地理学报

地理学报 ›› 2004, Vol. 59 ›› Issue (6): 1048-1057.doi: 10.11821/xb200406029

• 城市地理 • 上一篇    下一篇

基于ESDA-GIS的城镇群体空间结构

马晓冬1,2, 马荣华3, 徐建刚1   

  1. 1. 南京大学城市与资源学系,南京 210093;
    2. 徐州师范大学城市与环境学院,徐州 221009;
    3. 中国科学院南京地理与湖泊研究所,南京 210008
  • 收稿日期:2004-05-12 修回日期:2004-08-21 出版日期:2004-11-25 发布日期:2004-11-25
  • 作者简介:马晓冬 (1971-), 男, 江苏徐州人, 副教授, 博士生。主要从事城市与区域规划及GIS应用研究。 E-mail：maxiaodong@xznu.edu.cn
  • 基金资助:

    国家自然科学基金项目 (40301038; 40371038)

Spatial Structure of Cities and Towns with ESDA-GIS Framework

MA Xiaodong1,2, MA Ronghua3, XU Jiangang1   

  1. 1. Department of Urban and Resources Science, Nanjing University, Nanjing 210093, China;
    2. College of Urban and Environmental Science, Xuzhou Normal University, Xuzhou 221009, China;
    3. Nanjing Institute of Geography and Limnology, CAS; Nanjing 210008, China
  • Received:2004-05-12 Revised:2004-08-21 Online:2004-11-25 Published:2004-11-25
  • Supported by:

    National Nature Science Foundation of China, No.40301038; No.40371038

PDF (PC)

赞

被引次数

494

摘要：

基于ESDA-GIS的空间分析框架，利用江苏省1346个小城镇的统计数据，对其城镇群体空间结构进行了研究。首先对统计数据进行了概括性因子分析，抽取的规模与经济因子值的频率分布都是偏态的，且两因子之间不存在规模－效益的正相关性。随后，对小城镇群体的空间结构进行了探索性分析，由密度图显示：小城镇空间分布不均衡，其密度由东南部的沿长江、环太湖地区向西和向北递减。通过空间自相关测度分析得出：小城镇的经济因子的空间分布具有正相关性，呈现出空间集聚的特征；规模因子的空间相关特征不明显。最后，将小城镇的局域空间自相关系数按县 (市、区) 行政单元进行聚类，利用方差图的结论修正聚类结果，得出江苏省小城镇经济发展类型分区：三区、一环 (带)、一片，即苏北中部地区、苏中及宁镇区、苏锡常地区、苏北边缘环带、睢宁片区，对应的经济发展类型分别为：弥漫发展型、极核发展型、集群发展型、过渡发展型、持续贫困型。

关键词: ESDA, 城镇群体, 空间结构, 空间自相关, 江苏

Abstract:

Based on analysis with ESDA-GIS framework, by using statistical data of 1346 small cities and towns in Jiangsu province in 2002, their spatial structure was studied. We, at first, finished summary factor analysis from database, and drew the two principal components: economic factor and scale factor. By comparison, we found that the frequency distribution of the two factors was skewed and the skewed degree of scale factor was higher than that of economic factor. But there is no positive correlation as scale-benefit between the two factors. In this paper, exploratory spatial structure of the small cities and towns was analyzed. With the density map, we found their spatial distribution is imbalanced, whose density would descend gradually from Yangtze River zone and Taihu Lake area in the southeast to the west and the north, which showed a belt and circle structure. Based on the analysis with global SAMs and local SAMs, we came to some conclusions. The spatial distribution of economic factor of the small cities and towns showed positive autocorrelation and spatial cluster, but the autocorrelation of scale factor is very weak. Finally, after classifying the local SAMs coefficients of small cities and towns according to administrative units of country, we analyzed economic significance with anisotroptric variogram and improved the classified result. We found the spatial cluster of the economic development of the small cities and towns in Jiangsu province: three districts, one belt, and one part, i.e., the middle part of the northern Jiangsu area, the middle part of Jiangsu province and Nanjing-Zhenjiang area, Suzhou-Wuxi-Changzhou area, and the northern part of Jiangsu border belt, and district around Suining. The corresponding economic development types are diffusing development type, polarizing development type, clustering development type, transitional development type, and continual impoverished type.

Key words: ESDA, spatial structure, autocorrelation, Jiangsu province