Sustainability Assessment of Geotourism Consumption Based on Energy–Water–Waste–Economic Nexus: Evidence from Zhangye Danxia National Geopark

Abstract

The development of geotourism and the establishment of geoparks can generate new job opportunities, new economic activities, and additional sources of income, with great significance in achieving the United Nations Sustainable Development Goals. Tourists often consume more energy and water and generate more waste in order to seek a more comfortable state during their travels. This research took Zhangye Danxia National Geopark in the north slope of the Qinghai–Tibet Plateau in China as an example and combined field research with questionnaires to construct a non-radial model (SBM) based on energy-water-waste-economic nexus. From the perspective of consumption, this research focuses on the consumer characteristics of geotourism sustainability based on the energy–water–waste–economic nexus (GTS-EWWE) and the driving factors behind them under different degrees of consumption. The elderly, children, and high-income tourists can contribute more to the sustainability of geotourism. Stay duration in the geopark and energy consumption are the native driving factors behind the sustainability of geotourism. However, with the improvement of the tourism consumption level, the marginal effect of the negative influence of both factors is diminishing gradually. While tourism expenditure is a positive driving factor, the tourist’s travel mode and the consumer’s awareness of ecological–environmental actions will contribute more to the sustainability of geotourism. New energy technologies to promote the green development of geoparks is significant. This research aims to provide a reference for the sustainability assessment of geoheritage sights and to provide evidence for the appropriate management policy with respect to their sustainable development.

1. Introduction

Geotourism is a type of tourism that is based on geoheritage and geoparks, with “natural resources of landscape, landforms, fossil beds, rocks and minerals, with an emphasis on appreciating the processes that are creating and created such features” [1]. Geoparks and geotourism are relatively new activities within tourism [2]. Since the global growth of geotourism and geoparks, the importance of geoheritage has become more evident, especially from the perspective of the development of geotourism. The geological element focuses on geology and landscape and includes both “form”, such as landforms, rock outcrops, rock types, sediments, soils, and crystals; and “process”, such as volcanism, erosion, glaciation, etc. [3]. While developing geotourism, the establishment of geoparks can generate new job opportunities, new economic activities, and additional sources of income [4,5], with great significance in achieving the United Nations Sustainable Development Goals. Geotourism, by financing the conservation of geosites and stimulating local economic growth, presents itself as a comprehensive strategy [6]. Geological heritage is affected by rainfall and river systems, which are a natural point of connection between geological heritage and the geographical environment, especially in arid and semi-arid areas [7]. Some research has found that it is important for protecting geological heritage in the interest of sustainable development, proposing steps to improve coordination between protection and exploitation [8]. Geoparks play a crucial role in low-carbon tourism and sustainable development. The Xingwen Global Geopark has shown a constant promotion of low-carbon development, with an increasing carbon footprint but decreasing carbon intensity [9]. Similarly, the Batur Global Geopark in Indonesia aims to balance geology, environment, social–culture, economic, and conservation aspects, contributing to tourism development in the region [10]. These geoheritage sights highlight the importance of responsible tourism, dynamic carbon footprint measurement, and the need for sustainable development through the protection of geoheritage and promotion of geotourism. However, the growing economic significance of nature tourism may lead to negative environmental and sociocultural impacts, such as visitor overcapacity and inadequate knowledge during the construction of tourism infrastructures, potentially putting geoheritage resources at risk [11]. Moreover, related studies have shown that tourists often consume more energy and water and generate more waste in order to seek a more comfortable state during their travels [12]. For example, in terms of energy consumption, tourism accounts for about 3.2% of the global total, and it will continue to increase year by year [13]. In terms of water use, China is one of the most water-scarce countries listed by the United Nations, but the consumption of water in tourism consumption is increasing year by year. By 2050, the tourism industry’s water consumption will increase by 92% [14]. One study found that the water footprint of each tourist is about two to three times that of a local resident [15]. The impact of municipal solid waste management is one of the least investigated areas [16]. While waste represents one of the most visible impacts affecting tourism and the environment [17], previous findings show a wide fluctuation in the generation rate [18]. In conclusion, the consumption of energy and water, as well as the generation of waste, during tourism activities will threaten the geotourism system and environmental carrying capacity. The evaluation of tourism ecological security in geoparks underscores the importance of balancing tourism activities with environmental protection to improve ecological security [19]. Therefore, it is of vital importance to conduct in-depth research on the energy–water–waste–economic nexus (EWWE) in geoheritage sights or geoparks to achieve the sustainable development of geotourism.

Geotourism research concentrates in Europe, East Asia, the Middle East, and South America. The largest research communities are active in Italy, Brazil, China, and Poland [20]. Geotourism asset evaluation is a major research direction. Studies have been conducted on the geological value of geoheritage [21,22], its cultural and educational value [23], and market assessment [24]. The development of geotourism destinations is also a key aspect of geotourism research, encompassing studies such as the analysis of characteristics and tourism route design for tourist destinations [21] and geotourism products [25]. The assessment of the educational value of Earth protection and sustainable development in geoparks has emerged as a hot topic in recent research [26,27]. In recent years, with the growing emphasis on sustainable tourism and the role of tourism in achieving global sustainable goals, an increasing number of scholars have focused on the biodiversity and sustainable development of geoheritage and geoparks in the context of geotourism development. Simbaña-Tasiguano (2024) investigated the relationship between knowledge and the utilization of the geographic diversity of geoparks to promote Earth conservation and geotourism and discovered that the interaction between geographic diversity, biodiversity, and residents using geosites offers optimal conditions for promoting geotourism with cultural, natural, and adventure components [28]. Obradovic (2023) conducted a survey of Serbia’s national geoparks and found that residents generally support the development of geotourism, but there are certain concerns regarding the social costs of tourism development. This is a matter that tourism policymakers should pay particular attention to in order to mitigate social impacts and ensure ongoing support from local residents [29]. Xie (2024) analyzed 455 tourists in Chinese geoparks and found that responsibility attribution, attitude, values, and place attachment can influence the sustainability of geotourism behavior [30]. Carrillo (2024) proposed an integrated plan to promote sustainable geotourism by rationally designing tourism routes, guiding the participation of rural community residents, creating meaningful experiences, attracting tourists, and benefiting the local community, thereby facilitating overall and sustainable rural development [31]. Sen (2024) evaluated the geological heritage of the Tuwaiq Mountains in Saudi Arabia and found that the degradation risk of geological heritage indicates the need for Earth protection [32]. Anougmar (2024) discovered that Earth’s diversity is under threat. Most studies emphasize the recreational characteristics and economic value of geological resources, but there is a lack of significant oversight regarding the impact of tourism on geological resources [33].

The research on the relationships between energy, water, waste, and the economy is an important tool for analyzing the coordinated development of economic systems and ecosystems, as well as their impacts, and it has become a focus of attention for many countries and scholars. Studies show that the coordinated development of ecological environment and economy is crucial for sustainable development [34,35]. Energy structure transformation plays a vital role in promoting the coordinated development of economic and environmental systems, especially in regions with better economic development and high levels of human capital and urbanization [36]. Additionally, the coordinated development of social economic development and ecological environmental development is essential for sustainable development [37]. Scholars have found that there are complex, dynamic, and uncertain interactions between energy, water, waste, and the economy, making it difficult to coordinate management. Strategies based on any single resource will lead to unforeseen serious consequences [38]. Research has focused on the relationship between social and economic demand-driven resources and environmental footprint changes and economic policies based on consumption. Owen (2018) estimated the UK’s energy, water, and food consumption accounts for the period of 1997–2013 with the aim of reducing the environmental impact of product consumption [39]. Becken and Susanne (2017) studied the relationship between water and energy in the tourism accommodation sector and found that managing energy–water–waste synergies in the accommodation sector can improve energy investment efficiency and environmental outcomes [40]. Xiong (2022) used input–output tables and tourism statistical data to analyze the relationship between the tourism industry, energy, food, and water in Beijing and proposed that ecological innovation by tourism enterprises, the environmental regulation of tourism by the government, and guiding tourists to engage in green consumption can promote energy–food–water synergistic emission reduction [41]. Shi (2013) studied the impact of energy consumption on tourist sites and believes that promoting green development can be achieved by reducing waste and increasing the production of renewable energy [42]. Becken (2002) argued that optimizing the number of tourists, management style, technological equipment, and fuel combination can drive sustainable development in tourist destinations. Wan (2010) studied the impact of waste on tourist sites and proposed that changing transportation modes or using environmentally friendly vehicles is an important way to reduce the ecological footprint of waste in tourist sites [43]. Overall, these findings highlight the necessity of balancing economic growth with environmental protection to achieve tourism sustainable development and ensure the well-being of both economic systems and ecosystems.

There have been many efforts and progress made in researching EWWE, as well as in conducting empirical analysis of the tourism industry. However, there are some limitations. First, most studies about EWWE involve the tourism industry, its sectors, and tourism destinations, with relatively fewer studies focusing on geoparks or geotourism destinations. Second, the consumption angle mainly focuses on environmental behavior, and there has not been enough research focused on measuring the tourism process of different consumption groups. Third, the sustainability research on geoparks and geotourism concentrates on single environmental indicators such as water use, energy use, and carbon emissions, with less research based on the comprehensive EWWE.

Facing this existing knowledge gap, this research focuses on the following questions: What are the consumer characteristics of geotourism sustainability based on the energy–water–waste–economic nexus (GTS-EWWE)? What are the driving factors of geotourism sustainability under different degrees of consumption? How can geotourism achieve sustainability?

Geoparks are an important basis for geotourism, reflecting the interplay between various elements in the geotourism process and facilitating the unveiling of the sustainable development laws of geotourism. This research took Zhangye Danxia National Geopark in the north slope of the Qinghai–Tibet Plateau in China as an example and combined field research with questionnaires to construct a non-radial model (SBM) based on the energy–water–waste–economic nexus. Starting from the perspective of consumption, this research analyzed the sustainability of geotourism. The innovative aspects of this paper mainly are as follows: first, the geopark at the micro level is regarded as an ecological economic system, and the geotourism ecological economic relationship is studied; second, the consumer characteristics and driving factors of geotourism sustainability are explored, and recommendations are discussed. According to the above issues and the innovations, the purpose of this research is to pinpoint the crucial elements necessary to understanding how to generate and plan better future policies for geoparks management, especially through knowing consumer behavior, which can, in turn, be changed by educational and proactive campaigns of sensibilization.

The rest of the research is arranged as follows: Section 2 is the introduction of the study area, Section 3 is the method section, including the research framework, and the research methods; Section 4 details the results; and Section 5 presents the research discussion and conclusion.

2. Study Area: Zhangye Danxia National Geopark

Zhangye Danxia National Geopark of Gansu Province (northwestern China) has been protected for its stunning varicolored badlands of Early Cretaceous (Aptian–Albian) claystones [44]. The location of Zhangye Danxia National Geopark is shown in Figure 1. Named by Chinese geologists, Danxia landforms refer to an geoheritage developed on red beds, characterized by scarp slopes [45]. Exhibiting striking red cliffs and precipitous walls, these geoheritage sights bear similarities to karst topographies and constitute a unique yet fragile geographical and natural aesthetic [46]. These geoheritage sights not only feature distinctive terrains but also serve as sanctuaries for a variety of rare flora and fauna, thereby possessing considerable ecological and touristic significance [47]. Zhang (2007) conducted an analysis of the strata, structures, landforms, paleogeographic environment, and external dynamic conditions that contributed to the formation of the Danxia landform and the colored hills in Zhangye Danxia National Geopark and categorized the Danxia landform in this area into three types of landscape: “Window-Palace pattern, Pillar pattern, and Laneway pattern” [48]. The “Window-Palace” pattern is the most typical in Zhangye Danxia National Geopark. It is formed on the steep walls of horizontal and gently inclined red rock strata through the combined effect of water erosion along joints and lateral erosion in soft rock interlayers, including via wind. The “Pillar” pattern Danxia landform is affected by water erosion, which widens the vertical joints, causes the surrounding rock to collapse, and makes the steep cliff slope retreat continuously with an increasingly smaller area. The entire rock mass retreats to become “fort-like residual peaks” or isolated “stone pillars”. The “Laneway” pattern Danxia landform is a landscape formed by the downward incision and erosion of water along the original structural planes or vertical joints, resulting in the shapes of “alleyways” and “crevices through which sunlight shines”. As one of the most popular tourist attractions in China, Zhangye Danxia National Geopark has seen a surge in visitor numbers in recent years (the growing tourism trend is shown in Figure 1). In 2019, the park received 2.858 million visitors and generated operating revenue of 43.12 million yuan [49]. Growing tourism consumption is posing a serious challenge to the ecological environment. Under the global climate change background, the snow line of Qilian Mountain, an important ecological barrier in the Tibet Plateau [50], has risen significantly, leading to complex impacts on regional water resources crisis [51,52], desertification exacerbation [53], and the sustainable development of oases [54]. There is an urgent practical need to study the sustainability of geotourism in Zhangye Danxia National Geopark, and it also has typical representativeness for the sustainability research of similar geoheritage.

3. Methods and Data

3.1. Design and Methods of the Questionnaire Survey

In order to ensure the accuracy and reliability of the survey data, the survey subjects selected in this paper are all tourists to Zhangye Danxia National Geopark. According to the design of geotourism sustainability indicators, the survey questionnaire includes three parts: basic information of tourists, tourists’ consumption patterns, and tourists’ willingness to promote green geotourism development. The questionnaire is detailed in Supplementary Material. The questionnaire distribution period is from 30 September 2018 to 30 October 2018. The questionnaires were obtained through two methods: self-help filling and on-site manual guidance. A total of 871 valid questionnaires were collected. The sample structure is shown in

Table 1. Sample structure.

Category	Category Features	Frequency	Rate (%)
Gender	male	364	41.79
	female	507	58.21
Age	14 years old and under	9	1.03
	15–24 years old	404	46.33
	25–34 years old	231	26.50
	35–44 years old	96	11.01
	45–60 years old	126	14.45
	61–70 years old	3	0.34
	70 years old and above	3	0.34
Occupations	Civil servants	95	10.91
	Managers of enterprises and institutions	158	18.14
	Service workers	53	6.08
	Workers	41	4.71
	Farmers	20	2.30
	Military personnel	3	0.34
	Retirees	17	1.95
	Students	364	41.79
	Others	120	13.78
Educational attainment	Primary school and below	13	1.49
	Junior high school	40	4.59
	Senior high school	75	8.61
	Junior college	106	12.17
	Undergraduate	586	67.28
	Master’s degree and above	51	5.86
Income	CNY 0.5k and below	51	5.86
	CNY 0.5k–1k	92	10.56
	CNY 1.01k–2k	118	13.55
	CNY 2.01k–3k	181	20.78
	CNY 3.01k–4k	149	17.11
	CNY 4.01k–5k	89	10.22
	CNY 5.01k–6K	64	7.35
	CNY 6.01k–7k	38	4.36
	CNY 7.01k–8k	22	2.53
	CNY 8k and upon	67	7.69
Tourism Mode	Individual self-guided tour	369	42.37
	Family or travel with relatives and friends in a self-guided way	436	50.06
	Organized by employing company	32	3.67
	Organized by Travel Agencies	34	3.90

.

3.2. Sustainability Assessment Method of GST-EWWE

DEA (data envelopment analysis) is a non-parametric approach employed for assessing the performance of decision-making units by establishing an effective production frontier and determines whether the decision-making units are situated on the effective production frontier. The superiority of DEA lies in its ability to handle multi-input and multi-output issues of multiple decision-making units and is applicable to the efficiency evaluation of various types of organizations and industries. The Undesirable-SBM model, proposed by Tone in 2001, includes undesirable outputs [55] and overcomes the input–output surplus problem and the influence of slack variables on the results in traditional DEA models. The survey questionnaire data filled out by 871 tourists represent the decision units for this study. Each decision unit consists of three vectors: inputs, expected outputs, and undesirable outputs. The expression of the SBM model based on variable returns to scale is as follows:

$$\left\{\begin{array}{l}\min \phi =\left(1-{\displaystyle \frac{1}{n}}\sum _{i=1}^n{\displaystyle \frac{s_i^{-}}{x_{ik}}}\right)/\left[1+{\displaystyle \frac{1}{p_1+p_2}}\left(\sum _{r=1}^{p_1}{\displaystyle \frac{s_r^{+}}{y_{rk}}}+\sum _{r=1}^{p_2}{\displaystyle \frac{s_t^{b-}}{y_{tk}}}\right)\right]\\ s.t. x_k=X\lambda +s^{-},y_k=Y\lambda -s^{+},b_k=B\lambda +s^{b-}\\ \mu \ge 0,s_i^{-}\ge 0,s_r^{+}\ge 0,s_t^{b-}\ge 0\end{array}\right.,$$

(1)

where $\phi$ represents the level of sustainability of geotourism for each tourist based on the EWWE; $n$, $p_1$, and $p_2$ represent the number of input variables, expected output variables, and unintended output variables, respectively; $X$, $Y$, and $B$ represent the input, expected output, and undesirable output matrices; and μ represents the column vector. The range of $\phi$ value is [0, 1], and the closer it is to 1, the more efficient the decision unit is. Using this method, the sustainability of geotourism for 871 tourists was assessed.

3.3. Regression Analysis on Driving Factors

This research conducted an analysis of the driving factors of geotourism sustainability in Zhangye Danxia National Geopark using the Tobit model. The Tobit model was first proposed by Nobel Prize winner James Tobin in 1958 [56] as a method of restricted dependent variable regression that is suitable for regression where the explained variable takes values within the range of [0, 1]; hence, it is also known as the truncated regression model. This research applied StataMP17.0 to conduct a regression analysis of the driving factors of geotourism sustainability. The regression model formula is as follows:

$$\begin{gathered} y_{it}^{*}=\alpha x_{it}+{\beta }^T{x}_{it}+{\epsilon }_{it} \\ {y}_{it}=\left\{\begin{array}{l}{y}_{it}^{*}, y_{it}^{*}\ge 0 \\ 0, y_{it}^{*}\le 0\end{array} i=1,\cdots , N and t=\right.1,\cdots ,T \\ {\epsilon }_{it}~N\left(0,{\sigma }^2\right), \end{gathered}$$

(2)

where $\beta$ is the estimated parameter; $y_{it}$ is the censored observed value of the dependent variable $y_{it}^{*}$; ${\epsilon }_{it}$ is the random disturbance term; $x$ is the explanatory variable; i represents each individual tourism consumer; N indicates the quantity of tourism consumers; and t and T typically denote time or years. In this study, when analyzing the GTS-EWWE driving factors for all tourism consumption, T = 1. When analyzing the GTS-EWWE driving factors based on different consumption levels, T = 6.

3.4. Correlation Analysis of Impact Factors

The influence factors are realized through the Pearson correlation coefficient, which is a statistical method commonly used to reflect the degree of linear correlation between variables [57]. It evolved from a similar but slightly different idea proposed by Francis Galton in the 1880s. Pearson’s correlation coefficient is defined as the sum of the product of the covariance and standard deviation of two variables (x, y), as shown in Equation (3):

$${\rho }_{x,y}={\displaystyle \frac{{\sum }_{i=1}^n\left(x_i-\overline{x}\right)\left(y_i-\overline{y}\right)}{\sqrt{{\sum }_{i=1}^n{(x_i-\overline{x})}^2}\sqrt{{\sum }_{i=1}^n{(y_i-\overline{y})}^2}}},$$

(3)

where $\rho$ is the correlation coefficient, which takes values between −1 and 1, indicating the direction of change of the independent variable x and the dependent variable y. The direction and degree of the lead–lag correlation between the independent and dependent variables can be expressed by the sign and absolute value of the correlation coefficient, i.e., if $\rho$ > 0, it indicates that the dependent variable tends to increase as the independent variable increases; if $\rho$ < 0, it indicates that the dependent variable tends to decrease as the independent variable increases; if $\rho$ = 0, it means that the dependent variable shows no trend as the independent variable increases. In this research, the correlation coefficient is employed to analyze impact factors, offering a more intuitive understanding of how consumer characteristics and consumption behavior influence the sustainability of geotourism, thereby facilitating the optimization of the priority order for implementing policy recommendations.

3.5. Indicators and Data

Drawing on the SBM model’s input–output structure and incorporating tourists’ consumption in geoparks, a comprehensive set of indicators was developed, encompassing input indicators, output indicators, and undesirable output indicators. The input indicators consist of labor input, capital investment, energy consumption, and water resource utilization; the output indicators pertain to tourism expenditure; while the undesirable output indicators refer to waste emissions resulting from tourism activities. Alongside data collected through questionnaires, additional data were sourced from the official website of Zhangye City Government (http://www.zhangye.gov.cn/, accessed on 8 January 2020) and the management department of Zhangye Danxia National Geological Park. The sources for these indicators and data are presented in

Table 2. Indicators and data.

	Indicators	Data Source	Unit
Input	Tourism Consumption Labor Input	Labor force/Tourist population	Person
	Tourism Consumption Capital Input	Capital investment/Tourist population	Yuan
	Tourism Consumption Energy Input	Tourism activity energy consumption × project per capita energy consumption	Ton of Standard Coal
	Tourism Consumption Water Resource Input	Tourist stay duration × (Annual total water consumption/Tourist population)/Per capita tourist stay duration	Ton
Output	Tourism Consumption Revenue	Ticket + Other expenses for travel	Yuan
Undesirable-output	Tourism Consumption Waste Emission	Tourist stay duration × (Total amount of waste discharge/Number of tourists)/Per capita tourist stay duration	Ton

.

4. Results

4.1. The Characteristics of the Tourist Consumers for the Sustainability of Geotourism

The GST-EWWE in Zhangye Danxia National Geological Park is mainly concentrated at a relatively low level (Figure 2), and there is a considerable disparity in tourism consumption EWWE among different tourist groups. The distribution characteristics of the GST-EWWE exhibit a lighter left tail compared to the normal distribution, while the right tail is heavier, falling into a positively skewed distribution; i.e., the mean is greater than the median and the median is greater than the mode. The GST-EWWE are distributed within the range of [0.15, 1], with a mean of 0.1761. The GST-EWWE in Zhangye Danxia National Geological Park is relatively low. It is of the utmost urgency to identify the driving factors of the GST-EWWE and enhance geotourism sustainability in a more targeted manner. Additionally, when the GST-EWWE of the majority of tourists is below 0.2, there are still some tourists whose sustainability is distributed within the interval of [0.6, 1], indicating that individual tourists have achieved high sustainability in tourism consumption. This reveals significant variations in tourists’ tourism consumption behaviors, and it is necessary to further identify the sustainability of geotourism under different tourism consumption characteristics.

Based on the data characteristics of the GST-EWWE in Zhangye Danxia National GeoPark, it is categorized into four distinct levels: high score (CE ≥ 0.4); medium score (0.2 ≤ CE < 0.4); low score (0.1 ≤ CE < 0.2); and ultra-low score (CE < 0.1) (Figure 3). The characteristics of the population under different GST-EWWE score levels are analyzed, respectively. Tourists with low and ultra-low scores account for 45% and 26% of the total surveyed population, respectively, while there are only 47 tourists with high scores, accounting for 5% of the total surveyed subjects.

The analysis results of tourist characteristics with different GST-EWWE scores are presented in Figure 3. Males are more inclined to contribute to the sustainability of geotourism. Male visitors account for the majority of those with high GST-EWWE scores in the geological park, nearly twice the number of female visitors with high GST-EWWE. However, low-score female visitors predominate, and male visitors only amount to half the number of the female visitors. Teenagers and senior tourists are more inclined to contribute to the sustainability of geotourism. Military personnel and retirees contribute far more to GST-EWWE than those in other occupations, while students contribute relatively less to the GST-EWWE. The extent to which education contributes to the GST-EWWE sis uncertain. The income levels of visitors with high GST-EWWE scores are mainly concentrated between 4001 and 5000 yuan and above 8000 yuan. The contribution of the high-income group (i.e., above 8000 yuan) to high-scoring GST-EWWE is much higher than that of visitors with other income levels. The self-guided travel mode in family groups or groups of friends contributes relatively highly to high-scoring GST-EWWE. To obtain a more accurate driving mechanism of the GST-EWWE according to visitor characteristics, a correlation analysis and econometric regression analysis were carried out.

4.2. The Driving Factors for GST-EWWE

4.2.1. Correlation Analysis

From the perspective of consumers, the operation of the ecological economic system of geoparks depends on tourists’ travel activities. The GST-EWWE is influenced by the characteristics of the consumer population and consumption behaviors. Therefore, this paper selects influencing factors from four aspects, namely, the characteristics of tourism consumers, the travel patterns of consumers, the ecological and environmental impact of consumers, and the perception of green tourism by consumers for analysis. Specifically, the characteristics of the consumer population include four factors: gender, age, education attainment, and income level; the travel mode of consumers include three factors: the duration of stay near the geopark, the duration of stay in the geopark, and travel expenses; the ecological and environmental impact of consumers includes three factors: energy consumption, water consumption, and waste discharge; and the perception of green tourism includes three factors: cognition of green tourism, supportive attitude towards green tourism, and willingness to pay for green tourism. The data mainly come from questionnaires, and are all discrete types or degree indicators. The higher the value, the stronger the expression degree of the variable. The results of the correlation analysis are presented in

Table 3. Analysis of the correlative factors of GST-EWWE.

Number	Relevant Factors	Correlation Coefficient	Number	Relevant Factors	Correlation Coefficient
1	Gender (GEN)	−0.1103 *	8	Energy Consumption (ENERGY)	−0.1741 *
2	Age (AGE)	0.0125	9	Water Consumption (WATE)	−0.1052 *
3	Educational Attainment (EDU)	−0.0942 *	10	Waste Discharge (WAST)	−0.1052 *
4	Income (INCO)	0.1183 *	11	Cognition of Green Tourism (COGN)	0.0886 *
5	Stay Time near the Park (STAYNE)	0.1848 *	11	Supportive Attitude towards Green Tourism (SUPP)	0.0181
6	Stay Time in the Park (STAYIN)	−0.1052 *	13	Willingness to Pay for Green Tourism (WILL)	0.1481 *
7	Expenditure (EXPE)	0.8181 *

Note: * indicates significance at a p-value of 0.05.

.

Among the characteristics of the consumer group, gender and education level are significantly negatively correlated with the GST-EWWE; that is, the higher the proportion of females or the higher the education level, the lower the GST-EWWE. Income level positively affects GST-EWWE.

The duration of stay near geoparks in consumers’ tourism patterns exhibits a significant positive correlation with GST-EWWE. Conversely, the duration of stay within geoparks shows a significant negative correlation with GST-EWWE. Tourism expenditure has a significant positive correlation with the sustainability of geotourism consumption in terms of EWWE; that is, the higher the tourism expenditure, the better GST-EWWE.

Tourists’ ecological environmental impact, energy consumption, water consumption, and waste discharge are all significantly negatively correlated with the GST-EWWE; that is, an increase in the aforementioned variables will cause a reduction in the GST-EWWE. For instance, every 1% increase in waste discharge in scenic spots will lead to a 0.1052% decrease in the GST-EWWE.

Consumers’ cognition of green tourism and their willingness to pay both present a significant positive correlation with the GST-EWWE; that is, the higher the tourists’ cognitive level of green tourism and the stronger their willingness to pay, the higher the GST-EWWE will be in geological tourism. For example, for every 1% increase in the willingness to pay for green tourism, the GST-EWWE in the geopark will increase by 0.1481%.

4.2.2. Driving Factors

Based on the indicators with significant correlations, a Tobit regression model for the sustainability drivers of geological tourism consumption was constructed. To mitigate the heteroscedasticity in the model data, the tourism expenditure data were subjected to logarithmic processing, while the original values of the other variables were retained. Through multiple collinearity tests, it was discovered that the variance inflation factors of water consumption and waste discharge were relatively high. To prevent the distortion of model estimation resulting from the highly correlated relationship among variables, the two variables of water consumption and waste discharge were excluded from the regression analysis. The ultimate formula of GST-EWWE regression model is as follows:

$$CE={\gamma }_0+{\gamma }_{1}GEN+{\gamma }_{2}EDU+{\gamma }_{3}INCO+{\gamma }_{4}STAYNE+{\gamma }_{5}STAYIN+{\gamma }_{6}COGN+{\gamma }_{7}WILL+{\gamma }_{8}ENERGY+{\gamma }_{9}lnEXPE+\epsilon .$$

(4)

According to the GST-EWWE of different tourists, the driving factors were regressed, respectively. The regression results are presented in

Table 4. Regression results of driving factors GST-EWWE.

	EWWE Sustainability		High-Score EWWE Sustainability		Medium-Score EWWE Sustainability		Low-Score EWWE Sustainability		Ultra-Low-Score EWWE Sustainability
	Coef.	t	Coef.	t	Coef.	t	Coef.	t	Coef.	t
GEN	0.0001	0.0300	−0.0485	−1.6000	0.0178	3.2300 ***	0.0008	0.3400	−0.0018	−1.4800
EDU	−0.0027	−1.4000	−0.0251	−2.1200 **	0.0002	0.0900	−0.0007	−0.5600	0.0002	0.2200
INCO	−0.0005	−0.6400	−0.0003	−0.0400	0.0004	0.3300	−0.0002	−0.4400	−0.0008	−3.0300 ***
STAYNE	0.0008	0.4800	0.0043	0.5100	0.0017	0.8100	0.0010	1.0400	0.0005	0.7400
STAYIN	−0.0261	−18.8600 ***	−0.0843	−7.9000 ***	−0.0184	−9.0400 ***	−0.0114	−11.1700 ***	−0.0051	−10.2800 ***
COGN	−0.0024	−1.2300	−0.0074	−0.6900	−0.0005	−0.1600	−0.0016	−1.2800	−0.0001	−0.2300
WILL	0.0023	1.5500	−0.0105	−1.1900	−0.0016	−0.8100	0.0008	0.8800	−0.0004	−0.8100
ENERGY	−0.0013	−5.5700 ***	−0.0051	−4.2700 ***	−0.0008	−2.3600 **	−0.0006	−3.5700 ***	−0.0002	−2.8000 ***
LNEXPE	0.1528	50.0200 ***	0.2580	7.1300 ***	0.0974	11.6700 ***	0.0589	14.3900 ***	0.0380	12.8400 ***
_cons	−0.5529	−25.8200	−0.7653	−3.1600	−0.2859	−5.8600	−0.1263	−6.0300	−0.0787	−5.2300
Log likelihood	1282.1751		47.0777		389.9666		922.8448		760.7348

Note: ** and *** indicate significance at a p-value of 0.05 and 0.01, respectively.

.

The regression analysis of the driving factors for high-score GST-EWWE indicates that the education level is inversely correlated with the GST-EWWE. Specifically, for each additional level of education, GST-EWWE decreases by 0.025. The duration of stay in geological parks and energy consumption are also inversely correlated with GST-EWWE. For every increase of 1 kg of standard coal in energy consumption, the GST-EWWE reduces by 0.0051; and for every additional hour of stay by tourists, it decreases by 0.843. Expenditure is positively correlated with the GST-EWWE. For every 1% increase in tourists’ expenditure, the GST-EWWE by 0.25%. Evidently, for the high-scoring group in terms of the GST-EWWE, the effects of various driving factors are more pronounced.

The regression analysis of the driving factors behind the medium-score GST-EWWE reveals that gender is positively proportional to the GST-EWWE in the medium-score group; that is, tourist activities conducted by females can increase the GST-EWWE in the medium-score group by 0.0178. The influences of the duration of stay in geoparks, energy consumption, and tourism expenditure on the high-score GST-EWWE are all weaker than those on the other score groups.

The regression analysis of the driving factors being the low-score and ultra-low-score GST-EWWE sustainability reveals that the duration of stay in geoparks and energy consumption are inversely proportional to the GST-EWWE, while tourism expenditure is directly proportional. Moreover, the intensity of their effects is gradually decreasing. In the driving factors of the ultra-low-score EWWE sustainability of geotourism consumption, the monthly income of tourists becomes a negative driving factor for the GST-EWWE; that is, for every increase of 1000 yuan in income, the GST-EWWE decreases slightly by 0.0008.

4.3. Analysis of Drivers of GST-EWWE at Different Consumption Levels

4.3.1. Characteristics of GST-EWWE at Different Consumption Levels

In the analysis of the relevant influencing factors for the sustainability of geological tourism consumption EWWE, it was discovered that the correlation coefficient of tourist expenditure was the highest (0.8181, Table 2), demonstrating that tourist expenditure is the most crucial driving factor for GST-EWWE. Hence, tourist expenditure was employed as the grouping basis to further refine the analysis of the driving factors of GST-EWWE, enabling the 871 decision units to be re-grouped at different consumption levels. GST-EWWE was calculated independently at different consumption levels to make the distribution of GST-EWWE more balanced. Tourists were classified into six groups based on the amount they spent in the scenic area (

Table 5. Grouping of geotourism consumption in Zhangye Danxia National Geopark.

Group	Expenditure	Population	Percentage of Population	Statistical Characteristics of GTS-EWWE
				Mean	Standard Deviation	Minimum	Maximum
Low consumption	Less than 50 yuan	163	18.714	0.5803	0.1861	0.3824	1.0000
Medium-low consumption	50–100 yuan	189	21.699	0.5376	0.1407	0.3824	1.0000
Medium consumption	100–200 yuan	181	20.781	0.5271	0.1274	0.3824	1.0000
Medium-high consumption	200–300 yuan	127	14.581	0.5157	0.1416	0.3824	1.0000
High consumption	300–500 yuan	117	13.433	0.4590	0.1132	0.3530	1.0000
Ultra-high consumption	More than 500 yuan	94	10.792	0.3979	0.1431	0.2338	1.0000

). Figure 4 presents the histogram of the distribution of GST-EWWE in different consumption groups.

A total of 163 tourists—that is, 163 decision units—belonged to the low-consumption group of the scenic area. The histogram exhibits a positively skewed distribution; that is, the left tail is lighter than the normal distribution and the right tail is heavier than the normal distribution, as depicted in Figure 4. The GST-EWWE with low-consuming tourists is concentrated within the range of [0.3823, 1], with a mean value of 0.5803. Compared to other consumption groups, the standard deviation of the GST-EWWE with the low-consumption group is the largest. This suggests that the degree of dispersion of the GST-EWWE with the low-consumption group is the highest, and approximately 10% of the tourists are distributed in the high GST-EWWE range.

The medium-low consumption group had the largest number of tourists, totaling 189, and had the largest proportion among all consumption groups. The distribution characteristic remained a positively skewed distribution, with a mean greater than the mode. Compared to the low consumption group, the mean of the medium-low consumption group was slightly smaller at 0.5376. The number of decision units with the GST-EWWE ranging from 0.6 to 0.8 increased for the medium-low consumption group. There were 181 tourists in the medium consumption group. Compared to the low and medium-low consumption groups, the GST-EWWE was more evenly distributed between 0.4 and 0.7.

The GST-EWWE with the medium-high and high consumption groups is distributed more frequently at a relatively low level of approximately 0.4, and the dispersion degree of the GST-EWWE in the high consumption group is the lowest. There is a total of 94 tourists in the ultra-high consumption group, and the GST-EWWE is distributed within the range of [0.2338, 1], with the mean being the smallest among all groups at 0.3979. On the whole, the distribution of the GST-EWWE with ultra-high consumption tourists is more uniform and closer to a normal distribution. From the perspective of econometrics, the regression results of this group of data are more accurate, with smaller errors.

4.3.2. Analysis of the Correlation Factors of GST-EWWE Under Different Consumption Levels

The extent of the influence of GST-EWWE under different consumption levels is presented in

Table 6. Correlation analysis of the influencing factors of GST-EWWE.

Correlation Factors	GTS-EWWE with Low Consumption	GTS-EWWE with Medium-Low Consumption	GTS-EWWE with Medium Consumption	GTS-EWWE with Medium-High Consumption	GTS-EWWE with High-Consumption	GTS-EWWE with Ultra-High Consumption
GEN	−0.1463	0.0438	0.033	0.0452	−0.1221	−0.1243
AGE	−0.1737 *	−0.0078	0.0249	−0.0438	−0.0861	0.0911
EDU	−0.1806 *	−0.1324	0.0428	0.007	−0.0426	−0.0122
INCO	−0.2398 *	−0.1315	0.0148	0.0831	−0.1604	0.003
STAYNE	−0.0489	0.0469	−0.0973	−0.0544	−0.1256	−0.0812
STAYIN	−0.7015 *	−0.6011 *	−0.4766 *	−0.5796 *	−0.4881 *	−0.4484 *
ENERGY	−0.2180 *	−0.2225 *	−0.3313 *	−0.3796 *	−0.2762 *	−0.0659
WATE	−0.7015 *	−0.6011 *	−0.4766 *	−0.5796 *	−0.4881 *	−0.4484 *
WAST	−0.7015 *	−0.6011 *	−0.4766 *	−0.5796 *	−0.4881 *	−0.4484 *
COGN	0.0443	−0.0253	0.1377	−0.0565	0.075	−0.0465
SUPP	0.0198	−0.0594	−0.0293	0.1854 *	−0.0314	0.0115
WILL	−0.0714	0.0126	−0.1585 *	−0.1421	−0.0134	0.1042

Note: * indicates significance at a p-value of 0.05.

and Figure 5. When the amount of tourism consumption is less than 50 yuan, the characteristics of the consumer group represented by age, educational attainment, and income level have a significant negative correlation with the GST-EWWE of the geopark; that is, the older the age, the higher the educational attainment, and the higher the income level, the lower the level of the GST-EWWE. It is recommended that Zhangye Danxia National Geopark pay more attention to the above groups during the process of enhancing GST-EWWE, thereby achieving precise management and control of the GST-EWWE of specific tourists at a low consumption level.

At other consumption levels, the duration of stay, energy consumption, water consumption, and waste discharge within geological parks all exhibit a significantly negative correlation with the GST-EWWE; that is, the longer the stay within the geopark, the greater the energy consumption, water consumption, and waste discharge, and the lower the GST-EWWE. This suggests that the main driving factors influencing the GST-EWWE with different consumption levels are tourism patterns and the impact of tourism on the ecological environment. It is necessary to shorten the visiting time appropriately and to simplify the visiting routes as much as possible without compromising tourists’ experiences, while calling on tourists to consume in a low-carbon manner, to utilize water resources rationally, to enhance consumers’ environmental awareness, and to promote green consumption. Additionally, the willingness to pay for green tourism in tourism perception shows a distinct negative correlation with the GST-EWWE medium consumption group but a significantly positive correlation with the GST-EWWE of the medium-high level of tourism consumption. This indicates that the consumption categories of different consumer groups are clearly differentiated, exerting diverse effects on GST-EWWE.

4.3.3. Analysis of the Driving Factors of GST-EWWE Under Different Consumption Levels

To further explore the driving factors affecting the GST-EWWE at different consumption levels, the corresponding indicators with a strong influence on the EWWE sustainability were selected as the independent variables for the regression analysis. For instance, when investigating the GST-EWWE with respect to low-consumption tourists’ travel, the indicators with a more significant impact on the GST-EWWE with a low consumption level were chosen, mainly including age, education level, income level, duration of stay in the scenic area, energy consumption, water consumption, and waste emissions. Meanwhile, the indicators of water consumption and waste emissions with collinearity were eliminated. Eventually, the Tobit regression model for the GST-EWWE with low consumption established is as follows:

$$C{E}_1={\gamma }_{10}+{\gamma }_{11}AGE+{\gamma }_{12}EDU+{\gamma }_{13}INCO+{\gamma }_{14}STAYIN+{\gamma }_{15}ENERGY+\epsilon .$$

(5)

Similarly, based on the correlation factor analysis (3.4.1), separate regression models were established for other consumption tourists. The regression model for medium and low consumption with respect to the GST-EWWE with high consumption is as follows:

$$C{E}_2={\gamma }_{20}+{\gamma }_{21}STAYIN+{\gamma }_{22}ENERGY+\epsilon ,$$

(6)

$$C{E}_3={\gamma }_{30}+{\gamma }_{31}STAYIN+{\gamma }_{32}ENERGY+{\gamma }_{33}WILL+\epsilon ,$$

(7)

$$C{E}_4={\gamma }_{40}+{\gamma }_{41}STAYIN+{\gamma }_{42}ENERGY+{\gamma }_{43}ATTI+\epsilon ,$$

(8)

$$C{E}_5={\gamma }_{50}+{\gamma }_{51}STAYIN+{\gamma }_{52}ENERGY+\epsilon ,$$

(9)

$$C{E}_6={\gamma }_{60}+{\gamma }_{61}STAYIN+\epsilon .$$

(10)

The regression results of the driving factors of the GST-EWWE at different consumption levels are presented in

Table 7. Driving factors of GST-EWWE.

Indictor	GTS-EWWE with Low Consumption		GTS-EWWE with Medium-Low Consumption		GTS-EWWE with Medium Consumption		GTS-EWWE with Medium-High Consumption		GTS-EWWE with High-Consumption		GTS-EWWE with Ultra-High Consumption
	Coef.	t	Coef.	t	Coef.	t	Coef.	t	Coef.	t	Coef.	t
AGE	−0.0265	−2.3500 **
EDU	−0.0206	−1.5100
INCOME	−0.0097	−1.9600 **
STAYIN	−0.0893	−11.5200 ***	−0.0666	−10.4400 ***	−0.0467	−7.5600 ***	−0.0522	−6.8200 ***	−0.0395	−6.4500 ***	−0.0425	−4.8400 ***
ENERGY	−0.0050	−2.7700 ***	−0.0041	−3.5400 ***	−0.0058	−4.9500 ***	−0.0040	−3.1100 ***	−0.0036	−3.7100 ***
SUPP							0.0163	0.9000
WILL					−0.0071	−1.0300
_cons	1.0978	14.5600	0.7609	34.5900	0.7271	26.0500	0.7086	16.3000	0.6395	23.8100	0.5691	15.0800
Log likelihood	58.9295		131.6721		131.6721		83.4699		108.0056		53.5054

Note: ** and *** indicate significance at a p-value of 0.05 and 0.01, respectively.

.

It was discovered through the regression results that when the tourists’ consumption level was less than 50 yuan, the GST-EWWE presented a significantly inverse relationship with age, income, duration of stay in the park, and energy consumption. To be specific, for every 1000 yuan increase in tourists’ income, the GST-EWWE declined by 0.0097; for every one-hour increase in the duration of stay in the park, the GST-EWWE dropped by 0.0893; and for every additional 1 kg of standard coal in energy consumption, the GST-EWWE decreased by 0.005.

From low-to-medium and ultra-high consumption levels, the duration of stay and energy consumption of tourists in the park are significantly negatively correlated with the GST-EWWE. Specifically, for each additional hour of stay by tourists in the park, the GST-EWWE from low to medium to ultra-high consumption levels decreases by 0.0667, 0.0467, 0.0522, 0.0395, and 0.042, respectively. Furthermore, for each additional kilogram of energy consumption by tourists in the park, the GST-EWWE from low to high consumption levels decreases by 0.0041, 0.0058, 0.004, and 0.0036, respectively.

5. Discussion

The length of stay of tourists in a destination has a significant impact on the economic development of that destination. Research indicates that while total spending by tourists increases with higher-cost forms of tourism and length of stay, the locally retained spending is predicted only by the length of stay [58]. According to existing research, the length of tourists’ visits can have both positive and negative effects on the ecological environment of scenic areas, depending on factors like tourist perceptions, experiences [59], and tourists’ interactions [60] with the destination. From the empirical research in this paper, it is found that the longer the stay near but outside geoparks, the more beneficial it is for the improvement of the GST-EWWE, while the longer the stay in geoparks, the more detrimental it is. Among them, the stay time in the park has the most significant impact on the GST-EWWE with respect to low-consuming tourists. This is intuitive since the longer the low-consumption group stays in the geopark, the higher the energy and water consumption levels will be, and more garbage would be produced. Moreover, the limited economic value of geological parks from a low-consumption perspective cannot make up for the ecological–environmental impact caused by a longer stay, which is bound to result in a lower GST-EWWE. It is recommended to call on tourists to form good low-carbon tourism consumption behaviors, to stimulate tourists’ consumption demands from the demand side, and simultaneously to improve tourism products, rationally lay out the visiting routes within geoparks, and enhance the visiting efficiency of geoparks, thereby achieving an increase in the GST-EWWE.

Studies show that incorporating both direct and indirect energy use is crucial as each contributes significantly to the greenhouse emissions and climate impact of tourism [61]. Greenhouse emissions because of energy use have a significant impact on heritage sites [62]. In this research, we also found that energy consumption has the most obvious impact on the GST-EWWE with respect to medium-consuming tourists, suggesting that the consumption behavior of medium-consuming tourists is more common and widespread. The large number of tourists results in a large degree of energy consumption [63]. But exiting research has also found that the expansion of the tourism sector can offset the adverse effects of energy consumption stemming from trade openness, indicating a complex relationship between tourism indicators and energy demand [64]. Low-carbon behavior performance can help save energy usage [65]. It is suggested investors, the administration committees, tourists, and local governments be pushed to drive the low-carbon behavior of tourists in geoheritage parks.

UNESCO promotes the creation of UNESCO geoparks to conserve the environment and enhance sustainable economic development by increasing public awareness of the Earth’s geological heritage [66]. Some geological heritage sites have already conducted beneficial practices throughout the world. The protection of geological heritage sights in Xinjiang, China, involves coordinating protection and exploitation to ensure the sustainable development of local economies [8]. The Fernando de Noronha archipelago geoparks in Brazil take actions to enforce sustainability through ensuring appreciation of the heritage sights and promoting the development of sustainable tourism, along with environmental protection and interpretation [67]. The Siwa Oasis in Egypt is significant in terms of conservation and exploitation for research, education, and tourism purposes [68]. The Jeju Island geopark has taken action with respect to trail activation, community participation, awareness of local heritage, and tourist willingness to protect local heritage to promote local sustainable development [69]. This research constitutes an in-depth exploration of the sustainable development of geoheritage sights and tourism consumption. It scrutinizes the comprehensive effects on the economy and ecological environment of geological heritage sights from the perspective of tourists’ consumption behaviors and serves as a response and expansion to the above studies.

From the perspective of the sustainability of geotourism, the following suggestions are presented: First, the construction of the environmental interpretation system should be enhanced to reinforce the educational value of geoparks in terms of Earth’s protection, while guiding low-carbon tourism consumption behaviors. Second, geotourism routes and tourism products should be designed in a targeted manner. The design of tourism routes should be integrated with the evaluation of geoheritage value, and routes with high cultural and geoscientific values should be selected. Meanwhile, tourism products dominated by research and study should be developed. Third, based on tourists’ consumption habits, the location and duration of their stay within geoparks should be controlled and guided, and the flow of tourists should be rationally diverted to achieve a win–win development of both high consumption and sustainability. Fourth, geoparks need to establish a recycling system for energy, water, and waste to realize the recycling and reuse of resources and improve full-process experience and education with respect to ecological and environmental protection in geoparks. Fifth, the public Earth protection awareness of stakeholders, such as the communities and governments where the geoparks are located, should be cultivated to promote sustainable regional development.

6. Conclusions

The overall distribution of the GST-EWWE in geoparks is at a relatively low level of less than 0.4. This distribution takes on a pyramid form, with a greater number of low-score distributions at the bottom and fewer high-score distributions at the top. Among tourists with high GST-EWWE scores, the elderly, children, and high-income tourists are notable groups, while among tourists with low and ultra-low scores, middle school students and female tourists are prominent groups.

The main driving factors for the GST-EWWE primarily encompass the tourists’ stay duration in the geopark, energy consumption, and tourism expenditure. Along with the enhancement of the GST-EWWE, the marginal effects of energy consumption and stay duration in the geopark on the GST-EWWE increase, while the marginal effect of the tourism consumption level declines. The high-scoring groups with respect to the GST-EWWE can elevate the sustainability of a geopark more effectively by reducing energy consumption and stay duration in the geopark.

Tourism patterns, the impact of the tourism ecological environment, and green tourism perception have a relatively pronounced influence on the GST-EWWE at different consumption levels. The indicators that are significant for the GST-EWWE with different consumption levels are the duration of stay and the energy consumption within the park. This validates the previous analysis. As the tourism consumption level rises, the marginal effect of energy consumption and the duration of stay in the park on the GST-EWWE decreases. Moreover, tourists’ travel mode the consumers’ awareness of ecological environment action will contribute more to the GST-EWWE.

The aim of this research is to disclose the interaction between the economic system of supply and demand and the ecological–environmental system in geotourism by considering the characteristics of sustainable consumption in geotourism and the driving factors of its sustainable development, as well as to discuss methods of realizing the sustainable development of geotourism. This research provides a reference for the sustainable assessment of geotourism and proposes a direction for sustainable management in terms of consumption behaviors in geotourism for geoheritage sights and geoparks.

Geological heritage serves as a crucial window through which humanity can apprehend nature and comprehend the Earth as our home. The assessment of geological heritage constitutes a significant precondition for its sustainable management. This research provides a reference for the sustainability assessment of geoheritage, starting from the consumption behaviors of tourists under the framework of the EWWE, which reveals, more completely, the interaction intensity between the economic system and the ecosystem of geotourism in terms of supply and demand. In the future, it will be possible to continuously investigate and study geoparks, expanding our efforts to stakeholders such as the government, enterprises, and local communities. Through the Internet and big data means or data materials, we can comprehensively analyze and evaluate the driving and influential factors of the GST-EWWE, as well as the role and intensity of each factor, to reveal the driving mechanism behind the sustainability of geological tourism in a deeper manner and explore more paths towards improvement.