Ponente
Descripción
The El Niño - South Oscilation (ENSO) is an irregular oceanic-atmospheric oscillation that significantly impacts South America and Oceania weather patterns. The Oceanic Niño Index (ONI) categorizes the state of the system into three phases - El Niño (warm), La Niña (cold), and neutral - based on the difference of the sea surface temperature (SST) anomalies in the Niño 3.4 region of the Pacific ocean with respect to a mobile mean.
This study explores an alternative data analysis approach to ENSO phase classification using Topological Data Analysis. Specifically, the authors calculate Euler Characteristic Curves (ECC) from SST fields in a region of the Pacific (10°N-10°S, 160°E-90°W) from January 1950 to December 2022, and investigate whether these topological features can effectively discriminate between the three ENSO phases. Three machine learning models - Logistic Regression (LR), Support Vector Machines (SVM), and Extreme Gradient Boost (XGBoost)- were trained and evaluated on the task of classifying the ECC curves. The results indicate that the SVM model achieves the best performance, as measured by the Area Under the Curve metric.
