Date of Award


Level of Access Assigned by Author

Campus-Only Dissertation

Degree Name

Doctor of Philosophy (PhD)


Ecology and Environmental Sciences


Cynthia S. Loftin

Second Committee Member

William A. Halteman

Third Committee Member

Steven A. Sader


Initially my goal was to examine relationships between vegetation composition and structure and amphibian abundance within Okefenokee National Wildlife Refuge (ONWR). Unfortunately, the data were unsuitable for extrapolating species-habitat relationships beyond the sampled areas (Appendix A). Therefore, I examined ways of restoring natural processes (i.e., fire) responsible for creating vegetative diversity and patchiness within which amphibians have evolved. I used satellite data to provide managers with an accurate and efficient way to extract and monitor land cover changes due to fire. Results suggest ONWR is changing to a hardwood, fire intolerant system under a policy that disrupts the natural fire regime. I also tested two assumptions of fire policy: 1) wildfire activity has increased, and 2) increases are due to fuel accumulation following decades of fire suppression. Both assumptions are invalid for ONWR and burning is more dependent on fire weather than on age or spatial patterns of fuels. Therefore, I explored relationships among the El Niño Southern Oscillation (ENSO) cycle, local weather patterns, and fire frequency and extent. The La Niña phase of ENS0 was associated with lower dry season rainfall and lower water levels. These two variables are correlated with increases in the number of fires and area burned. Thus, large fires within ONWR are inevitable events, and current policy should be revised to consider global controls on fire. I developed a spatial model to investigate effects of different fire management scenarios on the composition and distribution of vegetation in ONWR. Results suggest current policies may be causing the ONWR system to shift into a different ecological state, resulting in a loss of vegetative diversity, patchiness, and resilience. Policy allowing a more natural fire cycle may prevent such a shift and could be accomplished through prescribed burns in specific communities or by allowing natural fires to burn. Finally, I gain perspective into the cultural attributes of research and management communities that combine to inhibit institutional change and critique institutional arrangements that determine fire policy. I propose a model to facilitate exchange of information between researchers and managers in hopes of implementing an adaptive management program at ONWR.