WHOI PO

Earth’s climate is currently experiencing El Niño conditions. Based on independent physical metrics and current forecasts, the event is expected to develop into an extreme event—a “super” El Niño—and could quite possibly become the strongest El Niño in the observational record. How far in advance could this event have been predicted? And how far into the future can we predict its subsequent evolution? In this talk, I will review the current climate state of the tropics and introduce DeepENSO—a deep learning model designed to predict three-dimensional (3D) upper-ocean temperature fields while incorporating ocean physics. DeepENSO produces skillful ENSO forecasts with lead times of up to 26 months, successfully overcoming the “spring predictability barrier” twice across all initialization seasons. Finally, I will discuss whether the apparent increase in ENSO activity over the past four decades might be linked to global warming.