Effect of air pollution on visibility in urban conditions. Warsaw case study

Abstract

The influence of air pollutants on visibility in Warsaw Agglomeration has been investigated. Following pollutants were considered: PM10, SO2, NO2 and O3, while meteorological parameters included: air temperatures (mean, minimum, maximum), solar radiation, relative air humidity, rainfall rates and wind speed. Initial analyses were performed with the use of principal component analysis (PCA). In next stages, the logistic regression (LR), the analysis of variance (ANOVA), one-way classification and a model path of generalized regression models (GRM) were applied. PCA analysis showed that in the cold season the visibility index depends on PM10, SO2, NO2 and the temperatures: T, Tmin, and Tmax. In the warm season, the index of visibility is mostly shaped by four elements: O3, T, Tmax and solar radiation. Logistic regression model indicated that in the warm season only two variables are significantly related to visibility: PM10 and relative humidity of air. Regularities in the cold season shown by the LR correspond with the conclusions from the PCA. Among meteorological conditions, the most important is air temperature, but only Tmax preserves the same direction of influence as the one pointed by the PCA model.