Deciphering the impacts of meteorology on surface ozone variability in eastern China using explainable machine learning models
4 affiliations
- 1Centre for Atmospheric Science, Yusuf Hamied Department of Chemistry, University of Cambridge, Cambridge, United Kingdom
- 2Laboratory for Climate and Ocean-Atmosphere Studies, Department of Atmospheric and Oceanic Sciences, School of Physics, Peking University, Beijing, China
- 3Peking University (Beijing, China)
- 4National Centre for Atmospheric Science, University of Cambridge, Cambridge, United Kingdom
Abstract
Understanding how meteorology influences surface ozone variability is critical for interpreting trends and designing effective air quality policies. This study employs explainable machine learning (XML) with SHapley Additive exPlanations (SHAP) to interpret daily ozone variations from 2013 to 2023 across three major regions in eastern China: North China Plain (NCP), Yangtze River Delta (YRD), and Pearl River Delta (PRD). An ensemble of five machine learning models (LightGBM, XGBoost, CatBoost, Random Forest, and Extra Trees) is trained using 14 meteorological variables and two temporal indicators. XML reveals nonlinear, region-specific ozone-meteorology relationships that are broadly consistent with physical understanding, while differences in SHAP attributions across algorithms highlight structural uncertainty arising from multicollinearity among input variables. We use SHAP-derived contributions to attribute warm-season ozone trends to meteorological versus non-meteorological drivers. Before 2019, ozone increases are mainly associated with the temporal proxy for non-meteorological influences (e.g., emission changes), whereas after 2019 meteorological variability dominates regional ozone trends. Exploiting the additive nature of SHAP, we develop a de-weathering framework that partitions daily ozone into a SHAP-based climatological baseline and a meteorology-induced ozone anomaly (MOA). Across all three regions, the magnitude of positive MOA events increases over 2013–2023, while their frequency and duration show no significant trends, indicating a strengthening meteorological amplification of pollution episodes rather than more frequent events. Our results demonstrate both the utility and limitations of XML for disentangling meteorological drivers of ozone pollution and provide new constraints on how meteorology shapes surface ozone under China’s clean air actions.
Related
- Tropical belt expansion and its impacts on climate variability in the Mediterranean from 1980 to 2022
- Meteorology shapes urban particle nucleation, while pollution-related factors modulate growth
- Advancing Near-Term Water Quality Forecasting with Explainable Machine Learning and Probabilistic Uncertainty Quantification
- Meteorology-driven Causal Nowcasting of Fugitive Landfill Emissions from Measured Coupling Timescales
- AirMPA: A Meteorology-to-Pollution Adapter for Global Air Quality Forecasting
BibTeX
@article{ye2026deciphering,
title = {{Deciphering the impacts of meteorology on surface ozone variability in eastern China using explainable machine learning models}},
author = {Xingpei Ye and Lin Zhang and Xiaolin Wang and Ni Lu and Sebastian H. M. Hickman and Guo Luo and Alexander Thomas Archibald},
journal = {Atmospheric chemistry and physics},
year = {2026},
month = may,
doi = {10.5194/acp-26-6377-2026},
url = {https://doi.org/10.5194/acp-26-6377-2026},
note = {CC BY},
}
RIS
TY - JOUR
TI - Deciphering the impacts of meteorology on surface ozone variability in eastern China using explainable machine learning models
AU - Xingpei Ye
AU - Lin Zhang
AU - Xiaolin Wang
AU - Ni Lu
AU - Sebastian H. M. Hickman
AU - Guo Luo
AU - Alexander Thomas Archibald
PY - 2026
DA - 2026/05/12
JO - Atmospheric chemistry and physics
DO - 10.5194/acp-26-6377-2026
UR - https://doi.org/10.5194/acp-26-6377-2026
AB - Understanding how meteorology influences surface ozone variability is critical for interpreting trends and designing effective air quality policies. This study employs explainable machine learning (XML) with SHapley Additive exPlanations (SHAP) to interpret daily ozone variations from 2013 to 2023 across three major regions in eastern China: North China Plain (NCP), Yangtze River Delta (YRD), and Pearl River Delta (PRD). An ensemble of five machine learning models (LightGBM, XGBoost, CatBoost, Random Forest, and Extra Trees) is trained using 14 meteorological variables and two temporal indicators. XML reveals nonlinear, region-specific ozone-meteorology relationships that are broadly consistent with physical understanding, while differences in SHAP attributions across algorithms highlight structural uncertainty arising from multicollinearity among input variables. We use SHAP-derived contributions to attribute warm-season ozone trends to meteorological versus non-meteorological drivers. Before 2019, ozone increases are mainly associated with the temporal proxy for non-meteorological influences (e.g., emission changes), whereas after 2019 meteorological variability dominates regional ozone trends. Exploiting the additive nature of SHAP, we develop a de-weathering framework that partitions daily ozone into a SHAP-based climatological baseline and a meteorology-induced ozone anomaly (MOA). Across all three regions, the magnitude of positive MOA events increases over 2013–2023, while their frequency and duration show no significant trends, indicating a strengthening meteorological amplification of pollution episodes rather than more frequent events. Our results demonstrate both the utility and limitations of XML for disentangling meteorological drivers of ozone pollution and provide new constraints on how meteorology shapes surface ozone under China’s clean air actions.
SN - 1680-7316
SN - 1680-7324
ER -