Improved sub-seasonal to seasonal (S2S) forecasts could enhance food security, the sustainable management of energy and water resources, and reduce disaster risk by providing earlier warnings for natural hazards.
The World Meteorological Organization (WMO) has launched a competition to improve, through Artificial Intelligence and/or Machine Learning techniques, the current precipitation and temperature forecasts for 3 to 6 weeks into the future from the best computational fluid dynamic models available today.
All the codes and scripts will be hosted at Renkulab, developed by the Swiss Data Science Center, and training and verification data will be accessible from the European Weather Cloud and IRI Data Library. Data access scripts will be provided. After the competition, open access will be provided to all the codes and results.
Timeline
- Open since: 1st June 2021
- Closes: 31st October 2021
- Winners announced: Early February 2022
Prizes for the top three submissions
- Winning team: CHF 15 000
- Second team: CHF 10 000
- Third team: CHF 5 000
Competition website: https://s2s-ai-challenge.github.io/
This challenge is part of the Subseasonal-to-Seasonal Prediction Project (S2S Project), coordinated by the World Weather Research Programme (WWRP)/World Climate Research Programme (WCRP), in collaboration with the Swiss Data Science Center (SDSC) and the European Centre for Medium-Range Weather Forecasts (ECMWF).
To get the announcement, click here.