Pith. sign in

REVIEW 2 major objections 1 minor 51 references

A two-layered mixed subgrid model combining AMD with the Gradient model matches DNS most closely in thermal large-eddy simulations of solar-receiver channel flows.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · grok-4.3

2026-06-26 21:57 UTC pith:SYKQ3I65

load-bearing objection The paper ranks twelve AMD-based T-LES models against DNS for anisothermal solar-receiver channel flow and flags one two-layer mixed variant as best, but the ranking rests entirely on the unshown quality of that DNS reference. the 2 major comments →

arxiv 2606.17596 v1 pith:SYKQ3I65 submitted 2026-06-16 physics.class-ph physics.flu-dyn

A posteriori study of Thermal-Large Eddy Simulation in solar receiver operating conditions

Yanis Zatout (PROMES , DATAFLOT , LISN) , Fran\c{c}oise Bataille , Adrien Toutant (UPVD , PROMES) This is my paper
classification physics.class-ph physics.flu-dyn
keywords thermal large-eddy simulationanisothermal channel flowsubgrid-scale modelingAMD modelsolar receivera posteriori evaluationmixed modellow-Mach Navier-Stokes
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper evaluates twelve subgrid-scale models, all derived from the Anisotropic Minimum Dissipation approach, inside low-Mach-number thermal large-eddy simulations of anisothermal turbulent channel flow at conditions typical of solar receivers. It computes a global error rate against existing direct numerical simulation data, then examines four models in greater detail under changes in mesh resolution and numerical scheme. The central finding is that a two-layered mixed formulation that adds the Gradient model on top of AMD for both momentum and scalar transport produces the smallest deviations from the reference DNS across the tested quantities.

Core claim

Under solar-receiver operating conditions, a two-layered mixed model that combines the AMD/AMD-scalar formulation with the Gradient model yields the best agreement with DNS data among the twelve AMD-based closures examined.

What carries the argument

The two-layered mixed AMD/AMD-scalar plus Gradient model, which supplies the subgrid stresses and scalar fluxes in the filtered low-Mach Navier-Stokes equations.

Load-bearing premise

The DNS data set provides an accurate and complete ground truth for every quantity compared in the a-posteriori tests.

What would settle it

A new direct numerical simulation at the same Reynolds and Prandtl numbers but with a different numerical method or domain size that produces mean profiles or fluctuation statistics differing from the reference DNS used here.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • The mixed model can be adopted directly for production T-LES of solar-receiver geometries at the tested resolution.
  • Coarser meshes remain usable provided the mixed model is retained, because the global error rate stays lower than for the other closures.
  • Both convective and diffusive numerical schemes interact with the model choice, so the mixed formulation should be paired with the same schemes used in the evaluation.
  • The advantage of the mixed model appears in both velocity and temperature statistics, indicating consistent improvement for the coupled thermal problem.

Where Pith is reading between the lines

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

  • The same mixed closure could be tested in geometries that include the actual absorber tubes or windows of a solar receiver rather than a periodic channel.
  • If the model remains superior when the flow includes radiation or variable properties, it would reduce the need for finer grids in engineering calculations of receiver efficiency.
  • Running the identical model set on an independent experimental data set at comparable conditions would confirm whether the ranking holds outside the DNS reference.
  • Extending the two-layer mixing strategy to other base models besides AMD might produce further gains at higher Reynolds numbers.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 1 minor

Summary. The manuscript presents an a posteriori evaluation of 12 subgrid-scale models for Thermal Large Eddy Simulation (T-LES) of anisothermal turbulent channel flows representative of solar receiver conditions. All models are variants of the Anisotropic Minimum Dissipation (AMD) model. The authors compute a global error rate to rank the models against Direct Numerical Simulation (DNS) data and select four for detailed analysis on mesh resolution, numerical schemes, and formulations. They conclude that a two-layered mixed model combining AMD/AMD-scalar with the Gradient model provides the best agreement with DNS.

Significance. If the DNS benchmark is a converged reference solution, the work supplies concrete guidance on SGS model selection for T-LES under high temperature-ratio conditions relevant to concentrated solar power. The systematic ranking of AMD-based variants and the identification of a specific mixed formulation constitute a falsifiable contribution to the literature on thermal turbulence modeling.

major comments (2)
  1. [Abstract] Abstract: The central claim that the two-layered mixed AMD/Gradient model yields the best agreement is established entirely via a posteriori comparison to DNS. The abstract supplies no information on DNS grid resolution, temporal convergence, or validation against known channel-flow statistics at the target Re, Pr, and temperature ratios. This assumption is load-bearing for the model ranking and must be documented with quantitative evidence.
  2. [Abstract] Abstract: The definition and normalization of the 'global error rate' used to evaluate and select among the 12 models are not stated. Without this, the quantitative basis for declaring one model superior cannot be assessed or reproduced.
minor comments (1)
  1. The abstract refers to 'solar receivers operating conditions' without listing the specific non-dimensional parameters (Re, Pr, temperature ratio) employed in the simulations.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We address each major comment below and will revise the abstract to improve clarity and self-containment.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The central claim that the two-layered mixed AMD/Gradient model yields the best agreement is established entirely via a posteriori comparison to DNS. The abstract supplies no information on DNS grid resolution, temporal convergence, or validation against known channel-flow statistics at the target Re, Pr, and temperature ratios. This assumption is load-bearing for the model ranking and must be documented with quantitative evidence.

    Authors: We agree that the abstract should include key quantitative details on the DNS to support the model ranking. In the revised manuscript we will add concise information on the DNS grid resolution (number of points and wall-normal spacing), temporal averaging period and convergence criteria, and confirmation that the DNS reproduces established channel-flow statistics at the target Re, Pr and temperature ratio. revision: yes

  2. Referee: [Abstract] Abstract: The definition and normalization of the 'global error rate' used to evaluate and select among the 12 models are not stated. Without this, the quantitative basis for declaring one model superior cannot be assessed or reproduced.

    Authors: The global error rate is defined in the body of the manuscript (Section 3) as a normalized aggregate L2 error across mean and rms profiles of velocity and temperature. To make the abstract self-contained we will insert a brief clause stating its definition and normalization while respecting length constraints. revision: yes

Circularity Check

0 steps flagged

No significant circularity: model ranking derived from external DNS comparison

full rationale

The paper conducts an a posteriori assessment of 12 T-LES subgrid-scale models (all variants of AMD) by direct numerical comparison to independent DNS data under solar-receiver conditions. Model selection and ranking rest on computed global error rates and detailed statistics against this external benchmark, with no parameter fitting, self-definitional equations, or load-bearing self-citations that reduce the central claim to its own inputs. The derivation chain consists of standard LES filtering, model implementation, and error quantification; none of these steps are equivalent to the reported best-model result by construction. DNS serves as an external reference, satisfying the criterion for non-circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides no information on free parameters, axioms, or invented entities; assessment limited to high-level description.

pith-pipeline@v0.9.1-grok · 5669 in / 1095 out tokens · 28013 ms · 2026-06-26T21:57:44.667104+00:00 · methodology

0 comments
read the original abstract

This study investigates Thermal-Large Eddy Simulations (T-LES) of anisothermal and turbulent channel flows under physical conditions representative of solar receivers. Solving the low-Mach number Navier-Stokes equations, T-LES results are evaluated a posteriori against Direct Numerical Simulation (DNS) data. We assess 12 subgrid-scale models. All models are based on the Anisotropic Minimum Dissipation (AMD) model. After computing a global error rate to evaluate all models, we select four for a detailed analysis regarding the effects of mesh resolution, numerical schemes, and model formulations. Results demonstrate that a two-layered mixed model combining the AMD/AMD-scalar with the Gradient model yields the best agreement with DNS.

Figures

Figures reproduced from arXiv: 2606.17596 by Adrien Toutant (UPVD, DATAFLOT, Fran\c{c}oise Bataille, LISN), PROMES), Yanis Zatout (PROMES.

Figure 1
Figure 1. Figure 1: Representation of the simulation domain 2.1. Geometry and mesh To model flows inside next-generation solar receivers, we use a bi-periodic channel, which represents the simplest possible geometry. Because the streamwise (x) and spanwise (z) directions are periodic, the grid spacing along these axes is uniform. In the wall-normal direction, the mesh is finer at the wall and coarser at the center of the chan… view at source ↗
Figure 2
Figure 2. Figure 2: Functional constant evolution as a function of the height in the channel. [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Normalized errors for the 12 tested models. The top panel represents the mean error, the middle one, the RMS error, and [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Relative error on first-order statistics ( [PITH_FULL_IMAGE:figures/full_fig_p014_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Mean streamwise velocity ⟨U⟩ +, wall-normal velocity ⟨V⟩ +, and temperature ⟨T⟩ + profiles for the selected simulations. Coarser mesh resolutions are represented by lighter line shades. 15 [PITH_FULL_IMAGE:figures/full_fig_p015_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Relative error on second-order statistics for the selected models (MA, M1, NA, F1). [PITH_FULL_IMAGE:figures/full_fig_p017_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Second-order statistics for the diagonal components of the deviatoric Reynolds stress tensor ( [PITH_FULL_IMAGE:figures/full_fig_p018_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Streamwise wall-normal Reynolds shear stress [PITH_FULL_IMAGE:figures/full_fig_p019_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Temperature transport terms (⟨U ′T ′ ⟩ +, ⟨V ′T ′ ⟩ +) and temperature variance (⟨T ′2 ⟩ +), alongside their corresponding subgrid-scale closure terms, for the selected models. 20 [PITH_FULL_IMAGE:figures/full_fig_p020_9.png] view at source ↗

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

51 extracted references · 7 canonical work pages

  1. [1]

    Abbà, A., Cercignani, A., and Valdettaro, L., Computers & Mathematics with Applications 46, 521 (2003), turbulence Modelling and Simulation

    2003

  2. [2]

    J., and Moin, P., Phys

    Abkar, M., Bae, H. J., and Moin, P., Phys. Rev. Fluids 1, 041701 (2016)

    2016

  3. [3]

    Leith, 151, 115 (2017)

    Aulery, F., Dupuy, D., Toutant, A., Bataille, F., and Zhou, Y ., Computers & Fluids A Special Issue in Honor of Cecil "Chuck" E. Leith, 151, 115 (2017)

    2017

  4. [4]

    Avellaneda, J., Bataille, F., and Toutant, A., International Journal of Heat and Fluid Flow 77, 40 (2019)

    2019

  5. [5]

    Bardina, J., Ferziger, J., and Reynolds, W., in 13th Fluid and Plasma Dynamics Conference, Fluid Dynamics and Co-Located Conferences (American Institute of Aeronautics and Astronautics, 1980)

    1980

  6. [6]

    Boussinesq, J., Essai sur la théorie des eaux courantes (Imprimerie nationale, 1877)

  7. [7]

    Calvin, C., Cueto, O., and Emonot, P., ESAIM: Mathematical Modelling and Numerical Analysis - Modélisation Mathématique et Analyse Numérique 36, 907 (2002)

    2002

  8. [8]

    Chapman, D. R. and Kuhn, G. D., Journal of Fluid Mechanics 170, 265 (1986)

    1986

  9. [9]

    David, M., Toutant, A., and Bataille, F., Physics of Fluids 33, 085111 (2021)

    2021

  10. [10]

    David, M., Toutant, A., and Bataille, F., Physics of Fluids 33, 045104 (2021)

    2021

  11. [11]

    David, M., Toutant, A., and Bataille, F., International Journal of Heat and Mass Transfer 164, 120599 (2021)

    2021

  12. [12]

    David, M., Toutant, A., and Bataille, F., International Journal of Heat and Mass Transfer 201, 123577 (2023)

    2023

  13. [13]

    David, M., Toutant, A., and Bataille, F., Physics of Fluids 35 (2023), 10.1063/5.0139433

    work page doi:10.1063/5.0139433 2023

  14. [14]

    Dupuy, D.,Analyse et Modélisation de l’interaction Entre Thermique et Turbulence Dans Les Récepteurs Solaires à Haute Température., Phd thesis, Perpignan (2018)

    2018

  15. [15]

    Dupuy, D., Toutant, A., and Bataille, F., Journal of Fluid Mechanics 834, 5 (2018)

    2018

  16. [16]

    Dupuy, D., Toutant, A., and Bataille, F., Physics of Fluids 31, 045105 (2019), https://pubs.aip.org/aip/pof/article-pdf/doi/10.1063/1.5091829/15748266/045105 _1_online.pdf

    work page doi:10.1063/1.5091829/15748266/045105 2019

  17. [17]

    Dupuy, D., Toutant, A., and Bataille, F., Physics of Fluids 31, 065113 (2019)

    2019

  18. [19]

    Dupuy, D., Toutant, A., and Bataille, F., International Journal of Thermal Sciences 145, 105999 (2019)

    2019

  19. [20]

    H., Physics of Fluids A: Fluid Dynamics 3, 1760 (1991)

    Germano, M., Piomelli, U., Moin, P., and Cabot, W. H., Physics of Fluids A: Fluid Dynamics 3, 1760 (1991)

    1991

  20. [21]

    A., and Domaradzki, J

    Hickel, S., Adams, N. A., and Domaradzki, J. A., Journal of Computational Physics 213, 413 (2006)

    2006

  21. [22]

    and Fischer, P

    Iliescu, T. and Fischer, P. F., Physics of Fluids 15, 3036 (2003), https://pubs.aip.org/aip/pof/article- pdf/15/10/3036/19131639/3036_1 _online.pdf . 22

    2003

  22. [23]

    Jiménez, J., Physica A: Statistical Mechanics and its Applications 263, 252 (1999), proceedings of the 20th IUPAP International Conference on Statistical Physics

    1999

  23. [24]

    Jiménez, J., Physics of Fluids 25, 101302 (2013), https://pubs.aip.org/aip/pof/article- pdf/doi/10.1063/1.4824988/16097093/101302 _1_online.pdf

    work page doi:10.1063/1.4824988/16097093/101302 2013

  24. [25]

    Fluid Engng 124, 829 (2002)

    von Kaenel, R., Adams, N., Kleiser, L., and V os, J., J. Fluid Engng 124, 829 (2002)

    2002

  25. [26]

    Kutta, W., Zeitschrift für Mathematik und Physik 46, 435 (1901)

    1901

  26. [27]

    18, edited by F

    Leonard, A., in Turbulent Diffusion in Environmental Pollution, Advances in Geophysics, V ol. 18, edited by F. Frenkiel and R. Munn (Elsevier, 1975) pp. 237–248

    1975

  27. [28]

    K., Physics of Fluids A: Fluid Dynamics 4, 633 (1992), https://pubs.aip.org/aip/pof/article- pdf/4/3/633/12443162/633_1_ online.pdf

    Lilly, D. K., Physics of Fluids A: Fluid Dynamics 4, 633 (1992), https://pubs.aip.org/aip/pof/article- pdf/4/3/633/12443162/633_1_ online.pdf

    1992

  28. [29]

    Liu, S., Meneveau, C., and Katz, J., Journal of Fluid Mechanics 275 (1994), 10.1017/S0022112094002296

    work page doi:10.1017/s0022112094002296 1994

  29. [30]

    and Ducros, F., Flow, Turbulence and Combustion 62, 183 (1999)

    Nicoud, F. and Ducros, F., Flow, Turbulence and Combustion 62, 183 (1999)

    1999

  30. [31]

    B., Cabrit, O., Bose, S., and Lee, J., Physics of Fluids 23, 085106 (2011)

    Nicoud, F., Toda, H. B., Cabrit, O., Bose, S., and Lee, J., Physics of Fluids 23, 085106 (2011)

    2011

  31. [32]

    Oberlack, M., Annual Research Briefs 3 (1997)

    1997

  32. [33]

    Paolucci, S., NASA STI/Recon Technical Report N 83 (1982)

    1982

  33. [34]

    Papalexandris, M., Continuum Mechanics and Thermodynamics 32 (2019), 10.1007/s00161-019-00785- z

    work page doi:10.1007/s00161-019-00785- 2019

  34. [35]

    (Springer, 2006)

    Pierre Sagaut„ Large Eddy Simulation for Incompressible Flows, 3rd ed. (Springer, 2006)

    2006

  35. [36]

    B., Turbulent Flows (Cambridge University Press, 2000)

    Pope, S. B., Turbulent Flows (Cambridge University Press, 2000)

    2000

  36. [37]

    J., Moin, P., and Verstappen, R., Physics of Fluids 27, 085107 (2015)

    Rozema, W., Bae, H. J., Moin, P., and Verstappen, R., Physics of Fluids 27, 085107 (2015)

    2015

  37. [38]

    Runge, C. D. T., Mathematische Annalen 46, 167 (1895)

  38. [39]

    H., Remmerswaal, R

    Silvis, M. H., Remmerswaal, R. A., and Verstappen, R., Physics of Fluids 29, 015105 (2017), https://pubs.aip.org/aip/pof/article-pdf/doi/10.1063/1.4974093/14927086/015105 _1_online.pdf

    work page doi:10.1063/1.4974093/14927086/015105 2017

  39. [40]

    Smagorinsky, J., Monthly Weather Review 91, 99 (1963)

    1963

  40. [41]

    G., Journal of Fluid Mechanics 156, 55 (1985)

    Speziale, C. G., Journal of Fluid Mechanics 156, 55 (1985)

    1985

  41. [42]

    and Adams, N., Phys

    Stolz, S. and Adams, N., Phys. Fluids 11, 1699 (1999)

    1999

  42. [43]

    B., Silvis, M

    Streher, L. B., Silvis, M. H., Cifani, P., and Verstappen, R. W. C. P., AIP Advances11, 015002 (2021)

    2021

  43. [44]

    Sutherland, W., The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science36, 507 (1893)

  44. [45]

    Toutant, A., Journal of Computational Physics 374, 822 (2018)

    2018

  45. [46]

    X., Folch, D., Gorobets, A., and Oliva, A., Physics of Fluids 27, 065103 (2015), https://pubs.aip.org/aip/pof/article-pdf/doi/10.1063/1.4921817/14742226/065103 _1_online.pdf

    Trias, F. X., Folch, D., Gorobets, A., and Oliva, A., Physics of Fluids 27, 065103 (2015), https://pubs.aip.org/aip/pof/article-pdf/doi/10.1063/1.4921817/14742226/065103 _1_online.pdf . 23

    work page doi:10.1063/1.4921817/14742226/065103 2015

  46. [47]

    Vreman, B., Geurts, B., and Kuerten, H., Physics of Fluids 6, 4057 (1994), https://pubs.aip.org/aip/pof/article-pdf/6/12/4057/19309476/4057_1 _online.pdf

    1994

  47. [48]

    Vreman, B., Geurts, B., and Kuerten, H., Journal of Fluid Mechanics 278, 351 (1994)

    1994

  48. [49]

    Vreman, B., Geurts, B., and Kuerten, H., Theoretical and Computational Fluid Dynamics 8, 309 (1996)

    1996

  49. [50]

    Vreman, B., Geurts, B., and Kuerten, H., Journal of Fluid Mechanics 339, 357–390 (1997)

    1997

  50. [51]

    L., and Koseff, J

    Zang, Y ., Street, R. L., and Koseff, J. R., Physics of Fluids A: Fluid Dynamics 5, 3186 (1993), https://pubs.aip.org/aip/pof/article-pdf/5/12/3186/12666067/3186_1 _online.pdf

    1993

  51. [52]

    24 Appendix A

    Zhou, Z., He, G., Wang, S., and Jin, G., Computers & Fluids 195, 104319 (2019). 24 Appendix A. Error rate tables Appendix A.1. First-order, second-order, and total error rate Model Category Scheme Mesh A Mesh B Mesh C Average M1 Mixed C4-C2 18.8 21.6 37.2 25.9 M2 Mixed C4-C2 25.8 22.8 33.6 27.4 M3 Mixed C2-C2 25.4 29.2 42.2 32.2 M4 Mixed C2-C2 31.7 36.7 2...

    2019