A new study published in Science demonstrates that an artificial intelligence weather forecasting model, Pangu-Weather, has achieved superior accuracy and speed compared to conventional numerical weather prediction methods. Developed by researchers at Huawei Cloud, the model was trained on 39 years of global weather data and can generate a 7-day forecast in seconds on a single GPU, a task that typically requires hours on supercomputers using traditional physics-based models. The AI system showed particular strength in predicting extreme weather events like typhoons and heatwaves, though experts note it still relies on initial data from traditional observations and may have limitations in predicting unprecedented events. The research highlights the growing role of machine learning in meteorology, potentially enabling faster and more accessible high-resolution forecasts. Read the full article at: https://sciencedaily.com/releases/2023/07/230705143123.htm