武汉市道路结构与商业集聚空间关联分析

城市道路结构是影响城市商业集聚的重要因素,研究二者之间的关联对于商业布局和道路规划具有重要支撑作用。本文以武汉市都市发展区为例,运用空间句法模型计算道路结构指标,采用GIS核密度法计算2014年商业兴趣点(C-POI)密度以分析商业集聚的空间结构,通过双变量相关分析道路结构变量与商业集聚程度的关联关系。结果显示：①道路结构指标中,与商业集聚关联最强的是道路全局集成度。在各商业类型中,与道路结构相关程度最大的是金融保险服务集聚。②道路连接值和全局集成度值分别与商业POI密度呈显著正相关,呈现“高—高”集聚的空间关联模式;总深度值与商业POI密度呈现显著负相关,呈现“高—低”集聚的空间关联模式;商业POI密度随控制值升高呈现先升高后降低的趋势。③商业集聚分布呈现为“核心—过渡区—边缘区”的多核心多层次结构,“高—高”集聚与“高—低”集聚模式集中分布于“核心”及其附近的“过渡区”。

Correlation analysis of road structure and commercial agglomeration in Wuhan City

Urban road structure is one of the most important factors influencing commercial agglomeration. Research on the relationship between urban road structure and commercial agglomeration plays a supporting role for the layout of services and traffic planning. Based on theories of space syntax combined with GIS and bivariate correlation analysis, we explored the correlation between road structure and commercial agglomeration in the Wuhan metropolitan area. The space syntax model was used to compute road structure indicators. The kernel density estimation method was used to calculate the density of commercial points of interests (C-POI) in 2014 to analyze the spatial structure of commercial agglomeration. We used the Pearson correlation coefficient to analyze the correlation between road structure and commercial agglomeration. The results show that: (1) Global integration showed the highest correlation with the commercial agglomeration among the four spatial syntactic variables. Finance and insurance services had the highest correlation with road structure. (2) Connectivity value and global integration value were significantly and positively correlated with C-POI density, with a spatial correlation pattern of "high-high" agglomeration. Total depth value showed a significant and negative correlation with C-POI density, with a spatial correlation pattern of "high-low" agglomeration. C-POI density increased first and then decreased with increasing control values. (3) The spatial distribution of commercial agglomeration presented a "multicore-transitional area-periphery" multiple level structure. "High-high" agglomeration and "high-low" agglomeration were concentrated in the "core area" and the "transitional area" in the vicinity of the cores.