Use of mineral magnetic measurements to investigate soil erosion and sediment delivery in a small agricultural catchment in limestone terrain

Understanding sediment delivery at the hillslope scale requires information on the spatial distributions and magnitudes of erosion and deposition. Empirical models such as the RUSLE may be useful for predicting erosion, but are poorly suited for quantifying deposition. Cesium-137 (Cs-137) is useful for quantifying both erosion and deposition, but is costly to inventory over a large area, and directly gauges soil redistribution only for recent decades. The use of rapidly acquired magnetic measurements represents a relatively new potential means of mapping and measuring soil redistribution. In this study, variations in surface magnetism are analyzed to determine patterns of erosion and sedimentation in a small agricultural catchment in northwestern Alabama (USA). Magnetic indicators of erosion are combined with published soil morphology, Cs-137 and short-term suspended sediment data from this former experimental watershed to evaluate long-term sediment delivery. All magnetic parameters measured could be related to soil erosion, although their patterns in space are not identical. Magnetic evidence suggests approximately 30 cm of soil loss on the steepest mid-slope portions of the catchment. Distributing soil loss across space according to magnetic patterns gives average long-term values lower than a prior estimate based on Cs-137 data. Long-term sediment delivery calculated using soil morphology to determine depositional volume ranges up to 45% depending on the magnetic parameter used to index soil loss, and assumptions regarding deposit geometry. These results suggest the need for continued refinement of magnetic techniques for purposes of erosion model validation and general sediment tracing applications. (C) 2001 Elsevier Science B.V. All rights reserved.