Source Apportionment of Volatile Organic Compounds with Initial Concentrations of Continuous Speciation Monitoring Data and Time-integrated Measurements of Organic Markers

YI-HSIEN LIU, Kuan-Lin Lai, Chia-Yang Chen, Chang-Fu Wu, National Taiwan University

     Abstract Number: 179
     Working Group: Source Apportionment

Abstract
Volatile organic compounds (VOCs) contribute to the formation of secondary organic aerosols (SOAs) and ozone, both of which are harmful to human health and the environment. Positive Matrix Factorization (PMF) is widely applied for VOC source apportionment; however, conventional PMF may overlook variations resulting from photochemical aging. To address this, some studies have incorporated photochemical parameterization to account for chemical transformations. Nevertheless, overlapping chemical profiles continue to hinder accurate source identification. While the inclusion of organic compounds (OCs) has improved PM₂.₅ source apportionment, limited research has integrated VOCs with organic molecular tracers. This study aims to enhance PMF analysis by incorporating initial VOC concentrations and selected organic tracers.

Hourly VOC monitoring and 12-hour integrated filter sampling were conducted during a 21-day period in summer at an industrial complex in southern Taiwan. A three-stage PMF modeling approach was applied, with the first stage focusing on VOC analysis; the second stage using initial concentration of VOCs, and third stage incorporating both initial VOC and OC markers. In Stage 2, seven factors were identified, with the three highest contributors attributed to Petrochemical Emissions, Synthetic Resin and Paints, and Industrial Emissions. In Stage 3, after incorporating OC data, additional sources were identified, including fungal sources, cooking sources and biomass burning, offering a more comprehensive source apportionment. These findings demonstrate that incorporating organic species can bridge gaps in VOC source apportionment, enhancing the resolution and accuracy of pollution source identification.