Content uploaded by Alberto Garcia
Author content
All content in this area was uploaded by Alberto Garcia on Sep 24, 2024
Content may be subject to copyright.
Integrating Biotic Interactions In Niche Analyses Unravels Patterns Of
Community Composition in Clownfishes
Alberto García Jiménez1, Antoine Guisan2,3, Olivier Broennimann2,3, Théo Gaboriau1* and Nicolas
Salamin1*
1*Department of Computational Biology, University of Lausanne, Lausanne, Switzerland.
2Department of Ecology and Evolution,University of Lausanne, Lausanne, Switzerland.
3Institute of Earth Science Dynamics, University of Lausanne, Lausanne, Switzerland.
*indicates co-last authorship.
Contributing authors: agarcia26286@gmail.com;
Abstract
Biotic interactions shape the ecology of species and communities, yet their integration into ecological niche modeling meth-
ods remains challenging. Despite being a central topic of research for the past decade, the impact of biotic interactions on
species distributions and community composition is often overlooked. Mutualistic systems offer ideal case studies for exam-
ining the effects of biotic interactions on species niches and community dynamics. This study presents a novel approach to
incorporating mutualistic interactions into niche modeling, using the clownfish-sea anemone system. By adapting existing niche
quantification frameworks, we developed a method to estimate the partial effects of known interactions and refine ecologi-
cal niche estimates. This approach allows for a more comprehensive understanding of how mutualistic relationships influence
species distributions and community assembly patterns. We also used mutualistic information to investigate the resource-use over-
lap, identitying patterns of competition within clownfish communities. Our results reveal significant deviations in niche estimates
when biotic interactions are considered, particularly for specialist species. Host partitioning among clownfish species reduces
resource-use overlap, facilitating coexistence in species-rich habitats and highlighting mutualism’s role in promoting and main-
taining diversity. We uncover complex dynamics in resource-use overlap among clownfish species, influenced by factors such
as species richness, ecological niche overlap, and host specialization. Specialist-generalist interactions strike an optimal balance,
supporting high species richness while minimizing competition. These insights enhance our understanding of clownfish biodiver-
sity patterns, demonstrating how diverse mutualistic strategies contribute to diversity build-up and mitigate competitive exclusion
in saturated communities. The analytical framework presented has broad applications beyond the clownfish-sea anemone sys-
tem, potentially extending to a broader range of interactions. It enables a more comprehensive understanding of biodiversity
maintenance in complex ecosystems and constitutes a valuable tool for conservation planning and ecosystem management.
Keywords: biotic interactions, clownfish, community composition, competition, mutualism, niche, species distribution, spatial ecology
Introduction
Diverse types of biotic interactions govern the inter-
connection between species in nature. Among
them, mutualism, a relationship in which different
species benefit from each other, has attracted the atten-
tion of biologists and the public at large. Mutualism plays a
substantial role in evolutionary processes. It has impacted
major evolutionary transitions like the origin of the eukary-
otic cell or the colonization of land by symbiotic plants,
and contributes to the increase of biodiversity (Bastolla
et al. 2009) by enhancing the survival and success of
interacting species (Benton 2009). It also has an impor-
tant ecological impact by favouring ecosystem stability,
and facilitating species dispersal and resilience (Hale et al.
2020; Le Roux et al. 2020), impacting species distributions
(Pellissier et al. 2013; Schleuning et al. 2015; Marjakangas
et al. 2020).
Mutualistic interactions present a gradient of inten-
sity from full generalists to exclusive specialists, leading
to diverse ecological and evolutionary dynamics (Bas-
compte & Jordano 2007; Sverdrup-Thygeson et al. 2017;
Gracia-Lázaro et al. 2018). Generalist mutualism pro-
motes ecosystem resilience through flexible responses to
environmental changes (Maia et al. 2021). In contrast,
specialist mutualism can drive co-evolution and lead to
highly adapted species pairs (Cook & Rasplus 2003).
Generalist-specialist transitions can foster diversification
through ecological speciation (Lunau 2004), where repro-
ductive isolation evolves as a by-product of adaptation
to different ecological niches (Chomicki et al. 2019; Fra-
chon et al. 2023). As species adapt to specific mutualistic
partners, they may undergo niche partitioning, reducing
niche overlap to avoid competitive exclusion (Salas-Lopez
et al. 2022), allowing closely related species to coexist
in the same ecosystem (Schoener 1974). Plant-pollinator
systems exemplify this process, with related plant species
evolving to attract different pollinators or flower at dif-
ferent times, reducing pollination competition (Van der
2Effects Of Mutualism On Clownfish Community Composition
Niet et al. 2014). In the marine environment, the clown-
fish (Amphiprioninae) mutualism with host sea anemones
might have triggered their rapid diversification through
ecological speciation, a process known as adaptive radia-
tion (Litsios et al. 2012).
The evolutionary success of this group of 30 reef fish
species, is attributed to their unique mutualistic associa-
tions with sea anemones in the Indo-Pacific Ocean. The
mutualism significantly enhances clownfish survival and
reproductive success (Lubbock 1980; Fautin 1991), with
each species developing specific associations with up to
ten sea anemone species, resulting in both generalist and
specialist behaviors. Despite their similar ecological char-
acteristics (habitat, diet and social structure), clownfishes
exhibit higher levels of coexistence in species-rich loca-
tions (Camp 2016; Elliott & Mariscal 2001), contrary to
expectations of increased competition. We hypothesize
that mutualistic interactions with sea anemones facili-
tate niche partitioning among clownfish species, reduc-
ing interspecific competition and enabling coexistence.
This mechanism would provide ecological support for
the adaptive radiation hypothesis, demonstrating mutu-
alism’ role in promoting clownfish diversity. However,
conclusive evidence of niche partitioning among clown-
fish species is lacking, hampering the testing of adaptive
radiation hypotheses and challenging our understanding of
clownfish diversification.
Furthermore, the interplay between generalist and spe-
cialist strategies likely influences clownfish distributions
and community assembly (Chesson 2020; Bartholomew
et al. 2022; Wandrag et al. 2022). Communities dom-
inated by specialist-specialist interactions are expected
to show limitations in species richness due to increased
competition for specific anemone hosts. Conversely, com-
munities with more generalist interactions could support
higher diversity by reducing direct competition and allow-
ing for more efficient resource utilization. Investigating this
interplay could reveal spatial patterns linked to species
richness, elucidating clownfish ecological roles, and deep-
ening our understanding of how mutualistic interactions
shape community structure, species coexistence, and bio-
diversity patterns.
To test these hypotheses, we need to estimate the real-
ized environmental niches of clownfish species, while
accounting for the effect of explicit mutualistic associa-
tions. However, current approaches quantifying ecological
niches and modeling species distributions involve tech-
niques like principal component analysis (PCA; Broenni-
mann et al. 2012) and niche-based spatial modelling of
species distributions (SDMs; see Guisan et al. 2014, Valavi
et al. 2021; Norberg et al. 2019), do not directly incor-
porate biotic interactions. This is despite the demonstrated
improvement in SDM predictions that the integration of
biotic interactions can bring (Wisz et al. 2013; Early &
Keith 2019; Kass et al. 2020; Jenkins et al. 2020) and the
recurrent calls to develop SDMs that can integrate these
interactions (see Soberón 2010; Boulangeat et al. 2012;
de Araujo et al. 2013; Leach et al. 2016; D’Amen et al.
2018; but see Konig et al. 2021). Consequently, the role
of biotic interactions is frequently overlooked (Anderson
2017; Palacio & Girini 2018), which can lead to misin-
terpretations of ecological niches and species distributions
(Zurell et al. 2020; Moullec et al. 2022). In our case,
such misinterpretations would hinder our capacity to fully
understand the significant role that mutualism plays in
shaping the ecological niche of clownfishes.
Here we built on an existing environmental niche
quantification approach and implemented a novel method
to reduce the biotic uncertainty, estimate the partial effects
of known interactions, and accordingly refine the esti-
mate of the ecological niche. To capture the mutualistic
dependence of clownfishes on sea anemones and under-
stand their effects on clownfishes niche, we adapted the
‘COUE’ framework (Guisan et al. 2014) by defining new
metrics describing the effect of biotic interactions on
niche quantification (Fig. 1). We compared the impact of
biotic interactions on niche quantification between gen-
eralist and specialist clownfishes, hypothesizing that sea
anemone associations significantly affect clownfish niches
and distributions, with a greater effect on specialists than
on generalists. Furthermore, we anticipate that incorpo-
rating mutualistic interactions as a limited resource into
the quantification of resource overlap will reflect current
competition dynamics among clownfishes, as well as their
relationship to species richness and community assembly
patterns.
Material & Methods
Data Collection
Clownfishes inhabit the Indo-Pacific Ocean, from the
East coast of Africa to Polynesia in the Pacific, and from
the coast of Japan to the South of Australia, while sea
anemones have a broader, worldwide distribution (Fautin
& Allen 1992). We collected 1,636 occurrences of ten host
sea anemone species (mean: 163.6; min: 68; max: 335)
and 4,258 occurrences of 30 clownfish species (mean:
146.8; min: 2; max: 860) from RLS, GBIF, OBIS and Hex-
acoral databases (Atlas of Living Australia 2017; GBIF.org
2018; OBIS 2017; Fautin 2008 respectively). Datasets were
filtered to remove duplicates and misplaced or misiden-
tified occurrences. After filtering, two clownfish species
(Amphiprion pacificus and thiellei) with fewer than five
occurrences were excluded.
Environmental data from GMED (Basher et al. 2018)
and Bio-Oracle (Tyberghein et al. 2012; Assis et al.
2018) were obtained at 0.083° resolution, representing
approximately 9,2 km near the equator, covering physical,
chemical, and biological factors. The 53 environmental
variables collected were restricted to shallow reefs and
the epipelagic zone above 50m depth, using the UNEP-
WCMC warm water coral reef map (UNEP-WCMC 2018).
After removing variables with excessive missing data and
discarding highly correlated variables (Pearson’s correla-
tion > 0.8), we retained eight environmental variables:
mean current velocity, mean salinity, mean temperature,
mean nitrate concentration, nitrate concentration range,
mean chlorophyll concentration, dissolved oxygen con-
centration range, and mean phytoplankton concentration.
Effects Of Mutualism On Clownfish Community Composition 3
Quantifying ecological niches
We constructed a global environmental space using
the first two principal components of a scaled PCA based
on the eight selected environmental variables, which
explained 39% and 22% of the total variance, respectively.
We used the ecospat R package (Broennimann et al. 2012)
to estimate the occurrence density o for both clownfishes
and sea anemones in a two-dimensional grid of 100 by 100
cells:
o=δ(n)
max(n)(1)
where δ(n)is the 2D kernel density estimation of
the number of occurrences on the defined environmental
space, and max(n)is the maximum number of occurrences
in any grid cell of the environmental space. The occurrence
density oranges from 0 for environments where the species
is not observed to 1 where it is most observed. It represents
the species ecological niche (i.e. realized environmen-
tal niche) and can be used as a proxy of the probability
of occurrence in a given environment (Drake & Richards
2018). .
Refining the ecological niche using mutualistic
interactions
We compiled the species-specific mutualistic inter-
actions between sea anemones and clownfishes from
scientific literature (Fautin 1985, 1991; Godwin &
Fautin 1992; Ollerton et al. 2007; Ricciardi et al.
2010; Litsios et al. 2012) and reputable online sources
(https://amphiprionology.wordpress.com,www.fishbase.
org,https://reeflifesurvey.com) to construct an association
matrix Awhere each clownfish species swas represented
by a vector of interactions (Fig. 1):
As=[α1· · · αn], α ∈ {0,1}
with nbeing the number of sea anemone species (here
10 in total). Given the obligate nature of the clownfish-
anemone mutualism, we assumed that environments
unsuitable for host anemones would be unavailable to
clownfish, resulting in αvalues of either 0 or 1 in our
case (see Supplementary Material & Methods for poten-
tial extensions). Environmental availability for clownfish
would thus depend on two factors: i) the association
between a present host species and the clownfish, as
defined by matrix A, and ii) the occurrence density of sea
anemone species in the environment. The host availability
ωwas estimated as
ω= 1 −
n
∏
k=1
(1 −okαk)(2)
where αkis the association between a clownfish and its
host anemone k(derived from the matrix A), while kis
the occurrence density of the host k.ωvalues range from
0 to 1 and represent the availability score across the two-
dimensional environmental space where a suitable host is
present. Then, the refined occurrence density o’ of the focal
clownfish species given the mutualistic associations was
o′=ωo (3)
where ois the occurrence density of the focal clownfish
species, and ωis the host availability across the environ-
mental space.
Given clownfishes’ limited dispersal capacity (Jones et
al. 2005; Almany et al. 2017) and local adaptation of both
clownfishes (Huyghe & Kochzius 2015; Clark et al. 2021;
Ducret et al. 2022) and sea anemones (Sachkova et al.
2020; Will et al. 2021; Prakash et al. 2021), we anticipated
regional ecological variations. We divided the study area
into 27 regions across 5 marine realms based on MEOW
(Spalding et al. 2007; Fig. S1). We estimated the ecolog-
ical niche ofor both clownfish and sea anemones, and
the mutualism-refined niche o’ for clownfishes, at regional
scale by assessing species niches within each marine
region’s environmental space subset. Main analyses were
conducted at a global scale and without environmental
variable selection to test the robustness of our findings
across spatial scales and variable inclusion (Fig. S2 and
S3). Additionally, we projected the occurrence densities of
each estimated niche into the geographical space for spa-
tial representation of the results (Supplementary Material &
Methods). We also conducted sensitivity analyses to eval-
uate the effect of the association matrix in our framework
(Fig. S4 and Supplementary Material & Methods).
Effect of mutualistic interactions on species niches
We compared the clownfishes niche estimates with
and without accounting for the sea anemones hosts occur-
rences (ecological niche ovs. mutualism-refined niche o’)
to understand the effect of implementing explicit mutual-
istic interactions in niche estimations. We measured differ-
ent characteristics of niche space by adapting the ‘COUE’
framework (Guisan et al. 2014). In particular, we quanti-
fied changes in species’ realized niches by defining three
categories of occupied environmental space, accounting
for the influence of sea anemone hosts. With Nrepre-
senting the number of cells in the environmental grid, we
defined ‘Unavailable’ as the proportion of environmental
space not occupied by the hosts of the given clownfish
species and thus not available for the clownfish, calculated
as N(o>0&o’=0)/N(o>0) , ‘Used’ as the proportion of envi-
ronmental space occupied by the given clownfish species,
calculated as N(o>0&o’>0)/N(o>0) , and ‘Unoccupied’ as
the proportion of environmental space inhabited by any
host of the given clownfish species but not by the clown-
fish, calculated as N(o’>0)/N(ω>0). To evaluate how the
generalist-specialist spectrum affects ecological niche esti-
mates incorporating mutualistic data, we classified species
based on host associations. Following Ollerton et al. (2007)
and Litsios et al. (2014), we defined generalists as species
interacting with three or more hosts, and specialists as
those with fewer than three hosts.
4Effects Of Mutualism On Clownfish Community Composition
Fig. 1: Scheme of the proposed frame-
work to assess the effect of biotic interac-
tions and whether differential mutualistic
behaviors may produce biases in the eco-
logical niche estimation. For each clown-
fish species, georeferenced occurrences
were collected, and ecological niche was
estimated from the environmental space
created using the selected environmen-
tal variables. Additionally, georeferenced
occurrences of all sea anemones species
were collected to infer their ecological
niches following the same procedure as
for the clownfish. Hosts ecological niches
were combined into a single multi-hosts
‘niche’ using the interaction vector fol-
lowing the provided formula to create an
envelope of host availability (ω). Finally,
we constrained the clownfish ecologi-
cal niche (o) by the host availability (ω)
to obtain a mutualism-refined ecological
niche (o′). Comparisons between the host
availability envelope and the estimated
clownfish ecological niche (dashed lines)
provided the UUU parameters, determin-
ing Unavailable environments (environ-
mentally suitable but unavailable due to
lack of host availability), Used environ-
ments as they were both suitable and
available, and Unoccupied environments
as those that were available but not suit-
able for the clownfish.
Effect of mutualistic interactions on species niche
overlaps
We examined how clownfish mutualistic interactions
influence niche overlap among clownfish species. Using
Schoener’s D, we calculated pairwise ecological niche
overlap between species, both with and without consider-
ing host anemone associations. Ecological and mutualism-
refined niche overlap inadequately represents actual com-
petition in clownfish, which is primarily driven by host use.
To estimate resource-use competition more accurately, we
developed a multi-layered approach. We replicated the
environmental space into layers corresponding to each
host, projected species ecological niches onto these layers,
and constrained them by host suitability (skipping equation
2 and taking each host’s occurrence density as ω). We
then averaged layer-specific overlap values to obtain a sin-
gle overlap measure for each species pair (Fig. S5). This
method accounts for the host-specific nature of clown-
fish competition and provides a more realistic estimate
of resource-use overlap. By integrating host specificity
and environmental preferences, it captures the interplay
between habitat requirements and resource competition
in the clownfish-sea anemone mutualism system. We
compared overlap estimates (ecological niche, mutualism-
refined niche, and resource-use) and examined differences
based on species pair specialization levels (generalist-
generalist, generalist-specialist, and specialist-specialist).
Spatial patterns of clownfish interspecific niche
overlap
We examined the relationship between species rich-
ness and both ecological niche and resource-use overlap,
as well as their geographical patterns. Species richness
was defined as the number of species predicted at a loca-
tion, while overlap intensity was calculated as the average
pairwise overlaps of all species at a site. Using spatial gen-
eralized linear mixed models (GLMMs) with the spaMM
R package (Rousset & Ferdy 2014), we assessed the effect
of species richness on both overlap estimates (ecological
niche and resource-use), accounting for spatial autocor-
relation (Fig. S6,S7 and Table S1). We then analyzed
subsets of ecological niche and resource-use overlap for
generalist-generalist, specialist-generalist, and specialist-
specialist interactions, estimating species numbers for each
interaction type. Finally, we conducted spatial GLMMs for
Effects Of Mutualism On Clownfish Community Composition 5
each subset to determine if niche overlap patterns varied
by interaction type.
Results
Effect of mutualistic interactions on species
niches
Incorporating mutualistic information significantly
altered ecological niche estimates in 60% of the 108
regional niche subsets (Table S2). Across all clownfish
species and regions, an average of 66% of a species eco-
logical niche remained used after integrating mutualistic
information (XUsed = 0.664 ±0.242). Approximately one-
sixth was unavailable due to unsuitable host environments
(XUnav ailable = 0.179 ±0.249), while a similar proportion
was suitable for hosts but outside the clownfish ecologi-
cal niche (XUnoccupied = 0.178 ±0.193). UUU proportions
however, exhibited considerable variability among species
(Fig. 2) and across regions (Fig. S8).
Fig. 2: Stacked bar plot showing the averaged UUU pro-
portions per clownfish species among regions. Colours
represent the different UUU parameters specified on the
legend on top. White vertical lines represent the standard
deviation across regions.
Specialists and generalists showed significant varia-
tions in UUU proportions (Fig. 3and Table S3). Specialists
had higher Unavailable (H= 23.061, df = 1, p < 0.001)
and lower Used (H= 17.129, df = 1, p < 0.001) propor-
tions in their ecological niches compared to generalists,
while no significant difference was detected for Unoccu-
pied proportions (H= 2.1735, df = 1, p = 0.14). The
stronger effects on specialist clownfish species Unavailable
and Used proportions were evident through higher niche
dissimilarity (H= 39.987, df = 1, p < 0.001), greater cen-
troid shift (H= 20.681, df = 1, p < 0.001), and larger
environmental shift (H= 17.704, df = 1, p < 0.001) when
compared to generalists (Fig. S9). These UUU patterns
were consistent at the spatial level (Fig. S10)
Fig. 3: Comparisons between generalists and specialists on
the proportions of Unavailable (a), Used (b), and Unoccu-
pied (c) proportions of the niche, adapted from the ‘COUE’
framework (Guisan et al. 2014). Violin plots show the dis-
tribution of the data. Statistical significance is represented
following the legend: no significant (n.s.); p< 0.05 (*); p<
0.01 (**); p< 0.001 (***); p< 0.0001 (****).
6Effects Of Mutualism On Clownfish Community Composition
Fig. 4: Pairwise species compar-
ison of between ecological and
mutualism-refined niche over-
laps (a), and ecological niche and
resource-use overlaps (b). Col-
ored lines in the main plots (left)
represent individual pairwise com-
parisons across categories, with
blue lines indicating a decrease
in overlap intensity and red lines
showing an increase. Boxplots illus-
trate the overall intensity of niche
overlap at each niche level, while
violin plots show the distribution of
pairwise overlaps for each category.
Histograms on the right display the
distribution of differences between
pairwise overlaps, where blue bars
indicate decreased overlap and
red bars indicate increased over-
lap compared to ecological niche
overlap. Statistical significance is
represented following the legend:
no significant (n.s.); p< 0.05 (*);
p< 0.01 (**); p< 0.001 (***); p<
0.0001 (****).
Effect of mutualistic interactions on species
niche overlaps
Clownfish ecological niche overlap, measured by
Schoener’s Dstatistic, was high (median Decological =
0.741; IQR = 0.199), indicating highly similar environ-
mental suitability. Accounting for host associations slightly
but significant (V= 14,723; df = 1; p= 0.015)
increased niche overlap (median Dmutualism−refined =
0.762; IQR = 0.220) compared to ecological niche
overlap. However, resource-use overlap was substantially
lower (median Dresource−use = 0.232; IQR = 0.304)
and significantly different from ecological and mutualism-
refined niche overlap (V= 35,327; df = 1; p < 0.001 and
V= 33,650; df = 1; p < 0.001, respectively; Fig. 4).
Decological showed no significant differences among
specialist-specialist, specialist-generalist, generalist-
generalist interactions (H= 4.193; df = 2; p= 0.122).
However, significant differences emerged after refin-
ing the ecological niche with mutualistic information
(H= 6.7509; df = 1; p= 0.034), with stronger differences
in resource-use overlap (H= 71.623; df = 1; p < 0.001).
In specialist-specialist pairs, 15 comparisons showed
zero resource-use overlap due to not shared hosts,
while the remaining 13 pairs exhibited high over-
lap (median Dresource−use = 0.456; I QR = 0.374).
Overlap decreased significantly in generalist-generalist
(median Dresource−use = 0.351; I QR = 0.212) and
generalist-specialist interactions (median Dresource−use =
0.201; IQR = 0.172), with the latter, being the most com-
mon among co-occurring clownfishes and exhibiting the
lowest resource-use overlap (Fig. S11).
Geographical patterns of niche overlap
The Eastern and Western Coral Triangle regions showed
the highest species richness and number of potential inter-
actions (Fig 5a). High environmental suitability overlap
was found in Pacific, Western and Central Indian Oceans
regions for both ecological and mutualism-refined niches
projections (Fig 5b; ecological niche projections shown).
Resource-use overlap was more pronounced in the Tropi-
cal North-western Pacific and Somali/Arabian sea (Fig 5c).
Generally, resource-use overlap was lower than environ-
mental suitability overlap for within regions (Fig 5d), with
the largest disparities in Temperate and Tropical Pacific
regions, Northeast Australian Shelf, Coral Triangle, and
Western Indian Ocean.
Resource-use overlap correlated positively with eco-
logical overlaps across regions (r= 0.800, p <
0.001) and negatively with species richness (r=
−0.287, p < 0.001). GLMMs revealed species rich-
ness negatively associated with resource-use overlap
(β=−0.013; 95%CI [−0.015,−0.017]), while environmen-
tal suitability overlap (β= 0.668; 95%CI [0.646,0.689])
showed a positive association. Their interaction had
a negative impact on resource-use overlap (β=
Effects Of Mutualism On Clownfish Community Composition 7
0.015; 95%CI [−0.011,0.019]; Fig. S12 and Table S4). Ana-
lyzing interaction subsets (generalist-generalist, specialist-
generalist, and specialist-specialist), resource-use over-
lap among generalist-generalist interactions showed pos-
itively associated with generalist species richness (β=
0.013; 95%CI [0.009,0.017]) and environmental suitabil-
ity overlap (β= 0.415; 95%CI [0.400,0.430]). Specialist-
generalist and specialist-specialist interactions exhibited
similar effects as the overall model, with varying magni-
tudes of association between environmental suitability and
resource-use overlap (Table S4).
Discussion
Our study introduces a novel approach to incorpo-
rate mutualistic interactions in the estimation of ecological
niches. Using clownfishes as a case study, we explored
how species-specific mutualistic interactions with sea
anemones influence the ecological niche. This allowed
us to assess niche overlaps among clownfishes based
on resource use, shedding light on competition patterns
within clownfish communities. We then evaluated the role
of mutualistic behaviour and host-partitioning in resource
competing communities.
Effect of mutualistic interactions in clownfish
species niches
Our study revealed significant misalignment between
clownfish and host anemone ecological niches (Fig 2).
Despite their obligate mutualism, we found only about
60% overlap between clownfish and host niches, con-
trary to our expectation of nearly-identical environmen-
tal suitability. This discrepancy exceeds that observed in
non-dependent interactions of other organisms (e.g. Aru-
moogum et al. 2023). While some mismatch could be
attributed to data limitations and methodological con-
straints, the extent of disagreement was unexpected given
the nature of their relationship.
Comparing ecological niches (based on environmental
factors) with mutualism-refined niches revealed deviations
due to the presence or absence of suitable host anemones.
These disparities can introduce potential biases into spa-
tial models, especially for specialists species heavily con-
strained by biotic interactions (Meineri et al. 2012). Our
approach incorporates explicit information about mutual-
istic interactions, refining ecological niche estimations and
enabling detailed examination of environmental suitability
discrepancies between clownfish and their host anemones.
Spatial projections of the UUU parameters showed
‘Used’ areas (suitable for both clownfish and host
anemones) as predominant in species distributions (Fig.
S13 and S14), indicating a strong agreement in environ-
mental suitability. ‘Unavailable’ (suitable for clownfish
only) and ‘Unoccupied’ (suitable for hosts only) areas rep-
resent less common environments, highlighting suitability
mismatches between mutualistic partners (Fig. S15). The
strong agreement between clownfish and host anemones
ecological niches in predominant environments suggests
an ecologically stable clownfish-sea anemone interaction,
Fig. 5: a) Estimated number of clownfish species occurring
per location. b) Averaged ecological niche overlap among
all co-occurring species per location. c) Average resource-
use overlap among of all co-occurring species per loca-
tion. d) Difference between ecological niche overlap and
resource-use overlap computed as Dresource-use - Decological.
Negative values represent higher ecological niche over-
lap than resource-use overlap and positive values represent
higher resource-use niche overlap than ecological niche
overlap.
likely maximizing habitat occupancy in a convergent man-
ner. This pattern may indicate co-evolution, with both part-
ners adapting to similar environmental conditions, driven
by their symbiotic benefits.
Regions with high ‘Unavailable’ proportions due to
host absence may indicate ecologically unstable or fluc-
tuating populations, potentially representing sink popu-
lations. These areas show inconsistent observations of
8Effects Of Mutualism On Clownfish Community Composition
both host anemones and clownfish. Such pattern could
indicate non-self-sustaining populations relying on immi-
gration from stable source populations. However, these
sink populations might foster new ecological adaptations
(Peniston et al. 2019). In source-sink dynamics, sink habi-
tats can play crucial roles in adaptation by harboring
increased genetic variation and experience reduced com-
petition, allowing for survival of novel phenotypes (Holt,
1996; Kawecki, 2008). Temporal variation in harsh sink
environments can facilitate niche evolution, with favorable
periods allowing population growth and fixation of ben-
eficial mutations (Lenormand, 2002; Holt et al. 2003). In
clownfish-anemone systems, populations in low host avail-
ability areas might evolve wider environmental tolerances,
utilize alternative hosts, or enhance dispersal capabilities.
Long-term population genetic studies are needed to fully
understand adaptive processes of clownfishes in marginal
habitats.
‘Unoccupied’ environments could facilitate niche
expansion for clownfish species (Bruno et al. 2003; Bulleri
et al. 2016). Sea anemones may create favorable microen-
vironments (Arossa et al. 2021) enabling clownfish larvae
to settle in otherwise harsh conditions. These areas could
foster new environmental adaptations and driving range
expansions (Chen et al. 2018; Álvarez et al. 2020). This
host-mediated niche expansion might have facilitated the
clownfish clade’s expansion into the West Indian Ocean
around 5 million years ago (Litsios et al. 2014), with
long-established sea anemones (Titus et al., 2019) provid-
ing an ecological opportunity for adapted clownfish. This
hypothesis highlights the interplay between biotic interac-
tions and abiotic factors in shaping species distributions
and evolution. It also suggests sea anemones act as niche
constructors, modifying environments for their symbionts,
similar to mycorrhizal fungi altering soil for plants or corals
building reef habitats.
Consistent with previous research on mutualistic net-
works (Bascompte & Jordano, 2007), generalist clownfish
species showed higher niche overlap with host anemones
compared to specialists. This suggests generalist niches
are less constrained by biotic interactions, while special-
ist niches are strongly shaped by the specificity of their
mutualistic relationship (Fig. 3). Specialist exhibited larger
proportions of unavailable niches, likely due to their tight
association with fewer hosts. This makes niche estima-
tions incorporating biotic interactions more likely to differ
from environmental-only models (Bateman et al. 2012),
particularly for specialists linked to rare or patchily dis-
tributed hosts. Specialist niche models without accounting
for biotic constrains may be prone to overfitting, potentially
leading to inaccurate predictions, oversimplified under-
standing of complex ecological dynamics and misguided
conservation efforts (Dormann et al. 2018). This is particu-
larly crucial given as specialist species are often more vul-
nerable to climate change and extinction risks. Conversely,
areas of overlapping predicted niches for clownfish and
host anemones enhance model robustness, increasing con-
fidence in distribution estimates and validating our under-
standing of their mutualism and ecological requirements.
While our niche estimations are considered reliable,
potential biases stemming from imbalance occurrence
data or incorrect biotic associations may exist. UUU
parameters can identify areas needing better sampling or
reveal misidentified associations, as seen in recent obser-
vations of A. latezonatus,A. chagosensis and A. sebae
(Gaboriau et al. 2024). These species showed associations
with more sea anemone species than previously known,
potentially explaining high levels of unavailable niche pro-
portions. A. sebae’s distribution, spanning from the Central
Indian Ocean to the Eastern Coral Triangle, shows decreas-
ing Unavailability proportions from west to east (100%
in Central Indian Ocean Islands, 52% in Andaman, and
14% in Western Coral Triangle). This gradient suggests
its western range could be facilitated by novel ecologi-
cal interactions, indicating a shift in ecological structuring
based on different host associations across its distribu-
tion. Similar patterns at smaller scales are observed for A.
chagosesis (40% Unavailability in Central Indian Ocean
Islands) and A. latezonatus (14% in East Central Australian
Shelf and 100% in Lord Howe Island). These analyses
provide a valuable tool for conservation strategies by iden-
tifying areas or populations requiring increased monitoring
efforts.
Effect of mutualistic interactions on species
niche overlaps
Our models provide insights into clownfish interspe-
cific competition dynamics. Competition relies on physical
interactions between species, making niche and distribu-
tion inferences crucial (Godsoe et al. 2015). Our approach
enhances clownfish distribution estimations by consider-
ing their nested presence within hosts and allows studying
interactions while accounting for resource partitioning due
to varying host use.
Ecological niche overlaps were high, regardless of
mutualistic associations and host use. As such, two clown-
fish in the same region would have similar environmen-
tal suitability despite inhabiting different sea anemones.
Tropical reef fishes have narrow environmental niches,
with limited tolerance to variations in temperature, salin-
ity, pH, and oxygen levels (Brandl et al. 2020; Munday
et al. 2008). Many species show reduced fitness when
conditions deviate from their optimal range (Johansen
et al. 2014; Rummer et al. 2013). Damselfishes exhibit
decreased maximum oxygen uptake and aerobic scope
at temperatures above 31°C (Habary et al. 2017). This
specialization makes reef fish sensitive to environmental
changes and vulnerable to climate impacts (Munday et al.
2008; Brandl et al. 2020). Resource-use overlap among
clownfish species varied based on shared host anemones
and mutualistic behaviours. Species sharing multiple host
anemones exhibited greater overlap compared to those
sharing fewer hosts. Generalist clownfish, associating with
a wider range of anemone hosts, demonstrated higher
resource-use overlap than specialists, restricted to fewer
host species.
Our analyses revealed an inverse relationship between
resource-use overlap and ecological niche overlap among
Effects Of Mutualism On Clownfish Community Composition 9
clownfishes. While sea anemone associations are critical
for clownfish survival and reproduction (Lubbock 1980;
Fautin 1991), factors driving species-specific host prefer-
ences remain unclear. We suggest that current mosaic of
host preferences arose from long-term competitive dynam-
ics aimed at minimizing resource-use overlaps, .facili-
tating species coexistence in optimal environments with
high ecological niche overlap. This competition avoid-
ance through resource partitioning is observed in various
mutualistic systems, such as yucca-yucca moths, acacia-
ant, plant-pollinator, and ant-myrmecophyte mutualisms
(e.g., Addicott 1998; Palmer et al. 2003; Lee et al. 2010;
Jeavons et al. 2020). In clownfishes, host resource-use
overlap likely drives interspecific competition, as these
species share fundamental ecological characteristics such
as trophic position, reproductive behavior, phenology, and
social structure. Host use and mutualistic behavior are pri-
mary differentiating factors, influencing coloration patterns
and species recognition (Gaboriau et al. 2024). This sug-
gests host use is a strong predictor of potential competitive
dynamics among clownfish species.
While other ecological factors may influence local
competition dynamics and the scale and resolution of the
niche estimations could limit detection of biotic interac-
tions (e.g. Pearson & Dawson 2003, Araujo & Rozenfeld
2014 and Fontoura et al. 2020), our analysis reveals poten-
tial effects of mutualistic interactions on clownfish co-
existence and competitive dynamics at the regional scale.
Our study reveals that generalist-specialist interactions
are the most prevalent, with 67 unique interactions, par-
ticularly in species-rich environments. These interactions
exhibit the lowest resource-use overlap, suggesting they
are favoured in saturated communities, reducing compe-
tition and enabling species coexistence. Such interactions
align with observations in plant-animal mutualistic sys-
tems, where generalist-specialist interactions contribute to
biodiversity maintenance (Bascompte & Jordano 2007).
Geographical patterns of resource-use overlap
We observed distinct coexistence patterns among
clownfish species across regions (Fig. S16 and Table S5,
each linked to contrasting levels of ecological niche and
resource-use overlap. These patterns reflect the complex
interplay between host specialization and environmen-
tal adaptation in shaping distributions, aligning with the
observed clownfish evolutionary dynamics (Litsios et al.
2014), suggesting a trade-off between host and environ-
mental specialization.
Our findings align with Camp et al. (2016), where
approximately 25% of clownfishes inhabiting the Coral
Triangle were found in interspecific cohabiting groups,
where clownfish species richness often exceeded host
availability. This region characterized by high species co-
occurrence and significant differences between ecological
and resource-use overlap, exhibits a common coexistence
pattern involving a generalist and a specialist with low
resource-use overlap, indicating opportunities for coex-
istence through niche partitioning (Polechová & Storch
2019). However, we identified two alternative coexis-
tence scenarios in which resource-use overlap was high,
suggesting on-going competition.
First, regions at distribution edges such as the Western
Pacific, Somali/Arabian Peninsula, and Central and West-
ern Indian Ocean, exhibited high resource-use overlap
driven mainly by closely related species. These regions,
characterized by more recent colonization events (Litsios
et al. 2014), likely represent areas of ongoing adaptation to
new environmental conditions. These species pairs share
host anemones while adapting to different environmental
conditions, indicating recent divergence and specializa-
tion in distinct ecological niches. Some of those interac-
tions have been reported as never coexisting within the
same location, suggesting fine-scale competitive exclusion
not captured by our regional-scale analyses.
Second, we found distantly related species, like A.
frenatus and A. biaculeatus or A. chrysopterus and A. per-
cula, converging at distribution edges and sharing hosts
despite evolving in different ecological niches, indicat-
ing secondary contact and convergence on common hosts
(Gaboriau et al. 2024). In these cases, we observed diver-
gence between species in other traits such as coloration,
morphology, or positioning relative to the host anemone.
This observation aligns with findings by Camp et al. (2016).
Such trait divergence likely serves to reduce competition
when species share host anemones, enabling cohabitation
on the same host while maintaining species recognition to
prevent hybridization.
Our study reveals complex resource-use overlap
dynamics influenced by species richness, ecological over-
lap, and specialization levels. Generalist-generalist inter-
actions show increased resource-use overlap as ecological
overlap and generalist species number rise. Coexistence
is most likely in communities with few generalists and
moderate ecological overlap. However, generalists’ flexi-
bility in host switching may mitigate competition locally.
Specialist-specialist interactions exhibit a different pattern,
where a larger number of specialists allows reduced com-
petition if ecological overlap is low. Specialist-generalist
interactions strike an optimal balance, supporting high
species richness and ecological overlap while maintain-
ing low resource-use overlap. This balance likely pro-
motes species coexistence and may explain the remarkable
clownfish diversity in the Coral Triangle, the most environ-
mentally competitive area. These findings underscore the
importance of considering specialization levels and eco-
logical factors in understanding biodiversity maintenance
mechanisms in complex ecosystems like coral reefs.
Conclusion
Our study introduces a novel framework for assessing
species interactions’ impact on niche estimation, with sig-
nificant implications for conservation strategies and under-
standing mutualistic networks. Applied to clownfishes, we
revealed the crucial role of resource partitioning in reg-
ulating competition, illuminating the evolution of diverse
clownfish-sea anemone associations. Our findings support
10 Effects Of Mutualism On Clownfish Community Composition
established hypotheses in mutualistic systems, demonstrat-
ing how diverse strategies promote ecosystem sustainabil-
ity and mitigate negative effects in saturated communi-
ties. Competition avoidance through resource partitioning
emerges as a central mechanism shaping clownfish com-
munities across the Indo-Pacific, aligning with broader
ecological principles. While our study highlights impor-
tance of resource partitioning, it also raises questions
about its evolutionary origins versus dynamic responses
to current host composition. This framework advances
clownfish ecology understanding and provides a valuable
tool for investigating mutualistic interactions broadly. It
can be extended to account for various interaction types,
contributing to a more comprehensive understanding of
biodiversity maintenance in complex ecosystems.
Supplementary information
Acknowledgements. We thank Daniele Silvestro, Pablo Duchen
and Thibault Latrille for their contribution to the discussions of this study.
Declarations
Funding. Financial support for this research was provided by the Uni-
versity of Lausanne funds and the Swiss National Science Foundation
(Grant Number: 310030_185223).
Conflict of interest statement. The authors declare no compet-
ing interests in the publication of this work.
Data availability. All data sets used and produced, figures and R
scripts can be retrieved from the DRYAD repository (https://doi.org/10.
5061/dryad.2bvq83bv8).
Code availability. The scripts used for the analyses will be deposited
on Dryad upon acceptance of the manuscript, ensuring the reproducibility
and accessibility of our research findings.
Author contribution. AGJ, TG and NS designed the study, AGJ col-
lected the data, developed the implementation, performed the modelling,
carried out the analyses, and wrote the initial manuscript. TG and NS con-
tributed to the development of the implementation, the interpretation of
the analyses and the writing of the manuscript, with help of all authors.
References
[1] Addicott, J. F. (1998). Regulation of mutualism
between yuccas and yucca moths: population level
processes. Oikos, 119-129.
https://doi.org/10.2307/3546474
[2] Almany G. R., Planes S., Thorrold S. R., Berumen M.
L., Bode M., Saenz-Agudelo P., … Jones G. P. (2017).
Larval fish dispersal in a coral-reef seascape. Nature
Ecology and Evolution,1(6), 1–7.
https://doi.org/10.1038/s41559-017-0148
[3] Álvarez I., Fernández I., Traoré A. et al. Genomic
scan of selective sweeps in Djallonké (West African
Dwarf) sheep shed light on adaptation to harsh envi-
ronments. Sci Rep 10, 2824 (2020).
https://doi.org/10.1038/s41598-020-59839-x
[4] Anderson R. P. (2017). When and how should biotic
interactions be considered in models of species
niches and distributions? Journal of Biogeography.
https://doi.org/10.1111/jbi.12825
[5] Araujo, M. B., & Rozenfeld, A. (2014).
The geographic scaling of biotic interac-
tions. Ecography, 37(5), 406-415. doi:DOI
https://doi.org/10.1111/j.1600-0587.2013.00643.x
[6] Araújo M.B., Anderson R.P., Márcia Barbosa A.,
Beale C.M., Dormann C.F., Early R., Garcia R.A.,
Guisan A., Maiorano L., Naimi B. and O’Hara
R.B., (2019). Standards for distribution models in
biodiversity assessments. Science Advances, 5(1),
p.eaat4858.
[7] Arossa S., Barozzi A., Callegari M., Klein S.
G., Parry A. J., Hung S-H., Steckbauer A.,
Aranda M., Daffonchio D, Duarte C. M. (2021).
The Internal Microenvironment of the Sym-
biotic Jellyfish Cassiopea sp. From the Red
Sea. Frontiers in Marine Science 8 2021. doi.
https://doi.org/10.3389/fmars.2021.705915
ISSN.2296-7745.
[8] Assis J., Tyberghein L., Bosch S., Verbruggen H.,
Serrão E. A., & De Clerck O. (2018). Bio-ORACLE
v2.0: Extending marine data layers for bioclimatic
modelling. Global Ecology and Biogeography,27(3),
277–284. https://doi.org/10.1111/geb.12693
[9] Atlas of Living Australia (2017). Reef Life Survey -
Survey Records. Occurrence dataset
https://doi.org/10.15468/4v5twn accessed via
GBIF.org on 2020-05-11.
[10] Bartholomew, D. C., Banin, L. F., Bittencourt, P. R.
L., Suis, M. A. F., Mercado, L. M., Nilus, R., Burslem,
D. F. R., & Rowland, L. (2022). Differential nutri-
ent limitation and tree height control leaf physiology,
supporting niche partitioning in tropical dipterocarp
forests. Functional Ecology, 36, 2084–2103.
https://doi.org/10.1111/1365-2435.14094
[11] Bascompte J., & Jordano P. (2007). Plant-Animal
Mutualistic Networks: The Architecture of Biodi-
versity. Annual Review of Ecology, Evolution, and
Systematics,38(1), 567–593.
https://doi.org/10.1146/annurev.ecolsys.38.091206.095818
[12] Basher Z., Bowden D. A., Costello M. J.
(2018). Global Marine Environment Datasets
(GMED). World Wide Web electronic publica-
tion. Version 2.0 (Rev.02.2018). Accessed at
http://gmed.auckland.ac.nz in June 2019
[13] Bastolla U., Fortuna M., Pascual-García A. et al.
(2009). The architecture of mutualistic networks
minimizes competition and increases biodiversity.
Nature 458, 1018–1020 (2009).
https://doi.org/10.1038/nature07950
Effects Of Mutualism On Clownfish Community Composition 11
[14] Bateman B.L., VanDerWal J., Williams S.E. and John-
son C.N. (2012), Biotic interactions influence the
projected distribution of a specialist mammal under
climate change. Diversity Distrib., 18: 861-872.
https://doi.org/10.1111/j.1472-4642.2012.00922.x
[15] Boulangeat I., Gravel D. and Thuiller W. (2012),
Accounting for dispersal and biotic interactions to
disentangle the drivers of species distributions and
their abundances. Ecology Letters, 15: 584-593.
https://doi.org/10.1111/j.1461-0248.2012.01772.x
[16] Brandl S.J., Johansen J.L., Casey J.M. et al. Extreme
environmental conditions reduce coral reef fish bio-
diversity and productivity. Nat Commun 11, 3832
(2020).
https://doi.org/10.1038/s41467-020-17731-2
[17] Broennimann O., Fitzpatrick M. C., Pearman P. B.,
Petitpierre B., Pellissier L., Yoccoz N. G., … Guisan,
A. (2012). Measuring ecological niche overlap from
occurrence and spatial environmental data. Global
Ecology and Biogeography,21(4), 481–497.
https://doi.org/10.1111/j.1466-8238.2011.00698.x
[18] Bruno J.F., Stachowicz J.J., Bertness M.D. (2003).
Inclusion of facilitation into ecological theory. Trends
in Ecology & Evolution, Volume 18, Issue 3, 2003,
Pages 119-125, ISSN 0169-5347,
https://doi.org/10.1016/S0169-5347(02)00045-9
[19] Bulleri F., Bruno J.F., Silliman B.R. and Stachowicz
J.J. (2016), Facilitation and the niche: implications for
coexistence, range shifts and ecosystem functioning.
Funct Ecol, 30: 70-78.
https://doi.org/10.1111/1365-2435.12528
[20] Camp E. F., Hobbs J.P. A., De Brauwer M., Dumbrell
A. J. and Smith D. J. (2016). Cohabitation promotes
high diversity of clownfishes in the Coral Triangle.
Proc. R. Soc. B. 2832016027720160277.
http://doi.org/10.1098/rspb.2016.0277
[21] Chen Y., Shenkar N., Ni P. et al. (2018). Rapid
microevolution during recent range expansion to
harsh environments. BMC Evol Biol 18, 187, 2018.
https://doi.org/10.1186/s12862-018-1311-1
[22] Chesson, P. (2020). Species coexistence, Chapter 2.
In: Theoretical Ecology: Concepts and Applications
(eds K.S. McCann & G. Gellner). Oxford University
Press, Oxford, pp. 5–27.
[23] Clark, René D. and Aardema, Matthew L. and Andol-
fatto, Peter and Barber, Paul H. and Hattori, Akihisa
and Hoey, Jennifer A. and Montes, Humberto R. and
Pinsky, Malin L. (2021). Genomic signatures of spa-
tially divergent selection at clownfish range margins.
Proc. R. Soc. B. 288:20210407.
https://doi.org/10.1098/rspb.2021.0407C
[24] Colwell, R. K., & Rangel, T. F. (2009). Hutchinson's
duality: The once and future niche. Proceedings of
the National Academy of Sciences of the United
States of America, 106, 19651-19658.
https://doi.org/10.1073/pnas.0901650106
[25] Cook, J. M., & Rasplus, J. Y. (2003). Mutualists with
attitude: coevolving fig wasps and figs. Trends in Ecol-
ogy & Evolution, 18(5), 241-248.
10.1016/S0169-5347(03)00062-4
[26] D’Amen M., Mod H. K., Gotelli N. J., & Guisan
A. (2018). Disentangling biotic interactions, environ-
mental filters, and dispersal limitation as drivers of
species co-occurrence. Ecography.
https://doi.org/10.1111/ecog.03148
[27] de Araújo C.B., Marcondes-Machado L.O. and Costa
G.C. (2014). The importance of biotic interactions
in species distribution models: a test of the Elto-
nian noise hypothesis using parrots. J. Biogeogr., 41:
513-523. https://doi.org/10.1111/jbi.12234
[28] Dormann C.F., Bobrowski M., Dehling D.M., et al.
Biotic interactions in species distribution modelling:
10 questions to guide interpretation and avoid false
conclusions. Global Ecol Biogeogr. 2018; 27: 1004–
1016. https://doi.org/10.1111/geb.12759
[29] Drake, J. M., and R. L. Richards. 2018. Estimating
environmental suitability. Ecosphere 9(9):e02373.
https://doi.org/10.1002/ ecs2.2373
[30] Ducret, H., Timm, J., Rodríguez-Moreno, M. et al.
(2022). Strong genetic structure and limited con-
nectivity among populations of Clark’s Anemone-
fish (Amphiprion clarkii) in the centre of marine
biodiversity. Coral Reefs 41, 599–609 (2022).
https://doi.org/10.1007/s00338-021-02205-8
[31] Early R, Keith SA. (2018). Geographically vari-
able biotic interactions and implications for
species ranges. Global Ecol Biogeogr.; 28: 42– 53.
https://doi.org/10.1111/geb.12861
[32] Elliott J.K. and Mariscal R.N. (2001). Coexistence
of nine anemonefish species: differential host and
habitat utilization, size and recruitment. Mar. Biol.
138:23–36.
https://doi:10.1007/s002270000441
[33] Fautin D. (1985). Competition by anemone fishes for
host actinians. Proceedings of the 5th International
Coral Reef Congress, Vol. 5, pp. 373–377.
[34] Fautin D. G. (1991). The anemonefish symbiosis
What is known and what is not. Symbiosis,10, 23–
46.
[35] Fautin D.G. and G.R. Allen (1992). Field guide to
anemonefishes and their host sea anemones. Western
Australian Museum, Francis Street, Perth.
12 Effects Of Mutualism On Clownfish Community Composition
[36] Fautin D. G. and Buddemeier R. W. (2008). Biogeoin-
formatics of the Hexacorals
http://www.kgs.ku.edu/Hexacoral/
[37] Fontoura, L., Cantor, M., Longo, G.O., Bender, M.G.,
Bonaldo, R.M. and Floeter, S.R. (2020), The macroe-
cology of reef fish agonistic behaviour. Ecography,
43: 1278-1290.
https://doi.org/10.1111/ecog.05079
[38] Frachon, L., Arrigo, L., Rusman, Q., Poveda, L., Qi,
W., Scopece, G., & Schiestl, F. P. (2023). Putative
signals of generalist plant species adaptation to local
pollinator communities and abiotic factors. Molecu-
lar Biology and Evolution, 40(3), msad036.
https://doi.org/10.1093/molbev/msad036
[39] Franklin J. (2010). Mapping Species Distributions:
Spatial Inference and Prediction (Ecology, Biodiver-
sity and Conservation). Cambridge University Press.
https://doi:10.1017/CBO9780511810602
[40] Gaboriau, T., Marcionetti, A., Garcia Jimenez, A.,
Schmid, S., Fitzgerald, L. M., Micheli, B., Titus, B., &
Salamin, N. (2024). Host-use drives convergent evo-
lution in clownfish and disentangles the mystery of
an iconic adaptive radiation. bioRxiv, 2024-07.
https://doi.org/10.1101/2024.07.08.602550
[41] GBIF.org (18 May 2018) GBIF Occurrence Download
https://doi.org/10.15468/dl.yohpbq
[42] Godsoe W., Jankowski J., Holt R. D., & Gravel D.
(2017). Integrating Biogeography with Contemporary
Niche Theory. Trends in Ecology and Evolution,32(7),
488–499.
https://doi.org/10.1016/j.tree.2017.03.008
[43] Godwin J. and Fautin D.G. (1992). Defense of Host
Actinians by Anemonefishes. Society, American, (3),
902–908.
[44] Gracia-Lázaro C., Hernánde, L., Borge-Holthoefer J.
et al. (2018). The joint influence of competition and
mutualism on the biodiversity of mutualistic ecosys-
tems. Sci Rep 8, 9253.
https://doi.org/10.1038/s41598-018-27498-8
[45] Guisan, A., Petitpierre B., Broennimann O., Daehler
C., & Kueffer C. (2014). Unifying niche shift studies:
insights from biological invasions. Trends in Ecology
& Evolution,29(5), 260–269.
https://doi.org/https://doi.org/10.1016/j.tree.2014.02.009
[46] Guisan A., Tingley R., Baumgartner J.B., Nau-
jokaitis�Lewis I., Sutcliffe P.R., Tulloch A.I., Regan
T.J., Brotons L., McDonald�Madden E., Man-
tyka�Pringle C. and Martin T.G. (2013). Predicting
species distributions for conservation decisions.
Ecology letters, 16(12), pp.1424-1435.
[47] Guisan A., Thuiller W., & Zimmermann N.
(2017). Habitat Suitability and Distribution Mod-
els: With Applications in R (Ecology, Biodiversity
and Conservation). Cambridge University Press.
https://doi:10.1017/9781139028271
[48] Habary, A., Johansen, J. L., Nay, T. J., Steffensen, J. F.,
& Rummer, J. L. (2017). Adapt, move or die–how will
tropical coral reef fishes cope with ocean warming?.
Global Change Biology, 23(2), 566-577.
https://doi.org/10.1111/gcb.13488
[49] Hale K.R.S., Valdovinos F.S. & Martinez N.D. Mutu-
alism increases diversity, stability, and function of
multiplex networks that integrate pollinators into
food webs.Nat Commun 11,2182 (2020).
https://doi.org/10.1038/s41467-020-15688-w
[50] Holt R.D., Gomulkiewicz R., and Barfield
M. (2003). The phenomenology of niche
evolution via quantitative traits in a ‘black-
hole’ sinkProc. R. Soc. Lond. B.270215–224.
http://doi.org/10.1098/rspb.2002.2219
[51] Hutchinson, G. E. (1957). Concluding remarks. Cold
Spring Harbor Symposia on Quantitative Biology 22:
415–427.
[52] Huyghe, F. and Kochzius, M. (2017). Highly
restricted gene flow between disjunct populations of
the skunk clownfish (Amphiprion akallopisos) in the
Indian Ocean. Mar Ecol, 38: e12357.
https://doi.org/10.1111/maec.12357
[53] Jeavons, E., van Baaren, J., & Le Lann, C. (2020).
Resource partitioning among a pollinator guild: A
case study of monospecific flower crops under high
honeybee pressure. Acta Oecologica, 104, 103527.
https://doi.org/10.1016/j.actao.2020.103527
[54] Jenkins D. A., Lecomte N., Andrews G., Yannic
G., Schaefer J. A. (2020). Biotic interactions govern
the distribution of coexisting ungulates in the Arc-
tic Archipelago – A case for conservation planning,
Global Ecology and Conservation, Volume 24, 2020,
e01239, ISSN 2351-9894.
https://doi.org/10.1016/j.gecco.2020.e01239.
[55] Johansen, J. L., Messmer, V., Coker, D. J., Hoey, A. S.,
& Pratchett, M. S. (2014). Increasing ocean tempera-
tures reduce activity patterns of a large commercially
important coral reef fish. Global Change Biology,
20(4), 1067-1074.
https://doi.org/10.1111/gcb.12452
[56] Jones G. P., Planes S., & Thorrold S. R. (2005). Coral
Reef Fish Larvae Settle Close to Home. Current Biol-
ogy,15(14), 1314–1318.
https://doi.org/https://doi.org/10.1016/j.cub.2005.06.061
Effects Of Mutualism On Clownfish Community Composition 13
[57] Kass J.M., Anderson R.P., Espinosa-Lucas A., Juárez-
Jaimes V., Martínez-Salas E., Botello F., Tavera
G., Flores-Martínez, J.J. and Sánchez-Cordero, V.
(2020), Biotic predictors with phenological informa-
tion improve range estimates for migrating monarch
butterflies in Mexico. Ecography, 43: 341-352.
https://doi.org/10.1111/ecog.04886
[58] Kawecki, T. J. (2008). Adaptation to marginal habi-
tats. Annual review of ecology, evolution, and
systematics, 39(1), 321-342.
https://doi.org/10.1146/annurev.ecolsys.38.091206.095622
[59] Konig C., Wuest R. O., Graham C. H., Karger D.
N., Sattler T., Zimmermann N. E., and Zurell D.
(2021). Scale dependency of joint species distri-
bution models challenges interpretation of biotic
interactions. Journal of Biogeography 48:1541-1551.
https://doi.org/10.1111/jbi.14106
[60] Leach K., Montgomery W. I., & Reid N. (2016).
Modelling the influence of biotic factors on species
distribution patterns. Ecological Modelling.
https://doi.org/10.1016/j.ecolmodel.2016.06.008
[61] Lee, C. T., & Inouye, B. D. (2010). Mutualism
between consumers and their shared resource can
promote competitive coexistence. The American Nat-
uralist, 175(3), 277-288.
https://doi.org/10.1086/650370
[62] Lenormand, T. (2002). Gene flow and the limits
to natural selection. Trends in ecology & evolution,
17(4), 183-189.
https://doi.org/10.1016/S0169-5347(02)02497-7
[63] Le Roux J.J. et al. (2020) Biotic Interactions as Medi-
ators of Biological Invasions: Insights from South
Africa. In: van Wilgen B., Measey J., Richardson D.,
Wilson J., Zengeya T. (eds) Biological Invasions in
South Africa. Invading Nature - Springer Series in
Invasion Ecology, vol 14. Springer, Cham.
https://doi.org/10.1007/978-3-030-32394-3 14
[64] Lira-Noriega A., Soberón J., Miller C.P. (2013).
Process-based and correlative modeling of desert
mistletoe distribution: a multiscalar approach. Eco-
sphere 4(8): 1-23.
https://doi.org/10.1890/ES13-00155.1
[65] Litsios G., Kostikova A., & Salamin N. (2014). Host
specialist clownfishes are environmental niche gen-
eralists. Proceedings of the Royal Society B: Biological
Sciences,281(1795).
https://doi.org/10.1098/rspb.2013.3220
[66] Litsios G., Pearman P.B., Lanterbecq D., Tolou N.
and Salamin N. (2014), The radiation of the clown-
fishes has two geographical replicates. J. Biogeogr.,
41: 2140-2149.
https://doi.org/10.1111/jbi.12370
[67] Litsios G., Sims C. A., Wüest R. O., Pearman P. B.,
Zimmermann N. E., & Salamin N. (2012). Mutual-
ism with sea anemones triggered the adaptive radia-
tion of clownfishes. BMC Evolutionary Biology, 12(1).
https://doi.org/10.1186/1471-2148-12-212
[68] Lubbock R. (1980). Why are clownfishes not stung
by sea anemones? Proc. R. Soc. Lond. B.20735–61.
http://doi.org/10.1098/rspb.1980.0013
[69] Lunau, K. (2004). Adaptive radiation and
coevolution—pollination biology case studies.
Organisms Diversity & Evolution, 4(3), 207-224.
https://doi.org/10.1016/j.ode.2004.02.002
[70] Marjakangas E-L., Abrego N., Grøtan V., et al. (2020).
Fragmented tropical forests lose mutualistic plant–
animal interactions. Divers Distrib. 2020; 26: 154–
168.
https://doi.org/10.1111/ddi.13010
[71] Mebs D. (2009). Chemical biology of the mutualistic
relationships of sea anemones with fish and crus-
taceans. Toxicon. 2009 Dec 15;54(8):1071-4. doi:
10.1016/j.toxicon.2009.02.027. Epub 2009 Mar 5.
PMID: 19268681.
[72] Meineri E., Skarpaas O., Vandvik V. (2012). Model-
ing alpine plant distributions at the landscape scale:
Do biotic interactions matter? Ecological Modelling,
Volume 231, 2012, Pages 1-10, ISSN 0304-3800,
https://doi.org/10.1016/j.ecolmodel.2012.01.021
[73] Moullec F., Barrier N., Drira S., Guilhaumon F., Hat-
tab T., Peck M. A., and Shin Y. J. (2022). Using species
distribution models only may underestimate climate
change impacts on future marine biodiversity. Eco-
logical Modelling 464:11.
https://doi.org/10.1016/j.ecolmodel.2021.109826.
[74] Munday, P. L., Jones, G. P., Pratchett, M. S., &
Williams, A. J. (2008). Climate change and the future
for coral reef fishes. Fish and Fisheries, 9(3), 261-285.
https://doi.org/10.1111/j.1467-2979.2008.00281.x
[75] Norberg A., Abrego N., Blanchet F. G., Adler F. R.,
Anderson B. J., Anttila J., Araújo M. B., Dallas T.,
Dunson D., Elith J., Foster S. D., Fox R., Franklin J.,
Godsoe W., Guisan A., O'Hara B., Hill N. A., Holt
R. D., Hui F. K. C., Husby M., Kålås J. A., Lehikoinen
A., Luoto M., Mod H. K., Newell G., Renner I., Roslin
T., Soininen J., Thuiller W., Vanhatalo J., Warton
D., White M., Zimmermann N. E., Gravel D., and
Ovaskainen O. (2019). A comprehensive evaluation
of predictive performance of 33 species distribution
models at species and community levels. Ecological
Monographs 89(3):e01370.
https://doi.org/10.1002/ecm.1370
[76] OBIS (2018). Ocean Biodiversity Information Sys-
tem Intergovernmental Oceanographic Commission
14 Effects Of Mutualism On Clownfish Community Composition
of UNESCO (2017)
[77] Ollerton J., McCollin D., Fautin D. G., & Allen
G. R. (2007). Finding NEMO: Nestedness engen-
dered by mutualistic organization in anemonefish
and their hosts. Proceedings of the Royal Soci-
ety B: Biological Sciences,274(1609), 591–598.
https://doi.org/10.1098/rspb.2006.3758
[78] Palacio F.X. and Girini J.M. (2018), Biotic interactions
in species distribution models enhance model perfor-
mance and shed light on natural history of rare birds:
a case study using the straight-billed reedhaunter
Limnoctites rectirostris.J Avian Biol, 49: e01743.
https://doi.org/10.1111/jav.01743
[79] Palmer T. M., Stanton M. L., and Young T. P. (2003).
Competition and Coexistence: Exploring Mecha-
nisms that Restrict and Maintain Diversity within
Mutualist Guilds. The American Naturalist 162(S4):
S63– 79. https://doi.org/10.1086/378682
[80] Pellissier L, Pinto-Figueroa E, Niculita-Hirzel H,
Moora M, Villard L, Goudet J, Guex N, Pagni
M, Xenarios I, Sanders I, Guisan A. (2013) Plant
species distributions along environmental gradients:
do belowground interactions with fungi matter? Front
Plant Sci. 2013 Dec 10;4:500.
doi: 10.3389/fpls.2013.00500. PMID: 24339830;
PMCID: PMC3857535.
[81] Peniston J.H., Barfield M., and Holt R.D. (2019).
Pulsed Immigration Events Can Facilitate Adaptation
to Harsh Sink Environments. The American Naturalist
2019 194:3, 316-333.
doi:10.1086/704608
[82] Pearson, R. G., & Dawson, T. E. (2003). Predicting
the impacts of climate change on the distribution
of species: are bioclimate envelope models useful?.
Global Ecology & Biogeography, 12, 361-372.
[83] Polechová J. & Storch D. (2019). Encyclopedia
of Ecology (2nd Edition). Elsevier, Volume 3, p.
72-80. https://doi.org/10.1016/B978-0-12-409548-
9.11113-3
[84] Prakash, S., Muthu, A., & Kumar, A. (2021). Pop-
ulation structure and reproductive performance in
the sea anemone associated shrimp Ancylocaris bre-
vicarpalis (Caridea: Palaemonidae). Journal of the
Marine Biological Association of the United Kingdom,
101(1), 109-116. doi:10.1017/S002531542000137X
[85] R Core Team (2018). R: A language and environment
for statistical computing. R Foundation for Statisti-
cal Computing, Vienna, Austria. Available online at
https://www.R-project.org/
[86] Ricciardi F., Boyer, M., & Ollerton J. (2010).
Assemblage and interaction structure of the
anemonefish-anemone mutualism across the
Manado region of Sulawesi, Indonesia. Envi-
ronmental Biology of Fishes,87(4), 333–347.
https://doi.org/10.1007/s10641-010-9606-0
[87] Rousset F. and Ferdy J-B. (2014). Testing envi-
ronmental and genetic effects in the presence of
spatial autocorrelation. Ecography 37(8): 781-790.
http://dx.doi.org/10.1111/ecog.00566
[88] Rummer, J. L., Stecyk, J. A., Couturier, C. S., Watson,
S. A., Nilsson, G. E., & Munday, P. L. (2013). Elevated
CO2 enhances aerobic scope of a coral reef fish. Con-
servation Physiology, 1(1), cot023.
https://doi.org/10.1093/conphys/cot023
[89] Sachkova, M.Y., Macrander, J., Surm, J.M. et al.
(2020). Some like it hot: population-specific adap-
tations in venom production to abiotic stressors in
a widely distributed cnidarian. BMC Biol 18, 121
(2020). https://doi.org/10.1186/s12915-020-00855-
8
[90] Salas-López, A., Violle, C., Munoz, F., Menzel, F.,
& Orivel, J. (2022). Effects of habitat and competi-
tion on niche partitioning and community structure in
neotropical ants. Frontiers in Ecology and Evolution,
10, 863080.
https://doi.org/10.3389/fevo.2022.863080
[91] Schlager S. (2017). “Morpho and Rvcg – Shape Anal-
ysis in R.” In Zheng G, Li S, Szekely G (eds.), Sta-
tistical Shape and Deformation Analysis, 217–256.
Academic Press. ISBN 9780128104934.
[92] Schleuning M., Fründ, J. and García, D. (2015), Pre-
dicting ecosystem functions from biodiversity and
mutualistic networks: an extension of trait-based con-
cepts to plant–animal interactions. Ecography, 38:
380-392. https://doi.org/10.1111/ecog.00983
[93] Schoener, T. W. (1974). Resource Partitioning in
Ecological Communities: Research on how similar
species divide resources helps reveal the natural reg-
ulation of species diversity. Science, 185(4145), 27-
39.
https://doi.org/10.1126/science.185.4145.27
[94] Soberón J. M. (2010). Niche and area of distribution
modeling: A population ecology perspective. Ecogra-
phy,33(1), 159–167. https://doi.org/10.1111/j.1600-
0587.2009.06074.x
[95] Spalding M. D., Fox H. E., Allen G. R., Davidson N.,
Ferdaña Z. A., Finlayson M., Halpern B.S., Jorge M.A.,
Lombana A., Lourie S.A., Martin K.D., McManus E.,
Molnar J., Recchia C.A., Robertson J. (2007). Marine
Ecoregions of the World: A Bioregionalization of
Coastal and Shelf Areas. BioScience,57(7), 573–583.
https://doi.org/10.1641/b570707
Effects Of Mutualism On Clownfish Community Composition 15
[96] Sverdrup-Thygeson A., Skarpaas, O., Blumen-
trath S., Birkemoe T., & Evju M. (2017). Habitat
connectivity affects specialist species richness
more than generalists in veteran trees. Forest
Ecology and Management,403(1432), 96–102.
https://doi.org/10.1016/j.foreco.2017.08.003
[97] Thioulouse J., Chessel D., Dolédec S., & Olivier J. M.
(1997). ADE-4: A multivariate analysis and graphical
display software. Statistics and Computing,7(1), 75–
83.
https://doi.org/10.1023/A:1018513530268
[98] Titus B. M., Benedict C., Laroche R., Gusmão L.
C., van Deusen V., Chiodo T., Meyer C. P., Beru-
men M. L., Bartholomew A., Yanagi K., Reimer J. D.,
Fujii T., Daly M., & Rodríguez E. (2019). Phyloge-
netic relationships among the clownfish-hosting sea
anemones. Molecular Phylogenetics and Evolution,
139. https://doi.org/10.1016/j.ympev.2019.106526
[99] Tyberghein L., Verbruggen H., Pauly K., Troupin C.,
Mineur F., & De Clerck O. (2012). Bio-ORACLE: A
global environmental dataset for marine species dis-
tribution modelling. Global Ecology and Biogeogra-
phy,21(2), 272–281. https://doi.org/10.1111/j.1466-
8238.2011.00656.x
[100] UNEP-WCMC, WorldFish Cen-
tre, WRI, TNC (2021). Global distribution of
warm-water coral reefs, compiled from multiple
sources including the Millennium Coral Reef Map-
ping Project. Version 4.1. Includes contributions
from IMaRS-USF and IRD (2005), IMaRS-USF (2005)
and Spalding et al. (2001). Cambridge (UK): UN
Environment World Conservation Monitoring Centre.
https://doi.org/10.34892/t2wk-5t34
[101] Valavi R., Guillera-Arroita G., Lahoz-Monfort
J. J., and Elith J. (2022). Predictive perfor-
mance of presence-only species distribution
models: a benchmark study with reproducible
code. Ecological Monographs 92(1):e01486.
https://doi.org/10.1002/ecm.1486
[102] Van der Niet, T., Peakall, R., & Johnson, S. D. (2014).
Pollinator-driven ecological speciation in plants: new
evidence and future perspectives. Annals of Botany,
113(2), 199-212.
https://doi.org/10.1093/aob/mct290
[103] Wandrag, E.M., Catford, J.A. and Duncan, R.P.
(2023), Niche partitioning overrides interspecific
competition to determine plant species distributions
along a nutrient gradient. Oikos, 2023: e08943.
https://doi.org/10.1111/oik.08943
[104] Will H Ryan, Jaclyn Aida, Stacy A Krueger-Hadfield
(2021). The Contribution of Clonality to Popu-
lation Genetic Structure in the Sea Anemone,
Diadumene lineata, Journal of Heredity, Vol-
ume 112, Issue 1, January 2021, Pages 122–139,
https://doi.org/10.1093/jhered/esaa050
[105] Wisz M. S., Pottier J., Kissling W. D., Pellissier L.,
Lenoir J., Damgaard C. F., … Svenning J. C. (2013).
The role of biotic interactions in shaping distributions
and realised assemblages of species: Implications for
species distribution modelling. Biological Reviews,
88(1), 15–30.
https://doi.org/10.1111/j.1469-185X.2012.00235.x
[106] Zurell D., Zimmermann N.E., Gross H., Bal-
tensweiler A., Sattler T., Wüest R.O. (2020). Testing
species assemblage predictions from stacked and
joint species distribution models. J Biogeogr. 2020;
47: 101– 113. https://doi.org/10.1111/jbi.13608
