Function Evolution of Oasis Cultivated Land and Its Trade-Off and Synergy Relationship in Xinjiang, China

Abstract

Cultivated land (CL) performs irreplaceable manifold functions in social stability, economic development, and ecological security, which is also essential for the accomplishment of the United Nations Sustainable Development Goals 2030 Agenda. China is the world’s most populous country, and it has important reference significance for the realization of the multi-function synergistic management of CL in China by revealing the evolution characteristics of cultivated land functions (CLFs) and the interaction between CLFs. However, the research to date has tended to focus on the eastern coastal areas and the central traditional agricultural areas of China. This study focuses specifically on Xinjiang, the main area of the arid region of northwest China. The connotations of social, economic, and ecological functions of oasis cultivated land (OCL) in Xinjiang were first discussed from a system theory perspective. Then, an evaluation index system of CLFs was constructed. On this basis, the evolution characteristics of CLFs and the interaction between CLFs in Xinjiang from 1990 to 2018 were quantitatively evaluated. Findings suggest that: (1) the economic function of the OCL in Xinjiang is strengthening, while the ecological function is degrading and the social function remains stable. Overall, the evolution of CLFs in Xinjiang was first dominated by ecological and social functions and then became economic-function-oriented; (2) the synergistic relationship between CLFs is weakening and the trade-off relationship is increasing over time. The trade-off effect between the economic function and other functions of OCL is strengthened gradually due to the OCL-use activities dominated by the economic function. This study not only enriches the regional content of CL multi-function research but can also provide reference for decision-making for the sustainable utilization and multi-function synergistic management of OCL in Xinjiang, China.

1. Introduction

A roadmap for “the future we want” in terms of human well-being and environmental sustainability has been constructed under the United Nations Sustainable Development Goals 2030 Agenda [1]. Satisfying the rising food consumption demand of humans on the premise of neither compromising multiple ecosystem services nor exceeding the safe limits of planetary boundaries will be a global challenge in sustainable land management [2,3,4,5]. As a natural resource essential for human survival and development, cultivated land (CL) has irreplaceable functions in safeguarding food security, maintaining ecosystem balance, supporting the livelihoods of farmers and herdsmen, and promoting economic development in rural areas [6]. Furthermore, these functions are correlated and interrelated [7]. The investigation of the sustainable management of CL from a multifunctional perspective has become a new paradigm in existing research on land science worldwide [8,9,10,11].

China is the world’s most populous and the largest developing country. In addition to concerning the food security of the 1.4 billion Chinese people, China’s sustainable management of CL is also immediately correlated with global food security and the United Nations Sustainable Development Goals 2030 Agenda. In that case, the way of achieving sustainable utilization and multifunctional synergistic management of CL in China during agricultural development is a crucial problem to be solved. To this end, various cultivated land functions (CLFs) should be first evaluated in a scientific manner. Then, the law of temporal and spatial evolution of CLFs and the interaction between functions and formation mechanisms need to be revealed. These are the scientific basis for promoting the realization of the multi-functional synergistic management of CL in China.

Considerable literature has grown up around the widely recognized multi-function property of CL. In terms of the functional connotation of CL, scholars hold the view that CL is a joint production system that produces both commercial and non-commercial outputs. Its underlying function is to provide products and services in conformity with the demands of human survival and development, which is normally associated with society, economy, and environment [12,13,14,15,16]. In terms of the classification of CLFs, there are currently two main classification approaches. One is to categorize CLFs in provisioning, regulating, supporting, and cultural functions with reference to the classification of ecosystem services [12,15,16,17,18]. The other is either to divide CLFs into social, economic, ecological, and cultural functions by inserting regional elements such as society, economy, ecology, and culture into the CL utilization system from the aspect of system theory [13,19,20,21,22,23,24] or to split CLFs into producing, living, and ecological functions [25,26]. In terms of the evaluation method of CLFs, the comprehensive functional index evaluation method is the most used quantitative method at present. Given its advantages of the application at multiple scales, the ability to describe multiple function types, and facilitating comparison between functions, it has been widely used in existing studies [12,13,16,19,20,22,25,26]. In addition, CLFs have also been evaluated using the value evaluation method [27], the material quality evaluation method [28], the energy analysis method [29], and the policy back-stepping method [30].

Affected by the changes in the products and services required for human survival and development in different periods, CLFs have shown distinct characteristics in different socio-economic stages [25,31], which is reflected in the shift from an emphasis on productivism-oriented functional needs to the more diversified functional requirements such as economy, society, culture, and ecology [9,10,32]. The manifestation of CLFs also varies in different regions. CLFs in Western developed countries, for instance, have been transformed markedly, with more emphasis on the functional requirements of CL in ecological services, landscape aesthetics, cultural entertainment, and the promotion of rural sustainable development [6,32]. By comparison, CL in China, a populous developing country, plays a momentous part in safeguarding food security and supporting farmers’ livelihoods [30]. Further, significant spatial differences can also be found in the CLFs in different regions of China [23,26]. To be specific, the economic function of CL in developed urban agglomerations and their hinterland areas is declining [19,31], while the recreational functions of their CL are on the rise [24,33]. In the meantime, farmers in major agricultural production regions cause particular stress on the functions of elderly care and employment of CL [34]. Conversely, the ecological function of CL is more significant in key ecological function areas to the west of the “Hu Line” [26]. In terms of the interaction between CLFs, the research methods of ecosystem service trade-off and synergy have been adopted for measurement, such as the correlation coefficient method [21,23], bivariate spatial autocorrelation method [22,28], gray T-relational model [35], and coupling coordination degree [19], etc. However, in specific research processes, the above quantitative methods need to be selectively used in combination with their respective research scales and data support. The correlation between functions becomes more complex because of the diversity of CLFs and the preferences of human demands and choices. The synergy and trade-off relationships can be observed extensively between CLFs, which are presented in significant spatio-temporal heterogeneity [19,21,22,23,36,37]. Hence, it is of great necessity to research multiple functions of CL based on local conditions.

To sum up, previous studies have suffered from shortcomings, albeit while producing fruitful results. Regarding research content, a unified classification and evaluation system has not yet been formed in academia, since the multi-functional connotation, classification, and evaluation of CL is under continuous discussion for improvement, which should be further investigated and perfected. Regarding research areas, coastal regions in eastern China and traditional agricultural regions in central China with high levels of urbanization and modernization have been studied in most cases, with few studies concentrated on economically backward and ecologically vulnerable areas in western China. Regarding research perspectives, many scholars focus their research on the evaluation of CL multifunctionality and the revelation of the spatial pattern characteristics of a single time section, lacking intensive research on the long-term multifunctional evolution characteristics of CL and the interaction between various functions. In fact, this should have been an important basis for measuring whether the utilization of CL in an area is reasonable and sustainable and also important decision-making reference for propelling the synergistic management of multiple CLFs.

Located in the northwest inland of China, Xinjiang is the largest provincial-level administrative region and the main area of the arid region in northwest China [38]. Oases are the substantial basis for human survival and development in the arid region [39]. As the main body of the modern oasis of Xinjiang, CL with distinct regional features performs irreplaceable multi-fold functions in social stability, economic development, and ecological security. However, Xinjiang has become a hot spot in China’s increment of CL over the past three decades [40,41]. The highly intensive CL activities lead to severe environmental problems while bringing about remarkable social–economic benefits [42]. Apparently, oasis cultivated land (OCL) utilization in Xinjiang is confronted with challenges in food, ecological, and social security, which should be transformed urgently towards sustainable utilization and multi-function synergistic management.

This study presents a case study of Xinjiang, the main area in the arid region of northwest China. Multi-function evolution characteristics of CL over a long period and the interaction between the functions are extensively examined in this paper. This study seeks to enrich the regional content of CL multi-function research and sets out to provide a decision-making reference for the sustainable utilization of CL and the multi-function synergistic management of CL in Xinjiang. The specific objectives of this study are: (i) investigating the functional connotation and classification of OCL in Xinjiang and constructing a multi-function evaluation index system for OCL, (ii) quantitatively evaluating and presenting the temporal and spatial evolution laws of the functions of OCL in Xinjiang from four temporal sections (1990, 2000, 2010, and 2018), (iii) investigating the trade-off and synergy relationship between functions of OCL in Xinjiang and their formation mechanisms, and (iv) making policy suggestions for the multi-function synergistic management of OCL in Xinjiang in accordance with the conclusions.

2. Materials and Methods

2.1. Study Area

Xinjiang, short for the Xinjiang Uygur Autonomous Region, 73°40′–96°23′ E and 34°25′–49°10′ N, is located in the middle of Eurasia and in the northwest of the People’s Republic of China, which has a typical temperate continental arid climate. It has three mountain ranges, namely the Altai Mountains, Tianshan Mountains, and Kunlun Mountains, as well as two inland basins—the Junggar Basin and the Tarim Basin. They form a topography and landform of “three mountain ranges surrounding two basins” and an ecological landscape pattern of “mountain–oasis–desert”, as shown in Figure 1; among them, CL is the main body of the artificial oasis of Xinjiang [38].

As of 2018, CL in Xinjiang reached an area of 5.24 million hectares, accounting for roughly 3% of the total land area of Xinjiang. The total population of the region has reached 24.87 million, with the share of the urban population being 50.91% [43]. Nearly half of Xinjiang’s population lives in rural areas due to lagging urbanization, with almost 80% of the total rural employees engaged in agriculture [43]. As the main spatial carrier of agricultural development and the production system most closely associated with oasis ecological security and rural social stability, OCL in Xinjiang bears typical regional characteristics and important research significance.

2.2. Methods

2.2.1. Connotation, Classification, and Evaluation Index of the OCL Function in Xinjiang

The arid region is a geographical whole consisting of three landscape systems (mountains, oases, and deserts) that are interrelated and mutually restricted in the ecological process [38,39]. To be specific, the mountain system provides resources such as water and soil to support the formation and development of oasis areas [38]. The oasis system, as an area with relatively higher productivity and a center of human production and life in the arid region, is a hub of gathering and exchanging materials and energy between mountain and desert systems. The oasis system includes four subsystems of population, society, economy, and the ecological environment [44]. The desert system has a vast area and harsh environment and is ecologically vulnerable [38].

The CL system is a complex system coupled with “nature–society–economy” [13,14,19,25]. The function of CL refers to the ability of CL to provide products and services that meet the needs of human survival and development during a certain period of development, utilization, and protection and under the comprehensive action of its constituent elements [25]. OCL is a key role in connecting the population, economy, society, and ecological subsystems of the oasis area [44]. When the population, society, economy, and ecological environment are embedded in the process of OCL utilization, the social, economic, and ecological functions of OCL are correspondingly formed (Figure 2). The use of OCL is essentially to meet the needs of human beings for the above three functions of CL. The rationality of the human’s CL-use activities in the oasis area is also indirectly reflected by the changes of the three functions of CL and the interaction between them.

Specifically, the social function of cultivated land (SFC) in Xinjiang is reflected in the important role of OCL in safeguarding food security and supporting farmers’ employment in the process of social and economic development of the oasis region, which is directly associated with the stability of Xinjiang. The Xinjiang oasis region is separate from the main grain-producing areas of China. Moving grain over long distances increases transportation costs and risks. Meanwhile, there are also uncertainties in grain production due to climatic aridity and vulnerable ecology. Therefore, OCL plays an important role in ensuring food self-sufficiency in oasis areas. The contracted farmland is the basic means of production and the workplace for farmers since the household contract responsibility system was implemented by the Chinese government in rural areas in the 1980s [45]. In particular, the employment of farmers is extremely reliant on OCL and agriculture due to the lagging development of secondary and tertiary industries in the oasis area of Xinjiang. Therefore, the OCL is prominent in promoting the active workforce in rural areas. The function of OCL in grain production and supply was described by two indicators, including the grain yield per unit area and the grain self-sufficiency rate. The function of OCL in employment security was described by two indicators, including the agricultural dependency of employment and the population-carrying capacity of OCL. The specific calculation methods of the above indicators are shown in

Table 1. Evaluation index system of oasis cultivated land function in Xinjiang.

Function Layer	Index Layer	Calculation Method	Unit
Social function	Grain yield per unit area	Total grain output/sown area of crops	ton/hm2
	Grain self-sufficiency ratio	Total grain output/(total regional population × 400 kg) × 100%	ton/hm2
	Agricultural dependency of employment	Number of agricultural employees/rural employees × 100%	%
	Population-carrying capacity of OCL	(Rural population − animal husbandry population)/CL area	person/hm2
Economic function	Contribution of agriculture to the national economy	Gross output of the planting industry/GDP × 100%	%
	Management benefit of OCL	Gross output of the planting industry/CL area	104 yuan/hm2
	Farmer’s income level	Per capita net income of rural households	yuan
	Planting proportion of non-food crops	Sown area of non-grain crops/total sown area of crops × 100%	%
Ecological function	Ecological service value of CL conversion area a	Calculation based on ArcGis software and Xie’s method	109 yuan
	Farmland biodiversity b	−∑pi·ln(pi), pi is the proportion of the sown area of various crops	-
	Agrochemical load of farmland c	(Chemical fertilizer + pesticide + plastic film) /CL area	1015 sej/hm2
	Net carbon sink level of farmland d	(CL carbon sink − CL carbon release)/CL area	ton C/hm2

Notes: The attributes of the evaluation indicators in the above table are all positive indicators, except the agrochemical load of farmland. ^a: The coefficients of ecological service value in the study period are calculated as 16,599, 59,085, 24,521, 95,289, 115,082, and 2921 for cultivated land, forests, grasslands, rivers/lakes, wetlands, and unused lands, respectively, in Xinjiang, using Xie’s calculation method [46], in combination with the food production and price levels in Xinjiang. The unit is yuan/hm². ^b: The biodiversity of cultivated land is calculated using Song’s calculation method [30], with twelve crop planting types including wheat, rice, corn, beans, potatoes, cotton, oil seeds, sugar beets, vegetables, melons for fruit, and alfalfa [43]. ^c: Usages of chemical fertilizers, plastic films, and pesticides were converted into solar energy values with a unified dimension before the sum calculation with reference to the method of Chang [47], and their conversion rates of solar energy values are 4.7 × 10⁹ sej/g, 6.38 × 10⁷ sej/g, and 2.69 × 10⁹ sej/g, respectively. ^d: The carbon sink of cultivated land in Xinjiang is calculated by using Tian’s calculation method [48]. Furthermore, the carbon emission of cultivated land involves the carbon emission caused by the utilization of cultivated land, with the carbon emission coefficient listed as fertilizer 1.285 kg C/kg [49], pesticide 4.9341 kg C/kg, plastic film 5.18 kg C/kg [48], agricultural diesel 0.5927 kg C/kg, irrigation 20.476 kg C/hm², and tillage 3.126 kg C/kg hm² [50].

.

The economic function of cultivated land (ENFC) in Xinjiang is reflected in the important role of OCL in promoting regional economic development and increasing farmers’ income by producing agricultural products and participating in market transactions to obtain economic benefits. The oasis is an area with relatively higher productivity in the arid region. Furthermore, agriculture is the initial economic form of the oasis, the foundation of national economic development, and the safeguard of industrial development in oasis areas [44]. On the one hand, agricultural producers increase the output and economic benefits of agricultural products through fully utilizing agricultural production conditions such as water, soil, light, and heat in the oasis. On the other hand, the economic output of OCL is also propelled by expanding the sown area of cash crops based on the continuous adjustment of the agricultural planting structure. The function of OCL in propelling regional economic development was described by two indicators, including the contribution of agriculture to the national economy and the management benefits of OCL. Furthermore, the function of OCL in increasing farmers’ incomes was described by two indicators, including the per capita net income of rural households and the planting proportion of non-food crops. The specific calculation methods of the above indicators are shown in Table 1.

The ecological function of cultivated land (ELFC) in Xinjiang is reflected in the important role of OCL in maintaining oasis stability, protecting biodiversity, maintaining farmland soil health, and increasing carbon sequestration. Oasisization and desertification are two elementary geographical processes with mutual feedback [39]. As the expansion of CL is a main form of oasisization, large quantities of natural ecological land will inevitably be sacrificed due to the excessive expansion of CL. As a result, it will result in breaking the dynamic balance between oasisization and desertification, threatening the stability of the oasis. Since diversified farmland crop is an ecological attribute of the OCL resource system, planting various crops is essential for protecting the biological diversity in the OCL [30]. Soil health is the basis for guaranteeing the function of OCL. Specifically, the increased soil nutrients, the enhanced soil fertility, and the control of weeds and pests can benefit from the rational utilization of agricultural chemicals. However, the irrational and excessive application of agricultural chemicals might bring forth negative feedback, such as damaging soil structure and exacerbating soil environmental load [11]. Oasis crops store carbon dioxide absorbed from the atmosphere via photosynthesis in an organic form, greatly contributing to carbon sequestration. Moreover, humans can also increase their carbon sink outputs through the management measures of OCL. Furthermore, there are also carbon emissions in the utilization of OCL in the agriculture production process [48]. Hence, the ELFC in Xinjiang was described by four indicators, including the ecological service value of the OCL conversion area, farmland biodiversity, farmland agrochemical load, and farmland net carbon sink level. The specific calculation methods of the above indicators are shown in Table 1.

2.2.2. Measurements for CLFs

To begin with, raw data were processed using MinMaxScaler to remove the influence of the index dimension. It is calculated as follows:

$$\mathrm{Positive} \mathrm{index}: x_{ij}=\frac{X_{ij}-\min \left\{X_j\right\}}{\max \left\{X_j\right\}-\min \left\{X_j\right\}}$$

(1)

$$\mathrm{Negative} \mathrm{index}: x_{ij}=\frac{\max \left\{X_j\right\}-X_{ij}}{\max \left\{X_j\right\}-\min \left\{X_j\right\}}$$

(2)

When the index is positive, the greater the better; when the index is negative, the smaller the better. County-level data of the same index at different temporal sections were processed for standardization to safeguard the comparability of data in different temporal sections.

Based on the function index, three functions of OCL were then calculated as shown below:

$$F\left(SFC\right)={\displaystyle \sum _{j=1}^{4}W\left(SF{C}_j\right)}\times f\left(SF{C}_j\right)$$

(3)

$$F\left(ENFC\right)={\displaystyle \sum _{j=1}^{4}W\left(ENF{C}_j\right)}\times f\left(ENF{C}_j\right)$$

(4)

$$F\left(ELFC\right)={\displaystyle \sum _{j=1}^{4}W\left(ELF{C}_j\right)}\times f\left(ELF{C}_j\right)$$

(5)

where $F\left(SFC\right)$, $F\left(ENFC\right)$, and $F\left(ELFC\right)$ are the social, economic, and ecological function indices of CL, respectively; $W\left(SF{C}_j\right)$, $W\left(ENF{C}_j\right)$, $W\left(ELF{C}_j\right)$ and $f\left(SF{C}_j\right)$, $f\left(ENF{C}_j\right)$, $f\left(ELF{C}_j\right)$ are the weight of the $j$-th characterization index and the standardized indices for the evaluations of social, economic, and ecological function evaluation of OCL, respectively. Moreover, all function indices ranged from 0 to 1, and the larger the index, the stronger the corresponding function. Characteristic indices corresponding to three CLFs were summarized using the equal-weight method. In other words, the weight value of each index was 0.25.

2.2.3. Analysis Method for the Evolution of CLFs

The spatio-temporal characteristics of the function evolution of OCL at the county scale of Xinjiang were described using the function grading method and the longitudinal comparison coefficient method. All CLFs indices in four-time sections (1990, 2000, 2010, 2018) of 84 counties in Xinjiang were first converted into a sequence of 336 units for each function through accumulation during function grading, and each CLF index was then divided into a high-value area, mid-value area, and low-value area using the natural breaking point grading method based on ArcGIS10.7 software.

The longitudinal comparison coefficient of the function is calculated as follows:

$$LC{C}_{ij,t}=F_{ij,t2}/F_{ij,t1}$$

(6)

where $LC{C}_{ij,t}$ is the longitudinal comparison coefficient of the $j$-th function of OCL at the period $t$ in the county (city) $i$ in the study area; $F_{ij,t1}$ and $F_{ij,t2}$ are the $j$-th CLF indices in the county $i$ of the study area in the base and end periods, respectively. If $LC{C}_{ij,t}$ is greater than 1, it indicates that the CLF is enhanced in the study period; if not, it indicates a gradual decline.

2.2.4. Analysis Method of Interaction between CLFs

As only four time sections (incl. 1990, 2000, 2010 and 2018) were involved in the calculation of the CLFs of varying counties in the study, the demands for sample size in the typical mathematical statistics methods cannot be satisfied. In that case, the trade-off and synergy among the functions of OCL were analyzed using the gray T-relational model improved by Huang and Wang [51]. With no strict requirement for sample size, the model adopted describes the strength, size, and order of the relations between factors using the order of gray correlation, which is superior in terms of uniqueness, symmetry, stability, and standardization. It can also be used for reflecting the positive and negative correlations of the sequence [51]. The model can be calculated as follows:

For the temporal section, $\left[a,b\right],b>a\ge 0$, assuming

$$\Delta t_k=t_k-t_{k-1},\begin{array}{c}\end{array}\left[a,b\right]={\displaystyle \underset{k=2}{\overset{n}{\cup }}\Delta t_k,\begin{array}{c}\end{array}}\Delta t_k{\displaystyle \cap \Delta t_{k-1}}=\varphi ,\begin{array}{c}\end{array}k=2,3,\cdots ,n$$

(7)

The two original time series set on $\left[a,b\right]$ are:

$$X_1=\left\{x_1\left(t_1\right),x_1\left(t_2\right),\cdots ,x_1\left(t_n\right)\right\};\begin{array}{c}\end{array}{X}_2=\left\{x_2\left(t_1\right),x_2\left(t_2\right),\cdots ,x_2\left(t_n\right)\right\}$$

(8)

Of which, $x_i\left(t_k\right),\begin{array}{c}\end{array}i=1,2;\begin{array}{c}\end{array}k=2,3,\cdots ,n$ is not equal to 0, then

$$y_i\left(t_k\right)=\frac{x_i\left(t_k\right)-x_i\left(t_{k-1}\right)}{x_i\left(t_{k-1}\right)},\begin{array}{c}\end{array}i=1,2;\begin{array}{c}\end{array}k=2,3,\cdots ,n$$

(9)

It represents the relative increment of the sequence $X_i$ from the time point $t_{k-1}$ to the time point $t_k$, that is, the relative initialization is performed on the original sequence.

$$D_i=\frac{{\displaystyle \sum _{k=2}^n\left|y_i\left(t_k\right)\right|}}{n-1},\begin{array}{c}\end{array}i=1,2$$

(10)

It represents the mean of the absolute value of the relative increment of the sequence in various periods of time.

$$z_i\left(t_k\right)=\frac{y_i\left(t_k\right)}{D_i},\begin{array}{c}\end{array}i=1,2;\begin{array}{c}\end{array}k=2,3,\cdots ,n$$

(11)

It represents the equalization of the relative increment of the sequence $X_i$ from the time point $t_{k-1}$ to the time point $t_k$.

The improved gray T-relational coefficient $\xi \left(t_k\right)$ of the sequences $X_1$ and $X_2$ from the time point $t_{k-1}$ to the time point $t_k$ is presented below:

i

When $z_1\left(t_k\right)$ and $z_2\left(t_k\right)$ are not zero simultaneously,

$$\frac{sign\left(z_1\left(t_k\right)·z_2\left(t_k\right)\right)}{\frac{1}{2}+\frac{1}{2}\left|\left|z_1\left(t_k\right)\right|-\left|z_2\left(t_k\right)\right|\right|+\frac{1}{2}\left[1-\frac{\min \left(\left|z_1\left(t_k\right)\right|,\left|z_2\left(t_k\right)\right|\right)}{\max \left(\left|z_1\left(t_k\right)\right|,\left|z_2\left(t_k\right)\right|\right)}\right]+\frac{1}{2}\left[\frac{\max \left(\left|y_1\left(t_k\right)\right|,\left|y_2\left(t_k\right)\right|\right)}{\min \left(\left|y_1\left(t_k\right)\right|,\left|y_2\left(t_k\right)\right|\right)}\right]}$$

(12)

ii

When $z_1\left(t_k\right)$ and $z_2\left(t_k\right)$ are not zero simultaneously, or when $y_1\left(t_k\right)$ and $y_2\left(t_k\right)$ are zero simultaneously, $\xi \left(t_k\right)=1$; when $y_1\left(t_k\right)$ and $y_2\left(t_k\right)$ are not zero simultaneously, then $\xi \left(t_k\right)=0$.

Where $sign\left(z_1\left(t_k\right)·z_2\left(t_k\right)\right)$ is the sign function, showing the positive and negative correlation. To be specific, when $sign\left(z_1\left(t_k\right)·z_2\left(t_k\right)\right)>0$, it indicates a positive correlation; when $sign\left(z_1\left(t_k\right)·z_2\left(t_k\right)\right)<0$, it shows a negative correlation.

The improved gray T-relational coefficient $r$ of the sequences $X_1$ and $X_2$ from the time point $t_{k-1}$ to the time point $t_k$ is presented below:

$$r=\frac{1}{b-a}{\displaystyle \sum _{k=2}^n\Delta t_k·\xi \left(t_k\right)}$$

(13)

When $t_k=k\begin{array}{c}\end{array}k=1,2,\cdots ,n$, it can be simplified as:

$$r=\frac{1}{n-1}{\displaystyle \sum _{k=2}^n\xi \left(t_k\right)}$$

(14)

When $-1\le r<0$, the sequences $X_1$ and $X_2$ are negatively correlated. That is to say, there is a trade-off relationship between both, and the closer the value approaching to −1, the stronger the trade-off. When $0<r\le 1$, the sequences $X_1$ and $X_2$ are positively correlated. In other words, there is a synergistic relationship between both. The closer the value approaches 1, the stronger the synergy is. When $r=0$, the sequences $X_1$ and $X_2$ are independent of each other.

2.3. Data Sources and Processing

Data of CL area in 1990 and 2000 were acquired from the Statistical Yearbook of Xinjiang Uygur Autonomous Region and those in 2010 and 2018 were from the Ledger Data of Land Change Survey of Xinjiang Uygur Autonomous Region. Other relevant socio-economic statistical data were obtained from the Xinjiang Statistical Yearbook and the Statistical Yearbook of Xinjiang Prefectures of the corresponding years. The missing data of some indices were replaced by the mean of adjacent years or the data of adjacent years. Outputs of the national economy and the planting industry were converted into comparable prices, with 1990 as the base period according to the price index of the corresponding year. Furthermore, the per capita net income of rural households was converted into the comparable price, with 1990 as the base period according to the consumer price index of rural residents of the corresponding year. The land-use remote sensing monitoring data of Xinjiang were obtained from the Resource and Environment Science Data Center of the Chinese Academy of Sciences (http://www.resdc.cn (accessed on 10 June 2021)) with a spatial resolution of 30 m. Administrative boundary vector data were from the basic geographic databases of China at a 1:1 million scale released by the China National Catalog Service for Geographic Information (http://www.webmap.cn (accessed on 5 January 2021)).

3. Results

3.1. Spatial and Temporal Evolution Characteristics of CLFs

3.1.1. General Characteristics

Evolution trends of the SFC, ENFC, and ELFC of Xinjiang varied significantly in the study period (Figure 3). The ENFC was increasing, with its function index increasing from 0.23 in 1990 to 0.44 in 2018, showing a relative increment of 91.30%. The ELFC was degrading gradually, with its function index declining from 0.51 in 1990 to 0.43 in 2018, showing a relative decrement of 15.69%. The SFC was stable as a whole, showing an increasing trend and then a decreasing trend. Its function index first increased from 0.40 in 1990 to 0.47 in 2010 and then decreased to 0.42 in 2018, showing a relative increment of 5%. To sum up, the evolution of CLFs in Xinjiang was first dominated by ecological and social functions and then became economic-function-oriented.

3.1.2. Spatial Differences

• Social function;

The spatio-temporal evolution characteristics of the SFC at the county level in Xinjiang are reported in Figure 4a–d. The numbers of counties included in the high-value, mid-value, and low-value areas of the SFC were 25, 38, and 21, respectively, in 1990. Specifically, the high-value area was mainly distributed in the Hotan Administrative Offices (HTAO) and Kashgar Administrative Offices (KSAO) in southern Xinjiang, in which most counties are national-level poverty-stricken areas, with OCL as the main source of livelihood. The low-value area was mainly distributed in the Altay Administrative Offices (ATAO) in northern Xinjiang. This is a traditional grassland pastoral area in Xinjiang, and its grassland resources performs a more significant social function than OCL. Furthermore, the mid-value area was mainly distributed in the main agricultural production regions along the southern and northern slopes of the Tianshan Mountains. By 2018, 31, 36, and 17 counties were contained in the high-value, middle-value, and low-value areas, respectively, for the SFC. Furthermore, their corresponding spatial distribution also changed. The high-value area changed into a scattered distribution from a centralized distribution. The high-value area in southern Xinjiang was mainly distributed in HTAO, western KSAO, and the northern Aksu Administrative Offices (ASAO), where the OCL was the main source of farmers’ employment and income increase due to the dense population and low urbanization. Furthermore, the high-value area in northern Xinjiang was mainly distributed in Tacheng Administrative Offices (TCAO), the Ili River Valley, and the Changji Hui Autonomous Prefecture (CHAP), where the social function of OCL was significant due to its rapid growth in grain production and supply capacity. Besides, the low-value area was scattered and located in animal husbandry counties along the border, and the mid-value area was distributed throughout the southern and northern regions of Xinjiang in clusters.

The longitudinal comparison coefficient of the SFC ranged from 0.51 to 6.06 from 1990 to 2018, as shown in Figure 4e. The SFC was strengthening in 48 counties and decreased in the remaining 36 counties in 2018, compared with that in 1990. Most of the 48 counties with strengthened social functions were distributed in northern Xinjiang, while the 36 counties with decreased social functions were mostly in southern Xinjiang. The SFC at the county level in Xinjiang presents the regional characteristics of “increasing in the north and decreasing in the south”, with an unchanged pattern of “weak in the north and strong in the south”.

2.

Economic function;

The spatio-temporal evolution characteristics of the ENFC at the county level in Xinjiang are reported in Figure 5a–d. There were 0, 19, and 65 counties included in the high-value, mid-value, and low-value areas in 1990, respectively, presenting a low ENFC as a whole. The mid-value area was distributed in Turpan, KSAO, and western HTAO. Other counties fell into the low-value area. The ENFC at the county level has been enhanced significantly by 2018, with 29, 43, and 12 counties included in the high-value, mid-value, and low-value areas for the ENFC, respectively. A part of the high-value area was distributed in the striped area in the Manas River Basin and the Tarim River, which are mostly agricultural production areas with satisfactory agricultural production suitability and high-level productivity of OCL. Another part is distributed in Bayangol Mongol Autonomous Prefecture (BYMAP), Turpan city, and Hami city, which presents a more prominent ENFC through shifting the planting structure of OCL from traditional food crops to high-yield economic crops, such as cotton and special forest fruits. The low-value area was dominated by part of the animal husbandry counties, which are rich in grassland resources, with a high proportion of animal husbandry in the agricultural industrial structure. The mid-value area was mainly distributed in HTAO, eastern CHAP, and ATAO.

The longitudinal comparison coefficient of the ENFC ranged from 1.01 to 5.51 from 1990 to 2018, as shown in Figure 5e. In 41 counties of Xinjiang, the ENFC has improved faster than Xinjiang’s average level, mainly distributed in ATAO, CHAP, Turpan city, Hami city, and BYMAP. To sum up, the ENFC at the county level in Xinjiang sees a generally upward trend, which is “faster in the northeast and slower in the southwest”, whereas the ENFC shows a distribution pattern that is “stronger in the southeast and weaker in the northwest”.

3.

Ecological function.

The spatio-temporal evolution characteristics of the ELFC at the county level in Xinjiang are reported in Figure 6a–d. There were 29, 37, and 18 counties included in the high-value, mid-value, and low-value areas, respectively, for the ELFC in 1990. The high-value area was distributed along the northern slope of the Tianshan mountains, ASAO, and eastern KSAO; the mid-value area was distributed in various prefectures in southern Xinjiang; the low-value area was distributed in northern and southern Xinjiang. As of 2018, there have been 29, 43, and 12 counties included in the high-value, mid-value, and low-value areas for the ELFC, respectively. The high-value area was distributed in eastern CHAP and the Ili River Valley, which has good biodiversity of farmland and a high level of net carbon sink of OCL. A part of the low-value area was located in BYMAP, KSAO, and ASAO, which have large populations and a high utilization rate of agricultural water and soil resources with a prominent imbalance between population and land. The other part of the low-value areas was distributed in Turpan city and Hami city, where economic crops such as cotton, special fruits, and vegetables have been substituted for food crops, with higher usage of agrochemicals for economic crops than that of food crops [52], causing more significant damages to ELFC due to the high environmental load of OCL in these areas.

The longitudinal comparison coefficient of the ELFC ranged from 0.32 to 1.68, as shown in Figure 6e. The ELFC was declining in 50 counties distributed in the western end of the northern slope of the Tianshan Mountains, BYMAP, ASAO, KSAO, Turpan city, and Hami city, and increasing in the remaining 34 counties. To sum up, the ELFC at the county level in Xinjiang is generally dominated by a downward trend, which is “faster in the southeast and slower in the northwest”, whereas the ELFC shows a distribution pattern that is “stronger in the northwest and weaker in the southeast”.

3.2. Trade-Off and Synergy Relationship between CLFs

3.2.1. Temporal Scale

Gray T-relational coefficients between two out of the three CLFs in Xinjiang are shown in

Table 2. Gray T-relational coefficients between CLFs in Xinjiang.

Function Group	Gray T-Relational Coefficients
	1990	2000	2010	2018
SFC—ENFC	0.1812	0.1039	0.0685	−0.0491
SFC—ELFC	0.0859	0.0530	0.1025	0.1160
ENFC—ELFC	0.1814	0.0716	−0.0779	−0.1021

. The findings show that the relationship between CLFs in Xinjiang is mainly synergistic, but the synergistic relationship between CLFs is weakening and the trade-off relationship is increasing over time. Also, there are differences in the interaction relationship between CLFs.

The SFC and ENFC were synergistic in 1990 (r = 0.1812), which gradually weakened in 2000 and 2010. As of 2018, the relationship between both has changed from synergy to trade-off (r = −0.0491). The SFC and ELFC were synergistic, while the synergistic intensity first increased and then decreased throughout the research period. As a whole, the synergy between the two is strengthened as the gray T-relational coefficient (r) developed into 0.1160 in 2018 from 0.0859 in 1990. The synergy between ENFC and ELFC in 1990 (r = 0.1814) weakened in 2000 and changed into a trade-off in 2010 (r = −0.0779) that further strengthened in 2018 (r = −0.1021).

In the early stage of the research period, OCL in Xinjiang was adopted to satisfy the regional food self-sufficiency, with a low proportion of economic crops. The agricultural production method was backward, and the increase in OCL production relied on intensive labor input and the low level of soil and water development. There were fewer inputs of production-increasing factors such as pesticides, chemical fertilizers, and plastic film; human activities exerted less influence on the ecosystem of OCL. Affected by the above factors, the three functions of OCL in Xinjiang showed a relatively low-level synergy at this stage.

In the later stage of the research period, some rural labor forces were replaced by labor-saving agricultural machines, along with the improved urbanization level and the progress in agricultural mechanization, weakening the SFC’s ability to absorb farmer employment. With the continuous improvement of grain production and self-sufficiency, local governments and farmers have expanded the proportion of cash crops in the planting structure under the inducement of economic benefits and widely used agrochemicals to increase farmland production. Besides, a large amount of ecological land was reclaimed to expand the scale of OCL and increase the benefits. In consequence, the ELFC showed a decline with the continuous strengthening of ENFC under the disturbance of high-intensity human agricultural activities. The SFC and ENFC have also gradually changed from a synergistic relationship to a trade-off relationship.

3.2.2. Spatial Scale

Trade-offs and synergies relationship between the three CLFs at the county level in Xinjiang and their spatial differences are reported in Figure 7.

The SFC and ENFC at the county level in Xinjiang were dominated by synergy (incl. 45 counties) and concurred with trade-offs (incl. 39 counties), as shown in Figure 7a. Those counties with a synergistic relationship between the two are distributed in TCAO, Ili River Valley area, ATAO, eastern CHAP, HTAO, Kizilsu Kyrgyz Autonomous Prefecture, and eastern KSAO in Xinjiang. Those counties with a trade-off relationship are distributed in BYMAP, Turpan city, Hami city, western CHAP, and western KSAO.

The SFC and ELFC at the county level in Xinjiang showed a pattern of equivalence between synergies (incl. 42 counties) and trade-offs (incl. 42 counties), as shown in Figure 7b. Those counties pertaining to the synergy are distributed in the TCAO, Bortala Mongol Autonomous Prefecture (BTMAP), Turpan city, Hami city, eastern CHAP, and southern BYMAP. Those counties belonging to the trade-off are distributed in HTAO, KSAO, northern Ili Valley, and ATAO.

The ENFC and ELFC at the county level in Xinjiang showed a pattern dominated by trade-offs (incl. 53 counties) and concurred with synergies (incl. 31 counties), as shown in Figure 7c. Those counties belonging to synergy are distributed in Ili Valley, the eastern section of the northern slope of Tianshan mountains, and eastern HTAO. Those counties belonging to trade-off are distributed in the Tacheng Basin, the western end of the northern slope of the Tianshan Mountains, Turpan city, Hami city, BYMAP, ASAO, KSAO, and Kizilsu Kyrgyz Autonomous Prefecture.

4. Discussion

4.1. Similarities and Differences between Xinjiang and Other Regions in the Functional Evolution of CL

Previous studies have shown that the overall evolution of CLFs has obviously different characteristics in different socio-economic stages [9,25,31,32]. This conclusion has also been ascertained in this case study of Xinjiang. The SFC and ELFC of Xinjiang were relatively significant between 1990 and 2010; the ENFC has been strengthened since 2010, more significantly than the SFC and ELFC. This also reflects that the CLFs in Xinjiang has shifted from serving food and survival needs in the past to serving the new needs of increasing farmers’ income and promoting rural economic development currently. The difference is that the CLFs in Western developed countries has been significantly transformed from productivism to post-productivism [6,32], and the CLFs in China has also transformed since 2006 and are moving towards the synergy of multiple functions [30]. However, the CLFs in Xinjiang are characterized by polarized evolution (with the strengthening economic functions and weakening ecological and social functions). The utilization of CL focusing on the pursuit of economic functions inevitably has a trade-off effect on other cultivated land functions, thereby affecting the sustainable use of OCL.

From the perspective of changes in various CLFs, previous studies have shown that the SFC are constantly weakening with the progress of urbanization and agricultural modernization, which has been ascertained in studies conducted nationwide [30], in the eastern coastal areas [31,35,36], and in the central traditional agricultural areas of China [19,28]. However, the SFC in Xinjiang was stable as a whole in the research period. It is different from the findings in other regions of China. There are two causes for this special phenomenon. On the one hand, Xinjiang is far away from China’s main grain-producing areas, so the SFC for Xinjiang’s grain self-sufficiency is relatively stable. On the other hand, the urbanization in Xinjiang (50.91% as of 2018) is low with the backward development of secondary and tertiary industries. Oasis farmland remains a stabilizer for carrying the employment of the majority of farmers when the abilities to absorb employment in urban and non-agricultural industries are insufficient.

The ENFC in Xinjiang has been enhanced. The trend is similar to previous findings in other regions of China [19,28]. However, there is a difference that the growth of economic function is more prominent in OCL. That is because the oasis area shows favorable collaboration between water, soil, light, and heat resources with flat and contiguously distributed land, as well as high productivity of CL [44]. More than that, the proportions of economic crops such as cotton, oilseeds, melons, and fruits are rising in the agricultural planting structure of Xinjiang on the basis of food self-sufficiency, contributing to the manifestation of the ENFC.

The degradation of the ELFC is commonly accompanied by the strengthening of the ENFC in different areas of China. Such a phenomenon is more prominent in Xinjiang, which can be boiled down to two reasons. First, Xinjiang has been a hot spot in China’s increment of CL over in the past 30 years [41]. The significant expansion of CL in Xinjiang has occupied a large amount of natural ecological land and scarce water resources, which has caused a severe negative impact on the ecological stability of the oasis area [39,42]. Second, the increased output in the CL of Xinjiang has a strong dependence on chemical fertilizers and pesticides. The mean application intensity of chemical fertilizers (358.23 kg) and pesticides (4.23 kg) per hectare of CL in Xinjiang is much higher than that of the fertilizers (125.53 kg) and pesticides (3.68 kg) used worldwide in the same period [37]. Meanwhile, Xinjiang is also a main cotton production area and the province with the largest consumption of plastic film in China. Its residual plastic film content of farmland is 4–5 times higher than the mean level in China [53], which leads to a high agrochemical load on CL and damages the ecological environment.

OCL in Xinjiang is essential for ensuring food security, supporting farmers’ employment and social stability, promoting economic development in rural areas, and maintaining ecological security. The rationality of regional CL-use activities can be reflected indirectly by the interaction between various functions of OCL. According to the research findings, the synergy among the three functions of Xinjiang’s CL is weakening over time, whereas the trade-off is strengthening. This indicates that the interaction between CLFs changes with human preferences for CLFs and changes in CL-use patterns and intensity, which has also been verified in some research in other regions [19,21,22,23,36].

4.2. Policy Recommendations of Multi-Function Synergistic Management of OCL in Xinjiang

The trade-off effect between the economic function and other functions of CL is strengthened gradually due to the CL-use activities dominated by the economic function in Xinjiang. This is in fact a revelation to us, that the unreasonable utilization of OCL in Xinjiang should be addressed by managing the trade-offs between CLFs and guiding the synergistic development of social, economic, and ecological functions of OCL in the process of future utilization. Four countermeasures and suggestions are proposed in response to the above research contents.

First, the SFC in the food production and self-sufficiency of Xinjiang should be stabilized. It is the geographical location of Xinjiang that determines the high cost and risk of long-distance grain transportation. Also, there are also uncertain risks in food production under global climate change and the vulnerability of regional ecological backgrounds. Therefore, it is imperative to depend on the OCL to deal with the problem of food supply of Xinjiang. On the one hand, the permanent basic CL protection system should be strictly implemented, with strict control over the occupation of high-quality CL resources during urbanization. On the other hand, the planting structure of CL should be optimized and adjusted in combination with the dynamics of population and grain demand in different counties and cities, preventing the excessive “non-grain” orientation of CL. Moreover, the construction of high-standard farmland should be also promoted in the oasis area to strengthen the irrigation security and agricultural mechanization of grain production, contributing to increasing the grain production capacity of OCL.

Second, the SFC in ensuring farmers’ employment of Xinjiang should be restrained. The level of urbanization in Xinjiang is relatively low, and the development of the non-agricultural economy is relatively backward. This has resulted in many rural laborers relying on OCL resources for their livelihoods, exacerbating artificial pressure on OCL and resulting in the dysfunction of OCL. Therefore, it is necessary to step efforts in channeling rural surplus labor to non-agricultural employment and reduce the pressure on oasis farmland. To be specific, Xinjiang’s population urbanization rate is far lower than the national level in the same period, which has a large room for improvement. Hence, new urbanization and the development of non-agricultural industries should be propelled for attracting surplus rural laborers to work in cities and industrial parks. In addition, villages and towns developing labor-intensive industries based on their actual conditions should be supported based on the policy of various provinces aiding Xinjiang, so as to guide rural laborers to participate in non-agricultural employment nearby.

Third, the ELFC in maintaining oasis stability and the soil health of Xinjiang should be restored. The past agricultural development model characterized by large-scale expansion in Xinjiang led to severe coercive effects on ecological land and scarce water resources in the arid region and threatened the stability of the oasis. Also, non-point source pollution of farmland and jeopardies to soil health were caused due to the over-reliance on agrochemicals for increasing the production of CL yields. In response to these, on the one hand, the strictest possible water resource management system should be implemented to ensure an appropriate scale of OCL through setting CL with water. On the other hand, the layout of OCL should be optimized based on agricultural suitability, with Xinjiang’s land space planning taken into consideration. Meanwhile, the encroachment of ecological space by CL should be strictly controlled. In addition, agricultural producers should rely on the progress of agricultural science and technology to improve the green and low-carbon utilization of CL, thereby reducing the unreasonable application of chemical fertilizers and pesticides and improving the recycling and resource utilization of waste agricultural film.

Fourth, the ENFC in promoting rural economic development and increasing farmers’ income in Xinjiang should be optimized. The fast growth of the ENFC in Xinjiang is achieved by the expansion of the agricultural scale and the adjustment of planting structure (the increasing proportion of economic crops planted). However, the growth potential is limited due to the single source of the economic benefits of OCL. Therefore, the integrated development of agriculture and the secondary and tertiary industries should be guided to extend the means of increasing farmers’ income from low-level CL planting to other links, such as processing and sales. By doing so, the income-increasing chain can be expanded by extending the industrial chain. Besides, ecological, green, and organic agriculture should be developed on the basis of advantageous resources, such as light, heat, and ecology in Xinjiang. Meanwhile, regional green agricultural brands should also be created to improve the market competitiveness of agricultural products, contributing to the conversion of ecological advantages to economic advantages.

4.3. Advantages, Limitations, and Prospects

This study selects Xinjiang, the main area of the arid region of northwest China, with unique regional features, as the study area. The connotations of social, economic, and ecological functions of OCL were first discussed based on the system theory. Then, an evaluation index system of CLFs was constructed. On this basis, the evolution characteristics of and the interaction between oasis CLFs in Xinjiang from 1990 to 2018 were investigated. The analytical framework proposed is applicable to related research on oasis CLFs in other arid regions apart from providing insights into the regional perspective of CL multi-function research.

Though the research has obtained certain results, there are also some deficiencies. The evaluation index system established in this paper might be deficient in some aspects due to the limited understanding of oasis farmland functions and the difficulty in data acquisition. It remains to be improved by considering the connotation and scientific evaluation methods of social, economic, and ecological functions of OCL together with multidisciplinary knowledge. What’s more, this paper only considers the change of CLFs at the county level in Xinjiang. In the future, more in-depth studies at the level of typical villages and towns in oasis should be considered, which will provide a corresponding decision-making reference for the differentiated policy formulation of the multi-function synergistic management of CL at different scales.

5. Conclusions

This paper constructs a multi-function evaluation index system for OCL on the basis of discussing the function connotation and classification of OCL in Xinjiang. On this basis, the evolution characteristics of and the interaction between CLFs in Xinjiang from 1990 to 2018 are quantitatively evaluated. The main conclusions include:

(1)

Evolution Features of CLFs: There were obvious differences in the evolution trends of the three functions of OCL in Xinjiang from 1990 to 2018. Among them, the economic function continued to increase, the ecological function was gradually degraded, and the social function was relatively stable. In general, the evolution of CLFs in Xinjiang was first dominated by ecological and social functions and then became economic-function-oriented. At the county level of Xinjiang, the SFC presents the regional characteristics of “increasing in the north and decreasing in the south”; the growth rate of the ENFC shows the regional characteristics of “growing fast in the northeast and slow in the southwest”; the deceleration of the ELFC presents the regional characteristics of “growing fast in the southeast and slow in the northwest”.

(2)

Interaction between CLFs: The relationship between SFC and ENFC in Xinjiang has evolved into a trade-off from synergy. The evolution characteristics of the relationship between ENFC and ELFC are the same as above. Although the relationship between SFC and ELFC are synergistic, it shows a different synergistic intensity tendency that first increased and then decreased. Spatially, the SFC and ENFC at the county level in Xinjiang are dominated by synergy. The ENFC and ELFC at the county level in Xinjiang shows a pattern dominated by trade-offs. The SFC and ELFC at the county level in Xinjiang show a pattern of equivalence between synergies and trade-offs.