Reproduction of coastal sea-level variations around Japan using a nested 2-km resolution model

The Japan Meteorological Agency has begun to operate a monitoring and forecasting system for the Seto Inland Sea, Japan. To update the system, we have been developing an ocean model, MRI.COM-JPN, that extends the model area to the entire coastal seas of Japan. In this presentation, the simulated sea-level variations are verified using AVISO altimetry and tide gauges at 186 sites.

The new model is based on the Meteorological Research Institute Community Ocean Model (MRI.COM) ver.4, and comprises three sub models, a global model, a North Pacific model and a 2-km resolution coastal model, which are coupled using a two-way nesting scheme. We have implemented processes that cause sea-level variations such as tides and depressing and suction of sea level pressure. After spinning up for one year, we used the experimental results of 2009 for analysis. Though this experiment is a so-called free-run experiment without data assimilation, basic sea conditions such as the seasonal development of the sea surface temperature field are well reproduced.

First, the mean sea level distribution in the model region agreed well with AVISO altimetry. The eddy kinetic energy of the sea surface geostrophic currents was also consistent with the AVISO data, although slightly larger. Next, harmonic analysis was carried out and compared to the tide gauge data to investigate the reproducibility of tides, which are the largest factor of the sea level variations. The root mean square error (RMSE) of the M2 tidal amplitude was 4.0 cm on average, which was only 9.2 % of the amplitude itself, 44 cm. The phase error was also as small as 10.2 degree. Third, we verified the daily-averaged sea levels to evaluate the reproducibility of the variations other than tides. The levels well followed the tide gauge data, as the RMSE was 7.3 cm and the correlation coefficient was 0.86 on average. These scores were further improved with the RMSE of 4.4 cm and the correlation coefficient of 0.94 by using a 4DVAR data assimilation technique. In addition to the overall indicators described above, some remarkable cases were also investigated. For example, we succeeded in reproduction of the sea level rising by more than 50 cm after several days from typhoon passage on the San'in Coast in September 2012.