Abstract
Since the Seoul metropolitan area is a highly developed megacity, many people are often exposed to high concentrations of particulate matter (PM), with mean aerodynamic diameters equal to or less than 10 μm (PM 10 ), in cold seasons. PM 10 concentrations can be influenced by a combination of various factors, including meteorological conditions, anthropogenic emissions, atmospheric chemical reactions, transboundary transport, and geographic characteristics. However, the establishment of an efficient air quality management plan remains challenging due to the limited understanding of the regional PM concentration characteristics. Here, the Seoul metropolitan (Seoul, Gyeonggi-do, and Incheon) and Chungcheongnam-do (Chungcheongnam-do, Daejeon, and Sejong) areas were regionally classified to identify the spatiotemporal air quality in areas where megacities and emission sources are mixed. The four representative regions were determined using the K-means clustering method based on the temporal variations in the observed PM 10 concentrations. The first cluster consisted of small cities in the southern and eastern parts of Gyeonggi-do and Chungcheongnam-do, respectively, while the second cluster consisted of Incheon, West Gyeonggi-do, and Seoul. In addition, the third and fourth clusters included West Chungcheongnam-do and East Gyeonggi-do, which are adjacent to the Yellow Sea and downstream area of the westerly wind, respectively. The characteristics of each cluster during the high PM 10 concentration events are explainable by wind patterns and the local air pollutant emissions, including nitrogen oxides and sulfur oxides. The obtained regional classification was different from the provincial-level administrative division of South Korea. Therefore, the present study is expected to be a scientific basis for overcoming the limitations of air quality management in administrative districts.