Articles | Volume 28, issue 3
https://doi.org/10.5194/npg-28-347-2021
© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/npg-28-347-2021
© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Research article
| 
30 Jul 2021
Research article |
| 30 Jul 2021
| 30 Jul 2021
Producing realistic climate data with generative adversarial networks
CERFACS, Toulouse, France
Institut National Polytechnique de Toulouse, Toulouse, France
Olivier Pannekoucke
CORRESPONDING AUTHOR
CNRM, Université de Toulouse, Météo-France, CNRS, Toulouse, France
Corentin Lapeyre
CORRESPONDING AUTHOR
CERFACS, Toulouse, France
CERFACS, Toulouse, France
Olivier Thual
CORRESPONDING AUTHOR
CERFACS, Toulouse, France
Institut de Mécanique des Fluides de Toulouse (IMFT), Université de Toulouse, CNRS, Toulouse, France
Related authors
No articles found.
Norman J. Steinert and Benjamin M. Sanderson
Earth Syst. Dynam., 16, 1711–1721, https://doi.org/10.5194/esd-16-1711-2025, https://doi.org/10.5194/esd-16-1711-2025, 2025
Short summary
Short summary
In this study, we explore how carbon emissions from thawing permafrost, known as the permafrost carbon feedback, affect two important climate metrics: how much the Earth warms per amount of carbon we emit and how much warming continues after we stop emitting carbon. Our study tackles a major gap in how we estimate future climate change. Using simplified climate models, we find a generalizable relationship between the permafrost carbon feedback and its additional warming impact on climate.
John P. Dunne, Helene T. Hewitt, Julie M. Arblaster, Frédéric Bonou, Olivier Boucher, Tereza Cavazos, Beth Dingley, Paul J. Durack, Birgit Hassler, Martin Juckes, Tomoki Miyakawa, Matt Mizielinski, Vaishali Naik, Zebedee Nicholls, Eleanor O'Rourke, Robert Pincus, Benjamin M. Sanderson, Isla R. Simpson, and Karl E. Taylor
Geosci. Model Dev., 18, 6671–6700, https://doi.org/10.5194/gmd-18-6671-2025, https://doi.org/10.5194/gmd-18-6671-2025, 2025
Short summary
Short summary
The seventh phase of the Coupled Model Intercomparison Project (CMIP7) coordinates efforts to answer key and timely climate science questions and facilitate delivery of relevant multi-model simulations for prediction and projection; characterization, attribution, and process understanding; and vulnerability, impact, and adaptation analysis. Key to the CMIP7 design are the mandatory Diagnostic, Evaluation and Characterization of Klima and optional Assessment Fast Track experiments.
Colin Jones, Isaline Bossert, Donovan P. Dennis, Hazel Jeffery, Chris D. Jones, Torben Koenigk, Sina Loriani, Benjamin Sanderson, Roland Séférian, Klaus Wyser, Shuting Yang, Manabu Abe, Sebastian Bathiany, Pascale Braconnot, Victor Brovkin, Friedrich A. Burger, Patrica Cadule, Frederic S. Castruccio, Gokhan Danabasoglu, Andrea Dittus, Jonathan F. Donges, Friederike Fröb, Thomas Frölicher, Goran Georgievski, Chuncheng Guo, Aixue Hu, Peter Lawrence, Paul Lerner, José Licón-Saláiz, Bette Otto-Bliesner, Anastasia Romanou, Elena Shevliakova, Yona Silvy, Didier Swingedouw, Jerry Tjiputra, Jeremy Walton, Andy Wiltshire, Ricarda Winkelmann, Richard Wood, Tokuta Yokohata, and Tilo Ziehn
EGUsphere, https://doi.org/10.5194/egusphere-2025-3604, https://doi.org/10.5194/egusphere-2025-3604, 2025
This preprint is open for discussion and under review for Geoscientific Model Development (GMD).
Short summary
Short summary
We introduce a new Earth system model experiment protocol to help researchers understand how Earth might respond to positive, zero, and negative carbon emissions. This protocol enables different models to be compared following similar warming and cooling rates. Researchers use the models to explore how the Earth reacts to different climate futures, including the risk of tipping points being exceeded and whether changes can be reversed. The results will support improved long-term climate policy.
Ken S. Carslaw, Leighton A. Regayre, Ulrike Proske, Andrew Gettelman, David M. H. Sexton, Yun Qian, Lauren Marshall, Oliver Wild, Marcus van Lier-Walqui, Annika Oertel, Saloua Peatier, Ben Yang, Jill S. Johnson, Sihan Li, Daniel T. McCoy, Benjamin M. Sanderson, Christina J. Williamson, Gregory S. Elsaesser, Kuniko Yamazaki, and Ben B. B. Booth
EGUsphere, https://doi.org/10.5194/egusphere-2025-4341, https://doi.org/10.5194/egusphere-2025-4341, 2025
This preprint is open for discussion and under review for Atmospheric Chemistry and Physics (ACP).
Short summary
Short summary
A major challenge in climate science is reducing projection uncertainty despite advances in models and observational constraints. Perturbed parameter ensembles (PPEs) offer a powerful tool to explore and reduce uncertainty by revealing model weaknesses and guiding development. PPEs are now widely applied across climate systems and scales. We argue they should be prioritized alongside complexity and resolution in model resource planning.
Benjamin M. Sanderson, Victor Brovkin, Rosie A. Fisher, David Hohn, Tatiana Ilyina, Chris D. Jones, Torben Koenigk, Charles Koven, Hongmei Li, David M. Lawrence, Peter Lawrence, Spencer Liddicoat, Andrew H. MacDougall, Nadine Mengis, Zebedee Nicholls, Eleanor O'Rourke, Anastasia Romanou, Marit Sandstad, Jörg Schwinger, Roland Séférian, Lori T. Sentman, Isla R. Simpson, Chris Smith, Norman J. Steinert, Abigail L. S. Swann, Jerry Tjiputra, and Tilo Ziehn
Geosci. Model Dev., 18, 5699–5724, https://doi.org/10.5194/gmd-18-5699-2025, https://doi.org/10.5194/gmd-18-5699-2025, 2025
Short summary
Short summary
This study investigates how climate models warm in response to simplified carbon emissions trajectories, refining the understanding of climate reversibility and commitment. Metrics are defined for warming response to cumulative emissions and for the cessation of emissions or ramp-down to net-zero and net-negative levels. Results indicate that previous concentration-driven experiments may have overstated the Zero Emissions Commitment due to emissions rates exceeding historical levels.
Susanne Baur, Benjamin M. Sanderson, Roland Séférian, and Laurent Terray
Earth Syst. Dynam., 16, 667–681, https://doi.org/10.5194/esd-16-667-2025, https://doi.org/10.5194/esd-16-667-2025, 2025
Short summary
Short summary
Stratospheric aerosol injection (SAI) could be used alongside mitigation to reduce global warming. Previous studies suggest that more atmospheric CO2 is taken up when SAI is deployed. Here, we look at the entire SAI deployment from start to after termination. We show how the initial CO2 uptake benefit, and hence lower mitigation burden, is reduced in later stages of SAI, where the reduction in natural CO2 uptake turns into an additional mitigation burden.
Marit Sandstad, Norman Julius Steinert, Susanne Baur, and Benjamin Mark Sanderson
EGUsphere, https://doi.org/10.5194/egusphere-2025-1038, https://doi.org/10.5194/egusphere-2025-1038, 2025
Short summary
Short summary
In this article we present METEORv1.0.0, a climate model emulator, that can be trained on full spacially resolved and widely available climate model data to reproduce climate variables, and make predictions from unseen emission trajectories. The methodology which consists of identifying patterns associated with various timescales of impact for one or more forcers using idealised experiments and anomaly calculations. Results for precipitation and temperature show good model performance.
Detlef van Vuuren, Brian O'Neill, Claudia Tebaldi, Louise Chini, Pierre Friedlingstein, Tomoko Hasegawa, Keywan Riahi, Benjamin Sanderson, Bala Govindasamy, Nico Bauer, Veronika Eyring, Cheikh Fall, Katja Frieler, Matthew Gidden, Laila Gohar, Andrew Jones, Andrew King, Reto Knutti, Elmar Kriegler, Peter Lawrence, Chris Lennard, Jason Lowe, Camila Mathison, Shahbaz Mehmood, Luciana Prado, Qiang Zhang, Steven Rose, Alexander Ruane, Carl-Friederich Schleussner, Roland Seferian, Jana Sillmann, Chris Smith, Anna Sörensson, Swapna Panickal, Kaoru Tachiiri, Naomi Vaughan, Saritha Vishwanathan, Tokuta Yokohata, and Tilo Ziehn
EGUsphere, https://doi.org/10.5194/egusphere-2024-3765, https://doi.org/10.5194/egusphere-2024-3765, 2025
Short summary
Short summary
We propose a set of six plausible 21st century emission scenarios, and their multi-century extensions, that will be used by the international community of climate modeling centers to produce the next generation of climate projections. These projections will support climate, impact and mitigation researchers, provide information to practitioners to address future risks from climate change, and contribute to policymakers’ considerations of the trade-offs among various levels of mitigation.
Benjamin M. Sanderson, Ben B. B. Booth, John Dunne, Veronika Eyring, Rosie A. Fisher, Pierre Friedlingstein, Matthew J. Gidden, Tomohiro Hajima, Chris D. Jones, Colin G. Jones, Andrew King, Charles D. Koven, David M. Lawrence, Jason Lowe, Nadine Mengis, Glen P. Peters, Joeri Rogelj, Chris Smith, Abigail C. Snyder, Isla R. Simpson, Abigail L. S. Swann, Claudia Tebaldi, Tatiana Ilyina, Carl-Friedrich Schleussner, Roland Séférian, Bjørn H. Samset, Detlef van Vuuren, and Sönke Zaehle
Geosci. Model Dev., 17, 8141–8172, https://doi.org/10.5194/gmd-17-8141-2024, https://doi.org/10.5194/gmd-17-8141-2024, 2024
Short summary
Short summary
We discuss how, in order to provide more relevant guidance for climate policy, coordinated climate experiments should adopt a greater focus on simulations where Earth system models are provided with carbon emissions from fossil fuels together with land use change instructions, rather than past approaches that have largely focused on experiments with prescribed atmospheric carbon dioxide concentrations. We discuss how these goals might be achieved in coordinated climate modeling experiments.
Colin G. Jones, Fanny Adloff, Ben B. B. Booth, Peter M. Cox, Veronika Eyring, Pierre Friedlingstein, Katja Frieler, Helene T. Hewitt, Hazel A. Jeffery, Sylvie Joussaume, Torben Koenigk, Bryan N. Lawrence, Eleanor O'Rourke, Malcolm J. Roberts, Benjamin M. Sanderson, Roland Séférian, Samuel Somot, Pier Luigi Vidale, Detlef van Vuuren, Mario Acosta, Mats Bentsen, Raffaele Bernardello, Richard Betts, Ed Blockley, Julien Boé, Tom Bracegirdle, Pascale Braconnot, Victor Brovkin, Carlo Buontempo, Francisco Doblas-Reyes, Markus Donat, Italo Epicoco, Pete Falloon, Sandro Fiore, Thomas Frölicher, Neven S. Fučkar, Matthew J. Gidden, Helge F. Goessling, Rune Grand Graversen, Silvio Gualdi, José M. Gutiérrez, Tatiana Ilyina, Daniela Jacob, Chris D. Jones, Martin Juckes, Elizabeth Kendon, Erik Kjellström, Reto Knutti, Jason Lowe, Matthew Mizielinski, Paola Nassisi, Michael Obersteiner, Pierre Regnier, Romain Roehrig, David Salas y Mélia, Carl-Friedrich Schleussner, Michael Schulz, Enrico Scoccimarro, Laurent Terray, Hannes Thiemann, Richard A. Wood, Shuting Yang, and Sönke Zaehle
Earth Syst. Dynam., 15, 1319–1351, https://doi.org/10.5194/esd-15-1319-2024, https://doi.org/10.5194/esd-15-1319-2024, 2024
Short summary
Short summary
We propose a number of priority areas for the international climate research community to address over the coming decade. Advances in these areas will both increase our understanding of past and future Earth system change, including the societal and environmental impacts of this change, and deliver significantly improved scientific support to international climate policy, such as future IPCC assessments and the UNFCCC Global Stocktake.
Rachid El Montassir, Olivier Pannekoucke, and Corentin Lapeyre
Geosci. Model Dev., 17, 6657–6681, https://doi.org/10.5194/gmd-17-6657-2024, https://doi.org/10.5194/gmd-17-6657-2024, 2024
Short summary
Short summary
This study introduces a novel approach that combines physics and artificial intelligence (AI) for improved cloud cover forecasting. This approach outperforms traditional deep learning (DL) methods in producing realistic and physically consistent results while requiring less training data. This architecture provides a promising solution to overcome the limitations of classical AI methods and contributes to open up new possibilities for combining physical knowledge with deep learning models.
Marit Sandstad, Borgar Aamaas, Ane Nordlie Johansen, Marianne Tronstad Lund, Glen Philip Peters, Bjørn Hallvard Samset, Benjamin Mark Sanderson, and Ragnhild Bieltvedt Skeie
Geosci. Model Dev., 17, 6589–6625, https://doi.org/10.5194/gmd-17-6589-2024, https://doi.org/10.5194/gmd-17-6589-2024, 2024
Short summary
Short summary
The CICERO-SCM has existed as a Fortran model since 1999 that calculates the radiative forcing and concentrations from emissions and is an upwelling diffusion energy balance model of the ocean that calculates temperature change. In this paper, we describe an updated version ported to Python and publicly available at https://github.com/ciceroOslo/ciceroscm (https://doi.org/10.5281/zenodo.10548720). This version contains functionality for parallel runs and automatic calibration.
Saloua Peatier, Benjamin M. Sanderson, and Laurent Terray
Earth Syst. Dynam., 15, 987–1014, https://doi.org/10.5194/esd-15-987-2024, https://doi.org/10.5194/esd-15-987-2024, 2024
Short summary
Short summary
The calibration of Earth system model parameters is a high-dimensionality problem subject to data, time, and computational constraints. In this study, we propose a practical solution for finding diverse near-optimal solutions. We argue that the effective degrees of freedom in the model performance response to parameter input is relatively small. Comparably performing parameter configurations exist and showcase different trade-offs in model errors, providing insights for model development.
Malte Meinshausen, Carl-Friedrich Schleussner, Kathleen Beyer, Greg Bodeker, Olivier Boucher, Josep G. Canadell, John S. Daniel, Aïda Diongue-Niang, Fatima Driouech, Erich Fischer, Piers Forster, Michael Grose, Gerrit Hansen, Zeke Hausfather, Tatiana Ilyina, Jarmo S. Kikstra, Joyce Kimutai, Andrew D. King, June-Yi Lee, Chris Lennard, Tabea Lissner, Alexander Nauels, Glen P. Peters, Anna Pirani, Gian-Kasper Plattner, Hans Pörtner, Joeri Rogelj, Maisa Rojas, Joyashree Roy, Bjørn H. Samset, Benjamin M. Sanderson, Roland Séférian, Sonia Seneviratne, Christopher J. Smith, Sophie Szopa, Adelle Thomas, Diana Urge-Vorsatz, Guus J. M. Velders, Tokuta Yokohata, Tilo Ziehn, and Zebedee Nicholls
Geosci. Model Dev., 17, 4533–4559, https://doi.org/10.5194/gmd-17-4533-2024, https://doi.org/10.5194/gmd-17-4533-2024, 2024
Short summary
Short summary
The scientific community is considering new scenarios to succeed RCPs and SSPs for the next generation of Earth system model runs to project future climate change. To contribute to that effort, we reflect on relevant policy and scientific research questions and suggest categories for representative emission pathways. These categories are tailored to the Paris Agreement long-term temperature goal, high-risk outcomes in the absence of further climate policy and worlds “that could have been”.
Susanne Baur, Benjamin M. Sanderson, Roland Séférian, and Laurent Terray
Earth Syst. Dynam., 15, 307–322, https://doi.org/10.5194/esd-15-307-2024, https://doi.org/10.5194/esd-15-307-2024, 2024
Short summary
Short summary
Most solar radiation modification (SRM) simulations assume no physical coupling between mitigation and SRM. We analyze the impact of SRM on photovoltaic (PV) and concentrated solar power (CSP) and find that almost all regions have reduced PV and CSP potential compared to a mitigated or unmitigated scenario, especially in the middle and high latitudes. This suggests that SRM could pose challenges for meeting energy demands with solar renewable resources.
Théo Defontaine, Sophie Ricci, Corentin J. Lapeyre, Arthur Marchandise, and Etienne Le Pape
EGUsphere, https://doi.org/10.5194/egusphere-2023-2621, https://doi.org/10.5194/egusphere-2023-2621, 2024
Preprint archived
Short summary
Short summary
This work presents how machine learning models predict discharge and outperforms the 6 h lead-time empirical model used operationally in Toulouse. The 40 k points database includes discharge and rainfall data, for 36 flood events. The approach also provides a reliable solution for extended 8 h lead-time. The scarcity and the heterogeneity of the data, especially in presence of outliers, heavily weigh on the learning strategy. Rainfall data processing increases the predictive performances.
Susanne Baur, Alexander Nauels, Zebedee Nicholls, Benjamin M. Sanderson, and Carl-Friedrich Schleussner
Earth Syst. Dynam., 14, 367–381, https://doi.org/10.5194/esd-14-367-2023, https://doi.org/10.5194/esd-14-367-2023, 2023
Short summary
Short summary
Solar radiation modification (SRM) artificially cools global temperature without acting on the cause of climate change. This study looks at how long SRM would have to be deployed to limit warming to 1.5 °C and how this timeframe is affected by different levels of mitigation, negative emissions and climate uncertainty. None of the three factors alone can guarantee short SRM deployment. Due to their uncertainty at the time of SRM initialization, any deployment risks may be several centuries long.
Benjamin M. Sanderson and Maria Rugenstein
Earth Syst. Dynam., 13, 1715–1736, https://doi.org/10.5194/esd-13-1715-2022, https://doi.org/10.5194/esd-13-1715-2022, 2022
Short summary
Short summary
Equilibrium climate sensitivity (ECS) is a measure of how much long-term warming should be expected in response to a change in greenhouse gas concentrations. It is generally calculated in climate models by extrapolating global average temperatures to a point of where the planet is no longer a net absorber of energy. Here we show that some climate models experience energy leaks which change as the planet warms, undermining the standard approach and biasing some existing model estimates of ECS.
Benjamin M. Sanderson, Angeline G. Pendergrass, Charles D. Koven, Florent Brient, Ben B. B. Booth, Rosie A. Fisher, and Reto Knutti
Earth Syst. Dynam., 12, 899–918, https://doi.org/10.5194/esd-12-899-2021, https://doi.org/10.5194/esd-12-899-2021, 2021
Short summary
Short summary
Emergent constraints promise a pathway to the reduction in climate projection uncertainties by exploiting ensemble relationships between observable quantities and unknown climate response parameters. This study considers the robustness of these relationships in light of biases and common simplifications that may be present in the original ensemble of climate simulations. We propose a classification scheme for constraints and a number of practical case studies.
Claudia Tebaldi, Kevin Debeire, Veronika Eyring, Erich Fischer, John Fyfe, Pierre Friedlingstein, Reto Knutti, Jason Lowe, Brian O'Neill, Benjamin Sanderson, Detlef van Vuuren, Keywan Riahi, Malte Meinshausen, Zebedee Nicholls, Katarzyna B. Tokarska, George Hurtt, Elmar Kriegler, Jean-Francois Lamarque, Gerald Meehl, Richard Moss, Susanne E. Bauer, Olivier Boucher, Victor Brovkin, Young-Hwa Byun, Martin Dix, Silvio Gualdi, Huan Guo, Jasmin G. John, Slava Kharin, YoungHo Kim, Tsuyoshi Koshiro, Libin Ma, Dirk Olivié, Swapna Panickal, Fangli Qiao, Xinyao Rong, Nan Rosenbloom, Martin Schupfner, Roland Séférian, Alistair Sellar, Tido Semmler, Xiaoying Shi, Zhenya Song, Christian Steger, Ronald Stouffer, Neil Swart, Kaoru Tachiiri, Qi Tang, Hiroaki Tatebe, Aurore Voldoire, Evgeny Volodin, Klaus Wyser, Xiaoge Xin, Shuting Yang, Yongqiang Yu, and Tilo Ziehn
Earth Syst. Dynam., 12, 253–293, https://doi.org/10.5194/esd-12-253-2021, https://doi.org/10.5194/esd-12-253-2021, 2021
Short summary
Short summary
We present an overview of CMIP6 ScenarioMIP outcomes from up to 38 participating ESMs according to the new SSP-based scenarios. Average temperature and precipitation projections according to a wide range of forcings, spanning a wider range than the CMIP5 projections, are documented as global averages and geographic patterns. Times of crossing various warming levels are computed, together with benefits of mitigation for selected pairs of scenarios. Comparisons with CMIP5 are also discussed.
Katherine Dagon, Benjamin M. Sanderson, Rosie A. Fisher, and David M. Lawrence
Adv. Stat. Clim. Meteorol. Oceanogr., 6, 223–244, https://doi.org/10.5194/ascmo-6-223-2020, https://doi.org/10.5194/ascmo-6-223-2020, 2020
Short summary
Short summary
Uncertainties in land model projections are important to understand in order to build confidence in Earth system modeling. In this paper, we introduce a framework for estimating uncertain land model parameters with machine learning. This method increases the computational efficiency of this process relative to traditional hand tuning approaches and provides objective methods to assess the results. We further identify key processes and parameters that are important for accurate land modeling.
Cited articles
Besombes, C.: Producing realistic climate data with GANs, Zenodo [data set], https://doi.org/10.5281/zenodo.4442450, 2021 (data available at: https://github.com/Cam-B04/Producing-realistic-climate-data-with-GANs.git, last access: January 2021). a
Beusch, L., Gudmundsson, L., and Seneviratne, S. I.: Emulating Earth system model temperatures with MESMER: from global mean temperature trajectories to grid-point-level realizations on land, Earth Syst. Dynam., 11, 139–159, https://doi.org/10.5194/esd-11-139-2020, 2020. a
Boukabara, S.-A., Krasnopolsky, V., Stewart, J. Q., Maddy, E. S., Shahroudi, N., and Hoffman, R. N.: Leveraging modern artificial intelligence for remote sensing and NWP: Benefits and challenges, B. Am. Meteorol. Soc.,
100, ES473–ES491, 2019. a
Chan, S. and Elsheikh, A. H.:
Parametric generation of conditional geological realizations using generative neural networks,
Computat. Geosci.,
23, 925–952, https://doi.org/10.1007/s10596-019-09850-7, 2019. a
Davies, T.:
Lateral boundary conditions for limited area models,
Q. J. Roy. Meteor. Soc.,
140, 185–196, 2014. a
Dramsch, J. S.: 70 years of machine learning in geoscience in review,
Adv. Geophys., 61, 1–55, 2020. a
Fraedrich, K., Kirk, E., Lunkeit, F., Luksch, U., and Lunkeit, F.: The portable university model of the atmosphere (PUMA): Storm track dynamics and low-frequency variability, Meteorol. Z, 14, 735–746, 2005b. a
Gagne, D. J., Christensen, H. M., Subramanian, A. C., and Monahan, A. H.:
Machine learning for stochastic parameterization: Generative adversarial networks in the Lorenz'96 model, J. Adv. Model. Earth Sy., 12, e2019MS001896, https://doi.org/10.1029/2019MS001896, 2020. a
Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y.: Generative adversarial networks, Communications of the ACM, 63, 139–144, 2020. a
He, K., Zhang, X., Ren, S., and Sun, J.:
Deep residual learning for image recognition, in: Proceedings of the IEEE conference on computer vision and pattern recognition, CVPR 2016, 770–778, 2016. a
Hergenrother, E., Bleile, A., Middleton, D., and Trembilski, A.:
The abalone interpolation: A visual interpolation procedure for the calculation of cloud movement,
in: Proceedings. XV Brazilian Symposium on Computer Graphics and Image Processing, Fortaleza, Brazil, 10 October 2002, 381–387, IEEE, 2002. a
Houtekamer, P. L. and Zhang, F.:
Review of the ensemble Kalman filter for atmospheric data assimilation,
Mon. Weather Rev.,
144, 4489–4532, 2016. a
Isola, P., Zhu, J.-Y., Zhou, T., and Efros, A. A.:
Image-to-image translation with conditional adversarial networks,
in: Proceedings of the IEEE conference on computer vision and pattern recognition, Honolulu, HI, USA, 21–26 July 2017, 1125–1134, 2017. a
Kantorovich, L. V. and Rubinshtein, S.: On a space of totally additive functions, Vestnik of the St. Petersburg University: Mathematics,
13, 52–59, 1958. a
Kingma, D. P. and Ba, J.:
Adam: A Method for Stochastic Optimization, arXiv [preprint],
arXiv:1412.6980, 22 December
2014. a
Lagerquist, R., McGovern, A., and Gagne II, D. J.:
Deep learning for spatially explicit prediction of synoptic-scale fronts,
Weather Forecast.,
34, 1137–1160, 2019. a
Ledig, C., Theis, L., Huszár, F., Caballero, J., Cunningham, A., Acosta, A., Aitken, A., Tejani, A., Totz, J., Wang, Z., and Shi, W.:
Photo-realistic single image super-resolution using a generative adversarial network, in: Proceedings of the IEEE conference on computer vision and pattern recognition, Honolulu, HI, USA, 21–26 July 2017,
4681–4690, 2017. a
Leinonen, J., Guillaume, A., and Yuan, T.:
Reconstruction of cloud vertical structure with a generative adversarial network,
Geophys. Res. Lett.,
46, 7035–7044, 2019. a
Li, J. and Heap, A. D.:
Spatial interpolation methods applied in the environmental sciences: A review,
Environ. Modell. Softw.,
53, 173–189, 2014. a
Lorenc, A. C.:
The potential of the ensemble Kalman filter for NWP–a comparison with 4D-Var,
Q. J. Roy. Meteor. Soc.,
129, 3183–3203, 2003. a
Nagarajan, V. and Kolter, J. Z.: Gradient descent GAN optimization is locally stable, arXiv [preprint], arXiv:1706.04156, 13 June 2017. a
Pannekoucke, O., Cebron, P., Oger, N., and Arbogast, P.:
From the Kalman Filter to the Particle Filter: A geometrical perspective of the curse of dimensionality, Adv. Meteorol.,
2016, 9372786, https://doi.org/10.1155/2016/9372786, 2016. a
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, E.:
Scikit-learn: Machine Learning in Python,
J. Mach. Learn. Res.,
12, 2825–2830, 2011. a
Peleg, N., Fatichi, S., Paschalis, A., Molnar, P., and Burlando, P.: An advanced stochastic weather generator for simulating 2-D high-resolution climate variables, J. Adv. Model. Earth Sy., 9, 1595–1627, 2017. a
Reichstein, M., Camps-Valls, G., Stevens, B., Jung, M., Denzler, J., Carvalhais, N., and Prabhat:
Deep learning and process understanding for data-driven Earth system science,
Nature,
566, 195–204, 2019. a
Requena-Mesa, C., Reichstein, M., Mahecha, M., Kraft, B., and Denzler, J.:
Predicting landscapes as seen from space from environmental conditions,
in: IGARSS 2018 – 2018 IEEE International Geoscience and Remote Sensing Symposium, Valencia, Spain, 22–27 July 2018, 1768–1771, IEEE, 2018. a
Scher, S.: Toward data-driven weather and climate forecasting: Approximating a simple general circulation model with deep learning, Geophys. Res. Lett., 45, 12616–12622, 2018. a
Springenberg, J. T., Dosovitskiy, A., Brox, T., and Riedmiller, M.:
Striving for simplicity: The all convolutional net, arXiv [preprint], arXiv:1412.6806, 21 December 2014. a
Vallis, G. K.: Atmospheric and Oceanic Fluid Dynamics, Cambridge University Press, Cambridge, UK, https://doi.org/10.2277/0521849691, 2006. a, b, c
Watson-Parris, D.: Machine learning for weather and climate are worlds apart,
Philos. T. R. Soc. A, 379, 20200098, https://doi.org/10.1098/rsta.2020.0098, 2021. a
Weyn, J. A., Durran, D. R., and Caruana, R.:
Can machines learn to predict weather? Using deep learning to predict gridded 500-hPa geopotential height from historical weather data,
J. Adv. Model. Earth Sy.,
11, 2680–2693, 2019. a
Weyn, J. A., Durran, D. R., and Caruana, R.:
Improving Data-Driven Global Weather Prediction Using Deep Convolutional Neural Networks on a Cubed Sphere,
J. Adv. Model. Earth Sy.,
12, e2020MS002109, https://doi.org/10.1029/2020MS002109, 2020. a
White, T.: Sampling Generative Networks, arXiv [preprint], arXiv:1609.04468, 14 September 2016. a
Wilks, D. S. and Wilby, R. L.:
The weather generation game: a review of stochastic weather models,
Prog. Phys. Geog.,
23, 329–357, 1999. a
Wu, J.-L., Kashinath, K., Albert, A., Chirila, D., Prabhat, and Xiao, H.:
Enforcing statistical constraints in generative adversarial networks for modeling chaotic dynamical systems, J. Comput. Phys., 406, 109209, https://doi.org/10.1016/j.jcp.2019.109209, 2020. a
Yeh, R. A., Chen, C., Yian Lim, T., Schwing, A. G., Hasegawa-Johnson, M., and Do, M. N.:
Semantic image inpainting with deep generative models,
in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA, 21–26 July 2017, 5485–5493, 2017. a
Zhang, X.-C., Chen, J., Garbrecht, J., and Brissette, F.:
Evaluation of a weather generator-based method for statistically downscaling non-stationary climate scenarios for impact assessment at a point scale,
T. ASABE,
55, 1745–1756, 2012. a
Short summary
This paper investigates the potential of a type of deep generative neural network to produce realistic weather situations when trained from the climate of a general circulation model. The generator represents the climate in a compact latent space. It is able to reproduce many aspects of the targeted multivariate distribution. Some properties of our method open new perspectives such as the exploration of the extremes close to a given state or how to connect two realistic weather states.
This paper investigates the potential of a type of deep generative neural network to produce...
