A Non-Linear Dynamical-Statistical Model for Reconstruction of the Air-Sea Element Fields in the Tropical Pacific Ocean

Aiming at tackling the difficulty in establishing a sea surface temperature (SST) dynamical model, this study develops a non-linear dynamical-statistical model of SST fields and their correlative factors based on Genetic Algorithms (GA) and the dynamical system reconstruction idea, which greatly improves the El Niño-Southern Oscillation (ENSO) forecast model. Using Hadley SST data, sea surface wind (SSW) and sea level pressure (SLP) data from the National Centers for Environmental Prediction-National Center for Environmental Research (NCEP-NCAR), with empirical orthogonal function (EOF) time-space for reconstruction, we carry out numerical integral forecasting experiments for SST, SSW, and SLP fields. By statistical analysis of the forecasting experiments, we find that forecasts for less than 25 months perform better than longer term forecasts. Based on the model, we forecast SST, SSW, and SLP fields in September, October, and November 2014 and predict a weak La Niña event. This study explores a novel method for the complex atmosphere-ocean system.