Integrated Diagnosis of Water Environment Security and Restoration Priorities in the Dongting Lake Basin, 2000–2020

Abstract

With the rapid advancement of industrialization and urbanization, the systematic assessment of water environment security in lake-type basins and the identification of key restoration zones have become critical scientific challenges for sustainable watershed management. This study focused on the Dongting Lake Basin, where a comprehensive evaluation system comprising 24 indicators was developed based on the Driving forces–Pressure–State–Impact–Response model. Indicator weights were determined using the entropy method. An obstacle degree model was applied to quantitatively identify the primary factors constraining water environment security. Additionally, spatial autocorrelation analysis was introduced to examine spatial dependency characteristics, enabling the delineation of priority restoration areas. The results demonstrated the following: (1) During 2000–2020, the Dongting Lake Basin exhibited significant spatial heterogeneity, with higher water environment security risks in the southeastern region, while the central-eastern region showed a continuous improvement trend. (2) Quantitative analysis identified the core obstacle factors affecting regional water environment security: wastewater treatment capacity (obstacle degree: 16.87%), ecological water use proportion (12.71%), effective irrigation area ratio (9.29%), environmental protection investment as a percentage of GDP (8.54%), and wastewater treatment rate (7.10%). (3) Based on these key constraints, targeted governance strategies are proposed, including enhancing wastewater treatment capacity, optimizing ecological water allocation, and increasing environmental protection investment. This study established an integrated “diagnosis–restoration–regulation” analytical framework for assessing water environment security and identifying priority restoration zones in lake-type basins, providing both theoretical foundations and practical references for global lake-type basin management.

1. Introduction

The accelerating pace of global industrialization and urbanization has exacerbated critical water-related challenges, including resource scarcity, pollution, and ecological degradation, posing significant threats to both human health and ecosystems [1]. The 2024 UN World Water Development Report identifies freshwater resource degradation as a major barrier to sustainable development [2]. This crisis is particularly severe in developing nations, where rapid urbanization and industrial expansion have worsened water quality deterioration. Inadequate water infrastructure further compounds these risks, as it contributes to approximately 2 million annual deaths from waterborne diseases such as diarrhea and arsenic poisoning [3,4]. As the world’s largest developing country, China faces significant water-related environmental challenges. Despite recent improvements in water quality, severe pollution persists in many rivers and lakes, with eutrophication trends threatening water supply safety and aquatic productivity in several regions [5]. In response, the Chinese government has implemented multiple policy measures, including wastewater treatment infrastructure development, ecological restoration initiatives, and the promotion of the environmental protection industry [6]. However, these macro-level interventions often lack spatial differentiation, limiting their effectiveness in targeted regional management.

The global water crisis has driven extensive research into comprehensive assessment frameworks. Since the 1970s, scholars have developed evaluation systems primarily focusing on water quality parameters [7,8] and water environment carrying capacity [9]. While these approaches provided valuable preliminary insights, their reliance on subjective indicator selection often failed to capture the systemic characteristics of aquatic ecosystems. The introduction of the Pressure–State–Response model and its derivatives marked a significant advancement, enabling systematic differentiation between environmental pressures, conditions, and mitigation measures. Applications by Guo et al. [10], Zhang et al. [11], and Wei [12] at the provincial, municipal, and reservoir scales, respectively, demonstrated the model’s effectiveness. The subsequent DPSIR (Driving forces–Pressure–State–Impact–Response) framework has gained particular prominence in complex watershed assessments due to its enhanced capacity for causal analysis and impact evaluation [13,14,15]. Compared with earlier frameworks such as the Pressure–State–Response model and carrying capacity models, DPSIR offers a more complete and logically coherent structure that captures the full spectrum of human–environment interactions. By explicitly incorporating “driving forces” and “impacts,” it enables a clearer delineation of causal pathways and feedback mechanisms between socioeconomic pressures and ecological responses. This makes it particularly effective for diagnosing water environmental issues that stem from both anthropogenic activities and systemic vulnerabilities. Moreover, the DPSIR model supports the integration of diverse data types—including economic, demographic, ecological, and policy-related indicators—into a unified analytical framework, facilitating both comprehensive assessments and spatially differentiated governance strategies. For these reasons, the DPSIR framework was selected as the foundation of this study’s evaluation system.

The accurate determination of indicator weights has been crucial for ensuring objective assessments in water environment security evaluations. To achieve this, researchers have employed various methodologies, including the Analytic Hierarchy Process [16], entropy weight method [17], fuzzy comprehensive evaluation [18], and improved fuzzy comprehensive evaluation [19]. Among these, the entropy weight method has gained prominence due to its strong objectivity, broad applicability, and operational simplicity. A comprehensive review of domestic and international studies has revealed that while significant progress had been made in evaluating water environment security and identifying key influencing factors across different scales (e.g., provincial, municipal, and reservoir levels), the applicability of assessment models for specific regions—such as arid zones, basins, mountainous areas, and rapidly urbanizing regions—as well as for particular water body types (e.g., lake-type basins and transboundary watersheds) still requires further validation and optimization.

The ultimate goal of water environment security research has been to accurately assess regional water conditions and propose actionable management strategies. Existing studies have primarily focused on influencing factors [20] and pollution sources [21] to derive policy recommendations. These studies have elucidated the causes and evolutionary patterns of water environment degradation to some extent, and they have also provided a scientific basis for regional governance. However, most research has remained confined to macro-level analyses, lacking the spatial precision necessary to identify and localize obstacle factors. This limitation has hindered the implementation of targeted restoration measures in specific areas. Consequently, a critical challenge in water environment security research has been the development of methodologies that leverage obstacle factors for spatial localization, thereby enabling the delineation of priority restoration zones with greater accuracy.

Current research on water environment security predominantly focuses on the provincial, municipal, and reservoir scales, with relatively fewer studies conducted at the watershed level. This disparity has arisen from the significant challenges of watershed-scale investigations, including difficulties in data collection across extensive areas, the complexity of watershed ecosystems, and operational and coordinative obstacles in water environment governance. However, due to the systemic and interconnected nature of water resources, a watershed-based approach that comprehensively considers upstream–downstream interrelationships as well as interactions among sub-watersheds is essential for effectively addressing transboundary water environment issues and achieving integrated watershed ecosystem management and optimization.

Lake-type basins pose unique research challenges due to their distinctive characteristics. These water bodies are not only directly affected by multiple anthropogenic activities, including agricultural, industrial, and urban discharges, but are also highly susceptible to eutrophication due to pollutant accumulation and retention. Additionally, the extended remediation periods required for these systems further complicate water environment research [22,23]. Dongting Lake, China’s second-largest freshwater lake, is located in the middle reaches of the Yangtze River and features a vast watershed spanning multiple provinces that exhibited typical lake-type basin characteristics [24]. As a crucial ecological functional zone within the Yangtze River Basin, the Dongting Lake watershed plays a vital role in water conservation, flood regulation, and climate modulation [25]. However, increasing pressures from population growth, economic development, and intensive water resource utilization have led to worsening water environment issues, highlighting the urgent need for scientific assessments to identify the key problems and to develop targeted governance strategies [26].

Building on this research background, the present study examines water environmental security in lake-type basins, selecting the Dongting Lake Basin (DLB) as a representative case. A comprehensive evaluation system based on the DPSIR model was developed to analyze data from 2000 to 2020. The entropy weight method was employed to determine the weighting coefficients of 24 selected indicators, enabling a systematic assessment of the spatiotemporal variations and multidimensional factors influencing water environmental security. To quantitatively identify key obstacles to water environmental security, an obstacle degree model was applied. Notably, this study introduced spatial autocorrelation analysis into the assessment of the spatial distribution of these primary obstacle factors, representing a novel approach in this research context. This integrated approach facilitated the precise delineation of priority restoration zones, thus providing targeted scientific support for formulating effective governance strategies and measures. The significance of this research is twofold: theoretically, it advances evaluation methodologies for watershed water environmental security; and practically, it establishes a comprehensive analytical framework that serves as a valuable reference for precision management and ecological restoration in other lake-type basins worldwide. The findings contribute both methodological innovations and practical solutions for addressing complex water environmental challenges in similar watershed systems.

2. Material and Methods

2.1. Study Area

This study focuses on the DLB in China. Located south of the Jingjiang River section of the Yangtze River, within the lake-network region of northern Hunan Province, Dongting Lake (111°14′–113°10′ E, 28°30′–30°23′ N) is China’s second-largest freshwater lake. The watershed spans the southern middle reaches of the Yangtze River and the northern Nanling Mountains, covering a vast area between 107°26′–114°20′ E and 24°36′–30°27′ N. With a total area of 261,879 km², it accounts for 14.6% of the drainage area of the Yangtze River Basin. The watershed extends beyond Hunan Province to include parts of six other administrative regions: Guizhou, Hubei, Guangxi, Chongqing, Jiangxi, and Guangdong (Figure 1). Characterized by a diverse topography—including plains, hills, and mountains—the basin experiences a subtropical monsoon climate with abundant but unevenly distributed precipitation, making it prone to extreme hydrological events such as floods and droughts. As a densely populated region with extensive river networks and intensive economic activities, the DLB serves as a crucial grain production base and economic development zone in China. However, these human pressures have led to significant ecological challenges, particularly water pollution and soil erosion. The complex interplay of natural and anthropogenic factors underscores the necessity of comprehensive and systematic research on water environmental security in the DLB to enhance our understanding of regional ecological issues and to inform future management strategies.

2.2. Data Sources

The study utilized two primary categories of data: socioeconomic statistics and natural geographic data. Socioeconomic indicators, including population figures and GDP, were obtained from government statistical yearbooks and official websites. Natural geographic data were collected from multiple authoritative sources: provincial administrative boundaries were sourced from the Resource and Environment Data Cloud Platform (http://www.resdc.cn, accessed on 12 September 2024), while annual precipitation data and multi-temporal land use/land cover remote sensing datasets were provided by the Geodata Center of the Chinese Academy of Sciences (http://gre.geodata.cn, accessed on 8 September 2024). To ensure comprehensive and reliable data coverage, this research incorporated additional sources, including water resource bulletins published by provincial water authorities, environmental statistics and status reports issued by provincial ecological environment departments, monthly surface water quality reports from the China National Environmental Monitoring Centre (http://www.cnemc.cn/, accessed on 6 November 2024), and hydrological network data from OpenStreetMap (http://download.geofabrik.de/, accessed on 23 September 2024). This multi-source approach ensured both the breadth and credibility of the dataset supporting the analytical framework of the study. To ensure maximum data consistency and comparability, this study prioritized the use of unified national-level statistical standards (e.g., China Statistical Yearbook on Environment, Ecological and Environmental Status Bulletin) during data collection. Regional data were rigorously examined for methodological consistency and subjected to time-series validation to ensure the reliability and coherence of all data sources.

2.3. Methodology

2.3.1. Water Environment Quality Assessment Based on the DPSIR Model

(1)

Construction of the Indicator System

The framework consisted of 5 criterion layers and 24 indicator layers (Figure 2) encompassing multiple dimensions and objectives. To ensure objective weighting, the entropy method was applied to determine indicator weights. This approach facilitated a systematic assessment of water environment security in the DLB from 2000 to 2020.

The components of the DPSIR framework are characterized as follows: The driving forces layer represents socioeconomic demands on the water environment in DLB, including population growth, economic activities, and urbanization. It primarily consists of population scale and economic conditions. The pressure layer examines direct human activities such as pollution discharge and excessive resource exploitation within the basin. It focuses on factors that include social and economic water consumption, as well as social and economic pollution emissions. The state layer evaluates actual water quality and ecological health, including pollution levels and ecosystem functioning. It primarily comprises resource supply status, wastewater discharge levels, and overall water environment quality. The impact layer assesses the comprehensive effects of water environment changes on ecosystems, human health, and socioeconomic systems. It encompasses both ecological and livelihood dimensions. The response layer encompasses management, technological, and policy measures designed to protect and improve the water environment through pollution control and resource optimization. Its core elements include management responses, economic responses, and ecological responses.

Indicator selection was based on four key criteria: alignment with the functional role of each DPSIR component, data availability and consistency, relevance to the environmental and governance context of the DLB, and support from existing studies and national technical guidelines. This approach ensured that the indicator system was both scientifically grounded and practically applicable to capturing the water environmental dynamics of the basin [15,27,28].

(2)

Indicator Processing

All indicators were classified using the natural breaks method. The natural breaks (Jenks) method was chosen for classification due to its ability to minimize within-class variance and maximize between-class variance; this allows it to better capture natural groupings and to reflect the spatial heterogeneity in the data. All indicators were classified using the natural breaks method [29,30]. The assignment of indicator values followed these principles: For indicators negatively affecting water environment security (where lower values indicate greater safety), the minimum value was assigned 1 and the maximum 5. For indicators positively affecting water environment security (where higher values indicate greater safety), the minimum value was assigned 5 and the maximum 1. Thus, in the final evaluation, lower composite values indicate higher water environment security. Higher composite values reflect more severe water environment degradation, necessitating enhanced regulation and protection measures. After grading and assigning values to each indicator, the evaluation indices for the five DPSIR criterion layers (Driving Forces, Pressure, State, Impact, and Response) were calculated separately. These indices were then weighted and aggregated to compute the Water Environment Security Index for the DLB.

2.3.2. Identification of Obstacle Factors

This study employed an obstacle degree model to identify key limiting factors affecting water environmental security in the DLB. The obstacle degree of each indicator exhibited a positive correlation with its constraining effect on regional water environmental security [31]. Diagnosing dominant obstacle factors allows for the development of more practical and targeted water environmental management measures, thereby improving overall water environmental conditions. The methodology primarily consisted of the following steps:

(1)

The calculation of factor contribution (F_i) and indicator deviation degree (I_i), as defined by the following formulas:

$$F_i=W_i\cdot X_{ij}$$

(1)

$$I_i=1-x_{ij}$$

(2)

where W_i represents the weight of the i-th indicator, X_ij denotes the weight of the j-th criterion layer to which the indicator belongs, while x_ij is the normalized value of the indicator.

(2)

The determination of obstacle degrees for both individual indicators (P_i) and criterion layers (U_j) affecting water environmental security was conducted as follows:

$$P_i=100\times \left(F_i\cdot I_i\right)/\left(\sum _{i=1}^m{F}_i\cdot I_i\right)$$

(3)

$$U_j=\sum P_{ij}$$

(4)

where m represents the total number of indicators.

2.3.3. Spatial Autocorrelation Analysis

Spatial autocorrelation analysis is a geostatistical method used to examine attribute correlations among neighboring locations. In this study, we employed GeoDa v1.14.0.0 software to analyze the spatial distribution patterns (i.e., dispersed or clustered) of major obstacle factors and water environmental security scores. This approach allowed us to identify priority areas for remediation.

The analysis involved two key steps: First, a first-order spatial weight matrix based on the Queen contiguity rule was adopted, in which regions sharing either a boundary or a vertex are considered neighbors. This method is well-suited for areal geographic units and effectively captures the spatial adjacency relationships between regions, aligning with the spatial structure of this study, which uses cities as the basic analytical units. Next, bivariate local Moran’s I was applied to examine localized spatial associations between the two variables. This approach identified four distinct clustering patterns: High-High (HH), High-Low (HL), Low-High (LH), and Low-Low (LL) (Formula (5)).

$$I_i=\left(y_i-\stackrel{⃐}{ y}\right)\sum _{j=1}^n{w}_{ij}\left(y_j-\stackrel{⃐}{ y}\right)$$

(5)

where I_i represents the local Moran’s Index for the i-th region, y_i and y_j represent the variable values for regions i and j, respectively, $\stackrel{⃐}{ y}$ is the mean value of the variable, n represents the total number of units and w_ij denotes the spatial weight value.

These clustering patterns effectively revealed the localized spatial relationships and distribution characteristics between major obstacle factors and water environmental security. In line with the research objectives, this study particularly emphasized regions facing the most severe water environmental security threats, characterized by high water environmental security scores. By analyzing the clustering patterns of major obstacle factors and water environmental security scores across five temporal periods, we systematically identified areas exhibiting HH and LH clustering patterns for more than three periods during the 20-year study. Spatial clustering distribution maps were then generated to delineate priority zones for remediation and restoration.

3. Results

3.1. Assessment Results of Water Environmental Security

Based on the grading criteria and the entropy method, the 24 indicators in the DLB from 2000 to 2020 were processed to determine their respective weights (

Table 1. Weight of water environmental security assessment indices in the DLB.

Target Layer	Criterion Layer	Weight	Indicator Layer	Impact Direction	Weight	Composite Weight
Water Environmental Security Evaluation Indicator System	D	0.0112	Population density	−	0.2626	0.0021
			GDP per capita	−	0.7374	0.0090
	P	0.3324	Domestic water consumption per capita	−	0.0527	0.0175
			Ecological water use proportion	+	0.3299	0.1089
			Domestic wastewater discharge per capita	−	0.0607	0.0199
			COD discharge from domestic wastewater per capita	−	0.1138	0.0386
			Ammonia nitrogen discharge from domestic wastewater per capita	−	0.0791	0.0260
			Water consumption per unit industrial added value	−	0.0464	0.0158
			Water consumption per unit agricultural added value	−	0.0665	0.0215
			Industrial wastewater discharge per unit industrial added value	−	0.0684	0.0232
			COD discharge from industrial wastewater per unit industrial added value	−	0.0930	0.0311
			Ammonia nitrogen discharge from industrial wastewater per unit industrial added value	−	0.0895	0.0299
	S	0.2177	Total water resources	+	0.2086	0.0439
			Water yield	+	0.1654	0.0368
			Annual precipitation	+	0.1558	0.0351
			Pesticide application per unit sown area	−	0.2537	0.0538
			Fertilizer application per unit sown area	−	0.0488	0.0102
			Proportion of river sections meeting Grade III water quality standards or better	+	0.1678	0.0379
	I	0.0928	Per capita available water resources	+	0.9317	0.0864
			Vegetation coverage	+	0.0683	0.0064
	R	0.3459	Wastewater treatment facility capacity	+	0.4005	0.1383
			Wastewater treatment rate	+	0.1634	0.0592
			Environmental protection investment as a percentage of GDP	+	0.2053	0.0710
			Proportion of irrigation area to crop sown area	+	0.2308	0.0774

Note: D denotes Driving forces, P denotes Pressure, S denotes State, I denotes Impact, and R denotes Response.

). Using weighted aggregation, the evaluation indices for each criterion layer and the overall water environmental security index of the basin were calculated. The assessment results for the key indicators are as follows (Figure 3):

(a)

Driving Forces Index

The results indicated that the driving forces index in DLB remained spatially stable over the 20-year study period while exhibiting distinct regional variations. Geographically, the eastern and southern regions of the basin, particularly Changsha City and Chenzhou City, consistently displayed higher driving forces index values, reflecting significantly greater environmental pressures from intensive economic activities and population growth. In contrast, the western areas exhibited comparatively lower driving forces index levels, indicating relatively reduced water environmental security risks.

(b)

Pressure Index

The overall water environmental pressure across the basin showed a gradual alleviation trend, although significant regional disparities persisted. In 2000, high-pressure areas were primarily concentrated in the northeastern and southern portions of the watershed. During 2005–2010, certain northern regions, particularly Jingzhou City, Enshi Tujia and Miao Autonomous Prefecture, and Changde City, exhibited elevated pressure indices, reflecting substantial challenges in water consumption and pollutant discharge. By 2015, intensified pressures emerged along the northern periphery and southwestern sectors, whereas Zhangjiajie City and Changsha City maintained relatively favorable conditions. By 2020, the spatial extent of high-pressure zones had further contracted, becoming predominantly confined to the southeastern watershed and western marginal areas.

(c)

State Index

The state index reflects the actual water quality and ecological health status of the basin. In 2000, 2005, and 2015, higher state index values were observed in the northeastern regions of the watershed, indicating poorer water environmental conditions and significant ecological pressures in these areas. In contrast, during 2010 and 2020, zones with elevated state index values gradually expanded toward the central and southwestern parts of the basin. This spatial shift highlights persistent water environmental security challenges in these regions, emphasizing the need for further optimization of governance measures and resource allocation to mitigate ecological risks and enhance aquatic ecosystem resilience.

(d)

Impact Index

The evaluation results of the Impact Index revealed distinct spatiotemporal patterns across DLB. In 2000, 2005, and 2015, the northeastern and central regions of the basin consistently exhibited higher Impact Index values, indicating more pronounced ecological and socioeconomic consequences of water environmental changes in these areas. In contrast, in 2010 and 2020, high-impact zones became predominantly concentrated in the central and southeastern portions of the watershed. This spatial evolution highlights progressively significant ecological and societal impacts in these regions, underscoring the urgent need for targeted conservation measures and adaptive management strategies to address growing environmental challenges.

(e)

Response Index

The response index reflects the basin’s capacity to implement governance measures and allocate resources for water environmental management. Evaluation results from 2000, 2005, and 2015 indicated relatively low response index values in the eastern basin, suggesting stronger institutional responsiveness in these regions. However, in 2010 and 2020, the spatial extent of high response index areas expanded eastward into the southeastern portions of the watershed, signaling a decline in response capacity due to inadequate mitigation measures and insufficient resource investments in these emerging hotspots.

The integrated water environmental security scores across DLB ranged between 2.1 and 3.9, with an overall fluctuation amplitude of approximately 1.8 points. Spatiotemporal analysis revealed persistent high-risk zones in the southeastern basin throughout the study period. In 2000, Yueyang City, Pingxiang City, and Hengyang City exhibited particularly vulnerable conditions. By 2005, emerging security declines became evident in Tongren City and Qiandongnan Miao-Dong Autonomous Prefecture. The 2010 assessment identified deteriorating conditions in northern and select southeastern areas, while 2015 saw improvements in the southeast contrasted by worsening conditions in the northeast. The most recent 2020 data indicated that Changde City achieved optimal security status, whereas elevated risks persisted along the watershed’s boundary areas.

3.2. Identification Results of Obstacle Factors

Through a systematic evaluation of the 24 indicators using the obstacle degree model, this study revealed relatively stable temporal variations in obstacle degrees across the 20-year study period (2000–2020). The findings are elaborated from two critical perspectives:

3.2.1. Key Obstacle Factors

The study identified five key indicators with the highest obstacle degrees that significantly constrain water environmental security in DLB (

Table 2. Barrier degree of water environmental security at the indicator level (Unit: %).

Year	2000	2005	2010	2015	2020	Average
Population density	0.0086	0.0086	0.0086	0.0086	0.0086	0.0086
GDP per capita	0.0354	0.0356	0.0355	0.0354	0.0357	0.0355
Domestic water consumption per capita	2.0163	2.0162	2.0138	2.0097	2.0119	2.0136
Ecological water use proportion	12.7185	12.7233	12.7095	12.7170	12.6711	12.7079
Domestic wastewater discharge per capita	2.3488	2.3511	2.3638	2.3608	2.3428	2.3535
COD discharge from domestic wastewater per capita	4.4608	4.4604	4.4595	4.4690	4.4513	4.4602
Ammonia nitrogen discharge from domestic wastewater per capita	2.9949	3.0056	3.0028	3.0080	3.0072	3.0037
Water consumption per unit industrial added value	1.8243	1.8315	1.8292	1.8239	1.8394	1.8297
Water consumption per unit agricultural added value	2.4741	2.4819	2.4803	2.4752	2.4963	2.4815
Industrial wastewater discharge per unit industrial added value	2.7186	2.7076	2.7174	2.7072	2.7203	2.7142
COD discharge from industrial wastewater per unit industrial added value	3.6240	3.6343	3.6485	3.6545	3.6820	3.6486
Ammonia nitrogen discharge from industrial wastewater per unit industrial added value	3.5090	3.5047	3.5055	3.5368	3.5555	3.5223
Total water resources	3.3170	3.3268	3.3222	3.3274	3.3190	3.3225
Water yield	2.7631	2.7746	2.7718	2.7638	2.7636	2.7674
Annual precipitation	2.6385	2.6476	2.6461	2.6396	2.6397	2.6423
Pesticide application per unit sown area	4.0698	4.0814	4.0805	4.0731	4.0782	4.0766
Fertilizer application per unit sown area	0.7747	0.7789	0.7801	0.7782	0.7750	0.7774
Proportion of river sections meeting Grade III water quality standards or better	2.8441	2.8523	2.8507	2.8445	2.8442	2.8471
Per capita available water resources	2.7860	2.7931	2.7912	2.7955	2.7950	2.7922
Vegetation coverage	0.2039	0.2045	0.2044	0.2040	0.2040	0.2042
Wastewater treatment facility capacity	16.9076	16.8184	16.8334	16.8691	16.8981	16.8653
Wastewater treatment rate	7.1524	7.1225	7.0876	7.0687	7.0684	7.0999
Environmental protection investment as a percentage of GDP	8.5268	8.5306	8.5566	8.5428	8.5256	8.5365
Proportion of irrigation area to crop sown area	9.2831	9.3086	9.3011	9.2872	9.2674	9.2895

; Figure 4): wastewater treatment facility capacity, ecological water use proportion, effective irrigation area ratio, environmental investment-to-GDP ratio, and wastewater treatment rate. Specifically, wastewater treatment facility capacity exhibited the highest average obstacle degree (16.87%). Spatial analysis identified major hotspots in Changde, Yiyang, Tongren, Qiandongnan Miao-Dong Autonomous Prefecture, and Yongzhou, highlighting substantial deficiencies in wastewater infrastructure construction and management in these areas. The ecological water use proportion exhibited an average obstacle degree of 12.71%, primarily concentrated along the northern basin periphery, indicating inadequate ecological water allocation that adversely affected water environmental security. Similarly, the effective irrigation area ratio had an average obstacle degree of 9.29%, with high-value zones predominantly located in northern agricultural regions, reflecting significant potential for improving agricultural water-use efficiency and irrigation management. The environmental investment-to-GDP ratio recorded an average obstacle degree of 8.54%, underscoring insufficient environmental funding and notable regional disparities in resource allocation, particularly in the eastern basin. Finally, the wastewater treatment rate presented an average obstacle degree of 7.10%, with scattered high-value clusters mainly in Zhangjiajie and Changsha, revealing suboptimal operational efficiency despite the presence of wastewater treatment infrastructure in these developed urban areas.

3.2.2. Key Obstacle Types

Analysis of the comprehensive obstacle degrees across all criterion layers revealed that response factors consistently represent the predominant obstacle to water environmental security in DLB. This finding fundamentally reflects systemic deficiencies in governance infrastructure, operational efficiency, and policy implementation—critical areas requiring urgent improvement. Pressure factors exhibit the second-highest obstacle degree, primarily driven by direct impacts from domestic and industrial wastewater discharges, as well as agricultural and industrial water consumption, which collectively pose substantial threats to water security. State factors exhibit intermediate obstacle levels, indicating that water quality and ecosystem health are influenced by multiple stressors, necessitating integrated remediation approaches. In contrast, impact and driving force factors demonstrate relatively lower obstacle degrees, suggesting that socioeconomic development exerts less direct influence on water security compared to response and pressure factors in the overall assessment. Synthesizing these obstacle factor identification results, this study proposes a prioritized management framework: (1) immediate enhancement of response capacities through infrastructure upgrades and policy enforcement; (2) targeted control of domestic and industrial pollution loads as well as agricultural water demands; and (3) progressive optimization of socioeconomic drivers through policy instruments (

Table 3. Barrier degree of water environmental security at the guideline level (Unit: %).

Year	D	P	S	I	R
2000	0.0439	38.6892	16.4071	2.9898	41.8699
2005	0.0442	38.7166	16.4615	2.9976	41.7800
2010	0.0441	38.7302	16.4514	2.9956	41.7787
2015	0.0440	38.7620	16.4267	2.9996	41.7678
2020	0.0443	38.7776	16.4196	2.9990	41.7595

Note: D denotes Driving forces, P denotes Pressure, S denotes State, I denotes Impact, and R denotes Response.

). This tripartite strategy provides a structured pathway for achieving holistic watershed security improvement while addressing the identified obstacle hierarchy.

3.3. Spatial Autocorrelation Analysis Results

Spatial autocorrelation analysis revealed significant clustering patterns between major obstacle factors and the water environmental security index across the DLB (Figure 5). The results were statistically significant at the 0.05 level, as determined through permutation tests (999 permutations).

The analysis identified nine prefecture-level regions—Changde, Yiyang, Tongren, Qiandongnan Miao-Dong Autonomous Prefecture, Yongzhou, Jingzhou, Yueyang, Pingxiang, and Hengyang—that consistently exhibited HH or LH clustering patterns in at least three separate study periods. This persistent spatial aggregation highlights the concentration of the five key obstacle factors constraining water security. Specifically, HH clusters for wastewater treatment capacity—indicating the co-occurrence of inadequate infrastructure and high environmental risk—were concentrated in Changde, Yiyang, Tongren, Qiandongnan, and Yongzhou. In contrast, LH clusters, which reflect persistent risks despite better infrastructure, appeared in Hengyang, Yueyang, and Pingxiang. Similarly, the ecological water use proportion exhibited HH clusters in Jingzhou, Yueyang, Pingxiang, Tongren, Qiandongnan, and Yongzhou, signaling ecological flow shortages. Meanwhile, LH clusters in Changde, Yiyang, and Hengyang suggested persistent risks despite greater ecological water allocations. The effective irrigation area ratio displayed HH clusters in Yueyang, Hengyang, Tongren, and Qiandongnan, revealing dual deficiencies in irrigation management and water security. Conversely, LH clusters in Jingzhou, Changde, Yiyang, and Yongzhou indicated that expanded irrigation systems failed to prevent water degradation. Parallel spatial patterns were observed for the environmental investment-to-GDP ratio and wastewater treatment rate, collectively demonstrating regional imbalances between management inputs and environmental outcomes across the basin.

4. Discussion

This study systematically analyzed the spatiotemporal evolution of water environmental security and its influencing factors in the DLB while examining the mechanisms of key obstacle types and identifying remediation measures for priority areas. The discussion focuses on four key aspects: (1) drivers of spatiotemporal variations in water environmental security, (2) mechanistic understanding of dominant obstacle types, (3) targeted restoration strategies for critical zones, and (4) study limitations and future research directions. This approach aims to provide scientifically grounded and actionable insights for watershed management.

4.1. Causes of Spatiotemporal Variations in Water Environmental Security in DLB

Further analysis of the spatial distribution of evaluation results across the DPSIR framework layers in DLB revealed that variations in driving forces, pressure, state, impact, and response are closely linked to regional economic development, resource utilization, and management capacity. The spatial pattern of the driving forces index strongly correlated with the intensity of regional economic activity. For example, from 2000 to 2020, Chenzhou City experienced steady primary industry GDP growth, accelerated secondary industry expansion (particularly after 2010), and consistent increases in per capita GDP and local fiscal expenditure. However, this economic expansion was accompanied by rising industrial wastewater discharge and other pollutant emissions, exacerbating water environmental security risks. Economic growth has driven intensified exploitation, utilization, and consumption of natural resources, placing significant stress on watershed ecosystems and water resources. This, in turn, has elevated water environmental security risks. Notably, the correlation analysis between economic indicators and the driving force index in Chenzhou City revealed a significant negative relationship (

Table 4. The correlation between economic development and driving forces index in Chenzhou City.

	The GDP of the Secondary Industry	Per Capita GDP	Local Fiscal Expenditure	Driving Forces Index
The GDP of the secondary industry	1	0.968 **	0.966 **	−0.949 *
Per capita GDP	0.968 **	1	0.998 **	−0.931 *
Local fiscal expenditure	0.966 **	0.998 **	1	−0.911 *
Driving forces index	−0.949 *	−0.931 *	−0.911 *	1

Note: * p < 0.05, ** p < 0.01.

), specifically manifested as: The strong negative correlation between GDP of the secondary industry and the driving force index (r = −0.949 *, p < 0.05) exposes an environmental paradox—economically disadvantaged regions, constrained by weak industrial foundations, cannot afford efficient pollution treatment facilities, forcing them to maintain extensive growth at the cost of environmental risks, resulting in lower industrialization levels correlating with higher water security risks. The highly significant positive correlation between GDP per capita and fiscal expenditure (r = 0.998, p < 0.01) confirms that economic weakness directly restricts fiscal capacity. The significant negative correlation between fiscal expenditure and the driving force index (r = −0.911 *, p < 0.05) reveals fiscal constraints as a key conduit for risk transmission. This implied that insufficient economic capacity may limit investments in environmental governance and infrastructure, resulting in more pronounced environmental drivers and a higher vulnerability of the regional water system [32]. The processes of industrial transformation and urbanization have profoundly altered the natural-social water cycle patterns, with eastern developed regions experiencing particularly severe human-water conflicts. Additionally, the mismatch between rapid economic growth and environmental protection investment has emerged as a major constraint, hindering sustainable improvements in water environmental security.

Second, the pressure index reflects the combined effects of water use and pollution loads from domestic, industrial, and agricultural activities across the basin. Although water environmental pressures were partially alleviated in some regions between 2000 and 2020 due to regional economic development and enhanced environmental management measures, areas with intensive economic activities and high water demand continued to experience elevated wastewater discharge loads. This persistent pattern highlights the ongoing conflict between socioeconomic development and water resource management, particularly in rapidly urbanizing and industrializing areas where water consumption and pollution generation remained disproportionately high relative to environmental carrying capacity. For example, analysis in Hengyang City (

Table 5. The correlation between economic development and pressure index in Hengyang City.

	The GDP of the Secondary Industry	Per Capita GDP	Local Fiscal Expenditure	Pressure Index
The GDP of the secondary industry	1	0.972 **	0.985 **	−0.880 *
Per capita GDP	0.972 **	1	0.996 **	−0.937 *
Local fiscal expenditure	0.985 **	0.996 **	1	−0.929 *
Pressure index	−0.880 *	−0.937 *	−0.929 *	1

Note: * p < 0.05, ** p < 0.01.

) showed significant negative correlations between key economic indicators (such as GDP of the secondary industry and local fiscal expenditure) and the pressure index. This strong negative correlation (r = −0.937 *, p < 0.05) manifests a development dilemma wherein fiscal limitations in low-income areas produce inadequate sewage pipe coverage and elevated direct wastewater discharge rates, thereby directly converting economic disadvantage into pollution load pressure. The significant negative correlation between local fiscal expenditure and the pressure index (r = −0.929 *, p < 0.05) directly links to pollution control investment gaps. This suggested that lower levels of economic development were associated with higher environmental pressure, a relationship that was likely due to insufficient investment in pollution control infrastructure and limited governance capacity, which exacerbated the mismatch between pollutant emissions and the basin’s carrying capacity [33]. The spatial heterogeneity of pressure indices underscores the need for differentiated management strategies that address both point-source pollution control in urban–industrial clusters and non-point-source mitigation in agricultural regions while considering the basin’s evolving development patterns and the impacts of climate variability on water availability.

Furthermore, water environmental security conditions exhibited distinct spatial variations and phased characteristics. Areas with higher state index values were primarily located in regions experiencing severe agricultural non-point-source pollution, high population density, and intensive water resource development and utilization. In some areas, excessive chemical inputs per unit of crop sown area led to water quality deterioration. Between 2010 and 2020, high-value state index areas gradually expanded into the central and southwestern parts of the basin, primarily due to significant land use changes, the expansion of agricultural irrigation areas, and industrial development, which exacerbated water resource supply-demand conflicts in these regions. Simultaneously, insufficient regional precipitation, reduced water yield, and relatively low water resource management levels further aggravated water environment degradation, progressively weakening the natural purification capacity of water bodies.

The northeastern and central parts of the basin exhibited higher impact indices, primarily due to their high population density, frequent economic activities, and substantial water resource demand and consumption. Additionally, geographical and climatic constraints led to uneven water resource distribution, exacerbating impacts on human life and ecosystems. In the central and southeastern regions, rapid economic growth and urbanization between 2010 and 2020 intensified resource consumption and ecological degradation, reducing per capita available water resources and making the impacts of water environmental changes on production and daily life more pronounced.

The response index revealed regional disparities in governance measures and resource allocation. The eastern DLB, benefiting from its developed economy and early implementation of environmental policies, demonstrated stronger response capabilities, particularly in wastewater treatment facility capacity and sewage treatment rates. However, due to imbalanced regional development, insufficient environmental management investment, and lagging policy supervision, high response index areas expanded southeastward in 2010 and 2020, indicating a decline in water environmental security response capacity in these regions. Overall, variations in water environmental security across different areas were closely linked to economic development levels, governance capacity, resource utilization efficiency, and natural conditions. The basin’s water environmental security risks exhibited a diffusion trend from central areas toward the periphery, reflecting imbalances in policy implementation and resource allocation among regions.

4.2. Mechanisms of Key Obstacle Factors

According to the obstacle degree analysis results, response factors consistently represented the primary obstacles to water environmental security in the DLB. This finding highlights deficiencies in governance infrastructure, operational efficiency, and policy implementation within the basin. To deepen the mechanistic understanding, we employed path analysis to quantify the causal relationships between economic drivers and water environmental security: Regional GDP per capita exerted a significantly negative effect on local fiscal expenditure. Insufficient fiscal expenditure further led to shortages in environmental governance investment, ultimately exacerbating water environmental risks. This mechanism is empirically validated by Chenzhou City’s data in Table 4; a strong negative correlation exists between GDP of the secondary industry and the driving forces index (r = −0.949 *). GDP per capita shows near-perfect positive correlation with fiscal expenditure (r = 0.998 **). These patterns confirm that economically disadvantaged areas, constrained by fiscal limitations, cannot support adequate governance investment, resulting in significantly elevated driving forces index. Although environmental protection investments have increased in recent years, they remain critically insufficient in economically vulnerable regions, as quantified by the path analysis above. Even when treatment facilities have been constructed, factors such as funding shortages, outdated technology, and poor management have often resulted in low operational efficiency, preventing these facilities from delivering their full environmental benefits. Moreover, inadequate policy enforcement at the local level, along with flawed supervision and accountability mechanisms, has weakened the effectiveness of environmental policies. For example, in the DLB, the cities of Yueyang, Changde, and Yiyang have failed to implement fertilizer reduction targets in accordance with national and provincial requirements in their efforts to control agricultural non-point source pollution. Problems include perfunctory work deployment, falsification during implementation, and distorted or inaccurate statistical reporting [34]. These governance failures are, in essence, the end-stage manifestation of the causal chain: economic constraints → fiscal limitations → insufficient investment.

Following response factors, pressure factors primarily manifest in domestic and industrial wastewater discharges, as well as in high water consumption in industrial and agricultural activities. Rapid population growth and accelerated urbanization in the basin have led to a significant increase in domestic wastewater discharge, while wastewater treatment capacity has failed to keep pace. Simultaneously, the expansion of heavy and chemical industries has generated large volumes of wastewater, with some enterprises discharging untreated or substandard effluent due to inadequate treatment facilities, severely polluting water bodies. Additionally, inefficient agricultural water use, along with the widespread application of fertilizers and pesticides, directly threatens the water environment through runoff and infiltration.

In contrast, state factors occupy an intermediate position, reflecting the combined influence of multiple factors on water quality and ecological health. The relatively lower obstacle degrees of impact and driving force factors suggest that the indirect effects of socioeconomic development are comparatively limited. Overall, although the basin has some natural purification capacity, the intensification of economic activities and population growth has overwhelmed this self-repair mechanism, rendering it ineffective against external pollution pressures.

4.3. Key Area Restoration Measures

To enhance the strategic relevance and policy applicability of the restoration measures, they are structured below according to short-term and long-term implementation priorities, reflecting differences in urgency, resource requirements, and systemic impact.

In the short term, priority should be given to addressing infrastructure deficiencies and operational inefficiencies that can yield rapid improvements in water environmental conditions. In regions with insufficient wastewater treatment capacity—such as Changde City, Yiyang City, Tongren City, Qiandongnan Miao-Dong Autonomous Prefecture, and Yongzhou City—investment in sewage infrastructure must be intensified to expand pipeline coverage, reduce untreated discharges, and enhance the joint control of point and non-point source pollution. In Hengyang City, Yueyang City, and Pingxiang City, attention should be focused on improving the operational efficiency of existing facilities to avoid overloading and to ensure stable treatment performance. At the same time, stronger control of agricultural non-point source pollution is needed across the basin, especially in areas like Jingzhou City, Yueyang City, Yiyang City, and Yongzhou City. These areas should enhance pollution source monitoring, enforce environmental regulations, and promote the adoption of green farming practices and water-saving technologies. Additionally, irrigation infrastructure should be upgraded in regions with low effective irrigation area ratios, such as Yueyang, Hengyang, Tongren, and Qiandongnan, in order to improve water use efficiency, reduce nutrient runoff, and support the coordinated development of agriculture and ecological security.

From a long-term perspective, the emphasis should shift toward systemic restoration of the ecological and governance foundations of the basin. Ecological water allocation must be optimized in cities such as Jingzhou, Yueyang, Pingxiang, Tongren, and Yongzhou, where adjustments in water distribution structure, wetland restoration, and enhanced hydrological connectivity can play key roles in restoring ecosystem integrity. In Changde, Yiyang, and Hengyang, comprehensive ecological scheduling mechanisms should be established to balance water use efficiency with pollution source reduction and ecological restoration goals. Meanwhile, regions with persistently low levels of environmental investment relative to GDP, such as Jingzhou, Yueyang, Changde, Yiyang, Hengyang, and Pingxiang, require sustained increases in fiscal inputs, alongside improvements in funding efficiency and industrial restructuring. In Tongren, Qiandongnan, and Yongzhou, attention should be directed toward optimizing fund utilization, advancing ecological compensation policies, and adjusting high-impact economic activities to reduce pressure on water ecosystems. Furthermore, in cities with underperforming wastewater treatment systems, including Jingzhou, Yueyang, Changde, and Pingxiang, governance measures such as pipeline expansion, multi-source pollution coordination, and ecological function rehabilitation must be prioritized. Even in areas such as Yiyang, Hengyang, Tongren, Qiandongnan, and Yongzhou, where wastewater treatment rates are relatively high, continuous efforts are needed to optimize system efficiency, strengthen non-point source control, and build long-term ecological resilience.

These recommendations are closely aligned with the priorities outlined in the National 14th Five-Year Plan for Ecological and Environmental Protection of Water Systems, which emphasizes investment in sewage treatment infrastructure, coordinated pollution control, wetland restoration, and upstream–downstream governance mechanisms. By anchoring the proposed restoration pathways within this national policy framework and tailoring them to local obstacle patterns, this study provides both scientific and actionable guidance for sustainable water environment management in lake-type basins.

4.4. Limitations and Prospects

This study, using the DLB as a case study, established a comprehensive water environmental security evaluation system based on the DPSIR model and systematically analyzed its spatiotemporal variations and key obstacle factors. Building upon previous studies of water environmental security, this research contributes methodological advancements and integrations in three key aspects. (1) In terms of evaluation system design, a multi-layered index framework based on the DPSIR model was constructed, incorporating 24 indicators and employing the entropy weight method for objective weighting. This improves the structural consistency and reproducibility of the evaluation system compared to conventional subjectively weighted approaches. (2) For obstacle factor identification, the study introduced the obstacle degree model to quantify key limiting factors. This enables the diagnosis of time- and space-specific constraints across the basin, extending the research focus from overall assessment to targeted factor analysis. (3) Regarding spatial heterogeneity, the integration of global and local Moran’s I statistics allowed the identification of clustered obstacle patterns and priority governance zones, addressing a common gap in previous studies concerning spatially explicit intervention targeting. Nevertheless, the study is subject to certain data and methodological limitations. Some indicators were constrained by inconsistent statistical standards across provinces or limited temporal coverage, which may have introduced uncertainties in the comparative analysis. The entropy weight method assumes that indicators with greater variability are more important, which may not fully reflect environmental relevance. In future research, we will consider introducing hybrid weighting methods that combine subjective and objective approaches, such as incorporating the Analytic Hierarchy Process (AHP) into the entropy-based framework, and applying sensitivity analysis to test the stability and consistency of evaluation results, thereby further enhancing the scientific validity and application reliability of the model. Similarly, the obstacle degree model relies on a linear additive structure, potentially overlooking nonlinear or synergistic interactions between variables. Moreover, the DPSIR framework simplifies causal relationships into a unidirectional flow, which may neglect the feedback loops or cross-scale interactions inherent in complex watershed systems. While the results reflect general patterns in lake-type basins, significant differences exist among basin types in hydrological processes, pollution sources, and socioeconomic drivers. Therefore, the findings may not be fully applicable to other basin types. Although the methodological framework developed in this study is transferable to other lake-type or semi-enclosed watersheds, the specific results and identified obstacle factors are context-dependent and not directly generalizable. Future research should conduct comparative analyses between lake-type and other basin types, identifying commonalities and distinct characteristics across different watersheds in regard to water environmental security. This would contribute to developing an evaluation framework that balances universality and specificity. Meanwhile, consideration will be given to incorporating climate scenario factors as external driving variables into the DPSIR model, or coupling hydrological simulation models with environmental models, to thoroughly investigate the evolution trends and risk warning mechanisms of the water environment under climate change, thereby enabling a more comprehensive assessment of the dynamic vulnerability of regional water environment security.

Additionally, building on the DPSIR model, future studies should incorporate the Social–Ecological Systems framework by integrating more socioeconomic and ecological process indicators to examine water environmental security from multiple 3 dimensions. Such an approach would provide stronger theoretical support and practical guidance for formulating more precise and effective watershed management strategies.

5. Conclusions

This study established a comprehensive evaluation system for water environmental security in lake-type basins and conducted an in-depth analysis of spatiotemporal variations and key obstacle factors in the DLB from 2000 to 2020. Additionally, it incorporated the novel approach of identifying priority restoration areas through spatial autocorrelation analysis. The results indicated that while overall water environmental security in the basin has improved, significant regional disparities persist, with higher risks in southeastern areas and better conditions in some central-eastern regions. The primary obstacle factors included wastewater treatment facility capacity, ecological water use proportion, effective irrigation area ratio, environmental investment-to-GDP ratio, and wastewater treatment rate. Accordingly, the following restoration strategies are proposed: (a) enhancing wastewater treatment capacity, particularly in low-treatment-rate areas, by increasing infrastructure investment and improving operational efficiency and coverage; (b) boosting environmental investment through stable fiscal mechanisms to gradually raise environmental funding as a percentage of GDP; and (c) optimizing ecological water resource utilization by ensuring a rational allocation between ecological and agricultural uses while mitigating agricultural pollution. Based on spatiotemporal characteristics, targeted measures are recommended for specific regions: strengthening wastewater treatment infrastructure and ecological restoration in Changde and Yiyang; optimizing ecological water allocation and pollution source control in Jingzhou and Yueyang; improving irrigation facilities and agricultural pollution management in Hengyang and Tongren; increasing environmental investment and fund utilization efficiency in Jingzhou and Changde; and enhancing wastewater treatment standards and non-point source pollution control in Yiyang and Hengyang. Implementing these measures through a multi-level integrated “diagnosis–restoration–regulation” management framework will ultimately promote sustainable development and ecological conservation in the DLB.