Source apportionment of fine particulate matter in Macao, China with and without organic tracers: A comparative study using positive matrix factorization

Abstract Source apportionment of PM 2.5 was performed using positive matrix factorization (PMF) based on chemical composition data from 24-h filters collected throughout the year of 2015 at two sites in Macao (TG and PE), where TG is a general station and PE is near the city's ferry terminal. The input data sets include major inorganic ions, organic carbon and elemental carbon, elements and organic tracers (non-polar organic compounds and sugar compounds). Eight factors were resolved, namely, (1) secondary sulfate formation process, (2) secondary nitrate formation process, (3) biomass burning, (4) industrial emissions, (5) residual oil combustion, (6) sea salt, (7) dust, and (8) vehicle + helicopter emissions. Among the resolved sources, secondary sulfate and secondary nitrate were the major PM 2.5 contributors, contributing to 33% and 23% of PM 2.5 mass annually. Vehicle + helicopter emissions contributed to 12%, followed by industrial emissions (10%). Back trajectory analysis and conditional probability function (CPF) analysis were performed on the PMF-resolved factor contributions to distinguish the local and regional sources. Industrial emissions and vehicle + helicopter emissions showed different air mass impact at the two sites. The CPF plots confirmed the site differences in the resolved vehicle + helicopter emissions and industrial emissions were consistent with the geographical features of Macao and known local emission sources. The role of organic tracers was examined by comparing PMF runs with and without the inclusion of organics, denoted PMF org and PMF t , respectively. The results showed that the same set of contributing sources were resolved while noticeable differences of absolute PM 2.5 contributions were observed between PMF org and PMF t and lower mass closure was noted for PMF t . PMF runs without source-indicative organic tracers tend to produce higher estimations of source contributions from biomass burning and vehicle exhaust factors than that with the organics while underestimate secondary nitrate, industrial emissions, and dust. This work highlights the importance of organic tracers in achieving more accurate source apportionment.