Handout (1.1 MB)
The formulation for the cross-wind variance proposed by Mahrt et al. (2001) is revisited using four newer datasets not used in tuning the original formulation. Improvements to the formulation are proposed. The model distinguishes between mesoscale and turbulence motions. Such partitioning is performed because the physics responsible for the fluctuations is very different, and their influence on the plume is very different. Mesoscale motions in the stable boundary layer include internal gravity waves, pulsating drainage flows, vortical modes, wake vortices and others. Turbulent motions are primarily due to the local vertical wind shear and temperature stratification. Turbulence is dispersive while the lower-frequency mesoscale motions typically advect the plume and may or may not be dispersive.
Fluctuations on turbulence time scales scale with the friction velocity and the scaled values exhibit minor variation between sites. The mesoscale motions do not scale with the local friction velocity. Systematic differences between sites are explored.