A review of ENSO prediction studies

摘要: A hierarchy of ENSO (El Nino/Southern Oscillation) prediction schemes has been developed which includes statistical and physical models. The models are, in general, based on advanced techniques can be classified into use either low-frequency variations the atmosphere (sea level pressure or surface wind) upper ocean heat content as predictors. consist coupled ocean-atmosphere varying degrees complexity, ranging from simplified ‘shallow water’-type to general circulation All models, physical, perform considerably better than persistence forecast predicting typical indices lead times 6 12 months. most successful schemes, fully show significant abilities at exceeding one year period. We therefore conclude that is predictable least advance. However, all this applies gross such Southern Oscillation Index. Despite demonstrated predictability, little known about predictability specific features associated with (e.g. Indian Monsoon rainfall, African drought, even off-equatorial sea temperature). Nor relative importance for different regional anomalies processes yet established. seasonal dependence well established, but responsible it are not understood.