Exploring Village Spatial Patterns for Sustainable Development: A Case Study of Diqing Prefecture

Abstract

Alexander’s A Pattern Language is an important text and focuses on the theory of diverse environmental spatial sustainability. With the contemporary digital development of villages, it is urgent that village spatial patterns are analyzed in a scientific and quantitative way in order to determine heritage village diversity. The village settlements in the Diqing region are typical representatives, having a changeable terrain, being large in number, and being multi-ethnic in China; in recent years, they have also faced slow development and limited conditions. However, few studies have focused on the multiple quantitative analysis of the diverse spatial patterns of village settlements in an ethnic minority region. Therefore, this study selects 2486 village settlements in Diqing and, using KED, NNI SSIA, etc., proposes a spatial pattern analysis framework (SPAF) based on pattern language theory. According to the spatial influencing factors, spatial analysis criteria are constructed to analyze the village spatial pattern types and subtypes. The results show that the region’s topographic conditions are the dominant factors that form the diversified village spatial patterns existent in the Diqing Prefecture. Among them, the dominant pattern of building villages along slopes with a small-population scale and large-dispersed settlements achieves a healthy and sustainable living environment that is oriented well, cost-saving, and conforms to nature. Meanwhile, the dominant pattern is also the reason for the inhibition of development due to inconvenient transportation and difficult management. Therefore, sustainable strategies should strike a balance between the two opposites. Based on the SPAF, spatial patterns can be effectively extracted for diverse village spaces, providing digital and visual references for the regeneration of contemporary rural areas.

1. Introduction

The spatial patterns of village settlements are an important part of the contemporary living environment. Research and protection can promote the common development of urban and rural areas and maintain and enhance the local ecological resources and environment [1]. Village settlements, as the material carrier of Chinese agricultural civilization, play a key role in serving the environment of local living space and ensuring sustainable cultural inheritance [2]. However, Yunnan village settlements are affected by ethnic minorities and the changing natural environment, reflecting the typical cultural identity of diverse local construction and ecological sustainability (

Table 1. Yunnan ethnic minority village settlement; edited by authors, pictures were photographed by the authors.

Type of Village	Yunnan Hani Ethnic Village Settlement	Yunnan Yi Ethnic Village Settlement
Ecologically sustainable intelligence	Mushroom house form, built with local thatch and rammed earth, with a variety of spatial textures, utilizing the terraced landscape for farming in the mountains.	Soil palm room form, built by using Yunnan Huangtu rammed earth, which is earthquake-resistant and connected to the roof, solving the problem of drying agricultural work in mountainous areas.

). This is particularly the case in the Diqing Tibetan Autonomous Prefecture, which is located in the southern part of the Qinghai-Tibet Plateau, bordering Yunnan, Sichuan, and Tibet. Among the 56 ethnic groups in China, 26 live in Diqing, including Zang, Lisu, Naxi, Bai, etc. Due to its geographical features and special cultural context, the village settlement planning pattern in Diqing is a typical example of multi-ethnic integration and environmentally sustainable adaptation (Figure 1) [3,4,5]. Nowadays, due to the rapid modernization and urbanization of China, village areas have shifted from experiencing inside growth to external forced development. This phenomenon has led to changes in the village population structure and movement towards urban areas, resulting in the fading of traditional cultural context [6]. Determining how to conserve village’s tangible form and culture has become an urgent issue in recent years. Diqing’s village settlement, as a typical case with a slow pace, large number, and high complexity, urgently needs to be protected using spatial digital information. The contemporary development of village settlements requires a reasonable quantity of spatial data for the further effective use and development of local resources, such as village data, natural resource data, spatial pattern data [7], etc. Therefore, by taking Diqing as the study area in order to determine the village spatial pattern, we propose exploring the sustainable context of village space and design of future human settlements in such complex geographical regions [8]. The spatial pattern of village settlements is formed under the comprehensive influence of various factors. In order to analyze the settlements’ spatial pattern, the unique terrain, road, and, river network of the area must be understood [7,9].

In response to the absence and disappearance of cultural ecology caused by industrialization at the end of the 20th century, Alexander proposes a research methodology that uses pattern languages to inspire design paths in the service of environment. However, limited to the context of Alexander’s age, pattern theory lacks practical application in village planning. Nonetheless, the theory satisfies the contemporary need for the sustainable preservation of diverse villages. Therefore, this study takes 2486 villages in Diqing as an example and proposes a spatial pattern analysis framework (SPAF) based on pattern language theory using KED, NNI, and SSIA to digitize and visualize the spatial pattern. This paper is organized as follows: firstly, Alexander’s pattern language theory and the necessity of this paper’s theoretical extension of it are explained through the literature review; secondly, the characteristics of Diqing, the research object, and the research method are demonstrated; thirdly, the spatial pattern of Diqing’s village settlements and its sustainable ecological intelligence are analyzed from three perspective; and finally, the results are summarized. These spatial patterns are the corresponding principles for contemporary village conservation and sustainable development.

2. Literature Review

In terms of trends in the research, through the retrieval of village-settlement-related literature in the past five years from WOS (Web of Science), 8644 articles were found (Figure 2). Among them, only 624 articles in the fields of architecture and urban planning related to the spatial form of the village. Citespace (5.1.R8 SE). software was used to analyze the high-frequency keywords in the articles (Figure 3); the smaller the number from 0 to 7, the higher the frequency of keywords [10]. Among them, informal settlements and big data were found to be the focus of attention in recent years. According to the analysis of the keyword timeline (Figure 4), these research areas were highly popular in 2019 and continue to have an impact on contemporary village settlements. This indicates that spatial analysis is an essential method and that research on village settlements presents an interdisciplinary trend, thus requiring a focus on the spatial diversity of village settlements [11,12,13,14].

2.1. A Pattern Language and Village Spatial Patterns

In the early 20th century, the research on village settlement form patterns was focused on the architectural form within the village and the method primarily used was typology; this classified the buildings according to their style, structure, and other material attributes [15]. Under the popularization of information technology, the empirical evidence on village spatial research emphasizes the integration of environment and space. Christopher Alexander proposed a methodology for spatial research in this context; this was a pattern language which collects 253 different spatial patterns and formulates reasonable design principles [16]. The pattern pointed out by Alexander is the principle of spatial type, which is the typical representative and basis for environment planning [17]. Although Alexander inspired new ways to think about design by integrating spatial forms, he neglected to consider the eco-sustainability behind spaces [18]. Based on Alexander’ work, some scholars have analyzed pattern languages for three types of spaces, namely settlements, public spaces, and residential courtyards, and considered that the following potential factors influence village spatial form: culture, living history, folk custom, etc. [18,19,20]. However, there is a lack of systematic, digital, and quantitative research methods [21,22,23]. While some research has analyzed village patterns from the perspective of geography and economics, providing a macro-level perspective with which to realize the village region, there remains a lack in terms of the illustration of the construct mechanism of village spatial pattern [24,25,26,27,28]. Therefore, our research aimed at the gap in terms of systematically determining the spatial sample and analysis mechanisms of village spatial sustainability and achieving the visualization and digitization of spatial patterns.

2.2. Visualization and Datatization of Village Settlement Spatial Pattern

Based on the analysis of the spatial pattern research method, it was found that contemporary village settlement research needs to increase the number of samples used in order to realize the corresponding digitalization and visualization. Through the literature review, we noticed that ArcGIS10.1 has the ability to manage and analyze large amounts of spatial data [29]. In recent years, ArcGIS has been widely employed in village settlement research [30]. This research has used kernel density, spatial autocorrelation (SA), and modeling approaches to analyze village settlements at the macro level, including public service facilities, community governance, rural geography, etc. Village settlement spatial patterns from the level of space require further explanation, especially with regard to how these natural factors influence the village form [31]. Meanwhile, some scholars have used GIS spatial analysis to evaluate the suitability of settlement space distribution [32]. This provides spatial planning guidance for the development and regeneration of traditional village settlements but the samples of spatial data still need to be expanded to focus on village diversity [25]. Research on village settlements in high-altitude areas has mostly focused on ecological resources [33]. Due to their complex terrain and roads, it is difficult to extract village space data in these areas. There is even less research on the spatial agglomeration characteristics and spatial form patterns of Diqing Prefecture.

Therefore, this study takes the villages in Diqing which are large in quantity and wide in area as the research object and introduces ArcGIS spatial analysis technology into the research of village settlement spatial patterns. Using more samples, it maximizes the extensive capacity of ArcGIS for fine analysis and explores the sustainability of contemporary village settlement spatial patterns on this theoretical and practical basis. It attempts to improve the quantitative level and digitization of village settlement research in the Diqing area. In this study, our research focused on three aspects:

(1)

Expanding the research sample and taking 2486 village settlements in Diqing Tibetan Autonomous Prefecture as an example, this study classifies the spatial data types and establishes a village spatial pattern analysis framework (SPAF) for such large in number, multi-ethnic, and topographically diverse areas;

(2)

Expanding the pattern language approach and basing it on the spatial relationship between the Diqing villages and topography, the vertical climate, and local construction, this study establishes scientific and quantitative analysis criteria using the village spatial data. Summarizing the village settlement patterns in Diqing by realizing the digitization and visualization of the space, this study improves the efficiency and scientificity of the traditional qualitative village spatial form analysis;

(3)

The analysis results pertaining to the village settlement patterns provide insights into rural planning in multi-ethnic areas and are applied to the sustainable development boundary of villages (including population size, morphology pattern, and environmental characteristics) to promote coordinated life, production, and ecology of villages in rural areas.

3. Materials and Methods

3.1. Research Area and Data Sources

The Diqing Prefecture has unique topographic and geographic conditions [34]. The rugged mountains and ancient plateaus wind up and down, forming the special World Natural Heritage and majestic landscape of the Diqing Prefecture with three mountains embracing two rivers, snow mountains like cities, and rivers like pools. It has a height difference of up to 5254 m within the topographic boundary line which causes a vertical climate level and three-dimensional ecological environment characteristics. This condition results in the village settlements in Diqing needing to integrate with the natural environment, reflecting the ecological sustainability [35]. Within the context of rural population decline in contemporary China, from 2014 to 2022, the population of the eight major ethnic minorities in Diqing, except for the Hui, have shown a distinctive trend of staying steady and slightly increasing by less than 10%; most of these ethnic minorities live in rural areas of Diqing (

Table 2. Data on the eight main ethnic minorities in Diqing; edited by authors, the origin data from Diqing official website, see reference [36,37], the origin map from Google Earth resource of Bigemap GIS Office.

Ethnic Group	Population Data in 2014 (Persons)	Population Data in 2022 (Persons)	Population Growth Rate	Ethnic Main Distribution Areas	Language	Religious Beliefs
Zang	130,735	134,199	2.64%	Deqin County; Shangri-La City; Tacheng Township and Badi Township, Vici County.	Tibetan Khampa dialect; Tibetan language.	Tibetan Buddhism
Lisu	109,697	112,385	2.45%	Vichy County; Xiaruo Township, Deqin County; Luoji Township, Shangri-La City.	Yi branch of the Tibetan–Myanmar family of the Sino–Tibetan language family; Lisu language.	The original religion of local people, Christianity, Catholics, and Tibetan Buddhism
Naxi	46,389	46,795	0.8%	Sanba, Jinjiang and Shangjiang in Shangri-La City; Yongchun, Tacheng, Pantiange and Yezhi in Weixi County; Fo Shan in Deqin County.	Yi branch of the Tibetan–Burmese language family of the Sino–Tibetan language family; Dongba script.	Dongbaism. Tibetan Buddhism
Yi	15,707	16,637	5.92%	Tiger Leaping Gorge, Sanba, Luoji and Jinjiang in Shangri-La City; Yongchun in Vixi County.	Yi branch of the Tibetan–Burmese language family of the Sino–Tibetan language family; Yi language.	Bhimaism
Bai	14,964	15,211	1.65%	Wideng, Zhonglu and Baiji floods in Vixi County.	Yi branch of the Tibetan–Burmese language family of the Sino–Tibetan language family; Bai language.	Bais’religious belief
Pumi	2115	2317	9.55%	Yongchun and Pantiange in Vixi County; Sanba, Luoji, Jinjiang and Shangjiang in Shangri-La City.	Qiang branch of the Tibetan–Burmese language family of the Sino–Tibetan language family.	Tibetan Buddhism, primitive religious worship
Miao	1457	1534	5.28%	Jinjiang Township, Shangri-La City.	Miao and Yao branch of the Sino–Tibetan language family, used only within the ethnic group, the rest in Chinese.	Primitive religious beliefs, such as ancestor worship
Hui	1137	1116	−1.84%	Jiantang Township and Sanba Township in Shangri-La City; Shengping Township in Deqin County; Baohe Township and Baiji Flood Township in Vixi County.	Chinese.	Islam

) [36,37]. In addition, Diqing belongs to an underdeveloped area in southwest China. As of 2022, the per capita disposable income of urban and rural residents in the Diqing Prefecture reached CNY22,214, an increase of about 5.8%. It ranks second to last in Yunnan Province’s economy [38]. As a region in which agriculture is the primary industry, its economic development is slow and there has been a slight change in the economy of its rural areas in recent years. Due to the area’s inconvenient transportation, weak agriculture industry, and slow economic situation, the traditional village spatial pattern in Diqing is both diverse and stable.

According to changeable geographical features of Diqing, the study area presents an evident cold temperate climate, which is characterized by an extreme temperature difference of about 52.5 °C, making it a rare extreme weather zone. Its special natural conditions shape three types of vertical ecological environments: alpine area (altitude of 2800–6740 m); mountain area (altitude of 2200–2800 m); and valley area (altitude of 1486–2200 m). Based on Diqing’s topography, vertical climate, and local construction, this study establishes scientific and quantitative analysis criteria using the traditional descriptive spatial pattern language. It analyzes the characteristics of the spatial patterns of village settlements in Diqing, summarizes the advantages and disadvantages of the traditional village space, and explores the sustainability of village settlement space (Figure 5).

The data for this study were obtained from three sources, including official websites and institutions, local book resources, and field research.

(1)

The village settlement name, address, and basic information data were obtained from the official website and local annual report books [4,5,36,37,39]. The satellite map and map data were obtained from the Google Earth map resource of Bigemap GIS Office which were high definition. The scale of the illustration is 1:1, meaning that the scale, outline, and location of the village can be recognized clearly;

(2)

The tangible space forms such as the village outline boundary, village roads, architectural group, etc., were obtained via on-site surveying and drawings at a scale of 1:1. Other materials that could not be obtained directly from books or official websites were obtained through field surveys, such as local legends, cultural features, etc. This information was obtained from the local village committees. The data were collected up until 2022. In this study, it was used as a social context for the research and was verified against the villager’s chronicles to prove their credibility.

3.2. Method

The methods used in this study include the following: the field survey method, ArcGIS spatial analysis, and spatial pattern analysis. This study comprises four stages (Figure 6). (1) Through the field survey method, spatial big data of village settlements in Diqing Prefecture were collected. (2) In terms of the type of spatial data, the spatial pattern analysis framework (SPAF) is refined. (3) The analysis criteria are established based on the impact elements of the spatial patterns. (4) The spatial analysis is performed using ArcGIS; the spatial agglomeration pattern and spatial layout pattern are determined using the f NNI, KDE, buffer analysis, and SOA methods and the spatial form pattern is determined using SSIA and SOA methods.

3.2.1. Field Survey and Village Settlement Spatial Big Data Collection and Collation

The field survey method is a social research method that involves participating in observation and immersing oneself in the actual space and regional context of the research object [40]. In the process of this study, through field surveys of village settlements in the study area, direct spatial data were obtained to supplement various data in the local literature. Meanwhile, the spatial data of the villages were classified; these data comprised the logical framework of the Diqing village spatial database (

Table 3. Classification of spatial data attributes of village settlements in Diqing; edited by authors.

Spatial Properties	Data Type
Name	Text
Settlement form	Int
Road accessibility	Int
Settlement density	Int
Size	Int
DEM_project1	Float
Slope_DEM	Float
Aspect_DEM	Float
Other qualitative data	Text

).

3.2.2. ArcGIS Spatial Analysis

(1)

Nearest Neighbor Index (NNI)

There are three types of spatial distribution in village settlement space: random, uniform, and agglomerative. This study uses the NNI method to determine the spatial agglomeration type of village settlements in the Diqing Prefecture. NNI uses the distribution of random patterns as a standard to measure the spatial distribution of point elements [41]. We calculate the nearest neighbor distance of each point feature in the study area and take the average value, that is, the nearest neighbor distance of point features, denoted by d. The average value of the nearest neighbor distance of the random pattern of point elements is the theoretical nearest neighbor distance, which is represented by d_min. In complete spatial randomness (CSR), the average NNI can also be obtained and its expectation is $\sum d_{min}$, represented by the following formula:

$$d_{min}=\frac{1}{n}\sum _{1=1}^n{d}_{\mathit{min}}$$

(1)

In terms of the study, the average NNI in the random mode is related to the area A of the study area and the number of events, n, considering the boundary correction of the study area. This is represented by the following formula:

$$\sum d_{min}=\frac{1}{2}\sqrt{\frac{A}{N}}+\left(0.051+\frac{0.041}{\sqrt{n}}\right)\frac{p}{n}$$

(2)

(2)

Kernel Density Estimation (KDE)

The KDE visually presents the density distribution of the village and is used to identify the scattered area and concentrated area of the settlement gathering [42]. This study uses KDE to focus on expressing the spatial agglomeration characteristics of village settlements in the Diqing Prefecture and, based on this, it extracts the spatial agglomeration pattern. The predicted density for the new (x, y) location uses the following:

$$Density=\frac{1}{{\left(radius\right)}^2}\sum _{i=1}^n\begin{array}{c}\left[\frac{3}{\pi }po{p}_i{\left(1-{\left(\frac{{dist}_i}{radius}\right)}^2\right)}^2\right] \\ \\ for {dist}_i<radius\end{array}$$

In the formula:

i = 1, …, and n is the input point. It only includes points in the sum if they are within a radius distance of the (x, y) location.

pop_i is the population field value of the I point and it is an optional parameter.

dist_i is the distance between point i and the (x, y) location.

(3)

Buffer Analysis

Buffer analysis is one of the basic functions of ArcGIS spatial analysis and is widely used in urban and village planning, such as in the extraction and establishment of mountain building parameters, urban planning and design databases, natural reserve areas, etc. [43]. In this paper, the buffer method is used to analyze the influence of key elements such as terrain, rivers, and roads on the spatial aggregation pattern of village settlements (

Table 4. The sample elements of village settlements in Diqing; edited and analyzed by authors, the origin map from Google Earth resource of Bigemap GIS Office.

Distribution of settlements

River system

Road system

).

3.2.3. Settlement Spatial Pattern Analysis

(1)

Spatial Overlay Analysis (SOA)

Spatial overlay analysis (SOA) is an operation that overlays data layers composed of related thematic layers to produce a new data layer, which integrates the attributes of the original two or more layers (

Table 5. The partial sample data layers of village settlements in Diqing; edited and analyzed by authors, the origin map from Google Earth resource of Bigemap GIS Office.

DEM data layer

Contour vector data layer

). In this study, SOA is used to analyze the relationship between the spatial distribution of village settlements and factors such as terrain (elevation, slope, and aspect) and to better present and extract the spatial pattern of village settlements.

(2)

Spatial Shape Index Analysis (SSIA)

In previous studies, the only classified village spatial forms are the cluster, belt, and finger types, while the spatial forms of Diqing’s village are diverse. Based on the spatial shape characteristics, we added the analysis criteria of the number of village settlements and the distance of settlement boundaries and classified the types of Diqing rural space, focusing more on the spatial planning characteristics. The formula for calculating the SSIA is as follows:

$$s=\frac{P}{P_0}=\frac{P}{1.5\lambda -\sqrt{\lambda }+1.5}\sqrt{\frac{\lambda }{A\pi }}$$

In the formula:

S—spatial shape index

P—the length of the settlement boundary obtained by surveying (unit: m)

P₀—the perimeter of an ellipse with the same area (unit: m)

A—the projected area of the village settlement space on the plane (unit: m²)

λ—the aspect ratio of the circumscribed rectangle of the settlement boundary

N1—the number of building groups within the village boundary (unit: pcs)

N2—the number of buildings in a unit building group (unit: pcs)

D—the average relative distance of buildings within the settlement (unit: m)

4. Results

4.1. The Spatial Agglomeration Pattern of Village Settlements

4.1.1. The Analysis Criteria of Spatial Agglomeration Pattern

The village settlement space, as the material foundation of the human living environment, needs the relevant vernacular resource conditions for spatial planning. The main factors affecting the village settlement spatial agglomeration pattern in Diqing Prefecture are natural factors [44].

(1)

Terrain

The factor affecting the agricultural production and living supplies of village settlement residents is the topography in which the settlement is located; this comprises the main impact elements for the formation of the spatial agglomeration pattern of village settlements [45]. The topographic conditions include three elements.

① Elevation

Elevation is a natural element and is also an essential factor affecting the distribution of village settlements [46]. Generally, the higher the altitude, the fewer the number of settlements. In terms of the topographic characteristics of the study area, we divided the elevation into 15 elevation levels according to the 200 m contour intervals, which include <1600 m, 1600–1800 m, 1800–2000 m, 2000–2200 m, 2200–2400 m, 2400–2600 m, 2600–2800 m, 2800–3000 m, 3000–3200 m, 3200–3400 m, 3400–3600 m, 3600–3800 m, 3800–4000 m, 4000–4200 m, and >4200 m. Using ArcGIS to perform SOA, we obtained statistical analysis data such as the number of settlements in each elevation zone. The calculated results are presented in Figure 7 and Figure 8 and

Table 6. Statistical table of the agglomeration of village settlements under different elevation levels; edited by authors, the data are the result calculated by ArcGIS through 15 elevation levels.

Elevation Zones	Number of Settlements (pcs)	Percent of Settlements (%)	Area (km²)	Settlement Density (pcs/km²)
1600–1800 m	80	3.22	108.1	0.74
1800–2000 m	282	11.34	381.1	0.74
2000–2200 m	331	13.31	591.1	0.56
2200–2400 m	356	14.32	847.6	0.42
2400–2600 m	361	14.52	1093.9	0.33
2600–2800 m	341	13.72	1311.5	0.26
2800–3000 m	189	7.60	1575	0.12
3000–3200 m	153	6.15	1700	0.09
3200–3400 m	239	9.61	2390	0.10
3400–3600 m	64	2.57	2133	0.03
3600–3800 m	45	1.81	2250	0.02
3800–4000 m	33	1.33	3300	0.01
>4000 m	12	0.48	1200	0.01
Total settlements (pcs)	2486	-	18,881.3	-

.

The results reveal the following. (a) Starting from 1486 m (the lowest point in the study area), the number of settlements gradually increases with the increase in elevation, reaching a maximum of 2400–2600 m. Afterward, as the elevation continues to rise, the number of settlements decreases. (b) At 1800–3400 m, there are a total of 2252 village settlements, accounting for 90.58% of all settlements. Among them, at 1800–2800 m, there are a total of 1671 village settlements, accounting for 67.22% of all settlements, which is the most concentrated area of village settlement distribution in the study area; at 2800–3400 m, there are a total of 581 village settlements, accounting for 23.37% of all settlements, which is the second highest concentrated distribution area of village settlements. (c) As the elevation increases, the agglomeration density of village settlements (the number of settlements per unit area) gradually decreases, that is, the higher the elevation, the more dispersed the spatial agglomeration of village settlements. (d) Above 3400 m, there are a total of 154 settlements, accounting for 6.19% of all settlements, and the settlement density is less than 0.02 per km². Therefore, areas above 3400 m are almost uninhabited.

② Slope

As with elevation, different areas with the same elevation have quite different levels of settlement agglomeration [30]. According to the characteristics of settlement agglomeration in Diqing, we divided the slope into 10 levels: 0–5°, 5–10°, 10–15°, 15–20°, 20–25°, 25–30°, 30–35°, 35–40°, 40–55°, and >45°. Using ArcGIS analysis, we obtained various statistical analysis data, such as the number of village settlements at different slope intervals. The calculated result is present in

Table 7. Statistical table of the agglomeration of settlements under different slopes; edited by authors, the data are the result calculated by ArcGIS through 10 levels of slope.

Slope Zones	Area (km²)	Number of Settlements (pcs)	Number of Settlements (%)	Settlement Density (pcs/km²)
0–5°	772.529	311	12.5%	0.40
5–10°	1389.824	353	14.2%	0.25
10–15°	2105.032	404	16.3%	0.19
15–20°	2823.770	443	17.8%	0.16
20–25°	3503.900	417	16.7%	0.12
25–30°	4041.572	338	13.6%	0.10
30–35°	4017.104	160	6.4%	0.04
35–40°	2717.320	47	1.8%	0.02
40–45°	1168.110	13	0.5%	0.01
>45°	577.602	0	0	0
Total settlements (pcs)	23,116.763	2486	-	-

and Figure 9 and Figure 10. The result reveals the following. (a) Starting from 0°, as the slope increases, the extent of settlements agglomeration gradually increases; the slope interval of 15–20° has the most settlements and then the number of settlements gradually decreases; after 35°, the number of settlements drops sharply. (b) Between 0° and 30° is the most concentrated area of village settlement agglomeration; there are still settlements distributed in areas above 30°. (c) As the slope increases, the agglomeration density of village settlements (the number of settlements per unit area) gradually decreases. On the contrary, the larger the slope, the more dispersed the spatial agglomeration of village settlements is. (d) The maximum slope of village settlement agglomeration in the Diqing Prefecture is 44.9°, which means that there is no settlement agglomeration above 45°.

In summary, the slope grading standard was as follows: 0–5° is a gentle slope, 5–15° is a moderate slope, 15–25° is a steep slope, 25–35° is a very steep slope, 35–40° is an extremely steep slope, 40–45° is an abrupt slope, and above 45° is a precipitous slope. In the study area, the terraces and dam lands on both sides of the valley have small slopes and are close to water sources and roads. Therefore, the living conditions are superior so the settlement density is large and the number is not insignificant. However, there are still numerous settlements distributed on moderate slopes, steep slopes, and very steep slope areas. The reasons for this are multifaceted. Firstly, Diqing belongs to the mountainous and gorges landform and the area with a gentle slope is limited. Secondly, the moderate slope, steep slope, and very steep slope areas in Diqing have rich land resources and are important agricultural and pastoral areas. Finally, this kind of settlement agglomeration may also have a relationship with the village settlement layout concept of ethnic minority areas.

③ Slope aspect

In different regions, the influence of the slope aspect on settlement agglomeration varies due to the comprehensive effects of natural, cultural, and other factors [47]. This study divides the slope aspect into 8 directions: north (337.5–22.5°), northeast (22.5–67.5°), east (67.5–112.5°), southeast (112.5–157.5°), south (157.5–202.5°), southwest (202.5–247.5°), west (247.5–292.5°), and northwest (292.5–337.5°). By using ArcGIS to analyze the attribute table of different slope aspects, the agglomeration of village settlements under different slope aspects can be systematically calculated, the results are presented in

Table 8. Settlement agglomeration numbers under different slope aspects; edited by authors, the data are the result calculated by ArcGIS through 8 directions.

Slope Aspects	Northwest 292.5–337.5°	North 337.5–22.5°	Northeast 22.5–67.5°	East 67.5–112.5°	Southeast 112.5–157.5°	South 157.5–202.5°	Southwest 202.5–247.5°	West 247.5–292.5°
Number of settlements (pcs)	212	199	243	379	354	340	390	368
Number of settlements (%)	8.5%	8.0%	9.8%	15.2%	14.2%	13.7%	15.7%	14.8%
Settlement density (pcs/km²)	0.081	0.072	0.084	0.119	0.120	0.121	0.135	0.123

and Figure 11 and Figure 12.

The results show the following. (a) There are 390 settlements distributed in the southwest direction (202.5–247.5°), with a settlement density of 0.135 per/km², this is the direction with the most villages and thus highest density of settlement agglomeration. (b) There are 654 settlements distributed in the north, northwest, and northeast directions, accounting for 26% of all settlements; there are 1084 settlements distributed in the south, southwest, and southeast directions, accounting for 43% of all settlements. (c) Simplifying the study area into the two directions of south and north, there are 1007 settlements distributed in the north direction (270–90°), accounting for 40% of all settlements, and 1479 settlements distributed in the south direction (90–270°), accounting for 60% of all settlements. Overall, the village settlement site selection and layout in the study area is mainly southward, followed by eastward and westward; northward is the direction with the least villages and the lowest density of settlement agglomeration.

(2)

River System

This study quantitatively analyzes the relationship between settlements and rivers by using the ArcGIS multiple-ring buffer tool. The specific method is as follows: buffer zones are established for all rivers (all rivers extracted from satellite images) and the main rivers (such as the Lancang River and Jinsha River and their first-level tributaries) in the study area at intervals of 500 m, totaling eight levels. Then, the settlement patch agglomeration layer and the river buffer zone layer are overlaid and analyzed in pairs. The calculated results are presented in Figure 13 and Figure 14 and

Table 9. Statistics of settlement agglomeration in different buffer zones of rivers; edited by authors, the data are the result calculated by ArcGIS through 8 level of buffer zones.

Statistics of All River Buffer Zones Analysis		Statistics of Main River Buffer Zones Analysis
Buffer Zones of All Rivers	Number of Settlements (pcs)	Buffer Zones of Main Rivers	Number of Settlements (pcs)
0–500 m	1146	0~500 m	635
500–1000 m	537	500~1000 m	389
1000–1500 m	295	1000~1500 m	249
1500–2000 m	167	1500~2000 m	179
2000–2500 m	103	2000~2500 m	137
2500–3000 m	63	2500~3000 m	115
3000–3500 m	61	3000~3500 m	88
>3500 m	136	>3500 m	331

.

The results show the following. (a) The number of settlements decreases sharply with an increase in the area’s distance from a river and shows an exponential decay pattern. (b) In total, 68% of the village settlements are distributed within 1000 m of rivers while only 18% of the settlements are in the buffer zone of 1000–2000 m. (c) It appears that the overall agglomeration form of settlements belongs to a typical linear-axis type; that is, village settlement areas tend to be distributed along both sides of major rivers and their tributaries, indicating that village site selection depends on water resources which satisfies the demand of traditional agricultural production and the safety of domestic water use [48].

(3)

Road System

This study uses the ArcGIS multiple-ring buffer tool to establish buffer zones at intervals of 500 m for the main roads in the study area. Then, the settlement patch agglomeration layer and the road buffer zone layer are overlaid and analyzed. The calculated results are shown in Figure 15 and

Table 10. Statistics of settlement numbers in different buffer zones of roads (500 m), the data analyzed by ArcGIS through 8 levels of road buffer zones.

buffer zones of road	0–500	500–1000	1000–1500	1500–2000	2000–2500	2500–3000	3000–3500	>3500
Number of settlements (pcs)	809	323	242	183	151	133	95	572

. The road buffer zone analysis results show the following. (a) The more settlements and the higher degree of aggregation, the closer they are to the road. (b) Contrarily, the fewer the settlements and the lower the density, the farther they are from the road. (c) The traditional village constructs relied on the accessibility of roads [49].

4.1.2. Pattern Types of Spatial Agglomeration

This study used the average nearest neighbor tool among the spatial statistics tools of ArcGIS to perform statistics according to the nearest distance index method. In order to more intuitively reflect the spatial aggregation law of the village settlements in the Diqing Prefecture, ArcGIS was used to make density maps of the village settlements in the Diqing. In ArcGIS10.1 software, there are three main ways to express distribution density, namely kernel density, point density, and line density [50]. In order to reflect the spatial distribution situation of settlements at different scale levels, after multiple experiments, the search radius was selected as 5000 m and the kernel density agglomeration map of village settlements in the Diqing Prefecture was generated, as the calculated results show in Figure 16. This indicates that the village settlements in the Diqing Prefecture show three spatial aggregation patterns.

(1)

Regional differences pattern

The settlement agglomeration has prominent regional differences. According to the kernel density agglomeration map of the settlements, the kernel density value has obvious regional differences. (a) There is a large difference in density between the north and south, with low density in the north and high density in the south. (b) There is a large difference in density between the east and west, with low density in the east and high density in the west. (c) There are obvious differences in settlement agglomeration showing an uneven distribution, as follows: Weixi County > Shangri-La City > Deqin County. This indicates that Weixi County is the region with the widest and most concentrated agglomeration of village settlements.

(2)

Two aggregation cores pattern

The settlement agglomeration in the study area has an obvious north–south zonal distribution but there are also two obvious aggregation cores. (a) The first core area is on both sides of the Lancang River in the central and western parts of Weixi County, with a core density of 1.06 per/km². (b) The second core area is at the junction of Deqin County and Weixi County on the west bank of the Jinsha River. (c) These two aggregation cores form the hot spot of the village (which means there is a high level of aggregation) settlement in Diqing. (d) In these two hot spots, an uneven distribution of the main ethnic groups is also shown, as follows: mixed by Hui, Lisu, and Bai > Lisu > Mixed by Hui, Lisu, and Bai. This suggests that mixed ethnic forms of living are concentrated in village settlements in the Diqing.

(3)

Orientation close to roads and rivers pattern

The influence of roads and rivers on settlement aggregation shows a consistency of spatial orientation, indicating the production and living needs of traditional village settlements along water and the construction mode of conforming to terrain in spatial planning. As the distance from rivers and roads increases, the spatial agglomeration of village settlements shows an exponential decay pattern. Therefore, settlements have a tendency to aggregate along rivers and roads, indicating that the abundance of water resources and the convenience of roads are important bases for spatial planning in rural construction.

4.2. The Spatial Layout Pattern of Village Settlement

4.2.1. The Analysis Criteria of Spatial Layout Pattern

Due to the influence of terrain, landform, and climate factors, the Diqing Prefecture has three regions with a vertical climate, with its temperature decreasing by around 0.37 to 0.75 °C for every 100 m increase in elevation: (a) valley region (altitude: 1486–2200 m, average annual temperature: 14~17 °C); (b) mountainous region (altitude: 2200–2800 m, average annual temperature: 8~14 °C); and (c) alpine region (altitude: 2800–6740 m. average annual temperature: −10~8 °C, with the lowest extreme temperature as low as −27.4 °C). Under these vertical climate conditions, this paper divides the spatial layout of settlements in the Diqing Prefecture into three types: valley flatland settlements (slope < 5°; valley region), gentle mountain slope settlements (slope 5–25°; mountainous region), and steep mountain slope settlements (slope > 25°; alpine region); see Figure 17.

4.2.2. Pattern Types of Spatial Layout

(1)

Valley Flatland Pattern

This type consists of small river valley alluvial plains, river valley terraces, flat small basins in the mountains, etc.; thus, they present diverse shapes and their sizes are determined by the terrain of the river valleys and flatlands, with slopes of less than 5°. The sections of the river in which it bends and swirls are those most likely to form river floodplains and eventually form large floodplain terraces (Figure 18). This settlement layout type accounts for 311 villages, accounting for 12.5% of all settlements. Among them, there are 107 settlements with a population size between 11 and 25 households, accounting for 34.4% of the valley flatland pattern, which is the highest population living form for this village settlement pattern. In addition, there are 173 settlements without a highway over more than half of number of totals, indicating that the slope of the valley flatland layout is less than 5° and that, due to the complex terrain and river system, the road is inconvenient.

(2)

Gentle Mountain Slope Pattern

There are 1617 gentle mountain slope settlements (slope between 5–25°) in the study area, accounting for 65% of all settlements, and they are the most important settlement distribution type in the Diqing Prefecture (Figure 19). Among them, the most important type of settlements’ population size is between 11 and 25 households, totaling 585 settlements and accounting for 36.1% of the gentle mountain slope pattern. This is followed by 404 settlements with a population size of less than 10 households, accounting for 25.0% of the total. There are 951 settlements that have the county roads and below; a 278 pcs settlement has the provincial roads level and above, in total accounting for 76.0%. This result indicates that the gentle mountain slope pattern is inevitably linked to accessible roads. However, due to the terrain conditions, most village settlement population sizes are not more than 25 households.

(3)

Steep Mountain Slope Pattern

Steep mountain slopes of more than 25° are one of the important village settlement distribution areas in the basin. These settlements are far away from large rivers or valleys, have high altitudes, and their layout structure is diverse (Figure 20). There are 558 settlements on steep mountain slopes, accounting for 23% of all settlements, meaning that this is one of the typical settlement distribution types in the Diqing Prefecture. Among them, the most important type of settlements’ population size is between 11 and 25 households, totaling 244 settlements, while there are 211 village settlements with a population of less than 10 households, totaling 82.3% of the steep mountain slope pattern. In addition, there are only 44 village settlements without a highway. This shows that this pattern of spatial layout has a small population size but the transportation is accessible.

In summary, in Diqing, 12.5% of the layout patterns are located in flat valley terrain, which is influenced by the water system. In addition, 87.5% of the layout patterns are influenced by mountains. To adapt to a changeable terrain, the layout pattern dominated by the water system can be divided into two subtypes: the tributary mouth layout pattern and the along-river layout pattern. The tributary mouth layout pattern and the position of the tributary, especially the confluence of the larger tributary and the river, lead to many sand and gravel deposits building up over the years, forming alluvial fans and flood fans. These places have a gentle terrain, fertile soil, and convenient transportation. Therefore, the tributary mouth becomes an important location for village settlement site selection (Figure 21). In the along-river layout pattern, due to the special geographical location and topography of the Diqing Prefecture, rivers and streams become the lifeline of village settlements. In order to close the water source and facilitate the access of water for production and living, village settlements are often laid out along both sides of rivers and streams. Under this pattern, settlements are mostly formed by extending along the natural line of the river and their main streets are parallel to the riverbank line (Figure 22).

The layout pattern dominated by mountains can be divided into two subtypes: the parallel contour line layout pattern and vertical contour line layout pattern. In the parallel contour line layout, horizontally, the settlements are arranged along the natural shape of the terrain on the same horizontal plane. The vertical extension along the contour line direction is smaller than the horizontal expansion b, see Figure 23. In addition, the vertical contour line layout pattern shows the vertical extension as the dominant feature. It shows a band-like trend of falling layer by layer along the contour line vertically. The elevation change in the main street is prominent and, in order to fully meet the use needs, several steps are set by intervals to conform to the terrain change. Overall, the vertical extension a along the contour line direction is larger than the horizontal expansion b, see Figure 24.

4.3. Spatial Form Pattern of Village Settlement

4.3.1. The Analysis Criteria of Spatial Form Pattern

We classified according to the morphological features of the closed shapes in the settlements, including the shape of village settlements under the influence of various natural geographical environments, ethnic characteristics, and human factors. According to the human influence factors and the local characteristics of Diqing [51], we added the limiting conditions of the number of settlements within the village area and the distance of the village settlement boundaries and used ArcGIS and AutoCAD to calculate the shape index of each village (

Table 11. The analysis criteria of the spatial form pattern and its classification index; edited by authors.

Spatial Planning Pattern		Classification Index
Clustered form	Normal Clustered	S < 2, λ < 2	N1 = 1, N2 > 10
	Group Clustered		N1 ≥ 2, N2 > 10
	Scattered Clustered		N1 ≥ 2, N2 < 10, D ≥ 15
Finger form		S ≥ 2, N1 = 1
Band form		S < 2, λ ≥ 2, N1 = 1

). The overall shape of the village settlements in Diqing Prefecture is divided into three types: clustered, finger, and band. With this, the cluster has three subtypes.

4.3.2. Pattern Types of Spatial Form

(1)

Clustered Form

① Normal Clustered

These spatial shape data conditions satisfy S < 2, in <2, and N1 = 1. The spatial features show that the closed shape of the village boundary presents a pattern similar to a circle, ellipse, square, or irregular polygon. The shape characteristics of the settlement show that the whole outline has no definite directional orientation and that there is no obvious development axis inside. There are 656 cluster village settlements in the Diqing Prefecture, accounting for 35.5% of the clustered form; these settlements are mainly distributed on both sides of the large river valley, major rivers, and roads, with a dense internal structure. The cluster settlements can be further divided into ordered clusters and unordered clusters. The former has a regular arrangement of buildings inside, an orderly organization of roads and lanes, and clear rules; the latter is the opposite (Figure 25). The most important type of settlement population size is between 11 and 25 households, totaling 248 settlements and accounting for 37.8%; of these, there are 196 settlements that have a public road. There are 211 village settlements with a population of less than 10 households, totaling 32.2% of the normal clustered pattern, of which 183 settlements have a public road. Furthermore, there are 510 village settlements with a slope less than 25°, accounting for 77.7% of this pattern. This result indicates that this pattern of traffic condition is convenient and has a flat terrain, while the population size of the village settlement is small.

② Group Clustered

These spatial shape data conditions satisfy S < 2, in <2, N1 ≥ 2, and N2 > 10. They are connected by roads and water systems due to terrain changes and each building group is linked by irregular paths, forming a free layout. This type of settlement tends to be large in scale to ensure a suitable farming radius (Figure 26). There are 259 groups in the village settlements in Diqing Prefecture. Among them, there are 160 village settlements with a population of over 25 households, accounting for 61.8% of the clustered group, of which 114 have a public road. Meanwhile, there are 99 village settlements with a population of less than 25 households, totaling 32.8%; of these, 82 settlements have a public road. Meanwhile, there are 209 village settlements with a slope of less than 25°, accounting for 80.7% of this pattern. This suggests that group-clustered villages are conducive to medium and large village settlements and that their site selection trends to conductive to good traffic conditions and a flat terrain.

③ Scattered Clustered

Scatter-clustered settlement patterns are located in variable terrain or narrow flat spaces, which pose a great challenge to building settlements (Figure 27). Their spatial shape data conditions satisfy S < 2, in <2, N1 ≥ 2, D ≥ 15, and N2 < 10. There are 933 scatter-clustered village settlements in Diqing. Among them, the most important type of settlements’ population size is less than 10 households, totaling 408 and accounting for 43.7%, of which 379 have a road. In addition, there are 373 village settlements with a population size between 11 and 25 households, totaling 40.0% of scatter-clustered settlements, of which 323 have a public road. Meanwhile, there are 657 village settlements with a slope of less than 25°, accounting for 70.4% of this pattern. This indicates that this pattern in the settlements’ population size is small and that the residential living form is flexible and unorganized. However, it has a convenient traffic condition and a flat terrain which can provide a habitable environment. Therefore, this village spatial pattern accounts for the largest number of villages due to its flexible adaptability in the changing environment of Diqing.

(2)

Finger form

These spatial shape data conditions satisfy S ≥ 2, N1 = 1. To adapt to the special natural conditions, they formed a special morphological feature of boundary divergence in a finger shape according to the mountains (Figure 28). There are 318 finger-form village settlements in the Diqing Prefecture, accounting for 12.8% of these appearing on the flat ground, but are caused by excessive local density differences. Among them, the most important type of settlements’ population size is between 11 and 50 households, totaling 227 settlements and accounting for 71.3%, of which there are only 75 without a public road. In addition, there are 280 village settlements with a slope of less than 25°, accounting for 88.0% of this pattern. This pattern indicates that the population size is small and that the traffic condition is convenient with a flat terrain.

(3)

Band form

These spatial shape data conditions satisfy S < 2, in ≥2, and N1 = 1, with a distinction between linear and curved types. Diqing has high mountain and canyon landforms, with dense river networks, forming band village settlements following roads and rivers (Figure 29). There are 320 band form village settlements in the Diqing Prefecture. Among them, the most important type settlement population size is between 11 and 50 households, totaling 217 settlements and accounting for 67.8%. However, there are 156 settlements without a public road, accounting for 48.8%. In addition, there are 274 village settlements with a slope of less than 25°, accounting for 85.6%. This pattern indicates that the population size is small and, even though the terrain is flat, that the traffic condition is inconvenient.

4.4. Optimization Sustainable Strategy for Village Settlements Based on Spatial Patterns

Based on the spatial information extraction of village settlements, this paper obtains the basic statistical data of three patterns in the study area using SOA and SSIA, among which the clustered pattern includes three subtypes: normal clustered, group clustered, and scatter clustered. By comparing and analyzing the settlement spatial pattern with the spatial data obtained by ArcGIS, the statistics results, shown in Figure 30 and Figure 31 and

Table 12. Settlements in the spatial pattern of village settlements under different elevation zones (pcs), edited by authors, the data were analyzed by ArcGIS.

Elevation Zones	Clustered Form			Finger Form	Band Form	Total
	Normal Clustered	Group Clustered	Scattered Clustered
<2000 m	115	36	58	73	80	362
2000–2500 m	229	101	325	117	110	872
2500–3000 m	165	82	334	66	59	706
3000–3500 m	113	40	135	62	75	425
3500–4000 m	31	0	72	0	6	109
>4000 m	3	0	9	0	0	12
Total	656	259	933	318	320	2486

, were found.

From the above charts, the following were indicated. (1) There are 933 scattered clustered pattern settlements, accounting for 38% of all settlements; this is the highest proportion among the five patterns. Meanwhile, except for villages below 2000 m in elevation, scattered clustered pattern settlements are also the largest proportion in all elevation zones. (2) The proportion of scattered clustered settlements in the same elevation zone is positively correlated with the altitude. This shows that elevation has an influence on the form pattern of village settlements and that the structure of the spatial form pattern of settlements in different elevation zones is different. (3) Among the scattered clustered settlements, 84% of them comprise less than 25 households and 44% of them have a scale of less than 10 households. This shows that most of the scattered clustered settlements in Diqing are not large in scale.

The statistics results based on the spatial analysis are presented in

Table 13. Statistical spatial planning of village settlement edited by authors; the data are the statistics result analyzed by ArcGIS through 5 types of village spatial pattern criteria.

Spatial planning	Pattern	Normal Clustered	Group Clustered	Finger form	Band form	Scattered Clustered
	Number of settlements (pcs)	656	259	318	320	933
Settlement density	Classification	High Density	Medium Density	Low Density	Scattered distribution	-
	Number of settlements (pcs)	616	715	435	720	-
Settlement scale	Classification	≤10 Household	11~25 Household	26~50 Household	>50 Household	-
	Number of settlements (pcs)	680	937	529	340	-
Traffic conditions	Classification	without highway	County roads level and below	Provincial roads level and above	-	-
	Number of settlements (pcs)	434	1447	605	-	-

. It is found that there are the following problems in the spatial pattern of village settlements in Diqing. (1) The spatial form pattern is diverse and the scattered pattern accounts for the highest proportion; this requires higher individualized planning. (2) The settlement scale is small, with settlements of less than 50 households accounting for 86%; these small and scattered living conditions increase the extent of engineering required to improve basic infrastructure, resulting in high construction costs and low efficiency. (3) The road traffic in settlements is inconvenient and only 24.3% of villages have provincial roads or above; these limit building a sound public service infrastructure and hinder the synergy of urban and village spaces.

According to the above problems, the corresponding spatial planning optimization sustainable strategies are proposed. (1) Targeting spatial pattern diversity involves optimizing village boundaries by incorporating ecological agriculture and natural landscapes into village protection. In particular, scatter-clustered villages need to be targeted to harmonize their variable terrain and narrow flat space. (2) Targeting small settlement scale involves improving residents’ usage data (including population age and social custom) via digital management, the transfer function of underutilized spaces, and reducing protection costs. (3) Targeting road traffic situation involves improving spatial accessibility based on the pattern information of hydrology, topography, and landscape; in addition, it involves planning reasonable roads according to the village pattern to decrease external transportation energy consumption and effectively optimize modernized infrastructure.

5. Discussion

Based on pattern language theory, this study established a spatial pattern analysis framework (SPAF) in order to analyze the data of village settlement spatial forms, focusing on those large in quantity and with a wide area and diversity. We conducted a quantitative analysis of village settlement spatial data by using ArcGIS and digitizing and visualizing the spatial pattern which, with large numbers and diversity, form effectively. This framework can be used to analyze the spatial patterns of village settlements in complex natural environments and can provide a reference for the sustainable protection and regeneration of the form. The following further explains the research results in combination with the historical context of village settlements, how the results are applicable to the contemporary village revitalization protection strategy, and reflects on the sustainable wisdom of settlement space:

(1)

The diversity of village settlements in Diqing is influenced by complex factors. This study proposes a quantitative spatial data analysis framework as the research focus. With the aim of examining the multiple village settlement spatial characteristics of Diqing, this study established an SPAF based on three aspects: topography, vertical climate, and local construction; these can allow objective spatial data to be visually obtained and are directly linked to the spatial shape form and character of the settlements. The research objective was to expand the practical application of Alexander’s pattern language, from constructing SPAF to centralized management and a targeted analysis unique to village diversity. Sustainability is the focus of this study. Through pattern analysis, the traditional low-tech ecological sustainability intelligence of village settlements for the characteristics of Diqing’s dynamic environment, multi-ethnicity, and low economy, the consumption of resources is recycled and reduced for village conservation and development. Additionally, the SPAF proposed in the study enabled us to find the same spatial type, adding new evaluation conditions that can be further analyzed in order to obtain the corresponding subtype; in the future culture, folklore and other factors can continue to be incorporated if the underlying integrity of the data integrity is ensured, reflecting methodological sustainability;

(2)

In the Diqing Prefecture, the natural factors that affect the aggregation pattern of village settlements include topographic conditions and hydrological conditions, which further affect agricultural production, site selection, construction material, etc. There are 2175 gentle mountain slope patterns (5–25°) and steep mountain slope pattern (>25°) settlements, accounting for 87.5% of all settlements in Diqing. This indicates that to sustain villages in variable and severe topographical environments in Diqing, for villages constructed along slopes that show sustainability and adaptivity in areas with variable topography, the utilization of flatland should be maximized, local earthen walls should be protected, and adequate lighting and ventilation for buildings should be ensured. Additionally, 68% of the village settlements were found to be distributed within 1000 m of a river, forming a settlement belt agglomeration along the river in Diqing; this is a special agglomeration pattern in Chinese traditional villages and is conducive to the sustainable self-circulation of water for agricultural production and domestic living. Overall, although the natural ecology of the area has a constraining effect on the spatial pattern of village, it provides the Diqing region with natural landscape resources that offer the region sustainable tourism and natural healing resources, including the following: three mountains embracing two rivers and snow mountains, Tiger Leaping Gorge, Ha Ba Snow Mountain, etc. The large population area mixes to form a stable, diverse, and sustainable local village spatial pattern, which is a typical example of the harmony between human beings and nature in the context of sustainable development;

(3)

The scattered clustered pattern is the most common settlement form in Diqing. Among the scatter-clustered settlements, 84% of them comprise less than 25 of households, forming the typical spatial form pattern of small scale and large dispersion in the Diqing area. Among all villages, there are 2715 settlements with a slope of more than 5°, accounting for 87.5% of in total. In addition, there are 2122 settlements with an elevation of 2000 m, accounting for 85.3% of in total. This means that more than 80% of villages’ spatial planning needs to be adapted to the changeable mountain terrain. This high altitude and steep slope cause a scattered village pattern, which is not conducive to the centralized management and planning of villages. The changing geographical conditions cause a diverse spatial pattern and provide a natural ecosystem. Balancing their natural ecology and flexible spatial patterns is a path to enhancing the sustainability of villages. For the Diqing area, in order to enhance the local living culture of individual settlements, centralized area division management can be utilized and targeted planning can be adopted [52]. Considering the integration of ethnic features, the protection and development of planning requires diversity and folk individualization, which leads to high construction costs; this can limit the regional economic development [53,54]. Large-scale centralized and targeted planning is conducive to the sustainable development of vernacular village characteristics and is able to maintain heritage unique traditional culture and reduce unnecessary consumption. In the future, it is necessary to consider dynamic changes in the sustainable development requirements of village settlements to further optimize spatial patterns.

6. Conclusions

This study analyzes the village settlement spatial patterns in the Diqing area, which reflect the sustainability of village settlements adapting to the natural environment and multiethic culture. This coincides with the sustainable intelligence of traditional cultural construction, such as the following: site selection according to the mountain conditions, road system, and river factors [55]; layout according to the local conditions [56]; diversified spatial forms according to the vernacular construct and population size [57]; etc. These patterns satisfy the spatial diversity under natural changes in the agricultural industry and present the intrinsic dynamics of spaces that require the balance of natural factors and adaptable planning [58]. Modern facilities penetrate the village area with the road system connecting villages to urban areas and there is an urgent need for a balance between modern and traditional sustainable spatial patterns for future development. These include, for example, the return demand pattern of the landscape living environment [59], the diversified development demand pattern of the village spatial form [60], the modernization demand pattern of the living comfortable pattern [61], cultural space activation, regeneration patten [62], etc. In order to ensure the vitality of a sustainable village regarding its regional culture and tangible form, it is essential to excavate the symbiotic cultural pattern [63]. Therefore, village planning and environmental design require the consideration of how the balance between tradition and modernization in the development of settlements might be achieved and how a healthy and sustainable development pattern language might be flexibly adjusted.

In terms of comparing the current situation of field investigation, this study shows that large-scale economic construction and rapid urbanization processes have brought many problems to the development of village settlements in the Diqing Prefecture, including the following. (1) Due to the focus on rapid modern development and the neglect of local culture and the village spatial pattern form, the ecological environment system boundary in Diqing has been damaged. The renewal of construction has caused diverse cultures of minority groups to fade to a single modern pattern and has seriously broken the original sustainable layout of villages adapted to nature [64]. (2) The lack of effective channels and convenient transportation for external communication has led to an underdeveloped industrial economy, a low quality of life for residents, a lack of modern public facilities and equipment, and insufficient development conditions. It has led to a lack of rational planning and an unintegrated management model for villages [65]. (3) Ignoring the regional characteristics and traditional living forms, blindly copying the similarity modern style architectural forms has resulted in the fading of the original ethnic culture of settlements and has not improved the quality of life and comfort of the local residents [66]. This situation is caused by the attempt to use a once-and-for-all pattern in order to address new changing space problems. Therefore, this spatial pattern is proposed to strengthen the link between the environment and living space and focus on the individual features and inner spirit of villages. This study used digital and visual techniques to recognize the changeable factors in village settlement space. The sustainability of village research is necessary to overlay invisible elements such as cultural heritage, the living habitat, vernacular custom, etc., in order to systematically and dynamically analyze the different characteristics of village spatial patterns.

In this study, although basic information on settlement building groups is collected, due to the severe natural conditions limiting the field research, there remains a lack of studies on information such as the vernacular architectural style, art structure, ethnological history, and local culture. In particular, Diqing is a multi-ethnic integration area and the exploration of the influence of integrated and diverse intangible cultures on village spatial patterns should be refined. This shows that the study of village settlements is a sustainable and dynamic process. Therefore, further research activities should include the following. (1) Combining vernacular characteristics data, such as ethnic customs, architectural styles, and social culture, to further explore the internal features of settlement spatial form types and diversification. (2) Combining urban morphology and conducting comparative studies to explore the spatial basis of urban design under urban–rural coordination.