Intraspecific variation in lizard heat tolerance alters estimates of climate impact.

Research addressing the effects of global warming on the distribution and persistence of species generally assumes that population variation in thermal tolerance is spatially constant or overridden by interspecific variation. Typically, this rationale is implicit in sourcing one critical thermal maximum (CT ) population estimate per species to model spatiotemporal cross-taxa variation in heat tolerance. Theory suggests that such an approach could result in biased or imprecise estimates and forecasts of impact from climate warming, but limited empirical evidence in support of those expectations exists. We experimentally quantify the magnitude of intraspecific variation in CT among lizard populations, and the extent to which incorporating such variability can alter estimates of climate impact through a biophysical model. To do so, we measured CT from 59 populations of 15 Iberian lizard species (304 individuals). The overall median CT across all individuals from all species was 42.8°C and ranged from 40.5 to 48.3°C, with species medians decreasing through xeric, climate-generalist and mesic taxa. We found strong statistical support for intraspecific differentiation in CT by up to a median of 3°C among populations. We show that annual restricted activity (operative temperature > CT ) over the Iberian distribution of our study species differs by a median of >80 hr per 25-km grid cell based on different population-level CT estimates. This discrepancy leads to predictions of spatial variation in annual restricted activity to change by more than 20 days for six of the study species. Considering that during restriction periods, reptiles should be unable to feed and reproduce, current projections of climate-change impacts on the fitness of ectotherm fauna could be under- or over-estimated depending on which population is chosen to represent the physiological spectra of the species in question. Mapping heat tolerance over the full geographical ranges of single species is thus critical to address cross-taxa patterns and drivers of heat tolerance in a biologically comprehensive way.