Assimilation of decomposed in situ directional wave spectra into a numerical wave model of typhoon waves

The purpose of this study was to enhance the accuracy of numerical wave forecasts through data assimilation during typhoon periods. A sequential data assimilation scheme was modified to enable its use with partitions of directional wave spectra. The performance of the system was investigated with respect to operational applications, specifically for typhoon waves. Two typhoons that occurred in 2006 around Taiwan (Kaemi and Shanshan) were used for this case study. The proposed data assimilation method increased the forecast accuracy in terms of wave parameters, such as wave height and period. After assimilation, the shapes of directional spectra were much closer to those reported from independent observations.