Spatiotemporal variations and relationship of PM and gaseous pollutants based on gray correlation analysis

Yong Y Zhang,Ying Jia,Ming Li,Li A Hou
DOI: https://doi.org/10.1080/10934529.2017.1383122
2018-01-28
Abstract:The present work characterizes the spatiotemporal variations of air pollution at four sites in Xi'an city. The investigations lasted for 1 year: January 1 through December 31, 2015. The concentrations of CO, NO2, O3, PM and SO2 were systematically monitored. The gray correlation analysis was used to correlate PM2.5 and gaseous pollutants. The formation mechanisms of sulfate and nitrate aerosols were discussed. Results clearly revealed severe air pollution by PM2.5, PM10 and NO2 at all four sites. Significant monthly variations were observed for PM2.5, PM10, SO2 and CO with the maximum values in December/January and minimum values in June/July. The O3 level showed an opposite trend. Spatially, high variations were observed for PM and some gaseous pollutants at individual sites. Gray correlation analysis revealed the significance of various major influencing factors on seasonal PM2.5 values at each site. In general, PM10, SO2 and NO2 affect the PM2.5 more than other pollutants. For a more efficient control of air pollution, a systematic spatial characterization of the seasonal variability of individual influencing factors is necessary.
What problem does this paper attempt to address?