Using wildlife communities to improve vegetation classification for conserving biodiversity

Determining which vegetation types of organisms perceive similarly and classifying these types into groups that function as similar habitats are necessary steps toward expanding the focus of conservation strategies from single species to ecosystems. Therefore, the methods used to determine these habitat classifications are crucial to the successful design and implementation of these conservation strategies. Typically, this process has been accomplished through best professional judgement. We used quantitative techniques to group vegetation types into habitats based on the occurrence of breeding wildlife species (n = 420) in Oregon. After calculating faunal similarities among all regional vegetation types (n = 130), we used cluster analysis to group vegetation types into wildlife habitats. We classified the original 130 vegetation types into 30 wildlife habitat types that we believe function similarly. We tested this classification to assess whether vegetation types could be correctly classified into habitat types based on wildlife species composition. Discriminant analysis correctly classified 95% of the vegetation types into their wildlife habitat types, strengthening our confidence in this approach. This approach for classifying habitat types allows consistent development of conservation strategies at coarse resolutions and aids in identifying vegetation types where additional biodiversity surveys are needed. Finally, this approach can be refined continuously as the precision of vegetation mapping and our understanding of organism-habitat associations improve.