An integrative approach for determination of air pollution and its health effects in a coal fired power plant area by passive sampling

Ambient concentrations of nitrogen dioxide (NO2), sulfur dioxide (SO2), ozone (O3) and volatile organic compounds (VOCs) were measured at several locations in Kütahya, a severely polluted city and also characterized as a thermal power plant city, in Turkey. Two-week extensive passive sampling campaigns were carried out in summer and winter at 108 sampling sites that were classified into three main groups as urban, rural and industrial. Spatial and seasonal distributions of the measured pollutants were evaluated employing Geographical Information System techniques. All pollutant concentrations showed an increasing pattern in winter, except for ozone. The concentrations of VOCs were substantially higher particularly at sampling sites with high traffic and population densities. Power plants were noted as important sources for VOCs since high concentrations were measured especially around the power plants. Highest NO2 levels were observed in the city center while there was a general decrease in the concentration levels far away from the city center. Considerably higher SO2 levels were observed in the settlements where local coal is used for residential heating. Seasonal variations in SO2 concentrations were quite low around the thermal power plants indicating their important effect on atmospheric levels. A basic population exposure assessment was conducted for two largest settlements of the province (Kütahya city center and Tavşanlı) by combining population density maps with pollutant distribution maps of NO2 and SO2. Exposure to NO2 and SO2 were assessed separately according to a classification made for different degrees of exposure. Cancer risks associated with inhalation of benzene were also estimated. Higher risk values were obtained from the sampling sites with higher population densities, especially in winter. Risk values estimated for 95 sampling sites were higher than EPA's acceptable risk value (1 × 10−6).