Coupling Relationship Analysis of Water Resources, Society, Economy, and Ecosystems in the Shule River Basin

Abstract

Climate change and human activities have intensified the imbalance between water supply and demand in the Shule River Basin. Prominent issues such as groundwater over-exploitation and insufficient ecological base flow have significantly constrained the high-quality development of the region. An evaluation system was developed comprising 20 indicators across four subsystems: water resources, society, economy, and ecosystems. The entropy weighting method was employed to determine the weights of each indicator. The coupling coordination degree of the water resource–society–economy–ecosystem system from 2003 to 2022 was assessed using a coupling coordination degree model. Network analysis was applied to evaluate the correlation and connectivity among indicators. A barrier diagnostic model based on indicator deviation was further constructed to identify key constraints within the system. The results showed that the overall coordination degree of the coupled system increased from 0.217 in 2003 to 0.409 in 2022, shifting from a moderately uncoordinated state to a weakly coordinated one. However, the coordination level remained low due to fluctuations in the water resource subsystem. The ecological and economic subsystems functioned as critical coupling hubs, while strong negative interactions within the water resource subsystem emerged as major constraints to coordinated development. Long-term dominant barriers included the proportion of water used for ecological and agricultural purposes, as well as per capita food production. After 2019, water resource-related indicators, such as per capita water availability and water production modulus, showed a marked increase in obstruction levels, highlighting the persistent challenges posed by water scarcity and inefficient utilization.

1. Introduction

As the second largest inland river in the Hexi Corridor, the Shule River Basin holds a pivotal position in Northwest China and serves as a key land node in the implementation of the national “Belt and Road” strategy [1]. Water resources, as the core support of the life system in arid regions, directly constrain regional development patterns [2,3]. According to the data of Gansu Province Water Resources Bulletin (2000–2022), water demand in the Shule River Basin increased by 11.53% in 2022 compared with 2000 and by 3.30% relative to the multi-year average from 2000 to 2022. In contrast, total water resources during the same period declined by 61.48% and 55.32%, respectively, indicating a sharply intensified imbalance between supply and demand. This pressure has been further exacerbated by two key factors: the high proportion of water allocated to agricultural irrigation and low overall water use efficiency [4]. Concurrently, the basin faces a range of water and ecological challenges, including groundwater over-exploitation, insufficient ecological base flow, continued expansion of artificial oases, and glacier retreats [5]. Therefore, investigating the coupling relationships among water resources, the economy, society, and ecosystems is essential for promoting sustainable water use and advancing high-quality regional development [6,7].

Many scholars have explored the interactions among water resources, society, economy, and ecosystems from various perspectives. Di Baldassarre et al. argued that imbalances in any single subsystem can disrupt the functioning of the entire system, ultimately impeding the sustainable development of the water–environment nexus. They emphasized the importance of adopting a systemic approach to environmental sustainability [8]. Voskamp et al. highlighted how rapid economic growth and irrational resource exploitation have intensified conflicts among water, socio-economic, and ecological subsystems, resulting in increasing incoherence within the water–environment system [9]. Khaizran et al. assessed the impacts of eco-efficiency and socio-economic activities on environmental sustainability by analyzing patterns of resource consumption and production in 19 Asia-Pacific countries, with adjustments for population size [10]. Qu et al. constructed a coupling framework that quantified the coordination among water resources, ecological environment, and socio-economic systems in the lower Yellow River receiving area and predicted the future evolution of system coordination [11]. Tu et al. evaluated the sustainability of the water resource–society–economy–ecosystem system using a machine learning-based clustering algorithm to classify system performance levels [12]. Cui et al. proposed a mechanistic model to identify states of coordination and misalignment, integrating a risk matrix to evaluate development and coordination states, and analyzed the spatial–temporal patterns and causes of uncoordinated development in Anhui Province from 2011 to 2020 [13]. Li et al. introduced the Haken model grounded in synergetic theory and applied a machine learning clustering algorithm to assess the synergistic evolution of the water–socio-economic–ecological composite system in the Yellow River Basin from 2009 to 2022 [14].

Existing studies have primarily examined the coupling and coordination among water resources, socio-economic systems, and the ecological environment by employing models such as coupling coordination analysis, spatial analysis techniques, and machine learning algorithms [15]. These studies have largely focused on measuring spatial–temporal characteristics, identifying driving factors, predicting future trends, and verifying interrelationships across different regions [16]. However, the coupling dynamics and synergistic evolution patterns of these systems in the arid zones of Northwest China remain insufficiently understood. A comprehensive analysis of the multidimensional interactions among water resources, socio-economic development, and ecosystems in arid regions is essential for designing a scientifically grounded water allocation framework, achieving water security, and fostering coordinated socio-economic and ecological development [17]. This study employs the coupling coordination degree model to evaluate the degree of coordination among water resources and social, economic, and ecological systems in the Shule River Basin from 2003 to 2022. The correlation and connectivity among system indicators are subsequently analyzed using a network analysis approach. An obstacle diagnostic model based on indicator deviation is constructed to identify the primary limiting factors within the water resource–society–economy–ecosystem system. Based on these analyses, the study provides a comprehensive depiction of the coupling relationships and synergistic dynamics of water resources, society, economy, and ecosystems in the arid regions of Northwest China.

2. Materials and Methods

2.1. Study Area

The Shule River Basin is an inland river basin located in the arid zone of northwest China, situated in the western segment of the Hexi Corridor, with geographic coordinates ranging from 93°30′ to 98°30′ E and 38°00′ to 42°30′ N [18], as shown in Figure 1. Bordered by the Heihe River Basin to the east, the Tarim Basin to the west, the Qilian Mountains to the south, and the Beishan Mountain System to the north, the basin forms a typical “alpine–oasis–desert” composite geomorphic system. The Shule River flows approximately 945 km from east to west, encompassing a drainage area of about 131,500 km², which accounts for 29.7% of the total area of the Hexi Corridor [19].

Precipitation in the Shule River Basin is highly uneven throughout the year, with the majority occurring between May and August. The average annual precipitation is approximately 80 mm. Water recharge in the basin primarily depends on snow and ice melt from the Qilian Mountains and atmospheric precipitation. Influenced by the constraints of a continental arid climate and complex topography [20], the basin exhibits significant inter-annual variability in water resources, with a multi-year average total volume of approximately 2.47 billion m³. According to the Gansu Province Water Resources Bulletin (2022), surface water resources for that year were estimated at 2.219 billion m³, while groundwater resources totaled 1.058 billion m³. Due to frequent surface–groundwater exchange, the overlapping volume reached 1.02 billion m³, resulting in a total water resource availability of 2.257 billion m³, representing a 7.8% fluctuation compared to the multi-year average.

The Shule River Basin spans five counties and cities in Gansu Province, Dunhuang, Yumen, Guazhou, Subei, Akesai, and part of Qinghai. The study area of this research primarily focuses on the Gansu region. According to the 2022 Gansu Provincial Statistical Yearbook, the total population of these five counties and cities reached 480,900 in 2022, marking a slight increase of approximately 0.04%. Among them, the urban population was 301,900, while the rural population was around 179,000.

Economic development in the Shule River Basin has generally exhibited an upward trend, with a temporary decline in 2014 and a gradual recovery beginning in 2016. In 2022, the regional GDP reached 50.130 billion yuan, with a per capita GDP of 104,200 yuan and a multi-year average GDP growth rate of approximately 12%.

Due to its unique geographic location and climatic conditions, the Shule River Basin faces multiple challenges related to water resources and ecological stability, including water scarcity, intensive water resource development, declining groundwater tables, inadequate assurance of ecological base flow, expansion of artificial oases, and glacier retreat [21].

2.2. Indicator Selection and Weighting Methods

By considering the water supply–demand relationship, development and utilization status, water use efficiency, socio-economic development, and water environment protection in the Shule River Basin, this study analyzes the main factors influencing the water resources–socio-economic–ecosystem system. Guided by the principles of systematicness, representativeness, practicability, and scientific rigor, an indicator system comprising 20 evaluation metrics across four subsystems—water resources, society, economy, and ecosystems—has been constructed [22], as shown in

Table 1. Indicator system for the water resource–society–economy–ecosystem assessment.

Subsystem	Indicator	Unit	Nature
Water Resources	Per capita water resources (m1)	m3/person	Positive
	Water production modulus (m2)	104 m3/km2	Positive
	Water resource utilization rate (m3)	\	Positive
	Annual precipitation (m4)	mm	Positive
	Per capita water storage (m5)	m3/person	Positive
Society	Urbanization rate (m6)	\	Positive
	Resident population (m7)	10,000 people	Positive
	Cultivated land area (m8)	km2	Negative
	Domestic water use per capita (m9)	m3/person	Negative
	Grain output per capita (m10)	kg/person	Positive
Economy	Per capita GDP (m11)	104 yuan/person	Positive
	Water consumption per 10,000 yuan GDP (m12)	m3/104 yuan	Negative
	Water consumption per 10,000 yuan industrial value added (m13)	m3/104 yuan	Negative
	GDP growth rate (m14)	\	Positive
	Proportion of agricultural water use (m15)	\	Negative
Ecosystems	Proportion of ecological water use (m16)	\	Positive
	Groundwater utilization rate (m17)	\	Negative
	Per capita wastewater discharge (m18)	kg/person	Negative
	Wastewater discharge per 10,000 yuan industrial value added (m19)	kg/104 yuan	Negative
	Area of water-saving irrigation (m20)	km2	Positive

.

The weights of the indicators are determined using the entropy weight method, which assigns weights based on the degree of variation among the indicators. This approach minimizes subjective bias and yields more objective and data-driven weight assignments.

The calculation process involves the following steps:

The original data were first standardized using the polar standardization method to eliminate the effects of differing dimensions.

For positive indicators:

$$a_{\mathrm{i}}={\displaystyle \frac{{\mathrm{z}}_{\mathrm{i}}-\min \left({\mathrm{z}}_{\mathrm{i}}\right)}{\max \left({\mathrm{z}}_{\mathrm{i}}\right)-\min \left({\mathrm{z}}_{\mathrm{i}}\right)}}$$

(1)

For negative indicators:

$$a_{\mathrm{i}}={\displaystyle \frac{\max \left({\mathrm{z}}_{\mathrm{i}}\right)-{\mathrm{z}}_{\mathrm{i}}}{\max \left({\mathrm{z}}_{\mathrm{i}}\right)-\min \left({\mathrm{z}}_{\mathrm{i}}\right)}}$$

(2)

where $a_{\mathrm{i}}$ denotes the standardized data of the ith indicator, and z_i denotes the specific value of the i-th indicator.

Calculate the proportion ${\mathrm{y}}_{\mathrm{i}}$ of the $\mathrm{i}$-th indicator within the system, where $\mathrm{n}$ denotes the number of indicators.

$${\mathrm{y}}_{\mathrm{i}}={\displaystyle \frac{a_{\mathrm{i}}}{{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{a}_{\mathrm{i}}}}$$

(3)

The information entropy of each indicator is calculated using the information entropy formula, where ${\mathrm{e}}_{\mathrm{i}}$ denotes the information entropy of the ith indicator.

$${\mathrm{e}}_{\mathrm{i}}=-{\displaystyle \frac{1}{\ln \mathrm{n}}}\sum _{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{y}}_{\mathrm{i}}\ln {\mathrm{y}}_{\mathrm{i}}, \left(0\le {\mathrm{e}}_{\mathrm{i}}\le 1\right)$$

(4)

The i-th indicator’s weight ${\mathrm{w}}_{\mathrm{i}}$ is then computed as follows:

$${\mathrm{w}}_{\mathrm{i}}={\displaystyle \frac{1-{\mathrm{e}}_{\mathrm{i}}}{{\sum }_{\mathrm{j}=1}^{\mathrm{n}}(1-{\mathrm{e}}_{\mathrm{i}})}}$$

(5)

2.3. System Coupling Coordination Degree Model

The coupling coordination degree model is commonly applied to assess the interrelationships among multiple subsystems and to evaluate the level of their coordinated development [23]. The degree of coupling coordination is denoted by D, with higher values indicating stronger system coordination. Based on the value of D value, the level of coordination can be classified into five categories: highly uncoordinated, moderately uncoordinated, weakly coordinated, moderately coordinated, and highly coordinated11, as shown in

Table 2. Classification of coupling coordination degree.

Coupling Coordination Degree D	Type of Coordinated Development
[0, 0.2)	Highly uncoordinated
[0.2, 0.4)	Moderately uncoordinated
[0.4, 0.6)	Weakly coordinated
[0.6, 0.8)	Moderately coordinated
[0.8, 1.0]	Highly coordinated

. The commonly used calculation formula is as follows:

$$\mathrm{D}=\sqrt{\mathrm{C}\mathrm{T}}$$

(6)

where C represents the coupling degree, and T is the comprehensive evaluation index. The general formula is expressed as follows:

$$\mathrm{C}={\displaystyle \frac{\mathrm{n}\sqrt[\mathrm{n}]{\prod _{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{A}}_i}}{\sum _{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{A}}_{\mathrm{i}}}}$$

(7)

$$\mathrm{T}=\sum _{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{w}}_{\mathrm{i}}{\mathrm{A}}_{\mathrm{i}}$$

(8)

where A is the comprehensive evaluation index of the subsystem.

According to the purpose of the study and the system characteristics of each subsystem, linear weighting is adopted to calculate the comprehensive evaluation index of each subsystem A. The specific formula is as follows:

$$\mathrm{A}=\sum _{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{w}}_{\mathrm{i}}{\mathrm{r}}_{\mathrm{i}}$$

(9)

2.4. Indicator Network Analysis Model

The network analysis method visualizes the internal relationships among indicators by representing them as nodes and their correlations as edges. This approach offers significant advantages in quantifying interactions within complex systems and has been widely applied in water resources and ecological research [24].

A sliding window technique was employed to analyze dynamic correlations. The water resource, society, economy, and ecosystem indicators were segmented into five-year time windows, resulting in 16 overlapping windows covering the period from 2003 to 2022 (Windows 1–16). Regression analysis was conducted within each window to evaluate the correlations among indicators. The use of sliding windows simplifies the search space and reduces redundant computations and time complexity [25].

Pearson’s correlation coefficient was used to analyze the relationships between indicators. The absolute value of the correlation coefficient reflects the strength of the correlation between two variables. Only correlations that passed a significance test (p < 0.05) were retained for further analysis. The formula for Pearson’s correlation coefficient is as follows:

$${\mathrm{r}}_{\mathrm{i}\mathrm{j}}={\displaystyle \frac{\mathrm{C}\mathrm{o}\mathrm{v}\left({\mathrm{X}}_{\mathrm{i}},{\mathrm{X}}_{\mathrm{j}}\right)}{{\mathsf{\sigma }}_{\mathrm{i}}{\mathsf{\sigma }}_{\mathrm{j}}}}$$

(10)

where ${\mathrm{r}}_{\mathrm{i}\mathrm{j}}$ denotes the correlation coefficient between indicator i and indicator j, $\mathrm{C}\mathrm{o}\mathrm{v}\left({\mathrm{X}}_{\mathrm{i}},{\mathrm{X}}_{\mathrm{j}}\right)$ is the covariance of indicators i and j, and ${\mathsf{\sigma }}_{\mathrm{i}}{, \mathsf{\sigma }}_{\mathrm{j}}$ denotes the standard deviation of indicators i and j.

The regression model used to estimate the relationships among indicators was constructed using the Least Squares Estimation (LSE) method. LSE minimizes the sum of squared differences between observed and predicted values to estimate the regression coefficients. The estimation formula is expressed as follows:

$${\mathrm{m}\mathrm{i}\mathrm{n}}_{\mathsf{\beta }}={‖\mathrm{Y}-\mathrm{X}\mathsf{\beta }‖}^2$$

(11)

$$\mathsf{\beta }={\left({\mathrm{X}}^{\mathrm{T}}\mathrm{X}\right)}^{-}{\mathrm{X}}^{\mathrm{T}}\mathrm{Y}$$

(12)

where β is the regression coefficient, X^T is the transposition of the indicator series values, and Y is the window year (e.g., 2003–2007).

The coupling relationships between indicators were classified based on the results of correlation and regression analyses, as summarized in

Table 3. Classification of coupling relationship types.

Indicator Attributes	rij	β	Coupling Type
Positive, Positive	>0	>0	Positive coupling
	>0	<0	Negative coupling
	<0	/	Balance coupling
Positive, Negative	<0	>0	Positive coupling
	<0	<0	Negative coupling
	>0	/	Balance coupling
Negative, Negative	>0	<0	Positive coupling
	>0	>=0	Negative coupling
	<0	/	Balance coupling

. Positive coupling indicates a complementary relationship in which an increase in one indicator is associated with an increase in the other, jointly promoting the development of the coupled water resource–society–economy–ecosystem system. A trade-off relationship reflects mutual constraint, where an increase in one indicator corresponds to a decrease in the other, thereby weakening their combined contribution to the system’s coordinated development. Negative coupling refers to a situation in which two indicators are positively correlated and promote each other, yet their interaction negatively affects the overall coordination of the water resource–society–economy–ecosystem system.

The graph package in R was used to perform network analysis [26]. A total of 20 indicators representing the water resource–society–economy–ecosystem system were designated as network nodes, while the coupling relationships between indicators served as edges. Different types of coupling were distinguished by edge color, and the Pearson correlation coefficient was used as the edge weight, with larger coefficients indicating stronger associations between nodes. The node degree, defined as the number of edges connected to a given node, was used to reflect the influence and strength of association of each indicator within the system. The network connectivity α was calculated as the ratio of actual connections to the total number of possible connections in the network, representing the overall intensity of coupling among the indicators. Higher connectivity values indicate stronger interdependence among system components. The connectivity α was computed using the following formula:

$$\propto =\sum {\displaystyle \frac{\mathrm{r}}{\frac{\mathrm{m}\left(\mathrm{m}-1\right)}{2}}}$$

(13)

where α is the connectivity, and m is the number of nodes.

2.5. Barrier Degree Model

The barrier degree model is used to identify the key factors hindering the coordinated development of a system. In this study, a barrier degree diagnostic model based on indicator deviation was constructed to analyze the primary obstacles affecting the water resource–society–economy–ecosystem system in the Shule River Basin [27]. The calculation process is outlined as follows:

Calculate the single-factor contribution F_i. Factor contribution indicates the degree of influence of the evaluation indicator on the overall objective.

$${\mathrm{F}}_{\mathrm{i}}={\mathrm{a}}_{\mathrm{i}}{\mathrm{W}}_{\mathrm{j}}$$

(14)

where W_j is the weight of the criterion layer to which the indicator belongs.

Calculate the indicator deviation P_i, which represents the distance of the calculated evaluation indicator from the water–socio-economic–ecosystem development objective.

$${\mathrm{P}}_{\mathrm{i}}=1-{\mathrm{a}}_{\mathrm{i}}$$

(15)

Calculate the barrier degree K_i. The greater the barrier degree, the greater the impact of the indicator on the development of the coupled water resources–social–economic–ecological system.

$${\mathrm{K}}_{\mathrm{i}}={\displaystyle \frac{{\mathrm{F}}_{\mathrm{i}}{\mathrm{P}}_{\mathrm{i}}}{{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{F}}_{\mathrm{i}}{\mathrm{P}}_{\mathrm{i}}}}$$

(16)

3. Results

3.1. Indicator Weights

The weights of the four subsystems, water resources, society, economy, and ecosystems, and their corresponding indicators, as determined by the entropy weighting method, are shown in Figure 2.

The weight distribution across the four subsystems was relatively balanced during the study period. Among them, the ecological subsystem held the highest weight at 0.303, followed by the social subsystem, the economic subsystem, and the water resource subsystem, which had the lowest weight at 0.221. The overall weighting order was ecological > social > economic > water resources. At the indicator level, the variables with the highest weights included annual precipitation (m4), per capita food production (m10), agricultural water use proportion (m15), ecological water use proportion (m16), and area of water-saving irrigation (m20). Among these, ecological water use proportion (m16) had the highest individual weight of 0.127.

3.2. Coupling Coordination Degree of the System

The results of the comprehensive evaluation index of the four major systems of water resources, society, economy, and ecosystems in the Shule River Basin from 2003 to 2022 are shown in Figure 3. The social, economic, and ecological subsystems exhibit a steady upward trend, indicating continuous improvement in their respective development levels. In contrast, the water resource subsystem shows significant fluctuations due to the influence of natural factors such as climate variability. Based on the trends in the comprehensive evaluation indices, the development process can be divided into three distinct phases. From 2003 to 2013, the water resources and social subsystems remained relatively stable, while the economic subsystem increased from 0.024 to 0.109, and the ecological subsystem rose from 0.056 to 0.104, showing a relatively balanced overall trend. During this period, the average ranking of the subsystem indices was water resources > society > economy > ecosystems. After 2013, the pattern changed significantly. The evaluation indices of the ecological and social subsystems reached higher levels, with the ecological index increasing to 0.155 after some fluctuations. In contrast, the economic and water resource subsystems remained at lower levels. As a result, the ranking shifted to society > ecosystems > water resources > economy. After 2019, the social subsystem showed signs of stabilization, while both the ecological and economic subsystems experienced further improvement. By the end of the study period, the ranking of the comprehensive evaluation indices became ecosystems > economy > society > water resources, reflecting the growing importance of ecological and economic development in the region’s overall coordination dynamics.

The coupling coordination degree of the water resource–society–economy–ecosystem system in the Shule River Basin from 2003 to 2022 is shown in Figure 4. Values ranged from 0.217 to 0.409, indicating a generally moderate level of incoherence. Despite fluctuations, the overall trend was upward, with an average annual growth rate of 0.86%. The evolution of the coordination degree can be divided into three stages. In the first stage (2003–2013), the coordination degree increased steadily from 0.217 to 0.320, with an average growth rate of 1.18%, remaining within the moderately uncoordinated range. In the second stage (2013–2019), a brief decline was followed by a gradual recovery, continuing the upward trend. In the third stage (2019–2022), the coordination degree increased more rapidly, reaching 0.395 with an average growth rate of 1.55%, indicating an improvement over the previous stage. After 2019, the coordination degree exceeded 0.4, transitioning from moderately uncoordinated to weakly coordinated.

3.3. Coupling Relationships Between Indicators

The coupling relationships among the evaluation indicators of the water resource–society–economy–ecosystem system from 2003 to 2022 are presented in Figure 5. Indicators with positive coupling relationships were most prevalent in Window 1 (2003–2007), followed by a sharp decline and a partial rebound, before reaching their lowest level in Window 14 (2016–2020). This trend indicates a dynamic evolution of coupling, decoupling, and re-coupling, with the extent of re-coupling remaining lower than the initial stage. Indicators exhibiting trade-off relationships showed a generally fluctuating upward trend throughout the study period, peaking in Window 11 (2013–2017). A subsequent decline was followed by a gradual increase, a trend that partially mirrors the changes observed in positive coupling relationships. In contrast, negative coupling relationships remained relatively stable and consistently low across most windows. Over time, the dominant interaction pattern among indicators gradually shifted from positive coupling to trade-off relationships, suggesting increasing internal constraints and complexity within the integrated water resource–society–economy–ecosystem system.

The connectivity of the coupling relationships is presented in Figure 6. The results exhibit clear stage-based characteristics consistent with the temporal evolution of the system’s coupling coordination degree. In the first stage, before 2013 (Window 7), positive coupling relationships displayed the highest connectivity and dominated the system, followed by trade-off relationships, while negative coupling relationships remained at the lowest level. In the second stage, the connectivity of trade-off relationships gradually increased, surpassing that of positive coupling relationships, reflecting the growing internal constraints and balancing effects among system indicators. Meanwhile, the connectivity of negative relationships remained low and stable, fluctuating between 0 and 0.02. In the third stage, after 2019 (Window 13), the connectivity levels of positive and negative coupling relationships tended to converge, while trade-off relationships remained at a relatively high level.

3.4. Barrier Factors

Temporal changes in the barrier degree of individual indicators within the water resource–society–economy–ecosystem system of the Shule River Basin from 2003 to 2022 are shown in Figure 7. Substantial variation is observed across both years and indicators. Prior to 2019, the system exhibited a pronounced polarization in barrier degrees, with a small number of indicators, such as the proportion of ecological water use (m16), the proportion of agricultural water use (m15), and per capita food production (m10), contributing between 33% and 56% of the total system obstacle, indicating a highly concentrated hindering effect. After 2019, the distribution of barrier degrees became more balanced, shifting from a few dominant constraints to a broader set of influential indicators. Notably, the influence of per capita water resources (m1), water production modulus (m2), annual precipitation (m4), cultivated land area (m8), and groundwater utilization rate (m17) increased significantly, suggesting a growing and more complex set of constraints affecting the coordinated development of the system.

The average barrier degrees of the evaluation indicators from 2003 to 2022 are presented in

Table 4. Ranking of indicator barrier degrees.

Indicator	Barrier Degree	Rank	Indicator	Barrier Degree	Rank
m16	0.1595	1	m17	0.0398	11
m15	0.124	2	m1	0.0324	12
m10	0.1095	3	m7	0.0281	13
m14	0.0864	4	m5	0.0274	14
m20	0.0641	5	m8	0.0263	15
m6	0.0472	6	m9	0.0244	16
m4	0.0459	7	m12	0.0241	17
m11	0.0458	8	m18	0.0138	18
m3	0.0423	9	m13	0.0107	19
m2	0.0414	10	m19	0.0068	20

. The top five major barrier factors are the ecological water use ratio (m16), agricultural water use ratio (m15), per capita food production (m10), GDP growth rate (m14), and area of water-saving irrigation (m20). Among them, two indicators belong to the ecological, two to the economic, and one to the social subsystem. In terms of the average barrier degree by subsystem, the overall ranking is economic subsystem > ecological subsystem > social subsystem > water resource subsystem.

The four subsystems of water resources, social, economic, and ecosystems can be divided into three phases, as shown in Figure 8. The first phase (2003–2013) is characterized by a slow overall change in barrier degrees. During this period, the ecological subsystem exerted the greatest influence, while the social and economic subsystems fluctuated in their relative impacts. The water resource subsystem consistently exhibited the lowest barrier degree, indicating that ecological indicators were the primary limiting factors for regional carrying capacity. The second phase (2013–2019) showed a significant increase in the barrier degree of the economic subsystem, while the ecological and social subsystems experienced slight declines, and the water resource subsystem showed a modest rise. In the third phase (2019–2022), the obstacle degrees of the ecological, social, and economic subsystems declined markedly, whereas the water resource subsystem experienced a sharp increase.

4. Discussion

4.1. System Coupling Coordination Degree

Except for the water resource subsystem, the other three subsystems show a consistent positive trend, reflecting the basin’s progressive achievements in social services, industrial structure optimization, and ecological and environmental management. In contrast, the water resource subsystem, influenced by climate fluctuations and natural endowments, exhibits significant instability. This instability remains a key factor limiting the coordinated development of the basin system.

The development of the basin’s integrated system has progressed through three stages: initial initiation (2003–2013), coordinated transition (2014–2018), and accelerated integration (2019–2022). This progression suggests that with the establishment of regional ecological compensation mechanisms and the transition to a greener industrial structure, the basin’s overall system coupling coordination is moving towards a higher level. However, the consistently low evaluation index of the water resource subsystem highlights its role as a limiting factor within the coupled system. Strengthening the resource foundation for coordinated development will be essential, including enhancing water resource allocation efficiency, promoting the construction of water-saving societies, and reinforcing mechanisms to ensure sufficient ecological water in the basin.

4.2. Indicator Coupling Relationships

Within the water resource subsystem, a distinct negative coupling characteristic is observed, with relatively high average correlation among nodes, yet its overall impact on the system manifests as a constraining negative influence. In contrast, the social subsystem exhibits weak internal coupling, indicating a stronger independence among its indicators. The economic subsystem demonstrates a high degree of coupling concentration within the overall system, particularly in terms of the tight coupling between per capita GDP and water consumption per 10,000 yuan of GDP with the ecological subsystem. Among the ecological subsystem indicators, the area under water-saving irrigation shows the highest degree of coupling in the system, highlighting its pivotal role as a key nexus in coordinating resource utilization and environmentally sustainable development.

The time series network further reveals the system’s coupling structure changes at different stages, which correlate closely with the evolution of the system’s coupling coordination degree. In the initial startup stage, positive coupling dominates, and the coupling coordination degree increases rapidly. In the second stage, the dominance of positive coupling gradually shifts towards a trade-off relationship, reflecting the system’s transition into a phase of multi-objective trade-offs and regulation. During this stage, the economic and ecological subsystems exert a greater influence on the system’s coupling coordination. In the third stage, the trade-off relationship becomes the dominant force, likely due to the lag in the overall score of the water resource subsystem, which restricts the overall development of the watershed system.

The balance coupling can be regarded as a key diagnostic signal indicating that the system is approaching or has already entered an unsustainable trajectory. It suggests that the intrinsic resilience of the system has been excessively depleted, and the marginal benefits of the existing development model have sharply declined. This foreshadows the potential for the system to become locked into a degenerative state, while also presenting a critical juncture for transitioning—through robust intervention—toward a new phase seeking synergistic gains. Through network analysis and the coupling coordination degree model, this theoretical expectation has been quantitatively identified and confirmed in the current developmental stage of the Shule River Basin.

4.3. Dynamic Evolution

The constraints within the water resource–society–economy–ecosystem system in the Shule River Basin have undergone a dynamic shift from polarization to equilibrium. The high average obstacle degree of the economic and ecological subsystem indicators reflects the basin’s ecological fragility and the high dependence of agriculture on water resources. However, with the advancement of ecological civilization construction and related policies, the dominant barrier factors have gradually shifted toward resource-based indicators. The uncertainty in water conditions and the efficiency of water utilization have emerged as new constraints on the system’s development.

Different stages of the obstacle degree reflect varying characteristics of system coupling, coordination, and the synchronized performance of indicator coupling relationships. In the first stage (2003–2013), the ecological subsystem exhibited the highest obstacle degree, with the social and economic subsystems fluctuating alternately, while the water resource subsystem had a relatively low obstacle degree. During this phase, positive coupling relationships dominated, indicating the primary ecological constraints on the watershed system. In the second stage (2013–2019), driven by policies following the 18th National Congress, the obstacle degree of the economic subsystem rose significantly, while the ecological and social subsystems experienced moderate declines. The degree of the water resource subsystem also increased, leading to a rise in both positive coupling and trade-off coupling relationships within the overall system. In the third stage (2019–2022), driven by the policies of the 19th National Congress on Ecological Civilization, the river-long system, and the “dual-carbon” goals, the obstacle degrees of the ecological, social, and economic subsystems decreased significantly. However, the obstacle degree of the water resource subsystem surged, highlighting the risks associated with water shortages and unequal distribution. This shift underscores the growing importance of water resources in balancing the system, with checks and balances becoming the dominant coupling relationship.

4.4. Limitations

This study systematically evaluated the coupling among water, social, economic, and ecological systems in the Shule River Basin using coupling coordination, network, and obstacle diagnosis models. However, limitations exist. Indicator weights rely solely on the entropy method based on data variability, potentially overlooking real-world significance. Data mainly from yearbooks and bulletins pose accuracy and continuity risks, while historical data limit future climate impact assessment.

Future work should integrate subjective weighting with the entropy method for robust indicators. Crucially, incorporating downscaled CMIP6 projections under RCP/SSP scenarios into dynamic models is needed to quantify climate impacts on water resources and system coupling, supporting adaptive planning. Multi-source data and uncertainty analysis should be enhanced.

The framework is transferable to other arid/semi-arid basins. For application, indicators should be adjusted locally, and region-specific climate scenarios must be included to improve foresight and reliability in cross-basin comparisons and long-term strategy.

5. Conclusions

The overall system coupling coordination degree increased from 0.217 in 2003 to 0.409 in 2022, transitioning from moderately uncoordinated to weakly coordinated. However, due to the volatility of the water resource subsystem, the system’s coordination remains at a low level.

Indicator coupling network analysis reveals that the ecological and economic subsystems act as the key coupling hubs, with water-saving irrigated area (m20) and per capita GDP (m11) serving as core nodes. A significant negative coupling within the water resource subsystem acts as a bottleneck, restricting the system’s coordinated development.

The diagnosis of barrier factors highlights that the proportion of water used for ecological purposes (m16), the proportion of water used for agriculture (m15), and per capita food production (m10) have long been the dominant barrier factors. However, after 2019, the obstacle degree of water resource-related indicators, such as per capita water resources (m1) and water production modulus (m2), increased significantly, underscoring the persistent pressure on water resource availability and utilization efficiency.

The Shule River Basin’s water resource–society–economy–ecosystem system evolved through three distinct stages: in the initial startup phase (2003–2013), the coupling coordination degree rose from 0.217 to 0.32 (+1.18% per year), with positive coupling dominating and ecological factors accounting for over 50% of the total obstacles. In the coordinated transition phase (2014–2018), the coordination degree fluctuated and rose to 0.395, with an increasing proportion of trade-off relations, highlighting economic barriers and emerging water resource constraints. In the accelerated integration phase (2019–2022), the coordination degree exceeded 0.40, with trade-off relationships dominating, while water resource barriers sharply increased, becoming the core bottleneck for further system integration.