Research on the Correlation Between Spatial Layout Characteristics and Geographical Conditions for Ethnic Minority Rural Settlements

Abstract

It is significant to study the correlation between the spatial distribution and topographic features for ethnic minority rural settlements, which can provide the theoretical basis and practical methods for the preservation of ethnic culture and scientific planning of territorial space. Liuzhou in Guangxi is a region with diverse ethnic groups and this paper takes Liuzhou as the case study. This study employs fractal theory, GIS spatial analysis, and correlation analysis methods to investigate the relationship between settlement spatial patterns and their surrounding geographical conditions. The findings reveal a significant positive correlation between the geographic location of ethnic minority rural settlements (including site selection and terrain features) and their geographical conditions (topographic and elevation factors). Additionally, significant associations exist between settlement slope, settlement orientation, and their positioning within mountainous terrain. The study also reveals strong correlations between planar morphological characteristics (or settlement scale) and settlement terrain for the settlements of the same ethnic group within the same region. Specifically, Dong settlements exhibit remarkable consistency in settlement scale, while Miao settlements demonstrate high similarity in terms of elevation distribution. The methodology developed in this study is applicable to correlation research on settlement characteristics across diverse ethnic groups and geographical regions. It not only reveals universal patterns of how physical-geographic environments influence the planar and spatial features of settlements, but also validates the logical coherence of investigating layout characteristics from both planar and spatial perspectives. The findings of this study not only provide practical guidance for the development and planning of settlements, but also offer recommendations for the cultural inheritance and settlement protection of ethnic minorities.

1. Introduction

In the current rapid urbanization process in China, the government has been making efforts to promote rural revitalization in recent years, putting forward new requirements for the planning, construction, and development of rural areas. Under these circumstances, rural layout planning should adopt approaches adapted to local conditions and rationally utilize local natural conditions. China’s ethnic minorities predominantly live in rural areas. The site selection and layout characteristics of ethnic minority settlements reflect the traditional culture and need to be inherited in the process of rural construction. At the same time, traditional ethnic minority settlements are also in urgent need of protection and development. The resolution of these issues fundamentally depends on a comprehensive understanding of geographical environments and in-depth research on village–geography relationships.

The cultural characteristics and conservation/development of ethnic minority rural settlements have attracted significant attention from scholars worldwide. International research has primarily focused on settlement typology and transformation process [1,2,3], the preservation of ethnic settlements and architecture [4,5], the protection and inheritance of ethnic culture [6,7], etc. Domestic scholars mainly concentrate on the conservation and development of ethnic minority villages [8,9,10], spatial morphological characteristics and evolutionary patterns of settlements [11,12,13,14,15,16], spatial distribution features and influencing factors [17,18,19]. Hotspot analysis, the geographic detector method, the nearest neighbor index, kernel density estimation, correlation analysis and clustering analysis are widely used in the research [20,21,22,23,24,25,26,27]. Numerous studies have found that the internal elements and spatial distribution features of rural settlements are closely related to natural environments [28]. Many of them have investigated the impacts of natural environments, production factors, and lifestyles on rural settlements, specifically including terrain, sunlight exposure [29], farming practices [30], land fertility [31], population density [32], transportation factors [33], socio-historical and cultural factors [34,35], etc. Relevant studies indicate that factors such as elevation, slope gradient, slope direction, and proximity to rivers [18,36] significantly influence the spatial morphology of settlements. Additionally, settlement location is mainly affected by the elevation, distance to cultivated land, and distance to main roads, while settlement scale is mainly affected by slope, the relief degree of land surface, and the distance to urban centers [21].

Previous studies have predominantly focused on the spatial distribution of settlements in relatively large geographical areas. Moreover, spatial layout characteristics have mainly been presented through inter-settlement positional relationships, while research on individual settlements and their three-dimensional spatial features remains scarce. Research on the distribution patterns of ethnic minority settlements and their relationship with geographical conditions still needs to be enriched. Due to the dual influence of culture and region [37,38,39], the relationship between spatial layout characteristics and geographical features of ethnic minority rural settlements has become increasingly complex, and their interrelationships are also affected by multiple factors. Therefore, it is necessary to conduct relevant research at a more detailed level.

The aforementioned findings demonstrate a clear correlation between the site selection and layout characteristics of rural settlements and their geographical environments, which has been further validated through studies on ethnic minority settlements. These established correlations can scientifically guide the planning and development of ethnic minority rural settlements, facilitating both the preservation and revitalization of traditional culture and the sustainable utilization of territorial resources. Based on this, the present study focuses on ethnic minority rural settlements in southern China, investigating their influencing factors from the perspectives of both planar and spatial layout characteristics at the individual settlement level. This study examines 34 representative ethnic minority rural settlements located in the multicultural convergence zone of Liuzhou, Guangxi. This paper employs fractal theory and a GIS analysis method to obtain the planar morphology, spatial layout, and topographic characteristics of representative settlements. In addition, statistical analysis methods were applied to investigate the interrelationships among key factors including ethnic composition, the natural environment, settlement morphology, and location selection. The influencing patterns of these factors on the spatial layout characteristics of settlements are systematically summarized. On the one hand, the findings of this study reveal the correlations between planar morphological characteristics and spatial layout features of settlements. On the other hand, they provide both theoretical foundations and practical guidance for research on the spatial patterns and influencing factors of ethnic minority rural settlements in general regions of southern China.

2. Study Area Overview and Data Sources

2.1. Study Area Overview

Guangxi is a typical multi-ethnic settlement area in China, and is characterized not only by its diverse ethnic groups but also by its complex mountainous terrain. Various ethnic groups coexist here, forming a convergence zone of multiple cultures, including Baiyue culture, Miao-Yao culture, Central Plains culture, and marine culture [40,41,42,43]. Liuzhou is located in north-central Guangxi. The city’s mountainous and hilly areas are home to a significant population of ethnic groups, particularly the Han, Zhuang, Miao, and Dong peoples. The local ethnic minorities in Liuzhou are also representative of those in Guangxi. The distribution of various ethnic groups in Guangxi is illustrated in Figure 1.

2.2. The Physical-Geographical Conditions of Guangxi and Liuzhou

Guangxi is located in the south of China and is surrounded by numerous mountain ranges, resulting in a general topographic pattern of higher elevations along the periphery and lower terrain in the central region. Overall, Guangxi can be characterized as a predominantly mountainous region with limited plains, which means that the northern areas consist of more rugged highlands while the southern parts are composed of plains and basins, as illustrated in Figure 2.

Overall, the distribution patterns and influencing factors of ethnic minority rural settlements in Guangxi can be studied through two primary aspects: the region’s natural environment and its multicultural interactions. Firstly, the predominantly mountainous terrain with limited plains has resulted in uneven economic development across Guangxi. The western and northern regions of Guangxi are covered with predominantly mountainous terrain with limited flatland; in addition, the climatic conditions are complex and the transportation infrastructure is poor. Under these circumstances, the production and livelihood systems are relatively underdeveloped. However, in the hilly river valleys of Sanjiang Dong Autonomous County in northwestern Guangxi, where the terrain gradually tends to be gentle and water resources are abundant, agricultural and economic development is more robust. Secondly, with the development of Guangxi and the migration of the population, the early indigenous ethnic groups in Guangxi engaged in cultural exchange and integration with the Han residents and other ethnic minorities who moved in, resulting in the formation of multicultural convergence in the settlement pattern. For instance, in northeastern Guangxi, the indigenous Baiyue settlements gradually adopted the Han characteristics in their form and distribution due to the migration of Han people. Similarly, the Miao settlements in Rongshui Miao Autonomous County have adopted the form of tile-roofed houses in their residential buildings, no longer being limited to stilt houses.

2.3. Research Data Sources and Selection of Representative Settlements

Liuzhou in Guangxi is a multi-ethnic cultural convergence area with rich and diverse ethnic characteristics. The ethnic minority settlements in Liuzhou exhibit both typical and representative features, which can provide a sufficient data basis for this study. Therefore, Liuzhou is selected as the research area in this paper. This study focuses on several major ethnic groups with large populations in Liuzhou (Han, Zhuang, Miao, Dong, and Yao) and selects regions with significant ethnic integration as the study areas. Representative rural settlements were selected as research samples within these typical ethnic groups and regions. When selecting samples, we included both single-ethnic settlements and multi-ethnic mixed settlements, while also taking into account the distribution of each ethnic group in various counties.

The research data resources of this paper are divided into planar land data, digital elevation data, and social development data. The planar land data is derived from the land use data and geographic information software of Liuzhou in 2023. In this way, geographic information such as rural settlements, river systems, and road topography can be obtained, which is used for geographic feature acquisition. The digital elevation data is derived from a geospatial database. On this basis, the terrain analysis is carried out. The social development data are derived from the “Liuzhou Statistical Yearbook-2023”. Social data such as the population and economic development status of each village in the study area are extracted from the book for the analysis of influencing factors.

Table 1. Ethnic population proportions in the districts and counties of Liuzhou.

District (County)	Chengzhong District	Yufeng District	Liunan District	Liubei District	Liujiang District	Liucheng County	Luzhai County	Rong’an County	Rongshui County	Sanjiang County
Proportion of Han (%)	67.15	58.10	59.16	65.58	26.80	42.05	44.37	58.17	28.58	17.01
Proportion of Zhuang (%)	27.83	34.71	35.41	25.93	70.71	52.74	51.03	33.06	13.07	5.88
Proportion of Miao (%)	1.58	0.8	0.81	1.39	0.26	0.68	0.44	2.97	38.77	18.39
Proportion of Dong (%)	1.07	0.97	0.81	1.19	0.03	0.64	0.41	3.25	11.28	55.01
Proportion of Yao (%)	1.32	1.25	0.96	1.14	0.4	0.39	1.97	2.05	6.74	3.61

presents the ethnic composition and distribution across Liuzhou’s districts and counties. The data reveals that urban districts and Liucheng County are predominantly inhabited by Han and Zhuang populations, with other ethnic minorities constituting minimal proportions. In contrast, Rongshui Miao Autonomous County and Sanjiang Dong Autonomous County exhibit greater ethnic diversity, with Miao and Dong ethnic groups forming the major populations, respectively. Based on the aforementioned sample selection principles and the actual conditions of Liuzhou, we selected 34 representative rural settlements and obtained basic data for these settlements. The selected ethnic settlements in this study comprehensively represent these ethnic groups and regions, effectively reflecting the correlations between the spatial layout features and geographical conditions. The sample size fully meets research requirements. The spatial distribution of the 34 studied settlements is shown in Figure 3.

3. Materials and Methods

The purpose of this study is to investigate the correlation between geographic characteristics and spatial layout patterns for ethnic minority rural settlements. Therefore, it is necessary to describe the geographical conditions and the spatial layouts of the settlements, respectively. We applied fractal theory and GIS analysis methods to quantitatively characterize the planar morphological features and topographic conditions of the settlements, respectively [45,46]. Correlation analysis is a quantitative method, which can be applied to assess the degree of association between two variables. The application requires representing both geographic features and spatial layout characteristics of settlements as quantifiable variables. Therefore, we will classify these features into distinct categories and employ numerical coding for systematic characterization.

Firstly, this section employs fractal theory and GIS analysis methods, utilizing AutoCAD 2022 and ArcGIS 10.2 software to calculate various characteristic parameters of rural settlements in Figure 3. Based on multi-source data including the land use, digital elevation, and social development of Liuzhou, this paper measures both planar morphological features (such as the length–width ratio) and spatial distribution characteristics (such as the slope gradient and slope direction) of each rural settlement. This analytical process integrates both planar morphological features and spatial layout characteristics into the framework of settlement spatial layout. On this basis, we can analyze the influencing factors shaping the spatial patterns of ethnic minority rural settlements in Liuzhou.

3.1. Fractal Theory

Fractal theory, which studies complex and irregular geometric objects [47], has been increasingly applied in recent years to investigate the planar morphological characteristics of rural settlements [48,49]. Researchers have applied relevant concepts in fractal geometry and landscape ecology to extract elements such as residential buildings, roads, rivers and lakes in the settlement, thereby obtaining the geometric dimension information of each element. In this way, the values of quantitative indicators are derived based on the corresponding definitions and calculation formulas. Scholars classify the morphological characteristics of settlements based on the magnitude and range of indicator values. The specific indicators include the length–width ratio, shape index, circularity ratio, aggregation degree [50,51,52], etc.

3.2. GIS Analysis Method

As an integrated system of geographic information, the GIS synthesizes the spatial data of diverse regions and presents them intuitively. In recent years, this methodology has been widely adopted by researchers across multiple disciplines. This study employs quantitative indicators derived from the literature [53] and utilizes ArcGIS 10.2 software to extract, calculate, and visualize the geographic information of rural settlements listed in Figure 3, with a focus on elevation, slope gradient, slope direction, and surface characteristics. Based on the quantitative results, the characteristic elements of ethnic minority rural settlements are classified and represented, enabling the quantitative analysis of the spatial distribution patterns. The quantitative indicators used in this study to characterize the planar morphology and topographic conditions of rural settlements are presented in

Table 2. Calculation methods and meanings of planar and spatial quantitative indicators.

Indicators	Calculation Formula	Symbol Description	Meaning
Length–width ratio	$\lambda =b/a$	Constructing the minimum bounding rectangle aligned parallel to the longest axis of the settlement, where b is the length of the bounding rectangle and a is the width	It represents the elongated nature of the settlement boundary, indicating whether it tends to be clustered or banded
Altitude and elevation	H	H is the height of Earth’s surface or altitude	It represents the height information of the settlement surface
Settlement orientation	$\begin{array}{l}{A}_s=57.29578\times \\ {\mathrm{atan}}^2\left(\mathrm{d}z/\mathrm{d}y-\mathrm{d}z/\mathrm{d}x\right)\end{array}$	dz is the amount of distance change in the height direction, dx is the amount of distance change in the x-direction, dy is the amount of distance change in the y-direction	It represents the slope orientation of the settlement location
Settlement slope	$\begin{array}{l}{S}_d=\\ \mathrm{atan}\left(\sqrt{{(\mathrm{d}z/\mathrm{d}x)}^2+{(\mathrm{d}z/\mathrm{d}y)}^2}\right)\\ \times 57.29578\end{array}$	dz is the amount of distance change in the height direction, dx is the amount of distance change in the x-direction, dy is the amount of distance change in the y-direction	It indicates the terrain gradient (gentle or steep) at the settlement location
Ground fluctuation	$R_{Fi}=H_{\max }-H_{\min }$	$R_{Fi}$ is ground fluctuation, $H_{\max }$ is the maximum elevation value in the analysis window, $H_{\min }$ is the minimum elevation value	It represents the elevation difference (or relief) within a certain area, indicating the degree of terrain undulation
Ground roughness	$\begin{array}{l}R={\displaystyle \frac{S_{curve}}{S_{hori}}}\\ \approx {\displaystyle \frac{1}{\cos (S\times 3.14159/180)}}\end{array}$	$S_{curve}$ is the surface area of the ground, $S_{hori}$ is the projected area of the ground on the horizontal plane and S is the slope factor (in degrees) in the DEM file	It represents the degree of surface folding

. Among them, the length–width ratio is derived from settlement planar maps, while all other quantitative indicators are computed using ArcGIS 10.2 software. The subsequent analysis will classify the geographic condition types of settlements based on the calculated indicators. Notably, this classification does not merely rely on a single indicator, but rather on the combination of indicators—an approach previously studied in Reference [45].

3.3. Pearson Correlation Analysis

Some scholars have found that the topography of settlement locations has a relatively significant impact on settlement morphology. However, the correspondence between the specific type of terrain and settlement morphology is neither fixed nor unique. This is because, in addition to topography, settlement morphology is influenced by other natural and human geographical factors, such as altitude, river distance, settlement scale, slope gradient, architectural layout, and other factors. Therefore, this correlation feature analysis requires the sample size to reach a certain level. Under these circumstances, we can summarize and deduce the relationship between settlement patterns and influencing factors from the distribution trends of a large number of samples. The correlation analysis method makes it possible to achieve the research objectives of this paper. Correlation analysis can only be conducted if there is a certain inherent connection or probability among the elements of correlation. Through correlation analysis, we can not only determine whether there is a clear mathematical relationship between different characteristic factors but also assess the strength of this correlation. Through an analysis of the variable types in the study objects of this paper, we adopted the Pearson correlation analysis method. The Pearson correlation coefficient is a statistical measure that evaluates the degree of linear correlation between two continuous variables [23,24,25].

Building upon current research [30], this study establishes an evaluation index system for factors influencing the spatial layout of rural settlements. Based on the spatial scale of rural settlements, 11 indicators are selected, covering aspects such as ethnic composition, natural environment, settlement patterns, and site selection characteristics. Pearson correlation coefficients between these variables are then calculated to explore the influencing factors of spatial layout characteristics. In summary, the technical research routine of this paper is illustrated in Figure 4, and the specific meanings of the 11 indicators in Figure 4 will be detailed in Section 4.3.

4. Results

4.1. Topographic and Geomorphic Analysis of Rural Settlements in Liuzhou

The original data is the digital elevation model (DEM) with a resolution of 12.5 m, as shown in Figure 5. This section utilizes ArcGIS 10.2 software to calculate topographic and geomorphic indicators for Liuzhou and the results are presented in Figure 6. Based on field surveys and GIS analytical data, ethnic minority rural settlements in Liuzhou demonstrate remarkable geomorphological diversity. According to the spatial relationships between rural settlements and nearby mountains/rivers, these settlements can be categorized as follows, with representative examples illustrated in

Table 3. Natural geographical features of representative settlements.

Settlement Type	Representative Settlement	Satellite Map	Aerial View	Topographic Features	Drainage Characteristics
Flatland and riverbank type	Longmiao Tun (Zhuang)			located in gently rolling hills, with undulating terrain across the area	river lies on one side, spreading along the periphery
Valley and riverbank type	Pingmao Tun (Dong)			clustered in the valley, nestled against the mountains and opening to low-lying flat terrain	built along the riverbank, evenly distributed on the gentle slopes
Valley agglomeration type	Gandong Tun (Miao)			distributed in line with the terrain, situated at the lowest elevation point of the area	determined by geographical features
Valley slope arrangement type	Gaoding Village (Dong)			distributed in both valleys and mountainous areas	far from the water source
Mountainside arrangement type	Linwang Tun (Yao)			built on a mountainside, with dwellings cascading down the slope	determined by the geographical terrain
Mountaintop agglomeration type	Songmei Tun (Miao)			built on the mountaintop and generally arranged along the contour lines	far from the water source

.

4.2. The Planar and Spatial Layout Characteristics of Rural Settlements

The traditional settlements in Liuzhou, Guangxi exhibit diverse overall morphological patterns. The length–width ratio serves as the most representative indicator of their distribution characteristics, allowing these settlements to be broadly categorized into three types: a clustered type, a clustered-banded type, and a banded type [45]. Moreover, the layout of these streets and alleys significantly influences the overall morphology of the settlements. The street and alley structures of traditional settlements in Liuzhou can generally be categorized into three types: a checkerboard pattern, a branch pattern, and an irregular network pattern. Detailed classifications with corresponding examples are presented in

Table 4. Street and alley structures of rural settlements.

Type	Checkerboard Pattern	Branch Pattern	Irregular Network Pattern
Illustration
Features	The road network unfolds in alignment with the layout of the settlement’s dwellings, exhibiting a well-ordered pattern with straight horizontal and vertical orientations.	The roads within the settlement serve to connect individual households, featuring a main thoroughfare that links multiple branch roads.	The road network within the settlement is disorganized, lacking clearly defined main or branch roads, though all paths remain interconnected.
Representative settlement	Yingshan Community (Zhuang/Han)	Guoli Tun (Miao)	Pingmao Tun (Dong)

. In this study, the spatial layout characteristics of rural settlements are mainly reflected in the slope gradient and slope direction, as well as the settlement siting patterns and topographic position. Specifically, settlement siting focuses on the mountain–water spatial configuration around the settlements, while settlement terrain examines the relative positional relationship between settlements and the mountains where they are located. The classification and characterization of these spatial layout features have been comprehensively elaborated on in the literature [46], and thus will not be reiterated here.

4.3. Classification and Feature Extraction of Ethnic Minority Rural Settlements in Liuzhou, Guangxi

Based on the preceding analysis of the natural environment and spatial layout characteristics of traditional ethnic minority settlements in Liuzhou, this study now proceeds to classify the spatial morphology of these rural settlements. Building upon previous research on the fundamental characteristics of traditional settlements in Liuzhou [53,54], this study selects 11 feature elements listed in

Table 5. Characteristic elements and numerical assignment.

Sequence Number	Characteristic Elements	Meaning	Numerical Assignment for the Elements
1	EC	The ethnic group component within the settlement	Han = 1; Zhuang = 2; Miao = 3; Dong = 4; Yao = 5; Multi-ethnic mixture = 6
2	TO	The topographic features of the settlement location	Flatland = 1; Hill = 2; Mountain = 3
3	AE	The elevation of the settlement location	90~150 m = 1; 150~200 m = 2; 200~450 m = 3; 450~550 m = 4; 550~700 m = 5; over 700 m = 6
4	RP	The spatial relationship between settlements and rivers	Leaning type = 1; Surrounding type = 2; Separated type = 3; Distant type = 4
5	SS1	The population size of the settlement	0~100 people = 1; 101~300 people = 2; 301~600 people = 3; 601~1000 people = 4; over 1000 people = 5
6	SO	The orientation of the settlement buildings	Facing east = 1; Facing south = 2; Facing west = 3; Facing north = 4; Facing east and west = 5; Facing south and north = 6; Facing all directions = 7
7	SS2	The slope of the settlement	Flat slope (0~5°) = 1; Gentle slope (6~15°) = 2; Inclined slope (16~25°) = 3; Steep slope (26~35°) = 4
8	PM	The architectural form within the settlement	Clustered type = 1; Clustered-banded type = 2; Banded type = 3; Atypical type = 4
9	SP	Distribution characteristics of streets and alleys	Checkerboard pattern = 1; Branch pattern = 2; Irregular network pattern = 3
10	SSS	Site selection methods for settlement residences	River valley and flatland type = 1; Mountain and river surrounding type = 2; Mountain intersection type = 3
11	TS	The spatial relationship between settlement architecture and the mountain	Flatland and riverside type = 1; Valley and riverside type = 2; Valley agglomeration type = 3; Valley slope arrangement type = 4; Mountainside arrangement type = 5; Mountaintop agglomeration type = 6

In the table, the full names of the acronyms are as follows: EC = Ethnic composition; TO = Topography; AE = Altitude and elevation; RP = River proximity; SS1 = Settlement scale; SO = Settlement orientation; SS2 = Settlement slope; PM = Planar morphology; SP = Street pattern; SSS = Site selection of settlement; TS = Terrain of settlement. The same applies to the following tables.

. By applying fractal theory and GIS analytical methods to calculate settlement distribution and spatial characteristics, specific value ranges are determined for each feature element, with numerical assignments as shown in Table 5. Based on the specific number of categories under each characteristic element, we represent them with an equal number. In Table 5, the assigned values represent the specific ranges of each characteristic element, and their magnitudes are also positively correlated with the values of each quantitative indicator. For indicators that cannot be quantified specifically, such as river proximity, settlement orientation, and planar morphology, values are assigned according to a definite qualitative sequence (e.g., river proximity is assigned in the order from near to far distance, and planar morphology is assigned in the order from a clustered to a banded pattern).

In this section, the characteristic elements of 34 field-surveyed traditional settlements are assigned values according to Table 5, and they are listed separately by county, as shown in

Table A1. Numerical assignment of characteristic elements for each settlement in Luzhai County.

No.	Settlement	EC	TO	AE	RP	SS1	SO	SS2	PM	SP	SSS	ST
1	Zhaishang Tun (Zhuang/Han)	6	2	2	3	5	7	1	3	2	1	1
2	Yingshan Community (Zhuang/Han)	6	2	1	2	5	7	1	1	1	1	1
3	Baodi Tun (Zhuang/Han)	6	2	1	4	2	7	1	1	1	1	1

,

Table A2. Numerical assignment of characteristic elements for each settlement in Rong’an County.

No.	Settlement	EC	TO	AE	RP	SS1	SO	SS2	PM	SP	SSS	ST
1	Tongban Tun (Han)	1	2	3	4	1	2	3	1	3	2	5
2	Gaoyang Tun (Han)	1	2	1	3	2	2	2	3	3	2	4
3	Longmiao Tun (Zhuang)	2	2	1	1	4	3	2	3	2	1	1
4	Xigupo Tun (Zhuang)	2	2	2	1	1	2	2	2	2	1	5
5	Dapao Tun (Miao)	3	2	2	2	3	5	3	2	2	1	2

,

Table A3. Numerical assignment of characteristic elements for each settlement in Rongshui Miao Autonomous County.

No.	Settlement	EC	TO	AE	RP	SS1	SO	SS2	PM	SP	SSS	ST
1	Baima Tun (Zhuang)	2	1	1	1	3	7	1	2	3	1	1
2	Xizhai Tun (Zhuang)	2	1	1	1	3	7	1	3	3	1	1
3	Liaodong Tun (Zhuang)	2	3	3	1	4	1	2	3	2	2	2
4	Pingmao Tun (Dong)	4	2	3	2	5	1	3	1	2	1	2
5	Xinzhai Tun (Dong)	4	3	4	3	3	7	1	1	3	2	2
6	Rongdi Village (Dong)	4	3	6	3	2	1	2	1	2	2	5
7	Gandong Tun (Miao)	3	2	5	1	5	1	2	1	3	3	3
8	Guoli Tun (Miao)	3	3	5	4	3	1	3	3	2	3	6
9	Songmei Tun (Miao)	3	3	6	4	4	2	3	3	3	3	6
10	Wuying Tun (Miao)	3	2	5	1	3	4	3	2	2	2	5
11	Peike Tun (Miao)	3	3	5	3	5	6	2	2	3	2	5
12	Jiman Tun (Miao)	3	3	5	3	5	2	2	2	3	3	4
13	Wuji Village (Miao)	3	3	5	3	5	2	2	2	3	2	5
14	Linwang Tun (Yao)	5	3	6	3	3	4	2	1	2	2	4
15	Dazhai Tun (Yao)	5	3	6	4	2	1	3	1	2	2	5
16	Zhengdiao Tun (Yao)	5	3	5	3	2	1	3	2	2	2	5

and

Table A4. Numerical assignment of characteristic elements for each settlement in Sanjiang Dong Autonomous County.

No.	Settlement	EC	TO	AE	RP	SS1	SO	SS2	PM	SP	SSS	ST
1	Gaoding Village (Dong)	4	3	4	4	5	4	2	1	3	3	3
2	Gaoyou Village (Dong)	4	2	4	4	5	5	4	2	2	3	4
3	Guandong Village (Dong)	4	3	2	2	5	7	2	3	2	2	2
4	Chengyang Village (Dong)	4	3	2	2	5	7	2	4	2	3	3
5	Mozhai Tun (Dong)	4	3	3	1	5	1	2	3	2	2	2
6	Zhiliao Tun (Dong)	4	3	5	4	3	2	3	3	2	3	6
7	Pingxi Tun (Dong/Miao)	6	3	2	1	4	4	2	3	3	2	4
8	Laoba Village (Miao)	3	3	4	3	5	4	3	1	3	2	5
9	Geliang Tun (Miao)	3	3	2	1	4	4	2	3	2	2	2
10	Menglong Tun (Miao/Yao)	6	3	4	4	3	4	3	3	2	2	5

in Appendix A.

4.4. Research on Correlation Analysis and Influence Laws

This section employs SPSS 19.0 to conduct bivariate correlation analysis on 11 quantitative indicators across 34 research samples, including ethnic composition, topography, elevation, river proximity, settlement scale, settlement orientation, settlement slope, planar morphology, street pattern, settlement site selection, and settlement terrain (as detailed in

Table 6. Correlation analysis of characteristic elements in traditional settlements and residences. Data source: processing results of SPSS software.

Characteristic Elements (Quantitative Indicators)		EC	TO	AE	RP	SS1	SO	SS2	PM	SP	SSS	ST
EC	Pearson correlation	1	0.310	0.116	0.269	0.188	0.268	−0.092	−0.128	−0.502 **	−0.041	−0.091
	Significance (two-tailed)	——	0.075	0.513	0.124	0.287	0.126	0.603	0.47	0.002	0.817	0.609
TO	Pearson correlation	0.310	1	0.588 **	0.352 *	0.203	−0.324	0.275	0.102	0.015	0.586 **	0.492 **
	Significance (two-tailed)	0.075	——	0.000	0.041	0.250	0.061	0.116	0.568	0.934	0.000	0.003
AE	Pearson correlation	0.116	0.588 **	1	0.482 **	−0.001	−0.514 **	0.505 **	−0.301	0.193	0.622 **	0.738 **
	Significance (two-tailed)	0.513	0.000	——	0.004	0.996	0.002	0.002	0.084	0.274	0.000	0.000
RP	Pearson correlation	0.269	0.352 *	0.482 **	1	−0.170	−0.091	0.375 *	−0.255	0.008	0.411 *	0.509 **
	Significance (two-tailed)	0.124	0.041	0.004	——	0.337	0.611	0.029	0.146	0.966	0.016	0.002
SS1	Pearson correlation	0.188	0.203	−0.001	−0.170	1	0.214	−0.077	0.164	0.114	0.249	−0.282
	Significance (two-tailed)	0.287	0.250	0.996	0.337	——	0.225	0.667	0.355	0.520	0.156	0.106
SO	Pearson correlation	0.268	−0.324	−0.514 **	−0.091	0.214	1	−0.525 **	0.076	−0.099	−0.336	−0.547 **
	Significance (two-tailed)	0.126	0.061	0.002	0.611	0.225	——	0.001	0.669	0.577	0.052	0.001
SS2	Pearson correlation	−0.092	0.275	0.505 **	0.375 *	−0.077	−0.525 **	1	0.008	−0.018	0.450 **	0.666 **
	Significance (two-tailed)	0.603	0.116	0.002	0.029	0.667	0.001	——	0.966	0.921	0.008	0.000
PM	Pearson correlation	−0.128	0.102	−0.301	−0.255	0.164	0.076	0.008	1	−0.017	0.144	−0.034
	Significance (two-tailed)	0.47	0.568	0.084	0.146	0.355	0.669	0.966	——	0.926	0.416	0.847
SP	Pearson correlation	−0.502 **	0.015	0.193	0.008	0.114	−0.099	−0.018	−0.017	1	0.310	0.245
	Significance (two-tailed)	0.002	0.934	0.274	0.966	0.520	0.577	0.921	0.926	——	0.074	0.162
SSS	Pearson correlation	−0.041	0.586 **	0.622 **	0.411 *	0.249	−0.336	0.450 **	0.144	0.310	1	0.580 **
	Significance (two-tailed)	0.817	0.000	0.000	0.016	0.156	0.052	0.008	0.416	0.074	——	0.000
ST	Pearson correlation	−0.091	0.492 **	0.738 **	0.509 **	−0.282	−0.547**	0.666 **	−0.034	0.245	0.580 **	1
	Significance (two-tailed)	0.609	0.003	0.000	0.002	0.106	0.001	0.000	0.847	0.162	0.000	——

Note: coefficients exceeding 0.3 with significance levels below 0.05 meet statistical requirements. Correlation coefficients range between 0 and 1, with values closer to 1 indicating a stronger pairwise correlation. ** Correlation is significant at the 0.01 level (two-tailed); * correlation is significant at the 0.05 level (two-tailed).

). Furthermore, given the substantial number of Miao settlements in Rongshui Miao Autonomous County and Dong settlements in Sanjiang Dong Autonomous County, this study specifically focuses on settlements within these two counties.

Table 7. Correlation analysis of characteristic elements for Miao settlements and residences in Rongshui Miao Autonomous County. Data source: processing results of SPSS software.

Characteristic Elements (Quantitative Indicators)		TO	AE	RP	SS1	SO	SS2	PM	SP	SSS	ST
TO	Pearson correlation	1	0.258	0.934 **	0.205	0.027	−0.091	0.636	0.300	0.091	0.548
	Significance (two-tailed)	——	0.576	0.002	0.659	0.954	0.846	0.124	0.513	0.846	0.203
AE	Pearson correlation	0.258	1	0.452	−0.132	−0.139	0.471	0.548	0.258	0.354	0.471
	Significance (two-tailed)	0.576	——	0.308	0.777	0.766	0.286	0.203	0.576	0.437	0.286
RP	Pearson correlation	0.934 **	0.452	1	−0.060	−0.136	0.213	0.826 *	0.117	0.284	0.711
	Significance (two-tailed)	0.002	0.308	——	0.898	0.771	0.646	0.022	0.803	0.537	0.073
SS1	Pearson correlation	0.205	−0.132	−0.060	1	0.083	−0.937 **	−0.580	0.923 **	−0.047	−0.609
	Significance (two-tailed)	0.659	0.777	0.898	——	0.860	0.002	0.172	0.003	0.921	0.147
SO	Pearson correlation	0.027	−0.139	−0.136	0.083	1	−0.123	−0.076	0.027	−0.737	0.135
	Significance (two-tailed)	0.954	0.766	0.771	0.860	——	0.793	0.871	0.954	0.059	0.773
SS2	Pearson correlation	−0.091	0.471	0.213	−0.937 **	−0.123	1	0.710	−0.730	0.167	0.708
	Significance (two-tailed)	0.846	0.286	0.646	0.002	0.793	——	0.074	0.062	0.721	0.075
PM	Pearson correlation	0.636	0.548	0.826 *	−0.580	−0.076	0.710	1	−0.354	0.194	0.936 **
	Significance (two-tailed)	0.124	0.203	0.022	0.172	0.871	0.074	——	0.437	0.677	0.002
SP	Pearson correlation	0.300	0.258	0.117	0.923 **	0.027	−0.730	−0.354	1	0.091	−0.411
	Significance (two-tailed)	0.513	0.576	0.803	0.003	0.954	0.062	0.437	——	0.846	0.360
SSS	Pearson correlation	0.091	0.354	0.284	−0.047	−0.737	0.167	0.194	0.091	1	−0.125
	Significance (two-tailed)	0.846	0.437	0.537	0.921	0.059	0.721	0.677	0.846	——	0.789
ST	Pearson correlation	0.548	0.471	0.711	−0.609	0.135	0.708	0.936 **	−0.411	−0.125	1
	Significance (two-tailed)	0.203	0.286	0.073	0.147	0.773	0.075	0.002	0.360	0.789	——

Note: coefficients exceeding 0.3 with significance levels below 0.05 meet statistical requirements. Correlation coefficients range between 0 and 1, with values closer to 1 indicating a stronger pairwise correlation. ** Correlation is significant at the 0.01 level (two-tailed); * correlation is significant at the 0.05 level (two-tailed).

and

Table 8. Correlation analysis of characteristic elements for Dong settlements and residences in Sanjiang Dong Autonomous County. Data source: processing results of SPSS software.

Characteristic Elements (Quantitative Indicators)		TO	AE	RP	SS1	SO	SS2	PM	SP	SSS	ST
TO	Pearson correlation	1	−0.270	−0.430	−0.200	−0.130	−0.878 *	0.316	0.200	−0.316	−0.217
	Significance (two-tailed)	——	0.605	0.385	0.704	0.805	0.021	0.541	0.704	0.541	0.680
AE	Pearson correlation	−0.270	1	0.787	−0.674	−0.638	0.592	−0.533	0.270	0.533	0.804
	Significance (two-tailed)	0.605	——	0.063	0.142	0.173	0.216	0.276	0.605	0.276	0.054
RP	Pearson correlation	−0.430	0.787	1	−0.430	−0.040	0.629	−0.631	0.430	0.777	0.733
	Significance (two-tailed)	0.385	0.063	——	0.395	0.940	0.181	0.179	0.395	0.069	0.097
SS1	Pearson correlation	−0.200	−0.674	−0.430	1	0.457	−0.293	−0.158	0.200	−0.316	−0.868 *
	Significance (two-tailed)	0.704	0.142	0.395	——	0.363	0.573	0.765	0.704	0.541	0.025
SO	Pearson correlation	−0.130	−0.638	−0.040	0.457	1	−0.095	0.206	−0.065	0.103	−0.301
	Significance (two-tailed)	0.805	0.173	0.940	0.363	——	0.857	0.695	0.902	0.846	0.563
SS2	Pearson correlation	−0.878 *	0.592	0.629	−0.293	−0.095	1	−0.231	−0.293	0.463	0.635
	Significance (two-tailed)	0.021	0.216	0.181	0.573	0.857	——	0.659	0.573	0.355	0.175
PM	Pearson correlation	0.316	−0.533	−0.631	−0.158	0.206	−0.231	1	−0.791	−0.250	−0.043
	Significance (two-tailed)	0.541	0.276	0.179	0.765	0.695	0.659	——	0.061	0.633	0.936
SP	Pearson correlation	0.200	0.270	0.430	0.200	−0.065	−0.293	−0.791	1	0.316	−0.108
	Significance (two-tailed)	0.704	0.605	0.395	0.704	0.902	0.573	0.061	——	0.541	0.838
SSS	Pearson correlation	−0.316	0.533	0.777	−0.316	0.103	0.463	−0.250	0.316	1	0.686
	Significance (two-tailed)	0.541	0.276	0.069	0.541	0.846	0.355	0.633	0.541	——	0.132
ST	Pearson correlation	−0.217	0.804	0.733	−0.868 *	−0.301	0.635	−0.043	−0.108	0.686	1
	Significance (two-tailed)	0.680	0.054	0.097	0.025	0.563	0.175	0.936	0.838	0.132	——

Note: coefficients exceeding 0.3 with significance levels below 0.05 meet statistical requirements. Correlation coefficients range between 0 and 1, with values closer to 1 indicating a stronger pairwise correlation. * correlation is significant at the 0.05 level (two-tailed).

present the correlation analysis results for Rongshui and Sanjiang autonomous counties, respectively. Since the data table for correlation analysis is symmetrical about the diagonal, only the data in the lower left corner are analyzed in this section. Statistically significant correlations are marked by bold fonts in the table for emphasis.

As shown in Table 6, when considering the traditional settlements of various ethnic groups and regions in Liuzhou as a whole, the sample size is sufficient to reflect the correlations among spatial layout characteristics. As the elevation of settlement locations progressively increases, the terrain gradually transitions from plains to hills and mountains. Concurrently, the distance between settlements and river systems grows larger, the slope gradient of settlement sites becomes steeper, and slope directions shift from dispersed to concentrated patterns. Notably, the evolutionary trend of slope directions is north→west→south→east. With the increase in settlement elevation, the site selection of settlements gradually shifts from valley floors to mountain peaks. Moreover, a significant positive correlation exists between the specific position of residential buildings on mountain slopes and the local elevation. It should be noted that the terrain of dwellings differs conceptually from their geographic location—even within the same mountainous area, diverse patterns emerge in site selection and spatial organization. Similarly, for valley-clustered settlements, their topographic and geomorphic conditions also vary significantly. In terms of street network patterns, distinct differences emerge among different ethnic groups. In Han and Zhuang mixed settlements, the street patterns are more regular, while Miao/Dong and Miao/Yao mixed settlements show irregular and chaotic street patterns. The remaining Miao, Dong, and Yao ethnic settlements are primarily characterized by dendritic and irregular network patterns. No significant correlation was observed between settlement scale/planar morphology and the aforementioned factors for the selected settlements, indicating dispersed distribution patterns across settlements at different elevations and topographic conditions. In summary, the correlation analysis demonstrates a clear relationship between the siting characteristics and their geographical environment. This indicates that natural environmental factors have significant influences on settlement location selection. Furthermore, the aforementioned correlation analysis has quantitatively substantiated this relationship.

Table 7 and Table 8 present the correlation analysis results conducted within the designed geographical regions and ethnic groups. Compared with Table 6, the current conclusions with clear correlations are fewer, but Table 7 and Table 8 demonstrate certain associations in specific aspects that were not revealed in the overall study. For instance, in Miao settlements of Rongshui Miao Autonomous County, a distinct correlation exists between the planar morphology and river proximity. As the distance from rivers increases, settlement configurations transition progressively from clustered patterns to banded patterns. In addition, the settlement scale has a relatively obvious impact on the slope and the layout of the streets and alleys where the settlement is located. Settlement scale shows significant negative correlation with slope gradient and positive correlation with street network complexity, which means that larger settlements would occupy gentler slopes and develop more disordered street patterns. As the elevation of settlement increases, its layout gradually shifts from a clustered form to a linear strip-like pattern. In the Dong settlements of Sanjiang Dong Autonomous County, there is a significant negative correlation between topography and settlement slope. Specifically, as the landform transitions from hills to mountains, the slope gradient of the settlement location becomes gentler. Additionally, a notable negative correlation exists between settlement scale and settlement terrain: the higher the settlement terrain, the smaller the scale of the settlement.

The weakening of correlation conclusions in Table 7 and Table 8 indicates that settlements within the same region and ethnic group demonstrate high degrees of similarity in these characteristic elements. For instance, the Dong settlements in Sanjiang Dong Autonomous County generally exhibit relatively large-scale layouts, with most adopting dendritic street networks. These settlements are predominantly sited at mountain–water intersections or along foothills. However, significant variations exist in their planar morphology and in their spatial relationships with rivers. The Miao settlements in Rongshui Miao Autonomous County generally occupy higher elevations, demonstrating notable consistency in slope gradients, street network patterns, and settlement siting. However, significant variations are observed in their planar morphology and spatial relationships with adjacent rivers. These findings indicate that even under similar topographic and site-selection conditions, different settlements can exhibit distinct planar morphological characteristics. This underscores the necessity of studying planar morphological features and their correlations with geographic elements, thereby verifying that planar and spatial characteristics form an organic entirety.

Based on the comprehensive correlation analysis, we can conclude that there exists a strong association between the spatial layout characteristics and geographical features. Although different regions and ethnic groups exhibit distinct influencing patterns, the overall trend of influence remains largely consistent. Firstly, the specific position of settlements on the mountains emerges as the primary determinant, exhibiting a positive correlation with elevation. Secondly, the settlement site selection of different ethnic groups is determined by their unique historical and cultural backgrounds. In addition, there exists a significant correlation between settlement slope gradient/direction and settlement terrain. Moreover, settlements located at higher relative positions on mountains tend to be situated farther from water systems. Thirdly, for single-ethnic settlements within a specific region, there exists a strong correlation between their planar morphological characteristics (or settlement scale) and the terrain features of the settlement site.

5. Discussion

This paper investigates the influencing factors of spatial layout characteristics in representative ethnic minority rural settlements of Liuzhou, conducting analysis from two distinct perspectives: the overall settlements and the settlements of the same ethnic group in the same region. However, the ethnic minority rural settlements in Liuzhou exhibit multi-ethnic coexistence characteristics, making them affected by dual influences from both geographical and ethnic factors [53,54]. Previous study on human–land relationships in multi-ethnic symbiotic settlements has adopted coupling approaches to incorporate various interrelationships, thereby addressing the limitations of single-factor analysis in superficial research [55].

This study selects 34 ethnic minority settlements within representative ethnic groups and regions as research samples. The sample size is sufficient to demonstrate the proposed methodology and meets the accuracy and rigor requirements for analyzing the correlation between spatial layout characteristics and geographical conditions for ethnic minority settlements. As evidenced in Table 7 and Table 8, the number of statistically significant correlations can be improved further. To enhance the comprehensiveness of research findings, it is recommended to increase the sample size of Miao and Dong settlements in Rongshui Miao Autonomous County and Sanjiang Dong Autonomous County, respectively. Furthermore, to enhance the scientific validity and generalizability of the research findings, the selection of settlement samples should thoroughly consider the complexity and diversity of topographic conditions. This approach will ensure that the research samples possess broader representativeness.

Section 4 investigates the correlation between spatial patterns and geographical features within settlements of the same ethnic group in the same area. However, considering that China’s ethnic minority settlements are typically distributed across multiple locations, it is worth further studying whether the spatial layout of the settlements within the same ethnic group will present completely different characteristics by integrating local customs and traditions after villagers migrate to different areas. One research approach involves conducting horizontal qualitative comparisons. This method is particularly suitable for settlements exhibiting common characteristics in spatial layout. By comparing the spatial features of settlements of the same ethnic group in different regions, whether differences exist can be determined. For settlements with dispersed spatial layout characteristics, quantitative research should be conducted. This involves employing correlation analysis to examine the relationship between the layout features and geographic conditions for settlements within the same ethnic group across different regions. By identifying both commonalities and distinctions in these correlations, relevant conclusions can be subsequently drawn.

This study focuses on Liuzhou, Guangxi, but the research methodology is not constrained by ethnic or regional factors. It remains applicable whether it is employed in single-ethnic regions or multi-ethnic settlements. The key distinction lies in separating ethnic factors based on the actual situation. Therefore, the research methods in this paper can be promoted and applied to future studies in other ethnic groups. The planar maps of settlements and the topographic data of their surrounding areas exhibit limited variations within a certain time frame. Since this study primarily focuses on the variation range of data and the correlations between datasets, these minor variations will not affect the validity of the conclusions. The research approaches in this paper are applicable within a certain ethnic group, region, and time range. While the research data may vary over time, such variation will not affect the applicability of the methods in this paper.

6. Conclusions

This study takes Liuzhou in Guangxi Zhuang Autonomous Region as a case study, and applies fractal theory and a GIS analysis method to calculate the planar morphological and spatial layout features of representative settlements. In terms of spatial morphology types, this study further extracts relevant factors influencing spatial layout characteristics and provides numerical assignment. Using Pearson correlation analysis, the research establishes the relationships between spatial morphological features and physical-geographical factors for traditional settlements. Furthermore, it analyzes the influencing patterns of various elements on settlement spatial morphology. The main research findings are summarized as follows:

(1) The ethnic minority traditional settlements in Liuzhou have evolved under the combined influence of multiple factors including transportation networks, topography, and rivers. When analyzing the traditional settlements of Han, Zhuang, Miao, Dong, and Yao ethnic groups as an integrated research entity, the study identifies significant correlations between their settlement siting/terrain characteristics and both geomorphological and altitude features.

(2) Given significant variations across ethnic groups and geographical regions, coupled with the wide distribution of planar characteristics and topographic features, this study specifically examines Miao settlements in Rongshui Miao Autonomous County and Dong settlements in Sanjiang Dong Autonomous County as separate case studies. The results demonstrate significant correlations between planar characteristics and topographic features in Miao settlements of Rongshui Miao Autonomous County. As mountain elevations increase and the distance from river systems grows, settlement planar morphology gradually transitions from clustered to linear patterns. For the Dong settlements in Sanjiang Dong Autonomous County, strong negative correlations exist between topography and settlement slopes, as well as those between settlement scale and terrain characteristics.

(3) The study reveals that rural settlements within the same region and ethnic group demonstrate a certain degree of concentration in terms of elevation, site selection, and terrain features, while exhibiting significant variations in river proximity and planar morphology. Furthermore, comparative analysis shows substantial differences in spatial layout between different ethnic settlements, which are influenced not only by physical-geographical factors but are also shaped by distinct ethnic cultures. This study establishes a general framework for investigating the influencing factors of spatial layout characteristics in ethnic minority rural settlements across southern China.

Based on the research on ethnic minority rural settlements, the following recommendations should be implemented for their planning and conservation: (1) The site selection characteristics of various ethnic settlements demonstrate significant diversity. Consequently, protection planning for different topographic regions should adopt unified standards. (2) When conducting new village planning, it is essential to incorporate the general characteristics of corresponding ethnic groups. For instance, Dong settlements typically feature larger scales, while Miao settlements are generally situated at higher elevations. Based on this, scientific planning can be conducted. (3) Areas adjacent to water sources and low-lying terrains constitute high-frequency settlement zones. It is necessary to strengthen environmental management and transportation construction to improve land utilization efficiency and residents’ quality of life.