Year: 2025
Type: article
Abstract: Sub-seasonal forecasting, which involves predicting weather conditions two to six weeks in advance, presents a major challenge for operational centres due to limited predictability from initial or slowly varying boundary conditions. In light of recent advances in artificial intelligence (AI) and machine learning (ML) methods for short-range forecasts (<14 d), the European Centre for Medium-Range Weather Forecasts, with support from the World Meteorological Organisation, has launched the AI Weather Quest: an international, open and scientifically rigorous research competition designed to evaluate AI/ ML methods for sub-seasonal forecasting. Targeting participants from a diverse range of organisations including academia, operational forecasting centres, and technology companies, the AI Weather Quest is a pioneering large-scale, multidisciplinary initiative aimed at advancing reliable sub-seasonal predictions through AI and ML. Competing teams submit weekly probabilistic forecasts of two-metre temperature, accumulated precipitation, and mean sea level pressure. Submissions are made in real time, allowing models to be evaluated under operational-like conditions. To foster a competitive environment, an evaluation system has been established that calculates global ranked probability skill scores and displays team performance on live leaderboards. Leading teams will be spotlighted quarterly and yearly, recognising both top-performing teams and innovative methodological contributions. Detailed information regarding registration, forecast submission, and the dedicated tools provided are available on the AI Weather Quest website.
Source: Machine Learning Earth
Institution: European Centre for Medium-Range Weather Forecasts
Language: English
Subfield: Atmospheric Science
Field: Earth and Planetary Sciences
Domain: Physical Sciences
SDG: Climate action
Open Access status: gold
Estimated APC paid: $3,125
Funder: European Commission
OpenAlex ID: W4413358013