Research Theme · 2026-02-28

ENSO Prediction, Predictability, and Climate Impacts

By Research Theme Editorial

This theme brings together ENSO predictability, the spring predictability barrier, climate-network impacts of El Nino, and physics-guided machine learning for long-lead climate prediction.

ENSO is one of the most influential modes of interannual climate variability. It affects the tropical Pacific and, through teleconnections, modulates rainfall, extremes, and climate risks across the globe.

The group approaches ENSO from two complementary directions: improving long-lead prediction across the spring predictability barrier, and identifying how El Nino impacts are organized across the global climate network.

Representative Work

Physics-guided deep echo state networks

Enhancing the predictability limits of ENSO with physics-guided deep echo state networks combines interpretable climate modes from the extended recharge oscillator framework with a deep echo state network, showing skillful Niño3.4 prediction at 16-20 month lead times and diagnosing the role of warm water volume and cross-basin mode coupling.

El Nino magnitude forecasting before spring

Complexity-based approach for El Niño magnitude forecasting before the spring predictability barrier uses complexity indicators of the Niño3.4 region to forecast El Nino magnitude before the spring predictability barrier.

Localized impacts of El Nino

Network analysis reveals strongly localized impacts of El Niño uses climate-network analysis to show that El Nino impacts are not spatially uniform, but exhibit strongly localized structures.

Why It Matters

This theme links prediction and mechanism: complexity indicators and machine learning support earlier forecasts, while climate-network analysis helps explain where and how ENSO impacts are organized.