NeuralOM: Neural Ocean Model for Subseasonal-to-Seasonal Simulation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gao, Yuan, Wu, Hao, Xu, Fan, Xiang, Yanfei, Gou, Ruijian, Shu, Ruiqi, Wen, Qingsong, Wu, Xian, Wang, Kun, Huang, Xiaomeng
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918294122921984
author Gao, Yuan
Wu, Hao
Xu, Fan
Xiang, Yanfei
Gou, Ruijian
Shu, Ruiqi
Wen, Qingsong
Wu, Xian
Wang, Kun
Huang, Xiaomeng
author_facet Gao, Yuan
Wu, Hao
Xu, Fan
Xiang, Yanfei
Gou, Ruijian
Shu, Ruiqi
Wen, Qingsong
Wu, Xian
Wang, Kun
Huang, Xiaomeng
contents Long-term, high-fidelity simulation of slow-changing physical systems, such as the ocean and climate, presents a fundamental challenge in scientific computing. Traditional autoregressive machine learning models often fail in these tasks as minor errors accumulate and lead to rapid forecast degradation. To address this problem, we propose NeuralOM, a general neural operator framework designed for simulating complex, slow-changing dynamics. NeuralOM's core consists of two key innovations: (1) a Progressive Residual Correction Framework that decomposes the forecasting task into a series of fine-grained refinement steps, effectively suppressing long-term error accumulation; and (2) a Physics-Guided Graph Network whose built-in adaptive messaging mechanism explicitly models multi-scale physical interactions, such as gradient-driven flows and multiplicative couplings, thereby enhancing physical consistency while maintaining computational efficiency. We validate NeuralOM on the challenging task of global Subseasonal-to-Seasonal (S2S) ocean simulation. Extensive experiments demonstrate that NeuralOM not only surpasses state-of-the-art models in forecast accuracy and long-term stability, but also excels in simulating extreme events. For instance, at a 60-day lead time, NeuralOM achieves a 13.3% lower RMSE compared to the best-performing baseline, offering a stable, efficient, and physically-aware paradigm for data-driven scientific computing. Code link: https://github.com/YuanGao-YG/NeuralOM.
format Preprint
id arxiv_https___arxiv_org_abs_2505_21020
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle NeuralOM: Neural Ocean Model for Subseasonal-to-Seasonal Simulation
Gao, Yuan
Wu, Hao
Xu, Fan
Xiang, Yanfei
Gou, Ruijian
Shu, Ruiqi
Wen, Qingsong
Wu, Xian
Wang, Kun
Huang, Xiaomeng
Machine Learning
Atmospheric and Oceanic Physics
Long-term, high-fidelity simulation of slow-changing physical systems, such as the ocean and climate, presents a fundamental challenge in scientific computing. Traditional autoregressive machine learning models often fail in these tasks as minor errors accumulate and lead to rapid forecast degradation. To address this problem, we propose NeuralOM, a general neural operator framework designed for simulating complex, slow-changing dynamics. NeuralOM's core consists of two key innovations: (1) a Progressive Residual Correction Framework that decomposes the forecasting task into a series of fine-grained refinement steps, effectively suppressing long-term error accumulation; and (2) a Physics-Guided Graph Network whose built-in adaptive messaging mechanism explicitly models multi-scale physical interactions, such as gradient-driven flows and multiplicative couplings, thereby enhancing physical consistency while maintaining computational efficiency. We validate NeuralOM on the challenging task of global Subseasonal-to-Seasonal (S2S) ocean simulation. Extensive experiments demonstrate that NeuralOM not only surpasses state-of-the-art models in forecast accuracy and long-term stability, but also excels in simulating extreme events. For instance, at a 60-day lead time, NeuralOM achieves a 13.3% lower RMSE compared to the best-performing baseline, offering a stable, efficient, and physically-aware paradigm for data-driven scientific computing. Code link: https://github.com/YuanGao-YG/NeuralOM.
title NeuralOM: Neural Ocean Model for Subseasonal-to-Seasonal Simulation
topic Machine Learning
Atmospheric and Oceanic Physics
url https://arxiv.org/abs/2505.21020