The driving mechanisms of industrial air pollution spatial correlation networks: A case study of 168 Chinese cities

Abstract

To enhance regional collaborative management and effectively address air pollution, this paper constructs the industrial air pollution network of 168 cities in China between 2010 and 2021. Social network analysis (SNA) and temporal exponential random graph model (TERGM) are used to investigate the structural characteristics and driving mechanisms of industrial air pollution spatial correlation networks. The results indicate the following: (1) Industrial air pollution levels in key Chinese cities showed a slight fluctuation between 2010 and 2015, but decreased after 2015. (2) The industrial air pollution linkage network exhibits small-world characteristics and path dependence, with different cities playing distinct roles in the network. (3) The evolution of the network is influenced by both exogenous and endogenous mechanisms. Local network structures such as reciprocity, connectivity, and transitivity make the network exhibit path-dependent characteristics. Cities’ attributes such as economic development level, green innovation capacity, and population density significantly influenced their status in the network and linkages between cities. Pollution linkages are more likely to occur in cities with similar levels of economic development, green technological innovation capacity, and institutional environments. Therefore, effective regional division for scientific collaborative governance should be considered from multiple perspectives. Based on these results, the paper provides several policy recommendations.