Predictability of ENSO

New York University, New York NY

Investigators

Abstract

The objectives of the proposed work are to (i) demonstrate that the techniques to determine and to understand the relaibility of dynamical predictions of ENSO, work in a representative cross section of current state of the art coupled ocean-atmosphere model forecast system; (ii) determine the physical basis for the variation in the skill amongst forecasts; (iii) investigate the usefulness of a collection of forecasts from different models in producing both a better forecast and an indication of forecast robustness; and (iv) develop a general theory of linear prediction in order to extend and refine the methodology developed to date. The PI proposes an investigation of the predictability of ENSO in a range of numerical models. The work is important because it will lead to better understanding of ENSO predictability and physical processes involved in tropical Pacific variability. Improvement in understanding is likely to lead to more accurate ENSO forecasts.