A Simulation-Based Study on the Coupling Coordination of Farmers’ Livelihood Efficiency and Land Use: A Pathway towards Promoting and Implementing the Rural Development and Rural Revitalization Strategy

Abstract

The interaction process between farmers’ livelihood and land use is a core link between the interaction and the coupling of the human–land system. It is a complex evolutionary process that involves several issues such as resource-intensive utilization and economic development. This study first constructs a dynamic model of the coupling system of farmers’ livelihood efficiency and land use and designs two types of 31 scenarios based on the farmer’s livelihood mode and land use. This study further simulates the coupling coordination relationship in different scenarios and then proposes suggestions for sustainable development. The findings of our study show that with the development of time, the livelihood capital, livelihood output, and land use level first showed a decline and then an increase. The results also reveal that livelihood efficiency and coordination degree are related to the livelihood mode and land use mode of farmers, while the land use level is not significantly associated with the livelihood mode of farmers. Pure-agriculture farmers have the lowest livelihood efficiency and coordination degree when they have no planting and breeding poultry, while part-time farmers have the highest land use level and coordination degree when they plant cash crops and breeding livestock. Besides, non-agriculture farmers have the highest livelihood efficiency and the lowest land use level when they neither plant crops nor breed livestock. To improve the level of coupling coordination, it is necessary not only to flexibly adjust the farmer’s livelihood and land use mode but also to optimize the allocation of various resources to promote the coordinated and sustainable development of farmers.

1. Introduction

Since industrial civilization, the wealth accumulation of human society has been accompanied by huge contradictions and conflicts between humans and land. Rapidly growing consumption demand and structural changes have created increasing pressure on the limited resource and environmental base, especially for countries and regions with large populations and strong development needs [1,2]. Up to now, global issues such as resource contention, environmental pollution, food crisis, resource shortage, and global warming are still the relationship between human activities and resources and the environment [3,4,5]. In this process, the human–land relationship has been studied by many fields such as ecology, computer science, environmental science, and economics [6,7]. Moreover, under different cultural and social backgrounds, scholars have different emphases on the study of human–land relationship. Among them, the research perspective of western countries has gradually shifted from the human–land relationship to more specific areas of energy consumption, urban sustainable development, and carbon dioxide emissions [8,9,10]. The study of the human–land relationship in China pays more attention to the impact of macro policies. Moreover, scholars have explored land space planning, ecological civilization construction, strategic environment evaluation, and other aspects of the country’s main strategic needs [11,12,13].

The countryside plays an important role in the national economy and social development. As human society gradually evolves into an industrial economy, knowledge economy, platform economy, and digital economy, rural areas are constantly impacted by changes in the external development environment such as globalization, industrialization, urbanization, and natural disasters, resulting in labor shortages, economic recession, poverty, inefficient use of resources, and other problems in some villages, which seriously restricts the high-quality development of rural areas [14,15,16]. For example, since the 1950s, the population of rural areas in Japan has shown a downward trend. A large number of young people have moved to cities, resulting in the stagnation of public services, a shortage of farmers, poor environmental management, and other problems in rural areas. Although Japan’s rural revitalization strategy has played a certain role in promoting sustainable agricultural development and revitalizing the rural economy, it has not fundamentally solved the phenomenon of rural agriculture recession and rural withering. The “New Village Movement” implemented in South Korea since 1970 has achieved a good cycle of rural economic revival and villagers’ spontaneous promotion of rural economic development, but the use of a large number of chemicals has caused harm to the rural environment, and the small production scale has also restricted the agricultural production efficiency. The proportion of the rural population of China in the national population decreased from 89.36% in 1949 to 36.11% in 2020; the number of the rural population of China decreased from 671.13 million in 2010 to 509.92 million in 2020 [17]. The livelihood of farmers has gradually changed from agricultural production to non-agricultural production. The efficiency of livelihood has been improved, but the level of land use has been declining. In 2017, China proposed a rural revitalization strategy, but the problems of rural hollowness, agricultural marginalization, and farmers’ aging are still growing. At present, the problem of rural recession has become a common challenge facing all countries of the world. Accelerating the implementation of a rural revitalization strategy is a fundamental issue related to the national economy and people’s life [18,19].

The human–land system is a dynamic, open and complex giant system composed of populations, society, economy, resources, the environment, and other factors. Scientific cognition of the human–land relationship and prediction and regulation of the unsustainable mode of the human–land relationship are major scientific issues related to regional sustainable development [20]. Among them, the rural human–land system is a rural spatial system composed of population, economy, and resources. Moreover, the rural human–land system emphasizes that under the interaction between the rural core system and the external development environment system, the rural social and economic structure has been reshaped [21]. With the development of urbanization, the human–land relationship in rural China, which integrates ecological fragility, poverty concentration, and frequent disasters, has undergone more dramatic and profound changes. The accelerating trend of socioeconomic activities and the differentiation of land use patterns also make the relationship between farmers’ livelihoods and land increasingly complex and changeable [22]. How to coordinate the human–land relationship, promote efficient land use and promote sustainable growth of farmers’ income has become the primary problem in promoting agricultural and rural development and implementing the rural revitalization strategy [23].

The livelihood problem is a direct reflection of human initiative and structure in the study of the human–land relationship, and a multidisciplinary cross-core issue to coordinate the human–land system, promote regional sustainable development, and solve the contradiction between people’s growing needs for a better life and unbalanced and insufficient development [24]. Based on the sustainable livelihood framework, scholars focused mainly on livelihood capital, livelihood mode, livelihood vulnerability, livelihood output, livelihood resilience, and other aspects [25,26,27] and pointed out that the lack of livelihood capital stock and unreasonable allocation, the weak feasibility of livelihood modes, and the lack of livelihood paths are the root causes of backward rural development and slow rural revitalization process [5,28,29]. Land use change and its driving mechanism are the frontier and key fields of global change and sustainable development research [30]. In previous studies, scholars explored land use type, land use mode, land use structure, land use degree, and land use efficiency [31,32], and organically combined land use with ecosystem services, urban and rural development, resources and environment, food security, and other issues [33,34,35,36].

In recent years, scholars have begun to pay attention to the symbiotic relationship between farmers’ livelihoods and land use [37,38,39], and have explored the relationship between livelihoods and land use from the perspective of farmer types, and geographical regions [40]. Among them, livelihood represents people’s use of various resources [41]. Differences in society, economy, resources, and other factors make farmers have diversified choices in land use [42]. The evolution of farmers’ livelihood capital endowment and the livelihood mode will drive them to change the mode, efficiency, intensity, and structure of land use, and have a direct or indirect impact on the ecological environment system [43,44]. Land is the most direct object for farmers to choose their livelihood strategies. Land degradation, changes in land use patterns, and changes in land use intensity will have an impact on farmers’ livelihood capital and livelihood strategies [45,46]. When the land use mode changes, the adaptability of farmers’ livelihoods will also change. Among them, the efficient use of land resources can increase farmers’ income, improve their economic status, and reduce their livelihood risks [47].

It can be found from previous studies that scholars mostly explore the relationship between farmers’ livelihood and land use from a qualitative perspective, while few research from a systematic perspective [48]. With the gradual popularization of computer technology, relevant research has gradually changed to quantitative evaluation and simulation research [49]. However, previous studies focused more on the relationship between a single livelihood link (livelihood capital, livelihood strategy, livelihood output) and land use [50], and more on the static coupling relationship between farmers’ livelihood and land use [51]. There was less research on the dynamic feedback relationship between the two and less on the comprehensive reflection of changes in the system from the perspective of process and efficiency. Furthermore, relevant previous achievements have mostly explored the principle of the human–land system in river basins, regions, and countries [52], such as the Qinghai–Tibet plateau, the Inner Mongolia plateau, South Africa, India, Nigeria [53,54,55], and paid less attention to the dynamic evolution between the livelihood and land use of farmers.

Therefore, this study takes the Qinba mountainous area in southern Shaanxi province as an example to build the coupling system dynamics model and evaluate the development of the two subsystems of livelihood and land use with the help of livelihood efficiency and land use level. By simulating the coupling coordination relationship under different livelihood and land use modes, we hope to provide targeted suggestions for improving the livelihood efficiency, land use level, and coordination of farmers in rural areas.

2. Materials and Methods

2.1. Research Area

The Qinba mountainous area in southern Shaanxi province, including Hanzhong, Ankang, and Shangluo (Figure 1), covers an area of 7.03 × 10⁴ km², accounting for 34.19% of the area of Shaanxi province [56]. It is a typical poverty-alleviated mountainous area, with rugged terrain, weak infrastructure, low-quality of cultivated land, a fragile environment, and other issues that constrain farmers’ land use behavior. Farmers can only go out to seek more development opportunities to maintain and pursue their own livelihood needs, making extensive land use, degradation of ecological functions of land and other issues increasingly prominent, and the human–land conflict is increasingly intensified [57]. In the period of comprehensively promoting high-quality development, how to balance the relationship between the survival and development of farmers and land use on the premise of resource conservation and environmental protection is not only the key to solving the “three rural issues”, but also an important basis for promoting the long-term revitalization of the countryside.

2.2. Data Sources

The parameters involved in this study (birth rate, death rate, among others) are mainly from the Shaanxi Statistical Yearbook, Hanzhong Statistical Yearbook, and other statistical data related to the research area from 2019 to 2020. The basic data involved (population size, educational level, health status, and land area, among others) are mainly from the data obtained by the research group in August 2020 in southern Shaanxi. The survey mainly adopted the stratified random sampling method to select the sample of the interviewed farmers. First, using principal component analysis and spatial cluster analysis, combined with the representativeness and typicality of counties, 25 sample counties were selected in southern Shaanxi province. Secondly, according to the per capita GDP and economic development of each township in the sample counties, two townships were selected for each sampled county. Thirdly, according to the economic development of each administrative village and the distance to the township government, two sample villages are selected for each sample township. Finally, six to eight sample farmers were randomly selected from each selected sample village. The steps of the random sampling procedure are as follows: ① The investigators communicated with the staff of the sample villages, obtained the list of permanent residents of the sample villages by households, and determine the sampling frame. ② Number the farmers in the sampling frame from 1, such as 1, 2, 3, 4, 5… 200. ③ Used the computer to conduct random sampling (for example, selected a cell in an Excel table and entered the formula “ = INT (RAND() ∗ (200 − 1 + 1)) + 1”. If we wanted to draw from 1 to 50, replace 200 with 50. Press F9 once to draw once) and draw 8 times for each sample village. ④ According to the sampled data, the sample farmers were determined and a household survey was conducted. In the process of the household survey, in order to ensure the quality of the questionnaire, the household head or the person familiar with the household situation is mainly selected as the interviewee. After the completion of each questionnaire, the investigator checked the questionnaire to prevent mistakes and omissions. ⑤ After the survey was completed, it was necessary to check and evaluate the representativeness of the sample farmers by comprehensively considering the income structure, family structure, production structure and other factors. When the farmers’ information was similar to the actual situation of the village, it was confirmed as valid data, and the valid questionnaire was coded and saved. ⑥ In the actual investigation process, if the eight households selected could obtain six to eight valid data, the investigation could be ended. If sufficient data could not be obtained, it was necessary to conduct supplementary sampling again with the help of computers until six to eight valid data were obtained.

The survey involved two parts of farmers’ livelihood (human capital, material capital, social capital, among others) and land use (planting type, breeding type, planting area, breeding area, among others), including 10 counties in Hanzhong City, 9 counties in Ankang City, and 6 counties in Shangluo City. A total of 639 questionnaires were obtained and 451 valid questionnaires were obtained after a rigorous data screening. In this survey, the proportion of males and females is basically the same. The proportion of farmers aged 21 to 65 years is 83.60%, the proportion of those with a level of education of junior high school and below is 68.96%, and the number of farmers is mostly three to five (

Table 1. Basic information of farmers.

Index	Category	Frequency Number	Frequency Rate
Gender	Male	247	54.77%
	Female	204	45.23%
Age groups	≤20 years old	37	8.20%
	21–65 years old	377	83.60%
	≥66 years old	37	8.20%
Education level	Junior high school and below	311	68.96%
	High school or technical secondary school	63	13.97%
	Junior college and higher	77	17.07%
Population size	≤2	29	6.43%
	3–5	283	84.92%
	≥5	39	8.65%
Region	Hanzhong	143	31.71%
	Ankang	180	39.91%
	Shangluo	128	28.38%

).

2.3. Research Methods

2.3.1. System Dynamics

System dynamics is a new interdisciplinary field that integrates cybernetics, system theory, information theory, and computer simulation technology [58]. It emphasizes the viewpoint of system, integrity, development, and movement and has significant advantages in exploring the problems of complex non-linear systems such as society, economy, and ecology that change over time [59]. As a comprehensive simulation model, it has been applied to explore land use change, land resource carrying capacity, land resource allocation, and land quality change [60,61,62]. This study mainly uses Vensim PLE software to build a system dynamics model based on the data of farmers’ livelihood and land use in the Qinba mountainous area in southern Shaanxi province, to explore the relationship between livelihood efficiency and land use.

2.3.2. Coupling Coordination Degree Model

Coupling refers to the phenomenon in which different systems interact with each other. The coupling degree refers to the degree of interaction between the systems. The coordination degree refers to the coordination state between systems [63]. This study mainly uses the coupling coordination degree model to measure the relationship between livelihood efficiency and land use. The calculation equations are as follows:

$$C=2\times {\left[\frac{u_1\times u_2}{(u_1+u_2)}\right]}^{\frac{1}{2}}$$

(1)

$$T=\alpha u_1+\beta u_2$$

(2)

$$D(u_1,u_2)=\sqrt{C\times T}$$

(3)

In Equations (1), (2) and (3), C is the coupling degree, reflecting the quality of the coupling development of the two systems. T is the development degree, reflecting the overall benefit or level of the two systems. D is the coordination degree, reflecting the coordinated development of the two systems. u₁ is the livelihood efficiency, u₂ is the land use level, α and β are weights [64]. Considering that livelihood efficiency and land use level are equally important to farmers, we believe that α and β take 0.5 for all.

3. Construction of the Coupling System Dynamics Model

3.1. System Structure Analysis

The interaction between the geographical environment and human activities constitutes the human–land system, and the differences in organizational units and research perspectives endow the human–land system with distinct hierarchical relationships [65] (Figure 2). From the system perspective, the “human–land system” at the farmer level, that is, the coupling system of livelihood efficiency and land use can be divided into two subsystems: livelihood and land use. Among them, livelihood is the way farmers earn a living, and land use is the way and state by which farmers use land resources. From a structural perspective, the operation of the livelihood and land use subsystems can be reflected in the livelihood efficiency and land use level, and the overall situation of the coupling system can be reflected in the degree of coordination degree of the livelihood efficiency and land use level. Among them, livelihood capital is the basis for farmers to choose livelihood modes and obtain livelihood results, livelihood mode is the way to allocate and use livelihood capital to achieve livelihood goals, livelihood output is the result of the combination of livelihood capital and livelihood mode, and livelihood efficiency is the comprehensive reflection of farmers’ livelihood ability and quality [66,67]. Land use intensity is the breadth and depth of land resource utilization, land use structure is the quantitative arrangement and spatial layout of land resources, land use benefit is the result of land resource utilization and allocation, and land use level is the comprehensive situation of land resource utilization and allocation by farmers [68,69]. From the element perspective, population, capital, land, society, and environment are the core elements of the composition and evolution of the coupling system of livelihood efficiency and land use. Among them, population, capital, and land are the basis for the interaction and coupling of the two subsystems of livelihood and land use, and society and the environment are important indicators of the integrated state of the coupling system [70]. From the indicator perspective, population size, educational level, health status, income, and other indicators are the most direct support for reflecting and evaluating the coupling system. Moreover, the development and evolution of the whole system can be clarified with the help of relevant indicators. Therefore, in order to accurately grasp the development and evolution law of the human–land system, which is the micro subject of farmers, it is critical to deeply analyze its dynamic mechanism, identify the core elements and system structure with the coupling system of livelihood efficiency and land use as the starting point.

In the coupling system of livelihood efficiency and land use, the livelihood efficiency and land use level of farmers will interact and influence each other. The livelihood capital of farmers will affect their livelihood modes, promote the differentiation of farmers’ production activities (agricultural activities and non-agricultural activities), and thus produce different livelihood outputs (employment opportunities, gross annual income, environmental quality) [71]. In this process, different livelihood needs will drive farmers to use land resources in various ways [72], and different livelihood capital will have an impact on farmers’ land use modes [50]. For example, with increasing age, the physical strength and health of farmers will decline and they will adjust the structure of the planting and breeding. With the improvement of knowledge and skills, farmers may engage in agricultural production in a modern and large-scale way, or they may extensively use land resources and engage in non-agricultural activities. With the increase in agricultural costs such as seeds, fertilizers, pesticides, and labor, farmers may choose agricultural activities or non-agricultural activities with lower input costs. At the same time, different modes of land use will restrict human capital, and natural capital, and encourage farmers to constantly adjust their modes of livelihood to increase the output of their livelihood [73]. Specifically, when farmers are engaged in non-agricultural production, it is difficult to effectively use land resources, which will cause a waste of livelihood resources to a certain extent and reduce the efficiency of livelihoods. When farmers are engaged in agricultural production, the restriction of land resources makes it difficult to effectively use their human capital, social capital, and financial capital. Moreover, farmers may introduce advanced production technology and expand production scale to better use livelihood capital and increase agricultural income. When farmers engage in part-time activities, they will improve their livelihood efficiency and level of land use by adjusting the proportion of input from agricultural and non-agricultural activities, and the structure of land use [74]. It can be said that with the rationalization of the way of livelihood and land use, a virtuous circle will be formed between the efficiency of livelihood and the level of land use, which will promote the sustainable and high-quality development of farmers.

3.1.1. Livelihood Subsystem

The purpose of the livelihood subsystem is to study the impact of different livelihood modes on farmers’ livelihood capital, livelihood output, livelihood efficiency, and land use level. Farmers’ livelihood subsystem mainly includes livelihood capital, livelihood mode, livelihood output, and livelihood efficiency, which can be measured by education level, health status, population size, number of livestock, and other indicators [75]. Based on the questionnaire data, the entropy method can be used to determine the weight of each indicator, and the DEA model can be used to measure the livelihood efficiency of farmers. The relevant indicators and weights are shown in

Table 2. Evaluation index system of farmers’ livelihood efficiency.

Evaluation Index		Variable	Weight
Livelihood capital	Human capital	Education level	0.513
		Health status	0.261
		Population size	0.226
	Physical capital	Number of breed	0.244
		Daily supplies	0.322
		Transportation tool	0.434
	Natural capital	Agricultural production area	0.706
		Housing area	0.294
	Financial capital	Gross annual income	0.139
		Loan channel	0.284
		Loan purpose	0.577
	Social capital	Family social work	0.527
		Trust of villagers	0.239
		Number of channels for help	0.234
	Information capital	Information tool	0.274
		Information channel	0.726
Livelihood output	Income level	Improvement of income level	0.253
	Education and medical level	Improvement of Education and medical	0.062
	Employment opportunities	Improvement of employment channel	0.367
	Rural attachment	Improvement of pride and attachment to hometown	0.184
	Ecological protection consciousness	Improvement of ecological protection awareness and values	0.134

.

3.1.2. Land Use Subsystem

The purpose of the land use subsystem is to study the impact of different land use modes on farmers’ land use levels, and the impact of land use level changes on farmers’ livelihoods. The farmers’ land use subsystem mainly includes land use structure, land use intensity, and land use benefit [76], which can be measured by per capita planting area, labor input, agricultural production, and other indicators [77]. Based on the questionnaire data, the entropy method can be used to determine the weight of each indicator, and the comprehensive evaluation method can be used to calculate the farmers’ land use level. The relevant indicators and weights are shown in

Table 3. Evaluation index system of farmers’ land use level.

Evaluation Index		Variable	Weight
Land use level	Land use structure	Per capita planting area	0.335
		Per capita breeding area	0.665
	Land use intensity	Land input	0.243
		Labor input	0.038
		Funds input	0.719
	Land use benefit	Agricultural output	0.374
		Agricultural output value	0.626

.

3.2. System Boundary Determination

This study mainly starts from the two subsystems of livelihood and land use and constructs the coupling system dynamics model. The spatial boundary of the model is the Qinba mountainous area in southern Shaanxi province, the time boundary is 2020–2070, and the simulation step is 1 year. In order to highlight the research object and the modeling purpose, this study assumes that the political, economic, social, and ecological environment of farmers’ life is stable, without major natural disasters, economic fluctuations, and policy changes. Moreover, farmers will choose appropriate livelihood modes and land use modes according to their own livelihood capital and livelihood needs.

3.3. Causality Analysis

Based on the analysis of the structure of the system, the causality diagram of the coupling system can be constructed (Figure 3). The main causal feedback loops are as follows.

(1) Population size → + Cultivated land area → + Agricultural income → + Gross annual income → + Education and medical level → + Rural attachment → + Population size.

(2) Gross annual income → + Education and medical level → + Rural attachment → + Population size → + Number of labor → + Non-agricultural income → + Agricultural income → + Gross annual income.

(3) Number of labor → + Non-agricultural income → + Gross annual income → + Ecological protection consciousness → + Rural attachment → + Population size → + Number of labor.

(4) Information channel → + Employment opportunities → + Non-agricultural income → + Agricultural income → + Gross annual income → + Education and medical level → + Rural attachment → + Population size → + Information channel.

(5) Social capital → + Information channel → + Employment opportunities → − Agricultural income → + Gross annual income → + Ecological protection consciousness → + Rural attachment → + Population size → + Social capital.

3.4. System Flow Diagram Establishment

Based on an analysis of the structure of the system and the causal feedback loop, a coupling system dynamics model can be built. The system flow stock diagram is shown in Figure 4. In general, the main variables involved in the model are as follows. (1) State variables, including population size, actual cultivated land area, and net income. (2) Rate variables, including an increase in population, increase in cultivation land, and increase in general income, among others. (3) Auxiliary variables, including planting area, planting expenditure, and the number of breeds, among others. (4) Constant, including the birth rate, death rate, land abandonment rate, and land transfer rate, among others.

3.5. Model Equation Design

According to the relationship and action principle of the variables in the system model, based on field survey data in the Qinba mountain area in the southern Shaanxi province, this study combined the entropy method, regression analysis, the arithmetic mean value method, the comprehensive evaluation method, and other methods to estimate parameters and derive model equations. After constantly debugging and operating the system, the coupling system dynamics model was established. The relevant parameters come mainly from statistical data and field survey data. The representative equations are as follows.

(1) Net income = INTEG (General income − aggregate expenditure, 28,895.82).

(2) Population = INTEG (Population increase − Population decrease, 3.97).

(3) Actual cultivated land area = INTEG (Cultivated land increase − Cultivated land decrease, 2.6).

(4) General income = Agricultural output value + Non-agricultural income.

(5) Aggregate expenditure = Other expenditure + Agricultural expenditure + Education and medical expenditure.

(6) Non-agricultural income = Number of labor ∗ 18,864.4 ∗ 1 + Employment opportunities ∗ 1000 ∗ 1 − Agricultural output value ∗ 1.25 ∗ 1.

(7) Human capital = Health status ∗ 0.261 + Education level ∗ 0.513 + Population size ∗ 0.226.

(8) Physical capital = Number of breeds ∗ 0.244 + Transportation tool ∗ 0.434 + Daily supplies ∗ 0.322.

(9) Natural capital = Housing area ∗ 0.294 + Agricultural production area ∗ 0.706.

(10) Financial capital = General income ∗ 0.139 ∗ 0.0002 + Loan channel ∗ 0.284 + Loan purpose ∗ 0.577.

(11) Social capital = Family social work ∗ 0.527 + Trust of villagers ∗ 0.239 + Number of channels for help ∗ 0.234.

(12) Information capital = Information tool ∗ 0.274 + Information channel ∗ 0.726.

(13) Income level = 2.468 + Information capital ∗ 0.182 + Physical capital ∗ 0.12 + Social capital ∗ 0.204 + Financial capital ∗ 0.142.

(14) Education and medical level = 3.033 + Information capital ∗ 0.137 + Social capital ∗ 0.164 + Financial capital ∗ 0.189.

(15) Employment opportunities =1.989 + Information capital ∗ 0.171 + Social capital ∗ 0.19.

(16) Rural attachment = 2.574 + Land use level ∗ 0.126 + Physical capital ∗ 0.144 + Social capital ∗ 0.215 + Natural capital ∗ 0.17 + Information capital ∗ 0.272.

(17) Ecological protection consciousness = 3.137 + Human capital ∗ 0.159 + Financial capital ∗ 0.116 + Natural capital ∗ 0.161 + Social capital ∗ 0.233 + Information capital ∗ 0.176 + Land use level ∗ 0.131.

(18) Land use structure = Per capita breeding area ∗ 0.665 + Per capita planting area ∗ 0.335.

(19) Land use intensity = Agricultural expenditure ∗ 0.719 ∗ 0.0001 + Agricultural production area ∗ 0.243 + Number of labor ∗ 0.038.

(20) Land use benefit = Agricultural output value ∗ 0.626 ∗ 0.0001 + Agricultural output ∗ 0.374 ∗ 0.0001.

(21) Land use level = Land use benefit ∗ 0.487 + Land use intensity ∗ 0.154 + Land use structure ∗ 0.359.

3.6. Model Inspection

To ensure the validity and rationality of the model, this study conducted a variety of tests, such as visual inspection, operation inspection, stability inspection, and historical inspection. Taking into account the degree of influence and the constituent elements of the variables, this study examined the output of three variables, net income, general income, and aggregate expenditure, in 3 months, 6 months, and 12 months. The test results are shown in Figure 5. In addition, this study selects 12 variables, including general income, aggregate expenditure, agricultural output value, agricultural output, agricultural expenditure, and other variables as test variables, and compares the system operation results with the average of previous data. The results are shown in

Table 4. Historical test results of some variables.

Variable	Simulation Value	Actual Value	Error
General income	61,595.70	61,824.27	−0.4%
Non-agricultural income	53,847.20	54,106.73	−0.5%
Agricultural output value	7748.48	7717.55	0.4%
Planting output value	2964.00	2939.07	0.8%
Breeding output value	4784.48	4778.48	0.1%
Agricultural output	2378.40	2349.90	1.2%
Planting yield	1820.00	1785.61	1.9%
Breeding yield	558.40	564.29	−1.0%
Aggregate expenditure	33,050.00	32,928.46	0.4%
Agricultural expenditure	3655.00	3662.53	−0.2%
Planting expenditure	2145.00	2193.45	−2.2%
Breeding expenditure	1510.00	1469.09	2.8%

.

It can be seen in Figure 5 that when the model inputs three different time steps, the variation range of each variable is small. Therefore, it can be considered that the model is at a stable level. It can be seen from Table 4 that the absolute error between the historical value and the analog value of the system model is controlled between 0.1% and 2.8%, and the precision of fitting is high, which can basically reflect the operation of the real system. In conclusion, the model established in this study can be used to reflect the real system and perform a simulation.

4. Simulation of the Coupling System

4.1. Scenario Settings

The coupling system of the efficiency of farming livelihood and land use is a complex system that involves social, economic, ecological, and other factors. The interaction process between the livelihood of farmers and the use of land is the core link of the development and change in this system. Drawing on relevant research [78,79] and combining the actual situation of farmers in the Qinba mountain area in southern Shaanxi province, this study designed 31 scenarios from the perspective of the mode of livelihood, the mode of land use, the mode of planting, and the mode of breeding (

Table 5. Scenario setting under different strategy combinations.

Scenario	Livelihood Mode	Land Use Mode	Planting Mode	Breeding Mode
1	Pure-agriculture farmers	Planting	Planting grain crops	No breeding
2			Planting cash crops
3			Planting grain crops and cash crops
4		Breeding	No planting	Breeding livestock
5				Breeding poultry
6				Breeding livestock and poultry
7		Both planting and breeding	Planting grain crops	Breeding livestock
8				Breeding poultry
9				Breeding livestock and poultry
10			Planting cash crops	Breeding livestock
11				Breeding poultry
12				Breeding livestock and poultry
13			Planting grain crops and cash crops	Breeding livestock
14				Breeding poultry
15				Breeding livestock and poultry
16	Part-time farmers	Planting	Planting grain crops	No breeding
17			Planting cash crops
18			Planting grain crops and cash crops
19		Breeding	No planting	Breeding livestock
20				Breeding poultry
21				Breeding livestock and poultry
22		Both planting and breeding	Planting grain crops	Breeding livestock
23				Breeding poultry
24				Breeding livestock and poultry
25			Planting cash crops	Breeding livestock
26				Breeding poultry
27				Breeding livestock and poultry
28			Planting grain crops and cash crops	Breeding livestock
29				Breeding poultry
30				Breeding livestock and poultry
31	Non-agriculture farmers	Neither planting nor breeding	No planting	No breeding

). Among them, farmers’ livelihood modes include pure-agriculture farmers, part-time farmers, and non-agriculture farmers. Land use modes include planting, breeding, both planting and breeding, and neither planting nor breeding. Planting modes include planting grain crops, planting cash crops, planting grain crops, cash crops, and no planting. Breeding modes include breeding livestock, breeding poultry, breeding livestock, poultry, and no breeding.

Among the 31 scenarios, the difference between livelihood mode and land use mode (planting mode and breeding mode) is the key to affecting the system state. Here, the actual cultivated land area (grain crop area, cash crop area), number of breeds (number of livestock, number of poultry) and non-agricultural income are selected as the scenario control factors, and two scenarios are set.

(1) Basic scenario. Without considering other factors, simulate how the coupling system will evolve under the most basic scenario according to the existing inertia development of the system. Based on arithmetic mean, regression fitting, and other methods, the parameters of related variables can be obtained. Among them, the actual cultivated land area is 2.6 (the proportion of grain crops is 50%, and the proportion of cash crops is 50%), the number of breeds is 12.88 (the number of livestock is 2.48, the number of poultry is 10.4), and the non-agricultural income (non-agricultural income = number of labor ∗ 18,864.4 ∗ 1 + employment opportunities ∗ 1000 ∗ 1 − agricultural output value ∗ 1.25 ∗ 1). That is to say, under the basic scenario, farmers will choose the relevant strategies in Scenario 30 (part-time farmers, planting grain crops and cash crops, breeding livestock and poultry).

(2) Simulation scenario. According to different modes of livelihood and land use, the change of the coupling system is discussed and the change of the factors of the regulation factors is discussed. For the convenience of comparison, this study only adjusts some parameters in the basic scenario (Scenario 30) and does not discuss the impact of planting scale and breeding scale on farmers temporarily. The relevant parameters are set as follows: In terms of livelihood mode, compared to part-time farmers, the non-agricultural income of pure-agriculture farmers is 0, and the actual cultivated land area of non-agriculture farmers is 0. In terms of mode of land use, compared to both planting and breeding farmers, the number of breeding farmers is 0, and the actual cultivated area of breeding farmers is 0. In terms of the mode of planting, compared to farmers who plant grain crops and cash crops, the proportion of grain crops from farmers who plant grain crops is 100%, the proportion of cash crops is 0, the proportion of grain crops from farmers who plant cash crops is 0, and the proportion of cash crops of farmers who plant cash crops is 100%. In terms of breeding mode, compared to farmers who breed livestock and poultry, the number of livestock from farmers who breed livestock is 4.96, the number of poultry is 0, the number of livestock from farmers who breed poultry is 0, and the number of poultry is 20.8.

4.2. Basic Scenario Analysis

To explore the development trend of the farmer’s livelihood and land use, this study simulated the coupling system of the farmer’s livelihood efficiency and land use. Under the existing variable relations and parameter values, the livelihood capital, livelihood output, and land use level of farmers are shown in Figure 6.

It can be seen from Figure 6a that with the development of time, the livelihood capital of farmers will decline first and then rise. Among them, human capital, physical capital, financial capital, social capital, and information capital will rise, while natural capital will first decrease and then rise. With the development of a social economy, the income level, education level, daily supplies, and information tools of farmers are also gradually increasing; their human capital and physical capital are also showing an increasing trend. Compared with other capital, natural capital has some differences. The reason is that with the acceleration of urbanization, the problem of rural land waste is becoming increasingly obvious. In the initial stage, the decrease in farmers’ cultivated land is significantly higher than the increase in cultivated land. In the later stage, the decrease in the cultivated land of farmers gradually slows down. Moreover, with the deepening of farmers’ attachment to the countryside, the increase in farmers’ cultivated land is gradually close to the decrease in cultivated land, and the housing area will show an upward trend with the increase in population, which makes the overall natural capital of farmers show a downward trend first and then a slow upward trend.

It can be seen from Figure 6b that with time development, the livelihood output of farmers will first decline and then rise. Among them, the income level, education, and medical level, and employment opportunities will increase, and rural attachment and awareness of ecological protection will first decline and then rise. The reason is that with the development of society, farmers have increasingly increased access to information, which expands the scope of employment of farmers and brings more employment opportunities and income to farmers. Moreover, with the increase in income and information capital, farmers have a deeper understanding of education and medicine and are willing and able to invest more funds to improve the level of education and medicine. The reason why rural attachment and awareness of ecological protection are declining first and then rising is that, with the gradual increase of the urban-rural income gap, farmers are more willing to work unexpectedly to seek higher income. In this process, a large amount of land has been transferred and abandoned, and farmers’ attachment to their hometown and ecological environment protection awareness has declined to some extent. As time passes, the income of farmers will be used to purchase production and living materials, which ties the relationship between farmers and their hometowns closer. Moreover, with the publicity of various channels, farmers’ awareness of ecological environmental protection has gradually increased and they are more and more willing to contribute to the development of their hometown.

It can be seen from Figure 6c that with time development, the level of land use of farmers will first decline and then increase. Among them, the land use structure shows a downward trend, and the land use intensity and land use benefit will first decline and then rise. The reason is that with an increasing population and decreasing cultivated land area, decreasing per capita cultivated land area has reduced the structure and scale of land use of farmers. Affected by the actual area of cultivated land, the trend of land input decreasing first and then increasing makes farmers’ land use intensity and land use benefit have the same development trend. Furthermore, with an increase in the labor force, the intensity of land use of farmers in the late growth rate is relatively high. However, if more labor is invested in less land, it will not bring more benefits to farmers and will cause a certain waste of resources.

4.3. Comparison and Analysis of Different Scenarios

For the above 31 scenarios, we take the average value of farmers’ livelihood and land use in each scenario in the simulation period and conduct standardization. Then the DEA model is used to measure farmers’ livelihood efficiency, and the coupling coordination degree model is utilized to measure the coupling coordination relationship between livelihood efficiency and land use (

Table 6. Farmers’ livelihood and land use under different scenarios.

Scenario	Livelihood Capital	Livelihood Output	Livelihood Efficiency	Land Use Structure	Land Use Intensity	Land Use Benefit	Land Use Level	Coupling Degree	Development Degree	Coordination Degree
1	0.196	0.066	0.267	0.515	0.600	0.201	0.375	0.986	0.321	0.563
2	0.198	0.068	0.272	0.515	0.637	0.235	0.397	0.982	0.335	0.573
3	0.197	0.067	0.269	0.515	0.619	0.218	0.386	0.984	0.328	0.568
4	0.158	0.100	0.501	0.458	0.502	0.749	0.607	0.995	0.554	0.742
5	0.025	0.000	0.000	0.128	0.244	0.120	0.142	0.000	0.071	0.000
6	0.091	0.050	0.435	0.293	0.373	0.434	0.374	0.997	0.405	0.635
7	0.368	0.190	0.409	0.981	0.939	0.954	0.961	0.915	0.685	0.792
8	0.232	0.089	0.304	0.645	0.674	0.322	0.492	0.972	0.398	0.622
9	0.300	0.139	0.367	0.813	0.806	0.638	0.727	0.944	0.547	0.719
10	0.371	0.192	0.410	0.981	0.977	0.988	0.984	0.911	0.697	0.797
11	0.235	0.091	0.307	0.645	0.711	0.355	0.514	0.968	0.411	0.630
12	0.303	0.142	0.371	0.813	0.844	0.672	0.749	0.941	0.560	0.726
13	0.301	0.141	0.371	0.813	0.825	0.655	0.738	0.944	0.554	0.723
14	0.234	0.090	0.305	0.645	0.693	0.338	0.503	0.969	0.404	0.626
15	0.274	0.120	0.347	0.812	0.823	0.654	0.737	0.933	0.542	0.711
16	0.896	0.951	0.841	0.538	0.625	0.212	0.393	0.932	0.617	0.758
17	0.893	0.945	0.838	0.538	0.664	0.248	0.416	0.942	0.627	0.768
18	0.894	0.948	0.840	0.538	0.644	0.230	0.404	0.937	0.622	0.763
19	0.782	0.895	0.907	0.457	0.502	0.749	0.606	0.980	0.757	0.861
20	0.718	0.883	0.974	0.128	0.244	0.120	0.142	0.666	0.558	0.610
21	0.750	0.889	0.939	0.293	0.373	0.434	0.374	0.903	0.657	0.770
22	0.985	0.969	0.779	1.000	0.960	0.964	0.976	0.994	0.878	0.934
23	0.919	0.957	0.825	0.667	0.698	0.333	0.509	0.972	0.667	0.805
24	0.952	0.963	0.801	0.834	0.829	0.648	0.743	0.999	0.772	0.878
25	0.981	0.963	0.778	1.000	1.000	1.000	1.000	0.992	0.889	0.939
26	0.916	0.951	0.823	0.667	0.738	0.368	0.532	0.977	0.678	0.814
27	0.949	0.957	0.799	0.833	0.869	0.684	0.766	1.000	0.783	0.885
28	0.983	0.966	0.779	1.000	0.980	0.982	0.988	0.993	0.884	0.937
29	0.918	0.954	0.823	0.667	0.718	0.350	0.521	0.974	0.672	0.809
30	0.950	0.960	0.801	0.834	0.849	0.666	0.754	1.000	0.778	0.882
31	0.694	0.876	1.000	0.000	0.000	0.000	0.000	0.000	0.500	0.000

). By comparing 31 scenarios, we can obtain high- and low-level scenarios for livelihood capital, livelihood output, livelihood efficiency, and other indicators (

Table 7. High level scenario and low level scenario of each index.

Index	High Level Scenarios	Low Level Scenarios
Livelihood capital	22, 28, 25	5, 6, 4
Livelihood output	22, 28, 25	5, 6, 1
Livelihood efficiency	31, 20, 21	5, 1, 3
Land use structure	22, 28, 25	31, 20, 5
Land use intensity	25, 28, 10	31, 5, 20
Land use benefit	25, 10, 28	31, 20, 5
Land use level	25, 28, 10	31, 20, 5
Coupling degree	27, 30, 24	5, 31, 20
Development degree	25, 28, 22	5, 1, 3
Coordination degree	25, 28, 22	5, 31, 1

).

4.3.1. Livelihood Efficiency under Different Scenarios

Table 6 and Table 7 show that when part-time farmers plant grain crops and breed livestock (Scenario 22), the livelihood capital and livelihood output are the highest. When non-agriculture farmers neither plant nor breed (Scenario 31), their livelihood efficiency is the highest. When pure-agriculture farmers do not plant and breed poultry (Scenario 5), the livelihood capital, income, and livelihood efficiency are the lowest. Comparing the livelihood efficiency of farmers under different scenarios, we can comprehend the following rules: (1) Livelihood efficiency of farmers is related to their livelihood mode, and there are the following rules under different livelihood modes: pure-agriculture farmers < part-time farmers < non-agriculture farmers. That is to say, the livelihood efficiency of non-agriculture farmers is the highest, while that of pure-agriculture farmers is the lowest. (2) Livelihood efficiency of farmers is related to their land use mode. Among them, the livelihood efficiency of pure-agricultural farmers when both planting and breeding is higher than that when planting, that is, the livelihood efficiency of scenarios 7–15 is higher than that of scenarios 1–3. The livelihood efficiency of part-time farmers has the following rules under different modes of land use: both planting and breeding < planting < breeding. That is, in scenarios 16–30, the livelihood efficiency of scenarios 19–21 is relatively high, while that of scenarios 22–30 is relatively low. (3) Livelihood efficiency of farmers is related to their planting mode. When the planting modes are the same, the livelihood efficiency of pure-agriculture farmers has the following rules under different planting modes: planting grain crops < planting grain crops and cash crops < planting cash crops. The livelihood efficiency of part-time farmers has the following rules under different planting modes: planting cash crops < planting grain crops and cash crops < planting grain crops. That is, increasing the proportion of grain crops planted can improve the livelihood efficiency of pure-agriculture farmers but will reduce the livelihood efficiency of part-time farmers. (4) Livelihood efficiency of farmers is related to their breeding mode. When the planting modes are the same, the livelihood efficiency of pure-agriculture farmers has the following rules under different breeding modes: breeding poultry < breeding livestock and poultry < breeding livestock. The livelihood efficiency of part-time farmers has the following rules under different breeding modes: breeding livestock < breeding livestock and poultry < breeding poultry. That is, increasing the breeding proportion of livestock can improve the livelihood efficiency of pure-agriculture farmers but will reduce the livelihood efficiency of part-time farmers. Therefore, in order to improve livelihood efficiency, farmers should not only be encouraged to participate in non-agricultural activities but also be encouraged to flexibly adjust planting and breeding modes according to their own livelihood modes.

4.3.2. Land Use Level under Different Scenarios

Table 6 and Table 7 further show that when non-agriculture farmers neither plant nor breed (Scenario 31), their land use structure, land use intensity, land use benefit and land use level are the lowest, and all are 0. When part-time farmers plant grain crops and raise livestock (Scenario 22), the structure of land use is the highest. When part-time farmers plant cash crops and breed livestock (Scenario 25), the land use intensity, land use benefit and land use level are the highest. When comparing the land use level of farmers under different scenarios, we can see the following rules. (1) The level of land use of farmers is not significantly related to their mode of livelihood. For example, although Scenario 25 and Scenario 20 are both part-time farmers, their land use levels differ greatly. (2) The level of land use of farmers is related to their land use mode. Compared to farmers-only planting, a farmers’ land use level is higher when both planting and breeding. For example, the land use level of scenarios 7–15 and 22–30 is higher than that of scenarios 1–3 and 16–18. (3) The level of land use of farmers is related to their planting mode. When the livelihood mode and the breeding mode are the same, the farmer’s land use level has the following rules under different planting modes: planting grain crops < planting grain crops and cash crops < planting cash crops. In other words, when farmers plant grain crops, their level of land use is relatively low. (4) The level of land use of farmers is related to their mode of breeding. When the livelihood mode and the planting mode are the same, the level of land use level has the following rules under different breeding modes: breeding poultry < breeding livestock and breeding poultry < breeding livestock. In other words, when farmers breed poultry, their level of land use is relatively low. Therefore, to improve the level of land use, farmers can be encouraged to plant and breed at the same time, and the planting scale of cash crops and the breeding scale of livestock can be appropriately increased.

4.3.3. Coupling Coordination Relationship under Different Scenarios

It can be seen from Table 6 and Table 7 that when pure-agriculture farmers do not plant and breed poultry (Scenario 5), their coupling degree, development degree, and coordination degree are the lowest. The coupling degree is the highest when part-time farmers plant cash crops and breed livestock and poultry (Scenario 27). When part-time farmers plant cash crops and breed livestock (Scenario 25), their development degree and coordination degree are the highest. By comparing the coordination degree under different scenarios, it can be seen that the coordination degree is related to their livelihood mode and land use mode. (1) When the land use mode (planting mode, breeding mode) is the same, the coordination degree of part-time farmers is greater than that of pure-agriculture farmers. For example, the coordination degree of Scenario 16 is greater than that of Scenario 1, and the coordination degree of Scenario 30 is greater than that of Scenario 15. (2) When the livelihood mode and the breeding mode are the same, the coordination degree of farmers has the following rules under different planting modes: planting grain crops < planting grain crops and cash crops < planting cash crops. For example, in Scenarios 1, 2, and 3, the coordination degree of Scenario 2 is the largest, followed by Scenario 3, and Scenario 1 is the smallest. (3) When the livelihood mode and planting mode are the same, the coordination degree of farmers has the following rules under different breeding modes: breeding poultry < breeding livestock and poultry < breeding livestock. For example, in scenarios 4, 5, and 6, the coordination degree of scenario 4 is the largest, followed by scenario 6, and scenario 5 is the smallest. Therefore, to improve the coupling coordination relationship between farmers’ livelihood efficiency and land use, it is not only necessary to optimize and innovate their livelihood modes and land use modes to improve their livelihood efficiency and land use level, but also to optimize the allocation of resources from a system perspective to achieve the coordinated development of the livelihood system and the land system.

5. Discussion

The human–land relationship is a pair of basic relationships born with the evolution of human development and a key problem facing today’s social development. This study takes the coupling system of farmers’ livelihood efficiency and land use as the starting point to explore the coupling coordination relationship between farmers’ efficiency and land use under different scenarios, which not only can promote the harmonious development of the human–land relationship, but also provide a reference for analyzing and predicting the future human–land relationship.

(1) Using the comprehensive research method of “theoretical induction, model analysis, simulation, empirical test” to carry out integrated research is an important direction of future human-land system research. The human-land system coupling model is a key tool for the study of human-land systems, and an important means to objectively understand the human-land relationship and provide decision making services [80]. Based on the micro-subject of farmers, this study analyzes the hierarchical structure of the coupling system of farmers’ livelihood efficiency and land use from different perspectives such as “system, structure, element, index”. The system dynamics method is used to simulate and analyze the changes of farmers’ livelihood and land use under different scenarios, which can not only promote and guide the coupling coordinated development of farmers’ livelihood and land use, but also carry out targeted prediction and research on the state and core relationship of the future farmers’ scale human-land system. In the process of research, this study objectively and scientifically simulated the coupling system of farmers’ livelihood efficiency and land use as far as possible, and the conclusions that pure agricultural farmers have low livelihood efficiency and part-time farmers have high coordination degree basically conform to our practical cognition. However, the semi-openness of the human-land system, the development of the individual situation of farmers, and the diversity of resources and environment determine that the specific values obtained from the simulation cannot be absolutely accurate, but the trend of changes in the human-land relationship obtained from the research, that is, the farmers’ livelihood capital, livelihood output, and land use level will first decline and then rise over time, which has great reference value for us to recognize the human-land system and regulate human-land relationship.

(2) It is an important content of future research on human–land systems to explore the coupling mechanism of human–land systems at different scales, such as individual, regional and global, at a deeper level. The coupling coordination relationship will be affected by many factors. In this study, we analyzed the livelihood and land use of farmers under general planting and breeding conditions and found that increasing the proportion of cash crops and livestock can effectively improve the level of coupling coordination between farmers’ livelihood efficiency and land use. Due to the direction and length of the research, this study did not conduct an in-depth analysis of farmers from the perspective of the planting scale and the breeding scale. However, in the actual investigation process, we found that when pure-agriculture farmers are planting and breeding on a large scale, their livelihood efficiency and land use level may be higher. Therefore, planting scale and breeding scale are also the key issues to be considered in the discussion of farmers’ livelihood and land use [81]. In addition, the coupling coordination relationship between farmers’ livelihood efficiency and land use is not only related to their own way of livelihood, planting and breeding but also affected by many factors such as regional development level, resource endowment, geographical location, infrastructure level and agricultural policies [82,83,84]. Therefore, in order to truly realize the coordinated development of the human–land system, more efforts should be made not only to optimize the way of livelihood and land use in the future, such as enriching and expanding non-agricultural activities, and optimizing the structure and type of planting and breeding but also to further explore the interaction between livelihood and land use on more scales, such as farmers, villages, counties. By building an integrated and cross-scale livelihood efficiency land use composite system, we hope to provide farmers, villages, counties, and other different subjects with a diversified and multi-classified comprehensive optimization measure system.

(3) The cross-integration of multidisciplinary, multi-fields, and multi-methods can provide important support for deepening the research of human–land systems. With the development of informatization and globalization, the physical boundary between people has been broken, communication between people has become more convenient and closer, and the human–land relationship has been endowed with a new connotation. If we want to deeply understand the human–land system, we need to re-understand the true meaning of the human–land relationship and its various elements in combination with the characteristics of the times. For example, we can combine the new concepts and models of the global and regional human–land system to explore the new human–land relationship in the context of urban-rural integration, COVID-19, ecological civilization construction, targeted poverty alleviation, rural revitalization, climate change, and food security. We can build a long-term cooperative mechanism involving multiple regions, departments, and disciplines to promote the coordinated and high-quality development of the human–land system in different types and regional scenarios. The modern human–land system is complex, involving nature, economy, and technology. Using complex system thinking and methods to analyze the human–land relationship, study human–land systems and explore human–land coupling coordination is not only conducive to the in-depth exploration of the evolution law and dynamic mechanism of human–land relationships but also conducive to promoting the coordinated development of human–land system and promoting regional sustainable development. Regarding the research methods of the human–land system, previous scholars mainly constructed the indicator system from the perspective of society, economy, and resources and conducted a comprehensive evaluation of the human–land system [22,85]. In addition, some scholars explored the human–land system with the help of the system dynamics model and the coupling coordination model [40,51,70,86]. In the future, the human–land system research needs a breakthrough in technology and methods. We can make full use of earth exploration, modern remote sensing, unmanned aerial vehicles, artificial intelligence, virtual reality, 5G network and other technologies and methods to create a systematic human–land network system, monitor the evolution data of the man earth system in a long-term and dynamic manner, and provide methods and technical support for realizing the real-time cognition and coordination of the human–land system [48,87,88,89].

6. Conclusions

In this study, a stratified random sampling method was adopted to conduct a field survey of farmers in southern Shaanxi, and relevant statistical data were referred to. Based on the relevant data, we built a system dynamics model to simulate the different modes of livelihood and modes and land use of farmers. Through comparative analysis of farmers’ livelihood capital, livelihood output, livelihood efficiency and land use level under different scenarios, the direction can be pointed out for improving the coupling coordination relationship. Relevant conclusions are as follows.

(1) According to the system inertia development, the livelihood capital, livelihood output, and land use level of farmers will first decrease and then increase. Specifically, with the development of time, human capital, physical capital, financial capital, social capital, information capital, income level, education and medical level, and employment opportunities will rise. This is because with economic development and technological progress, the income level, living materials and information tools of farmers are gradually increasing, and the education services and medical services provided by the state and society will continue to increase. Accordingly, the human capital and physical capital of farmers will also show a growth trend. Natural capital, rural attachment, ecological protection awareness, land use intensity, and land use benefit will decline first and then rise. Moreover, land use structure will decline due to the fact that under the condition of a certain family land area, the number of family population will increase over time, but the per capita planting area and the breeding area will continue to decrease.

(2) The efficiency of the livelihood of farmers is related to their mode of livelihood and their mode of land use mode. On the one hand, the livelihood efficiency of farmers will be affected by their livelihood modes, and the following rules will appear under different livelihood modes: pure-agriculture farmers < part-time farmers < non-agriculture farmers. This is because engaging in non-agricultural activities can bring more income to farmers, and the increase in income will further improve people’s education level, medical level, and employment opportunities. At present, more non-agriculture farmers live in cities, where there is better education, medical conditions, better employment opportunities, and higher livelihood efficiency. Pure-agriculture farmers live mainly in rural areas, with low agricultural income and poor infrastructure, which makes their livelihood efficiency low. On the other hand, land use modes will have different impacts on the livelihood efficiency of pure-agriculture farmers and part-time farmers. When the proportion of cash crop planting and livestock breeding is increased, the livelihood efficiency of pure-agriculture farmers will be improved, while that of part-time farmers will be reduced. This is because farmers need to invest more time costs to increase the proportion of cash crops and livestock breeding. For pure-agriculture farmers, this will further increase their income and improve their livelihood efficiency. For part-time farmers, the increase in agricultural activity time results in a decrease in non-agricultural activities, which will reduce their income level, employment opportunities and livelihood efficiency.

(3) The level of land use of farmers is mainly related to their planting mode and breeding mode. The land use level farmers have the following rules under different planting and breeding modes: planting grain crops < planting grain crops, and cash crops < planting cash crops, breeding poultry < breeding livestock, and poultry < breeding livestock. When farmers plant cash crops and breed livestock at the same time, their level of land use is high. When farmers do not plant and only breed poultry or breed only planting grain crops, their level of land use is low. The reason is that, compared to food crops, the output and output value of cash crops are larger. Moreover, planting cash crops can bring more benefits to farmers and improve their land use level. Compared with breeding poultry, the cost of capital, labor, and land invested in breeding livestock is higher, and the corresponding output and output value are also higher. The higher land use structure, land use degree, and land use efficiency make the land use level of breeding livestock higher.

(4) The coordination degree is related to their livelihood mode and land use mode. For farmers with the same land use mode, the coordination degree of part-time farmers is greater than that of pure-agriculture farmers. The reason is that part-time farmers have a higher income and more employment opportunities than pure farmers and can allocate resources more flexibly and improve livelihood efficiency and coupling coordination. For farmers with the same livelihood mode and breeding mode, the coordination degree is the smallest when planting grain crops, and the coordination degree is the largest when planting cash crops. For farmers with the same livelihood mode and planting mode, the coordination degree is the smallest when breeding poultry, and the coordination degree is the largest when breeding livestock. The reason is that planting cash crops and breeding livestock can improve the income level, livelihood efficiency and land use level of farmers, and promote the coupling coordinated development of livelihood efficiency and land use.