Research On The Impact Of Urban Centrality On Air Quality In Chinese Cities From A Social Network Perspective

With the rapid development of Chinese cities,the flow of production factors has broken through geographical and administrative boundaries,and the connections between cities have become increasingly close.The complex network of human and material exchanges between cities also affects air pollution.In the past,the pattern of controlling air pollution was regulated as administrative units.And there has been a lack of thinking about the relationship between the cities.China’s urban correlation network has gradually improved with the migration of production factors in the urban network.How to systematically analyze the role of cities in this correlation network,as well as the impact of this role status on air quality,that is,the impact of urban centrality on air quality,and its mechanism,has become the focus of this study.The generation of air pollution not only originates from local production and living activities,but is also affected by the complex human and material exchange network between cities.It has the characteristics of being trans-regional,complex and dynamic,and its spillover effects are far beyond the scope of local government management and control.Therefore,based on the perspective of social network analysis,this study starts from the concept and theoretical analysis of urban correlation network,and uses the revised gravity model to construct urban correlation network based on the traditional gravity model.On this basis,urban centrality is included in the research framework,and the IPAT model is extended based on multiple theories.The panel data of 283 cities in China from 2015 to 2021 are used to discuss the effects of urban centrality on air quality.The impact of urban centrality on air quality is analyzed from different perspectives,such as two types of networks and different development stages of urban agglomerations.Furthermore,this paper discusses the specific mechanism of urban centrality on air quality from three aspects:threshold effect of economic growth,moderating effect of industrial structure and spatial spillover effect,and provides policy suggestions.The main conclusions of this paper are as follows:（1）Systematicly sorting air quality index（AQI）in China from the macro level shows that the spatial distribution pattern of air pollution in China is relatively stable,and the overall distribution of AQI is high in the central part of China and low in the southern and northern parts of China.From 2015 to 2021,the AQI continued to decline,and the air quality has been significantly improved across the country,but PM_2.5,PM₁₀and O₃ are still the main pollutants in most cities.（2）The flow of factors broke through the administrative boundary restrictions,and the overall network relationship of Chinese cities was complicated.The results show that the cities in the coastal areas are relatively closely connected,while the cities in other areas are relatively loosely connected.Chinese cities are interconnected in the correlation network,the overall correlation degree of the urban network is steadily.There are many redundant relationships between cities,and multiple spillover effects are obvious.Cities with high urban centrality can be divided into municipalities directly under the central government or provincial capitals with high political status,developed cities with high economic level,transportation hub cities with high transportation capacity,and satellite cities providing services to large cities.（3）Using the extended IPAT model,the panel data of 283 cities from 2015 to 2021are used to empirically analyze the effect of urban centrality on air quality.The results show that urban centrality has a significant positive effect on AQI,which means the congestion effect and siphoning effect caused by the increase of urban centrality will lead to a decrease of air quality.Increased environmental awareness and collaborative governance policies in urban agglomerations can significantly mitigate the impact of urban centrality on air quality.The impact of urban centrality on air quality varies across agglomerations of different maturity.（4）The transmission of the impact of urban centrality on air quality can be effected by the threshold effect of economic growth and the moderating effect of industrial structure.The results show that there is an S-shaped threshold effect of economic growth in the process of the influence of urban centrality on air quality,and the double thresholds indicate that there are differences in the degree of influence of urban centrality on air quality at different stages of economic growth.Industrial structure has a strong positive moderating effect on the relationship between urban centrality and air quality,which means the larger the proportion of secondary industry in the urban economy,the stronger the positive impact of urban centrality on air quality.（5）The transmission of the impact of urban centrality on air quality can be effected through the spatial spillover effects of urban centrality and air quality.The results show that the spatial spillover effect exists in both the adjacency matrix and the inverse linear distance matrix.At the same time,urban centrality also has a spatial spillover effect.As cities improve their own urban centrality,government power spills over and pollution is transferred to other cities,and the costs of environmental governance are also transferred to other cities,worsening the air quality of neighbouring cities.There is a strong correlation between cities in a geographical perspective,whether they are neighbouring or just not far.This proves once again that the control of air pollution cannot rely on a single city,and therefore the control of air pollution should adopt a regional cooperative way.Finally,based on the above studies,this paper comprehensively examines the current situation of air quality in China,the impact of urban centrality on air quality and its mechanism.Furthermore,recommendations for trans-regional collaborative control of air pollution are presented.