OceanLight: Efficient Global Ocean Forecasting via Geometry-Adaptive Unstructured Mesh Representation

2026-08-17

AI Summary

This study addresses the high computational cost of physical models and grid redundancy in deep learning for global ocean forecasting by proposing a novel framework integrating geometry-adaptive unstructured mesh tokenization with graph neural networks. Overcoming structured grid limitations, this approach establishes a scalable modeling paradigm that simultaneously preserves geostrophic balance and mesoscale eddy representation. Experimental results demonstrate that the proposed method surpasses state-of-the-art baselines in prediction accuracy and energy spectrum fidelity while achieving optimal geostrophic consistency. Furthermore, it significantly reduces computational overhead, decreasing GPU memory usage by 62% and FLOPs by 70%, thereby enabling efficient and high-precision global ocean forecasting.

Abstract

Reliable global ocean forecasting is critical for climate monitoring, marine navigation, and extreme event early warning. Physics-based ocean forecasting models impose prohibitive computational costs, while existing deep learning approaches predominantly rely on structured-grid architectures, incurring unnecessary computation on masked land cells and enforcing uniform resolution across dynamically heterogeneous ocean regions regardless of local flow complexity. Here we present OceanLight, an efficient global ocean forecasting framework innovatively combining geometry-adaptive unstructured mesh tokenization with a graph neural network (GNN) backbone. OceanLight achieves pointwise forecast accuracy and kinetic energy spectral fidelity exceeding both operational numerical analyses and state-of-the-art AI-based models, while surpassing all AI-based ocean models in geostrophic balance consistency. Furthermore, OceanLight demonstrates reliable mesoscale eddy representation, capturing coherent ocean structures beyond pointwise statistical optimization. These capabilities are delivered with a 62% reduction in GPU memory consumption and 70\% reduction in FLOPs relative to structured-grid baselines. Our unstructured mesh representation establishes a generalizable paradigm for scalable data-driven oceanography.

Problem

• Global Ocean Forecasting
• Computational Efficiency
• Structured Grid Limitations
• Adaptive Resolution

Innovation

• Geometry-Adaptive Unstructured Mesh
• Graph Neural Network
• Efficient Global Ocean Forecasting
• Mesh Tokenization
• Geostrophic Balance Consistency