Phenotypic signatures of urbanization are scale-dependent: A multi-trait study on a classic urban exploiter

Understanding at which spatial scales anthropogenic selection pressures operate most strongly is a prerequisite for efficient conservation and management of urban biodiversity. Heterogeneity in findings on the strength and direction of urbanization effects may result from a lack of consensus on which spatial scales are most adequate when studying biotic effects of urbanization. Therefore, here, using the house sparrow (Passer domesticus) as model, we test the hypothesis that more than one spatial scale will explain variation among phenotypic stress markers. By applying a unique hierarchical sampling design enabling us to differentiate between local and regional effects of urbanization, we here show that the strength and direction of relationships with the percentage of built-up area - a simple structural measure of urbanization - vary among phenotypic stress markers and across the spatial range over which urbanization is measured. While inverse relationships with scaled body mass and bill height of adult house sparrows (Passer domesticus) were strongest when the degree of urbanization was quantified at city-level, similar relationships with corticosterone concentrations in feathers were only detected at the scale of individual home ranges. In contrast, tarsus length, wing length, and two measures of feather development were not significantly related to urbanization at any spatial scale. As the suite of phenotypic stress markers applied in this study revealed signatures of urbanization over a broad spatial range, we conclude that measures aimed at mitigating impacts of urbanization on free-ranging populations should best be implemented at multiple spatial scales too.