Date of Award


Level of Access Assigned by Author

Open-Access Thesis

Degree Name

Master of Science (MS)


Wildlife Ecology and Wildlife Conservation


Raymond J. O'Connor

Second Committee Member

George J. Jacobson, Jr.

Third Committee Member

Daniel J. Harrison


With mounting evidence that global temperatures have increased significantly over the last century and the projections of greater changes in climate by the end of this century, understanding the potential consequences of these changes for species is essential to conservation efforts. Here I evaluate the potential response of birds to projected climate change by using regression tree analysis to create models of species distributions under current conditions from Breeding Bird Survey data and then project these models onto General Circulation Model (GCM) scenarios of global climate change. Before modeling species responses to climate change, I selected seventeen bird species to evaluate several considerations that could influence the ability to effectively model species distributions. First, I addressed the spatial resolution of the analysis. GCM data are readily available at a relatively coarse-grain compared to bird data. Since the spatial resolution of an analysis can greatly affect the outcome, I, therefore, assessed the consequences of modeling bird abundance at the 640-km2 hexagonal grid (fine-grain) and the county resolution (coarse-grain) in the eastern United States. The results indicated that county resolution models produced good predictions of current bird distributions. Next, I compared two sets of climate data to ensure that the climate outputs from GCMs were as effective in modeling bird distributions as climate variables currently used in ecological studies. There were no differences of the overall model goodness of fit between the two sets of species models. The results from these analyses indicated that effective models of bird distributions at the county resolution could be constructed, provided both climate and land cover variables were present as predictors. Following these evaluations I was able to model current abundance for 152 bird species. These models were projected onto two GCM scenarios. The projected response of birds under the two GCM scenarios varied greatly among species. Overall, both GCM scenarios projected approximately 49% of the species to decrease markedly and 22% to increase in their eastern United States populations. These results indicate the potential for large shifts in bird distributions in response to global climate change. The heightened awareness of our Earth's increasing temperature has been linked to the rapid increase in greenhouse gases. Planting forests on marginal agricultural land has emerged as a promising proposal to sequester excess carbon dioxide, but none of these afforestation studies have considered the costs or benefits associated with impacts on wildlife. By combining information on current forest and farmland bird abundances with the results from simulations of carbon sequestration policies in South Carolina, Maine, and southern Wisconsin, it is possible to quantify the impacts of land use decisions on bird populations. I estimated losses respectively of 12.2 %, 10.8 %, and 1 1.7 % in farmland birds and gains of 2.5 %, 3.2 %, and 21.8 % in forest species in South Carolina, Maine, and southern Wisconsin. The results from this analysis reveal the importance of considering the effects of large-scale land use decisions on wildlife.