SPATIAL SIMPLIFICATION OF INPUT DATA FOR HYDROLOGIC MODELS: ITS EFFECT ON MAP ACCURACY AND MODEL RESULTS

Statistical generalization can be used on map data to reduce the complexity of the data before analysis. The generalization scheme tested here calculates the dominant cover class (mode) of a theme within specified areas. In this work, the specified areas are agricultural fields, and the themes are soils, slope, and aspect. The purpose of this study was to assess the impact of using a vector-based mode amalgamation scheme on the map polygon count, map accuracy, and hydrologic model results at field and watershed scales. The Groundwater Loading Effects of Agricultural Management Systems (GLEAMS) model was initially run on fields using inputs based on the original coverages describing soils, slope, aspect, and field boundaries. Polygons representing one or more themes were amalgamated and the simulations were then repeated. The map accuracy of the amalgamated themes were calculated. The original model results were compared to those based on the amalgamated data. The amalgamation technique decreased the number of polygons, but introduced substantial map errors. The soils amalgamation introduced error into most measures, but more so at the field scale than the watershed scale Amalgamations of the slope and aspect themes had less effect on the model outputs investigated. We concluded that the aggregation method and extent should depend on the model being used and on the modeling objectives