Mapping nationwide concentrations of sulfate and nitrate in ambient PM2.5 in South Korea using machine learning with ground observation data

Abstract

Particulate matter (PM) is a major air pollutant in Northeast Asia, with frequent high PM episodes. To investigate the nationwide spatial distribution maps of PM_2.5 and secondary inorganic aerosols in South Korea, prediction models for mapping SO₄^2- and NO₃^- concentrations in PM_2.5 were developed using machine learning with ground-based observation data. Specifically, the random forest algorithm was used in this study to predict the SO₄^2- and NO₃^- concentrations at 548 air quality monitoring stations located within the representative radii of eight intensive air quality monitoring stations. The average concentrations of PM_2.5, SO₄^2-, and NO₃^- across the entire nation were 17.2 ± 2.8, 3.0 ± 0.6, and 3.4 ± 1.2 μg/m³, respectively. The spatial distributions of SO₄^2- and NO₃^- concentrations in 2021 revealed elevated concentrations in both the western and central regions of South Korea. This result suggests that SO₄^2- concentrations were primarily influenced by industrial activities rather than vehicle emissions, whereas NO₃^- concentrations were more associated with vehicle emissions. During a high PM_2.5 event (November 19-21, 2021), the concentration of SO₄^2- was primarily influenced by SO_X emissions from China, while the concentration of NO₃^- was affected by NO_X emissions from both China and Korea. The methodology developed in this study can be used to explore the chemical characteristics of PM_2.5 with high spatiotemporal resolution. It can also provide valuable insights for the nationwide mitigation of secondary PM_2.5 pollution.

Keywords: Machine learning; Nitrate; PM(2.5); Secondary inorganic ions; Sulfate.