Testing a Grid-Based Soil Erosion Model across Topographically Complex Landscapes

This study was conducted to evaluate the overall performance of the Precision Agricultural-Landscape Modeling System (PALMS) for calculating runoff and soil loss under cropped conditions. The PALMS model uses a lognormal distribution of saturated hydraulic conductivity across the fields to simulate typical soil heterogeneity within soil texture classes. Runoff and soil loss data were collected in three farm fields for a total of 75 runoff events during 2 yr under six cropping scenarios (alfalfa [Medicago sativa L.] and corn [Zed mays L.] no-tilled, corn and soybean [Glycine max (L.) Merr.] moldboard plowed, and alfalfa and corn chisel plowed). For individual storms, calculated runoff and sediment loss from PALMS were compared with corresponding measurements for each farm during the entire cropping season. The coefficient of determination (r(2)) between runoff calculations and measurements was 0.84. The r(2) between soil loss calculations and measurements with the storm-by-storm simulations during both seasons was 0.78. Based on these continuous simulation results, the PALMS calculations appear to have lower relative errors with large events than small events, a desirable result because large events are most important in assessing the environmental consequences of management practices.