地理学报 ›› 2008, Vol. 63 ›› Issue (12): 1257-1267.doi: 10.11821/xb200812003

• 城市地理 • 上一篇    下一篇

世界城市地铁发展历程与规律

曹小曙, 林强   

  1. 中山大学地理科学与规划学院, 广州510275
  • 收稿日期:2008-05-04 修回日期:2008-10-29 出版日期:2008-12-25 发布日期:2010-08-03
  • 作者简介:曹小曙(1970-), 男, 甘肃人, 博士, 博士生导师, 中国地理学会会员, 主要从事交通地理与土地利用研究。 E-mail: caoxiaoshu964@sohu.com
  • 基金资助:

    国家自然科学基金项目(40571052)

The Evolution of Worldwide Metro Systems: A Study on Their Scales and Network Indexes

CAO Xiaoshu, LIN Qiang   

  1. School of Geography and Planning, Sun Yat-sen University, Guangzhou 510275, China
  • Received:2008-05-04 Revised:2008-10-29 Online:2008-12-25 Published:2010-08-03
  • Supported by:

    National Natural Science Foundation of China, No.40571052

摘要:

基于地铁系统数据统计分析, 对地铁建设的空间时序以及网络发育差异进行归纳, 结 合相关数据推导出地铁规模的匡算模型, 并依据地铁系统的属性指标组和网络指标组对世界 121 个城市地铁系统样本进行聚类分析, 将其划分为3 个类别, 并讨论城市人口规模和城市 用地规模对地铁规模的影响, 最后对中国53 个城市进行地铁规模匡算和等级划分。结果显 示, 世界城市地铁建设在时空上存在波动性, 地区间地铁系统的网络发育不均衡, 欧洲和北 美洲的地铁网络发育较为成熟, 而亚洲和南美洲的地铁网络则有较大的扩展潜力。第一类别 地铁系统规模大, 网络发育成熟, 但仍有较大的扩展潜力, 地铁规模与城市用地规模的相关 性更强; 第二类别地铁系统网络的发育程度较好, 但是扩展潜力一般, 地铁规模与城市人口 规模和城市用地规模的相关性较弱; 第三类别地铁系统的规模普遍较小, 地铁网络发育处于 较低等级水平, 地铁网络仍然有很大的成环潜力和结合潜力, 地铁规模与城市人口规模的相 关性更强。中国城市地铁的匡算规模介于37.4~129.9 km, 根据模型的匡算结果中国地铁系统 可以划分为4 个等级, 未来中国的地铁建设仍然有很大潜力。

关键词: 城市地铁, 网络结构, 属性指标, 聚类分析, 世界, 中国

Abstract:

Abstract: Based on the statistics of 121 metro systems in the world, this paper analyses the worldwide metro construction from 1863 to 2003. Linear regression is introduced to show the relationship between the metro scale and urban population, as well as surface area. And by applying the hierarchical cluster analysis, three types of metro systems are identified in terms of their scales and network indexes: the first metro system with 10 samples, which is huge and with complex network, such as London and New York; the second metro system with four samples, which is relatively big but with limited network patulous potential, such as Athens and Vienna; and the third metro system with 107 samples, which is relatively small but has great patulous potential in network, such as Beijing and Montreal. Our findings suggest that metro scale, as well as network indexes, show great differences among five major continents. Metro systems in Europe and North America embrace a more mature network; however, metro systems in Asia and South America have greater network patulous potential. Also we find that in most cases urban population has more impacts on metro scale, especially to the samples in the third metro system. However, when it comes to the first metro system, surface area has significant impacts on metro scale. Finally three linear regression models are implemented to calculate the theoretical scales of 53 Chinese major cities, and it shows that theoretic metro scales in these cities range from 37.2 km to 129.9 km. These metro systems can be divided into four classes based on their model calculations. According to the result, Beijing, Shanghai and Chongqing should develop the biggest metro systems in China. With the progress of urbanization in China, metro systems will grow rapidly in the coming years, especially in the developed regions.

Key words: metro systems, network, attribute indexes, hierarchical cluster analysis, world, China