Disentangling the effects of climate and landscape change on bird population trends in the western U.S. and Canada

Changes in climate are often assumed result in changes to species’ ranges, with potential impacts on natural system functioning and ecosystem services. ‘Climate envelope models’, which rely on correlations between climate and species distributions, have been used to predict the future of biodiversity under these assumptions. However, other factors including land-cover, dispersal ability and interspecific competition and facilitation may play an important role in driving species distributions and population trends either alone or in combination with climate. In an ongoing project, we used long-term data on bird distributions and abundance to develop climate envelope and land-use models for 161 species in order to provide a multi-species test of the degree to which climate envelope versus land-use models are useful in predicting species distributions and population trends of birds in forest ecosystems of the western U.S. and Canada. Our results suggest that models describing associations between climatic variables and abundance patterns can be used for some species to predict changes through time, and that changes in climate have already driven shifts in the geographic patterns of abundance of bird populations in western North America. For other species, models using land-use variables including raw remote-sensing variables may provide the best predictions for abundance change. The results of this research showing the reliability of models across multiple species will aid managers in understanding which species are most vulnerable to changes from climate, land-use change and their interaction.