Multifractal Characteristics of Criteria Air Pollutant Time Series in Urban Areas

The complexity of spatial-temporal air pollutant concentration dynamics requires innovative modeling investigation approaches. The details of non-linear nature of pollutant behavior cannot be revealed by conventional approaches, but fractal and Hurst rescaled analyses allow the quantification of pollutant dynamics structure via self-similarity and scale invariance. In this study, we applied multiscale multifractal analysis (MMA) to investigate the complex time-series of criteria air pollutants (PM10, PM2.5, NOX, SO2, CO and O3). The results showed that PM10 and PM2.5 concentrations are more stable compared to gaseous oxides and exhibit less prominent multifractality. Out of gaseous contaminants, CO is confirmed to be less chemically reactive than NO, NO2, NOx, SO2 and O3 under the same atmospheric conditions in urban and semi-urban area. As concluded, the multifractal analysis presented herein can enhance our understanding of specific pollutant dynamic features and support relevant sectors to control the pollutant release and distribution.