Water allocation sustainability assessment in climate change: a modeling approach using water footprint and just policy

Water resources have been over-allocated in some basins worldwide, which has become a significant concern in water resource management and the environment (Douez et al. 2020). In addition, climate change and population growth have increased water demands and intensified this problem, especially in arid and semi-arid regions (Nivesh et al. 2023). Therefore, climate change and water scarcity have risked food security and sustainable development in arid regions. It still needs to be determined whether the existing water allocation regime fits these challenges and how it should be reformed (Hellegers & Leflaive 2015).

The fact that the primary consumer of available water resources is irrigated agriculture necessitates water management in the agricultural sector (Doulgeris et al. 2015). Governments are setting policies to reallocate water to environmental uses in many major river basins worldwide that were previously developed for irrigation (Bark et al. 2014). These reallocation policies are also developed in the form of market-based strategies for other environmental purposes, such as surface or groundwater quality management (Panagopoulos & Giannika 2022, 2023; Souri et al. 2023).

Water reallocation (RA) commonly reduces the amount allocated for different demands, especially in the agricultural sector, where it consumes up to 80% of available water (Berbel & Esteban 2019). Meanwhile, pressuring the irrigation sector for adjustments to reallocation policies can be perceived as a conflict with food security. Therefore, sustaining the balance between water irrigation and food security is a critical objective (Ren et al. 2018). Moreover, agriculture and food security are predicted to be significantly affected by climate change, though the impacts will vary by region and crops (Pilevneli et al. 2023). Based on the findings of the ADAPT project in seven basins, it was revealed that the agricultural sector requires effective farm management strategies, specifically in irrigation, to cope with water scarcity and adapt to climate change (Droogers & Aerts 2005). In this regard, deficit irrigation (DI) is widely considered a sustainable strategy to maintain food security and save water in arid and semi-arid regions (Doulgeris et al. 2015; Romero et al. 2022). DI application rate is 15–50% for different crops, especially wheat and apple (Delavar et al. 2020; Babaeian et al. 2021).

However, reducing the allocated water may face challenges in the implementation because of potential social consequences. Previous studies have demonstrated that social justice significantly impacts accepting such decisions within society (Syme et al. 1999; Hophmayer-Tokich & Krozer 2008). Users will more likely accept RA policies with reduced water allocation if they believe these policies will lead to a just distribution of water resources (distributive justice) through a fair decision-making process, even if it results in economic loss (Jakeman et al. 2016). Thus, considering water issues as matters of justice and utilizing distributive justice principles to assess the reallocating scenario is crucial. One of the most relevant distributive justice principles for RA and DI is the equality of resources. In Ronald Dworkin's proposed principle, everyone is responsible for their choices if the water resource is equally shared (Lamont & Favor 2017). A study has yet to be conducted on implementing Dworkin's theory in the allocation/reallocation of water resources. Despite the importance of justice in water resources allocation/reallocation, its concept has attracted little attention among scholars and policy-makers (Thaler 2021). Researchers usually use multi-objective models to allocate water resources in which justice is defined as a utility function or considered by simplified indicators like the Gini coefficient (Hu et al. 2016; Roa-García & Brown 2017; Xu et al. 2019; Vail Castro 2022), Bentham–Rawls criterion (Xu et al. 2019), social welfare, Rawlsian welfare, and Nash scheme (Munguía-López et al. 2019; Ochoa-Barragán et al. 2021). The shortcomings of Gini and other indicators are that only the inequality principle is emphasized, while justice requires a more extensive picture (Martin et al. 2015). In the case of water resource allocation, justice is about (1) ensuring everyone's access to a healthy environment and ensuring the human right to water and (2) satisfying the fair distribution of resources beyond basic needs, which reflects the concept of human development (Movik 2014). One of the principles of modern distributive justice is equality of opportunity and resources. As a first attempt to promote just sustainability (Agyeman & Evans 2004), this study will translate this principle into water reallocation rules to promote just sustainability.

This research initially aims to develop a modeling framework for the sustainability of river basins with dominant agriculture through climate change. Using this framework, we can afterwards reassess current water allocation strategies to be more adaptive against climate change impacts and food security in the future. For these purposes, WF sustainability, as accounting indices, is emphasized coupled with insight on just water redistribution. Dworkin's principle of equality is referred to as the pivotal philosophy for just water reallocation by DI. Therefore, the originality of this research mainly lies in its integrated approach to water reallocation by considering hydrological modeling, climate change impacts, WF assessment, just water reallocation, food production yields and security. Here, the Soil and Water Assessment Tool (SWAT) was used to simulate hydrological processes, production yields, and WF sustainability assessment in both the current and future climate change scenarios. In this regard, the Karkheh River Basin (KRB) was chosen where different stakeholders are potentially in conflict. The developed methodology is explained in detail in the following section. The paper is divided into four sections. Section 2 introduces the proposed meteorology and the KRB as a study area. Section 3 shows and discusses the results in different scenarios and, finally, section 4 concludes the main achievements of this research.

The SWAT model is used to assess RA scenarios between provinces. Therefore, it was initially set to configure the KRB according to the political and hydrological boundaries. Accordingly, the basin was simulated for the period of 1985–2015. Here, crop management operations, including the plant-growing season, fertilization, irrigation, and harvesting for the six main crops (winter wheat, alfalfa, apple, corn, sugar beet, and tomato) were applied to the model. Moreover, some adjustments were made in the SWAT code for appropriate hydrological simulation. Parameters such as manual irrigation efficiency, the interaction between aquifers and groundwater level, and the karst features’ simulation are modified in the model. More details about these modifications are explained in previous studies (Delavar et al. 2020, 2022).

To evaluate the performance of the SWAT model, the sensitivity analysis, calibration, and validation of the model have been carried out with the SUFI-2 algorithm in SWAT-CUP software (Abbaspour 2013). Here, the parameters with the highest impact on the simulated stream flows were defined. If a parameter has relatively considerable t-state value and its p-value is close to zero, it would be more effective on and sensitive to stream flows. The results of the sensitivity analysis with the p-value and t-state of parameters are presented in Table 1. They show that the SCS runoff curve number (CN2), groundwater parameters (GWQMN), and soil moisture parameters (SOL_Z) are the most influential parameters affecting stream flows in this study.

We used the monthly observed streamflow data of 17 hydrometric stations from 1985 to 2015 for model warmup (1985–1989), calibration (1990–2004) and validation (2005–2015). Table 2 shows the model performance for simulating stream flows in the 17 aforementioned stations using the coefficient of determination (R²) and Nash-Sutcliffe Efficiency (NSE). R² and NSE are more than 0.5 in most stations, showing the model performance is almost satisfactory (Moriasi et al. 2007). However, they have reported that other factors, such as quality and quantity of data collection, model calibration procedure, time step, and project scope and scale, might affect the model evaluation (Babaeian et al. 2021) as our results are affected by the high numbers of sub-basins (153) and HRUs (1,658).

In this research, the WFS indicator will be applied to analyze the sustainability of the water resources and agriculture system in the current condition and possible future situations by the SWAT model. WFS indicates whether the current water used in the basin ensures the availability of sufficient water for future generations (Pellicer-Martínez & Martínez-Paz 2016). The WF includes blue water footprint, green water footprint, and grey water footprint (Hoekstra et al. 2017). According to the purpose of this research, the sustainability of green, blue, and groundwater will be assessed.

Here, ET_green is direct evapotranspiration of precipitation, SW_green is soil moisture from precipitation, ET_irr is evapotranspiration from irrigation and precipitation together, PCP is the total precipitation, and EFR is environmental flow requirements, which is the aggregate of 1,283 million cubic meters required for Hour-al-Azim (Roozbahani et al. 2011) and the primary environmental flow of the river in the period t obtained by the Montana method.

This section projected temperature and precipitation data as output variables fed into the SWAT model. It is needed for future streamflow simulations and to quantify and project the effects of climate change on the future availability of water. For this purpose, the best model of the CMIP6 GCMs (IPSL-CM6A-LR) was used to analyze precipitation and temperature for the near future (2021–2050) under Shared Socioeconomic Pathways (SSPs) 5-8.5. The SSP5-8.5 scenario is high-force, representing the combination of high social vulnerability and high radiative forcing, and it is the only path to achieve the man-made radiative forcing level of 8.5 W m⁻² by 2100 (Sun et al. 2022). The criteria for selecting the GCMs model was based on their highly acceptable performance in the previous studies investigating climate change impacts on the KRB (Ashraf Vaghefi et al. 2015; Fereidoon & Koch 2018). Downscaling the GCMs outputs and generating weather variables was accomplished by the LARS-WG weather generator (Semenov et al. 2002). The model's performance was evaluated at two synoptic stations, Ravansar and Khoramabad (Figure 1), from 1990 to 2015.

The Principle of Natural Resource Equality means that everyone has a claim to an equally valuable share of Earth's natural resources. Ronald Dworkin proposed the theory of equality in resources as one of the first and most essential theories of Luck egalitarianism (Dworkin 1981a, 1981b). Dworkin proposed that people begin with equal resources but be allowed to end up with unequal economic benefits due to their own choices (Lamont & Favor 2017). Accordingly, after equal distribution of resources, if individuals (farmers in this study) are disadvantaged because of someone's choices, it cannot be included as an injustice despite its inevitable inequality in welfare (Barry 2017). Therefore, treating people as equals ensures that income and wealth distribution at any given moment is ambition-sensitive but not endowment-sensitive. Regarding sensitivity to ‘ambitions,’ Dworkin argues that provided people have an ‘equal’ starting point (in Dworkin's case, resources), they should live with the consequences of their choices (Lamont & Favor 2017). Concerning ‘endowments,’ Dworkin proposes a hypothetical compensation scheme in this scheme, people who are unlucky in the ‘natural lottery’ and disadvantaged in the natural distribution provided before equal sharing of resources (irrigation water).

Regardless of planting type, each farmer receives equal water for one hectare of irrigated land based on Equation (8) and Dworkin's principle in the starting point. Therefore, farmers can each make a unique profit by choosing their desired planting and irrigation methods. This inequality in economic benefits, which results from how to use equally allocated resources initially, is fair. In addition, according to the Dworkin principle, the need of the environment as the disadvantaged sector should be met before reallocating water to irrigated agriculture.

The KRB is Iran's third-largest and most productive river basin in the southwest of Iran. The KRB is located between a longitude of 46°06 to 49°10 E and a latitude of 30°58 to 34°56 N that approximately covers 42,267 km² surface area. This basin administratively includes seven provinces of Kurdistan, Kermanshah, Hamadan, Ilam, Lorestan, Markazi, and Khuzestan (Figure 2). Its highest and lowest points are about 3,623 m and 182 m above sea level. The long-term annual average precipitation in the basin varied from 300 mm in the arid region of the south to 800 mm in the semi-arid north (Fereidoon & Koch 2018). About half of the precipitation falls in the cool months of January–March, and the warmest months of June–September receive less than 2% of the total precipitation.

The Karkheh River is the main river of the basin that originates from the central zone of the Zagros Mountains and flows into the Hour-al-Azim wetlands on the border with Iraq. The KRB is one of the most productive agricultural areas known as the food basket of Iran, making up 9% of the irrigation fields of Iran and producing about 10% of the country's wheat (Ashraf Vaghefi et al. 2015). However, the water use efficiency in this section is the most challenging issue. Significant decline in the Karkheh River discharge in recent years increases concerns for drinking water and agriculture and the environmental needs of the downstream areas (Delavar et al. 2022). Accordingly, the KRB is a notable example of an agriculturally heavily exploited basin with typical water challenges faced similarly in other regions worldwide (Fereidoon & Koch 2018).

The Dworkin equality of resource theory was chosen as the basis for just water reallocation to improve basin sustainability. According to this principle, farmers should equally have access to irrigated water per land area (m³/ha). It should be noted that this water reallocation for agriculture should be carried out after allocating the required drinking (250 MCM) and minimum ecological water (1,600 MCM) as a human right to water. After satisfying the drinking and environmental needs, the agricultural sector in the basin will face a decrease in allocated water. The next step includes reallocating agricultural water based on the equality of resource principle (equal water per hectare of farmland). This may result in reduced allocated water for some provinces, while others may receive more water for agriculture.

It should be noted that the total produced calories in the current water allocation is approximately 37.56 Mcal/ha. By SSP5-8.5 without and with RA, this value is 36.24 and 32.83 Mcal/ha, respectively. It means that climate change reduces 3.5% of overall production, and RA adds more reduction by 9%. Nevertheless, these reductions may not endanger food security in this basin which equals 9 Mcal/ha. Therefore, RA would not be a real threat to food security, but future studies should consider revising WFS indices. For example, the index of food environmental footprint (FEF) is one step forward (Jamshidi & Naderi 2023), considering the combined effects of water pollution and nutritious production with a water–food nexus.

This study used the SWAT model to evaluate the three sustainability indices of blue, green, and black water scarcity of six agricultural productions in the Karkheh basin in three scenarios of current climate change (SSP5-8.5) and water reallocation policy (RA). According to the results, we can conclude that:

• Water footprint and its sustainability indices (BuWS, GnWS, and BkWS) can be modeled by the SWAT in different conditions, like climate change. They can provide an accounting framework to outline and compare the basin status with a water conservation perspective.

• BkWS is the main challenge of all provinces in the Karkheh basin (>1), while BuWS is also critical for some provinces. These two indices get worse by climate change, while GnWS remains rather constant. Here, corn, tomato, and sugar beet experience the highest reduction, up to 17–21%, which makes them vulnerable to production in climate change concerns. In Iran and the KRB, only corn is considered a strategic crop for self-sufficiency, while the others are not.

• The Dworkin principle is a recommended rule for fair water reallocation, allowing greater water allocations from agriculture to environmental purposes. WF per land area is equalized for all stakeholders here, in provinces. RA could improve basin sustainability in both the current and climate change scenarios. Here, it could also ensure the available water for drinking and ecological purposes. Under RA policy, water withdrawal in stakeholders (five provinces) was reduced by 21–48%, making BuWS becomes sustainable (<1), while BkWS remains unsustainable (>1).

• Despite improving water sustainability in the basin by RA, the production yield would be adversely influenced. Consequently, overall food production decreases. According to a research study, climate change reduces 3.5% of overall production, and RA adds more reduction by 9%. However, this reduction does not pose a threat to food security in this particular case. As food security is a crucial objective for sustainability, conducting further investigations into the impact of RA rules on food security in other regions is crucial.

This research provides a practical framework for water reallocation that promotes sustainability for both present and future generations. This integrated approach can be applied to regions that face environmental challenges and dominant agriculture. However, it is crucial to consider the economic implications of transferring water from agricultural to urban and industrial regions. Any significant changes to water reallocation could impact local and regional economies and the global food supply and demand. Furthermore, exploring different distributive justice principles, sustainability of grey water footprint, pollution reduction, and multi-objective optimization is recommended when studying water reallocation.

Iran Water Resource Management Company partially supports this study under contract No. S.0012.1401 with University of Tehran.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.