Delhi has 37 continuous air quality monitoring stations that measure several pollutants out of which particulate matter is of the most concern. As several agencies are involved in setting up of these stations, there is no single design that is being followed to determine the sites which should be a part of this monitoring network. This study aims to estimate the redundancy in particulate matter measurements made across the monitoring stations in Delhi along with grouping areas that have a similar air quality through usage of multivariate analysis techniques of principal component analysis (PCA) and cluster analysis (CA). Also, through a set of indices (PC scores) the crucial hours and days in the years of 2018 and 2019 have been presented with regard to air quality of Delhi. It was found that two and three principal components only were being explained by PM10 and PM2.5 hourly concentration data respectively, and stations could be clustered based on the type of land use in the sites at which they lay along with their geographical location. The indices indicated that the peak traffic and early morning hours in the days of summer and pre-winter seasons along with the rainy season months had the maximum impact on the particulate matter concentrations in Delhi.
Keywords- monitoring stations, particulate matter, principal component analysis, cluster analysis, hourly concentrations.