Spatial Characteristics of Land Use Multifunctionality and Their Trade-Off/Synergy in Urumqi, China: Implication for Land Space Zoning Management

Abstract

Identifying and exploring the spatial characteristics of land use multifunctionality (LUMF) and their trade-off/synergy are the basis for promoting the coordinated development of LUMF, and have significant implications for land space zoning management. In this study, we integrated multi-source data to construct a multi-functional identification system of land use, and quantitatively identified agricultural production function (APF), urban life function (ULF), and ecological function (EF) from grid units. We used the mechanical equilibrium model and Spearman correlation variable analysis to explore the trade-off/synergy between the primary and secondary function of land use. The results show that LUMF has obvious spatial differentiation characteristics and significant composite characteristics. Functionality interweaves and overlaps spatially, creating trade-off/synergy between LUMF. Urumqi as a whole was at a coordinated level (73%). High urban life–low agricultural production and high ecology–low agricultural production were the main types of trade-off/synergy. APF and EF were dominant functions, and there was a significant synergistic relationship. APF and urban life-bearing function had a trade-off relationship. Based on the research results, zoning attempts were made as a reference. Finally, under the framework of regional function theory, we considered the sequential selection process and competition process of LUMF, and put forward proposals for land space zoning management.

1. Introduction

The United Nations 2030 Agenda for Sustainable Development states that more than half of the world’s population currently lives in urban areas. Therefore, urban construction and management are inseparable from the realization of sustainable development goals [1]. The agglomeration of LUMF is the essence of the formation of land space [2,3]. Since the reform and opening up in China, the economic growth-oriented social development has led to great changes in the spatial pattern of the land space and exacerbated conflicts between urban land, ecological land, and agricultural land, such as the disorderly expansion of urban development boundaries, agricultural land encroachment on agricultural land, and ecological environmental degradation [4,5,6,7]. In the context of constructing ecological civilization and deepening reform of the spatial planning system [8], Optimizing and coordinating life, ecology, and production space has become the core essence of zoning management of land space [9]. Under the framework of territorial functions, revealing the interaction between LUMF plays an important role in optimizing and managing territorial space [10].

The formation of regional function reflects the carrying function and feedback mechanism of the ecosystem to human activities, as well as the utilization and dependence of human needs on the ecosystem space [11]. LUMF assumes a comprehensive function in the sustainable development of the human–land relationship, which is comprehensive and diverse. LUMF originated from the agricultural multifunctional theory [12], which was developed and inherited based on the studies of ecosystem services and landscape functions [13,14,15]. The research on LUMF has a different focus from that of agriculture, ecosystem services, and landscape multi-function, in which the former focuses on the coordinated properties of society, economy, and ecology [4,16,17], while the latter focuses more on eco-environmental attributes [18,19,20]. The identification of LUMF is the basis of land space zoning management [4]. Scholars have constructed a LUMF evaluation index system from the aspects of production (economy), life (society), ecology (environment), etc., based on the regional practical problems, research objectives, and policy orientations [21,22]. The methods, including comprehensive index method, fuzzy comprehensive evaluation method, improved mutation series method, full arrangement polygon graphic method, etc., were used to spatially quantify the land use function [23,24,25]. However, the identification method of LUMF is singular, and the complex properties of functions are ignored in the identification process, which weakens the practical applicability of the identification results.

Trade-off/synergy analysis is an efficient method to obtain the relationship between multifunctionalities [26]. Trade-off refers to the increase in one function that leads to a decrease in other functions, showing a competitive relationship between functions. Synergy refers to the increase or decrease in two or more functions, showing a symbiotic relationship in the coordinated development of functions [27]. The analysis of the development law of LUMF must focus on the interaction between natural ecosystems and artificial ecosystems [28]. Artificial ecosystems and natural ecosystems jointly promote the sustainable development of the world, which is important for land and space management [29,30]. At present, trade-off/synergy research mostly focuses on natural ecosystems or artificial ecosystems, and the research to reveal the relationship of trade-off/synergy between natural ecosystems and artificial ecosystems is rare. Currently, the trade-off/synergy relationship of natural ecosystems is mainly discussed in terms of manifestations, research methods, scale effects, etc. [31,32,33,34]. The research on the trade-off/synergy of artificial ecosystems focuses on the cultivated land, and commonly used methods include coupling coordination degree, mathematical, scenario simulation, and prediction [18,35,36]. Regional function takes the sustainable development of comprehensive ecosystems and the maximization of overall human welfare as the basic starting point, and originating from the practice of land space development and taking the identification of regional function and policy formulation as the research objects [37,38]. The principle of regional function serves the optimization of the spatial pattern of land and is mainly applied in the major function zoning, the carrying capacity of resources and environment, and the changing mechanism of the functional pattern. Based on the regional function theory, it has a high operability for the land space division, which can clarify the functional orientation and effectively guide the protection and development of the land space [24]. The competition among LUMF hinders the coordinated development of land space. At present, the research to explore the relationship between LUMF from the perspective of trade-off/synergy and suggestions on land space zoning management for the coordinated development of LUMF are sufficient. Therefore, it is necessary to explore the trade-off/synergy relationship of LUMF in the integrated ecosystem in the framework of regional functional theory research, to explore the nature of the formation of land space and to provide reference for zoning management.

In the construction of the “Belt and Road” strategy, the location advantage and policy orientation accelerated the urbanization process, and the radiation effect of functional agglomeration was significant, which inevitably brought about conflict between LUMF. This phenomenon is more pronounced in large cities in northwestern China, especially Urumqi. During the construction of the “Silk Road Economic Belt”, the city of Urumqi experienced a surge in population and expansion of urban boundaries. Rapid LUMF transformation, making urban land encroach on agricultural land and ecological land, and persecuting local agro-ecological and ecological functions. Therefore, it is necessary to clarify the trade-off/synergy effect of LUMF to provide a theoretical basis for the management of land space zoning in Urumqi. The purpose of this study is to reveal the spatial differentiation characteristics of LUMF in Urumqi, the core node of the “Silk Road Economic Belt”, and its trade-off/synergy relationship. Under the guidance of the spatial ordering rule of regional function, in order to minimize the degree of functional trade-off and the spatial synergy of functional combinations, the purpose of putting forward the management of territorial space division is to enrich the research in this area.

2. Materials and Methods

2.1. Study Area

Urumqi, the capital of Xinjiang Uygur Autonomous Region, consists of 7 districts and 1 county, as shown in Figure 1. It is located in the middle of the northern foot of Tianshan Mountain and the southern margin of Jungar basin, with a geographical range of 86°37′33″~88°58′24″ E and 42°45′32″~44°08′00″ N. It has a total area of 13,800 km², including 522 km² of built-up area.

As the core city of the “Silk Road Economic Belt”, Urumqi is the fastest-growing city in Xinjiang. The unique geographic strategic location provides advantages for its population growth and socio-economic development. In 2020, the permanent resident population was 4.05 million, with an increase of 1.39 million from 2016. In the same year, the GDP was 3337.32 billion yuan, increasing by 916.67 billion yuan compared with 2016. By 2020, the greening coverage rate of the built-up area reached 40.5%, and the urbanization rate exceeded 90%. The rapid growth of GDP and population has prompted dramatic changes in the pattern of land use. From 2000 to 2020, urban land expanded by about 0.58-fold, and grassland and unused land were mostly occupied for the expansion of urban land. The rapid transformation of land use types leads to fierce conflicts between LUMF, which lead to uneven development of land space. Therefore, we took Urumqi as a typical case to explore the trade-off/synergy relationship between LUMF, hoping to provide theoretical support for regional decision makers to optimize land space zoning management.

2.2. Data Sources

The data used in this study mainly include socio-economic data, road traffic data, point of interest (POI), net primary production (NPP), normalized difference vegetation index (NDVI), DEM, population density, land use type, and soil type data (

Table 1. Data sources and description for this study.

Type of Data	Data Content	Data Sources
Vector data	Road traffic data	Chinese Academy of Sciences Resource and Environmental Science Data Center (www.resdc.cn, accessed on 15 July 2020)
	Point of interest (POI) data	Use JAVA programming to crawl Gaode map points of interest
Raster data	Land use/Land cover data (30 m × 30 m)	Chinese Academy of Sciences Resource and Environmental Science Data Center (www.resdc.cn, accessed on 25 July 2020)
	Net primary production (NPP) (500 m × 500 m)	Chinese Academy of Sciences Resource and Environmental Science Data Center (www.resdc.cn, accessed on 30 July 2020)
	Normalized Difference Vegetation Index (NDVI) (1 Km × 1 Km)	Chinese Academy of Sciences Resource and Environmental Science Data Center (www.resdc.cn, accessed on 15 July 2020)
	Population density (1 Km × 1 Km)	Chinese Academy of Sciences Resource and Environmental Science Data Center (www.resdc.cn, accessed on 15 June 2020)
	DEM data (30 m × 30 m)	Geospatial Data Cloud (www.gscloud.cn, accessed on 20 July 2020)
	Soil data (1 Km × 1 Km)	Chinese Academy of Sciences Resource and Environmental Science Data Center (www.resdc.cn, accessed on 15 June 2020)
Socio-economic statistics data	Individual output value of agriculture, animal husbandry, forestry	Urumqi Statistical Yearbook (2019) (www.wlmq.gov.cn, accessed on 20 June 2020)

). The ArcGIS software (version 10.2, Esri, Redlands, CA, USA) platform was used to resample the required spatial data to 500 m, unify the spatial resolution, and convert it into a unified projected coordinate system (WGS_1984_UTM_Zone_46N). A grid cell of 500 m × 500 m with 57,750 grids was created using the fishnet tool to unify the analysis and evaluate the experimental data.

2.3. Research Framework

With this purpose, we comprehensively constructed a LUMF evaluation index system from the aspects of agricultural production function (APF), urban life function (ULF), and ecological function (EF) by integrating multi-source data, and spatialized the secondary functional indicators in geographic grid cells to analyze their spatial characteristics. The level and type of the trade-off/synergy of LUMF were determined based on the mechanical equilibrium model and Spearman variable correlation analysis method. Considering the function sequential selection process and the competition process comprehensively, the trade-off degree between functions is alleviated from the factor level of LUMF formation and development. The purpose of this work is to optimize the land use structure in the functional trade-off area and improve the land function, in order to provide theoretical support and practical guidance for land space zoning management. The research flow chart is shown in Figure 2.

2.4. Identification of LUMF

The agglomeration of land functions is the basis for the formation of land space. There is a “one-to-one” or “many-to-one” relationship between land functions and land space [39,40]. Only relying on land type to determine land function has the problem of imperfect basic expression of land space formation, and it is necessary to refine the land use function with the help of multi-functional evaluation indicators. Some indicators use administrative divisions as statistical units, which cannot directly reflect the secondary functions of land use at the micro-scale. In order to solve these problems, we used the spatial analysis method to quantitatively identify and spatially visualize the APF, ULF, and EF indicators based on the national strategic guidance document “National Territory Planning Outline (2016–2030)” issued by the State Council. LUMF was evaluated on the grid scale, and the results are listed in

Table 2. Identification index system of LUMF.

Primary Functions	Sub-Functions	Indicators	Weights (%)	Quantification Method	Units	References
Agricultural production	Agricultural product supply	Crop supply	34.58	$A_1\times \frac{F_1}{S_1}$	Kg/m2	A1, A2, A3 represent the areas of cultivated land, grassland, forest land, respectively, in a 500 m × 500 m grid cell. F1, F2, F3 represent the total output value of agriculture, animal husbandry, and forestry, respectively, in each district in 2019. S1, S2, S3 represent the total area of arable land, water area, and grassland, respectively, in each district [46].
		Animal husbandry supply	33.45	$A_2\times \frac{F_2}{S_2}$	Kg/m2
		Forestry supply	31.97	$A_3\times \frac{F_3}{S_3}$	Kg/m2
Urban life	Life bearing	Residential carrying	10.62	#	Pop/m2	Characterization of population density.
		Traffic carrying	8.10	#	Km/Km2	The ratio of traffic land mileage to grid area in grid cells.
	Life support	Life service	14.09	ArcGIS overlay analysis	#	Euclidean distance spatialization of POI data of shopping malls, supermarkets (stores), market shopping services, and living services.
		Employment service	12.63		#	Euclidean distance spatialization of POI data of primary, secondary, and tertiary industries.
		Medical service	15.20		#	Euclidean distance spatialization of POI data in general hospitals, specialized hospitals, health centers, and clinics.
		Security services	15.07		#	Euclidean distance spatialization of POI data of government agencies such as alarm points and fire points.
	Cultural education	Education service	13.80		#	Euclidean distance spatialization of POI data of science, education, and culture in kindergartens, primary schools, middle schools, and museums.
		Leisure and entertainment services	10.49		#	Euclidean distance spatialization of POI data of scenic spots and sports leisure.
Ecology	Ecological service	Biodiversity	24.92	$NP{P}_{maen}\times F_{pre}\times F_{tem}\times \left(1-F_{alt}\right)$	#	$NP{P}_{maen}$ is the average annual vegetation net primary productivity. $F_{pre}$ is the annual average precipitation. $F_{tem}$ is the annual average temperature, $F_{alt}$ is the altitude factor [47].
		Windbreak and sand fixation	24.42	$\left(NDVI-NDV{I}_{min}\right)/\left(NDV{I}_{max}-NDV{I}_{min}\right)$	#	NDVI normalized index.
		Soil and water conservation	25.06	${\displaystyle \sum }NPP\times \left(1-K\right)\times \left(1-F_{SLO}\right)$	#	Standardized processing of NPP data, soil erodibility factor and slope factor raster data superimposed.
		Climate adjustment	25.60	${\displaystyle \sum }NPP\times N\times \beta$	g/m2	C is the carbon content of CO2 fixed by vegetation in the atmosphere. N is the proportion of carbon in CO2, which is 27.27%. β = 1.63, which means that vegetation needs to fix 1.63 g of carbon per 1 g of dry matter [48].

.

APF aims to ensure the supply of agricultural products and provide services for human development [41,42]. To enrich the types of agricultural products and ensure national food security, Urumqi vigorously develops the planting industry, breeding industry, and agricultural product processing industry. Cultivated land, grassland, and forest land provide the basis for the development of fruit and vegetable cultivation, poultry breeding, and fish breeding. Therefore, APF was divided into crop production, animal husbandry production, and forest product supply functions. ArcGIS (version 10.2.0.3348, Esri, Redlands, CA, USA) was used to quantify the secondary functions, and the specific quantification methods are listed in Table 2.

ULF undertakes economic and social development in the land space and provides life and spiritual security for the basic survival of human beings. The rational allocation of specific life functions in the land space provides convenient services for residents’ life. With the acceleration of urbanization, life bearing, life support, and cultural education have emerged, reflecting people’s pursuit of a better life. In the spatialization of ULF, the focus was on the convenience of different service types of places to provide service functions to residents; that is, the closer the distance to these places, the stronger the ULF [43]. Population density and road accessibility can reflect the bearing function. The spatial layout characteristics of POI data, such as a company’s industrial structure, hospitals at all levels, supermarkets, fire stations, and police stations, can reflect the life support function. The spatial distribution distance of POI data such as schools and entertainment and leisure places can show the function of cultural education. Based on the Euclidean distance spatial analysis of POI data in ArcGIS (version 10.2.0.3348, Esri, Redlands, CA, USA) software, the smaller the distribution distance between points with the same attribute, the stronger the function. The traffic road density was also calculated in ArcGIS (version 10.2.0.3348, Esri, Redlands, CA, USA). The specific quantification method is shown in Table 2.

Stabilizing ecosystems and improving ecosystem service capabilities are the prerequisites for the harmonious coexistence of human and nature and the sustainable development of society. In Urumqi, the shortage of water resources limits the development of the city due to its location in the arid and semi-arid area of northwest China. Soil erosion and soil desertification limit the development of the agricultural economy. Ecological and agricultural land are occupied for urban expansion. For these reasons, we selected the functions of biodiversity, windbreak and sand fixation, climate adjustment, and soil and water conservation to represent the EF (Table 2). According to the “Technical Guidelines for the Delineation of Ecological Protection Red Lines”, the NPP index method was used to quantify the biodiversity, water and soil conservation, and climate regulation function [44]. We found that there was a negative correlation between vegetation coverage and soil wind erosion; that is, the higher the vegetation coverage, the lower the likelihood and risk of topsoil providing sand and dust caused by strong winds. Among these indicators, the NDVI index can better reflect the level of wind and sand fixation function [45].

After removing the outliers of indicators, spatial processing was performed according to the specific quantification method of each index function (Euclidean distance has been spatialized). The range method is to normalize each sub-index so that the value range of each index was between 0 and 1. Based on the SPSSAU online analysis platform (https://spssau.com, accessed on 15 July 2020), principal component analysis was used to determine the weight of each index (Table 2). The spatial visualization of APF, ULF, and EF was performed in ArcGIS 10.2 software by the comprehensive evaluation method.

$$Y_{xj}=\left(X_{ij}-X_{ij min}\right)/\left(X_{ij max}-X_{ij min}\right)$$

(1)

where $Y_{xj}$ indicates the normalized value, $X_{ij}$ is the original value, and $X_{ij max}$ and $X_{ij min}$ represent the maximum and minimum values in the original data, respectively.

$$APFI={{\displaystyle \sum }}_{n=1}^{3}S{I}_{\left(x,n\right)}\times W_n$$

(2)

$$ULFI={{\displaystyle \sum }}_{n=1}^{3}S{I}_{\left(x,n\right)}\times W_n$$

(3)

$$EFI={{\displaystyle \sum }}_{n=1}^{3}S{I}_{\left(x,n\right)}\times W_n$$

(4)

where $APFI$, $ULFI$, and $EFI$ represent the APF index, ULF index, and EF index, respectively; $S{I}_{\left(x,n\right)}$ represents the index value x in the nth fishing net; and W is the weight of the index n.

2.5. Trade-Off/Synergy between LUMF

The key to the coordinated development of land space depends on the degree of synergy between functions. The unreasonable development of any function can affect the trade-off/synergy of the overall function of the region [49]. Therefore, in this study, we explored the trade-off/synergy between LUMF by citing the mechanical equilibrium in physics [36,50].

As shown in Figure 3, APF, ULF, and EF are the forces distributed in three different directions in the Cartesian quadrant, and the forces are equiangular (120°). From the perspective of land and space management, when each force reaches the average level, the resultant force F acts on the origin O, indicating that the LUMF develops synergistically. If one or more of the forces do not reach an average level, the resultant force F deviates from the coordination point. Specifically, the longer the length of the resultant force, the higher the degree of trade-off between functions. In this study, the polar angle (β) was used to analyze the type of function bias, and the degree of deviation can intuitively reflect which function is dominant in the area. The polar angles of OA, OB, and OC are 90°, 210°, and 330°, respectively. OA stands for APF, OB stands for EF, and OC stands for ULF. F₁ is the resultant force of OA and OC, and F is the resultant force of F₁ and OB. I, II, III, IV, V, and VI represent different functional advantage quadrants (

Table 3. Characteristics of land use multifunctional quadrant.

Quadrant	Polar Angle Range	Agricultural Production Function	Urban Life Function	Ecological Function	Description
I	[330°,0°)∪[0°,30°)	/	−	+	Function of high ecology–low urban life
II	[30°,90°)	+	−	/	Function of high agricultural production–low urban life
II	[90°,150°)	+	/	−	Function of high agricultural production–low ecology
IV	[150°,210°)	/	+	−	Function of high urban life–low ecology
V	[210°,270°)	−	+	/	Function of high urban life–low agricultural production
VI	[270°,330°)	−	/	+	Function of high ecology–low agricultural production

Note: “+”: positive function (above average); “−”: negative function (below average); “/”: uncertain size and direction.

).

$$F1=\sqrt{O{A}^2+O{B}^2+\left|\left(2\times OA\times OB\right)\times \mathit{cos}\left(\angle XOA-\left|\angle XOB\right|\right)\right|}$$

(5)

$$\alpha ={\mathit{sin}}^{-1}\frac{\left|0C\right|\times \mathit{sin}\left(\left|\left(\angle XOB\right)\right|\right)}{F1}$$

(6)

$$F=\sqrt{F{1}^2+O{C}^2+\left|\left(2\times F1\times OC\right)\times \mathit{cos}\left(\left|\left(\angle XOA-\alpha \right)-\angle XOC\right|\right)\right|}$$

(7)

$$\angle FOB={\mathit{sin}}^{-1}\frac{F1\times \mathit{sin}\left(\left|\left(\angle XOA-\alpha \right)-\angle XOC\right|\right)}{F}$$

(8)

$$\beta =\angle XOC-\angle FOC$$

(9)

The Spearman correlation analysis method can quantitatively describe the correlation strength and direction between variables and is more suitable for the analysis of geographic data with nonlinear characteristics. Based on the Spearman rank correlation coefficient to explore the correlation between the secondary functions of land use, we further analyzed the trade-off/synergy between the secondary functions on the basis of the primary functions.

$$r_{ij}=\frac{r_{ij}-r_{ik}\times r_{jk}}{\sqrt{\left(1-r_{ik}^2\right)\times \left(1-r_{jk}^2\right)}}$$

(10)

where r(ij,k) refers to the partial correlation coefficient between two LUMF indicators obtained by determining one LUMF indicator. $r_{ij}$, $r_{ik}$, and $r_{jk}$ are the Spearman correlation coefficients between each pair of LUMF indicators. When the correlation coefficient is greater than 0 and passes the significance test, it is considered that there is a synergistic relationship between the functions. Conversely, there are trade-offs between functions [51,52].

3. Results

3.1. Spatial Pattern of LUMF

3.1.1. Spatial Distribution Characteristics of a Single Function of Land Use

From Figure 4a, the APF index ranges from 0.0 to 0.7, mainly reflecting Toutunhe District, Xinshi District, and the central part of Urumqi County. The animal husbandry supply function in Midong District and Urumqi County is relatively strong. The crop supply function in the Toutunhe area is strong. Urumqi County and Midong District have strong forestry supply functions.

From Figure 4b, the ULF index ranges between 0.0 and 0.9, and the effect of functional agglomeration is significant. In terms of life security function, life-bearing function, and spiritual purification function, Tianshan District, Shayibak District, Shuimogou District, and other central urban areas are at the forefront. The construction land mainly carries the activities of human daily life, so its ULF index is high. They are mainly concentrated in the city center and near the main traffic arteries, and also in the enriched areas of the primary industry. Its spatial distribution is consistent with that of APF.

From Figure 4c, under the dual influence of the natural ecological environment background limit and human activities, the EF index ranges between 0.0 and 0.6. The EF index of the central urban area is weaker than other areas, and has obvious spatial differentiation characteristics. Among them, soil and water conservation functions and windbreak and sand fixation functions occupy a high proportion of ecological service functions, concentrated in the southern mountains of Tianshan Mountain in Urumqi County.

3.1.2. Spatial Distribution Characteristics of Composite Land Use Functions

With the development of society and the continuous improvement of spiritual life, land use types are becoming more and more diversified. Each type of land use gradually assumes multiple functions (Figure 5). According to the judgement of the single land use function, the composite land use function was divided. APF, ULF, and EF values were divided into primary, secondary, and tertiary levels according to the natural discontinuity method. Based on the principle of functional strength and area dominance, nine types of functional combinations in Urumqi were determined. The kernel density map was plotted for spatial visualization, as shown in Figure 6.

From the kernel density map, the function of main ecology–secondary urban life as well as main ecology–secondary urban life–tertiary agricultural production are similar in spatial distribution, and the contiguous areas are concentrated in the southern mountainous area of Urumqi County. The function of main urban life–secondary agricultural production–tertiary ecology as well as balanced agricultural production–urban life–ecology show intertwined and mosaic characteristics in spatial distribution, and they are concentrated in the north of Urumqi. The function of balanced ecology–urban life as well as main urban life–secondary ecology–tertiary agricultural production are scattered in the densely populated areas. The function of main urban life–secondary ecology is concentrated in the center of each district in a reunion shape, which is consistent with the spatial pattern of economic and social agglomeration development zone in Urumqi. The function of main urban life–balance agricultural production and ecological as well as balanced urban life and ecology–secondary agricultural production are complementary in spatial pattern.

In general, the functional combination types dominated by urban life and agricultural production are consistent in spatial distribution and are complementary to the functional combination type dominated by ecology. The former is mainly concentrated in the urban center, while the latter is distributed around the central urban area. That is, the agglomeration effect of the same type of functional combination type is strong, and the spatial agglomeration effect of different functional combination types is weak, which means that there may be a trade-off relationship between regional land use and multifunctionality.

Various human needs lead to the generation of LUMF, which make functions interweave and overlap in space, resulting in conflict and competition among LUMF. Therefore, quantitative identification and trade-off/synergy diagnosis of LUMF are urgently needed to achieve land space zoning management.

3.2. Characteristics of Trade-Off/Synergy Relationship of LUMF

3.2.1. Trade-Off/Synergy Level of LUMF

As shown in Figure 7, the resultant force (F) reflects the trade-off/synergy level of LUMF. In ArcGIS (version 10.2.0.3348, Esri, Redlands, CA, USA), the resultant force F is classified into four types by the natural discontinuity method, namely, high synergy type (0.00 ≤ F < 0.22), low synergy type (0.22 ≤ F < 0.38), mild trade-off type (0.38 ≤ F < 0.55), and severe trade-off type (0.55 ≤ F ≤ 1.26).

From Figure 8a, the trade-off/synergy index between LUMFs is in the range of 0.00~1.26, which is generally at the synergy level (73%), and there is a significant functional trade-off relationship. The type of trade-off in the Toutun River District (area a) is mainly reflected in agricultural land, which accounts for 67% of the land patch. The trade-off type in Urumqi County (area b) is mainly reflected in the natural pasture and woodland in the southern Tianshan Mountains, accounting for 90% of the trade-off type land patch, showing that the southern mountainous area is the core of heavy trade-off, and the degree of functional trade-off gradually weakens and spreads outward. The trade-off type in area c is mainly reflected in the natural pasture at the border of Shuimogou and Midong district, accounting for 13% of the trade-off type of land patches. There is contiguous trade-off type in area d, and the trade-off type is mainly reflected in sandy land, accounting for 74% of the trade-off type land patches, showing that the trade-off degree increases with the increase in urban distance.

3.2.2. Trade-Off/Synergy Type of LUMF

According to the characteristics of resultant force (F) and polar angle (β), the trade-off/synergy types of LUMF were divided into urban life–ecology (quadrants I and IV), urban life–agricultural production (quadrants II and V), and agricultural production–ecology (quadrants III and VI) (Table 3). The resultant force of LUMF is distributed in the whole functional quadrant, mostly in quadrants V and VI, indicating that the dominant function is urban life and ecology. The trade-off type is mainly high urban life–low agricultural production and high ecology–low agricultural production (Figure 9).

1. Functional trade-off/synergy types of urban life–ecology. This functional type is subdivided into the function of high ecology–low urban life (I) (0.65%) and high urban life–low agricultural production (IV) (16.93%). Among them, the functional trade-off/synergy type of high urban life–low ecology is the main type in Urumqi. Midong (45.78%), Shayibak (26.67%), and Tianshan (21.45%) have a high proportion of this functional type.

2. Functional trade-off/synergy types of urban life–agricultural production. This functional type is subdivided into high agricultural production–low urban life (II) (1.71%) and high urban life–low agricultural production (V) (34.68%). Among them, high urban life–low agricultural production is the main functional trade-off/synergy type. The proportion of this type is 70.66% in Shayibak, 48.01% in Tianshan, 47.01% in Xinshi, 39.15% in Urumqi County, and 33.62% in Dabanchen.

3. Functional trade-off/synergy types of agricultural production–ecology. This functional type is subdivided into high agricultural production–low ecology (III) (4.98%) and high ecology–low agricultural production (VI) (41.05%). Among them, high ecology–low agricultural production is the main functional trade-off/synergy type. The proportion of this type is 61.79% in Dabanchen, 55.06% in Shuimogou, 48.22% in Toutunhe, 44.67% in Urumqi County, and 39.22% in Xinshi.

4. Discussions

4.1. Trade-Off/Synergy of LUMF

The objective existence of ecological occupiability, resource and environmental carrying capacity, spatial organization of economic activity, and concentration and dispersion of population are important factors for the generation and development of LUMF [11]. The complex interrelationships between regional natural ecosystem systems and artificial ecosystems have led to a trade-off/synergy between LUMF [53]. Geopolitical factors such as geographical advantages and the natural environment can promote changes in land use functions, while objective factors such as national policies and regional planning can influence the direction and size of functional changes [54]. The mechanical balance model can reflect the trade-off/synergy type between primary functions, but it cannot characterize the trade-off/synergy relationship between secondary functions. Therefore, the Spearman correlation analysis method was introduced to explore the trade-off/synergy between secondary functions.

In this study, we verified the global existence of synergies between LUMF (

Table 4. Spearman correlation between primary land use functions.

	Average Value	Standard Deviation	Agricultural Production	Urban Life	Ecology
Agricultural production	0.026	0.076	1
Urban life	0.622	0.215	0.332 **	1
Ecology	0.263	0.211	0.416 **	0.383 **	1

** p < 0.01.

) and different degrees of trade-offs/synergies between sub-functions (Figure 10), showing the characteristics of spatial overlapping and synergistic development of functions.

From Table 4, there is a significant synergistic relationship between APF and EF, and EF is the dominant function. Ecological occupancy and resource and environmental carrying capacity serve as constraints, restricting the quality of the development of the primary industry, resulting in different advantageous functions in different regions. The research results show that animal husbandry supply function and soil and water conservation function have high synergy, indicating that grassland, forest land, and cultivated land also exert ecological effects while producing economic benefits [55]. The synergistic effect of animal husbandry supply function and climate regulation function is consistent with the conclusion that grassland is the main carbon sink type [56]. The abuse of chemical fertilizers and pesticides reduces the quality of cultivated land and causes ecological deterioration in the surrounding areas, thus showing a weak synergy between crop supply function and ecological function [57].

The synergistic relationship between ULF and EF is the second significant, and ULF is the dominant function, indicating that the living space has some ecological functions. The spatial organization of economic activities has caused conflicts between urban development boundaries and ecological protection red lines, resulting in a weak synergistic effect between secondary functions of urban life and secondary ecological functions [58]. The industrial-based economic development model reduces the quality of the ecological environment in the built-up area and makes the climate adjustment function of the construction land lower [59]. The greening rate of the built-up area has reached 37.4%, indicating that the ecological environment is also at a good level while undertaking the functions of life, which is closely related to the improvement of local residents’ awareness of ecological protection and the government’s ecological priority planning policy [60].

The synergistic relationship between ULF and APF is the weakest, and ULF is the dominant function. Except for the trade-off relationship between the life-bearing function and the secondary function of agricultural production, other secondary functions are synergistic. There is a strong competitive relationship between urban living land and agricultural production land, which is consistent with the research results of the increase in the area of construction land and the occupation of cultivated land in Urumqi [61]. The urbanization process has caused the expansion of built-up areas, encroaching on agricultural land, and reducing the potential economic value of forest land and grassland. The improvement of basic service facilities and the increase in the urban greening rate make the urban life function index higher than the agricultural production function index [56].

It is worth noting that the trade-off/synergy of LUMF always takes the ULF as the dominant function in Urumqi, which is due to the difference in the LUMF classification system, and it also reflects the sequential selection of function. In this study, the construction of the classification system includes the land planning factors mentioned in the text and the layout factors of the secondary and tertiary industries in urban living areas. The layout of secondary and tertiary industries in urban areas mainly serves to solve the employment problem of urban population, ensure the living standard of residents, and then improve the urban population capacity and increase urban attraction. Therefore, in this classification system, the secondary and tertiary industries are divided into life security functions, and the production function only refers to the APF provided by land types such as agriculture, forestry, and animal husbandry.

4.2. Suggestions on LUMF Zoning and Sustainable Development

Determining the trade-off/synergy relationship between LUMF is the basis of zoning, which is the most common means of land space zoning management [62]. Coupling coordination degree, comparative advantage index, cluster analysis, and correlation analysis have been well applied in the spatial division of national land; however, these methods are mostly used at the county level as the evaluation unit [63,64]. The mechanical balance model has been demonstrated to be effective in quantifying the trade-off/synergy level of LUMF and identifying its types under the grid cells [65]. However, when such methods are used to evaluate and analyze trade-off/synergy relationships of the LUMF, they are mostly concentrated in economically developed areas, and there are few studies and analyses on economically underdeveloped areas. Therefore, taking the core node city of the “Silk Road Economic Belt” as an example, the trade-off/synergy relationship of LUMF was studied using the mechanical balance model, as shown in Figure 11. On that basis, some suggestions on land space zoning management are put forward.

In the process of optimizing land space zoning management, it is necessary to consider the sequential selection process and competition process of LUM, prevent the evolution of regions with synergy levels to regions with trade-off levels, alleviate functional trade-offs, and promote the balanced development of functional space. According to the constraints of regional ecological occupiability and carrying capacity of resources and the environment, as well as the leading role of society and culture, efficient and intensive land use policies and zoning management schemes are proposed for areas with low functional synergy and trade-off, so as to achieve coordinated development of regional functions.

(1) Functional trade-off/synergy optimization zone for urban life–ecology. How to coordinate APF and EF on the basis of ensuring ultra-low emissions is a key issue that needs to be addressed in such areas. In the process of functional sequence selection, ecological function is used as the first sequence. Therefore, under the condition of satisfying the constraints of ecological occupiability and carrying capacity of resources and the environment, social and cultural factors are fully considered. Our recommendations are to scientifically use the types of land use in urban and rural development areas, and expand the proportion of ecological land in living space; rationally develop natural reserves and scientifically understand the impact of natural tourism development on the ecological environment, so as to mitigate the contradiction between them; and develop a factory pollution emission supervision mechanism and establish an ecological corridor between industrial land and other lands to ensure regional ecological security.

(2) Functional trade-off/synergy optimization zone for urban life–agricultural production. Therefore, the spatial organization of economic activities and the spatial concentration of population have become the determinants of regional LUMF development. Under the basic conditions constrained by natural factors, the function of life has the right of priority over the function of production. Our recommendations are to restrict the development of areas with strong ecological occupiability and large carrying capacity of resources and environment, so as to alleviate the contradiction between urban development boundaries and ecological protection redlines, and actively promote the construction of high-standard farmland; change the traditional agricultural planting mode, enhance the accessibility of traffic roads, and improve the infrastructure service facilities in rural areas, so that APF and ULF can be coordinated; and scientifically plan and accurately utilize scattered land use types in urban areas while ensuring the quality of production and life in the region [66].

(3) Functional trade-off/synergy optimization zone for agricultural production–ecology. In the sequential selection of functions, the constraints of ecological occupiability are given priority, followed by the carrying capacity of resources and environment. Our recommendations are to protect basic farmland in areas of concentrated agricultural development, avoiding the damage of inefficient agricultural production to the ecological environment; understand regional characteristics of agriculture, guide the symbiotic development of tertiary industry and agricultural production, and promote the large-scale industrialization of agricultural and sideline product processing; and develop an ecological compensation mechanism for returning farmland to the forest (grassland) and promote the development of ecological agriculture, leisure agriculture, and forest agriculture [67].

4.3. Uncertainties and Challenges for Further Research

This study has uncertainties in the following aspects. (1) Selection and quantification of index factors. This is because there is no unified standard for classifying the LUMF in academia [68]. The selection of indicators has obvious data limitations and personal subjective tendency, and the quantitative method of indicators has limitations. In the Euclidean distance spatialization of POI data, the influence of terrain and actual road conditions is ignored. (2) Temporal factors in LUMF trade-off/synergy. Land use has significant changes over time, so the functional trade-off/synergy relationship has functional intensity changes and evolution trends in time series. This is mainly due to the limitation of data acquisition, such as continuous POI data, precipitation, and temperature data. (3) Limitations on the use of the mechanical equilibrium model. This is due to the requirement that there is no collinearity among the multifunctional variables of land use.

In future research, multidisciplinary theoretical knowledge will be used to enrich the connotation of LUMF. Moreover, we will quantify the secondary functional indicators from the dimension of multi-source data, so that the quantitative results of the indicators are closer to the actual situation. According to the major function zoning, we will scientifically delineate the “three-zones and three-lines” and implement the management of land space zoning.

5. Conclusions

In this study, we took Urumqi, the core city of the “Silk Road Economic Belt”, as an example, identified LUMF from geographical grid cells by integrating multi-source data, and then analyzed its spatial characteristics. We used the mechanical equilibrium model and Spearman correlation variable analysis to explore the trade-off/synergy between the primary and secondary functions of land use. The conclusions are as follows.

LUMF had obvious spatial differentiation characteristics, and the composite characteristics between functions were significant. The APF, ULF, and EF indices were in the ranges of 0.0–1.7, 0.0–0.9, and 0.0–0.6, respectively. APF was concentrated in Toutunhe District, Xinshi District, and Urumqi County. ULF was concentrated in old towns such as Tianshan District. EF was weaker in the central urban area than in other areas. The trade-off/synergy index between LUMF varied from 0.00 to 1.26 and was generally at the synergy level (73%). Areas with a high level of synergy usually had a better ecological environment, and the trade-off level of central urban areas with high population density and economic activity was higher. The trade-off/synergy types were mainly high urban life–low agricultural production and high ecology–low agricultural production. APF and EF were the dominant functions, and there was a significant synergistic relationship between them. In addition, there was a trade-off between APF and urban life-bearing function. Based on the trade-off/synergy relationship between LUMF, Urumqi was divided into urban life–ecology, urban life–agricultural production, and agricultural production–ecology functional trade-off/synergy optimization zones. Finally, based on the framework of regional function theory, we comprehensively considered the sequential selection process and competition process of LUMF, and put forward proposals for land space zoning management.

The principle of mechanical equilibrium proves that it shows applicability in the study of primary functional trade-off/synergy in land use, and the problems of quantification of trade-off/synergy and the determination of dominant function can be better solved, but there is deficiency in the study of trade-off/synergy of secondary function, which needs to be supplemented by the theories and methods of statistics and other disciplines.