A new artificial intelligence weather forecasting model developed by Google DeepMind, named GraphCast, has demonstrated superior accuracy and speed compared to the world’s leading traditional system in a comprehensive global evaluation. The study, published in the journal Science, shows GraphCast significantly outperformed the European Centre for Medium-Range Weather Forecasts (ECMWF) high-resolution forecast (HRES) system on over 90% of 1,380 test variables and forecast lead times. The AI model, which can generate a 10-day forecast in under one minute on a single Google TPU v4 cloud computer, excels at predicting severe weather events, including tropical cyclones and atmospheric rivers, with greater precision days in advance. Researchers emphasize that AI models like GraphCast are not intended to replace but to complement established physics-based systems, potentially leading to more reliable and accessible weather predictions worldwide. For the full details, read the complete article at https://sciencedaily.com/releases/2024/04/240411125735.htm