Water Footprint and Water Resources Sustainability: A Review

3.1. Composition of the Water Footprint

The water footprint is a multidimensional indicator of water use that encompasses blue, green, and grey water [8,20]. Blue water refers to liquid water in rivers, lakes, wetlands, and aquifers and is the type most commonly examined in traditional water resource studies. Green water includes two components: green water flow, which refers to the water that evaporates from soil, water bodies, and vegetation; and green water storage, which relates to soil moisture maintained in the unsaturated zone, formed by precipitation and available for plant use. Grey water is a measure of the extent of water contamination and is defined as the amount of freshwater needed to absorb pollutants to meet existing environmental water quality standards [12]. In terms of water use boundaries, the water footprint includes an internal water footprint and an external water footprint [8], where the internal water footprint of a region is the amount of water used, excluding imported products and services, and the external water footprint is the amount of water included in imported products and services in the form of virtual water. Water footprints are characterized by different temporal and spatial scales, including global, national, provincial, municipal, and watershed scales, as well as temporal scales such as years, months, or days.

3.2. Accounting for the Water Footprint

The extent to which humans have reached or even exceeded the planetary boundaries of freshwater resources is a central concern for global sustainable development [21]. As illustrated in Figure 1, balancing food, energy and water are all important resources that are essential for human survival. These three are interdependent and interact with each other, forming the Food-Energy-Water (FEW) system [22]. In the context of climate change and societal transformation, maintaining the health of the ecosystem and providing sufficient food, energy, and water resources for the increasing population and managing the resulting water footprint are of great importance to ensure people’s water, food, and energy security.

3.2.1. Method for Calculating the Water Footprint of Food Production in the Agricultural Production Process

As defined, the agricultural water footprint comprises green, blue, and grey components. According to Allen et al. [23], the agricultural water footprint is usually calculated as follows:

(1) Green water footprint: if evapotranspiration (ET) exceeds adequate precipitation (P) during the growing season, the green water footprint equals the effective P; if ET is less than effective P, the green water footprint equals the actual ET.

(2) Blue water footprint: This depends on the crop’s tolerance to water scarcity, irrigation efficiency, and the availability of green water. If evapotranspiration (ET) exceeds precipitation (P), the blue water footprint equals ET minus P; otherwise, the blue water footprint is zero. Complete information on irrigation, soil, and cultivation is required to calculate the blue water footprint.

(3) The grey water footprint: This quantifies the freshwater needed to dilute pollutants from agricultural activities to meet environmental quality standards. Nitrogen fertilizers are considered as one of the major contributors to grey WF [24]. Phosphorus, also from fertilizers, has a significant impact on water bodies as it causes eutrophication and requires considerable dilution [25]. Two examples of emerging pollutants that are increasingly significant yet difficult to quantify due to their complex interactions and limited data are pharmaceuticals originating from manure or wastewater irrigation, and microplastics derived from mulch films [16]. Sophisticated modelling tools (e.g., SPARE: WATER) are required to capture the persistence of these pollutants in the environment [26]. Together, these components constitute the overall water footprint. While traditionally based on nitrate leaching, the grey water footprint increasingly accounts for multiple pollutants. It is calculated using the following formula:

$$GWF={\displaystyle \frac{L}{C_{max}-C_{nat}}}$$

where C_max is the permissible concentration in the receiving water, C_nat is the natural background concentration, and L is the pollutant load (e.g., nitrogen or phosphorus). The robustness of the assessment is significantly enhanced by the inclusion of multiple pollutant loads and region-specific water quality criteria.

According to various researchers, there are different models for simulating the water footprint of crop production. These models include the CROPWAT model [27,28], the GEPIC model [29] (GIS-based environmental policy integrated climate model), the Environmental Policy Integrated Climate (EPIC) and the AquaCrop model [30]. The most widely used model for calculating the water footprint of crops is the CROPWAT model [31,32]. The primary advantage of the CROPWAT model is its flexibility and adaptability, enabling its application to a wide range of crops in nearly any location worldwide, while allowing for model localization and adjustment of calculation accuracy. Table 1 summarizes the main tools for assessing the water footprint of agricultural production. At the international level, water footprint studies are usually conducted either at the global or country level. For example, Tuninetti et al. [33] project the global consumption of green and blue water for wheat, rice, maize, and soybeans. These assessments reveal that crop production primarily relies on green water, with blue water becoming more efficient when irrigation is applied [33]. WF estimates are sensitive to various input parameters, particularly reference evapotranspiration (ET₀) and crop coefficients (K_c). Blue water footprints (WFs) tend to be more sensitive to input variability than green water footprints. The sensitivity of WFs varies across crop types, water supply methods, and regional climates. Uncertainties in WF estimates due to combined input uncertainties can be around ±26% at a 95% confidence level [34,35]. Figure 2 illustrates the calculations of agricultural water footprints.

Table 1:

Summary of water footprint assessment tools for agricultural and energy sectors [18, 27-30, 36-38].

Model/Tool	Developer/Source	Main Features	Strengths	Limitations	References
CROPWAT	FAO	Calculates crop water requirements and irrigation needs based on climatic and crop data	Widely used, easy to operate, adaptable to various locations, supports multiple crops	Limited pollutant modelling requires accurate local climate and soil data	[27,28]
GEPIC	IIASA & CAU	Integrates GIS data with crop models and environmental data to estimate crop production and water use	Spatially explicit; considers land use, hydrology, and environmental impacts	Complex setup; limited access to input data in data-scarce regions	[29]
EPIC	USDA	Simulates crop growth and environmental impacts, including nutrient runoff and sediment transport	Includes nutrient and pesticide leaching; suited for policy assessment	Requires detailed input data; complex calibration	[36]
AquaCrop	FAO	Focuses on the relationship between water and yield in response to different water conditions	Simple interface; water productivity-focused; suitable for scenario analysis	Limited pollutant tracking; does not account for grey water components directly	[30]
SPARE: WATER	ZALF, Germany	Integrative tool for analyzing WF and water productivity at the farm level using a systems approach	Holistic integration of social-ecological dynamics; suitable for comparative studies	Still under limited use, high data and calibration demands	[37]
WaterStat/WF Assessment Tool	Water Footprint Network	Online tool following Hoekstra’s methodology for calculating green, blue, and grey WF	Transparent methodology; consistent with global WF standards	Limited crop-specific detail; primarily used for awareness/policy rather than detailed farm management	[18]
WaNuLCAS	ICRAF	Simulates nutrient and water dynamics in agroforestry systems	Good for mixed cropping/agroforestry; nutrient and water flow dynamics	Limited to agroforestry systems; not standardized for WF estimation	[38]

Research on water footprints spans global and regional scales and addresses different policy objectives. Global studies aim to identify countries with high agricultural water consumption to reduce the overall water footprint [39]. For instance, in China, a modified water footprint assessment framework revealed that 24.9% of water consumed by grain crops is transferred to animal products through feed grain [40]. Regional studies, such as those of the Chinese province of Liaoning, use input-output analysis to compare water footprints from both production and consumption perspectives. They help identify sectors with high water footprints and provide policy recommendations, such as changes in industrial structure and trade, water-saving technologies, and capacity-building to reduce the stress on water scarcity [41].

3.3. Food-Energy-Water (FEW) Relevance

The water footprint method is designed to facilitate the integrated management of water resources. The Harvard Water Resources Project Milliman (1962) was the first to propose an interdisciplinary, integrated social and environmental approach to water resources management. However, the implementation of many water resources management plans, including this project and Integrated Water Resources Management (IWRM), has often fallen short of expectations. There are three main reasons for this: (1) The IWRM project attempts to integrate all seemingly water-related resource elements, but their scope is not clearly defined; (2) The IWRM project focuses only on water resources and has a single accounting purpose; (3) The issue of unclear or inconsistent scope. In the past, water resource accounting methods mainly were limited to the river basin scale. For resources such as energy and food, which are based on statistical data from administrative units, accounting methods are limited. FEW also follows a cross-sectoral integration approach, but its more advanced features are: (1) FEW has a clearer accounting object, namely food, energy and water resources; (2) FEW does not need to focus only on water, and stakeholders from the agriculture and energy sectors are more likely to participate; (3) The choice of scale is more flexible and free, both river basins and administrative units can be used as units of analysis. The 2011 World Conference on Green Economic Development emphasized that food, energy, and water resources are fundamental pillars of sustainable development, with water playing a central role within the FEW nexus [52]. The inclusion of WF in the FEW models is an essential step to enable effective management of resources because it quantifies the consumption of water by sectors and points out trade-offs and synergies. One such significant real-world example is the Sponge City Initiative in China, launched in 2014 to address urban flooding, water pollution, and water scarcity with the help of low-impact development (LID) and green infrastructure. In Shenzhen, a pilot Sponge City, urban planning was informed by optimizing rainwater retention, reducing the blue WF for urban water supply by 15% (from 1.2 billion m³ to 1.02 billion m³ annually) and the need for energy-intensive water treatment by 10% [53]. These water savings supported urban agriculture by reallocating the saved water for irrigation, improving local food security, and reducing the virtual WF of food imports by 8%. By integrating permeable pavements, green roofs, and wetlands, the initiative minimized runoff, reduced energy consumption for drainage systems, and increased food production, demonstrating the policy relevance of the FEW Nexus [54]. Such examples highlight the importance of WF metrics in operationalizing the FEW frameworks for sustainable urban planning, even if scaling these practices in different hydrological contexts remains a challenge [55].

In the past, most studies have focused on the relationship between the two, especially the application of the water footprint to the study of the “water-food” Nexus and the “water-energy” Nexus. After the World Conference on Greening the Economy in 2011, the study of the FEW has dramatically increased, and the research contents include three aspects: (1) the concept of the FEW Nexus; (2) the modelling and simulation of the FEW Nexus; (3) the study of use cases, the identification of trade-offs and synergies of the FEW Nexus, and the formulation of cross-sectoral strategies for sustainable regional development. In the literature, the most commonly used methods for studying the FEW nexus can be classified into four main categories: (1) the ecological footprint method by Vanham [56], who studied the applicability of the water footprint to the FEW nexuses; (2) economic methods, such as the Computable General Equilibrium (CGE) method. Computable general equilibrium (CGE) models are particularly well suited to analyzing environmental policy and the food–energy–water (FEW) nexus, as they can evaluate both the costs and benefits of policy changes across multiple economic levels [57,58]. Integrated modelling methods, such as those described by Dale et al. [59], the authors systematically analyzed the water and energy use of the electricity generation system in California, USA, in the context of climate change, combining this with a hydrological model and long-term energy substitution planning; (3) online modelling platforms specifically used to assess the FEW Nexus include the “FEW Nexus Assessment Tool 2.0” [60]. There are more studies at the global, national, and cross-regional levels, but fewer studies at the city level. Ramaswami et al., [61] analyzed the relationship between food, energy, and water in New Delhi, India, from the perspective of urban systems. Domestic FEW correlation research started late but is receiving increasing attention. Figure 3 demonstrates the integration of the water Footprint into FEW nexuses.

In a series of System Dynamics Models (SDMs), different frameworks have been applied to investigate the sustainability of the food-energy-water (FEW) Nexus in various contexts. Aquaponics systems have been studied in peri-urban food production systems [62], in general aquaponics research [63], and in urban areas, where they have been used for systemic modelling to investigate sustainable practices [31,64]. SDM can reveal different interactions in FEW systems and thus becomes a tool for decision-making and informing policy processes about sustainability [62,63]. This has been used to describe the imbalanced relationships between supply and demand, to evaluate specific policies, or conduct scenario analysis to secure the resources of megacities optimally [31]. At the same time, SDM has also been used in considering the sustainability transition for urban FEW infrastructures, with particular attention to climate, policy drivers, and land use context [64]. Documented examples increasingly demonstrate the generalizability of SDM in addressing the challenges of the FEW Nexuses, including the sustainable management of resources at various scales and in diverse environments.

In summary, while there is a consensus on the need to include blue, green, and grey water in the water footprint assessment, there are differences in the choice of accounting methods, with CROPWAT preferred for agriculture and LCA for the energy sector, reflecting sectoral differences in data availability and complexity.

5.1. Current Challenges

(1) Data limitations are one of the biggest challenges in calculating the water footprint. They affect accuracy both in agriculture (data on grey water pollution from pesticides, nitrogen, phosphorus, and microplastics) [16] and in the energy sector (data on production and the supply chain obscure the distinction between water demand and actual consumption) [44]. The complexity of the estimation methods and deviations caused by overlaps among multiple departments affect the accuracy of water footprint calculations [83]. Insufficient data on water conservation facilities, for example, make it impossible to assess whether “physical water scarcity” or “economic water scarcity” influences policy decisions. The timeliness of WF assessments for dynamic resource management is limited, and substantial delays result from the conventional dependence on product input-output tables, which are updated every five years [84]. On the other hand, recent developments have significantly increased the water data’s temporal resolution. To monitor water balance in real time, the GRWNOD platform, launched in 2023, offers high-frequency, open-access, near-real-time data on water consumption in urban, industrial, and agricultural areas [83]. In areas where traditional monitoring is infrequent or delayed, it enables dynamic tracking of greywater contamination and provides reliable data for water footprint modeling. Complementary tools such as IoT-based smart metering supported by Low-Power Wide-Area Networks provide granular data for urban and industrial WF calculations and improve the accuracy of consumption tracking [85]. Remote sensing platforms, such as OpenET, utilize satellite data to provide field-scale estimates of evapotranspiration. According to Thaler et al. [84], there are three areas where the accuracy of water footprint models needs to be improved: (i) data acquisition and pre-processing, (ii) calibration of model parameters, and (iii) methodological innovations tailored to sectoral and regional conditions. These limitations must be addressed for water footprint (WF) sustainability studies to be optimally utilized in decision-making, as highlighted in Section 4 [83]. To provide context to the challenges, Table 2 summarizes the primary water footprint research studies and highlights their methods, results, and identified gaps to improve the accuracy of agricultural WF [86]. Even when environmental water demand is included in the calculation, most calculations assume it is a fixed percentage of water flow. Ecological water demand varies across regions and seasons, and techniques for measuring or simulating regional ecological water demand remain underdeveloped. The reliability and accuracy of the water footprint calculation results are the basis for the sustainability assessment (Section 3.1) and the formulation of policy measures.

Therefore, improved methods for simulating regional ecological water demand, combined with high-frequency data, are needed to increase the reliability of the water footprint.

(2) A major challenge is the mismatch between the scales at which water is managed (often administrative) and the scales at which water processes and impacts occur (often seasonal and hydrological). Song et al. [102] in their research found that the spatial mismatch is greater than the temporal mismatch, especially in regions of high urbanization. Temporal scales are unable to distinguish among the different seasons, while geographical scales are unable to account for the movement of water and its different forms, such as virtual water. To manage water effectively, the correct scale must be chosen since different scales can initiate different management practices [103]. Seasonal water shortages are commonly underappreciated in interannual studies and therefore lead to false predictions of water supply and demand [104]. For example, for India’s Indo-Gangetic Plain, seasonal WF assessments for rice revealed 28% higher green WF during the monsoon months (July–September) compared with interannual averages (2500 m³/ton vs. 1950 m³/ton) due to significant variability in rainfall [105]. In sub-Saharan Africa, the estimated water demand for sorghum varied by 22% between the wet and dry seasons (1600 m³/ton vs. 2050 m³/ton), with interannual models underestimating water stress in the dry season by 12% [106]. These examples underscore the need for an adaptive timescale for WF research, as both short- and long-term climatic variability play dominant roles in water supply [107]. Water footprint accounting can support sustainable water management in large, semi-arid catchments by mapping the spatial and temporal patterns of sectoral water use [108]. Recent findings emphasize the links between climate variability, water resources, and socio-economic status. Short-term climatic fluctuations, particularly heavy rainfall, play a critical role in determining drinking water quality in developing countries [109]. High/low temperature and drought in farming regions negatively impact farmers’ psychological status, especially drought [110]. Prolonged hydrological extremes have persistent impacts on Sub-Saharan African wealth disparities, and dryness and wetness reduce regional GDP per capita [111]. Climate change is altering rainfall patterns, resulting in shorter and more intense rainy seasons that threaten water and food security in vulnerable regions [112]. These results show that adaptation measures and investments are needed to prevent future impacts on water resources and socio-economic well-being. Effective water resource management and policy decisions, therefore, depend on the careful selection of appropriate spatial and temporal scales for specific research objectives. In this context, the length of the study changes the results. For example, China’s evolving agricultural water footprint between 2000 and 2019 was influenced by significant route dependence and geographical dependence [113].

Calculation of the water footprint at the selected spatial level is complicated by the flow of water resources and the virtual availability of water [108]. This complication arises because both direct and indirect water consumption, as well as the multiple impacts of human management and weather conditions across different locations, must be considered. For example, Mao et al. [114] state that the water footprint is distributed differently across spatial scales (e.g., crop field, county, river basin) and is influenced by local management practices and climatic conditions. These findings are in agreement with those of Zhuo et al. [115], who found that different spatial scales yield distinct responses to water use in agriculture. The virtual water trade, that is, water embedded in exported and imported goods, must be updated to estimate regional water dependence [116]. Accurate calculation of virtual water is often hindered by a lack of data and methodological sophistication, making national water resources planning challenging [47]. Effective water management, therefore, requires a nuanced understanding of both local and global water flows, highlighting the need for spatially explicit assessments. Consequently, the precise choice of spatial scale is often independent of regulatory decisions, which in turn could alter the perceived upper limits of water resource use [107].

Aligning spatio-temporal scales with the characteristics and objectives of specific research topics facilitates a better understanding of water distribution and use, ultimately supporting the development of effective response strategies. Spatio-temporally disaggregated information can provide opportunities for water saving and offer strategic planning insights for further sustainable water management policies in large basins [108].

While choosing the correct spatial and temporal scale is crucial, it is equally important to consider the broader context of water resources management. This requires a good understanding of the socio-economic and environmental factors that affect water use and availability, in the wider context of how policy decisions affect the sustainability of water management. However, these issues require more global solutions as they evolve, i.e., by bringing together diverse data and methodologies to examine the dynamics of water resources holistically.

(3) Grey and green water footprint assessments are essential for sustainable water management under favorable conditions in urban landscapes and agriculture. The water footprint of crops has mostly been analyzed in the green domain, often overlooking the contributions of trees, pastures, and urban greenery. The analysis of grey water also needs improvement, especially regarding the self-purifying capacity of water bodies and the increasingly complex nature of pollutants. These topics are described in more detail in the following sections.

The green water footprint is the amount of rainwater consumed during plant growth, especially in agriculture. However, it is essential to know that urban green spaces within the metropolitan area also contribute to the water footprint. Unfortunately, these contributions are not consistently considered when valuing such areas [16]. Extending the green water footprint to recreational areas and forests could be one way to address the failure of urban water management systems [117].

In contrast, the grey water footprint quantifies the volume of water required to assimilate pollutants. Current methods are simple and do not capture the complex interaction of different pollutants [37]. Complex models such as SPARE: WATER provide extensive calculations due to the factors of the environment in which it exists, combined with the complexity of the pollutants [17]. Future studies should focus on integrating multiple variables into the assessment of the grey water footprint to provide a more accurate representation of environmental impacts [16].

Conversely, some argue that too much emphasis on the water footprint would undermine other large-scale approaches to optimizing the economical and efficient use of water. One school of thought suggests that a holistic approach is needed to use and manage water resources efficiently.

(4) The unequal distribution of water resources is evident in the disparities in per capita water consumption across countries worldwide, posing an additional challenge. Political dynamics, socio-economic development, and geographical conditions contribute to these differences. In this context, the work by Yalew et al. [118] highlighted that areas with higher levels of economic development have larger water footprints. The authors attributed this to the increased consumption and industrial activity. Furthermore, the relocation of production to less developed regions leads to greater pollution and resource degradation, which perpetuates inequalities [119]. A study of 27 countries from 1960 to 2010 found an N-shaped relationship between urban water use and GDP, known as the cubic water Kuznets curve [120]. In China, the urban water footprint was found to be 1.6 to 3.7 times higher than the rural footprint, with consumption levels accounting for the differences between provinces [121]. Globally, around 70% of countries exhibit an inverted U-shaped trend in water consumption due to technological progress and the optimization of production structures. The Organization for Economic Co-operation and Development (OECD) is a 36-member intergovernmental economic organization established in 1961 with the objective of economic growth and international trade [35]. OECD nations have abolished competition-restricting regulations and harmonized their product market policies to a greater extent [122]. The OECD countries with the highest water consumption typically generate over 60% of their GDP in the service sector, have an urbanization rate of 70%, and a GDP per capita of over $20,000 [108]. The results above show the complex nature of the relationships among economic growth, urbanization, and water use patterns.

Current studies focus on implementing distributive justice principles to manage water resources in ways that optimize equity and consider hydrological efficiency. Traditional models tend to overlook socio-economic impacts, thereby further aggravating inequalities [118]. Incorporating considerations of morality and distributive justice into the IWRM framework could promote equitable and sustainable water distribution, particularly in regions experiencing water scarcity [123]. Equity principles in running water systems can reconcile equity and efficiency, expand the space for solutions, and indirectly benefit poor people [124]. Equity in water distribution is considered central to a more equal future, and equity takes precedence over efficiency and markets in water policy [108]. This research highlighted the need for interdisciplinary collaboration and robust policy support to develop models that are technically sound and incorporate social equity provisions, aligning water resources management with higher social values of equity and sustainability.

Indices of equality, such as the Gini coefficient and Theil index, have been used in recent studies to quantify fair access to water sources across provinces and to provide numerical data on spatial and population inequality [125]. The Gini coefficient, a standard economic measure of income inequality, is a statistical indicator of dispersion ranging from 0 (equality) to 1 (maximum inequality). In the context of water resource distribution, it is assessed by plotting the Lorenz curve of cumulative water supply against cumulative population or supply area and calculating the ratio of the area between the line of equality and the Lorenz curve to the total area under the line of equality.

Similarly, the Theil Index is an entropy-based measure of inequality that breaks down overall inequality into within-group and between-group components. For this reason, it is particularly suitable for analyzing nested spatial scales, such as municipal, provincial, and national levels. The following formula is used for the calculation:

These indices are helpful in identifying areas experiencing oversupply (i.e., uneven water supply to a population or region) or undersupply (i.e., uneven allocation). For example, Yu et al. [125] employed the Gini coefficient to reveal substantial inequalities in the water supply systems of rapidly growing Chinese provinces. Li et al. [31] advanced this approach by developing a composite framework for evaluating water resource sustainability, which includes security (measured by resilience indicators), efficiency (measured by water productivity indicators), and equity (measured by the Gini coefficient) to assess the multi-layered sustainability of water resource systems. In the context of advancing water sustainability, this holistic perspective encompasses not only water availability and use efficiency but also equitable and just access for all population groups.

All these interventions have promoted a more equitable distribution of water. However, managing the international water regime and establishing mechanisms for international cooperation remain significant challenges. Addressing these concerns, therefore, requires an approach that effectively balances local needs and international sustainability requirements.

5.2. Future Research Perspectives

Future research must focus on exploring the FEW Nexuses by integrating ecosystem services and developing effective, comprehensive management practices. The Food–Energy–Water (FEW) nexus is recognized as a key driver of sustainable development, reflecting the complex interdependencies among these critical resources [99]. Integrated management practices should be implemented to enhance resource efficiency and minimize inefficiencies [126]. Analysts have employed predictive mathematical modeling tools and techniques to support FEW nexus planning, including advanced multi-criteria optimization methods and composite decision-making indices [126,127]. These techniques enable the identification of Nexus synergies and trade-offs, leading to improved resource management, policy planning, and formulation. The interconnections, linkages, and efficiency within the FEW nexus have been examined across most sectors, with gaps and opportunities for future improvement identified [128]. Stakeholders adopting a Nexus perspective can drive resource efficiency and sustainable action, thereby facilitating the achievement of the Sustainable Development Goals [99].

The Nexus approach to FEW addresses the interconnections between these sectors explicitly and recognizes that a shift in one industry significantly affects the other two drivers [129]. This holistic approach is necessary for sustainable resource management and for achieving the Sustainable Development Goals [99]. Various studies have suggested different quantitative models to characterize the FEW Nexus, such as input-output models to analyze interlinkages among sectors [130] and multi-criteria decision-making models to optimize production and resource use [131]. These help to identify synergies and trade-offs in agricultural land use systems [126]. Agricultural crops such as alfalfa are susceptible to energy and water use [130]. For example, according to Nie et al. [126], Pareto solutions could maximize food production, save water and energy, and function efficiently under a wide range of climatic conditions.

Political will and commitment are essential to foster synergies among stakeholders pursuing complementary goals within the basin. Ideally, cross-sectoral stakeholder engagement ensures the representation of different perspectives that would improve the coherence and efficiency of resource management strategies [132]. The concept of the FEW Nexuses seems promising to address some of the sustainability challenges. However, there are still critical bottlenecks in its operationalization, and much more research is needed to conceptualize and effectively use these linkages. Research linking theories and methods across food, energy, water, and ecosystem services is still at an early stage, and there is limited literature to date that integrates ecosystem services into a broader, more comprehensive system. There is a need for multidisciplinary and interdisciplinary comprehensive studies and breakthroughs from many case studies. There is also a great need for cooperation and coordination at transboundary and regional levels, as well as for pursuing compensation mechanisms to make water resources sustainable.

Use the “environmental footprint family” indicator to assess the environmental sustainability dimension of the human-environment system. The sustainability of the human-environment system includes social, economic, and ecological dimensions, with environmental sustainability as the foundation of social and economic sustainability. The family of ecological footprints includes the water footprint, carbon footprint, nitrogen footprint, phosphorus footprint, land footprint, energy footprint, and others [133]. Their common feature is that they can quantify human use of natural resources and comprehensively assess the ecological sustainability dimension of the coupled human-environment system. The concept of the ecological environmental footprint family is well-suited for addressing the limitations of focusing on a single resource while ignoring others, as it takes into account the environmental impacts of multiple resource use, which is particularly important for achieving sustainable regional development. Therefore, including the environmental impacts of social and economic growth in the assessment of the “footprint family” [134] is a new research idea and direction that considers sustainable development, the comprehensive development of society, economy, and environment.

The primary challenges in water footprint research include data limitations, methodological inconsistencies, and issues related to spatial and temporal scales. While there is a consensus on the need for improved data and methods, future research should focus on developing standardized protocols and integrated models to fill these gaps.

In summary, despite several problems and challenges in existing research, water footprint accounting, particularly the theory and method of linking food, energy, water, and other ecosystem services, can provide important scientific and technological support for fulfilling human well-being, reducing trade-offs in resource use, promoting mutual synergies, and fostering sustainable regional development.