A new artificial intelligence weather forecasting model, developed by researchers at the University of California, has demonstrated superior accuracy in global weather predictions compared to established numerical models. The system, named Aurora, was trained on over a million hours of diverse weather and climate simulations, enabling it to generate a 10-day global forecast in under a minute on a single graphics processing unit. In benchmark tests, Aurora consistently matched or exceeded the accuracy of the European Centre for Medium-Range Weather Forecasts’ high-resolution model, particularly in predicting extreme weather events like hurricanes and atmospheric rivers. The researchers highlight the model’s efficiency and scalability as a significant advancement, though they note it is designed to complement rather than replace traditional physics-based systems. The full study detailing the model’s architecture and performance is available in the journal Science. Read the full article at https://sciencedaily.com/releases/2024/05/240523130426.htm