Stream capture is a major driver of the retreat of the Blue Ridge Escarpment, but timescales of capture are not well understood. This study examines stream sediment geochemistry to establish a set of sediment source fingerprints which can be used to identify and date the capture of the Tallulah River. Statistical analyses show significant differences in U, Th, and certain REE enrichment. These differences result from variations in bedrock along the lengths of each river and a shift in relative stream powers after capture to favor mobilization or deposition of heavy elements. The observed differences should be sufficient to identify where Tallulah sediment appears in floodplains of the capturing Tugaloo River, facilitating future dating of the capture event. Understanding the timing of river capture will provide insight into the ongoing reshaping and redistribution of river systems and interactions of geomorphic processes in the continuing evolution of the southern Appalachian Mountains.

Long, linear, high-relief escarpments mark many of the world's passive margins. These Great Escarpments have been interpreted to be the result of isostatic flexure, parallel slope retreat, and divide migration which accompanies rifting. It is unclear whether all these escarpments share this origin. Also uncertain is whether these features are formed via stable, steady-state processes or by climatic shifts or tectonic rejuvenation. The Blue Ridge Escarpment, eastern North America's great escarpment, is no different. A number of hypotheses attempt to explain the Blue Ridge Escarpment. These include lithologic variation between Blue Ridge and Piedmont rocks, the distance to ultimate base level, as well as, escarpment retreat resulting from post/syn-rift warping or faulting. We approach this problem from two directions. The first involves topographic comparisons and geologic observations to recognize and track divide migration. The second approach uses U-Th/He thermochronometry along two scarp-normal transects. Topographic analysis used data extracted from DEMs to compare three zones - the Upland, the Piedmont and the scarp zone itself. Parameters such as relief, drainage density, hypsometry, and slope are often used as proxies for relative erosion rates and the degree of maturity of a landscape. Results from these analyses indicate that the Upland and Piedmont zones are distinct landscapes, sharing very few topographic similarities, yet neither appears significantly more erodible than the other. Examination of parameters in the proximity of the escarpment point toward more rapid erosion here. Field evidence of this rapid scarp erosion (and thus divide migration) lies in the presence of beheaded stream channels, cobble roundness, and clast provenance. U-Th/He thermochronometry is a low temperature technique that allows us to calculate when rock cooled below 60-70C. Temperature is used as a proxy for depth, from which we can extract an exhumation rate. This method allows us to further test scarp genesis hypotheses. Preliminary results show older ages (~160) from the Upland surface than on the Piedmont lowland (~100 Ma). This confirms that the Piedmont surface is distinct from the Upland and demonstrates that it has experienced greater erosion. There is also some indication that ages "jump" across the Bowens Creek/Brevard fault system. Lastly, the ages appear to become younger approaching the escarpment which is indicative of scarp migration. As these results are preliminary, more data is required to prove or disprove these conclusions. / Master of Science