The Coupling and Spatial-Temporal Evolution of High-Quality Development and Ecological Security in the Middle Route of South-to-North Water Diversion Project

Abstract

The South-to-North Water Diversion Project constitutes a fundamental initiative designed to enhance water resource distribution and foster regional coordinated development. To investigate the coupling coordination and its spatiotemporal evolution between high-quality development and ecological security (HQD-ES) within the project’s water source areas, this research established a dedicated evaluation index system. Employing coupling coordination, spatial autocorrelation, and Geographically Weighted Regression (GWR) models, the study analyzed the coupled coordination state and its spatiotemporal characteristics across these water source areas for the period 2010–2023. The findings demonstrated that (1) the high-quality development trend remained generally positive, rising from 0.253 to 0.377; ecological safety level showed sustained improvement, increasing from 0.365 to 0.731. (2) The coupling degree (CD) was in a high coupling stage on the whole; the coupling coordination degree (CCD) increased significantly, from imminent imbalance to good coordination state, and the space pattern showed “prominent in the middle and stable in the north and south”. (3) There was no obvious spatial correlation existing between the CCD of HQD-ES in Nanyang City. Tongba, Fangcheng, and Xinye displayed spatial correlation characteristics of low-high aggregation and high-low aggregation. GWR results showed that industrial structure, urbanization, and greening level promoted CCD, while economic level, population density, and environmental regulation inhibited it.

1. Introduction

The Middle Route of South-to-North Water Diversion (SNWD) represents a major water transfer initiative operating across basins, over long distances, and with large flow capacity. As an inter-basin, large-flow, long-distance, and super-large water diversion project, the project has effectively alleviated the constraints of water shortage on the urbanization development in the northern region. As of March 2023, the water supply of Danjiangkou Reservoir to the north had reached 54.814 billion m³, and the water quality had continued to maintain or exceed the Class II standard of surface water, benefiting more than 24 major cities and 200 counties and municipalities along its route, directly serving a population exceeding 85 million.

High-quality development constitutes a fundamental strategy underpinning national modernization, characterized by greater comprehensiveness and equilibrium in socioeconomic progress. Ecosystem health refers to the state that each part of the system can coordinate with each other and coexist in balance, and maintain a certain elasticity and stability to external disturbances. As an inter-basin transfer scheme, the SNWD extracts major water resources from source zones, generating detrimental effects on their socio-economic development and ecological environments. To address this imbalance and promote coordinated progress in the Middle Route Project’s source area, this work analyzes critical factors for securing ecological integrity alongside high-quality development.

This research formulated a multidimensional index framework for high-quality development, integrating water resources, economic, social, and environmental subsystems. Using the Pressure-State-Response (PSR) conceptual model, we concurrently established an ecological security evaluation system. To explore the synergistic linkage and spatiotemporal dynamics between these dual systems within the water source region, coupling coordination modeling, spatial autocorrelation analysis, and Geographically Weighted Regression (GWR) model were employed.

2. Literature Review

2.1. Research on HQD-ES

It was essential to comprehend its core meaning and attributes, including the new development concept as well as the path, structure, and strategy of economic development, to assess high-quality development [1,2,3,4,5]. Based on the people-centered development philosophy and revolving around the theoretical essence of shared development outcomes, a comprehensive indicator system encompassing five dimensions was constructed: income, health, education, sustainable development, and livelihood improvements [6,7]. Ecological security refers to the health and integrity of ecosystems and the ability to adapt to environmental changes in the face of threats such as ecological damage and environmental pollution [8]. Rapid advances in the socioeconomic sphere contributed to ecological problems, for example, vegetation attenuation, land degradation, air pollution, and biodiversity decline had become increasingly prominent, posing a serious threat to local people‘s life and production [9,10,11,12,13]. Based on the PSR model, scholars had built an indicator system to comprehensively analyze the pressure status of urban population, land, resources, and environment [14,15,16]. Multiple studies based on the PSR model conducted evaluations and spatiotemporal analyses of ecological security in various regions across China, including Lake Chaohu, the Shule River Basin, Zhuhai City, and the Nandu River Basin. These studies identified industrial development, urbanization, and socio-economic indicators as the primary factors affecting ecological security and proposed corresponding improvement strategies [17,18,19,20,21].

2.2. Research on Coupling Coordination

The concept of coupling, which traces its roots to physics, denotes a state where two or more systems are interconnected through mutual interaction [22,23,24,25]. Coupling coordination interactions are typically multifaceted and dynamic, encompassing various subsystems and evolving continuously over time. As one of the methods to measure the interaction effect, coupling has been extensively applied in various fields. In recent years, scholars studied the coupling coordination between many systems, including economic development, urbanization, ecological security, and other issues [26,27,28,29,30]. Investigations into the coordinated interaction between economy and ecosystem has been quite extensive [31,32,33]. Concurrently, research on the interplay between the ecological environment and urbanization has established a well-structured foundation in theory [34,35,36]. Multiple studies employing the coupling coordination degree (CCD) model demonstrated an overall improvement or upward trajectory in the coordination between urbanization, economic development, high-quality development, and the ecological environment across diverse regions within China and transnationally. Findings indicated that most regions had reached above-moderate coordination levels by the time of the studies. However, variations exists in the coordination performance among different elements (economic, resource, and environmental). Crucially, regional collaborative policies have proved highly effective in enhancing the overall coordination degree [25,37,38,39,40,41].

2.3. Research on Spatiotemporal Evolution

To explore the spatiotemporal evolution of coupling coordination across diverse systems, a common methodology, CCD calculation, was employed both domestically and internationally, followed by an analysis of evolutionary characteristics using temporal sequences and spatial distribution maps. ESDA is the abbreviation of exploratory spatial analysis. The model method aimed to explore the spatial correlation and dependence of single or multiple variables in different geographical spaces among various regions within the study area. It was based on the first and second laws of geography [42,43]. The ESDA method mainly uses the spatial autocorrelation index to realize its spatial correlation measure. The prevailing spatial autocorrelation metrics employed were Moran’s indices (global Moran’s I and local Moran’s I) alongside Geary’s C coefficient. The former primarily assessed global spatial patterns across the entire study area, whereas the latter focused on identifying localized spatial configurations within sub-regions [44]. It had applications in the environment, population density, and other aspects [45,46,47,48]. Through analysis using Moran’s I index and LISA cluster maps, researchers revealed the spatial differentiation patterns of high-quality development coordination, urbanization–ecological system relationships, and digital economy development within the Yangtze River Delta, Beijing–Tianjin–Hebei region, and the Yangtze River Economic Belt. The analysis identified significant internal spatial agglomeration and a “high-central, low-periphery” distribution pattern across these regions, reflecting the uneven spatial development across China’s key regions [35,49,50]. Some scholars implemented the GWR to examine spatial analysis in coupling coordination relationship across different fields [51,52,53,54]. The GTWR model analysis revealed the spatiotemporal heterogeneity patterns of the various systems. It verified the interactive effects of spatial dependence and temporal non-stationarity. High-value areas clustered predominantly in core areas, such as provincial capitals. Key temporal nodes triggered potential turning points in trends [55,56].

2.4. Literature Summary

When evaluating high-quality development, many scholars often only considered the economic development of cities in the region or other unilateral situations in the impact evaluation of water source areas and did not comprehensively evaluate the impact of water source areas. Methodologically, the entropy weight-TOPSIS model demonstrated a comprehensive assessment architecture with robust outcome precision. Therefore, this research employed the entropy weight-TOPSIS approach to assess the high-quality development and ecological security (HQD-ES) status within the water source region, combined with a coupling coordination and GWR model to analyze their coupling coordination relationship over the period 2010–2023, along with an analysis of influencing factors. As a critical resource, water availability, demand, and environmental conditions progressively had a significant influence on the stable growth of national and regional economies. Studying the coupled coordination dynamics between water resources systems and other systems was of great help for us to comprehend the relevant development status of our cities and regions and formulate reasonable development plans to take advantage of water resources. Consequently, this research prioritized analyzing the spatiotemporal coordination between HQD-ES, aiming to furnish scientifically grounded recommendations for achieving sustainable and harmonious regional development.

3. Construction of HQD-ES Index System

3.1. High-Quality Development Index System

High-quality development called for a transition from an extensive model of growth toward an intensive one, focusing on green, innovative, coordinated, open, and shared development. Consequently, to capture its essence, corresponding indicators were selected covering the water resources, economic, social, and environmental systems. The resulting index system was presented in

Table 1. High-quality development evaluation index system.

System	Index	Unit	Indicator Type
Water resources system	Water resources	Billion m3	+
	Precipitation	mm	+
	Per capita water resources	m3	+
	Yield of groundwater	Billion m3	-
Economic system	Proportion of the secondary industry to GDP	%	-
	Per capita GDP	CNY	+
	Water consumption Per 10,000 CNY of GDP	m3	-
	Total retail sales of social consumer goods	Billion	+
Social system	Urbanization rate	%	+
	Engel’s coefficient of urban residents	%	-
	Engel’s coefficient of rural residents	%	-
	Per capita domestic water consumption	L/d	+
	Population density	People/km2	-
Environmental system	Green coverage rate in built-up areas	%	+
	Harmless treatment capacity of household waste	10,000 tons	+
	Sewage discharge	10,000 m3	-
	Road cleaning area	10,000 m2	+

+ indicates that this indicator system is positive, while - indicates that it is negative.

.

3.2. Ecological Security Evaluation Index

The PSR model focused on the interactions between the ecological environment and human activities. The selection of specific indicators took into account the prevailing ecological security conditions within the water source area and adhered strictly to principles demanding scientific validity, systematic coherence, practical operability, and indicator independence. The resulting index system was presented in

Table 2. Ecological security evaluation index system.

System	Index	Unit	Indicator Type
Pressure	Application amount of agricultural fertilizers	t	-
	Application amount of agricultural plastic film	t	-
	Pesticide application rate	t	-
	Per capita urban road area	m2	-
	Per capita daily water consumption	L	-
	Electricity consumption per 10,000 CNY of GDP	KWh/10,000 CNY	-
	Energy consumption per 10,000 CNY of GDP	Tons of standard coal/10,000 CNY	-
State	Per capita park green area	m2	+
	Grain yield per unit area of cultivated land	kg/m2	+
	Investment in fixed assets	Billion CNY	+
	Per capita disposable income of urban residents	CNY	+
Respond	Sewage treatment rate	%	+
	Harmless treatment rate of household waste	%	+
	Proportion of the tertiary industry to GDP	%	+

+ indicates that this indicator system is positive, while - indicates that it is negative.

. The indicator weight was shown in Figure 1.

3.3. Data Source

The total amount of water resources, precipitation, per capita water resources and groundwater supply used in this study were all from Nanyang Water Resources Bulletin. The secondary industry accounted for GDP, GDP per capita, total retail sales of social good, urbanization rate, Engel coefficient of urban residents, Engel coefficient of rural residents, per capita daily water consumption, population density, greening of built-up areas, harmless treatment of domestic waste, sewage discharge, road cleaning and cleaning area, agricultural chemical fertilizer usage quantities, plastic film usage quantities, pesticide application intensity, per capita daily domestic water consumption, per capita urban road area, electricity consumption per 10,000 CNY of GDP, energy consumption per 10,000 CNY of GDP, the per capita park green space area, grain yield per unit area of cultivated land, per capita disposable income of urban residents, the investment in fixed assets of the whole society, the harmless treatment rate of domestic waste, the sewage treatment rate, and the proportion of the tertiary industry in GDP were all from Nanyang Statistical Yearbook.

4. Research Methods and Models

4.1. Entropy Weight-TOPSIS Method

The calculation process of entropy weight-TOPSIS method was as follows:

(1)

Standardization:

Positive indexes:

$${{\displaystyle X}}_{ij}={\displaystyle \frac{{{\displaystyle x}}_{ij}-\min ({{\displaystyle x}}_{1j},{{\displaystyle x}}_{2j},\dots ,{{\displaystyle x}}_{nj})}{\max ({{\displaystyle x}}_{1j},{{\displaystyle x}}_{2j},\dots ,{{\displaystyle x}}_{nj})-\min ({{\displaystyle x}}_{1j},{{\displaystyle x}}_{2j},\dots ,{{\displaystyle x}}_{nj})}}$$

(1)

Negative indexes:

$${{\displaystyle X}}_{ij}={\displaystyle \frac{\max ({{\displaystyle x}}_{1j},{{\displaystyle x}}_{2j},\dots ,{{\displaystyle x}}_{nj})-{{\displaystyle x}}_{ij}}{\max ({{\displaystyle x}}_{1j},{{\displaystyle x}}_{2j},\dots ,{{\displaystyle x}}_{nj})-\min ({{\displaystyle x}}_{1j},{{\displaystyle x}}_{2j},\dots ,{{\displaystyle x}}_{nj})}}$$

(2)

In the formula, it was the j indicator value of the first i city before standardization, the j index value of the standardized i city, n represented the number of cities, and m represented the number of indicators. Within the formula:

(2)

Calculation of the proportion of indicators:

P_tij was set as the proportion of the j indicator in the j index of the i city in the t year. The calculation formula was

$${{\displaystyle P}}_{tij}={{\displaystyle X}}_{tij}/{\displaystyle {\sum }_{t=1}^i{\displaystyle {\sum }_{i=1}^m{{\displaystyle x}}_{tij}}}(j=1,2,\dots n)$$

(3)

where m characterized the number of evaluation city unit i.

(3)

Information entropy:

$$E_j=\ln {\displaystyle \frac{1}{n}}{\displaystyle \sum _{i=1}^n\left(\frac{X_{ij}}{{\displaystyle \sum _{i=1}^n{X}_{ij}}}\ln \frac{X_{ij}}{{\displaystyle \sum _{i=1}^n{X}_{ij}}}\right)}$$

(4)

(4)

The weight of each evaluation index:

$$W_j=(1-E_j)/{\displaystyle \sum _{j=1}^m(1-E_j)}$$

(5)

(5)

Calculation of the comprehensive index U_ij.

Let U_ij be the composite indicator in the i city of the t year.

$${{\displaystyle U}}_{ij}={\displaystyle {\sum }_{j=1}^m({{\displaystyle W}}_{ij}}{{\displaystyle X}}_{tij})$$

(6)

where X_tij was the normalized data.

4.2. Research Model

(1)

Coupling degree model

$$C=2\sqrt{{(E+Q)(E\times Q)}^2}$$

(7)

C was the coupling degree (CD), which could measure the CD between HQD-ES, and the value range was (0, 1). E and Q represented the comprehensive score index of the two systems of HQD-ES, respectively.

(2)

Development degree model

$$T=aE+bQ$$

(8)

T was the development degree, which represented the contribution of the overall synergy, and value range was (0, 1); a and b were the weight values of each subsystem, and a + b = 1. This study considered the two systems to be of equivalent significance, so a = b = 0.5; E and Q represented the comprehensive score index of HQD-ES, respectively.

(3)

Coordination degree model

$$D=\sqrt{CT}$$

(9)

D was the degree of coordination, which could judge whether there was benign coordination in the relationship of mutual coupling. Values fell within the range (0, 1), where higher values corresponded to superior coordination.

The specific division conditions of CD and CCD were provided in Appendix A.

4.3. Spatial Autocorrelation Analysis

Spatial autocorrelation measures the interdependence of values for a single variable observed at different geographical locations, acting as an indicator of spatial clustering. To investigate the temporal and spatial changes in CCD within multiple water source areas, this research utilized both global and local autocorrelation models.

(1)

Global spatial autocorrelation

Global spatial autocorrelation described the comprehensive spatial configuration of attribute values throughout the study area. This method examined whether the value of a spatial variable displayed correlation with those at adjacent locations. Commonly, this analysis concentrated on the spatial distribution of a specific attribute among regional spatial units.

$$I(d)={\displaystyle \frac{n}{{\displaystyle \sum _{i=1}^n{\displaystyle \sum _{j\ne i}^n{W}_{ij}}}}}\times {\displaystyle \frac{{\displaystyle \sum _{i=1}^n{\displaystyle \sum _{j\ne i}^n{W}_{ij}}}(X_i-\overline{X})(X_j-\overline{X})}{{\displaystyle \sum _{i=1}^n{(X_i-\overline{X})}^2}}}$$

(10)

W_ij was the spatial connection matrix between spatial units i and j, n was the total number of spatial units, X_i was the attribute value of spatial unit i, Xj was the attribute value of spatial unit j, and $\overline{X}$ was the average value of spatial unit attributes. The value range of I was from −1 to 1: when I = 0, the representation space was not related; when I was positive, it was a positive correlation; when I was negative, it was a negative correlation.

(2)

Local spatial autocorrelation

Local spatial autocorrelation represented the correlation of a specific attribute between each spatial position and its surrounding adjacent positions within the study area. Global spatial autocorrelation assumed spatial homogeneity, implying that only a single trend pervaded the entire region. In reality, spatial heterogeneity was more common, and local spatial autocorrelation offered a more precise method for identifying the heterogeneous characteristics of spatial elements. The local Moran’s I index, a statistical measure for a given spatial unit i, was defined as follows:

$$I_i={\displaystyle \frac{{\displaystyle \sum _{i=1}^n{\displaystyle \sum _{j\ne i}^n{W}_{ij}}}\times (X_i-\overline{X})(X_j-\overline{X})}{\frac{1}{n}{\displaystyle \sum _{i=1}^n{(X_i-\overline{X})}^2}}}$$

(11)

(3)

GWR analysis

The GWR model, a spatial statistical method, captured spatially varying relationships by estimating local regression parameters at each location. This was achieved through weighted least squares regression using neighboring observations. The approach enabled the exploration of spatial variation patterns and facilitated the modeling of non-stationary relationships inherent in spatial data. Consequently, this study employed the GWR method to investigate the spatial heterogeneity of coupling coordination.

$$Y_i={\beta }_0\left(u_i,v_i\right)+{\sum }_{k=1}^n{\beta }_k\left(u_i,v_i\right)x_{ik}+{\epsilon }_i, i=1,2\dots n$$

(12)

$Y_i$ was the global dependent variable, $x_{ik}$ was the independent variable, ${\beta }_0\left(u_i,v_i\right)$ was a constant term, $\left(u_i,v_i\right)$ was the spatial coordinates of the i area, and ${\beta }_k\left(u_i,v_i\right)$ was the variable parameter of $x_{ik}$.

5. Results and Analysis

5.1. HQD-ES Level

5.1.1. High-Quality Development Analysis

The high-quality development level was measured comprehensively across water resources, economy, society, and environment. Utilizing a framework of 17 indicators, this study evaluated this level across the water source area. It revealed disparities in high-quality development among the thirteen counties and cities within the study area from 2010 to 2023 (Figure 2). The findings revealed a general upward trend in the high-quality development indices across all counties, rising from 0.253 to 0.377. In 2023, Wolong and Wancheng had higher levels of high-quality development, with comprehensive indexes of 0.685 and 0.624, respectively. The reason was that Wolong and Wancheng districts implemented multi-tiered initiatives, including provincial special funds for high-quality manufacturing development, municipal interest subsidy programs for innovation, and district-level star-rated enterprise evaluation incentives to stimulate progress. The lowest level in Sheqi was 0.281. In 2010, the areas with higher levels were Xixia, Wolong, and Wancheng, which were 0.379, 0.376, and 0.357, respectively. Simultaneously, Sheqi recorded the minimum value of 0.156. After 2015, the comprehensive index in all counties and districts increased. The “Nanyang Support for Sci-tech Innovation Policy List”, the “Nanyang Reward Measures of the Cultural Tourism Industry for High-Quality Development” and the “Nanyang Several Measures to Further Boost Consumption and Expand Domestic Demand” collectively boosted regional high-quality development.

Significant disparities existed in the level of high-quality development across different regions of Nanyang City, and the spatial characteristics of central prominence were gradually formed. Areas exhibiting high levels were primarily located in the urban center of Nanyang. As a development hub, it enjoyed the best resources in economy, culture, science, and education. This enabled it to provide momentum for high-quality development, drove the growth of surrounding cities to some extent, and contributed to a gradual decrease in internal disparities in high-quality development levels within the study region. The medium-level areas were Xichuan and Xixia. The source of water for the SNWD was Xichuan, which accelerated the economic development and industrial structure to be greener. Adjacent to Xichuan, Xixia relied predominantly on agriculture and steadfastly adhered to the development philosophy of “ecologizing the economy and economizing the ecology”. Regarding spatial pattern, high-quality development showed the structural features of “the highest in the central zone, the relatively high in the western zone, and the lowest in the eastern zone”.

5.1.2. Ecological Security Analysis

From the trend of the comprehensive index of ecological security in the water source area, the ecological security composite index for counties and districts displayed a general upward trajectory with fluctuations in 2010–2023 (Figure 3). The composite ecological security index climbed from 0.365 in 2010 to 0.731 by 2023, showing a good development model. Among them, the ecological status of Wolong, Wancheng, and Xichuan was at the highest level, with the means of 0.649, 0.645, and 0.579, respectively. Among them, the largest increase in the comprehensive index was Neixiang and Tanghe. From 2010 to 2015, the growth rate of each county slowed down. Between 2016 and 2023, the comprehensive ecological security index for every county rose steadily. The outcomes were largely attributable to the execution of “Nanyang Municipal Environmental Protection Comprehensive Rectification Plan” and “Three-Year Implementation Scheme for Steady Ecological Quality Improvement”. Spatially, counties with lower ecological security levels were primarily located in the southern and northeastern parts of the region, including Fangcheng, Tanghe, and Xinye. Conversely, counties exhibiting higher ecological security levels within the water source area, such as Wolong, Wancheng, and Xichuan, were mainly situated in the central urban core and the western and southeastern zones. Overall, the ecological environment showed improvement moving from east to west. In summary, the water source area’s ecological security level displayed a distinct spatial pattern: highest in the central, relatively high in the west and southeast, and lowest in the south.

5.2. CCD Analysis

Applying the CCD model, we assessed the coupling level and coordination state between HQD-ES in the water source area during 2010–2023. The analysis revealed progressive enhancement in the HQD-ES coupling degree throughout the study period, but the increase did not change much, staying mainly in a high coupling stage. With a mean CD of 0.9933, Wolong exhibited the highest level and consistently remained in a state of high coupling. Substantial changes were observed in its CCD type over time, including good coordination, intermediate coordination, primary coordination, marginal coordination, and finally imminent imbalance. As time went on, the coupling coordination stage changed from imminent imbalance to good coordination. Compared with 2010, the research area in 2023 was mainly dominated by primary coordination. Wolong and Wancheng were in a good coordination state, and the overall CCD trend was good.

Based on shifts in coupling coordination types and their numerical evolution, the spatial evolution patterns of the CCD linking regional HQD-ES in the water source area were visualized for representative years (Figure 4). In the early stage, the coordination degree was generally low, largely because economic development was mainly based on industry and agriculture, the level of economic development quality was low, the energy demand and consumption were large, the capacity of resource development and utilization was limited, the pollutants were discharged in large quantities, and the lack of understanding of environmental protection caused the ecological environment to be affected by economic development. Therefore, the two systems exhibited low coordination initially. However, societal development, technological progress, and accelerated construction of an environmentally friendly society led to an overall improvement in their coordination after 2015. In general, Wolong and Wancheng had the highest coordination degree. As the central urban areas, they had made rapid progress in social development and scientific and technological level, which had promoted HQD-ES coordination. As a county-level region, Tanghe’s economic development was mainly based on agriculture, which would cause a certain degree of non-point source pollution and affect the environment, resulting in the lowest CCD. In general, the CCD between counties and districts was on the rise, indicating positive progress in urban development.

To quantify spatial heterogeneity in CCD between systems, trend surface analysis was implemented across the water source area during 2010–2023. The X axis represented east, while the Y axis denoted north. The curve corresponded to the CCD variation fit line. (Figure 5). Through Arcgis, the trend line of coordination degree in typical years was drawn, and the differences in regional coordination degree in time and space were analyzed. From the trend of the curve, the spatial distribution characteristics of the CCD in the north–south direction changed from “high in the north and south, low in the south” to “prominent in the middle and balanced in the north and south”, and these spatial characteristics tended to be stable with the passage of time; in the east–west direction, the coordination degree in the early stage of the study showed the characteristics of “high in the west and low in the east”, and gradually tended to be gentle with the passage of time, indicating that the disparity was slightly alleviated and the CCD of HQD-ES gradually narrowed, and the coordination between the zones in the east–west direction was improved.

5.3. Spatial Correlation Analysis of CCD

5.3.1. Global Autocorrelation Analysis

Spatial correlation analysis employed Geoda software 1.20 to calculate annual CCD values for Nanyang City’s HQD-ES systems. This computation utilized Queen contiguity spatial weights matrix and calculated the global Moran’s I index, with its significance verified through Z value and p value. The values are compiled in

Table 3. The global Moran‘s index of CCD in 2010–2023.

Year	Moran’s I	p	Z
2010	−0.1264	0.427	−0.2295
2011	−0.1780	0.344	−0.4978
2012	−0.1937	0.293	−0.6103
2013	−0.1848	0.315	−0.5690
2014	−0.2076	0.253	−0.7075
2015	−0.2101	0.244	−0.7191
2016	−0.1769	0.335	−0.5211
2017	−0.2064	0.256	−0.6832
2018	−0.1619	0.377	−0.4391
2019	−0.1231	0.483	−0.2065
2020	−0.1193	0.468	−0.1809
2021	−0.0878	0.423	−0.0053
2022	−0.1182	0.498	−0.1711
2023	−0.1667	0.369	−0.4617

.

It could be seen that all p values from 2010 to 2023 were greater than 0.1 and | Z | was less than 1.65, indicating that the global Moran‘s index of the CCD between the two systems in the study area did not pass the significance test during the study period. The Moran‘s index from 2010 to 2023 was less than 0, indicating that similar features were often distributed in distant locations, and the characteristics of adjacent locations were quite different, which was manifested as spatial dispersion. The CCD between the two systems during this period displayed a negative spatial correlation as a whole. Therefore, there was no obvious spatial correlation between the two systems.

To better illustrate the spatial distribution characteristics of CCD between HQD-ES, Moran’s I scatter plots for the years 2010, 2013, 2016, 2019, and 2023 were generated using GeoDa software (Figure 6). The plots revealed that points in the second and fourth quadrants significantly outnumbered others, with the second quadrant being dominant. Over time, the CCD exhibited a negative correlation within the region and displayed attributes of random distribution. This pattern indicated the presence of spatial heterogeneity and overall fluctuating development, primarily attributable to the absence of strong spatial polarization effects driven by geographical disparities in regional coupling coordination.

5.3.2. Local Spatial Autocorrelation Analysis

The global spatial autocorrelation reflected the correlation characteristics of the coordination degree in the spatial range, and it could not reflect the local spatial aggregation type. Therefore, through the use of Geoda software, Moran scatter plots, LISA clustering maps, and spatial saliency maps were generated according to the Queen spatial weight matrix. The aggregation types were classified into four distinct classes: low-low (L-L), low-high (L-H), high-low (H-L), and high-high (H-H) aggregation. To further analyze the significance of local spatial correlations between the HQD-ES coupling and coordination levels in Nanyang City, the LISA clustering diagram was utilized for analysis. The LISA charts for 2010, 2013, 2016, 2019, and 2023 were selected, as shown in Figure 7.

It could be known from the above figure that the cities displaying spatial relationships on the LISA cluster diagram indicated that they had passed the significance test of the local Moran’s index. In 2010–2023, there were L-H aggregation and H-L aggregation, and the significant areas were Tongbai, Fangcheng, and Xinye. Tongbai exhibited an L-H aggregation, where its low CCD neighbored areas with high coordination levels, resulting in an uneven spatial distribution. However, interactions and connections existed between these regions. The low-coordination areas had opportunities to learn from and develop toward high-coordination counterparts, which partially alleviated the polarization effect. The H-L aggregation showed that the areas were Fangcheng and Xinye. Specifically, the H-L aggregation suggested that the CCD of Fangcheng and Xinye were high, while those of their adjacent cities were low. Due to insufficient inter-regional coordination, this type of aggregation area was susceptible to polarization, which failed to generate significant spillover effects for less coordinated neighboring cities.

Spatial statistical analysis confirmed clustering patterns in the HQD-ES coupling coordination across the study region. Local spatial autocorrelation was characterized by L-H and H-L cluster types, with significant heterogeneity.

5.3.3. GWR Model Results Analysis

To further analyze the spatial heterogeneity of the CCD between high-quality development and ecological security within the study area, this study examined key factors influencing their coupling coordination. Specific indicators included industrial structure, economic level, urbanization, population density, green level, and environmental regulation (

Table 4. The index system of influencing factors in the HQD-ES coupling coordination.

Influencing Factor	Indicator Calculation	Unit
Industrial Structure	Proportion of Secondary/Tertiary Industry in GDP	%
Economic Level	Per Capita GDP	CNY
Urbanization	Urban Population to Total Population Ratio	%
Population Density	Permanent Residents per Square Kilometer	persons/km2
Greening Level	Per Capita Park Green Space Area	m2
Environmental Regulation	Sewage Treatment Rate	%

). Using the GWR model, we analyzed the drivers behind spatial variations in Nanyang City’s CCD. The CCD served as the dependent variable, while the six aforementioned influencing factors were treated as independent variables.

To provide a more intuitive visualization of the spatial variations in how each influencing factor affected the CCD, regression coefficients for the six factors were categorized and mapped using ArcGIS software 10.8.2. The resulting spatial patterns of influence are illustrated in Figure 8. Overall, industrial structure, urbanization, and green level exerted positive effects on Nanyang City’s CCD, whereas economic level, population density, and environmental regulation exerted negative effects.

In the western districts, Xixia, Xichuan, and Neixiang, industrial structure exerted the strongest positive effect on enhancing the CCD (Figure 8a). These areas possessed favorable ecological foundations but were relatively less developed economically, with relatively homogeneous industrial structures. Conversely, in Tongbai, Sheqi, and Tanghe, industrial structure yielded a comparatively lower positive effect on CCD. These regions featured strong agricultural foundations and relatively mature economies, where the scope for further optimization of the industrial structure was more limited. Consequently, the marginal coordination benefits derived from changes in industrial structure diminished.

Within the studied area of Nanyang City, an increase in economic development level significantly inhibited the enhancement of the CCD (Figure 8b). This finding was counterintuitive to the conventional theoretical expectation that economic growth promotes coordination. Furthermore, this negative impact was not uniform but exhibited significant spatial heterogeneity. In counties like Xinye, Tanghe, and Tongbai, the conflict between economic development and ecological protection became exceptionally acute due to associated resource consumption and environmental pollution. Consequently, economic development exerted the strongest negative effect on CCD in these areas. While economic growth still posed potential environmental pressures elsewhere, its adverse impact on CCD was comparatively lower in regions such as Xixia and Xichuan. This resulted in a relatively diminished negative effect associated with per capita GDP growth in those western counties.

Urbanization demonstrated the strongest positive effect on enhancing CCD in Xixia, Xichuan, and Neixiang (Figure 8c). This outcome stemmed from substantial policy backing linked to their ecologically sensitive status and clearer pathways for realizing ecological value. In contrast, urbanization provided weaker impetus in Tongbai and Sheqi. This limitation was primarily constrained by restricted development space under ecological conservation mandates, lagging transformation of traditional industrial structures, and relatively limited targeted policy and financial support. This pattern confirmed that urbanization’s impact on regional sustainable development displays high “Place-Context Dependency”, necessitating full consideration of spatial heterogeneity in geographical location, development stage, policy environment, and industrial structure.

In ecologically sensitive and fragile areas like Nanzhao and Fangcheng, population pressure pushed against the ecological security baseline (Figure 8d). Simultaneously, the local economy lacked sufficient momentum for transformation to effectively mitigate this pressure, resulting in significant negative effects. In regions such as Xichuan, Dengzhou, and Xinye, comparatively superior natural conditions, stronger economic capacity, and robust external policy support created more possibilities and greater room for maneuvering. This allowed for buffering population pressure and seeking a balance between development and conservation, thereby mitigating the adverse impacts of population density to a greater extent.

In the plain–hill areas, specifically Tanghe and Tongbai (Figure 8e), where ecological conditions were relatively weak, the need for ecological improvement driven by development was urgent, and significant potential existed for enhancing vegetation cover, afforestation efforts could rapidly and significantly improve the coordination level between development and ecological preservation. In contrast, in the deep mountainous counties of Xixia and Neixiang, where ecological conditions were excellent, vegetation coverage approached saturation, and development faced strong ecological constraints, marginal improvements in greening levels contributed relatively little to enhancing the coordination level.

In core water source protection zones such as Xixia, Xichuan, and Neixiang, the development pressure signified by increased wastewater treatment rates, along with its associated potential ecological risks, significantly exceeded the direct environmental benefits it brought (Figure 8f). This imbalance became the primary constraint on the coordination level. In areas with more urgent development needs and relatively weak environmental governance infrastructure, like Tanghe and Tongbai, wastewater treatment served as a fundamental environmental management investment. Its positive effects, improving environmental quality and supporting development, were far more pronounced. Consequently, its negative impact on the coordination level was minimal.

5.4. Discussions

The ecological security index constructed in this study preliminarily reveals the ecological pattern characteristics of the research area. However, to enhance its practical utility, subsequent research should prioritize deepening and expanding the following key directions:

(1)

Deepening the quantitative assessment of hydrological processes and infrastructure impacts. The region’s ecology is profoundly influenced by water conservancy facilities and climate change. Current models require more systematic integration of key hydrological data, including variations in river discharge, the severity and duration of drought events, and regional water budget changes. Crucially, it is essential to clarify the cascading effects of operational releases from upstream reservoirs on downstream flow regimes, water availability, and ecosystems [57].

(2)

Integrating the driving mechanisms of terrain and climate complexity on ecological patterns. The research area constitutes a typical mountainous water source region. Variations in mountain elevation and complex weather patterns fundamentally shape ecological characteristics. Future research must explicitly incorporate elevational differences and utilize specialized hydrological methods to investigate how regional precipitation patterns and moisture transport pathways influence water resource distribution. This, in turn, drives the spatial heterogeneity of ecological patterns and functions [58].

(3)

Exploring water diplomacy and collaborative governance mechanisms under large-scale water conservancy projects. Large water conservancy projects within or spanning the research area not only impact ecology and hydrology but also involve complex issues of water allocation across upstream–downstream relationships, transboundary regions, and even internationally. Future research should introduce the concept of “water diplomacy”. It needs to explore how ecological security assessment results can be effectively integrated into collaborative water resource decision-making. Key research focuses on establishing effective cross-regional coordination mechanisms, benefit compensation schemes, and risk-sharing strategies. This aims to balance project benefits, potential ecological disruption risks, and diverse regional needs, ensuring the sustainability of the ecological security pattern and equitable regional development [59,60].

6. Conclusions and Recommendations

6.1. Conclusions

An evaluation index system was established for both HQD-ES within the water source area. The CCD model was then adopted to quantitatively assess the CCD across the study area from 2010 to 2023. Subsequently, spatial trend surface analysis, spatial autocorrelation and GWR were applied to explore the spatiotemporal evolution characteristics and patterns of coupling coordination of HQD-ES. The principal conclusions derived were as follows:

(1)

In the study area from 2010 to 2023, the level of high-quality development showed a slow upward trend, and the level of ecological security maintained an upward trend. In terms of spatial pattern, high-quality development showed the structural characteristics of “the highest in the central zone, the relatively high in the western zone, and the lowest in the eastern zone”. The level of ecological security displayed the pattern characteristics of “highest in the central, relatively high in the west and southeast, and lowest in the south”.

(2)

The CD increased from 2010 to 2023, but the increase was small, and the whole was in a highly coupled stage. The mean value of CD in Wolong was the highest, 0.9933, which was always a state of high coupling during this time. The average value of CCD increased significantly, from imminent imbalance to good coordination state. Compared with 2010, the research area in 2023 was mainly dominated by primary coordination. Wolong and Wancheng were in a good coordination state, and the overall CCD trend was good. The spatial pattern exhibited characteristics of “prominent in the middle and stable in the north and south”. The gap in coordination degree along the east–west direction gradually narrowed, and the overall coupling coordination level of central cities was higher than that of county-level cities.

(3)

In general, no obvious spatial correlation existed between the CCD of HQD-ES in Nanyang City. Tongba, Fangcheng, and Xinye displayed spatial correlation characteristics of L-H aggregation and H-L aggregation, which indicated that spatial heterogeneity in the study area was more pronounced than homogeneity. According to GWR results, the coupling coordination degree of Nanyang City was significantly positively affected by industrial structure, urbanization, and greening level, but significantly negatively affected by economic level, population density, and environmental regulation.

This study found that Nanyang’s central district exhibited high development quality and strong ecological security, serving as the primary driver for sustainable regional development. Although regional coordination improved overall, the pace of enhancement was slow, and the development gap areas required further reduction. Furthermore, the level of development coordination varied considerably across localities. This spatial heterogeneity indicated a need for tailored approaches based on specific local conditions.

6.2. Recommendations

Due to varying resource abundance, spatial structures, socioeconomic development foundations, policy directions, and other factors, significant differences existed in the comprehensive development levels and coupling coordination between HQD-ES in the SNWD Middle Route Project’s water source area. To achieve higher-level coordinated development, the following measures were propounded:

(1)

Central Region’s Leading Role as Demonstration Zones. Key central areas, such as Wolong, served as demonstration zones. Their integrated experiences in green industrial transformation, ecological management, and highly efficient resource utilization were summarized to create a replicable model for broader applications.

(2)

Implementation of an Eastern Advancement Plan. Targeted initiatives addressed underdeveloped areas in eastern Nanyang City. Infrastructure investment increased, green industries were fostered, and dedicated technical assistance and industrial transfer channels with the central region were established. This specifically enhanced economic performance and resource efficiency in these areas.

(3)

Establishment of a City–County Collaboration Network. Utilizing the strong development strengths of central cities (Nanyang’s core urban area), a cooperation network covering surrounding county-level cities was built. Platforms for ecological compensation, technology sharing, and joint monitoring were created, which facilitated the flow of resources like talent, technology, and capital from the central city to the counties, promoting their shared development.