%0 Journal Article %1 Bi2023 %A Bi, Kaifeng %A Xie, Lingxi %A Zhang, Hengheng %A Chen, Xin %A Gu, Xiaotao %A Tian, Qi %D 2023 %J Nature %K idea:big_data_geo_2 idea:rem pangu todo:read %N 7970 %P 533--538 %R 10.1038/s41586-023-06185-3 %T Accurate medium-range global weather forecasting with 3D neural networks %U https://doi.org/10.1038/s41586-023-06185-3 %V 619 %X Weather forecasting is important for science and society. At present, the most accurate forecast system is the numerical weather prediction (NWP) method, which represents atmospheric states as discretized grids and numerically solves partial differential equations that describe the transition between those states1. However, this procedure is computationally expensive. Recently, artificial-intelligence-based methods2 have shown potential in accelerating weather forecasting by orders of magnitude, but the forecast accuracy is still significantly lower than that of NWP methods. Here we introduce an artificial-intelligence-based method for accurate, medium-range global weather forecasting. We show that three-dimensional deep networks equipped with Earth-specific priors are effective at dealing with complex patterns in weather data, and that a hierarchical temporal aggregation strategy reduces accumulation errors in medium-range forecasting. Trained on 39þinspaceyears of global data, our program, Pangu-Weather, obtains stronger deterministic forecast results on reanalysis data in all tested variables when compared with the world's best NWP system, the operational integrated forecasting system of the European Centre for Medium-Range Weather Forecasts (ECMWF)3. Our method also works well with extreme weather forecasts and ensemble forecasts. When initialized with reanalysis data, the accuracy of tracking tropical cyclones is also higher than that of ECMWF-HRES.

@article{Bi2023, abstract = {Weather forecasting is important for science and society. At present, the most accurate forecast system is the numerical weather prediction (NWP) method, which represents atmospheric states as discretized grids and numerically solves partial differential equations that describe the transition between those states1. However, this procedure is computationally expensive. Recently, artificial-intelligence-based methods2 have shown potential in accelerating weather forecasting by orders of magnitude, but the forecast accuracy is still significantly lower than that of NWP methods. Here we introduce an artificial-intelligence-based method for accurate, medium-range global weather forecasting. We show that three-dimensional deep networks equipped with Earth-specific priors are effective at dealing with complex patterns in weather data, and that a hierarchical temporal aggregation strategy reduces accumulation errors in medium-range forecasting. Trained on 39{\thinspace}years of global data, our program, Pangu-Weather, obtains stronger deterministic forecast results on reanalysis data in all tested variables when compared with the world's best NWP system, the operational integrated forecasting system of the European Centre for Medium-Range Weather Forecasts (ECMWF)3. Our method also works well with extreme weather forecasts and ensemble forecasts. When initialized with reanalysis data, the accuracy of tracking tropical cyclones is also higher than that of ECMWF-HRES.}, added-at = {2023-12-12T13:05:50.000+0100}, author = {Bi, Kaifeng and Xie, Lingxi and Zhang, Hengheng and Chen, Xin and Gu, Xiaotao and Tian, Qi}, biburl = {https://www.bibsonomy.org/bibtex/2c9fe5ca62abdc6b5e58605a150d28ae1/annakrause}, day = 01, description = {Accurate medium-range global weather forecasting with 3D neural networks | Nature}, doi = {10.1038/s41586-023-06185-3}, interhash = {cd125f655e3313e81d0abf69610e73df}, intrahash = {c9fe5ca62abdc6b5e58605a150d28ae1}, issn = {1476-4687}, journal = {Nature}, keywords = {idea:big_data_geo_2 idea:rem pangu todo:read}, month = jul, number = 7970, pages = {533--538}, timestamp = {2023-12-14T14:31:37.000+0100}, title = {Accurate medium-range global weather forecasting with 3D neural networks}, url = {https://doi.org/10.1038/s41586-023-06185-3}, volume = 619, year = 2023 }