@misc{keisler_forecasting_2022,
	title = {Forecasting {Global} {Weather} with {Graph} {Neural} {Networks}},
	copyright = {Creative Commons Attribution 4.0 International},
	url = {http://arxiv.org/abs/2202.07575},
	doi = {10.48550/arXiv.2202.07575},
	abstract = {We present a data-driven approach for forecasting global weather using graph neural networks. The system learns to step forward the current 3D atmospheric state by six hours, and multiple steps are chained together to produce skillful forecasts going out several days into the future. The underlying model is trained on reanalysis data from ERA5 or forecast data from GFS. Test performance on metrics such as Z500 (geopotential height) and T850 (temperature) improves upon previous data-driven approaches and is comparable to operational, full-resolution, physical models from GFS and ECMWF, at least when evaluated on 1-degree scales and when using reanalysis initial conditions. We also show results from connecting this data-driven model to live, operational forecasts from GFS.},
	urldate = {2023-02-16},
	publisher = {arXiv},
	author = {Keisler, Ryan},
	month = feb,
	year = {2022},
	note = {arXiv:2202.07575 [physics]},
	keywords = {Atmospheric and Oceanic Physics (physics.ao-ph), Computer Science - Machine Learning, FOS: Computer and information sciences, FOS: Physical sciences, GNN, Graph Neural Networks, Machine Learning (cs.LG), NN, Neural Networks, Physics - Atmospheric and Oceanic Physics},
}

{"_id":"knnYgeqsm4WQMtw7k","bibbaseid":"keisler-forecastingglobalweatherwithgraphneuralnetworks-2022","author_short":["Keisler, R."],"bibdata":{"bibtype":"misc","type":"misc","title":"Forecasting Global Weather with Graph Neural Networks","copyright":"Creative Commons Attribution 4.0 International","url":"http://arxiv.org/abs/2202.07575","doi":"10.48550/arXiv.2202.07575","abstract":"We present a data-driven approach for forecasting global weather using graph neural networks. The system learns to step forward the current 3D atmospheric state by six hours, and multiple steps are chained together to produce skillful forecasts going out several days into the future. The underlying model is trained on reanalysis data from ERA5 or forecast data from GFS. Test performance on metrics such as Z500 (geopotential height) and T850 (temperature) improves upon previous data-driven approaches and is comparable to operational, full-resolution, physical models from GFS and ECMWF, at least when evaluated on 1-degree scales and when using reanalysis initial conditions. We also show results from connecting this data-driven model to live, operational forecasts from GFS.","urldate":"2023-02-16","publisher":"arXiv","author":[{"propositions":[],"lastnames":["Keisler"],"firstnames":["Ryan"],"suffixes":[]}],"month":"February","year":"2022","note":"arXiv:2202.07575 [physics]","keywords":"Atmospheric and Oceanic Physics (physics.ao-ph), Computer Science - Machine Learning, FOS: Computer and information sciences, FOS: Physical sciences, GNN, Graph Neural Networks, Machine Learning (cs.LG), NN, Neural Networks, Physics - Atmospheric and Oceanic Physics","bibtex":"@misc{keisler_forecasting_2022,\n\ttitle = {Forecasting {Global} {Weather} with {Graph} {Neural} {Networks}},\n\tcopyright = {Creative Commons Attribution 4.0 International},\n\turl = {http://arxiv.org/abs/2202.07575},\n\tdoi = {10.48550/arXiv.2202.07575},\n\tabstract = {We present a data-driven approach for forecasting global weather using graph neural networks. The system learns to step forward the current 3D atmospheric state by six hours, and multiple steps are chained together to produce skillful forecasts going out several days into the future. The underlying model is trained on reanalysis data from ERA5 or forecast data from GFS. Test performance on metrics such as Z500 (geopotential height) and T850 (temperature) improves upon previous data-driven approaches and is comparable to operational, full-resolution, physical models from GFS and ECMWF, at least when evaluated on 1-degree scales and when using reanalysis initial conditions. We also show results from connecting this data-driven model to live, operational forecasts from GFS.},\n\turldate = {2023-02-16},\n\tpublisher = {arXiv},\n\tauthor = {Keisler, Ryan},\n\tmonth = feb,\n\tyear = {2022},\n\tnote = {arXiv:2202.07575 [physics]},\n\tkeywords = {Atmospheric and Oceanic Physics (physics.ao-ph), Computer Science - Machine Learning, FOS: Computer and information sciences, FOS: Physical sciences, GNN, Graph Neural Networks, Machine Learning (cs.LG), NN, Neural Networks, Physics - Atmospheric and Oceanic Physics},\n}\n\n","author_short":["Keisler, R."],"key":"keisler_forecasting_2022","id":"keisler_forecasting_2022","bibbaseid":"keisler-forecastingglobalweatherwithgraphneuralnetworks-2022","role":"author","urls":{"Paper":"http://arxiv.org/abs/2202.07575"},"keyword":["Atmospheric and Oceanic Physics (physics.ao-ph)","Computer Science - Machine Learning","FOS: Computer and information sciences","FOS: Physical sciences","GNN","Graph Neural Networks","Machine Learning (cs.LG)","NN","Neural Networks","Physics - Atmospheric and Oceanic Physics"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"misc","biburl":"https://bibbase.org/zotero/iagogv","dataSources":["ovn29uG6Mbp3JWCRR"],"keywords":["atmospheric and oceanic physics (physics.ao-ph)","computer science - machine learning","fos: computer and information sciences","fos: physical sciences","gnn","graph neural networks","machine learning (cs.lg)","nn","neural networks","physics - atmospheric and oceanic physics"],"search_terms":["forecasting","global","weather","graph","neural","networks","keisler"],"title":"Forecasting Global Weather with Graph Neural Networks","year":2022}