay, USA AM FX Bay 6 17 1.00 M 1.1 V x Ley & McIvor [j] (2001)
Curacao, Netherlands Antilles AM FX Bay 3 11 0.00 D, M Var S,V Nagelkerken et al. (2001)
Mar Ecol Prog Ser 318: 1–18, 2006
St. Croix, US Virgin Islands AM HZ Site 10 10 1.00 M 2.0 S,T x Tobias (2001)
Rio da Fazenda, Brazil AM nr Region 3 3 0.33 BM 1.0 A,T,V x Uieda & Uieda (2001)
Sepetiba Bay, Brazil AM RM Region 3 158 0.00 M 3.0 TW Araujo et al. (2002)
Caete River Estuary, Brazil AM FX Site 3 3 1.00 M 1.0 TP x Barletta-Bergan et al. (2002)
Curacao, Netherlands Antilles AM FX Biotope 3 11 3.00 S nr V x Cocheret de la Morinière et al.
[k] (2002)
Lower Volta, Ghana WA FX Site 7 7 0.71 nr nr C,G,S x x Dankwa & Gordon (2002)
Curacao, Netherlands Antilles AM FX Biotope 8 17 0.71 S 1.0 V x Nagelkerken & van der Velde
[k] (2002)
Eastern Caribbean AM FX Biotope 3 23 0.00 Var Var V x Nagelkerken et al. (2002)
Bimini, Bahamas AM FX Biotope 4 4 0.25 nr B,S x Neuman & Gruber (2002)
Sikao Creek Estuary, Thailand SA FX Site 6 6 0.67 Q 26.7 F,R,S x x x Tongnunui et al. [l] (2002)
Johor Strait, Malaysia SA FX Site 4 4 0.25 M 1.5 S x Hajisamae & Chou (2003)
Sikao Creek, Thailand SA FX Site 3 3 0.67 Q 2.4 S x x Ikejima et al. [l] (2003)
Biscayne Bay, USA AM RM Region 2 129 1.00 S 2.0 V x Serafy et al. (2003)
SE Everglades, USA AM FX Site 3 12 1.00 BM 3.2 V x Faunce et al. (2004)
Rio Lagartos, Mexico AM FX Site 28 28 0.00 Var Var S,TW Vega-Cendejas & Santillana
(2004)
Chwaka Bay, Tanzania EA FX Biotope 5 5 0.40 Var Var S,V x Lugendo et al. (2005)
7
Faunce & Serafy: Mangrove-fish literature review
Fig. 2. Location of studies of mangrove fishes used in the present review (coded by geographic region)
marized their data based on the number of species fishes (21.7%) or identify explanatory variables for
belonging to various salinity regimes and trophic observed utilization patterns (15.6%). Less than 10%
levels for Bahía de Cartagena, Ciénaga de Tesca, and of studies were concerned with the remaining topics
Ciénaga Grande de Santa Marta. Similarly, Cervigón (Fig. 3a). Most studies aimed to identify temporal pat-
(1985) used data from an earlier study and historical terns, and typically achieved this goal through monthly
records to generate a list of fish species for the Ori- sampling for a period of 0.5 to 1.5 yr (Fig. 3b,c). Sam-
noco estuary (Venezuela) according to salinity regime. pling durations of more than 2 yr were uncommon
Finally, the ecology of the Itamaracá ecosystem (Brazil) (< 5%) with the longest published survey spanning 5 yr
was summarized by Paranagua & Eskinazi-Leça (1985), (Lorenz 1999). In addition, most studies sampled, or
who provided a family and species list of fishes. For otherwise quantified, fishes at a small number of loca-
more complete summaries of fish studies conducted tions. Only 4 studies sampled mangroves at more than
from large tropical estuarine systems, the reader 20 locations: Serafy et al. (2003) sampled 129 locations,
should consult the comprehensive works of Blaber Claro & García-Arteaga (1993) sampled 63 locations,
(2000) and Yáñez-Arancibia (1985). Ley et al. (1999) sampled 42 locations, and Lorenz
(1999) sampled 24 locations (Table 1). Fixed sampling
designs were employed much more often (81%) than
random-stratified (8.5%) or haphazard designs (1.8%)
Study design
or various other sampling designs (1.8%). In fixed- and
Study design incorporates a study’s purpose with its mixed-design surveys, the rationale for selecting site
methodology. Over half of the examined surveys of locations was rarely provided.
mangrove fishes aimed to identify spatial and/or temp- These results highlight some limitations with our
oral patterns of mangrove utilization, while a lesser knowledge of mangrove habitat utilization by fishes.
For example, if not selected remotely, or a priori, the
proportion were conducted to provide an inventory of
Table 2. Comparison of mangrove area and the number of published studies from within each geographic region. The selection
ratio (wi) of Savage (1931) was used to compare the proportion of studies to the proportion of mangrove area within each region
Region Mangrove Proportion of Number of Proportion of wi
area (km2) total area total studies total studies
Americas 49 096 0.271 53 0.477 1.762
Australasia 18 789 0.104 26 0.234 2.252
East Africa and Middle East 10 024 0.055 7 0.063 1.147
South and Southeast Asia 75 173 0.415 18 0.162 0.391
West Africa 27 995 0.155 7 0.063 0.407
Total 181 0770 1110
8 Mar Ecol Prog Ser 318: 1–18, 2006
a) Purpose ment of what constitutes a ‘typical’ year, or of the
0.6
extent of inter-annual variability in both the fish
0.5
assemblages and their environment. While the sea-
sonal dynamics of mangrove use by fishes has received
0.4
attention, only a few studies have focused on shorter
0.3
temporal scales. In the future, researchers should bet-
ter describe how and why sampling locations were
0.2
selected, and when possible include the rationale
0.1
behind their sampling intensity/allocation decisions.
How fish samples have been acquired from man-
0.0
In rns
y
eo ar.
hy
n
.
e
groves is of particular importance due, in part, to issues
gt
or
yp
io
v
ap
m
at
tte
nt
rt
y
of species- and size-selectivity. Indeed, one of the
gr
or
ve
er
or
ea
Pa
st
at
at
G
Re
W
an
major reasons mangroves have received relatively lit-
og
pl
Bi
Ex
tle attention as fish habitats is that it is inherently diffi-
cult to quantitatively sample fishes within them. Con-
b) Frequency
0.5
sequently, our understanding of the role(s) that these
Proportion of studies
habitats play in the lives of fishes has been hindered by
0.4
the fact that the same sampling methods have rarely
been used from one study to the next. Over one-third
0.3
of all studies we reviewed used towed gears, which
0.2 is a consistent finding among geographic regions
(Fig. 4a,b). While towed nets can be effective in sea-
0.1 grass beds, they are of little or no use within the dense,
rigid, entangled roots of mangrove trees. Use of pas-
0.0
lly
y
y
ly
ly
Va ly
us
d
al
l
rte
th
th
ht
ht
a) Gears: type
ua
rio
on
on
on
ig
ig
po
nn
0.35
as
rtn
rtn
-m
M
re
i-a
Se
Fo
Fo
Bi
ot
m
0.30
N
<
Se
0.25
c) Duration
0.25 0.20
0.15
0.20
0.10
0.15
0.05
Proportion of studies
0.00
0.10
ed
ve
e
al
t
s
en
er
ur
su
i
w
ss
gl
m
os
Vi
To
0.05
An
Pa
le
cl
ng
En
ta
En
0.00
b) Gears: region
5
0
5
0
5
0
3+
us
d
0.
1.
1.
2.
2.
3.
rte
1.0
rio
po
Anglers
Va
re
ot
0.8
N
Entanglement
Fig. 3. (a) Purpose, (b) sampling frequency, and (c) sampling
0.6
duration (in yr) of studies of mangrove fishes. Var: variable; Visual
mgt: management. Note difference in y-axis scales
0.4 Enclosure
choice of sampling locations may be biased, which in Passive
0.2
turn can be reflected in the data gathered (Rozas &
Towed
Minello 1997). In addition, the limited spatial extent 0.0
AM AU EA SA WA
and/or small number of sites sampled in most studies
Region
may not be representative of a given area or region,
Fig. 4. The gear type used to collect fishes from mangrove
and thus may be of limited use to coastal resource
habitats (a) among studies, and (b) by geographic region. AM:
managers if they must determine the costs and benefits Americas; AU: Australasia; EA: East Africa and the Middle
of developing one mangrove area over another. Simi- East; SA: South and Southeast Asia; WA: West Africa. Note
larly, the lack of multi-year studies precludes assess- difference in y-axis scales
9
Faunce & Serafy: Mangrove-fish literature review
sive gears has also been common (27% of studies). coral reef environments (e.g. Lindeman & Snyder 1999,
Most passive gears are difficult to place in dense man- Nagelkerken et al. 2000b, Russ et al. 2005). An impor-
grove prop-roots without constituting additional struc- tant distinction between visual surveys and other
ture, and like towed gears are often actually employed methods is that the former does not capture fishes, and
at the periphery of the mangrove shoreline. This con- thus is advantageous for studying threatened or
sidered, one out of 5 purported studies of mangrove endangered species. Limitations of visual surveys stem
fishes that we reviewed failed to sample within man- from variations in visibility, size- and species-specific
grove habitat. This has undoubtedly produced unrep- responses of fish to those performing the survey,
resentative data in specific cases, and has probably led observer experience, recording errors, as well as safety
to unfounded conclusions about the nature and extent concerns (Cheal & Thompson 1997, Thompson & Map-
of fish utilization of mangroves in general. stone 1997, Ley et al. 1999). While precision varies by
There are notable exceptions, however, where more methodology (e.g. roving, timed, or belt transect, writ-
appropriate sampling techniques have been applied to ten or audio recording media), accuracy problems can
quantitatively sample fish within the mangroves be reduced by performing observer training and using
proper. These include enclosure, entanglement, and a limited number of personnel (Bell et al. 1985, Greene
visual techniques, which have each been employed & Alevizon 1989, St. John et al. 1990). Visual survey
with similar frequency (i.e. 11 to 13% of studies). A techniques were used far more often in studies con-
common type of enclosure gear uses a fine mesh net to ducted in the Americas (Caribbean), than in either
encompass a mangrove area and fishes are subse- Australasia or East Africa, and were absent from South
quently removed with poison and/or a smaller net (e.g. and Southeast Asia and West Africa (Fig. 4b).
Bell et al. 1984, Thayer et al. 1987, Blaber & Milton Given the above, there is clearly no ‘best’ method for
1990). Such block netting can result in estimates of sampling fishes within mangrove habitats. The optimal
both abundance and biomass per unit area (standing method will vary according to study constraints and
crop), and is an especially effective method for collect- have bias and precision that should be weighed in
ing small, cryptic fishes. However, sampling efficiency accordance with the goals of the project. In studies that
is dependent on the clearing method used. Two draw- focus on analyzing the entire assemblage, the applica-
backs of enclosure samplers are that they often involve tion of multiple gear types has been used with success
both short- and long-term disturbance of the habitat and is preferred (Blaber et al. 1985). With respect to
under study (e.g. prop-root and canopy removal), and single gear studies, we agree with Rozas & Minello
are relatively labor intensive. In contrast, passive sam- (1997) that enclosure samplers are superior for quanti-
plers such as fyke, flume, or channel nets do not fying fishes in structurally-complex habitats, especially
greatly modify the mangrove habitat, but are limited to in turbid waters, and add that visual surveys are par-
situations where tides are sufficient to drain the habitat ticularly useful in clearer waters.
and effectively force fishes into these capture devices
(McIvor & Odum 1986). Most traps, like entanglement
gears, can be rapidly deployed and cause minimal Habitat metrics
habitat disturbance. These gears can also be effective
for catching relatively large (ca. 10 cm total length) An historical summary of abiotic and biotic measures
mobile fishes often missed by other gears; however, collected aids our present ability to assess the value of
there are size- and species-selectivity constraints. Size mangroves as fish habitat. Unlike experimental studies
selectivity problems can be reduced by sampling with that benefit from being able to isolate and manipulate
traps of different mesh sizes and openings, or by specific variables for study, in the field it is difficult to
sampling with nets with multiple panels composed of choose appropriate abiotic (habitat) factors to measure
different meshes (Sogard et al. 1989, Sheaves 1995). since they are often autocorrelated with one another.
Unfortunately, like all passive samplers, only relative While the appropriate environmental variables mea-
abundance or biomass, rather than density or biomass sured in a study should differ depending on the goals
per unit area, can be estimated using traps and entan- of the project and on local conditions, over half of the
glement nets. Finally, underwater visual fish census habitat metrics recorded in the literature database con-
can be a rapid and effective technique for gathering sisted of temperature and salinity measurements, and
data and making quantitative comparisons of fish dis- this trend was observed across all geographic regions
tribution, abundance, and size-structure within and (Fig. 5a,b). Certainly temperature is linked to broad
among habitat types. Visual fish census has been uti- spatial patterns in the use of mangroves by fishes, as
lized in seagrass beds, mangroves, and hardbottom more fish species are noted from tropical estuaries
communities, and has become the most accepted than from sub-tropical estuaries, and a positive corre-
method for estimating fish abundance and diversity in lation between temperature and overall assemblage
10 Mar Ecol Prog Ser 318: 1–18, 2006
b) Habitat metrics: region
a) Habitat metrics: type
0.30 1.0
pH
0.25 0.8 Structure
Rain
0.20
0.6 DO
0.15 Turbidity
0.4 Depth
0.10
Temperature
0.2
0.05 Salinity
Proportion of studies
0.00 0.0
AM AU EA SA WA
ity
p
h
ity
DO
in
e
pH
ur
pt
m
Ra
lin
id
ct
Te
De
Region
rb
Sa
ru
Tu
St
d) Fish metrics: region
c) Fish metrics: type 1.0
0.35
0.30 G/Z
0.8
FO
0.25
Standing crop
0.6
0.20 Density
Biomass
0.15 0.4 Size
Percent
0.10
0.2 Presence
0.05
0.0
0.00
AM AU EA SA WA
ce
t
ze
s
ity
op
FO
/Z
n
en
as
tio
G
Si
ns
en
cr
Region
rc
om
uc
De
es
Pe
ng
od
Bi
Pr
di
Pr
an
St
Fig. 5. Summary of (a,b) habitat and (c,d) fish metrics collected from mangroves (a,c) among studies and (b,d) by geographic
region. DO: Dissolved oxygen; FO: frequency of occurrence; G/Z: growth and/or mortality estimates; rain: rainfall; temp:
temperature. Geographic regions abbreviated as in Fig. 4
richness, diversity, and abundance has been noted by ture on mangrove fishes (Connell 1978, Leigh 1990).
several authors (Robertson & Duke 1987, Williamson However, there is scant evidence suggesting such
et al. 1994, Lin & Shao 1999). While seasonal and relationships may also hold for fishes inhabiting
diel changes in temperature are typically predictable, mangroves (Serafy et al. 2003). Although regime
changes in salinity within a mangrove habitat can be characterization requires that dynamic abiotic vari-
more dynamic. Salinity can either remain relatively ables are measured over longer periods of time than
stable throughout the year (e.g. along well-connected just on the day of fish sampling, few studies in the liter-
oceanic islands), exhibit seasonal changes resulting ature have examined fish-habitat relationships on
from fluvial runoff, or change dramatically as a result multiple time scales (Bell et al. 1984, Lorenz 1999,
of anthropogenic freshwater releases (Faunce et al. Faunce et al. 2004).
2004). In addition, observed fish patterns in mangrove As in the case for temporal scales, examination of
habitats may be correlated with: (1) water depth, multiple spatial scales may be integral to determining
where the habitat is temporally inundated (e.g. which fishes utilize mangrove habitats and why; yet
Robertson & Duke 1990a, Laegdsgaard & Johnson this has also been largely ignored in the literature on
1995); (2) turbidity, where sediment transport is (a) mangrove fishes. At the smallest scales, structural
high (e.g. Blaber 1980) or (b) in areas without large complexity may be important, but this was reported in
salinity fluxes; and (3) dissolved oxygen in areas less than 5% of studies. It is likely that few field-based
with poor water flow or located downstream of large studies measured and reported structural measure-
industrial or agricultural areas (e.g. Claro et al. 2001). ments because early attempts failed to find meaning-
Abiotic regime (i.e. mean, range, and stability) may ful correlations with fish measures (Sheridan 1992,
be of more importance in structuring the assemblage Mullin 1995). Interestingly, experimental studies have
of mangrove fishes than ‘snapshot’ metrics collected demonstrated that the increased structural complexity
at the time of sampling. For example, a negative rela- of mangroves reduces the efficiency of predators (Pri-
tionship between environmental stability and species mavera 1997, Laegdsgaard & Johnson 2001). At larger
diversity has been well documented outside the litera- scales, many studies have sampled sites located at
11
Faunce & Serafy: Mangrove-fish literature review
various distances from features upstream (e.g. fresh Duke 1990b), and no estimates of habitat-specific mor-
water) versus downstream (e.g. coral reefs). However, tality or production have appeared in the literature.
distance values were reported in less than 10% of Because such studies may have been specifically
studies, and in only 2 instances were they used in focused on these biological metrics, it is possible that
analyses with fish metrics (Nagelkerken et al. 2000b, such studies exist outside the realm of this review.
Hajisamae & Chou 2003). Such analyses can be per- Other biotic factors such as larval supply, predation,
formed readily given the advances in global posi- competition, and food supply are difficult to consis-
tioning satellites and geographic information system tently and reliably measure, and we were unable to
technology (e.g. Kendall et al. 2003). find studies that reported these measures in the litera-
ture on mangrove fishes.
Fish metrics
Data analyses
Recently, several fish metrics have been reviewed
and ranked according to their usefulness for determin- Having summarized where, when, and how fish data
ing the importance of fish habitats, including man- have been collected from mangrove habitats, the ensu-
groves. In 1996, the US government mandated that all ing discussion is concerned with what has been done
stock assessments include EFH provisions and con- with them. Fish abundances were usually analyzed at
sider 4 levels of information (USDOC 1996). On a habi- one of 3 levels: the entire assemblage, ‘dominant taxa’
tat-specific basis, these include fish presence-absence (e.g. the 10 most abundant species grouped), or indi-
(Level 1), densities (Level 2), growth, reproduction, vidual species. A measure of the entire assemblage
and survival rates (Level 3), and secondary production (e.g. total fish density or biomass) was the most com-
rates (Level 4). A refined definition of ‘nursery habitat’ monly used level of analysis (52%), followed by the
emerged with a paper by Beck et al. (2001). They analysis of dominant taxa (31.2%) and then individual
contend that a nursery habitat contains one or more of species abundances (16.8%). As most studies aimed to
the following traits compared to other non-nursery identify patterns of fish use, analyses were focused on
habitats: (1) greater densities of young fishes; (2) lower examining temporal and spatial variation (Fig. 6a).
predation rates; (3) higher growth rates; and (4) more Fish–habitat correlations were examined less often,
successful migration to subsequent habitats (Beck et and were completely absent from the West African
al. 2001). region. Regardless of the focus, the type of data analy-
In light of these developments, this literature review sis conducted was typically in the form of simple
can be used to answer the question: What information side-by-side comparisons (Fig. 6b). Similarity indices,
is available to assess the value of mangroves as fish ANOVA, and ordination techniques were applied
habitat? Presence/absence information was the most with equal frequency among studies with spatial or
widely reported form of fish data, followed closely by temporal emphasis, and less often in studies with a
percentage composition (Fig. 5c). These 2 metrics fish-habitat focus; the latter investigation type alone-
accounted for over half the reported entries (31 and utilized multiple linear regression techniques.
24%, respectively) and were available from almost all Simple data comparisons dominated the literature.
surveys of mangrove fishes that we examined. Size Comparisons of fish data by family (38.1%) predomi-
information was less prevalent, and was present most nated, probably because this information is presented
often as part of a description of collected fishes. Only in species lists (Fig. 7a). Comparing fish groups
1 publication presented detailed size information for according to life-history (maturity) stage or as either
several species over time (Robertson & Duke 1990b). residents or transients (residency) was also common
Biomass information was more prevalent in the litera- (23.7 and 19.5%, respectively). Comparisons accord-
ture than either density or standing crop (i.e. numbers ing to trophic groups and diel period were among the
and biomass per unit area, respectively), both of which least reported in the database. The characterization
require information about the area sampled (Fig. 5c). and comparison of fishes according to their trophic
Remarkably, frequency of occurrence information (i.e. level can be a valuable tool for revealing their role in
the proportion of sites or repeated samples that con- system energy flow. The concept of using such func-
tained at least 1 individual), available from any survey, tional groups as a basis for site and ecosystem com-
went unreported in > 90% of studies. Density, standing parison and evaluation has recently been reviewed
crop, and frequency of occurrence data have not been for coral reefs (Bellwood et al. 2004) and holds
reported from the West African region (Fig. 5d). Per- promise for application to mangrove systems as well.
haps most limiting to mangrove fish habitat assess- Evidence is mounting that mangroves primarily serve
ment is that only 1 estimate of growth (Robertson & as daytime refugia for a major component of fishes
12 Mar Ecol Prog Ser 318: 1–18, 2006
b) Analysis type
a) Analysis focus
0.6
1.0
Habitat
Habitat
Proportion of studies
Spatial
0.5
0.8
Temporal
Spatial 0.4
0.6
Temporal 0.3
0.4
0.2
0.2 0.1
0.0
0.0
s
y
n
ns
VA
AM AU EA SA WA
on
rit
tio
o
O
ila
ris
si
na
Region
AN
m
es
pa
i
rd
Si
gr
om
O
Re
C
Fig. 6. Data analyses performed within studies of mangrove fishes organized by (a) geographic region and (b) type (i.e. statistical
test). Data are organized according to the purpose of analyses: identifying spatial patterns (spatial), identifying temporal patterns
(temporal), or exploring fish–habitat interactions (habitat). Non-statistical comparisons of data types (e.g. density) are labeled
comparisons. Similarity refers to indices of similarity, diversity, and evenness. Geographic regions abbreviated as in Fig. 4
occupying mangrove shorelines (Rooker & Dennis data from 19 studies that quantified fishes within
1991, Nagelkerken et al. 2000a, Valdés-Muñoz & mangroves and at least one other habitat.
Mochek 2001). This suggests for some species that
fish production attributed to mangroves may not nec-
essarily derive from this habitat alone (Adams et al. Current limitations
2006). Linkages between mangrove shorelines and
the proximity, size, and availability of nocturnal forag- Our review reveals that (1) certain regions, specifi-
ing areas, such as seagrass beds or mudflats, deserve cally South and Southeast Asia and West Africa, are
greater attention. under-represented in the literature, (2) the majority of
surveys were spatially restrictive and/or of short dura-
tion, and (3) numerous purported surveys of mangrove
fishes failed to sample within mangrove habitats per
SYNOPSIS
se. In defense of these studies, most were designed
This work represents the first attempt to assemble with modest goals in mind: (1) to identify which taxa
and examine a substantial number of published studies were present, and (2) to determine their abundances
of mangrove fishes. In contrast, previous reviews of among locations and/or sequential samples. While
the literature on mangrove fishes have been more these studies have been useful in identifying the com-
limited in scope. Thayer & Sheridan (1999) examined ponents and dynamics of various assemblages of man-
the methods and results of less than 12 studies from grove fishes, this type of data provides little informa-
Florida (USA), while Sheridan & Hays (2003) compiled tion with which to compare and evaluate the
a) Fish comparisons: type b) Fish comparisons: region
0.5 1.0
Proportion of studies
Diurnal
0.4 0.8
Trophic
0.3 0.6
Residency
0.2 0.4
Maturity
0.1 0.2
Family
0.0 0.0
AM AU EA SA WA
ily
y
y
c
l
na
it
nc
hi
m
ur
op
ur
Region
de
Fa
at
Di
Tr
si
M
Re
Fig. 7. How fish metrics were grouped for data comparisons (a) among studies, and (b) by geographic region. Geographic regions
abbreviated as in Fig. 4
13
Faunce & Serafy: Mangrove-fish literature review
importance of mangrove habitats for ecosystem diver- Future directions
sity and productivity.
Where fish–habitat correlations have been exam- The limitations above are important to consider
ined, most studies performed analyses at the assem- when planning future studies. Given the history of the
blage level using environmental data obtained at the literature, it appears likely that future studies will
time of sampling. Consequently, there is meagre infor- continue to examine its spatio-temporal patterns of
mation on how individual species respond to environ- mangrove use by fishes. In general, studies that exam-
mental variability. This is unfortunate, given that recent ine fish and habitat features at multiple spatial and
works suggest that the nature of the relationship be- temporal scales will be more valuable than those that
tween habitat features and fishes is species- and size- examine at only one scale. Irrespective of scale, studies
specific, and that only with this level of understanding that examine both the mean and variance of abiotic
can insight into ecological processes be gained (Bene- and biotic metrics will provide more insight than those
detti-Cecchi 2003, Cocheret de la Morinière et al. that only consider ‘static’ measures at the time of
2004). Furthermore, the type of fish metrics reported sampling. Power analysis and/or sampling efficiency
did not often include those most desired by decision evaluation are exceedingly rare in the literature, but
makers. For example, although simple presence/ are needed for mangrove fishes research to gain the
absence and percentage composition data were com- attention it deserves (Ley et al. 1999). It is possible that
monly reported, size and frequency of occurrence were statistical treatments such as these have been lacking
not, despite the fact that these metrics should be avail- in the literature partly because, at the species-specific
able from any survey that repeatedly samples fishes. level, ‘zero-laden’ fish abundance datasets are typical
The more detailed, difficult-to-collect data needed for and less than desirable for traditional statistical treat-
mangrove valuation were effectively absent from the ments. However, the problem of rarity is widespread in
database. With possibly 1 exception (i.e. Robertson & ecological studies (Gaston 1994), including those on
Duke 1990b), no study of mangrove fishes recorded ad- coral reef fishes (Jones et al. 2002). Researchers exam-
equate numbers over time, numbers at age, size-fre- ining species-specific fish abundance data will benefit
quencies, or tag-recapture data needed for the tradi- from the work of Aitchison (1955), Pennington (1983),
tional assessment of growth, mortality, or secondary Lo et al. (1992) and Johnson et al. (1999).
production. As a result, comparisons of EFH or nursery Several scientists are moving the study of mangrove
data from studies of mangrove fishes will likely be lim- fishes beyond pattern recognition towards more eco-
ited to density or biomass values only. If the USDOC logically meaningful landscape-scale approaches,
(1996) or Beck et al. (2001) definitions of these terms are including habitat connectivity, suitability, and the
to be used to assess the habitat value of mangroves, fu- contribution of mangrove habitats in support of adult fish
ture efforts must strive to track cohorts of fish over populations (e.g. Pittman et al. 2004, Sheaves 2005,
space and time. Mumby 2006). However, the need for species- and life-
Among the most significant conclusions to draw from stage-specific information on growth, mortality, and sec-
this review is that surveys of mangrove fishes are not ondary production rates remains. Although sufficient
readily comparable. Hence, the findings of any new age–length, biomass and size-distribution data exist for
study may be either bolstered or refuted using selected several species to generate habitat-specific production
references from the relevant literature. For example, estimates, this step has yet to be taken for mangrove
findings are mixed in studies relating habitat features to fishes. Attaining accurate home ranges and movement
assemblages of mangrove fishes with respect to water rates for mangrove fishes represents an additional and
temperature (Wright 1986, Lin & Shao 1999), salinity significant challenge towards linking juvenile and adult
(Quinn 1980, Ikejima et al. 2003), and turbidity (Little et stocks. Studies that examine habitat quality and avail-
al. 1988b, Kimani et al. 1996). Different conclusions re- ability are also needed to determine what makes some
garding the fish assemblage may be reached even when mangroves more important fish habitats than others.
2 studies have been conducted within the same body of
water. For example, Williamson et al. (1994), who sam-
Acknowledgements. This work was derived from a portion of a
pled with 8 mm mesh beach seine in Raby Bay, Australia,
doctoral dissertation by C. H. F. We thank the present and past
reported, ‘the majority of fish captures were either small library staff of the University of Miami’s Rosenstiel School of
species or juveniles’, while Moreton (1990) — who sam- Marine and Atmospheric Science (RSMAS), especially Helen
Albertson and Gail Clement. We also appreciate the efforts of
pled with 18 mm mesh seines and 100 to 150 mm mesh
numerous students and staff of the international community at
gill nets — stated, ‘most species .… were present as both
RSMAS for translating numerous publications. The comments
juveniles and adults’ and ‘standing-crop estimates for and additional literature sources provided by 4 anonymous
the fishes occurring within the mangroves were amongst reviewers greatly improved this manuscript. This paper is
the highest recorded values for estuarine areas’. Sustainable Fisheries Division Contribution SFD-2006-024.
14 Mar Ecol Prog Ser 318: 1–18, 2006
LITERATURE CITED Blaber SJM, Blaber TG (1980) Factors affecting the distribution
of juvenile estuarine and inshore fish. J Fish Biol 17:143–162
Acosta A (1997) Use of multi-mesh gillnets and trammel nets to Blaber SJM, Milton DA (1990) Species composition, commu-
estimate fish species composition in coral reef and man- nity structure and zoogeography of fishes of mangrove
groves in the southwest coast of Puerto Rico. Caribb J Sci estuaries in the Solomon Islands. Mar Biol 105:259–267
33:45–57 Blaber SJM, Young JW, Dunning MC (1985) Community struc-
Adams A, Dahlgren C, Kellison GT, Kendall MS, Layman CA, ture and zoogeographic affinities of the coastal fishes of the
Ley JA, Nagelkerken I, Serafy JE (2006) The juvenile con- Dampier Region of north-western Australia. Aust J Mar
tribution function of back reef systems. Mar Ecol Prog Ser Freshw Res 36:247–266
318:287–301 Blaber SJM, Brewer DT, Salini JP (1989) Species composition
Aitchison J (1955) On the distribution of a positive random vari- and biomasses of fishes in different habitats of a tropical
able having a discrete probability mass at the origin. J Am northern Australian estuary: their occurrence in the adjoin-
Stat Assoc 50:901–908 ing sea and estuarine dependence. Estuar Coast Shelf Sci
Alongi DM (2002) Present state and future of the world’s 29:509–531
mangrove forests. Environ Conserv 29:331–349 Blaber SJM, Brewer DT, Harris AN (1994) Distribution, bio-
Alvarez León R, Blanco Racedo J (1985) Composition of fish mass and community structure of demersal fishes of the
communities in the lagoon and estuarine complex of Carta- Gulf of Carpentaria, Australia. Aust J Mar Freshw Res 45:
gena Bay, Cienaga de Tesca and Cienaga Grande of Santa 375–396
Marta, Colombian Caribbean. In: Yáñez-Arancibia A (ed) Boeseman M (1963) An annotated list of fishes from the Niger
Fish community ecology in estuaries and coastal lagoons: delta. Zool Verh 61:1–54
towards an ecosystem integration. UNAM Press, Mexico Boltachev AR (1991) Species composition in the estuarine
City, p 535–656 ichthyocene of the mangrove zone of Guinea. J Ichthyol
Amadi AA, Kilham P, Mavuti KM (1990) A comparative ecol- 31:67–75
ogy of estuaries in Nigeria. Hydrobiologia 208:27–38 Boujard T, Rojas-Beltran R (1988) Longitudinal zonation of the
Araujo FG, De Azevedo MCC, Silva MA, Pessanha ALM, fish population of the Sinnamary River (French Guiana).
Gomes ID, Da Cruz-Filho AG (2002) Environmental influ- Rev Hydrobiol Trop 21:47–61
ences on the demersal fish assemblages in the Sepetiba Boulon RH Jr (1992) Use of mangrove prop root habitats by fish
Bay, Brazil. Estuaries 25:441–450 in the northern US Virgin Islands. Proc Gulf Caribb Fish
Atiqullah M, Shaukat SS, Qureshi N (1997) Diversity of fish Inst 41:189–204
communities at Sandspit Karachi coast, Pakistan. Oebalia Branch GM, Grindley JR (1979) Ecology of southern African
23:45–59 estuaries. Part XI. Mngazana: a mangrove estuary in
Austin HM (1971) A survey of the ichthyofauna of the man- Transkei. S Afr J Zool 14:149–170
groves of western Puerto Rico during December, 1967– Cervigón F (1985) The ichthyofauna of the Orinoco estuarine
August, 1968. Caribb J Sci 11:27–39 water delta in the West Atlantic coast, Caribbean. In:
Baran E, Hambrey J (1998) Mangrove conservation and coastal Yáñez-Arancibia A (ed) Fish community ecology in estuar-
management in southeast Asia: What impact on fishery ies and coastal lagoons: towards an ecosystem integration.
resources? Mar Pollut Bull 37:431–440 UNAM Press, Mexico City, p 57–78
Barletta-Bergan A, Barletta M, Saint-Paul U (2002) Community Chaves P, Bouchereau JL (1999) Biodiversity and dynamics
structure and temporal variability of ichthyoplankton in of ichthyic communities in the mangrove of Guaratuba,
North Brazilian mangrove creeks. J Fish Biol 61:33–51 Brazil. Oceanol Acta 22:353–364
Batista VS, Rego FN (1996) Análise de associações de peixes, Chaves P, Correa MFM (1998) Composicao ictiofaunistical da
em igarapés do estuário do Rio Tibiri, Maranhão. Rev Bras area de manguezal da Baia de Guaratuba, Parana, Basil.
Biol 56:163–176 Rev Bras Biol 15:195–202
Beck MW, Heck KL, Able KW, Childers DL and 9 others (2001) Chaves TC, Vendel AL (2001) Nota complementar sobre a
The identification, conservation, and management of estu- composicao ictiofaunistica da Baia de Guaratuba, Parana,
arine and marine nurseries for fish and invertebrates. Bio- Brasil. Rev Bras Biol 18:349–352
Science 51:633–641 Cheal AJ, Thompson AA (1997) Comparing visual counts of
Bell JD, Pollard DA, Burchmore JJ, Pease BC, Middleton MJ coral reef fish: implications of transect width and species
(1984) Structure of a fish community in a temperate tidal selection. Mar Ecol Prog Ser 158:241–248
mangrove creek in Botany Bay, New South Wales. Aust J Chong VC, Sasekumar A, Leh MUC, D’Cruz R (1990) The fish
Mar Freshw Res 35:33–46 and prawn communities of a Malaysian coastal mangrove
Bell JD, Craik GJS, Pollard DA, Russell BC (1985) Estimating system, with comparisons to adjacent mud flats and inshore
length frequency-distributions of large reef fish under- waters. Estuar Coast Shelf Sci 31:703–722
water. Coral Reefs 4:41–44 Claro R, García-Arteaga JP (1993) Fish community structure on
Bellwood DR, Hughes TP, Folke C, Nystrom M (2004) Con- mangroves from the Insular Group Sabana-Camaguey,
fronting the coral reef crisis. Nature 429:827–833 Cuba. Avicennia 0:60–83
Benedetti-Cecchi L (2003) The importance of the variance Claro R, Lindeman KC, Parenti LR (2001) Ecology of the marine
around the mean effect size of ecological processes. fishes of Cuba. Smithsonian Institution Press, Washington,
Ecology 84:2335–2346 DC
Blaber SJM (1980) Fish of the Trinity Inlet system of north Cocheret de la Morinière E, Pollux BJA, Nagelkerken I, van
Queensland with notes on the ecology of fish faunas of der Velde G (2002) Post-settlement life cycle migration pat-
tropical Indo-Pacific estuaries. Aust J Mar Freshw Res 31: terns and habitat preference of coral reef fish that use sea-
137–146 grass and mangrove habitats as nurseries. Estuar Coast
Blaber SJM (2000) Tropical estuarine fishes: ecology, exploita- Shelf Sci 55:309–321
tion, and conservation. Blackwell Science, London Cocheret de la Morinière E, Nagelkerken I, van der Meij H,
Blaber SJM (2002) ‘Fish in hot water’: the challenges facing fish van der Velde G (2004) What attracts juvenile coral reef fish
and fisheries research in tropical estuaries. J Fish Biol 61:1–20 to mangroves: habitat complexity or shade? Mar Biol 144:
15
Faunce & Serafy: Mangrove-fish literature review
139–145 Johnson DR, Funicelli NA, Bohnsack JA (1999) Effectiveness of
Collette BB (1983) Mangrove fishes of New Guinea. In: Teas HJ an existing estuarine no-take fish sanctuary within the
Biology and ecology of mangroves. Dr. W. Junk Publishers, Kennedy Space Center, Florida. North Am J Fish Manag
The Hague, p 91–102 19:436–453
Connell JH (1978) Diversity in tropical rain forests and coral Jones GP, Caley MJ, Munday PL (2002) Rarity in coral reef fish
reefs. Science 199:1302–1310 communities. In: Sale PF (ed) Coral reef fishes: dynamics
Dankwa HR, Gordon C (2002) The fish and fisheries of the and diversity in a complex ecosystem. Academic Press, San
Lower Volta mangrove swamps in Ghana. Afr J Sci Technol Diego, CA, p 81–102
(Sci Eng Ser) 3:25–32 Kendall MS, Christensen JD, Hillis-Starr Z (2003) Multi-scale
Davis TLO (1988) Temporal changes in the fish fauna entering data used to analyze the spatial distribution of French
grunts, Haemulon flavolineatum, relative to hard and soft
a tidal swamp system in tropical Australia. Environ Biol
Fish 21:161–172 bottom in a benthic landscape. Environ Biol Fish 66:19–26
Day JH (1974) The ecology of Morrumbene estuary, Mozam- Kimani EN, Mwatha GK, Wakwabi EO, Ntiba JM, Okoth BK
bique. Trans R Soc S Afr 41:43–97 (1996) Fishes of a shallow tropical mangrove estuary, Gazi,
Dennis GD (1992) Island mangrove habitats as spawning and Kenya. Mar Freshw Res 47:857–868
nursery areas for commercially important fishes in the Krishnamurthy K, Jeyasslan MJP (1981) The early life history
Caribbean. Proc Gulf Caribb Fish Inst 41:205–225 of fishes from Pichavaram mangrove ecosystem of India. In:
Faunce CH, Serafy JE, Lorenz JJ (2004) Density-habitat rela- Lasker R, Sherman K (eds) The early life history of fish:
tionships of mangrove creek fishes within the southeastern recent studies. Second ICES Symposium. Cons Int Explor
saline Everglades (USA), with reference to managed fresh- Mer, Palegade 2–4, Copenhagen, p 416–423
water releases. Wetlands Ecol Manag 12:377–394 Kuo S, Lin H, Shao K (1999) Fish assemblages in the mangrove
Flores-Verdugo F, González-Farias F, Ramírez-Flores O, creeks of northern and southern Taiwan. Estuaries 22:
Amezcua-Linares A, Yáñez-Arancibia A, Alvarez-Rubio M, 1004–1015
Day JW Jr (1990) Mangrove ecology, aquatic primary pro- Laegdsgaard P, Johnson CR (1995) Mangrove habitats as nurs-
ductivity, and fish community dynamics in the Teacapan- eries: unique assemblages of juvenile fish in subtropical
Agua Brava lagoon-estuarine system (Mexican Pacific). mangroves in eastern Australia. Mar Ecol Prog Ser 126:67–81
Estuaries 13:219–230 Laegdsgaard P, Johnson CR (2001) Why do juvenile fish utilize
Gaston KJ (1994) Rarity. Chapman & Hall, London mangrove habitats? J Exp Mar Biol Ecol 257:229–253
Gillanders BM, Able KW, Brown JA, Eggleston DB, Sheridan Lal PN (1984) Coastal fisheries and the management of man-
PF (2003) Evidence of connectivity between juvenile and grove resources in Fiji. SPC Fish Newsl:15–23
adult habitats for mobile marine fauna: an important com- Laroche J, Baran E, Rasoanandrasana NB (1997) Temporal pat-
ponent of nurseries. Mar Ecol Prog Ser 247:281–295 terns in a fish assemblage of a semiarid mangrove zone in
Gonzalez-Acosta AF, De la Cruz Agüero J, Ruiz-Campos G Madagascar. J Fish Biol 51:3–20
(1999) Ichthyofauna associated to El Conchalito mangrove Lasserre G, Toffart JL (1977) Echantillonnage et structure des
swamp, Ensenada de La Paz, Baja California Sur, Mexico. populations ichthyologiques des mangroves de Guade-
Investigaciones Marinas CICIMAR 14:121–131 loupe en Septembre 1975. Cybium 3:115–127
Greene LE, Alevizon WS (1989) Comparative accuracies of Leigh EG (1990) Community diversity and environmental
visual assessment methods for coral reef fishes. Bull Mar Sci stability— a reexamination. Trends Ecol Evol 5:340–344
44:899–912 Ley JA, McIvor CC (2001) Linkages between estuarine and
Hajisamae S, Chou LM (2003) Do shallow water habitats of an reef fish assemblages: enhancement by the presence of
impacted coastal strait serve as nursery grounds for fish? well-developed mangrove shorelines. In: Porter JW, Porter
Estuar Coast Shelf Sci 56:281–290 KG (eds) The Everglades, Florida Bay, and coral reefs of the
Halliday IA, Young WR (1996) Density, biomass and species Florida Keys: an ecosystem sourcebook. CRC Press, Boca
composition of fish in a subtropical Rhizophora stylosa Raton, FL, p 539–562
mangrove forest. Mar Freshw Res 47:609–615 Ley JA, McIvor CC, Montague CL (1999) Fishes in mangrove
He B, Fan H, Mo Z (2001) Study on species diversity of fishes in prop-root habitats of northeastern Florida Bay: distinct
mangrove area of Yingluo Bay, Guangxi province. J Trop assemblages across an estuarine gradient. Estuar Coast
Oceanogr 20:74–79 Shelf Sci 48:701–723
Heck JL, Hays G, Orth RJ (2003) Critical evaluation of the Lin HJ, Shao KT (1999) Seasonal and diel changes in a subtrop-
nursery role hypothesis for seagrass meadows. Mar Ecol ical mangrove fish assemblage. Bull Mar Sci:775–794
Prog Ser 253:123–136 Lindeman KC, Snyder DB (1999) Nearshore hardbottom fishes
Holm RF (1977) The standing crop of fishes in a tropical marine of southeast Florida and effects of habitat burial caused by
lagoon. Fla Sci 40:258–261 dredging. Fish Bull 97:508–525
Ikejima K, Tongnunui P, Medej T, Taniuchi T (2003) Juvenile Little MC, Reay PJ, Grove SJ (1988a) Distribution gradients
and small fishes in a mangrove estuary in Trang Province, of ichthyoplankton in an East African mangrove creek.
Thailand: seasonal and habitat differences. Estuar Coast Estuar Coast Shelf Sci 26:669–677
Shelf Sci 56:447–457 Little MC, Reay PJ, Grove SJ (1988b) The fish community of an
Inger RE (1955) Ecological notes on the fish fauna of a coastal East African mangrove creek. J Fish Biol 32:729–747
drainage of North Borneo. Fieldiana Zool 37:47–90 Lo NCH, Jacobson LD, Squire JL (1992) Indexes of relative
Janekarn V (1993) Species composition and annual population abundance from fish spotter data dased on Delta-Lognormal
growth of fishes in front of a mangrove in Phang-Nga Bay, models. Can J Fish Aquat Sci 49:2515–2526
the Andaman Sea, Thailand. Phuket Mar Biol Cent (Spec. Lorenz JJ (1999) The response of fishes to physicochemi-
Publ.) 12:131–140 cal changes in the mangroves of northeast Florida Bay.
Janekarn V, Boonruang P (1986) Composition and occurrence Estuaries 22:500–517
of fish larvae in mangrove areas along the east coast of Louis M, Bouchon C, Bouchon-Navaro Y (1992) L’ichtyofaune
Phuket Island, western peninsular, Thailand. Phuket Mar de mangrove dans la Baie de Fort-de-France (Martinique).
Biol Cent Res Bull 44:1–22 Cybium 16:291–305
16 Mar Ecol Prog Ser 318: 1–18, 2006
Louis M, Bouchon C, Bouchon-Navaro Y (1995) Spatial and tions in catch from two mangrove habitats in the Lagos
temporal variations of mangrove fish assemblages in Mar- Lagoon. Environ Ecol 13:121–128
tinique (French West Indies). Hydrobiologia 295:275–284 Odum WE, McIvor CC, Smith TJ, III (1982) Ecology of the
Louis M, Lam Hoai T, Lasserre G (1985) Résultats préliminaires mangroves of south Florida: a community profile. Office of
sur le recrutement en poissons dans deux lagunes des man- Biological Services, Fish and Wildlife Service, US Dept. of
groves de Guadeloupe: Belle-Plaine et Manche-à-Eau. Rev Interior, Washington, DC
Hydrobiol Trop 18:249–265 Paranagua MN, Eskinazi-Leça E (1985) Ecology of a northern
Lugendo BR, Pronker A, Cornelissen I, de Groene A, Nagel- tropical estuary in Brazil and technological perspectives
kerken I, Dorenbosch M, van der Velde G, Mgaya YD in fish culture. In: Yáñez-Arancibia A (ed) Fish community
(2005) Habitat utilisation by juveniles of commercially ecology in estuaries and coastal lagoons: towards an
important fish species in a marine embayment in Zanzibar, ecosystem integration. UNAM Press, Mexico City,
Tanzania. Aquat Living Resour 18:149–158 p 595–614
Lugo AE, Snedaker SC (1974) The ecology of mangroves. Pennington M (1983) Efficient estimators of abundance, for fish
Annu Rev Ecol Syst 5:39–64 and plankton surveys. Biometrics 39:281–286
Manly BFJ, McDonald LL, Thomas DL (1993) Resource selec- Phillips PC (1981a) Annotated checklist of fishes at Jiquilisco
tion by animals: statistical design and analysis for field Bay, El Salvador. Rev Biol Trop 29:45–58
studies. Chapman & Hall, London Phillips PC (1981b) Diversity and fish community structure in
Martínez AM, Contreras S, Maravilla O (1982) Abundancia y a central American mangrove embayment. Rev Biol Trop
diversidad de la ictiofauna, en tres areas de manglar de la 29:227–236
Bahía de la Paz, México. Trans Cibcasio 6:138–151 Pinto L (1987) Environmental factors influencing the occur-
McIvor CC, Odum WE (1986) The flume net: a quantitative rence of juvenile fish in the mangroves of Pagbilao, Philip-
method for sampling fishes and macrocrustaceans on tidal pines. Hydrobiologia 150:283–301
marsh surfaces. Estuaries 9:219–224 Pinto L, Punchihewa NN (1996) Utilization of mangroves and
Morton RM (1990) Community structure, density and standing seagrasses by fishes in the Negombo Estuary, Sri Lanka.
crop of fishes in a subtropical Australian mangrove area. Mar Biol 126:333–345
Mar Biol 105:385–394 Pittman SJ, McAlpine CA, Pittman KM (2004) Linking fish
Mullin SJ (1995) Estuarine fish populations among red man- and prawns to their environment: a hierarchical landscape
grove prop roots of small overwash islands. Wetlands 15: approach. Mar Ecol Prog Ser 283:233–254
324–329 Pollard DA, Hannan JC (1994) The ecological effects of
Mumby PJ (2006) Connectivity of reef fish between mangroves structural flood mitigation works on fish habitats and fish
and coral reefs: algorithms for the design of marine communities in the lower Clarence River system of south-
reserves. Biol Conserv 128:215–222 eastern Australia. Estuaries 17:427–461
Nagelkerken I, van der Velde G (2002) Do non-estuarine Primavera JH (1997) Fish predation on mangrove-associated
mangroves harbour higher densities of juvenile fish than penaeids: the role of structures and substrate. J Exp Mar
adjacent shallow-water and coral reef habitats in Curaçao Biol Ecol 215:205–216
(Netherlands Antilles)? Mar Ecol Prog Ser 245:191–204 Quinn NJ (1980) Analysis of temporal changes in fish assem-
Nagelkerken I, Dorenbosch M, Verberk WCEP, Cocheret de la blages in Serpentine Creek, Queensland. Environ Biol Fish
Morinière E, van der Velde G (2000a) Day–night shifts of 5:117–133
fishes between shallow-water biotopes of a Caribbean bay, Quinn NJ, Kojis BJ (1985) Does the presence of coral reefs in
with emphasis on the nocturnal feeding of Haemulidae and proximity to a tropical estuary affect the estuarine fish
Lutjanidae. Mar Ecol Prog Ser 194:55–64 assemblage? In: Harmelin-Vivien M, Salvat B, La Croix C,
Nagelkerken I, Dorenbosch M, Verberk WCEP, Cocheret de la Gabrie C, Toffart JL (eds) Proc Fifth Int Coral Reef Congr,
Morinière E, van der Velde G (2000b) Importance of shal- Tahiti, 27 May–1 Jun 1985, Miscellaneous papers (A),
low-water biotopes of a Caribbean bay for juvenile coral Vol 5. Antenne Museum-EPHE, Moorea, p 445–450
reef fishes: patterns in biotope association, community Robertson AI, Duke NC (1987) Mangroves as nursery sites:
structure and spatial distribution. Mar Ecol Prog Ser comparisons of the abundance and species composition of
fish and crustaceans in mangroves and other nearshore
202:175–192
habitats in tropical Australia. Mar Biol 96:193–205
Nagelkerken I, van der Velde G, Gorissen MW, Meijer GJ,
Robertson AI, Duke NC (1990a) Mangrove fish-communities in
van’t Hof T, den Hartog C (2000c) Importance of man-
tropical Queensland, Australia: spatial and temporal pat-
groves, seagrass beds and the shallow coral reef as a nurs-
terns in densities, biomass and community structure. Mar
ery for important coral reef fishes, using a visual census
Biol 104:369–379
technique. Estuar Coast Shelf Sci 51:31–44
Robertson AI, Duke NC (1990b) Recruitment, growth and resi-
Nagelkerken I, Kleijnen S, Klop T, van den Brand RACJ,
dence time of fishes in a tropical Australian mangrove
Cocheret de la Morinière EC, van der Velde G (2001)
system. Estuar Coast Shelf Sci 31:723–743
Dependence of Caribbean reef fishes on mangroves and
Rojas JRM, Castro M, Pizarro JF (1994a) Lista agregada de
seagrass beds as nursery habitats: a comparison of fish fau-
peces en tres zonas de manglar del Golfo de Nicoya, Costa
nas between bays with and without mangroves/seagrass
Rica. Uniciencia 11:89–96
beds. Mar Ecol Prog Ser 214:225–235
Rojas JRM, Pizarro JF, Castro MV (1994b) Diversidad y abun-
Nagelkerken I, Roberts CM, van der Velde G, Dorenbosch M,
dancia íctica en tres áreas de manglar en el Golfo de
van Riel MC, Cocheret de la Morinière E, Nienhuis PH
Nicoya, Costa Rica. Rev Biol Trop 42:663–672
(2002) How important are mangroves and seagrass beds for
Rönnbäck P, Troell M, Kautsky N, Primavera JH (1999) Distrib-
coral-reef fish? The nursery hypothesis tested on an island
ution pattern of shrimps and fish among Avicennia and
scale. Mar Ecol Prog Ser 244:299–305
Rhizophora microhabitats in the Pagbilao Mangroves,
Newman SP, Gruber SH (2002) Comparison of mangrove and
Philippines. Estuar Coast Shelf Sci 48:223–234
seagrass fish and macroinvertebrate communities in
Rooker JR, Dennis GD (1991) Diel, lunar and seasonal changes
Bimini. Bahamas J Sci 9:19–27
in a mangrove fish assemblage off southwestern Puerto
Nwadukwe FO (1995) Species abundance and seasonal varia-
17
Faunce & Serafy: Mangrove-fish literature review
Rico. Bull Mar Sci 49:684–698 inventories of coral reefs, soft bottoms and mangroves in
Rozas LP, Minello TJ (1997) Estimating densities of small fishes Saint-Vincent Bay (New Caledonia). Proc Sixth Int Coral
and decapod crustaceans in shallow estuarine habitats: a Reef Symp 2, Townsville, p 613–618
review of sampling design with focus on gear selection. Thompson AA, Mapstone BD (1997) Observer effects and
Estuaries 20:199–213 training in underwater visual surveys of reef fishes. Mar
Russ GR, Stockwell B, Alcala AC (2005) Inferring versus Ecol Prog Ser 154:53–63
measuring rates of recovery in no-take marine reserves. Tobias WJ (2001) Mangrove habitat as nursery grounds for re-
Mar Ecol Prog Ser 292:1–12 creationally important fish species — Great Pond, St. Croix,
Sasekumar A, Chong VC, Leh MU, D’Cruz R (1992) Man- U.S. Virgin Islands. Proc Gulf Caribb Fish Inst 52:468–487
groves as a habitat for fish prawns. Hydrobiologia 247: Tongnunui P, Ikejima K, Yamane T, Horinouchi M, Medej T,
195–207 Sano M, Kurokura H, Taniuchi T (2002) Fish fauna of the
Savage RE (1931) The relation between the feeding of the her- Sikao Creek mangrove estuary, Trang, Thailand. Fish Sci
ring off the east coast of England and the plankton of the 68:10–17
surrounding waters. Fish Invest Lond Ser II 12:1–88 Tzeng W, Wang Y (1992) Structure, composition and seasonal
Sedberry GR, Carter J (1993) The fish community of a shallow dynamics of the larval and juvenile fish community in the
tropical lagoon in Belize, Central America. Estuaries mangrove estuary of Tanshui river, Taiwan. Mar Biol 113:
16:198–215 481–490
Serafy JE, Faunce CH, Lorenz JJ (2003) Mangrove shoreline Uieda VS, Uieda W (2001) Species composition and spatial dis-
fishes of Biscayne Bay, Florida. Bull Mar Sci 72:161–180 tribution of a stream fish assemblage in the east coast of
Sheaves MJ (1992) Patterns of distribution and abundance of Brazil: comparison oftwo field study methodologies. Brazilian
fishes in different habitats of a mangrove-lined tropical J Biol 61:377–388
estuary, as determined by fish trapping. Aust J Mar Freshw USDOC (1996) Magnusen-Stevens Fishery Conservation and
Res 43:1461–1479 Management Act as amended through October 11, 1996,
Sheaves M (1995) Large lutjanid and serranid fishes in tropical National Oceanic and Atmospheric Administration Techni-
estuaries: Are they adults or juveniles? Mar Ecol Prog Ser cal Memorandum NMFS-F/SPO-23
129:31–40 Valdés-Muñoz E (1981) Estructura y diversidad de la ictiofauna
Sheaves MJ (1998) Spatial patterns in estuarine fish faunas in de los manglares de Punta del Este, Isla de la Juventud.
tropical Queensland: a reflection of interaction between Cienc Biol 6:111–124
long-term physical and biological processes? Mar Freshw Valdés-Muñoz E, Mochek AD (2001) Behavior of marine fishes
Res 49:31–40 of the Cuban shelf. In: Claro R, Lindeman KC, Parenti LR
Sheaves M (2005) Nature and consequences of biological (eds) Ecology of the marine fishes of Cuba. Smithsonian
connectivity in mangrove systems. Mar Ecol Prog Ser 302: Institution Press, Washington, DC, p 73–114
293–305 van der Velde G, Gorissen MW, Denhartog C, Vanthoff T,
Sheridan PF (1992) Comparative habitat utilization by estuar- Meijer GJ (1992) Importance of the Lac-Lagoon (Bonaire,
ine macrofauna within the mangrove ecosystem of Rookery Netherlands-Antilles) for a selected number of reef fish
Bay, Florida. Bull Mar Sci 50:21–39 species. Hydrobiologia 247:139–140
Sheridan P, Hays C (2003) Are mangroves nursery habitat for Vance DJ, Haywood MDE, Heales DS, Kenyon RA, Loneragan
transient fishes and decapods? Wetlands 23:449–458 NR, Pendrey RC (1996) How far do prawns and fish move
Snedaker SC (1989) Overview of ecology of mangroves and into mangroves? Distribution of juvenile banana prawns
Penaeus merguiensis and fish in a tropical mangrove forest
information needs for Florida Bay. Bull Mar Sci 44:341–347
Sogard SM, Powell GVN, Holmquist JG (1989) Spatial- in northern Australia. Mar Ecol Prog Ser 131:115–124
distribution and trends in abundance of fishes residing in Vega-Cendejas ME, de Santillana MH (2004) Fish community
seagrass meadows on Florida Bay mudbanks. Bull Mar Sci structure and dynamics in a coastal hypersaline lagoon: Rio
44:179–199 Lagartos,Yucatan, Mexico. Estuar Coast Shelf Sci 60:285–299
Spalding M, Blasco F, Field C (1997) World Mangrove Atlas. Vega-Cendejas ME, Ordóñez U, Hernández M (1994) Day-
The International Society for Mangrove Ecosystems Night variation of fish population in the mangrove of
(ISME), Okinawa Celestun lagoon, Mexico. Int J Ecol Env Sci 20:99–108
St. John J, Russ GR, Gladstone W (1990) Accuracy and bias of Vidy G (2000) Estuarine and mangrove systems and the nurs-
visual estimates of numbers, size structure and biomass of a ery concept: which is which? The case of the Sine Saloum
coral reef fish. Mar Ecol Prog Ser 64:253–262 system (Senegal). Wetl Ecol Manag:37–51
Stoner AW (1986) Community structure of the demersal fish Vose FE, Bell SS (1994) Resident fishes and macrobenthos in
species of Laguna Joyuda, Puerto Rico. Estuaries 9:142–152 mangrove-rimmed habitats: evaluation of habitat restora-
Thayer GW, Sheridan P (1999) Fish and aquatic invertebrate tion by hydrologic modification. Estuaries 17:585–596
use of the mangrove prop-root habitat in Florida: a review. Warburton K (1978) Community structure, abundance and
In: Yáñez-Arancibia A, Lara-Domínguez AL (eds) Ecosis- diversity of fish in a Mexican coastal lagoon system. Estuar
temas de Manglar en America Tropical. Instituto de Ecolo- Coast Mar Sci 7:497–519
gia, A.C. Mexico, UICN/ORMA Costa Rica, NOAA/NMFS Weinstein MP, Courtney CM, Kinch JC (1977) The Marco
Silver Spring, MD, p 380 Island estuary: a summary of physicochemical and biologi-
Thayer GW, Colby DR, Hettler WF Jr (1987) Utilization of the cal parameters. Fla Sci 40:97–124
red mangrove prop root habitat by fishes in South Florida. Weng HT (1990) Fish in shallow areas in Moreton Bay,
Mar Ecol Prog Ser 35:25–38 Queensland and factors affecting their distribution. Estuar
Thia-Eng C (11973) An ecological study of the Ponggol estuary Coast Shelf Sci 30:569–578
in Singapore. Hydrobiol 43:3–4 Williamson I, King C, Mather PB (1994) A comparison of fish
Thollot P (1992) Importance des mangroves pour la faune communities in unmodified and modified inshore habitats
ichtyologique des récifs coralliens de Nouvelle-Calédonie. of Raby Bay, Queensland. Estuar Coast Shelf Sci 39:
Cybium 16:331–344 401–411
Thollot P, Kulbicki M (1988) Overlap between the fish fauna Wright JM (1986) The ecology of fish occurring in shallow
18 Mar Ecol Prog Ser 318: 1–18, 2006
water creeks of a Nigerian mangrove swamp. J Fish Biol mass and diversity of estuarine fishes coupled with tropical
29:431–441 habitat heterogeneity (southern Gulf of Mexico). J Fish Biol
Yáñez-Arancibia A (1985) Fish community ecology in estuaries 33:191–200
and coastal lagoons: towards an ecosystem integration. Yáñez-Arancibia A, Lara-Domínguez AL, Day JW Jr (1993) In-
UNAM Press, Mexico teractions between mangrove and seagrass habitats medi-
Yáñez-Arancibia A, Lara-Domínguez AL, Rojas-Galaviz JL, ated by estuarine nekton assemblages: coupling of primary
Sanchez-Gil P, Day JW Jr, Madden CJ (1988) Seasonal bio- and secondary production. Hydrobiologia 264:1–12
Editorial responsibility: Howard I. Browman (Associate Submitted: February 17, 2005; Accepted: May 24, 2006
Editor-in-Chief), Storebø, Norway Proofs received from author(s): July 12, 2006
Vol. 318: 1–18, 2006 Published August 3
Mar Ecol Prog Ser
OPEN
PEN
ACCESS
CCESS
FEATURE ARTICLE: REVIEW
Mangroves as fish habitat: 50 years of field studies
Craig H. Faunce1, 3,*, Joseph E. Serafy1, 2
1
Division of Marine Biology and Fisheries, Rosenstiel School of Marine and Atmospheric Science, University of Miami,
4600 Rickenbacker Causeway, Miami, Florida 33149, USA
2
Southeast Fisheries Science Center, National Marine Fisheries Service, 75 Virginia Beach Drive, Miami, Florida 33149, USA
3
Present address: Florida Fish and Wildlife Conservation Commission, Fish and Wildlife Research Institute,
Tequesta Field Laboratory, 19100 SE Federal Highway, Tequesta, Florida 33469, USA
ABSTRACT: Mangroves dominate undisturbed nat-
ural shorelines of many sub-tropical and tropical
regions, yet their utilization by fishes is poorly
understood. To provide the first comprehensive list
of empirical field studies for comparative and ref-
erence purposes, we assembled and quantified
aspects of 111 mangrove-fish surveys published
between 1955 and 2005. Differences in the location,
purpose, methodology, data gathered, and analyses
performed among studies have resulted in a frag-
mented literature making cross-study comparisons
difficult, at best. Although the number of published
studies has increased over time, a geographical
bias in the literature has persisted towards studies
performed in the USA and Australia, and against
studies performed in Southeast Asia and West
Africa. The typical survey design has examined
<10 fixed locations on a monthly or bimonthly basis
for a period of less than 2 yr. Water temperature
and salinity measurements have been the most
reported habitat variables; others, such as structural
and landscape measures, continue to be rare.
Moreover, the focus to date has been on identifying Faunce & Serafy review mangrove fish studies, focusing on
sampling methodology, and on the types of fish and habitat
assemblage-level patterns of fish use, with very few
data reported. While most studies have addressed spatio-
studies providing species-specific estimates of
temporal patterns in fish assemblage structure, species-
abundance, growth, mortality, and secondary pro-
specific estimates of fish mortality, growth and secondary
duction. Unless future studies strive towards ob-
production are still required for appraisal of the importance
taining such estimates, gauging the importance of of mangroves as fish habitat.
mangroves as fish habitat and their broader contri- Photo: Jiangang Luo
bution to ecosystem diversity and production will
remain elusive.
INTRODUCTION
KEY WORDS: Fishes · Essential fish habitat · Man-
Mangrove wetlands are a dominant feature of undis-
groves · Nursery · Review
turbed tropical and subtropical shorelines around the
Resale or republication not permitted without
globe. Throughout their range, however, these habi-
written consent of the publisher
*Email: craig.faunce@myfwc.com © Inter-Research 2006 · www.int-res.com
2 Mar Ecol Prog Ser 318: 1–18, 2006
tats are in a state of decline. Approximately one-third words ‘mangrove(s)’ and/or ‘fish(es)’. Records were
of the world’s mangrove forests has been lost to coastal selected from the earliest available time period (1971)
development over the past 50 yr (Alongi 2002). While to January, 2005. The resulting list of over 500 publica-
there is general agreement that mangroves provide a tions was reduced to relevant works according to 2
buffer against storm surges, reduce shoreline erosion main criteria: (1) the study must have been published
and turbidity, absorb and transform nutrients, and are in readily available outlet (i.e. in the ‘primary litera-
inhabited by a variety of organisms, opinions vary as to ture’); and (2) each publication must have contained a
the importance of mangrove habitats to fishes and, by field-based survey of the ichthyofauna that was con-
extension, to offshore fisheries (Thollot & Kulbicki ducted within a natural mangrove system. Second, the
1988, Blaber et al. 1989, Thollot 1992, Nagelkerken et Science Citation Index (Web of Science; http://isi5.
al. 2001). For example, the sub-tidal prop-root habitats newisiknowledge.com) was used to identify articles
of mangroves are often cited as nurseries for fishes of that cited works from our reduced ASFA list. Again,
economic importance. Today, the protection of man- any additional publications were vetted according to
groves worldwide is based almost entirely on their the selection criteria above. Third, articles from
purported importance to fisheries and/or a number of the authors’ personal libraries and those introduced
rare and endangered species (Snedaker 1989, Baran & through peer review (of this paper) were added. The
Hambrey 1998). However, because the same man- references cited in each relevant article were exam-
grove species can often occur under marine, estuarine, ined for new items and this process was continued
and freshwater conditions, a wide variety of fish until no additional publications emerged.
assemblages can be found among their inundated Study locations were grouped into 5 geographic
prop-roots (hereafter termed ‘mangrove habitats’). As regions following the World Mangrove Atlas (Spalding
such, mangrove habitats likely play a variety of roles in et al. 1997): (1) South and Southeast Asia (Pakistan to
the lives of associated fishes; feeding areas for some the west, China and Japan to the northeast, including
species or life stages, daytime refugia for others, nurs- Indonesia), (2) Australasia (Australia, Papua New
ery and/or nesting areas for yet more. This situation Guinea, New Zealand, and the South Pacific islands),
suggests that questions regarding the contribution of a (3) the Americas (north, central, and south), (4) West
given mangrove habitat to the diversity, productivity Africa, and (5) East Africa and the Middle East (Iran to
and stability of broader fish communities (and their South Africa eastwards, including the islands in the
exploited components) must be carefully qualified, or, Indian Ocean). Using the selection (foraging) ratio of
in some cases, may be premature. Savage (1931), geographical bias in the literature was
The purpose of this paper is to address some of the expressed as the proportion of total studies realized
most basic questions regarding the body of literature per region relative to the area of mangrove coverage
on mangrove fishes that has been published over the within each region (Manly et al. 1993).
past 5 decades. These questions include: How many The study purpose, methodology, data gathered, and
field studies have been conducted, why and where analyses performed were extracted and tabulated
were they performed, and what techniques were used? using vote-counting procedures, where ‘present’ was
What types of measurements have been made of the given a value of 1 and ‘absent’ was given a value of
fish assemblages, their component species and their 0. Data were expressed as proportions of the total
habitats? Is there sufficient basis for comparing assem- number of votes per attribute. Study purposes included
blages of mangrove fishes with those associated with identifying spatial or temporal patterns, generating
other, structurally-complex habitats, such as seagrass species lists, identifying explanatory variables, biogeo-
beds and coral reefs? The answers to these questions graphic comparisons, restoration, water management,
are pertinent to researchers about to embark on new and gear evaluations. Methodologies included the
studies, as well as to those making efforts to balance sampling design (fixed, random, haphazard, or vari-
natural resource protection with pressing socio-eco- ous), sampling frequency (daily, weekly, fortnightly,
nomic considerations. monthly, bimonthly, quarterly, seasonally, semi-annu-
ally) sampling duration, and gear type. Gear types were
classified according to Rozas & Minello (1997) and
included entanglement gear (gill or trammel nets),
METHODS
towed nets (trawls, seines), passive samplers (fyke
Publications for this review were selected from 3 nets, flume nets, rotenone-used with or without nets,
databases. First, a search of the Aquatic Sciences fish traps, e.g. breder, plankton), and ‘enclosure sam-
and Fisheries Abstracts (ASFA) electronic database plers’ (block or drop nets, drop traps, and cast nets).
was conducted (Cambridge Scientific Abstracts; www. Visual surveys and angling were added as additional
csa.com) using keyword and title searches for the gears. The type of mangrove forest sampled was noted
3
Faunce & Serafy: Mangrove-fish literature review
using the classification scheme of Lugo & Snedaker mangrove coverage, it is encouraging that more litera-
(1974), which included fringing, riverine and/or basin ture is emerging from this region where coastal fish as-
forest. The data gathered in each study included biotic semblages are also relatively diverse (Blaber 2002).
and abiotic habitat metrics. Fish metrics included However, a literature void remains for the West Africa
groupings by family, maturation stage, residency sta- region — an area that is likely to continue to be under-
tus, trophic level, or diel habits as well as the type of represented unless specifically targeted for study.
fish data gathered and analyzed. We classified the Interestingly, the first 2 studies of mangrove fishes
types of fish metric data reported in each study accord- were conducted in the 2 regions that are least repre-
ing to the criteria recommended for determining sented today.
‘essential fish habitat’ (EFH) (USDOC 1996) which Although disproportionate, the spatial distribution of
included: presence/absence, frequency of occurrence, studies we examined covers much of the known global
percent composition, size, biomass (g), density (num- distribution of mangroves (Fig. 2). Specific areas that
ber/area), standing crop (g/area), growth and mortality have received the most thorough study include Florida
rates, and rates of secondary production. Finally, the (USA) and Moreton Bay (Australia). Similar dominance
focus of analyses (e.g. defining spatial and/or temporal of studies from the USA and Australia in the literature
patterns, examining fish–habitat correlations) was tab- has beset previous reviews of fishes occupying sea-
ulated and the type of statistical test(s) or data treat- grass beds (Heck et al. 2003), mangroves (Sheridan &
ment(s) performed (similarity measures, analysis of Hays 2003), and studies of ontogenetic fish movements
variance, ordination, or regression) were recorded. (Gillanders et al. 2003). Regions outside US and Aus-
tralian waters where data are particularly lacking
include: (1) Pacific Panama; (2) Colombia; (3) Central
Brazil; (4) the Red Sea; (5) Mozambique; (6) the Bay of
RESULTS AND DISCUSSION
Bengal and the Andaman Sea; and (7) Borneo (Fig. 2).
However, there are several important cases where
Chronology and geography
fishes inhabiting tropical estuaries from these areas
A total of 111 publications were examined from 104 have been reported. In such cases, data summaries
independent field surveys of mangrove fishes pub- were made from multiple surveys of fishes that rarely
lished between 1955 and 2005 (Table 1). The earliest mentioned study objectives, methods, gear, or habi-
records of mangrove-associated fishes were species tat(s) sampled, making their inclusion here difficult, at
lists compiled by Inger (1955) and Boeseman (1963) as best. In Florida (USA), species lists of mangrove fishes
part of broad ecological inventories of forests in Borneo were compiled from numerous studies by Odum et
(South and Southeast Asia region) and the Niger Delta al. (1982) and presented for tidal streams, estuarine
(West Africa region), respectively. Austin (1971) pro- bays, and oceanic bays. Off Columbia, Alvarez León
vided the first inventory of mangrove fishes from & Blanco Racedo (1985) reviewed aspects of 31 studies
Puerto Rico (American region), and Day (1974) from conducted in the Cartagena Bay system. They pro-
Mozambique (i.e. the East Africa and Middle East vided an overall species list for the system, and sum-
region). Blaber (1980), and Blaber & Blaber
(1980) published the first of many works on W Africa
S & SE Asia
100
assemblages of mangrove fishes from Aus- E Africa
Cumulative number of publications
tralia (Australasian region). Australasia
Americas
While the cumulative number of publica- 80
tions has grown steadily since the mid 1980s,
the sharpest increases occurred for the
regions of South and Southeast Asia and the 60
Americas (Fig. 1). Selection indices indicate
that the geographic distribution of studies
40
among regions has not been commensurate
with the proportion of the world’s mangrove
acreage that they contain (Table 2). Nearly 20
70% of studies have been conducted in either
the Americas or Australasia, and the South
0
and Southeast Asia and West Africa regions 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
Year
are clearly under-represented in the litera-
ture. Because South and Southeast Asia con- Fig. 1. Cumulative number of publications on mangrove fishes by each
tains the largest proportion of the world’s geographic region from 1955 to 2005 (n = 111)
4
Table 1. Chronological list of surveys of mangrove fishes conducted during 1955 to 2005. Design codes: FX: fixed; HZ: haphazard; RM: random; VAR: various. SU: sample
unit; Prop. mangrove: proportion of sites that were within mangrove habitats. Frequency codes: BM: bimonthly; D: daily; F: fortnightly; M: monthly; S: seasonal; Q: quar-
terly; Var: various. Gear codes: A: anglers; B: block net; C: cast net; F: fyke net; G: gill or trammel net; P: plankton net; R: rotenone; S: seine net; T: trap; TP: trap or flume net;
TW: trawl; V: visual surveys. Published studies with same letter [in brackets] are from the same fish survey. nr = not reported
Location Region Design SU No. No. Prop. Fre- Duration Gear Basin Fringe Riverine Source
SU sites mangrove quency (yr)
Dewhurst Bay, Borneo SA FX Site 11 11 0.36 1 nr A,C,R,S x Inger (1955)
Niger Delta WA nr Region 4 4 1.00 Var NR x Boseman (1963)
Western Puerto Rico AM FX Estuary 8 8 1.00 Var G,S x x Austin (1971)
Ponggol Estuary, Singapore SA FX Site 3 3 1.00 M 1.0 S x Thia-Eng (1973)
Morrumbene Estuary, Mozambique EA FX Site 6 6 0.50 3 A,S x x Day (1974)
Old Rhodes Key lagoon, USA AM FX Site 6 6 M 0.4 A,T,V Holm (1977)
Guadeloupe AM FX Site 12 12 0.33 M 0.1 F,TP x Lasserre & Toffart [a] (1977)
Marco Island Estuary, USA AM FX Site 11 11 0.00 M 4.1 TW Weinstein et al. (1977)
Huizache and Caimanero lagoons, AM FX Site 5 5 0.80 M 1.3 C,S x Warburton (1978)
Mexico
Mngazana Estuary, S Africa EA FX Site 7 7 1.00 3 G,P,S x x Branch & Grindley (1979)
Trinity Inlet System, Australia AU FX Site 18 18 0.39 nr B,G,P,S x Blaber (1980)
Moreton Bay, Australia AU FX Site 4 4 0.50 M 1.0 B,P,S x Blaber & Blaber (1980)
Serpentine Creek, Australia AU FX Site 10 10 0.00 F 1.2 TW Quinn (1980)
Tamil Nadu coast, India SA nr Region 1 nr P x Krishnamurthy &
Prince Jeyaseelan (1981)
Jiquilisco Bay, El Salvador AM FX Site 6 6 0.00 M 1.3 TW Phillips [b] (1981b)
Jiquilisco Bay, El Salvador AM FX Site 6 6 0.00 M 1.3 C,G,TW Phillips [b] (1981b)
Isle of Youth, Cuba AM FX Site 4 4 1.00 S 1.0 V x x Valdéz-Muñoz (1981)
Bahia de la paz, Mexico AM FX Site 3 7 0.00 M 0.5 TW Martínez et al. (1982)
Papua New Guinea AU FX Site 14 14 1.00 nr R x x x Collette (1983)
Botany Bay, SE Australia AU FX Site 1 1 1.00 BM 1.5 B,R x Bell et al. (1984)
Wairiki Creek, Fiji AU FX Site 6 6 1.00 M 1.0 G x Lal (1984)
Mar Ecol Prog Ser 318: 1–18, 2006
Dampier region, NW Australia AU FX Biotope 2 9 0.67 Q 1.0 G,R,S,V x x x Blaber et al. (1985)
Guadeloupe AM FX Bay 2 12 0.33 Var Var TP x Louis et al. [a] (1985)
Papua New Guinea AU FX Estuary 2 5 0.00 F 2.0 TW Quinn & Kojis (1985)
Phuket Island, Thailand SA FX Site 4 4 0.50 M 1.6 P,TW x x Janekarn & Boonruang (1986)
Laguna Joyuda, Puerto Rico AM FX Site 5 5 0.00 M 1.0 TW Stoner (1986)
Elichi Creek, Nigeria, W Africa WA FX Site 2 2 1.00 M 1.1 B x Wright (1986)
Pagbilao Bay, Phillipines SA FX Site 6 6 1.00 M 1.5 G,TW x Pinto (1987)
Alligator Creek, Australia AU nr Biotope 3.5 8 0.50 BM 1.2 S x Robertson & Duke [c] (1987)
Western Florida Bay, USA AM FX Site 8 8 1.00 Var Var B,R,TW x x Thayer et al. (1987)
Sinnamary river, French Guiana AM nr Region 3 M Var C,G,TP,R,TW x x Boujard & Beltran (1988)
Leanyer Swamp, N Australia AU FX Site 2 2 1.00 F 0.5 R,T x x Davis (1988)
Tudor Creek, Kenya, E Africa EA FX Site 4 4 0.75 F 0.6 P,S x Little et al. [d] (1988b)
Tudor Creek, Kenya, E Africa EA FX Site 5 5 0.60 nr nr P x Little et al. [d] (1988a)
Saint-Vincent Bay, New Caledonia AU nr Biotope 3 121 0.01 Var Var F,G,TW,R,V x x Thollot & Kulbicki (1988)
Teminos Lagoon, Mexico AM FX Biotope 2 2 0.00 M 1.0 TW Yáñez-Arancibia et al. [e]
(1988)
Embley Estuary, NE Australia AU FX Biotope 5 6 0.83 Q 2.0 B,G,R,S,TW x Blaber et al. (1989)
Table 1 (continued)
Location Region Design SU No. No. Prop. Fre- Duration Gear Basin Fringe Riverine Source
SU sites mangrove quency (yr)
Nigerian Coast, W Africa WA FX Estuary 4 nr P,S nr nr nr Amadi et al. (1990)
Solomon Islands, W Pacific AU FX Estuary 13 13 1.00 nr B,G,R,S x x Blaber & Milton (1990)
Selangor, Malaysia SA FX Biotope 4 41 0.22 Var Var F,G,TW x x Chong et al. [f] (1990)
Tecapan-Aqua Brava, Mexico AM FX Site 15 15 1.00 M 1.0 S,TW Flores-Verdugo et al. (1990)
Moreton Bay, E Australia AU FX Site 1 3 1.00 M 1.2 B,G,S x x Morton (1990)
Alligator Creek, Australia AU nr Estuary 1 BM 1.2 G,S,TP x x Robertson & Duke [c] (1990b)
Alligator Creek, Australia AU nr Estuary 4 BM 1.2 G,S,TP x x Robertson & Duke [c] (1990a)
Moreton Bay, Australia AU FX Biotope 5 5 0.60 M 0.8 TW x Weng (1990)
Dubreka and Tabunsu estuary,
Guinea WA FX Estuary 2 nr A,C,G,TW nr nr nr Boltachev (1991)
Cayo Collado, Puerto Rico AM FX Site 8 8 1.00 S 0.2 V x Rooker & Dennis (1991)
Northern US Virgin Islands AM FX Site 5 5 1.00 M 1.8 T,V x Boulon (1992)
La Parguera, Puerto Rico AM FX Biotope 4 4 0.50 M 1.1 P x Dennis (1992)
Fort-de-France, Martinique AM FX Site 8 8 1.00 Q 2.3 TP x Louis et al. [g] (1992)
Selangor, Malaysia SA FX Biotope 3 41 0.22 Var Var F,G,TW x x Sasekumar et al. [f] (1992)
Alligator Creek, Australia AU RM Region 3 6 0.50 M 0.4 A,S,T x Sheaves (1992)
Rookery Bay, USA AM VAR Biotope 3 30 0.33 Q 1.3 TP x Sheridan (1992)
Saint-Vincent Bay, New Caledonia AU VAR Biotope 2 577 0.03 M 1.0 G,R,TM,TP,V x Thollot (1992)
Tanshui River, Taiwan SA FX Site 3 3 0.00 M 2.8 B x Tzeng & Wang (1992)
Bonaire, Netherlands Antilles AM nr Biotope 6 nr V x van der Velde et al. [h] (1992)
Sabana-Camaguey, N Cuba AM RM Region 3 63 1.00 1 nr V x Claro & García-Arteaga (1993)
Phang-Nga Bay, Thailand SA FX Site 1 1 0.00 M 1.2 TW Janekarn (1993)
Ambergis Cay, Belize AM FX Biotope 3 3 0.33 BM 1.0 TW x Sedberry & Carter (1993)
Terminos Lagoon, Mexico AM FX Biotope 2 2 0.00 BM 1.2 TW Yáñez-Arancibia et al. [e]
(1993)
Gulf of Carpentaria, N Australia AU FX Biotope 4 S Var B,F,G,R,S,TW x x x Blabler et al. (1994)
Clarence River, Australia AU FX Site 18 18 0.28 Q 2.5 B,R x Pollard & Hannan (1994)
Gulf of Nicoya, Costa Rica AM FX Site 3 3 1.00 M 0.9 G,S x Rojas et al. [i] (1994a)
Gulf of Nicoya, Costa Rica AM FX Site 3 3 1.00 M 0.9 G,S x Rojas et al. [i] (1994b)
Faunce & Serafy: Mangrove-fish literature review
Celestum Lagoon, Mexico AM FX Site 1 1 1.00 S 1.5 B,R x x Vega-Cendejas et al. (1994)
Upper Tampa Bay, USA AM RM Bay 2 24 M 2.9 T x x x Vose & Bell (1994)
Raby Bay, E Austalia AU FX Biotope 2 7 0.43 M 1.1 S x Williamson et al. (1994)
Moreton Bay, E Australia AU FX Biotope 2 6 0.33 M 2.0 S,TP x x Laegdsgaard & Johnson (1995)
Fort-de-France, Martinique AM FX Site 8 8 1.00 Q 2.3 TP x Louis et al. [g] (1995)
Placido Bayou, USA AM FX Site 6 6 1.00 Var B,S x x Mullin (1995)
Lagos Lagoon, Nigeria WA FX Site 2 2 0.50 M S x Nwadukwe (1995)
Sao Luis, Brazil AM FX Site 4 4 1.00 M 0.8 TP x Batista & Rego (1996)
Tin Can Bay, NE Austalia AU FX Site 4 4 1.00 BM 2.2 B x Halliday & Young (1996)
Gazi Creek, Kenya, E Africa EA FX Site 3 3 0.67 M 0.8 S x Kimani et al. (1996)
Negombo Estuary, W Sri Lanka SA FX Site 6 4 1.00 M 2.1 S x Pinto & Punchihewa (1996)
Embley River, N Australia AU FX Site 4 4 1.00 D,S Var B x Vance et al. (1996)
La Parguera, Puerto Rico AM RM Biotope 3 43 0.21 nr G,V x Acosta (1997)
Karachi, Pakistan SA FX Site 2 2 0.50 F 0.9 C nr nr nr Atiqullah et al. (1997)
5
6
Table 1 (continued)
Location Region Design SU No. No. Prop Fre- Duration Gear Basin Fringe Riverine Source
SU sites mangrove quency (yr)
Tulear Lagoon, SW Madagascar EA FX Site 1 1 1.00 M 1.0 G Laroche et al. (1997)
Guaratuba, Brazil AM RM Region 1 1 1.00 nr C,S,TW Chaves & Correa (1998)
Sao Luis, Brazil AM FX Site 4 4 1.00 M 0.8 TP x Batista & Rego (1996)
Tin Can Bay, NE Australia AU FX Site 4 4 1.00 BM 2.2 B x Halliday & Young (1996)
Gazi Creek, Kenya, E Africa EA FX Site 3 3 0.67 M 0.8 S x Kimani et al. (1996)
Negombo Estuary, W Sri Lanka SA FX Site 6 4 1.00 M 2.1 S x Pinto & Punchihewa (1996)
Embley River, N Australia AU FX Site 4 4 1.00 D, S Var B x Vance et al. (1996)
La Parguera, Puerto Rico AM RM Biotope 3 43 0.21 nr G,V x Acosta (1997)
Karachi, Pakistan SA FX Site 2 2 0.50 F 0.9 C nr nr nr Atiqullah et al. (1997)
Tulear Lagoon, SW Madagascar EA FX Site 1 1 1.00 M 1.0 G Laroche et al. (1997)
Guaratuba, Brazil AM RM Region 1 1 1.00 nr C,S,TW Chaves & Correa (1998)
NE tropical Australia AU HZ Region 4 12 1.00 Q 2.3 T x Sheaves (1998)
Yingluo Bay, China SA FX Site 2 2 1.00 Q 1.0 B x x He et al. (2001)
Guaratuba, Brazil AM RM Region 1 1 1.00 M 2.9 TW Chaves & Bouchereau (1999)
Bay of La Paz, Mexico AM FX Site 1 1 1.00 nr F x Gonzalez-Acosta et al. (1999)
N & S Taiwan SA FX Region 6 6 1.00 BM 1.0 F x Kuo et al. (1999)
NE Florida Bay, USA AM FX Bay 6 42 1.00 M 1.1 B,R,V x x Ley et al. [j] (1999)
N Taiwan SA FX Site 1 1 1.00 M 1.0 F x Lin & Shao (1999)
NE Florida Bay, USA AM FX Site 4 24 1.00 BM 5.2 R,T x Lorenz (1999)
Pagbilao, Philippines SA FX Site 4 4 1.00 D 0.0 B x Rönnbäck et al. (1999)
Curacao, Netherlands Antilles AM FX Biotope 6 12 1.00 Var V x Nagelkerken et al. [k] (2000a)
Curacao, Netherlands Antilles AM FX Biotope 6 12 1.00 Var V x Nagelkerken et al. (2000b)
Bonaire, Netherlands Antilles AM FX Biotope 6 27 0.22 M 0.6 V x Nagelkerken et al. [h] (2000c)
Sine Saloum, Senegal WA FX Site 3 6 1.00 M 0.2 F Vidy (2000)
Guaratuba, Brazil AM RM Region 1 1 1.00 nr F,S,TW Chaves & Vendel (2001)
NE Florida Bay, USA AM FX Bay 6 17 1.00 M 1.1 V x Ley & McIvor [j] (2001)
Curacao, Netherlands Antilles AM FX Bay 3 11 0.00 D, M Var S,V Nagelkerken et al. (2001)
Mar Ecol Prog Ser 318: 1–18, 2006
St. Croix, US Virgin Islands AM HZ Site 10 10 1.00 M 2.0 S,T x Tobias (2001)
Rio da Fazenda, Brazil AM nr Region 3 3 0.33 BM 1.0 A,T,V x Uieda & Uieda (2001)
Sepetiba Bay, Brazil AM RM Region 3 158 0.00 M 3.0 TW Araujo et al. (2002)
Caete River Estuary, Brazil AM FX Site 3 3 1.00 M 1.0 TP x Barletta-Bergan et al. (2002)
Curacao, Netherlands Antilles AM FX Biotope 3 11 3.00 S nr V x Cocheret de la Morinière et al.
[k] (2002)
Lower Volta, Ghana WA FX Site 7 7 0.71 nr nr C,G,S x x Dankwa & Gordon (2002)
Curacao, Netherlands Antilles AM FX Biotope 8 17 0.71 S 1.0 V x Nagelkerken & van der Velde
[k] (2002)
Eastern Caribbean AM FX Biotope 3 23 0.00 Var Var V x Nagelkerken et al. (2002)
Bimini, Bahamas AM FX Biotope 4 4 0.25 nr B,S x Neuman & Gruber (2002)
Sikao Creek Estuary, Thailand SA FX Site 6 6 0.67 Q 26.7 F,R,S x x x Tongnunui et al. [l] (2002)
Johor Strait, Malaysia SA FX Site 4 4 0.25 M 1.5 S x Hajisamae & Chou (2003)
Sikao Creek, Thailand SA FX Site 3 3 0.67 Q 2.4 S x x Ikejima et al. [l] (2003)
Biscayne Bay, USA AM RM Region 2 129 1.00 S 2.0 V x Serafy et al. (2003)
SE Everglades, USA AM FX Site 3 12 1.00 BM 3.2 V x Faunce et al. (2004)
Rio Lagartos, Mexico AM FX Site 28 28 0.00 Var Var S,TW Vega-Cendejas & Santillana
(2004)
Chwaka Bay, Tanzania EA FX Biotope 5 5 0.40 Var Var S,V x Lugendo et al. (2005)
7
Faunce & Serafy: Mangrove-fish literature review
Fig. 2. Location of studies of mangrove fishes used in the present review (coded by geographic region)
marized their data based on the number of species fishes (21.7%) or identify explanatory variables for
belonging to various salinity regimes and trophic observed utilization patterns (15.6%). Less than 10%
levels for Bahía de Cartagena, Ciénaga de Tesca, and of studies were concerned with the remaining topics
Ciénaga Grande de Santa Marta. Similarly, Cervigón (Fig. 3a). Most studies aimed to identify temporal pat-
(1985) used data from an earlier study and historical terns, and typically achieved this goal through monthly
records to generate a list of fish species for the Ori- sampling for a period of 0.5 to 1.5 yr (Fig. 3b,c). Sam-
noco estuary (Venezuela) according to salinity regime. pling durations of more than 2 yr were uncommon
Finally, the ecology of the Itamaracá ecosystem (Brazil) (< 5%) with the longest published survey spanning 5 yr
was summarized by Paranagua & Eskinazi-Leça (1985), (Lorenz 1999). In addition, most studies sampled, or
who provided a family and species list of fishes. For otherwise quantified, fishes at a small number of loca-
more complete summaries of fish studies conducted tions. Only 4 studies sampled mangroves at more than
from large tropical estuarine systems, the reader 20 locations: Serafy et al. (2003) sampled 129 locations,
should consult the comprehensive works of Blaber Claro & García-Arteaga (1993) sampled 63 locations,
(2000) and Yáñez-Arancibia (1985). Ley et al. (1999) sampled 42 locations, and Lorenz
(1999) sampled 24 locations (Table 1). Fixed sampling
designs were employed much more often (81%) than
random-stratified (8.5%) or haphazard designs (1.8%)
Study design
or various other sampling designs (1.8%). In fixed- and
Study design incorporates a study’s purpose with its mixed-design surveys, the rationale for selecting site
methodology. Over half of the examined surveys of locations was rarely provided.
mangrove fishes aimed to identify spatial and/or temp- These results highlight some limitations with our
oral patterns of mangrove utilization, while a lesser knowledge of mangrove habitat utilization by fishes.
For example, if not selected remotely, or a priori, the
proportion were conducted to provide an inventory of
Table 2. Comparison of mangrove area and the number of published studies from within each geographic region. The selection
ratio (wi) of Savage (1931) was used to compare the proportion of studies to the proportion of mangrove area within each region
Region Mangrove Proportion of Number of Proportion of wi
area (km2) total area total studies total studies
Americas 49 096 0.271 53 0.477 1.762
Australasia 18 789 0.104 26 0.234 2.252
East Africa and Middle East 10 024 0.055 7 0.063 1.147
South and Southeast Asia 75 173 0.415 18 0.162 0.391
West Africa 27 995 0.155 7 0.063 0.407
Total 181 0770 1110
8 Mar Ecol Prog Ser 318: 1–18, 2006
a) Purpose ment of what constitutes a ‘typical’ year, or of the
0.6
extent of inter-annual variability in both the fish
0.5
assemblages and their environment. While the sea-
sonal dynamics of mangrove use by fishes has received
0.4
attention, only a few studies have focused on shorter
0.3
temporal scales. In the future, researchers should bet-
ter describe how and why sampling locations were
0.2
selected, and when possible include the rationale
0.1
behind their sampling intensity/allocation decisions.
How fish samples have been acquired from man-
0.0
In rns
y
eo ar.
hy
n
.
e
groves is of particular importance due, in part, to issues
gt
or
yp
io
v
ap
m
at
tte
nt
rt
y
of species- and size-selectivity. Indeed, one of the
gr
or
ve
er
or
ea
Pa
st
at
at
G
Re
W
an
major reasons mangroves have received relatively lit-
og
pl
Bi
Ex
tle attention as fish habitats is that it is inherently diffi-
cult to quantitatively sample fishes within them. Con-
b) Frequency
0.5
sequently, our understanding of the role(s) that these
Proportion of studies
habitats play in the lives of fishes has been hindered by
0.4
the fact that the same sampling methods have rarely
been used from one study to the next. Over one-third
0.3
of all studies we reviewed used towed gears, which
0.2 is a consistent finding among geographic regions
(Fig. 4a,b). While towed nets can be effective in sea-
0.1 grass beds, they are of little or no use within the dense,
rigid, entangled roots of mangrove trees. Use of pas-
0.0
lly
y
y
ly
ly
Va ly
us
d
al
l
rte
th
th
ht
ht
a) Gears: type
ua
rio
on
on
on
ig
ig
po
nn
0.35
as
rtn
rtn
-m
M
re
i-a
Se
Fo
Fo
Bi
ot
m
0.30
N
<
Se
0.25
c) Duration
0.25 0.20
0.15
0.20
0.10
0.15
0.05
Proportion of studies
0.00
0.10
ed
ve
e
al
t
s
en
er
ur
su
i
w
ss
gl
m
os
Vi
To
0.05
An
Pa
le
cl
ng
En
ta
En
0.00
b) Gears: region
5
0
5
0
5
0
3+
us
d
0.
1.
1.
2.
2.
3.
rte
1.0
rio
po
Anglers
Va
re
ot
0.8
N
Entanglement
Fig. 3. (a) Purpose, (b) sampling frequency, and (c) sampling
0.6
duration (in yr) of studies of mangrove fishes. Var: variable; Visual
mgt: management. Note difference in y-axis scales
0.4 Enclosure
choice of sampling locations may be biased, which in Passive
0.2
turn can be reflected in the data gathered (Rozas &
Towed
Minello 1997). In addition, the limited spatial extent 0.0
AM AU EA SA WA
and/or small number of sites sampled in most studies
Region
may not be representative of a given area or region,
Fig. 4. The gear type used to collect fishes from mangrove
and thus may be of limited use to coastal resource
habitats (a) among studies, and (b) by geographic region. AM:
managers if they must determine the costs and benefits Americas; AU: Australasia; EA: East Africa and the Middle
of developing one mangrove area over another. Simi- East; SA: South and Southeast Asia; WA: West Africa. Note
larly, the lack of multi-year studies precludes assess- difference in y-axis scales
9
Faunce & Serafy: Mangrove-fish literature review
sive gears has also been common (27% of studies). coral reef environments (e.g. Lindeman & Snyder 1999,
Most passive gears are difficult to place in dense man- Nagelkerken et al. 2000b, Russ et al. 2005). An impor-
grove prop-roots without constituting additional struc- tant distinction between visual surveys and other
ture, and like towed gears are often actually employed methods is that the former does not capture fishes, and
at the periphery of the mangrove shoreline. This con- thus is advantageous for studying threatened or
sidered, one out of 5 purported studies of mangrove endangered species. Limitations of visual surveys stem
fishes that we reviewed failed to sample within man- from variations in visibility, size- and species-specific
grove habitat. This has undoubtedly produced unrep- responses of fish to those performing the survey,
resentative data in specific cases, and has probably led observer experience, recording errors, as well as safety
to unfounded conclusions about the nature and extent concerns (Cheal & Thompson 1997, Thompson & Map-
of fish utilization of mangroves in general. stone 1997, Ley et al. 1999). While precision varies by
There are notable exceptions, however, where more methodology (e.g. roving, timed, or belt transect, writ-
appropriate sampling techniques have been applied to ten or audio recording media), accuracy problems can
quantitatively sample fish within the mangroves be reduced by performing observer training and using
proper. These include enclosure, entanglement, and a limited number of personnel (Bell et al. 1985, Greene
visual techniques, which have each been employed & Alevizon 1989, St. John et al. 1990). Visual survey
with similar frequency (i.e. 11 to 13% of studies). A techniques were used far more often in studies con-
common type of enclosure gear uses a fine mesh net to ducted in the Americas (Caribbean), than in either
encompass a mangrove area and fishes are subse- Australasia or East Africa, and were absent from South
quently removed with poison and/or a smaller net (e.g. and Southeast Asia and West Africa (Fig. 4b).
Bell et al. 1984, Thayer et al. 1987, Blaber & Milton Given the above, there is clearly no ‘best’ method for
1990). Such block netting can result in estimates of sampling fishes within mangrove habitats. The optimal
both abundance and biomass per unit area (standing method will vary according to study constraints and
crop), and is an especially effective method for collect- have bias and precision that should be weighed in
ing small, cryptic fishes. However, sampling efficiency accordance with the goals of the project. In studies that
is dependent on the clearing method used. Two draw- focus on analyzing the entire assemblage, the applica-
backs of enclosure samplers are that they often involve tion of multiple gear types has been used with success
both short- and long-term disturbance of the habitat and is preferred (Blaber et al. 1985). With respect to
under study (e.g. prop-root and canopy removal), and single gear studies, we agree with Rozas & Minello
are relatively labor intensive. In contrast, passive sam- (1997) that enclosure samplers are superior for quanti-
plers such as fyke, flume, or channel nets do not fying fishes in structurally-complex habitats, especially
greatly modify the mangrove habitat, but are limited to in turbid waters, and add that visual surveys are par-
situations where tides are sufficient to drain the habitat ticularly useful in clearer waters.
and effectively force fishes into these capture devices
(McIvor & Odum 1986). Most traps, like entanglement
gears, can be rapidly deployed and cause minimal Habitat metrics
habitat disturbance. These gears can also be effective
for catching relatively large (ca. 10 cm total length) An historical summary of abiotic and biotic measures
mobile fishes often missed by other gears; however, collected aids our present ability to assess the value of
there are size- and species-selectivity constraints. Size mangroves as fish habitat. Unlike experimental studies
selectivity problems can be reduced by sampling with that benefit from being able to isolate and manipulate
traps of different mesh sizes and openings, or by specific variables for study, in the field it is difficult to
sampling with nets with multiple panels composed of choose appropriate abiotic (habitat) factors to measure
different meshes (Sogard et al. 1989, Sheaves 1995). since they are often autocorrelated with one another.
Unfortunately, like all passive samplers, only relative While the appropriate environmental variables mea-
abundance or biomass, rather than density or biomass sured in a study should differ depending on the goals
per unit area, can be estimated using traps and entan- of the project and on local conditions, over half of the
glement nets. Finally, underwater visual fish census habitat metrics recorded in the literature database con-
can be a rapid and effective technique for gathering sisted of temperature and salinity measurements, and
data and making quantitative comparisons of fish dis- this trend was observed across all geographic regions
tribution, abundance, and size-structure within and (Fig. 5a,b). Certainly temperature is linked to broad
among habitat types. Visual fish census has been uti- spatial patterns in the use of mangroves by fishes, as
lized in seagrass beds, mangroves, and hardbottom more fish species are noted from tropical estuaries
communities, and has become the most accepted than from sub-tropical estuaries, and a positive corre-
method for estimating fish abundance and diversity in lation between temperature and overall assemblage
10 Mar Ecol Prog Ser 318: 1–18, 2006
b) Habitat metrics: region
a) Habitat metrics: type
0.30 1.0
pH
0.25 0.8 Structure
Rain
0.20
0.6 DO
0.15 Turbidity
0.4 Depth
0.10
Temperature
0.2
0.05 Salinity
Proportion of studies
0.00 0.0
AM AU EA SA WA
ity
p
h
ity
DO
in
e
pH
ur
pt
m
Ra
lin
id
ct
Te
De
Region
rb
Sa
ru
Tu
St
d) Fish metrics: region
c) Fish metrics: type 1.0
0.35
0.30 G/Z
0.8
FO
0.25
Standing crop
0.6
0.20 Density
Biomass
0.15 0.4 Size
Percent
0.10
0.2 Presence
0.05
0.0
0.00
AM AU EA SA WA
ce
t
ze
s
ity
op
FO
/Z
n
en
as
tio
G
Si
ns
en
cr
Region
rc
om
uc
De
es
Pe
ng
od
Bi
Pr
di
Pr
an
St
Fig. 5. Summary of (a,b) habitat and (c,d) fish metrics collected from mangroves (a,c) among studies and (b,d) by geographic
region. DO: Dissolved oxygen; FO: frequency of occurrence; G/Z: growth and/or mortality estimates; rain: rainfall; temp:
temperature. Geographic regions abbreviated as in Fig. 4
richness, diversity, and abundance has been noted by ture on mangrove fishes (Connell 1978, Leigh 1990).
several authors (Robertson & Duke 1987, Williamson However, there is scant evidence suggesting such
et al. 1994, Lin & Shao 1999). While seasonal and relationships may also hold for fishes inhabiting
diel changes in temperature are typically predictable, mangroves (Serafy et al. 2003). Although regime
changes in salinity within a mangrove habitat can be characterization requires that dynamic abiotic vari-
more dynamic. Salinity can either remain relatively ables are measured over longer periods of time than
stable throughout the year (e.g. along well-connected just on the day of fish sampling, few studies in the liter-
oceanic islands), exhibit seasonal changes resulting ature have examined fish-habitat relationships on
from fluvial runoff, or change dramatically as a result multiple time scales (Bell et al. 1984, Lorenz 1999,
of anthropogenic freshwater releases (Faunce et al. Faunce et al. 2004).
2004). In addition, observed fish patterns in mangrove As in the case for temporal scales, examination of
habitats may be correlated with: (1) water depth, multiple spatial scales may be integral to determining
where the habitat is temporally inundated (e.g. which fishes utilize mangrove habitats and why; yet
Robertson & Duke 1990a, Laegdsgaard & Johnson this has also been largely ignored in the literature on
1995); (2) turbidity, where sediment transport is (a) mangrove fishes. At the smallest scales, structural
high (e.g. Blaber 1980) or (b) in areas without large complexity may be important, but this was reported in
salinity fluxes; and (3) dissolved oxygen in areas less than 5% of studies. It is likely that few field-based
with poor water flow or located downstream of large studies measured and reported structural measure-
industrial or agricultural areas (e.g. Claro et al. 2001). ments because early attempts failed to find meaning-
Abiotic regime (i.e. mean, range, and stability) may ful correlations with fish measures (Sheridan 1992,
be of more importance in structuring the assemblage Mullin 1995). Interestingly, experimental studies have
of mangrove fishes than ‘snapshot’ metrics collected demonstrated that the increased structural complexity
at the time of sampling. For example, a negative rela- of mangroves reduces the efficiency of predators (Pri-
tionship between environmental stability and species mavera 1997, Laegdsgaard & Johnson 2001). At larger
diversity has been well documented outside the litera- scales, many studies have sampled sites located at
11
Faunce & Serafy: Mangrove-fish literature review
various distances from features upstream (e.g. fresh Duke 1990b), and no estimates of habitat-specific mor-
water) versus downstream (e.g. coral reefs). However, tality or production have appeared in the literature.
distance values were reported in less than 10% of Because such studies may have been specifically
studies, and in only 2 instances were they used in focused on these biological metrics, it is possible that
analyses with fish metrics (Nagelkerken et al. 2000b, such studies exist outside the realm of this review.
Hajisamae & Chou 2003). Such analyses can be per- Other biotic factors such as larval supply, predation,
formed readily given the advances in global posi- competition, and food supply are difficult to consis-
tioning satellites and geographic information system tently and reliably measure, and we were unable to
technology (e.g. Kendall et al. 2003). find studies that reported these measures in the litera-
ture on mangrove fishes.
Fish metrics
Data analyses
Recently, several fish metrics have been reviewed
and ranked according to their usefulness for determin- Having summarized where, when, and how fish data
ing the importance of fish habitats, including man- have been collected from mangrove habitats, the ensu-
groves. In 1996, the US government mandated that all ing discussion is concerned with what has been done
stock assessments include EFH provisions and con- with them. Fish abundances were usually analyzed at
sider 4 levels of information (USDOC 1996). On a habi- one of 3 levels: the entire assemblage, ‘dominant taxa’
tat-specific basis, these include fish presence-absence (e.g. the 10 most abundant species grouped), or indi-
(Level 1), densities (Level 2), growth, reproduction, vidual species. A measure of the entire assemblage
and survival rates (Level 3), and secondary production (e.g. total fish density or biomass) was the most com-
rates (Level 4). A refined definition of ‘nursery habitat’ monly used level of analysis (52%), followed by the
emerged with a paper by Beck et al. (2001). They analysis of dominant taxa (31.2%) and then individual
contend that a nursery habitat contains one or more of species abundances (16.8%). As most studies aimed to
the following traits compared to other non-nursery identify patterns of fish use, analyses were focused on
habitats: (1) greater densities of young fishes; (2) lower examining temporal and spatial variation (Fig. 6a).
predation rates; (3) higher growth rates; and (4) more Fish–habitat correlations were examined less often,
successful migration to subsequent habitats (Beck et and were completely absent from the West African
al. 2001). region. Regardless of the focus, the type of data analy-
In light of these developments, this literature review sis conducted was typically in the form of simple
can be used to answer the question: What information side-by-side comparisons (Fig. 6b). Similarity indices,
is available to assess the value of mangroves as fish ANOVA, and ordination techniques were applied
habitat? Presence/absence information was the most with equal frequency among studies with spatial or
widely reported form of fish data, followed closely by temporal emphasis, and less often in studies with a
percentage composition (Fig. 5c). These 2 metrics fish-habitat focus; the latter investigation type alone-
accounted for over half the reported entries (31 and utilized multiple linear regression techniques.
24%, respectively) and were available from almost all Simple data comparisons dominated the literature.
surveys of mangrove fishes that we examined. Size Comparisons of fish data by family (38.1%) predomi-
information was less prevalent, and was present most nated, probably because this information is presented
often as part of a description of collected fishes. Only in species lists (Fig. 7a). Comparing fish groups
1 publication presented detailed size information for according to life-history (maturity) stage or as either
several species over time (Robertson & Duke 1990b). residents or transients (residency) was also common
Biomass information was more prevalent in the litera- (23.7 and 19.5%, respectively). Comparisons accord-
ture than either density or standing crop (i.e. numbers ing to trophic groups and diel period were among the
and biomass per unit area, respectively), both of which least reported in the database. The characterization
require information about the area sampled (Fig. 5c). and comparison of fishes according to their trophic
Remarkably, frequency of occurrence information (i.e. level can be a valuable tool for revealing their role in
the proportion of sites or repeated samples that con- system energy flow. The concept of using such func-
tained at least 1 individual), available from any survey, tional groups as a basis for site and ecosystem com-
went unreported in > 90% of studies. Density, standing parison and evaluation has recently been reviewed
crop, and frequency of occurrence data have not been for coral reefs (Bellwood et al. 2004) and holds
reported from the West African region (Fig. 5d). Per- promise for application to mangrove systems as well.
haps most limiting to mangrove fish habitat assess- Evidence is mounting that mangroves primarily serve
ment is that only 1 estimate of growth (Robertson & as daytime refugia for a major component of fishes
12 Mar Ecol Prog Ser 318: 1–18, 2006
b) Analysis type
a) Analysis focus
0.6
1.0
Habitat
Habitat
Proportion of studies
Spatial
0.5
0.8
Temporal
Spatial 0.4
0.6
Temporal 0.3
0.4
0.2
0.2 0.1
0.0
0.0
s
y
n
ns
VA
AM AU EA SA WA
on
rit
tio
o
O
ila
ris
si
na
Region
AN
m
es
pa
i
rd
Si
gr
om
O
Re
C
Fig. 6. Data analyses performed within studies of mangrove fishes organized by (a) geographic region and (b) type (i.e. statistical
test). Data are organized according to the purpose of analyses: identifying spatial patterns (spatial), identifying temporal patterns
(temporal), or exploring fish–habitat interactions (habitat). Non-statistical comparisons of data types (e.g. density) are labeled
comparisons. Similarity refers to indices of similarity, diversity, and evenness. Geographic regions abbreviated as in Fig. 4
occupying mangrove shorelines (Rooker & Dennis data from 19 studies that quantified fishes within
1991, Nagelkerken et al. 2000a, Valdés-Muñoz & mangroves and at least one other habitat.
Mochek 2001). This suggests for some species that
fish production attributed to mangroves may not nec-
essarily derive from this habitat alone (Adams et al. Current limitations
2006). Linkages between mangrove shorelines and
the proximity, size, and availability of nocturnal forag- Our review reveals that (1) certain regions, specifi-
ing areas, such as seagrass beds or mudflats, deserve cally South and Southeast Asia and West Africa, are
greater attention. under-represented in the literature, (2) the majority of
surveys were spatially restrictive and/or of short dura-
tion, and (3) numerous purported surveys of mangrove
fishes failed to sample within mangrove habitats per
SYNOPSIS
se. In defense of these studies, most were designed
This work represents the first attempt to assemble with modest goals in mind: (1) to identify which taxa
and examine a substantial number of published studies were present, and (2) to determine their abundances
of mangrove fishes. In contrast, previous reviews of among locations and/or sequential samples. While
the literature on mangrove fishes have been more these studies have been useful in identifying the com-
limited in scope. Thayer & Sheridan (1999) examined ponents and dynamics of various assemblages of man-
the methods and results of less than 12 studies from grove fishes, this type of data provides little informa-
Florida (USA), while Sheridan & Hays (2003) compiled tion with which to compare and evaluate the
a) Fish comparisons: type b) Fish comparisons: region
0.5 1.0
Proportion of studies
Diurnal
0.4 0.8
Trophic
0.3 0.6
Residency
0.2 0.4
Maturity
0.1 0.2
Family
0.0 0.0
AM AU EA SA WA
ily
y
y
c
l
na
it
nc
hi
m
ur
op
ur
Region
de
Fa
at
Di
Tr
si
M
Re
Fig. 7. How fish metrics were grouped for data comparisons (a) among studies, and (b) by geographic region. Geographic regions
abbreviated as in Fig. 4
13
Faunce & Serafy: Mangrove-fish literature review
importance of mangrove habitats for ecosystem diver- Future directions
sity and productivity.
Where fish–habitat correlations have been exam- The limitations above are important to consider
ined, most studies performed analyses at the assem- when planning future studies. Given the history of the
blage level using environmental data obtained at the literature, it appears likely that future studies will
time of sampling. Consequently, there is meagre infor- continue to examine its spatio-temporal patterns of
mation on how individual species respond to environ- mangrove use by fishes. In general, studies that exam-
mental variability. This is unfortunate, given that recent ine fish and habitat features at multiple spatial and
works suggest that the nature of the relationship be- temporal scales will be more valuable than those that
tween habitat features and fishes is species- and size- examine at only one scale. Irrespective of scale, studies
specific, and that only with this level of understanding that examine both the mean and variance of abiotic
can insight into ecological processes be gained (Bene- and biotic metrics will provide more insight than those
detti-Cecchi 2003, Cocheret de la Morinière et al. that only consider ‘static’ measures at the time of
2004). Furthermore, the type of fish metrics reported sampling. Power analysis and/or sampling efficiency
did not often include those most desired by decision evaluation are exceedingly rare in the literature, but
makers. For example, although simple presence/ are needed for mangrove fishes research to gain the
absence and percentage composition data were com- attention it deserves (Ley et al. 1999). It is possible that
monly reported, size and frequency of occurrence were statistical treatments such as these have been lacking
not, despite the fact that these metrics should be avail- in the literature partly because, at the species-specific
able from any survey that repeatedly samples fishes. level, ‘zero-laden’ fish abundance datasets are typical
The more detailed, difficult-to-collect data needed for and less than desirable for traditional statistical treat-
mangrove valuation were effectively absent from the ments. However, the problem of rarity is widespread in
database. With possibly 1 exception (i.e. Robertson & ecological studies (Gaston 1994), including those on
Duke 1990b), no study of mangrove fishes recorded ad- coral reef fishes (Jones et al. 2002). Researchers exam-
equate numbers over time, numbers at age, size-fre- ining species-specific fish abundance data will benefit
quencies, or tag-recapture data needed for the tradi- from the work of Aitchison (1955), Pennington (1983),
tional assessment of growth, mortality, or secondary Lo et al. (1992) and Johnson et al. (1999).
production. As a result, comparisons of EFH or nursery Several scientists are moving the study of mangrove
data from studies of mangrove fishes will likely be lim- fishes beyond pattern recognition towards more eco-
ited to density or biomass values only. If the USDOC logically meaningful landscape-scale approaches,
(1996) or Beck et al. (2001) definitions of these terms are including habitat connectivity, suitability, and the
to be used to assess the habitat value of mangroves, fu- contribution of mangrove habitats in support of adult fish
ture efforts must strive to track cohorts of fish over populations (e.g. Pittman et al. 2004, Sheaves 2005,
space and time. Mumby 2006). However, the need for species- and life-
Among the most significant conclusions to draw from stage-specific information on growth, mortality, and sec-
this review is that surveys of mangrove fishes are not ondary production rates remains. Although sufficient
readily comparable. Hence, the findings of any new age–length, biomass and size-distribution data exist for
study may be either bolstered or refuted using selected several species to generate habitat-specific production
references from the relevant literature. For example, estimates, this step has yet to be taken for mangrove
findings are mixed in studies relating habitat features to fishes. Attaining accurate home ranges and movement
assemblages of mangrove fishes with respect to water rates for mangrove fishes represents an additional and
temperature (Wright 1986, Lin & Shao 1999), salinity significant challenge towards linking juvenile and adult
(Quinn 1980, Ikejima et al. 2003), and turbidity (Little et stocks. Studies that examine habitat quality and avail-
al. 1988b, Kimani et al. 1996). Different conclusions re- ability are also needed to determine what makes some
garding the fish assemblage may be reached even when mangroves more important fish habitats than others.
2 studies have been conducted within the same body of
water. For example, Williamson et al. (1994), who sam-
Acknowledgements. This work was derived from a portion of a
pled with 8 mm mesh beach seine in Raby Bay, Australia,
doctoral dissertation by C. H. F. We thank the present and past
reported, ‘the majority of fish captures were either small library staff of the University of Miami’s Rosenstiel School of
species or juveniles’, while Moreton (1990) — who sam- Marine and Atmospheric Science (RSMAS), especially Helen
Albertson and Gail Clement. We also appreciate the efforts of
pled with 18 mm mesh seines and 100 to 150 mm mesh
numerous students and staff of the international community at
gill nets — stated, ‘most species .… were present as both
RSMAS for translating numerous publications. The comments
juveniles and adults’ and ‘standing-crop estimates for and additional literature sources provided by 4 anonymous
the fishes occurring within the mangroves were amongst reviewers greatly improved this manuscript. This paper is
the highest recorded values for estuarine areas’. Sustainable Fisheries Division Contribution SFD-2006-024.
14 Mar Ecol Prog Ser 318: 1–18, 2006
LITERATURE CITED Blaber SJM, Blaber TG (1980) Factors affecting the distribution
of juvenile estuarine and inshore fish. J Fish Biol 17:143–162
Acosta A (1997) Use of multi-mesh gillnets and trammel nets to Blaber SJM, Milton DA (1990) Species composition, commu-
estimate fish species composition in coral reef and man- nity structure and zoogeography of fishes of mangrove
groves in the southwest coast of Puerto Rico. Caribb J Sci estuaries in the Solomon Islands. Mar Biol 105:259–267
33:45–57 Blaber SJM, Young JW, Dunning MC (1985) Community struc-
Adams A, Dahlgren C, Kellison GT, Kendall MS, Layman CA, ture and zoogeographic affinities of the coastal fishes of the
Ley JA, Nagelkerken I, Serafy JE (2006) The juvenile con- Dampier Region of north-western Australia. Aust J Mar
tribution function of back reef systems. Mar Ecol Prog Ser Freshw Res 36:247–266
318:287–301 Blaber SJM, Brewer DT, Salini JP (1989) Species composition
Aitchison J (1955) On the distribution of a positive random vari- and biomasses of fishes in different habitats of a tropical
able having a discrete probability mass at the origin. J Am northern Australian estuary: their occurrence in the adjoin-
Stat Assoc 50:901–908 ing sea and estuarine dependence. Estuar Coast Shelf Sci
Alongi DM (2002) Present state and future of the world’s 29:509–531
mangrove forests. Environ Conserv 29:331–349 Blaber SJM, Brewer DT, Harris AN (1994) Distribution, bio-
Alvarez León R, Blanco Racedo J (1985) Composition of fish mass and community structure of demersal fishes of the
communities in the lagoon and estuarine complex of Carta- Gulf of Carpentaria, Australia. Aust J Mar Freshw Res 45:
gena Bay, Cienaga de Tesca and Cienaga Grande of Santa 375–396
Marta, Colombian Caribbean. In: Yáñez-Arancibia A (ed) Boeseman M (1963) An annotated list of fishes from the Niger
Fish community ecology in estuaries and coastal lagoons: delta. Zool Verh 61:1–54
towards an ecosystem integration. UNAM Press, Mexico Boltachev AR (1991) Species composition in the estuarine
City, p 535–656 ichthyocene of the mangrove zone of Guinea. J Ichthyol
Amadi AA, Kilham P, Mavuti KM (1990) A comparative ecol- 31:67–75
ogy of estuaries in Nigeria. Hydrobiologia 208:27–38 Boujard T, Rojas-Beltran R (1988) Longitudinal zonation of the
Araujo FG, De Azevedo MCC, Silva MA, Pessanha ALM, fish population of the Sinnamary River (French Guiana).
Gomes ID, Da Cruz-Filho AG (2002) Environmental influ- Rev Hydrobiol Trop 21:47–61
ences on the demersal fish assemblages in the Sepetiba Boulon RH Jr (1992) Use of mangrove prop root habitats by fish
Bay, Brazil. Estuaries 25:441–450 in the northern US Virgin Islands. Proc Gulf Caribb Fish
Atiqullah M, Shaukat SS, Qureshi N (1997) Diversity of fish Inst 41:189–204
communities at Sandspit Karachi coast, Pakistan. Oebalia Branch GM, Grindley JR (1979) Ecology of southern African
23:45–59 estuaries. Part XI. Mngazana: a mangrove estuary in
Austin HM (1971) A survey of the ichthyofauna of the man- Transkei. S Afr J Zool 14:149–170
groves of western Puerto Rico during December, 1967– Cervigón F (1985) The ichthyofauna of the Orinoco estuarine
August, 1968. Caribb J Sci 11:27–39 water delta in the West Atlantic coast, Caribbean. In:
Baran E, Hambrey J (1998) Mangrove conservation and coastal Yáñez-Arancibia A (ed) Fish community ecology in estuar-
management in southeast Asia: What impact on fishery ies and coastal lagoons: towards an ecosystem integration.
resources? Mar Pollut Bull 37:431–440 UNAM Press, Mexico City, p 57–78
Barletta-Bergan A, Barletta M, Saint-Paul U (2002) Community Chaves P, Bouchereau JL (1999) Biodiversity and dynamics
structure and temporal variability of ichthyoplankton in of ichthyic communities in the mangrove of Guaratuba,
North Brazilian mangrove creeks. J Fish Biol 61:33–51 Brazil. Oceanol Acta 22:353–364
Batista VS, Rego FN (1996) Análise de associações de peixes, Chaves P, Correa MFM (1998) Composicao ictiofaunistical da
em igarapés do estuário do Rio Tibiri, Maranhão. Rev Bras area de manguezal da Baia de Guaratuba, Parana, Basil.
Biol 56:163–176 Rev Bras Biol 15:195–202
Beck MW, Heck KL, Able KW, Childers DL and 9 others (2001) Chaves TC, Vendel AL (2001) Nota complementar sobre a
The identification, conservation, and management of estu- composicao ictiofaunistica da Baia de Guaratuba, Parana,
arine and marine nurseries for fish and invertebrates. Bio- Brasil. Rev Bras Biol 18:349–352
Science 51:633–641 Cheal AJ, Thompson AA (1997) Comparing visual counts of
Bell JD, Pollard DA, Burchmore JJ, Pease BC, Middleton MJ coral reef fish: implications of transect width and species
(1984) Structure of a fish community in a temperate tidal selection. Mar Ecol Prog Ser 158:241–248
mangrove creek in Botany Bay, New South Wales. Aust J Chong VC, Sasekumar A, Leh MUC, D’Cruz R (1990) The fish
Mar Freshw Res 35:33–46 and prawn communities of a Malaysian coastal mangrove
Bell JD, Craik GJS, Pollard DA, Russell BC (1985) Estimating system, with comparisons to adjacent mud flats and inshore
length frequency-distributions of large reef fish under- waters. Estuar Coast Shelf Sci 31:703–722
water. Coral Reefs 4:41–44 Claro R, García-Arteaga JP (1993) Fish community structure on
Bellwood DR, Hughes TP, Folke C, Nystrom M (2004) Con- mangroves from the Insular Group Sabana-Camaguey,
fronting the coral reef crisis. Nature 429:827–833 Cuba. Avicennia 0:60–83
Benedetti-Cecchi L (2003) The importance of the variance Claro R, Lindeman KC, Parenti LR (2001) Ecology of the marine
around the mean effect size of ecological processes. fishes of Cuba. Smithsonian Institution Press, Washington,
Ecology 84:2335–2346 DC
Blaber SJM (1980) Fish of the Trinity Inlet system of north Cocheret de la Morinière E, Pollux BJA, Nagelkerken I, van
Queensland with notes on the ecology of fish faunas of der Velde G (2002) Post-settlement life cycle migration pat-
tropical Indo-Pacific estuaries. Aust J Mar Freshw Res 31: terns and habitat preference of coral reef fish that use sea-
137–146 grass and mangrove habitats as nurseries. Estuar Coast
Blaber SJM (2000) Tropical estuarine fishes: ecology, exploita- Shelf Sci 55:309–321
tion, and conservation. Blackwell Science, London Cocheret de la Morinière E, Nagelkerken I, van der Meij H,
Blaber SJM (2002) ‘Fish in hot water’: the challenges facing fish van der Velde G (2004) What attracts juvenile coral reef fish
and fisheries research in tropical estuaries. J Fish Biol 61:1–20 to mangroves: habitat complexity or shade? Mar Biol 144:
15
Faunce & Serafy: Mangrove-fish literature review
139–145 Johnson DR, Funicelli NA, Bohnsack JA (1999) Effectiveness of
Collette BB (1983) Mangrove fishes of New Guinea. In: Teas HJ an existing estuarine no-take fish sanctuary within the
Biology and ecology of mangroves. Dr. W. Junk Publishers, Kennedy Space Center, Florida. North Am J Fish Manag
The Hague, p 91–102 19:436–453
Connell JH (1978) Diversity in tropical rain forests and coral Jones GP, Caley MJ, Munday PL (2002) Rarity in coral reef fish
reefs. Science 199:1302–1310 communities. In: Sale PF (ed) Coral reef fishes: dynamics
Dankwa HR, Gordon C (2002) The fish and fisheries of the and diversity in a complex ecosystem. Academic Press, San
Lower Volta mangrove swamps in Ghana. Afr J Sci Technol Diego, CA, p 81–102
(Sci Eng Ser) 3:25–32 Kendall MS, Christensen JD, Hillis-Starr Z (2003) Multi-scale
Davis TLO (1988) Temporal changes in the fish fauna entering data used to analyze the spatial distribution of French
grunts, Haemulon flavolineatum, relative to hard and soft
a tidal swamp system in tropical Australia. Environ Biol
Fish 21:161–172 bottom in a benthic landscape. Environ Biol Fish 66:19–26
Day JH (1974) The ecology of Morrumbene estuary, Mozam- Kimani EN, Mwatha GK, Wakwabi EO, Ntiba JM, Okoth BK
bique. Trans R Soc S Afr 41:43–97 (1996) Fishes of a shallow tropical mangrove estuary, Gazi,
Dennis GD (1992) Island mangrove habitats as spawning and Kenya. Mar Freshw Res 47:857–868
nursery areas for commercially important fishes in the Krishnamurthy K, Jeyasslan MJP (1981) The early life history
Caribbean. Proc Gulf Caribb Fish Inst 41:205–225 of fishes from Pichavaram mangrove ecosystem of India. In:
Faunce CH, Serafy JE, Lorenz JJ (2004) Density-habitat rela- Lasker R, Sherman K (eds) The early life history of fish:
tionships of mangrove creek fishes within the southeastern recent studies. Second ICES Symposium. Cons Int Explor
saline Everglades (USA), with reference to managed fresh- Mer, Palegade 2–4, Copenhagen, p 416–423
water releases. Wetlands Ecol Manag 12:377–394 Kuo S, Lin H, Shao K (1999) Fish assemblages in the mangrove
Flores-Verdugo F, González-Farias F, Ramírez-Flores O, creeks of northern and southern Taiwan. Estuaries 22:
Amezcua-Linares A, Yáñez-Arancibia A, Alvarez-Rubio M, 1004–1015
Day JW Jr (1990) Mangrove ecology, aquatic primary pro- Laegdsgaard P, Johnson CR (1995) Mangrove habitats as nurs-
ductivity, and fish community dynamics in the Teacapan- eries: unique assemblages of juvenile fish in subtropical
Agua Brava lagoon-estuarine system (Mexican Pacific). mangroves in eastern Australia. Mar Ecol Prog Ser 126:67–81
Estuaries 13:219–230 Laegdsgaard P, Johnson CR (2001) Why do juvenile fish utilize
Gaston KJ (1994) Rarity. Chapman & Hall, London mangrove habitats? J Exp Mar Biol Ecol 257:229–253
Gillanders BM, Able KW, Brown JA, Eggleston DB, Sheridan Lal PN (1984) Coastal fisheries and the management of man-
PF (2003) Evidence of connectivity between juvenile and grove resources in Fiji. SPC Fish Newsl:15–23
adult habitats for mobile marine fauna: an important com- Laroche J, Baran E, Rasoanandrasana NB (1997) Temporal pat-
ponent of nurseries. Mar Ecol Prog Ser 247:281–295 terns in a fish assemblage of a semiarid mangrove zone in
Gonzalez-Acosta AF, De la Cruz Agüero J, Ruiz-Campos G Madagascar. J Fish Biol 51:3–20
(1999) Ichthyofauna associated to El Conchalito mangrove Lasserre G, Toffart JL (1977) Echantillonnage et structure des
swamp, Ensenada de La Paz, Baja California Sur, Mexico. populations ichthyologiques des mangroves de Guade-
Investigaciones Marinas CICIMAR 14:121–131 loupe en Septembre 1975. Cybium 3:115–127
Greene LE, Alevizon WS (1989) Comparative accuracies of Leigh EG (1990) Community diversity and environmental
visual assessment methods for coral reef fishes. Bull Mar Sci stability— a reexamination. Trends Ecol Evol 5:340–344
44:899–912 Ley JA, McIvor CC (2001) Linkages between estuarine and
Hajisamae S, Chou LM (2003) Do shallow water habitats of an reef fish assemblages: enhancement by the presence of
impacted coastal strait serve as nursery grounds for fish? well-developed mangrove shorelines. In: Porter JW, Porter
Estuar Coast Shelf Sci 56:281–290 KG (eds) The Everglades, Florida Bay, and coral reefs of the
Halliday IA, Young WR (1996) Density, biomass and species Florida Keys: an ecosystem sourcebook. CRC Press, Boca
composition of fish in a subtropical Rhizophora stylosa Raton, FL, p 539–562
mangrove forest. Mar Freshw Res 47:609–615 Ley JA, McIvor CC, Montague CL (1999) Fishes in mangrove
He B, Fan H, Mo Z (2001) Study on species diversity of fishes in prop-root habitats of northeastern Florida Bay: distinct
mangrove area of Yingluo Bay, Guangxi province. J Trop assemblages across an estuarine gradient. Estuar Coast
Oceanogr 20:74–79 Shelf Sci 48:701–723
Heck JL, Hays G, Orth RJ (2003) Critical evaluation of the Lin HJ, Shao KT (1999) Seasonal and diel changes in a subtrop-
nursery role hypothesis for seagrass meadows. Mar Ecol ical mangrove fish assemblage. Bull Mar Sci:775–794
Prog Ser 253:123–136 Lindeman KC, Snyder DB (1999) Nearshore hardbottom fishes
Holm RF (1977) The standing crop of fishes in a tropical marine of southeast Florida and effects of habitat burial caused by
lagoon. Fla Sci 40:258–261 dredging. Fish Bull 97:508–525
Ikejima K, Tongnunui P, Medej T, Taniuchi T (2003) Juvenile Little MC, Reay PJ, Grove SJ (1988a) Distribution gradients
and small fishes in a mangrove estuary in Trang Province, of ichthyoplankton in an East African mangrove creek.
Thailand: seasonal and habitat differences. Estuar Coast Estuar Coast Shelf Sci 26:669–677
Shelf Sci 56:447–457 Little MC, Reay PJ, Grove SJ (1988b) The fish community of an
Inger RE (1955) Ecological notes on the fish fauna of a coastal East African mangrove creek. J Fish Biol 32:729–747
drainage of North Borneo. Fieldiana Zool 37:47–90 Lo NCH, Jacobson LD, Squire JL (1992) Indexes of relative
Janekarn V (1993) Species composition and annual population abundance from fish spotter data dased on Delta-Lognormal
growth of fishes in front of a mangrove in Phang-Nga Bay, models. Can J Fish Aquat Sci 49:2515–2526
the Andaman Sea, Thailand. Phuket Mar Biol Cent (Spec. Lorenz JJ (1999) The response of fishes to physicochemi-
Publ.) 12:131–140 cal changes in the mangroves of northeast Florida Bay.
Janekarn V, Boonruang P (1986) Composition and occurrence Estuaries 22:500–517
of fish larvae in mangrove areas along the east coast of Louis M, Bouchon C, Bouchon-Navaro Y (1992) L’ichtyofaune
Phuket Island, western peninsular, Thailand. Phuket Mar de mangrove dans la Baie de Fort-de-France (Martinique).
Biol Cent Res Bull 44:1–22 Cybium 16:291–305
16 Mar Ecol Prog Ser 318: 1–18, 2006
Louis M, Bouchon C, Bouchon-Navaro Y (1995) Spatial and tions in catch from two mangrove habitats in the Lagos
temporal variations of mangrove fish assemblages in Mar- Lagoon. Environ Ecol 13:121–128
tinique (French West Indies). Hydrobiologia 295:275–284 Odum WE, McIvor CC, Smith TJ, III (1982) Ecology of the
Louis M, Lam Hoai T, Lasserre G (1985) Résultats préliminaires mangroves of south Florida: a community profile. Office of
sur le recrutement en poissons dans deux lagunes des man- Biological Services, Fish and Wildlife Service, US Dept. of
groves de Guadeloupe: Belle-Plaine et Manche-à-Eau. Rev Interior, Washington, DC
Hydrobiol Trop 18:249–265 Paranagua MN, Eskinazi-Leça E (1985) Ecology of a northern
Lugendo BR, Pronker A, Cornelissen I, de Groene A, Nagel- tropical estuary in Brazil and technological perspectives
kerken I, Dorenbosch M, van der Velde G, Mgaya YD in fish culture. In: Yáñez-Arancibia A (ed) Fish community
(2005) Habitat utilisation by juveniles of commercially ecology in estuaries and coastal lagoons: towards an
important fish species in a marine embayment in Zanzibar, ecosystem integration. UNAM Press, Mexico City,
Tanzania. Aquat Living Resour 18:149–158 p 595–614
Lugo AE, Snedaker SC (1974) The ecology of mangroves. Pennington M (1983) Efficient estimators of abundance, for fish
Annu Rev Ecol Syst 5:39–64 and plankton surveys. Biometrics 39:281–286
Manly BFJ, McDonald LL, Thomas DL (1993) Resource selec- Phillips PC (1981a) Annotated checklist of fishes at Jiquilisco
tion by animals: statistical design and analysis for field Bay, El Salvador. Rev Biol Trop 29:45–58
studies. Chapman & Hall, London Phillips PC (1981b) Diversity and fish community structure in
Martínez AM, Contreras S, Maravilla O (1982) Abundancia y a central American mangrove embayment. Rev Biol Trop
diversidad de la ictiofauna, en tres areas de manglar de la 29:227–236
Bahía de la Paz, México. Trans Cibcasio 6:138–151 Pinto L (1987) Environmental factors influencing the occur-
McIvor CC, Odum WE (1986) The flume net: a quantitative rence of juvenile fish in the mangroves of Pagbilao, Philip-
method for sampling fishes and macrocrustaceans on tidal pines. Hydrobiologia 150:283–301
marsh surfaces. Estuaries 9:219–224 Pinto L, Punchihewa NN (1996) Utilization of mangroves and
Morton RM (1990) Community structure, density and standing seagrasses by fishes in the Negombo Estuary, Sri Lanka.
crop of fishes in a subtropical Australian mangrove area. Mar Biol 126:333–345
Mar Biol 105:385–394 Pittman SJ, McAlpine CA, Pittman KM (2004) Linking fish
Mullin SJ (1995) Estuarine fish populations among red man- and prawns to their environment: a hierarchical landscape
grove prop roots of small overwash islands. Wetlands 15: approach. Mar Ecol Prog Ser 283:233–254
324–329 Pollard DA, Hannan JC (1994) The ecological effects of
Mumby PJ (2006) Connectivity of reef fish between mangroves structural flood mitigation works on fish habitats and fish
and coral reefs: algorithms for the design of marine communities in the lower Clarence River system of south-
reserves. Biol Conserv 128:215–222 eastern Australia. Estuaries 17:427–461
Nagelkerken I, van der Velde G (2002) Do non-estuarine Primavera JH (1997) Fish predation on mangrove-associated
mangroves harbour higher densities of juvenile fish than penaeids: the role of structures and substrate. J Exp Mar
adjacent shallow-water and coral reef habitats in Curaçao Biol Ecol 215:205–216
(Netherlands Antilles)? Mar Ecol Prog Ser 245:191–204 Quinn NJ (1980) Analysis of temporal changes in fish assem-
Nagelkerken I, Dorenbosch M, Verberk WCEP, Cocheret de la blages in Serpentine Creek, Queensland. Environ Biol Fish
Morinière E, van der Velde G (2000a) Day–night shifts of 5:117–133
fishes between shallow-water biotopes of a Caribbean bay, Quinn NJ, Kojis BJ (1985) Does the presence of coral reefs in
with emphasis on the nocturnal feeding of Haemulidae and proximity to a tropical estuary affect the estuarine fish
Lutjanidae. Mar Ecol Prog Ser 194:55–64 assemblage? In: Harmelin-Vivien M, Salvat B, La Croix C,
Nagelkerken I, Dorenbosch M, Verberk WCEP, Cocheret de la Gabrie C, Toffart JL (eds) Proc Fifth Int Coral Reef Congr,
Morinière E, van der Velde G (2000b) Importance of shal- Tahiti, 27 May–1 Jun 1985, Miscellaneous papers (A),
low-water biotopes of a Caribbean bay for juvenile coral Vol 5. Antenne Museum-EPHE, Moorea, p 445–450
reef fishes: patterns in biotope association, community Robertson AI, Duke NC (1987) Mangroves as nursery sites:
structure and spatial distribution. Mar Ecol Prog Ser comparisons of the abundance and species composition of
fish and crustaceans in mangroves and other nearshore
202:175–192
habitats in tropical Australia. Mar Biol 96:193–205
Nagelkerken I, van der Velde G, Gorissen MW, Meijer GJ,
Robertson AI, Duke NC (1990a) Mangrove fish-communities in
van’t Hof T, den Hartog C (2000c) Importance of man-
tropical Queensland, Australia: spatial and temporal pat-
groves, seagrass beds and the shallow coral reef as a nurs-
terns in densities, biomass and community structure. Mar
ery for important coral reef fishes, using a visual census
Biol 104:369–379
technique. Estuar Coast Shelf Sci 51:31–44
Robertson AI, Duke NC (1990b) Recruitment, growth and resi-
Nagelkerken I, Kleijnen S, Klop T, van den Brand RACJ,
dence time of fishes in a tropical Australian mangrove
Cocheret de la Morinière EC, van der Velde G (2001)
system. Estuar Coast Shelf Sci 31:723–743
Dependence of Caribbean reef fishes on mangroves and
Rojas JRM, Castro M, Pizarro JF (1994a) Lista agregada de
seagrass beds as nursery habitats: a comparison of fish fau-
peces en tres zonas de manglar del Golfo de Nicoya, Costa
nas between bays with and without mangroves/seagrass
Rica. Uniciencia 11:89–96
beds. Mar Ecol Prog Ser 214:225–235
Rojas JRM, Pizarro JF, Castro MV (1994b) Diversidad y abun-
Nagelkerken I, Roberts CM, van der Velde G, Dorenbosch M,
dancia íctica en tres áreas de manglar en el Golfo de
van Riel MC, Cocheret de la Morinière E, Nienhuis PH
Nicoya, Costa Rica. Rev Biol Trop 42:663–672
(2002) How important are mangroves and seagrass beds for
Rönnbäck P, Troell M, Kautsky N, Primavera JH (1999) Distrib-
coral-reef fish? The nursery hypothesis tested on an island
ution pattern of shrimps and fish among Avicennia and
scale. Mar Ecol Prog Ser 244:299–305
Rhizophora microhabitats in the Pagbilao Mangroves,
Newman SP, Gruber SH (2002) Comparison of mangrove and
Philippines. Estuar Coast Shelf Sci 48:223–234
seagrass fish and macroinvertebrate communities in
Rooker JR, Dennis GD (1991) Diel, lunar and seasonal changes
Bimini. Bahamas J Sci 9:19–27
in a mangrove fish assemblage off southwestern Puerto
Nwadukwe FO (1995) Species abundance and seasonal varia-
17
Faunce & Serafy: Mangrove-fish literature review
Rico. Bull Mar Sci 49:684–698 inventories of coral reefs, soft bottoms and mangroves in
Rozas LP, Minello TJ (1997) Estimating densities of small fishes Saint-Vincent Bay (New Caledonia). Proc Sixth Int Coral
and decapod crustaceans in shallow estuarine habitats: a Reef Symp 2, Townsville, p 613–618
review of sampling design with focus on gear selection. Thompson AA, Mapstone BD (1997) Observer effects and
Estuaries 20:199–213 training in underwater visual surveys of reef fishes. Mar
Russ GR, Stockwell B, Alcala AC (2005) Inferring versus Ecol Prog Ser 154:53–63
measuring rates of recovery in no-take marine reserves. Tobias WJ (2001) Mangrove habitat as nursery grounds for re-
Mar Ecol Prog Ser 292:1–12 creationally important fish species — Great Pond, St. Croix,
Sasekumar A, Chong VC, Leh MU, D’Cruz R (1992) Man- U.S. Virgin Islands. Proc Gulf Caribb Fish Inst 52:468–487
groves as a habitat for fish prawns. Hydrobiologia 247: Tongnunui P, Ikejima K, Yamane T, Horinouchi M, Medej T,
195–207 Sano M, Kurokura H, Taniuchi T (2002) Fish fauna of the
Savage RE (1931) The relation between the feeding of the her- Sikao Creek mangrove estuary, Trang, Thailand. Fish Sci
ring off the east coast of England and the plankton of the 68:10–17
surrounding waters. Fish Invest Lond Ser II 12:1–88 Tzeng W, Wang Y (1992) Structure, composition and seasonal
Sedberry GR, Carter J (1993) The fish community of a shallow dynamics of the larval and juvenile fish community in the
tropical lagoon in Belize, Central America. Estuaries mangrove estuary of Tanshui river, Taiwan. Mar Biol 113:
16:198–215 481–490
Serafy JE, Faunce CH, Lorenz JJ (2003) Mangrove shoreline Uieda VS, Uieda W (2001) Species composition and spatial dis-
fishes of Biscayne Bay, Florida. Bull Mar Sci 72:161–180 tribution of a stream fish assemblage in the east coast of
Sheaves MJ (1992) Patterns of distribution and abundance of Brazil: comparison oftwo field study methodologies. Brazilian
fishes in different habitats of a mangrove-lined tropical J Biol 61:377–388
estuary, as determined by fish trapping. Aust J Mar Freshw USDOC (1996) Magnusen-Stevens Fishery Conservation and
Res 43:1461–1479 Management Act as amended through October 11, 1996,
Sheaves M (1995) Large lutjanid and serranid fishes in tropical National Oceanic and Atmospheric Administration Techni-
estuaries: Are they adults or juveniles? Mar Ecol Prog Ser cal Memorandum NMFS-F/SPO-23
129:31–40 Valdés-Muñoz E (1981) Estructura y diversidad de la ictiofauna
Sheaves MJ (1998) Spatial patterns in estuarine fish faunas in de los manglares de Punta del Este, Isla de la Juventud.
tropical Queensland: a reflection of interaction between Cienc Biol 6:111–124
long-term physical and biological processes? Mar Freshw Valdés-Muñoz E, Mochek AD (2001) Behavior of marine fishes
Res 49:31–40 of the Cuban shelf. In: Claro R, Lindeman KC, Parenti LR
Sheaves M (2005) Nature and consequences of biological (eds) Ecology of the marine fishes of Cuba. Smithsonian
connectivity in mangrove systems. Mar Ecol Prog Ser 302: Institution Press, Washington, DC, p 73–114
293–305 van der Velde G, Gorissen MW, Denhartog C, Vanthoff T,
Sheridan PF (1992) Comparative habitat utilization by estuar- Meijer GJ (1992) Importance of the Lac-Lagoon (Bonaire,
ine macrofauna within the mangrove ecosystem of Rookery Netherlands-Antilles) for a selected number of reef fish
Bay, Florida. Bull Mar Sci 50:21–39 species. Hydrobiologia 247:139–140
Sheridan P, Hays C (2003) Are mangroves nursery habitat for Vance DJ, Haywood MDE, Heales DS, Kenyon RA, Loneragan
transient fishes and decapods? Wetlands 23:449–458 NR, Pendrey RC (1996) How far do prawns and fish move
Snedaker SC (1989) Overview of ecology of mangroves and into mangroves? Distribution of juvenile banana prawns
Penaeus merguiensis and fish in a tropical mangrove forest
information needs for Florida Bay. Bull Mar Sci 44:341–347
Sogard SM, Powell GVN, Holmquist JG (1989) Spatial- in northern Australia. Mar Ecol Prog Ser 131:115–124
distribution and trends in abundance of fishes residing in Vega-Cendejas ME, de Santillana MH (2004) Fish community
seagrass meadows on Florida Bay mudbanks. Bull Mar Sci structure and dynamics in a coastal hypersaline lagoon: Rio
44:179–199 Lagartos,Yucatan, Mexico. Estuar Coast Shelf Sci 60:285–299
Spalding M, Blasco F, Field C (1997) World Mangrove Atlas. Vega-Cendejas ME, Ordóñez U, Hernández M (1994) Day-
The International Society for Mangrove Ecosystems Night variation of fish population in the mangrove of
(ISME), Okinawa Celestun lagoon, Mexico. Int J Ecol Env Sci 20:99–108
St. John J, Russ GR, Gladstone W (1990) Accuracy and bias of Vidy G (2000) Estuarine and mangrove systems and the nurs-
visual estimates of numbers, size structure and biomass of a ery concept: which is which? The case of the Sine Saloum
coral reef fish. Mar Ecol Prog Ser 64:253–262 system (Senegal). Wetl Ecol Manag:37–51
Stoner AW (1986) Community structure of the demersal fish Vose FE, Bell SS (1994) Resident fishes and macrobenthos in
species of Laguna Joyuda, Puerto Rico. Estuaries 9:142–152 mangrove-rimmed habitats: evaluation of habitat restora-
Thayer GW, Sheridan P (1999) Fish and aquatic invertebrate tion by hydrologic modification. Estuaries 17:585–596
use of the mangrove prop-root habitat in Florida: a review. Warburton K (1978) Community structure, abundance and
In: Yáñez-Arancibia A, Lara-Domínguez AL (eds) Ecosis- diversity of fish in a Mexican coastal lagoon system. Estuar
temas de Manglar en America Tropical. Instituto de Ecolo- Coast Mar Sci 7:497–519
gia, A.C. Mexico, UICN/ORMA Costa Rica, NOAA/NMFS Weinstein MP, Courtney CM, Kinch JC (1977) The Marco
Silver Spring, MD, p 380 Island estuary: a summary of physicochemical and biologi-
Thayer GW, Colby DR, Hettler WF Jr (1987) Utilization of the cal parameters. Fla Sci 40:97–124
red mangrove prop root habitat by fishes in South Florida. Weng HT (1990) Fish in shallow areas in Moreton Bay,
Mar Ecol Prog Ser 35:25–38 Queensland and factors affecting their distribution. Estuar
Thia-Eng C (11973) An ecological study of the Ponggol estuary Coast Shelf Sci 30:569–578
in Singapore. Hydrobiol 43:3–4 Williamson I, King C, Mather PB (1994) A comparison of fish
Thollot P (1992) Importance des mangroves pour la faune communities in unmodified and modified inshore habitats
ichtyologique des récifs coralliens de Nouvelle-Calédonie. of Raby Bay, Queensland. Estuar Coast Shelf Sci 39:
Cybium 16:331–344 401–411
Thollot P, Kulbicki M (1988) Overlap between the fish fauna Wright JM (1986) The ecology of fish occurring in shallow
18 Mar Ecol Prog Ser 318: 1–18, 2006
water creeks of a Nigerian mangrove swamp. J Fish Biol mass and diversity of estuarine fishes coupled with tropical
29:431–441 habitat heterogeneity (southern Gulf of Mexico). J Fish Biol
Yáñez-Arancibia A (1985) Fish community ecology in estuaries 33:191–200
and coastal lagoons: towards an ecosystem integration. Yáñez-Arancibia A, Lara-Domínguez AL, Day JW Jr (1993) In-
UNAM Press, Mexico teractions between mangrove and seagrass habitats medi-
Yáñez-Arancibia A, Lara-Domínguez AL, Rojas-Galaviz JL, ated by estuarine nekton assemblages: coupling of primary
Sanchez-Gil P, Day JW Jr, Madden CJ (1988) Seasonal bio- and secondary production. Hydrobiologia 264:1–12
Editorial responsibility: Howard I. Browman (Associate Submitted: February 17, 2005; Accepted: May 24, 2006
Editor-in-Chief), Storebø, Norway Proofs received from author(s): July 12, 2006