Patterns in Wildlife–Vegetation Association Matrices

Abstract

We explored patterns in species–vegetation associations for vertebrate species found in forest communities in the northeastern U.S. Associations compiled by Degraaf et al.were used as the basis for our analyses. Three hundred and thirty-six vertebrate species and 45 vegetative cover types were investigated. Vertebrates were grouped as mammals, amphibians, reptiles and birds, and vegetation classes were aggregated hierarchically into 3, 6, 11 or 45 vegetation cover types. We used correspondence analysis to explore these data for patterns, with the objective of identifying dual ordinations of species and vegetation cover types expressing the strongest possible pattern of association between species and vegetation. Each level of vegetative aggregation was used in combination with each vertebrate grouping in the analyses. We found that associations were not particularly strong for any of the vertebrate groupings at any level of vegetative aggregation. Of the four taxonomic groups, birds appeared to be the most strongly associated with vegetation cover type. For all vertebrate groups the strength of the association with vegetative cover types declined with the level of vegetative aggregation. Results suggest that the vegetative cover types in Degraaf et al.are of marginal value in representing patterns of vertebrate biodiversity, and that ecological attributes other than vegetative cover types are more important in recognizing patterns of species occurrence. They also suggest that the level of detail in a classification of vegetative cover can influence patterns of association exhibited by vertebrate species. This must be kept in mind when biologists use species–vegetation associations to develop maps of vertebrate diversity.