Spatio-Temporal Pattern and Influence Mechanism of Cultivated Land System Resilience: Case from China

Abstract

The study of cultivated land systems from the perspective of resilience is of great significance for the innovation of the research paradigm of cultivated land use and the rational utilization and protection of cultivated land. This study aims to explain the theoretical connotations of cultivated land system resilience (CLSR), construct an evaluation system and zoning rules for CLSR, and take 30 provinces of China as case study areas to explore the influencing factors of CLSR, so as to provide a reliable governance plan for the sustainable development of cultivated land. The results show that: (1) CLSR refers to a sustainable development ability that CLS—by adjusting the structure and scale of internal elements—absorbs and adapts to internal and external disturbances and shocks to the maximum possible extent, abandons the original inapplicable state, creates a new recovery path, achieves a new balance, and avoids system recession. (2) The overall CLSR of the 30 provinces showed an upward trend, and the degree of polarization of the distribution pattern was gradually intensified and experienced a transition process from “leading by resource and ecological resilience—equilibrium of each resilience—leading by production and scale structural resilience”. (3) In the north, east, and south coastal areas of China, CLSR mainly consists of the major evolution areas and the stable development areas; the potential excitation areas of CLSR are mainly concentrated in the central and western regions of China; the CLSR-sensitive lag areas and degraded vulnerable areas are mainly distributed in the northwest and southwest of China. (4) Water resource endowment has a strong influence on CLSR, while social economy mainly influences CLSR through ‘economic foundation-superstructures’ and ‘economic development-factor agglomeration’. (5) According to the different CLSR zones, CLSR was strengthened mainly from the aspects of driving factor agglomeration, building factor free-flow systems, and multi-means support.

1. Introduction

Sustainable cultivated land use is vital to food security and is also an important link in maintaining the sustainable development of the economy and society. However, with the rapid progress of industrialization and urbanization, quality, quantity, and ecological crises of cultivated land use have appeared in many regions of the world [1,2].The operation rules of the cultivated land use system (CLS) require further study in order to stimulate the system’s resistance to external interference, thus better dealing with this crisis. The coping ability of the CLS can be interpreted as cultivated land use system resilience (CLSR).

From the rise of physical resilience, the derivation of ecological resilience, and the sublimation of evolutionary resilience, the resilience theory has gone through a developmental process from the pursuit of a single equilibrium to the exchange of multiple equilibrium states, and finally to the abandonment of equilibrium and an emphasis on cross-scale dynamic interactions. Its connotation has evolved from an initial emphasis on the efficiency and speed of recovery to an emphasis on the ability to adapt and transform [3]. Under the social–ecological system, a resource system constantly adjusts its structure to adapt to external interferences and stimulation, reduce damage, and create new development paths by virtue of its self-regulation ability and external protection power—endowed by social and economic systems—so as to improve its ability of sustainable development, which can be understood as resilience. Resilience is different from single recovery and elasticity, and is in fact the coupling of the two [4]. An overview of international resilience studies shows that their research objects are gradually shifting from abstract economic resilience [5,6] and ecological resilience [7] to concrete urban resilience [8,9], community resilience [10,11] energy resilience [12,13], and landscape resilience [14,15]. Their research perspectives have focused on public management [16], disaster prevention and mitigation [17,18], planning guidelines [19,20], industrial development [21,22], and supporting infrastructure construction [23,24]. The purpose of the current study is mainly focused on the classification and evaluation of resilience [25,26] and its optimization strategy [27,28], and the research ideas of previous studies show the development from theoretical construction to practical guidance.

The CLS involves the coupling of many subsystems such as ecology, the economy, and society, including the whole process of ‘input and output’, which can be controlled manually and has the condition of resisting external forces. It is an organic complex with special functions formed by the interaction of cultivated land used for human production activities and the elements contained in the upper and lower space of a certain area [29]. Considering economic and social development and the construction of ecological civilization to be equally important, and that food security has become an important component of national security, the transformation of cultivated land use [30,31], cultivated land quality assessment and protection [32], sustainable intensification of cultivated land use [33], functional evolution of cultivated land [34], and the biodiversity of cultivated land use systems [35] have become the focus of international attention and research. Based on the perspectives of correlations between and the compositions of the morphological structure of cultivated land [36], the evaluation of driving factors [37], changes in property rights relationships [38], and the coupling of functions [34], understanding the problems that exist in cultivated land use and exploring the optimization paths of cultivated land protection and sustainable use have become a research topic with important theoretical and practical significance. However, the level of socio-economic development [39], farming households’ behavior and utilization efficiency [40], the degree of cultivated land fragmentation [41,42], cultivated land carrying capacity [43], land consolidation, farmland property rights, and classification and grading policies and systems [44,45] all have a significant impact on the transformation of cultivated land use and the development of cultivated land multifunctionality [46] (Bahar, Kirmikil, 2021).

Many international scholars have evaluated the resilience of agricultural systems and regions from the perspectives of the economy, stakeholder, and land use based on the adaptation theory, and have comprehensively summarized the process of agricultural adaptative changes through a variety of methods; the results show that resilience is the most important factor in the comparison of agricultural vulnerability and resilience, and that adaptability represented by resilience is an important link to achieving sustainable agricultural development—which is positively correlated with sustainability, as they complement each other [47]. The attribute of farm resilience includes the various activities of farmers in organizing, developing, and utilizing farms, which is expressed as the diversity of livelihoods [48]. In the framework of the social ecosystem, understanding the interactions between farm change characteristics and the social natural system, and cultivating the ability of farm systems to renew and reorganize themselves in the face of external disturbances is an important part of realizing the sustainable development of farms [49]. Farms face disturbances from conflict, natural disasters, climate change, demand for food, price fluctuations, and health crises [50]; its resilience can be improved by the number of farm facilities, farmers’ age, agricultural terms of trade, agricultural output, and other aspects [51,52].

The main objectives of this study are to (i) try to define the conceptual connotations of CLSR, and to construct an index system and zoning rules based on the characteristics of the CLS itself from the perspective of resilience theory; (ii) take 30 provinces of China as evaluation units, analyze the temporal and spatial evolution characteristics of CLSR in China, divide the types of CLSR according to zoning rules, and clarify its influence mechanisms; (iii) provide new ideas and a scientific basis for the promotion of cultivated land protection and sustainable utilization by discussing management ideas and setting future research directions for CLSR.

2. Theoretical Analysis of CLSR

2.1. Definition of the Connotation

The social system is in constant evolution [53], and the CLS is a crucial link. The CLS is constantly disturbed by natural, social, economic, institutional, and other factors in the process of its operation, which shows spiral evolution dynamics and exists in a non-linear and non-equilibrium state. In case of external interferences—under the joint action of internal and external organizations—the CLS will adjust its own structure, form, scale, function, input, and output, and form a system organization in line with the new environment through the reconstruction of the internal elements of its system; find a new development path to resist, absorb, and slow down the damage caused by interference; and continue to maintain the stable operation of the system. Resilience corresponds to transformation, which applies to the ability of any process in the framework of a social–ecological system to improve its own adaptability [3,4]. Therefore, in the face of crisis, the CLS abandons the original equilibrium state and takes the initiative to learn, adapt, and transform—showing resilience in the form of self-recovery and renewal, which can also be interpreted as ‘social–ecological system resilience’. Therefore, based on resilience theory and thinking, CLSR refers to a sustainable development ability that the CLS, by adjusting the structure and scale of its internal elements, absorbs and adapts to internal and external disturbances and shocks to the maximum possible extent, abandons the original inapplicable state, creates a new recovery path, achieves a new balance, and avoids system recession.

2.2. The CLS Adaptive Cycle Process

The adaptive cycle theory is the basis of resilience theory and is an important theory for understanding the internal structural evolution of a system, the response of a system to disturbances, and the transformation of a system under the framework of the social–ecological system, which includes four cycle stages—namely, exploitation, conservation, release, and reorganization [6,54,55]. The process of the CLS creating a new recovery path and transforming to a new equilibrium can be regarded as the adaptive cycle of cultivated land [49]. In the exploitation stage, the internal input and output, the cultivated land scale and form, and the other elements and structures of the CLS respond to external stimulation; the combination and collocation of elements begins to change, the system structure changes from single to complex, and the CLS continues to develop—at which time, CLSR plays a higher role. After entering the conservation stage, the CLS develops rapidly and becomes less sensitive to external disturbances. The internal components of the system begin to solidify and the connections among them become closer. The development of the system begins to slow down and gradually reaches a new equilibrium state, and the functions of the cultivated land also change accordingly. At this point, the role of CLSR begins to decrease, which is the period when the system is most vulnerable to external disturbances and is most out of balance. Stimulated by the new environmental dilemma, CLS enters the release stage; creative destruction makes the structure of the cultivated land use no longer fit for the current social environment, which impacts the connections and structures between system elements, and makes it lose its elasticity to resist interference. The function of the cultivated land begins to decline, and CLSR starts to play a greater role. A CLS with strong resilience will enter the reorganization stage, the internal element structure of the system will be completely broken, and each element will become independent. The original function of the CLS will fade away, and a new development will begin; the adaptive cycle will be realized again so as to promote the transformation of the cultivated land use. However, less resilient cultivated lands will break out of the cycle and fail to play their role in the reorganization stage, and so will completely collapse—such as cultivated lands that lose their fertility and cannot be recovered due to disasters like drought, floods, and fires caused by natural or man-made factors, and cultivated lands that are abandoned because the rural labor force leaves or is far away from large-scale cultivated land areas.

2.3. The Composition of CLSR

CLS is an important part of the social production system. When the input of people, land, technology, and money within the system meets the requirements for the sustainable utilization of cultivated land, the production and ecological functions of the cultivated land can be fully exerted, which indicates that the transformation of the cultivated land has reached an ideal state. At this time, the resource input, scale structure, ecological regulation, and production capacity of the cultivated land will have all reached their optimal states, and the CLS will have the strongest ability to resist external interferences. As the existing relevant studies [37] suggest, the internal resource input, resource endowment, property right system, and external social, economic, and political factors are the main drivers leading to the transformation of a CLS, and the output, structure, and function of the CLS—such as its utilization form and scale, production levels, and ecological function—indicate whether or not a CLS transformation has been successful. CLSR mainly reflects the adaptability of a CLS’s transformation. Therefore, it is appropriate to measure CLSR by resource input, ecological function, production capacity, and scale structure based on the driving force and result of the CLS transformation. In general, CLSR is mainly reflected in the correlations and interactions between resource resilience, ecological resilience, production resilience, and scale structure resilience of a cultivated land, covering multi-dimensional systems such as cultivated land production, input, ecological security, morphological structure, and so on.

Among these, resource resilience is primarily the input of production capital, such as labor, agricultural machinery, irrigation, and transportation—which can resist the strong attraction of urbanization for various factors of production. Ecological resilience is mainly reflected in the quality of a cultivated land, its biological environment, and the strength of its ecological regulation and function, which can resist unstable changes in ecological factors such as climate. Production resilience is reflected in whether the output of a cultivated land can meet the needs of society; this measures the production and social security functions of the cultivated land, which are mainly to resist the increasing demand for food from the outside world. Scale structural resilience represents the quantity, production form, and spatial structure of a cultivated land, mainly reflecting the resistance of the cultivated land to fragmentation and abandonment.

In recent years, due to tension in the global trade environment, the continuous increase of the population, and ongoing environmental deterioration, the global food crisis has intensified, the cultivated land area has changed dramatically, and CLSR has continued to decrease. Therefore, policies and systems such as cultivated land protection, high-standard cultivated land construction, cultivated land circulation, and agricultural supply-side structural reform are being formed at the right moment, and CLSR can be improved from the aspects of improving the quantity and quality of the labor force, doubling investment into machinery, science, and technology, capital investment, and optimizing the scale structure of cultivated lands; with the strengthening of ecological civilization construction, the ecological function of cultivated land has been paid more attention, which is conducive to the improvement of CLSR. To sum up, external socio-economic factors such as globalization, urbanization, and industrialization will affect the composition structure of various elements within CLSs, unbalance CLSs, stimulate CLSR, better promote the adaptive cycle process of CLSs, stimulate the transformation of cultivated land utilization, and realize the sustainable development of CLS under the coordination of external forces such as financial support and policy inclination (Figure 1).

3. Research Methods and Data Sources

3.1. Research Methods

3.1.1. Comprehensive Measures of CLSR

Based on the above analysis, CLSR can be comprehensively measured across four dimensions: resource resilience, production resilience, ecological resilience, and scale structure resilience of a cultivated land. The basic idea is to construct an evaluation index system from four dimensions, and Stata15.0 software was used to normalize these indicators. The entropy method was used to weight, and the resilience evaluation equation was constructed using the comprehensive weighted summation model to evaluate CLSR.

Resilience evaluation equation. System resilience is the weighted sum of various resiliences in the system, and its formula is:

$$\mathrm{Positive} \mathrm{direction} U_{\mathit{ij}}=(x_i-\min x_i)/(\max x_i-\min x_i)$$

(1)

$$\mathrm{Negative} \mathrm{direction} U_{\mathit{ij}}=(\max x_i-x_i)/(\max x_i-\min x_i)$$

(2)

$$e_j=-k{\displaystyle \sum _{i=1}^m{U}_{ij}\ln U_{ij}}$$

(3)

$$W_j=\frac{(1-e_j)}{{\displaystyle \sum _{j=1}^n(1-e_j)}}$$

(4)

$$R_j={\displaystyle \sum _{i=1}^m{W}_j\times U_{ij}};\hspace{1em}R={\displaystyle \sum _{j=1}^n{R}_j};\hspace{1em}{R}^{\prime }=\frac{R_j}{R}$$

(5)

In the formula, U_ij is the index normalization value, m is the number of research units; e_j denotes the fuzzy entropy value of the j-th feature; W_j is the entropy; R_j is the resource resilience, production resilience, ecological resilience, and scale structure resilience of CLS; and n is the number of components of CLSR—n is 4 here. R is the total resilience of the system, which is expressed by the sum of normalized index variable values—its value range is [0, 4]. However, because the results in the calculation were too small, in order to better show regional changes, the evaluation results were simultaneously enlarged 100-fold. R′ represents the contribution rate of resource resilience, ecological resilience, production resilience, and scale structure resilience to the total resilience.

3.1.2. Spatial and Temporal Pattern of CLSR and Its Driving Mechanism Measures

(1)

Spatial Autocorrelation Analysis

A spatial regression model was built on the basis of testing whether there is a spatial correlation between elements, and Moran’s I was used to test the spatial autocorrelation between variables:

$$I=\frac{n{\displaystyle \sum _{i=1}^n{\displaystyle \sum _{j=1}^n{W}_{ij}(R_a-\overline{R})(R_b-\overline{R})}}}{({\displaystyle \sum _{i=1}^n{\displaystyle \sum _{j=1}^n{W}_{ij}}}){\displaystyle \sum _{i=1}^n{(R_b-R)}^2}}$$

(6)

In the formula, I is the global Moran index; $\overline{R}$ is the average value of resilience; R_a and R_b are the resilience values of provinces a and b, respectively, (a is not equal to b); W_ij is the weight distance; and n is the number of spatial regions studied. The value range of I is [–1, 1]. When the value of I is closer to 1, the spatial agglomeration of the elements is stronger; when the value of I is closer to −1, the spatial dispersion of the elements is stronger. When I = 0, it indicates that the elements are evenly distributed in the region. We used Stata15.0 software for calculations.

(2)

Spatial Durbin Model

$$R_{jit}=\delta {\displaystyle \sum _{j=1}^n{W}_{ij}{R}_{jit}+\phi +\beta x_{it}}+{\displaystyle \sum _{j=1}^n{W}_{ij}{x}_{ijt}\theta +c_i+}{\alpha }_t+{\epsilon }_{it}$$

(7)

In this formula, δ is the spatial regression coefficient, representing the spatial spillover effect and the mutual influence degree between the resilience values of adjacent spatial units; β is the regression coefficient of the explanatory variables, which indicates the influence degree of the explanatory variables on the explained variables; θ is the spatial lag coefficient of the explanatory variables, representing the influence degree of the explanatory variables in adjacent regions on the explained variables in the region. We used Stata15.0 software for calculations.

3.1.3. Evaluation Index System for CLSR

Previous studies have mainly focused on cultivated land use transition and function evaluations and have established an evaluation index system from the characteristics of the cultivated land use system, such as its economic–social–spatial, quantity-yield, and morphological functions, and so on [30,32]. This study comprehensively considered the characteristics of the cultivated land use system, and constructed an evaluation index system covering the whole process of the input and output of cultivated land use. In addition, the cultivated land ecological farmland circulation was organically integrated. While paying attention to the utilization of cultivated land, the impact of human actions on it is also emphasized. The evaluation system contains more dimensions and covers more comprehensive contents than previous systems.

According to the connotation framework of CLSR, an index system was established based on the four dimensions of: cultivated land resource resilience, ecological resilience, production resilience, and scale and structure resilience, for comprehensive representation. The resource resilience of cultivated land is mainly derived from the input of manpower, machinery, equipment, capital, and other production factors in a CLS. The higher the input level is, the stronger the resource resilience of the cultivated land is. Therefore, the resource resilience of the cultivated land was measured by the labor input per unit of cultivated land area, the total mechanical power per unit area, the effective irrigation rate, and the average land expenditure on agriculture. The ecological resilience of a cultivated land is closely related to the pressure of the ecological regulation undertaken by the cultivated land, which reflects the ability of the cultivated land to resist disasters and carry out ecological recovery. Excessive input of chemical fertilizers, pesticides, and other substances will cause damage to cultivated land. Therefore, excessive levels of chemical fertilizer application, the disaster rate of a cultivated land, and the proportion of cultivated land in an ecological area were selected to reflect the ecological resilience of the cultivated land. Increases in the levels of excessive chemical fertilizer application and the disaster area of a cultivated land lead to low ecological resilience, which is particularly prominent when the proportion of the cultivated land in an ecological area is high. The production resilience of a cultivated land stresses whether the output of a CLS can meet people’s living needs and bring certain economic benefits. The higher the grain yield and the agricultural output value are, the higher the production resilience of the cultivated land is; so, the grain yield per unit area, the agricultural output value per unit of cultivated area, and the per capita grain guarantee rate were selected to calculate the production resilience of the cultivated land. The scale structure resilience of a cultivated land is an attribute of the quantity of cultivated land, its planting structure, and the scale management degree of the cultivated land. If cultivated land resources become richer, more food crops are planted, and transfers of the cultivated land become more frequent; then, the quality and large-scale operation degree of arable land become higher, its functions are better performed, and the scale structure resilience becomes stronger. Therefore, the proportion of cultivated land, the proportion of the grain-sown area, and the cultivated land circulation rate were selected to represent the scale structure resilience of the cultivated land (

Table 1. Cultivated Land Use System Resilience (CLSR) Evaluation Index System.

Criterion Layer	Index Layer	Definition of Phase	Direction	Units	Weight
Resource Resilience	Labor input per unit cultivated land area	Rural agricultural employment personnel/Cultivated area	+	Person/hm2	0.34
	Total mechanical power per unit area	Total agricultural machinery power/Total cultivated area	+	(kw·h)/hm2	0.25
	Effective irrigation rate	Effective irrigation area/Total cultivated area	+	%	0.22
	Average land expenditure on agriculture	Agricultural expenditure/Total cultivated area	+	ten thousand yuan/hm2	0.19
Ecological Resilience	Excess amount of chemical fertilizer application	Yield of chemical fertilizer application/Total cultivated area—225kg/hm2	-	kg	0.30
	Disaster rate of cultivated land	The disaster area occupies the total area of cultivated land	-	%	0.43
	Proportion of cultivated land area to ecological land	Cultivated area/Area of ecological land	-	%	0.27
Production Resilience	Grain yield per unit area	Grain total output/Cultivated area	+	t/hm2	0.27
	Agricultural output value per unit cultivated area	Total value of agricultural output/Cultivated area	+	ten thousand yuan/hm2	0.32
	Per capita grain guarantee rate	Grain total output/Permanent resident population × 400 kg	+	%	0.41
Scale Structural Resilience	Proportion of cultivated land	Cultivated area/Total area of land	+	%	0.26
	Cultivated land circulation rate	Cultivated land circulation area/The total area of farmland contracted by households	+	%	0.59
	Proportion of grain sown area	The sown area of grain/Total cultivated area	+	%	0.15

Note: The upper limit of the international chemical fertilizer safety standard is 225 kg/hm². Ecological land refers to the sum of cultivated land, woodland, grassland, water area, and unused land.

).

3.1.4. CLSR Zoning Rules

CLSR is closely related to the sustainable use of cultivated land. It is very important for the sustainable use of cultivated land to strengthen the comprehensive management of CLS from the perspective of ‘society–ecology’, based on the analysis of the resilience of the cultivated land. Therefore, considering the evolution patterns and trends of cultivated land resilience, the following cultivated land resilience zoning rules could be preliminarily defined (

Table 2. Zoning Rules of Cultivated Land Use System Resilience (CLSR).

Zoning Type	Characteristics
Major Evolution Area	High and medium resilience value; rising speed is high.
Stable Development Area	High and medium resilience value; rising speed is stable or low.
Potential Excitation Area	Medium and low resilience value, but the growth rate is relatively high.
Sensitive Lag Area	Medium and low resilience value, but the growth rate is relatively low.
Degraded Vulnerable Area	The resilience value is medium and low, and it shows a downtrend.

). Based on the natural breakpoint method, the 2018 CLSR in the study area was divided into three grades: high, middle, and low. The growth rate of the CLSR from 2015 to 2018 was calculated to represent its improvement rate, and it was divided into a deceleration zone and a low-speed zone by the natural breakpoint method. Then, the two types of regions were superimposed to measure the future development trend of the regional CLSR. When CLSR is high, its lifting speed is high, and this region is the major evolution area, whereas if the lifting speed is slow, it is a stable development area; when CLSR is at a medium or low level and its promotion speed is high, this region is the potential excitation area, whereas if the lifting speed is slow, it is the sensitive lag area; if the lifting speed is negative, it is the degraded vulnerable area.

3.2. Data Sources

The data on cultivated land, ecology, and total land area in this research work were obtained mainly from the Resource and Environment Science and Data Center of the Chinese Academy of Sciences [56]. By consulting the China Statistical Yearbook, China Statistical Yearbook on Environment, China Rural Statistical Yearbook, Statistics on the National Rural Economic Situation, Annual Statistical Report of China Rural Operation and Management, and provincial statistical yearbooks from 2005 to 2018, we obtained data on cultivated land production, input, transfer, natural ecological environment, and social and economic development in each province [57]. Constrained by missing data, some data is replaced by the data from adjacent years or obtained by proportional calculation. However, due to the deficiency of data on China’s Tibet, China’s Hong Kong, Macao, and Taiwan, the scope of this study only includes 30 provinces in the Chinese mainland.

3.3. Research Area

As the world’s largest developing country, China’s economic growth rate from 2013 to 2018 reached 7.0%, which is much higher than the world average of 2.9% [57]. This rapid economic development has greatly increased China’s per capita national income and has significantly improved the level of people’s livelihoods. China’s agricultural modernization has also been a rapid development; agricultural production has increased substantially, grain and meat output ranks among the top in the world, and farmland mechanization and large-scale utilization have also been greatly improved. As time goes by, during the 13 years from 2005 to 2018, the average growth rate of total power of agricultural machinery in all provinces of China was 812,000 kWh/year, and the average growth rate of production in the planting industry was 10.6 billion yuan/year [58]. In addition, China is characterized by the diversity in its topography, climate, resource types, and land resource utilization. Taking the whole of China as an example is more universal and applicable [59,60].

Cultivated land is a basic requirement to guarantee human survival and development, which is related to one country’s social stability and sustainable development. China’s per capita cultivated land area is small, and reserves of cultivated land are insufficient, with a shortage of high-quality cultivated land. Most of the cultivated land resources are concentrated in plains, basins, and hilly areas such as Northeast China, North China, the middle and lower reaches of the Yangtze River and the Pearl River Delta, while cultivated land areas in the West are small and scattered. As society and the economy advance, the situation with the cultivated land resources in China is becoming very severe. Excessive use of chemical fertilizers and pesticides, endless cultivation of cultivated land, and a shortage of water resources all cause cultivated land desertification and serious pollution, and the quality of cultivated land continues to decline; continuous population increases make the load of the cultivated land in China increase continuously. However, due to the continuous loss of the rural labor force, the imperfect agricultural infrastructure, the continuous low market price of agricultural products, the continuous expansion, and the occupation of construction land, the phenomenon of cultivated land abandonment, waste, and occupation occurs frequently. The contradiction between man and land is increasingly sharp and the protection of cultivated land is facing great pressure [61,62]. Therefore, promoting the sustainable and intensive use of cultivated land, giving full play to the functions of cultivated land, and improving the resilience of cultivated land facing interferences are important ways to alleviate the contradiction between man and land (Figure 2).

4. Result Analysis

4.1. Temporal and Spatial Patterns of CLSR

CLSR was graded using the natural breakpoint method. Evaluation results showed that CLSR in China is on the rise as a whole, and the polarization of the distribution pattern is becoming intensified. In terms of quantity, the average values of cultivated land resilience in 2005, 2010, 2015, and 2018 were 2.69, 3.04, 3.67, and 3.93, respectively, and the period from 2010 to 2015 saw the fastest growth. From a spatial perspective, CLSR showed strong regional agglomeration and developed in a continuous sheet shape, while CLSR differences among local areas were obvious. The high and low values were concentrated in eastern coastal areas and central and western inland areas, respectively. As time went on, the low value agglomeration areas showed a trend of diffusion from the central areas to the western areas. The northern black soil area became the dominant area of CLSR, while the strength of CLSR in the southern coastal areas gradually weakened (Figure 3).

Regarding the contribution of the cultivated land resource resilience, ecological resilience, production resilience, and scale structure resilience to CLSR, the CLSR of 30 provinces experienced the transformation process of ‘leading by resource and ecological resilience—equilibrium of each resilience—leading by production and scale structural resilience’ at the four research time points. To be specific, in 2005, the resource resilience of cultivated land was dominant in eastern coastal areas, but its distribution area was gradually shrinking. By 2018, due to the urban–rural dualization, only Guangxi, Shandong, and Shanghai, with higher population density and economic levels, had a higher contribution rate of resource resilience. During the study period, the contribution of ecological resilience in eastern coastal areas was low, and the phenomenon of uncontrolled occupation of cultivated land and excessive application of chemical fertilizers and pesticides caused by economic growth was intensified. Regions with a low contribution rate of ecological resilience gradually expanded westward, and finally, regions with a high contribution of ecological resilience were concentrated in Xinjiang and Qinghai. Regions with a high contribution of production resilience were mainly concentrated in the northeast black land region, and by virtue of their natural advantages, they expanded to the central and western regions. However, the contribution rate of scale structure resilience has been at a low level at the time of the study. As the government vigorously promotes the transfer of agricultural land, the large-scale operation of cultivated land keeps expanding, and its contribution rate keeps increasing—but it is mainly concentrated in eastern coastal areas. Thus, the cultivated land in the central, western, and northeast regions of China was mainly showing ecological resilience, while the cultivated land in eastern coastal regions was mainly showing resource resilience in 2005. In 2010 and 2015, the western and northern parts of China were dominated by ecology and production resilience, while the southeastern parts were dominated by resources and scale structure resilience. In 2018, areas with high values of cultivated land resilience in the north and east of China were mainly showing production structure resilience and scale structure resilience; China’s western regions were in a relatively diverse ecological environment, and the cultivated land there bore a low ecological adjustment pressure, so its ecological resilience was high—but its resistance to economic, social, and environmental disturbances was low, and so its overall resilience was low (Figure 4).

4.2. CLSR Zoning Characteristics

The growth rate of CLSR in the study area from 2015 to 2018 was calculated to represent development trends, and the natural breakpoint method was applied to classify and superimpose them with the status quo of CLSR in 2018. On this basis, CLSR zoning rules in Table 2 were combined to delimit CLSR zoning in some areas of the Chinese mainland (Figure 5).

In combination with the results in Figure 4 and the zoning results, it appears that CLSR in China’s northern, eastern, and southern coastal regions, and most of the area in the middle reach of the Yangtze River, mainly belongs to major evolution areas and stable development areas. The CLSR value in these areas is relatively high and increases at a relatively high or stable speed. Most of them are dominated by production resilience and scale structure resilience, while resource resilience and ecological resilience are in a relatively weak state. It shows that these regions pursue agricultural economic output under the conditions that the population, capital, and machinery tend to flow into secondary and tertiary industries, and—at the same time—that the trend of intensive management of cultivated land has been strengthened, and the ecological function of cultivated land has been restrained. The distribution of potential excitation areas is more dispersed, being mainly concentrated in northern China’s Liaoning, Tianjin; mid-western China’s Qinghai, Shaanxi; and southern China’s Hainan and other places. Although the resilience value is low, the growth rate is high, and the type of distribution areas are mainly ecological resilience; the resource and production resilience are relatively low, the contribution of scale structure resilience is in an up-trend, and CLSR shows great potential. Sensitive lag areas and degraded vulnerable areas are mainly distributed in China’s northwest and southwest regions, and places such as Shanxi and Fujian, with low resilience values and a low growth rate—even with a negative growth state. Each kind of resilience value is weak; most of the area is dominated by ecological resilience, and its CLS is extremely vulnerable to external interference (Beijing’s CLSR is low and stagnant due to its political and social functions).

4.3. Influence Mechanism of CLSR Evolution

In order to explore the influence mechanism of CLSR, this study started from the natural ecological environment and social and economic development and took measurements of the endowment of regional natural resources such as water and soil, the quality of the ecological environment, and the protection intensity as its principal measurements. Five explanatory variables including cultivated land area (CLA), total water resources (TWR), proportion of natural reserve area (PNRA), proportion of desertification of cultivated land area (PDCLA), and number of environmental emergencies (NEE) were selected to represent the regional natural ecological environment status. In order to measure the current situation of regional economic, social, industrial, and financial development, five explanatory variables including per capita disposable income of farmers (PCDIF), the urbanization rate (UR), fiscal revenue (FR), the proportion of secondary and tertiary industries (PSTI), and the rural permanent resident population (RPRP) were selected to represent regional socio-economic development, and a regression model was established with these 10 explanatory variables (Appendix A). The driving effects of natural, ecological, financial, and population factors on the resilience of cultivated land were analyzed. Spatial correlations between provinces were verified by calculating the CLSR and Moran’s I of resilience of cultivated land resources, ecology, production, and scale structure at each research point. It was found that there was a positive spatial correlation between CLSR and resilience among provinces, and all passed the significance test of 99% (

Table 3. Spatial Correlation of Cultivated Land Use System Resilience (CLSR).

	2005	2010	2015	2018
CLSR	0.365 ***	0.316 ***	0.42 ***	0.348 ***
Resource Resilience	0.495 ***	0.507 ***	0.488 ***	0.304 ***
Ecological Resilience	0.569 ***	0.556 ***	0.527 ***	0.519 ***
Production Resilience	0.245 ***	0.245 ***	0.247 ***	0.272 ***
Scale Structure Resilience	0.432 ***	0.334 ***	0.384 ***	0.413 ***

Note: *, ** and *** indicate that the results are significant at the level of 10%, 5%, and 1%, respectively.

). This shows that the resilience of cultivated land in provinces is not randomly distributed, and is characterized by significant spatial dependence and aggregation. Therefore, it is necessary to introduce a spatial regression model to analyze the driving mechanism.

An appropriate spatial econometric model was selected by spatial correlation diagnostic methods such as LM, LR, and the Hausman test, and the results are shown in

Table 4. LM, R-LM, and Hausman test results.

	LM Test				LR Test		Hausman Test
	LM	R-LM	LMLAG	R-LMLAG	SDE Simplified to SAR	SDE Simplified to SEM	Random Effect	Individual Fixed Effect Model	Time Fixed Effect
CLSR	21.972 *** (0.000)	9.3 *** (0.002)	19.027 *** (0.000)	6.355 ** (0.012)	19.88 ** (0.0187)	25.66 *** (0.0023)	36.52 ** (0.019)	23.76 *** (0.000)	2.78 ** (0.045)
Resource Resilience	14.17 *** (0.000)	0.776 (0.378)	27.908 *** (0.000)	14.515 *** (0.000)	18.36 ** (0.0493)	25.33 *** (0.0048)	36.96 ** (0.017)	4.60 *** (0.000)	9.44 *** (0.000)
Ecological Resilience	51.483 *** (0.000)	21.589 *** (0.000)	32.853 *** (0.000)	2.959 * (0.085)	39.77 *** (0.000)	55.18 *** (0.000)	72.56 *** (0.000)	7.1 *** (0.000)	5.15 *** (0.002)
Production Resilience	25.134 *** (0.000)	4.215 *** (0.000)	27.809 *** (0.000)	6.89 *** (0.009)	37.3 *** (0.0001)	42.33 *** (0.0001)	78.29 *** (0.000)	16.45 *** (0.000)	0.76 *** (0.021)
Scale Structure Resilience	18.1 *** (0.000)	5.402 ** (0.020)	18.284 *** (0.000)	5.587 ** (0.018)	31.13 *** (0.000)	38.82 *** (0.000)	17.93 * (0.065)	38.50 *** (0.000)	1.20 * (0.089)

Note: *, ** and *** indicate that the results are significant at the level of 10%, 5%, and 1%, respectively.

. According to the LM test results, the spatial error effect of the resource resilience of cultivated land was not significant, and the spatial error and lag effect of the explained variables all passed the significance test. The LR test results all rejected the original hypothesis of SDE model simplification, indicating that SAR and SEM models cannot replace SDE models. According to the results of the Hausman test, all resilience models rejected the original hypothesis and chose the fixed effect. Finally, the joint significance test was used to test the SDE models, and the results showed that the original hypothesis was rejected in all the models, and the time–space double-fixed model was selected for the spatial analysis.

According to the results of spatial Durbin model test (

Table 5. Regression Results of the Spatial Durbin Model.

	CLSR		Resource Resilience		Ecological Resilience		Production Resilience		Scale Structure Resilience
	β	θ	β	θ	β	θ	β	θ	β	θ
Total water resources	0.0045 *	−0.0012 **	0.00046 ***	−0.0012 ***	−0.00002	0.00103 **	0.00002 ***	0.00005	0.00001	0.00008
The proportion of natural reserve area	0.001	0.17	0.0023	0.034	0.00012	0.0124 ***	0.0021	0.0293 *	−0.003 *	−0.0443 ***
Number of environmental emergencies	−0.001	0.0001	−0.0013 **	−0.0014	0.00002	−0.00024 *	0.00009	0.000304	0.0006 **	0.0007
The proportion of desertification cultivated land area	−0.05	0.02	0.0001	0.00001	0.00001 ***	−0.00001	−0.00002 **	−0.00006	0.00001	0.00002
Cultivated Land area	0.0001	−0.0003	0.0034	−0.002	0.00005	−0.00009	0.00012	0.002	0.00003	0.0014
Per capita disposable income of farmers	0.0003	−0.0001	0.0001	−0.0001 **	0.00008 **	0.00007 *	0.00003 *	−0.00001	0.00002	0.00008 ***
Urbanization rate	−0.015	−0.0001	−0.016	0.043 **	0.0012	0.0018	0.024 ***	−0.0315 **	−0.004	−0.011
Fiscal revenue	−0.002	0.008	0.0001	0.00017 **	−0.00003	0.00002 *	−0.00002	0.00004 *	−0.00005 ***	−0.00007 **
The proportion of secondary and tertiary industries	−0.05	−0.021	−0.03	−0.08 **	0.00042	−0.0043	−0.022 *	0.0414	0.0174 **	0.046 ***
Rural permanent resident population	0.002 ***	−0.0031	0.007 *	−0.00007	−0.00001 **	0.00002 *	0.00007 ***	−0.00008 **	0.00007 ***	0.00005
δ	0.21 **		0.15 ***		0.304 *		0.595 ***		0.22 *
R2	0.757		0.776		0.45		0.79		0.88
sigma2_e	0.04 ***		0.40 ***		0.0451 ***		0.005 ***		0.008 ***

Note: *, ** and *** indicate that the results are significant at the level of 10%, 5%, and 1%, respectively.

), the δ values of the spatial spillover effects of CLSR and the resilience of resources, ecology, production, and scale structure were all positive, and they passed the significance test to different degrees—indicating that the improvement of CLSR and resilience in adjacent provinces will promote the development of CLSR in a province. Geographical factors had a positive impact on the spatial distribution of CLSR, so there is a significant spatial dependence of CLSR, which was the same with the spatial cluster development trend of China’s economy.

From the natural ecological environment, the total water resources had significant positive influences on total resilience, resource resilience, and production resilience of cultivated land, and had a negative and positive spatial lag effect on the total resilience, resource resilience, and ecological resilience of cultivated land in the neighborhood, respectively, showing that the spatial distribution of water resources difference is the important factor in the differences in cultivated land resilience layout. The proportion of the natural reserve area had a significant negative impact on the scale structure resilience of the cultivated land in the province and its neighbors, and the main reason for this was that the establishment of nature reserves can hinder the large-scale continuous management of cultivated land; however, due to the extensive ecological benefits, the ecological and production resilience of the neighboring cultivated land will be promoted. The number of environmental emergencies had a negative and positive impact on the resilience of the cultivated land resources and the scale structure, respectively, in the province, and had a negative spatial lag effect on the ecological resilience of the cultivated land in the neighborhood, which shows that environmental events will affect farming households’ input and management willingness to cultivate land, cause ecological damage, and affect the process of resisting disturbances and the orderly transformation of cultivated land. The higher the proportion of desertification of cultivated land is, the lower the area of cultivated land is, the weaker the resilience of cultivated land production is, and the lower the pressure of ecological regulation of cultivated land is. However, there is a negative impact on the whole ecosystem, so although the regression coefficient is positive, there is a negative impact on the ecological resilience of cultivated land. The area of cultivated land had no significant effect on the resilience of the cultivated land, which indicates that the quantity of cultivated land is not the main factor affecting the resilience of cultivated land.

From social and economic development, the per capita disposable income of farmers had a significant positive impact on the cultivated land ecology, production resilience, and neighborhood ecological resilience, but a negative impact on the neighborhood cultivated land resource resilience—indicating that as economy develops, farming households’ environmental protection consciousness is gradually strengthened and the agglomeration effect of economic factors is also significant. The urbanization rate had a positive impact on the production resilience of the province and the resource resilience of the neighborhood, mainly because urbanization brings certain funds and technologies to agricultural development, which increases the agricultural output—but at the same time, it has a negative impact on the production resilience of the cultivated land in the neighborhood. Fiscal revenue had a significant positive influence on the cultivated land resources, ecology, and production resilience of the neighborhood, and had a significant negative influence on the scale structure resilience of the cultivated land of this province and the neighborhood. The main reason for this is that land finance is the main source of government finance, and it is easier to conduct land transactions between adjacent regions, which destroys the trend of large-scale operation of cultivated land to some extent. The proportion of secondary and tertiary industries had a significant negative impact on the production resilience of cultivated land in the province and the resilience of cultivated land resources in the neighborhood, but had a significant positive impact on the scale structure resilience of the cultivated land in the province, which reflects the attraction of industrial development to the population in terms of region and employment, and the promotion of cultivated land transfer. The rural permanent resident population is one of the main factors that influences cultivated land resilience; it has a significant positive driving effect on the total resilience, resources, production, and scale structure resilience of cultivated land in the province, and the ecological resilience of cultivated land in the neighborhood, while having a negative impact on the ecological resilience of cultivated land in the province and the production resilience of cultivated land in the neighborhood—showing that in areas where the rural population is concentrated, the cultivated land is also highly concentrated, which attracts the transfer of cultivated land from neighboring areas to the province and stimulates the development of agriculture in the province.

In general, areas with rich natural resources, a large permanent rural resident population, and good farmers’ livelihood have higher CLSR. The natural ecological environment and socio-economic development in these regions provide the essential basis for the maintenance of CLS, meeting the needs of agricultural cultivation, promoting farmers’ willingness to transfer cultivated land, and enhancing farmers’ awareness of ecological protection; the natural ecological environment and social-economic development status of the neighborhood mainly depend on the factor attraction and the regional radiation and driving effect of the economy and ecology, affecting the CLS of the region. Through the above ‘economic foundation—superstructure’ and ‘economic development—factor agglomeration’ and other forms of driving, the resource ecology, production, and scale structural resilience of the CLS have been improved, and thus, CLSR has been enhanced (Figure 6).

4.4. CLSR Governance Idea Based on Zoning Characteristics and Impact Mechanism

Based on the characteristics of each zoning and the analysis of the CLSR impact mechanism, the following governance ideas can be summed up:

Major evolution area and stable development area: It should maintain the advancement and stable development of CLSR, maintain its advantages and make up for any shortcomings. To address the dilemma of industrial bias in economically developed areas, based on the practice area, when using the cultivated land rationally, increasing agricultural output efficiency, improving the system of cultivated land transfer, promoting more intensive utilization of cultivated land scale, and maintaining regional CLSR advantages, attention should be paid to maintain the ecological function of cultivated land. It should enhance the ecological protection of cultivated land, open up the front end of agricultural production and attract the input of various elements from the aspects of technological innovation and utilization system perfection. It should give full play to the driving effects of ‘economic foundation—superstructure’ and ‘economic development—factor agglomeration’ through technical training and market effects, so as to make up for any shortcomings, and the resources and ecological resilience of this zoning can be improved to achieve an all-round improvement of CLSR.

Potential excitation area: To promote CLSR in a multi-means and focused way, and to broaden the channel of factor flow in multiple ways. In view of the characteristics of low resource and production resilience in potential excitation areas, under the principle of not reducing ecological resilience, flexible use of personnel training, organization construction, financial credit, planning guidance, scientific and technological support, facilities improvement, market construction, and other means of enhancing resource and production resilience, boost the growth of scale structure resilience, and take the multi-scale joint mechanism of the country/province/city as the starting point, it expands the flow of people, money, technology, and other elements to the CLS through ways such as stimulating the market by demand and seeking guiding policies, and strives to form a free-flow system of elements within and outside the province between provinces at the national level, and between cities and prefectures at the provincial level to achieve the effect of factor guarantee and resource drive to stimulate the potential of CLSR.

Sensitive lag area and degraded vulnerable area: Multiple measures should be taken to support the improvement of CLSR and build a regional driving mechanism for CLSR. CLSs in this area have a weak ability to resist external interference and are very vulnerable to damage, and in combination with the actual regional development, their development momentum is weak, and urgently needs external support. Therefore, from the perspective of the economy, society, and politics, we should strengthen the preferential policy and investment of capital and technology, improve the infrastructure and organization construction of CLS, build the support system of CLSR in backward areas, and give full play to the spatial linkage effects of CLSR and its positive influencing factors. The development of the neighborhood can drive the development of cultivated land resources, ecology, production, and scale structure resilience in backward areas, form a regional driving mechanism of CLSR, and play a radiation effect, so as to cultivate the endogenous power of CLSs in backward areas and facilitate the recovery and generation of CLSR in sensitive lag areas and degraded vulnerable areas.

5. Discussion

In the new period of rapid reconstruction of urban and rural patterns, CLSs are up against severe internal and external disturbances, and the instability, vulnerability, and unsustainability of CLSs are increasingly prominent. Therefore, it is of great importance to study the historical and current difficulties of the diversified development of CLSs with the help of cultivated land resilience to ensure the effective fulfilment of the various functions of CLSs, meeting the needs of social development, and to explore the adaptive transformation and sustainable development of cultivated land. It is the starting point of CLSR research to explore the governance ideas of CLSR within the framework of ‘society–ecology’ by using resilience theory to deepen the related research of cultivated land utilization, to establish an index system from the perspectives of resource input, production efficiency, sustainable development capacity, etc., to analyze temporal and spatial changes in CLSR, and to divide CLSR according to the evolution and speed of the CLSR, and to clarify the impact mechanism of CLSR. This study preliminarily defined the concept of CLSR and emphasized the adaptability of cultivated land systems to create a new path of transformation on the basis of previous studies on cultivated land use transformation; it also evaluated the future development capacity of cultivated land systems. Compared to other similar studies [50,63], the innovation here lies in the establishment of an empirical research framework of theoretical analysis–index evaluation factor detection and governance countermeasures from the aspects of constructing all aspects of CLSR indicators and provincial-scale comparison and regional classification and grading, which complements CLSR research system and is part of socio-ecosystem resilience research. Similar to previous studies, we all believe that resilience emphasizes adaptive learning and transformation ability, which is an important supplement to sustainability. In the face of disturbances, the system needs to go through an adaptive cycle process to achieve factor reconstruction. Internationally, studies on farm resilience focus on the construction of its conceptual theory and the evaluation of its governance strategies, but a perfect evaluation index system for farm resilience has not been established, and empirical studies are relatively weak [51,64]. In addition, compared with studies that focus on farmers and users to explore changes in resilience in relation to the behavior of subjects (Hossard, Fadlaoui, Ricote, Belhouchette, 2021), our research mainly starts from the properties of the CLS itself, from the multi-dimensions of its input–output, ecology, and morphology—fully highlighting the main position of land elements in the system—and we evaluated CLSR from the spatio-temporal dimension, delineating resilience zones in space and indicating how each region should improve CLSR in the future, and explored the spatial spillover effects of CLSR-influencing factors, which were all lacking in previous studies.

All this was extensive and applicable to strengthening and innovating the current research into CLSR. From an international perspective, cultivated land is a scarce resource recognized by all countries. As the global demand for food continues to rise and urban expansion accelerates, the world is facing a dilemma of cultivated land abandonment, fragmentation, and non-agricultural utilization, as well as the gradual decline of cultivated land’s quality, fertility, and ecological functions. In response to this dilemma, some countries have innovated the utilization and management of cultivated land by issuing common agricultural policies, delegating cultivated land rights, innovating cultivated land trading systems, regulating land markets, strengthening agricultural land investment, and other means. However, the problem of urbanization encroaching on highly suitable agricultural areas and the conflict between cultivated and construction land has not been effectively addressed [1,44].The continuous expansion of construction land scale leads to continuous occupation of cultivated land, and the contradiction between economic development and food production is becoming increasingly serious. At the same time, more and more high-quality cultivated land is occupied by construction land, and it is difficult to rely on reclamation of the same quality of cultivated land to compensate, resulting in a continuous decline in the quality of cultivated land. In addition, with the spread of the urban fringe, the cultivated land ecosystem is being inevitably destroyed by the pollution of urban waste water, waste gas, and waste. The contradiction between cultivated land and construction land also has a negative impact on the territorial space management system, the phenomenon of management failure and inapplicability frequently occurs, and the conflict between multiple management systems has become an obstacle to the coordinated development of space. In order to enhance the adaptability of CLS, we should evaluate and identify the size and type of regional CLSR from the perspectives of input, output, ecology, and structure of the CLS itself, clarify the shortcomings of regional CLS development, and discuss the key points and directions of CLS governance in different types of regions in combination with the impact mechanism of CLSR, which will undoubtedly serve as a scientific reference for countries in the world to aim at the direction of CLS governance. However, this study has some limitations in the innovation of the CLSR evaluation model. How to innovate research methods to better represent the characteristics of system adaptive transformation is still a problem worth considering. In addition, the data used in this study are all secondary data released by Chinese authorities, and there is a lack of primary data and empirical cases to support our research results, which is also what we need to further study.

The study of CLSR does not stop there. It will need to integrate and highlight the characteristics of resilience corresponding to adaptive transformation, further innovate the methods and models of CLSR evaluation and zoning, explore the correlation mechanism between various levels of CLSR, establish a spatial CLSR network pattern, explore the means of enhancing cultivated land resilience in multiple ways and focuses, explore the resilience range matched with the region, predict the existing problems of regional cultivated land utilization based on cultivated land resilience, and put forward scientific and reasonable prevention and control suggestions in the next step of CLSR research. In the future, combining CLSR research with intensive utilization of the cultivated land scale, transfer of cultivated land use, and sustainable development of CLS, explaining the path and logical origin of its change in depth, establishing the theoretical connection and logical framework of CLSR and cultivated land use research, forming a research paradigm of cultivated land resilience, and opening up the perspective of cultivated land resilience are important and innovative links in the theoretical system of cultivated land research. It is of great significance to maintain international food security to look at future research on cultivated land resilience from the perspective of development, and to organically link cultivated land resilience with agricultural resilience and regional resilience.

6. Conclusions

In this study, we conducted a preliminary discussion on the connotations of cultivated land resilience, established an index system from multiple dimensions, analyzed the spatial–temporal evolution pattern of cultivated land resilience in 30 provinces in China, and explored the driving mechanism of cultivated land resilience, and reached the following conclusions:

(1)

Cultivated land system resilience can be defined as: by adjusting the structure and scale of internal elements, CLS is a kind of sustainable development ability, which absorbs and adapts internal and external disturbances and shocks to the maximum extent, abandons the original inapplicable state, creates a new recovery path, transforms to a new balance, and avoids system recession. Cultivated land resilience is an important force to promote the transformation of CLS. It conforms to the adaptive cycle process and includes four parts: resource resilience, ecological resilience, production resilience, and scale structure resilience.

(2)

The cultivated land system resilience of the 30 provinces showed an upward trend as a whole, the polarization degree of the distribution pattern gradually intensified, and it experienced a transformation process of ‘leading by resource and ecological resilience—equilibrium of each resilience—leading by production and scale structural resilience’.

(3)

CLSR in China’s northern, eastern, southern coastal regions, and most of the area in the middle reach of Yangtze River mainly belong to major evolution areas and stable development areas. The distribution of potential excitation areas is more dispersed, mainly concentrated in northern China’s Liaoning, Tianjin, mid-western China’s Qinghai, Shaanxi, southern China’s Hainan, and other places. CLSR-sensitive lag areas and degraded vulnerable areas are mainly distributed in China’s northwest and southwest regions and places like Shanxi and Fujian.

(4)

Through influencing the structure and scale of CLS internal factors, the natural ecological environment and social–economic development status affect the resilience of cultivated land, and the spatial effect is significant. The resource endowment, mainly dominated by water resources, has a strong impact on cultivated land resilience, while social economy mainly affects cultivated land resilience through several forms, such as ‘economic foundation—superstructure’ and ‘economic development—factor agglomeration’.

(5)

Based on the analysis of CLSR zoning and its influence mechanism, the following ideas of CLSR governance can be obtained: it should maintain the advance and stable development of CLSR; maintain the advantages and make up for the shortcomings in the major evolution areas and stable development areas; it should promote CLSR in a multi-means and focused way and broaden the channel of factor flow in multiple ways in potential excitation areas; and it should take multiple measures to support the improvement of CLSR and build a regional driving mechanism for CLSR in sensitive lag areas and degraded vulnerable areas.