Use of neural networks for stable, accurate and physically consistent parameterization of subgrid atmospheric processes with good performance at reduced precision
A promising approach to improve climate-model simulations is to replace traditional subgrid parameterizations based on simplified physical models by machine learning algorithms that are data-driven. However, neural networks (NNs) often lead to instabilities and climate drift when coupled to an atmos...
Main Authors: | Yuval, Janni, O'Gorman, Paul A, Hill, Chris N |
---|---|
Other Authors: | Massachusetts Institute of Technology. Department of Earth, Atmospheric, and Planetary Sciences |
Format: | Article |
Language: | English |
Published: |
American Geophysical Union (AGU)
2021
|
Online Access: | https://hdl.handle.net/1721.1/135562 |
Similar Items
-
Neural‐Network Parameterization of Subgrid Momentum Transport in the Atmosphere
by: Janni Yuval, et al.
Published: (2023-04-01) -
Non‐Local Parameterization of Atmospheric Subgrid Processes With Neural Networks
by: Wang, Peidong, et al.
Published: (2023) -
Stable machine-learning parameterization of subgrid processes for climate modeling at a range of resolutions
by: Yuval, Janni, et al.
Published: (2020) -
Parameterization of subgrid plume dilution for use in large-scale atmospheric simulations
by: A. D. Naiman, et al.
Published: (2010-03-01) -
Applications of deep learning to ocean data inference and subgrid parameterization
by: Bolton, T, et al.
Published: (2019)