Detachment rates of purple soil rills affected by subsurface water flow depth

Abstract

Subsurface water flow, considered as a specific hydrological condition, can couples with overland flow, which likely lead to notable effects on soil detachment process in rills. Clarifying the change in soil detachment rate under different subsurface flow conditions is beneficial to understand the mechanisms of soil erosion and improve process-based erosion model. In this study, the maximum sediment delivery rates at sediment transport capacity were used to determine the conditions when soil detachment rates reached zero. Then, a numerical method (NM) and modified numerical method (MNM) were utilized to estimate the detachment rates through sediment delivery amounts along a rill affected by subsurface water flow depths (5 cm, 10 cm, and 15 cm). The experiments were performed under combined conditions of four slope gradients (5°, 10°, 15°, and 20°) and three flow rates (2, 4, and 8 L min−1). The results showed that the detachment rates could be fitted by a linear function of the sediment concentration computed by the NM and MNM under different subsurface water flow depths. When the depths of subsurface water were 5 cm, 10 cm and 15 cm below the soil surface, soil detachment rates estimated by the NM were approximately 1.23, 1.31 and 1.12 times those obtained by the MNM, respectively. The MNM yielded more reasonable detachment rates than the NM. This method is applicable to estimate detachment rates affected by different subsurface water flow depths. Soil detachment rates of subsurface water flow 5 cm below the surface were similar to those of saturated soil with a 20° slope gradient and a flow rate of 8 L min−1. Soil detachment rate for subsurface water flow at 15 cm below the surface were similar to those of unsaturated soil under low hydrodynamic conditions. The results of this study are helpful to understand the mechanisms of soil erosion affected by subsurface flow and provide a theoretical basis for improving the soil erosion model.