Hydrodynamic Flood Forecasting using Kalman Filtering: - I. Model Development -

Kalman Filtering technique was used for real-time update in the hydraulic flood routing model. Optimal updating estimates were obtained by combining the results of hydraulic flood routing model with real time data using Kalman Filter gain factor. The study area was the main Han River which is from Paldang Dam to the river mouth and the 6 flood/non-flood events were applied for Kalman Filtering technique. The results showed that RMS error has been improved 52.76% at Jamsu bridge, 55.03% at Hangang bridge on average. The rivetted of the main Han River was simplified to reduce running time used for using Kalman Filtering technique and the improvement rate of simulation time was represented as 76.6% approximately. It could be concluded that the combination of Kalman Filtering technique and hydraulic flood routing model could improve the accuracy of real-time flood forecasting.