The comparative evidence for urban species sorting by anthropogenic noise.

Anthropogenic noise is more intense at lower sound frequencies, which could decrease urban tolerance of animals with low-frequency vocalizations. Four large comparative studies tested whether anthropogenic noise filters bird species according to the sound frequencies they use and produced discrepant results. We reanalysed data from these studies to explain their different results. Urban tolerance of bird species (defined here as often occurring and breeding in cities) is very weakly related to urban preference or relative abundance (defined based on changes in population density from urban to nearby rural environments). Data on urban preference/abundance are potentially accurate for individual cities but differ among cities for the same species, whereas existing data on urban tolerance are coarser but provide a more global synthesis. Cross-species comparisons find a positive association between the sound frequency of song and urban tolerance, but not urban preference/abundance. We found that showing an association between song frequency and urban tolerance requires controlling for additional species traits that influence urban living. On the contrary, controlling for other species traits is not required to show a positive association between song frequency and use of noisy relative to quiet areas within the same type of environment. Together, comparative evidence indicates that masking by urban noise is part of a larger set of factors influencing urban living: all else being equal, species with high-frequency sounds are more likely to tolerate cities than species with low-frequency sounds, but they are not more likely to prefer, or to be more abundant in, urban than non-urban habitats.