A neural-dynamical hybrid coupled model has been developed for giving seasonal predictions of the tropical Pacific sea surface temperatures. A 6-layer dynamical ocean model of the tropical Pacific is driven by the FSU wind stress, then during the forecasting period (Tang and Hsieh, 2002a), the ocean model is coupled to a nonlinear neural network atmospheric model, which estimates the surface wind stress anomalies from the upper ocean heat content anomalies (Tang et. al 2001).
The hybrid coupled model uses the NCEP sea level height anomaly (SLHA) data to initialize the forecasts (Tang and Hsieh, 2002b). The assimilation of the NCEP SLHA was found to yield as great an improvement in the forecast correlation skills as the assimilation of heat content anomalies (Tang and Hsieh 2002b). Fig. 1 shows the correlation skills of the predicted SST anomalies (SSTA) in the NINO3 region in the equatorial eastern Pacific during 1990-1999 using our model with SLHA assimilation. The predictions were made at three months intervals (starting on 1 January, 1 April, 1 July and 1 October) and continued until a lead time of 15 months.