Using biophysical factors to predict regional biodiversity potential in the pacific and inland northwest
The distribution of biodiversity across the Pacific and Inland Northwest (PINW) is currently not well known. This knowledge is important for setting management priorities within agency and industrial forest holdings. Previous attempts to map biodiversity across regions largely relied on scaling up from our understanding of stand-scale habitat relationships. This approach does not consider the effects of climate, topography, vegetation productivity, and other biophysical variables that have been found to be important predictors of biodiversity at continental scales. The goal of this study is to integrate consideration of traditional environmental variables such as habitat type and recently available data on factors such as primary productivity derived from satellite data and computer simulation models to assess predictors of bird, tree, and shrub, species richness across the PINW. Bird species richness, Shannon's Diversity Index, and sensitive species richness were generated from the USGS Breeding Bird Survey data (BBS). Predictor data on vegetation composition, topography, climate, and vegetation productivity were derived or generated from many sources. Model selection procedures were used to evaluate relationships between the predictor data and the biodiversity response variables. The resulting best models were used to extrapolate bird, tree, and shrub richness across the study area. These maps were integrated to depict areas of high biodiversity potential and to evaluate the locations of forest industry lands relative to these areas. The best model for bird richness (R 2=.58) included terms for land cover heterogeneity, percent of conifer along the route, net primary productivity, solar radiation in May and June, vapor pressure deficit in July, precipitation in May and June, and slope. The best model for shrubs (R 2=.82) also included topographic heterogeneity, maximum leaf area, and log area sampled. For tree richness, the best model (R 2=.57) also included topographic heterogeneity and log area sampled. Observed bird species richness was high in an arc extending from the Siskiyou Mountains in southwest Oregon, northward along the east side of the Cascades, across the Okanogan Region of north central and northeastern Washington and into northern Idaho and Northwest Montana. Richness of sensitive bird species was highest in the forested ecoregions east of the Cascades and in the Siskiyou Mountains. The major industrial forest lands occupy environmental settings that are relatively high in predicted bird richness and are in proximity to areas of high bird, tree, and shrub biodiversity potential. The results suggest that there is no single overarching factor that controls bird and tree richness across the PINW. Rather, habitat composition, vegetation productivity, climate, and topography interact to drive diversity, with the strength of each factor varying across the region.
