Research papersProbabilistic trade-off assessment between competing and vulnerable water users – The case of the Senegal River basin

Author, Amaury Tilmant, Jasson Pina, Maher Salman 

Highlights

• •The distributional impacts of infrastructural projects are assessed using trade-offs charts.
• •Probabilistic trade-offs charts reveal the vulnerability of water users.
• •In the Senegal River basin, impacts are not homogeneously distributed.
• •Traditional food production is more vulnerable to external stressors.
• •Coalitions of countries are emerging based on their competitive advantages.

Abstract

The development of river basins often involves trading-off competing objectives in an uncertain environment. Through a stochastic analysis, the trade-off discovery can be enriched to identify vulnerabilities; that is, the sensitivity of those losses with respect to changing conditions in the basin. Using the Senegal River basin as a case study, we generate probabilistic trade-offs to analyze the competition between objectives that are vulnerable to both natural and anthropogenic factors. Here the natural factors essentially correspond to the hydrologic processes and their natural variability, while the anthropogenic factors include future water allocation policies as well as the level of development of the water resources system. Our analysis of the trade-off relationships reveals the existence of two coalitions of objectives: traditional food production (agriculture and floodplain fisheries) versus hydropower-navigation. The probabilistic trade-offs also show that of the two main coalitions of objectives, the one dealing with traditional food production is much more vulnerable to changes in both hydro-climatic conditions and allocation policies. Since traditional food production mostly concerns economically and politically-marginal communities in downstream riparian countries, this larger vulnerability should be factored in when negotiating water and benefits sharing schemes.

Introduction

Certain regions of the globe are experiencing robust population growth rates that are likely to persist for decades to come. For example, according to Cotillon and Gray (2016), over the past 60 years, West Africa has witnessed the world’s fastest population growth with a fivefold increase (compared to a threefold increase for the rest of the world). In arid and semi-arid regions, such a large increase in rural population is therefore expected to add pressure on limited land and water resources, which in turn will have detrimental impacts on fragile ecosystems. As highlighted in many regions around the globe, the tension over land and water resources is likely to intensify if new hydraulic infrastructure are constructed to meet the energy and water needs of urbanites. This is the case for example in the Mekong River basin where tradeoffs between dam benefits and their undesirable societal and ecological impacts are quite contentious (Hecht et al., 2019).

In a river basin, arbitrating such distributional conflict is challenging due to heterogeneous endowments, highly variable precipitations, the presence of multiple water users and unidirectional externalities. As river basins develop and their water resources are increasingly committed, an ex-ante assessment of the trade-off relationship between competing uses may become instrumental for mitigating burgeoning conflicts (Hajkowicz and Collins, 2007). By providing a description of the gains and losses associated with particular water allocation policies and/or infrastructural development, the trade-off relationship contributes to the emergence of a common understanding among water users. It also provides policy makers with an incentive to increase the scope of negotiation beyond water allocation and seek creative solutions that will transform the zero-sum game into a positive-sum one through, for example, the development of benefits sharing mechanisms (Sadoff and Grey, 2002).

Because the discovery of those trade-offs is not easy, especially in complex water resources systems with multiple supplies and evolving water demands, it often requires the help of system-wide modelling and multi-criteria analysis. The approach adopted in this paper aims at identifying Pareto optimal solutions; that is, the set of solutions where the performance of at least one objective must be degraded in order to improve the performance of another one. In the objective space, the set of Pareto optimal solutions define the Pareto front, the frontier between dominated and unfeasible solutions. The common approach to trace out this Pareto front transforms the multi-objective problem into many single objective problems that are solved independently using standard mathematical programming methods (Cohon and Marks, 1975). This approach was used, for example, by Oven-Thompson et al. (1982) to trace out the trade-off relationship between irrigation and hydropower generation for the High Aswan dam in Egypt. The Pareto front was obtained after identifying optimal hydropower production levels associated with increasing water withdrawals for irrigation purposes. A similar approach was adopted in Tilmant and Kelman (2007) to assess trade-offs between energy generation and irrigated agriculture in the Euphrates River basin. Trade-offs between hydropower generation and environmental flows in the Zambezi River basin have been analyzed by Giuliani and Castelletti (2013). Grafton et al. (2011) optimized the trade-off in water allocation between irrigated agriculture and environmental flows in the Murray River basin in Australia.

Recent developments in bio-inspired multi-objective optimization methods, such as multi-objective evolutionary algorithms (MOEA), aim at constructing the Pareto frontier in a single run instead of solving multiple independent optimization problems (Siegfried and Kinzelbach, 2006). When combined with visual analytics, MOEAs allow exploring multi-dimensional trade-offs with more than 4 objectives (Reed et al., 2013). These methods were used, for example, by Hurford et al. (2014) to analyze trade-offs between economic, ecological and livelihood benefits as well as power generation, irrigated agriculture and water supply in a semi-arid river basin in Brazil. In Giuliani et al. (2015), that approach was used to discover trade-offs between urban water supply, atomic power plant cooling, hydropower generation and environmental flows in the Lower Susquehanna (USA). However, as pointed out by Castelletti et al. (2013), those simulation-based optimization methods may become computationally demanding when the number of objectives increases, and difficult to parametrize when the water resources system is large and the network complex.

In water resources engineering, vulnerability often measures the likely magnitude of failure with respect to some pre-defined thresholds (Hashimoto et al., 1982). In this study, we adopt a broader definition: vulnerability is the degree to which a system, subsystem, or system component is likely to experience harm due to exposure to a hazard, either a perturbation or stress/stressor (Turner et al., 2003).

The last decade has witnessed the development of methods to better identify and then assessed vulnerabilities in the water sector. The main driver behind this growing interest for vulnerability assessment is climate change and more specifically how to exploit climate information to inform decision-making (Weaver et al., 2013). This research line has led to the development of bottom-up approaches which rely on a stress test to identify the conditions under which the system requires adaptation measures (see e.g. Brown et al., 2012, Turner et al., 2014, Culley et al., 2016). To achieve this, the approach relies on an integrated hydrological and system model to determine the system’s performance with respect to the climate projections used in the stress test. Another driver is the need to assess vulnerabilities in non-cooperatively managed river basins where data are either absent or not shared (Rougé et al., 2018). Here too, a system model was needed to determine under which hydrologic conditions, and where, the system would most likely fail.

This paper examines the use of probabilistic trade-offs to measure not only the extent of the gains and losses among competing objectives but also their vulnerability with respect to natural and anthropogenic factors affecting those trade-offs. In this study, we restrict the natural factors to the hydrologic processes and their natural variability, but the analysis could be expanded and include other processes such as land degradation, climate change, etc. The anthropogenic factors include, for example, alternative (future) water allocation policies as well as the level of development of the selected water resources system. This paper therefore contributes to a growing body of work on best practices in strategic planning, which broadly encompasses the development of national policies, plans, programmes and strategies for resource development and benefit distribution (Wilson, 2019). More specifically, it directs attention to the characterization of elements needed to ensure greater equity when designing water and benefit sharing arrangements (Grey and Sadoff, 2007): Who wins (loses) and where? How sensitive are those losses to human-induced and natural stressors? These elements are particularly relevant when the ultimate goal of the benefit sharing arrangement is to build more resilient and adaptive communities (Dinar et al., 2015), especially when stakeholders involve economically and politically marginal groups (Morgera, 2016), when relations between stakeholders are characterized by power asymmetries (Molle et al., 2010).

To illustrate those concepts, we use the Senegal River basin as a case study. The Senegal River basin is a largely underdeveloped river basin in West Africa hosting two reservoirs, two power plants and several irrigation schemes mainly located in the lower reach. Currently, the main uses are irrigated agriculture (including flood-recession), hydropower generation and fisheries; industrial uses (mining) and domestic water supply can also be found in the basin but involve small volumes. This picture might completely change in the next two or three decades as riparian countries have plans to develop the hydropower and irrigation potential of the basin, expand mining activities and transform the lower reach into a 900 km inland fluvial route.

As pointed out by Kliot et al. (2001), the Senegal is one of the few transboundary river basins managed by a genuine active joint-management organization (OMVS, the Senegal River Basin Authority) with personnel coming from the four riparian countries. OMVS’ mandates involve not only the planning but also the operation of water resources in the basin. However, despite the high level of cooperation, turning this underdeveloped river basin into a major energy-food-transportation hub in West Africa will challenge policy makers and OMVS for years to come as it involves intra and inter-country trade-offs. As mentioned earlier, this paper focuses on the characterization of elements needed to ensure greater equity when designing water and benefit sharing arrangements, which requires exploring those trade-offs and showing how they may evolve in the next three decades starting from the current situation in the basin.

To characterize those elements, we first determine the optimal allocation policies that may be adopted to share water resources in an increasingly complex system. Then, policies are used in simulation and the hydrologic and economic impacts assessed. From this assessment, trade-offs are discovered, coalitions of objectives revealed and vulnerable objectives, and hence water users, identified. Moreover, those probabilistic trade-offs can also be used to inform a nexus dialogue between the respective sectors in order to improve cross-sectoral planning and achieve equitable trade-offs (Bhaduri et al., 2015). In the Senegal River basin, for example, those trade-offs not only show the interdependence between water, food, energy and the environment but also the vulnerability of some of the sub-sectors contributing to food security. Moreover, because they overlook the entire situation in the basin, river basin authorities such as OMVS might be in a unique position to bring together stakeholders from the water, energy, food and environment sectors in order to share those trade-offs, discuss synergies, and identify optimal solutions (Cai et al., 2018).

The paper is organized as follows. It starts with a presentation of the Senegal River basin, which is followed by a description of the development and management scenarios. Section four is devoted to the optimization model that has been used to determine the optimal allocation policies in the basin. Then, section five presents the procedure implemented to derive probabilistic trade-offs charts. Section six discusses the simulation results with a focus on the trade-offs. Finally, concluding remarks are given in section seven.

Section snippets

The senegal river basin

The Senegal River drains an area of 337.000 km²in western Africa (Fig. 1). The basin is shared by four countries: Guinea, Mali, Mauritania and Senegal. The headwaters are located in Guinea, the water tower of the Senegal River basin (SRB), where the Bafing River runs north until it merges with the Bakoye in Mali. From there, the Senegal River runs north-west through a series of falls and gorges before arriving in Kayes. Downstream of Kayes, the hydraulic gradient is much lower and the river

Development and management scenarios

Scenarios represent alternative potential development and management pathways of the basin for the horizon 2040–2050. Here, the term development implies alternative levels of water resources’ commitment in the basin, i.e. the degree with which water is diverted, controlled and used. Management indicates alternative allocation policies between competing uses.

Two levels of development are considered: the first level corresponds to the current situation in the basin; the second level represents an 

Determining optimal allocation policies

The hydrologic and economic impacts of alternative development and management scenarios are assessed using a river basin (hydroeconomic) model. Generally speaking, a hydroeconomic model combines economic management concepts with an engineering-level of understanding of a hydrologic system (Heinz et al., 2007), including water quality (Davidsen et al., 2015). In the Senegal River basin, the model must determine optimal allocation policies, e.g. reservoir releases, water withdrawals for offstream 

Analysis of simulation results

Fig. 3 shows how the trade-off relationship between the main economic activities in the Senegal River basin changes with respect to the development and management scenarios. Results for the Baseline scenario are given for indicative purposes only; this is the yardstick to compare the magnitude of changes in the various sectors, their sensitivity to the hydrologic variability and to allocation policies, when second generation dams and the new irrigation schemes will be built.

Starting with

Conclusions

While new water infrastructure are being built to sustain economic development in emerging or developing economies, we should also direct our attention towards the development of adequate institutions to mitigate their negative impacts on the environment, and to equitably share their benefits and costs. To achieve this, we need, among other things, to understand the distributional impacts of water development, but also the stakeholders’ vulnerability to natural and anthropogenic factors

CRediT authorship contribution statement

Amaury Tilmant: Conceptualization, Methodology, Software, Writing – original draft. Jasson Pina: Software, Formal analysis, Investigation, Visualization, Writing – review & editing. Maher Salman: Project administration, Funding acquisition, Writing – review & editing. Claudia Casarotto: Formal analysis, Investigation, Writing – review & editing. Fethi Ledbi: Formal analysis, Investigation, Writing – review & editing. Eva Pek: Project administration, Funding acquisition, Resources.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The work was supported by a project from the Food and Agriculture Organization of the United Nations (FAO) entitled FAO TCP/INT/360 ”Renforcement de la gestion des ressources en eau transfrontalières dans le Bassin du Fleuve Sénégal”. This study is an independent assessment commissioned by the FAO. It does not represent the views of any riparian country or of the FAO.

References (66)

• X. Cai et al.Understanding and managing the food-energy-water nexus – opportunities for water resources researchAdv. Water Resour.(2018)
• C. Davidsen et al.Hydroeconomic optimization of reservoir management under downstream water quality constraintsJ. Hydrol.(2015)
• S. Dinar et al.Climate change, conflict, and cooperation: Global analysis of the effectiveness of international river treaties in addressing water variabilityPolitical Geography(2015)
• J.S. Hecht et al.Hydropower dams of the mekong river basin: A review of their hydrological impactsJ. Hydrol.(2019)
• A.P. Hurford et al.Using many-objective trade-off analysis to help dams promote economic development, protect the poor and enhance ecological healthEnviron. Sci. Policy(2014)
• N. Kliot et al.Institutions for management of transboundary water resources: their nature, characteristics and shortcomingsWater Policy(2001)
• J.B. Kollat et al.A framework for visually interactive decision-making and design using evolutionary multi-objective optimization (video)Environmental modelling and Software(2007)
• F. Molle et al.River basin closure: processes, implications and responsesAgric. Water Manag.(2010)
• C. Sadoff et al.Beyond the river: the benefits of cooperation on international riversWater Policy(2002)
• A. AdamsFleuve Sénégal: gestion de la crue et avenir de la vallée(2000)

View more references

Cited by (12)

• MUWOS – Multiple use water optimization system for the power generation and navigation trade-offs analysis2023, Renewable EnergyCitation Excerpt :These trade-offs have been described and discussed in the literature [10,11,59–62]. Tilmant et al. [11] and Hashemi et al. [62] developed an object-oriented Bayesian network to integrate impacts computed in quantitative submodels of hydropower revenues. Teichert et al. [63] evaluated the water trade-offs arising from energy generation and aquatic habitat conservation policies.Show abstract
• Coupling coordination degree spatial analysis and driv