Large eddy simulation of reactive pollutant dispersion : the analysis of physical and chemical interacting processes over idealized urban street canyons

Abstract

The dispersion of inert pollutants over urban areas has been studied for a long time. However, the dispersion of chemically reactive pollutants is not well known. Nowadays, the popular research topics mainly focus on the phenomena of the pollutant dispersion over complicated city morphologies, but limited studies have attempted to explore the mechanism of reactive pollutant dispersion.

In this thesis, the incompressible large eddy simulation (LES) coupled with chemical transport equations is developed to simulate the turbulent flows and the associated chemical reactions. The isothermal condition is applied to create the neutral stratification. The titration of ozone (O3) and the photolysis of nitrogen dioxide (NO2) take place in the computational domain filled with O3 of 1 ppm. The source of nitric oxide (NO) with the mixing ratio of 1 ppm, 10 ppm, 100 ppm or 1000 ppm is placed on the ground surface of the first street canyon. The idealized 2D street canyons are used so as to provide a parameterized description of the urban roughness, and the roughness is controlled by the aspect ratio (AR) of the building height to the street width with the range from 1 to 1/8.

The results show that the physical parameters such as the friction factor, the Reynolds shear stress, and the turbulent kinetic energy increase significantly with decreasing AR from 1 to 1/8. However, the influence of the AR value to the turbulent timescales is limited to a relatively small range. Although the pollutant is actually emitted at the ground level, it is reasonable to assume that the pollutant is effectively emitted and reflected at the roof level when the plume of the pollutant dispersion in the roughness sublayer (RSL) are studied for idealized street canyon models with uniform buildings.

The plume size analysis shows that the dispersion of nonreactive pollutants is enhanced by higher friction factors. However, based on the NO mixing ratio and the AR value, the available configurations with the ozone titration can be divided into group (A) and group (B), which are respectively characterised by the timescale ratios of NO and O3. In group (A), the NO dispersion is suppressed, but the NO2 dispersion is enhanced. In group (B), the NO dispersion is enhanced, while the NO2 dispersion keeps in line with nonreactive pollutants. For the more realistic NOx - O3 cycle, group (A) has similar relations as the ozone titration, but the NO and NO2 dispersions in group (B) are not significantly changed by chemical reactions. According to the similarity analysis, the findings in this study can be generalized to other situations with similar Damköhler numbers.