SENSITIVITY OF A SHENANDOAH WATERSHED MODEL TO DISTRIBUTED RAINFALL DATA

Abstract

Precipitation is a primary input for hydrologic modeling. In this study, the use of spatially varying gridded precipitation data obtained from the North American Land Data Assimilation System (NLDAS) is investigated for possible improvements to stream flow predictions within the Shenandoah River watershed over the use of gage based precipitation data. The sensitivity of hydrologic responses to gridded and gage precipitation data is analyzed using the Hydrologic Engineering Center's Hydrologic Modeling System (HEC-HMS). Precipitation is the only input which varies between models. Model performance of each precipitation input is assessed by comparing predicted and measured stream flow during the 1995 to 1996 water year and calculating a number of goodness of fit measures. Results indicate that the use of spatially varying gridded precipitation data can improve stream flow predictions within the Shenandoah River watershed. Future research directions include the calibration and validation of the HEC-HMS model using the gridded precipitation data.