长江经济带空气质量指数的时空特征及驱动因素分析——基于贝叶斯空间计量模型的实证

Spatio-temporal Characteristics of Air Quality Index and Its Driving Factors in the Yangtze River Economic Belt: An Empirical Study Based on Bayesian Spatial Econometric Model

基于2015年长江经济带126个城市空气质量监测数据,首先利用探索性空间数据分析方法揭示了空气质量指数(AQI)的时空演变特征,然后采用贝叶斯空间滞后模型探讨了长江经济带空气质量指数的社会经济驱动因素。研究结果表明:①长江经济带年AQI在空间上整体具有东高西低,长江以北高长江以南低的分布特点,具有明显的空间集聚特征。空气污染严重的热点地区主要集中长三角城市群的江苏省、浙北地区、皖北大部分地区以及上海市。空气质量较好的冷点地区则主要集中在云南省、四川的攀枝花以及贵州的大部分地区。②长江经济带AQI在季节上呈现冬春高、夏秋低的季节变化趋势。总体而言,四季的高值集聚主要分布在鄂皖苏,低值集聚主要分布在云贵地区。③贝叶斯空间滞后模型回归结果显示,长江经济带空气质量存在显著的空间溢出效应。此外,模型结果证实了“环境库兹涅兹曲线”假说;FDI回归系数为正,支持了“污染避难所”假说;人口密度、公路客运量均是导致空气污染加剧的重要因素,而第三产业比重和建成区绿化覆盖率增加有利于长江经济带空气质量的改善。

Based on a cross-sectional data at the city level in the Yangtze River Economic Belt,this paper firstly employs the exploratory spatial data analysis method to investigate the spatio-temporal variation characteristics of air quality index(AQI)of 126 cities in 2015,and then applies the Bayesian spatial econometric model to explore the socio-economic driving factors of air quality index of the Yangtze River Economic Belt.The findings are the following:1)The distribution of the annual average AQI values in the Yangtze River Economic Belt exhibits a significant spatial cluster pattern,specifically high AQI values in the north and low AQI values in the south.Moreover,the hot spot analysis results show that the most polluted areas are mainly Jiangsu province,the northern Zhejiang province,Shanghai and the most areas of northern Anhui province while Yunnan province,Panzhihua city of Sichuan and the most areas of Guizhou province show better air quality.2)Regarding the seasonal characteristics of AQI values in the Yangtze River Economic Belt,it also shows a typical seasonal characteristic,specifically,high AQI values in both winter and spring,and low AQI values in both summer and autumn.Overall,during the whole year high AQI values are mainly concentrated on Hubei province,Anhui province and Jiangsu province while low AQI values on Yunnan province.3)The Bayesian spatial lag model indicates that the urban AQI values of the Yangtze River Economic Belt show significant spatial spillover effects.Moreover,the environmental Kuznets curve hypothesis has been verified,indicating that as GDP per capita increases,air quality worsens.However,when GDP per capita continues to increase,air quality becomes better.Besides,the coefficient of FDI variable is significant and positive,indicating the pollution haven hypothesis.Finally,the increase in population density and highway passenger transportation are important driving factors worsening air quality while the higher proportion of the tertiary industry and green coverage ratio improvements contribute to improving air quality of the Yangtze River Economic Belt.