Diffusion Models Bridge Deep Learning and Physics in ENSO Forecasting

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xu, Weifeng, Zhu, Xiang, Li, Xiaoyong, Yao, Qiang, Ren, Xiaoli, Deng, Kefeng, Wu, Song, Shao, Chengcheng, Xu, Xiaolong, Zhao, Juan, Zhao, Chengwu, Cao, Jianping, Wang, Jingnan, Wang, Wuxin, Li, Qixiu, Gao, Xiaori, Wu, Xinrong, Wang, Huizan, Cao, Xiaoqun, Zhang, Weiming, Song, Junqiang, Ren, Kaijun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908689804296192
author Xu, Weifeng
Zhu, Xiang
Li, Xiaoyong
Yao, Qiang
Ren, Xiaoli
Deng, Kefeng
Wu, Song
Shao, Chengcheng
Xu, Xiaolong
Zhao, Juan
Zhao, Chengwu
Cao, Jianping
Wang, Jingnan
Wang, Wuxin
Li, Qixiu
Gao, Xiaori
Wu, Xinrong
Wang, Huizan
Cao, Xiaoqun
Zhang, Weiming
Song, Junqiang
Ren, Kaijun
author_facet Xu, Weifeng
Zhu, Xiang
Li, Xiaoyong
Yao, Qiang
Ren, Xiaoli
Deng, Kefeng
Wu, Song
Shao, Chengcheng
Xu, Xiaolong
Zhao, Juan
Zhao, Chengwu
Cao, Jianping
Wang, Jingnan
Wang, Wuxin
Li, Qixiu
Gao, Xiaori
Wu, Xinrong
Wang, Huizan
Cao, Xiaoqun
Zhang, Weiming
Song, Junqiang
Ren, Kaijun
contents Accurate long-range forecasting of the El \Nino-Southern Oscillation (ENSO) is vital for global climate prediction and disaster risk management. Yet, limited understanding of ENSO's physical mechanisms constrains both numerical and deep learning approaches, which often struggle to balance predictive accuracy with physical interpretability. Here, we introduce a data driven model for ENSO prediction based on conditional diffusion model. By constructing a probabilistic mapping from historical to future states using higher-order Markov chain, our model explicitly quantifies intrinsic uncertainty. The approach achieves extending lead times of state-of-the-art methods, resolving early development signals of the spring predictability barrier, and faithfully reproducing the spatiotemporal evolution of historical extreme events. The most striking implication is that our analysis reveals that the reverse diffusion process inherently encodes the classical recharge-discharge mechanism, with its operational dynamics exhibiting remarkable consistency with the governing principles of the van der Pol oscillator equation. These findings establish diffusion models as a new paradigm for ENSO forecasting, offering not only superior probabilistic skill but also a physically grounded theoretical framework that bridges data-driven prediction with deterministic dynamical systems, thereby advancing the study of complex geophysical processes.
format Preprint
id arxiv_https___arxiv_org_abs_2511_01214
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Diffusion Models Bridge Deep Learning and Physics in ENSO Forecasting
Xu, Weifeng
Zhu, Xiang
Li, Xiaoyong
Yao, Qiang
Ren, Xiaoli
Deng, Kefeng
Wu, Song
Shao, Chengcheng
Xu, Xiaolong
Zhao, Juan
Zhao, Chengwu
Cao, Jianping
Wang, Jingnan
Wang, Wuxin
Li, Qixiu
Gao, Xiaori
Wu, Xinrong
Wang, Huizan
Cao, Xiaoqun
Zhang, Weiming
Song, Junqiang
Ren, Kaijun
Geophysics
Accurate long-range forecasting of the El \Nino-Southern Oscillation (ENSO) is vital for global climate prediction and disaster risk management. Yet, limited understanding of ENSO's physical mechanisms constrains both numerical and deep learning approaches, which often struggle to balance predictive accuracy with physical interpretability. Here, we introduce a data driven model for ENSO prediction based on conditional diffusion model. By constructing a probabilistic mapping from historical to future states using higher-order Markov chain, our model explicitly quantifies intrinsic uncertainty. The approach achieves extending lead times of state-of-the-art methods, resolving early development signals of the spring predictability barrier, and faithfully reproducing the spatiotemporal evolution of historical extreme events. The most striking implication is that our analysis reveals that the reverse diffusion process inherently encodes the classical recharge-discharge mechanism, with its operational dynamics exhibiting remarkable consistency with the governing principles of the van der Pol oscillator equation. These findings establish diffusion models as a new paradigm for ENSO forecasting, offering not only superior probabilistic skill but also a physically grounded theoretical framework that bridges data-driven prediction with deterministic dynamical systems, thereby advancing the study of complex geophysical processes.
title Diffusion Models Bridge Deep Learning and Physics in ENSO Forecasting
topic Geophysics
url https://arxiv.org/abs/2511.01214