Introduction: A Note on Passive Scalar Transport in Turbulence

Abstract

The first manifestation of turbulence is certainly its dramatic enhancement of diffusion for transported quantities such as smoke, tracers, or heat, for instance. In fact, our intuitive visual notion of turbulence without any recourse to velocity measurements will rely in these diffusion observations. This is true in particular for the spatially organized structures, whose evidence is generally based on visualizations of a dye marking the flow. It is therefore of prior importance to investigate the characteristics of the velocity and vorticity fields corresponding to observed diffusing passive or active scalars. Conversely, one may also study the diffusion in a turbulent flow whose velocity field is well documented experimentally. This is done in the present section for a grid turbulence (see papers by Saetran et al., Gibson et al.), a turbulence submitted to a constant mean velocity (Stapountzis and Britter), and for the turbulent channel flow (Kim and Moin). In the latter case, it is interesting to note that the “experiment” is in fact a numerical simulation. This paper shows a close correlation between the passive scalar and longitudinal velocity fluctuations in the near-wall region, resulting in streaky structures characteristic of the turbulent boundary layer close to the wall. This is more evidence of the fact that most of the “thermal coherent structures” are the signature of “velocity coherent structures”, corresponding to well-identified vortex structures. The same result has been found in the two-dimensional numerical simulations of free shear flows such as mixing layers or plane jets and wakes done by Comte et al. (see also [1] and the paper by Comte et al., these proceedings). Hence, it seems that a passive scalar (at least at Schmidt number greater than or of the order of one) or a passive temperature may be the best experimental way of characterizing the topology of turbulence.