Optimal water allocation through a multi-objective compromise between environmental, social, and economic preferences

Highlights

• Sustainable water allocation requires the triple bottom line approach.

• Proposed multi-objective model covers economic, social, and environmental aspects.

• Our model focuses on water allocation of shared basins with competing stakeholders.

• We analysed the sensitivity of stakeholders to the objectives' weights.

• We applied the model to a real cases study to demonstrate its applicability.

Abstract

Raising water demands and insufficient freshwater resources are the main reasons of water conflicts in transboundary watersheds. Sustainable water allocation can be a resolution for water disputes as it addresses simultaneously economic, social and environmental benefits. In this paper, a multi-objective model is introduced, which leads to sustainable water allocation of transboundary watersheds taking into account all these three aspects. Five water allocation objectives are proposed for this model in which three of them address the social factors and others represent the economic and environmental preferences. The Compromise Programming technique is employed to solve the applied model to the Sefidrud Basin, Iran and several water allocation schemes are provided based on various weights combinations. The results of the model elucidate that the proposed model can allocate 83 percent of the Basin's water resources, to its stakeholders in a sustainable way while the environmental demand is satisfied.

Introduction

Water is an important factor in conflicts among stakeholders of shared watersheds (Nandalal and Simonovic, 2002). The conflicts inevitably occur when the stakeholders' water demands are not satisfied due to insufficient shared water resources (Palmer et al., 1999). Apart from enormous negative impacts on the environment, water conflicts can jeopardize economic prospects and the social welfare of the stakeholders (Nandalal and Simonovic, 2002). Water allocation as an important part of integrated catchment management (Jakeman and Letcher, 2003) is one of the most effective water management alternatives for dealing with the contradiction of increasing water demand and inadequate surface supplies (Khare et al., 2007) if it addresses economic, environmental, and social factors, simultaneously (UNESCAP, 2000). Hence, it can be considered as a resolution of water conflicts in transboundary watersheds. It should be emphasized that due to the limitation of water resources in watersheds, they must be used in a way that creates more benefits for societies.

Operations Research has a long tradition of supporting natural resources allocation (Plà et al., 2014); and thus, water allocation modeling are mainly formulated based on optimization techniques. Single objective optimization has been tried for resolving water conflicts of transboundary watersheds through water allocation (e.g., Kucukmehmetoglu and Guldmann, 2004, Pulido-Velázquez et al., 2006; Griffith et al., 2009, Zoltay et al., 2010, Housh et al., 2013, Roozbahani et al., 2013); however, it is not able to provide a sustainable water allocation due to its consideration of only one water allocation objective. Most of these models have utilized economic objectives in water allocation modeling and left out the consideration of the social and the environmental factors (see more discussion for example in Scoccimarro et al., 1999).

Multi-Objective Programming (MOP) has frequently been used to handle various issues related to water resource management (Haimes and Hall, 1974, Cai et al., 2004) such as water allocation, due to its multidisciplinary nature and complexity. This technique has been frequently employed for water allocation of unshared basins. For instance, Babel et al. (2005) developed a MOP model for optimal water allocation of the Nong Pla Lai Reservoir, Thailand. The model maximized the aggregate satisfaction of water demands; as well as, the aggregate net economic benefit achieved from the demand sectors. Two approaches; weighting technique and simultaneous compromise constraint were used to turn the two objectives into a single objective function. Although, the environment was taken into account as a sector in this study and the model implied to maximize its satisfaction, the environmental water satisfaction was not actually maximized, due to the maximization of aggregate satisfaction of water demands. Xevi and Khan (2005) introduced a MOP model for water conflict resolution between the agricultural sector and in stream environmental flow. The objective functions of the introduced model were to maximize water benefit, minimize variable cost and minimize total supplementary groundwater pumping requirements to meet crop demand from the irrigated areas. The weighted version of goal programming was employed for solving this model, which was applied to the hypothetical Irrigation Area using real data at Berembed weir on the Murrumbidgee River, Australia. In this model, the environmental water satisfaction is considered as a firm constraint, however, there was no objective representing the social factor. Liu et al. (2010) presented a MOP model for the optimal allocation of water resources in saltwater intrusion areas by considering three objectives: maximizing economic interest and social satisfaction and minimizing the amounts of polluted water. The model was examined in the Pearl River Delta in China while ignoring the environmental water satisfaction. Ahmadi et al. (2012) introduced a genetic algorithm–based multi-objective model for water allocation of a case study in the Aharchay watershed, Iran. The model provided desirable water quality and quantity for various sectors while maximizing agricultural production in the upstream region, mitigating the unemployment (social) impacts of land use changes, and providing reliable water supply to the downstream region. In this study, the quality of water was taken into consideration rather than the satisfaction of environmental water requirements. Anghileri et al. (2013) proposed a MOP model for conflict resolution between the agricultural sector and hydropower production in the Alpine watershed in Italy. The model's objective functions were to minimize the shortage of irrigation demand and maximize the hydropower production. The formulated problem was solved using the weighted sum method while the inflows of reservoirs in this basin were considered stochastic. In this study, satisfying the environmental water supply was a firm constraint while the model allocated water to these competing sectors without taking into consideration the economic factor. Rezapour Tabari and Yazdi (2014) developed a multi-objective optimization model based on inter-basin water resources and restoration of outer-basin water resources. They considered three water allocation objectives in their model: supplying inter-basin water demand, reducing the amount of water output from a basin boundary, and increasing water transfer to adjacent basins. These researchers failed to include the environmental and economic factors in their study.

Water allocation of transboundary watersheds are more complicated than unshared basins where stakeholders of transboundary basins are often limited by administrative boundaries, for instance states/provinces at the national level, or countries at the international level. In these cases, unsustainable water allocation can cause serious water disputes between the stakeholders and somehow brings about political problems between the stakeholders. Hence, the participation of social, economic, and environmental factors (UNESCAP, 2000) in water allocation modeling of transboundary basins is a strong need. In the case of water allocation of shared watersheds, MOP has only been implemented for a few watersheds.

The Aral Sea basin is a transboundary watershed. Several mathematical models were developed to allocate water between its stakeholders. McKinney and Cai (1997) introduced a MOP model for water allocation of this basin. The amount of flow to the Aral Sea, the basin's demands satisfaction, and equalizing the distribution of water deficits between stakeholders were considered water allocation indicators in this study. Cai et al. (2002) presented a water allocation model based on MOP for sharing water of the Syr Darya River basin (one of the two major rivers feeding the Aral Sea) among its stakeholders where the risk of water supply to stakeholders, the Aral Sea environmental water satisfaction, equity in water allocation, and economic efficiency in water infrastructure development were the model objectives. Schlüter et al. (2005) did the same study for optimizing water allocation of the Amu Darya River (second major tributary of the Aral Sea). The study's water allocation objectives comprised the deficits of water delivery to all stakeholders, the planned flow to the Aral Sea, the degree of filling of the reservoirs, and the demand for stability of the system. In all three studies, weighted sum technique was utilized for finding the solutions of the developed water allocation models. These studies addressed economic and social factors in their water allocation objectives; however, the maximization of the basin benefit from water usage was not selected as the economic factor. Moreover, they only focused on the satisfaction of the Aral Sea water requirements, which is situated in the downstream of the basin, rather than the water supply to the environment for the whole body of watershed.

The Euphrates and Tigris River basin is a transboundary watershed, of which three countries, Turkey, Iraq, and Syria are its stakeholders and share its water resources. Kucukmehmetoglu and Guldmann (2010) introduced a MOP model for sharing water resources of this basin between these three countries. The model's advantage was to involve all stakeholders' profits (the profits of Turkey, Iraq, and Syria) in water allocation modeling instead of maximization of the basin's profit. The model's objective functions were the maximization of stakeholders' profits from water consumption by different sectors. In this study, the environmental water satisfaction was disregarded. They used the weighted sum technique and constraint technique to find the solutions of their model and selected three different weights for the objective functions while the justification of these weights was not transparent.

This paper first proposes five water allocation indicators to be considered in water allocation formulation of transboundary watersheds in order to achieve sustainable water allocation. Then, it introduces a water allocation model based on the MOP approach to determine the water portions of stakeholders (administrative divisions) of a transboundary basin, incorporating all proposed indicators. The implementation of the proposed model is illustrated in a real case study. In this part of the paper, a correlation analysis based approach is introduced in order to select the objectives creating highest conflict for the developed model for the real case study. Then, the Compromise Programming technique is utilized for finding the solutions for the developed model.

The structure of the paper is organized as follows. In the next section, the model formulation is presented. The case study of this research is briefly discussed in Section 3. The results of the model are presented in Section 4, followed by the conclusion in Section 5.