Rates of Erosion Determined From 10Be Analysis of Alluvial Sediments, Great Smoky Mountains, Tennessee and North Carolina

We measured ¹⁰Be in quartz extracted from sediment samples (n=22) collected from a network of 22 rivers and streams draining the Great Smoky Mountains and calculated model erosion rates that average 31+/-5 m My^-1. Network analysis and mass balance calculations verify two main assumptions on which model erosion rate calculations depend: no significant storage within the sampled basin (also indicated by field observations) and good mixing of sediments from different sources within the sampled basins. Mass removal rates from each sampled basin were calculated from ¹⁰Be model erosion rates and basin size. For the two basins that we studied in detail (i.e. most of the main tributaries and the main stem were sampled; Oconaluftee River and Raven Fork), total area and mass removal rates of the tributaries were compared with area and calculated mass removal rates of the main stem. Tributaries supplied a fraction of the total mass removed from the basin equal to the area they cover within the drainage system. This correspondence suggests that erosion rates are spatially uniform throughout the mountain range. We analyzed different sand and gravel fractions (250-850 um, 850-2000 um, and >2000 um) to test whether different grain sizes of sediment have different cosmogenic nuclide activities. Grain size tests of two samples in the Oconauftee River (GSCO-1, GSCO-7) yielded similar nuclide activities for the three different fractions, suggesting that different grain sizes have similar cosmic-ray dosing histories. However, the >2000 um fraction in one of the samples (GSCO-1) had a nuclide activity 40% lower than the <2000 um fractions, implying less cosmic ray dosing for the larger than for the smaller grains. The low ¹⁰Be activity of this sample might result from a single mass wasting event that rapidly delivered large grain size, previously shielded sediment to the channel (c.f., Brown et al., 1995, EPSL, 129: 193-202). Rates of erosion do not correlate with drainage basin area and variance diminishes as drainage basin area increases; the 3 largest basins have ¹⁰Be model erosion rates (30.2+/-2.2 m My^-1) similar to the mean erosion rate (27.5+/-4.6 m My^-1) of the 11 smaller headwater basins. We found a very weak correlation between rates of erosion and maximum drainage basin relief (R²= 0.13) and are currently testing the relationship between nuclide activity and other basin parameters including drainage density and slope.