EUVIMEDEUVIMED entdecken
Crossref

Quantifying the drivers of ozone formation using localized photochemical reactivity and double machine learning

Mengdi Song, Yafei Liu, Xiaoning Ru, Hua Yang, Yu Qu, Kai Qin, Yuqing Qiu, Guoxian Zhang, Xingang Liu

Environmental Research · 2026-12 · 310 · 125841

DOI: 10.1016/j.envres.2026.125841 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Environmental Research, 2026-12. Verfasst von Mengdi Song und 8 weiteren Autor:innen. Beteiligte Einrichtung: School of Environment and Spatial Informatics, China University of Mining and Technology, Xuzhou, 221116, PR China. Hauptthemen laut MEDLINE: Ozone, Machine Learning, Air Pollutants, Environmental Monitoring. Die Arbeit nennt 47 Literaturangaben. Gefördert von: China Postdoctoral Science Foundation, National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Surface ozone (O3) pollution remains severe in many Chinese cities despite declining precursor emissions, reflecting nonlinear interactions among emissions, meteorology, chemistry, and transport. Here, we integrated hourly observations from Yuncheng in the Fenwei Plain during 2020-2022 with a localized observation-constrained box model to develop a photochemically constrained double machine learning (DML) framework. From 2020 to 2022, the annual mean O3 concentration in Yuncheng increased from 76.2 to 79.9μg m-3. Compared with clean days, O3-polluted days had stronger southeasterly winds, higher temperatures, lower humidity. Box-model calculations indicated that O3 formation in Yuncheng was VOC-limited during the pollution season. Using MCM v3.3.1, we derived a localized maximum incremental reactivity scale for Yuncheng (MIR-YC), with reactivity decreasing in the order isoprene > anthropogenic alkenes > aromatics > oxygenated VOCs > alkanes. The resulting localized ozone formation potentials were subsequently used in place of measured VOC concentrations to constrain the machine-learning analysis. The optimized random forest model achieved a test R2 of 0.923, and SHAP analysis attributed 44.6% and 26.4% of predictive importance to local precursors and photochemical drivers, respectively. DML then quantified conditional marginal responses beyond predictive attribution, associating a 1 °C temperature increase with a 4.04μg m-3 increase in O3 after adjustment for observed covariates. This integrated framework links localized chemical sensitivity with observation-based atmospheric responses, providing a quantitative basis for locally tailored VOC-NOx control strategies.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

OzoneMachine LearningAir PollutantsEnvironmental MonitoringChinaVolatile Organic CompoundsPhotochemical ProcessesCities

Stichwörter der Autor:innen

Double machine learning (DML)Localized box modelOzone formationVolatile organic compounds (VOCs)

Substanzen

OzoneAir PollutantsVolatile Organic Compounds

Autor:innen und Institutionen

  • School of Environment and Spatial Informatics, China University of Mining and Technology, Xuzhou, 221116, PR China.
  • State Key Laboratory of Water Environment Simulation, School of Environment, Beijing Normal University, Beijing, 100875, PR China; China CDC Key Laboratory of Environment and Population Health, National Institute of Env…
  • Yuncheng Municipal Ecological Environment Bureau, Yuncheng, 044000, PR China.
  • State Key Laboratory of Atmospheric Boundary Layer Physics and Atmospheric Chemistry, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, 100029, PR China.
  • School of Environment and Spatial Informatics, China University of Mining and Technology, Xuzhou, 221116, PR China. Electronic address: qinkai@cumt.edu.cn.
  • College of Environmental Sciences and Engineering, Peking University, Beijing, 100875, PR China.

Förderung

  • China Postdoctoral Science Foundation
  • National Natural Science Foundation of China

Zitieren

APA

Song, M., Liu, Y., Ru, X., Yang, H., Qu, Y., Qin, K., Qiu, Y., Zhang, G., & Liu, X. (2026). Quantifying the drivers of ozone formation using localized photochemical reactivity and double machine learning. Environmental Research, 310, 125841. https://doi.org/10.1016/j.envres.2026.125841

Vancouver

Song M, Liu Y, Ru X, Yang H, Qu Y, Qin K, et al. Quantifying the drivers of ozone formation using localized photochemical reactivity and double machine learning. Environmental Research. 2026;310:125841. doi:10.1016/j.envres.2026.125841

BibTeX
@article{Song2026,
  author = {Song, Mengdi and Liu, Yafei and Ru, Xiaoning and Yang, Hua and Qu, Yu and Qin, Kai and Qiu, Yuqing and Zhang, Guoxian and Liu, Xingang},
  title = {Quantifying the drivers of ozone formation using localized photochemical reactivity and double machine learning},
  journal = {Environmental Research},
  year = {2026},
  volume = {310},
  pages = {125841},
  doi = {10.1016/j.envres.2026.125841}
}
Kostenlos starten →