THE UNDERSTANDING OF ENSO CYCLE MECHANISM AND ENSO POTENTIAL PREDICTION ABILITY

<正> In recent years,the dynamic coupled models of ocean-atmosphere and statistical models havebeen used in routine operation for issuing long-lead forecasts.The dynamic coupled models consistof models with varying degrees of complexity,ranging from simplified coupled models of theshallow water to coupled general circulation models.During the period of 1980—1992,somemodels performed considerably better than the persistence forecast on predicting typical indices ofENSO for lead time of 6 to 12 months.It seems that ENSO is predictable at least one year inadvance.However.nearly all the models have lost their skill of forecasting sea surface temperature(SST)changes in the eastern equatorial Pacific since 1992.It is a challenge not only to the dynamicmodels but also to the understanding of the ENSO cycle mechanism.This paper examines multipletime-space scales of the ocean-atmosphere interactions and potential prediction ability of ENSOevent by using data analysis and model study.