Search for a command to run...
Spatial scale is fundamental in understanding species–landscape relationships because species’ responses to landscape characteristics typically vary across scales. Nonetheless, such scales are often unidentified or unreliably predicted by theory. Many landscapes worldwide are urbanizing, yet the spatial scaling of species’ responses to urbanization is poorly understood. We investigated the spatial scaling of urbanization effects on a community of 15 mammal species using ~60 000 wildlife detections collected from a constellation of 207 camera traps across an extensive urban park system. We embedded a bivariate Gaussian kernel in hierarchical multi‐species models to determine two scales of effect (a scale of maximal effect and a broader scale of cumulative landscape effect) for two biological responses (occupancy and site visit frequency) across two seasons (winter and summer) for each species. We then assessed whether scales of effect varied according to theoretical predictions associated with biological responses and species traits (body size and mobility). Scales of effect ranged from < 50 m to > 9000 m and varied among species, but not as predicted by theory. Species’ occupancy generally showed a weak response to urbanization and the scale of this effect was both highly uncertain and consistent across species. We did not detect any relationship between scales of effect and species’ body size or mobility, nor was there any evident pattern of scaling across biological response or seasons. These results imply that 1) urbanization effects on mammals manifest across a very broad spectrum of spatial scales, and 2) current theories that a priori predict the scale at which urbanization affects mammals may be of limited use within a given system. Overall, this study suggests that developing general theory regarding the scaling of species–landscape relationships requires additional empirical work conducted across multiple species, systems and timescales.