A Hybrid Deep-Learning Model for El Niño Southern Oscillation in the Low-Data Regime

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Schloer, Jakob, Newman, Matthew, Thuemmel, Jannik, Capotondi, Antonietta, Goswami, Bedartha
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909595261206528
author Schloer, Jakob
Newman, Matthew
Thuemmel, Jannik
Capotondi, Antonietta
Goswami, Bedartha
author_facet Schloer, Jakob
Newman, Matthew
Thuemmel, Jannik
Capotondi, Antonietta
Goswami, Bedartha
contents While deep-learning models have demonstrated skillful El Niño Southern Oscillation (ENSO) forecasts up to one year in advance, they are predominantly trained on climate model simulations that provide thousands of years of training data at the expense of introducing climate model biases. Simpler Linear Inverse Models (LIMs) trained on the much shorter observational record also make skillful ENSO predictions but do not capture predictable nonlinear processes. This motivates a hybrid approach, combining the LIMs modest data needs with a deep-learning non-Markovian correction of the LIM. For O(100 yr) datasets, our resulting Hybrid model is more skillful than the LIM while also exceeding the skill of a full deep-learning model. Additionally, while the most predictable ENSO events are still identified in advance by the LIM, they are better predicted by the Hybrid model, especially in the western tropical Pacific for leads beyond about 9 months, by capturing the subsequent asymmetric (warm versus cold phases) evolution of ENSO.
format Preprint
id arxiv_https___arxiv_org_abs_2412_03743
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Hybrid Deep-Learning Model for El Niño Southern Oscillation in the Low-Data Regime
Schloer, Jakob
Newman, Matthew
Thuemmel, Jannik
Capotondi, Antonietta
Goswami, Bedartha
Machine Learning
Atmospheric and Oceanic Physics
While deep-learning models have demonstrated skillful El Niño Southern Oscillation (ENSO) forecasts up to one year in advance, they are predominantly trained on climate model simulations that provide thousands of years of training data at the expense of introducing climate model biases. Simpler Linear Inverse Models (LIMs) trained on the much shorter observational record also make skillful ENSO predictions but do not capture predictable nonlinear processes. This motivates a hybrid approach, combining the LIMs modest data needs with a deep-learning non-Markovian correction of the LIM. For O(100 yr) datasets, our resulting Hybrid model is more skillful than the LIM while also exceeding the skill of a full deep-learning model. Additionally, while the most predictable ENSO events are still identified in advance by the LIM, they are better predicted by the Hybrid model, especially in the western tropical Pacific for leads beyond about 9 months, by capturing the subsequent asymmetric (warm versus cold phases) evolution of ENSO.
title A Hybrid Deep-Learning Model for El Niño Southern Oscillation in the Low-Data Regime
topic Machine Learning
Atmospheric and Oceanic Physics
url https://arxiv.org/abs/2412.03743