Overland flow velocity is an important parameter in soil erosion models. However, most existing flow velocity equations are mainly derived from laboratory simulation experiments using artificial vegetation. The purposes of this study were to evaluate the applicability of current overland flow velocity models and to develop and calibrate a field-based overland flow velocity prediction equation for the Loess Plateau. In-situ scouring experiments were carried out in the Zhifanggou small watershed under a fixed 10° slope, with five flow rates (0.06—1 L s⁻ 1 ) and five stem cover levels (0—15%). Our experimental findings indicate that stem cover significantly regulates overland flow hydrodynamic characteristics. As stem cover increased, mean flow velocity decreased notably, along with reduced Reynolds ( Re ) and Froude ( Fr ) numbers. This indicates that vegetation suppresses turbulence and shifts flow from supercritical to subcritical regimes. Compared to bare slopes, average flow velocity decreased by 12% to 48% when stem cover increased from 2.5% to 15%, and at 1 L s⁻ 1 , flow transitioned from turbulent to transitional when stem cover exceeded 5%, with a tendency toward laminar flow at higher cover. Complementing these experimental results, our modeling findings show that among three evaluated velocity models, the modified formula u = a q −b C s+c J d C s+e (a = 0.312, b = 1.506, c = 0.731, d = 1.54, e = 0.3), calibrated with experimental data, performed best in predicting field overland flow velocity ( NSE = 0.95, R 2 = 0.96, RMSE = 0.03), and this modified formula is recommended for overland flow velocity simulation on the Loess Plateau. Collectively, this study quantifies the impact of natural vegetation stem cover on hydraulic processes, providing key field evidence for velocity parameterization in Loess Plateau soil erosion models and scientific support for soil and water conservation measures.
Wei et al. (Tue,) studied this question.