Welcome to the American Ornithological Society 2018 Annual Conference. We are pleased to have you join us at the lovely Hilton El Conquistador Resort in Tucson, AZ.
Studies investigating avian habitat preferences typically measure spatial use patterns at one or more spatial scales and draw inferences about the animals' preference for some resource or habitat type based on a resource selection function (RSF). By designing avian habitat selection studies that first assume the operative spatial scale(s) for consideration (e.g. home-range scale, micro-site-scale) and failing to verify the behavioral relevance of the selected spatial scales, researchers may be attributing selection behaviors to focal organisms that are, in truth, an artifact of the study design. In a recent analysis, Yanco and Linkhart measured habitat preferences of Flammulated Owls (Psiloscops flammeolus) at both home-range and micro-site scales in a post-fire environment. They suggested that different proportions of high-severity burned forest in owl home ranges, as compared to the entire burn scar, represented habitat selection behavior by the owls at the home-range scale. Here we use an agent-based model to explore whether the patterns of habitat proportions described by Yanco and Linkhart could have derived from processes unrelated to selection at the home-range scale. Specifically, we show that supplying a simulated “agent” with a set of decision rules that do not account for any home-range-scale decisions can produce RSFs that suggest habitat selection at home range scales. These results suggest that RSFs may incorrectly ascribe behaviors to focal systems because of mismatched scales, and we encourage researchers to use caution when interpreting results via a single explanation.
