A new artificial intelligence weather forecasting model developed by Google DeepMind has demonstrated superior performance compared to traditional physics-based systems in global medium-range weather predictions. The GraphCast model, detailed in a study published in Science, uses machine learning to analyze decades of historical weather data and generate 10-day forecasts in under a minute. Researchers report that GraphCast outperformed the European Centre for Medium-Range Weather Forecasts’ high-resolution forecast system (HRES) on more than 90% of 1,380 test variables, including critical measurements of temperature, pressure, wind speed, and humidity. The AI system showed particular strength in predicting severe weather events, accurately forecasting the path of Hurricane Lee nine days in advance, three days earlier than conventional models. While the technology shows promise for faster, more efficient forecasting, experts note that AI models still rely on data from traditional systems and face challenges in predicting rare or unprecedented weather phenomena. The development represents a significant shift toward data-driven approaches in meteorology, though operational integration with established forecasting centers remains a future goal. Read the full article at: https://sciencedaily.com/releases/2023/11/231114143522.htm