Global Wave Data Assimilation Model using Optimal Interpolation Method

To construct a reanalyzed global ocean wave data set with improved accuracy, which is important for the better understanding and simulation of various near-surface ocean dynamics, a data assimilation method has been embedded to the global spectral wave model based on WW3. The major factors controlling the wave simulation accuracy are the wind condition and the parameterization on the wave energy development, dissipation and nonlinear processes between wave components. However, the atmospheric prediction accuracy is still not sufficient, and the parameterization cannot be generalized due to the local geographic conditions.

In detail, the data assimilation using the optimal interpolation method has been applied, verification through the comparison with satellite altimeters and buoy observations has been made with examination of the data assimilation effects. The significant wave heights computed by the integration of wave energy spectra are showed to be quite similar with observed results. However, the wave periods and directions related to the shape of wave energy spectra are not sufficiently comparable. Generally there have been difficulties in predicting the propagation of long period waves such as swells.

The wave energy spectra on wave number and direction domains was multiplied by optimal interpolation method with the ratio of observed significant wave heights on first guessed simulated results. The energy spectra was thereafter shifted by the difference between simulated and observed peak wave periods and directions. From then examination of the reanalysis simulation during 1 year, it could be seen that the accuracy of the model with the data assimilation shows better results than that without data assimilation.