1. Introduction
Spatial zoning is a public policy to achieve strategic guidance, rigid management, and control. The question of how to coordinate spatial zoning has become a global topic [
1]. Meanwhile, the content and scale of spatial zoning are distinctly different from the national conditions and development. Overall, the spatial zoning of developed countries such as in Europe, the United States and Japan, is relatively mature and complete [
2,
3,
4], while that of developing countries such as China is relatively backward and unsystematic. In terms of the content of spatial zoning, developed countries have turned from material construction planning to social development planning, and the social and environmental factors involved have been given much more attention [
5]. From the scope of spatial zoning, more attention has been paid to wide range spatial zoning. For example, the regional development planning of Holland, Britain, Germany, and Hungary take the whole country as the planning object [
2,
6,
7,
8]. While undertaking a comprehensive survey of the spatial zoning of developed countries, we found that they had three important characteristics: (1) This spatial zoning was relatively complete [
2,
3,
4] and the level was clear [
1]. Taking Japan as an example, the national comprehensive development plan has been compiled since the 1960s and a round of planning is compiled every ten years. At present, it has established a perfect planning and management system and formed spatial zoning for the full coverage of the country to wide areas, and then to city–town–village, which was adapted to the Japanese administrative system [
9]. (2) Spatial zoning has changed from a traditional concept with economic growth as the primary goal to the comprehensive and sustainable development of the economic, social, and ecological environments [
10,
11,
12,
13,
14,
15,
16,
17]. (3) Much attention has been paid to the major existing problems of these countries such as the industrial structure adjustment after economic growth, resource and environment pressures, the imbalance of regional and intra-regional development, energy shortages, global climate change, and so on [
18,
19,
20,
21,
22,
23,
24]. In all, the spatial zoning of developed countries has made positive contributions to the local development in several ways such as a balance of regional development, control of land use, environmental protection, and so on [
4]. In comparison, China is a developing country with a high population density and the land ownership of China is land nationalization. Meanwhile, Chinese urbanization is developing at a high speed [
25,
26]. All of these natures make Chinese spatial zoning unique and very different from that of developed countries. The study of Chinese spatial zoning can not only be used to supplement and refine the existing spatial zoning of western developed countries, but can also provide strong support and a basis for the local spatial planning (zoning) of all levels of governments in China.
Municipal administrative areas (MAAs) are a key component of administrative zoning systems and an important field of zoning adjustment in China. Spatial zoning based on MAAs plays an important role in administrative systems and promotes the implementation of space control policies. However, in actual urban spatial zoning, plans are often contradictory and the effect of management overlap is particularly serious. Meanwhile, the planning system is hierarchical, and strategies where the government value the city but neglect the rural areas, and extensive land use have caused great difficulties for governments at all levels [
27]. To overcome these problems, the construction of a unified spatial system to identify ecological–agricultural production–living spaces (EPLs) [
28,
29] is crucial for spatial plan coordination [
30]. Here, the EPLs include the ecological space mainly used for ecological services and ecosystem maintenance (such as woodlands, grasslands, and so on), the agricultural production space that is mainly for agricultural production (such as basic farmlands, general farmlands for agricultural production, etc.), and the living space, which mainly undertakes the construction and economic development of the urban and rural areas (such as the urban built-up area, urban planning area and development zones, and rural living land).
As for the empirical research into spatial zoning reported previously, most do not use a pattern of equilibrium development [
31], which gives priority to the development of MAAs. Objectively speaking, this pattern helps society achieve a higher economic growth [
32,
33]. However, there are growing gaps of economic, societal, and environmental factors between urban and rural MAAs, resulting in problems such as the deterioration of urban environmental quality, house and traffic crowding, and an increase in urban crime rates. Meanwhile, the amount of agricultural and ecological land has decreased, and the ecological environment has been damaged [
34,
35,
36,
37]. Recently, with the proposal of a more scientific outlook on development, decreasing the gaps between urban and rural areas and ensuring the equilibrium development of the economy, society, and the environment in MAAs under limited resources have become the main principles of urban spatial zoning [
38]. Similarly, developed countries such as Germany, France, the United States, and the United Kingdom have encountered the same difficulties in the process of industrial development. They have also adopted strategies of equilibrium development to balance the development of land space and environmental protection [
39,
40,
41,
42].
There has been a great deal of research on MAA spatial zoning models, and three main models have been established: the urban structure zoning model, the urban growth boundary model, and the function-oriented zoning model. Of them, the urban structure zoning model has been widely used for early MAA spatial zoning. Using this model, the mono-centric model and multi-centric model have been used to study the influence of city sprawl in Seoul and the Mediterranean during different periods [
43,
44,
45,
46,
47,
48]. For the urban structure zoning model, its mathematical expression has advantages in exploring the influence of spatial dynamic change factors and their relationships. With the intensification of urban expansion to agriculture and ecology, empirical research on urban growth boundaries have been studied using many methods such as GIS and cellular automaton, and the urban growth boundary model became the mainstream model for the zoning of regional living space [
49,
50,
51,
52,
53,
54,
55]. Generally, the urban growth boundary model realizes the boundary simulation of urban living space in different periods using different zoning methods and establishes an evaluation system of subjective weight settings. Therefore, it is useful in guiding urban expansion and land use. With the increasing chaos of urban spatial functions, the function-oriented zoning model has become increasingly useful. This model is divided into two categories. The first is the Major Function-Oriented zoning model such as the spatial equilibrium model for regional development, which is the most representative model [
56]. This model is based on the main influencing factors in the economy, society, environment, and politics, and the model uses the method of setting a subjective weight to zone the major function of the region, providing the basis for regional function orientation and a future development direction. It has been widely used in land development [
57,
58,
59,
60,
61] in many of the municipal–county administrative areas of China, and the correctness of the model has been verified empirically. The second is the EPLs zoning model. In recent years, with the proposal of multi-plan coordination, growing attention has been paid to a number of EPLs zoning models using subjective weight settings based on function zones. For example, Liu and colleagues explored an evaluation method to classify and evaluate EPLs based on the classifications of land use [
62]. This evaluation system is a reference for the classification planning and optimization of EPLs. The function-oriented zoning model has aided in the construction of a set of subjective evaluation systems, and due to its practicality, it has been widely applied in the field of spatial zoning.
Generally, the urban structure zoning model studies the factors that affect the urban form and the relationship between them based on strong economic logic derivation. It has been widely used in urban spatial zoning and urban sprawl. However, the factors it studies belong to the economic field and its combination with urban planning is not strong, and many factors cannot be quantified, so it cannot effectively solve the practical problems of spatial plan coordination. The key purpose of the urban growth boundary model is to construct an evaluation system of subjective weight settings and to simulate the urban development boundary during different periods using different models. Given the existence of subjective weight settings, the zoning of the spatial boundary obtained with this model is relatively subjective. As a result, it cannot be used effectively to guide the spatial plan coordination. The function-oriented zoning model can be used to effectively classify optimal development zones, key development zones, prohibited zones, limited development zones, and EPLs in a city, and gives a qualitative evaluation of urban functional attributes from the view of development. Nevertheless, in this model, there is a strong subjective influence of subjective weight settings on zoning [
62].
In all, the existing MAA spatial zoning cannot effectively guide the implementation of MAA planning, which leads to various intensifying contradictions between the planning content and spatial governance. Meanwhile, the land-use planning indicator of MAA cannot be used directly, construction projects are difficult to perform, and the management of ecological space is disordered. These problems have led to the resistance of social and economic development in MAAs. In order to solve these problems, it is of utmost importance to establish a scientific and effective MAA spatial zoning model (M-MSZ) for the zoning of MAAs. In this study, based on the spatial equilibrium model for regional development, we constructed an M-MSZ under the framework of spatial equilibrium theory [
63,
64,
65]. The obtained model was optimized by a statistical analysis and compared with the existing spatial zoning model (including the urban structure zoning model, urban growth boundary model, and major function-oriented zoning model) under the same factors. It was found that the model showed advantages related to its quantitative, objective, scientific mechanism, and easily accessible data sources. Moreover, it provided an efficient and unified EPLs zoning map for MAA spatial plan coordination, and it showed a higher cost-effectiveness, and was more efficient and precise in the simulation of urban land use planning.
It is necessary to explain that in this paper, the major function-oriented zones’ delimitation of MAA was used to coordinate with the M-MSZ to formulate a zoning standard of MAA. EPLs zoning was carried out based on this zoning standard, as shown in
Figure 1. Meanwhile, the M-MSZ realized the EPLs spatial zoning of MAAs by constructing an evaluation system of spatial development suitability. Therefore, the M-MSZ can not only carry out the delimitation of EPLs, but can also realize the zoning of spatial development suitability. In the following part of this paper, the applications in both EPLs’ delimitation and spatial development suitability of MAA were conducted.
4. Discussion and Conclusions
As we know, unbalanced development is the root of the increasing gap between urban and rural areas, which resists the urban and rural symbiosis in most cities and regions of the world [
94,
95]. Scientific and rational MAA spatial zoning will help to solve many imbalance problems of MAA development such as urban and rural development and protection, land use, etc. Recently, the question of how to build a scientific and reasonable MAA spatial zoning model and coordinate spatial planning has received much attention. Currently, several spatial zoning models have been reported such as the urban structure zoning model, urban growth boundary model, and major function-oriented zoning model, and so on, and they have been successfully used to zone different functional zones and to guide the planning of MAA. Particularly for western developed countries, these models have been applied widely. The experience of them has provided strong support and reference for Chinese spatial zoning. However, it cannot be used directly for the spatial zoning of China for several reasons. First, each country chooses the model suitable for their development and makes the model work in their national conditions. Hence, different national conditions will inevitably create differences in the content and structure of spatial zoning, as shown in
Figure 10. Here, we selected several representative developed countries that have shown advanced research in spatial planning to compare with China [
2,
3,
4,
6,
7]. As for China, this is a country with a unique national condition (public ownership of land, vast expanses of land, and a huge population) that has been undergoing the high-speed development of urbanization [
96]. Therefore, we cannot mechanically apply the spatial zoning models of western developed countries to China. Second, the assignment of factors of the existing models is often difficult and their subjective weight settings are strong, which makes them ineffective to guide the EPLs zoning and spatial plan coordination of China. In this paper, we launched theoretical and empirical research on the spatial zoning model based on Chinese national conditions and development stages.
Existing spatial zoning models often show shortcomings such as difficult data acquisition and limited application due to their high requirements of the related parameters, insufficient objectivity because of their strong subjective weight settings, and lack of comprehensive zoning on EPLs. In this study, we attempted to construct a quantitative and easily operated MAA spatial zoning model (M-MSZ) by using the organic integration of economic logic deduction and urban planning practice under the framework of a spatial equilibrium theory, based on the planning experience on the balanced development of ecology, humanity, protection, and development. The M-MSZ and its construction method provide a new perspective and technological support to alleviate and even solve the extensive economic behaviors such as the inefficient operation of land resources (the speed of land urbanization is faster than the speed of population urbanization) and the deterioration of the ecological environment in China and other developing countries with rapid economic growth. It also demonstrates a new feasible demonstration for China to promote the implementation of new urbanization strategies in the new period and new normal situation [
99]. First, the selection of influence factors for the model was carried out from the literature statistics, the questionnaire survey of MAA participants (residents, government, developers, and farmers) about the leading factors of choosing a living space, and expert advice, which made the model factors more accurate. Second, during the construction of the model, the iceberg transport costs and Marshallian demand theory were introduced, and the development suitability and major functions of MAA were taken into synthetic consideration, which ensured the rationality and feasibility of the zoning model. Then, the model was optimized by statistical analysis and the rationality of the model was verified by the numerical simulation analysis, which guaranteed the scientific application of the model construction. Finally, typical MAAs were used as samples to complete the comparison and induction with the existing spatial zoning models, and the scientificity and ease of operation of the M-MSZ were verified again. The results of optimization, verification, and application are shown as follows.
During the construction of the model, the M-MSZ combined the advantages of the three existing spatial zoning models to reduce their disadvantages. For example, four factors that represent a high contribution rate and easy collection in the urban structure zoning model were retained, these being: economic situation, population agglomeration, location advantage, and transport superiority, respectively. At the same time, the total land commodities, urban structure, and the preference variable of MAA participants to land commodities were optimized, which made these three factors easier to acquire. As a result, the M-MSZ improved the practical application of the urban structure zoning model and its application was much easier to extend. On the other hand, the M-MSZ reduced the subjectivity of the urban growth boundary model and the major function-oriented zoning model (including the spatial equilibrium model for regional development and the existing EPLs zoning model) while maintaining their high practicality. The ratio of quantitative parameters was enhanced significantly by using the method of economic derivation. Many quantitative parameters such as the economic situation, population agglomeration, location advantage, transport superiority, available land volume, the pending development zones, and the preference variable of MAA participants to land commodities were combined in the model to make it more accurate and more objective in the mechanism expression. After comparison, it can be seen that the M-MSZ showed that it was more efficient, comprehensive, cost-effective, and widely used in the coordination of spatial planning and the EPLs zoning than the three existing spatial zoning models. Moreover, this model could also effectively guide the hierarchical scale zoning of the urban spatial system as well as the zoning of land use functions.
The correlation analysis showed that the relevance of unavailable land and the pending development zones to the suitability of urban spatial development was very small. From the comparison of the F values and the Fa (k–1, n–k) of each factor after regression analysis, all factors were F > Fa (k–1, n–k) except for the unavailable land and pending development zones, which was consistent with the results of the correlation analysis. After eliminating the unavailable land and pending development zones, the rest of the 11 factors including the economic situation, location advantage, transport superiority degree, population quantity, topographic and terrain, land quantity, environmental capacity, ecosystem vulnerability, utilization of water resources, natural disaster, and urban structure could be used to effectively explain the spatial development suitability.
The results of the numerical simulation showed that the economic development level, population quantity, and land resource were the main factors that affected the suitability of urban spatial development, while location advantage, transport superiority degree, and urban structure had a smaller influence on the suitability of urban spatial development. The five factors in the preference of MAA participants to land commodity including topographic and terrain, natural disaster, utilization of water resources, environmental capacity, and ecosystem vulnerability had the smallest influence on the suitability of the urban spatial development. Importantly, the accuracy of the M-MSZ was confirmed by the numerical simulation results.
The comparison of the M-MSZ and the three existing models (urban structure zoning model, urban growth boundary model, and major function-oriented zoning model) [
43,
49,
61] showed that the Kappa values of the consistency test were 85.9%, 88.2%, and 85.2%, respectively, with an average of 86.4%. We believe that the M-MSZ efficiently inherited and extended the advantages and interpretations of the existing zoning modes. Moreover, they avoided the limitations of the application range of the existing models and looked like an organic integration of the three existing spatial zoning models. In addition, the EPLs zoning of the M-MSZ was clearer than the three existing models. Therefore, the M-MSZ could be used as a supplement and extension to the research of current existing spatial zoning. From the perspective of the application, this model can not only solve the problem of the EPLs zoning and spatial plan coordination in MAA, but can also effectively guide urban land use planning from two dimensions of space and time and effectively promote the coordination and sustainable development of spatial planning. When the application scope is concerned, the method and construction concept of M-MSZ are more practical to the spatial planning of regions including developing countries or underdeveloped areas (which occupy more than 70% of the world′s land area and the total population), and regions that are in rapid urbanization [
100,
101].
In conclusion, the M-MSZ can be used to solve the problems of urban and rural co-ordination and regional coordination such as delimiting EPLs of urban and rural areas in MAAs, determining land use functions of MAAs, evaluating the suitability of land exploitation in MAAs, and formulating the timetable of land development in MAAs. However, the M-MSZ is limited in its contribution to various transmission flows such as in the planning of city traffic, population, economy, and tourism [
102]. Although the M-MSZ is based on the mathematical derivation and comprehensive determination, there are still some subjective judgments such as the expert survey, the questionnaire survey of the factors affecting the choice of living space and its examination, and requires further verification of many more samples and regions. This is a potential area for our future research.