Introduction

The protection of traditional villages has become an important task for Chinese countryside revitalization since it is emphasized that “the importance of preserving traditional villages and rural features” at the 2020 Central Rural Work Conference. According to the definition given in a document jointly issued by four national ministries and commissions, including China’s Ministry of Housing and Urban-Rural Development, traditional villages are villages that were formed earlier, have richer traditional resources, have certain historical, cultural, scientific, artistic, social and economic values and should be protected because they are rich in traditional culture and are the greatest legacy of China’s agrarian civilization (Feng et al. 2017; Hu et al. 2014). The protection of traditional villages is related to the inheritance of history and culture (Katapidi 2021), the development of rural characteristics and the enhancement of rural appearance. At present, China has accumulated certain achievements in traditional village protection, and till 2020, 6,819 traditional villages have been included in the national list, forming the world’s largest cluster for the protection of the heritage of agricultural civilization.

However, due to rapid industrialization and urbanization in past decades, traditional villages have been declining or even disappearing (Wang et al. 2021). Thus, how to protect the traditional villages during the process of urban-rural integration is still a research question. Especially when the population dynamic is taking place greatly, it needs to strengthen the research and protection of rural areas to keep their traditional spatial patterns and social-economical characters (Balsa-Barreiro et al. 2021). For solving this problem, planning approaches are usually employed, such as develo** spatial protection policies to keep traditional spatial patterns (Ma et al. 2021), economical revitalize policies to attract labors (Ian and Yue 2022), etc. Although researchers and governors paid great efforts on develo** protection policies for accelerating village development and in the meantime, kee** its traditional characters, some traditional villages are still facing problems and challenges such as showing a singularization development paths (Daskon 2010) and the fragmentation of historical resources (Zhang et al. 2020). Due to the reasons, to some extent, lack of spatial characteristics and its evolution process of traditional villages on spatial and social-economical aspects would be the key. In this context, it is important to study the spatial characteristics of traditional villages at a macroscopic regional scale and to analyze the develo** history and influencing factors on their distribution and evolution (Ma and Tong 2022) of traditional villages. This is also the purpose of present work.

To a certain extent, a comprehensive multifactor analysis would provide a basis for recognizing the effective factors for the conservation and sustainable development of traditional villages in China and provide a basis for further in-depth case studies of traditional villages. Existing literatures in this field mainly covers macroscopic scales such as national (Kang et al. 2021), especially ship** by rivers is recognized as the key influencing factor and also one of the main characters of Fujian’s traditional village distribution among China (Zhang and Yu, 2021; Que et al. 2023). Some scholars in the existing literature believe that a relatively large feature of Fujian Province compared with Zhejiang and Guangdong Province (Feng et al. 2017; Wu and Yu, 2021) is that the spatial order of Fujian Province is back to the mountains and facing the sea (Hu et al. 2021). The distributions and development of traditional villages have a great correlation in natural factors of water and ship**. But relevant researches seldom explain the strong and weak sequence of influencing factors quantitatively. Lacking of complete data support across natural and cultural of traditional villages and systematic research on influencing factors make them relatively weak in explain the distribution and characteristics of traditional villages in Fujian province, and thus, supporting government recognize the key factor for making protection policies. It is also supposed to be the contribution of present work.

Based on this, present work takes Fujian Province as the study area, systematically composes the base map of traditional villages in Fujian Province by integrating multiple elements across natural and cultural aspects. From the two dimensions of natural conditions and humanistic conditions, present research selects four indicators, including terrain and landscape factors system, river system, ancient and modern transportation system, and socio-economic system, in order to comprehensively analyze the influencing factors of the spatial distribution of traditional villages in Fujian Province, and systematically study the influence degree of these influencing factors on the spatial distribution of traditional villages. These series of calculation indicators and methods are objective and effective. On the basis of making full use of traditional village data at all levels in Fujian Province, present research rationally uses relevant calculation indicators. Present research is supposed to provide reference and certain guiding suggestions for the follow-up special protection planning of territorial space, and provide ideas for the study of traditional villages at the provincial scale, and provide reference for the rural development of other countries and regions in the world.

The remainder of present research is organized as follows: the overview of the study area, various data sources, research framework and research methodology are introduced in Section Data Sources and Research Methodology. In the section of Spatial distribution characteristics of traditional villages, the spatial distribution characteristics, pattern, density and trend of traditional villages in Fujian Province are analyzed clearly. In the section of Analysis of Influencing Factors, it analyzes the impact of various influencing factors on the spatial distribution of traditional villages in Fujian Province and the strength of the impact from both natural and human aspects. Finally, in Section Conclusions and recommendations, the research results are discussed, and the targeted protection suggestions are put forward.

Data sources and research methodology

Study area overview and data sources

Study area

Fujian Province is located on the southeast coast of China, with an irregular trapezoid shape with two parallel sides and concave sides, and is governed by nine districts and cities, with a total area of about 124,000 hm2 (Fig. 1). The unique mountain and sea natural geographical environment and Hakka social culture in Fujian Province have bred a large number of valuable ancient villages, including traditional villages at various levels, famous historical and cultural villages, and traditional villages with high conservation value that have not yet been included in the above list (Fig. 2) (Dai, 2000). At the same time, the total number of national traditional villages in Fujian Province ranks sixth in China, and its quantity and quality occupy a leading position in the country, with outstanding value for conservation research. The most famous traditional village in Fujian Province is the Hakka traditional village in the west of Fujian Province, which was formed when the Han people in the Central Plains fled the war and moved south five times. For safety reasons, the Hakka people choose to live in the valleys or mountains in the southwest of Fujian Province. The earliest Hakka settlement was in Shibi Village, Ninghua County, and then moved along the Tingjiang River valley in Fujian Province, and integrated with the local ethnic groups to form a unique cultural characteristics of the Hakka people. Typical houses of Hakka people generally present tulou and fort, which is also for the need of defense and the consideration of life privacy. This paper takes the various types of traditional villages and famous historical and cultural villages① in Fujian Province at all levels in the list as the object of study, including seven batches of national-level famous historical and cultural villages, a total of 57 villages, six batches of provincial-level famous historical and cultural villages, a total of 138 villages, five batches of national-level traditional villages, a total of 494 villages and four batches of provincial-level traditional villages, a total of 917 villages. A few villages exist in both the traditional villages and famous historical and cultural villages in the list due to their special value and significance However, due to the necessity of classification research, they are not screened out. According to the criteria for village classification and selection in China, both traditional villages and historical and cultural villages are villages with important historical value. In addition, this study also needs comprehensive village data as support, so as to draw relatively reliable conclusions. Therefore, the total number of traditional villages in Fujian Province studied in this paper reaches 1606 (Table 1).

Fig. 1: Location map of Fujian Province in China.
figure 1

The figure shows the position of Fujian Province in China. It can be clearly seen that Fujian Province is in the southeast of China. The red line is the border of Fujian Province, whose capital is Fuzhou.

Full size image
Fig. 2: Spatial distribution of traditional villages in Fujian Province.
figure 2

This figure shows the specific distribution of traditional villages at various levels in Fujian Province, China. The dark red dots represent national traditional villages, the light red dots represent provincial traditional villages, the dark green dots represent national historical and cultural villages, and the light green dots represent provincial historical and cultural villages.

Full size image
Table 1 Statistics of Traditional Villages in Fujian Province at all levels.
Full size table

Data sources

The data studied in this paper includes three major categories: (1) Map data is based on DEM data of Fujian Province, and spatial information such as elevation, slope, slope direction, water system and roads are extracted to establish a basic database of Fujian maps, and the data source is the open source website of geographic data. (2) The village data was obtained from the official websites of the Ministry of Housing and Urban-Rural Development of China and the Department of Housing and Urban-Rural Development of Fujian Province. The geographical coordinates of traditional villages were mainly obtained through online map queries, and the spatial distribution map of traditional villages in Fujian Province was drawn by overlay. (3) Socio-economic data related to cities and counties as well as data on ancient post roads were obtained through statistical yearbooks, local chronicles and historical atlases, among which statistical yearbooks were obtained through the official website of Fujian Provincial Bureau of Statistics by searching 《Fujian Statistical Yearbook – 2021》.

Research framework

Firstly, the traditional village data of Fujian Province at all levels and types and the map data of Fujian Province were used as data sources to construct a basic data-base of traditional villages in Fujian Province. Secondly, the spatial distribution characteristics were analysed in depth. Specifically, the statistical analysis model was used to determine the type of spatial distribution and analyse the overall distribution status of traditional villages; by using the spatial autocorrelation analysis tool, the regional relationship pattern of the spatial distribution of traditional villages was measured; on this basis, the kernel density estimation method was used to sort out the spatial distribution density of traditional villages and analyse the clustering area of traditional villages. On the basis of this, the spatial distribution density of traditional villages is sorted out using the kernel density estimation method, and the specific situation of traditional village clusters is analysed. The spatial distribution trend is determined by using the standard ellipse difference tool, and the expansion trend of traditional villages is analysed. Finally, by combining the spatial distribution characteristics with the correlation characteristics of the influencing factors, the distribution pattern of traditional villages in Fujian Province is revealed and summarised, and targeted planning and protection suggestions are made (Fig. 3).

Fig. 3: Spatial distribution characteristics of traditional villages in fujian province and analysis framework of influencing factors.
figure 3

This figure shows the research framework from the construction of the database to the analysis of the specific spatial distribution characteristics of traditional villages to the analysis of the influencing factors to the proposal of conservation strategies.

Full size image

Research methodology

Statistical analysis models

The statistical analysis models such as the nearest point index, geographical concentration index and imbalance index were used for quantitative analysis to determine the types of spatial distribution of various types of traditional villages at different levels in Fujian Province (Table 2) (Zhang et al. 2020). The innovation of this statistical analysis model is that it can accurately determine the spatial distribution type of traditional villages in Fujian Province by using a variety of indicators and combining the data of all kinds of traditional villages in Fujian Province for comprehensive analysis.

Table 2 A statistical analysis model of the spatial distribution of traditional villages in Fujian Province.
Full size table

Spatial autocorrelation analysis

Spatial autocorrelation analysis is an analytical method to examine the interactions and interdependencies between regions in space, including global autocorrelation and local autocorrelation. The former is commonly used to analyse the global Moran’s I index as a measure of the degree of spatial autocorrelation, while the latter is commonly used to explore in detail the distribution state of spatial elements in local space with the local Moran’s I index. In particular, the global Moran’s I index represents the spatial characteristics of an attribute of a study object across the study area, and is used to measure the overall spatial correlation between regions, and is calculated as follows (Wu, 2014):

$$I = \frac{n}{{S_0}} \times \frac{{\mathop {\sum }\nolimits_{i = 1}^n \mathop {\sum }\nolimits_{j = 1}^n w_{ij}\left( {y_i - \bar{y}} \right)\left( {y_j - \bar{y}} \right)}}{{\mathop {\sum }\nolimits_{i = 1}^n \left( {y_i - \bar{y}} \right)^2}}$$
(4)

where: yi and yj denote the attribute values of the ith spatial unit and the jth spatial unit respectively; Wij is the spatial weight value; S is the sum of Wij; n is the total number of spatial units under study. the value of Moran’s I index is between [−1, 1], its value greater than 0 is positive correlation, indicating the existence of agglomerative distribution characteristics, less than 0 is negative correlation, indicating the existence of dispersive distribution characteristics, equal to 0 indicates no correlation, indicating a random distribution.

While the local Moran’s I index indicates the specific spatial location and distribution of agglomerative centres, which can better reflect the homogeneity and heterogeneity of spatial elements, five relationships usually emerge: high-high clustering (HH), i.e. high-value centres and surrounded by high-value areas; high-low clustering (HL), i.e. high-value centres and surrounded by low-value areas; low-high clustering (LH), i.e. low-value centres and surrounded by high-value areas ; low-low clustering (LL), i.e. low value centres surrounded by low value areas; and no obvious clusters. In this paper, the global Moran’s I index is used to measure the overall association characteristics of the spatial distribution of traditional villages in Fujian Province, and on this basis, the local Moran’s I index is used to analyse the homogeneity and heterogeneity characteristics of each spatial unit of traditional villages in Fujian Province and its neighbouring units.

Nuclear density estimation method

The kernel density estimation method is a non-parametric approach that assumes that geographic events can occur at any spatial location, but with different probabilities of occurrence at different locations. The denser the point element, the higher the kernel density value, the higher the probability of a geographical event occurring; conversely, the lower the probability. The degree of discrete concentration of point elements is reflected by calculating the density of point elements around each output raster, which is calculated as (Hu et al. 2016):

$$f\left( s \right) = \mathop {\sum }\limits_{i = 1}^n \frac{1}{{h^2}}k\left( {\frac{{s - c_i}}{h}} \right)$$
(5)

Where: f(s) represents the kernel density function; h is the bandwidth; n indicates the number of points whose linear distance from point s is less than or equal to h; the k function indicates the distance relationship between each element point s and the core ci; the estimation of kernel density depends on the selection of the kernel function and bandwidth h. In this paper, by calculating the kernel density of various types of traditional villages at various levels in Fujian Province, the analysis reflects the gathering area and distribution characteristics of traditional villages in Fujian Province.

Superposition analysis method

Overlay analysis is an analytical method that superimposes two or more groups of elements from the same area to produce new features (Long and Zhao, 2022). In this paper, the spatial distribution maps of various types of traditional villages at various levels in Fujian Province are overlaid with digital map elevation (DEM), river system and ancient post road data respectively to form a new multi-factor layer, and the spatial distribution of traditional villages under the influence of different factors and their correlation with each factor are analysed through relevant charts.

Spatial distribution characteristics of traditional villages

Type of spatial distribution

Considering the traditional villages in Fujian Province as point shapes, the spatial distribution types of these points are studied, i.e. the spatial distribution types of traditional villages are studied. According to the nearest neighbour index (Clark and Evans, 1954) model in Table 2, it can be seen that cohesive, uniform and random distribution are the three types of spatial distribution of point elements, and the nearest neighbour distance and nearest neighbour index can be applied to measure the type of spatial distribution of point elements. According to Eq. (1), the theoretical nearest distance of traditional villages in Fujian Province is 13.33 km, the actual average nearest distance is 4.11 km, the nearest point index R = 0.31, R < 1, and Z = −53 calculated by Z test, (Z scores are used to test the statistical significance of spatial autocorrelation analysis, with a negative Z score representing agglomeration, a positive score representing dispersion, and a zero score being uncharacteristic), The significance level P value is 0.00, indicating that the probability of randomly generating agglomeration patterns is less than 1%. Therefore, the spatial distribution of traditional villages in Fujian Province is in a significant cohesive distribution. Secondly, according to equation (2), the G-value of the municipal-level geographical concentration index of traditional villages in Fujian Province is 39.48, indicating that there is a strong tendency of concentration in the distribution of traditional villages in Fujian Province. Again, according to equation (3), the municipal imbalance index S of traditional villages in Fujian Province is 0.38, indicating that the spatial distribution of traditional villages in Fujian Province is uneven at the municipal level. Finally, using the Lorenz curve (Fig. 4), it was verified that the traditional villages in Fujian Province are mainly distributed in the five cities of Ningde, Nan**, Longyan, Sanming and Fuzhou, with the number of traditional villages accounting for 80% of the total number of traditional villages in the province, which is in line with the above analysis of the concentrated but uneven distribution of traditional villages in Fujian Province. In summary, the spatial distribution of traditional villages in Fujian Province is cohesive and uneven.

Fig. 4: Lorentz curve of spatial distribution of traditional villages in Fujian Province.
figure 4

The yellow oblique line in the figure represents the average cumulative distribution of traditional villages in prefecture-level cities of Fujian Province under ideal conditions, while the blue curve represents the actual cumulative distribution of traditional villages in prefecture-level cities of Fujian Province.

Full size image

Spatial distribution pattern

On the basis of identifying traditional villages in Fujian Province as cohesive and uneven spatial types, spatial autocorrelation analysis was used to measure the spatial distribution pattern of traditional villages in Fujian Province and to examine the interdependence and interactions between regions in space. Firstly, based on the number of traditional villages distributed in each administrative unit of Fujian Province, the global Moran’s I index was used to analyse the overall characteristics of regional differences in the spatial distribution of villages in Fujian Province, and it was calculated that the global Moran’s I index of the spatial distribution of traditional villages in Fujian Province was 0.53, and Z = 7.22, p = 0.00, and the test satisfied |Z | > 1.96 and p < 0.05, The null hypothesis can be rejected. The spatial distribution of traditional villages in Fujian Province has a significant positive spatial correlation, i.e. the units with more traditional villages have more traditional villages in their surrounding units, and vice versa.

Secondly, on the basis of clarifying the relationship that the spatial distribution of traditional villages in Fujian Province is positively correlated, the local Moran’s I index was applied to further reveal the local characteristics and distribution status of spatial units homogeneous or heterogeneous with their neighbouring units, and the local spatial autocorrelation LISA map of traditional villages in Fujian Province was obtained. As can be seen from Fig. 5, the high-high clustering areas (HH) are mainly distributed in the central part of Ningde City and the contiguous area formed in the western part of Fuzhou City (including Shouning County, Zhouning County, Fuan City, Jianou City, **nan County, Jiaocheng District, Gutian County, Yan** District, Youxi County, Minqing County and Dehua County) and ** City, Longyan City, Sanming City and Fuzhou City, but in terms of distribution density, Longyan City and Sanming City fall out of the top 5 positions, because The reason for this is that their municipalities are too large, resulting in smaller distribution densities and a relatively low position in the distribution density ranking. Therefore, the following discussion is conducted to accurately analyse the spatial distribution density of traditional villages in Fujian Province.

Table 3 Statistics on the Distribution of Traditional Villages in Fujian Province by Municipality.
Full size table

Secondly, in order to visually represent the aggregation characteristics of traditional villages in Fujian Province, the Kernel Density estimation method was introduced, and the Kernel Density tool was used to carry out kernel density analysis. After repeated experiments, 10 km was chosen as the search radius to generate a kernel density distribution map of traditional villages in Fujian Province at all levels. From Fig. 6, it can be seen that traditional villages at all levels in Fujian Province exhibit a spatial distribution pattern of “large dispersion and small concentration” and the concentration tends to increase, i.e. on the whole, traditional villages at all levels are distributed in all regions of the province, but there is an obvious clustering effect in local areas. Specifically, there is one major catchment area and two sub-catchment areas for the spatial distribution of national and provincial historical and cultural villages. One major catchment area is in the southwest of Ningde City (Gutian County, **nan County, Fuan County, Zhouning County and Jiaocheng District); two sub-catchment areas are in the northwest of Longyan City (Changting County) and at the junction of western Zhangzhou and southeastern Longyan (Yongding District, Nan**g County and **he County). There are three main clusters and one sub-cluster in the spatial distribution of national and provincial traditional villages. The three main clusters are: west-central Ningde (Fuan City, Jiaocheng District, Zhouning County), south-western Fuzhou (Yongtai County), and the intersection of western Zhangzhou and southeastern Longyan (Yongding District, Nan**g County, **he County); the one sub-cluster is: the intersection of north-western Zhangzhou and southeastern Sanming City (Dehua County, Yongchun County, Dada County). There-fore, it can be concluded that the core agglomerations in central Ningde show a tendency to complement and strengthen, while the other secondary agglomerations also have a slightly complementary character. In summary, the overall spatial distribution of traditional villages in Fujian Province is characterised by one core agglomeration, two major agglomerations and two subagglomerations, with a clear core-edge distribution pattern. Nan**g County, **he County); 2 sub-catchment areas: northwestern Zhangzhou and southeastern Sanming City intersection (Dehua County, Yongchun County, Dada County), northern Sanming City (Liancheng County).

Fig. 6: Map of the nuclear density distribution of traditional villages in Fujian Province.
figure 6

(a) shows the distribution of nuclear density of historical and cultural villages at the national and provincial levels; (b) shows the distribution of nuclear density of traditional villages at the national and provincial levels; (c) shows the distribution of comprehensive nuclear density of villages, including historical and cultural villages and traditional villages.

Full size image

Spatial distribution trend

Using the standard deviation ellipse analysis tool to analyse the trend and direction of the traditional village point data set and the geometric centre location in Fujian Province, the expansion trend of traditional villages in Fujian Province can be grasped from a spatial and temporal perspective. As can be seen from Fig. 7, the ellipses of historical and cultural villages, traditional villages and integrated traditional villages all show a northeast-southwest development trend, and the ellipses of each category of traditional villages basically overlap; each ellipse direction is consistent with the direction of the

Fig. 7: Ellipse analysis of standard deviation of traditional villages in Fujian Province.
figure 7

The picture shows the distribution of mean centers and standard deviations of all kinds of traditional villages at all levels in Fujian Province, reflecting the distribution centers and trends of all kinds of traditional villages at all levels.

Full size image

provincial boundary. In terms of the centre of gravity of distribution, the centre of gravity of traditional villages in each category is located in Youxi County, northeast of Sanming City. In general, the distribution trend of traditional villages in Fujian Province is mainly from the northeast to the southwest, and Youxi County is the geometric centre of the distribution of traditional villages in Fujian Province.

In summary, traditional villages in Fujian Province are cohesive and uneven spatial types, with significant overall positive spatial correlations and only a few spatial units with negative correlations, with one core, two main and two sub-colonies and an obvious core-edge distribution, and with a northeast to southwest spatial expansion trend.

Analysis of influencing factors

Terrain and landscape factors

Physical geography is one of the most important factors influencing the distribution of villages, especially the topographical factors. In this paper, we obtained DEM data of Fujian Province from the open source geographic data cloud website, processed the images with relevant software, obtained the elevations and then coupled and overlaid them with the traditional villages in Fujian Province (Fig. 8). The results show that the traditional villages in Fujian Province are mainly distributed in the northeast, central and southwest of the province along the mountainous areas, which is consistent with the northeast to southwest expansion trend analysed above. The northeast-southwest mountain range running through the province is the main axis of village clustering, with most of the traditional villages clustered along the Daiyun Mountains in the central part of the province and the Lufeng Mountains in the northeast (Hu et al. 2021), as well as along the Tortoiseshell Mountains and the Bo** Ridge in the southwest. According to Fig. 9, there are 258 traditional villages in Fujian Province with elevations below 100 m, accounting for 16.6% of the total, and 43 villages above 1000 m, accounting for only 2.7% of the total, the vast majority of which are in these mountainous areas within the range of mountainous hilly elevations.

Fig. 8: Association map of traditional villages and elevation in Fujian Province.
figure 8

The points in the figure represent the distribution of traditional villages. The greener the lower the altitude, and the yellow the higher the altitude.

Full size image
Fig. 9: Statistical map of the elevation of traditional villages in Fujian Province.
figure 9

In the picture, the horizontal coordinate represents the altitude segment and the vertical coordinate represents the quantity, which overall reflects the distribution of traditional villages in various altitude segments in Fujian Province.

Full size image

From the above analysis, it can be seen that most of the traditional villages in Fujian Province are distributed along the central mountains from northeast to southwest and other mountainous areas. Its distribution law can be roughly as follows: at high altitude, there are more villages; At lower elevations, there are fewer villages. The natural topography of these mountainous and hilly areas is a protective barrier, which is conducive to the long and complete preservation of traditional villages. On the contrary, in the flatter coastal areas of eastern and southern Fujian Province, the distribution of traditional villages is relatively small, probably due to the fact that the original traditional villages on the plain terrain are more affected by social and economic development and external environmental influences, leading to the interruption and extinction of the villages.

River system factors

River systems are both the main source of water for residents’ production and living, as well as undertaking certain transport functions, and play an important guiding and driving role in the siting and clustering of traditional villages. Fujian Province mainly has five rivers and one stream, specifically the six river systems of Minjiang, Jiulongjiang, **jiang, Tingjiang, Saijiang and Mulanxi. In this paper, the lo-cation overlay and buffer analysis of the river network and traditional villages in Fujian Province (Fig. 10) are used to reflect the spatial relationship between the spatial distribution of traditional villages and river system factors. From Fig. 8 and buffer statistics, it can be seen that: 509 traditional villages in Fujian Province are located within 500 m of major water systems, accounting for 31.7% of all villages; 812 traditional villages are located within 1 km, accounting for 50.6% of all villages. Most of the traditional villages are located in mountainous areas near rivers, and only a few of them are far away from big rivers and streams. The reason for this is that the inhabitants of villages located in mountainous areas with high altitudes mostly use the springs in the mountains, and the other is that the inhabitants of villages located near the sea mostly dig wells to get water or process seawater, so they can still solve problems such as using water for eating while being far away from big rivers and streams.

Fig. 10: Association map of traditional villages and river systems in Fujian Province.
figure 10

The blue lines in the figure represent the buffer areas of Fujian river system, which are buffered 500 meters, 1 km, 2 km, 3 km and 5 km according to different degrees of blue. Combined with the distribution of traditional villages, the overall correlation between traditional villages and river system is reflected.

Full size image

In general, the river system in Fujian Province is strongly correlated with the distribution of traditional villages, and the proximity to rivers is one of the leading environmental factors in the location of traditional villages. Traditional villages that are located along rivers have better agricultural production conditions and are able to obtain a better production and living environment, which also improves the chances of the development and continuation of traditional villages.

Ancient and modern transportation factors

The transportation factor is an important indicator of regional development and has an important influence in the formation and conservation and development of traditional villages. Firstly, analysing the correlation between ancient post roads and traditional villages is conducive to gras** the distribution characteristics of traditional villages from the perspective of their formation and development. In this paper, by consulting historical maps and local chronicles of Fujian Province, the ancient post roads in Fujian Province were compiled and mapped, and the buffer analysis was conducted using 1 km, 5 km and 10 km distance ranges to obtain the correlation degree map between ancient post roads and traditional villages in Fujian Province (Fig. 11) (Fujian Provincial Local History Editorial Committee, 2004). The results show that there are 33 traditional villages located within 1 km of ancient post roads in Fujian Province, accounting for 2.05% of all traditional villages in Fujian Province; 162 traditional villages within 5 km, accounting for 10.09% of all traditional villages in Fujian Province; and 330 traditional villages within 10 km, ac-counting for 20.55% of all traditional villages in Fujian Province. It can be seen that, on the one hand, the ancient post roads are somewhat related to the traditional villages under the existing system, but not as much as the influence of river systems and topography on the location and layout of villages; on the other hand, the overall shape of the ancient post roads is in the shape of an unsealed “8”, laying a certain foundation for the formation of the modern transportation system in Fujian Province, which is the historical skeleton of the modern transportation system in Fujian Province.

Fig. 11: Association diagram of traditional villages and ancient post roads in Fujian Province.
figure 11

The red lines in the figure represent the ancient post roads in Fujian Province, the white area represents the 1 km buffer zone, the light purple area represents the 5 km buffer zone, and the dark purple area represents the 10 km buffer zone, which overall reflects the distribution correlation between the ancient post roads and traditional villages in Fujian Province.

Full size image

Secondly, the data related to the number of road miles of each region in Fujian Province in 2020 were obtained from 《Fujian Provincial Statistical Yearbook-2021》 (Fujian Provincial Bureau of Statistics, 2021), on which the data were collated and the road density of each region was calculated to obtain Table 4, which was analysed in conjunction with Table 2 above. The results show that the road densities of Ningde, Longyan, Sanming and Nan** are small, ranking in the bottom four, while the distribution of the number of traditional villages in these four regions ranks in the top four; the distribution of the number of traditional villages is smaller in areas such as Quanzhou, Putian and **) being in the bottom four in terms of disposable income and household deposits of farmers, and the level of urbanization and gross regional product also being at a lower level, lagging behind the provincial average; the areas with less traditional villages (i.e. ** traditional villages in China (Xu et al. 2023). Under the premise of green and low-carbon development, we should comprehensively and harmoniously protect the traditional village clusters affected by different physical geography and human and social factors. Focus on overall consideration of the organic connection between individual traditional villages and provinces, districts and village groups, and explore the establishment of control and protection mechanisms at all levels by using new technologies (Yang et al. 2021) and new methods (Li et al. 2023).

This paper studies the spatial distribution characteristics of various types of traditional villages at all levels in Fujian Province from the macro level, and conducts a spatial analysis of their influencing factors based on mathematical statistics model and geographic information system. Under the constant changes of various environmental factors today, the positive and negative correlations between the distribution of traditional villages and different factors such as topography, river system, traffic elements and social economy are identified. The invariance of the distribution law of traditional villages is discussed in order to provide basic support for the protection and utilization of traditional villages and the planning of village agglomeration area in the future.