Source apportionment of fine particles based on combined numerical model and receptor model

Abstract

In recent years, atmospheric particulate matter has become the primary air pollutant in most cities of China. In order to support an efficient control for particles reduction, it is of great importance to investigate the contribution of different sources to particulate matter. Source apportionment has been a conventional technique for seeking the emission sources of particulate matter. There are many ways to investigate the source of particles, such as receptor models, emission inventories, trajectory analysis, dispersion models, photochemical models and source models. Receptor models were shown to be an effective tool for source apportionment. As one of the most popular receptor models, the Positive Matrix Factorization (PMF) model estimates the sources contribution rate based on chemical characteristics of particulate samples. Traditionally, apportioning various sources of particulate matter is mainly through offline chemical composition analysis in combination with measurements in receptor regions. This method is limited by observational periods and locations and is applicable only for historical events. So researches on the source apportionment of PM2.5 with high spatial and temporal resolution in urban scale would be of great significance to control air pollution scientifically and improve urban air quality. The aim of this paper is to develop a source apportionment method of fine particulate matter based on a combination of numerical air quality model and receptor model. This method will simulate the chemical components of fine particulate matter accurately and evaluate the source contributions quantitatively at the same time. In this study, a new method is developed based on a numerical model (RegAEMS model) and a receptor model (PMF) to enhance the temporal and spatial resolutions of apportioning particulate matter sources. This method is applied to the period during the Youth Olympic Games (YOG) in Nanjing for apportioning fine particulate matter sources. With RegAEMS, the concentrations of PM2.5 and its main chemical compositions are simulated. We find that the simulations agree well with the results from the offline chemical composition analysis. Using PMF model, the components of fine particulate matter during the 2014 YOG (July to September) are identified, which include secondary organic aerosols (25.9%), combustion dust (16.5%), secondary sulfate aerosols (14.5%), secondary nitrate aerosols (12.6%), vehicle exhaust (12.0%), dusts (11.7%), and industrial activities (6.9%). Such results are in good agreement with sampling measurements. This study suggests that the contributions from combustion dust and industrial activities are lower during YOG than before and after YOG, indicating that the control measures of industrial activities during YOG are effective in reducing air pollution. This study suggests that integrated numerical modeling and receptor modeling can effectively assess contributions of different pollution sources, and thus be useful to source apportionments of heavy pollution events, providing scientific basis for emergency controls of air pollution. This finding will be useful for the local government to create efficient control strategies to reduce emissions of different sources of PM2.5.