HYDRO-METEOROLOGICAL FLOOD FORECASTING USING TANK MODEL WITH SATELLITE-BASED RAINFALL INPUT FOR KEMAMAN RIVER CATCHMENT

Flooding has always been one of the worst natural disasters in Kemaman Terengganu Malaysia. A reliable, timely and accurate flood forecasting system can improve flood preparedness. In this study, Tank Model was applied for flood forecasting at the upper catchment of the Kemaman River Basin. Hourly rainfall and water level data were used as input to the Tank Model. Autoregressive corrections were carried out to improve the simulated flood level at Rantau Panjang forecasting station. Five sets of data from selected storm events were used to calibrate the Tank Model and then the performance of the model was verified by using two (2) months of hourly rainfall data. A set of tank coefficients that suit the tank configuration selected for the Kemaman River Basin were determined by trial and error calibrations. Flood levels at Rantau Panjang, Kemaman were simulated with actual measured catchment rainfall and the Mean Absolute Error (MAE) was found to be 0.167m and the model efficiency was 92%. The Tank Model was found to be able to simulate and forecast the rising limb of the flood hydrograph as well as the runoff peak for the next 1hr to 4 hr. Once the tank model had been calibrated, satellite-based rainfall estimates were applied to evaluate the simulation and forecasting capability of the Tank Model. The forecast results show that direct application of the QPE from the satellite images needs further enhancement to improve the accuracy. The model developed is able to provide timely (lag time) and reliable warning at an average of 6 hours before the flood peak reaches Rantau Panjang Station, however the accuracy of water level forecast depends on the accuracy of the rainfall forecast