Interspecies physiological variation as a tool for cross-species

Global warming is now recognized as the domi-nant threat to biodiversity because even pro-tected populations and habitats are susceptible.Nonetheless, current criteria for evaluatingspecies’ relative endangerment remain purelyecological, and the accepted conservationstrategies of habitat preservation and populationmanagement assume that species can mountecological responses if afforded protection. Theinsidious threat from climate change is that itwill attenuate or preclude ecological responsesby species that are physiologically constrained;yet, quantitative, objective criteria for assessingrelative susceptibility of diverse taxa to warming-induced stress are wanting. We explored theutility of using interspecies physiological vari-ation for this purpose by relating species’ phys-iological phenotypes to landscape patterns ofecological and genetic exchange. Using a sala-mander model system in which ecological,genetic and physiological diversity are wellcharacterized, we found strong quantitativerelationships of basal metabolic rates (BMRs) toboth macroecological and phylogeographic pat-terns, with decreasing BMR leading to dispersallimitation (small contemporary ranges withmarked phylogeographic structure). Measuresof intrinsic physiological tolerance, which varysystematically with macroecological and phylo-geographic patterns, afford objective criteria forassessing endangerment across a wide rangeof species and should be incorporated into con-servation assessment criteria that currently relyexclusively upon ecological predictors.Keywords: climate change; phylogeography;range size; stress; conservation assessment;salamander1. INTRODUCTIONAnthropogenic warming is now the dominant threat toglobal biodiversity, but objective, operational criteriafor assessing species’ relative endangerment from cli-mate change remain wanting (IUCN 2001; Willott T Thomas et al. 2004). Present criteria areexclusively ecological (e.g. population size, range size),underlain by the implicit assumption that protectedspecies can mount biological responses to climatechange if afforded protection. This conservationscheme ignores species differences in physiologicaloptima and tolerances that determine their ecologicalpotential during climate change (Bernardo & Spotila2006). There is a pressing need in conservation biologyand climate change research for cross-species, quan-titative, objective criteria for assessing the susceptibilityof species to climate change-induced extinction.One attempt to address this shortcoming is phe-nomenological, using climatic descriptions of contem-porary ranges coupled with expected temperatures topredict future ranges. These bioclimatic models arebest case scenarios because, while they reflect existinginterspecific differences in projected ranges underglobal warming, they also implicitly assume that allspecies have similar potential to access and exploitpredicted climate space.Here we explore an alternative, mechanisticapproach that uses intrinsic properties of species topredict their responses to climate change. The first stepof this approach (examined here) is to assess whetherphysiological differences among species are quan-titatively related to population ecological and geneticdynamics. If so, such functional relationships could beused to make explicit quantitative predictions about notonly potential ranges, but also species’ ecological andpopulation genetic fates as warming ensues. Ecologistshave long appreciated that physiological diversity mustrelate to interspecific differences in habitat occupancyand use, but there are surprisingly few quantitative datademonstrating such a relationship. As Gaston (2003)observed: ‘impressions as to such differences abound,but seem little more developed than that.’ Comparedwith widely studied extrinsic regulators of dispersal andestablishment (e.g. interspecific interactions, climate),the impact of interspecific physiological differences onmacroecological patterns is underappreciated (Brown1995; Kirkpatrick & Barton 1997; Clarke 2003).Similarly, while ecological variables are being incorpor-ated into comparative phylogeographic analyses, theeffects of interspecific physiological variation arenot studied. Thus, demonstrations of quantitativeconnections between physiology and macroecologicalpatterns or, especially, phylogeographic patterns, aregenerally lacking.We studied physiological variation amongDesmognathus salamanders endemic to the Appala-chian Mountains of eastern North America. This