以全国321个地级以上行政区为研究对象,基于城市间汽车、火车和航空班次联系测算各城市对外联系强度,通过核密度估计、位序-规模及探索性空间数据分析,对多元交通流视角下城市间交通流强度的空间格局进行提取和解析。研究表明:在整个城市体系中,高位次城市规模凸显,中小规模城市有待于进一步发育。基于公路和铁路联系的城际交通流分别表现为由沿海向内陆逐级减弱、以国家铁路大动脉沿线为中心的核心-边缘结构;基于航空联系的城际交通流表现出高度极化和点状镶嵌特征。

This paper aims to reveal the spatial pattern of urban external connection intensity from the respective of traffic flow. Previous urban studies on traffic flow laid more emphasis on single type of traffic flow while multi-flow synthesis was not taken into account. Moreover, the scale of prefecture-level city was not be covered before. Therefore, a further analysis on spatial pattern of multi-type traffic intensity between cities in China is required. To detail and comprehensively consider the spatial pattern of cities in China, we narrow the study scale to prefecture-level city. According to the list of prefecture-level divisions of China, this paper firstly takes 321 cities as basic study units. Data crawling was implemented under the C# language environment, which collects the runs number of three transport modes as basic data. Then we measure external connection index of each city respectively through bus, railway and flight schedules. In the following analysis, kernel density estimation was employed to describe spatial distribution pattern of urban external connection intensity, which implies a concentration trend. The number of runs shows bus > train > flight. Furthermore, Rank-size Rule was applied to portrait distribution variations of urban external connection intensity, among them flight schedules show significant rank-size feature, while train schedules come last. From the perspective of the whole urban system, the distribution of urban size is concentrated. Specifically, top-ranking cities have considerable scales, while small and medium-size cities need further development. After that, we use ESDA (Exploratory Spatial Data Analysis) to examine the spatial agglomeration. There is apparently positive spatial correlation between urban external connection index based on bus schedules data and that based on train schedules data. Spatially, a gradually weakening trend appears from coastal to inland. At the same time, core-periphery structure can be recognized, which identifies the national railway artery as core area and areas along the railway as periphery.