A storm surge hindcast for the Northwest Pacific Ocean from 1979 to 2023

A storm surge driven by meteorological phenomena such as low atmospheric pressure and strong winds is an abnormal rise in sea level that can result in disasters, including coastal flooding, damage to coastal structures, ecosystem destruction, and beach deformation. This phenomenon arises from complex interactions between the atmosphere, ocean, and topography, making it essential to reconstruct and analyze past storm surge events to assess their impacts and identify vulnerable areas. Therefore, this study produced storm surge hindcast for the Northwest Pacific region, covering 115-150°E, 20-52°N, from 1979 to 2023. The dataset was generated using pressure and wind fields from ERA5 as external forcing conditions for the Delft3D-FM model. It features a spatial resolution of approximately 40 km in the open sea and 800 m in coastal areas, with a temporal resolution of 1 hour. The topographic data for modeling comprised GEBCO2023 for the open sea and the latest nautical chart data for Korea’s coastal regions. The accuracy of the data was evaluated for short-term events (individual typhoons) and long-term trends (multi-year statistical values), using observational data from 45 tide gauge stations along the Korean coast as the evaluation standard. The root mean squared error (RMSE), correlation coefficient (R), and variance ratio (VR) were employed as evaluation metrics. The analysis of data accuracy for short-term events revealed that it varied depending on topographical features, such as water depth, and the specific characteristics of the typhoon. Long-term trends were evaluated for the annual average, as well as the 99-percentile and maximum values, both representing extreme events. The analysis confirmed that storm surges should be analyzed using the concept of extreme values rather than average values. It was also identified that both short-term events and long-term trends tended to be underestimated by the model compared to the observations. This is likely due to the inability of ERA5, used as the external forcing condition, to accurately simulate extreme weather conditions such as typhoons. So, in the future, it is considered necessary to conduct storm surge hindcast simulations by applying high-resolution meteorological reanalysis data (e.g., JRA-55) for pressure and wind fields as external forcing conditions in the same numerical modeling environment.