🌍 5th ECMWF–ESA Machine Learning Workshop | Bologna 13-17/04/2026

We had the opportunity to attend the 5th ECMWF–ESA Machine Learning Workshop at the DAMA Technopole in Bologna — a key event at the intersection of AI, Earth Observation, and climate science.

A highlight of the experience was participating in a guided visit to the European Centre for Medium-Range Weather Forecasts (ECMWF) infrastructure, where we explored the high-performance computing systems that support operational weather and climate prediction. Seeing these systems up close gave valuable insight into the scale and complexity behind modern forecasting.


🎯 Workshop motivation and scope

The workshop focused on the rapidly growing role of Machine Learning (ML) in Earth System Observation and Prediction (ESOP).

ML is now transforming the field:

  • From supporting tools → to end-to-end data-driven forecasting systems
  • From research concepts → to operational applications

At the same time, important challenges remain:

  • How far can purely data-driven models go without physics?
  • How to best combine ML with physical models for reliable predictions?

In parallel, the field is evolving technologically:

  • Transition from CPU → GPU/TPU-based HPC systems
  • Emerging paradigms like quantum, edge, and neuromorphic computing

đź§  Key thematic areas

The workshop covered a wide range of topics, including:

  • Hybrid ML–physics systems for weather and climate prediction
  • End-to-end ML frameworks for forecasting
  • ML applications for Earth Observation data
  • High-performance computing and next-generation architectures
  • Machine learning for Digital Twins of the Earth system

The programme provided a comprehensive snapshot of the state of the art, combining keynotes, technical sessions, and discussions across the ESOP community


🚀 Impact for our work

Beyond the technical content, the workshop was highly motivational. It reinforced the importance of:

  • Developing AI-based downscaling approaches
  • Combining CAMS, satellite, and in-situ data
  • Moving towards impact-driven applications (e.g. health, environment, policy)

This experience strongly motivates us to push forward and deliver high-quality, innovative solutions within our projects at the Eratosthenes Centre of Excellence.


Share your love