Performance of the SWEEP Model Affected by Estimates of Threshold Friction Velocity

Abstract

Abstract. The Wind Erosion Prediction System (WEPS) is a process-based model and needs to be verified under a broad range of climates, soils, and land management practices. Occasional poor performance of the WEPS erosion submodel (Single-event Wind Erosion Evaluation Program, or SWEEP) in simulating small amounts of erosion of loessial soils, which contribute to poor air quality in the Columbia Plateau region of the U.S. Pacific Northwest, has been partly attributed to overestimation of threshold friction velocity (u*t). Few studies have been conducted to improve the poor performance of the SWEEP in this region. Therefore, the objective of this study was to test the performance of the SWEEP in simulating the occurrence of erosion when the simulated u*t (SWEEP u*t) was replaced with measured values of u*t or with estimates of u*t derived from algorithms used in the Lu and Shao model, Shao model, Texas Tech Erosion Analysis Model (TEAM), and Wind Erosion on European Light Soils (WEELS) model. The occurrence of erosion from agricultural lands was better estimated by the SWEEP when SWEEP u*t was replaced with the u*t algorithm used in the Lu and Shao model. However, soil loss was better estimated by the SWEEP when using SWEEP u*t rather than the measured u*t or other u*t algorithms. Since prediction of both the occurrence and magnitude of wind erosion is important to overall model performance, there is a need to further improve the SWEEP in the Columbia Plateau and other regions where small amounts of erosion can impact soil resources and environmental quality.