Predicting Sea Surface Temperature Patterns in the Pacific Ocean

The most dominant mode of oceanic climate variability on an interdecadel scale is the El Niño Southern Oscillation (ENSO). ENSO events are characterized by the anomalous sea surface temperatures (SSTs) in the equatorial Pacific. Large anomalies in this region are known to have high impact on global climate, as for instance, intense rainfall over Peru and droughts and wildfires in Northern Australia during the ENSO warm phase, known as El Niño. Although ENSO is an important climate phenomena, its cause and variability is not yet fully understood, which makes it hard for climate models to capture and predict its occurrence. In this project, we aim to forecast the equatorial SST anomaly patterns in the Pacific based on a data driven approach used together with probabilistic generative models (PGMs). PGMs allow us to make spatial and temporal predictions with uncertainties as well as reveal insights into underlying drivers of the modeled phenomenon. Our final goal is to be able to provide local SST forecasts from a global perspective.