PhyDA: Physics-Guided Diffusion Models for Data Assimilation in Atmospheric Systems

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Wang, Hao, Han, Jindong, Fan, Wei, Zhang, Weijia, Liu, Hao
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910953112600576
author Wang, Hao
Han, Jindong
Fan, Wei
Zhang, Weijia
Liu, Hao
author_facet Wang, Hao
Han, Jindong
Fan, Wei
Zhang, Weijia
Liu, Hao
contents Data Assimilation (DA) plays a critical role in atmospheric science by reconstructing spatially continous estimates of the system state, which serves as initial conditions for scientific analysis. While recent advances in diffusion models have shown great potential for DA tasks, most existing approaches remain purely data-driven and often overlook the physical laws that govern complex atmospheric dynamics. As a result, they may yield physically inconsistent reconstructions that impair downstream applications. To overcome this limitation, we propose PhyDA, a physics-guided diffusion framework designed to ensure physical coherence in atmospheric data assimilation. PhyDA introduces two key components: (1) a Physically Regularized Diffusion Objective that integrates physical constraints into the training process by penalizing deviations from known physical laws expressed as partial differential equations, and (2) a Virtual Reconstruction Encoder that bridges observational sparsity for structured latent representations, further enhancing the model's ability to infer complete and physically coherent states. Experiments on the ERA5 reanalysis dataset demonstrate that PhyDA achieves superior accuracy and better physical plausibility compared to state-of-the-art baselines. Our results emphasize the importance of combining generative modeling with domain-specific physical knowledge and show that PhyDA offers a promising direction for improving real-world data assimilation systems.
format Preprint
id arxiv_https___arxiv_org_abs_2505_12882
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PhyDA: Physics-Guided Diffusion Models for Data Assimilation in Atmospheric Systems
Wang, Hao
Han, Jindong
Fan, Wei
Zhang, Weijia
Liu, Hao
Machine Learning
Artificial Intelligence
Data Assimilation (DA) plays a critical role in atmospheric science by reconstructing spatially continous estimates of the system state, which serves as initial conditions for scientific analysis. While recent advances in diffusion models have shown great potential for DA tasks, most existing approaches remain purely data-driven and often overlook the physical laws that govern complex atmospheric dynamics. As a result, they may yield physically inconsistent reconstructions that impair downstream applications. To overcome this limitation, we propose PhyDA, a physics-guided diffusion framework designed to ensure physical coherence in atmospheric data assimilation. PhyDA introduces two key components: (1) a Physically Regularized Diffusion Objective that integrates physical constraints into the training process by penalizing deviations from known physical laws expressed as partial differential equations, and (2) a Virtual Reconstruction Encoder that bridges observational sparsity for structured latent representations, further enhancing the model's ability to infer complete and physically coherent states. Experiments on the ERA5 reanalysis dataset demonstrate that PhyDA achieves superior accuracy and better physical plausibility compared to state-of-the-art baselines. Our results emphasize the importance of combining generative modeling with domain-specific physical knowledge and show that PhyDA offers a promising direction for improving real-world data assimilation systems.
title PhyDA: Physics-Guided Diffusion Models for Data Assimilation in Atmospheric Systems
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2505.12882