Modelling sheet erosion on steep slopes of clay loess soil using a rainfall simulator

The process of sheet erosion has a dramatic influence on the development of soil erosion. To successfully model sheet erosion it is necessary to establish a suitable erosion prediction model for regions with clay loess soil. This study strives to achieve the following: 1) understanding and quantifying sheet erosion patterns caused by different variations of the rainfall intensity, slope steepness, and flow velocity, and 2) evaluate the influence of hydrodynamic parameters on sheet erosion for clay loess soil. A series of simulation experiments of rainfall on clay loess soil were carried out using a range of values of rainfall intensity (54, 78, 102, 126 and 150 mm h−1) and gradient (8.7, 17.6, 26.8, 36.4 and 46.6%). The simulation outcomes demonstrated that the rate of sheet erosion followed a power function of rainfall intensity and also gradient with a coefficient of determination R2 of 0.98 and a Nash–Sutcliffe efficiency (NSE) of 0.96. The sensitivity of the sheet erosion rate to rainfall intensity was higher than its sensitivity to slope steepness. Furthermore, the sheet erosion rate could be predicted using its power-law dependence on flow velocity with R2 of 0.88 and NSE of 0.82. The sheet erosion rate was optimally and linearly fitted with the shear stress with R2 of 0.95 and NSE of 0.95, it was optimally fitted through power functions with either the stream power (with metrics of R2 of 0.94 and NSE of 0.93) or the unit stream power (with R2 of 0.74 and NSE of 0.68). Thus stream power represented a reliable predictor of the sheet erosion rate, and the shear stress was the better predictor. By contrast, the unit stream power was found to be a poor predictor of the sheet erosion rate. The findings of this study can help to accurately predict sheet erosion for clay loess soil on steep slopes.