The spatio-temporal variability of the COVID-19 lockdown impact on NO2 concentrations: A satellite-based modelling study

Keywords: air pollution, COVID-19, NO2, spatio-temporal variability, satellite data
Abstract Type: Paper Abstract

Abstract

Existing studies have indicated the positive effects of the lockdown measures on air quality in global cities. However, less attention has been paid to examining the multi-scale spatial variability of its impact. Moreover, empirical evidence is lacking as to how quickly the impact of the lockdown began and how it evolved in different cities. Here, using satellite data and surface measurements, we developed a regional-based random forest method to derive daily nitrogen dioxide (NO2) data at 1 km2 spatial resolution in the Beijing-Tianjin-Hebei (referred to as Jingjinji) region. By leveraging this high-resolution air pollution data, this study adopts a difference-in-difference strategy to answer three questions: 1) How fine-scale air data can help in reducing the uncertainty in the impact evaluation; 2) what is the spatial heterogeneity in NO2 dynamics during the COVID-19 pandemic at different spatial scales; 3) with the recovery of social and economic activities, how fast the air pollution bounce back to the pre-COVID-19 period level? Thus, we can evaluate the spatio-temporal variability of the COVID-19 lockdown impact on air quality in the Jingjinji region. We expect this study to provide the scientific basis for air pollution control at a multi-scale level and, more importantly, contribute significantly to promoting high-quality air monitoring and modelling in cities, benefiting the entire community.