首页 > , Vol. , Issue () : -
With the successful implementation of the "Air Pollution Prevention and Control Action Plan"(2013-2017) and “Three-year Action Plan to Win the Blue Sky Defense War” (2018-2020), the concentrations of five major pollutants (i.e., PM2.5, PM10, SO2, NO2 and CO), except ozone, significantly dropped for most cities in China. The increasing ground-level ozone concentrations has been a key factor restricting the improvement in the ambient air quality, especially in summer. Compared with the measurements from ground-based monitoring sites, satellite remote sensing technology can obtain spatially continuous total column ozone. However, as ozone is abundantly distributed in the stratosphere, the ground-level ozone has a very low contribution to the total column ozone observed from space, Therefore, the information provided by the satellite total column ozone product is limited for estimating ground-level ozone concentrations. In this study, combined with TROPOMI ozone precursor (NO2 and HCHO) products, ERA5 meteorological parameters and ground-based monitoring data, a machine learning model was developed to estimate daily maximum 8-hour average ground-level ozone concentration over China during the period of 2019 - 2020. By comparing the performance of three ensemble learning methods, i.e., Extreme Gradient Boosters (XGBoost), Extreme Random Trees (ERT) and Gradient Boost Regression Tree (GBRT), the averaged overall 10-fold cross-validation R2 of 2019 and 2020 are all larger than 0.89. Although the estimated results by XGBoost showed the best agreement between model predictions and observations with an averaged RMSE and MAE of 15.77μg/m3 and 10.53μg/m3, respectively, ERT method was finally selected to model the estimation of daily maximum 8-hour average ground-level ozone concentration by considering the rationalization of spatial distribution. Due to the implementation of proactive emission reduction measures carried out by Chinese government, as well as the impact of the COVID-19 epidemic, the rising trend of ozone concentration over the years has been reversed. The annual average value of ground-level ozone concentration in 2020 reached 107.41±18.6μg/m3 over China, which is 1.85μg/m3 less than that 2019(109.26±19.71μg/m3). Sever surface ozone pollution events frequently occur from May to September every year because the high temperature can further promote photochemical reactions. The estimated ground-level ozone concentrations of Beijing-Tianjin-Hebei, Yangtze River Delta, Pearl River Delta and Chengdu-Chongqing regions are significantly higher than their surrounding areas, which are the key areas for ozone pollution prevention and control.