01.10.2026

Around 35 participants, including challenge leads, PhD students, postdoctoral researchers and experienced professionals, are taking part in the hackathon, both in person and remotely. They represent around 30 institutions across Europe, Africa, Asia, North America and Oceania, reflecting the event’s international scope. 

The hackathon focuses on a growing challenge in climate research: how to efficiently analyse the increasingly large and distributed datasets generated by high-resolution climate simulations and large ensemble experiments. EXPECT is developing new data infrastructure to support this work and make the analysis of distributed climate data more efficient.

During the three-day event, participants first receive an introduction to the new distributed climate data processing infrastructure. They then apply the tools in hands-on projects, working in small groups on specific climate science challenges.

The projects cover a range of topics at the interface of climate science, data analysis and artificial intelligence. Participants investigate changes in atmospheric circulation patterns associated with heat extremes in Europe, with a particular focus on double-jet regimes. Other projects explore the use of deep learning for climate research, climate data downscaling with the ECMWF Weather Generator, and explainable AI methods to better understand the drivers of changes in climate.

For their hands-on work, participants have access to DKRZ’s high-performance computing system Levante, providing the computing resources needed for the data-intensive challenges. The programme also includes a guided tour of DKRZ’s high-performance computing facilities, giving participants a closer look at the infrastructure behind climate research at DKRZ.

By combining new data infrastructure with concrete scientific questions, the event provides an opportunity to test new approaches directly on climate research problems.