---
title: "Spherical Physics-Informed Neural Operator with Multi-Scale Coupling for Meteorological Downscaling"
authors:
  - "Yiqiang Ye"
  - "Yichi Wang"
  - "Jiawei Wen"
  - "Jiahui Jiang"
  - "Zhaoyu Zhong"
  - "Jiangjian Yu"
  - "Chunxia Xiao"
  - "Haodi Zhang"
doi: "10.24963/ijcai.2026/734"
url: "https://ijcai.org/proceedings/2026/734"
pdf: "https://ijcai.org/proceedings/2026/0734.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "6600-6608"
track: "AI4Tech: AI Enabling Technologies"
keywords:
  - "Domain-specific AI4Tech: Other AI4Tech applications"
---

# Spherical Physics-Informed Neural Operator with Multi-Scale Coupling for Meteorological Downscaling

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

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 6600-6608. DOI: [10.24963/ijcai.2026/734](https://doi.org/10.24963/ijcai.2026/734)

## Abstract

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.