Nest-site selection of passerines: effects of geographic scale and public and personal information

Abstract

Nest-site selection is an important determinant of individual fitness in birds. Understanding what information individuals use to choose nest sites is therefore important for understanding the evolution of nest-site selection, the dynamics of populations, and the conservation of species. We used five years of mark-recapture data for Mountain Bluebirds (Sialia currucoides) to examine how dispersal probability and nest-site selection vary with potential cues of nest-site quality. Dispersal distance between breeding seasons and nest-site selection were modeled as a function of personal reproductive success, conspecific density, conspecific reproductive success, and habitat type. Between years, the dispersal probability was related to personal reproductive success, not conspecific information, and individuals fledging fewer young dispersed longer distances. For dispersing individuals, the probability that a nest site was selected in year i was negatively related to distance from the nest site selected in year i - 1 for all age and sex classes, and positively related to conspecific density and reproductive success in year i - 1 for both second-year (SY) and after-hatch-year (AHY) females. However, nest-site selection in year i was more strongly related to conspecific density in year i- 1 for hatch-year (HY) females and was much more strongly related to the reproductive success of conspecifics in year i - 1 for AHY females. Nest-site selection of HY and AHY males was not consistently related to the metrics of conspecific information, but we suspect that relationships were obscured by competitive interactions. We found no evidence indicating that individuals respond differently to conspecific information at longer distances, suggesting that individuals limit dispersal to areas where they have prior knowledge. We predict that these patterns of nest-site selection will allow birds to loosely track nest-site quality and maintain an ideal free distribution, where average fitness is equal in all habitat types.