The phylogenetic and functional diversity of regional breeding bird assemblages is reduced and constricted through urbanization

Aim: Urbanization broadly affects the phylogenetic and functional diversity of natural communities through a variety of processes including habitat loss and the introduction of non-native species. Due to the challenge of acquiring direct measurements, these effects have been studied primarily using “space-for-time” substitution where spatial urbanization gradients are used to infer the consequences of urbanization occurring across time. The ability of alternative sampling designs to replicate the findings derived using space-for-time substitution has not been tested. Location: Global. Methods: We contrasted the phylogenetic and functional diversity of breeding bird assemblages in 58 cities worldwide with the corresponding regional breeding bird assemblages estimated using geographic range maps. Results: Compared to regional assemblages, urban assemblages contained lower phylogenetic diversity, lower phylogenetic beta diversity, a reduction in the least evolutionary distinct species and the loss of the most evolutionarily distinct species. We found no evidence that these effects were related to the presence of non-native species. Urban assemblages contained fewer aquatic species and fewer aquatic foraging species. The distribution of body size and range size narrowed for urban assemblages with the loss of species at both tails of the distribution, especially large bodied and broadly distributed species. Urban assemblages contained a greater proportion of species classified as passerines, doves or pigeons; species identified as granivores; species that forage within vegetation or in the air; and species with more generalized associations with foraging strata. Main conclusions: Urbanization is associated with the overall reduction and constriction of phylogenetic and functional diversity, results that largely replicate those generated using space-for-time substitution, increasing our confidence in the quality of the combined inferences. When direct measurements are unavailable, our findings emphasize the value of developing independent sampling methods that broaden and reinforce our understanding of the ecological implications of urbanization.