Factors influencing the variability of SO2 concentrations in Istanbul

The correlation between sulfur dioxide (SO2) concentrations measured at the European and Asian sides of Istanbul and meteorological parameters is inve stigated using principal component analysis (PCA) and multiple regression a nalysis techniques. Several meteorological parameters are selected to repre sent the atmospheric conditions during two winter periods: 1993-1994 and 19 94-995. Six principal components are found to explain the majority of the o bserved meteorological variability. Surface pressure, 850-mb temperature, a nd surface zonal (east-west) and meridional (north-south) winds show high l oadings on separate factors identified by PCA. We seek dominant meteorologi cal parameters that control the SO2 levels at each monitoring station. Seve ral multiple regression analysis models are fitted to the data from each mo nitoring station using six principal components and previous-day SO2 concen trations as independent variables. Results suggest that the most important parameters, highly correlated with SO2 concentrations in the Istanbul metropolitan area, are atmospheric press ure and surface zonal and meridional winds. These components have more infl uence on the determination of the air pollution levels at the Asian side th an at the European side.