Does climate or species interactions differentially drive northern vs. southern range limits?

Background/Question/Methods: Determining the factors that limit the distribution of species is a central goal of ecology and evolutionary biology. Biologists are being called upon to predict species distributional responses to climate change, biological invasions, habitat fragmentation, and a number of other anthropogenic pressures. Generalizations about range determinants that are applicable to multiple species at coarse spatial scales would be useful when predicting species distributional responses. A number of ecological and evolutionary processes have been identified as playing a role in determining range limits yet there are few widely applicable generalizations that can be made. The predictive application of existing biogeographic hypotheses is often hindered by a disparity between literature addressing issues from a purely theoretical standpoint and those examining empirical patterns. One such biogeographic hypothesis concerns the role of climate vs. species interactions at species’ northern and southern distributional limits. The pervasive view is that northern range boundaries are determined by physiological tolerance limits driven by climate, while species interactions are more important in defining and limiting the southern distributional limits, at least for those organisms in the Northern Hemisphere. We used ecological niche modeling (ENM) in Maxent to investigate the role of climate in determining the northern and southern distributional limits of all North American amphibian and reptile species. Specifically we address geographic discordance between species' known distributions and those distributions as predicted by the ENM models. 

Results/Conclusions: If climatic conditions are the dominant factor limiting northern distributions we predict ENMs to show little to no overprediction beyond species known northern range boundaries. Likewise if species interactions override the importance of climate at the southern range limits, we predict ENMs to show greater overprediction beyond known southern range boundaries. We summarize the results for amphibians, reptiles, snakes, lizards, salamanders, and frogs/toads. Results are not consistent with the predicted pattern; rather, in general, the patterns are of overprediction at the northern boundary.