Influence of Livelihood Capitals on Landscape Service Cognition and Behavioral Intentions in Rural Heritage Sites

Abstract

Farmers’ livelihoods are critical for global sustainable development and the conservation and transmission of rural heritage. However, neglecting farmers’ livelihoods increases the risks to living heritage conservation. Therefore, it is essential to explore the impact of livelihood capital on farmers’ landscape services cognition and their behavioral intentions. Based on the Sustainable Livelihoods Approach, this study examines the rural areas of Mulanbei irrigation district, which was recognized as a World Heritage Irrigation Structure in 2014, and uses a structural equation model to investigate the relationships between farmers’ livelihood capital, landscape services cognition and behavioral intentions. The study’s key findings include the following: (1) The levels of human capital (0.541), social capital (0.671), and cultural capital (0.645) are relatively high, while the levels of natural, physical, and financial capital are comparatively low. (2) There are significant differences in landscape service cognition and behavioral intentions among farmers of different livelihood strategies, with diversified livelihood farmers demonstrating the highest levels of both cognition and intentions, while subsidy-dependent farmers show the lowest levels. (3) Natural, cultural and financial capital play a crucial role in influencing farmers’ landscape services cognition and their behavioral intentions. Landscape service cognition mediates the relationship between livelihood capital and behavioral intentions. (4) To increase farmers’ willingness to protect and promote rural heritage, efforts should focus on enhancing natural, physical, and financial capital while fostering cultural capital to promote advocacy.

1. Introduction

A livelihood comprises the assets (including both material and social resources), capabilities, and activities that an individual or household possesses for making a living and improving long-term well-being [1]. Farmers’ livelihood conditions directly affect their quality of life and long-term welfare. Rural heritage is closely related to farmers’ settlement behaviors and their agricultural and pastoral activities [2], reflecting the positive interaction between farmers’ livelihoods and rural resource management. It serves as a valuable model for sustainable human–environment relationships. Recently, due to the influences of urbanization, industrialization, and commercialization, farmers’ income sources have become more diversified, resulting in changes in household income, employment prospects, and educational and training opportunities. Simultaneously, farmers’ perceptions and attitudes toward traditional resource management practices have undergone significant shifts. On the one hand, consensus holds that rural heritage contributes to diversified livelihoods [3,4], promoting the development of part-time farming and increasing income levels [5,6,7]. On the other hand, the single-directional transformation of farmers’ livelihoods in the context of heritage tourism may reduce livelihood resilience and stability and even damage the authenticity and vitality of heritage due to the excessive pursuit of economic benefits [8]. Nevertheless, the long historical evolution of rural heritage has deeply ingrained a sense of connection between local farmers and their heritage. However, in developing countries, government authorities often lead heritage conservation efforts. When the responsibility for heritage protection and utilization is divided, the intended benefits for local farmers may not be fully realized, and farmers may not recognize the social and economic advantages. Recent studies emphasize that farmers’ active involvement is crucial for effective heritage conservation, as they are the smallest decision-making units in these efforts [9,10]. Furthermore, farmers’ behavioral intentions are shaped by their livelihood characteristics, with a tendency to optimize their livelihood capital [11,12]. Therefore, understanding the challenges farmers face in selecting livelihood strategies is essential for achieving the dual goals of heritage conservation and community development.

Sustainable livelihoods can cope with and recover from stress and shocks, maintain or enhance its capabilities and assets, and provide sustainable livelihood opportunities for the next generation while contributing net benefits to other livelihoods at the local and global levels and in the short and long term [1,13]. In 1997, the UK Department for International Development (DFID) issued a white paper, ‘Eliminating Global Poverty: The Challenge for the 21st Century’, which explicitly proposed promoting human development and environmental protection by creating sustainable livelihoods for the poor. Since then, extensive theoretical and practical exploration has emerged concerning rural livelihoods [14], such as sustainable livelihood analysis [15], and livelihoods in tourism destinations [16]. These efforts have led to the development of livelihood analysis frameworks that emphasize the interrelationships such as “livelihood context–livelihood capital–institutions and processes–livelihood strategies–livelihood outcomes.” Among these theoretical frameworks, the Sustainable Livelihoods Approach (SLA) developed by the DFID has been widely adopted, dividing livelihood capital into five fundamental types: human, natural, physical, financial, and social [15]. This approach was later expanded to include cultural and tourism capital, allowing for a more comprehensive characterization of household features and the diagnosis of livelihood development bottlenecks. Sustainable livelihoods have emerged as an important approach for addressing gaps in traditional ecosystem services and resource management, particularly in terms of recognizing social choices and stakeholder perspectives [17]. This approach focuses on exploring win–win solutions that optimize landscape functions and improve well-being by considering the diverse needs of farmers and their livelihoods [18].The SLA has been acknowledged as a vital tool for examining rural issues of poverty alleviation and development, effectively linking environmental characteristics, social demands, and individual decision-making. This not only helps us to understand how micro-level forces shape landscape evolution [19,20] but also ensures that landscape decision-making is aligned with the interests of relevant stakeholders [21,22], providing a theoretical foundation for investigating the relationship between rural heritage landscape provision and farmers’ behavioral decisions.

World Heritage Irrigation Structures (WHISs) represent a prime example of rural productive landscape heritage. Over time, irrigation districts have evolved into an agricultural-based landscape structure and a rural society centered on water management, making significant contributions to increasing food production, improving farmers’ livelihoods, and promoting rural prosperity. These WHISs, consisting of heritage sites and their surroundings, are always vast and complex. Their extensive surface canal systems and continued functioning heavily depend on the spontaneous maintenance efforts of local farmers. Despite the strong attachment to and sense of identity that farmers have with these heritage sites, livelihood pressures often lead to behaviors conflicting with the goals of sustainable heritage development. Examples include encroaching on water bodies for expanding residential and industrial areas, overusing agricultural fertilizers, or opting out of local cultural activities. Previous studies have examined how external factors such as individual characteristics, household structures, income levels, social networks, and agricultural policies influence farmers’ behavioral intentions [23,24,25].

As the autonomous initiative of farmers continues to strengthen, research on internal cognitive motivations and the dynamic processes driving behavior change has developed progressively. Li et al. examined how farmers’ perceptions impact their willingness to engage in green agricultural practices, identifying specific behavioral factors [26,27]. Building upon this, Wang et al. investigated the mediating role of livelihood capital and value perception in influencing pro-environmental behavior, adding depth to the understanding of these motivations [28]. More recently, Lyu et al. expanded this work by exploring the relationship between farmers’ cognition, intentions, and behaviors in the context of sustainable land intensification in Shandong Province, offering a comprehensive view of these dynamics [23,29]. Despite these advances, there is still limited research on how farmers perceive the value of landscape services and how this perception influences their engagement and acceptance of conservation measures. This might lead to a mismatch between policy recommendations and the actual needs and expectations of farmers. As research on rural heritage conservation increasingly shifts from a purely physical focus to exploring the interaction between farmers’ cognition and their behavioral intentions [4,30], an important question arises: how can enhancing farmers’ awareness of landscape services guide their behavioral intentions to achieve optimal livelihood outcomes? This question has become a central focus in both academic and practical discussions.

This study aims to examine farmers’ cognition of the landscape services provided by WHISs and their associated behavioral intentions. It seeks to uncover the key factors influencing farmers’ behavioral intentions and explore how these intentions differ among farmers with various livelihood strategies. The results will offer valuable insights for developing tailored policies that balance heritage conservation and utilization. Specifically, this study addresses three main questions:

(1)

How do farmers with different livelihood strategies differ in their cognition of landscape services and behavioral intentions?

(2)

What are the specific pathways through which different types of livelihood capital influence landscape service cognition and behavioral intentions?

(3)

Does landscape service cognition mediate the relationship between livelihood capital and farmers’ behavioral intentions?

2. Research Framework and Research Hypotheses

2.1. Research Framework

The core idea of livelihood studies is to affirm that small-scale farmers possess the wisdom and potential to make effective behavioral decisions. Livelihood capital serves as a key foundation for these decisions, providing a multidimensional representation of farmers’ assets and capabilities. For instance, the accumulation of human capital may enable farmers to participate more actively in conservation and promotional activities, while abundant natural capital can enhance their awareness of environmental protection. From the perspective of the SLA, farmers can adopt livelihood strategies in response to changes in the external environment, which subsequently influence both the characteristics and the condition of their livelihood capital [31]. However, recent studies suggest that farmers’ livelihood strategies are not always entirely rational, especially in the context of rural tourism development. Changes in livelihood strategies may lead to disturbances in the socio-ecological system, resulting in short-term increases in livelihood capital but heightened vulnerability [32]. Moreover, research has found that when faced with decisions regarding resource development and conservation, farmers tend to opt for familiar approaches rather than experimenting with innovative methods or new policy tools [33]. This cognitive bias can affect farmers’ long-term judgments about the value of ecosystems, leading to irrational decisions.

Cognition and Behavior Theory (CBT) posits that cognitive processes play a crucial role in behavioral modification, emphasizing that behavioral decisions stem from an individual’s behavioral intentions (BI). BI, in turn, is shaped by cognitive perceptions formed during the process of receiving and evaluating information [34]. Numerous studies have demonstrated that CBT significantly enhances the explanatory power of behavioral intentions research, particularly in fields such as health psychology and social behavior [35]. Its applicability has also been increasingly recognized in environmental behavior studies [36]. Some scholars have applied CBT to explore farmers’ behavioral intentions, finding that farmers’ attachment to their homeland and sense of belonging can increase their willingness to participate in rural environmental protection [37,38,39]. However, its connection with farmers’ specific livelihood needs remains under explored. Therefore, this study integrates CBT with SLA to comprehensively analyze the mechanisms underlying farmers’ behavioral intentions.

Specifically, rural heritage, as a form of living heritage, relies on the important and continuous landscape services it provides to local farmers. Over centuries, WHISs have offered key landscape services to farmers in the irrigation districts, such as storm-water management, climate regulation, food supply, and biodiversity conservation. With socio-economic development and changes in resource utilization, the heritage’s primary functions have gradually shifted from being predominantly ecological and productive to a combination of ecological and social functions [40]. Farmers’ behavioral intentions are generally aligned with the goal of increasing livelihood capital and harmonizing with livelihood strategies. At the same time, their perception of the landscape services provided by heritage also influences their behavioral intentions. Some farmers may even be willing to sacrifice personal interests to contribute to the collective goals of the local community. This perception is not only linked to the objective landscape services provided by the heritage but is also influenced by farmers’ specific livelihood strategies.

To capture these dynamics, this study incorporates CBT into the SLA, constructing an analytical model that encompasses livelihood capital, landscape service cognition, and behavioral intentions. As shown in Figure 1, livelihood capital and farmers’ cognition directly influence behavioral intentions and play a role in long-term livelihood strategies. Additionally, considering that rural heritage sites are typically shaped by unique cultural characteristics and socio-ecological relationships [41], this study introduces “cultural capital” as a distinct component of livelihood capital, alongside the traditional five types of livelihood capital.

2.2. Research Hypotheses

Farmers’ willingness to engage in rural heritage protection and utilization reflects their attitudes and decision-making processes regarding environmental and resource management. Based on the proposed research framework, the relationships among farmers’ livelihood capital, landscape service cognition, and their willingness to engage in heritage protection and utilization behaviors are articulated as follows:

Hypothesis 1.

Different types of livelihood capital have a significant positive impact on the willingness to participate, conserve and promote.

Hypothesis 2.

Different types of livelihood capital have a significant positive impact on landscape service cognition.

Hypothesis 3.

Landscape service cognition has a significant positive impact on the willingness to participate, to conserve and to promote.

Hypothesis 4.

Landscape service cognition mediates the effects of different types of livelihood capital on willingness to participate, conserve, and promote.

3. Materials and Methods

3.1. Study Area

The Mulanbei Water Conservancy Project (MWCP), located in Putian, Fujian Province, China, started construction in 1064 BC and was completed in 1083 BC during the Northern Song Dynasty, with three rounds of site re-selections and reconstructions. Having been in operation for more than 930 years, it has largely preserved both the original position and form of its engineering structures from the historical period. It is recognized as the largest dam diversion project with saltwater rejection and freshwater conservation functions in ancient Fujian Province. In 2014, it was listed as one of the first World Heritage Irrigation Structures (WHISs). The main engineering heritages include three parts: the weir, the canal system, and the embankment project. The weir is primarily represented by the Mulanbei dam, which spans a length of 219.13 m. The total length of the canal system is 309.5 km, while the embankment project extends for 87.48 km. According to statistics from the International Commission on Irrigation and Drainage (ICID), the effective irrigated area covers 91.33 km² and benefits over 500,000 people. The MWCP is acknowledged as a unique plain granary and an important water conservation area in the southeast coast of China, as well as one of the regions experiencing the fastest coastal urbanization development [42].

In response to increasing urban development pressures and the lack of guidance for internal village construction, Putian City initiated four rounds of “Ecological Green Center Planning” from 2008 to 2023. Land use planning drawings can be found in Appendix A. This initiative aims to protect areas with a concentrated distribution of cultural and natural resources within the Mulanbei irrigation district. The master planning, confirmed in September 2023, delineates a core protected area of 65 km², encompassing 61 villages and 3 non-village areas (including one marine area and 2 state-owned or town-owned tidal flats). Among these, 10 villages possess distinctive protective value, 21 villages exhibit industrial characteristics, and 4 are recognized as provincial historical and cultural villages. Considering factors such as village location, industrial characteristics, landscape features, and land use, 7 villages with the highest protective value were selected as research samples. These include 3 villages with industrial characteristics—Daili Village (a Taobao village), Qibu Village (shoe-making industry), and Chenqiao Village (furniture industry); one provincial traditional village—Dongyang Village; one key cultural tourism development village—Wujiang Village; and 2 provincial water conservancy scenic spots—Jiangdong Village and Dongjia Village. This is shown in Figure 2.

3.2. Questionnaire Design

The questionnaire content encompasses four key aspects:

• Basic Socioeconomic Characteristics of the Respondents: This consisted of education level, household size, etc.
• Livelihood Capital Status of the Respondents: This aspect examines various types of capital, including human capital, natural capital, physical capital, financial capital, social capital, and cultural capital.
• Respondents’ Cognition of Landscape Services in the Area: Respondents are asked to evaluate their awareness and understanding of landscape services based on their own perceptions.
• Behavioral Willingness of the Respondents: This consisted of willingness to participate, willingness to protect, and willingness to promote.

Aspects 1 and 4 in the list above included respondents’ scores according to a Likert scale.

3.3. Variable Selection and Measurement

This study builds upon the research of Ren et al. [43] and, in conjunction with the framework established earlier, designs a measurement scale for farmers’ livelihood capital, landscape service cognition, and behavioral willingness. This scale encompasses three categories of variables:

• Independent Variable (IV): Farmers’ livelihood capital, which includes 12 indicators across six dimensions: human capital, natural capital, physical capital, financial capital, social capital, and cultural capital.
• Mediating Variable (MV): Farmers’ cognition of landscape services, comprising 11 measurement indicators within four dimensions: ecology, production, society, and landscape culture.
• Dependent Variable (DV): Farmers’ behavioral intentions, which include three dimensions: willingness to participate, willingness to protect, and willingness to publicize, totaling 11 items.

The measurement of livelihood capitals and the specific variable settings are presented in

Table 1. Indicators and quantification of livelihood capital.

Livelihood Capital	Indicator	Magnitude Definition
Human capital	Education level (A1)	Illiteracy = 1; Elementary school and below = 2; Junior high school = 3; High school/technical secondary school = 4; Junior college/higher vocational college and above = 5
	Proportion of household labor force (A2)	0~≤1/5 = 1; 1/5~≤2/5 = 2; 2/5~≤3/5 = 3; 3/5~≤4/5 = 4; 4/5~≤1 = 5
Natural capital	Household cropland area (A3)	0~≤1 mu = 1; 1~≤2 mu = 2; 2~≤3 mu = 3; 3~≤4 mu = 4; >4 mu = 5
	Water and fertilizer conditions of cropland (A4)	Very poor = 1; Poor = 2; Neutral = 3; Good = 4; Very good = 5
Physical capital	Types of household appliances (A5)	0 ~ 2 items = 1; 3 ~ 4 items = 2; 5 ~ 6 items = 3; 7 ~ 8 items = 4; More than 8 items = 5
	Types of residential housing (A6)	Old house = 1; Renovated old house = 2; Refurbished house = 3; Newly built house = 4
Financial capital	Household savings status (A7)	30,000~60,000 yuan = 1; 60,000 to 100,000 yuan = 2; 100,000 to 300,000 yuan = 3; 300,000 to 500,000 yuan = 4; More than 500,000 yuan = 5
	The difficulty of lending money to others (A8)	Very difficult = 1; Difficult = 2; Average = 3; Easy = 4; Very easy = 5
Social capital	The closeness with relatives and friends (A9)	Never interact = 1; Rarely interact = 2; Occasionally interact = 3; Fairly often interact = 4; Frequently interact = 5
	Willingness to participate in village activities (A10)	Never participate = 1; Occasionally participate passively = 2; Occasionally participate actively = 3; Frequently participate = 4; Lead participation = 5
Cultural capital	The level of understanding of folklore (A11)	Not familiar at all = 1; Slightly unfamiliar = 2; Neutral = 3; Somewhat familiar = 4; Very familiar = 5
	The degree of recognition of rural cultural values (A12)	Do not approve at all = 1; Somewhat disapprove = 2; Neutral = 3; Somewhat approve = 4; Strongly approve = 5

and

Table 2. Variable setup and assignment.

Variable Setup			Magnitude Definition
Landscape service cognition	Climate regulation	The climate here is comfortable and pleasant (B1)	1 = Strongly disagree 2 = Disagree 3 = Neutral 4 = Agree 5 = Strongly agree
	Storm-water management	There is no risk of flooding here (B2)
	Freshwater supply	The water quality here is excellent (B3)
	Habitat maintenance	There is a great diversity of species here (B4)
	Food supply	There is a large production of grains, vegetables, and fruits here (B5)
	Residential support	There are many people living here (B6)
	Employment security	There are many ways to earn money here (B7)
	Transportation	The transportation here is very convenient (B8)
	Cultural value	The culture here makes me feel proud (B9)
	Landscape aesthetics	The scenery here is beautiful (B10)
	Recreation and leisure	This place meets the needs for recreation and leisure (B11)
Behavioral Intentions	Participation intention	Support eco-agriculture and leisure tourism for rural heritage conservation. (C1)	1 = Strongly unwilling 2 = Unwilling 3 = Neutral 4 = Willing 5 = Strongly willing
		Invest time/funds in eco-agriculture. (C2)
		Invest time/funds in handicrafts and processing. (C3)
		Invest time/funds in tourism. (C4)
	Conservation intention	The overall willingness to protect the Mulanbei rural heritage. (C5)
		Actively engage in environmental protection. (C6)
		Accept industrial land transition subsidies to protect rural heritage. (C7)
		Embrace farmland transfer benefits for heritage. (C8)
		Rent and renovate idle homes for preservation. (C9)
		Donate to protect rural heritage. (C10)
	Promotion intention	Participate in local cultural promotions. (C11)
		Encourage others to preserve rural characteristics. (C12)

.

3.4. Data Analysis

Considering the arrangements of activities such as employment and farming, and to ensure the representativeness of the surveyed farmers, in this study, we communicated with village leaders, who recommended representative farming households to participate in the questionnaire survey. Random household surveys were then conducted along the main streets of the villages. Given the average number of households (approximately 900) in the sample villages—Chenqiao Village (955 households), Daili Village (356 households), Dongjia Village (1213 households), Dongyang Village (558 households), Jiangdong Village (1371 households), Qibu Village (1182 households), and Wujiang Village (705 households)—around 50 (5%) households were selected from each village for investigation. A total of 371 questionnaires were collected, with 350 valid responses, yielding an effective rate of 94.3%. The distribution of valid questionnaires was as follows: Chenqiao Village (45), Daili Village (59), Dongjia Village (43), Dongyang Village (33), Jiangdong Village (58), Qibu Village (67), and Wujiang Village (45).

Structural equation model (SEM) is a multivariate statistical method designed to analyze complex relationships among multiple indicators. It is suitable for three-dimensional and multi-level analyses, allowing for the quantification of causal relationships between various factors, making it ideal for examining intricate data structures [44]. In practice, the structural equation model has been widely applied in the fields of social sciences, psychology, and management, as well as in other research. In this study, the structural equation model analyzes the path coefficients of latent and observed variables, revealing the influence of farmers’ livelihood capital on their cognition of and behavioral intentions toward landscape services. The structural equation model consists of two components: (1) the measurement model, which reflects the relationship between observed and latent variables, and (2) the structural model, which describes the relationships among the latent variables. The equations are as follows.

$$\left\{\begin{array}{c}{Y}_i={\Lambda }_y{\eta }_i+{\epsilon }_i\\ X_i={\Lambda }_x{\xi }_i+{\delta }_i\end{array}\right.$$

(1)

$${\eta }_i={\rm B}{\eta }_i+\Gamma {\xi }_i+{\zeta }_i$$

(2)

In the equations, $Y_i$ is the endogenous manifest variables; $X_i$ is the exogenous manifest variables; ${\eta }_i$ represents the endogenous latent variables; ${\xi }_i$ is the exogenous latent variable; ${\Lambda }_y$ refers to the factor loading matrix for the cognition and behavioral intentions indicators; ${\Lambda }_x$ is the factor loading matrix for the livelihood capital indicators; ${\epsilon }_i$ and ${\delta }_i$ represent variable errors; ${\rm B}$ and $\Gamma$ are the coefficient matrices between the endogenous latent variable the exogenous latent variable; and ${\zeta }_i$ denotes the residuals.

4. Results

4.1. Descriptive and Analysis

As shown in Figure 3, 56.3% of respondents were male. The majority of respondents were aged 35 and above, accounting for 89.7%. The education level was generally low, with 72.6% having completed junior high school or below. Most of the respondents were engaged in farming, migrant work, or self-employment (including handicrafts, processing, and commerce), comprising 74.9% of the total.

Based on the classification of livelihood strategies in existing studies [14,45], farmers were categorized into five types: diversified livelihood, wage-operated, purely agricultural, labor-led, and subsidy-dependent. As depicted in

Table 3. Classification of farmer strategies.

Livelihood Strategies	Quantity (Household)	Proportion (%)	Living Mode
Diversified livelihood	39	11.1	Mainly to civil servants, public institutions (including village cadres)
Wage-operated	149	42.6	Enterprise employees, engaging in handicrafts, processing, tourism, commerce, etc.
Purely agricultural	82	23.4	Mainly farming
Labor-led	52	14.9	Mainly for workers
Subsidy-dependent	28	8	Other (mostly retired, unemployed)

, the wage-operated type is the most prevalent, while the traditional purely agricultural type still accounts for more than 20%. A smaller portion of households rely on stable public services or social security, falling into the diversified livelihood and subsidy-dependent categories.

Based on the dimensional normalization calculation of the livelihood capital of farmers with different livelihood strategies, the specific results are presented in Figure 4. In general, the relatively high average values of social capital and cultural capital among farmers indicate strong cohesion, resource access, and cultural identity. In contrast, the relatively low levels of natural capital, physical capital, and financial capital suggest weak physical and economic foundations, which may hinder agricultural productivity and economic development.

Farmers with different livelihood strategies show varied performance across different dimensions of livelihood capital. Compared with the overall average, diversified livelihood farmers perform well across all types of capital, with especially high levels of social capital and cultural capital, reflecting their balanced development capabilities. Wage-operated farmers have livelihood capital levels that are close to the overall average. Purely agricultural farmers excel in natural capital, social capital, and cultural capital. Labor-led farmers exhibit higher levels of human capital and financial capital but lower levels of social capital and cultural capital. However, subsidy-dependent farmers exhibit obviously lower levels of natural capital.

4.2. Differences in Landscape Service Cognition Among Farmers with Different Livelihood Strategies

The difference analysis in Figure 5 shows that the variation in cultural function cognition is relatively significant, while the differences in ecological function cognition are less pronounced. In terms of cultural landscape services, diversified livelihood farmers have a significantly higher recognition of local cultural values compared to wage-operated, subsidy-dependent, and labor-led farmers. Wage-operated farmers show lower awareness of landscape aesthetics and leisure and recreation functions compared to diversified livelihood and purely agricultural farmers. In the dimension of ecological landscape services, only wage-operated farmers exhibit a significantly lower cognition of climate regulation functions compared to diversified livelihood farmers, while the cognitive differences in other ecological functions (such as storm-water management and habitat maintenance) are not significant among farmers of different livelihood strategies. Overall, diversified livelihood farmers demonstrate a higher level of awareness across multiple dimensions of landscape services, particularly in cultural and ecological aspects, making them the group with the most comprehensive and in-depth cognition of landscape services. Additionally, although there are no significant differences in the cognition of transportation (social) and freshwater supply (ecological) among farmers of different livelihood strategies, the overall scores in these aspects are moderately high.

4.3. Differences in the Behavioral Intentions Among Farmers with Different Livelihood Strategies

As shown in Figure 6, farmers of different livelihood strategies exhibit significant differences across three dimensions: participation intention (C1–C4), conservation intention (C5–C10), and promotion intention (C11–C12). In terms of participation intention, diversified livelihood and purely agricultural farmers show the highest overall recognition and support for participating in ecological agriculture, handicrafts, processing industries, and tourism. Purely agricultural farmers, in particular, demonstrate a strong enthusiasm for participating in ecological agricultural production. Regarding protection intention, diversified livelihood farmers express a greater willingness to protect rural heritage sites and the environment. Specifically, their willingness to utilize idle resources and pay fees to protect heritage sites is significantly higher than that of other livelihood strategies. Lastly, for promotion intention, diversified livelihood farmers show a stronger commitment to promoting activities and educating others, reflecting their active involvement in rural protection and development.

Overall, diversified livelihood farmers exhibit higher levels of recognition and willingness across all aspects, indicating that they are more proactive in supporting rural protection and development. In contrast, subsidy-dependent farmers display lower enthusiasm for participation and protection but show a relatively higher willingness to engage in the promotion of local cultural traditions.

4.4. Results of Structural Equation Model Regression

The data were analyzed using the statistical software SPSS 27.0. The Cronbach’s α coefficient for each latent variable ranged from 0.686 to 0.826, the KMO value was not less than 0.5, and Bartlett’s spherical test reached a significance level of 0.01. The results indicate that the sample data have good reliability and validity. The specific results are shown in

Table 4. Reliability and validity testing of questionnaire across dimensions.

Dimension	Cronbach’ s α	KMO
Landscape service cognition	0.773	0.819
Participation intention	0.686	0.648
Conservation intention	0.815	0.780
Promotion intention	0.826	0.500

. Additionally, the measurement of livelihood capital in this study includes both tangible and intangible dimensions, using context-specific indicators that combine qualitative and quantitative data. Since these factors are not based on a typical Likert scale, traditional reliability tests are not applicable [9]. Instead, the indicators are adapted from well-established frameworks in the literature, carefully selected to align with the study’s context and theoretical foundation, ensuring they comprehensively capture the dimensions of livelihood capital [43].

The structural equation model was constructed using Amos 27.0 software, with parameter estimation performed through the maximum likelihood estimation method. As shown in

Table 5. Reliability and validity testing of questionnaire across dimensions.

Dimension Fit Index	Recommended Value	Fit Value
X2	The smaller the better	860.531
X2/df	<3.0	2.502
GFI	>0.9	0.853
AGFI	>0.8	0.814
RMSEA	<0.08	0.066

, the model was revised based on modification indices. The X²/df, AGFI, and RMSEA all met the required standards, while the GFI was close to the recommended value. Therefore, the overall model demonstrates a good fit.

As shown in

Table 6. Assuming results and path coefficient.

Hypothesis	Path Relationship	Estimated Value	S.E.	C.R	p-Value	Result
H1a	Human capital → Participation intention	0.090	0.630	1.423	0.155	Not accept
H1b	Human capital → Conservation intention	0.048	0.029	1.626	0.104	Not accept
H1c	Human capital → Promotion intention	0.018	0.057	0.321	0.749	Not accept
H1d	Natural capital → Participation intention	0.128	0.046	2.794	*	Accept
H1e	Natural capital → Conservation intention	0.056	0.022	2.593	**	Accept
H1f	Natural capital → Promotion intention	0.003	0.041	0.063	0.950	Not accept
H1g	Physical capital → Participation intention	−0.002	0.062	−0.033	0.973	Not accept
H1h	Physical capital → Conservation intention	−0.063	0.030	−2.112	*	Accept
H1i	Physical capital → Promotion intention	−0.110	0.057	−2.090	*	Accept
H1j	Financial capital → Participation intention	0.081	0.060	1.342	0.180	Not accept
H1k	Financial capital → Conservation intention	0.112	0.030	3.724	***	Accept
H1l	Financial capital → Promotion intention	0.113	0.055	2.069	*	Accept
H1m	Social capital → Participation intention	0.070	0.058	1.189	0.234	Not accept
H1n	Social capital → Conservation intention	−0.008	0.028	−0.276	0.783	Not accept
H1o	Social capital → Promotion intention	0.059	0.054	1.089	0.276	Not accept
H1p	Cultural capital → Participation intention	0.051	0.070	0.724	0.469	Not accept
H1q	Cultural capital → Conservation intention	−0.052	0.033	−1.546	0.122	Not accept
H1r	Cultural capital → Promotion intention	0.150	0.070	2.127	*	Accept
H2a	Human capital → Landscape service cognition	0.016	0.040	0.395	0.693	Not accept
H2b	Natural capital → Landscape service cognition	0.024	0.029	0.821	0.412	Not accept
H2c	Physical capital → Landscape service cognition	0.084	0.041	2.044	*	Accept
H2d	Financial capital→ Landscape service cognition	0.181	0.040	4.511	***	Accept
H2e	Social capital → Landscape service cognition	0.118	0.039	3.039	**	Accept
H2f	Cultural capital → Landscape service cognition	0.208	0.047	4.435	***	Accept
H3a	Landscape service cognition → Participation intention	0.772	0.152	5.076	***	Accept
H3b	Landscape service cognition → Conservation intention	0.401	0.077	5.182	***	Accept
H3c	Landscape service cognition → Promotion intention	0.614	0.123	4.994	***	Accept

Note: ***, **, * indicate significance at 1%, 5% and 10% levels, respectively.

and Figure 7, regarding the impact of various types of livelihood capital on behavioral intentions, the paths H1d, H1e, H1h, H1i, H1k, H1l, and H1r are significantly established, while the other paths are not significant. Among these, the path coefficients for H1d, H1e, H1k, H1l, and H1r are positive, indicating that natural capital has a significant positive impact on participation and protection intentions; financial capital has a significant positive impact on conservation and promotion intentions; and cultural capital has a significant positive impact on promotion intention. However, the path coefficients for H1h and H1i are negative, suggesting that physical capital has a significant negative impact on conservation and promotion intentions.

Second, concerning the impact of different types of livelihood capital on landscape service cognition, the paths H2c, H2d, H2e, and H2f are significantly established, and all have positive path coefficients, while the other paths are not significant. This suggests that physical capital, financial capital, social capital, and cultural capital exert a significant positive influence on landscape service cognition. Additionally, in terms of the effect of landscape service cognition on behavioral intention, the paths H3a–H3c are significantly established, with positive coefficients, indicating that cognition of landscape services has a significant positive effect on participation, conservation, and promotion intentions.

In summary, farmers’ natural capital, financial capital, and cultural capital stand out in this pathway, indicating their key roles in promoting farmers’ active participation, conservation, and promotion behaviors.

This study conducted mediation effect analysis using the bootstrap test method, with 1000 repeated samples, and the results are shown in

Table 7. Mediation effect test results.

Hypothesis	Hypothesized Path	Standardized Indirect Effect	Bias-Corrected		Percentile		Result
			Upper Limit	Lower Limit	Upper Limit	Lower Limit
H4	Natural capital → Landscape service cognition → Participation intention	0.026	0.088	−0.020	0.083	−0.038	Not accept
	Natural capital → Landscape service cognition → Conservation intention	0.024	0.082	−0.017	0.077	−0.021	Not accept
	Physical capital → Landscape service cognition → Conservation intention	0.060 *	0.129	0.011	0.123	0.007	Accept
	Physical capital → Landscape service cognition → Promotion intention	0.049 *	0.108	0.010	0.103	0.006	Accept
	Financial capital → Landscape service cognition → Conservation intention	0.146 **	0.226	0.073	0.232	0.077	Accept
	Financial capital → Landscape service cognition → Promotion intention	0.119 **	0.193	0.062	0.191	0.060	Accept
	Cultural capital → Landscape service cognition → Promotion intention	0.131 **	0.215	0.061	0.222	0.063	Accept

Note: **, * indicate significance at 5% and 10% levels, respectively.

. The findings indicate that both physical capital and financial capital can enhance farmers’ willingness to engage in conservation and promotion by increasing their cognition of landscape services. Cultural capital also has a positive indirect effect on promotion intention through the cognition of landscape services. However, the indirect effect of natural capital on participation and conservation intention through landscape service cognition did not reach significant levels, suggesting that other unrecognized pathways or factors may influence the effect of natural capital on farmers’ behavioral intentions. Additionally, as shown in Table 6 and Table 7, the direct and indirect effects of physical capital and financial capital exhibit consistency and significance across different behavioral intentions pathways, further highlighting the critical role these capitals play in shaping farmers’ behavioral intentions.

5. Discussion

5.1. Differences in Landscape Service Cognition and Behavioral Intentions Among Farmers with Different Livelihood Strategies

Livelihood strategies refer to the action plans that farmers develop by leveraging and utilizing their resources to achieve livelihood goals or pursue livelihood outcomes [46]. To explore the differences in farmers’ landscape service cognition and behavioral intentions, this study categorized respondents based on different livelihood strategies. Existing studies generally classify farmers’ livelihoods using two main approaches: (1) the proportion of non-agricultural income to total income or the proportion of non-agricultural labor to total labor input [47], and (2) the sources of farmers’ income or their employment status [48]. The first method is commonly used in economically developed areas or urban–rural integrated regions where comprehensive socioeconomic data are available, allowing for the accurate validation of farmers’ income through data comparison. The second approach, based on income sources or employment status, is more applicable to regions with diverse industrial structures. For example, in this study, some villages have substantial industrial land, showcasing a development trend toward “residential areas integrated with handicraft industrial parks.” Therefore, farmers’ livelihoods rely not only on centuries-old irrigated agriculture but also on multiple income sources, including traditional handicrafts and processing industries. In this study, we classified the sample farmers into five livelihood strategies: diversified livelihood, wage-operated, purely agricultural, labor-led, and subsidy-dependent. This classification provides valuable insights for understanding livelihood diversity in regions with mixed industrial development.

The results of this study reveal significant differences in farmers’ cognition of landscape services based on their livelihood strategies, particularly in the ecological and cultural dimensions. Zhang et al. found that these differences in ecological and cultural cognition are closely related to the type and degree of farmers’ livelihood activities [36], which is further supported by our findings. Notably, in our study, diversified livelihood farmers demonstrated a much higher cognition of cultural value compared to those with other livelihood strategies. Most of these farmers hold public sector jobs, such as civil servants or village cadres, which gives them broader exposure to both agricultural and non-agricultural activities. This multidimensional participation enables them to better understand and appreciate the various values of landscape services, making them more reliant on traditional cultural resources and more committed to their protection and preservation [49]. In terms of the ecological dimension, while there were no substantial differences in the recognition of most ecological functions among the various livelihood strategies, wage-based farmers showed a significantly lower awareness of climate regulation compared to part-time equilibrium farmers. This finding aligns with studies by Wang et al. [50] and Guo et al. [51], which suggests that economically driven farmers tend to place less emphasis on ecological functions, focusing more on short-term economic gains rather than long-term ecological benefits.

Moreover, this study found that diversified livelihood farmers exhibited the highest levels of engagement in participation, protection, and promotion, particularly in ecological agriculture and tourism development. Farmers with diversified livelihoods are generally more willing to engage in regional activities and ecological conservation [24]. In contrast, subsidy-dependent farmers demonstrated a relatively low willingness to participate, which could be attributed to their heavy reliance on government subsidies and a lack of motivation for self-driven development [52]. In terms of protection intentions, diversified livelihood farmers also displayed a stronger willingness to engage in environmental conservation and the utilization of idle resources. This behavior may be driven by both their emotional attachment to rural heritage and their potential economic benefits. These farmers were also more inclined to contribute financially to the maintenance of rural heritage sites and the sustainable development of the environment. Regarding promotion intentions, diversified livelihood farmers showed higher enthusiasm than other groups, especially in educating and encouraging others, likely due to their greater social capital.

5.2. The Impact of Farmers’ Livelihood Capital and Landscape Service Cognition on Behavioral Intentions

Farmers’ livelihood capital and cognition of landscape services significantly influence their behavioral intentions. Livelihood capital, which includes human, natural, physical, financial, social, and cultural capital, collectively shapes farmers’ decision-making when responding to different situations [53,54]. The results of this study indicate that different types of livelihood capital have varying degrees of impact on farmers’ behavioral intentions and their understanding of landscape services.

First, natural capital was found to have a significant positive effect on farmers’ willingness to participate in the protection of the landscape. This is consistent with the findings of Li et al., who demonstrated that the availability and quality of natural capital directly influence farmers’ attitudes and behaviors toward cropland conservation measures [55]. When natural capital is abundant or in urgent need of protection, farmers are more inclined to participate in conservation efforts and seek appropriate compensation models. The positive effect of natural capital indicates that when farmers have access to sufficient natural resources, they are more inclined to participate in conservation activities to sustain both their livelihoods and the resources themselves. However, other studies suggest that in regions where natural resources are abundant and easily accessible, the impact of natural capital on farmers’ behavior may diminish [56]. Additionally, cultural capital also had a notable positive impact on farmers’ willingness to promote heritage conservation, further highlighting its importance in encouraging community participation and protecting heritage [13]. Similarly, financial capital had a clear positive impact on both the willingness to protect and promote, suggesting that greater economic capacity increases farmers’ readiness to invest in rural heritage conservation and promotion. However, it is important to note that physical capital may have negative externalities. This study suggests that the accumulation of physical capital could actually decrease farmers’ willingness to participate in protection and promotion efforts.

In contrast, this study found no significant effects of human capital and social capital on behavioral intentions, which contrasts with other research that highlights the importance of human capital in shaping environmental protection intentions [13]. This discrepancy could be due to the profit-driven nature of farmers’ investment in human capital in the study area, where heritage conservation is viewed more as a public good. Furthermore, the influence of social capital varies considerably depending on its form—such as social trust or social networks—and the context in which it operates. In resource-limited environments, other types of capital, like natural and financial capital, tend to have a stronger influence on farmers’ behavioral intentions, thereby diminishing the impact of social capital [57].

The study also found that landscape service cognition plays a key mediating role between livelihood capital and behavioral intentions. Physical, financial, and cultural capital indirectly increased farmers’ intentions to protect and promote by enhancing their cognition of landscape services, which aligns with the findings of Yu et al. [58]. Although natural capital had some influence on landscape service cognition, its indirect effect on behavioral intentions through this cognition was not significant. Instead, it directly impacted farmers’ participation and protection intentions. This result suggests that the influence pathways of natural capital are more complex and may involve additional mediating variables, which require further investigation.

Moreover, this study confirmed the direct and significant impact of landscape service cognition on farmers’ behavioral intentions, particularly regarding their recognition of cultural and ecological functions, which significantly enhanced their willingness to engage in protection and promotion efforts. This aligns with the historical development of irrigation heritage landscapes, which have gradually evolved toward a more socio-ecological model.

5.3. Policy Recommendations

To address the previously discussed relationships between livelihood capital, landscape service cognition, and behavioral intentions, and to effectively enhance farmers’ willingness to participate in, conserve, and promote rural heritage while achieving sustainable livelihood outcomes, this study recommends the following:

(1)

Promoting the Synergistic Development of Cultural and Financial Capital.

This study highlights the crucial role that cultural capital plays in fostering farmers’ willingness to engage in protective behaviors. Local governments can promote cultural tourism and festivals, while also creating cultural promotion platforms such as museums and traditional handicraft workshops. These efforts can increase farmers’ cultural engagement and strengthen their sense of identity with local traditions. In doing so, not only can farmers’ willingness to protect rural heritage sites be enhanced, but more diversified livelihood opportunities can also be created [41]. For diversified livelihood farmers, it is important to develop cultural resources and enhance the value-added aspects of cultural industries, allowing them to take a leading role in promoting sustainable livelihood outcomes. Financial capital, which has a significant impact on farmers’ behavioral intentions, should also be addressed. Current policies tend to focus on singular subsidies, but financial support should be more diversified. For example, diversified livelihood farmers could benefit from targeted funding that enables them to balance diversified livelihood strategies with greater ecological responsibility [59]. For wage-operated and subsidy-dependent farmers, long-term ecological subsidies and development funds could be used to enhance financial capital, ensuring their livelihood security.

(2)

Enhancing Livelihood Capital to Improve Landscape Service Cognition.

Enhancing farmers’ physical and natural capital, while emphasizing the role of social networks within social capital, can significantly improve their understanding of landscape services and increase their behavioral intentions. The reasonable development of physical capital helps improve farmers’ production conditions and quality of life. Policies should guide farmers in upgrading or establishing infrastructure (such as energy-efficient and environmentally friendly facilities, green buildings, and ecological agriculture equipment) to ensure the efficient use of resources and environmental sustainability [60]. Improving natural capital is also key to strengthening farmers’ willingness to engage in conservation efforts. Providing training in ecological agriculture can help farmers use natural resources more efficiently, which not only increases their awareness of landscape services but also boosts their willingness to participate in ecotourism, biodiversity conservation, and other related activities. Expanding social networks plays an important role in diversifying farmers’ livelihoods and enhancing their behavioral intentions. Through cooperatives and mutual aid organizations, farmers can benefit from knowledge sharing and resource integration, and this mutual support system can increase their participation and enthusiasm in heritage conservation efforts [61]. The expansion of social capital strengthens farmers’ sense of social identity and cooperation, thereby increasing their willingness to engage in and promote rural heritage conservation [62].

(3)

Differentiated Strategies for Farmers with Different Livelihood Strategies.

Diversified livelihood farmers, with their strong understanding of the ecological and cultural value of landscape services, along with their high willingness to participate, protect, and promote, are well-positioned to take on a leading role in rural heritage conservation. Therefore, the government and relevant organizations should further empower them by encouraging participation in rural tourism, cultural preservation, and ecological agriculture projects. Village cadres, such as party branch secretaries and village committee directors, play a critical role in this process. Given that village leaders in China are selected through party elections and direct village-level elections, their influence and continuity in office provide a stable foundation for long-term rural development initiatives. By actively involving village cadres—who are often re-elected based on their performance and leadership—these projects can be better managed and more effectively implemented, ensuring that local resources are efficiently utilized and that rural communities benefit from sustainable practices. Wage-operated farmers, who tend to have weaker cognition of ecological functions and focus more on short-term economic gains, should be guided towards rural heritage conservation through economic incentives such as subsidies. For example, establishing special funds to subsidize industries related to ecological protection, particularly ecotourism and green agriculture, could attract this group to participate in heritage protection and development. Additionally, enhancing their awareness of ecological values could gradually shift their focus toward long-term ecological benefits. For purely agricultural farmers, who show a strong willingness to engage in ecological agriculture, supporting their involvement in rural heritage protection through ecological agricultural projects would be beneficial. This could include providing agricultural training, technical support, and market promotion to improve their productivity while encouraging the adoption of sustainable farming practices that protect the environment. This would enhance their contribution to the protection of rural heritage sites. Labor-led farmers, whose livelihoods largely depend on off-farm work, are less involved in ecological protection and cultural activities. For this group, strengthening their social networks could increase their intention to participate, especially by encouraging them to engage in heritage conservation and cultural transmission during holidays or off-peak farming seasons, alongside the local community. Finally, subsidy-dependent farmers, who generally show low behavioral intentions and are less active in rural heritage protection, may benefit from improved social security measures to stabilize their livelihoods. At the same time, enhanced awareness campaigns could help increase their understanding of the cultural and ecological value of rural heritage sites.

5.4. Limitations and Future Works

This research uncovers important insights into the relationship between farmers’ livelihood capital, landscape service cognition, and their behavioral intentions in rural heritage conservation. It offers both theoretical and practical guidance for improving sustainable livelihood strategies and heritage preservation practices. However, two limitations persist in this research. First, physical capital data remain difficult to obtain, requiring refined survey designs for more detailed information. Second, the influence of the industrial economy, natural environment, and cultural characteristics on farmers’ behaviors has not been fully explored. Future research might delve deeper into these areas to better understand the feedback mechanisms involved.

6. Conclusions

The study of sustainable livelihoods for farmers in rural heritage sites holds significant importance for heritage conservation, transmission, and the sustainable development of farmers. This study constructs a research framework within the Sustainable Livelihoods Approach, integrating Cognition and Behavior Theory, and develops a scale to assess landscape service cognition and farmers’ behavioral intentions. Using a sample of 371 farmers from representative villages within Putian’s Ecological Green Heart, this study explores the differences among farmers with various livelihood strategies and analyzes key factors influencing their behavioral intentions from the perspectives of livelihood capital and landscape service cognition. The key findings are as follows:

(1)

The normalized levels of human capital (0.541), social capital (0.671), and cultural capital (0.645) are relatively high among farmers in the study area, while the levels of natural capital, physical capital, and financial capital are comparatively low.

(2)

Diversified livelihood farmers exhibit the highest levels of overall landscape service cognition and behavioral intentions. They stand out in their recognition of ecological and cultural services and demonstrate a strong enthusiasm for participating in ecological agriculture and the handicrafts and processing industries. In contrast, subsidy-dependent farmers have the lowest level of behavioral intentions.

(3)

Natural capital, financial capital, and cultural capital play key roles in influencing farmers’ landscape service cognition and behavioral intentions. Additionally, landscape service cognition mediates the relationship between livelihood capital and farmers’ behavioral intentions.

(4)

To enhance farmers’ willingness to protect rural heritage, the focus should be on improving and accumulating natural, physical, and financial capital. In contrast, to increase farmers’ willingness to promote rural heritage, the accumulation of cultural capital is particularly crucial.