Abstract
Recently, due to global warming, there are growing concerns about possible tremendous disaster caused by furious typhoons more than ever. It is therefore very important to develop an accurate numerical model for storm surge. Recently, some circulation models, e.g. Princeton Ocean Model (POM), Regional Ocean Modeling System (ROMS), etc. are applied to the storm surge simulations for enabling more accurate estimations. Among these existing models, Finite Volume Coastal Ocean Model (FVCOM) developed by Chen et al. (2003) has an advantage in that it employs unstructured grid system so as to compute tidal currents in complicated inner bays with high accuracy. Thus, the FVCOM seems to be superior to other models in storm surge simulations. In this study, first, the FVCOM was modified to deal with the spatiotemporal changes of pressure fields in the momentum equations. Second, in order to investigate the influence of computational conditions (e.g. domain size, grid size, etc.) on the computational results, the storm surge simulations were conducted for the past severe typhoons which actually caused the storm surges in Ariake Sea and Yatsushiro Sea. Then, computational results were compared with observed data as well as other results computed by the POM to clarify the advantages of the FVCOM in storm surge simulations.Unstructured Grids can re-create complex topography better than Structured Grids, hence FVCOM is more applicable to compute storm surges in inner bay.
