Experimental study on soil erosion by concentrated waterflow affected by thawed soil depth

Abstract The rill erosion of a partially thawed soil slope is of great importance for understanding erosion dynamics and supplying datasets to estimate rill erosion model parameters. This study aimed to investigate the effects of thawed soil depth on rill erosion and to quantify the functional relationship between rill length and erosion. A series of laboratory experiments were conducted to measure sediment delivery under four slope gradients of 5°, 10°, 15°, and 20°, three flow rates of 1, 2, and 4 L/min, and three thawed depths of 1, 2, and 5 cm at nine slope lengths of 0.5, 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, and 8.0 m. Each individual experimental run harvested four runoff samples to measure sediment concentration. A total of 3,888 data points were produced. The experimental data showed that sediment concentrations along the rill can be well fitted with piecewise functions of linear and exponential sections. The sediment concentrations on the slopes of shallow-thawed depths of 1 and 2 cm initially increased linearly before exponentially increasing with rill length. Under condition of 5 cm thawed depth, the sediments exponentially increased with slope length. The lack of sediment supply over partially thawed soil slope determined the linear increase in sediment at the initial slope part. When water flow traveled a certain slope length to pick up sufficient sediments, the sediment concentration started to increase exponentially. The sediment concentration increased with slope gradient and flow rate. A longer slope was needed to achieve a higher maximum sediment concentration on a shallower thawed depth, a gentler soil slope, and a higher flow rate. The experimental datasets and their functional relationship can supply the basis for estimating the model parameters of rill erosion, such as soil detachment rate and soil erodibility, on partially thawed frozen-soil slope.