Using TROPOMI-Based Estimation of Daily Ozone Levels to Assess the Impact of COVID-19 on Ozone Concentrations in China

Abstract:

While China’s strict quarantine policy during the COVID-19 pandemic reduced the production of ozone precursors from transportation and industrial sectors, ground observations have reported an increase in ozone levels in many Chinese cities. However, few study has been able to evaluate the ozone level change in the entire country. We developed a machine learning model using tropospheric total ozone column from the TROPOspheric Monitoring Instrument (TROPOMI), ozone profiles from the Ozone Monitoring Instrument (OMI), as well as metrological and land use data to estimate full-coverage daily ground ozone concentrations in China at 0.05° spatial resolution. We built two separate models for the pandemic period (11/2019 – 04/2020) and the reference period (11/2018 – 04/2019), respectively. There were 209654 daily measurements from a total of 1500 AQS monitor during the study period. The out of bag R² (RMSE) was 86.7% (12.52) in 2018-2019 model and 90.06% (10.92) in 2019-2020 model. Scaled TROPOMI boundary layer ozone is a highly important predictor, together with wind fields and model-simulated VOCs, such as C2H6 and CO. In the 3^rd phase of Covid-19 (03/2020-04/2020) we defined as high level quarantine phase, a significant change of concentration of ozone took place comparing to the concentration in reference episode (March 2019 to Apr 2019) in China (95% CI: 2.44, 2.53; p < 0.0001). Our study demonstrate the possibility and utilization of TROPOMI product for modelling Ozone at a fine spatial and temporal resolution, which will allow us for construction of long-term daily Ozone measurements at 0.05° spatial resolution and thus support further epidemiological and environmental studies about ground Ozone.