Flume experiment to verify WEPP rill erosion equation performances using loess material

PURPOSE: This study aims to verify the performances of Water Erosion Prediction Project (WEPP) rill erosion equation using loess material by investigating the variations of soil detachment rate with sediment load by rill flow, quantifying the response of soil detachment rate to sediment load, and comprehensively examining WEPP rill erosion equation, so as to provide scientific basis for the application of WEPP model on the loess plateau and to sufficiently understand the response of soil detachment rate to sediment load. MATERIALS AND METHODS: The experiment was conducted in a rill flume with a soil-feeding hopper and was specifically designed to isolate the effect of sediment load on detachment rate. Loess material was collected from a typical hilly region of the Loess Plateau, Ansai, Shaanxi, China. The test soil was quantitatively fed into rill flow by a soil-feeding hopper to produce different sediment loads. Seven unit flow discharges (1.11, 1.56, 2.00, 2.44, 2.89, 3.33, and 3.78 × 10⁻³ m² s⁻¹) were combined with six slopes (10.51, 15.84, 21.26, 26.79, 32.49, and 38.39 %). The sediment transport capacity was measured for each combination. The detachment rate was measured for each combination under seven sediment loads, which were 0, 10, 25, 50, 75, 90, and 100 % of the sediment transport capacity. RESULTS AND DISCUSSION: Soil detachment rate decreased with the increase of sediment load. Levels of sediment load in 0, 10, 25, 50, 75, 90, and 100 % reduced detachment rate in rates of 0, 18.93, 36.36, 56.28, 70.15, 83.42, and 92.19 %, respectively. The response relationship of detachment rate to sediment load by rill flow was described well by a negative linear equation (R ² range from 0.8489 to 0.9982, P < 0.01), and the vertical and horizontal intercepts of the linear equation represented the detachment and transport capacities, respectively (R ² = 0.9955, NSE = 0.9788 for D c ; R ² = 0.9957, NSE = 0.9635 for T c), as expressed by the WEPP rill erosion equation. The WEPP rill erosion equation predicted the soil detachment rate very well (R ² = 0.9667, NSE = 0.9611). CONCLUSIONS: Sediment load transported by rill flow has a negative influence on soil detachment rate in rills. Introducing sediment load as a factor in model equation of detachment is essential for developing an accepted erosion model. The WEPP rill erosion equation could correctly reflect the response relationship of detachment rate to sediment load in this flume experiment and has a good applicability to loess material.