ABSTRACT

Rill erosion has serious impacts on arable soils in Chinese Loess Plateau. It is imperative to clarify the key processes of rill erosion to implement effective soil conservation practices. In this study, rill erosion on a 26.8% slope was evaluated using a Terrestrial Laser Scanner (TLS) and CA-Rill model under simulated rainfall. The results showed that (1) the TLS method and the CA-Rill model could effectively describe slope erosion processes, including the splash and sheet erosion, drop pits, intermittent rills, and rill network development stages; (2) compared with the traditional method, the average error of the CA-Rill model for estimating runoff and soil loss was 6.8% and 26.8%, respectively, the CA-Rill model could accurately simulate runoff processes, but poorly simulated sediment processes, especially in splash and sheet erosion stages; and (3) rill erosion evolved into the dominant erosion pattern after 30 min of rainfall, then rill morphological parameters – rill plane density, average rill depth, maximum rill length, and average rill width enlarged, and increased by more than 3.7, 1.8, 2.9, and 4.0 times by the end of rainfall. The methods used to estimate sediment yields and the variation of morphological parameters could provide reference for soil erosion monitoring and modelling.