Improvement of sub-seasonal probabilistic forecasts of European high-impact weather events using machine learning techniques

Each year Europe is subject to drought, heatwaves and periods of persistent rainfall that could lead to the flooding of rivers. Although short-range weather forecasts have improved substantially over the last decennia, long-range weather forecasts have improved less. The goal of this project is to improve these long-term probabilistic forecasts of extreme weather. Warnings can then be given earlier and more reliably. Long meteorological datasets and newly developed statistical post-processing methods enable us to better integrate the relevant information, and correct shortcomings of operational ensemble prediction systems.

Chiem van Straaten is a PhD student at The Institute for Environmental Studies ( and KNMI, who is working with principal investigator Dr Maurice Schmeits (KNMI), Professor Bart van den Hurk (VU and Deltares), Dr Dim Coumou (VU and KNMI) and Dr Kirien Whan (KNMI).

IMPRINT is concerned with extremes that occur in the tails of the distribution.

The project is funded by NWO project number ALWOP.395.


  • van Straaten, C., Whan, K., Coumou, D., van den Hurk, B., and Schmeits, M. 2019. The influence of aggregation and statistical post-processing on the sub-seasonal predictability of European temperatures. QJRMS (under review)

Further reading: