EXCLAIM Symposium 2025: Is AI the Future of Weather and Climate Modeling?
Venue
Rämistrasse 101
8092 Zürich
8092 Zürich
The EXCLAIM symposium explores the transformative potential of AI-driven and physics-based models in advancing weather and climate prediction. Join for an exciting three-day symposium of speakers from academia and industry at the forefront of the field. Through in-depth discussions and analysis, he latest breakthroughs in numerical and AI-based modeling techniques, highlighting key trends, challenges, synergies and opportunities that are shaping the future of this field will be examined.
Key Objectives
Analyze and discuss the latest advancements in numerical and AI-based weather and climate modeling. Identify key trends, challenges, and opportunities shaping the field. Leverage the synergies between numerical methods and AI-based approaches to advance weather and climate modeling, with a strong commitment to ethical, responsible, and reproducible AI practices. Strengthen personal connections and collaborations through networking.
Program
The three-day event features tutorials, keynote talks, oral presentations, poster sessions, and networking opportunities. It is organised along a tutorial, four main topics and a public keynote.
Tutorial - Machine learning for weather and climate
- Dr. Matthew Chantry (Strategic Lead for Machine Learning, ECMWF)
- Dr. Mihai Alexe (Scientist for Machine Learning, ECMWF)
Session A - Creating the foundations of AI-based modeling of weather and climate
- Dr. Peter Dueben (Head of Earth System Modelling, ECMWF; Honorary Professor, University of Cologne): "Yes – AI is a Part of the Future of Weather and Climate Modeling"
- Dr. Maria J. Molina (Assistant Professor, University of Maryland): "Machine Learning for Uncovering Patterns in the Earth Systems"
Session B - Harnessing the power of AI in modeling weather and climate
- Dr. Mike Pritchard (Director of Climate Simulation Research, NVIDIA Research; Professor of Earth System Science, University of California, Irvine): "Beyond Weather Emulators and High-Resolution Simulations: The Unique Challenges of ML for Climate"
Session C - Harnessing the power of physics-based modeling of weather and climate
- Dr. Cathy Hohenegger (Independent Research Group Leader, Max Planck Institute for Meteorology): "Learning from Kilometre-Scale Global Simulations of the Earth System"
- Prof. Pier Luigi Vidale (Professor of Climate System Science, University of Reading): "Weather and Climate Models at km-scale: complementing and challenging observations"
Session D - Merging AI and physics-based modeling of weather and climate
- Dr. Stephan Hoyer (Senior Staff Software Engineer, Google Research): "How AI Will Transform Earth System Modeling"
- Dr. Inna Polichtchouk (Senior Scientist, ECMWF) "Nudging the Large Scales of a Physics-Based Model Toward ML Forecasts"
Public Keynote
- Martin Vetterli (Full Professor of Audiovisual Communications Laboratory, EPFL): "AI for Science, AI for Society: Two Sides of the Same Coin?"
The keynote by Martin Vetterli is open to the public and free of charge. However, registration is required (Registration deadline: May 31, 2025). Participants already registered for the EXCLAIM Symposium do not need to register again.
For full details on the schedule, session topics, and presentations, please refer to the Program.
Target Audience
- Scientific communities working on both numerical and AI-based modeling approaches
- Representatives from companies working on high performance computing (HPC) and machine learning methods
- Early career researchers in the above mentioned fields
Format
The EXCLAIM symposium is organized as an on-site event with 150 to 200 expected participants. Additionally, the sessions will be live-streamed (one-way streaming). Registration is mandatory for all participants.
The official language of the symposium is English.
