Print

Reuter, Joanna M.

Geology

2005

MS

I use cosmogenic ¹⁰Be analyses to address both applied and basic science questions regarding rates and patterns of erosion in the 71,250 km² Susquehanna River Basin of New York, Pennsylvania, and Maryland. Measurements of in situ-produced ¹⁰Be from 88 fluvial sediment samples constrain basin-scale erosion rates on a 104 to 105 year time scale, and four bedrock samples provide ridge-top erosion rates. Sediment samples are from two groups: (1) 60 samples are from small (0.6 to 25 km²), non-glaciated basins underlain by a single lithology; these were selected through geographic information systems (GIS) analysis; (2) 28 samples are from USGS stream gages and represent complex basins of multiple lithologies and varying degrees of present-day land use; some of the USGS basins were glaciated during the Pleistocene. Erosion rates range from 4 to 54 m/My in the southern, non-glaciated part of the Susquehanna River Basin. The broadest range of erosion rates occurs among the small, GIS-selected basins, but the average erosion rate of this group (16 ± 10 m/My, mean ± standard deviation) is similar to that of the larger USGS basins (14 ± 4 m/My). The erosion rates from the Susquehanna River Basin are consistent with rates from other regions of relatively low relief and tectonic quiescence, as determined through a comparison with more than 360 other basins for which ¹⁰Be data are available worldwide. My analysis of erosion rate patterns in the Susquehanna River Basin utilized GIS-selected basins to test for relationships between erosion rate, mean basin slope, lithology, and physiographic province. Overall, erosion rate correlates positively with slope (R² = 0.57), but correlations vary by physiographic province, with rrogressively weaker relations in a down-basin direction (Appalachian Plateaus, R² = 0.72; Valley and Ridge, R² = 0.37; Piedmont, R²= 0). After accounting for slope, lithology does not appear to affect basin-scale erosion rates, based on comparisons between sandstone and shale basins in the Valley and Ridge. The relationships established among the small basins lend confidence that the inferred erosion rates for the lithologically complex and human-impacted USGS basins that are not glaciated are robust. However, samples from glaciated basins yield ¹⁰Be concentrations (0.5-1.2 x 10⁵ atoms g⁻¹quartz) that are consistently lower than those for similarly sized basins south of the glacial margin (1.7-4.9 x 10⁵ atoms g⁻¹ quartz). This discrepancy results from violation of the steady-state erosion assumption in previously glaciated basins. Thus, data from these basins are not directly interpretable as erosion rates. USGS basins have sediment yield records that can be compared with ¹⁰Be erosion rates to assess whether background rates of sediment generation are in equilibrium with contemporary sediment yield. These comparisons indicate that contemporary sediment yields exceed ¹⁰Be sediment generation rates by up to an order of magnitude. Sediment yields are particularly high relative to ¹⁰Be sediment generation rates in the agricultural southeastern part of the Susquehanna River Basin. Extrapolating the ¹⁰Be data to longer time scales allows for an assessment of geomorphic models of landscape change. I infer that the central Appalachian landscape is dynamic, conforming to the models of neither Davis nor Hack. The ¹⁰Be results imply that on a 10⁴ to 10⁵ year time scale, the topography and relief of the Susquehanna landscape are changing as valleys lower faster than ridges and steep slopes erode more quickly than gentle slopes. The spatial patterns of erosion rates suggest that the basin is not in steady state and may be experiencing a transient response to a drainage network perturbation, perhaps one initiated in the Miocene as suggested by other work. These results lend insight into how relief is maintained in a passive margin setting.