Abstract
Climate modelling and analysis are facing new demands to enhance projections and climate information. Here we argue that now is the time to push the frontiers of machine learning beyond state-of-the-art approaches, not only by developing machine-learning-based Earth system models with greater fidelity, but also by providing new capabilities through emulators for extreme event projections with large ensembles, enhanced detection and attribution methods for extreme events, and advanced climate model analysis and benchmarking. Utilizing this potential requires key machine learning challenges to be addressed, in particular generalization, uncertainty quantification, explainable artificial intelligence and causality. This interdisciplinary effort requires bringing together machine learning and climate scientists, while also leveraging the private sector, to accelerate progress towards actionable climate science.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$29.99 /Â 30Â days
cancel any time
Subscribe to this journal
Receive 12 print issues and online access
$209.00 per year
only $17.42 per issue
Buy this article
- Purchase on SpringerLink
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
References
Eyring, V. et al. Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization. Geosci. Model Dev. 9, 1937â1958 (2016).
Lee, J.-Y. et al. in Climate Change 2021: The Physical Science Basis (eds Masson-Delmotte, V. et al.) 553â672 (Cambridge Univ. Press, 2021).
Lam, R. et al. Learning skillful medium-range global weather forecasting. Science 382, 1416â1421 (2023).
Price, I. et al. GenCast: diffusion-based ensemble forecasting for medium-range weather. Preprint at https://doi.org/10.48550/arXiv.2312.15796 (2023).
Monteleoni, C. et al. in Computational Intelligent Data Analysis for Sustainable Development (eds Yu, T. et al.) 81â126 (CRC Press, 2013).
Rolnick, D. et al. Tackling climate change with machine learning. ACM Comput. Surv. 55, 42 (2022).
Gentine, P., Eyring, V. & Beucler, T. in Deep Learning for the Parametrization of Subgrid Processes in Climate Models (Camps-Valls, G. et al.) Ch.â21, 307â314. (John Wiley & Sons, 2021).
Reichstein, M. et al. Deep learning and process understanding for data-driven Earth system science. Nature 566, 195â204 (2019).
Eyring, V., Gentine, P., Camps-Valls, G., Lawrence, D. M. & Reichstein, M. AI-empowered next-generation multiscale climate modelling for mitigation and adaptation. Nat. Geosci. https://doi.org/10.1038/s41561-024-01527-w (2024).
Watson-Parris, D. et al. ClimateBench v1. 0: A benchmark for data-driven climate projections. J. Adv. Model. Earth Syst. 14, e2021MS002954 (2022).
Yu, S. et al. ClimSim: an open large-scale dataset for training high-resolution physics emulators in hybrid multi-scale climate simulators. Preprint at https://doi.org/10.48550/arXiv.2306.08754 (2023).
Beucler, T. et al. Climate-invariant machine learning. Sci. Adv. 10, eadj7250 (2024).
Wu, D. et al. Quantifying uncertainty in deep spatiotemporal forecasting. In Proc. 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining 1841â1851 (Association for Computing Machinery, 2021).
McGovern, A. et al. Making the black box more transparent: understanding the physical implications of machine learning. Bull. Am. Meteorol. Soc. 100, 2175â2199 (2019).
Runge, J., Gerhardus, A., Varando, G., Eyring, V. & Camps-Valls, G. Causal inference for time series. Nat. Rev. Earth Environ. 10, 2553 (2023).
Iglesias-Suarez, F. et al. Causally-informed deep learning to improve climate models and projections. J. Geophys. Res. Atmos. 129, 2023â039202 (2024).
Eyring, V. et al. Reflections and projections on a decade of climate science. Nat. Clim. Change 11, 279â285 (2021).
Eyring, V. et al. in Climate Change 2021: The Physical Science Basis (eds Masson-Delmotte, V. et al.) 423â552 (Cambridge Univ. Press, 2021).
Stevens, B. et al. DYAMOND: the DYnamics of the Atmospheric general circulation Modeled On Non-hydrostatic Domains. Prog. Earth Planet. Sci. 6, 61 (2019).
Gentine, P., Pritchard, M., Rasp, S., Reinaudi, G. & Yacalis, G. Could machine learning break the convection parameterization deadlock? Geophys. Res. Lett. 45, 5742â5751 (2018).
Grundner, A. et al. Deep learning based cloud cover parameterization for ICON. J. Adv. Model. Earth Syst. 14, 2021â002959 (2022).
Behrens, G. et al. Improving atmospheric processes in earth system models with deep learning ensembles and stochastic parameterizations. Preprint at https://doi.org/10.48550/arXiv.2402.03079 (2024).
Bretherton, C. S. et al. Correcting coarse-grid weather and climate models by machine learning from global storm-resolving simulations. J. Adv. Model. Earth Syst. 14, e2021MS002794 (2022).
Rasp, S., Pritchard, M. S. & Gentine, P. Deep learning to represent sub-grid processes in climate models. Proc. Natl Acad. Sci. USA 115, 9684â9689 (2018).
Brenowitz, N. D. et al. Machine learning climate model dynamics: offline versus online performance. In NeurIPS 2020 Workshop on Tackling Climate Change with Machine Learning (2020).
Beucler, T. et al. Enforcing analytic constraints in neural networks emulating physical systems. Phys. Rev. Lett. 126, 98302 (2021).
Wang, R., Walters, R. & Yu, R. Incorporating symmetry into deep dynamics models for improved generalization. Preprint at https://doi.org/10.48550/arXiv.2002.03061 (2020).
Rasp, S. Coupled online learning as a way to tackle instabilities and biases in neural network parameterizations: general algorithms and Lorenz 96 case study (v1.0). Geosci. Model Dev. 13, 2185â2196 (2020).
Grundner, A., Beucler, T., Gentine, P. & Eyring, V. Data-driven equation discovery of a cloud cover parameterization. J. Adv. Model. Earth Syst. 16, 2023â003763 (2024).
Camps-Valls, G. et al. Discovering causal relations and equations from data. Phys. Rep. 1044, 1â68 (2023).
Couldrey, M. P. et al. What causes the spread of model projections of ocean dynamic sea-level change in response to greenhouse gas forcing? Clim. Dyn. 56, 155â187 (2021).
Zanna, L. & Bolton, T. Data-driven equation discovery of ocean mesoscale closures. Geophys. Res. Lett. 47, e2020GL088376 (2020).
Zhang, C. et al. Implementation and evaluation of a machine learned mesoscale eddy parameterization into a numerical ocean circulation model. J. Adv. Model. Earth Syst. 15, e2023MS003697 (2023).
Guillaumin, A. P. & Zanna, L. Stochastic-deep learning parameterization of ocean momentum forcing. J. Adv. Model. Earth Syst. 13, e2021MS002534 (2021).
Bolton, T. & Zanna, L. Applications of deep learning to ocean data inference and subgrid parameterization. J. Adv. Model. Earth Syst. 11, 376â399 (2019).
Ross, A., Li, Z., Perezhogin, P., Fernandez-Granda, C. & Zanna, L. Benchmarking of machine learning ocean subgrid parameterizations in an idealized model. J. Adv. Model. Earth Syst. 15, e2022MS003258 (2023).
Friedlingstein, P. et al. Global carbon budget 2023. Earth Syst. Sci. Data 15, 5301â5369 (2023).
Dagon, K., Sanderson, B. M., Fisher, R. A. & Lawrence, D. M. A machine learning approach to emulation and biophysical parameter estimation with the Community Land Model, version 5. Adv. Stat. Climatol. Meteorol. Oceanogr. 6, 223â244 (2020).
Humphrey, V. et al. Soil moistureâatmosphere feedback dominates land carbon uptake variability. Nature 592, 65â69 (2021).
Shen, C. et al. Differentiable modelling to unify machine learning and physical models for geosciences. Nat. Rev. Earth Environ. 4, 552â567 (2023).
Zhao, W. L. et al. Physics-constrained machine learning of evapotranspiration. Geophys. Res. Lett. 46, 14496â14507 (2019).
Yang, T. et al. Evaluation and machine learning improvement of global hydrological model-based flood simulations. Environ. Res. Lett. 14, 114027 (2019).
Wang, N., Zhang, D., Chang, H. & Li, H. Deep learning of subsurface flow via theory-guided neural network. J. Hydrol. 584, 124700 (2020).
Kraft, B., Jung, M., Körner, M., Koirala, S. & Reichstein, M. Towards hybrid modeling of the global hydrological cycle. Hydrol. Earth Syst. Sci. 26, 1579â1614 (2022).
Xie, K. et al. Physics-guided deep learning for rainfall-runoff modeling by considering extreme events and monotonic relationships. J. Hydrol. 603, 127043 (2021).
Nathaniel, J., Liu, J. & Gentine, P. MetaFlux: meta-learning global carbon fluxes from sparse spatiotemporal observations. Sci. Data 10, 440 (2023).
Peherstorfer, B., Willcox, K. & Gunzburger, M. Survey of multifidelity methods in uncertainty propagation, inference, and optimization. SIAM Rev. 60, 550â591 (2018).
Cutajar, K., Pullin, M., Damianou, A., Lawrence, N. & González, J. Deep Gaussian processes for multi-fidelity modeling. Preprint at https://doi.org/10.48550/arXiv.1903.07320 (2019).
Delaunay, A. & Christensen, H. M. Interpretable deep learning for probabilistic MJO prediction. Geophys. Res. Lett. 49, 2022â098566 (2022).
Kurth, T. et al. FourCastNet: accelerating global high-resolution weather forecasting using adaptive Fourier neural operators. Preprint at https://doi.org/10.48550/arxiv.2208.05419 (2022).
Beusch, L., Gudmundsson, L. & Seneviratne, S. I. Emulating Earth system model temperatures with MESMER: from global mean temperature trajectories to grid-point-level realizations on land. Earth Syst. Dyn. 11, 139â159 (2020).
Doury, A., Somot, S., Gadat, S., Ribes, A. & Corre, L. Regional climate model emulator based on deep learning: concept and first evaluation of a novel hybrid downscaling approach. Clim. Dyn. 60, 1751â1779 (2023).
Quilcaille, Y., Gudmundsson, L., Beusch, L., Hauser, M. & Seneviratne, S. I. Showcasing MESMER-X: spatially resolved emulation of annual maximum temperatures of earth system models. Geophys. Res. Lett. 49, 2022â099012 (2022).
Ham, Y.-G., Kim, J.-H. & Luo, J.-J. Deep learning for multi-year ENSO forecasts. Nature 573, 568â572 (2019).
Immorlano, F. et al. Transferring climate change knowledge. Preprint at https://doi.org/10.48550/arXiv.2309.14780 (2023).
IPCC. Summary for Policymakers. In Climate Change 2021: The Physical Science Basis (eds Masson Delmotte, V. et al.) 3â32 (Cambridge Univ. Press, 2021).
OâBrien, T. A. et al. Increases in future AR count and size: overview of the ARTMIP tier 2 CMIP5/6 experiment. J. Geophys. Res. Atmos. 127, e2021JD036013 (2022).
Kurth, T. et al. Exascale deep learning for climate analytics. Preprint at https://doi.org/10.48550/arxiv.1810.01993 (2018).
Salcedo-Sanz, S. et al. Analysis, characterization, prediction, and attribution of extreme atmospheric events with machine learning and deep learning techniques: a review. Theor. Appl. Climatol. 155, 1â44 (2024).
OâBrien, T. A. et al. Detection of atmospheric rivers with inline uncertainty quantification: TECA-BARD v1.0.1. Geosci. Model Dev. 13, 6131â6148 (2020).
Prabhat, K. et al. ClimateNet: an expert-labeled open dataset and deep learning architecture for enabling high-precision analyses of extreme weather. Geosci. Model Dev. 14, 107â124 (2021).
Liu, Y. et al. Application of deep convolutional neural networks for detecting extreme weather in climate datasets. Preprint at https://doi.org/10.48550/arXiv.1605.01156 (2016).
Muszynski, G., Kashinath, K., Kurlin, V. & Wehner, M. Topological data analysis and machine learning for recognizing atmospheric river patterns in large climate datasets. Geosci. Model Dev. 12, 613â628 (2019).
Kim, S. et al. Deep-Hurricane-Tracker: tracking and forecasting extreme climate events. In 2019 IEEE Winter Conference on Applications of Computer Vision 1761â1769 (IEEE, 2019).
Molina, M. J. et al. A review of recent and emerging machine learning applications for climate variability and weather phenomena. Artif. Intell. Earth Syst. https://doi.org/10.1175/AIES-D-22-0086.1 (2023).
Molina, M. J., Gagne, D. J. & Prein, A. F. A benchmark to test generalization capabilities of deep learning methods to classify severe convective storms in a changing climate. Earth Space Sci. 8, e2020EA001490 (2021).
Vandal, T., Kodra, E. & Ganguly, A. R. Intercomparison of machine learning methods for statistical downscaling: the case of daily and extreme precipitation. Theor. Appl. Climatol. 137, 557â570 (2018).
Miloshevich, G., Cozian, B., Abry, P., Borgnat, P. & Bouchet, F. Probabilistic forecasts of extreme heatwaves using convolutional neural networks in a regime of lack of data. Physical Review Fluids 8, 040501 (2023).
Prodhan, F. A. et al. Projection of future drought and its impact on simulated crop yield over South Asia using ensemble machine learning approach. Sci. Total Environ. 807, 151029 (2022).
Eyring, V. et al. Earth System Model Evaluation Tool (ESMValTool) v2.0 â an extended set of large-scale diagnostics for quasi-operational and comprehensive evaluation of Earth system models in CMIP. Geosci. Model Dev. 13, 3383â3438 (2020).
Dijkstra, H., Hernandez-Garcia, E., Masoller, C. & Barreiro, M. Networks in Climate (Cambridge Univ. Press, 2019).
Karmouche, S. et al. Regime-oriented causal model evaluation of AtlanticâPacific teleconnections in CMIP6. Earth Syst. Dyn. 14, 309â344 (2023).
Nowack, P., Runge, J., Eyring, V. & Haigh, J. D. Causal networks for climate model evaluation and constrained projections. Nature Commun. 11, 1415 (2020).
Barnes, E. A., Barnes, R. J., Martin, Z. K. & Rader, J. K. This looks like that there: interpretable neural networks for image tasks when location matters. Artif. Intell. Earth Syst. 1, e220001 (2022).
Schlund, M. et al. Constraining uncertainty in projected gross primary production with machine learning. J. Geophys. Res. Biogeosci. 125, e2019JG005619 (2020).
Mooers, G. et al. Comparing storm resolving models and climates via unsupervised machine learning. Sci. Rep. 13, 22365 (2023).
Lopez-Gomez, I., McGovern, A., Agrawal, S. & Hickey, J. Global extreme heat forecasting using neural weather models. Artif. Intell. Earth Syst. 2, 220035 (2023).
Boulaguiem, Y., Zscheischler, J., Vignotto, E., van der Wiel, K. & Engelke, S. Modeling and simulating spatial extremes by combining extreme value theory with generative adversarial networks. Environ. Data Sci. 1, 5 (2022).
Jiang, C. et al. MeshfreeFlowNet: a physics-constrained deep continuous space-time super-resolution framework. In Proc. International Conference for High Performance Computing, Networking, Storage and Analysis (IEEE, 2020).
Kendall, A. & Gal, Y. What uncertainties do we need in Bayesian deep learning for computer vision? Preprint at https://doi.org/10.48550/arXiv.1703.04977 (2017).
Gal, Y. & Ghahramani, Z. Dropout as a Bayesian approximation: representing model uncertainty in deep learning. In Proc. 33rd International Conference on Machine Learning (eds Balcan, M. F. & Weinberger, K. Q.) 1050â1059 (Proceedings of Machine Learning Research, 2016).
Fort, S., Hu, H. & Lakshminarayanan, B. Deep ensembles: a loss landscape perspective. Preprint at https://doi.org/10.48550/arXiv.1912.02757 (2019).
Osband, I., Blundell, C., Pritzel, A. & Van Roy, B. Deep exploration via bootstrapped DQN. In Advances in Neural Information Processing Systems (eds Lee, D. et al.) Vol. 29 (Curran Associates, 2016).
Kingma, D. P. & Welling, M. Auto-encoding variational Bayes. Preprint at https://doi.org/10.48550/arXiv.1312.6114 (2013).
Cachay, S. R., Zhao, B., Joren, H. & Yu, R. DYffusion: a dynamics-informed diffusion model for spatiotemporal forecasting. Preprint at https://doi.org/10.48550/arXiv.2306.01984 (2023).
Hall, D. et al. NVIDIAâs Earth-2: an interactive digital twin of the Earth and its subsystems. In XXVIII General Assembly of the International Union of Geodesy and Geophysics (IUGG) (2023); https://doi.org/10.57757/IUGG23-3980
van Straaten, C., Whan, K., Coumou, D., van den Hurk, B. & Schmeits, M. Using explainable machine learning forecasts to discover subseasonal drivers of high summer temperatures in Western and Central Europe. Mon. Weather Rev. 150, 1115â1134 (2022).
Mamalakis, A., Ebert-Uphoff, I. & Barnes, E. A. in xxAI - Beyond Explainable AI (eds Holzinger, A. et al.) 315â339 (Springer, 2022).
Rader, J. K., Barnes, E. A., Ebert-Uphoff, I. & Anderson, C. Detection of forced change within combined climate fields using explainable neural networks. J. Adv. Model. Earth Syst. 14, e2021MS002941 (2022).
Toms, B. A., Barnes, E. A. & Hurrell, J. W. Assessing decadal predictability in an earth-system model using explainable neural networks. Geophys. Res. Lett. 48, e2021GL093842 (2021).
Mamalakis, A., Barnes, E. A. & Ebert-Uphoff, I. Carefully choose the baseline: lessons learned from applying XAI attribution methods for regression tasks in geoscience. Artif. Intell. Earth Syst. 2, 220058 (2023).
Rudin, C. Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nat. Mach. Intell. 1, 206â215 (2019).
McGraw, M. C. & Barnes, E. A. Memory matters: a case for granger causality in climate variability studies. J. Clim. 31, 3289â3300 (2018).
Bauer, P., Stevens, B. & Hazeleger, W. A digital twin of Earth for the green transition. Nat. Clim. Change 11, 80â83 (2021).
Foundational Research Gaps and Future Directions for Digital Twins (National Academies Press, 2024).
Nguyen, T., Brandstetter, E. J., Kapoor, A., Gupta, J. K. & Grover, A. ClimaX: a foundation model for weather and climate. Preprint at https://doi.org/10.48550/arxiv.2301.10343 (2023).
Bonev, B. et al. Spherical Fourier neural operators: learning stable dynamics on the sphere. Preprint at https://doi.org/10.48550/arXiv.2306.03838 (2023).
Mardani, M. et al. Residual diffusion modeling for km-scale atmospheric downscaling. Preprint at https://doi.org/10.48550/arXiv.2309.15214 (2023).
Brenowitz, N. D. et al. A practical probabilistic benchmark for AI weather models. Preprint at https://doi.org/10.48550/arXiv.2401.15305 (2023).
Nevo, S. et al. Flood forecasting with machine learning models in an operational framework. Hydrol. Earth Syst. Sci. 26, 4013â4032 (2022).
Acknowledgements
We acknowledge the Aspen Global Change Institute for hosting a workshop on Exploring the Frontiers in Earth System Modeling with Machine Learning and Big Data in June 2022 as part of its traditionally landmark summer interdisciplinary sessions (https://www.agci.org/event/22s3). The workshop was funded by NASAâs (National Aeronautics and Space Administration) Earth Science Division, the National Oceanic and Atmospheric Administrationâs Climate Program Office, the National Science Foundationâs (NSF) Directorate for Geosciences (GEO), and travel support from the World Climate Research Programme. V.E., P.G. and F.I.-S. were funded by the European Research Council (ERC) Synergy Grant âUnderstanding and Modelling the Earth System with Machine Learning (USMILE)â under the EU Horizon 2020 research and innovation programme (grant agreement 855187). V.E. and K.W. were also supported by the Deutsche Forschungsgemeinschaft (German Research Foundation) through the Gottfried Wilhelm Leibniz Prize awarded to V.E. (EY 22/2-1). W.D.C. was supported by the Director, Office of Science, Office of Biological and Environmental Research of the US Department of Energy (contract DE-AC02-05CH11231) and used resources of the National Energy Research Scientific Computing Center, also supported by the Office of Science of the US Department of Energy (DOE; contract DE-AC02-05CH11231). P.G. acknowledges additional funding from the NSF Science and Technology Center Learning the Earth with Artificial Intelligence and Physics (LEAP; award 2019625-STC) and DOE Advanced Scientific Computing Research programme. Additional support was received for E.A.B.: NSF (AGS-1749261) and US Department of Energy supported by the Regional and Global Model Analysis programme; C.S.B.: Allen Institute for AI; H.M.C.: Natural Environment Research Council (NE/P018238/1); K.D.: US DOE (DE-SC0022070) and NSF (IA 1947282) and NSF (1852977); D.J.G.: NSF (1852977) and NSF (ICER-2019758); D.âHall: NVIDIA; D.âHammerling: US DOE (DE-FE0032311); M.C.M.: NSF (ICER-2019758); G.A.M.: US DOE (DE-SC0022070) and NSF (IA 1947282); M.J.M.: US DOE (DE-SC0022070) and NSF (IA 1947282) and University of Maryland Grand Challenges Grants Program (GC17-2957817); C.M.: NSF (2153040); J.M.: US DOE (DE-AC36-08GO28308); M.S.P.: NSF LEAP (2019625-STC) and DOE (DE-SC0023368 and DE-SC0022255); D.R.: Canada CIFAR AI Chairs programme; J.R.: EU Horizon 2020 research and innovation programme XAIDA (101003469) and ERC Starting Grant CausalEarth (948112); P.S.: ERC project RECAP under the EU Horizon 2020 research and innovation programme (724602) and FORCeS project under the EU Horizon 2020 research programme (821205); O.W.-M.: Allen Institute for AI; R.Y.: US DOE (DE-SC0022255) and NSF (IIS-2146343); L.Z.: Schmidt Futures (G-23-65185) and NSF (GG017158-01). We thank J.âSnyder (Lawrence Berkeley National Laboratory) for his work to create Fig. 1.
Author information
Authors and Affiliations
Contributions
V.E. and W.D.C. jointly led the writing of the Perspective, and coordinated and oversaw the execution of the study. They conceived the concept of pushing the frontiers of climate modelling and analysis with ML for this study with support from P.G. and contributed to all sections of the Perspective. All authors contributed to the concept, drafting and writing of the manuscript and to the discussions.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Climate Change thanks Peter Bauer, Peter Gibson and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Additional information
Publisherâs note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Eyring, V., Collins, W.D., Gentine, P. et al. Pushing the frontiers in climate modelling and analysis with machine learning. Nat. Clim. Chang. 14, 916â928 (2024). https://doi.org/10.1038/s41558-024-02095-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41558-024-02095-y