Geographic variations in eco-evolutionary factors governing urban birds: The case of university campuses in China.

Urbanization alters natural habitats, restructures biotic communities and serves as a filter for selecting species from regional species pools. However, empirical evidence of the specific traits that allow species to persist in urban areas yields mixed results. More importantly, it remains unclear which traits are widespread for species utilizing urban spaces (urban utilizers) and which are environment-dependent traits. Using 745 bird species from 287 university/institute campuses in 74 cities and their species pools across China, we tested whether species that occur in urban areas are correlated with regards to their biological (body mass, beak shape, flight capacity and clutch size), ecological (diet diversity, niche width and habitat breadth), behavioural (foraging innovation) and evolutionary (diversification rate) attributes. We used Bayesian phylogenetic generalized linear mixed models to disentangle the relative roles of these predictors further, and to determine the extent to which the effects of these predictors varied among different cities. We found that urban birds were more phylogenetically clustered than expected by chance, and were generally characterized by a larger habitat breadth, faster diversification rate, more behavioural innovation and smaller body size. Notably, the relative effects of the attributes in explaining urban bird communities varied with city temperature and elevation, indicating that the filters used to determine urban species were environment dependent. We conclude that, while urban birds are typically small-sized, generalists, innovative and rapidly diversifying, the key traits that allow them to thrive vary spatially, depending on the climatic and topographic conditions of the city. These findings emphasize the importance of studying species communities within specific cities to better understand the contextual dependencies of key traits that are filtered by urban environments.