Robust predictions of species distribution: Spatial habitat models for a brood parasite

摘要: Empirical habitat selection models aid managers by predicting distribution patterns, but recent critiques urge a fuller application of available methods. We used suite modelling techniques to identify factors influencing the brown-headed cowbirds (Molothrus ater), brood parasite capable causing population declines in some host species. data on cowbird occurrence Southern Gulf Islands, Canada, estimate influence landscape features beginning with set priori based literature. Best were also constructed using stepwise logistic and autologistic regression, only widely data, these their predictive maps then evaluated. Cowbirds predicted proximity potential feeding areas context. The best model included landcover, cattle, distance urban area agriculture. Model performance was improved restricted slightly poorer. All robust internal validation bootstrapping when compared an independent empirical dataset parasitism rates 12 sites monitored for up 8 years. Our should help land make spatial predictions about impact particular hosts, as well evaluate land-use decisions that could abundance fitness. evaluated several approaches interest conservationists suggestions future studies are likely affect conservation use planning.