Constructing Local Religious Landscapes: Spatiotemporal Evolution of Tibetan Buddhist Temples in the Tibetan–Yi Corridor

Abstract

Situated in the mountainous and gorge-ridden region at the junction of the Tibet Autonomous Region, Sichuan Province, and Yunnan Province, the Tibetan–Yi Corridor is home to the Kham Tibetan area, one of China’s three traditional Tibetan areas. Tibetan Buddhism and the establishment of its temples in this region have evolved and propagated from nothing to a diverse landscape since the 8th century. Existing studies, however, have paid little attention to the intricate interplay between the formation of this sacred religious landscape and the specific geographic and sociocultural contexts in which it is situated. By taking temple architecture as a research vehicle, this study begins by extracting spatial data from historical GIS network data resources and 276 local gazetteers of 45 counties in the Tibetan–Yi Corridor. Secondly, it digitalizes and quantifies the geographic information, construction dates, sectarian affiliations, and sizes of 1479 Tibetan Buddhist temples in the region, establishing a database covering four historical periods. Finally, it employs GIS technology to visualize the spatial distribution of these temples, revealing their spatial and temporal patterns and evolution. From a religious geographical perspective, this study reconstructs the historical trajectories and diffusion patterns of the Nyingma, Kagyu, Sakya, Gelug, Jonang, and Bon sects in the Tibetan–Yi Corridor, revealing the complex interplay, succession, and ebb and flow of these sects over time. The research results show that the historical spread and development of Tibetan Buddhism in the Tibetan–Yi Corridor were influenced by a complex interplay of geographical, social, political, and economic factors, including the unique topography of the Qinghai–Tibet Plateau and Hengduan Mountains, the complex interplay of agriculture and pastoralism, the historical influence of dynastic changes and central government policies on border regions, and ancient pilgrimage and trade routes. At the same time, as a multi-ethnic region inhabited by over 20 minorities, including Tibetans, Yi, Qiang, Naxi, and Nu, the Tibetan–Yi Corridor has a cultural identity dominated by religion, which has become an important factor in maintaining multi-ethnic symbiosis throughout its history, highlighting the unique historical status and role of the Tibetan–Yi Corridor in the entire Tibetan Buddhist cultural circle.

1. Introduction

Religious landscapes are tangible manifestations of human religious activities within natural landscapes and a superorganic construct that constitutes cultural landscapes (Isaac 1960; Sauer 1925, pp. 315–50). The immense and enigmatic natural terrain of the Qinghai–Tibet Plateau has significantly influenced the origin, spread, and distribution of Tibetan Buddhism (Feng 2005). Tibetan Buddhist temples serve as the most tangible and significant material representations of Tibetan Buddhist culture within the religious landscapes of the Qinghai–Tibet Plateau. These Tibetan Buddhist temples not only play a role in supporting religious culture and propagating doctrines, but also reveal the interplay between the sacred and the secular (Jordan et al. 1999, p. 71), as well as the spatial relationships between religious culture, the natural environment, and the cultural environment through their physical locations (Kong 2006).

The spatial interaction of religious diffusion and distribution is a core component of religious geography (Stump 2009). Unlike scholars who emphasize the supernatural forces of religion in shaping cultural landscapes (Kong 1990), P. Jackson and others prefer to view landscapes as veils, texts, and objects of gaze, and to incorporate concepts such as symbolism and identity in the study of religious geography, with a particular focus on the construction of locality (Elden 2009). A “place” can be understood as both a specific point on the Earth’s surface and a spatial arena for human activities (Stump 2009). When tracing the significance of religious changes in a specific place (Ivakhiv 2006), time is often introduced as a variable to examine the dynamic interplay between religion and space (Tulasiewicz and Forsman 2022). In other words, there is an empirical research trend that focuses on the spatial order, temporal sequence, and causal mechanisms of regional systems formed by the correlation between religious activities and the geographical environment (Brace et al. 2006). In recent years, the integration of Geographic Information Systems (GIS) has enabled in-depth, data-driven spatial analyses of the temporal and spatial patterns of religious change and its underlying causes (Lu et al. 2019; Zhang et al. 2022).

The Tibetan–Yi Corridor, also known as the “Ethnic Corridor”, is home to the Kham Tibetan area, one of China’s three traditional Tibetan areas1. Its distinctive geography and diverse ethnic makeup set it apart from Ü–Tsang and Amdo (Fei 1980). During the reign of Trisong Detsen of the Tubo Dynasty in the mid-8th century, the eminent Buddhist Master Vairocana, a key figure in the early spread of Tibetan Buddhism, traveled from Ü–Tsang to the Tibetan–Yi Corridor to disseminate Buddhist doctrines and spread them by leveraging the cultural influence of the religious origin (Rogers 1962, pp. 355–68). The propagation of Tibetan Buddhism and the establishment of its temples in the Tibetan–Yi Corridor have evolved from nothing to a diverse landscape. While existing research primarily focuses on the development of individual sects (Wang 2011; Bai 2006) and the influence of sects, doctrines, politics, and culture on the dimensions, style, and color scheme of religious buildings (Feng 2005; Heatwole 1989), few studies have explored the intricate interplay between the formation of the sacred religious landscape in this region and the specific geographic context in which it is situated and a historical and social context marked by the intertwining and evolving development of diverse sects within a multi-ethnic society.

Therefore, this paper utilizes historical Geographic Information Systems (HGIS) to quantify historical geographic processes and develop corresponding spatiotemporal models. By creating historical slices, we can spatially accumulate and overlay the distribution of temples across different periods. By employing visual analysis, we can reconstruct the spatial distribution patterns of Tibetan Buddhist temples and trace the spatial diffusion of different sects within the Tibetan–Yi Corridor. This enables us to examine the relationship between the spatial distribution of religion and the expansion of religious beliefs and practices (Sopher 1967, p. 22), and explore the profound impact of its historical driving forces on the formation of local religious landscapes.

2. Study Area

The Tibetan–Yi Corridor2 is geographically defined as the Hengduan Mountains region in eastern Tibet, Sichuan, and western Yunnan, encompassing latitudes 22° to 32°05′ north and longitudes 97° to 103° east (Fei 1980) (Figure 1). The region has a northwest–to–southeast slope, with an average elevation of 3500–4500 m; With an average altitudinal difference of over 5000 m between ridges and valleys (Zhao 1985, p. 42), it is characterized by a landscape of north–south trending mountains and valleys, dissected by seven mountain ranges and six major rivers (Nujiang River, Lantsang River, Jinsha River, Ya-lung River, Dadu River, and Minjiang River) flowing southward (Figure 2). As a natural geographic corridor, the Tibetan–Yi Corridor has been characterized by east–west barriers and north–south accessibility (Shi and Zou 2013). This unique geographic feature has made it a historical crossroads for multiple ethnic groups. This region is home to over 20 ethnic groups, including Tibetans, Yi, Qiang, Naxi, Nu, Lisu, and Buyi, all belonging to the Tibeto–Burman language family (https://www.zgbk.com/ecph/words?SiteID=1&ID=128399&Type=bkzyb&Preview=false, accessed on 15 April 2024). It is one of China’s most ethnically diverse regions, and one of the best places to find the most authentic and well–preserved examples of ethnic cultures and historical sites (Fei 1980).

The Tibetan–Yi Corridor essentially coincides with the historical Kham Tibetan area of China. While Kham is a traditional region without definitive borders, its core political, economic, and cultural heart is generally considered to be the area stretching from Kangding, Litang, and Garze in Sichuan to Qamdo in Tibet (Wang 2006, p. 2). Given this, the spatial scope of this study follows the historical path of Tibetan Buddhism’s expansion from Ü–Tsang into the Kham Tibetan area (Shi and Zou 2013). To facilitate quantitative analysis, this study has defined its spatial scope based on China’s current administrative divisions. Specifically, the study area includes 11 counties under Qamdo Prefecture in Tibet Autonomous Region, 17 counties under Garze Tibetan Autonomous Prefecture and 11 counties under Aba Tibetan and Qiang Autonomous Prefecture in Sichuan Province, 3 counties under Diqing Tibetan Autonomous Prefecture and 3 counties under Lijiang Prefecture in Yunnan Province. In total, the study covers 5 prefectural–level administrative divisions and 45 county–level administrative units. The study area spans approximately 388,226 square kilometers and is home to roughly 4.337 million residents, including the Tibetan autonomous prefecture area accounted for about 87.1%, the Tibetans with a ratio of about 60.8% (https://www.stats.gov.cn, accessed on 15 April 2024).

By adopting the Tibetan–Yi Corridor, a region that integrates both natural–geographical and cultural–geographical dimensions (Zhao 1985, pp. 403–5; Fei 1980), as the focus of this study, we depart from traditional research methods centered around administrative divisions or single ethnic groups. This shift offers a tangible geographic context for exploring the concept of “historically formed multi-ethnic regions” within the framework of cultural geography (Shi and Zou 2013). Furthermore, adopting the Tibetan–Yi Corridor as the focus of this study makes a significant contribution to having a new understanding of the spread of Tibetan Buddhism in this region and the importance of religious–based cultural identity in fostering and preserving harmonious coexistence among diverse ethnic groups throughout history.

3. Data

3.1. Data Source

The historical geographic information data collected for this study on Tibetan Buddhist temples in the Tibetan–Yi Corridor encompasses three primary categories: basic temple data, natural resource data, and sociocultural environmental data (Tulasiewicz and Forsman 2022). The basic temple data encompasses the temple’s location, historical names, construction date, reconstruction or destruction date (if any), affiliated sect, and documented changes in sectarian affiliation (if any). The natural resource data encompasses digital elevation models (DEMs), slope data, hydrological data, and land use information. The sociocultural environmental data encompasses population data, governance policy data, economic data, and data on significant historical events.

Based on the Chinese government’s official website, China had more than 3800 Tibetan Buddhist temples as of 2018 (http://www.scio.gov.cn/ztk/dtzt/37868/38146/38148/Document/1626659/1626659.htm, accessed on 1 May 2024). However, detailed data on the distribution of Tibetan Buddhist temples at the provincial, municipal, and county levels is generally not publicly disclosed. Furthermore, the fragmented administrative divisions within the Tibetan–Yi Corridor have resulted in a dearth of systematic data that encompasses all temples in the region, it also has posed significant challenges for conducting holistic cultural–geographical studies of Tibetan Buddhist temples in the region (Li 2016, p. 37). And, after more than 1100 years of development, in addition to the currently known temple buildings, there are also numerous historical temple buildings that were either damaged (with sites) or extinct (without sites) (Longzhu 2016, p. 14).

For the aforementioned reasons, our research team has been systematically gathering data since 2020 by employing historical Geographic Information Systems, conducting comprehensive literature reviews, and undertaking extensive field studies. We collected data on 1479 Tibetan Buddhist temples located within the Tibetan–Yi Corridor. Of all the temples, 72.2% were located with a high degree of precision. Field investigations were carried out to refine the location data of 15.5% of the temples that had less precise historical records. For 12.3% of the temples, the coordinates were determined using the geographic centroid of the smallest administrative unit, ensuring precision down to the town, township, or village level. To correlate historical changes in temple names, document comparisons were conducted. To supplement the missing construction dates of temples, we referenced historical documents that mentioned the builders or living Buddhas associated with each temple. The dates of temple destruction and restoration were crucial for pinpointing the various stages of historical development. The data sources are listed below:

(1)

Historical geographic information data: the historical Geographic Information System (Historical GIS) has created an open access database that provides a year-by-year record of China’s historical geographic changes, starting from the earliest documented historical geographic data (from approximately 223 BCE) and continuing up to 1911 CE, and offers researchers a foundational data platform for historical GIS studies. These data were drawn from the China historical GIS (http://www.fas.harvard.edu/~chgis, accessed from 5 October 2020 to 10 July 2024).

(2)

Local gazetteers: these serve as comprehensive encyclopedias of regional history across Chinese dynasties, offering chronologically continuous data on geographical changes (Ji 2011). This study draws on a variety of local gazetteer compilations, including Complete Collection of Chinese Local Gazetteers· A Collection of Sichuan Prefecture and County Gazetteers 中国地方志集成·四川府县志辑and A Synopsis of Research Materials on the Social and Historical Survey of the Kham Tibetan Area 康区藏族社会历史调查资料辑要, in conjunction with the China Local Gazetteers Database, to retrieve a total of 276 local gazetteers from 45 counties across five regions: Qamdo, Garze, Aba, Diqing, and Lijiang. Spanning general, departmental, and specialized categories, these local gazetteers have a long history, dating back to the Tang Dynasty’s Yuanhe County Gazetteer 元和郡县图志.

Table 1. Quantitative distribution analysis of 276 retrieved local gazetteers in the Tibetan–Yi Corridor.

	Total	Time Distribution					Administrative Division Level
		Unknown	Ming Dynasty and Before (Before 1644)	Qing Dynasty (1644–1911)	The Republic of China (1912–1949)	People’s Republic of China (After 1949)	City	County	Township
Qamdo	19	2	–	2	3	12	2	14	3
Garze	102	21	8	14	7	52	28	72	2
Aba	84	17	5	9	18	35	8	66	10
Diqing	25	6	2	5	4	8	9	5	11
Lijiang	46	9	1	4	5	27	17	19	10

presents the temporal and administrative hierarchical distribution of the local gazetteers retrieved for this study.

(3)

Maps and geographic data: In this study, all the geographical processing programs use the GCS_WGS_1984 coordinate system. All maps drawn during the research process, as well as the administrative boundaries and national borders of relevant regions, use maps provided by the Ministry of Natural Resources of the People’s Republic of China (https://www.mnr.gov.cn/, accessed on 10 May 2024) as base maps. The geographic elevation data were obtained from NASA’s SRTMDEM public data with a resolution of 90m and processed by a geospatial data cloud site created by the Computer Network Information Center, Chinese Academy of Sciences (http://www.gscloud.cn/, accessed on 10 May 2024). Relevant topographic, geomorphic, water system, transportation, and other basic geographic data were obtained from the National Geomatics Database (2018 version) and the National Natural Resources and Geospatial Basis Information Repository (https://www.sgic.net.cn/portal/index.html#/Home, accessed on 15 April 2024). Historical maps and historical place names were obtained from the ancient map library, which is part of a historical place name model analysis service system of China historical GIS (https://timespace-china.fudan.edu.cn/FDCHGIS/homePage, accessed on 15 April 2024).

3.2. Data Preprocessing and Classification Coding of Temple Geographic Information

First, the collected historical data pertaining to temples were categorized by attribute. Second, we standardized the temple data using a combination of information coding and visual interpretation (Liu et al. 2010). Third, the terrain data were clipped and projected, and a geographic index based on topography and elevation was used to analyze the influence of topographic conditions on the distribution of Tibetan Buddhist temples.

The purpose of classification coding of temple geographic information is to assign a unique code to each temple that encompasses multiple attributes (Mertel et al. 2021). Through standardized preprocessing and categorical analysis of the collected data, this study found that the “affiliated sect” and “geographic location” of Tibetan Buddhist temples in the Tibetan–Yi Corridor were key characteristics that, respectively, indicated the religious and spatial attributes of these temples. Based on this, a hierarchical classification coding system was designed for this study, consisting of five code segments and 16 code positions (

Table 2. Data label application for five code segments and 16 code points.

Code Segment	Code Point	Example Application
	Name	Aga Temple	Wangdui Temple	Dalu Temple	Qiangbalin Temple	Songzanlin Temple	Zerichui Temple
Attribute Code (A)	Age	798	924	1326	1437	1681	1590
	Scale	4030 m2	75,200 m2	5290 m2	137,100 m2	338,200 m2	6310 m2
	Historical Name	Yadeng Temple	Lakonda Longma	—	—	Guihua Temple	—
Sect Code (S)	Sect	Nyingma	Sakya;	Sakya	Gelug	Gelug	Gelug;
	Sect Conversion	—	originally Bon temple; converted to Nyingma in 1025 and to Sakya in 1276	—	—	—	originally Kagyu; converted to Gelug in 1710
Time Code (T)	Temporal Scale	Stage I	Stage I	Stage II	Stage III	StageIV	Stage III
Location Code (L)	Province (autonomous region)	Sichuan	Sichuan	Sichuan	Tibet	Yunnan	Sichuan
	Region (state)	Garze	Garze	Aba	Qamdo	Diqing	Garze
	City (county)	Xinlong	Dege	Jiuzhaigou	Qamdo	Shangri-La	Litang
	Town (township)	Kailasi	Wangbuding	Dalu	Qamdo	Jiantang	Shangmula
	Village	—	Xuetong	Dalu	—	—	—
Coordinate Code (C)	Longitude	101°04′	98°42′	103°68′	97°19′	99°71′	100°46′
	Latitude	32°47′	31°84′	33°56′	31°15′	27°82′	29°62′
	Altitude	5230 m	3140 m	2478 m	3812 m	3300 m	4120 m
	Slope	26.05°	16.35°	31.92°	14.02°	12.35°	21.41°

provides applications and examples of these data labels).

(1)

The Attribute Code (A) comprises the temple’s name, historical names, construction date, and scale.

(2)

The Sect Code (S) comprises the temple’s six affiliated sects (Nyingma, Kagyu, Sakya, Gelug, Jonang, and Bon3), and documented changes in sectarian affiliation. Sects are a primary criterion for categorizing Tibetan Buddhist temples, with each temple affiliated with a specific sect. Beyond the four major sects—Nyingma, Kagyu, Sakya, and Gelug—the Tibetan–Yi Corridor, for historical reasons, also houses a considerable number of Bon temples (Bai 2006), as well as Jonang temples, which have disappeared from Tibet and have mainly survived in the Tibetan–Yi Corridor (Yang 2019).

(3)

The Time Code (T) comprises four historical stages, as outlined in

Table 3. Four temporal stage indices.

	Temporal Scale	Description
Stage I:	8th century–1246	From the mid-8th century, when the eminent Buddhist Master Vairocana entered the Kham Tibetan area to disseminate Buddhist doctrines and establish temples4, until just before the establishment of the Sakya local regime.
Stage II:	1247–1367	From the establishment of the Sakya local regime (when Tibet was incorporated into the Yuan dynasty) until the end of the Yuan dynasty.
Stage III:	1368–1643	The Ming dynasty
Stage IV:	1644–1911	The Qing dynasty

. The time period spans from the mid-8th century, when Tibetan Buddhism was introduced, to the end of the Qing Dynasty. By combining the historical spread of Tibetan Buddhism in the Tibetan–Yi Corridor with the ebb and flow of sects influenced by central government support from different dynasties, this study employs chronological categorization and historical periodization (Min and Zhang 2024).

(4)

The Location Code (L) provides details about the administrative divisions where the temple is situated, including the specific province (autonomous region), district (prefecture), city (county), town (township), and village.

(5)

Coordinate Code (C) provides the temple’s geographic coordinates (longitude and latitude), elevation, and slope.

3.3. Establishing a Historical GIS Spatial Database of Temples

A historical GIS spatial database for Tibetan Buddhist temples in the Tibetan–Yi Corridor was constructed based on a Geodatabase data model by importing preprocessed point data of temples and related geographic feature data. This study treats the historical information of 1479 temples, categorized as extant, ruined with remains, and completely vanished, as a point feature within a geographic information system (Mertel et al. 2021). All thematic data, vector data, and raster data are unified and stored in a relational database (Lu et al. 2019). The data are categorized and stored in specific datasets according to their structure and geometry, including Feature Datasets, Raster Datasets, and TINs. A coding system is used to link non-spatial attributes, like sects and time periods, to the thematic data of the temple (Cascon-Katchadourian and Alberich-Pascual 2021). The textual data are organized and stored in a Table dataset, serving as additional attributes for the temple’s thematic data attribute table (Guo et al. 2024). Image data are stored as Raster Datasets and linked to the temple’s thematic data using a naming convention (Zhang et al. 2022). For instance, the image “N138–G017–230–001” is associated with Litang Temple, a Nyingma sect temple in Garze Prefecture, where “N138–G017–230” is the temple’s code and “001” is the sequence number of the image. This facilitates the coding and linking of thematic data, attribute data, and image data.

4. Methods

4.1. Nearest Neighbor Index

The nearest neighbor index (NNI) is a geographic measure used to assess the degree of clustering among different datasets. In this study, it is applied to objectively reflect the spatial distribution patterns (random, dispersed, or clustered) and spatial relationships among the temples belonging to six different sects in the Tibetan–Yi Corridor. The calculation method is based on the ratio of the observed nearest neighbor distance to the expected nearest neighbor distance. The formula is as follows:

$$R={\displaystyle \frac{{\bar{r}}_1}{{\bar{r}}_E}}=2\sqrt{D}\times {\bar{r}}_1$$

(1)

In the equation, R represents the nearest neighbor index, ${\bar{r}}_1$ denotes the average distance between nearest neighbors of distance r₁, ${\bar{r}}_E$ signifies the expected nearest neighbor distance, and D indicates the point density. When R = 1, indicating ${\bar{r}}_1$ = ${\bar{r}}_E$, the point features exhibit a random distribution. When R > 1, meaning ${\bar{r}}_1$ > ${\bar{r}}_E$, the point features exhibit a dispersed distribution. When R < 1, meaning ${\bar{r}}_1$ < ${\bar{r}}_E$, the point features exhibit a clustered distribution. A smaller R value implies a higher degree of clustering. The calculation for ${\bar{r}}_E$ is as follows:

$${\bar{r}}_E={\displaystyle \frac{1}{2\sqrt{n/A}}}={\displaystyle \frac{1}{2\sqrt{D}}}$$

(2)

In the equation, A denotes the area of the study region and n denotes the number of point features within the region.

4.2. Dot Density Estimation

Dot density estimation (DDE) is used to fit the spatial clustering patterns and hotspot regions of Tibetan Buddhist temples in the Tibetan–Yi corridor. By treating all temples as vector points in space, we use the point density analysis tool in ArcGIS 10.8 to create a grid with 1000 m by 1000 m cells. The point density is calculated by dividing the count of points within a neighborhood by the area of that neighborhood. By conducting a linear regression analysis on the spatial variation function of point density, we can derive the density distribution characteristics of temples. The formula used is as follows:

$$F(x)={\displaystyle \frac{1}{nh}}{\sum }_{i=1}^n k\left[{\displaystyle \frac{d\left(x-x_i\right)}{h}}\right]$$

(3)

In the equation, F(x) denotes the DDE value at point x, n denotes the number of temples of various sects, k denotes the point function, d denotes the dimensionality, and (x − x_i) denotes the distance between the estimated point x and the sample point x_i. By calculating the density contribution of each sample point within specified grid cells using a density function, we find that the DDE value generally increases with proximity to the central sample point, indicating a more clustered distribution of temples.

4.3. Standard Deviation Ellipse Analysis

Standard deviation ellipse (SDE) is a statistical method used to accurately measure the diffusion direction and dispersion degree across a spatial distribution of geographic features. Due to historical factors, the spatial distribution of Buddhist temples belonging to six different sects in the Tibetan–Yi Corridor is neither uniform nor consistent. To examine the propagation trajectories, directions, and influences of different sects from a cultural–geographical perspective, it is necessary to quantitatively measure the centrality, extensiveness, and orientation of their temple distributions. Starting from the average center of each sect’s temples, SDE assesses the overall spatial distribution pattern and directional trend of temples by measuring the distances and azimuth between the ellipse’s centroid, x-axis, and y-axis. The calculation is as follows:

$$\mathrm{t}\mathrm{a}\mathrm{n}\theta ={\displaystyle \frac{\left({\sum }_{i=1}^n{\tilde{x}}_i^2-{\sum }_{i=1}^n{\tilde{y}}_i^2\right)+\sqrt{{\left({\sum }_{i=1}^n{\tilde{x}}_i^2-{\sum }_{i=1}^n{\tilde{y}}_i^2\right)}^2+4{\left({\sum }_{i=1}^n{\tilde{x}}_i{\tilde{y}}_i\right)}^2}}{2{\sum }_{i=1}^n{\tilde{x}}_i{\tilde{y}}_i}}$$

(4)

$${\sigma }_x=\sqrt{2}\sqrt{{\displaystyle \frac{{\sum }_{i=1}^n{\left({\tilde{x}}_i\mathrm{c}\mathrm{o}\mathrm{s} \theta -{\tilde{y}}_i\mathrm{s}\mathrm{i}\mathrm{n} \theta \right)}^2}{n}}}$$

(5)

$${\sigma }_y=\sqrt{2}\sqrt{{\displaystyle \frac{{\sum }_{i=1}^n{\left({\tilde{x}}_i\mathrm{s}\mathrm{i}\mathrm{n} \theta +{\tilde{y}}_i\mathrm{c}\mathrm{o}\mathrm{s} \theta \right)}^2}{n}}}$$

(6)

The azimuth θ represents the primary direction of the spatial distribution of features, while the x-axis (major axis) indicates the degree of deviation from the centroid in the primary direction, and the y-axis (minor axis) indicates the degree of deviation in the secondary direction. The greater the difference between the lengths of the major and minor axes (i.e., the higher the oblateness), the more pronounced the directionality and dispersion of the data; conversely, the smaller the difference, the more centripetal the data.

4.4. Thiessen Polygon Analysis

The Thiessen polygon is a polygon that is created by connecting the perpendicular bisectors of the sides of triangles formed by a set of discrete points. The spatial distribution pattern of a point set on a plane, including its influence range, interrelationships, and boundaries with neighboring areas (Li et al. 2022), can be used to accurately reveal the distribution patterns and spatial relationships of geographic features. The patterns of spatiotemporal distribution and evolution of different Tibetan Buddhist temples of various sects in the Tibetan–Yi Corridor can be visualized by constructing a Thiessen polygon for four historical stages, revealing the geographical distribution and spatial relationships of different sects in various historical stages.

5. Results and Analysis

5.1. The Spatial Distribution Pattern

5.1.1. Number

This study collected data on 1479 Tibetan Buddhist temples in the Tibetan–Yi Corridor and found significant differences in both the distribution of temples across regions and the distribution of different sects. As shown in Figure 3, the distribution of temples across the five regions is as follows: Garze Prefecture has the largest number of temples (690, accounting for 46.7%), followed by Qamdo (478, 32.3%), Aba (255, 17.2%), Diqing (38, 2.6%), and Lijiang (18, 1.2%). As illustrated in Figure 4, the distribution of temples among the six sects is as follows: Nyingma has the largest number (508, 34.3%), followed by Gelug (428, 28.9%), Kagyu (194, 13.1%), Sakya (177, 12.1%), Bon (141, 9.5%), and Jonang (31, 2.1%).

5.1.2. Density

Using dot density estimation (DDE), this study visualizes the degree of spatial clustering of Tibetan Buddhist temples and quantifies the proportional distribution of temple density within the region. Geographical factors have a significant impact on the spatial distribution pattern of temples in the Tibetan–Yi Corridor, as shown in Figure 5. Temples tend to cluster along watercourses. The spatial clustering of temples within the corridor formed by the valleys of six major rivers is becoming more pronounced. Multiple density cores are connected to form a band-like “cluster” along the river, which serves as a major center for religious activities in the region. The main religious center is located along the Jinsha River, with secondary centers along the Lantsang, Ya-lung, and Dadu Rivers.

Table 4. Weighted statistics for spatial distribution of dot density of Tibetan Buddhist temples at the county level.

Dot Density Ranking	Country, State, Province/Autonomous Region	Adjacent River	Dot Density	Number of Temples
1	Dege, Garze, Sichuan	Jinsha River, Ya-lung River	0.0083	71
2	Jiangda, Qamdo, Tibet	Jinsha River	0.0078	71
3	Chaya, Qamdo, Tibet	Lantsang River	0.0077	65
4	Gongjue Qamdo, Tibet	Jinsha River	0.0071	62
5	Malkang, Aba, Sichuan	Dadu River	0.0062	60
6	Shiqu, Garze, Sichuan	Jinsha River, Ya-lung River	0.0059	76
7	Xinlong, Garze, Sichuan	Ya-lung River	0.0055	59
8	Baiyu, Garze, Sichuan	Jinsha River	0.0054	57
9	Garze, Garze, Sichuan	Ya-lung River	0.0050	54
10	Dingqing, Qamdo, Tibet	Nujiang River	0.0047	52
11	Qamdo, Qamdo, Tibet	Lantsang River	0.0044	57
12	Litang, Garze, Sichuan	Ya-lung River	0.0041	50

presents the dot density ranking of temples at the county level. Dege County has the highest density (0.0083), followed closely by Jiangda County (0.0078) and Chaya County (0.0077). Figure 6 shows that Dege, Jiangda, Gongjue, and Baiyu counties situated along the Jinsha River; Chaya and Qamdo counties situated along the Lantsang River; Dingqing County situated along the Nujiang River; Shiqu County situated in the upper reaches of the Ya-lung River, Garze and Xinlong counties situated in the middle reaches; Yajiang County situated in the lower reaches; Aba and Malkang counties situated in the upper reaches of the Dadu River; and Danba County situated in the middle reaches of the Dadu River exhibit clear spatial polarization. Additionally, smaller clusters are evident along tributaries such as the Dingqu, Xianshui, and Duke Rivers. Temples are more dispersed outside the clustered river valley areas mentioned above, with considerable fluctuations in their numbers. The spatial distribution patterns of temples are shaped by geographical factors. The distribution of Tibetan Buddhist temples in the Tibetan–Yi Corridor shows a decreasing trend from west to east and from north to south. This trend is particularly evident in the eastern part of Aba Prefecture and Diqing and Lijiang in the south, as these locations are furthest from the religious origin of Ü–Tsang.

5.1.3. Direction

Based on the overall distribution of Tibetan Buddhist temples in the Tibetan–Yi Corridor, standard deviation ellipse analysis was conducted to explore the distribution of six sects. The results are shown in Figure 7. Based on the diffusion direction of the standard deviation ellipse, the geographic distribution of the six sects exhibits significant directional characteristics, mainly manifested in two dominant directions: west–east (azimuth θ between 2.3° and 10.6°) and northwest–southeast (azimuth θ between 38.1° and 59.2°). In terms of the area of the standard deviation ellipse, the distribution area of temples in the upper reaches of the six major rivers is approximately twice that of other regions. Based on the centroid coordinates of the standard deviation ellipses, the centroids of Nyingma, Sakya, and Bon temples are all located near Xilong County in central Ganzi Prefecture today, the centroid of Kagyu is located in Gongjue County, Qamdo Prefecture, the centroid of Gelug is located at the junction of Baiyu and Litang Counties in Ganzi Prefecture, and the centroid of Jonang is located in Rangtang County, Aba Prefecture. With the exception of the Jonang sect, the centroids of the other five sects are relatively concentrated. There is a significant overlap in the spatial distributions of temples belonging to different sects. This demonstrates that, historically, all Tibetan Buddhist sects have developed fully throughout the Tibetan–Yi Corridor and exhibit strong spatial correlations with each other.

In terms of dispersion, Jonang temples exhibit the highest level of concentration (standard deviation ellipse oblateness of 0.17), with their distribution limited to Rangtang, Malkang, and Aba counties in Aba Prefecture. The Gelug temples exhibit the second highest level of concentration (oblateness of 0.18), with their distribution concentrated in the middle reaches of the Lantsang and Dadu Rivers, from northwest to southeast. Sakya temples exhibit the widest spatial distribution (oblateness of 0.25), extending across the region from west to east. Most Kagyu temples (oblateness of 0.32) are found in the Nujiang and Ya-lung river basins, following a northwest–southeast pattern along the rivers. Despite being the most numerous, Nyingma temples exhibit a moderate level of spatial clustering (oblateness of 0.39). Their distribution is primarily concentrated in the northern region, extending from west to east along the upper reaches of the Lantsang and Dadu Rivers. Bon temples exhibit the widest spatial dispersion (oblateness of 0.62), and their distribution is primarily oriented in a west–east direction from Qamdo to Aba.

5.2. Temporal Evolution Characteristics

5.2.1. Four Historical Stages

Table 5. Weighted statistics of the number of Tibetan Buddhist temples in four historical stages.

	Time Scale	Scale Length (Year)	Number of Temples	Proportion	Construction Density
Stage I	8th century–1246 AD	Approximately 500	258	17.4%	0.24
Stage II	1247–1367 AD	120	244	16.5%	0.96
Stage III	1368–1643 AD	275	451	30.5%	0.45
Stage IV	1644–1911 AD	267	526	35.6%	0.59

shows a general increasing trend in the number of Tibetan Buddhist temples in the Tibetan–Yi Corridor over time. From the mid-8th century, when Master Vairocana entered Kham to disseminate Buddhist doctrines and build temples, until the establishment of the Sakya local regime in the 13th century (Stage I), the number of temples increased from 258 to 1479 by the end of the Qing dynasty (Stage IV), with an average growth rate of approximately 15.4% across the four historical stages. Of these, the Qing dynasty (Stage IV) witnessed the highest number of temple constructions (526) and the greatest construction density (0.0043); The Ming dynasty (Stage III) ranked second in terms of the number of temples built, with a total of 451. Despite having the shortest duration of 120 years, from the establishment of the Sakya local regime to the end of the Yuan dynasty (Stage II), this period saw the highest construction density of 0.96.

Figure 8 analyzes the distribution changes of Tibetan Buddhist temples from one point to an entire area and the shift in hotspot regions from west (Ü–Tsang region) to east and south in the four historical periods. At the beginning of Stage I, there were only 12 early temples established by Master Vairocana (Dzapa 1990) within the region. The average nearest neighbor distance (${\bar{r}}_1$) was calculated to be 116.4 km, and the nearest neighbor index (R) was 1.37, indicating a dispersed distribution of temples across 11 counties in present-day Qamdo and Garze. By the end of Stage I, the 258 temples began to exhibit a clustered spatial distribution pattern (R = 0.92). As they were still in the early stages of diffusion, the distribution hotspots were primarily concentrated in Dingqing County, located in the upper reaches of the Nujiang River, which is closest to Ü–Tsang. Simultaneously, small-scale clustered distributions were formed in the upper reaches of the Lantsang, Jinsha, and Dadu Rivers. With the eastward spread of Tibetan Buddhism in Stage II, the hotspots of temples shifted eastward to the Jinsha and Ya-lung river basins, exhibiting a homogeneous diffusion pattern from a primary center (R = 1). Starting from Stage III, Tibetan Buddhism entered a period of rapid expansion in the Tibetan–Yi Corridor, with the number of temples nearly doubling compared to the previous stage. There was a further eastward shift in the distribution hotspots, with a concentration in the upper reaches of the Ya-lung and Dadu Rivers in Garze and Aba Prefectures (R = 0.78). By the end of the Ming dynasty, Tibetan Buddhist temples had achieved 100% coverage of the 45 county units in the Tibetan–Yi Corridor. In Stage IV, the hotspots returned to the Lantsang and Jinsha river basins, and the distribution of temples also exhibited multiple hotspots (R = 0.53). Additionally, the number of temples in Diqing and Lijiang, located in the south, increased by 64% and 100%, respectively. The historical pattern of Tibetan Buddhism’s diffusion along the six major rivers of the Tibetan–Yi Corridor, from west to east and north to south, has been established.

5.2.2. Six Sects

The number of temples constructed by each sect is a hallmark of its influence and growth in the region. As shown in Figure 9, Figure 10 and Figure 11, the composition and evolution of different sects across the four historical stages exhibit significant temporal variations.

In Stage I, there were 114 Nyingma (44.2%), 88 Bon (34.1%), 53 Kagyu (20.1%), and 3 Sakya (1.2%) temples, respectively. The Nyingma and Kagyu sects experienced the most significant development. The Nyingma sect was the first to introduce Buddhism to the Tibetan–Yi Corridor. The 12 temples founded by Master Vairocana in the 8th century belonged to the old Nyingma system of esoteric Buddhism (Yang 1994, p. 42) and did not experience significant development thereafter. It was not until the early 12th century, with the establishment of Gatuo Temple in Baiyu County, Garze Prefecture, that the region had its first representative Nyingma temple. As the affiliated temples grew in size, they gradually formed the main center of the Nyingma sect along the Jinsha River in this stage. The Kagyu sect has numerous subsects; one of these, the Karma Kagyu subsect, originated in Qiwuqi, Qamdo. The Phaktru and Tsalpa Kagyu subsects were mainly distributed in Dege, Litang, and Daocheng counties of Garze Prefecture, forming a distribution pattern in which they were relatively clustered on the west bank of the Jinsha River and more dispersed on the east bank. Bon religion, the ancient religion, was first introduced to the Tibetan–Yi Corridor in the 2nd century and then again in the 6th century. After losing the Bon–Buddhist conflict in Tibet in the 8th century, Bon followers were forced to migrate eastward, making the Tibetan–Yi Corridor the largest gathering place for Bon followers outside Tibet (Dhondup 2007, p. 156). This resulted in a clustered distribution pattern, with Dingqing County in the upper reaches of the Nujiang River, Zuogong County in the middle reaches, and Jinchuan County (Yongzhong Lading Temple) in the Dadu river basin serving as the main centers.

Stage II saw a surge in the construction of Sakya temples, which, due to theocracy, accounted for the highest proportion at 38.5% (94 temples). The Nyingma, Kagyu, and Bon temples followed, with 36% (88), 17.4% (44), and 7.4% (18), respectively. During his three journeys through Kham between 1264 and 1274, Phags-pa, the founder of the Sakya dynasty and an imperial tutor of the Yuan dynasty, not only built numerous temples but also converted existing temples like Tagong (originally Kagyu) in Kangding County, Baiya (originally Kagyu) in Dege County, and Gengqing (originally Nyingma) to the Sakya sect. In the early 14th century, the Sakya sect built numerous temples in the Dadu River basin, leading to a clustered distribution pattern in areas such as Dege and Jiangda in the upper reaches of the Jinsha River, and Litang, Yajiang, and Kangding in the lower reaches of the Ya-lung River. During this period, the Sakya sect’s extensive temple construction, facilitated by its political power, significantly suppressed the growth of other sects (Yang 2007). For example, the Nyingma sect experienced a 22.8% decrease in the number of temples, and the establishment of the Jingang Temple in Kangding County in 1273 marked an eastward shift in its development focus.

The most distinctive feature of Stage III was the decline of the Sakya sect and the rise of the Gelug sect. The Gelug sect surpassed all others in terms of the number of temples, which was 180 (40%). The Nyingma sect followed with 130 temples (28.9%), then the Kagyu sect with 57 (12.6%), the Sakya sect with 48 (11.1%), the Bon sect with 18 (3.8%), and the Jonang sect with 18 (3.6%). In 1354, the Sakya regime in Ü–Tsang was overthrown by the Phaktru Kagyu subsect. Originating in Kham, the Karma Kagyu subsect, with the support of Ming emperors, not only flourished in its birthplace, Qiwuqi, but also expanded its influence eastward across the Jinsha River, reaching as far south as Diqing and Lijiang. Statistical data show that by the late Ming dynasty, the Karma Kagyu subsect faced significant challenges due to joint pressure from the Gelug sect and Mongol forces (Wang 2011), resulting in the destruction or conversion of many of its temples. Despite being founded as late as 1409, the Gelug sect quickly rose to prominence, reshaping the temple distribution within the Tibetan–Yi Corridor. After its introduction to the Qamdo region in the early 15th century (Li 2013), its influence never extended beyond the Jinsha River. It was not until the Third Dalai Lama, Sonam Gyatso, converted a Bon temple in Litang to Gelug in 1580 (establishing Changqingchun Keer Temple), that the Gelug sect gained a significant foothold east of the Jinsha River, leading to the development of a dispersed network of subsidiary centers from the Ya-lung River to the lower reaches of the Dadu River. The Jonangpa sect emerged in the Ü–Tsang region in the early Yuan dynasty. With the rise of the Gelug sect, the Jonangpa were forced to either convert or migrate eastward (Wang 2013, p. 210). From then on, the sect was confined to the northern counties of Dartang, Markam, and Aba in the Tibetan–Yi Corridor, and thus had a smaller number of temples and a more restricted geographical distribution than other sects.

Gelug temples accounted for nearly half (248 out of 526, or 47.1%) of all temples established in Stage IV, while other sects, such as Nyingma (176, 33.5%), Kagyu (40, 7.6%), Sakya (32, 6.1%), Bon (17, 3.2%), and Jonang (13, 2.5%), were in a state of recovery or catching up, resulting in multi-sect coexistence. The most significant development in this stage was the widespread dissemination of Gelug and Nyingma Buddhism. The rapid growth of the Gelug sect was largely due to the widespread construction of its own temples, such as Donggu Temple and Sangzhu Temple in Garze, and Jueri Temple and Shouning Templein Luhuo. In addition, the Gelug sect expanded by converting temples of other sects, particularly in southern Garze regions like Batang, Xiangcheng, Derong, and Daocheng, where nearly half of the temples of other sects converted to the Gelug sect in this stage. This was largely achieved through the sect’s political and military influence. The Nyingma sect, which was the first to enter Kham, experienced a construction boom in the early Qing dynasty, with a 35.4% increase in the number of temples compared to Stage III. The three major temples, which were built during the early Qing dynasty, formed a clustered center in the upper reaches of the Ya-lung and Dadu Rivers in the northern region, with Seda County as a significant foothold, except Gatuo Temple. Relying on its traditional regional advantages, the Kagyu sect experienced a resurgence and shifted its focus southward to Diqing and Lijiang. The Bon sect saw a decline in the construction of new temples in its later stages. Before the Qing Emperor Qianlong suppressed Bon (Dhondup 2007, p. 221), Jinchuan was its dominant region. Afterward, its focus shifted eastward to Songpan, the easternmost area.

5.3. Diffusion Patterns

The temporal evolution of the spatial distribution of temples reflects the process of religious diffusion, with the origin and diffusion patterns being the two most significant factors (Park 1994, pp. 1–123). Ü–Tsang served as the cradle of Tibetan Buddhism, which subsequently spread to the Tibetan–Yi Corridor. Figure 12 quantifies the spatial extent, areas of influence, and spatial relationships of temples of different sects in the Tibetan–Yi Corridor using Thiessen polygons for various historical periods to examine the diffusion patterns of Tibetan Buddhism in this region.

5.3.1. Relocation Diffusion

Figure 12a shows that the diffusion pattern of Buddhism into the Tibetan–Yi Corridor in Stage I (8th century–1246) was primarily a result of relocation diffusion (Park 1994, pp. 1–123). By combining the three historical stages of distinguished monks journeying to propagate the dharma, spreading the dharma in “times of persecution”, and monks from Kham returning to their homeland to spread the Dharma, it becomes clear that the diffusion paths closely aligned with the relocation routes of the sect’s followers. Moreover, the spatial correlation among different temple centers was relatively low.

During the reign of Trisong Detsen of Tubo (755–797), Vairocana, one of the “Seven Enlightened Ones” (七觉士, the first ordained monks in Tubo), was exiled to the Jiarong region of the Dadu River Basin due to internal sectarian disputes. Thus, he began to build temples and take on disciples in this region, leaving his footprints in places like Malkang, Jinchuan, Ruoergai, and Lixian (Tchodan 1995, p. 324). It can be said that the spread of Buddhism in the Tibetan–Yi Corridor originated from Vairocana’s personal migration. However, the spatial influence of this diffusion model, which relied on individual relocation diffusion, was relatively weak. By the 11th century, only 14 counties had Buddhist temples (decreasing from west to east: seven in Qamdo Prefecture, four in Garze Prefecture, and three in Aba Prefecture), representing less than one-third coverage, and most counties had only one temple.

Under the influence of the “annihilation of Buddhism” movement during the reign of Langdarma of Tubo (838–842), Buddhism in central Tubo suffered a severe blow. A large number of monks fled to the Tibetan–Yi Corridor and continued to propagate Buddhism (Zhang 2020), forming clustered centers of Buddhist activities in Dege, Dengke (Shiqu), and Xikang in northern Kham. Therefore, Buddhism in Kham in the “times of persecution” not only survived but also experienced sustained and accelerated development as an important refuge, becoming the primary source for the subsequent “Eastern propagation” (下路弘法) to revive Tibetan Buddhism. This phenomenon is precisely the relocation diffusion pattern triggered by the cultural transfer caused by Langdarma’s annihilation of Buddhism (Shi 1996).

A notable feature of the post-revival period of Tibetan Buddhism, which began in 978 AD, was the formation of various sects. The 11th to 13th centuries also marked a prosperous period for the development and spread of Tibetan Buddhism (Ga 2001, pp. 1–5). In the early 12th century, a monk originating from Kham, Karmapa Deshin Shekpa, brought the Nyingma sect back to Kham. In 1160, Gatuo Temple, the ancestral temple of the Nyingma sect, was built in Baiyu County. Supported by tribal leaders, 114 Nyingma temples were subsequently constructed in various regions. As shown in Figure 13a, a diffusion pattern with a relatively clustered east–west distribution (oblateness of 0.49) was formed in the upper reaches of the Nujiang River, the middle reaches of the Jinsha River, and the upper reaches of the Dadu River. In the same stage, three monks from Kham, Songqinba, Pamdrukpa, and Caiba Zhunzhuzha, introduced three different subsects of the Kagyu sect to Kham. They successively built 53 temples, including Karma Dansa Temple (the ancestral temple of the Karma Kagyu sect) in Qiwuqi, Bapang Temple (Figure 14, with its affiliated temples later spread across the Tibetan–Yi Corridor) in Dege, and Lenggu Temple in Litang. As shown in Figure 13b, the early spread of the Kagyu sect exhibited a strong tendency to propagate in a northwest–southeast direction (oblateness of 0.72), characterized by a large geographical span and a high degree of dispersion. It is evident that the endogenous relocation diffusion pattern of Kham-origin monks returning to their homeland to preach provided the talent pool and local connections necessary for the early development of Tibetan Buddhism in the Tibetan–Yi Corridor. By the end of Stage I, the coverage rate of county-level units within the Tibetan–Yi Corridor had reached 77.8%.

5.3.2. Expansion Diffusion

From Stage II to Stage IV (1247–1911), the primary pattern of diffusion of Tibetan Buddhism in the Tibetan–Yi Corridor was characterized by continuous spatial expansion (Park 1994, pp. 1–123). Figure 12b–d demonstrate an increasing trend in the radiation range, influence area, and spatial correlation of temples belonging to various sects in these three historical stages. Each sect, rooted in its own base, continuously expanded to surrounding areas. By mid-Stage III (the 15th century), the coverage rate of county-level units within the Tibetan–Yi Corridor had reached 100%.

Simultaneously, the trend of regionalization for each sect within the Tibetan–Yi Corridor began to emerge from Stage II onwards. The various sects established a series of relatively stable bases by utilizing the power of local tribes in various ways. For instance, the Nyingma sect in Dege and Baiyu, the Kagyu sect in Qiwuqi and along the banks of the Jinsha River, the Sakya sect in Dege and Kangding, and the Gelug sect in Chaya, Litang, and Huoer (present-day Luhuo), all exhibited distinct regional characteristics. These relatively dispersed bases served as the origins of diffusion for various sects in the Tibetan–Yi Corridor. Driven by the influence of their original bases, the power of various sects expanded continuously into surrounding areas, and the spatial scope they occupied became increasingly larger. By Stage IV, the multiple regional groups formed by the temples of various sects, as well as the clustered and homogeneous distribution patterns within these regions, were the result of this expansion diffusion pattern. For example, in 1675, Rinzin Kunsang Jigme founded the renowned Nyingma Temple, Baiyu Temple, in Baiyu County, Garze Prefecture. In the following 100 years, more than 100 affiliated temples were developed in Garze, Aba, and Jiangda of Qamdo (Yang 2007). Due to its exclusive adherence to the Nyingma sect (Figure 15), Seda County in Garze Prefecture has developed a distinctive pattern of expansion diffusion, with highly regional clustering characteristics.

6. Discussion: Driving Factors on the Formation of Local Religious Landscapes

6.1. Geographical Environment

After the establishment of the Tubo Empire in the 7th century, Buddhism was introduced from the Tang dynasty, Nepal, and other places (Li 1993). After undergoing a localization process from the 7th to the 10th century, which involved integration with the indigenous Bon religion and intense competition with local powers, Tibetan Buddhism gradually evolved into a system with distinct regional characteristics in terms of temple organization, scriptures, and temple education. It subsequently spread to the Kham and Amdo Tibetan areas (Yan and Dan 1993, p. 12). Amdo and Ü–Tsang are both on the same geographic plate, the Qinghai–Tibet Plateau. Located in the Tibetan–Yi Corridor, Kham serves as a transitional zone between the Qinghai–Tibet Plateau, the Yunnan–Guizhou Plateau, and the Sichuan Basin, which is characterized by a gradual decrease in altitude. Its unique geographical features have significantly influenced the spread of religion and the distribution of temples.

6.1.1. Rivers

Influenced by its topography, the Tibetan–Yi Corridor, with its natural transportation network formed by the “Six Parallel Rivers” and numerous tributaries, serves as a vital lifeline for local people, facilitating the transportation of goods, and also as a major channel for the inter-regional spread of Tibetan Buddhism. Moreover, with a high concentration of the population located in river valleys, particularly those with convenient transportation, temples are relatively dense, serving as significant centers for religious activities. The significance of rivers in the spread of religion can be clearly seen in the early routes of religious transmission. The 11 counties where the 12 temples were founded by Master Vairocana in the 8th century are all situated along the Lantsang, Jinsha, and Ya-lung rivers. By the late period, rivers flowed through 42 of the 45 counties in the Tibetan–Yi Corridor where Tibetan Buddhist temples were located, accounting for 93.3% of the total. By analyzing the straight-line distance between temples and rivers, Figure 16 demonstrates a strong tendency for temples to cluster in river valleys. A remarkable 95.7% of temples are located within 5 km of a river, with 61.2% being within 1 km and 40.2% within 500 m. The linear nature of rivers has also determined the north–south linear diffusion of religion within the region and the band-like distribution of temples along the rivers.

6.1.2. Mountains

Ninety-eight percent of the Tibetan–Yi Corridor is composed of mountainous terrain. Of the 45 counties where Tibetan Buddhism is prevalent, all but the southernmost ones (Weixi, Lijiang, and Ninglang) are characterized by high-altitude mountainous terrain, mirroring the topography of the religion’s origin. Figure 17 illustrates the spatial relationship between elevation and temple distribution in the Tibetan–Yi Corridor. Analysis of the impact of topographic relief on temple distribution revealed that Stage I Tibetan Buddhist temples were primarily constructed in counties with an average elevation below 2500 m. Notably, 23.5% of counties with an average elevation below 2500 m were found during Stage II–IV. Counties with elevations between 2500 and 3500 m had the highest count, making up 38.2% of the total, while 26.5% of the counties were located above 3500 m. The smallest number of counties were located above 4000 m, making up only 11.8% of the total. Given the rugged terrain, the local population tends to cluster in gentler areas that are suitable for agriculture and settlement, primarily in river valleys around 2600 m above sea level and other relatively flat regions. Therefore, Tibetan Buddhism would initially prioritize regions with larger populations and better economic conditions during its introduction period. In particular, the relatively flat terrain and relatively concentrated population in the northern pastoral areas created good regional conditions for the development of early Tibetan Buddhism. Later, to further expand the influence of its doctrines, the focus of its propagation gradually shifted to regions with more rugged terrain. To overcome the challenges posed by such terrain, people in these regions often congregated in concentrated areas more suitable for farming and daily life. Consequently, these small settlements served as the main centers for the subsequent spread of Tibetan Buddhism in the Tibetan–Yi Corridor.

6.1.3. The Relationship Between Temple Site Selection and the Environment

“The emergence of these lamaseries is attributable to a variety of factors, with geography being the fundamental cause.” (Hu 1941). The size, style, and materials of Tibetan Buddhist temples, as iconic symbols of religious landscapes in Tibetan area, are shaped by the unique topographical conditions of high mountains and valleys, and the interplay of agriculture and pastoralism, while simultaneously reflecting the distinctive characteristics of their respective regions. This showcases the interplay between temples and the surrounding natural landscape.

Numerous temples built in mountains, valleys, grasslands, and flat fields were carefully situated to harmonize with the surrounding environment, and particular attention was paid to their relationship with mountains, rivers, and other natural features. Originating from the Bon religion, the ancient indigenous faith of Tibet, the worship of sacred mountains has endowed mountain ranges with profound symbolic significance. Scattered throughout the Tibetan–Yi Corridor are 18 sacred mountains revered by Tibetans as the abodes of protective deities. Temples built in close proximity to these holy sites are believed to be under the protection of these deities. Rivers are often seen as symbols of purification and nourishment. Building temples near clear rivers symbolizes the nurturing of the human spirit by Buddhist doctrines (Longzhu 2010).

Table 6. Four temple types integrated with the surroundings.

Type	Characteristics	Photo
The mountaintop type	Tibetan ancestors had a long-standing custom of building castles and palaces atop mountains, combining grandeur with strong defensive capabilities. After the implementation of the state religion system, the tradition of building temples on mountaintops continued, and the elevated terrain was fully utilized to underscore the supreme sanctity of Buddhism. Typically, the layout is centered around the Tsokchen Assembly Hall, with monastic residences and other ancillary buildings surrounding it. Examples include Baiyu Temple and Gatuo Temple in Baiyu County and Bapang Temple in Dege County.
The hillside type	Skillfully utilizing the mountainous terrain, the temples are constructed in tiers, cascading upwards to create a harmonious and visually striking landscape, reflecting Buddhist principles of living in harmony with nature. Climatically, this location also minimizes the impact of cold winter monsoons on the plateau. Examples include Changqingchun Keer Temple in Litang County and Songzanlin Temple in Shangri-La.
The river valley type	The layouts of these temples, situated in river valleys, are often flexible and adapted to the contours of the valley. River valley temples are often the economic and cultural centers of their communities, providing economic support to local residents through activities such as agriculture, livestock farming, and handicrafts. Examples include Yaqing Temple in Baiyu County and Tagong Temple in Kangding County.
the nomadic tent type	To cater to the nomadic lifestyle of herders in vast grasslands, portable tent temples were developed, allowing for the easy relocation of religious communities and the practice of Buddhism in various environments. The nomadic lifestyle in Tibetan areas is closely intertwined with the tradition of tent temples. Examples include Gegong in Shiqu County and Laduo Temple in Qamdo County.

categorizes the four main ways in which temples in the Tibetan–Yi Corridor are integrated with their surroundings: the mountaintop type, hillside type, river valley type, and nomadic tent type. The relationship between temples and their surroundings exhibits distinct characteristics: of a grand scale, harmoniously blending with the surrounding natural environment; Built along the contour of the mountain, facing the river, and rising in tiers. The layout mimics the natural contours of the mountain, with surrounding ancillary buildings creating a visual replication of the mountain’s form and emphasizing the central position of the main structure.

6.2. The Intervention of Secular Rulers

The feudal warlords on the Qinghai–Tibet Plateau have historically been marked by a unique blend of secular and religious authority, with a longstanding tradition of the union of church and state (Su 2021, p. 45). During the Tubo dynasty, temples accumulated significant wealth and power, and many monks played active roles in politics. After its formation, Tibetan Buddhism formed close alliances with local powers, creating a scattered network of settlements, large and small, with temples as their focal points. These temples served not only as religious and political centers but also as hubs for agriculture, livestock herding, and handicrafts. Consequently, since Tibet was incorporated into the Yuan dynasty’s central government in the 13th century, the Yuan, Ming, and Qing dynasties all employed religious influence to consolidate their rule and effectively govern the border regions. This involved granting privileges to religious leaders, supporting religious sects, and constructing numerous temples. With the direct intervention of secular rulers and the implementation of a local system combining religious and political authority, Tibetan Buddhism rapidly spread throughout the Tibetan–Yi Corridor.

6.2.1. The Yuan Dynasty: The “Imperial Tutor” System

The Yuan Dynasty instituted the “Imperial Tutor” system. The Outline of the History of China’s Ethnic Relations notes that “the strategy of leveraging religious leaders to maintain control over Tubo or Xifan (present–day Tibet), pioneered by Kuyuk, was further refined and employed on a larger scale by Kublai Khan.” (Weng 2001, p. 278). Upon assuming imperial power, Kublai Khan conferred upon the inaugural Sakya Dharma King, Phags-pa, the prestigious title of “Imperial Tutor”. The Sakya sect’s rapid expansion in the Tibetan–Yi Corridor was largely due to the patronage it received from secular powers. The Yuan dynasty established the Xuanzheng Yuan in 1264 to oversee Tibetan affairs, appointing Phags-pa as its chief administrator (Ngawang 2002, p. 54). The Duogan Si Xuanwei Si Duyuanshuai Fu, a subordinate agency, was tasked with governing the Qamdo and Garze regions. Phags-pa’s frequent travels between Dadu (Beijing) and Ü–Tsang between 1264 and 1274, which invariably took him through Kham, were instrumental in establishing a network of temples that served as the bedrock of his political authority in the region (Chen 2007, p. 103). Motivated by his devotion to the Sakya sect, the Dege Tusi constructed the Sama Temple in Baiyu County. This temple has been revered as the ancestral temple by the Dege Tusi lineage for generations.

The Karma Kagyu sect’s significant influence in Kham secured it a prominent position within the Yuan court, leading to several generations of its lineage holders being summoned by the emperor. The sect’s growing influence was demonstrated through its vigorous construction of new temples such as Zexieduoding Temple in Garze County and Zhata Temple in Batang County. The Sengegang Temple in Muya County was instrumental in the eastward expansion of this sect. As documented in the History of Garze Tibetan Autonomous Prefecture, “the Karma Kagyu sect saw a substantial increase in its influence and power in Kham during the Yuan dynasty; it was primarily during the Yuan dynasty that the Karma Kagyu sect emerged as the most prominent Kagyu sect in Tibetan areas, with the greatest number of temples and followers” (Geler 1984, pp. 116–8).

6.2.2. The Ming Dynasty: Supporting Multiple Buddhist Sects, Including the Construction of Many Temples and the Employment of Numerous Monks

Taking a more rational approach, the Ming dynasty marked a departure from the Yuan dynasty’s excessive reverence for the Sakya sect and diversified its religious patronage, supporting multiple Buddhist sects, including the construction of many temples and the employment of numerous monks. By granting titles to multiple religious leaders and establishing a monastic bureaucracy, the Ming dynasty fostered the development and coexistence of various religious sects in Tibet (Chen 2000). The Ming dynasty maintained the Tusi system in frontier regions, “allowing native chieftains to govern local ethnic minorities”. In Kham, the combination of religious and secular authority created a theocratic system, with the Karma Kagyu sect receiving special favor from the court due to its close ties to the region. The Yongle Emperor of the Ming dynasty bestowed the title of Great Dharma King upon the Fifth Karmapa, Dezin Shepa, of the Karma Kagyu sect in Nanjing in 1406. His subsequent journey back to Qiwuqi, marked by extensive temple building, elevated the Karma Kagyu sect to new heights of influence in the Tibetan–Yi Corridor (Ji 2011). A History of Garze Tibetan Autonomous Prefecture 甘孜藏族自治州史话, “the extent of the Karma Kagyu sect’s influence in the Dege region that the rise of the Sakya sect in the Yuan dynasty coincided with the Dege tusi’s adoption of Sakya Buddhism; the fall of the Yuan dynasty led to the rise of the Karma Kagyu sect in Tibet; due to the Dege tusi’s patronage of the Kagyu sect, several of his family’s temples were transformed into significant centers for this sect.” (Geler 1984, pp. 116–8). The construction of Babang Temple sparked widespread building of Karma Kagyu temples throughout the Dege Tusi’s domain. By the 1950s, Babang Temple still had over 500 monks and nearly 80 affiliated temples. This sect has spread significantly in southern regions, including Batang, Litang, Diqing, and Lijiang, which have become important centers for its propagation (Chen 2000).

The political alliance between the Third Dalai Lama, Sonam Gyatso, and the Mongolian tribes of Qinghai in the 16th century significantly contributed to the rapid expansion of the Gelug sect. The expansion of the Gelug sect into the Tibetan–Yi Corridor was initiated by the Mu clan Tusi of Lijiang (Naxi), who invited the Third Dalai Lama to teach and ordain disciples in his territories of Litang and Batang. He also transformed a Bon temple in Litang into Changqingchun Keer Temple (Figure 18), adding the famous Dazhao Hall and Wu Wang Hall to the complex. Over time, this temple grew to become the largest Gelug temple in the Tibetan–Yi Corridor by the Qing Dynasty’s Qianlong period. The General Annals of Ü–Tsang 卫藏通志 records that “545 li east of Batang lies Litang, home to the Changqingchun Keer Temple. This temple was of considerable size, with the Abbot in residence and over 3000 lamas under his jurisdiction” (Feng 2005).

6.2.3. The Qing Dynasty: A Policy of Land Grants and Patronage and Propagating Religious Doctrine Through the Construction of Temples

To consolidate its rule over the borderlands, the Qing dynasty actively promoted Tibetan Buddhism through policies such as granting fiefdoms, bestowing rewards, and building temples. In 1652, following the establishment of the Gandan Pozhang local regime, the Fifth Dalai Lama led a grand entourage of over 3000 people to Beijing for an audience with the Shunzhi Emperor. By granting him the title of “His Holiness the Great Compassionate Buddha of the Western Paradise, Universal Vajradhara Dalai Lama”, the Shunzhi Emperor formally recognized the Gelug sect as the dominant religious and political force in Tibet. The History of Buddhism 佛教史 states that “following the dissolution of the alliance between the Gelug sect and the Hoshuot Mongols, the Qing court subsequently backed the Gelug monastic order to take control of the Kashag government, resulting in a theocratic system in Tibet where monks and nobles jointly ruled under Qing authority”. (Ren and Du 2006, p. 382). Against this backdrop, the Dege Tusi, having adopted the Gelug sect of Tibetan Buddhism, supplanted the Bon-practicing Baili Tusi to become the paramount indigenous power in the Tibetan–Yi Corridor. At the behest of the Fifth Dalai Lama, Kugi Angong Pengco was dispatched to Kham during the Kangxi Emperor’s reign to undertake a massive temple-building campaign. Backed by the Dege Tusi, the construction of the influential “Thirteen Temples of Huoer” led to a widespread conversion of other sects to the Gelug sect throughout the region (Wang 2011). Concurrently, the central government lent its support to the building of Gelug temples. In 1728, on the order of the Yongzheng Emperor, the Seventh Dalai Lama was escorted to Litang, and a substantial sum of 160,000 taels of silver was allocated for the construction of Hhuiyuan Temple in Gada (present-day Daofu).

Although the Qing government’s dominant policy was “to promote Yellow Sect (Gelug) Buddhism to pacify the Mongols”, it simultaneously offered support to other sects to maintain a balance of power among different groups. For example, the early Qing dynasty witnessed a surge in temple construction by the Nyingma sect, with the construction of three more main temples in cooperation with the Dege Tusi, facilitating the spread of Tibetan Buddhism to southern Sichuan and northern Yunnan. The Dege Tusi’s territory alone boasted over 70 Nyingma temples and nearly 20,000 monks by the end of the Qing dynasty (Li 2013). The alliance between secular powers and upper-level religious power greatly facilitated the spread of Tibetan Buddhism, overcoming geographical constraints and fostering a diverse landscape of Tibetan Buddhism in the Tibetan–Yi Corridor during this period.

6.3. Connection to Ancient Trade Routes and Pilgrimage Paths

The Tibetan–Yi Corridor served as both a north–south corridor and an east–west bridge between the Tibetan region and the central dynasty. Figure 6, Figure 7 and Figure 8 indicate that the upper reaches of the Lantsang, Jinsha, and Ya-lung Rivers in the Tibetan–Yi Corridor have the highest concentration of temples. The phenomenon can be attributed to the cultural diffusion networks established by the region’s historical transportation routes.

6.3.1. The Ancient Tea-Horse Road

The ancient tea-horse road5, while originating in the Tang and Song dynasties, truly flourished during the Ming and Qing dynasties. Historically, tea sold to the Qinghai–Tibet Plateau primarily came from Ya’an in Sichuan and Dali and Pu’er in Yunnan, and was transported through the Tibetan–Yi Corridor to various destinations on the plateau (Shi and Zou 2013). The development of the Sichuan–Tibet and Yunnan–Tibet lines, with their numerous branches, created a transportation and trade network connecting Sichuan, Yunnan, and Tibet, and this network extended into Nepal, Bhutan, and India. Figure 19 illustrates the ancient tea-horse road’s Yunnan–Tibet branch, which travels through Dali, Lijiang, Shangri-La, Deqin, and enters Tibet via Mangkang. It then proceeds to Lhasa via Chaya, Qamdo, Luolong, and Linzhi. The Sichuan–Tibet line branches into two routes from Yaan to Kangding. The northern route, historically known as the “Commercial Road”, passes through Kangding, Daofu, Luhuo, Garze, Dege, and enters Tibet via Qamdo before reaching Lhasa via Qiwuqi, Dingqing, and Naqu. The southern route, historically known as the “Official Road”, served as the main conduit for central government administration in Tibet from the Zhengtong era of the Ming dynasty onward (Zhang et al. 2010). It traversed Yajiang, Litang, Batang, and entered Tibet via Mangkang, meeting the northern route at Qamdo. Zhang Qiqin, a Qing dynasty official, authored The Latest Survey of Tachienlu–Tibet Road 炉藏道里最新考, which recorded the northern route from Kangding to Tibet. “It is heard that besides the southern route from Luding (Kangding) to Tibet, there is also a northern route, also known as the trade route. Traveling north from the pass, one passes through Huoer (Luhuo), Zhuwo (Luhuo), Garze, Zhandui (Xinlong), Dege, and Reya (Qamdo), and then continues through Qiwuqi to Nakeshu and the lands of the Thirty–Nine Tribes, finally reaching Front Tibet.” (Zhang 1906).

The ancient tea-horse road’s Yunnan–Tibet and Sichuan–Tibet lines not only facilitated trade between Tibet and the south and east but also played a crucial role in spreading Tibetan Buddhism to the Tibetan–Yi Corridor. The ancient tea-horse road traversed nearly half the counties in this region, making it a vital corridor for the spread of religion, as evidenced by the high concentration of temples. According to

Table 7. Temple weights in counties along the ancient tea-horse road and the Ancient Beijing–Tibet Road.

Line Name	Number of Counties	County Weight	Number of Temples	Temple Weight
Tea-horse Road (Yunnan–Tibet)	7	15.6%	321	21.7%
Tea-horse Road (North Sichuan–Tibet Route)	8	17.8%	525	35.5%
Tea-horse Road (South Sichuan–Tibet Route)	8	17.8%	421	28.5%
Beijing–Tibet Road	7	15.6%	478	32.3%

, the Yunnan–Tibet line covers seven counties and houses 321 temples, constituting 21.7% of the total. Of the total temples, 62.4% (923 temples) are located in the 14 counties along the northern and southern routes of the Sichuan–Tibet Line. The density cores along the Kangding–Qamdo line overlap with the area that has the most concentrated temples in the region. The junctions of these routes were not only important transportation hubs but also the original entry points for Buddhism. For example, Qamdo County, a key intersection of the north and south Sichuan–Tibet routes and the three Yunnan–Tibet routes, was home to the region’s first temples. Despite its distance from the Ü–Tsang region, Kangding County, the starting point for Sichuan tea into Tibet, was home to Buddhist temples as early as the Tang Dynasty and, uniquely, 2 of the 12 early temples built by Master Vairocana were located in this area.

6.3.2. The Ancient Beijing–Tibet Road

After the Yuan Dynasty established administrative control over Tibet, the central government built 72 large and 10 small postal stations along the ancient Tang–Tubo road, extending from Dadu (Beijing) to Sakya Temple in Shigatse, Tibet. This new network, known as the ancient Beijing–Tibet road, facilitated population censuses and effective governance, and currently connects eight provinces (Pang 1999). According to Garwa Awang Sangbo’s field research on the postal stations along the ancient Beijing–Tibet road, these stations were primarily established to facilitate the return of the Imperial Tutor Phags-pa to Tibet and to ensure the smooth transmission of the Emperor’s imperial edicts and the Imperial Tutor’s decrees. To support these objectives, many stations were located in densely populated areas, which was conducive to the propagation of Buddhism (Zheng and Gyalwa 2021, pp. 3–7). Of the 27 large postal stations along the ancient Beijing–Tibet road located in the three major Tibetan areas, 8 were situated within the Tibetan–Yi Corridor, primarily along major rivers. Three of these stations were located in historic Dege (comprising present-day Dege, Jiangda, Shiqu, and Baiyu counties), and were also crucial factors in the significant development of the Nyingma, Kagyu, Sakya, and Gelug sects in Dege.

The ancient Beijing–Tibet road overlaps with part of the northern route of the Sichuan–Tibet line of the ancient tea-horse road. The two roads generally follow the same direction and meet in Qiwuqi after entering Tibet. It is worth noting that after Litang and Batang, the southern Sichuan–Tibet line traverses the central range of the Hengduan Mountains (the Shaluli Mountain), a particularly arduous journey, before reaching Qamdo County. Historically, the northern route along the Ya-lung River, bypassing the Shaluli Mountain, was more frequently chosen. This made the northern route of the Sichuan–Tibet line of the ancient tea-horse road and the ancient Beijing–Tibet road the primary east–west corridors into Tibet and the most important paths for the spread of Tibetan Buddhism. Combined with the aforementioned tendency of temples being concentrated in densely populated areas along river valleys, the Tibetan Buddhist temples in the Tibetan–Yi Corridor are concentrated in the historic Dege region, and form a dense center of gravity in the upper reaches of the Lancang–Jinsha–Yalong river.

6.4. Cultural Identification and Transmission

The Tibetan–Yi Corridor is primarily inhabited by Tibeto–Burman ethnic groups, whose origins can be traced back to the Hehuang region in northern China. Originating from the divergence of the Tibeto–Burman and Sino–Tibetan language families in the Neolithic era, the ancestors of the Tibeto–Burman people migrated southward along the Tibetan–Yi Corridor, eventually settling in Southeast Asia, including Vietnam and Burma (Shi and Zou 2013). Unlike the Ü–Tsang and Amdo Tibetan areas, which are predominantly Tibetan, the Tibetan–Yi Corridor is a multi-ethnic area where Tibetans make up the majority (60.8%). Historically, the diverse ethnic groups within the region have always influenced each other’s interactions through their varying degrees of shared identity. Moreover, the mountainous terrain and deep valleys of the Tibetan–Yi Corridor have resulted in geographic fragmentation, making it difficult for people within the region to travel and hindering the formation of strong, stable political and economic entities over a long period.

Given the aforementioned issues, the doctrines and rituals of Tibetan Buddhism have served to bridge these divides. The unified spiritual belief and lifestyle, as well as the cultural identity dominated by religious forces, have been instrumental in maintaining the coexistence of these minorities and constructing local religious landscapes. Since the late 13th century, Tibetan Buddhism has expanded beyond its original ethnic boundaries (Ga 2001, p. 16), spreading to multiple ethnic groups such as Qiang, Yi, and Naxi, and gradually covering the entire Tibetan–Yi Corridor. The section on “Customs” in the Overview of Daocheng County 稻城县概况 highlights “the deep reverence for Buddhism among the local people”, stating that “the Kham people have a profound faith in lamas. Education, medicine, rituals, divination, art, and precious objects are all entrusted to the monks.” (Ren 1930).

6.4.1. Temples and Their Influence on Society

Due to the generous support and preferential treatment provided by successive central dynasties, Tibetan Buddhist temples often held vast tracts of land and numerous serfs in local areas. For instance, based on a survey of monastic landholdings in Garze Prefecture in 1958 (

Table 8. Selected survey data of temple-owned cultivated land in Ganzi.

County	Township	Township Arable Land (mu)	Temple Arable Land (mu)	Percent
Dege	Gongya	3975	742.5	18.7%
Dege	Gengqing	3750	1227.5	32.7%
Dege	Baiya	2797.5	1332.5	43.8%
Dege	Maisu	6745	1400	20.8%
Dege	Zhake	7597.5	647.5	8.5%
Baiyu	Baiyu	6090	2737.5	44.7%
Baiyu	Hebo	3837.5	1300	35.4%

) (Ran 1994, pp. 253–4), monastic land in five surveyed townships of Dege County accounted for an average of 24.9% of the total cultivated land in each township, while in two surveyed townships of Baiyu County, the average was 40.1%. Historically, the location of temples and the holding of Buddhist meetings also contributed to the development of towns and markets. Due to the underdeveloped commodity economy in the past, Tibet lacked cities in the strict sense. However, the large gatherings of monks and laypeople during Buddhist meetings led to the formation of local economic, commercial, and residential centers around the temples. The beginnings of towns were thus established, and these gatherings also fostered greater social interaction among the local population. Concurrently, in a society dominated by religious forces, people generally adhered to a uniform, ritualized way of life. For instance, starting from the seventh day of the Tibetan New Year, people would dress up and go to the temples to turn the prayer wheel 300 times, a task that needed to be completed by the evening of the sixteenth day. Starting in April each year, pastoral areas would begin preparations for nomadic herding, while agricultural regions would get ready for spring planting. The nomads would pitch white tents in the wilderness and recite the “Mani Mantra” for 5 to 20 days. Farmers would go to the temple to recite the “Mani Mantra” and turn the prayer wheel, praying for prosperity, abundant livestock, and a bountiful harvest (Zhang 2020). It is evident that the temple, as a large-scale political, cultural, and economic entity, exerted its influence on all aspects of local society and people’s lives.

6.4.2. Temples as Agents of Cultural Diffusion

In the history of Tibet, temples functioned as the primary educational establishments. Temples provided the sole avenue for acquiring literacy, scriptural knowledge, and a wide range of subjects, “from Buddhist philosophy to practical skills”, culminating in a distinctive and complete temple educational system (Ran 1994, p. 278). In particular, the rise of the Gelug sect led to significant improvements in the temple education system. Temples have evolved into vibrant hubs of knowledge and culture, marked by advancements in scripture printing, the arts, and architectural techniques. Tibetan temples function not only as religious centers but also as important institutions for printing and publishing in Tibet. Established in 1731 under the initiative of the 42nd Dege Tusi, the Dege Printing House eventually became one of the most significant and productive printing institutions in Tibet. This printing house produced a diverse range of materials, including not only religious scriptures but also numerous books and treatises on Tibetan astronomy, medicine, literature, and architecture, as well as a vast collection of woodblock prints of Buddhist images. The historical predominance of the Gelug sect in Ü–Tsang resulted in the Potala Palace Printing House and the Shigatse Printing House, two of the three major printing houses, primarily collecting and printing Gelug texts, and printing fewer non-Gelug works (Yang 2007). This is not the case with the Dege Printing House, which, under the supportive and inclusive policy of the Dege Tusi, printed a wide range of scriptures and works from various sects (Zhang 2020), significantly contributing to the dissemination of these sects in the Kham–Yi Corridor.

Monks in Tibetan Buddhist temples serve as both religious practitioners and cultural transmitters, and they have multiple identities as writers, historians, painters, and doctors. Through literature, Tibetan opera, and medicine, among other forms, they have effectively spread religious culture. For instance, Tibetan opera, with its diverse artistic forms, is closely linked to temples. It is frequently performed during Buddhist meetings. Its content is often based on Buddhist stories, and its performance style includes religious elements (Shi and Zou 2013). Therefore, performing Tibetan opera before a large audience of believers strengthens the social status and educational role of religion and deepens the inseparable reliance between the multi-ethnic groups and religions in the region.

7. Conclusions

The Tibetan–Yi Corridor is both a crucial part of China’s traditional Tibetan areas and an independent cultural area of Tibetan Buddhism, representing a secondary central area where Tibetan Buddhism spread from its original location (Sauer 1925, pp. 315–50). By leveraging historical GIS technology, this study offered a quantitative and visual exploration of the spatial distribution of 1479 Tibetan Buddhist temples in the Tibetan–Yi Corridor, intuitively revealed their spatial patterns and temporal evolution, employed a religious–geographical perspective to restore the historical diffusion patterns and development picture of the six Tibetan Buddhist sects—Nyingma, Kagyu, Sakya, Gelug, Jonang, and Bon—in the region, and deeply explored the driving factors and influencing mechanisms of the construction of this local religious landscape. The following conclusions were drawn.

(1)

The spatial distribution of temples is shaped by geographical determinants. The cultural diffusion from the religious origin of Ü–Tsang to the diffusion areas exhibits an attenuation trend from west to east and from north to south. This trend is particularly evident in the eastern part of Aba Prefecture and Diqing and Lijiang in the south, which are furthest from the origin.

(2)

The spatial distribution of temples is clustered in the river valleys of the six major rivers. The corridor space formed by the six major rivers determines that the density core of temples is distributed along the north–south direction of the rivers and concentrates in the river valley areas. The easily accessible river valleys form the important centers of religious activities in the region, creating a pattern of dense distribution centered on the historical Dege area (characterized by a large number of temples, with a high density, a diverse range of sects, and the presence of many ancestral and important temples) in the upper reaches of the Lancang–Jinsha–Yalong river.

(3)

The spatial diffusion of temples from various sects has evolved from a localized to a widespread pattern, and from a homogeneous to a heterogeneous one. Stage I was characterized by a relocation diffusion pattern formed through distinguished monks journeying to propagate the dharma, the cultural transfer in “times of persecution” and monks from Kham returning to their homeland to spread the Dharma. From Stage II to Stage IV, the regionalization trend of various sects began to emerge, forming an expansion diffusion pattern where multiple diffusion sources continuously expanded to surrounding areas. The radiation range of temples, the influenced areas, and the spatial relationships among them are all showing an increasing trend.

(4)

The development of various sects demonstrates strong historical continuity, and the spatial distribution of their temples exhibits significant regional variation. The Nyingma sect established the earliest and most extensive temple network, predominantly in the pastoral regions along the Jinsha River, with significant concentrations in specific regions of Qamdo (Gongjue), Garze (Dege, Xinlong, Baiyu and Seda), and Aba (Jinchuan, Malkang). The Kagyu sect is divided into numerous subsects, with the Karma Kagyu, originating in Qiwuqi, Qamdo, exhibiting a distinctive spatial distribution pattern where the temples were relatively clustered on the west bank of the Jinsha River and more dispersed on the east bank. The majority of Sakya temples were established during the 13th and 14th centuries when the sect held theocratic power, concentrated in Derge and Jiangda in the upper reaches of the Jinsha River and Litang, Yajiang, and Kangding in the lower reaches of the Yalong River. By building monasteries and converting monasteries of other sects, Gelug quickly formed a pattern of concentration and distribution of major and secondary centers in Qamdo, Chaya, and Mangkang along the Lancang River, Ganzi, and Luhuo along the Yalong River, and Aba, Danba, and Jinchuan along the Dadu River. The Jonang sect is now geographically restricted to the three northern counties of Rangtang, Malkang, and Aba, which are the only remnants of its historical presence. The Bon religion, after experiencing early expansion and subsequent decline due to religious conflicts in the 8th century, underwent an eastward migration. Consequently, the Tibetan–Yi Corridor emerged as the primary region for Bon practitioners outside of Ü–Tsang, with significant concentrations in Dingqing, Dege, and Jinchuan.

(5)

The spatial configurations of temples have undergone dynamic transformations over time, shaped by a multitude of driving forces. (a) Geographical factors: The flat terrain and concentrated populations in pastoral areas facilitated the early development of Tibetan Buddhism. Subsequent alliances with political elites and local tribal entities facilitated rapid expansion into high-altitude regions. The human–land relationship and topographical features have significantly influenced the development of four distinct temple types: the mountaintop, hillside, river valley, and nomadic tent types. (b) Political factors: In their efforts to consolidate power and extend their governance to border regions, the Yuan, Ming, and Qing dynasties strategically leveraged religious influence during periods of the intervention of secular rulers in religious propagation. In diverse political contexts, various sects expanded their forces with the help of royal power, or sought development in competition with other sects, which together constituted the diversified pattern of expansion, contraction, and succession in this period. (c) Traffic factors: The spatial distribution of ancient temples, predominantly oriented east–west, was shaped by the convergence of pilgrimage paths and trade routes. The junctions of postal stations and lines along pilgrimage paths and trade routes served as significant focal points for the diffusion of religious beliefs. (d) Cultural factors: Tibetan Buddhism has undergone a process of diffusion from a single ethnic group to multiple ethnic groups. The shared religious cultural identity has been instrumental in maintaining the coexistence of multiple ethnic groups in this region. This accentuates the distinctive historical significance of the Tibetan–Yi Corridor as a pivotal node within the broader Tibetan Buddhist cultural network.

It can be seen that the complicated relations of elements in geographical space have shaped the religious landscape in different fields. In this paper, the spatial and temporal model is combined to carry out a visual reconstruction in the way of historical slices, providing a reference for the narrative method and research paradigm to the field of religious geography. In contrast to traditional Tibetan Buddhist studies that primarily focus on deities and ritual practices, this study emphasizes the complex interplay between the formation of this sacred landscape and its unique geographical, political, and social contexts from an interdisciplinary perspective, and describes the overall evolution process from a macro spatiotemporal narrative dimension. On this basis, we will further develop a temporal sequence forecasting model to quantitatively assess the dimensions and depth of variables that account for the cultural diversity reflected in the architectural styles and forms of temples across different sects.