基于PPA方法的农村居民点空间分布模式研究——以重庆市主城区为例

Spatial distribution pattern of rural settlements based on point pattern analysis methods——A case study of chongqing main district

针对农村居民点的空间分布影响着区域经济发展的规模、方向以及发展的可持续性问题,引入最邻近距离指数(nearest neighbor index,NNI)、核密度估计(kernel density estimation,KDE)和Voronoi图及其变异系数(CV值)3种点模式分析(point pattern analysis,PPA)方法,从不同的角度揭示重庆市农村居民点空间分布模式。研究显示,重庆市主城区农村居民点NNI是0.072,总体上属于典型的集聚分布模式;在空间分布上,农村居民点形成了不同等级的热点分布区域,而且具有较强的方向性;在聚散特征上,离市中心越远,农村居民点分布越趋于均匀分布,离市中心越近,农村居民点集聚性越强。从研究方法来看,3种PPA方法各有优缺点:NNI方法便于从数量上揭示空间点的总体特征;KDE在空间上探测地理事件的热点区域方面具有优势;Voronoi多边形CV分析法更关注一定统计单元的地理现象分布特征。三者有机结合使得对点空间格局的研究更深入、更全面、更具体。

The nearest neighbor distance index（NNI）,kernel density estimation（KDE）,Voronoi diagram and its coefficient of variation（CV） are introduced in this study.These methods can be called point pattern analysis methods.The distribution pattern point of the space of rural residents in Chongqing city is revealed by these methods from different angles.Study shows that： the nearest neighbor index（NNI） is 0.072,belonging to the aggregated typical distribution pattern;in the spatial distribution aspect,there are lots of hot regional distribution in different grades,and to some extent,the spatial distribution has directional property;in the clutch characteristics,the farther away from the city center,rural the settlements distribution is more evenly distributed,the more close to the centre,the more aggregated.From the research point of view,there are some advantages and disadvantages by using different point pattern analysis methods： NNI method can reveal the general characteristics of spatial points from the quantity;KDE has advantages to check the hot spots of geographical event;Voronoi polygon method focuses on analysis the geographical phenomenon of a certain statistical unit distribution characteristics.The integration of the above three kinds of point pattern analysis methods,spatial point pattern can be understood more in-depth,more comprehensive.