A study on the generalised space of urban–rural integration in Beijing suburbs during the present day

Abstract

Urban–rural integration is a complex process in the giant urban–rural system of the Beijing suburbs with various evolutions in modern suburban functions. Therefore, the functions and processes of the whole Beijing suburb system cannot be treated in isolation. Based on the theory of generalised space proposed by Chen (2008), the urban–rural integration process in the Beijing suburbs was analysed in real space (R-space), phase space (P-space) and order space (O-space). The results show that the macro urban–rural gap is the dominant factor in the O-space. A multifunctional policy at the meso-level will promote the scientific evolution of the P-space. These factors in turn influence the evolution of the urban–rural integration in Beijing suburbs at various levels in the R-space. Finally, new features of spatial organisation, for example, fractals, are generalised from the interactions of those three spaces. Therefore, the perspective of the R-P-O space fusion can provide theoretical guidance to investigators for the urban–rural integration of Beijing suburbs.

Keywords

Beijing suburbs complex giant system generalised space multifunctionality urban–rural integration

Introduction

Since the reforms and opening up of 1978, the suburbs of Beijing, Shanghai, Guangzhou and other metropolitan areas in China have rapidly developed, and the basic driving forces are urbanisation, industrialisation and de-agriculturalisation (Chen, 2010a). The urban–rural integration processes in these metropolitan suburbs not only involved population movement, economic development and social integration, but also biological production processes and environmental reconstruction processes. Therefore, the urban–rural integration of metropolitan suburbs can be further divided into five aspects as follows: economic integration, population integration, social integration, spatial integration and ecological integration (Gu et al., 2006; Wu, 2001; Zhang et al., 2010). However, those processes coexist with the sound development of urban and rural areas and the ‘three rural issues’ (the problems concerning agriculture, farmer and rural area) as well. There are outstanding conflicts in adjusting the agricultural structure, social-economic transformation, ecological environmental conservation and the capacity of land, populations and the ecology of the environment. Other challenges such as agricultural land loss and inharmonious development among functions and mechanisms also exist in the suburbs. Therefore, the locations of multi-functionalisation and the coupling of multiple functions must lead to a variety of urban–rural integration processes, which have profoundly affected the evolution of agricultural land/areas, rural residential areas and the urban system.

The metropolitan suburbs of China are complex systems that are influenced by both human and natural events, which have been studied in a number of fields such as geography, economics and sociology to understand the developmental mechanism of modern suburban agricultural areas (Song and Liu, 2010). However, investigations that integrate multiple disciplines are basically at the theoretical discussion stage, and they have not yet achieved substantial results (Gu et al., 2006; Long et al., 2010; Wu et al., 2006).

The evolution of the Beijing suburb has become more complex since the 1980s and has tended to perform multiple complex functions. Therefore, regional disorder has been strengthened since the end of the 20th century. Moreover, the developmental gap between urban and rural areas has been further expanded. Thus, a systematic analysis of the suburbs by using complex system theory has been adopted. However, there have not yet been any studies that can scientifically reflect the status-factor relations of the Beijing suburbs. Studies based on the fusion mechanism of the multi-domain/system are very few. This study employed Beijing as a case study. A systematic analysis of the Beijing suburbs was implemented on the basis of generalised space and the development of a status quo in multiple fields to explain and analyse the complexity of the spatial-temporal evolutions of Beijing suburbs.

Literature review and method

Literature review

Modern peri-urban agriculture emerged in most international metropolitan suburbs with various beneficial contradictions because of accelerated urbanisation and industrialisation. Studies on modern peri-urban agriculture also reached a climax in the 20th century. For example, Wilson (2009) and Potter and Tilzey (2007) studied the evolution of UK regions with regards to aspects of the industrial transitional pathways of agriculture and the evolution of social and economic infrastructure in agricultural areas. The versatility of modern suburban areas in developed countries has involved various challenges such as food safety issues, environmental issues and intensified socio-economic contradictions. However, there are seldom successful cases that overcome the above problems in agricultural areas merely based on their own development.

As new features of the metropolitan suburban strip have appeared in recent years, a number of investigations have focused on the micro and meso developmental base of the modern peri-urban strip. Roca Cladera et al. (2009) and Vasanen (2012) analysed the macro developmental background of metropolitan suburbs from a ‘functional polycentricity’ view, and they concluded that the evolution of this area is the result of interactions among a number of subsystems. This ‘polycentric’ spatial structure is clearly different from the traditional monocentric city (Roca Cladera et al., 2009; Vasanen, 2012). Little (2000) concluded that this region will display the features of a post-industrial society on the basis of his new study on communication technologies, which will provide new thinking for analysing the evolution of metropolitan suburb areas. Marsden (2003) argued that the commercialisation process in modern suburban agricultural areas plays an important role in promoting local development. Clark (2005) showed that social capital and regional changes in social and economic infrastructure further enhance the complexity of evolution in the modern suburban zone. However, existing investigations only focus on one specific real space, or the socio-economic developmental status or the micro-evolutionary base, lacking the integration of multiple disciplines. Therefore, a systematic case study is needed.

Studies from various perspectives such as regional development, agriculture, rural development and ecology have been conducted to analyse the features of modern suburban agricultural areas in Beijing. Zhang et al. (2010) and Long et al. (2010) analysed the ecological, economic and social characteristics of the evolutionary mechanisms and evolution processes in Beijing suburbs. Their results show that there are still numerous problems, for example, unreasonable spatial development and rough mechanisms, although the overall situation of the Beijing suburbs is getting better. Zhang and Zhao (2007), Zhao et al. (2009) and Zhao (2011) and Wang (2005) analysed land use in Beijing suburbs from different perspectives. Their results showed that the social infrastructure, economic development and spatial structure of the Beijing suburbs have dramatically changed during the last 30 years (Wang, 2005; Zhang and Zhao, 2007; Zhao, 2011; Zhao et al., 2009). Jiang and Zhang (2006), Guan et al. (2010), Song and Liu (2012a) and Song et al. (2013) studied the multiple suitability, spatial characteristics and driving forces of the development in Beijing suburban areas by taking farmland and rural settlements as a case study. However, existing studies have usually focused on one aspect of the Beijing suburban area, which cannot fully reflect the overall situation of Beijing suburbs. For example, studies by Zhang et al. (2010) and Long et al. (2010) did not analyse agricultural land and rural settlements, which are two important characteristics of Beijing suburbs. Therefore, some investigators began to analyse the evolutionary mechanism and complex features of this region. For example, Yuan et al. (2008) analysed the evolutionary trend of the Beijing suburban region by using the multiple-objective programming method. Chen (2010b, 2012) and Chen and Wang (2013) further clarified the self-organisation processes in Beijing’s development and concluded that the city’s spatial structure can be roughly divided into three layers from the centre to its periphery. The space complexity of these three layers is clearly different, which provides specific theoretical guidance for analysing the evolution of the Beijing suburb. However, more studies are needed to analyse the driving forces of Beijing suburban development, especially regarding external incentives for evolution, action models and action levels of a variety of land types.

Existing studies on Beijing suburbs have generally focused on analysing the status quo, the contradictions and operational mechanisms of development under conditions occurring after industrialisation, urbanisation and ecological conservation. Special attention has been paid to the urban system and to the social, economic and ecological evolutionary mechanisms of Beijing suburbs. However, investigations and practices in urban–rural integration or the overall evolution of space in Beijing suburbs are relatively rare.

Methods

The spatial complexity of modern suburban agricultural areas varies across different spatial evolutionary patterns and at multiple spatial scales. Three categories of corresponding data are required to analyse the spatial complexity of modern suburban agricultural areas, including spatial data at different scales, temporal data and categorised data. However, existing studies have usually employed one or two types of data. However, some attributes of data from modern suburban agricultural areas were not correctly identified, which led to confusion in the analytical results of different research in the same research area (such as the Beijing suburbs) and ‘misleading’ follow-up studies. Chen (2008) proposed a new concept of generalised space in one of his Chinese books that was based on three sets of fractal dimensions with the aim of understanding geospatial changes at multiple scales. In this new space theory, the geographical space is classified into the following three types: real space (R-space), phase space (P-space) and order space (O-space). Three specific cases were used to illustrate the three types of spaces and fractal dimensions of geographical systems (Chen, 2008). The results showed that a new generalised space theory can be employed to explain parameters of geographical scaling laws, such as the scaling exponent of the allometric growth law for cities and fractal dimensions. Chen and Jiang (2009) further illustrated the concept of R-P-O space by addressing the spatio-temporal evolution of Hangzhou, China based on allometric and fractal ideas. Moreover, Chen discussed his three spaces theory from the viewpoint of allometric growth and revealed that this theory can help scientists to understand Zipf’s law and the fractal structure of cities in addition to the complex evolutions of a city or region (Chen, 2014).

The concept of generalised space was introduced systematically in Chen (2012). As shown in Table 1 (Chen, 2012), the generalised space framework consists of three space types:

Table 1.

The three types of geographical spaces – RPO framework (Chen, 2012).

Space type	Attribute	Allometry	Phenomenon (example)	Relation
Real space (R-space)	Space	Dilation allometry	Form, network, pattern, spatial distribution, etc.	y(r) ∝x(r)^a
Phase space (P-space)	Time	Longitudinal allometry	Process, evolution, etc.	y(t) ∝x(t) ^a
Order space (O-space)	Class (hierarchy)	Cross-sectional allometry	Size distribution, hierarchical distribution, etc.	y(k) ∝x(k) ^a

Note: In the formulae, r = radius, t = time and k = rank or order.

R-space: this is common geographical space that can be represented by maps and images, and it can be measured by using the Cartesian coordinate systems familiar to geographers. Geographical phenomena such as real forms, networks, patterns and spatial distribution can be analysed in R-space.

P-space: can be reflected by time series of geographical systems. Therefore, the regularity of geographical evolution in temporal series can be analysed in P-space. P-space can be represented by a 2n-dimensional space with coordinates (x1, x2,…, xn, y1, y2, …, yn), where xi describes a time series of a physical system (e.g. an urban area) and yi represents the time series of the state of a corresponding physical system (e.g. urban population) (i=1,2,…, n).

O-space: the definition of O-space is similar to that of P-space. However, it is characterised by data based on rank-size distributions or hierarchical structures. Therefore, O-space can be used to analyse the size distribution and hierarchical distribution among other parameters.

This concept solved the problems in which fractal dimensions cannot correspond with each other and abstract spaces are often unmeasurable. Therefore, changes in a specific area can be analysed comprehensively from multiple perspectives. To integrate existing research results systematically and to understand the regional development of Beijing suburbs, Chen’s concept of generalised space was used in this study to define the range of Beijing suburbs (section ‘Regional overview’).

Regional overview

The spatial scope of urban–rural integration

The urban–rural areas of Beijing can be further divided into different functional regions, different agricultural spheres and different socio-economic radiation circles, respectively. These three spatial scopes can reflect the macro-meso background of Beijing suburbs and help to explain the impacts of the micro socio-economic basis of urban–rural integration in Beijing suburban areas. According to Beijing’s agricultural layout planning, Beijing’s suburban areas can be divided into the following circles: The outskirt plain agricultural circle (near the suburbs), the exurb plain agricultural circle (middle suburb) and the mountain ecosystem-conservation agricultural circle (outer suburbs) as shown in Figure 1 (Zhang and Zhao, 2007). The Beijing suburban areas in this paper refer to the comprehensive spatial scope that is based on the division of agricultural spheres, taking into account the other two division spheres, which can reflect complex socio-economic changes.

Figure 1.

A map of the Beijing suburbs.

The current state of urban–rural integration

The urban–rural integration of Beijing suburbs relies on five modern suburban evolutionary functions and realisation processes, including the regional social function, the ecological function, the modern agricultural production function, the non-agricultural production function and the landscape function. As shown in Table 2, the developmental goals of the Beijing suburbs during different periods and the actual achievements of urban–rural functions have dramatically changed since 1978 (Song and Zhu, 2013b). According to Beijing’s Urban Master Plan (2004–2020) and the 12th Five-Year Plan (2011–2015), the Beijing suburbs are the subject of several goals and are supposed to form a special structure with multi-centres and multi-levels.

Table 2.

The urban–rural functional implementation of Beijing suburbs at different periods. (Song and Zhu, 2013b).

	Economic function	Industrialisation range	Social function	Urbanisation range	Ecological function
1978 to the early 1990s	Emphasis	Four districts of the near suburb (radiation)	General	Near edge of four districts of the near suburb (radiation)	Neglect
1990s	Emphasis	The outer edge of the middle suburb (radiation)	Fewer emphasis	The centre of the middle suburb (radiation)	General
First 10 years of the 21st century	Less emphasis	The adjacent areas of the middle-outer suburb (enclave)	Fewer emphasis	The outer edge of the middle suburb (rating transition)	Emphasis
After 2010	Less emphasis	The clumps + fan style + circle tendency	Emphasis	The tendency of multi-level, multi-centre	Emphasis

Note: Industrialisation and urbanisation were used to represent the achievement of economic function and social function, and the degree of attention was used to represent the realisation of the government-led ecological functions.

The evolution of the spatial layout for urban–rural functions was significantly influenced by the terrain, location, urban system and route system. The regional system displays a complex spatial pattern dominated by the above factors and exhibits spatial self-organisation and fractal characteristics (Song and Liu, 2012a; Song et al., 2013). The urban and rural industrial models, developmental functions and income differences have profound impacts on Beijing’s urban–rural integration process, producing large urban–rural differences in regional affluence, resident behaviours and the structures of developmental functions.

Unlike Shanghai, Guangzhou and other metropolitan suburbs, the scale of towns in Beijing’s suburbs has rapidly decreased, which directly impacts the distribution of urban–rural functions. The current situation of urban–rural functions was analysed by Song and Liu (2010) with data that included existing achievements (Liu and Zhu, 2006; Yang et al., 2005; Zhang and Liang, 2009), urban plans (Beijing’s Urban Master Plan (2004–2020) and the 12th Five-Year Plan (2011–2015)). As shown in Table 3 (Song and Liu, 2010), the mountainous region of Beijing (approximately 64 km outside the core of Beijing) is primarily planned to serve ecological functions. The two plains regions (approximately 22–64 km outside the core of Beijing) have relatively complex functions and multifunctional transitions, which formed the spatial basis for the regional evolution of urban–rural integration. The industrial restructuring and rural de-agriculturalisational processes simultaneously form the urban–rural local environment (Chen, 2010a, 2012; Wang, 2005).

Table 3.

A multifunctional realisation degree evaluation of Beijing suburbs (Song and Liu, 2010).

	Economic integration	Population integration	Social integration	Spatial integration	Ecological environment integration	The total score of the urban and rural integration
Middle suburb	0.2373	0.2432	0.2150	0.1993	0.1521	1.0469
Outer suburb	0.2237	0.2176	0.2458	0.1097	0.1569	0.9537

An overview of the regional development system

At present, the primary challenges faced in the urban–rural integration of Beijing suburbs includes the contradiction between urban sprawl and resource constraints and the contradictions among unbalanced regional developments. Both the above contradictions are macro characteristics of urban–rural functions during the micro-meso fusion progress.

System functions and government actions

The ideal situation of Beijing suburban urban–rural integration is to form a spatial layout with optimised urban–rural functions. The functional activities aim to maximise the ecological-economic-social comprehensive benefits and directly affect the social and economic infrastructures. The five functions of Beijing suburbs have been greatly improved since the 1980s. The layout of urban and rural areas tends to be more rational (Song and Liu, 2010). The above functions were uncoordinated in the coupling process, leading to the excessive expansion of non-agricultural land, the ‘three rural issues’ and conflicts between industrial restructuring and the destruction of the ecological environment. Moreover, changes in the environment, the agricultural layout, new countryside construction, the rural medical security system and the employment of farmers in Beijing suburbs required assumptions of the urban–rural construction functions from various perspectives (Song and Liu, 2011). As shown in Figure 2 (Song and Liu, 2010), Table 2 and Table 3, Beijing suburbs and their operational mechanisms are complex.

Figure 2.

A map of development in the Beijing suburbs (Song and Liu, 2010).

The government and the market economy will form a joint force in regional development and promote the emergence of healthy macro features. However, the market economy is complex. The government’s policies may contain errors in judgement and behavioural uncertainties. Thus, the normal state of long-term development in Beijing’s suburbs tends towards disorder with multiple uncertainties.

Regional development and functional systems

A great number of studies have been conducted to analyse the natural differences, regional differences, industry differences and social differences in Beijing suburbs (Gu et al., 2006; Guan et al., 2010; Han et al., 2010; He et al., 2006; Huang et al., 2010; Yang et al., 2005). Results by Qi and Zhang (2007) show that the order of labour input should be as follows: non-agricultural industries→cash crops→animal husbandry→vegetable fields→gardens→field crops. Agricultural labour tends to shift towards non-agricultural industries. Labour in field crop production tends towards the profitable crop. Those trends form the basic law of spatial distribution in Beijing’s suburban agricultural space (Qi and Zhang, 2007). However, the production function or the sum of every isolated functional system cannot fully reflect the socio-economic characteristics and spatial development characteristics of Beijing suburbs. Furthermore, these factors cannot effectively reflect the dynamic characteristics of dominant and recessive regional characteristics (Tian et al., 2010; Yuan et al., 2008). The regional system is sometimes greater or stronger than the sum of the functional system of the Beijing suburbs. For example, enhancing the socio-economic complexity makes urban systems and post-industrial features grow through the multiplier effect and makes the fractal possible. Moreover, the regional system is sometimes weaker than the sum of the functional system. For instance, the lag from the ‘three rural issues’ and ecological construction objectively make the agricultural production function disadvantageous. Furthermore, the ecological construction effects are often not as good as expected. As shown in Table 3, the combined effects of system functions that are evaluated separately cannot represent the whole.

The urban–rural integration mode

Over 30 years, Beijing has not only become more ordered in terms of macro development, but it is also self-organised into micro economic-social activities. Song and Liu (2010) analysed the macro urban–rural integration development model of Beijing’s suburbs by principal component analysis (PCA), cluster analysis and logistic regression analysis. As shown in Figure 2 (Song and Liu, 2010), the regional macro-meso-micro integration is affected by many factors such as the deterioration of ecological environments, industrial structure adjustments, resident attribute changes and population movements. Moreover, the characteristics and existence mode of Beijing suburbs differ from that of the central region of the city. Therefore, the complexity of the urban–rural integration of Beijing suburbs increased, which requires a relevant and consistent research paradigm.

The new features of urban–rural integration in Beijing suburbs

The micro-activities or single subsystems of Beijing suburbs do not usually display obvious regular behaviours. New features in the macro-evolution of the overall system include the following: (1) Beijing suburbs tend to be more self-organised on micro than macro scales. For example, agricultural production decreases gradually from the village→township→county→municipal area. (2) The evolution of Beijing suburbs follows a simple developmental route of agriculture→township enterprises→industry→business services→high-tech industry, which leads to complex behaviours in the system (Song and Liu, 2012b; Song and Zhu, 2013a). (3) The complexity of a single functional system is not greater than that of the whole system of Beijing suburbs. All the above features form a developmental process phenomenon in which the urban–rural integration process appears to be in contradiction between the ordered state and disordered state. For instance, as shown in Table 4 (Zhang et al., 2010), the spatial, economic and population integration of the middle suburb is more favoured than that of the outer suburb. However, the social integration of the industrialisation, urbanisation and de-agriculturalisation superposition process in the middle suburb are less advantaged than that of the outer suburb. Thus, the rural and urban differences form a relative chaos of urban–rural integration.

Table 4.

The urban–rural integration degree between the middle suburb and the outer suburb in 2009 (Zhang et al., 2010).

	Economic integration	Population integration	Social integration	Spatial integration	Ecological environment integration	The total score of the urban and rural integration
Middle suburb	0.2373	0.2432	0.2150	0.1993	0.1521	1.0469
Outer suburb	0.2237	0.2176	0.2458	0.1097	0.1569	0.9537

The space evolution mechanisms

Although the transpositions of urban–rural functions (see Table 3) partly reflect the new features of the region, there remains a lack of studies on the socio-economic system in the parameter space. In the generalised space concept, the functions and behaviours of the Beijing suburb can be expressed in R-space (for instance, the natural spatial structure), P-space (for instance, the different developmental scenarios) and O-space (for instance, the hierarchical sequence). These spaces cover the factors stated in the first three subsections of section ‘An overview of the regional development system’, measuring the evolution of Beijing’s urban–rural integration in time, space and order. Research on the structure of the generalied space can articulate the in-depth complexity of Beijing suburbs (Chen, 2010b; Chen and Song, 2011; Chen and Wang, 2013).

The R-space of Beijing suburbs

The land types of Beijing’s suburbs can be divided into cultivated land, forest land, rural settlements and urban land in R-space. The outer suburb of Beijing is dominated by forest land. Changes in agricultural land, construction land and ecological land from 1975 to 2002 in Beijing are shown in Figure 3. The built-up area in Beijing rapidly expanded during this period. The extrapolation of the suburban edge was rapid. Beijing’s ecological construction and development grew rapidly with gradually decreased agricultural land.

Figure 3.

Land use changes in Beijing.

Cultivated land and forest land in the R-space

In 2004, the nearest suburb of Beijing was 15% cultivated land and 18% forest. The middle suburb was 37% cultivated land and 22% forest. The soil texture, terrain slope, soil depth and soil organic matter was used to evaluate the natural attributes of cultivated land. Those factors were obtained from a 1:10,000 DEM map (2006). The results showed that the area of cultivated land patches increases with distance and the curve is indicated with the lower goodness of fit (R ²). Overall, the natural attributes of cultivated land are in a disordered state.

However, if the land resources and labour were distributed into different urban–rural functions to maximise the economic efficiency of cultivated land, the results would be different. As shown in Figure 4 (Song and Liu, 2012a), the spatial distribution of the cultivated land is significantly different from the farmland, the landscape corridor belt, the landscape greenbelt and the beautifying farmland area. These different areas form a macro-layered structure of cultivated land and are approximately 23–44 km, 44–66 km and 66–126 km away from Tiananmen Square, respectively (Song and Liu, 2012a). As shown in Figure 1, the layered structure is consistent with the range of near, middle and outer suburbs of the agricultural circle. The R-space consists of the combined result of a variety of socio-economic factors.

Figure. 4.

A cycle distribution map shows five types of cultivated land in the Beijing suburbs (Song and Liu, 2012a).

The Beijing Municipal Government plays a significant role in forest planning. In the closest suburb, the forest is primarily employed for its landscaping, greening and ecological functions. The middle suburban forest land performs both ecological and economic functions. The outer suburban forest land is primarily used for ecological conservation. The value of forest land tends to increase. Therefore, the R-space of Beijing suburbs is characterised by mandatory or organisational features.

At present, Beijing suburbs primarily serve landscape and ecosystem service functions. The agricultural production function has been and continues to be secondary (Zhang and Zhao, 2007). The agricultural lands have been rendered into spheres, primarily focusing on vegetable production in the near suburbs, producing food and animal products in the middle suburbs and producing fruit products in the outer suburbs. The outer suburbs serve as ecological barriers and provide a water conservation function. There are significant differences between the natural and economic attributes of the agricultural land in Beijing’s suburbs (Song and Liu, 2010). Therefore, the sphere difference is manifested through the agricultural layout and fluctuations in agricultural product prices.

Rural settlements in R-space

Of all rural residents 18% are located in the near suburb and 13% are in the middle suburb, which together form an important part of Beijing’s suburbs. The number of patches in the middle suburban settlements in 2006 was large (19,774). However, the shape was most complex when the patch area reached 3.40 hm² because the layout of the settlements was relatively more free than other land. The number of settlement patches in the near suburb was lower (7972) than that of the farther suburbs. However, the shape of settlement patches was more regular and larger (4.58 hm²).

The spatial distribution of rural settlements is the result of interactions among the convenience of agricultural production, the de-agriculturalisation process, the historical basis, the infrastructure, land planning and other factors. As shown in Figure 5 (Song et al., 2013), the rural settlement plaque shows a staggered distribution in the area centred in Qianmen. The fractal and Hurst exponent indicate that the plaque of overall rural settlements does not show fractal characteristics, although it was impacted by various factors. However, the spatial suture of the plains area is made of simple terrain and primarily serves an agricultural function (7∼76 km to Qianmen). The rural settlements in the near suburb are affected by urbanisation, industrialisation and de-agriculturalisation and have fractal characteristics. Overall, the settlement area of Beijing is the key that links the spatial expansion of the metropolis and the development of agriculture, rural areas and farmers. The spatial planning of the rural settlement area varies across space (Song et al., 2013). Figure 5 (Song et al., 2013) shows a map of the circle distribution for rural residential areas. The rural settlements in Beijing suburbs are also space vectors that support non-agricultural industry, the construction of social functions and the landscape.

Figure. 5.

A map of the circle distribution for rural residential areas in Beijing (Song et al., 2013).

The overall evolution of R-space

As shown in Figures 4 and 5, industrialisation, urbanisation and de-agriculturalisation occurred in the agricultural land and rural settlement areas of Beijing. The fractal characteristics show that (1) the fractal dimension of agricultural land in the near-middle suburbs, which primarily produce grain, vegetables and livestock products, is 1.809. In other words, agricultural land varies quickly and with complexity in this space. The self-affine dimension indicates that the near-middle settlement is 1.757, which is close to that of agricultural land. (2) The dimension of self-affine records for the agricultural land in the middle-outer suburban area, which serves as an ecological corridor, sightseeing/facilities for agriculture and fruit production, is 1.601. However, the fractal dimension of the middle-outer settlement is 1.673. Therefore, the rural settlement in the middle-outer suburb is more complex than that of agricultural land. (3) Overall, the rural settlement is more complex in the R-space (Song and Liu, 2012a; Song et al., 2013). These spatial characteristics inevitably led to changes in the social and economic activities of the suburban zone.

The P-space of Beijing suburbs

Because of the different driving forces of industrialisation, urbanisation and de-agriculturalisation, the Beijing suburb has undertaken various developmental goals at different periods, as shown in Table 2. Furthermore, the spatial evolution of the Beijing suburb system presented a time lag. The importance of the different functions at different periods caused R-spatial variations, which achieved remarkable results in the latter phase.

Characteristics of the development process

The macro development of Beijing’s suburbs was performed in different stages, including the pursuit of economic benefits in the 1980s, the balance of community interests in the 1990s and the ecological-economic-social comprehensive benefits at present. The development of Beijing suburbs from the late 20th century can be summarised as follows: (1) social-economic-ecological comprehensive functions were gradually replaced by purely economic functions or social functions. (2) Non-agricultural industries and settlements transferred from urban area to suburbs, which directly impacted the development of suburban agriculture and rural settlements. (3) Non-agricultural industries were greatly developed. Agriculture was not the backbone of regional development and must be further optimised. (4) To protect the ecological environment, the existing agricultural spatial structure was continued. However, the social and economic infrastructures in all regions had changed (Song and Zhu, 2013a).

Development status and trends

Development during the last years of the 20th century forms the basis for integrating industries and regional development. The industrial structure, regional spatial development and evolutionary rules are simulated by using the VPμGA in the author’s early study (Chen and Song, 2011). The industrial structure, regional spatial states and evolutionary rules of Beijing’s suburbs were evaluated on the basis of the developmental process described in section ‘Characteristics of the development process’, taking the township and village enterprises (TVEs) of seven industrial sectors in 2008 as the case. The results show that the developmental scenarios of Beijing’s suburbs can be described as Scenario 1→Scenario 2→Scenario 3 as follows: (1) Scenario 1: the developmental scale of TVEs has been stepped down from close to far. However, the volatility of the development in the near suburb is larger than that in the farther suburb. The regional development displays an overall strong disorder. (2) Scenario 2: the development of the near suburb is better than that of the outer suburbs. However, there is the possibility of rapid development in the outer suburbs. Middle suburban development is relatively slow. (3) Scenario 3: in comparison with the status quo, the primary function of urban and rural areas (social function) has been enhanced. However, the developmental state of non-agricultural industries is gradually getting better on the plains from the near suburbs to the outer suburbs, where all regions in Beijing’s suburbs have achieved connotative development. This developmental process (Scenario 1→Scenario 2→Scenario 3) realises the urban–rural integration to achieve not only the radiation of development in the near suburb but also the multi-centre layout of suburban towns, the urban–rural integration of settlements in terms of social-economic activities and the sound development of industry.

In recent years, the non-agricultural industrial development in the near suburb has tended to be good. However, as shown in the above simulation, a policy that completely relies on the market economy is undesirable. Furthermore, governmental management is necessary for effective protection for the development of Beijing urban–rural integration. Based on this conclusion, the government has already selected major supporting industries according to the functions undertaken by all counties. The developmental model is shown in Figure 2. This model was created to promote the rational allocation of multiple urban–rural functions.

The O-space of Beijing suburbs

The basic state of the O-space

Since 1980, the urban, transportation and business systems of the Beijing suburbs have formed a complete net-link structure. However, industrialisation, urbanisation and de-agriculturalisation have not been consistent. Therefore, the dominant factor in urban–rural integration and spatial evolutionary characteristics varies at different periods, as shown in Table 2. These characteristics and the variety of plans have formed the current spatial pattern of urban–rural functions (Table 2).

According to urban planning, the future spatial structure of urban development in Beijing’s suburbs can be summarised as ‘two axis–two bands–multi-centres’. However, the current status is that traffic and commercial centres form a regional hierarchical network. The O-space is the basis that leads to evolution according to the rules of urban–rural integration, especially economic and social integration in Beijing suburbs. These factors not only result in the net-link distribution of land use in R-space, but also the self-hierarchical law of the region’s development in P-space. According to experience, the O-space tends to be in a rational developmental state when the level of the regional transportation network is approaching the level of the urban system. The closer the level of regional transportation networks is to the level of the urban system, the more reasonable the sequence of the O-space is. Overall, the O-space of the near suburb is relatively poor at present. However, the O-space of the middle and outer suburbs is relatively good (Zhang et al., 2010).

The O-space based on the R-space and P-space

The Beijing suburbs spread from the centre of the city to counties in the O-space. This feature is closely related to the integration of multiple functions and their evolutions in P-space and the final space state in R-space. For example, the economic types of small towns exhibit an obvious ‘radiation + clumps + fan’ shape (Song, 2012b). Han et al. (2010) showed that O-space is generated during the development of the R-space and P-space. Therefore, the urban and rural populations were integrated by the industrial evolution. In other words, urban–rural integration is a specific reflection of the R-space, P-space and O-space. The fractal analysis showed that the near suburb changed quickly and is disordered in P-space because it is affected by various factors. The rural settlements are gradually suitable to the development of agricultural industry and the evolution of the urban–rural system. The middle suburb is relatively stable in the R-space and P-space. The agricultural land and rural settlements in the outer suburb are randomly distributed and do not display clearly self-organised and self-evolving features (Song and Liu, 2012a; Song et al., 2013), which was also confirmed in Chen and Wang’s study (2013) on the subregions of Beijing on the basis of multifractal characterisation. There are many uncoordinated issues because the suburb focused on different developmental functions at different times. Moreover, the realisation of development functions varied.

Conclusions

Beijing suburban development based on generalised space

From the perspective of systems science, the urban–rural integration of a Beijing suburb is a complex, giant system. Its evolution is accompanied by the realisation of multiple urban–rural functions and the generation of new features. Therefore, multifunctional problems should be analysed from the generalised space. Only in this way can the overall state and situation of the urban–rural integration of Beijing suburbs be reflected. Although a basic evaluation framework of Beijing suburbs is given in this paper, the fusion of multiple functions and the regional evolutionary mechanism (in R-space, P-space and O-space) still remains unclear.

From a generalised space perspective, the urban–rural integration of Beijing suburbs has achieved sound development in the R-space, P-space and O-space during the multifunctional process. The micro socio-economic activities tend to be at a high level, which was achieved by the three transition stages as follows: the low level of balanced development→the low level of non-balanced development→the high level of non-balanced development. Furthermore, the Beijing suburbs tend to develop balanced, generalised space with a distinct functional layout.

The Beijing suburbs are now generally in a fusion situation of various development scenarios and are in disorder states in most cases. These results show that the macro urban–rural gap of Beijing suburbs is the dominant factor in O-space. Furthermore, the multifunction guide at the meso-level in different regions will promote the scientific evolution of the P-space. Both will in turn influence the evolution of the R-space.

The spatial complexity evolution of Beijing suburbs

Beijing suburbs are not a simple regional system in which the problem can be simplified by cybernetic methods. As a feature of complex giant systems, the non-additivity viewpoint (such as fractal features) would provide theoretical guidance regarding the true characteristics, state and evolutionary mechanisms of the urban–rural integration of Beijing suburbs, including economic, population, social, spatial and ecological environment integration.

One of the most important issues for urban–rural integration is the search for the self-organisation rules of urban–rural region evolution based on R-P-O space fusion. Analyses have shown that the development of the P-space and the R-space of Beijing suburbs are correlated. Moreover, the P-space and R-space both generate self-organisation in various fields such as cultivated land, rural settlements, industrial structure and regional spatial states because of the direction of the O-space. Moreover, the socio-economic development and the impact of causal factors will lead to changes in regional development patterns. Therefore, the integration of urban and rural areas can be divided into state stabilities in the P-space and structural stabilities in the Q-space.

The evolution of Beijing suburbs in the R-P-O space should be studied as a whole on the basis of the generalised space concept. This paper argues that the Beijing suburban system has a universal geography law in the O-space. The developmental drivers differs greatly in the P-space and are disordered in the R-space. The coarse-graining method is an effective way to analyse the self-organisation process, the R-P-O space fusion and the new features of the Beijing urban and rural area.

Evaluation criteria and government actions

The integration of urban–rural functions shows new features that are summarised as mentioned above. The evaluated features include: (1) the integration of multiple urban–rural functions to produce efficient R-P-O space evolution capacity and adaptability, (2) the effective implementation of urban–rural integration processes in multifunction/multi-space and (3) the sustainable development capacity of the ecological-economic-social system. These features of rapid metropolitan urbanisation appeared in Beijing. Changes in the R-space, the P-space and the O-space were the purposes of the evaluation. Those features can be reflected by the self-organisation characteristics of objects in the R-space and the evolutions of the state of the P-space. Based on an analysis of the cultivated land, rural settlement (section ‘The R-space of Beijing suburbs’), industrial structure, regional spatial states (section ‘The P-space of Beijing suburbs’), and the overall effects of the R-P-O space fusion according to new features in the Beijing suburbs, this region is doing well at present.

However, the problems of different areas are not identical. The urban–rural integrations in different areas cannot be addressed in a unified or fixed mode. The innovative development of Beijing suburbs has made great progress. However, governmental innovation is stronger. Innovation in economic development and industrial structure adjustment is insufficient, which reduces the innovative strength of Beijing suburbs. The pressures of sustainable development in Beijing’s suburbs remain strong. Therefore, narrowing the income gap between urban and rural residents is still the top priority of governmental policies. Key issues for governmental actions include changes in the ecological/social/economic attributes at the micro level and the evolution of the phase/order space at the macro and meso levels.

Footnotes

Funding

This study was supported by the National Natural Science Foundation of China (Grant No. 41401208).

References

Chen

Song

(2011) Spatial analysis for functional region of suburban-rural area using micro genetic algorithm with variable population size. Expert Systems with Applications 39: 6469–6475.

Chen

(2008) Fractal Urban Systems: Scaling, Symmetry, and Spatial Complexity. Beijing: Scientific Press (in Chinese).

Chen

(2010a) The corresponding relationships between the demographic transition, urbanization process, and industrial development of China. Geographical Research 29(12): 2109–2120 (in Chinese).

Chen

(2010b) Exploring fractal parameters of urban growth and form with wave-spectrum analysis. Discrete Dynamics in Nature and Society DOI:10.1155/2010/974917.

Chen

(2012) On the spaces and dimensions of geographical systems. Journal of Geography and Geology 4(1): 118–135.

Chen

(2014) The spatial meaning of Pareto’s scaling exponent of city-size distributions. Fractals 22(1–2): 1–15.

Chen

Jiang

(2009) An analytical process of the spatio-temporal evolution of urban systems based on allometric and fractal ideas. Chaos, Solitons & Fractals 39(1): 49–64.

Chen

Wang

(2013) Multifractal characterization of urban form and growth: The case of Beijing. Environment and Planning B: Planning and Design 40(4): 884–904.

Clark

(2005) The ‘New Associationalism’ in agriculture: Agro-food diversiﬁcation and multifunctional production logics. Journal of Economic Geography 5: 475–498.

10.

C L

Chan Roger

Liu

(2006) Beijing’s socio-spatial restructuring: Immigration and social transformation in the epoch of national economic reformation. Progress in Planning 66: 249–310.

11.

Guan

Zhang

Zhao

(2010) Regionalization and patterns of rural residential rearrangement in suburban area of Beijing. Areal Research and Development 29(3): 114–118 (in Chinese).

12.

Han

Cai

Liu

(2010) Regional economic types and spatial differentiation of small towns in peri-urban Beijing. Urban Studies 17(4): 123–128 (in Chinese).

13.

Norio

Zhang

(2006) Modeling urban expansion scenarios by coupling cellular automata model and system dynamic model in Beijing, China. Applied Geography 26: 323–345.

14.

Huang

Zhang

(2010) Degree of urbanization influences the persistence of Dorytomus weevils (Coleoptera: Curculionoidae) in Beijing, China. Landscape and Urban Planning 96: 163–171.

15.

Jiang

Zhang

(2006) Relationship between distribution changes of rural residential land and environment in mountainous areas of Beijing. Transactions of the CSAE 22(11): 85–92 (in Chinese).

16.

Little

(2000) Networks and neighbourhoods: Household, community and sovereignty in the global economy. Urban Studies 37(10): 1813–1825.

17.

Liu

Zhu

(2006) Evaluating the geographic concentration of manufacturing industries of Beijing based on distance-based methods. Acta Geographica Sinica 61(12): 1247–1258 (in Chinese).

18.

Long

Liu

(2010) Building new countryside in China: A geographical perspective. Land Use Policy 27: 457–470.

19.

Marsden

(2003) The Condition of Rural Sustainability. Assen: Royal Van Gorcum.

20.

Potter

Tilzey

(2007) Agricultural multifunctionality, environmental sustainability and the WTO: Resistance or accommodation to the neoliberal project for agriculture? Geoforum 38: 1290–1303.

21.

Zhang

(2007) Characteristics of input of suburban farmers of metropolises and variance analysis. Resources Science 29(4): 13–17 (in Chinese).

22.

Roca Cladera

Marmolejo Duarte

Moix Bergadà

(2009) Urban structure and polycentrism: Towards a redefinition of the sub-centre concept. Urban Studies 46(13): 2841–2868.

23.

Song

Liu

(2010) Study on the functions and research method of modern suburban agricultural regions in China. Journal of China Agricultural University 15(6): 120–126 (in Chinese).

24.

Song

Liu

(2011) Spatial characteristics analysis for the multifunctional transition of suburban agricultural areas in Beijing. Scientia Geographica Sinica 31(4): 427–433 (in Chinese).

25.

Song

Liu

(2012a) Study on the spatial structure of cultivated land in Beijing based on the spectrum and fractal characteristics. Geographical Research 31(8): 1399–1410 (in Chinese).

26.

Song

Liu

(2012b) The process and enlightenment of Beijing urban–rural integration since 1988. Progress in Geography 31(8): 1399–1410 (in Chinese).

27.

Song

Zhu

(2013a) Study on the temporal-spatial evolvement of structure in Beijing suburban agricultural spheres from 2000 to 2008. Journal of China Agricultural University 18(1): 202–209 (in Chinese).

28.

Song

Zhu

(2013b) Evolution and evaluation of urban–rural integration of suburban agricultural areas in Beijing. Economic Geography 31(10): 1723–1727 (in Chinese).

29.

Song

Guan

Zhu

(2013) Study on the spatial fractal characteristics and complexity of rural residential areas in Beijing. Scientia Geographica Sinica 32(12): 52–60 (in Chinese).

30.

Tian

Yang

(2010) Spatial pattern of urban functions in the Beijing metropolitan region. Habitat International 34: 249–255.

31.

Vasanen

(2012) Functional polycentricity: Examining metropolitan spatial structure through the connectivity of urban sub-centre. Urban Studies 49(16): 3627–3644.

32.

Wang

(2005) The political ecology of rural reform in metropolitan area in China: A case study of Beijing city. Geographical Research 24(3): 453–463 (in Chinese).

33.

Wilson

(2009) The spatiality of multifunctional agriculture: A human geography perspective. Geoforum 40: 269–280.

34.

(2001) China’s recent urban development in the process of land and housing marketisation and economic globalization. Habitat International 25: 273–289.

35.

Wang

(2006) Monitoring and predicting land use change in Beijing using remote sensing and GIS. Landscape and Urban Planning 78: 322–333.

36.

Yang

Zheng

Wen

(2005) Studies on service value evaluation of agricultural ecosystem in Beijing region. Journal of Natural Resources 20(4): 564–571 (in Chinese).

37.

Yuan

Chen

(2008) Urban dynamics and multiple-objective programming: A case study of Beijing. Communications in Nonlinear Science and Numerical Simulation 13: 1998–2017.

38.

Zhang

Zhao

(2007) Layout of Metropolitan Modern Agriculture. China University of Petroleum Press (in Chinese).

39.

Zhang

Han

. (2010) Comprehensive evaluation and cluster analysis of the urban–rural integration. Rural Economy 2010(12): 49–52 (in Chinese).

40.

Zhang

Liang

(2009) Influencing mechanisms of industrial agglomeration on urban spatial structure – The case of Beijing Metropolitan Region. Areal Research and Development 28(5): 6–11 (in Chinese).

41.

Zhao

Lü

Woltjer

(2009) Conflicts in urban fringe in the transformation era: An examination of performance of the metropolitan growth management in Beijing. Habitat International 33(2009): 347–356.

42.

Zhao

(2011) Managing urban growth in a transforming China: Evidence from Beijing. Land Use Policy 28: 96–109.