Introduction

Traditional Tibetan villages in western Sichuan are renowned for their distinct cultural landscapes, integral to China’s diverse ethnic mosaic1. These villages epitomize the Tibetan ethos of harmony with nature and embody profound cultural and ethnic significance2,3. Recognized for their unique cultural attributes, these areas are invaluable treasures of global cultural heritage4. However, rapid urbanization and globalization threaten their traditional lifestyles, architectural uniqueness, and cultural integrity5. Although scholars have investigated the sustainability of these cultural landscapes, existing research predominantly focuses on isolated cultural or geographic elements, lacking a systematic approach to examining multicultural landscape characteristics and conducting regional comparisons6,7. Despite their affiliation with different Tibetan tribes, these villages share common characteristics in their cultural landscapes while preserving distinct ethnic and cultural attributes. This interplay between shared features and local variations is central to understanding the coexistence of cultural diversity and unity within the Tibetan landscape. The similarities reflect shared historical and environmental influences, whereas the differences safeguard the unique identities of each tribe, enriching Tibetan culture. To address these gaps, this study investigates villages from different Tibetan tribes in western Sichuan, integrating Geographic Information Systems (GIS) with cultural landscape gene theory. This approach establishes a comprehensive framework for analyzing and preserving these landscapes, highlighting the interplay between spatial and cultural dimensions while elucidating the structural and functional aspects of cultural landscape genes8,9,10. The findings enhance understanding of Tibetan cultural characteristics and patterns, illustrating how shared and distinct features collectively contribute to the resilience and diversity of these landscapes. The study proposes innovative strategies for heritage conservation and provides valuable references for preserving the cultural landscapes of minority regions worldwide, especially amid modernization and environmental challenges.

The concept of a cultural landscape encapsulates the intricate interplay between cultural practices, beliefs, and the surrounding environment. It extends beyond physical elements to encompass intangible cultural factors, including traditions, religious practices, and social behaviors within a spatial context. In contrast to “culture” or “landscape,” which are frequently examined independently, a cultural landscape synthesizes these dimensions to illustrate how communities shape, utilize, and perceive their environment. This integrated perspective underpins the theoretical foundation of cultural landscape gene theory, which deconstructs landscapes into their smallest, most stable, and adaptive components to facilitate systematic study and conservation9,11.

Tibetan villages in western Sichuan are vital to China’s rich tapestry of ethnic minority cultures, embodying a cultural landscape abundant in tangible and intangible dimensions. This landscape encompasses the geographical setting, traditional architecture, folk festivals, religious beliefs, and ethnic arts2. The concept of cultural landscape genes, drawing inspiration from genetics, offers a structured framework for analyzing these elements, identifying their core components and interrelations that shape village cultures11. These “genes” represent the smallest stable units of the cultural landscape, and their preservation is critical to sustaining the authenticity and diversity of Tibetan village cultures.

Recent studies have increasingly emphasized identifying and analyzing the components of traditional village cultural landscapes, with a particular focus on conservation and sustainable development12,13,14. Methodologies such as hierarchical analysis, fuzzy comprehensive evaluation, and Geographic Information Systems (GIS) have played a pivotal role in developing structured models to assess the interplay between natural and cultural resources. These models, rooted in cultural landscape gene theory, have demonstrated their significance in mapping and analyzing the fundamental components of cultural landscapes, facilitating their adaptation to environmental changes and modernization pressures15,16,17.

However, most studies have examined individual landscape components in isolation, often focusing on specific cultural elements, such as architecture or rituals, while overlooking the interconnectedness of these elements16,18,19. For instance, Fan et al. (2023) applied fractal theory to investigate the formation mechanisms of Tibetan villages, providing novel insights into cultural heritage preservation2. Nevertheless, the study primarily focuses on structural patterns, potentially neglecting the dynamic cultural practices that bring vitality to these spaces. Similarly, Wei (2019) integrated landscape gene theory and historical analysis to examine the relationship between architectural landscapes and settlement environments in Tibetan and Qiang villages, revealing how these relationships evolve in response to environmental and social changes20. Nonetheless, this approach is constrained in its ability to capture the complexity of cultural interactions and the diverse influences that shape these landscapes.

While previous studies have made valuable contributions21,22,23, a significant gap remains in comparative analyses of Tibetan tribal villages. Such analyses could illuminate the territorial and systemic characteristics of these landscapes. This study addresses these gaps by systematically applying cultural landscape gene theory to compare the cultural landscapes of eight nationally recognized Tibetan villages representing five distinct tribes24,25. Incorporating methodologies such as gene identification, encoding, classification, and comparison26,27,28,29, this research aims to uncover the intricate interactions among regional landscape features, ethnic cultural networks, and structural elements, thereby proposing a novel framework for understanding and conserving Tibetan cultural heritage30.

This comprehensive approach not only addresses the limitations identified in prior studies but also contributes to the discourse on preserving cultural landscapes in ethnic minority regions23, providing valuable insights into the challenges and opportunities faced by these communities amid modernization pressures31,32,33. Future research could build upon these findings by applying similar frameworks to other ethnic regions, further investigating the dynamic interplay between cultural heritage and modernity.

Methods

Overview of the study area

The Tibetan villages selected for this study are located in western Sichuan Province, including the Garzê Tibetan Autonomous Prefecture, the Ngawa Tibetan and Qiang Autonomous Prefecture, Pingwu County (Mianyang City), and Shimian County (Ya’an City), as shown in Fig. 1. These sites were selected for their cultural characteristics, historical significance, and roles in local societies, as detailed in the Catalog of Traditional Chinese Villages. The study examines eight representative Tibetan traditional villages: Che Ma Village and Re La Village (Kham Tibetans), Xi Suo Village and Song Gang Village (Jiarong Tibetans), Shen Zuo Village (Amdo Tibetans), Min Zu Village and Er Li Village (Baima Tibetans), and Jiang Ba Village (Ersu Tibetans). Located in high-altitude mountainous regions, these villages feature distinctive architectural styles and rich natural and cultural resources34,35. Over the past decade, implementing the rural revitalization strategy has significantly improved the ecological environment and social development in these Tibetan villages.

Fig. 1: Research area.
figure 1

(Fig. 1 was created using ArcMap 10.8, QGIS 3.22.5. The base map in Fig. 1 is sourced from OpenStreetMap contributors and used under the Open Database License (ODbL), available at https://www.openstreetmap.org/copyright).

Full size image

Research approach

The flowchart (Fig. 2) illustrates the research framework employed in this study, which integrates multiple methods to analyze the cultural landscapes of Tibetan villages systematically. The following sections provide a detailed explanation of each step in the framework.

Fig. 2
figure 2

Research process flowchart.

Full size image

Data resource and type

This study adopts a mixed-methods approach, integrating quantitative and qualitative analyses to examine the cultural landscapes of traditional Tibetan villages1,36.

The quantitative analysis primarily utilizes Geographic Information Systems (GIS) with data from GIS Cloud and Map Open Platform, including 30-meter resolution Digital Elevation Models (DEM) and Points of Interest (POI) data. Additional data sourced from the Traditional Chinese Villages website and the Sichuan Statistical Yearbook website quantify the villages’ geographic locations, topographical environments, settlement patterns, facility distributions, spatial usage, and foundational sociocultural information. Qualitative data collected through in-depth interviews, observations, and literature reviews complement the quantitative data by offering insights into the sociocultural dynamics underlying the numbers, providing essential information for constructing the dataset37.

Interviews were conducted from January to December 2022, ensuring comprehensive data collection across seasons and enabling the study to capture potential temporal variations in cultural practices. For qualitative interviews, 24 participants were selected using purposive sampling to ensure representation across community roles, including village elders, cultural practitioners, and local leaders. The criteria for participant selection included: (1) residency in the village for over 10 years, (2) active participation in cultural practices, and (3) knowledge of local traditions. This sampling strategy ensured comprehensive data collection, capturing diverse cultural perspectives while allowing for cross-village comparisons. Informed consent was obtained from all participants, and their identities were anonymized for confidentiality38,39. Interviews lasted between 32 and 110 minutes and were conducted via online video calls or in-person group discussions, depending on participants’ availability and preferences40. This approach enabled a dynamic exploration of topics, including family tradition transmission, local festival characteristics, and the impact of modern rural development on traditional culture.

All interview sessions were audio-recorded with participants’ consent and transcribed for subsequent textual analysis. Data anonymization measures included removing identifiable details and assigning unique identifiers to each dataset (see Table 1). The qualitative data analysis software NVivo was used to facilitate content analysis. Recorded content was systematically coded to identify themes, including community cultural transmission, policy impact, and the interaction between cultural resources and village development41.

Table 1 Thematic classification and coding
Full size table

To reduce interviewer bias and minimize the effects of cultural differences that could distort data interpretation, a culturally diverse team of researchers conducted the interviews42. The team regularly reviewed and discussed findings to challenge and refine thematic interpretations. They also revisited key interview segments to ensure balanced data representation and understanding.

Although the interviews yielded valuable insights, they inherently have limitations, as the perspectives are specific to certain groups and do not represent all relevant communities43. The interviewer’s potential subjectivity could influence question framing and interpretation. Additionally, cultural differences between researchers and respondents may result in misunderstandings44,45. To address these challenges, we integrated survey data with literature sources to enhance the study’s comprehensiveness and supplement quantitative data and observational findings.

Dataset construction and processing

The dataset for this study is constructed through a systematic analysis of both material and immaterial cultural landscape elements. Field surveys, interviews, and literature reviews were conducted to extract 17 variables, capturing and quantifying the cultural landscape characteristics of the villages.

The collected data comprises eight material cultural landscape elements, including village area, road network layout, water system distribution, architectural patterns, and others46. These elements are numerically processed using standardized measures and ratios to enable analysis in subsequent statistical models. The remaining nine immaterial cultural landscape elements, such as religious beliefs, traditional customs, and festival activities, are compiled from literature reviews and interview data to construct the dataset47. Most of these data are text-based qualitative data, which require encoding strategies to transform them into quantitative indicators for statistical analysis; that is, different weights and values are assigned based on the impact level of each immaterial cultural element. We use the StandardScaler from the Sklearn Python preprocessing module, which standardizes features by mean removal and unit variance scaling. This normalization ensures that data of varying scales are compared using a consistent standard. This process improves data operability and analytical precision, providing a solid foundation for subsequent complex analyses and model construction, as shown in Table 2 below.

Table 2 Data set of cultural landscape elements in sample villages
Full size table

Identification and stratification of genetic units

To systematically analyze the data and uncover the cultural landscape genes of Tibetan villages, we applied a gene theory framework previously used by other researchers22,48,49. This approach allows us to identify groups with similar characteristics within complex datasets and pinpoint the cultural landscape gene units to which each element belongs. We identified two types of genes in traditional Tibetan villages: material cultural landscape genes and immaterial cultural landscape genes16,17,50. To facilitate this, we constructed an index system with analytical units such as settlement patterns, architectural features, religious beliefs, and traditional crafts. These indicators encompass multidimensional cultural elements, ranging from village-scale features to festival activities. This structured approach systematically categorizes and clarifies the various components of the cultural landscape of Tibetan villages, providing a comprehensive framework for analysis and conservation efforts (as shown in Table 3). The indicator system proposed in this study serves as a foundational framework for systematically characterizing cultural landscapes. While the indicators are tailored to reflect the interplay between cultural and spatial dimensions, they also allow for adaptation and refinement in future research to accommodate localized or dynamic cultural features.

Table 3 Indicator system for genetic characterization of cultural landscapes
Full size table

Gene encoding

After defining the structure of landscape elements and units, we applied gene theory to encode these elements. This process aims to precisely identify and delineate the key components of the traditional Tibetan village cultural landscapes in Western Sichuan. By converting complex cultural characteristics into quantifiable data, we can more effectively systematize the analysis and understanding of the structure and features of cultural landscapes51. This method helps distill the core elements of the cultural landscape, revealing the distinctive features of Tibetan cultural landscapes and providing clear indicators for their preservation and transmission.

In biology, bases are arranged in specific sequences to form DNA, which constitutes essential materials and coenzymes of life, carrying rich genetic information52. Similarly, in our research, the elements of rural cultural landscapes are regarded as “bases,” combined in various sequences to form landscape gene units, producing diverse cultural landscape phenomena and rural characteristics. Following the principle of like-with-like merging, we applied existing information categorization and encoding techniques to encode the cultural landscape genes of five Tibetan tribal villages. According to the landscape gene encoding model53, the codes are represented using the “English initial + Arabic numeral format.” The coding includes three levels: significant categories (represented by the letters A-Z), middle categories (represented by numbers 1–9), such as \({x}_{1}\), and subcategories (ranging from 1–1 to 1–9), such as \({x}_{(1-1)}\), constructing a chain of village cultural landscape gene information (as shown in Fig. 3).

Fig. 3
figure 3

Cultural gene coding model.

Full size image

This systematic coding process allows us to effectively organize and analyze cultural landscape data, contributing to a deeper understanding of the unique cultural elements and their interactions in Tibetan villages.

Gene classification and comparison

This section employs a combination of Pearson correlation, Principal Component Analysis (PCA), and the Analytical Hierarchy Process (AHP) to examine latent variables and their roles within the cultural landscape of traditional Tibetan villages in Western Sichuan54,55. The aim is to use these methods to identify the fundamental types and impacts of cultural genes and the similarities and differences between the cultural landscapes of different villages56. These methods facilitate understanding the diversity and continuity of cultural elements across various village cultures. Additionally, the insights gained are invaluable for formulating strategies for cultural revitalization and protection and promoting regional economic and social development.

Firstly, correlation analysis explores the specific relationships between these identified elements. This method allows us to quantify the strength and direction of relationships between elements, allowing us to explain their direct correlations or indirect influences. The formula for calculating Pearson correlation is as follows:

$${\rm{r}}=\frac{\sum ({x}_{i}-\bar{x})({y}_{i}-\bar{y})}{\sqrt{\sum {({x}_{i}-\bar{x})}^{2}\sum {({y}_{i}-\bar{y})}^{({y}_{i}-\bar{y})}}}$$
(1)

Where, \({x}_{i}\) and \({y}_{i}\) represent the observed values while \(\bar{x}\) and \(\bar{y}\) are their respective means. The symbol r denotes the Pearson correlation coefficient, which measures the strength and direction of the linear relationship between two quantitative variables. The r value ranges from -1 to 1, where 1 indicates a perfect positive correlation, -1 indicates a perfect negative correlation, and 0 indicates no linear correlation. Correlation analysis is a fundamental tool for examining relationships among variables. It is used to test linear relationships among elements57, providing insights into how different variables interact.

Second, Principal Component Analysis (PCA) is a dimensionality reduction technique used to effectively extract and summarize key variables in data, uncovering underlying structures or patterns58. Calculating eigenvalues and loadings is a critical step in PCA. A high eigenvalue indicates significant variation in data along that direction, suggesting that it contains substantial information and holds considerable importance in the dataset. The magnitude of loadings reflects the contribution of original variables to the principal component, representing their importance in forming the principal component59. The equation for PCA involves the following steps:

$$\left|{\rm{C}}-\lambda I\right|=0$$
(2)

Where C is the covariance matrix, λ is the eigenvalue, and I is the unit matrix. The eigenvalues are the values corresponding to the eigenvectors of the covariance matrix, which this equation can obtain.

$${{Load}}_{{ij}}=\sqrt{{\lambda }_{j}}\times {v}_{{ij}}$$
(3)

where \({\lambda }_{j}\) is the jth eigenvalue and \({v}_{{ij}}\) is the ith element of the jth eigenvector. The loadings are critical to explain the actual meaning of each principal component.

Subsequently, the Analytic Hierarchy Process (AHP) is applied to evaluate the weights of various landscape elements, enabling precise definition of gene types. AHP is a multi-criteria decision-making tool that tackles complex issues by conducting pairwise comparisons of elements across levels, ranking LINGcategorizing extracted elements, and assigning weights to systematically evaluate the relative importance of each cultural landscape gene60. This method enables a more precise understanding and interpretation of the multidimensional characteristics of cultural landscapes in traditional Tibetan villages of western Sichuan61. The weight distribution basis is formulated based on the structure of the cultural landscape gene characteristic index system established in the previous section (Table 4) and integrates information from consultations with folklore scholars and field surveys. Next, a rating matrix A is constructed to quantify the relative importance of various factors. Each pair of factors is assigned a scale value from 1 to 9, reflecting their relative importance62. Then, compute the weight vector W = (\({w}_{1,}{w}_{2,}{w}_{3,}\ldots {w}_{n,}\)), obtained by normalizing the eigenvectors corresponding to the largest eigenvalues of the judgment matrix A with the expression:

$${AW}={\lambda }_{{ma}x}W$$
(4)

where A is the judgment matrix, W is the weight vector, and \({\lambda }_{{ma}x}\) is the maximum eigenvalue of A.

Table 4 Clustering and coding information of cultural landscape genes
Full size table

To ensure the rationality of the assessment process, we further perform a consistency test on the judgment matrix. The consistency index (CI) and consistency ratio (CR) are calculated as follows:

$${\rm{CI}}=\frac{\,{\lambda }_{\max }-n}{n-1}$$
(5)
$${\rm{CR}}=\frac{{CI}}{{RI}}$$
(6)

where \({\lambda }_{\max }\) is the maximum eigenvalue of the judgment matrix, n is the dimension or order of the judgment matrix, and RI is the random consistency index. The consistency of the judgment matrix is considered acceptable only if the value of CR is less than 0.1. This step of consistency testing ensured that our judgments in assessing the relative importance of different cultural landscape factors were robust and credible60.

Additionally, network structure analysis was employed to examine the commonalities and differences in cultural landscape genes. This method analyzes complex interactions among elements from a macro perspective, revealing structural patterns that establish cultural connections and highlight the uniqueness of each gene characteristic across different villages63,64. In this analysis, a network of cultural landscape elements is constructed by defining nodes (i.e., cultural elements) and edges (i.e., connections between components). Centrality analysis evaluates the importance of a node within the network using two key metrics: Degree Centrality, which measures the number of connections a node has, and Betweenness Centrality, which reflects how often a node serves as a bridge along the shortest path between other nodes. The equation is as follows:

$${C}_{D}(i)=\frac{deg (i)}{{\rm{N}}-1}$$
(7)

Here, \({C}_{D}(i)\) is the degree centrality of node i, deg(i) is the degree of node i (i.e., the number of direct connections), and N is the total number of nodes in the network.

$${C}_{B}(i)=\sum _{j < k}\frac{{gjk}(i)}{{gjk}}$$
(8)

Here, \({C}_{B}(i)\) is the median centrality of node i, gjk is the number of shortest paths from node j to node k, and gjk(i) is the number of paths that pass through i among these paths.

We applied community detection algorithms to reveal clusters in the network that represent villages with similar cultural traits. Methods such as modularity optimization were used to evaluate the density of connections within communities, refining the group structures of artistic elements and highlighting their commonalities and differences. This process employs the NetworkX module in Python65.

The integration of the methods has improved the accuracy of classifying the cultural landscape genes of Tibetan villages and enhanced the understanding of their cultural diversity, continuity, and differences. These methods streamline the data processing workflow and establish a systematic and scientific analytical framework. This framework offers robust theoretical and methodological support for dissecting and comprehensively assessing cultural landscape genes and formulating conservation strategies.

Results

Characterization of the genetic units of the cultural landscape

Tibetan villages on the western Sichuan Plateau present a distinctive cultural landscape integrating elements from Han, Yi, Qiang, and Miao ethnic groups. This fusion is reflected in local architecture, art, and traditions, arising from prolonged interaction and adaptation between the community and its environment. These interactions underscore the dynamic processes of cultural gene inheritance, variation, and selection, showcasing cultural diversity and resilience53,66. This section examines the material and immaterial “gene units” of these villages, focusing on spatial patterns, architectural styles, traditional crafts, religious customs, language, and festival traditions. This study aims to extract and analyze these gene units through case studies, revealing their distinctive roles within the cultural landscape67,68. This approach underscores the vitality and resilience of cultural integration, highlighting the complex processes by which these cultural elements are preserved and transformed.

Spatial pattern gene unit

The spatial layout of Tibetan villages on the western Sichuan Plateau is predominantly influenced by topography, road networks, and architectural styles1. Typically situated in high-altitude areas, these villages blend seamlessly into the surrounding local rivers, woodlands, and grasslands. This integration reflects a long-standing symbiosis between humans and nature, highlighting the region’s complex and diverse environmental landscape2. Village streets and alleys, which form the backbone of the village structure, connect residential zones with public spaces, enhancing connectivity and accessibility. Figure 4a illustrates this, showing that Che Ma Village and Xi Suo Village have dense building and population clusters influenced by their road layouts. In contrast, Re La Village and Jiang Ba Village align their road layouts with natural water systems, reflecting minimal impact on the terrain. Similarly, Shen Zuo Village and Erli Village exhibit minimal terrain influence on their road designs. By contrast, Song Gang Village and Min Zu Village feature winding roads adapted to hilly landscapes69. These variations underscore the unique adaptations of Tibetan villages to their geographic settings, highlighting their role in shaping the cultural landscape. This is illustrated by field-photographed settlement patterns, using Xisuo, Jiangba, and Shenzuo as examples (see Fig. 4b).

Fig. 4: Information on spatial patterns and architectural landscape.
figure 4

a Spatial pattern of village buildings and roads(self-illustration); b Sample village settlements(self-timer); c Tibetan residential architecture of different forms and materials(Field collection); d Spatial Location Characteristics of Temples(self-illustration); e The temple building elevation(self-illustration). (The author took photographs in this collection during a field trip. All images are original).

Full size image

Architectural landscape gene unit

Tibetan village architecture is primarily divided into residential houses and temples. It is influenced by local terrain, road networks, and water systems, resulting in complex and diverse spatial layouts70,71. Residential structures are typically “broadly dispersed yet closely grouped in small clusters”. Architecturally, Khampa, Jiarong, and Amdo Tibetans favor towered houses, ranging from simple square forms to complex polygons. In contrast, the Ershu and Baima Tibetans typically adopt Sichuan’s architectural layout’s “enclosed” design. Common building materials, such as rammed earth, slate, and raw timber, are used with varying construction techniques. Roofs are flat, sloped, or hipped, incorporating traditional and modern elements (see Fig. 4c). The typical Tibetan home layout comprises a ground floor for livestock, a second floor for living, and a third floor serving as a prayer hall, reflecting the spiritual hierarchy of gods, humans, and animals. This architecture meets practical needs for living, worship, and animal husbandry while reflecting the esthetic values and social structures of different tribes, highlighting the historical and cultural evolution of these ethnic groups.

Temple architecture plays a central role in Tibetan villages, shaped by Buddhist doctrines and ethnic culture. Temples are typically constructed at higher elevations, often centered around mountains, reflecting the Mandala-inspired Buddhist cosmology and the concept of the three realms72. These temples serve as centers for religious activities and are integral to the community’s cultural life. For instance, temples such as Litang Monastery, Manzhou Monastery, Changlie Monastery, and Danda Lun Monastery are either located next to or surrounding villages or adopt a “temple above, village below” layout (see Fig. 4d). The spatial positioning of temples in Tibetan villages underscores their integration with both daily life and the natural landscape73.

Typically located at elevated sites near mountains, these temples embody Mandala-inspired cosmology and the concept of the three realms, symbolizing harmony between humans, spirituality, and nature. This strategic placement underscores the community’s deep reverence for religious culture and emphasizes temple architecture as a focal point for both spiritual and social life72. Prominent examples, such as the Litang, Manzhou, Changlie, and Danda Lun Monasteries, are often arranged in a “temple above, village below” configuration (Fig. 4d), overlooking or surrounding the villages. This layout seamlessly integrates religious practices with daily life, reflecting the community’s profound respect for nature and spirituality while reinforcing the temples’ role as centers of cultural cohesion73.

Among the Khampa, Jiarong, and Amdo tribes, Buddhist temples exhibit distinctive architectural styles, categorized as Tibetan flat-roofed and Sino-Tibetan hybrid designs (see Fig. 4e). These structures are meticulously constructed with traditional materials, including stone, wood, earth, and brick concrete, prioritizing durability and esthetic harmony. Key architectural components include the main Buddha Hall, scripture halls, monks’ quarters, stupas, and auxiliary buildings. Additionally, decorations vary and feature traditional Buddhist motifs, including the “Buddha Light Vase” and “Eight Auspicious Symbols”11,62,74. These elements are designed to accommodate religious practices and community gatherings.

In villages such as Erli and Minzu (among the Baima Tibetans) and Jiangba (among the Ersu Tibetans), where the Bon religion is predominant, natural elements are venerated as incarnations of spirits. Consequently, these villages lack prominent Buddhist temples and instead feature smaller, widely distributed shrines, such as mountain god temples and land god temples, situated along mountainsides or within fields and courtyards75,76. These settings illustrate the integration of residential and religious spaces, highlighting the unified relationship among spirit, culture, and nature within Tibetan villages.

Traditional art gene unit

Tibetan art forms on the western Sichuan Plateau are integral to the cultural landscape, reflecting a rich tapestry of heritage and customs across tribes and highlighting the unique fusion of multi-ethnic integration77. These art forms encompass a diverse range of practices, including dyeing, weaving, calligraphy, music, dance, and metal casting. Thangka painting is particularly renowned for its intricate craftsmanship and profound religious and mythological themes78,79. These paintings portray religious activities and traditional life scenes, embodying intertribal beliefs and artistic expressions. According to Sanlang Ruo’erwu, a Thangka inheritor from Xi Suo Village, Thangka is both an art form and a crucial component of Tibetan intangible cultural heritage, preserving historical and religious narratives.

Tibetan clothing is designed for the high-altitude climate and reflects the unique traits of each tribe through distinctive styles and decorations. For example, Khampa and Amdo Tibetans wear loose robes adorned with heavy gemstones, reflecting their regional esthetic. In contrast, Jiarong and Ersu Tibetans incorporate tailoring techniques and embroidery from Qiang and Han cultures. Baima Tibetans uniquely adorn their hats with white rooster feathers, symbolizing spiritual protection. These garments are practical for daily use and reinforce tribal identity and esthetic preferences.

Tibetan utensils, furniture, and handicrafts are frequently adorned with auspicious symbols, including the Buddha Light Vase and the Eight Auspicious Symbols80. These designs transcend mere embellishment; they enrich daily life and convey the profound layers of Tibetan cultural expression.

Religious beliefs gene unit

In Tibetan rural areas, religious beliefs form the foundation of village culture, profoundly influencing architecture, art, and daily life81. Temples serve as centers for religious activities and key venues for cultural heritage and community interaction82. The dominant Tibetan Buddhist sects in the Khampa, Jiarong, and Amdo regions are Gelug and Nyingma, whose doctrines significantly influence villages’ spatial organization and public life patterns83.

The religious belief gene in Tibetan culture is characterized by a diverse array of totemic patterns, including Buddhist deities, natural elements, and heroic figures, which symbolize the culture and reflect the fusion of religious and folk traditions78,84. These totems are prominently integrated into the architectural designs and interior layouts of temples, embodying the spiritual essence of religious beliefs and providing a lasting visual representation of faith85.

Furthermore, various religious ceremonies and festivals vividly showcase the diversity of religious cultures and their pivotal role in sustaining community cultural identity. These events strengthen community bonds and act as vital links between the past and present, as well as tradition and modernity.

Folk culture gene unit

The folk culture of Tibetan regions in western Sichuan is vividly expressed through language, script, festivals, and rituals. The Tibetan language, including dialects such as Kham, Amdo, and Jiarong, is pivotal in enriching the region’s cultural tapestry86. The principal Buddhist scriptures, the Kangyur and Tengyur, document over three thousand years of Tibetan cultural history and are written in distinctive Tibetan scripts, Kham and Ume, both of which are recognized as national intangible cultural heritage 87.

Field interviews in Minzu Village and Jiangba Village revealed that, despite lacking a formal written language, the Baima and Ersu Tibetans sustain their cultural heritage through oral traditions and artistic expressions, including songs, dances, and paintings (as illustrated in Figs. 5, 6).

Fig. 5
figure 5

Interview with villagers and folklorists of Crab Snail Fortress, Jiangba Village (Collected on 22nd January,2022).

Full size image
Fig. 6: Interview with Minzu Village Folklorist (Collected on 12th December, 2022).
figure 6

(These photos were captured during the field research interviews).

Full size image

Tibetan festivals and folk activities, such as the Buddha Sunning Festival, Tibetan New Year, Mountain Worship Festival, and Archery Festival, underscore the cultural and religious distinctiveness of the tribes while serving as vital conduits for cultural preservation, reflecting the deep-rooted values and history of the Tibetan people88. These diverse aspects of folk culture are crucial for understanding the cultural dynamics and diversity in Tibetan villages of western Sichuan89. They further highlight the significance of these artistic forms in fostering community cohesion, identity, and continuity90.

Cultural landscape genetic coding and classification

Building on a previously established index system and gene information chain structure18,28, along with gene unit characteristics and a coding structure concept(as presented in Table 3 and Fig. 3), we have classified cultural landscape genes into five major categories, 17 subcategories, 33 minor categories, and 98 landscape unit components, as detailed in Table 4.

This classification employs a top-down hierarchical framework, beginning with major categories that encompass broad cultural and spatial dimensions, including spatial patterns, architectural features, traditional arts, religious beliefs, and cultural practices. These categories were selected for their representativeness, stability, and relevance to the cultural landscape of Tibetan villages17,48. Subcategories and unit components were further refined to improve specificity and establish a solid foundation for comprehensive analysis.

The system emphasizes the most significant and representative components, serving as a foundational framework for systematically studying cultural landscapes. It does not exclude the potential influence of other cultural elements but prioritizes key components for systematic study. This framework can be refined and expanded in future research to include additional elements as necessary, accommodating new insights and evolving cultural contexts. (Fig. 7)

Fig. 7: Clustering of cultural landscape gene units.
figure 7

(Fig. 7 illustrates the hierarchical clustering of the cultural landscape’s material and immaterial elements into various groups, reflecting the gene coding structure depicted in Fig. 3. This structure demonstrates how core characteristics of cultural genes are represented and vary across different artistic expressions.).

Full size image

As shown in Fig. 8, Principal Component Analysis (PCA) and clustering algorithms were applied to reduce the dimensionality of mixed data, extract critical cultural landscape gene variables, and group them based on their intrinsic relationships. PCA simplified the data structure, revealed underlying patterns, and explained most of the variation through a few principal components59. Subsequently, cluster analysis categorized the gene units based on the similarity of these components, as depicted in Fig. 8, clarifying the types of cultural landscape gene units. The densely clustered groups in the figure indicate the presence of primary genes with shared core characteristics91. Gene units on the periphery indicate the presence of mixed genes, while those outside the primary clusters may represent new landscape elements shaped by external socio-economic factors.

Fig. 8
figure 8

Clustering of gene components after principal component (PCA) based dimensionality reduction.

Full size image

This analysis identified four types of cultural landscape genes: central, attachment, mixed, and variant. Primary genes are fundamental, defining the main characteristics of the cultural landscape, such as architectural styles, religious beliefs, and traditional arts, which play a crucial role in shaping the cultural landscape of Tibetan villages92. Attachment genes enhance core genes by adding features or depth. Mixed genes represent the blending and interaction of different artistic elements, while variant genes reflect the introduction of new cultural elements, potentially driven by external socio-economic changes93.

Genotypes of cultural landscape elements

This section employs correlation analysis, Principal Component Analysis (PCA), and the Analytical Hierarchy Process (AHP) to examine the five cultural landscape gene units in Tibetan villages of western Sichuan. The objective is to classify and understand the impact of these gene types on the cultural characteristics and identity of the villages. This analysis provides a scientific basis for the cultural preservation and development strategies of these villages.

Initially, the Pearson correlation coefficient (based on Eq. (1)) was applied to explore linear correlations among various cultural variables, which were visualized through a correlation matrix and a heatmap in Fig. 9. This foundational step identifies clusters of gene types and examines their significance within the cultural landscape94, revealing how these elements interact and shape the overall cultural framework of the villages.

Fig. 9: Heat map of correlation matrix between variables in gene units.
figure 9

a Correlation Matrix for Spatial Pattern; b Correlation Matrix for Building Landscape; c Correlation Matrix for Traditional Arts; d Correlation Matrix for Religious Belifes; e Correlation Matrix for Culture Folk; f Correlation Matrix for latent Variable.

Full size image

As shown in Fig. 9, the heatmaps illustrate the relationships between different cultural landscape gene variables, as measured by Pearson correlation coefficients. These variables are grouped into five categories: spatial patterns, architectural features, traditional arts, religious beliefs, and cultural customs. In the spatial pattern group (Fig. 9a), village size shows a positive correlation with road layout and a negative one with building arrangement. Architectural features (Fig. 9b) and cultural customs (Fig. 9e) exhibit strong intra-group correlations, reflecting internal coherence. Religious and artistic variables (Figs. 9c, d) also display notable associations, suggesting deep cultural linkages. Figure 9f integrates all five categories, illustrating a complex network of both positive and negative intergroup relationships.

To investigate the statistical structure of the cultural gene variables and reduce dimensionality, we applied principal component analysis (PCA)62,95. PCA transforms the original correlated variables into a set of uncorrelated principal components that collectively retain as much variance as possible from the dataset96. Each element is associated with an eigenvalue, representing the amount of variance it explains; the larger the eigenvalue, the more significant the component’s contribution97.

Two indicators are central to interpreting the PCA results: the loadings matrix and the explained variance. The loadings matrix quantifies the contribution of each original variable to each principal component, with higher absolute values indicating stronger associations98. Identifying the component with the highest loading for a given variable allows for a preliminary grouping of variables. Explained variance, meanwhile, denotes how much of the dataset’s total variance each component accounts for, as calculated using Eqs. 2 and 399. Components with higher explained variance highlight the principal axes of variation. Together, these indicators clarify the underlying structure and support the classification of variables based on shared statistical features.

As shown in Fig. 10, the statistical structure of landscape variables is revealed through principal component analysis (PCA). In Figure (a), the heatmap presents the loading coefficients of each variable on the first five principal components (PC1-PC5). In Figure (b), the scree plot quantifies the individual and cumulative variances explained by each principal component, highlighting the primary position of PC1.

Fig. 10: PCA results.
figure 10

a PCA loading heatmap(PC1–PC5); b Explained variance byprincipal components.

Full size image

PC1 accounts for 74.51% of the total variance, indicating its centrality in encapsulating the shared structural essence of core cultural elements. Variables related to “Religious beliefs and Religious sects” exhibit strong positive loadings on PC1, corroborating their foundational status within the cultural system. PC2 explains 7.56% of the variance and centers on architectural and functional elements (e.g., building typologies, forms, functions, and ornamentations), embodying relatively stable cultural attributes, thus justifying its preliminary classification as an Attachment Gene. PC3 (5.16% variance explained) and PC4 (3.28%) highlight dynamic, transitional cultural patterns (e.g., traditional art forms, water system configurations, and linguistic practices), reflecting intercultural interaction and adaptive flexibility, which is why they are designated as Mixed Genes. PC5 explains only 1.44% of the variance. It involves elements such as “traditional crafts, totemic worship, and festive” practices, embodying emergent or variant characteristics, which warrant its tentative classification as a Variant Gene. Components beyond PC5 yield negligible variance and lack interpretive clarity, justifying their exclusion from further analysis.

Although PCA can objectively reveal the statistical structure of variables and the primary sources of variation, as a purely data-driven method, it may not accurately reflect the importance of each variable in terms of cultural values100. Some variables may contribute minimally to statistical variance yet possess high cultural value. To address this gap, we supplemented the PCA with an expert-based evaluation using the Analytic Hierarchy Process (AHP). This method incorporates subjective judgment into a structured decision-making framework101. AHP facilitates the integration of qualitative insights with quantitative analysis by constructing a pairwise comparison matrix, allowing for the assignment of weights based on expert assessment12. This approach enables the inclusion of culturally meaningful but less statistically prominent variables, offering a more holistic evaluation.

Accordingly, we developed an AHP framework encompassing three evaluation dimensions. Experts were invited to assign weights to each dimension based on their knowledge and experience. The dimensions include: (1) Cultural functionality, which refers to the role of variables in cultural inheritance and functional performance, reflecting their significance and value as cultural elements102; (2) Adaptability, which refers to the ability of variables to adjust and inherit in response to changes in the environment and times, demonstrating their strength in maintaining existence and influence in different contexts103; (3) Statistical relevance, which refers to the degree of importance of variables in statistical analysis, such as their performance in PCA, including the variance contribution and loading levels of the corresponding principal components104.

Following the evaluation framework mentioned above, we conducted a weighted summary of the expert scores for each variable across three dimensions (cultural functionality, adaptability, and statistical relevance), ultimately obtaining a comprehensive weight for each variable (based on Eqs. 5–6). Experts independently rated all variables on a scale of 1–9, and the AHP method was used to calculate the overall scores, reflecting the importance of each variable in terms of the integrity, continuity, and adaptability of the cultural landscape system. As shown in Table 5, “water system layout” (0.0712), “road network layout” (0.0690), and “architectural decoration” (0.0669) received relatively high weights, indicating that their cultural value is more prominent in the comprehensive assessment across the three dimensions.

Table 5 Combined factor analysis and weighting of genotypes
Full size table

Based on the principal components and the weights of their variables, we determine and classify the types of effects of various cultural gene variables97,100, as shown in Fig. 11. Variables that exhibit high loadings on a single principal component and rank highly in expert evaluation are considered Main genes. These variables contribute substantially to cultural characteristics, statistically reflecting a dominant factor, significant variance explanation, and strong functional or symbolic relevance. Conversely, variables with more evenly distributed loadings across multiple components, implying influence from several latent factors, and lower expert-assigned weights are categorized as Mixed genes. These typically serve as supportive or transitional elements within the broader cultural system.

Fig. 11
figure 11

Criteria for determining genotypes of landscape elements (self - drawn).

Full size image

From above, we conducted an integrated classification of cultural gene types for each variable, as presented in Table 5 and Fig. 12. For instance, religious belief and religious sect are both associated with the first principal component (PC1), which accounts for 74.51% of total variance, indicating a strong statistical correlation. However, in the AHP evaluation, religious belief received higher scores in dimensions such as cultural functionality and institutional stability, marking it as a Main gene for its foundational role in shaping spatial logic, institutional frameworks, and spiritual order in Tibetan villages. In contrast, although a religious sect shares a similar statistical profile, its cultural significance is more localized and historically contingent, leading to a lower expert-assigned weight and its classification as an Attachment gene.

Fig. 12: Interpretation of genotypes of variables based on PCA and AHP.
figure 12

a Variance explained by principal component(PCA); b Combined weight values redefine the genotype of the variable.

Full size image

“Religious belief”, as a cultural core, not only informs spatial generation mechanisms but also influences architecture, community structure, and ritual practices. Likewise, variables like “Building features and traditional arts” are also categorized as Attachment genes, reflecting their supportive roles in everyday practices and in sustaining cultural diversity.

“Language and text”, though showing only moderate statistical loadings in PCA, was assigned a higher cultural value in the AHP due to its heritage significance and adaptability to sociocultural transitions. It was thus identified as a Mixed gene, representing continuity and flexibility in intercultural communication. Similarly, “water system layout” exhibited notable loadings across multiple components and was initially categorized as a Mixed gene. Although its expert evaluation score was moderate, its enduring influence on village morphology, irrigation systems, and spatial governance led to its reclassification as an Attachment gene.

By contrast, “village size” was marked as a Variant gene due to its weak statistical correlation and high sensitivity to external socio-economic and environmental fluctuations12,101. As illustrated in Fig. 12b, some variables, including “water system layout”, “language and text”, and “traditional art forms”, shifted in gene classification based on their final weight values, highlighting the evolution from purely statistical categorization toward a more holistic evaluative framework.

This classification correction highlights the dynamic role and interactive structure of variables within the Tibetan cultural landscape. The mixed and variant genes demonstrate the tension between tradition and adaptation, while the stability of the main genes and attached genes provides structural support for the reproduction of village community culture and spatial order. Over time, these cultural genes not only carry historical continuity but also promote cultural renewal, allowing Tibetan villages to retain their artistic essence and cultural identity while responding to external changes.

Through this integrated analytical approach, the study constructs a structured and hierarchical system for classifying cultural landscape variables102. This not only sharpens the accuracy in identifying each variable’s functional role but also offers a solid theoretical and practical base for understanding how a wide range of cultural factors interact. Ultimately, it contributes to a deeper grasp of what sustains the resilience and long-term vitality of rural Tibetan communities103,104.

Commonalities and differences in the cultural landscape of the village

This section utilizes the information gathered from the field study and interviews, as well as the previous exploration of the genotypes of landscape elements, to map the network structure (Based on Eqs. (7–8)) and demonstrate the connections and distinctions between the villages (shown in Fig. 13). Religious beliefs as subject genes are integral to cultural identity, enhancing community cohesion and maintaining social order105. This influence is prominent in the spatial arrangement and architectural layout design during religious activities and traditional festivals, which are crucial for the transmission of tradition and community strengthening. Despite the diversity in religious sects, there is a uniformity in the values expressed through various religious activities, ranging from temple ceremonies to the creation of shrines and prayer rooms in homes106.

Fig. 13
figure 13

Analyzing network structure.

Full size image

Festivals like the “Buddha Sunning Festival” and the Tibetan New Year, although different in execution, share a common spiritual and cultural foundation. These events enhance the stability and continuity of village culture, reinforcing a shared cultural identity. The architectural style also displays significant similarities across villages, as seen in the types of buildings, decorative elements, and functional designs. For example, temples in villages such as Re La, Shen Zuo, and Xi Suo share a foundational faith that influences their spatial layouts. Although architectural details differ, all villages employ adaptations suitable for the high-altitude environment, such as heavy roofing and adequate insulation. These choices reflect a prudent adaptation to complex climates while supporting traditional family and agricultural lifestyles107. The architecture is shaped by geographical, historical, and climatic factors, allowing each village to develop distinct cultural characteristics and evolutionary paths108.

Under the influence of mixed and attachment genes, villages exhibit both diversity and unique traits. Traditional clothing styles vary significantly among them. For example, Tibetan attire in Shen Zuo Village features minimalist decoration and a subdued color palette, contrasting sharply with the vibrant, ornate clothing of the Jiarong Tibetans in Xi Suo Village. Moreover, the layout and modeling of the dwellings in these villages are also unique, reflecting a respect for the natural landscape and geographical environment while also demonstrating functionality and cultural significance109.

Language diversity in the villages also highlights cultural exchange and nuances of independence. All villagers speak Tibetan, yet tribal dialects show remarkable vocabulary and grammar differences due to geographical isolation and historical migration. These dialects offer more than just linguistic variety—they encapsulate unique cultural insights and descriptions of social phenomena and the natural environment, serving as an intangible cultural heritage that enlightens us about historical interactions and communication styles between villages110.

Moreover, socio-economic developments and policy initiatives continue to reshape the cultural landscape of these villages, which needs to be a point of concern in subsequent studies. At the same time, integrating traditional practices with modern influences and infrastructure has led to the emergence of new cultural identities and shaped social and economic dynamics. These changes present challenges and opportunities that require a careful balance between cultural preservation and development111.

This deep dive into the cultural landscape genes of Tibetan villages in Western Sichuan unveils their complex cultural fabric and highlights the critical need to protect and evolve these unique heritages in the face of modernization. This exploration reveals the rich cultural complexity and the ongoing conservation challenges these villages encounter during modernization.

Cultural landscape heritage and challenges in Tibetan villages

Cultural landscapes are a central component of cultural heritage, encompassing rich histories and diverse artistic expressions. These landscapes are vital for education and community engagement, promoting the younger generation’s appreciation of the uniqueness and diversity of local culture. However, they face threats from foreign cultural influences and modern lifestyles that undermine their traditional features and essence112.

Community education programs are crucial in this context. During field visits, interactions with local communities revealed a deep respect for traditional cultural systems, emphasizing the importance of oral histories and artisanal skills vital for preserving cultural continuity. Incorporating these Traditional forms of knowledge transmission into formal education can strengthen community cohesion and preserve unique cultural identities105. Cultural practices, rituals, and festivals foster a sense of identity and belonging, serving as mechanisms for social cohesion and intergenerational communication108,110. Our field research highlighted the prominent role of these cultural expressions in community life, shaping the social fabric and fostering cultural continuity in the face of a changing external environment.

Addressing these challenges requires a stronger focus on the sustainability of cultural landscapes. Field research and data analysis have provided essential insights into the geographical environment, community development, and artistic preservation. In traditional Tibetan villages, natural features such as mountains, rivers, and forests shape the physical layout and communities’ cultural and social structures. Water sources and terrain significantly shape residents’ daily activities, architectural styles, and road networks. These natural conditions underpin the villages’ self-sufficiency and sustainable growth, prioritizing them for preservation113. Our analyses enhance understanding of the complexity and diversity involved in preserving cultural landscapes, enabling the creation of protective measures that balance tradition and modern needs.

Conservation efforts should prioritize the sustainability of the natural environment and its role in supporting traditional practices, thereby providing crucial data for further research. A multifaceted approach is recommended, including the establishment of community workshops, cultural festivals, and the revival of traditional crafts to enhance community engagement and secure policy support114,115. Furthermore, integrating local cultural and environmental research into community education and media outreach raises public awareness of the need to preserve both cultural and natural environments116. Striking a balance between heritage conservation and development is essential117,118. Sustainable practices that respect traditional methods are essential, such as utilizing modern building techniques to preserve traditional structures and leveraging cultural and ecological benefits to enhance local landscapes119,120.

Preserving and developing cultural landscapes in Tibetan villages requires extensive community participation, robust policy support, and regional collaboration. This comprehensive strategy both safeguards and rejuvenates unique cultural heritages121. Studying the cultural genome of these landscapes enables the preservation of Tibetan village cultures’authenticity and the development of strategies to address contemporary challenges.

Discussion

This study integrates gene identification techniques with GIS spatial analysis and extensive field research to systematically investigate the cultural landscape of traditional Tibetan villages in western Sichuan. By categorizing cultural landscape genes into core, attachment, mixed, and variant types, this research establishes a structured and replicable framework enabling multidimensional analysis of cultural landscapes across levels such as units, hierarchies, coding, and categorization122,123. This methodological integration significantly advances the understanding of cultural genotypes and their essential role in preserving the authenticity and diversity of village cultures124.

The findings reveal that various gene types have distinct roles in shaping the cultural landscape. Core genes, such as religious beliefs, serve as the cultural backbone, shaping spatial organization, architectural styles, and community rituals. Attachment genes, such as architectural features and traditional arts, provide stability and enhance the cultural landscape through daily practices and craftsmanship. Mixed genes, such as language and festivals, demonstrate the adaptability of Tibetan culture, reflecting its evolution through cross-cultural interactions. Variant genes, such as village size, illustrate the community’s adaptability to socio-economic and environmental changes. These findings provide a nuanced understanding of the interplay between tradition and adaptation, offering fresh perspectives on the dynamics of cultural landscapes125,126,127.

This research advances the theoretical understanding of cultural landscape genetics by establishing a robust framework that integrates spatial and cultural dimensions. It highlights the essential role of cultural norms and symbols in sustaining community resilience under globalization pressures and proposes practical strategies for preserving cultural heritage. Furthermore, the study emphasizes the need to balance the preservation of tangible and intangible cultural elements to sustain cultural diversity and promote the sustainable development of ethnic communities. Enriching existing methodologies provides a replicable model for analyzing cultural landscapes in other ethnic areas, bridging theoretical insights and practical applications in heritage conservation.

Despite its contributions, this study acknowledges certain limitations. The research scope was limited to a small number of Tibetan villages, which may constrain the generalizability of the findings. A broader survey encompassing more communities and diverse cultural landscapes could yield a more comprehensive understanding. Additionally, the cross-sectional nature of this study does not capture long-term cultural dynamics, such as the impacts of policy changes and economic development on cultural landscapes42,44. Future research should extend the application of this framework to encompass a broader range of villages and integrate additional variables, such as socio-economic, environmental, and policy factors. Longitudinal studies would also be valuable for capturing the temporal evolution of cultural landscapes, offering a dynamic perspective on cultural adaptation and resilience. These efforts will enhance the applicability and depth of the findings, establishing new benchmarks for cultural heritage conservation research114,128,129,130.

Based on a systematic analytical framework, this article explores the role of tangible and intangible cultural elements in shaping and maintaining village identity. It comprehensively depicts the environmental, social, and cultural dimensions of Tibetan tribal villages. It elucidates how these dimensions interact to preserve the integrity and resilience of ethnic cultures, ensuring their continuity amidst modernization. The study provides empirical support for the theoretical development of cultural landscape genetics. Additionally, it establishes a strong foundation for developing more precise and feasible strategies for conserving cultural heritage. These strategies emphasize the protection of cultural heritage and its flexible and sustainable integration into contemporary social development. The findings further confirm that cultural landscape genes play a central role in safeguarding and perpetuating the traditions of ethnic rural communities, highlighting their critical importance in maintaining cultural diversity and long-term resilience.