大气污染物SO_2空间相关性的空间集聚分析
|
摘要
以东北三省2017年的大气污染物SO_2为研究对象,通过全局指标(全局Moran指数、Geary系数)、区域型指标(Moran’s I、局部Geary’s C、局部Getis’s G)等,对SO_2的空间聚集情况进行分析计算,比较两种指标的探测结果。结果表明,在全局型空间自相关的分析中,Moran指数、Geary’s C两个指标均表明东北三省SO_2存在显著的空间自相关性;Moran散点图、LISA集聚图、局部G系数集聚图等均揭示了东北地区36个地级市SO_2的局部空间相关性,即低值集聚区(冷点)主要集中在研究区东部,(热点)高值集聚区集中在研究区的西南部;通过对两种指数的分析可发现,在研究区的西南部,营口、大连、铁岭3个地区在Moran指数中为低-高集聚区,黑河为不相关地区,但在局部G系数中,营口、大连、铁岭为热点(高-高集聚),黑河为冷点(低-低集聚区),结合实际情况,对分析SO_2空间相关性来说,Moran指数相对G系数的分析结果更优。
Taking the atmospheric pollutant SO_2 of the three northeastern provinces in 2017 as the research object, through global indicators(global Moran index, Geary coefficient), and regional indicators(Moran'I, local Geary's C, local Getis' s G), the spatial aggregation of SO_2 was analyzed and calculated. The detection results of the two indexes were compared. The results showed that in the analysis of spatial autocorrelation, the Moran index and Geary's C index both indicated that there was significant spatial autocorrelation in SO_2 in the three northeastern provinces; Moran scatter plot, LISA agglomeration map,and local G cluster agglomeration etc. all revealed the local spatial correlation of SO_2 in prefecture-level cities in 36 prefecture-level cities in northeast China, That is, the low-value clusters(cold point) are mainly concentrated in the eastern part of the research area, and the high-value clusters(hot point) are concentrated in the southwest part of the research area; Through the analysis of the two indices, it could be found that in the southwestern part of the research area,Yingkou, Dalian and Tieling are low-high agglomeration areas in the Moran index, and Heihe is an unrelated area.However, in the local G coefficient, Yingkou, Dalian, Tieling are hot spots(high-high agglomeration) and Heihe is a cold spot(low-low agglomeration area). According to the actual situation, Moran index is better than G coefficient in analyzing the spatial correlation of SO_2.
Taking the atmospheric pollutant SO_2 of the three northeastern provinces in 2017 as the research object, through global indicators(global Moran index, Geary coefficient), and regional indicators(Moran'I, local Geary's C, local Getis' s G), the spatial aggregation of SO_2 was analyzed and calculated. The detection results of the two indexes were compared. The results showed that in the analysis of spatial autocorrelation, the Moran index and Geary's C index both indicated that there was significant spatial autocorrelation in SO_2 in the three northeastern provinces; Moran scatter plot, LISA agglomeration map,and local G cluster agglomeration etc. all revealed the local spatial correlation of SO_2 in prefecture-level cities in 36 prefecture-level cities in northeast China, That is, the low-value clusters(cold point) are mainly concentrated in the eastern part of the research area, and the high-value clusters(hot point) are concentrated in the southwest part of the research area; Through the analysis of the two indices, it could be found that in the southwestern part of the research area,Yingkou, Dalian and Tieling are low-high agglomeration areas in the Moran index, and Heihe is an unrelated area.However, in the local G coefficient, Yingkou, Dalian, Tieling are hot spots(high-high agglomeration) and Heihe is a cold spot(low-low agglomeration area). According to the actual situation, Moran index is better than G coefficient in analyzing the spatial correlation of SO_2.
引文
[1]邓敏,刘启亮,李光强,等.空间聚类及分析应用[M].北京:科学出版社,2011.
[2] TOBLER W R. A Computer movie simulating urban growth in the detroit region[J].Economic geography,1970,46:234-240.
[3] GETIS A,ORD J K. The analysis of spatial association by use of distance statistics[J].Geographical analysis,1992,24(3):189-206.
[4]王劲峰,李连发,葛咏,等.地理信息空间分析的理论体系探讨[J].地理学报,2000,55(1):92-103.
[5]李连发,王劲峰.地理空间数据挖掘[M].北京:科学出版社,2014.
[6] MORAN P A P. The interpretation of statistical maps[J].Journal of the royal statistical society B,1948,37:24-251.
[7] GEARY R C. The contiguity ratio and statistical mapping[J].The incorporated statistician,1954(5):115-145.
[8]程度胜.经济发展与环境污染相互关系的空间统计分析[D].长沙:湖南大学,2015.
[9]徐志伟,常旭蕊.投资增长加重了中国工业二氧化硫的排放吗——基于空间面板模型的总量判断与结构比较[J].贵州财经大学学报,2016(2):90-99.
[10]郭梦梦,仇方道.淮海经济区工业SO2排放的空间相关性及影响因素分析[J].云南地理环境研究,2017,29(2):33-40.
[11] GOODCHILD M. Spatial autocorrelation concepts and techniques in modern geography[M].Norwich,UK:Geo Books,1986.
[12]陈强.高级计量与经济学及Stata应用[M].北京:高等教育出版社,2014.
[13]张松林,张昆.全局空间自相关Moran指数和G系数对比研究[J].中山大学学报(自然科学版),2007(4):93-97.
[2] TOBLER W R. A Computer movie simulating urban growth in the detroit region[J].Economic geography,1970,46:234-240.
[3] GETIS A,ORD J K. The analysis of spatial association by use of distance statistics[J].Geographical analysis,1992,24(3):189-206.
[4]王劲峰,李连发,葛咏,等.地理信息空间分析的理论体系探讨[J].地理学报,2000,55(1):92-103.
[5]李连发,王劲峰.地理空间数据挖掘[M].北京:科学出版社,2014.
[6] MORAN P A P. The interpretation of statistical maps[J].Journal of the royal statistical society B,1948,37:24-251.
[7] GEARY R C. The contiguity ratio and statistical mapping[J].The incorporated statistician,1954(5):115-145.
[8]程度胜.经济发展与环境污染相互关系的空间统计分析[D].长沙:湖南大学,2015.
[9]徐志伟,常旭蕊.投资增长加重了中国工业二氧化硫的排放吗——基于空间面板模型的总量判断与结构比较[J].贵州财经大学学报,2016(2):90-99.
[10]郭梦梦,仇方道.淮海经济区工业SO2排放的空间相关性及影响因素分析[J].云南地理环境研究,2017,29(2):33-40.
[11] GOODCHILD M. Spatial autocorrelation concepts and techniques in modern geography[M].Norwich,UK:Geo Books,1986.
[12]陈强.高级计量与经济学及Stata应用[M].北京:高等教育出版社,2014.
[13]张松林,张昆.全局空间自相关Moran指数和G系数对比研究[J].中山大学学报(自然科学版),2007(4):93-97.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |