Characteristics of secondary inorganic aerosols and contributions to PM2.5 pollution based on machine learning approach in Shandong Province

Tianshuai Li; Qingzhu Zhang; Xinfeng Wang; Yanbo Peng; Xu Guan; Jiangshan Mu; Lei Li; Jiaqi Chen; Haolin Wang; Qiao Wang

doi:10.1016/j.envpol.2023.122612

Characteristics of secondary inorganic aerosols and contributions to PM_2.5 pollution based on machine learning approach in Shandong Province

Environ Pollut. 2023 Nov 15:337:122612. doi: 10.1016/j.envpol.2023.122612. Epub 2023 Sep 25.

Authors

Tianshuai Li¹, Qingzhu Zhang², Xinfeng Wang¹, Yanbo Peng³, Xu Guan⁴, Jiangshan Mu¹, Lei Li¹, Jiaqi Chen¹, Haolin Wang¹, Qiao Wang¹

Affiliations

¹ Big Data Research Center for Ecology and Environment, Environment Research Institute, Shandong University, Qingdao, 266003, PR China.
² Big Data Research Center for Ecology and Environment, Environment Research Institute, Shandong University, Qingdao, 266003, PR China. Electronic address: zqz@sdu.edu.cn.
³ Big Data Research Center for Ecology and Environment, Environment Research Institute, Shandong University, Qingdao, 266003, PR China; Shandong Academy for Environmental Planning, Jinan, 250101, PR China.
⁴ Shandong Academy for Environmental Planning, Jinan, 250101, PR China.

PMID: 37757930
DOI: 10.1016/j.envpol.2023.122612

Abstract

Primary emissions of particulate matter and gaseous pollutants, such as SO₂ and NO_x have decreased in China following the implementation of a series of policies by the Chinese government to address air pollution. However, controlling secondary inorganic aerosol pollution requires attention. This study examined the characteristics of the secondary conversion of nitrate (NO₃^-) and sulfate (SO₄^2-) in three coastal cities of Shandong Province, namely Binzhou (BZ), Dongying (DY), and Weifang (WF), and an inland city, Jinan (JN), during December 2021. Furthermore, the Shapley Additive Explanation (SHAP), an interpretable attribution technique, was adopted to accurately calculate the contributions of secondary formations to PM_2.5. The nitrogen oxidation rate exhibited a significant dependence on the concentration of O₃. High humidity facilitates sulfur oxidation. Compared to BZ, DY, and WF, the secondary conversion of NO₃^- and SO₄^2- was more intense in JN. The light-gradient boosting model outperformed the random forest and extreme-gradient boosting models, achieving a mean R² value of 0.92. PM_2.5 pollution events in BZ, DY, and WF were primarily attributable to biomass burning, whereas pollution in Jinan was contributed by the secondary formation of NO₃^- and vehicle emissions. Machine learning and the SHAP interpretable attribution technique offer a precise analysis of the causes of air pollution, showing high potential for addressing environmental concerns.

Keywords: Machine learning; PM(2.5); SHAP; Secondary formation; Water-soluble ions.

MeSH terms

Aerosols / analysis
Air Pollutants* / analysis
Air Pollution* / analysis
China
Environmental Monitoring / methods
Particulate Matter / analysis
Seasons

Substances

Air Pollutants
Particulate Matter
Aerosols