Secondary Forecast Models—The ENSO Example

In this note, a procedure to construct predictors is suggested that could take advantage of the increasing predictive skill of ENSO models. These so-called secondary forecast models (SFM) attempt to produce a forecast for a particular object application based on independent forecasts of ENSO. Although their predictive skill may be modest, these models can provide an alternate and objective forecast to the entirely subjective schemes more widely used. An example of the application of SFM to the forecast of the seasonal precipitation of a typical station of the Mediterranean Californias shows promising results.