Influence of local meteorological conditions on source contributions of PM2.5: A comparison of Conventional Positive Matrix Factorization (C-PMF) and Dispersion Normalized PMF (DN-PMF) models in Daejeon and Gwangju, South Korea.

PM_2.5 consists of  various chemical constituents originating from multiple sources and is associated with adverse health effects, including cardiovascular and respiratory diseases. Effective source specific management strategies are essential for mitigating these impacts. Positive Matrix Factorization (PMF) has been widely used as a receptor model to identify sources and quantify their contributions. However, conventional PMF (C-PMF) often overestimates or underestimates source contributions due to meteorological influences. To address this limitation, the Dispersion-Normalized PMF (DN-PMF) model has been introduced. This advanced approach accounts for meteorological conditions, providing more accurate source contributions.

In this study, hourly data for 28 chemical constituents of PM_2.5, measured from 2019 to 2022 at National Intensive Monitoring Stations (NIMS) in Daejeon and Gwangju, South Korea, were used as input data for both C-PMF and DN-PMF. The study aimed to identify sources whose contributions are significantly influenced by meteorological factors and to compare regional variations.  Ten sources were resolved by both models in each city, and differences in source contributions between the two approaches were calculated. Seasonal and temporal variations were also examined to determine meteorologically influenced sources and regional differences.

In Daejeon, a significant difference in secondary nitrate contributions was observed between the models, particularly during winter, when the atmospheric conditions favor its formation. In contrast, contributions of secondary sulfate showed minimal differences, suggesting it is primarily affected by long-range transport and less sensitive to local meteorological conditions.  In Gwangju, secondary nitrate and sulfate contributions showed relatively small differences, indicating lower sensitivity to local meteorological factors. Additionally, differences in contributions were observed for sources influenced by local emissions, highlighting variations between the two regions, for the same sources. These regional differences are likely attributable to the specific emission characteristics, meteorological conditions, and sources locations of each city. Supporting data, including emission inventories, meteorological parameters, and the Conditional bivariate probability function (CBPF) were used to explain the observed variations. This study underscores the influence of local meteorological conditions on source contributions and provide valuable insights for developing region-specific PM_2.5 management strategies.

Acknowledgement

This research was supported by “Study on the analysis of medium- and long-term factors affecting PM_2.5 emission changes” funded by National Air Emission Inventory and Research Center of the Ministry of Environment under grant, South Korea. This work was supported by the National Institute of Environmental Research (NIER) of the Ministry of Environment under grant, South Korea No. NIER-2021-03-03-001. This research was supported by Particulate Matter Management Specialized Graduate Program through the Korea Environmental Industry & Technology Institute (KEITI) funded by the Ministry of Environment (MOE).