Estimation of Particulate Matter (PM2.5) Over Kolkata

Particulate matter has a diverse range of effects on human health and climate due to which it has emerged as a key parameter in monitoring air quality. The current study explores and estimate the concentration of particulate matter (PM2.5) from MODIS AOD product over the city of Kolkata for a period of 3 years (2019–2021). PM2.5 concentration dataset was acquired from seven CPCB ground stations spread throughout Kolkata. Further, the study utilized the 1-km MODIS AOD product and meteorological parameters from MERRA-2. Considering the statistical analysis of data, four regression models were derived and considered for PM estimation. Daily estimated PM2.5 concentrations were compared against respective observations. The developed models were evaluated with the help of statistical methods. Model-2 based on the multi-linear regression equation was found to be the best fit model having a strong positive correlation between the estimated and observed PM2.5 values (R = 0.814). The root mean square error (RMSE) was estimated at 22.54 µg/m3. The estimated PM2.5 values were able to capture the trend of PM2.5 concentrations on the ground level. The normalized mean bias (NMB) value was − 0.315 and the mean absolute error was 18.94. The mean absolute percent error is estimated at around 5.16%. The results demonstrated that the developed model thus can be used to study the particulate matter concentration over areas where ground-based observation sites are sparse on the city level.