1 Introduction
Traditional villages refer to those villages built in a long time, which have undergone a long historical evolution and are still mainly inhabited and engaged in agricultural production by farmers, and retain the traditional living forms and cultural forms. Traditional villages have long history, rich traditional resources, and important historical, cultural, scientific, artistic, social and economic values. Compared with other villages, the most distinctive features of traditional villages are the existing traditional architectural features are intact, the village site selection and pattern maintain traditional characteristics, and the intangible cultural heritage stays alive and well (
Hu et al., 2014). Therefore, traditional villages in China provide important historical evidence for early humans who adapted and learned to live in harmony with nature in a period of time, representing the precious heritage left by Chinese ancestors to the future generations (
Tong, 2014). The modern world has seen rapid economic development, industrialization, and urbanization, which has led to the gradual disappearance of regional and traditional ways of living. Thus, existing traditional villages face a serious existential threat. The protection of traditional villages and cultural heritage in China is a critical and ongoing effort. The protection of traditional village cultural resources has attracted wide attention from all sectors of society. Relevant governmental authorities of China began conducting survey and appraisal of “Chinese Traditional Villages” (defined as “traditional villages” hereafter) in 2012 and have already added 6819 traditional villages to their directory of traditional villages. The Jiangxi Provincial Department of Housing and Urban-Rural Development announced the first batch of 248 provincial-level traditional villages in August of 2017.
Modern research of Chinese traditional villages began broadly in the 1980s, with significant findings coming from many fields of study. Investigations of the spatial form of traditional villages primarily focus on two areas (
Fu and Huang, 2016): the spatial distribution patterns of traditional villages from a geographical perspective (
Liu et al., 2014;
Sun et al., 2017;
Li et al., 2018a), and the spatial relationships with various elements within these traditional villages from an architectural perspective (
Wang and Zhou, 2011). At the same time, with the development of spatial analysis computing theory and intelligent technology, a variety of spatial quantitative research ideas and methods produced in the fields of mathematics, computer science, and complex systematic science have gradually penetrated into village morphology research (
Yang, 2020). Since the early 2000s, GIS has been applied to multiple levels of historical and cultural geography (
Ebert, 2004;
Lv and Guo, 2016). Combined with GIS spatial statistics methods, spatial patterns and their changes can be quantified (
Entrikin, 2001;
He et al., 2018). Space syntax can rationally and efficiently do quantitatively analyze on the settlement space from the spatial relationship point of view, so as to analyze and summarize the spatial structure of traditional villages. Fractal theory breaks the traditional method of qualitatively dividing the form of villages. It simplifies the complex and disordered village space into irregular geometric figures, and uses geometric eigenvalues to improve the accuracy of the description of village morphological characteristics (
Li et al., 2020).
In summary, compared with traditional research methods, interdisciplinary quantitative analysis methods and mathematical models have expanded the ideas and fields of traditional village spatial morphology research (
Deng et al., 2021). The depth and breadth of quantitative analysis methods of spatial morphology are also increasing. However, due to the lack of a large number of field verification samples and data, most traditional village studies are only limited to qualitative analysis, and do not study the characteristics of spatial form and law of villages from a quantitative perspective. Thus, we take Fuzhou area of Jiangxi Province, which contains many distinct cultural characteristics and prominent regional integrity as an example, apply the methods of GIS, space syntax and fractal geometry, and use a combination of qualitative and quantitative approaches to analyze the spatial distribution and spatial morphological characteristics of traditional villages in Fuzhou from a cultural geographic perspective, with the objective of providing improvements for the future protection, construction, and development of traditional villages.
2 Study area and data
2.1 Study area overview
Fuzhou City, located in eastern Jiangxi Province, China, includes 11 counties and districts of Guangchang County, Nanfeng County, Lichuan County, Nancheng County, Zixi County, Jinxi County, Yihuang County, Chongren County, Le’an County, Linchuan District, and Dongxiang District (Fig.1). The area has a subtropical monsoon climate with four distinct seasons (average temperature = 16.9°C/yr−18.2°C/yr), abundant rainfall (precipitation = 1600−1900 mm/yr) and sufficient sunlight (sunshine duration = 1582−1928.1 h/yr). The terrain of Fuzhou is high in south and low in north, surrounded by mountains in east, south and west, and gradually inclined to Poyang Lake Basin in north, forming a relatively closed environment. The landscape is complex, with mountains, hills, hillocks, and valley plains; hills are the dominant feature. The region has a dense network of river systems, with the Fuhe River and its tributaries being the major waterway.
Fuzhou area is within the Linchuan cultural region, a regional culture that is the product of the Jiangyou people from eastern Jiangxi Province. It appeared during the Qin and Han Dynasties, reached its peak during the Song Dynasty, and lasted up until the Ming and Qing Dynasties. The culture made brilliant achievements in education, literature, the arts, science, and other fields, and developed into a unique and prominent regional culture throughout the long course of its history. During the Ming and Qing Dynasties, the commodities economy of Fuzhou was dynamic. There are a large number of country fairs in Fuzhou, and in terms of their number and scale, they exceed the level of Yangtze River and Pearl River Delta during the same period (
Li and Wang, 2002). Ancient post roads and rivers have built a bridge between merchants and the country fairs, providing the possibility for the prosperity of the market and long-term trade. So, a large accumulated wealth derived from highly developed trade networks, and a prosperous economy and commercial wealth which provided material guarantees for settlement construction. Overall, the unique physical geographic conditions and profound cultural deposits of the area gave rise to a unique settlement culture. There are numerous traditional villages in Fuzhou area that are densely distributed, creating distinct regional characteristics, which provide an ideal domain for the study of traditional villages in China.
2.2 Data sources
Fuzhou area had 96 designated national-level traditional Chinese villages as of December 2018, according to the Ministry of Housing and Urban-Rural Development of the People’s Republic of China and 83 designated provincial-level traditional villages according to the Department of Housing and Urban-Rural Development of Jiangxi Province. Our study will analyze these two designated groups of villages, excluding duplicate data, with a total of 114 are considered. The location coordinates of traditional villages are queried using Google Earth Pro; Digital Elevation Model (DEM) comes from the ASTER DEM digital elevation data set with a resolution of 30 m and the ALOS DEM digital elevation data set with a resolution of 12.5 m in the Geospatial Data Cloud. The water system data comes from the basic data set of 1:250000 in Jiangxi Province. Ancient post road comes from the Chinese Historical Atlas edited by
Tan (
1982).
3 Macroscopic regional level: spatial distribution characteristics of traditional villages
3.1 Overall spatial distribution of traditional villages
Heat map is good at showing where point features are concentrated. The kernel density method was used to calculate the heat map density. The work of kernel density is to calculate the density of point features around each output raster cell. So, we used the kernel density method (bandwidth = default value) to form a heat map of traditional villages in Fuzhou area (Fig.2.). The layout of the spatial distribution of traditional villages is in clusters, with a single high-density cluster seen, and having an uneven overall distribution. The high-density cluster is located in north region of Fuzhou, in Jinxi County. Jinxi County contains the most villages (52) of the 11 counties and districts considered here, accounting for 45.6% of the total. This area is located on flat land with an average elevation less than 100 m. The southern parts of Nancheng County and northern regions of Nanfeng County, as well as Le’an County, also have a noticeable but lesser density of traditional villages.
ArcGIS Pro can apply the closest proximity point index method, which displays traditional villages as abstract points to distinguish spatial types. We consider three types of spatial distribution which are typical in nature and society: random, dispersed, and clustered. Z-score (Z), or standard score, is the number of standard deviations a given data point lies above or below the mean. When −1.65<Z<1.65, the distribution is considered random; when Z>1.65 it is a dispersed distribution and when Z<−1.65 it is a clustered distribution. Calculation of the closest proximity index for the 11 counties and districts showed that the highest proximity index of the traditional villages in the Fuzhou area was 0.813, signifying a random distribution. Statistical results (Tab.1) of each county and district show that the traditional village distribution in Jinxi County is clustered and Nanfeng County is random, while the distribution in the other counties is classified as dispersed.
3.2 The relationship between landform and the spatial distribution of traditional villages
Altitude has a fundamental influence on the distribution of traditional villages, as increases of altitude bring decreases in air temperature and pressure. Settlements are thus constrained by ranges of temperature and pressure, they are “suitable” for human habitation. The east, south and west of Fuzhou area is surrounded by mountains, while central Fuzhou alternates between hills and basin valleys. The terrain is high in the north and low in the south, inclining gradually toward the plains of Poyang Lake. The landforms here are complex, formed primarily of hills interspersed by mountains, hillocks, and valley plains. Mountainous areas with an altitude over 500 m account for 30% of the total area, hills between an altitude of 200−500 m account for 50%, and hillocks and valley plains below 200 m account for 20%.
ArcGIS Pro was used in combination with an elevation map to compare the spatial distribution of traditional villages in Fuzhou region, binning the results as a thematic map of elevation distribution between 0 and 200 m, 200−500 m, 500−1000 m and over 1000 m (Fig.3). The average altitude of all traditional villages in the Fuzhou area is 258 m, lower than the area’s overall average altitude. 86% (98 villages) of the traditional villages in Fuzhou area are found in the hillocks and valley plains below 200 m, while 8% (9 villages) are found in the hilly areas between 200 and 500 m and around 6% (7 villages) are distributed in the mountainous areas between 500 and 1000 m. Plain areas and the low-lying slopes of hillocks are much better for construction and there are more resources of cultivatable land; therefore, the site selection of these villages and the suitability of these sites for human existence show clear connections.
3.3 Relationships between water systems and the spatial distribution of traditional villages
Water is one of the most important resources for human survival. When selecting a village site, people would therefore need to consider the relative abundance of water sources, and not exposing themselves to excessive flooding risk. Water systems provide humans with vital water sources, and they were also an important mode of transportation in traditional agricultural societies; therefore, there is a remarkably close relationship between the distribution of traditional villages and water systems.
Fuzhou area has well-developed water systems. The major water system of the region is Fuhe River and its tributaries, which cover the entire area. Tributaries of Ganjiang River and Xinjiang River are also prominent in Fuzhou. The Fuhe River is second only to Ganjiang River in Jiangxi Province, with a length of 349 km and a drainage area covering 15800 km2. ArcGIS Pro was applied to map the current water systems in the Fuzhou area at the 500 m, 1000 m and 1500 m buffer boundaries and these maps were then combined with the traditional village distribution data to explore the relationships between traditional villages and water systems (Fig.4). The majority of traditional villages in Fuzhou area are located beside water systems, with 84% (96 villages) found within a 500 m buffer zone from rivers. The remaining 16% of traditional villages are found within a range between 500 and 1500 m. Thus, all traditional villages are located within 1.5 km from water systems or tributaries.
3.4 Spatially stratified heterogeneity analysis
Geographic detectors are statistical tools which can explore spatial heterogeneity and reveal their determining factors (
Wang et al., 2017). Geographic detectors are used here to study the main driving factors of traditional village distribution in Fuzhou. Geographic detectors employ multiple methods, including the measurement of stratified heterogeneity among data, testing of the coupling between two variables
X and
Y, and the investigation of the interactions between two explanatory variables
X1 and
X2 to a response variable
Y, without any specific form of interaction. Each of the tasks can be accomplished by the Geodetector
q-statistic:
where h = 1, 2, …, L is the strata of variables X or Y, N and 2 represent the number of units and the variance of Y in, respectively. The value of q is limited to the interval of [0, 1]. The larger the value of q, the stronger the impact of independent variable X on dependent variable Y. This method is used to study the main driving factors which influence the distribution of traditional villages. The DEM (X1) and the river system (X2) are used to investigate the spatially stratified heterogeneity of the traditional villages. These results show that both DEM and water system distribution significantly affect the spatial distribution of traditional villages, and the influence of elevation is greater than that of water system proximity (Tab.2). In other words, people first prefer to settle in a flat area, and then consider the distance of the water system.
4 Microscopic (individual) level: morphological characteristics of traditional villages in Fuzhou area
From the micro level or the individual level, we study the four aspects of the traditional village morphology, the village scale, the village landing appearance, the village boundary form and the village living patterns.
4.1 Village scale
The morphological contours of the 114 villages are sketched on Google Earth and input into the GIS system to determine the scale of each village. The size distribution of the scale of traditional villages in Fuzhou area is split into 0−10, 10−20, and 20−40 km2 and classified as small, medium, and large, respectively. Statistical results show that the majority of traditional villages in the Fuzhou area are of medium or small scale, with large-scale villages accounting for only around 10%. An example of a medium-scale village, Dongyuan Village in Jinxi County, is shown in Fig.5. The majority of counties and districts have on average one or two large-scale settlements which are relatively equally distributed regionally. There are 11 large-scale villages in total, including Huwan Village in Jinxi County, Liyang Village in Dongxiang District, Shangtang Village in Nancheng County, among others. The majority of these 11 traditional villages are located in market towns. The scale of settlements is determined by physical geographic conditions, and as the area is limited by its landform, cultivated land resources, and other conditions. Overall, the majority of settlements considered here are medium to small.
4.2 Village landscape
Early settlers in China placed upmost importance on the landscape of the area. On one hand, the village’s surroundings were influenced by physical geographic features, and on the other hand, people were influenced by traditional Chinese concept of
feng shui. The concept of
feng shui embodies ancient Chinese understanding of physical geography and the rational utilization of terrain, wind, water, etc., which advocates a harmonious coexistence with nature. The ideal model for selecting the site of a village under the
feng shui perspective is “mountains behind and water in front” (
Chen and Song, 2009). Thus, many village sites were selected with a mountain to the back or built beneath a hillock, with a natural water source in front or excavated ponds, which can form a three-way spatial landscape consisting of a water body, followed by the village itself, and then the mountains or raised land behind it.
We use ArcGIS Pro to convert the DEM raster data (12.5 m resolution) in Fuzhou City into contour lines with a spacing of 50 m, to display the terrain elevation around traditional villages. The buffer analysis tool of ArcGIS Pro is used to limit the distance to 1 km around the traditional village as a buffer, and environmental factors such as the water system and mountains within 1 km of the village are considered. Fuzhou area has a complex topography and geomorphology, and a large overall elevation range. Thus, we here define three distinct classifications. When the absolute elevation of the surrounding environment of the village is less than 500 m and the relative height difference within the 1 km buffer area around the village is less than or equal to 50 m, it is a “flat” (D1) village. When the absolute elevation is less than 500 m and the relative elevation difference in the surrounding 1 km buffer area is greater than 50 m, it is a “hilly” (D2) village. When the absolute elevation of the surrounding environment of the village is greater than 500 m, it is a “mountain” (D3) village. The water system in Fuzhou is well-developed, and the location of traditional villages is highly correlated with water bodies. Villages with rivers and lakes within 1 km are classified as villages W1, villages with ponds within 1 km are classified as villages W2, and villages without water systems or ponds within 1 km are classified as villages W3. With the overall consideration of elevation and water systems, we classified Fuzhou’s traditional village environments into eight modes: D1W1 (flat river banks), D1W2 (flat land with ponds), D1W3 (flat land on all sides), D2W1 (river bank hillocks), D2W2 (hillocks behind and ponds in front), D2W3 (hillocks on all sides), D3W1 (mountains behind and water in front) type, D3W3 (mountains on all sides). Fig.6 shows examples of each of these types.
4.3 Village boundary morphology
The boundary of the spatial shape of the village can be quantitatively described by the cluster shape, the band shape, and the shape index. The value of the aspect ratio of the village boundary (indicated by
λ) is expressed as the length and narrowness of the village boundary (
Xie, 2015). We extract the various endpoints of village elements such as buildings, courtyard walls, and structures, and connect the enclosed two-dimensional closed figure planes as the village boundary (
Cao and Zhu, 2019). We use “
b” to represent the longest axial distance between any two endpoints of the boundary and “
a” to represent the shortest distance between the two ends of the boundary; the intersection angle between the long axis and the short axis is constrained to be between 70° and 110°. The specific calculation of
λ is:
When the traditional village boundaries are round or square, the value of
λ of the village is closer to 1 (
Xie et al., 2015); when the value of
λ is larger, the village boundary graphics tends to be more banded. Therefore, we set
λ = 2 as a critical point. If the value of
λ is greater than 2, the spatial morphology of the village tends to be band-shaped; if the value of
λ is less than 2, the spatial morphology of the village tends to be cluster-shaped.
The village shape index can reflect the unevenness of the village boundary and the overall irregularity. The larger the value of the shape index, the more irregular the village boundary, according to
where
S is the shape index of the village boundary,
P is the perimeter of the village boundary shape,
λ is the aspect ratio of the village boundary shape, and
A is the area of the village boundary shape (
Li et al., 2018b). The minimum value of the shape index,
S, is 1. When the value of the shape index is close to 1, the shape of the village boundary approximates a circle; when the value of the shape index is greater than 1, the shape of the village boundary is more complex and irregular (
Wang, 2006). Our results in Fuzhou area show that when the shape index is greater than 1.5 (
S >1.5), the village boundary shape has a finger-like divergence pattern. Therefore, the village with a shape index greater than 1.5 (
S >1.5) is defined as a finger-shaped village.
The combination of the above-defined indicators leads to four rough categories, “lumpy villages”, “banded villages”, “finger-like villages with banded features”, and “finger-like villages with cluster characteristics” (Fig.7; Tab.3).
According to statistical analysis of all the traditional villages, 41% of the villages are generally finger-shaped and have the largest number. Lumpy shaped villages and band shaped villages divide the remaining part equally.
4.4 Village living patterns
The general form of traditional villages in Fuzhou area is based on a history of clans living together with a centralized distribution. The theory of spatial syntax can be used to select several distinct types of typical villages to draw axis models, and villages can be roughly divided into “comb”, “row”, “crisscross” and “freestyle” living forms according to their synergy values (Tab.4). Synergy represents the relationship between local integration degree (
R = 3) and global integration degree (
R =
n), expresses the degree of connection between local space and overall space and measures the degree of synergy via correlation analysis using scatterplots. The closer the correlation value (
R2) of synergy is to 1, the closer the space tends to be a single core space, and if closer to 0, the overall layout and optimization strategy is weaker (
Chen et al., 2018). When
R2 is about 0.7, there is a significant correlation between the two variables, and the village living pattern is closest to a single-core space layout. When
R2 is about 0.5, there is a general correlation between the two variables, and the village living pattern is relatively close to a single core. When
R2 is about 0.3, there is a lower correlation between the two variables, the village residence pattern is a crisscross layout close to a multicore space, and when
R2 is about 0.1, the village living form is a freestyle layout closest to a multicore space. In summary, traditional villages can be divided into four types of living patterns: comb (i.e., columns), rows, crisscross, and freestyle (Fig.8).
Comb-type settlements have been extensively studied in Guangdong Province (
Hong, 2018). However, Fuzhou area of Jiangxi Province has not received as much attention. Comb settlements in Fuzhou area of Jiangxi Province have similar morphological characteristics to the comb settlements in Guangfu area of Guangdong Province, with a structured shape. The street texture is in a “one horizontal, many vertical” layout, with a main street in front of the village and several lanes distributed perpendicular to the main street, parallel to the prevailing wind direction in summer and planned so as to adapt to the local climate. Another feature of the comb layout is that living units are arranged in depth, with the layout of living units running parallel in the direction that the lanes extend. This layout is suitable for an environment that has mountains behind and a water system in front, and where the terrain is low in front and high in back, as it helps drain the water into the excavated ponds in front of the village. The layout of row settlements is neat and orderly and differs from comb settlements in that the living units are arranged in horizontal rows, with the lanes orientated in the same direction as the rows of living units and the living units face the streets. Planning for the layout of crisscross settlements has characteristics of both the comb and row patterns, with the lanes distributed in a grid shape and present at either side of or in front of living units. The freestyle layout is when the layout of the buildings is more dispersed with lanes irregularly distributed at random.
The traditional villages in the Fuzhou area are primarily of the comb pattern and distributed over most of the Fuzhou area. There are only two traditional villages of a row pattern in Fuzhou and neighboring Ji’an area. The layout of row pattern villages is more traditional in the Ji’an area. Crisscrossing layout villages occur primarily in mountainous areas and are also relatively small in number.
5 Relationships between the macroscopic regional level and microscopic individual level of traditional villages
The characteristics of traditional villages in Fuzhou were analyzed above from a macro-regional distribution perspective (topography and water system) and a micro-village morphology perspective (the scale, the village landscape environment, the village boundary morphology and the village living form). However, there are many factors that can affect the morphology of the traditional villages. These include natural and geographical factors, social and human factors, and policy factors. Temporal correlations of macro-distribution characteristics, micro-morphological characteristics and influencing factors are investigated here to determine formation patterns of traditional villages. Its formation mode is determined, and its spatial distribution and its relationship with the natural environment are revealed.
We first combine the scale distribution map with elevation maps, water system maps and ancient road maps (Fig.9). Although most traditional villages in Fuzhou are small and medium-sized, larger villages are generally located in flat areas with lower elevations and are located close to water systems and traffic arteries. This is due to the fact that ancient transportation relied on water and land transportation. The economy was highly developed with convenient transportation and the population was concentrated, and large-scale villages were able to develop.
We next compare village landscape maps with elevation, water systems and ancient post roads. There are six types of village landscape patterns found in low elevation areas (average elevation <500 m), while two main types were generally found in relatively higher elevation areas (average elevation >500 m). The low elevation types are D1W1, D1W2, D1W3, D2W1, D2W2, D2W3, while the relatively high elevation types are only D3W1 and D3W3. Thus, the microscopic landscape of the village is consistent with the macroscopic topography, but different kinds of landscapes exist in different places due to the differences in the micro-topography.
Next, we compare living patterns map combined with elevation, water system and ancient post roads (Fig.10). This investigation shows that free-style settlements are mainly located in areas with higher altitudes and are greatly affected by topographical factors. The other three kinds of patterns were mostly distributed in low-altitude areas.
We sum the number of different village scales in different living patterns to study the relationship between village scale and living patterns (Tab.5). Villages with crisscross patterns are mostly found in middle or large-scale villages, as these village patterns are mostly located along rivers or close to traffic arteries which are economically developed, and because village construction was mainly affected by different cultural and economic factors, which, over time, led to larger village size. The scale of free-style patterned villages is mostly small, due to surrounding terrain restrictions and insufficient arable area.
The percentage of different living landscapes in different living patterns is presented as a pie chart in Fig.11. Most of the villages with crisscross patterns are of the D1W1, D1W2, and D2W1 types, generally close to water systems. Freestyle-patterned villages were mainly found in mountainous regions and are affected by topographical factors. Comb and row patterned villages were widely distributed among mountain and water environments, meaning that topography was not the decisive factor for these village patterns.
There are two major factors affect living patterns: the natural geographical factors and social or human factors. Fuzhou is part of the greater Linchuan cultural area, and Jiangxi Province was the birthplace of neo-Confucianism thought. Thus, this area was profoundly affected by Confucian ritual culture with a corresponding importance given to compact layouts, a keen sense of order. The orderly formation of the comb layout of traditional villages is a material reflection of the spatial order of ritual culture, and a spatial expression of social order. This area has also been heavily influenced by cultural communication. Linchuan, Jinxi, and Dongxiang belonged to the core area of Linchuan culture, while Le’an, Guangchang, and Lichuan belonged to marginal areas of culture. Le’an County is greatly influenced by the adjacent culture of Luling. Comb pattern villages were the main living pattern used in the Linchuan Cultural Area which reflected its distinctive regional cultural characteristics. Row pattern villages are mainly located in Le’an County, which is adjacent to the Luling Culture Area, and is influenced by this culture.
A superposition analysis of the village landscape environment, village boundary morphology, and village scale found that 90% of the villages in Fuzhou are located in areas near water systems and are in flat and hilly areas with an altitude of less than 500 m; very few are located in the mountains. Areas far away from water systems are still found at lower elevations (below 700 m,) and the distance to water systems is always within 1500 m. The landscape environment type of the most villages in Fuzhou is D2W1(river bank hillocks), accounting for 40.36%, the second is D2W2 (Hillocks behind and ponds in front), accounting for 28.07%, and D1W1 (flat river banks) is next, accounting for 15.78%. The D1W3 (flat land on all sides) type is the least common, which only accounts for 0.87%. This is due to the fact that the topography of Fuzhou area is dominated by hillocks. Even if it is far away from rivers and lakes, there are usually common ponds found around the villages.
Most of the traditional villages in Fuzhou have the forms of finger-shape boundaries. 76.6% of the finger-shaped villages are surrounded by hilly landform, and 23.4% of the finger-shaped villages are located in flat areas. Due to its cohesive nature within the settlement and because it is not restricted by the topography during the expansion process, the irregularity of the village boundary is weaker, and the boundary of the village in the flat terrain is often more “lumpy”. Villages in the form of “belt” are generally affected by water systems. They are built along the river, and their expansion is limited by the water system, presenting a zonal boundary shape. The boundary morphology of villages in hilly areas or mountain areas tends to be finger-shaped due to the influence of topography, especially in mountainous environments and river ravine areas. The elevation difference limits the free expansion of the village boundary and even the village size, which will also result in small village sizes in general.
If we only consider large-scale villages, we find 12 large villages with an area of more than 20 km2, which are all located in hilly or flat areas, and 92% are close to water systems. There are four types of landscape environment found for large-scale villages: D1W1 (flat river banks), D2W1 (riverbank hillocks), D2W2 (hillocks behind and ponds in front), and D2W3 (Hillocks on all sides). There are also a sizable number of large-scale villages of the D2W1 (riverbank hillocks) type, and the D2W2 (hillocks behind and ponds in front) type. The low-altitude hill landform has little restriction on the expansion of the village, and sufficient water resources ensure the environmental capacity of the village. Villages near rivers can also meet the transportation needs of the villages. Therefore, hills near the water system are the environmental foundation for the development of most large villages in Fuzhou. Flat land is relatively rare in the macro-topography of Fuzhou, but it is the environment where most large villages are found (Fig.12). We thus conclude here that the location of large villages in Fuzhou is biased toward low-altitude, hillock, or flat land near water systems.
Finally, there is an uneven spatial distribution of traditional villages in the counties and districts of Fuzhou, with Jinxi County forming a high-density cluster. This is due to two main reasons. First, this area is located on plains and hillocks, and the number of traditional villages here was high to start with. Second, there was a special fund of 220000 yuan each year to reward individuals who repair ancient buildings and a total of more than 100 million yuan was secured for the protection of various cultural relics which led to the restoration of a large number of ancient buildings in Jinxi County. Thus, effective government policies and a variety of energetic approaches helped to formulate this “Jinxi model”, which has helped to preserve a large quantity of traditional villages.
6 Conclusions
The spatial relationship of traditional villages developed over thousands of years was the combined result of people adapting to production and living, their spiritual, social, and cultural needs, and specific physical geographic conditions. We have here applied GIS technology using qualitative and quantitative methods to conduct analyses on both macroscopic and microscopic levels to the morphological characteristics of the traditional villages in the Fuzhou area. Our study found the following conclusions.
1) The spatial distribution of traditional villages on the macroscopic regional level is unevenly distributed, generally found at low elevation, and typically seen in close proximity to water systems. A high-density area is found in Jinxi County, owing to strong protection and restoration policies. The majority of traditional villages on a microscopic (individual) level are medium to small scale, with eight different landscape classifications. Village layouts are primarily comb-like patterns.
2) Village scales are limited by physical geographic conditions and economic and cultural factors. The limitations of the hilly terrain and dense population led to the majority of traditional villages in this area being medium to small scale, with the majority of large-scale villages appearing as market towns located in the low altitude and being close to water areas and traffic arteries.
3) The micro landscape environment of village site selection is consistent with the macro topography, but due to the difference of micro landforms, village landscapes are primarily dictated by the rich variety of landforms in Fuzhou area. However, the majority of villages were based on the “mountains behind, water in front” model, which is influenced by the traditional Chinese concept of feng shui and the desire to build habitats that are suitable for living.
4) The boundary morphology and scale of villages are greatly affected by proximity to water systems and elevation of the surrounding environment. Village boundaries in flat areas tend to be clumped or finger-shaped, while village boundaries near water systems tend to be belt-shaped. When the village is located in mountains and hills, the border of the village has a high degree of irregularity due to the influence of the topography, and is also affected by local water systems, especially river gulleys. Thus, the overall border morphology of these villages which often are finger-like villages with cluster characteristics.
5) The primary living pattern of traditional villages in Fuzhou is the comb pattern. Free-style and crisscross pattern villages are greatly affected by terrain factors. Crisscross patterns are often found in large scale villages, while free-style patterns are often found in smaller villages located in high-altitude areas, as their expansion is limited by local landform environments. Conversely, comb and row pattern villages are more affected by cultural factors. Comb pattern villages are typically located in the core cultural area, while row pattern villages are typically located in the cultural outskirt areas.
PDF
(13563KB)
