A new artificial intelligence weather forecasting model developed by Google DeepMind has demonstrated superior accuracy compared to conventional physics-based systems in global weather prediction. The GraphCast model, detailed in a Science journal study, uses machine learning to analyze decades of historical weather data, learning complex atmospheric patterns to generate 10-day forecasts in under one minute. Researchers report that GraphCast outperformed the European Centre for Medium-Range Weather Forecasts’ high-resolution system (HRES) in over 90% of 1,380 test metrics, including critical variables like temperature, pressure, wind speed, and humidity. The AI model particularly excelled at predicting extreme weather events, such as tropical cyclones and atmospheric rivers, with greater lead time and precision. While traditional numerical weather prediction models rely on supercomputers solving complex physical equations over hours, GraphCast generates forecasts faster using less computational power. Experts note that AI models like GraphCast represent a complementary tool rather than a replacement for established systems, with potential to enhance early warning systems and disaster preparedness globally. Read the full article at: https://sciencedaily.com/releases/2023/11/231114143522.htm