Long-term prediction of ENSO with physics-guided Deep Echo State Networks

TL;DR

This paper introduces a physics-guided Deep Echo State Network that predicts ENSO events up to 20 months ahead by leveraging climate physics, revealing a predictability horizon of about 30 months with minimal computational resources.

Contribution

The study develops a novel physics-guided Deep Echo State Network that enhances long-term ENSO prediction accuracy and interpretability using climate physics-based modes.

Findings

Achieves ENSO prediction up to 16-20 months ahead.

Identifies a finite predictability horizon of approximately 30 months.

Demonstrates the effectiveness of physics-guided reservoir computing for climate prediction.

Abstract

The El Ni\~{n}o-Southern Oscillation (ENSO) is a dominant mode of interannual climate variability, yet the mechanisms limiting its long-lead predictability remain unclear. Here we develop a physics-guided Deep Echo State Network (DESN) that operates on physically interpretable climate modes selected from the extended recharge oscillator (XRO) framework. DESN achieves skillful Ni\~{n}o3.4 predictions up to 16-20 months ahead with minimal computational cost. Mechanistic experiments show that extended predictability arises from nonlinear coupling between warm water volume and inter-basin climate modes. Error-growth analysis further indicates a finite ENSO predictability horizon of approximately 30 months. These results demonstrate that physics-guided reservoir computing provides an efficient and interpretable framework for diagnosing and predicting ENSO at long lead times.