Next Article in Journal
Identifying Optimal Summer Microclimate for Conifer Seedlings in a Postfire Environment
Previous Article in Journal
Predicting Optimal Sites for Ecosystem Restoration and Assisted Migration of Abies pinsapo Boiss. Using Species Distribution Modelling
Previous Article in Special Issue
Effect of Forest Greening on Carbonate Rock Weathering Carbon Sink in the Subtropical Humid Zone
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Forests
Volume 16
Issue 12
10.3390/f16121804
forests-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Landscape Aesthetics Quality in Subalpine Forests of Eastern Tibetan Plateau Will Greatly Decrease by the End of the Century?

by
Junyan Liu
1,2,3,
Jie Du
3,
Chenfeng Zhang
4,
Benedicte Bachelot
5,
Yiling Yang
4,
Tingfa Dong
4 and
Yan Wu
1,*
1
Chengdu Institute of Biology, Chinese Academy of Sciences, Chengdu 610041, China
2
Yunnan Urban Agricultural Engineering and Technological Research Center, College of Agronomy, Kunming University, Kunming 650214, China
3
Jiuzhaigou Nature Reserve Administrative Bureau, Jiuzhaigou 623402, China
4
State Key Laboratory of Vegetation Structure, Functions and Construction (Veglab), Ministry of Education Key Laboratory for Transboundary Ecosecurity of Southwest China, Yunnan Key Laboratory of Biological Adaptation, Conservation and Utilization, Institute of Biodiversity, School of Ecology and Environmental Science, Yunnan University, Kunming 650500, China
5
Department of Plant Biology, Ecology, and Evolution, Oklahoma State University, Stillwater, OK 74078-3013, USA
*
Author to whom correspondence should be addressed.
Forests 2025, 16(12), 1804; https://doi.org/10.3390/f16121804
Submission received: 16 October 2025 / Revised: 17 November 2025 / Accepted: 20 November 2025 / Published: 30 November 2025
(This article belongs to the Special Issue Social and Cultural Benefits of Forests Contributing to Human Health and Well-Being)
Browse Figures
Versions Notes

Abstract

Landscape aesthetic quality (LAQ) is a vital cultural ecosystem service in global forests, particularly in the subalpine forests across the Tibetan Plateau, which are considered popular tourist destinations due to their unique cultural services. However, the explicit spatial localization and spatial–temporal dynamics of LAQ in subalpine forests in the Tibetan Plateau remain largely unexplored. Herein, we introduced a method for assessing LAQ that integrates the species’ biophysical attributes with spatial landscape characteristics, allowing for a spatially explicit quantification of LAQ. We further employ this approach to project changes in LAQ under forest landscape dynamics (2016–2096) in Jiuzhaigou, eastern Tibetan Plateau. Most regions exhibited moderate or low LAQ, with high values ible, while over half of low-LAQ regions were not. The high-value zone of LAQ is projected to rise slightly by 2056 but decline sharply by 2096. These results reveal strong spatial heterogeneity in LAQ and indicate that future landscape dynamics will substantially reshape its distribution in the subalpine forests of the eastern Tibetan Plateau. Our findings provide early evidence of declining cultural ecosystem quality in subalpine forests and offer guidance for adaptive management in similar mountain ecosystems worldwide.

1. Introduction

Landscape aesthetic quality (LAQ) was highlighted in the Millennium Ecosystem Assessment in 2005 as a crucial ecosystem service valued by humans but seldom fully integrated into assessments following the ecosystem service framework [1,2,3]. However, the direct and intuitive appreciation of LAQ has played a significant role in garnering public support for ecosystem conservation [4,5,6]. Importantly, LAQ also drives tourist and recreational visitation to natural areas, often substantially impacting local economies [5,7,8]. Therefore, LAQ should be strongly considered in landscape planning and decision-making [3,9]. Nevertheless, LAQ is seldom quantified and mapped despite the increasing trend of LAQ research [8,10].
The quantitative evaluation of LAQ is often challenging due to its dependence on biophysical attributes and human perception and cognition [8,9,11,12,13]. Although LAQ studies have expanded across various disciplines such as psychology, anthropology, evolutionary biology, and landscape planning [14], there is still a lack of well-developed quantitative and standardized assessment approaches [15,16,17]. The approaches from “expert” and “perception-based” to translating people’s preferences into the biophysical attributes used for expert judgments [13,16]. Various indicators, such as landscape naturalness, diversity, and uniqueness, measured using GIS tools, have been utilized to assess LAQ at regional [4,18] and national scales [19,20]. However, using these indicators for mapping and provisioning LAQ might not accurately reflect the states of different landscapes at finer scales. In forest landscapes, using indicators derived from widely available geospatial data, especially land use/land cover categories, may result in similarities between different types of forest communities, affecting the precision of the assessment [21]. Research on “near-views” within forest landscapes has consistently demonstrated that species’ biophysical attributes, including trunk forms, leaf colors, and leaf density, have the most significant impact on aesthetic judgments [22]. Therefore, developing indicators linking species’ biophysical attributes to aesthetic preferences is crucial for assessing forest LAQ.
Understanding the spatial–temporal variation in LAQ is beneficial for forest management [23]. A spatially explicit map and temporal dynamics of forest LAQ could help identify areas with high aesthetic quality and protection values, contributing to establishing and adjusting forest policies and plans [24]. The distribution of forest LAQ is generally heterogeneous in space [25] due to spatial variation in forest landscapes resulting from species distribution patterns responding to local environmental factors (e.g., soil types, topographic conditions, and disturbance history) [26]. Furthermore, forest LAQ may shift over time due to forest succession [27]. Despite the extensive research on forest LAQ, little work has been done to use geospatial analysis to assess spatial–temporal changes in LAQ resulting from changes in forest landscape dynamics.
Subalpine forests in the eastern Tibetan Plateau are crucial in providing ecological services for China and Asia, including biodiversity, soil conservation, and water provision [28,29]. Additionally, these forests are popular destinations for mass tourism and ecotourism [30,31]. However, as one of the most sensitive ecosystems to climate change, subalpine forests across the Tibetan Plateau will undergo significant spatial–temporal shifts in forest communities at the landscape scale in the future [32]. Consequently, the ecosystem services provided by subalpine forests, including LAQ, will likely change with forest landscape dynamics [24]. Understanding the spatial–temporal changes in LAQ under future forest landscape dynamics is benefit for local administrators to promptly adjust their forest management strategies. However, previous characterization of LAQ in the subalpine forests has relied on static data sources and lacks spatially explicit information [30]. To date, explicit spatial localization of LAQ and their spatial–temporal dynamics from subalpine forests across the Tibetan Plateau remains unknown.
Jiuzhaigou World Natural Heritage Site (hereafter Jiuzhaigou), located in subalpine regions on the eastern Tibetan Plateau, is one of the most popular tourist attractions in China due to its exceptional natural beauty and aesthetics. Jiuzhaigou is renowned for its autumn color-leaved sightseeing forest landscapes, primarily due to the widespread presence of “color forests” (i.e., deciduous broad-leaved forests and redwood forests in our study area), which exhibit vibrant autumn colors that can evoke positive emotional responses in individuals [32,33]. In contrast, evergreen coniferous forests like spruce-fir forests, which are prevalent in Jiuzhaigou, lack autumn colors, resulting in lower aesthetic appeal. However, the forest landscapes in this region are projected to change due to forest succession [32], along with variations in LAQ among different forest types, leading to spatial–temporal fluctuations in LAQ over the next few decades. Understanding the specific spatial distribution of LAQ in Jiuzhaigou and predicting their spatial–temporal dynamics could significantly aid in developing landscape strategies. However, to our knowledge, there have been no previous attempts to map and predict the spatial–temporal dynamics of LAQ in this region.
In this study, we developed an innovative methodology to quantify and map LAQ in Jiuzhaigou based on the relationship between species’ biophysical attributes and aesthetic preferences. Subsequently, we conducted a viewshed analysis of LAQ spatial characteristics to better assess its accessibility potential, aiding decision-making by providing information on visual potential and incidence [4]. Finally, we predicted the spatial–temporal variations in LAQ over the next 80 years under future forest landscape dynamics, guided by the following questions: (1) What are the spatial distribution patterns of LAQ across the entire area and within viewshed areas, respectively? (2) How will projected LAQ change spatially and temporally under future forest landscape dynamics? These results can fill this knowledge gap by undertaking a comprehensive spatial mapping of LAQ and forecasting its spatiotemporal dynamics within subalpine forests.

2. Materials and Methods

2.1. Study Area

Jiuzhaigou National Nature Reserve is located on the eastern edge of the Tibetan Plateau (32°53′–33°20′ N, 103°46′–104°05′ E; Figure 1). The elevation ranges from 2000 to 4880 m above sea level, with an annual mean temperature of 7.3 °C and annual mean precipitation of 760 mm [32]. The region is renowned for its forests with colorful autumn leaves, high biodiversity, and abundant tourism resources. The vegetation is dominated by eight tree species, including four evergreen coniferous species (Picea asperata (Mast.), Pinus tabuliformis (Carrière), Sabina saltuaria (Rehder & E. H. Wilson), and Abies faxoniana (Rehder & E. H. Wilson)), and other deciduous tree species (Populus davidiana (Dode), Betula platyphylla (Sukaczev), Larix mastersiana (Rehder & E. H. Wilson), and Quercus liaotungensis (Koidz.), also known as autumn color-leaved species) [32] (Table S1). The soil types in the study area exhibit notable vertical variation, primarily comprising cinnamomic (<2200 m), brunisolic (2200–2700 m), dark brown (2700–3200 m), podzolic (3200–3800 m) and meadow (>3800 m) from low to high altitudes [32].

2.2. Approach Used for Assessing LAQ

Following Casado-Arzuaga et al. [4], we computed the LAQ in Jiuzhaigou by considering three elements: the aesthetic quality of species, namely the perceived value of different tree species in our study, and two landscape characteristics (landscape diversity and relief) (Figure 2). Detailed explanations and calculations for each element are provided in the subsequent sections. Ultimately, utilizing the weights assigned to the different elements mentioned in Casado-Arzuaga et al. [4], we derived the LAQ as the weighted sum of the three elements (Table S2).

2.2.1. Species Aesthetic Quality

(a)
Determinants and indicators
To identify indicators for evaluating the artistic quality of species, we initially conducted a literature review to explore the factors influencing artistic quality at the species level and the commonly used indicators to measure them. The literature reviewed mainly focused on assessments of artistic quality (additional reviews are provided by Felton et al. [34], but we also incorporated studies based on the evaluation of scenic beauty, which presented methods for forest landscape aesthetics assessments [28,30]. Tveit et al. [35] provided a comprehensive review of visual concepts, encompassing the key factors influencing aesthetic quality, along with their corresponding references. We utilized this framework as the foundation for selecting and subsequently developing the indicators for implementation. Specifically, the selected indicators align with the widely recognized perceptions and preferences related to the aesthetic quality of species, are based on the biophysical features of species, and utilize available and eligible data. Finally, we selected a set of indicators that apply at the species level to develop a formal method for assessing the aesthetic quality of species in Jiuzhaigou (see Table 1).
(b)
Data collection
Three indicators, encompassing ten sub-indicators, were chosen to assess species aesthetic quality (Table 1 and Table S3). Four quantitative species attributes—discoloration period, crown width, leaf area, and leaf density—were derived from prior research [36,37]. We hypothesize that these quantitative attributes positively impact species’ aesthetic value. Color elements—hue, saturation, and brightness—were quantified from images using a software-based quantification method [28]. Specifically, 139 images of the eight dominant tree species were collected from the Jiuzhaigou National Nature Reserve Administration (https://www.jiuzhai.com/ (accessed on 20 October 2020)) and the Plant Photo Bank of China (http://ppbc.iplant.cn/ (accessed on 23 October 2020)). Among these, 14 images were captured for each evergreen coniferous species (i.e., P. asperata, P. tabuliformis, and A. faxoniana), 20 for each color-leaved species (i.e., P. davidiana, B. platyphylla, L. mastersiana and Q. liaotungensis), and 14 for S. saltuaria. Images of the four color-leaved species were captured during autumn to emphasize their peak aesthetic quality. To mitigate the influence of varying environments, lighting, and other factors, Adobe Photoshop Version 21.0.1 (Adobe, CA, USA) was utilized to eliminate non-plant elements from the images. Subsequently, each species’ dominant color elements—hue, saturation, and brightness—were identified using ColorImpact V4.1.2. for image processing [38]. Finally, other qualitative indicators such as crown shape, trunk clarity, and leaf morphology were primarily collated through a literature review, consultations with local forestry experts, and Flora Reipublicae Popularis Sinicae (https://www.iplant.cn/frps (accessed on 15 November 2020)).
(c)
Questionnaire survey
An electronic questionnaire was developed to gain insight into the public’s perception of species’ biophysical attributes. The questionnaire survey categorized color elements and qualitative indicators into distinct groups based on the attribute values of the eight tree species. Initially, we utilized Adobe Photoshop CS6 to generate diverse images for each sub-indicator, varying in type (outlined in Table 1). These images were then integrated into an electronic questionnaire (accessible at https://www.wjx.cn/vm/OtgFwxr.aspx (accessed on 15 March 2021)) via the Wenjuanxing web questionnaire platform. The questionnaire encompassed three sections. Section one encompassed the basic demographics of the participants, detailed in Table S4. In the second section, respondents were prompted to assess the images based on their initial intuitive perception, following a 5–10 s observation period [30]. Additionally, they were requested to rank the attractiveness of each image series. Given that prior knowledge or recognition of tree species may influence non-expert aesthetic judgments, participants were not informed that the images represented different tree species in the questionnaire. The third section requested participants to assess the aesthetic quality of various forest types at the stand level, as outlined in Section 2.2.1 (e). This evaluation aimed to validate our approach, which is grounded in species’ biophysical attributes. The survey was accessible online in China from February to March 2021, and the participants were randomly selected from the Internet. A total of 577 questionnaires were randomly distributed via email, resulting in the collection of 532 responses. Among the respondents, 52.82% were male and 47.18% were female (Table S4). Regarding the age distribution, individuals aged between 18 and 60 constituted the largest group (72.94%), followed by those under 18 (17.67%), while respondents aged over 60 represented the smallest proportion (9.4%). For more detailed socio-demographic information on the respondents, see Table S4. Quantitative measures of each sub-indicator aesthetic factor were derived from the questionnaire by averaging ratings among observers, as per Daniel et al. [2]. Numerous studies, notably those conducted by Frank et al. [15], Schirpke et al. [18], and Mu et al. [39], consistently indicate that demographic variables, including age, gender, academic discipline, and educational attainment, do not significantly impact the outcomes of aesthetic assessments. Consequently, the present study did not investigate the association between the respondents’ backgrounds and the evaluation outcomes. Cronbach’s test was employed to assess the reliability of the questionnaire results [40], and the alpha value obtained in this study was 0.776, indicating satisfactory reliability and internal consistency of the questionnaire content.
(d)
Standardized and weighted
To standardize the values of different sub-indicators, ranging from 0 (no relevant capacity) to 1 (very high relevant capacity), it is necessary to eliminate the differences between them. The weight of each indicator was determined by experts using the Delphi method [11,41,42]. Twenty experts in ecology, forestry, tourism, and landscape aesthetics were requested to assign preference weights (e.g., 0–1) to selected sub-indicators to reflect stakeholder preferences (Table 1). Since the Delphi method heavily relies on the knowledge and judgments of the appraisers, all selected experts have approximately 20 years of research experience in the field. The final score for species aesthetic quality is the weighted sum of all sub-indicators:
E   =   ∑ i = 1 m ( a i × b i )
where E is the final species aesthetic quality score, m is the total number of sub-indicators, a i and b i   correspond respectively to the value and weight of each sub-indicator. The higher the value is, the higher the species’ aesthetic quality.
(e)
Method validation
The aesthetic quality of species was calculated based on their biophysical attributes and validated against the traditional method of evaluating aesthetic quality at the forest stand level using photo questionnaires. The questionnaire included 8 images, each representing a different forest stand (Figure 1) of eight tree species. The most representative image for each forest type was chosen from various sceneries taken in October 2020 during the peak period of autumn leaf coloration to depict the autumn forest landscape. These images were selected from a small visual field region to minimize the impact of terrain aspects on the perception of forest landscape quality. Participants rated each picture from 1 (very ugly) to 5 (very beautiful) based on their opinion of the aesthetic value of the forest landscapes. The forest aesthetic quality of each stand was calculated as the median score given by participants in the 532 questionnaires analyzed. Pearson’s correlation analysis validated our approach by comparing species’ aesthetic quality with forest stand-level questionnaires. We found a significant positive correlation (R = 0.925 **) between the aesthetic quality based on species biophysical attributes and the forest stand-level questionnaires, indicating that our approach can assess the aesthetic quality of forest stands. Subsequent analyses were based only on the former (Figure S1).

2.2.2. Landscape Aesthetic Quality

The aesthetic quality of each species was calculated as described in Section 2.2.1. The spatial distribution of tree species (30 m resolution for this study) was then combined to obtain the map of the species’ aesthetic quality. We also considered two other landscape characteristics, diversity and relief, when calculating LAQ. This is because the diversity of the landscape and rough landscapes exhibit heightened visual attraction in addition to the species’ aesthetic quality. Following Casado-Arzuaga et al. [4], landscape diversity was calculated as the number of different tree species surrounding each raster pixel. Additionally, relief was determined by calculating the difference between the altitude of each pixel and the average altitude within a 300-m radius. The DEM with 30 m resolution from Geospatial Data Cloud (http://www.gscloud.cn/ (accessed on 20 May 2020)) was used to calculate relief. We hypothesized that higher diversity and relief differences were associated with a higher LAQ value [4,15]. After processing each metric, the map layers of the three elements (species aesthetic quality, landscape diversity, and relief) were combined to create the final LAQ distribution map by using the weighted linear combination method. Based on the approach described by Casado-Arzuaga et al. [4] and in consultation with local administrators and relevant experts, the weights assigned to each element are presented in Table S2. By using the natural breaks (Jenks) method, the resulting map values were divided into five classes (very low, low, medium, high, and very high) to visualize the spatial pattern of LAQ.

2.3. Viewshed and Spatial–Temporal Dynamic Analysis

Using ArcGIS 10.2 software, we calculated the visible area from the sightseeing road based on its topographic position, known as a viewshed. Viewshed area represents the land visible from a single point without obstruction [43]. Participants generally preferred unimpeded views and a broader field of view [44]. This variable is measured in ArcGIS using the 30 m resolution DEM, considering the location, angle of view, and obstructions. We hypothesized that greater viewshed overlap indicates a higher probability of visibility in the overlapping area [45]. The final output map identified areas with the best visual accessibility potential in Jiuzhaigou. We categorized the viewshed map values into five classes (very low, low, medium, high, and very high) using the natural breaks (Jenks) method to reflect visual accessibility probability. Within each viewshed type, we analyzed the spatial distribution pattern of LAQ.
Additionally, we examined the spatial–temporal changes in LAQ from 2016 to 2096. In this study, with relief and species aesthetic quality remaining constant, the only variable for recalculating future landscape diversity is using different periods of the species distribution map. Subsequently, we can calculate the future LAQ and analyze the spatial–temporal changes in LAQ from 2016 to 2096. The initial species distribution map was obtained from the forestry resource survey data that was collected in 2016 for the subcompartment division of Sichuan Province. Subsequently, we simulated future species distribution maps using the LANDIS-II model and the PnET-II model, as described in our previous study [32]. The LANDIS-II model simulates forest dynamics by monitoring different species–age cohorts as they engage based on key characteristics. These include factors such as fire resistance, shade tolerance, longevity, seed dispersal capabilities, and the potential for vegetative sprouting [46]. We employed the LANDIS-II model to simulate forest landscape dynamics at the grid cell level, incorporating species-specific life-history traits, establishment probabilities (SEP), and maximum aboveground net primary production (ANPP) [47]. Within the LANDIS-II framework, values for ANPP and SEP for each species were obtained using the PnET-II model. In this model, ANPP is estimated by incorporating both woody growth and foliage production, whereas SEP relates to processes such as seed germination, seedling development, and mortality. These processes can be influenced by factors like light intensity, temperature, and soil moisture [48]. In our previous research, we employed these models to simulate forest landscape dynamics over the entire 80-year simulation period, including species composition and spatial distribution. We generated species distribution maps at 10-year intervals (2016–2096) under the current climate scenario in Jiuzhaigou, without considering natural disturbances (e.g., wildfires, windthrows, and insect outbreaks) or anthropogenic disturbances (e.g., harvesting, deforestation, and afforestation). The species distribution maps were depicted as a grid of 30 m × 30 m pixels populated by eight dominant species. Subsequently, we calculated the landscape diversity for different future time periods using the method described in Section 2.2.2 within ArcGIS 10.2. Finally, by overlaying the relief and species aesthetic quality, we obtained the future spatial distribution map of LAQ.
This research examined the impact of future forest landscapes on LAQ and, as a result, only the current climate scenario was evaluated while the effects of climate change were not considered. We utilized thirty-year climatological means in the simulations to reflect current climate conditions. Specifically, monthly climate records (temperature and precipitation) from 1986 to 2016 were sourced from 17 meteorological stations located near the study area (China Meteorological Administration, http://data.cma.cn (accessed on 30 April 2020)). These climate data were then combined with topographic information, and the point-based climate data were spatially interpolated at a 90-m resolution using thin-plate smoothing splines through the ANUSPLIN 4.3 software [49].
Validating spatial landscape models is challenging due to limited long-term, region-specific data. We evaluated the LANDIS model by comparing its simulated 2016 aboveground biomass against forestry inventory observations. Stand volume was converted to biomass [50], and 200 random sampling points were used for comparison. A strong positive correlation (R2 = 0.6384, p < 0.001) confirmed the model’s reliability. Details are provided in Liu et al. [32].

2.4. Mann–Kendall Test for Trend Analysis

To examine long-term trends in the LAQ, the non-parametric Mann–Kendall test was employed [51,52]. This method is rank-based and nonparametric, and has been widely applied for detecting monotonic trends within time series data related to environmental, hydrological, and climatic variables [53,54]. Within the Mann–Kendall test, the slope calculated using the Theil-Sen estimator [55] is commonly employed to identify a monotonic trend and to quantify the change in the variable per unit time. The estimator is given as
β = M e d i a n ( x j − x l j − l )     ∀ 1 < 1 < j
where 1 < l < j < n, and x denotes the overall median derived from all record pair combinations in the dataset, thereby ensuring resistance to extreme observations. A positive value of β signifies an upward trend, whereas a negative β indicates a downward trend.

3. Results

3.1. Mapped Elements for Landscape Aesthetic Quality

In our study area, species aesthetic quality scores ranged from 2.52 to 5, with an average pixel value of 3.03 (Table 2). Color-leaved species such as B. platyphylla, Q. liaotungensis and P. davidiana generally have a high aesthetic quality, while evergreen coniferous species (e.g., P. asperata and A. faxoniana) forests have relatively low values. The spatial pattern of variations in species aesthetic quality (Figure 3a) aligns with species distribution (Figure 1). Most areas have low or mediocre scores, particularly in the Danzu Valley, Zechawa Valley, and Zangmalongli Valley (Figure 3a). However, a few areas with high scores are found in the Shuzheng Valley and Zharu Valley. These areas also exhibit high scores for landscape diversity, as depicted in the resulting map, despite diversity being a minor weighting factor in LAQ assessment (Figure 3b). The mean score of landscape diversity is only 0.37, with a peak at 1. However, the map presents a varied picture; landscape diversity scores are generally higher in the north of the study area at low altitudes, and lower values are found in the south of Jiuzhaigou at high altitudes, where extensive spruce-fir forests result in low diversity. The relief score demonstrates a strong tendency toward high values, with a peak at 1 and a mean of 0.7 (Table 2). Spatially, most analyzed areas are characterized by high relief scores (Figure 3c).

3.2. Landscape Aesthetic Quality Map

LAQ scores in Jiuzhaigou ranged from 2.69 to 7, with prevailing scores below 4.5 and a mean score of 4.53 (Table 2). The 2016 LAQ map for Jiuzhaigou (Figure 4) indicates particularly high scores in the naturally low-altitude northern regions, notably in Shuzheng Valley and Zharu Valley, where all three elements exhibit high scores. However, contrasting examples can be found in the southern regions with high altitudes. Specifically, 68.0% of the analyzed area had a “low” or “very low” aesthetic value, while only 5.4% had a “very high” value when the LAQ scores were divided into five levels. The regions with “very low” LAQ were situated along the upper edge of the forest distribution. Adjacent to them, the colder-climate coniferous forest landscapes in high altitudes, as one of the most common landscape types in our study area, are typically monotonous and exhibit low perceived aesthetic.

3.3. Viewshed Analysis

The spatial distribution of viewshed areas is depicted in Figure 5a. Overall, most areas along the sightseeing road have good visual accessibility, especially in Shuzheng and Rize Valleys, which are assigned high or very high values. However, some areas in Danzu Valley and the southern parts of Jiuzhaigou are not visible despite having low LAQ in these areas. Notably, around 56.3% of the visually accessible areas are covered by the viewshed compared to the LAQ-analyzed areas. The LAQ pattern in viewshed areas closely resembles that observed in all of Jiuzhaigou territory (Figure 5b), but the average scores in viewshed areas (4.68, range: 3.20–7) are slightly higher than in the entire territory (4.53, range: 2.69–7) (Table 2). Additionally, a high proportion of areas (72.8%) with “very high” LAQ had visual accessibility, while more than half of areas with “very low” LAQ are invisible (Figure 5b). Approximately 70% of the regions exhibiting LAQ were observed from viewshed areas of “very low” and “low” classification. The proportion of area with LAQ showed a decreased trend with increasing visual accessibility of the viewshed (Table 3).

3.4. Spatial–Temporal Dynamic

Over the next 80 years, the area percentages of “very low”, “low”, and “medium” LAQ categories are consistently higher than those of “high” and “very high” LAQ categories (Figure 6). We note a divergence in LAQ types relative to the trends in area percentages. Generally, the area percentages of “very low”, “low”, and “medium” LAQ categories exhibit an upward trend over the next 80 years (2016–2096) (p < 0.001), with predicted increases of 12.0%, 11.0%, and 7.10% by the end of the 21st century, respectively. Although the predicted area percentages of “high” and “very high” LAQ categories demonstrated a statistically significant upward trend (p < 0.001), their percentages decreased in the final decade, declining by 0.44% and 9.20%, respectively, by 2096 compared to 2016. Across the entire region with LAQ, the overall area percentage gradually increased by 8.68% by 2096.
Upon analyzing forest landscape dynamics, we observed no significant changes in LAQ scores in 2056, as the average values in 2056 (4.52) remained close to those of 2016 (4.53). However, we predicted a significant decreasing trend in LAQ by the end of the 21st century (p < 0.001). The spatial distribution of LAQ in the future (Figure 7) mirrors that of 2016 (Figure 4), with high LAQ values concentrated in the northern low-altitude regions and relatively lower values in the southern high-altitude regions, exhibiting a diverse spatial pattern. Notably, the proportion of the high LAQ zone increased from 2.44% in 2016 to 2.51% in 2056—especially in the Shuzheng and Zharu Valleys—and then declined significantly to 2.21% by 2096 (Figure 7). We have observed that the LAQ in these areas will improve by 2056, but will deteriorate by the end of the century (Figure 8). Nevertheless, the LAQ in other regions with high altitudes, especially at the upper edge of forest distribution, will improve by both 2056 and 2096, even in areas with a low LAQ score. The area percentage of these regions shows an increasing trend under future forest landscape dynamics.

4. Discussion

Our study presented a reliable approach for assessing LAQ in subalpine forest ecosystems across the Tibetan Plateau. As far as we know, this is the first study to predict the spatiotemporal changes in LAQ in this region. More importantly, by mapping the current and future LAQ, our findings provide a critical scientific basis for addressing the growing challenges of tourism pressure and for formulating adaptive land management policies that enhance the long-term resilience of landscape aesthetics in this iconic World Heritage Site. Our study also emphasizes the need to manage these forests better to overcome the potential decrease in LAQ predicted by the end of the 21st century.

4.1. Methodological Insights

Landscape aesthetics is grounded in objective parameters and indicators, which quantitatively assess the perceptual evaluation of aesthetic landscape characteristics [20]. The psychophysical paradigm defended that those preferences for and the attractiveness of a particular landscape are ostensibly rooted in the landscape’s biophysical characteristics [4,13]. Following this paradigm, we developed an innovative methodology that linked species’ biophysical attributes with human perceptions assessed by a questionnaire survey. We also considered landscape characteristics: landscape diversity and relief, following Casado-Arzuaga et al. [4], as terrain ruggedness may increase aesthetic quality because of high visual attraction and more opportunities, notably in mountain areas [56]. The selection of indicators was grounded in a comprehensive review of aesthetic quality and scenic beauty assessments of forest landscapes, incorporating insights derived from perception theory and empirical research on visual preferences [34,35]. Despite such an approach not being conducted in previous studies, we argue that ample evidence exists to substantiate the validity of our selected criteria. Our findings support the assumption of a general consensus among the public regarding LAQ, specifically that aesthetic quality in color-leaved species is higher than other evergreen species.
Previous studies, relying on photo projections and questionnaires, primarily concentrate on specific geographical regions [18,28,30]. Consequently, they lack an efficient batch evaluation system. In contrast, by using digital spatial data, GIS tools can reconstruct landscapes visible to observers, which has been employed in aesthetic quality research [57]. Nevertheless, these studies primarily concentrate on aspects of ecological landscapes while disregarding human preferences. Compared to these methods, our approach demonstrates its strength by directly highlighting the impact of species’ biophysical attributes on forest LAQ provision. Additionally, our method can be implemented in GIS Version 10.2 software due to the consistent aesthetic quality of species, which extends beyond specific site characteristics. This enables batch processing for LAQ evaluation and analysis, facilitating monitoring and scenario analysis. However, it is worth noting that our method does not account for community characteristics at the stand level, such as tree density, stand age, and vegetation coverage, which can vary significantly across regions and even raster pixels.
Taking into account all factors, spatially explicit LAQ assessments can encounter challenges due to the reliance on simplistic proxies that overlook unique environmental conditions, as highlighted by Paracchini et al. [58]. Conversely, when analyzing temporal LAQ variations in the future, the exclusion of community structure indicators from forest landscape models poses integration challenges. Furthermore, our quantification of aesthetic quality focuses solely on the color elements of autumn foliage, disregarding seasonal variations. While seasonal variations in LAQ are indeed significant in our study areas [59], our primary focus is on long-term LAQ changes resulting from forest landscape dynamics over decades or centuries. Therefore, interannual variations are not the primary concern of this study. Additionally, the influence of water bodies on LAQ is neglected in our study, despite their recognized significance as a key factor [58,59]. As a consequence, the underestimation of LAQ, particularly in high-altitude forest landscapes adjacent to water bodies, may occur due to the clustered distribution of water bodies primarily in the low-altitude regions of Shuzheng, Rize, and Zechawa Valleys (Figure 1). Despite these limitations, our approach offers novel insights into societal preferences for forest landscapes by correlating aesthetic quality with species biophysical attributes. It can also be applied to LAQ assessments of other subalpine forest ecosystems. Future research should explore potential influencing factors and integrate specific surveys to better understand human perceptions and preferences. This study serves as a foundation for such endeavors.

4.2. Aesthetic Quality Map

Our findings demonstrate significant spatial variations in LAQ distribution. High-value zones are predominantly concentrated in the northern areas with low altitude, while the southern areas with high altitude mostly receive low or mediocre scores. This disparity is primarily due to the spatial distribution of tree species, which affects two key elements of LAQ assessment: species aesthetic quality and diversity. In contrast, relief has minimal impact on the spatial distribution of LAQ variation, as the substantial altitude drop across our mountain landscapes results in most areas receiving high relief scores.
Our results align with previous studies [22,28], indicating a preference for color-leaved species over other evergreen coniferous species regarding aesthetic quality. Notably, color carries the greatest weight, comprising nearly half of all weights, highlighting its stronger role in aesthetic service compared to other biophysical attributes, consistent with prior research [28]. This can be explained by the relatively obvious effect of color, as a visual attribute triggered by light, on the transmission of species’ visual traits. Furthermore, color exerts a more potent stimulation on human visual nerves than other attributes, such as size and shape [60]. Previous studies have also found that yellow or orange hues, high brightness, and saturation are associated with higher aesthetic scores for color-leaved species compared to evergreen coniferous species [61], a finding supported by our study. Furthermore, our results indicate a higher species diversity in the northern part of our study area compared to other regions, revealing a spatial distribution variation in LAQ in combination with species aesthetic quality.
The average LAQ accessibility values in the viewshed analysis are slightly higher than those in the entire Jiuzhaigou territory; this can be explained by the visual accessibility of most areas with high LAQ, mainly located at low altitudes or by the wayside, where color-leaved species are concentrated [62]. Therefore, preserving these areas is essential for maintaining the local natural landscape values. However, it is noted that some scattered regions with high potential for LAQ, located at higher altitudes, are not visible from sightseeing roads due to their high roughness and limited long-distance visibility. Additionally, some areas with high LAQ in Zharu Valley are not visible due to the inaccessibility of sightseeing roads. Although most areas with high LAQ are visible in our study area, only a small percentage is visible from the viewshed areas classified as “high” and “very high.” This suggests that these visually accessible areas could be prioritized as forest management areas when local administrators plan to improve LAQ.
Placing LAQ in a spatiotemporal context is crucial. Although there are no significant changes in the overall mean scores, the resulting spatial variation indicates an increase in the high-value zone of LAQ in 2056, followed by a decrease in 2096 due to future forest landscape dynamics. These dynamics impact forest cover, alter tree species composition and distribution, and consequently affect forest ecosystem services, including LAQ [63]. A previous study within the same region observed that the area percentages of color-leaved species increased in 2056 but significantly decreased by the end of the 21st century [32]. Evergreen coniferous species (i.e., Picea asperata forest and Abies faxoniana forest) are considered the climax species in succession in our study area [32], indicating that species in the study area will likely converge to these species without anthropogenic disturbances (e.g., deforestation, harvesting, and afforestation). Therefore, changes in tree species distribution, combined with differences in their aesthetic quality, lead to spatiotemporal variation in LAQ over the entire 80-year simulation period. Additionally, it is noteworthy that LAQ in those regions with high altitudes, especially at the upper edge of forest distribution, shows an enhanced trend under future forest landscape dynamics. Since our LAQ assessment excluded shrub species, assuming they have no LAQ value, the evident upward range shift of forest distribution in our study area led to enhanced LAQ in these regions [32]. However, these regions, mostly above 3700 m, are predominantly covered by Rhododendron shrubs, which have high aesthetic value. Therefore, our results may exaggerate the impact of forest landscape dynamics on LAQ in these areas. Despite this limitation, we still have confidence in our results because our objective was not to precisely predict LAQ changes in response to forest landscape dynamics, but to assess general future trends.

4.3. Implications for Management and Planning

Mapping the LAQ can help local administrators become more aware of the need to establish and implement forest management strategies [64,65]. Our findings reveal a diverse spatial pattern of LAQ within the study region, providing valuable insights for forest management and policy formulation. Areas with high LAQ values, indicative of superior visual aesthetics, ought to be prioritized for inclusion within landscape conservation areas. Given the regions’ status as a major tourist destination, these areas are under constant threat from tourism-related infrastructure development and visitor impact. Proactive policies are needed to strictly regulate human activities in these zones to preserve their aesthetic integrity. At the same time, areas with particularly low LAQ need to focus on improving aesthetic quality, as local administrators tend to retain or increase areas with high LAQ for tourism development purposes [62].
The relative importance of each indicator in the final LAQ assessment system can be reflected by its numerical weight, which could guide landscape construction for further improvement. Changing forest structure, adjusting, or replacing tree species with color-leaved species seems practical, as the color element has the highest weight value in LAQ assessment, and color-leaved species are confirmed to have higher aesthetic quality than other evergreen coniferous species. However, such interventions must be designed with long-term ecosystem resilience in mind. For example, the selected species should be both aesthetically appealing and ecologically adapted to future climatic conditions while exhibiting resistance to pests and diseases. Additionally, diversity plays an essential role in LAQ; therefore, considering management methods for creating appropriately mixed forest landscapes can improve LAQ. Not all areas with low LAQ need to be reconstructed, as it is labor-intensive and time-consuming. Areas with high visual accessibility should be prioritized to effectively increase the visible probability of high LAQ.
Our projection of a decrease in LAQ by the end of the century is a critical warning. Forest managers can use these spatiotemporal assessments to monitor resources and anticipate changes. Regions projected to transition from high to low LAQ warrant further attention. To mitigate these adverse trends, policymakers may proactively adopt measures such as preemptive thinning, assisted species migration, and the development of aesthetically oriented buffer zones. Specifically, we propose that reserve management authorities incorporate an “LAQ spatial distribution map” into regional spatial planning frameworks. In areas with high LAQ, no new roads or ancillary facilities should be constructed, and any proposed developments must undergo a rigorous visual impact threshold review. This proactive strategy is critical for preserving the landscape’s aesthetic resilience amid the challenges posed by climate change and escalating anthropogenic pressures.

5. Conclusions

Spatially explicit assessment of LAQ necessitates linking landscape properties with human perceptions. Our approach, based on species’ biophysical attributes, enables us to estimate the forest LAQ at any location to monitor its spatial–temporal dynamic changes over time. Our findings indicate that most areas with high LAQ scores had visual accessibility, which could be strategically leveraged by decision-makers and landscape planners for tourism development purposes. Nevertheless, LAQ will decrease under future forest landscape dynamics. To our knowledge, this is the first study to predict the spatiotemporal changes in LAQ in subalpine areas across the Tibetan Plateau, which can help local forest administrators make timely policy adjustments in response to the potential decrease in LAQ predicted by the end of the 21st century. Additionally, as similar developments may occur in other subalpine regions all over the World Natural Heritage Site, our findings provide a useful information basis to assist forest authorities and managers in developing more effective forest management strategies to enhance aesthetic experiences.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/f16121804/s1, Figure S1: Comparison between species aesthetic quality and forest aesthetic quality. Figure S2: The temporal changes in LAQ of Jiuzhaigou from 2016 to 2096. Table S1: Distribution of tree species in Jiuzhaigou National Nature Reserve. Table S2: Weighting of different elements considered to map the LAQ. Table S3: Value of each indicator of the eight tree species in species aesthetic quality evaluation. Table S4: Socio-demographic characteristics of the participants (N = 532). Landscape Aesthetic Quality Questionnaire.

Author Contributions

J.L.: Conceptualization, Data curation, Formal analysis, Visualization, Writing—original draft. J.D.: Data curation, Writing—review and editing. C.Z.: Data curation. B.B.: Writing—review and editing. Y.Y.: Writing—review and editing. T.D.: Funding acquisition, Writing—review and editing. Y.W.: Conceptualization, Funding acquisition, Supervision. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by by Jiuzhaigou Post-Disaster Restoration and Reconstruction Program (grant number 5132202020000046), Yunnan Fundamental Research Projects (202401BF070001-003), the Foundation of Ministry of Education Key Laboratory of Southwest China Wildlife Resources Conservation (XNYB25-01), Fund of Ministry of Education Key Laboratory for Transboundary Ecosecurity of Southwest China (YNUECO2024003) and Doctor Fund of Kunming University (YJL25002).

Data Availability Statement

Data available on request from the corresponding author.

Acknowledgments

We thank Heng-Xing Zou and Xiushan Li for providing valuable comments on the manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Millennium Ecosystem Assessment (Program) (Ed.) Ecosystems and Human Well-Being: Synthesis; Island Press: Washington, DC, USA, 2005. [Google Scholar]
  2. Daniel, T.C.; Muhar, A.; Arnberger, A.; Aznar, O.; Boyd, J.W.; Chan, K.M.A.; Costanza, R.; Elmqvist, T.; Flint, C.G.; Gobster, P.H.; et al. Contributions of cultural services to the ecosystem services agenda. Proc. Natl. Acad. Sci. USA 2012, 109, 8812–8819. [Google Scholar] [CrossRef]
  3. Albert, C.; Galler, C.; Hermes, J.; Neuendorf, F.; Von Haaren, C.; Lovett, A. Applying ecosystem services indicators in landscape planning and management: The ES-in-Planning framework. Ecol. Indic. 2016, 61, 100–113. [Google Scholar] [CrossRef]
  4. Casado-Arzuaga, I.; Onaindia, M.; Madariaga, I.; Verburg, P.H. Mapping recreation and aesthetic value of ecosystems in the Bilbao Metropolitan Greenbelt (northern Spain) to support landscape planning. Landsc. Ecol. 2014, 29, 1393–1405. [Google Scholar] [CrossRef]
  5. Zheng, H.; Wang, L.; Wu, T. Coordinating ecosystem service trade-offs to achieve win–win outcomes: A review of the approaches. J. Environ. Sci. 2019, 82, 103–112. [Google Scholar] [CrossRef]
  6. Plieninger, T.; Shamohamadi, S.; García-Martín, M.; Quintas-Soriano, C.; Shakeri, Z.; Valipour, A. Community, pastoralism, landscape: Eliciting values and human-nature connectedness of forest-related people. Landsc. Urban Plan. 2023, 233, 104706. [Google Scholar] [CrossRef]
  7. Zhang, M.; Xiong, K.; Zhao, X.; Lyu, X. Mapping and assessment of karst landscape aesthetic value from a world heritage perspective: A case study of the Huangguoshu Scenic area. Herit. Sci. 2024, 12, 195. [Google Scholar] [CrossRef]
  8. Tribot, A.-S.; Deter, J.; Mouquet, N. Integrating the aesthetic value of landscapes and biological diversity. Proc. R. Soc. B Biol. Sci. 2018, 285, 20180971. [Google Scholar] [CrossRef] [PubMed]
  9. Schirpke, U.; Mölk, F.; Feilhauer, E.; Tappeiner, U.; Tappeiner, G. How suitable are discrete choice experiments based on landscape indicators for estimating landscape preferences? Landsc. Urban Plan. 2023, 237, 104813. [Google Scholar] [CrossRef]
  10. Veerkamp, C.J.; Schipper, A.M.; Hedlund, K.; Lazarova, T.; Nordin, A.; Hanson, H.I. A review of studies assessing ecosystem services provided by urban green and blue infrastructure. Ecosyst. Serv. 2021, 52, 101367. [Google Scholar] [CrossRef]
  11. Ha, S.; Yang, Z. Evaluation for landscape aesthetic value of the Natural World Heritage Site. Environ. Monit. Asse. 2019, 191, 483. [Google Scholar] [CrossRef]
  12. Zhang, W.; Yu, Y.; Wu, X.; Pereira, P.; Esteban Lucas-Borja, M. Integrating preferences and social values for ecosystem services in local ecological management: A framework applied in Xiaojiang Basin Yunnan province, China. Land Use Policy 2019, 91, 104339. [Google Scholar] [CrossRef]
  13. Wang, L.; Zheng, H.; Chen, Y.; Ouyang, Z.; Hu, X. Systematic review of ecosystem services flow measurement: Main concepts, methods, applications and future directions. Ecosys. Serv. 2022, 58, 101479. [Google Scholar] [CrossRef]
  14. Gosal, A.S.; Ziv, G. Landscape aesthetics: Spatial modelling and mapping using social media images and machine learning. Ecol. Indic. 2020, 117, 106638. [Google Scholar] [CrossRef]
  15. Frank, S.; Fürst, C.; Koschke, L.; Witt, A.; Makeschin, F. Assessment of landscape aesthetics—Validation of a landscape metrics-based assessment by visual estimation of the scenic beauty. Ecol. Indic. 2013, 32, 222–231. [Google Scholar] [CrossRef]
  16. Kang, N.; Liu, C. Towards landscape visual quality evaluation: Methodologies, technologies, and recommendations. Ecol. Indic. 2022, 142, 109174. [Google Scholar] [CrossRef]
  17. Gan, Q.; Liao, L.; Kang, X.; Xu, Z.; Fu, T.; Cao, Y.; Feng, Y.; Dong, J. Cultural ecosystem services and disservices in protected areas: Hotspots and influencing factors based on tourists’ digital footprints. Ecosyst. Serv. 2024, 70, 101680. [Google Scholar] [CrossRef]
  18. Schirpke, U.; Tasser, E.; Tappeiner, U. Predicting scenic beauty of mountain regions. Landsc. Urban Plan. 2013, 111, 1–12. [Google Scholar] [CrossRef]
  19. Hermes, J.; Albert, C.; von Haaren, C. Assessing the aesthetic quality of landscapes in Germany. Ecosyst. Serv. 2018, 31, 296–307. [Google Scholar] [CrossRef]
  20. Kalinauskas, M.; Miksa, K.; Inacio, M.; Gomes, E.; Pereira, P. Mapping and assessment of landscape aesthetic quality in Lithuania. J. Environ. Manag. 2021, 286, 112239. [Google Scholar] [CrossRef]
  21. Balzan, M.V.; Caruana, J.; Zammit, A. Assessing the capacity and flow of ecosystem services in multifunctional landscapes: Evidence of a rural-urban gradient in a Mediterranean small island state. Land Use Policy 2018, 75, 711–725. [Google Scholar] [CrossRef]
  22. Li, Q.; Du, Y.; Liu, Y.; Chen, J.; Zhang, X.; Liu, J.; Tao, J. Canopy Gaps Improve Landscape Aesthetic Service by Promoting Autumn Color-Leaved Tree Species Diversity and Color-Leaved Patch Properties in Subalpine Forests of Southwestern China. Forests 2021, 12, 199. [Google Scholar] [CrossRef]
  23. Felipe-Lucia, M.R.; Soliveres, S.; Penone, C.; Manning, P.; van der Plas, F.; Boch, S.; Prati, D.; Ammer, C.; Schall, P.; Gossner, M.M.; et al. Multiple forest attributes underpin the supply of multiple ecosystem services. Nat. Commun. 2018, 9, 4839. [Google Scholar] [CrossRef]
  24. Graves, R.A.; Pearson, S.M.; Turner, M.G. Landscape dynamics of floral resources affect the supply of a biodiversity-dependent cultural ecosystem service. Landsc. Ecol. 2017, 32, 415–428. [Google Scholar] [CrossRef]
  25. Zheng, D.; Wang, Y.; Hao, S.; Xu, W.; Lv, L.; Yu, S. Spatial-temporal variation and tradeoffs/synergies analysis on multiple ecosystem services: A case study in the Three-River Headwaters region of China. Ecol. Indic. 2020, 116, 106494. [Google Scholar] [CrossRef]
  26. Wu, Y.; Wang, D.; Qiao, X.; Jiang, M.; Li, Q.; Gu, Z.; Liu, F. Forest dynamics and carbon storage under climate change in a subtropical mountainous region in central China. Ecosphere 2020, 11, e03072. [Google Scholar] [CrossRef]
  27. Xu, M.; Liu, T.; Xie, P.; Chen, H.; Su, Z. Temporal Changes in Community Structure over a 5-Year Successional Stage in a Subtropical Forest. Forests 2020, 11, 438. [Google Scholar] [CrossRef]
  28. Zhang, Z.; Qie, G.; Wang, C.; Jiang, S.; Li, X.; Li, M. Relationship between Forest Color Characteristics and Scenic Beauty: Case Study Analyzing Pictures of Mountainous Forests at Sloped Positions in Jiuzhai Valley, China. Forests 2017, 8, 63. [Google Scholar] [CrossRef]
  29. Wang, Y.; Li, Z.; Deng, X. Assessment of cultural ecosystem services of the Qinghai-Tibetan Plateau: Guarding the beauty of the Plateau, co-creating a better future. Ecol. Indic. 2024, 166, 112405. [Google Scholar] [CrossRef]
  30. Jia, H.; Luo, P.; Yang, H.; Luo, C.; Li, H.; Wu, S.; Cheng, Y.; Huang, Y.; Xie, W. Exploring the Relationship between Forest Scenic Beauty with Color Index and Ecological Integrity: Case Study of Jiuzhaigou and Giant Panda National Park in Sichuan, China. Forests 2022, 13, 1883. [Google Scholar] [CrossRef]
  31. Huang, S.; Xu, R.; Wang, N.; Qian, J.; Tu, W.; Luo, P.; Xing, A.Z.F. Unveiling the potential supply of cultural ecosystem services on the Qinghai-Tibet Plateau: Insights from tourist hiking trajectories. Ecosyst. Serv. 2025, 73, 101711. [Google Scholar] [CrossRef]
  32. Liu, J.; Zou, H.; Bachelot, B.; Dong, T.; Zhu, Z.; Liao, Y.; Plenković-Moraj, A.; Wu, Y. Predicting the responses of subalpine forest landscape dynamics to climate change on the eastern Tibetan Plateau. Glob. Change Biol. 2021, 27, 4352–4366. [Google Scholar] [CrossRef] [PubMed]
  33. Kuper, R. Effects of Flowering, Foliation, and Autumn Colors on Preference and Restorative Potential for Designed Digital Landscape Models. Environ. Behav. 2020, 52, 544–576. [Google Scholar] [CrossRef]
  34. Felton, A.; Petersson, L.; Nilsson, O.; Witzell, J.; Cleary, M.; Felton, A.M.; Björkman, C.; Sang, Å.O.; Jonsell, M.; Holmström, E.; et al. The tree species matters: Biodiversity and ecosystem service implications of replacing Scots pine production stands with Norway spruce. Ambio 2020, 49, 1035–1049. [Google Scholar] [CrossRef]
  35. Tveit, M.; Ode, A.; Fry, G. Key concepts in a framework for analysing visual landscape character. Landsc. Res. 2006, 31, 229–255. [Google Scholar] [CrossRef]
  36. Lu, X.; Liang, E.; Wang, Y.; Babst, F.; Leavitt, S.W.; Camarero, J.J. Past the climate optimum: Recruitment is declining at the world’s highest juniper shrublines on the Tibetan Plateau. Ecology 2019, 100, e02557. [Google Scholar] [CrossRef]
  37. Xu, Z.; Liu, Q.; Du, W.; Zhou, G.; Qin, L.; Sun, Z. Modelling leaf phenology of some trees with accumulated temperature in a temperate forest in northeast China. For. Ecol. Manag. 2021, 489, 119085. [Google Scholar] [CrossRef]
  38. Moretti, G.; Marsland, S.; Lyons, P. Computational production of colour harmony. Part 2: Experimental evaluation of a tool for gui colour scheme creation. Color Res. Appl. 2013, 38, 218–228. [Google Scholar] [CrossRef]
  39. Mu, Y.; Lin, W.; Diao, X.; Zhang, Z.; Wang, J.; Lu, Z.; Guo, W.; Wang, Y.; Hu, C.; Zhao, C. Implementation of the visual aesthetic quality of slope forest autumn color change into the configuration of tree species. Sci. Rep. 2022, 12, 1034. [Google Scholar] [CrossRef]
  40. Cronbach, L.J. Coefficient alpha and the internal structure of tests. Psychometrika 1951, 16, 297–334. [Google Scholar] [CrossRef]
  41. Shi, J.; Han, D.; Chen, C.; Shen, X. KTMN: Knowledge-driven Two-stage Modulation Network for visual question answering. Multimed. Syst. 2024, 30, 350. [Google Scholar] [CrossRef]
  42. Han, D.; Shi, J.; Zhao, J.; Wu, H.; Zhou, Y.; Li, L.-H.; Khan, M.K.; Li, K.-C. LRCN: Layer-residual Co-Attention Networks for visual question answering. Expert Syst. Appl. 2025, 263, 125658. [Google Scholar] [CrossRef]
  43. Grêt-Regamey, A.; Bishop, I.D.; Bebi, P. Predicting the scenic beauty value of mapped landscape changes in a mountainous region through the use of GIS. Environ. Plan. B Plan. Des. 2007, 34, 50–67. [Google Scholar] [CrossRef]
  44. Riitters, K.H.; O’Neill, R.V.; Hunsaker, C.T.; Wickham, J.D.; Yankee, D.H.; Timmins, S.P.; Jones, K.B.; Jackson, B.L. A factor analysis of landscape pattern and structure metrics. Landsc. Ecol. 1995, 10, 23–39. [Google Scholar] [CrossRef]
  45. Yoshimura, N.; Hiura, T. Demand and supply of cultural ecosystem services: Use of geotagged photos to map the aesthetic value of landscapes in Hokkaido. Ecosyst. Serv. 2017, 24, 68–78. [Google Scholar] [CrossRef]
  46. Mladenoff, D.J. LANDIS and forest landscape models. Ecol. Model. 2004, 180, 7–19. [Google Scholar] [CrossRef]
  47. Scheller, R.M.; Domingo, J.B.; Sturtevant, B.R.; Williams, J.S.; Rudy, A.; Gustafson, E.J.; Mladenoff, D.J. Design, development, and application of LANDIS-II, a spatial landscape simulation model with flexible temporal and spatial resolution. Ecol. Model. 2007, 201, 409–419. [Google Scholar] [CrossRef]
  48. Castro, J.; Zamora, R.; Hódar, J.A.; Gómez, J.M. Seedling establishment of a boreal tree species (Pinus sylvestris) at its southernmost distribution limit: Consequences of being in a marginal Mediterranean habitat. J. Ecol. 2004, 92, 266–277. [Google Scholar] [CrossRef]
  49. Hijmans, R.J.; Cameron, S.E.; Parra, J.L.; Jones, P.G.; Jarvis, A. Very high resolution interpolated climate surfaces for global land areas. Int. J. Climatol. 2005, 25, 1965–1978. [Google Scholar] [CrossRef]
  50. Fang, J.; Chen, A.; Peng, C.; Zhao, S.; Ci, L. Changes in forest biomass carbon storage in China between 1949 and 1998. Science 2001, 292, 2320–2322. [Google Scholar] [CrossRef]
  51. Mann, H.B. Nonparametric tests against trend. Econometrica. 1945, 13, 245–259. [Google Scholar] [CrossRef]
  52. Kendall, M.G. Rank Correlation Methods; Charles Griffin: London, UK, 1975. [Google Scholar]
  53. Geng, M.M.; Wang, K.L.; Yang, N.; Li, F.; Zou, Y.A.; Chen, X.S.; Deng, Z.M.; Xie, Y.H. Evaluation and variation trends analysis of water quality in response to water regime changes in a typical river-connected lake (Dongting Lake), China. Environ. Pollut. 2021, 268, 115761. [Google Scholar] [CrossRef]
  54. Mohammed, S.; Gill, A.R.; Alsafadi, K.; Hijazi, O.; Yadav, K.K.; Hasan, M.A.; Khan, A.H.; Islam, S.; Cabral-Pinto, M.; Harsanyi, E. An overview of greenhouse gases emissions in Hungary. J. Clean. Prod. 2021, 314. [Google Scholar] [CrossRef]
  55. Theil, H. A rank invariant method of linear and polynomial regression analysis. Proc. R. Neth. Acad. Sci. 1950, 53, 1397–1412. [Google Scholar]
  56. Weyland, F.; Laterra, P. Recreation potential assessment at large spatial scales: A method based in the ecosystem services approach and landscape metrics. Ecol. Indic. 2014, 39, 34–43. [Google Scholar] [CrossRef]
  57. Palacio Buendia, A.V.; Perez-Albert, Y.; Serrano Gine, D. Mapping Landscape Perception: An Assessment with Public Participation Geographic Information Systems and Spatial Analysis Techniques. Land 2021, 10, 632. [Google Scholar] [CrossRef]
  58. Paracchini, M.L.; Zulian, G.; Kopperoinen, L.; Maes, J.; Schägner, J.P.; Termansen, M.; Zandersen, M.; Perez-Soba, M.; Scholefield, P.A.; Bidoglio, G. Mapping cultural ecosystem services: A framework to assess the potential for outdoor recreation across the EU. Ecol. Indic. 2014, 45, 371–385. [Google Scholar] [CrossRef]
  59. Li, J.; Huang, Z.; Zhu, Z.; Ding, G. Coexistence Perspectives: Exploring the impact of landscape features on aesthetic and recreational values in urban parks. Ecol. Indic. 2024, 162, 112043. [Google Scholar] [CrossRef]
  60. Park, S.S. Handbook of Vitreo-Retinal Disorder Management: A Practical Reference Guide; World Scientific: Singapore, 2015. [Google Scholar] [CrossRef]
  61. Sun, Y.M. Color Quantization and Evaluation Research on Fall-Color Trees in Beijing; Beijing Forestry University: Beijing, China, 2015. [Google Scholar]
  62. Bossard, C.C.; Cao, Y.; Wang, J.; Rose, A.; Tang, Y. New patterns of establishment and growth of Picea, Abies and Betula tree species in subalpine forest gaps of Jiuzhaigou National Nature Reserve, Sichuan, southwestern China in a changing environment. For. Ecol. Manag. 2015, 356, 84–92. [Google Scholar] [CrossRef]
  63. Solomon, N.; Segnon, A.C.; Birhane, E. Ecosystem Service Values Changes in Response to Land-Use/Land-Cover Dynamics in Dry Afromontane Forest in Northern Ethiopia. Int. J. Environ. Res. Public Health 2019, 163, 4653. [Google Scholar] [CrossRef]
  64. Hersperger, A.M.; Bürgi, M.; Wende, W.; Bacău, S.; Grădinaru, S.R. Does landscape play a role in strategic spatial planning of European urban regions? Landsc. Urban Plan. 2020, 194, 103702. [Google Scholar] [CrossRef]
  65. Gao, Z.; Wu, C.; Li, N.; Wang, P.; Li, J. Visual Aesthetic Quality of Qianjiangyuan National Park Landscapes and Its Spatial Pattern Characteristics. Forests 2024, 15, 1289. [Google Scholar] [CrossRef]
Forests 16 01804 g001
Figure 1. Location (a) and forest types (b) of the study area. (c) Photograph of the “color forest” in the study area (from the Jiuzhaigou National Nature Reserve Administration: https://www.jiuzhai.com/ (accessed on 16 May 2020)).
Figure 1. Location (a) and forest types (b) of the study area. (c) Photograph of the “color forest” in the study area (from the Jiuzhaigou National Nature Reserve Administration: https://www.jiuzhai.com/ (accessed on 16 May 2020)).
Forests 16 01804 g001
Forests 16 01804 g002
Figure 2. The framework for the evaluation of landscape aesthetic quality (Note: the numbers on the arrows indicate the weight of each sub-indicator).
Figure 2. The framework for the evaluation of landscape aesthetic quality (Note: the numbers on the arrows indicate the weight of each sub-indicator).
Forests 16 01804 g002
Forests 16 01804 g003
Figure 3. Mapped elements for landscape aesthetics quality in Jiuzhaigou (species aesthetics quality (a), landscape diversity (b) and relief (c)).
Figure 3. Mapped elements for landscape aesthetics quality in Jiuzhaigou (species aesthetics quality (a), landscape diversity (b) and relief (c)).
Forests 16 01804 g003
Forests 16 01804 g004
Figure 4. Landscape aesthetic quality map of Jiuzhaigou in 2016.
Figure 4. Landscape aesthetic quality map of Jiuzhaigou in 2016.
Forests 16 01804 g004
Forests 16 01804 g005
Figure 5. Viewshed areas (a) and landscape aesthetic quality viewshed analysis in 2016 (b).
Figure 5. Viewshed areas (a) and landscape aesthetic quality viewshed analysis in 2016 (b).
Forests 16 01804 g005
Forests 16 01804 g006
Figure 6. Area percentages of LAQ over the next 80 years.
Figure 6. Area percentages of LAQ over the next 80 years.
Forests 16 01804 g006
Forests 16 01804 g007
Figure 7. The spatial distribution of LAQ at year 2056 and 2096.
Figure 7. The spatial distribution of LAQ at year 2056 and 2096.
Forests 16 01804 g007
Forests 16 01804 g008
Figure 8. The spatial distribution change in LAQ by 2096.
Figure 8. The spatial distribution change in LAQ by 2096.
Forests 16 01804 g008
Table 1. Description and weighting of the indicators and sub-indicators used in species aesthetic quality evaluation.
Table 1. Description and weighting of the indicators and sub-indicators used in species aesthetic quality evaluation.
IndicatorWeightSub-IndicatorWeightDescription
Color elements0.4830Hue0.1411Hue was classified into three categories according to the color of eight tree species: yellow, orange and green.
Saturation0.1022Saturation was classified into three categories—high, medium, and low—based on the saturation ranges of eight tree species using the natural breaks method.
Brightness0.0984Brightness was classified into three categories: high brightness, medium brightness and low brightness using the natural breaks method.
Discoloration period0.1413The duration of discoloration period.
Tree characteristics0.3035Crown shape0.1423Tree species were classified into four categories according to the crown shape: ovate, wide ovate, conical and umbrella.
Trunk clarity0.0732Tree species were classified into three categories according to the extent of trunk clarity: high-clarity, low-clarity and unclarity.
Crown width0.0880The width of crown
Leaf characteristics0.2135Leaf shape0.0937Leaf form were classified into three categories according to the leaf form of eight tree species: ovate, obovate and striped.
Leaf area0.0588
Leaf density0.0610We used the biomass ratio of branches and leaves to reflect the extent of dense tree canopy.
Table 2. Descriptive statistics of LAQ indicators (species aesthetic quality, landscape diversity and relief), LAQ and the viewshed of LAQ. Minimum (Min), maximum (Max), standard deviation (SD), coefficient of variation (CV %).
Table 2. Descriptive statistics of LAQ indicators (species aesthetic quality, landscape diversity and relief), LAQ and the viewshed of LAQ. Minimum (Min), maximum (Max), standard deviation (SD), coefficient of variation (CV %).
MeanMinMaxSDCV %
Species aesthetic quality3.032.5250.6822.44
Landscape diversity0.370.1710.1643.24
Relief0.7010.3448.57
LAQ4.532.6970.8418.54
LAQ (Viewshed)4.683.2070.8718.59
Table 3. Area (% of each LAQ) of LAQ in different types of viewshed areas (km2).
Table 3. Area (% of each LAQ) of LAQ in different types of viewshed areas (km2).
Visual AccessibilityLandscape Aesthetic Quality
Very LowLowMediumHighVery High
Very low14.4 (18.2)26.0 (20.7)18.1 (31.7)4.80 (23.0)4.21 (26.9)
Low10.1 (12.8)17.8 (14.2)13.8 (24.3)4.45 (21.3)3.66 (23.4)
Medium6.64 (8.39)10.2 (8.11)6.16 (10.8)2.15 (10.3)1.93 (12.4)
High3.78 (4.78)6.58 (5.25)4.61 (8.05)1.82 (8.70)1.01 (6.45)
Very high0.38 (0.47)1.78 (1.42)2.34 (4.10)0.50 (2.41)0.58 (3.70)
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Liu, J.; Du, J.; Zhang, C.; Bachelot, B.; Yang, Y.; Dong, T.; Wu, Y. Landscape Aesthetics Quality in Subalpine Forests of Eastern Tibetan Plateau Will Greatly Decrease by the End of the Century? Forests 2025, 16, 1804. https://doi.org/10.3390/f16121804

AMA Style

Liu J, Du J, Zhang C, Bachelot B, Yang Y, Dong T, Wu Y. Landscape Aesthetics Quality in Subalpine Forests of Eastern Tibetan Plateau Will Greatly Decrease by the End of the Century? Forests. 2025; 16(12):1804. https://doi.org/10.3390/f16121804

Chicago/Turabian Style

Liu, Junyan, Jie Du, Chenfeng Zhang, Benedicte Bachelot, Yiling Yang, Tingfa Dong, and Yan Wu. 2025. "Landscape Aesthetics Quality in Subalpine Forests of Eastern Tibetan Plateau Will Greatly Decrease by the End of the Century?" Forests 16, no. 12: 1804. https://doi.org/10.3390/f16121804

APA Style

Liu, J., Du, J., Zhang, C., Bachelot, B., Yang, Y., Dong, T., & Wu, Y. (2025). Landscape Aesthetics Quality in Subalpine Forests of Eastern Tibetan Plateau Will Greatly Decrease by the End of the Century? Forests, 16(12), 1804. https://doi.org/10.3390/f16121804

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop