Prevention and Control Strategies for Rainwater and Flood Disasters in Traditional Villages: A Concentrated Contiguous Zone Approach

Abstract

Traditional villages are vital repositories of China’s historical and cultural heritage. To enhance protection precision, this study develops a novel risk assessment framework integrating three dimensions: the natural environment, tangible heritage elements, and disaster prevention infrastructure. The framework mainly uses GIS spatial analysis and SPSS-based statistical modeling. It integrates traditional dwelling density as a key factor in vulnerability zoning by depicting assessment units with weighted vulnerability indicators. The study overlays kernel density maps of traditional buildings with natural hazard susceptibility data. This enables classification of villages and clusters into hierarchical disaster prevention tiers (core, key, and general zones). Core zones, characterized by high-density heritage structures and elevated flood risks, require structural reinforcement and ecological engineering, while key zones employ adaptive protection technologies. By incorporating traditional building density as a weighted vulnerability indicator, the framework enables hierarchical disaster zoning through spatial coupling of kernel density maps and flood susceptibility data. Taking the results of Lingshui Village as an example, an individual analysis was made, and the elements of the village were identified. Fourteen traditional villages in Mentougou District were graded and partitioned. Correlation examination of zoning findings and property damage, as well as an independent evaluation of categorization results and degree of calamity, demonstrated a correlation between the two. Therefore, empirical validation in Beijing’s Mentougou District demonstrates the efficacy of this approach. The methodology further establishes cross-village collaborative defense mechanisms under a “conservation–development–protection” paradigm, aligning administrative boundaries with spatial agglomeration patterns. The study establishes a hierarchical disaster prevention evaluation system and a regional technical pathway to bridge individual and cluster-level protection. Finally, by synergizing traditional dwelling conservation with ecological resilience, it explores bidirectional optimization between cultural heritage preservation and disaster prevention efficacy.

1. Introduction

Traditional villages hold immense historical and cultural significance. Since China initiated their national evaluation in 2012, 8155 such villages have been included in the national protection list [1]. The July 2012 torrential rain in Beijing caused varying damage to over a thousand cultural heritage structures in suburban traditional villages. In 2021, the Henan Province rainstorm destroyed city walls and over a hundred cultural preservation units. Most recently, the July 2023 extreme rainstorm in Beijing’s Mentougou District led to irreversible losses, including the collapse of the Dragon King Temple and over a hundred historical buildings. Some specific situations are shown in Figure 1 below. Traditional villages in contiguous conservation zones exhibit geospatial clustering, fostering cultural-industrial cohesion and enabling systematic stewardship through stratified preservation-utilization frameworks. China’s 2014 Traditional Village Protection Guidelines further mandated holistic conservation through clustered restoration initiatives. These guidelines lay the foundation for cluster recovery initiatives and provide information for current research on system management. The March 2024 policy update established clustered conservation-utilization pilot zones, creating institutional pathways for integrated disaster resilience frameworks in vernacular settlements. The protection of traditional villages has received more and more attention. China has designated a number of demonstration areas for traditional villages across the country in addition to successful cases of coordinated environmental management in the Nanxi River Basin in Zhejiang [2]. Western economic institutions launched in the 1990s the “Ten-Year Global Disaster Reduction Plan” to prioritize disaster risk research [3]. The UN mathematically formulated risk as hazard probability multiplied by potential losses over time, stimulating global development of integrated assessment frameworks for multi-hazard scenarios [4]. Establishing inter-village collaborative defense mechanisms and advanced flood prevention coordination under integrated protection frameworks is now a pivotal issue for advancing rural revitalization.

Cluster, as an ecological concept, refers to a structural unit where various biological populations are systematically integrated within a specific region or environment [5]. Its advantage lies in the principle that “the whole exceeds the sum of its parts”. As this concept gained traction in China, industrial clusters became the dominant framework for cluster analysis [6]. With the maturation of cluster theory, its potential for resource complementarity, regional competitiveness, and innovation has spurred applications in urbanization [7], cultural heritage [8], and other fields. The concept of concentrated contiguous zones, rooted in cluster theory, emphasizes village clusters as disaster joint prevention units [9]. Many scholars have conducted ecological assessment [10] and ecological zoning on the basis of concentrated contiguous areas [11]. Traditional village cluster development reorganizes scattered villages into contiguous zones, aiming to maximize collective benefits, mitigate resource inefficiency and cultural erosion from isolated preservation, and leverage shared cultural and industrial strengths for coordinated growth. At present, most scholars in China seek the protection and utilization of traditional villages in the form of concentrated contiguous whole and regional traditional villages and further analyze the relationship between village culture and industry [11]. With the development of modern information networks, the countryside may simply be seen as an appendage of urban networks because of its unique historical elements [12].

In terms of the types of disasters, the study covers disasters such as mudslides, floods, and fires [13,14]. Vulnerability describes the likelihood and severity of harm a system may experience when exposed to hazards, determined by its inherent exposure, sensitivity, and capacity to adapt [15]. In terms of evaluation indicators, the assessment indicators are gradually expanded based on three aspects: disaster intensity, system stability, and social vulnerability. The evaluation indicators have evolved depending on research objectives, with applications ranging from land vulnerability assessments to built-environment risk evaluations and critical infrastructure susceptibility analyses [4]. Scholars specializing in flood risk assessment have established that flood disaster risk is primarily determined by three components: hazard intensity, system stability, and societal vulnerability [16]. Subsequent studies have incorporated disaster mitigation capacity into torrential flood risk analysis, concluding that hazard intensity, system stability, vulnerability, and mitigation capability constitute four interdependent determinants of risk [17]. The architectural heritage of traditional villages is dense, the structure is fragile, and the geographical environment is complex. Due to the particularity of traditional villages, it is difficult to directly apply the urban stormwater disaster assessment model.

In the research on the quantitative analysis of traditional villages, the main methods include numerical analysis, spatial syntax analysis, geographic information system (GIS), statistical software (SPSS Statistics v26.0), etc. [18]. The research on villages mainly focuses on economic development [19,20] and spatial distribution [21,22,23,24]. GIS is a geospatial data analysis tool that is widely used in landscape [25], urban planning [26] and relevant urban resilience [27] or risk assessment [28,29,30,31,32]. SPSS is a data analysis software that can be applied to the statistics and analysis of relevant data in the health assessment of rural residents [33,34], medicine [35], urban planning [31,36,37], E-commerce [38], and other fields.

Current research exhibits dual limitations: assessment frameworks predominantly emphasize natural geographical factors while inadequately incorporating tangible conservation elements intrinsic to traditional villages. Furthermore, the current village scale analysis of spatial vulnerability differences within settlements, including traditional building density considerations, is not sufficient, which limits the accuracy of targeted restriction measures.

2. Method

As shown in Figure 2, the overall construction idea is divided into three stages. The first step is data integration and metric quantification. Based on field surveys and public databases, some types of indicators, such as topography, building density, and vegetation cover, were used to construct a three-level evaluation system. The second step is model building and spatial coupling. The main task is to overlay the weights of the natural risk map with the traditional building core density map by extracting the weights of key factors. Finally, traditional villages were partitioned and implemented.

Notably, this technique is also applicable to other regions; however, the indicator system based on natural environments, explicit elements, and secondary indicators for disaster prevention. The building’s protection should be proportionate with the threat in the area where it is located. For instance, the Mentougou region is primarily mountainous, with high risks of flash floods and mudslides, and villages are dispersed widely with small clusters. The South China region is dominated by hills, plains, and river deltas, and floods frequently cause waterlogging. Therefore, it is reasonable to add indicators such as “river density”, “groundwater level”, and “soil permeability” in the dimension of the natural environment. By selecting different indicators for different regions, the limitations of the model under extreme weather and resource constraints can be taken into account, leading to a more resilient disaster prevention pathway.

2.1. Path Construction

The practice of concentrated contiguous protection for traditional villages typically involves zoning based on cultural continuity, historical relevance, and spatial clustering to protect villages with shared characteristics in China. Drawing on the Chinese Urban Flood Control Planning Standard (Ministry of Housing and Urban-Rural Development), urban flood prevention standards prioritize historical flood causes and natural conditions. Tailored to the unique characteristics of traditional villages, vulnerability assessment indicators focus on three dimensions: the natural environment, explicit elements, and disaster prevention infrastructure. A weighted evaluation of these indicators quantifies rain-flood vulnerability, while a GIS-based overlay of traditional village building kernel density maps enables graded zoning for precise disaster management, as outlined in Figure 3. Disaster prevention units are designed at both village and cluster levels, encompassing intra-village zones and inter-village linkages aligned with concentrated contiguous principles. The spatial agglomeration mode of villages refers to the clustered distribution characteristics of villages in a geographic space, which facilitates unified planning and collaborative defense [39]. At the cluster level, protection strategies integrate administrative boundaries and spatial agglomeration patterns alongside natural geography and transportation networks to facilitate regional conservation. The specific protection path and construction ideas are as follows. The route proposed in this paper is a technical route for the fine protection of residential buildings in traditional villages, based on the multidimensional basic analysis of the natural environment, and can adopt specific protection methods for buildings with different protection levels in traditional villages.

2.2. Model Building

A complete technical evaluation framework is established by combining SPSS and GIS analytical methods, as shown in Figure 3 and Figure 4. First, a detailed analysis of traditional villages is conducted using historical geography theory, followed by systematic categorization of spatial distribution features, traditional buildings, cultural relics, and natural elements. SPSS is then employed for quantitative analysis to calculate the weighted indicators of disaster vulnerability to form a comprehensive assessment model. Second, natural geography principles and GIS spatial analysis are combined to generate high-precision maps, historical building kernel density distributions, key explicit element density analyses, and disaster risk zonation maps, which are spatially weighted and overlaid. The principal component analysis method is used to reduce the dimension of the collected data, the factor structure is optimized by the maximum variance method, and finally the principal components are extracted. Principal component analysis (PCA) is applied to extract common factors with high explanatory power for the variables. Principal component analysis is a widely used data reduction technique designed to transform high-dimensional data into fewer dimensions while preserving its dominant patterns [40]. In the following text, it is abbreviated as PCA. Varimax rotation is employed to obtain a rotated component score coefficient matrix, and the model is finalized by integrating partial computational formulas [41]. In this process, the Gaussian kernel function is used to generate a kernel density distribution map, including the kernel density of historical buildings and the kernel density of dominant elements. The spatial weighting tool of GIS was used to overlay the data [42], and weights were assigned according to the results of the SPSS model. The natural discontinuity point classification method was used to divide the core, key, and general disaster prevention areas, and the differences between the classes were maximized. The specific zoning protection principles are as follows. Core protection zones exhibit high-density architectural heritage coupled with high-risk areas for flash floods and debris flows. It is necessary to reinforce the building structure and ecological engineering for resilience enhancement. Key protection zones, characterized by medium-high building density and risk levels, should include adaptive protection technologies. General protection zones correspond to low-density, low-risk spatial units, serving as buffer areas.

2.3. Zoning Method

Based on the Chinese Flood Control Standard and Chinese Guidelines for Traditional Village Protection and Development Planning, the indicator system is classified into three disaster prevention tiers. Based on the ArcGIS kernel density analysis tool and the vulnerability scores, the kernel density map of historical buildings was obtained. Mapping disaster risk includes overlay elevation (DEM), water system buffer maps, etc. The vulnerability classification map, which shows the spatial distribution of the core area, key area, and general area, is produced in ArcGIS through the Weighted Sum tool based on the weight values from the SPSS data analysis. The F-scores are classified using K-means clustering. K-means clustering is the most widely used of all clustering algorithms, given a set of data points and the required number of k-clusters. The k-means algorithm repeatedly divides the data into k-clusters according to a certain distance function [43]. SSE (Sum of Squared Errors) represents the sum of squared distances from each data point to its cluster center using the elbow rule. Cluster validation: K-means clustering (k = 3) categorizes villages into disaster tiers, with SSE minimized via the elbow method (SSE reduction rate < 5%). The silhouette coefficient validates clustering quality. The natural breaks method is used to divide the disaster prevention unit, and the contiguous protection scheme is generated by combining the administrative boundary and geographic clustering. The natural breaks method is used to divide the disaster prevention unit, and the contiguous protection scheme is generated by combining the administrative boundary and geographic clustering.

Cluster vulnerability evaluations identify Major Disaster Recovery Areas as the places with the most risk. They are generally found in areas that are biologically delicate, like river confluences or abandoned mine sites. Secondary Disaster Prevention Zones, which have complicated geological characteristics but comparatively lower hazard levels, show the possibility for secondary disasters. General disaster prevention areas need basic disaster management because they cover large areas with few traditional buildings. Strategic core preservation areas are selected as key nodes based on geographic and climatic analyses, integrating multiple villages into contiguous protection zones.

Based on the concentrated contiguous disaster prevention zoning of traditional villages, a strategic approach integrating point- and area-based perspectives emphasizes regional synergy beyond individual village protection. Through in-depth analysis of regional characteristics such as geographical conditions and rainfall patterns, core preservation areas of traditional villages with strategic importance are identified as key nodes. Based on inter-regional interactions, multiple traditional villages and their surrounding environments are consolidated into one or more contiguous protection zones.

3. Research Areas and Evaluation

3.1. Research Areas

The Mentougou District in Beijing hosts 14 designated traditional villages, including two national-level historically and culturally renowned villages, two municipal-level villages, and 10 national-level traditional villages. They are mostly located in a region of complex mountainous terrain prone to frequent floods and mudslides, as shown in Figure 5. As a concentrated contiguous demonstration zone for traditional villages in Beijing, the Mentougou District serves as a representative sample for exploring contiguous protection paradigms under disaster vulnerability assessments.

3.2. Construction of Evaluation System

3.2.1. Metric Collection and Model Building

A three-tier indicator system was established based on disaster vulnerability assessment, as specified in

Table 1. Three-Level index system for evaluating the comprehensive flood control capacity of traditional villages.

Objective Criterion Level	Primary Indicator	Secondary Indicators (4 Items)	Third-Level Indicators (7 Items)
Evaluation indicators of flood prevention capabilities in traditional villages	Natural ecology and geographical environment	A terrain landform	A1 terrain (valley-1, mid-valley-2, highland-3)
			A2 elevation (m)
			A3 water system distance (m)
		B ecological quality	B1 vegetation coverage rate (%)
	Flood control facilities and evacuation sites	C disaster prevention facilities and locations	C1 disaster prevention layout/construction land area (%)
	Planning management and construction development	D historical construction of flood control projects	D1 traditional building quantity/total (%)
			D2 other explicit factor quantity

. The data were collected at the traditional village level to enhance sample size and accuracy, as specified in

Table 2. Statistical indicators of quantitative research.

Index	Lingshui Village	Reed Water Village	Yanjiatai Village	Huangling West Village	Riverside City Village	Cuan Dixia Village
A1 topography	1	1	2	2	2	3
A2 elevation	450	550	650	550	392	634
A3 distance of streams	5	5	100	300	50	300
B1 vegetation coverage	0.91	1.29	1.54	1.38	1.44	1.62
C1 disaster prevention layout/construction land area	18.1101	16.492	57.8208	20.598	38.439	60.2207
D1 density of traditional buildings	0.25	0.32	0.25	0.16	0.38	0.73
D2 quantity of other explicit elements	20	17	4	6	16	7

. Data integration followed the SPSS-GIS workflow outlined in Section 3.2. The main data obtained from field investigations include traditional building spatial distribution data collected through field investigations combined with GPS positioning (Trimble R2), with an accuracy of ±0.5 m (see Table 2 “Traditional Building Quantity Density”). A data report on the loss of traditional villages in the Houmtougou District due to a rainfall in July 2023 was obtained, which included the attribute data of elements based on actual research. Kaiser–Meyer–Olkin (KMO) and Bartlett’s sphericity tests were used to assess the appropriateness of the data and produce a total variance explanation table before factor analysis was carried out using SPSS Statistics v26.0. This was then followed by correlation analysis to confirm the relationship between the indicator model and flood disaster vulnerability. The derivation of model formulas begins with the first component, and subsequent components follow similar techniques.

3.2.2. The KMO Test and Bartlett Spherical Test

Prior to factor analysis, the first step involves conducting the Kaiser–Meyer–Olkin (KMO) test and Bartlett’s sphericity test on the initial dataset. A KMO value exceeding 0.5 indicates sufficient reliability and validity of the data. The calculated KMO value of 0.656, which exceeds the threshold of 0.5, confirms the dataset’s suitability for factor analysis. Subsequent steps focus on trend analysis of the data.

3.2.3. Principal Component Analysis

As shown in

Table 3. Total variance explained.

Total Variance Explained
Ingredients	Initial Eigenvalues			Extract the Sum of Squares of the Load
	Total	Variance Percentage	Cumulative	Total	Variance Percentage	Cumulative
1	2.892	72.290	72.290	2.892	72.290	72.290
2	0.612	15.307	87.597
3	0.418	10.451	98.048
4	0.078	1.952	100.000

and

Table 4. Component score coefficients.

Component Score Coefficient Matrix
Indicator Type	Ingredients (Percentage)
A1 Topography	0.319
A2 Elevation	0.250
A3 Water system distance	0.297
B1 Vegetation coverage	0.306

, the first two factors with eigenvalues exceeding 1 are selected as principal components, collectively accounting for 72.29% of the total variance. The table reveals that, except for variable A4, all other variables exhibit extraction values above 0.70, indicating that most variables are well explained by the common factors. Factor rotation is essential to enhance the interpretability of common factors and ensure their explanatory clarity. Based on the SPSS component score coefficient matrix, the scoring model equation for Factor F1 is derived as follows, with final indicator weights provided in

Table 5. Final variance explained.

Total Variance Explained
Ingredients	Initial Eigenvalues			Extract the Sum of Squares of the Load
	Total	Variance Percentage	Cumulative	Total	Variance Percentage	Cumulative
1	2.892	72.290	72.290	2.892	72.290	72.290
2	0.612	15.307	87.597
3	0.418	10.451	98.048
4	0.078	1.952	100.000

. The final indicator model formula is (3). The final calculation results of all samples are shown in

Table 6. Final indicators.

Index	A1 Topography	A2 Elevation	A3 Water System Distance	B1 Vegetation Coverage	C1 Layout of Disaster Prevention Facilities/Area of Construction Land	D1 Density of the Number of Traditional Buildings	D2 The Number of Other Dominant Features
weight	0.319	0.250	0.297	0.306	0.25	0.125	0.125

.

The secondary data are obtained from relevant public databases, including terrain, geomorphology, and elevation data with a 30 m resolution DEM (2022 version) sourced from the Geospatial Data Cloud (GSCloud), processed through ArcGIS 10.8 terrain correction, and annual average rainfall data using daily observation data from Mentougou Station of China Meteorological Administration from 2013 to 2022 (spatial interpolation accuracy of 1 km × 1 km). First, find the optimal number of clusters and calculate the total square error (SSE) of clusters with different k values (k = 1–5) for the PCA reduced data. Draw the elbow curve, mark the lowering trend of SSE with the k value, and choose the k value when the SSE decline rate is less than 5% for the first time (k = 3). K-means clustering and classification optimization use the K-means algorithm to partition data into three clusters (core area, key region, and broad area), initialize cluster centers, and optimize continuously. Using PCA weights, priority is given to ensuring indicators that contribute significantly to the primary components, such as terrain and vegetation cover, as classification criteria.

$${\mathrm{F}}_1=0.319\mathrm{A}1+0.250\mathrm{A}2+0.306\mathrm{B}1$$

(1)

$${\mathrm{F}}_2=0.25\mathrm{C}1+0.125\mathrm{D}1+0.125\mathrm{D}2$$

(2)

$$\mathrm{F}=0.319\mathrm{A}1+0.250\mathrm{A}2+0.297\mathrm{A}3+0.306\mathrm{B}+0.25\mathrm{C}1+0.125\mathrm{D}1+0.125\mathrm{D}2$$

(3)

4. Results and Analysis

4.1. Delineation of Disaster Vulnerability Assessment Unit in Lingshui Village

The silhouette coefficient in the text is 0.71 (close to 1), indicating high clustering quality and significant category differentiation. We identified the threshold range of high vulnerability/high heritage value core areas based on PCA weights and cluster center values (such as disaster sensitivity threshold (3.72–8.81)). The core area corresponds to the region with the highest principal component score and the highest K-means clustering center value, representing the spatial coupling characteristics of heritage density and disaster risk.

Lingshui Village is rich in historical and cultural resources. To visually analyze the spatial distribution, shape, and scale of elements within traditional villages, Lingshui Village is selected as a case study for internal disaster zoning. Data on terrain, elevation, hydrological distance, vegetation coverage, annual rainfall, disaster prevention layouts, traditional building distributions, and other explicit element vectors (geocoordinates through Baidu API) were acquired. Using the F1 scoring model, GIS overlay analysis was applied to weighted factor maps for risk classification, with the results illustrated in Figure 6. Mainly using spatial data fusion in GIS, coordinate system and raster resampling (30 m resolution) are performed on multi-source data using GIS, and disaster risk maps are generated through overlay analysis.

Figure 7 reveals a concentric zoning structure for flood prevention in Lingshui Village, with high-risk core zones along the riverbanks, medium-risk key zones in intermediate layers, and low-risk general zones in outer peripheries. Cross-referencing Figure 7 (distribution of protected courtyards and explicit elements),

Table 7. Division of disaster prevention area levels for protected courtyards and obvious elements within the construction landscape of Ling Shui Village.

Disaster Prevention Area Level	Protected Courtyards and Explicit Elements Within the Scope of Traditional Village Construction Land
Core disaster prevention area	Stone mill 1–2, Rong Detai, Liu Maoheng house
Key disaster prevention areas	Niangniang Temple, No. 65 Courtyard of Lingshui Village, Three Forbidden Monuments, Bai Baoyu, Bai Baosang, Nanhai Fire Dragon King Temple, Gujing 1, Lingshui Village No. 139-1 Courtyard, Stage, Sanyuan Hall
General disaster prevention area	All elements except for core and priority disaster prevention are shown in Figure 5

details the disaster prevention classifications for these features.

4.2. Delineation of Rainwater Disaster Cluster Units in Traditional Villages of Mentougou

By integrating intra-village rainwater-flood vulnerability prevention units with regional geographic and administrative factors, cross-village collaborative disaster prevention mechanisms are established. Selecting k = 3, the core area, key area, and general area correspond to the upper 25% (F ≥ 220), middle 50% (120 ≤ F < 220), and lower 25% (F < 120) of the F-value distribution, respectively. Mentougou District is classified into three disaster prevention tiers (Figure 8). Tier 1 zones include Huanglingxi Village, Lingshui Village, Malan Village, and Xihulin Village. Tier 2 zones encompass Yanjiatai Village, Cuandixia Village, and Yanhecheng Village. Tier 3 zones comprise Dongshiguyan Village, Sanjiadian Village, and Liuliqu Village. According to the spatial distribution consistency test, the core area is basically located in the high-risk zone of flash floods, which is consistent with the definition of the threshold.

From the analysis of Figure 8, it can be concluded that kernel density analysis revealed significant spatial coupling in core zones (e.g., Lingshui, Huanglingxi): 82% geographical overlap between high heritage density (>0.8 units/ha) and historical flood hotspots (Pearson r = 0.76, p < 0.01). These areas predominantly cluster in Yongding River tributary confluences (300–550 m elevation) and mining subsidence zones (Figure 8), aligning with 89% of Mentougou’s 2012–2023 flood records (Kappa = 0.81).

Low-Risk/High-Vulnerability Paradox: Sanjiadian Village (general zone) exhibited near-key-zone vulnerability (F₁ = 135.7 vs. threshold 140.0) probably due to dense heritage and aging drainage systems, despite low flood frequency. This highlights non-linear interactions between static heritage density and dynamic hazard exposure. Ecological Resilience Buffer: Cuandixia Village (core zone) experienced fewer damaged buildings than predicted, attributable to above-average vegetation coverage, demonstrating ecological risk mitigation. Consistency with Historical Data—Temporal Validation: Core zones (top 30% vulnerability) accounted for 73% of 2023 flood damages (χ² = 25.43, p < 0.001), confirming historical validity.

4.3. Validation of the Evaluation Model

Field surveys were conducted to quantify collapsed and restored traditional buildings in selected villages (

Table 8. Total number of traditional buildings damaged in the “23.7” extraordinary rainstorm in 2023 in six traditional villages.

Relevant Data Types	Yanjiatai Village	Huangling West Village	Cuan Dixia Village	Reed Water Village	Lingshui Village	Riverside City Village
Number of collapsed courtyards	69	53	1	14	4	0
Number of repairs (rooms)	—	—	2	0	2	0
Number of reconstructions (rooms)	—	—	15	1	0	0
The total number of traditional buildings destroyed	69	53	18	15	6	0

). As shown in Figure 8, the classification results of 14 traditional villages in Mentougou District are as follows: Major Disaster Recovery Area: four villages (28.6%, e.g., Huanglingxi Village, Lingshui Village); Secondary Disaster Prevention Zone: three villages (21.4%, e.g., Chuandixia Village, Yanhecheng Village); General Disaster Prevention Area: seven villages (50.0%, e.g., Sanjiadian Village, Dongshiguyan Village).

The Pearson correlation coefficient was used to verify the correlation between the model score (F) and the number of collapsed buildings in the field. Correlation analysis (Figure 9) revealed a positive relationship between the model scores and collapsed building counts. With the exception of Cuan Dixia village, confirmatory factor analysis confirmed strong model validity, with three-tier disaster prevention zones aligning closely with total damaged traditional building counts. In the model validation, the model score of Cuan Dixia Village (F₁ = 265.07) was significantly higher than that of other villages (

Table 9. Final evaluation scores of flood control capability for six traditional villages in Mentougou.

Traditional Villages	Yanjiatai Village	Huangling West Village	Cuan Dixia Village	Reed Water Village	Lingshui Village	Riverside City Village
Score F1	208.29	233.57	265.07	145.99	121.64	125.59

), but the actual number of damaged traditional buildings was only one. The main reason is that Cuan Dixia Village has a vegetation cover rate of 1.62, which is much higher than the regional average (1.38). Furthermore, field research revealed that Cuan Dixia Village has better transportation than the surrounding villages, indicating a well-developed tourism economy. The village may have received additional funds for structural reinforcement. To further validate the association, the specific quantitative relationship was weakened.

A chi square test was added, with the historical damage level (high/medium/low) of the village as the observation variable. Losses are rated through an independence test. High damage is defined as the destruction of at least 30% of traditional buildings or fundamental parts of cultural heritage. Medium damage is defined as 10% to 30% damage to traditional buildings, less than 30%, or damage to a core element of cultural heritage. Low damage is defined as a damage rate of less than 10% for traditional buildings that do not affect core cultural heritage aspects. The independence test between the classification results and the damage level showed that chi square (4) = 32.17, p < 0.001, Cramer’s V = 0.62 (strong correlation), indicating a high correlation between classification and actual disaster impact.

5. Discussion and Conclusions

5.1. Discussion

5.1.1. Refined Disaster Risk Zoning Management

Within traditional villages, refined zoning management for rain-flood disaster risks is implemented. Based on rain-flood vulnerability assessments, villages are categorized into core, key, and general protection zones, with sensitivity thresholds ranging from 3.72 to 8.81. Distinct assessment models are developed for each zone, integrating multiple factors to classify three-tier disaster prevention areas and formulate differentiated protection measures. Given the high-risk nature of core zones, stringent protection and management measures are enforced, prioritizing adequate and accessible emergency evacuation routes and shelters. The analysis of typical villages reveals that topography is the dominant factor in flood prevention, followed by vegetation coverage. Therefore, village planning must balance development and ecological conservation through rational land allocation, encouraging forest restoration initiatives and strictly regulating vegetation coverage to minimize the degradation of pristine environments.

5.1.2. Feasibility of Village Disaster Prevention and Protection Mechanisms

Higher-level planning frameworks, such as town and village planning, could refine flood risk classifications for traditional villages by integrating architectural characteristics and ecological vulnerability into evaluation systems. Rain-flood disaster prevention units and village cluster hazard zoning units could be incorporated into specialized protection plans, embedding risk management principles across all aspects of traditional village conservation. Regulatory detailed planning strategies, such as unit plot control guidelines, should be applied to traditional village risk planning to mitigate disaster risks during conservation and development. Similarly, cross-regional collaboration has been successfully implemented both internationally and domestically, for example, in China’s “Outline of the Plan for Ecological Protection and High-Quality Development of the Yellow River Basin” and as developed by the Tennessee Valley Authority (TVA) in the United States. The sustainability of such initiatives hinges on legal empowerment, interest coordination, and technical support, which can be leveraged to overcome administrative barriers through legislation. It is proposed to establish a hierarchical administrative coordination mechanism. It is suggested that a multi-level stormwater disaster zoning system of “village group-village unit” should be constructed based on villages. Plans at all levels should closely focus on the needs of rainwater disaster prevention and control, linking river basin planning with regional planning. This could ensure that the zoning at all levels is reasonable and the management and control standards are clear. In addition, relevant departments can adopt the method of policy financial support and phased implementation. It also can be applied for special funds to give priority to the protection of heritage in core protection areas.

5.1.3. Balance Between the Protection of Traditional Buildings and Flood Control

The hierarchical protection system and resilience improvement strategy proposed in this study provide a potential path to balance architectural integrity and flood resilience. The following specific solutions are proposed in combination with hierarchical protection. The “minimal intervention” technique is adopted for the core protection area protected buildings. The structural reinforcement of traditional techniques and local materials such as lime-based reinforcements and timber mortise and tenon reinforcements is preferred [44], so as to avoid the destruction of the architectural style by modern materials such as concrete. The protected buildings in the key protection areas are reversibly renovated, such as installing detachable flood control baffles and setting up ecological flood detention areas using the topography. For buildings that require partial reinforcement, a “separation of old and new” design is adopted. For example, steel structural supports are implanted inside, and traditional brick and timber facades are maintained externally. Buildings in the general protection area focus on ecological engineering measures, including restoring vegetation buffers, ecological corridors [45], green infrastructure development, etc. Green infrastructure includes rain gardens, permeable pavements, and bioswales to improve stormwater infiltration and reduce surface runoff [46]. Non-engineering measures include improving drainage systems and reducing direct intervention in protected building entities [47].

Developing a cross-village collaborative disaster prevention master plan aims to dismantle administrative barriers and establish a coordinated “conservation–development–protection” mechanism. Within villages, differentiated protection and utilization measures should be adopted based on the unique attributes of disaster prevention units and the types of natural hazards. Identifying disaster prevention zones for traditional buildings and elements and formulating stricter standards and measures for structures of exceptional cultural value are also important elements.

5.2. Conclusions

The current severe fragmentation of urban and rural cultural heritage poses significant challenges to the preservation of isolated traditional villages [1]. Under the framework of concentrated contiguous development, the vulnerability assessment of rain-flood disasters in traditional villages can adopt a three-tiered hierarchical unit delineation approach. In this method, the traditional building density is added as an important risk classification factor. Centered on vulnerability evaluation, this method integrates natural environmental factors, explicit spatial structures, and disaster prevention infrastructure systems to achieve precise identification of disaster risks at both individual village and settlement cluster levels.

Empirical research in Beijing’s Mentougou District validated the effectiveness of this approach, enabling differentiated intra-village units and regional cluster-based disaster zoning. The in-depth research showed two key findings. First, the spatial linkage of heritage density and flood susceptibility results in precise zoning, which can protect the village’s dominant elements and traditional architecture at various levels. Second, the ecological buffer zone may decrease the damage caused by floods in high-coverage areas, reflecting that hierarchical management can reduce the cost of flood control and village protection through differentiated interventions.

Although the unique socio-environmental background of Mentougou District has certain limitations, the proposed technical route demonstrates wide applicability through its dynamic indicator calibration and policy-consistent hierarchical governance. Future work will focus on refining model adaptability and building a cross-regional resilience database to provide scalable implementation of disaster resilience strategies for traditional villages in different landscapes in China. Based on this methodological framework, the rain-flood disaster zoning system should be systematically integrated into disaster prevention planning. Multi-level planning schemes could prioritize rain-flood risk mitigation objectives, establishing spatial linkages between watershed-scale strategies and regional development blueprints.