Pith. sign in

REVIEW 3 major objections 3 minor 37 references

Manipulating the direction of turbulent energy flux via tensor geometry in a two-dimensional flow

T0 review · 3 major / 3 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 The paper argues that steering the angle between small-scale stress and large-scale rate-of-strain tensors in a two-dimensional flow controls the net spectral energy flux, and reports 2D Navier-Stokes turbulence with net forward energy…

desk verdict A clean experimental and numerical demonstration that the sign of the measured subgrid energy flux in 2D flow follows the mechanical angle between small-scale forcing and background shear, but the stronger claim of having produced net forward energy flux is not established because the reported flux drops the Leonard and cross terms. read the letter →

arxiv 2411.15581 v1 pith:DZBVKTHP submitted 2024-11-23 physics.flu-dyn

Xinyu Si , Filippo De Lillo , Guido Boffetta , Lei Fang This is my paper · ORCID
classification physics.flu-dyn
keywords two-dimensionalturbulenceenergycascadedirectionspectralfluxtensorgeometrystress-strainalignmentforwardquasi-2Dflowmanipulationfilter-spacetechnique
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper sets out to show that the direction of turbulent energy transfer across scales is not fixed by the dimensionality of the flow. The central relation is $\Pi^{(L)}=-2\sigma\varepsilon\cos(2\theta^{(L)})$: the spectral energy flux at a cutoff scale $L$ is controlled by the angle $\theta^{(L)}$ between the extensional eigenvectors of the small-scale stress tensor and the large-scale rate-of-strain tensor, with $\theta^{(L)}<\pi/4$ giving inverse flux and $\theta^{(L)}>\pi/4$ giving forward flux. The paper demonstrates control of $\theta^{(L)}$ by superimposing a directionally biased small-scale forcing on a background shear whose strain orientation is known. With the forcing set roughly perpendicular to the background strain, the authors report the first experimental and numerical realization of two-dimensional Navier-Stokes turbulence with a net forward energy flux. If this stands, energy-flux direction becomes a design variable rather than a fixed property of the flow.

What carries the argument

The load-bearing object is the tensor-geometry identity $\Pi^{(L)}=-2\sigma\varepsilon\cos(2\theta^{(L)})$ for two-dimensional flows, where $\theta^{(L)}$ is the angle between the extensional eigenvector of the deviatoric subgrid stress and that of the filtered rate-of-strain tensor. The argument runs through four steps: (i) the direction of a small-scale velocity perturbation determines the extensional eigenvector of the subgrid Reynolds stress; (ii) a background shear flow supplies a known, ordered large-scale strain eigenframe; (iii) superimposing the two fixes $\theta^{(L)}$ through the mechanical angle between the forcing direction and the shear strain axis; and (iv) the cosine identity turns that angle into the sign and magnitude of the spectral energy flux. The filter-space technique and the Leonard/cross/subgrid decomposition of the stress are the measurement tools used to evaluate $\Pi^{(L)}$.

What would settle it

Compute the full filtered flux $\Pi^{(L)}=\Pi_L^{(L)}+\Pi_C^{(L)}+\Pi_\tau^{(L)}$ on the same experimental or numerical velocity fields used in the paper, and check whether the total flux changes sign with the forcing angle as the subgrid component does; a positive total at $\theta\approx\pi/2$ would confirm the cascade reversal, while a sign mismatch would show that the subgrid-only analysis produced the claimed direction.

Watch

Extended reading notes

Core claim

In the authors' formulation, the cascade in a 2D flow is a mechanical process: the subgrid stress tensor acts as a force and the filtered rate-of-strain tensor as a generalized displacement, and their product gives the spectral energy flux. Rewriting the flux in the eigenframes of the two tensors yields $\Pi^{(L)}=-2\sigma\varepsilon\cos(2\theta^{(L)})$, where $\sigma$ and $\varepsilon$ are the largest eigenvalues of the deviatoric stress and rate-of-strain tensors. Because both eigenvalues are positive, the sign of $\Pi^{(L)}$ is entirely determined by the angle $\theta^{(L)}$ between the extensional eigenvectors. The paper then engineers $\theta^{(L)}$: a steady hydrodynamic shear fixes the orientation of the large-scale rate-of-strain eigenframe, while a directionally biased small-scale perturbation creates small-scale stress whose extensional eigenvector points along the forcing direction. The measured flux varies sinusoidally with the mechanical forcing angle in the form $\cos(2\theta)$, and at $\theta\approx\pi/2$ both experiment and simulation show a positive net energy flux, a regime the authors describe as 2D weak turbulence with a net forward energy flux, which they state has not been produced before.

Load-bearing premise

The load-bearing premise is that the subgrid stress component alone represents the true spectral energy flux between scales; if the discarded Leonard and cross terms carry significant net flux in these shear-plus-forcing flows, the reported sign of the net forward energy flux could come from the chosen flux definition rather than from the actual cascade direction.

Editorial extensions

If this is right

  • In a two-dimensional flow, the mechanical angle of the small-scale forcing acts as a dial for the cascade: $\theta=0$ gives inverse flux, $\theta=\pi/2$ gives forward flux, and $\theta=\pi/4$ gives nearly zero flux; the paper verifies this sinusoidal dependence in both experiments and simulations.
  • Because the flux sign is set by local tensor alignment rather than by the global dimensionality of the system, the same framework suggests that a three-dimensional flow with an engineered strain background could be driven toward an inverse cascade; the authors state this extension explicitly.
  • A new regime of 2D Navier-Stokes turbulence with net forward energy flux is now available as a testbed for cascade theory, complementing the traditional inverse-cascade regime of two-dimensional turbulence.
  • Applications follow wherever cascade direction matters: enhanced mixing in microfluidic devices, controlled coupling of biologically generated turbulence with background shear, weakening of coastal transport barriers, and a second-order route by which wind stress can redistribute energy inside oceanic flows.

Reading between the lines

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

  • A natural next step, not reported in the paper, is to compute the full filtered flux including the Leonard and cross terms in the same flows; because the paper isolates the subgrid component, that calculation would show whether the cosine law holds for the total flux as well.
  • If the cosine law is robust, the measured angle $\theta^{(L)}$ could become a real-time observable for feedback control: a sensor reading strain alignment could adjust the forcing direction to keep the flux at a desired sign.
  • The paper's estimate for weakening transport barriers, built on a periodic cellular flow with a localized forcing patch, can be read as a quantitative prediction that roughly one part in two thousand of the sustaining power suffices to start dismantling a barrier; testing it would require measuring barrier strength after forcing in that flow.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 3 minor

Summary. The paper proposes that the direction of the spectral energy flux in a two-dimensional flow can be controlled by manipulating the alignment between the extensional eigenvector of the small-scale stress tensor and that of the large-scale rate-of-strain tensor. The theoretical basis is Eq. (1), Π(L) = −2σε cos(2θ(L)), which is an exact tensor-geometry identity for symmetric deviatoric tensors. The authors test the idea in an electromagnetically driven thin-layer experiment and in pseudospectral DNS, both consisting of a steady shear background perturbed by a 5×5 array of rods or force monopoles whose mechanical direction is varied. They report that the measured alignment angle θ(L) follows the mechanical angle, that the subgrid flux changes sign accordingly, that three control cases show no net flux, and that the third-order structure function changes sign consistently. They conclude that they have produced, for the first time, two-dimensional Navier-Stokes turbulence with a net forward energy flux, and they discuss applications to microfluidics, biological mixing, coastal transport barriers, and ocean-atmosphere coupling.

Significance. If fully supported, the central claim is significant: it would demonstrate a geometric, parameter-free mechanism for reversing the direction of the turbulent cascade in a 2D flow, contrary to the usual dimensionality argument. The positive features are substantial: Eq. (1) is a mathematical identity rather than a fitted model; the experimental design includes three meaningful control cases; the simulations vary the mechanical angle in a controlled way; and the observed dependence of the measured alignment angle on the mechanical angle is a falsifiable prediction that is borne out. The main risk is that the reported 'net energy flux' is only the subgrid component Π_τ, not the full filtered flux, and the justification for ignoring the Leonard and cross components is not established for the manipulated flows. Since the headline claim is the generation of a new type of 2D turbulence with net forward energy flux, this flux-definition issue is load-bearing. The paper is therefore promising but requires a substantial check before the central claim can be accepted.

major comments (3)
  1. [Methods 5.1.2, Eqs. (S5)-(S9), Fig. S1] The quantity reported as Π(L) in the main text is only the subgrid component Π_τ; the Leonard and cross components are discarded with the justification that they are negligible in homogeneous turbulence and that boundary padding contaminates them. The verification shown in Fig. S1 is performed on a checkerboard magnet flow, not on the shear-plus-rod or shear-plus-monopole flows used for the manipulation claim. In the manipulated flows the large-scale shear is coherent and the small-scale forcing is localized and time-periodic, so the spatial averaging that cancels Π_L and Π_C in homogeneous turbulence is not guaranteed to apply. Because the headline claim is a net forward energy flux, the full flux Π = Π_L + Π_C + Π_τ should be reported, at least for the DNS where boundary padding can be avoided or controlled; if Π_L + Π_C is comparable in magnitude or opposite in sign to Π_τ, the sign of the true total flux is not established.
  2. [Section 3, Fig. 3d,k] The sign of the third-order structure function S3 is offered as independent evidence for the flux direction, but the manuscript itself notes that inertial-range scaling laws do not apply at Re≈210 (experiment) and Re≈184 (DNS). The expected relation S3 = −Cεr with the sign of C depending on cascade direction is an inertial-range result; at these Reynolds numbers, S3 is only a qualitative indicator and cannot quantitatively settle whether the total spectral flux is forward. A direct computation of the full flux in the DNS, as requested above, would resolve this ambiguity.
  3. [Section 3, paragraph beginning 'Significantly...'] The claim that the authors have 'both experimentally and numerically produced 2D weak turbulence with net forward energy flux—a type of NS turbulence that has never been generated before' is stronger than what is demonstrated. The measured quantity is Π_τ, not the total spectral energy flux, and the flows are weakly turbulent at moderate Reynolds number with periodic forcing. Please either verify the full-flux sign or qualify this statement explicitly as a claim about the subgrid-scale component of the flux.
minor comments (3)
  1. [Section 3, first paragraph] The text says 'we observed a salient shift of θ(L) toward 0 (Fig. 1a)', but Fig. 1 is the schematic figure; the experimental PDF data appear in Fig. 3a (and the numerical results in Fig. 3h). Please correct the cross-reference.
  2. [Methods 5.2.3, Eq. (S12)] The statement that 'the filtering process will not affect the eigenvector direction' holds for the idealized case in which the small-scale velocity direction is uniform over the filter support. For the general spatially varying fields, convolution can mix tensor components and rotate the eigenframe; the statement should be qualified, even if the experimental and numerical visualizations in Figs. S2 and S3 support the approximation in the present configuration.
  3. [Methods 5.5, Eq. (S15) and following estimate] The LCS manipulation estimate depends on several freely chosen parameters (α=0.1, ‖u′‖=0.08, and the π/5×π/5 region), and the reported 0.05% energy figure follows from these choices. This is acceptable as an order-of-magnitude illustration, but it should be explicitly labeled as such, and a brief sensitivity statement would help readers assess the robustness of the application claim.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: Eq. 1 is an exact identity and the empirical control of tensor alignment is measured independently.

full rationale

The paper's core relation, Π(L) = −2σε cos(2θ(L)) (Eq. 1; derived in Methods, Eq. S10), is an exact linear-algebra identity for the inner product of two symmetric deviatoric tensors expressed in their eigenframes. It is not a fit, not a statistical inference, and not a parameterized ansatz. The empirical content lies in the separately, independently measured quantities: the mechanical angle of the rod/monopole array is set by the experimenter, the resulting eigenvector alignment θ(L) is measured from the velocity fields, and the spectral flux is computed from the same fields by filtering. The paper shows that changing the mechanical angle moves the measured θ(L) through π/4 and, correspondingly, the measured subgrid flux changes sign. No step in this chain amounts to fitting a scalar to a target and then reading back a prediction; the sign relation is a mathematical identity, while the controllability of θ(L) is an observational result. The only self-citations (refs. 10, 13, and related prior work) support either the efficiency measure η = cos(2θ) or the setup of electromagnetically driven thin-layer flows. The efficiency measure is re-derived in the Supplementary Methods, and the experimental apparatus is described in the present paper, so the cited results are not load-bearing in the sense of importing an unverified premise. The Methods also disclose a real limitation: the analysis uses the subgrid component Π_τ of the filtered flux instead of the full Π(L), justified by homogeneity arguments and a checkerboard-flow check. This caveat bears on whether the measured positive Π_τ represents a true net positive total cascade flux in the manipulated shear-plus-forcing flows, but that is an empirical/measurement-validity concern, not a circularity: the sign of Π_τ is not imposed by fitting or by definition. No circular step was found.

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

The central theoretical relation (Eq. 1) is an exact identity and has no fitted parameters. The parameters listed are hand-chosen operating points for the simulations and for the speculative LCS application estimate, not quantities fit to make the theory match data. The axioms are standard filter-space turbulence results plus domain assumptions about quasi-2D behavior and the dominance of the subgrid flux component.

free parameters (6)
  • Simulation monopole forcing amplitude A = 0.38 (arbitrary units)
    Chosen in Methods 5.4 to provide close to maximum measured effect Π_τ/(2νΩ); kept fixed across all simulated angles. This is a hand-selected operating point, not fitted to data.
  • Simulation monopole forcing frequency ω = 2π/5
    Same selection rationale as amplitude: chosen for near-maximum effect and kept fixed.
  • LCS estimation: large-scale frictional damping coefficient α = 0.1
    Assumed in Methods 5.5 for the Taylor-Green flow manipulation estimate.
  • LCS estimation: small-scale disturbance magnitude ‖u′‖ = 0.08
    Assumed in Methods 5.5 to compute the power and energy flux needed to weaken Lagrangian coherent structures.
  • LCS estimation: manipulated region area = π/5 × π/5 square
    Chosen region around a hyperbolic point in Methods 5.5; result scales with this choice.
  • Filter cutoff length L/W = 0.8
    Selected for the main analysis; the manuscript does not show the sensitivity of the conclusion to this choice.
assumptions (6)
  • standard math The filtered Navier-Stokes equations and the expression Π(L) = -τ_ij S_ij for the spectral energy flux (Methods 5.1.1).
    Standard filter-space derivation, referenced to Germano (29) and Leonard (31).
  • standard math The tensor geometry identity Π(L) = -2σε cos(2θ(L)) for symmetric deviatoric 2D tensors (Eq. 1, Methods 5.2.1).
    Exact eigen-decomposition identity; follows from linear algebra.
  • domain assumption The flow in the thin-layer experiment is adequately described by 2D incompressible Navier-Stokes dynamics (Methods 5.3.1).
    Standard for quasi-2D electromagnetic thin-layer flows; paper asserts 'the 2D was well-kept throughout' but provides no quantitative measure of three-dimensionality.
  • domain assumption The subgrid (small-small) component Π_τ dominates the total spectral energy flux after spatial averaging, so Leonard and cross components can be discarded (Methods 5.1.2).
    Cited to Liao & Ouellette (32) and verified by the authors on a checkerboard-forced 2D flow (Fig. S1b), but not directly verified for the manipulated shear-plus-rod flows.
  • domain assumption The extensional eigenvector of the subgrid stress aligns with the mechanical forcing direction of the rods/monopoles (Methods 5.2.3).
    Derived for a single plane-wave disturbance; assumed to hold for the 5×5 array after filtering. Empirically supported by Fig. 2c and the correlated PDFs in Fig. 3a,h.
  • domain assumption The low-Reynolds-number flows (Re ≈ 184-210) can be considered weakly turbulent such that the cascade concepts and structure-function sign arguments apply qualitatively (Sections 2-3).
    The paper acknowledges 'at our relatively low Reynolds numbers, the scaling results do not apply' and uses 'weak turbulence' terminology.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Manipulating the direction of turbulent energy flux via tensor geometry in a two-dimensional flow." pith.science (2026). https://pith.science/paper/DZBVKTHP

@misc{pith2026241115581,
  author       = {Pith},
  title        = {Pith review of: Manipulating the direction of turbulent energy flux via tensor geometry in a two-dimensional flow},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DZBVKTHP}},
  note         = {Machine review of arXiv:2411.15581}
}
read the original abstract

In turbulent flows, energy flux refers to the transfer of kinetic energy across different scales of motion, a concept that is a cornerstone of turbulence theory. The direction of net energy flux is prescribed by the dimensionality of the fluid system.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

37 extracted references · 37 canonical work pages

  1. [1]

    A. N. Kolmogorov, The local structure of turbulence in inc ompressible viscous fluid for very large Reynolds. Numbers. In Dokl. Akad. Nauk SSSR 30, 301 (1941)

    work page 1941

  2. [2]

    R. H. Kraichnan, Inertial ranges in two-dimensional turb ulence. The Physics of Fluids 10 (7), 1417–1423 (1967)

    work page 1967

  3. [3]

    G. K. Batchelor, Computation of the energy spectrum in hom ogeneous two-dimensional tur- bulence. The Physics of Fluids 12 (12), II–233 (1969)

    work page 1969

  4. [4]

    A. N. Kolmogorov, A refinement of previous hypotheses conc erning the local structure of tur- bulence in a viscous incompressible fluid at high Reynolds number. Journal of Fluid Mechanics 13 (1), 82–85 (1962)

    work page 1962

  5. [5]

    S. B. Pope, Turbulent Flows (Cambridge University Press) (2000), doi:10.1017/ CBO9781316179475

    work page 2000

  6. [6]

    L. F. Richardson, The supply of energy from and to atmosphe ric eddies. Proceedings of the Royal Society of London. Series A, Containing Papers of a Mathematical and Physical Character 97 (686), 354–373 (1920)

    work page 1920

  7. [7]

    H. Xia, D. Byrne, G. Falkovich, M. Shats, Upscale energy tr ansfer in thick turbulent fluid layers. Nature Physics 7 (4), 321–324 (2011)

    work page 2011

  8. [8]

    L. Fang, N. T. Ouellette, Multiple stages of decay in two-d imensional turbulence. Physics of Fluids 29 (11), 111105 (2017)

    work page 2017

Show all 37 references
  1. [9]

    L. Fang, N. T. Ouellette, Spectral condensation in labora tory two-dimensional turbulence. Physical Review Fluids 6 (10), 104605 (2021)

    2021

  2. [10]

    L. Fang, N. T. Ouellette, Advection and the efficiency of sp ectral energy transfer in two- dimensional turbulence. Physical review letters 117 (10), 104501 (2016)

    2016

  3. [11]

    Y . Liao, N. T. Ouellette, Geometry of scale-to-scale ene rgy and enstrophy transport in two- dimensional flow. Physics of Fluids 26 (4) (2014). 17

    2014

  4. [12]

    X. Si, L. Fang, Interaction between swarming active matter and flow: The impact on Lagrangian coherent structures. Physical Review Fluids 9 (3), 033101 (2024)

    2024

  5. [13]

    X. Si, L. Fang, Biologically generated turbulent energy flux in shear flow depends on tensor geometry. PNAS nexus 3 (2), pgae056 (2024)

    2024

  6. [14]

    N. T. Ouellette, H. Xu, E. Bodenschatz, A quantitative st udy of three-dimensional Lagrangian particle tracking algorithms. Experiments in Fluids 40, 301–313 (2006)

    2006

  7. [15]

    J. P . Boyd, Chebyshev and Fourier spectral methods (Courier Corporation) (2001)

    2001

  8. [16]

    V . J. Valad ˜ao, G. Boffetta, M. Crialesi-Esposito, F. D. Lillo, S. Musacc hio, Spectrum correction on Ekman-Navier-Stokes equation in tw o-dimensions (2024), https://arxiv.org/abs/2408.15735

    2024 arXiv

  9. [17]

    A. N. Kolmogorov, Dissipation of energy in the locally is otropic turbulence. Proceedings of the Royal Society of London. Series A: Mathematical and Physical Sciences 434 (1890), 15–17 (1991)

    1991

  10. [18]

    Alexakis, L

    A. Alexakis, L. Biferale, Cascades and transitions in tu rbulent flows. Physics Reports 767, 1–101 (2018)

    2018

  11. [19]

    A. D. Stroock, et al., Chaotic mixer for microchannels. Science 295 (5555), 647–651 (2002)

    2002

  12. [20]

    H. A. Stone, A. D. Stroock, A. Ajdari, Engineering flows in small devices: microfluidics toward a lab-on-a-chip. Annu. Rev. Fluid Mech. 36 (1), 381–411 (2004)

    2004

  13. [21]

    I. A. Houghton, J. R. Koseff, S. G. Monismith, J. O. Dabiri, Vertically migrating swimmers generate aggregation-scale eddies in a stratified column. Nature 556 (7702), 497–500 (2018)

    2018

  14. [22]

    Kunze, J

    E. Kunze, J. F. Dower, I. Beveridge, R. Dewey, K. P . Bartle tt, Observations of biologically generated turbulence in a coastal inlet. Science 313 (5794), 1768–1770 (2006)

    2006

  15. [23]

    Katija, Biogenic inputs to ocean mixing

    K. Katija, Biogenic inputs to ocean mixing. Journal of Experimental Biology 215 (6), 1040– 1049 (2012). 18

    2012

  16. [24]

    Olascoaga, et al

    M. Olascoaga, et al. , Persistent transport barrier on the West Florida Shelf. Geophysical research letters 33 (22) (2006)

    2006

  17. [25]

    L. Fang, N. T. Ouellette, Influence of lateral boundaries on transport in quasi-two-dimensional flow. Chaos: An Interdisciplinary Journal of Nonlinear Science 28 (2), 023113 (2018)

    2018

  18. [26]

    W. K. Dewar, G. R. Flierl, Some effects of the wind on rings. Journal of physical oceanography 17 (10), 1653–1667 (1987)

    1987

  19. [27]

    Renault, M

    L. Renault, M. J. Molemaker, J. Gula, S. Masson, J. C. McWi lliams, Control and stabilization of the Gulf Stream by oceanic current interaction with the at mosphere. Journal of Physical Oceanography 46 (11), 3439–3453 (2016)

    2016

  20. [28]

    S. Rai, M. Hecht, M. Maltrud, H. Aluie, Scale of oceanic ed dy killing by wind from global satellite observations. Science Advances 7 (28), eabf4920 (2021)

    2021

  21. [29]

    Germano, Turbulence: the filtering approach

    M. Germano, Turbulence: the filtering approach. Journal of Fluid Mechanics 238, 325–336 (1992)

    1992

  22. [30]

    Rivera, W

    M. Rivera, W. Daniel, S. Chen, R. Ecke, Energy and enstrop hy transfer in decaying two- dimensional turbulence. Physical review letters 90 (10), 104502 (2003)

    2003

  23. [31]

    Leonard, Energy cascade in large-eddy simulations of turbulent fluid flows, in Advances in geophysics (Elsevier), vol

    A. Leonard, Energy cascade in large-eddy simulations of turbulent fluid flows, in Advances in geophysics (Elsevier), vol. 18, pp. 237–248 (1975)

    1975

  24. [32]

    Y . Liao, N. T. Ouellette, Spatial structure of spectral transport in two-dimensional flow. Journal of Fluid Mechanics 725, 281–298 (2013)

    2013

  25. [33]

    J. G. Ballouz, N. T. Ouellette, Tensor geometry in the tur bulent cascade. Journal of Fluid Mechanics 835, 1048–1064 (2018)

    2018

  26. [34]

    Z. Xiao, M. Wan, S. Chen, G. Eyink, Physical mechanism of t he inverse energy cascade of two-dimensional turbulence: a numerical investigation. Journal of Fluid Mechanics 619, 1–44 (2009)

    2009

  27. [35]

    N. T. Ouellette, P . O’Malley, J. P . Gollub, Transport of fi nite-sized particles in chaotic flow. Physical review letters 101 (17), 174504 (2008). 19

    2008

  28. [36]

    cheerios effect

    D. Vella, L. Mahadevan, The “cheerios effect”. American journal of physics 73 (9), 817–825 (2005)

    2005

  29. [37]

    cheerios effect

    K. Schneider, M. Farge, Decaying two-dimensional turbu lence in a circular container. Physical review letters 95 (24), 244502 (2005). Funding: L.F. thanks the U.S. National Science Foundation under Gran t No. CMMI-2143807 and CBET-2429374. Author contributions: L.F. conceived ...

    2005

Pith tools

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