Infiltration and deep flow over sloping surfaces: comparison of numerical and experimental results

Infiltration and deep flow on sloping surfaces were studied by combining controlled laboratory experiments with mathematical models. Experimental variables included soil types, rainfall intensities, and surface slopes. Preliminary analyses of the data indicated that increased slope has a positive influence on surface flow, and that this influence increases relatively with decreasing rainfall rate. However, current theories could not adequately explain the observed behavior of deep and surface flows for varying slopes. Three mathematical models of varying complexity were employed to supplement the experimental results. These models were (i) the 2-D Hydrus numerical model, (ii) numerical solution of 1-D saturated–unsaturated flow equation on sloping surfaces, and (iii) a simplified 1-D sharp-front model for sloping surfaces. For the latter two models, a surface flow component based on the kinematic wave approximation for shallow flows was externally coupled to the subsurface flows to route water over the soil surface. For each soil, one experiment at the lowest slope and rainfall rate was utilized for estimation of model parameters that could not be measured independently, while all the other results were used for model corroboration. The Hydrus model indicated that a 1-D analysis would be adequate as the water front moves essentially parallel to the slope. To account for the influence of slope and soil-type on experimental results, an effective saturated conductivity was proposed. Model results were found to be in reasonable agreement with observations of surface flow, deep flow, and water contents in the soil profile with the use of the proposed effective saturated conductivity. Limitations on the applicability of the sharp-front model in this context were discussed.