Turbulence length scales in a low-roughness near-neutral atmospheric surface layer

Abstract

This paper investigated the integral length scales of turbulence in a low-roughness atmospheric surface layer (ASL), characterised by very smooth terrain in the Utah desert during near-neutral conditions, and evaluated the Engineering Sciences Data Unit (ESDU) 85020 and 86010 predictions for the turbulence length scales in a low-roughness ASL. The correlation integral method was used to estimate the integral length scales of the velocity components with longitudinal, lateral and vertical separations from sonic measurements on a vertical tower and spanwise array in the Surface Layer Turbulence and Environmental Science Test (SLTEST) field experiment. It was found that the longitudinal integral length scales calculated using near-neutral SLTEST data followed a logarithmic relationship with height proportional to the mean velocity profile with approximately constant integral time scale, however the sizes of the longitudinal components of the energy-containing eddies in the low-roughness flat terrain were 2–3 times smaller than those previously measured during field experiments in open country terrains. The calculated length scales with longitudinal separations over the very smooth terrain characteristics of the salt flats at Dugway were not consistent with those predicted by ESDU 85020. In contrast, the scaling of the lateral and vertical components of the three-dimensional turbulence structure with respect to the longitudinal component in the low-roughness ASL were consistent with similarity theory predictions in ESDU 86010 that the scaling ratios are independent of terrain roughness. Furthermore, this confirms the large dependence of the longitudinal turbulence length scales on the upstream terrain roughness and highlights the large variation of turbulence length scales observed at different low-roughness sites in the literature.