Hydrologic response of a small watershed model to generated precipitation

Watershed models such as SWAT (Soil and Water Assessment Tool) simulate wat er quality impacts of land and water resource management alternatives. To s imulate these impacts, long-term daily rainfall data are necessary. In the absence of measured rainfall data, watershed models use weather generators to simulate rainfall events. The objective of this study is to examine seve ral daily precipitation generators in terms of the hydrologic response of S WAT SWAT is generally applied to large river basins but has been validated and applied on the small watershed scale as well. Daily rainfall inputs inc luded a 60-year measured rainfall record from 1939 to 1998 for Riesel, Texa s, in the heart of the Blackland Prairie, and data generated with the preci pitation components of three weather generation programs: WGEN, WXGEN, and USCLIMATE. Measured and generated rainfall were input into SWAT and run for a 53 ha watershed near Riesel, Texas. Rainfall totals, extreme rainfall ev ents, and the resulting hydrologic responses of runoff volume and peak flow s were then examined. For this study scenario, WXGEN was able to more close ly match observed rainfall than WGEN and USCLIMATE, In terms of resulting S WAT hydrologic response, WXGEN rainfall best reproduced runoff volumes simu lated with measured rainfall, and USCLIMATE performed better in reproducing peak runoff rates. These are important results as probabilities of exceedi ng runoff volume or peak flow thresholds are often questions of interest in watershed projects.