Content uploaded by Simon James Pittman
Author content
All content in this area was uploaded by Simon James Pittman
Content may be subject to copyright.
282
Caribbean Journal of Science, Vol. 45, No. 2-3, 282-303, 2009
Copyright 2009 College of Arts and Sciences
University of Puerto Rico, Mayagüez
Introduction
Many fish species in tropical coral reef
ecosystems connect multiple habitats
through regular nocturnal migrations into
neighboring habitats, with some species
using several distinct resources across a
compositionally complex mosaic of habi-
tats (Parrish 1989; Kramer and Chapman
1999; Nagelkerken et al. 2000; Cocheret de
la Moriniere 2002; Dorenbosch et al. 2005;
Unsworth et al. 2008). For instance, Meyer
et al. (1983) reported at least 15 fish fami-
lies that leave coral reefs to forage in neigh-
boring areas. Haemulidae (grunts) and
Lutjanidae (snappers) in the Caribbean
have frequently been observed to under-
take sun-synchronous migrations by leav-
ing their daytime shelter on coral reefs and
mangroves at dusk to migrate to adjacent
seagrass and sand beds to forage at night
before returning to the more structured hab-
itat types by dawn (Ogden and Ehrlich 1977,
Helfman et al. 1982; Rooker and Dennis
1991; Nagelkerken et al. 2000; Monaco et al.
2009). Seagrass beds provide a high abun-
dance of food and suitable refuge in low
light conditions, thus functioning as a com-
plementary or supplementary resource for
many multi-habitat species (Pittman et al.
2004, Pittman et al. 2007). This pattern of
day-night resource use has likely evolved to
maximize growth while minimizing mor-
tality through predation (Dahlgren and
Eggleston 2000, Grol et al. 2008). Community
studies have shown that nocturnal excur-
sions can result in pronounced diel shifts in
fish assemblage composition across inter-
connected coral reef ecosystems (Kopp et al.
Nocturnal fish movement and trophic flow across habitat
boundaries in a coral reef ecosystem (SW Puerto Rico)
Randall D. Clark
1
* , Simon Pittman
1
, Chris Caldow
1
, John Christensen
1
,
Bryant Roque
2
, Richard S. Appeldoorn
2
, and Mark E. Monaco
1
1
National Oceanic and Atmospheric Administration, Biogeography Branch, 1305 East-West Highway,
Silver Spring, MD 20910 USA
2
Department of Marine Sciences, University of Puerto Rico, Mayaguez, Puerto Rico 00681-9013 USA
* Email: randy.clark@noaa.gov
A BSTRACT. — Few studies have quantified the extent of nocturnal cross-habitat movements for fish, or the
influence of habitat adjacencies on nutrient flows and trophodynamics. To investigate the patterns of noc-
turnal cross-boundary movements of fish and quantify trophic connectivity, fish were sampled at night
with gillnets set along the boundaries between dominant habitat types (coral reef/seagrass and mangrove/
seagrass) in southwestern Puerto Rico. Fish movement across adjacent boundary patches were equivalent at
both coral reefs and mangroves. Prey biomass transfer was greater from seagrass to coral reefs (0.016 kg/km)
and from mangroves to seagrass (0.006 kg/km) but not statistically significant, indicating a balance of flow
between adjacent habitats. Pelagic species (jacks, sharks, rays) accounted for 37% of prey biomass transport
at coral reef/seagrass and 46% at mangrove/seagrass while grunts and snappers accounted for 7% and 15%,
respectively. This study indicated that coral reefs and mangroves serve as a feeding area for a wide range of
multi-habitat fish species. Crabs were the most frequent prey item in fish leaving coral reefs while molluscs
were observed slightly more frequently than crabs in fish entering coral reefs. For most prey types, biomass
exported from mangroves was greater than biomass imported. The information on direction of fish movement
together with analysis of prey data provided strong evidence of ecological linkages between distinct adjacent
habitat types and highlighted the need for greater inclusion of a mosaic of multiple habitats when attempting
to understand ecosystem function including the spatial transfer of energy across the seascape.
K EYWORDS. — connectivity , habitat boundaries , coral reef ecosystems , mangroves , fish prey , nocturnal fish
movement , seascape
283
NOCTURNAL CORAL REEF FISH MOVEMENT, SW PUERTO RICO
2007; Unsworth et al. 2007). Since most stud-
ies of fish movement occur during daylight
hours, little is known about the identity and
abundance of nocturnal trans-boundary
movements. Earlier studies of migrating
resident species (e.g. grunts, snappers) sug-
gest net trophic flow of prey biomass to be
greater coming back to the resting habitat
type (Ogden and Ehrlich 1977; Ogden and
Zieman 1977; McFarland et al. 1979).
Fish migrations have long been consid-
ered to be important conduits of organic
and inorganic material to and from coral
reefs, mangroves and surrounding areas
(Birkeland 1985; Meyer and Schultz 1985;
Parrish 1989; Sheaves 2005), yet since the
publication of Randall’s (1967) dietary sur-
vey of 212 Caribbean fish relatively little
quantitative information is available on
multi-species foraging and diets. Even fewer
studies have focused on the exchanges
of organisms and dietary material across
boundaries between neighboring habitats.
This is an important knowledge gap since
fishes have been shown to be a significant
source of organic carbon and other nutrients
in tropical marine ecosystems (Bray et al.
1981; Ogden and Gladfelter 1983). For exam-
ple, early experiments on the effects of excre-
tion and defecation from migrating schools
of fish over coral reefs indicated significant
inputs of ammonium and phosphorous that
may enhance the growth of macroalgae
(Meyer et al. 1983; Meyer and Schultz 1985).
In addition, foraging fish have been identi-
fied as a key redistributer of sediment par-
ticles in coral reef ecosystems (Alheit 1981)
and the transport of nutrients and transfer
of energy away from mangroves by mobile
animals can have important consequences
for recycling in mangroves (Sheaves and
Molony 2000; Sheaves 2005).
Parrish (1989) and Polis et al. (1997)
argue that trophic interactions that con-
nect discrete habitat types can exert a major
influence on the local abundance and distri-
bution of organisms through both “bottom-
up” processes as a result of cross-habitat
transport of materials and nutrients and
top-down processes as a result of preda-
tion. As we progress toward understanding
seascape structure and the dynamics and
energy pathways across the seascape, there
is an urgent need for quantitative infor-
mation capable of identifying pathways of
energy flow and determining the influence
of habitat boundaries and habitat adjacen-
cies on these processes. In this way, we begin
to link ecological patterns with dynamic
ecological processes across structural mosa-
ics of habitat in coral reef ecosystems. This
fundamental ecological information can be
applied to help understand factors includ-
ing human modifications that may enhance
or limit energy flow across coral reef eco-
systems and may also help determine the
optimal design of Marine Protected Areas
(MPAs) and help refine designations of
Essential Fish Habitat (EFH) (Polunin and
Roberts 1993; Murray et al. 1999).
This paper examines inter-habitat trophic
connectivity across the seascape of La
Parguera, SW Puerto Rico by examining the
flow of fish biomass and their associated
prey. Numerous adult and sub-adult fish
species were captured during their noctur-
nal excursions across: (A) the boundary of
coral reefs and seagrasses ( Cr-Sg ), and (B)
the boundary of mangroves and seagrasses
( Mg-Sg ) and quantitative data on their prey
consumption were collected. Trophic flow,
defined here as the cross-boundary move-
ment of fish and prey biomass, was exam-
ined at Cr-Sg and Mg-Sg boundaries with
special emphasis on the role of Haemulidae
(grunts) and Lutjanidae (snappers) due to
their abundance in coral reef ecosystems
and well documented nocturnal migrations.
Three main questions were addressed
through quantitative descriptions and hypo-
th esis testing:
(1) What are the quantities of fish and the
biomass of their consumed prey that
are moving into and out from coral
reefs and mangroves through noctur-
nal excursions?
(2) Are haemulids (grunts) and lutjanids
(snappers) the primary conduits of
fish biomass and nutrients into and
out from coral reefs and mangroves
through nocturnal excursions?
(3) What are the dominant prey items
being transported into and out from
coral reefs and mangroves through
nocturnal excursions?
284
R. D. CLARK, ET AL.
METHODS
Study Area
Gillnet sampling was conducted dur-
ing nine surveys from June 2000-December
2002 across the insular shelf off La Parguera,
Puerto Rico ( Figure 1 ). Survey missions
were conducted every three to four months.
The shoreline and islands of the area are
lined with mangrove communities domi-
nated by the red mangrove Rhizophora man-
gle . Adjacent sediments support seagrasses
(dominated by Thalassia testudinum ), mac-
roalgae and unvegetated sand and sandy
mud interspersed with coral reefs and
patch reefs, which vary in habitat size and
benthic community composition. The tidal
range was relatively small (<0.5 m), and
mangrove prop roots at the seaward edge
of the mangrove stands were continually
immersed throughout the tidal cycle. Water
depths for sampled areas ranged from
3-20 m at Cr-Sg sites and 1-3 m at Mg-Sg
sites.
Sampling fish
A digital benthic habitat map (Kendall
et al. 2001) displaying only the major hab-
itat types of coral reefs, seagrasses and
mangroves was used to randomly select hab-
itat boundaries for sampling within the La
Parguera coral reef ecosystem. The habitat
map was used to stratify the study area into
two unique strata: (1) coral reef/seagrass
( Cr-Sg ) and (2) mangrove/seagrass ( Mg-Sg ).
During each survey five to eight gillnets
were set at each boundary. Gillnets were
100 m long and had 5 x 5 cm nylon mesh size
and were deployed along the Cr-Sg habitat
boundary running parallel to the reef edge
and along the Mg-Sg habitat boundary run-
ning parallel to the mangrove edge. Nets
were set by boat at dusk and retrieved at
Fig. 1. Location of study area, dominant benthic habitat types using NOAA’s benthic habitat map (Kendall
et al., 2002) and gillnet sampling sites.
285
NOCTURNAL CORAL REEF FISH MOVEMENT, SW PUERTO RICO
dawn, with soak times ranging from 12-14
hours.
During net retrieval, the orientation of the
fish in the net was recorded; if orientation
could not be decided the fish was not used
in subsequent analyses. Orientation was
used as a predictor of the direction of local
movements (Hall et al., 1979). For example,
at the Cr-Sg boundary, fish that entered the
net from the coral reef side were considered
to have been leaving the coral reef, while fish
on the other side of the net were considered
to have been entering the coral reef from
adjacent areas dominated by seagrasses.
All fish were stored on ice and promptly
transported to the laboratory where fish
were identified, weighed (wet weight) and
lengths measured. Standard length (SL) was
used to measure fishes, shark length was
measured from the tip of the snout to the
precaudal pit, and pectoral fin width was
measured for rays. Fishes were assigned to
one of four trophic guilds (herbivore, inverti-
vore, piscivore or zooplanktivore) based on
prey information from this research, and
where information was limited groups were
assigned based on Randall (1967) and other
sources obtained from FishBase (Froese and
Pauly 2009). Entire gastro-intestinal organs
were excised and stored in 10% formalin for
24 hours, then transferred to a 30% ethanol
solution. Stomach contents were removed,
wet weighed to the nearest .001 g, and
dietary items were identified to the low-
est possible taxon. Ultimately, prey items
were grouped into higher taxonomic cat-
egories, such as algae, crabs, shrimp, other
crustaceans (including isopods, ostracods,
or unidentifiable crustaceans), cephalopods,
echinoderms, fish, molluscs (including gas-
tropods and bivalves), and seagrass to facili-
tate statistical analyses. Prey item frequency
of occurrence in fish stomachs was used
to quantify the importance of prey items
to Cr-Sg and Mg-Sg fish communities.
Data analyses
Mean biomass, density, species rich-
ness, prey biomass and prey frequency of
ocurrence in fish stomachs was calculated
for each net site and segregated by direc-
tion of movement. Mean fish morphomet-
rics and gut content biomass (including
± 1 standard error) and frequency of occur-
rence was compared for each direction at
all sites and compared to determine net
transfer between habitat types. Gut content
biomass was examined by total fish move-
ment, the most abundant families, resident
species versus transient species, and for
taxa within the families Haemulidae and
Lutjanidae. Statistical analyses were con-
ducted using JMP software (SAS Institute,
2006). Differences for paired comparisons
between fish movemement and habitat
types and gut content directional flow were
examined using the Wilcoxon rank sums
nonparametric test, since the data could
not be transformed to satisfy ANOVA crite-
ria (Zar 1999). Multiple comparisons were
conducted using Kruskall-Wallis rank sums
test to detect significant differences in fish
morphometrics within and among the habi-
tats, flow of prey biomass among the taxo-
nomic groupings, followed by the Nemenyi
test to identify the significant factor; statis-
tics were considered significant at p<0.05.
Correspondence analysis (CA) using recip-
rocal averaging (Zar 1999) and represented
as a two-dimensional ordination plot was
used to visually compare the family compo-
sition of fish assemblages at Cr-Sg sites with
those at Mg-Sg sites.
Results
Overall, 57 gillnet deployments at Cr-Sg
and 64 at Mg-Sg sites were used for this
study. A total of 458 bony and cartilaginous
fishes representing 67 species and 33 fami-
lies were captured using gillnets in the La
Parguera study area ( Tables 1 and 2 ). Cr-Sg
sites yielded 267 individuals from 56 spe-
cies and 24 families exhibiting a total bio-
mass of 210.99 kg. Mean fish length
(standard length, SL), excluding cartilagi-
nous species was 24.2 cm (SE ± 0.59), with
the smallest fish measuring 9.4 cm and the
largest 73 cm. Seven individuals from three
species of cartilaginous fishes were cap-
tured, with a total biomass of 43.53 kg.
Mg-Sg sites yielded 191 individuals from 29
species and 21 families, with a total biomass
of 196.29 kg. Mean fish length (SL), exclud-
ing cartilaginous fishes, at Mg-Sg sites was
286
R. D. CLARK, ET AL.
Table 1 . Summary morphometrics for all species captured at Cr-Sg boundaries. Species names followed by –c
indicate cartilaginous fishes. Trophic groups: inv=invertivore, h=herbivore, p=piscivore, z=zooplanktivore.
Length is standard length (SL) for fishes; snout to precaudal pit length for sharks; and pectoral fin width for rays.
Species Family
Trophic
group Abundance
Length
min (cm)
Length
max (cm)
Biomass
min (kg)
Biomass
max (kg)
Acanthurus bahianus Acanthuridae h 2 16.4 17.9 0.142 0.241
Acanthurus chirurgus Acanthuridae h 6 16.4 22.2 0.195 0.395
Acanthurus coeruleus Acanthuridae h 13 14.3 18.3 0.17 0.35
Aetobatus narinari-c Myliobatidae inv 1 50.6 50.6 8.9 8.9
Albula vulpes Albulidae inv 1 27.2 27.2 0.345 0.345
Anisotremus virginicus Haemulidae inv 1 21.5 21.5 0.352 0.352
Calamus bajonado Sparidae inv 3 21.1 22.2 0.35 0.397
Calamus calamus Sparidae inv 20 17.5 35.1 0.196 1.634
Calamus penna Sparidae inv 9 19.7 22.8 0.3 0.446
Calamus pennatula Sparidae inv 3 19.5 22.4 0.276 0.365
Caranx bartholomaei Carangidae p 1 31.4 31.4 0.704 0.704
Caranx hippos Carangidae p 3 28.3 47.4 0.649 2.473
Caranx latus Carangidae p 7 26.1 54.8 0.499 3.592
Caranx ruber Carangidae p 12 18.7 62 0.166 5.525
Carcharhinus perezi-c Carcharhinidae p 1 74.8 74.8 2.245 2.245
Chaetodipterus faber Ephippidae inv 3 15.6 16.8 0.257 0.329
Chloroscomburs chrysurus Carangidae p 8 11.2 17.2 0.032 0.109
Dactylopterus volitans Dactylopteridae inv 1 23.3 23.3 0.254 0.254
Dasyatis americana-c Dasyatidae inv 5 43.3 65.8 3.958 9.45
Diodon holocanthus Tetraodontidae inv 3 25.5 42.5 0.746 2.921
Diodon hystrix Tetraodontidae inv 8 13.7 44.1 0.227 3.96
Echeneis naucrates Echeneidae inv 4 20.4 47.5 0.193 3.066
Gerres cinereus Gerreidae inv 8 21.7 26.9 0.341 0.543
Haemulon aurolineatum Haemulidae inv 3 12.5 13 0.036 0.055
Haemulon chrysargyreum Haemulidae inv 1 9.4 9.4 0.019 0.019
Haemulon flavolineatum Haemulidae inv 3 12.1 13.2 0.054 0.076
Haemulon parra Haemulidae inv 2 27.1 29.9 0.603 0.632
Haemulon plumieri Haemulidae inv 17 16 24.2 0.113 0.4
Haemulon sciurus Haemulidae inv 17 14.4 26.5 0.092 0.471
Haemulon striatum Haemulidae inv 2 11.3 13.5 0.035 0.069
Holocentrus adscensionis Holocentridae inv 3 10.2 13.6 0.047 0.322
Holocentrus marianus Holocentridae inv 1 16.2 16.2 0.086 0.086
Holocentrus rufus Holocentridae inv 2 15 15.7 0.078 0.08
Kyphosus sectatrix/incisor Kyphosidae h 23 22 32.4 0.417 1.179
Lachnolaimus maximus Labridae inv 2 20.2 22.3 0.342 0.436
Lactophrys trigonus Ostraciidae inv 1 32.5 32.5 0.898 0.898
Lutjanus analis Lutjanidae p 3 28 31.3 0.597 0.839
Lutjanus apodus Lutjanidae p 1 25.8 25.8 0.598 0.598
Lutjanus cyanopterus Lutjanidae p 1 29.2 29.2 0.66 0.66
Lutjanus griseus Lutjanidae p 8 28 34.4 0.556 0.974
Lutjanus jocu Lutjanidae p 5 21.8 29.8 0.317 0.912
Lutjanus synagris Lutjanidae p 1 16.8 16.8 0.148 0.148
Mulloidichthys martinicus Mullidae inv 2 15 22 0.093 0.268
Myripristis jacobus Holocentridae z 4 12.1 12.9 0.075 0.098
Ocyurus chrysurus Lutjanidae p 2 31.5 31.9 0.612 0.732
Pomacanthus arcuatus Pomacanthidae inv 4 10.5 19 0.094 0.528
Scarus guacamaia Scaridae h 1 25.9 25.9 0.664 0.664
Scarus vetula Scaridae h 2 29.8 29.9 0.912 1.007
Scomberomorus regalis Scombridae p 4 41.1 45.6 0.824 1.085
Scorpaena plumieri Scorpaenidae p 6 14.3 20.2 0.147 0.409
(Continued)
287
NOCTURNAL CORAL REEF FISH MOVEMENT, SW PUERTO RICO
Species Family
Trophic
group Abundance
Length
min (cm)
Length
max (cm)
Biomass
min (kg)
Biomass
max (kg)
Selar crumenophthalamus Carangidae z 9 17.1 18.5 0.103 0.143
Sparisoma chrysopterum Scaridae h 1 27.6 27.6 0.724 0.724
Sparisoma aurofrenatum Scaridae h 4 14 35.1 0.133 1.548
Sparisoma viride Scaridae h 2 24.3 27.5 0.542 0.649
Synodus intermedius Synodontidae p 5 24.2 28.4 0.21 0.318
Trachinotus falcatus Carangidae inv 2 63.4 73 9.5 11.6
Table 1. Continued.
Table 2 . Summary morphometrics for all species captured at Mg-Sg boundaries. Species names followed by
–c indicate cartilaginous fishes. Trophic groups: inv=invertivore, h=herbivore, p=piscivore, z=zooplanktivore.
Length is standard length (SL) for fishes; snout to precaudal pit length for sharks; and pectoral fin width for rays.
Species Family
Trophic
group Abundance
Length
min (cm)
Length
max (cm)
Biomass
min (kg)
Biomass
max (kg)
Aetobatus narinari-c Myliobatidae inv 2 42.1 44.5 0.94 1.157
Archosargus rhomboidalis Sparidae h 36 16.6 21.3 0.197 0.319
Bothus lunatus Bothidae inv 3 13.7 17.5 0.059 0.124
Caranx hippos Carangidae p 1 55.6 55.6 3.752 3.752
Caranx ruber Carangidae p 1 33.5 33.5 0.781 0.781
Centropomus undecimalis Centropomidae p 4 39.1 43 0.651 1.323
Chaetodipterus faber Ephippidae inv 7 13.9 30.4 0.187 1.852
Dasyatis americana-c Dasyatidae inv 10 45.1 66.3 3.921 12
Diodon holocanthus Tetraodontidae inv 5 22.5 35.2 0.548 1.97
Diodon hystrix Tetraodontidae inv 14 22.5 41.6 0.556 3.052
Gerres cinereus Gerreidae inv 11 20.3 25.9 0.249 0.589
Ginglymostoma cirratum-c Rhincodontidae p 1 130 130 10 10
Gymnothorax funebris Muraenidae p 1 96.7 96.7 1.75 1.75
Haemulon sciurus Haemulidae inv 24 10.6 21.6 0.038 0.267
Lactophrys trigonus Ostraciidae inv 2 17.4 18.6 0.196 1.593
Lutjanus analis Lutjanidae p 3 28.4 29.5 0.25 0.693
Lutjanus apodus Lutjanidae p 1 17.7 17.7 0.177 0.177
Lutjanus cyanopterus Lutjanidae p 2 44.2 49.6 2.551 3.512
Lutjanus griseus Lutjanidae p 27 16.9 41.3 0.147 1.754
Lutjanus jocu Lutjanidae p 13 21.5 35.2 0.307 1.16
Megalops atlanticus Elopidae p 5 37.9 44.9 0.876 1.397
Mugil curema Mugilidae inv 1 28.9 28.9 0.508 0.508
Peprilus alepidotus Stromateidae inv 1 14.2 14.2 0.165 0.165
Scorpaena plumieri Scorpaenidae p 5 12.9 20.3 0.111 0.424
Selene vomer Carangidae z 2 12.8 21.1 0.069 0.325
Sphyraena barracuda Sphyraenidae p 2 25.5 30.9 0.147 0.251
Synodus intermedius Synodontidae p 2 16.9 21 0.075 0.119
Trachinotus falcatus Carangidae inv 2 35.9 64 1.697 8.1
Trinectes maculatus Soleidae inv 3 14.2 18 0.149 0.302
24.54 cm (SE ± 0.81), where the smallest fish
was 10.6 cm and the largest was 64.0 cm.
Thirteen individuals from three species of
cartilaginous species were captured at
Mg-Sg with a total biomass of 77.82 kg.
Abundance and species richness were sig-
nificantly greater at Cr-Sg sites (p=0.0014
and p=0.0005, respectively) than at Mg-Sg
sites. Mean fish biomass did not differ
between the two habitats.
Invertivorous fish were the most abun-
dant trophic group, with highest biomass
(p<0.0001) at both Cr-Sg and Mg-Sg sites
( Figure 2a,b ). Piscivores were significantly
(p<0.0001) more abundant and with greater
biomass than herbivores only at Mg-Sg sites.
288
R. D. CLARK, ET AL.
Zooplanktivores were comparatively rare
in gillnet samples, exhibiting lower abun-
dance and biomass than all other trophic
groups.
Fish assemblage composition was signifi-
cantly different (c
2
<0.0001) between Cr-Sg
and Mg-Sg sites ( Figure 3 ). Although over-
lap among families occurred frequently
between the two habitat strata, actual spe-
cies composition often differed. For exam-
ple, Sparidae were abundant at both Cr-Sg
and Mg-Sg boundaries, but Archosargus spp.
were more prevalent at Mg-Sg sites, while
Calamus spp. were more prevalent at Cr-Sg
sites ( Tables 1 and 2 ). Lutjanidae, Sparidae,
and Haemulidae comprised 55% of the total
abundance at Mg-Sg sites, while at Cr-Sg
sites Haemulidae, Carangidae, Sparidae,
Kyphosidae and Lutjanidae comprised 62%
of the total abundance ( Table 3 ). Biomass
was mainly comprised of individuals from
Dasyatidae, Lutjanidae, and Tetraodontidae
accounting for 67% of total fish biomass at
Mg-Sg sites, while Carangidae, Dasyatidae,
Tetraodontidae, and Kyphosidae yielded
the most biomass (60% of fish biomass) at
Cr-Sg sites.
Nocturnal fish movements in and out of
mangroves and coral reefs
Fish assemblage. — Trophic flow, or the
flow of fish species, their biomass, and
ingested prey biomass into or out of a par-
ticular habitat type, was greatest at Cr-Sg
sites ( Figure 4 ). Mean fish density, biomass
and species richness were greater leaving
than entering Cr habitats at night, yet only
fish species richness was statistically signif-
icant (p=0.0018). Mean prey biomass enter-
ing Cr habitats at night (26.8, SD=72.25) was
almost three times higher than that leaving
Cr (10.2, SD=16.47) resulting in a mean net
transfer of prey biomass of 0.016 kg from
Sg to Cr ; however this relationship was not
statistically significant. Mean fish abun-
dance, biomass, and species richness were
greater leaving Mg habitats at night than
entering Mg, although none were statisti-
cally significant. Furthermore, mean prey
biomass was not statistically greater enter-
ing (mean=4.79, SD=11.22) or leaving Mg
(mean=11.36, SD=27.42) but was almost
three times greater leaving Mg ( Figure 4 ).
Total prey biomass was comparable for fish
leaving both Cr and Mg habitats; however,
prey biomass transport (not statistically sig-
nificant) into Cr was five times greater than
that observed at Mg .
Fish trophic groups .— Examining prey
biomass by trophic group revealed that
invertivores were the primary transport-
ers of prey biomass ( Figure 5 ). Invertivores
accounted for 48% of prey biomass enter-
ing Cr habitats at night and 41% leaving Cr .
Invertivores accounted for 35% of prey bio-
mass entering Mg and 81% leaving Mg at
night. Herbivores were the second highest
source of incoming prey biomass from fish
averaging 36% of the prey biomass entering
Fig. 2. Mean bidirectional (a) abundance and
(b) biomass (± 1SE) for fish trophic guilds captured at
Cr-Sg boundaries and Mg-Sg boundaries. h=herbivore,
inv=invertivore, p=piscivores, z=zooplanktivore.
289
NOCTURNAL CORAL REEF FISH MOVEMENT, SW PUERTO RICO
Cr , but exhibited high variability. Piscivores
accounted for 20% of the total prey biomass
entering Cr at night and 12% of the prey bio-
mass leaving Cr s. The prey biomass within
piscivores entering Mg from adjacent areas
at night was 37% of the total and piscivore
prey accounted for 11% of the total prey
leaving Mg .
Fish families. — Of the 24 families cap-
tured at Cr-Sg , prey transport was pri-
marily accomplished by eight families
( Figure 6a ). Approximately 70% of prey bio-
mass entering Cr habitats was transported
by Kyphosidae and Carangidae, while
only 4% was transported by Sparidae and
Lutjanidae. Fish from Kyphosidae were the
predominant movers of prey biomass out
of Cr habitats (33%) followed by Lutjanidae
(11%), Sparidae (10%) and Carangidae
(7%). Seven of twenty one families trans-
ported the majority of prey biomass across
Mg-Sg boundaries ( Figure 6b ). Sparidae,
Lutjanidae, and Dasyatidae accounted for
79% of the total prey biomass imported
into Mg ( Figure 6b ). Fish from the families
Dasyatidae, Carangidae, Tetraodontidae,
and Sparidae accounted for 80% of total
prey biomass leaving Mg at night.
The families were further pooled to rep-
resent resident and transient species. As a
whole, resident families transported 62.7%
of bidirectional prey biomass at Cr-Sg . For
resident species, transfer of prey biomass
( Figure 7a ) was greater from Sg to Cr (0.014
kg ± 0.006) compared to Cr-Sg (0.008 kg ±
0.001). Transient species displayed a greater
prey biomass net transfer from Sg to Cr
(0.012 ± 0.007) compared to Cr to Sg (0.0016
± 0.0012). Prey biomass transfer was not
statistically significant for resident or tran-
sient directional flow. Only 53% of prey
biomass was transported by resident spe-
cies at Mg-Sg. Resident families prey bio-
mass ( Figure 7b ) was greater from Mg to
Sg (0.005 ± 0.0011) in comparison to Sg to
Mg (0.003 ± 0.0009) and was not statistically
significant. Conversely, transient fami-
lies transferred significantly greater prey
biomass (p<0.0001) from Mg to Sg (0.006
± 0.002) compared with Sg to Mg (0.0012 ±
0.0011).
Fig. 3. Correspondence analysis (CA) ordination plot displaying fish family and trophic group composition for
all gillnet samples at Cr-Sg and Mg-Sg boundaries.
290
R. D. CLARK, ET AL.
Table 3 . Total abundance and biomass for fish families captured at Cr-Sg and Mg-Sg boundaries.
Reef Mangrove
Family Abundance Biomass (kg) Abundance Biomass (kg)
Haemulidae 46 12.38 24 2.11
Carangidae 42 57.46 6 14.72
Sparidae 35 14.59 36 8.93
Kyphosidae 23 16.39
Lutjanidae 21 14.08 46 40.53
Acanthuridae 21 5.29
Tetraodontidae 11 21.07 19 25.69
Scaridae 10 6.41
Holocentridae 10 1.00
Scorpaenidae 6 1.60 5 1.28
Dasyatidae 5 32.39 10 65.73
Synodontidae 5 1.29 2 0.19
Echeneidae 4 4.28
Scombridae 4 3.64
Pomacanthidae 4 1.06
Labridae 2 0.78
Mullidae 2 0.36
Myliobatidae 1 8.90 2 2.10
Ostraciidae 1 0.90 2 1.79
Carcharhinidae 1 2.25
Albulidae 1 0.35
Dactylopteridae 1 0.25
Elopidae 5 5.84
Centropomidae 4 4.25
Muraenidae 1 1.75
Soleidae 3 0.69
Mugilidae 1 0.51
Sphyraenidae 2 0.40
Bothidae 3 0.26
Stromateidae 1 0.17
Fig. 4. Transfer of mean fish density, biomass, species richness and prey biomass between Cr-Sg and Mg-Sg bound-
aries. Arrows indicate the direction in which fish were travelling when captured. Broken line represents the gillnet.
Importance of grunts and snappers in the
nocturnal trophic flux in coral reef ecosystems
Lutjanidae. — As a family, snappers were
captured in 25% of Cr-Sg gillnet sets and
51% of Mg-Sg . Of the total fish captured,
14.6% were snappers (67 individuals from
seven species) ( Tables 1 and 2 ). Lutjanus
griseus (gray snapper) and L. jocu (dog
snapper) were the most abundant lutjanid
species at both the Cr-Sg boundary and the
291
NOCTURNAL CORAL REEF FISH MOVEMENT, SW PUERTO RICO
Mg-Sg boundary, but overall lutjanids were
more abundant at Mg-Sg sites. Mean size
and biomass of L. griseus at Cr -30.6 cm SL
(± 0.65), 0.76 kg (± 0.04), and at Mg -31.6 cm
SL (± 1.1), 0.89 kg (± 0.07) ( Cr -30.6 cm SL,
0.76 kg; Mg -31.6 cm SL, 0.89 kg) were not
significantly different between L. jocu size
and biomass at Cr -25.6 cm SL (±1.3), 0.59
kg (±0.09) or Mg -27.8 cm SL (±1.2), 0.64 kg
(±0.06). Snapper movements resulted in a
net transfer of 0.36 kg (± 0.27 kg) biomass
from Cr to Sg and a net transfer of 0.29 kg
(± 0.21 kg) from Mg to Sg . Total snapper prey
biomass flow exhibited similar patterns
where net transfer from Cr to Sg of 0.001 kg
(± 0.003 kg) was observed and net transfer
of 0.0007 kg (± 0.0015 kg) from Mg to Sg . Of
the most abundant snappers, L. jocu exhib-
ited net transfers of prey biomass from Sg
to Cr and Sg to Mg , while L. griseus exhib-
ited net transfer of biomass from Cr and
Mg to Sg ( Figure 8a ). Snappers transported
5% of total prey biomass at Cr-Sg sites and
14% at Mg-Sg. Fish comprised the majority
of snapper prey biomass at both Cr-Sg and
Mg-Sg sites ( Figure 9a ) and represent the
essential prey item for the family as a whole
( Table 4 ). Fish were the principal dietary
item for L. griseus and L. jocu at Cr-Sg, while
crabs appeared to be the primary prey item
for L. analis. Crab frequency and biomass
was greater for snappers at Mg-Sg; how-
ever, fish were the most important for the
family as a group and for L. griseus . Crabs
were the primary prey item for
L. jocu at
Mg-Sg. Shrimp were a minor item for snap-
pers at Cr-Sg, but increased considerably
for snappers at Mg-Sg ( Table 4 ).
Fig. 5. Mean prey biomass (± 1SE) transported by fish trophic groups between Cr-Sg and Mg-Sg boundaries.
Negative values indicate movement into Sg , positive values indicate movement into Cr or Mg .
292
R. D. CLARK, ET AL.
Fig. 6. Mean prey biomass (± 1SE) grouped by family for fish moving in and out of: (a) Cr-Sg , and (b) Mg-Sg .
Negative values indicate movement into Sg , positive values indicate movement into Cr or Mg .
293
NOCTURNAL CORAL REEF FISH MOVEMENT, SW PUERTO RICO
Haemulidae. — Grunts were captured at
47% of Cr-Sg sites and 28% of Mg-Sg sites.
Grunts accounted for 15.3% of all fish cap-
tured (70 individuals from eight species).
Eight species of grunt were captured at the
Cr-Sg boundary, while only one species,
Haemulon sciurus (bluestriped grunt) was
captured at Mg-Sg sites ( Tables 1 , 2 ). Mean
H. sciurus size and biomass was significantly
greater (p<0.0001) at Cr-Sg sites SL=20.6 cm
(± 0.89 cm), biomass=0.28 kg (± 0.31 kg)
than Mg-Sg sites SL=13.9 cm (± 0.56 cm),
biomass=0.08 kg (± 0.11 kg). Grunt move-
ments resulted in a net transfer of 0.05 kg
(± 0.07 kg) biomass from Cr to Sg while net
transfer of 0.01 kg (± 0.03 kg) biomass from
Sg to Mg was observed. Overall, grunts con-
tributed a net transfer of prey biomass from
Cr to Sg (0.0005 kg, ± 0.0007 kg) and was
demonstrated by each species except for
H. aurolineatum ( Figure 8b ). H. sciurus was the
only grunt species exhibiting a net transfer
of prey biomass, 0.00002 kg (± 0.0001 kg),
from Mg to Sg . Grunts accounted for 1.6%
of prey biomass movement at Cr-Sg sites
and 0.8% of prey biomass movement at
Mg-Sg sites.
In terms of biomass, crabs, echinoderms,
other crustacea, polychaetes, and shrimp
were the most common prey items in grunt
stomach contents. Crabs and echinoderms
were the principal prey items at Cr-Sg sites,
while shrimp, other crustacea, and mol-
luscs were primary items at Mg-Sg sites
( Figure 9b ). Most prey items exhibited net
transfer from Cr to Sg except echinoderms,
seagrass, and other crustacea. Crabs, algae,
molluscs, and shrimp exhibited net trans-
fer of biomass from Sg to Mg while sea-
grass, other crustacea, polychaetes, and
fish exhibited net transfer from Mg to Sg .
Echinoderms were not a prey item for
grunts at Mg-Sg . Despite relatively high
echinoderm biomass at Cr-Sg ( Figure 9b ),
Fig. 7. Mean prey biomass (± 1SE) transported by resident and transient species between Cr-Sg and Mg-Sg
boundaries. Negative values indicate movement into Sg , positive values indicate movement into Cr or Mg .
294
R. D. CLARK, ET AL.
Fig. 8. Mean net flow of consumed prey biomass for: (a) snappers and (b) grunts. Negative values indicate
movement into Sg , positive values indicate movement into Cr or Mg .
295
NOCTURNAL CORAL REEF FISH MOVEMENT, SW PUERTO RICO
Fig. 9. (a) Snapper mean prey item biomass (± 1SE) flow at Cr-Sg and Mg-Sg boundaries. (b) Grunt mean prey
item biomass (± 1SE) flow at Cr-Sg and Mg-Sg boundaries. Negative values indicate movement into Sg , positive
values indicate movement into Cr or Mg .
296
R. D. CLARK, ET AL.
frequency of occurrence was low and was
considered less important for grunt prey
items ( Table 4 ). Other crustaceans and crabs
and were the prevalent dietary source for
grunts at Cr-Sg . Crabs were most impor-
tant for bluestriped grunts ( H. sciurus ) with
lesser importance attributed to other crus-
taceans, while other grunt species exhibited
similar frequencies between crabs and other
crustaceans. Crabs were less prominent
in grunt stomach contents at Mg-Sg while
other crustaceans and molluscs were more
significant dietary items compared to grunt
stomach contents at Cr-Sg .
Prey types transported by fish across coral
reef ecosystem boundaries. — A total of 458 fish
stomachs were examined ( Cr-Sg n=266, Mg-
Sg n=192). Twenty percent of all stomachs
were empty at each boundary. Crabs, algae,
echinoderms, and fish exhibited the great-
est prey biomass at Mg-Sg sites representing
60% of total biomass. For most prey types,
biomass exported from Mg was greater than
biomass imported into Mg at night ( Figure 10 );
only polychaetes and seagrass plant material
exhibited net gains for biomass. Algae, echi-
noderm and fish biomass composed 70%
of total prey biomass at Cr-Sg . Nearly all
prey items exhibited net gains of biomass at
Cr-Sg ; other crustacea, mollusc, and shrimp
biomass exhibited net losses.
Crabs and molluscs ranked first or sec-
ond in frequency of occurrence among prey
movement at Cr - Sg and Mg - Sg ( Table 5 ).
Crabs were the most frequently observed
prey at Mg and were observed in over 50%
of the fish captured leaving Mg. Crabs were
the most frequent prey item in fish leaving
Cr while molluscs were observed slightly
more frequently than crabs in fish entering
Cr . Fish and algae ranked third at Mg-Sg
while algae were more frequently observed
than fish at Cr-Sg . Fish ranked fourth and
third for prey importance at Cr-Sg and
Mg-Sg , respectively.
Discussion
Many tropical fish species conduct migra-
tions of varying spatial and temporal scales
(daily, ontogenetic) across a mosaic of habi-
tat types within a seascape, yet little is
known of the regular crepuscular and noc-
turnal patterns of movement that take place
in fish communities across habitat boundar-
ies. This paper measures several compo-
nents of functional connectivity, especially
as it pertains to energy transfer through the
flux of living biomass across the seascape
and the type and quantity of associated bio-
mass in the form of ingested prey. Most
trophic flow studies have focused on indi-
vidual taxa, thus this study represents a
Table 4 . Frequency of occurrence of prey items in snapper and grunt stomach contents at Cr-Sg and Mg-Sg
boundaries. Prey items include al-algae, cr-crab, oc-other crustaceans, ec-echinoderms, fi-fish, mo-molluscs,
po-polychaetes, se-seagrass, sh-shrimp.
Nal cr oc ec fi mo po se sh
Cr-Sg
All snappers 21 4.76 33.33 14.29 57.14 4.76 19.05 14.29
L. analis 3 133.33 33.33 66.67 33.33
L. griseus 8 12.50 12.50 12.50 62.50 12.50
L. jocu 5 18.89 0.02 79.59 1.48
All grunts 46 8.51 46.81 51.06 19.15 8.51 29.79 27.66 6.38 21.28
H. plumieri 17 5.56 38.89 33.33 22.22 11.11 27.78 22.22 16.67
H. sciurus 17 17.65 70.59 64.71 23.53 5.88 35.29 35.29 11.76 23.53
H. flavolineatum 3 33.33 66.67
H. aurolineatum 3 33.33 66.67 33.33
Mg-Sg
All snappers 46 2.17 39.13 4.35 43.48 17.39
L. griseus 27 3.70 25.93 3.70 44.44 22.22
L. jocu 13 76.92 46.15 15.38
All grunts H. sciurus 24 8.33 37.50 66.67 4.17 50.00 25.00 4.17 29.17
297
NOCTURNAL CORAL REEF FISH MOVEMENT, SW PUERTO RICO
unique community-wide perspective of fish
movement, their prey and associated bio-
mass in a coral reef ecosystem. Trophic flow
information is vital to understanding eco-
system function (Christensen and Pauly
2004) and is considered the key method that
determines (a) ecosystem state and (b) the
connectivity of biomass between ecological
groups or trophic levels (Monaco and
Ulanowicz 1997; Gascuel et al. 2008).
In the present study, we detected the
movements and prey transport of 67 fish
species representing four trophic groups.
Our results support previous observations
that many fish species use coral reefs and
mangroves as refuge during the day and
venture out to adjacent habitat types (sea-
grass, algal beds, unconsolidated sedi-
ments) to feed during low light levels
(Ogden and Ehrlich, 1977; Birkeland, 1985;
Parrish et., 1989). However, our results do
not support that prey biomass flow into
resting habitats are significantly greater
than that leaving. Mean flow at Cr-Sg
habitats followed the pattern suggested by
earlier trophic studies (Ogden and Erhlich
1977; Ogden and Zieman 1977; McFarland
et al. 1979) with net flow of prey biomass
greater for fish returning to resting habitats
after nocturnal foraging, however, flow was
not statistically significant which suggests
that the net transfer of prey biomass
between reef and seagrass is essentially
zero. Whether this is an artifact of the sam-
pling design and error associated with dif-
ferential digestive rates needs to be further
investigated. It is possible that there is a
dependence on habitat quality, such as reef
size or reef and seagrass benthic composi-
tion that may have influenced these
patterns.
Mangroves have been considered feeding
areas for fishes that reside on adjacent mud
flats and seagrass beds (Sheaves and
Molony 2000); however, Nagelkerken and
van der Velde (2004a) claim that mangroves
are less important for fish residing in adja-
cent habitats than for fish that reside in
mangroves. This study indicated that man-
grove habitats serve as a feeding area (or
Fig. 10. Mean prey item biomass (± 1SE) for all fishes captured at Cr-Sg and Mg-Sg boundaries. Negative val-
ues indicate movement into Sg , positive values indicate movement into Cr or Mg .)
298
R. D. CLARK, ET AL.
the habitats are frequently visited) for a
wide range of multi-habitat fish species
including sharks, jacks, rays, and snappers.
Species level prey trophic flow may also be
influenced by adult/juvenile migration pat-
terns and diurnal feeding within the man-
groves (Nagelkerken and van de Velde
2004b). Thus, the flow at mangroves is com-
plex, where residents may feed within the
mangroves and import prey biomass from
adjacent seagrass habitats while other fishes
extract biomass (Nagelkerken and van der
Velde 2004a, b).
Using results from this study we estimate
47.6 individuals/km (SD ± 36.4) migrating
between coral reefs and adjacent seagrasses.
This estimate includes 3.75 snappers/km
(SD ± 7.2), 8.21 grunts/km (SD ± 12.8), and
7.8 pelagic/non-residents/km (SD ± 13.1).
Fish movement at the Mg-Sg boundary
yielded 29.8 individuals/km (SD ± 18.2),
which include 7.1 snappers/km (SD ± 8.2),
3.7 grunts/km (SD ± 7.2), and 3.4 pelagic/
non-resident individuals/km (SD ± 5.6).
These estimates are likely underestimates of
the entire communities that reside in these
habitats due to the selectivity of the sam-
pling gear and further studies should target
smaller species. Similarly, prey biomass is
likely underestimated due to differential
digestive rates, and the duration of net soak
times. Therefore, our estimates of prey bio-
mass flow exhibiting net transfer of 0.016
kg/km from Sg to Cr and 0.006 kg/km from
Mg to Sg are likely underestimated, but
combined with prey frequency of occur-
rence provide insight into relative transfer
of prey and their biomass. Prey item fre-
quency of occurrence for the fish communi-
ties sampled highlight the importance of
crabs, molluscs, and algae being trans-
ported among the habitats.
We observed net transfer of prey biomass
from reef to seagrass for L. griseus, L. syn-
agris, L. cyanopterus , and all but one grunt
species. Our results primarily represent
subadults and adults and thus do not
include the many juveniles that are also
likely to move across patch boundaries to
forage at night. More intensive field work
with smaller gillnet mesh size supple-
mented by acoustic tracking surveys and
gut content analysis could be integrated to
gain more direct evidence of species trophic
flows across a broader size range. Acoustic
surveys have shown differences between
adult and juvenile migrations, where adults
were more active on the resting reef and
off-reef migrations were not as structured
as that seen for juveniles (Tulevich and
Recksiek 1994; Beets et al. 2003). In fact,
Tulevich and Recksiek (1994) observed
adult white grunts ( H. plumieri) roaming
around reef habitats during the day and did
not move off the reef at night. At present,
the question of migration timing is not well
understood, and patterns could vary by
species, by size, and by habitat arrangement
(size, relief, etc). Manual acoustic tracking
of individual H. sciurus and L. apodus in
St. John (U.S. Virgin Islands) has shown sev-
eral individuals exhibiting high site fidelity
Table 5 . Directional frequency of occurrence and rank (in parentheses) of prey items for all fish captured at Cr-
Sg and Mg-Sg boundaries. Prey items include al-algae, cr-crab, oc-other crustaceans, ec-echinoderms, fi-fish, mo-
molluscs, po-polychaetes, se-seagrass, sh-shrimp.
Sg-Cr Cr-Sg Sg-Mg Mg-Sg
Total fish catch 123 143 89 103
crab 27.64 (2) 33.57 (1) 30.34 (1) 54.81 (1)
mollusc 28.45 (1) 25.87 (2) 23.60 (2) 39.42 (2)
algae 21.95 (3) 20.28 (4) 19.10 (3) 16.35 (3)
fish 17.88 (4) 13.29 (5) 19.10 (3) 16.35 (3)
other crustacean 17.88 (4) 23.08 (3) 14.61 (5) 15.38 (5)
shrimp 6.50 (8) 13.29 (5) 12.36 (6) 12.50 (6)
seagrass 7.32 (7) 4.20 (9) 3.37 (8) 8.65 (7)
polychaete 7.32 (6) 9.09 (7) 8.99 (7) 6.73 (8)
echinoderm 4.88 (9) 6.99 (8) 0 (9) 3.85 (9)
299
NOCTURNAL CORAL REEF FISH MOVEMENT, SW PUERTO RICO
to a specific location on a coral reef during
both day and night (S. Hitt and S.J. Pittman,
unpublished data). While our results were
observations collected over a shorter dura-
tion compared to acoustic surveys, results
cannot address migration consistency. Most
of the clear patterns have been observed
with grunts, and may not be a common
representation for reef and mangrove
communities.
Many factors work synergistically to
determine fish community structure within
a given reef or mangrove site, such as the
composition and spatial configuration of the
seascape as a whole, as well as the within-
habitat structure of coral reefs, mangroves
and seagrasses (Kendall et al. 2003; Pittman
et al. 2004; Pittman et al. 2007a) and the
bathymetric complexity of the surrounding
area (Pittman et al. 2007b). These data rep-
resent a unique insight into multi-species
and multi-habitat fish movement across
patch boundaries within the region and the
first estimates of trophic flow among the
mosaic of habitat types. Additionally, fish
movements can vary by taxa and life stage,
therefore the results provided here need
to be complemented with information on
the abundance, movements and nutrient
transport patterns of smaller fishes to pro-
vide a better estimate of community-wide
energy flow.
The community structure at both Cr-Sg
and Mg-Sg sites was strongly influenced
by transient pelagic species, such as jacks,
sharks, and rays. Their abundance was
generally low, however their large size
accounted for almost 50% of the total bio-
mass. When transient species were excluded
from estimates of trophic flow at Cr-Sg sites,
net gains of prey biomass declined by 63%.
Similarly, exclusion of non-residents at
mangrove sites reduced net prey biomass
flow by 75%. Our results indicate that pre-
dation (top-down processes) on coral reefs
was a significant process however our
results indicate that coral reefs and man-
groves are likely to function as both a sink
and a source for nutrients transported by
fishes. Predation by carnivorous jacks and
sharks play an important role in structur-
ing reef fish communities (Hixon 1991)
and also for fish using mangroves (Ogden
and Gladfelter 1983). The obvious signifi-
cance of these results is the top-down pro-
cess of predation, removal of biomass from
habitats. Pelagic species in coral reef eco-
systems can have home ranges from 1-5
km (Honebrink 2000; Cartamil et al. 2003),
thus habitat connectivity could have a large
spatial context connecting multiple habi-
tat types through frequent transboundary
movements.
Invertivores were the dominant trophic
group among both habitat types. Jones et
al. (1991) summarized that benthic inver-
tebrate predators were the most speciose
among all trophic groups within coal reef
systems. In concordance with this observa-
tion, invertebrate prey biomass was the pri-
mary source of trophic flow at all habitat
boundaries. Seagrass and mangrove habi-
tats generally produce an abundant inver-
tebrate fauna (Weinstein and Heck 1977;
Wahbeh 1982) and we observed a wide
diversity of invertebrate organisms among
the stomach contents. Parrish et al. (1985)
noted that nearly 75% of fish captured
in the Northwest Hawaiian Islands con-
sumed crustaceans, predominantly crabs
and shrimp. In this study, crabs were dom-
inant prey types in terms of total biomass
and frequency of occurrence. Herbivores
usually comprise the majority of fish bio-
mass on reefs and are important for bio-
mass turnover from primary producers to
higher trophic levels (Ferreira et al. 2004).
In this study, herbivores were considerably
less abundant than invertivores and pisci-
vores combined at both reef and mangrove
sites. Zooplanktivores were rare at both reef
and mangrove sites and this group is prob-
ably not efficiently sampled by gillnets due
to their schooling behavior above reefs and
smaller size.
Grunts and snappers were abundant in
gillnets, and nocturnal migrations to sea-
grass beds are well documented (Pollard
1984; Nagelkerken et al. 2000). Snappers
were significant transporters of prey mate-
rial in and out of coral reefs and mangroves
as both consumers and providers of energy.
Grunts were less significant contributors,
accounting for only 1-2% of total prey bio-
mass. Overall snappers contributed a net
loss of prey biomass at reef sites. L. jocu and
300
R. D. CLARK, ET AL.
L. griseus were the most abundant snappers
and L. jocu contributed net gains of prey
biomass into reefs while L. griseus yielded
net loss of prey biomass. Starck and Davis
(1966) report that juvenile and sub-adult
gray snapper exhibit diurnal foraging into
adjacent seagrass habitats. Additionally,
diurnal feeding on the reef may also occur
where grunts and snappers may opportu-
nistically forage in sandy reef interstices
(Tulevech and Recksiek 1994).
Frequency of occurrence results indicated
that fish were the dominant prey type for
snappers at coral reef habitats, while fish
and crabs were equally important in snap-
per stomachs from the mangroves. Crabs,
echinoderms, and other crustaceans were
the most important prey items for grunts on
coral reefs, while molluscs, shrimp, and
other crustacea were most important for
grunts in the mangroves. While this may
indicate ontogenetic differences, L. jocu and
L. griseus size was not significantly different
between the two habitats. Thus, prey den-
sity-dependent factors probably influence
snapper diet at both sites. Individuals of
H. sciurus were significantly smaller in man-
groves and prey items were significantly
different at reefs versus mangroves, which
suggests ontogenetic shifts in habitat and
diet (See Clark et al. 2003). This pattern has
also been observed in southwest Puerto
Rico (Dennis 1992) and other regions of the
Caribbean (Mumby 2004; Cocheret de la
Moriniere et al. 2003).
Gillnets are highly selective to the size of
fish they capture, but generally unselective
to the suite of fishes captured (Hickford and
Schiel 2008). Gillnet surveys have not been
efficient at characterizing reef communities;
however, they are effective at capturing
transient pelagic species and migratory reef
species (Hickford and Schiel 2008), which
was the primary objective for this study.
There is little information about size selec-
tivity or capture efficiency in tropical coral
reef ecosytems (Acosta and Appeldoorn
1995). Acosta and Appeldoorn (1995) com-
pared capture efficiency between a variety
of gillnet types and mesh sizes at reef and
mangrove sites in the same study area.
Resulting finfish catch (excluding sharks
and rays) using gillnets with similar mesh
size (7.6 cm stretched mesh, compared to
our 5 cm) but lower sample size, yielded
similar catch at reef sites in terms of biomass
(2.8 kg/net) compared with the results pre-
sented here (2.9 kg/net). However, their
observed biomass at mangrove sites was
lower (1.1 kg/net) compared to this study
(1.8 kg/net).
Gillnets are inefficient sampling tools for
estimating fish density and size distribu-
tion, while other methods such as visual
surveys can underestimate pelagic and
cryptic species while also underestimating
size distribution. For example, during this
study mean fish size (FL) at reef and man-
grove habitats was 27.8 and 26.5 cm, respec-
tively. Visual survey data collected from
2000-2006, yielded mean fish size (FL) less
than 10 cm for both reef and mangrove hab-
itats (Biogeography Branch [ http://ccma.
nos.noaa.gov/about/biogeography ],
unpublished data.). As such, the selectivity
of the gear employed here is not representa-
tive of the community at large, but may be
sufficient to comprehensively estimate sub-
adult and adult movements in the ecosys-
tem. Gillnet selectivity precluded us from
making any estimates of fish movement and
energy transport for fishes generally less
than 10 cm. Smaller fishes on reefs (such as
grunts) are typically more abundant than
larger fishes and have been shown to pro-
vide beneficial nutrients to benthic inverte-
brates on their reef schooling sites (Meyer
and Schulz 1985). Additionally, the survey
methods did not adequately sample zoo-
planktivores which are generally abundant
in coral reef ecosystems.
Coral reef ecosystems are generally olig-
otrophic, thus the energy provided by fishes
could be a significant source of nutrient
enrichment for coral reefs. The transfer of
energy between habitats may enhance other
pathways of energy flow. For example, sec-
ondary production may be stimulated by
migrating organisms (Ogden and Gladfelter
1983). Coral reefs are considered among the
highest productive systems in the world due
to their ability to efficiently recycle nutrients
(Szmant 1983). Fishes that reside on reefs
deposit waste products onto the reef and
provide a substantial amount of organic
material that is important for corals and other
301
NOCTURNAL CORAL REEF FISH MOVEMENT, SW PUERTO RICO
sessile invertebrates (Szmant 1983; Meyer
and Schulz 1985). Top down processes, such
as piscivory by large predators, may act to
neutralize net gains of energy on reefs, but
our study indicates that, while pelagic spe-
cies do have an influence, the resident taxa
import more than what is extracted.
Conculsions
In this study, we examined movements
and diets of a multispecies assemblage that
provide insight into energy flow through
the ecosystem. We observed that energy
flow at night was multi-directional at both
Cr-Sg and Mg-Sg sites where foraging activ-
ity among adjacent habitat types was high.
Additionally, we infer that pelagic species
play a considerable role in the flow of
energy at both habitats and may connect
habitats at scales of kms. Snappers and
grunts combined accounted for approxi-
mately 7% of prey biomass transfer at Cr-Sg
and 15% at Mg-Sg . We suggest that future
energy flow research be conducted in con-
juction with tracking data to further
strengthen the understanding of connectiv-
ity in the region and provide support for
resource management where connections
are strongest. Furthermore, the broader
scale patterning of the seascape surround-
ing the patch boundaries sampled in this
study may explain some of the between site
variability in abundance, diversity and bio-
mass of fish thus adding a significant source
of unquantified variability (Pittman et al.
2007b; Grober-Dunsmore et al. 2009). Future
studies are needed that link fish movement
behavior, trophic ecology and landscape
ecology to comprehensively quantify the
patterns of trophic connectivity across trop-
ical marine seascapes.
Acknowledgements. — Funding for this
project was provided by NOAAs Center
for Sponsored Coastal Ocean Research.
Additional support was provided by
NOAAs Coral Reef Conservation Program
and the University of the Virgin Islands.
Sincere thanks to M. Prada and K. Foley for
assisting with logistics to support the field-
work and to all of R. Appeldoorn’s staff
for putting up with the variety of smells in
the lab. Gracious thanks to K. Buja, T. Gill,
M. Kendall, M. Nelson, and J. Waddell who
contributed significantly to the fieldwork.
Literature Cited
Alheit, J. 1981. Feeding interactions between coral reef
fishes and zoobenthos. In Proc. 4
th
Int. Coral Reef
Symp. ed. E. D. Gomez et al., 545-552, Manila.
Birkeland, C. 1985. Ecological interactions between
tropical coastal ecosystems. UNEP Regional Seas
Reports and Studies No. 73.
Bray, R. N., A. C. Miller, and G. G. Geese. 1981. The fish
connection: A trophic link between planktonic and
rocky reef communities. Science 214:204-205.
Cartamil, D. P., J. J. Vaudo, C. G. Lowe, B. M. Wetherbee,
and K. N. Holland. 2003. Diel movement patterns
of the Hawaiian stingray, Dasyatis lata : implications
for ecological interactions between sympatric elas-
mobranch species. Mar. Biol. 142:841-847.
Clark, R., M. E. Monaco, R. S. Appeldoorn, and B. Roque.
2005. Fish habitat utilization in a Puerto Rico coral reef
ecosystem. Proc. 56
th
Gulf Carib. Fish. Inst. 467-485.
Cocheret de la Morinière E., B. J. A. Pollux, I. Nagelkerken,
and G. van der Velde. 2002.Post-settlement life cycle
migration patterns and habitat preference of coral
reef fish that use seagrass and mangrove habitats as
nurserie. Est. Coast. Shelf Sci. 55:309-321.
Cocheret de la Moriniere, E., B. J. A. Pollux, I.
Nagelkerken, and G. Van der Velde. 2003. Diet
shifts of Caribbean grunts (Haemulidae) and snap-
pers (Lutjanidae) and the relation with nursery-to-
coral reef migrations. Est. Coast. Shelf Sci. 57:1-11.
Cortes, E. 1997. A critical rewiew of methods studying
fish feeding based on analysis of stomach contents:
application to elasmobranch fishes. Can. J. Fish.
Aquat. Sci. 54:726-738.
Dahlgren, C. P., and D. B. Eggleston. 2000. Ecological
processes underlying ontogenetic habitat shifts in a
coral reef fish. Ecology 81:2227-2240.
Dennis, G. D. 1992. Resource utilization by members
of a guild of benthic feeding coral reef fish. PhD
Dissertation. University of Puerto Rico, Mayaguez.
224pp.
Dorenbosch, M., M., G. G. Grol, M. J. A. Christiansen,
I. Nagelkerken, and G. van der Velde. 2005. Indo-
Pacific seagrass beds and mangroves contribute to
fish density and diversity on adjacent coral reefs.
Mar. Ecol. Prog. Ser. 302:63-76.
Dulvy, N. K., R. P. Freckleton, and N. V. C. Polunin. 2004.
Coral reef cascades and the indirect effects of preda-
tor removal by exploitation. Ecol. Lett. 7:410-416.
Ferreira, C., E. L., S. R. Floeter, J. L. Gasparini, B. P.
Ferreira, and J. C. Joyeux. 2004. Trophic structure
patterns of Brazilian reef fishes: a latitudinal com-
parison. J. Biogeogr. 31:1093-1106.
Gascuel, D., L. Morissette, M. L. D. Palomares, and
V. Christensen. 2008. Trophic flow kinetics in marine
ecosystems: Toward a theoretical approach to eco-
system functioning. Ecol. Model. 217:33-47.
302
R. D. CLARK, ET AL.
Grober-Dunsmore, R, S. J. Pittman, C. Caldow, M. A.
Kendall, and T. Fraser. 2009. A landscape ecology
approach for the study of ecological connectiv-
ity across tropical marine seascapes. In Ecological
Connectivity Among Tropical Coastal Ecosystems , ed.
I. Nagelkerken, 493-530. Springer, Netherlands.
Grol, M. G. G., M. Dorenbosch, E. M. G. Kokkelmans,
and I. Nagelkerken. 2008. Mangroves and seagrass
beds do not enhance growth of early juveniles of a
coral reef fish. Mar. Ecol. Prog. Ser. 366:137-146.
Hall, D. J., E. E. Werner, J. F. Gilliam, G. G. Mittelbach,
D. Howard, C. G. Doner, J. A. Dickerman, and
A. J. Stewart. 1979. Diel foraging behavior and prey
selection in the golden shiner ( Notemigonus cryso-
leucas ). J. Fish. Res. Board Can. 36 (9):1029-1039.
Helfman, G. S., J. L. Meyer, and W. N. McFarland.
1982. The ontogeny of twilight migration patterns
in grunts (Pisces:Haemulidae). Anim. Behav. 30:
317-326.
Hickford, M. J. H., and D. R. Schiel. 2008. Experimental
gill-netting of reef fish: Species-specific responses
modify capture probability across mesh sizes.
J. Exp. Mar. Biol. Ecol. 358:163-169.
Hixon, M. A. 1991. Predation as a process structuring
coral reef fish communities. In The Ecology of Fishes
on Coral Reefs. ed. P.F. Sale, 475-508. Academic Press,
Inc., San Diego, CA.
Honebrink, R. R. 2000. A review of the biology of the
family Carangidae, with emphasis on species found
in Hawaiin waters. Division of Aquatic Resources,
Dept. of Land & Natural Resources, State of Hawaii.
DAR Technical Report 20-01 .
Jones, G. P., D. J. Ferrell, and P. F. Sale. 1991. Fish pre-
dation and its impact on the invertebrates of coral
reefs and adjacent sediments. In The Ecology of
Fishes on Coral Reefs , ed. P.F. Sale, 156-179. Academic
Press, Inc., San Diego, CA.
Kendall, M. S., C. R. Kruer, K., R. Buja, J. D. Christensen,
M. Finkbeiner, R. Warner, and M. E. Monaco. 2002.
Methods used to map the benthic habitats of Puerto
Rico and the US Virgin Islands. NOAA Tech. Mem.
152 . Silver Spring, MD.
Kendall, M. S., J. D. Christensen, and Z. Hillis-Starr.
2003. Multi-scale data used to analyze the spatial
distribution of French grunts, Haemulon flavolinea-
tum , relative to hard and soft bottom in a benthic
landscape. Environ. Biol. Fish. 66:19-26.
Kramer, D. L., and M. R. Chapman. 1999. Implications
of fish home range size and relocation for marine
reserve function. Environ. Biol. Fish. 55:65-79.
Kopp, D., Y. Bouchon-Navaro, M. Louis, and C.
Bouchon. 2007. Diel differences in the seagrass fish
assemblages of a Caribbean island in relation to
adjacent habitat types. Aquat. Bot. 87:31-37.
Kulbicki, M., Y. M. Bozec, P. Labrosse, Y. Letourneur,
G. M. Tham, and L. Wantiez. 2005. Diet composi-
tion of carnivorous fishes from coral reef lagoons of
New Caledonia. Aquat. Living Resour. 18:231-250.
Meyer J. L., and E. T. Schultz. 1985. Migrating haemu-
lid fishes as a source of nutrients and organic mat-
ter on coral reefs. Limnol. Oceanogr. 30(1):146-156.
Meyer, J. L., E. T. Schultz, and G. S. Helfman. 1983.
Fish schools: An asset to corals. Science 220(4601):
1047-1049.
Monaco, M. E., and R. E. Ulanowicz. 1997. Comparative
ecosystem trophic structure of three U.S. mid-
Atlantic estuaries. Mar. Ecol. Prog. Ser. 161:239-254.
Monaco, M. E., A. Friedlander, S. D. Hile, S. J. Pittman,
and R. H. Boulon. 2009. The coupling of St. John,
US Virgin Islands marine protected areas based on
reef fish habitat affinities and movements across
management boundaries. Proc. 11
th
Int. Coral Reef.
Symp. Ft. Lauderdale, FL.
Mumby, P. J., A. J. Edwards, J. E. Arias-Gonzalez,
K. C. Lindeman, P. G. Blackwell, A. Gall, M. I.
Gorczynska, A. R. Harborne, C. L. Pescod, H.
Renken, C. C. C. Wabnitz, and G. Llewellyn. 2004.
Mangroves enhance the biomass of coral reef fish
communities in the Caribbean. Nature 427:533-536.
Murray, S. N., R. F. Ambrose, J. A. Bohnsack, L. W.
Botsford, M. H. Carr, G. E. Davis, P. K. Dayton,
D. Gotshall, D. R. Gunderson, M. A. Hixon, J.
Lubchenco, M. Mangel, A. MacCall, D. A. McArdle,
J. C. Ogden, J. Roughgarden, R. M. Starr, M. J.
Tegner, and M. M. Yoklavich. 1999. No-take reserve
networks: protection for fishery populations and
marine ecosystems. Fisheries 24(11):11-25.
Nagelkerken, I., M. Dorenbosch, W. Verberk, E. C. de la
Moriniere, and G. van der Velde. 2000. Importance
of shallow-water biotopes of a Caribbean bay for
juvenile coral reef fishes: Patterns in biotope asso-
ciation, community structure and spatial distribu-
tion. Mar. Ecol. Prog. Ser. 202:175-192.
Nagelkerken, I., and G. van der Velde. 2004a. Are
Caribbean mangroves important feeding grounds
for juvenile reef fish from adjacent seagrass beds?
Mar. Ecol. Prog. Ser. 274:143-151.
Nagelkerken, I., and G. van der Velde. 2004b. Relative
importance of interlinked mangroves and seagrass
beds as feeding habitats for juvenile reef fish on a
Caribbean island. Mar. Ecol. Prog. Ser. 274:153-159.
Ogden, J. C., and P. R. Ehrlich. 1977. The behaviour
of heterotypic resting schools of juvenile grunts
(Pomadasyidae). Mar. Biol. 42:273-280.
Ogden, J. C., and E. H. Gladfelter.1983. Coral Reefs,
Seagrass Beds, and Mangroves: Their Interaction
in the Coastal Zones of the Caribbean. UNESCO
Reports in Marine Science 23 , 133 pp.
Parrish, J. D. 1989. Fish communities of interacting
shallow-water habitats in tropical oceanic regions.
Mar. Ecol. Prog. Ser. 58:143-160.
Parrish, J. D., M. W. Callahan, and J. E. Norris. 1985.
Fish trophic relationships that structure reef com-
munities. Proc. 5
th
Int. Coral Reef Congr. 4:73-78.
Pittman, S. J., C. A. McAlpine, and K. M. Pittman. 2004.
Linking fish and prawns to their environment: a
hierarchical landscape approach. Mar. Ecol. Prog.
Ser. 283:233-254.
Pittman, S. J., J. D. Christensen, C. Caldow, C. Menza,
and M. E. Monaco. 2007a. Predictive mapping of
fish species richness across shallow-water sea-
scapes in the Caribbean. Ecol. Model. 204:9-21.
303
NOCTURNAL CORAL REEF FISH MOVEMENT, SW PUERTO RICO
Pittman, S. J., C. Caldow, S. D. Hile, and M. E. Monaco.
2007b. Using seascape types to explain the spatial
patterns of fish in the mangroves of SW Puerto
Rico. Mar. Ecol. Prog. Ser. 348:273-284.
Polis, G. A., W. B. Anderson, and R. D. Holt. 1997.
Toward an integration of landscape and food web
ecology: The dynamics of spatially subsidized food
webs. Ann. Ecol. Syst. 28:289-316.
Pollard, D. A. 1984. A review of ecological studies on
seagrass–fish communities, with particular refer-
ence to recent studies in Australia. Aquatic Botany
18:3-42.
Polunin, N. V. C., and C. M Roberts. 1993. Greater bio-
mass and value of target coral-reef fishes in two
small Caribbean marine reserves. Mar. Ecol. Prog.
Ser. 100:167-176.
Randall, J. E. 1967. Food habits of reef fishes of the
West Indies. Stud. Trop. Ocean. 5:665-847.
Rooker J. R., and G. D. Dennis. 1991. Diel, lunar and
seasonal changes in a mangrove fish assemblage
off southwestern Puerto Rico. Bull. Mar. Sci. 49:
684-698.
Sale, P. F. 2004. Connectivity, recruitment variation,
and the structure of reef fish communities. Integr.
Comp. Biol. 44:390-399.
SAS Institute. 2006. SAS/STAT
©
Software. SAS
Institute, Inc. Cary, NC. USA.
Sheaves, M. 2005. Nature and consequences of biologi-
cal connectivity in mangrove systems. Mar. Ecol.
Prog. Ser. 302:293-305.
Sheaves M., and B. Molony. 2000. Short-circuit in the
mangrovefood chain. Mar. Ecol. Prog. Ser. 199:97-109.
Szmant-Froelich, A. 1983. Functional aspects of nutri-
ent cycling on coral reefs. In The Ecology of Deep and
Shallow Coral Reefs , ed, M. L. Reaka, 133-139. NOAA
Undersea Research Program, Silver Spring, MD.
Unsworth, R. K. F., E. Wylie, D. J. Smith, and J. J. Bell.
2007. Diel trophic structuring of seagrass bed fish
assemblages in the Wakatobi Marine National Park,
Indonesia. Est. Coast. Shelf Sci. 72:81-88.
Unsworth, R. K. F., P. S. DeLeon, S. L. Garrard, J. Jompa,
D. J. Smith, and J. J. Bell. 2008. High connectivity of
Indo-Pacific seagrass fish assemblages with man-
grove and coral reef habitats. Mar. Eco. Prog. Ser.
353:213-224.
Wahbeh, M. I. 1982. Distribution, biomass, biometry
and some associated fauna of the seagrass commu-
nity in the Jordan Gulf of Aqaba. Proc. 4th Int. Coral
Reef Symp. 2:453-459.
Weinstein, M. P., and K. L. Heck. 1979. Ichthyofauna
of seagrass meadows along the Caribbean coast of
Panama and in the gulf of Mexico: composition,
structure and community ecology. Mar. Biol. 50:
97-107.
