Neural general circulation models for weather and climate
Open accessThe DOI matches its Crossref record; 54 of the 54 references it cites were found in a registry or an index; 13 of the 26 sources its claims cite were checked for support.
Abstract
General circulation models (GCMs) are the foundation of weather and climate prediction. GCMs are physics-based simulators that combine a numerical solver for large-scale dynamics with tuned representations for small-scale processes such as cloud formation. Recently, machine-learning models trained on reanalysis data have achieved comparable or better skill than GCMs for deterministic weather forecasting. However, these models have not demonstrated improved ensemble forecasts, or shown sufficient stability for long-term weather and climate simulations. Here we present a GCM that combines a differentiable solver for atmospheric dynamics with machine-learning components and show that it can generate forecasts of deterministic weather, ensemble weather and climate on par with the best machine-learning and physics-based methods. NeuralGCM is competitive with machine-learning models for one- to ten-day forecasts, and with the European Centre for Medium-Range Weather Forecasts ensemble prediction for one- to fifteen-day forecasts. With prescribed sea surface temperature, NeuralGCM can accurately track climate metrics for multiple decades, and climate forecasts with 140-kilometre resolution show emergent phenomena such as realistic frequency and trajectories of tropical cyclones. For both weather and climate, our approach offers orders of magnitude computational savings over conventional GCMs, although our model does not extrapolate to substantially different future climate…
Overview
CoChat Summary
## TL;DR
NeuralGCM is a fully differentiable hybrid general circulation model trained on ERA5 weather trajectories that combines a physical dynamical core with learned parameterizations. It achieves competitive performance for weather forecasting and multi-decade climate simulation, while the paper notes it does not extrapolate to substantially different future climates.
## Key findings
- On medium-range weather forecasting, NeuralGCM-0.7° and GraphCast achieved the best RMSE for the first 1–3 days (depending on variable), while NeuralGCM models compared favorably on persistent bias, including notably lower specific-humidity bias in the tropics.
- The stochastic NeuralGCM-ENS at 1.4° had lower error than ECMWF-ENS across almost all variables, lead times, and vertical levels for ensemble-mean RMSE, RMSB, and CRPS; its spread-skill ratio was approximately 1.
- In climate simulations with prescribed SST, NeuralGCM captured seasonal climate metrics and long-term warming trends: the ensemble-mean global temperature RMSE was 0.16 K versus 0.45 K for climatology, and all 22 stable 40-year NeuralGCM-2.8° runs had smaller spatial bias than the 22 CMIP6 AMIP runs.
- For tropical cyclones, NeuralGCM at 1.4° produced 83 cyclones versus 86 in ERA5 for the corresponding period, while X-SHiELD produced 40 after regridding to 1.4°.
## Methods
- The model is a fully differentiable hybrid GCM with a JAX-implemented dynamical core solving the hydrostatic primitive equations with moisture on sigma coordinates, plus a learned physics module that operates on single atmospheric columns.
- Weather evaluation used 732 forecasts from noon and midnight UTC in 2020 (held out from training for machine-learning models), regridded to 1.5°, with comparisons against ERA5 or ECMWF operational analysis depending on the baseline.
- Deterministic models were trained with a combination of losses for accuracy, spectral matching, and bias; stochastic models used CRPS and injected Gaussian random fields, with two ensemble members per forecast for training.
- Climate experiments used deterministic models at 2.8° and 1.4° with prescribed historical SST and sea ice, including 37 initial conditions for 2-year seasonal-cycle runs and 37 initial conditions for 40-year AMIP-like runs, of which 35 and 22 were stable, respectively.
## Limitations
- The paper explicitly states that NeuralGCM does not extrapolate to substantially different future climates.
- The authors note that for more substantial SST increases (+4 K), NeuralGCM’s response diverges from expectations, and that AMIP simulations with increased SST show climate drift.
- The text makes evident that the current version cannot distinguish precipitation from evaporation because it directly predicts column tendencies; this is stated in the paper.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures












Verification
Claims the paper makes
6 Valid · 1 Not valid · 10 Not checked · 53 Own results and interpretations
Some cited sources support this claim and some don’t — read each source’s verdict below.
In the paper
“Our results provide strong evidence for the disputed hypothesis37–39 that learning to predict short-term weather is an effective way to tune parameterizations for climate.”
§ Neural general circulation models for weather and climate > Discussion
What the sources say
- Partly supportedcites [37] Ruiz, J. J., Pulido, M. & Miyoshi, T. Estimating model parameters with ensemble-based data assimilation: a review. J. Meteorol. Soc. Jpn Ser. II 91, 79–99 (2013). sourceExperimental — accuracy not yet measured
The Conclusions (p. 98) do carry the substance attributed here — 'Parameter estimation also has the potential to improve model climatology... Model parameters can be trained during relatively short periods (i.e., a couple of years) using a data assimilation cycle, and then the estimated parameters can be used for climate simulations' — and the Introduction (p. 90) lists climate-model optimization among the applications of parameter estimation. But the source states this as a potential, not as demonstrated evidence: the Abstract (p. 89) and Section 3.4 (pp. 93-94) limit the paper's own results to a positive effect on short-to medium-range numerical weather prediction (RMSE/bias of 24-h precipitation, spread-error relationship), with no climate simulation performed, so 'strong evidence ... is an effective way to tune parameterizations for climate' overstates it, and nothing in our copy describes the hypothesis as disputed. (LLM, conf=0.75)
- Supportedcites [38] Schneider, T., Lan, S., Stuart, A. & Teixeira, J. Earth system modeling 2.0: a blueprint for models that learn from observations and targeted high-resolution simulations. Geophys. Res. Lett. 44, 12–396 (2017). source
Section 2.1 (Information Sources for Parameterization Schemes) records that parameters in cloud, convection and precipitation schemes 'have been estimated by minimizing errors in short-term weather forecasts over time scales of hours or days', i.e. the hypothesis the citing sentence names. Section 2.3 (Objectives) shows it is indeed contested: 'This has given rise to hopes that it may suffice to evaluate objective functions by weather hindcasts over time scales of only hours ... But experience has shown that such short-term optimization may not always lead to the desired improvements in climate simulations (Schirber et al., 2013)', and the authors consequently choose accumulation windows wide compared with the O(10 days) forgetting time. The citation is attached only to the disputed hypothesis itself, not to the citing paper's claim that their results support it, so our copy supports what is attributed here. (LLM, conf=0.80)
- Disputedcites [39] Schneider, T., Leung, L. R. & Wills, R. C. J. Opinion: Optimizing climate models with process knowledge, resolution, and artificial intelligence. Atmos. Chem. Phys. 24, 7041–7062 (2024). sourceExperimental — accuracy not yet measured
Our copy of Schneider et al. (2024) argues the opposite of what the claim attributes to it: in Sect. 5 ('AI for learning parameterizations') it states that supervised learning focused on the short-term residual has limitations, that 'focusing solely on minimizing the short-term residual ϵ' means the learned parameterization 'may not necessarily minimize the climate-relevant loss function... concerned with longer-term statistics,' and cites Schirber et al. (2013) as an example where 'short-term optimization over trajectories can lead to no improvement or even the degradation of the longer-term climate statistics.' The same section says holistic learning from data 'is more likely to be successful in weather prediction, where short-term accuracy is prioritized... In contrast, climate prediction faces two primary challenges,' and that learned parameterizations 'often struggle with conserving essential quantities such as energy and exhibit instabilities during extended integrations... because minimizing the short-term residual ϵ does not inherently ensure conservation or stability.' The paper also never frames this as a merely 'disputed hypothesis' but advocates process-based, online, climate-relevant-loss learning instead (Sect. 3 and Sect. 5). (LLM, conf=0.85)
Checked by a reasoning model — may be wrong.
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Recently, machine-learning models trained on reanalysis data have achieved comparable or better skill than GCMs for deterministic weather forecasting3,4.”
§ Neural general circulation models for weather and climate > Abstract General circulation models (GCMs) are the foundation of weather and climate prediction1,2. GCMs are physics-based simulators that combine a numerical solver for large-scale dynamics with tuned representations for small-scale processes such as cloud formation. Recently, machine-learning models trained on reanalysis data have achieved comparable or better skill than GCMs for deterministic weather forecasting3,4. However, these models have not demonstrated improved ensemble forecasts, or shown sufficient stability for long-term weather and climate simulations. Here we present a GCM that combines a differentiable solver for atmospheric dynamics with machine-learning components and show that it can generate forecasts of deterministic weather, ensemble weather and climate on par with the best machine-learning and physics-based methods. NeuralGCM is competitive with machine-learning models for one- to ten-day forecasts, and with the European Centre for Medium-Range Weather Forecasts ensemble prediction for one- to fifteen-day forecasts. With prescribed sea surface temperature, NeuralGCM can accurately track climate metrics for multiple decades, and climate forecasts with 140-kilometre resolution show emergent phenomena such as realistic frequency and trajectories of tropical cyclones. For both weather and climate, our approach offers orders of magnitude computational savings over conventional GCMs, although our model does not extrapolate to substantially different future climates. Our results show that end-to-end deep learning is compatible with tasks performed by conventional GCMs and can enhance the large-scale physical simulations that are essential for understanding and predicting the Earth system.
What the sources say
- Supportedcites [3] Lam, R. et al. Learning skillful medium-range global weather forecasting. Science 382, 1416–1421 (2023). source
The paper’s opening summary says GraphCast was trained on reanalysis data and significantly outperformed an operational deterministic forecasting system on 90% of 1,380 verification targets. The Conclusions also describe machine-learning weather prediction as competitive with traditional methods. The paper compares GraphCast specifically with HRES, rather than using the term “GCMs.” (LLM, conf=0.90)
- Partly supportedcites [4] Bi, K. et al. Accurate medium-range global weather forecasting with 3D neural networks. Nature 619, 533–538 (2023). sourceExperimental — accuracy not yet measured
The Discussion and Deterministic global weather forecast sections report that Pangu-Weather, trained on reanalysis data, achieved better deterministic forecast accuracy than ECMWF’s operational IFS. The paper identifies IFS as a numerical weather prediction system, however, and does not report a comparison with a system it calls a GCM. (LLM, conf=0.82)
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“So far, hybrid models have mostly been limited to idealized scenarios such as aquaplanets22,23.”
§ Neural general circulation models for weather and climate > Main
What the sources say
- Supportedcites [22] Rasp, S., Pritchard, M. S. & Gentine, P. Deep learning to represent subgrid processes in climate models. Proc. Natl Acad. Sci. USA 115, 9684–9689 (2018). source
The Climate Model and Neural Network Setup section describes this hybrid model as using an aquaplanet setup. The Discussion calls its training climate idealized and says extending the method to a less idealized setup is an immediate follow-up task. The paper thus supports the cited example of the limitation, though it does not quantify how many hybrid models use such setups. (LLM, conf=0.87)
- Supportedcites [23] Yuval, J. & O’Gorman, P. A. Stable machine-learning parameterization of subgrid processes for climate modeling at a range of resolutions. Nat. Commun. 11, 3295 (2020). source
The Introduction describes earlier neural-network climate-model work in an aquaplanet configuration, and the Results state that this study also uses an aquaplanet configuration. The Discussion identifies training over land with topography as a challenge for future research, supporting the characterization of hybrid models as largely limited to idealized settings. (LLM, conf=0.86)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Deterministic machine-learning models using a mean-squared-error loss are rewarded for averaging over uncertainty, producing unrealistically blurry predictions when optimized for multi-day forecasts3,13.”
§ Neural general circulation models for weather and climate > Main
What the sources say
- Supportedcites [3] Lam, R. et al. Learning skillful medium-range global weather forecasting. Science 382, 1416–1421 (2023). source
The Conclusions (p. 10) state that GraphCast’s MSE objective encourages it to express uncertainty by spatially blurring predictions. The paper also reports that training for longer forecast horizons produces blurrier forecasts at longer lead times. This supports the citing sentence’s account of why MSE-optimized deterministic forecasts can become blurry. (LLM, conf=0.90)
- Partly supportedcites [13] Keisler, R. Forecasting global weather with graph neural networks. Preprint at https://arxiv.org/abs/2202.07575 (2022). sourceExperimental — accuracy not yet measured
Keisler (2022) supplies the empirical setting and the observation but not the claimed mechanism. Section 3.3.3 (p. 5) confirms the model is trained with a mean-squared-error loss, and Section 3.3.2 (p. 5) confirms it is optimized for multi-day forecasts via a multi-step rollout loss (4-, 8-, 12-step, i.e. 1-3 day rollouts). Section 4.2 (p. 6) reports that over a 12-step (3-day) rollout 'the data-driven forecast does become smoother over time' — but the paper nowhere says the model is 'rewarded for averaging over uncertainty', never frames the smoothing as a consequence of irreducible forecast uncertainty under MSE, and never calls the predictions 'unrealistically blurry' (it treats the 3-day output as generally tracking large-scale flows, and its only degradation past ~6 days is hexagon-patterned instability, Section 4.3, p. 6). So the blur/averaging-over-uncertainty causal claim attributed here exceeds what this source states. (LLM, conf=0.72)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“With prescribed sea surface temperature, NeuralGCM can accurately track climate metrics for multiple decades,”
- Also stated in: § Abstract General circulation models (GCMs) are the foundation of weather and climate prediction1,2. GCMs are physics-based simulators that combine a numerical solver for large-scale dynamics with tuned representations for small-scale processes such as cloud formation. Recently, machine-learning models trained on reanalysis data have achieved comparable or better skill than GCMs for deterministic weather forecasting3,4. However, these models have not demonstrated improved ensemble forecasts, or shown sufficient stability for long-term weather and climate simulations. Here we present a GCM that combines a differentiable solver for atmospheric dynamics with machine-learning components and show that it can generate forecasts of deterministic weather, ensemble weather and climate on par with the best machine-learning and physics-based methods. NeuralGCM is competitive with machine-learning models for one- to ten-day forecasts, and with the European Centre for Medium-Range Weather Forecasts ensemble prediction for one- to fifteen-day forecasts. With prescribed sea surface temperature, NeuralGCM can accurately track climate metrics for multiple decades, and climate forecasts with 140-kilometre resolution show emergent phenomena such as realistic frequency and trajectories of tropical cyclones. For both weather and climate, our approach offers orders of magnitude computational savings over conventional GCMs, although our model does not extrapolate to substantially different future climates. Our results show that end-to-end deep learning is compatible with tasks performed by conventional GCMs and can enhance the large-scale physical simulations that are essential for understanding and predicting the Earth system. · plain
“With prescribed sea surface temperature, NeuralGCM can accurately track climate metrics for multiple decades”
§ Neural general circulation models for weather and climate > Abstract
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § Abstract General circulation models (GCMs) are the foundation of weather and climate prediction1,2. GCMs are physics-based simulators that combine a numerical solver for large-scale dynamics with tuned representations for small-scale processes such as cloud formation. Recently, machine-learning models trained on reanalysis data have achieved comparable or better skill than GCMs for deterministic weather forecasting3,4. However, these models have not demonstrated improved ensemble forecasts, or shown sufficient stability for long-term weather and climate simulations. Here we present a GCM that combines a differentiable solver for atmospheric dynamics with machine-learning components and show that it can generate forecasts of deterministic weather, ensemble weather and climate on par with the best machine-learning and physics-based methods. NeuralGCM is competitive with machine-learning models for one- to ten-day forecasts, and with the European Centre for Medium-Range Weather Forecasts ensemble prediction for one- to fifteen-day forecasts. With prescribed sea surface temperature, NeuralGCM can accurately track climate metrics for multiple decades, and climate forecasts with 140-kilometre resolution show emergent phenomena such as realistic frequency and trajectories of tropical cyclones. For both weather and climate, our approach offers orders of magnitude computational savings over conventional GCMs, although our model does not extrapolate to substantially different future climates. Our results show that end-to-end deep learning is compatible with tasks performed by conventional GCMs and can enhance the large-scale physical simulations that are essential for understanding and predicting the Earth system. · plain
29 more claims
Create a free account to see all 70 claims and to verify the 9 references still open.
- It may state two separate claims as one. 31
- Probably not a substantial claim. 6
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?
The cited source says what the claim attributes to it.
In the citing paper
“An exception is NeuralGCM20, a hybrid NWP–MLWP method, which combines the dynamical core of a traditional NWP with local ML-based parameterizations and shows competitive performance with operational ensemble forecasts.”
From Probabilistic weather forecasting with machine learning (2025) · cites this paper as [20]
Citations
Cites
Hersbach, H., Bell, B., Berrisford, P., Hirahara, S., Horányi, A., Muñoz‐Sabater, J., Nicolas, J., Peubey, C., Radu, R., Schepers, D., et al. (2020) · Quarterly Journal of the Royal Meteorological Society · doi:10.1002/qj.3803 · Cited by 31,882 · cited 1×
“These models rely solely on machine-learning techniques, using roughly 40 years of historical data from the European Center for Medium-Range Weather Forecasts (ECMWF) reanalysis v5 (ERA5) 14 for model training and forecast initialization.”
MethodEyring, V., Bony, S., Meehl, G. A., Senior, C. A., Stevens, B., Stouffer, R. J., & Taylor, K. E. (2016) · Geoscientific Model Development · doi:10.5194/gmd-9-1937-2016 · Cited by 11,981 · cited 3×
“34 and Supplementary Table 6.”
MethodInfluential(2012) · doi:10.1017/cbo9781139177245 · Cited by 7,236 · cited 1×
“The difficulty of reducing uncertainty in long-term climate projections 9 and estimating distributions of extreme weather events 10 presents major challenges for climate mitigation and adaptation 11 .”
BackgroundStrictly Proper Scoring Rules, Prediction, and Estimation
Gneiting, T., & Raftery, A. E. (2007) · Journal of the American Statistical Association · doi:10.1198/016214506000001437 · Cited by 5,793 · cited 10×
“Our stochastic loss is based on the CRPS 28,48,49 .”
BackgroundMethodInfluentialDeep learning and process understanding for data-driven Earth system science
Reichstein, M., Camps-Valls, G., Stevens, B., Jung, M., Denzler, J., Carvalhais, N., & Prabhat (2019) · Nature · doi:10.1038/s41586-019-0912-1 · Cited by 5,673 · cited 1×
“…longer timescales 17,18 , these models have not demonstrated the ability to outperform existing GCMs. Hybrid models that combine GCMs with machine learning are appealing because they build on the interpretability, extensibility and successful track record of traditional atmospheric models 19,20 .”
BackgroundThe quiet revolution of numerical weather prediction
Bauer, P., Thorpe, A., & Brunet, G. (2015) · Nature · doi:10.1038/nature14956 · Cited by 2,729
Accurate medium-range global weather forecasting with 3D neural networks
Bi, K., Xie, L., Zhang, H., Chen, X., Gu, X., & Tian, Q. (2023) · Nature · doi:10.1038/s41586-023-06185-3 · Cited by 1,670 · cited 4×
“Recent advances in machine learning have presented an alternative for weather forecasting 3,4,12,13 .”
BackgroundMethodInfluentialLearning skillful medium-range global weather forecasting
Lam, R., Sanchez-Gonzalez, A., Willson, M., Wirnsberger, P., Fortunato, M., Alet, F., Ravuri, S., Ewalds, T., Eaton-Rosen, Z., Hu, W., et al. (2023) · Science · doi:10.1126/science.adi2336 · Cited by 1,466 · cited 5×
“Specifically, we diagnose precipitation minus evaporation (Supplementary Information section H.5) rather than directly predicting these as in machine-learning-based approaches 3 .”
BackgroundMethodInfluentialSelf-citationMarine boundary layer clouds at the heart of tropical cloud feedback uncertainties in climate models
Bony, S., & Dufresne, J. (2005) · Geophysical Research Letters · doi:10.1029/2005gl023851 · Cited by 1,302 · cited 1×
“Tuning GCMs to match historical data remains a manual process 5 , and GCMs retain many persistent errors and biases 6–8 .”
BackgroundDeep learning to represent subgrid processes in climate models
Rasp, S., Pritchard, M. S., & Gentine, P. (2018) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1810286115 · Cited by 1,044 · cited 2×
“9–22.”
BackgroundSelf-citation
Research works printed but not linked: 2
- Holton, J. R. An introduction to Dynamic Meteorology 5th edn (Elsevier, 2004).
- Sutskever, I., Vinyals, O. & Le, Q. V. Sequence to sequence learning with neural networks. Adv. Neural Inf. Process. Syst. 27, 3104–3112 (2014).
Cited by
Probabilistic weather forecasting with machine learning
Price, I., Sanchez-Gonzalez, A., Alet, F., Andersson, T. R., El-Kadi, A., Masters, D., Ewalds, T., Stott, J., Mohamed, S., Battaglia, P., et al. (2025) · Nature · doi:10.1038/s41586-024-08252-9 · Cited by 434
Self-citationWeatherBench 2: A Benchmark for the Next Generation of Data‐Driven Global Weather Models
Rasp, S., Hoyer, S., Merose, A., Langmore, I., Battaglia, P., Russell, T., Sanchez‐Gonzalez, A., Yang, V., Carver, R., Agrawal, S., et al. (2024) · Journal of Advances in Modeling Earth Systems · doi:10.1029/2023ms004019 · Cited by 245
Self-citationInterpretable machine learning for weather and climate prediction: A review
Yang, R., Hu, J., Li, Z., Mu, J., Yu, T., Xia, J., Li, X., Dasgupta, A., & Xiong, H. (2024) · Atmospheric Environment · doi:10.1016/j.atmosenv.2024.120797 · Cited by 139
Early warning of complex climate risk with integrated artificial intelligence
Reichstein, M., Benson, V., Blunk, J., Camps‐Valls, G., Creutzig, F., Fearnley, C. J., Han, B., Kornhuber, K., Rahaman, N., Schölkopf, B., et al. (2025) · Nature Communications · doi:10.1038/s41467-025-57640-w · Cited by 239
Guo, Q., He, Z., Wang, Z., Qiao, S., Zhu, J., & Chen, J. (2024) · Water · doi:10.3390/w16192870 · Cited by 155
Metadata
- Authors
- Kochkov, D., Yuval, J., Langmore, I., Norgaard, P., Smith, J., Mooers, G., Klöwer, M., Lottes, J., Rasp, S., Düben, P., Hatfield, S., Battaglia, P., Sanchez-Gonzalez, A., Willson, M., Brenner, M. P., & Hoyer, S.
- Affiliations
- Google (United States) (US)
- Massachusetts Institute of Technology (US)
- European Centre for Medium-Range Weather Forecasts (GB)
- Google DeepMind (United Kingdom) (GB)
- Corresponding author
- Kochkov, Dmitrii
- ORCID
- Venue
- Nature
- Published
- 2024-07-22
- Volume
- 632
- Issue
- 8027
- Pages
- 1060-1066
- ISSN
- 0028-0836 (ISSN-L)
- 1476-4687
- PMC
- PMC11357988
- OpenAlex
- W4400880443
- Cited by
- 454
- References
- 54
- Citations by year
- 2026: 188 · 2025: 209 · 2024: 57
- FWCI
- 82.79
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- creativecommons.org/licenses/by/4.0 (tdm)
- creativecommons.org/licenses/by/4.0 (vor)
- Locations
- Nature (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- RePEc: Research Papers in Economics (submittedVersion)
- arXiv (Cornell University)
- 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 › Physics and Astronomy › Statistical and Nonlinear Physics › Model Reduction and Neural Networks
- Keywords
- weather forecasting, climate prediction, ensemble forecasting, tropical cyclones, long-term climate simulation, computational efficiency
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (qwen3.8-omni-flash and gpt-6-sol).
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Neural general circulation models for weather and climate. https://app.cochat.ai/papers/10.1038/s41586-024-07744-y
- The paper
- Kochkov, D., Yuval, J., Langmore, I., Norgaard, P., Smith, J., Mooers, G., Klöwer, M., Lottes, J., Rasp, S., Düben, P., Hatfield, S., Battaglia, P., Sanchez-Gonzalez, A., Willson, M., Brenner, M. P., & Hoyer, S. (2024). Neural general circulation models for weather and climate. Nature. https://doi.org/10.1038/s41586-024-07744-y
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.