New estimation of overland flow velocity based on Manning's equation and equivalent roughness

Abstract

The overland flow velocity is a fundamental hydraulic variable that directly impacts soil separation, sediment transport, and material deposition during soil erosion. Accurately predicting this velocity is challenging due to the significant spatial variability of overland flow hydraulic characteristics resulting from vegetation covers. To quantify the influence of vegetation stem coverage on mean flow velocity, fixed-bed scouring tests were conducted with eleven stem coverages (0-20.42%), four slope gradients (3.49%-20.78%), and five unit discharges (0.278-2.222 L·m −1 ·s −1). The results indicated a trend of mean flow velocity initially increasing and then decreasing as stem coverage rises, with a critical threshold value identified at 2.72%. Compared to studies on bare slopes, an increase in mean velocity ranging from 8.22% to 49.91% was observed on slopes with a stem coverage of 2.72%. The accuracy of velocity prediction can be improved by utilizing both discharge and slope gradient, as opposed to methods that rely solely on unit discharge for prediction. The adjusted correction coefficient (adj. R²) increased from 0.205-0.610 to 0.754-0.968. Finally, a dimensionally harmonious flow velocity prediction model based on the adjusted Manning's equation was established using the equivalent roughness, where flow resistance was found to be a combination of grain resistance and stem resistance. The developed model has been proven accurate in predicting experimental results (adj. R² = 0.831, Nash-Sutcliffe efficiency coefficient = 0.830). The results obtained here enhance our understanding and prediction of soil erosion, providing further insights into the phenomenon.