A new artificial intelligence weather forecasting model developed by Google DeepMind, called GraphCast, has demonstrated superior performance compared to the world’s leading traditional numerical weather prediction system. In a comprehensive evaluation, GraphCast provided more accurate 10-day forecasts for over 90% of 1,380 test variables, including critical atmospheric conditions like temperature, pressure, and humidity. The AI model operates significantly faster, generating a 10-day forecast in under a minute on a single Google TPU, whereas conventional systems can take hours on extensive supercomputers. GraphCast was trained on nearly four decades of historical weather data from the European Centre for Medium-Range Weather Forecasts (ECMWF). While promising, meteorologists emphasize that AI models like GraphCast should complement, not replace, established physics-based systems, as they rely on the high-quality data produced by traditional methods. The research was published in the journal Science. Read the full article at: https://sciencedaily.com/releases/2023/11/231114143522.htm