[A-TT35] Machine Learning Techniques in Weather, Climate, Ocean, Hydrology and Disease Predictions

[ATT35-P04] Skillful prediction of Atlantic Niño index using machine learning models

*Venkata Ratnam Jayanthi¹, Ingo Richter¹ (1.Application Laboratory, JAMSTEC)

Keywords:ATL3, ANN

Several machine learning models were evaluated for their skill in predicting the June-Aug averaged Atlantic Niño index (ATL3) at long lead time of 1-6 months. The input attributes to the models were derived based on a lag correlation analysis between the observed ATL3 index and sea surface temperature, sea surface height, vertically averaged (0-100m) potential temperature. Of all the models evaluated, the model based on artificial neural networks (ANN) had better skill in predicting the ATL3 index at all the lead times of 1-6 months, with a correlation coefficient of about 0.6. It is found that the machine learning model based on ANN outperforms the state-of the art dynamical models in predicting the ATL3 index.