First-principles foundation of the log-universality of turbulent velocity structure functions: derivation of the Reynolds-number factor beta(Re), the effective number of degrees of freedom N_eff = 1/beta(Re), and the validity of the multifractal thermodynamic framework of Geneste et al. (2019)

Velocity structure functions measured in experiments and direct numerical simulations (DNS) collapse onto universal curves when rescaled by a logarithmic function of the Reynolds number, through a single fitted factor beta(Re) whose inverse plays the role of an effective number of degrees of freedom, N_eff = 1/beta(Re) ~ (4/3) log(R_lambda/5). This log-universality, established empirically by Geneste, Faller, Nguyen, Shukla, Laval, Daviaud, Saw and Dubrulle (Entropy 21, 326, 2019) using von Karman-flow stereoscopic PIV and DNS up to order p = 9, and interpreted by them through a multifractal-thermodynamic analogy with free energy, pressure, and a phase-transition-like signature, currently rests on fitting: beta(Re) has no first-principles derivation, its low-Reynolds-number behavior is disputed between experiments, the role of the unknown scale L(Re) entering the formulae is undetermined, the identification of N_eff with the dimension of a turbulence attractor is an untested conjecture, and the phase-transition-like signatures and the low-Reynolds-number breakdown of the collapse are unconfirmed and unexplained. The problem is to give this empirical framework a theoretical foundation and to establish its domain of validity. The problem is open.

Background

In fully developed turbulence, the statistics of velocity differences across a distance $\ell$ are characterized by the structure functions $S_p(\ell) = \langle|\delta_\ell u|^p\rangle$. Kolmogorov's 1941 theory (K41) predicts $S_p(\ell) \sim \ell^{p/3}$ in the inertial range (the range of scales between the forcing scale and the dissipation scale), but measured exponents $\zeta(p)$ deviate at large $p$ because the energy dissipation is intermittent, i.e. spatially fluctuating. Kolmogorov's 1962 refinement (K62) models these fluctuations as log-normal, and the multifractal formalism of Parisi and Frisch generalizes this: the velocity field is locally self-similar with a spatially varying Holder exponent $h$ (the local scaling exponent of velocity increments, $|\delta_\ell u| \sim \ell^h$), the probability of observing $h$ at scale $\ell$ decays like a power of the scale ratio, $(\ell/L)^{C(h)}$, where $C(h)$ is the multifractal spectrum (the large-deviation rate function of $h$), and the local dissipation cutoff fluctuates as $\eta_h \sim L\,\mathrm{Re}^{-1/(1+h)}$, which reduces to the Kolmogorov scale at $h = 1/3$.

Frisch and Vergassola (EPL 14, 439, 1991) showed that these fluctuating dissipation scales produce an intermediate dissipation range and imply a form of universality in which structure functions, expressed in Kolmogorov units, depend on $\log(\ell/\eta)$ only through the combination $\beta \cdot \log(\ell/\eta)$ with $\beta$ scaling like $1/\log(L/\eta)$, i.e. like $1/\log(\mathrm{Re})$ since $L/\eta \sim \mathrm{Re}^{3/4}$. Castaing, Gagne and Marchand (Physica D 68, 387, 1993) verified this log-similarity empirically in turbulent jets and found that the fitted parameter $1/\beta$ grows like $\log(R_\lambda)$ at large Taylor-scale Reynolds number $R_\lambda$ but saturates at low $R_\lambda$ in their liquid-helium jet data; here $R_\lambda$ is the Reynolds number based on the Taylor microscale, a standard measure of the scale separation in the flow.

Geneste et al. (Entropy 21, 326, 2019) extended this program using stereoscopic particle image velocimetry (SPIV, a three-component velocity-field measurement technique) in a turbulent von Karman flow (a flow between counter-rotating disks) and DNS of the Navier-Stokes equations. They showed that structure functions up to order $p = 9$ collapse on universal curves under a generalized rescaling

$$\beta(\mathrm{Re})\,\frac{\log(\tilde S_p/S_{0p})}{\log(L_0/\eta)} = F\!\left(p,\; \beta(\mathrm{Re})\,\frac{\log(\ell/\eta)}{\log(L_0/\eta)}\right),$$

where $\tilde S_p$ is normalized by Kolmogorov velocity and length units (the dissipation scale $\eta$ and the velocity $u_K = (\varepsilon\nu)^{1/4}$), $L_0$ is the integral (forcing) scale, $S_{0p}$ and $\beta(\mathrm{Re})$ are fitted, and the fitted law is $1/\beta(\mathrm{Re}) \approx \beta_0 \log(R_\lambda/5)$ with $\beta_0 \approx 4/3$, compatible with the earlier jet data at large $R_\lambda$ (the paper's inline statement of this law is printed with a typographical inversion, $1/\beta \sim \beta_0/\log(R_\lambda)$, contradicted by its own Figure 6 caption and Conclusion). Interpreting $F = \log \tilde S_{3p}$ as a multifractal free energy with temperature $1/p$ and volume $V = \log(\ell/\eta)$, they showed that the collapse is equivalent to extensivity, $F(T,V,N) = N F(T,V/N,1)$, provided the number of degrees of freedom is identified as $N = (1/\beta(\mathrm{Re}))\,\log(L_0/\eta)$; their Conclusion instead summarizes the effective degree-count as $N_{\mathrm{eff}} = 1/\beta(\mathrm{Re})$, which grows only like $\log(R_\lambda)$ — far smaller than the classical Landau estimate $N \sim \mathrm{Re}^{9/4}$, the number of grid nodes needed to resolve a turbulent flow down to the Kolmogorov scale (the paper's bookkeeping varies between these identifications). The conjugate multifractal pressure $P = \partial F/\partial V$ develops an inflection and a sign change of $\partial P/\partial V$ at large enough $R_\lambda$ and order $p$, reminiscent of a vapor-liquid phase transition between rough (turbulent) and relaminarized flow regions, a phenomenon previously observed only for Lagrangian (trajectory-based) statistics by the International Collaboration for Turbulence Research (Arneodo et al., PRL 100, 254504, 2008), and expected to disappear in multifractal models without a fluctuating dissipation scale (Biferale et al., PRL 93, 064502, 2004).

The authors explicitly acknowledge that this framework rests on unresolved foundations: the scale $L$ entering the formulae is not known a priori and may itself depend on Re; the empirical law for $\beta(\mathrm{Re})$ has no theoretical derivation and its low-$R_\lambda$ behavior does not saturate as the jet data did; the conjecture that $N_{\mathrm{eff}} = 1/\beta(\mathrm{Re})$ corresponds to the dimension of a suitable attractor of turbulence is unproven; and the collapse fails for the DNS at the lowest Reynolds number ($R_\lambda = 25$) in the far dissipative range. Note on conventions: because $\beta(\mathrm{Re})$ multiplies a logarithm whose normalization is fitted, $1/\beta$ is determined only up to an additive shift depending on the reference amplitude used in the collapse procedure; the constant $\beta_0 \approx 4/3$ is meaningful only together with the shift $R_\lambda/5$ in the argument of the logarithm.

Problem Statement

The single scientific objective is to give the empirically observed log-universality of Eulerian velocity structure functions a first-principles foundation and to establish its physical interpretation and domain of validity. Concretely, within the framework of Geneste et al. (2019), where structure functions up to order 9 from experiments and DNS collapse under the rescaling $\beta(\mathrm{Re})\,\log(\tilde S_p/S_{0p})/\log(L_0/\eta) = F(p,\beta(\mathrm{Re})\,\log(\ell/\eta)/\log(L_0/\eta))$ with $1/\beta(\mathrm{Re}) \approx \beta_0 \log(R_\lambda/5)$, $\beta_0 \approx 4/3$, and where the multifractal free energy $F = \log \tilde S_{3p}$ (temperature $1/p$, volume $\log(\ell/\eta)$) satisfies the extensivity relation $F(T,V,N) = N F(T,V/N,1)$ with a degree-count built from $1/\beta(\mathrm{Re})$ — the effective degree-count posed against the attractor dimension being $N_{\mathrm{eff}} = 1/\beta(\mathrm{Re})$ — the problem asks: (a) determine whether $\beta(\mathrm{Re})$ admits a first-principles derivation (from the Navier-Stokes equations and/or the multifractal formalism with fluctuating dissipation scales) and, if so, derive its functional form — including the value and universality of the constant $\beta_0$ and the additive reference-amplitude shift under which $1/\beta$ is defined only up to a constant — going beyond the Frisch-Vergassola heuristic $\beta \sim 1/\log(\mathrm{Re})$; (b) settle the disputed low-Reynolds-number behavior of $1/\beta$ (saturation in the liquid-helium jet data of Castaing, Gagne and Marchand versus no saturation in the von Karman and DNS data); (c) determine the scale $L(\mathrm{Re})$ entering the multifractal and K62 formulae, which differs from the integral scale and is currently a free fitting parameter entangled with $\beta$ and $S_{0p}$; (d) establish or refute the physical interpretation of $N_{\mathrm{eff}} = 1/\beta(\mathrm{Re})$ as the dimension of a suitable attractor of turbulence (the invariant set in phase space on which the long-time dynamics settle); (e) confirm or refute the phase-transition-like behavior of the multifractal pressure $P = \partial F/\partial V$ (inflection and coexistence of rough and relaminarized regions), in particular its universality across flows and its distinction from finite-Reynolds-number and measurement-resolution artifacts; and (f) explain the observed failure of the collapse at the lowest Reynolds numbers (DNS at $R_\lambda = 25$, and the analogous lowest-Re PIV datasets) in the far dissipative range (scales of order the Kolmogorov cutoff $\eta$), i.e. whether it reflects insufficient scale separation, resolution effects, or a genuine breakdown of the log-universality. Each of these components corresponds to an open foundational issue that Geneste et al. (2019) explicitly leaves unresolved. All six components (a)–(f) are in scope as constituent parts of this single objective: a complete solution addresses the objective as a whole, and a resolution of any proper subset of the components leaves the remaining ones open.

References

1. Damien Geneste, Hugues Faller, Florian Nguyen, Vishwanath Shukla, Jean-Philippe Laval, Francois Daviaud, Ewe-Wei Saw, Berengere Dubrulle, 'About Universality and Thermodynamics of Turbulence', Entropy 21(3), 326 (2019), DOI 10.3390/e21030326, arXiv:1903.00769, https://doi.org/10.3390/e21030326 (source paper)
2. H. Faller, D. Geneste, T. Chaabo, A. Cheminet, V. Valori, Y. Ostovan, L. Cappanera, Ch. Cuvier, F. Daviaud, J.-M. Foucaut, J.-L. Guermond, J.-Ph. Laval, C. Nore, V. Padilla, C. Wiertel, B. Dubrulle, 'On the nature of intermittency in a turbulent von Kármán flow', Journal of Fluid Mechanics 914, A2 (2021), DOI 10.1017/jfm.2020.908, https://doi.org/10.1017/jfm.2020.908
3. Berengere Dubrulle, 'Multi-Fractality, Universality and Singularity in Turbulence', Fractal and Fractional 6(10), 613 (2022), DOI 10.3390/fractalfract6100613, https://doi.org/10.3390/fractalfract6100613
4. Arjun Berera, Daniel Clark, 'Information production in homogeneous isotropic turbulence', Physical Review E 100, 041101 (2019), DOI 10.1103/PhysRevE.100.041101, https://doi.org/10.1103/PhysRevE.100.041101
5. Benjamin Musci, Sebastien Aumaitre, Enzo Francisco, Adam Cheminet, Berengere Dubrulle, 'The Excess Dissipation of Energy in a Turbulent Boundary-Layer and its Departure from Log-Normality' (2026), arXiv:2607.02974, https://arxiv.org/abs/2607.02974
6. U. Frisch, M. Vergassola, 'A Prediction of the Multifractal Model: the Intermediate Dissipation Range', Europhysics Letters 14, 439-444 (1991), DOI 10.1209/0295-5075/14/5/009, https://doi.org/10.1209/0295-5075/14/5/009
7. B. Castaing, Y. Gagne, M. Marchand, 'Log-similarity for turbulent flows?', Physica D: Nonlinear Phenomena 68, 387-400 (1993), DOI 10.1016/0167-2789(93)90132-k, https://doi.org/10.1016/0167-2789(93)90132-k
8. A. Arnèodo, R. Benzi, J. Berg, L. Biferale, E. Bodenschatz, A. Busse, E. Calzavarini, B. Castaing, M. Cencini, L. Chevillard, R. T. Fisher, R. Grauer, H. Homann, D. Lamb, A. S. Lanotte, E. Lévèque, B. Lüthi, J. Mann, N. Mordant, W.-C. Müller, S. Ott, N. T. Ouellette, J.-F. Pinton, S. B. Pope, S. G. Roux, F. Toschi, H. Xu, P. K. Yeung (International Collaboration for Turbulence Research), 'Universal Intermittent Properties of Particle Trajectories in Highly Turbulent Flows', Physical Review Letters 100, 254504 (2008), DOI 10.1103/PhysRevLett.100.254504, https://doi.org/10.1103/PhysRevLett.100.254504
9. L. Biferale, G. Boffetta, A. Celani, B. J. Devenish, A. Lanotte, F. Toschi, 'Multifractal Statistics of Lagrangian Velocity and Acceleration in Turbulence', Physical Review Letters 93, 064502 (2004), DOI 10.1103/PhysRevLett.93.064502, https://doi.org/10.1103/PhysRevLett.93.064502
10. E.-W. Saw, P. Debue, D. Kuzzay, F. Daviaud, B. Dubrulle, 'On the universality of anomalous scaling exponents of structure functions in turbulent flows', Journal of Fluid Mechanics 837, 657-669 (2018), DOI 10.1017/jfm.2017.848, https://doi.org/10.1017/jfm.2017.848
11. Berengere Dubrulle, 'Beyond Kolmogorov cascades', Journal of Fluid Mechanics 867, P1 (2019), DOI 10.1017/jfm.2019.98, https://doi.org/10.1017/jfm.2019.98
12. Dhawal Buaria, Katepalli R. Sreenivasan, 'Saturation and Multifractality of Lagrangian and Eulerian Scaling Exponents in Three-Dimensional Turbulence', Physical Review Letters 131, 204001 (2023), DOI 10.1103/PhysRevLett.131.204001, https://doi.org/10.1103/PhysRevLett.131.204001
13. Kartik P. Iyer, 'Dependence of the asymptotic energy dissipation on third-order velocity scaling', Physical Review Fluids 8, L082601 (2023), DOI 10.1103/PhysRevFluids.8.L082601, arXiv:2211.00554, https://doi.org/10.1103/PhysRevFluids.8.L082601
14. Benjamin Musci, Berengere Dubrulle, Jean LeBris, Damien Geneste, Pierre Braganca, Jean-Marc Foucaut, Christophe Cuvier, Adam Cheminet, 'Experimental measurement of the vorticity–strain alignment around extreme energy transfer events', Journal of Fluid Mechanics 1016, A53 (2025), DOI 10.1017/jfm.2025.10447, arXiv:2503.22627, https://doi.org/10.1017/jfm.2025.10447
15. Eun-jin Kim, 'Intermittency and Self-Organisation in Turbulence and Statistical Mechanics', Entropy 21(6), 574 (2019), DOI 10.3390/e21060574, https://doi.org/10.3390/e21060574
16. P. Debue, V. Valori, Ch. Cuvier, F. Daviaud, J.-M. Foucaut, J.-P. Laval, Ch. Wiertel, V. Padilla, B. Dubrulle, 'Three-dimensional analysis of precursors to non-viscous dissipation in an experimental turbulent flow', Journal of Fluid Mechanics 914, A9 (2021), DOI 10.1017/jfm.2020.574, https://doi.org/10.1017/jfm.2020.574