Influencing Factors and Transmission Mechanisms of Pro-Environmental Behavior: Evidence from Tea Farmers in Wuyishan National Park

Abstract

Tea farmers in Wuyishan National Park face the dual challenges of promoting sustainable tea production while adhering to strict ecological protection policies. This study investigates the key factors influencing tea farmers’ pro-environmental behavior and the transmission mechanisms that encourage the adoption of sustainable development. Based on the theory of planned behavior, the theory of externalities, and place attachment theory, 346 valid questionnaires were collected through household interviews across 12 villages within Wuyishan National Park and its surrounding areas. The results indicate that environmental responsibility and concern for community well-being are major motivators of pro-environmental behavior. Market orientation, production intensification, and adoption of clean technologies significantly enhance environmental outcomes at the farm level. While ecological compensation policies help offset the costs of conservation, their impact is constrained by inconsistencies in standard-setting and implementation. The establishment of the national park has enhanced pro-environmental behavior among farmers within the park and influenced those in peripheral areas by strengthening place attachment and social norms. Tea farmers in the core areas of the national park exhibit higher levels of pro-environmental behavior compared to those on the periphery. This study offers several policy recommendations and contributes new insights into understanding the mechanisms behind tea farmers’ pro-environmental behavior within national park contexts, providing valuable reference for pro-environmental practices in the global protected area system.

1. Introduction

The challenge of harmonizing agricultural production with ecological conservation has become a central issue in the era of global sustainable development. While agriculture underpins food security and economic growth, it is also a leading cause of environmental degradation, including soil erosion, biodiversity loss, and pollution. Effectively guiding farmers to adopt environmentally sustainable practices is an essential challenge for both environmental management and agricultural economics. Their actions directly influence the health of natural ecosystems. However, research on farmers’ environmental behavior in specific policy contexts, such as national park systems, remains limited. There is an urgent need to uncover the driving mechanisms behind these behaviors and their regional variations.

The establishment of national parks marked the start of protected areas. Over more than a century, countries have developed distinctive national park systems [1]. Since 2015, China has been piloting its national park system to protect key ecosystems and balance resource conservation with sustainable use [2,3]. Wuyishan National Park, one of China’s first national parks, highlights the dual goals of ecological protection and economic development. Because of the geographical location and climatic conditions, Wuyishan has been a famous tea area since ancient times and is considered the birthplace of black and oolong tea [4]. The development of the tea industry has created huge economic value, and local farmers are enthusiastic about planting tea [5]. However, numerous ecological problems have been exposed with the development of the tea industry [6], such as deforestation, the application of prohibited fertilizers, toxic biological pesticide residues, fuel consumption based on wood and coal, and the use of Masson pine wood in nematode endemic areas for smoke tea production. Currently, there is considerable tension between the tea farmers’ goal of maximizing livelihood benefits and the national park’s goal of “ecological protection first” [7]. The misalignments translate into several concrete challenges for tea farmers. These include restrictions on expanding tea plantations, constraints on traditional tea-smoking methods due to the ban on using Masson pine, and rising compliance costs linked to the adoption of cleaner production technologies. In addition, ongoing uncertainty regarding the criteria and timing of ecological compensation schemes heightens farmers’ concerns about income security and discourages long-term investment in sustainable practices.

Tea farmers are important carriers and management subjects of the tea culture and industry in Wuyishan National Park [8]. Guiding and regulating the production behaviors of tea farmers not only improves the quality of tea but also effectively reduces the negative impacts of human production activities on the ecological environment [9]. However, it is still unclear in which ways the government can effectively encourage tea farmers to adopt pro-environmental behaviors so that they can maintain economic benefits and protect the ecological environment simultaneously. Especially in specific agricultural practices, questions such as how to translate the ecological concept into the practical actions of farmers and how to protect the ecology without damaging the benefits of farmers have not been well-answered [10]. Therefore, to formulate more targeted and operational policies and measures, it is crucial to study the mechanisms underlying farmers’ pro-environmental behaviors from the perspective of their behaviors. In this study, based on planned behavior, externality, and place attachment theory, we explore the combination path of pro-environmental behaviors and regional non-wood forest products from a micro perspective [11].

This study focuses on the following questions:

(1)

What is the level of pro-environmental behavior of tea farmers in Wuyishan National Park?

(2)

What are the factors influencing the pro-environmental behavior of tea farmers in Wuyishan National Park?

(3)

What is the transmission pattern of tea farmers’ pro-environmental behavior?

This study aims to “assess behavior levels—identify key factors—clarify transmission mechanisms—inform policy design,” with the ultimate goal of supporting more balanced ecological, economic, and social outcomes in protected area governance. Although focused on Wuyishan National Park, the findings provide insights into the global challenge of balancing agriculture with environmental protection in protected areas. This study offers a practical framework for promoting sustainable agricultural practices in regions facing similar ecological and economic challenges.

2. Theoretical Foundation

This study employs three core theoretical frameworks to explain the complex motivations behind tea farmers’ pro-environmental behavior, while also considering the dual objectives of ecological conservation and economic development within the context of national parks. These frameworks are integrated to explore the driving mechanisms behind pro-environmental behaviors. Such integration responds to recent calls in environmental psychology and behavioral economics for multi-theory approaches that reflect the social, emotional, and structural complexity of sustainability behaviors [12,13,14].

• Theory of Planned Behavior

The theory of planned behavior posits that individual actions are not solely the result of rational decisions, but are also influenced by social norms and external control factors [15]. For tea farmers, decisions to engage in pro-environmental behavior are shaped by their attitudes toward environmental protection, the expectations of their social groups, and their available resources and capabilities to implement sustainable practices. This framework emphasizes both cognitive and social determinants, providing insights into the psychological mechanisms that underpin behavior. The TPB can be further strengthened when combined with affective and place-based factors, such as environmental identity and place attachment [16,17]. This is particularly relevant for rural communities where spatial belonging shapes both attitudes and norms.

• Theory of Externalities

The theory of externalities examines how individual actions indirectly affect others, particularly in contexts involving shared resources and ecosystem protection [18]. Tea farmers’ production practices, such as the use of fertilizers or ecological restoration efforts, generate both positive and negative externalities for neighboring communities and the natural environment. In the case of Wuyishan National Park, the actions of tea farmers impact their economic outcomes but also influence the broader sustainability of the surrounding ecosystem and society. Combining externalities with behavioral frameworks helps explain how farmers respond to institutional incentives and ecological compensation when their private costs do not align with public environmental benefits [19,20].

• Place Attachment Theory

Place attachment theory focuses on the emotional bonds between individuals and their environment [21]. This emotional response is often rooted in an individual’s fundamental need for security and a sense of belonging, which are particularly relevant for tea farmers in Wuyishan. Furthermore, place attachment can serve as a catalyst for environmental responsibility and positive behaviors among individuals [22]. As both economic participants and stewards of the unique ecological environment of Wuyishan, this emotional connection is especially pertinent in areas like national parks, which hold both cultural and ecological value.

These three theories intersect in a multidimensional manner, collectively shaping the pro-environmental behavior of tea farmers. Tea farmers’ production decisions are influenced not only by their personal commitments to environmental protection but also by economic incentives [23]. Under the TPB, their environmental attitudes are closely tied to market demand, especially when green products yield higher returns, making farmers more inclined to adopt sustainable practices. Market incentives enhance farmers’ perceptions of the economic benefits of environmental actions, thereby reducing psychological barriers to implementation.

From the perspective of the externalities, the costs of ecological maintenance reflect the conflict between external benefits and internal costs faced by tea farmers [24]. When ecological compensation or incentive policies are implemented, cost reduction can provide farmers with stronger economic motivation to protect the environment. Tea farmers’ pro-environmental behavior is also influenced by social norms, particularly when group members share common interests [25]. The increased attention to injunctive norms and a better understanding of injunctive norms can encourage tea farmers to rectify their behaviors [26]. In such cases, the consistency of behavior and peer pressure encourage individuals to make decisions aligned with collective expectations [27]. Nevertheless, due to the influence of acquaintance society, relatives and friends may tolerate and forgive each other’s bad behaviors, resulting in an opposite social sensitivity to the expected situation [28].

The emotional bond between tea farmers and the land and community plays a crucial role in motivating them to adopt pro-environmental practices. In the context of Wuyishan National Park, tea farmers’ sense of belonging and responsibility toward the ecological environment drives them to actively protect the ecosystem. Tea farmers’ attachment to the land increases their willingness to invest more effort in ecological conservation. Their awareness of environmental issues and understanding of future ecological responsibilities directly influence the behaviors they adopt in production processes [29]. High environmental awareness and a strong sense of responsibility lead tea farmers to evaluate not only economic benefits but also the long-term ecological impact of their actions. Additionally, concern for others’ well-being plays a significant role as well. Tea farmers care not only about their own economic interests but also about the welfare of others and the broader society. This concern drives them to consider the wider social impact of their actions, thereby increasing their likelihood of engaging in pro-environmental behavior.

Social norms and market incentives work in tandem, encouraging tea farmers to comply with societal expectations while also exploring economically sustainable production methods. This dual-driven mechanism ensures that tea farmers’ pro-environmental behaviors are grounded in both social responsibility and economic feasibility. The complementarity between place attachment and psychological factors strengthens farmers’ sense of responsibility and environmental sensitivity, while their heightened environmental awareness and concern for others’ well-being further drive pro-environmental actions. In summary, the three theories form an interconnected analytical framework. Each theory offers deep insights into the underlying drivers of behavior. In practice, these factors are interwoven, creating a complex and dynamic decision-making process.

3. Materials and Methods

3.1. Study Area and Data Source

Wuyishan National Park of China is located in the northern section of the Wuyishan Mountain Range between Fujian and Jiangxi and covers a total area of 1280 km². The Fujian section of the national park spans 29 administrative villages, with more than 700 households. In 2023, the output value of the tea industry chain in Wuyishan reached 13.5 billion yuan, and nearly 50% of the per capita income of farmers was derived from this industry. According to data from the Agricultural and Rural Bureau of Wuyishan City for 2023, there were 148,000 mu of tea gardens and 21,118 registered tea market entities in the city, including 43 tea enterprises above the designated size and 2 national leading enterprises. Among the 260,000 residents of the city, 120,000 were involved in the tea industry.

In 2023, we had several informal discussions with the Wuyishan National Park Administration to understand the basic situation of tea industry development and tea farmers’ production behaviors in the Wuyishan area. Based on this understanding, we selected 12 major tea farming villages within the Fujian section of Wuyishan National Park of China. These villages were chosen for their large population, high proportion of permanent residents, and their concentration of tea production. Due to the widespread migration of tea farmers, we focused on surveying all household members who remained in the villages, ensuring that our sample covered the permanent farming population. We conducted household interviews in these villages, collecting a total of 354 questionnaires. After excluding 8 invalid questionnaires with missing or abnormal key variables, we obtained a total of 346 valid questionnaires, with an effective rate of 97.74%. The distribution of the research sample is shown in

Table 1. Distribution of the research sample.

County (City, District)	Township	Village	Sample Size	Effective Sample Size	Effective Rate
Wuyishan	Xingcun	Tongmu	6	6	100.00%
		Huang	21	20	95.24%
		Hongxing	34	34	100.00%
		Xing	34	32	94.12%
		Chengdun	41	41	100.00%
		Caodun	33	32	96.97%
		Zhoutou	40	39	97.50%
		Lixin	22	21	95.45%
		Liyuan	16	14	87.50%
	Wuyi	Huangbai	49	49	100.00%
	Xingtian	Nanyuanling	34	33	97.06%
	Yangzhuang	Daan	24	23	95.83%
Total	4	12	354	346	97.74%

. Figure 1 shows the location map of Wuyishan National Park.

China has delimited clear lines of demarcation when building national parks. Based on edge effect theory and the practical situation of Wuyishan National Park, some tea farmers in nearby villages remain significantly connected to the national park. Even those located outside the park’s boundary are still influenced by the presence of the protected area, which affects the decision-making of local communities and governments, leading to a certain consistency in policy formulation [30,31]. For example, although some tea farmers have moved out of the national park (based on the landmark boundary), their residences are only one street or one river away from the national park. These tea farmers are still closely linked to the national park. Additionally, while some farmers have relocated outside the park, they had previously lived within the park, or their production resources remain inside the park, meaning their production activities are still influenced by the national park’s policies. Nanyuanling village is a typical relocation village. Therefore, we treat these farmers as a group along with those inside the park, referring to them as Group 1.

Group 2 specifically includes tea farmers who are entirely within the park’s boundaries. These farmers are subject to the more stringent management policies within the national park, which have a more direct and uniform impact on their production activities and policy enforcement. By comparing these two groups, it enabled us to better explore the national park system and its edge effect.

3.2. Variable Description

3.2.1. Dependent Variable

In this study, the level of pro-environmental behaviors of tea farmers in Wuyishan National Park was taken as the dependent variable. Because tea planting and processing is a complex process, not every link will have a great influence on the ecological environment. Hence, based on the actual situation of tea production and processing of tea farmers in Wuyishan National Park, we extracted a total of 13 behaviors that greatly impact the ecological environment from all links of pre-production, in-production, and post-production to measure the level of pro-environmental behaviors of tea farmers. The selection was grounded in their ecological impact, as confirmed by previous research, expert consultations, and field observations specific to tea farming practice in Wuyishan National Park [32]. For details, see

Table 2. Scale for measuring pro-environmental behavior of tea farmers.

Dependent Variable	Production Stage	Behavioral Indicators
Pro-environmental behavior	Pre-production	The expansion of tea plantations in ten years (Y1)
		The main basis for selecting tea seedling varieties (Y2)
		The main basis for the purchase of fertilizers (Y3)
	Production	Fuel for roasted tea (Y4)
		The way of smoking tea (Y5)
		Fertilizer use (Y6)
		Types of fertilizers (Y7)
		The way of fertilizing or sprinkling (Y8)
		The way to clean up weeds (Y9)
		The main irrigation method (Y10)
	Post-production	Tea categories (Y11)
		Tea packaging strength (Y12)
		Waste disposal methods (Y13)

.

3.2.2. Core Independent Variables

Many studies have shown that farmers can be encouraged to adopt more environmentally friendly production methods through efforts in terms of institutional design, market mechanisms, and cultural identity [33,34,35]. For example, some regions encourage farmers to adopt organic farming methods by establishing ecological compensation mechanisms [36]. The market recognition of ecological products is improved through certification system introduction and brand building, and farmers’ environmental awareness and skills are enhanced through community participation and free education and training. Overall, these above measures have promoted the formation of pro-environmental behaviors to a certain extent, but there are still challenges in practice, such as insufficient capital investment, insufficient technical support, and poor market docking [37]. The differences among individual tea farmers also affect the consistency of policy effects [38].

Pro-environmental behaviors are behaviors caused by an individual’s internal self-drive and external pressure and influence [39]. The selection of core independent variables mainly considers the following aspects: First, as the market management subject, tea farmers pursue the maximization of income and the minimization of cost at the monetary level [40]. In this regard, tea farmers’ perception of the cost and income of choosing pro-environmental behaviors should be investigated [41]. If the pro-environmental behaviors are profitable, farmers will be more likely to choose behaviors with higher pro-environmental levels. Second, the environment in which tea farmers live may affect their behaviors [42]. The close-range social grid can produce information pressure and obedience-induced convergence [43], whereas knowledge sharing among families can promote the circulation of environmental knowledge [44]. Environmental knowledge and concern have a positive and significant impact on individuals’ pro-environmental behaviors directly or through the potential generation mechanism of behaviors [45]. Technology diffusion is more convenient for the acquaintance society [46], and the imperative norms and descriptive norms can produce the imperceptibility of individual behaviors [47]. Third, the individual’s choice is driven by consciousness, thinking, and psychological activities [48]. Due to the lack of a sense of security and gain, anxiety and sadness about migration confirm a psychological connection to local space [49]. Through the establishment of a psychological contract between individuals and specific places [50], natural empathy is generated [51], and individuals will abandon short-sighted behaviors and comprehensively consider the benefit value of individual behaviors to places and individuals based on a long-term perspective [52]. The tea farmers in Wuyishan National Park are mostly workers who live in local areas all year round, with a low migration frequency. If they have a higher degree of environmental awareness [6], a higher sense of environmental responsibility [53], a greater incentive expectation for pro-environmental behaviors [54], and more concern about the well-being of others [55], they will be more likely to adopt high-level pro-environmental behaviors.

Combining the above theoretical analysis with the literature review, this study summarizes the core variables that may cause tea farmers to choose higher levels of pro-environmental behaviors into economic, social, and psychological factors; each category of factors includes six secondary indicators. Each secondary indicator is assigned with a specific value in the range of 1 to 5, and different values represent the degree of agreement of the respondents: “strongly disagree” = 1, “fairly disagree” = 2, “unconcerned or uncertain” = 3, “fairly agree” = 4, and “strongly agree” = 5. For details, see

Table 3. Description, assignment, and expectation of indicators.

Grade 1	Grade 2	Indicator Description	Expectation
Economic factors	Excessive use of chemical fertilizers	Excessive use of fertilizers and pesticides will make it difficult to sell tea.	+
	Intensive management	Intensive cultivation of tea gardens will boost income.	+
	Prohibition of Chinese red pine	Banning the use of Chinese red pine for tea smoking will have a significant negative impact on the income from tea operations.	−
	Ecological maintenance costs	Even though it will cost more money to maintain the ecosystem of the tea plantations, ecological maintenance will still be carried out.	+
	Production restriction	The establishment of the Wuyishan National Park has caused inconvenience to the cultivation and processing of tea.	+
	Market orientation	When tea is purchased, the purchase price will be increased because the tea is greener and pollution-free.	+
Social factors	Place attachment	Proud of the tea culture in my hometown, I am willing to produce more natural tea.	+
	Social norms	I often notice more signs and posters around my life and try to understand the meaning of these signs and posters.	+
	Group pressure	When growing and processing tea, I will actively refer to my neighbors’ practices and want to be consistent with them.	+
	Application of clean technology	If a new technology or method can reduce ecological pollution in tea planting and processing, I will actively try it.	+
	Environmental regulation	If I know the relevant regulations on ‘ecological environment protection’, I will abide by them and ask myself to implement them.	+
	Establishment of Wuyishan National Park	After the establishment of Wuyishan National Park, I will pay more attention to the protection of the environment in tea planting and processing.	+
Psychological factors	Environmental awareness	When planting and processing tea, I will care about the impact of my actions on the ecosystem.	+
	Environmental responsibility	The protection of the ecological environment in Wuyishan National Park requires each of us to take up our own responsibility.	+
	Social sensitivity	If I pollute the environment during tea planting and processing, I will worry about my neighbors’ comments.	+
	Incentive expectation	If rewarded, I will be happy to carry out ecological maintenance.	+
	Reputational appeals	Expect to get a good reputation for doing ecologically sound behavior.	+
	Concern for the well-being of others	I will be happy to carry out eco-maintenance of my own tea gardens, which will benefit everyone.	+

Note: + denotes that higher indicator values correspond to stronger pro-environmental behavior, while − indicates the opposite direction.

.

3.2.3. Control Variables

The control variables include the group characteristics of tea farmers and the management situation of tea farms [56]. Household characteristics such as gender, age, education level, and occupation may be the variables that affect the level of behaviors. Tea farm management based on actual tea production mainly involves six aspects of indicators from the perspective of production material design, including the tea farm area, whether to join the cooperative, and whether there is machinery on the farm [57]. For details, see

Table 4. Description, assignment, and expectations of indicators (in order of questionnaire topics).

Grade 1	Grade 2	Indicator Description	Indicator Assignment	Expectation
The group characteristics	Gender	Female or male	Female = 0; male = 1	+/−
	Age	Length of time alive	Age (in years) of the surveyed tea farmer household head	+/−
	Education level	Highest level of education	None = 0; primary education = 1; secondary education = 2; tertiary education = 3; postgraduate education = 4	+
	Occupation	Work	None = 0; farming = 1; non-farm employment = 2; part-time work = 3	+/−
	Whether there are village cadres	Consideration for household situation	No = 0; yes = 1	+
	Main source of family income		Labor outside the park = 1; income from tourism type in the park = 2; tea cultivation = 3; tea business = 4; wage income from forest protection, law enforcement, etc. = 5	+
The management situation of tea farms	Tea farm area	Land area used for tea production (including land area of self-reserved hills, leased others’ tea gardens, etc.)	0 acres ~ no upper limit	+
	Whether to join the cooperative	Join a tea-related co-operative.	No = 0; yes = 1	+
	Whether there is machinery on the farm	Large-scale machinery related to tea planting and processing, such as lawn mowers, tea frying machines, and tea smoking machines.		+
	Whether to participate in technical training	Participation in tea cultivation training, tea processing training, etc.		+
	Frequency of participation in technical training	Interval of participation in the above tea-related training techniques.	No = 0; once a year (including above) = 1; once in half a year = 2; once in a quarter = 3; once in a month (including below) = 4	+
	Whether receiving ecological compensation	Economic or material subsidies received from the village office or other sources for ecological protection.	No = 0; yes = 1	+

Note: + denotes that higher indicator values correspond to stronger pro-environmental behavior, while − indicates the opposite direction.

.

3.3. Analytical Methods

3.3.1. Factor Analysis

The observable production behavior of tea farmers is the dependent variable component, and the pro-environmental behavior of tea farmers in Wuyishan National Park is the latent variable of this study. For the convenience of factor analysis, objective weighted factor analysis was used to conduct dimension reduction and value assignment on the dependent variable matrix, and the KMO test and Bartlett sphericity test were performed. Using SPSS 22.0, we extracted the common factors for specific production behaviors of tea farmers. The extraction method was selected as the main component, the maximum variance method was employed for orthogonal rotation, and the Bartlett method was used for score assignment. According to the production performance of tea farmers, the letters A to D represent the levels of pro-environmental behaviors of tea farmers from high to low, of which A, B, C, and D stand for excellent level, good level, fair level, and unqualified level. The quantification process for the dependent variables are detailed in Appendix A.

3.3.2. Ordered Probit Model

The Ordered probit model (Oprobit) was chosen for its suitability in handling ordinal dependent variables, which is crucial for analyzing pro-environmental behavior, as it is naturally categorized into varying levels of sustainable practices. Unlike binary or continuous variables, pro-environmental behavior follows a clear order of intensity. The Oprobit model enables us to estimate the likelihood of tea farmers adopting different levels of environmental behavior based on various factors, providing a deeper understanding of how each factor influences the probability of specific actions at different levels of pro-environmental behavior. The latent regression model was established as follows:

$$y^{\ast }=x^{\prime }\mathsf{\beta }+\epsilon , \epsilon \left|X\right.~N(0, 1),$$

where $y^{\ast }$ is an unobservable latent variable and $\epsilon$ is the error term.

3.3.3. Interpretive Structural Modeling Method

The Interpretive Structural Modeling (ISM) method was used to examine the interrelationships and causal pathways between factors influencing pro-environmental behavior. ISM allows us to map the complex interactions between socio-economic and environmental variables, identifying key drivers to tea farmers’ behavior. By structuring these relationships hierarchically, ISM provides deeper insights into how certain factors enable the promotion of pro-environmental behavior. The specific steps were as follows:

First, we determined the adjacency matrix $R$.

$$R_{ij}=\left\{\begin{array}{l}1(S_i has influences on S_j)\\ 0(S_i has no influences on S_j)\end{array}\right.(i,j=1, 2, 3\dots k),$$

where $k$ is the number of factors affecting the pro-environmental behaviors of tea farmers in Wuyishan National Park in the results of the regression model, and $S_i$ represents an influencing factor. The logical matrix output by the Delphi method served as the adjacency matrix ($R$) for the ISM model, and the hierarchical structure was generated through the reachability matrix calculation, as shown in Appendix B.

Subsequently, according to the adjacency matrix, the reachable matrix $M$ can be calculated as follows.

$$M={(R+I)}^{\lambda +1}=(R+I{)}^{\lambda }\ne {(R+I)}^{\lambda -1}\ne \dots \ne {(R+I)}^2\ne (R+I)(2\le \lambda \le k),$$

where $I$ represents the unit matrix of the influencing factors of pro-environmental behaviors, by which the hierarchical structure among the influencing factors can be further obtained using the following equation:

$$L=\left\{S_i\left|P\left(S_i\right)\cap \right.\right.Q\left(S_i\right)=P\left.(S_i)\right\}(i=1,2,\dots ,k),$$

where $P\left(S_i\right)$ is the reachable set of the factors affecting the pro-environmental behaviors of tea farmers in Wuyishan National Park, that is, the set of all factors that can be reached from the influencing factor $S_i$.

4. Results

4.1. Descriptive Statistics and Reliability Analysis

4.1.1. Descriptive Statistics

Tea farm sizes varied widely, with the average farm covering 19.85 mu. A large standard deviation was observed due to three factors: (1) land consolidation and leasing by large households; (2) earlier government encouragement for land reclamation and tea plantation; (3) expansion by the “privileged persons” in the village (such as the village chief or secretary) through land mergers. Only 14.5% of farmers joined cooperatives, despite a high mechanization rate (77.2%) in production and processing. However, participation in technical training was low (21.1%), with limited frequency. Ecological compensation reached 43.6% of households, with an average compensation of CNY 1032.66 per household annually. The average annual income from tea businesses was CNY 165,400, with a large variance. The average income per mu of tea land was CNY 8396, and annual input costs per mu totaled CNY 1975.5, resulting in an estimated annual profit of CNY 6420.5 per mu. In addition, the descriptive statistics of the group characteristics of tea farmers are shown in

Table 5. Basic characteristics of farm households.

Variable	Category	Frequency	Proportion
Gender	Male	323	93.35%
	Female	23	6.65%
Age	44 years old and below	55	15.90%
	45–59 years old	159	45.95%
	60–74 years old	105	30.35%
	75–89 years old	26	7.51%
	90 and above	1	0.29%
Education level	None	24	6.93%
	Primary education	135	39.02%
	Secondary education	172	49.71%
	Tertiary education	15	4.34%
	Postgraduate education	0	0.00%
Occupation	None	1	0.29%
	Farming	90	26.01%
	Non-farm employment	19	5.49%
	Part-time work	236	68.21%
Whether there are village cadres	Yes	33	9.54%
	No	313	90.46%
Main source of family income	Labor outside the park	31	8.96%
	Income from tourism type in the park	13	3.76%
	Tea cultivation	220	63.58%
	Tea business	79	22.83%
	Wage income from forest protection, law enforcement, etc.	3	0.87%

.

4.1.2. Pro-Environmental Behaviors of Tea Farmers

The factor analysis for pro-environmental behaviors showed valid results (KMO = 0.683, Bartlett’s significance = 0.000). Among the respondents, 42.78% demonstrated moderate pro-environmental behaviors (level C), while 28.32% exhibited good behaviors (level B), and 16.18% showed excellent behaviors (level A). However, 12.72% of farmers exhibited insufficient pro-environmental behaviors, indicating substantial room for improvement toward meeting the ecological goals of the national park.

Table 6. Distribution of the levels of pro-environmental behaviors of tea farmers.

Level of Pro-Environmental Behavior	Score	Frequency	Proportion (%)
A. Excellent	>0.6	56	16.18
B. Good	0–0.6	98	28.32
C. Fair	−0.6–0	148	42.78
D. Unqualified	<−0.6	44	12.72

Note: The scores for pro-environmental behavior are provided in Appendix A.

shows the distribution of the levels of pro-environmental behaviors of tea farmers in Wuyishan National Park. The pro-environmental behavior standards are detailed in Appendix A.

4.1.3. Reliability and Validity Assessment

The core dependent variable in this study is a composite variable, which requires evaluation of its reliability and validity. To assess these aspects, reliability analysis was conducted using Cronbach’s alpha coefficient, and validity was assessed using the KMO measure. The results presented in

Table 7. The result of Cronbach’s alpha coefficient.

Variable	α Coefficient	Variable	α Coefficient	Variable	α Coefficient	Cronbach α Coefficient
E1	0.801	S1	0.784	P1	0.787	0.796
E2	0.791	S2	0.777	P2	0.793
E3	0.800	S3	0.802	P3	0.774
E4	0.781	S4	0.786	P4	0.773
E5	0.810	S5	0.784	P5	0.774
E6	0.782	S6	0.785	P6	0.767

demonstrate that the variable has high reliability, with Cronbach’s alpha coefficients for individual items ranging from 0.767 to 0.810, indicating strong internal consistency. Additionally, the KMO value of 0.841 and Bartlett’s test of sphericity indicate that the study data is highly suitable for factor extraction, indirectly reflecting good validity.

4.2. Results for Group 1

Prior to regression estimation, the multicollinearity test was conducted on the variables of the data used in this study. According to the test results, the tolerance values were all greater than 0.1 and below 1, and the VIF values were all below 10, with a mean of 1.65, indicating that there was no multicollinearity.

4.2.1. Factors Influencing the Pro-Environmental Behaviors of Group 1 and the Marginal Contribution of Variables

The regression results of Group 1 and the marginal contribution of variables are shown as Coefficient 1 in

Table 8. Regression results and marginal contribution of factors influencing pro-environmental behavior in Group 1.

Variable	Coefficient 1 (t-Value)	D (Std.Err.)	C (Std.Err.)	B (Std.Err.)	A (Std.Err.)
E1	−0.044 (−0.26)	0.003 (0.011)	0.002 (0.008)	0.000 (0.001)	−0.005 (0.017)
E2	0.233 ** (2.39)	−0.015 ** (0.007)	−0.011 ** (0.004)	0.001 (0.002)	0.025 ** (0.010)
E3	−0.035 (−0.41)	0.002 (0.005)	0.002 (0.004)	0.000 (0.001)	−0.004 (0.009)
E4	0.246 ** (2.24)	−0.016 ** (0.007)	−0.011 ** (0.005)	0.001 (0.002)	0.026 ** (0.012)
E5	0.130 * (1.77)	−0.008 * (0.005)	−0.006 * (0.003)	0.001 (0.001)	0.014 * (0.008)
E6	0.393 *** (3.19)	−0.025 *** (0.008)	−0.018 *** (0.007)	0.002 (0.003)	0.042 *** (0.013)
S1	0.215 (1.34)	−0.014 (0.010)	−0.010 (0.008)	0.001 (0.002)	0.023 (0.017)
S2	0.339 *** (2.60)	−0.022 ** (0.009)	−0.016 *** (0.006)	0.002 (0.002)	0.036 *** (0.014)
S3	−0.225 *** (−2.75)	0.015 *** (0.005)	0.010 *** (0.004)	−0.001 (0.001)	−0.024 *** (0.009)
S4	0.323 *** (3.83)	−0.021 *** (0.006)	−0.015 *** (0.004)	0.002 (0.002)	0.034 *** (0.009)
S5	0.250 * (1.66)	−0.016 (0.010)	−0.012 * (0.007)	0.001 (0.002)	0.026 * (0.016)
S6	0.121 (1.31)	−0.008 (0.006)	−0.006 (0.004)	0.001 (0.001)	0.013 (0.010)
P1	−0.066 (−0.43)	0.004 (0.010)	0.003 (0.007)	0.000 (0.001)	−0.007 (0.016)
P2	0.755 *** (7.03)	−0.049 *** (0.006)	−0.035 *** (0.007)	0.004 *** (0.005)	0.080 *** (0.010)
P3	0.069 (0.59)	−0.004 (0.008)	−0.003 (0.006)	0.000 (0.001)	0.007 (0.012)
P4	0.411 ** (2.57)	−0.027 *** (0.010)	−0.019 ** (0.009)	0.002 (0.003)	0.044 *** (0.017)
P5	0.171 (1.43)	−0.011 (0.008)	−0.008 (0.006)	0.001 (0.001)	0.018 (0.013)
P6	0.889 *** (7.43)	−0.058 *** (0.009)	−0.041 *** (0.006)	0.005 (0.006)	0.094 *** (0.012)
Gender	−0.550 ** (−2.19)	0.036 ** (0.017)	0.026 ** (0.012)	−0.003 (0.004)	−0.058 ** (0.026)
Age	−0.004 (−0.55)	0.000 (0.000)	0.000 (0.000)	0.000 (0.000)	0.000 (0.001)
Education level	−0.125 (−0.91)	0.008 (0.009)	0.006 (0.006)	−0.001 (0.001)	−0.013 (0.015)
Occupation	−0.069 (−0.71)	0.004 (0.006)	0.003 (0.004)	0.000 (0.001)	−0.007 (0.010)
Whether there are village cadres	−0.087 (−0.37)	0.006 (0.015)	0.004 (0.011)	0.000 (0.001)	−0.009 (0.025)
Main source of family income	−0.019 (−0.19)	0.001 (0.006)	0.001 (0.005)	0.000 (0.001)	−0.002 (0.010)
Tea farm area	−0.015 *** (−3.72)	0.001 *** (0.000)	0.001 *** (0.000)	0.000 (0.000)	−0.002 *** (0.000)
Whether to join the cooperative	0.075 (0.28)	−0.005 (0.018)	−0.003 (0.012)	0.000 (0.001)	0.008 (0.029)
Whether there is machinery on the farm	−0.289 (−1.45)	0.019 (0.013)	0.014 (0.009)	−0.002 (0.002)	−0.031 (0.021)
Whether to participate in technical training	0.357 (1.08)	−0.023 (0.022)	−0.017 (0.016)	0.002 (0.003)	0.038 (0.035)
Frequency of participation in technical training	−0.539 ** (−2.31)	0.035 ** (0.015)	0.025 ** (0.012)	−0.003 (0.004)	−0.057 ** (0.024)
Whether receiving ecological compensation	−0.392 ** (−2.34)	0.025 ** (0.011)	0.018 ** (0.008)	−0.002 (0.003)	−0.042 ** (0.018)

Note: *** p < 0.01, ** p < 0.05, * p < 0.1; A, B, C and D represent the grading of environmentally friendly behavior.

. Among the core variables, market orientation (E6), social norms (S2), group pressure (S3), application of clean technology (S4), environmental responsibility (P2), and concern for the well-being of others (P6) were significant at the level of 1%. Intensive management (E2), ecological maintenance cost (E4), and incentive expectation (P4) were significant at the level of 5%, and production restriction (E5) and environmental regulation (S5) were significant at the level of 10%.

The factors that had the greatest influences on promoting the marginal contribution rates of tea farmers’ A-level and B-level pro-environmental behaviors were concern for the well-being of others (9.4%) and environmental responsibility (0.4%), respectively. The factor with the most pronounced influence on reducing the marginal contribution rates of tea farmers’ C-level and D-level pro-environmental behaviors was also concern for the well-being of others (P6), with marginal contribution rates of 4.1% and 5.8%, respectively.

4.2.2. Transmission Pathways of Pro-Environmental Behaviors of Group 1

The significant variables affecting the pro-environmental behaviors of tea farmers were substituted into the ISM model to further determine the transmission relations and pathways among these variables. Following the Delphi method and considering the actual situation of Wuyishan National Park, we constructed the logical relationship matrix of key factors influencing the pro-environmental behaviors of tea farmers in Group 1.

The factors of Group 1 were divided into three levels, namely L1 = {gender, tea farm area, ecological compensation}, L2 = {training frequency, ecological maintenance cost, production restriction, market orientation, social norms, group pressure, clean technology application, environmental regulation, environmental responsibility, incentive expectation}, and L3 = {intensive management, concern for the well-being of others}. Based on the hierarchical decomposition results, the relational structure diagram is shown in Figure 2.

For Group 1, gender, tea farm area, and ecological compensation constituted the sources of pro-environmental behavior of tea farmers. These economic, social, and psychological situational factors can strongly guide tea farmers to implement intensive management and have concern for the well-being of others, thus producing pro-environmental behavior. Specifically, the main pathways for the occurrence of pro-environmental behavior are as follows:

Pathway 1: Gender → expected incentives, training frequency, group pressure → clean technology application → intensive management → pro-environmental behavior.

In Pathway 1, the factor of gender is viewed as the starting point of the transmission mechanism. The traditional understanding of male household heads makes men more susceptible to group pressure, and they are therefore more likely to participate in training and more eager for incentives. The increase in training frequency can improve the learning quality and learning ability of individuals, make it easier to use clean technology energy for production activities, and leads to the improvement of the production intensification level, thus producing pro-environmental behavior.

Pathway 2: (1) Tea farm area → intensive management → pro-environmental behavior. (2) Tea farm area → ecological maintenance cost, production restrictions → environmental regulations → concern for the well-being of others → pro-environmental behavior.

The tea farm area is an important means of production, and pro-environmental behavior can be directly produced by the intensive management of tea farms through intensive cultivation. With the increase in tea farm area, ecological maintenance costs and tea production restrictions also increase, affecting the compliance degree of tea farmers with environmental regulations. The effective constraints of environmental regulations can encourage tea farmers to care about the well-being of others, thus producing pro-environmental behavior.

Pathway 3: Ecological compensation → ecological maintenance cost, production restrictions → environmental regulations, social norms → concern for the well-being of others → pro-environmental behavior.

Pathway 3 emphasizes the different influences of ecological compensation on ecological maintenance cost and production restrictions. Ecological compensation can mitigate the ecological maintenance cost to a certain extent, but this compensation will also aggravate the tea farmers’ concerns about production restrictions. In addition, ecological compensation policies will make tea farmers pay more attention to the mandatory social norms issued by the government or the village department, thus producing pro-environmental behavior.

Pathway 4: Market orientation → Clean technology application, intensive management → pro-environmental behavior.

Market orientation is closely related to the income of tea farmers. The market demand for intensive management and clean technology application can increase the willingness of tea farmers to apply new production methods. The application of new technologies can provide positive feedback as an incentive indicating that the behavior changes of tea farmers can bring considerable economic profits, thus improving the levels of their pro-environmental behavior.

Pathway 5: Environmental responsibility → ecological maintenance cost, production restrictions → environmental regulations, social norms → concern for the well-being of others → pro-environmental behavior.

Tea farmers’ sense of environmental responsibility can make them less sensitive to ecological maintenance cost and production restrictions. Consequently, they will pay more attention to environmental regulations and social norms, obey these regulations and norms more faithfully, and actively have concern for the well-being of others, thus producing pro-environmental behavior.

4.3. Results for Group 2

4.3.1. Factors Influencing the Pro-Environmental Behaviors of Group 2

For the policy and historical reasons mentioned above, we further restricted the sample variables and observed differences in the results. Some samples were excluded from Group 1 according to the preset conditions, and the remaining 271 sample variables were termed Group 2. The influencing factors of the level of pro-environmental behaviors of Group 2 and their marginal contribution rates changed, and the results are shown in

Table 9. Regression results of factors influencing pro-environmental behavior in Group 2 and changes in marginal contributions.

Variable	Coefficient 2 (t-Value)	Change in Significance Level	D (Std.Err.)	C (Std.Err.)	B (Std.Err.)	A (Std.Err.)
E2	0.274 ** (2.21)	Unchanged	−0.018 ** (0.008)	−0.009 ** (0.004)	0.004 (0.002)	0.024 ** (0.010)
E4	0.172 (1.42)	No longer significant
E5	0.235 ** (2.27)	Increased	−0.016 ** (0.007)	−0.008 ** (0.003)	0.003 (0.002)	0.020 ** (0.009)
E6	0.447 *** (3.21)	Unchanged	−0.030 *** (0.009)	−0.015 ** (0.006)	0.006 * (0.003)	0.038 *** (0.012)
S1	0.578 *** (2.77)	Increased	−0.039 *** (0.015)	−0.019 ** (0.008)	0.008 * (0.004)	0.050 *** (0.018)
S2	0.365 ** (2.49)	Decreased	−0.024 ** (0.010)	−0.012 ** (0.005)	0.005 (0.003)	0.031 *** (0.012)
S3	−0.351 *** (−3.26)	Unchanged	0.023 *** (0.008)	0.012 *** (0.004)	−0.005 ** (0.002)	−0.030 *** (0.009)
S4	0.242 *** (2.60)	Unchanged	−0.016 ** (0.007)	−0.008 *** (0.003)	0.003 ** (0.002)	0.021 ** (0.008)
S5	−0.036 (−0.19)	No longer significant
S6	0.520 *** (3.19)	Increased	−0.035 *** (0.011)	−0.017 *** (0.006)	0.007 * (0.004)	0.045 *** (0.013)
P2	0.791 *** (5.90)	Unchanged	−0.053 *** (0.007)	−0.027 *** (0.008)	0.011 ** (0.005)	0.068 *** (0.010)
P4	0.426 ** (2.24)	Unchanged	−0.028 ** (0.012)	−0.014 * (0.008)	0.006 (0.004)	0.037 ** (0.016)
P6	0.885 *** (6.39)	Unchanged	−0.059 *** (0.011)	−0.030 *** (0.007)	0.012 (0.005)	0.076 *** (0.010)
Gender	−1.040 *** (−3.30)	Increased	0.029 *** (0.012)	0.015 *** (0.008)	−0.013 (0.006)	−0.039 *** (0.015)
Tea farm area	−0.014 *** (−3.11)	Unchanged	0.001 *** (0.000)	0.000 *** (0.000)	0.000 ** (0.000)	−0.001 *** (0.000)
Whether receiving ecological compensation	−0.413 ** (−2.09)	Decreased	0.028 ** (0.013)	0.014 * (0.008)	−0.006 (0.004)	−0.036 ** (0.017)
Frequency of participation in technical training	−0.43 (−1.42)	No longer significant

Note: *** p < 0.01, ** p < 0.05, * p < 0.1; A, B, C and D represent the grading of environmentally friendly behavior.

.

Place attachment (S1) and the establishment of Wuyishan National Park (S6) were newly-added influencing factors and significant at the 1% level. The significance levels of production restriction (E5) and gender both increased from being significant at the 10% level to being significant at the 1% level. The significance levels of social norms (S2) and ecological compensation both decreased from being significant at the 1% level to being significant at the 5% level. Ecological maintenance cost (E4), environmental regulation (S5), and training frequency were no longer significant. The factor of concern for the well-being of others (P6) contributed most significantly to adjusting the level of pro-environmental behaviors at levels A, C, and D, indicating the marginal importance of this variable.

4.3.2. Transmission Pathways of Pro-Environmental Behaviors of Group 2

There were more levels for the influencing factors of Group 2, and the factors were mainly divided into eight levels, namely L1 = {gender}, L2 = {tea farm area, group pressure, national park establishment}, L3 = {production restriction}, L4 = {place attachment}, L5 = {social norms}, L6 = {market orientation, environmental responsibility}, L7 = {clean technology applications}, and L8 = {ecological compensation, intensive management, incentive expectation, concern for the well-being of others}. Based on the hierarchical decomposition results, the relational structure diagram of the influencing factors of tea farmers’ pro-environmental behaviors in Wuyishan National Park was obtained (Figure 3).

The pathways of pro-environmental behavior of Group 2 have some similarities to those of Group 1, but there are also some heterogenous characteristics. The most special characteristic of the transmission pathways of Group 2 is that the pathways of the pro-environmental behavior of Group 2 all start from the establishment of the national park. The pathways are as follows:

Pathway 1: National park establishment → pro-environmental behavior.

Pathway 2: National park establishment → concern for the well-being of others → pro-environmental behavior.

Pathway 3: National park establishment → Ecological compensation → pro-environmental behavior.

Pathway 4: National park establishment → production restrictions → place attachment → social norms, environmental responsibility → concern for the well-being of others → pro-environmental behavior.

First, the establishment of national parks has a direct effect on the level of pro-environmental behavior of tea farmers. Second, the connotation concept of national park establishment is consistent with the ecological concept of pro-environmental behavior, and the establishment of national parks can enhance the awareness of tea farmers about the well-being of others and improve the level of pro-environmental behavior. Third, the establishment of supporting ecological compensation policies with national park establishment can reduce the cost of tea farmers’ pro-environmental behavior on the one hand, but on the other hand, unreasonable ecological compensation standards may affect the adoption of pro-environmental behavior. In addition, the establishment of Wuyishan National Park can cause greater production restrictions for tea farmers within the national park. The restrictions for tea production can impair the place attachment of tea farmers or cause their “relative deprivation”. However, even if the level of place attachment fluctuates, place attachment can still enhance the attention of tea farmers to the changes in their living environment so that they can better adapt to social norms and improve their environmental responsibility. As a result, these tea farmers care more about the well-being of others and improve their level of pro-environmental behavior.

5. Discussion and Conclusions

5.1. Key Findings and Policy Implications

5.1.1. The Duality of Market Orientation: The Synergistic Role of Profit Motivation and Institutional Constraints

This study finds that market orientation significantly influences the pro-environmental behavior of tea farmers. The research demonstrates that tea farmers with higher market expectations are more inclined towards pro-environmental behaviors, thereby confirming market-based incentives. However, the study also uncovers a more complex market mechanism, where downstream buyers impose strict constraints through chemical testing and punitive contracts, which are combined with softer incentives like geographical indication certification for tea. This “carrot and stick” approach effectively mitigates information asymmetry within the supply chain and enhances information transfer efficiency. However, in the context of smallholder economies, this model creates tension due to the vulnerability of tea farmers. Farmers may face an “ecological poverty trap” due to the costs of upgrading equipment. To address this challenge, it is recommended to consider the establishment of an ecological behavior credit bank, where actions like adopting clean technologies and participating in training could be converted into tradable credits. Furthermore, the establishment of an ecological tiered certification system that links the intensity of environmental efforts to market price gradients could further incentivize sustainable practices.

5.1.2. The Governance Paradox of National Parks: Spatial Regulation and the Transformation of Social Capital

This study further validates the boundary effect and identifies the role of tea culture. Although tea farmers in the core areas of the national park face stricter production restrictions, they form a compensatory mechanism through ecological identity reconstruction by viewing national park certification as a quality signal that allows them to command a premium price. This finding challenges the conventional assertion that production restrictions inevitably lead to negative emotions and shows how institutional identities can be transformed into cultural capital. On the periphery of the national park, however, tea farmers’ levels of pro-environmental behavior decline significantly, reinforcing the gradient effect of spatial governance. This suggests that national park policies need to design a buffer zone governance model that provides ecological compensation based on geographic distance to avoid weakening constraints or incentives for farmers in marginal areas due to lax management. Therefore, a concentric governance system should be designed with differentiated ecological compensation based on geographic distance and cultural capital.

5.1.3. Reconstructing Environmental Ethics: From Individual Rationality to Relational Rationality

The study reveals that the marginal effect of “concern for the well-being of others” is the most significant, demonstrating the unique impact of China’s rural relational communities on pro-environmental behavior. Tea farmers use relations networks to establish informal mechanisms for promoting pro-environmental practices, with this relational governance playing a critical role in driving pro-environmental behavior. This finding emphasizes that, within the Chinese context, environmental behavior is not solely the result of rational decision-making but also serves as a means of relationship reproduction. The study also finds that the tolerance effect in close-knit societies leads to an insignificant social sensitivity, pointing to a complex relationship between traditional ethics and modern environmental governance. As a result, the effective utilization of relational governance tools, the cultivation of networks of ecological local elites, and the leveraging of hierarchical social structures to accelerate the dissemination of sustainable technologies are imperative. However, this approach must be complemented by the implementation of anonymous supervision mechanisms to mitigate the protective behaviors that may emerge within close-knit communities.

5.1.4. Gendered Ecological Labor: Implicit Environmental Governance in the Productive Domain

The study further corroborates the substantial impact of gender on tea farmers’ pro-environmental conduct. Female tea farmers demonstrate higher levels of pro-environmental behavior compared to their male counterparts, which may be attributed to the clean aesthetics observed in domestic labor, extending household ethics into the production field and forming a distinctive form of eco-mothering. This finding calls into question the conventional gender-neutral assumption in environmental behavior studies and prompts a reevaluation of the implicit contribution of reproductive labor to environmental governance. However, extant ecological compensation policies have not adequately captured this latent contribution, resulting in diminished policy efficacy. Consequently, there is a necessity for policies to prioritize the environmental governance value of gendered labor. One potential solution is the establishment of family ecological accounts to quantify women’s environmental labor as compensatory credits. This would serve to strengthen policy support for women’s pro-environmental behaviors.

5.2. Comparative Analysis with Global Studies

“Pro-environmental behavior” is regarded as an extension of “pro-social behavior” in the relationship between human beings and environmental resources [58]. It is generally believed that the concept of “pro-environmental behavior” was first proposed by Hines et al. [59]. These authors explored the possible pattern of responsible environmental behavior from the perspectives of individual environment-related knowledge, individual environmental attitude, and self-responsibility, and believed that human beings can make positive changes in the environment by improving individual cognition and actively intervening in the ecosystem and ecological structure. The origin of pro-environmental behavior in China can be traced back to the thoughts of “Heaven, earth, and human beings are born together” and “All things and I are a community with a shared future” of Zhuangzi, an ancient Chinese philosopher [60]. His philosophy emphasizes conformity to nature (Taoist culture) and opposes artificial creation and coercion [61]. Numerous modern environmental movements and individual pro-environmental actions are inspired and supported by such traditional thinking [62].

In recent years, the global literature on pro-environmental behavior has increasingly emphasized the role of nature connectedness, social identity, and behavioral spillover mechanisms. This study focuses on the influencing factors and underlying mechanisms of pro-environmental behavior among tea farmers in Wuyishan National Park. The findings show both consistencies and differences compared with existing studies. Capaldi et al. [63] conducted a meta-analysis demonstrating that nature connectedness is positively associated with happiness and well-being. While their analysis focused on conceptualizing nature contact at an individual and psychological level, our findings highlight place-based cultural identity and land dependency as context-specific expressions of nature connectedness in rural China. Martin et al. [64] further distinguished between nature contact and nature connectedness, showing that emotional connectedness predicts PEB more robustly. This aligns with our observation that tea farmers’ long-term residence and multigenerational ties to the tea-growing landscape increase their responsiveness to environmental messages, especially when reinforced by national park branding and ecological regulations.

Fritsche et al. [65] proposed the Social Identity Model of Pro-environmental Action, arguing that individuals are more likely to adopt sustainable behaviors when such behaviors align with group identity and norms. Our findings support this framework, but also emphasize the village-level embeddedness of group pressure and social norms, which are shaped by government policy, cooperative dynamics, and interpersonal networks. Thøgersen and Crompton [66] critiqued the reliance on “simple and painless” behavioral nudges in environmental campaigning, warning that such strategies may not lead to durable behavioral change. Consistent with this, we found that group conformity alone is insufficient to sustain high-level PEB. Rather, internalized environmental responsibility, market orientation, and concern for others’ well-being are more reliable drivers of change.

Importantly, recent empirical studies in other developing country contexts provide further resonance. For instance, Ataei et al. [67], using Norm Activation Theory in Iranian agriculture, showed that awareness of consequences and ascription of responsibility significantly predict environmental sustainability behavior. Our study similarly finds that perceived responsibility and anticipated reputational or communal outcomes are major contributors to behavioral change. Likewise, Menatizadeh et al. [68] explored cognitive and behavioral factors underlying water conservation among Iranian farmers, emphasizing the role of environmental awareness, perceived control, and social commitment. These cognitive–behavioral pathways mirror our model’s emphasis on incentive expectations, environmental regulation, and clean technology adoption, particularly under conditions of limited resources and policy pressure.

Compared to these international cases, our study is unique in integrating institutional mechanisms such as national park systems, market incentives, and relational governance structures into the analysis of farmer behavior. It contributes to bridging the gap between individual cognition-oriented models and institutional or collective action frameworks, offering a holistic explanation in complex protected area settings. In conclusion, our research contributes a place-sensitive, policy-coupled, and culture-informed perspective on how environmental behaviors emerge and evolve in rural China. This enriches the global understanding of environmental governance in agricultural frontiers.

5.3. Limitations and Future Research Directions

This study employs cross-sectional data to quantitatively analyze the various factors influencing pro-environmental behavior among tea farmers. However, the use of cross-sectional data limits the strength of causal inferences. Future research endeavors may benefit from the incorporation of longitudinal data or experimental designs to enhance the validity of causal conclusions. Due to constraints in the size of the questionnaire and in field research conditions, this study did not explore the cost–benefit structure of pro-environmental behaviors. In practice, smallholder farmers face not only technical choice dilemmas but also the challenge of balancing livelihood and risk. Consequently, subsequent studies should prioritize the identification of the “critical threshold” of incentives, determining the most efficacious combinations of market and policy interventions in stimulating pro-environmental behaviors among diverse farmer demographics. The integration of participatory cost assessment and stratified simulation experiments would facilitate the development of incentive policies with enhanced operational feasibility.

Furthermore, while this study addresses the issue of eco-migrant identity, particularly in the context of their roles and the restructuring of interests within national park management, it has not thoroughly incorporated this issue into the framework of environmental justice theory. As the national park system progresses, eco-migrants often face conflicts in identity and inequities in resource distribution. Future research should prioritize examining the socio-economic conditions of eco-migrants, their sense of environmental governance, and their level of participation. This would illuminate the ethical dimensions of policy implementation, particularly in relation to the identity and social exclusion of marginalized groups. Such an endeavor would contribute to the development of a more comprehensive framework for environmental justice.

While this study provides empirical evidence that lends support to the relevant theories, it is important to acknowledge the significant contextual differences in the study of protected area farmer behavior from a global perspective. These contextual differences pertain to the impact of varying national institutional backgrounds and the mechanisms of cultural capital transformation on farmers’ pro-environmental actions. The region under study functions under a collective land tenure system, which facilitates effective spatial governance. However, in regions with private land ownership, the applicability of existing policy tools may need to be reassessed. Therefore, while the findings offer valuable insights for understanding pro-environmental behavior within the Chinese national park context, their generalizability to other geographic or institutional settings remains limited. Future research is encouraged to test and refine the proposed analytical framework in diverse protected areas, particularly in regions with different governance systems, land tenure arrangements, and sociocultural contexts.

5.4. Conclusions

This study explores the multifaceted factors that influence tea farmers’ pro-environmental behavior, and the underlying mechanisms through which their actions are transmitted, with a particular focus on the interaction between market orientation, national park policies, social norms, and psychological factors. This study demonstrates that tea farmers’ pro-environmental behaviors are not solely driven by market incentives, but are also significantly influenced by institutional constraints, cultural capital, and psycho-social mechanisms. Empirical research on tea farmers in Wuyishan National Park was undertaken to provide multi-dimensional insights to promote the theoretical development and practical application of farmers’ environmental behavior.