Constraints on vertebrate range size predict extinction risk

Abstract

AbstractAimThe only factor in the fossil record that consistently buffers against extinction risk is large geographical range. We ask whether extant vertebrate species with the smallest geographical range for their body size have a higher extinction risk, and thus whether the lower bound of the modern range–body size relationship could serve as an effective conservation prioritization tool.LocationGlobal in scope.Time periodModern.Major taxa studiedSix classes of vertebrates.MethodsWe compiled a database of geographical range, body size and extinction risk for six vertebrate classes (n = 26,076). We characterized the shape of the relationship between geographical range and body size for each class, using 90% and 10% quantile regression to describe the upper and lower bounds, respectively. We then evaluated the degree of extinction vulnerability of species at the lower bound of the regression using generalized linear mixed models. All analyses accounted for phylogenetic dependence between related species.ResultsThe relationships between species ranges and body sizes were generally positive at both the upper and the lower bounds, and segmented (nonlinear) relationships were common. Despite this variability, species near the lower boundary of the relationship were more often in higher extinction risk categories, and this remained true when the role of range size in the International Union for Conservation of Nature Red Listing criteria was accounted for.Main conclusionsVariability in the upper and lower bounds of the range–body size relationship suggests that some classes of vertebrates exhibit combinations of ranges and body sizes that might not reflect historical patterns. Nonetheless, the range–body size relationship remains a reliable and useful predictor of extinction risk, more so than range size does alone. The range–body size relationship could therefore be used to track the trajectories of species towards or away from an extinction threshold and allow the tracking of how different human activities alter the range–body size relationship.