Author(s): Amalia Wijayanti; Abe Shiori; Nakamura Yosuke
Linked Author(s):
Keywords: Flood forecasts; RRI Model; Weather predictions
Abstract: This study evaluates the performance of the European Centre for Medium-Range Weather Forecasts (ECMWF) Numerical Weather Prediction (NWP) Integrated Forecasting System (IFS) and the Artificial Intelligence Forecasting System (AIFS) in flood forecasting. The objective is to compare the accuracy of these precipitation forecast products in predicting flood magnitude and time of arrival. The Rainfall-Runoff Inundation (RRI) Model is used for hydrological modeling. All model configurations are held constant except for the rainfall input variable. Results indicate that for the initial 24-hour forecast lead, there is no significant difference in root mean square error (RMSE) between NWP and AIFS. As the forecast lead time increases, AIFS shows superior performance with a progressively larger difference in RMSE. The F1-score and Frequency Bias (FB) consistently indicate more favorable results for AIFS compared to NWP across all forecast steps. The F1-score for AIFS ranges from 0.216 to 0.717, while NWP ranges from 0.121 to 0.692.
Year: 2026