CoChat
Sign Up
Cited by 283Meteorological Phenomena and SimulationsClimate variability and modelsOceanographic and Atmospheric Processes

Stable machine-learning parameterization of subgrid processes for climate modeling at a range of resolutions

Open access
Yuval, Janni, O’Gorman, Paul A.·2020·Nature Communications
DOI 10.1038/s41467-020-17142-3PMC PMC7335176
Open this paper in CoChat — free accountOpen PDF

Abstract

Global climate models represent small-scale processes such as convection using subgrid models known as parameterizations, and these parameterizations contribute substantially to uncertainty in climate projections. Machine learning of new parameterizations from high-resolution model output is a promising approach, but such parameterizations have been prone to issues of instability and climate drift, and their performance for different grid spacings has not yet been investigated. Here we use a random forest to learn a parameterization from coarse-grained output of a three-dimensional high-resolution idealized atmospheric model. The parameterization leads to stable simulations at coarse resolution that replicate the climate of the high-resolution simulation. Retraining for different coarse-graining factors shows the parameterization performs best at smaller horizontal grid spacings. Our results yield insights into parameterization performance across length scales, and they also demonstrate the potential for learning parameterizations from global high-resolution simulations that are now emerging.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41467-020-17142-3

Cites

3 of 3 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • So far, hybrid models have mostly been limited to idealized scenarios such as aquaplanets.

    “So far, hybrid models have mostly been limited to idealized scenarios such as aquaplanets22,23.”

    From Neural general circulation models for weather and climate (2024) · cites this paper as [23]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 35

  1. [1]Webb MJ Lambert FH Gregory JM Origins of differences in climate sensitivity, forcing and feedback in climate models Clim. Dyn. 2013 40 677 707 10.1007/s00382-012-1336-x · DOI
  2. [2]Sherwood SC Bony S Dufresne J Spread in model climate sensitivity traced to atmospheric convective mixing Nature 2014 505 37 42 10.1038/nature12829 24380952 · DOI · Paper page
  3. [5]Ceppi P Hartmann DL Clouds and the atmospheric circulation response to warming J. Clim. 2016 29 783 799 10.1175/JCLI-D-15-0394.1 · DOI
  4. [6]Schneider T Climate goals and computing the future of clouds Nat. Clim. Change 2017 7 3 5 10.1038/nclimate3190 · DOI · Paper page
  5. [7]Bretherton CS Khairoutdinov MF Convective self-aggregation feedbacks in near-global cloud-resolving simulations of an aquaplanet J. Adv. Model. Earth Sys. 2015 7 1765 1787 10.1002/2015MS000499 · DOI
  6. [8]Stevens B DYAMOND: the DYnamics of the Atmospheric general circulation Modeled on Non-hydrostatic Domains Prog. Earth Planet. Sci. 2019 6 61 10.1186/s40645-019-0304-z · DOI · Paper page
  7. [9]Stouffer RJ Time scales of climate response J. Clim. 2004 17 209 217 10.1175/1520-0442(2004)017<0209:TSOCR>2.0.CO;2 · DOI
  8. [10]Krasnopolsky VM Fox-Rabinovitz MS Belochitski AA Using ensemble of neural networks to learn stochastic convection parameterizations for climate and numerical weather prediction models from data simulated by a cloud resolving model Adv. Artif. Neural Syst. 2013 2013 1 13 10.1155/2013/485913 · DOI
  9. [11]Gentine P Pritchard M Rasp S Reinaudi G Yacalis G Could machine learning break the convection parameterization deadlock? Geophys. Res. Lett. 2018 45 5742 5751 10.1029/2018GL078202 · DOI · Paper page
  10. [12]Rasp S Pritchard MS Gentine P Deep learning to represent subgrid processes in climate models Proc. Natl Acad. Sci. U.S.A. 2018 115 9684 9689 10.1073/pnas.1810286115 30190437 PMC6166853 · DOI · Paper page
  11. [13]O’Gorman PA Dwyer JG Using machine learning to parameterize moist convection: Potential for modeling of climate, climate change, and extreme events J. Adv. Model. Earth Syst. 2018 10 2548 2563 10.1029/2018MS001351 · DOI · Paper page
  12. [14]Brenowitz ND Bretherton CS Prognostic validation of a neural network unified physics parameterization Geophys. Res. Lett. 2018 45 6289 6298 10.1029/2018GL078510 · DOI
  13. [16]Bolton T Zanna L Applications of deep learning to ocean data inference and subgrid parameterization J. Adv. Model. Earth Syst. 2019 11 376 399 10.1029/2018MS001472 · DOI
  14. [17]Arakawa A The cumulus parameterization problem: past, present, and future J. Clim. 2004 17 2493 2525 10.1175/1520-0442(2004)017<2493:RATCPP>2.0.CO;2 · DOI
  15. [18]Arakawa A Wu C-M A unified representation of deep moist convection in numerical modeling of the atmosphere. part i J. Atmos. Sci. 2013 70 1977 1992 10.1175/JAS-D-12-0330.1 · DOI
  16. [19]Ahn M-S Kang I-S A practical approach to scale-adaptive deep convection in a GCM by controlling the cumulus base mass flux npj Clim. Atmos. Sci. 2018 1 1 8 10.1038/s41612-018-0021-0 · DOI
  17. [20]Pearson K Modelling the diurnal cycle of tropical convection across the ‘grey zone’ Q. J. R. Meteorol. Soc. 2014 140 491 499 10.1002/qj.2145 · DOI
  18. [23]Beucler, T., Rasp, S., Pritchard, M. & Gentine, P. Achieving conservation of energy in neural network emulators for climate modeling. preprint at https://arxiv.org/abs/1906.06622 (2019). · arXiv
  19. [24]Breiman L Random forests Mach. Learn. 2001 45 5 32 10.1023/A:1010933404324 · DOI · Paper page
  20. [25]Hastie, T., Tibshirani, R. & Friedman, J. The Elements of Statistical Learning 2nd edn (Springer, 2001)
  21. [26]Khairoutdinov MF Randall DA Cloud resolving modeling of the ARM summer 1997 IOP: Model formulation, results, uncertainties, and sensitivities J. Atmos. Sci. 2003 60 607 625 10.1175/1520-0469(2003)060<0607:CRMOTA>2.0.CO;2 · DOI
  22. [27]Neale RB Hoskins BJ A standard test for AGCMs including their physical parametrizations: I: The proposal Atmos. Sci. Lett. 2000 1 101 107 10.1006/asle.2000.0019 · DOI
  23. [28]Kuang Z Blossey PN Bretherton CS A new approach for 3D cloud-resolving simulations of large-scale atmospheric circulation Geophys. Res. Lett. 2005 32 L02809 10.1029/2004GL021024 · DOI
  24. [29]Garner ST Frierson DMW Held IM Pauluis O Vallis GK Resolving convection in a global hypohydrostatic model J. Atmos. Sci. 2007 64 2061 2075 10.1175/JAS3929.1 · DOI
  25. [30]Boos WR Fedorov A Muir L Convective self-aggregation and tropical cyclogenesis under the hypohydrostatic rescaling J. Atmos. Sci. 2016 73 525 544 10.1175/JAS-D-15-0049.1 · DOI
  26. [31]Fedorov AV Muir L Boos WR Studholme J Tropical cyclogenesis in warm climates simulated by a cloud-system resolving model Clim. Dyn. 2019 52 107 127 10.1007/s00382-018-4134-2 · DOI
  27. [32]Möbis B Stevens B Factors controlling the position of the intertropical convergence zone on an aquaplanet J. Adv. Model. Earth Syst. 2012 4 M00A04 10.1029/2012MS000199 · DOI
  28. [33]Grell GA A scale and aerosol aware stochastic convective parameterization for weather and air quality modeling Atmos. Chem. Phys. 2014 14 5233 5250 10.5194/acp-14-5233-2014 · DOI
  29. [34]Belochitski A Tree approximation of the long wave radiation parameterization in the NCAR CAM global climate model J. Computational Appl. Math. 2011 236 447 460 10.1016/j.cam.2011.07.013 · DOI
  30. [35]Kiehl JT The National Center for Atmospheric Research community climate model: CCM3 J. Clim. 1998 11 1131 1149 10.1175/1520-0442(1998)011<1131:TNCFAR>2.0.CO;2 · DOI
  31. [36]Ma D Boos W Kuang Z Effects of orography and surface heat fluxes on the south asian summer monsoon J. Clim. 2014 27 6647 6659 10.1175/JCLI-D-14-00138.1 · DOI
  32. [38]Geurts, P. Some enhancements of decision tree bagging. In European Conference on Principles of Data Mining and Knowledge Discovery (eds Zighed, D. A., Komorowski, J. & Żytkow, J.), 136–147, (Springer, 2000).
  33. [39]Bernard, S., Heutte, L. & Adam, S. On the selection of decision trees in random forests. IEEE International Joint Conference on Neural Networks (IJCNN), Jun 2008, Atlanta, United States. pp. 302–307, 10.1109/IJCNN.2009.5178693ff. · DOI
  34. [40]Painsky A Rosset S Lossless compression of random forests J. Comput. Sci. Technol. 2019 34 494 506 10.1007/s11390-019-1921-0 · DOI
  35. [41]Chen CT Knutson T On the verification and comparison of extreme rainfall indices from climate models J. Clim. 2008 21 1605 1621 10.1175/2007JCLI1494.1 · DOI · Paper page

Cited by

Sorted by citations

Authors
Yuval, J., & O’Gorman, P. A.
Affiliations
Massachusetts Institute of Technology (US)
Corresponding author
Yuval, Janni
ORCID
Venue
Nature Communications
Publisher
Springer Science and Business Media LLC
Published
2020-07-03
Volume
11
Issue
1
Pages
3295-3295
ISSN
2041-1723 (ISSN-L)
DOI
10.1038/s41467-020-17142-3
PMC
PMC7335176
OpenAlex
W3040129451
Cited by
283
References
3
Citations by year
2026: 26 · 2025: 31 · 2024: 74 · 2023: 48 · 2022: 40 · 2021: 57 · 2020: 7
FWCI
12.46
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • DSpace@MIT (Massachusetts Institute of Technology) (submittedVersion, cc-by-nc)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • arXiv (Cornell University)
Funders
MIT Environmental Solutions Initiative EAPS Houghton-Lorenz postdoctoral fellowship
Topics
  • Physical Sciences › Earth and Planetary Sciences › Atmospheric Science › Meteorological Phenomena and Simulations
  • Physical Sciences › Environmental Science › Global and Planetary Change › Climate variability and models
  • Physical Sciences › Earth and Planetary Sciences › Oceanography › Oceanographic and Atmospheric Processes
Keywords
subgrid-scale modeling, machine learning parameterization, random forest, climate drift, convection parameterization, coarse-graining, grid resolution dependence, climate model uncertainty

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review