地理学报 ›› 2014, Vol. 69 ›› Issue (6): 759-770.doi: 10.11821/dlxb201406004

• 人口与城市地理 • 上一篇    下一篇

中国城市体系等级结构及其空间格局——基于DMSP/OLS夜间灯光数据的实证

吴健生1, 刘浩1, 彭建2, 马琳1,2   

  1. 1. 北京大学城市人居环境科学与技术重点实验室, 深圳518055;
    2. 北京大学城市与环境学院, 北京100871
  • 收稿日期:2013-08-15 修回日期:2014-04-12 出版日期:2014-06-20 发布日期:2014-06-20
  • 通讯作者: 刘浩(1987-),男,硕士,主要从事城市与区域规划、区域经济研究。E-mail:liuhao4680@126.com
  • 作者简介:吴健生(1965-),男,博士,教授,主要从事遥感与GIS、景观生态学与土地利用规划研究。E-mail:wujs@szpku.edu.cn
  • 基金资助:
    国家自然科学基金项目(41330747)

Hierarchical structure and spatial pattern of China’s urban system:Evidence from DMSP/OLS nightlight data

WU Jiansheng1, LIU Hao1, PENG Jian2, MA Lin1,2   

  1. 1. Key Laboratory of Human Environmental Science and Technology, Peking University, Shenzhen 518055, Guangdong, China;
    2. College of Urban and Environmental Sciences, Peking University, Beijing 100871, China
  • Received:2013-08-15 Revised:2014-04-12 Online:2014-06-20 Published:2014-06-20
  • Supported by:
    National Natural Science Foundation of China, No.41330747

摘要: 完善的城市体系可以不断优化各个城市的时空资源并强化区域城市的集聚效应,切实推动区域可持续发展与城市综合效益最大化。受惠于城市因素流与城市作用潜力的显著相关关系,节点体系研究可以准确测度城市体系内部各个城市之间的相互作用关系,又避免了功能体系研究所需矢量数据的难以获取与测算庞杂,而DMSP/OLS夜间灯光遥感数据可以综合地定量测度城市地区人类活动的广度与强度。基于夜间灯光指数的引力模型定量测度中国341个城市之间的相互作用关系,并通过二阶段聚类法综合分析其城市体系等级结构与空间格局。中国节点城市被划分为7 大国家节点城市(包括北京、天津、上海、广州、重庆、哈尔滨与沈阳)、26 个区域节点城市与107 个省域节点城市,而中国城市体系则被划分为2 个国家城市体系(北方城市体系与南方城市体系)、8 个区域城市体系(华北城市体系、东北城市体系、西北城市体系、黄河中游城市体系与华东城市体系、华南城市体系、西南城市体系、长江中游城市体系) 与31 个省域城市体系。城市体系的地域范围是相对的、局部的,没有明确的识别界线,不过通过对基于DMSP/OLS夜间灯光数据的中国城市体系等级结构与空间格局与顾朝林实证研究的分析结果进行进一步的综合比对评估,总体而言,基于DMSP/OLS夜间灯光数据对城市体系的等级结构与空间格局进行实证研究是可信的和可行的。

关键词: 城市体系, 空间格局, DMSP/OLS, 中国, 等级结构, 引力模型

Abstract: A perfect urban system can promote the urban agglomeration effect, and the prerequisite for urban system is that regional socio-economic development should actually reach a certain stage. Thanks to the strong correlation between the urban element flow and the mass of spatial units, the node system study in general should scientifically measure the interaction intensity among spatial units, and also avoid the difficulty that the vector data of the function system study is still hard to obtain and measure. As a special node system study data, DMSP/OLS nighttime light data can be used to measure urban human's activity intensity and activity breadth comprehensively, having a strong positive correlation with local GDP, population size, urbanization level and so on. So the urban nighttime light index can make a perfect combination between a single statistical index and a comprehensive index system. The gravity model based on DMSP/OLS nighttime light data is used to measure the interaction intensity among 341 cities in mainland China, and then the two step cluster analysis is made to analyze hierarchical structure and spatial pattern of mainland China urban system comprehensively. There are 7 national node cities (including Beijing, Tianjin, Shanghai, Guangzhou, Chongqing, Harbin and Shenyang), 26 regional node cities and 107 provincial node cities in mainland China. Mainland China urban system are divided into 2 national urban systems (the northern national urban system and the southern national urban system) and 8 regional urban systems (North China urban system, Northeast China urban system, Northwest China urban system, Middle Yellow River urban system, and East China urban system, South China urban system, Southwest China urban system, Middle Yangtze River urban system), as well 31 provincial urban systems. Every urban system is closely linked, interdependently and interactionally, whose slope is relative and local area, and then some urban systems and many cities may be define accurately again, but the empirical study of hierarchical structure and spatial pattern of mainland China urban system evidenced from DMSP/OLS nightlight data is rather objective and compellent. All in all, to analyze hierarchical structure and spatial pattern of urban system based on DMSP/OLS nightlight data is obviously feasible and scientific.

Key words: hierarchical structure, spatial pattern, DMSP/OLS, China, urban system, gravity model