Leveraging an Atmospheric Foundational Model for Subregional Sea Surface Temperature Forecasting

2025-10-29

AI Summary

Traditional numerical ocean models suffer from high computational costs and poor scalability, hindering high-resolution sea surface temperature (SST) forecasting. Method: This study innovatively adapts the pre-trained atmospheric foundation model Aurora to the ocean domain, leveraging high-resolution ocean reanalysis data to develop a purely data-driven SST forecasting model tailored to the Canary Upwelling System. We employ a staged fine-tuning strategy, latitude-weighted loss function, and systematic hyperparameter optimization to enhance spatiotemporal modeling efficiency and generalization. Contribution/Results: The model achieves a root-mean-square error of 0.119 K and an anomaly correlation coefficient of 0.997 on the test set, accurately reproducing large-scale SST patterns. It establishes a scalable, computationally efficient paradigm for climate research and marine resource management, bridging the gap between foundation models and operational oceanography.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Cognitive Modeling & Cognitive Systems: Other Foundations of Cognitive Modeling & Systems
• Computer Vision: Diffusion Models for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

The accurate prediction of oceanographic variables is crucial for understanding climate change, managing marine resources, and optimizing maritime activities. Traditional ocean forecasting relies on numerical models; however, these approaches face limitations in terms of computational cost and scalability. In this study, we adapt Aurora, a foundational deep learning model originally designed for atmospheric forecasting, to predict sea surface temperature (SST) in the Canary Upwelling System. By fine-tuning this model with high-resolution oceanographic reanalysis data, we demonstrate its ability to capture complex spatiotemporal patterns while reducing computational demands. Our methodology involves a staged fine-tuning process, incorporating latitude-weighted error metrics and optimizing hyperparameters for efficient learning. The experimental results show that the model achieves a low RMSE of 0.119K, maintaining high anomaly correlation coefficients (ACC $approx 0.997$). The model successfully reproduces large-scale SST structures but faces challenges in capturing finer details in coastal regions. This work contributes to the field of data-driven ocean forecasting by demonstrating the feasibility of using deep learning models pre-trained in different domains for oceanic applications. Future improvements include integrating additional oceanographic variables, increasing spatial resolution, and exploring physics-informed neural networks to enhance interpretability and understanding. These advancements can improve climate modeling and ocean prediction accuracy, supporting decision-making in environmental and economic sectors.

Problem

• Adapting atmospheric deep learning model for sea temperature prediction
• Reducing computational costs in ocean forecasting using AI
• Improving subregional SST forecasts in Canary Upwelling System

Innovation

• Adapting atmospheric deep learning model for ocean forecasting
• Fine-tuning with high-resolution oceanographic reanalysis data
• Using staged fine-tuning process with weighted error metrics