Estimating the Impact of COVID-19 on the PM2.5 Levels in China with a Satellite-Driven Machine Learning Model Open Access

Li, Qiulun (Spring 2021)

Permanent URL:


China implemented an aggressive nationwide lock down procedure immediately after the COVID-19 outbreak in January 2020. As China emerges from the impact of COVID-19 on national economic and industrial activities, it became the site of a large-scale natural experiment to evaluate the impact of COVID-19 on regional air quality. However, ground measurements of PM2.5 concentrations do not offer comprehensive spatial coverage especially in suburban and rural regions. In this study, we developed a machine learning method with satellite aerosol remote sensing data, meteorological fields and land use parameters as major predictor variables to estimate spatiotemporally resolved daily PM2.5 concentrations in China. Our study period consists of a reference semester (November 1, 2018 – April 30, 2019) and a pandemic semester (November 1, 2019 – April 30, 2020), with six modeling months in each semester. Each period was then divided into sub-period 1 (November and December), sub-period 2 (January and February) and sub-period 3 (March and April). The reference semester model obtained a 10-fold cross-validated R2 (RMSE) of 0.79 (17.55 μg/m3) and the pandemic semester model obtained a 10-fold cross validated R2 (RMSE) of 0.83 (13.48 μg/m3) for daily PM2.5 predictions. Our prediction results showed high PM2.5 concentrations in the North China Plain, Yangtze River Delta, Sichuan Basin and Xinjiang Autonomous Region during the reference semester. PM2.5 levels were lowered by 4.8 μg/m3 during the pandemic semester comparing to the reference semester and PM2.5 levels during sub-period 2 decreased most by 18%. The Southeast region was affected most by the COVID-19 outbreak with PM2.5 levels during sub-period 2 decreased by 31%, following by the Northern Yangtze River Delta (29%) and Pearl River Delta (24%).

Table of Contents

Table of Contents

1. Introduction 1

2. Materials and Methods 3

2.1. Study Area and Time Periods 3

2.2. Data 4

2.2.1. PM2.5 monitoring data 4

2.2.2. MAIAC AOD data 5

2.2.3. Meteorological parameters 5

2.2.4. Land use data 6

2.3. Data integration 6

2.4. Spatial cluster analysis 7

2.5. PM2.5 modeling 8

3. Results 10

3.1. Descriptive statistics 10

3.2. model performance and variable importance 10

3.3. PM2.5 predictions 12

4. Discussion 14

5. Conclusions 19

6. References 19

7. Supplementary Materials 24

About this Master's Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
  • English
Research Field
Committee Chair / Thesis Advisor
Last modified

Primary PDF

Supplemental Files