Europe's AI-Powered Weather Forecasting System Promises Improved Accuracy and Efficiency

Weather forecasting takes big step forward with Europe's new AI system

Improved weather prediction using artificial intelligence promises to take a big step forward with the launch of a new European system, which can outperform conventional forecasting methods for up to 15 days ahead. While tech companies and meteorological offices around the world are already applying AI to the weather, the European Centre for Medium-range Weather Forecasts (ECMWF) said its operational model broke new ground by making global predictions freely available to everyone at any time. "This milestone will transform weather science and predictions," said Florence Rabier, director-general of ECMWF, an intergovernmental organisation. "Making the AI Forecasting System operational produces the widest range of parameters using machine learning available to date." An experimental version tested over the past 18 months showed the system was about 20 per cent more accurate on key predictions than the best conventional methods, which feed millions of worldwide weather observations into supercomputers and crunch them with physics-based equations. The new European system could predict the track of a tropical cyclone 12 hours further ahead, giving valuable extra warning time for severe events, said Florian Pappenberger, ECMWF director of forecasts. The world experienced its hottest temperatures on record in 2024, and Europe has become the fastest warming continent, triggering extreme weather events. The agency has been at the forefront of observations and public awareness about the effects of climate change. Other medium-range AI forecasting systems under development include GenCast and GraphCast from Google DeepMind, Pangu-Weather from Huawei, FourCastNet from Nvidia and FuXi from Shanghai Academy of AI for Science and Fudan University. All were trained on a database of weather observations compiled by the ECMWF over 40 years. Comparing the accuracy of competing AI forecasting systems was hard, Pappenberger said, because their relative performance differed according to the variables and timescales assessed. Scores published by the ECMWF give some idea of performance but do not identify an overall champion. But Pappenberger noted that its system stood out for predicting many more features than standard temperature, precipitation and wind. For example, it also forecasts solar radiation and wind speeds at 100 metres -- the height of a typical turbine -- helpful for the renewable energy sector. Although ECMWF forecasts are freely available, the agency does not issue severe weather alerts nor tailor-made predictions to industry users, leaving the specialised forecasts to national or local authorities and private companies. ECMWF and a group of European national met offices have created an open-source technical framework for AI weather systems called Anemoi, after the Greek god of the winds. The underlying machine-learning architecture is based on the same "graph neural network" as Google DeepMind's forecasting models. Peter Battaglia, research director at DeepMind, said it was "impressive" to see how the ECMWF had adapted to the AI wave that had reshaped the field in recent years, and the latest open model would add to the pool of knowledge. The ECMWF plans to improve its system further by increasing its spatial resolution and moving from the present version, which generates one forecast at a time, to "ensemble forecasting" -- or creating a collection of 50 forecasts simultaneously with slightly different starting conditions to provide a range of possible outcomes. In the future, said Kirstine Dale, chief AI officer at the UK Met Office, a mix of physics-based and data-based simulations would be needed for "their combined strengths to provide accurate, fast, reliable and trustworthy forecasts". Today the boundaries of reliable day-to-day weather forecasts in Europe are six to seven days ahead for precipitation and wind, and up to 14 or 15 days for temperature, said Pappenberger. "Machine learning models have a fair chance of extending that because they may be able to extract something out of the data that we may not currently represent well enough in physics-based models."

Europe’s New AI Weather Model Is Faster, Smarter, and Freeâ€"Here’s What to Know

The European Center for Medium-Range Weather Forecasts has debuted its own forecast AI model on the heels of a cutting-edge Google model released in December. The European Center for Medium-Range Weather Forecasts (ECMWF) just launched an AI-powered forecasting model, which the center says outperforms state-of-the-art physics-based models by up to 20%. The model is dubbed the Artificial Intelligence Forecasting System (AIFS). According to an ECMWF release, the new model operates at faster speeds than physics-based models and takes approximately 1,000 times less energy to make a forecast. The ECMWF, now in its 50th year of operation, produced ENS, one of the world's leading medium-range weather prediction models. Medium-range forecasting includes weather predictions made between three days and 15 days in advance, but ECMWF also forecasts weather up to a year ahead. Weather forecast models are essential for states and local governments to stay prepared for extreme weather eventsâ€"as well as for more daily needs, like knowing what the weather will be like on your upcoming vacation. Traditional weather prediction models make forecasts by solving physics equations. A limitation of these models is that they are approximations of atmospheric dynamics. A compelling aspect of AI-driven models is that they could learn more complex relationships and dynamics in weather patterns directly from the data, rather than relying only on previously known and documented equations. The ECMWF's announcement comes on the heels of Google DeepMind's GenCast model for AI-powered weather prediction, the next iteration of Google's weather prediction software that includes NeuralGCM and GraphCast. GenCast outperformed ENS, the ECMWF's leading weather prediction model, on 97.2% of targets across different weather variables. With lead times greater than 36 hours, GenCast was more accurate than ENS on 99.8% of targets. But the European Center is innovating, too. The launch of AIFS-single is just the first operational version of the system. "This is a huge endeavour that ensures the models are running in a stable and reliable way," said Florian Pappenberger, Director of Forecasts and Services at ECMWF, in the center release. "At the moment, the resolution of the AIFS is less than that of our model (IFS), which achieves 9 km [5.6-mile] resolution using a physics-based approach." "We see the AIFS and IFS as complementary, and part of providing a range of products to our user community, who decide what best suits their needs," Pappenberger added. The team will explore hybridizing data-driven and physics-based modeling to improve the organization's ability to predict weather with precision. "Physics-based models are key to the current data-assimilation process," said Matthew Chantry, Strategic Lead for Machine Learning at ECMWF and Head of the Innovation Platform, in an email to Gizmodo. "This same data-assimilation process is also vital to initialize every day machine learning models, and allow them to make forecasts." "One of the next frontiers for machine learning weather forecasting is this data-assimilation step, which if solved would mean that the full weather forecasting chain could be based on machine learning," Chantry added. Chantry is a co-author of a study awaiting peer review that describes a data-driven, end-to-end forecast system that does not rely on physics-based reanalysis. Called GraphDOP, the system uses observable quantities such as brightness temperatures from polar orbiters "to form a coherent latent representation of Earth System state dynamics and physical processes," the team wrote, "and is capable of producing skillful predictions of relevant weather parameters up to five days into the future." Integrating artificial intelligence methods with physics-driven weather prediction modeling is a promising venue for more precise forecasting. Testing to date indicates that AI-powered forecasting can outperform historical models, but so far those models have relied on reanalysis data. Observations on the ground were essential for training the models, and it remains to be seen just how impressive the technology's forecasting abilities will be when it's forced to go off-script.