A Cost-Effective Methodology for Spatial Concentration Distributions of Urban Air Pollutants

Abstract

A methodology to determine economically the spatial concentration distribution of the air pollutants of carbon monoxide (CO), sulphur dioxide (SO2), nitrogen monoxide (NO), nitrogen dioxide (NO2), oxides of nitrogen (NOx) and traffic rates (TR) is described. It involves the immediate transfer of samples from field to analysers for measurement and a subsequent statistical treatment. The proposed methodology has been applied in Patras using 5 and 50-l Teflon air sample bags, sampling at least 12 to 36-l actual volumes within a 20-min time interval. Totally, 221 pairs of 5-l and 112 single 50-l samples were randomly picked in morning rush hours of working days from 64 locations of a 40.0-km2 area during a winter period, when peaks of primary air pollutants usually occur due to high traffic rates and systematic inversions. Measurements were used to statistically calculate spatial average levels approximating 1-h mean concentrations with acceptable mean probable errors less than 25 % for indicative random sampling. The 1-h levels were strongly correlated to the corresponding traffic rates. Iso-concentration diagrams indicated possible zones susceptible to high pollution levels and helped to check the location appropriateness of the existing monitoring stations for (a) fixed urban-background measurements at the Vas. Georgiou A’ Sq., which was ideal, and (b) fixed traffic-oriented measurements, which should be relocated to the Ipsilon Alonion Sq. In addition, data helped to determine other points where indicative measurements should be performed. Data could be very useful for the Patras air quality assessment in conjunction with model predictions and/or objective estimation methods.