Open-Source
Weather Nowcasting
with Machine Learning

A grassroots, open-source effort advancing short-term weather prediction through machine learning — now also the official Working Group 6 on Nowcasting within EUMETNET's E-AI Programme.

[OVERVIEW]

What is MLCast?

MLCast Community is a decentralized ecosystem of researchers, engineers, and meteorologists building open-source infrastructure for high-precision weather nowcasting. We leverage modern machine learning techniques to process vast amounts of radar data in real-time, providing critical lead time for severe weather events.

We currently gather radar precipitation across Europe, growing country by country. Drag, scroll, or use the controls to explore where data is live — and where we still need your help.

Satellite data support is coming soon — broadening MLCast beyond radar for richer, global nowcasting.

Dataset Catalog

Radar precipitation

Expanding
Live Wanted

Countries

5

Years of rain

~70

Cadence

5m

Help us expand coverage — is your country missing?

[MISSION]

Collaborative Research Ecosystem

Our mission is to democratize access to advanced meteorological prediction tools. By fostering an open-source community, we aim to accelerate innovation in weather nowcasting, bridging the gap between academic research and real-world requirements.

  • check_circle Standardized, documented ARCO datasets for Machine Learning weather forecasts.
  • check_circle Modular approch, supporting customizable architectures.
  • check_circle Transparent evaluation frameworks preventing "black box" science.
[EUROPEAN NETWORKS]

Integrated into European Meteorological Networks, Open to the world.

MLCast is embedded in the European meteorological networks EUMETNET with the Nowcasting and E-AI Programmes. This connection supports coordination between national meteorological services, research communities, and operational nowcasting experts, helping to align open-source machine-learning developments with operational requirements.

The community leverages the European Weather Cloud for hosting and accessing shared datasets. In collaboration with EUMETSAT, MLCast also supports the provision of satellite data, including Meteosat Third Generation products, for machine-learning-based nowcasting.

These European ties extend the community's reach, but they are not a boundary: MLCast is open to contributions from any individual or organization, irrespective of geographical location.