Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

OMG-HD: A High-Resolution AI Weather Model for End-to-End Forecasts from Observations

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A model trained only on raw observations outperforms two operational weather systems for 12-hour surface forecasts across the US.

desk verdict A promising E2E AI forecast model whose headline claim is undercut by a circular RTMA evaluation; the idea is worth a serious referee but the evidence needs independent verification. read the letter →

arxiv 2412.18239 v1 pith:5BYDJ73D submitted 2024-12-24 physics.ao-ph cs.LG

Pengcheng Zhao , Jiang Bian , Zekun Ni , Weixin Jin , Jonathan Weyn , Zuliang Fang , Siqi Xiang , Haiyu Dong
show 4 more authors
Bin Zhang Hongyu Sun Kit Thambiratnam Qi Zhang
This is my paper · ORCID
classification physics.ao-phcs.LG PACS 92.60.Wc
keywords AIweatherpredictionend-to-endforecastingobservationaldataOMG-HDassimilationregionalRTMAsurfacevariables
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper claims that a kilometer-scale weather forecast can be made directly from raw observations—surface stations, radar, and satellite channels—without any numerical weather model or data assimilation step. It introduces OMG-HD, a two-block neural network that first converts six hours of heterogeneous observations into a gridded atmospheric state and then steps that state forward in time. Evaluated against RTMA and station data, OMG-HD reports lower RMSE than HRRR and IFS-HRES for 2-m temperature, wind speed, specific humidity, and surface pressure, with gains up to 48% for humidity. If correct, this would mean an operational forecast system could be updated in real time from incoming observations, bypassing the latency and information loss of traditional data assimilation.

What carries the argument

The central mechanism is the two-block architecture: an Assimilating Block built on Shifted-Window Transformer (Swin) layers fuses the past six hours of station, radar, satellite, time-encoding, and topography inputs into a gridded 26-channel initial state, with 6 channels supervised against RTMA surface variables and 20 free latent channels; a Forecasting Block using Adaptive Fourier Neural Operator (AFNO) layers then advances this state in 6-hour autoregressive steps, with a combined loss L1+L2+L3 supervising the analyzed state and the 6- and 12-hour forecasts against RTMA. This learned initial state plays the role of a data-assimilation analysis, constructed not by solving observation operators but by end-to-end training with the forecast objective.

What would settle it

Evaluate forecasts from OMG-HD and HRRR against raw station observations for all four variables at stations withheld from both training and the RTMA analysis; if the reported RMSE margin shrinks below statistical significance, the claimed advantage would be, at least in part, an artifact of the verification target.

Watch

Extended reading notes

Core claim

OMG-HD establishes, on the paper's own terms, that an end-to-end AI weather prediction model can learn to make kilometer-scale, twelve-hour forecasts directly from raw observations—station measurements, radar reflectivity, and satellite channels—without using any NWP output as input or as a training target. The model learns its own assimilation step in a dedicated Assimilating Block, mapping sparse heterogenous observations into a dense 26-channel gridded state whose hidden channels encode information such as upper-air conditions, and then makes autoregressive forecasts with a Fourier-based Forecasting Block. Compared against the operational HRRR and IFS-HRES baselines over CONUS, the paper reports average RMSE reductions of 13% for 2-m temperature, 17% for 10-m wind speed, 48% for 2-m specific humidity, and 32% for surface pressure, and shows the model remains competitive even when up to 50% of input stations are masked and on hold-out stations not seen during training.

Load-bearing premise

The results are measured against RTMA, the same analysis product OMG-HD was trained on, so the comparison assumes RTMA is an unbiased ground truth that does not systematically favor a model fitted to it.

Editorial extensions

If this is right

  • Short-range operational forecasts could be produced directly from new observations within minutes, without waiting for an NWP cycle with data assimilation.
  • AI weather prediction need not depend on reanalysis products for input or initialization; model updates could track instrument networks in real time.
  • The 20 latent channels in the analyzed state suggest a learned analysis can carry information about the atmosphere beyond the supervised surface variables.
  • If the result holds under independent verification, it sets a new baseline that operational short-range regional systems would be measured against.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural next test is to run the same architecture over a global domain with full satellite coverage, to see whether the learned analysis generalizes beyond CONUS and beyond 12-hour lead times.
  • Because OMG-HD was trained on RTMA and verified against RTMA, an independent evaluation against raw station observations—especially for surface pressure, which was partly replaced by RTMA values—would separate true forecast skill from fit to the analysis.
  • The strong humidity improvement suggests the satellite and radar channels carry moisture information that station-only inputs miss; an ablation isolating each input source could identify which channel drives the gain and guide future observation targeting.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper introduces OMG-HD, a high-resolution AI weather forecasting model that takes raw observations (surface stations, GOES-16 satellite channels, MRMS radar, topography, time encodings) over the CONUS region and produces 12-hour forecasts of 2-meter temperature, 10-meter wind speed, 2-meter specific humidity, and surface pressure. The architecture consists of an Assimilating Block, which converts sparse multi-source observations into a gridded latent state, and a Forecasting Block, which generates autoregressive 6-hour forecasts. The model is trained end-to-end with losses against RTMA labels. The evaluation compares OMG-HD against HRRR, IFS-HRES, and GFS using both RTMA-gridded verification and station observations, with additional experiments on hold-out stations, masked input stations, and two case studies of extreme events. The central claim is that OMG-HD outperforms operational NWP models at lead times up to 12 hours across CONUS.

Significance. If the reported results are sound, the paper would demonstrate a noteworthy proof of concept: an AI model trained and run directly from observations can compete with operational high-resolution NWP systems for short-range surface forecasting over a large domain. The architectural choices are described in detail, the data sources and training procedure are specified, and the masking ablation addresses an operationally relevant failure mode. The case studies are illustrative and the authors are candid in the discussion about limitations such as the CONUS-only scope, the use of RTMA labels, and the lack of boundary conditions. However, the main significance is conditional: the headline comparison is measured against RTMA, which is also the training target, so the reported margins over HRRR and IFS-HRES are not an independent measure of forecast skill. The hold-out station evaluation does not fully resolve this problem, as the station observations used for verification are part of the same observational network that feeds the RTMA labels.

major comments (3)
  1. [Section 3.1 and Fig. 2a] The primary evaluation in Fig. 2a verifies forecasts against RTMA, and Section A.1 states that RTMA is used to derive the training labels. Because the optimization losses L1, L2, and L3 are computed directly against RTMA, the RMSE of OMG-HD against RTMA is not an unbiased comparison with HRRR, IFS-HRES, or GFS, none of which were trained on RTMA. The reported 13%, 17%, 48%, and 32% improvements over HRRR are therefore partly a measure of how well OMG-HD fits the RTMA analysis, not an independent measure of forecast quality. The paper should either provide an evaluation against a truth dataset that was not used as a training label, or explicitly reframe the claim as skill relative to the RTMA analysis rather than a general statement of forecast superiority.
  2. [Section 3.3.1 and Fig. 2c, with Section A.1] The hold-out station test does not break the label-verification coupling. The hold-out stations are selected from the same ISD/Synoptic station network that supplies the input observations, and RTMA assimilates those station reports into the gridded labels used for training. Excluding a station from the direct station-label loss does not remove its influence from the RTMA target at that location, because the RTMA analysis was generated with those very observations. Furthermore, for the hold-out evaluation the variable Q is derived from RH, T, and SP, and Section A.1 states that RTMA SP is used as a substitute for station SP, which re-introduces the training analysis into the verification. Thus Fig. 2c does not provide the independent out-of-sample evidence needed to support the generalization claim.
  3. [Section 3.2 and Fig. 5] The comparison of the embedded assimilation state in Fig. 5 is also evaluated against RTMA labels, so it inherits the same circularity as Fig. 2a. The claim that OMG-HD produces a more accurate initial state than the baselines is essentially a statement that the model's gridded output, trained to match RTMA, is closer to RTMA than the NWP model backgrounds. This may be true, but it does not demonstrate that the latent state is meteorologically superior in an independent sense; the figure should be interpreted with that caveat or replaced with a comparison using independent observations.
minor comments (5)
  1. [Abstract and Section 3.1] The abstract says 'up to a 13% improvement' for 2-meter temperature, while Section 3.1 says the 13% is the performance gain 'averaged over lead times.' These are different quantities, and since the RMSE varies with lead time in Fig. 2a, the abstract should state whether the percentage is a lead-time-averaged value or a maximum, and should identify the lead times at which the improvement is not achieved.
  2. [Section A.2.1] The subsection header contains a typo: 'F orecasting block' should read 'Forecasting block.'
  3. [Section A.2.1 and Table 2] The text refers to 'Digital Evaluation Models (DEMs)' in Section A.1; this should be 'Digital Elevation Models.'
  4. [Figure 2] The main RMSE plots show no confidence intervals or significance tests. Given that the headline claim is about outperforming operational NWP models, reporting uncertainty intervals or at least the number of verification times would help assess whether the differences are robust.
  5. [Section A.1] The paper says 'The labels used to train OMG-HD are derived from the Real-Time Mesoscale Analysis (RTMA)' and later notes that RTMA has 'small discrepancy from raw observations.' This is a useful caveat, but the evaluation sections should reference it directly when interpreting the RTMA-based results, since the discrepancy is exactly what makes the training-verification coupling consequential.

Circularity Check

3 steps flagged · score 6.0 of 10

Primary evaluation is trained-target versus trained-target: OMG-HD is trained on RTMA labels and its headline RMSE is verified against RTMA; the hold-out station Q metric additionally substitutes RTMA SP, re-importing the training label.

  1. fitted input called prediction [Section 2 ('labels are derived from RTMA'), Section 3.1 ('measured against the RTMA as truth'), Section A.2.2 (L1/L2/L3 versus RTMA)]
    "During the training phase, the labels are derived from the Real-Time Mesoscale Analysis (RTMA) [24], which is a dataset specifically designed to align more closely with observations than traditional data assimilation schemes [25]. ... Fig. 2a shows the root mean squared error (RMSE) of the models’ forecasts as a function of lead time, measured against the RTMA as truth."

    All three training losses (L1, L2, L3) are computed against RTMA labels, so the model's parameters are directly optimized to minimize error on RTMA fields. The headline grid RMSE and the abstract's 13/17/48/32% improvements are then computed with the very same RTMA product as truth, while the operational baselines are not fitted to RTMA. The comparison therefore measures, in large part, how well the model matches its own training target; the reported advantage over IFS-HRES and HRRR is statistically forced and is not an independent verification of forecast skill.

  2. fitted input called prediction [Section 3.3.1 and Section A.1 (hold-out station setup)]
    "During the training phase, the labels are derived from the Real-Time Mesoscale Analysis (RTMA) [24], which is a dataset specifically designed to align more closely with observations than traditional data assimilation schemes [25]. ... we use a hold-out mechanism [21] to isolate specific stations from the training dataset, reserving them as hold-out stations to test how well the model predicts unseen data."

    The paper calls the hold-out test 'unseen data' because the stations are removed from the training inputs, but the training labels are RTMA analyses that are built from surface observations and are explicitly described as aligning more closely with observations. Nothing in the paper excludes the held-out station reports from the RTMA labels used as targets; consequently the model can have been trained to reproduce the very station values used for verification. The hold-out evaluation is therefore not a clean out-of-sample test and does not break the RTMA-label coupling of the primary comparison.

1 more flagged steps
  1. self definitional [Section A.1, last paragraph (Q conversion for station and hold-out evaluation)]
    "Similarly, for the hold-out evaluation in Fig. 2c, the same conversion process is applied; however, SP from RTMA is used as a substitute for station SP."

    The hold-out Q score is not an independent station observation: it is computed from station RH and T with RTMA SP substituted for the missing station SP. RTMA SP is one of the grid variables the model is trained to predict (the L1/L2/L3 targets). So the 'station' verification for Q is partly defined as a function of the model's own training target; any forecast matching RTMA SP will score well on this metric by construction.

full rationale

The central circular step is explicit: OMG-HD is trained end-to-end against RTMA labels, and the primary grid evaluation (Fig. 2a) and the abstract's improvement percentages are verified against the same RTMA product. Since the operational baselines were not fitted to RTMA, the reported margin is partly a property of the training objective rather than an independent measure of short-range forecast skill. The station-based evaluation was intended to provide an independent check, but it does not fully escape the coupling: the hold-out Q metric uses RTMA SP as a substitute for station SP, and the paper does not establish that held-out station reports were excluded from the RTMA labels used as training targets. The architecture and end-to-end training procedure themselves are not circular, and no load-bearing self-citation chain is present. A score of 6 reflects that the central abstract claim rests substantially on a trained-target versus trained-target comparison, while the station, hold-out, and case-study results provide some independent, if incomplete, evidence.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The central claim depends on a large trained model whose parameters are fitted to RTMA labels, plus a set of domain assumptions about RTMA reliability and station representativeness. No new physical entities are introduced.

free parameters (1)
  • Trained neural network weights = Not disclosed
    The forecast skill depends on millions of parameters fitted to RTMA labels during training; the paper reports no model weights or checkpoints.
assumptions (4)
  • domain assumption RTMA is a reliable approximation of true surface conditions for training and evaluation.
    The model is trained on RTMA and the primary verification in Fig. 2a uses the same product as truth (Sections 2, 3.1). If RTMA is biased, those biases are learned by the model and inflate its apparent skill.
  • domain assumption Station observations after mapping to grid cells and averaging provide an unbiased evaluation target.
    The station-level evaluation in Section 3.1 uses about 1,600 high-quality stations, and hold-out stations are excluded from training. However, the mapping and averaging (Section A.1) could systematically smooth extremes, and Q is derived from RH, T, and SP rather than measured directly.
  • ad hoc to paper The 20 latent channels in the Assimilating Block can encode useful upper-air information without explicit supervision.
    The paper postulates that these channels capture 'possible upper-air dynamics' (Section A.2.1), but there is no direct evidence or supervision to guarantee this.
  • domain assumption The temporal split (train 2018-2022, test 2023) ensures the model is not overfit to the test period.
    Standard practice, but the paper does not analyze potential regime shifts or distribution drift that could affect 2023 skill.

how reviews work

0 comments
Cite this review

Pith. "Pith review of OMG-HD: A High-Resolution AI Weather Model for End-to-End Forecasts from Observations." pith.science (2026). https://pith.science/paper/5BYDJ73D

@misc{pith2026241218239,
  author       = {Pith},
  title        = {Pith review of: OMG-HD: A High-Resolution AI Weather Model for End-to-End Forecasts from Observations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5BYDJ73D}},
  note         = {Machine review of arXiv:2412.18239}
}
read the original abstract

In recent years, Artificial Intelligence Weather Prediction (AIWP) models have achieved performance comparable to, or even surpassing, traditional Numerical Weather Prediction (NWP) models by leveraging reanalysis data. However, a less-explored approach involves training AIWP models directly on observational data, enhancing computational efficiency and improving forecast accuracy by reducing the uncertainties introduced through data assimilation processes. In this study, we propose OMG-HD, a novel AI-based regional high-resolution weather forecasting model designed to make predictions directly from observational data sources, including surface stations, radar, and satellite, thereby removing the need for operational data assimilation. Our evaluation shows that OMG-HD outperforms both the European Centre for Medium-Range Weather Forecasts (ECMWF)'s high-resolution operational forecasting system, IFS-HRES, and the High-Resolution Rapid Refresh (HRRR) model at lead times of up to 12 hours across the contiguous United States (CONUS) region. We achieve up to a 13% improvement on RMSE for 2-meter temperature, 17% on 10-meter wind speed, 48% on 2-meter specific humidity, and 32% on surface pressure compared to HRRR. Our method shows that it is possible to use AI-driven approaches for rapid weather predictions without relying on NWP-derived weather fields as model input. This is a promising step towards using observational data directly to make operational forecasts with AIWP models.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Deep Learning and Foundation Models for Weather Prediction: A Survey

    cs.LG 2025-01 conditional novelty 4.0 of 10

    A survey that organizes deep learning weather prediction models into three training paradigms: deterministic, generative, and pre-train-fine-tune.

Reference graph

Works this paper leans on

50 extracted references · 33 canonical work pages · cited by 1 Pith paper

  1. [1]

    Nature 525(7567), 47–55 (2015)

    Bauer, P., Thorpe, A., Brunet, G.: The quiet revolution of numerical weather prediction. Nature 525(7567), 47–55 (2015)

    work page 2015

  2. [2]

    In: Pro- ceedings of the Platform for Advanced Scientific Computing Conference

    Kurth, T., Subramanian, S., Harrington, P., Pathak, J., Mardani, M., Hall, D., Miele, A., Kashinath, K., Anandkumar, A.: Fourcastnet: Accelerating global high- resolution weather forecasting using adaptive fourier neural operators. In: Pro- ceedings of the Platform for Advanced Scientific Computing Conference. PASC ’23. Association for Computing Machinery...

    arXiv 2023

  3. [3]

    arXiv preprint arXiv:2304.02948 (2023)

    Chen, K., Han, T., Gong, J., Bai, L., Ling, F., Luo, J.-J., Chen, X., Ma, L., Zhang, T., Su, R., et al.: Fengwu: Pushing the skillful global medium-range weather forecast beyond 10 days lead. arXiv preprint arXiv:2304.02948 (2023)

    arXiv 2023

  4. [4]

    npj Climate and Atmospheric Science 6(1), 190 (2023)

    Chen, L., Zhong, X., Zhang, F., Cheng, Y., Xu, Y., Qi, Y., Li, H.: Fuxi: A cas- cade machine learning forecasting system for 15-day global weather forecast. npj Climate and Atmospheric Science 6(1), 190 (2023)

    work page 2023

  5. [5]

    Nature 619(7970), 533–538 (2023)

    Bi, K., Xie, L., Zhang, H., Chen, X., Gu, X., Tian, Q.: Accurate medium-range global weather forecasting with 3d neural networks. Nature 619(7970), 533–538 (2023)

    2023

  6. [6]

    Science 382(6677), 1416–1421 (2023)

    Lam, R., Sanchez-Gonzalez, A., Willson, M., Wirnsberger, P., Fortunato, M., Alet, F., Ravuri, S., Ewalds, T., Eaton-Rosen, Z., Hu, W., et al.: Learning skillful medium-range global weather forecasting. Science 382(6677), 1416–1421 (2023)

    2023

  7. [7]

    arXiv preprint arXiv:2405.13063 (2024)

    Bodnar, C., Bruinsma, W.P., Lucic, A., Stanley, M., Brandstetter, J., Garvan, P., Riechert, M., Weyn, J., Dong, H., Vaughan, A., et al.: Aurora: A foundation model of the atmosphere. arXiv preprint arXiv:2405.13063 (2024)

    arXiv 2024

  8. [8]

    Quarterly Journal of the Royal Meteorological Society 146(730), 1999–2049 (2020)

    Hersbach, H., Bell, B., Berrisford, P., Hirahara, S., Hor´ anyi, A., Mu˜ noz-Sabater, J., Nicolas, J., Peubey, C., Radu, R., Schepers, D., et al.: The era5 global reanal- ysis. Quarterly Journal of the Royal Meteorological Society 146(730), 1999–2049 (2020)

    2020

Show all 50 references
  1. [9]

    part ii: Recent years

    Eyre, J., Bell, W., Cotton, J., English, S., Forsythe, M., Healy, S., Pavelin, E.: Assimilation of satellite data in numerical weather prediction. part ii: Recent years. Quarterly Journal of the Royal Meteorological Society 148(743), 521–556 (2022)

    2022

  2. [10]

    Quarterly Journal of the Royal Meteorological Society 144(713), 1218–1256 (2018)

    Gustafsson, N., Janji´ c, T., Schraff, C., Leuenberger, D., Weissmann, M., Reich, H., Brousseau, P., Montmerle, T., Wattrelot, E., Buˇ c´ anek, A., et al.: Survey 13 of data assimilation methods for convective-scale numerical weather prediction at operational centres. Quarterl...

    2018

  3. [11]

    Journal of Advances in Modeling Earth Systems 14(6), 2022–003016 (2022)

    Laloyaux, P., Kurth, T., Dueben, P.D., Hall, D.: Deep learning to estimate model biases in an operational nwp assimilation system. Journal of Advances in Modeling Earth Systems 14(6), 2022–003016 (2022)

    2022

  4. [12]

    Bulletin of the American Meteorological Society 104(4), 756–767 (2023)

    Valmassoi, A., Keller, J.D., Kleist, D.T., English, S., Ahrens, B., ˇDur´ an, I.B., Bauernschubert, E., Bosilovich, M.G., Fujiwara, M., Hersbach, H., et al.: Current challenges and future directions in data assimilation and reanalysis. Bulletin of the American Meteorological S...

    2023

  5. [13]

    Atmospheric Science Letters 24(1), 1130 (2023)

    Hu, G., Dance, S.L., Bannister, R.N., Chipilski, H.G., Guillet, O., Macpherson, B., Weissmann, M., Yussouf, N.: Progress, challenges, and future steps in data assim- ilation for convection-permitting numerical weather prediction: Report on the virtual meeting held on 10 and 12...

    2023

  6. [14]

    arXiv preprint arXiv:2408.09695 (2024)

    Fu, Y., Wang, F., Shao, Z., Yu, C., Li, Y., Chen, Z., An, Z., Xu, Y.: Lightweather: Harnessing absolute positional encoding to efficient and scalable global weather forecasting. arXiv preprint arXiv:2408.09695 (2024)

    2024

  7. [15]

    arXiv preprint arXiv:2312.12455 (2023)

    Xiao, Y., Bai, L., Xue, W., Chen, K., Han, T., Ouyang, W.: Fengwu-4dvar: Cou- pling the data-driven weather forecasting model with 4d variational assimilation. arXiv preprint arXiv:2312.12455 (2023)

    2023 arXiv

  8. [16]

    Nature Machine Intelligence 5(6), 602–611 (2023)

    Wu, H., Zhou, H., Long, M., Wang, J.: Interpretable weather forecasting for worldwide stations with a unified deep model. Nature Machine Intelligence 5(6), 602–611 (2023)

    2023

  9. [17]

    arXiv preprint arXiv:2404.00411 (2024)

    Vaughan, A., Markou, S., Tebbutt, W., Requeima, J., Bruinsma, W.P., Anders- son, T.R., Herzog, M., Lane, N.D., Hosking, J.S., Turner, R.E.: Aardvark weather: end-to-end data-driven weather forecasting. arXiv preprint arXiv:2404.00411 (2024)

    2024 arXiv

  10. [18]

    arXiv preprint arXiv:2407.15586 (2024)

    McNally, A., Lessig, C., Lean, P., Boucher, E., Alexe, M., Pinnington, E., Chantry, M., Lang, S., Burrows, C., Chrust, M., et al.: Data driven weather forecasts trained and initialised directly from observations. arXiv preprint arXiv:2407.15586 (2024)

    2024 arXiv

  11. [19]

    arXiv preprint arXiv:2312.12462 (2023)

    Chen, K., Bai, L., Ling, F., Ye, P., Chen, T., Chen, K., Han, T., Ouyang, W.: Towards an end-to-end artificial intelligence driven global weather forecasting system. arXiv preprint arXiv:2312.12462 (2023)

    2023

  12. [20]

    arXiv preprint arXiv:2408.05472 (2024)

    Sun, X., Zhong, X., Xu, X., Huang, Y., Li, H., Feng, J., Han, W., Wu, L., Qi, Y.: Fuxi weather: An end-to-end machine learning weather data assimilation and 14 forecasting system. arXiv preprint arXiv:2408.05472 (2024)

    2024

  13. [21]

    arXiv preprint arXiv:2306.06079 (2023)

    Andrychowicz, M., Espeholt, L., Li, D., Merchant, S., Merose, A., Zyda, F., Agrawal, S., Kalchbrenner, N.: Deep learning for day forecasts from sparse observations. arXiv preprint arXiv:2306.06079 (2023)

    2023 arXiv

  14. [22]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp

    Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., Guo, B.: Swin trans- former: Hierarchical vision transformer using shifted windows. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 10012–10022 (2021)

    2021

  15. [23]

    In: International Conference on Learning Representations (2021)

    Guibas, J., Mardani, M., Li, Z., Tao, A., Anandkumar, A., Catanzaro, B.: Effi- cient token mixing for transformers via adaptive fourier neural operators. In: International Conference on Learning Representations (2021)

    2021

  16. [24]

    Weather and Forecasting26(5), 593–612 (2011)

    De Pondeca, M.S., Manikin, G.S., DiMego, G., Benjamin, S.G., Parrish, D.F., Purser, R.J., Wu, W.-S., Horel, J.D., Myrick, D.T., Lin, Y., et al.: The real- time mesoscale analysis at noaa’s national centers for environmental prediction: current status and development. Weather a...

    2011

  17. [25]

    Weather and Forecasting 35(3), 977–996 (2020)

    Morris, M.T., Carley, J.R., Col´ on, E., Gibbs, A., De Pondeca, M.S., Levine, S.: A quality assessment of the real-time mesoscale analysis (rtma) for aviation. Weather and Forecasting 35(3), 977–996 (2020)

    2020

  18. [26]

    ECMWF Newsletter 147, 14–19 (2016)

    H´ olm, E., Forbes, R., Lang, S., Magnusson, L., Malardel, S.: New model cycle brings higher resolution. ECMWF Newsletter 147, 14–19 (2016)

    2016

  19. [27]

    part i: Motivation and system description

    Dowell, D.C., Alexander, C.R., James, E.P., Weygandt, S.S., Benjamin, S.G., Manikin, G.S., Blake, B.T., Brown, J.M., Olson, J.B., Hu, M., et al.: The high- resolution rapid refresh (hrrr): An hourly updating convection-allowing forecast model. part i: Motivation and system des...

    2022

  20. [28]

    part ii: Forecast performance

    James, E.P., Alexander, C.R., Dowell, D.C., Weygandt, S.S., Benjamin, S.G., Manikin, G.S., Brown, J.M., Olson, J.B., Hu, M., Smirnova, T.G.,et al.: The high- resolution rapid refresh (hrrr): an hourly updating convection-allowing forecast model. part ii: Forecast performance. ...

    2022

  21. [29]

    STI Climate Bulletin, 159–163 (2019)

    Zhou, X., Zhu, Y., Fu, B., Hou, D., Peng, J., Luo, Y., Li, W.: The development of next ncep global ensemble forecast system. STI Climate Bulletin, 159–163 (2019)

    2019

  22. [30]

    Monthly Weather Review 151(4), 897–911 (2023) 15

    Bouall` egue, Z.B., Cooper, F., Chantry, M., D¨ uben, P., Bechtold, P., Sandu, I.: Statistical modeling of 2-m temperature and 10-m wind speed forecast errors. Monthly Weather Review 151(4), 897–911 (2023) 15

    2023

  23. [31]

    In: Artificial Intelligence, pp

    Bewoor, L.A., Bewoor, A., Kumar, R.: Artificial intelligence for weather forecast- ing. In: Artificial Intelligence, pp. 231–239. CRC Press, ??? (2021)

    2021

  24. [32]

    Tellus 21(3), 289–307 (1969)

    Lorenz, E.N.: The predictability of a flow which possesses many scales of motion. Tellus 21(3), 289–307 (1969)

    1969

  25. [33]

    Science 363(6425), 342–344 (2019)

    Alley, R.B., Emanuel, K.A., Zhang, F.: Advances in weather prediction. Science 363(6425), 342–344 (2019)

    2019

  26. [34]

    arXiv preprint arXiv:2409.13598 (2024)

    Schmude, J., Roy, S., Trojak, W., Jakubik, J., Civitarese, D.S., Singh, S., Kuehn- ert, J., Ankur, K., Gupta, A., Phillips, C.E., et al.: Prithvi wxc: Foundation model for weather and climate. arXiv preprint arXiv:2409.13598 (2024)

    2024 arXiv

  27. [35]

    Nature Machine Intelligence 5(12), 1340–1341 (2023)

    Gromov, A.: Deconstructing the generalization gap. Nature Machine Intelligence 5(12), 1340–1341 (2023)

    2023

  28. [36]

    Lad, V., Gurnee, W., Tegmark, M.: The remarkable robustness of llms: Stages of inference? arXiv preprint arXiv:2406.19384 (2024)

    2024 arXiv

  29. [37]

    Technical report (2014)

    Lui, M.-C.M.: Complete decoding and reporting of aviation routine weather reports (metars). Technical report (2014)

    2014

  30. [38]

    Nature Communications 15(1), 6931 (2024)

    Bontempi, D., Nuernberg, L., Pai, S., Krishnaswamy, D., Thiriveedhi, V., Hosny, A., Mak, R.H., Farahani, K., Kikinis, R., Fedorov, A., et al.: End-to-end repro- ducible ai pipelines in radiology using the cloud. Nature Communications 15(1), 6931 (2024)

    2024

  31. [39]

    Nature Reviews Materials 7(11), 887–907 (2022)

    Ates, H.C., Nguyen, P.Q., Gonzalez-Macia, L., Morales-Narv´ aez, E., G¨ uder, F., Collins, J.J., Dincer, C.: End-to-end design of wearable sensors. Nature Reviews Materials 7(11), 887–907 (2022)

    2022

  32. [40]

    International Journal of Computer Vision, 1–23 (2024)

    Liao, B., Chen, S., Zhang, Y., Jiang, B., Zhang, Q., Liu, W., Huang, C., Wang, X.: Maptrv2: An end-to-end framework for online vectorized hd map construction. International Journal of Computer Vision, 1–23 (2024)

    2024

  33. [41]

    arXiv preprint arXiv:2409.09371 (2024)

    Jin, W., Weyn, J., Zhao, P., Xiang, S., Bian, J., Fang, Z., Dong, H., Sun, H., Thambiratnam, K., Zhang, Q.: Weatherreal: A benchmark based on in- situ observations for evaluating weather models. arXiv preprint arXiv:2409.09371 (2024)

    2024 arXiv

  34. [42]

    Bulletin of the American Meteorological Society 92(6), 704–708 (2011)

    Smith, A., Lott, N., Vose, R.: The integrated surface database: Recent develop- ments and partnerships. Bulletin of the American Meteorological Society 92(6), 704–708 (2011)

    2011

  35. [43]

    In: Earth Observing Systems XXIV, vol

    Tan, B., Dellomo, J., Wolfe, R., Reth, A.: Goes-16 and goes-17 abi inr assessment. In: Earth Observing Systems XXIV, vol. 11127, pp. 290–301 (2019). SPIE

    2019

  36. [44]

    Bulletin of the American Meteorological Society 97(4), 621–638 (2016)

    Zhang, J., Howard, K., Langston, C., Kaney, B., Qi, Y., Tang, L., Grams, H., 16 Wang, Y., Cocks, S., Martinaitis, S., et al.: Multi-radar multi-sensor (mrms) quantitative precipitation estimation: Initial operating capabilities. Bulletin of the American Meteorological Society ...

    2016

  37. [45]

    University of Mysore (2017)

    Balasubramanian, A.: Digital elevation model (dem) in gis. University of Mysore (2017)

    2017

  38. [46]

    Remote sensing of environment 257, 112349 (2021)

    Cawse-Nicholson, K., Townsend, P.A., Schimel, D., Assiri, A.M., Blake, P.L., Buongiorno, M.F., Campbell, P., Carmon, N., Casey, K.A., Correa-Pab´ on, R.E., et al.: Nasa’s surface biology and geology designated observable: A perspective on surface imaging algorithms. Remote sen...

    2021

  39. [47]

    Weather and Forecasting 26(5), 593–612 (2011) https://doi.org/10.1175/ W AF-D-10-05037.1

    Pondeca, M.S.F.V.D., Manikin, G.S., DiMego, G., Benjamin, S.G., Parrish, D.F., Purser, R.J., Wu, W.-S., Horel, J.D., Myrick, D.T., Lin, Y., Aune, R.M., Keyser, D., Colman, B., Mann, G., Vavra, J.: The real-time mesoscale analysis at noaa’s national centers for environmental pr...

    2011

  40. [48]

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

    Liu, Z., Hu, H., Lin, Y., Yao, Z., Xie, Z., Wei, Y., Ning, J., Cao, Y., Zhang, Z., Dong, L., et al.: Swin transformer v2: Scaling up capacity and resolution. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 12009–12019 (2022)

    2022

  41. [49]

    arXiv preprint arXiv:1412.6980 (2014)

    Kingma, D.P.: Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)

    2014 arXiv

  42. [50]

    In: Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp

    Rasley, J., Rajbhandari, S., Ruwase, O., He, Y.: Deepspeed: System optimiza- tions enable training deep learning models with over 100 billion parameters. In: Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp. 3505–3506 (2020) ...

    2020

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.