Meteorological normalization of surface ozone variability across warm and cold seasons in China during 2015–2024

Surface ozone (O ₃ ) concentrations in China have continued to increase, but the roles of meteorological variability and longer-term changes remain uncertain, particularly across seasons and regions. Here, we combined LightGBM-based meteorological normalization with SHapley Additive exPlanations (SHAP) to investigate warm- and cold-season variations in maximum daily 8 h average O ₃ (MDA8  O ₃ ) across China during 2015–2024. The models reproduced daily MDA8  O ₃  variability reasonably well, with mean testing R ²  values of 0.62 and 0.68 in the warm and cold seasons, respectively. Observed national mean MDA8  O ₃ increased by 2.0 and 2.2 µgm-3yr-1 in the two seasons, respectively. After meteorological normalization, the corresponding trends were 1.1 and 1.2 µgm-3yr-1, indicating that meteorological variability alone cannot fully explain the observed increases. The normalized long-term component may reflect changes in precursor emissions, background O ₃ , regional transport, and other processes not explicitly represented in the model. SHAP analysis revealed distinct seasonal and regional meteorological associations. During the warm season, temperature and solar radiation were more important in northern inland and basin regions, whereas relative humidity and wind were more important in southern coastal regions. During the cold season, solar radiation dominated across most regions, while relative humidity was more important in the Pearl River Delta. Meteorological contributions generally shifted from negative during 2015–2019 to positive after 2019, amplifying O ₃  pollution in several regions. These findings highlight the importance of accounting for seasonal meteorological variability when evaluating long-term O ₃  trends in China.