Bayesian Network Model for flood forecasting based on atmospheric ensemble forecasts
The purpose of this study is to propose the Bayesian Network (BN) model to estimate flood peak from Atmospheric Ensemble Forecasts (AEFs). The Weather Research and Forecasting model was used to simulate historic storms using five cumulus parameterization schemes. The BN model was trained to forecast flood peak from AEFs. Mean Absolute Relative Error was calculated as 0.076 for validation data while it was calculated as 0.39 in artificial neural network (ANN) as a widely used model. It seems that BN is less sensitive to small data set, thus it is more suited for forecasting flood peak than ANN.