COMPARATIVE ACCURACY ASSESSMENT OF DELFT3D STORM SURGE SIMULATIONS DURING TROPICAL CYCLONE ERROL ALONG THE COAST OF ROTE ISLAND USING FNL AND ERA5 DATA

Abstract

Accurate representation of storm surge is crucial for comprehending sea level responses to tropical cyclone events, especially in coastal regions susceptible to extreme weather conditions. This study seeks to compare storm surge simulations using atmospheric data from the fifth generation of the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis and the Final Operational Global Analysis (FNL) as atmospheric forcing in the Delft3D-FLOW model during Tropical Cyclone Errol around Rote Island from 9 to 18 April 2025. The data utilized include tropical cyclone track data, atmospheric data, bathymetry data from the Indonesian National Bathymetry dataset, tidal data from the TPXO global ocean tide model, and sea level observations from the Batutua Tide Gauge Station. The simulation results were assessed using the Root Mean Square Error (RMSE) and correlation coefficient against the observed sea level data. The findings indicate that both atmospheric datasets effectively represented the observed sea level fluctuation patterns. The simulation employing the fifth generation of the ECMWF reanalysis yielded a RMSE of 0.4324 and a correlation coefficient of 0.8676, whereas the FNL produced values of 0.4322 and 0.8677, respectively. Spatial analysis reveals that the storm surge distribution corresponded with the movement of Tropical Cyclone Errol, transitioning from wind setup conditions at the onset of the event to negative storm surge as the cyclone moved away from the study area. The analogous spatial patterns suggest that both atmospheric datasets offer comparable representations of storm surge characteristics during the event.