Michel Stefanello

Characterization of Wind Meandering in Low-Wind-Speed Conditions

Investigation of low-wind cases observed during the Urban Turbulent Project campaign (Torino, Italy) and at the Santa Maria meteorological station (Santa Maria, Brazil) provides insight into the wind-meandering phenomenon, i.e. large, non-turbulent oscillations of horizontal wind speed and temperature. Meandering and non-meandering cases are identified through analysis of the Eulerian autocorrelation functions of the horizontal wind-velocity components and temperature. When all three autocorrelation functions oscillate, meandering is present. As with weak turbulence, meandering shows no dependence on stability but is influenced by presence of buildings and depends on wind speed. We show that, while the standard deviation of the horizontal velocity is always large in low-wind conditions, the standard deviation of the vertical velocity shows very different behaviour in meandering and non-meandering conditions. In particular, the value of the ratio of the standard deviations of the vertical and horizontal velocities typifies the meandering condition.

DIRECT NUMERICAL SIMULATION OF AN STABLY STRATIFIED EKMAN FLOW FROM THE INCOMPACT3D CODE

In a diurnal cycle, distinct thermal and mechanical forcing generates different manifestations of a planetary boundary layer. The stable boundary layer occurs when the soil surface has a lower temperature than the air above. In this layer, wind shear is the main mechanism of turbulence generation. In the present study, a direct numerical simulation of an Ekman layer over a smooth wall is presented to investigate the different turbulent patterns that occur during evolution from a neutral boundary layer to a weakly stable boundary layer. The preliminary study shows the appearance of turbulent structures near the surface, due to the imposition of a stratification.

OBTAINING THE MEANDERING LOOP PARAMETER IN LOW WIND SPEED CONDITION

The looping parameter m is the main value to characterize the meandering phenomenon. This parameter is relationship with negative lobes in the observed autocorrelation function generated from components of horizontal speed. In this work, we present a study comparing the mean values of the looping parameter between 2 diverse sites in the Brazilian sector.

Eddy diffusivities for the convective boundary layer derived from LES spectral data

Large Eddy Simulation (LES) spectral data and Taylor statistical diffusion theory are used to obtain Eddy diffusivities in a convective boundary layer. The derivation employs a fitting expression obtained from LES data for the vertical peak frequency. The vertical Eddy diffusivities are well behaved and show similar patterns and magnitudes as those derived from experimental spectral peak frequency data. In addition, this new vertical Eddy diffusivity was introduced into an advection diffusion equation which was solved by Generalized Integral Laplace Transform Technique (GILLT) method and validated with observed contaminant concentration data of the Copenhagen experiment. The results of this new approach are shown to agree with the measurements of Copenhagen.