Multivariate methods for ground-level ozone modeling

The aim of this study is to apply multivariate statistical methods in predicting ozone (O3) concentrations at the ground level of the troposphere as the function of pollution and meteorological parameters. PM10, SO2, NO, NO2, CO, O3, CH4, NMHC, temperature, rainfall, humidity, pressure, wind direction, wind speed and solar radiation were measured hourly for one year period in order to predict O3 concentrations of 1 h later. In the study, relationships between O3 data and other variables were investigated by bivariate correlation analysis. CH4, NMHC, NO2 exhibited considerable negative correlations with O3 described with the Pearson correlation coefficients of − 0.67, − 0.55, − 0.51, respectively whereas highest positive correlation was noted for temperature with correlation coefficient of 0.60. Multiple regression analysis (MLR) was used for modeling annual and seasonal O3 concentrations. Adjusted R2 values were determined as 0.90, 0.85 and 0.92 respectively for annual period, cooling and warming seasons. In order to decrease the number of input variables principle component analysis (PCA) was applied by using annual data. MLR analysis was repeated using four principle components and new adjusted R2 was calculated as 0.63.

Research highlights► We have used multivariate statistical methods for modeling tropospheric O3. ► Relations between O3 and other variables were examined by bivariate correlation analysis. ► Multiple regression analysis (MLR) was used for modeling O3 concentrations. ► Principle component analysis was applied in order to reduce the number of input variables. ► Using principle components in MLR analysis provided lower standard error value.