Wave information is of primary importance for design of maritime structures and for the project implementation. NOWPHAS, a wave observation network along the Japanese coast, has been widely used for development, utilization and disaster prevention in coastal area. However, it is also necessary to estimate wave information using a numerical wave model if there is no observational data near the study site. It has been pointed out that the existing third generation wave model has a problem with regard to the reproduction of ocean swells. In this study, we validate the third generation wave model and discuss the applicability of it to coastal wave phenomena “YoriMawari-wave (YM-wave)” by conducting two hindcast experiments. In the Experiment 1, we conducted 3-year wave hindcast for the Japanese coast, and investigated the accuracy of waves statistics estimated with different source term functions. The Experiment 2 focused on an YM-wave event in the Toyama bay and investigated the impact of directional resolution on model results. These experiments demonstrated that the different wave models did not provide significant difference on wave statistics, and neither model can reproduce the wave height during the YM-wave events. On the other hand, the estimated spectral shapes are largely different depending on the directional resolution. The incident wave spectrum at the time of the YM-wave was extremely narrow which can be quantified by Smax parameter as 560, and therefore, it is suggested that the amplitude modulation due to the coherent interference of the quasi-monochromatic wave train could be related to the YM-wave generations.

Key Words: Third generation wave model, Swell dominant wave field standing wave,Wave hindcast, YoriMawari-waves in the Toyama bay