A new study published in Science demonstrates that artificial intelligence weather prediction models can now match or exceed the accuracy of conventional physics-based models. The research, led by a team from the University of California, Berkeley, evaluated several AI systems, including Google’s GraphCast and Huawei’s Pangu-Weather, over a year-long period. The AI models proved particularly adept at predicting severe weather events like hurricanes and atmospheric rivers several days in advance, often with greater precision and at a fraction of the computational cost. While traditional models still hold advantages in certain areas, such as long-range seasonal forecasting, the findings signal a significant shift in meteorological science. The rapid advancement suggests AI will become a core component of operational weather forecasting worldwide. Read the full article at: https://sciencedaily.com/releases/2024/10/241024163456.htm