Workshop on AI in Weather and Climate Science – Highlights & Recordings​

​​​​

On 7 Oct. 2026, this workshop brought together researchers from the Netherlands working at the intersection of atmospheric science, climate research, and machine learning to exchange on recent advances in the field. Machine learning is rapidly transforming atmospheric and climate science. The topics of the workshop included but were not limited to explainability, climate modelling, weather forecasting, and nowcasting applications. The workshop fostered exchange on recent scientific advances and connected researchers in the field as part of AIMET. The program provided especially early-career researchers with an opportunity to present and discuss their work. Recordings and presentation slides are being made available below alongside the workshop programme.

​

Photo gallery

Group photo 1    photo 2    photo 3    photo 4

​​

​​

​

​

​

​

​

​

​

​

​​

​

​

​

​

​

​

​

​

​

 

​

 

Speakers included

• Serban Vadineanu (KNMI) 

• Max Zhdanov (Google DeepMind, University of Amsterdam) 

• Giannis Dravilas (University of Amsterdam, EUMETSAT)

• Alice Abramowicz (KNMI)

• Francesco Pinto (TU Delft, BFH)

• Sophie Buurman (KNMI)

• Jelmer van der Graaff (Wageningen University, Northpool B.V.)

• Max Pierzyna (TU Delft)

• Francesco Pasquini (Utrecht University, KNMI)

• Tage Tykesson (Chalmers, TU Delft)

​​​

Programme

11:00    Opening

11:10    AI for weather forecasting

• Serban Vadineanu: Pretraining high-resolution European ML weather ensembles with a stretched-grid framework

• Francesco Pasquini: Assessing the ability of a stretched-grid deep-learning weather prediction model to capture physical balances

• Alice Abramowicz: Advancing data assimilation capacities with deep learning​

• Giannis Dravilas: Deep learning-based radar reflectivity estimation from MTG-FCI and LI observations

12:15    Lunch

13:30    AI for weather forecasting (cont.)

• Sophie Buurman: Towards a data-driven weather model for forecasting on-demand extremes at hectometric scale 

• Max Zhdanov: (Sparse) Attention to the details: preserving spectral fidelity in ML-based weather forecasting models

• Francesco Pinto: Satellite-based offshore wind nowcasting: addressing spatio-temporal irregularities of scatterometer observations with deep learning

• Carlo Saccardi: Diffusion models as learned priors for inverse problems, with an application to weather

14:30    Coffee break

15:15    Atmosphere, Climate and AI

• Jelmer van der Graaff: A new method to identify and explain sources of precipitation modification, illustrated for the western Netherlands

• Max Pierzyna: Estimating high-resolution profiles of atmospheric turbulence from reanalysis using deep learning

• Tage Tykesson: Towards efficient climate model emulation with transformer-based Earth system models

16:00    Drinks
​​​​

​

We thank all speakers and participants for their valuable contributions and engaging discussions.​​