pith. sign in

arxiv: 2606.06348 · v1 · pith:FCIHMZFZnew · submitted 2026-06-04 · 💻 cs.LG

Performance Evaluation of GraphCast for Medium-Range Weather Forecasting over Brazil

Wolfgang R. Rowell Jr. , Lucas S. Kupssinsk\"u This is my paper

Pith reviewed 2026-06-28 02:59 UTC · model grok-4.3

classification 💻 cs.LG
keywords GraphCastmachine learning weather predictionBrazilmedium-range forecastingECMWF IFSregime-dependent skillaustral winteraustral summer
0
0 comments X

The pith

GraphCast underperforms the IFS on winter medium-range Z500 over southern Brazil but regains skill at longer leads and leads on summer moisture transport.

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

The paper benchmarks GraphCast against the deterministic ECMWF IFS high-resolution forecast across four Brazilian climatic sub-regions using IFS analysis as ground truth. It computes error metrics on T850, Q850 and Z500 over selected seasonal windows. The results show a regime-dependent pattern: GraphCast has more difficulty resolving fast baroclinic systems at medium range in austral winter but benefits from smoothing at extended range, while in the summer wet season it captures large-scale moisture transport more accurately yet damps high-frequency convective signals that degrade temperature forecasts. These findings supply a regional performance baseline and identify the physical regimes where further adaptation of ML weather models would be needed.

Core claim

GraphCast exhibits a regime-dependent skill profile relative to the IFS. During austral winter it underperforms on Z500 in the medium range (lead days 2-7) when resolving fast-propagating baroclinic systems over southern Brazil, yet regains an advantage in the extended range where its smoothing of chaotic small-scale variability improves deterministic scores. During the austral summer wet season it accurately captures large-scale moisture transport while intrinsically dampening the high-frequency convective variability that degrades deterministic NWP temperature forecasts.

What carries the argument

The regime-dependent skill comparison of GraphCast versus IFS across seasons, variables and Brazilian sub-regions, computed from statistical metrics against operational IFS analysis.

If this is right

  • GraphCast's smoothing effect becomes beneficial for deterministic skill at extended lead times.
  • The model captures large-scale moisture transport more reliably than IFS during the summer wet season.
  • High-frequency convective variability is intrinsically reduced, improving temperature forecast consistency in summer.
  • The identified seasonal and regional boundaries can guide targeted adaptation of ML weather models for tropical and subtropical domains.

Where Pith is reading between the lines

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

  • Similar regime-dependent evaluations could be performed for other convective regions in the Global South to test generality.
  • Ensemble versions of GraphCast might reduce the observed damping of convective variability while retaining the smoothing benefit.
  • The same seasonal contrast might appear in other chaotic prediction tasks where one model smooths small scales.
  • Replacing the IFS analysis reference with purely observational verification would provide a stronger test of real-world utility.

Load-bearing premise

The operational IFS analysis is treated as accurate ground truth for calculating statistical metrics for both models.

What would settle it

Recomputing the same metrics against an independent observational dataset (for example radiosonde or satellite retrievals withheld from IFS) would show whether the reported relative skill differences persist.

read the original abstract

The paradigm of global weather forecasting is rapidly shifting with the emergence of Machine Learning Weather Prediction models (MLWP). While these data-driven architectures demonstrate remarkable global skill, regional benchmarks in the Global South remain scarce, leaving their efficacy in complex, highly convective environments largely unverified. This study evaluates the performance of GraphCast operational against the deterministic ECMWF IFS HRES as baseline across four distinct Brazilian climatic sub-regions. Utilizing a scalable, cloud-native pipeline and the WeatherBench-X framework for benchmarking weather models, we assess selected tropospheric variables ($T_{850}$, $Q_{850}$, $Z_{500}$) over four selected seasonal windows, employing the operational IFS analysis as the ground truth to calculate the statistical metrics for both models. Results reveal a regime-dependent skill profile. During the austral winter, GraphCast underperforms in the medium range (lead days 2-7) for $Z_{500}$ when resolving fast-propagating baroclinic systems over southern Brazil, but regains an advantage in the extended range, where its inherent smoothing of chaotic small-scale variability becomes beneficial under deterministic skill metrics. Conversely, during the austral summer wet season, GraphCast accurately captures large-scale moisture transport while intrinsically dampening the high-frequency convective variability that degrades deterministic NWP temperature forecasts. These findings establish a baseline for Brazil and define the specific physical boundaries that will guide future ``tropicalization'' efforts, aiming to optimize these foundational AI models for regional resilience.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

1 major / 1 minor

Summary. The manuscript evaluates GraphCast against deterministic ECMWF IFS HRES over four Brazilian climatic sub-regions using a cloud-native pipeline and WeatherBench-X. It employs operational IFS analysis as ground truth for T_850, Q_850, and Z_500 across seasonal windows and reports a regime-dependent skill profile: GraphCast underperforms on Z_500 for fast-propagating baroclinic systems in austral winter medium range (lead days 2-7) but regains advantage in the extended range due to smoothing; in the austral summer wet season it captures large-scale moisture transport while damping high-frequency convective variability that affects NWP temperature forecasts.

Significance. If the central claims hold after addressing verification concerns, the work supplies a needed regional baseline for MLWP models in the Global South and identifies concrete physical regimes (baroclinic systems, convective variability) that can guide targeted tropicalization. The scalable pipeline and WeatherBench-X usage are strengths for reproducibility.

major comments (1)
  1. [Abstract] Abstract: The regime-dependent skill claims (GraphCast underperformance on Z_500 for baroclinic systems in austral winter medium range, advantage in extended range; superior moisture transport but damped convection in summer) are computed exclusively against operational IFS analysis as ground truth for both models. Because IFS HRES is dynamically consistent with the IFS assimilation system while GraphCast was trained on ERA5, systematic differences in analysis increments or small-scale representation will favor IFS HRES in deterministic scores; this choice directly underpins the reported performance boundaries and is not neutralized by the four sub-regions or seasonal windows.
minor comments (1)
  1. [Abstract] Abstract: The description of the pipeline and framework is high-level; the full text should include explicit references to the exact seasonal windows, spatial domains of the four sub-regions, and the precise lead-time ranges used for the medium- versus extended-range comparisons.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive comment on verification methodology. We address the concern point-by-point below.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The regime-dependent skill claims (GraphCast underperformance on Z_500 for baroclinic systems in austral winter medium range, advantage in extended range; superior moisture transport but damped convection in summer) are computed exclusively against operational IFS analysis as ground truth for both models. Because IFS HRES is dynamically consistent with the IFS assimilation system while GraphCast was trained on ERA5, systematic differences in analysis increments or small-scale representation will favor IFS HRES in deterministic scores; this choice directly underpins the reported performance boundaries and is not neutralized by the four sub-regions or seasonal windows.

    Authors: We agree this is a substantive methodological limitation. Using operational IFS analysis as ground truth for both models creates a consistency advantage for HRES that is not present for GraphCast (trained on ERA5), and the sub-regional/seasonal stratification does not remove this bias. In the revised manuscript we will add an explicit discussion of this issue in the Methods and Discussion sections, including its potential effect on the reported regime-dependent skill differences. We will also note the limitation briefly in the abstract. A full re-computation against ERA5 is not feasible within the current experimental design due to data access constraints for the full set of lead times and variables, but we will flag this as an important direction for follow-on work. revision: partial

Circularity Check

0 steps flagged

No circularity detected in evaluation methodology

full rationale

The paper is a performance benchmark comparing GraphCast and IFS HRES against operational IFS analysis as external ground truth, using standard statistical metrics (e.g., for T850, Q850, Z500) over fixed regions and seasons. No equations, predictions, or central claims reduce by construction to fitted parameters, self-definitions, or self-citation chains. The regime-dependent skill profiles are computed outputs from these independent comparisons rather than inputs renamed as results. The methodology is self-contained against external benchmarks with no load-bearing self-referential steps.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Only the abstract is available; no equations, full methods, or additional details are provided to identify free parameters or further axioms.

axioms (1)
  • domain assumption The operational IFS analysis serves as ground truth for verification metrics.
    Explicitly stated in the abstract as the basis for calculating statistical metrics for both models.

pith-pipeline@v0.9.1-grok · 5797 in / 1283 out tokens · 57080 ms · 2026-06-28T02:59:47.547666+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

75 extracted references · 29 canonical work pages · 1 internal anchor

  1. [1]

    Essentials of meteorology: An invitation to the atmosphere

    Ahrens, C Donald and Henson, Robert. Essentials of meteorology: An invitation to the atmosphere

  2. [2]

    Climatology

    Rohli, Robert V and Vega, Anthony J. Climatology

  3. [3]

    Firmament: The hidden science of weather, climate change and the air that surrounds us

    Clark, Simon. Firmament: The hidden science of weather, climate change and the air that surrounds us

  4. [4]

    Dynamics of The Tropical Atmosphere and Oceans

    Webster, Peter J. Dynamics of The Tropical Atmosphere and Oceans

  5. [5]

    Fundamentals of numerical weather prediction

    Coiffier, Jean. Fundamentals of numerical weather prediction

  6. [6]

    The origins of computer weather prediction and climate modeling , volume =

    Lynch, Peter , year =. The origins of computer weather prediction and climate modeling , volume =. Journal of Computational Physics , publisher =. doi:10.1016/j.jcp.2007.02.034 , number =

    work page doi:10.1016/j.jcp.2007.02.034 2007

  7. [7]

    2024 , url =

    ECMWF , title =. 2024 , url =

    2024

  8. [8]

    doi:10.21957/85FC99659C , url =

    de Elia, Ramon and Haiden, Thomas and Matsudo, Cynthia and Otero, Federico and Gascon, Estibaliz and Castro, Lucia and Magnusson, Linus and Godoy, Alejandro and Coelho, Caio and Bechis, Hernan and Casati, Barbara and D'Amen, Daniela and Frassoni, Ariane and Galgano, Santiago and Osman, Marisol and Osores, Soledad and Righetti, Silvina and Salio, Paola and...

    work page doi:10.21957/85fc99659c

  9. [9]

    doi:10.21957/OPEN-DATA , url =

    Open data , howpublished =. doi:10.21957/OPEN-DATA , url =

    work page doi:10.21957/open-data

  10. [10]

    2025 , url =

    ECMWF , title =. 2025 , url =

    2025

  11. [11]

    2002 , url =

    BUIZZA, Roberto , title =. 2002 , url =

    2002

  12. [12]

    2001 , url =

    BUIZZA, Roberto , title =. 2001 , url =

    2001

  13. [13]

    2012 , url =

    ECMWF , title =. 2012 , url =

    2012

  14. [14]

    2020 , url =

    ECMWF , title =. 2020 , url =

    2020

  15. [15]

    2025 , url =

    Copernicus , title =. 2025 , url =

    2025

  16. [16]

    2021 , url =

    INPE , title =. 2021 , url =

    2021

  17. [17]

    2021 , url =

    CPTEC , title =. 2021 , url =

    2021

  18. [18]

    2011 , url =

    CPTEC , title =. 2011 , url =

    2011

  19. [19]

    2012 , url =

    IGBE , title =. 2012 , url =

    2012

  20. [20]

    Ben Bouallègue, Zied and Clare, Mariana C. A. and Magnusson, Linus and Gascón, Estibaliz and Maier-Gerber, Michael and Janoušek, Martin and Rodwell, Mark and Pinault, Florian and Dramsch, Jesper S. and Lang, Simon T. K. and Raoult, Baudouin and Rabier, Florence and Chevallier, Matthieu and Sandu, Irina and Dueben, Peter and Chantry, Matthew and Pappenberg...

    work page doi:10.1175/bams-d-23-0162.1

  21. [21]

    2023 , url =

    Hoyer, Stephan , title =. 2023 , url =

    2023

  22. [22]

    2024 , url =

    Rasp, Stephan , title =. 2024 , url =

    2024

  23. [23]

    AI-Weather SotA vs Time , publisher =

    Rasp, Stephan , keywords =. AI-Weather SotA vs Time , publisher =. 2024 , copyright =. doi:10.6084/M9.FIGSHARE.28083515.V1 , url =

    work page doi:10.6084/m9.figshare.28083515.v1 2024

  24. [24]

    2023 , url =

    Google Research , title =. 2023 , url =

    2023

  25. [25]

    2025 , url =

    Google Research , title =. 2025 , url =

    2025

  26. [26]

    and Scher, Sebastian and Weyn, Jonathan A

    Rasp, Stephan and Dueben, Peter D. and Scher, Sebastian and Weyn, Jonathan A. and Mouatadid, Soukayna and Thuerey, Nils , keywords =. WeatherBench: A benchmark dataset for data-driven weather forecasting , publisher =. 2020 , copyright =. doi:10.48550/ARXIV.2002.00469 , url =

    work page doi:10.48550/arxiv.2002.00469 2020

  27. [27]

    2024 , eprint=

    WeatherBench 2: A benchmark for the next generation of data-driven global weather models , author=. 2024 , eprint=

    2024

  28. [28]

    and Bauer, Peter , year =

    Dueben, Peter D. and Bauer, Peter , year =. Challenges and design choices for global weather and climate models based on machine learning , volume =. Geoscientific Model Development , publisher =. doi:10.5194/gmd-11-3999-2018 , number =

    work page doi:10.5194/gmd-11-3999-2018 2018

  29. [29]

    and Durran, Dale R

    Weyn, Jonathan A. and Durran, Dale R. and Caruana, Rich , year =. Can Machines Learn to Predict Weather? Using Deep Learning to Predict Gridded 500‐hPa Geopotential Height From Historical Weather Data , volume =. Journal of Advances in Modeling Earth Systems , publisher =. doi:10.1029/2019ms001705 , number =

    work page doi:10.1029/2019ms001705

  30. [30]

    and Scher, Sebastian and Weyn, Jonathan A

    Rasp, Stephan and Dueben, Peter D. and Scher, Sebastian and Weyn, Jonathan A. and Mouatadid, Soukayna and Thuerey, Nils , year =. WeatherBench: A Benchmark Data Set for Data‐Driven Weather Forecasting , volume =. Journal of Advances in Modeling Earth Systems , publisher =. doi:10.1029/2020ms002203 , number =

    work page doi:10.1029/2020ms002203

  31. [31]

    Data-driven medium-range weather prediction with a Resnet pretrained on climate simulations: A new model for WeatherBench , publisher =

    Rasp, Stephan and Thuerey, Nils , keywords =. Data-driven medium-range weather prediction with a Resnet pretrained on climate simulations: A new model for WeatherBench , publisher =. 2020 , copyright =. doi:10.48550/ARXIV.2008.08626 , url =

    work page doi:10.48550/arxiv.2008.08626 2020

  32. [32]

    and Durran, Dale R

    Weyn, Jonathan A. and Durran, Dale R. and Caruana, Rich , year =. Improving Data‐Driven Global Weather Prediction Using Deep Convolutional Neural Networks on a Cubed Sphere , volume =. Journal of Advances in Modeling Earth Systems , publisher =. doi:10.1029/2020ms002109 , number =

    work page doi:10.1029/2020ms002109

  33. [33]

    and Jamil, Omar and Morcrette, Cyril J

    Clare, Mariana C.A. and Jamil, Omar and Morcrette, Cyril J. , year =. Combining distribution‐based neural networks to predict weather forecast probabilities , volume =. Quarterly Journal of the Royal Meteorological Society , publisher =. doi:10.1002/qj.4180 , number =

    work page doi:10.1002/qj.4180

  34. [34]

    and Durran, Dale R

    Weyn, Jonathan A. and Durran, Dale R. and Caruana, Rich and Cresswell‐Clay, Nathaniel , year =. Sub‐Seasonal Forecasting With a Large Ensemble of Deep‐Learning Weather Prediction Models , volume =. Journal of Advances in Modeling Earth Systems , publisher =. doi:10.1029/2021ms002502 , number =

    work page doi:10.1029/2021ms002502

  35. [35]

    Forecasting Global Weather with Graph Neural Networks , publisher =

    Keisler, Ryan , keywords =. Forecasting Global Weather with Graph Neural Networks , publisher =. 2022 , copyright =. doi:10.48550/ARXIV.2202.07575 , url =

    work page doi:10.48550/arxiv.2202.07575 2022

  36. [36]

    FourCastNet: A Global Data-driven High-resolution Weather Model using Adaptive Fourier Neural Operators

    Pathak, Jaideep and Subramanian, Shashank and Harrington, Peter and Raja, Sanjeev and Chattopadhyay, Ashesh and Mardani, Morteza and Kurth, Thorsten and Hall, David and Li, Zongyi and Azizzadenesheli, Kamyar and Hassanzadeh, Pedram and Kashinath, Karthik and Anandkumar, Animashree , keywords =. FourCastNet: A Global Data-driven High-resolution Weather Mod...

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2202.11214 2022

  37. [37]

    Lorenz , title =

    E.N. Lorenz , title =. 1995 , journal =

    1995

  38. [38]

    The ERA5 global reanalysis , volume =

    Hersbach, Hans and Bell, Bill and Berrisford, Paul and Hirahara, Shoji and Horányi, András and Muñoz Sabater, Joaquín and Nicolas, Julien and Peubey, Carole and Radu, Raluca and Schepers, Dinand and Simmons, Adrian and Soci, Cornel and Abdalla, Saleh and Abellan, Xavier and Balsamo, Gianpaolo and Bechtold, Peter and Biavati, Gionata and Bidlot, Jean and B...

  39. [39]

    , year = 2015, month = sep, volume = 525, pages =

    Bauer, Peter and Thorpe, Alan and Brunet, Gilbert , year =. The quiet revolution of numerical weather prediction , volume =. Nature , publisher =. doi:10.1038/nature14956 , number =

    work page doi:10.1038/nature14956

  40. [40]

    2020 , url =

    Copernicus.EU , title =. 2020 , url =

    2020

  41. [41]

    2019 , url =

    Copernicus.EU , title =. 2019 , url =

    2019

  42. [42]

    2019 , url =

    World Meteorological Organization , title =. 2019 , url =

    2019

  43. [43]

    Accurate medium-range global weather forecasting with 3D neural networks , volume =

    Bi, Kaifeng and Xie, Lingxi and Zhang, Hengheng and Chen, Xin and Gu, Xiaotao and Tian, Qi , year =. Accurate medium-range global weather forecasting with 3D neural networks , volume =. Nature , publisher =. doi:10.1038/s41586-023-06185-3 , number =

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

  44. [44]

    Liu, Cheng-Chin and Hsu, Kathryn and Peng, Melinda S. and Chen, Der-Song and Chang, Pao-Liang and Hsiao, Ling-Feng and Fong, Chin-Tzu and Hong, Jing-Shan and Cheng, Chia-Ping and Lu, Kuo-Chen and Chen, Chia-Rong and Kuo, Hung-Chi , year =. Evaluation of five global AI models for predicting weather in Eastern Asia and Western Pacific , volume =. npj Climat...

    work page doi:10.1038/s41612-024-00769-0

  45. [45]

    and Kirchengast, Gottfried and Schw\"

    Scherllin‐Pirscher, Barbara and Steiner, Andrea K. and Kirchengast, Gottfried and Schw\". The power of vertical geolocation of atmospheric profiles from GNSS radio occultation , volume =. Journal of Geophysical Research: Atmospheres , publisher =. 2017 , month = feb, pages =. doi:10.1002/2016jd025902 , number =

    work page doi:10.1002/2016jd025902 2017

  46. [46]

    Learning skillful medium-range global weather forecasting , volume =

    Lam, Remi and Sanchez-Gonzalez, Alvaro and Willson, Matthew and Wirnsberger, Peter and Fortunato, Meire and Alet, Ferran and Ravuri, Suman and Ewalds, Timo and Eaton-Rosen, Zach and Hu, Weihua and Merose, Alexander and Hoyer, Stephan and Holland, George and Vinyals, Oriol and Stott, Jacklynn and Pritzel, Alexander and Mohamed, Shakir and Battaglia, Peter ...

    work page doi:10.1126/science.adi2336

  47. [47]

    and El-Kadi, Andrew and Masters, Dominic and Ewalds, Timo and Stott, Jacklynn and Mohamed, Shakir and Battaglia, Peter and Lam, Remi and Willson, Matthew , year =

    Price, Ilan and Sanchez-Gonzalez, Alvaro and Alet, Ferran and Andersson, Tom R. and El-Kadi, Andrew and Masters, Dominic and Ewalds, Timo and Stott, Jacklynn and Mohamed, Shakir and Battaglia, Peter and Lam, Remi and Willson, Matthew , year =. Probabilistic weather forecasting with machine learning , volume =. Nature , publisher =. doi:10.1038/s41586-024-...

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

  48. [48]

    Neural general circulation models for weather and climate , volume =

    Kochkov, Dmitrii and Yuval, Janni and Langmore, Ian and Norgaard, Peter and Smith, Jamie and Mooers, Griffin and Kl\". Neural general circulation models for weather and climate , volume =. Nature , publisher =. 2024 , month = jul, pages =. doi:10.1038/s41586-024-07744-y , number =

    work page doi:10.1038/s41586-024-07744-y 2024

  49. [49]

    Lang, Simon and Alexe, Mihai and Chantry, Matthew and Dramsch, Jesper and Pinault, Florian and Raoult, Baudouin and Clare, Mariana C. A. and Lessig, Christian and Maier-Gerber, Michael and Magnusson, Linus and Bouallègue, Zied Ben and Nemesio, Ana Prieto and Dueben, Peter D. and Brown, Andrew and Pappenberger, Florian and Rabier, Florence , keywords =. AI...

    work page doi:10.48550/arxiv.2406.01465 2024

  50. [50]

    Lang, Simon and Alexe, Mihai and Clare, Mariana C. A. and Roberts, Christopher and Adewoyin, Rilwan and Bouallègue, Zied Ben and Chantry, Matthew and Dramsch, Jesper and Dueben, Peter D. and Hahner, Sara and Maciel, Pedro and Prieto-Nemesio, Ana and O'Brien, Cathal and Pinault, Florian and Polster, Jan and Raoult, Baudouin and Tietsche, Steffen and Leutbe...

    work page doi:10.48550/arxiv.2412.15832 2024

  51. [51]

    Aeronautical Software - PDAS , title =

    U.S. Aeronautical Software - PDAS , title =. 1976 , url =

    1976

  52. [52]

    Um Modelo Metodológico de Classificação de Climas , volume =

    Nimer, Edmon , year =. Um Modelo Metodológico de Classificação de Climas , volume =. Revista Brasileira de Geografia , publisher =

  53. [53]

    and Arnal, Louise and Boucher, Marie-Amélie and Chang, Annie Y.-Y

    Slater, Louise J. and Arnal, Louise and Boucher, Marie-Amélie and Chang, Annie Y.-Y. and Moulds, Simon and Murphy, Conor and Nearing, Grey and Shalev, Guy and Shen, Chaopeng and Speight, Linda and Villarini, Gabriele and Wilby, Robert L. and Wood, Andrew and Zappa, Massimiliano , year =. Hybrid forecasting: blending climate predictions with AI models , vo...

    work page doi:10.5194/hess-27-1865-2023 2023

  54. [54]

    Managing Renewable Energy Risk in the Brazilian Forward Market: The Impact of Hourly Hedging , year=

    Street, Alexandre and Fanzeres, Bruno and Carvalho, Gustavo , journal=. Managing Renewable Energy Risk in the Brazilian Forward Market: The Impact of Hourly Hedging , year=

  55. [55]

    2025 , url =

    INMET , title =. 2025 , url =

    2025

  56. [56]

    2021 , publisher =

    Cavalcanti, Iracema Fonseca de Albuquerque and Ferreira, Nelson Jesuz , title =. 2021 , publisher =

    2021

  57. [57]

    Accessing WeatherNext forecasts , year =

  58. [58]

    GRIB Parameter Database: Parameter 156 -- Geopotential Height , year =

  59. [59]

    2006 , publisher =

    Atmospheric Science: An Introductory Survey , author =. 2006 , publisher =

    2006

  60. [60]

    , year =

    Geer, Alan J. , year =. Significance of changes in medium-range forecast scores , volume =. Tellus A: Dynamic Meteorology and Oceanography , publisher =. doi:10.3402/tellusa.v68.30229 , number =

    work page doi:10.3402/tellusa.v68.30229

  61. [61]

    2024 , month =

    Chantry, Matthew and Adewoyin, Rilwan and Alexe, Mihai and Ben Bouallègue, Zied and Clare, Mariana and Dramsch, Jesper and Hahner, Sara and Lang, Simon and Lessig, Christian and Magnusson, Linus and Maier-Gerber, Michael and Mertes, Gert and Moldovan, Gabriel and Prieto Nemesio, Ana and O'Brien, Cathal and Pinault, Florian and Raoult, Baudouin and Santa C...

    work page doi:10.21957/c45e3f67e9 2024

  62. [62]

    On Some Limitations of Current Machine Learning Weather Prediction Models , volume =

    Bonavita, Massimo , year =. On Some Limitations of Current Machine Learning Weather Prediction Models , volume =. Geophysical Research Letters , publisher =. doi:10.1029/2023gl107377 , number =

    work page doi:10.1029/2023gl107377

  63. [63]

    An approximation algorithm for dissecting a rectangle into rectangles with specified areas , volume=

    Hiroshi Nagamochi and Yuusuke Abe , journal=. An approximation algorithm for dissecting a rectangle into rectangles with specified areas , volume=

  64. [64]

    and Leiserson, Charles E

    Cormen, Thomas H. and Leiserson, Charles E. and Rivest, Ronald L. and Stein, Clifford , howpublished=. Introduction to Algorithms, Second Edition , year=

  65. [65]

    Skiena , publisher=

    Steven S. Skiena , publisher=. The algorithm design manual , year=

  66. [66]

    Beautiful Code: Leading Programmers Explain How They Think , editor=

    Jon Bentley , title=. Beautiful Code: Leading Programmers Explain How They Think , editor=. 2007 , pages=

    2007

  67. [67]

    Adaptive photo collection page layout , volume=

    Brian Clayton Atkins , booktitle=. Adaptive photo collection page layout , volume=

  68. [68]

    Two algorithms for automatic document page layout , year=

    Jo. Two algorithms for automatic document page layout , year=. Proceedings of the eighth ACM symposium on Document engineering , location=

  69. [69]

    A novel physics-based interaction model for free document layout , booktitle=

    Piccoli, Ricardo Farias Bidart and Chamun, Rodrigo and Cogo, Nicole Carrion and de Oliveira, Jo. A novel physics-based interaction model for free document layout , booktitle=. 2011 , isbn=. doi:10.1145/2034691.2034725 , acmid=

    work page doi:10.1145/2034691.2034725 2011

  70. [70]

    and Prikladnicki, Rafael

    Borges, Olimar Teixeira and Couto, Julia and Ruiz, Duncan D. and Prikladnicki, Rafael. How Machine Learning has been Applied in Software Engineering?. Proceedings of the International Conference on Software Engineering and Knowledge Engineering. 2020

    2020

  71. [71]

    Um algoritmo determin\'

    Ricardo Farias Bidart Piccoli , numpages=. Um algoritmo determin\'. 2008 , url=

    2008

  72. [72]

    Automatic layout of variable-content print data , url=

    Eldan Goldenberg , month=. Automatic layout of variable-content print data , url=

  73. [73]

    Optimal pagination techniques for automatic typesetting systems , year=

    Michael Frederick Plass , numpages=. Optimal pagination techniques for automatic typesetting systems , year=

  74. [74]

    Active Layout Engine: Algorithms and Applications in Variable Data Printing , url=

    Xiaofan Lin , institution=. Active Layout Engine: Algorithms and Applications in Variable Data Printing , url=

  75. [75]

    Coffman and J

    E. Coffman and J. Galambos and S. Martello and D. Vigo , title=