Urban street canyons: Coupling dynamics, chemistry and within-canyon chemical processing of emissions

Abstract

Street canyons, formed by rows of buildings in urban environments, are associated with high levels of atmospheric pollutants emitted primarily from vehicles, and substantial human exposure. The street canyon forms a semi-enclosed environment, within which emissions may be entrained in a re-circulatory system; chemical processing of emitted compounds alters the composition of the air vented to the overlying boundary layer, compared with the primary emissions. As the prevailing atmospheric chemistry is highly non-linear, and the canyon mixing and predominant chemical reaction timescales are comparable, the combined impacts of dynamics and chemistry must be considered to quantify these effects. Here we report a model study of the coupled impacts of dynamical and chemical processing upon the atmospheric composition in a street canyon environment, to assess the impacts upon air pollutant levels within the canyon, and to quantify the extent to which within-canyon chemical processing alters the composition of canyon outflow, in comparison to the primary emissions within the canyon. A new model for the simulation of street canyon atmospheric chemical processing has been developed, by integrating an existing Large-Eddy Simulation (LES) dynamical model of canyon atmospheric motion with a detailed chemical reaction mechanism, a Reduced Chemical Scheme (RCS) comprising 51 chemical species and 136 reactions, based upon a subset of the Master Chemical Mechanism (MCM). The combined LES–RCS model is used to investigate the combined effects of mixing and chemical processing upon air quality within an idealised street canyon. The effect of the combination of dynamical (segregation) and chemical effects is determined by comparing the outputs of the full LES–RCS canyon model with those obtained when representing the canyon as a zero-dimensional box model (i.e. assuming mixing is complete and instantaneous). The LES–RCS approach predicts lower (canyon-averaged) levels of NOx, OH and HO2, but higher levels of O3, compared with the box model run under identical chemical and emissions conditions. When considering the level of chemical detail implemented, segregation effects were found to reduce the error introduced by simplifying the reaction mechanism. Chemical processing of emissions within the canyon leads to a significant increase in the Ox flux from the canyon into the overlying boundary layer, relative to primary emissions, for the idealised case considered here. These results demonstrate that within-canyon atmospheric chemical processing can substantially alter the concentrations of pollutants injected into the urban canopy layer, compared with the raw emission rates within the street canyon. The extent to which these effects occur is likely to be dependent upon the nature of the domain (canyon aspect ratio), prevailing meteorology and emission/pollution scenario considered.