A new artificial intelligence weather forecasting model developed by Google DeepMind has demonstrated superior accuracy 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 forecasts up to 10 days ahead in under one minute on a single cloud-based computer. In head-to-head tests with the European Centre for Medium-Range Weather Forecasts’ high-resolution forecasting system (HRES), GraphCast provided more accurate predictions for over 90% of 1,380 test variables, including critical atmospheric measurements like geopotential, temperature, and humidity. The AI system excels at predicting severe weather events, showing particular skill in forecasting tropical cyclone tracks and atmospheric rivers. Researchers emphasize that AI models like GraphCast are designed to complement rather than replace traditional numerical weather prediction systems, offering a faster, more computationally efficient alternative for certain applications. The model’s code has been made open-source to encourage further development and integration into global weather forecasting infrastructure. Read the full article: https://sciencedaily.com/releases/2023/11/231114155020.htm