Avian response to changes in structure and composition of riparian and upland land cover in the Great Basin

We examined alternative scenarios of land cover that may affect status of breeding birds in the Great Basin and success of interventions to maintain their viability. Scientists and managers often assume that species of birds with similar resource needs and life history characteristics (guilds) are likely to have similar responses to environmental change, but rarely validate that assumption. We investigated whether avian guilds grouped by level of riparian dependence or nesting requirements have comparable, predictable responses to land use and climate. We modeled species occurrences using five years of empirical data on bird distributions, vegetation, and geographic coordinates. We used geospatial tools and satellite images to derive data on elevation and primary productivity, and used MARS and GLMMs to generate one set of environmental basis functions for each guild.  Expected predictive capacity of our models was evaluated with bootstrapping. Because our goal was prediction, not identification of key ecological variables for individual species, we reduced the potential suite of variables in our models a priori to six candidate predictors selected by experts on the biology of birds in the Great Basin. Birds typically had asymptotic relationships with vegetation variables, but more complex spatial patterns with latitude, longitude, and elevation. Responses of riparian obligates to vegetation were generally similar, and more predictable than responses of species guilds based on nest location. By predicting how different species are likely to respond to environmental change, we increase our capacity to facilitate management that will maintain a high proportion of native taxa.