The IRI seasonal climate prediction system and the 1997/98 El Nino event

The International Research Institute for Climate Prediction (IRI) was formed in late 1996 with the aim of fostering the improvement, production, and use of global forecasts of seasonal to interannual climate variability for the explicit benefit of society. The development of the 1997/98 El Nino provided an ideal impetus to the IRI Experimental Forecast Division (IRI EFD) to generate seasonal climate forecasts on an operational basis. In the production of these forecasts an extensive suite of forecasting tools has been developed, and these are described in this paper. An argument is made for the need for a multimodel ensemble approach and for extensive validation of each model's ability to simulate interannual climate variability accurately. The need for global sea surface temperature forecasts is demonstrated. Forecasts of precipitation and air temperature are presented in the form of 'net assessments,' following the format adopted by the regional consensus forums. During the 1997/98 El Nino, the skill of the net assessments was greater than chance, except over Europe, and in most cases was an improvement over a forecast of persistence of the latest month's climate anomaly.