EUMETSAT to contribute to European meteorological AI initiative
Programme aims to harness technological advances and help member states make greater use of EUMETSAT data
Europe’s meteorological satellite agency, EUMETSAT, will contribute its expertise in data curation to a Europe-wide initiative aiming to enhance the use of artificial intelligence and machine learning in weather and climate applications.
26 June 2024
26 June 2024
EUMETSAT’s 30-member-state council today approved the organisation’s involvement in the EUMETNET Artificial Intelligence and Machine Learning for Weather, Climate and Environmental Applications programme. EUMETNET is a grouping of 33 European national meteorological services that provides a framework for cooperation on meteorological activities.
“This programme will help our member states further exploit EUMETSAT satellite data and get the most out of the advances made possible by artificial intelligence (AI) and machine learning,” EUMETSAT Chief Scientist Dr Paolo Ruti said.
“Machine learning methods have already proven capable of producing reliable short-to-medium-range weather forecasts. The application of AI and ML technologies is not limited to weather prediction, however, and will extend to other areas including analysis of satellite data and fusion of satellite imagery into a single, more interpretable product.”
EUMETSAT’s contribution to the programme will mainly include:
- User engagement, to gauge needs for training data, tools and cloud-based collaboration
- Constructing robust training datasets for machine learning applications, such as nowcasting and weather forecasting
- Ensuring and maintaining the quality of machine learning training datasets
- Ensuring satellite data are accessible through easy-to-use portals or application programming interfaces
- Providing infrastructure to allow member states to retrain ML-based forecasting models currently available at national scale to European scale, and making them available to member states for easy installation and development
“A key objective of the EUMETNET programme is to widely share its achievements to foster a culture of openness and collaboration and facilitate the uptake of AI methods within the European meteorological community,” Ruti said.
“This is consistent with EUMETSAT’s own strategic approach to develop AI and ML action areas that are relevant to our members’ capabilities and internal EUMETSAT applications.”
Main image: EUMETSAT headquarters in Darmstadt, Germany.
Find out more about EUMETSAT
Press release translations
