Key toxic components and sources affecting oxidative potential of atmospheric particulate matter using interpretable machine learning: Insights from fog episodes

Ruiyu Li; Caiqing Yan; Qingpeng Meng; Yang Yue; Wei Jiang; Lingxiao Yang; Yujiao Zhu; Likun Xue; Shaopeng Gao; Weijian Liu; Tianxing Chen; Jingjing Meng

doi:10.1016/j.jhazmat.2023.133175

Key toxic components and sources affecting oxidative potential of atmospheric particulate matter using interpretable machine learning: Insights from fog episodes

J Hazard Mater. 2024 Mar 5:465:133175. doi: 10.1016/j.jhazmat.2023.133175. Epub 2023 Dec 7.

Authors

Ruiyu Li¹, Caiqing Yan², Qingpeng Meng¹, Yang Yue³, Wei Jiang¹, Lingxiao Yang¹, Yujiao Zhu¹, Likun Xue¹, Shaopeng Gao⁴, Weijian Liu⁵, Tianxing Chen⁶, Jingjing Meng⁷

Affiliations

¹ Environment Research Institute, Shandong University, Qingdao 266237, China.
² Environment Research Institute, Shandong University, Qingdao 266237, China. Electronic address: cyan0325@sdu.edu.cn.
³ School of Environmental Science and Engineering, Shandong University, Qingdao 266237, China.
⁴ Institute of Tibetan Plateau Research, Chinese Academy of Sciences, Beijing 100101, China.
⁵ College of Environmental and Safety Engineering, Qingdao University of Science and Technology, Qingdao 266042, China.
⁶ College of Engineering, University of Washington, 1410 NE Campus Pkwy, Seattle, WA 98195, USA.
⁷ College of Environment and Planning, Liaocheng University, Liaocheng 252000, China.

PMID: 38086305
DOI: 10.1016/j.jhazmat.2023.133175

Abstract

Fog significantly affects the air quality and human health. To investigate the health effects and mechanisms of atmospheric fine particulate matter (PM_2.5) during fog episodes, PM_2.5 samples were collected from the coastal suburb of Qingdao during different seasons from 2021 to 2022, with the major chemical composition in PM_2.5 analyzed. The oxidative potential (OP) of PM_2.5 was determined using the dithiothreitol (DTT) method. A positive matrix factorization model was adopted for PM_2.5. Interpretable machine learning (IML) was used to reveal and quantify the key components and sources affecting OP. PM_2.5 exhibited higher oxidative toxicity during fog episodes. Water-soluble organic carbon (WSOC), NH₄⁺, K⁺, and water-soluble Fe positively affected the enhancement of DTT_V (volume-based DTT activity) during fog episodes. The IML analysis demonstrated that WSOC and K⁺ contributed significantly to DTT_V, with values of 0.31 ± 0.34 and 0.27 ± 0.22 nmol min^-1 m^-3, respectively. Regarding the sources, coal combustion and biomass burning contributed significantly to DTT_V (0.40 ± 0.38 and 0.39 ± 0.36 nmol min^-1 m^-3, respectively), indicating the significant influence of combustion-related sources on OP. This study provides new insights into the effects of PM_2.5 compositions and sources on OP by applying IML models.

Keywords: Chemical composition; Fine particulate matter; Machine learning; Positive matrix factorization; Reactive oxygen species.