Global patterns and predictors of avian population density

Abstract

AbstractAimHow population density varies across animal species in the context of environmental gradients, and associated migratory strategies, remains poorly understood. The recent influx of avian trait data and population density estimates allows these patterns to be described and explored in unprecedented detail. This study aims to identify the main macroecological drivers of population density in birds.LocationGlobal.Time period1970–2021.Major taxa studiedBirds (Aves).MethodsWe collated a dataset of 5072 local population density estimates for 1853 species and modelled population density as a function of trait and environmental predictors in a Bayesian framework accounting for phylogenetic and spatial autocorrelation. We explored the influence of body mass, diet, primary lifestyle, mating system, nesting behaviour, territoriality, and migratory behaviour on population density, accounting for a range of environmental variables, including preferred habitat type, primary productivity, precipitation and temperature. Based on this empirical baseline, we then predicted the mean population density for 9089 species of birds and estimated global geographic patterns of bird population density.ResultsPopulation density was lower in species with larger body mass and higher trophic levels, and also declined in territorial species, migratory species, brood parasites and species inhabiting resource‐poor habitat types (e.g., deserts). Conversely, population density increased in cooperative breeders. Environmental drivers were most influential for migratory birds, with precipitation and temperature both associated with higher population density. Overall, bird population densities were higher at lower latitudes.Main conclusionsOur results support previous findings on the role of body mass, diet and environmental gradients, but also reveal novel species‐specific drivers of avian densities related to reproduction, migration and resource‐holding behaviour. Substantial fine‐scale variation remains unexplained. We provide a global dataset of population density predictions for use in macroecological analyses and conservation assessments.