Testing the role of patch openness as a causal mechanism for apparent area sensitivity in a grassland specialist.

Area sensitivity, species being disproportionately present on larger habitat patches, has been identified in many taxa. We propose that some apparently area-sensitive species are actually responding to how open a habitat patch is, rather than to patch size. We tested this hypothesis for Bobolinks (Dolichonyx oryzivorus) by comparing density and occupancy to a novel openness index, patch area, and edge effects. Bobolink density and occupancy showed significant relationships with openness, but logistic models based on an openness occupancy threshold had greater explanatory power. Thresholds remained approximately consistent from June through August, and shifted to be more open in September. Variance partitioning supported the openness index as unique and relevant. We found no relationships between measures of body condition (body mass, body size, circulating corticosterone levels) and either openness or area. Our findings have implications for studies of area sensitivity, especially with regards to inconsistencies reported within species: specifically, (1) whether or not a study finds a species to be area sensitive may depend on whether small, open sites were sampled, and (2) area regressions were sensitive to observed densities at the largest sites, suggesting that variation in these fields could lead to inconsistent area sensitivity responses. Responses to openness may be a consequence of habitat selection mediated by predator effects. Finally, openness measures may have applications for predicting effects of habitat management or development, such as adding wind turbines, in open habitat.