The role of spatial scale in breeding habitat selection of neotropical migrant birds in Albemarle County, Virginia

The determination of variables controlling species distributions has been a fundamental area of study in ecology. Traditionally, species-habitat relationships were studied on small spatial scales, i.e. the microhabitat. With the advances of island biogeography theory in landscape ecology and the development of Geographic Information Systems (GIS), the examination of species-habitat relationships expanded to encompass a range of spatial scales from microhabitat to watershed levels. When studying highly mobile animals such as birds, a multiple spatial scale approach is highly desirable. The overall objective of this research was to determine the spatial scale(s) of habitat variables that influence distributions of Neotropical migrant birds in Albemarle County, Virginia. Neotropical migrant species were targeted because many species appear to be suffering population declines on regional and continental scales. Hypotheses in this study tested the definition and breadth of variables and the distinct spatial scales critical to breeding habitat associations by individual species and species groups. This study is unique because it cross examines: (1) species groups - forest interior and scrub-associated species; (2) spatial scales - microhabitat and landscape scales (<100m to 3km); and (3) land cover types - three forest types, scrub, herbaceous and human disturbed. Among the forest interior species the correlation with human disturbance and fragmentation varied by species. Although Ovenbird distributions were dictated by the presence of large contiguous, undisturbed forests, Worm-eating Warbler and Wood Thrush distributions relied on different variables. Fragmentation did not emerge as a fundamental determinant in forest interior species distributions. In contrast to the forest interior species, scrub species possessed common habitat variables. This research underscored this group's complex relationship with human disturbance. Scrub species prefer early successional habitat, often created through human disturbance; however, humans have eliminated most of the scrub habitat in the county, decreasing overall habitat availability. No single spatial scale dominated this analysis. Microhabitat variables were more influential on species distributions, but landscape scale variables factored into habitat associations for all but two species. An analysis of wind bias exposed a source of variability and error when conducting animal population surveys based on their vocalizations. Wind speeds as low as 2–6 mph altered the propagation and detection of sound (birdcalls) through the atmosphere, effectively changing the sampling plot radius. Thus, for some species, sampling in windy versus calm conditions changed the hierarchy of important habitat variables and spatial scales defining species distributions. The biodiversity map developed in this research indicates regions of the county that should be considered for protection of Neotropical migrant birds. Because this analysis suggested that local Brown-headed Cowbird populations are relatively low, conservation management in Albemarle County may gain significance and have implications on regional and continental scales. This local scale approach is the best, and perhaps only practical strategy for conservation of species because land use management, development and conservation decisions are usually made at the local (county or smaller) level.