ORCAst: Operational High-Resolution Current Forecasts

2025-01-21

AI Summary

This work addresses the limitations in accuracy and real-time capability of current one-week, high-resolution global sea surface current forecasts. We propose a novel multi-stage, multi-branch deep learning framework. Methodologically: (1) we introduce a staged supervision paradigm—first training on dense satellite altimetry data to learn the sea surface height–geostrophic current mapping, then transferring knowledge to sparse in-situ drifter velocity observations; (2) we design a multi-arm encoder-decoder architecture that jointly fuses heterogeneous nadir and SWOT altimetry data with drifter measurements, enabling regional adaptivity and end-to-end learning under physical constraints. Experiments demonstrate that our method consistently outperforms state-of-the-art approaches in both sea surface current assimilation and 7-day forecasting, achieving significant improvements in forecast accuracy and real-time responsiveness.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

We present ORCAst, a multi-stage, multi-arm network for Operational high-Resolution Current forecAsts over one week. Producing real-time nowcasts and forecasts of ocean surface currents is a challenging problem due to indirect or incomplete information from satellite remote sensing data. Entirely trained on real satellite data and in situ measurements from drifters, our model learns to forecast global ocean surface currents using various sources of ground truth observations in a multi-stage learning procedure. Our multi-arm encoder-decoder model architecture allows us to first predict sea surface height and geostrophic currents from larger quantities of nadir and SWOT altimetry data, before learning to predict ocean surface currents from much more sparse in situ measurements from drifters. Training our model on specific regions improves performance. Our model achieves stronger nowcast and forecast performance in predicting ocean surface currents than various state-of-the-art methods.

Problem

• ocean current prediction
• accuracy improvement
• future forecasting

Innovation

• ORCAst Model
• Satellite Data Integration
• Drifter Measurements