Identifying the spatiotemporal dynamics of PM2.5 concentration and its implications for national sustainable development experimental zone of China

Abstract

The accelerated pace of urbanization and industrialization in China has given rise to air pollution in the form of PM2.5. This pollution poses significant threats to the atmospheric environment and human health. Traditional statistical models often lack the required data precision for medium or small-scale epidemiological and pollutant exposure studies. Therefore, alternative approaches must be developed. This study employs the Random Forest (RF) model, utilizing measured PM2.5 concentration and auxiliary data, to simulate PM2.5 concentration with 1 km spatial resolution for 2015–2019. The results showed that: (1) The RF model was sufficiently accurate with a 10-fold CV, resulting in a coefficient of determination (R2) of 0.75, a root-mean-square error (RMSE) measuring at 13.24 μg/m3, and a mean absolute error (MAE) of 9.12 μg/m3 (2) A notable variation and dynamic pattern in the concentration of PM2.5 were observed. The geographical distribution displayed elevated levels in the northern regions and subdued levels in the southern regions, with the most elevated PM2.5 values recorded in Xinjiang and Northern China. (3) Pollution levels in the five major urban agglomerations, ranked from high to low, were as follows: the Beijing-Tianjin-Hebei (BTH), the Guanzhong Plain (GZP), the Yangtze River Delta (YRD), the Chengdu-Chongqing (CY), and the Pearl River Delta (PRD). (4) The PM2.5 concentration of the whole country generally showed a downward trend. These findings offer valuable scientific insights to support atmospheric environmental protection and epidemiological research endeavors. Finally, policy implications for the coordinated development of the economy and environment in the national sustainable development experimental zone major urban agglomerations were proposed to achieve more equitable and balanced development. This paper provides policy recommendations and empirical evidence for further promoting environmental balance in national sustainable development experimental zones.