A conditional Generative Adversarial network model for the Weather4Cast 2024 Challenge

Atharva Deshpande; Hrishikesh Simu; Jeet Shah; Kaushik Gopalan

REVIEW 3 major objections 5 minor 1 cited by

A conditional Generative Adversarial network model for the Weather4Cast 2024 Challenge

T0 review · 3 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read A two-stage optical-flow-plus-cGAN pipeline tops the Weather4Cast 2024 rainfall nowcasting leaderboard.

desk verdict A short, honest competition report whose first-place CRPS is externally verified; the method is a standard cGAN+optical flow pipeline, and the main weakness is the missing ablation isolating what the cGAN actually contributes. read the letter →

arxiv 2412.00451 v1 pith:JSHSDQSK submitted 2024-11-30 cs.CV

Atharva Deshpande , Kaushik Gopalan , Jeet Shah , Hrishikesh Simu This is my paper · ORCID

classification cs.CV MSC 68T0786A10

keywords precipitationnowcastingconditionalGANopticalflowSEVIRIOPERAradarWeather4CastCRPSimage-to-imagetranslation

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 argues that precipitation up to four hours ahead can be forecast competitively by decomposing the task into two steps: extrapolate satellite infrared radiance images with dense optical flow, then translate each extrapolated radiance frame into a rainfall map with a conditional GAN. The model scored about 7.34 CRPS in the Weather4Cast 2024 core challenge, the only entry to beat the baseline (10.84), and took first place. The authors attribute this to the two-step decomposition and to preprocessing that focuses the model on cloudy regions, which keeps the generator relatively small. The claim matters because operational nowcasting often relies on optical flow alone or on heavy learned models, and this shows a hybrid can be competitive at short leads.

What carries the argument

The load-bearing mechanism is the two-stage decomposition. Stage one is dense optical flow: sparse Lucas-Kanade motion vectors from blob features are interpolated with radial basis functions into a dense field that propagates the four input SEVIRI radiance frames forward sixteen steps, covering four hours. Stage two is image translation: a Pix2Pix-style conditional GAN, with a U-Net generator containing dilated convolutions in the bottleneck and a PatchGAN discriminator, maps each extrapolated radiance frame to an OPERA rainfall map; training uses adversarial, pixel-wise MAE, and VGG-16 perceptual losses. Preprocessing, namely averaging four highly correlated infrared channels and applying Otsu-based foreground segmentation to isolate cloudy regions, is what allows the generator to stay small while still learning the radiance-to-rain relationship.

What would settle it

Compare the model against a persistence forecast (rainfall stays as the latest observed frame) and against a no-motion variant (radiance held constant for four hours) on the same CRPS and on a scale-selective score like fractions skill score at 5–20 km. If the optical-flow version does not beat persistence and the no-motion variant, the reported skill comes from the cGAN translation rather than the extrapolation; if no variant beats the baseline at full OPERA resolution, the competition result is an artifact of the 32x32 averaging.

Watch

Extended reading notes

Core claim

The central claim is that a modest conditional GAN can produce usable four-hour cumulative rainfall estimates when fed with radiance frames that are themselves forecast by dense optical flow, rather than by a learned temporal model. The authors report a CRPS of 7.34 versus 10.84 for the best baseline in the Weather4Cast 2024 competition, and first place on the core leaderboard. The method averages four infrared channels, masks out cloud-free regions using Otsu thresholding, propagates radiance sixteen time steps with dense Lucas-Kanade optical flow, and then uses a U-Net-style generator with a PatchGAN discriminator to translate each predicted radiance frame into an OPERA rainfall frame. The paper also notes that the model tends to underestimate peak rainfall intensity and to flag deep cloud as rain, and that the optical flow extrapolation fragments cloud systems at longer lead times.

Load-bearing premise

The whole pipeline depends on dense optical flow extrapolating cloud evolution faithfully enough for the cGAN to translate the predicted radiance into rainfall; the paper's own figure shows cloud systems fragmenting as the optical-flow lead time grows.

Editorial extensions

If this is right

If the claim holds, precipitation nowcasting can be assembled from off-the-shelf optical flow plus a small image-to-image GAN, without a large recurrent or video model.
Because the cGAN sees only one radiance frame at a time, its rainfall output inherits any defects in the optical-flow extrapolation, such as the cloud fragmentation the paper documents.
The competition CRPS is computed on rain rates averaged over 32x32 pixel blocks, so the reported margin over baseline is at a coarse scale; the paper itself cautions that location errors may be hidden by this averaging.
Being the only submission to outperform the baseline suggests that, at least at the competition's resolution, the two-stage pipeline is a strong point of reference for future learned nowcasting models.

Reading between the lines

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

A testable extension would replace the optical-flow stage with a learned motion model trained on the same radiance data; if fragmentation is the main error source, this should improve CRPS beyond 7.34 without changing the cGAN stage.
Because the generator is frame-independent, conditioning it on the four input frames or on the optical-flow displacement field could let it correct rain-streak discontinuities, a direction the paper's own discussion of temporal variability suggests.
The coarse CRPS averaging implies the model's real-world utility should be judged on scale-selective scores, such as fractions skill score for small rain cells, before operational use is considered.
If the two-step decomposition is as competitive as reported, end-to-end differentiable optical flow plus a small GAN may offer a cheaper alternative to large video-prediction transformers for short-range rainfall.

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.

Desk Editor's Note

read the letter

Colleague,

Quick read: this is a competition report, not a research paper, and judged as a report it's mostly fine. The headline result — 7.34 CRPS, first place in Weather4Cast 2024 core challenge, only submission beating baseline — is externally verified by the leaderboard, so the central claim is solid. What's new is the specific configuration: average of four IR channels, Otsu cloud masking, dense Lucas-Kanade/RBF optical flow extrapolation, then a Pix2Pix-style cGAN with VGG perceptual loss. Each piece is standard, but the combination as a winning challenge entry is new and the code is on GitHub.

The paper is also appropriately honest. It calls its own qualitative results 'decidedly mixed,' shows the optical flow fragmentation in Fig 3, and flags that the 32x32 averaging may hide misalignment in the CRPS. That's good scientific hygiene for a short report.

The soft spots are real but mostly what you'd expect from the genre. First, the abstract says 'approximately 7.5' while Table 2 says 7.34 — sloppy, should be fixed. Second, there's exactly one visual example, no error bars, no sensitivity analysis on the Otsu threshold or normalization constants. Third, and this is the one substantive issue: the cGAN is trained on observed SEVIRI frames and applied to optical-flow-extrapolated frames, which the paper itself shows become increasingly fragmented with lead time. So the 7.34 is an end-to-end score; we can't attribute it to the learned radiance-to-rainfall mapping without an ablation or fine-tuning on flow outputs. The paper doesn't do that, and doesn't quantify the optical flow stage at all. The stress-test note is right on this point. It doesn't falsify the leaderboard claim, but it does mean the title's implication — that the cGAN is the thing that wins — is underdetermined.

Who's this for: someone entering Weather4Cast or a similar nowcasting challenge who wants a working baseline recipe. It's not a method paper for the general CV/weather audience.

My take: send it to peer review, but with the expectation of revision — add the discrepancy fix, an ablation or at least a quantification of the optical-flow stage, and a sensitivity check on the preprocessing. The external score makes it worth referees' time; the missing analysis keeps it from being more than a solid challenge write-up.

Recommendation: accept for review with major revision.

Referee Report

3 major / 5 minor

Summary. The paper describes a two-stage pipeline for precipitation nowcasting in the Weather4Cast 2024 challenge: dense optical flow (Lucas-Kanade via pySTEPS) extrapolates SEVIRI infrared radiance frames four hours ahead, and a Pix2Pix-style conditional GAN translates each extrapolated radiance frame into an OPERA rainfall field. The rainfall fields are aggregated to produce cumulative rainfall, and the submission is reported to have achieved a CRPS of 7.34, placing first on the core challenge leaderboard and outperforming the baseline (10.84). The paper includes qualitative sample outputs, a discussion of optical-flow-induced fragmentation, and candid statements about the mixed quality of the results and the insensitivity of the competition metric to spatial misalignment.

Significance. If the leaderboard claim is taken at face value, the paper reports a competitive win in an externally evaluated nowcasting challenge and demonstrates that a relatively simple optical-flow-plus-cGAN pipeline can beat the challenge baseline. The authors are unusually candid about limitations, and the code is made publicly available. However, the scientific significance is currently limited by the absence of any statistical analysis, the lack of an ablation separating the optical-flow and cGAN stages, and the acknowledged insensitivity of the official metric to spatial errors. The central result is externally verified as a competition score, but the paper does not yet substantiate what component of the pipeline contributes to that score.

major comments (3)

[§2.2 and Figure 3] The cGAN is trained on observed SEVIRI frames ('The training process involved a single input frame as input and a single corresponding OPERA rainfall frame as output'), but at inference it is applied to Lucas-Kanade extrapolated frames, which the paper's own Figure 3 shows become increasingly fragmented with lead time. Because no ablation, fine-tuning, or distribution-shift analysis is provided, the end-to-end CRPS of 7.34 cannot be attributed to the learned radiance-to-rainfall mapping; the optical-flow stage may be contributing positively, negatively, or negligibly. I request either a per-stage ablation (e.g., cGAN on observed versus extrapolated inputs) or quantitative evidence that the extrapolated inputs are within the training distribution.
[§3, Table 2] The only quantitative result is a single leaderboard point estimate (7.34 versus 10.84), with no uncertainty interval, no number of test samples, no definition of the baseline model, and no per-lead-time or per-region breakdown. Given the paper's own Section 4 caveat that the official metric averages over 32x32 blocks and may be insensitive to misalignment, the leaderboard number alone does not establish spatial forecast skill; additional verification (e.g., confidence intervals from the competition's test set or complementary metrics such as FSS) is needed.
[§2.2 and §3] The optical-flow stage is never quantitatively evaluated despite being half of the pipeline and the source of the degradation shown in Figure 3. Reporting a quantitative comparison of extrapolated radiance frames against observed future frames (e.g., MAE or structural similarity per lead time) is necessary to understand how much of the final error originates in each stage.

minor comments (5)

[Figure 2] The caption says the prediction corresponds to a '2 hour lead time' while the text says the example is for a lead time of 4 hours; these should be reconciled.
[Throughout] There are several typographical errors: 'Probablity' in Section 3, 'allong' in Section 2, and 'an single' in Section 2.2.
[References] Reference [3], the Weather4Cast 2024 challenge paper, is missing venue and year information; it should be completed.
[§4] The Discussion describes foreground/background segmentation as 'the other innovation introduced', but this technique is not novel in the image-processing or nowcasting literature; the claim should be softened or supported with a comparison.
[§2] The normalization constants (dividing by 150 for input and by 5 for rainfall) appear arbitrary; a sentence justifying these choices or noting their sensitivity would help reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found; the CRPS claim is anchored externally and no prediction reduces to its own inputs.

full rationale

The paper's headline result—CRPS 7.34 and 1st place on the Weather4Cast 2024 core challenge leaderboard—is measured against an external competition benchmark, not derived from the model's own equations or from any fitted parameter. Table 2 reports competition leaderboard scores, so the central claim is independently anchored outside the paper. The preprocessing and modeling choices (channel averaging, Otsu cloud segmentation, normalization constants, Lucas-Kanade optical-flow extrapolation, cGAN losses) are domain-motivated and are not fitted to the CRPS in a way that makes the reported score follow by construction. The paper is transparent about limitations: it calls the feature-selection procedure 'admittedly ad-hoc', reports 'decidedly mixed results' in basic validation, and cautions that because submissions contain 32x32 averaged rain rates, 'misalignments in the location and extent of the rain bands in the model estimates may not significantly affect the CRPS metric considered in the competition.' These caveats bear on whether the cGAN contributes positively to the end-to-end score, but they do not show that a prediction was defined in terms of its own output or that a fitted value was renamed as a prediction. There are no load-bearing self-citations: the model adapts publicly described Pix2Pix and pySTEPS components, and the cited Weather4Cast competition paper is an external benchmark reference, not an unverified author-specific result used to force the conclusion. No circular step can be exhibited with a specific reduction, so the appropriate score is 0.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central claim rests on a chain of domain assumptions about cloud physics and optical flow that the paper explicitly acknowledges are ad hoc or imperfect. The only quantitative anchor is the externally scored CRPS from the competition leaderboard, which is why the circularity burden is low despite several hand-chosen preprocessing constants.

free parameters (5)

IR channel set and averaging = IR_097, IR_108, IR_120, IR_134 averaged
Chosen by hand from a correlation matrix (Table 1); the paper states the two window channels are most reliable and adds neighboring absorption channels, then averages them. This reduces input dimensionality but discards potentially useful spectral information.
Otsu threshold = computed per image
Threshold derived from the radiance histogram per image to segment cloudy vs cloud-free regions; it is a data-dependent choice not validated against other segmentation methods.
Input normalization constant = 150
Dividing radiance pixel values by 150 and subtracting 1 maps to [-1,1]; chosen ad hoc to match a plausible radiance range.
Rainfall normalization constant = 5
Scaling OPERA rates by dividing by 5 and subtracting 1; chosen ad hoc and affects the balance of the pixel-wise loss.
Model hyperparameters = 200 epochs, batch size 16, Adam lr 2e-4, cyclic schedule
Standard settings but no hyperparameter search or sensitivity analysis is reported.

assumptions (4)

domain assumption Cloud-top temperature is a sufficient proxy for surface rainfall.
The paper discards visible and water-vapor channels and relies on IR window channels; this neglects warm-rain processes and is acknowledged in Section 4 as a cause of false rain detections over deep clouds.
domain assumption Optical flow extrapolation of radiance images over 4 hours is a valid forward model.
Used in Section 2 to create inputs for the cGAN; the paper's own Figure 3 shows increasing fragmentation of cloud systems with lead time, indicating the extrapolation degrades.
domain assumption Otsu thresholding identifies all and only rain-relevant cloudy regions.
Cloud-free pixels are set to the maximum radiance value, implying they carry no rainfall information; the assumption is not tested and may erase developing convection.
domain assumption Per-frame translation from radiance to rainfall is temporally independent.
The generator maps single radiance frames to single rain frames (Section 2.2); the paper notes this is sub-optimal because temporal evolution matters.

how reviews work

0 comments

Cite this review

Pith. "Pith review of A conditional Generative Adversarial network model for the Weather4Cast 2024 Challenge." pith.science (2026). https://pith.science/paper/JSHSDQSK

@misc{pith2026241200451,
  author       = {Pith},
  title        = {Pith review of: A conditional Generative Adversarial network model for the Weather4Cast 2024 Challenge},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JSHSDQSK}},
  note         = {Machine review of arXiv:2412.00451}
}

read the original abstract

This study explores the application of deep learning for rainfall prediction, leveraging the Spinning Enhanced Visible and Infrared Imager (SEVIRI) High rate information transmission (HRIT) data as input and the Operational Program on the Exchange of weather RAdar information (OPERA) ground-radar reflectivity data as ground truth. We use the mean of 4 InfraRed frequency channels as the input. The radiance images are forecasted up to 4 hours into the future using a dense optical flow algorithm. A conditional generative adversarial network (GAN) model is employed to transform the predicted radiance images into rainfall images which are aggregated over the 4 hour forecast period to generate cumulative rainfall values. This model scored a value of approximately 7.5 as the Continuous Ranked Probability Score (CRPS) in the Weather4Cast 2024 competition and placed 1st on the core challenge leaderboard.

Figures

Figures reproduced from arXiv: 2412.00451 by the authors.

**Figure 1.** Conditional GAN Architecture 2.2 Training Procedure The training process involved a single input frame as input and a single corresponding OPERA rainfall frame as output. The trained model was then applied to the input frames generated through optical flow to predict rainfall rates for the 16 future time steps. The generator was then trained to predict rainfall maps (OPERA) for each SEVIRI frame, effectively learnin… view at source ↗

**Figure 2.** Sample predicted image after performing optical flow with a 2 hour lead time. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Illustration of behavior of Optical flow algorithm [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

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

A Space-Time Transformer for Precipitation Nowcasting
cs.CV 2025-11 conditional novelty 5.0 of 10

A full space-time attention video transformer recast as 64-class rainfall prediction with log-frequency class weighting won the Weather4Cast 2025 Cumulative Rainfall challenge (CRPS 3.135).

Reference graph

Works this paper leans on

22 extracted references · 20 canonical work pages · cited by 1 Pith paper

[1]

Paul Menzel, Frances C

W. Paul Menzel, Frances C. Holt, Timothy J. Schmit, Robert M. Aune, Anthony J. Schreiner, Gary S. Wade, and Donald G. Gray. Application of GOES-8 / 9 Soundings to Weather Forecasting and Nowcasting. Bulletin of the American Meteorological Society, 79(10):2059–2077, October 1998

work page 1998
[2]

Nystrom, Xingchao Chen, Shian-Jian Lin, and Lucas M

Fuqing Zhang, Masashi Minamide, Robert G. Nystrom, Xingchao Chen, Shian-Jian Lin, and Lucas M. Har- ris. Improving Harvey Forecasts with Next-Generation Weather Satellites: Advanced Hurricane Analysis and Prediction with Assimilation of GOES-R All-Sky Radiances. Bulletin of the American Meteorological Society, 100(7):1217–1222, July 2019

work page 2019
[3]

Kreil, and Sepp Hochreiter

Aleksandra Gruca, Pilar Rípodas, Xavier Calbet, Llorenç Lliso, Federico Serva, Bertrand Le Saux, David P. Kreil, and Sepp Hochreiter. Weather4cast 2024 – Multi-task Challenges for Rain Movie Prediction on the Road to Hi-Res Foundation Models

work page 2024
[4]

Bipasha Paul Shukla, C. M. Kishtawal, and Pradip K. Pal. Satellite-Based Nowcasting of Extreme Rainfall Events Over Western Himalayan Region. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 10(5):1681–1686, May 2017

work page 2017
[5]

Operational Application of Optical Flow Techniques to Radar-Based Rainfall Nowcasting

Wang-chun Woo and Wai-kin Wong. Operational Application of Optical Flow Techniques to Radar-Based Rainfall Nowcasting. Atmosphere, 8(3):48, March 2017. 5 cGAN Model for W4C’24 Challenge Deshpande et al

work page 2017
[6]

I. V . Sideris, L. Foresti, D. Nerini, and U. Germann. NowPrecip: localized precipitation nowcasting in the complex terrain of Switzerland. Quarterly Journal of the Royal Meteorological Society, 146(729):1768–1800, April 2020

work page 2020
[7]

Subpixel- Based Precipitation Nowcasting with the Pyramid Lucas–Kanade Optical Flow Technique

Ling Li, Zhengwei He, Sheng Chen, Xiongfa Mai, Asi Zhang, Baoqing Hu, Zhi Li, and Xinhua Tong. Subpixel- Based Precipitation Nowcasting with the Pyramid Lucas–Kanade Optical Flow Technique. Atmosphere, 9(7):260, July 2018

work page 2018
[8]

A rain-type adaptive optical flow method and its application in tropical cyclone rainfall nowcasting

Jiakai Zhu and Jianhua Dai. A rain-type adaptive optical flow method and its application in tropical cyclone rainfall nowcasting. Frontiers of Earth Science, 16(2):248–264, June 2022

work page 2022

Show all 22 references

[9]

Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting

Xingjian Shi, Zhourong Chen, Hao Wang, Dit-Yan Yeung, Wai-kin Wong, and Wang-chun Woo. Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting. In Advances in Neural Information Processing Systems, volume 28. Curran Associates, Inc., 2015

2015
[10]

Skillful precipitation nowcasting using deep generative models

Shreya Ravuri, Piotr Miłos, Johannes Langguth, Guillem Alenya, Roger Casals, Alexander Pritzel, Daniel Guo, Tobias Pfaff, Christopher Burgess, Pierre Humbert, et al. Skillful precipitation nowcasting using deep generative models. Nature, 597(7878):672–677, 2021

2021
[11]

A Deep Learning Model for Precipitation Nowcasting Using Multiple Optical Flow Algorithms

Ji-Hoon Ha and Hyesook Lee. A Deep Learning Model for Precipitation Nowcasting Using Multiple Optical Flow Algorithms. Weather and Forecasting, 39(1):41–53, December 2023

2023
[12]

OFAF-ConvLSTM: An Optical Flow Attention Fusion-ConvLSTM Model for Precipitation Nowcasting

Tengfei Nie, Xiang Ji, and YuYing Pang. OFAF-ConvLSTM: An Optical Flow Attention Fusion-ConvLSTM Model for Precipitation Nowcasting. In 2021 3rd International Academic Exchange Conference on Science and Technology Innovation (IAECST), pages 283–286, December 2021

2021
[13]

An introduction to meteosat second generation (msg)

Johannes Schmetz, Paul Pili, Stéphane Tjemkes, Dennis Just, Jochen Kerkmann, Simone Rota, and Alain Ratier. An introduction to meteosat second generation (msg). Bulletin of the American Meteorological Society, 83(7):977–992, 2002

2002
[14]

Opera the radar project

Elena Saltikoff, Günther Haase, Laurent Delobbe, Nicolas Gaussiat, Maud Martet, Daniel Idziorek, Hidde Leijnse, Petr Novák, Maryna Lukach, and Klaus Stephan. Opera the radar project. Atmosphere, 10(6):320, 2019

2019
[15]

The operational weather radar network in europe

Asko Huuskonen, Elena Saltikoff, and Iwan Holleman. The operational weather radar network in europe. Bulletin of the American Meteorological Society, 95(6):897–907, 2014

2014
[16]

A threshold selection method from gray-level histograms

Nobuyuki Otsu et al. A threshold selection method from gray-level histograms. Automatica, 11(285-296):23–27, 1975

1975
[17]

G. Bradski. The OpenCV Library. Dr. Dobb’s Journal of Software Tools, 2000

2000
[18]

pysteps - a Community-Driven Open-Source Library for Precipitation Nowcasting

Seppo Pulkkinen, Daniele Nerini, A Pérez Hortal, C Velasco-Forero, A Seed, Urs Germann, and Loris Foresti. pysteps - a Community-Driven Open-Source Library for Precipitation Nowcasting. 2019

2019
[19]

Phillip Isola, Jun-Yan Zhu, Tinghui Zhou, and Alexei A. Efros. Image-to-Image Translation with Conditional Adversarial Networks, 2016

2016
[20]

RainNet v1.0: a convolutional neural network for radar-based precipitation nowcasting

Georgy Ayzel, Tobias Scheffer, and Maik Heistermann. RainNet v1.0: a convolutional neural network for radar-based precipitation nowcasting. Geoscientific Model Development, 13(6):2631–2644, June 2020

2020
[21]

Patch-based generative adversarial network for single image haze removal

Qianli Jia and Zhikang Ma. Patch-based generative adversarial network for single image haze removal. In 2020 International Conference on Computer Engineering and Application (ICCEA), pages 882–886, 2020

2020
[22]

Estimation of the Continuous Ranked Probability Score with Limited Information and Applications to Ensemble Weather Forecasts

Michaël Zamo and Philippe Naveau. Estimation of the Continuous Ranked Probability Score with Limited Information and Applications to Ensemble Weather Forecasts. Mathematical Geosciences, 50(2):209–234, February 2018. 6

2018

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

Paul Menzel, Frances C

W. Paul Menzel, Frances C. Holt, Timothy J. Schmit, Robert M. Aune, Anthony J. Schreiner, Gary S. Wade, and Donald G. Gray. Application of GOES-8 / 9 Soundings to Weather Forecasting and Nowcasting. Bulletin of the American Meteorological Society, 79(10):2059–2077, October 1998

work page 1998

[2] [2]

Nystrom, Xingchao Chen, Shian-Jian Lin, and Lucas M

Fuqing Zhang, Masashi Minamide, Robert G. Nystrom, Xingchao Chen, Shian-Jian Lin, and Lucas M. Har- ris. Improving Harvey Forecasts with Next-Generation Weather Satellites: Advanced Hurricane Analysis and Prediction with Assimilation of GOES-R All-Sky Radiances. Bulletin of the American Meteorological Society, 100(7):1217–1222, July 2019

work page 2019

[3] [3]

Kreil, and Sepp Hochreiter

Aleksandra Gruca, Pilar Rípodas, Xavier Calbet, Llorenç Lliso, Federico Serva, Bertrand Le Saux, David P. Kreil, and Sepp Hochreiter. Weather4cast 2024 – Multi-task Challenges for Rain Movie Prediction on the Road to Hi-Res Foundation Models

work page 2024

[4] [4]

Bipasha Paul Shukla, C. M. Kishtawal, and Pradip K. Pal. Satellite-Based Nowcasting of Extreme Rainfall Events Over Western Himalayan Region. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 10(5):1681–1686, May 2017

work page 2017

[5] [5]

Operational Application of Optical Flow Techniques to Radar-Based Rainfall Nowcasting

Wang-chun Woo and Wai-kin Wong. Operational Application of Optical Flow Techniques to Radar-Based Rainfall Nowcasting. Atmosphere, 8(3):48, March 2017. 5 cGAN Model for W4C’24 Challenge Deshpande et al

work page 2017

[6] [6]

I. V . Sideris, L. Foresti, D. Nerini, and U. Germann. NowPrecip: localized precipitation nowcasting in the complex terrain of Switzerland. Quarterly Journal of the Royal Meteorological Society, 146(729):1768–1800, April 2020

work page 2020

[7] [7]

Subpixel- Based Precipitation Nowcasting with the Pyramid Lucas–Kanade Optical Flow Technique

Ling Li, Zhengwei He, Sheng Chen, Xiongfa Mai, Asi Zhang, Baoqing Hu, Zhi Li, and Xinhua Tong. Subpixel- Based Precipitation Nowcasting with the Pyramid Lucas–Kanade Optical Flow Technique. Atmosphere, 9(7):260, July 2018

work page 2018

[8] [8]

A rain-type adaptive optical flow method and its application in tropical cyclone rainfall nowcasting

Jiakai Zhu and Jianhua Dai. A rain-type adaptive optical flow method and its application in tropical cyclone rainfall nowcasting. Frontiers of Earth Science, 16(2):248–264, June 2022

work page 2022