Conflicting Objectives in Non-Conventional Water Valorization in the Mediterranean ^†

Abstract

In recent years, non-conventional water (NCW) has been providing a promising alternative against increasing water scarcity in the Mediterranean. However, little work is available regarding the socioeconomic effects of its use. The purpose of this study is to present the effects of different levels of availability of irrigation water on four different Mediterranean areas relevant to the valorization of NCW. The analysis is based on technical and economic data from four Mediterranean Living Labs (LLs): one in Italy, one in Spain, one in Egypt and one transboundary between Tunisia and Algeria. The methodological approach is based on different versions of mathematical programming (linear programming, parametric programming, multi-objective programming). The results of the analysis showed that future scenarios of water deficiencies will have serious implications on the cropping pattern and will severely affect certain farm types, equally affecting employment, incomes and input use.

1. Introduction

Water scarcity is one of the most severe challenges for the Mediterranean basin. Future climate change projections point to a fast acceleration of the problem, but also, the increasing phenomena of extreme weather events illustrate that actions and solutions are needed in a pressing way. It is expected that all water-demanding sectors, most of which are crucial for Northern and Southern Mediterranean economies, will be negatively impacted [1]. A promising option is to experiment and develop the use of non-conventional waters (NCW) (wastewater, runoff water harvesting, desalination) by developing technologies or rediscovering traditional techniques. More importantly, however, NCW has entered mainstream discussions about water management, and its relevance to the development of governance schemes is increasing [2,3].

Within this framework, AG-WaMED project seeks to support evidence-based water management in four Living Labs (LLs) located in Mediterranean watersheds (including a transboundary case) in Spain (Campo de Cartagena) and Italy (Val d’ Orcio) (Northern Mediterranean), as well as in Egypt (Wadi Naghamish) and a transboundary river catchment in Algeria and Tunisia (Wadi El Kebir) (southern Mediterranean). As part of the AG-WaMED project, a common framework has been developed to assess the economic and social importance of NCW in the four LLs. This paper presents the development and implementation of such a framework under the consideration that each LL faces specific challenges, but that there are also common external drivers that significantly affect the use of water (and particularly NCW) in each area and shape future prospects. In particular, mathematical programming methods are used to model the challenges in each LL and to suggest ways to optimize allocation of irrigation water among farming activities, and also to predict how the sector will balance different levels of water availability and how other conflicting objectives at the economic, social and environmental levels will be affected.

2. Methodological Approach

The first step of the methodological approach involved a detailed description of the farming sector in each LL based on published information and data, as well as on the input of stakeholders and the experience pf AGWAMED project partners. The initial stages of this analysis indicated that there were considerable differences in scale and sectoral organization in the four LLs, where agricultural production, in particular, differed considerably in terms of cropping patterns and product commercialization. In addition, social and environmental challenges were also present. In this context, the main challenges and objectives were co-decided with stakeholders in each one of the four LLs through a co-creation approach. The analysis in this paper is, thus, oriented to respond to a stated main challenge that has been defined separately for each LL.

After defining the main challenges in the current situation in the LLs, the collection of technical and economic data was the following step. In each LL, detailed technical and economic data were collected for main crops and farm types. According to availability, these technical and economic data were collected either directly from farmers—by means of questionnaire interviews—or were retrieved from published sources. Using these data, indicators of the socioeconomic performance of crops and/or farm types were calculated. These indicators were then used for an investigation of the current situation, as well as future scenarios, through mathematical programming models to evaluate the performance of each model of NCW use at the LL scale in terms of its effectiveness to induce benefits such as employment, farm income and less use of agrochemicals. The exact type of mathematical programming model differed across LLs according to the kind and nature of the specific challenge and conditions in each LL.

The specific types of mathematical programming that were considered and employed in this analysis were linear programming (LP), parametric programming (PP) and multi-objective programming (MOP). LP is a nonparametric mathematical procedure for optimal resource allocation which maximizes or minimizes a linear function of variables (objective function) that are subject to linear inequalities (constraints) and must assume non-negative levels [4]. The solution produces an optimal combination of activities in terms of cost minimization or output maximization. The PP method constitutes an extension of LP, adding more flexibility to the method, as it basically permits a sensitivity analysis of the solution of the LP model. In particular, in the PP model, one of the constraints in the model is allowed to vary within a predefined range, and the model yields a set of solutions reflecting optimizations under different levels of availability of the constraint. Multi-objective programming (MOP) is an optimization method that examines the possibilities of simultaneously achieving conflicting objectives (such as maximum gross margin (GM) and employment in contrast to minimum water consumption or use of fertilizers and pesticides) [5]. The purpose of the MOP model is to evaluate the performance of each model of NCW use at the Living Lab scale, taking conflicting objectives into consideration. This model (MOP) yields a set of solutions that accomplish various objectives at different levels, a “space” of non-inferior compromising solutions which will allow LLs to examine them and decide on which one is the best [6].

3. Results and Discussion

Figure 1 presents three selected solutions of a MOP model in the cross-boundary LL of Algeria and Tunisia with the main question “How to cover growing demand without burdening water resources, and what is the role of smallholders versus larger ones?”. In this LL, NCW involves a combination of boreholes, treated- and floodwater to face the challenges of climate change. Water is mostly used for agriculture which principally covers local and regional demand. Family farms are predominant, and in the last few years, smart agriculture has gradually been adopted, including the expansion of irrigated areas, which increases risks of groundwater salinization and the use of noxious agrochemicals. This process is hindered by the lack of labor as younger family members often migrate and leave the primary sector, and skepticism regarding modernization. Figure 1 presents three indicative solutions which compromise conflicting objectives (maximization of GM, minimization of water and agrochemical use) under low labor availability. Cereal and trees increase alongside a reduction in potatoes and vegetable cultivation, and this leads to notably less use of agrochemicals. In extreme conditions of water unavailability, the expansion of olives can lead to water savings, while incomes are reduced in case of water or agrochemical use reduction.

In the case of the Egyptian LL, water is managed collectively, and runoff water is stored in cisterns and behind dikes. This is an old practice which is evolving to deal with increasing droughts and water scarcity. Small family farms are predominant in the study area of the LL, and specialize into two tree species (figs and olives) which are very important for their survival. Although the production is quite extensive, intensification is ongoing, with new production technologies which face cultural barriers. In this context, an MOP model was chosen to investigate whether the adoption of new technologies can be justified in the future, even under water scarcity scenarios. As shown in Figure 2, where selected solutions are illustrated from an MOP model, a part of the land remains uncultivated, even in the case of increased water availability, while significant water deficiency can lead to desertification (solution PP_Water_2), where only olives remain in the cropping pattern. Only marginal changes for agrochemical use are possible, linked with less olive cultivation.

Figure 3 presents the results of a PP model for Italian LL where the availability of irrigation water is allowed to vary. In Val d’ Orcia, rainwater harvesting was the main NCW source, with the use of small ponds to collect rainwater that can be available for periods of water shortage. Nevertheless, these ponds are becoming outdated and need maintenance, while longer periods of drought are being witnessed year on year. The area is dominated by small- and medium-sized family farms with a variety of tree and arable crops. However, the focus is on durum wheat and vineyards, which are products with important market shares and exports. The main question for this LL is, therefore, how to protect these crops under volatile water availability in the future. As illustrated in Figure 3, the actual and optimal situations are almost the same, but water availability brings very significant changes. More water availability favors the predominance of vineyards, while a 29% reduction in water will lead to a complete loss of viticulture, with gains for arable farms. In future extreme draught conditions, olive production seems to be the only viable activity.

In the Spanish LL (Campo de Cartagena), extensive use of NCW was reported, with the aim to balance the increasing water scarcity. The use of desalinized water, water reuse and water trading are well-established in the area. The LL includes a large area with more than 35,000 ha cultivated per year, and very intensive production oriented towards exports. Vegetables, citrus fruit and olives are the main crops, while almonds are especially typical for the area. Under these conditions, the main question for the analysis was “How can intensification be balanced with protection of water resources?”. Figure 4 presents the results of PP under different levels of irrigation water availability. It is illustrated that specialization of vegetables and other tree crops can lead to 10% water savings, while the gross margin can increase by 13%. However, less water availability implies that 5000–20,000 ha are in danger of abandonment, which is particularly relevant for almond trees, while olives, vegetables and some fruits will expand as water becomes scarce.

4. Conclusions

The analysis, by means of mathematical programming, showed that future possible water (un)availability is very crucial for all LLs, and can have significant effects for key crops (e.g., for PDO vineyards in Italy and figs in Egypt). In this context, other social and economic challenges, such as labor availability, are important for areas where specialized crops with high-quality characteristics are present (e.g., in Italy LL), and can be a significant limiting factors in the value of NCW solutions. Furthermore, more in-depth analysis remains necessary to reveal the exact details of possible investments in NCW valorization. However, at this stage, the analysis has shown that water losses could entail very significant losses of economic returns and employment, while they could mitigate production uncertainties.

There are significant discrepancies in the achievement of conflicting objectives, and input from stakeholders, as well as balancing priorities from policy-makers is crucial to design effective strategies to secure water availability. As future steps, the scenarios derived for each LL will be presented to stakeholders, who will provide feedback based on their priorities and needs, while parts of these results will be revised or complemented with more details to examine additional possible future scenarios. In order to support this process, the gross margin calculations in this study have not taken income support or other subsidies into account in order to demonstrate the market potential of cropping patterns without possible distortions relating to policy.