"时空接近"空间自相关模型构建及其应用——以长三角区域经济差异分析为例

地理学第一定律是空间自相关统计学的核心和基石,传统的空间自相关以邻接性或空间距离函数确定空间权重矩阵。但随着快速交通网络变得具有渗透性,空间距离内涵已发生质的变化,"接近"概念有必要从"空间接近"拓展为"时空接近"。基于这一新的内涵,提出"时空接近"空间自相关概念并构建相应计算模型,结合表征时间距离的可达性指标将该模型应用于长三角区域经济差异分析。通过计算全局Moran's I和局域Moran's I,结果表明:在全局空间自相关指标上,"时空接近"空间自相关能更好地揭示地理事物间的关联动态变化。在局域空间自相关指标上,当时间距离截取在合理范围时,邻接性表现出较强的稳健性,但"时空接近"与传统"空间接近"的空间相关性存在较大差异,两者结合能更客观揭示地理事物空间关联的现象和本质。

Construction of spatial autocorrelation method of spatial-temporal proximity and its application: Taking regional economic disparity in the Yangtze River Delta as a case study

Tobler's First Law is the core and the basis of spatial autocorrelation statistics.Spatial weight matrix is decided by the traditional spatial autocorrelation in the way of proximity or the given distance function.The rapid transportation networks are becoming saturated;hence,the meaning of spatial distance changes essentially.The connotation of proximity should be extended from "spatial proximity" to "spatial-temporal proximity".Based on such connotation,we propose a new concept-spatial autocorrelation of ...