中文摘要：

基于春运人口流动大数据,选取对外联系度、优势流、城市位序-规模分析等方法对转型期中国城市网络特征进行分析。结果显示：1城市网络层级结构中蕴藏着位序-规模规律,但与理想的帕累托分布有所区别,城市规模彼此差异相对较小;2空间距离与城市等级在城市网络联系中发挥支配性作用,保证了城市网络的层级性与有序性;3中国城市网络核心联系呈现＂两横三纵＂特征,该特征与铁路大动脉的空间分布高度吻合;4东部地区城市网络联系更加密切,而西北、西南地区则相对稀疏,基本上以＂胡焕庸线＂为界,而＂兰新线＂是突破这一限制的潜在力量;5中国东北地区未形成明显的区域性中心,城市联系形成带状网络;6华北与华南地区的＂灯下黑＂现象值得警惕,缓解这一问题的可行办法是核心城市功能的对外疏散,加强核心城市与周边城市之间的联系;7带状区域发展或许将成为未来中国区域经济发展的流行模式和中坚力量。总体上看,针对于揭示转型期中国城市网络结构特征,春运人口流动数据具有一定的研究价值,是城市与人口研究领域一个值得深入挖掘的重要数据源。

英文摘要：

The space of flows theory proposed by Manuel Castells has given birth to the network perspective of city network. In contrast with the traditional perspective of urban system which is based on the central place theory, city network perspective pays more attention to the interactions and linkages between cities and regions. Thus, the city network is becoming one of the new and hot topics in the field of urban geography.＂Chunyun＂is a well-known socio-economic phenomenon unique to transitional China, which refers to a blooming of population flows during the Spring Festival, or refers to the period when the blooming happens. Previously,the real population flows data in＂Chunyun＂were difficult to obtain. The situation had not changed until the＂Big data on human migration during the spring festival from Baidu map＂（＂Baidu migration data＂in short）came up. The big data were gathered from the locations provided by hundreds of millions smart phone users through Location Based Service（LBS） Baidu map data source, and was published in the form of interactive heat map that displays people＇s travel routes in China during the Chunyun period. Based on Baidu migration data, using degree of external linkages, dominant flows and network-based rank-size analysis, the spatial pattern of city network in transitional China was studied in this article. The study process certificates that Baidu migration data is indeed a high quality data sources for the study of city network, and has turned up some interesting results： 1） The distribution of external linkage degree of cities in China follows Zipf＇s law, but differs from the ideal Pareto distribution. 2） The factor of spatial distance and city level play key roles in the formation of urban network of China, and ensure the hierarchy and regularity of the network. 3） The spatial distribution of core linkages in the city network could be summarized as＂Three-horizontal Two-longitudinal＂, which almost coincides with the rail arteries in China. The＂Three-