Mercator Ocean International Makes AI Ocean Forecasting Operational with GLONET Validated by OceanBench at NeurIPS 2025

At NeurIPS 2025, one of the world’s leading AI conferences, Mercator Ocean unveiled OceanBench, the first open benchmark designed to assess AI Ocean forecasting models. This milestone follows the publication of Mercator Ocean’s AI model GLONET in September. Together, OceanBench and GLONET represent a decisive step toward integrating artificial intelligence into operational ocean forecasting, strengthening Europe’s leadership in trustworthy, AI-driven predictions at global scale.

Presented by Anass El Aouni, Oceanographer and machine learning (ML) researcher, and Quentin Gaudel, ML Systems Architect, OceanBench establishes a unified standard to evaluate AI models against fundamental ocean dynamics. GLONET is the first forecasting system to be validated through OceanBench for operational use.

Anass El Aouni : “OceanBench provides a transparent standard to evaluate AI-driven ocean forecasts. By comparing artificial intelligence with physics-based models, we can highlight strengths, identify areas for improvement, and guide the development of more reliable forecasting tools. There is no single ‘best’ model, but the benchmark ensures fair, science-based comparisons and helps users select the most suitable solution for each application.”

Quentin Gaudel: “Artificial intelligence is transforming the speed and scale of ocean prediction, but trust requires careful validation. OceanBench allows the community to rigorously measure performance while keeping AI models grounded in ocean physics. Combined with physics-based systems, AI can accelerate forecasts and broaden access to ocean information. This is critical for advancing operational forecasting and supporting Europe’s Digital Twin Ocean vision.”

A new standard for AI Ocean prediction

While AI has revolutionized weather forecasts, its operational use in ocean forecasting has been limited by the lack of shared standards for quality, validation, and physical realism. OceanBench fills this gap by providing:

• Open data and open-source tools for reproducible evaluation,
• Three evaluation tracks: Model-to-Reanalysis, Model-to-Analysis, Model-to-Observations,
• Physical diagnostics that verify forecasts against known ocean dynamics.

The first benchmark round includes four global forecasting systems: the high-resolution physics-based GLO12, and three ML ocean emulators — GLONET, XiHe, and Wenhai. Early results indicate AI models deliver rapid, accurate forecasts in many areas, but high-resolution physics-based systems remain essential for capturing fine-scale and high-frequency ocean processes. This combined approach is key to building reliable, operational solutions.

GLONET, Mercator Ocean’s new AI global forecasting system

Published in September 2025, and following full validation through OceanBench, GLONET is ready for operational deployment. Its key features include:

- Daily 10-day global forecasts generated within seconds,
- Forecast variables including sea surface height, temperature, salinity, and currents across 21 depth levels,
- Trained on Mercator Ocean’s GLORYS12 reanalysis, ensuring physical consistency.

Marie Drevillon, Head of Operations at Mercator Ocean and co-author of OceanBench: “With GLONET validated through OceanBench, AI becomes part of our operational forecasting toolkit. These capabilities enable faster simulations, broader access to ocean information, and ultimately more robust, science-based decision support.”

Supporting Europe’s Digital Ocean Vision

The launch of OceanBench and deployment of GLONET mark significant steps toward Europe’s vision for a Digital Twin Ocean capable of providing fast, reliable, and physically robust predictions. By establishing a community-driven standard for AI evaluation, OceanBench strengthens Europe’s leadership in trustworthy, operational AI ocean science.

By transitioning into an intergovernmental organization (IGO) representing its Member States, Mercator Ocean fosters long-term European collective action in support of this shared digital ocean vision.

Resources

• OceanBench Paper (NeurIPS 2025): NeurIPS Poster OceanBench: A Benchmark for Data-Driven Global Ocean Forecasting systems
• OceanBench Benchmark: https://oceanbench.lab.dive.edito.eu/
• GLONET : GLONET: Mercator's End‐to‐End Neural Global Ocean Forecasting System - El Aouni - 2025 - Journal of Geophysical Research: Machine Learning and Computation - Wiley Online Library

About OceanBench V1.0
“OceanBench: A Benchmark for Data-Driven Global Ocean Forecasting Systems”: published and presented at NeurIPS 2025. Authors include Anass El Aouni, Quentin Gaudel, Juan Emmanuel Johnson, Charly Regnier, Julien Le Sommer, Simon van Gennip, Ronan Fablet, Marie Drevillon, Yann Drillet, and Pierre-Yves Le Traon. OceanBench is implemented by Mercator Ocean International in collaboration with: Programme prioritaire de recherche Océan & Climat, IMT Atlantique, Université de Grenoble Alpes, IGEO Institute de Geociencias.

About Mercator Ocean
Mercator Ocean International is a global leader in digital oceanography and ocean forecasting. Since 2014, it has operated the Copernicus Marine Service on behalf of the European Commission, providing free, reliable ocean data and forecasts worldwide. The organization develops advanced digital ocean systems and leads the European Digital Twin Ocean initiative, supporting ocean decision-making with predictive, science-based insights. Currently evolving into a new intergovernmental organization (IGO), Mercator Ocean is strengthening collaboration across Europe and globally to advance digital ocean systems for a sustainable ocean. www.mercator-ocean.eu

View source version on businesswire.com: https://www.businesswire.com/news/home/20251210914358/en/