Study on the Distribution Characteristics and Influencing Factors of Traditional Villages in the Yunnan, Guangxi, and Guizhou Rocky Desertification Area

Abstract

This paper aims to analyze the distribution patterns of the Yunnan, Guangxi, and Guizhou rocky desertification area, and provide efficient protection and development strategies. This region has a complex geographical environment, and it is distinguished by hosting China’s largest ethnic minority population and the highest concentration of autonomous ethnic counties among contiguous impoverished areas, with numerous traditional villages. Thus, it is significant to conduct a comprehensive study of traditional villages within this domain, with a particular focus on their centralized preservation and strategic utilization. This research employed ArcGIS and Geodetector software for a rigorous analysis of the spatial distribution characteristics and influential factors of traditional villages in the Yunnan, Guangxi, and Guizhou rocky desertification area. The key findings can be summarized as follows. (1) The traditional villages in this region predominantly exhibit an agglomerative distribution pattern, with pronounced concentrations in southeast Guizhou and secondary concentrations in Anshun and Guilin. (2) Natural environmental factors, social economic factors, and national cultural factors impact the distribution of traditional villages synthetically by positive, median, or negative correlation. (3) The results of the Geodetector show that, significantly, social economic and national cultural factors exert a more pronounced influence than natural environmental factors, especially population density and intangible heritage quantity. The interaction of multiple factors shows an enhanced trend. (4) From the perspective of formation mechanism, natural environmental factors serve as foundational elements shaping the original distribution pattern; national cultural factors act as dominant determinants, accentuating spatial distribution distinctions across various regions and social economic factors emerge as critical catalysts for the sustainable development of traditional villages. The interaction factors can have a more profound impact. Furthermore, it is expected that this study will contribute to the effectiveness of ecology and economy in this area and more analogous regions.

1. Introduction

Originating from agrarian civilization, villages epitomize the outcome of harmonious integration between human societies and their natural surroundings, thereby constituting the rural social milieu [1]. Traditional villages represent habitats of early human settlements, encapsulating ancestral architectural acumen and boasting profound historical lineage intertwined with cultural wealth. The march of progress, propelled by industrialization, while enhancing the rural living environment through technological advancement, concurrently disrupted the traditional production methods and indigenous lifestyles of rural inhabitants. This perturbation inflicted severe attrition upon cultural and natural reservoirs, precipitating in the erosion of the enduring ecological equilibrium nurtured by traditional villages and a consequent dwindling in their numbers [2].

The advent of the 1982 “Law of the People’s Republic of China” on the Protection of Cultural Relics signified the elevation of Chinese ancient villages and associated cultural heritage to a pivotal conservation status, therefore, through landscape beautification, architectural construction, and the theoretical research of rural areas, threatened villages began to be concerned [3]. Commencing in 2003, China’s Ministry of Housing and Urban-Rural Development embarked on the assessment of historic and cultural villages for inclusion into official registers. This work established a basic frame for prospective protection, development, and research, turning the protection of traditional villages from professional to the national stage [4]. Subsequently, across seven successive iterations, a total of 487 villages were enlisted. In 2012, a watershed decision by the Expert Committee on the protection and development of traditional villages sanctioned the transformation of “ancient villages” nomenclature to “traditional villages”, precipitating an expansive embrace of protected traditional settlements. The State Administration of Cultural Heritage, in conjunction with the Ministry of Housing and Urban-Rural Development, has subsequently consecrated six successive enumerations of traditional village inventories, validated 8155 settlements and catalyzed ongoing safeguarding endeavors. The seminal proposition of the Rural Revitalization Strategy in 2017 lent renewed impetus to the safeguarding of traditional villages. This thrust was invigorated in 2018 when “The State Council promulgated the Strategic Plan for Rural Revitalization (2018–2022)”, where primacy was accorded to agrarian and rural domain development. Encompassing the period 2020–2023, the cooperative efforts of the Ministry of Finance and the Ministry of Housing and Urban-Rural Development are spotlighting the conservation and prudent exploitation of contiguous clusters of traditional villages. This accentuation resonates with the ethos of “contextual responsiveness” underscored in the rural revitalization strategy, signifying that the preservation of traditional villages’ holistic coherence and uninterrupted continuum is accorded salient significance.

The study of cultural relics and architectural sites has long been an integral part of humanity’s engagement with history. The research on traditional villages characterizes cultural heritage to keep the historic culture which symbolizes a unique civilization alive. Connecting with practice, these researchers improved rural sustainability from cultural, agricultural, ecological, and tourism resources which, when probed, provide constant motivation for the city [3]. However, it was not until the 19th century that traditional villages, as a distinct branch within the field of human geography, garnered dedicated attention from the academic community. In the 20th century, research on traditional villages evolved from a primarily physical geography focus to encompass a broader examination of societal aspects. Scholars began emphasizing internal motivations, spatial-behavior relationships, and other multidimensional aspects [5,6,7]. Over time, the study of traditional villages has grown into a multidisciplinary field, incorporating geography, sociology, architecture, ecology, and more. Extensive foundational research has paved the way for endeavors such as cultural value assessment, preservation efforts, and tourism development. During this time, the remaining economic situation of traditional villages began to improve. From the excellent cases emerging progressively, a constant driver was provided, with rural, traditional villages in this region establishing a popular cultural and tourism image. Examining the distribution characteristics of traditional villages allows for macro-level assessments of settlement formation, site selection criteria, and influential factors. In terms of traditional village location, numerous researchers have already carried out nation-wide investigations, and found multifarious forms from different areas and ethnic groups using qualitative analysis. This research found that villages were relevant in terms of ecology, landscape, and Geomantic Omen; moreover it revealed that the wisdom responded to the nature of the original human [8,9]. In the past decade, visual quantitative analysis, often leveraging ArcGIS (v10.8.0), has emerged as the predominant approach for investigating spatial distribution patterns [10,11,12,13,14]. Subsequently, technologies like Geodetector have been integrated, enabling the calculation of driving forces and the exploration of cross-interactions among multiple factors, thus enhancing the comparability and relevance of influential factors [15]. Research in this domain has taken various scopes, with some studies centered on administrative divisions, typically at the provincial level. These encompass regions such as Sichuan, Guizhou, Yunnan, Hubei, Fujian, Guangdong, Jiangxi, and even the entire expanse of China [16,17,18,19,20,21]. The selection of administrative divisions as research boundaries often aligns with comprehensive environmental factors, including natural, cultural, and economic aspects. Such an approach helps unveil the distribution patterns of traditional villages in similar holistic contexts. Conversely, other studies have adopted specific geographical environments as their research focus. These may encompass river basins like the Yellow River and the Yangtze River, or distinctive terrains such as the Qinghai–Tibet Plateau and the hilly regions of northern Shaanxi [22,23,24,25]. Research on distribution characteristics typically delves into parameters like distribution types, core densities, geographical concentrations, and disparities. Meanwhile, examinations of influential factors encompass topographical features, climatic conditions, socioeconomic factors, and aspects of national culture. These comprehensive investigations collectively contribute to a deeper understanding of the complex dynamics shaping traditional village distributions.

Yunnan, Guangxi, and Guizhou have the most traditional villages and widest variety of ethnic minorities in China, thus these areas have plenty of resources for scientific research. The Yunnan, Guangxi, and Guizhou rocky desertification area holds unique significance as one of the 14 centralized and contiguous impoverished regions highlighted in the “2011–2020 Outline of Poverty Alleviation and Rural Development in China”. It distinguishes itself by possessing the most extensive rocky desertification expanse, the largest minority population, and the most substantial ethnic autonomous territories. Under the threat of rocky desertification, traditional villages there are suffering from severe poverty and general disappearance, therefore it is an emergency to preserve them and propel development. This region continues to be of paramount importance in social development, underscored by its renewed emphasis in the 2021 “14th Five-Year Plan for National Economic and Social Development and the Vision Goals of 2035”. These documents prioritize “scientifically promoting comprehensive rocky desertification control and enhancing the ecological security barrier system.” However, despite extensive research on the ecological and economic aspects providing a practiced path to deal with environmental and financial problems progressively [26,27,28,29], there is a noticeable lack of research pertaining to traditional villages in rural areas which have the most poverty within this region. Research on the distribution characteristics of traditional villages in this region can comprehend the issues there more specifically and organically.

This study is designed to address this lack through the following objectives:

• Describing and analyzing the spatial distribution characteristics of traditional villages in the Yunnan, Guangxi, and Guizhou rocky desertification area.
• Analyzing its influencing factors and evaluating whether they are positive or negative, and explaining the influence extent of its influencing factors, and estimating which ones are dominant.
• Summarizing and probing the reasons for the generation of these distribution patterns.
• Proposing specific protection and development strategies.

By pursuing these objectives, this study seeks to enhance the understanding of the site location driver of traditional villages in this ecologically and socially significant region; creatively analyzing the correlation between rocky desertification and traditional villages in terms of quantification can specifically recognize the influencing factors of this unique environment. Furthermore, this study aims to offer practical recommendations that can facilitate the region’s continued development and ecological conservation, and improve the financial situation of rural poverty while serving as a model for similar regions confronting analogous issues.

2. Materials and Methods

2.1. Study Area

In southwestern China, an expansive karst landscape, coupled with ample rainfall, results in surface water runoff, soil erosion, substantial rock exposure, and the swift emergence of rocky desertification [30]. Since the 1980s, global warming and large-scale, unsustainable human development have exacerbated the rocky desertification issue. The region’s delicate environment, combined with the complexities of land resource utilization, poses challenges to achieving sustainable development. Consequently, villages in this context often experience economic disparities and lower living standards. Since 2005, the government has initiated rocky desertification control projects, notably the “clear water and green mountains are golden mountains and silver mountains” initiative, resulting in observable improvements in the rocky desertification situation. However, a significant challenge remains in how to enhance living conditions in these villages, elevate economic standards, and formulate development plans that foster progress without adversely impacting the ecological environment or exacerbating rocky desertification. Hence, a fundamental imperative lies in the analysis of the distribution characteristics of traditional villages within the Yunnan, Guangxi, and Guizhou rocky desertification area. Additionally, investigating the key factors influencing these distribution patterns assumes paramount importance. Such inquiries serve as crucial foundations for facilitating the sustainable development of regions prone to rocky desertification.

The Yunnan, Guangxi, and Guizhou rocky desertification area have the most objects of poverty and the most extensive rocky desertification area of China’s 14 centralized and contiguous poverty-stricken regions. Situated in southwestern China, it spans from 21°35′ N to 27°31′ N latitude and from 103°24′ E to 110°55′ E longitude, covering an extensive area of approximately 2.34 million square kilometers. This region encompasses 15 prefecture-level administrative divisions and 91 county-level administrative regions across Yunnan, Guangxi, and Guizhou provinces (Figure 1). Predominantly situated within the Yunnan–Guizhou Plateau, the landscape is characterized by karst topography, marked by rugged terrain, low mountains, and hills, along with a significant expanse affected by rocky desertification. The climate here falls within the subtropical monsoon category, with an annual average temperature ranging from 5 °C to 24 °C and annual precipitation levels spanning between 800 and 1800 mm. As of August 2022, a total of 705 traditional villages from this region have been officially recognized and included in the List of Chinese Traditional Villages. Most of them are minority traditional villages generated during the Ming and Qing dynasties. Experiencing repeated migration and self-derivation, they remain multiple sites cultural heritage recording historic civilization. At present, preserving and developing contiguous clusters of traditional villages is a crucial project within this region.

2.2. Data Source

The data for the 705 villages included in this study were sourced from six successive editions of the Chinese Traditional Village Directories, which were officially published by the Ministry of Housing and Urban-Rural Development, the Ministry of Culture, and the Ministry of Finance. These village locations were geospatially pinpointed using Baidu Maps to determine their geometric centers or village committee coordinates, which were subsequently mapped using ArcGIS. Elevation data for the Yunnan, Guangxi, and Guizhou rocky desertification area were obtained from the Geospatial Data Cloud (https://www.gscloud.cn/, accessed on 3 August 2023). Data regarding water systems, transportation networks, temperature, and precipitation were retrieved from the National Data Center for Earth System Science (http://www.geodata.cn/, accessed on 3 August 2023). Population and GDP data for each administrative region was sourced from regional almanacs and the Bulletin of China’s Seventh Population Census. Information related to intangible cultural heritage was accessed through the Digital Museum of China Intangible Cultural Heritage (https://www.ihchina.cn/, accessed on 5 August 2023). Topographic relief data were computed utilizing ArcGIS software based on the available elevation data. The data concerning stony desertification sensitivity in each county were derived from a research paper authored by Hu Yunfeng, a researcher affiliated with the Chinese Academy of Sciences [31] (

Table 1. Data source.

Data	Description	Sources
Traditional village	705 point features	Traditional Chinese village Digital Museum: http://www.dmctv.cn/ (accessed on 25 July 2023)
Elevation	Spatial resolution of 90 m	Geospatial Data Cloud: https://www.gscloud.cn/ (accessed on 3 August 2023)
River	Line features of main river	The National Earth System Science Data Center: http://www.geodata.cn/ (accessed on 3 August 2023)
Road	Line feature of railways and highways
Temperature	Spatial resolution of 1 km
Rainfall
Population	Quantity and Structure of population	The Ministry of Natural Resources of the P.R.C.: http://www.stats.gov.cn/ (accessed on 5 August 2023)
GDP	Annual GDP of each county	The Resource and Environmental Science Data Center of the Chinese Academy of Sciences: http://www.resdc.cn/ (accessed on 5 August 2023)
Intangible cultural heritage	Quantity of each county	Intangible Cultural Heritage Net of China https://www.ihchina.cn/ (accessed on 5 August 2023)

).

2.3. Methods

The study used various visualization and quantitative methods to investigate the distribution characteristics of traditional villages and to delve into the factors influencing these characteristics.

For analyzing the distribution characteristics of traditional villages, the average nearest neighbor index and Voronoi diagram are used to analyze the distribution type of traditional villages; moreover, these two models can verify each other’s result. Kernel density analysis, geographical concentration index, and Moran’s I index are used to analyze the nucleus, extent, and correlation of agglomeration. Imbalance index and Lorentz curve are used to evaluate the spatial distribution balance or imbalance of traditional villages [24,25].

For clearly probing the influencing factors of distribution, Buffer analysis, overlay analysis, and more visualization tools in ArcGIS are used. In addition, Geodetector applies both single and interaction factors detection methods to assess the degree of influence exerted by different factors on the distribution of traditional villages [20].

More details of the methods used in this study can be found in

Table 2. Statistical analysis models and their geographical interpretation.

No.	Indicator	Formula	Definition	Interpretation
1	Average Nearest Neighbor Index	$R=\frac{\overline{D}}{\overline{D}i}$	$\overline{D}$ is the actual nearest neighbor distance; $\overline{D}i$ is the theoretical nearest neighbor distance.	Measures the spatial distribution type of point elements. When R > 1, it is uniformly distributed; when R = 1, it is randomly distributed; when R < 1, it is a clustered distribution.
2	Voronoi Diagram	$Cv=\frac{R}{\overline{s}}$	R is the standard deviation of the polygon area of Tyson, and $\overline{s}$ is the average of the polygon area of Tyson.	Measures an element’s relative degree of spatial variation. If 33% < $Cv$ < 64%, the dotted elements are randomly distributed. If $Cv$ ≥ 64%, the point elements are agglomerated. If $Cv$ ≤ 33%, the dotted features are evenly distributed.
3	Kernel Density Analysis	$f\left(x\right)=\frac{1}{nh}{\displaystyle \sum _{i=1}^n}k\left(\frac{\chi -{\chi }_i}{h}\right)$	n represents the number of points in the neighborhood; h represents the bandwidth; $k\left(\frac{\chi -{\chi }_i}{h}\right)$ represents the kernel function.	Measures the density of point elements in their surrounding neighborhood; the larger the $f\left(x\right)$ value, the denser the distribution of point.
4	Geographic Concentration Index	$G=100\times \sqrt{{\displaystyle \sum _{i=1}^n}{\left(\frac{{\chi }_i}{T}\right)}^2}$	$x_i$ is the number of point elements in the ith city area; T is the total number of point elements; and n is the total number of cities.	Reflects the degree of concentration in a certain area of point elements; the value of G is between 0 and 100. The larger the value of G, the more concentrated the distribution of point elements.
5	Global Moran’s I	$$\begin{gathered} \mathrm{M}\mathrm{o}\mathrm{r}\mathrm{a}\mathrm{n}\mathrm{’}\mathrm{s}\mathrm{I}= \\ \frac{\sum _{i=1}^n\sum _{j=1}^n{w}_{ij}\left(x_i-\overline{x}\right)\left(x_j-\overline{x}\right)}{\sum _{i=1}^n\sum _{j=1}^n{w}_{ij}\sum _{i=1}^n{\left(x_i-\overline{x}\right)}^2} \end{gathered}$$	$x_i$ and $x_i$ are the number of villages in No.i and No.$j$ counties; $\overline{x}$ is the average number; $w_{ij}$ is the spatial adjacent weight matrix of the counties; and n is the total number of counties.	The values of Moran’s I is [0, 1], and the closer to 1, the more positive the spatial autocorrelation of the factor, the closer to 1, the more negative the spatial autocorrelation of the factor, 0 meaning a random pattern.
6	Imbalance Index	$S=\frac{\sum _{i=1}^n{Y}_i-50(n+1)}{100\times n-50(n+1)}$	n is the number of study areas; $Y_i$ is the cumulative proportion of the ith rank after the proportion of point elements in each region is ranked from large to small.	Reflects the degree of imbalance in a certain area of point elements; 0 < S < 1 indicates uneven distribution; S = 1 indicates uniform distribution; and S = 0 indicates highly concentrated distribution.
7	Geodetector	$q=1-\frac{\sum _{h=1}^L{N}_h{\sigma }_h^2}{N{\sigma }^2}$	L is the stratification of the independent or dependent variable, $N_h$ and ${\sigma }_h^2$ are the number of elements and the variance of layer h, respectively, and N and ${\sigma }^2$ are the number of units and the variance of the whole, respectively.	Analyze the strength of each factor on spatial differentiation. The q value is a measure of the detection force of the independent variable. The q value is [0, 1], and the closer to 1, the greater the influence of the factor.

, and the study framework is shown in Figure 2.

3. Results

3.1. Spatial Distribution Characteristics of Traditional Villages in Yunnan, Guangxi, and Guizhou Rocky Desertification Area

3.1.1. Spatial Distribution Type

The spatial distribution of traditional villages typically falls into three categories: clustered distribution, random distribution, and uniform distribution. In the case of the Yunnan, Guangxi, and Guizhou rocky desertification area, an analysis of the average nearest neighbor in ArcGIS was conducted. The expected average distance ($\overline{Di}$) was 10,452.50 m, the observed average distance ($\overline{D}$) was 5257.31 m. The nearest neighbor ratio (R) was calculated to be 0.50, which is less than 1, and the z = −25.25, p = 0.00 (Figure 3a). This result indicates that the traditional villages in the Yunnan, Guangxi, and Guizhou rocky desertification area clustered. To further validate this clustering pattern, a Voronoi Diagram was constructed using the Thiessen Polygon tool. The analysis revealed a standard deviation of polygon areas at 1441.02 square kilometers, with a mean area of 332.80 square kilometers. The coefficient of variation (CV) was calculated at 433.00%, which is significantly greater than 64% (Figure 3b). This additional analysis also indicates that the traditional villages in the Yunnan, Guangxi, and Guizhou rocky desertification area clustered.

3.1.2. Spatial Agglomeration Situation

In the Yunnan, Guangxi, and Guizhou rocky desertification area, there exist a total of 15 prefecture-level administrative regions. Remarkably, among these regions, two cities stand out for their absence of traditional villages. By calculating with the statistics in

Table 3. Statistics related to traditional villages in city.

Province	City	Quantity of Counties/Total	Quantity of Villages	Proportion of Villages	Cumulative Proportion of Villages	Average Quantity of Villages (per 1000 km2)
Guizhou	Southeast Guizhou	16/16	415	58.87%	58.87%	13.48539
	Anshun	6/6	78	11.06%	69.93%	8.186519
	Southern Guizhou	12/13	78	11.06%	80.99%	3.116398
Guangxi	Guilin	2/17	39	5.53%	86.52%	8.805058
	Liuzhou	3/10	33	4.68%	91.21%	3.26619
Yunnan	Wenshan	8/8	23	3.26%	94.47%	0.704571
Guizhou	Southwest Guizhou	8/8	11	1.56%	96.03%	0.632938
Yunnan	Honghe	2/3	8	1.13%	97.16%	2.190636
	Qujing	2/9	6	0.85%	98.01%	0.997198
Guizhou	Liupanshui	3/4	5	0.71%	98.72%	0.822106
Guangxi	Hechi	10/11	4	0.57%	99.29%	0.132178
	Baise	12/12	3	0.43%	99.72%	0.080464
	Chongzuo	4/7	2	0.28%	100.00%	0.177923
	Nanning	3/12	0	0.00%	100.00%	0
	Laibing	1/6	0	0.00%	100.00%	0

, the geographical concentration index G is 61.47. This figure surpasses the average geographical concentration index Ga of 47.00. Such a disparity indicates that traditional villages in this area tend to be clustered within these prefecture-level regions. Moving to a more granular level of analysis, we find that there are 91 county-level administrative regions within this rocky desertification area. Out of these, 55 counties are known to harbor traditional villages. The calculated geographical concentration index G at the county level is 25.50, markedly higher than the average geographical concentration index of 7.75. This outcome underscores the concentration of traditional villages at the county level within this region. Expressed per 1000 km², the density of traditional villages is notably highest in the southeastern areas of Guizhou, particularly Anshun and Guilin, as indicated in Table 3.

Utilizing nuclear density analysis, Figure 4a underscores the concentrated distribution of traditional villages within Southeast Guizhou and the Anshun region. Notably, the apex of this concentration is observed in Leishan County within Southeast Guizhou, with Guilin serving as a secondary focal point. In stark contrast, the central, southern, and western sectors, particularly the southern segment, lack substantial traditional village clusters Furthermore, employing the “Global Moran’s I” methodology, spatial autocorrelation assessment reveals a positive correlation in traditional village distribution, as evidenced by a Moran’s Index of 0.39, a z-score of 6.03 (exceeding the significance threshold of 2.58), and a p-value of 0.00 (indicating a non-random clustering pattern). Building upon this foundation, “Anselin Local Moran’s I” delineates significant agglomeration zones. Figure 4b highlights the High–High cluster and the Low–Low cluster, which means highly agglomerative counties concentrate in Southeastern Guizhou, with contrasting lowly agglomerative counties concentrated in the southwestern regions of Baise and Hechi, and the value of agglomeration around it is positive correlated with it. It is noteworthy that Libo County exhibits a distinctive profile, falling within the category of a Low–High outlier, which means its agglomeration is low with contrasting higher agglomeration around. In summation, Southeast Guizhou emerges as the primary city where the pronounced traditional village concentration is autocorrelative.

3.1.3. Spatial Distribution Equilibrium

Table 3 reveals that southeast Guizhou hosts most traditional villages, constituting 58.87% of the total, a notably higher proportion compared to other regions. Following closely are Anshun and Southern Guizhou, each with a substantial number of traditional villages, accounting for 11.06% each. In addition, the Nanning and Laibin regions do not have any traditional villages. Utilizing the imbalance index method in conjunction with the distribution data of traditional villages within the Yunnan, Guangxi, and Guizhou rocky desertification area, the imbalance index S for the spatial distribution of traditional villages at the prefecture-level administrative regions was computed as 0.82. This value underscores a significant degree of unevenness in the distribution of traditional villages across prefecture-level administrative divisions. Moreover, the imbalance index S at the county-level administrative regions is computed at 0.81, indicating a similar level of uneven distribution as observed at the prefecture level. The generated Lorentz Curve (Figure 5) corroborates this assessment. It distinctly deviates from the Average Curve, displaying substantial curvature, which signifies a pronounced imbalance in the county-level distribution of traditional villages.

3.2. Factors Influencing the Spatial Distribution Characteristics of Traditional Villages in the Yunnan, Guangxi, and Guizhou Rocky Desertification Area

3.2.1. Natural Environment

The Yunnan, Guizhou, and Guangxi rocky desertification area exhibits an average elevation of approximately 1499.47 m. This region predominantly features a karst landform, characterized by significant ground undulations. Geographically, it encompasses the Yunnan–Guizhou Plateau, with terrain transitioning from hilly in the northwest to Zhongshan in the southeast; the highest elevations are found in the northwestern areas, specifically in Liupanshui, Guizhou Province, and Qujing and Honghe in Yunnan Province. These locations, situated within the Wumeng Mountains, reach elevations exceeding 2500 m. In contrast, the lowest elevations are observed around Chongzuo, Nanjing, and Guilin in Guangxi, where elevations are less than 100 m. Hechi City, located in the eastern part of this region, displays the most prominent, extensive, and dense ground undulations. East of Baise City, noticeable canyon basins are encircled by mountains to the southwest and northeast, resulting in lower elevations and ground relief variations. To analyze these elevation and ground relief characteristics, ArcGIS was employed through the “extracting value to point” method, enabling the extraction of pertinent data from traditional village locations for subsequent statistical analysis.

From an elevation perspective, traditional villages in the Yunnan, Guizhou, and Guangxi rocky desertification area are primarily situated within the 500 to 1000 m middle-altitude range, encompassing 459 villages, or 65.11% of the total. These are concentrated mainly in the Miaoling Mountain Range and Anshun in southeast Guizhou. Other traditional villages are dispersed around the Miaoling Mountain range in southeast Guizhou (Figure 6a). This pattern is historically rooted; higher elevations served as strategic locations, distancing these villages from urban development, offering protection during ancient conflicts, and preserving their traditional character amidst modernization [32]. Regarding topographic relief, 551 traditional villages are located on moderately relieved hills, constituting around 60.28% of the total (Figure 6b). The intricate terrain has limited infrastructure development, reducing external interactions and nurturing unique cultural environments, thus preserving the historical essence of these villages [33]. Additionally, in these rocky desertification-prone regions with scarce cultivable land, selecting high-elevation areas with fluctuation allows for maximizing flat land availability for agriculture, satisfying agricultural production needs.

Rocky desertification is a land degradation process characterized by reduced vegetation and extensive soil erosion, resulting in the exposure of large bedrock areas on the surface. This phenomenon often leads to a decline in land productivity, rendering agricultural production, upon which traditional villages depend, challenging. The Yunnan, Guangxi, and Guizhou rocky desertification area is one of the most severely affected regions in China. Lava coverage accounts for 48.70% of its total area, a major contributor to this desertification issue. The extent of rocky desertification is dynamic, influenced by both human activities and environmental factors. In this study, the concept of sensitivity to rocky desertification is employed to assess the degree and likelihood of ecological problems related to rocky desertification within the disturbed areas of the Yunnan, Guangxi, and Guizhou rocky desertification regions. The methodology hinges on an ecological environmental sensitivity evaluation model [34]. Utilizing data acquired through ArcGIS, in conjunction with the locations of traditional villages, a spatial distribution map illustrating the sensitivity of each county to rocky desertification was generated (Figure 6c). A significant finding is that 62 counties, or 68.13% of the total, exhibit marked sensitivity to rocky desertification. The northern region of southern Guizhou shows particularly severe sensitivity, whereas the central and eastern portions exhibit relatively lower sensitivity. This outcome underscores the fact that most traditional villages tend to be strategically positioned in areas with low sensitivity to rocky desertification. These regions boast favorable soil conditions for cultivation, facilitating sustainable village development. However, certain villages are in areas with higher rocky desertification sensitivity, notably in the high rocky desertification sensitivity zone of Anshun. This distribution pattern is attributed to their historical origins, particularly as “military tun” settlements during the Ming Dynasty. These villages require minimal cultivated land and exert limited impact on rocky desertification due to their military focus [35].

The availability of water systems has historically been a pivotal factor in the establishment and development of early human settlements, contributing significantly to the formation of traditional villages. In the Yunnan, Guangxi, and Guizhou rocky desertification area, a network of rivers crisscrosses the landscape. Key water systems include the Pearl River, Yangtze River, and Red River, along with their tributaries such as the Qingshui River, Durliu River, Hongshui River, Zuojiang River, Youjiang River, Rongjiang River, Wujiang River, Nanpanjiang River, and Beipanjiang River. These rivers collectively provide abundant water resources that have been essential for the sustenance of villages in this region. To understand the impact of these rivers on the distribution of traditional villages, various spatial analyses were conducted using ArcGIS, including “buffer zone analysis”, “superposition analysis”, and “near neighbor analysis”. These analyses divided the distance from the main rivers into three regions: 0~5 km, 5~10 km, and 10~15 km. The number of traditional villages within these different distance categories was tallied to explore the influence of rivers on village distribution (Figure 6d). The findings reveal a concentration of villages near the Qingshuijiang River and Dulujiang River, a phenomenon closely tied to the historical migration patterns of the Miao and Dong ethnic minorities along these two rivers [36]. Specifically, 181 villages (25.67%) were situated within 5 km of a river, 169 villages (23.97%) were within 5–10 km, and 127 villages (18.01%) were within 10–15 km. This distribution indicates that the closer villages are to a river in this region, the more numerous they tend to be, underlining their dependence on these water sources. Interestingly, the number of villages does not sharply decline with increasing distance from the main river, and some villages even cluster near Leigong Mountain, a location situated 15 km away from the mainstream. This clustering is attributed to the area’s susceptibility to severe flooding [37]. To mitigate flood risks, many villages have chosen to establish themselves in areas more distant from the main river, where wells and streams can still provide water. Consequently, the main rivers in this region play a multifaceted role in traditional villages, serving not only as a vital water source but also as a means of flood control.

Climate exerts a dual influence on villages by affecting the living environment’s comfort and influencing crop production. Two critical climate factors, temperature and precipitation, were assessed using ArcGIS with annual average temperature and annual total precipitation data for the rocky desertification area encompassing Yunnan, Guangxi, and Guizhou in the year 2020. These data were superimposed with traditional village locations to create a distribution map. The annual average temperature across the region demonstrates an upward trend from northwest to southeast, particularly pronounced in the canyon area of Baise City (Figure 7a). Conversely, total annual precipitation decreases gradually from west to east, with the highest levels recorded in Guilin and Liuzhou City (Figure 7b). Analyzing the data extracted from traditional villages reveals that 590 villages in this region primarily occupy the annual average temperature range of 15–18 °C, constituting 83.69% of the total. This temperature range is comfortable and suitable for human habitation. Regarding precipitation, traditional villages predominantly inhabit areas receiving 1000 mm to 1500 mm of annual rainfall, accounting for 82.13% of the total. This precipitation range is ideal for agriculture, offering suitable conditions for farming while also mitigating the risk of flood disasters to a certain extent. In Guilin and Guangxi, located in the eastern part of the region, some villages are concentrated in areas experiencing annual rainfall exceeding 1500 mm. This region has implemented extensive terracing to counteract the soil erosion caused by heavy rainfall [38]. Longsheng terrace, for instance, has been under development since the Tang and Song Dynasties and was recognized as a “Global Important Agricultural Cultural Heritage” in 2018.

3.2.2. Social Economy

Towards the end of the 20th century, the Yunnan, Guangxi, and Guizhou region was primarily reliant on the primary industry. However, over the course of more than two decades, a shift in focus from the agricultural sector to other industries occurred. This transition witnessed gradual improvements in transportation infrastructure. However, these developments were primarily concentrated in urban centers, leading to a gradual increase in urbanization rates and contributing to an imbalance in social and economic development [39]. The rocky desertification area in Yunnan, Guangxi, and Guizhou has faced numerous challenges, including rocky desertification, resulting in underdeveloped and underutilized resources and limited economic capacity. These factors have contributed to widespread and deep-seated poverty. Population density, per capita GDP, and urbanization rates were calculated for each county within this rocky desertification area. The data were then layered onto traditional village locations to create corresponding distribution maps. Most counties in this region exhibit a moderate population density, ranging from 40 to 140 people per square kilometer. The northern areas near Anshun and Kaili, however, have relatively high population densities, ranging from 210 to 651 people per square kilometer. Notably, many of the counties housing traditional villages have low population densities, with 343 counties (48.65%) having fewer than 100 people per square kilometer and 621 counties (88.09%) having fewer than 200 people per square kilometer. This negative correlation between population density and traditional village concentration is due to the rural-to-urban migration driven by urban economic opportunities, resulting in lower population densities in areas with numerous traditional villages (Figure 8a). In terms of average annual per capita GDP, most regions in this area fall below CNY 50,000, which is below the national average. Traditional villages are primarily located in counties with per capita GDP ranging from CNY 29,800 to CNY 39,700, constituting 83.55% of the total. None of the traditional villages have a per capita GDP lower than CNY 29,800. This indicates that counties hosting traditional villages tend to have lower per capita GDP, which can negatively impact the preservation of these villages and even lead to their disappearance (Figure 8b). Considering the urbanization rate, it varies across counties from 20.33% to 90.00%, with regions such as Wenshan Autonomous Prefecture and Baise City in the southwest having several counties with low urbanization rates. Most traditional villages are situated in areas with urbanization rates below 40%, suggesting that lower urbanization rates are more conducive to the preservation of traditional villages (Figure 8c). In summary, socio-economic development in the Yunnan, Guangxi, and Guizhou rocky desertification area is generally negatively correlated with the retention of traditional villages. This correlation is driven by urbanization stemming from higher economic levels, leading to significant rural-to-urban migration. Traditional villages, unable to keep pace with urban development, either retain their original character, culture, and production modes or face extinction as populations migrate. Additionally, many former settlements have transformed from villages into cities, erasing their traditional appearances and cultures in favor of modernization. An exception to this overall trend is the Anshun region, which exhibits high socio-economic indicators while still preserving a significant number of traditional villages. The reason for this exceptional case lies in the migration history of the Tunpu traditional villages in Anshun, which have experienced fewer conflicts with Han-dominated cities. Furthermore, Anshun has implemented advanced tourism development measures that protect traditional culture while promoting local economic growth, thus achieving a positive feedback loop that radiates to surrounding areas and maintains compatibility between traditional villages and social and economic development [40].

To evaluate the impact of national expressways and railways on the distribution of traditional villages, “buffer zone analysis”, “superposition analysis”, and “near neighbor analysis” were employed in ArcGIS. These analyses segmented the length of the national expressways and railways into three regions: 0~5 km, 5~10 km, and 10~15 km. The number of traditional villages within these various distance ranges was tabulated to assess the influence of public transportation on traditional village distribution (Figure 8d). The findings reveal that there are 297 traditional villages located within 5 km of public transportation routes, constituting 42.13% of the total. In addition, 176 traditional villages are situated within 5–10 km from public transportation, accounting for 24.96%, and 102 traditional villages are located within 10–15 km from public transportation, making up 14.47% of the total. These results indicate that, as the distance to public transportation decreases, the number of villages increases, highlighting the positive impact of public transportation on the preservation of traditional villages. Notably, among the traditional villages within a buffer distance of less than 5 km, half of them are positioned within 1 km of public transportation routes. This underscores that public transportation in this area not only does not disrupt traditional villages but also provides convenience and accessibility to these villages, facilitating their continued existence and ease of maintenance.

3.2.3. National Culture

Historically, Yunnan, Guangxi, and Guizhou have consistently been major regions in China where ethnic minorities have gathered [41]. This area also holds the distinction of having the largest number of ethnic minorities among the 14 contiguous poverty-stricken areas in China. The dominant ethnic groups include the Zhuang, Miao, Buyi, Yao, and Dong, among others. Over time, various ethnicities have thrived and intermingled in this region, giving rise to a wealth of unique cultural heritages, including the influence of “Tunpu” military culture. Therefore, this study seeks to measure the impact of ethnic culture on traditional villages through the variables of ethnic minority population and national intangible cultural heritage. Using ArcGIS, the proportion of ethnic minority populations and the quantity of national intangible cultural heritage in each county were visualized and overlaid with traditional village locations. The analysis reveals that the proportion of the ethnic minority population in the Yunnan, Guangxi, and Guizhou rocky desertification area decreases as one moves from the southeast to the northwest. Most areas have ethnic minority populations exceeding 50%, with 13 counties in Guangxi even reaching levels above 90%. Of the 476 (67.52%) traditional villages in this region, the majority have ethnic minority populations exceeding 80%, underscoring the significant influence of ethnic minority populations on traditional villages. An exception is noted in Anshun, where traditional villages have relatively smaller minority populations, primarily because most “Tunpu” settlements in Anshun were established by Han Chinese (Figure 9a). Examining the presence of national intangible cultural heritage, the Southeast Guizhou region in the northeast boasts the highest number, with a decreasing radiating effect outward from this central region. In contrast, some areas in the western part of the region lack national intangible cultural heritage. The results indicate that traditional villages are predominantly concentrated in areas with intangible cultural heritage, especially in regions with a high quantity of such heritage. Specifically, 555 traditional villages are in counties with more than two intangible cultural heritages (78.72%), 368 traditional villages are situated in counties with more than five intangible cultural heritages (52.20%), and only 59 traditional villages are found in counties with no intangible cultural heritage (8.37%). This underscores the substantial influence of national intangible cultural heritage on the spatial distribution of traditional villages in this area (Figure 9b).

Overall, the distribution of traditional villages is positively correlated with the number of national intangible cultural heritages and the proportion of the ethnic minority population, which is negatively correlated with the sensitivity of stony desertification, river distance, traffic distance, population density, and urbanization rate, and correlated with the median elevation, temperature, precipitation, relief degree, and per capita GDP.

3.3. Quantitative Analysis of the Influencing Factors of the Yunnan, Guangxi, and Guizhou Rocky Desertification Area

3.3.1. Single Factor Detection

The study utilized ArcGIS to extract data for 12 factors that influence traditional village points. These factors included elevation (X1), relief degree (X2), river distance (X3), county stony desertification sensitivity (X4), air temperature (X5), precipitation (X6), traffic distance (X7), county population density (X8), county per capita GDP (X9), county urbanization rate (X10), county minority population proportion (X11), and county intangible heritage quantity (X12). The Geodetector software was employed to assess the impact of each factor on traditional villages, with a p-value < 0.05 indicating the significance of each factor’s influence. The explanatory power of each factor was as follows, in descending order: the number of intangible cultural heritage > population density > urbanization rate > per capita GDP > minority population proportion > precipitation > air temperature > rocky desertification > elevation > river distance > road distance > relief degree (

Table 4. Geodetector results of spatial differentiation of traditional villages.

	X1	X2	X3	X4	X5	X6	X7	X8	X9	X10	X11	X12
q statistic	0.103	0.045	0.078	0.118	0.208	0.260	0.048	0.430	0.399	0.424	0.321	0.430
p value	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000

). In summary, social-economic and ethnic cultural factors held greater explanatory power than natural environmental factors in this analysis.

For a more profound cognition, it is necessary to analyze the cause of this result. Remarkably, this contrasted with previous studies on Yunnan and Guangxi, which often emphasized the significant impact of the natural environment on traditional villages [18,42]. Moreover, research on southwestern China also indicated that natural environment factors perform the most crucial role in traditional villages [43]. In comparison, research on the distribution of traditional villages in Guizhou indicate that socio-economic factors and national culture factors dominate, which can explain this notable phenomenon [17,44]. In addition, the traditional Guizhou villages of this study amount to 587, constituting 83.69% of the total, especially the traditional villages of Southeast Guizhou constituting 58.87% of the total which proved the severe impact of urbanization [36], so that they dominate the result primarily. The lower explanatory power of natural environmental factors in this study can be attributed to the unique characteristics of the selected Yunnan, Guangxi, and Guizhou rocky desertification area, which is one of China’s concentrated contiguous areas. This region exhibits significant differences in socio-economic and environmental conditions compared to other areas. The relatively lower economic development level makes villages more sensitive to changes in social-economic factors [45]. This sensitivity is further supported by correlation analyses of population density, urbanization rate, and per capita GDP in this region. Urbanization-driven economic changes often lead to rural-to-urban migration, which can disrupt the development environment of traditional villages and result in their degradation or disappearance. However, the low explanatory power of road distance suggests that many villages in this region have limited external connectivity, which contributes to the preservation of their original character. Additionally, ethnic culture’s explanatory power is notably high, with the number of intangible cultural heritages standing out as the most influential factor among all variables. This region boasts rich ethnic culture, which developed relatively late in the country’s history, giving rise to a wealth of regional cultural heritage [46]. These ethnic intangible cultural heritages serve not only as cultural foundations for ethnic groups’ development over time but also as significant economic resources, particularly as urbanization increases. This synergy has established a positive feedback loop, contributing to the preservation and continuity of traditional villages.

3.3.2. Interaction Factor Detection

The study employed various visualization and quantitative methods to investigate the distribution.

Geodetector’s interaction factor detection assesses how the explanatory power of the dependent variable changes when two influencing factors operate together. In this study, 12 influencing factors were chosen for interactive impact factor detection to examine alterations in their explanatory power on traditional villages (Figure 10). Based on the relationship between the obtained q-value and the independent explanatory power of the two impact factors, five types of interactions were identified:

• q(A∩B) < min[q(A), q(B)]—This suggests that the nonlinearity of the two factors is weakened when they work together.
• min[q(A), q(B)] < q(A∩B) < max[q(A), q(B)]—This indicates that the nonlinearity of a single factor is reduced when it interacts with another.
• q(A∩B) > max[q(A), q(B)]—This implies that the dual-factor interaction is enhanced.
• q(A∩B) = q(A) + q(B)—This suggests that the two factors are independent of each other.
• q(A∩B) > q(A) + q(B)—This indicates that the two factors exhibit a nonlinear interaction.

The results of this study revealed only two instances of nonlinear enhancement or dual-factor enhancement. Three indicators, namely “population density ∩ urbanization rate”, “population density ∩ intangible heritage number”, and “urbanization rate ∩ minority population proportion,” exhibited explanatory power exceeding 0.6. This further underscores the predominant influence of socio-economic and ethnic cultural factors on traditional villages.

Although the individual impact of factors, such as river distance and stony desertification sensitivity, was relatively small, most of these factors exhibited nonlinear interactions with others, where the explanatory power exceeded the sum of the two interacting factors (Figure 10).

4. Discussion

4.1. Factors Influencing the Spatial Distribution Pattern of Traditional Villages

The factors influencing traditional villages have undergone continuous changes throughout history. Before the Yuan dynasty, the regions of Yunnan, Guangxi, and Guizhou were vast, and the populations there were sparse, so that the mode of production in villages was original and agriculture was the primary [47]. Therefore, factors such as rivers, precipitation and temperature held particular significance during the early stages of traditional village development. The villages in the Yunnan, Guangxi, and Guizhou rocky desertification area suffered from rocky desertification and showed more cravings for arable land, so they were often situated at relatively high elevations with significant geomorphic relief to release more flatlands. Therefore, natural environmental factors constituted the original factor shaping the spatial distribution of traditional villages.

After entering the Yuan Dynasty, the initial distribution pattern of traditional villages was established under the influence of the natural environment; ethnic culture emerged as the dominant factor deepening the variations in the spatial distribution patterns of traditional villages across different regions. Many traditional villages in the Yunnan, Guangxi, and Guizhou rocky desertification area generated from ethnic migrations, while the populations of migrations increased faster and faster, especially in the Ming and Qing dynasty when this region constructed efficient traffic [46]. These ethnic groups often practiced a strategy of “living in steep environments” as a means of avoiding potential threats of war, especially Miao village concentrated at the Leigong mountain in Leishan. Various ethnic groups congregated based on group identity, while differences in ethnicity sometimes led to estrangement or, conversely, integration due to complex historical factors. For example, the Han migrations of the Ming dynasty distanced the minorities, and formed the distribution pattern of minority villages far away from “Tunpu”. On the side, the loose policy for minorities ultimately turned this separation to integration [48].

While urbanization gained momentum, ethnic cultural factors continued to play a role in preserving traditional villages as sources of group cohesion and economic development. After the Qing dynasty, China entered the primary stage of modernization with the global environment; the rapid pace of social and economic development encouraged rural-to-urban population migration, significantly diminishing the vibrancy of traditional villages. Particularly in the Yunnan, Guangxi, and Guizhou rocky desertification area, where traditional villages tend to have lower economic levels, the impact of urban economic development is particularly sensitive. Advanced industrial development altered the original character of traditional villages and transformed their modes of production, resulting in the gradual erosion of traditional culture and the loss of their survival space, so that the distribution of traditional villages in the Yunnan, Guangxi, and Guizhou rocky desertification area are distanced from the urban. However, economics also exerts a positive influence. Economics, combined with cultural or ecological resources, can positively influence the preservation of traditional villages; just as state departments started to consecrate enumerations of traditional village inventories, they took advantage of cultural and agricultural industrialization. In addition, convenient transport performs a crucial role in agricultural trade and rural tourism with natural environmental factors, which provide a primary income for rural villages, especially the Yunnan, Guangxi, and Guizhou rocky desertification area, which has a complex geography; thus, the villages near the road are more prosperous.

Consequently, natural environment was the original factor, which initially formed the basic distribution pattern of the traditional village in this region; national ethnicity factors are the ensuing determinate driver, which deepens the difference of the distribution pattern over time in this region. The socio-economic factor is a crucial driver after urbanization, which determines if the traditional villages are sustainably developing in this region. Moreover, the interaction among each factor can influence the distribution pattern of traditional villages of this region more profoundly.

4.2. The Protection and Development Strategy of Traditional Villages

As is learned from the distribution pattern of the traditional village in the Yunnan, Guangxi, and Guizhou rocky desertification area, the strategies for preserving and developing traditional villages should emphasis the agricultural and ecological issues by considering the natural environment factors, especially the rocky desertification. The numerous national cultures of the traditional villages should be treasured as the dominated driver to ensure their authenticity, which provides a constant developing motivation. Appropriate financial aid can foster the development of the traditional village without being urbanized, by propelling trade and tourism. However, it should consider both the common characteristics within the same contiguous region and classify and discuss them based on the varying conditions of influencing factors for each village. To foster sustainable development in this region, the author proposes the following three development approaches:

• Ecological Preservation and Tourism: Considering the wide area of rocky desertification, this mode would entail managing the ecological problem. Based on this, leveraging the rich landscapes of karst landform propels this original ecological scenic spot, promoting tourism initiatives that emphasize the area’s unique natural landscapes. To protect the natural environment from disaster, government overall planning would be a key element of this strategy. By creating sustainable eco-tourism experiences that respect and preserve the local culture and environment, traditional villages can benefit economically while safeguarding their natural environment.
• Cultural Heritage Conservation and Revitalization: This approach centers on preserving and revitalizing the cultural heritage of the region. It involves targeted efforts to safeguard traditional practices, art forms, and intangible cultural heritage. Initiatives could include heritage festivals, cultural education programs, and heritage tourism. Flexible operation among businesses would be crucial for insuring the finances. Thus, selecting literate, responsible, and proficient businesses can ensure the longevity and vibrancy of these cultural traditions.
• Eco-Agriculture and Sustainable Farming Practices: Given the historic importance of agriculture, a sustainable eco-agriculture model can be explored. This approach would encourage environmentally friendly and sustainable farming practices that align with the region’s unique ecological characteristics. Promoting organic farming, traditional crop varieties, and agro-tourism could enhance economic prospects for local farmers while preserving the natural environment. Farmers’ spontaneous organization would be pivotal in advancing this model.

In essence, these strategies aim to strike a balance between economic development and the preservation of the ecological and cultural heritage of the Yunnan, Guangxi, and Guizhou rocky desertification area. By tailoring approaches to the specific conditions and resources of each traditional village, it becomes possible to foster sustainable development while maintaining the unique character of these communities.

5. Conclusions and Future Work

This study investigated 705 settlements located within the Yunnan, Guangxi, and Guizhou rocky desertification area, all of which are officially recognized as Chinese traditional villages. This research aimed to understand their distribution patterns and identify the key factors influencing their characteristics. Here are the main conclusions drawn from this investigation:

• Traditional villages within the Yunnan, Guangxi, and Guizhou rocky desertification area exhibit a non-uniform and clustered distribution pattern. The primary concentrations of these villages are found in the southeastern regions of Guizhou, particularly in Anshun. Additionally, Guilin represents a secondary cluster area for traditional villages. Notably, there is a clear autocorrelation pattern observed in most counties located within southeastern Guizhou, emphasizing the clustering nature of these villages in this specific geographical area.
• Traditional villages in the Yunnan, Guangxi, and Guizhou rocky desertification area are intricately tied to environmental, social, and cultural factors:
  In terms of natural environment factors, these villages are primarily found at altitudes of 500–1000 m, in areas with moderate terrain relief and low rocky desertification sensitivity. They are typically located within 5 km of rivers, emphasizing the importance of water resources. The climate is ideal, with average annual temperatures of 15–18 °C and annual precipitation between 1000 and 1500 mm, favoring agriculture. In terms of socio-economic factors, traditional villages are concentrated in counties with low population densities (<100/sq. km), average annual GDP per capita between CNY 29,800 and CNY 39,700, and urbanization rates below 40%. Accessibility to other villages and transportation networks, within 5 km buffer zones, is common. In terms of national cultural factors, these villages often thrive in counties where ethnic minorities make up over 50% of the population and boast at least two national intangible cultural heritage items.
  Among these factors, ethnic populations and intangible cultural heritage are positively correlated with village distribution. Conversely, factors like stony desertification sensitivity, distance from rivers and major roads, high population densities, and high urbanization rates negatively affect distribution. Elevation, temperature, precipitation, terrain relief, and GDP per capita are median correlation.
  Notably, in Anshun, historical factors, such as rocky desertification, high urbanization, population density, ethnic composition, and intangible cultural heritage, have adversely impacted traditional villages.
• The results show that the socio-economic factors and national culture factors are greater than the natural environment, and the explanatory power of each factor is as follows: number of intangible cultural heritage > population density > urbanization rate > per capita GDP > minority population proportion > precipitation > temperature > rocky desertification > elevation > river distance > road distance > relief degree. The interaction of river distance and rocky desertification sensitivity with most other factors can play a nonlinear role, and the explanatory power is greater than the sum of the two factors.
• From the perspective of the formation of its distribution characteristics, the undulating karst landform, low economic level, and rich cultural characteristics are the main reasons for the difference between the Yunnan, Guangxi, and Guizhou rocky desertification areas and other areas. The natural environment was the original factor of forming the initial distribution pattern of traditional villages, national culture is the dominant driver of deepening the differences in the spatial distribution pattern of different regions, and socio-economic factors are the crucial driver for the sustainable development of traditional villages. Meanwhile, the interactions among each factor create more profound effects on traditional villages in this region.
• According to the study on the distribution characteristics and influencing factors of traditional villages in the Yunnan, Guangxi, and Guizhou rocky desertification area, it is necessary to provide three types of protection and development strategies: ecological preservation and tourism through overall government planning; cultural heritage conservation and revitalization through flexible businesses; and eco-agriculture and sustainable farming practices through spontaneously organized farmers.

This study delves into the distribution characteristics of traditional villages in the rocky desertification regions of Yunnan, Guangxi, and Guizhou. It takes a macro-level approach, scrutinizing natural environmental, socio-economic, and national cultural factors to unearth the underlying influencing factors and distribution patterns. Furthermore, the study plays a basic role in micro-level research in terms of spatial layout, architecture characteristics, and ecological management in this region. The difficulties of reproducing the environment of ancient times and quantifying historic culture are the limitations of this study; ensuing works will explore more specific methodologies to tackle this. The rocky desertification area in these provinces is ecologically sensitive, economically low, and culturally diverse, making it a priority for sustained protection and development within rural revitalization initiatives. Thus, this research aims to offer a theoretical foundation and strategic insights to solve the noticeable rural poverty and rocky desertification of this region in the future, while serving as a paradigm for similar regions confronting analogous issues.