Climate Change Data Portal
DOI | 10.1029/2019MS001896 |
Machine Learning for Stochastic Parameterization: Generative Adversarial Networks in the Lorenz '96 Model | |
Gagne D.J.; II; Christensen H.M.; Subramanian A.C.; Monahan A.H. | |
发表日期 | 2020 |
ISSN | 19422466 |
卷号 | 12期号:3 |
英文摘要 | Stochastic parameterizations account for uncertainty in the representation of unresolved subgrid processes by sampling from the distribution of possible subgrid forcings. Some existing stochastic parameterizations utilize data-driven approaches to characterize uncertainty, but these approaches require significant structural assumptions that can limit their scalability. Machine learning models, including neural networks, are able to represent a wide range of distributions and build optimized mappings between a large number of inputs and subgrid forcings. Recent research on machine learning parameterizations has focused only on deterministic parameterizations. In this study, we develop a stochastic parameterization using the generative adversarial network (GAN) machine learning framework. The GAN stochastic parameterization is trained and evaluated on output from the Lorenz '96 model, which is a common baseline model for evaluating both parameterization and data assimilation techniques. We evaluate different ways of characterizing the input noise for the model and perform model runs with the GAN parameterization at weather and climate time scales. Some of the GAN configurations perform better than a baseline bespoke parameterization at both time scales, and the networks closely reproduce the spatiotemporal correlations and regimes of the Lorenz '96 system. We also find that, in general, those models which produce skillful forecasts are also associated with the best climate simulations. ©2020. The Authors. |
英文关键词 | climate; generative adversarial networks; lorenz; machine learning; stochastic parameterization; weather |
语种 | 英语 |
scopus关键词 | Climate models; Machine learning; Parameterization; Stochastic systems; Adversarial networks; Climate simulation; Data assimilation techniques; Data-driven approach; Machine learning models; Recent researches; Spatiotemporal correlation; Structural assumption; Stochastic models; correlation; data assimilation; detection method; machine learning; mapping method; model; parameterization; sampling; simulation; stochasticity; uncertainty analysis |
来源期刊 | Journal of Advances in Modeling Earth Systems
![]() |
文献类型 | 期刊论文 |
条目标识符 | http://gcip.llas.ac.cn/handle/2XKMVOVA/156741 |
作者单位 | National Center for Atmospheric Research, Boulder, CO, United States; Atmospheric, Oceanic and Planetary Physics, University of Oxford, Oxford, United Kingdom; Department of Atmospheric and Oceanic Sciences, University of Colorado, Boulder, CO, United States; School of Earth and Ocean Sciences, University of Victoria, Victoria, BC, Canada |
推荐引用方式 GB/T 7714 | Gagne D.J.,II,Christensen H.M.,et al. Machine Learning for Stochastic Parameterization: Generative Adversarial Networks in the Lorenz '96 Model[J],2020,12(3). |
APA | Gagne D.J.,II,Christensen H.M.,Subramanian A.C.,&Monahan A.H..(2020).Machine Learning for Stochastic Parameterization: Generative Adversarial Networks in the Lorenz '96 Model.Journal of Advances in Modeling Earth Systems,12(3). |
MLA | Gagne D.J.,et al."Machine Learning for Stochastic Parameterization: Generative Adversarial Networks in the Lorenz '96 Model".Journal of Advances in Modeling Earth Systems 12.3(2020). |
条目包含的文件 | 条目无相关文件。 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。