Leveraging PMF Time Series Characteristics from Multi-PAMS Measurements for NMHC Source Investigation and Ozone Formation Insights

Abstract. Ozone pollution is a persistent concern in Taiwan’s Kaoping region, where high industrial emissions contribute to poor air quality. Although non-methane hydrocarbons (NMHCs) are key ozone precursors, their sources and seasonal dynamics remain insufficiently resolved in this complex environment. This study aimed to characterize NMHC sources and quantify their contributions to ozone formation across seasons using high-resolution measurements from three Photochemical Assessment Monitoring Stations (PAMS) combined with Positive Matrix Factorization (PMF). Eight distinct source profiles were resolved, including petrochemical factors I & II, refinery, gasoline evaporation, mixed sources (vehicular/solvent), acetylene, aged air mass, and biogenic emissions. The model effectively captured source-specific signatures, notably the acetylene factor at Linyuan (R² = 0.99 with observations), serving as an intrinsic check on PMF performance.  Source contributions varied by site and season, with the petroleum industry as the dominant contributor (33–71 %), especially at Xiaogang, while aged air mass (12–30 %) and mixed sources (2–29 %) also played important roles. Despite petroleum dominance, highly reactive species in the mixed source factor drove higher ozone formation potential (OFP). Seasonal and pollution-level analyses revealed that even under moderate ozone conditions (MDA8 40–60 ppb), urban-industrial emissions remained significant contributors to OFP. These findings advance understanding of source-specific NMHC dynamics and highlight the value of multi-site, year-round monitoring for constraining ozone precursor sources. The results underscore the need to prioritize controls on petroleum and urban-industrial emissions to mitigate ozone in industrialized regions.