PREDICTION OF GLOBAL RAINFALL PROBABILITIES USING PHASES OF THE SOUTHERN OSCILLATION INDEX

THE El Nino/Southern Oscillation (ENSO) is a quasi-periodic interannua l variation in global atmospheric and oceanic circulation patterns, kn own to be correlated with variations in the global pattern of rainfall (1-3), Good predictive models for ENSO, if they existed, would allow a ccurate prediction of global rainfall variations, thus leading to bett er management of world agricultural production(4,5), as well as improv ing profits and reducing risks for farmers(6,7). But our current abili ty to predict ENSO variation is limited, Here we describe a probabilis tic rainfall 'forecasting' system that does not require ENSO predictiv e ability, but is instead based on the identification of lag-relations hips between values of the Southern Oscillation Index, which provides a quantitative measure of the phase of the ENSO cycle, and future rain fall, The system provides rainfall probability distributions three to six months in advance for regions worldwide, and is simple enough to b e incorporated into management systems now.