How Integrated Are Water and Food Systems in China? Assessing Coupling Mechanisms and Geographic Disparities

Abstract

Water resources are of vital importance to human survival and development. This study systematically analyzed the coupling coordination mechanism between China’s food security (FS) and water resource management (WRM) from 2010 to 2022 using the TOPSIS model, Dagum Gini coefficient, coupling coordination model, and fixed effects regression model. The results indicate that FS exhibited a “U-shaped” evolution: an average annual decline of 1.4% before 2017 followed by recovery to 2.39% due to policy optimization and technological upgrades, though significant regional disparities persisted with 15 provinces maintaining ecological vulnerability scores below 0.3. WRM showed an average annual increase of 1.33%, later accelerating to 1.76% driven by projects like the South-to-North Water Diversion Project, which significantly improved 28 provinces. The FS-WRM coupling coordination degree escalated from mild imbalance to near imbalance, forming a spatial pattern of “central region leading–northeast following–eastern fluctuation–western catching up”, with 10 provinces reaching barely coordinated levels in 2022. The study reveals that policy support, infrastructure development, technological innovation, and management model transformation are key influencing factors for FS-WRM coupling coordination.

1. Introduction

China’s total water resources are relatively abundant, but per capita water resources are far below the world average, making it a country with relatively scarce water resources. Moreover, the spatial and temporal distribution of water resources in China is highly uneven [1,2]. Temporally, precipitation is concentrated in summer with significant inter-annual variations, leading to an uneven intra-annual distribution of water resources and frequent seasonal droughts and floods [3] (See Figure 1). Spatially, southern regions are relatively water-rich, while northern regions, particularly the North China and Northwest regions, face severe water shortages, with a prominent contradiction between water supply and demand [4,5]. This uneven distribution poses significant challenges to the rational development, utilization, and management of water resources. With China’s continuous population growth and rapid economic development, food demand has been on the rise. On the one hand, the increase in population directly leads to a rise in total food consumption. On the other hand, improvements in living standards have raised people’s demands for food quality and variety, further stimulating the development of food production [6,7]. However, food production heavily relies on water resources, especially the development of irrigated agriculture, which requires significant amounts of water. In water-scarce northern regions, to meet food production demands, large amounts of groundwater have been extracted, leading to issues such as declining groundwater levels and land subsidence [8,9]. Additionally, low agricultural water use efficiency exacerbates the water resource shortage. Traditional irrigation methods, such as flood irrigation, result in severe water waste, making it difficult to efficiently utilize limited water resources and further increasing the pressure on water resources.

Globally, water, energy, and food form an intricate system that profoundly influences the achievement of sustainable development. Many countries and regions face the issue of uneven water resource distribution, which makes the interconnectedness among water, energy, and food even more pronounced [10]. For instance, in arid regions, more energy is often consumed to extract groundwater or desalinate seawater in order to meet the demands of food production and consumption. Additionally, energy-intensive food production and processing techniques further exacerbate the pressure on water resources. At the international level, water resource management and food production systems frequently encounter coordination deficiencies, with widespread issues such as multi-departmental management, unclear responsibilities, and coordination difficulties. For example, in agricultural irrigation, the lack of a unified management and coordination mechanism often leads to excessive water use in upstream areas, resulting in water shortages in downstream regions [11,12]. Meanwhile, with the advancement of urbanization, the expansion of urban construction land encroaches on a significant amount of agricultural land, further compressing the space for food production and posing challenges to the integrated development of water resource management and food security. To address these challenges, the international community is placing increasing emphasis on the rational development, utilization, and management of water resources, as well as the sustainability of food production systems. Measures such as improving agricultural water use efficiency, promoting water-saving technologies, and optimizing the agricultural industrial structure have become crucial pathways for alleviating water resource pressure and ensuring food security. Simultaneously, advancing energy efficiency and the utilization of renewable energy sources also contribute to reducing reliance on and consumption of water resources. In terms of food production, the world is gradually transitioning towards more sustainable models, such as precision agriculture and ecological agriculture, to reduce water consumption while enhancing food yield and quality. These measures not only help balance the relationships among water, energy, and food but are also vital for achieving global sustainable development goals. Through these comprehensive measures, the interactions among water, energy, and food can be optimized, propelling the world towards a more sustainable development trajectory.

China’s current water resource management system and food production system are somewhat uncoordinated. In water resource management, there are issues of multi-departmental management, unclear responsibilities, and coordination difficulties. Water resource management involves multiple departments such as water conservancy, environmental protection, and agriculture, with a lack of effective communication and collaboration among them, leading to numerous contradictions and problems in the planning, allocation, utilization, and protection of water resources. For example, in the allocation of agricultural irrigation water, due to the lack of a unified management and coordination mechanism, upstream areas often overuse water, causing downstream areas to face water shortages [13]. In the food production system, agricultural production methods are relatively extensive, with a high dependence on water resources, a lack of water-saving awareness, and the application of water-saving technologies. Meanwhile, the agricultural industrial structure is unreasonable, with excessive areas planted with highly water-consuming crops, exacerbating the water resource shortage [14]. Furthermore, with the acceleration of urbanization, urban construction land continues to expand, encroaching on a large amount of agricultural land, further compressing the space for food production, and posing challenges to the coordinated development of water resource management and food security [15].

Scholars in the field of water resource management and food security have conducted early research and achieved rich results. The study of water resources and food security emerged as a response to escalating global ecological crises and the growing urgency for sustainable development. From the publication of the seminal work “The Limits to Growth” in 1972, through the World Bank’s 1990s research on water–energy–agriculture linkages in countries like India and Mexico, to the emergence of new concepts such as “virtual water” and “water footprint,” academic understanding of the water–energy–food nexus has deepened significantly. The 2011 Bonn Conference on Water–Energy–Food Security first proposed the interconnected “bonding relationship” between water, energy, and food (WEF), sparking extensive research into WEF coupling systems—including single-system, dual-system, and triple-system configurations—resulting in numerous studies and international collaborations. For instance, in 2013, the Asia-Pacific Water–Energy–Food Linkages Report was released, proposing that the core of WEF linkages lies in comprehensively quantifying the interrelationships among water, energy, and food [16]. The AsianDevelopmentBank (ADB) highlighted the social commodity nature of water resources and emphasized the need to establish a rational and well-developed market for water resource utilization [17]. The UN Food and Agriculture Organization (FAO) developed a decision-making framework for WEF coupling systems in 2014 [18], while the World Water Forum incorporated ecosystems into this framework in 2018. Statistics show that nearly 300 organizations participated in WEF bonding research between 2011 and 2015, particularly in China and the United States. Many scholars have also studied China as a case study, with the Woodrow Wilson International Center for Scholars in the U.S. analyzing China’s WEF-related issues in 2015 and establishing a WEF Security Index framework. In recent years, many scholars have built models to explore the impact of climate change on water security or water–energy–food security, demonstrate the importance of sustainable development, and put forward solutions such as transboundary water resource management and option contracts to provide reference for policy making. In theoretical research, some scholars have put forward the connotation definition, organizational composition, and operation mechanism of the WEF bond relationship, while some scholars have proposed theories such as the sustainable utilization of water resources and food security guarantees, providing a theoretical basis for research in these areas [19,20,21,22,23,24]. Despite the extensive research conducted by scholars in this field, there are still notable deficiencies. In terms of research content, many scholars have studied the coupling mechanism, influencing factors, simulation, and safety risk measurement of WEF bond relationships. However, most existing studies focus on either water resources or food security independently, with relatively few studies examining the coupled and coordinated relationship between the two. This study, however, innovatively delves into the intricate interactions and internal laws governing the coupling and coordination of water resource management and food security in China, filling a significant gap in the literature. In terms of research methods, although various models and methods have been utilized in previous studies, their capabilities in simulating and predicting complex systems, especially when considering multi-factor and multi-scale interactions, still need improvement. This study employs a comprehensive approach that integrates the entropy weight method, TOPSIS model, and coupling coordination model to assess the coupling mechanisms and geographic disparities between water resource management and food security in China from 2010 to 2022. This methodological innovation allows for a more nuanced and accurate understanding of the dynamic relationships between these two systems. In terms of research scale, existing studies mostly concentrate on the national or regional level, with relatively little research on the coupled and coordinated relationships between water resource management and food security at the grassroots level, lacking a systematic analysis of the relationships between different scales. This study not only analyzes the situation at the national and regional levels but also delves into provincial-level data, providing a more granular and systematic analysis of the coupling and coordination mechanisms across different scales.

Therefore, it is necessary to further strengthen research on the coupling and coordination mechanisms and key driving factors of water resource management and food security in China, as this study does, to fill the gaps in existing studies and provide a more comprehensive understanding of the complex relationships between these two critical systems. This study’s innovative approach and findings contribute significantly to the field, offering valuable insights for policymakers and researchers alike.

2. Overview of the Study Area

China is located in eastern Asia and the western Pacific, with a total land area of approximately 9.6 million square kilometers and a total sea area of approximately 4.73 million square kilometers. This study focuses on 31 provincial-level administrative regions in China, excluding Taiwan Province and the two Special Administrative Regions. These 31 regions comprise 22 provinces, five autonomous regions, and four municipalities directly under the central government. Significant differences exist among regions in terms of geographical location, topography, and climatic conditions, which have profound impacts on the distribution of water resources and the pattern of food production (See Figure 2). China is the most populous country in the world, and its huge population size poses enormous pressure on food demand. With rapid economic development and accelerated urbanization, a large number of rural populations have migrated to cities, leading to an increase in urban population size. As of 2024, China’s urbanization rate has exceeded 67%, and the increase in urban population has led to changes in food consumption structures and total demand, posing new requirements for food security guarantees [25]. Meanwhile, water resources have faced varying degrees of pressure during the urbanization process, further exacerbating regional development tensions.

3. Data Sources and Construction of the Index System

The study period selected for this research is from 2010 to 2022. The data mainly originate from the China Statistical Yearbook, China Environmental Statistical Yearbook, China Water Resources Statistical Yearbook, China Rural Statistical Yearbook, and government statistical bulletins from 2011 to 2023. The selection of the following indicators follows the selection of systematism, representativeness, accuracy, and availability and refers to relevant research methods in the academic field (See

Table 1. Representation of indicators.

Dimension	Index	Unit	Index Description
Food security	Production input	Grain sown area	Thousand hectares, reflecting the scale of food production
		Grain output	Million tons, the core indicator measuring food production achievements
		Fertilizer application volume	Million tons, an important input for increasing crop yields
		Pesticide use volume	Million tons, crucial for safeguarding grain yields and quality
		Agricultural film use volume	Million tons, used to cover farmland to improve soil conditions
		Agricultural machinery power	Ten thousand kilowatts, reflecting the level of agricultural mechanization
	Disaster and resources	Disaster-affected area	Thousand hectares, indicating the area of crops affected by natural disasters
	Socioeconomic	Rural residents’ disposable income	CNY, reflecting the living standards of rural residents
		Rural population	Ten thousand people, indicating the size of the main food consumption group
Water resource management	Water use and penetration rate	Per capita comprehensive water use	Cubic meters, reflecting overall water use levels
		Coverage rate of water supply in villages	%
		Urban water use penetration rate	%
	Water resources and emission	Per capita water resources	Cubic meters per person, reflecting water resource abundance
		Per capita domestic water use	Cubic meters per person, an important indicator of residents’ domestic water needs
		Per capita chemical oxygen demand (COD) emission	Tons per person, a measure of water pollution levels

) [26]. For missing data, a linear interpolation method is mainly used to estimate and fill in values.

The analysis of food security is conducted through three dimensions: production inputs, disasters and resources, and socioeconomic factors. In the production inputs dimension, grain cultivation area and yield serve as fundamental indicators that directly reflect agricultural scale and efficiency. The use of fertilizers, pesticides, agricultural films, and farm machinery power demonstrate the intensiveness level and technological advancement of modern agriculture. The disaster and resources dimension focuses on affected areas as a key indicator, reflecting the vulnerability of the food system and its climate risk resilience. From the socioeconomic perspective, rural residents’ disposable income and population size influence the sustainability of food production through economic incentives and labor supply dynamics, respectively.

The development of a water resource management indicator system should holistically reflect three key dimensions: resource endowment, utilization efficiency, and sustainability. In terms of water consumption and accessibility, per capita comprehensive water use measures regional resource intensity, closely tied to industrial structures (e.g., agricultural water consumption ratio). Village water supply coverage and urban water access rates demonstrate infrastructure equity, revealing disparities in public services between urban and rural areas. Regarding water resources and emissions, per capita water availability serves as the core indicator of resource carrying capacity. Per capita domestic water consumption reflects residents’ consumption patterns, while per capita COD emissions directly quantify water environmental stress.

4. Model Methods

The software versions for the Entropy Weight Method, TOPSIS model, and coupling coordination model are sourced from SPSSAU (website: https://spssau.com/index.html). Dagum Gini Coefficient used MATLAB 2023a.

4.1. Entropy Weight Method

The entropy weight method is extensively utilized in various multi-objective comprehensive evaluation studies. It is primarily employed to assign weights to each index within an index system for two key reasons: firstly, to uncover insignificant (or hidden) information within each index, and secondly, to prevent selective bias resulting from negligible differences among indices, thereby comprehensively and objectively reflecting the situations of the indices. Generally, the larger the entropy value, the smaller the entropy weight, which suggests a lower importance of the index. The entropy weight method typically leverages data entropy information, that is, the quantity of information, for weight calculation [27].

(1)

Normalization Processing

For the construction of a standardized evaluation matrix, let the evaluation index matrix be defined as follows:

$$\mathrm{V}=\left[\begin{array}{cccc}{\mathrm{v}}_{11}& {\mathrm{v}}_{12}& \dots & {\mathrm{v}}_{13}\\ {\mathrm{v}}_{21}& {\mathrm{v}}_{22}& \dots & {\mathrm{v}}_{2\mathrm{n}}\\ \text{⋮}& \text{⋮}& \text{⋮}& \text{⋮}\\ {\mathrm{v}}_{\mathrm{m}1}& {\mathrm{v}}_{\mathrm{m}2}& \dots & {\mathrm{v}}_{\mathrm{mn}}\end{array}\right]$$

(1)

To construct a standardized evaluation matrix, the original data needs to be normalized. For positive indices (the larger the better) and negative indices (the smaller the better) the following equations are used:

$${\mathrm{r}}_{\mathrm{i}\mathrm{j}}={\displaystyle \frac{{\mathrm{v}}_{\mathrm{i}\mathrm{j}}-\min ({\mathrm{v}}_{\mathrm{i}\mathrm{j}})}{\max ({\mathrm{v}}_{\mathrm{i}\mathrm{j}})-\min ({\mathrm{v}}_{\mathrm{i}\mathrm{j}})}}$$

(2)

$${\mathrm{r}}_{\mathrm{ij}}={\displaystyle \frac{\max ({\mathrm{v}}_{\mathrm{ij}})-{\mathrm{v}}_{\mathrm{ij}}}{\max ({\mathrm{v}}_{\mathrm{ij}})-\min ({\mathrm{v}}_{\mathrm{ij}})}}$$

(3)

$$\mathrm{R}=\left[\begin{array}{cccc}{\mathrm{v}}_{11}& {\mathrm{v}}_{12}& \dots & {\mathrm{v}}_{13}\\ {\mathrm{v}}_{21}& {\mathrm{v}}_{22}& \dots & {\mathrm{v}}_{2\mathrm{n}}\\ \text{⋮}& \text{⋮}& \text{⋮}& \text{⋮}\\ {\mathrm{v}}_{\mathrm{m}1}& {\mathrm{v}}_{\mathrm{m}2}& \dots & {\mathrm{v}}_{\mathrm{mn}}\end{array}\right]$$

(4)

In the formula, V represents the initial evaluation matrix, where vij denotes the initial value of the i-th indicator in the j-th year; R represents the standardized evaluation matrix, where rij denotes the standardized value of the i-th indicator in the j-th year; i = 1, 2, …, m, with m being the number of evaluation indicators; and j = 1, 2, …, n, with n being the number of evaluation years [28].

(2)

Index Weight Determination

By normalizing the information utility value, the entropy weight of each index can be obtained. The calculation formula is as follows:

$${\mathrm{w}}_{\mathrm{i}}={\displaystyle \frac{1-{\mathrm{H}}_{\mathrm{i}}}{\mathrm{m}-\sum _{\mathrm{i}=1}^{\mathrm{m}}{\mathrm{H}}_{\mathrm{i}}}}$$

(5)

$${\mathrm{H}}_{\mathrm{i}}=-{\displaystyle \frac{1}{\ln (\mathrm{n})}}\sum _{\mathrm{j}=1}^{\mathrm{n}}{\mathrm{f}}_{\mathrm{ij}}\ln {\mathrm{f}}_{\mathrm{ij}}$$

(6)

$${\mathrm{f}}_{\mathrm{ij}}={\displaystyle \frac{{\mathrm{r}}_{\mathrm{ij}}}{\sum _{\mathrm{j}=1}^{\mathrm{n}}{\mathrm{r}}_{\mathrm{ij}}}}$$

(7)

4.2. TOPSIS Model Overview

The TOPSIS model is a commonly used evaluation method in socioeconomic systems to solve multi-objective decision-making problems. It evaluates and comprehensively analyzes various socioeconomic problems by judging the proximity of index values to positive and negative ideal values based on specific measurements in unit values [29].

(3)

Construction of an Evaluation Matrix Based on Entropy Weights

To reflect the authenticity and objectivity of the evaluation matrix, a weighted normalized evaluation matrix Y is constructed using entropy weights wi. The specific calculation formula is as follows:

$$\mathrm{Y}=\left[\begin{array}{cccc}{\mathrm{y}}_{11}& {\mathrm{y}}_{12}& \dots & {\mathrm{y}}_{1\mathrm{n}}\\ {\mathrm{y}}_{21}& {\mathrm{y}}_{22}& \dots & {\mathrm{y}}_{2\mathrm{n}}\\ \text{⋮}& \text{⋮}& \text{⋮}& \text{⋮}\\ {\mathrm{y}}_{\mathrm{m}1}& {\mathrm{y}}_{\mathrm{m}2}& \dots & {\mathrm{y}}_{\mathrm{mn}}\end{array}\right]=\left[\begin{array}{cccc}{\mathrm{r}}_{11}\text{⋅}{\mathrm{w}}_1& {\mathrm{r}}_{12}\text{⋅}{\mathrm{w}}_1& \dots & {\mathrm{r}}_{1\mathrm{n}}\text{⋅}{\mathrm{w}}_1\\ {\mathrm{r}}_{21}\text{⋅}{\mathrm{w}}_2& {\mathrm{r}}_{22}\text{⋅}{\mathrm{w}}_2& \dots & {\mathrm{r}}_{2\mathrm{n}}{\mathrm{w}}_2\\ \text{⋮}& \text{⋮}& \text{⋮}& \text{⋮}\\ {\mathrm{r}}_{\mathrm{m}}{\mathrm{w}}_{\mathrm{m}}& {\mathrm{r}}_{\mathrm{m}2}\text{⋅}{\mathrm{w}}_{\mathrm{m}}& \dots & {\mathrm{r}}_{\mathrm{mn}}\text{⋅}{\mathrm{w}}_{\mathrm{m}}\end{array}\right]$$

(8)

(4)

Solution of Positive and Negative Ideal Solutions

First, the positive and negative ideal solutions for each index need to be determined. The positive ideal solution represents the optimal value for each index in the sample, while the negative ideal solution represents the worst value. Let Y+ be the maximum value of the ith index in the jth year within the evaluation data, i.e., the optimal preference scheme, known as the positive ideal solution. Y- is the minimum value of the ith index in the jth year within the evaluation data, i.e., the worst preference scheme, known as the negative ideal solution. The specific calculation methods are as follows [29]:

$${\mathrm{Y}}^{+}=\left\{\max ({\mathrm{y}}_{\mathrm{ij}})|\mathrm{i}=1,2,\dots ,\mathrm{m}\right\}=\left\{{\mathrm{y}}_1^{+},{\mathrm{y}}_2^{+},\dots ,{\mathrm{y}}_{\mathrm{n}}^{+}\right\}$$

(9)

$${\mathrm{Y}}^{-}=\left\{\min ({\mathbf{y}}_{\mathrm{ij}})|\mathrm{i}=1,2,\dots ,\mathrm{m}\right\}=\left\{{\mathbf{y}}_1^{-},{\mathbf{y}}_2^{-},\dots ,{\mathbf{y}}_{\mathrm{n}}^{-}\right\}$$

(10)

This study uses the Euclidean distance calculation formula. Let Dj+ be the distance between the ith index and yi+ and Dj− be the distance between the ith index and yi−. The calculation methods are as follows:

$${\mathrm{D}}_{\mathrm{j}}^{+}=\sqrt{\sum _{\mathrm{i}=1}^{\mathrm{m}}({\mathbf{y}}_{\mathrm{i}}^{+}-{\mathbf{y}}_{\mathrm{ij}}{)}^2}$$

(11)

$${\mathrm{D}}_{\mathrm{j}}^{-}=\sqrt{\sum _{\mathrm{i}=1}^{\mathrm{m}}({\mathbf{y}}_{\mathrm{i}}^{+}-{\mathbf{y}}_{\mathrm{ij}}{)}^2}$$

(12)

$${\mathrm{T}}_{\mathrm{j}}={\displaystyle \frac{{\mathrm{D}}_{\mathrm{j}}^{-}}{{\mathrm{D}}_{\mathrm{j}}^{+}-{\mathrm{D}}_{\mathrm{j}}^{-}}}$$

(13)

In the above formula, yij represents the weighted normalized value of the i-th indicator in the j-th year. yi+ and yi− Denote the best ideal solution and the worst ideal solution, respectively, for the i-th indicator among the values taken over n years. Tj represents the degree to which the evaluation value in the j-th year approaches the optimal ideal value, commonly referred to as the closeness degree. Its value generally ranges between 0 and 1. The larger Tj is, the closer that year is to the optimal level. When Tj is closer to 1, the evaluation effect is higher; when Tj is closer to 0, the evaluation is lower. The closeness degree is used to reflect the comprehensive evaluation value and determine the order of superiority and inferiority [30].

4.3. Coupling Coordination Model

The concept of coupling originated in physics and refers to the strength of interdependence between two or more entities. Drawing on existing research, this paper constructs a “Water Resources Management Index—Food Security Index” coupling coordination degree model to explore the degree of coordinated development between the two. Let Ui represent the comprehensive evaluation value of the i-th dimension. Then, the coupling degree C and the coordination degree T can be calculated using the following formulas [31,32]:

$$\begin{array}{c}\mathrm{C}={\displaystyle \frac{4\sqrt[4]{{\mathrm{U}}_1\text{×}{\mathrm{U}}_2\text{×}{\mathrm{U}}_3\text{×}{\mathrm{U}}_4}}{{\mathrm{U}}_1+{\mathrm{U}}_2+{\mathrm{U}}_3+{\mathrm{U}}_4}}\end{array}$$

(14)

$$\begin{array}{c}\mathrm{T}={\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{n}}}{\mathsf{\alpha }}_{\mathrm{i}}\text{×}{\mathrm{U}}_{\mathrm{i}},{\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{n}}}{\mathsf{\alpha }}_{\mathrm{i}}=1\end{array}$$

(15)

$$\begin{array}{c}\mathrm{D}=\sqrt{\mathrm{C}\text{×}\mathrm{T}}\end{array}$$

(16)

In the above formulas, C represents the coupling degree, T represents the coordination degree, and D represents the coupling coordination degree. α is the weight. To eliminate the influence of subjective factors on the evaluation results, the entropy weight method, which extracts and analyzes objective information from the index data itself, is used to determine the index weights. Generally, the coupling degree focuses more on the correlation intensity between systems rather than development quality. The coordination degree focuses more on the development quality between systems, emphasizing harmony and efficiency [33].According to the general practice of academia, the coupling coordination degree is divided into 10 categories (See

Table 2. Types of indicators for coupling coordination degree.

Coupling Coordination Degree (D)	Class Type	Coupling Coordination Degree (D)	Class Type
0 ≤ D < 0.1	Extremely imbalanced	0.1 ≤ D < 0.2	Severely imbalanced
0.2 ≤ D < 0.3	Moderately imbalanced	0.3 ≤ D < 0.4	Mildly imbalanced
0.4 ≤ D < 0.5	On the verge of imbalance	0.5 ≤ D < 0.6	Barely coordinated
0.6 ≤ D < 0.7	Primary coordination	0.7 ≤ D < 0.8	Intermediate coordination
0.8 ≤ D < 0.9	Good coordination	0.9 ≤ D ≤ 1	Excellent coordination

).

4.4. Dagum Gini Coefficient

The Dagum Gini coefficient is used to measure the regional difference G of FS-WRM coupling coordination and can be decomposed into regional difference Gw, inter-regional difference Gnab, and super density variation Gt. The calculation formula of the Dagum Gini coefficient is as follows:

$$\mathrm{G}={\sum }_{\mathrm{j}=1}^{\mathrm{k}}{\sum }_{\mathrm{h}=1}^{\mathrm{k}}{\sum }_{\mathrm{i}=1}^{{\mathrm{n}}_{\mathrm{j}}}{\sum }_{\mathrm{r}=1}^{{\mathrm{n}}_{\mathrm{h}}}\left|{\mathrm{y}}_{\mathrm{ji}}-{\mathrm{y}}_{\mathrm{hr}}\right|/2{\mathrm{n}}^2\stackrel{-}{\mathrm{y}}$$

(17)

$${\mathrm{G}}_{\mathrm{w}}={\sum }_{\mathrm{j}=1}^{\mathrm{k}}{\mathrm{G}}_{\mathrm{jj}}{\mathrm{P}}_{\mathrm{j}}{\mathrm{S}}_{\mathrm{j}}$$

(18)

$${\mathrm{G}}_{\mathrm{nb}}={\sum }_{\mathrm{j}=2}^{\mathrm{k}}{\sum }_{\mathrm{h}=1}^{\mathrm{j}-1}{\mathrm{G}}_{\mathrm{jh}}\left({\mathrm{p}}_{\mathrm{j}}{\mathrm{s}}_{\mathrm{h}}+{\mathrm{p}}_{\mathrm{h}}{\mathrm{s}}_{\mathrm{j}}\right){\mathrm{D}}_{\mathrm{jh}}$$

(19)

$${\mathrm{G}}_{\mathrm{t}}={\sum }_{\mathrm{j}=2}^{\mathrm{k}}{\sum }_{\mathrm{h}=1}^{\mathrm{j}-1}{\mathrm{G}}_{\mathrm{jh}}\left({\mathrm{p}}_{\mathrm{j}}{\mathrm{s}}_{\mathrm{h}}+{\mathrm{p}}_{\mathrm{h}}{\mathrm{s}}_{\mathrm{j}}\right)\left(1-{\mathrm{D}}_{\mathrm{jh}}\right)$$

(20)

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

(21)

$${\mathrm{G}}_{\mathrm{jj}}={\displaystyle \frac{1}{2{\stackrel{-}{\mathrm{Y}}}_{\mathrm{j}}}}{\sum }_{\mathrm{i}=1}^{{\mathrm{n}}_{\mathrm{j}}}{\sum }_{\mathrm{r}=1}^{{\mathrm{n}}_{\mathrm{j}}}\left|{\mathrm{y}}_{\mathrm{ji}}-{\mathrm{y}}_{\mathrm{jr}}\right|/{\mathrm{n}}_{\mathrm{j}}^2$$

(22)

In the formula, G represents the overall Gini coefficient, Gw denotes the regional difference contribution, Gnb indicates the inter-regional difference contribution, and Gt stands for the super density variation contribution. yji and yhr represent the FS-WRM coupling coordination degree of province i (r) within region j (h), while $\overline{\mathrm{y}}$ denotes the average coupling coordination degree [34].

5. Research Results

5.1. Comprehensive Score Indices of Water Resource Management and Food Security

To evaluate the overall stability and reliability of the research data and validate the effectiveness of the constructed assessment framework, this study employed Cronbach’s α coefficient to test the internal reliability at each level. The results for the “food security” dimension showed an ideal Cronbach’s α coefficient of 0.8893 through factor analysis, passing the reliability test. For the “water resources” dimension, the Cronbach’s α coefficient reached 0.5267 via factor analysis, indicating acceptable reliability.

5.1.1. Comprehensive Score Index of Water Resource Management Level

During the study period, the average value of China’s water resource management level comprehensive score index gradually increased from 0.1796 to 0.2104, with an annual average growth rate of 1.33%. The growth was slow from 2010 to 2016, with an annual average increase of 0.90%, but accelerated significantly from 2016 to 2022, with an annual average increase of 1.76% (See Figure 3). The main reason was that during the policy window period or transition period from 2010 to 2016, the effects of institutional constraints and cross-regional water resource allocation projects (such as the South-to-North Water Diversion Project) were not immediately apparent. After 2016, water pollution control and evaluation measures became more stringent. Technological innovation capabilities in water conservation and water environment treatment continued to grow stronger. Resource allocation projects were continuously refined and improved. Moreover, significant enhancements were witnessed in the urban water usage penetration rate, the village water supply penetration rate, and the rate of excellent water quality.

During the study period, except for those in Guangxi, Shanghai, and Tibet, the comprehensive score indices of water resource management levels in the other 28 provincial administrative regions all showed steady improvements (See

Table 3. Comprehensive score index of water resources for each province from 2010 to 2022.

Province	2010	2012	2014	2016	2018	2020	2022	Compound Growth Rate
Anhui	0.1155	0.1219	0.1272	0.1377	0.1562	0.171	0.176	3.57%
Beijing	0.2083	0.2055	0.208	0.2075	0.2216	0.2102	0.2218	0.52%
Fujian	0.1716	0.1975	0.2094	0.2179	0.2198	0.2174	0.2128	1.81%
Gansu	0.1121	0.1137	0.1167	0.1273	0.1501	0.1618	0.1669	3.37%
Guangdong	0.2069	0.1997	0.1985	0.2018	0.2073	0.2164	0.2331	1.00%
Guangxi	0.2229	0.1841	0.193	0.1949	0.2068	0.1951	0.204	−0.74%
Guizhou	0.1121	0.1036	0.1215	0.1281	0.1536	0.1593	0.168	3.43%
Hainan	0.1961	0.2008	0.2117	0.2272	0.2288	0.2157	0.2322	1.42%
Hebei	0.1371	0.1369	0.1462	0.1525	0.1622	0.1644	0.1674	1.68%
Henan	0.092	0.0905	0.0964	0.1092	0.1299	0.1501	0.1571	4.56%
Heilongjiang	0.1389	0.1451	0.1554	0.1554	0.1823	0.1904	0.1942	2.83%
Hubei	0.1316	0.1294	0.1617	0.2024	0.2198	0.2157	0.2222	4.46%
Hunan	0.1539	0.1352	0.1418	0.151	0.1675	0.1694	0.1777	1.21%
Jilin	0.1268	0.1016	0.1139	0.131	0.1401	0.162	0.1726	2.60%
Jiangsu	0.2054	0.2015	0.2059	0.2112	0.2177	0.2215	0.2256	0.78%
Jiangxi	0.1409	0.135	0.1417	0.1483	0.1632	0.1691	0.1767	1.90%
Liaoning	0.1291	0.1206	0.1316	0.1431	0.1533	0.1572	0.1653	2.08%
Inner Mongolia	0.147	0.1235	0.1291	0.141	0.1571	0.1642	0.1657	1.00%
Ningxia	0.1522	0.1545	0.1599	0.1586	0.1888	0.2051	0.2013	2.36%
Qinghai	0.201	0.1813	0.1816	0.1846	0.2121	0.2174	0.2025	0.06%
Shandong	0.1538	0.1541	0.1606	0.1657	0.1666	0.1699	0.1731	0.99%
Shanxi	0.1369	0.1411	0.1436	0.1457	0.1575	0.1617	0.1617	1.40%
Shaanxi	0.1229	0.1249	0.131	0.1392	0.1606	0.1689	0.1712	2.80%
Shanghai	0.2537	0.2534	0.246	0.2499	0.2446	0.24	0.2407	−0.44%
Sichuan	0.0953	0.1101	0.1089	0.1302	0.1672	0.1726	0.1826	5.57%
Tianjin	0.163	0.1667	0.1653	0.1681	0.1797	0.1737	0.1843	1.03%
Tibet	0.8098	0.7167	0.7268	0.7741	0.7801	0.7458	0.6929	−1.29%
Xinjiang	0.2766	0.2839	0.2773	0.275	0.2699	0.2799	0.285	0.25%
Yunnan	0.1303	0.1236	0.1322	0.1483	0.1651	0.1771	0.1849	2.96%
Zhejiang	0.1877	0.1907	0.189	0.1992	0.1981	0.1963	0.2095	0.92%
Chongqing	0.135	0.1253	0.1384	0.1469	0.1807	0.1881	0.1931	3.03%

). Guangxi, as a water-rich region with a per capita water resource of 4380.2 cubic meters, ranking third in the country, faced issues such as far higher values than the national averages in terms of water consumption per CNY 10,000 of industrial added value, actual water consumption for cultivated land irrigation, per capita domestic water use, per capita chemical oxygen demand emission, and per capita comprehensive water use. Additionally, the village water supply penetration rate was lower than the national average, indicating an urgent need to strengthen water-saving efforts, irrigation infrastructure construction, and sewage control and treatment. Shanghai, as a megacity, faced structural water shortages but had the highest per capita domestic water use in the country, indicating an unreasonable water use structure. Tibet’s per capita water resource was 20 times the national average, but due to ecological protection policies and the scattered residence of farmers and herders, the water resource utilization rate was low.

It is particularly noteworthy that in 2010, the comprehensive score indices for water resources in Henan and Sichuan stood at a mere 0.092 and 0.0953, respectively. Despite experiencing annual average growth rates of 4.56% and 5.57% during the study period, these figures failed to reach the national average. The primary contributing factors were the severe water scarcity issues in both provinces, which are major agricultural hubs. In these regions, flood irrigation practices resulted in low effective coefficients for farmland irrigation water.

5.1.2. Comprehensive Score Index of Food Security

During the study period, China’s food security level comprehensive score index showed an overall slow upward trend, with an annual average growth rate of 0.16%, manifesting as a “U”-shaped evolution trend. The annual average decrease was 1.4% from 2010 to 2017, but the annual average increase was 2.39% from 2017 to 2022 (See Figure 4).

Shandong, Henan, and Heilongjiang provinces, despite comprising only 8.3% of China’s total land area, account for 24%, 26.7%, and 29.3% of the national total in terms of grain sown areas, agricultural machinery power, and grain outputs, respectively. Situated in the hinterlands of the North China Plain and the Northeast Plain, these provinces benefit from favorable geographical conditions, fertile soil, high levels of agricultural mechanization, robust cultivated land protection, and rapid progress in the construction of high-standard farmland. With the rapid development of agricultural scale, technology, and industrialization, as well as the mitigated impacts of extreme weather, multiple factors have collectively contributed to Henan (0.5527), Shandong (0.5001), and Heilongjiang (0.4711) consistently ranking among the top three provinces nationwide in terms of food security levels, far outpacing Hebei (0.4129), which ranks fourth, and leading the country for an extended period (See

Table 4. Comprehensive score index of food security for each province from 2010 to 2022.

Province	2010	2012	2014	2016	2018	2020	2022	CAGR
Anhui	0.3989	0.4065	0.4159	0.4203	0.3848	0.4386	0.4075	0.18%
Beijing	0.2142	0.1979	0.1885	0.1956	0.2132	0.2371	0.2545	1.45%
Fujian	0.2460	0.2301	0.2242	0.2277	0.2141	0.2496	0.2593	0.44%
Gansu	0.2879	0.2999	0.3192	0.3324	0.2160	0.3132	0.2427	−1.41%
Guangdong	0.3086	0.3020	0.2997	0.3023	0.2879	0.3146	0.3184	0.26%
Guangxi	0.2915	0.3019	0.2854	0.3048	0.2769	0.3045	0.3218	0.83%
Guizhou	0.2554	0.2444	0.2409	0.2398	0.2276	0.2577	0.2689	0.43%
Hainan	0.2012	0.1823	0.1749	0.2140	0.1857	0.2143	0.2297	1.11%
Hebei	0.4673	0.4795	0.4928	0.4616	0.3926	0.4461	0.4129	−1.03%
Henan	0.5971	0.6178	0.6384	0.6347	0.5401	0.6377	0.5527	−0.64%
Heilongjiang	0.4059	0.4315	0.4462	0.4489	0.4540	0.4888	0.4711	1.25%
Hubei	0.3435	0.3361	0.3446	0.3401	0.3217	0.3453	0.3600	0.39%
Hunan	0.3702	0.3730	0.3768	0.3793	0.3454	0.3867	0.3845	0.32%
Jilin	0.2935	0.2985	0.3052	0.3132	0.3011	0.3346	0.3382	1.19%
Jiangsu	0.3936	0.3988	0.4174	0.4157	0.3545	0.4186	0.3893	−0.09%
Jiangxi	0.2971	0.2942	0.2798	0.2709	0.2600	0.2886	0.2980	0.03%
Liaoning	0.3137	0.3225	0.3159	0.3112	0.2475	0.3174	0.2819	−0.89%
Inner Mongolia	0.2966	0.2975	0.3149	0.3211	0.3072	0.3619	0.3568	1.55%
Ningxia	0.1984	0.1786	0.1692	0.1735	0.1847	0.2098	0.2095	0.46%
Qinghai	0.1922	0.1716	0.1615	0.1661	0.1784	0.2012	0.2113	0.79%
Shandong	0.6322	0.6475	0.6565	0.6384	0.4769	0.6483	0.5001	−1.94%
Shanxi	0.2637	0.2532	0.2469	0.2341	0.2231	0.2519	0.2581	−0.18%
Shaanxi	0.2635	0.2568	0.2497	0.2855	0.2452	0.2707	0.2775	0.43%
Shanghai	0.2134	0.1997	0.2395	0.2619	0.2339	0.2621	0.2699	1.98%
Sichuan	0.4191	0.4257	0.4296	0.4337	0.3665	0.4606	0.3997	−0.39%
Tianjin	0.2106	0.1939	0.1855	0.2363	0.2401	0.1684	0.2391	1.06%
Tibet	0.1925	0.1717	0.1610	0.2034	0.2123	0.2083	0.2170	1.01%
Xinjiang	0.3210	0.3330	0.3998	0.4098	0.2466	0.4173	0.3000	−0.56%
Yunnan	0.3174	0.3231	0.3319	0.3391	0.2775	0.3439	0.3118	−0.15%
Zhejiang	0.2636	0.2542	0.2546	0.2600	0.2499	0.2945	0.3009	1.11%
Chongqing	0.2344	0.2184	0.2113	0.2155	0.2111	0.2880	0.2496	0.53%

).

According to 2022 data, there were 15 provincial administrative regions with comprehensive score indices for food security levels below 0.3 (See Table 4). Among these regions, Beijing (0.2545), Shanghai (0.2699), and Tianjin (0.2391), leveraging their unique political and economic statuses, vigorously promoted the development of secondary and tertiary industries. They ranked among the top three nationwide in urbanization rates, at 89.3%, 87.6%, and 85.11% respectively, far exceeding the national average of 65.2%. Consequently, they faced a shortage of agricultural labor (with the rural population in Beijing accounting for only 12% of the total resident population). As an old industrial base in Northeast China, Liaoning’s (0.2819) grain sown area accounts for only 2.5% of the national total. With a 73% urbanization rate, high-quality cultivated land is being occupied by construction land. High-intensity farming has accelerated black soil degradation. Fujian (0.2593), Hainan (0.2297), and Chongqing (0.2496) are constrained by topographical conditions and shifts in development models, leading to an expansion in the cultivation areas of high-value economic crops. Northwest regions such as Shaanxi (0.2775), Gansu (0.2427), and Ningxia (0.2095) are constrained by fragile ecological environments and backward economic conditions, facing water shortages, cultivated land quality degradation, backward irrigation technologies, and spatial competition between ecological protection and restoration policies (such as returning farmland to forests and grasslands) and food production. Similarly, Shanxi (0.2581) is also significantly affected by the “reducing grain and increasing forage” policy. Guizhou (0.2689), Tibet (0.2170), and Qinghai (0.2427) are mainly constrained by natural conditions, with fragmented topography, poor soil quality, and mountainous and hilly landscapes in Guizhou, located on the Yunnan-Guizhou Plateau, making large-scale agricultural machinery operations difficult and prone to soil erosion. The perennial frozen soil on the Qinghai-Tibet Plateau in Tibet and Qinghai limits the construction of high-standard farmland, resulting in single grain crop varieties and low yields.

During the study period, two-thirds of the provincial administrative regions achieved increases in their food security level comprehensive score indices, with Shanghai, Beijing, Hainan, and six other provincial administrative regions experiencing increases exceeding 1%. However, except for Inner Mongolia, Jilin, and Heilongjiang, all of which possess black soil and have food security level comprehensive score indices greater than 0.3, the other six provincial administrative regions are close to or below 0.3, indicating that the foundation of food security needs to be consolidated and that there is still significant room for improvement in food security level comprehensive score indices. Ten provincial administrative regions experienced decreases in their food security level comprehensive score indices, with Shandong (−1.94%), Gansu (−1.41%), and Hebei (−1.03%) experiencing decreases exceeding 1%. Among them, Shandong’s growth started from a high point and was sluggish (See Table 4).

It is noteworthy that in 2017, the composite score index for food security levels in 25 provincial administrative regions reached an inflection point (See Figure 4), with 13 of the regions experiencing a decline of more than 10% compared to 2016. This was primarily influenced by the structural reforms on the agricultural supply side (including adjustments to the planting structure in the “Sickle-shaped Bend” region and a systematic reduction in the use of pesticides and fertilizers), the expansion of disaster-affected areas in key grain-producing regions such as Shandong and Henan (particularly severe drought conditions), and adjustments to the standards in the Third National Agricultural Census.

5.2. Coupling and Coordination Degree Between Water Resource Management and Food Security

For the study period, the mean values of the score index for water resource management, the score index for food security, and the coupling coordination degree D are illustrated in Figure 5. It is evident that changes in either the water resource management level or the food security level will cause fluctuations in the coupling coordination degree. There is a clear coupling and coordination relationship between the two. Water resources serve as a rigid constraint for grain production, with water resource management playing a supportive and safeguarding role for food security. Conversely, food security acts as a weathercock and touchstone for the level of water resource management. To achieve coupling coordination and sustainable development between the two, it is imperative to simultaneously promote high-quality development in both areas.

5.2.1. Spatial Distribution

Observations show (See

Table 5. Dagum Gini coefficient of coupling and coordination between food security and water resources management from 2010 to 2022.

Year		2010	2011	2012	2013	2014	2015	2016	2017	2018	2019	2020	2021	2022
Overall difference		0.104	0.107	0.119	0.121	0.132	0.120	0.113	0.104	0.096	0.085	0.094	0.074	0.074
Inter-regional difference	East–central	0.086	0.096	0.101	0.111	0.119	0.098	0.085	0.086	0.077	0.076	0.074	0.076	0.072
	East–west	0.091	0.090	0.121	0.099	0.108	0.106	0.083	0.092	0.093	0.081	0.083	0.072	0.063
	East–northeast	0.125	0.130	0.142	0.149	0.164	0.144	0.133	0.110	0.102	0.092	0.094	0.085	0.081
	Central–west	0.057	0.069	0.092	0.083	0.079	0.075	0.070	0.093	0.090	0.080	0.086	0.064	0.068
	Central–northeast	0.105	0.113	0.106	0.129	0.136	0.117	0.120	0.115	0.104	0.090	0.095	0.081	0.082
	West–northeast	0.102	0.103	0.125	0.114	0.125	0.122	0.119	0.117	0.111	0.090	0.104	0.076	0.077
Intra-regional difference	Eastern region	0.095	0.104	0.114	0.118	0.129	0.118	0.092	0.070	0.066	0.076	0.062	0.072	0.059
	Central region	0.048	0.074	0.053	0.091	0.086	0.061	0.066	0.090	0.078	0.066	0.066	0.061	0.068
	Western region	0.057	0.045	0.104	0.052	0.052	0.079	0.059	0.086	0.088	0.067	0.090	0.061	0.060
	Northeast region	0.119	0.122	0.130	0.136	0.149	0.143	0.151	0.117	0.110	0.091	0.112	0.078	0.082
Sources and contribution rates of regional disparities	Contribution rate of intra-regional factors	29.025	28.998	28.729	28.661	28.623	29.848	30.436	28.544	28.616	28.618	29.917	28.125	28.605
	Contribution rate of inter-regional factors	28.153	30.136	31.029	33.651	31.726	27.522	25.156	27.833	22.173	24.609	11.484	23.607	22.962
	Contribution rate of hypervariable density	42.822	40.866	40.243	37.688	39.651	42.630	44.409	43.623	49.211	46.774	58.599	48.268	48.433

) that the overall difference in China’s FS-WRM coupling coordination Gini coefficient decreased from 0.104 in 2010 to 0.074 in 2022, representing a decline of 28.85%. This fluctuating downward trend indicates narrowing disparities in FS-WRM coupling coordination levels both within and between regions, demonstrating a trend towards regional coordinated development. However, significant regional gaps remain. During the study period, inter-regional differences in the FS-WRM coupling coordination Gini coefficient showed a descending trend of “east–northeast > east–west > west–northeast > central–northeast > east–central > central–west”, while intra-regional differences followed a descending pattern of “east > northeast > west > central”.

Distribution Characteristics and Differences Among the Four Major Regions

During the study period, the coordinated development of water resource management and food security in China’s four major regions exhibited significant spatial heterogeneity. This characteristic reflects the nonlinear coupling relationship among policy support, technological innovation, and natural endowments. Overall, the coupling coordination degrees in all four regions increased year by year, yet they all remained at the brink of imbalance, showing a gradient pattern of Central China > Northeast China > Eastern China > Western China (See Figure 6). The average annual growth rates differed markedly, with Northeast, Eastern, Western, and Central China recording rates of 1.63%, 0.81%, 1.39%, and 1.91% (See

Table 6. Changing characteristics of regional coupling coordination.

Province	2010	2014	2018	2022	CAGR
Liaoning	0.3732	0.3796	0.3623	0.408	0.75%
Jilin	0.3562	0.3329	0.3857	0.4571	2.10%
Heilongjiang	0.4395	0.4869	0.5304	0.5533	1.94%
Regional average	0.3896	0.3998	0.4261	0.473	1.63%
Beijing	0.3739	0.3231	0.3817	0.4375	1.32%
Tianjin	0.3288	0.284	0.3841	0.3874	1.38%
Hebei	0.4604	0.4884	0.473	0.4907	0.53%
Shanghai	0.4026	0.4371	0.4285	0.4683	1.27%
Jiangsu	0.5282	0.5414	0.5172	0.5463	0.28%
Zhejiang	0.4168	0.409	0.4124	0.4706	1.02%
Fujian	0.3822	0.3898	0.3821	0.4352	1.09%
Shandong	0.548	0.5677	0.5175	0.5367	−0.17%
Guangdong	0.4742	0.4591	0.4577	0.5054	0.53%
Hainan	0.3421	0.2845	0.3284	0.4143	1.61%
Regional average	0.4257	0.4184	0.4283	0.4692	0.81%
Inner Mongolia	0.3938	0.374	0.416	0.4591	1.29%
Guangxi	0.4747	0.4418	0.4475	0.4813	0.12%
Chongqing	0.3231	0.3009	0.3466	0.4075	1.95%
Sichuan	0.304	0.3705	0.4666	0.5048	4.32%
Guizhou	0.2962	0.3057	0.3411	0.4003	2.54%
Yunnan	0.378	0.3899	0.4042	0.454	1.54%
Tibet	0.5165	0.306	0.5697	0.5621	0.71%
Shaanxi	0.3271	0.3311	0.3689	0.4115	1.93%
Gansu	0.318	0.3481	0.3222	0.3736	1.35%
Qinghai	0.3272	0.1957	0.297	0.3648	0.91%
Ningxia	0.2977	0.2287	0.3005	0.3609	1.62%
Xinjiang	0.5408	0.5968	0.4615	0.5284	−0.19%
Regional average	0.3748	0.3491	0.3952	0.4424	1.39%
Shanxi	0.353	0.3484	0.3401	0.3829	0.68%
Anhui	0.3808	0.4185	0.46	0.5001	2.30%
Jiangxi	0.3843	0.3733	0.3865	0.4344	1.03%
Henan	0.3193	0.3619	0.4683	0.5294	4.30%
Hubei	0.395	0.4464	0.4963	0.5251	2.40%
Hunan	0.4488	0.4316	0.4549	0.4905	0.74%
Regional average	0.3802	0.3967	0.4344	0.4771	1.91%

), respectively. This presents a spatial pattern of “steady growth and leading in Central China, steady progress in Northeast China, fluctuations in high-value areas in Eastern China, and catching up from behind in Western China.”

Among the main regions, the Central region has traditionally been a major grain-producing area. Benefiting from natural resources such as the Yangtze and Huaihe Rivers, as well as policy dividends from national strategies like the “Food Security Industrial Belt” and the “Yangtze River Economic Belt,” the coupling coordination degree in the Central region has risen rapidly, securing a leading position since 2017. It is on the verge of transitioning from a state near imbalance to a barely coordinated level (with an average of 0.4771 in 2022). During the study period, the coupling coordination degree of Food Security—Water Resources Management (FS-WRM) in the Northeast region steadily increased from 0.3896 to 0.473. The “stability” in Northeast China is mainly attributed to its solid agricultural foundation, favorable natural conditions, and policy support. Meanwhile, the “progress” stems from technological and institutional innovations, such as the national action plan “Outline for the Protection of Black Soil in Northeast China (2017–2030)” and other guidelines, which have accelerated the restoration and governance of black soil, infrastructure construction, and the implementation of the strictest water resource management system.

From 2010 to 2016, the coupling coordination degree of FS-WRM in Eastern China outpaced that of the other three major regions. In fact, in 2014, there was a significant difference of 0.069 compared to the Western region. However, this pattern changed in 2017. Eastern China, at the forefront of reform, opening up, and high-quality development, has experienced significant fluctuations in its coupling coordination degree due to factors such as the acceleration of urbanization, which has led to the encroachment of construction land on farmland; intense competition for water resources among industry, domestic use, and agriculture; increased instability in the climate system; and uneven spatial distribution of arable land. Western China ranks first among the four regions in terms of both cultivated land area and total water resources. Nevertheless, its complex and fragile ecological foundation, relatively backward economic development and technological strength, low irrigation water utilization coefficient, and low grain output pose challenges in achieving coordinated development in terms of “ecological protection, food security, and water resource security” as a whole.

Distribution Characteristics and Differences Within the Four Major Regions

At the provincial level, during the study period, except for Shandong and Xinjiang, the coupling coordination degrees of the other 29 provinces and municipalities showed varying degrees of improvement, with significant disparities among them. In 2022, five provinces and municipalities, including Tianjin and Gansu, remained in the mildly imbalanced category, while sixteen, including Hebei and Liaoning, were on the brink of imbalance. Ten provinces and municipalities, including Heilongjiang, achieved a barely coordinated level (See Figure 7).

In Northeast China, Heilongjiang (0.5533) had a significantly higher coordination degree than Liaoning (0.408) and Jilin (0.4571), reaching the barely coordinated level and forming a gradient pattern that decreased from north to south. This pattern is positively correlated with the area of black soil, reflecting the positive effects of black soil protection policies on the resource system. Eastern China exhibited a “core–periphery” differentiation. Major agricultural provinces like Shandong and Jiangsu continued to lead, maintaining a stable barely coordinated level for a long time. In contrast, Beijing, Tianjin, and Fujian only crossed from mild imbalance to the brink of imbalance in 2019, highlighting the squeezing effect of economic development on the allocation of water and land resources. The internal disparities within Western China were the most pronounced. Frontier provinces such as Xinjiang (0.5284) and Tibet (0.5621) have formed high-value areas owing to their ecological advantages and policy preferential treatment, creating a disparity of over 0.2 with provinces like Ningxia (0.3609) and Qinghai (0.3648), which are hindered by ecological and economic constraints. As a vital grain production base in China, Central China boasts well-developed water systems and abundant resources. Covering 10.69% of the country’s land area, it contributes over 30% of the national grain output. In particular, core provinces like Henan, Anhui, and Hubei, leveraging their geographical advantages, strategic policies, and large-scale grain production capabilities, have jointly shaped a “golden triangle” driving the rise of Central China. This region has established a high-value contiguous area centered around the Huang-Huai Plain and the middle and lower reaches of the Yangtze River Plain. In contrast, Shanxi (0.3829), which has long been reliant on the energy industry and constrained by its traditional resource-based economy, has emerged as a regional lowland (See Table 6).

5.2.2. Temporal Characteristics

From 2010 to 2022, the overall coupling coordination degree between food security (FS) and water resource management (WRM) in China remained relatively low, with an annual average of just 0.4181, indicating a stage of impending imbalance. This suggests that China has not yet achieved coordinated and high-quality development in FS and WRM, and the mechanism for positive interaction between the two needs further improvement. From a temporal perspective, the average coupling coordination degrees for FS-WRM in 2010, 2014, 2018, and 2022 were 0.3926, 0.3910, 0.4210, and 0.4654, respectively. The coordination level steadily improved from mild imbalance to impending imbalance, gradually approaching a barely coordinated state (with 10 provinces reaching this level in 2022). The impending imbalance level was achieved in 2015 (with a national average of 0.4029), demonstrating a close connection between FS and WRM, characterized by mutual promotion and joint development. From 2010 to 2017, the average coupling coordination degree for FS-WRM increased from 0.3926 to 0.4114, with an annual growth rate of 0.67%. From 2017 to 2022, it rose from 0.4114 to 0.4654, with an annual growth rate of 2.5% (See Figure 8). Through long-term exploration and continuous breakthroughs, China has identified a more systematic, effective, and technologically advanced coordination path to address the core contradiction between ensuring food security and coping with water scarcity.

At the policy and institutional level, in 2012, the State Council issued “Opinions on Implementing the Strictest Water Resource Management System,” establishing the “Three Red Lines” (total water use, water use efficiency, and pollutant discharge limits in water function zones). After 2017, the implementation and assessment standards for this policy became stricter. In 2015, the structural reform of the agricultural supply side was first proposed, with a core focus on optimizing the agricultural structure and layout. Under rigid resource constraints (especially water resources), efforts were made to reduce corn planting areas, adjust the production layout of animal husbandry and fisheries, and consolidate and enhance grain production capacity. In 2016, the “River Chief and Lake Chief System” was fully implemented nationwide, clarifying the responsible entities for water resource management. In 2019, the “National Water Conservation Action Plan” was issued, emphasizing the closer integration of water resource management and food security in the national top-level strategy through water-saving irrigation, optimization of crop planting structures, promotion of water-saving methods in animal husbandry and fisheries, and rural domestic water conservation. Meanwhile, economic means such as comprehensive reform of agricultural water pricing and agricultural water rights trading have also promoted the standardization and scientific management of water resources.

At the technological innovation level, after 2017, the application cost of efficient water-saving irrigation technologies decreased, and subsidies were strengthened, entering a stage of comprehensive popularization. New artificial intelligence technologies such as smart irrigation began to penetrate the agricultural sector, significantly improving water use efficiency. At the infrastructure construction level, the benefits of major water conservancy projects, high-standard farmland construction, large-scale land management, and supporting agricultural infrastructure policies have gradually emerged.

5.2.3. Influencing Factors

This study constructed an analytical framework comprising 15 explanatory variables and 1 explained variable based on Chinese provincial panel data from 2010 to 2022 (See

Table 7. Descriptive statistics.

Variable Identifier	Variables	Number	Minimum	Maximum	Mean	Standard Deviation
x1	Expenditure on agriculture, forestry, and water resources (CNY 10,000)	403	6.714 × 105	1.359 × 107	5.411 × 106	2.880 × 106
x2	Average years of education for rural residents (year)	403	3.819	12.6	7.849	1.004
x3	Technology market development level	396	2.707 × 10−5	0.191	0.017	0.03
x4	Agricultural disaster-affected area (thousands of hectares)	403	2	4224	747.18	760.666
x5	Per capita disposable income of rural residents (yuan)	403	3747	39,729	13,282.39	6394.789
x6	Number of legal entities in primary industry	403	26	159,565	37,953.16	33,572.809
x7	Increase in primary industry (100 million year)	403	69	6299	2050.99	1469.468
x8	Effective irrigation area (1000 hectares)	403	109	6535	2131.49	1680.922
x9	Disaster resistance	403	0.1	1	0.554	0.164
x10	Research and development intensity	403	0.002	0.068	0.017	0.0116
x11	Proportion of added value of the primary industry in the gross regional product (%)	403	0.2	26.3	9.755	5.1360
x12	Urban area and cultivated land requisitioned for construction (sq.km.)	403	0	165.950	27.415	27.423
y	FS-WRM coupling coordination degree	403	0.196	0.609	0.415	0.082

). The mean value of the explained variable, the coupling coordination degree of Food Security—Water Resource Management (FS-WRM) (y), is 0.415 (with a standard deviation of 0.082), indicating that the overall coordination level between food security and water resource management in China is still in a state of “impending imbalance” (within the range of 0.4–0.5). Among the key variables, the R&D intensity (x10), rural residents’ per capita disposable income (x5), effective irrigation area (x8), and agricultural disaster-affected area (x4) showed a significant positive correlation with the coupling coordination degree between China and FS-WRM (See

Table 8. Benchmark regression analysis.

Variable Identifier	Y
x1	0.022
	(1.425)
x2	0.061
	(1.097)
x3	0.078
	(0.287)
x4	0.006 **
	(2.315)
x5	0.288 ***
	(2.940)
x6	−0.005
	(−0.559)
x7	−0.018
	(−1.077)
x8	0.063 **
	(2.402)
x9	0.027
	(1.100)
x10	3.389 ***
	(3.249)
x11	0.035 *
	(1.651)
x12	0.005
	(1.395)
Constant	−3.232 ***
	(−3.358)
Time fixed effects	Control
Provincial fixed effects	Control
N	403
adj. R2	0.852

Notes: t statistics in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01.

).

Due to the correlation between error terms, the standard errors in the benchmark regression model may be biased, leading to distorted regression results. To address this issue, this study conducted cluster adjustment on the standard errors at the provincial-year level. Additionally, considering the unique administrative status and resource allocation advantages of municipalities directly under the central government, these regions were excluded from the robustness test due to their special administrative characteristics. Finally, to prevent extreme values from interfering with model coefficient regression results, a 1% truncation method was employed for robustness testing. The robustness test results in

Table 9. Robustness test.

	Change the Clustering Hierarchy	Excluding Municipalities	Tail Trimming
	(1)	(2)	(3)
x1	0.022	0.020	0.025
	(1.425)	(1.129)	(1.607)
x2	0.061	0.050	0.038
	(1.097)	(0.699)	(0.818)
x3	0.078	0.074	−0.004
	(0.287)	(0.275)	(−0.016)
x4	0.006 **	0.008 ***	0.005 **
	(2.315)	(3.082)	(2.288)
x5	0.288 ***	0.240 **	0.193 ***
	(2.940)	(2.008)	(2.734)
x6	−0.005	−0.003	0.005
	(−0.559)	(−0.312)	(0.573)
x7	−0.018	0.023	−0.010
	(−1.077)	(1.110)	(−0.589)
x8	0.063 **	0.125 ***	0.041 *
	(2.402)	(3.970)	(1.761)
x9	0.027	−0.011	0.027
	(1.100)	(−0.393)	(1.079)
x10	3.389 ***	5.028 ***	2.951 ***
	(3.249)	(4.304)	(2.965)
x11	0.035 *	0.037 *	0.021
	(1.651)	(1.684)	(1.053)
x12	0.005	0.005	0.005
	(1.395)	(1.460)	(1.426)
Constant	−3.232 ***	−3.536 ***	−2.275 ***
	(−3.358)	(−2.998)	(−3.308)
Time fixed effects	control	control	control
Provincial fixed effects	control	control	control
N	403	351	403
adj. R2	0.846	0.857	0.847

Notes: t statistics in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01.

further validate the conclusions of the benchmark regression, demonstrating strong stability of the findings.

Based on the solid-state effect model analysis results, the influencing factors of the coupling coordination degree between food security (FS) and water resource management (WRM) (Y) exhibit significant differences (See Table 8). It can be observed that FS-WRM is primarily influenced by the following four aspects:

Technological innovation capability: The coefficient of R&D intensity (x10) was the largest and passed the 1% significance test (t = 3.249), indicating that scientific and technological innovation investment played a core role in improving the coupling coordination degree of FS-WRM. Faced with the superposition of new and old problems, the supporting role of scientific and technological innovation in the coordinated development of FS-WRM has become increasingly prominent. During the “13th Five-Year Plan” period (2016–2020), the contribution rate of agricultural science and technology progress in China reached 60%. Compared to 2015, the coupling coordination degree of FS-WRM in China increased by 14.66% in 2020. Economically developed eastern cities such as Jiangsu and Zhejiang, where cultivated land area is limited, industrialization and urbanization are rapidly developing, and water resources are severely scarce, have continuously enhanced grain production efficiency by selecting and promoting new varieties; advancing the mechanization, intelligence, and digitization of crop cultivation, harvesting, and planting; and applying new water-saving irrigation technologies. This has effectively resolved the contradiction between urbanization, industrialization development, and grain production. In 2022, Shanghai ranked first in the country in terms of grain yield per unit area. Another example is the Sanjiang Plain, an important rice production base in China, where the per capita cultivated land area is five times the national average. Since the 1990s, the paddy field area in the Sanjiang Plain has increased sevenfold, and the problem of groundwater over-extraction has become increasingly severe. In response, Heilongjiang Province has widely promoted water-saving controlled irrigation technology for rice. Coupled with its good agricultural foundation, the coupling coordination degree for FS-WRM in Heilongjiang Province increased by 13.4% from 2016 to 2022.

Agricultural infrastructure construction: The effective irrigated area (x8) has a significant positive impact on the coupling coordination degree of FS-WRM (t = 2.402), and agricultural infrastructure directly improves the water resource allocation capacity and disaster resistance capacity. Given China’s vast territory and significant differences in natural endowments, water scarcity is largely due to uneven temporal and spatial distribution. Water conservancy projects, covering functions such as irrigation area construction, inter-basin water transfer, flood control and drought relief, and water-saving and efficiency enhancement efforts have effectively addressed this issue, achieving precise allocation of water resources and promoting steady growth in grain production. For example, the Hetao Irrigation Area has improved irrigation efficiency, the South-to-North Water Diversion Project has alleviated groundwater over-extraction in the North China Plain, and the Three Gorges Project has relieved irrigation pressure on farmland in the middle and lower reaches of the Yangtze River. Meanwhile, China has carried out large-scale construction of high-standard farmland, transforming permanent basic farmland into modern high-standard farmland through optimized infrastructure and systematic engineering. According to estimates, high-standard farmland construction can increase grain production capacity by 10% and save water by 20–30%. Technology has empowered the improvement of comprehensive agricultural production capacity, ensuring stable yields despite droughts and floods.

Modernized management mode: The coefficient of rural residents‘ per capita disposable income (x5) is statistically significant (t = 2.940), indicating that income growth can enhance agricultural technology adoption and optimize resource management. Meanwhile, China has achieved victory in the poverty alleviation campaigns and vigorously implemented a rural revitalization strategy. The urbanization rate has accelerated significantly (from 49.95% in 2010 to 65.22% in 2022), with rural residents’ average years of education rising markedly (from 7.5 years in 2010 to 9.8 years in 2022). The skill level and overall quality of rural practitioners are improving day by day. This demonstrates the increasingly diversified and modernized development model of rural economies. Through water-saving irrigation technology, rural tourism, expanding sales channels of agricultural products by relying on online platforms, and establishing geographical indications of agricultural products, the transformation of agricultural production modes and management modes has been promoted, and the sustainable development of agriculture and the coordinated development of FS-WRM coupling have been promoted.

Policy support: It is worth noting that the coupling coordination degree between crop disaster area (x4) and FS-WRM showed a significant positive correlation, reflecting that the expansion of the disaster area led to the strengthening of policy intervention (insurance compensation and disaster relief input), which indirectly improved the coordination degree. Policy support has facilitated the transformation of China’s food security from “ensuring output” to “ensuring production capacity + quality” and water resource management from “development-oriented” to “protection-oriented,” gradually transitioning to “smart water use” at the top-level framework level, thereby providing direction for the coordinated development of FS-WRM. On the one hand, the guidance and incentive effects of full-chain policies, such as the marketization of water use rights, comprehensive reform of agricultural water pricing, and minimum purchase price policy for rice and wheat, offer direction for the coordinated development of the two. For instance, Xinjiang has widely implemented a precise subsidy and water-saving reward model for the comprehensive reform of agricultural water pricing, promoting the popularization of drip irrigation technology through economic levers and significantly improving water use efficiency. According to estimates, from 2015 to 2022, the effective utilization coefficient of farmland irrigation water in Xinjiang increased by 12%, and the proportion of agricultural water use in total societal water use decreased from 95% to 91%.

6. Research Conclusions

(1)

During the study period, China’s food security level showed a “U-shaped” evolution. Before 2017, it decreased by an average of 1.4% annually (affected by policy adjustments and disasters), and after 2017, it increased by an average of 2.39% annually. Shandong, Henan, and Heilongjiang consistently ranked among the top three due to their fertile plains and mechanization. However, 15 provinces scored below 0.3 due to ecological fragility in the northwest, topographical limitations in the southwest, and black soil degradation in the northeast. Two-thirds of the provinces saw an increase in scores, but regions with high starting points like Shandong experienced sluggish growth. In 2017, 25 provinces reached an inflection point due to dual pressures from policies and disasters.

(2)

From 2010 to 2022, China’s water resource management level steadily improved, with the comprehensive score index increasing by an average of 1.33% annually. The growth rate significantly accelerated in the later period (1.76%). Regional differences were evident. Delayed policy implementation and technological innovation were the main reasons for the slow growth in the early stages, while strict water pollution control, the promotion of water-saving technologies, and inter-regional water transfer projects (such as the South-to-North Water Diversion Project) yielded remarkable results in the later period. At the provincial level, except for Guangxi, Shanghai, and Tibet, the water resource management level in 28 provinces showed significant improvement. Guangxi and Shanghai faced challenges such as insufficient water conservation and structural issues, respectively, while Tibet experienced low utilization rates due to ecological protection constraints. Major agricultural provinces like Henan and Sichuan had low comprehensive score indices at the beginning of the study period due to low irrigation efficiency.

(3)

During the study period, the coordinated development of water resource management and food security in China’s four major regions exhibited significant spatial heterogeneity. Overall, it increased year by year but remained at the impending imbalance level, presenting a spatial pattern of “steady growth and leading in Central China, steady progress in Northeast China, fluctuations in high-value areas in Eastern China, and catching up from behind in Western China.” From a regional internal perspective, Northeast China formed a gradient pattern that decreased from north to south; Eastern China showed a “core–periphery” differentiation, with agricultural provinces like Shandong and Jiangsu continuing to lead; Western China was internally differentiated due to ecological, policy, and economic factors; and Central China’s Henan, Anhui, and Hubei jointly formed a high-value contiguous area centered around the Huang-Huai Plain and the middle and lower reaches of the Yangtze River Plain. From a temporal characteristic perspective, from 2010 to 2022, China’s FS-WRM coordination level steadily improved from mild imbalance to impending imbalance, gradually approaching a barely coordinated level (with 10 provinces reaching this level in 2022). FS and WRM were closely connected, with mutual promotion and joint development. Among them, the coupling coordination degree for FS-WRM grew faster from 2017 to 2022, with an annual growth rate of 2.5%.

(4)

The coupling coordination of food security–water resource management (FS-WRM) is mainly influenced by four factors: policy support provides financial and institutional guarantees for the coupling coordination of the two; infrastructure construction creates foundational conditions for their coupling coordination; technological innovation capability injects core driving forces into their coupling coordination; and a modernized management mode establishes an operational framework for their coupling coordination.

7. Prospects

Research on the coupling coordination mechanism and key driving factors of water resource management and food security in China needs to be based on the realistic background of intensifying global climate change and resource constraints, deepening multi-scale collaborative governance paths with systemic thinking. In the future, differentiated strategies need to be implemented based on regional resource endowment differences. For example, in the arid regions of Northwest China, efforts should be made to strengthen the linkage between water-saving and efficiency enhancement efforts and ecological restoration; in the major grain-producing areas of Northeast China, water–grain coordination and black soil protection should be promoted; and in the water-rich regions of South China, water rights trading and industrial upgrading should be optimized. Meanwhile, cross-departmental coordination mechanisms should be constructed, and resource conflicts between departments should be quantified through the water–energy–food (WEF) coupling model. This will accelerate the connection between the innovative river and lake chief system [35] and the fallow land system [36], promote cross-basin ecological compensation mechanisms, improve relevant infrastructure, and solve the problem of cross-regional allocation of water resources. At the technological level, the popularization of digital twin and smart irrigation technologies needs to be accelerated, and unconventional water resource utilization systems should be developed. Research should focus on the interactions among natural, social, and institutional multi-dimensional driving factors, quantify the risk transmission paths of climate change on the water-grain system, assess the compound impacts of land use change and urbanization processes, and achieve incentive compatibility through innovative combinations of policy tools and legal guarantees. Facing rigid challenges such as water scarcity, cultivated land quality degradation, and water pollution, it is necessary to break through technological bottlenecks in smart irrigation equipment and water–fertilizer integration through scientific and technological innovation, establish a public participation mechanism for a “water-saving society,” and deepen international cooperation to introduce advanced global experiences. Only by constructing a modern governance system for the coordinated development of “water–grain–ecology–economy” can the dual goals of ensuring food security for 1.4 billion people and enabling sustainable utilization of water resources be achieved, providing a sustainable “Chinese solution” for global resource security governance.