Factors Influencing the Conservation Intentions of Visitors to a World Heritage Site: A Case Study of Libo Karst

Abstract

The behavioral intention of tourists to protect the environment is a popular topic in tourism geography research. Understanding the mechanisms that influence the post-tour behavioral intentions of heritage site visitors can effectively promote the generation of positive behavioral intentions for the sustainable development of heritage sites. Starting from the “cognitive–emotional–behavioral” path of attitude theory, we introduced environmental knowledge awareness (EKA), geomorphological value perception (GVP), place attachment (PA) and behavioral intention (BI) variables to construct a model of the influence mechanism of tourists’ behavioral intentions towards conservation in heritage sites; we test this model on the Libo World Natural Heritage Site, a karst in southern China, as an example. The results show the following: (1) environmental knowledge awareness (EKA) is the basis of place attachment (PA) and behavioral intention (BI), and place attachment (PA) plays an influential mediating role. (2) Geomorphological value perception (GVP) has a positive effect on behavioral intention. (3) The “GEPB” conceptual model fits the “cognitive–emotional response–behavioral intention” chain of tourists in natural world heritage sites and passes the empirical test; (4) The direct positive effect of place attachment on behavioral intention is significant as an essential emotional factor that promotes behavioral intention. This study provides an empirical foundation for the synergistic development of conservation and tourism in World Heritage Karst Sites. It also discusses theoretical and managerial implications based on the findings and provides directions for future research.

1. Introduction

World Heritage (WH) is the most precious treasure of humanity, has Outstanding Universal Value (OUV), is an irreplaceable source of life and inspiration [1,2], and is the “identity card” of a nation. As of 9 March 2023, 1154 World Heritage sites have been inscribed worldwide, including 897 cultural heritage, 218 natural heritage, and 39 mixed heritage sites. Conradin et al. stated that World Heritage sites (WHSs) are places of significant conservation and iconic value, and are essential for tourism and regional economic development [3]. At the same time, issues related to the WHSs have attracted great academic interest. To date, China has 56 WHSs located in 30 provinces, which have contributed to the balanced development of tourism, and especially to the economic development of poor areas. However, the development of tourism over the past 20 years, which has benefited from the WHSs, has also had a major impact on their conservation and use. WHSs are currently facing numerous threats from human pressures and natural factors, such as tourism, development and construction projects, the overuse of natural resources, and natural disasters [4,5]. Therefore, how to ensure the authenticity and integrity of WH properties and effectively protect them during development is an issue that deserves serious study.

The conservation objective of the World Natural Heritage Sites (WNHSs) is to ensure that the OUV and integrity of the heritage site are preserved and developed in perpetuity. Presently, natural heritage sites regularly carry the OUV of WNHSs, which are rich in geological, hydrological, and biological resources. Among the existing WNHSs, there are 93 sites that meet the criteria (viii), 33 of which are related to karst. Karst is a topography with peculiar geomorphological and hydrological characteristics, due to the rock’s high solubility and the secondary pore space developed [6]. Over thousands of years, the process of dissolution leads to typical karst landscapes with notable surface features, including poljes, dolines, sinkholes, towers, caves and shafts, dry valleys, and underground rivers [7,8]. The peculiar karst landscape and its caves have strong cultural, historical, scientific, aesthetic, and recreational values [9]. In order to preserve the WHSs and bring their influence into full play, many scholars at home and abroad have studied the functions, values, types, conservation and management of WHSs and their buffer zones [10,11,12,13]. The karst landscape, as an intrinsic part of the value of the Libo WHKS, is a core element of the heritage site’s tourism development. Moreover, tourists, as the main element of heritage tourism activities and as the stakeholders in heritage conservation, have a direct impact on the ecological, conservation and environmental carrying capacity of heritage sites. For tourists, there are extremely few who have the professional scientific knowledge to understand the formation and evolution of karst landscapes and recognize their scientific value. Large numbers of visitors to natural heritage sites are sightseeing types, who come to heritage sites for the limited purpose of sightseeing and have a lower perception of the value of World Heritage Karst Sites (WHKSs). Therefore, does the visitor’s perception of the value of the karst landscape affect their conservation intentions? What are some of the factors that influence a visitor’s intention to act in conservation? How are these factors related and how do they affect behavioral intentions?

Tourist activities may have an impact on the authenticity and integrity of WNHSs and modify the value of the WHSs itself [14]. Therefore, conservation actions must start from WNHSs’ value and the conservation of the carriers of their value (mountains, water, forests, fields, and lakes) in order to achieve the perpetual preservation of the heritage [15]. The natural landscape of a heritage site, especially the special landscape, scientific significance, value perception, and emotional attachment to it are all factors that create “place attachment” among tourists. Environmental knowledge perceptions and the place attachment of tourists have a positive impact on the conservation and presentation of the values of WHSs [16,17,18,19,20,21]. Researchers have studied value orientation, visitor perceptions, behavioral intentions, and heritage conservation from the perspective of tourists [22,23,24,25], and have also analyzed the impact of visitor satisfaction and loyalty on heritage conservation [26]. The perception of the geomorphological value of the WHSs is an element to assess the OUV attractiveness of the site, determine whether it affects visitors’ satisfaction, and determine their heritage conservation intentions [27].

This study examines the factors influencing tourists’ willingness to conserve heritage from the perspective of tourists, using their knowledge of the WHSs’ environment as an antecedent variable, and using geomorphological value perception and local attachment as mediators. Based on the existing literature and the analysis of SPSS25.0 and AMOS27.0, we explore the degree of influence of geomorphological value perception, environmental knowledge perception, local attachment, and conservation behavior on tourists’ intention to conserve heritage sites through the actual exploration of WHSs, the collection and analysis of questionnaire data, and the analysis of the main influencing factors; this is in order to objectively understand the problems that currently exist in heritage sites. Targeted recommendations for heritage site conservation and tourism development are made to enhance the attractiveness and conservation of heritage sites, and scientific references are provided to WHSs for the conservation and tourism development of karst-like heritage sites. This research is an important contribution to the recognition and promotion of the importance of karst heritage landscapes in terms of social development, as well as to strengthening education that propounds that tourism to WHKSs should be based on the karst landscapes and on sustainability.

2. Theoretical Background and Hypotheses

Environmental psychologists have been studying the relationship between the natural environment and human behavior for decades. Mehrabian and Russell (M–R) proposed a valuable theoretical model to demonstrate the influence of the physical environment on human behavior [28]. Through the S–O–R structure, the model assumes that external environmental stimuli (S) elicit emotional responses (O) from individuals, and these emotional responses further induce approach or avoidance behaviors (R) towards the environment (Figure 1). By exploring the mechanism of tourists’ “perception–emotion–behavior” behavior, tourists’ perception of heritage sites’ geomorphological value and environmental knowledge are considered as components of M–R model S. The attached particles can be thought of as components of the M–R model O, and environmental behavioral intention can be considered as components of M–R model R. Generally, visitors acquire knowledge through environmental education, recognize the value of WNHSs, and then develop emotions, such as place attachment (PA), which eventually determine heritage conservation behavioral intention (HCBI).

2.1. Environmental Knowledge Awareness

Tourist environmental knowledge is an indication of the level of ecological awareness among tourists, and if they have extensive environmental knowledge, they will be more concerned about environmental issues in their tourist destinations. Tourism development focuses on minimizing the impact on the local environment, conservation, and cultural heritage, and promoting income generation and community development while promoting environmental education and political awareness [29,30,31]. The main aim of environmental education in WNHSs is to educate and explain nature and the environment to the general public for the purpose of environmental protection, in order to improve the environmental literacy of citizens [32]. Li and Yin et al. proposed the influence path of local identity–nature, empathy–pro-environmental behavior and local dependence–environmental education, perception–pro-environmental behavior, based on the environmental education perceptions of tourists in ecotourism scenic spots [33]. Aguilar also proposed that the major purpose of environmental education is to alter individual environmental attitudes and environmental behaviors, and seek to alter individual environmental behaviors while driving group synergistic environmental behaviors [34]. Environmental awareness is an educational awareness that visitors acquire during their interactions with the natural environment, with some guidance on environmental conservation.

An essential venue for environmental education is the environment itself. Although environmental education is a slow-acting model of environmental protection, it is more effective in reaching people’s hearts and minds [35]. Environmental education can be a key tool to address environmental issues because it is an effort to achieve the goal of environmental protection and conservation [36,37]. In contrast, heritage tourism is one of the most effective ways to reconcile the contradictions between heritage preservation and utilization, due to its less destructive nature, outstanding heritage education function and ability to achieve economic, social and environmental benefits at the same time. In fact, education is considered an indispensable requirement if we want to promote the sustainable development of WNHSs [38]. In general, environmental education provides a high level of information and awareness about environmental problems and solutions, which leads to sustainable and eco-friendly behaviors [39,40]. Environmental education increases understanding and sensitivity to environmental issues, broadens knowledge and contributes to the development of a positive attitude towards the preservation of WHSs.

2.2. Heritage Geomorphology Value Perception

The value of heritage represents the quality and character of heritage and is one of the central themes of heritage discourse and practice. Although the concept of value is central to WHSs, little is known about it [41]. The karst heritage landform landscape is developed in a long-term, specific geological environment and is a landscape attribute that combines aesthetic, ecological, scientific, and economic values, and has particular ornamental, science education, fitness, and special tourism functions [42]. Visitor value perceptions, i.e., how visitors feel at a heritage site and their perceptions of what they hope to gain before arriving at the site, consist of effective, functional, and social values [43,44]. Drivers of value include experience/service quality, motivation, and engagement [45,46]. A large body of empirical evidence confirms the importance of perceived value as an essential antecedent of satisfaction and behavioral intention, as well as the positive correlation between visitors’ perception of value and loyalty [47,48,49,50,51,52].

As a spatial vehicle for heritage tourism and the development of displays of heritage, few scholars have focused on the transmission of heritage values among tourists. WNHSs have an elevated ornamental value compared to other tourist sites, but still suffer from poor integration, low perception and lack of interaction among visitors. The perception of the value of a heritage site at the time of tourists’ visit is an essential factor in determining their willingness to revisit, behavioral intentions, satisfaction and awareness of their heritage preservation responsibilities. Thus, a correct perception of heritage value is fundamental for dealing with the relationship between preservation and the utilization of heritage sites, and also affects visitor attitudes towards heritage and environmental conservation behavior.

2.3. Place Attachment

Place attachment is an emotional attachment that arises through an individual’s interaction with a place, and is also an identity sub-structure of perceptions, attitudes, and values about a specific region or place [53,54]. Place attachment can be categorized as an attachment, affect, identity, and social connectivity, suggesting that these dimensions significantly influence place satisfaction and environmental behavior. The purpose of studying tourists’ sense of place is to explore the meaning and value of how tourists perceive tourism destinations, and the results of existing studies have shown that place attachment has a positive effect on resource conservation and protection in tourism destinations [55]. While tourism serves as an essential way for humans to perceive value and understand the natural environment, the natural environment is the intersection of the individual and the place, and has great symbolic significance for the visitor. The psychological attachment people have to their local area will motivate visitors to act more responsibly in the environment, for example, by volunteering to pick up litter and protect animals.

Kyle et al. used emotional attachment items to measure psychological commitment factors (e.g., location involvement–location identification) and concluded that location attachment is a determinant of behavioral loyalty [56]. Nan et al., taking Sanqingshan National Park as an example, suggested that local attachment plays a key role in promoting heritage conservation and improving community satisfaction and well-being [16]. Place attachment, as an essential basis for a visitor’s intention to protect the environment, has a significant positive effect on a visitor’s environmentally responsible behavior and appreciation of the destination. In addition, place attachment, as an emotional attachment and connection to nature, has been found to be a significant predictor of pro-environmental behavioral intentions in several studies [57,58].

2.4. Behavior Intentions

Recognizing the need for tourist retention and sustainable tourism development, research on tourists’ behavioral intentions is increasingly emerging [59,60]. Revisiting and recommending behaviors is now considered a central issue for future tourism development [61]. Much of the research on tourists’ behavioral intentions has focused on studying the relationship between behavioral intentions and other tourism elements, such as destination image, perceived value, satisfaction, and behavioral intentions [27,62,63]. Tourism scholars normally define tourists’ behavioral intentions as the “willingness to revisit” and “willingness to recommend” [63]. Chen argued that behavioral intention is a tourist’s judgment about the possibility of revisiting the same destination or the willingness to recommend it to others [61].

In the field of tourism, most scholars use behavioral intentions as the final variable in the decision-making process to predict tourists’ future behavioral tendencies, and most of them use tourists’ willingness to revisit and their willingness to recommend sites to measure the actual behavior of tourists [64]. Tourist loyalty can be defined and assessed in both an attitude and a behavioral way. In practice, behavioral loyalty is difficult to measure, so most researchers use behavioral intention, i.e., intentional loyalty, as a trade-off for behavioral loyalty [65], and the degree of destination loyalty is normally reflected in tourists’ willingness to revisit and willingness to recommend [61,66].

2.5. Assumptions and Conceptual Models

The behavioral intention of visitors to aid in heritage preservation is the starting point for the preservation of heritage sites. Tourism, the main use of WNHSs, is also the main threat posed by human activity at numerous natural heritage sites. In terms of tourism activities in WHSs, tourism attractiveness, cognitive and emotional images, perceived value, recreational involvement, conservation commitment, and environmental orientation are essential factors influencing the environmentally responsible behaviors of tourists [67]. The core attractions of WHSs OUV and their supporting attractions, such as tourism services and local communities, have a positive impact on attraction attachment, perceived value, and tourists’ attitudes and behaviors towards destination conservation [17,68,69]. Chubchuwong studied destination attachment and property ownership in nature tourism and showed that place attachment has a positive effect on environmental behavioral intentions [21]. Heritage education motivates most stakeholders to better protect the world’s natural and cultural heritage sites by interpreting and promoting the OUV value perceptions and environmental knowledge perceptions of heritage [70]. Several scholars have studied the environmental behavior of tourists and their behavioral intentions to protect WHSs based on the theory of planned behavior (TPB), showing that attitudes have a significant effect on behavioral intentions [71,72]. The present study is based on the stimulus–organism–response (S–O–R) framework of the environmental stimulus–emotional state–behavioral response theory, i.e., tourists are stimulated by external sources and their internal states are modified, leading to positive environmental behaviors.

Hypotheses 1 (H1). 

The higher the awareness of environmental knowledge among heritage visitors, the higher their perceived geomorphological value.

Hypotheses 2 (H2). 

The higher the heritage visitor’s awareness of the environment, the higher their place attachment.

Hypotheses 3 (H3). 

The higher the perceived environmental knowledge of heritage visitors, the more positive their behavioral intentions towards environmental protection.

Hypotheses 4 (H4). 

The higher the geomorphological value is perceived by heritage visitors, the higher their place attachment.

Hypotheses 5 (H5). 

The higher the geomorphological value is perceived by a heritage visitor, the more positive their behavioral intention to protect the environment.

Hypotheses 6 (H6). 

The higher the place attachment of heritage visitors, the more positive their behavioral intention to protect the environment.

Hypotheses 7 (H7). 

Geomorphological value perception has a mediating role in the influence of environmental knowledge perception on behavioral intentions.

Hypotheses 8 (H8). 

Local attachment has a mediating role in the influence of environmental knowledge perception on behavioral intention.

Hypotheses 9 (H9). 

Geomorphological value perception and place attachment has a mediating role in the influence of environmental knowledge perception on behavioral intentions.

The proposed theoretical model is shown in Figure 2.

3. Study Area and Methods

3.1. Study Area

The Karst landscape is one of the most outstanding landscapes on Earth. Southern China Karst is the largest continuous area of karst landforms in the world, and is one of the most typical areas of tropical and subtropical karst landforms in China and the world. Its heritage sites are the best preserved and most complete evolutionary sequences of subtropical carbonate rocks formed by water erosion and dissolution in the world; these have a high natural aesthetic value, excellent geomorphological value, a high potential for ecological processes, and a great level of biodiversity [73]. Not only are they of great scientific interest, but numerous karst landscapes exhibit extraordinary natural beauty. The karst mountainous region of southern China, centered on Guizhou, is one of the largest and most concentrated contiguous karst regions in the world, with the most typical and complex karst development areas and the richest landscape types.

This paper is a study of the Libo WHKS, located in Guizhou Province, China (Figure 3). The Libo WHKS contains numerous towering conical peaks and deep sunken funnels, as well as underground rivers and lengthy underground caves, and its conical karst is also a model site for other karst sites of its kind in the world. The terrain is high in the west and low in the east, and it is a typical representative of cone-shaped karst in the transition zone between the Guizhou plateau and the Guangxi lowland, which stands out in the world because of the complete morphological spectrum that is formed by the gradual transition from plateau karst to lowland karst. With an average elevation of 747 m, but varying between 385 and 1109 m, it is a subtropical karst plateau with large internal differences and is deeply cut by rivers. The heritage site covers an area of 295.618 hm², with a buffer zone of 434.98 hm². The Libo WHKS includes two areas of the Maolan National Nature Reserve and the Zhangjiang National Key Scenic Area [74]. The heritage site demonstrates the mutual evolution and progression of the peak cluster–peak fore landscape.

3.2. Survey Instrument

In this study, a self-administered questionnaire was used to collect empirical data from tourists in a WHKS (Libo) in southern China. The questionnaire questions were designed based on a review of the literature and specific characteristics of heritage tourism. The questionnaire was pre-tested and revised to ensure its validity. The questionnaire used in this survey consists of two parts. The first one is related to the economic and social characteristics of the tourist, including demographic characteristics (gender, age, occupation, education, monthly personal income, place of residence) and additional information (place of origin, number of trips, travel routes, knowledge, purpose of travel, tourist facilities recorded), thus determining the social attributes of the tourist. The second part is an investigation of the visitor’s perceptions, emotions, and behavioral intentions. In order to minimize excessive bias in the data when using the structural equation model SEM, the research scales in the questionnaire were measured on a five-point Likert scale corresponding to measures 1–5 (1 = heavily disagree; 5 = fully agree), except for the respondent information, which was measured on a categorical scale (Figure 4).

3.3. Sampling Instructions

The questionnaire was distributed to domestic visitors to the WNHSs (few international visitors during the New Crown epidemic) by eight geography PhD and MSc students trained in research questionnaire methodology. The survey was conducted between 1 October and 10 October 2021, and the questionnaires were distributed at the main entrances and exits of the Daqikong, Xiaoqikong Scenic Area and Maolan national nature reserve. To improve the quality and recall of the questionnaire, visitors were given the questionnaire after visiting the two main exits of a scenic area. According to the convenience sampling method, 600 questionnaires were distributed and after excluding incomplete or illogical questionnaires, we obtained 558 valid questionnaires with a valid rate of 93%. The research team interviewed visitors and other stakeholders on issues such as heritage conservation and heritage tourism development in order to gain a deeper understanding of conservation and tourism development at the WNHSs.

4. Results

4.1. Descriptive Statistical Analysis

As shown in

Table 1. Descriptive statistics of the demographic characteristics of the samples.

Demographic	Type	Frequency	Percentage (%)	Demographic	Type	Frequency	Percentage (%)
Gender	Man	245	43.9	Dwell time (day)	1	69	12.4
	Women	313	56.1		2–3	376	67.3
Age	18–25	164	29.4		>3	113	20.3
	26–45	251	45.1	Place of residence	In Guizhou province	319	57.2
	46–60	122	21.7		Other provinces	239	42.8
	>60	21	3.8	Mode of travel	Travel agency	49	8.8
Education level	Middle School and below	43	7.7		Group trips	69	12.4
	Secondary education	254	45.5		Self-driving travel	248	44.4
	Bachelor’s degree	221	39.6		public transport	166	29.7
	Master’s degree or above	40	7.2		Other	26	4.7
Average annual income (CNY)	<10,000	24	4.3	Career	Students	162	29.0
	10,000–60,000	315	56.5		Workers	50	9.0
	>60,000	219	39.2		Service industry operators	72	12.9
Frequency of interaction	First time	355	63.6		Professional person	140	25.1
	Second time	149	26.7		Retirees	92	16.5
	Three times or more	54	9.7		Other	42	7.5

, the demographic characteristics of the visitors include nine attributes such as gender, age, education level, occupation, average annual income, place of residence, number of visits, length of stay and mode of travel. Of these, 43.9% were male and 56.1% were female; the largest age group was 26–45 years old (45.1%), in which 29.4% were 18–25 years old, 21.7% were 46–60 years old, and 3.8% were over 60 years old. Most of the respondents had secondary education, with 39.6% having undergraduate education; in terms of average annual income, the annual income of 56.5% of visitors was concentrated between RMB 10,000–60,000, which is in the middle range of China’s per capita annual income; 63.6% of visitors were visiting the first time, 57.2% of visitors came from the province, 67.3% of visitors said they were willing to stay in the heritage site for 2–3 days, and the mode of travel was mainly concentrated on driving oneself (44.4%).In terms of occupational characteristics, 29.0% of visitors were students, mainly related to the time of visit.

In addition, as part of the South China Karst Heritage Site, the Libo WHKS has unique geomorphological and aesthetic values. In total, 49.7% of the visitors interviewed learned about the Southern China Karst Heritage Site through WeChat, Weibo and video; 88.7% of visitors found the uniqueness of the WHKSs and the splendid and colorful natural scenery to be its most appealing aspect; a total of 51.3% of visitors rated the Libo WHKS as slightly deficient in terms of scientific tourism.

4.2. Structural Equation Model

4.2.1. Reliability and Validity

In this paper, all variables except basic demographic information were potential variables that needed to be evaluated before testing the structural equation model [75]. In this study, latent and measured variables were at the core of the research related to tourism development and environmental protection, based on visitor perceptions and the protection of the development and geomorphological values of WNHSs. Primarily, 21 core potential variables, in terms of heritage geomorphology value perceptions, environmental knowledge perceptions, local attachment, and behavioral intentions, were studied and analyzed (e.g.,

Table 2. Reliability and validity analysis of the questionnaire.

Number of Projects	Cronbach’s α		KMO and Bartlett Test
21	Cronbach alpha based on standardized terms	0.935	Kaiser–Meyer–Olkin Measurement of Sampling Appropriateness		0.947
	Cronbach’s α	0.935	Bartlett’s sphericity test	Approximate chi-square	6609.166
				Degree of freedom	210
				Significance	0.000

). To ensure the stability and reliability of the scale, reliability and validity tests were performed on the reliability of the sample containing the 21 question items prior to sample analysis. The internal consistency test was conducted by calculating the reliability coefficient of Cronbach’s alpha for the scale, and the results showed that the scale had a strong combined reliability (CR), with CR = 0.935; the reliability coefficient of the scale was above 0.9, which indicated that the scale had excellent reliability. Meanwhile, the applicability of the Kaiser–Meyer–Olkin (KMO) test and Bartlett’s spherical test for a factor analysis of the sample data was explored, and the results showed that the KMO value was 0.947 and that Bartlett’s spherical test chi-square value was 6609.166, with a significance level of 0.000 < 0.001; a significant statistical test of validity for 21 question items verified the suitability of the data for factor analysis [76].

4.2.2. Exploratory Factor Analysis

To ensure the quality of the scaling, the 21 extracted variables were tested for structural validity and internal consistency, respectively. The results are shown in

Table 3. Exploratory factor analysis of the measurement project.

Latent Variable	Dimensions	Observational Variable	Factor Loadings	KMO and Bartlett Test	Cronbach’s α
Heritage Geomorphology Value Perception	Value cognition	Libo meets the geological, landform and aesthetic standards (GVP1)	0.793	KMO = 0.880 Sig. = 0.000	0.864
		Libo peak depressions, valleys, canyons and other geomorphic landscape combinations are rich, complete and typical (GVP2)	0.821
	Effective management	Perfect management system and measures for heritage protection (GVP3)	0.808
		Integration of tourism development status and value conservation of heritage sites (GVP4)	0.703
	Cognition of heritage value	The heritage site has the significance of scientific research, cultivation and popularization (GVP5)	0.667
		The heritage sites are rich in demonstrations of outstanding universal values and educational activities (GVP6)	0.834
Environmental Knowledge Awareness	Self-directed education	Interesting knowledge of heritage sites presented through brochures, video clips, etc. (EKA1)	0.871	KMO = 0.818 Sig. = 0.000	0.873
		Publicity and explanation of heritage sites contribute to heritage conservation management. (EKA2)	0.858
	Other-directed education	Tour guide’s presentation helped me to learn some knowledge. (EKA3)	0.860
		Lecture training on heritage conservation management. (EKA4)	0.817
Place Attachment	Place dependency	I really like this place and the heritage tour is unforgettable. (PA1)	0.800	KMO = 0.887 Sig. = 0.000	0.885
		Compared with other sites, the heritage sites of Shibing and Libo are more satisfying. (PA2)	0.790
		I think I will never forget the beautiful natural scenery here. (PA3)	0.797
	Place identity	Heritage tourism sites are very important tourism activities for me. (PA4)	0.812
		I consider the site to be of high natural heritage value. (PA5)	0.798
		I have a strong sense of identification with the Libo and Shibing heritage sites. (PA6)	0.785
Behavior Intentions	Compliance behavior intentions	When I visit, I will observe the heritage site notes. (BI1)	0.765	KMO = 0.833 Sig. = 0.000	0.847
		I will respect local customs, cultural traditions and religious beliefs. (BI2)	0.764
	Positive behavior intentions	I will discourage any behavior that I see on the tour that breaks the rules of heritage sites. (BI3)	0.828
		I would like to support or participate in volunteer activities for the conservation and management of heritage sites if they are available. (BI4)	0.788
		After learning about heritage sites, I will take the initiative to promote relevant environmental knowledge to others. (BI5)	0.794

Note: Heritage Geomorphology Value Perception (GVP); Environmental Knowledge Awareness (EKA); Place Attachment (PA); Behavior Intentions (BI).

. The internal consistency coefficient Cronbach’s α was greater than 0.8, indicating that the reliability of the scales was excellent; the KMO coefficients of the latent variables were all larger than 0.8 for the sample data, indicating the strong validity of the sample data. The KMO coefficients were all greater than 0.8 for the sample data, indicating the strong validity of the sample data, and significant consistency and validity of the measurements [77].

4.3. Validation Factor Analysis

In this study, the validation factor analysis was performed using the AMOS 26.0 software and the model was tested for structural validity for the first time. For the 558 samples analyzed, x²/df was a better indicator of model fit than x². The more consistent the model fit is, the more “successful” the model is and the more useful it is. According to Hayduck, among the myriad measures used to fit the model to the structural equations; the cardinality minimum difference value (CMID), degree of freedom (DF), canonical cardinality value (x²/df), goodness-of-fit indicator (GFI), adjusted goodness-of-fit indicator (AGFI), comparative goodness-of-fit indicator (CFI), non-canonical goodness-of-fit indicator (TL), root mean squared error of approximation (RMSEA) error (RMSEA) and standardized root mean square residuals (RMSR) can be used to evaluate the fit of the model to the data [78,79,80]. Numerical results and recommended values for the metrics of the models proposed in this study are given in

Table 4. Test results of goodness-of-fit indices for SEM.

Fit Indices	Model Index Values	Reference Range	Conclusion	Reference Sources
CMID	474.802	The smaller the better
DF	183	The smaller the better
CMID/DF	2.595	<3 Excellent, <5 Acceptable	Excellent	Hayduck, 1987 [78]
GFI	0.925	>0.8 Acceptable, >0.9 Good fit	Good fit	Bagozzi and Yi, 1988 [79]
AGFI	0.906	>0.8 Acceptable, >0.9 Good fit	Good fit	Hu and Bentler, 1998 [80]
CFI	0.952	>0.9	Good fit	Bagozzi and Yi, 1988 [79]
TLI(NNFI)	0.945	>0.9	Good fit
RMSEA	0.054	<0.08 Excellent, <0.1Acceptable	Excellent	Bagozzi and Yi, 1988 [79]
SRMR	0.043	<0.08	Good fit	Hu and Bentler, 1998 [80]

, and the comparative analysis allows us to conclude that the test superiority metrics meet the recommended level, indicating that the models are well-suited to the collected data.

According to the test results of the model fit in Table 4, a x²/df below 5 is acceptable (below 3 is excellent). The value calculated in this paper is 2.595, which is in the excellent range. The values of gfi, agfi, cfi, and tli are all larger than 0.9, indicating a favorable model fit. The RMSEA is 0.054, which is in the excellent range. The standardized root mean square residual (SRMR) is 0.043, which is less than 0.08, indicating a favorable model fit. From this, it can be judged that the structural model is stable and the discussed causal relations are present. Therefore, from an overall point of view, the structural equation model in this paper is in perfect agreement and the model can be studied and analyzed.

4.3.1. Portfolio Validity

After obtaining an excellent validation model by fitting a structural equation model, the reliability and validity of each dimension were tested to verify the applicability of the model. Through the analysis, the results are shown in

Table 5. Convergent validity test table.

Dimensionality	Title Item	Significance Estimation				Topic Validity		Component Reliability	Convergent Validity
		Unstd.	S.E.	z-Value	p	Std.	SMC	CR	AVE
Heritage Geomorphology Value Perception	GVP1	1.000				0.751	0.564	0.866	0.523
	GVP2	1.042	0.056	18.634	***	0.787	0.619
	GVP3	1.023	0.057	18.053	***	0.764	0.584
	GVP4	0.837	0.058	14.430	***	0.621	0.386
	GVP5	0.784	0.058	13.537	***	0.585	0.342
	GVP6	1.070	0.056	18.965	***	0.800	0.640
Place Attachment	PA6	1.000				0.729	0.531	0.885	0.562
	PA5	1.021	0.060	17.037	***	0.750	0.563
	PA4	1.066	0.060	17.793	***	0.783	0.613
	PA3	1.012	0.060	16.883	***	0.743	0.552
	PA2	1.005	0.060	16.753	***	0.737	0.543
	PA1	1.021	0.059	17.156	***	0.755	0.570
Environmental Knowledge Awareness	EKA4	1.000				0.708	0.501	0.873	0.633
	EKA3	1.161	0.064	18.073	***	0.817	0.667
	EKA2	1.158	0.064	18.185	***	0.822	0.676
	EKA1	1.164	0.064	18.324	***	0.829	0.687
Behavior Intentions	BI1	1.000				0.704	0.496	0.848	0.527
	BI2	0.983	0.065	15.174	***	0.695	0.483
	BI3	1.075	0.065	16.600	***	0.764	0.584
	BI4	1.049	0.064	16.268	***	0.748	0.560
	BI5	1.009	0.064	15.652	***	0.718	0.516

Note: *** p < 0.001.

; the standardized factor loadings (Std) of the topic validity range from 0.585~0.829, which is significant (p < 0.001); the CR of each latent variable is greater than 0.8; and the average variance extracted (AVE) reaches the standard of 0.5. the higher the AVE, the better the convergence effect.

4.3.2. Distinguishing Validity

The validity test looks at the discriminative validity between variables, where discriminative validity refers to low correlations and significant differences between latent variables. This can be assessed by comparing the magnitude of the square root of the mean squared integer, extracted with the correlation coefficient of the desired variable. According to the criteria proposed by Fornell and Larcker, if the correlation coefficient of a variable with other variables is smaller than the square root of the AVE of that variable, it indicates the favorable discriminant validity of that variable [81]. As shown in

Table 6. Distinct validity scale.

	Convergent Validity	Distinct Validity
	AVE	Place Attachment	Environmental Knowledge Awareness	Heritage Geomorphology Value Perception	Behavior Intentions
Place Attachment	0.633	0.796
Environmental Knowledge Awareness	0.562	0.748	0.749
Heritage Geomorphology Value Perception	0.523	0.719	0.473	0.723
Behavior Intentions	0.527	0.770	0.491	0.852	0.726

Note: The diagonal bold numbers are the open root values of AVE.

, the results indicated that the square root of the AVE of each latent variable was higher than the correlation coefficient of that latent variable with other latent variables, except for the correlation coefficient value of the perceived value of heritage geomorphology and behavioral intention (0.852), which was higher than the AVE of behavioral intention (0.527). In summary, the apparent validity among the latent variables is better. Therefore, the discriminative validity of the measurement model in this study is appropriate and additional hypothesis testing can be performed.

4.4. Mediation Effect Check

The mediation effect between GVP, PA, and BI is also explored by the hypothesis testing results. SEM was plotted with AMOS 27.0 to set a 95% confidence interval, and 5000 bootstrap samples were taken to test the mediation effect using the bootstrap ML method. The results showed that the standardized indirect effect coefficients between environmental knowledge perceptions and tourists’ behavioral intentions for heritage conservation were both 0.000, with significant indirect effects, and both bias-corrected and percentile tests did not contain 0 in the 95% confidence interval; this indicates that geomorphological value perceptions and local attachment mediated the relationship between environmental knowledge perceptions and tourists’ behavioral intentions for heritage conservation. The role of the standardized indirect effect coefficient between environmental knowledge perceptions, geomorphological value perceptions, place attachment and heritage conservation behavioral intentions was 0.165, with 0 found in both the bias-corrected and percentile tests at 95% confidence intervals, indicating that geomorphological value perceptions and place attachment do not mediate the effects of environmental knowledge perceptions and tourists’ heritage conservation behavioral intentions. role. In summary, hypotheses H7 and H8 hold and hypothesis H9 does not.

4.5. Structural Model Hypothesis Validation Evaluation and Results Analysis

The overall model relationship path diagram generated by Amos 27 (Figure 5) explains the degree of influence and the structural relationships among the latent variables in the structural equation model constructed in the paper.

Table 7. Model path analysis results.

Assumptions	Unstd.	S.E.	C.R.	p	Std. (β)	R2	Verification Results
EKA→GVP	0.764	0.058	13.103	***	0.719	0.518	Support
EKA→PA	0.858	0.079	10.843	***	0.844	0.568	Support
GVP→PA	0.431	0.058	9.543	***	0.602		Support
GVP→BI	0.563	0.058	9.758	***	0.605	0.782	Support
PA→BI	0.401	0.054	5.023	***	0.408		Support
EKA→BI	0.408	0.078	5.244	***	0.413		Support

Note: *** p < 0.001.

and

Table 8. Results of intermediate effect test.

Assumptions	Intermediary Path	Indirect Effect Coefficient	Bootstrapping				Verification Results
			Bias-Corrected 95%		Percentile 95%
			Lower Limit	Upper Limit	Lower Limit	Upper Limit
H7	EKA → GVP → BI	0.000	0.244	0.496	0.241	0.491	Support
H8	EKA → PA → BI	0.000	0.017	0.240	0.018	0.237	Support
H9	EKA → GVP → PA → BI	0.165	−0.047	0.001	−0.037	0.004	Rejection

reports the results of the hypothesis testing. Overall, all 8 hypotheses are supported.

5. Results and Discussion

5.1. Results

The environmental knowledge perceptions factor had significant positive effects on geomorphological value perceptions, place attachment, and positive environmental protection behavioral intentions (H1 = 0.72, CR = 13.103; H2 = 0.84, CR = 10.843 and H3 = 0.41, CR = 5.244). Thus, H1, H2 and H3 are supported. In other words, environmental education leads to a deeper understanding of the scientific value of a heritage site when visitors visit it, makes them feel attached to it, and directly motivates their positive environmental behavior and conservation intentions. As hypothesized, geomorphological value perception had a significant positive effect on place attachment (H4 = 0.431, CR = 9.543) and on positive environmental behavioral conservation intentions (H5 = 0.563, CR = 9.758), thus supporting H4 and H5. Place attachment had a significant positive effect on positive behavioral intentions to protect the environment (H6 = 0.563, CR = 9.758), supporting H6. Personal emotional identity played an essential role in protecting visitors in nature and in bringing them closer to nature. Finally, H7 and H8 indicate the mediating role of geomorphological value perception and place attachment between environmental knowledge perception and behavioral intention, which indicates that value perception and place attachment are key factors in the development of heritage conservation and tourism, and have an important role in the conservation and development of heritage sites; thus H9 is not supported. Once the identity bond is established, the visitors themselves generate a self-regulatory force that makes the willingness to limit unfriendly behavioral intentions stronger.

5.2. Theoretical Contributions

This study contributes to the existing literature in several ways. The first is the visitor’s perception of the geomorphological value of the heritage site. The complex scientific concepts of karst landscapes (dissolution of soluble rocks by water) are not suitable for tourists who are not knowledgeable about earth sciences. Moreover, some supporting materials (panels, leaflets, etc.) are poorly scientifically explained and lack methods to effectively communicate scientific concepts to the so-called general public. Therefore, tourists’ perception of forms (landscape, topography, sediments, rocks, fossils), processes (tectonic activity, erosion, sedimentation) and tourism (attractions, accommodation, tours, activities, interpretation, planning and management) in WHKSs becomes particularly valuable. Visitor perception can be enhanced through technological measures; for example, new technologies (such as audio commentary applets, human–computer interaction, virtual reality, etc.) can be used to facilitate the absorption and perception of geomorphological values by visitors. Second, environmental knowledge perception is a precursor to ensuring attachment and legacy-conserving behavioral intentions, and the perception of geomorphological values plays a crucial mediating role. Furthermore, tourists’ place attachment, heritage attachment and destination attachment facilitate scenic conservation and value demonstration, which have an integrated impact on heritage conservation [16,82]. Finally, legacy-conserving behavioral intentions are a direct consequence of geomorphological value perception and place attachment. It is only by developing emotional attachment and our value perceptions during a trip to a WHKS that we can better constrain our behavioral intentions. Environmental knowledge perception, geomorphological value perception, and place attachment together contribute to creating positive behavioral intentions for heritage preservation.

5.3. Managerial Contributions

The study provides several practical implications for heritage site management and marketing organizations by understanding the factors found to have a direct impact on tourists’ conservation behavioral intentions in heritage sites. First, in terms of WHKSs’ geomorphological values, tourism at heritage sites should be sustainable (i.e., economically viable, promote community development, and contribute to landscape conservation) and educational (achieved through geographic interpretation), based on karst landscapes. This contributes to the communication and presentation of the geomorphological value of WHKSs and the realization of the educational value of heritage sites. Education is the least expensive and most beneficial form of conservation in WHSs [83]. Second, heritage site managers should focus on conserving natural resources, and designing and promoting their integrity and authenticity to visitors. Finally, heritage tourism should promote the preservation of the geomorphological value of WHKSs, with tourism and tourism development having value preservation at its core [84]. The organic integration of humanities, social sciences, and natural sciences into the management and conservation of WHSs will enable green and sustainable development.

5.4. Limitations and Future Research Directions

From the perspective of geomorphological value conservation, if people are made more aware of and connected to the geological landscape of a heritage site through more significant and memorable experiences, they are more likely to value it and help manage it sustainably [85]. This study explores and analyzes the preservation of the geomorphological value of heritage sites and the development of tourism in relation to the perceived geomorphological value of WHKSs. Some limitations provide potential avenues for future research. First, this paper focuses on quantitative analysis, which, in conjunction with qualitative studies, allows for an in-depth analysis of the formation mechanism of legacy-conserving behavioral intentions in the future. Second, the introduction of heritage site geomorphological value perception and environmental knowledge perception into the antecedents that drive the behavioral intentions of visitors for heritage preservation is explored. While the scale design and empirical evidence demonstrate that the model fits accurately, the applicability of the scale to additional WNHSs needs to be validated by adding empirical case studies.

6. Conclusions

In this study, the tourism questionnaire of Libo WHKS was selected, and the stimulus–organism–response structure of the Mehrabian and Russell model was used to discover the influence mechanisms of tourists’ perceptions of geomorphological value and environmental knowledge on tourists’ place attachment and behavioral intentions towards environmental protection in the heritage site. First, we construct a theoretical framework for “GEPB” for geomorphological value perception, environmental knowledge perception, place attachment, and behavioral intention by establishing perception scales for both visitor value perception and visitor behavioral intention. Second, the perception of environmental knowledge has a significant effect on the perception of geomorphological value, suggesting that the perception of the cognitive level of the visitor affects the perception of geomorphological value. Environmental knowledge perceptions and value perceptions have a significant impact on place attachment and influence visitor behavioral intentions, suggesting that visitor perceptions of the value of a heritage site can lead to place attachment, which, in turn, affects visitor conservation behaviors and intentions to revisit the site. The role of place attachment and value perception in mediating visitor behavioral intentions and environmental knowledge suggests that place attachment and value perception play a key role in promoting heritage preservation and tourism development. Finally, based on the theoretical framework of the “cognitive–emotional–behavioral” chain, we provide a direction for the synergistic development of WHKSs’ conservation and tourism, and make an important contribution to tourism development, heritage conservation and tourism education based on the sustainability of karst landscapes.