Research

Streamflow Modeling for Flood Events

NWS River Forecast Centers (RFC) have the operational responsibilities of forecasting streamflow and therefore forecasting floods. RFCs use lumped parameter hydrological models for estimation of streamflow. We have focused our attention on two important aspect of flood forecasting: improvements in precipitation forecasts, and application of a sophisticated spatially distributed hydrologic model. We used a Kalman filter formulation to improve the RFC quantitative precipitation forecast (QPF). The Kalman-Improved QPFs are input into our spatially distributed hydrologic model SHEELS. Model parameters were calibrated using radar rainfall data (stage III). Figure 1 shows comparison between the SHEELS streamflow forecasts produced using 6-hr advance Lower Mississippi RFC and Kalman improved QPF for Tropical Storm Allison. With all the parameters held constant, Kalman improved QPF product improves streamflow forecast timing as well as magnitude over the traditional LMRFC forecast. Kalman improved streamflow forecast had better mean absolute bias, RMSE, and r2 than those computed for LMRFC forecast.


Technical Contact: Dr. Ashutosh Limaye (ashutosh.limaye@msfc.nasa.gov)
Responsible Official: Dr. James L. Smoot (James.L.Smoot@nasa.gov)
Page Curator: Diane Samuelson (diane.samuelson@nasa.gov)