In 2008 the International Collaboration for Turbulence Research compared eight data sets — three particle-tracking experiments and five direct numerical simulations spanning Taylor-scale Reynolds numbers Rλ=120R_\lambda=120–740740 — and found that Lagrangian velocity structure functions up to order p=4p=4 collapse onto a common behavior, and that a multifractal description extended to the dissipative range, with a free crossover parameter β\beta controlling the transition near the Kolmogorov time, captures intermittency over three decades of time lags. The issues the Letter itself identified remain unresolved. First, it is not known whether a strong mean flow or persistent large-scale anisotropy, as in turbulent jets and wall-bounded turbulence, breaks the recovery of small-scale Lagrangian universality, or whether the multifractal parametrization must be modified in such flows. Second, the proposed physical identification of β\beta — different values corresponding to different statistical weights of vortex-filament trapping along particle trajectories — has never been established quantitatively. Third, higher-order Lagrangian moments (p>4p>4) were beyond statistical convergence in 2008; direct numerical simulations now reach Rλ≈1300R_\lambda\approx1300 and orders near p≈10p\approx10, but the multifractal prediction for dissipative-range and acceleration statistics is contested, and converged experimental high-order data remain scarce. The problem is to settle the universality and the physical content of the multifractal description of Lagrangian intermittency outside the regime in which it was validated.

Background

A fluid particle advected by a turbulent flow defines Lagrangian velocity increments δτv=v(t+τ)−v(t)\delta_\tau v=v(t+\tau)-v(t); the moments S(p)(τ)=⟨(δτv)p⟩S^{(p)}(\tau)=\langle(\delta_\tau v)^p\rangle are the Lagrangian velocity structure functions (LVSF). Scale separation is measured by the Taylor-scale Reynolds number RλR_\lambda, since the ratio of the large integral time TLT_L to the Kolmogorov dissipative time τη\tau_\eta (the shortest dynamically relevant time, set by the mean energy dissipation rate and the viscosity) grows as TL/τη∼RλT_L/\tau_\eta\sim R_\lambda. Dimensional (Kolmogorov 1941) reasoning predicts S(p)(τ)∼(ϵτ)p/2S^{(p)}(\tau)\sim(\epsilon\tau)^{p/2} in the inertial range τη≪τ≪TL\tau_\eta\ll\tau\ll T_L, but measured increment distributions develop increasingly heavy, non-Gaussian tails at small τ\tau — intermittency — which manifests as anomalous scaling exponents ξ(p)≠p/2\xi(p)\neq p/2. Intermittency can be monitored scale by scale through the local exponent ζ(p,τ)=dlog⁡S(p)/dlog⁡S(2)\zeta(p,\tau)=d\log S^{(p)}/d\log S^{(2)}, which flattens toward an inertial-range plateau only for τ\tau well above τη\tau_\eta and shows a pronounced dip around τ∼τη\tau\sim\tau_\eta.

In the Parisi–Frisch multifractal (MF) picture, Eulerian velocity increments scale as δru∼rh\delta_r u\sim r^h on sets of fractal dimension D(h)D(h), and the dimensional bridge τ∼r/δru\tau\sim r/\delta_r u extends this to time increments along trajectories, with the dissipative time itself fluctuating, τη(h)/TL∼Rλ2(h−1)/(1+h)\tau_\eta(h)/T_L\sim R_\lambda^{2(h-1)/(1+h)}. To describe Lagrangian data at all lags, including the dissipative range, Chevillard et al. (2003) and Biferale et al. (2004) introduced a smooth crossover formula for the conditional increment δτv(h)\delta_\tau v(h) — Eq. (4) of the source Letter — built from the powers τ/TL\tau/T_L and τη(h)/TL\tau_\eta(h)/T_L joined through a free parameter β\beta that controls the sharpness of the transition near τ∼τη\tau\sim\tau_\eta: it reduces to inertial-range scaling δτv∼(τ/TL)h\delta_\tau v\sim(\tau/T_L)^h at large lags and to smooth, differentiable behavior at small lags, and it fixes the depth of the dip of ζ(p,τ)\zeta(p,\tau) around τη\tau_\eta. Vortex filaments — intense, thin, tube-like vortical structures at dissipative scales — can trap tracer particles for times of order τη\tau_\eta; simulations cited in the Letter showed that depleting or enhancing the probability of particle trapping in filaments depletes or enhances exactly this dip.

The source Letter (International Collaboration for Turbulence Research, Phys. Rev. Lett. 100, 254504 (2008)) compared eight data sets — three experiments (particle tracking velocimetry, the optical reconstruction of tracer trajectories, and acoustic Doppler tracking, which infers tracer velocity from the Doppler shift of sound scattered by the tracers; Rλ=124R_\lambda=124–740740) and five direct numerical simulations (DNS, Rλ=140R_\lambda=140–650650) — and found (i) that the local exponents ζ(4,τ)\zeta(4,\tau), fourth order being the highest order converged in all data sets, collapse across flows with isotropy recovered at small lags, evidencing universality of the Lagrangian energy cascade; and (ii) that the MF description, with β=4\beta=4 and Eulerian D(h)D(h) taken from longitudinal or transverse fits to the log-Poisson cascade spectrum (a standard intermittent-cascade ansatz for D(h)D(h)), captures the data across three decades of τ\tau for Rλ∈[150,800]R_\lambda\in[150,800] (with asymptotic predictions ζ(4)≃1.71\zeta(4)\simeq1.71 and ≃1.59\simeq1.59 from the two spectra). The validated regime is thus statistically quasi-isotropic homogeneous flows and moments up to p=4p=4, with the crossover parameter β\beta a fitted constant of contested physical meaning (a related review, Chevillard et al. 2012, denotes the analogous sharpness parameter γ\gamma and reports fitted values from about 1 to 4 across data sets, attributing some of the variation to finite-tracer-size filtering).

Problem Statement

Determine whether the universal multifractal description of small-scale Lagrangian intermittency established by the International Collaboration for Turbulence Research (Phys. Rev. Lett. 100, 254504 (2008)) extends beyond the regime in which it was validated — statistically quasi-isotropic flows and velocity-increment moments up to order p=4p=4 — by addressing three linked questions on the validation and physical grounding of that description:

(1) Strong mean flow and persistent anisotropy. Does the observed collapse of the normalized local exponents ζ(p,τ)\zeta(p,\tau) across independent data sets — the recovery of small-scale Lagrangian universality, with isotropy restored at small time lags — survive flows with a strong mean flow or persistent large-scale anisotropy, such as turbulent jets and wall-bounded turbulence? If it does not, determine how the multifractal parametrization (including the crossover parameter β\beta) must be modified for such flows.

(2) Physical content of the crossover parameter. The source Letter proposed that different values of β\beta correspond to different statistical weights of vortex filaments along particle trajectories. Establish or refute this quantitatively: relate β\beta — or equivalently the crossover shape it controls near τ≈τη\tau\approx\tau_\eta, including the dip of ζ(p,τ)\zeta(p,\tau) — to measurable statistics of particle trapping in vortex filaments (for example, residence-time or trapping-event statistics conditioned on flow topology along trajectories).

(3) Higher-order statistics. Test whether (i) universality across flows and (ii) the multifractal description, including its dissipative-range crossover, hold at orders p>4p>4 (and for derived high-order statistics such as acceleration moments), using the converged Lagrangian velocity-structure-function moments available at Taylor-scale Reynolds numbers beyond those of the 2008 data sets — recent DNS already provides Lagrangian exponents up to p≈10p\approx10 at Rλ≤1300R_\lambda\le1300 (Buaria and Sreenivasan 2023), while converged experimental high-order data and independent cross-data-set coverage remain the outstanding gap; existing experimental data, where available, may be re-analyzed.

These three questions are jointly required: resolving the objective requires addressing all three together, and an answer to any subset resolves only that component. Reducing the methodological biases that limit the 2008 comparison (finite measurement-volume effects in experiments, tracer-interpolation and subgrid-dissipation differences in simulations) is not a separate requirement but a scope qualifier that sets the lag ranges, orders, and flow configurations over which the comparison can be made.

References

  1. International Collaboration for Turbulence Research: 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. Lévêque, B. Lüthi, J. Mann, N. Mordant, W.-C. Müller, S. Ott, N. T. Ouellette, J.-F. Pinton, S. B. Pope, G. S. Roux, F. Toschi, H. Xu, and P. K. Yeung, 'Universal intermittent properties of particle trajectories in highly turbulent flows', Phys. Rev. Lett. 100, 254504 (2008). DOI: 10.1103/PhysRevLett.100.254504; arXiv:0802.3776. https://doi.org/10.1103/PhysRevLett.100.254504
  2. L. Chevillard, S. G. Roux, E. Lévêque, N. Mordant, J.-F. Pinton, and A. Arnéodo, 'Lagrangian velocity statistics in turbulent flows: effects of dissipation', Phys. Rev. Lett. 91, 214502 (2003). DOI: 10.1103/PhysRevLett.91.214502. https://doi.org/10.1103/PhysRevLett.91.214502
  3. L. Biferale, G. Boffetta, A. Celani, B. J. Devenish, A. Lanotte, and F. Toschi, 'Multifractal statistics of Lagrangian velocity and acceleration in turbulence', Phys. Rev. Lett. 93, 064502 (2004). DOI: 10.1103/PhysRevLett.93.064502. https://doi.org/10.1103/PhysRevLett.93.064502
  4. Mark Peter Rast and Jean-Francois Pinton, 'Point-vortex model for Lagrangian intermittency in turbulence', Phys. Rev. E 79, 046314 (2009). DOI: 10.1103/PhysRevE.79.046314. https://doi.org/10.1103/PhysRevE.79.046314
  5. Federico Toschi and Eberhard Bodenschatz, 'Lagrangian properties of particles in turbulence', Annu. Rev. Fluid Mech. 41, 375–404 (2009). DOI: 10.1146/annurev.fluid.010908.165210. https://doi.org/10.1146/annurev.fluid.010908.165210
  6. Jacob Berg, Søren Ott, Jakob Mann, and Beat Lüthi, 'Experimental investigation of Lagrangian structure functions in turbulence', Phys. Rev. E 80, 026316 (2009). DOI: 10.1103/PhysRevE.80.026316. https://doi.org/10.1103/PhysRevE.80.026316
  7. Laurent Chevillard, Bernard Castaing, Alain Arneodo, Emmanuel Lévêque, Jean-François Pinton, and Stéphane G. Roux, 'A phenomenological theory of Eulerian and Lagrangian velocity fluctuations in turbulent flows', C. R. Physique 13, 899–928 (2012). DOI: 10.1016/j.crhy.2012.09.002; arXiv:1112.1036. https://doi.org/10.1016/j.crhy.2012.09.002
  8. Michael Wilczek, Haitao Xu, Nicholas T. Ouellette, Rudolf Friedrich, and Eberhard Bodenschatz, 'Generation of Lagrangian intermittency in turbulence by a self-similar mechanism', New J. Phys. 15, 055015 (2013). DOI: 10.1088/1367-2630/15/5/055015. https://doi.org/10.1088/1367-2630/15/5/055015
  9. Brian L. Sawford and P. K. Yeung, 'Direct numerical simulation studies of Lagrangian intermittency in turbulence', Phys. Fluids 27, 065109 (2015). DOI: 10.1063/1.4922205. https://doi.org/10.1063/1.4922205
  10. Nickolas Stelzenmuller, Juan Ignacio Polanco, Laure Vignal, Ivana Vinkovic, and Nicolas Mordant, 'Lagrangian acceleration statistics in a turbulent channel flow', Phys. Rev. Fluids 2, 054602 (2017). DOI: 10.1103/PhysRevFluids.2.054602. https://doi.org/10.1103/PhysRevFluids.2.054602
  11. Lukas Bentkamp, Cristian C. Lalescu, and Michael Wilczek, 'Persistent accelerations disentangle Lagrangian turbulence', Nat. Commun. 10, 3550 (2019). DOI: 10.1038/s41467-019-11060-9. https://doi.org/10.1038/s41467-019-11060-9
  12. R. Watteaux, G. Sardina, L. Brandt, and D. Iudicone, 'On the time scales and structure of Lagrangian intermittency in homogeneous isotropic turbulence', J. Fluid Mech. 867, 438–481 (2019). DOI: 10.1017/jfm.2019.127. https://doi.org/10.1017/jfm.2019.127
  13. Ron Shnapp, 'On small-scale and large-scale intermittency of Lagrangian statistics in canopy flow', J. Fluid Mech. 913, R2 (2021). DOI: 10.1017/jfm.2021.33. https://doi.org/10.1017/jfm.2021.33
  14. S. Santosh Kumar, Xinyi Huang, Xiang Yang, and Jiarong Hong, 'Three dimensional flow motions in the viscous sublayer', Theor. Appl. Mech. Lett. 11, 100239 (2021). DOI: 10.1016/j.taml.2021.100239. https://doi.org/10.1016/j.taml.2021.100239
  15. Valentina Valori, Paul Debue, Adam Cheminet, Tarek Chaabo, Yaşar Ostovan, Christophe Cuvier, Jean-Marc Foucaut, Jean-Philippe Laval, Cécile Wiertel, Vincent Padilla, François Daviaud, and Bérengère Dubrulle, 'On Lagrangian intermittency from 4-D particle tracking velocimetry measurements of a turbulent von Kármán flow', arXiv:2104.04222 (2021). https://arxiv.org/abs/2104.04222
  16. Bianca Viggiano, Thomas Basset, Stephen Solovitz, Thomas Barois, Mathieu Gibert, Nicolas Mordant, Laurent Chevillard, Romain Volk, Mickaël Bourgoin, and Raúl Bayoán Cal, 'Lagrangian diffusion properties of a free shear turbulent jet', J. Fluid Mech. 918, A25 (2021). DOI: 10.1017/jfm.2021.325. https://doi.org/10.1017/jfm.2021.325
  17. Bianca Viggiano, Thomas Basset, Mickaël Bourgoin, Raúl Bayoán Cal, Laurent Chevillard, Charles Meneveau, and Romain Volk, 'Lagrangian modeling of a nonhomogeneous turbulent shear flow: Molding homogeneous and isotropic trajectories into a jet', Phys. Rev. Fluids 9, 044604 (2024). DOI: 10.1103/PhysRevFluids.9.044604; arXiv:2310.04330. https://doi.org/10.1103/PhysRevFluids.9.044604
  18. Dhawal Buaria and Katepalli R. Sreenivasan, 'Scaling of acceleration statistics in high Reynolds number turbulence', Phys. Rev. Lett. 128, 234502 (2022). DOI: 10.1103/PhysRevLett.128.234502; arXiv:2202.09682. https://doi.org/10.1103/PhysRevLett.128.234502
  19. Dhawal Buaria and Katepalli R. Sreenivasan, 'Saturation and multifractality of Lagrangian and Eulerian scaling exponents in three-dimensional turbulence', Phys. Rev. Lett. 131, 204001 (2023). DOI: 10.1103/PhysRevLett.131.204001; arXiv:2307.04846. https://doi.org/10.1103/PhysRevLett.131.204001