pipette
ENEnglish

Spherical Physics-informed Neural Operator with Multi-scale Coupling for Meteorological Downscaling

Yiqiang Ye, Yichi Wang, Haodi Zhang, Jiawei Wen, Jiahui Jiang, Zhaoyu Zhong, Jiangjian Yu, Chunxia Xiao

Preprint con versión publicadaAfirmaciones fuertes, leer con cuidadoUso en el mundo real

En palabras de los autores

Meteorological downscaling is crucial for high-resolution regional climate forecasting and disaster early warning. While neural operators have emerged as a promising paradigm for modeling complex spatiotemporal mappings, existing frameworks often struggle with spherical manifold geometric distortions, inherent atmospheric multi-scale coupling mismatches, and lack of explicit atmospheric laws. We propose the Spherical Physics-informed Neural Operator, which utilizes a Spherical Laplacian Decomposition to partition atmospheric fields into hierarchical frequency components, maintaining exact point-wise correspondence across scales. To evaluate these representations at arbitrary locations, we introduce a localized spherical integral operator that approximates continuous kernel transforms via geometry-aware attention. Dynamical consistency is further enforced by embedding differentiable constraints into the learning process. Extensive experiments demonstrate that our framework attains superior accuracy and zero-shot generalization across various meteorological variables and unseen queries, representing a robust and interpretable solution for global-to-regional meteorological downscaling.

Resultado principalEl resumen no menciona limitaciones.

Apareció: martes, 20 de octubre. arXiv. Preprint con versión publicada.

DOI: 10.24963/ijcai.2026/734

Versión publicada: Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence(IJCAI-26), pp. 6600-6608, 2026

Comentario de los autores: Published in IJCAI 2026 (AI4Tech: AI Enabling Technologies). 9 pages, 4 figures