To understand the generation mechanism of a hierarchy of multiscale vortices in a high-Reynolds-number turbulent boundary layer, we conduct direct numerical simulations and educe the hierarchy of vortices by applying a coarse-graining method to the simulated turbulent velocity field. When the Reynolds number is high enough for the premultiplied energy spectrum of the streamwise velocity component to show the second peak and for the energy spectrum to obey the $-5/3$ power law, small-scale vortices, that is, vortices sufficiently smaller than the height from the wall, in the log layer are generated predominantly by the stretching in strain-rate fields at larger scales rather than by the mean-flow stretching. In such a case, the twice-larger scale contributes most to the stretching of smaller-scale vortices. This generation mechanism of small-scale vortices is similar to the one observed in fully developed turbulence in a periodic cube and consistent with the picture of the energy cascade. On the other hand, large-scale vortices, that is, vortices as large as the height, are stretched and amplified directly by the mean flow. We show quantitative evidence of these scale-dependent generation mechanisms of vortices on the basis of numerical analyses of the scale-dependent enstrophy production rate. We also demonstrate concrete examples of the generation process of the hierarchy of multiscale vortices.

Abstract

To investigate the transport of heavy small particles (inertial particles) in high-Reynolds- number wall turbulence, we conduct direct numerical simulations of inertial particles in turbulent channel flow at the friction Reynolds number $Re_\tau =1000$ . In the statistically steady state, particles distribute inhomogeneously; particles with different relaxation times form voids with different sizes in the bulk of the flow, whereas they form streak-like clusters with different widths along the streamwise direction. To explore the origin of the multiscale voids and clusters of particles, we identify objectively the axes of tubular vortices and the spines of streaks with different sizes. These identifications enable us to show quantitatively that (i) vortices sweep out the particles of the relaxation time comparable with their turnover time, irrespective of their size and existing height, and that, among these particles, (ii) those swept out by wall-detached tubular vortices form clusters isotropically around them, whereas (iii) those swept out by wall-attached vortices are attracted by a nearby low-speed streak. These explain the reason why the multiscale clusterings are well described in terms of the local Stokes number defined by the turnover time of multiscale vortices. Furthermore, these descriptions of the particle transport give us a clear view to understand velocity statistics and wall-deposition mechanism of inertial particles in high-Reynolds-number wall turbulence.

We have conducted direct numerical simulations of a turbulent boundary layer for the momentum-thickness-based Reynolds number Re θ = 1804600. To extract the largestscale vortices, we coarse-grain the uctuating velocity elds by using a Gaussian lter with the lter width comparable to the boundary layer thickness. Most of the largestscale vortices identied by isosurfaces of the second invariant of the coarse-grained velocity gradient tensor are similar to coherent vortices observed in low-Reynolds-number regions, that is, hairpin vortices or quasi-streamwise vortices inclined to the wall. We also develop a percolation analysis to investigate the threshold-dependence of the isosurfaces and objectively identify the largest-scale hairpin vortices in terms of the coarsegrained vorticity, which leads to the quantitative evidence that they never disappear even in fully developed turbulent regions. Hence, we conclude that hairpin vortices exist in the largest-scale structures irrespective of the Reynolds number.

No abstract