Modelling of urban air pollution in the Edmonton Strathconoa Industrial Area using artificial neural networks

Exposure to air pollution has become an exceedingly inescapable part of urban living. An important facet of the control and abatement of urban air pollution (UAP) is the use of modelling. Current modelling techniques generally use mathematics to describe transport and dispersion mechanisms of pollutants, and predict their levels at given locations away from a source. However, due to mathematical constraints, these models have limited success in dealing with complex airsheds containing many point and non-point sources of UAP.;Artificial Neural Network (ANN) modelling is a technique that effectively models non-linear type processes, such as those governing complex urban air pollution situations. The objective of this study was to investigate the use of ANN to model UAP, namely oxides of nitrogen, in the Strathcona Industrial Area, east of the City of Edmonton. This study found that ANN is a promising and effective technique for modelling hourly oxide of nitrogen fluctuations in an urban environment.