Updated: July 1, 2010
Overview of research at OREDA
Ensemble data assimilation: application of simulation techniques
Models are computer programs to simulate the atmosphere or the ocean. Models can produce spatiotemporally consistent data based on the physical laws. By combining simulated and observed data more accurate and easy-to-use analysed (assimilated) datasets are created. This analysis is used in climate variability research and as initial values of simulations.
We develop ensemble-based methods with collaborative researchers outside JAMSTEC. One of the advantages of ensemble-based methods is that an estimate of the analysis error is obtained as the analysis ensemble spread, which has not been available in traditional analysis techniques.
Fig. 2: Distribution of the wind at 850 hPa at 0 UTC on 17 July 2005 (arrows) and estimated flow-dependent analysis error of the zonal wind (color shading). The cool (warm) color shading represents regions with small (large) errors.