Global S&T Development Trend Analysis Platform of Resources and Environment
DOI | 10.1073/pnas.1810286115 |
Deep learning to represent subgrid processes in climate models | |
Rasp, Stephan1,2; Pritchard, Michael S.2; Gentine, Pierre3,4 | |
2018-09-25 | |
发表期刊 | PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA |
ISSN | 0027-8424 |
出版年 | 2018 |
卷号 | 115期号:39页码:9684-9689 |
文章类型 | Article |
语种 | 英语 |
国家 | Germany; USA |
英文摘要 | The representation of nonlinear subgrid processes, especially clouds, has been a major source of uncertainty in climate models for decades. Cloud-resolving models better represent many of these processes and can now be run globally but only for shortterm simulations of at most a few years because of computational limitations. Here we demonstrate that deep learning can be used to capture many advantages of cloud-resolving modeling at a fraction of the computational cost. We train a deep neural network to represent all atmospheric subgrid processes in a climate model by learning from a multiscale model in which convection is treated explicitly. The trained neural network then replaces the traditional subgrid parameterizations in a global general circulation model in which it freely interacts with the resolved dynamics and the surface-flux scheme. The prognostic multiyear simulations are stable and closely reproduce not only the mean climate of the cloud-resolving simulation but also key aspects of variability, including precipitation extremes and the equatorial wave spectrum. Furthermore, the neural network approximately conserves energy despite not being explicitly instructed to. Finally, we show that the neural network parameterization generalizes to new surface forcing patterns but struggles to cope with temperatures far outside its training manifold. Our results show the feasibility of using deep learning for climate model parameterization. In a broader context, we anticipate that data-driven Earth system model development could play a key role in reducing climate prediction uncertainty in the coming decade. |
英文关键词 | climate modeling deep learning subgrid parameterization convection |
领域 | 地球科学 ; 气候变化 ; 资源环境 |
收录类别 | SCI-E |
WOS记录号 | WOS:000445545200040 |
WOS关键词 | RESOLUTION DEPENDENCE ; RESOLVING MODEL ; CLOUDS ; PARAMETERIZATION ; PRECIPITATION ; CIRCULATION ; SENSITIVITY ; ATMOSPHERE |
WOS类目 | Multidisciplinary Sciences |
WOS研究方向 | Science & Technology - Other Topics |
URL | 查看原文 |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | http://119.78.100.173/C666/handle/2XK7JSWQ/204990 |
专题 | 地球科学 资源环境科学 气候变化 |
作者单位 | 1.Ludwig Maximilians Univ Munchen, Meteorol Inst, D-80333 Munich, Germany; 2.Univ Calif Irvine, Dept Earth Syst Sci, Irvine, CA 92697 USA; 3.Columbia Univ, Earth Inst, Dept Earth & Environm Engn, New York, NY 10027 USA; 4.Columbia Univ, Data Sci Inst, New York, NY 10027 USA |
推荐引用方式 GB/T 7714 | Rasp, Stephan,Pritchard, Michael S.,Gentine, Pierre. Deep learning to represent subgrid processes in climate models[J]. PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA,2018,115(39):9684-9689. |
APA | Rasp, Stephan,Pritchard, Michael S.,&Gentine, Pierre.(2018).Deep learning to represent subgrid processes in climate models.PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA,115(39),9684-9689. |
MLA | Rasp, Stephan,et al."Deep learning to represent subgrid processes in climate models".PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA 115.39(2018):9684-9689. |
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。
修改评论