Hydrobiologia DOI 10.1007/s10750-017-3208-1
PRIMARY RESEARCH PAPER
Projected compositional shifts and loss of ecosystem services in freshwater fish communities under climate change scenarios Shekhar R. Biswas . Richard J. Vogt . Sapna Sharma
Received: 3 November 2016 / Revised: 17 April 2017 / Accepted: 18 April 2017 Ó Springer International Publishing Switzerland 2017
Abstract What are the projected impacts of climate change on community composition and consequentially 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 ‘‘bestcase’’ and ‘‘business-as-usual’’ climate change scenarios. 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 commercially 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 community functional structure, which is likely to affect important aquatic ecosystem services.
Communicated by Handling editor: Begon˜a Santos
Electronic supplementary material The online version of this article (doi:10.1007/s10750-017-3208-1) contains supplementary 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
Keywords Compositional shift Ecosystem services 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 environments 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
123
Hydrobiologia
(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 (Dı´az 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 community trait structure for present versus projected communities. 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 organisms 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 distributions will have ramifications for overall ecosystem productivity (Millennium-Ecosystem-Assessment, 2005). Further, future trait distributions can be forecasted 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
123
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, coolwater 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 extirpation 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 richness, (iii) functional diversity and (iv) functional composition. We compared present day communities with those forecast under business-as-usual and bestcase 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. Correspondingly, lake functional diversity will increase if warm-water invaders introduce new functional characteristics 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
Methods Study area and data The study area is inland lakes in Ontario, Canada (Fig. 1). These lakes support a wide range of commercial 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 Monitoring (OMNRF-BsM) Program aims to collect data on *700 Ontario lakes every five years to inform landscape level management; and to-date has collected 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 largeand small-mesh gillnets (Sandstrom et al., 2010). We obtained the most recent (2008–2012) georeferenced fish community dataset from OMNRFBsM 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 disproportionate influence on statistical analysis (Jackson & Harvey, 1989). For the same set of lakes, we obtained historical climate data (monthly mean temperature and precipitation) represented as climate averages from 1950 to 2000 from WorldClim 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)
123
Hydrobiologia
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 occurrence 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
123
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 presence) and specificity (ability to correctly predict species absence) for each species distribution model. This procedure is recommended when species presences 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 CCMA GCM. 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 communities 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 (Arau´jo & 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; specificity range: 0.51–0.94; see Table 1) but over prediction for species richness (correlation between projected richness and actual richness, r = 0.61, n = 129 validation lakes), which is a well acknowledged phenomenon in the stacking approach of community generation (Zurell et al., 2016). As such, our projected communities should be considered a
Pearl dace Brown bullhead Burbot Ninespine stickleback
Margariscus margarita (Cope, 1867)
Ameiurus nebulosus (Lesueur, 1819)
Lota lota (Linnaeus, 1758)
Pungitius pungitius (Linnaeus, 1758)
Spottail shiner
Notropis hudsonius (Clinton, 1824)
Creek chub
Blacknose shiner
Notropis heterolepis Eigenmann & Eigenmann, 1893
Semotilus atromaculatus (Mitchill, 1818)
Blackchin shiner
Notropis heterodon (Cope, 1865)
Fathead minnow
Common shiner
Luxilus cornutus (Mitchill, 1817)
Pimephales promelas Rafinesque, 1820
Emerald shiner
Notropis atherinoides Rafinesque, 1818
Mimic shiner
Golden shiner
Notemigonus crysoleucas (Mitchill, 1814)
Bluntnose minnow
Lake chub
Couesius plumbeus (Agassiz, 1850)
Notropis volucellus (Cope, 1865)
Northern redbelly dace
Chrosomus eos Cope, 1861
Pimephales notatus (Rafinesque, 1820)
Short head redhorse
Moxostoma macrolepidotum (Lesueur, 1817)
1=ð1 þ e
Longnose sucker
Esox masquinongy Mitchill, 1824 White sucker
Northern pike Muskellunge
Esox lucius Linnaeus, 1758
Catostomus catostomus (Forster, 1773)
Rainbow smelt
Osmerus mordax (Mitchill, 1814)
Catostomus commersonii (Lacepe`de, 1803)
1=ð1 þ e
Cisco
Coregonus artedi Lesueur, 1818
Þ
Þ
1=ð1 þ e Þ
Þ
0.07
ð1:29þ3:02 logðsdÞ0:76 logðsaÞ0:23tmeanson Þ
0.28 0.47 0.11
1=ð1 þ eð16:15þ0:88tmeanjja ÞÞ Þ 1=ð1 þ eðð4:67þ2:01 logðsaÞþ3:76 logðmdÞ0:29tmeanjja Þ Þ 1=ð1 þ eð8:05þ1:46 logðsaÞþ3:11 logðsdÞ0:22tmeanat Þ Þ
Þ
0.05
1=ð1 þ e
0.07
1=ð1 þ eð3:902:65 logðsaÞþ0:06precson þ2:75 logðmdÞÞ Þ
0.10
1=ð1 þ eð7:611:79 logðsaÞ0:49tmeanjja þ3:71 logðsdÞÞ Þ
Þ
0.24
1=ð1 þ e
0.47
0.17
0.05
1=ð1 þ eð17:71þ0:79tmeanjja þ1:30 logðmdÞþ2:58 logðsdÞÞ Þ
ð22:05þ0:13precjja þ2:39 logðsdÞþ0:43tmeanjja Þ
1=ð1 þ eð1:81þ1:72 logðsaÞ1:54 logðmdÞ0:38tmeanson Þ Þ
1=ð1 þ e
ð2:160:15tmeanat þ1:86 logðsdÞÞ
1=ð1 þ eð20:94þ1:07tmeanjja Þ Þ
0.17
0.29
ð7:63þ0:07precson þ1:20 logðmdÞÞ
0.23 Þ
1=ð1 þ eð2:96þ1:15 logðsaÞ1:28 logðsdÞ0:22tmeanson ÞÞ Þ
1=ð1 þ e
0.14
1=ð1 þ eð5:76þ3:00 logðsdÞ0:57tmeanjja Þ Þ ð12:52þ0:84tmeanjja 0:81 logðmdÞ0:75 logðsaÞÞ
0.16
0.91
0.17
1=ð1 þ eð3:252:03 logðsaÞþ0:07precson Þ Þ
Þ
Þ
0.22
0.71
0.20
ð2:550:54tmeanson þ1:34 logðsaÞÞ
ð8:910:76tmeanjja þ2:42 logðmdÞ1:53 logðsdÞÞ
Þ
Þ
ð22:18þ0:96tmeanjja þ1:18 logðsaÞ2:25 logðmdÞþ2:87 logðsdÞÞ
ð0:80þ1:81 logðsaÞ0:24tmeanat 4:19 logðsdÞÞ
1=ð1 þ eð13:64þ1:43 logðsaÞþ0:10precjja 1:59 logðmdÞÞ Þ
1=ð1 þ e
1=ð1 þ e
1=ð1 þ e
0.57 0.11
Þ
ð19:71þ2:39 logðmdÞþ0:08precson þ0:93 logðsaÞþ0:37tmeanjja Þ
1=ð1 þ e
0.56
1=ð1 þ eð4:44þ1:73 logðsaÞþ2:29 logðmdÞ2:53 logðsdÞ0:17tmeanat Þ Þ
ð0:61þ1:92 logðsaÞ0:38tmeanjja þ0:70 logðmdÞÞ
0.47
Lake whitefish
1=ð1 þ e
Lake trout
Coregonus clupeiformis (Mitchill, 1818)
ð2:270:42tmeanjja þ7:04 logðmdÞþ5:20 logðsdÞÞ
Salvelinus namaycush (Walbaum, 1792)
0.20
1=ð1 þ eð11:870:64tmeanjja 2:23logðsaÞþ3:41logðsdÞÞ Þ
Brook trout
Salvelinus fontinalis (Mitchill, 1814)
Optimal threshold
Model used to predict future presence
Local name
Species
Table 1 Species-wise logistic models for predicting future distributions of fishes in inland lakes of Ontario
0.71
0.70
0.72
0.70
0.75
0.71
0.68
0.75
0.69
0.67
0.75
0.70
0.66
0.61
0.67
0.50
0.64
0.88
0.50
0.67
0.80
0.55
0.82
0.72
0.82
0.60
Sensitivity
0.81
0.78
0.85
0.69
0.80
0.81
0.79
0.79
0.69
0.51
0.56
0.64
0.79
0.72
0.69
0.91
0.88
0.57
0.85
0.93
0.63
0.82
0.68
0.72
0.83
0.92
Specificity
55
266
136
42
37
37
102
41
320
114
45
146
147
146
87
33
83
617
77
42
488
62
402
360
286
64
No of occurrence
Hydrobiologia
123
123 1=ð1 + e 1=ð1 + e
Bluegill Smallmouth bass Largemouth bass Black crappie Yellow perch Sauger Walleye Johnny darter Logperch Mottled sculpin Slimy sculpin
Lepomis macrochirus Rafinesque, 1819
Micropterus dolomieu Lacepe`de, 1802
Micropterus salmoides (Lacepe`de, 1802)
Pomoxis nigromaculatus (Lesueur, 1829)
Perca flavescens (Mitchill, 1814)
Sander canadensis (Griffith & Smith, 1834)
Sander vitreus (Mitchill, 1818)
Etheostoma nigrum Rafinesque, 1820
Percina caprodes (Rafinesque, 1818)
Cottus bairdii Girard, 1850
Cottus cognatus Richardson, 1836
Þ
0.09 0.06
1=ð1 þ eð5:86þ1:55 logðmdÞþ0:61 logðsaÞÞ Þ 1=ð1 þ eð6:75þ2:33 logðmdÞþ0:03precson Þ Þ
0.86
0.86
0.60
0.75
0.82
0.63
0.84
0.75
0.86
0.79
0.62
0.73
0.70
0.79
Sensitivity
0.59
0.78
0.70
0.87
0.57
0.88
0.67
0.86
0.94
0.76
0.89
0.67
0.81
0.74
Specificity
49
45
171
33
469
38
567
65
101
329
59
196
294
282
No of occurrence
sd secchi depth, sa surface area, md mean depth, tmeanjja mean summer temperature (June–July–August), tmeanson mean fall temperature (September–October–November), tmeandjf mean winter temperature (December–January–February), tmeanmam mean spring temperature (March–April–May)
0.08 0.31
1=ð1 þ eð8:09þ1:18 logðsaÞþ1:54 logðsdÞþ0:17tmeanjja ÞÞ Þ
0.64
1=ð1 þ eð8:450:33tmeanat þ4:18 logðsdÞþ1:23 logðsaÞÞ Þ
0.07
1=ð1 þ eð4:21þ1:99 logðsaÞ1:93 logðmdÞþ0:18tmeanjja 2:70 logðsdÞÞ Þ
0.20
1=ð1 þ eð6:475:74 logðsdÞþ2:02 logðsaÞ0:32tmeanat ÞÞ Þ
Þ 0.85
ð22:58þ1:27tmeanjja 1:75 logðmdÞÞ
0.30
0.54
0.23
0.32
1=ð1 + eð11:97þ2:41logðsaÞþ0:61tmeanjja 3:88logðsdÞÞ Þ
1=ð1 þ e
1=ð1 þ e
Þ
ð5:36þ1:21tmeanmam 3:55 logðmdÞþ3:32 logðsdÞþ0:60 logðsaÞÞ
Þ
ð19:74þ0:98tmeanjja þ0:82logðsaÞþ1:55logðmdÞÞ
Þ
ð24:26þ1:29tmeanjja 2:54logðmdÞþ3:67logðsdÞÞ
1=ð1 þ eð20:65þ1:19tmeanjja ÞÞ Þ
1=ð1 þ e
Pumpkinseed
Lepomis gibbosus (Linnaeus, 1758)
0.48 0.54
ð17:93þ0:92tmeanjja þ0:91 logðsaÞÞ
Rock bass
Ambloplites rupestris (Rafinesque, 1817)
1=ð1 þ eð2:61þ1:93 logðsaÞ3:19 logðsdÞ0:32tmeanson Þ Þ
Trout perch
Percopsis omiscomaycus (Walbaum, 1792)
Optimal threshold
Model used to predict future presence
Local name
Species
Table 1 continued
Hydrobiologia
Hydrobiologia
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 categorical traits, we used the verbatim categories found in the database. Functional diversity of all traits and communityweighted 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, respectively. 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 represents 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 variance 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 summarize species composition and used 999 permutations 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’’.
123
Hydrobiologia
maximum body length, maximum body weight, maximum fecundity, species’ disturbance tolerance and trophic niche) differ between present and two forecasted 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 implemented in R as, gls (response * predictor, method = ’’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 indicates that the composition of projected fish communities of Ontario under climate change scenarios is different from present (P = 0.01, R2 = 0.29). The difference is also evident in ordination space (Electronic 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 scenario (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-asusual 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).
123
Functional composition Community-weighted mean trait values differed between present and projected communities for traits related to thermal guild, economic importance, maximum length, maximum weight, average fecundity, trophic niche and disturbance tolerance (Fig. 3; Electronic Supplementary Materials, Tables S2–4). We did not detect significant responses from traits related to spawning season and habitat use (Electronic Supplementary Materials, Fig. S4). Presently, with respect to the thermal guild functional trait, the proportion of lakes dominated by warmwater, 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 Materials, 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 Supplementary Materials, Table S4). ‘‘Forage and bait’’ fishes are projected to predominate the fish community representing approximately 93% of species composition based on species occurrence (Fig. 3e). Presently, the means (±1 SD) of communityweighted 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 Supplementary Materials, 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
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 for a 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
Supplementary Materials, Table S2). Disturbance tolerance in future communities is projected to increase only under the business as usual climate change scenario (RCP 8.5), but the mean trait value for all three scenarios remains \1.0, suggesting that communities will still have ‘‘moderate tolerance’’ in both present and forecasted 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;
123
Hydrobiologia Fig. 3 Communityweighted trait states or values for lake fish communities. a mosaic plot showing proportion of lakes dominated by cold-water (species that prefer \19°C during summer months), cool-water (19–25°C) and warm-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
123
Hydrobiologia
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-associated changes in ecosystems services based on changes in species occurrence data rather than changes in abundances of the different species within the communities. 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; Mene´ndez 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 & Boe¨t, 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 consistent across all sizes of lakes (Fig. S5). But projected increases in lake-wise species richness 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 largebodied, 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, smallerbodied 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 bestcase 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 coldwater species will be lost entirely in the business as usual warming scenario. These climate-mediated projected shifts in community 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
123
Hydrobiologia
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 competitive 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 climate 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 climates. 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 smallerbodied 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 temperature (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 ‘‘bestcase’’ climate change scenarios will 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, whitefish 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
123
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 refugia with 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 fecundity 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 configuration 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 considerations of species diversity and functional composition are useful means of assessing how aquatic communities are expected to change with a changing climate, and can be helpful in crafting informed conservation and management strategies under circumstances of global environmental change (Ville´ger 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, Begon˜a 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
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. Proceedings 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. Arau´jo, 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 biological 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¨ okonomie 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. Arau´jo, 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. Sobel & 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. Ohlemu¨ller, 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. Boe¨t, 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. Proceedings of the National academy of Sciences of the United States of America 106: 12788–12793. Dı´az, S., S. Lavorel, F. De Bello, F. Que´tier, 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. Gue´gan, 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. Vandermeer & 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,
123
Hydrobiologia Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Jackson, D. A. & H. H. Harvey, 1989. Biogeographic associations 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 Society, 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. Go¨ßling-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. Mene´ndez, R., A. G. Megı´as, 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 Applications 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. Ville´ger, V. Parravicini, M. Kulbicki, J. E. Arias-Gonza´lez, M. Bender, P. Chabanet, S. R. Floeter, A. Friedlander, L. Vigliola & D. R. Bellwood, 2014. Functional over-redundancy and high functional
123
vulnerability in global fish faunas on tropical reefs. Proceedings 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. Proenc¸a, D. Raffaelli, K. B. Suttle, G. M. Mace, B. Martı´n-Lo´pez, B. A. Woodcock & J. M. Bullock, 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 Communications 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. Naumenko, 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 Large Mesh 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 approaches. 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 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. Scho¨n, 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 Communications 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. Ville´ger, S., J. R. Miranda, D. F. Herna´ndez & D. Mouillot, 2010. Contrasting changes in taxonomic vs. functional diversity of tropical fish communities after habitat degradation. 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: tweleve explanations for Hutchinson’s paradox of the plankton: 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. Schro¨der, 2016. Effects of functional traits on the prediction 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.
123