Estimation of PM10 concentration from Landsat 8 OLI satellite imagery over Delhi, India

Air Pollution in India is a major issue with the majority of Indian cities having very poor air quality and reeling under severe particulate pollution. Comprehensive monitoring of air pollution is necessary to counteract this great environmental health hazard. The aim of this study is to develop a multispectral empirical model to estimate the concentration of particulate matter less than ten micrometres (PM10) using Landsat 8 Operational Land Imager (OLI) satellite data over Delhi, India. PM10 concentration data provided by the Central Pollution Control Board from the ground monitoring stations in New Delhi was obtained corresponding to the acquisition date of Landsat 8 satellite data. Atmospheric reflectance in the visible bands of Landsat 8 imagery was calculated and correlated with the PM10 measurements from ground stations. The feasibility of the proposed algorithm was analysed based on the correlation coefficient, and the root mean square error value. The results show that the suggested multispectral PM10 model can predict particulate matter concentrations with an acceptable level of accuracy.