The prediction of the probability density function (PDF) of a pollutant concentration within atmospheric flows is of primary importance in estimating the hazard related to accidental releases of toxic or flammable substances and their effects on human health. This need motivates studies devoted to the characterization of concentration statistics of pollutants dispersion in the lower atmosphere, and their dependence on the parameters controlling their emissions. As is known from previous experimental results, concentration fluctuations are significantly influenced by the diameter of the source and its elevation. In this study, we aim to further investigate the dependence of the dispersion process on the source configuration, including source size, elevation and emission velocity. To that end we study experimentally the influence of these parameters on the statistics of the concentration of a passive scalar, measured at several distances downwind of the source. We analyze the spatial distribution of the first four moments of the concentration PDFs, with a focus on the variance, its dissipation and production and its spectral density. The information provided by the dataset, completed by estimates of the intermittency factors, allow us to discuss the role of the main mechanisms controlling the scalar dispersion and their link to the form of the PDF. The latter is shown to be very well approximated by a Gamma distribution, irrespective of the emission conditions and the distance from the source. Concentration measurements are complemented by a detailed description of the velocity statistics, including direct estimates of the Eulerian integral length scales from two-point correlations, a measurement that has been rarely presented to date.

No abstract

No abstract

An analytical model has been developed for the flow along a street canyon (of height H and width W ), generated by an external wind blowing at any angle relative to the axis of the street. Initially, we consider the special case of a wind blowing parallel to the street. The interior of the street is decomposed into three regions, and the flow within each region is assumed to depend only on the external wind and the distance to the closest solid boundary. This decomposition leads to two different flow regimes: one for narrow streets (H/W > 1/2) and one for wide streets (H/W < 1/2). The theoretical model agrees well with results obtained from numerical simulations using a Reynolds-Averaged Navier-Stokes model. We then generalize the model to the case of arbitrary wind direction. Numerical solutions show that the streamlines of the mean flow in the street have a spiral form, and for most angles of incidence, the mass flux along the street scales on the component of the external wind resolved parallel to the street. We use this result to generalize the model derived for wind blowing parallel to the street, and the results from this model agree well with the numerical simulations. The model that has been developed can be evaluated rapidly using only very modest computing power, so it is suitable for use as an operational tool.

No abstract

No abstract

The turbulent exchange of momentum between a two-dimensional cavity and the overlying boundary layer has been studied experimentally, using hot-wire anemometry and particle image velocimetry (PIV). Conditions within the boundary layer were varied by changing the width of the canyons upstream of the test canyon, whilst maintaining the square geometry of the test canyon. The results show that turbulent transfer is due to the coupling between the instabilities generated in the shear layer above the canyons and the turbulent structures in the oncoming boundary layer. As a result, there is no single, unique velocity scale that correctly characterizes all the processes involved in the turbulent exchange of momentum across the boundary layer. Similarly, there is no single velocity scale that can characterize the different properties of the turbulent flow within the canyon, which depends strongly on the way in which turbulence from the outer flow is entrained into the cavity and carried round by the mean flow. The results from this study will be useful in developing simple parametrizations for momentum exchange in the urban canopy, in situations where the street geometry consists principally of relatively long, uniform streets arranged in gridlike patterns; they are unlikely to be applicable to sparse geometries composed of isolated three-dimensional obstacles.

We investigate the reliability of a meandering plume model in reproducing the passive scalar concentration statistics due to a continuous release in a turbulent boundary layer. More specifically, we aim to verify the physical consistency of the parametrizations adopted in the model through a systematic comparison with experimental data. In order to perform this verification, we take advantage of the velocity and concentration measurements presented in part I of the present study (Nironi et al., Boundary-Layer Meteorol, 2015) particularly concerning estimates of the Eulerian integral length scales and the higher order moments of the concentration probability density function. The study is completed by a sensitivity analysis in order to estimate the effects of the variations of the key parameter to the model results. In the light of these results, we discuss the benefits and shortcomings of this modelling approach and its suitability for operational purposes.