Analysis of Spatial and Temporal Variation and Influencing Factors of Rural Land Dependence from the Perspective of Human-Land Relationship

Abstract

Face the issues of rural youth outflow, increase of aging population, and rural decline, this research explored the spatiotemporal evolution characteristics and influencing factors of the rural human-land relationship using the quantification index of rural land dependence. The research defines the concept of rural land dependence and constructs an evaluation index system for rural land dependence. By using the entropy weight method, the research quantitatively measures the high and low levels of rural land dependence in 30 provincial administrative units in China from 2006 to 2020. Besides, the GIS spatial analysis and geodetector are used to explore and analyze the spatial-temporal pattern and influencing factors of China’s land dependence. The results show that: (1) Temporal evolution: Since 2006, the level of rural land dependence in China has shown a decreasing trend, with economic dependence > population dependence > spatial dependence in terms of partial dependence differences. (2) Spatial pattern: During the study period, there are regional differences in the spatial distribution of rural land dependence in China, with a “high in the southwest and low in the eastern coastal areas” characteristic. The distribution of high-dependence areas has changed from being concentrated to being dispersed, while the distribution of low-dependence areas has changed from being dispersed to being concentrated. (3) Influencing factors analysis: From 2006 to 2018, the number of large agricultural employees is the main influencing factor of land dependence. With the decrease of population dependence and the increase of spatial dependence indexes, as well as the influence of the social environment, the area of rural areas and the total area of the region became the main influencing factor of land dependence from 2018 to 2020. The results can provide scientific basis for deepening the research on rural land use efficiency, and also provide decision support for promoting rural sustainable development and achieving common prosperity.

1. Introduction

Internationally, rural youth outflow and increase of aging population caused by urbanization are mutually causal with rural decline. For example, Maxwell Hartt explained in his book “Quietly shrinking cities” why Canadian cities in the era of growth and population loss, and emphasized the importance of considering both urban size and population loss rate [1,2]. Sabato Vinci et al. take the metropolitan area (Attica, Greece) as the research object, and provide the dynamic performance of regional population decline through the study of the interaction of population aging, low fertility rate, counter-urbanization and crisis-driven migration from 2010 to 2019 [3]. Niushan Jinger, a Japanese scholar, pointed out that from the 1950s to the 1970s, Japan entered a period of high economic growth. The rural surplus population was rapidly absorbed by industry and commerce, and the phenomenon of part-time employment, mixed residence and aging of the rural population gradually emerged, which led to the decline of autonomous villages [4]. Marcos Carchano et al. point out that population decline is a serious problem facing Spain, especially in rural areas. Due to the reduction of infrastructure and services, the deterioration of quality of life, the low inflow of new residents, the low level of local development and the aging of the population, there have been some vicious cycles [5]. In order to prevent this situation from appearing or becoming worse, the report of the 19th National Congress of the Communist Party of China clearly stated for the first time that “implementing the strategy of rural revitalization” has also become the overall strategy of China’s agricultural and rural modernization. In view of China’s rural revitalization strategy, Chinese scholar Ye Xingqing’s research suggests that rural revitalization should pay special attention to the three key factors of “human, land, and money” [6]. He believes that if the rural areas remain limited to the development of agriculture and the development of farming is limited to the development of crops and livestock, farmers cannot be fully employed and rural areas cannot be prosperous and developed. Analysis shows that there are problems in current human-land relationship in rural areas: the rapid development of urbanization and agricultural technology has liberated rural labor forces on the one hand, and on the other hand, it has also promoted the transfer of rural populations to urban areas and agricultural labor forces to non-agricultural industries. This group of people is significantly better than those who stay in agriculture and rural areas in terms of age, education level, and gender ratio. The quantity and quality of rural man resources are decreasing, which is the “human” issue of rural revitalization. With the gradual exit of “farmers in the 1940s and 1950s” and “farmers in the late 1980s” not being engaged in farming, in addition to the construction of urban-rural social security systems, the livelihood security function of rural land is also declining, and the risk of land lying fallow and idle in rural areas is increasing, which is the “land” issue of rural revitalization. Therefore, quantitatively studying the dependence relationship between population and land in rural areas from the perspective of human-land relationship can help make more scientific judgments about the status and changes of human-land relationship, and provide reference for rural revitalization and its sustainable development. Among them, the sustainable development of rural areas must be based on the active interdependence of human-land.

The issue of human-land relationship is the core of geography [7,8,9], which mainly refers to the dependence of human on nature and the active status of human [10]. As human’s ability to understand, utilize and transform nature, the connotation of human-land relationship has also changed to the breadth and depth with the development of productivity [9]. In the quantitative study of human-land relationship, the system dynamics analysis method is the most commonly used, mainly for the static evaluation of the relationship, coupling degree and coordination mode between population, resource environment and development in the regional system of human-land relationship [11,12,13,14]. It is an evaluation of the results of the interaction between human-land relationship, but it does not describe the degree of “human” dependence on “land”, or cannot reflect the level of “land” status on “human”. Therefore, Wang Yingli and others [15,16,17,18,19] introduce the concept of “interdependence” in economics, which reflects the degree of interdependence between one country’s economy and another country’s economy or the world economy [20,21,22]. They constructed a model of the dependence of economic and social development on agriculture, evaluated and analyzed the agricultural dependence of Jiangsu Province and even the entire country, and found that the level of agricultural dependence is negatively correlated with the level of regional economic development. In order to understand the degree of dependence of people and the economy on land in the human-land relationship system in rural areas and the spatial and temporal variation laws, the study puts forward the hypothesis of human-land relationship (Figure 1): Hypothesis 1 (time difference): There are three stages of farmers’ dependence on land: the first stage, in the early stage of reform and opening up, farmers have a high degree of dependence on land. At this stage, farmers lack the opportunity to go out to work and can only passively rely on land. In the second stage, based on national policy support and urban economic development, farmers have more opportunities to go out to work, and their dependence on rural land will decrease. In the third stage, with the development of rural economy and the change of farmers’ main needs, some farmers may return to their hometowns to help the poor, contract land and carry out some e-commerce activities of agricultural products. At this time, farmers’ dependence on land rises again and turns into active dependence. Hypothesis 2 (spatial difference): The progress of farmers’ dependence on land in developed and underdeveloped regions is different. In the early stage, farmers in developed areas can break away from the passive dependence on land faster and earlier; comparatively speaking, it takes longer time for farmers in underdeveloped areas to get rid of passive dependence and enter the active dependence stage later. In the context of the above hypothesis, a very important point is to quantify the dependency relationship, so this study referred to Wang Yingli’s model of agricultural dependence and attempted to construct a model of land dependence in rural areas. The spatial and temporal evolution characteristics of land dependence in rural areas of 30 provincial administrative units in China from 2006 to 2020 were analyzed. It should be noted that the land dependence studied here does not specifically refer to the degree of dependence on a certain type of land use in rural areas, but refers to the degree of dependence of people on land functions (such as population carrying capacity) or products (such as agricultural and economic output) in the spatial scope based on rural land. The improvement of the research lies in drawing on the method of economic dependence, quantifying the dependence of “human-land” and analyzing long-term changes in time series, which is conducive to a clearer portrayal of the relationship between human and land.

2. Materials and Methods

Establishing a systematic and perfect index system to comprehensively evaluate the rural land dependence of China’s provincial administrative units is the basis for quantifying the relationship between rural people and land, solving rural land problems, and promoting rural revitalization and development. The research uses the six indicators of the number of large agricultural employees in rural areas, the total population of the region at the end of the year, the area of rural areas, the total area of the region, the large agricultural production, and the gross production of the region in the statistical data. Starting from human life, life and production, the evaluation index system of rural land dependence is constructed based on population dependence, spatial dependence, and economic dependence. The entropy weight method is used to assign weights to each index and calculate the rural land dependence. Then the time evolution process of rural land dependence is analyzed, and the spatial pattern change of rural land dependence is analyzed by GIS spatial visualization. Finally, the geodetector model is used to analyze the influencing factors of rural land dependence. The specific framework is shown in Figure 2.

2.1. The Concept of Dependence and the Evaluation Model of Rural Land Dependence

2.1.1. The Concept of Dependency and Related Applications

In 1946, the American economist W. A. Brown proposed “interdependence” in his book Reinterpretation of the International Gold Standard System 1914–1943. It reflects the degree of interdependence of a country’s economy with other countries’ economy or with the world economy. It was later widely used in international economics, econometrics and development economics. Dependence is actually a proportion index. For example, the quantitative performance of foreign trade dependence is the ratio of a country’s total import and export trade to its GDP, which can reflect the degree of a country’s economy relying on foreign trade to a certain extent. Some scholars have also applied it to agriculture-related fields. For example, Zhu Huajun [23] constructed the dependence of agricultural mechanization development on financial investment, and Wang Yingli [15,16,17,18,19] constructed the dependence of economic and social development on agriculture.

2.1.2. Evaluation Model of Rural Land Dependence

(1)

Evaluation Index System

With reference to the definition of the concept of dependence, land dependence in rural areas refers to the dependence of economic and social development on rural land in terms of population absorption, space carrying and economic output in the regional system of rural human-land relationship. Therefore, the evaluation index system of rural land dependence is mainly constructed from three aspects: population dependence, spatial dependence and economic dependence (

Table 1. Evaluation index system of land dependence in rural areas.

Evaluation Standard	Evaluation Index	Weight
Population dependence	the number of large agricultural employees in rural areas/the total population of the region at the end of the year	0.368
Spatial dependence	the area of rural areas/the total area of the region	0.173
Economic dependence	the large agricultural production/the gross production of the region	0.459

Note: The number of large agricultural employees in rural areas is the number of rural agricultural, forestry, animal husbandry and fishery practitioners. The number of agricultural, forestry, animal husbandry and fishery practitioners in the first industry in the statistical data is subtracted from the number of agricultural, forestry, animal husbandry and fishery practitioners in urban units. The total population of the region at the end of the year is the population at the end of the year in the China Statistical Yearbook; the area of rural areas is the sum of the actual area of rural housing and the area of agricultural land in the statistical yearbook of urban and rural construction in China. The total area of the region is the provincial area; the large agricultural production is the total production of agriculture, forestry, animal husbandry and fishery in statistical data; the gross production of the region is the gross regional product in statistical data.

). This is because on the one hand, land can provide production function for farmers, and provide economic output, bring economic benefits and create economic dependence value for farmers through production function; provide living space and create space-dependent value for farmers through social security function; create population-dependent value by absorbing labor. On the other hand, land can provide livelihood strategies for the rural poor, enhance the creativity of the poor in rural areas, and achieve sustainable livelihoods, which is also a key and important way to achieve sustainable development [24,25,26].

In the human-land relationship system, human have subjective initiative [27], and are the premise of economic and social development. All development is man-based, while land can carry all things and has a certain capacity to accommodate. It is the material foundation and spatial carrier [28]. Generally, economic benefits are an important goal pursued in the human-land relationship. In the rural human-land relationship territorial system, the number of rural populations directly relates to the development of agriculture and rural areas. For example, with the acceleration of urbanization, there are some problems in rural areas, such as population hollowing out [29,30], population aging [31,32] and so on. The problem of rural labor loss [33] is derived. However, it is unclear when and where such loss occurs and what the spatial differences and characteristics are. Therefore, it is necessary to understand the absorption level and changes of rural space on the population within the system. Although the proportion of rural population engaged in non-agricultural industries has increased, the characteristics of rural population outflow are obvious. However, in some places, there is a phenomenon of “not leaving the hometown but leaving the land”. Because in the absence of sound social security functions, rural land not only has spatial carrying capacity but also plays an important role in social security. Therefore, within a certain period and scope, people who leave the countryside still “occupy” rural space. In some areas, there is even a phenomenon of rural hollowing-out and expansion coexisting, causing waste of land resources. Therefore, it is necessary to understand the spatial carrying status of rural land. Lastly, economic benefits are also an important goal of human-land relationships in rural areas, and the income brought by land is a key factor in measuring whether it can retain farmers. If the income brought by land can offset the cost and have a certain surplus, it can promote the human-land relationship to some extent [34]. Farmers invest a lot of time, labor, and equipment, and the output and benefits of land must meet certain standards, satisfy farmers’ living expenses, and be economically feasible, so that rural areas can retain famers.

(2)

Evaluation Method

The research draws on the evaluation method of the dependence proportion index, and uses the number of large agricultural employees in rural areas/the total population of the region at the end of the year to represent population dependence, the area of rural areas/the total area of the region to represent spatial dependence, and the large agricultural production/the gross production of the region to represent economic dependence. The weight of each dependence index is calculated by the entropy weight method. The basic idea of the entropy weight method is to determine the objective weight according to the size of the index variability, and finally use the comprehensive weighted average method to calculate the total dependence. The specific steps are:

（1)

Selection of Index

Suppose there are $r$ years, $n$ provinces and $m$ indicators, then $X_{\theta ij}$ denotes the year $\theta$ and the $j$ th indicator of the $i$ th province.

(2)

Standardization

The range method is used to standardize each index to eliminate the influence of dimension between different units on the data results.

$$X_{\theta ij}^{\prime }=\frac{X_{\theta ij}-X_{\min }}{X_{\max }-X_{\min }}$$

(1)

(3)

Index Information Entropy Calculation

$$e_j=-\ln {(rn)}^{-1}\sum _{\theta }^r\sum _i^n\left(P_{\theta ij}\times \ln \left(P_{\theta ij}\right)\right)$$

(2)

In the formula, $P_{\theta ij}$ is the intermediate quantity and has no practical significance, $P_{\theta ij}=\frac{x_{\theta ij}^{\prime }}{{\displaystyle \sum _{\theta }^r\sum _i^n}{x}_{\theta ij}^{\prime }}$, $e_j$ is the information entropy of the $j$ index;

(4)

Weight Calculation of Indexes

$$w_j=\frac{1-e_j}{{\displaystyle \sum _j^m}{e}_j}(j=1,2\cdots m)$$

(3)

(5)

Comprehensive Weighted Average

$$F=\sum {\mathrm{X}}_{\theta ij}^{\prime }\times w_j$$

(4)

2.2. Geodetector

Geodetector is a new statistical method for detecting spatial heterogeneity and revealing its underlying driving factors. The basic idea is to assume that the study area is divided into several sub-regions. If the sum of the variances of the sub-regions is less than the total variance of the region, then there is spatial heterogeneity. If the spatial distribution of two variables tends to be consistent, then they have statistical correlation. The $q$ value can be used to measure spatial heterogeneity, detect explanatory factors, and analyze the interaction between variables [35], and the expression is shown in Formula (5).

$$q=1-\frac{{\displaystyle \sum _{h=1}^L}{N}_h{\sigma }_h^2}{N{\sigma }^2}=1-\frac{SSW}{SST}\hspace{1em}\hspace{1em}\hspace{1em}SSW=\sum _{h=1}^L{N}_h{\sigma }_h^2,SST=N{\sigma }^2$$

(5)

In the formula, $L$ is the stratification of variable $Y$ or independent variable $X$; $N_h$ and $N$ are the layer $h$ and the unit number of the whole region; ${\sigma }_h^2$ and ${\sigma }^2$ are the variance of layer $h$ and the $Y$ value of the whole region respectively; $SSW$ and $SST$ are the sum of intra-layer variance and the total variance of the whole region, respectively. The range of $q$ statistic is (0, 1], and the larger the $q$ value is, the more significant the spatial differentiation of land dependence change is [36].

Interaction detector mainly judges whether there is interaction between two factors, as well as the strength, direction, linearity or non-linearity of the interaction, by separately calculating and comparing the $q$ values of each single factor and the $q$ value after the two factors are superimposed. The superposition of two factors includes both multiplication and other relationships, and the specific judgment basis are shown in

Table 2. Interaction between independent variables and dependent variables.

Judgment Basis	Interaction
$q\left(X_1\cap X_2\right)<Min\left(q\left(X_1\right),q\left(X_2\right)\right)$	Non-linear weakening
$Min\left(q\left(X_1\right),q\left(X_2\right)\right)<q\left(X_1\cap X_2\right)<Max\left(q\left(X_1\right),q\left(X_2\right)\right)$	Single-factor nonlinear weakening
$q\left(X_1\cap X_2\right)>Max\left(q\left(X_1\right),q\left(X_2\right)\right)$	Two-factor enhancement
$q\left(X_1\cap X_2\right)=q\left(X_1\right)+q\left(X_2\right)$	Independence
$q\left(X_1\cap X_2\right)>q\left(X_1\right)+q\left(X_2\right)$	Non-linear enhancement

.

2.3. Data Sources

(1) The data used in the research mainly includes the China Statistical Yearbook from 2007 to 2021 and the China Urban-Rural Construction Statistical Yearbook from 2006 to 2020, which are used to obtain data on the number of large agricultural employees in rural areas, the total population of the region at the end of the year, the area of rural areas, the total area of the region, the large agricultural production, the gross production of the region, the per capita gross domestic product, the per capita net income of farmers, the registered population of villages and the provincial statistical yearbook are obtained. All statistical data are downloaded through the National Bureau of Statistics, provincial bureaus or people’s governments, the Ministry of Housing and Urban-Rural Development of the People’s Republic of China. China’s administrative units such as Tibet Autonomous Region, Hong Kong Special Administrative Region, Macao Special Administrative Region and Taiwan Province are not included in the data collection. Therefore, they are not studied in the research.

(2) The vector base map data used in the study were downloaded from the standard map service website of the Ministry of Natural Resources (http://bzdt.ch.mnr.gov.cn/index.html (accessed on 14 September 2022)), and the map review number is GS (2019) 1822.

2.4. Summary of Second Section

This section mainly introduces the analysis ideas of rural land dependence, the methods used in the research (entropy weight method, GIS spatial analysis, geodetector) and data (statistical data, vector data).

3. Results and Analysis

Combined with the analysis framework of rural land dependence in the second part, this section analyzes the evolution and development stage of rural land dependence in China from the perspective of time, analyzes the distribution change of rural land dependence in China from the perspective of space, and explores the influencing factors of this change.

3.1. Evolution of Rural Land Dependence over Time

(1) Total dependence analysis. According to the evaluation results of rural land dependence, the changing trend chart of rural land dependence from 2006 to 2020 (Figure 3) is drawn to describe the temporal evolution characteristics of rural land dependence in China. The details are as follows:

(1)

The overall dependency values change in the range of [4.01%~82.88%], and the amplitude of variation of each province is in the range of [−29.6%~21.1%].

(2)

Since 2006, the rural land dependence of 30 provincial administrative units has shown a continuous downward trend, but the decline process also has a “phased” evolution law, which is roughly divided into four stages (Figure 4): fluctuation decline period (2006–2012), stable decline period (2012–2018), rebound increase period (2018–2019), rapid decline period (2019–2020).

(3)

Among the provinces, Hainan Province has the highest rural land dependence, reaching the highest in 2006, with the highest value of 82.88%. In addition, Yunnan Province, Guizhou Province and Guangxi Province have relatively high rural land dependence, all above 60%. The rural land dependence of Shanghai, Beijing and Tianjin have always been relatively low, all below 20%. Shanghai has the lowest rural land dependence in 2020, with a value of 4.01%.

(2) Dependency index analysis. The changing trend chart of each index of dependence from 2006 to 2020 is drawn (Figure 5). The analysis has found that:

(1)

On the whole, the population dependence and economic dependence indexes show a downward trend, and the spatial dependence index increases slightly.

(2)

The changes of population dependence and economic dependence indexes have both commonalities and respective characteristics. They are all mainly based on fluctuation changes. Among them, the population dependence index shows a gradual downward trend, which is divided into three stages (Figure 6), that is, first fluctuating decline, then steady decline, and finally rapid decline; the economic dependence index shows a fluctuating downward trend, which is divided into two stages (Figure 7), that is, the fluctuating decline first and then fluctuating rise.

(3)

The spatial dependence change is relatively stable, which can be divided into three stages (Figure 8), that is, the medium level remaining stable, and remaining stable at a low level after declining in 2012, and finally rising to a high level and remaining stable in 2018.

3.2. Spatial Pattern Change of Rural Land Dependence

Based on the ArcGIS platform, comprehensively considering the mode and frequency, using the natural breakpoint method to divide the rural land dependence into three types: low dependence (dependence ≤ 0.39, light green), middle dependence (0.39 < dependence ≤ 0.57, green), high dependence (dependence > 0.57, deep green), and display its spatial visualization (Figure 9). The analysis has found that:

(1) The high dependence area changes from a centralized state in the southwest to a discrete state in the 15 years on research. In 2006, the high dependence shows a “U” shaped distribution, and mainly concentrates in Gansu in the northwest, Sichuan, Yunnan, Guizhou, Guangxi in the southwest, and Hunan, Hubei, Henan, Anhui, Hainan in the central and southern regions; until 2013, the high dependence provinces in the southwest show a significant contraction, and Heilongjiang in the northeast begins to show high dependence. By 2020, the country with high dependence shows a “southwest-northeast” diagonal discrete distribution, and the high dependence is reduced to Yunnan in the southwest and Heilongjiang in the northeast; Hainan has maintained a state of high dependence during the study period.

(2) The middle dependence area is concentrated in the northern region and spreads from north to south, with the number increasing first and then decreasing. In 2006, there were 12 middle dependence regions. Apart from Jiangxi and Fujian provinces, the rest of the provinces are concentrated in the northern region, mainly including Xinjiang, Qinghai, Inner Mongolia, Ningxia, Heilongjiang, Liaoning, Hebei, Shandong, Shaanxi and Chongqing. From 2007 to 2009, Jilin Province becomes middle dependence; from 2010 to 2012, the number of middle dependence areas remain unchanged at 14, but there are slight differences in distribution. Chief among them is that Henan and Anhui change from high dependence to middle dependence, and Qinghai changes from middle dependence to low dependence. In 2013, the spatial pattern of middle dependence changes from northeast concentration to northwest concentration, and the number also increases. By 2014, the number of middle dependence reaches the maximum of 18. The number of middle dependent regions begins to decrease from 2017, and by 2020 it is 13. The overall distribution spreads from north to south.

(3) The low dependence changes from a discrete state to a concentrated distribution in the eastern coast, and the number gradually increases. In 2006, the low dependence areas mainly include Shanxi, Beijing, Tianjin, Jiangsu, Shanghai, Zhejiang and Guangdong. In 2015, Qinghai in the west changes from middle dependence to low dependence. In 2018, the low dependence begins to expand to the south and east with Beijing, Tianjin and Hebei as the center. By 2020, the low dependence is concentrated in the eastern coastal areas, and the central and western regions are scattered and discrete distribution.

3.3. Analysis of the Influencing Factors of Rural Land Dependence

3.3.1. Single Factor Detection Analysis

In this study, the geodetector model was used to analyze the influencing factors of land dependence by taking the number of large agricultural employees in rural areas (X1), the total population of the region at the end of the year (X2), the area of rural areas (X3), the total area of the region (X4), the large agricultural production (X5) and the gross regional production value (X6) in the evaluation index as six factors (Figure 10). The results show that: Overall, the six influencing factors showed fluctuating changes, among which X1 decreased after fluctuating, and X2, X3, X4, X5, and X6 increased after fluctuating. The influencing force of X1 is the highest in 2008, with a q value of 0.460. Since 2018, the ranking of X1’s q value has started to decline, while the ranking of X2, X3, and X4’s q values has started to rise. The q values of X3 and X4 are among the top two, which may be related to the sudden decline in the population dependence index and the sudden increase in the spatial dependence index in 2018. This indicates that with the development of technology and the progress of the times, the main influencing factors of land dependence will also undergo some changes under the influence of the social environment. Spatial dependence factors (X3 and X4) are gradually replacing the number of large agricultural employees (X1) as the dominant factor of land dependence. Secondly, in terms of each dependence index, X1 and X2 in the population dependence factors show opposite changes. From 2006 to 2020, the q value of X1 shows a downward trend, while the q value of X2 shows an upward trend, even surpassing X1 from 2019. The spatial dependence factors (X3 and X4) show a trend of increasing in the same direction, but the q value suddenly decreased in 2010, which may be due to the influence of other factors. The economic dependence factors (X5 and X6) show a relatively consistent upward trend, among which X5 slightly fluctuated upward, while X6 fluctuated greatly.

3.3.2. Interaction Factor Detection Analysis

As shown in Figure 11, in the interactive detection results of rural land dependence in 30 provincial administrative regions in China from 2006 to 2020, the combined effect of any two factors is greater than that of a single factor. The specific results are as follows:

(1) On the whole, the interaction between X6 and other factors is more intense. In 2006–2013 and 2017–2018, the interaction with X1 is very strong, and the interaction in 2008 is the strongest, and the interaction q value is 0.913. In 2014–2016, the interaction with X3 is very strong, indicating that after the interaction between X6 and other factors, the impact on land dependence has increased greatly. From 2019 to 2020, X4 respectively has strong interaction with X2 and X1, and the interaction q values respectively are 0.795 and 0.892. The interaction between X3 and X4 is weak in 2007 and 2010–2020, and the interaction is the weakest in 2010, and the interaction q value is 0.195, indicating that after the interaction of spatial dependence factors, the degree of influence on land dependence increases little. The interaction between X2 and X5 is weak in 2006 and 2008–2009.

(2) Firstly, in terms of each dependence index, with the change of time, the interaction between population dependence factors (X1, X2) and spatial dependence factors (X3, X4) is on the rise, and the interaction is the strongest in 2020, with an average q value of 0.848. Secondly, the interaction between population dependence factors (X1, X2) and economic dependence factors (X5, X6) shows a fluctuating downward and then fluctuating upward trend. The interaction is the weakest in 2014, with an average q value of 0.568, whereas the interaction is the strongest in 2020, with an average q value of 0.669. Thirdly, the interaction between spatial dependence factors (X3, X4) and economic dependence factors (X5, X6) shows a fluctuating upward trend, and the interaction is the strongest in 2020, with an average q value of 0.843. The above changes show that the spatial dependence factors (X3, X4), as a limiting factor, have a very strong impact on the land dependence after interacting with the population dependence factors (X1, X2) and the economic dependence factors (X5, X6).

The above results show that the interaction of various factors on the change of rural land dependence is dynamic and complex. Therefore, the synergy of various factors must be comprehensively considered in exploring the evolution of land dependence.

3.4. Summary of Third Section

This section constructs the evaluation index system of rural land dependence, and uses the entropy weight method to measure the level of rural land dependence of 30 provincial administrative units in China from 2006 to 2020. It also uses GIS spatial analysis method to explore the spatial and temporal pattern evolution characteristics of China’s land dependence, and uses geodetector to analyze the influencing factors, and quantitatively reveals the influencing factors of the evolution of rural land dependence in 30 provincial administrative regions of China from 2006 to 2020.

4. Discussion

(1) In terms of time series changes, rural land dependence has shown a continuous downward trend since 2006, which is consistent with the hypothesis mentioned in the introduction. This is because the development of urbanization provides more non-agricultural employment opportunities for the agricultural labor force, and the progress of agricultural science and technology further liberates the rural labor force, both of which promote the rural population into the city and the transfer of agricultural labor force to non-agricultural industries. However, from a systematic point of view, urban development is not to completely submerge rural areas, and there is no need for rural areas, so the sustainable development of rural areas is indispensable. At present, the problem of the loss of labor quantity and the decline of labor quality, as well as the resulting idle and abandoned rural land, is an important problem that needs to be solved in stages and regions. On the one hand, the government may support and encourage the development of farmers’ professionalization, and implement policies according to circumstances. This encouragement and support mainly include three aspects: legal system, policy support, and education and training [37]: (1) The government may introduce relevant laws and regulations on farmers’ professionalization to ensure that new professional farmers can exercise their rights and protect the rights and interests of new professional farmers; (2) The government may introduce preferential policies and treatment related to the professionalization of farmers, attract farmers to return home actively, and even attract urban silver talents to become professional farmers and take the road of urban nurturing rural areas; (3) The government may be able to provide professional farmers with sound education and training, provide farmers with a large platform for education, solve the problem of low education level of rural population, and encourage farmers to actively learn agricultural technology, improve agricultural production, and realize mechanized operations; keep up with the trend of the times, and actively promote farmers to rely on their own to achieve and complete the whole process of rural agricultural products from production (such as farming, cultivation, harvesting) to the combination of network and offline sales (such as packaging, sales). On the other hand, the power of all sectors of society may provide a big stage for farmers to enhance the well-being, work value and social recognition of professional farmers. On the other hand, the rural grassroots should actively respond to national policies and revitalize the population and land elements in the village. The village collective should adjust measures to local conditions, according to the characteristics of the village, actively develop the rural green industry, drive the farmers, stimulate the enthusiasm and vitality of the farmers’ labor, and use the idle agricultural land and homestead to form the village characteristic industry, improve the competitiveness of the village itself, the competitiveness of the village agriculture, and promote the sustainable development of agriculture and rural areas.

(2) In terms of spatial pattern, the degree of rural land dependence during the study period showed a regional difference of “high in the southwest and low in the eastern coast”. This is because the natural conditions in the eastern coast of China are favorable, the secondary and tertiary industries are developing rapidly, and the benefits are high compared with the primary industry, resulting in low dependence of farmers on land. However, most of southwest China is an underdeveloped area. Due to the limitation of terrain and traffic conditions, there are few opportunities for employment in other industries, which leads to a high dependence of farmers on land. The spatial distribution of high dependence was initially concentrated in the underdeveloped areas in the southwest, showing a continuous distribution. With the change of time, it turned into a discrete distribution, and the number decreased. The dependence is concentrated in the farming-pastoral ecotone in central and northern China, and the agriculture and animal husbandry are relatively developed. It spreads from the central and northern parts of China to the south, and the number increases first and then decreases. The spatial distribution of low dependence has changed from dispersion to concentration in the eastern coast, and the number has gradually increased.

(3) In terms of influencing factor analysis, from 2006 to 2018, X1 was the main influencing factor of land dependence. The population dependence index shows a downward trend and the spatial dependence index shows an upward trend. This may be because the urbanization development at this stage has not yet formed a certain scale. Most farmers adopt the form of part-time work and plant agriculture while working, resulting in a slight loss of land labor force. From 2018 to 2020, the spatial dependence factor (X3, X4) has gradually become the main influencing factor affecting the degree of land dependence. This may be because with the rapid development of urbanization, most farmers choose to give up farming and go to the city to work, but they have certain feelings for the land. They still “occupy” a certain space in the countryside, that is, “not leaving the hometown but leaving the land”, which leads to the increase of rural space instead of decreasing. As a limiting factor, spatial dependence factor will have a very strong impact on land dependence after interaction with population dependence factors (X1, X2) and economic dependence factors (X5, X6). The average q values after interaction are 0.848 and 0.843, respectively. This shows that rural land space is very suitable, but the idle rural land caused by a series of problems such as labor loss will be a very worthy of attention and attention.

In the human-land relationship system, human have subjective initiative and can understand, use, change and protect the geographical environment. Human have a strong dependence on the land [38], which is manifested in the population carrying function, agricultural output function and economic output function of farmers relying on the land. The land is the material basis and space carrier for human survival, and the geographical environment restricts the depth, breadth and speed of human social activities [39]. From the perspective of human-land relationship, this study aims to solve the problems in rural areas and help rural sustainable development. Internationally, some countries have also done research to solve the sustainable development of rural areas. For example, Clark N. Melendres et al. evaluate the effectiveness of the Bannay Island Highlands Sustainable Rural Development Project and find that the project significantly increased the rice yield and farm income of beneficiaries [40]. Aceleanu Mirela Ionela et al. promote rural sustainable development through research and analysis of renewable energy in rural areas of Romania [41]. Gema Cárdenas Alonso et al. find through research that the development of rural areas in Estremadura, Spain, is not sustainable enough [42]. Wiebke Wellbrock et al. believe that it is necessary for sustainable development to let the public participate in and develop rural innovation system [43].

5. Conclusions

(1) In the system of human-land relationship, the dependence of human and land is mutual. Not only does human depend on land, but also depends on the accommodation space provided by land and the output demand given by land to human within a certain range. At the same time, land also has a certain dependence on human, which depends on human’s management, maintenance and governance. The specific performance is that the quality of land, the function of land and the output benefit are strictly dependent on the management and maintenance of farmers. This study only explores the dependence and spatial-temporal changes of some functions (such as population carrying capacity) or products (such as agricultural output and economic output) of human to land under the interaction of human-land relationship in the spatial scope of rural land as the carrier, and believes that active high dependence is a state conducive to the sustainable development of agriculture and rural areas, which has certain reference value for quantifying human-land relationship to a certain extent, and also provides ideas and methods for academic discussion and research of human-land relationship. In future research, we will focus on exploring the degree index and difference of bringing well-being to human.

(2) In addition, the incompleteness of statistical data and the inconsistency of statistical caliber to some extent affect the collection of index data. To ensure the consistency of data connotation, the following treatments were made to the data: (1) The number of large agricultural employees in rural areas from 2017 to 2020 was determined by subtracting the number of employees in non-private urban agricultural, forestry, animal husbandry, and fishery units from the number of primary industry employees. Due to inconsistency with other provinces’ statistical indicators, the number of rural large agricultural employees in Liaoning Province in 2019 and in Shanxi Province from 2007 to 2009 was used for research. (2) Since specific data on the gross agricultural output value of large agricultural production in rural areas were not published, the rural and urban total output values were used for research on economic dependence indicators. The replacement of the above statistical data may have a slight impact on the research results.