Study on Spatial Changes in PM2.5 before and after the COVID-19 Pandemic in Southwest China

Abstract

Coronavirus disease 2019 (COVID-19) swept the world at the beginning of 2020, and strict activity control measures were adopted in China’s concentrated and local outbreak areas, which led to social shutdown. This study was conducted in southwest China from 2019 to 2021, and was divided into the year before COVID-19 (2019), the year of COVID-19 outbreak (2020), and the year of normalization of COVID-19 prevention and control (2021). A geographically and temporally weighted regression (GTWR) model was used to invert the spatial distribution of PM2.5 by combining PM2.5 on-site monitoring data and related driving factors. At the same time, a multiple linear regression (MLR) model was constructed for comparison with the GTWR model. The results showed that: (1) The inversion accuracy of the GTWR model was higher than that of the MLR model. In comparison with the commonly used PM2.5 datasets “CHAP” and “ACAG”, PM2.5 inverted by the GTWR model had higher data accuracy in southwest China. (2) The average PM2.5 concentrations in the entire southwest region were 32.1, 26.5, and 28.6 μg/m3 over the three years, indicating that the society stopped production and work and the atmospheric PM2.5 concentration reduced when the pandemic control was highest in 2020. (3) The winter and spring of 2020 were the relatively strict periods for pandemic control when the PM2.5 concentration showed the most significant drop. In the same period of 2021, the degree of control was weakened, and the PM2.5 concentration showed an upward trend.