Spatial and Temporal Divergence of Water Resource Carrying Capacity in Hubei Province, China, from the Perspective of Three Major Urban Agglomerations

Abstract

Water resource carrying capacity is indispensable for sustainable development, acting as a crucial determinant for harmonizing ecological preservation with socio-economic development. This study centers on Hubei Province, which is an important water conservation area in the Yangtze River Basin and is one of the core water source areas for the South-to-North Water Diversion Project, and evaluates the water resource carrying capacity of the three major urban agglomerations in Hubei Province from 2005 to 2020 based on the four dimensions of water resources, economics, society, and ecology, using the entropy weighting method and the TOPSIS model to construct an evaluation index system. We then employ the kernel density estimation method, ArcGIS visualization, and the Dagum Gini coefficient method to perform a comprehensive analysis of spatial and temporal differences, dynamic evolution, and contribution sources. The results show that (1) the water resource carrying capacity of Hubei Province as a whole increased from a severe overload to overload level during the study period. The water resource carrying capacity of the three major urban agglomerations shows a regional distribution pattern where the Yi-Jing-Jing-En agglomeration’s capacity surpasses that of the Wuhan urban agglomeration, which is bigger than Xiang-Shi-Sui-Shen urban agglomeration. A lower ecological water use rate primarily constrains the enhancement of the carrying capacity of water resources in Hubei Province. (2) The kernel density estimation reveals an increase in the overall water resource carrying capacity across Hubei Province’s three major urban agglomerations during the study period, alongside a pronounced trend towards polarization. (3) While the overall Gini coefficient, indicating an imbalance in water resource carrying capacity in Hubei Province, remains high, it demonstrates a declining trend, suggesting a growing disparity in water resource carrying capacity across the province’s three major urban agglomerations. Hubei Province’s water resource carrying capacity faces challenges of an overall imbalance and localized vulnerability. Strategies should aim to enhance synergy, address these deficiencies directly, and devise targeted measures tailored to the distinct features of various urban clusters.

1. Introduction

Water resources are not only the most fundamental material guarantee for human survival, but also a crucial production means for socio-economic development, occupying a central position among various resources [1]. The study of water resource carrying capacity represents a systematic organization of natural resources and social development, serving as crucial foundational research for national development. This research provides significant references for national policy decisions concerning the population, ecology, the economy, and social development. As a country that supports 20% of the world’s population with only 6% of the world’s water resources [2], China faces significant challenges due to the imbalance between water supply and demand caused by population growth. Hubei Province, a major economic province in China and one of the most developed provinces in the Yangtze River Basin, is a leader in the country’s GDP and holds substantial strategic importance nationwide. As one of the initial “resource-saving and environmentally friendly society” comprehensive supporting reform pilot zones in China, Hubei Province also confronts a series of contradictions and conflicts between water resources, the environment, and socio-economic development [3]. Therefore, it is necessary to analyze the regional differences in the spatial and temporal evolution of the water resource carrying capacity of the three major urban agglomerations in Hubei Province based on both temporal and spatial dimensions. This analysis aims to provide guidance for promoting the coordinated development of water resources and the socio-economic development in Hubei Province and offer references for addressing the imbalance of water supply and demand both in China and globally.

Overseas research on the carrying capacity of resources and the environment is extensive and often integrates water resources with other environmental or resource elements to explore regional planning, management, and sustainable development [4,5]. China also started early in studying the carrying capacity of water resources, accumulating numerous research achievements in defining the concept, influencing factors, and evaluation methods of water resource carrying capacity.

In terms of definition and connotation, early studies often used natural elements such as water scarcity, sustainable utilization levels, ecological limits, or biosphere limits to describe water resource carrying capacity, primarily guiding agricultural production [6]. With the acceleration of urbanization and the increasing proportion of national industrial output, the definition of water resource carrying capacity has expanded to include demographic and economic factors to better explain the urban environment [7,8]. Currently, water resource carrying capacity is mainly summarized as “the maximum amount of water resources that can be exploited in an ecological region at a given stage of socio-economic development, focusing on the interactions between water resources and socio-economics” [9]. In terms of influencing factors, existing research results can be summarized into two main aspects: natural factors and social factors. Among natural factors, the quantity and quality of water resources are decisive factors [10]. Social factors consider the importance of regional production and consumption levels, as well as the different effects of population density, urbanization rate, and urbanization quality on the water resource carrying capacity in different regions [3,11]. In constructing the indicator system, scholars have adopted the “PSR model [12]”, the “DPSIR model [13]”, the “DPSIRM model [14]”, and the “VPOSRM model [15]” to construct the water resource carrying capacity evaluation index system from the perspective of ecological security of watersheds [16]. With the continuous enrichment of the theory, some scholars began to combine various measurement methods to evaluate the water resource carrying capacity of different regions. Currently, the main evaluation methods include subjective weighting methods and objective weighting methods, such as the analytic hierarchy process (AHP) [17], the entropy weighting method [18], and the TOPSIS method [19]. For example, Deng Zhenghua used the TOPSIS model combined with AHP and the entropy weighting method to measure the water resource carrying capacity of the Dongting Lake Basin from 2009 to 2018 [20].

In summary, the existing studies provide useful references for this paper, but there are still the following aspects to be expanded upon and improved: first, in terms of research scale, existing studies mainly focus on economic zones [21], watersheds [22], or cities [23] and counties [24], and there is little research on the microscale from the perspective of “urban agglomeration”, especially concerning the water resource carrying capacity of urban agglomerations in Hubei Province, which is relatively weak in guiding and demonstrating, and urgently needs to be expanded upon and deepened. Secondly, concerning the research focus, existing studies analyzing spatial and temporal differences following the evaluation of water resource carrying capacity are limited. Water resource carrying capacity is influenced by temporal and spatial factors and is characterized by heterogeneity and regional variability. Neglecting these aspects could compromise the accuracy of the research findings [25]. Therefore, this study takes the three major urban agglomerations in Hubei Province as the research object, combining the entropy weighting method with the TOPSIS model based on systems theory, constructing a four-dimensional evaluation index system of water resources, society, the economy, and ecology, and adopting methods such as kernel density estimation and the Dagum Gini coefficient to comprehensively assess and analyze the spatial and temporal differences, dynamic evolution, and contributing sources of the water resource carrying capacity of the three major urban agglomerations in Hubei Province from 2005 to 2020. This study thus aims to optimize the regional economic structure and water resources allocation and provide theoretical references with higher precision and greater instructive power for the sustainable and coordinated development of water resource carrying capacity.

2. Materials and Methods

2.1. Overview of the Study Area

Hubei Province is situated in the middle reaches of the Yangtze River, spanning from 108°21′ E to 116°07′ E and from 29°05′ N to 33°20′ N (Figure 1). The province administers 12 provincial municipalities, one autonomous prefecture, three provincial-level cities, and one forest area. According to the “Hubei Province New Urbanization Plan (2021–2035)”, it is segmented into three major urban agglomerations: the Yi-Jing-Jing-En urban agglomeration (comprising Yichang, Jingzhou, Jingmen, Enshi), the Xiang-Shi-Sui-Shen urban agglomeration (including Xiangyang, Shiyan, Suizhou, Shennongjia), and the Wuhan urban agglomeration (encompassing Wuhan, Huanggang, Ezhou, Huangshi, Xianning, Xiantao, Qianjiang, Tianmen, Xiaogan). Located in a subtropical monsoon humid climate zone, Hubei hosts three primary river basins: the Yangtze River, Han River, and Qing River. In terms of total volume, Hubei is rich in water resources; for instance, in 2020, the province recorded an average precipitation of 1642.6 mm, totaling 175.471 billion m³ of water resources, with a per capita availability of 3038 m³. However, concerning the production, domestic, and ecological utilization of water resources, there are still regions where development and utilization are not harmonized. Conducting a scientific evaluation of the water resource carrying capacity of Hubei Province in order to understand the current status and potential is crucial for rationally determining industrial scale and urban layout.

2.2. Research Methodology

2.2.1. Entropy Weight Method

Different methods for calculating weights can lead to significant variations in evaluation results. The determination of weights mainly includes the analytic hierarchy process (AHP) [26], RANCOM [27], the TOPSIS method [28], and the entropy weighting method [29]. The AHP simplifies the complex analysis process of influencing factors, but its research results may have low credibility due to an excessive reliance on qualitative analysis. The RANCOM method integrates multiple opinions and expert knowledge, making it particularly suitable for decision-making problems that span multiple disciplines or fields. However, it depends considerably on the selection and opinions of experts, which may introduce subjective biases. In extensive research areas, it is nearly impossible for experts to possess detailed knowledge of every location within the study area.

In contrast, the entropy weighting method is an objective method based on mathematical statistics. It determines weights according to the information provided by the data observed from each indicator, thus overcoming the influence of subjective factors on weight determination. This method has been widely used across various disciplines. The TOPSIS method utilizes the information from all indicators to objectively and accurately evaluate the pros and cons of the subjects by comparing the relative distance of each evaluated object from the best and worst possible outcomes for quantitative ranking. Its advantages include simple computation and reasonable results. Combining the entropy weighting method with the TOPSIS method can refine the evaluation formulas for the objects assessed and for the positive and negative ideal solutions, thereby aligning the evaluation results more closely with the actual situation. Additionally, this approach mitigates the shortcomings of the AHP and RANCOM methods, which primarily depend on the subjective opinions of experts for weight determination. The calculation steps are as follows:

• Selection of indicators

Assuming there are h years, m cities, and n evaluation indicators, while x_λij is the indicator value of the jth indicator for the ith city in the λth year.

2.

Dimensionless processing of indicators

The method of range standardization is adopted to dimensionlessly normalize each indicator.

Positive indicator treatments:

$${\mathrm{Z}}_{\mathsf{\lambda }\mathrm{i}\mathrm{j}}=\frac{\left({\mathrm{x}}_{\mathsf{\lambda }\mathrm{i}\mathrm{j}}-{\mathrm{x}}_{\mathrm{m}\mathrm{i}\mathrm{n}}\right)}{\left({\mathrm{x}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{x}}_{\mathrm{m}\mathrm{i}\mathrm{n}}\right)}$$

(1)

Negative indicator treatments:

$${\mathrm{Z}}_{\mathsf{\lambda }\mathrm{i}\mathrm{j}}=\frac{\left({\mathrm{x}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{x}}_{\mathsf{\lambda }\mathrm{i}\mathrm{j}}\right)}{\left({\mathrm{x}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{x}}_{\mathrm{m}\mathrm{i}\mathrm{n}}\right)}$$

(2)

where i = 1, 2, 3, … , m; j = 1, 2, 3, … , n; and i and j are the total number of evaluation objects and evaluation indexes, respectively. x_max and x_min are the maximum and minimum values of different indexes j in all evaluation objects. x_λij are the different indicators i after dimensionless and before dimensionless processing.

3.

Normalization of indicators

$${\mathrm{P}}_{\mathsf{\lambda }\mathrm{i}\mathrm{j}}=\frac{{\mathrm{Z}}_{\mathsf{\lambda }\mathrm{i}\mathrm{j}}}{\sum _{\mathsf{\lambda }=1}^{\mathrm{h}}\sum _{\mathrm{i}=1}^{\mathrm{m}}{\mathrm{Z}}_{\mathsf{\lambda }\mathrm{i}\mathrm{j}}}$$

(3)

To address cases where it is essential to maintain the relative proportionality between data points, this study employs decoupling normalization. This method involves dividing the value of each data point by the total sum of all values in the dataset. Consequently, each normalized value falls within a predetermined range, ensuring that the sum of all normalized values is equal to one. This approach effectively preserves the proportional relationships among the data points, thereby facilitating more accurate subsequent analyses.

4.

Calculation of the entropy value of each indicator

$${\mathrm{E}}_{\mathrm{j}}=-\mathrm{k}\sum _{\mathsf{\lambda }=1}^{\mathrm{h}}\sum _{\mathrm{i}=1}^{\mathrm{m}}{\mathrm{P}}_{\mathsf{\lambda }\mathrm{i}\mathrm{j}}\ln {\mathrm{P}}_{\mathsf{\lambda }\mathrm{i}\mathrm{j}}$$

(4)

where $\mathrm{k}=\frac{1}{\ln \left(\mathrm{h} \times \mathrm{m}\right)}$.

This research employs the Cartesian product approach to compute the contribution of each data point. It involves multiplying the probability of each event by its corresponding logarithm, akin to calculating the Cartesian product in a multidimensional vector space. This technique is intended to measure more accurately the contributions of individual probabilities to the overall entropy, thereby effectively reflecting the data’s uncertainty and dispersion.

5.

Calculation of the redundancy of the entropy value of each indicator

$$\mathrm{D}=\mathrm{I}-{\mathrm{E}}_{\mathrm{j}}$$

(5)

6.

Calculation of the weights of the indicators

$${\mathrm{W}}_{\mathrm{j}}=\frac{{\mathrm{D}}_{\mathrm{j}}}{\sum _{\mathrm{j}=1}^{\mathrm{n}}{\mathrm{D}}_{\mathrm{j}}}$$

(6)

2.2.2. TOPSIS Method

The TOPSIS model, introduced by Hwang et al. in 1981 [30], is a multi-criteria decision analysis method also known as the Technique for Order of Preference by Similarity to Ideal Solution [31]. This method effectively utilizes information from the original data, accurately reflecting differences among evaluation schemes and facilitating the ranking of evaluation targets based on their merits. The basic process of the TOPSIS model includes normalizing the data to create a standardized matrix, calculating the distance and closeness coefficient (C_i) between the evaluation objects and the optimal and worst solutions, and then ranking the evaluation objects to determine their relative merits. A smaller C_i value indicates weaker water resource carrying capacity, highlighting a greater contradiction between water resources supply and demand; conversely, a higher C_i value suggests stronger water resource carrying capacity, indicating a balanced relationship between water supply and demand. This study employs the TOPSIS method in conjunction with entropy weighting to comprehensively evaluate the water resource carrying capacity of the three major urban agglomerations in Hubei Province. The calculation results are as follows:

• Construction of a weighting matrix

$$\begin{gathered} \mathrm{R}={\left({\mathrm{r}}_{\mathrm{i}\mathrm{j}}\right)}_{\mathrm{m}\times \mathrm{n}}, \\ {\mathrm{r}}_{\mathrm{i}\mathrm{j}}={\mathrm{W}}_{\mathrm{j}}·{\mathrm{x}}_{\mathrm{i}\mathrm{j}}\left(\mathrm{i}=1, 2, \dots ,\mathrm{m}; \mathrm{j}=1, 2, \dots ,\mathrm{n}\right) \end{gathered}$$

(7)

2.

Determining the optimal and worst solutions

$${\mathrm{S}}_{\mathrm{j}}^{+}=\max \left({\mathrm{r}}_{\mathrm{i}\mathrm{j}}, {\mathrm{r}}_{2\mathrm{j}}, \dots , {\mathrm{r}}_{\mathrm{n}\mathrm{j}}\right)$$

(8)

$${\mathrm{S}}_{\mathrm{j}}^{-}=\min \left({\mathrm{r}}_{\mathrm{i}\mathrm{j}}, {\mathrm{r}}_{2\mathrm{j}}, \dots , {\mathrm{r}}_{\mathrm{n}\mathrm{j}}\right)$$

(9)

3.

Calculation of the Euclidean distance of each solution from the optimal solution and the worst solution

$${\mathrm{D}}_{\mathrm{i}}^{+}=\sqrt{\sum _{\mathrm{j}=1}^{\mathrm{n}}{\left({\mathrm{s}}_{\mathrm{j}}^{+}-{\mathrm{r}}_{\mathrm{i}\mathrm{j}}\right)}^2}$$

(10)

$${\mathrm{D}}_{\mathrm{i}}^{-}=\sqrt{\sum _{\mathrm{j}=1}^{\mathrm{n}}{\left({\mathrm{s}}_{\mathrm{j}}^{-}-{\mathrm{r}}_{\mathrm{i}\mathrm{j}}\right)}^2}$$

(11)

4.

Calculating the closeness Ci

$${\mathrm{C}}_{\mathrm{i}}=\frac{{\mathrm{s}\mathrm{e}\mathrm{p}}_{\mathrm{i}}^{-}}{{\mathrm{s}\mathrm{e}\mathrm{p}}_{\mathrm{i}}^{+}+{\mathrm{s}\mathrm{e}\mathrm{p}}_{\mathrm{i}}^{-}},{\mathrm{C}}_{\mathrm{i}}\in [0,1]$$

(12)

2.2.3. Sensitivity Analysis

Multi-criteria decision analysis (MCDA) is a systematic approach used in complex decision-making environments to analyze and select among alternatives [32]. Sensitivity analysis, a tool integral to MCDA, assesses how sensitive decision outcomes are to changes in various parameters. This analysis helps to determine which criteria or weights significantly impact the final decision and how stable these outcomes are under varying assumptions. In our study, we employed MCDA to test the weights derived from the TOPSIS method. Initially, we conducted a Monte Carlo simulation to generate 1000 sets of random weights to explore different weight combinations. These random weights were then applied to the decision matrix for TOPSIS evaluation. We calculated the preference scores for each option under varying weight combinations and ranked them accordingly. Subsequently, we used the fuzzy ranking method to statistically analyze the ranking results from each simulation, calculating the mean ranking and standard deviation for each option. Through these statistical analyses, we assessed the sensitivity of different options to changes in weights, identified the options most sensitive to weight variations, and thus validated the stability and reliability of the weights produced by the TOPSIS method. This comprehensive approach provides a more robust foundation for decision-making.

2.2.4. Kernel Density Estimation Methods

Kernel density estimation is a commonly used technical tool in statistics, which is usually used to visualize the distributional characteristics of a variable in the form of a kernel density plot, where the horizontal coordinates of the kernel density plot indicate the range of the variable’s values, and the vertical coordinates indicate the probability density value of the range, which indicates the likelihood that the variable will occur within the range, and the distributional characteristics of the samples are compared at different points in time. Kernel density estimation involves the use of a continuous density curve to characterize the distribution form of a random variable to estimate its density. Assuming that the density function of the random variable x (water resource carrying capacity) is f(x), the probability density at point x can be obtained from Equation (13):

$$\mathrm{f}(\mathrm{x})=\frac{1}{\mathrm{d}·\mathrm{N}}\sum _{\mathrm{i}=1}^{\mathrm{N}}\mathrm{K}\left(\frac{{\mathrm{x}}_{\mathrm{i}}-\overline{\mathrm{x}}}{\mathrm{d}}\right)$$

(13)

where N is the number of observations; d is the bandwidth; K(-) is the kernel function; X_i is the observed value of water resource carrying capacity level obeying independent distribution; and $\overline{\mathrm{x}}$ is the mean value of water resource carrying capacity level. In this paper, we chose the commonly used Gaussian kernel function to describe the time-series evolution pattern of regional disparities in water resource carrying capacity level in Hubei Province.

2.2.5. Dagum Gini Coefficient Decomposition

In this paper, the Dagum Gini coefficient was chosen to calculate the difference between regions of water resource carrying capacity in Hubei Province, and the Gini coefficient is defined as Equation (14):

$$\mathrm{G}=\frac{1}{2·{\mathrm{n}}^2·\overline{\mathrm{Y}}}·\sum _{\mathrm{j}=1}^{\mathrm{k}}\sum _{\mathrm{h}=1}^{\mathrm{k}}\sum _{\mathrm{i}=1}^{{\mathrm{n}}_{\mathrm{j}}}\sum _{\mathrm{r}=1}^{{\mathrm{n}}_{\mathrm{h}}}\left|{\mathrm{y}}_{\mathrm{j}\mathrm{i}}-{\mathrm{y}}_{\mathrm{h}\mathrm{r}}\right|$$

(14)

where G indicates the overall Gini coefficient. A larger G indicates a larger regional gap in the water carrying capacity level of Hubei Province.

n is the number of cities and k is the number of sub-clusters, which in this paper are the Wuhan urban agglomeration, the Xiang-Shi-Sui-Shen urban agglomeration, and the Yi-Jing-Jing-En urban agglomeration.

nj_(nh) denotes the number of cities within a subgroup of j_(h), j and h are the number of subgroup divisions, and i and r are the number of cities within the subgroups. yji_(yhr) denotes the water resource carrying capacity level of any one of the cities within the Wuhan urban agglomeration of j_(h), the Xiang-Shi-Sui-Shen urban agglomeration, and the Yi-Jing-Jing-En urban agglomeration. $\overline{\mathrm{Y}}$ denotes the average value of water resource carrying capacity level of all cities, as in Equation (15):

$$\overline{\mathrm{Y}}=\frac{\sum _{\mathrm{j}=1}^{\mathrm{k}}\sum _{\mathrm{i}=1}^{{\mathrm{n}}_{\mathrm{j}}}{\mathrm{y}}_{\mathrm{j}\mathrm{i}}}{\mathrm{n}}$$

(15)

To measure the Dagum Gini coefficient for the level of water carrying capacity within and between subgroups, equations can be constructed as shown in Equations (16) and (17):

$${\mathrm{G}}_{\mathrm{j}\mathrm{j}}=\frac{{\left(2·\overline{{\mathrm{Y}}_{\mathrm{j}}}\right)}^{-1}·\sum _{\mathrm{i}=1}^{{\mathrm{n}}_{\mathrm{j}}}\sum _{\mathrm{r}=1}^{{\mathrm{n}}_{\mathrm{j}}}\left|{\mathrm{y}}_{\mathrm{j}\mathrm{i}}-{\mathrm{y}}_{\mathrm{j}\mathrm{r}}\right|}{{\mathrm{n}}_{\mathrm{j}}^2}$$

(16)

$${\mathrm{G}}_{\mathrm{j}\mathrm{h}}=\frac{\sum _{\mathrm{i}=1}^{{\mathrm{n}}_{\mathrm{j}}}\sum _{\mathrm{r}=1}^{{\mathrm{n}}_{\mathrm{h}}}\left|{\mathrm{y}}_{\mathrm{j}\mathrm{i}}-{\mathrm{y}}_{\mathrm{h}\mathrm{r}}\right|}{{\mathrm{n}}_{\mathrm{j}}{\mathrm{n}}_{\mathrm{h}}\left(\overline{{\mathrm{Y}}_{\mathrm{j}}}+\overline{{\mathrm{Y}}_{\mathrm{h}}}\right)}$$

(17)

where G_jj and G_jh denote the Dagum’s Gini coefficients for the level of water carrying capacity within and between subgroups, respectively. $\overline{{\mathrm{Y}}_{\mathrm{j}}}$ + ($\overline{{\mathrm{Y}}_{\mathrm{h}}}$) denotes the average value of water carrying capacity within subgroup j_(h).

Dagum Gini coefficient has a good decomposition performance, which can decompose the source of the gap in the level of water resource carrying capacity into three parts: the contribution of intra-region Gini coefficient to the overall Gini coefficient (G_w), the contribution of inter-region net value gap to the overall Gini coefficient (G_nb), and the contribution of hyper-variable density to the overall Gini coefficient (G_t), and the formulae of the three measures are shown in Equations (18)–(20).

$${\mathrm{G}}_{\mathrm{w}}=\sum _{\mathrm{i}=1}^{{\mathrm{n}}_{\mathrm{j}}}{\mathrm{G}}_{\mathrm{j}\mathrm{j}}{\mathrm{P}}_{\mathrm{j}}{\mathrm{S}}_{\mathrm{j}}$$

(18)

$${\mathrm{G}}_{\mathrm{n}\mathrm{b}}=\sum _{\mathrm{i}=2}^{\mathrm{k}}\sum _{\mathrm{h}=1}^{\mathrm{j}-1}{\mathrm{G}}_{\mathrm{j}\mathrm{h}}\left({\mathrm{P}}_{\mathrm{j}}{\mathrm{S}}_{\mathrm{h}}+{\mathrm{P}}_{\mathrm{h}}{\mathrm{S}}_{\mathrm{j}}\right){\mathrm{D}}_{\mathrm{j}\mathrm{h}}$$

(19)

$${\mathrm{G}}_{\mathrm{t}}=\sum _{\mathrm{i}=2}^{\mathrm{k}}\sum _{\mathrm{h}=1}^{\mathrm{j}-1}{\mathrm{G}}_{\mathrm{j}\mathrm{h}}\left({\mathrm{P}}_{\mathrm{j}}{\mathrm{S}}_{\mathrm{h}}+{\mathrm{P}}_{\mathrm{h}}{\mathrm{S}}_{\mathrm{j}}\right)\left(1-{\mathrm{D}}_{\mathrm{j}\mathrm{h}}\right)$$

(20)

where Pj_(Ph) denotes the proportion of the number of cities within subgroup j_(h), i.e., P_j = nj/n, P_h = nh/n; Sj_(Sh) denotes the proportion of the water resource carrying capacity level within subgroup j_(h), i.e., ${\mathrm{S}}_{\mathrm{j}}=\left({\mathrm{n}}_{\mathrm{j}}\overline{{\mathrm{Y}}_{\mathrm{j}}}\right)/\left(\mathrm{n}\overline{\mathrm{Y}}\right)$, ${\mathrm{S}}_h=\left({\mathrm{n}}_h\overline{{\mathrm{Y}}_h}\right)/\left(\mathrm{n}\overline{\mathrm{Y}}\right)$; and D_jh is the interaction of water resource carrying capacity between subcluster j and subcluster h.

2.3. Indicator Selection

This study employed a systems theory approach, viewing water resources, society, economy, and ecology as integrated components. By fully accounting for the natural water resource endowments in Hubei Province and referencing the sustainable development indicator system proposed by the Committee on Environmental Issues, along with related studies [5,26,33], we established a comprehensive evaluation index system for assessing the water resource carrying capacity of Hubei Province. This system adheres to the principles of scientific validity, availability, diversity, representativeness, hierarchy, and dynamism in the selection of relevant indicators.

Initially, the analytic hierarchy process (AHP) and the entropy weighting method were applied to assign comprehensive weights to 20 evaluation indicators of water resource carrying capacity. Subsequently, the TOPSIS model was used to estimate the water resource carrying capacity of the three major urban agglomerations in Hubei Province from 2005 to 2020. The index system is structured into three levels: the target level, which reflects the specific water resource carrying capacity in Hubei Province; the criterion level, which encompasses four dimensions—water resources, society, economy, and ecology; and the indicator level, which includes specific indicators that adequately describe the attributes of the criterion level. The weights indicate the impact of each indicator on the water resource carrying capacity, as detailed in

Table 1. Evaluation index system of water resource carrying capacity of Hubei Province.

Target Level	Standardized Layer	Specific Indicators (Units)	Calculation Method	Selection of Significance	Weight (%)	Causality
Water carrying capacity	Water resources subsystem	Precipitation intensity (billion m3/km2)	Annual precipitation/area	Reflects the degree of natural precipitation recharge in the study area	15.206	+
		Modulus of water supply (10,000 m3/km2)	Water supply/total regional area	Reflecting the security of the supply of water resources	5.118	+
		Modulus of water production (million (m3/km2))	Water production/total regional area	Reflecting the dynamics of water resources	3.327	+
		Water resources per capita (m3)	Total water resources/total population	Reflecting per capita availability of water resources	4.16	+
		Water resources development and utilization	Water supply/total water resources	Reflecting the extent of water resources development and utilization	0.8	_
	Social subsystem	Population density (persons/km2)	Total population/total regional area	Degree of population concentration	0.521	_
		Urbanization rate	Total urban population/total population	Reflecting the degree of urbanization	1.512	_
		Drainage length per capita (persons/km)	Length of drainage pipes/total population	Reflecting the level of infrastructure	9.325	+
		Per capita urban domestic water consumption (L/person-day)	Urban domestic water consumption/urban population	Reflecting water stress on urban residents	0.253	_
	Economic subsystem	GDP per capita ($)	Total GDP/total population	Reflecting economic developments	6.491	+
		Water consumption per million GDP (m3)	Total water consumption/total GDP	Reflecting the economic benefits of water resources	0.661	_
		Share of tertiary sector	Tertiary GDP/Gross Regional Product	Reflecting the degree of optimization of the regional economic structure	0.581	+
		Agricultural output intensity	Total agricultural GDP/area	Reflecting regional agricultural development	4.372	+
		Water consumption of CNY 10,000 of industrial added value (m3)	Total industrial water consumption/total industrial value added	Reflecting the efficiency of industrial water use	1.466	_
		Average acreage water used for agricultural irrigation (m3)	statistical data	Reflecting agricultural water use efficiency	2.078	_
	Ecological subsystem	Ecological water use rate	Ecological water use/total water use	Reflecting ecological water use	37.814	+
		Wastewater discharge per capita for urban residents (tons)	Domestic wastewater discharge of urban residents/urban population	Reflecting environmental pressures from domestic water use	0.253	_
		Discharge of secondary industry wastewater per unit of output value (tons)	Secondary wastewater discharge/secondary GDP	Reflecting environmental pressures from domestic water use	1.327	_
		Greening coverage in built-up areas	statistical data	Reflecting the degree of greening of the area	0.587	+
		Urban domestic sewage treatment rate	Sewage treatment/total sewage discharge	Reflecting urban wastewater treatment	4.146	+

. A “+” signifies a positive indicator, indicating that a higher value at the indicator level results in a higher comprehensive evaluation value at the criterion level; a “−” signifies a negative indicator, indicating that a higher value at the indicator level results in a lower comprehensive evaluation value at the criterion level. The original data were normalized to eliminate differences among the indicators, ensuring comparability.

The water resource carrying capacity system is a complex system formed by the coupling of the water resources, society, economy, and ecology subsystems. From a systemic relationship perspective, the material conditions of the social subsystem stem from the economic system, the water resources subsystem supports the existence of the social system, and the ecological subsystem provides a living environment for the social subsystem. The water resources subsystem requires an analysis of both the natural endowment of water resources and their production capacity, as well as the status of regional water resource development and utilization. This characterizes the baseline situation of the water resources system in Hubei Province, which benefits from abundant natural precipitation and strong water supply security capacity. Thus, the water resources subsystem includes five indicators: precipitation intensity, the modulus of water supply, and the modulus of water production, among others. The development of the social subsystem is influenced by other systems while also being supported by the water resources subsystem. The social subsystem’s impact on water resource carrying capacity primarily involves population size, social development level, and urban infrastructure. Therefore, this study selected four indicators: population density, urbanization rate, drainage length per capita, and per capita urban domestic water consumption. Hubei Province’s traditional industries account for a significant proportion, and the development of high-water-consuming industrial structures such as agriculture and the secondary industry negatively impacts water resource carrying capacity. Based on this, this study selected six economic development-related indicators: GDP per capita, water consumption per million GDP, and agricultural output intensity, among others. The ecological environment subsystem and the water resources subsystem complement each other. Hubei Province is rich in forest and wetland resources, which play a significant role in conserving water resources, while industrial production and living activities pose certain threats to the ecological subsystem. Accordingly, this study chose five indicators to reflect the carrying capacity of the ecological subsystem: ecological water use rate, per capita urban domestic wastewater discharge, and wastewater discharge per unit output value of the secondary industry, among others.

2.4. Data Sources

The data for the 17 cities and states in Hubei Province covered in this paper mainly came from the “Water Resources Statistics Bulletin” and “Soil and Water Loss Bulletin” of Hubei Province and each city and state from 2005 to 2020, while the data for the social subsystems and the economic subsystems mainly came from the “Hubei Statistical Yearbook”, the “China Urban Construction Statistics Yearbook”, and the EPS database from 2005 to 2020 in Hubei. Considering the factor of data availability, the raw data for 2005–2020 were finally selected. For the very few missing data, the mean substitution method or regression substitution method was used to fill in the gaps.

3. Results

3.1. Analysis of the Temporal Evolution of Water Resource Carrying Capacity

In this study, the kernel density estimation method is applied to explore the distribution dynamic characteristics of four aspects of the distribution location, posture, ductility, and polarization trend of the water resource carrying capacity of Hubei Province and the three major urban agglomerations, and several years with the same intervals are selected for presentation, and the results are shown in Figure 2.

3.1.1. Hubei Province

Figure 2a describes the evolution trend of the water resource carrying capacity in Hubei Province during the sample observation period. In terms of the distribution of the center of gravity, the center of gravity of the curve is obviously shifted to the right. The height of the main peak mainly experiences “a significant decline–a significant rebound–a slight decline”, and the width of the main peak experiences “a significant increase–a slight narrowing–a slight widening”; in general, the height of the main peak decreases and the width of the main peak becomes larger, while the left boundary trends toward obvious convergence, indicating that the water resource carrying capacity of Hubei Province continued to rise during the sample period, with a certain magnitude of the absolute difference expanding, and the growth rate in some cities significantly accelerating. The reason for this may be that Hubei Province in recent years has successively issued normative documents such as “Opinions on the Implementation of the Strictest Water Resources Management System” and “Assessment Measures for the Implementation of the Strictest Water Resources Management System in Hubei Province”, which provide top-level design for water resource protection and management in Hubei Province, and provide rule of law guarantees for the improvement of water resource carrying capacity.

At the same time, it is also observed that the left skirt of the main peak of the carrying capacity distribution curve is significantly shortened, while the right skirt is relatively lengthened, and there continues to be a significant right trailing phenomenon, with a certain degree of broadening of the distribution of extensibility, which means that the gap between the cities with high water resource carrying capacity and the average in the province continues to widen, and there is a significant polarization phenomenon. In addition, specifically in terms of the evolution of its peak, the distribution curve in general experiences a change from a double peak to a single peak, and then from a single peak to a double peak, which means that bipolar and multi-polar polarization tends to strengthen the characteristics. For 2005, the two sides of the main peak are steep and smooth, and there are obvious side peaks, indicating that the water resource carrying capacity of Hubei Province has obvious hierarchical differentiation characteristics, and the right side of the main peak gradually undulates and slows down from 2006 to 2015, indicating that with the establishment of the water resources management system in 2012, the absolute differences in water resource carrying capacity among the three major urban agglomerations of Hubei Province decreased. In 2020, the right side of the main peak rises again, indicating that the water resource carrying capacity of Hubei Province is once again characterized by hierarchical differentiation. This shows that although the water resource carrying capacity of Hubei Province has been significantly improved with the implementation of the strategy of protecting the Yangtze River, due to the differences in the level of economic development, resource endowment, population density, and ecological protection policies of the cities, it is difficult for cities with low water resource carrying capacity to catch up in the short term, and it is likely that the gap with the leading city will continue to widen.

3.1.2. Three Major Urban Agglomerations

Figure 2b,c,d, respectively, depict the evolutionary trends of the water resource carrying capacity of the Yi-Jing-Jing-En urban agglomeration, the Wuhan urban agglomeration, and the Xiang-Shi-Sui-Shen urban agglomeration during the sample observation period.

Firstly, from the distribution position, the kernel density estimation curve of the water resource carrying capacity for all three urban agglomerations shifts to the right, consistent with the overall trend in Hubei Province. This indicates that the water resource carrying capacity of these urban agglomerations generally increased.

Secondly, from the distribution pattern, the height of the main peak of the distribution curve for the Wuhan urban agglomeration decreases year by year, mirroring the overall trend in Hubei Province. This suggests that while the water resource carrying capacity of the Wuhan urban agglomeration rose during the sample period, there was an increasing absolute difference. The Xiang-Shi-Sui-Shen urban agglomeration shows a trend where the height of the main peak first rises and then falls, with the width of the curve becoming larger. This implies an expanding absolute difference in water resource carrying capacity for this urban agglomeration. The Yi-Jing-Jing-En urban agglomeration exhibits a “rising–declining” evolution in the height of the main peak and a “narrowing–widening” evolution in the width, indicating a slight increase in the absolute difference.

From the viewpoint of distribution extension, the right trailing phenomenon in the Wuhan urban agglomeration is consistent with Hubei Province’s trend. In contrast, the Xiang-Shi-Sui-Shen and Yi-Jing-Jing-En urban agglomerations show a mitigating trailing phenomenon, suggesting that while some cities’ water resource carrying capacity lowers the overall level, this trend is easing. The extensibility of the estimated kernel density curves differs among the three agglomerations. The Wuhan urban agglomeration shows a rightward broadening trend, while the Xiang-Shi-Sui-Shen and Yi-Jing-Jing-En urban agglomerations exhibit a “convergence–broadening” change process, indicating a slight convergence overall. This means that the gap between cities with high and low water resource carrying capacity in Xiang-Shi-Sui-Shen and Yi-Jing-Jing-En narrowed to some extent, but this narrowing is not significant.

Finally, in terms of the polarization phenomenon, the three urban agglomerations show significant differences. During the observation period, the distribution of water resource carrying capacity in the Wuhan urban agglomeration evolved from an initial single peak to a “one main side” bimodal state, with the right peak gradually declining, indicating a polarization trend. The Xiang-Shi-Sui-Shen urban agglomeration experienced a “single peak–double peak–single peak–one main side and two sides of multiple peaks” evolution. In 2005, it had a single peak with a low value; in 2010, a double peak emerged; in 2015, it reverted to a single peak; and by 2020, it had a main peak and two side peaks, indicating a clear gradient effect and an ongoing polarization trend. The Yi-Jing-Jing-En urban agglomeration consistently showed a single peak state, with the main peak height rising in 2010 and 2015. In 2020, a side peak emerged, and both the main and side peak heights decreased significantly, indicating a clear internal gradient effect and continued prominent polarization. In summary, the three major urban agglomerations are mainly characterized by decentralized regional agglomeration.

3.2. Analysis of the Spatial Evolution of Water Resource Carrying Capacity

To analyze the spatial evolution of the water resource carrying capacity of the three major urban agglomerations in Hubei Province, based on the established evaluation index system and the results of the proximity degree Ci, and referring to the research results of Deng Zhenghua et al.’s research results [20], this paper divides the water resource carrying capacity into five levels, namely low level, lower level, medium level, higher level and high level, as shown in

Table 2. Carrying capacity interval of water resources at each carrying level.

Carrying Capacity Interval	[0.15, 0.25)	[0.25, 0.35)	[0.35, 0.45)	[0.45, 0.55)	[0.55, 0.65)
Carrying status	Serious overload	Weak overload	Balance	Loadable	Suitable load
Carrying grade	Low level	Lower level	Medium level	Higher level	High level

.

To explore the spatial distribution and evolution of water resource carrying capacity in Hubei Province, the Geographic Information System (GIS) was employed to categorize the water resource carrying capacity into five grades: serious overload, weak overload, balanced, loadable, and suitable load. This study focused on the years 2005, 2010, 2015, and 2020, utilizing spatial visualization and comparative analysis. As illustrated in Figure 3, the carrying capacity of water resources in Hubei Province exhibits significant spatial disparities, with notable inter-regional differences. This distribution is specifically characterized by a pattern of “high in the southwest and low in the northeast”.

In detail, (1) the water resources along the Yangtze River city belt, particularly the Yi-Jing-Jing-En urban agglomeration, have consistently been at the forefront, with Enshi maintaining a high level of water resource carrying capacity. The Wuhan city cluster predominantly exhibits a medium level of capacity. With socio-economic development, the pressure on water resource supply and demand has shifted from densely populated first- and second-tier cities to developing third- and fourth-tier cities. In cities like Yichang, Jingmen, and Wuhan, which are densely populated with industrial zones, the pressure on water resources has been alleviated through advancements in economic development, science and technology, and increased water resource utilization rates. (2) In the north–south non-Yangtze River flow area, the water resource carrying capacity of the Xiang-Shi-Sui-Shen urban agglomeration was at a low level before 2015. During this period, the industrial development in cities such as Shiyan and Xiangyang was in its initial stages, with immature production technologies, low water use efficiency, and excessive exploitation of water resources. This led to negative growth in water resource stocks and a low carrying capacity. After 2015, with improved water use efficiency and the emergence of environmentally friendly industries, the water resource carrying capacity in the Xiang-Shi-Sui-Shen urban agglomeration dropped to a lower level.

3.3. Overall Variance Analysis

To further analyze the differences in the level of water resource carrying capacity in Hubei Province and their sources, this paper applies the Dagum Gini coefficient and decomposition method to calculate and decompose the overall differences. Figure 4 illustrates the evolving trend of the overall differences in water resource carrying capacity in Hubei Province during the observation period. From the trend of the overall Gini coefficient, it is found that the overall gap in water resource carrying capacity among the three major urban clusters in Hubei Province showed a general downward trend before 2019 and an upward countertrend after 2019. This indicates that there is a relatively obvious imbalance among the three major urban clusters in Hubei Province, and the degree of imbalance is continuously deepening.

Specifically, from 2007 to 2018, the regional Gini coefficient of water resource carrying capacity in Hubei Province was relatively stable, basically fluctuating around 0.085, with an average annual decrease of 1.96%. This “stability” was broken in 2019. During the period from 2019 to 2020, the Gini coefficient increased from 0.084 to 0.12, with an increase of 0.036, representing a growth rate of 42.86%. This is mainly because the “Eleventh Five-Year Plan”, “Twelfth Five-Year Plan”, and the “Thirteenth Five-Year Plan for Water Development in Hubei Province” achieved certain results in coordinating water resources, water ecology, water environment, and water disaster control strategies, narrowing the gap between regions within the province. From 2019 to 2020, Hubei Province was affected by the COVID-19 pandemic, and the overall economy declined, leading to a decrease in per capita GDP in the economic subsystem of the corresponding index system, making the imbalance in water resource carrying capacity between regions more apparent. However, the differences in water resource endowment among various urban clusters are significant.

Therefore, narrowing the regional differences in water resource carrying capacity is a lengthy process, and governments at all levels should continue to promote the implementation and realization of water security assurance work with high quality and efficiency.

3.4. Analysis of Variations within Regions

Figure 5 illustrates the evolutionary trend of intra-regional differences in water resource carrying capacity among the three major urban agglomerations in Hubei Province during the examination period.

Regarding the three major urban agglomerations, the regional Gini coefficient of the Yi-Jing-Jing-En urban agglomeration consistently surpasses the overall Gini coefficient, contributing significantly to the considerable disparities in water resource carrying capacity in Hubei Province. The sample mean value of intra-regional differences in the Yi-Jing-Jing-En urban agglomeration is as high as 0.11, ranking first among all three major urban agglomerations and continuing to rise rapidly after 2018. This trend may primarily result from Hubei Province’s successful selection of the ecological protection and restoration project in the Three Gorges area of the Yangtze River as one of the third batch of national pilot projects in 2018. This project covers the three municipalities of Yichang, Enshi, and Jingzhou and has been systematically implemented, contributing to ecological protection and restoration in the Three Gorges area.

The regional Gini coefficients of the Xiang-Shi-Sui-Shen urban agglomeration, except for 2006, do not exceed the overall Gini coefficient of Hubei Province, indicating a relatively low degree of imbalance within the region. The average value of the sample from the Xiang-Shi-Sui-Shen urban agglomeration is 0.06, within a reasonably acceptable range, and started to decline after peaking at 0.14 in 2006. The issue of upgrading heavy industry and manufacturing industry while promoting green development warrants attention to enhance the water resource carrying capacity in the Xiang-Shi-Sui-Shen urban agglomeration.

The sample average value of the Gini coefficient in the Wuhan urban agglomeration is 0.07, second only to the Yi-Jing-Jing-En urban agglomeration, showing an evident fluctuating trend closely linked to Wuhan city’s vigorous implementation of the Yangtze River protection strategy and pollution prevention and control measures. This effort comprehensively advances water environment governance and leads to initiatives such as the Drainage Network Hidden Disease Detection Project in Wuhan City’s Central District. Additionally, regulations such as the “Wuhan Lake Protection Regulations Implementation Rules” in 2005 and the “Wuhan Lake Protection Regulations (Amendment)” in 2015 impose more stringent lake protection requirements. However, the regional Gini coefficient of the Wuhan urban agglomeration from 2019 to 2020 follows a growth curve similar to the overall Gini coefficient trend. This may be attributed to the industrial system paralysis resulting from the COVID-19 pandemic, causing a sharp decline in data related to water consumption per CNY 10,000 of industrial added value and ecological water usage rates, consequently leading to a steep rise in the overall Gini coefficient after 2019.

3.5. Analysis of Interregional Differences

Figure 6 visually depicts the inter-regional differences in water resource carrying capacity among the three major urban agglomerations in Hubei Province and their evolutionary trends during the observation period. The variation trend of inter-regional differences among the three major urban agglomerations can be divided into three stages: a continuous decline from 2005 to 2012, fluctuation from 2013 to 2018, and a sudden increase from 2019 to 2020.

Specifically, between 2005 and 2019, a trend of “slight increase–continuous decrease–relatively stable–sudden increase” was observed between the Yi-Jing-Jing-En urban agglomeration and the Wuhan urban agglomeration. The Wuhan urban agglomeration and the Xiang-Shi-Sui-Shen urban agglomeration showed an initial increase followed by a decrease from 2005 to 2007, followed by an overall decline until 2019, with a sudden increase thereafter. Between the Yi-Jing-Jing-En urban agglomeration and the Xiang-Shi-Sui-Shen urban agglomeration, there was an upward trend between 2005 and 2006, followed by a fluctuating downward trend from 2006 to 2019, with a sudden increase after 2019.

Since the implementation of the Scientific Development Concept strategy in 2003, Hubei Province has attached great importance to water resources protection. In 2005, the “Wuhan Lake Protection Regulations Implementation Rules” were promulgated, and in 2012, the Standing Committee of the Provincial People’s Congress issued local regulations—the “Hubei Province Lake Protection Regulations”—gradually achieving the goal of “no pollution, no reduction in quantity, no shrinking of area” of water bodies. In 2017, a midterm evaluation of the implementation of the “13th Five-Year Plan” for environmental protection in Hubei Province was conducted, and a three-year action plan for pollution prevention and control in Hubei Province was formulated and implemented, comprehensively improving the quality of the atmospheric, water, and soil environments, and reducing the inter-regional differences in water resource carrying capacity among the three major urban agglomerations. The Gini coefficient between the Wuhan urban agglomeration and the Xiang-Shi-Sui-Shen urban agglomeration decreased from 0.086 in 2005 to 0.067 in 2013, a decrease of 0.019, or approximately 22.09%. The Gini coefficient between the Yi-Jing-Jing-En urban agglomeration and the Xiang-Shi-Sui-Shen urban agglomeration decreased from 0.137 in 2005 to 0.087 in 2013, a decrease of 0.05, or approximately 36.50%. In 2020, cities such as Yichang, Jingmen, Xiangyang, and Suizhou in Hubei Province were affected by heavy rainfall, leading to a significant increase in indicators such as precipitation intensity and exacerbating inter-regional differences. The sample mean of the Gini coefficient between the Yi-Jing-Jing-En urban agglomeration and the other two urban agglomerations is as high as 0.11, while the sample mean of inter-regional differences is 0.08. Although the water demand in the Wuhan urban agglomeration is high, sufficient technology and funds ensure relatively high water resource utilization efficiency. On the other hand, the Xiang-Shi-Sui-Shen urban agglomeration is influenced by the natural environment and urban industrial development, resulting in lower water resource reserves and utilization efficiency, thus making the differences in water resource carrying capacity more pronounced.

3.6. Analysis of Variance Contribution

The analysis of the decomposition of differences reveals that from 2005 to 2020, the average annual contribution rates of intra-group differences, inter-group differences, and hyper-variation density to the overall differences in the water resource carrying capacity of the three major urban agglomerations in Hubei Province are 34.58%, 20.13%, and 45.29%, respectively, indicating that hyper-variation density is the primary source of regional disparities. As depicted in Figure 7, the contribution rate of intra-group differences shows a stable and slight upward trend over the sample period, rising from 31.65% in 2005 to 35.70% in 2020, second only to hyper-variation density, making it a significant factor contributing to regional differences in water resource carrying capacity. Inter-group differences increased from 0.014 in the initial sample period to 0.04 in 2020, yet their contribution rate consistently remained below 33.65% and continued to diminish, suggesting that differences between urban agglomerations are not the primary cause of regional differences in water resource carrying capacity. Although the contribution rate of hyper-variation density decreased from 54.71% in 2005 to 30.65% in 2020, a decline of 24.06%, its contribution rate to overall differences remained above 30.65% in the long term.

Therefore, hyper-variation density is the primary source of overall differences in water resource carrying capacity among the three major urban agglomerations in Hubei Province over the sample period, indicating a prominent issue of regional overlap and overlap in water resource carrying capacity among these urban agglomerations, ultimately leading to relatively large disparities between cities with high water resource carrying capacity and those with low water resource carrying capacity. This may be attributed to the following two reasons: (1) Economic development and policy differences among overlapping cities in Hubei Province affect the economic subsystem and social subsystem, thereby widening the gap in water resource carrying capacity. For example, the siphoning effect causes economically developed areas to absorb more advanced water-saving irrigation technologies and more substantial financial support from neighboring areas, thereby widening the gap in water resource carrying capacity. (2) Pollution diffusion increases environmental governance pressure on certain cities. For example, water pollution from upstream spreads to downstream cities, and industrial upgrades lead some cities to relocate heavily polluting enterprises to other cities, which may further widen the overall gap in water resources through the resulting water pollution.

4. Discussion

This article provides a scientific basis for determining planning objectives by examining the maximum population and economic scale that water resources can support. The selection of water carrying capacity indicators, the construction of the indicator system, the calculation of weights, and the application of quantitative evaluation methods all introduce uncertainties into the estimates of water carrying capacity. Therefore, it is necessary to analyze the rationality of the results estimated in this article. By testing the results estimated using the TOPSIS model based on earlier research findings (in parts of Hubei), it is found that the analysis of the water carrying capacity of Hubei’s three major urban agglomerations (e.g., Yi-Jing-Jing-En ranked first) is generally consistent with the values measured by Yang et al. using the variable fuzzy method [34], and the estimated results for water carrying capacity in Shiyan provided by this article are similar to those of Cheng et al. based on the DPSIRM framework [35]. This suggests that the water carrying capacity indicator system constructed in this article and the application of the TOPSIS model to estimate water carrying capacity in Hubei Province are relatively ideal.

This article outlines the regional disparities in water carrying capacity, providing guidance for spatial layout and control. The research shows significant regional differences in water carrying capacity in Hubei Province, with Yi-Jing-Jing-En performing better than Wuhan and Xiang-Shi-Sui-Shen. These disparities are primarily attributed to variations in ecological water usage rates within the ecosystem, which contrasts with the findings of Lv Bo et al. for the eastern region of Heilongjiang Province [36]. Potential reasons for these differences include (1) significant regional and climatic differences between the two areas, (2) notable disparities in water resource abundance, and (3) a higher proportion of agricultural water use in the eastern region of Heilongjiang compared to Hubei, where both agricultural and ecological water usage are high. Hence, the research results more accurately match the specific conditions of these two locations.

Existing studies primarily focus on evaluating water carrying capacity and struggle to represent the spatial disparities constrained by water resources, which do not meet the precision requirements of spatial planning. However, this article evaluates from an urban agglomeration perspective, better serving urban spatial planning layouts. It is noteworthy that for the case study area in the subtropical monsoon humid climate zone, an indicator system consisting of 20 indicators is designed, establishing a weight structure primarily based on ecological and economic categories. This can provide a reference for other regions to conduct related research, ensuring that the selection of evaluation indicators and weights should be comprehensively determined based on local water resource conditions, water user characteristics, and water infrastructure foundations.

5. Conclusions

Employing a multidimensional evaluation index system developed using the entropy weighting method and the TOPSIS model, this research work investigates the interactions among water resources, society, economics, and ecology. This study utilizes advanced analytical tools such as kernel density estimation, ArcGIS for spatial visualization, and the Dagum Gini coefficient method to explore the spatiotemporal variations, dynamic evolution, and contributing factors to the water resource carrying capacity of the three major urban agglomerations in Hubei Province from 2005 to 2020. The principal findings are as follows:

From a temporal perspective, the overall water resource carrying capacity in Hubei Province gradually improved from severe overload to just overload between 2005 and 2020. In recent years, the coordination between water supply and demand in these urban agglomerations has been enhanced, indicating a clear upward trend in water resource carrying capacity. This study also revealed a polarizing trend in water resource carrying capacity among the urban agglomerations, as evidenced by kernel density estimation. Before 2015, the water resource carrying capacity in the Xiang-Shi-Sui-Shen urban agglomeration was relatively low. After 2015, with improvements in water resource utilization efficiency and the rise of environmentally friendly industries, the water resource carrying capacity of this agglomeration dropped to lower levels. The industrial structure of the Xiang-Shi-Sui-Shen urban agglomeration urgently needs transformation, as it still relies on high water-consuming and pollution-intensive industries, leading to significant discrepancies between water resource conditions and socio-economic water demands, resulting in a prolonged low level of water resource carrying capacity.

From a spatial distribution perspective, there is a pronounced spatial imbalance in water resource carrying capacity among Hubei’s three major urban agglomerations, characterized by “high in the southwest and low in the northeast”. Despite a consistently high Gini coefficient that continues to decrease, indicating a narrowing gap, a significant imbalance persists, highlighting overall and localized vulnerabilities in Hubei’s water resource management strategy. The Gini coefficient ranking for water resource carrying capacity among the three urban agglomerations showed that the Yi-Jing-Jing-En agglomeration has superior capacity compared to the Wuhan and Xiang-Shi-Sui-Shen agglomerations, with the overall disparity decreasing before 2019. However, since 2019, this trend has reversed, increasing under severe natural environmental constraints, with some cities experiencing a decline in water resource carrying capacity due to uncoordinated water resource utilization and social development.

From the perspective of influencing factors, the ecological water use rate has the most significant impact on the water resource carrying capacity of Hubei Province, followed by precipitation intensity, per capita drainage pipe length, and per capita GDP. The use of water in agriculture and industry, along with domestic water use, exacerbates water resource pressures and is the primary cause of the water supply demand imbalance in Hubei Province. Establishing how to coordinate water ecosystem protection and promote sustainable socio-economic and ecological development remains a common challenge faced by the three major urban agglomerations in Hubei Province.

6. Policy Recommendations

Based on the aforementioned analysis and conclusions, differentiated strategies should be formulated to reduce the vulnerability of the overall and localized water resource carrying capacity in the three major urban agglomerations of Hubei Province.

Firstly, guided by the enhancement of coordination, the protection of the water ecological environment should be comprehensively planned. Given the issue of low ecological water usage rates in certain cities, the spatial and temporal layout of water resources must be thoroughly planned. Initially, the connectivity system of rivers and lakes among the three urban agglomerations should be improved, constructing an ecological community of water resources to achieve a “long vine and melon, spider web layout, interconnected” water management pattern. By leveraging the strengths to support the weaknesses, the water resource carrying capacity of tail-end urban agglomerations such as Xiang-Shi-Sui-Shen can be enhanced, thereby reducing regional disparities. Additionally, as the core entity in water resource governance, the government needs to eliminate coordination pressures arising from cross-departmental interests to achieve effective water ecosystem governance. This can be accomplished by establishing a robust intergovernmental collaboration network and decomposing governance responsibilities through legislation, thereby forming a “responsibility chain” among different government departments to ensure effective coordination as a community of shared interests.

Secondly, focusing on addressing weaknesses and accelerating industrial transformation and upgrading is crucial. To tackle the issue of widening absolute regional disparities in water resource carrying capacity and the emerging trend of polarization, priority should be given to improving the carrying capacity of the Xiang-Shi-Sui-Shen urban agglomeration. Initially, sustained efforts should be made in project implementation to accelerate the advancement of key water source projects, ecological water usage, wastewater treatment technologies, and related special initiatives in the Xiang-Shi-Sui-Shen urban agglomeration. This will improve the existing water resource management and distribution model, enhance supply security, and reduce industrial water use pressure. Moreover, industrial rectification, especially the adjustment of high water-consuming industries and the green transformation of pollution-intensive industries, is urgently needed. Promoting the green transformation and upgrading of industrial enterprises should initiate a shift from scale-driven to technology innovation-driven approaches. Industrial structure should be adjusted, production water tools should be renovated, and production water efficiency should be improved. The technological transformation of water-saving in high energy-consuming industries should be accelerated, and outdated wastewater treatment equipment should be eliminated.

Thirdly, implementing precise strategies to promote balanced water resource allocation is essential. Addressing the spatial imbalance in water resource carrying capacity in Hubei Province requires full consideration of the differences in water resource endowments among different urban agglomerations. Adhering to the principle of “adapting measures to local conditions and advancing by zones”, efforts should focus on improving sewage treatment rates and water resource utilization efficiency in the Wuhan urban agglomeration through continuous efforts in management and technology, thereby promoting the construction of water-saving cities. In the Yi-Jing-Jing-En urban agglomeration, a scientifically improved water resource allocation pattern should be pursued, coordinating the surrounding basins of the urban agglomeration, achieving cascade development of all river segments, scientifically coordinating the storage and development of water resources, constructing water conservancy hubs, and improving intra-regional water allocation methods. Additionally, strengthening the coordination of main and tributary rivers and flood and drought water resource management between regions is necessary. For the relatively lower water resource carrying capacity of the Xiang-Shi-Sui-Shen urban agglomeration, it is essential to scientifically plan the regional water usage, ensuring the water needs for key industries, basic livelihoods, and ecological purposes.