EUROSEM: An Evaluation with Single Event Data from the C5 Watershed, Oklahoma, USA

Abstract

For the GTCE model comparison EUROSEM was evaluated against data for the C5 watershed, Oklahoma, USA. Model calibration was carried out using data for events from the training period with similar rainfall patterns and soil conditions to that of the test events. Calibration was conducted by modifying the saturated hydraulic conductivity and initial moisture content of the soil until broad agreement was reached between the runoff simulated by the model and the observed data Once this had been achieved the model was independently calibrated to simulate soil loss by modifying the parameters describing the cohesion and erodibility of the soil, and rill density. During the calibration phase parameter values were maintained within realistic limits for the watershed concerned. Once calibrated, the model was applied to four test events. EUROSEM simulated the total runoff and soil loss from three of these events quite well, but failed to reproduce the hydrographs and sedigraphs. Observed hydrographs and sedigraphs must be supplied at the calibration phase if catchment response is to be successfully simulated.