Content uploaded by Adrian C Stier
Author content
All content in this area was uploaded by Adrian C Stier on Mar 25, 2014
Content may be subject to copyright.
REPORT
Predators alter community organization of coral reef cryptofauna
and reduce abundance of coral mutualists
A. C. Stier •M. Leray
Received: 14 July 2013 / Accepted: 19 August 2013 / Published online: 3 September 2013
!Springer-Verlag Berlin Heidelberg 2013
Abstract Coral reefs are the most diverse marine systems
in the world, yet our understanding of the processes that
maintain such extraordinary diversity remains limited and
taxonomically biased toward the most conspicuous species.
Cryptofauna that live deeply embedded within the inter-
stitial spaces of coral reefs make up the majority of reef
diversity, and many of these species provide important
protective services to their coral hosts. However, we know
very little about the processes governing the diversity and
composition of these less conspicuous but functionally
important species. Here, we experimentally quantify the
role of predation in driving the community organization of
small fishes and decapods that live embedded within Po-
cillopora eydouxi, a structurally complex, reef-building
coral found widely across the Indo-Pacific. We use surveys
to describe the natural distribution of predators, and then,
factorially manipulate two focal predator species to quan-
tify the independent and combined effects of predator
density and identity on P. eydouxi-dwelling cryptofauna.
Predators reduced abundance (34 %), species richness
(20 %), and modified species composition. Rarefaction
revealed that observed reductions in species richness were
primarily driven by changes in abundance. Additionally,
the two predator species uniquely affected the beta diver-
sity and composition of the prey assemblage. Predators
reduced the abundance and modified the composition of a
number of mutualist fishes and decapods, whose benefit to
the coral is known to be both diversity- and density-
dependent. We predict that the density and identity of
predators present within P. eydouxi may substantially alter
coral performance in the face of an increased frequency
and intensity of natural and anthropogenic stressors.
Keywords Predation !Decapods !Neocirrhites
armatus !Caracanthus maculatus !Trapezia !
Dascyllus
Introduction
An extraordinary diversity and biomass of invertebrates
live inconspicuously embedded within the matrix of coral
reefs (Grassle 1973; Brock and Brock 1977; Klumpp et al.
1988; Stella et al. 2011; Enochs and Manzello 2012). For
Communicated by Biology Editor Dr. Glenn Almany
A. C. Stier and M. Leray have equal contribution to this work.
Electronic supplementary material The online version of this
article (doi:10.1007/s00338-013-1077-2) contains supplementary
material, which is available to authorized users.
A. C. Stier
Department of Biology, University of Florida, 223 Bartram Hall,
PO Box 118525, Gainesville, FL 32611-8525, USA
Present Address:
A. C. Stier (&)
Department of Zoology, University of British Columbia,
#4200-6270 University Blvd., Vancouver, BC V6T 1Z4, Canada
e-mail: adrian.stier@gmail.com
M. Leray
Laboratoire d’Excellence ‘‘CORAIL’’, USR 3278 CNRS EPHE,
CRIOBE-CBETM, Universite
´de Perpignan,
66860 Perpignan Cedex, France
e-mail: leray.upmc@gmail.com
Present Address:
M. Leray
Department of Invertebrate Zoology, National Museum of
Natural History, Smithsonian Institution, Washington, DC, USA
123
Coral Reefs (2014) 33:181–191
DOI 10.1007/s00338-013-1077-2
example, a single coral colony of Pocillopora damicornis
can house more than 5000 crustaceans per square meter
(Grassle 1973). Yet, we know relatively little about the
ecological processes that contribute to the diversity,
dynamics, and persistence of these more cryptic reef
communities. Studies in temperate systems point to pre-
dation as a key processes promoting marine community
dynamics (i.e., Keystone Predation, Paine 1966), and some
evidence suggests an increase in the strength of predation
at lower latitudes may contribute to high tropical diversity
(Pianka 1966; Schemske et al. 2009; Freestone et al. 2010).
However, empirical evidence supporting the positive
effects of predators on tropical diversity is mixed. For
example, predators can reduce coral and fish diversity by
disproportionately consuming rare prey (Chesher 1969;
Glynn 1976; Almany and Webster 2004). In contrast,
predation can also increase diversity. For example, Stier
et al. (2013b) showed that predatory fish can increase prey
fish diversity by promoting priority effects (Stier et al.
2013b). Additionally, Marhaver et al. (2012) showed that
species-specific microbes promote coral diversity through
negative frequency-dependent predation on coral recruits
(i.e., a Janzen-Connell effect Janzen 1970; Connell 1971).
However, studies examining the effects of predation on
tropical marine diversity have been taxonomically biased
toward conspicuous and economically valuable corals and
fishes. These groups make up only a fraction of tropical
marine biodiversity, which is heavily dominated by more
cryptic invertebrate species (Stella et al. 2011). Indeed, a
survey quantifying the taxonomic breadth of species stud-
ied in kelp forests, mangroves, seagrasses, and coral reefs
found that species studied in coral reefs have the lowest
taxonomic breadth across these four systems (Fisher et al.
2011). Such poor taxonomic coverage in studies of coral
reefs restricts our understanding of a major component of
tropical biodiversity and may therefore limit our appreci-
ation of the drivers of ecosystem function. In this study, we
focus on decapod crustaceans and small-bodied fish that
live within corals of the South Pacific. Decapod remains
litter the reef and are commonly found in fish stomach
dissections (Kulbicki et al. 2005; Enochs and Manzello
2012; Kramer et al. 2012; Leray et al. 2012a,b,2013),
suggesting that crustaceans are major nodes in a vibrant
web of tropical marine trophic interactions and that pre-
dation may be a key process driving decapod dynamics.
In addition to being a diverse yet underrepresented
group in the ecological literature on tropical marine bio-
diversity, tropical decapods are also relevant, because a
subset of decapod species play a crucial role in the resil-
ience of their coral host. Certain species of Trapezia and
Alpheus promote coral growth and survival by alleviating
stressors such as seastar predation (Glynn 1983; Pratchett
2001), sedimentation (Stewart et al. 2006), and mucus-
producing snails (Stier et al. 2010). Furthermore, these
decapods often co-occur with damselfishes that indirectly
promote coral growth through oxygenation (Goldshmid
et al. 2004) and nutrient subsidy (Meyer and Schultz 1985;
Liberman et al. 1995; Holbrook et al. 2008,2011). The
beneficial effects of mutualists on their host are often
dependent on mutualist density and identity (Holland et al.
2002); previous studies show increased coral performance
with increased density of and diversity of decapods
(McKeon et al. 2012; Stier et al. 2012) and increased
density of damselfish (Holbrook et al. 2008). The benefits
each mutualist species provides are also highly dependent
upon the fish and decapod identity (Holbrook et al. 2011;
McKeon et al. 2012; Stier et al. 2012). Therefore, if pre-
dation affects the density, diversity, and composition of
coral mutualists or if different predator species differen-
tially affect certain mutualist species, then, the perfor-
mance and resilience of the coral host may be
fundamentally altered by predator presence, density, or
identity (Knight et al. 2006; Romero et al. 2011).
Here, we quantify the effect of predation in driving the
abundance, diversity, and composition of decapods and
fishes that live cryptically embedded between the branches
of P. eydouxi, a dominant, structurally complex, reef-
building coral distributed widely across the South Pacific.
Our study combines (1) a survey to explore the spatial
distribution of our two focal predators: the coral croucher
(Scorpaenidae; Caracanthus maculatus) and the flame
hawkfish (Cirrhitidae; Neocirrhites armatus) and (2) a
recruitment experiment to examine the independent and
combined effects of each predator species on P. eydouxi-
associated communities. We show that both flame hawkfish
and coral crouchers substantially alter fish and crustacean
community organization, and we discuss the implications
of shifts in community structure for coral growth and
resilience.
Materials and methods
Study site and species
We conducted our study in the lagoon of Opunohu bay on
the North shore of Moorea, French Polynesia (17"300S,
149"500W). Coral reefs are highly diverse ecosystems, and
tens to hundreds of species can co-occur on a single coral
colony (Grassle 1973; Lassig 1977; Castro 1988; Leray
et al. 2012b). One such coral host is the coral genus
Pocillopora, which is an important reef-building coral
distributed widely across the Indo-Pacific. Multiple trophic
levels of fishes and invertebrates live among the branches
of Pocillopora, which provides structural habitat for sev-
eral co-occurring carnivorous fishes and a wide diversity of
182 Coral Reefs (2014) 33:181–191
123
potential invertebrate prey (*80 % decapods, Odinetz
1983). We focused on the two most common predatory
fishes that are strictly associated with pocilloporid corals
and co-occur in coral heads at shallow depths: the flame
hawkfish (Neocirrhites armatus) and the coral croucher
(Caracanthus maculatus) (Randall 2005). Both coral
crouchers and flame hawkfish are ambush predators that
restrict their movement to the inside of the coral host. Thus,
both predators and prey share structural refuge from larger
predators (a common phenomena in many predator–prey
systems, Gotelli and Ellison 2006), causing predators to
live in close proximity with a diversity of coral-associated
decapod prey species. The two focal predators exhibit
slight differences in microhabitat use; coral crouchers
remain deep among the coral branches, while the flame
hawkfish are more mobile. These two predators primarily
comprised of crustaceans and also includes gastropods and
fishes (morphological observations in gut contents; Randall
2005; Bacchet et al. 2006), DNA barcoding (Leray et al.
2013), and DNA metabarcoding of prey tissue remains
(Leray M, pers comm).
Predator surveys
To describe natural variation in predator density, identity,
and co-occurrence, we conducted a visual census of fish
predators living among the branches of 93 haphazardly
selected live Pocillopora eydouxi colonies on the forereef
between 6 and 10 m depth. Since adult predators occur on
larger P. eydouxi, we focused on corals with a maximum
diameter (l) and perpendicular diameter (w) ranging from
20 to 40 cm and height (h) ranging from 15 to 30 cm.
Experimental design
We executed a field experiment to quantify the response of
the prey community to shifts in the density, identity, and
co-occurrence of coral crouchers and flame hawkfish. We
collected 60 dome-shaped P. eydouxi (size range,
l9w9h: 25 925 915 to 35 935 920) from the
forereef and removed all decapod and fish species using a
low concentration of anesthetic to minimize coral stress
(0.02 % clove oil: Leray et al. 2012b). After invertebrate
removal, we transplanted corals to a 6 910 array of cinder
blocks spaced 8 m apart in a nearby sand flat within
Opunohu lagoon. Immediately following coral deployment,
we collected 20 solitary fish and 10 male–female pairs of
each of the focal predator species from P. eydouxi colonies
in a nearby lagoonal site using hand nets and anesthetic
clove oil (10 %). After tagging each individual fish with a
unique color combination of elastomer (VIE; Northwest
Marine Technology, Shaw Island, Washington, USA)
above and below the caudal peduncle, we measured and
randomly assigned fishes to six predator treatments
(n=10) (see Electronic Supplemental Material, ESM A
for pictures of experimental set-up and predators). We used
a combination of an additive and substitutive design:
Control (no predator), 1 coral croucher, 2 coral crouchers, 1
flame hawkfish, 2 flame hawkfish, and 1 coral crou-
cher ?1 flame hawkfish. For treatments with two preda-
tors of the same species, we retained the original pairing
from collection because each of these species is often
found in male–female pairs, and the placement of two
males on a colony is impossible due to intense aggressive
interactions. Predators were deployed from April to August
for 126 days, and predator treatments were pressed to
maintain initial predator densities. Over the course of the
experiment, seven flame hawkfish and five coral crouchers
migrated. We returned these fish to their original coral
head. Five flame hawkfish and eight coral crouchers died or
emigrated from the array and were replaced immediately
following their noted disappearance during predator cen-
suses conducted three times per week. At the end of the
experiment, all macro-invertebrates were collected using
clove oil, identified to the lowest possible taxonomic level,
and counted.
Statistical analysis
To test the hypothesis that co-occurrence of coral crouchers
and flame hawkfish in P. eydouxi is random, we generated
10,000 null distributions of the surveyed data by randomly
assigning each of the observed 224 fish predators to 93
corals. We, then compared observed distributions of
predator density and co-occurrence (presence/absence) to
permuted distributions.
We quantified the abundance and diversity of fishes and
decapods to measure the effect of the different predator
treatments. For abundance, we pooled the total number of
fishes and decapods together for a single analysis. For
diversity, we partitioned effects of predators into within and
between reef diversity (alpha and beta diversity, respec-
tively). We examined two metrics for within-reef diversity.
First, we assessed effects of predators on the mean number
of species per individual P. eydouxi colony (hereafter spe-
cies richness). Second, because predator treatments differed
substantially in total prey abundance and shifts in abun-
dance can cause corresponding shifts in diversity through
sampling effects, we quantified rarefied within-reef diver-
sity using individual-based rarefaction (i.e., alpha diversity
adjusted for differences among reefs in number of indi-
viduals) (Gotelli and Colwell 2001). Although less studied,
predators can affect beta diversity (i.e., the amount that prey
diversity varies from reef to reef) by modifying community
size, preferential prey consumption, or changing the
strength of priority effects (Chase et al. 2009; Stier et al.
Coral Reefs (2014) 33:181–191 183
123
2013a). To assess how predator treatments affected beta
diversity and composition, we calculated community dis-
tance matrices for the 60 coral heads. Differences in beta
diversity can occur through shifts in species incidence (i.e.,
presence-absence) or through differences in species relative
abundance. Additionally, differences in alpha diversity can
also confound beta diversity estimates (Chase et al. 2011).
We therefore examined communities using three distance
metrics that emphasize shifts in species incidence (Jaccard
index), shifts in species incidence adjusted for differences
in alpha diversity (Raup-Crick index), and relative abun-
dance (Manhattan index) (Anderson et al. 2011; Chase et al.
2011).
For each response variable, we conducted four planned
orthogonal contrasts to assess the effects of coral crouchers
(CC) and flame hawkfish (FL) on the prey community: (1)
the effect of predators: [Control] vs. [1 CC,2 CC,1 FL, 2
FL, 1 CC ?1 FL]; (2) the main effect of predator density:
[1CC, 1FL] vs. [2CC, 2FL]; (3) the main effect of predator
identity: [1 CC,2 CC]vs. [1 FL, 2 FL]; and (4) the
interaction between predator density and identity: [1 CC,2
FL]vs. [2 CC, 1 FL]. One additional (non-orthogonal)
contrast was made to determine whether mixed versus
monospecific assemblages of predators had differential
effects on prey abundance, diversity, and composition. We
used a ttest to compare monospecific and mixed treat-
ments, which were contrasted at the same density: [2 CC,
2FL] vs. [1CC ?1FL].
We modeled abundance, richness, and rarefied richness
using a general linear model with a Gaussian error distri-
bution. We tested for differences in beta diversity and
mean community composition using nonparametric multi-
variate permutation analogs of univariate Leven’s test and
ANOVA: PERMADISP (beta diversity: Anderson et al.
2006) and PERMANOVA (composition: Anderson 2001),
respectively.
We examined beta diversity using the 10 most abundant
species in the dataset (which accounted for 86 % of the
total abundance) because rare species increase ‘‘stress’’ in
distance calculations (Legendre and Legendre 1998). Note
that results for beta diversity and community composition
reported below were qualitatively identical whether or not
rare species were included. We used the statistical pro-
gramming environment R 3.0.0 for the computation of all
statistics (R Development Core Team 2013) and the
‘‘Vegan’’ package (Oksanen et al. 2013) for community
analysis and graphics.
Results
Our surveys showed that predatory fishes frequently co-
occurred and were abundant on natural reefs. We found
coral couchers and flame hawkfish in 65 and 81 % of the
93 surveyed corals, respectively. Observed patterns of
occurrence and co-occurrence (i.e., presence-absence) of
the two predator species among the branches of P. ey-
douxi are no different than expected by chance (Fig. 1b).
However, observed density distributions of predatory
fishes differed from the expected density distribution
estimated from randomizations. Coral croucher densities
naturally varied from a single individual to a triplet, with
a higher frequency of pairs and a lower frequency of
singlets than expected by chance (Fig. 1a). Flame hawk-
fish, however, exhibited a greater range (1–5) of intra-
specific density, and solitary individuals occurred more
frequently than expected by chance (Fig. 1c). Both
hawkfish and coral crouchers have sophisticated mating
systems. Facultative monogamy has been documented in
studies of flame hawkfish (Donaldson 1989) and in studies
of Caracanthus unipinna, a closely related congener of
the focal coral croucher in this study (Caracanthus mac-
ulatus) (Wong et al. 2005). Such flexible mating systems
likely contribute to the substantial variation in density of
the two predators.
In our recruitment experiment, four reefs were con-
sumed by the corallivorous seastar Acanthaster planci and
were removed from all subsequent analyses, reducing the
sample sizes for the 2CC treatment to 8 and the 1Fl and
1CC1Fl treatments to 9. A total of 5921 individuals from
73 species (54 decapods and 19 fishes) were sampled
across the 56 experimental reefs after 18 weeks (ESM B).
For decapods, 80.4 % of all individuals came from 4
families (Palaemonidae: 50.2 %, Trapeziidae: 10.9 %,
Alpheidae: 10.5 %, Galatheidae: 8.8 %), and for fishes,
86 % of individuals were damselfish. Because of the large
number of statistical tests present in the analysis, the main
text focuses on statistically significant effects (i.e.,
p\0.05), and ESM C provides the test statistics and
p-values for all contrasts. Notably, at a density of two
predators per coral, there was no difference in abundance,
diversity, or composition between mixed and monospecific
predator assemblages (ESM C); therefore we focus on the
effects of the other contrasts below.
Predators reduced total abundance
The aggregate effect of predation on prey abundance was a
34 % reduction in abundance (t
54
=2.98, p=0.003,
Fig. 2a). However, doubling predator densities had
opposing effects for each focal predator, which produced
an interaction between predator density and identity
(t
54
=2.64, p=0.008). Two flame hawkfish decreased
prey abundance by an additional 76 % relative to one flame
hawkfish, while a doubling in coral croucher density led to
a 19 % increase in prey abundance (Fig. 2a).
184 Coral Reefs (2014) 33:181–191
123
Predators reduced species richness
On average, predators reduced species richness by 20 %
(Fig. 2b) (t
54
=2.95, p=0.005); however, there was no
effect of predators when richness estimates were adjusted
for differences in abundance using rarefaction (Fig. 2c)
(t
54
=0. 30, p=0.77), suggesting reduction in prey
diversity in the presence of predators was driven by
reductions in prey abundance (i.e., a ‘‘sampling’’ effect).
There was no significant interaction or main effects of
predator density and identity for richness or rarefied rich-
ness (ESM C).
Experimental effect of predators on beta diversity
Predator density and identity each significantly affected the
beta diversity based on species incidence (identity: Fig. 3,
F
p
=11.97;p=0.003; density F
p
=4.12;p=0.01).
Reefs with flame hawkfish had 70 % higher beta diversity
than those with coral crouchers, and reefs with two pre-
dators had 43 % greater beta diversity than those with one
predator (Fig. 3b). These effects of predator density and
identity can be attributed to the interdependence of alpha
and beta diversity when using the Jaccard index. Control-
ling for differences in species richness using the Raup-
Crick metric, we found no effect of predator identity
(F
p
=1.18;p=0.28) or density (F
p
=0.07;p=0.81).
There was no significant effect of predators or interaction
between predator density and identity for beta diversity
based on species incidence or relative abundance, and there
were no main effects of predator density or identity based
on species relative abundance (ESM C).
Predator species produced unique decapod
communities
Predators shifted community composition based on prey
species relative abundance (F
p
=4.41;p=0.002).
Additionally, there was a significant main effect of predator
density on composition based on both species incidence
(F
p
=1.84;p=0.05, Fig. 3d) and relative abundance
(F
p1,54
=3.25;p=0.008, Fig. 3c). Relative to corals
with a single predator, corals with two predators had a
lower relative abundance of Chromis viridis and Galathea
mauritiana, and a higher relative abundance of Harpili-
opsis spinigera (Fig. 3c). Furthermore, the presence of two
predators reduced the incidence (i.e., number of reefs
Species
Number of Corals
CC CC+FL FL
Number of FishNumber of Fish
201 345012345
0
10
20
30
40
50 ab c
Fig. 1 Natural variation in
density of acoral crouchers
(CC) and cflame hawkfish (FL)
from surveys of 93 naturally
occurring Pocillopora eydouxi.
Patterns of co-occurrence are
also presented (b). Colored bars
(CC: blue, FL: red, CC ?FL:
green) represent 95 % quantiles
from 10,000 randomizations of
observed distribution of
predator co-occurrence
50
100
150
Control 1FL 2FL 1CC 2CC 1CC1FL
Treatment
Abundance
12
14
16
18
20
Richness
6
7
8
9
10
Rarefied Richness
a
b
c
Fig. 2 Effect of predator treatments on total abundance (a) richness
(b), and rarefied richness (mean ±1 SE). Symbols and colors
represent predator type (FL: Flame Hawkfish—red; CC: Coral
Croucher—blue, mixed assemblage (1FL1CC)—green) and density
(one predator—square, 2 predators, triangle, mixed assemblage—
diamond)
Coral Reefs (2014) 33:181–191 185
123
occupied) by the majority of common prey species (e.g.,
Alpheus lottini,Trapezia bidentata, Dascyllus flavicaudus,
and Galathea mauritiana), but increased the incidence of
Fennera chacei (Fig. 3d). The two predators differed sig-
nificantly in how they affected species composition based
on species relative abundance (F
p
=3.44;p=0.005,
Fig. 3c). Relative to coral crouchers, flame hawkfish
treatments had a greater relative abundance of T. bidentata,
Athanas dijboutensis, and H. spinigera, but had fewer F.
chacei. There was, however, no aggregate effect of pre-
dators on composition based on species incidence, inter-
action between predator density and identity based on
Periclimenes gonioporae
Trapezia bidentata
Trapezia serenei
Dascyllus avicaudus
Chromis viridis
a
c
b
d
Alpheus djiboutensis
0.10
0.15
0.20
0.25
One Two
Beta Diversity (Jaccard)
Croucher Flame
Predator Density Predator Identity
Control 1Fl 2Fl 1CC 2CC 1CC1Fl
P. gonioporae
F. chacei
Fennera chacei
H. spinigera
Palaemonidae
C. viridis
D. avicaudus
Pomacentridae
G. mauritiana
Galatheidae
A. lottini
A. djiboutensis
Alpheidae
T. bidentata
T. serenei
Trapeziidae
0.50
1.00
0.00
Proportion of Abundance
C. viridis
G. mauritiana
F. chacei
D. avicaudus
T. bidentata
A. lottini
P. gonioporae
A
. djiboutensis
H. spinigera
T. serenei
0.50
1.00
0.00
Proportion of Reefs Occupied
0
One
Two
Density
Harpiliopsis spinigera
Galathea mauritiana
Alpheus lottini
Fig. 3 The effect of predator treatments on beta diversity and species
composition of fish and decapod communities: predator density
(a) and identity (b) on beta diversity based on species incidence (a,
b). In aand b,symbols correspond to predator contrasts (see Fig. 2
legend for more detail). Also, shown are pictures of ten most common
species and their composition based on relative abundance (c) and
incidence (d). Colors in cand dcorrespond to colors on borders of
species photos
186 Coral Reefs (2014) 33:181–191
123
species incidence or relative abundance, or main effect of
predator identity on composition based on species inci-
dence (ESM C).
Predation reduces coral mutualist abundance
Both predator species negatively affected the abundance of
the majority of known mutualists species present in our
study. Predators reduced the abundance of damselfish by
60 %, with strongest effects observed with two flame
hawkfish and the mixed predator treatments (Fig. 4a). All
predator treatments had similar effects on Alpheus lottini,
reducing abundance by an average of 30 % (Fig. 4b). The
effects of predators on Trapezia were more context
dependent. On average, predators reduced small Trapezia
species (T. tigrina, T. speciosa, T. bella, T. serenei, T.
guttata, T. bidentata, and T. areolata) by 25 %. These
effects were relatively consistent with the exception of the
1 coral croucher treatment, which showed no difference
from the control treatment. There was, however, no
apparent effect of predators on the abundance of large
Trapezia spp. (T. flavopunctata and T. rufopunctata), a
group known to have strong defensive benefits during
outbreaks of the crown of thorns seastar (Acanthaster
planci) (Glynn 1983; Pratchett 2001; McKeon 2010).
Discussion
Our study explored the independent and combined effects
of two predatory fish species on a diverse decapod prey
assemblage. We showed novel effects of predatory fishes
on the abundance, diversity, and composition of coral-
dwelling fishes and decapods. Predators negatively affected
the abundance of nearly all prey species in our experiment,
with strong effects on a number of species that are known
coral mutualists. Furthermore, predators also negatively
affected two common groups from the families Palaemo-
nidae and Galatheidae, but the ecological role of these
organisms are not well known and require further study.
The strong negative effect of predators on decapod and
fish abundance is likely a product of both consumptive and
non-consumptive effects. In addition to eating prey, pre-
dators may have changed prey abundance and composition
by affecting prey activity (Stankowich and Blumstein
2005), behavior (Preisser et al. 2005), or the strength of
interactions among indirectly interacting prey (Orrock
et al. 2008). While reefs with predators had fewer species
present (i.e., lower richness, Fig. 2b), rarefied diversity was
equivalent across treatments (Fig. 2c), suggesting that
lower richness on predator treatments was driven by a
sampling effect. The two focal predators differed in their
effect on total prey and had unique effects on community
composition. The stronger negative effects of flame
hawkfish on abundance may be attributable to their larger
metabolic demand due to higher activity and larger body
size. Lastly, we speculate that the differences in prey
community composition between the two predator species
is a product of predator microhabitat preference (spatial
niche partitioning, Munday et al. 2001) with coral crou-
chers living and hunting deeper among the coral branches
than the flame hawkfish (pers. obs.).
Effects of predator density, identity, and diversity
Previous studies examining the effects of predator diversity
in modifying prey communities have concluded that both
antagonism and cooperation among predators can modify
prey community structure and ecosystem function (e.g.,
Byrnes et al. 2006; Cardinale et al. 2006). However, dis-
tinguishing the effects of predator density from diversity
and quantifying whether predators are antagonizing or
cooperating requires explicitly incorporating the nonlin-
earities of the predator functional response and the rate of
prey depletion throughout an experiment (McCoy et al.
2012). Although uncommon, the functional response and
level of prey depletion in the presence of each focal
0
10
20
30
40
6
8
10
12
14
2.0
2.5
3.0
3.5
Control
1FL
2FL
1CC
2CC
1CC1FL
0.5
1.0
1.5
2.0
2.5
Control
1FL
2FL
1CC
2CC
1CC1FL
Abundance
Damselfish
Small Trapezia spp.
Alpheus lottini
Large Trapezia spp.
ba
cd
Fig. 4 Effect of six predator treatments on the four major groups of
mutualists found in Pocillopora eydouxi: damselfish (a), Alpheus
lottini (b), small-bodied Trapezia sp. (c), and large-bodied Trapezia
sp. We distinguish between small and large Trapezia groups (see
Results for species identities), because size class is related to
characteristic differences in functional roles. Large Trapezia are
thought to be more critical for anti-predator defense against large
seastars, whereas smaller Trapezia are better known for defending
against smaller gastropods and removing sediment. Symbol shapes
and colors are identical to Fig. 2
Coral Reefs (2014) 33:181–191 187
123
predator species can be readily estimated, and then, mul-
tiple predator species functional response curves can be
integrated to generate expected proportions of prey killed
under the assumption that predator effects combined
independently. However, due to the destructive nature of
our sampling in our system in the present study, we were
unable to calculate each predator’s functional response or
the level of prey depletion, because the final prey density
was measured after recolonization rather than an instanta-
neous per capita survival of prey. This same concern
applies to testing whether predators antagonize or cooper-
ate as intraspecific predator density changes (sensu Skalski
and Gilliam 2001). Our inference about the mecha-
nism(s) underlying the changes in community structure
under different predator assemblages (e.g., density-identity
interactions, main effects of density, or monospecific ver-
sus mixed contrast) is therefore analytically constrained.
For example, flame hawkfish had stronger negative effects
on abundance with two predators relative to one predator,
whereas the effects of coral crouchers decreased with a
doubling in intraspecific density. The increased abundance
of prey in the presence of high coral croucher density may
be evidence for intraspecific antagonism, or alternatively, a
single predator may indirectly affect the foraging ability of
another predator in the absence of antagonism by simply
depleting the prey base and increasing the amount of time
another predator spends searching for prey (McCoy et al.
2012). The increased negative effect of high-density pre-
dators on species richness is similarly difficult to interpret.
We can, however, say that monospecific and mixed pred-
ator treatments exhibited statistically identical species
richness, a finding that concurs with the only other study
that has examined community-level response to predator
diversity in demographically open marine systems
(O’Connor and Bruno 2009).
The consequences of predation on mutualists
Mutualist diversity often increases host production. For
example, trees increase their productivity when the diver-
sity of root microbes increase (Van der Heijden et al.
1998), and in ant-plant mutualisms, acacia tree fitness is
maximized when a diversity of ant mutualists are present
throughout the tree’s ontogeny (Palmer et al. 2010). It is
becoming clear that predation can fundamentally alter
mutualisms by modifying mutualist density, behavior, or
composition (Anderson and Midgley 2002; Knight et al.
2006; Romero et al. 2011). In this system, mutualist ser-
vices by decapods and fishes are known to increase with
mutualist density and diversity (Holbrook et al. 2008;
McKeon et al. 2012; Stier et al. 2012); therefore, we
hypothesize that the negative effects of each of our focal
predators on mutualist abundance and diversity are likely
to have negative indirect effects on coral growth and sur-
vival. One caveat to this hypothesis is that one of the large
Trapezia species (T. rufopunctata) is known to increase the
mortality of newly settled damselfishes (Schmitt et al.
2009); however, larger Trapezia were seemingly unaf-
fected by the presence of our two focal predators (Fig. 4d).
The effects of the predators differed somewhat
depending on the predator treatment and mutualist group.
For example, the two flame hawkfish reefs and the mixed-
species predator treatment nearly eliminated damselfish
from the community, whereas the other predator treatments
maintained somewhat larger populations. The effects of
predators on abundance of Alpheus lottini and small-bodied
Trapezia spp. were more uniform, with the exception that a
single coral croucher had no obvious effect on small-bod-
ied Trapezia. At the scale of the array, the decreased
incidence of certain mutualists (e.g., Alpheus lottini,Tra-
pezia bidentata, Dascyllus flavicaudus) in treatments with
two predators relative to one predator will likely add to the
reduced level of mutualist services. Our findings comple-
ment a recent study that focused the role of a third addi-
tional predator found in this system, the arc-eye hawkfish
(Paracirrhites arcatus), affecting the density of mutualistic
damselfish. Arc-eye hawkfish reduced damselfish density,
which has cascading negative indirect effects on coral
growth (Holbrook et al. 2011).
Antagonistic and synergistic interactions among mutu-
alists can be common in systems where a diverse set of
mutualists co-occur. The presence of predators may either
disrupt or magnify these mutualist–mutualist interactions
(e.g., due to changes in mutualist density, behavior, or
composition), thereby altering host performance. In this
system, the decreased density and incidence of A. lottini in
the presence of predators may compound the negative
indirect effect of predators on the host, because T. serenei
and A. lottini are known to synergize while defending
corals from predatory seastars (McKeon et al. 2012).
Although untested, synergies within or among other deca-
pod or fish species may be similarly modified by the
presence of predators.
Predation on mutualists may also alleviate the host from
the energetic costs of symbiosis, which, depending on
environmental conditions, may be so costly as to have a net
negative effect on host performance. In this system, the
hypothesized negative indirect effects of coral crouchers
and flame hawkfish on corals may be mitigated by the fact
that corals provide lipid bodies to their decapod mutualists
in exchange for the decapod’s protective services (Stimson
1990). These lipid bodies are a marine analog to extrafloral
nectaries produced by the plants in ant-plant mutualisms
where plants often subsidize ants with nectar in exchange
for ant protective services against herbivores. Additionally,
nitrogenous waste excretion by the predators may
188 Coral Reefs (2014) 33:181–191
123
positively affect the coral by acting as a nutrient subsidy. In
contrast to the decapod mutualists, the fish-coral mutualism
is unknown to cost the host coral energy. Given the suite of
direct, indirect, and higher-order interactions present in P.
eydouxi communities, it is difficult to translate the conse-
quences of the short-term interactions measured here and in
other studies for the lifetime reproductive success of host
corals. Future studies should focus on disentangling the
effects of trophic and mutualistic interactions across the
coral ontogeny to measure the relative contribution of
certain interactions to host fitness (Palmer et al. 2010).
In conclusion, our findings add to a growing body of
literature that suggests that predators, despite their strong
effects, do not act as stabilizing agents for diverse tropical
marine systems. Although predation in this system does not
stabilize prey diversity (e.g., through frequency-dependent
or keystone predation), the unique compositional differ-
ences produced by each predator treatment provides novel
insight into the processes governing spatio-temporal vari-
ation in the size and structure of key coral mutualist
communities. Indeed, the mutualist fishes and decapods
studied here do act as stabilizing forces by facilitating the
growth and survival of the coral in the presence of a suite
of anthropogenic and natural stressors, and this allows P.
eydouxi to persist and provide structure for a number of
other non-mutualist species that hide within the branches of
the coral. Furthermore, our study expands upon the limited
taxonomic breadth in studies of reef ecosystems (Fisher
et al. 2011) by examining the ecological drivers of a
diverse, but poorly studied group of fishes and decapods
that are a major component of the diet for commercially
important fisheries species (e.g., snappers and groupers)
(Kulbicki et al. 2005). Developing a mechanistic under-
standing of the processes governing the structure and
dynamics of small but critically important cryptofauna
communities is a key to understanding the dynamics and
stability of diverse coral reef ecosystems.
Acknowledgments We thank M. Be
´raud and volunteers of ‘‘Plane
`te
Urgence’’ for assistance in the field, A. Anker, G. Paulay, and J.
Poupin for help with specimen identification, B. M. Bolker for help
with analysis, C. W. Osenberg and S. C. Mills, and two anonymous
reviewers for helpful comments on the manuscript. This work was
supported by the Partnership University Fund and the Killam Foun-
dation. Field work was conducted at CRIOBE and Richard B. GUMP
station.
References
Almany GR, Webster MS (2004) Odd species out as predators reduce
diversity of coral-reef fishes. Ecology 85:2933–2937
Anderson MJ (2001) A new method for non-parametric multivariate
analysis of variance. Austral Ecol 26:32–46
Anderson B, Midgley JJ (2002) It takes two to tango but three is a
tangle: mutualists and cheaters on the carnivorous plant Rori-
dula. Oecologia 132:369–373
Anderson MJ, Ellingsen KE, McArdle BH (2006) Multivariate
dispersion as a measure of beta diversity. Ecol Lett 9:683–693
Anderson MJ, Crist TO, Chase JM, Vellend M, Inouye BD, Freestone
AL, Sanders NJ, Cornell HV, Comita LS, Davies KF, Harrison
SP, Kraft NJB, Stegen JC, Swenson NG (2011) Navigating the
multiple meanings of beta diversity: a roadmap for the practicing
ecologist. Ecol Lett 14:19–28
Bacchet P, Zysman T, Lefe
`vre Y (2006) Guide des poissons de Tahiti
et ses ı
ˆles. Au vent des ı
ˆles, Tahiti
Brock RE, Brock JH (1977) A method for quantitatively assessing the
infaunal community in coral rock. Limnol Oceanogr 22:948–951
Byrnes J, Stachowicz JJ, Hultgren KM, Hughes AR, Olyarnik SV,
Thornber CS (2006) Predator diversity strengthens trophic
cascades in kelp forests by modifying herbivore behaviour. Ecol
Lett 9:61–71
Cardinale BJ, Srivastava DS, Emmett Duffy J, Wright JP, Downing
AL, Sankaran M, Jouseau C (2006) Effects of biodiversity on the
functioning of trophic groups and ecosystems. Nature
443:989–992
Castro P (1988) Animal symbioses in coral reef communities - a
review. Symbiosis 5:161–184
Chase JM, Biro EG, Ryberg WA, Smith KG (2009) Predators temper
the relative importance of stochastic processes in the assembly of
prey metacommunities. Ecol Lett 12:1210–1218
Chase JM, Kraft NJ, Smith KG, Vellend M, Inouye BD (2011) Using
null models to disentangle variation in community dissimilarity
from variation in a-diversity. Ecosphere 2 art24. doi:10.1890/
ES10-00117.1
Chesher RH (1969) Destruction of Pacific corals by sea star
Acanthaster planci. Science 165:280–283
Connell JH (1971) On the role of natural enemies in preventing
competitive exclusion in some marine animals and in rain forest
trees. In: Den Boer PJ, Gradwell G (eds) Dynamics of
populations: Proceedings of the advanced study institute on
dynamics of numbers of populations. PUDOC, Wageningen,
pp 298–310
Donaldson T (1989) Facultative monogamy in obligate coral-dwelling
hawkfishes (Cirrhitidae). Environ Biol Fish 26:295–302
Enochs IC, Manzello DP (2012) Species richness of motile cryptofa-
una across a gradient of reef framework erosion. Coral Reefs
31:653–661
Fisher R, Knowlton N, Brainard RE, Caley MJ (2011) Differences
among major taxa in the extent of ecological knowledge across
four major ecosystems. PLoS ONE 6:e26556
Freestone AL, Osman RW, Ruiz GM, Torchin ME (2010) Stronger
predation in the tropics shapes species richness patterns in
marine communities. Ecology 92:983–993
Glynn PW (1976) Some physical and biological determinants of coral
community structure in the Eastern Pacific. Ecol Monogr
46:431–456
Glynn PW (1983) Increased survivorship in corals harboring crusta-
cean symbionts. Mar Biol Lett 4:105–111
Goldshmid R, Holzman R, Weihs D, Genin A (2004) Aeration of
corals by sleep-swimming fish. Limnol Oceanogr 49:1832–1839
Gotelli NJ, Colwell RK (2001) Quantifying biodiversity: procedures
and pitfalls in the measurement and comparison of species
richness. Ecol Lett 4:379–391
Gotelli NJ, Ellison AM (2006) Food-web models predict species
abundances in response to habitat change. PLoS Biol
4:1869–1873
Grassle J (1973) Variety in coral reef communities. In: Jones OA,
Endean R (eds) Biology and geology of coral reefs. Academic
Press, New York, p 480
Coral Reefs (2014) 33:181–191 189
123
Holbrook SJ, Brooks AJ, Schmitt RJ, Stewart HL (2008) Effects of
sheltering fish on growth of their host corals. Mar Biol
155:521–530
Holbrook SJ, Schmitt RJ, Brooks AJ (2011) Indirect effects of species
interactions on habitat provisioning. Oecologia 166:739–749
Holland JN, DeAngelis DL, Bronstein JL (2002) Population dynamics
and mutualism: functional responses of benefits and costs. Am
Nat 159:231–244
Janzen DH (1970) Herbivores and the number of tree species in
tropical forests. Am Nat 104:501–528
Klumpp DW, McKinnon AD, Mundy CN (1988) Motile cryptofauna
of a coral reef - abundance, distribution, and trophic potential.
Mar Ecol Prog Ser 45:95–108
Knight TM, Chase JM, Hillebrand H, Holt RD (2006) Predation on
mutualists can reduce the strength of trophic cascades. Ecol Lett
9:1173–1178
Kramer MJ, Bellwood DR, Bellwood O (2012) Cryptofauna of the
epilithic algal matrix on an inshore coral reef, Great Barrier
Reef. Coral Reefs 31:1007–1015
Kulbicki M, Bozec Y-M, Labrosse P, Letourneur Y, Mou-Tham G,
Wantiez L (2005) Diet composition of carnivorous fishes from
coral reef lagoons of New Caledonia. Aquat Living Resour
18:231–250
Lassig BR (1977) Communication and coexistence in a coral
community. Mar Biol 42:85–92
Legendre P, Legendre L (1998) Numerical ecology. 2nd English
edition. Elsevier Science, Amsterdam
Leray M, Boehm J, Mills SC, Meyer C (2012a) Moorea BIOCODE
barcode library as a tool for understanding predator–prey
interactions: insights into the diet of common predatory coral
reef fishes. Coral Reefs 31:383–388
Leray M, Be
´raud M, Anker A, Chancerelle Y, Mills SC (2012b)
Acanthaster planci outbreak: decline in coral health, coral size
structure modification and consequences for obligate decapod
assemblages. PLoS ONE 7:e35456. doi:10.1371/journal.pone.
0035456
Leray M, Agudelo N, Mills SC, Meyer CP (2013) Effectiveness of
annealing blocking primers versus restriction enzymes for
characterization of generalist diets: unexpected prey revealed
in the gut contents of two coral reef fish species. PLoS ONE
8:e58076
Liberman T, Genin A, Loya Y (1995) Effects on growth and
reproduction of the coral Stylophora pistillata by the mutualistic
damselfish Dascyllus marginatus. Mar Biol 121:741–746
Marhaver KL, Vermeij MJA, Rohwer F, Sandin SA (2012) Janzen-
Connell effects in a broadcast-spawning Caribbean coral: distance-
dependent survival of larvae and settlers. Ecology 94:146–160
McCoy MW, Stier AC, Osenberg CW (2012) Emergent effects of
multiple predators on prey survival: the importance of depletion
and the functional response. Ecol Lett 15:1449–1456
McKeon CS (2010) Diversity in a tropical marine mutualism. Ph.D.
thesis, University of Florida, p123
McKeon CS, Stier A, McIlroy S, Bolker B (2012) Multiple defender
effects: synergistic coral defense by mutualist crustaceans.
Oecologia 169:1095–1103
Meyer JL, Schultz ET (1985) Tissue condition and growth rate of
corals associated with schooling fish. Limnol Oceanogr
30:157–166
Munday PL, Jones GP, Caley MJ (2001) Interspecific competition and
coexistence in a guild of coral-dwelling fishes. Ecology
82:2177–2189
O’Connor MI, Bruno JF (2009) Predator richness has no effect in a
diverse marine food web. J Anim Ecol 78:732–740
Odinetz MO (1983) Ecologie et structure des peuplements de
crustaces decapodes associes aux coraux du genere Pocillopora
en Polynesie Francasise et en Micronesie (Guam). These de 3
rd
cycle, Universite Pierre et Marie Curie, Paris
Oksanen J, Guillaume Blanchet F, Kind tR, Legendre P, O’Hara RB,
Simpson GL, Solymos P, Stevens MH, Wagner H (2013) vegan:
Community Ecology Package
Orrock JL, Grabowski JH, Pantel JH, Peacor SD, Peckarsky BL, Sih
A, Werner EE (2008) Consumptive and nonconsumptive effects
of predators on metacommunities of competing prey. Ecology
89:2426–2435
Paine RT (1966) Food web complexity and species diversity. Am Nat
100:65–75
Palmer TM, Doak DF, Stanton ML, Bronstein JL, Kiers ET, Young
TP, Goheen JR, Pringle RM (2010) Synergy of multiple partners,
including freeloaders, increases host fitness in a multispecies
mutualism. Proc Natl Acad Sci USA 107:17234–17239
Pianka ER (1966) Latitudinal gradients in species diversity: A review
of concepts. Am Nat 100:33–46
Pratchett MS (2001) Influence of coral symbionts on feeding
preferences of crown-of-thorns starfish Acanthaster planci in
the western Pacific. Mar Ecol Prog Ser 214:111–119
Preisser EL, Bolnick DI, Benard MF (2005) Scared to death? The
effects of intimidation and consumption in predator-prey inter-
actions. Ecology 86:501–509
R Development Core Team (2013) R: A language and environment
for statistical computing. R Foundation for Statistical Comput-
ing, Vienna, Austria
Randall JE (2005) Reef and shore fishes of the South Pacific.
University of Hawaii Press
Romero GQ, Antiqueira PAP, Koricheva J (2011) A meta-analysis of
predation risk effects on pollinator behaviour. PLoS ONE
6:e20689
Schemske DW, Mittelbach GG, Cornell HV, Sobel JM, Roy K (2009)
Is there a latitudinal gradient in the importance of biotic
interactions? Annu Rev Ecol Evol Syst 40:245–269
Schmitt RJ, Holbrook SJ, Brooks AJ, Lape JCP (2009) Intraguild
predation in a structured habitat: distinguishing multiple-pred-
ator effects from competitor effects. Ecology 90:2434–2443
Skalski GT, Gilliam JF (2001) Functional responses with predator
interference: viable alternatives to the Holling Type II model.
Ecology 82:3083–3092
Stankowich T, Blumstein DT (2005) Fear in animals: a meta-analysis
and review of risk assessment. Proc R Soc B-Biol Sci
272:2627–2634
Stella JS, Pratchett MS, Hutchings PA, Jones GP (2011) Coral-
associated invertebrates: diversity, ecology importance and
vulnerability to disturbance. Oceanogr Mar Biol Annu Rev
49:43–104
Stewart HL, Holbrook SJ, Schmitt RJ, Brooks AJ (2006) Symbiotic
crabs maintain coral health by clearing sediments. Coral Reefs
25:609–615
Stier AC, McKeon CS, Osenberg CW, Shima JS (2010) Guard crabs
alleviate deleterious effects of vermetid snails on a branching
coral. Coral Reefs 29:1019–1022
Stier AC, Gil MA, McKeon CS, Lemer S, Leray M, Mills SC,
Osenberg CW (2012) Housekeeping mutualisms: Do more
symbionts facilitate host performance? PLoS ONE 7:e32079
Stier AC, Geange SW, Bolker BM (2013a) Predator density and
competition modify the benefits of group formation in a shoaling
reef fish. Oikos 122:171–178
Stier AC, Geange SW, Hanson KM, Bolker BM (2013b) Predator
density and timing of arrival affect reef fish community
assembly. Ecology 94:1057–1068
Stimson J (1990) Stimulation of fat-body production in the polyps of
the coral Pocillopora damicornis by the presence of mutualistic
crabs of the genus Trapezia. Mar Biol 106:211–218
190 Coral Reefs (2014) 33:181–191
123
Van der Heijden MGA, Klironomos JN, Ursic M, Moutoglis P,
Streitwolf-Engel R, Boller T, Wiemken A, Sanders IR (1998)
Mycorrhizal fungal diversity determines plant biodiversity,
ecosystem variability and productivity. Nature 396:69–72
Wong ML, Munday P, Jones G (2005) Habitat patch size, facultative
monogamy and sex change in a coral-dwelling fish, Caracanthus
unipinna. Environ Biol Fish 74:141–150
Coral Reefs (2014) 33:181–191 191
123
Stier & Leray 2013
ESM - A: Focal predator species and experimental manipulations
Two predatory fishes: A) the coral croucher (Caracanthus maculatus) and B) the flame hawkfish
(Neocirrhites armatus) inside of Pocillopora sp. coral heads. Photo credits: Thomas Vignaud.
Experimental coral zip-tied to a cinder block (C) in the lagoon of Moorea. Coral croucher tagged
with a unique color combination of elastomer just above the anal fin (see arrow in D).
Stier & Leray 2013
Page 2 of 7
ESM - B: Effect of predators on 10 most abundant decapod
Response of the ten most common species to each predator treatment (control – black, flame
hawkfish: FL – red, coral croucher: CC – blue, and mixed species assemblage - green).
Figure B1
Abundance
0
20
40
60
10
30
50
5
15
25
0
5
10
15
0
2
4
6
8
0
2
4
0
1
2
3
0
2
4
6
0
1
2
3
4
0
1
2
3
4
1FlControl 2Fl 1CC 2CC 1CC1Fl
Predator Treatment
1FlControl 2Fl 1CC 2CC 1CC1Fl
P. goniporae T. bidentata
H. spinigera S. charon
T. serenei H. depressa
Athanas sp. Galathea sp.
A. lottini T. tigrina
a) f)
b) g)
c)
d)
e)
h)
i)
j)
Stier & Leray 2013
Page 3 of 7
Plate B1b Photographs of decapod species that recruited on experimental reefs.
A: Alpheus dolerus (Alpheidae); B: Alpheus pachychirus (Alpheidae); C: Alpheus cf. sizou (Alpheidae); D: Alpheus
diadema (Alpheidae); E: Alpheus collumianus (Alpheidae); F: Synalpheus cf. gracilirostris (Alpheidae); G: Arete cf.
indicus (Alpheidae); H: Cuapetes cf. ensifrons (Palaemonidae); I: Periclimenes gonioporae (Palaemonidae); J:
Harpiliopsis depressa (Palaemonidae); K: Fennera chacei (Palaemonidae); L: Chlorocurtis jactans (Pandalidae); M:
Thor amboinensis (Hippolytidae); N: Saron marmoratus (Hippolytidae); O: Thinora maldivensis (Hippolytidae); P:
Calaxius sp. (Axiidae); Q: Trapezia bidentata (Trapeziidae); R: Trapezia serenei (Trapeziidae); S: Trapezia tigrina
(Trapeziidae); T: Trapezia flavopunctata (Trapeziidae); U: Trapezia areolata (Trapeziidae); V: Liomera
monticulosa (Xanthidae); W: Galathea mauritiana (Galatheidae); X: Perinia tumida (Majidae)
Stier & Leray 2013
Page 4 of 7
Table B1 Mean (± 1SD ) abundance of all fishes and decapods inside P. eydouxi at the end of the recruitment study. Empty cells
represent 0. CC: Caracanthus maculatus; FL: Neocirrhites armatus.
Control
1CC
2CC
1FL
2FL
1CC1FL
Family
Species
Mean
SD
Mean
SD
Mean
SD
Mean
SD
Mean
SD
Mean
SD
Palaemonidae
P. gonioporae
29.30
23.20
18.30
14.36
33.63
25.61
16.22
19.66
5.40
6.15
22.33
14.57
Palaemonidae
H. spinigera
20.70
8.96
19.80
9.48
15.13
17.24
18.00
9.15
15.70
11.77
15.89
8.88
Palaemonidae
F. chacei
4.60
13.51
18.80
40.79
49.25
84.73
2.67
5.98
3.90
4.12
6.44
17.85
Galatheidae
G. mauritiana
19.30
17.68
14.50
8.06
8.00
14.17
7.89
7.25
2.00
3.74
2.33
2.74
Trapeziidae
T. serenei
1
4.85
11.90
6.30
6.75
2.76
6.33
4.56
5.90
3.07
6.44
3.05
Alpheidae
A. djiboutensis
7.10
5.90
4.00
4.97
4.50
2.98
5.56
3.17
4.60
4.01
3.44
2.83
Alpheidae
A. lottini
3.20
1.99
2.10
1.45
2.63
2.39
2.56
0.73
2.50
1.43
2.33
1.12
Trapeziidae
T. bidentata
2.00
1.63
1.10
0.99
1.75
1.75
2.33
1.32
1.30
1.70
1.00
1.22
Alpheidae
S. charon
1.40
0.70
1.20
0.92
1.63
0.74
1.44
0.88
1.20
1.03
1.22
1.09
Palaemonidae
H. depressa
1.60
1.51
0.90
1.10
1.13
1.55
0.89
1.83
1.30
1.83
0.89
0.93
Alpheidae
A. pachychirus
1.30
1.34
1.10
1.45
0.25
0.46
1.22
0.97
0.90
0.99
1.33
1.32
Portunidae
T. admete
1.00
0.94
1.20
1.23
0.63
0.74
0.78
0.97
0.90
0.88
0.67
0.87
Trapeziidae
T. tigrina
0.70
0.95
1.30
1.70
0.50
0.76
1.22
1.64
0.90
1.20
0.56
0.53
Hyppolitidae
T. paschalis
1.10
2.47
0.70
2.21
2.50
7.07
0.11
0.33
0.60
0.84
0.33
0.71
Alpheidae
A. diadema
0.50
0.53
0.80
0.92
0.38
0.74
1.11
1.05
0.20
0.42
0.44
1.01
Xanthidae
D. hispida
1.10
1.29
0.30
0.67
0.50
0.93
0.78
0.83
0.20
0.42
0.33
0.50
Trapeziidae
T. guttata
0.40
0.70
0.40
0.70
0.25
0.71
0.56
1.33
0.40
0.97
0.67
1.41
Palaemonidae
C. cf. ensifrons
0.20
0.63
0.20
0.63
0.63
1.19
0.56
0.88
0.50
0.53
0.44
1.01
Trapeziidae
T. areolata
0.50
0.85
0.50
0.71
0.13
0.35
0.33
0.50
0.10
0.32
0.22
0.44
Epialtidae
M. monoceros
0.30
0.67
0.10
0.32
0.25
0.71
0.44
0.88
0.22
0.44
Alpheidae
S. gracilirostris
0.20
0.63
0.10
0.32
0.13
0.35
0.22
0.67
0.30
0.67
Hyppolitidae
T. amboinensis
0.30
0.67
0.40
0.52
0.11
0.33
Palaemonidae
C. jactans
0.10
0.32
0.11
0.33
0.60
1.90
Alpheidae
A. collumianus
0.50
1.58
0.25
0.71
Alpheidae
A. equalis
0.20
0.42
0.30
0.48
0.22
0.44
Trapeziidae
T. speciosa
0.50
1.58
0.25
0.71
Trapeziidae
T. rufopunctata
0.40
0.70
0.25
0.71
0.10
0.32
Alpheidae
A. dolerus
0.40
0.70
0.22
0.44
Pilumnidae
P. tahitensis
0.30
0.95
0.30
0.48
Galatheidae
Phylladiorhynchus sp
0.11
0.33
0.44
1.33
Xanthidae
P. speciosa
0.10
0.32
0.44
0.88
Stier & Leray 2013
Page 5 of 7
Alpheidae
A. parvirostris
0.20
0.63
0.11
0.33
0.11
0.33
Trapeziidae
T. bella
0.10
0.32
0.13
0.35
0.22
0.67
Xanthidae
C. laevissima
0.10
0.32
0.13
0.35
0.11
0.33
0.10
0.32
Xanthidae
X. lamarcki
0.10
0.32
0.25
0.46
0.11
0.33
Alpheidae
Alpheus sp.
0.20
0.63
0.13
0.35
Xanthidae
Cymo sp.
0.10
0.32
0.20
0.63
Xanthidae
C. quadrilobatus
0.20
0.63
0.11
0.33
Epialtidae
P. tumida
0.10
0.32
0.10
0.32
Epialtidae
M. orientalis
0.10
0.32
0.10
0.32
Porcellanidae
Petrolisthes sp.
0.20
0.42
Trapeziidae
T. flavopunctata
0.10
0.32
0.11
0.33
Xanthidae
P. paumotensis
0.20
0.42
Xanthidae
L. monticulosa
0.10
0.32
0.10
0.32
Alpheidae
A. indicus
0.11
0.33
Alpheidae
S. tumidomanus
0.11
0.33
Alpheidae
A. sizou
0.11
0.33
Axiidae
Calaxius sp.
0.10
0.32
Hyppolitidae
S. marmoratus
0.11
0.33
Palaemonidae
Onycocaris sp.
0.10
0.32
Portunidae
T. coeruleipes
0.10
0.32
Xanthidae
M. nudipes
0.10
0.32
Xanthidae
P. pugil
0.10
0.32
Xanthidae
P. semigranosa
0.13
0.35
Pomacentridae
D. flavicaudus
16.20
6.61
11.70
9.62
10.88
8.08
11.00
12.09
3.40
4.58
2.33
2.69
Pomacentridae
C. viridis
14.70
35.82
11.11
33.33
Pomacentridae
D. aruanus
3.10
2.60
1.80
1.55
1.38
1.69
0.22
0.67
0.40
0.70
0.11
0.33
Pomacentridae
P. modestus
1.20
1.55
0.70
0.82
0.25
0.71
0.89
1.27
0.30
0.67
0.33
0.50
Scorpaenidae
S. fowleri
0.40
0.70
0.75
1.16
0.56
0.88
Pomacentridae
P. pavo
0.30
0.48
0.50
0.71
0.40
0.97
0.11
0.33
Labridae
T. hardwicke
0.20
0.42
0.30
0.48
0.25
0.46
0.22
0.44
0.10
0.32
0.22
0.44
Scorpaenidae
A. coccineus
0.40
0.84
0.38
0.52
0.22
0.44
0.11
0.33
Gobiidae
Eviota sp.
0.60
0.84
0.10
0.32
Acanthuridae
Acanthurus sp
0.20
0.42
0.13
0.35
0.10
0.32
Scorpaenidae
S. tinkhami
0.20
0.63
0.11
0.33
0.11
0.33
Chaetodontidae
C. citrinellus
0.10
0.32
0.20
0.63
Apogonidae
A. semipunctata
0.10
0.32
0.13
0.35
0.11
0.33
Stier & Leray 2013
Page 6 of 7
Holocentridae
S. microstoma
0.13
0.35
0.22
0.67
Cirrhitidae
P. arcatus
0.13
0.35
0.11
0.33
Pomacentridae
S. bandanensis
0.10
0.32
0.11
0.33
Apogonidae
O. angustatus
0.10
0.32
Balistidae
P. flavimarginatus
0.10
0.32
Scorpaenidae
P. antennata
0.10
0.32
Stier & Leray 2013
Page 7 of 7
ESM - C: Community response to five contrasts among six predator treatments.
We do not report main effects of predator density and identity (ID) when there is a significant interaction between density
and identity (Density*ID). Degrees of freedom are not available for beta diversity or composition tests because they are
perumetation based. Bolded values are significant (i.e. p < 0.05)
Predation
Density*ID
Density
ID
Diversity
Response
test
statistic
p value
test
statistic
p value
test
statistic
p value
test
statistic
p value
test
statistic
p value
abundance
t = 2.98,
df = 54
0.003
t = 2.64,
df = 54
0.008
NA
NA
NA
NA
t = 0.90,
df = 54
0.376
richness
t = 2.98,
df = 54
0.008
t = 0.13,
df = 54
0.901
t = 1.58,
df = 54
0.121
t = 0.77,
df = 54
0.446
t = 0.13,
df = 54
0.895
rarefied
richness
t = 0.30,
df = 54
0.77
1.86,
df = 54
0.068
t = 0.88,
df = 54
0.379
t = 1.91,
df = 54
0.061
t = 0.03,
df = 54
0.98
Beta
(Jaccard)
Fπ
= 2.47
0.137
Fπ
= 1.06
0.371
Fπ=
11.97
0.003
Fπ= 4.12
0.01
Fπ = 0.02
0.898
Composition
(Jaccard)
Fπ
= 2.45
0.085
Fπ
= 1.91
0.171
Fπ
= 4.12
0.049
Fπ
=
11.97
0.001
Fπ = 1.23
0.243
Beta
(Manhattan)
Fπ
= 2.47
0.104
Fπ
= 1.92
0.189
Fπ
= 0.34
0.573
Fπ
= 0.91
0.362
Fπ = 0.01
0.918
Composition
(Manhattan)
Fπ
= 4.41
0.004
Fπ
= 0.83
0.551
Fπ
= 3.25
0.006
Fπ
= 3.44
0.009
Fπ = 0.58
0.659
Beta
(Raup-Crick)
Fπ
= 3.81
0.069
Fπ
= 0.15
0.148
Fπ
= 0.07
0.807
Fπ
= 1.18
0.282
Fπ = 0.18
0.788
