Assessment and prediction of benzene concentrations in a street canyon using a variety of models, in the means of environmental management purposes

Urban air quality nowadays is one of the major environmental issues, due to its impact to
various health problems, caused by the daily exposure of the population in dangerous air
pollutants. The highest levels of air pollutants are usually observed in street canyons, emitted
by urban traffic. To understand the way that pollutants dispersed in a street canyon
environment, various modeling techniques are used. Both the accuracy of the predictions and
the quickness of the calculations, are significant factors to adopt a modeling technique.
In the present study, we used benzene as an indicator of traffic pollution. Two different
modeling techniques, an artificial neural network (ANN) and a semi empirical deterministic
model (DET) are used, to predict benzene concentrations in a street canyon environment.
The ANN was based on a training procedure using measurements collected in a specific
street canyon (benzene concentrations, traffic density, vehicle’s type distribution), while the
DET model was based on road traffic emission rate, wind speed and direction, and the
geometrical characteristics of the road.
After the validation of models, their response to preselected “what if” scenarios was
attempted.
Although both models produced very good results, given the limited amount of data available,
the ANN succeeded slightly better than DET in predicting benzene concentrations. On the
contrary, ANN presents lower response in predicting the effect of significant changes in traffic
flow patterns on benzene concentrations.
The results from the simulations indicate that the ANN is a promising technique for benzene
modeling in an urban environment, since DET seems to be ideal for environmental
management purposes.