Source apportionment of particulate matter based on numerical simulation during a severe pollution period in Tangshan, North China

Facing serious air pollution problems, the Chinese government has taken numerous measures to prevent and control air pollution. Understanding the sources of pollutants is crucial to the prevention of air pollution. Using numerical simulation method, this study analysed the contributions of the total local emissions and local emissions from different sectors (such as industrial, traffic, resident, agricultural, and power plant emissions) to PM2.5 concentration, backward trajectory, and potential source regions in Tangshan, a typical heavy industrial city in north China. The impact of multi-scale meteorological conditions on source apportionment was investigated. From October 2016 to March 2017, total local emissions accounted for 46.0% of the near-surface PM2.5 concentration. In terms of emissions from different sectors, local industrial emissions which accounted for 23.1% of the near-surface PM2.5 concentration in Tangshan, were the most important pollutant source. Agricultural emissions were the second most important source, accounting for 10.3% of the near-surface PM2.5 concentration. The contributions of emissions from power plants, traffic, residential sources were 2.0%, 3.0%, and 7.2%, respectively. The contributions of total local emissions and emissions from different sectors depended on multi-scale meteorological conditions, and static weather significantly enhanced the contribution of regional transport to the near-surface PM2.5 concentration. Eight cluster backward trajectories were identified for Tangshan. The PM2.5 concentration for the 8 cluster trajectories significantly differed. The near-surface PM2.5 in urban Tangshan (receptor point) was mainly from the local emissions, and another important potential source regionwas Tianjin. The results of the source apportionment suggested the importance of joint prevention and control of air pollution in some areas where cities or industrial regions are densely distributed. (C) 2020 Elsevier Ltd. All rights reserved.