1. School of Resource and Environmental Sciences, Wuhan University, Wuhan 430079, China
2. Key Laboratory of Geographic Information System, Ministry of Education, Wuhan University, Wuhan 430079, China
3. State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan 430079, China
National Natural Science Foundation of China, No.41571385;
Different urban elements may exhibit various aggregation patterns. It is of great significance to quantitatively investigate the disparity and connection among various aggregation patterns of urban elements for understanding the mechanism of urban development and supporting urban planning. Taking Wuhan city, Central China, as a case, we collected five types of urban elements, namely, impervious surface, population density, Point of Interest (POI), plot ratio, and road network, to explore their spatial distribution and characteristics of aggregation patterns. We first used Kernel Density Estimation (KDE) method to identify city centers and we found that there is one major center and seven sub-centers in Wuhan. Then we partitioned the study area by gradient analysis, and calculated the densities of urban elements. The density of urban element decreases outward from the centers, which is fitted well using "Inverse S-shaped" function. We used Concentration Degree Index (CDI) to reflect the aggregation degree of urban elements. The results indicate that the degrees of the aggregation of urban elements are: Commercial POI > population > plot ratio > road. Commercial POI and population are highly aggregated in the urban core area, while plot ratio and road are moderately aggregated in the urban core area. The spatial patterns of highly aggregated commercial POI and population are the combination of point pattern and zonal pattern, while the spatial patterns of moderately aggregated road network and plot ratio are the combination of point pattern, ring pattern and axial pattern. As for the lowly aggregated impervious surface, it shows a ring pattern. The attraction effect of city centers and vertical development promote the center-oriented agglomeration of urban elements. At the same time, the transportation lines and corridor effects of the scarce landscape resources reshape the spatial distribution pattern of urban elements.
本文以典型的多中心城市武汉市为例,选取不透水表面、人口密度、兴趣点（Point Of Interest, POI）、容积率、道路密度等数据来研究城市要素的空间集聚特征,运用核密度估计法定量识别武汉市城市主次中心,结合梯度分析法划定城市圈层并计算各圈层内城市要素密度,定量拟合各圈层城市要素密度分布形态,采用集聚度指数划分各要素集聚强度类型,分析各要素空间分布和集聚特征,探讨各要素空间分布形成的机理,有助于理解城市要素集聚规律并为城市规划提供决策支持。
JacobsJ.The Death and Life of Great American Cities. , 1961.
LiJinyan.Urban agglomeration: Theory and evidence [D]. , 2008.
[李金滟. 城市集聚:理论与证据[D]. , 2008.]
BattyM, Kim KS.Form follows function: Reformulating urban population density functions. , 1992, 29(7): 1043-1069.
In this paper, we argue that the most appropriate form for urban population density models is the inverse power function, contrary to conventional practice, which is largely based upon the negative exponential. We first show that the inverse power function has several theoretical properties which have hitherto gone unremarked in the literature. Our main argument, however, is based on the notion that a density function should describe the extent to which the space available for urban development is filled. To this end, we introduce ideas from urban allometry and fractal geometry to demonstrate that the inverse power model is the only function which embodies the fractal property of self-similarity which we consider to be a basic characteristic of urban form and density. In short, we show that the distance parameter 伪 of the inverse power model is a measure of the extent to which space is filled, and that its value is determined by the basic relation D + 伪 = 2 where is the fractal dimension of the city in question. We then test this model using four data sets which measure the density and morphology of the city of Seoul. Using a variety of estimation methods such as loglinear regression, dimensional approximation and entropy-maximising, we estimate dimension and density parameters for 136 variants of the function and its data sets. From these estimates, 125 are within the values hypothesised and this suggests fairly conclusively that the value of the density parameter 伪 for the inverse power function should be within the range 0鈥1 and probably between 0.2 and 0.5. Many related questions are raised by this analysis which will form the subject of future research.
EllisonG, Glaeser EL.Geographic concentration in U.S. manufacturing industries: A Dartboard approach. , 1997, 105(5): 889-927.
This paper discusses the prevalence of Silicon Valley-style localizations of individual manufacturing industries in the United States. Several models in which firms choose locations by throwing darts at a map are used to test whether the degree of localization is greater than would be expected to arise randomly and to motivate a new index of geographic concentration. The proposed index controls for differences in the size distribution of plants and for differences in the size of the geographic areas for which data is available. As a consequence, comparisons of the degree of geographic concentration across industries can be made with more confidence. We reaffirm previous observations in finding that almost all industries are localized, although the degree of localization appears to be slight in about half of the industries in our sample. We explore the nature of agglomerative forces in describing patterns of concentration, the geographic scope of localization, and the extent to which agglomerations involve plants in similar as opposed to identical industries.
MarconE, PuechF.Evaluating the geographic concentration of industries using distance-based methods. , 2003, 3(4): 409-428.
We propose new methods for evaluating the spatial distribution of firms. To assess whether firms are concentrated or dispersed, economists have tradi-tionally used indices that analyse the heterogeneity of a spatial structure at a single geographic level. We introduce distance-based methods, Besag's L function (derived from Ripley's K function) and Diggle and Chetwynd's D function to describe simultaneously spatial distribution at different geographical scales. Our empirical applications consider the distribution of French manufacturing firms in the Paris area and in France generally. For some geographic levels, results show significant concentration or dispersion of firms according to their sector of activity.
HuallcháinB.Agglomeration of services in American metropolitan areas. , 1989, 20(3): 34-49.
The growth of many service industries among American metropolitan areas stem from an eclectic set of forces. These include market penetration effects of increasing importance of services throughout the economy, agglomeration effects in immature and deregulated industries, and institutional and infrastructure constraints. These diverse forces are interpreted as urbanization and localization economies. A cycle of centralization of 27 fast growing service industries is documented for MS As in the period 1977-84. The two agglomeration effects are tested directly, using a power function model that relates employment and establishment growth to MSA size and initial level of employment in a local industry. Localization economies rather than the general advantages of metropolitan size best explain the growth patterns. This result implies that service industrial complexes are rapidly emerging in American metropolitan areas.
Han SS, QinB.The spatial distribution of producer services in Shanghai. , 2009, 46(4): 877-896.
This paper examines the location characteristics of the newly emerged producer services in Shanghai. The research questions are: how does the location of producer services fit into Shanghai's spatial context? What are the factors shaping the spatial distribution pattern? Are the spatial distribution and its determinants similar to those observed in other cities? Analyses of data collected from field reconnaissance, interviews and secondary sources reveal that the spatial distribution of producer services in Shanghai is characterised by concentration and dispersion. The concentration of producer services in an extensive central-city core is similar to that observed in other cities. However, the wide spread of producer services over a fairly big region is unexpected. The paper adds insight to the study of producer services by contending that contextual factors such as an indigenous path of development, state intervention and market institution, determine the location of producer services in Shanghai.
LiShaoying, WuZhifeng, LiBiying, et al.The spatial and temporal characteristics of residential floor area ratio in metropolitan at multi-scales based on Internet real estate data: Case study of Guangzhou. , 2016,35(4): 770-780.
The spatial information of urban land use in Changchun city, China in 1900,1930,1954,1976,1990 and 2004 is extracted by integrating SPOT5 images, relief maps and urban planning maps and other spatial data based on remote sensing and GIS. The land use information entropy and fractal dimension and mechanism of urban spatial evolving process is analyzed in Changchun city in the 100 years based information entropy, fractal dimension and spatial variation model. The results show that the land use information entropy increased except for 1990 and urban land types diversifies and its structure became balanced in Changchun in the 100 years. The Spatial variation function of land use information entropy accord with exponential model. The ratio of nugget and sill increases, which indicates its spatial auto-correlation decreased and the homogeneous character became weakened. Commercial land gathers toward the center of the internal circle. By contrary, industry land clusters in the urban fringe. The fractal dimension of urban evolving indicates the character of self-organizing. Social regime, urban roads, urban land price and urban planning are the important factors to the land use information entropy and fractal dimension of urban spatial evolving process.
Jiao LM.Urban land density function: A new method to characterize urban expansion. , 2015, 139: 26-39.
Density analysis lies at the core of studies on urban expansion; however, many methods in urban land density analysis are arbitrary and suffer from the lack of an established foundation. We observed an 鈥淚nverse S-shape Rule鈥 for urban land density that varies outward from an urban center by investigating 28 major cities in China at three time points. We proposed an inverse S-shaped function to formulate urban land density, which fit well for all of the cities in our sample using a nonlinear least squares fitting method. The parameters of the function explicitly describe the basic properties of an urban form. Based on the fitted functions, we derived an established method for the concentric partitioning of urban area and further proposed indicators to measure the urban compactness, urban expansion rate, and degree of urban sprawl. These indicators are practical for characterizing urban form and urban sprawl for either a single city or for multiple cities. A case study on major Chinese cities from 1990 to 2010 reveals that most of the cities expanded rapidly and became less compact and more dispersed during those two decades. However, most of the cities grew faster and showed more sprawl in the second decade compared to the first one. Discussions show that the model is also applicable for non-monocentric cities and possibly can be applied to many other geographical phenomena.
城市设施兴趣点(POI)在局部地理空间下往往呈现聚集型分布特征(即热点),表达该特征的核密度法(kernel density estimation)是最常用到的可视化工具。考虑到核密度方法中缺少量化统计分析,提出了一种城市设施POI分布热点探测的新方法。首先基于"距离衰减效应"计算地理单元的属性值;然后采用Getis-Ord <i>G</i><sub><i>i</i></sub><sup>*</sup>统计指数定量分析设施POI点的局部空间相关性特征。与传统基于样方法的空间自相关相比,核密度法由于顾及了地理学第一定律的区位影响,计算获得的地理单元属性值可保留空间的细节信息,热点的空间自相关分析结果可以反映设施服务影响的连续性特征。通过实际金融设施数据的自相关分析实验,表明该方法能有效提取POI基础设施在城市区域中的分布热点范围。
FlahautB, MouchartM, San ME, et al.The local spatial autocorrelation and the kernel method for identifying black zones:A comparative approach., 2003, 35(6): 991-1004.
This article aims to determine the location and the length of road sections characterized by a concentration of accidents (black zones). Two methods are compared: one based on a local decomposition of a global autocorrelation index, the other on kernel estimation. After explanation, both methods are applied and compared in terms of operational results, respective advantages and shortcomings, as well as underlying conceptual elements. The operationality of both methods is illustrated by an application to one Belgian road.
Over the past two decades, metropolises in China witnessed unprecedented growth rates in urban areas. The rapid growth of urban space could lead to a series of problems. The fact that China has the largest population manifests the negative consequences of urban expansion. A rigorous and quantitative comparison of urban growth among metropolitan areas will lend basic support to policy making on regional urban development and will help city planners, economists, environmentalists, etc. This study analyzed the spatiotemporal patterns of urban expansion at different distance from city center in 27 metropolises in China using multi-temporal remotely sensed data. We use three satellite images obtained from the Landsat Thematic Mapper (TM) and Enhanced Thematic Mapper Plus (ETM+) sensor system circa 1990, 2000 and 2010 for each city to develop maps of urban areas. A concentric partitioning method was employed to define urban core area and urban fringe on the basis of built-up density calculated in a series of 1km concentric rings. Urbanization Concentricity Index was proposed to measure the degree to which residential or non-residential development is close to urban core area. Then we calculated and analyzed Urbanization Intensive Index across a series of 1km buffer zones for each city to identify the location-based spatiotemporal patterns of urbanization. UII offers a measure of speed and intensity of urban expansion. The results showed that the spatial structure of these cities was closely related to national policies for regional development, demonstrating obvious regional characteristics. In general, the UCI value of the cities in western and northeastern China is higher in central and eastern China. Most cities exhibited lower UCI value in 2010 compared with 1990, showing a tendency toward dispersion. The highest UII for each city usually occurs around the urban core area and move outward over time. The UII value for second- and third-tier cities became quite small in rings more than 20 km away from city center. The area close to city center witnessed higher UII value in the first decade as compared with the second one. On the contrary, the UII value for the buffer zones away from the city center was much larger in the second decade. Urban form also correlated to the level of economic development to some extent. In the first decade, first-tier cities expanded more intensely and broadly than other cities. Some Second-and third-tier cities experienced significant urban expansion in the second decade, demonstrating a tendency towards decentralization in their development pattern. In 1990, almost all metropolises in China exhibited a trend of comprising one single large central area for the entire landscape. With the expansion of urban land and rapid new nuclei development, some developed cities exhibited a transition of urban space from being mono-centric to multi-nucleated in form, with the emergence of some sub-centers. However, some relatively less developed cities still retained their mono-centric urban space. It can be inferred that urban structure tends to transform from being monocentric to polycentric with the continuous development of economy.
Yu WH, Ai TH, Shao SW.The analysis and delimitation of Central Business District using network kernel density estimation. , 2015, 45: 32-47.
Central Business District (CBD) is the core area of urban planning and decision management. The cartographic definition and representation of CBD is of great significance in studying the urban development and its functions. In order to facilitate these processes, the Kernel Density Estimation (KDE) is a very efficient tool as it considers the decay impact of services and allows the enrichment of the information from a very simple input scatter plot to a smooth output density surface. However, most existing methods of density analysis consider geographic events in a homogeneous and isotropic space under Euclidean space representation. Considering the case that the physical movement in the urban environment is usually constrained by a street network, we propose a different method for the delimitation of CBD with network configurations. First, starting from the locations of central activities, a concentration index is presented to visualize the functional urban environment by means of a density surface, which is refined with network distances rather than Euclidean ones. Then considering the specialties of network distance computation problem, an efficient way supported by flow extension simulation is proposed. Taking Shenzhen and Guangzhou, two quite developed cities in China as two case studies, we demonstrate the easy implementation and practicability of our method in delineating CBD.
Leslie TF.Identification and differentiation of urban centers in phoenix through a multi-criteria kernel-density approach. , 2010, 33(2): 205-235.
Research concerning geographical centers of economic activity has sought to explain patterns of development and interaction in cities. This article presents a new method of defining intraurban centers within a spatial economic framework as a combination of both employment and establishment kernel-smoothed patterns. The method is applied in Phoenix, a postmodern metropolis that has grown by more than a factor of thirty between 1950 and 2005 and is one of the largest and fastest growing metropolitan areas in the United States. Centers are found to vary considerably in their sectoral composition and are grouped based on their focus of secondary, retail, or high-order emphasis. A conditional logit model is used to show how each center differentiates with regard to establishment size and sector as well as the importance of center characteristics.
Cladera JR, Duarte C R M, Moix M. Urban structure and polycentrism: Towards a redefinition of the Sub-centre concept. , 2009, 46(13): 2841-2868.
VeneriP.The identification of sub-centres in two Italian metropolitan areas: A functional approach. , 2013, 31(343): 177-185.
This work introduces a method aimed at identifying metropolitan sub-centres based on an interaction approach, using commuting flows. In addition, it compares the results with other approaches based on employment and/or density thresholds commonly used in the literature. Although the scientific literature and the policy debate on polycentricity are increasingly devoted to its functional dimension, empirical research aimed at identifying metropolitan sub-centres mainly relies on employment density, which refers to a more morphological approach. After discussing the concept of centre and the contemporary literature on its empirical identification, the proposed methodology is applied to the metropolitan areas of Rome and Milan. Such a method shows a better fit compared to other approaches based on job density. Results are found to be remarkably sensitive to the method applied and are dependent, to some extent, on the different concept of centre that is adopted in the characterisation of metropolitan spatial structures.
Sun YR, Fan HC, LiM, et al.Identifying the city center using human travel flows generated from location-based social networking data. , 2015, 43(3): 480-498.
Since cities have become more complex and some large cities are likely to be polycentric, a better understanding of cities requires a clear topology that reveals how city centers are spatially distributed and interacted with. The identification of a city center that aims to find the accurate location of the city center or delineate the city center with a precise boundary becomes vital. This work attempts to achieve this by using a new type of movement data generated from location-based social networks, whereby three different methods are deployed for clustering and compared regarding identification of city centers and delineation of their boundaries. Experiments show that city centers with precise boundaries can be identified by using the proposed approach with location-based social network data. Furthermore, the results show that the three methods for clustering have different advantages and disadvantages during the process of city center identification, and thus seem to be suitable for cities with different urban structures.
Tresserra JM.Does employment density death? Towards a new integrated methodology to identify and characterize sub-centres//52nd Congress of the European Regional Science Association: "Regions in Motion - Breaking the Path", August 21-25, 2012. Bratislava, , 2012.
MaZuopeng, LiChenggu, ZhangJing, et al.The relationship between different types of traffic route and functional urban land use in Changchun. , 2016, 35(9): 1687-1700.