Helicity subgrid-scale model s and their numerical validation Preprint: AIP/123-QED
N. Yokoi, P.D. Mininni, A. Pouquet, D. Rosenberg, R. Marino · 2026
Read Helicity subgrid-scale model s and their numerical validation Preprint: AIP/123-QED by N. Yokoi, P.D. Mininni, A. Pouquet, D. Rosenberg, R. Marino (2026) 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-2603-26559.txt.
Introduction
Yokoi and co-authors test helicity-based subgrid models for large-eddy simulation of turbulence. The classic Smagorinsky model often drains too much energy, even with dynamic tuning. Helicity, the link between velocity and vorticity, adds shape data that plain energy measures miss. The team checks helicity models against direct numerical simulations, with and without net rotation. Forcing holds large-scale helicity gradients in place during the runs. They find helicity effects should join the eddy viscosity term to model small-scale turbulence well.
Catalog record
| CREATOR | N. Yokoi, P.D. Mininni, A. Pouquet, D. Rosenberg, R. Marino |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 114835 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.26559
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Helicity subgrid-scale model s and their numerical validation Preprint: AIP/123-QED N. Yokoi, P.D. Mininni, A. Pouquet, D. Rosenberg, R. Marino Abstract Large-eddy simulations (LES) with an appropriate subgrid-scale (SGS) model provide a powerful tool for investigating real-world turbulence. The Smagorinsky model, one of the simplest and most used SGS models, often shows an over-dissipative behavior even when using dynamic procedures to adjust the model coefficient. By incorporating the structural or geometrical information of turbulence provided by helicity (velocity–vorticity correlations), the helicity SGS model is expected to alleviate these issues in the standard Smagorinsky framework , in which only information of turbulence intensity is considered through the turbulent energy. The validity of helicity SGS model s is investigated here with the aid of direct numerical simulations (DNSs). Using configurations with and without net rotation, and with large-scale helicity gradients sustained by a mechanical forcing, we show that to better model SGS turbulence, SGS helicity effects should be incorporated into the model together with the Smagorinsky-like eddy viscosity. I Introduction Astrophysical, geophysical, plasma physics, and engineering flows are almost invariantly extremely turbulent. Because of huge spatial dimensions, high speeds, and very small diffusivities (including viscosity, thermal diffusivity, resistivity, etc.), the Reynolds numbers of such flows are huge. And due to the strong nonlinearity, the vast scales of motions are coupled with each other. The ratio of the largest energy-containing scale L to the smallest viscous dissipation scale \ell_{\textrm{D}} is expressed in terms of the Reynolds number Re as L/\ell_{\rm{D}}\sim Re^{3/4} . The number of grid points required for three-dimensional (3D) direct numerical simulations (DNSs), N_{\rm{G}} , scales as a result as N_{\rm{G}}\sim Re^{9/4} . Even when considering the flow associated with our own walking, the Reynolds number can be as large as Re=UL/\nu=10^{6} ( U=1\ {\rm{m}}\ {\rm{s}}^{-1} , L=1\ {\rm{m}} , \nu=10^{-6}\ {\rm{m}}^{2}\ {\rm{s}}^{-1} ) . It becomes then impossible to simultaneously solve all scales in flows of astro/geophysical or engineering interest, with even larger Re . As a result, we have to reduce the number of degrees of freedom needed to perform numerical simulations. Thus, turbulence modeling is an indispensable approach to turbulence flows. In the Reynolds-averaged Navier–Stokes (RANS) numerical approach and its associated modeling, a statistical averaging of field quantities is adopted. The statistically averaged fields are called the mean fields. In this formulation, the fluctuations are not directly calculated. They are represented instead by a turbulence model with a small number of degrees of freedom, and with the fluctuation effects in the mean-field equations given by turbulent fluxes such as the Reynolds stress, the turbulent mass flux, the turbulent heat flux, and so on. RANS models are used mostly for engineering calculations, and in some cases for some geo/astrophysical applications. Because of the statistical character of the averaging, the applicability of RANS for a fine description of unsteady turbulence is limited. Similar to the Reynolds-averaged turbulence model, large-eddy simulations (LESs) of turbulence with subgrid-scale (SGS) models provide a powerful tool for analyzing turbulent flows at very large Reynolds numbers. In LESs, by adopting an appropriate filter (with filter width denoted by \Delta ), the turbulent motions are divided into grid-scale (GS) and SGS components. If the filtering is operated in the space domain, the GS component represents the turbulent motions at large scales, while the SGS component corresponds to small scales. In a simulation, the large-scale or GS motions (large eddies) are explicitly solved, while the small-scale (SGS) turbulence motions are somehow modeled. The scales of motions in the SGS model are much smaller than their counterparts in RANS models. As a result, SGS fluctuations tend to be more homogeneous and isotropic. As a result, the SGS models in LESs are expected to be much simpler than the models in RANS simulations. There are several approaches to SGS modeling. Sagaut 2006 One of them is the so-called functional modeling, in which the effect of the small-scale (SGS) fluctuations on the large-scale (GS) motions is represented through the energetic action associated with the forward energy cascade. The GS energy is drained using a SGS tensor and given to the SGS energy. This SGS stress tensor, which represents the effect of unresolved scale (SGS) fluctuations on the resolved scale (GS) motions, is expressed specifically in terms of the GS strain rate tensor. Another approach is the so-called structural modeling, in which a part of the SGS structural information is reconstructed by assuming some relation between the statistical structure of the flow at different filtering levels. Because of the additional information somehow preserved from the unresolved scales, the prediction of the energy transfer is often greatly improved in structural models, without resorting to any prior knowledge of the nature of the interaction between the GS and SGS fluctuations. Perhaps the simplest existing functional model is the Smagorinsky model. Smagorinsky 1963 In this model, the SGS stress is expressed by the GS strain rate tensor, \mbox{\boldmath${\overline{s}}$}=\{{\overline{s}^{ij}}\} , using a SGS viscosity which is determined by \nu_{\rm{S}}=(C_{\rm{S}}\Delta)^{2}\overline{s} with \overline{s}=\sqrt{{\overline{s}}^{ij}{\overline{s}}^{ij}/2} . The model constant C_{\rm{S}} is called the Smagorinsky constant, whose value can be evaluated from the Kolmogorov energy spectrum as C_{\rm{S}}\approx 0.18 assuming the Kolmogorov constant to be equal to 1.4 . There are some drawbacks of the Smagorinsky model. The first one is the need to adjust its constant depending on the flow considered. The optimised Smagorinsky constant C_{\rm{S}} is 0.18 for homogeneous isotropic turbulence, Antonopoulos-Domis 1981 0.15 for mixing layer turbulence, Hamba 1987 and 0.10 for wall turbulence Deardorff 1970 (see Table 1 ). Since the SGS viscosity \nu_{\rm{S}} depends on C_{\rm{S}}^{2} , if we adopt the value of 0.18 for wall turbulence, the SGS viscosity is overestimated by more than three times [ (0.18/0.10)^{2}=3.24 ] compared to the corresponding \nu_{\rm{S}} in the bulk with the optimised value of 0.10 . This is not at all good from the viewpoint of the universality of the model constant. The second drawback of the Smagorinsky model is also related to this point: The Smagorinsky model is generally over-dissipative in regions of large GS strain. As a result, it tends to exaggerate the dissipative properties of turbulence transport. Finally, the Smagorinsky representation of the SGS stress does not automatically vanish very close to walls where the turbulence effects vanish. In other words, the Smagorinsky model in its simple form has no self-adaptivity in the near-wall region. We therefore have to implement a self-adaptive scheme, such as a dynamic procedure to determine the model coefficient, or adopt a damping function to satisfy the near wall turbulence behavior. In dynamic procedures some of these drawbacks are alleviated by adopting a sequential application of two filters at different scales, and by evaluating the effective coefficient through a minimization of errors. Germano, et al. 1991 However, even with this dynamic procedure, it is known that if the flow displays strong turbulent structures, e.g., in the streamwise vorticity, results obtained with the Smagorinsky model still have strong limitations. In this work we address the drawbacks of the lack of universality in the C_{\rm{S}} constant and the over-dissipative properties of the Smagorinsky model from another physical perspective. In wall turbulence, localized and very strong vortical structures develop mostly in the streamwise direction. They are positively and negatively aligned to the streamwise