Projected compositional shifts and loss of ecosystem …. R. Biswas (&) Faculty of Natural Resources...
Transcript of Projected compositional shifts and loss of ecosystem …. R. Biswas (&) Faculty of Natural Resources...
PRIMARY RESEARCH PAPER
Projected compositional shifts and loss of ecosystem servicesin freshwater fish communities under climate changescenarios
Shekhar R. Biswas . Richard J. Vogt . Sapna Sharma
Received: 3 November 2016 / Revised: 17 April 2017 / Accepted: 18 April 2017
� Springer International Publishing Switzerland 2017
Abstract What are the projected impacts of climate
change on community composition and consequen-
tially on the distribution of functional traits? Answers
to these questions are somewhat unclear but critical for
designing ecological management strategies. Here we
forecast potential impacts of climate change on
freshwater lake fish communities of Ontario, Canada,
by contrasting species composition, species richness,
functional diversity and functional composition for
present versus projected communities under ‘‘best-
case’’ and ‘‘business-as-usual’’ climate change sce-
narios. Results indicate that the composition of
projected communities differs from present, and
includes a shift from cold- and cool-water species to
warm-water species. Species richness in projected
communities is estimated to increase by 60–81%, but
functional diversity is estimated to decline. These
projected communities are estimated to have on
average 22% shorter mean body length, 38% lighter
body weight and 36% less fecundity than present.
Also, the present configuration of sport and commer-
cially important fishes are projected to decline in their
distribution, potentially impacting ecosystem services
associated with commercial and recreational fisheries.
Together, climate change may initiate a compositional
shift that may result in an important shift in commu-
nity functional structure, which is likely to affect
important aquatic ecosystem services.
Keywords Compositional shift � Ecosystemservices � Species distribution models � Trait analysis
Introduction
Broad-scale shifts in species distributions in response
to climate change are well documented (Parmesan &
Yohe, 2003; Chen et al., 2011) and are typically a
result of species having to respond to local environ-
ments according to their thermal habitat requirements
(Tonn & Magnuson, 1982). Of course, changes in
species distributions will have an impact on local
community dynamics, composition and diversity
Communicated by Handling editor: Begona Santos
Electronic supplementary material The online version ofthis article (doi:10.1007/s10750-017-3208-1) contains supple-mentary material, which is available to authorized users.
S. R. Biswas � S. Sharma
Department of Biology, York University, 4700 Keele
Street, Toronto, ON M3J 1P3, Canada
S. R. Biswas (&)
Faculty of Natural Resources Management, Lakehead
University, 955 Oliver Road, Thunder Bay, ON P7B 5E1,
Canada
e-mail: [email protected]
R. J. Vogt
Department of Biological Sciences, University of Quebec
at Montreal, CP 8888, Succ. Centre Ville, Montreal,
QC H3C 3P8, Canada
123
Hydrobiologia
DOI 10.1007/s10750-017-3208-1
(Walther, 2010; Simpson et al., 2011) via ecological
interactions that result as species begin to invade novel
habitats. Since each species can play relatively unique
ecological functions and deliver somewhat unique
ecological services (Dıaz et al., 2007), any such
change in local species composition or diversity
should impact ecosystem functions and services
(Cardinale et al., 2012; Hooper et al., 2012), or the
resilience of ecological communities (Oliver et al.,
2015a, b). These community-scale consequences of
climate change are still somewhat poorly understood
but important for designing ecological management
strategies for the continued provision of important
ecosystem services.
One way to anticipate the impact of climate change
on community-scale ecosystem functions and services
is to project future communities under climate change
scenarios (Guisan & Thuiller, 2005; Botkin et al.,
2007), followed by comparative analyses of commu-
nity trait structure for present versus projected com-
munities. Here functional traits allow quantification of
species characteristics (i.e. morphological, structural
and behavioural.) that influence species performance
or fitness (Violle et al., 2007; Nock et al., 2016). By
emphasizing the characteristics that define how organ-
isms interact with one another and their changing
environments, a functional trait approach can provide
insights into the ecological consequences of regional
shifts in species distributions (Hooper et al., 2005). In
fact, as the pattern of trait distribution of a lake
changes, inferences can be made regarding systematic
changes in ecosystem functioning or provision of
ecosystem services (Civantos et al., 2012; Thuiller
et al., 2014; Barbet-Massin & Jetz, 2015; Gauzere
et al., 2015; Mokany et al., 2015). For example, a
large-scale shift in species’ size or body mass distri-
butions will have ramifications for overall ecosystem
productivity (Millennium-Ecosystem-Assessment,
2005). Further, future trait distributions can be fore-
casted once present day relationships between traits
and local environments are understood (Mokany et al.,
2016).
Here we focus on modelling freshwater fish species
and trait distributions, which define how co-occurring
species interact with one another and their habitats,
across a landscape of lakes and for various regional
climate warming scenarios. Changes in freshwater fish
species distributions in response to warming climates
and changing availability of suitable thermal habitat
have been well documented in lakes (Shuter et al.,
1980; Magnuson et al., 1990, 1997; Alofs et al., 2014).
Typically, coldwater fishes are defined as fishes with
an optimal thermal range between 8 and 12�C, cool-water fishes as fishes with an optimal thermal range
between 16 and 20�C, and warm-water fishes as fishes
with an optimal thermal range between 22 and 26�C(Christie & Regier, 1988; Magnuson et al., 1990).
Under warming climates, warm-water fishes, such as
centrarchids (e.g. smallmouth bass and rock bass),
have invaded lakes at the northern extent of their range
(Alofs et al., 2014), and coldwater fishes, such as
cisco, have become extirpated from lakes near the
southern extents of their range (Sharma et al., 2011).
Species range shifts like these are predicted to become
exacerbated under scenarios of climate change with
forecasts of range expansion of predatory warm-water
fishes, northern shifts of cool-water fishes and extir-
pation of coldwater fishes from inland lakes (Chu
et al., 2005; Sharma et al., 2007; Van Zuiden et al.,
2016). Ramifications for the remaining fishes in the
community are still poorly understood.
We therefore aim to understand how changing
climates may impact freshwater fish community
composition and functional trait distributions across
a landscape of lakes to assess the impacts of climate
change on (i) species composition, (ii) species rich-
ness, (iii) functional diversity and (iv) functional
composition. We compared present day communities
with those forecast under business-as-usual and best-
case climate change scenarios in 2070 using a
700-lake dataset from Ontario, Canada. This region
is compositionally dynamic, as it includes both the
present northern range limit of warm-water fishes and
the southern extent of the range of cold-water species
(Chu et al., 2005; Sharma et al., 2007; Van Zuiden
et al., 2016). By 2070, as warm-water habitats become
more prevalent, we expect that species richness will
increase in many lakes as new species begin to invade
what were once predominantly colder habitats. Cor-
respondingly, lake functional diversity will increase if
warm-water invaders introduce new functional char-
acteristics to community trait distributions in lakes
altered by a warming climate. Further, we explore the
ramifications of changes to fish community structure
and trait distributions for provision of ecosystem
services that may result from climate warming.
Hydrobiologia
123
Methods
Study area and data
The study area is inland lakes in Ontario, Canada
(Fig. 1). These lakes support a wide range of com-
mercial and recreation fisheries including walleye
(Sander vitreus), whitefish (Coregonus clupeiformis),
perch (Perca flavescens), pike (Esox Lucius) and lake
trout (Salvelinus namaycush). The Ontario Ministry of
Natural Resources and Forestry Broad-scale Monitor-
ing (OMNRF-BsM) Program aims to collect data on
*700 Ontario lakes every five years to inform
landscape level management; and to-date has col-
lected detailed information on lake fish communities
and lake characteristics from 2008 to 2012 for over
700 lakes (Sandstrom et al., 2010). Data that were
collected included latitude, longitude, mean lake
depth, lake surface area, secchi depth, pH and
presence/absence of 129 fish species. Fish data were
collected between June–September, using both large-
and small-mesh gillnets (Sandstrom et al., 2010).
We obtained the most recent (2008–2012) geo-
referenced fish community dataset from OMNRF-
BsM program for 722 inland lakes. Our final lake
dataset consisted of 645 lakes, following the removal
of lakes that were species poor (species richness was
\5% of total richness in the dataset). We also removed
rare species (occurrence in less than 5% of lakes)
resulting in a regional species richness of 40 species
(i.e. a total of 40 species were included in the final
dataset). Rare species and sites can have a dispropor-
tionate influence on statistical analysis (Jackson &
Harvey, 1989).
For the same set of lakes, we obtained historical
climate data (monthly mean temperature and precip-
itation) represented as climate averages from 1950 to
2000 fromWorldClim database (Hijmans et al., 2005),
Fig. 1 Map of Ontario highlighting study lakes and lake-wise present mean annual air temperature (�C) averaged over the period of
1950–2000. Climatic data were obtained from the WorldClim database, with a spatial resolution of 30 arc-seconds (*1 km)
Hydrobiologia
123
with a resolution of 30 arc-seconds (*1 km). When
deriving climate estimates and estimating future
climate change projections, the IPCC recommends
the use of climate normals from 1950 to 2000 to reduce
the influence of inter-annual variability in weather on
climate forecasts (IPCC, 2013). By using the averaged
climate over decades, we are focusing on the effects of
climate variability on fish populations, rather than
weather. The timeline difference for climate data
(1950–2000) and fish data (2008–2012) should not be
a problem because species distributional shift is a
gradual process; this climatic baseline is widely used
in the species distribution modelling; and presently,
this is the most updated climate normal (at a fine
spatial resolution of 1 km2) we obtained. The most
conservative greenhouse gas emissions or ‘‘best-case’’
scenario (Representative Concentration Pathway,
RCP = 2.6) and ‘‘business-as-usual’’ greenhouse gas
emissions scenarios (RCP = 8.5) were obtained from
the Canadian Centre for Climate Modelling and
Analysis (CCMA) CMIP5 (IPPC Fifth Assessment)
Global Climate Model (GCM) for 2070. Here the year
2070 represents climate projections averaged for
2060–2080. The mean annual air temperatures for
our study lakes are predicted to increase by
1.81 ± 1.35�C for RCP 2.6 and 3.63 ± 1.31�C for
RCP 8.5 (Fig. S1 in Electronic Supplementary
Materials).
Projecting species composition and generating
future communities
Generalized linear models were developed for all 40
fish species to understand the environmental factors
structuring the occurrence of each species in the
community (Table 1). Eighty percent of the dataset
was used to train the models (N = 516 lakes), and the
remaining 20% of the data were used to validate the
models (N = 129 lakes). That is, we first randomly
split the whole dataset into training and validation sets;
and then, the same training dataset was used for
modelling the occurrence of each species and the same
validation dataset was used for validating the occur-
rence of each species. While the response variable was
the presence/absence of each species, explanatory
variables included lake surface area, mean depth,
secchi depth, seasonal mean air temperatures, seasonal
precipitation and mean annual air temperatures and
precipitation. These environmental and climatic
variables are known to be relevant for structuring
Ontario fish communities (Jackson & Harvey, 1989).
Environmental variables were tested for normality and
multicollinearity (see Electronic Supplementary
Materials, Fig. S2). Forward selection with a dual
criterion (a = 0.05 and R2adj) was used to identify
significant environmental predictor variables for each
species (Blanchet et al., 2008).
Receiver Operating Characteristics (ROC) curves
were used to identify thresholds that maximize the
sensitivity (ability to correctly predict species pres-
ence) and specificity (ability to correctly predict
species absence) for each species distribution model.
This procedure is recommended when species pres-
ences and absences are not equal within the data
(Sharma & Jackson Sharma & Jackson, 2008). We
retained the best model for each species based on the
lowest Akaike Information Criterion (AIC) value.
Following the development and validation of species
distribution models and appropriate ROC thresholds
for each species, we projected the generalized linear
model in 2070 using the most conservative scenario
(RCP = 2.6) and business-as-usual scenario
(RCP = 8.5) of the CCMAGCM. Finally, by stacking
the predicted occurrence of all 40 species in each lake,
lake-wise future communities were generated (Guisan
& Zimmermann, 2000; Dubuis et al., 2011; Guisan &
Rahbek, 2011). One of the major advantages of this
stacking approach to construct lake-wise future com-
munities is that they yield lake-specific species
composition (Zurell et al., 2016), which is essential
for the follow up functional trait analyses involving
both quantitative and categorical traits. However, our
bio-climate based projections essentially reflect the
aspect of habitat suitability (Araujo & Peterson, 2012)
and did not consider other constrains that could also
limit species occurrence at a particular lake, such as
biotic interactions, dispersal or historical factors
(Guisan & Rahbek, 2011; Zurell et al., 2016). We
therefore evaluated our projection efficiency (model
uncertainty) and found satisfactory at the level of
individual species (sensitivity range: 0.50–0.86; speci-
ficity range: 0.51–0.94; see Table 1) but over predic-
tion for species richness (correlation between
projected richness and actual richness, r = 0.61,
n = 129 validation lakes), which is a well acknowl-
edged phenomenon in the stacking approach of
community generation (Zurell et al., 2016). As such,
our projected communities should be considered a
Hydrobiologia
123
Table
1Species-wiselogisticmodelsforpredictingfuture
distributionsoffishes
ininlandlakes
ofOntario
Species
Localnam
eModel
usedto
predictfuture
presence
Optimal
threshold
Sensitivity
Specificity
Noof
occurrence
Salvelinusfontinalis(M
itchill,1814)
Brooktrout
1=ð1
þe�
ð11:87�0:64tm
eanjja�2:23logðsaÞþ3:41logðsdÞÞÞ
0.20
0.60
0.92
64
Salvelinusnamaycush
(Walbaum,1792)
Laketrout
1=ð1
þe�
ð�2:27�0:42tm
eanjjaþ7:04logðm
dÞþ
5:20logðsdÞÞÞ
0.47
0.82
0.83
286
Coregonusclupeiform
is(M
itchill,1818)
Lakewhitefish
1=ð1
þe�
ð0:61þ1:92logðsaÞ�0:38tm
eanjjaþ0:70logðm
dÞÞÞ
0.56
0.72
0.72
360
CoregonusartediLesueur,1818
Cisco
1=ð1
þe�
ð�4:44þ1:73logðsaÞþ2:29logðm
d�
2:53logðsdÞ�
0:17tm
eanatÞ Þ
0.57
0.82
0.68
402
Osm
erusmordax(M
itchill,1814)
Rainbow
smelt
1=ð1
þe�
ð�19:71þ2:39logðm
dÞþ
0:08prec s
onþ0:93logðsaÞþ0:37tm
eanjjaÞ Þ
0.11
0.55
0.82
62
EsoxluciusLinnaeus,1758
Northernpike
1=ð1
þe�
ð�0:80þ1:81logðsaÞ�0:24tm
eanat�4:19logðsdÞÞÞ
0.71
0.80
0.63
488
EsoxmasquinongyMitchill,1824
Muskellunge
1=ð1
þe�
ð�22:18þ0:96tm
eanjjaþ1:18logðsaÞ�2:25logðm
dÞþ
2:87logðsdÞÞÞ
0.22
0.67
0.93
42
Catostomuscatostomus(Forster,1773)
Longnose
sucker
1=ð1
þe�
ð8:91�0:76tm
eanjjaþ2:42logðm
d�
1:53logðsdÞÞÞ
0.17
0.50
0.85
77
Catostomuscommersonii(Lacepede,1803)
Whitesucker
1=ð1
þe�
ð2:55�0:54tm
eansonþ1:34logðsaÞÞ Þ
0.91
0.88
0.57
617
Moxostomamacrolepidotum
(Lesueur,
1817)
Shorthead
redhorse
1=ð1
þe�
ð�13:64þ1:43logðsaÞþ0:10prec jja�1:59logðm
dÞÞÞ
0.20
0.64
0.88
83
ChrosomuseosCope,
1861
Northernredbelly
dace
1=ð1
þe�
ð�3:25�2:03logðsaÞþ0:07prec s
onÞ Þ
0.16
0.50
0.91
33
Couesiusplumbeus(A
gassiz,
1850)
Lakechub
1=ð1
þe�
ð5:76þ3:00logðsdÞ�
0:57tm
eanjjaÞ Þ
0.14
0.67
0.69
87
Notemigonuscrysoleucas(M
itchill,1814)
Golden
shiner
1=ð1
þe�
ð�12:52þ0:84tm
eanjja�0:81logðm
d�
0:75logðsaÞÞ Þ
0.23
0.61
0.72
146
Notropisatherinoides
Rafinesque,
1818
Emeraldshiner
1=ð1
þe�
ð�2:96þ1:15logðsaÞ�1:28logðsdÞ�
0:22tm
eansonÞÞÞ
0.29
0.66
0.79
147
Luxiluscornutus(M
itchill,1817)
Commonshiner
1=ð1
þe�
ð�7:63þ0:07prec s
onþ1:20logðm
dÞÞÞ
0.17
0.70
0.64
146
Notropisheterodon(Cope,
1865)
Blackchin
shiner
1=ð1
þe�
ð�20:94þ1:07tm
eanjjaÞ Þ
0.05
0.75
0.56
45
Notropisheterolepis
Eigenmann&
Eigenmann,1893
Blacknose
shiner
1=ð1
þe�
ð�2:16�0:15tm
eanatþ1:86logðsdÞÞÞ
0.17
0.67
0.51
114
Notropishudsonius(Clinton,1824)
Spottailshiner
1=ð1
þe�
ð�1:81þ1:72logðsaÞ�1:54logðm
d�
0:38tm
eansonÞ Þ
0.47
0.69
0.69
320
Notropisvolucellus(Cope,
1865)
Mim
icshiner
1=ð1
þe�
ð�22:05þ0:13prec jjaþ2:39logðsdÞþ
0:43tm
eanjjaÞ Þ
0.10
0.75
0.79
41
Pimephalesnotatus(Rafinesque,
1820)
Bluntnose
minnow
1=ð1
þe�
ð�17:71þ0:79tm
eanjjaþ1:30logðm
dÞþ
2:58logðsdÞÞÞ
0.24
0.68
0.79
102
PimephalespromelasRafinesque,
1820
Fatheadminnow
1=ð1
þe�
ð7:61�1:79logðsaÞ�0:49tm
eanjjaþ3:71logðsdÞÞÞ
0.07
0.71
0.81
37
Sem
otilusatromaculatus(M
itchill,1818)
Creek
chub
1=ð1
þe�
ð�3:90�2:65logðsaÞþ0:06prec s
onþ2:75logðm
dÞÞÞ
0.05
0.75
0.80
37
Margariscusmargarita
(Cope,
1867)
Pearldace
1=ð1
þe�
ð�1:29þ3:02logðsdÞ�
0:76logðsaÞ�0:23tm
eansonÞ Þ
0.07
0.70
0.69
42
Ameiurusnebulosus(Lesueur,1819)
Brownbullhead
1=ð1
þe�
ð�16:15þ0:88tm
eanjjaÞÞÞ
0.28
0.72
0.85
136
Lota
lota
(Linnaeus,1758)
Burbot
1=ð1
þeð
�ð�
4:67þ2:01logðsaÞþ3:76logðm
d�
0:29tm
eanjjaÞ Þ
0.47
0.70
0.78
266
Pungitiuspungitius(Linnaeus,1758)
Ninespine
stickleback
1=ð1
þe�
ð�8:05þ1:46logðsaÞþ3:11logðsdÞ�
0:22tm
eanatÞ Þ
0.11
0.71
0.81
55
Hydrobiologia
123
Table
1continued
Species
Localnam
eModel
usedto
predictfuture
presence
Optimal
threshold
Sensitivity
Specificity
Noof
occurrence
Percopsisomiscomaycus(W
albaum,1792)
Troutperch
1=ð1
þe�
ð�2:61þ1:93logðsaÞ�3:19logðsdÞ�
0:32tm
eansonÞ Þ
0.48
0.79
0.74
282
Ambloplitesrupestris
(Rafinesque,
1817)
Rock
bass
1=ð1
þe�
ð�17:93þ0:92tm
eanjjaþ0:91logðsaÞÞ Þ
0.54
0.70
0.81
294
Lepomisgibbosus(Linnaeus,1758)
Pumpkinseed
1=ð1
þe�
ð�20:65þ1:19tm
eanjjaÞÞÞ
0.32
0.73
0.67
196
LepomismacrochirusRafinesque,
1819
Bluegill
1=ð1
+e�
ð�24:26þ1:29tm
eanjja�2:54logðm
dÞþ
3:67logðsdÞÞÞ
0.23
0.62
0.89
59
MicropterusdolomieuLacepede,
1802
Smallm
outh
bass
1=ð1
+e�
ð�19:74þ0:98tm
eanjjaþ0:82logðsaÞþ1:55logðm
dÞÞÞ
0.54
0.79
0.76
329
Micropterussalmoides
(Lacepede,
1802)
Largem
outh
bass
1=ð1
þe�
ð�5:36þ1:21tm
eanmam�3:55logðm
dÞþ
3:32logðsdÞþ
0:60logðsaÞÞ Þ
0.30
0.86
0.94
101
Pomoxisnigromaculatus(Lesueur,1829)
Black
crappie
1=ð1
þe�
ð�22:58þ1:27tm
eanjja�1:75logðm
dÞÞÞ
0.20
0.75
0.86
65
Perca
flavescens(M
itchill,1814)
Yellow
perch
1=ð1
+e�
ð�11:97þ2:41logðsaÞþ0:61tm
eanjja�3:88logðsdÞÞÞ
0.85
0.84
0.67
567
Sander
canadensis(G
riffith
&Smith,1834)
Sauger
1=ð1
þe�
ð�6:47�5:74logðsdÞþ
2:02logðsaÞ�0:32tm
eanatÞÞÞ
0.07
0.63
0.88
38
Sander
vitreus(M
itchill,1818)
Walleye
1=ð1
þe�
ð�4:21þ1:99logðsaÞ�1:93logðm
dÞþ
0:18tm
eanjja�2:70logðsdÞÞÞ
0.64
0.82
0.57
469
Etheostomanigrum
Rafinesque,
1820
Johnnydarter
1=ð1
þe�
ð�8:45�0:33tm
eanatþ4:18logðsdÞþ
1:23logðsaÞÞ Þ
0.08
0.75
0.87
33
Percinacaprodes
(Rafinesque,
1818)
Logperch
1=ð1
þe�
ð�8:09þ1:18logðsaÞþ1:54logðsdÞþ
0:17tm
eanjjaÞÞÞ
0.31
0.60
0.70
171
CottusbairdiiGirard,1850
Mottledsculpin
1=ð1
þe�
ð�5:86þ1:55logðm
dÞþ
0:61logðsaÞÞ Þ
0.09
0.86
0.78
45
CottuscognatusRichardson,1836
Slimysculpin
1=ð1
þe�
ð�6:75þ2:33logðm
dÞþ
0:03prec s
onÞ Þ
0.06
0.86
0.59
49
sdsecchidepth,sa
surfacearea,mdmeandepth,tmeanjjameansummer
temperature
(June–July–August),tmeansonmeanfalltemperature
(September–October–Novem
ber),
tmeandjfmeanwintertemperature
(Decem
ber–January–February),tmeanmammeanspringtemperature
(March–April–May)
Hydrobiologia
123
generous estimate and should be interpreted with
caution.
Trait data and community-weighted traits
Rao’s quadratic entropy, a measure of functional
diversity, was computed by using nine traits relevant
to trophic and thermal niche, yield, demography,
habitat use, commercial importance and adaptation
potential (Table 2). Trait data were compiled from
locally available Ontario freshwater fishes life history
database (http://www.ontariofishes.ca/) and global
fish database (http://www.fishbase.ca/). For categori-
cal traits, we used the verbatim categories found in the
database.
Functional diversity of all traits and community-
weighted mean value or dominant states of individual
traits were computed using function ‘‘dbFD’’, which
implements a flexible distance-based framework to
compute multidimensional functional diversity
indices, in the R library ‘‘FD’’ (Laliberte & Legendre,
2010); we used Gower’s distance. Traditionally,
correlated traits are avoided in computing functional
diversity. Among the selected nine traits in our study,
only species length and weight traits were correlated to
each other. However, we kept both length and weight
in our analyses because species length and weight are
important proxies for body size and biomass, respec-
tively. Moreover, we were mainly interested in
understanding changes in the values of individual
traits. As implemented in the FD package, for numeric
traits, lake-wise community-weighted mean repre-
sents the average trait value across all species in a lake,
and for categorical traits, community-weighted mean
represents the most frequently (i.e. mode) represented
trait state in a lake.
Data analyses
We used permutational multivariate analysis of vari-
ance test using distance matrices to test whether fish
species composition differs between present and
future climate change scenarios (Anderson, 2001).
We used a Bray-Curtis dissimilarity matrix to sum-
marize species composition and used 999 permuta-
tions to determine statistical significance. We also
used non-metric multidimensional scaling to visualize
the compositional trends for different time periods and
climate change scenarios.
We then tested whether species richness, functional
diversity and the mean values of community-weighted
quantitative traits (functional composition; i.e.
Table 2 List of species life history traits used to quantify functional diversity and community-weighted trait means in inland lake
fish communities of Ontario
Trait Ecological significance Coding Trait states/units
Trophic breadth Trophic niche Numeric Number of prey phyla consumed from diet studies
Environment Habitat use Category Benthic, Benthopelagic and Pelagic
Thermal guild Thermal niche Category Cold-water, Cool-water and Warm-water speciesa
Spawning season Demography Category Fall, Spring, Summer, Spring-summer, Winter
Average fecundity Demography Numeric The average number of mature eggs produced
by a female fish per year
Maximum length Body size/yield Numeric Cm
Maximum weight Body size/yield Numeric Kg
Economic importance Ecological services Category Bait; Commercial; Panb; Sportc; Commercial and sport;
Forage and bait; Forage, bait, commercial and sport;
Forage and commercial; Forage, commercial and sport
Disturbance tolerance Extinction risk Ordinal Intolerant (0), Moderate (1), Tolerant (2)
a Coldwater fishes: optimal thermal range between 8 and 12�C; cool-water fishes: optimal thermal range between 16 and 20�C;warm-water fishes: optimal thermal range between 22 and 26�C (Christie & Regier, 1988; Magnuson et al. 1990)b Pan: ‘‘Small fish species easily captured by angling, often found in large numbers, which is harvested for fun or food. Harvest limits
are generous or unlimited’’.c Sport: ‘‘Species that are harvested for personal use, recreation or challenge. Harvest of these species is usually regulated’’.
Hydrobiologia
123
maximum body length, maximum body weight, max-
imum fecundity, species’ disturbance tolerance and
trophic niche) differ between present and two fore-
casted communities, using generalized least square
regression with an induced restrictive correlation
structure (Zuur et al., 2009). The restrictive correlation
structure (i.e. compound symmetric structure)
accounts for the correlation between observations
within the same lakes for different time periods, i.e.
present versus under the scenarios of climate change
(Zuur et al., 2009). The generic model was imple-
mented in R as, gls (response * predictor, metho-
d = ’’REML’’, correlation = corCompSymm(-
form = *1|lake_id), data) using the package
‘‘nlme’’. Chi square tests were used to test if the
community-weighted categorical traits (i.e. thermal
guild, habitat use and economic importance) differ
between present and two forecasted communities
under climate scenarios.
Results
Species composition, richness and functional
diversity
Permutational multivariate analyses of variance indi-
cates that the composition of projected fish commu-
nities of Ontario under climate change scenarios is
different from present (P = 0.01, R2 = 0.29). The
difference is also evident in ordination space (Elec-
tronic Supplementary Materials, Fig. S3), in terms of
separation of lakes between present versus projected
communities (P\ 0.01; R2 = 0.31). The present
species richness of 10.75 ± 3.68 per lake is projected
to increase (P\ 0.01) on average by 60% (6.4 new
species per lake) based on the best-case climate
change scenario and 81% (8.7 new species per lake)
based on the business-as-usual climate change sce-
nario (Fig. 2a; Electronic Supplementary Materials,
Table S1). Under the most conservative scenario of
climate change, 80% of lakes are projected to have
increased species richness, whereas in a business-as-
usual scenario, 95% of lakes are projected to have
increased species richness (Fig. 2b–c). By contrast,
functional diversity is projected to decline (P\ 0.01)
between present and climate change scenarios
(Fig. 2d; Electronic Supplementary Materials,
Table S1).
Functional composition
Community-weighted mean trait values differed
between present and projected communities for traits
related to thermal guild, economic importance, max-
imum length, maximum weight, average fecundity,
trophic niche and disturbance tolerance (Fig. 3; Elec-
tronic Supplementary Materials, Tables S2–4). We did
not detect significant responses from traits related to
spawning season and habitat use (Electronic Supple-
mentary Materials, Fig. S4).
Presently, with respect to the thermal guild func-
tional trait, the proportion of lakes dominated bywarm-
water, cold-water and cool-water species are 4.8, 13.18
and 82%, respectively. According to our projection,
lakes dominated by the trait for warm-water species is
likely to increase by 10–40%,whereas lakes dominated
by the trait for cool-water species is likely to decrease
by up to 27% under scenarios of climate change. Lakes
dominated by the trait for cold-water fish is projected to
decrease by 80%, and will comprise only 0.2–1.8% of
total lakes (Fig. 3a; Electronic Supplementary Mate-
rials, Table S3). The trait for ‘‘commercial and sport’’
fish, including walleye, whitefish, yellow perch,
northern pike and lake trout, is projected to decline
from being found in 27% of lakes to 0.5% under
climate change scenarios (Fig. 3e; Electronic Supple-
mentaryMaterials, Table S4). ‘‘Forage and bait’’ fishes
are projected to predominate the fish community
representing approximately 93% of species composi-
tion based on species occurrence (Fig. 3e).
Presently, the means (±1 SD) of community-
weighted maximum body length, maximum body
weight and average fecundity are 68.29 ± 13.22 cm,
9.09 ± 3.24 kg and 100344.92 ± 45760.57 eggs/per
year. The projected communities under climate change
scenarios are estimated to have, on average, 21–24%
shorter community-weighted mean body length (P\0.01), 38–39% lighter mean body weight (P\ 0.01)
and 36–37% less mean fecundity (P\ 0.01) (Fig. 3b–
d; Electronic SupplementaryMaterials, Table S2). Such
reduction in community-weighted mean body size
suggests reduced yield under climate change scenarios.
The projected communities under the best-case climate
change scenario are estimated to have on average lower
mean trophic breadth (i.e. the average number of phyla
consumed by a species) than the present communities,
indicating the potential decline of predatory species in
projected communities (Fig. 3e; Electronic
Hydrobiologia
123
Supplementary Materials, Table S2). Disturbance tol-
erance in future communities is projected to increase
only under the business as usual climate change scenario
(RCP8.5), but themean trait value for all three scenarios
remains \1.0, suggesting that communities will still
have ‘‘moderate tolerance’’ in both present and fore-
casted scenarios (Fig. 3f).
Discussion
This study reinforces the idea that climate change is
likely to initiate a compositional shift that may favour
warm-water species at the expense of presently
abundant cool- and cold-water species (Chu et al.,
2005; Sharma et al., 2007; Van Zuiden et al., 2016;
Fig. 2 Species richness and functional diversity for present
versus projected fish communities of year-2070. a lake species
richness, b–c distributions of change in lake species richness fora given RCP, d lake functional diversity, e–f distributions of
changes in lake functional diversity for a given RCP. The values
of change in species richness or functional diversity (b–c, e–f) are derived by subtracting lake-specific present species
richness or functional diversity from future species richness or
functional diversity. In the boxplots (a, d), each box contains
middle half of the raw data for a given variable and for a given
climate scenario, horizontal line within each box represents the
median value, and whiskers, as a measure of spread, represent
the inter quartile ranges. Box marked with same letter did not
differ significantly (a = 0.05) among each other, as identified
by Tukey’s post hoc test
Hydrobiologia
123
Fig. 3 Community-
weighted trait states or
values for lake fish
communities. a mosaic plot
showing proportion of lakes
dominated by cold-water
(species that prefer\19�Cduring summer months),
cool-water (19–25�C) andwarm-water ([25�C) fish.b–e community-weighted
maximum length, weight,
fecundity and trophic
breadth. f economic
importance. g disturbance
tolerance. Boxes within the
same plot (b–g) marked with
same letter did not differ
significantly at a = 0.05,
identified by Tukey’s post
hoc test
Hydrobiologia
123
Van Zuiden & Sharma, 2016). This forecasted shift in
community composition is likely to result in an
important shift in community functional structure, as
supported by the changes in community-weighted
means for traits depicting thermal guild, maximum
length, maximum weight, average fecundity, trophic
breadth and economic importance (Fig. 3). Because
these traits are closely related to fisheries productivity
and profitability (Sumaila et al., 2011), the projected
shifts in functional composition can be useful in
developing strategies to buffer fisheries against the
impacts of climate change, and to ensure the continued
provision of aquatic ecosystem services into the
future. However, although our study offers the much
needed initial insights into the compositional shift and
potential changes in community trait structure, our
results should be interpreted with caution, given firstly
that we did not consider the potential negative effects
of inter-specific interactions (Alofs & Jackson,
2015a, b) or potential dispersal constraints (Melles
et al., 2015) in generating future communities.
Secondly, our study attributes climate change-associ-
ated changes in ecosystems services based on changes
in species occurrence data rather than changes in
abundances of the different species within the com-
munities. The latter is a more likely outcome of
climate change impacts given that refugia and fisheries
regulations could also limit the impacts of climate
change on inland lakes. Further studies incorporating
processes, such as inter-specific interactions, dispersal
and species abundance information, in climate change
community composition models would be
worthwhile.
Species richness and functional diversity
We project a 60–81% increase in species richness in
future freshwater fish communities as warm-water
habitat availability increases under warming scenarios
(Hawkins et al., 2003; Menendez et al., 2006).
Increases in species richness in warming aquatic
habitats (Sagarin et al., 1999) is in part attributable to
the gradual nature of the change in thermal habitat
(Daufresne & Boet, 2007). While coldwater fishes are
likely be lost under the most extreme predicted future
changes in temperature (Fig. 3a), gradual warming is
likely to ensure a mix of cool- and warm-water species
until such time as a future equilibrium state might
allow for competitive exclusion of fishes adapted to
cooler habitats. In fact, multiple generations will likely
be necessary before species richness might be
expected to decline (Wilson, 1990). Although larger
lakes in Ontario typically have greater species richness
than small lakes, the projected increase in species
richness under climate change scenarios was consis-
tent across all sizes of lakes (Fig. S5).
But projected increases in lake-wise species rich-
ness are not met with concomitant increases in
functional diversity. This discrepancy occurred
because several traits (2 out of 9) did not change
under climate change scenarios, and the remaining
traits shifted to smaller, lighter, smaller trophic
breadth (i.e. prey species) and less fecund species,
with some traits (describing cold-water, and large-
bodied, economically important fishes) being lost
entirely in forecasted communities (Mouillot et al.,
2014).
Species and functional composition
Present configuration of community composition is
forecasted to change to favour warm-water, smaller-
bodied fishes directly in response to increasing lake
water temperatures (Magnuson et al., 1990; Jaeger
et al., 2014; Melles et al., 2015; O’reilly et al., 2015).
Under warming scenarios, thermal habitat availability
for warm-water fishes is usually projected to increase,
which should result in an increase of warm-water
fishes at the expense of those adapted to cold waters
(Hondzo & Stefan, 1991; Chu et al., 2005; Van Zuiden
& Sharma, 2016). Accordingly, we found that the
proportion of lakes in which warm-water species will
be found may increase by 10% and 40% under best-
case and business as usual scenarios, respectively. The
proportion of lakes in which cool-water species are
found is likely to decrease by up to 27%, and cold-
water species will be lost entirely in the business as
usual warming scenario.
These climate-mediated projected shifts in com-
munity composition are expected to result in a
northward range expansion of both cool-water and
warm-water fishes (Chu et al., 2005; Alofs et al., 2014;
Van Zuiden et al., 2016), which may have strong
impacts on community-wide species interactions. For
example, the invasion of a warm-water fish like
smallmouth bass is projected to result in the loss of
25,000 forage fish populations in Ontario by 2050
(Jackson & Mandrak, 2002), the loss of up to 20,000
Hydrobiologia
123
lake trout populations across Canada by 2100 (Sharma
et al., 2009), and a three-fold reduction in walleye
abundance across Ontario lakes by 2070 (Van Zuiden
& Sharma, 2016) through predator–prey and compet-
itive interactions.
Our models suggest that the present configuration
of several traits describing important metrics of
fisheries yield may change as lakes warm in this
region. Specifically, we project that, on average,
future inland freshwater fish communities in Ontario’s
lakes will be 21–24% shorter, 38–39% lighter and
42–45% less fecund in 2070. These results are
consistent with a meta-analysis suggesting that cli-
mate change benefits smaller-bodied fishes (Daufresne
et al., 2009). Daufresne et al. (2009) presented three
ecological theories at community, population and
individual scales to infer mechanisms that explain why
smaller-bodied fishes are favoured in warming cli-
mates. For example, Bergmann’s rule operates at the
community scale, and suggests that warmer regions
favour smaller-bodied species (Bergmann, 1847). At
the population scale, James’ rule posits that smaller-
bodied individuals will be favoured within populations
found in warmer regions (James, 1970). Finally, at the
individual scale, the temperature-size rule suggests
that body size will decrease with increasing temper-
ature (Atkinson, 1994). Each of these ecological
theories suggest that body size distributions should
decrease with warming climates. However, though
Ontario fishes are typically known to attain greater
sizes in the north than south, we did not notice any
latitudinal bias regarding community-weighted body
size distributions under climate change scenarios
(Fig. S6).
We project that ‘‘business-as-usual’’ and ‘‘best-
case’’ climate change scenarioswill modify the present
configuration of cold-water commercial and sport
fisheries in Ontario’s inland lakes. Affected species
will include lake trout, northern pike, walleye, white-
fish and yellow perch, which are presently found in
27% of lakes. We forecast that the trait describing the
distribution of these species (commercial and sport
fish) will decline at the landscape scale, and will only
be found in 0.5% of lakes under projected climate
change. In contrast, the trait describing ‘‘forage and
bait’’ species will ultimately represent approximately
93% of species composition based on occurrence.
However, one must take into account that some of the
lakes that support commercial fisheries in Ontario, e.g.
Lake Nipigon, Rainy Lake and Lake of the Woods are
not sampled as part of the Broad-scale Monitoring
program (and the data used in this study), and that some
of the deep lakes that support many commercial
fisheries, e.g. Lake Nipigon may still offer coldwater
refugiawith climate change.Nevertheless, the increase
in dominance of forage and bait species combined with
community-wide shrinking body size may impact fish
yield and the provision of regional ecosystem services
under climate change scenarios (Sheridan & Bickford,
2011; Koenigstein et al., 2016). Alternatively, the
increasing dominance of small-bodied pan fish under
climate change scenarios may actually create alternate
fishing opportunities in the future.
One vital ecosystem service provided by fisheries is
food provision, which will be affected with this
forecasted reduction in species body size (Sumaila
et al., 2011). In 2014, approximately 11,684 tonnes of
freshwater fish was commercially harvested from
inland lakes in Ontario, and this catch was worth
*$33,434,000 CDN (http://www.dfo-mpo.gc.
ca). Reductions in the species body size and fecun-
dity may have important implications for fish yield,
and this could be a costly consequence of climate
change. In addition, sport or recreation fisheries also
represent important ecosystem services that will be
disrupted in Ontario under future scenarios of climate
change, and people who rely on the present configu-
ration of the cold-water commercial and sport fishery
are likely to experience consequential socio-economic
consequences of a warming climate (Nelson et al.,
2013). We however suggest that integrating consid-
erations of species diversity and functional composi-
tion are useful means of assessing how aquatic
communities are expected to change with a changing
climate, and can be helpful in crafting informed con-
servation and management strategies under circum-
stances of global environmental change (Villeger
et al., 2010; Cadotte et al., 2011).
Acknowledgements We thank Ontario Ministry of Natural
Resources for the fish data; Thomas Van Zuiden and Miranda
Chen for the climate data; Saiful Khan for the map of the study
area; and John Magnuson, Begona Santos and two anonymous
reviewers for valuable comments on an earlier version of this
manuscript. Funding for this research was provided by Natural
Sciences and Engineering Research Council Canada Discovery
Grant to SS and York University.
Hydrobiologia
123
References
Alofs, K. M. & D. A. Jackson, 2015a. The abiotic and biotic
factors limiting establishment of predatory fishes at their
expanding northern range boundaries in Ontario, Canada.
Global Change Biology 21: 2227–2237.
Alofs, K. M. & D. A. Jackson, 2015b. The vulnerability of
species to range expansions by predators can be predicted
using historical species associations and body size. Pro-
ceedings of the Royal Society B: Biological Sciences 282:
20151211.
Alofs, K. M., D. A. Jackson & N. P. Lester, 2014. Ontario
freshwater fishes demonstrate differing range-boundary
shifts in a warming climate. Diversity and Distributions 20:
123–136.
Anderson, M. J., 2001. A new method for non-parametric
multivariate analysis of variance. Austral Ecology 26:
32–46.
Araujo, M. B. & A. T. Peterson, 2012. Uses and misuses of
bioclimatic envelope modeling. Ecology 93: 1527–1539.
Atkinson, D., 1994. Temperature and organism size – a bio-
logical law for ectotherms? Advances in Ecological
Research 3: 1–58.
Barbet-Massin, M. &W. Jetz, 2015. The effect of range changes
on the functional turnover, structure and diversity of bird
assemblages under future climate scenarios. Global
Change Biology 21: 2917–2928.
Bergmann, C., 1847. Uber die verh altnisse der warme¨ oko-
nomie der thiere zuihrer grosse. Gott. Stud. 1: 595–708.
Blanchet, F. G., P. Legendre & D. Borcard, 2008. Forward
selection of explanatory variables. Ecology 89:
2623–2632.
Botkin, D. B., H. Saxe, M. B. Araujo, R. Betts, R. H. W.
Bradshaw, T. Cedhagen, P. Chesson, T. P. Dawson, J.
R. Etterson, D. P. Faith, S. Ferrier, A. Guisan, A. S. Hansen,
D. W. Hilbert, C. Loehle, C. Margules, M. New, M. J. So-
bel & D. R. B. Stockwell, 2007. Forecasting the effects of
global warming on biodiversity. BioScience 57: 227–236.
Cadotte, M. W., K. Carscadden & N. Mirotchnick, 2011.
Beyond species: functional diversity and the maintenance
of ecological processes and services. Journal of Applied
Ecology 48: 1079–1087.
Cardinale, B. J., J. E. Duffy, A. Gonzalez, D. U. Hooper, C.
Perrings, P. Venail, A. Narwani, G. M. Mace, D. Tilman,
D. A. Wardle, A. P. Kinzig, G. C. Daily, M. Loreau, J.
B. Grace, A. Larigauderie, D. S. Srivastava & S. Naeem,
2012. Biodiversity loss and its impact on humanity. Nature
486: 59–67.
Chen, I.-C., J. K. Hill, R. Ohlemuller, D. B. Roy & C.
D. Thomas, 2011. Rapid range shifts of species associated
with high levels of climate warming. Science 333:
1024–1026.
Christie, G. C. & H. A. Regier, 1988. Measures of optimal
thermal habitat and their relationship to yields for four
commercial fish species. Canadian Journal of Fisheries and
Aquatic Sciences 45: 301–314.
Chu, C., N. E. Mandrak & C. K. Minns, 2005. Potential impacts
of climate change on the distributions of several common
and rare freshwater fishes in Canada. Diversity and
Distributions 11: 299–310.
Civantos, E., W. Thuiller, L. Maiorano, A. Guisan & M.
B. Araujo, 2012. Potential impacts of climate change on
ecosystem services in Europe: the case of pest control by
vertebrates. BioScience 62: 658–666.
Daufresne, M. & P. Boet, 2007. Climate change impacts on
structure and diversity of fish communities in rivers. Global
Change Biology 13: 2467–2478.
Daufresne, M., K. Lengfellner & U. Sommer, 2009. Global
warming benefits the small in aquatic ecosystems. Pro-
ceedings of the National academy of Sciences of the United
States of America 106: 12788–12793.
Dıaz, S., S. Lavorel, F. De Bello, F. Quetier, K. Grigulis & T.
M. Robson, 2007. Incorporating plant functional diversity
effects in ecosystem service assessments. Proceedings of
the National Academy of Sciences 104: 20684–20689.
Dubuis, A., J. Pottier, V. Rion, L. Pellissier, J.-P. Theurillat &A.
Guisan, 2011. Predicting spatial patterns of plant species
richness: a comparison of direct macroecological and
species stacking modelling approaches. Diversity and
Distributions 17: 1122–1131.
Gauzere, P., F. Jiguet & V. Devictor, 2015. Rapid adjustment of
bird community compositions to local climatic variations
and its functional consequences. Global Change Biology
21: 3367–3378.
Guisan, A. & C. Rahbek, 2011. SESAM – a new framework
integrating macroecological and species distribution
models for predicting spatio-temporal patterns of species
assemblages. Journal of Biogeography 38: 1433–1444.
Guisan, A. &W. Thuiller, 2005. Predicting species distribution:
offering more than simple habitat models. Ecology Letters
8: 993–1009.
Guisan, A. & N. E. Zimmermann, 2000. Predictive habitat
distribution models in ecology. Ecological Modelling 135:
147–186.
Hawkins, B. A., R. Field, H. V. Cornell, D. J. Currie, J.-F.
Guegan, D. M. Kaufman, J. T. Kerr, G. G. Mittelbach, T.
Oberdorff, E. M. O’brien, E. E. Porter & J. R. G. Turner,
2003. Energy, water, and broad-scale geographic patetrns
in species richness. Ecology 84: 3105–3117.
Hijmans, R. J., S. E. Cameron, J. L. Parra, P. G. Jones & A.
Jarvis, 2005. Very high resolution interpolated climate
surfaces for global land areas. International Journal of
Climatology 25: 1965–1978.
Hondzo, M. & H. G. Stefan, 1991. Three case studies of lake
temperature and stratification response to warmer climate.
Water Resources Research 27: 1837–1846.
Hooper, D. U., F. S. Chapin, J. J. Ewel, A. Hector, P. Inchausti,
S. Lavorel, J. H. Lawton, D. M. Lodge, M. Loreau, S.
Naeem, B. Schmid, H. Setala, A. J. Symstad, J. Vander-
meer & D. A. Wardle, 2005. Effects of biodiversity on
ecosystem functioning: a consensus of current knowledge.
Ecological Monographs 75: 3–35.
Hooper, D. U., E. C. Adair, B. J. Cardinale, J. E. K. Byrnes, B.
A. Hungate, K. L. Matulich, A. Gonzalez, J. E. Duffy, L.
Gamfeldt & M. I. O’Connor, 2012. A global synthesis
reveals biodiversity loss as a major driver of ecosystem
change. Nature 486: 105–108.
IPCC, 2013. Climate Change 2013: The Physical Science Basis.
Contribution of Working Group I to the Fifth Assessment
Report of the Intergovernmental Panel on Climate Change,
Hydrobiologia
123
Cambridge University Press, Cambridge, United Kingdom
and New York, NY, USA.
Jackson, D. A. & H. H. Harvey, 1989. Biogeographic associa-
tions in fish assemblages: local vs regional processes.
Ecology 70: 1472–1484.
Jackson, D. A. & N. E. Mandrak (eds), 2002. Changing Fish
Biodiversity: Predicting the Loss of Cyprinid Biodiversity
Due to Global Climate Change. American Fisheries Soci-
ety, Bethesda, MD.
Jaeger, K. L., J. D. Olden&N. A. Pelland, 2014. Climate change
poised to threaten hydrologic connectivity and endemic
fishes in dryland streams. Proceedings of the National
Academy of Sciences 111: 13894–13899.
James, F. C., 1970. Geographic size variation in birds and its
relationship to climate. Ecology 51: 365–390.
Koenigstein, S., F. C. Mark, S. Goßling-Reisemann, H. Reuter
& H.-O. Poertner, 2016. Modelling climate change impacts
on marine fish populations: process-based integration of
ocean warming, acidification and other environmental
drivers. Fish and Fisheries 17: 972–1004.
Laliberte, E. & P. Legendre, 2010. A distance-based framework
for measuring functional diversity from multiple traits.
Ecology 91: 299–305.
Magnuson, J. J., J. D. Meisner & D. K. Hill, 1990. Potential
changes in the thermal habitat of great lakes fish after
global climate warming. Transactions of the American
Fisheries Society 119: 254–264.
Magnuson, J. J., T. K. Kratz, T. F. Allen, D. E. Armstrong, B.
J. Benson, C. J. Bowser, D. W. Bolgrien, S. R. Carpenter,
T. M. Frost, S. T. Gower, T. M. Lillesand, J. A. Pike & M.
G. Turner, 1997. Regionalization of long-term ecological
research (LTER) on north temperate lakes. International
Association of Theoretical and Applied Limnology 26(Pt 2
26): 522–528.
Melles, S. J., C. Chu, K. M. Alofs & D. A. Jackson, 2015.
Potential spread of Great Lakes fishes given climate change
and proposed dams: an approach using circuit theory to
evaluate invasion risk. Landscape Ecology 30: 919–935.
Menendez, R., A. G. Megıas, J. K. Hill, B. Braschler, S.
G. Willis, Y. Collingham, R. Fox, D. B. Roy & C.
D. Thomas, 2006. Species richness changes lag behind
climate change. Proceedings of the Royal Society of
London B: Biological Sciences 273: 1465–1470.
Millennium-Ecosystem-Assessment, 2005. Ecosystems and
Human Well-being: Policy Responses. Island Press,
Washington, DC.
Mokany, K., J. J. Thomson, A. J. J. Lynch, G. J. Jordan & S.
Ferrier, 2015. Linking changes in community composition
and function under climate change. Ecological Applica-
tions 25: 2132–2141.
Mokany, K., S. Ferrier, S. R. Connolly, P. K. Dunstan, E.
A. Fulton, M. B. Harfoot, T. D. Harwood, A. J. Richardson,
S. H. Roxburgh, J. P. W. Scharlemann, D. P. Tittensor, D.
A.Westcott & B. A.Wintle, 2016. Integrating modelling of
biodiversity composition and ecosystem function. Oikos
125: 10–19.
Mouillot, D., S. Villeger, V. Parravicini, M. Kulbicki, J.
E. Arias-Gonzalez, M. Bender, P. Chabanet, S. R. Floeter,
A. Friedlander, L. Vigliola & D. R. Bellwood, 2014.
Functional over-redundancy and high functional
vulnerability in global fish faunas on tropical reefs. Pro-
ceedings of the National Academy of Sciences 111:
13757–13762.
Nelson, E. J., P. Kareiva, M. Ruckelshaus, K. Arkema, G.
Geller, E. Girvetz, D. Goodrich, V. Matzek, M. Pinsky, W.
Reid, M. Saunders, D. Semmens&H. Tallis, 2013. Climate
change’s impact on key ecosystem services and the human
well-being they support in the US. Frontiers in Ecology and
the Environment 11: 483–493.
Nock, C. A., R. J. Vogt & B. E. Beisner, 2016. Functional Traits.
In eLS. Wiley.
Oliver, T.H.,M. S.Heard, N. J. B. Isaac, D.B. Roy,D. Procter, F.
Eigenbrod, R. Freckleton, A. Hector, C. D. L. Orme, O.
L. Petchey, V. Proenca, D. Raffaelli, K. B. Suttle, G.
M. Mace, B. Martın-Lopez, B. A. Woodcock & J. M. Bul-
lock, 2015a. Biodiversity and resilience of ecosystem
functions. Trends in Ecology & Evolution 30: 673–684.
Oliver, T. H., N. J. B. Isaac, T. A. August, B. A. Woodcock, D.
B. Roy & J. M. Bullock, 2015b. Declining resilience of
ecosystem functions under biodiversity loss. Nat Com-
munications 6: 10122.
O’reilly, C. M., S. Sharma, D. K. Gray, S.E. Hampton, J.
S. Read, R. J. Rowley, P. Schneider, J. D. Lenters, P.
B. Mcintyre, B. M. Kraemer, G. A. Weyhenmeyer, D.
Straile, B. Dong, R. Adrian, M. G. Allan, O. Anneville, L.
Arvola, J. Austin, J. L. Bailey, J. S. Baron, J. D. Brookes, E.
De Eyto, M.T. Dokulil, D. P. Hamilton, K. Havens, A.
L. Hetherington, S. N. Higgins, S. Hook, L. R. Izmest’eva,
K. D. Joehnk, K. Kangur, P. Kasprzak, M. Kumagai, E.
Kuusisto, G. Leshkevich, D. M. Livingstone, S. Macintyre,
L. May, J. M. Melack, D. C. Mueller-Navarra, M. Nau-
menko, P. Noges, T. Noges, R. P. North, P.-D. Plisnier, A.
Rigosi, A. Rimmer, M. Rogora, L. G. Rudstam, J.
A. Rusak, N. Salmaso, N. R. Samal, D.E. Schindler, S.
G. Schladow, M. Schmid, S. R. Schmidt, E. Silow, M.
E. Soylu, K. Teubner, P. Verburg, A. Voutilainen, A.
Watkinson, C. E.Williamson&G. Zhang, 2015. Rapid and
highly variable warming of lake surface waters around the
globe. Geophysical Research Letters 42: 10773–10781.
Parmesan, C. & G. Yohe, 2003. A globally coherent fingerprint
of climate change impacts across natural systems. Nature
421: 37–42.
Sagarin, R. D., J. P. Barry, S. E. Gilman & C. H. Baxter, 1999.
Climate-related change in an intertidal community over
short and long time scales. Ecological Monographs 69:
465–490.
Sandstrom, S., M. Rawson & N. Lester, 2010. Manual of
Instructions for Broad-scale Fish Community Monitoring;
using LargeMesh Gillnets and Small Mesh Gillnets. 34. In:
O.M.O.N. Resources (ed.), Peterborough, ON.
Sharma, S. & D. A. Jackson, 2008. Predicting smallmouth bass
(Micropterus dolomieu) occurrence across North America
under climate change: a comparison of statistical approa-
ches. Canadian Journal of Fisheries and Aquatic Sciences
65: 471–481.
Sharma, S., D. A. Jackson, C. K. Minns & B. J. Shuter, 2007.Will northern fish populations be in hot water because of
climate change? Global Change Biology 13: 2052–2064.
Sharma, S., D. A. Jackson & C. K. Minns, 2009. Quantifying the
potential effects of climate change and the invasion of
Hydrobiologia
123
smallmouth bass on native lake trout populations across
Canadian lakes. Ecography 32: 517–525.
Sharma, S., M. J. Vander Zanden, J. J. Magnuson & J. Lyons,
2011. Comparing Climate Change and Species Invasions
as Drivers of Coldwater Fish Population Extirpations. Plos
One 6: e22906.
Sheridan, J. A. & D. Bickford, 2011. Shrinking body size as an
ecological response to climate change. Nature Climate
Change 1: 401–406.
Shuter, B. J., J. A. Maclean, F. E. J. Fry & H. A. Regier, 1980.
Stochastic simulation of temperature effects on first-year
survival of smallmouth bass. Transactions of the American
Fisheries Society 109: 1–34.
Simpson, S. D., S. Jennings, M. P. Johnson, J. L. Blanchard, P.-J.
Schon, D. W. Sims & M. J. Genner, 2011. Continental
shelf-wide response of a fish assemblage to rapid warming
of the sea. Current Biology 21: 1565–1570.
Sumaila, U. R., W. W. L. Cheung, V. W. Y. Lam, D. Pauly & S.
Herrick, 2011. Climate change impacts on the biophysics
and economics of world fisheries. Nature Climate Change
1: 449–456.
Thuiller, W., S. Pironon, A. Psomas, M. Barbet-Massin, F.
Jiguet, S. Lavergne, P. B. Pearman, J. Renaud, L. Zupan &
N. E. Zimmermann, 2014. The European functional tree of
bird life in the face of global change. Nature Communi-
cations 5: 3118.
Tonn, W. M. & J. J. Magnuson, 1982. Patterns in the species
composition and richness of fish assemblages in Northern
Wisconsin lakes. Ecology 63: 1149–1166.
Van Zuiden, T. M., M. M. Chen, S. Stefanoff, L. Lopez & S.
Sharma, 2016. Projected impacts of climate change on
three freshwater fishes and potential novel competitive
interactions. Diversity and Distributions 22: 603–614.
Van Zuiden, T. M. & S. Sharma, 2016. Examining the effects of
climate change and species invasions on Ontario walleye
populations: can walleye beat the heat? Diversity and
Distributions 22: 1069–1079.
Villeger, S., J. R. Miranda, D. F. Hernandez & D. Mouillot,
2010. Contrasting changes in taxonomic vs. functional
diversity of tropical fish communities after habitat degra-
dation. Ecological Applications 20: 1512–1522.
Violle, C., M.-L. Navas, D. Vile, E. Kazakou, C. Fortunel, I.
Hummel & E. Garnier, 2007. Let the concept of trait be
functional! Oikos 116: 882–892.
Walther, G. R., 2010. Community and ecosystem responses to
recent climate change. Philosophical Transactions of the
Royal Society B-Biological Sciences 365: 2019–2024.
Wilson, J. B., 1990. Mechanisms of species coexistence: twe-
leve explanations for Hutchinson’s paradox of the plank-
ton: evidence from New Zealand plant communities. New
Zealand Journal of Ecology 13: 17–42.
Zurell, D., N. E. Zimmermann, T. Sattler, M. P. Nobis & B.
Schroder, 2016. Effects of functional traits on the predic-
tion accuracy of species richness models. Diversity and
Distributions 22: 905–917.
Zuur, A. F., E. N. Ieno, N. J. Walker, A. A. Saveliev & G.
M. Smith, 2009. Mixed Effects Models and Extensions in
Ecology with R. Springer.
Hydrobiologia
123