Pith. sign in

REVIEW 4 major objections 7 minor 67 references

Regional Weather Variable Predictions by Machine Learning with Near-Surface Observational and Atmospheric Numerical Data

T0 review · 4 major / 7 minor · reviewed 2026-08-11 · deepseek-v4-flash

Pith's one-line read A machine-learning model fuses five-minute ground observations with hourly atmospheric model output to forecast weather variables at fine time resolution, and a regional variant extends this accuracy to ungauged locations.

desk verdict Solid station-level nowcasting with public code, but the 'ungauged' claim leans on borrowed observations from a nearby station and the gains lack a persistence baseline. read the letter →

arxiv 2412.10450 v2 pith:5EYN6TJL submitted 2024-12-11 physics.ao-ph cs.AIcs.LG

classification physics.ao-phcs.AIcs.LG MSC 68T0786A10
keywords weatherforecastingmachinelearningMesonetWRF-HRRRencoder-decoderLSTMregionaltransfernowcasting
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

The paper sets out to show that pairing frequent local observations with hourly atmospheric model output yields accurate, fine-resolution weather forecasts. Each dedicated 'modelet'—an encoder–decoder LSTM—ingests five-minute Mesonet readings and spatially aligned WRF-HRRR numerical fields to predict one variable at one station with five- or fifteen-minute resolution. Across eleven Kentucky stations and four variables, MiMa achieves the lowest RMSE in 39 of 44 comparisons, beating several baselines including the Micro-only model and raw WRF-HRRR. The regional Re-MiMa variant, trained on three representative stations with elevations as inputs, predicts eight held-out stations accurately in 22 of 32 cases, supporting the claim that a few regional models can replace many station-specific ones.

What carries the argument

The central object is the MiMa 'modelet': a per-variable, per-station (or per-region) encoder–decoder LSTM with two encoders. The Micro Encoder consumes the most relevant five-minute Mesonet parameters; the Macro Encoder consumes the most relevant hourly WRF-HRRR parameters, temporally downscaled to the same five-minute grid by fitting a quadratic polynomial to the last three hourly values. Encoder hidden states are concatenated into a single context vector that initializes the decoder, which predicts the target variable sequentially over the horizon. Re-MiMa modelets add elevation as an extra input channel to both encoders and randomly shuffle the order of representative-station streams during training, so that a single modelet can generalize across a region's elevation range.

What would settle it

Retrain Re-MiMa on the same three representative stations but, at inference for an ungauged site, replace the borrowed closest-elevation observational data with either (a) no micro input at all or (b) data from the farthest training station, then measure RMSE at the eight held-out stations; if accuracy remains at the Table X levels, the regional claim is robust, and if it degrades sharply, the ungauged result depends on proximity to a live station.

Watch

Extended reading notes

Core claim

The central claim is that integrating near-surface observational data and atmospheric numerical outputs at aligned geo-grids yields accurate, fine-grained short-term weather forecasts, and that this accuracy transfers to ungauged locations when elevation is included as a training feature. MiMa achieves this with a two-encoder decoder LSTM: a Micro Encoder processes the station's five-minute multivariate observations, a Macro Encoder processes the corresponding hourly WRF-HRRR fields downscaled to five-minute intervals via a quadratic polynomial fit, and the hidden states of both encoders are concatenated to initialize a decoder that produces sequential forecasts. Re-MiMa appends station elevation to both encoder inputs and randomly shuffles groups of representative-station data frames during training, preventing bias toward any single site. In evaluation, MiMa attains the smallest RMSE in 39 of 44 station-parameter instances (Table VI), and Re-MiMa outperforms location-specific MiMa at 22 of 32 ungauged station-parameter combinations (Table X).

Load-bearing premise

For Re-MiMa, the load-bearing assumption is that a target ungauged site can be served by borrowing real-time observational data from the training station closest in elevation; if that station's live data are unavailable, the regional forecast loses its ground-input channel and the claimed ungauged accuracy collapses.

Editorial extensions

If this is right

  • Weather nowcasting at five- or fifteen-minute resolution becomes practical in any region that already has a Mesonet-like observation network and an hourly numerical model such as WRF-HRRR.
  • A single regional modelet per weather variable can replace many station-specific models, reducing the number of models that must be trained and maintained.
  • Forecasts for ungauged locations can be issued without local instruments, provided the nearest elevation-closest training station has live observations to borrow.
  • Prediction error grows slowly as the lead time extends from one to four hours, so the approach supports useful forecasts over a several-hour horizon.
  • Raw WRF-HRRR output, despite covering the whole United States, is far less accurate than the fused modelet, showing that the fusion step itself adds substantial value.

Reading between the lines

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

  • Re-MiMa's ungauged capability depends on borrowing live data from a nearby station; a truly uninstrumented area with no nearby operational station would not be covered as described, and the paper does not quantify how accuracy degrades with station distance or elevation mismatch.
  • The parameter-relevance subsets were found to be identical across grids; if that stability holds beyond Kentucky, transferring MiMa to another region may require only a new relevance analysis, but whether the same subsets apply in different climates is untested.
  • The temporal downscaling of WRF-HRRR via quadratic extrapolation could be compared against simple interpolation or higher-order methods; such an ablation would isolate how much of the gain comes from downscaling versus from the fusion of the two data sources.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 7 minor

Summary. This paper proposes MiMa, an LSTM-based encoder-decoder model with separate Micro and Macro encoders, for predicting one of four weather variables (temperature, humidity, wind speed, pressure) at a 5-minute temporal resolution at individual Kentucky Mesonet stations, using 5-minute Mesonet observations and hourly WRF-HRRR outputs. It also proposes Re-MiMa, in which the encoders additionally receive station elevation and are trained on data from three representative stations, with the intent of forecasting at ungauged locations. The authors report that MiMa achieves the lowest RMSE in 39 of 44 station-parameter instances compared with five baselines, and that Re-MiMa outperforms location-specific MiMa in 22 of 32 cases at eight stations not used in its training.

Significance. The practical goal—accurate 5-minute-scale nowcasting at Mesonet stations and its regional extension—is worthwhile, and the paper has concrete strengths: the code and datasets are made publicly available; the WRF-HRRR preprocessing is described at the level of files and computational cost; the evaluation spans many stations and includes an ablation, an ensemble illustration, and an extreme-weather test. If the accuracy claims survive a more rigorous comparison, the contribution would be a useful fine-grained regional nowcasting tool. At present, however, the significance is limited by the lack of a persistence baseline, absence of statistical uncertainty measures, and the fact that the 'ungauged' Re-MiMa evaluation actually borrows live observations from nearby gauged stations.

major comments (4)
  1. [§V-B, Table VI] The central claim that MiMa 'significantly outperforms current models' is not established by the reported comparisons. The table contains no persistence baseline (e.g., last-observation carry-forward or diurnal persistence), which is the standard reference for 5-minute-ahead nowcasting; the evaluation covers only 16 days in one season; and no error bars, confidence intervals, or significance tests are given for the RMSE/MAE differences. Because the gaps over SARIMA and the Micro model are often small (e.g., TEMP at CCLA: 0.28 vs 0.24 for Micro; HUMI at LXGN: 1.03 vs 1.02), the 'significant' language is unsupported. Please add a persistence baseline, report uncertainty across the 16 days, and test whether the differences are statistically meaningful.
  2. [§VI, Table X] The ungauged-location claim is not demonstrated. Section VI states that for a target station not in the training set, 'observational data borrowed from the training station closest in elevation' are used as the Micro input. Thus every tested 'ungauged' station actually receives a live 5-minute observation stream from a nearby gauged station, and the model is never run without surface observations. A genuinely observation-free site would have no Micro input, leaving only the Macro encoder and elevation, a regime never evaluated. The abstract's statement that Re-MiMa provides accurate predictions 'even in areas without observational stations' is therefore not supported by Table X. Please either evaluate Re-MiMa with the Micro channel disabled at held-out locations or revise the claims to describe the actual capability: regional transfer using borrowed live observations from the elevation-closest gauged station.
  3. [§V-A, Experiment Setup Details] The train/validation/test split is described ambiguously: the model is trained on the third season of 2018 and 2019, then 'predicts the weather conditions for 80% of the 2020 data in the same season', with the remaining 20% used for validation and early stopping. The evaluation in Table VI is over '16 days chosen arbitrarily in the third season of 2020'. It is unclear whether those 16 days are a subset of the 80% test portion or overlap the 20% validation portion. If the validation period used for early stopping includes any of the reported test days, the early-stopping criterion leaks test information into model selection. Please clarify the exact dates and ensure the validation and test sets are disjoint.
  4. [§IV-B, Temporal Downscaling] The Macro encoder relies on WRF-HRRR values temporally downscaled from hourly to 5-minute resolution by a quadratic polynomial fit to the last three hourly points. This is a strong assumption about sub-hourly atmospheric evolution and is not validated against the 5-minute Mesonet observations, nor is the model's sensitivity to the choice of l=3 and the polynomial form reported. Since the Macro input is a core component of the MiMa architecture, please provide a validation of the downscaling or an ablation demonstrating that the results are insensitive to the downscaling procedure.
minor comments (7)
  1. [Abstract and Section I] The abstract and introduction call MiMa an 'encoder-decoder transformer structure', but Section IV-B describes it as an LSTM-based encoder-decoder and explicitly states that a transformer with attention would be unsuitable; please align the terminology.
  2. [§V-B] The text contains the typo 'third season of 202' and should read '2020'.
  3. [§V-C] The phrase 'different time sans' appears to mean 'different time steps' or 'time stamps'; please correct it.
  4. [§V-B] The phrase 'when all four predicted parameters at each station are taken into aggressive consideration' should be 'aggregate consideration'.
  5. [§V-A, Spatial Alignment of Micro and Macro Datasets] The WRF-HRRR dataset is described as 'gridded satellite data'; WRF-HRRR is a numerical weather prediction model output, not satellite data, so this characterization is misleading.
  6. [Table VI footnote] The footnote states that entries where MiMa is not smallest are underlined, but underlining is not visible in the manuscript text; please ensure the final typeset table renders the underlining.
  7. [References] Reference [3] appears to be a teaching document about equations rather than the HRRR data archive used in the paper; please cite the actual HRRR data source.

Circularity Check

0 steps flagged · score 2.0 of 10

No load-bearing circular step: predictions are evaluated on held-out observations with independent WRF-HRRR inputs; the Re-MiMa ungauged claim is a scoping limitation, not circularity.

full rationale

The MiMa and Re-MiMa results are empirical forecasts, not a derivation: models are trained on 2018-2019 Mesonet and independently computed WRF-HRRR data, then scored against 2020 Mesonet observations not used in training (Section V-A, Tables VI and X), so no fitted parameter is renamed as a prediction. The only self-citation, the preliminary ECML-PKDD version [1], is a pointer rather than load-bearing evidence, because the same comparative results appear in Tables VI and XI. The notable caveat is in Section VI: Re-MiMa's ungauged evaluation supplies real-time observations borrowed from the elevation-closest training station as Micro-Encoder input, with the paper stating that observational data borrowed from the training station closest in elevation to the target station are used for prediction. Therefore the abstract's claim of prediction even in areas without observational stations is not demonstrated for a fully observation-free region. This is an evidentiary limitation of the regional generalization claim, not circularity by construction, because the forecast variable is still scored against the target station's own held-out observations and is not equal to the borrowed input. Score 2 reflects the minor self-citation and the overbroad ungauged framing rather than any equation-level circularity.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The method rests on several pragmatic choices: hand-set feature selection thresholds, ad hoc quadratic downscaling of HRRR data, and the assumption that elevation alone captures station-to-station variation. No new physical entities are introduced. The ledger reflects the engineering nature of the contribution.

free parameters (6)
  • Relevance threshold Theta = 0.3
    Hand-set feature-selection cutoff used for all modelets; Section V-A.
  • Maximum relevant parameters gamma = 10
    Hand-set cap on number of micro and macro inputs per modelet; Section V-A.
  • Polynomial downscaling window l = 3
    Number of past hourly HRRR values used to fit quadratic for 5-minute interpolation; Section IV-B.
  • LSTM hidden sizes = micro and macro encoders 256, decoder 512
    Architecture hyperparameters chosen without reported sensitivity analysis; Section V-A.
  • Training hyperparameters = batch 64, epochs 60, dropout 0.5, learning rate 0.001, patience 5 and 10
    Standard but hand-chosen; the patience number is inconsistent in the text.
  • Extreme threshold = 5%
    Used to shrink time series for extreme-weather evaluation; Section V-E.
assumptions (5)
  • domain assumption Nearest HRRR grid cell is representative of each Mesonet station site.
    Spatial alignment procedure in Section V-A selects one 3-km grid cell by distance; complex terrain may make this cell unrepresentative.
  • ad hoc to paper Quadratic fit to the last three hourly HRRR values adequately captures sub-hourly atmospheric evolution.
    Temporal downscaling in Section IV-B creates all 5-minute macro inputs; no validation against 5-minute truth is shown.
  • ad hoc to paper For Re-MiMa, borrowing the elevation-closest training station's observations is a valid proxy for an ungauged site.
    Inference procedure in Section VI; without it, Re-MiMa cannot run at an ungauged location.
  • domain assumption Elevation is the only station-specific attribute needed to generalize across the region.
    Re-MiMa input includes elevation only; land cover, slope, and exposure are ignored.
  • standard math LSTM gradient training and standard supervised learning assumptions hold.
    Background ML machinery used in Eqs. (1)-(3); not claim-specific.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Regional Weather Variable Predictions by Machine Learning with Near-Surface Observational and Atmospheric Numerical Data." pith.science (2026). https://pith.science/paper/5EYN6TJL

@misc{pith2026241210450,
  author       = {Pith},
  title        = {Pith review of: Regional Weather Variable Predictions by Machine Learning with Near-Surface Observational and Atmospheric Numerical Data},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5EYN6TJL}},
  note         = {Machine review of arXiv:2412.10450}
}
read the original abstract

Accurate and timely regional weather prediction is vital for sectors dependent on weather-related decisions. Traditional prediction methods, based on atmospheric equations, often struggle with coarse temporal resolutions and inaccuracies. This paper presents a novel machine learning (ML) model, called MiMa (short for Micro-Macro), that integrates both near-surface observational data from Kentucky Mesonet stations (collected every five minutes, known as Micro data) and hourly atmospheric numerical outputs (termed as Macro data) for fine-resolution weather forecasting. The MiMa model employs an encoder-decoder transformer structure, with two encoders for processing multivariate data from both datasets and a decoder for forecasting weather variables over short time horizons. Each instance of the MiMa model, called a modelet, predicts the values of a specific weather parameter at an individual Mesonet station. The approach is extended with Re-MiMa modelets, which are designed to predict weather variables at ungauged locations by training on multivariate data from a few representative stations in a region, tagged with their elevations. Re-MiMa (short for Regional-MiMa) can provide highly accurate predictions across an entire region, even in areas without observational stations. Experimental results show that MiMa significantly outperforms current models, with Re-MiMa offering precise short-term forecasts for ungauged locations, marking a significant advancement in weather forecasting accuracy and applicability.

Figures

Figures reproduced from arXiv: 2412.10450 by the authors.

Figure 1
Figure 1. Kentucky Mesonet weather observational stations denoted by yellow circles, with those stations chosen for MiMa model evaluation and pointed by red line segments tagged with their latitudes, longi￾tudes, and elevations. (i.e., variables) per hour over large geo-grids (e.g., 3 km × 3 km), with coarse temporal granularity (hourly forecasts) often deemed insufficient for applications requiring predictions in the interva… view at source ↗
Figure 2
Figure 2. Overview of the MiMa (short for Micro-Macro) model inputted with data from both an individual station and WRF-HRRR modeling computation to yield the weather variable predictions. itation [14], [15], [16], [17], [18], air quality [19], weather changes [20], [21], [22], [23], etc. However, existing ML forecasting models have not yet achieved accurate predictions of weather variables at fine temporal resolutions (as se… view at source ↗
Figure 3
Figure 3. Structure of the Micro model, with the hidden state Ht obtained by inputting Xmicro to an encoder and the output Ot+1 obtained by inputting Ht plus Y0 to a decoder. Output Ot+1 is then passed to a fully connected layer which generates the predicted parameter value Yt+1 via a fully connected network. forecasting in consecutive future time points, an Encoder￾Decoder structure with the Long Short-Term Memory (LSTM) net… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Structure of MiMa model, with its Micro Encoder and its Decoder identical to those depicted in [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Prediction results for 24 hours for those four prediction parameters with the Micro, MiMa, and WRF models compared to the observed data. TABLE VII: RMSE values of MiMa modelets at each 15-minute interval over a 3-hour horizon under 1-hour (4-hour) lead time Station Par…
Figure 6
Figure 6. Figure 6: Ensemble temperature prediction plot of MiMa modelet for Station FARM. progresses from the first time point (at 15 min.) to the 12th time point (at 180 min.) over the prediction horizon for all four weather parameters. The table also reveals that larger lead times yiel…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

67 extracted references · 58 canonical work pages

  1. [1]

    Precise weather parameter predictions for target regions via neural networks,

    Y . Zhang, X. Yuan, S. K. Kimball, and et. al., “Precise weather parameter predictions for target regions via neural networks,” in Machine Learning and Knowledge Discovery in Databases. Applied Data Science Track: European Conference, ECML PKDD 2021, September 2021 , 2021, pp. 151–167

  2. [2]

    WRF Resources,

    “WRF Resources,” http://home.chpc.utah.edu/ ∼u0553130/Brian Blaylock/wrf.html, 2021

  3. [3]

    National Center for Atmospheric Research, Understanding the earth sys- tem, http://www.cgd.ucar.edu/staff/islas/teaching/2 Equations.pdf, 2019

  4. [4]

    Kentucky Mesonet,

    “Kentucky Mesonet,” https://www.kymesonet.org, 2021

  5. [5]

    Mesonet,

    “Mesonet,” https://www.mesonet.org/, 2021

  6. [6]

    Long short-term memory,

    S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997

  7. [7]

    Generative adversarial nets,

    I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y . Bengio, “Generative adversarial nets,” Advances in Neural Information Processing Systems , vol. 27, 2014

  8. [8]

    Deep neural networks for ultra- short-term wind forecasting,

    M. Dalto, J. Matu ˇsko, and M. Va ˇsak, “Deep neural networks for ultra- short-term wind forecasting,” in Proceedings of IEEE International Conference on Industrial Technology (ICIT) , 2015, pp. 1657–1663

Show all 67 references
  1. [9]

    Transfer learning for short-term wind speed prediction with deep neural networks,

    Q. Hu, R. Zhang, and Y . Zhou, “Transfer learning for short-term wind speed prediction with deep neural networks,” Renewable Energy, vol. 85, pp. 83–95, 2016

  2. [10]

    Predicting hurricane trajectories using a recurrent neural network,

    S. Alemany, J. Beltran, A. Perez, and S. Ganzfried, “Predicting hurricane trajectories using a recurrent neural network,” in Proceedings of AAAI Conference on Artificial Intelligence , vol. 33, 2019, pp. 468–475

  3. [11]

    Data-driven weather forecast- ing models performance comparison for improving offshore wind tur- bine availability and maintenance,

    R. K. Pandit, A. Kolios, and D. Infield, “Data-driven weather forecast- ing models performance comparison for improving offshore wind tur- bine availability and maintenance,” IET Renewable Power Generation , vol. 14, no. 13, pp. 2386–2394, 2020

  4. [12]

    Deep learning for solar power forecasting—an approach using autoencoder and LSTM neural networks,

    A. Gensler, J. Henze, B. Sick, and N. Raabe, “Deep learning for solar power forecasting—an approach using autoencoder and LSTM neural networks,” in Proceedings of International Conference on Systems, Man, and Cybernetics (SMC) , 2016, pp. 2858–2865

  5. [13]

    Weather based photovoltaic energy generation prediction using LSTM networks,

    S. Arshi, L. Zhang, and B. Strachan, “Weather based photovoltaic energy generation prediction using LSTM networks,” in Proceedings of International Joint Conference on Neural Networks (IJCNN) , 2019

  6. [14]

    Convolutional LSTM network: A machine learning approach for precipitation nowcasting,

    S. Xingjian, Z. Chen, H. Wang, D.-Y . Yeung, W.-K. Wong, and W.- C. Woo, “Convolutional LSTM network: A machine learning approach for precipitation nowcasting,” in Proceedings of Advances in Neural Information Processing Systems , 2015, pp. 802–810

  7. [15]

    Rainfall prediction: A deep learning approach,

    E. Hern ´andez, V . Sanchez-Anguix, V . Julian, J. Palanca, and N. Duque, “Rainfall prediction: A deep learning approach,” in Proc. of Interna- tional Conference on Hybrid Artificial Intelligence Systems , 2016, pp. 151–162

  8. [16]

    Improving precipitation estimation using convolutional neural network,

    B. Pan, K. Hsu, A. AghaKouchak, and S. Sorooshian, “Improving precipitation estimation using convolutional neural network,” Water Resources Research, vol. 55, no. 3, pp. 2301–2321, 2019

  9. [17]

    Near real-time hurricane rainfall forecasting using convolutional neural network models with in- tegrated multi-satellite retrievals for gpm (imerg) product,

    T. Kim, T. Yang, L. Zhang, and Y . Hong, “Near real-time hurricane rainfall forecasting using convolutional neural network models with in- tegrated multi-satellite retrievals for gpm (imerg) product,” Atmospheric Research, vol. 270, 2022

  10. [18]

    Evaluation of subseasonal-to-seasonal (s2s) precipitation forecast from the north amer- ican multi-model ensemble phase ii (nmme-2) over the contiguous us,

    L. Zhang, T. Kim, T. Yang, Y . Hong, and Q. Zhu, “Evaluation of subseasonal-to-seasonal (s2s) precipitation forecast from the north amer- ican multi-model ensemble phase ii (nmme-2) over the contiguous us,” Journal of Hydrology, vol. 603, p. 127058, 2021

  11. [19]

    Deep distributed fusion network for air quality prediction,

    X. Yi, J. Zhang, Z. Wang, T. Li, and Y . Zheng, “Deep distributed fusion network for air quality prediction,” in Proceedings of 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2018, pp. 965–973

  12. [20]

    A deep hybrid model for weather forecasting,

    A. Grover, A. Kapoor, and E. Horvitz, “A deep hybrid model for weather forecasting,” in Proceedings of 21th ACM SIGKDD International Con- ference on Knowledge Discovery and Data Mining , 2015, pp. 379–386

  13. [21]

    Improving data-driven global weather prediction using deep convolutional neural networks on a cubed sphere,

    J. A. Weyn, D. R. Durran, and R. Caruana, “Improving data-driven global weather prediction using deep convolutional neural networks on a cubed sphere,” Journal of Advances in Modeling Earth Systems , vol. 12, no. 9, 2020

  14. [22]

    A large- scale comparison of artificial intelligence and data mining (ai&dm) techniques in simulating reservoir releases over the upper colorado region,

    T. Yang, L. Zhang, T. Kim, Y . Hong, D. Zhang, and Q. Peng, “A large- scale comparison of artificial intelligence and data mining (ai&dm) techniques in simulating reservoir releases over the upper colorado region,” Journal of Hydrology, vol. 602, 2021. 18

  15. [23]

    Simulating hydropower discharge using multiple decision tree methods and a dynamical model merging technique,

    T. Yang, X. Liu, L. Wang, P. Bai, and J. Li, “Simulating hydropower discharge using multiple decision tree methods and a dynamical model merging technique,” Journal of Water Resources Planning and Manage- ment, vol. 146, no. 2, 2020

  16. [24]

    Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting,

    H. Wu, J. Xu, J. Wang, and M. Long, “Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting,” in Proceedings of 35th Conference on Neural Information Processing Systems (NeurIPS 2021) , vol. 34, pp. 22 419–22 430, November 2021

  17. [25]

    Fedformer: Frequency enhanced decomposed transformer for long-term series fore- casting,

    T. Zhou, Z. Ma, Q. Wen, X. Wang, L. Sun, and R. Jin, “Fedformer: Frequency enhanced decomposed transformer for long-term series fore- casting,” in Proceedings of 39th International Conference on Machine Learning (ICML 2022) . PMLR, 2022, pp. 27 268–27 286

  18. [26]

    Interpretable weather forecasting for worldwide stations with a unified deep model,

    H. Wu, H. Zhou, M. Long, and J. Wang, “Interpretable weather forecasting for worldwide stations with a unified deep model,” Nature Machine Intelligence, vol. 5, pp. 602–611, June 2023

  19. [27]

    Fourcastnet: Accelerating global high-resolution weather forecasting using adaptive fourier neural operators,

    T. Kurth, S. Subramanian, P. Harrington, J. Pathak, M. Mardani, D. Hall, A. Miele, K. Kashinath, and A. Anandkumar, “Fourcastnet: Accelerating global high-resolution weather forecasting using adaptive fourier neural operators,” in Proceedings of the platform for advanced scien...

  20. [28]

    Skillful precipitation nowcasting using deep generative models of radar,

    S. Ravuri, K. Lenc, M. Willson, D. Kangin, R. Lam, P. Mirowski, M. Fitzsimons, M. Athanassiadou, S. Kashem, S. Madge et al., “Skillful precipitation nowcasting using deep generative models of radar,” Nature, vol. 597, pp. 672–677, 2021

  21. [29]

    MiMa codes and pertinent information,

    “MiMa codes and pertinent information,” https://prefer-nsf.org/ publications software.html, 2023

  22. [30]

    Challenges and design choices for global weather and climate models based on machine learning,

    P. D. Dueben and P. Bauer, “Challenges and design choices for global weather and climate models based on machine learning,” Geoscientific Model Development, vol. 11, no. 10, pp. 3999–4009, 2018

  23. [31]

    Toward data-driven weather and climate forecasting: Ap- proximating a simple general circulation model with deep learning,

    S. Scher, “Toward data-driven weather and climate forecasting: Ap- proximating a simple general circulation model with deep learning,” Geophysical Research Letters , vol. 45, no. 22, pp. 12–616, 2018

  24. [32]

    Can machines learn to predict weather? using deep learning to predict gridded 500-hpa geopotential height from historical weather data,

    J. A. Weyn, D. R. Durran, and R. Caruana, “Can machines learn to predict weather? using deep learning to predict gridded 500-hpa geopotential height from historical weather data,” Journal of Advances in Modeling Earth Systems , vol. 11, no. 8, pp. 2680–2693, 2019

  25. [33]

    Data-driven forecasting of high-dimensional chaotic systems with long short-term memory networks,

    P. R. Vlachas, W. Byeon, Z. Y . Wan, T. P. Sapsis, and P. Koumoutsakos, “Data-driven forecasting of high-dimensional chaotic systems with long short-term memory networks,” in Proc. of Royal Society A: Mathemat- ical, Physical and Engineering Sciences , vol. 474, no. 2213, 2018

  26. [34]

    Sevir: A storm event imagery dataset for deep learning applications in radar and satellite meteorology,

    M. Veillette, S. Samsi, and C. Mattioli, “Sevir: A storm event imagery dataset for deep learning applications in radar and satellite meteorology,” Advances in Neural Information Processing Systems , vol. 33, 2020

  27. [35]

    U-net: Convolutional networks for biomedical image segmentation,

    O. Ronneberger, P. Fischer, and T. Brox, “U-net: Convolutional networks for biomedical image segmentation,” in Int’l Conference on Medical Image Computing and Computer-assisted Intervention , 2015, pp. 234– 241

  28. [36]

    Improving nowcasting of convective development by incorporating polarimetric radar variables into a deep-learning model,

    X. Pan, Y . Lu, K. Zhao, H. Huang, M. Wang, and H. Chen, “Improving nowcasting of convective development by incorporating polarimetric radar variables into a deep-learning model,” Geophysical Research Letters, vol. 48, no. 21, 2021

  29. [37]

    Msdm v1. 0: A machine learning model for precipitation nowcasting over eastern china using multisource data,

    D. Li, Y . Liu, and C. Chen, “Msdm v1. 0: A machine learning model for precipitation nowcasting over eastern china using multisource data,” Geoscientific Model Development , vol. 14, no. 6, pp. 4019–4034, 2021

  30. [38]

    Micro-climate prediction-multi scale encoder-decoder based deep learning framework,

    P. Kumar, R. Chandra, C. Bansal, S. Kalyanaraman, T. Ganu, and M. Grant, “Micro-climate prediction-multi scale encoder-decoder based deep learning framework,” in Proc. of the 27th ACM SIGKDD Confer- ence on Knowledge Discovery & Data Mining , 2021, pp. 3128–3138

  31. [39]

    Comprehensive transformer-based model architecture for real- world storm prediction,

    F. Lin, Y . Zhang, L. Chen, L. Peng, P. Sigdel, X. Yuan, and N.-F. Tzneg, “Comprehensive transformer-based model architecture for real- world storm prediction,” in Proceedings of European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Data...

  32. [40]

    Skillful nowcasting of extreme precipitation with nowcastnet,

    Y . Zhang, M. Long, K. Chen, L. Xing, R. Jin, M. Jordan, and J. Wang, “Skillful nowcasting of extreme precipitation with nowcastnet,” Nature, vol. 619, pp. 526–532, 2023

  33. [41]

    itrans- former: Inverted transformers are effective for time series forecasting,

    Y . Liu, T. Hu, H. Zhang, H. Wu, S. Wang, L. Ma, and M. Long, “itrans- former: Inverted transformers are effective for time series forecasting,” arXiv preprint arXiv:2310.06625 , 2023

  34. [42]

    Neural networks for postpro- cessing ensemble weather forecasts,

    D. Kochkov, J. Yuval, and I. Langmore, “Neural networks for postpro- cessing ensemble weather forecasts,” Monthly Weather Review, vol. 146, pp. 2885–3900, August 2018

  35. [43]

    Deep uncertainty quantification: A machine learning approach for weather forecasting,

    B. Wang, J. Lu, Z. Yan, H. Luo, T. Li, Y . Zheng, and G. Zhang, “Deep uncertainty quantification: A machine learning approach for weather forecasting,” in Proc. of the 25th ACM SIGKDD international conference on knowledge discovery & data mining , 2019, pp. 2087–2095

  36. [44]

    Weather-5k: A large- scale global station weather dataset towards comprehensive time-series forecasting benchmark,

    T. Han, S. Guo, Z. Chen, W. Xu, and L. Bai, “Weather-5k: A large- scale global station weather dataset towards comprehensive time-series forecasting benchmark,” arXiv preprint arXiv:2406.14399 , 2024

  37. [45]

    Deriving accurate surface meteorological states at arbitrary locations via observation-guided continous neural field modeling,

    Z. Liu, H. Chen, L. Bai, W. Li, K. Chen, Z. Wang, W. Ouyang, Z. Zou, and Z. Shi, “Deriving accurate surface meteorological states at arbitrary locations via observation-guided continous neural field modeling,” IEEE Transactions on Geoscience and Remote Sensing , 2024

  38. [46]

    Deep- physinet: Bridging deep learning and atmospheric physics for accurate and continuous weather modeling,

    W. Li, Z. Liu, K. Chen, H. Chen, S. Liang, Z. Zou, and Z. Shi, “Deep- physinet: Bridging deep learning and atmospheric physics for accurate and continuous weather modeling,” arXiv preprint arXiv:2401.04125 , 2024

  39. [47]

    Deep learning for day forecasts from sparse observations,

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

  40. [48]

    Multi-modal graph neural networks for localized off-grid weather forecasting,

    Q. Yang, J. Giezendanner, D. S. Civitarese, J. Jakubik, E. Schmitt, A. Chandra, J. Vila, D. Hohl, C. Hill, C. Watson et al. , “Multi-modal graph neural networks for localized off-grid weather forecasting,” arXiv preprint arXiv:2410.12938, 2024

  41. [49]

    The High-Resolution Rapid Refresh (HRRR),

    “The High-Resolution Rapid Refresh (HRRR),” https://rapidrefresh. noaa.gov/hrrr/, 2022

  42. [50]

    A description of the advanced research wrf model version 4,

    W. C. Skamarock, J. B. Klemp, J. Dudhia, D. O. Gill, Z. Liu, J. Berner, W. Wang, J. G. Powers, M. G. Duda, D. M. Barker et al., “A description of the advanced research wrf model version 4,” National Center for Atmospheric Research: Boulder, CO, USA , vol. 145, p. 145, 2019

  43. [51]

    WRF Model Users’ Page,

    “WRF Model Users’ Page,” https://www.mmm.ucar.edu/ weather-research-and-forecasting-model, 2022

  44. [52]

    A north american hourly assimilation and model forecast cycle: The rapid refresh,

    S. G. Benjamin, S. S. Weygandt, J. M. Brown, M. Hu, C. R. Alexander, T. G. Smirnova, J. B. Olson, E. P. James, D. C. Dowell, G. A. Grell et al., “A north american hourly assimilation and model forecast cycle: The rapid refresh,” Monthly Weather Review, vol. 144, no. 4, pp. 166...

  45. [53]

    upcoming HRRRv2/RAPv3 imple @ NCEP - sched 8/23,

    “upcoming HRRRv2/RAPv3 imple @ NCEP - sched 8/23,” https://ruc. noaa.gov/forum/f2/Welcome.cgi/noframes/read/2478, 2022

  46. [54]

    Attention is all you need,

    A. Vaswani, “Attention is all you need,” Advances in Neural Information Processing Systems, 2017

  47. [55]

    Feature selection for machine learn- ing: comparing a correlation-based filter approach to the wrapper

    M. A. Hall and L. A. Smith, “Feature selection for machine learn- ing: comparing a correlation-based filter approach to the wrapper.” in Proceedings of the 12th International Florida Artificial Intelligence Research Society, vol. 1999, 1999, pp. 235–239

  48. [56]

    Is mutual information adequate for feature selection in regression?

    B. Fr ´enay, G. Doquire, and M. Verleysen, “Is mutual information adequate for feature selection in regression?” Neural Networks, vol. 48, pp. 1–7, 2013

  49. [57]

    R. J. Hyndman and G. Athanasopoulos, Forecasting: principles and practice. OTexts, 2018

  50. [58]

    Atmospheric temperature prediction using support vector machines,

    Y . Radhika and M. Shashi, “Atmospheric temperature prediction using support vector machines,” International journal of computer theory and engineering, vol. 1, no. 1, p. 55, 2009

  51. [59]

    Nowcasting: The promise of new technologies of communi- cation, modeling, and observation,

    C. Mass, “Nowcasting: The promise of new technologies of communi- cation, modeling, and observation,” Bulletin of the American Meteoro- logical Society, vol. 93, no. 6, pp. 797–809, 2012

  52. [60]

    Neural networks for postprocessing ensemble weather forecasts,

    S. Rasp and S. Lerch, “Neural networks for postprocessing ensemble weather forecasts,” Monthly Weather Review, vol. 146, no. 11, pp. 3885– 3900, 2018

  53. [61]

    Ensemble methods for neural network-based weather forecasts,

    S. Scher and G. Messori, “Ensemble methods for neural network-based weather forecasts,” Journal of Advances in Modeling Earth Systems , vol. 13, no. 2, 2021

  54. [62]

    The ensemble approach to forecasting: A review and synthesis,

    H. Wu and D. Levinson, “The ensemble approach to forecasting: A review and synthesis,” Transportation Research Part C: Emerging Technologies, vol. 132, 2021

  55. [63]

    Computing the ensemble spread from de- terministic weather predictions using conditional generative adversarial networks,

    R. Brecht and A. Bihlo, “Computing the ensemble spread from de- terministic weather predictions using conditional generative adversarial networks,” Geophysical Research Letters , vol. 50, no. 2, 2023. Yihe Zhang (M’24) received M.S. degree from SYSU-CMU Joint Institute of Engi...

  56. [1986]

    His current research interest includes high-performance computer systems, computer net- works, and parallel and distributed processing

    He has been with the Center for Advanced Computer Studies, School of Computing and In- formatics, the University of Louisiana at Lafayette, since 1987. His current research interest includes high-performance computer systems, computer net- works, and parallel and distributed p...

  57. [2016]

    degree in the School of Computing and Informatics, Uni- versity of Louisiana at Lafayette

    He is currently working toward a Ph.D. degree in the School of Computing and Informatics, Uni- versity of Louisiana at Lafayette. His research inter- ests include machine learning applications, knowl- edge graphs, and data-driven security. He received the Best Paper Award and ...

  58. [2018]

    His research includes stochastic and ap- proximate computing, unary processing, in-memory computing, and hyperdimensional computing

    He is an Assistant Professor at the School of Computing and Informatics, University of Louisiana at Lafayette. His research includes stochastic and ap- proximate computing, unary processing, in-memory computing, and hyperdimensional computing. He received the Best Paper Award ...

  59. [2022]

    His research interests lie in weather fore- casting, radar analysis, and remote sensing

    He currently works as a meteorologist at the National Oceanic and Atmospheric Administration (NOAA). His research interests lie in weather fore- casting, radar analysis, and remote sensing. Sytske Kimball received her Ph.D. in Meteorology from Pennsylvania State University in ...

Pith tools

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