Can climate explain interannual local extinctions among bird species?




Geographical variations in species richness are strongly related to temperature and precipitation. On ecological time-scales, these variations in species richness should reflect rates of immigration and local extinction (extirpation). Here we ask whether the probability of local extinction in passerine birds covaries with climate. Specifically, we test whether local extinctions increase with climatic harshness or with the climatic distance from a species' optimal climate.


USA and Canada.


We obtained bird counts from the North American Breeding Bird Survey (BBS) from 1967 to 2012. For each BBS route, we calculated the probability of interannual local extinction for each of 206 passerine birds. We then used linear mixed-effects models and structural equation modelling to relate local extinction rates to our hypothesized predictor variables: temperature, precipitation and their distance from the species' most occupied temperature and precipitation.


We found that local extinctions are nearly independent of temperature and precipitation: no climate is inherently more extinction-prone than any other. Similarly, the climatic distance from a species' maximally occupied temperature and precipitation has only an extremely weak positive effect on the probability of local extinction. We found that only abundance has a strong negative effect on the probability of local extinction.

Main conclusions

Although variations in local extinctions are typically spatially structured, we conclude that they are not related to contemporary climate in a consistent way among species. Broad-scale geographical gradients of species richness are unlikely to be driven by higher extinction rates in climatically harsh areas.