Abstract
Urban bird communities are homogenized across large spatial scales, suggesting that the urban environment acts as an environmental filter. We hypothesize that large scale commonness is a better predictor of urban affinity of birds than any particular species trait.
We estimated the relative importance of taxonomy, reproductive, ecological and morphological traits, and commonness of individual bird species. We compiled data on i) breeding bird communities of 41 European cities from urban bird atlases, and ii) regional bird assemblages defined by nine grid cells of the Atlas of European Breeding Bird around each city, and quantified the urban affinity of each species by comparing its incidence in cities and in randomly drawn communities from respective regional assemblages.
Conditional inference tree-based random forest analysis was utilized to assess the importance of individual predictors. A sign test was used to detect differences between congeneric pairs of species with contrasting affinity to cities.
Birds associated with woody habitats and those having altricial chicks had higher affinity for cities. Of the other reproductive traits, only clutch size showed an association with urban affinity.
Different bird orders differed significantly in their urban affinity, exemplifying the homogenizing effect of cities. However, by far the most important factor associated with bird tolerance to the urban environment was species commonness, indicating that either the traits associated with commonness, or population effects driven by commonness, are responsible for their presence in cities.