Resource partitioning and functional diversity
of worldwide freshwater fish communities
Lise Comte,1,2,† Julien Cucherousset,2 Stéphanie Boulêtreau,3,4 and Julian D. Olden1
1School
of Aquatic and Fishery Sciences, University of Washington, Seattle, Washington 98105 USA
EDB (Laboratoire Evolution & Diversité Biologique), Université Toulouse III Paul Sabatier,
ENFA, 118 route de Narbonne, F-31062, Toulouse, France
3INP, UPS, EcoLab (Laboratoire Ecologie Fonctionnelle et Environnement), Université de Toulouse,
118 route de Narbonne, F-31062, Toulouse, France
4CNRS, EcoLab, F-31062, Toulouse, France
2CNRS, UMR5174
Citation: Comte, L., J. Cucherousset, S. Boulêtreau, and J. D. Olden. 2016. Resource partitioning and functional diversity
of worldwide freshwater fish communities. Ecosphere 7(6):e01356. 10.1002/ecs2.1356
Abstract. Despite significant progress in recent years, understanding the rules governing the assembly
of natural communities is still challenging and knowledge of how the integration of nonnative species may
disrupt community structure and function is needed. To address this challenge, we collated stable isotope
data for 159 freshwater fish communities around the world with and without nonnative species and quantified spatial variation in both community isotopic functional diversity and intraspecific variation in species niches. Using a null model and partial least squares path analysis, we then evaluated how the interplay
between abiotic (historical, energetic, climatic, habitat size) and biotic (niche segregation) factors shape
community structure and functional diversity, and how these relationships have changed, and with what
consequences, in the presence of nonnative species. We found that niche partitioning is a primary force
underlying the structure and functional redundancy of native fish communities, which may be ­governed
by a synergism between contemporary climate, productivity and habitat size. We also found ­evidence of
a legacy of historical climate on functional diversity, independent from species richness. By contrast, path
models of communities containing nonnative species demonstrated lower explanatory power and had no
clear association with any of the abiotic or biotic factors. In conclusion, we demonstrated that strong spatial
patterns in community structure and functional diversity of freshwater fish communities exist at the global
scale, underlined by the complex interplay between external and internal filters, but that these patterns
may be blurred by anthropogenic species introductions. Our results further highlighted the importance
of accounting for realized species niches and species status (i.e., native and nonnative) when investigating
questions related to the assembly and functional diversity of multitrophic communities.
Key words: biotic interactions; community structure; environmental filtering; functional redundancy; lakes; niche differentiation; processes of coexistence; realized niche; resource use plasticity; stable isotopes.
Received 19 October 2015; revised 7 February 2016; accepted 25 February 2016. Corresponding Editor: D. P. C. Peters.
Copyright: © 2016 Comte et al. This is an open access article under the terms of the Creative Commons Attribution
­License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
† E-mail: [email protected]
but their relative contribution in shaping the
observed patterns of biodiversity is still largely
unresolved (Cavender-­Bares et al. 2009, Swenson
2013). This is particularly true for present-­day
communities in which continued introductions
of nonnative species may be trumping natural
Introduction
Environmental conditions, species interactions, and historical processes have been widely
recognized to simultanouesly influence the taxonomic and functional structure of communities,
v www.esajournals.org
1
June 2016 v Volume 7(6) v Article e01356
COMTE ET AL.
drivers of biodiversity patterns and overriding
long-­evolved networks of biotic interactions (Gotelli and Arnett 2000, Leprieur et al. 2008, Olden
et al. 2008). Species introductions may reshuffle
the number, abundance, and functional characteristics of species within local communities, by
causing declines or extirpations of native populations and by adding new and/or redundant
functional traits (Olden et al. 2004, Leprieur et al.
2009, Matsuzaki et al. 2013, Azzurro et al. 2014).
In turn, interactions between coexisting species
may cause rapid and drastic changes in community organization and functional diversity, even
if the number of species remains unchanged
(Gotelli and Arnett 2000, Sanders et al. 2003, Sagouis et al. 2015). Additional research is therefore
needed to more clearly elucidate the rules governing the assembly of natural communities and
understand how the integration of nonnative
species may disrupt community structure and
functional biogeography (Hobbs et al. 2006, Sax
et al. 2007).
The concept of environmental filtering posits
that local abiotic conditions selects for species according to their environmental tolerances, thus
limiting the number and functional attributes of
species in a community (e.g., “physiological tolerance” hypothesis; Currie et al. 2004, “digestive
constraint” hypothesis; González-­Bergonzoni
et al. 2012). Similarly, evidence suggests that
Quaternary climatic changes have induced taxonomic and functional legacies in regional species
pool through differential processes of speciation,
extinction, and dispersal (“historical climate stability” hypotheses; Araújo et al. 2008, Ordonez
and Svenning 2015). By contrast, the principles
of competitive exclusion and limiting similarity
are expected to favor the coexistence of functionally dissimilar species by promoting the
exploitation of different resources (“niche partitioning” hypothesis; Pianka 1974, Schoener 1974,
Mason et al. 2008). However, previous investigations have demonstrated that the environmental
­context, such as microhabitat heterogeneity or
­resource availability, can greatly influence the
partitioning of resources among species (Chesson 2000, Mouquet et al. 2002). Taken together,
the functional diversity of communities is expected to result from a hierarchical process whereby external abiotic filters first sort species from
a regional species pool and internal biotic filters
v www.esajournals.org
(themselves influenced by abiotic factors) then
regulate species coexistence within local communities through intraspecific niche differentiation
(Farias and Jaksic 2009, Violle et al. 2012).
Trait-­based approaches have been increasingly used to explore the potential mechanisms responsible for structuring functional diversity at
both local and global scales (Mouillot et al. 2007,
Swenson 2013, Violle et al. 2014). Empirical evidence points to both functional clustering (co-­
occurrence of functionally similar species) and
overdispersion (co-­occurrence of functionally
dissimilar species), but the role of intraspecific
niche variability in controlling the coexistence
of species is still under-­appreciated (Jung et al.
2010, Violle et al. 2012). This is partly because
of the challenges associated with measuring the
realized ecological niche of species in a community and the partitioning of niche space among
species. Indeed, most studies measuring community functional space do so according to mean
and fixed trait values of species to infer resource
use and potential competition, thus disregarding inter-­ and intra-­population niche variability. However, although an association between
species traits similarity and resource overlap in
animal communities is intuitively appealing, it
does not capture the inherent, and often considerable, plasticity in animal resource exploitation
(Albouy et al. 2011, Zhao et al. 2014). Stable isotopes ratios, by integrating dietary history and
providing an estimate of realized trophic interactions (Bearhop et al. 2004, Newsome et al. 2007),
represent an appealing alternative to indirect
measures of species niches based on ecomorphological traits (e.g., Villéger et al. 2008, Azzurro
et al. 2014). Furthermore, stable isotope analyses
provide insight into complementarity aspects of
species realized niches (i.e., the origin of the carbon used by consumers and their trophic position within the food chain) and therefore relate
isotopic niche space to the n-­dimensional niche
concept (Layman et al. 2012, Cucherousset and
Villéger 2015, Rigolet et al. 2015). While patterns
in resource use are likely captured by isotopic
niches, isotopic overlap does not necessarily indicate the sole existence of direct interactions
between organisms, as isotope values do not
provide information regarding species feeding
behavior and microhabitat use. Nonetheless,
understanding the macro-­ecological patterns
2
June 2016 v Volume 7(6) v Article e01356
COMTE ET AL.
Fig. 1. Global distribution of the freshwater fish communities used in the present study (n = 159) with blue
for communities containing nonnative species (n = 80) and orange for communities without nonnative species
(n = 79). The lines indicate the densities of communities along the latitudinal gradient.
in isotopic community structure may provide a
promising way to bring new insights on how local species interactions contribute to community
assembly and functional diversity.
Here, we quantify the relative importance of
both external abiotic factors and internal biotic-­
mediated filters in structuring freshwater fish
communities and patterns of isotopic functional
diversity across the globe, and evaluate the role
of nonnative species in altering these relationships. We collated data from a comprehensive
review of stable isotope studies spanning a large
gradient in species richness and environments
to quantify both the isotopic functional diversity and the level of intraspecific variation in realized niches of freshwater fish communities with
and without nonnative species. We first assessed
patterns of functional redundancy across communities, and then evaluated how the interplay
between abiotic (i.e., historical, energetic, climatic, habitat size) and biotic (i.e., niche segregation)
factors shape community structure and functional diversity. Next, we explored how the putative
mechanisms identified above may change in the
presence of nonnative species.
isotopes was compiled from both peer-­reviewed
and gray literature. We focused exclusively on
standing (lentic) water ecosystems to take advantage of the relative isolation of lacustrine
ecosystems (i.e., lakes, reservoirs, and ponds)
compared to flowing (lotic) water ecosystems.
Studies were included only if the entire fish community was examined (species richness ranged
from 4 to 40), excluding data from experimental
manipulations. From this database, we extracted quantitative information on both mean stable isotope values (i.e., niche position) and their
standard deviation (i.e., intraspecific niche variability) for each species in each community. Species were further classified as native or nonnative
according to their known indigenous range on a
country or state basis based on FishBase (Froese
and Pauly 2015) and the USGS Nonindigenous
Aquatic Species database (http://nas.er.usgs.
gov) (see Sagouis et al. 2015 for details). The final database included 159 independent freshwater fish communities, among which half (n = 80)
contained at least one nonnative species contributing between 6% and 100% of the total species
richness (Fig. 1; Appendix S1: Table S1). Communities were located across six continents and
a broad diversity of environmental conditions.
The average human footprint index (1 km grid;
LWP-­2, 2005) calculated for each hydrographic
basin did not differ significantly (t-­test, P = 0.63)
between communities with (24.67 ± 16.84 SD)
and without (25.82 ± 16.84 SD) nonnative species.
Methods
Data sources
Stable isotopes.—A database quantifying the
isotopic structure of freshwater fish communities
using nitrogen (δ15N) and carbon (δ13C) stable
v www.esajournals.org
3
June 2016 v Volume 7(6) v Article e01356
COMTE ET AL.
Environmental descriptors.—We quantified a
set of environmental variables expected to influence freshwater fish communities and patterns of
isotopic functional diversity at the global scale.
Based on previous research on macro-­ecological
patterns of freshwater fish taxonomic and functional diversity (Mason et al. 2008, Oberdorff
et al. 2011, Schleuter et al. 2012), we tested three
hypotheses related to: (1) climate history, (2) environmental filtering, and (3) niche partitioning
(see Appendix S2: Table S1 for details about the
hypotheses). To this end, we considered historical, energetic, climatic, and habitat size variables. Remotely sensed data were extracted from
existing environmental layers in a Geographic
Information System based on the geographical
coordinates of each fish community, which were
either reported in the original publications or determined using locality names. Historical factors
were calculated as the climatic conditions during
the last glacial maximum (CCSM3; Collins et al.
2006) minus the contemporary climatic conditions (Worldclim; Hijmans et al. 2005): quaternary anomalies in mean annual air temperature,
maximum temperature of the warmest month
and minimum temperature of the coldest month.
Contemporary climate was characterized by the
aforementioned temperature variables and also
included temperature stability (Worldclim; Hijmans et al. 2005). Productivity estimates in lakes
are not available at the macroecological scale but
patterns of ecosystem productivity in terrestrial and aquatic ecosystems are likely to covary
(Shurin et al. 2006). Consequently, energy availability was estimated using net terrestrial primary productivity (NPP; Imhoff et al. 2004) that was
calculated using a buffer of 10 km around the
sites to account for potentially diverse resource
inputs from the drainage network. Habitat size
was measured as lake surface area (km2; from
original publications or other sources).
variability. Isotopic functional diversity and
the degree of niche segregation were calculated
within the bi-­dimensional δ13C and δ15N isotopic space characterizing the range of resources
used and the number of trophic levels within a
community respectively (Layman et al. 2012).
Isotopic functional diversity therefore encompasses information on the structure of communities, the realized interactions among species
and the transfers of energy within communities,
whereas the degree of niche segregation is integrative of the extent of overlap in realized niche
and resource partitioning across species. To account for potential spatial or temporal variability in the isotopic composition of baselines across
food web, all species isotopic values were standardized according to their community centroid
following Schmidt et al. (2011).
At the community level, isotopic functional
­diversity was then calculated using the standard ellipse area (SEAc) based on population
mean isotopic values that unlike the convex
hull methods, is less sensitive to outliers (Jackson et al. 2011). At the species level, intraspecific niche variability was first quantified for each
species within each community from the SEAc
using intraspecific population variability along
the two isotopic axes. This represents a ­powerful
approach as it relies on measures of standard
deviation from the species centroid in two-­
dimensional isotopic space, a formulation that is
faithful to the concept of ecological niches (Bearhop et al. 2004, Newsome et al. 2007). As stable
isotope values of individual fish sampled in the
community were rarely reported, we used the
sample variance as an unbiased estimate of the
population variance along the two isotope axes.
Population isotopic niches were drawn using a
Monte Carlo simulation comprising 1000 random draws for each species of a given community according to two ­normal distributions based
on the sample mean and standard deviation of
δ13C and δ15N, and propagating an error of 10−3.
As the realized functional niche of a species is inherently community-­dependent as it can include
only the range of resources available in a given
community, the degree of niche segregation was
calculated as the ratio between each population
SEAc and the total community SEAc and averaged across species (i.e., mean value per community). Hence, this ratio quantifies the strength
Quantification of community structure and
functional diversity
For each community, we characterized: (1)
species richness, representing the number of locally interacting species, (2) isotopic functional
diversity, representing the amount of functional
niche space occupied by all co-­occurring species,
and (3) the degree of niche segregation, representing within-­community intraspecific niche
v www.esajournals.org
4
June 2016 v Volume 7(6) v Article e01356
COMTE ET AL.
of ecological interactions and local processes on
species coexistence, such as micro-­environmental
heterogeneity or competition, with values of zero
indicating complete niche segregation with non-­
overlapping niches (Pianka 1974, Violle et al.
2012).
of freshwater fish communities (Esposito-­Vinzi
et al. 2010). Path analysis seeks to elucidate the
causal processes underlying the observed relationships and estimates the relative importance
of alternative paths of influence. Specifically, we
tested the (1) direct effects of abiotic factors on
niche segregation, (2) direct effects of abiotic
factors on species richness and isotopic functional diversity, (3) direct effects between niche
segregation, species richness and isotopic functional diversity.
To limit model complexity while simultaneously gaining a better understanding of the underlying drivers of functional diversity, we grouped
correlated abiotic variables into latent variables
(historical factors, contemporary climate, energy availability, and habitat size) according to the
hypothesized causal mechanisms (see Appendix
S2 for details on model construction). Through
an iterative procedure, the algorithm calculates
the path coefficients (i.e., standardized beta coefficients between variables in the structural
model) and their significance via bootstrapping.
Non-­significant paths were then dropped until
the goodness of fit decreased (GOF) to obtain
the most parsimonious model (see Appendix S2:
Tables S2–S5 for details on model evaluation).
Lastly, to test whether the inferred causal mechanisms were affected by species introduction, potential differences in the strength and direction of
the path coefficients between communities with
and without nonnative species were assessed using bootstrap t-­tests (Chin 2003).
All statistical analyses were performed with R
(R Development Core Team 2015) using the siar
(Jackson et al. 2011) and plspm (Sanchez 2013)
packages.
Statistical analyses
Testing for redundancy.—In each ­community,
potential redundancy was estimated using a
null model approach that simulates randomly
assembled communities (Mouchet et al. 2010). In
each case, null distributions were generated by
randomly sampling 1000 times the same number
of species from a global species pool defined by
all species without distinction between native
and nonnative species and recalculating the metrics each time. Standardized effect sizes (SES)
were then calculated for each community by subtracting the mean of the null distribution from
the observed value and dividing by the standard deviation of the null distribution. SESs thus
quantified community functional redundancy
relative to the global species pool independently
from any sampling effect due to species richness.
Positive isotopic functional diversity SES values
were considered overdispersed, exhibiting higher diversity than expected by random assembly
as a consequence of limiting similarity, and negative SES values were considered underdispersed,
exhibiting lower diversity than expected under
random assembly as a consequence of environmental filtering (Mouillot et al. 2007). Similarly,
negative niche segregation SES values indicated
that species niches were more complementary than expected by chance, while positive SES
values indicated that species niches were more
aggregated and thus more redundant with one
another than expected by chance.
To assess the potential influence of nonnative
species on community functional redundancy,
we related the SESs for the degree of niche segregation and isotopic functional diversity to species
richness using a linear model including a pairwise interaction with nonnative species presence.
Determinants of community structure and functional diversity.—We used partial least squares
path modeling, a robust form of structural
equation modeling, to examine the relative
importance of external and internal factors to
predict the structure and functional diversity
v www.esajournals.org
Results
There was evidence for both overdispersion
and clustering on the degree of niche segregation and isotopic functional diversity (Fig. 2a).
We found a negative relationship between species richness and the degree of niche similarity
of the co-­existing species in native communities
but not in communities containing nonnative
species (F = 8.23, P < 0.01; Fig. 2b; Appendix S3:
Table S1). Similarly, while the isotopic functional
diversity increased with increasing species richness in communities with only native species,
5
June 2016 v Volume 7(6) v Article e01356
COMTE ET AL.
Fig. 2. (a) Functional structure of the communities described by the isotopic functional diversity (FiD) and
the degree of niche segregation among coexisting species (Niche). FiD and Niche were expressed as standardized
effect sizes (SES) relative to a null model of random community assembly. Positive FiDSES values indicate higher
diversity (overdispersion) and negative values indicate lower diversity (clustering) than random expectations.
Positive NicheSES values indicate niche aggregation and negative values indicate niche segregation. (b, c)
Relationships between niche segregation and isotopic functional diversity with species richness fitted using
linear models with an interaction term accounting for the presence of nonnative species. Blue indicates
communities with nonnative species and orange communities without nonnative species.
Species richness was positively associated with
contemporary climatic factors (0.33) and habitat
size (0.34), indicating that warm and climatically
stable areas, as well as larger habitats, harbor the
highest number of species. The negative
association between species richness and the
degree of niche segregation (−0.25) further
supports the role of niche specialization and
niche complementarity in promoting species
coexistence in communities. Similarly, the degree
of niche segregation appeared to promote isotopic
functional diversity (−0.63). Isotopic functional
diversity was also negatively correlated with
historical factors (−0.15), independently from
the effect of niche segregation. As a result,
we found that global patterns in fish taxonomic
and functional diversity along environment
gradients were the product of both external
abiotic factors and internal biotic-­mediated filters
(Fig. 4a, b).
indicating that communities were more functionally complementary than expected at random in species-­rich communities, communities
with nonnative species displayed higher level
of functional redundancy (F = 9.99, P < 0.01;
Fig. 2c; Appendix S3: Table S2).
We also found evidence that the structural
mechanisms underlying community structure
and functional diversity differed between
communities with and without nonnative species
(bootstrap t-­tests; P < 0.05; Appendix S2: Table S4).
In communities containing only native species,
the path model indicated that the degree of niche
segregation was significantly related to net
primary productivity (path coefficient = 0.41) and
contemporary climate (−0.27) and marginally
related to habitat size (−0.17, 95% CL: −0.35 to
0.06) (Fig. 3a). Species niches were therefore more
segregated in larger and less productive habitats
and under warmer and more stable climates.
v www.esajournals.org
6
June 2016 v Volume 7(6) v Article e01356
COMTE ET AL.
Fig. 3. Path diagram used to assess how the interplay between abiotic (historical, energetic, climatic, habitat
diversity) factors and the degree of niche segregation (Niche) shape species richness (SR) and isotopic functional
diversity (FiD) of global freshwater fish communities (a) without and (b) with nonnative species. Values are path
coefficients estimated by partial least squares path modeling analysis. The thickness of the arrows indicates
effect strength and asterisks represent statistical significance based on 95% percentile confidence intervals
calculated on 1000 bootstrap samples. Climate: contemporary temperature and stability; Historical: quaternary
climate anomalies since the Last Glacial Maximum; Habitat: lake surface area; Energy: net primary productivity.
FiD and Niche were expressed as standardized effect sizes (SES) relative to a null model of random community
assembly. Productivity, Habitat and SR were log-­transformed prior to analysis to linearize the relationships, but
the original names are kept for simplicity.
In contrast with native-­only communities,
those containing nonnative species appeared less
structured, as indicated by the decrease in the
GOF of the path model and the limited and
weaker significant associations with both abiotic
and biotic factors (Figs. 3b, 4c, d). In particular,
we found little evidence of strong associations
between abiotic factors, the degree of niche
segregation and the isotopic functional diversity,
or between the degree of niche segregation and
species richness. We also observed a decrease in
the strength of the relationships between
contemporary climate and species richness (0.24),
and between the degree of niche segregation
and the isotopic functional diversity (−0.32).
These results suggest that the presence of
nonnative species modify community structure
by reshuffling both the number and functional
identity of species within communities, in
addition to disrupting the internal organization
of biotic interactions.
v www.esajournals.org
Discussion
Our study highlights the importance of niche
segregation in determining the structure of
freshwater fish communities (Tedesco et al. 2007,
Mason et al. 2008, Matsuzaki et al. 2013) and
supports that habitat and environmental filters
interact with niche specialization through multiple pathways to mediate the strength of interspecific competition (Chesson 2000, Adler
et al. 2013). We found evidence that broad-­scale
variations in species richness and niche segregation were directly associated with contemporary climate. Although the causal mechanisms
underlying these patterns elude correlative
studies such as ours (and others), these results
are consistent with the hypothesis that warm
areas with stable climate are within the physiological range of more species, and that increase
specialization may amplify this underlying gradient (Currie et al. 2004). In turn, this may
7
June 2016 v Volume 7(6) v Article e01356
COMTE ET AL.
Fig. 4. Relative importance of both external abiotic factors (direct effect) and internal biotic-­mediated filters
(indirect effects) on (a–c) species richness and (b–d) isotopic functional diversity for communities (a, b) without
and (c, d) with nonnative species estimated by partial least squares path modeling analysis.
and the degree of niche segregation suggests that
elevated resource opportunities (e.g., availability and/or diversity of resources) may promote
coexistence of functionally similar species, and
therefore may promote greater functional redundancy at the community level. High overlap in
resource use might be well-­accommodated by
separation along other dimensions of the niche
(Schoener 1974) or through individual specialization (Araújo and Costa-­Pereira 2013). This might
allow more functional strategies to coexist within
local communities, thus enhancing functional
diversity. Our study provides the first support
for this finding at the global scale.
Our results indicate that the positive association between fish species richness and habitat
size may be the result of increased niche availability, as well as the addition of functionally differentiated species (Evans et al. 2005). Moreover,
the negative relationship between productivity
v www.esajournals.org
8
June 2016 v Volume 7(6) v Article e01356
COMTE ET AL.
be especially true for isotopic niches as species
with highly overlapping niches may also display microhabitat or temporal niche partitioning
(e.g., Codron et al. 2015). Coupling stable isotope
data with behavioral and habitat use information
should allow to gain a better understanding of
the mechanisms underlying the observed variation in species’ isotopic niches. Nonetheless, our
results still suggest that the role of ecological interactions in shaping the assembly and functional
diversity of multitrophic communities should
not be dismissed in macro-­ecological studies, as
their role may be discernable even at broad spatial scales.
We provide evidence supporting the hypothesis that historical processes have played a non-­
negligible role in determining the functional
diversity of natural communities at macroecological scales (Ordonez and Svenning 2015).
Whereas freshwater fish communities that have
been subjected to higher Quaternary climatic
anomalies showed signs of functional clustering,
communities in historical stable climatic areas
displayed more functionally dissimilar species.
This suggests that such effects are likely to occur
directly via filtering of unsuitable trophic strategies and/or differential postglacial colonization lags, rather than through historical legacies
on species richness (e.g., via a sampling effect).
This is surprising as historical factors, especially
Quaternary temperature anomalies, have been
previously proposed as important predictors of
species richness (Tedesco et al. 2005, Araújo et al.
2008). Nevertheless, at a global scale, paleoclimate variations are likely to covary with other
spatial gradients, which highlights the inherent
difficulty in distinguishing between historical
and environmental filters in shaping present-­day
distribution patterns of diversity. In addition,
other historical factors that were not included in
the present study, such as historical connectivity
and distance to climate refuges, are potentially
involved in the spatial variation in freshwater
species richness (Dias et al. 2014). Despite the
convincing patterns observed, further investigations are thus required to confirm our conclusions regarding the relative importance of these
underlying mechanisms and their consequences
on community functional diversity.
Retrospective analyses of species invasions
can provide new insight into the processes
v www.esajournals.org
­ nderlying community assembly (Sax et al.
u
2007). Here, we found that the taxonomic and
functional diversity of communities with nonnative species were poorly predicted by both abiotic and biotic factors; this contrasts with communities containing only native species where these
relationships were considerably stronger. Our
results thus support previous studies reporting
that distribution patterns of nonnative species
are more likely to reflect the context-­dependence
and ­idiosyncratic nature of the introduction (and
subsequent establishment), rather than natural processes (Blackburn et al. 2008, Olden et al.
2008). For instance, the weaker association between species richness and contemporary climate
implied that at the global scale, fish introductions
may not co-­vary with the biogeography of native
species (Leprieur et al. 2008). Similarly, the lack
of relationship between Quaternary climatic oscillations and functional diversity may indicate
that human-­mediated dispersal have overcome
historical legacies in shaping large-­scale distribution patterns of freshwater fish (Leprieur et al.
2009, Wilson et al. 2009).
We showed that species introductions may not
only alter the number and functional identity
of species, but could also disrupt the structure
of biotic interactions among co-­existing species.
In the presence of nonnative species, the degree
of niche segregation was unrelated to environmental and species richness gradients, and communities displayed higher levels of functional
redundancy. This may indicate that changes in
functional structure were associated with the
nonrandom selection of particular feeding strategies. For instance, Sagouis et al. (2015) suggested
that in lakes, humans tend to introduce omnivorous freshwater fish species, which may increase
overlap in species functional roles. An alternative
hypothesis might be that increased competition
with nonnative species may lead to increased diet
variation in order for species to maintain energy
requirements (Svanbäck and Bolnick 2007). It is
also likely that in such recently assembled communities, nonnative species have not evolved in
their environment for a sufficient long period for
co-­evolutionary adjustments to occur. Therefore,
it may still be expected that niche partitioning
play an important role in structuring fish communities in the presence of nonnative species.
Most communities have been the ­recipient of
9
June 2016 v Volume 7(6) v Article e01356
COMTE ET AL.
several nonnative species with different introduction histories and a further step would be to
investigate whether time since introduction affects the dynamic of species’ niches and to quantify the effects of assembly history on community
structure.
An important consideration of our study
is that our conclusions regarding the consequences of species introductions on community
­functional diversity does not benefit from comparable community structure data before and
after species ­introductions. Using the history of
fish species invasions in Japanese fresh waters,
Matsuzaki et al. (2013) found that nonnative
species ­increased the overall functional space of
historical communities, seemingly through the
selection of larger bodied-­size, piscivorous species with wider niche breadth. Similarly, Azzurro
et al. (2014) showed that successful marine fish
invaders tended to be morphologically differentiated from native species. We therefore cannot eliminate the possibility that the presence of
nonnative species was the consequence of other
forms of ecosystem changes, rather than being
the drivers of the observed change themselves
(Didham et al. 2005). It is important to note, however, that in this study we documented changes
at the local, rather than regional (e.g., watershed)
scale, and using species realized niches instead
of fixed traits-­values across communities. This
suggests that nonnative species might induce a
greater functional overlap even if they show differences in their mean ecomorphological traits
(see also Walsworth et al. 2013).
There has been a greater application of trait-­
based approaches to tackle questions related to
community assembly and diversity (Mouchet
et al. 2010, Adler et al. 2013, Violle et al. 2014).
Although such approaches are well-­suited to
this challenge, by often describing the potential
niche of species based on a priori functional attributes and by disregarding intraspecific variability, trait-­based methods may confound the
interpretation of the eco-­evolutionary dynamics
of communities and associated ecosystem functions (Farias and Jaksic 2009, Jung et al. 2010,
Violle et al. 2012). By integrating information
about the realized niche of interacting species at
a local scale, we were able to document changes in several aspects of community structure
along resource, species richness, and ­species
v www.esajournals.org
10
i­ ntroduction gradients. However, potential
variations in basal resources across communities may render the interpretation of changes
in isotopic space sometimes difficult and this is
because differences in species’ isotopic niches
may be driven by baseline variations that could
be linked to environmental conditions (Hoeinghaus and Zeug 2008). Although some challenges
remain, the development of new analytical procedures (e.g., Cucherousset and Villéger 2015,
Rigolet et al. 2015) coupled with mathematical
null models may offer a promising avenue to reveal species coexistence processes and assembly
rules driven by classical niche-­based coexistence
theory, and to improve our understanding of the
functional consequences of non native species introductions.
Acknowledgments
We are grateful to all the authors who shared their
stable isotope data to make our study possible and
to Xingli Giam, Gaël Grenouillet and two anonymous
reviewers whose comments improved the manuscript.
Financial support was provided by a H. Mason Keeler
Endowed Professorship (School of Aquatic and Fishery
Sciences, University of Washington) to JDO (also
supporting LC) and by an “ERG Marie Curie” grant
(PERG08-­GA-­2010-­276969) to JC in the lab EDB, part
of the Laboratoire d’Excellence (LABEX) entitled TULIP
(ANR-­10-­LABX-­41).
Literature Cited
Adler, P. B., A. Fajardo, A. R. Kleinhesselink, and N.
J. B. Kraft. 2013. Trait-­based tests of coexistence
mechanisms. Ecology Letters 16:1294–1306.
Albouy, C., F. Guilhaumon, S. Villéger, M. Mouchet, L.
Mercier, J. M. Culioli, J. A. Tomasini, F. Le Loc’h,
and D. Mouillot. 2011. Predicting trophic guild
and diet overlap from functional traits: statistics,
opportunities and limitations for marine ecology.
Marine Ecology Progress Series 436:17–28.
Araújo, M. S., and R. Costa-Pereira. 2013. Latitudinal
gradients in intraspecific ecological diversity. Biology Letters 9:20130778.
Araújo, M. B., D. Nogués-Bravo, J. A. F. Diniz-­
Filho, A. M. Haywood, P. J. Valdes, and C. Rahbek.
2008. Quaternary climate changes explain diversity among reptiles and amphibians. Ecography
31:8–15.
Azzurro, E., V. M. Tuset, A. Lombarte, F. Maynou, D.
Simberloff, A. Rodríguez-Pérez, and R. V. Solé.
June 2016 v Volume 7(6) v Article e01356
COMTE ET AL.
Evans, K. L., P. H. Warren, and K. J. Gaston. 2005.
Species-­energy relationships at the macroecological scale: a review of the mechanisms. Biological
Reviews 80:1–25.
Farias, A. A., and F. M. Jaksic. 2009. Hierarchical determinants of the functional richness, evenness and
divergence of a vertebrate predator assemblage.
Oikos 118:591–603.
Froese, R., and D. Pauly. 2015. FishBase. www.­fishbase.
org
González-Bergonzoni, I., M. Meerhoff, T. A. Davidson,
F. Teixeira-de Mello, A. Baattrup-Pedersen, and E.
Jeppesen. 2012. Meta-­analysis shows a consistent
and strong latitudinal pattern in fish omnivory
across ecosystems. Ecosystems 15:492–503.
Gotelli, N. J., and A. E. Arnett. 2000. Biogeographic effects of red fire ant invasion. Ecology Letters
3:257–261.
Hijmans, R. J., S. E. Cameron, J. L. Parra, P. G. Jones,
and A. Jarvis. 2005. Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology 25:1965–1978.
Hobbs, R. J., et al. 2006. Novel ecosystems: theoretical
and management aspects of the new ecological world
order. Global Ecology and Biogeography 15:1–7.
Hoeinghaus, D. J., and S. C. Zeug. 2008. Can stable isotope ratios provide for community-­wide measures
of trophic structure? Comment. Ecology 89:2353–
2357.
Imhoff, M. L., L. Bounoua, T. Ricketts, C. Loucks, R.
Harriss, and W. T. Lawrence. 2004. Global patterns
in human consumption of net primary production.
Nature 429:870–873.
Jackson, A. L., R. Inger, A. C. Parnell, and S. Bearhop.
2011. Comparing isotopic niche widths among
and within communities: SIBER – Stable Isotope
Bayesian Ellipses in R. Journal of Animal Ecology
80:595–602.
Jung, V., C. Violle, C. Mondy, L. Hoffmann, and S.
Muller. 2010. Intraspecific variability and trait-­
based community assembly. Journal of Ecology
98:1134–1140.
Layman, C. A., et al. 2012. Applying stable isotopes to
examine food-­web structure: an overview of analytical tools. Biological Reviews 87:545–562.
Leprieur, F., O. Beauchard, S. Blanchet, T. Oberdorff,
and S. Brosse. 2008. Fish invasions in the world’s
river systems: when natural processes are blurred
by human activities. PLoS Biology 6:e28.
Leprieur, F., J. D. Olden, S. Lek, and S. Brosse. 2009.
Contrasting patterns and mechanisms of spatial
turnover for native and exotic freshwater fish in
Europe. Journal of Biogeography 36:1899–1912.
LWP-2. 2005. Global human footprint dataset
(­Geographic). Last of the Wild Project, Version 2
2014. External morphology explains the success
of biological invasions. Ecology Letters 17:1455–
1463.
Bearhop, S., C. E. Adams, S. Waldron, R. A. ­Fuller,
and H. MacLeod. 2004. Determining trophic
niche width: a novel approach using stable
isotope analysis. Journal of Animal Ecology
73:1007–1012.
Blackburn, T. M., P. Cassey, and J. L. Lockwood. 2008.
The island biogeography of exotic bird species.
Global Ecology and Biogeography 17:246–251.
Cavender-Bares, J., K. H. Kozak, P. V. A. Fine, and S. W.
Kembel. 2009. The merging of community ecology
and phylogenetic biology. Ecology Letters 12:693–
715.
Chesson, P. 2000. Mechanisms of maintenance of species diversity. Annual Review of Ecology and Systematics 31:343–366.
Chin, W. 2003. A permutation procedure for multigroup comparison of PLS models. Pages 33–43 in
M. Vilares, M. Tenenhaus, P. Coelho, V. Esposito-Vinzi and A. Morineau, editors. PLS and related
methods. Proceedings of the International Symposium PLS03. Decisia, Paris, France.
Codron, J., K. J. Duffy, N. L. Avenant, M. Sponheimer,
J. Leichliter, O. Paine, P. Sandberg, and D. Codron.
2015. Stable isotope evidence for trophic niche partitioning in a South African savanna rodent community. Current Zoology 61:397–411.
Collins, W. D., et al. 2006. The community climate system model version 3 (CCSM3). Journal of Climate
19:2122–2143.
Cucherousset, J., and S. Villéger. 2015. Quantifying the
multiple facets of trophic diversity: new integrative metrics for stable isotope ecology. Biological
Indicators 56:152–160.
Currie, D. J., et al. 2004. Predictions and tests of
climate-­based hypotheses of broad-­scale variation
in ­taxonomic richness. Ecology Letters 7:1121–1134.
Dias, M. S., T. Oberdorff, B. Hugueny, F. Leprieur, C.
Jézéquel, J.-F. Cornu, S. Brosse, G. Grenouillet, and
P. A. Tedesco. 2014. Global imprint of historical
connectivity on freshwater fish biodiversity. Ecology Letters 17:1130–1140.
Didham, R. K., J. M. Tylianakis, M. A. Hutchison, R.
Ewers, and N. J. Gemmel. 2005. Are invasive species the drivers of ecological change? Trends in
Ecology & Evolution 20:471–474.
Esposito-Vinzi, V., L. Trinchera, and S. Amato. 2010.
PLS path modeling: from foundations to recent developments and open issues for model assessment
and improvement. Pages 47–83 in V. E. Vinzi, W.
Chin, J. Henseler and H. Wang, editors. Handbook
of partial least squares. Springer-Verlag, Berlin,
Heidelberg, Germany.
v www.esajournals.org
11
June 2016 v Volume 7(6) v Article e01356
COMTE ET AL.
ing of benthic communities. Functional Ecology
29:1350–1360.
Sagouis, A., J. Cucherousset, S. Villéger, F. Santoul, and
S. Boulêtreau. 2015. Non-­native species modify the
isotopic structure of freshwater fish communities
across the globe. Ecography 38:979–985.
Sanchez, G. 2013. PLS path modeling with R. Trowchez Editions, Berkeley, California, USA.
Sanders, N. J., N. J. Gotelli, N. E. Heller, and D. M. Gordon. 2003. Community disassembly by an invasive
species. Proceedings of the National Academy of
Sciences USA 100:2474–2477.
Sax, D. F., et al. 2007. Ecological and evolutionary insights from species invasions. Trends in Ecology &
Evolution 22:465–471.
Schleuter, D., M. Daufresne, J. Veslot, N. W. H. Mason, C. Lanoiselée, S. Brosse, O. Beauchard, and
C. ­Argillier. 2012. Geographic isolation and climate govern the functional diversity of native fish
communities in European drainage basins. Global
Ecology and Biogeography 21:1083–1095.
Schmidt, S. N., C. J. Harvey, and M. J. Vander Zanden.
2011. Historical and contemporary trophic niche
partitioning among Laurentian Great Lakes coregonines. Ecological Applications 21:888–896.
Schoener, T. W. 1974. Resource partitioning in ecological communities. Science 185:27–39.
Shurin, J. B., D. S. Gruner, and H. Hillebrand. 2006. All wet
or dried up? Real differences between aquatic and terrestrial food webs. Proceedings of the Royal Society of
London Series B: Biological Sciences 273:1–9.
Svanbäck, R., and D. I. Bolnick. 2007. Intraspecific
competition drives increased resource use diversity within a natural population. Proceedings of the
Royal Society of London Series B: Biological Sciences 274:839–844.
Swenson, N. G. 2013. The assembly of tropical tree
communities – the advances and shortcomings of
phylogenetic and functional trait analyses. Ecography 36:264–276.
Tedesco, P. A., T. Oberdorff, C. A. Lasso, M. Zapata,
and B. Hugueny. 2005. Evidence of history in explaining diversity patterns in tropical riverine fish.
Journal of Biogeography 32:1899–1907.
Tedesco, P. A., et al. 2007. Local-­scale species-­energy
relationships in fish assemblages of some forested
streams of the Bolivian Amazon. Comptes Rendus
– Biologies 330:255–264.
Villéger, S., N. W. H. Mason, and D. Mouillot. 2008.
New multidimensional functional diversity indices
for a multifaceted framework in functional ecology. Ecology 89:2290–2301.
Violle, C., et al. 2012. The return of the variance: intraspecific variability in community ecology. Trends
in Ecology & Evolution 27:244–252.
(LWP-2). NASA Socioeconomic Data and Applications Center (SEDAC), Palisades, New York, USA.
Mason, N. W. H., P. Irz, C. Lanoiselée, D. Mouillot,
and C. Argillier. 2008. Evidence that niche specialization explains species-­energy relationships in
lake fish communities. Journal of Animal Ecology
77:285–296.
Matsuzaki, S. I. S., T. Sasaki, and M. Akasaka. 2013.
Consequences of the introduction of exotic and
translocated species and future extirpations on the
functional diversity of freshwater fish assemblages.
Global Ecology and Biogeography 22:1071–1082.
Mouchet, M. A., S. Villéger, N. W. H. Mason, and D.
Mouillot. 2010. Functional diversity measures: an
overview of their redundancy and their ability to
discriminate community assembly rules. Functional Ecology 24:867–876.
Mouillot, D., O. Dumay, and J. A. Tomasini. 2007.
Limiting similarity, niche filtering and functional
diversity in coastal lagoon fish communities. Estuarine, Coastal and Shelf Science 71:443–456.
Mouquet, N., J. L. Moore, and M. Loreau. 2002. Plant
species richness and community productivity: why
the mechanism that promotes coexistence matters.
Ecology Letters 5:56–65.
Newsome, S. D., C. Martinez del Rio, S. Bearhop, and
D. L. Phillips. 2007. A niche for isotopic ecology.
Frontiers in Ecology and the Environment 5:429–
436.
Oberdorff, T., P. A. Tedesco, B. Hugueny, F. Leprieur,
O. Beauchard, S. Brosse, and H. H. Dürr. 2011.
Global and regional patterns in riverine fish species richness: a review. International Journal of
Ecology 2011:1–12.
Olden, J. D., N. L. Poff, M. R. Douglas, M. E. Douglas,
and K. D. Fausch. 2004. Ecological and evolutionary consequences of biotic homogenization. Trends
in Ecology & Evolution 19:18–24.
Olden, J. D., M. J. Kennard, and B. J. Pusey. 2008. Species invasions and the changing biogeography of
Australian freshwater fishes. Global Ecology and
Biogeography 17:25–37.
Ordonez, A., and J.-C. Svenning. 2015. Geographic patterns in functional diversity deficits are linked to
glacial-­interglacial climate stability and accessibility. Global Ecology and Biogeography 24:826–837.
Pianka, E. R. 1974. Niche overlap and diffuse competition. Proceedings of the National Academy of Sciences USA 71:2141–2145.
R Development Core Team. 2015. R: a language and
environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.
Rigolet, C., E. Thiébaut, A. Brind’Amour, and S. F. Dubois. 2015. Investigating isotopic functional indices
to reveal changes in the structure and function-
v www.esajournals.org
12
June 2016 v Volume 7(6) v Article e01356
COMTE ET AL.
Wilson, J. R. U., E. E. Dormontt, P. J. Prentis, A. J. Lowe,
and D. M. Richardson. 2009. Something in the way
you move: dispersal pathways affect invasion success. Trends in Ecology & Evolution 24:136–144.
Zhao, T., S. Villéger, S. Lek, and J. Cucherousset. 2014.
High intraspecific variability in the functional
niche of a predator is associated with ontogenetic shift and individual specialization. Ecology and
Evolution 4:4649–4657.
Violle, C., P. B. Reich, S. W. Pacala, B. J. Enquist,
and J. Kattge. 2014. The emergence and promise
of functional biogeography. Proceedings of the
National Academy of Sciences USA 111:13690–
13696.
Walsworth, T. E., P. Budy, and G. P. Thiede. 2013. Longer food chains and crowded niche space: effects of
multiple invaders on desert stream food web structure. Ecology of Freshwater Fish 22:439–452.
Supporting Information
Additional Supporting Information may be found online at: http://onlinelibrary.wiley.com/doi/10.1002/
ecs2.1356/supinfo v www.esajournals.org
13
June 2016 v Volume 7(6) v Article e01356