Acting as “networking 'nodes'” (
Journal of Geographical Sciences, Volume. 30, Issue 7, 1060(2020)
Mega-towns in China: Their spatial distribution features and growth mechanisms
As a special outcome of urbanization, mega-towns not only play an important role in the process of socio-economic development, but also are important contributors to urbanization. Based on a spatial database of mega-towns in China, this paper explores the spatial distribution features and growth mechanisms of China’s 238 mega-towns using the nearest neighbour distance method, kernel density estimation, regression analysis, global autocorrelation, local autocorrelation and other spatial analysis methods. Results of spatial distribution features show that: (1) on the national scale, the existing 238 mega-towns mainly gathered in the southeast coastal areas of China; they formed two spatial core agglomerations, several secondary ones and a southeast coastal agglomeration belt; (2) on the regional scale, each economic region’s index was less than 1, indicating that mega-towns in each region tended to be spatially agglomerated due to the close relationship with regional development level and their number; (3) on the provincial scale, 68% of provincial-level units in China tended to be a spatial agglomeration of mega-towns; only one province had a random distribution; the number of mega-towns in those evenly-distributed provinces was generally small. The growth of mega-towns was determined by a combination of various natural and humanistic factors, including topography, location, economy, population, traffic, and national policy. This paper chose digital elevation model (DEM), location advantage, economic density, population density, and highway density distribution as corresponding indicators as quantitative factors. By combining their local autocorrelation analysis, these factors all showed certain influence on the spatial growth of mega-towns and together scheduled it. In the future, provinces and cities should make full use of the mega-town functions to promote their socioeconomic development, especially the central and western regions in China.
1 Introduction
Acting as “networking 'nodes'” (
Mega-town is a special form of socio-economic development that often emerges in the process of rural industrialization and urbanization (
China has a huge land mass, and its natural background conditions and socio-economic levels among various regions are distinct (
Based on the systematic construction of a spatial database of China's mega-towns, this paper aims to explore the above two questions to lay a foundation for promoting their sustainable development. Using the nearest neighbour distance and kernel density estimation methods, we analysed mega-town's spatial distribution pattern, examining their national and regional differences. Furthermore, by combining Excel software with ArcGIS software, regression analysis and spatial autocorrelation method were applied based on the overlapping spatial elements, to quantitatively analyse the growth mechanism of mega-towns from topography, location, economy, population, road traffic and other five aspects. Besides, national policy was selected to qualitatively analyse the growth mechanism. This research can also provide certain reference for other developing countries.
2 Key literature review
According to the
Currently, the concept of small towns has not been clearly defined. Based on the availability of town data, many countries or current studies often applied a rather simple definition of towns as settlements with a certain population size, which was non-uniform due to different population thresholds of small towns among various countries (
As a special kind of town in the world, mega-town's population and functional form often reach or even exceed the general level of cities. This is closely related to national standards for setting up towns and cities. For instance, a British town can evolve into a city when three main criteria are satisfied: “a minimum population of 300,000”, “a record of good local government” and “a 'local metropolitan character'”. Sometimes, it also involves other factors, such as the results of democratic elections, government ruling standards, etc. So far, a number of large towns (such as those with over 200,000 residents) have existed in the UK, which cannot legitimately call themselves a city without the royal designation. Similarly, 238 mega-towns are formed in China, which refers to a town with more than 100,000 inhabitants in the township according to
Mega-towns in most foreign countries rarely have conflicts and problems between management and development, as the local governments have autonomous rights on lawmaking, financial allocation, infrastructure construction, and other aspects. Therefore, foreign research on mega-towns is relatively limited, mainly about development processes, social issues, land uses, etc. For instance,
In view of the non-municipal self-government systems in China, mega-towns have certain particularity and restrictive nature: they have more responsibility but less power, which drives them into a dilemma of weak function and low efficiency (
3 Data and methods
3.1 Study area and data
China is located in East Asia on the west coast of the Pacific. It has 34 provincial administrative regions (Hong Kong, Macao, and Taiwan are excluded due to the lack of data), as shown in
Figure 1.
3.2 Methods
3.2.1 Nearest neighbour distance
The nearest neighbour distance is the distance between any point and its nearest neighbour, indicating the extent of geographical proximity between points in geographic space. The nearest neighbour index is mainly analysed by the average distance between a point and its nearest neighbour, and the expected average distance of hypothetical random distribution, to determine whether the point is randomly distributed or concentrated. The specific calculation method is as follows:
where
3.2.2 Kernel density estimation
The kernel density estimation (KDE) method mainly uses a moving cell to estimate the density of points or line patterns. It is to study the distribution characteristics of points by examining the spatial changes of point density in a regular area, whose calculation formula can be expressed as:
where
Spatial analysis tools of ArcGIS software can realize the KDE method and perform spatial visualization of results. This study uses the KDE method to estimate the spatial density of 238 mega-towns in China, and analyses the overall spatial distribution characteristics of mega-towns on the national scale.
3.2.3 Regression analysis
Regression analysis is a method of analysing the linear correlation between one independent variable
where
where
This study is to judge the correlation between each factor and the growth of mega-towns, by using each factor's data to conduct the regression analysis with the number of mega-towns, laying a foundation for spatial autocorrelation analysis.
3.2.4 Spatial autocorrelation
Spatial autocorrelation analysis is an analytical method that tests whether a unit's observations are correlated with its neighbour's observations (
where
Local autocorrelation is to reveal the similarity or heterogeneity between observations of adjacent spatial units in the study area. It can quantitatively identify the distribution of “hot spot, cold spot” of a certain attribute of area, and then detect the spatial pattern of regional polarization. The formula of the local Moran's
where
Global autocorrelation and local autocorrelation methods are mainly used to analyse the spatial growth mechanism of mega-towns. Among them, this paper chose global autocorrelation to determine whether each relevant factor of mega-towns was spatially agglomerated. Further, this research adopted local autocorrelation to analyse the spatial distribution characteristics of various factors. By comparing and analysing the consistency or similarity between the spatial distribution features of factors and that of mega-towns, this paper further explored the influence mechanism of related factors on the growth of mega-towns.
4 Results
4.1 National scale
As shown in
China's mega-towns listed by scale hierarchies in 2015
China's mega-towns listed by scale hierarchies in 2015
|
Based on formula (1) and ArcGIS software, the theoretical nearest neighbour distance of China's mega-towns is 99,828.3 m, and the nearest neighbour index
To further analyse the spatial concentration characteristics of mega-towns, this study obtained the kernel density of mega-towns characterizing the spatial distribution pattern of mega-towns based on kernel density estimation method, as shown in
Figure 2.
(1) The overall distribution of China's mega-towns was a pattern of “sparse in the Northwest and dense in the Southeast”. From
(2) Mega-towns formed two spatial core agglomerations and several secondary ones. The two core clusters were namely the Yangtze River Delta urban cluster and the Pearl River Delta urban cluster, where both the number and population size of these mega-towns were relatively large. The two medium-city-level mega-towns were both concentrated in the Pearl River Delta region, where the total number of mega-towns was 34, which accounted for 14.3% of all mega-towns. The Yangtze River Delta region had 65 mega-towns, which made up a relatively significant share (27.3%) of mega-towns. Among them, there were 12 mega-towns with a population of over 200,000, including Dachang Town, Qibao Town, Xinzhuang Town, etc. There were secondary clusters which were mainly distributed in the Beijing-Tianjin-Hebei urban agglomeration, the Sichuan Basin, the central area of Hunan and Hubei, and the central area of Guizhou. These secondary clusters had a relatively large number of small-sized mega-towns with a contiguous distribution. It can be seen that the spatial agglomeration features of China's mega-towns were comparatively remarkable.
(3) A southeast coastal agglomeration belt of mega-towns has formed. China's southeast coastlands had a prominent geographical location advantage, rapid economic development, and a high level of urbanization. Production factors in these areas such as labour, capital, technology, etc., were highly concentrated, which provided unique conditions for the growth and development of mega-towns. A number of mega-towns have already grown up in the coastal strip from the eastern part of Jiangsu Province to Hainan Island, forming a coastal belt.
4.2 Regional scale
According to features of socio-economic development in various regions, China's national land space is divided into eight comprehensive economic zones, as shown in
Socio-economic situation of eight comprehensive economic zones in 2015
Socio-economic situation of eight comprehensive economic zones in 2015
|
This study counted the number of mega-towns in each area and summarized the economic contribution and demographic contribution of mega-towns to the corresponding region, as shown in
Figure 3.
In addition, this study recorded the nearest neighbour distance indexes of mega-towns of eight economic regions, as shown in
Regional nearest neighbour distance index r of China's mega-towns in 2015
Regional nearest neighbour distance index r of China's mega-towns in 2015
|
4.3 Provincial scale
Based on data availability, the following analysis did not include Hong Kong, Macao, and Taiwan. This study counted the number of mega-towns in 31 provincial administrative units and calculated their nearest neighbour distance index
Provincial nearest neighbour distance index r of China's mega-towns in 2015
Provincial nearest neighbour distance index r of China's mega-towns in 2015
|
(1) Mega-towns of most provincial-level units in China tended to be in spatial agglomeration distribution, but degrees of agglomeration were not high.
(2) Only one province had a random distribution of mega-towns. Based on calculation results, Anhui Province was the only province among all provincial units, whose nearest neighbour distance index
(3) The number of mega-towns in those provinces where mega-towns were evenly distributed was generally small.
5 Analysis of growth mechanisms of mega-towns
The spatial growth of mega-towns is a result of the combination roles of many influencing factors. Considering natural factors and regional natural resources, terrain and environment play fundamental roles in the development and growth of urban settlements by affecting production efficiency, life quality and ecological protection. With regard to humanistic and social factors, regional economic development conditions, population distribution, geographical location, infrastructure level and government policy, etc. all have an important effect on the growth of mega-towns. According to the principles of correlation factor typicality, data availability, and operability, this study selected five aspects to complete a quantitative analysis on growth mechanisms of China's mega-towns, involving topography, location, economic development, population concentration degree, and traffic. Meanwhile, the national policy factor was chosen for the qualitative analysis.
In terms of quantitative analysis, this study would take a regression analysis to judge the correlation between the selected influencing factors and the number of mega-towns. This study used each factor's indicator data to make regression analysis with mega-towns. By determining those factors that are related to the number of mega-towns according to the result of regression analysis, this study will continue to perform spatial autocorrelation analysis to analyse in-depth the specific impact of factors on the spatial growth of mega-towns.
With regard to qualitative analysis, this paper will analyse the impact of national policy factors on the spatial growth of mega-towns.
Finally, this study will summarize the growth mechanism of China's mega-towns by comprehensively sorting out the results of qualitative and quantitative analysis.
5.1 Quantitative factor analysis
5.1.1 Regression analysis
This paper applied digital elevation model (DEM) indicator to characterize terrain; the distance between a mega-town and its neighbouring core city was indicative of the location advantage level of the mega-town; economic density represented the degree of regional economic development; population density indicated the regional population concentration degree; highway network density characterized the level of regional road traffic convenience. One thing to note, there were 90 core cities consisting of 34 provincial capital cities and 56 prefecture-level cities above the scale of a large city, whose urban resident population was over 1 million. In addition, the economic density, population density, and highway network density data all adopted municipal-scale data. The specific data processing methods are as follows:
This paper collected national DEM raster data, vector surface data of national prefecture-level spatial units, vector point data of national prefecture-level city seats, and vector point data of the 238 mega-towns' geographical layout, vector line data of the national highway network distribution, urban GDP and resident population statistical data. This paper took the extraction tool to assign the DEM raster data to the corresponding extra-towns, to obtain the DEM vector database of mega-towns in China. Also, it measured the distance between a mega-town and its neighbouring core city, constructing the location advantage vector database. Furthermore, it calculated economic density, population density, and road network density at the prefecture-level, and built the vector databases of national municipal-scale economic density, population density, and road network density. Finally, it superimposed the three vector maps with the spatial distribution vector point map of mega-towns through the interception tool of ArcGIS, to make mega-towns obtain the local attributes including economic density, population development levels, and traffic conditions. Therefore, 238 mega-towns obtained all attributes of the five selected influencing factors.
This study used each factor's attribute data to make regression analysis with the number of mega-towns based on formulas (4)-(7), to judge the correlation between each factor and the growth of mega-towns. The results showed that the correlation coefficient
Figure 4.
It denoted that there all were interrelationships between the five factors and mega-towns: (1) The economic density, population density, and highway density were positively correlated with mega-towns. Except for road density, the results of model tests were very significant, indicating that densely populated and economically developed areas were more conducive to the cultivation of mega-towns. The improvement of traffic conditions might also benefit the development of mega-towns. (2) DEM and distance from core cities both had a negative correlation with mega-towns. The DEM's model test was generally significant, and the model test of distance from the core city was very significant. This indicated that the flatter the terrain and the closer to the core city, the more conducive to the growth and development of mega-towns. Therefore, this paper determined that these five factors chosen were related to the growth of mega-towns.
5.1.2 Spatial autocorrelation analysis
To further explore the specific impact of factors on the spatial growth of mega-towns and the growth mechanism of mega-towns, this paper implemented spatial autocorrelation analysis, involving global autocorrelation and local autocorrelation.
First, using the formula (8), this paper carried out global autocorrelation processing on those influencing factors on mega-towns' growth to determine whether there were some spatial agglomeration characteristics. The results of the global autocorrelation showed that the Moran's
Second, this paper made local autocorrelation processing on the spatial distribution of each correlation factor based on the formula (9). By further confirming whether there was consistency or similarity between each factor's concentration characteristics and mega-towns' distribution features, we analysed their influence on the spatial growth of mega-towns. The results are as follows:
(1) Terrain factor. This paper conducted local autocorrelation measures on the DEM distribution map of mega-towns in China. The results are displayed in
Figure 5.
In addition, this study analysed the elevation values of mega-towns and found that nine mega-towns with an altitude of over 1000 m accounted for 3.78% of the total number of mega-towns; six between 500 and 1000 m occupied 2.52%; 16 with 300-500 m occupied 6.72%; eight between 200 and 300 m consisted of 3.36%; 199 were below 200 m, accounting for 83.6%. This indicates that more than 80% of mega-towns are distributed in plain areas, implying that the flatter the terrain, the more likely that mega-towns would flourish.
(2) Location advantage factor. The location of a mega-town refers to the spatial connection between a mega-town and a developed city or core economic body. The closer a mega-town is to a developed city or core economy, the more opportunity it has to accept the radiating effect, which provides advantages to location. This study utilized the distance between a mega-town and its neighbouring core city to roughly characterize the location advantage of the mega-town. The local autocorrelation result of the location advantages shows that mega-towns mainly formed two agglomeration types, namely High-High disadvantage-type and Low-Low advantage-type (
Figure 6.
(3) Economic development factor. Since the reform and opening up of China, there have been significant regional differences in China's socioeconomic development (Sui, 2017). The economic development in the eastern coastal areas such as the Yangtze River Delta and the Pearl River Delta is rapid as compared to other areas. Mega-towns in these regions have become the main choice for industrial transfer and functional evacuation of core cities. The local autocorrelation processing on the economic density distribution is shown in
Figure 7.
(4) Population concentration degree factor. Working as producers and consumers, the population is the primary development body of a town, whose spatial agglomeration could promote the growth and development of the town.
Figure 8.
(5) Traffic condition factor. Road traffic, working as an important transmission medium for urban economic and social relations, plays a vital role in the development of towns (
Figure 9.
5.2 Qualitative factor analysis
There are many qualitative factors affecting the growth of mega-towns, including urban institutional mechanisms, management systems, government policies, etc. This research mainly selected national policies for qualitative analysis. The Chinese government has introduced preferential policies for small towns since the reform and opening-up (
In addition, the latest standard for setting up towns, made by the Ministry of Civil Affairs in 2000, does not restrict an upper limit on the population size of a town, which means that a town's status would not change regardless of any population size. In light of the division criteria of city size in China, 198 of 238 mega-towns have reached the size standard of type-II small city with a population size of less than 200,000, 38 fit the scale demand of type-I small city whose scale is between 200,000 and 500,000, and another two meet the criterion of medium-sized cities with 500,000-1,000,000 residents. However, the Chinese government does not upgrade them into corresponding cities, because a town is the lowest one in Chinese administrative levels, which cannot fulfill the normal management duties of a city government. Hence, China has formed a unique administrative unit-mega-towns.
5.3 Summary of growth mechanisms of mega-towns
Based on the qualitative and quantitative analysis, this paper found that there are different levels of correlations between all related factors and small towns. The natural resources and environment, especially the topographical conditions, laid the foundation for the growth of mega-towns. The economic factor, population factor, and traffic factor all played positive roles in the development of mega-towns. Among them, the concentration features of the three main agglomeration types of economic density and population density were both relatively consistent with the overall spatial concentration characteristics of mega-towns. In addition, the absolute value of correlation coefficient between population density and mega-towns was the largest according to the results of regression analysis. This reflected that the population concentration degree was the main driving force for the growth and development of the mega-towns. Location advantage's absolute value was the second biggest, signifying that the mega-town was easy to grow around big cities, working as a satellite town or a carrier of the population and industrial transfer. In addition, the national policy orientation and administrative division system played a certain role in the formation of mega-towns.
In fact, there were intricate interactions between the relevant factors, which together scheduled the growth and development of mega-towns. On the one hand, economic development and external traffic conditions have a positive impact on population agglomeration and diffusion (
Figure 10.
6 Conclusions and discussion
6.1 Conclusions
There were 238 mega-towns in China in 2015. The total population in mega-towns was more than 38 million, occupying 11.82% of the whole small towns although their number only accounted for 1.18%, which reflected their important contributions to urbanization. This paper aims at analysing the spatial distribution features of mega-towns in China, and exploring their growth mechanism to provide certain reference for making town development policies. It adopted the nearest neighbour distance, kernel density estimation, regression analysis, global autocorrelation, local autocorrelation and other spatial analysis methods based on a spatial database of mega-towns in China.
The results showed that the nearest neighbour index
With regard to the growth mechanism of mega-towns, this paper mainly adopted quantitative analysis methods, combining with qualitative analysis. It chose topography, location, economy, population and traffic as related factors and selected DEM, location advantage, economic density, population density, and highway density as corresponding indicators of the five quantitative factors. According to regression analysis results, the correlation coefficient
6.2 Discussion
Mega-town works as a key link to realize the interaction between new urbanization and new rural construction, which has a unique role in advancing regional economic growth, absorbing labour employment and other aspects. Provinces and prefecture-level units should make full use of the mega-town functions to promote their socioeconomic development, and adjust the urban development strategies according to the spatial distribution characteristics of mega-towns. China is the largest developing country, which still has the arduous task of urbanization. Fostering mega-towns would benefit the realization of in-situ urbanization, which is conducive to the orderly advancement of the Chinese urbanization process. Particularly, the central and western regions in China, which are dominated by small towns with relatively backward development, need to cultivate mega-towns as new development nodes based on the grasp of their growth mechanism. Meanwhile, the cultivation of mega-towns will help alleviate the population and economic pressures of large cities, to mitigate the problem of “city diseases” to improve the sustainability of large cities. As mega-towns have more responsibility but less power, Zhejiang Province has taken the lead in launching the strategy of “cultivating pilot-towns into small cities”, and achieved remarkable results in terms of systems and mechanisms, development quality and efficiency, ecological environment, etc. In recent years, the state government of China has also adopted this strategy and carried out practical work. This might provide some reference for other developing countries.
[1] Small cities? Towards a research agenda. International Journal of Urban and Regional Research, 33, 683-699(2009).
[4] Testing for spatial autocorrelation among regression residuals. Geographical Analysis, 4, 267-284(1972).
[5] The role of small towns in the local economy and some implications for development policy. Local Economy, 15, 280-301(2000).
[6] et alSmall towns as ‘sub-poles’ in English rural development: Investigating rural-urban linkages using sub-regional social accounting matrices. Geoforum, 38, 1219-1232(2007).
[7] The role of small towns in local place making. European Countryside, 427(2016).
[8] The unit of social organization in a large town. The Sociological Review, 7, 23-36(1914).
[9] Exploration on promoting system reform of large towns. China Economic & Trade Herald, 26(2015).
[10] . The Division of Labor in Society(2000).
[11] Visions and actions for medium-size cities: Reports from the European Workshops of Alicante, Volos and Oviedo. European foundation for the improvement of living and working conditions. Luxemburg(1994).
[12] Kleinstädte als Wachstumsmotoren ländlich-peripherer regionen: Das Beispiel Waldviertel. disP-The Planning Review, 48, 63-76(2012).
[13] Promoting the scientific development of urbanization through the reform of the city of mega-towns. Regional Economic Review, 10-11(2017).
[14] Contemporary perspectives of small towns in India: A review. Habitat International, 86, 19-27(2019).
[15] et alPreliminary thoughts on the cultivation of mega-towns as new and medium-sized cities: Taking Zhejiang Province as an example. Urban Studies, 23, 8-13(2016).
[16] Formality versus reality: An exploration of the development process for mega townships in India. The Towers of New Capital: Mega Townships in India, 38-49(2016).
[17] Small Town Sustainability: Economic, Social, and Environmental Innovation. Birkhauser(2009).
[18] Re-urbanity: Urbanising the rural and ruralising the urban. Environment & Planning A, 39, 728-747(2007).
[19] Time and space differentiation of the development of urbanization in China’s urban population. Geographical Research, 38, 85-101(2019).
[20] Linear governance, participatory governance and development governance: A study of social governance in coastal large towns. Journal of the Party School of the Central Committee of the C.T.C, 22, 94-102(2018).
[21] Small-town sustainability: Prospects in the second modernity. European Planning Studies, 18, 1545-1565(2010).
[22] . National Report on the New Urbanization in 2015(2015).
[23] . Beyond the Metropolis: Urban Geography as if Small Cities Mattered(2007).
[24] Analysis of regional economic growth difference and its influencing factors in Anhui Province. National Circulation Economy, 63-65(2018).
[25] Towns (and villages): Definitions and implications in a historical setting. Cliometrica, 269(2017).
[26] Approaches to planning for secondary cities in developing countries. Washington, DC: Rivkin Associates Inc(1982).
[27] Secondary Cities in Developing Countries: Policies for Diffusing Urbanization. Beverly Hills, NJ: Sage(1983).
[28] Alcohol outlets, social disorganization, land use, and violence in a large college town: Direct and moderating effects. Criminal Justice Review, 38, 29-49(2013).
[29] The role and development perspectives of small towns in Central Serbia. Spatium, 8-15(2006).
[31] Research and realization about spatial analysis in the SVG-based WebGIS. Journal of Chinese Computer Systems, 33, 1770-1773(2012).
[32] Concentration and decentralization of population in the Beijing-Tianjin-Hebei metropolitan region and its determinants: A regional density function approach. Acta Geographica Sinica, 64, 956-966(2009).
[33] World Urbanization Prospects: The 2018 Revision. New York: United Nations(2018).
[34] Small towns of hope and glory InL Vaz T N, van Leeuwen E, Nijkamp P (eds.). Towns in Rural World. Farnham: Ashgate, 3-10(2013).
[35] Research on the transformation and upgrading of administrative divisions of big towns. Intelligence, 226(2018).
[36] The development of small towns in the process of urbanization: Taking large towns as an example. China Economic & Trade Herald, 60-61(2014).
[37] The development history and new situation of small towns in the past 60 years in New China. Economic Geography, 29, 1605-1611(2009).
[38] Reform of administrative division system in mega-towns in economically developed areas. Planners, 32, 34-39(2016).
[40] Introduction: Reclaiming small towns. In: Denis E, Zerah M-H (eds.). Subaltern Urbanisation in India: An Introduction to the Dynamics of Ordinary Towns. New Delhi: Springer(2017).
Get Citation
Copy Citation Text
Xueqin WANG, Shenghe LIU, Wei QI. Mega-towns in China: Their spatial distribution features and growth mechanisms[J]. Journal of Geographical Sciences, 2020, 30(7): 1060
Category: Research Articles
Received: Aug. 28, 2019
Accepted: Jan. 10, 2020
Published Online: Apr. 21, 2021
The Author Email: LIU Shenghe (liush@igsnrr.ac.cn)