Ambient PM2.5, well-known for its adverse impacts on human health, is a very heterogeneous pollutant. Its chemical composition and attributable sources vary by region, influenced by meteorological and geographical conditions as well as emission sources. However, administrative policies are currently focused on mass concentrations. However, not all PM2.5 sources provide equally toxic particles. Thus, those sources that should be the focus of controls has not been the priority. In the present study, we conducted source apportionment utilizing positive matrix factorization (PMF) and investigated the association of PM2.5 source contributions with emergency department visits (EDVs) in major megacities in South Korea. Overall, an interquartile range (IQR) increment in source contribution increased the number of emergency room visits. Industry and coal combustion sources, marked by heavy metals, were principally associated with the adverse health impacts. However, the sources showing significant associations with EDVs differed across the study area. In addition, we found that region-specific relationships between PM2.5 sources and morbidity were plausible, considering the existence of relevant sources such as industrial complexes and coal-fired power plants. The analysis of source contributions according to wind conditions also supported the source-morbidity relationships. These findings suggest that administrative policies for PM2.5 control should be established and implemented considering region-specific characteristics of the links between PM2.5 sources and health impacts to maximize the control's public health effects. Furthermore, the results of the present study indicate that PMF was an effective method for linking acute exposure to PM2.5 source types with health outcomes to prioritize its sources.
Keywords: Emergency department visit (EDVs); General linear models (GLM); Heterogeneity; PM(2.5); Positive matrix factorization (PMF).
Copyright © 2024. Published by Elsevier Inc.