Scale invariance of intermittency in LES turbulence

B. Magacho 1, S. Thalabard 2, M. Buzzicotti 3, F. Bonaccorso 3, L. Biferale 3, A. A. Mailybaev 1 · 2025

Read Scale invariance of intermittency in LES turbulence by B. Magacho 1, S. Thalabard 2, M. Buzzicotti 3, F. Bonaccorso 3, L. Biferale 3, A. A. Mailybaev 1 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-15874.txt.

Introduction

Magacho and five co-authors show, in 2025 work, that a hidden sameness in wild fluid flow survives inside the cheap sums used to fake big flows. Real wild flow breaks plain self-sameness, with bursts at all sizes. A fresh view found the sameness restored when flow is read in a frame that rides and warps with the fluid.

That restored sameness, the team proves, holds in the large-eddy method too, where fine swirls are modelled, not solved. Both built-in and added fine-swirl models keep it, at solved sizes and below the grid alike. Screen runs bear the sums out with high care. Past its plant use, the method may serve as a clean lab for deep flow study.

Catalog record

CREATORB. Magacho 1, S. Thalabard 2, M. Buzzicotti 3, F. Bonaccorso 3, L. Biferale 3, A. A. Mailybaev 1
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE35 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-15874

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Scale invariance of intermittency in LES turbulence B. Magacho 1, S. Thalabard 2, M. Buzzicotti 3, F. Bonaccorso 3, L. Biferale 3, A. A. Mailybaev 1 Abstract Turbulent flows exhibit large intermittent fluctuations from inertial to dissipative scales, characterized by multifractal statistics and breaking the statistical self-similarity. It has recently been proposed that the Navier-Stokes turbulence restores a hidden form of scale invariance in the inertial interval when formulated for a dynamically (nonlinearly) rescaled quasi-Lagrangian velocity field. Here we show that such hidden self-similarity extends to the Large-Eddy Simulation (LES) approach in computational fluid dynamics (CFD). In particular, we show that classical subgrid-scale models, such as implicit or explicit Smagorinsky closures, respect the hidden scale invariance at all scales – both resolved and subgrid. In the inertial range, they reproduce the hidden scale invariance of Navier-Stokes statistics. These properties are verified very accurately by numerical simulations and, beyond CFD, turn LES into a valuable tool for fundamental turbulence research. keywords turbulence, intermittency, large-eddy simulation, hidden symmetry † † affiliation: 1 Instituto de Matemática Pura e Aplicada – IMPA, Rio de Janeiro, Brazil 2 Université Côte d’Azur, CNRS, Institut de Physique de Nice, 06200 Nice, France 3 University of Rome " Tor Vergata", Department of Physics and INFN, Rome, Italy 1 Introduction Large-eddy simulations are used in applied studies to simulate large-scale turbulent dynamics, relying on empirical modeling at small (subgrid) scales ( Pope, 2000 ; Meneveau & Katz, 2000 ; Sagaut, 2005 ; Fureby, 2008 ) . They are capable of generating intermittent (multifractal) velocity fields {\boldsymbol{u}}({\boldsymbol{x}},t) , reproducing key features of homogeneous 3D turbulence ( Cerutti & Meneveau, 1998 ; Biferale et al., 2019 ; Linkmann et al., 2018 ) . Classical description of intermittency involves, in particular, the distributions of longitudinal velocity increments \Delta_{\ell}u_{\parallel}=\big[{\boldsymbol{u}}({\boldsymbol{x}}+\ell{\boldsymbol{e}})-{\boldsymbol{u}}({\boldsymbol{x}})\big]\cdot{\boldsymbol{e}} (with {\boldsymbol{e}} arbitrary unit vector) becoming increasingly flat-tailed at small scales and leading to anomalous scaling of structure functions ( Frisch, 1995 ; Chevillard, 2015 ; Iyer et al., 2017 ) ; see Fig. 1 . The multifractal phenomenology is rooted in the scaling symmetries of the Navier-Stokes (NS) equations, suggesting that these symmetries should drive any modeling strategies of turbulence including LES. Figure 1: (a) 3D rendering of the velocity norm |{\boldsymbol{u}}| in units of the root-mean-squared velocity u_{rms}\simeq 2.4 for LES at resolution N^{3}=1024^{3} . (b) Broken self-similarity. Shown is the flatness of the longitudinal velocity increments \Delta_{\ell}u_{\parallel} for different resolutions in units of L=\pi and with forcing scale L_{f}=4L/3 ; see the implicit line in Tab. 1 for numerical details. For reference, in purple we show the anomalous scaling \propto\ell^{-0.12} . The inset shows their PDFs normalised by their standard deviation \sigma(\Delta_{\ell}u_{\parallel}) (shifted vertically for better visibility) for the 1024^{3} run at scales ranging geometrically from \delta to 128\delta from top to bottom. (a) (b) Here we investigate the Hidden Scale Invariance (HS) approach developed recently by Mailybaev (2022a) ; Mailybaev & Thalabard (2022) , which describes scaling properties of the NS in a quasi-Lagrangian reference frame ( Belinicher & L’vov, 1987 ; Mitra & Pandit, 2004 ; Biferale et al., 2011 ) after dynamic (nonlinear) rescaling. By revealing unbroken (hidden) self-similarity of intermittent statistics, this approach provides theoretical support to the anomalous scaling of structure functions ( Mailybaev, 2022b ; Thalabard & Mailybaev, 2024 ; Calascibetta et al., 2025 ) and the universal statistics of multipliers, the ratios like \Delta_{\ell}u_{\parallel}/\Delta_{\ell^{\prime}}u_{\parallel} , as conjectured by Kolmogorov in his refined theory of turbulence ( Kolmogorov, 1962 ; Chen et al., 2003 ) . Extension of the HS to SGS closures has been recently studied in shell models ( Mailybaev, 2023 ) . In this work, we prove that properly designed SGS models, for which the Smagorinsky is the simplest example, preserve the HS. This implies that the rescaled quasi-Lagrangian statistics does not depend on the observation scale, \ell , and the filtering (cut-off) scale, \delta , separately but only through their ratio \alpha=\ell/\delta . Besides, for those resolved scales \delta\ll\ell\ll L_{f} far from the cut-off and the forcing scale L_{f} , e.g. in the inertial range, the statistics becomes independent of \alpha and coincides with the analogous hidden-symmetric statistics of the NS. Performing high-resolution simulations, we provide very accurate numerical verification of HS. Such accuracy is attained because the HS in SGS models extends to subgrid scales, unlike the NS where it only refers to the inertial range. This stronger form of HS has implications for both practical (development of SGS models) and theoretical studies of turbulence. 2 Kinematic rescaling of a quasi-Lagrangian velocity Navier-Stokes-Smagorinsky equation In practice, the closure problem takes as inputs a filtering scale \delta and a filter G . It aims to predict the dynamics of the filtered field as the convolution {\boldsymbol{u}}_{\delta}=G_{\delta}*{\boldsymbol{u}} with the kernel G_{\delta}({\boldsymbol{x}})=\delta^{-3}G({\boldsymbol{x}}/\delta) , providing implicit or explicit parametrizations for the so-called subgrid scale tensor {\boldsymbol{R}}_{\delta}=\left({\boldsymbol{u}}\otimes{\boldsymbol{u}}\right)_{\delta}-{\boldsymbol{u}}_{\delta}\otimes{\boldsymbol{u}}_{\delta} . We start with the simplest example of the Smagorinsky model, in which {\boldsymbol{R}}_{\delta} is expressed as the viscous stress with the effective eddy-viscosity \nu_{s} . Dropping \delta ’s from the subscripts, the resulting Navier-Stokes-Smagorinsky (NSS) dynamics for the filtered velocity becomes \partial_{t}{\boldsymbol{u}}+{\boldsymbol{u}}\cdot\bnabla{\boldsymbol{u}}=-\bnabla p+\bnabla\cdot\left(2\nu_{s}{\boldsymbol{s}}\right)+{\boldsymbol{f}},\quad\bnabla\cdot{\boldsymbol{u}}=0, (1) where p({\boldsymbol{x}},t) is the pressure and {\boldsymbol{f}}({\boldsymbol{x}},t) is a large-scale force with correlation length L_{f}\gg\delta . The eddy viscosity is defined in terms of the filtered rate-of-strain tensor {\boldsymbol{s}} as \nu_{s}=(c_{s}\delta)^{2}\|{\boldsymbol{s}}\|,\quad{\boldsymbol{s}}=\frac{1}{2}\left(\bnabla{\boldsymbol{u}}+(\bnabla{\boldsymbol{u}})^{T}\right), (2) with the Smagorinsky constant c_{s}=0.1 and the norm \|{\boldsymbol{s}}\|^{2}=\sum 2s_{ij}^{2} . Considering the filtering scale \delta to be much larger than the viscous scale, the kinematic viscosity \nu\ll\nu_{s} is small and, therefore, omitted in the NSS equations ( 1 ). The filtering operation leading to Eq. ( 2 ) is formal only. While the NSS dynamics aims to mimic the Navier-Stokes (NS) equations for the resolved scales \ell\gg\delta , the role of eddy viscosity is to replace the energy transfer at the subgrid scales \ell\lesssim\delta by a dissipative scheme ( Cruz & Lamballais, 2023 ) . Kinematic rescaling Similarly to the NS system previously discussed by Mailybaev (2022a) ; Mailybaev & Thalabard (2022) , the starting point of our study is the projection of the velocity field {\boldsymbol{u}}({\boldsymbol{x}},t) into the rescaled velocity field {\boldsymbol{U}}({\boldsymbol{X}},\tau) . At a heuristic level, the new field represents a quasi-Lagrangian velocity increment ( Belinicher & L’vov, 1987 ) with a dynamically rescaled amplitude and timeframe. The course of proper time \tau is determined by the local level of fluctuations, running fast when the latter are high and slowly otherwise. The rescaling is parametrized by an observational scale \ell and a label {\boldsymbol{x}}_{0}\in\mathbb{R}^{3} . The label selects a Lagrangian trajectory as \dfrac{d{\boldsymbol{x}}_{*}}{dt}={{\boldsymbol{u}}}({\boldsymbol{x}}_{*},t),\quad{\boldsymbol{x}}_{*}(0)={\boldsymbol{x}}_{0}. (3) Fluctuations at scale \ell are measured as amplitudes of quasi-Lagrangian increments as a_{\ell}(t)=\sqrt{\int_{|\mathbf{X}|\leq 1}\big|\Delta_{\ell}{\boldsymbol{u}}\big({\boldsymbol{x}}_{*}(t),{\boldsymbol{X}},t\big)\big|^{2}d^{3}{\boldsymbol{X}}},\quad\Delta_{\ell}{\boldsymbol{u}}({\boldsymbol{x}},{\boldsymbol{X}},t)={\boldsymbol{u}}\big({\boldsymbol{x}}+\ell{\boldsymbol{X}},t\big)-{\boldsymbol{u}}\big({\boldsymbol{x}},t\big). (4) The space-variable {\boldsymbol{X}} represents the vectorial separation with respect to the Lagrangian trajectory in units of the observational scale \ell . The velocity field is then rescaled as {\boldsymbol{U}}({\boldsymbol{X}},\tau)=\dfrac{\Delta_{\ell}{\boldsymbol{u}}\big({\boldsymbol{x}}_{*}(t),{\boldsymbol{X}},t\big)}{a_{\ell}(t)},\quad\tau=\frac{1}{\ell}\int_{0}^{t}a_{\ell}(s)\,ds. (5) Here the first formula introduces the rescaled velocity {\boldsymbol{U}} depending on the rescaled time \tau , which is expressed in terms of the velocity difference \Delta_{\ell}{\boldsymbol{u}} at original time t . The second formula defines the relation between t and \tau , which is one-to-one for a_{\ell}(t)>0 . Physically, \tau represents a Lagrangian clock measuring time in units of the local turnover time \ell/a_{\ell}(t) encountered along the trajectory {\boldsymbol{x}}_{*}(t)

Read the complete text