Spatial Morphology Evolution of Rural Settlements in the Lower Yellow River Plain: The Case of Menggang Town in Changyuan City, China

Abstract

This study investigated the spatial pattern evolution of the rural settlement system in the town of Menggang, China, based on settlement patches extracted from remote sensing data for four time points between 1990 and 2018. Five typical villages were selected to study their spatial textures from point to surface. The reasons for the evolution of the rural settlements were examined as well. The results indicated that the number of settlement patches in Menggang decreased, the area of a single patch expanded, and the scale of the rural settlements continued to expand. Settlements were distributed randomly, but an agglomeration trend began to appear. The plain terrain had little restriction on settlement site selection. Cofferdams and embankments were the limiting factors for settlement expansion, which led to the compact development of the settlements. Economic development, population growth, and family miniaturization were positive factors for the expansion of the rural settlements. In the typical villages, cluster settlements and cluster settlements with a banded tendency had higher spatial utilization efficiency. Several high-density building centers were formed in the settlements, and the trend of residential buildings was mainly northeast–southwest. The plain terrain had little effect on the orientation of buildings, single-building areas, or the public spaces of the settlements. Traditional culture and traffic routes affected the orientation of the main buildings of folk houses. This study has theoretical and reference value for the structure optimization and sustainable development of rural settlements in the lower Yellow River plain.

1. Introduction

With regional development, the rural–urban divide is widening and rural decline is becoming a global issue [1]. China’s rural problems have several commonalities, but are also specific in terms of structural complexity and time concentration. The rural decline is a new manifestation of the “three rural issues” and a new challenge to the modernization process [1,2,3], while China’s rural revitalization strategy is a response to rural population aging, rural hollowing out, and disparities in public services between urban and rural areas [4,5]. Rural settlements are the places where people live, participate in social activities, and engage in production, forming the composite system of the human–land relationship [6]. In the process of utilizing resources and environment to develop rural society and economy, rural residents have constantly explored natural laws and formed their rural settlements, a composite system of the human–land relationship with local characteristics. Therefore, the study of typical regional rural settlements can invert its evolution process, analyze the interaction mechanism between human and nature with regional characteristics, and find its evolution law. Such research has good theoretical and practical significance for rural sustainable development and rural revitalization. In the context of climate change and food security concerns, the sustainable development of rural areas has received increasing attention, giving rise to research on sustainable livelihood, community resilience, vulnerability, eco-economy, and bio-economy [7,8,9,10]. At the same time, against the background of rural revitalization, there is consensus among researchers on the need to strengthen research on the evolution theory of rural settlement, change trends in spatial structure, the transformation of professional villages, the improvement and optimization of settlements, and the landscape of settlements [9].

Environmental protection and the high-quality development of the Yellow River basin [11,12] are important national strategies for China. Changes in the Yellow River channel have significantly affected the sustainable development of rural settlements in the Huang-Huai Plain [13,14,15]. In the more than 40 years since China’s “reform and opening up”, social and economic changes have had profound effects on the sustainability of rural settlements in the region. Questions remain, however, about the specific characteristics of rural settlements in the lower Yellow River plain under the influence of socioeconomic development. How do such factors affect the future trends and sustainability of these rural settlements? Few studies have specifically investigated these issues. To address this knowledge gap, it is necessary to select typical cases in the region for relevant research. Changyuan is the representative for studying rural settlement evolution under the influence of the Yellow River and socioeconomic development. The town of Menggang is located in the eastern part of Changyuan, close to the Yellow River and about 9 km from Changyuan’s central urban area. Menggang can be regarded as the typical case for studying rural settlements at the township scale. Such work has theoretical and practical significance for the sustainable development and revitalization of rural settlements in this area. More broadly, it has reference value for the study of village forms and sustainable development in similar environments in China and internationally.

The study of rural settlements in terms of geography emphasizes the temporal and spatial characteristics of settlement distribution. The study of rural settlements from an architectural perspective focuses on two-dimensional plane shapes and emphasizes the characteristics of spatial texture. Neither perspective can deeply capture the evolution of rural settlements on its own. Thus, the present study’s main contributions are as follows. First, adopting an interdisciplinary approach, geography and architecture were integrated to conduct the investigation. Second, based on rural settlement patches, the spatial texture characteristics of rural settlements in the town area were studied in terms of points and planes.

This study extracted settlement patches from 1990 to 2018 in the case region to study the evolution of spatial patterns in the rural settlements, which were divided into five types. Typical representative villages were selected for each type, and their spatial texture characteristics were analyzed, which could reflect the common characteristics of all villages in the region. This approach can present a more comprehensive and systematic characterization of the whole rural settlement system.

Next, Section 2 reviews studies that analyze the spatial characteristics of rural settlements and their mechanisms. Section 3 describes the study area and research methods, while Section 4 presents the results and analysis. Section 5 is the discussion, and Section 6 concludes.

2. Literature Review

The evolution of rural settlement patterns reflects the adaptability of humans to various economic and geographical conditions, the natural environment, and historical developments. Accordingly, it is an important topic in rural geography research [16,17]. The study of traditional settlement morphology in China began within the past 20 years [18]. The study of rural settlement morphology has mainly focused on chronological evolution, morphological evolution patterns, and the factors affecting the evolution of rural settlements [18,19,20,21,22,23,24]. The evolution of the spatial structure of rural settlements has been divided into four models: discrete isomorphism, centripetal stratification, branch duality, and multiple network systems [25]. The main factors affecting the evolution of rural settlements include geographic location, natural environment constraints, infrastructure, regional cultural inheritance and integration, urbanization, rural industrial transformation, land-use reform and innovation, farmers’ behavioral transformations, and macro-control policies [26,27,28,29,30]. The effect of roads on rural settlements has gradually weakened, the influence of towns has gradually increased, and the effects of water systems and terrains have fluctuated [31]. More recently, rural settlements have been studied at a more micro-level scale. With regard to architecture, studies have focused on sustainable design [32], the architectural characteristics of specific regions [33], and the spatial characteristics and architectural culture of traditional dwellings. Regarding the spatial morphology of traditional villages, quantitative research on the two-dimensional plane morphology of settlements has utilized landscape ecology, fractal geometry, and mathematical statistics [34]. The relationship and order between settlements have been studied in terms of residential orientation, residential area, and residential spacing [35]. The spatial morphological characteristics of ancient towns or traditional villages have been identified based on spatial grammar and other methods [36,37]. Furthermore, the historical spatial processes of villages have been analyzed and compared using the multitemporal analyses of spatial pattern morphological indicators and regression analyses of morphological evolution [16].

Most research on rural settlement morphology has been conducted from the perspectives of geography and architecture. Geography-based research has mostly focused on the types [38] and reconstruction [39] of rural settlements, settlement hollowing [40], morphological evolution [16,41,42], and spatial features and influencing factors [21,43]. In terms of architecture, studies have focused on architectural features [33], village spatial forms [37], and village texture [36]. Although there are differences according to discipline, settlement morphology systems remain an important direction for rural settlement research. In addition, trends in related research have been influenced by factors on the regional and spatial scales. Most of the relevant research in China has focused on Jiangsu [44], southern Fujian [45], and the northwest regions [24] at the province, county, and city scales. There is a lack of research on the Huang-Huai plain, and few studies have focused on the spatial morphology of typical villages at the microscale. Moreover, studies have tended to focus on either the spatial pattern of rural settlements or the textural characteristics of villages. Comparatively few studies have investigated the interaction between the spatial patterns of rural settlements and the textural characteristics of village morphology. The village form has good continuity and inheritance and remains relatively stable over time. Furthermore, the “trace” under the influence of internal and external factors is more obvious. Therefore, studying the settlement form at the township scale can place individual villages within the rural settlement system and explore the evolution of the spatial form of rural settlements. In addition, typical samples selected from the settlement system can be used to study the spatial form of individual villages. This approach can accurately and comprehensively capture the detailed characteristics of settlement evolution and provide a reference for the sustainable development of settlements based on analyses of the influencing factors.

3. Materials and Methods

3.1. Study Area: Menggang

Menggang is located in the middle-eastern part of Changyuan (Figure 1). Changyuan is located in northeastern Henan Province, in the alluvial plain of the lower Yellow River, which belongs to the warm temperate continental monsoon climate. The county is between 114°29′–114°59′ E and 34°59′–35°23′ N, with an area of 1051 km². The county governs eleven towns, two townships, and five offices, with a total population of 878,300 at the end of 2018. Changyuan has a good industrial foundation, and its economic development is mainly based on the private industrial economy, which started in the 1970s. It now has seven characteristic industries: lifting machinery, medical equipment, construction, culinary, green food, and seedlings flower industries [46].

As a case area, Menggang has two typical structural characteristics. First, Menggang has the typical characteristics of rural settlements in the lower Yellow River. There is a section of the lower Yellow River called the “Tofu Waist”, which extends nearly 200 km and is the place where the Yellow River is most likely to burst and flood. The reason for floods in the “Tofu Waist” is that the soil quality of the levee is poor (it is “soft”, as with tofu), and it cannot withstand wind and waves. Therefore, even small and medium-sized floods may cause major disasters. The town of Menggang in Changyuan is located in the “Tofu Waist” of the Yellow River, and its settlement development is greatly affected by the Yellow River. Because of its proximity to the Yellow River, Menggang has faced disasters while also enjoying the convenience afforded by the Yellow River. The development and form of settlements are deeply influenced by the Yellow River. The Guanmeng Dike, built in the 1990s to prevent Yellow-River-related disasters, was named based on the fact that it runs through the whole of Menggang. The Yellow River thus has a close relationship with Menggang and affects its formation, development, and morphological evolution. Morphological changes in the Yellow River basin are closely related to settlement changes. Most of the lower Yellow River is plain agricultural area, and the settlement forms have high similarity. Therefore, Menggang can be considered representative of the evolution of rural settlements in the lower Yellow River.

Second, Menggang is a typical representative of the evolution of rural settlements under the influence of the private sector. Since China’s reform and opening up, industry in Changyuan has significantly developed. Changyuan gradually evolved from an agricultural city into an industrial city, forming the “Changyuan mode” of economic development. Settlements are therefore deeply affected by industrial development in the area. Menggang is part of Changyuan and is about 9 km from the central urban area. In recent years, urbanization has had an increasingly strong influence, greatly affecting the settlement form and style of Menggang. Therefore, Menggang can be considered representative of the evolution of rural settlements under the influence of the private sector.

3.2. Data Sources and Use

The data used in this study came from Landsat remote sensing images, satellite images, aerial photos, and the socioeconomic data of settlements.

Table 1. Data sources and uses.

Data Types	Year	Data Sources	Data Usage
Landsat remote sensing image	1990, 2000, 2008, 2018	Geospatial data cloud http://www.gscloud.cn/ (accessed on 10 November 2019)	Extracting patch information of rural settlements
Satellite image	1990, 2000, 2008, 2018	Google Maps http://www.gditu.net/ (accessed on 10 November 2019)	Check patch information of rural settlements
Aerial Photos	2018	Planning and Design Institute of Henan University	Draw the building and road axis map of the village
Land-use map	2016	Planning and Design Institute of Henan University	Extract the administrative boundary information of villages and township in 2016
Statistical data of social economy	1990, 2000, 2008, 2018	Henan Provincial Bureau of Statistics http://www.ha.stats.gov.cn/ (accessed on 10 December 2019)	Calculate relevant social and economic indicators

shows the specific data sources and how they were used.

3.3. Methods

3.3.1. Coefficient Variation of the Voronoi Diagram

A Voronoi diagram of the settlements in Menggang was created using ArcGIS. The coefficient of variation (CV) was calculated according to the area of the Voronoi, which can measure the relative change in the settlement in space [47]. The formula is as follows:

$$\mathrm{CV}=\frac{{\mathsf{\sigma }}^{\mathrm{P}}}{\sum {\mathrm{T}}_{\mathrm{i}}{\mathrm{M}}_{\mathrm{i}}/\sum {\mathrm{M}}_{\mathrm{i}}},$$

(1)

where $\mathrm{CV}$ represents the coefficient of variation for the scenario, ${\mathsf{\sigma }}^{\mathrm{P}}$ is the standard deviation of accessibility values ${\mathrm{T}}_{\mathrm{i}}$, and ${\mathrm{M}}_{\mathrm{i}}$ is the weight of the settlement patch. Following Duyckaerts (2000), the point set is randomly distributed when CV ∈ (33–64); it is clustered when CV ≥ 64; it is evenly distributed when CV ≤ 33.

3.3.2. Compactness Index

Cole proposed the compactness index [48], which is the ratio of a region’s area to its minimum outer circle area. A region’s area A and minimum outer circle area D is calculated as follows:

S = A/D.

(2)

The compactness index reflects the complexity of the region’s shape, which considers the circle to be the most compact shape. The closer compactness S is to 1, the more compact the area; the smaller the compactness S, the more complex the area pattern.

3.3.3. Space Syntax

Space syntax, proposed by Hillier, quantitatively describes the spatial attributes of traffic routes in free space through morphological analysis variables. Connection value, control degree, depth value, and integration degree are the main morphological analysis variables in traditional spatial syntax models [49]. The road network of rural settlements is a linear layout on a two-dimensional plane. The axis map of typical village roads was drawn in AutoCAD and imported into Depthmap to calculate the integration value, which was used to describe the aggregation or dispersion degree of typical villages.

3.3.4. Settlement Outer Boundary Shape Analysis Index

The shape index [34] formula with the same aspect ratio ellipse as the frame of reference is

$$\mathrm{S}=\frac{\mathrm{P}}{\left(1.5\mathsf{\lambda }+1.5-\sqrt{\mathsf{\lambda }}\right)}\sqrt{\frac{\mathsf{\lambda }}{\mathrm{A}\mathsf{\pi }}},$$

(3)

where P is the perimeter, A is the area, and λ is the aspect ratio. Based on this, we calculated the shape index Sbig, Smid, and Ssmall of the closed graph of the settlement boundary at three virtual-edge scales of 100 m, 30 m, and 7 m, respectively, and obtained Save as follows:

$${\mathrm{S}}_{\mathrm{ave}}={\mathrm{S}}_{\mathrm{big}}\times 1.4010\times 0.25+{\mathrm{S}}_{\mathrm{mid}}\times 0.5+{\mathrm{S}}_{\mathrm{small}}\times 0.5611\times 0.25.$$

(4)

The flat pattern of the settlement was classified using the graph’s aspect ratio λ and shape index Save. Firstly, the S index is used to judge whether the village shape is a finger-like settlement. Secondly, the λ index is used to judge whether the village is in a cluster settlement or band shape settlement or cluster settlement with a band tendency. Specific classification standards are shown in

Table 2. Rural settlement flat-pattern classification.

S Value	λ Value	Settlements Category
S ≥ 2	λ < 1.5	Finger-like settlements with cluster tendency
	1.5 ≤ λ < 2	Finger-like settlements without obvious tendency
	λ ≥ 2	Finger-like settlements with band tendency
S < 2	λ < 1.5	Cluster settlements
	1.5 ≤ λ < 2	Cluster settlements with band tendency
	λ > 2	Band settlements

[34].

4. Results

4.1. Evolution Process of the Formation of the Township Settlement System

4.1.1. Small Restriction of the Plain Terrain and Random Distribution of Rural Settlements

The settlement distribution of Menggang was found to be random, and the production and living behaviors of people in the plain areas were less restricted by topography. The centrifugation point of the settlement patch in Menggang for 4 years was extracted, and ArcGIS was used to draw the Voronoi diagram. The standard deviation and average value of the actual area of Menggang were calculated. This enabled finding the CV value for the 4 years. As shown in

Table 3. Calculation results of the Voronoi diagram, 1990–2018.

Year	Area Standard Deviation	Average Area (m2)	Coefficient Variation CV	Distribution Type of Rural Settlement
1990	491,393.35	967,228.62	50.80	random distribution
2000	553,651.97	1,220,550.31	45.36	random distribution
2008	512,807.76	1,314,439.44	39.01	random distribution
2018	574,694.69	1,385,489.76	41.48	random distribution

, the area standard deviation of the Voronoi diagram in 1990 was 491,393.35, the average area was 967,228.62 m², and the CV value was 50.80; the values in 2000 were 553,651.97, 1,220,550.31 m², and 45.36, respectively; those in 2008 were 512,807.76, 1,314,439.44 m², and 39.01, respectively. In 2018, they were 574,694.69, 1,385,489.76 m², and 41.48, respectively. In summary, the 28-year period saw large changes in the settlement patch size, and the CV values fluctuated, though they remained between 33 and 64. This indicates that the overall distribution type of rural residential areas in Menggang was random. Menggang is part of the Yellow River alluvial plain, and the plain terrain has little restriction on the site selection of its settlements. The random distribution of settlements in Menggang thus conforms to the settlement distribution pattern in the plain area.

4.1.2. Expansion of Patch Area of Individual Settlements, Decrease in Number, and Expansion of Rural Settlement Scale

The number, total area, and density of patches were selected to reflect the characteristics of settlement expansion. The number of settlement patches reflects the consolidation and expansion of settlements; the total area of patches reflects the overall expansion of township settlements, and patch density represents the sparseness of settlements.

Table 4. Patch information for Menggang, 1990–2018.

Year	Number of Patches	Total Area of Patches (hm2)	Patch Density	Area Growth Rate
1990	53	402.86	0.131	-
2000	42	529.49	0.079	31.43%
2008	39	644.19	0.061	21.66%
2018	36	869.93	0.041	35.04%

shows that the total number of patches decreased from 53 in 1990 to 42 in 2000. The total patch area increased from 402.86 hm² in 1990 to 529.49 hm² in 2000, an increase of 31.43%. Patch density was reduced from 0.131 to 0.079 in 2000, with an expansion rate of 31.43%. The total number of patches decreased from 42 in 2000 to 39 in 2008, the patch area increased from 529.49 hm² in 2000 to 644.19 hm² in 2008, the patch density decreased from 0.079 to 0.061 (expansion rate: 21.66%), and the expansion speed of rural settlement was reduced. The number of patches declined from 39 in 2008 to 36 in 2018, and the patch area increased from 644.19 hm² in 2008 to 869.93 hm² in 2018, an increase of 35.04%. Overall, the number and density of patches in the settlements of Menggang declined over the 28-year period, while the total area of patches continued to increase. The area expansion rate appeared to be faster first, then slower, and then faster.

There are two main reasons for the expansion of the rural settlement scale in Menggang. First, economic development is an important factor affecting the expansion of rural settlements. Economic development provides external economic support for rural settlement expansion, making expansion possible. Further, increases in income give residents an internal impulse toward settlement expansion. According to the Changyuan Yearbook, the per capita gross domestic product of Menggang increased rapidly from CNY 7127 in 2001 to CNY 24,081 in 2018, indicating the rapid development of rural economy. With the development of the rural economy and the improvement of farmers’ income level, there had been a “house building fever” in rural settlement. The expansion and new construction of housing have led to an increase in rural settlement scale. At the same time, the phenomenon of rural “hollowing out” caused by abandoned houses and idle land in the rural settlement indicates that a large number of abandoned homesteads have not been optimized for use [50]. Of course, the scale expansion in rural settlement may also be due to social welfare issues, traffic congestion, housing shortages, environmental pollution, and other issues arising from urban development, making many rural residents more willing to use the money earned in the city to build housing in their rural hometown [51]. Second, population growth and family miniaturization are the endogenous demands of settlement land expansion. With rapid increases in population, the trend of family miniaturization has become increasingly obvious. The superposition of the two leads to rapid growth in the number of households. When the demand for construction land brought about by population increases cannot be supported simply by improving spatial utilization, this will inevitably lead to an increase in the total area of residential land. According to the data from the Changyuan Yearbook and Xinxiang Yearbook, the population of Menggang increased from 47,000 in 2001 to 49,000 in 2018, and the number of households increased from 11,158 in 2000 to 13,957 in 2018. During the same period, the total patch area expanded by 340.44 hm². The expansion of settlement patches was consistent with the population growth. Moreover, following rural village tradition, when building new homesteads, the old homesteads are often retained for various reasons, and the homesteads of people who move to other places are also retained for a long time. In short, growth in population and households is another important factor in the expansion of rural settlements.

4.1.3. Gradual Connection of Rural Settlements into Patches and Decreased Patch Compactness

With the expansion of settlement patches, residential areas gradually join together into pieces. The settlement center of Menggang gradually migrated from the northwest corner to the north of Wenyanqu, and the density value of the middle area also gradually increased. The nuclear density value calculated by ArcGIS can intuitively characterize the agglomeration or dispersion of rural settlement patches. Comparing the nuclear density distribution maps under bandwidths of 2000 m, 1000 m, and 500 m revealed that the bandwidth of 1000 m was the most intuitive and effective. Thus, a bandwidth of 1000 m was selected to protract the density distribution of rural settlements in Menggang for 4 years (Figure 2). In 1990, two high-density centers appeared in the northwest of the town. In 2000, the high-density area was still in the northwest corner of the town, while the secondary high-density area moved to the northern zone of Wenyanqu. In 2008, the two high-density areas in the northwest corner of the town and the northern part of Wenyanqu split into two high-value areas, while another high-value area began to form in the middle zone. In 2018, the northwest high-value area shifted westward, while the degree of agglomeration weakened, but the agglomeration centers in the north of Wenyanqu and the middle zone were strengthened.

Overall patch compactness in Menggang showed a downward trend. Using the natural point fracture method in ArcGIS, based on the compactness index, the settlement patches in Menggang were divided into different grades, representing the morphological complexity of the settlement patches. The proportion of each grade patch in different years was calculated (Figure 3 and Figure 4). As shown in Figure 4, the overall rural settlement compactness index of Menggang decreased from 0.55 in 1990 to 0.50 in 2018, showing little change. The number of second-grade patches always occupied a dominant position. Except for a large fluctuation in 2000, the difference between the first-grade patch and the second-grade patch was not obvious. The third-grade patch was in a weak position for a long time. Figure 3 shows that the first-grade patches gradually transited from being relatively random in 1990 to being relatively concentrated in the western region in 2018. The second-grade patches were relatively balanced in the east, north, and south of the town. The third-grade patches were distributed between the first- and second-grade patches.

Cofferdams and embankments are important factors related to the compact development of rural settlements. The Yellow River changes course to the north in Changyuan. In the past, the Yellow River had burst nine times in 10 years, and Changyuan was affected by this three times. Many villages along the Yellow River have built cofferdams to prevent floods caused by river channel changes. Village cofferdams often become the enclosure boundary of villages and limit their expansion. The Guanmeng Dike, a part of the Yellow dike, and the natural Wenyan Canal, in the east of Menggang, protect the town from Yellow River flooding and provide production and domestic water for residents along the coast. They are also a factor limiting village development and comprise the boundaries of towns. Cofferdams and embankments guide rural settlements to expand around the original patches, leading to the consolidation of adjacent patches, which is also conducive to the intensive development of villages.

4.2. Typical Morphological Characteristics of Villages

4.2.1. Classification of Rural Settlement Morphology and Selection of Typical Villages

Compared with mountainous areas, settlements are relatively compact in the plains. When calculating the morphological index, the minimum width of 50 m was selected to generate an external rectangular boundary, which can not only reflect the local outline of the village in detail, but also not trivial, and can best reflect the overall shape of the village. Therefore, the morphological index S and the boundary length–width ratio λ were calculated with the minimum bandwidth of 50 m for the 35 villages in Menggang (

Table 5. Village morphology classification table based on λ and the shape index S value.

S Value	λ Value	Settlements Category	Village Name
S ≥ 2	λ < 1.5	Finger-like settlements with a cluster tendency	Shitouzhuang
	1.5 ≤ λ < 2	Finger-like settlements without an obvious tendency	Jiuji, Beichen
	λ ≥ 2	Finger-like settlements with a band tendency
S < 2	λ < 1.5	Cluster settlements	Qiucun, Shangxiaozhai, Wangshitou, Buzhai, Kongcun, Erlang, Zhangxiaozhai, Yezhai
	1.5 ≤ λ < 2	Cluster settlements with a band tendency	Fulou, Zhifang, Xichen, Dawangzhuang, Zhangzhuang, Niannan, Boyu, ZhaOzhuang, Sunzhai, Xianglizhang, Lihuzhai
	λ > 2	Band settlements	Tianshitou, Yangzhai, Fengwan, Weiyuan, Gaolou, Shiwulihe

). According to the judgment standard in Section 3.3.4, the 35 villages in Menggang were divided into five categories: finger-like settlements with a cluster tendency, finger-like settlements without an obvious tendency, cluster settlements, cluster settlements with a banded tendency, and banded settlements. The first two of five categories are finger-like settlements, whose boundary morphology is dominated by multiple directional extension. However, the boundary shape of finger-like settlements with a cluster tendency is relatively similar to that of circle, square, or irregular polygon, while the boundary shape of finger-like settlements without an obvious tendency has no obvious tendency. Cluster settlements and cluster settlements with a banded tendency are all group-packed settlements; that is, the boundary shape is relatively similar to round, square, or irregular polygon. However, the boundary shape of cluster settlements with a banded tendency is dominated by only one directional extension; that is, it presents a banded tendency. Banded settlements are typical zonal settlements; that is, their boundary shapes are dominated by only one directional extension.

One representative village was selected from each category to study its village form. After considering the history of the villages and the integrity of architectural preservation, the five representative villages selected were Shitouzhuang (finger-like settlements with a cluster tendency), Beichen (finger-like settlements without an obvious tendency), Dawangzhuang (cluster settlements with a banded tendency—from the Yongle period of the Ming Dynasty), Gaolou (banded settlements; Ming Dynasty), and Kongcun (cluster settlements). Because the Menggang government is located in Menggang village, it was not selected so as to reduce the effect of institutional differences between villages.

4.2.2. Analysis of Village Public Activity Spaces

The road followed the principle of “longest and least” by mapping the axis of a typical village in AutoCAD, importing the axis map into Depthmap, and calculating the integration of the village space (Figure 5). Warm colors indicate higher integration, and cooler colors (blue) indicate lower integration. As shown in Figure 5a, the two roads in Shitouzhuang were highly integrated, red, east–west, and north–south. North–south road utilization was more reasonable, with shops, green parks, and some public service facilities on both sides of the road. As shown in Figure 5b, there were three vertical main roads in Beichen and two horizontal main roads. The southernmost horizontal main road integration was the highest, the road axis color in the southwest area was cool overall, and integration was low. As shown in Figure 5c, the integration degree of Dawangzhuang was slightly higher than that of Kongcun, with more roads with yellow and red colors, fewer cold-toned axes, and two of the most integrated axes in the easternmost and northernmost parts of the village; these play an important role in the village road system. As shown in Figure 5d, the overall road integration of Gaolou was low, and the central part included a main road through the village. There were many entrances and buildings in the southern part of the main road, but the low degree of integration indicated that at the southern part of the main road, there was a need to build a new main road to improve the accessibility of the southern part of the village. As shown in Figure 5e, there were five roads with high integration in Kongcun, including two north–south roads and three east–west roads. They are the main roads of the village and play an important connecting role in its road system. Among them, the east–west main road in the middle was the most important, with the village’s elderly care center, shops, sanitary rooms, public toilets, parking lots, and gardens distributed on both sides of the road.

Traffic lines are the skeleton of the development of rural settlements. The roads connecting counties and townships are generally important roads in the town and have the highest utilization rate. Rural settlements tend to develop along the strip around the two sides of these roads, and public service facilities are also arranged along the traffic lines. Because the village is located in the plain area, the residential area of single farmers is large, and the public space area of the village is generally large.

4.2.3. Building Density and Direction

(1)

Different types of villages have different building densities

Figure 6 shows a map of the core density of five typical villages at a bandwidth of 30 m. To calculate the building nuclear density of a typical village, the natural point fracture method was divided into five levels in ArcGIS; the dividing thresholds were 0.0015, 0.0035, 0.0055, and 0.0085. The darker the color of the illustration, the denser the building distribution.

As shown in Figure 6a, the building density of Shitouzhuang village was higher in the east than in the west, the distribution of buildings was more scattered, the area of internal public space was larger, and the phenomenon of building clusters was obvious. The main factors restricting the development of the village to the southwest were found to be administrative boundaries, with the road in the south of Shitouzhuang also restricting its expansion. As shown in Figure 6b, the architectural nuclear density distribution map of Beichen was divided into three groups: the center of the village had the highest density of buildings, followed by Xiaozhai in the southwest (belonging to the natural village of Beichen), and the lowest was Beilizhai village in the northeast (belonging to the natural village of Beichen). There was an obvious zonal region with a high core density (b①) in the western part of Beichen and the eastern part of Xiaozhai, mainly relying on the village’s main road, convenient transportation, and densely distributed buildings. Figure 6c shows that the core density of the buildings in Dawangzhuang was on the high side as a whole, indicating that its buildings were clearly gathered together. The high-value area formed a spatial Y pattern. Observing the aerial images, building mass was small in places with higher building density, and the locations were mainly on both sides of the road. The lower-density area included old village houses with a small building volume, large courtyards, and dense vegetation. As shown in Figure 6d, the distribution of buildings in Gaolou formed two high-value conjoined areas, showing a pattern of higher in the north than the south and expansion along the road. The area along the main road was narrow north–south and had a longer east–west pattern. Aerial images showed that the northwest part of the village was a residential gathering area, and the east was close to the township, mostly commercial land with large buildings. Figure 6e shows that most area-density values in Kongcun were greater than 0.0035, with a high-value area surrounded by a low-value area. Two large-scale, high-density areas were formed in the northwest and southwest (Figure 6e ①②), characterized by a small building mass, small public land area, diverse housing types, and different shapes. The low-density area in the middle of the village was mainly public land, with dense trees, no building distribution, or less building areas in the courtyard.

In summary, the banded settlement Gaolou, the cluster settlement Kongcun, and the cluster settlement with banded tendency Dawangzhuang had more high-density areas, and areas with a density value of 0.0035 or more covered a larger area and formed multiple high-density centers. The building densities of Shitouzhuang, a finger-like settlement with a cluster tendency, and Beichen, a finger-like settlement with no obvious tendency, were relatively low, and building distribution was more uniform.

(2)

The orientation of typical villages buildings is similar

Most of China’s northern buildings are oriented from north to south. The orientations of building have a certain tilt as a result of local geographical environments and customs. Figure 7 shows that the overall construction direction of the five typical villages was north–east, and there were slight differences between villages. The tilt angle of the three natural village buildings in Beichen was different. The northern structure tilt of Xiaozhai in the southwest was more obvious, with the north–south axis being at a 15° angle, and the southern tilt angle was reduced. Beilizhai in the northeast had a relatively small tilt angle of 5–10° with a north–south axis. The western buildings in Beichen were north–south, and the east was similar to Beilizhai. Most buildings in Gaolou were tilted 5–10°; there was a greater tilt angle in the westerly buildings, with the maximum reaching 30–40°. Kongcun basically had a north–south architectural layout, with no obvious architectural tilt. Shitouzhuang can be divided into two parts from the middle. The western buildings as a whole tilted to the east by 5–10°, and the eastern buildings presented an S-shaped tilt feature. The overall building direction of Dawangzhuang was slightly inclined to the east, with a greater tilt angle more eastward.

The orientation of buildings is influenced by many factors, such as cultural traditions, traffic, and natural factors. In China’s traditional architectural culture, considering the need to coordinate life and climate, folk houses in northern China are often built in a southerly direction. However, building orientations may have a certain inclination influenced by local geographical environments and customs. The plain terrain was found to have little effect on the orientation of the main houses in Menggang. Village trunk roads will have an effect on building orientation. For travel convenience, building orientation generally forms a vertical relationship with the road. Of course, when building construction occurs before road construction, building orientation might also be inclined toward the road. For example, the building orientation in Gaolou was obviously inclined, but it was basically the same as that of the main road in the village.

5. Discussion

This paper expands the dimensions of the research on the spatial morphology of rural settlements. This study investigated the spatial morphological evolution characteristics of rural settlements in three dimensions: the macroscopic scale of geography, the microscopic scale of morphology, and the microscopic scale of architecture, and to a certain extent, conducted a beneficial exploration on the evolution of rural settlements on a small spatial scale. Settlement form is the result of the combined effect of the regional geographical environment and the economic environment [52,53]. This paper studies the spatial distribution characteristics of rural settlements in the lower reaches of the Yellow River plain, and the result shows that the rural settlements in the plain have the characteristics of agglomeration, which is similar to the research results of rural settlements in other plain areas [29,44,53]. At the same time, this study expands the research on the spatial texture characteristics of typical villages. This paper studies the spatial morphological characteristics of typical rural settlements of various morphological types in plain areas. It is obvious that natural and human factors have different effects on the spatial morphology of rural settlements of different geomorphic types region [34,37]. Different from other studies on spatial morphology of settlements [32,54], this paper is based on geographical spatial distribution so as to better integrate geographical factors into the study of spatial morphology of settlements. Studying the spatial evolution characteristics of rural settlements in small-town areas has theoretical and practical significance for protecting the traditional style of the Yellow River basin, implementing the rural revitalization strategy, and sustainably developing small towns [55,56]. Of course, there are some limitations owing to the influence of various factors, which can be explored in future research, as described below.

(1)

Data are a constraining factor for small-scale spatial research. The spatial form of rural settlements is the result of long-term, multifactor effects [52]. Research on small-scale rural settlements in typical regions can clearly and accurately describe their evolution processes and analyze their mechanisms. What role does the Yellow River play in the sustainable development of rural settlements? Undoubtedly, it has both restrictive effect and positive contribution [15,57,58]. The role of the Yellow River in the sustainable development of rural settlements may vary with different time scales. Obviously, the study of long time series may explain this problem more effectively. However, it is difficult to obtain remote sensing data or socioeconomic data of long time series. Therefore, relevant research needs to be conducted with the help of alternative means.

(2)

Quantitative research is a common method for mechanism analysis. However, the limitation of small-scale spatial data acquisition restricts the verification of quantitative research reliability. Qualitative research has become an alternative method for this type of research. It is also worth considering how to obtain relevant research materials corresponding to the time nodes of rural settlement evolution to ensure the reliability of qualitative research.

6. Conclusions

From the perspectives of geography and architecture, this study investigated the spatial morphological evolution characteristics of the rural settlement system in the town of Menggang based on settlement patches extracted from remote sensing data covering 1990–2018. The spatial texture characteristics of five representative typical villages were analyzed to understand the common characteristics of the village spatial form in the case area from point to area. The spatial evolution mechanism of rural settlements was studied as well. The main conclusions are as follows.

(1)

Cofferdams and embankments have led to a clustering trend in the distribution of rural settlements in the lower Yellow River plain, and the development of economy and society has promoted the continuous spatial-scale expansion of these rural settlements. The distribution pattern of rural settlements in Menggang Town has changed from random to agglomeration. Moreover, the overall scale of settlements continued to expand, but the number of settlement patches decreased, and its area expanded. The plain terrain is the natural supporting condition for the random distribution of settlements, while cofferdams and embankments along the Yellow River became important factors inducing the agglomeration distribution of rural settlements. Economic development was an external factor affecting rural settlement expansion, and population growth and family miniaturization represented the endogenous demand for the expansion of settlement land.

(2)

Improving the efficiency of street and alley space utilization is an important feature of the development of plain villages, and roads and traditional culture are important influencing factors for village expansion and architectural orientation. In typical villages, the clustered settlements and clustered settlements with a banded tendency were the street and lane space models with high-utilization efficiency. These settlements often formed one or more high-density centers of buildings, and the building inclination direction was mainly northeast–southwest. Traffic lines represent the skeleton of the development of rural settlements. Traffic roads are the guiding factors affecting the layout and internal expansion of rural settlements. Plain terrain has less restrictions on the space and orientation of single buildings and the public space of the settlements, while regional culture and traffic roads will affect the orientation of architectural development.