A new artificial intelligence system called GraphCast, developed by Google DeepMind, has demonstrated superior accuracy in medium-range weather forecasting compared to the world’s leading traditional system. The AI model, described in a paper published in Science, can generate a 10-day forecast in under a minute on a single machine, a task that typically requires hours of computation on a supercomputer for conventional physics-based models. GraphCast was trained on nearly four decades of historical weather data and outperformed the European Centre for Medium-Range Weather Forecasts’ high-resolution forecast (HRES) system on more than 90% of over 1,300 test variables. The system excels at predicting severe weather events, including tropical cyclones and extreme temperatures, with greater precision and lead time. While not intended to replace existing systems, it is designed to complement them, offering a faster, more efficient tool for meteorologists. The model’s code has been made publicly available to aid the broader weather prediction community. Read the full article at: https://sciencedaily.com/releases/2023/11/231114143522.htm