The ECMWF System 3 ocean analysis system

TitleThe ECMWF System 3 ocean analysis system
Publication TypeMiscellaneous
Year of Publication2007
AuthorsAlonso-Balmaseda, M, Vidard, A, Anderson, DLT
Secondary TitleTechnical Memorandum
Date PublishedFebruary

A new operational ocean analysis system (system 3 or S3) has been implemented at ECMWF. It consists of two analysis streams: (i) a historical reanalysis from 01/01/1959 which is continuously maintained up to 11 days behind real time and is used to initialize seasonal forecasts, and (ii) an early delivery ocean analysis, produced daily in real time, used to initialize the monthly forecasts. The S3 ocean analysis has several innovative features, including an on-line bias correction algorithm, the assimilation of salinity data on T-surfaces and assimilation of altimeter-derived sea level anomalies and global trends. Two main criteria have been considered in the design of the assimilation algorithm: making optimal use of the observation information at the same time as avoiding spurious climate variability in the resulting ocean reanalysis due to the non-stationary nature of the observing system. The new S3 ocean analysis system outperforms the previous operational system (S2) in the tropics; the biases in both temperature and salinity are reduced, and the representation of the interannual variability is improved. In the extratropics S3 has larger interannual and decadal variability than S2, since the relaxation to climatology has been reduced, but the biases are larger. It is shown that data assimilation has a large impact on the mean state of the first guess, and consistently reduces the bias. In the tropics, the interannual variability is improved, especially the ENSO related variability. In the extratropics, the variability is increased. Data assimilation has a favourable impact on the skill of seasonal forecasts of SST, especially in the western Pacific, where the forecast skill in terms of RMS error is improved at all lead times. In the first 3 months, the forecast skill of the coupled model is improved by more than 20% by using data assimilation in the initialization of the ocean.