Weather radar rainfall data-based flood forecasting model

The system error always exists in the real-time flood forecasting system, which includes the model error and the error of observation. In order to diminish the system error, this study tried to introduce the weather radar rainfall technique and BP neural network technique into River Basin flood forecasting, and a flood forecasting model was developed based on the distributed hydrological model. The real-time forecasting model was applied to the Shiguanhe River Basin, and the result showed that the model could well realize the coupling of radar remote sensing data with the hydrological model. The model developed provided a basis for research on the application of weather radar rainfall technique to river basin flood forecasting.