Assessment of Cultural Ecosystem Service Values in Mountainous Urban Parks Based on Sex Differences

Abstract

Urban parks are vital for providing cultural ecosystem services (CESs) to residents. However, few studies have explored sex-based differences in CES demand, particularly within mountainous urban parks. This study aimed to elucidate sex-based differences in the perceptions and preferences for CESs and landscape elements and explore their relationship in mountainous urban parks. Using value-labeled photographs from an equal number of male and female volunteer visits to Eling Park in Chongqing, China, the SolVES model was employed to investigate the relationship between sex-specific perceptions of CESs and landscape elements. The results showed that males preferred slopes and steps, garden facilities, and recreation facilities, whereas females preferred overlooks that offer multiple CES values, including science and ecological education, and aesthetic and cultural heritage values. Females perceived social relational value at the lotus pond and Kansheng Tower, whereas males perceived inspirational value at Kansheng Tower, the entrance, and the cliffside path. Males linked inspirational value to fort-like ridges and cliffs. Females associated inspirational value with slopes and steps. Based on the findings, the study recommends enhancing CESs specific to mountainous landscapes and incorporating sex-sensitive design elements. Ultimately, these efforts aim to position parks as key components of urban sustainable development, promoting overall resident well-being.

1. Introduction

In April 2005, the United Nations released the Millennium Ecosystem Assessment, which, for the first time, proposed that cultural ecosystem services (CESs) are one of the four ecosystem services [1]. CESs encompass the non-material benefits that humans derive from ecosystems, including spiritual enrichment, cognitive development, aesthetic enjoyment, and recreational opportunities [2]. CESs are inherently intangible and context-dependent, relying on physical environments, such as farmlands, wetlands, and mountains, for their actualization [3]. Urban parks, acting as indispensable components of urban ecosystems, significantly contribute to urban ecological functions and enhance resident well-being [4,5], serving as critical spatial carriers of urban CESs [6]. However, with the rapid pace of urbanization, city expansion has led to a reduction in the natural environment, adversely affecting the supply capacity of green space CESs. Concurrently, the increasing demand for CESs among urban residents has exacerbated the imbalance between the supply and demand of urban green space CESs [7]. Research on park CESs is essential for improving residents’ physical and mental well-being, enhancing urban aesthetics, and promoting a balanced human–nature relationship [8].

CES evaluation methods mainly involve monetization and non-monetization assessments [9]. Monetization is often limited to values like recreation, ecotourism, and cultural values [10,11], while non-monetization is widely used for its broader applicability across categories. Consequently, scholarly interest in non-monetary valuation methods has grown significantly due to their ability to optimize environmental resource allocation. Current research on park CESs employs various non-monetary methods, each with unique strengths and limitations in capturing user perceptions and preferences. Traditional methods like questionnaires [12] and field surveys [13] effectively gather detailed insights but are labor-intensive and hard to scale [14]. Social media analysis offers abundant user-generated content but often lacks socio-demographic data, limiting group-specific analysis [15,16,17]. Participatory mapping provides precise spatial data but requires extensive participant training [17]. The SolVES model effectively reveals CESs spatial distribution but relies on photographic data with detailed background information [18,19]. VEP produces numerous object-linked photos but is costly and quality-demanding. The advantage of integrating the SoIVES model with the VEP method [20,21] lies in the fact that VEP can enhance the SoIVES model’s superior processing of spatial locations, allowing each point on the generated CES distribution map to correspond to actual photographs and the landscape elements depicted on them. VEP ensures that the digitization of the SoIVES model is not confined to a mere point containing valuable data but rather represents the tangible existence of the actual landscape.

Studies have demonstrated an association between CESs and the physical characteristics of green space landscapes [22,23,24]. For instance, Li et al. [22] found that natural landscape elements, such as plants, animals, and water features, along with supporting facilities in urban green spaces, significantly influence overall satisfaction when personally participating in services provided by urban green spaces. Similarly, Oteros-Rozas et al. [23] demonstrated that trees and lawns in five major European cultural heritage sites are related to recreation, social relations, and cultural heritage values. Various natural and cultural landscape elements collectively underpin the provision of multiple cultural services [24]. However, challenges persist in delineating the value chain of green space ecosystem landscapes, functions, and spaces, as identifying which specific CES values are attributed to particular landscape elements remains difficult [25]. Urban parks, characterized by their diverse natural and cultural landscapes, provide a broader range of CESs than other types of green spaces, such as forests, greenways, and front-yard greenery, with higher perception frequencies for place attachment, aesthetics, recreation, cultural identity, and education [26]. Mountainous urban parks, in particular, offer more complex and varied relationships between landscape features and cultural services owing to their unique natural environmental conditions. For instance, the high altitude and steep terrain of mountainous parks create unique visual landscapes and influence visitor experiences and the types of cultural activities available, such as hiking and rock climbing [27]. Concrete landscape features and biophysical attributes in these parks play a significant role in narratives about individual CES experiences. Notable or memorable places, owing to striking features, are likely to be more prominently sought out during visits [28].

CESs emerge from the interaction between physical landscape attributes and the subjective perceptions of individuals. Recent research emphasizes the necessity of evaluating CESs in terms of both physical landscape features and the subjective differences among various groups [29,30]. Studies have demonstrated that demographic and sociological factors, such as age, sex, occupation, and ethnicity, influence individual values, preferences, and behaviors, leading to different perceptions of CESs [20]. However, Velandia et al. [31] found that the characteristics of parks played a more significant role in shaping socio-cultural values than the socio-demographic attributes of their stakeholders, such as age, education level, and sex, which were found to have no substantial impact. Zeratsion et al. [32] also found that education level, income, and occupation significantly influenced the appreciation of recreational and educational services provided by urban forests, whereas sex and age showed no significant correlation with identified CESs. Fortnam et al. [33] explored how ecosystem services contribute to various aspects of well-being, revealing sex-specific differences in how men and women value these services through an empirical study conducted across eight coastal communities. Swapan et al. [34] also demonstrated that males generally prioritize recreational services and focus less on habitat and historical services. In the complex terrain and ecologically diverse environment of mountainous parks, these sex-based value differences may become even more intricate [35].

Sex is a crucial component of socio-demographic attributes. The perceived differences in cultural ecosystem services (CESs) based on sex determine the varying landscape needs of urban parks. Conducting research on sex differences in CESs ensures that park design and management can incorporate nuanced, differentiated planning to meet the needs and expectations of different sex groups [36]. Studies on sex differences in CESs help adjust the planning of landscape elements and functional configurations in parks based on preferences. Understanding the relationship between landscape elements and CESs from a sex perspective is essential for current landscape practices in mountainous urban parks [33]. By calculating and mapping the actual supply and demand of ecosystem services and their spatial mismatches, planners can identify key development areas for enhancing the availability of CESs for both men and women in mountainous parks, thereby improving land use and management efficiency. Simultaneously, research from a sex perspective can externalize the deep-seated differences in sex characteristics within the field of CES perception, thereby constructing a sex-equitable ecological governance system in urban parks and breaking down sex stereotypes in cultural landscape concepts. Based on the SoIVES model with the VEP method, the combination enhances the authenticity and scientific rigor of the data, clearly delineating the differences in landscape preferences and value demands among different sex groups.

Despite the increasing recognition of the importance of evaluating CESs in urban parks, a significant research gap remains, particularly regarding CESs in mountainous urban parks. Challenges include uncertainties in evaluation scales and depth, limited studies on the relationship between value services and mountainous landscape elements, and the constraints of objective methods [37,38]. Socio-demographic factors such as age, sex, and cultural background also influence CES perception; however, previous studies have not thoroughly explored the impact of sex differences on CES perception in mountainous urban environments. Furthermore, this study aligns with the United Nations’ Sustainable Development Goals, particularly those focused on promoting well-being (Goal 3), achieving sex equality (Goal 5), and ensuring inclusive, safe, and accessible urban spaces (Goal 11). These goals underscore the urgent need to understand and address sex-specific needs and perceptions in urban green spaces to foster sustainable and equitable urban development. Therefore, this study aims to develop a classification index for CESs tailored to mountainous urban parks and to elucidate the relationships between CES perceptions and landscape elements among young individuals of different sexes under mountainous conditions. Additionally, this study seeks to provide a theoretical basis for optimizing spatial layout, rational land use, and effective landscape element configuration in mountainous urban parks. Compared to other mountainous parks in Chongqing, such as Pipashan Park (too small in area), Zou Rong Park (limited landscape elements), and Binjiang Park (lacking mountainous features), Eling Park stands out with its moderate size, rich landscape elements, distinct mountainous characteristics, and well-defined main trails, making it an ideal site for our experimental needs.

2. Methods

2.1. Study Area

Eling Park, located at the highest point of the Yuzhong Peninsula in the mountainous city of Chongqing, China, is a notable example of the traditional Bayu garden style. The construction of the park, covering an area of 6.63 hectares at an elevation of 300–388 m, began in 1909, was completed in 1911, and was opened to the public in 1958 [39]. This iconic park is renowned for its rich diversity of plant species, which thrive in the subtropical monsoon humid climate of the region, which is characterized by hot summers, cold winters, and high humidity with frequent overcast conditions.

Eling Park features over 10 scenic spots and three monuments, including Kansheng Tower, Exiang Mountain Villa, Sanyou Pavilion, and the lotus pond. The park encompasses approximately 0.3 hectares of water area, 0.85 hectares of paved area, and 0.983 hectares of green area, providing a rich array of natural and historical resources for relaxation, recreation, education, and commemoration. Based on its landscape distribution and mountainous terrain characteristics, the landscape elements of Eling Park are categorized into non-mountainous and mountain-specific landscape elements. The non-mountainous landscape elements include 11 categories: ponds, flowers and lawns, shrubs and trees, animals, modern architecture, antique architecture, recreation facilities, garden facilities, roads, squares and platforms, and people (visitors). In contrast, the mountain-specific landscape elements, particularly considering the unique ecological and topographical features of the park, are divided into five categories: slopes and steps, fort-like ridges and cliffs, air-raid shelters, tree roots, and overlooks (distant views from high points) (Figure 1).

2.2. Data Acquisition

2.2.1. Experimental Design

Based on the experimental purpose of exploring the impact of sex variables on CESs, we established a framework comprising visitor-employed photography (VEP), questionnaires, and semi-structured interviews, such that the experimental methods can meet the requirements of wide applicability and high effective data [21]. Visitor-employed photography (VEP) involves hiring tourists to capture photos that reflect the value of cultural ecosystem services, with software like “2bulu 7.5.5” used to record the photos and their location data. “2bulu 7.5.5” generates publicly accessible KML files with walking paths, photo points, walking time, speed, and poster height, relying on mobile phone positioning for data accuracy and reliability. We collected basic demographic information on volunteers through a questionnaire survey, including age, sex, local or non-local place of residence, and whether or not it was their first visit to the park. During interviews, volunteers discussed their visit motivations and memorable experiences with the research team, enhancing survey reliability and offering valuable insights for further analysis.

We posted volunteer recruitment information on university social media platforms. Eligible volunteers underwent pre-experiment training, covering CES concepts, “2bulu 7.5.5” usage, designated routes, and timing/weather requirements. To ensure the absolute influence of sex factors, we selected college students aged 18–22 with similar educational backgrounds. Volunteers conducted the experiment on sunny afternoons during rest days in October and November 2023 to maintain environmental consistency, relaxed conditions, and photo clarity. They visited Eling Park under specified conditions, captured at least 50 photos meeting at least one CESs value, categorized them, and submitted the photos to the research team (Figure 2). We set the number of male and female participants equally at 45 each.

2.2.2. Pre-Survey

On 25 September 2023, the research team, encompassing four individuals, conducted a preliminary survey to assess the reliability of the “2bulu 7.5.5” app and the feasibility of the survey procedures.

Based on field research in Eling Park and the preliminary analysis, combined with historical research on relevant park CESs [1,32], we selected and defined CES values suitable for mountainous urban parks, ultimately establishing 11 relevant indicators (

Table 1. Eleven established CES values.

CES Values	Description
Cultural diversity	I can feel the coexistence of urban culture, modern culture, literature, art, religion, and tourism.
Spiritual and religious	The landscape allows me to experience religious and spiritual meanings, and I feel a sense of reverence from it.
Science and ecological education	I can gain scientific and ecological knowledge from this place, and I believe it can provide an avenue for education.
Inspiration	I can be inspired to create cultural inspirations and artistic expressions.
Aesthetic	I can feel beauty from visual, auditory, sensory, and other aspects.
Social relations	I can feel that the place provides an environment for social relationships, and I believe that the site creates unique social relationships.
Sense of place	I can emotionally connect with the place and feel a sense of security and belonging from there, and I believe that the place is unique.
Cultural heritage	I can feel the existence of historical landscapes (nature and humanities) or species with cultural value in the area, and I can feel their historical and cultural value [10,40].
Future	I have been awakened to a sense of responsibility to protect natural ecosystems and cultural spiritual wealth, and I hope that future generations can also understand and experience them.
Health	I can breathe fresh air, exercise, relax my mind, and nurture my body here.
Recreation	I plan to visit the area because of its unique features and natural or cultural landscapes that are of tourist value.

CESs, cultural ecosystem service.

). Additionally, an appropriate tour route was provided for the volunteers, covering most areas of the park. The path nodes included all CES values, directing volunteers to capture various landscape elements within the park.

2.2.3. Recruitment of Volunteers and Experimental Conduct

All young volunteers were voluntarily participants and showed enthusiasm for this study. They were physically healthy, had normal color vision, good physical strength, and met the experimental requirements. The research team provided informed consent forms and obtained authorization to use the path files and photos of the records, ensuring anonymity and confidentiality of the data. Before the formal survey, we distributed basic information and statistical questionnaires to the volunteers, informing them of the 11 CES values and their meanings in printed form. At the same time, we informed the volunteers that they could choose to travel alone or in groups. However, for the photography and classification of indicators, they could not be influenced by their peers but should have completed the task based on their own judgment.

The formal survey was conducted on sunny days between 1:00 p.m. and 5:00 p.m. from 10 October to 8 November 2023. During this period, the landscape of Eling Park remained almost unchanged, with stable temperature, humidity, and lighting conditions and no significant noise interference. The research team collected 90 path files and categorized 5106 photographs, including 2670 from the male group and 2436 from the female group.

2.3. Data Analysis and Quantification

2.3.1. SolVES 4.0 Model

The SolVES 4.0 model evaluates public perceptions of ecosystem services using non-monetary valuation by integrating spatial and social survey data to quantify, spatially analyze, and predict ecosystem services. It comprises three sub-modules: the Ecosystem Services Social-Values Model, the Value Mapping Model, and the Value Transfer Mapping Model. The Value Transfer Mapping Model relies on pre-existing and validated models, whereas this study aims to propose a new CESs value model. Therefore, only the first two modules were utilized. Among them, the Ecosystem Services Social-Values Model uses a gradient scale of 1–10 to measure the level of CES values. Additionally, the model incorporates average nearest neighbor analysis to assess spatial clustering. SolVES 4.0 utilizes the MaxEnt maximum entropy model to calculate the receiver operating characteristic curve. The “area under the curve” (AUC) represents the reliability of the evaluation results of the model, with AUC values ranging from 0 to 1. A training AUC value greater than 0.7 indicates effective evaluation and higher AUC values denote better model fit [41].

The model requires four data types: CES perception points, study area boundaries, environmental layers, and survey data. Based on the characteristics of Eling Park, 23 environmental layer indicators were selected (

Table 2. Environmental data inputs for the SolVES model.

Data	Format	Resolution	Source
CES points	Shapefile	0.1 m	Volunteers visited Eling Park and used the “2bulu 7.5.5” app to capture data through photography.
Study area	Shapefile	0.1 m	Drawn according to the CAD map of Eling Park.
Land use type	Raster	0.1 m
Distance to water landscape	Raster	0.1 m	Based on MAP WORLD (China National Platform for Common Geospatial Information Services) satellite images, digitized in ArcGIS 10.8, and calculated using Euclidean distance.
Distance to roads	Raster	0.1 m
Slope	Raster	0.1 m
Elevation	Raster	0.1 m
Hill shade	Raster	0.1 m
Distance to tower	Raster	0.1 m
Distance to air-raid shelter	Raster	0.1 m
Distance to landscape sculpture	Raster	0.1 m
Distance to pavilion	Raster	0.1 m
Distance to circular corridor	Raster	0.1 m
Distance to bridge	Raster	0.1 m
Distance to rockery	Raster	0.1 m
Distance to square and platform	Raster	0.1 m
Distance to modern architecture	Raster	0.1 m
Distance to retaining wall	Raster	0.1 m
Distance to flower	Raster	0.1 m
Distance to antique architecture	Raster	0.1 m
Distance to green space	Raster	0.1 m
Distance to cliff	Raster	0.1 m
Distance to hillside	Raster	0.1 m

CESs, cultural ecosystem service; CAD, computer-aided diagram.

) [20]. An Eling Park computer-aided design (CAD) map was created using an Eling Park visitor map and field surveys, and environmental layer indicators were processed into Raster format data in ArcGIS 10.8. Using PostgreSQL 16.0 for data storage, the MaxEnt model for maximum value analysis, and QGIS 3.26 as the mapping software, the pixel size in SolVES 4.0 was set to 0.1 m for the 6456 m² area of Eling Park, with a search radius parameter of 10 m.

2.3.2. Content Analysis of the Photography Dataset

The research team analyzed the photography dataset captured by the volunteers to identify three distinct types of information: the geographical coordinates of each photograph, the CES values identified by the visitors in the photograph titles, and the landscape elements present in the photographs. Landscape elements have been previously analyzed based on their proportion in a photograph, with the criterion that the element occupies more than 20% of the image [20]. The analysis in this study was performed by four researchers in pairs, with each pair making joint decisions. In cases of disagreement, all four researchers collaborated to determine the final extraction results.

2.3.3. Data Analysis

First, the frequency of landscape elements and CES perceptions in the photographs provided by each volunteer were statistically analyzed using SPSS 28.0. Box plots were employed to visually compare these frequencies between the two sex subgroups. Second, taking the CESs, mountain elements, and number of occurrences of these landscape elements in the photographs as the three variables, correspondence analysis (CA) was conducted to elucidate the relationship between the CESs and landscape elements. The Chi-square test of the four-fold table data was used to analyze categorical variables, including the differences in socio-demographic variables and the frequency of CESs perceived between the two groups. Results with a p-value < 0.05 were regarded as statistically significant [42,43]. Finally, Sankey diagrams were used to visualize the distribution and transfer of resources, representing quantities with varying widths of flowing lines, allowing observers to intuitively understand the complex system dynamics.

3. Results

3.1. Demographic Profile of Volunteers

Notable differences were observed between the male and female subgroups in terms of the number of visits, family address, number of stops (>60), total time consumption, and average speed (

Table 3. Socio-demographic characteristics of the 90 study volunteers.

Categories	Responses	Male	Percentage (%)	Female	Percentage (%)
Travel companions	Alone	9	20	7	16
	Accompanied	36	80	38	84
Number of visits	First-time	32	72	20	44
	Non-first time	13	28	25	56
Family address	Chongqing	7	16	16	36
	Non-Chongqing	38	84	29	64
Number of stops	>60	4	26	11	74
	≤60	41	46	34	54
Total time consumption (min)	>30	18	40	29	64
	≤30	27	60	11	36
Average speed (km/h)	>2	33	73	19	42
	≤2	12	27	26	58

). Approximately 25% of the volunteers were Chongqing residents, and over 80% were accompanied. However, a higher proportion of first-time visits were observed in the male subgroup than in the female subgroup. Females tended to make more stops, especially >60, and spent more total time, with a significantly lower average speed than males (Table 3).

3.2. Photography Dataset Interpretation

To explore the perceptual differences between mountainous landscapes and non-mountainous landscapes, we divided the 16 landscape elements into 5 landscape elements with mountainous features and 11 without mountainous features, providing corresponding schematic diagrams. Landscape elements with mountainous features refer to those that are distinctive or unique to mountainous areas, while non-mountain elements are those characteristic of plains or found in both environments. Therefore, non-mountain elements may also appear in mountainous regions. All the images were captured by volunteers who were solicited for this purpose (Figure 3).

Significant differences were observed in the frequency of landscape elements captured in photographs between the male and female subgroups (Figure 4). In the male subgroup, the least perceived elements were air-raid shelters and animals, whereas the most perceived were shrubs and trees, followed by flowers and lawns, garden facilities, and roads. The male subgroup had dispersed perceptions of fort-like ridges and cliffs, flowers and lawns, and squares and platforms, with more uniform perceptions of other elements. Overall, fort-like ridges and cliffs, ponds, animals, garden facilities, and squares and platforms had skewed distributions, whereas the other elements were normally distributed. In the female subgroup, the most perceived elements were overlooks and shrubs and trees, followed by roads and flowers and lawns, whereas the least perceived were air-raid shelters, tree roots, animals, and modern architecture. The female subgroup had more dispersed perceptions of fort-like ridges and cliffs, overlooks, squares and platforms, and people, with more uniform perceptions of other elements. Tree roots, squares, and platforms had normal distributions, whereas the distribution of other elements was skewed. Regarding mountain-specific elements, females perceived overlooks and fort-like ridges and cliffs more frequently than males, with overlooks reaching a maximum frequency of approximately 0.175 compared to the 0.025 observed by males. In terms of non-mountain elements, females had higher frequencies for shrubs and trees, ponds, roads, and people, whereas other elements were less frequently perceived than those for males (Figure 4).

Significant differences were observed in the evaluation of CES values between male and female subgroups, with females generally having higher frequencies of perceptions across various CES values (Figure 5). For the male subgroup, the highest perceived value was aesthetics, followed by recreation, inspiration, cultural heritage, science and ecological education, health, social relations, sense of place, and future values, with the lowest being cultural diversity and spiritual and religious values. Perceptions of cultural diversity, spiritual and religious, and future values were uniform, whereas that of social relations was the most dispersed. Overall, recreation value showed a normal distribution, whereas the distribution of the other values was skewed. In the female subgroup, almost all volunteers perceived aesthetic value, which also had the highest frequency, followed by recreation, social relations, science and ecological education, sense of place, health, cultural heritage, inspiration, spiritual and religious, and cultural diversity values, with future value being the least perceived. Perceptions of cultural diversity, spiritual and religious, and health values were uniform, whereas that of aesthetic value was the most dispersed. There was a skewed distribution among the female groups (Figure 5).

The value points and path KML files generated using “2bulu 7.5.5” were exported and merged in ArcGIS 10.8 to create a GPS point map (Figure 6). The Kernel Density Tools in ArcGIS 10.8 were used to perform kernel density estimation of the value points (Figure 6). The GPS point map and kernel density estimation map initially reflected the spatial distribution and clustering differences of the value points. The female group had a higher point density at the park entrance than the male group, lower density near Kansheng Tower, and higher density near the lotus pond. Minimal differences were observed in the overall routes between sexes; however, both groups repeatedly passed through the roads in the southern area near the park entrance with minimal stops.

3.3. Spatial Arrangement of CESs and Subgroup Disparities

The average nearest neighbor analysis, integrated within QGIS 3.26, was used to effectively quantify the clustering degree of the value points. When the average nearest neighbor ratio (R) was less than 1, with a significantly large Z-score, the cultural service type exhibited an ‘aggregation pattern’. In this study, all CESs demonstrated this pattern. Moreover, all AUCs exceeded 0.7, with some AUCs exceeding 0.9, indicating high reliability in the model evaluation (

Table 4. Average nearest neighbor statistics (R-value and Z-score) and performance of the SolVES model.

CES Values	R-Value		Z-Score		AUC (Test)		MVI
	Male	Female	Male	Female	Male	Female	Male	Female
Cultural diversity	0.63	0.59	−7.07	−12.82	0.88 (0.72)	0.88 (0.75)	3	3
Spiritual and religious	0.46	0.43	−10.62	−9.06	0.92 (0.77)	0.90 (0.78)	4	4
Science and ecological education	0.45	0.57	−16.08	−11.99	0.90 (0.78)	0.90 (0.76)	7	6
Inspiration	0.27	0.33	−22.01	−17.51	0.95 (0.88)	0.88 (0.70)	7	5
Aesthetic	0.42	0.44	−20.45	−25.01	0.91 (0.80)	0.90 (0.76)	10	10
Social relations	0.29	0.34	−21.59	−20.04	0.94 (0.88)	0.90 (0.72)	6	6
Sense of place	0.45	0.66	−14.82	−9.63	0.93 (0.83)	0.88 (0.79)	5	5
Cultural heritage	0.46	0.39	−16.66	−13.44	0.90 (0.81)	0.89 (0.77)	6	7
Health	0.49	0.60	−14.39	−11.13	0.89 (0.75)	0.89 (0.70)	4	5
Future	0.62	0.59	−9.59	−8.50	0.89 (0.72)	0.87 (0.77)	5	4
Recreation	0.27	0.34	−23.76	−12.72	0.91 (0.78)	0.90 (0.77)	9	6

CESs, cultural ecosystem services, AUC, area under the curve; MVI, maximum value index.

).

Three representative environmental metrics were selected to analyze the AUC results: distance to the tower (DTT), distance to the cliff (DTCL), and distance to water bodies (DTW) [44]. The response curves illustrate the relationships between the environmental metrics (DTT, DTCL, and DTW) and the 11 selected CES types. For males, the value index (VI) for inspiration and health values increased with DTT (Figure 7a). Conversely, for females, the VI for cultural diversity, spiritual and religious, inspiration, and health values sharply increased with DTT (Figure 7b). For males, the VI for cultural heritage, social relations, and recreation values decreased with increasing DTCL (Figure 7c). Conversely, for females, significant VI responses were observed for aesthetics, social relations, and health values (Figure 7d). Owing to the small water area in Eling Park, VI responses to DTW were somewhat random (Figure 7e), with females exhibiting a higher affinity for water bodies than males (Figure 7f).

The spatial distribution of CESs reflects the values at specific locations, represented by 1–10 value gradients. The highest value in the figures indicates the maximum value index (MVI) for each CES (Table 4). This index evaluates the maximum CES value by considering the aggregation of value points, subjective evaluations of volunteers, and the number of value points. Males had a higher MVI for science and ecological education, inspiration, future, and recreation values, whereas females had a higher MVI for cultural heritage and health values. Both sexes showed a similar MVI for cultural diversity, spiritual and religious, aesthetic, social relations, and sense of place.

Figure 8 shows the spatial distribution of the 11 CESs perceived by the two sex groups, with colors ranging from blue to red indicating increasing values. Generally, the CES distribution followed an experimental route, with high-value areas at the park entrance, Kansheng Tower, cliffside path, central lawn, and lotus pond. Lesser-value areas were located along the lesser-travelled paths and pavilions. Sex-based differences significantly affected the spatial distribution of CESs. Both groups showed high and widely distributed aesthetic values, but low cultural diversity, future, and spiritual and religious values. Similar MVI and spatial patterns were observed for aesthetic and cultural heritage values across both groups. Key spatial nodes, such as the central lawn, Kansheng Tower, cliffside path, and park entrance, were associated with high aesthetic, recreation, and sense of place values. Monuments, such as the Anti-Japanese War Monument, Eling Monument, and Sino-Soviet Monument, were linked to cultural heritage value. The female subgroup perceived social relations at the lotus pond and Kansheng Tower, whereas the male subgroup did not show significant perceptions at these places. The male subgroup perceived higher health value at Kansheng Tower, whereas the female subgroup had a significantly higher perception of social relations value at Kansheng Tower, the central lawn, and the park entrance. Males had a higher perception of inspiration value at Kansheng Tower, the park entrance, and the cliffside path, whereas females displayed a more dispersed perception of inspiration value. Overall, both groups had similar perceptions of the cultural diversity value. However, males perceived a higher cultural diversity value at the entrance, on uphill paths, and near Kansheng Tower, whereas females had higher perceptions at Kansheng Tower, the lotus pond, and the entrance. At the lotus pond, females had higher perceptions of spiritual, religious, and social relations values than males. Females also perceived a significantly higher sense of place value at Kansheng Tower, the cliffside path, the central lawn, and the lotus pond. Males exhibited a broader perception range for future and recreation values than females. However, females showed a broader perception range for cultural diversity, spiritual and religious, and sense of place values. The various CESs at the cliffside path had positive impacts, with notable values for aesthetic, recreation, future, and health values. As volunteers returned to the entrance along the downhill path, their perceptions of the various CESs tended to decline.

3.4. Association Between CESs, Landscape Elements, and Differences Between Subgroups

In both the male and female subgroups, the CESs and park landscape elements in this analysis were not independent and showed significant relationships (x² = 17,081.612, p < 0.001; x² = 121,178.325, p < 0.001). The CA biplots showed relationship patterns between these variables (Figure 9). In the male group, aesthetic, scientific, and ecological education values were associated with flowers and lawns, overlooks, and shrubs and trees. Recreation and health values were related to overflows and shrubs and trees; future value was closely related to tree roots. Inspiration value was closely related to fort-like ridges and cliffs. Social relations value was associated with squares and platforms. Cultural heritage value was closely linked to people; cultural diversity value was linked to recreation facilities. In the female group, the social relations value tended to be associated with modern architecture. Recreation and health values were closely linked to shrubs and trees; cultural heritage value was closely related to road and garden facilities. Inspiration value was closely linked to slopes and steps, as well as animals and tree roots; future value was related to tree roots and overlooks. Science and ecological education value was associated with flowers and lawns; cultural heritage value was linked to antique architecture. Additionally, air-raid shelters in the male and female biplots, as well as animals in the male biplots, were distant from other CESs, whereas roads in the male biplot were close to the origin, suggesting that these landscape elements were not significantly linked to any CESs.

To examine the significance of park landscape elements in delivering CESs, Sankey diagrams were utilized to analyze and visualize the flow relationships and proportional distribution between landscape elements and CESs, with the width of the connecting lines reflecting the flow magnitude [45]. The three CESs that exhibited comparable spatial distribution patterns also had similar landscape composition elements in both subgroups; both the male and female subgroups perceived aesthetic value from shrubs and trees, flowers and lawns, and roads, with health and recreation value perceived mainly from shrubs and trees. The remaining eight CESs (cultural diversity, spiritual and religious, science and ecological education, inspiration, social relations, sense of place, cultural heritage, and future) exhibited significant differences in spatial distribution, with considerable variation in landscape composition between the groups. For instance, in perceiving inspiration value, the male subgroup emphasized shrubs and trees, whereas the female subgroup emphasized roads, flowers, and lawns, and overlooks. Similarly, for cultural heritage value, the male subgroup prioritized recreation facilities, squares and platforms, and garden facilities, whereas the female subgroup prioritized roads and overlooks. Notably, for mountain elements, females generally valued overlooks more than males, perceiving a range of CESs, such as science and ecological education, aesthetic, and cultural heritage values (Figure 10).

4. Discussion

4.1. Landscape Environmental Value Orientation

4.1.1. Sex Differences in Landscape Element Preferences

In terms of perceived landscape elements, both the male and female subgroups demonstrated perceptions of shrubs and trees, flowers and lawns, and roads, which align with the most abundant landscape features in Eling Park. For males, the preferred mountain landscape elements were slopes and steps, whereas the preferred non-mountain elements were garden and recreation facilities. These landscape preferences may be linked to males’ general tendency towards physical activity, their socializing needs through sports, and their emphasis on the practicality and convenience of the park [46,47,48]. For females, the preferred mountain element was the overlook, and the preferred non-mountain elements were ponds, squares, platforms. These preferences may be linked to females’ greater focus on mental health and emotional regulation, their tendency to build and maintain social connections through conversation and sharing, and their heightened sensitivity to aesthetics [49,50,51].

4.1.2. Sex Differences in CES Value Orientation

Both the male and female subgroups clearly perceived the aesthetic, recreation, cultural heritage, and social relations values of the park. This aligns with Eling Park’s primary functions of providing natural and cultural landscapes, as well as serving as a public space for leisure, fitness, and community interaction. This conclusion differs from that of previous studies, such as those by Kicic et al. [26] on park forests and Dou et al. [52] on the National Wetland Park, likely owing to differences in environmental characteristics and the surveyed population. For the male subgroup, the most highly perceived CESs were recreation and aesthetic values, whereas the least perceived were cultural diversity and spiritual and religious values, consistent with the findings of Zhou et al. [40] and Swapan et al. [34]. Males typically engage in activities such as sports, running, and family gatherings in community parks, which offer relaxation and social opportunities. In contrast, males might be less attuned to cultural diversity and spiritual and religious values, which involve deeper cultural and spiritual experiences that may not be as immediately engaging as recreation and aesthetics. For the female subgroup, the most highly perceived value was aesthetics while the least perceived was future value. Females often have a keener sense of beauty and seek relaxation and joy through aesthetic experiences during leisure, rendering them more likely to prefer spaces that offer high aesthetic value over spaces that offer abstract future value during visits to parks [53].

4.1.3. Association Logic Between Landscape Elements and CESs

Analysis of the relationship between landscape features and CESs indicated that both males and females recognized recreation, aesthetic, and social relations values in Eling Park (Figure 8 and Figure 9). In the male subgroup, inspiration was associated with landscape elements, such as fort-like ridges and cliffs, which may be linked to their need for physical activity, their spatial awareness, and the visual impact of these features. Consistent with the findings of Swapan et al. [34], the males frequently perceived recreation. Additionally, males typically visit parks for physical exercise, challenging activities, and social interactions [54], leading to the association between social relations and squares and platforms. CA analysis revealed that air-raid shelters and ponds were less associated with other CESs for males, likely owing to their infrequent appearance. Females generally prefer aesthetic and artistic enjoyment, comfort, security, and micro landscape features [55], as evidenced by the relationship between recreation and shrubs and trees. Science and ecological education value was associated with flowers and lawns, indicating that these elements can stimulate enthusiasm for ecological education in females [56]. Inspiration was often associated with artistic tree roots exposed on the surface, animals, and slopes and steps, confirming that females focus more on detailed aesthetic value than men [56]. Garden facilities are linked to cultural heritage value, reflecting their role in prompting thoughts on cultural heritage [57].

4.1.4. Influencing Factors of Differences

Preferences for CESs in mountainous versus plain cities can be attributed to factors such as geographic features, socio-cultural background, personal experience, and psychological perceptions. The steep terrain, rich vegetation, and unique landscapes of mountainous areas are more likely than the plains area to evoke inspiration, aesthetics, and cultural diversity values [58]. Public preferences for CESs are influenced by the interplay of natural and infrastructural elements with cultural and social backgrounds, particularly in regions with distinct cultural histories, such as Chongqing [59]. Additionally, personal experiences and psychological needs further shape individual landscape perceptions. Individuals who frequently hike in mountainous areas often prioritize inspiration and aesthetic values in natural landscapes [57], whereas urban dwellers focus on the wellness and social functions of plain parks [60]. Mountainous environments cater to the psychological needs of adventure and discovery, whereas plains environments fulfil the need for relaxation and interaction [61]. These preferences reflect the diverse demands and expectations of visitors for both natural and cultural landscape across different geographic environments and sexes.

4.2. Spatial Distribution Pattern of CESs

Each CES exhibited high- and low-value points across various locations, corresponding to specific landscape areas. For instance, the Anti-Japanese War Monument and Eling Monument were associated with cultural heritage value; the flowerbeds in front of Kansheng Tower were linked to aesthetic value; the lotus pond was associated with inspiration; multiple squares were linked to social relations; and the entrance square and Kansheng Tower were associated with a sense of place. Kansheng Tower, the entrance square, the lotus pond, and the cliffside path were favored by both subgroups owing to their accessibility and landmark features. Kansheng Tower, located at the highest point of Eling Park, offers views of the Jialing River and Yuzhong Peninsula from its upper levels. This location embodies a composite of aesthetic, inspiration, social relations, and spiritual and religious values. The distribution of aesthetic, recreation, and inspiration values at the park entrance corresponds with the initial sense of novelty upon arrival.

Regarding the spatial perception of a sense of place within mountainous city parks, the female subgroup also demonstrated a higher perception of social relations value than the male subgroup, particularly at key space nodes, such as the entrance plaza, lotus pond, and cliffside path. This difference in social relations perception suggests that females show a greater need for social interactions and the temporary connections that mountainous parks provide. However, males exhibited greater sensitivity to science and ecological education and recreation values than that of females, which was evident at locations such as the central lawn, lotus pond, and Kansheng Tower. The values associated with science and ecological education and recreation are relatively more “rational” and “straightforward” than the sense of place, social relations, and inspiration values, which reflects the tendency of females to have more emotion-driven needs and males to have more rational exploratory needs [62,63]. Additionally, not all CESs showed spatial distribution differences between the two subgroups; aesthetic and cultural heritage values displayed similar spatial distribution patterns, concentrated in areas such as Kansheng Tower and the Anti-Japanese War Monument. For inspiration, health, spiritual and religious, social relations, cultural heritage, and future values, differences in overall sensitivity were observed between the subgroups; however, specific spatial nodes, such as the entrance plaza, Kansheng Tower, and cliffside path, showed a certain degree of consistency in perception.

Analysis of the effect of mountainous features on CESs revealed that volunteers had a higher CES perception for the uphill path than for the downhill path. As volunteers moved from the park entrance to Kansheng Tower, the increasing elevation and denser distribution of landscape nodes along the uphill path resulted in a higher concentration of aesthetic, inspiration, recreation, and science and ecological education values. Conversely, during the return journey along the cliffside path to the entrance, both subgroups experienced lower perceptions of CES values. The uphill terrain significantly affected perceptions of sense of place, cultural heritage, and aesthetic values. The unique topographical features of the mountainous park provided a distinctive sense of place, contrasting with the monotonous perspectives of flatland parks, enriching the aesthetic experiences of volunteers. Elevated areas, often housing commemorative buildings, enhanced the cultural heritage value of the park. However, the terrain also posed challenges to the overall CES perception. The increased physical exertion and time investment required for mountainous park exploration, compared to those required for flatland parks of equivalent areas, led to a faster decline in the perceptual sensitivity of the volunteers on the downhill path as they approached the entrance.

4.3. Research Limitations

In the social research method used in this study, VEP can better represent the participants’ perception of the connection between landscape and CESs in the park; a large number of photos render landscape elements concrete. The questionnaire accurately collected the basic information of 90 volunteers. The semi-structured interview had a significant advantage in reflecting the subjective perceptions of the experimenter. The three methods combined to meet the needs of the experiment [21].

However, in the process of the volunteer experiment, there were some problems, such as a small sample size (90) and insufficient consideration of the group effect, among others. In the discussion of the factors affecting CESs, we did not account for the impact of whether they went alone or whether they went to the park for the first time on their perception. Although the age group and educational background of the population were controlled, subjectivity among volunteers can introduce bias, a common challenge in CES research [34,35,36]. Despite the importance of CESs to many residents, this category remains the least explored among the four ecosystem service categories (provisioning, regulating, cultural services, and supporting services), particularly in mountainous regions. Utilizing the VEP method can improve the accuracy of the results by increasing subgroup sample sizes. However, this approach relies heavily on immediate feedback, complicating the comprehensive understanding of the broader perspectives of volunteers.

The SolVES model addresses the previous limitations of CES research by integrating public perception with real-world scenarios and specific environmental indicators, thereby enhancing the differentiation of CESs across various regions [9]. However, the model presents notable limitations in analyzing small areas, mountainous terrains, questionnaire design flexibility, and the accuracy of subjective evaluations. First, the accuracy of SolVES is significantly affected by the land use type distribution. In this study, the small park area required manually drawn land use maps, differing from those used in land planning, which may have reduced the overall precision. Second, although SolVES requires terrain elevation data, it relies solely on Euclidean distance tools that consider linear distances without vertical coordinates, limiting the ability of the model to factor in terrain elevation. This limitation necessitates manual map interpretation to assess the effect of mountainous terrain on CES distribution. Third, when evaluating multiple CESs, volunteers may experience confusion when categorizing the types of CES for each photograph, which can affect the accuracy of the results. Finally, the SolVES model employs spatial distance to illustrate the connection between landscape features and CESs, which can introduce bias. For example, photographs captured near cliffs might focus on distant views or vegetation rather than on the cliff itself, potentially diluting relevant CES data.

Combining photography, interviews, and observations is crucial for exploring public perceptions of mountain landscapes and understanding attitudes toward nature. Current CESs research frameworks do not adequately address sex variables. Furthermore, the classification of CESs may have issues of incompleteness and potential bias. Informing participants of the definitions in advance might influence their perception of the landscape. Future studies should further investigate the effect of sex on perception, particularly in mountainous landscapes. The complexity of study subjects, methodological accuracy, and comprehensiveness of research frameworks are essential for understanding CES perception. Additionally, integrating artificial intelligence and machine learning with multimodal data, such as social media and ergonomics experimental data, could enhance the exploration of differences in CES perception in mountainous versus non-mountainous city landscapes and across sexes [64]. In future research, we will increase the sample size and focus on diversity, including people of different ages, education levels, and health conditions. We will use stratified or quota sampling to better understand how different groups perceive the value of cultural ecosystem services (CESs) in mountain parks. To reduce bias in CES self-reports, we will combine lab experiments with field studies to more accurately measure participants’ perceptions [65]. Additionally, urbanization and climate change can significantly affect how people perceive CESs. Extreme weather, like heatwaves or droughts, may change how residents value green spaces [11]. In Chongqing’s mountainous area, hot summers and cold winters could make these effects stronger. Future studies will examine how seasonal changes and long-term climate shifts influence people’s preferences, such as the need for shade in summer or sunlight in winter. These improvements will make the findings more reliable and provide useful insights for better park planning.

4.4. Implications for Landscape Design and Management

4.4.1. Enhancing CES-Oriented Landscape Elements

Mountainous terrains should be leveraged to improve inspiration, aesthetic, and cultural heritage values. Viewing platforms with seating should be integrated along cliff trails to enhance scenic appreciation, while historical features, such as air-raid shelters and fortress-like ridges, should be preserved and equipped with interpretive signage to strengthen cultural heritage value.

4.4.2. Balancing CES Supply and Demand

Spatial analysis of CES values should guide targeted interventions. Males, who prioritize recreation, would benefit from the addition of climbing walls and fitness zones along steep paths. Females, who value social interaction and cultural heritage, require more shaded seating, garden spaces, and semi-private pavilions near key social hubs like lotus ponds and overlooks.

4.4.3. Establishing Sex-Sensitive Development Zones

By mapping CES perception hotspots, planners can designate areas that cater to specific needs. High-traffic areas should integrate features that promote both social engagement and recreation, ensuring an inclusive environment. Meditation zones and cultural festivals can enhance spiritual and religious values, particularly in spaces where such CESs are currently underrepresented.

4.4.4. Incorporating CES Values into Sustainable Development Goal

Embed CESs into SDG frameworks by linking health-centric landscapes (Goal 3), equitable greenway networks (Goal 11), and community co-design workshops to ensure inclusive access. Strengthening connectivity between Eling Park and other green spaces through ecological corridors will enhance CES accessibility. Promote community participation by regularly assessing resident needs and dynamically adjusting landscape configurations to align CESs supply with urban sustainability goals.

5. Conclusions

Sex is a crucial factor in the spatial design of urban parks. However, differences in the perceptions of CESs between males and females are often overlooked in the landscape planning of mountain urban parks. This oversight can limit the activities and social participation of different sexes in mountainous parks, thereby affecting their perceptions and experiences. This study primarily explored the distribution characteristics of CESs in an urban, mountainous park based on sex differences and their associations with landscape elements. By collecting and analyzing data from both sex groups, spatial distribution maps of CESs were constructed, which revealed potential differences in the association between landscape elements and CESs.

Males and females show distinct preferences in landscapes and CES perception, reflecting their different needs. Male favor slopes and recreational facilities, emphasizing physical activity and practicality, while females prefer viewing platforms and ponds, highlighting their need for emotional comfort and social interaction. Males had lower perceptions of cultural diversity and spiritual and religious values, whereas females had lower perceptions of future value. Regarding the space nodes in Eling Park, areas such as the central lawn, Kansheng Tower, and the park entrance were noted for high aesthetic, recreation, and sense of place values, with the cliffside path positively influencing various values. Females were more likely to perceive social relations value at space nodes such as the lotus pond and Kansheng Tower, whereas males were more likely to perceive inspiration value at Kansheng Tower, the park entrance, and the cliffside path. Regarding the relationship between landscape elements and CESs, for males, inspiration value was closely related to fort-like ridges and cliffs, whereas the social relations value was closely related to squares and platforms. For females, recreation value was closely related to shrubs and trees; science and ecological education value was closely related to flowers and lawns; inspiration value was related to tree roots, animals, and slopes and steps; and cultural heritage value was related to garden facilities. These differences shape how individuals use park spaces and affect the delivery of intangible services like cultural heritage and ecological education. By optimizing mountain-specific features—such as adding viewing platforms or integrating cultural relics—parks can better meet the needs of different sexes, improving CESs delivery.

Sex-sensitive park design supports urban sustainability across ecological, social, and economic dimensions. Ecologically, integrating CES evaluation into urban green space planning and building culturally focused ecological corridors can help mountainous parks and other green spaces work together, forming a sustainable ecological network. Socially, sex-sensitive park optimization can foster inclusive urban development. For example, creating sex-specific zones—like climbing areas for men and semi-private social spaces for females—breaks traditional stereotypes and combines functionality with inclusive design. This approach enhances spatial fairness and strengthens community ties by increasing residents’ connection to nature. Economically, it enhances park appeal and efficiency, stimulating local green economies. These measures align with the UN Sustainable Development Goals on well-being, sex equality, and inclusive spaces, providing practical solutions for balancing human, natural, and social needs in rapidly urbanizing mountainous cities.