Socio-Ecological Coupling and Multifunctional Spatial Differentiation in Watershed Rural Systems: Toward Coordinated Development

Abstract

Socio-ecological systems in basin regions characterized by diverse cultural traditions and hierarchical village spatial structure are undergoing profound transformation driven by multifunctional demands and spatial restructuring. This study develops an analytical framework encompassing economic production, socio-cultural functions, and ecological potential to examine the spatial differentiation and socio-ecological coupling mechanisms within the Yilong Lake Basin, Yunnan Province. Through the entropy weighting method and a coupling coordination model, the framework evaluates the “lake–mountain–village” gradient of spatial differentiation. The results indicate that: (1) the overall coordination level of multifunctional systems in the region remains relatively low, exhibiting a decreasing trend from lakeshore to the mountain periphery; (2) village-level dependencies of spatial functions can be summarized into three coupling categories—associated with institutional embedding, self-organization, and value mismatch—revealing distinct socio-ecological interaction patterns; and (3) three coupling categories correspond to three differentiated governance pathways, namely coupling optimization, functional transition, and conflict mitigation. The study advances theoretical and methodological insights into the spatial differentiation and evolution of complex village systems, highlighting the nonlinear coexistence of interdependence and constraint among economic, social, and ecological functions. It further provides practical guidance for coordinated governance and sustainable spatial planning in similar rural and basin environments worldwide.

1. Introduction

Since the mid-20th century, traditional villages have been reconceptualized as “arenas” of habitation and “repositories” of culture [1,2], embodying spatiotemporal transformation and cultural landscapes [3]. Both the UNESCO World Heritage Convention and the 2030 Agenda for Sustainable Development [4] highlight their multifunctional value in ecological protection, heritage preservation, and sustainable development. This multifunctionality lies in the integrated interactions of ecological, economic, and cultural dimensions [5], while sustaining local knowledge and historical continuity [6,7].

Digital globalization is rapidly reshaping rural structures, accelerating depopulation and threatening the disintegration of traditional settlements. The UN projects that by 2050, 68% of the global population will live in cities, sharply contracting rural regions. In China, many rural communities face profound shrinkage, reflecting the structural erosion of social cohesion and collective vitality. Although six editions of the List of Chinese Traditional Villages have designated 8155 villages for protection [8], many continue to face functional decline, poor services, and cultural discontinuity. The situation is most severe in southwestern China, where difficult terrain and limited transport intensify marginalization. Yunnan, a highland region of diverse ethnic groups, contains lake-area villages such as those around Yilong Lake in Shiping County (Figure 1), where settlement systems are shaped by terrain, ecological vulnerability, and cultural diversity. Villages around Yilong and Qilu Lakes have developed multifunctional systems based on lake ecology, agriculture, and ethnic culture, offering a key case for studying functional differentiation and nonlinear coupling in traditional settlements [9].

“Differentiation” captures variations in ecological, economic, and cultural functions across spatial scales, while “coupling” highlights dynamic interactions among functional systems. Yunnan Province has introduced multiple policies to coordinate spatial functions in traditional villages—such as urban development and heritage-preservation regulations [10], tiered protection planning for towns and villages [11], practical planning initiatives [12], and county-level village guidelines [13]. However, these measures mainly stress ecological red lines and rigid land-use control, while overlooking flexible, nonlinear spatial-functional interactions within villages [14].

This limitation is evident in Shiping County’s Yilong Lake and Qilu Lake watersheds, where ecological vulnerability and development constraints are closely intertwined, exposing the structural heterogeneity of multifunctional village systems [15]. Here, villages face pressures from economic growth, ecological protection, and cultural preservation, producing complex, overlapping spatial functions. However, existing studies remain focused on static classification, neglecting dynamics of coupling, differentiation, and feedback between villages and their watershed environments [16,17].

By reviewing literature on multifunctional differentiation and nonlinear coupling, this study analyzes the spatial evolution of the “water–paddy–residence” system and builds a theoretical basis for understanding how natural rhythms and social functions adapt through spatial coupling.

1.1. Multifunctional Differentiation in Traditional Village Systems

Multifunctional systems in traditional villages exhibit strong heterogeneity and complex cross-system feedbacks [18], making multifunctionality a key lens for analyzing spatial characteristics [19]. Some studies take a static, cross-sectional view, classifying villages [20] by function (e.g., agricultural- or tourism-oriented) [21,22]. Others examine functional coordination or links between multifunctionality and land-use transitions [19]. Recent work also highlights challenges, including cultural shifts from tourism, fragmented governance, and policy-function mismatches [23,24].

From a systems perspective, current research faces three limitations. First, most studies focus on the static identification regarding the “existence” or “strength” of functions, constrained by rigid village classifications and neglecting the dynamic evolution of functional structures across scales and contexts [25]. Second, uncoordinated multifunctional systems risk “functional conflict traps,” undermining spatial governance [26]. Third, recurring issues in current territorial spatial planning, such as “policy misalignment” and fragmented governance, arise from overlooking nested structures, dominant variables, and nonlinear interactions [27]. For instance, cultural functions are often reduced to “performative consumption” under tourism, weakening sustainable value recognition and leading to the misuse of cultural resources [28]. Under the combined influence of complex factors such as ethnic and religious dynamics, population mobility, and ecological disturbances, ignoring the nonlinear characteristics of villages as dynamic “human-land-institution” systems risks producing governance strategies that fall into path dependence and ineffective interventions [29].

Therefore, research should shift from “function identification” to “coupling mechanisms” [30], and from “linear classification” toward “nonlinear differentiation” [31]. This represents a key step toward making the governance of traditional villages more adaptive and multifunctional.

1.2. Nonlinear Coupling in Regional Human–Land Systems

Recent studies increasingly apply machine learning and statistical models to examine complex multifunctional relationships in traditional village human-land systems, including function classification [32], cluster analysis [33] and probabilistic networks [34] reveal spatiotemporal fluctuations, abrupt changes, and dynamic mechanisms [35,36]. However, regional village clusters exhibit more complex function evolution than current models capture, with most studies emphasizing static differentiation and ignoring thresholds and transition mechanisms [37]. This limits spatial regulation and functional coordination [38]. A mechanism-driven approach is needed to capture coupling dynamics and evolutionary patterns of multifunctional systems for villages [27].

Coupling is generally understood as a dynamic linkage between interacting systems, emerging over time through reciprocal adaptation, alignment, and information exchange rather than static or linear associations [39]. Beyond synchronous co-variation, coupling emphasizes temporal coherence and mutual responsiveness among subsystems. In human–land systems, nonlinear coupling is introduced to identify development trajectories that differ fundamentally from linear causality or simple positive correlations, in which functional dimensions interact through feedbacks, thresholds, and context-dependent mechanisms, producing differentiated and path-dependent outcomes across space and time [40]. This conceptualization highlights the limitations of static linear models in representing the complex dynamics of multifunctional evolution in watershed villages and underscores the necessity of a nonlinear coupling perspective to capture conditional, adaptive, and context-sensitive interdependencies among functional subsystems.

In Yunnan’s lake-region ethnic villages [41], multifunctional evolution unfolds within ecology-led development trajectories [42] embedded in cultural and institutional contexts, rendering static or linear analytical approaches insufficient for effective spatial governance [43]. For instance, in the traditional village cluster of Yilong Lake’s watersheds in Shiping County, Honghe Prefecture, key influencing variables—such as cultural identity, land attachment, and ecological memory—operate as unstructured and context-dependent variables. Although difficult to operationalize within conventional quantitative statistical models, these factors critically shape functional transitions and coupling dynamics, and their roles may be insufficiently revealed in the absence of a coupling-oriented analytical perspective [44].

1.3. Contextual Conditions of Multifunctional Coupling in Multi-Ethnic Watershed Villages

Honghe Prefecture exhibits a typical multi-ethnic settlement region, comprising 11 historically established ethnic groups at the prefectural level. Shipping County features a socio-cultural setting primarily influenced by Yi people, coexisting with Han and other minority groups over an extended period [45]. Within this context, human–land relationships in watershed villages are shaped by both quantifiable material factors, such as natural resources and geographical position, and non-material factors, including culturally ingrained institutional frameworks, social conventions, and historically established settlement practices. Instead of forming a hierarchical link, these two types of elements function concurrently, impacting human–land coupling processes through separate yet interconnected pathways [46].

More specifically, ethnic composition and historical trajectories condition multifunctional evolution through their influence on land-use strategies, collective action, and local governance. As non-material factors characterized by limited observability, weak comparability, and strong context dependence, they are difficult to quantify and incorporate into a standardized indicator system comparable to material variables such as economic development, land resources, industrial structure, or accessibility.

On this basis, this study considers quantifiable spatial geographical environmental factors at the village level as the foundational inputs for the coupling coordination model, while conceptualizing ethnic and historical diversity as contextual constraints for analyzing coupling outcomes in governance pathways. This analytical perspective integrates material and non-material aspects at the mechanistic level, collectively influencing the patterns of human-land interaction and the varied governance pathways in traditional watershed villages.

Building on the above conceptual and contextual framework, taking the 45 traditional villages in the Yilong Lake region of Shiping County, Honghe Prefecture, Yunnan Province (Figure 1), as research samples, this study examines how spatial functions in traditional villages adapt through coupling mechanisms, revealing dynamic interactions under multivariable, multiscale conditions. This analysis process focuses on the underlying logic of “multifunctional differentiation” in traditional villages under the influence of lake systems, as well as the interactive patterns of “regional nonlinear coupling”. It first analyzes spatial differentiation between interior and peripheral plateau-lake villages, then applies a coupling evaluation model to identify tensions, imbalances, and pathways of functional change, constructing a framework for nonlinear regional coupling assessment. Finally, it explores adaptive coupling modes and proposes multi-system coupling strategies, addressing gaps in understanding multifunctional dynamics and providing theoretical and practical guidance for spatial governance and coordinated development in traditional villages [47]. This approach addresses current needs in village governance for refined “living heritage” and “dynamic adaptation.”

2. Materials and Methods

This study examines 45 traditional villages in Shiping County, Honghe Prefecture, Yunnan Province, all listed in the National Catalogue of Chinese Traditional Villages [48]. Shiping County [49] spans 3041.81 km² with 271,900 residents, known as a historic “land of literature and culture” where Yi and Han traditions intersect. As of March 2025, the villages cluster around Yilong Lake, Qilu Lake, and surrounding mountains, forming a characteristic lake-mountain-village pattern (Figure 2). They illustrate the co-evolution of water, ecology, economy, and culture, while reflecting both shared traits of lake-region villages and local heterogeneity [50]. This sample thus provides a strong basis for empirical analysis and policy insight.

The research followed a methodological framework (Figure 3). First, key indicators across the “lake-mountain-village” dimensions were selected, and feature data capturing the current status and potential functions were collected. Subsequently, a stepwise data processing and coupling evaluation procedure was conducted (Figure 4). Data were preprocessed and indicator weights were determined using the entropy weighting method. A coupling coordination model was then constructed using the coupling function (C), comprehensive coordination index (T), and coordination degree (D), normalized and classified into four ranges. Spatial differentiation, including adjacency-dependence and clustering-periphery, was analyzed via kernel density, and mechanisms underlying multifunctional differences were identified. Finally, regional coordination pathways were proposed, providing a scientific basis for governance and guidance for the protection and development of traditional villages in Shiping County.

2.1. Data Collection and Indicator System

2.1.1. Construction of Indicator System

The data utilized in this study are sourced from the most recent available materials (primarily post-2020), pertaining to the period following the National Territorial Spatial Planning (2021–2035) [51] which marks a pivotal institutional transition in China’s high-level spatial governance reform. Accordingly, this study assessed multifunctionality in traditional villages using three core systems—economic production, socio-cultural development, and ecological environment (

Table 1. Summary of sub-item data.

System	Sub-Item	Date Source
Economic production	Distance to county seat (km)	China Statistical Yearbook (regional data) Yunnan Statistical Yearbook Honghe Prefecture Statistical Yearbook Shiping County Statistical Yearbook
	Total annual economic income (10,000 CNY)
	Annual per capita net income (CNY)
	Total land area of the village (km2)
	Rural population
	Total arable land (mu)
	Per capita arable land (mu)
	Annual income from secondary and tertiary industries (10,000 CNY)
Socio-cultural development	Night-time light intensity (nW/cm2/sr)	NOAA National Centers for Environmental Information: https://www.ncei.noaa.gov/news/sunset-nighttime-lights-noaa (accessed on 11 January 2025)
	Distance from national cultural relics (m)	List of national key cultural relics protection units in eight batches (State Administration of Cultural Heritage Comprehensive Administrative Management Platform): https://ncha.gjzwfw.gov.cn/ (accessed on 11 January 2025)
	Distance from adjacent railway stations (m)	GRDC (https://www.gis5g.com/data-resource) (accessed on 11 January 2025) 2024 vector dataset for national railways and stations (2024)
	Distance from adjacent roads	National Geomatics Center of China (https://www.gis5g.com/data-resource) (accessed on 11 January 2025) 2023 vector dataset for the national road network
	Administrative rank of adjacent roads
Potential of environmental resource	Distance from adjacent lakes (m)	GRDC (https://www.gis5g.com/data-resource) (accessed on 11 January 2025) 2024 Level-5 standard river system shapefile (SHP) data for China
	Elevation of adjacent lakes (m)
	Elevation difference between village and adjacent lake (m)
	Distance from adjacent rivers (m)
	Elevation difference between village and adjacent rivers (m)

). First, this aligns with the Outline of the National Territorial Spatial Planning (2021–2035), which promotes coordinated “production-living-ecological spaces” (PLES) and multi-plan integration, and with Shiping County’s designation of agricultural, urban, and ecological spaces as primary carriers. Second, the coupled evolution of these systems reflects spatial restructuring, interrelations among geographic elements, and functional space classification [52]. Collectively, they capture the material, cultural, and ecological foundations of traditional village development and provide a basis for identifying villages with high coupling degrees, especially those influenced by lake–mountain interactions in basin environments.

2.1.2. Key Points in Data Selection

The selection of sub-item data for analyzing the multifunctional characteristics of traditional villages focuses on three aspects.

Economic production indicators capture the material basis of traditional villages’ spatial functions, reflecting rural development and supporting multifunctional evolution [53]. Aligned with the requirements for “optimized production space” under the PLES framework, these indicators assess development potential through village location, economic development (income), and production capacity (arable land and secondary/tertiary industry income),and production capacity (arable land and secondary/tertiary industry income). In particular, village location is operationalized by distance to the county center, which directly conditions market accessibility, labor mobility, and opportunities for non-agricultural income generation, and is therefore treated as a core economic-related factor rather than a purely locational variable.

Socio-cultural development indicators reflect villages’ service capacity and cultural value [54]. They include socioeconomic network position, accessibility of cultural resources, night-time light intensity as a proxy for economic activity [55], distance to national heritage sites, and horizontal/vertical proximity to transport nodes, capturing both cultural access and regional connectivity. In the multi-ethnic context of Shiping County, proximity to cultural heritage sites represents opportunities for cultural participation, social interaction, and heritage-related tourism.

Environmental resource indicators are based on villages’ dependence on and access to Yilong Lake, highlighting interactions with natural resources [56]. Lakes and rivers provide living resources and shape ecosystem services, with water-related measures—distance and elevation to lakes and tributaries—reflecting resource use and potential for landscape-based development under watershed conditions.

2.2. Data Preprocessing

After constructing the indicator tables for the three systems, all the raw data processing was carried out according to the procedure (Figure 4). All the raw data were standardized using Stata 17. The goal of standardization was to convert indicators with different units into a common scale for comparison and analysis. For positive indicators (global positives: x₂, x₃, x₄, x₅, x₆, x₇, x₈, y₁, y₂, y₃), standardized values closer to 1 indicate better performance. For negative indicators (global negatives: x₁, y₄, y₅, z₁, z₂, z₃, z₄, z₅), higher standardized values correspond to poorer performance.

$$x_i=\left\{\begin{array}{c}\begin{array}{cc}{\displaystyle \frac{x_i-\min \left(x_i\right)}{\max \left(x_i\right)-\min \left(x_i\right)}},& \mathrm{Positive} \mathrm{Indicators}\end{array}\\ \begin{array}{cc}{\displaystyle \frac{\max \left(x_i\right)-x_i}{\max \left(x_i\right)-\min \left(x_i\right)}},& \mathrm{Negative} \mathrm{Indicators}\end{array}\end{array}\right.$$

(1)

The standardized values ${x_i}^{\prime }$ (Equation (1)) range from 0 to 1, reflecting the relative position of each indicator within its range of values.

Weight calculation. The weights $y_i$ (Equation (2)) of the standardized indicators were calculated to represent the relative importance of each indicator within the overall dataset.

$$y_i={\displaystyle \frac{x_i}{{\sum }_{i=1}^m{x}_i}}$$

(2)

where i represents the indicator characteristics of different samples, m represents the total number of indicators in each system: the first system contains 8 indicators, the second system contains 5 indicators, and the third system contains 5 indicators.

2.3. Scoring Using the Entropy Weight Method

Indicator weights were assigned using the entropy weight method rather than subjective approaches like the Analytic Hierarchy Process (AHP), which can produce unstable, non-reproducible results. The entropy method objectively derives weights from data by analyzing each indicator’s information entropy, ensuring transparency, traceability, and greater rigor.

The entropy weight method was applied in Stata to calculate standardized indicator weights. Information entropy E (Equation (3)) quantifies each indicator’s dispersion across the sample, providing an objective measure while avoiding subjective bias.

$$E_i=-{\displaystyle \frac{1}{\ln N}}{\displaystyle \sum _{i=0}^m}\ln y_i$$

(3)

where N represents the 45 traditional villages, and m represents the total number of indicators in each system (8, 5, and 4 for the first, second, and third systems, respectively), and $i=1$ denotes the first sample.

The information entropy E measures the uniformity of the distribution of indicator i:

• $E\to 1$: The values of this indicator are similar across all samples, indicating little variation and low information content.
• $E\to 0$: The values of this indicator vary widely among the samples, indicating high variation and rich information content.

The weight W of each indicator is calculated using the difference coefficient D (Equation (4)) according to Equation (5):

$$D_i=1-E_i$$

(4)

$$W_i={\displaystyle \frac{D_i}{{\sum }_{i=1}^m{D}_i}}$$

(5)

A larger difference coefficient D corresponds to a higher weight W, indicating that indicator i has a greater contribution to the overall evaluation.

Next, Standardized indicator values are multiplied by their weights to compute composite scores U₁, U₂, and U₃ (Equations (6) and (7)), providing an accurate representation of each system’s relative importance and interrelationships.

$$\mathrm{Score}=\sum _{i=1}^m{W}_i{x_i}^{\prime }$$

(6)

$$\sum _{j=1}^m{W}_i=1$$

(7)

2.4. Coupling Evaluation Model

Coupling can be applied to traditional villages’ multifunctional systems, reflecting interactions among economic, social, and ecological subsystems to reveal spatial differentiation [57]. In this study, a coupling coordination model was constructed using Stata, as described below.

2.4.1. Coupling Function (C)

The coupling function C (Equation (8)) is used to measure the strength of interactions among the three systems (e.g., economic, social, and environmental). The principle is as follows:

$$C={\left({\displaystyle \frac{\mathrm{Numerator}}{\mathrm{Denominator}}}\right)}^{{\displaystyle \frac{1}{3}}}$$

(8)

The numerator (product of subsystem indicators) reflects interaction strength; the denominator normalizes unit and scale differences, keeping C between 0 and 1 for comparability.

The coupling function C for the three systems is calculated as follows (Equation (9)):

$$C={\left[{\displaystyle \frac{U_1\times U_2\times U_3}{\left(\frac{U_1+U_2+U_3}{3}\right)}}\right]}^{{\displaystyle \frac{1}{3}}}$$

(9)

where

• $C=0$: The three systems are completely uncorrelated or disconnected.
• $C=1$: The three systems are fully coupled, representing the strongest coupling and optimal coordination.
• The larger the value of C, the stronger the balance among the three systems.

2.4.2. Comprehensive Coordination Index (T)

The comprehensive coordination index T (Equation (10)) is used to evaluate the average level of integrated development across the three systems. The rationale is as follows:

$$T={\beta }_i{U}_1+{\beta }_2{U}_2+{\beta }_3{U}_3$$

(10)

$$T={\displaystyle \frac{1}{3}}{U}_1+{\displaystyle \frac{1}{3}}{U}_2+{\displaystyle \frac{1}{3}}{U}_3$$

(11)

T is the average of standardized scores (U₁, U₂, U₃) for economic, cultural, and environmental systems; The coefficient β represents the relative importance of each subsystem in the comprehensive development index (T). At this stage, economic production, social and cultural development, and ecological environment are assigned equal weight (β = 1/3; Equation (11)), aiming to ensure the three subsystems hold equal standing at the system integration level. This prevents subjective prioritization in subsequent steps lacking empirical validation. This consideration arises from the following rationale: the equal-weighting assumption is applicable only during the system integration phase of developing the comprehensive development index (T) and does not affect weight distribution at the indicator level, which is exclusively determined by the entropy method [58].

This methodological choice is especially relevant in multi-ethnic, multi-watershed, and multi-scale village contexts, where expert assessments of subsystem importance often diverge and lack stable consensus. Although methods such as the Analytic Hierarchy Process (AHP) can be employed to assign differential system weights, they depend on expert judgment and may introduce subjectivity and sensitivity to evaluator preferences [59].

Thus, the actual contribution differences among subsystems are inherently reflected in U₁, U₂ and U₃ through the weight allocation of the entropy method. In this sense, this stage is a neutral integration rule designed to enhance the transparency, reproducibility, and comparability of coupling results. Higher T indicates greater integration.

2.4.3. Calculation of the Coupling Coordination Degree (D)

The coupling coordination degree (Equation (12)) reflects interaction strength and overall development among the three systems in the 45 villages; higher D indicates stronger coordination and integration.

$$D=\sqrt{CT}$$

(12)

By calculating C, T, and D, we assess the development characteristics of villages. The possible scenarios are as follows (Figure 5):

• Low C and low T: Weak coordination and unbalanced coupling.
• High C and low T: Strong coordination but limited overall development.
• Low C and high T: High development but poor coordination among systems.
• High C and high T: Both coordination and development are strong.

System performance depends on both development (T) and coordination (C); low values reduce D. In practice, D integrates these factors to rank villages and reveal coordination, providing a rigorous basis for analyzing interactions and spatial patterns.

3. Data Analysis

3.1. Features of Spatial Differentiation

The kernel density map (Figure 6 and

Table 2. Village numbers and corresponding names.

No.	Village Name	No.	Village Name
1	Lulai Village	24	Taiyue Village
2	Mocedian Village	25	Potoudian Village
3	Mushan Village	26	Baxin Village
4	Quzuo Village	27	Liujiashan Village
5	Xiaochong Village	28	Longpeng Village
6	Longhei Village	29	Samazha Village
7	Wuying Village	30	Doudiwan Village
8	Sujiazhai Village	31	Baoxiu Village
9	Zhangbenzhai Village	32	Yangxinzhai Village
10	Zhuchong Village	33	Tala Village
11	Panying Village	34	Bailang Village
12	Taoyuan Village	35	Xinjie Village
13	Baisafen Village	36	Yuejiawan Village
14	Shaochong Village	37	Diemulong Village
15	Shuiguachong Village	38	Shigang Village
16	Laoxudian Village	39	Yiheiji Village
17	Longgang Village	40	Lanziying Village
18	Sewan Village	41	Maohe Village
19	Fujiaying Village	42	Lijiazhai Village
20	Guanshang Village	43	Song Village
21	Fengshan Village	44	Dashui Village
22	Dazhai Village	45	Xiaoruicheng Village
23	Zhengying Village

) identifies the first differentiation cluster in the Yilong Lake dam area (

Table 3. Spatial patterns of villages in the dam area.

Name	Master Plan	Aerial View of Village Layout
Lijiazhai Village
Song Village
Dashui Village
Xiaoruicheng Village
Spatial Characteristics	Alluvial plain buffer zone	Linear extension (tributaries/roads)

), covering Baoxiu, Yilong, and Baxin. Kernel density estimation is used here to visualize the spatial concentration intensity of traditional villages, reflecting the degree of settlement agglomeration under combined natural and infrastructural conditions. Higher density values indicate stronger spatial clustering and functional interaction among villages. Favorable terrain, fertile soils, abundant water, and the lake-based irrigation system support intensive farming, while transport, administrative resources, and infrastructure enhance integration [57]. Yilong functions as the political–economic center, Baoxiu forms a “lake–town–village” pattern (Table 3), and Baxin retains stability as a farming hub. Collectively, they constitute a highly coordinated settlement belt [60].

The second differentiation type occurs in high-altitude northern and southern villages, including Shaochong, Longpeng, and Niujie towns (

Table 4. Spatial patterns of villages in the mountainous area.

Name	Master Plan	Aerial View of Village Layout
Tala Village (belt-like shape)
Yiheji Village (belt-like shape)
Belt-like Differentiation	Terrain-adapted	Contour-adapted
Diemulong Village (clustered shape)
Clustered Differentiation	High-altitude terraces	Arable–transport link

). In these areas, lower kernel density values reflect a more dispersed settlement pattern, indicating weaker spatial connectivity and limited functional interaction among villages. These low-density, fragmented settlements are constrained by mountain terrain, limited transport, weak infrastructure, and restricted land development, hindering multifunctional coordination [61]. Land development is also restricted, making it difficult to achieve the multifunctional coordination seen in the dam area. While ecological resources and traditional culture are well-preserved, these villages face marginalization risks—population outflow, inadequate services, and potential functional decline or shrinkage.

In Shiping County, the evolution of the multifunctional system shows a “concentrated dam area-dispersed mountainous area” pattern. Villages around Yilong Lake have developed an integrated “water-paddy-residence” system, with clustering influenced by topography, transport, and proximity to administrative centers, reflecting a natural-social evolutionary trajectory. Since the Ming and Qing dynasties, the lake’s irrigation system has supported stable agricultural settlements [62], while modern governance and infrastructure have further reinforced clustering in the dam area [49]. Mountain villages, limited by rugged terrain and poor transport, lag in integrating into the resource-institution development axis, producing a “basin core-mountain periphery” spatial pattern and a basin-oriented tension structure.

3.2. Classification of Coupling and Coordination

Previous studies have classified multifunctional coupling data using methods such as breakpoint segmentation [63], coupling state classification [64], spatial heterogeneity partitioning [65], and settlement evolution classification [51], reflecting context-dependent approaches. To evaluate the coupling and coordination degree (D) in traditional villages, the quartile method is preferred for its clear criteria, even intervals, and balance between granularity and interpretability. Quantile classification was adopted to ensure balanced group sizes and comparability across villages, making it more suitable for the present sample structure [66].

Given the D range in this study (0.4032–0.7465), the quartile method was adopted (

Table 5. Classification of coupling coordination degree.

Traditional Village Sample	Coupling Coordination Degree (D)	Value Range	Number of Villages at Different Stages
Lulai Village	0.4032153	0.40–0.48	Stage 4: antagonistic stage, 11 villages
Mocedian Village	0.4253063
Mushan Village	0.4265351
Quzuo Village	0.4274328
Xiaochong Village	0.459585
Longhei Village	0.4694006
Wuying Village	0.4832362
Sujiazhai Village	0.4846746
Zhangbenzhai Village	0.4866254
Zhuchong Village	0.4872925
Panying Village	0.4879586
Taoyuan Village	0.4941365	0.48–0.57	Stage 3: breaking-in stage, 21 villages
Baisafen Village	0.4954966
Shaochong Village	0.5120823
Shuiguachong Village	0.5187631
Laoxudian Village	0.525607
Longgang Village	0.5293064
Sewan Village	0.5297481
Fujiaying Village	0.5325361
Guanshang Village	0.5335027
Fengshan Village	0.5412129
Dazhai Village	0.5429253
Zhengying Village	0.5447065
Taiyue Village	0.5468364
Potoudian Village	0.5531332
Baxin Village	0.5533727
Liujiashan Village	0.5558993
Longpeng Village	0.5595284
Samazha Village	0.5622892
Doudiwan Village	0.5626635
Baoxiu Village	0.5707847
Yangxinzhai Village	0.5720714
Tala Village	0.5832748	0.57–0.66	Stage 2: primary coordination stage, 9 villages
Bailang Village	0.5983039
Xinjie Village	0.598406
Yuejiawan Village	0.6045035
Diemulong Village	0.6107176
Shigang Village	0.6213369
Yiheiji Village	0.6220282
Lanziying Village	0.6377032
Maohe Village	0.6568747
Lijiazhai Village	0.6636602	0.66–0.75	Stage 1: coherent stage, 4 villages
Song Village	0.7119327
Dashui Village	0.7206658
Xiaoruicheng Village	0.7465541

) to enhance stability and explanatory power. In stage 1 (coherent), 4 villages have the highest D values (0.66–0.75). Stage 2 (primary coordination) includes 9 villages (0.57–0.66), stage 3 (breaking-in) has 21 villages (0.48–0.57), and stage 4 (antagonistic) contains 11 villages (0.40–0.48), where coordination nears breakdown.

Based on kernel density and coupling coordination analyses (Table 5 and Figure 7), the four value ranges are grouped into three multifunctional coupling mechanisms—institutional embedding, self-organization, and resource matching—showing a generative, differentiated, and dynamic pathway of coordination (Figure 8).

Category I: Functional integration under institutional embedding. The coupling mechanism under Category I encompasses a total of four villages (Stage 1 in Figure 7 and Table 5), all situated within the core area of Yilong Town (i.e., Xiaoruicheng Village, Dashui Village, Song Village, and Lijiazhai Village). These villages have active secondary and tertiary industries, with D values of 0.75, 0.72, 0.71, and 0.66 (The red section in Figure 9)—well above the overall average of 0.55—indicating a high level of multifunctional coupling. They share proximity to Yilong Lake (<1 km), low elevation differences (<30 m), and strong transport accessibility (<5 km to the county seat) (Figure 10). Lijiazhai Village, for example, restored ecological farmland in 2021 and developed homestay tourism and lakeside agriculture, creating a model of “production-settlement integration” with closely linked land, transport, and industry. This high coupling was achieved through the combined effect of internal resources and external policy support [67].

Category I villages follow a high-coupling, policy-driven pathway, integrating external institutions with local endowments. Natural features (lake basins, alluvial plains) enable reclamation and irrigation [68], while governments guide resource allocation through planning and policy. In line with Ostrom’s “institutional fit,” clear property rules align institutions with resource use, producing a positive-effect integration pathway.

Category II: Endogenous adaptation via self-organization. Nine villages (Stage 2 in Figure 7 and Table 5) fall under Category II. Three mountainous villages in Niujie Town (Tala, Diemulong, and Yiheji) show notable coupling (D = 0.58, 0.61, 0.62, respectively), despite being over 35 km from the county seat, far from Yilong Lake and the nearest railway, and having limited transport (the yellow section in Figure 9). Category II villages reflect self-organized SES units where clan networks, traditional practices, and ecological farming sustain stability. Guided by SES theory, knowledge transfer, institutional innovation, and self-organization foster ecological resilience and social integration [69]. For example, Tala Village combines understory planting, terraced farming, and land rituals to regulate ecological pressures.

Category III: Antagonistic pattern under value mismatch. Category III includes 32 villages (The blue and green section in Figure 9): 21 in the breaking-in stage (Stage 3 in Figure 7 and Table 5) and 11 in the antagonistic stage (Stage 4 in Figure 7 and Table 5). All are near-dysfunctional, with D values below the regional average. Stage 3 villages often have rich natural resources but suffer functional mismatches associated with limited accessibility and insufficient institutional embedding in infrastructure provision and development planning. For example, Samazha Village has ample arable land and forests, yet its lack of industrial integration and market links keeps per capita income over 30% below the county average.

Lulai Village (Stage 4) in the southern Yilong Lake mountains faces ecological limits and low investment, hindering tourism. From Lefebvre’s spatial triad view, such villages show an antagonistic pattern: resource-rich but functionally ineffective, trapped between potential and dysfunction [70].

4. Results: Socio-Ecological Coupling Mechanisms and Governance Pathways

In summary, coupling classification enables evidence-based allocation of village functions, balancing equity, ecology, and cultural preservation. It provides practical guidance within a “macro-guidance–micro-implementation” framework. Based on Categories I–III, three tailored strategies with supporting policies are proposed.

4.1. Institutional Enhancement for Coupling Optimization in Basin Core Areas

To summarize the characteristics and adjustment strategies of Category I, box plots compared coupling coordination and key functions across village types (Figure 11a–c). Results show that from Category I to III, total arable land and population decline while per capita arable land rises, revealing a nonlinear link between resources and population, with economic functions marked by structural heterogeneity and stage-specific imbalances.

The results reveal a threshold effect of location: most highly coupled villages cluster around Yilong Lake, yet proximity alone does not ensure development. Some lakeside Category II villages still show low incomes and weak coupling due to industrial monotony. Thus, natural endowments provide the base, but institutional support and industrial guidance are essential to turn potential into development.

Category I villages (e.g., Xiaoruicheng, Dashui, Lijiazhai, and Song) exhibit stable, efficient multifunctionality, highlighting the regulatory role of institutional capital in the “field–capital–practice” framework. Their development follows a heterogeneous, non-equilibrium path shaped by resource integration, coordination, and institutional mobilization. Lijiazhai integrates agriculture and services via an “ecological rice-agritourism” model [71], while Dashui reshapes space via transportation upgrading.

To strengthen Category I villages, focus on three priorities:

• Integrate county-level and village planning
• Foster multifunctional industries (“one village, one product” model).
• Empower stakeholders, recognize cultural inheritors, ensure cultural justice

4.2. Adaptive Allocation for Functional Transition in Semi-Peripheral Mountainous Villages

To examine Category II villages, box plots compared coordination degree and key spatial elements across types (Figure 11d–f). From Category I to III, the distance from the county seat and income from secondary and tertiary industries decline, while the land area remains similar. Thus, village development is not determined by land size alone but by resource accessibility, industrial balance, integration capacity, and institutional intervention.

Some remote Category II villages, such as Lanziying, Yiheji, and Diemulong, still exhibit high coupling coordination (above 0.6). This indicates that, despite geographic and institutional constraints, peripheral villages can optimize multifunctional coupling through autonomous adjustment, exemplifying “non-central-non-homogeneous” coupling.

For Category II villages, spatial adjustment should adopt evidence-based, function-specific pathways. Key actions are:

• Improve peripheral infrastructure and services through county-level compensation [72], and revitalization funds to address fragmented spatial structures.
• Build “district–village” industrial chains that integrate cultural and tourism resources, tailored to local differences and coupling degrees.

4.3. Equity Compensation for Conflict Mitigation in Mountain Hinterlands

Among Shiping County’s 45 traditional villages, 32 fall into a low-coupling state (the blue and green section in Figure 9), corresponding to Category III. This includes 21 villages in Stage 3 and 11 in Stage 4. They show wide variability in socio-cultural and ecological functions, reflecting internal antagonism and a lack of coordinated spatial systems. These conflicts, marked by “reverse antagonism, obstruction, and imbalance”, underscore the need for positive adjustment strategies.

Mitigating reverse antagonism: Mitigating reverse antagonism: Category III mountain villages, especially in southern Niujie and peripheral mountainous areas of Baxin, average a low coupling value of 0.46, below the county mean of 0.54. Poor transport—over 30 km from the county seat, >100 m elevation differences, and in some cases no all-season roads—creates functional isolation. While median distances to rail or road nodes differ little across categories (Figure 11g), road hierarchy shapes spatial coupling: villages near high-grade roads show stronger coordination, while those on rural or dead-end roads often become functional “islands,” even if close to core infrastructure. Thus, accessibility quality matters more than physical distance.

Mountain towns like Shaochong and Longpeng also have to deal with environmental problems that make their economies weakened, which adds to the problem of poor coupling. To make these kinds of systems stronger, it is suggested that ecological pay and the growth of green industries (like terraced agrotourism) be used together. According to new studies, the prices of carbon allowances under the national ETS have stayed low compared to other foreign markets. They have mostly been around tens of Chinese yuan per tonne [73], which is about what the market is worth right now. This price range gives policymakers a way to measure how much pay should be given in rural areas that are sensitive to the environment.

In ecological redline zones (e.g., Yilong Lake east belt), adopt rice–fish–duck systems to cut chemical inputs and boost resilience. According to the Territorial Spatial Master Plan of Shiping County (2021–2035), priority should go to road extension and hierarchy upgrading (Niujie, Baxin) to link heritage, ecology, and culture, embedding tourism services within a transport–tourism–ecology framework to strengthen system functions.

5. Discussion

5.1. Validation of Nonlinear Multifunctional Coupling Mechanisms

This study empirically validates the core research question of nonlinear multifunctional coupling in rural watershed systems. The result reveals differentiated human–land coupling mechanisms within Yunnan’s plateau lake watershed systems, showing how agrarian resources (economic production), socio-cultural attributes (regional resources), and policy-driven institutional adjustments jointly, rather than from environmental endowment alone, shape the multifunctional evolution of traditional villages. By integrating a coupling coordination model with spatial differentiation analysis, the study advances theoretical understanding of nonlinear coupling processes in multifunctional rural systems under watershed and mountainous geomorphological conditions.

The findings identify a coupling gradient from irrigated lake- and riverbed-based core areas to semi-peripheral zones and mountainous hinterlands, highlighting the dynamic interaction between natural environmental conditions and human governance structures. Importantly, the results confirm that multifunctional coupling does not adhere to a straightforward positive linear correlation with natural resource advantages. Even villages with advantageous environmental conditions may still have adverse coupling effects due to inadequate institutional intervention and integrated resource planning. Conversely, villages located in tributary or mountainous areas, despite limited access to lake-based resources, may exhibit relatively stable or even positive coupling states due to effective self-organized agrarian adaptation and functional adjustment.

It should be noted that the nonlinearity discussed in this study does not refer to statistically tested nonlinear functional forms in a strict mathematical sense. Rather, it is used to describe heterogeneous, stage-differentiated, and context-dependent coupling patterns observed across villages under different resource, institutional, and spatial conditions. Given the cross-sectional nature of the data and the use of a coupling coordination framework, the results highlight differentiated coupling responses and threshold-like contrasts between village groups, rather than formally estimating nonlinear parameters. In this sense, the findings should be interpreted as evidence of structural heterogeneity and non-equilibrium coupling behaviors, which deviate from simple linear or resource-deterministic assumptions and underscore the importance of governance context and adaptive mechanisms.

Overall, the coupling relationships observed in this study encompass both synergistic coordination and antagonistic obstruction, thereby confirming the nonlinear and context-dependent nature of multifunctional rural development in watershed environments.

5.2. Generalization of Governance Pathways and Practical Implications

Beyond the case of Shiping County, the proposed framework demonstrates applicability across watershed systems in Yunnan and the broader ethnically diverse regions of Southwest China. The lake–mountain coupling perspective abstracts general human–land interaction patterns beyond basin-specific uncertainty and, when situated within the broader context of karst watersheds and plateau lake systems in Yunnan and Southwest China, provides a transferable framework for optimizing coupling strategies across regions characterized by pronounced geomorphological and rural development heterogeneity.

Within this broader regional context, distinct geomorphological settings correspond to differentiated coupling pathways. Tourism growth and industrial agglomeration exacerbate land competition and ecological constraints [74] in the central Yunnan plateau lake-basin region (such as the Dianchi, Fuxian, Xingyun, and Yangzonghai basins), necessitating an institutional embedding pathway to achieve multifunctional coupling in Basin Core Areas. The Lancang River watershed and the Nanla River tributary valley in southwestern Yunnan demonstrate significant ecological sensitivity alongside diverse livelihoods, where agriculture, forestry, and cross-regional mobility coexist, thereby enhancing the effectiveness of adaptive allocation pathways for semi-mountainous regions in bolstering functional system resilience) [75]. In the southeastern Yunnan karst Honghe watershed region (e.g., the Yilong and Qilu Lake basins) and the northwestern Yunnan high-altitude gorge and plateau lake region (e.g., the Erhai, Lugu, and Chenghai basins), rural development is simultaneously hindered by ecological vulnerability and significant risks of rocky desertification [76]. In these contexts, ecological and cultural landscape functions predominate village development, while equity- and compensation-focused governance strategies for mountain hinterlands are essential in alleviating development conflicts, maintaining multifunctional systems, and preserving ecosystem services.

Taken together, despite substantial variations in geomorphology and development stages, aligning governance strategies with terrain constraints, ecological susceptibility, livelihood foundations, and institutional contexts enables the differentiated coupling mechanisms identified in this study to generate interpretable and transferable theoretical value across diverse regions of Yunnan.

From a global sustainability perspective, this study identifies three adaptive governance paths that form a structured decision-support framework for promoting Sustainable Development Goals (SDGs) in rural watershed systems [77]. Institutional enhancement in basin core areas promotes compact and multifunctional rural settlements in alignment with SDG 11 (sustainable human settlements), adaptive allocation in semi-peripheral mountainous villages enhances climate resilience in accordance with SDG 13 (climate adaptation), and equity-oriented compensation in mountainous hinterlands safeguards ecological functions and sustainable land use in compliance with SDG 15 (life on land). Collectively, these routes convert human-land coupling mechanisms into distinct and scalable governance principles that facilitate integrated watershed management and evidence-based decision-making.

5.3. Limitations

However, the plateau lakes and their watershed habitats in Yunnan Province demonstrate more intricate human-environment interactions in the context of urbanization. Yilong Lake in Honghe Prefecture was chosen as a representative case study for plateau lake village systems, exhibiting notable representativeness at both spatial and sociological levels; however, it did not sufficiently differentiate the varied topographical environments and cultural contexts present across wider county-level or prefectural administrative areas. Consequently, subsequent study ought to broaden the analytical framework to neighboring counties or regional areas characterized by varied geography and distinct ethnic cultures to evaluate the robustness and transferability of the found coupling patterns.

In addition to constraints concerning spatial scope, methodological ambiguities persist regarding the comprehensiveness of the indicator system and the validity of the assumptions behind the three subsystems. Due to data confidentiality and availability limitations, specific ecological factors (such as soil quality and biodiversity) [78] and subjective cognitive elements (including cultural identity and practice-based perceptions) lack standardized and comparable quantitative metrics, complicating their consistent incorporation into the existing indicator framework. Future research should investigate: (i) the potential introduction of multicollinearity and endogeneity biases through the inclusion of non-material factors in coupling models, which may diminish model interpretability and explanatory power [79]; and (ii) the efficacy of capturing the intricate feedbacks and temporal dynamics of nonlinear human–land systems by integrating multi-source ecological data and hybrid modeling frameworks (e.g., sensitivity analysis, scenario simulation, or robustness testing), to identify stable analytical units within coupling coordination models and to rigorously evaluate the robustness and applicability of coupling outcomes under varying model assumptions [80].

Moreover, embracing a polycentric network perspective might envision traditional settlements as multifunctional nodes within regional systems. This method enables the comparative examination of coupling mechanisms, coordination efficacy, and the development of polycentric spatial ordering [81], thereby offering a more solid theoretical framework and empirical support for resilience-focused governance and tailored spatial management.

6. Conclusions

This study applied a coupling coordination model with entropy weighting to evaluate spatial heterogeneity and multifunctional dynamics in traditional villages within a plateau lake watershed context. By moving beyond static zoning research, the research emphasizes the dynamic interplay of resources, culture, institutions, and practices, and it underscores the role of institutional allocation in spatial coupling.

The empirical data results validate the internal consistency and logical completeness of the proposed research framework, demonstrating that multifunctional rural development in watershed environments is governed by nonlinear coupling mechanisms rather than linear resource-dependent relationships. The identification of coupling gradients, synergistic coordination, and antagonistic obstruction along lake-core, semi-peripheral, and mountainous hinterland villages substantiates the theoretical assumptions and methodological design of the study, thus providing a coherent analytical logic for examining human–land interactions in complex rural systems. Overall, by linking empirical evidence with governance pathways and sustainable development objectives, it offers both theoretical advancement and practical guidance for integrated watershed planning, resilience-oriented governance, and evidence-based rural development policy.