UNESCO Global Geoparks’ impact on China’s tourism: Scarcity,synergy and saturation

Abstract

Integrating scarcity theory and the Tourism Area Life Cycle (TALC) model, this study develops a theoretical framework to investigate how UNESCO Global Geoparks (UGGps) influence domestic and inbound tourism in China. We further analyse their nonlinear (inverted U-shaped) effects and synergistic interactions with other prestigious designations—World Heritage Sites (WHS) and China’s National 5A-level Tourist Attractions. Using panel data from 30 Chinese provinces for the period 2000–2020, we find: (1) UGGps significantly boost both domestic and international tourist arrivals; (2) the impact on total and domestic arrivals follows an inverted U-shaped curve, indicating diminishing returns beyond an optimal level; and (3) co-location with National 5A-level attractions produces strong synergies, especially for domestic tourism. These results advance tourism theory by unravelling the interplay between scarcity and destination life cycles, while providing actionable guidance for geopark sustainability and policy development.

Keywords

UNESCO Global Geoparks tourist arrivals panel data nonlinear relationship brand synergistic effects

Introduction

UNESCO Global Geoparks (UGGps) represent one of UNESCO’s three flagship initiatives for global sustainable development, serving as internationally recognized brands for geo-tourism resources (Li et al., 2021). The Global Geoparks concept originated at the 1991 Digne Convention, with UNESCO formally establishing the Geoparks Programme in 1997. This initiative reached a significant milestone in 2015 when 195 UNESCO member states officially approved the UGGps designation, marking the first new UNESCO brand since the World Heritage designation in 1972 (Han et al., 2017). Currently, 213 UGGps operate across 48 countries, each representing unified geographical areas where internationally significant geological sites are managed through an integrated approach emphasizing protection, education, and sustainable development.

Within tourism research, UGGps are considered rare tourism resource brands comparable in scope to World Heritage Sites (WHS) (Dowling, 2010). Although both programs operate under UNESCO’s umbrella, they serve distinct purposes. WHS primarily aim to identify and protect cultural and natural heritage demonstrating Outstanding Universal Value (OUV), with preservation of integrity and authenticity as core objectives (Ringbeck, 2017). In contrast, UGGps complement the World Heritage List by recognizing geological sites that may not meet strict OUV criteria but still possess significant value (Migoń, 2018). While both programs share conservation goals, UGGps place greater emphasis on balancing protection with sustainable utilization of geological heritage (Catana and Brilha, 2020).

The fundamental operational difference between these programs lies in their priorities. WHS focus primarily on protecting heritage irreplaceability and integrity, employing tourism primarily as a tool for conservation and education. UGGps, however, explicitly integrate economic development with conservation, pursuing threefold goals of protection, education, and sustainable development (Catana and Brilha, 2020). These geoparks actively promote community participation and specifically aim to stimulate local economic growth, particularly through geotourism initiatives that support poverty reduction and employment generation in remote areas (Farsani et al., 2011).

While the relationship between tourism development and regional economic growth is well-established, research has focused predominantly on WHS rather than UGGps (Lin et al., 2020a). Early studies suggested inherent conflicts between heritage conservation and tourism development (Aas et al., 2005), though recent research demonstrates that WHS designation can create mutually beneficial relationships (Yang et al., 2010). Quantitative studies have shown positive, asymmetric relationships between WHS designation and tourism revenue (Lin et al., 2019), following nonlinear patterns (Lin et al., 2020a). Despite these advances in understanding WHS impacts, the relationship between UGGps and tourism development remains understudied, particularly given UGGps’s rapid global expansion and potential as drivers of sustainable tourism.

As foundations for geotourism development (Dowling, 2010), UGGps can generate employment opportunities, stimulate economic activity, and create new income sources in rural areas (Farsani et al., 2011). These benefits help address challenges such as population decline and low resident incomes (Bonachea et al., 2023). Moreover, UGGps provide host communities with prestigious branding opportunities for their rare geological heritage, enabling strategic marketing while fostering local participation in geopark development (Bonachea et al., 2023).

However, UGGps development faces significant challenges. The 2012 delisting of Australia’s Kanawinka Global Geopark and yellow-card warnings issued to three Chinese UGGps in 2013 demonstrate ongoing tensions between conservation goals and development pressures (Lewis, 2023; Luo et al., 2021). These cases reveal critical governance risks, including overemphasis on brand accreditation at the expense of operational management, and potential conflicts between branding objectives and sustainability goals.

Existing research on UGGps’ impacts has primarily employed qualitative methods and case studies, examining micro-level aspects such as resource characteristics (Yahya Sheibani and Zamanian, 2023), promotion strategies (Cai et al., 2023), poverty alleviation effects (Deng and Zou, 2021), and visitor satisfaction (Cai et al., 2023). While valuable, these approaches limit generalisability. At the regional level, robust empirical evidence about UGGps’ effects on tourism development remains scarce.

China’s experience with UGGps, beginning in 2004, provides an ideal research context. With 41 UGGps as of 2023 - the highest number globally - China’s geoparks are predominantly located in rural, economically disadvantaged areas where they serve as important tools for sustainable development (Han et al., 2017). Notably, many Chinese UGGps overlap with other prestigious designations: 12 are also WHS, while 24 hold National 5A-level tourist attraction status (Cai et al., 2023). This unique concentration of high-status tourism brands makes China particularly suitable for examining interactions between scarce tourism resources and development outcomes.

This study addresses three critical gaps in understanding UGGps’ tourism impacts: (1) differential brand effects across domestic versus international markets; (2) potential nonlinear marginal effects on tourism demand; and (3) synergistic interactions with other premium resource brands (particularly WHS and National 5A-level attractions). Leveraging China’s comprehensive UGGp network as an empirical setting, this research provides the first systematic analysis of UGGp–tourism relationships while pioneering investigation into multi-brand synergies. The results advance theoretical understanding of geotourism branding and offer evidence-based policy recommendations for UGGp management.

This study makes several key contributions. First, it integrates Resource Scarcity Theory and the Tourism Area Life Cycle (TALC) model to develop a dynamic framework that explains tourism resource-demand relationships. This theoretical advancement builds upon prior research on scarce tourism resources and economic growth (Lin et al., 2019, 2020a, 2020b; Luo et al., 2021).

Second, panel regression models are applied to provide robust empirical evidence on the linear and nonlinear effects of UGGps, complementing existing qualitative research by offering a quantitative analysis of UGGps’ impact on China’s tourism expansion. Third, analysing the brand synergistic effects between UGGps and other international or national brands—such as WHS and National 5A-level attractions—enhances our understanding of multi-brand destination development mechanisms.

Theoretical framework and research hypotheses

Theoretical foundation

Scarcity is a fundamental concept in economics, stating that the limited availability of a commodity enhances its subjectively perceived value. This principle later gave rise to commodity theory (Brock, 1968), which holds that the value of any good depends on the degree to which it is unavailable. When people perceive something as scarce, they tend to appreciate it more (Gholipour et al., 2014). In other words, higher scarcity leads to greater attractiveness. The influence of scarcity is also evident in the supply–demand dynamics of tourism resources and products (Zeng et al., 2024).

In tourism contexts, scarcity can be classified into three types: supply-driven, demand-driven, and resource scarcity (Hosany and Hosany, 2025). Supply-driven scarcity arises from the objectively limited availability of a resource; demand-driven scarcity refers to perceived scarcity triggered by high demand, herd behavior, or brand awareness (Huang et al., 2020); and resource scarcity denotes limitations caused by time, money, or space (Cannon et al., 2019).

This study argues that UGGps, as representatives of precious geotourism resources and prominent tourism brands under the UNESCO umbrella, derive their attractiveness from both supply-driven and demand-driven scarcity. Moreover, the roles of these two types of scarcity shift dynamically across different stages of destination development, thereby exerting varying influences on tourism growth.

First, UGGps constitute a supply-driven scarce resource. When a province hosts very few UGGps, objective supply scarcity is strong. As an informational frame, supply-driven scarcity may lead consumers to associate a product’s exclusivity with high quality when making purchase decisions (Aguirre et al., 2015; Huang et al., 2020). Consumers who are able to own or consume limited-supply products tend to perceive themselves as enjoying higher social status—a phenomenon known as the snob effect (Amaldoss and Jain, 2005).

Furthermore, compared with ordinary tourism resources, scarce tourism resource brands typically signify unique natural, cultural, or historical value and offer distinctive experiences that are difficult to replicate elsewhere (Wang and Yuan, 2020). Scarce tourism resources can therefore significantly stimulate demand by satisfying the need for exploration, fostering a sense of unique value, and enhancing market awareness (Cuccia et al., 2016). As the number of UGGps within a province grows, destinations can form “geopark clusters,” triggering demand-driven scarcity. At this stage, rising brand awareness, greater social media exposure, and intensified tourist herd behaviour (i.e., Cialdini’s principle of “social proof”) combine to enhance the overall appeal of UGGps (Park et al., 2022).

However, when the number of UGGps in a destination exceeds a certain threshold, two things happen. On the one hand, the supply scarcity of UGGps declines. On the other hand, UGGps with limited spatial capacity attract excessive numbers of tourists, exposing visitors to another form of scarcity that negatively affects destination attractiveness—spatial scarcity, or tourism crowding—which reduces both experience quality and destination appeal (Jacobsen et al., 2019). In summary, the dynamic evolution of tourism resource scarcity is closely intertwined with the life-cycle pattern of tourism destinations (Haywood, 1986; Marsiglio and Tolotti, 2024).

Stage-specific features of UGGps’ impact on tourism

According to the Tourism Area Life Cycle (TALC) theory, a tourist destination, much like an industrial product, progresses from market entry through maturity and eventually toward decline, with the number of tourists replacing product sales volume as the key indicator (Butler, 2008). The evolution of the destination life cycle is shaped by a combination of socio-economic forces, including resource endowment, changes in tourist demand, brand image building, political support or opposition, and the development of new destinations (Agarwal, 2002; Haywood, 1986; Hovinen, 2002). The perceived scarcity of tourism resources likewise changes across different TALC stages under the influence of these factors (Marsiglio and Tolotti, 2024; Zhong et al., 2008).

Therefore, we integrate scarcity theory with TALC theory and, building on the three-stage (Stansfield, 1978) and six-stage (Butler, 2008) TALC models, propose a four-stage theoretical framework to analyse the dynamic relationship between the number of scarce tourism resources (represented by UGGps) and tourism demand.

In the Demand emergence stage, a region possesses very few scarce resources such as UGGps or WHS, so supply scarcity is strong. Destination awareness and reputation have not yet been established, and destination attractiveness relies primarily on the perceived value generated by supply scarcity (Zeng et al., 2024). Meanwhile, following the distance-decay law of tourism, destinations at this stage are more attractive to nearby source markets but have limited appeal to more distant markets, especially international ones (Lee et al., 2012; McKercher, 2021; Nicolau and Mas, 2006). UGGps at this stage also attract some genuine geo-tourists and adventurous backpackers; overall, however, these groups remain relatively small. Tourist arrivals show an increasing trend, but the total volume remains modest.

Subsequently, in the Demand growth stage, due to the lag in information dissemination from the previous stage, the influence of scarce tourism resource brands on more distant international markets only becomes fully evident (Lin et al., 2020a). Considering that we live in the Internet era, the time between the transmission of marketing information and its reception by consumers has shortened considerably compared to the past (Rosado González et al., 2020). The destination enters a new phase in which domestic and international demand grow simultaneously. Tourist arrivals increase notably but have not yet entered a phase of truly rapid expansion.

In the Demand expansion stage, as the concentration of scarce resources increases, brand effects and scale effects begin to emerge (Gnoth, 2002; Huang et al., 2020). Brand effects transmit perceived scarcity to global tourists, enhancing the destination’s international appeal. Simultaneously, destination managers can achieve scale effects through joint marketing and promotion of multiple scarce resource brands (Lee & Jayakumar, 2021; Pásková and Zelenka, 2018). Destination awareness rises accordingly, and tourism demand expands rapidly. Thus, the destination enters a new phase dominated by demand-driven scarcity.

Finally, in the Demand saturation stage, the massive influx of tourists begins to saturate environmental carrying capacity. Managers may introduce measures to control the growth of tourist arrivals for purposes such as resource protection and sustainable utilisation (Chen et al., 2022; Kastenholz, 2004). At the same time, when tourist numbers reach a certain threshold, visitors’ psychological carrying capacity also becomes saturated, and the tourism experience starts to decline (Zhong et al., 2008). The marginal utility brought by an increasing number of UGGps within the same region also begins to diminish at this stage, and tourism demand gradually stabilises and saturates. If this situation persists, as resource scarcity weakens and substitution among similar tourism resource brands takes place, tourists will start to seek out new destinations and alternative tourism products, and total demand will enter a phase of overall saturation (Hassan, 2000). Figure 1 presents the theoretical framework of this study. Based on this framework, we propose Hypotheses 1 and 2.

Figure 1.

Dynamic relationship between UGGps and tourism demand.

Hypothesis 1

The number of UGGps exhibits a quadratic (inverted U-shaped) effect on tourist arrivals.

Hypothesis 2

The number of UGGps exerts varying effects on tourist arrivals across different development stages.

Differences in the impact of UGGps on domestic and international tourism

UGGps are established on the basis of geological heritage resources that represent 4.6 billion years of Earth’s evolution (Migoń, 2018). These resources possess significant scientific and educational value, attracting both specialised geo-tourists and general tourists drawn by their scarcity (Amaro et al., 2023). Moreover, the prestigious UNESCO designation enhances destination awareness and perceived value. However, the influence of UGGps on tourism demand is moderated by the distance-decay effect (Lee et al., 2012; McKercher, 2021; Nicolau and Mas, 2006), whereby closer markets experience lower travel costs, fewer cultural barriers (Okafor et al., 2018), and stronger destination image perceptions. Consequently, we expect UGGps to demonstrate more immediate effects on domestic tourism markets.

While research indicates that UGGps’ impact on international markets exhibits a temporal lag (Lin et al., 2020a), their UNESCO designation functions as a powerful international branding tool, comparable to that of WHS. This official recognition enhances global visibility and perceived scarcity (Adie et al., 2018). Additionally, clustering multiple UGGps within a region enables cost-sharing through joint promotion and collaborative governance, creating economies of scale that boost market competitiveness (Lee & Jayakumar, 2021; Pásková and Zelenka, 2018). These combined brand and scale effects should positively influence inbound tourism. Thus, we propose:

Hypothesis 3

The number of UGGps demonstrates differential effects on domestic and international tourist arrivals.

Synergistic effects of UGGps with other brands

China’s UGGps frequently overlap with other prestigious tourism brands, including WHS and National 5A-level attractions (Cai et al., 2023). As of 2020, nine UGGps in China were also designated as WHS, while 22 held 5A-level scenic spot status. Branding serves as a critical tool in protected area management, particularly for conserving globally scarce resources (Luo et al., 2021). Moreover, brand awareness significantly influences tourist decision-making, and destinations possessing multiple high-profile brands benefit from enhanced recognition and tourism growth (Wang and Yuan, 2020).

UNESCO explicitly emphasises that UGGps should foster connections between geological heritage and other natural, cultural, and intangible assets. The co-branding of UGGps with National 5A-level attractions or WHS can create a reinforcement mechanism that amplifies a destination’s overall appeal (Luo et al., 2021). While some researchers argue that mature tourism destinations with established market recognition may experience diminishing returns from additional branding (Wang and Yuan, 2020), UGGps—as a relatively recent designation—can synergistically enhance tourism attractiveness and demand when combined with existing high-status brands. This leads to our fourth hypothesis:

Hypothesis 4

UGGps have a positive effect on tourist arrivals when they overlap with WHS or National 5A-level attractions.

Data and variables

This study examines 30 provincial-level administrative divisions in mainland China that had established at least one UNESCO Global Geopark (UGGp) by 2020. The sample was selected on the basis of UGGp presence, with Tibet excluded due to incomplete data availability. As shown in Figure 2, these provinces represent the geographic distribution of UGGps across China. Annual data were collected from provincial statistical yearbooks for the period 2000 to 2020, yielding a balanced panel dataset of 630 observations (30 provincial-level units × 21 years).

Figure 2.

Distribution of UGGps in China in 2020.

Dependent and independent variables

The key independent variable is the number of UGGps in each province, which serves as a proxy for the UGGp brand and reflects the strength of scarce tourism resources in shaping tourism attractiveness. These data were obtained from the official website of the UNESCO Global Geoparks Network.

To comprehensively assess tourism development, we employ two standard dependent variables widely used in tourism economics research: inbound tourist arrivals (ITA) for international tourism and domestic tourist arrivals (DTA) for domestic tourism (Demir et al., 2020; Kim et al., 2006; Wang and Chen, 2021; Yang et al., 2010).

Control variables

To ensure that the model accurately captures the actual impact of the independent variables on the dependent variables, we include the following control variables based on a review of the existing literature.

Foreign Direct Investment (FDI) is measured as the ratio of FDI to GDP. Business tourists often constitute a significant component of inbound tourist arrivals and are closely related to FDI (Davidson and Sahli, 2015). At the same time, FDI can indirectly promote domestic tourist arrivals by facilitating regional infrastructure construction (Barrowclough, 2007). The number of star-rated hotels (HOTEL) serves as a proxy for regional tourism reception capacity, controlling for the impact of accommodation conditions on tourist arrivals. Better accommodation facilities can enhance tourist comfort and length of stay, thereby potentially increasing both domestic and inbound tourist numbers (Shi et al., 2016).

Urbanization rate (UR) measures the impact of urbanization on tourism development. Higher urbanization levels are generally associated with better transportation, accommodation, catering, and other tourism-related infrastructure (Luo et al., 2015). Car ownership per capita (CAR) reflects residents’ income levels and the prevalence of private vehicles (Yang et al., 2017). Regions with higher economic levels tend to offer better facilities for self-drive tourism, which may attract more domestic tourists (Lin et al., 2020a). Road network density (HRD) acts as a proxy for transportation convenience, controlling for the impact of transport infrastructure on tourist arrivals. Greater road density can reduce travel costs and increase the probability of tourist visits (Lin et al., 2020a).

Variable statistical analysis

Table 1 reports descriptive statistics and multicollinearity diagnostics for all variables in our models. Notably, total tourist arrivals (lnTTA) and domestic tourist arrivals (lnDTA) display extremely high variance inflation factors (VIF) of 4509.21 and 4364.38, respectively, with tolerance values approaching zero. This is expected, as domestic tourism accounts for the vast majority of total tourist arrivals in China, resulting in near-perfect collinearity between lnTTA and lnDTA. Apart from this structural overlap, the VIF values for all other variables remain well below 10, and tolerance statistics exceed 0.1, indicating no concerning multicollinearity issues in our dataset.

Table 1.

Descriptive statistics of variables.

Variables	Observations	Mean	S.D.	Minimum	Maximum	VIF	Tolerance
lnTTA	630	9.321	1.275	5.550	11.640	4509.21	0.000
lnDTA	630	9.303	1.278	5.443	11.638	4364.38	0. 000
lnITA	630	4.672	1.576	−1.204	8.229	4.25	0.235
UGGps	630	0.703	0.984	0	4	1.33	0.751
FDI	630	0.024	0.021	0	0.122	1.40	0.714
lnHOTEL	630	5.666	0.691	2.708	7.069	2.08	0.480
UR	630	1.925	3.603	0.050	26.13	3.65	0.274
CAR	630	2.689	6.191	0.023	56.185	3.86	0.259
ROAD	630	0.743	0.488	0.020	2.190	2.37	0.422

Table 2 presents a comparative analysis of provinces grouped by the presence of UNESCO Global Geoparks. The data reveal a clear disparity: provinces with at least one UGGp register substantially higher average tourist arrivals—across total, domestic, and inbound metrics—compared to provinces without any UGGps. Moreover, these UGGp-hosting provinces exhibit significantly more developed tourism infrastructure, as reflected in higher numbers of star-rated hotels and greater road network density. This infrastructure advantage likely underpins the higher visitor volumes observed.

Table 2.

Summary of data.

Variables	(1)	(2)	(3)
Variables	Provinces with at least one UGGps	Provinces without UGGps	Whole sample
lnTTA	10.039 (0.880)	8.792 (1.263)	9.321 (1.275)
lnDTA	10.020 (0.886)	8.776 (1.267)	9.303 (1.278)
lnITA	5.339 (1.362)	4.182 (1.545)	4.672 (1.576)
FDI	0.022 (0.013)	0.026 (0.025)	0.024 (0.021)
lnHOTEL	5.982 (0.503)	5.434 (0.718)	5.666 (0.691)
UR	2.118 (4.519)	1.783 (2.739)	1.925 (3.603)
CAR	3.847 (8.903)	1.837 (2.574)	2.689 (6.191)
ROAD	0.855 (0.456)	0.662 (0.496)	0.743 (0.488)
Observation	267	363	420

Notes: Standard errors are in parentheses.

Table 3 presents the correlation coefficients between key variables. The number of UGGps shows moderate positive correlations with total (r = 0.458), inbound (r = 0.319), and domestic (r = 0.456) tourist arrivals, indicating a stronger relationship with domestic tourism than with the inbound market. Star-rated hotel counts demonstrate the strongest overall correlations with tourist arrivals across all categories, while road network density is particularly strongly associated with domestic tourism, highlighting the importance of transportation infrastructure for domestic visitor flows.

Table 3.

Correlation analysis of variables.

Variables	lnTTA	lnDTA	lnITA	UGGPS	FDI	lnHOTEL	UR	CAR
lnTTA
lnDTA	1.000***
lnITA	0.684***	0.674***
UGGPS	0.458***	0.456***	0.319***
FDI	0.086**	0.080**	0.334***	−0.065
lnHOTEL	0.574***	0.567***	0.670***	0.284***	0.089**
UR	0.098**	0.098**	0.093**	−0.038	−0.077*	0.051
CAR	0.257***	0.258***	0.120***	0.057	−0.079**	0.074*	0.834***
ROAD	0.690***	0.687***	0.557***	0.248***	0.345***	0.330***	−0.005	0.113***

***p < 0.01, **p < 0.05, *p < 0.1.

Methodology

This study employs panel regression analysis to investigate the relationship between UGGps and tourism development across Chinese provinces. Panel data methods are particularly suitable for this research because they offer several advantages over conventional cross-sectional or time-series approaches. First, by combining both cross-sectional and temporal dimensions, panel data provides a larger number of observations, which improves the efficiency and accuracy of parameter estimates (Cortes-Jimenez and Pulina, 2010). Second, panel methods allow us to account for both time-varying effects and persistent differences across provinces, enabling a more nuanced analysis of dynamic relationships (Vuković et al., 2023). Third, these techniques help control for unobserved heterogeneity through fixed or random effects specifications (Hausman and Taylor, 1981), reducing potential omitted variable bias.

To address potential endogeneity concerns arising from simultaneity or measurement error, we implement the System Generalized Method of Moments (SYS-GMM) estimator. This approach uses lagged values of the dependent variables as instruments, which helps mitigate reverse causality problems while maintaining estimation efficiency. Diagnostic tests, including the Hansen test for instrument validity and tests for autocorrelation, confirm the appropriateness of our instrumental variable strategy.

Linear relationship model

We begin by estimating linear specifications to test Hypotheses 1–4 regarding the basic relationship between UGGps and tourism flows. The baseline empirical model takes the following form:

{l n T T A}_{i t} = α_{0} + α_{1} {U G G p s}_{i t} {+ α}_{2} {l n F D I}_{i t} + α_{3} {l n H O T E L}_{i t} + α_{4} {U R}_{i t} + α_{5} {C A R}_{i t} + α_{6} {R O A D}_{i t} + \sum_{k} γ κ {Y e a r}_{k t} + μ_{i} + ε_{i t},

(1)

where lnTTA_it represent the logarithmic form of tourism development, α₀ is a constant term, UGGps_it represents the number of UGGps, lnFDI_it, lnHOTEL_it, UR_it, CAR_it, and HRD_it represent the logarithmic forms of foreign direct investment, number of star-rated hotels, urbanization rate, car ownership per capita, and road network density, respectively. μ_i is the individual effect,

ε

_it is the random error term, both normally distributed. α represents the impacts of the control variables. Subscripts i and t denote province and time, respectively. If the coefficients are statistically significant and positive, UGGps have a positive impact on tourism development.

Nonlinear relationship model

For Hypothesis 1, which posits diminishing returns to additional geopark designations, we augment the linear specification with a quadratic term for UGGp count. This allows us to test for potential nonlinearities in the relationship between geoparks and tourism demand.

{l n T T A}_{i t} = α_{0} + α_{1} {U G G p s}_{i t} + {α_{2} U G G p s}_{i t}^{2} {+ α}_{3} {l n F D I}_{i t} + α_{4} {l n H O T E L}_{i t} + α_{5} {U R}_{i t} + α_{6} {C A R}_{i t} + α_{7} {R O A D}_{i t} + \sum_{k} γ κ {Y e a r}_{k t} + μ_{i} + ε_{i t},

(2)

According to the regression results, if α₂ is significantly different from 0, a non-linear relationship exists between UGGps and tourism development. If α₂ is significantly negative, an inverted U-shaped relationship exists between UGGps and tourism development.The turning point of this relationship can be calculated from the estimated coefficients to identify the optimal number of geoparks beyond which additional designations provide increasing or decreasing benefits.

To test Hypothesis 2, we adopt a dummy variable grouping approach. The number of UGGps is transformed into four binary variables (UGGps = 1, UGGps = 2, UGGps = 3, UGGps = 4), with provinces having zero UGGps serving as the reference group. The following fixed-effects model is specified:

{l n T T A}_{i t} = α_{0} + \sum_{j = 1}^{4} α_{1} D_{i t}^{j} {+ α}_{2} {l n F D I}_{i t} + α_{3} {l n H O T E L}_{i t} + α_{4} {U R}_{i t} + α_{5} {C A R}_{i t} + α_{6} {R O A D}_{i t} + \sum_{k} γ κ {Y e a r}_{k t} + μ_{i} + ε_{i t} .

(3)

If the coefficients $α_{1}$ are significantly positive and successively larger, the effect of UGGps on tourism demand follows a non-uniform, first increasing then plateauing pattern. This finding would support Hypothesis 2 and suggest that the marginal benefit of additional UGGps begins to diminish after a certain number of geoparks.

Brand synergistic effect model

To test Hypothesis 4 regarding potential complementarities between different tourism designations, we extend the baseline model by incorporating interaction terms between UGGp and other prestigious labels. Specifically, interaction terms between UGGps and dummy variables for National Category 5A Scenic Areas (5A_D) and World Heritage Sites (WHS_D) are introduced to examine the superposition effects of brands on inbound and domestic tourist arrivals. The models are specified as follows:

{l n T T A}_{i t} = α_{0} + α_{1} {U G G p s}_{i t} + {α_{2} ({U G G p s}_{i t} * {5 A_{D}}_{i t}) + α}_{3} {l n F D I}_{i t} + α_{4} {l n H O T E L}_{i t} + α_{5} {U R}_{i t} + α_{6} {C A R}_{i t} + α_{7} {R O A D}_{i t} + \sum_{k} γ κ {Y e a r}_{k t} + μ_{i} + ε_{i t},

(4)

and

{l n T T A}_{i t} = β_{0} + β_{1} {U G G p s}_{i t} + {β_{2} ({U G G p s}_{i t} * {W H S_D}_{i t}) + β}_{3} {l n F D I}_{i t} β_{4} {l n H O T E L}_{i t} + β_{5} {U R}_{i t} + β_{6} {C A R}_{i t} + β_{7} {R O A D}_{i t} + \sum_{k} γ κ {Y e a r}_{k t} + μ_{i} + ε_{i t},

(5)

where 5A_D and WHS_D are dummy variables: the value is 1 if province i has at least one Category 5A Scenic Area or World Heritage Site overlapping with a UGGps in year t; otherwise, it is 0. If the interaction term coefficients α₂ (Model 4) and β₂ (Model 5) are significantly positive, this indicates a synergistic reinforcement effect of brand superposition on tourism attractiveness. If they are insignificant or negative, no reinforcing effect is present. All control variables are defined as before, μ_i is the individual effect, and

ε_{i t}

is the random error term.

Empirical results

Relationship between UGGps and tourist arrivals

The regression results presented in Tables 4 –6 comprehensively demonstrate the relationship between UGGps and tourist arrivals, testing Hypotheses 1–3. First, to test Hypothesis 1, we introduced a quadratic term into the models. As shown in Table 4 for total tourist arrivals, the linear term is significantly positive (β = 0.7071, p < 0.01) while the quadratic term is negative (β = −0.1249, p < 0.1). Table 6 reveals an identical pattern for domestic tourist arrivals (lnDTA), with a significant negative quadratic coefficient (−0.1254, p < 0.1). However, a striking divergence appears in the inbound tourism model: while the linear term remains positive (β = 0.2970, p < 0.05), the quadratic term is completely insignificant (β = −0.0388, p > 0.1). This partially confirms Hypothesis 1, indicating that the inverted U-shaped diminishing returns exist robustly for the domestic market, whereas the inbound market responds in a more continuous, linear fashion without hitting a capacity threshold within the observed data range. This nonlinear pattern is further visualized in Figure 3, which graphically illustrate the inverted U-shaped relationship between the number of geoparks and tourism arrivals.

Table 4.

Relation between UGGps and total tourist arrivals.

Variables	(1)	(2)	(3)	(4)	(5)
Variables	OLS	RE	FE	FE	FE
UGGps	0.2411***	0.3554***	0.3675***	0.7071***
UGGps	(0.0303)	(0.1059)	(0.1104)	(0.1726)
UGGps ²				−0.1249*
UGGps ²				(0.0667)
UGGps = 1					0.5095***
UGGps = 1					(0.1533)
UGGps = 2					0.8917***
UGGps = 2					(0.1851)
UGGps = 3					1.4187***
UGGps = 3					(0.2254)
UGGps = 4					0.0818
UGGps = 4					(0.2956)
FDI	−5.6992***	−4.8959*	−4.2492	−3.1916	−2.7691
FDI	(1.4022)	(2.5197)	(2.7996)	(2.6649)	(2.7146)
lnHOTEL	0.6693***	0.4510***	0.3983**	0.3272*	0.3376*
lnHOTEL	(0.0437)	(0.1568)	(0.1829)	(0.1826)	(0.1726)
UR	−0.0394***	−0.0508	−0.0789*	−0.0838*	−0.0796*
UR	(0.0142)	(0.0309)	(0.0413)	(0.0420)	(0.0400)
CAR	0.0479***	0.0443**	0.0447**	0.0400*	0.0395**
CAR	(0.0084)	(0.0207)	(0.0206)	(0.0201)	(0.0188)
ROAD	1.3473***	1.6052***	1.6501***	1.6796***	1.7416***
ROAD	(0.0667)	(0.2082)	(0.2444)	(0.2459)	(0.2516)
year_2003	−0.4493***	−0.3091***	−0.2933***	−0.2409***	−0.2444***
year_2003	(0.1313)	(0.0614)	(0.0619)	(0.0597)	(0.0565)
year_2008	−0.5304***	−0.4720***	−0.4549***	−0.4432***	−0.4293***
year_2008	(0.1293)	(0.0536)	(0.0608)	(0.0560)	(0.0552)
year_2009	−0.4541***	−0.4058***	−0.3904***	−0.3824***	−0.3692***
year_2009	(0.1291)	(0.0489)	(0.0569)	(0.0537)	(0.0505)
year_2020	0.206	0.0397	0.0135	0.0131	−0.0347
year_2020	(0.1337)	(0.0889)	(0.0908)	(0.0801)	(0.0860)
Constant	4.5015***	5.4742***	5.7672***	6.0850***	5.9896***
Constant	(0.2353)	(0.8708)	(0.9780)	(0.9832)	(0.9208)
N	630	630	630	630	630
R-sq	0.7183		0.7234	0.7333	0.7427

Notes: Standard errors are in parentheses. *p < 0.1, **p < 0.05, ***p < 0.01.

Table 5.

Statistics of number of UGGps.

No. of UGGps	Freq.	Percent	Cum.
0	363	57.62%	57.62%
1	139	22.06%	79.68%
2	95	15.08%	96.74%
3	18	2.86%	97.62%
4	15	2.38%	100.00%

Table 6.

Difference between domestic and international markets.

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	Linear regression		Non-linear regression
	lnDTA	lnITA	lnDTA		lnITA
UGGps	0.3710***	0.1914**	0.7120***		0.2970**
UGGps	(0.1112)	(0.0742)	(0.1730)		(0.1405)
UGGps ²			−0.1254*		−0.0388
UGGps ²			(0.0671)		(0.0480)
UGGps = 1				0.5108***		0.1623
UGGps = 1				(0.1539)		(0.1111)
UGGps = 2				0.9009***		0.4519***
UGGps = 2				(0.1855)		(0.1590)
UGGps = 3				1.4284***		0.7869***
UGGps = 3				(0.2288)		(0.1733)
UGGps = 4				0.0861		0.0007
UGGps = 4				(0.2976)		(0.1854)
FDI	−4.5585	3.3525	−3.4966	−3.0710	3.6813	3.9863
FDI	(2.8219)	(2.4026)	(2.6821)	(2.7291)	(2.4064)	(2.4831)
lnHOTEL	0.3980**	0.1751	0.3266*	0.3379*	0.153	0.1756
lnHOTEL	(0.1838)	(0.1493)	(0.1835)	(0.1734)	(0.1568)	(0.1534)
UR	−0.0794*	0.0026	−0.0843*	−0.0800*	0.001	0.006
UR	(0.0416)	(0.0154)	(0.0423)	(0.0403)	(0.0157)	(0.0154)
CAR	0.0451**	0.0084	0.0404*	0.0399**	0.007	0.007
CAR	(0.0208)	(0.0074)	(0.0203)	(0.0190)	(0.0072)	(0.0062)
ROAD	1.6532***	1.4341***	1.6828***	1.7454***	1.4433***	1.4919***
ROAD	(0.2464)	(0.1417)	(0.2481)	(0.2537)	(0.1401)	(0.1470)
year_2003	−0.2895***	−0.5487***	−0.2369***	−0.2407***	−0.5324***	−0.5386***
year_2003	(0.0620)	(0.0716)	(0.0599)	(0.0566)	(0.0711)	(0.0692)
year_2008	−0.4571***	−0.2327***	−0.4454***	−0.4316***	−0.2291***	−0.2227***
year_2008	(0.0609)	(0.0655)	(0.0561)	(0.0553)	(0.0655)	(0.0647)
year_2009	−0.3910***	−0.2236***	−0.3830***	−0.3700***	−0.2212***	−0.2167***
year_2009	(0.0572)	(0.0510)	(0.0539)	(0.0508)	(0.0517)	(0.0508)
year_2020	0.0225	−2.2170***	0.022	−0.0261	−2.2172***	−2.2510***
year_2020	(0.0910)	(0.1898)	(0.0803)	(0.0862)	(0.1894)	(0.1914)
Constant	5.7533***	2.5235***	6.0724***	5.9716***	2.6223***	2.4678***
Constant	(0.9837)	(0.8195)	(0.9884)	(0.9256)	(0.8558)	(0.8386)
N	630	630	630	630	630	630
R-sq	0.7217	0.7153	0.7316	0.741	0.7166	0.7241

Notes: Standard errors are in parentheses. *p < 0.1, **p < 0.05, ***p < 0.01.

Figure 3.

Total visitor arrivals.

Second, to examine Hypothesis 2 (effects across different development stages), we employed stepwise dummy variables (Models 5 in Table 4 and Table 6). The results for total and domestic tourism show a clear non-linear scaling pattern. Provinces with one UGGp experience an increase of approximately 66% in domestic arrivals compared to those without. The marginal effect escalates significantly for provinces with two (β = 0.9009, p < 0.01) and three UGGps (β = 1.4284, p < 0.01). Crucially, when the count reaches four UGGps, the coefficient drops sharply to 0.0861 and loses all statistical significance. This strongly validates Hypothesis 2, pinpointing the third geopark as the inflection point where marginal promotional benefits drastically diminish. In contrast, the stepwise coefficients for inbound tourism are weaker in magnitude and lack the clear progressive escalation seen in the domestic market.

Third, Hypothesis 3 posits different effects between domestic and international arrivals. The data overwhelmingly support this. Comparing the baseline linear models, the coefficient for domestic tourists (β = 0.3710, p < 0.01) is nearly double that for international tourists (β = 0.1914, p < 0.05). This disparity is further magnified when observing the nonlinear stage models: the domestic market’s peak response at the third geopark (β = 1.4284) is substantially higher than the inbound market’s peak (β = 0.7869). This verifies Hypothesis 3, demonstrating that UGGps exert differential effects on domestic and international arrivals.

The control variables in Table 6 further reveal the distinct mechanisms driving these two markets. Accommodation infrastructure (lnHOTEL) significantly boosts domestic arrivals (β = 0.3266, p < 0.1) but has no significant effect on inbound arrivals, suggesting that domestic tourists rely heavily on local service quality. Conversely, Foreign Direct Investment (FDI) shows a significant positive association with inbound tourism (β = 3.6813, p < 0.01) but a negative relationship with domestic tourism. Road network density (ROAD) remains a universally critical driver for both segments. Furthermore, the urbanisation rate (UR) negatively impacts domestic tourism (β = −0.0843, p < 0.1) while having no effect on inbound tourism. Finally, year dummies accurately capture exogenous shocks: the 2003 SARS outbreak and the 2008 Financial Crisis suppressed both markets, while the 2020 pandemic caused a catastrophic drop in inbound tourists (β = −2.2172, p < 0.01) without a significant impact on domestic flows.

Brand synergistic effects between UGGp and WHS/national 5A-level attractions

The analysis of brand interaction effects, presented in Table 7, reveals heterogeneous synergies between UGGps and other prestigious tourism designations. The interaction term between UGGps and National 5A-level attractions yields consistently positive and statistically significant coefficients across all model specifications. Specifically, the marginal contribution of this co-branding strategy is most pronounced for domestic tourist arrivals (β = 0.0216, p < 0.01), followed by inbound arrivals (β = 0.0157, p < 0.01) and total arrivals (β = 0.0211, p < 0.01). This finding partially supports Hypothesis 4, suggesting that tourists—particularly those from the domestic market—assign additive value to destinations that successfully integrate geological heritage with the comprehensive service standards typically associated with 5A-rated attractions.

Table 7.

Regression analysis considering 5A tourist attraction and WHS.

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	lnTTA	lnDTA	lnITA	lnTTA	lnDTA	lnITA
	5A			WHS
UGGps	0.2111*	0.2109*	0.0754	0.2906*	0.2933*	0.1408
UGGps	(0.1099)	(0.1105)	(0.0726)	(0.1657)	(0.1673)	(0.1056)
UGGps × 5A	0.0211***	0.0216***	0.0157***
UGGps × 5A	(0.0053)	(0.0053)	(0.0039)
UGGps × WHS				0.0262	0.0264	0.0172
UGGps × WHS				(0.0364)	(0.0366)	(0.0298)
FDI	−4.7041	−5.0241*	3.0151	−4.5586	−4.8710*	3.149
FDI	(2.9155)	(2.9318)	(2.3503)	(2.8359)	(2.8634)	(2.4082)
lnHOTEL	0.5095***	0.5118***	0.2576	0.4314**	0.4314**	0.1969
lnHOTEL	(0.1793)	(0.1797)	(0.1598)	(0.1798)	(0.1806)	(0.1530)
UR	−0.0716*	−0.0719*	0.008	−0.0784*	−0.0789*	0.0029
UR	(0.0388)	(0.0391)	(0.0137)	(0.0413)	(0.0416)	(0.0153)
CAR	0.0426**	0.0430**	0.0069	0.0458**	0.0462**	0.0091
CAR	(0.0194)	(0.0195)	(0.0065)	(0.0205)	(0.0206)	(0.0076)
ROAD	1.5570***	1.5579***	1.3650***	1.6200***	1.6227***	1.4143***
ROAD	(0.2496)	(0.2513)	(0.1424)	(0.2489)	(0.2508)	(0.1495)
year_2003	−0.3325***	−0.3296***	−0.5778***	−0.3048***	−0.3011***	−0.5563***
year_2003	(0.0579)	(0.0578)	(0.0711)	(0.0619)	(0.0621)	(0.0736)
year_2008	−0.4311***	−0.4327***	−0.2150***	−0.4510***	−0.4532***	−0.2302***
year_2008	(0.0576)	(0.0576)	(0.0614)	(0.0600)	(0.0601)	(0.0653)
year_2009	−0.3568***	−0.3567***	−0.1987***	−0.3825***	−0.3830***	−0.2184***
year_2009	(0.0567)	(0.0569)	(0.0479)	(0.0587)	(0.0589)	(0.0520)
year_2020	−0.0131	−0.0048	−2.2368***	0.0094	0.0183	−2.2198***
year_2020	(0.1078)	(0.1086)	(0.1940)	(0.0905)	(0.0907)	(0.1896)
Constant	5.2260***	5.1995***	2.1222**	5.6127***	5.5974***	2.4220***
Constant	(0.9382)	(0.9415)	(0.8692)	(0.9519)	(0.9566)	(0.8329)
N	630	630	630	630	630	630
R-sq	0.7332	0.7319	0.723	0.7246	0.723	0.716

Notes: Standard errors are in parentheses. *p < 0.1, **p < 0.05, ***p < 0.01.

However, the interaction between UGGps and WHS fails to achieve statistical significance in any model specification for total, domestic, or inbound tourism. Furthermore, even when controlling for these interaction effects, the standalone coefficient of UGGps remains positive and statistically significant for total and domestic arrivals, confirming the independent baseline value of the geopark designation in driving regional tourism development.

Endogenous and robustness test

To address potential endogeneity concerns arising from the correlation between explanatory variables and the error term, this study employs the System Generalized Method of Moments (SYS-GMM) estimator to examine the effect of UGGps on tourism flows. The SYS-GMM approach offers a robust solution by exploiting lagged levels and differences of the dependent variable as internal instruments, thereby improving the reliability of coefficient estimates.

As presented in Table 8, the number of UGGps exhibits a statistically significant and positive association with total, domestic, and inbound tourist arrivals. These results remain consistent across specifications, indicating that the influence of UGGps persists even when endogeneity bias is controlled for. In particular, the use of lagged variables as instruments in the SYS-GMM framework effectively mitigates potential reverse causality and omitted variable bias, reinforcing the causal interpretation of the findings. The models successfully pass both the AR (2) test, indicating no second-order autocorrelation, and the Hansen test (p = 1.000), confirming the validity of the instrument selection.

Table 8.

Endogenous and robustness test.

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	lnTTA	lnDTA	lnITA	lnTTA	lnDTA	lnITA
	SYS-GMM			FE
L.lnTTA	0.9710***
L.lnTTA	(0.0164)
L.lnDTA		0.9703***
L.lnDTA		(0.0167)
L.lnITA			0.7404***
L.lnITA			(0.0514)
UGGps	0.0440**	0.0439**	0.2046***
UGGps	(0.0215)	(0.0207)	(0.0672)
UGGps_d				0.6351***	0.6389***	0.2577**
UGGps_d				(0.1404)	(0.1411)	(0.1150)
FDI	−1.624	−1.6314	8.0268**	−3.0061	−3.3056	3.9343
FDI	(1.6258)	(1.6019)	(3.2749)	(2.6302)	(2.6581)	(2.6174)
lnHOTEL	0.0644**	0.0615*	0.2255***	0.3017	0.3011	0.1459
lnHOTEL	(0.0303)	(0.0303)	(0.0542)	(0.1801)	(0.1811)	(0.1564)
UR	0.0007	0.0007	0.0307***	−0.0883*	−0.0888*	−0.0003
UR	(0.0019)	(0.0019)	(0.0077)	(0.0452)	(0.0456)	(0.0174)
CAR	−0.0004	−0.0004	−0.0130***	0.0389*	0.0392*	0.0062
CAR	(0.0011)	(0.0011)	(0.0033)	(0.0209)	(0.0211)	(0.0077)
ROAD	−0.0594	−0.0529	0.2391*	1.7779***	1.7831***	1.5327***
ROAD	(0.0563)	(0.0559)	(0.1182)	(0.2406)	(0.2425)	(0.1538)
year_2003	−0.2483***	−0.2432***	−0.5441***	−0.2565***	−0.2529***	−0.5458***
year_2003	(0.0255)	(0.0255)	(0.0930)	(0.0580)	(0.0582)	(0.0701)
year_2008	−0.0701***	−0.0681***	−0.2698***	−0.4547***	−0.4571***	−0.2390***
year_2008	(0.0176)	(0.0175)	(0.0698)	(0.0538)	(0.0539)	(0.0643)
year_2009	−0.0205	−0.0186	−0.1926***	−0.3865***	−0.3873***	−0.2271***
year_2009	(0.0122)	(0.0126)	(0.0369)	(0.0521)	(0.0524)	(0.0509)
year_2020	−0.6181***	−0.6132***	−2.4416***	0.0679	0.0776	−2.1808***
year_2020	(0.0403)	(0.0402)	(0.1684)	(0.0814)	(0.0816)	(0.1875)
Constant	0.1167	0.1362	−0.4655	6.2080***	6.1950***	2.6371***
Constant	(0.2186)	(0.2219)	(0.3026)	(0.9733)	(0.9793)	(0.8587)
N	600	600	600	630	630	630
R-sq				0.718	0.7161	0.7075
AR (2)	0.564	0.549	0.184
Hansen_p	1.000	1.000	1.000

Notes: Standard errors are in parentheses. *p < 0.1, **p < 0.05, ***p < 0.01.

To further evaluate the robustness of the results, alternative model specifications were estimated by replacing the continuous UGGps variable with a dummy variable (UGGps_d), which equals 1 if a province has at least one UNESCO Global Geopark and 0 otherwise. As reported in Table 8, the estimated effects of UGGps on tourism outcomes remain stable. In all cases, the UGGps_d variable demonstrates a consistent positive and statistically significant impact on all three categories of tourist arrivals. More importantly, the results reaffirm the market heterogeneity identified in the baseline models: the positive effect of UGGp designation is substantially larger for domestic tourist arrivals (β = 0.6389, p < 0.01) than for inbound tourist arrivals (β = 0.2577, p < 0.05). This stability enhances the credibility of the main findings regarding differential market responses to geopark development.

Discussion and implications

Discussion

This study confirms that UNESCO Global Geopark (UGGp) designation significantly increases both domestic and inbound tourist arrivals, consistent with prior research (Bonachea et al., 2023; Farsani et al., 2011). However, the panel regression analysis reveals important market-specific nuances. As shown in Tables 4 and 6, the effect on domestic tourist arrivals is substantially stronger than on inbound arrivals, validating our theoretical framework. Influenced by tourism distance decay (McKercher, 2021) and lagged international brand recognition (Lin et al., 2020a), UGGp designation initially stimulates domestic demand, while the impact on international tourism becomes prominent only in the second stage of the lifecycle, shifting the pattern from primarily domestic growth towards synchronized domestic and international expansion.

Within the nonlinear models, the positive linear term and negative quadratic term for UGGps unequivocally point to an inverted U-shaped relationship, demonstrating the applicability of diminishing marginal utility to tourism resource supply. This pattern is consistent with Lin et al.’s (2020a) findings concerning World Heritage Sites. Based on the fixed-effects regression for total tourist arrivals (Model 4 in Table 4), the estimated relationship is:

{l n T T A}_{i t} = 6.085 + 0.707 {U G G p}_{i t} - 0.124 {U G G p}_{i t}^{2} - 3.191 {l n F D I}_{i t} + 0.327 \ln {H o t e l}_{i t} - {0.083 U R}_{i t} - 0.040 {C A R}_{i t} + 1.679 {R O A D}_{i t} .

(6)

To identify the optimal point, we derive the marginal effect by taking the partial derivative of equation (6) with respect to UGGps:

\frac{\partial \ln I T A}{\partial U G G p s} = 0.707 - 0.249 U G G p s = 0

As illustrated in Figure 3, the turning point is 2.83. This indicates that when a province hosts approximately 2.83 UGGps, the marginal contribution to total tourist arrivals is maximised; beyond this, marginal returns diminish. Notably, as reported in Table 5, 96.74% of the sample observations have three or fewer UGGps, and the turning point of 2.83 falls slightly below three. Consequently, the overall effect of UGGps in China remains in a growth phase of increasing returns, with only a few provinces (those with four UGGps) possibly entering the zone of diminishing returns.

The study further examined synergy effects between UGGps and other designations. A significant positive synergy emerges between UGGps and National 5A-level attractions in the domestic market, whereas it is not statistically significant in the inbound market. Conversely, the interaction between UGGps and World Heritage Sites yields no significant synergy in either market. This helps explain, from another perspective, why the optimal threshold for domestic tourism is higher. The 5A designation is a highly influential domestic brand, and its complementary interaction with UGGps enhances domestic appeal, slows diminishing returns, and delays the turning point. In contrast, both WHS and UGGps fall under the UNESCO heritage umbrella, and their overlapping resource attributes may prevent tourists from perceiving a differentiated experience, thus failing to generate a detectable synergy effect (Wang and Yuan, 2020).

Policy implications

The findings yield several important policy recommendations. First, destination marketers should adopt a strategic approach to UGGp development that accounts for the observed nonlinear effects. The identified optimal number of approximately 2.83 UGGps for total arrivals provides concrete guidance for regional planning. Since most provinces currently remain below this threshold, continued strategic expansion is warranted. However, regions approaching or exceeding it should shift focus from quantity expansion to quality enhancement, emphasising thematic differentiation and functional specialisation to avoid product homogenisation.

Second, carefully designed brand integration strategies are needed to maximise synergies while minimising redundancy. The significant co-branding benefits between UGGps and 5A-rated attractions, particularly for domestic tourism, suggest that tourism authorities should develop coordinated marketing campaigns and unified visual identity systems that highlight complementary strengths. In contrast, the lack of synergy with World Heritage status indicates that simple brand stacking is ineffective; managers should instead develop clear differentiation strategies that emphasize each designation’s unique value proposition.

Third, marketing efforts should employ distinct approaches for domestic and international audiences. For Chinese domestic tourists, digital platforms such as social media and short-video applications are particularly effective for raising awareness. International marketing should leverage UNESCO’s global networks and emphasize the universal scientific significance of geological heritage. Developing specialized products such as geo-tours and scientific expeditions can help attract high-value international niche markets.

Fourth, the governance framework should evolve from an emphasis on accreditation attainment towards long-term performance management. Lessons should be drawn from the delisting of Kanawinka Global Geopark and the yellow-card warnings issued to Zhangjiajie, Lushan, and Wudalianchi in China. We recommend establishing comprehensive monitoring systems that track key indicators including visitor satisfaction, scientific output, educational impact, and community benefits. Introducing independent third-party evaluation mechanisms and robust public participation channels would help ensure that UGGps deliver meaningful sustainable development outcomes rather than merely serving as prestige markers.

Theoretical implications

This study makes three significant contributions to tourism and resource economics theory. First, by employing rigorous panel data methods, we demonstrate that UGGp establishment generates measurable increases in both domestic and international tourist arrivals, with a stronger effect on domestic tourism. These results validate and extend previous work suggesting that geopark development contributes to sustainable regional tourism economies (Bonachea et al., 2023; Dowling, 2010; Farsani et al., 2011). While existing research has predominantly consisted of micro-level qualitative case studies, our macro-scale quantitative approach offers stronger generalizability and more robust evidence.

Second, our research advances theoretical understanding of how scarce tourism resources influence destination development over time. Previous studies have linked prestigious designations such as WHS to tourism growth (Lin et al., 2019, 2020a, 2020b) but lacked a comprehensive framework explaining how these impacts evolve across lifecycle stages. We bridge this gap by integrating Butler’s (2008) Tourism Area Life Cycle model with resource scarcity theory, producing a dynamic four-stage mechanism that explains the evolving attractiveness of UGGps.

The demand emergence stage features limited brand recognition, with appeal stemming mainly from scarcity and attracting nearby domestic visitors. In the demand growth stage, international awareness develops, leading to more balanced growth across market segments. The demand expansion stage sees compounding effects as brand recognition and synergistic relationships accelerate tourism growth (Gnoth, 2002). Finally, the demand saturation stage emerges when diminishing marginal returns and binding carrying capacity constraints lead to stabilized or declining visitation (Marsiglio and Tolotti, 2024). The finding that domestic tourism exhibits an inverted U-shape while inbound tourism remains linear (Table 6) aligns with this framework: domestic demand in some provinces may have entered the saturation stage, whereas international demand remains in the expansion stage.

Third, our investigation of brand synergy effects contributes to destination branding theory. The significant synergy between UGGps and National 5A-level attractions, and its absence between UGGps and WHS, provides empirical support for the concept of brand attribute convergence (Lewis, 2023; Wang and Yuan, 2020). These results imply that successful multi-brand strategies require careful management of brand differentiation to avoid substitutability effects, with implications extending beyond geoparks to destinations holding multiple prestigious designations.

Conclusion, limitations, and future research directions

This study systematically examined the relationship between UGGps and tourism demand using provincial-level data from China. Our analysis confirms that UGGp designation significantly increases tourist arrivals, with particularly strong effects on domestic tourism. The relationship for total arrivals is inverted U-shaped, with the optimal number of UGGps being approximately 2.83. As 96.74% of the sample has three or fewer UGGps—below this turning point—the overall impact of UGGps on Chinese tourism remains in a growth phase of increasing returns. We also identify positive synergistic effects between UGGps and National 5A-level attractions but no significant complementarity with World Heritage status.

Several limitations suggest productive directions for future research. First, our provincial-level analysis cannot capture sub-regional variations or cross-boundary spillover effects. Second, all UGGps are treated as homogeneous, whereas different geopark types (e.g., volcanic, fossil, coastal) may have distinct tourism impacts. Third, reliance on aggregate visitation data leaves room for more nuanced understanding of visitor motivations and experiences. Fourth, although we have included a rich set of control variables, the possibility of omitted variable bias cannot be completely ruled out.

Three particularly promising research directions emerge. Spatial econometric methods such as geographically weighted regression could uncover regional variations and spillover effects. Incorporating big data sources like social media activity or mobile device tracking could provide more granular behavioral insights. Finally, visitor-level research using surveys or experimental methods could yield important insights into how tourists perceive and respond to multiple destination brands.

To conclude, this study advances both theoretical understanding and practical management of UGGps’ impact while identifying important avenues for future investigation. By combining rigorous empirical analysis with innovative theoretical synthesis, it provides a stronger foundation for evidence-based decision-making in geotourism development and destination branding.

Footnotes

ORCID iDs

Hongwei Zhong

Ming-Hsiang Chen

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Yunnan Fundamental Research Projects (202601CI07017).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Author biographies

Hongwei Zhong is an Associate Professor at School of Tourism and Hospitality Management, Yunnan University of Finance and Economics, China. His areas of research are tourism economics and tourism geography.

Liang Liu is a Lecturer at School of Tourism and Hospitality Management, Yunnan University of Finance and Economics, China. His research focuses on tourism economics and tourism geography.

Tie Wang is a professor at the College of Geography and Tourism, Qufu Normal University, China. His research focuses on tourism geography and community-based tourism.

Ming-Hsiang Chen is a Professor of Hospitality Business Management at Washington State University, USA. His areas of research are hospitality finance, hospitality philanthropy and tourism economics.

Wei Pu is an Associate Professor at School of Tourism and Hospitality Management, Yunnan University of Finance and Economics, China. Her research focuses on regional tourism economics.

References

Aas

Ladkin

Fletcher

(2005) Stakeholder collaboration and heritage management. Annals of Tourism Research 32(1): 28–48. https://doi.org/10.1016/j.annals.2004.04.005

Adie

Hall

Prayag

(2018) World heritage as a placebo brand: a comparative analysis of three sites and marketing implications. Journal of Sustainable Tourism 26(3): 399–415. https://doi.org/10.1080/09669582.2017.1359277

Agarwal

(2002) Restructuring seaside tourism: the resort lifecyle. Annals of Tourism Research 29(1): 25–55.

Aguirre

Mahr

Grewal

, et al. (2015) Unraveling the personalization paradox: the effect of information collection and trust-building strategies on online advertisement effectiveness. Journal of Retailing 91(1): 34–49. https://doi.org/10.1016/j.jretai.2014.09.005

Amaldoss

Jain

(2005) Pricing of conspicuous goods: a competitive analysis of social effects. Journal of Marketing Research 42(1): 30–42. https://doi.org/10.1509/jmkr.42.1.30.56883

Amaro

Chaves

Henriques

, et al. (2023) Motivation-based segmentation of visitors to a UNESCO Global Geopark. Geoheritage 15(2): 79. https://doi.org/10.1007/s12371-023-00848-3

Barrowclough

(2007) Foreign investment in tourism and small Island developing states. Tourism Economics 13(4): 615–638. https://doi.org/10.5367/000000007782696122

Bonachea

González-Díez

Hernández-Blanco

, et al. (2023) Suitability of valleys of Cantabria area for a UGGps proposal. Land 12(12): 2177.

Brock

(1968) Implications of commodity theory for value change. Psychological Foundations of Attitudes 243–275.

10.

Butler

(2008) The concept of a tourist area cycle of evolution: implications for management of resources. Canadian Geographies/Géographies Canadiennes 24(1): 5–12. https://doi.org/10.1111/j.1541-0064.1980.tb00970.x

11.

Cai

Han

, et al. (2023) Promoting geotourism in Dunhuang UNESCO Global Geopark. Geoheritage 15(1): 22. https://doi.org/10.1007/s12371-023-00796-y

12.

Cannon

Goldsmith

Roux

(2019) A self‐regulatory model of resource scarcity. Journal of Consumer Psychology 29(1): 104–127. https://doi.org/10.1002/jcpy.1035

13.

Catana

Brilha

(2020) The role of UNESCO Global Geoparks in promoting geosciences education for sustainability. Geoheritage 12(1): 1. https://doi.org/10.1007/s12371-020-00440-z

14.

Chen

Zheng

Zhang

, et al. (2022) Spatio-temporal evolution and obstacle factors analysis of tourism ecological security in Huanggang Dabieshan UNESCO Global Geopark. International Journal of Environmental Research and Public Health 19(14): 8670. https://doi.org/10.3390/ijerph19148670

15.

Cortes-Jimenez

Pulina

(2010) Inbound tourism and long-run economic growth. Current Issues in Tourism 13(1): 61–74. https://doi.org/10.1080/13683500802684411

16.

Cuccia

Guccio

Rizzo

(2016) The effects of UNESCO world heritage list inscription on tourism destinations performance in Italian regions. Economic Modelling 53: 494–508. https://doi.org/10.1016/j.econmod.2015.10.049

17.

Davidson

Sahli

(2015) Foreign direct investment in tourism, poverty alleviation, and sustainable development: a review of the Gambian hotel sector. Journal of Sustainable Tourism 23(2): 167–187. https://doi.org/10.1080/09669582.2014.957210

18.

Demir

Simonyan

Chen

, et al. (2020) Asymmetric effects of geopolitical risks on Turkey’s tourist arrivals. Journal of Hospitality and Tourism Management 45: 23–26. https://doi.org/10.1016/j.jhtm.2020.04.006

19.

Deng

Zou

(2021) Geotourism and geoparks for sustainable rural development and poverty alleviation: Huanggang Dabieshan UNESCO Global Geopark, China. Australian Journal of Earth Sciences 69(2): 286–301. https://doi.org/10.1080/08120099.2021.1965023

20.

Dowling

(2010) Geotourism’s global growth. Geoheritage 3(1): 1–13. https://doi.org/10.1007/s12371-010-0024-7

21.

Farsani

Coelho

Costa

(2011) Geotourism and geoparks as novel strategies for socio-economic development in rural areas. International Journal of Tourism Research 13(1): 68–81. https://doi.org/10.1002/jtr.800

22.

Gholipour

Tajaddini

Al-mulali

(2014) Does personal freedom influence outbound tourism? Tourism Management 41: 19–25. https://doi.org/10.1016/j.tourman.2013.08.010

23.

Gnoth

(2002) Leveraging export brands through a tourism destination brand. Journal of Brand Management 9: 262–280. https://doi.org/10.1057/palgrave.bm.2540077

24.

Han

Tian

, et al. (2017) From geopark to sustainable development: heritage conservation and geotourism promotion in the huangshan UNESCO Global Geopark (China). Geoheritage 10(1): 79–91. https://doi.org/10.1007/s12371-017-0227-2

25.

Hassan

(2000) Determinants of market competitiveness in an environmentally sustainable tourism industry. Journal of Travel Research 38(3): 239–245. https://doi.org/10.1177/004728750003800305

26.

Hausman

Taylor

(1981) Panel data and unobservable individual effects. Econometrica: Journal of the Econometric Society 49: 1377–1398. https://doi.org/10.2307/1911406

27.

Haywood

(1986) Can the tourist-area life cycle be made operational? Tourism Management 7(3): 154–167. https://doi.org/10.1016/0261-5177(86)90002-6

28.

Hosany

(2025) On scarcity and tourism: theoretical perspectives and implications. Journal of Travel Research 00472875251406552. Available at: https://doi.org/10.1177/00472875251406552

29.

Hovinen

(2002) Revisiting the destination lifecycle model. Annals of Tourism Research 29(1): 209–230. https://doi.org/10.1016/s0160-7383(01)00036-6

30.

Huang

Liu

Kandampully

, et al. (2020) Consumer responses to scarcity appeals in online booking. Annals of Tourism Research 80: 102800. https://doi.org/10.1016/j.annals.2019.102800

31.

Jacobsen

JKS

Iversen

Hem

(2019) Hotspot crowding and over-tourism: antecedents of destination attractiveness. Annals of Tourism Research 76: 53–66. https://doi.org/10.1016/j.annals.2019.02.011

32.

Kastenholz

(2004) ‘Management of demand’ as a tool in sustainable tourist destination development. Journal of Sustainable Tourism 12(5): 388–408. https://doi.org/10.1080/09669580408667246

33.

Kim

Chen

Jang

(2006) Tourism expansion and economic development: the case of Taiwan. Tourism Management 27(5): 925–933. https://doi.org/10.1016/j.tourman.2005.05.011

34.

Lee

Jayakumar

(2021) Economic impact of UNESCO Global Geoparks on local communities: comparative analysis of three UNESCO Global Geoparks in Asia. International Journal of Geoheritage and Parks 9(2): 189–198. https://doi.org/10.1016/j.ijgeop.2021.02.002

35.

Lee

Guillet

Law

, et al. (2012) Robustness of distance decay for international pleasure travelers: a longitudinal approach. International Journal of Tourism Research 14(5): 409–420. https://doi.org/10.1002/jtr.861

36.

Lewis

(2023) Evolution of geotourism in Australia from Kanawinka Global Geopark and Australian national landscapes to GeoRegions and geotrails:a review and lessons learned. Land 12(6): 1190. https://doi.org/10.3390/land12061190

37.

Ding

(2021) Study on methodology of assessing synergy between conservation and development of karst protected area in the case of the Diehong bridge scenic area of Jiuxiang Gorge Cave Geopark, Yunnan, China. Environment, Development and Sustainability 24(4): 5867–5886. https://doi.org/10.1007/s10668-021-01688-3

38.

Lin

Chen

Lin

, et al. (2019) Asymmetric effects of cultural and natural world heritage sites on tourism receipts. Current Issues in Tourism 23(24): 3134–3147. https://doi.org/10.1080/13683500.2019.1697649

39.

Lin

Chen

Lin

, et al. (2020a) Nonlinear impact of world heritage sites on China’s tourism expansion. Tourism Economics 27(4): 795–819. https://doi.org/10.1177/1354816620904886

40.

Lin

Chen

, et al. (2020b) 5A tourist attractions and China’s regional tourism growth. Asia Pacific Journal of Tourism Research 25(5): 524–540. https://doi.org/10.1080/10941665.2020.1741411

41.

Luo

Qiu

Lam

(2015) Urbanization impacts on regional tourism development: a case study in China. Current Issues in Tourism 19(3): 282–295. https://doi.org/10.1080/13683500.2015.1033385

42.

Luo

Mou

, et al. (2021) Exploring China’s 5A Global Geoparks through online tourism reviews: a mining model based on machine learning approach. Tourism Management Perspectives 37: 100769. https://doi.org/10.1016/j.tmp.2020.100769

43.

Marsiglio

Tolotti

(2024) The tourism area life cycle hypothesis: a micro-foundation. Tourism Economics 30(2): 345–360. https://doi.org/10.1177/13548166221138645

44.

McKercher

(2021) The impact of distance on tourism: a tourism geography law. In: Tourism Spaces. Routledge, pp. 137–141.

45.

Migoń

(2018) Geoheritage and world heritage sites. In: Geoheritage. Elsevier, pp. 237–249.

46.

Nicolau

Mas

(2006) The influence of distance and prices on the choice of tourist destinations: the moderating role of motivations. Tourism Management 27(5): 982–996. https://doi.org/10.1016/j.tourman.2005.09.009

47.

Okafor

Khalid

Then

(2018) Common unofficial language, development and international tourism. Tourism Management 67: 127–138. https://doi.org/10.1016/j.tourman.2018.01.008

48.

Park

Kim

(2022) Evolutionary aspects of scarcity information with regard to travel options: the role of childhood socioeconomic status. Journal of Travel Research 61(1): 93–107. https://doi.org/10.1177/0047287520971040

49.

Pásková

Zelenka

(2018) Sustainability management of UNESCO Global Geoparks. Sustainable Geoscience and Geotourism 2: 44–64. https://doi.org/10.18052/www.scipress.com/sgg.2.44

50.

Ringbeck

(2017) The world heritage convention and its management concept. Aspects of Management Planning for Cultural World Heritage Sites: Principles, Approaches and. Practices. Cham: Springer International Publishing, 15–24.

51.

Rosado-González

Sá

Palacio-Prieto

(2020) UNESCO Global Geoparks in Latin America and the Caribbean, and their contribution to agenda 2030 sustainable development goals. Geoheritage 12(2): 36. https://doi.org/10.1007/s12371-020-00459-2

52.

Shi

Wang

, et al. (2016) Impacts of star-rated hotel expansion on inbound tourism development: evidence from China. Applied Economics 48(32): 3033–3048. https://doi.org/10.1080/00036846.2015.1133900

53.

Stansfield

(1978) Atlantic City and the resort cycle background to the legalization of gambling. Annals of Tourism Research 5(2): 238–251. https://doi.org/10.1016/0160-7383(78)90222-0

54.

Vuković

Maiti

Petrović

(2023) Tourism employment and economic growth: dynamic panel threshold analysis. Mathematics 11(5): 112. https://doi.org/10.3390/math11051112

55.

Wang

Chen

(2021) Nonlinear impact of air quality on tourist arrivals: new proposal and evidence. Journal of Travel Research 60(2): 434–445. https://doi.org/10.1177/0047287519899993

56.

Wang

Yuan

(2020) Harmonizing the branding strategy of world natural heritage in China: visitors’ awareness of the multiple brands of Wulingyuan, Zhangjiajie. Geoheritage 12(2): 41. https://doi.org/10.1007/s12371-020-00462-7

57.

Yahya Sheibani

Zamanian

(2023) Geodiversity and geological treasure of Tabas UNESCO Global Geopark for geotourism development, new UGGps from Iran. Geoheritage 15(3): 106.

58.

Yang

Lin

Han

(2010) Analysis of international tourist arrivals in China: the role of world heritage sites. Tourism Management 31(6): 827–837. https://doi.org/10.1016/j.tourman.2009.08.008

59.

Yang

Jia

Liu

, et al. (2017) Car ownership and urban development in Chinese cities: a panel data analysis. Journal of Transport Geography 58: 127–134. https://doi.org/10.1016/j.jtrangeo.2016.11.015

60.

Zeng

Zhang

Ruan

, et al. (2024) Does the scarcity marketing of tourism cultural and creative products trigger impulsive consumption? Journal of Vacation Marketing 32(1): 52–69.

61.

Zhong

Deng

Xiang

(2008) Tourism development and the tourism area life-cycle model: a case study of Zhangjiajie National Forest Park, China. Tourism Management 29(5): 841–856. https://doi.org/10.1016/j.tourman.2007.10.002