A new artificial intelligence weather forecasting model developed by Google DeepMind, called GraphCast, has demonstrated superior accuracy over the world’s leading traditional numerical weather prediction system. The AI model, described in a Science journal paper, produces 10-day weather predictions in under a minute on a single machine, compared to the hours required by conventional supercomputer-run systems. In a comprehensive evaluation against the European Centre for Medium-Range Weather Forecasts’ high-resolution forecast (HRES), GraphCast was more accurate in over 90% of 1,380 test variables, including critical predictions for severe weather events like tropical cyclones and atmospheric rivers. The model was trained on nearly four decades of historical weather data to learn cause-and-effect relationships within the Earth’s climate system. While not intended to replace current operational systems, researchers highlight its potential as a powerful complementary tool that could enhance early warning systems for extreme weather. The model’s code has been made publicly available to support further scientific advancement. Read the full article at: https://sciencedaily.com/releases/2023/11/231114143522.htm