Source apportionment of PM_2.5 size distribution and composition data from multiple stationary sites using a mobile platform

Several source apportionment studies considering multiple sites showed spatial variability of source contributions. However, setting up multiple fixed sites to collect comprehensive chemical speciation data is resource demanding. In this study, field campaigns were conducted at multiple receptor sites in the Mailiao and Taishi townships in Yunlin County using a mobile platform to demonstrate the feasibility of receptor modeling with particle size distribution and PM_2.5 speciation data. Sources of air pollutants to all of the monitoring sites were identified and quantified using a modified positive matrix factorization (PMF) model. Modeling results indicated that a mixed source dominated by secondary aerosol was the largest contributor to PM_2.5 at most sites. Adding VOC measurements with high time resolution helped to improve the source separation. Different patterns of source contributions among sites and seasons were observed showing both spatial heterogeneity and seasonal variation.