Scientia Geographica Sinica  2016 , 36 (12): 1843-1849 https://doi.org/10.13249/j.cnki.sgs.2016.12.009

论文

基于Ripley’s K函数的南京市ATM网点空间分布模式研究

王结臣123, 卢敏1, 苑振宇1, 芮一康12, 钱天陆1

1.南京大学地理信息科学系, 江苏 南京 210023
2. 江苏省地理信息科学重点实验室, 江苏 南京 210023
3. 江苏省地理信息资源开发与利用协同创新中心, 江苏 南京 210023

Point Pattern Analysis of ATMs Distribution Based on Ripley’s K-Function Method in Nanjing City

Wang Jiechen123, Lu Min1, Yuan Zhenyu1, Rui Yikang12, Qian Tianlu1

1. Department of Geographic Information Science, Nanjing University, Nanjing 210023, Jiangsu, China
2. Jiangsu Province Key Laboratory of Geographic Information Science and Technology, Nanjing University, Nanjing 210023, Jiangsu, China
3. Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, Nanjing 210023, Jiangsu, China

中图分类号:  P208

文献标识码:  A

文章编号:  1000-0690(2016)12-1843-07

收稿日期: 2015-11-9

修回日期:  2016-03-25

网络出版日期:  2016-12-20

版权声明:  2016 《地理科学》编辑部 本文是开放获取期刊文献,在以下情况下可以自由使用:学术研究、学术交流、科研教学等,但不允许用于商业目的.

基金资助:  国家自然科学基金项目(41571377,41401450)资助

作者简介:

作者简介:王结臣(1973-),男,安徽太湖人,教授,博导,主要研究GIS理论与应用、地理空间分析。E-mail:wangjiechen@nju.edu.cn

展开

摘要

运用Ripley’s K函数的相关理论,以南京市ATM网点为研究对象,分别从平面与网络空间两种视角,在中心城区范围与主城区范围两种空间尺度上,通过单变量K函数法分析ATM网点的分布模式,通过双变量K函数法分析ATM网点与地铁站点的空间关联情况,最后对计算结果进行评价与分析。研究表明,ATM网点在南京主城区与中心城区均呈现出较强的集聚状态;在一定的距离范围内,ATM网点与地铁站点之间也有较强的依赖关系。同时,对于沿着路网分布的地理空间点状对象而言,利用网络K函数法进行空间点模式分析比用平面K函数法更加符合实际情况。

关键词: Ripley’s K函数 ; ATM网点分布 ; 点模式分析

Abstract

Since distributions of many types of urban objects are not random but in some particular patterns, analyzing and revealing the spatial distribution pattern of these points in urban space are essential to understand social, economic and geographical factors behind the distributions, and analysis results are conductive to wide applications such as facility layout and aided decision support. In point pattern analysis, the results may be biased by merely calculating the nearest neighbor distance. The Ripley’s K-function was therefore proposed with advantages of considering the distance between any pair of points. Because many urban points associated with human activities are constrained by road networks, a network K-function, as an extension of traditional planar K-function, is then presented by applying a network distance, i.e., the shortest path distance between any pair of two points. In this article, the Ripley’s K-function was applied to analyze the spatial distribution characteristics of ATMs in Nanjing City. First of all, we used both planar univariate K-function and network univariate K-function to analyze the distribution pattern of ATMs at spatial scales of downtown areas and main urban districts. Then we used planar bivariate K-function and network bivariate K-function to investigate the spatial correlation between ATMs and metro stations in main urban districts. Local bivariate (cross) K-functionwas finally applied to explore the impact of metro stations on the ATMs in local areas. The results show that ATMs are highly clustered in both planar and network space and the cluster characteristic is more significant in downtown areas than in main urban areas. Besides, ATMs and metro stations are highly correlated in the study area. With the increase of the measuring distance, the relationship between ATM and metro station distributions shows more obvious characteristic of spatial aggregation within a certain distance. In the analysis of local cross K-function method, ATMs are clustered with metro stations in downtown areas while there is no significant clustering characteristic between ATMs and metro stations on the outskirts. It implies that the distribution of ATMs is mainly determined by the regional commercial development. In addition, for spatial pattern analysis on point objects distributed along road networks, network K-function method is more practical than planar K-function in terms of revealing appropriate distribution pattern and relationship between two types of point objects.

Keywords: Ripley’s K-Function ; ATMs distribution ; point pattern analysis

0

PDF (722KB) 元数据 多维度评价 相关文章 收藏文章

本文引用格式 导出 EndNote Ris Bibtex

王结臣, 卢敏, 苑振宇, 芮一康, 钱天陆. 基于Ripley’s K函数的南京市ATM网点空间分布模式研究[J]. , 2016, 36(12): 1843-1849 https://doi.org/10.13249/j.cnki.sgs.2016.12.009

Wang Jiechen, Lu Min, Yuan Zhenyu, Rui Yikang, Qian Tianlu. Point Pattern Analysis of ATMs Distribution Based on Ripley’s K-Function Method in Nanjing City[J]. Scientia Geographica Sinica, 2016, 36(12): 1843-1849 https://doi.org/10.13249/j.cnki.sgs.2016.12.009

城市中很多事件的发生地点往往不是随机分布,而是在特定空间呈现一定的集聚状态。如果将事件抽象成点,则可应用空间点模式的相关理论进行分析。 Ripley’s K函数是空间点模式分析中最常用的方法之一[1,2]。国内外相关学者对 K函数的理论和应用进行了许多富有成效的研究,Ripley最早提出空间点模式分析理论,它基于连续平面上欧式距离假设[2,3];高凯等基于多年TM遥感数据,运用平面 K函数法对武汉市多尺度动态景观格局的特征及其变化情况进行了分析,以此作为景观指数法的有效补充[4];Miller等指出,基于连续平面上欧式距离的假设对于分析受限空间内要素分布的空间特征来说存在一定局限[5];Okabe和Yamada等人将传统的K函数法应用于网络空间结构的地理事件点群研究中,并研发了SANET工具,以支持网络结构下点群分布模式分析[6,7]。网络 K函数可应用于分析沿街分布的犯罪事件[8]和市内经济单元的集聚度[9]等,邬伦等利用单变量与双变量相结合的网络 K函数法分析了香港餐饮业空间格局以及是否受交通站点与旅游景点的影响,探索与分析了不同空间尺度下餐饮店选址和空间分布规律[10];葛莹等将K函数与微观经济学中的边际分析法相结合,估算了长江三角洲地区主要城市的集聚度与边际集聚两个指标,并据此划分城市群的经济空间[11]

城市ATM网点作为重要的便民设施,数量巨大并广泛服务于商业及购物消费场所,然而其空间分布模式等信息仍有待研究挖掘。本文针对南京市ATM网点,基于平面和网络空间两种视角,分别采用单变量、双变量以及局部 K函数等多种方法对其空间分布模式进行分析,研究ATM网点的分布与重要交通站点(如地铁站点)之间的依赖关系,并探讨平面 K函数与网络 K函数对分析结果差异的影响。

1 Ripley’s K函数分析的基本理论

对实际的地理对象点集进行集聚模式分析时,仅仅使用最邻近距离会掩盖结果中的其他模式。为了解决这一问题,Ripley对事件之间的所有距离进行研究,提出了 K函数方法,后来人们将其命名为 Ripley’s K函数。 K函数的优点在于其通过考虑所有点之间的距离关系,来分析整个区域所有空间尺度上的点分布模式。这种方法可以较精确地识别点集在不同空间尺度下的集聚或分散程度。该理论发展至今,形成了许多 Ripley’s K函数的变体,如网络 K函数、单变量与双变量的 K函数、“O-Ring统计量”[12]等,对分布模式可提供更多有用的信息。几种较常用的 K函数介绍如下。

1.1 普通Ripley’s K函数

Ripley’s K函数的定义是:假想在研究区域内放置一系列圆(半径为 d),圆的中心依次放在每一个事件点上,然后计算每一个半径为 d的圆内的事件数量。对所有事件计算其平均值,最后用该平均值除以整个研究区域的事件密度就得到了距离为 d下的K函数值 K(d)。对一系列 d值重复该过程。点包含于圆内可以通过比较事件之间的距离 dij来确定,而研究区域的事件密度可根据区域面积 A和事件数量 n去估计;同时假设 Id为一个标记量,当 dijIdId为1,否则为0。据此化简得到 K(d)的估计函数为:

K(d)=An2ijId(dij)(1)

在“完全空间随机”(Completely Spatial Random, CSR)的零假设条件下,即观测模式与随机模式之间不存在统计上显著的差异[13],很容易确定 K函数的期望值[14]。采用蒙特卡洛模拟检验方法[15],通过设置合理的置信区间,构建包络线。如果观测值位于包络线的置信区间内,则认为该观测值是随机分布模式;如果位于包络线的上限之上,则认为观测值在空间上是集聚的分布模式;而位于包络线下限之下的则为均匀分布模式。

1.2 网络Ripley’s K函数

作为对平面 K函数方法的一种扩展,网络 K函数方法中的距离采用的是网络距离,即两点之间的最短路径距离[16]。在给定的网络空间下,检测每个事件点周围一定距离下的事件点的数量是显著的多(或少),也就是点模式显著地趋向于集聚(或分散)。对于事件p中任意一点 pi, L˜(t|pi)表示到 pi网络距离小于等于 t的点所构成的子网, nt|pi表示落在子网 L˜(t|pi)上除 pi之外的事件点的数量, pi处的网络K函数的定义为[17,18]

K(t|pi)=1ρn(t|pi)(2)

式中, ρ表示 L˜(t|pi)上的点的密度,即 ρ=(n-1)L˜L˜表示网络的长度)。在零假设模型下,在整个网络 L˜上产生 n-1个独立、随机的事件点,则 nt|pi服从一个参数为 n-1L˜(t|pi)L˜的二项分布。根据该二项分布,可以求得 piK函数的期望、方差等统计指标。

接下来考虑 p中所有事件点,可以根据 nt|pi,i=1,,n的平均值来定义整个网络空间下的K函数,于是有:

K(t)=1ρi=1nn(t|pi)n(3)

为了检验零假设,需要计算 K(t)的期望、方差等统计量[19]。在实际应用中,通常还需要进行显著性检验,从而确定某一距离下点模式是趋向于集聚还是分散。

1.3 单变量与双变量K函数

K函数可分为单变量的 K函数与双变量的 K函数两种类型。单变量 K函数方法主要用来研究单一地理点群对象的内在空间格局分布模式,该方法试图仅仅通过地理对象的坐标信息来探索其在空间上的分布模式。双变量 K函数方法主要研究一类事件点的分布模式是否依赖于另外一类事件点,或者一种事件的分布模式是否对另一种事件的分布产生影响。本文的双变量K函数方法主要用来分析商业网点与其他设施点之间,如地铁站点、大中型超市等,是否存在空间上的某种依赖关系。

2 ATM网点空间分布模式分析

2.1 数据来源

本研究采用的数据包括:南京市主城区范围及中心城区边界范围、道路网络、各大中型银行ATM网点数据等。主城区边界是以南京市区影像图(2012版)为底图,以绕城公路和长江岸线为界通过屏幕数字化方法获取的,如图1中外边界所示;中心城区范围主要参照南京城墙分布范围绘制形成,即图1中内部区域的边界。对影像图上标注的主要道路经矢量化和拓扑构建,获得主城区的简单道路网络模型(不考虑道路等级、通行状况等)。银行ATM数据主要选择了南京市主城区内网点数量超过50的大中型商业银行ATM网点,包括工商银行、建设银行等合计749个ATM网点,网点分布与数量见图1表1

图1   南京ATM网点、地铁站点的分布

Fig.1   Distribution of ATMs and metros in Nanjing City

表1   银行ATM网点数量统计

Table 1   Statistics of ATMs in Nanjing City

银行名称ATM网点数量银行名称ATM网点数量
中国工商银行130中国邮政储蓄银行96
中国建设银行117中国银行86
中国农业银行102中国民生银行62
交通银行97招商银行59

新窗口打开

2.2 单变量K函数分析

在南京市主城区范围内,ATM网点的单变量 K函数分析结果如图2a、b所示。从图中可以看出,在平面和网络两种空间视角下,ATM网点分布格局的 K函数曲线均完全位于CSR模式的置信区间之外,说明ATM网点的分布模式和CSR模式存在显著差异。同时 K函数的观测曲线在CSR模式蒙特卡洛模拟的上限之上,证明观测值存在明显的集聚特征,可以判断南京主城区范围内的ATM网点分布在平面和网络空间下均呈现显著的集聚性。通过比较平面和网络 K函数曲线发现,在相同搜索距离下平面 K函数方法的累积点数明显多于网络 K函数方法的累积点数,这反映平面 K函数所表征的ATM网点集聚强度更大。

图2   ATM网点单变量K函数分析
Obs曲线为观测的K函数曲线;Exp(Mean)曲线为随机分布的K函数期望曲线;Exp(Upper5.0%)曲线和Exp(Lower5.0%)曲线分别表示显著性水平为0.05时蒙特卡洛模拟的上限和下限;横轴表示搜索距离d;纵轴表示K函数值与区域内平均强度的积,即到每个点距离小于搜索距离d的累积点个数的平均值
a. 主城区平面; b.主城区网络; c.中心城区平面; d.中心城区网络

Fig.2   Univariate K function analysis for ATMs

在中心城区范围内,ATM网点的平面和网络单变量 K函数分析结果如图2c、d所示。可以看到,中心城区ATM网点的平面 K函数曲线依然表现出显著集聚特性,但集聚强度要小于主城区的整体强度;与主城区的网络 K函数曲线相比,中心城区的曲线离蒙特卡洛模拟的上限更接近,这表明ATM网点在中心城区内也表现出了一定的集聚特征,但这种特征不及在主城区范围内显著。

总体而言,平面 K函数和网络 K函数方法所得出的结果有一定差异,前者对于一些沿着网络分布的事件会产生过度集聚的判断,而在路网空间下这些事件并未表现出相应程度的集聚特性。

2.3 双变量K函数分析

ATM网点在空间上的集聚规律主要与商业设施的分布相关。对于商业中心而言,为有效吸引客流量,更好地实现其商业效益,在进行区位规划时最看重的因素之一就是交通可达性,因此商业中心一般选址于交通便捷的城市道路主干道上。南京市于2005年开始铺设并逐步完善城市轨道交通线路网络,沿路所设的地铁站点位置也主要参考了商业中心所在的位置,因此地铁站点与商业中心之间存在较大程度上的关联。为此,我们进一步针对ATM网点与地铁站点之间关系进行双变量 K函数分析,分别采用平面和网络空间下的双变量 K函数方法,以749家ATM网点和南京地铁1、2号线上35个地铁站点为研究对象,在南京市主城范围进行比较与探索分析。平面双变量 K函数方法通过R语言spatstat包中的“Kcross”方法实现,网络双变量 K函数方法采用“Global cross K function”方法实现,利用R语言进行结果可视化。

平面和网络双变量 K函数方法分析结果如图3a、b所示。图中可以看出,两种方法下ATM网点双变量 K函数曲线均从100 m左右开始位于蒙特卡洛模拟的上限之上,这表明在双变量 K函数曲线上的ATM网点,在地铁站点附近表现出集聚特性,随着距离的增加,其分布与地铁站点之间表现出更显著的集聚关系。从 K函数曲线的增长趋势看,随着距离的增加,曲线增长的速度有所放缓,但仍保持在蒙特卡洛模拟的上限之上,并且观测值与CSR随机模式的期望值之间的差距越来越大,表明ATM网点与地铁站点之间所表现出的空间集聚特性越来越明显。

图3   ATM网点与地铁站点的双变量K函数分析
a.平面; b.网络 (图3图例同图2

Fig.3   Bivariate cross-K function analysis along ATM networks and metro stations

2.4 局部K函数分析

局部 K函数方法可以用来进一步分析地铁站点在局部区域对ATM网点的影响。这里分别选择新街口、奥体东、迈皋桥、孝陵卫4个地铁站点,在网络空间视角下采用局部 K函数方法(Local cross K function method)进行探测分析(图4)。

图4   单个地铁站点的局部网络K函数曲线
a.新街口; b.奥体东站; c.孝陵卫; d.迈皋桥 (图4图例同图2

Fig.4   The local K function for single metro station

图4可以看出,位于不同位置的地铁站点表现出来的分布模式有较大差异。其中,位于城区中心地带的新街口地铁站点,其局部 K函数曲线表明在很小的尺度下其周围的ATM网点分布就表现出非常强烈的集聚特性;而位于中心城区之外的奥体东、迈皋桥、孝陵卫3个地铁站的 K函数曲线,反映其周围的ATM网点几乎未能呈现集聚分布模式。区位因素是形成这种差异的主要原因。作为南京市最繁华的商业中心,新街口的商业区位优势明显;奥体中心、孝陵卫、迈皋桥这3个地铁站远离中心城区,在区位条件上缺乏优势,周边ATM网点的分布亦相对分散。在较小空间尺度(500~600 m)上,孝陵卫与迈皋桥附近的ATM网点呈现出较弱的集聚特征;随着距离的增加,ATM网点与地铁站点的双变量观测值曲线基本位于CSR模式的置信区间之内,呈现出一种随机分布的空间模式。

综合平面与网络的双变量 K函数以及局部双变量 K函数的结果,可以推断出ATM网点在空间分布的总体趋势上与地铁站点存在着集聚的关系。在较大的空间尺度下,双变量 K函数表现出更强的集聚特性,这种集聚特性并非来自两者之间的直接关系,更可能是因为商业网点和地铁站点之间的空间相关关系;在较小的尺度范围内,商业网点在中心城区的地铁站周围表现出显著的集聚特性,而近郊地铁站点附近的商业网点集聚特征相对较弱。商业网点的空间分布模式主要由区域商业发育程度决定,地铁站点对商业网点的布局存在一定影响,但由于地铁线路投入营运时间较短,目前这种影响还比较有限。

3 结语

本文以南京市ATM网点为研究对象,在平面和网络空间两种视角下,采用单变量和双变量 K函数方法,分别对ATM网点在空间上的分布模式及其与地铁站点之间的空间关系进行了分析。研究表明,ATM网点在空间上呈现出显著的集聚特性,其网点分布与地铁站点也存在较明显的依赖关系。不同研究区域(主城区和中心城区)和不同空间视角(平面和网络)的对比分析表明,平面 K函数对于一些沿着网络分布的事件会产生过度集聚的误判,利用网络 K函数法进行空间点模式分析比用平面 K函数法更加符合实际情况。同时,ATM网点集聚特征在主城区范围内要比在中心城区内更显著,反映了不同的研究区范围对分析结果的影响。

The authors have declared that no competing interests exist.


参考文献

[1] Lotwick H W, Silverman B W.

Methods for analysing spatial processes of several types of points

[J]. Journal of the Royal Statistical Society. 1982,44(3): 406-413.

URL      [本文引用: 1]      摘要

Two approaches are described to the analysis of spatial patterns consisting of several types of points. The first approach uses a method of asymptotically unbiased estimation of the second moment distribution; the second uses methods based on regions of empty space in the patterns. Some distributional results are given in each case, and a method of Monte Carlo testing conditional on the marginal structure is described. The methods are illustrated by being applied to some physiological data.
[2] Dixon P M.

Ripley’s K function

[J]. Encyclopedia of environmetrics, 2002, 3: 1796-1803.

https://doi.org/10.1002/9780470057339.var046      URL      [本文引用: 2]      摘要

Ripley's K function summarizes spatial point process data. It can be used to describe a set of locations, test hypotheses about patterns, and estimate parameters in a spatial point process model. For a stationary point process, K ( t ) is the expected number of additional points within distance t of a focal point divided by the intensity of the process. A univariate version is used for one set of locations and a multivariate version is used when points can be labeled by a small number of groups. This article reviews the properties of Ripley's K function and two related functions, then illustrates the computation and interpretation using data on the locations of trees in a swamp hardwood forest.
[3] Ripley B D.

Spatial statistics

[M]. Chichester, U.K.:John Wiley & Sons, 2005.

[本文引用: 1]     

[4] 高凯, 周志翔, 杨玉萍, .

基于Ripley’s K函数的武汉市景观格局特征及其变化

[J]. 应用生态学报, 2010, 21(10): 2621-2626.

URL      [本文引用: 1]      摘要

基于1987、1996和2007年武汉市tm遥感影像,应用点格局分析方法之一的ripley-k函数,分析了武汉市市域景观格局的特征及其变化.结果表明:1987—2007年间,耕地是武汉市的景观基质,而林地、水体、草地、城乡建设用地和其他未利用地则以斑块或廊道的形式镶嵌其中;林地、水体、草地和城乡建设用地在所有研究尺度下均呈现出显著的聚集空间格局,且林地、草地和城乡建设用地的空间聚集性强于水体;耕地在小尺度下呈现出聚集的空间格局,随着尺度的增大,逐渐变为随机、均匀的空间格局.研究期间,武汉市林地、城乡建设用地面积逐次增大,水体、草地和耕地面积逐次减少;各景观类型的空间格局特征也发生了改变,总体表现为林地和城乡建设用地空间聚集度降低,分布的均匀程度增加,而草地、水体和耕地的空间分布变得更加不均匀,聚集程度增加.与样方法、样线(带)法相比,利用样点分析景观格局具有简单、准确、易用等优点.ripleyk函数是多尺度景观格局分析的有效手段,是景观指数法的支持和补充.

[Gao Kai, Zhou Zhixiang, Yang Yuping et al.

Characteristics and changes of landscape pattern in Wuhan City based on Ripley’s K function

.Chinese Journal of Applied Ecology, 2010, 21(10): 2621-2626.]

URL      [本文引用: 1]      摘要

基于1987、1996和2007年武汉市tm遥感影像,应用点格局分析方法之一的ripley-k函数,分析了武汉市市域景观格局的特征及其变化.结果表明:1987—2007年间,耕地是武汉市的景观基质,而林地、水体、草地、城乡建设用地和其他未利用地则以斑块或廊道的形式镶嵌其中;林地、水体、草地和城乡建设用地在所有研究尺度下均呈现出显著的聚集空间格局,且林地、草地和城乡建设用地的空间聚集性强于水体;耕地在小尺度下呈现出聚集的空间格局,随着尺度的增大,逐渐变为随机、均匀的空间格局.研究期间,武汉市林地、城乡建设用地面积逐次增大,水体、草地和耕地面积逐次减少;各景观类型的空间格局特征也发生了改变,总体表现为林地和城乡建设用地空间聚集度降低,分布的均匀程度增加,而草地、水体和耕地的空间分布变得更加不均匀,聚集程度增加.与样方法、样线(带)法相比,利用样点分析景观格局具有简单、准确、易用等优点.ripleyk函数是多尺度景观格局分析的有效手段,是景观指数法的支持和补充.
[5] Miller H J.

Market area delimitation within networks using geographic information systems

[J]. Geographical Systems,1994, 1(2): 157-173.

URL      [本文引用: 1]      摘要

Abstract Market area delimitation is important for assessing a commercial facility's performance or for identifying candidates for new facility location. However, current market area delimitation methods rely on an unrealistic representation of space as planar and isotropic. This paper develops generic procedures for conducting market area delimitation within street network structures using geographic information systems (GIS). First, this paper formulates network based versions of the Huff visit probability model. Then, restatement of these models in the format of relational databases and structured query language (SQL) provides a generic algorithm that bridges the gap between mathematics and the capabilities of a computerized information system. Finally, this paper discusses general implementation issues and a case study using a commercial GIS package. -Author
[6] Okabe A, Yamada I.

The K-function method on a network and its computational implementation

[J]. Geographical Analysis, 2001, 33(3): 271-290.

https://doi.org/10.1111/j.1538-4632.2001.tb00448.x      URL      [本文引用: 1]      摘要

Abstract This paper proposes two statistical methods, called the network K-function method and the network cross K-function method, for analyzing the distribution of points on a network. First, by extending the ordinary K-function method defined on a homogeneous infinite plane with the Euclidean distance, the paper formulates the K-function method and the cross K-function method on a finite irregular network with the shortest-path distance. Second, the paper shows advantages of the network K-function methods, such as that the network K-function methods can deal with spatial point processes on a street network in a small district, and that they can exactly take the boundary effect into account. Third, the paper develops the computational implementation of the network K-functions, and shows that the computational order of the K-function method is O(n2Q log nQ) and that of the network cross K-function is O(nQ log U3Q), where nQ is the number of nodes of a network.
[7] Okabe A, Sugihara K.

Spatial Statistics along Networks: Statistical and Computational Methods

[M]. Chichester, U.K.: John Wiley & Sons, 2012.

[本文引用: 1]     

[8] Lu Y, Chen X.

On the false alarm of planar K-function when analyzing urban crime distributed along streets

[J]. Social Science Research, 2007, 36(2): 611-632.

https://doi.org/10.1016/j.ssresearch.2006.05.003      URL      [本文引用: 1]      摘要

Abstract Many social and economic activities, especially those in urban areas, are subject to location restrictions imposed by existing street networks. To analyze the spatial patterns of these urban activities, the restrictions imposed by the street networks need to be taken into account. K-function is a method commonly used for general point pattern analysis as well as crime pattern study. However, applying the planar K-function to analyze the spatial autocorrelation patterns of urban activities that are typically distributed along streets could result in false alarm problems. Depending on the nature of the urban street networks and the distribution of the urban activities, either positive or negative false alarm might be introduced. Acknowledging that many urban crimes are typically distributed along streets, this paper compares the traditional planar K-function with a network K-function for crime pattern analysis. The patterns of vehicle thefts in San Antonio, Texas are examined as a case study.
[9] Garrocho-Rangel C, Álvarez-Lobato J A, Chávez T.

Calculating Intraurban Agglomeration of Economic Units with Planar and Network K-Functions: A Comparative Analysis

[J]. Urban Geography, 2013, 34(2): 261-286.

https://doi.org/10.1080/02723638.2013.778655      URL      [本文引用: 1]      摘要

ABSTRACT According to recent research, one of the most promising strategies for intraurban job growth lies promoting localized clusters that produce goods and services which are primarily sold within a single city, metropolitan area, or urban region. However, in order to design urban policies to create or reinforce local clusters, the first challenge is to measure in a reliable way the clustering tendencies of different kinds of economic units in intraurban space. The aim is to compare the similarities and differences in results obtained from two methods designed to measure global clustering tendencies (the planar and network K-functions) in terms of characterization, scale, and intensity of intraurban localization patterns for tertiary economic units in a Latin American metropolis. It is concluded that the network K-function is a more appropriate method for measuring agglomeration patterns, scale, and intensity at the intra-urban level.
[10] 邬伦, 刘亮, 田原, .

基于网络K函数法的地理对象分布模式分析——以香港岛餐饮业空间格局为例

[J]. 地理与地理信息科学, 2013, 29(5): 7-11.

https://doi.org/10.7702/dlydlxxkx20130502      URL      [本文引用: 1]      摘要

很多地理对象的空间分布与空间上呈现网状结构的地理现象高度相 关,分析这些地理对象的分布模式,在地理研究中有重要意义.该文采用由平面空间扩展到网状结构空间的网络K函数法,以香港岛餐饮业地理空间格局为例开展研 究.应用单变量K函数法分析餐饮店在网状结构空间中的分布模式,应用双变量交叉K函数法分析餐饮店分布是否受交通站点及旅游景点影响,并对不同尺度下餐饮 店地理选址和空间分布规律进行探索与分析.

[Wu Lun, Liu Liang, Tian Yuan et al.

Spatial Pattern Analysis of Geographic Feature Using Network K—Function Methods with a Case Study of Restaurant Distribution in Hong Kong Island. Geography and

Geo-Information Science, 2013, 29(5): 7-11. ]

https://doi.org/10.7702/dlydlxxkx20130502      URL      [本文引用: 1]      摘要

很多地理对象的空间分布与空间上呈现网状结构的地理现象高度相 关,分析这些地理对象的分布模式,在地理研究中有重要意义.该文采用由平面空间扩展到网状结构空间的网络K函数法,以香港岛餐饮业地理空间格局为例开展研 究.应用单变量K函数法分析餐饮店在网状结构空间中的分布模式,应用双变量交叉K函数法分析餐饮店分布是否受交通站点及旅游景点影响,并对不同尺度下餐饮 店地理选址和空间分布规律进行探索与分析.
[11] 葛莹, 蒲英霞, 赵慧慧, .

基于边际K函数的长三角地区城市群经济空间划分

[J]. 地理学报, 2015,70(4), 528-538.

https://doi.org/10.11821/dlxb201504002      URL      [本文引用: 1]      摘要

基于空间点模式分析的Ripley's K函数,结合微观经济学的边际分析法,通过集聚度和边际集聚两个指标,从理论上探索城市群集聚效应的定量测度方法,提出一种城市群的经济空间划分方法。本 文的不同在于,多尺度估算城市的集聚度和边际集聚,以城市的边际集聚极值点时的城市区域布局模式为最优,据此划分城市群的经济空间,并以长江三角洲地区为 例,对本文提出的方法进行实证。研究结果表明:①城市的集聚度估计显示,长三角地区2010年城市空间布局为随机分布型,但随着观测尺度的增加,城市的集 聚度呈快速上升趋势。②边际集聚估计揭示,当城市区位和城市规模的集聚尺度分别为173 km和185 km时,城市区位或规模的集聚效应达到峰值,此时长三角地区城市空间布局出现最优模式。③空间聚类分析展示,在城市区域布局的最优空间模式下,长三角地区 呈现“中心—外围”的经济空间结构,高集聚度子群是区域经济发展中心,全部位于上海经济辐射圈,而低集聚度子群是外围欠发达地区,全部位于区际行政边界, 暗示边际负效应仍阻碍着地区内外人员的往来。

[Ge Ying, Pu Yingxia, Zhao Huihui et al.

Dividing economic space into urban agglomerations usingthe marginal K-function:A case study of the Yangtze River Delta region

. Acta Geographica Sinica,2015,70(4), 528-538. ]

https://doi.org/10.11821/dlxb201504002      URL      [本文引用: 1]      摘要

基于空间点模式分析的Ripley's K函数,结合微观经济学的边际分析法,通过集聚度和边际集聚两个指标,从理论上探索城市群集聚效应的定量测度方法,提出一种城市群的经济空间划分方法。本 文的不同在于,多尺度估算城市的集聚度和边际集聚,以城市的边际集聚极值点时的城市区域布局模式为最优,据此划分城市群的经济空间,并以长江三角洲地区为 例,对本文提出的方法进行实证。研究结果表明:①城市的集聚度估计显示,长三角地区2010年城市空间布局为随机分布型,但随着观测尺度的增加,城市的集 聚度呈快速上升趋势。②边际集聚估计揭示,当城市区位和城市规模的集聚尺度分别为173 km和185 km时,城市区位或规模的集聚效应达到峰值,此时长三角地区城市空间布局出现最优模式。③空间聚类分析展示,在城市区域布局的最优空间模式下,长三角地区 呈现“中心—外围”的经济空间结构,高集聚度子群是区域经济发展中心,全部位于上海经济辐射圈,而低集聚度子群是外围欠发达地区,全部位于区际行政边界, 暗示边际负效应仍阻碍着地区内外人员的往来。
[12] Li L, Huang Z, Ye W et al.

Spatial distributions of tree species in a subtropical forest of China

[J]. Oikos, 2009, 118(4): 495-502.

https://doi.org/10.1111/j.1600-0706.2009.16753.x      URL      [本文引用: 1]      摘要

ABSTRACT The spatial dispersion of individuals in a species is an important pattern that is controlled by many mechanisms. In this study we analyzed spatial distributions of tree species in a large-scale (20 ha) stem-mapping plot in a species-rich subtropical forest of China. O-ring statistic was used to measure spatial patterns of species with abundance >10. 惟0鈥10, the mean conspecific density within 10 m of a tree, was used as a measure of the intensity of aggregation of a species. Our results showed: (1) aggregated distribution was the dominant pattern in the plot. The percentage of aggregated species decreased with increased spatial scale. (2) The percentages of significantly aggregated species decreased from abundant to intermediate and to rare species. Rare species was more strongly aggregated than common species. Aggregation was weaker in larger diameter classes. (3) Seed traits determined the spatial patterns of trees. Seed dispersal mode can influence spatial patterns of species, with species dispersed by both modes being less clumped than species dispersed by animal or wind, respectively. Considering these results, we concluded that seed dispersal limitation, self-thinning and habitat heterogeneity primarily contributed to spatial patterns and species coexistence in the forest.
[13] Wiegand T, Kisslin W D, Cipriotti P A et al.

Extending point pattern analysis for objects of finite size and irregular shape

[J]. Journal of Ecology, 2006, 94(4): 825-837.

https://doi.org/10.1111/j.1365-2745.2006.01113.x      URL      [本文引用: 1]      摘要

1 We use a grid- and simulation-based approach to extend point pattern analysis to deal with plants of finite size and irregular shape, and compare the results of our approach with that of the conventional point approximation. The plants are approximated by using an underlying grid and may occupy several adjacent grid cells depending on their size and shape. Null models correspond to that of point pattern analysis but need to be modified to account for the finite size and irregular shape of plants.
[14] Wiegand T, A Moloney K.

Rings, circles, and null-models for point pattern analysis in ecology

[J]. Oikos, 2004, 104(2): 209-229.

[本文引用: 1]     

[15] Besag J, Diggle P J.

Simple Monte Carlo tests for spatial pattern

[J]. Applied Statistics, 1977,26(3):327-333.

https://doi.org/10.2307/2346974      URL      [本文引用: 1]      摘要

ABSTRACT The Monte Carlo approach to testing a simple null hypothesis is reviewed briefly and several examples of its application to problems involving spatial distributions are presented. These include spatial point pattern, pattern similarity, space-time interaction and scales of pattern. The aim is not to present specific “recommended tests” but rather to illustrate the value of the general approach, particularly at a preliminary stage of analysis.
[16] Spooner P G, Lunt I D, Okabe A et al.

Spatial analysis of roadside Acacia populations on a road network using the network K-function

[J]. Landscape Ecology, 2004, 19(5): 491-499.

https://doi.org/10.1023/B:LAND.0000036114.32418.d4      URL      [本文引用: 1]      摘要

Spatial patterning of plant distributions has long been recognised as being important in understanding underlying ecological processes. Ripley’s K-function is a frequently used method for studying the
[17] Okabe A, Kitamura M.

A computational method for market area analysis on a network

[J]. Geographical Analysis, 1996, 28(4): 330-349.

https://doi.org/10.1111/j.1538-4632.1996.tb00939.x      URL      [本文引用: 1]      摘要

ABSTRACT This paper shows a computational method for market area analysis assuming that stores and consumers are distributed over a network and the distance between two points on the network is given by the route distance. First, we consider five basic questions often raised in market area analysis, and show a general method, called the network transformation method, that gives an intuitive way of looking at computational methods for solving these questions. Second, assuming that consumers follow the Huff model, we consider four questions concerning market area delineation and market potential often discussed in market area analysis in practice. We show that the network transformation method is also useful to develop computational methods for solving these questions. One of the notable results is that market area delineation of the Huff model (which is analytically difficult to obtain on a plane) can be exactly obtained on a network.
[18] 杨珏婕, 刘世梁, 赵清贺, .

基于网络K函数的西双版纳人工林空间格局及动态

[J]. 生态学报, 2011, 31(22): 6734-6742.

URL      [本文引用: 1]      摘要

Abstract 区域植被格局的分布特征受诸多要素影响,但其空间格局和动态具有一定规律或自相关性,道路网络作为景观中显著的人工线性要素,在很大程度上影响着区域的植被格局特征,特别是人工植被的分布特征。运用网络K函数,分析了道路网络和人工林空间格局分布的相互关系,并且用二元网络K函数研究了人工林扩展对针叶林和阔叶林的影响。结果表明:人工林在1970—2000年间种群分布格局有非常明显的变化,特别是从1990到2000年,种群面积不断扩大,主要从北部地区扩展到西北和东南地区。1970—1990年人工林扩展主要集中在低海拔的道路网络附近,沿道路网络呈现明显的集聚分布,公路效应明显。但后期逐渐向距公路较远、海拔较高的地区扩展,到2000年在大尺度下人工林斑块呈显著随机分布。同时,人工林面积的增长对针叶林影响显著,对阔叶林有影响但是并不显著。二元网络K函数表明,在1970到1990年人工林与针叶林沿道路网络在小尺度为负关联,在局部地区存在着竞争,但在大尺度上对环境条件的要求具有一致性为正关联。到2000年,在大尺度上人工林与针叶林的种群分布格局呈显著负相关,人工林面积的不断扩展导致了针叶林面积的下降。

[Yang Juejie, Liu Shiliang, Zhao Qinghe et al.

Spatial and dynamic analysis of plantations in Xishuangbanna using network K-function

. Acta Ecologica Sinica, 2011, 31(22): 6734-6742. ]

URL      [本文引用: 1]      摘要

Abstract 区域植被格局的分布特征受诸多要素影响,但其空间格局和动态具有一定规律或自相关性,道路网络作为景观中显著的人工线性要素,在很大程度上影响着区域的植被格局特征,特别是人工植被的分布特征。运用网络K函数,分析了道路网络和人工林空间格局分布的相互关系,并且用二元网络K函数研究了人工林扩展对针叶林和阔叶林的影响。结果表明:人工林在1970—2000年间种群分布格局有非常明显的变化,特别是从1990到2000年,种群面积不断扩大,主要从北部地区扩展到西北和东南地区。1970—1990年人工林扩展主要集中在低海拔的道路网络附近,沿道路网络呈现明显的集聚分布,公路效应明显。但后期逐渐向距公路较远、海拔较高的地区扩展,到2000年在大尺度下人工林斑块呈显著随机分布。同时,人工林面积的增长对针叶林影响显著,对阔叶林有影响但是并不显著。二元网络K函数表明,在1970到1990年人工林与针叶林沿道路网络在小尺度为负关联,在局部地区存在着竞争,但在大尺度上对环境条件的要求具有一致性为正关联。到2000年,在大尺度上人工林与针叶林的种群分布格局呈显著负相关,人工林面积的不断扩展导致了针叶林面积的下降。
[19] Xie Z, Yan J.

Detecting traffic accident clusters with network kernel density estimation and local spatial statistics: An integrated approach

[J]. Journal of Transport Geography, 2013, 31, 64-71.

https://doi.org/10.1016/j.jtrangeo.2013.05.009      URL      [本文引用: 1]      摘要

Kernel density estimation (KDE) has long been used for detecting traffic accident hot spots and network kernel density estimation (NetKDE) has proven to be useful in accident analysis over a network space. Yet, both planar KDE and NetKDE are still used largely as a visualization tool, due to the missing of quantitative statistical inference assessment. This paper integrates NetKDE with local Moran’I for hot spot detection of traffic accidents. After density is computed for road segments through NetKDE, it is then used as the attribute for computing local Moran’s I. With an NetKDE-based approach, conditional permutation, combined with a 100-m neighbor for Moran’s I computation, leads to fewer statistically significant “high-high” (HH) segments and hot spot clusters. By conducting a statistical significance analysis of density values, it is now possible to evaluate formally the statistical significance of the extensiveness of locations with high density values in order to allocate limited resources for accident prevention and safety improvement effectively.

/