A new artificial intelligence weather forecasting model developed by Google DeepMind has demonstrated superior accuracy to traditional numerical weather prediction systems in global medium-range forecasts. The GraphCast model, detailed in a study published in Science, uses machine learning trained on decades of historical weather data to predict hundreds of weather variables up to 10 days in advance. In head-to-head comparisons with the European Centre for Medium-Range Weather Forecasts’ high-resolution forecast system (HRES), GraphCast provided more accurate predictions for over 90% of 1,380 test variables. The AI system was particularly skilled at predicting severe weather events, accurately forecasting the path of Hurricane Lee in September 2023 nine days in advance, three days earlier than traditional approaches. Researchers emphasize that AI models like GraphCast should complement rather than replace conventional physics-based systems, offering faster and potentially more efficient forecasting capabilities. Read the full article at https://sciencedaily.com/releases/2023/11/231114155020.htm