Temporal spatial characteristics and key influencing factors of pm2.5 concentrations in china based on stirpat model and kuznets curve

Wenyang Huang1,2, Huiwen Wang1,3, Hongrui Zhao4, Yigang Wei1,2

1 School of Economics and Management, Beihang University, Beijing, 100191, China;
2 Beijing Key Laboratory of Emergency Support Simulation Technologies for City Operations, Beijing, 100191, China.
3 Beijing Advanced Innovation Center for Big Data and Brain Computing, Beijing, 100191, China
4 Department of Civil Engineer, Tsinghua University, Beijing, 10084, China 53S Center, Tsinghua University, Beijing, 10084, China


China's tremendous economic developments are achieved at the expense of environmental deterioration. In recent years, PM2.5 pollution has become increasingly serious in China, attracting widespread attention from the citizens and the government. This study aims to investigate the temporal spatial characteristics of PM2.5 in China in 1998 2012, fully considering the potential influencing factors of PM2.5. The satellite remote sensing data compensate for the deficiency of surface monitoring data in China, such as minimising bias from data contamination, broad coverage and representing long-term temporal spatial resolution data. A largely expanded list of potential impacting factors is selected based on environmental Kuznets curve (EKC) theory and Stochastic Impacts by Regression on Population, Affluence and Technology (STIRPAT) framework. Moran s I tests are used to examine the spatial correlation of PM2.5, and the pooled regression, spatial lag and time-fixed effects spatial lag models are used and compared to explore the influencing factors of PM2.5. The thresholded first-order inverse distance spatial weight matrix can measure the spatial spillover effect of PM2.5 more accurately by fully considering the effect of distance on spatial influence level. Several important findings are derived. Firstly, China s PM2.5 shows a distinct positive spatial correlation. The local Moran s I test shows that the significant high high PM2.5 agglomeration regions include Beijing Tianjin Hebei region, Yangtze River Delta and central China, connecting the two economic urban agglomerations. Secondly, the regression results of the time-fixed effects spatial lag model indicate that that PM2.5 of a given region increases by 0.362% if the PM2.5 of its ambient region increases by 1%. Thirdly, factors from all perspectives of STIRPAT model are very effective in explaining PM2.5. Fourthly, no inverted U shape EKC is found between the overall economic development level and the PM2.5 concentration in China in 1998-2012.


PM2.5; EKC; influencing factors; spatial econometric model; spatial autocorrelation; STIRPAT model

Full Text:

 Subscribers Only