Comparison of soil erosion and deposition rates using radiocesium, RUSLE, and buried soils in dolines in East Tennessee

Abstract

Three dolines (sinkholes), each representing different land uses (crop, grass, and forest) in a karst area in East Tennesse, were selected to determine soil erosional and depositional rates. Three methods were used to estimate the rates: fallout radiocesium (137Cs) redistribution, buried surface soil horizons (Ab horizon), and the revised universal soil loss equation (RUSLE). When 137Cs redistribution was examined, the average soil erosion rates were calculated to be 27 t ha–1 yr–1 at the cropland, 3 t ha–1 yr–1 at the grassland, and 2 t ha–1 yr–1 at the forest. By comparison, cropland erosion rate of 2.6 t ha–1 yr–1, a grassland rate of 0.6 t ha–1 yr–1, and a forest rate of 0.2 t ha–1 yr–1 were estimated by RUSLE. The 137Cs method expressed higher rates than RUSLE because RUSLE tends to overestimate low erosion rates and does not account for deposition. The buried surface horizons method resulted in deposition rates that were 8 t ha–1 yr–1 (during 480 yr) at the cropland, 12 t ha–1 yr–1 (during 980 yr) at the grassland, and 4 t ha–1 yr–1 (during 101 yr) at the forest site. By examining 137Cs redistribution, soil deposition rates were found to be 23 t ha–1 yr–1 at the cropland, 20 t ha–1 yr–1 at the grassland, and 16 t ha–1 yr–1 at the forest site. The variability in deposition rates was accounted for by temporal differences;137Cs expressed deposition during the last 38 yr, whereas Ab horizons represented deposition during hundreds of years. In most cases, land use affected both erosion and deposition rates – the highest rates of soil redistribution usually representing the cropland and the lowest, the forest. When this was not true, differences in the rates were attributed to differences in the size, shape, and closure of the dolines.