Journal of Biogeography (J. Biogeogr.) (2015) 42, 2187–2196
ORIGINAL ARTICLE
Global variation in thermal physiology of birds and mammals: evidence for phylogenetic niche conservatism only in the tropics Imran Khaliq1,2*a, Susanne A. Fritz1, Roland Prinzinger3, Markus Pfenninger1,2, Katrin B€ ohning-Gaese1,2 and Christian Hof1
1
Senckenberg Biodiversity and Climate Research Centre (BiK-F), Senckenberganlage 25, Frankfurt 60325, Germany, 2Department of Biological Sciences, Institute for Ecology, Evolution and Diversity, Goethe University, Max-von-Laue-Straße 13, Frankfurt 60438, Germany, 3Metabolic Physiology, Department of Biological Sciences, Institute for Ecology, Evolution and Diversity, Goethe University, Max-von-Laue-Straße 13, Frankfurt 60438, Germany
ABSTRACT
Aim Physiological traits that approximate the fundamental climatic niche – the climatic conditions where a species can survive – are the outcome of adaptation to the environment under historical and current environmental constraints. If a large amount of the variation in physiological traits among species can be explained by their phylogeny rather than by contemporary environmental conditions, this would indicate phylogenetic conservatism in physiological traits, i.e. the tendency of species to retain their ancestral physiology over time. Here, we evaluate the relative contributions of phylogeny and environment to explain the variation in physiological traits of birds and mammals at the global level, as well as separately for tropical versus temperate species. Location Global. Methods We compiled a large data set from the literature, on the thermal traits and basal metabolic rates of 552 endotherms (255 bird and 297 mammal species) as measured in physiological experiments, along with phylogenetic, geographical and climatic data. Our analyses, which were performed separately for birds and mammals, partitioned the variation in comparative physiological data into the relative contributions of phylogenetic and environmental distance matrices. Results Overall, the current environment explained a larger amount of variation in thermal traits among species than the phylogeny. However, we found that phylogeny was much more important than current environment for explaining the variation in physiological traits in the tropics, whereas environment was more important than phylogeny in temperate species.
*Correspondence: Imran Khaliq, Senckenberg Biodiversity and Climate Research Centre (BiK-F), Senckenberganlage 25, Frankfurt 60325, Germany. E-mail:
[email protected] a
Present address: Government Degree College Vehova, Punjab, Pakistan
Main conclusions While evidence for phylogenetic conservatism in physiological traits at the global level was weak, results for tropical species suggest phylogenetic conservatism in their physiological traits. These results indicate a stronger tendency in tropical species to retain their ancestral thermal traits, which might in turn imply a lower physiological adaptability of tropical species to ongoing and future climate change. Keywords Climate change, ecological niche, endotherms, macrophysiology, physiological traits, temperate regions, thermal adaptation, thermal tolerances, tropical climate.
INTRODUCTION Why most species are restricted to certain environmental conditions has long been a central question in ecology, bioª 2015 John Wiley & Sons Ltd
geography and evolution (Darwin, 1859). In a time of rapid climate change, this question has become even more important, because accurate estimates of the environmental conditions where species can survive (i.e. their fundamental http://wileyonlinelibrary.com/journal/jbi doi:10.1111/jbi.12573
2187
I. Khaliq et al. climatic niche; Hutchinson, 1957; Sober on, 2007; Colwell & Rangel, 2009) are needed to guide conservation efforts under a changing climate. The environmental conditions that species occupy (i.e. the realized ecological niche; Hutchinson, 1957; Guisan & Zimmermann, 2000) are the outcome of dispersal limitations, for instance due to geographical barriers, their interactions with other species, and the environmental conditions that species are adapted to (Sober on, 2007). Closely related species are expected to have similar ecological niches for two non-exclusive reasons: because they have adapted to similar environmental conditions or because they share a common evolutionary history (Freckleton & Jetz, 2009). Hence, the observed similarity of ecological niches of closely related species should contain both current environmental and evolutionary signals. Hereafter, we use the terms ‘environmental’ for adaptation to contemporary environmental conditions and ‘phylogenetic’ for adaptation to historical environmental conditions that are captured by the phylogeny of the living species. Accurate quantifications of the relative contributions of evolutionary and environmental signals to the observed patterns of species’ ecological niches are becoming increasingly important for understanding species’ potential responses to climate change (Quintero & Wiens, 2013). Niche conservatism is the tendency of species to retain ancestral niche characteristics (Wiens & Graham, 2005; Cooper et al., 2011; Grigg & Buckley, 2013; Kamilar & Cooper, 2013). However, similar traits in related or unrelated species can also be an outcome of living in (and adaptation to) similar climatic conditions (i.e. convergence; Freckleton & Jetz, 2009). For example, closely related species living in the tropics may share similar values of their thermal traits because climatic conditions in the tropics are seasonally and spatially less variable than in the temperate regions and, as a consequence of living under homogeneous climatic conditions, species in the tropics may show a strong pattern of phylogenetic niche conservatism (Pianka, 1966; Terborgh, 1973; Wiens & Donoghue, 2004). On the other hand, as the climatic conditions in the temperate regions generally show larger seasonal and spatial variation, temperate species are expected to show more variation in their trait values. Such a larger degree of dissimilarity in their trait values could derive from a lower level of phylogenetic conservatism and a stronger signal of adaptation to the environment. Hence, different levels of phylogenetic niche conservatism in the physiological traits of birds and mammals may be expected in different geographical regions. In recent years, the potential of phylogenetic conservatism of the climatic niche as a measure of species’ capacity to adapt to climate change has been recognized (Hof et al., 2010; Cooper et al., 2011; Grigg & Buckley, 2013). To estimate the degree of phylogenetic conservatism in climatic niches, an accurate quantification of species’ climatic niches is crucial. For these niche quantifications, most researchers to date have used correlations of species distribution records with climatic conditions (Hof et al., 2010; Cooper et al., 2011; Olalla-Tarraga et al., 2011). This approach has been 2188
criticized for several reasons (Dormann et al., 2010). First, the approach assumes a close link between the climatic conditions experienced by a species and its physiological tolerances, which may not be the case (Sunday et al., 2012; Khaliq et al., 2014). Second, the set of climatic conditions inferred from species’ geographical ranges is by definition only a subset of the species’ fundamental niche (Sober on, 2007; Colwell & Rangel, 2009). Therefore, as physiological traits determined from experiments are more direct measures of species’ physiological adaptation to climatic conditions ~ez & Ara (Higgins et al., 2013; Ferri-Yan ujo, 2015), these physiological traits should be more appropriate estimates of the fundamental climatic niches of species (Cooper et al., 2011). For ectotherms, researchers increasingly use experimentally measured physiological variables as measures for fundamental climatic niches in order to estimate the level of phylogenetic conservatism (Grigg & Buckley, 2013). For endotherms, however, there are no such studies, and all recent publications on phylogenetic niche conservatism in endotherms have used niche estimates based on climate only (Cooper et al., 2011; Olalla-Tarraga et al., 2011). Here, we investigate patterns of phylogenetic conservatism of climatic niches in the two most species-rich endotherm groups, i.e. birds and mammals, with the largest data set of experimentally measured thermal traits to date (Khaliq et al., 2014). Different statistical approaches have been adopted to detect phylogenetic niche conservatism. These approaches can be broadly grouped into two categories. The first approach measures the rate of evolution in different traits, and a slower rate is assumed to indicate phylogenetic conservatism (Losos, 2008; Cooper et al., 2010, 2011). The second approach evaluates the role of phylogeny or taxonomy in explaining trait variation across species, and the explanatory power of phylogeny is expected to be positively related to the level of phylogenetic conservatism (Freckleton & Jetz, 2009; Cooper et al., 2010; Hof et al., 2010; Olalla-Tarraga et al., 2011). In this study, we adopt the latter approach, because it enables us to partition the variation in traits into phylogenetic and environmental components (Freckleton & Jetz, 2009). If phylogeny explains more variation in physiological traits than the environment that species experience, we can interpret this as an indication of phylogenetic conservatism in the climatic niches of endotherms. On the contrary, if environment explains more variation in the physiological traits than phylogeny, we can assume that climatic niches are not phylogenetically conserved, and that species have adapted to the contemporary environmental conditions they experience. Assuming that higher levels of phylogenetic conservatism in their physiological traits imply a lower ability to adapt successfully to changing climatic conditions (Losos, 2008; Wiens, 2008; Cooper et al., 2011), our analyses may allow us to assess the vulnerability of species and their potential physiological responses to predicted climate change. Here, we investigate the relative contributions of phylogeny and environment in experimentally measured thermal traits of 552 bird and mammal species. Endotherms maintain Journal of Biogeography 42, 2187–2196 ª 2015 John Wiley & Sons Ltd
Phylogenetic conservatism in endotherm thermal physiology constant body temperatures and are therefore generally considered to be thermal specialists (Huey et al., 2012). Owing to this high level of thermal specificity, we expect that the variation in the thermal traits of endotherms would be better explained by the phylogeny than by the environment. Specifically, we test the following predictions: we expect (1) that phylogeny will explain more variation in thermal traits of birds and mammals than the environment at the global scale, which would provide support for strong phylogenetic conservatism of their climatic niches, and (2) that the relative influence of phylogeny on thermal tolerances is stronger in tropical than in temperate species, which may reflect the higher environmental variation experienced by temperate species. MATERIALS AND METHODS We collated the following thermal traits from the literature at the species level, as experimental measures of the thermal tolerances of birds and mammals: breadth of thermal neutral zone (TNZ), upper critical temperature (UCT) and lower critical temperature (LCT) (see Khaliq et al., 2014; for a detailed description). TNZ represents the thermal comfort zone for endotherms; within TNZ individuals can maintain their body temperatures with minimum energy expenditure (i.e. at the basal metabolic rate, BMR; Scholander et al., 1950a). Additionally, we compiled data on BMR from the literature (see Appendix S1 in Supporting information). Higher metabolic rates in endotherms in comparison to ectotherms have long been considered to be an adaptation to cold climatic conditions (Bozinovic & Rosenman, 1989). While BMR has an impact on multiple aspects of a species’ physiology and life history, it is also an important component of the physiological thermal niche of endotherms (Scholander et al., 1950b; Sßekercioglu et al., 2012; Ara ujo et al., 2013). BMR is highly correlated with body mass (Prinzinger & H€anssler, 1980), which is a highly phylogenetically conserved trait as well (Freckleton et al., 2002; Cooper & Purvis, 2010; Kamilar & Cooper, 2013). Consequently, in our analysis we are using mass-corrected BMR to account for the potentially confounding effects of body mass. Mass-corrected BMR was calculated as BMRcorr = BMR/Mb where ‘M’ is body mass and ‘b’ is the allometric scaling exponent (McNab, 2012). The data set contained 255 bird species with 161 resident and 94 migratory species, of which 63 were collected from their breeding grounds and the remaining 31 from their wintering grounds (representing 25 of 28 orders and 72 of 170 families of birds) and 297 mammal species (representing 24 of 29 orders and 75 of 153 families of mammals; taxonomy following Wilson & Reeder, 2005; Jetz et al., 2012). In the case of mammals, data for BMR were available for only 277 species (see Appendix S1). The bird data set was further split into subsets of 94 migratory and 161 resident birds, as migratory birds shuttle regularly between their breeding and wintering grounds, which may require special physiological adaptations to differing environmental conditions (Jetz et al., Journal of Biogeography 42, 2187–2196 ª 2015 John Wiley & Sons Ltd
2008). From each study that provided physiological data (see Appendix S1), we extracted the geographical coordinates of the sites where the individuals used in the experiments were captured (‘capture sites’ hereafter). Data on thermal conditions (mean annual temperature, monthly average of daily minimum temperature of the coldest month, monthly average of daily maximum temperature of the hottest month) were derived for these capture sites from the CliMond data set for current climate, which is averaged over 30 years from 1961 to 1991 and centred at 1975 (Kriticos et al., 2012). The temperature values included in our analysis ranged from 35 to 41 °C for birds, and from 32 to 41 °C for mammals. Thus, the variation in ambient temperature captured in our analyses should be large enough to include most of the thermal conditions that species experience across the globe. We selected these three climatic variables as they are considered relevant for the physiologically critical temperatures of endotherms (McNab & Morrison, 1963; Ara ujo et al., 2013). To classify species as tropical or temperate, we used the latitude of their capture sites. A species was classified as tropical if its capture site was located below 23.5° absolute latitude and as temperate if its capture site was located above 23.5° absolute latitude. For species for which we found more than one study, we included the study with the most extreme thermal trait values. The method used in this study to estimate the relative influence of phylogeny and environment on thermal traits and mass-corrected BMR uses pairwise phylogenetic and geographical distances among species, and assumes that geographical distances between species pairs evolved along the phylogeny (Freckleton & Jetz, 2009). For all of these distance-based analyses, we used the capture sites of the species as their geographical occurrences or to infer their preferred environmental conditions. To quantify the contributions of the environment to the variation in thermal traits and mass-corrected BMR, we used four surrogates for the environment, namely one pairwise geographical distance matrix among species (based on the geographical coordinates) and three pairwise climatic distance matrices among species (based on the three climatic variables), assuming that the geographical or climatic distance matrices among species would capture the environmental signal (Freckleton & Jetz, 2009). Before the analyses, we standardized all the physiological variables to a mean of 0 and a variance of 1 to enable comparison between different groups. We used published species-level phylogenies for birds and mammals as phylogenetic data (Fritz et al., 2009; Kuhn et al., 2011; Jetz et al., 2012). To evaluate the consistency of results across different tree topologies, we randomly sampled 100 trees from the pseudo-posterior distribution of these supertrees (Kuhn et al., 2011; Jetz et al., 2012) and ran the analyses for the 100 trees. These analyses enabled us to estimate 95% confidence intervals for each parameter (see below) across these 100 trees in our analyses, which quantify the influence of phylogenetic uncertainty as captured in the 2189
I. Khaliq et al. distributions of trees on our analyses. For the mammalian trees, polytomies were resolved using a birth–death model of diversification (Kuhn et al., 2011). We then estimated the contributions of phylogeny and each one of the four surrogates of environmental conditions to the interspecific variation in TNZ, UCT, LCT and masscorrected BMR of birds and mammals. We used a method that partitions the variance into phylogenetic and environmental components rather than estimating the rates of thermal trait evolution, for several reasons. First, we wanted to tease apart the contributions of phylogenetic constraints and of adaptation to environment, which is done less explicitly in rate analyses. Second, we use subsets of global specieslevel phylogenies, where the uncertainties of topology and branch lengths are high, confidence in rate estimates low, and rate heterogeneity across the phylogeny is very likely, complicating model fit for evolutionary rate analyses. Third and most importantly, estimation of evolutionary rates should be based on complete trees, but we only have data for a small proportion of species. The method of Freckleton & Jetz (2009) is therefore most appropriate for our data, as it tries to estimate the relative contributions of phylogeny and environment to the current distribution of trait values across the tips of the phylogeny. The method produces three positive parameters (φ, k0 and c; equation 2.7 in Freckleton & Jetz, 2009) that sum to 1. The value of φ represents the relative contribution of spatial effects (here: contribution of environmental conditions). The value of k0 = (1φ) k represents the relative contribution of phylogeny (where k is the branch length transformation defined by Pagel, 1999). Finally, c = (1φ) (1k) represents the relative contribution of unexplained effects independent of phylogeny and environment (Freckleton & Jetz, 2009; Cooper et al., 2011; Grigg & Buckley, 2013). In the analyses using the geographical distance matrix for birds, we ran three separate analyses for resident birds only, migratory birds only, and resident and migratory birds together. For the analyses with climatic distance matrices, we only used resident birds because migratory bird species experience different climatic conditions in their breeding and wintering grounds, and information about the geographical locations of the breeding and the wintering grounds for the individuals used in the experiments were not available. As our data set only represents relatively small proportions of the known mammal and bird species, we assessed how representative these species were by evaluating how the species in our data set were distributed across the complete species-level phylogenies. To assess whether our species were a random selection from the phylogenetic trees of all bird and mammal species, we used the D statistic (Fritz & Purvis, 2010). We tested the value of D for significant departure from 1; with D = 1 if the species are randomly distributed across the phylogeny and D = 0 if the species are as phylogenetically clustered as expected under a threshold model of Brownian motion (Fritz & Purvis, 2010). Additionally, we used D to test whether tropical or temperate species within 2190
our data set were phylogenetically clustered with respect to each other, because under global niche conservatism and a tropical origin for both taxa, phylogenetic clustering would be expected for the temperate species (Wiens & Donoghue, 2004). RESULTS The evaluation of the relative contributions of geography (as surrogate for environment) and phylogeny to the variation in thermal traits across endotherm species showed contrasting results globally and for separate climatic regions. At the global level, the influence of geography was much more pronounced on the thermal traits of birds and mammals than the influence of phylogeny (Figs 1a–f & 2a). In contrast, the global variation in mass-corrected BMR appears to be largely driven by the phylogeny (Figs 1g,h & 2a). When evaluating the relative contributions of phylogeny and geography to thermal traits and mass-corrected BMR in tropical and temperate species separately, results for mass-corrected BMR were similar to the global analysis and showed a very strong influence of phylogeny in both regional groups (Fig. 2c,e). However, thermal traits showed contrasting influence of phylogeny and geography in the regional analyses; tropical species displayed relatively strong phylogenetic influence in all thermal traits, whereas temperate species generally showed a stronger influence of geography (Fig. 2c,e). Relative phylogenetic contributions for birds and mammals were similar in all analyses except for LCT, for which phylogeny showed a higher contribution in tropical mammals than in tropical birds (Fig. 2c,e). When using the three different climatic distance matrices as surrogates for environment, their relative contributions were in most cases very weak (Fig. 2b,d,f & Fig. S1 in Appendix S2) compared to geographical contributions in the first set of analyses (Fig. 2a,c,e). Furthermore, the influence of phylogeny on the global distribution of thermal traits and mass-corrected BMR across resident birds and mammals was higher than the climatic influence (Fig. 2b). When evaluating the relative contributions of phylogeny and climate to thermal trait and mass-corrected BMR variation at the regional scale, we found patterns similar to the geographical analysis: phylogeny contributed more to the variation in thermal tolerances across tropical species than across temperate species (Fig. 2d,f & Fig. S1d,f in Appendix S2). For migratory birds, we found similar patterns of relative contributions of phylogeny and geography or climate as reported above (see Fig. S2 in Appendix S2). Testing whether species were randomly distributed across the phylogenies also revealed differences between the global and regional analysis. In the global analysis, the bird and mammal species in our data set were distributed relatively randomly across the respective super-trees of all species, although the D values were significantly different from 1 in both groups (birds: D = 0.846, P < 0.001; mammals: D = 0.804, P < 0.001). However, the temperate species were highly Journal of Biogeography 42, 2187–2196 ª 2015 John Wiley & Sons Ltd
Phylogenetic conservatism in endotherm thermal physiology (a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
Figure 1 Geographical and phylogenetic distribution of thermal traits and mass-corrected basal metabolic rate (BMR) in (a, c, e, g) birds (n = 255) and (b, d, f, h) mammals (n = 297; nBMR = 277). (a, b) Breadth of thermal neutral zone (TNZ, in °C), (c, d) upper critical temperature (UCT, in °C), (e, f) lower critical temperature (LCT, in °C) and (g, h) mass-corrected BMR (in kJ h1). Points on the maps indicate the location of the sites where species’ individuals were captured for the physiological experiments in which thermal traits were measured. For improved visual clarity, we separated out the symbols for overlapping species.
clustered across the bird trees in our data set (D = 0.05, P = 0.52 for departure from 0, Fig. S3 in Appendix S2) and also, to a lesser extent, across the mammal trees (D = 0.49, P < 0.001 for departure from 0, Fig. S4 in Appendix S2). DISCUSSION Contrary to our expectations, we found that at the global scale, environment was more important than phylogeny for explaining the variation in thermal traits of birds and mammals, except for BMR. However, the separate analyses for the tropical and temperate subgroups supported our prediction of a stronger influence of phylogeny on the thermal traits of tropical, than on those of temperate species. This difference between tropical and temperate species was similar in birds and mammals, irrespective of the environmental surrogates used, and of whether or not migratory bird species were included. Journal of Biogeography 42, 2187–2196 ª 2015 John Wiley & Sons Ltd
No evidence for global phylogenetic niche conservatism in thermal traits Current environmental conditions appeared to be more influential than the past evolutionary history for the variation in the thermal tolerances of endotherm species in our global analysis, thus providing no support for a general phylogenetic conservatism of climatic niches (Wiens & Donoghue, 2004). While there is mixed support for climatic niche conservatism in the literature for mammals (Dormann et al., 2010; Olalla-Tarraga et al., 2011), this has seldom been tested in birds. Our results are in accordance with the studies that found no evidence of climatic niche conservatism in mammals (Cooper et al., 2010; Dormann et al., 2010; but see Olalla-Tarraga et al., 2011). Recently, Ara ujo et al. (2013) used data sets on critical temperatures for birds and mammals similar to ours, and discussed the amount of variation in these critical temperatures with regard to the level of 2191
I. Khaliq et al.
Geographic distances (a)
Climatic distances (b)
1.00
Unexplained Geography/Climate Phylogeny
0.75 0.50
Trop = Tropical Temp = Temperate
0.25 0 TNZ UCT LCT BMR TNZ UCT LCT BMR
TNZ UCT LCT BMR TNZ UCT LCT BMR
(d)
(c) 1.00 0.75 0.50 0.25 0 Trop TempTrop Temp Trop TempTrop Temp TNZ
UCT
LCT
Trop TempTrop Temp Trop TempTrop Temp
BMR
(e)
TNZ
UCT
LCT
BMR
(f)
1.00 0.75 0.50 0.25 0 TropTemp Trop TempTrop TempTrop Temp TNZ
UCT
LCT
BMR
Trop TempTrop Temp Trop TempTrop Temp TNZ
UCT
LCT
BMR
Figure 2 Relative phylogenetic, environmental and unexplained contributions to the variation in thermal variables (i.e. breadth of thermal neutral zone, TNZ; UCT, upper critical temperature; LCT, lower critical temperature; BMR, basal metabolic rate). Environmental distances were quantified in two ways: (a, c, e) with geographical distance matrices and (b, d, f) with climatic distance matrices (based on annual mean temperature in this case). We show the results of (a, b) the global analysis, conducted separately for resident birds and mammals, (c, d) the regional analysis for tropical (n = 76) and temperate (n = 85) resident birds, and (e, f) the regional analysis for tropical (n = 109) and temperate (n = 188) mammals. The variance was partitioned into phylogenetic (k0 , black), geographical or climatic (φ, white), and unexplained residual components (c, grey). Error bars indicate 95% confidence intervals across 100 pseudo-posterior trees for each parameter separately. All analyses were performed separately for birds and mammals.
phylogenetic conservatism in these traits. Although our results do not support their suggestion that especially UCTs are phylogenetically conserved, we do find support for their hypothesis about the low phylogenetic conservatism of LCTs (Ara ujo et al., 2013). Differences between our results and those of previous studies might have arisen because of the following reasons. First, while we assess phylogenetic contributions on the 2192
variation of actual thermal tolerances which were measured in physiological experiments, previous studies were based on species’ climatic preferences estimated from climatic conditions across their geographical ranges (Olalla-Tarraga et al., 2011). However, the climatic conditions a species can tolerate (i.e. its fundamental climatic niche; approximated by its tolerances to climate) do not need to match the geographically realized climatic space a species occurs in (i.e. Journal of Biogeography 42, 2187–2196 ª 2015 John Wiley & Sons Ltd
Phylogenetic conservatism in endotherm thermal physiology its realized climatic niche), which has been shown in both conceptual and empirical studies (Hutchinson, 1957; Colwell & Rangel, 2009; Khaliq et al., 2014). Second, previous studies either only tested for phylogenetic signal while ignoring environmental influences on the different measures of climatic niches (Olalla-Tarraga et al., 2011), or only looked at the variation of thermal traits and their relationship to conditions, without explicitly quantifying the evolutionary (i.e. phylogenetic) signal (Ara ujo et al., 2013). In contrast, we deliberately aimed to tease apart phylogenetic and environmental contributions to the variation in thermal tolerances, which should give more valuable insight on the evolutionary trajectories of species’ thermal tolerances, and thus on components of their fundamental climatic niche. Finally, the only previous study that did use a data set of thermal tolerances similar to ours (Ara ujo et al., 2013) contains far fewer species of endotherms than our study. Tropical but no temperate phylogenetic niche conservatism The phylogenetic contribution to thermal trait variation was much stronger in tropical than in temperate endotherms; this highlights the importance of regional differences which can be obscured in global analyses (Hof et al., 2010). In fact, the strong environmental contribution to thermal trait variation in temperate species also dominates the global pattern, masking the strong phylogenetic influence on thermal tolerances in tropical species (which are, however, numerically dominant globally, but not in our data set). The strong contribution of phylogeny in the case of the thermal traits of tropical species can be an outcome either of comparatively homogeneous climatic conditions in the tropics, or of shared ancestry (Wiens & Donoghue, 2004), or of both. However, we emphasize that the method used here accounts for the relative contributions of the environment and phylogeny, so the result should not be an artefact of homogeneous climatic conditions alone. The higher phylogenetic contribution to the variation in UCT in the tropics suggests that, if exposed to rising temperatures, tropical species may face challenges in responding by means of adaptive changes (Wiens & Graham, 2005; Deutsch et al., 2008; Ara ujo et al., 2013; but see Logan et al., 2014). However, the results for cold tolerances in tropical birds are contrary to the expectations under phylogenetic niche conservatism (Wiens & Donoghue, 2004), as they showed a strong environmental component (however, only when using geographical distance matrices). This indicates that tropical birds and mammals may have developed different capabilities for tolerating cold temperatures (Khaliq et al., 2014). Overall, our results suggest that cold tolerance is probably a labile trait in tropical birds, and unlike tropical mammals, tropical birds may be able to shift their ranges towards colder habitats under a changing climate.
Journal of Biogeography 42, 2187–2196 ª 2015 John Wiley & Sons Ltd
Basal metabolic rate versus thermal traits The strong influence of environment on TNZ breadths and critical temperatures, but not on mass-corrected BMR, could indicate that the evolution of the former responds faster to climatic conditions. Endotherms also respond seasonally to variation in climatic conditions by altering the thickness of their body covering (Scholander et al., 1950b); as a consequence, the values of critical temperatures may vary seasonally as well (McNab & Morrison, 1963; Wilson et al., 2011). On the contrary, changes in mass-corrected BMR may be more difficult to achieve because that might require alterations in the basic molecular architecture of species’ metabolism (White & Kearney, 2013). If our assumptions hold, these results suggest that thermal traits such as TNZ and critical temperatures are more labile evolutionarily and can thus, in response to changing environmental conditions, potentially adapt faster than mass-corrected BMR. This might in turn imply that both groups of endotherms are able to meet the challenge of predicted future rising temperatures by adjusting their critical temperatures. However, we note that there is a discrepancy between the temporal scale of the phylogenetic analyses (100,000s to millions of years) of our or similar studies, and the time-scale of species’ responses to ongoing climate change (decades to centuries). Therefore, direct conclusions on adaptability that are derived from phylogenetic signal may not be warranted (Quintero & Wiens, 2013), and explicit studies addressing this temporal mismatch are required. Environmental surrogates and species sampling To assess the strength of environmental contributions to the interspecific variation of thermal traits, we used either geographical or climatic distance matrices, which led to different results: the relative contributions of the environmental distance matrices were lower when using a climatic instead of a geographical distance matrix. These differences may be due to the fact that thermal traits can be influenced not only by temperature (which was used to calculate the climatic distance matrices), but also by other climatic and environmental factors (i.e. amount of precipitation, habitat, resource availability, topography etc.) in a complex manner. In other words, geographical distance can be viewed as a proxy for many unmeasured environmental variables, so its better predictive power probably reflects the fact that the climatic variables used in the other analyses were not the only ones that matter. This is also evident from the fact that the three climatic matrices, i.e. minimum, mean and maximum temperatures, contributed to different extents to explaining the variation in thermal traits. Thus, these complex effects of a multitude of different environmental factors may be captured by geographical distances between species more comprehensively than by distances in thermal conditions alone.
2193
I. Khaliq et al. Another reason for the relatively low contribution of temperature variation to physiological trait variation may be a mismatch between the temperature data used and accurate estimates of the thermal environments experienced by a species’ individuals. For example, average or extreme values of air temperature may insufficiently capture the temperature preferences of species that behaviourally regulate the temperatures they are exposed to (e.g. fossorial or arboreal mammals, understorey birds, etc.). Although this goes beyond the aim of our study, there is certainly a need for more detailed studies accounting for this behavioural variation as well as for using climatic data with improved spatial and temporal resolutions (see Kearney et al., 2014). We acknowledge that our results and possibly those of previous studies may be to some extent confounded by a bias in sampling. The relatively low number of species in our analyses – as compared to the total number of extant species of birds and mammals – may have made the detection of any phylogenetic influence difficult. However, estimations of the D statistic indicated that species included in our analysis are fairly randomly distributed across the complete supertrees of the respective taxonomic groups. Moreover, Pagel’s k (Pagel, 1999), as applied in our analysis to estimate phylogenetic contributions, has been shown to be robust to small sample size with random sampling under a Brownian process of trait evolution (M€ unkem€ uller et al., 2012). Therefore, we assume that the trends revealed by our analyses are robust to sampling biases. CONCLUSIONS Assuming that the phylogenetic component in the variance of thermal traits across species is a measure for their adaptability to climate change (Wiens, 2004, 2008; Wiens & Donoghue, 2004), our results for the global-level analysis indicate that many endotherms may be less vulnerable to rising temperatures than previously expected (Thomas et al., 2004) because of two main reasons. First, we show low global phylogenetic conservatism of thermal tolerances, so species may be able to respond to the rising temperatures via adaptation of their physiology. Second, most species, particularly from temperate regions, have higher upper thermal tolerances than the maximum temperatures predicted for the year 2080 under climate change (Khaliq et al., 2014). In contrast, the strong phylogenetic influence on the thermal traits of tropical species indicates that many tropical endotherm species may face challenges when confronted with changing temperature regimes as projected by climate change scenarios. Most of the tropical species are already living close to their upper critical temperatures, and even small increases in temperatures may be challenging for them (Colwell et al., 2008; Deutsch et al., 2008; Khaliq et al., 2014; but see Bonebrake & Deutsch, 2012; Helmuth et al., 2014). Changing temperatures might be more challenging for tropical mammals than for tropical birds, as tropical mammals showed a stronger level of conservatism in their cold tolerances: this suggests 2194
that, unlike tropical birds, tropical mammals might find it difficult to shift their ranges towards colder habitats. Although we recognize that species can evolve rapidly to new climatic conditions if they have high levels of genetic variation, for tropical endotherms with long generation times, small numbers of offspring, low dispersal abilities and low BMR levels (Wiersma et al., 2007), the evolutionary pace might be too slow to off-set negative effects of climate change. ACKNOWLEDGEMENTS We are grateful to Robert P. Freckleton for providing R code, Bob O’Hara for help with statistical analyses, and to Sßerban Prochesß, Natalie Cooper and one anonymous referee for their comments on previous versions of the text. We also thank Tanja Caprano and Cornelia Weist for their support with data preparation and analyses. I.K. is supported by the Higher Education Commission of Pakistan and the German Academic Exchange Service (DAAD). This work was supported by the research funding programme “LOEWE Landes-Offensive zur Entwicklung Wissenschaftlich-€ okonomischer Exzellenz” of Hesse’s Ministry of Higher Education, Research, and the Arts. REFERENCES ~ez, F., Bozinovic, F., Marquet, P.A., Ara ujo, M.B., Ferri-Yan Valladares, F. & Chown, S.L. (2013) Heat freezes niche evolution. Ecology Letters, 16, 1206–1219. Bonebrake, T.C. & Deutsch, C.A. (2012) Climate heterogeneity modulates impact of warming on tropical insects. Ecology, 93, 449–455. Bozinovic, F. & Rosenman, M. (1989) Maximum metabolic rate of rodents : physiological and ecological consequences on distributional limits. Functional Ecology, 3, 173–181. Colwell, R.K. & Rangel, T.F. (2009) Hutchinson’s duality: the once and future niche. Proceedings of the National Academy of Sciences USA, 106, 19651–19658. Colwell, R.K., Brehm, G., Cardel us, C.L., Gilman, A.C. & Longino, J.T. (2008) Global warming, elevational range shifts, and lowland biotic attrition in the wet tropics. Science, 322, 258–261. Cooper, N. & Purvis, A. (2010) Body size evolution in mammals: complexity in tempo and mode. The American Naturalist, 175, 727–738. Cooper, N., Jetz, W. & Freckleton, R.P. (2010) Phylogenetic comparative approaches for studying niche conservatism. Journal of Evolutionary Biology, 23, 2529–2539. Cooper, N., Freckleton, R.P. & Jetz, W. (2011) Phylogenetic conservatism of environmental niches in mammals. Proceedings of the Royal Society B: Biological Sciences, 278, 2384–2391. Darwin, C. (1859) On the origin of species by means of natural selection, or the preservation of favoured races in the struggle for life. John Murray, London. Journal of Biogeography 42, 2187–2196 ª 2015 John Wiley & Sons Ltd
Phylogenetic conservatism in endotherm thermal physiology Deutsch, C.A., Tewksbury, J.J., Huey, R.B., Sheldon, K.S., Ghalambor, C.K., Haak, D.C. & Martin, P.R. (2008) Impacts of climate warming on terrestrial ectotherms across latitude. Proceedings of the National Academy of Sciences USA, 105, 6668–6672. Dormann, C.F., Gruber, B., Winter, M. & Herrmann, D. (2010) Evolution of climate niches in European mammals? Biology Letters, 6, 229–232. ~ez, F. & Ara Ferri-Yan ujo, M.B. (2015) Lizards could be warming faster than climate. Ecography, 38, 437–439. Freckleton, R.P. & Jetz, W. (2009) Space versus phylogeny: disentangling phylogenetic and spatial signals in comparative data. Proceedings of the Royal Society B: Biological Sciences, 276, 21–30. Freckleton, R.P., Harvey, P.H. & Pagel, M. (2002) Phylogenetic analysis and comparative data: a test and review of evidence. The American Naturalist, 160, 712–726. Fritz, S.A. & Purvis, A. (2010) Selectivity in mammalian extinction risk and threat types: a new measure of phylogenetic signal strength in binary traits. Conservation Biology, 24, 1042–1051. Fritz, S.A., Bininda-Emonds, O.R.P. & Purvis, A. (2009) Geographical variation in predictors of mammalian extinction risk: big is bad, but only in the tropics. Ecology Letters, 12, 538–549. Grigg, J.W. & Buckley, L.B. (2013) Conservatism of lizard thermal tolerances and body temperatures across evolutionary history and geography. Biology Letters, 9, 20121056. Guisan, A. & Zimmermann, N.E. (2000) Predictive habitat distribution models in ecology. Ecological Modelling, 135, 147–186. Helmuth, B., Russell, B.D., Connell, S.D., Dong, Y., Harley, C.D.G., Lima, F.P., Sara, G., Williams, G.A. & Mieszkowska, N. (2014) Beyond long-term averages: making biological sense of a rapidly changing world. Climate Change Responses, 1, 1–6. Higgins, J.K., MacLean, H.J., Buckley, L.B. & Kingsolver, J.G. (2013) Geographic differences and microevolutionary changes in thermal sensitivity of butterfly larvae in response to climate. Functional Ecology, 28, 982–989. Hof, C., Rahbek, C. & Ara ujo, M.B. (2010) Phylogenetic signals in the climatic niches of the world’s amphibians. Ecography, 33, 242–250. Huey, R.B., Kearney, M.R., Krockenberger, A., Holtum, J.A.M., Jess, M. & Williams, S.E. (2012) Predicting organismal vulnerability to climate warming: roles of behaviour, physiology and adaptation. Philosophical Transactions of the Royal Society B: Biological Sciences, 367, 1665–1679. Hutchinson, G.E. (1957) Concluding remarks. Cold Spring Harbor Symposium on Quantative Biology, 22, 415–427. Jetz, W., Freckleton, R.P. & McKechnie, A.E. (2008) Environment, migratory tendency, phylogeny and basal metabolic rate in birds. PLoS ONE, 3, e3261.
Journal of Biogeography 42, 2187–2196 ª 2015 John Wiley & Sons Ltd
Jetz, W., Thomas, G.H., Joy, J.B., Hartmann, K. & Mooers, A.O. (2012) The global diversity of birds in space and time. Nature, 491, 444–448. Kamilar, J.M. & Cooper, N. (2013) Phylogenetic signal in primate behaviour, ecology and life history. Philosophical Transactions of the Royal Society B: Biological Sciences, 368, 20120341. Kearney, M.R., Isaac, A.P. & Porter, W.P. (2014) microclim: global estimates of hourly microclimate based on longterm monthly climate averages. Scientific Data, 1, 1–9. Khaliq, I., Hof, C., Prinzinger, R., B€ ohning-Gaese, K. & Pfenninger, M. (2014) Global variation in thermal tolerances and vulnerability of endotherms to climate change. Proceedings of the Royal Society B: Biological Sciences, 281, 20141097. Kriticos, D.J., Webber, B.L., Leriche, A., Ota, N., Macadam, I., Bathols, J. & Scott, J.K. (2012) CliMond: global highresolution historical and future scenario climate surfaces for bioclimatic modelling. Methods in Ecology and Evolution, 3, 53–64. Kuhn, T.S., Mooers, A.Ø. & Thomas, G.H. (2011) A simple polytomy resolver for dated phylogenies. Methods in Ecology and Evolution, 2, 427–436. Logan, M.L., Cox, R.M. & Calsbeek, R. (2014) Natural selection on thermal performance in a novel thermal environment. Proceedings of the National Academy of Sciences USA, 111, 14165–14169. Losos, J.B. (2008) Phylogenetic niche conservatism, phylogenetic signal and the relationship between phylogenetic relatedness and ecological similarity among species. Ecology Letters, 11, 995–1003. McNab, B.K. (2012) Extreme measures: the ecological energetics of birds and mammals. The University of Chicago Press, Chicago, IL. McNab, B.K. & Morrison, P. (1963) Body temperature and metabolism in subspecies of Peromyscus from arid and mesic environments. Ecological Monographs, 33, 63–82. M€ unkem€ uller, T., Lavergne, S., Bzeznik, B., Dray, S., Jombart, T., Schiffers, K. & Thuiller, W. (2012) How to measure and test phylogenetic signal. Methods in Ecology and Evolution, 3, 743–756. Olalla-Tarraga, M., McInnes, L., Bini, L.M., Diniz-Filho, J.A.F., Fritz, S.A., Hawkins, B.A., Hortal, J., Orme, C.D.L., Rahbek, C., Rodriguez, M.A. & Purvis, A. (2011) Climatic niche conservatism and the evolutionary dynamics in species range boundaries: global congruence across mammals and amphibians. Journal of Biogeography, 38, 2237–2247. Pagel, M. (1999) Inferring the historical patterns of biological evolution. Nature, 401, 877–884. Pianka, E.R. (1966) Latitudinal gradients in species diversity: a review of concepts. The American Naturalist, 100, 33–46. Prinzinger, R. & H€anssler, I. (1980) Metabolism–weight relationship in some small nonpasserine birds. Experientia, 36, 1299–1300. Quintero, I. & Wiens, J.J. (2013) Rates of projected climate change dramatically exceed past rates of climatic niche
2195
I. Khaliq et al. evolution among vertebrate species. Ecology Letters, 16, 1095–1103. Scholander, P.F., Hock, R., Walters, V. & Irving, L. (1950a) Adaptation to cold in arctic and tropical mammals and birds in relation to body temperature, insulation, and basal metabolic rate. The Biological Bulletin, 99, 259–271. Scholander, P.F., Walters, V.V., Hock, R., Irving, L. & Johnson, F. (1950b) Body insulation of some arctic and tropical mammals and birds. The Biological Bulletin, 99, 225–236. Sßekercioglu, C ß .H., Primack, R.B. & Wormworth, J. (2012) The effects of climate change on tropical birds. Biological Conservation, 148, 1–18. Sober on, J. (2007) Grinnellian and Eltonian niches and geographic distributions of species. Ecology Letters, 10, 1115–1123. Sunday, J.M., Bates, A.E. & Dulvy, N.K. (2012) Thermal tolerance and the global redistribution of animals. Nature Climate Change, 2, 686–690. Terborgh, J. (1973) On the notion of favorableness in plant ecology. The American Naturalist, 107, 481–501. Thomas, C.D., Cameron, A., Green, R.E., Bakkenes, M., Beaumont, L.J., Collingham, Y.C., Erasmus, B.F.N., de Siqueira, M.F., Grainger, A., Hannah, L., Hughes, L., Huntley, B., van Jaarsveld, A.S., Midgley, G.F., Miles, L., Ortega-Huerta, M.A., Peterson, A.T., Phillips, O.L. & Williams, S.E. (2004) Extinction risk from climate change. Nature, 427, 145–148. White, C.R. & Kearney, M.R. (2013) Determinants of interspecific variation in basal metabolic rate. Journal of Comparative Physiology B: Biochemical, Systemic, and Environmental Physiology, 183, 1–26. Wiens, J.J. (2004) Speciation and ecology revisited: phylogenetic niche conservatism and the origin of species. Evolution, 58, 193–197. Wiens, J.J. (2008) Commentary on Losos (2008): niche conservatism deja vu. Ecology Letters, 11, 1004–1005. Wiens, J.J. & Donoghue, M.J. (2004) Historical biogeography, ecology and species richness. Trends in Ecology and Evolution, 19, 639–644.
2196
Wiens, J.J. & Graham, C.H. (2005) Niche conservatism: integrating evolution, ecology, and conservation biology. Annual Review of Ecology, Evolution, and Systematics, 36, 519–539. Wiersma, P., Mu~ noz-Garcia, A., Walker, A. & Williams, J.B. (2007) Tropical birds have a slow pace of life. Proceedings of the National Academy of Sciences USA, 104, 9340–9345. Wilson, A.-L., Brown, M. & Downs, C.T. (2011) Seasonal variation in metabolic rate of a medium-sized frugivore, the Knysana turaco (Tauraco corythaix). Journal of Thermal Biology, 36, 167–172. Wilson, D.E. & Reeder, D.M. (2005) Mammal species of the world: a taxonomic and geographic reference. John Hopkins University Press, Baltimore, MD. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article: Appendix S1 Data table containing physiological and geographical data for bird and mammal species. Appendix S2 Additional figures (Figs S1–S4). BIOSKETCH Imran Khaliq is a PhD student and is broadly interested in macrophysiology, macroecology and macroevolution. His work employs physiological, phylogenetic and biogeographical approaches to understand the distribution of species. Author contributions: I.K., S.F. and C.H. conceived the idea, R.P. contributed data. I.K. collected and analysed the data and led the writing. All authors discussed the results and contributed to the writing. Editor: Sßerban Prochesß
Journal of Biogeography 42, 2187–2196 ª 2015 John Wiley & Sons Ltd
