SOON: Symmetric Orthogonal Operator Network for Global Subseasonal-to-Seasonal Climate Forecasting

TL;DR

The paper introduces SOON, a novel neural network architecture that improves global climate forecasting by better modeling anisotropic atmospheric dynamics, leading to more accurate and efficient predictions.

Contribution

It proposes the Symmetric Orthogonal Operator Network with anisotropic embedding and symmetric decomposition, addressing limitations of isotropic models in climate prediction.

Findings

SOON outperforms existing models in accuracy.

SOON achieves higher computational efficiency.

The method effectively models anisotropic atmospheric processes.

Abstract

Accurate global Subseasonal-to-Seasonal (S2S) climate forecasting is critical for disaster preparedness and resource management, yet it remains challenging due to chaotic atmospheric dynamics. Existing models predominantly treat atmospheric fields as isotropic images, conflating the distinct physical processes of zonal wave propagation and meridional transport, and leading to suboptimal modeling of anisotropic dynamics. In this paper, we propose the Symmetric Orthogonal Operator Network (SOON) for global S2S climate forecasting. It couples: (1) an Anisotropic Embedding strategy that tokenizes the global grid into latitudinal rings, preserving the integrity of zonal periodic structures; and (2) a stack of SOON Blocks that models the alternating interaction of Zonal and Meridional Operators via a symmetric decomposition, structurally mitigating discretization errors inherent in long-term integration. Extensive experiments on the Earth Reanalysis 5 dataset demonstrate that SOON establishes a new state-of-the-art, significantly outperforming existing methods in both forecasting accuracy and computational efficiency.