Abstract
This study proposes a model on the relationship between cultural ecosystem service (CES) and visitors’ subjective well-being (SWB). The model consists of the four constructs of CES, SWB, place attachment, and event attachment, which are developed from literature of multiple disciplines. The model is tested with primary data collected in a China’s national park and its 10th Flower Expo. The results indicate that perceived value of CES has strong direct effect on visitors’ SWB, as well as a serial mediating effect through place attachment and event attachment. Perceived value of CES is an important antecedent of place attachment, event attachment, and visitors’ SWB. When controlling the effect of CES, event attachment exhibits a significant effect on visitors’ SWB, but the effect of place attachment is insignificant. In addition to the proposed model, the study’s findings provide fresh empirical evidence to further understand each of the four constructs and their interrelationship.
Keywords
Introduction
A growing number of tourism and recreation studies is emerging that investigate the interdependence relationship among tourism and recreation activities, ecosystem protection, and human well-being from ecosystem service perspective (Chen 2020; Drius et al. 2019; Ghermandi, Camacho-Valdez, and Trejo-Espinosa 2020; Zhang and Xu 2020). Ecosystem services can be classified in supporting, provisioning, regulating, and cultural categories (Millenium Ecosystem Assessment [MEA] 2005). Supporting services include those that reflect biophysical structure and process of ecosystem such as nutrient cycling, soil formation, and primary production. Provisioning services refer to the materials human obtains from ecosystem such as food, fresh water, and fuel. Regulating services consists of those climate regulation, flood regulation, and water purification etc. Cultural services refer to the non-material benefits people receive from ecosystem through spiritual enrichment, cognitive development, reflection, recreation, and esthetic experience (MEA 2003, 2005). Nature-based tourism, recreation and esthetic are important cultural ecosystem services (CES) (Willis 2015).
The relationship between ecosystem services and human well-being has been examined in a broad spectrum of fields including environmental sciences, ecology, biodiversity conservation, water resources, green sustainable science technology, forestry, geography, economics, recreation, and tourism (Chen, Lehto, and Cai 2013; Wang, Zhang, and Cui 2021; Willis 2015). As such, a wide range of cross-disciplinary methodology such as medical (physiological) indication, psychological method, spatial databases, economic modeling has been used (Pröbstl-Haider 2015). These studies from various disciplines report a general association between ecosystem service and human health and well-being. For example, Wheeler et al. (2012) found a positive correlation between coastal proximity and human health. Van Herzele and de Vries (2012) also noted that the neighborhood with more green space showed better health indicators. Yet “clear evidence is not always found” (Martinez-Juarez et al. 2015, 66).
Of the four types of ecosystem services, the supporting, provisioning, and regulating services are measured more in economic terms and easily quantifiable in their relationship to human well-being. CES, however, is less so and its relationship to human well-being has been underrepresented in literature (Chan, Satterfield, and Goldstein 2012; Kosanic and Petzold 2020; Zhang et al. 2020). The MEA’s definition of CES equates service with benefit, which may lead to confusion of blurring the border between CES and well-being. For example, Gladkikh, Gould, and Coleman (2019) identified 16 CES typologies, among them, mental health and social relations are also belong to human well-being according to MEA. Similarly, Bryce et al. (2016) expressly viewed CES as subjective well-being (SWB) indicators. Therefore, this definition method has been criticized and an alternative definition is proposed, which views CES as the interaction between ecosystem and human (Fish, Church, and Winter 2016). Based on this new definition, a significant amount of extant research on CES and well-being operationalizes CES as exposure to nature or accessibility to nature area (Lee et al. 2011; Martinez-Juarez et al. 2015).
However, the exposure is not equal to benefits, but just provides a possibility to experience them. The cultural benefits from ecosystem services in general are highly complex and subjective, influenced by various individual, contextual, and cultural value factors (Zhang and Xu 2020). Understanding how CES influences well-being has even been more elusive. Extant CES and well-being studies usually use horizontal comparison (i.e., nature vs. urban setting) or vertical comparison (i.e., before vs. after design) approach to examine the benefits, which are mainly based on two theories of the attention restoration theory (ART, Kaplan and Kaplan 1989) and the stress reduction theory (SRT, Ulrich 1983; Ulrich et al. 1991). However, these approaches fail to reveal the complexity and subjectivity of CES benefits. Exposure to nature may enhance restoration of diminished functional resources and capabilities. The ART and SRT explain the restorative effects of nature from two different processes. The ART can explain the recovery from directed attention fatigue, while the SRT can describe the recovery from psychophysiological stress. However, both theories are limited in capturing the CES benefits. They are concerned with the characteristics of nature, while ignoring individual and social factors which influence individual perception and feeling of the natural environment.
The current study leverages the setting of forest ecosystem and the 10th China flower Expo in addressing the complexity and subjective nature of CES benefits and its influence on tourist well-being. The Expo took place between 21st May and 2nd July of 2021 in Dongping National Forest Park, Chongming Island, Shanghai, China. The forest park is the biggest plain plantation (3.55 km2) in the middle and lower reaches of Yangtze River. It has flat terrain and rich plant resources with nearly a thousand of species such as ginkgo and camphor. Its forest cover is up to about 80% of the park. China Flower Expo is known as country’s “Olympic” for its flower industry, being held every four years. The 10th China Flower Expo was hosted in Dongping National Forest Park, the first time for the event to be staged in a forest on a rural island.
The forest park and the flower expo together construct a compound natural environment where tourists experience various CES. Specifically, the study introduces CES concept into the frontier area of tourist well-being (Zhang, Zhang, and Bai 2021) by linking with other important concepts in tourism via a relational model. The study focuses on domain subjective well-being (SWB), defined as the happiness feeling from park visit and nature-themed event participation. The study also examines if nature-themed events staged in national forest park could enhance the effect of perceived CES on tourist SWB. These research objectives cumulates in a proposition of modeling the effects of CES on tourist SWB with place attachment and event attachment as mediators.
Literature Review and Theoretical Framework
CES and Its Valuation
CES is defined as the non-material benefits human obtain from ecosystem (Ament et al. 2017; MEA 2003). MEA recognizes 10 types of CES, including spiritual and religious values, esthetic values, recreation and ecotourism, cultural diversity, knowledge systems, educational values, inspiration, social relations, sense of place, and cultural heritage values (MEA 2003). Other have proposed additions, such as artistic, bequest, existence, identity, ingenuity, life teaching, mental health, perspective (Gladkikh, Gould, and Coleman 2019; Kosanic and Petzold 2020). Of all, recreation, esthetics, and cultural heritage are the three most mentioned in extant literature. Since it is difficult to assign a monetary value to CES due to their intangibility and subjectivity (Chan, Satterfield, and Goldstein 2012), social-cultural evaluation methods are often used for CES measures. The methods include interview, questionnaire survey, focus group, photo-questionnaire, and participatory GIS. One recent trend is the use of geolocated social media data as quantitative assessment of CES by tabulating the number of photos of CES occurrences in geographic unit (Zhang et al. 2020). Such approach does not enable any analysis of relationships between CES and tourist well-being at individual level.
A close relative of CES is social value of ecosystem services (SolVES), defined as the non-market values perceived by ecosystem stakeholders (Sherrouse, Semmens, and Clement 2014). CES is one type of ecosystem services, while SolVES is a valuation method of social values in ecosystem services. In the forest setting, researchers have identified 13 aspects of SolVES, including esthetic, biodiversity, cultural, economic, future, historic, intrinsic, learning, life sustaining, recreation, spiritual, subsistence, therapeutic (Brown and Reed 2000; Clement and Cheng 2011; Rolston and Coufal 1991; Sherrouse, Semmens, and Clement 2014). Some of these types such as recreation, spiritual, and esthetic are related to CES, others such as biodiversity and life sustaining are related to supporting, provisioning, and regulating services. The present study integrated the typology of social value by Sherrouse, Semmens, and Clement (2014) and that of social-cultural values by Breyne, Dufrêne, and Maréchal (2021) in developing a series of quantifiable CES values. They are measured in the study in terms of tourist perceptions, that is tourist perceived value of CES.
CES and Subjective Well-Being
MEA’s definition and category of CES is not without critics (Bryce et al. 2016; Fish, Church, and Winter 2016). Bryce et al. (2016) pointed to the conflation of services and benefits in uniquely defining CES and human well-being. Fish, Church, and Winter (2016) provided a place-based CES framework and defined CES as the interactions between environmental spaces and the cultural or recreational practices that take place within them. As such, cultural benefits including identities (e.g., sense of place), experiences (e.g., inspiration), and capabilities (e.g., knowledge) are viewed as indicators of reflecting place-based SWB, rather than CES. Bryce et al. (2016) evaluated 15 SWBs (i.e., benefits of CES), including spiritual, therapeutic, esthetic, sense of place. These SWBs are similar to the aforementioned social value or social-cultural value of CES.
In addition to the blurring overlaps between CES and well-being, various definitions from different disciplines further complicate the concept of well-being. In ecology and environment fields, human well-being is the often used term, and defined as “a human experience that includes the basic materials for a good life, freedom of choice and action, health, good social relationships, a sense of cultural identity, and a sense of security” (Díaz et al. 2006, 1301). Human well-being can be explored from the two perspectives of SWB and objective well-being. SWB refers to the subjective evaluation or satisfaction of basic materials, health, security, and social relations (Yang et al. 2019), while objective well-being is reflected by objective indicators like income and jobs (Vladisavljević and Mentus 2019).
As a concept of positive psychology which studies valued subjective experiences such as well-being and positive individual traits, SWB focuses on happiness and can be defined as pleasure attainment and pain avoidance (Ryan and Deci 2001; Seligman and Csikszentmihalyi 2000) in the three dimensions of life satisfaction, positive affect, and negative affect (Dekhili and Hallem 2020; Diener 1984). Psychological and sociological approaches to well-being are also present in positive psychology (Zhang, Zhang, and Bai 2021). Some well-being frameworks have been proposed to integrate SWB and psychological well-being, such as PERMA. Standing for positive emotion, engagement, relationships, meaning, and accomplishment, PERMA has been adopted in tourism context (Mirehie and Gibson 2020; Seligman 2002). Tourists’ well-being and leisure well-being are domain specific, contributing to the general well-being. In present study, visitors’ SWB refers to visitors emotional experience and life satisfaction of visiting the national forest and the flower expo.
Based on MEA’s framework of the relationship between ecosystem service and human well-being, some studies empirically investigated the relationship between ecosystem service and human well-being (Berbés-Blázquez 2012; Ciftcioglu 2017; Daw et al. 2011; Santos-Martín et al. 2013; Wu 2013). Yang et al. (2019) applied social cultural values in examining the relationship between the respondents’ perceptions of ecosystem service (i.e., provisioning, regulating, and cultural service) and their well-being (i.e., satisfaction of basic materials, health, security, and social relations). However, they did not find the significant connection between CES and social relations in urban area like in other studies such as Berbés-Blázquez (2012). Huang et al. (2020) quantified four types of well-being similar to Yang et al. (2019). They found that some ecosystem services such as water retention at regional scale were significantly associated with SWB. Some other studies examined CES and SWB based on the place-based CES framework of Fish, Church, and Winter (2016). For example, Buchecker and Degenhardt (2015) revealed that regular nearby outdoor recreation had a significant but rather marginal effect on respondents’ self-reported well-being (e.g., positive and negative emotion). Different from the above two approaches, the CES is the central theme of the current study and is operationalized as perceived values of seven cultural ecosystem benefits. This study therefore expands the CES framework of Fish, Church, and Winter (2016) by integrating the subjectivity of nature experiences. In so doing, this study looks into the relationship between perceived value of CES and SWB in national forest park context and hypothesizes that:
H1: Perceived value of CES positively influences park visitors’ SWB.
CES and Place Attachment
Place attachment or sense of place has been explored in various disciplines, such as environmental psychology, environmental sociology, human geography, health sciences, environmental resource management, and tourism and leisure (Boley et al. 2021; Cao, Qu, and Yang 2021; Hausmann et al. 2016). Sense of place is only recently introduced into ecosystem service field and recognized as a type of ecosystem service (MEA 2005). Although different terms and definitions are used in different disciplines, its primary meaning reflects the bond between people and place or ecosystem (Hausmann et al. 2016; Hernandez, Hidalgo, and Ruiz 2014; Stedman 2002; Tuan 1990). Some previous studies in ecosystem service view sense of place as a CES (Ryfield et al. 2019), others regard it as SWB (Bryce et al. 2016). The present study differentiates the CES, SWB, and sense of place. Following the tradition of environmental psychology, this study adopts the term of place attachment, equate it with sense of place, and conceptualizes it in the two dimensions of place identity and place dependence (Boley et al. 2021; Hernandez, Hidalgo, and Ruiz 2014). This study defines CES as visitors’ perception of cultural benefits provided by forest ecosystem service, place attachment as the positive affective bond between visitors and the forest park, and SWB as visitors’ emotional experience of the forest park and the flower expo.
The relationship between CES and place attachment has been explored in previous studies (e.g., Faccioli et al. 2020; Kaltenborn, Linnell, and Gómez-Baggethun 2020). Kaltenborn, Linnell, and Gómez-Baggethun (2020) examined how attachment to place influenced the appreciation of cultural ecosystem services benefits in northern Norway by resident survey. They found that place attachment positively influenced the perceived importance of ecosystem service benefits for well-being. Different from previous studies with residents as their subjects, this present study examines the relationship between park visitors’ place attachment and their perceived value of CES. When visitors perceive more value from the forest ecosystem service benefits, they may be more likely to form an affective connection to and dependence on the forest park, leading to greater place attachment. The study hypothesizes that:
H2: Perceived value of CES positively influences park visitors’ place attachment.
CES and Event Attachment
Attachment reflects the emotional bond connecting an individual with a particular target (Bowlby 1979). The target could be person, place, brand, or event. The affective bond can be developed if the target satisfies individual’s basic security needs (Bowlby 1958). The concept of event attachment is based on the psychological attachment theory (Ouyang, Gursoy, and Sharma 2017). Limited event attachment research mainly focuses on sport events, such as FIFA World Cup and Rugby World Cup (Ouyang, Gursoy, and Sharma 2017; Prayag et al. 2020). This present study extends the event type to nature-themed events (i.e., the flower expo) and defines event attachment as visitors’ psychological attachment to it (Brown, Smith, and Assaker 2016; Filo, Funk, and O’Brien 2010; Ouyang, Gursoy, and Sharma 2017; Prayag et al. 2020). The attachment to an event could be formed through internalizing it into self-concept, which creates and represents self-image or self-identity (Ouyang, Gursoy, and Sharma 2017).
Previous studies in sport area tested the antecedences of event attachment such as team satisfaction and team attachment (Prayag et al. 2020) and its moderating effects of event attachment on the relationships between trust in government and other event variables (Ouyang, Gursoy, and Sharma 2017). The relationship between event satisfaction and place attachment is also examined (Lee, Kyle, and Scott 2012). No literature has been found to date that explores the relationship between CES and event attachment. When nature-themed events like the flower expo are hosted in a forest ecosystem, both could be blended into a whole, making the nature-themed event become part of the ecosystem. Visitors’ perceived value of CES cannot stand aloof from the nature-themed event. When visitors perceive greater value of CES of the forest park, they may develop greater affective bond with the nature-themed event hosted in the park. The study hypothesizes that:
H3: Perceived value of CES positively influences park visitors’ event attachment.
The Mediating Effect of Place Attachment
Place attachment is essential to various aspects of life (Relph 1976). An increasing evidence demonstrates the contribution of place attachment to human well-being at the neighborhood, community, and city scales (Eyles and Williams 2008; Rollero and De Piccoli 2010; Scannell and Gifford 2017). Scannell and Gifford (2017) reported 13 categories of psychological benefits to explain the ways of place attachment influencing well-being, including memories, belonging, and comfort-security. Place attachment could evoke individual’s memories and belonging, further leading to positive emotion and enhancement of well-being.
Besides their home, tourism destination is another important and meaningful place for tourists (Zhang et al. 2019). Tourists often develop cognitive and affective bond to destinations through memorable experience (Lee and Jeong 2021; Vada, Prentice, and Hsiao 2019; Wang et al. 2020). Empirical studies investigated different influential paths of tourist experience on place attachment and well-being. For example, Wu et al. (2019) examined the effect of tourist authenticity experience on SWB through the mediating effect of place attachment. They found tourists’ place identity of the Palace Museum positively enhanced their SWB, but place dependence had no significant effect. However, Vada, Prentice, and Hsiao (2019) found that hedonic and eudaimonic well-being fully mediated the relationship between memorable tourism experiences and place attachment. In addition, Lee and Jeong (2021) proved the influence of hedonic and eudaimonic well-being on place attachment. Studies focusing on destination residents also showed the two causal directions between place attachment and well-being (Li, Pan, and Hu 2021; Liu et al. 2021). This present study views place attachment as the antecedent variable of SWB and hypothesizes that
H4a: Place attachment positively influences visitors’ SWB.
The empirical evidence on the causal relationship between place attachment and well-being in tourism remains limited; so does the role of place attachment as a mediator on the relationship between CES and SWB. Its mediating effects have been explored in extant literature on environmentally responsible behavior and well-being (Basu, Hashimoto, and Dasgupta 2020; Cheng and Wu 2015). Wu et al. (2019) revealed the mediating effect of place identity between authenticity and subjective well-being of Chinese museum tourists, but did not find the mediating effect of place dependence. Basu, Hashimoto, and Dasgupta (2020) examined and proved the mediating effect of place attachment on the relationship between nature connectedness and human well-being based on a national-scale survey in Japan. Both CES and nature connectedness reflect the connection of people and nature. This study extends this mediating effect to CES and well-being and hypothesizes that:
H4b: Place attachment mediates the relationship between perceived value of CES and SWB.
The Mediating Effect of Event Attachment
The effects of festivals and events on visitors’ well-being have been under-researched, although literature is more abundant on the social and economic impacts of festivals and events (Andersson, Armbrecht, and Lundberg 2017; Yozukmaz, Bertan, and Alkaya 2020). Previous studies have also examined the relationship between event variables such as event quality and event satisfaction and their outcomes such as expenditure and place attachment (Andersson, Armbrecht, and Lundberg 2017; Kim, Choe, and Petrick 2018; Lee, Kyle, and Scott 2012; Zhang, Fong, and Li 2019). The findings of Nghiêm-Phú, Kiều, and Hoàng (2020) supported the direct effect of event satisfaction on life satisfaction of nature-themed festival tourists. Armbrecht and Andersson (2020) found the influences of event experience and event satisfaction on SWB of sport event participants. Yolal et al. (2016) revealed the positive effects of social benefits on SWB of local attendees of a film festival. These studies indicate the positive impacts of hosting events on tourists or residents’ SWB. The present study carries forward the line of inquiries and hypothesizes:
H5a: Event attachment positively influences visitors’ SWB.
The extant literature does not provide evidence on the mediating effect of event attachment on the relationship between CES and SWB, although a limited amount of research examin its moderating effect (Ouyang, Gursoy, and Sharma 2017). The present study attempts to investigate its mediating effect in CES and SWB context. Based on earlier discussion about the relationships among CES, event attachment and SWB, the study posits that event attachment mediates the relationship between CES and SWB. Higher value perception of CES in forest parks increases visitors’ event attachment hosted in the parks, which then enhances visitors’ SWB. The study hypothesizes that:
H5b: Event attachment mediates the relationship between perceived value of CES and SWB.
Place Attachment and Event Attachment
Previous studies found that event satisfaction could lead to attachment to the hosting place or destination (Kim, Choe, and Petrick 2018; Lee, Kyle, and Scott 2012; Nghiêm-Phú, Kiều, and Hoàng 2020 Zhang, Fong, and Li 2019). These studies provide evidence indicating the significance of hosting events to destination marketing. Yet, extant literature is scarce on the relationship between place attachment and event attachment, even though place attachment is not an unfamiliar construct in event field. One study investigated the antecedents of event attachment and its effects on place attachment and place loyalty in the Sky Lantern Festival. They found that event attachment exerted positive effects on place attachment (Tsaur et al. 2019). The present study examines their relationship and role in CES and visitors’ well-being in the setting of a nature-themed event blended into the host forest park and posits that visitors’ attachment to the host park may transfer to the event. The study hypothesizes that:
H6a: Place attachment positively influences event attachment.
Place attachment and event attachment may have a serial mediating role on the relationship between visitors’ perceived value of CES and their SWB. When visitors perceive higher value of CES, they can develop higher affective attachment to the place providing those values. This affective attachment is then transferred to the event hosted in the place, forming event attachment and further leading to higher sense of SWB. This study hypothesizes that:
H6b: Place attachment and event attachment has a serial mediating role on the relationship between perceived value of CES and SWB.
Figure 1 present the conceptual framework of the study. It summarizes the aforementioned nine hypotheses and illustrates their inter- and intra-relationships.

The conceptual model and hypotheses.
Methodology
Measures
Four main constructs are measured in this study. The construct of perceived value of CES was based on the work of Clement and Cheng (2011). Their original scale included 13 values of ecosystem service. In their study, the respondents divided a hypothetical $100 to these values. In the present study, 7 of their 13 ecosystem service values were adapted to measure perceived values of CES. They are esthetic, recreation, learning, spiritual, historical, therapeutic, and cultural. These values are measured on 5-point Likert scales with 1 = strongly disagree and 5 = strongly agree. The construct of place attachment was based on the work of Lee, Kyle, and Scott (2012). The present study adopted six items, three of which are for place identity. They are “I identify strongly with this park,” “This park means a lot to me,” and “I am very attached to this park.” The other three items are for place dependence, including “For the recreation/leisure activities that I enjoy, this park is the best,” “I prefer this park over other places for recreation/leisure activities that I enjoy,” and “When others suggest alternatives to this park for the recreation/leisure activities that I enjoy, I still choose this park.”
The construct of event attachment was adapted from Ouyang, Gursoy, and Sharma (2017) and Prayag et al. (2020). It was operationalized as unidimensional construct and measured using five items. The construct of SWB included three items adapted from Yolal et al. (2016) and Chen (2020). All items were measured using 5-point Likert scales where 1 = strongly disagree and 5 = strongly agree. Since the references from which the scales of the current study were adapted or adopted are in English, the study conducted a procedure of back-to-back translations to arrive at the scales in Chinese. They were then pretested and linguistically refined for the language suitability for Chinese respondents. Besides the four main constructs, some questions about respondents’ demographic and behavioral variables were included in the questionnaire.
Data Collection
On-site survey was conducted at Dongping National Forest Park, Chongming, Shanghai, China during the 10th Flower Expo. The Expo consists of 180 outdoor parks with 2,000 floral varieties and over 300 varieties of magaphanerophytes, as well as 64 indoor parks with near 1,000 floral varieties. The visitors expected to experience not only the vast scale of the Expo, but also learn about the diverse array of exhibits from different regions and new species cultivated by associations and companies.
During the period from May 21st to July 2nd, 2021, the Expo attracted over 2,120,000 on-site visitors and 24,080,000 online visitors. The survey was conducted onsite by trained postgraduates wearing a surgical mask twice on June 2–6th and 18–20th. Since April 2020, the Chinese tourism industry has gradually started to recover from COVID-19. According to the report of the Ministry of Culture and Tourism of China, the number of domestic tourists went up to 2,300 million during a five-day holiday on May Day in 2021, which was 103.2% of the number of tourists in the same period in 2019. Convenient sampling method was used. The potential respondents were approached after they exited the Expo throughout afternoon and during early evening and were asked if they were willing to fill the questionnaire. On average,1in 10 were unwilling to do so. The interviewers stood nearby when the respondents filled out the questionnaire. A total of 496 questionnaires were returned by the respondents. In checking the data quality, 67 of them were identified as having missing values in any responses including demographic questions. These questionnaires were therefore removed from the data. The remaining 429 questionnaires were retained as the study’s sample after no careless responses were detected. The detecting procedure was a combination of inconsistency approach and response pattern approach as described in Huang et al. (2012).
In the sample, male visitors (52%) were slightly more than females (48%). The main age groups were 19–30 years old (39.6%) and 31–50 years old (36.4%). The numbers of visitors aged less 18 (11.9%) and over 51 (12.1%) were relatively small. Most of the sample had four-year bachelor degree or diploma (56.4%), followed by high school (21.9%), two-year postgraduate (15.4%), and middle school (6.3%). The income distribution of the sample was 21.9% for less 2,000 RMB (about $313) per month, 25.9% for 2,000–6,000 RMB (about $940), 27.7% for 6,000–10,000 RMB (about $1,567), and 24.5% for over 10,000 RMB. About 65.5% respondents visited the Expo with friends or colleagues, 24.5% with family. About 79% respondents were from outside of Chongming Island, 21% were from within the region.
Data Analysis
SPSS and AMOS tools were used to analyze the data. Descriptive statistics (i.e., skewness and kurtosis) were applied for multivariate normality diagnosis; and Harmon’s single-factor exploratory factor analysis (EFA) was conducted for common method variance test. According to Kline (2011), multivariate normality could be assumed through assessing the univariate skewness of each variable as long as the z-values are less than 3.0. An EFA was then executed to identify the structure of CES, and a confirmatory factor analysis (CFA) was followed to identify the structure of place attachment. Finally, the two-stage testing procedure recommended by Anderson and Gerbing (1988) was run for structural equation modeling (SEM). In the first stage, the total measurement model was estimated using CFA. In the second stage the structural model and research hypotheses were evaluated by structural model analysis.
Results
Multivariate Normality and Common Method Variance Tests
Absolute skewness values for all items were between 0.053 and 1.468, and absolute kurtosis values for all items were between 0.117 and 2.049, indicating normal distribution was satisfied (Kline, 2011). All items were grouped for exploratory factor analysis. The analysis resulted in four factors with eigenvalues greater than 1, and 67.07% of the total variance was explained. Out of them, the first factor explained 44.53% of the total variance, which was lower than the 50% threshold, indicating the results were not biased substantially by common method variance.
The Measurement Model
Exploratory factor analysis (EFA) was employed first to check the structure of CES variables, as they were measured by different approach from its reference literature. The Kaiser-Meyer-Olkin (KMO) value was 0.879, and the significance of Bartlett’s test was lower 0.001, indicating sampling adequacy. The components analysis approach and orthogonal rotation were used to extract the common factor, revealing that perceived value of CES was an unidimensional construct. This common factor explained 57.85% of total variance. Confirmatory factor analysis (CFA) was run for place attachment with the two dimensions of place identity and place dependence. The result demonstrated an ideal fit between the model and the data, with λ2/df at 1.241 (less than 5), Goodness-of-Fit (GFI) at 0.993, Incremental Fit Index (IFI) at 0.999, Comparative Fit Index (CFI) at 0.999, the Root Mean Square Error of Approximation (RMSEA) at 0.024 (less than 0.08). All loadings were between 0.81 and 0.93, above the threshold level of 0.60. The means of three items for place identity and three items for place dependence were calculated respectively. They were used as indicators of place attachment in the total measurement model and structural model.
The total measurement model was tested using CFA. The results showed that the model fit indices were not good (λ2/df = 4.373, GFI = 0.878, IFI = 0.911, CFI = 0.910, RMSEA = 0.089). According to the modification indices, two items in event attachment with low loading values were deleted. The new model was rerun and the fit indices met the recommended values (λ2/df = 3.605, GFI = 0.912, IFI = 0.942, CFI = 0.942, RMSEA = 0.078). The convergent validity and discriminant validity for each construct was evaluated. The composite reliability (CR) values for the four constructs were between 0.85 and 0.89 (over 0.7 threshold), combined with all factor loadings were between 0.65 and 0.91 (over 0.5 threshold). The above-threshold results confirmed the convergent validity of all constructs (Hair et al. 2010) (Table 1). The discriminant validity was confirmed by comparing the correlation coefficients between the constructs with the square roots of average variance extracted (AVE). All AVE values (0.51–0.73) were over 0.50, and their square roots were greater than the relative correlation coefficients (Table 2).
Total Measurement Model Test Results.
Note: All items were measured using 5-point Likert scales where 1 = strongly disagree and 5 = strongly agree.
The Square Roots of Average Variance Extracted and Correlation Coefficients.
Note: The values (in bold) in diagonal line are the square roots of AVE.
Overall, visitors held a very high level of perceived values of CES experience in the park (mean = 4.18). The top three were esthetic, recreational, and spiritual values. Visitors also felt a high level of subjective well-being (mean = 4.17). The Expo enriched their life and made them feel happy. Comparatively, visitors developed a moderate level of attachment to the park (mean = 3.59) and the Expo (mean = 3.29). Their affective connections to the Park and the Expo were not so high.
The Structural Model and Multiple Mediating Effect Test
The structural model was evaluated before hypothesis testing. The output showed an satisfactory fit indices between the structural model and data, with λ2/df = 3.605, GFI = 0.912, IFI = 0.942, CFI = 0.942, RMSEA = 0.078 (Figure 1). Then the proposed hypotheses were assessed. Perceived value of CES had a significant effect on visitors’ SWB (β = 0.62, p < .001), which supported H1. Perceived value of CES also exerted significant effects on place attachment (β = 0.67, p < .001) and event attachment (β = 0.23, p < .001). H2 and H3 were therefore supported. When controlling for the effects of perceived value of CES and event attachment, place attachment had no significant effect on visitors’ SWB (β = −0.02, p > .05). H4a was rejected. However, when controlling for the effects of perceived value of CES and place attachment, event attachment had a significant effect on visitors’ SWB (β = 0.17, p < .05). H5a was supported. In addition, place attachment had a significant effect on event attachment (β = 0.55, p < .001) when the effect of perceived value of CES was controlled. H6 was supported. The structural model explained 51% of total variance on visitors’ SWB, indicating the model had a good explanation power for visitors’ SWB. Meanwhile, the model also explained 46% of variance of place attachment and 53% of variance of event attachment, indicating perceived value of CES was an important explanatory variable for place attachment, event attachment, and SWB.
The multiple and serial mediating effects in the model were tested by AMOS 24.0. The bootstrapping with 5,000 samples was performed, and bias-corrected confidence intervals at 95% confidence level were evaluated. The results demonstrated two mediational paths were significant. The first path was perceived value of CES → event attachment → visitors’ SWB (effect size = 0.045, p < .05), the second path was perceived value of CES → place attachment → event attachment → visitors’ SWB (effect size = 0.074, p < .05). Therefore, H5b and H6b were supported. However, the mediation path of perceived value of CES → place attachment → visitors’ SWB (effect size = −0.019, p > .05) was not significant, which indicated that place attachment did not directly mediate the relationship between perceived value of CES and SWB, but rather exerted an indirectly serial mediation effect through event attachment; thus, H4b was rejected. While controlling for the two significant mediation effects, the direct effect of perceived value of CES still had a strong effect on visitors’ SWB (β = 0.62, p < .001). This result suggested that place attachment and event attachment just had a partial mediation effect on the relationship between perceived value of CES and visitors’ SWB.
Discussions and Conclusion
Theoretical Contribution
Understanding a particular phenomenon through cross-discipline lens is more informative than a single discipline approach. The construct of CES does not draw enough attention from tourism research community, even though its importance has long been recognized in the fields of recreation, nature-based tourism, and ecotourism. This study introduces CES into nature-based tourism and event contexts. The findings confirm its significance in explaining visitors’ place attachment, event attachment, and SWB. Therefore, this study expands tourists’ SWB research by providing broader explanation perspectives from CES. Meanwhile, this study enriches the understanding of the relationship between ecosystem service and human well-being by revealing the influence mechanism of CES on visitors’ SWB in nature-based tourism and event contexts.
Previous researches showed that outdoor recreation or exposure to nature could enhance well-being (Buchecker and Degenhardt 2015; Lee et al. 2011). These studies operationalized CES as the interactions between environmental spaces and the recreational practices. The present study goes further to operationalize CES values from visitors’ perception which can better address the subjectivity of CES. The effects of exposure to nature depend on not only environmental characteristics but also individual subjective perception and experience. The examination of the dual effects enables a more comprehensive understanding on the relationship between natural ecosystem and human well-being.
The extant valuation of ES in ecological, environmental and recreational areas often uses spatial analysis approach and asks the respondents to assign a fixed money (e.g., $100) to various ecosystem service (Clement and Cheng 2011). This valuation method is based on the trade-off assumption of ecosystem service. However, the relationships among different types of CES (e.g., historic value, cultural value) could not be traded of exchanged. They are synergetic and complementary. By employing the approach of Likert scale to quantify CES, the respondents can rate the importance of each CES to them to reflect the synergetic assumption of ecosystem service.
This synergetic assumption approach also makes it possible to evaluate CES relationships with other constructs using the multivariate analyses that are well established in tourism research. The proposed model in this study builds a nomological network by integrating CES, place attachment, event attachment, and SWB from different fields. The findings show that perceived value of CES has a good nomological validity. Strong correlation coefficients between CES and the other three constructs (0.68, 0.60, and 0.71) confirm the rationality of using Likert scale to quantify CES and strong explanation power of CES on tourists’ SWB and other constructs (Hair et al. 2010; Otoo, Kim, and Choi 2021).
Place attachment is an important construct to represent people-place relationship. The limited amount of previous research has explored its influence on CES (Kaltenborn, Linnell, and Gómez-Baggethun 2020), but the evidence is lacking on the effect of CES on place attachment. This study empirically illustrates that higher perceived value of CES leads to higher place attachment. The relationship between CES and event attachment has not been explored in previous research. This study is the first attempt to examine their relationship in a nature-themed event hosted in a national forest park. The finding indicates that higher perceived value of CES could lead to higher event attachment. This supports that CES can also be introduced into event research.
The serial mediating effect analysis reveals that the direct effect of visitors’ perceived value of CES on their SWB is still the main path of influence. The serial mediating effect is relatively low, even though there exists two significant indirect paths. An interesting finding is that place attachment has no significant effect on SWB in the model. They have significant correlation when CES does not be introduced, however, when controlling the effect of CES, the effect of place attachment becomes insignificant. This suggests that the influence of CES on SWB is more direct rather than indirect through place attachment. This result is inconsistent with previous studies which argue that place attachment is an important mediating variable in explaining well-being and pro-environmental behavior (Basu, Hashimoto, and Dasgupta 2020; Cheng and Wu 2015). The finding from this study indicates that more research in different contexts is called for to further explore the mediating role of place attachment.
Managerial Implications
This study also makes several contributions for visitors’ SWB enhancement, forest park operations, and event management. There are dual management objectives of forest parks to protect and serve. On one hand, the administrative organizations have the responsibility to keep the forest ecosystem healthy; on the other hand, the forest parks should provide ecosystem service including CES to contribute to visitors’ overall well-being including their SWB. This study found that visitors’ perceived value of CES had a strong direct effect on their SWB. Therefore, the forest parks could enhance visitors’ SWB through improving their value perception from the aspect of esthetic, recreation, learning, spiritual, historic, therapeutic and cultural value. Visitors value natural scenery, sights, sounds, and smells, so the forest park should take strict natural environment protection measures to keep the naturalness. Although Dongping National Forest Park is an artificial forest park near a big city, its forest coverage has been increasing since 1959. The park should continue to keep and improve its forest coverage rate.
At the same time, the park needs to provide outdoor recreation and natural learning opportunities such as walking, camping, sightseeing, and various forms of natural interpretation. Natural interpretations should be developed both online and onsite. The park could build its own online platforms such as Wechat public account to post natural scientific knowledge. In addition to static message boards, onsite interpretations could leverage mobility technology to provide immersive experience to enhance visitors’ interest and interpretive effect. Onsite educational activities can be organized both routinely and on special schedules, as exemplified by Shanghai Botanic Garden. The garden hosted four different courses to promote the learning of eco-culture during COP-15 (2020 UN Biodiversity Conference). The learning course of “seeking creatures in limited time” proved to be effective to guide people to discover the biodiversity. The park can also develop therapeutic appeal to reduce visitors’ stress by improving such environmental characteristics as being spaciously private and sensually peaceful through landscape and facilities design. In addition, the park could better exhibit its historic and cultural value by telling an attracting story about the experiences of urban students who were dispatched to impoverished rural areas for reeducation during the Cultural Revolution (https://www.sohu.com/a/455835511_100123001). The park is on the former site of a ZhiQingDian.
Hosting nature-themed event in forest parks could improve the park’s attractiveness. In return, the event heritages also could enrich park services for a long term. This study’s findings indicate that event attachment positively influences visitors’ SWB, even when controlling the effect of CES. This suggests that the forest parks should consider as part of their routine operations the staging of events. The events, however, should be selective so they can develop visitors’ event attachment and strengthen their self-concept. It is therefore essential for the parks to have staff or have access to the professionals who understand potential visitors’ values, preferences, and motivations. The choices of event themes and activities should base on the knowledge of visitors’ profile. For those forest parks not so popular yet, hosting a successful hallmark or unique event may improve their popularity, at the same time building and improving the infrastructure and facilities. Intensive media coverage is also essential to put the host park and the event in the spotlight. In addition, the hallmark events create heritages. The host park should leverage these heritages to satisfy more visitors’ needs. Upon successfully hosting and concluding the country’s 10th Flower Expo, Dongping Forest Park has initiated a permanent venue named Flower Expo Cultural Park, which provides a good example for event heritage preservation and utilization.
Limitations
This study adopted seven items from SolVES to measure the perceived value of CES. It did not include other variables proposed by MEA and other sources, such as inspiration and social relations. By adopting an existing scale, this study may have also inherited some items that were double-barreled. Future studies should improve the scale by considering more categories of CES and exploring its structure or dimensions. The multi-dimension conceptualization approach of CES can solve the double-barreled issue well. Additionally, the study examined only the serial mediating effects of place attachment and event attachment. Albeit the significant amount of variance explained in the model, the study would have been more informative by introducing other constructs and variables, such as event satisfaction and nature connection, that may influence the relationships between CES and SWB. The 10th Flower Expo was a mega event on the national scale. Nature-themed festivals and events come in various sizes and at different locations, be they in urban, rural and natural area. Future studies on other types of nature-themed events in other settings will further advance the knowledge on the effects of cultural ecosystem services on visitors’ subjective well-being. Even the COVID-19 in China has been controlled very well, cross-provincial travels are still limited, which may influence the sample demographic characteristics and further influence the results, thus the finding should be retested in post COVID-19.
Conclusion
Since the link between ecosystem service and human well-being was proposed by MEA in 2005, this theoretical framework has guided scholars from a variety of disciplines to empirically examine their relationship and the underlying mechanism. This study expanded the framework by proposing and testing a theoretical model that integrated four constructs from different academic areas. They are cultural ecosystem service (CES) from ecology and environmental management; place attachment from geography, environmental psychology, recreation and tourism; event attachment from festival and event; and subjective well-being (SWB) from positive psychology.
In the setting of a national forest park and a flower expo in China, the study has revealed the interrelationships among the four constructs. Perceived value of CES is found to be an important and direct antecedent of visitors’ SWB. There is also an indirect effect of perceived value of CES on SWB through event attachment and place attachment as mediators. By operationalizing CES from visitors’ perspective in quantitative measures, the study addresses the subjectivity of CES and the synergy of different benefits of CES as well. Based on this approach, a nomological network model has been built. The findings confirmed the important role of CES in this network model, They present important implications for forest park operations and event management. CES warrants further inquiries in nature-based tourism and events.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research is supported by the National Natural Science Foundation of China (Grant No. 41971254).
