Abstract
The health of urban communities is under threat from poor air quality arising from motor vehicle exhaust emissions. Air quality monitoring is usually undertaken in a few fixed sites, giving little information on the exposure of individual people. This paper describes a modelling strategy designed to provide more precise exposure data in order to quantify the links between air quality and human health. Effective, semi-empirical box models have been developed for street canyons and provide a good fit to data collected in Glasgow. More advanced CFD models in two and three dimensions have also been developed using PHOENICS and give quantitatively accurate results for pollutant concentrations and vortex formation in street canyons.