Aerosol dispersion in the area surrounding an existing biological treatment facility is investigated using large-eddy simulation, with the objective to investigate the applicability of computational fluid dynamics to complex real-life problems. The aerosol sources consist of two large aeration ponds that slowly diffuse aerosols into the atmosphere. These sources are modelled as dilute concentrations of a non-buoyant non-reacting pollutant diffusing from two horizontal surfaces. The time frame of the aerosol release is restricted to the order of minutes, justifying a statistically steady inlet boundary condition. The numerical results are compared to wind-tunnel experiments for validation. The wind-tunnel flow characteristics resemble neutral atmospheric conditions with a Reynolds number, based on the boundary-layer thickness, of Re δ ≈ 2 × 10 5 . The numerical inflow conditions are based upon the wind-tunnel flow field. The predicted decay of both the mean and root-mean-square concentrations are in good agreement with experimental data; at 3 m from the ground, the plume mean concentration 200 m downwind of the source is approximately 2% of the source strength. The numerical data in the near-surface layer (0-50 m from the ground) correspond particularly well with the wind-tunnel data. Tentative deposition simulations suggest that there seems to be little difference in the deposition rates of large (1.8 × 10 −5 m) and small (3 × 10 −6 m) particles in the near-field under the flow conditions considered.

rotating frame of reference both have a tendency to suppress turbulence fluctuations and even cause relaminarization. In the latter case, both internal frictional losses and the mixing of momentum are substantially reduced. The implications are a significantly reduced rate of conversion of turbulence kinetic energy to internal energy (by the action of viscosity), reduced scale separation between small and large scale turbulence, and possibly increased large-scale anisotropy.Transport and dispersion of passive contaminants are governed by turbulence and mean flow advection, rather than by molecular diffusion; the contaminant simply follows the large-scale three-dimensional and time-dependent velocity field. Passive contaminant transport is therefore expected to respond significantly to changes in the componental structure of the flow field caused by the imposition of a stably stratified background. An analogous situation occurs for dense-gas releases in an isothermal environment. In this case, the density variation is caused by the density difference between the contaminant itself and the ambient air.Regardless of origin, spatial variation of fluid density generally affects all turbulence scales of the flow, changing the spatial and componental structure and in some cases even the rate of transfer of kinetic energy to internal energy. These particular features make stably stratified fluid flows challenging to model, both using statistically based turbulence closures and eddy-resolving simulations that rely on subgrid-scale models. The practical importance of stably stratified shear turbulence in general and its relevance to contaminant transport and dispersion constitute the primary motivations for the present study.The componental structure of the flow is most commonly characterized by considering the anisotropy of the turbulence, e.g., the departure from an equipartitioning of turbulence energy in the three spatial directions. Several quantities can be derived and examined in order to analyze the anisotropy of a turbulent flow. In the work of Smyth and Moum, 1 anisotropy tensors are derived from the Reynolds stress tensor and the viscous dissipation rate tensor, among others. Investigation of the anisotropy tensor components, the invariants, and the principal axes of these anisotropy tensors is conducted to shed light on the anisotropy of the flow. Smyth and Moum 1 also consider various spectral measures of the flow, but no examination of the vertical variations within the shear layer is carried out.A more straight-forward analysis of turbulence anisotropy is found in Thoroddsen and Van Atta, 2 where the ratios of various components of the Reynolds stress and viscous dissipation rate tensors are reported. Reynolds-stress correlation coefficients, cf., e.g., Kim, Moin, and Moser, 3 may also shed light on turbulence anisotropy. Lumley 4 demonstrates the use of the well-known "anisotropy invariant map" in order to characterize the anisotropy of a tensor, whereas Kassinos, Reynolds, and Rogers 5 show how the structu...

[1] A model for the turbulence dissipation rate in stably stratified shear turbulence is developed and validated. The functional dependence of the model is derived from first principles and it represents a conceptually new approach in that it depends on the background temperature field rather than on the fluctuating velocity field. This novel feature makes the proposed model a viable candidate for dissipation rate estimates in measured real-life flows. Direct numerical simulation data are used for a priori assessment of the proposed model. It is demonstrated that the proposed model performs very well, particularly in cases where the background stratification becomes dynamically important. Also, a generalized expression for the mixing coefficient has been rigorously derived from first principles assuming local isotropy of incompressible turbulent flows. The mixing coefficient is shown to depend on the Prandtl number and values are in correspondence with previous studies.