Estimation of Near-Surface Ozone Concentration Across China and Its Spatiotemporal Variations During the COVID-19 Pandemic

China has made remarkable progress in controlling particulate matter, while O
₃
pollution over China has become increasingly severe in recent years according to ground observations. Continuous monitoring of dynamic changes in O
₃
concentrations on regional and national scales can provide valuable insights for pollution control policies. Therefore, an improved similarity distance-based space-time random forest (SDSTRF) model was developed to estimate the near-surface O
₃
concentration using the surface measurements, satellite O
₃
precursors, meteorological variables, and other auxiliary information. The O
₃
concentration data over China were generated based on the developed model with a spatial resolution of 10 km and a temporal resolution of 1 day from 2016 to 2022. The validation results against the ground measurements indicate that the developed SDSTRF model effectively captures O
₃
variations, achieving a coefficient of determination of 0.83 and a root mean square error of 20.37 μg/m
³
. The spatiotemporal variations of O
₃
concentrations were investigated using the generated dataset. A significant increasing trend of 1.243 μg/m
³
/yr in O
₃
concentrations was observed in eastern China during the COVID-19 pandemic, which was attributed to changes in NO
_x
concentrations. In this study, the possible reasons for the increase in O
₃
concentrations are also discussed. Overall, the improved SDSTRF model and the comprehensive analysis of the spatiotemporal variations of near-surface O
₃
will significantly contribute to achieving clean air in China.