DYNAMIC COMPLEX NETWORK ANALYSIS OF PM2.5 IN HENAN PROVINCE OF CHINA

At present, air pollution has become a major environmental problem threatening human health. PM2.5 concentration is an important indicator to measure air pollution. Studying the distribution and interaction of PM2.5 concentration between cities can provide a scientific basis for air quality monitoring, air pollution control, and the formulation of collaborative strategy for economy and environment in Henan Province of China. According to the PM2.5 concentration data of each prefecture-level city in 2018, we analyze the correlation of PM2.5 concentration between cities in Henan Province of China. Further, we construct a directed complex network of PM2.5 interaction based on Granger causality to explore the directivity of the impact between cities in Henan Province of China. Then, we introduce the "trophic coherence" method in biology to infer the hierarchical structure and stability of the network. The research indicates: (1) there are the evident of seasonal differences in PM2.5 concentration in Henan Province of China. The mean of PM2.5 concentration in the four seasons shows different trends, and there is the relatively obvious holiday effect. (2) In different seasons, the cross-correlation of PM2.5 concentration between cities is different. The cross-correlation between cities in spring and summer shows obvious spatial heterogeneity, and PM2.5 concentration between cities in autumn and winter shows higher spatial embeddedness. (3) The impact of PM2.5 concentration between cities in Henan Province of China has obvious causal directivity. The trophic coherence of the PM2.5-directed network is the smallest in autumn, with the most stable structure, while is with the largest vulnerability in summer.