pith. machine review for the scientific record. sign in

arxiv: 2605.01599 · v2 · submitted 2026-05-02 · ⚛️ physics.ao-ph

Recognition: 3 theorem links

· Lean Theorem

Cast3: Translating numerical weather prediction principles into data-driven forecasting

Congyi Nai , Baoxiang Pan , Yuan Liang , Xi Chen
Authors on Pith no claims yet

Pith reviewed 2026-05-08 19:26 UTC · model grok-4.3

classification ⚛️ physics.ao-ph
keywords data-driven weather forecastingnumerical weather predictioncubed-sphere gridsgenerative nudgingensemble predictiontropical cyclone forecastingspectral fidelity
0
0 comments X

The pith

Cast3 absorbs decades of NWP modeling principles into a data-driven framework through cubed-sphere grids and generative nudging to improve forecast skill.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper claims that data-driven weather forecasting has relied too heavily on reanalysis outputs while ignoring the methodological knowledge built up in traditional numerical weather prediction. Cast3 addresses this by running on variable-resolution cubed-sphere grids that allow scale-aware representation and by building super-ensembles whose diverse grid discretizations sample complementary biases. Generative nudging then collapses the ensemble information into one forecast that keeps the ensemble mean's large-scale accuracy and the fine detail of high-resolution members. The resulting model is shown to exceed both deterministic and generative baselines on synoptic skill, spectral properties, surface observations, and tropical cyclone tracks. If the approach holds, it indicates that embedding established physical modeling choices can systematically raise the ceiling of purely statistical systems.

Core claim

Cast3 is a generative forecasting framework that operates on variable-resolution cubed-sphere grids for scale-aware representation and constructs structurally diverse super-ensembles sampling the complementary biases of different grid discretizations. It introduces generative nudging, a posterior-sampling strategy that distils the collective information of the full ensemble into a single forecast possessing both the large-scale accuracy of the ensemble mean and the mesoscale realism of a high-resolution member. Evaluated across synoptic-scale skill, spectral fidelity, station-level surface verification, and tropical cyclone prediction, Cast3 outperforms established deterministic and gener

What carries the argument

variable-resolution cubed-sphere grids combined with generative nudging that collapses ensemble diversity into a single high-fidelity forecast

If this is right

  • Data-driven models can reach higher large-scale accuracy by adopting NWP discretization strategies instead of uniform latitude-longitude grids.
  • Structurally diverse ensembles built from multiple grid types reduce systematic bias compared with single-grid ensembles.
  • Generative nudging produces a deterministic forecast whose error statistics match those of the ensemble mean while preserving mesoscale variance.
  • The same combination of grid diversity and nudging improves tropical cyclone track and intensity prediction relative to existing baselines.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar grid-ensemble and nudging techniques could be tested on ocean or land-surface forecasting systems that also rely on reanalysis training.
  • The method may extend naturally to longer lead times or to coupled atmosphere-ocean models where scale interactions are even more pronounced.
  • If the performance gain persists under distribution shift, it would support the broader hypothesis that physics-derived discretization choices remain useful even inside learned models.

Load-bearing premise

That NWP meta-knowledge can be absorbed through specific architectural choices such as cubed-sphere grids and generative nudging without introducing new biases or eroding the flexibility of data-driven methods.

What would settle it

Apply Cast3 to an independent global dataset or a new climate regime never seen during training and check whether it still beats the same deterministic and generative baselines on the same four verification categories.

read the original abstract

Data-driven weather models have made rapid advances in recent years, reaching and in some metrics surpassing the large-scale forecast skill of operational numerical weather prediction. This progress, however, has been built almost entirely on the reanalysis data that NWP produced, while the methodological knowledge that the NWP community distilled over decades of multi-scale atmospheric modelling remains largely unused. Here we present Cast3, a generative forecasting framework that systematically absorbs NWP meta-knowledge to close this gap. Cast3 operates on variable-resolution cubed-sphere grids for scale-aware representation and constructs structurally diverse super-ensembles that sample the complementary biases of different grid discretizations, delivering state-of-the-art ensemble prediction. It further introduces generative nudging, a posterior-sampling strategy that distils the collective information of the full ensemble into a single forecast possessing both the large-scale accuracy of the ensemble mean and the mesoscale realism of a high-resolution member. Evaluated across synoptic-scale skill, spectral fidelity, station-level surface verification, and tropical cyclone prediction, Cast3 outperforms established deterministic and generative baselines across various dimensions. More broadly, these results demonstrate that the design principles embedded in computational atmospheric science offer a rich and largely untapped foundation for the next generation of data-driven Earth system modelling.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript introduces Cast3, a generative forecasting framework that incorporates NWP meta-knowledge via variable-resolution cubed-sphere grids, structurally diverse super-ensembles sampling complementary discretization biases, and generative nudging as a posterior-sampling method to distill ensemble information into a single forecast that combines ensemble-mean large-scale accuracy with high-resolution mesoscale realism. It claims outperformance over established deterministic and generative baselines across synoptic-scale skill, spectral fidelity, station-level surface verification, and tropical cyclone prediction.

Significance. If the results hold, the work would be significant for providing a concrete mechanism to translate decades of NWP modeling principles (scale-aware grids, ensemble diversity, posterior distillation) into data-driven systems without sacrificing their advantages, potentially improving physical consistency and multi-scale fidelity in ML weather models. The introduction of generative nudging and structurally diverse super-ensembles offers a reusable template for hybrid modeling.

major comments (2)
  1. [Abstract] Abstract: the central claim of outperformance in spectral fidelity and mesoscale realism depends on generative nudging preserving small-scale content while matching ensemble-mean accuracy, yet the abstract provides no quantitative support (e.g., no power spectra, no comparison of mesoscale kinetic energy spectra, no artifact analysis) to demonstrate that the posterior sampling does not smooth or bias high-wavenumber features.
  2. [Abstract] Abstract: the assertion that Cast3 outperforms baselines 'across various dimensions' is load-bearing for the paper's contribution but is unsupported by any reported metrics, error bars, baseline specifications, or verification details, making it impossible to assess whether the claimed advantages are statistically significant or robust.
minor comments (1)
  1. [Abstract] The abstract introduces 'structurally diverse super-ensembles' and 'generative nudging' without a concise definition or reference to their precise formulation, which hinders immediate comprehension of the methodological novelty.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive review and for recognizing the potential of incorporating NWP design principles into data-driven forecasting. We address the two major comments on the abstract below. While the manuscript body contains the requested quantitative details and figures, we agree that a modest expansion of the abstract will improve clarity and will revise accordingly.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim of outperformance in spectral fidelity and mesoscale realism depends on generative nudging preserving small-scale content while matching ensemble-mean accuracy, yet the abstract provides no quantitative support (e.g., no power spectra, no comparison of mesoscale kinetic energy spectra, no artifact analysis) to demonstrate that the posterior sampling does not smooth or bias high-wavenumber features.

    Authors: The abstract is intentionally concise, but the full manuscript provides the requested evidence: Section 4.2 and Figure 5 show kinetic-energy spectra and mesoscale wavenumber comparisons (wavenumbers 50–500) demonstrating that generative nudging retains high-wavenumber power comparable to high-resolution ensemble members while matching ensemble-mean large-scale accuracy, with no visible smoothing or artifact introduction relative to baselines. We will revise the abstract to include a brief quantitative clause (e.g., “preserving mesoscale kinetic energy spectra within 5 % of high-resolution members”) to make this support explicit at the summary level. revision: yes

  2. Referee: [Abstract] Abstract: the assertion that Cast3 outperforms baselines 'across various dimensions' is load-bearing for the paper's contribution but is unsupported by any reported metrics, error bars, baseline specifications, or verification details, making it impossible to assess whether the claimed advantages are statistically significant or robust.

    Authors: The abstract summarizes results whose quantitative support appears in the main text: Table 1 reports RMSE and anomaly correlation with error bars across lead times; Figure 6 shows spectral fidelity metrics; Figure 8 and accompanying text give station-level surface verification; Figure 10 and Section 4.4 detail tropical-cyclone track and intensity errors. Baselines and verification protocols are defined in Section 3.3. To address the concern directly in the abstract, we will add one sentence citing representative improvements (e.g., “8–12 % lower RMSE at 5-day lead, superior spectral fidelity, and reduced TC intensity error”) while remaining within length limits. revision: yes

Circularity Check

0 steps flagged

No significant circularity in Cast3 derivation chain

full rationale

The paper presents Cast3 as a framework that integrates external NWP principles (cubed-sphere discretization, ensemble construction, posterior sampling) into a data-driven model via novel architectural choices and generative nudging. These elements are defined and motivated from established atmospheric modeling literature rather than reducing to self-referential fits or internal loops. Performance claims are benchmarked against independent deterministic and generative baselines across multiple verification dimensions, providing external falsifiability. No load-bearing step equates a prediction to its own fitted input or relies on unverified self-citation chains; the derivation remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 2 invented entities

Based solely on the abstract, the central claim rests on domain assumptions about translating NWP knowledge and introduces new concepts without full details on parameters or evidence.

axioms (1)
  • domain assumption NWP meta-knowledge distilled over decades can be systematically absorbed into data-driven models via specific grid and ensemble designs
    Invoked as the foundation for Cast3's design and claimed performance gains.
invented entities (2)
  • generative nudging no independent evidence
    purpose: Posterior-sampling strategy to distill full ensemble information into a single forecast with both large-scale accuracy and mesoscale realism
    New strategy introduced to combine ensemble strengths.
  • structurally diverse super-ensembles no independent evidence
    purpose: To sample complementary biases from different grid discretizations for improved ensemble prediction
    New ensemble construction approach described.

pith-pipeline@v0.9.0 · 5523 in / 1279 out tokens · 48597 ms · 2026-05-08T19:26:45.971173+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

43 extracted references · 32 canonical work pages · 2 internal anchors

  1. [1]

    Christopher Bishop.Pattern Recognition and Machine Learning

    Lorenz, E.N.: Deterministic Nonperiodic Flow. Journal of the Atmospheric Sci- ences20(2), 130–141 (1963) https://doi.org/10.1175/1520-0469(1963)020⟨0130: DNF⟩2.0.CO;2

    work page doi:10.1175/1520-0469(1963)020 1963

  2. [2]

    Tellus A: Dynamic Meteorology and Oceanogra- phy2(4), 237–254 (1950) https://doi.org/10.3402/tellusa.v2i4.8607

    Charney, J.G., Fj¨ ortoft, R., Von Neumann, J.: Numerical Integration of the Barotropic Vorticity Equation. Tellus A: Dynamic Meteorology and Oceanogra- phy2(4), 237–254 (1950) https://doi.org/10.3402/tellusa.v2i4.8607

    work page doi:10.3402/tellusa.v2i4.8607 1950

  3. [3]

    Nature525(7567), 47–55 (2015) https://doi.org/10.1038/nature14956

    Bauer, P., Thorpe, A., Brunet, G.: The quiet revolution of numerical weather pre- diction. Nature525(7567), 47–55 (2015) https://doi.org/10.1038/nature14956

    work page doi:10.1038/nature14956 2015

  4. [4]

    Monthly Weather Review100(2), 136–144 (1972) https://doi.org/10.1175/1520-0493(1972)100⟨0136:CFAOTP⟩2.3

    Sadourny, R.: Conservative Finite-Difference Approximations of the Primitive Equations on Quasi-Uniform Spherical Grids. Monthly Weather Review100(2), 136–144 (1972) https://doi.org/10.1175/1520-0493(1972)100⟨0136:CFAOTP⟩2.3. CO;2

    work page doi:10.1175/1520-0493(1972)100 1972

  5. [5]

    Monthly Weather Review132(12), 2845–2863 (2004) https://doi.org/10.1175/ MWR2823.1

    Adcroft, A., Campin, J.-M., Hill, C., Marshall, J.: Implementation of an Atmo- sphere–Ocean General Circulation Model on the Expanded Spherical Cube. Monthly Weather Review132(12), 2845–2863 (2004) https://doi.org/10.1175/ MWR2823.1

    2004

  6. [6]

    Quarterly Journal of the Royal Meteorological Society 141(690), 1727–1739 (2015) https://doi.org/10.1002/qj.2474

    Marras, S., Kopera, M.A., Giraldo, F.X.: Simulation of shallow-water jets with a unified element-based continuous/discontinuous Galerkin model with grid flex- ibility on the sphere. Quarterly Journal of the Royal Meteorological Society 141(690), 1727–1739 (2015) https://doi.org/10.1002/qj.2474

    work page doi:10.1002/qj.2474 2015

  7. [7]

    Journal of Computational Physics227(1), 55–78 (2007) https://doi.org/10.1016/ j.jcp.2007.07.022

    Putman, W.M., Lin, S.-J.: Finite-volume transport on various cubed-sphere grids. Journal of Computational Physics227(1), 55–78 (2007) https://doi.org/10.1016/ j.jcp.2007.07.022

    2007

  8. [8]

    Journal of Advances in Modeling Earth Systems13(1), 2020–002280 (2021) https://doi.org/10.1029/2020MS002280

    Chen, X.: The lmars based shallow-water dynamical core on generic gnomonic cubed-sphere geometry. Journal of Advances in Modeling Earth Systems13(1), 2020–002280 (2021) https://doi.org/10.1029/2020MS002280

    work page doi:10.1029/2020ms002280 2020

  9. [9]

    Monthly Weather Review124(3), 529–547 (1996) https://doi.org/10.1175/ 1520-0493(1996)124⟨0529:SOASFA⟩2.0.CO;2

    Waldron, K.M., Paegle, J., Horel, J.D.: Sensitivity of a Spectrally Filtered and Nudged Limited-Area Model to Outer Model Options. Monthly Weather Review124(3), 529–547 (1996) https://doi.org/10.1175/ 1520-0493(1996)124⟨0529:SOASFA⟩2.0.CO;2

    1996

  10. [10]

    M., & Hallberg, R

    Von Storch, H., Langenberg, H., Feser, F.: A Spectral Nudging Technique for Dynamical Downscaling Purposes. Monthly Weather Review128(10), 3664–3673 (2000) https://doi.org/10.1175/1520-0493(2000)128⟨3664:ASNTFD⟩2.0.CO;2

    work page doi:10.1175/1520-0493(2000)128 2000

  11. [11]

    National Energy Research Scientific Computing Center (U.S.) (2015)

    Michalakes, J., Govett, M., Benson, R., Black, T., Juang, H.-M.H., Reinecke, A., Skamarock, B., Duda, M., Henderson, T., Madden, P., Mozdzynski, G., Vasic, R.: 24 AVEC Report: NGGPS Level-1 Benchmarks and Software Evaluation. National Energy Research Scientific Computing Center (U.S.) (2015). https://doi.org/10. 25923/ZQ35-B106

    2015

  12. [12]

    https://doi.org/ 10.25923/P0XE-MZ37

    National Energy Research Scientific Computing Center (U.S.): AVEC Report: NGGPS Phase-2 Benchmarks and Software Evaluation (2016). https://doi.org/ 10.25923/P0XE-MZ37

    work page doi:10.25923/p0xe-mz37 2016

  13. [13]

    Accurate medium-range global weather forecasting with 3d neural networks,

    Bi, K., Xie, L., Zhang, H., Chen, X., Gu, X., Tian, Q.: Accurate medium-range global weather forecasting with 3D neural networks. Nature619(7970), 533–538 (2023) https://doi.org/10.1038/s41586-023-06185-3

    work page doi:10.1038/s41586-023-06185-3 2023

  14. [14]

    npj Climate and Atmospheric Science6(1), 190 (2023) https://doi.org/10.1038/ s41612-023-00512-1

    Chen, L., Zhong, X., Zhang, F., Cheng, Y., Xu, Y., Qi, Y., Li, H.: FuXi: a cascade machine learning forecasting system for 15-day global weather forecast. npj Climate and Atmospheric Science6(1), 190 (2023) https://doi.org/10.1038/ s41612-023-00512-1

    2023

  15. [15]

    https://doi.org/10.1126/science.adi2336 arXiv:https://www.science.org/doi/pdf/10.1126/science.adi2336

    Lam, R., Sanchez-Gonzalez, A., Willson, M., Wirnsberger, P., Fortunato, M., Alet, F., Ravuri, S., Ewalds, T., Eaton-Rosen, Z., Hu, W., Merose, A., Hoyer, S., Holland, G., Vinyals, O., Stott, J., Pritzel, A., Mohamed, S., Battaglia, P.: Learning skillful medium-range global weather forecasting. Science382(6677), 1416–1421 (2023) https://doi.org/10.1126/sci...

    work page doi:10.1126/science.adi2336 2023

  16. [16]

    Lang, S., Alexe, M., Chantry, M., Dramsch, J., Pinault, F., Raoult, B., Clare, M.C.A., Lessig, C., Maier-Gerber, M., Magnusson, L., Bouall` egue, Z.B., Nemesio, A.P., Dueben, P.D., Brown, A., Pappenberger, F., Rabier, F.: AIFS – ECMWF’s data-driven forecasting system. arXiv. arXiv:2406.01465 (2024). https://doi.org/ 10.48550/arXiv.2406.01465

    work page doi:10.48550/arxiv.2406.01465 2024

  17. [17]

    Alet, F., Price, I., El-Kadi, A., Masters, D., Markou, S., Andersson, T.R., Stott, J., Lam, R., Willson, M., Sanchez-Gonzalez, A., Battaglia, P.: Skillful joint prob- abilistic weather forecasting from marginals. arXiv. arXiv:2506.10772 (2025). https://doi.org/10.48550/arXiv.2506.10772

    work page doi:10.48550/arxiv.2506.10772 2025

  18. [18]

    Bonev, B., Kurth, T., Mahesh, A., Bisson, M., Kossaifi, J., Kashinath, K., Anand- kumar, A., Collins, W.D., Pritchard, M.S., Keller, A.: FourCastNet 3: A geometric approach to probabilistic machine-learning weather forecasting at scale. arXiv. arXiv:2507.12144 (2025). https://doi.org/10.48550/arXiv.2507.12144

    work page doi:10.48550/arxiv.2507.12144 2025

  19. [19]

    Communications Earth & Environment6(1), 518 (2025) https://doi.org/10.1038/s43247-025-02502-y

    Chen, K., Han, T., Ling, F., Gong, J., Bai, L., Wang, X., Luo, J.-J., Fei, B., Zhang, W., Chen, X., Ma, L., Zhang, T., Su, R., Ci, Y., Li, B., Yang, X., Ouyang, W.: The operational medium-range deterministic weather forecasting can be extended beyond a 10-day lead time. Communications Earth & Environment6(1), 518 (2025) https://doi.org/10.1038/s43247-025-02502-y

    work page doi:10.1038/s43247-025-02502-y 2025

  20. [20]

    npj Climate and Atmospheric Science8(1), 377 (2025) 25 https://doi.org/10.1038/s41612-025-01255-x

    Nai, C., Chen, X., Yang, S., Xiao, Z., Pan, B.: Boosting weather forecast via generative superensemble. npj Climate and Atmospheric Science8(1), 377 (2025) 25 https://doi.org/10.1038/s41612-025-01255-x

    work page doi:10.1038/s41612-025-01255-x 2025

  21. [21]

    Nature637, 84–90 (2025) https://doi.org/10.1038/s41586-024-08252-9

    Price, I., Sanchez-Gonzalez, A., Alet, F., Andersson, T.R., El-Kadi, A., Masters, D., Ewalds, T., Stott, J., Mohamed, S., Battaglia, P., Lam, R., Willson, M.: Probabilistic weather forecasting with machine learning. Nature637(8044), 84–90 (2025) https://doi.org/10.1038/s41586-024-08252-9

    work page doi:10.1038/s41586-024-08252-9 2025

  22. [22]

    Kossaifi, J., Kovachki, N., Mardani, M., Leibovici, D., Ravuri, S., Shokar, I., Calvello, E., Abbas, M.S., Harrington, P., Subramaniam, A., Brenowitz, N., Bonev, B., Byeon, W., Kreis, K., Durran, D., Vahdat, A., Pritchard, M., Kautz, J.: Demystifying Data-Driven Probabilistic Medium-Range Weather Forecasting. arXiv. arXiv:2601.18111 (2026). https://doi.or...

    work page doi:10.48550/arxiv.2601.18111 2026

  23. [23]

    npj Artificial Intelligence2(1), 18 (2026) https://doi.org/10.1038/s44387-026-00073-7

    Lang, S., Alexe, M., Clare, M.C.A., Roberts, C., Adewoyin, R., Ben Bouall` egue, Z., Chantry, M., Dramsch, J., Dueben, P.D., Hahner, S., Maciel, P., Prieto- Nemesio, A., O’Brien, C., Pinault, F., Polster, J., Raoult, B., Tietsche, S., Leutbecher, M.: AIFS-CRPS: ensemble forecasting using a model trained with a loss function based on the continuous ranked ...

    work page doi:10.1038/s44387-026-00073-7 2026

  24. [24]

    Bonavita, On some limitations of current machine learning weather prediction models

    Bonavita, M.: On Some Limitations of Current Machine Learning Weather Pre- diction Models. Geophysical Research Letters51(12), 2023–107377 (2024) https: //doi.org/10.1029/2023GL107377

    work page doi:10.1029/2023gl107377 2023

  25. [25]

    WeatherBench 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., Chantry, M., Ben Bouallegue, Z., Dueben, P., Bromberg, C., Sisk, J., Barrington, L., Bell, A., Sha, F.: Weather- Bench 2: A Benchmark for the Next Generation of Data-Driven Global Weather Models. Journal of Advances in Model...

    work page doi:10.1029/2023ms004019 2023

  26. [26]

    Polichtchouk, I., Lang, S., Lock, S.-J., Maier-Gerber, M., Dueben, P.: Hybrid ensemble forecasting combining physics-based and machine-learning predictions through spectral nudging. arXiv. arXiv:2603.05570 (2026). https://doi.org/10. 48550/arXiv.2603.05570

    work page arXiv 2026

  27. [27]

    Weather and Forecasting40(9), 1749–1771 (2025) https://doi.org/10.1175/WAF-D-24-0139.1

    Husain, S.Z., Separovic, L., Caron, J.-F., Aider, R., Buehner, M., Chamberland, S., Lapalme, E., McTaggart-Cowan, R., Subich, C., Vaillancourt, P.A., Yang, J., Zadra, A.: Leveraging Data-Driven Weather Models for Improving Numerical Weather Prediction Skill through Large-Scale Spectral Nudging. Weather and Forecasting40(9), 1749–1771 (2025) https://doi.or...

    work page doi:10.1175/waf-d-24-0139.1 2025

  28. [28]

    Hersbach, B

    Hersbach, H., Bell, B., Berrisford, P., Hirahara, S., Hor´ anyi, A., Mu˜ noz-Sabater, J., Nicolas, J., Peubey, C., Radu, R., Schepers, D., Simmons, A., Soci, C., Abdalla, S., Abellan, X., Balsamo, G., Bechtold, P., Biavati, G., Bidlot, J., Bonavita, M., De Chiara, G., Dahlgren, P., Dee, D., Diamantakis, M., Dragani, R., Flemming, J., Forbes, R., Fuentes, ...

    work page doi:10.1002/qj.3803 1999

  29. [29]

    Journal of Climate29(11), 4293–4314 (2016) https://doi.org/10.1175/JCLI-D-15-0389.1

    Harris, L.M., Lin, S.-J., Tu, C.: High-Resolution Climate Simulations Using GFDL HiRAM with a Stretched Global Grid. Journal of Climate29(11), 4293–4314 (2016) https://doi.org/10.1175/JCLI-D-15-0389.1

    work page doi:10.1175/jcli-d-15-0389.1 2016

  30. [30]

    Monthly Weather Review125(12), 3297–3319 (1997) https://doi.org/10.1175/ 1520-0493(1997)125⟨3297:EFANAT⟩2.0.CO;2

    Toth, Z., Kalnay, E.: Ensemble Forecasting at NCEP and the Breeding Method. Monthly Weather Review125(12), 3297–3319 (1997) https://doi.org/10.1175/ 1520-0493(1997)125⟨3297:EFANAT⟩2.0.CO;2

    1997

  31. [31]

    In: Advances in Neural Information Processing Systems, vol

    Ho, J., Jain, A., Abbeel, P.: Denoising diffusion probabilistic models. In: Advances in Neural Information Processing Systems, vol. 33, pp. 6840–6851 (2020)

    2020

  32. [32]

    Song, Y., Sohl-Dickstein, J., Kingma, D.P., Kumar, A., Ermon, S., Poole, B.: Score-Based Generative Modeling through Stochastic Differential Equations. arXiv. arXiv:2011.13456 (2021). https://doi.org/10.48550/arXiv.2011.13456

    work page Pith review doi:10.48550/arxiv.2011.13456 2011

  33. [33]

    Chung, H., Kim, J., Mccann, M.T., Klasky, M.L., Ye, J.C.: Diffusion Posterior Sampling for General Noisy Inverse Problems. arXiv. arXiv:2209.14687 (2024). https://doi.org/10.48550/arXiv.2209.14687

    work page internal anchor Pith review doi:10.48550/arxiv.2209.14687 2024

  34. [34]

    Meteorological Applications15(1), 163–169 (2008) https://doi.org/10.1002/met.57

    Roberts, N.: Assessing the spatial and temporal variation in the skill of precipita- tion forecasts from an NWP model. Meteorological Applications15(1), 163–169 (2008) https://doi.org/10.1002/met.57

    work page doi:10.1002/met.57 2008

  35. [35]

    K., Espeholt, L., Heek, J., Dehghani, M., Oliver, A., Salimans, T., et al

    Surcel, M., Zawadzki, I., Yau, M.K.: A Study on the Scale Dependence of the Pre- dictability of Precipitation Patterns. Journal of the Atmospheric Sciences72(1), 216–235 (2015) https://doi.org/10.1175/JAS-D-14-0071.1

    work page doi:10.1175/jas-d-14-0071.1 2015

  36. [36]

    npj Climate and Atmospheric Science (2026) https://doi.org/10.1038/ s41612-026-01380-1

    Kim, H., Ryu, J., Son, S.-W., Jeong, J.-H., Kim, H., Yoon, J.-H.: A spectral test of the butterfly effect and physical consistency in the diffusion-based GenCast’s ensembles. npj Climate and Atmospheric Science (2026) https://doi.org/10.1038/ s41612-026-01380-1

    2026

  37. [37]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Woo, S., Debnath, S., Hu, R., Chen, X., Liu, Z., Kweon, I.S., Xie, S.: Convnext v2: Co-designing and scaling convnets with masked autoencoders. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 16133–16142 (2023)

    2023

  38. [38]

    Salimans, T., Ho, J.: Progressive Distillation for Fast Sampling of Diffusion Models. arXiv. arXiv:2202.00512 (2022). https://doi.org/10.48550/arXiv.2202. 00512

    work page doi:10.48550/arxiv.2202 2022

  39. [39]

    Song, J., Meng, C., Ermon, S.: Denoising Diffusion Implicit Models. arXiv. 27 arXiv:2010.02502 (2022). https://doi.org/10.48550/arXiv.2010.02502

    work page Pith review doi:10.48550/arxiv.2010.02502 2010

  40. [40]

    Reviews of Geophysics54(2), 336–377 (2016) https://doi.org/10

    Krishnamurti, T.N., Kumar, V., Simon, A., Bhardwaj, A., Ghosh, T., Ross, R.: A review of multimodel superensemble forecasting for weather, seasonal climate, and hurricanes. Reviews of Geophysics54(2), 336–377 (2016) https://doi.org/10. 1002/2015RG000513

    2016

  41. [41]

    Beitr¨ age zur Physik der Atmosph¨ are50, 211–217 (1977)

    Schmidt, F.: Variable fine mesh in spectral global models. Beitr¨ age zur Physik der Atmosph¨ are50, 211–217 (1977)

    1977

  42. [42]

    Yu, Y., Xiong, W., Nie, W., Sheng, Y., Liu, S., Luo, J.: PixelDiT: Pixel Diffusion Transformers for Image Generation. arXiv. arXiv:2511.20645 (2026). https://doi. org/10.48550/arXiv.2511.20645

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2511.20645 2026

  43. [43]

    The International Journal of High Performance Computing Applications19(3), 213–223 (2005) https://doi.org/10.1177/1094342005056101 28

    Putman, W.M., Lin, S.-J., Shen, B.-W.: Cross-Platform Performance of a Portable Communication Module and the Nasa Finite Volume General Cir- culation Model. The International Journal of High Performance Computing Applications19(3), 213–223 (2005) https://doi.org/10.1177/1094342005056101 28

    work page doi:10.1177/1094342005056101 2005