Accurate medium-range global weather forecasting with 3D neural networks
Huawei Cloud, Shenzhen, China
Nature volume 619, pages 533–538 (2023)
https://www.nature.com/articles/s41586-023-06185-3
Discussion
In this paper, we present Pangu-Weather, an AI-based system that trains deep networks for fast and accurate numerical weather forecasting. The major technical contributions include the design of the 3DEST architecture and the application of the hierarchical temporal aggregation strategy for medium-range forecasting. By training the models on 39 years of global weather data, Pangu-Weather produces better deterministic forecast results on reanalysis data than the world’s best NWP system, the operational IFS of ECMWF, while also being much faster. In addition, Pangu-Weather is excellent at forecasting extreme weather events and performing ensemble weather forecasts. Pangu-Weather reveals the potential of using large pre-trained models for various downstream applications, showing the same trend as other AI scopes, such as computer vision26,27, natural language processing28,29, cross-modal understanding30 and beyond.
