Explaining Local-Scale Species Distributions: Relative Contributions of Spatial Autocorrelation and Landscape Heterogeneity for an Avian Assemblage

摘要: Understanding interactions between mobile species distributions and landcover characteristics remains an outstanding challenge in ecology. Multiple factors could explain including endogenous evolutionary traits leading to conspecific clustering habitat features that support life history requirements. Birds are a useful taxon for examining hypotheses about the relative importance of these among community. We developed hierarchical Bayes approach model relationships bird occupancy local variables accounting spatial autocorrelation, similarities, partial observability. fit alternative models detections 90 observed during repeat visits 316 point-counts forming 400-m grid throughout Patuxent Wildlife Research Refuge Maryland, USA. Models with performed significantly better than our autologistic null models, supporting hypothesis heterogeneity is important as exogenous driver distributions. Conspecific alone was comparatively poor descriptor community composition, but there evidence autocorrelation all species. Considerable uncertainty whether combined most parsimonious describing at scale. Spatial structuring may be weaker intermediate scales within which dispersal less frequent, information flows localized, types become spatially diversified therefore exhibit little aggregation. Examining such across assemblages contributes understanding community-level associations conspecifics landscape composition.