Lognormality of gradients of diffusive scalars in homogeneous, two-dimensional mixing systems

Kolmogorov’s third hypothesis, as extended by Gurvich and Yaglom, is found to be obeyed by a diffusive scalar for a class of homogeneous, two-dimensional mixing models. The mixing models all involve the advection of fluid by discrete vortices distributed in a square region with periodic boundary conditions. By computer simulation, it is found that the squared gradient of a diffusive scalar so advected is lognormally distributed, obeys the predicted scaling when a spatial smoothing is applied, and exhibits a power-law range in the spatial autocorrelation. In addition, it is found that the scaling property cuts off at the Batchelor length, as predicted by Gibson. Since the mixing models employed do not incorporate the dynamical features of high-Reynolds-number turbulence, these results suggest that scalar lognormality and associated scaling behavior may be more robust or persistent than the scaling laws of the flow field.