A Forecasting system of RIAM for the Japan/East Sea

Hirose, Naoki*; Sergey, V.*; Watanabe, Tatsuro*; Kawamura, Hideyuki  ; Yamamoto, Masaru*

The Japan Sea holds many features that are similar to world's oceans. We have been developing an ocean data assimilation and prediction system. There are currently two versions of the model aiming at short- and long-term forecasts. The short-term version is driven by tidal forcings and atmospheric predictions, and enables accurate forecasts for up to five days including tidal changes. Short-term outputs will be used for material drift simulations such as an oil spill. The long-term version is sequentially corrected by assimilating the sea surface temperature and height of satellites by the nudging and reduced-order Kalman filter, respectively. The assimilated model estimates realistic nowcast and forecast for a few months. We are also successful to predict a migration of giant jellyfish in 2005. The system will be extended to the East Asian Marginal Seas in the near future.