The influence of thermal tolerances on geographical ranges of endotherms

Abstract

Aim To understand how climatic conditions influence the geographical distributions of species and their potential responses to climate change, we investigated the relationships between the thermal tolerances of species and the size and limits of their distributions. We tested two hypotheses for endotherms: the climatic variability hypothesis, which predicts increases in range size with increasing breadth of thermal tolerance, and the climatic extreme hypothesis, which predicts that range limits are related to thermal tolerance limits. Furthermore, we tested whether these relationships differ between temperate and tropical areas. Location Global. Time period Present. Major taxa studied Birds and mammals. Methods We compiled data on thermal tolerances that had been measured in physiological experiments for 453 endothermic species, along with information on geographical ranges and climatic conditions. We applied phylogenetic generalized least square regressions to test for relationships between thermal tolerance and (a) range size or limits and (b) breadth and extremes of the climatic conditions that each species experiences across its distribution. Results We found that range size was not related to the breadth thermal tolerance for endotherms. However, the range limits at high latitudes as well as the minimum temperatures experienced by species were closely related to the physiological cold tolerances of species. These relationships were particularly strong in temperate regions, but these patterns were not found in the tropics. Main conclusions Our results are inconsistent with the predictions of the climatic variability hypothesis, but are in line with the predictions of the climatic extreme hypothesis. Furthermore, the factors determining species distributions do not appear to be the same in tropical and temperate regions. Our study emphasizes the need to combine spatially explicit distribution models with information from physiological experiments in order to capture regional differences and improve predictions of the responses of species to climate change.

Addressing common pitfalls does not provide more support to geographical and ecological abundant‐centre hypotheses

“…Categorizing species into these two extremes is challenging as territorial behaviour is uncertain for most species, or it can manifest only under certain circumstances, at the individual or group level, or only within one of the two sexes. forest margin), or biogeographic barriers such as mountains or coastlines (but see Lee-Yaw et al 2016, Khaliq et al 2017). If this is the case, the test of the theory would be extremely challenging because the same definition of geographic range boundaries is blurry and depends on the methodological approach used and the data available (Fortin et al 2005).…”

“…For example the geographic range shape can be bounded by a change in habitat type (e.g. forest margin), or biogeographic barriers such as mountains or coastlines (but see Lee-Yaw et al 2016, Khaliq et al 2017). An influence of any of these factors would not only violate the premises of the geographical abundant-centre interpretation, but it would also bias the niche suitability measurements.…”

Parasite intensity is driven by temperature in a wild bird

“…Calosi et al 2010) and in endotherms there is evidence that range limits relate to thermal tolerance limits (e.g. Khaliq et al 2017). P. azurea and P. falcozi, the two species present in our study area, are both widespread over the Palaearctic in a wide range of host species (Wesolowski 2001, Matyukhin andKrivosheina 2008).…”

An oversimplification of physiological principles leads to flawed macroecological analyses

“…In light of the rapidly changing climate, there is an urgent need to develop a mechanistic understanding of how physiological functioning mediates ecological patterns. Recently, there has been a spate of papers using analyses that scale up from a standard physiological model, the Scholander–Irving model, to make predictions about range constraints on endothermic vertebrates (Buckley, Khaliq, Swanson, & Hof, ; Fristoe et al, ; Khaliq, Böhning‐Gaese, Prinzinger, Pfenninger, & Hof, ; Khaliq, Hof, Prinzinger, Böhning‐Gaese, & Pfenninger, ). Here, we argue that oversimplifications of the Scholander–Irving model and the use of questionable datasets lead to questionable macrophysiological analyses.…”