Key Insights from The Third Workshop on Machine Learning for Earth System Modelling

From 25–27 August 2025, the third workshop on Machine Learning for the Earth System took place at the Universitätsclub Bonn, Germany. This focused workshop brought together approximately 140 on-site participants and 100 virtual attendees, fostering a dynamic exchange of ideas among researchers and practitioners in Earth system science and machine learning.

The workshop was organized by the Center for Earth System Observation and Computational Analysis (CESOC), the European Centre for Medium-Range Weather Forecasts (ECMWF), and with the support of the TRA Modelling at the University of Bonn, the Forschungszentrum Jülich and the University of Cologne. It featured a comprehensive programme that included:

Thematic Sessions: Focused discussions where experts shared insights on advanced machine learning techniques and their applications in Earth system modelling and topics such as numerical weather prediction, climate emulation, and Earth system component modelling.
Poster Sessions: Opportunities for participants to present and discuss their research findings in an interactive format.
Networking Opportunities: Facilitated connections among attendees to encourage collaboration and knowledge sharing.

The programme covered a range of topics

Machine learning-based numerical weather prediction.
Climate emulators and machine learning-based climate models.
Machine learning-based Earth system component models, such as ocean and land models.
Incorporation of Earth system observations in the above models.
Evaluation and explainability of the above models.
Datasets for model training, evaluation, and benchmarking.

The event not only highlighted the growing importance of integrating machine learning into Earth system research but also strengthened a community committed to ongoing collaboration. Building on this momentum, plans to continue the workshop series are being considered for next year.

For more details on the programme and speakers, visit the workshop page.