Prediction and spatio – Temporal analysis of ozone concentration in a metropolitan area

Abstract

Abstract Increases in the concentrated amounts of ozone (O3) on the earth’s surface have become a major matter of concern for human health and may have dire consequences. The purpose of this study is to predict the ozone concentration and demonstrate its spatio-temporal changes in Tehran. A large amount on meteorology, concentration of other pollutants, traffic information, industrial agglomeration and remote sensing data were used for this purpose. Study parameters on ozone concentrations were initially analyzed in order to find meaningful relationships between the study parameters and ozone concentrations. Ozone concentration was then predicted using multivariate linear regression and spatio-temporal changes of ozone concentration were finally investigated. Results indicated that the proposed regression model was able to account for 89 percent of changes in 2015. Spatio-temporal analysis showed that the best performance of the model, in regards to ozone concentration, occurred in autumn (R2 = 0.84) at Park Rose (R2 = 0.89). Furthermore, the spatial distribution of estimated ozone concentrations at the park were consistent with the actual amounts observed. Finally, in accordance with spatial dispersion of ozone concentrations throughout the 22 districts of Tehran, areas 8, 13, and 20, located in the eastern and southern parts of Tehran were indicated as areas with the highest ozone concentration.