干旱区生态环境知识资源中心

Adequately simulating the occurrence and magnitude of erosion using the Wind Erosion Prediction System is critical for resource conservation and improving environmental quality in the United States and other regions of the world. Past observations indicate an insensitivity of the Wind Erosion Prediction System erosion submodel (Single-event Wind Erosion Evaluation Program (SWEEP)) to simulate the occurrence of wind erosion from loess soils during intensifying winds in the Pacific Northwest United States. This insensitivity is attributed to either underestimation of friction velocity (u(*)) or overestimation of threshold friction velocity (u(*)t), but no definitive work has been conducted to validate u(*) and u(*t) in the SWEEP. The objective of this study was to compare measured u(*) and u(*t) with those simulated by the SWEEP. The SWEEP begins to simulate erosion when u(*) exceeds u(*t), thus we compared measured and simulated u(*) and u(*t) of agricultural fields during wind erosion events in eastern Washington. No evidence indicated that the SWEEP consistently overestimated or underestimated u(*). The SWEEP was very sensitive to the stem area index, and estimation of u(*) was better when the SWEEP was used at field sites characterized by a lower rather than a higher stem area index. Our results, however, indicate that the SWEEP overestimated u(*t) during all observed erosion events. Therefore, simulating the occurrence of wind erosion of loess soils in the Pacific Northwest and other arid and semiarid regions of the world may be improved through better estimation of u(*t) in the SWEEP.