Climate Change Mitigation Behaviors in Tourists in Chinese Mountains

Abstract

In this study, we employed a Protection Motivation Theory (PMT) framework and Stimulus–Organism–Response (SOR) model to explore respondents’ emotional and behavioral responses to threats posed by climate change in high mountain areas. Data were collected from 391 valid questionnaires and analyzed using partial least squares structural equation modeling (PLS-SEM) to rigorously test the proposed hypotheses. The results indicate that threat appraisal and coping appraisal are significantly associated with stronger tourist intentions toward climate change mitigation, as they foster positive emotional responses. Specifically, the pathway involves awe as a self-transcendent emotion, which appears to play a crucial role in predicting climate change mitigation behavioral intentions. Climate change knowledge is found to negatively moderate the relationship between threat appraisal and awe. These findings provide new insights by revealing the psychological mechanisms underlying mountain tourists’ behaviors against climate change. Practically, they highlight the need to use diverse audiovisual elements to evoke awe among viewers and communication messages that focus on the severity of the threats posed by climate change in highly mountainous areas.

1. Introduction

During the period from 23 June to 2 July 2025, soaring temperatures across Europe resulted in 2305 deaths across 12 major cities, including Barcelona, Milan, and London. Among these, temperature changes attributable to climate change were responsible for 1504 fatalities [1]. Climate change has become a pressing issue of our time [2], and its impact on highly mountainous areas around the world is particularly pronounced. Characterized by high elevation, steep topography, cold temperatures, and remoteness, highly mountainous regions show a prominent sensitivity to climate change [3], with impacts such as accelerated glacier retreat, frequent hazards caused by extreme weather, and the gradual loss of biodiversity. These impacts severely threaten the vulnerable ecosystems and high-altitude environments of alpine regions, and, without immediate and effective climate change mitigation measures, they are likely to result in irreversible damage. Home to approximately 10 percent of the global population [4], highly mountainous areas have developed unique cultural identities, which, combined with their striking natural beauty, unique biodiversity and ecosystems, and abundant recreational opportunities, represent a strong, multi-faceted appeal for tourists. Mountain tourism has brought highly significant economic and social development opportunities to local communities, which are usually located in less developed and remote regions [5]. Climate change could jeopardize the future of mountain tourism, a vital economic sector that provides income, supports conservation efforts, and fosters cross-cultural appreciation for these unique landscapes. The sustainability of this important form of tourism is now under unprecedented threat.

As China includes vast and topographically diverse mountainous territories, it faces severe impacts from these issues. Iconic highly mountainous areas such as the Himalayas, the Kunlun Mountains, and the Tianshan are not only geological marvels but also crucial reservoirs of biodiversity and water resources. Among these, the Hengduan Mountains in southwestern China stand out as a region of exceptional ecological and touristic value yet one that remains comparatively understudied in the context of climate change and tourist behaviors. The Hengduan Mountains, spanning parts of the autonomous region of Tibet, Yunnan and Sichuan provinces, host a remarkable concentration of 249 A-rated tourist attractions, forming a renowned and economically critical tourism cluster [6]. With profound topographic variation and a broad spectrum of climatic conditions, this area has become renowned for its rich biodiversity [7]. However, the flora and fauna species are impacted severely by changes in precipitation and increasing temperatures [8]. This area is also home to a variety of ethnic groups, primarily Tibetan people, which has contributed to its rich and unique cultural heritage, both tangible and intangible. The dearth of research focusing specifically on this region represents a significant oversight, limiting our understanding of tourist perceptions and behaviors in a geographically and culturally vital area. This study is designed to address this gap by focusing on the Hengduan Mountains, thereby contributing empirical insights gleaned in a previously neglected context.

It is crucial to understand what can motivate tourists to behave in a way that fosters pro-environmental action. However, engaging tourists in climate change mitigation represents a unique challenge [9]. Unlike local residents, who witness the gradual, cumulative effects of climate change over years, tourists are transient visitors. Their brief stay typically prevents them from directly observing the long-term, persistent environmental degradation. This temporal disconnect can lead to a failure to recognize the severity of the threat. Although public perception and awareness of climate change are growing in China [10], it is worth discussing whether increased subjective knowledge in this field could influence people’s threat perception and coping efficacy in the face of climate change. As a matter of fact, the very grandeur and pristine beauty of the mountain environment, such as towering peaks, deep valleys, and majestic glaciers, can evoke a powerful emotional response in tourists, most notably, the feeling of awe. Awe, a self-transcendent emotion elicited by perceptually vast stimuli that overwhelm current mental structures [11], is thought to promote cognitive accommodation, thus encouraging individuals to care more about the external environment. In the context of mountain tourism, this emotional and cognitive response holds significant potential. It can bridge the gap created by temporal disconnect, encouraging tourists to adopt behaviors to protect the source of their awe-inspiring experience.

This study employs Protection Motivation Theory (PMT) and the Stimulus–Organism–Response (SOR) model to empirically investigate the formation and impact mechanisms of awe within the specific context of the Hengduan Mountains. The PMT model has been widely used to study the motives behind pro-environmental behaviors. However, its application in highly mountainous contexts has been restricted to a cognitive-orientated framework. Prior studies have largely focused on how these cognitive appraisals directly lead to behaviors or behavioral intentions, overlooking the potential mediating emotional role that appraisal processes may elicit. We posit that in highly mountainous settings featuring striking landscapes, threat and coping appraisal is not the sole driver of actions. Rather, it first elicits awe, a transcendent emotion that is closely associated with stimuli that are vast and lead to fear appeal, which in turn motivates intentions toward climate change mitigation behavior. Considering this, we propose an extended PMT framework that utilizes awe as the organism element in the SOR model to explain the psychological pathway leading to climate change mitigation behavior intentions. This study is designed to (1) investigate the direct impact of protection motivations on climate change mitigation behavior intention; (2) examine the mediating role of awe in the relationship between protection motivations and climate change mitigation behavior intention; (3) assess the moderating effect of subjective knowledge on climate change in the process. For this study, we created an online survey targeting tourists who visited the most prestigious destinations in Hengduan Mountains. The findings are expected to contribute to the mountain tourism and climate change literature while offering actionable insights for destination managers, marketers, and policymakers to inform the design of tourism experiences and communication strategies that promote sustainable behaviors and contribute to long-term resilience and sustainability in tourism in highly mountainous areas in China and beyond.

2. Literature Review and Hypothesis Development

2.1. Climate Change and Mountain Tourism

Climate change refers to long-term changes in weather patterns that result in direct effects, such as rising temperatures, and secondary effects, such as rising sea levels and reduced snow cover [12]. With the climate being a crucial attraction in many destinations, variations in temperature and precipitation will reduce the attractiveness of tourist destinations, thus changing tourist behaviors and negatively affecting the overall travel experience [13]. Mountain tourism relies heavily on natural environments, the quality of which is largely affected by the climate [14]. The negative effects of climate change, such as increased temperatures and reduced snow cover, are particularly pronounced in mountain destinations such as the Alps, which rely on snow as the main driver of tourism development [15]. In addition, the decline in climate comfort will lead to a sharp decrease in tourism demand, and this change will affect the seasonal distribution of mountain tourism [16]. To address these changes, some regions have begun to adopt a systems approach to adjusting tourism activities, such as summer glacier skiing in the Swiss Alps, to enhance adaptability through diversified tourism activities and infrastructure improvement, in the face of the huge challenges posed by climate change [17].

In addition, the economic and social impacts of climate change on mountain tourism are still significant. Mountain tourism is not only a pillar of the local economy but also provides a large number of employment opportunities for local communities, especially in remote areas. As the number of tourists decreases, the economic income and living standards of these areas are threatened [18]. For instance, summer glacier skiing activities in Switzerland are severely threatened by climate change, and most ski resorts have closed. These scenic areas are facing huge economic and environmental pressures [17]. This change has not only affected tourist engagement but has also affected local economic activity [19]. Although there have been many studies on the impact of climate change on mountain tourism, there are still some research gaps, especially in southern regions, where comprehensive studies on the impact of climate change on ecosystem services, local communities, and tourists are relatively scarce [20]. At the same time, there is a lack of systematic and multidisciplinary in-depth research on sustainable development and adaptive measures in mountain tourism [19]. In summary, the impacts of climate change on mountain tourism are complex and diverse, and local governments and relevant stakeholders must work together and take effective adaptation and mitigation measures in order to ensure long-term sustainability in mountain tourism. Therefore, in the face of climate change and the continuous development of mountain tourism, we should also pay attention to tourism behaviors to enable sustainable development in scenic spots.

2.2. PMT and Climate Change Mitigation Behavior Intentions

The “Protective Motivation Theory” (PMT) was originally developed by Rogers [21] to explain the way in which individuals respond to health threats. Since then, the application of PMT has been extended to a broad range of fields, from injury prevention and political issues to environmental concerns and the protection of others [22]. The theory suggests that an individual’s decision to take protective behavior is driven by two main cognitive assessments: threat appraisal and coping appraisal. Threat appraisal assesses the severity and vulnerability of a threat, while coping appraisal assesses an individual’s perception of the threat and the effectiveness of the proposed response. According to PMT, individuals are more likely to engage in protective behavior if they perceive the threat as serious and have information about their ability to respond effectively [21]. PMT was further refined in the 1980s with a focus on understanding the different dimensions of threat appraisal, as well as coping appraisal, resulting in a more nuanced understanding of how individuals assess and respond to health threats [23].

It was not until the beginning of the 21st century that PMT began to be applied to the study of pro-environmental behaviors, especially in the context of natural disasters and climate change. Research based on PMT frequently forms the basis for empirical studies that examine how fear appeals regarding environmental threats can motivate behavioral change [24]. As the focus on climate change intensifies, researchers are trying to apply PMT to understand how individuals respond to environmental risks, including natural disasters and long-term environmental threats. This shift in focus is driven by the need to understand what an individual might do to mitigate the effects of climate change, such as reducing greenhouse gas emissions and adopting better sustainable practices [25]. In the domain of tourism, researchers found that when tourists perceive higher severity and vulnerability, as well as higher self-efficacy and response efficacy, pro-environmental behaviors will be stimulated [24,26], such as consuming plant-based food [27], reducing one’s carbon footprint when traveling [28], adopting fully electric vehicles [26], and learning more about the surrounding environment [29]. This highlights the importance of threat and coping appraisal in pro-environmental behaviors [30].

However, when applied to the mitigation of climate change, PMT often ignores wider societal influences on individual behavior. Economic and institutional factors must be reflected in research on farmers’ preparedness to deal with climate-related disasters, where risk perception is important but barriers such as lack of resources and knowledge often limit effective action [31]. In addition, the induction of self-efficacy, which is part of coping appraisal, also plays an important role in determining whether individuals feel empowered to take action against climate change [32]. Recently, in addition to the core components of PMT, moral obligation, the feeling of personal responsibility to act, has been frequently integrated as an extension of PMT [33]. While our study focuses on testing the fundamental structure of PMT, it is widely recognized that moral obligation represents a potent parallel driver, which suggests that individuals participate in climate change mitigation not only because they feel that they are in danger or believe that their actions are effective but also because they believe that they have a moral responsibility to protect the environment and future generations [33].

Although the research on PMT is expanding, there are many gaps, such as the fact that PMT focuses on the cognitive processes involved in protective behavior but ignores the role of emotions in motivating action; emotional responses such as awe or guilt may encourage individuals to take protective action, especially if they feel empowered to do so [24]. This means that the combination of emotional responses and PMT can enhance predictive power and provide a more comprehensive research framework for understanding climate change mitigation behaviors. Overall, PMT provides valuable insights into assessing and responding to climate change threats, not only to guide behavioral intent but also to influence actual action. However, there are still many challenges limiting the application of PMT to climate change research, and the theory’s emphasis on individual-level factors needs to be extended to consider the role of social fees, policies, knowledge, and collective action. The combination of moral motivation and emotional response represents a promising direction for future research, especially in ways to encourage people to adopt long-term sustainable behaviors. By filling in these gaps and combining PMT with other behavioral theory models, we can develop more effective strategies to promote climate change mitigation behaviors at the individual and societal levels [34,35].

H1. 

Threat appraisal is positively related to climate change mitigation behavior intentions.

H2. 

Coping appraisal is positively related to climate change mitigation behavior intentions.

2.3. Mediating Effect of Awe

Awe is a complex psychological response, usually triggered by stimuli that are considered to be extremely grandiose, beyond one’s existing frame of mind, and that require adaptation and adjustment [11]. Awe triggers a sense of insignificance and the need to adjust our frame of mind to make sense of the vast world that we encounter [36,37]. Spectacular mountain scenes are typical examples of awe-inspiring nature [38]. As an iconic natural landscape, mountains are often awe-inspiring due to their size and beauty, and they are seen as places of both spiritual and esthetic significance [39]. The core characteristics of awe are “perceived vastness” and “need for accommodation” [11], which allow people to shift their perspective, thus focusing less on themselves and more on the wider world around them [40]. Based on the Stimulus–Organism–Response model, it is found that when tourists are exposed to object and social stimuli, awe is evoked as an intrinsic positive emotion, which, in turn, leads to changes in how they respond to their destination [41].

First proposed by Mehrabian and Russell [42], the Stimulus–Organism–Response (S-O-R) model has been widely applied in studying tourist behaviors [43,44,45]. In traditional S-O-R research, “stimuli” are regarded as external factors, such as the natural environment, a retail ambiance, or service quality. The S-O-R model suggests that particular environmental stimuli directly influence an individual’s cognitive and emotional conditions, resulting in either approach or avoidance behaviors [46]. However, in this research, we conceptualize cognitive appraisals as internalized psychological stimuli. Upon their visit to highly mountainous areas, tourists receive lots of environmental clues and mentally process them. This internal assessment generates cognitive outcomes [47], i.e., threat appraisal and coping appraisal in our study. As direct reflection of the external reality of highly mountainous contexts, threat and coping appraisal function as psychological events that subsequently stimulate the emotional state of the “organism”. Recently, an increasing number of studies have suggested that subjective perceptions, such as cognitive assessment, personal beliefs, and emotions, can serve as stimuli to foster emotional and cognitive responses, thereby influencing subsequent behaviors. For instance, in the field of virtual tourism research, tourists’ subjective perception of authenticity plays a stimulating role, influencing their emotions and thereby affecting their willingness to visit the destination [48]. Further research has found that tourists perceiving a destination as having restorative qualities can have a positive impact on their mental image of the destination and their loyalty to it through the experience of finding pleasure and meaning in life [49]. Similarly, studies have demonstrated how tourists’ subjective perception of destination credibility leads to emotional responses and environmentally responsible behaviors, highlighting the crucial role of subjective perception as an internal stimulus [50]. In a study on online consumer behavior based on the S-O-R model, it was found that potential customers appraised or evaluated online reviews, which then served as stimuli to foster different emotions, resulting in various empathy behaviors [51]. When they extended the research model to the context of online travel agency apps such as Tripadvisor and Booking.com, the authors found the same Stimulus–Organism–Response mechanism [52]. Therefore, threat appraisal and response appraisal in PMT can also be regarded as subjective stimuli that trigger emotional responses (awe) and lead to specific behavioral reactions.

H3a. 

Threat appraisal is positively related to awe.

H3b. 

Coping appraisal is positively related to awe.

As an emotion usually aroused by magnificent natural scenery, awe is increasingly seen as an important emotional experience in the tourism industry. This type of awe is usually associated with other positive emotions, such as gratitude and joy, and is classified as unthreatened awe [53]. When tourists visit highly mountainous areas, the vastness of nature is likely to elicit this type of awe. In the meantime, when tourists perceive and appraise the likelihood and severity of climate change in highly mountainous areas, threatened awe is likely to be aroused, which is characterized by an increased feeling of fear [54]. Therefore, in this study, we conceptualize and measure both threatened and unthreatened awe as one variable, “awe”. The sense of awe in natural environments greatly increases the overall satisfaction of visitors and strengthens their connection with the environment, leading them to adopt environmentally friendly behaviors to protect the environment [39,55]. Often evoked in tranquil, awe-inspiring landscapes, awe inspires a desire for self-improvement and prompts visitors to take actions that benefit the environment [56]. A sense of awe can make visitors more aware of the importance of environmental protection, thereby increasing their environmental awareness and promoting sustainable tourism activities [57].

The role of “awe” as an emotional response has been increasingly recognized in influencing environmental behaviors, especially in the fight against climate change. An awe-inspiring sanctum enables visitors to feel that they are a part of the scenic landscape by reducing the focus on themselves, which increases their sense of responsibility for the environment [57]. This shift in self-perception motivates visitors to enact eco-friendly behaviors such as reducing waste or supporting conservation efforts [55,56]. In spectacular natural settings such as high mountains, awe triggers a “small-self” effect, which drives visitors to feel smaller. The feeling of “being small” subsequently reinforces their focus on environmental issues and encourages sustainable behavior [36,40]. In summary, the current research emphasizes the role of awe in tourism. Awe can significantly mediate the relationship between tourists’ perception of the natural environment and their environmental behaviors. This demonstrates the key role played by awe in cultivating environmental awareness in tourism environments [58].

H3c. 

Awe is positively related to climate change mitigation behavior intentions.

H3d. 

Awe mediates the positive relationship between threat appraisal and climate change mitigation behavior intentions.

H3e. 

Awe mediates the positive relationship between coping appraisal and climate change mitigation behavior intentions.

2.4. Subjective Knowledge on Climate Change as a Moderator

Individuals’ environmental behaviors are driven by their environmental knowledge. Current research indicates that knowledge of environmental issues and mitigation measures plays an essential role in shaping attitudes and behavioral patterns toward climate change mitigation [58]. For instance, knowledge of the causes and consequences of climate change plays a key role in shaping attitudes toward climate change mitigation [59]. Knowledge can make people aware of the urgency of environmental problems, which in turn strengthens their willingness to take steps to protect the environment [60]. Furthermore, knowledge of the environmental impact of various behaviors, such as saving energy or reducing one’s carbon footprint, has been identified as a key factor driving more significant actions, especially when high-impact behaviors are desired [61]. Nevertheless, the role of knowledge is complex, particularly when the distinction between subjective and objective knowledge is considered [62]. Although both types of knowledge play important roles in the cognitive process, subjective knowledge has been proven to be strongly associated with consumer behaviors and is easier to measure than objective knowledge [63]. Thus, the present study measures subjective knowledge only. Some studies have found that elevated subjective knowledge can paradoxically weaken risk perception [64,65], often attributed to the development of overconfidence [66], which leads to individuals underestimating threats. In contrast, other research has reported a positive association, aligning with the conventional view that knowing more makes one feel more vulnerable [67]. Yet another body of work has found non-significant relationships, suggesting the influence of contingent factors [28,63]. This inconsistency underscores a critical research gap, i.e., understanding the specific mechanism through which subjective knowledge influences cognitive appraisals within protection motivation frameworks and moderates the feeling of awe.

In the meantime, researchers have studied the interplay between awe and environmental knowledge. As a self-transcendent emotion, awe is elicited through encounters with perceptually vast stimuli that surpass and disrupt an individual’s current knowledge structures [68]. Learning about nature promotes the experience of awe. It was revealed that participants with a higher level of knowledge about nature reported elevated levels of both dispositional awe and nature connectedness [69]. Actually, awe has long been regarded as a “epistemic emotion” [70] or “knowledge emotion” [11]. However, as an emotion often triggered by vastness, awe inspires individuals to think outside of the box and depend less on previous knowledge. The complex interplay and relationship between awe and knowledge make it worthwhile to continue to explore the dynamic interactions between knowledge as a moderating variable and other factors and to further elucidate how knowledge can be utilized to promote the experience of awe and thus encourage more sustainable environmental protection behaviors. Thus, we propose a first-stage moderated mediation model. Specifically, we hypothesize that the relationship between cognitive appraisals, i.e., threat and coping, and awe is moderated by subjective knowledge on climate change. This theoretical positioning implies that the strength of the predictive power of threat appraisal and coping appraisal in behavior intentions through awe depends on the level of subjective knowledge.

H4a. 

Subjective knowledge on climate change moderates the relationship between threat appraisal and awe.

H4b. 

Subjective knowledge on climate change moderates the relationship between coping appraisal and awe.

As depicted in Figure 1, the hypotheses listed above were tested.

3. Methodology

This study employs Partial Least Squares Structural Equation Modeling (PLS-SEM) to test the proposed theoretical model. PLS-SEM proves particularly effective in identifying multiple key constructs and their influencing factors. It allows the flexible specification of the relationships between model structures and variables, while providing strong explanatory power for endogenous variables. These advantages have led to it being widely applied in empirical research in recent years. PLS-SEM is explicitly designed to predict and explain variance in the dependent variables, which is aligned with the purpose of the current study, i.e., to predict tourists’ climate change mitigation behavior intentions and to explain the key antecedents of this outcome [71]. Furthermore, PLS-SEM has relatively flexible requirements regarding sample size and data distribution and demonstrates high reliability in estimating models with multiple constructs and path relationships. Thus, without imposing a strict distributional assumption on the data, it provides a robust method for testing interaction effects. PLS-SEM can also estimate complex models with higher-order constructs due to its component-based approach, while avoiding issues such as model identification and estimation instability [72]. These characteristics adequately support the complex modeling requirements of this study, including the specification of second-order latent variables and the incorporation of interaction terms among the measured variables.

According to the theoretical model, perceived vulnerability, perceived severity, self-efficacy, and response efficacy constitute the first-order latent variables. These first-order latent variables, in turn, formatively construct the two second-order latent variables: threat appraisal and coping appraisal. The measurement indicators, the first-order latent variables, the second-order latent variables, and the connections between them collectively form a reflective–formative second-order model. To operationalize these higher-order constructs, we employed the repeated indicators approach [73], where all indicators of the lower-order components, including all items measuring the first-order variables, are repeatedly used as indicators for the higher-order components. This method is well suited for our model as it allows for a direct estimation of the relationships among constructs in the model. As presented in

Table 1. Second-order constructs’ reliability and validity.

Construct	Cronbach’s Alpha	Composite Reliability (CR)	Average Variance Extracted (AVE)
Coping appraisal	0.762	0.834	0.458
Threat appraisal	0.803	0.859	0.506

, the assessment of threat appraisal revealed an AVE of 0.506, which demonstrated its validity. Although coping appraisal revealed an AVE of 0.458, slightly below the strict threshold of 0.5, its composite reliability was high, confirming the internal consistency of the measure.

3.1. Measurement Scale and Questionnaire Design

The research instruments for measuring the constructs were adapted from the existing literature, considering contextual factors and wording concerns. The questionnaire consisted of three parts. The first part comprised a screening question that asked respondents whether they have visited highly mountainous areas in China’s Hengduan Mountains. Those answering “Yes” proceeded with the survey, while those answering “No” were directed to exit. The second part covered demographic variables, including age, gender, educational background, and monthly income.

The third part involved the measurement of the key variables, comprising a total of 23 items. One of the independent variables, “threat appraisal”, comprised two first-order variables: perceived vulnerability and perceived severity. The items measuring the two constructs were adapted from existing studies [3,28]. Three items measuring perceived vulnerability were rated on a 5-point Likert scale, ranging from 1, indicating “definitely impossible”, to 5, indicating “definitely will”. Another three items measuring perceived severity were also rated on a 5-point Likert scale, ranging from 1, indicating “not at all”, to 5, indicating “extremely”. Another independent variable, “coping appraisal”, also consisted of two first-order variables: response-efficacy and self-efficacy. For the measurement of the two constructs, six items adapted from existing studies [24,74,75,76] were used.

The scale of awe, the mediator, was adapted from [77]. For the measurement of subjective knowledge on climate change, the moderator, four items adapted from [28] were used. The dependent variable “climate change mitigation behavior intention” measured the respondents’ willingness to perform the proposed behaviors. Three items that included behaviors both in daily life and on the mountain trip were developed from previous studies [24,74,75,76]. All the other items are measured on a 5-point Likert scale, from 1, indicating “strongly disagree”, to 5, indicating “strongly agree”.

This study adopted a double-blind back-translation procedure for the English scales to minimize subjective bias in the translation process. Two experts in tourism research were then invited to review and refine the questionnaire. Prior to the formal survey distribution, a pilot test was conducted using 63 questionnaires. The reliability of the scale was tested using the item total correlation coefficient and Cronbach’s α coefficient. Feedback from the pretest proved that the questionnaire was generally reliable. The measurement items are presented in

Table 2. Variables and measures.

Construct	Measurement Items	Source
Perceived vulnerability (PV)	How likely do you think it is that climate change will result in natural hazards in mountainous areas?	[3,28]
	How likely do you think it is that climate change will result into changes in accessibility in mountainous areas?
	How likely do you think it is that climate change will reduce landscape attractiveness due to reduced snow cover in mountainous areas?
Perceived severity (PS)	How severely will your tourism experience be influenced by natural hazards in mountainous areas?	[3,28]
	How severely will your tourism experience be influenced if accessibility in mountainous areas is altered?
	How severely will your tourism experience be influenced by reduced landscape attractiveness due to reduced snow cover?
Response efficacy (RE)	Living a green lifestyle helps to mitigate climate change.	[24,74,76]
	Spreading climate change-related knowledge helps to mitigate climate change.
	Reducing one’s carbon footprint while traveling helps to mitigate climate change.
Self-efficacy (SE)	I know how to take actions to mitigate climate change.	[75]
	I have the ability to take actions that could mitigate climate change.
	My individual actions will contribute to mitigating climate change.
Awe	The natural landscape during the visit appeared grand to me.	[77]
	The natural landscape during the visit evoked awe in me.
	The natural landscape during the visit made me feel awe.
	The natural landscape during the visit was astonishingly beautiful.
Subjective knowledge on climate change (SK)	I have some general knowledge about climate change.	[28]
	I know about the causes of climate change.
	I know about the impacts of climate change.
	I know how to respond to climate change.
Climate change mitigation behavior intention (CB)	I am interested in learning more about climate change in mountainous areas.	[24,74,76]
	I am willing to live a green lifestyle.
	I am willing to reduce my carbon footprint when traveling.

.

3.2. Data Collection and Sample Descriptive Analysis

This study focused on the impact of climate change on intentions toward mitigation behaviors among mountain tourists. The research area, the Hengduan Mountains in southwestern China, spans three provinces (Tibet Autonomous Region, Yunnan Province, and Sichuan Province) and boasts 249 A-level scenic spots. With an average altitude of 4000 to 5000 m, this region features spectacular landscapes, glacial wonders, and rich biodiversity. However, it is also a typical ecologically fragile area that is highly vulnerable to the effects of climate change [6]. Therefore, this study specifically targets respondents who have visited at least one of the eight most renowned scenic spots within the Hengduan Mountains. These sites were selected based on a comprehensive set of criteria, including their officially certified 5A level, annual visitor numbers, and prominence in regional tourism marketing. The final list comprised the following eight destinations: Daocheng Yading, Yulong Snow Mountain, Siguniang Mountain, Zheduo Mountain, Genie Sacred Mountain, Gongga Snow Mountain, Haizi Mountain, and Shiziwang Peak. To verify whether respondents had visited one of these specific scenic areas, the first question in the survey presented this list and asked for confirmation.

The sample size required for this study was calculated using G*Power 3.1.9.2 (Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany). Following the conventions of previous research, a t-test and linear multiple regression method were selected in G*Power. With reference to [78], the effect size was set to a medium level, and the significance level was set at 0.05. The calculation results indicated that a minimum 89 participants were required to achieve a statistical power of 0.95.

Given that the potential respondents were dispersed throughout China, an online survey method was adopted. The survey was disseminated via “Wenjuanxing”, a professional online survey platform, as well as social media platforms such as “Xiaohongshu” and “Weibo”, between 11 July and 9 August 2025, which resulted in the participation of 782 individuals. To enhance sample diversity, we employed a soft quota method by disseminating surveys across a diverse range of accounts and groups, such as regional community forums and interest groups pertaining to different age groups. We specifically monitored the responses in real time against key demographic variables to avoid the predominance of one single demographic group. Through the first screening question, 447 tourists who had visited the Hengduan Mountains were identified. Following this, their questionnaires were screened, eliminating those with a per-question completion time of less than 10 s or showing obvious response patterns. In addition, an attention check item was embedded in the questionnaire, which required the respondent to complete a simple calculation problem. Those who gave an incorrect answer were also eliminated. Ultimately, 391 valid questionnaires were collected, representing a validity rate of 87.5%.

Table 3. Demographic profiles.

Variable	Characteristics	Frequency	Percentage
Gender	Male	172	44
	Female	219	56
Age	<25	134	34.3
	25–40	128	32.7
	41–65	122	31.2
	>65	7	1.8
Monthly salary	<3000	109	27.9
	3000–5000	58	14.8
	5001–8000	88	22.5
	>8000	136	34.8
Education	Junior high school and below	7	1.8
	Senior high school or technical secondary school	13	3.3
	Junior college or bachelor’s degree	291	74.4
	Master’s degree or above	80	20.5

summarizes the sample’s demographics. There were slightly more female than male respondents, accounting for 56%. The respondents were relatively evenly distributed across the three age groups of under 25, 25–40, and 40–65 years old, which is consistent with the physical profile of typical adventure tourists visiting high-altitude destinations, for whom such travel is often more accessible. Only 1.79% were seniors aged 65 and above, which is consistent with the fact that the high altitude of the mountain attractions in this survey deterred many elderly tourists. Concurrently, it is also possible that this distribution mirrors the user base of the social media platforms, such as Xiaohongshu and Weibo, employed for recruitment, which are predominantly utilized by younger, urban demographics in China. Slightly over half of the respondents (57.34%) reported a monthly income exceeding RMB 5000. In terms of educational background, the overwhelming majority held an associate or bachelor’s degree, accounting for 74.42%.

4. Results and Empirical Study

4.1. Common Method Bias

Common method bias was assessed using Harman’s single-factor test [79]. An exploratory factor analysis was conducted on all measured variables without rotation. Seven factors were extracted, which collectively accounted for 73.604% of the total variance. The first factor explained 26.663% of the variance, which was well below the critical threshold of 50% [80]. This indicates that a single factor did not account for the majority of the variance, thus confirming the absence of significant common method bias in the data. Furthermore,

Table 4. Item descriptive statistics and full collinearity VIF.

Construct	Items	Loadings	Mean	SD	Skewness	Kurtosis	VIF
SK	SK1	0.722	3.34	0.816	−0.216	0.083	2.042
	SK2	0.825	3.39	0.795	−0.345	0.254	2.299
	SK3	0.928	3.76	0.723	−1.028	2.186	1.720
	SK4	0.625	3.45	0.776	−0.263	0.048	1.462
PV	PV1	0.799	3.91	0.743	−0.562	0.606	1.597
	PV2	0.86	3.99	0.778	−0.681	1.112	1.791
	PV3	0.8	3.84	0.878	−0.525	0.105	1.393
PS	PS1	0.86	3.88	0.849	−0.803	1.013	1.988
	PS2	0.906	3.76	0.896	−0.560	0.233	2.434
	PS3	0.832	3.48	0.919	−0.428	0.109	1.756
RS	RE1	0.817	4.24	0.843	−1.249	1.934	1.580
	RE2	0.853	4.01	0.786	−0.650	0.610	1.617
	RE3	0.825	4.10	0.840	−0.790	0.551	1.626
SE	SE1	0.855	3.65	0.818	−0.554	0.529	1.854
	SE2	0.875	3.45	0.890	−0.383	0.175	2.034
	SE3	0.837	3.70	0.844	−0.707	0.692	1.670
AWE	AWE1	0.864	4.38	0.679	−1.026	1.660	2.407
	AWE2	0.925	4.42	0.655	−0.966	0.999	4.667
	AWE3	0.9	4.44	0.680	−1.101	1.130	3.962
	AWE4	0.869	4.45	0.630	−1.009	1.784	2.393
CB	CB1	0.799	4.10	0.640	−0.270	0.093	1.662
	CB2	0.924	4.28	0.641	−0.675	1.401	2.833
	CB3	0.891	4.20	0.725	−1.180	3.186	2.422

Note: CB = climate change mitigation behavior intentions; SK = Subjective knowledge on climate change; PS = perceived severity; PV = perceived vulnerability; RE = response efficacy; SE = self-efficacy.

summarizes the item-level statistics, including collinearity assessed via the full VIF procedure. Cronbach’s α estimates of all the seven constructs were greater than the suggested cut-off values of 0.70, ranging from 0.757 to 0.913. A vast majority of items met Kock’s 3.3 rule [81], indicating the collinearity is not a general concern in the measurement model. Although two items (AWE2 and AWE 3) exhibited VIF values above the threshold (4.667 and 3.962 respectively) but below 5.0, it is important to note that these items demonstrated high outer loadings and the corresponding construct demonstrated high composite reliability. Therefore, we retained these items for their theoretical relevance. Their inclusion did not pose a critical threat to the model’s interpretation.

4.2. Exploratory Factor Analysis

An EFA was performed using the principal component extraction technique and varimax rotation method on 23 measurement items in SPSS 27.0 (IBM Corp., Chicago, IL, USA) to assess the factor structure of these items [72]. The Kaiser–Meyer–Olkin (KMO) value and Bartlett’s test of sphericity were used to test the suitability of factor analysis for this study. As presented in

Table 5. KMO and Bartlett’s test.

Kairser–Meyer–Olkin Measure of Sampling Adequacy		0.812
Bartlett’s Test of Sphericity	Approx. Chi-Squared	4974.957
	df	325
	Sig.	<0.001

, the KMO value was 0.812, and Bartlett’s test of sphericity was highly significant (p < 0.001), affirming that factor analysis is appropriate for this study [73]. The EFA revealed seven components with eigenvalues greater than 1, accounting for 73.604% of the total variance. The EFA results confirmed that all factors were significantly loaded under the corresponding constructs specified by the relevant literature and higher than the threshold of 0.5 [82].

4.3. Measurement Model

In this study, we conducted data analysis with Smart PLS 4.0 (SmartPLS GmbH, Hamburg, Germany). For parameter estimation, we adopted the path weighting scheme with a maximum of 300 iterations, while for significance testing, we employed the bootstrapping method with 5000 samples. The analysis was configured to generate bias-corrected and accelerated confidence intervals, which provided a more robust estimate. Two-tailed tests were used for all significance tests, with a significance level of 0.05. Through calculation, the factor loading coefficients and the reliability and validity of the model were obtained, and the analysis results are presented in Table 4 and

Table 6. Measurement model’s reliability, convergent validity and discriminant validity.

Construct	AWE	CB	SK	PS	PV	RE	SE
Awe (AWE)	0.890	0.667 [0.592, 0.735]	0.167 [0.108, 0.274]	0.321 [0.230, 0.411]	0.390 [0.279, 0.488]	0.264 [0.158, 0.373]	0.307 [0.207, 0.400]
Climate mitigation behavior Intention (CB)	0.589	0.873	0.311 [0.193, 0.434]	0.191 [0.094, 0.289]	0.285 [0.154, 0.409]	0.313 [0.188, 0.434]	0.461 [0.351, 0.565]
Subjective knowledge on climate change (SK)	0.200	0.288	0.783	0.067 [0.065, 0.152]	0.210 [0.113, 0.331]	0.085 [0.066, 0.200]	0.412 [0.302, 0.522]
Perceived severity (PS)	0.282	0.16	0.065	0.867	0.524 [0.427, 0620]	0.156 [0.076, 0.268]	0.197 [0.116, 0.290]
Perceived vulnerability (PV)	0.323	0.228	0.178	0.421	0.820	0.153 [0.111, 0.252]	0.191 [0.085, 0.307]
Response efficacy (RE)	0.226	0.26	0.086	0.129	0.099	0.832	0.353 [0.230, 0.462]
Self-efficacy (SE)	0.267	0.386	0.325	0.162	0.149	0.287	0.856
Cronbach’s alpha	0.913	0.843	0.822	0.833	0.757	0.778	0.817
rho_c	0.939	0.905	0.861	0.900	0.860	0.871	0.891
AVE	0.793	0.762	0.613	0.751	0.673	0.692	0.732

Note: The bold values on the diagonal are the square roots of the AVE, the values below the diagonal are the correlation coefficients, and the values above the diagonal are the HTMT values (in italics) with CIs in the square brackets.

.

As displayed in Table 6, all the constructs revealed good reliability. Cronbach’s α estimates of all seven constructs were greater than the suggested cut-off value of 0.70, ranging from 0.757 to 0.913. Composite reliability (CR) is a robust indicator for assessing the internal consistency of a model’s constructs. The results suggest that CR estimates of all the constructs ranged from 0.861 to 0.939, well above the 0.7 threshold, confirming that the model has good reliability.

Afterwords, construct validity was evaluated via convergent and discriminative validity. Average variance extracted (AVE) and factor loadings serve as indicators of the model’s convergent validity [82]. All the factor loadings had significance greater than 0.6, ranging from 0.625 to 0.928. AVE measures the proportion of variance in the observed indicators that is explained by the underlying construct. As shown in Table 5, the results suggested that the AVE estimates of all constructs ranged from 0.613 to 0.793, above the recommended threshold of 0.5 [83]. Convergent validity was supported. In addition, the square root of each construct’s AVE was greater than its correlations with all other constructs, fulfilling the Fornell-Larcker criterion [83]. Moreover, all heterotrait-monotrait (HTMT) ratios were below the threshold of 0.85, confirming that the measurement model exhibited good discriminant validity.

4.4. Hypothesis Testing

To ensure the reliability of path coefficient estimation, in this study, we employed Smart PLS 4.0 (SmartPLS GmbH, Hamburg, Germany) to conduct a path-level variance inflation factor (VIF) test on the structural model. As shown in

Table 7. Direct paths between constructs and inner VIF values in structural model.

Direct Path	β	SD	t	p	95% CI	VIF
H1: TA→CB	0.173	0.033	5.234	<0.001	[0.130, 0.324]	1.160
H2: CA→CB	0.126	0.034	3.743	<0.001	[0.073, 0.181]	1.119
H3a: TA→AWE	0.294	0.051	5.706	<0.001	[0.269, 0.440]	1.054
H3b: CA→AWE	0.213	0.053	4.049	<0.001	[0.127, 0.299]	1.157
H3c: AWE→CB	0.588	0.040	14.794	<0.001	[0.524, 0.655]	1.230
H4a: SK × TA→AWE	−0.102	0.048	2.130	0.033	[−0.276, −0.021]	1.165
H4b: SK × CA→AWE	0.018	0.050	0.349	0.727	[−0.241, 0.029]	1.235

Note: TA, threat appraisal; CB, climate change mitigation behavior; CA, coping appraisal; AWE, awe; SK, Subjective knowledge on climate change.

, the VIF values for all variables were below the critical threshold of 3.0. These results confirm the absence of significant multicollinearity issues in the model, ensuring the unbiasedness and stability of path coefficient estimation and providing a reliable foundation for hypothesis testing. Further model fit tests revealed that the R² values for the endogenous variables AWE and behavioral intention were 0.209 and 0.405, respectively, and R² adjusted values for the two variables were 0.199 and 0.400, respectively, both exceeding the standard acceptance value of 0.10, indicating that the model had strong predictive capability. Additionally, all Q² values for the endogenous variables (Q²_awe = 0.177, Q²_CB = 0.170) were above the standard threshold of 0, demonstrating that the model was effect in predicting correlations among variables. Moreover, the SRMR value of the model in this study was 0.072, which is below the accepted standard of 0.08, indicating a good model fit. In summary, the model exhibits satisfactory adaptability and is suitable for subsequent analysis.

For this study, we employed the bootstrap sampling method, setting the maximum number of iterations for path weighting to 300 and the number of bootstrap samples to 5000. The standardized effects of the main paths in the structural model are summarized in Figure 2. The results show that threat appraisal (β = 0.173, t = 5.234, p < 0.001) had a significant positive impact on climate change mitigation behavior intention. H1 is thus supported. In the meantime, coping appraisal (β = 0.126, t = 3.743, p < 0.001) also played a significant positive role. Therefore, H2 is supported. Coping appraisal (β = 0.213, t = 4.049, p < 0.001) and threat appraisal (β = 0.294, t = 5.706, p < 0.001) both had a significant positive effect on awe. H3a and H3b are hence supported. In addition, awe (β = 0.588, t = 14.794, p < 0.001) played a significant positive role in climate change mitigation behavior intention. H3c is thus supported. The strong positive relationship between awe and climate change mitigation behavior intention is not only the closest in the model but also dominates the direct effects. To illustrate the practical implications, a one-standard-deviation increase in awe is associated with a 0.588 standard deviation increase in behavior intention. This suggests that interventions that successfully elicit the emotion of awe can be more effective in motivating climate change mitigation behavior intentions than those that focus solely on encouraging cognitive appraisal. The interaction between subjective knowledge on climate change and threat appraisal (β = −0.102, t = 2.130, p < 0.05) had a significant negative effect on awe. Thus, H4a is supported. The interaction between Subjective knowledge on climate change and coping appraisal had a positive impact on awe. This impact, however, was not significant. Therefore, H4b is not supported.

As shown in

Table 8. Indirect paths in structural model.

Indirect Path	β	SD	t	p	95% CI
H3d: TA→AWE→CB	0.151	0.03	4.968	0.000	[0.103, 0.203]
H3e: CA→AWE→CB	0.11	0.027	4.025	0.000	[0.069, 0.159]

, the indirect effect of threat appraisal through awe (β = 0.151, t = 4.968, p < 0.001) on climate change mitigation behavior intention was positive and significant. H3d is supported as a result. Similarly, the indirect effect of coping appraisal through awe (β = 0.11, t = 4.025, p < 0.001) on climate change mitigation behavior was also positive and significant, thus supporting H3e.

After confirming the significant mediating effect of awe, we estimated the indirect effects for different levels of the moderator. The significant negative interaction between threat appraisal and subjective knowledge indicates that the positive relationship between threat appraisal and behavior intentions though awe is weakened as subjective knowledge increases. As illustrated in Figure 3, this relationship was stronger when subjective knowledge was low (−1SD, simple slop = 0.203) and the relationship became weaker when subjective knowledge was high (+1SD, simple slope = 0.098).

5. Discussion and Implications

5.1. General Discussion

Conducted in the Hengduan Mountain area in Southwest China, this study explored the generation mechanism underlying Chinese mountain tourists’ intentions toward climate change mitigation behaviors. The results reveal that mountain tourists’ understanding of climate change-induced risks and their perception of the efficacy of climate change mitigation behaviors directly and positively influence their mitigation behavior intentions. Meanwhile, through mountain tourism experience, tourists are likely to focus less on themselves, appreciate the magnificence of high mountains, and feel a sense of their own smallness, leading to a feeling of awe. This prompts them to reflect on the importance of preserving the ecosystems and landscapes of highly mountainous areas, which enhances their intentions to enact behaviors in daily life and while traveling to fight climate change. Overall, this study underscores the major positive effects of PMT antecedents and awe on climate change mitigation behavior intentions among tourists.

Specifically, the two antecedent constructs derived from the PMT framework, namely, threat appraisal and coping appraisal, significantly predict mountain tourists’ climate change mitigation behavior intentions, consistent with the results of [84,85]. This study extends these findings to the high mountain tourism context. The high mountain tourism experience can help tourists understand their role within the broader natural context and instill in them a heightened sense of responsibility and obligation toward the natural environment, thus prompting them to go through the threat appraisal and coping appraisal processes. Threat appraisal allows tourists to evaluate the severity of, and vulnerability to, consequences of climate change in highly mountainous areas. Due to global warming and snow cover declines, mountain tourism destinations are undergoing challenges such as increased natural hazards, changes in accessibility, and reduced attractiveness of landscapes [3]. The results show that respondents perceive high levels of severity and vulnerability, suggesting that they believe that climate change will adversely impact their mountain tourism experience. Consequently, a high level of threat appraisal increases their willingness to enact climate change mitigation behaviors. Climate change mitigation encompasses a suite of measures aimed at reducing or preventing greenhouse gas emissions [86]. Such measures can be taken both in daily life and while traveling, such as acquiring more information about climate change, adopting a green lifestyle, and reducing one’s carbon footprint during traveling. Similarly, through coping appraisal, tourists assess the response efficacy and self-efficacy of undertaking the aforementioned mitigation behaviors. Thus, their confidence in their ability to fight against climate change and in its effectiveness also enhances their mitigation behavior intentions.

Tourists’ threat appraisal and coping appraisal significantly contribute to their experience of awe. This relationship underscores the effectiveness of a well-preserved natural environment and thoughtfully presented mountain attractions. High mountain attractions typically feature unique landscapes, breathtaking scenery, and rich biodiversity; however, these aspects also mean that highly mountainous areas are among the most vulnerable to climate change. Visits to those areas connect tourists with striking scenery and the power of nature and provide opportunities such as mountaineering, hiking, skiing, or simply sightseeing, which can evoke feelings of smallness and insignificance, laying the foundation for the generation of awe. Moreover, threat appraisal allows tourists to reflect on how severely climate change will negatively impact the mountain tourism experience, whereas coping appraisal leads them to contemplate on how effective mitigation behaviors can help in the fight against climate change. Rich in reflection and contemplation, the cognitive processes of threat appraisal and coping appraisal function as powerful stimuli for strong emotional responses. The magnificence of scenery and invincibility of natural power can create an awe-inspiring experience when tourists visit highly mountainous areas. Additionally, while threat appraisal and coping appraisal both demonstrate a significant influence on the generation of awe, threat appraisal shows a comparatively stronger influencing power, which is consistent with prior research proving that threat plays a role in the activation of emotional responses [24]. Notably, the findings reveal a discrepancy that emotional response demonstrates a stronger influence on climate change mitigation behaviors while threat appraisal and coping appraisal exhibit limited direct impacts. This is consistent with the framework of the SOR model, which suggests that stimuli work as catalysts for human responses rather than direct influencers [42]. The cognitive process of threat appraisal and coping appraisal does not immediately lead tourists to feel the urge to enact mitigation behaviors. Rather, this is uncovered through emotional reflection. As a matter of fact, threat appraisal and coping appraisal do provide a psychological environment for the emergence of mitigation behavioral intentions. However, it is the emotional response associated with the natural setting during the cognitive process that more directly impacts behavioral intentions, exhibiting a strong mediating effect.

Respondents’ Subjective knowledge on climate change displays a differentiated moderating effect. It negatively moderates the relationship between threat appraisal and the generation of awe in this study. Although counterintuitive, this finding aligns with prior research that revealed that respondents who reported that they were more informed regarding global warming perceived a lower level of climate change risk [87]. It is true that a lack of factual knowledge would stop people from enacting pro-environmental behaviors [88]. However, it was revealed that perceived knowledge was not equal to objectively assessed factual knowledge [89]. Objective knowledge refers to how much an individual actually knows about a topic, while subjective knowledge represents the degree to which an individual believes that they understand a topic [90]. The construct of knowledge about climate change used in our study represents a typical type of subjective knowledge since it was measured based on the respondents’ perception of their knowledge about climate change. Influenced by the Dunning–Kruger effect [91], people tend to overestimate their climate knowledge. A lower level of actual knowledge was related to the highest overestimation [92]. Therefore, people who claimed to possess abundant knowledge regarding climate change were very likely to overestimate their actual knowledge level, which led to a negative relationship with threat appraisal, thus reducing the generation of awe. Furthermore, alternative theoretical perspectives also support this finding. Information avoidance [93] might prevent individuals with high subjective knowledge from fully processing threatening information. This defensive avoidance could be a coping strategy to manage the “eco-anxiety” [94] that is associated with fully confronting the crisis. In addition, the phenomenon of risk normalization [95] could provide another explanation. Repeated exposure to climate-related information though both media and personal travel experience may reduce its novelty or salience. This normalization process would increase the emotional impact required for awe to emerge, as the stimulus is gradually appraised as a familiar and normalized risk.

On the other hand, the moderating effect of Subjective knowledge on climate change between coping appraisal and awe is not significant. This finding is inconsistent with the many previous studies that found that prior knowledge about climate change was very likely to enhance perceived efficacy [84] and encourage pro-environmental behaviors [92]. However, a study on young people living in Australia found that the relationship between Subjective knowledge on climate change and self-efficacy was subject to gender and age differences [96]. Similarly, it was found that trust played a crucial mediating role between knowledge and climate change concern, with knowledge being uncorrelated with climate change concern among respondents who had a lower level of trust [97]. In this sense, it can be concluded that the relationship between knowledge and coping appraisal is complex. In summary, a significant moderating effect by knowledge was not observed.

5.2. Theoretical Implications

This study has several important theoretical implications. Firstly, it extended the Protection Motivation Theory (PMT) through an integrated Stimulus–Organism–Response model (SOR), thereby developing a comprehensive framework that clearly identifies the psychological mechanism between protection motivation constructs and climate change mitigation behavior intentions. We posit that threat and coping appraisal function as distal catalysts and not as immediate drivers of behavior intentions. Instead, mountain tourists experienced awe due to cognitive processes in response to risks and challenges brought on by climate change, which acts as the proximal driver in directly motivating behavior intention. Furthermore, this study specifies specific boundary conditions for this theoretical model. The emotional pathway is likely to be particular to awe-eliciting settings, especially high-altitude and spectacle-rich contexts such as the Hengduan Mountains. In such settings, the vastness and grandeur of nature can overwhelm cognitive processing and make emotion a more direct motivator for behavior intention. Secondly, although climate change mitigation has been widely studied, research in the context of highly mountainous tourism destinations is relatively limited. High mountain tourism provides opportunities for both leisure and education. By examining the relationships among threat appraisal, coping appraisal, awe, and climate change mitigation behavior intentions, this study empirically validated the significant positive influence of high mountain tourism experiences on tourists’ behavioral intentions against climate change. This is crucial for people living in cities built from steel and concrete who are seldom able to feel the power of nature. Lastly, this study aligns with prior research demonstrating that unlike objectively assessed knowledge, subjective knowledge on climate change negatively moderates the relationship between threat appraisal and awe while displaying no significant moderating effect between coping appraisal and awe. This finding contradicts the public assumption that simply raising people’s awareness of climate change is an effective way of enhancing their intention to combat this pressing global issue.

5.3. Practical Implications

For mountain tourism destination managers and marketers aiming to encourage tourists to adopt climate change mitigation behaviors, this study offers the following insights. Firstly, our findings indicate that placing greater emphasis on the threats posed by climate change to mountain tourism destinations can enhance the positive effect of awe in promoting behavioral change. Thus, marketing communication strategies should delve into the audience’s cognitive processes when responding to climate change threats and emphasize the use of diverse audiovisual elements to evoke awe among viewers. Specifically, instead of a sign that simply reads “Protect the Glaciers”, on-site interpretive signage can play a more effective educational role by communicating information with a focus on clarifying the severity of climate change threats to mountain ecosystems, such as glacial retreat or biodiversity loss, and their anthropogenic impacts. In addition, audio and visual cues using AR or VR technologies could translate the severity of climate change impacts into tangible tourist experiences. For example, an AR application can enable tourists to view the contemporary landscape with an overlay showing how it looked decades ago to demonstrate the process of glacier retreat. The application of soundscapes can also create an immersive and emotionally resonant context for presenting severity information. Soundscape stations can be built at pivotal locations, playing examples of natural sounds that are diminishing with climate change. Highlighting accessible pathways to implement these pro-environmental actions is also important. When tourists begin a hike, an in-app nudge could lead to a notification that clearly communicates the sensitivity of the ecosystem and the effectiveness of personal emission-reduction behaviors, such as carpooling or choosing public transportation, as coping responses. During the trip-planning process, official websites or OTAs could provide low-carbon itinerary options rather than conventional routes. The prompt on the website could suggest that theses itineraries prioritize electric shuttle transport, local food-based meal options, and certified ecolodges.

In addition, building upon general public awareness campaigns, climate change educational initiatives should evolve from fostering “awareness” toward promoting “actionable competencies”, specifically aimed at providing accurate knowledge of climate change and practical skills. Our findings support the need for educational programs to move beyond the dissemination of general knowledge and instead emphasize the provision of actionable, context-specific information. To maximize effectiveness, such programs should be tailored to distinct behavioral domains (e.g., transportation, food consumption) and carefully address both the real and perceived constraints that individuals encounter within each respective domain. For example, instead of vague appeals such as “Fight against Climate Change”, more concrete and context-specific actions could be suggested during travel. At a trailhead, a sign could read “Visit the central area on foot or by rental bicycle” or “Buy a locally made souvenir to reduce transportation emissions”. Through this targeted and context-aware educational intervention, tourists can transform subjective self-perceptions of understanding into an objective knowledge framework guiding specific mitigation behaviors, thereby overcoming barriers to action and genuinely enhancing the adoption rates of low-carbon practices. Our findings caution against simply increasing tourists’ subjective knowledge, which may foster overconfidence. To address this, educational interventions should aim to calibrate the gaps between tourists’ subjective knowledge and actual or objective knowledge. One effective way of achieving this could be the use of interactive quizzes with immediate feedback on DMO websites or in visitor centers. Quizzes should cover effective and actionable climate change behaviors rather than general climate science. The completion of the quizzes with a satisfactory score could earn “green points”, which could be converted into a small monetary donation to a climate change cause or redeemed in the form of low-carbon experiential rewards.

6. Conclusions

This empirical investigation of tourists in China’s Hengduan Mountains provides robust evidence for a novel cognitive–emotional pathway to climate change mitigation and yields several distinctive theoretical and practical insights that advance the current understanding of pro-environmental behaviors in eco-sensitive tourism contexts. Our study makes a significant contribution by bridging the PMT and S-O-R frameworks, demonstrating that in awe-inspiring highly mountainous contexts, cognitive appraisals serve as distal catalysts, while awe acts as the proximal driver or behavioral intentions. This finding underscores the necessity of incorporating discrete emotions into cognitive models to fully understand pro-environmental action in spectacular natural settings. Another intriguing finding lies in the moderating role of subjective knowledge, which weakened the link between threat appraisal and awe, revealing a potentially counterproductive effect of overconfidence. These findings have significant implications for destination management practices, particularly regarding developing awe-elicited and fact-based intervention strategies. Destination management organizations should design strategies that collaboratively elicit awe through the grandeur of the landscape while simultaneously providing clear and feasible mitigation action.

7. Limitations and Recommendations for Future Research

This study has three major limitations. Firstly, the reliance on self-reporting raises potential endogeneity concerns. Although we have rigorously tested our hypothesized model, the risk of endogeneity cannot be fully ruled out due to omitted variable bias, reverse causality, and measurement errors. Future research should address these issues by incorporating longitudinal or experimental designs to establish temporal precedence and causality. Furthermore, multi-group or interaction analyses should also be employed to compare groups based on prior visits, age, or gender, to test the model’s robustness. Secondly, the adoption of an online data collection method in a single country could have led to limitations in sample representativeness. There is a possibility that only those who had a certain level of digital literacy were able to participate in the survey. Future research should adopt a mixed-methods approach, incorporating both online and on-site data collection so as to enhance the generalizability of our findings to the broader population of mountain tourists. Thirdly, the measurement of awe did not account for the timing of participants’ most recent mountain tourism experience. According to Construal Level Theory [98], the elicitation of awe is closely tied to temporal distance. Future research could incorporate this temporal variable to examine how different levels of temporal distance influence the experience and effects of awe. In addition, future studies could also distinguish between state and trait awe to enable greater measurement precision. A final limitation related to the study model is that the construct of subjective knowledge on climate change focused on participants’ own perceptions of their climate change knowledge, and this variable was revealed to be non-significant in moderating the relationship between coping appraisal and awe. Future research should add an objective climate knowledge construct into the model to test the different roles played by objectively assessed knowledge in this model.