Background/Question/Methods Anthropogenic change has altered the composition and function of ecological communities across the globe. As a result, there is a need for studies examining observed community compositional change and determining whether and how anthropogenic change drivers may be influencing that turnover. In particular, it will also be important to determine to what extent community turnover is idiosyncratic or if turnover can be explained by predictable responses across species based on traits or niche characteristics. Avian communities are experiencing compositional change as well as observed declines in many species, however the importance of different anthropogenic change drivers in explaining compositional changes across scales remains unknown. Here, we measured avian compositional change in species abundances across North America from 1990 to 2016 in the Breeding Bird Survey using an ordination method to describe the directionality in the trajectory of the community in ordination space, and model that directionality as a function of land use and climate change drivers from local to regional scales. We also examined how community directionality may be attributed to changes in individual species or species groups, including foraging guilds, trophic groups, migratory distance, and breeding biomes. Results/Conclusions We found that at local scales, land use change (change in proportion cover) explained a greater proportion of variance in community directionality than climate change variables, while as scale increased to regional communities, trends in temperature explained a greater proportion of variance in directionality. Overall, our models explained a greater total proportion of variance in directionality at larger spatial scales as well. We also found across the study region, changes in directionality could be attributed to one of a handful of species undergoing strong expansions over the study time period (such as Eurasian Collared-dove, Streptopelia decaocto) or strong declines, such as Northern Bobwhite (Colinus virginianus). We did not observe consistent patterns in compositional change in any trait groups we examined except for those that included previously identified highly influential species. Our results have two important implications: first, the relative importance of different anthropogenic change drivers may vary with scale, which should be considered in studies modeling impacts of global change on biodiversity. Second, in North American avian communities, individual species undergoing large shifts in population may drive signals in compositional change, and composite community turnover metrics should be carefully selected as a result.
