【摘要】 以长江经济带为研究对象,基于2015年城市空气质量监测数据,首先利用自然正交函数分析了126个城市空气质量指数（AQI）的时空演变特征,然后采用地理加权回归方法从空间异质性视角探讨了空气质量的社会经济影响因素。研究结果表明:（1） EOF第一模态特征向量的高值出现在长三角的江苏省,长江中游的武汉都市圈、长株潭地区以及成渝城市群。低值区则集中在云南的大部分城市、贵州的黔南州、四川的阿坝州以及浙江的舟山、台州和丽水。（2）月尺度的AQI时间系数变化大致呈U形分布,即1—5月呈现波动下降趋势,6—8月AQI总体处于"U"的谷底,9—12月逐渐波动上升。就年内季节变化来看,冬季空气质量最差,夏季空气质量最好,春秋两季居中。（3）人均地区生产总值、人口密度、PM_2.5浓度和SO₂浓度的增加均是导致空气污染加剧的重要因素,而FDI和建成区绿化覆盖率的增加有利于空气质量的改善,并且不同城市的影响因素对空气质量作用存在显著的空间异质性。更多还原

【Abstract】 We employed a cross-sectional data of the Yangtze River Economic Belt at the city level to investigate spatiotemporal variation characteristics of air quality index（AQI） of 126 cities by means of the empirical orthogonal function（EOF） method, and then applied the geographically weighted regression method to analyze socio-economic factors of AQI from a perspective of spatial heterogeneity. The findings are the followings.（1） The first mode of EOF has high eigenvector values in Jiangsu Province, Wuhan metropolitan region, Chang-Zhu-Tan region and Chengdu-Chongqing urban cluster while low high eigenvector values are shown in most cities of Yunnan Province, Qiannan of Guizhou Province, Aba of Sichuan Province and Zhoushan, Taizhou and Lishui of Zhejiang Province.（2） On the monthly time scale, the time coefficient of AQI displays a U-shaped curve. Specifically, it first declines with fluctuations from January to May, and then keeps the lowest values from June to August, and finally increases from September to December. In other words, air quality is the worst in winter and the best in summer within the year of 2015.（3） The increase in per capita GDP, population density, PM_2.5 concentration and SO₂ concentration are the important factors worsening air quality while FDI and green coverage ratio contribute to improving air quality. Finally, socio-economics factors may have impacts on air quality from city to city, presenting significant spatial heterogeneity.更多还原