Dynamics and Predictability of Coastal El Nino Events, and Implications for El Niño/Southern Oscillation (ENSO) Diversity

El Nino events warm the ocean along the coast of Peru and Ecuador but the warming is typically far more extensive, covering much of the equatorial Pacific. Basin-wide warming is expected due to the "Bjerknes feedback", a self-reinforcing interaction between the equatorial surface winds and the sea surface temperature (SST) contrast between the "warm pool" of the western Pacific and the "cold tongue" along the equatorial eastern Pacific.

But recent work by the Principal Investigators (PIs) of this award and others has identified another kind of El Nino event in which the warming is in fact largely confined near the coast, with neutral conditions or even a weak La Nina, with cooler than normal SST, over most of the equatorial Pacific. These coastal El Nino events are of scientific interest because of what they tell us about the dynamics of El Nino and the diversity of its possible manifestations.

They are also of practical interest because they cause heavy rainfall and flooding, in part because they peak during the March-to-May rainy season of equatorial South America.

Research conducted here explores the underlying dynamics of coastal El Nino events, their relationship to basin-scale El Nino events, and the extent to which they can be predicted at lead times of several months or longer. The research builds on earlier work in which the PIs examined the sensitivity of coastal SST to subsurface ocean waves (Kelvin waves) produced by wind bursts occurring at various locations along the equatorial Pacific, finding that some locations are particularly effective at generating coastal warming.

Research on predictability is conducted in part using a Machine Learning (ML) model trained on maps of SST, ocean heat content, and surface winds. The model can be used to identify the regions and variables which contribute the most to forecast skill. The predictors identified by the ML model are further explored using simulations conducted with the Community Earth System Model (CESM), which is used to identify the dynamical mechanisms through which the ML-derived predictors influence coastal SSTs.

The work is of societal as well as scientific interest due to the impacts of coastal El Nino events on the adjacent countries of South America. The work on predictability is particularly relevant as it involves the development of an ML model that can be used for operational prediction. More generally, the exploration of El Nino dynamics in the project is relevant to the prediction of basin-wide El Nino events, which affect weather and climate in the US and other parts of the world.

In addition, the PIs are working with an undergraduate student to create an online educational version of the ML model in which users can explore the sensitivity of coastal events to various forecast inputs, and are also developing ML-based predictions for local precipitation impacts. Outreach is also conducted through the Open House of the School of Ocean and Earth Science and Technology (SOEST). In addition, project provides support and training to a graduate student.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.