Dynamic estimation of air pollutants

Abstract

Quantitative analysis based on a measure of central tendencies and variability of air pollutants is carried out. The pollutant data is skewed at all sites. The large value of standard deviation emphasis the complexity involved in modeling the data. The trend percentage of 8-hourly and 24-hourly concentration of pollutants and meteorological parameters are compared during the road rationing scheme in Delhi. The influence of the scheme is being analyzed at Dwarka for its two phases- winter (I) and summer (II). The findings suggest that phase I is efficacious than phase II in curbing pollution. It is concluded that during the winter phase, there was an increase of 24% in PM2.5 and 26% in NOx. However, the increase was 38% in PM2.5 and 23% in NOx during phase II. The trend analysis depicts that there was a significant difference between the 8 hour and 24-hour trend percentage which may contribute to an error for the prediction of pollutants. Meteorological conditions and environmental effects were the reported causes for the air quality rise. Thus in the long run, apart from vehicular emissions, other significant factors are contributing to the rise of air pollution. Chaos theory is used further to study the behavior of air pollutants and depict the applicability of models proposed for air pollutants. Phase Space Reconstruction allows the computation of the entire system of pollutant time series. Positive largest Lyapunov exponents, low correlation dimensions, and values of correlation coefficient close to 1 in 0-1 chaos test prove quantitatively that pollutants are chaotic. Recurrence Plot graphically depicts the complexity of pollutant dynamics. The strong nonlinear characteristics are determined by the meteorological environment, the source of the pollutants, the underlying dynamic surface, and the combination of physical and chemical processes. The uncertainty of pollutants makes precise predictions of air quality a challenging task. The time-space analysis is used to extract the features of pollutants