A large eddy simulation study of mean dispersion and concentration fluctuations from a point source

In the framework of the ERC project COMTESSA (camera observation and modelling of 4D tracer dispersion in the atmosphere) a set of high-resolution Large Eddy Simulation (LES) numerical experiments has been performed to investigate the characteristics of the scalar field dispersing from a localized small source. We performed a set of simulations reproducing the dispersion from a continuous tracer release in a neutrally stratified boundary layer generated in a wind tunnel. The turbulent velocity field and scalar concentration statistics have been compared with the wind tunnel measurements of Nironi et al. (2015) and Fackrell and Robins (1982). Regarding the scalar field the comparison included the mean concentration, the higher moments up to the fourth order and the shape of the concentration probability density function (PDF) at several downwind positions. Two Eulerian and one fully Lagrangian numerical methods for the solution of the filtered scalar advection and diffusion equation have been tested with respect to their capability in reproducing the statistical characteristic of the scalar field. As expected, the non-diffusive behaviour of the Lagrangian method clearly appears in the representation of the intensity of concentration fluctuations. Unexpectedly, the Eulerian methods seem to have somewhat poorer performances compared to the Lagrangian one in the prediction of the mean concentration field. Our results show that if the first and second concentration moments are correctly predicted the LES is a reliable method to estimate the higher order moments and the concentration PDF.