Predicting erosion patterns using a spatially distributed erosion model with spatially variable and uniform parameters

Distributed models are generally viewed as the best type of model for evaluating the erosional response of a drainage basin to changes in land use, climate, and other controlling factors. In many instances, however, information about the spatial variability of the model parameters is not available so that, even though the model is spatially explicit, parameters are treated as though they were uniform across the basin. The objective of this study was to compare the results of model runs with spatially variable parameters with those of model runs with uniform parameters. The model used is Cascade 6, a grid-based erosion model which requires data on elevation, saturated conductivity, surface roughness, and soil cohesion for each grid cell. The model was applied to the 45-ha Catsop basin located in the province of South-Limburg, The Netherlands. This loess-covered basin has a gently to moderately sloping topography, and land use is predominantly agricultural. The model is calibrated for peak discharge and sediment rating curve, and calibration is carried out separately with spatially variable and uniform runoff and erosion parameters. Model results indicate that erosion rates for individual grid cells modelled with variable parameters can be predicted from the erosion rate modelled with uniform parameters for grid cells with a drainage area from I to 1000 grid cells. Thus, the spatial pattern of erosion in the basin can be modelled over 85% of the basin area even in the absence of information about the spatial variability of the runoff and erosion parameters.