Profiles of high-order moments of longitudinal velocity explained by the random sweeping decorrelation hypothesis

Under the assumptions that the random sweeping decorrelation hypothesis applies and that the velocity statistics are near Gaussian, the logarithmic variation of high-order moments of longitudinal velocity with distance from a boundary in the inertial region (where the logarithmic law holds for the mean longitudinal velocity) is explained by the existence of a -1 power law in the longitudinal velocity spectrum. During the idealized horizontal planar array study for quantifying surface heterogeneity, measurements and profiles of longitudinal velocity were collected within the first meter from the surface under mild atmospheric thermal stratification. These measurements show good agreement with the proposed theory. Further investigation into the validity of the random sweeping decorrelation hypothesis reveals that it is not strictly valid across all scales but can be viewed as operationally viable due to inherent cancellation in its interaction terms. More importantly, deviations from the random sweeping decorrelation hypothesis predictions appear consistent across the logarithmic region and captured by a quasiconstant, suggesting possible avenues for correction in the modeling of high-order moments.

Duke Scholars

Author Gabriel G. Katul Civil and Environmental Engineering