Deep Learning to Represent Subgrid Processes in Climate Models

TitleDeep Learning to Represent Subgrid Processes in Climate Models
Publication TypeJournal Article
Year of Publication2018
JournalProceedings of the National Academy of Sciences
Volume115
Number39
Pages9684-9689
Date Published11/2018
Abstract / Summary

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 short-term 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.

URLhttp://dx.doi.org/10.1073/pnas.1810286115
DOI10.1073/pnas.1810286115
Journal: Proceedings of the National Academy of Sciences
Year of Publication: 2018
Volume: 115
Number: 39
Pages: 9684-9689
Date Published: 11/2018

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 short-term 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.

DOI: 10.1073/pnas.1810286115
Citation:
Rasp, S, M Pritchard, and P Gentine.  2018.  "Deep Learning to Represent Subgrid Processes in Climate Models."  Proceedings of the National Academy of Sciences 115(39): 9684-9689.  https://doi.org/10.1073/pnas.1810286115.