Response of a distributed watershed erosion model to variations in input data aggregation levels

Distributed watershed models require that mapped variables be characterized by some number of discrete units. The data resolution, which affects the number and size of units, used to spatially represent the watershed variables are important factors that affect the results of distributed hydrologic and geomorphic process response models. The ANSWERS model, run for a typical piedmont watershed in central North Carolina, integrated soil, terrain, hydrography, and landcover information on a cell by cell basis for estimating erosion and deposition patterns within the watershed and for defining model responses at the basin outlet. A raster-based GIS, interfaced to ANSWERS, resampled, formatted, and organized the model input data for iterative calculations of erosion and deposition patterns and magnitudes through model responses for selected data aggregation levels. Spatial aggregation levels evaluated in this research included 30, 60, 120, 180, 240, 300, 420, and 600 m square cells. Semivariograms and fractal dimensions were calculated for each model input variable for the range of aggregation levels evaluated. The magnitude of changes in the model outputs at the selected levels of aggregation followed closely the changes in the spatial dependence of the input variables estimated through semivariograms.