Measuring the Sustainability of Water Plans in Inter-Regional Spanish River Basins
Abstract
This paper analyses and compares the sustainability of the water plans in the Spanish River basins according to the objectives of the Water Framework Directive. Even though the concept of sustainability has been traditionally associated with the triple bottom line framework, composed of economic, environmental, and social dimensions, in this paper sustainability has been enlarged by including governance aspects. Two multicriteria decision analysis approaches are proposed to aggregate the sustainability dimensions. Results show that the environmental dimension plays the most important role in the whole sustainability (40%) of water basins, followed by both economic and social criteria (25%). By contrast, the dimension of governance is the least important for sustainability (11%). A classification of the Spanish basins according to their sustainability indicates that the water agency with the highest sustainability is Western Cantabrian, followed by Eastern Cantabrian and Tagus. By contrast, Minho-Sil, Jucar, and Douro are the least sustainable.
Full text
water Article Measuring the Sustainability of Water Plans in Inter-Regional Spanish River Basins María M. Borrego-Marín †and Laura Riesgo *,† Department of Economics, Pablo de Olavide University, Ctra. de Utrera, km. 1, 41013 Seville, Spain; [email protected] *Correspondence: [email protected]; Tel.: +34-954-349851 † The authors contributed equally to this work. Academic Editors: Giacomo Zanni, Davide Viaggi and Meri Raggi Received: 30 April 2016; Accepted: 5 August 2016; Published: 11 August 2016 Abstract: This paper analyses and compares the sustainability of the water plans in the Spanish River basins according to the objectives of the Water Framework Directive. Even though the concept of sustainability has been traditionally associated with the triple bottom line framework, composed of economic, environmental, and social dimensions, in this paper sustainability has been enlarged by including governance aspects. Two multicriteria decision analysis approaches are proposed to aggregate the sustainability dimensions. Results show that the environmental dimension plays the most important role in the whole sustainability (40%) of water basins, followed by both economic and social criteria (25%). By contrast, the dimension of governance is the least important for sustainability (11%). A classification of the Spanish basins according to their sustainability indicates that the water agency with the highest sustainability is Western Cantabrian, followed by Eastern Cantabrian and Tagus. By contrast, Minho-Sil, Jucar, and Douro are the least sustainable. Keywords: sustainability; Water Framework Directive; integral water management; multicriteria decision analysis; water policy design 1. Introduction A modern water management system must be not only effectively provide water security, but also be sustainable, combining economic progress with social development and the conservation of habitats and ecosystems. The Water Framework Directive (WFD)—Directive 2000/60/EC [ 1 ]—and the introduction of river basin districts may help to fulfil such objectives. The environmental objectives are defined in Article 4—the core article—of the WFD, aiming to achieve a sustainable water management system on the basis of a high level of protection of the aquatic environment. Achieving such sustainability requires some boundaries, as through the definition of river basin districts. These districts are hydrological units selected on the basis of the spatial catchment area of the river, and not depending on any administrative or political boundary. Spain has a wide tradition in water management through agencies called basin water agencies (BWAs), which have been operative since 1920. BWAs play an important role in water planning, resource management and land use, protection of the public water domain, management of water use rights, water quality control, planning and execution of new water infrastructure, dam safety programs, etc. The WFD sets out clear deadlines for each of the requirements as can be consulted in [ 2 ]. Within such milestones, water administration agencies from each member state have to report each issue to the European Commission on time, with 2015 being a relevant date in the WFD implementation. Thus, the first management plan (River Basin Management Plan 2009–2015) has been finalised and Water 2016,8, 342; doi:10.3390/w8080342 www.mdpi.com/journal/water
Water 2016,8, 342 2 of 14 the second management plan (River Basin Management Plan 2015–2021) and the First Flood Risk Management Plan have just started. Since the first River Basin Management Plan has finalised quite recently, it is of particular interest analysing the sustainability of Spanish BWAs in water management and their contribution to fulfil the WFD objectives. In this sense [ 3 ], it is recommended to strengthen the links between water planners and academics in order to improve future revisions of the River Basin Management Plans. More concretely, it is proposed that the assessment and the selection of methods were done jointly in order to design and implement new water policies in Spain. In addition, the role of BWAs is highlighted as potential coordinators of such evidence-based policy-making. Considering this framework, the objective of this paper is to analyse and compare the sustainability of water plans in the Spanish river basins according to the objectives of the WFD. In addition, dimensions that may be enhanced to improve the basins’ sustainability are analysed, being this analysis a starting point to improve water management sustainability in the following management plans. After this brief introduction, Section 2reviews some of the previous works on assessing sustainability by using multicriteria decision-making methods. In the Section 3the case study is presented. Sections 4and 5include the methods used to assess the sustainability of water plans and results. Finally, Section 6concludes the paper. 2. Literature Review Sustainability has been used as a criterion to analyse water resource management quite often in the literature. In order to assess such sustainability, multicriteria decision analysis (MCDA) has been commonly used since the 1970s. It is possible to find a considerable number of applications related to water management on different river basins. Thus, Hajkowicz and Collins [ 4 ] reviewed 113 studies that used MCDA for analysing water resource management. They found that these methods are of relevance since the annual publication rate has been steadily growing since the late 1980s. The majority of applications are related to the fields of water policy, supply planning and the evaluation of major infrastructure. Regarding the evaluation of different water management strategies, it is worth highlighting [ 5 ], in which a three-step process is developed to evaluate different water management strategies in a river basin in Brazil. The analytical hierarchy process (AHP) was used to help identifying the groups of interest, articulate their preferences and find the dominant preferences of the community within the river basin, as well as to get a consistent evaluation of management strategies. In addition, Martín-Ortega et al. [ 6 ] performed a multicriteria analysis of water management under the WFD. They selected some measures for a sustainable and socially accepted water management in the Guadalquivir river basin in order to test the applicability of the AHP in the new WFD context. A survey was carried out in the context of a future enlargement of a reservoir. Results suggest that the AHP is an adequate tool for the WFD purposes and a useful complement for the cost-effectiveness analysis. There are other works that analyse different water management strategies to address concrete problems in some areas. In this line, Jaber and Mohsen [ 7 ] proposed a support system for decision evaluation and selection of nonconventional water resources in the river Jordan. They include desalination of saline and seawater, treated waste water, importation of water across boundaries, and water harvesting. Using AHP, they found that water desalination was ranked the highest, being the most promising resource, followed by water harvesting. Freiras and Magrini [ 8 ] presented a selection of sustainable water management strategies for a mining complex located in the southeast region of Brazil, which concentrates most of the country’s population and the mining facilities, but a small portion of the water available in the territory. A stepwise process for incorporating environmental risks into the decision-making using a multicriteria approach and AHP was developed and applied in this case study. Da Cruz and Marques [ 9 ] used the MACBETH multicriteria model to determine sustainability level of urban water cycle services (UWCS). They show that it is possible to assess both global sustainability
Water 2016,8, 342 3 of 14 and performance of UWCS in each particular dimension of the sustainability, taking into account the values and judgments of the legitimate stakeholders. Recently, Marques et al. [ 10 ] discussed the concept of sustainable water services and suggested using MACBETH multicriteria method to assess it. They illustrated a real-world application of the method in urban water services (UWSs) in Portugal and used a simple additive aggregation model to calculate the sustainability score of each UWS. Finally, the work of [ 11 ] implemented MCDA in an irrigated area in Spain. They found six factors to define alternative strategies (policies) that could change the planning scenario of the irrigation system: irrigation system, water pricing, water allocation, crop distribution, fertiliser use and subsidies received. Five different MCDA techniques were used and results indicated that all techniques choose the same alternative strategy as the preferred one: sprinkler irrigation system, with no change in the existing water pricing and water allocation schemes, growing wheat and barley as the main crops with organic fertilisers and without any change in the subsidy policy. 3. Case Study The main Spanish BWAs exceed a single region, being called as inter-regional water agencies (IRWAs). We can distinguish ten different IRWAs in Spain, that is, Western and Eastern Cantabrian (Cantábrico oriental y occidental), Minho-Sil (Miño-Sil), Douro (Duero), Tagus (Tajo), Guadiana, Guadalquivir, Segura, Jucar, and Ebro. In addition, there are minor basins comprised in one single region, and called intra-regional water agencies, such as Galician Coast, Andalusian Mediterranean Basin, Tinto, Odiel and Piedras, Guadalete and Barbate, inland basins of Catalonia, Balearic Islands, and Canary Islands. The location of BWAs is showed in Figure 1. Water2016,8,3423of14 particulardimensionofthesustainability,takingintoaccountthevaluesandjudgmentsofthe legitimatestakeholders.Recently,Marquesetal.[10]discussedtheconceptofsustainablewater servicesandsuggestedusingMACBETHmulticriteriamethodtoassessit.Theyillustrateda real‐worldapplicationofthemethodinurbanwaterservices(UWSs)inPortugalandusedasimple additiveaggregationmodeltocalculatethesustainabilityscoreofeachUWS.Finally,theworkof [11]implementedMCDAinanirrigatedareainSpain.Theyfoundsixfactorstodefinealternative strategies(policies)thatcouldchangetheplanningscenariooftheirrigationsystem:irrigation system,waterpricing,waterallocation,cropdistribution,fertiliseruseandsubsidiesreceived.Five differentMCDAtechniqueswereusedandresultsindicatedthatalltechniqueschoosethesame alternativestrategyasthepreferredone:sprinklerirrigationsystem,withnochangeintheexisting waterpricingandwaterallocationschemes,growingwheatandbarleyasthemaincropswith organicfertilisersandwithoutanychangeinthesubsidypolicy. 3.CaseStudy ThemainSpanishBWAsexceedasingleregion,beingcalledasinter‐regionalwateragencies (IRWAs).WecandistinguishtendifferentIRWAsinSpain,thatis,WesternandEasternCantabrian (Cantábricoorientalyoccidental),Minho‐Sil(Miño‐Sil),Douro(Duero),Tagus(Tajo),Guadiana, Guadalquivir,Segura,Jucar,andEbro.Inaddition,thereareminorbasinscomprisedinonesingle region,andcalledintra‐regionalwateragencies,suchasGalicianCoast,AndalusianMediterranean Basin,Tinto,OdielandPiedras,GuadaleteandBarbate,inlandbasinsofCatalonia,BalearicIslands, andCanaryIslands.ThelocationofBWAsisshowedinFigure1. Figure1.Locationofinter‐regionalandintra‐regionalbasinsinSpain.Source:Adaptedfrom[12]. Thispaperisfocusedontheanalysisofthesustainabilityofintegralwatermanagementin IRWAs,whichaccountfor87%oftheSpanishareaand64%ofpopulation.AmongtheIRWAswe canseehighdifferencesintheareaandpopulationcovered.Tagusistheriverbasinthatsupplies watertothehighestpercentageofpopulation,mainlybecauseitincludesoneofthebiggestSpanish cities,Madrid,withametropolitanareapopulationofaround6.5million.Regardingthesizeofthe IRWA,Ebroextendsfornineregions,beingthelargestbasininSpain.Bycontrast,Eastern Cantabrianisthelowestbasinandcoversthelowestratioofpopulation. Figure 1. Location of inter-regional and intra-regional basins in Spain. Source: Adapted from [12]. This paper is focused on the analysis of the sustainability of integral water management in IRWAs, which account for 87% of the Spanish area and 64% of population. Among the IRWAs we can see high differences in the area and population covered. Tagus is the river basin that supplies water to the highest percentage of population, mainly because it includes one of the biggest Spanish cities, Madrid, with a metropolitan area population of around 6.5 million. Regarding the size of the IRWA, Ebro extends for nine regions, being the largest basin in Spain. By contrast, Eastern Cantabrian is the lowest basin and covers the lowest ratio of population. The main characteristics of the inter-regional water basins under study are summarized in Table 1.
Water 2016,8, 342 4 of 14 Table 1. Main characteristics of the Spanish inter-regional water basins. River Basin Area (km2) Area over Spain (%) * Population (No. of Inhabitants) Population over Spain (%) ** Number of Regions Involved in Spain Western Cantabrian 19,002 3.8 1,656,626 3.6 5 Eastern Cantabrian 6405 1.3 1,297,494 2.8 3 Minho-Sil 17,619 3.5 825,851 1.8 3 Douro 78,859 15.6 2,222,532 4.8 8 Ebro 85,569 16.9 3,226,921 6.9 9 Tagus 55,781 11.1 7,273,871 15.6 5 Jucar 42,851 8.5 5,178,000 11.1 4 Guadiana 55,527 11.0 1,443,707 3.1 3 Guadalquivir 57,527 11.4 4,480,321 9.6 4 Segura 20,234 4.0 1,884,220 4.3 4 Notes: * This percentage shows the area that each river basin represents in the total area of Spain; ** This percentage shows the population in each basin over the total population in Spain. Source: River Basin Management Plans 2015–2021 [13–22]. 4. Methods Within the framework of the MCDA, this paper assesses the sustainability of inter-regional water agencies (IRWAs). Sustainability is assessed by considering the traditional economic, environmental, and social dimensions (Triple Bottom Line [ 23 ]), but also governance. Each of the sustainability dimensions has been analysed using a number of indicators that will be presented below in detail. In a second step, the relative importance of indicators and dimensions/criteria is assessed through the analytical hierarchy process (AHP). Later, the IRWAs are classified in a ranking in terms of their sustainability according to the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) (see Figure 2). In summary, MCDA allows us to aggregate the performance of each attribute in each dimension, and afterwards to get a sustainability measure on the basis of the aggregation of each dimension. Water2016,8,3424of14 Themaincharacteristicsoftheinter‐regionalwaterbasinsunderstudyaresummarizedinTable1. Table1.MaincharacteristicsoftheSpanishinter‐regionalwaterbasins. RiverBasinArea(km2)Areaover Spain(%)* Population (No.ofInhabitants) Populationover Spain(%)** NumberofRegions InvolvedinSpain WesternCantabrian19,0023.81,656,6263.65 EasternCantabrian64051.31,297,4942.83 Minho‐Sil17,6193.5825,8511.83 Douro78,85915.62,222,5324.88 Ebro85,56916.93,226,9216.99 Tagus55,78111.17,273,87115.65 Jucar42,8518.55,178,00011.14 Guadiana55,52711.01,443,7073.13 Guadalquivir57,52711.44,480,3219.64 Segura20,2344.01,884,2204.34 Notes:*Thispercentageshowstheareathateachriverbasin representsinthetotalareaofSpain; **ThispercentageshowsthepopulationineachbasinoverthetotalpopulationinSpain.Source: RiverBasinManagementPlans2015–2021[13–22]. 4.Methods WithintheframeworkoftheMCDA,thispaperassessesthesustainabilityofinter‐regional wateragencies(IRWAs).Sustainabilityisassessedbyconsideringthetraditionaleconomic, environmental,andsocialdimensions(TripleBottomLine[23]),butalsogovernance.Eachofthe sustainabilitydimensionshasbeenanalysedusinganumberofindicatorsthatwillbepresented belowindetail.Inasecondstep,therelativeimportanceofindicatorsanddimensions/criteriais assessedthroughtheanalyticalhierarchyprocess(AHP).Later,theIRWAsareclassifiedina rankingintermsoftheirsustainabilityaccordingtotheTechniqueforOrderPreferencebySimilarity toIdealSolution(TOPSIS)(seeFigure2).Insummary,MCDAallowsustoaggregatethe performanceofeachattributeineachdimension,andafterwardstogetasustainabilitymeasureon thebasisoftheaggregationofeachdimension. Figure2.Outlineofthemethodologicalapproach. Selection of criteria & indicators to assess IRWA sustainability Decision matrix Weighting criteria Dataset (IRWA) AHP Ranking IRWA Literature review & experts 1 st step 2 nd step 3 rd step TOPSIS Figure 2. Outline of the methodological approach. Table 2shows the dimensions/criteria and indicators selected to assess IRWAs’ sustainability.
Water 2016,8, 342 5 of 14 Table 2. Dimensions and indicators to assess the sustainability of BWA. Dimension/Criterion Indicators Economic Ratio of cost recovery for water services. Water productivity, measured as the ratio between the gross values added of economic sectors (GVA) and the volume of water supplied to each sector. Budget limits, measured as the maximum expenditure in investments. Environmental Water stress, measured as the ratio of the volume of water consumed and existing water resources in the basin. Number of measures aimed at achieving environmental objectives. Efficiency: losses in distribution infrastructures. Volume of reused water in the total amount of water supplied. Social Additional population served over the resident population in the basin. Number of measures aimed at satisfying demands. Employment relative to the volume of water supplied in the basin. Governance Number of measures to improve governance. Number of administrations involved in the management, implementation and/or financing measures. Number of initiatives to encourage active participation of the public. The selection of indicators in each dimension has been based on both a literature review [ 24 – 26 ] and the expertise of a panel of experts. The economic dimension is measured through three indicators: 1. Ratio of cost recovery for water services. The concept of cost recovery appears in the WFD (Article 9) in the sense that member states shall take into account such principles, including environmental and resource costs, having regard for the economic analysis, and in accordance to the polluter-pays principle. Member states shall report in the river basin management plans the steps towards implementing the recovery of the costs of water services. Taking into account the WFD, the ratio of cost recovery is calculated as the ratio between revenues and costs for water services, including financial, environmental, and resource costs. An estimation of the cost recovery ratio of financial costs related to water services can be found in [ 27 ]. Environmental costs are related to the externalities that occur mainly in water extraction and discharge processes when affecting other users or ecosystems. Resource costs refer to the value of water scarcity. More information about environmental and resource cost in the context of the European WFD can be found in [ 28 ]. The higher the ratio of cost recovery, the higher the economic sustainability of the IRWA. 2. Water productivity, measured as the ratio between the gross value added (GVA) of economic sectors and the volume of water supplied to each sector. More information about the estimation of water productivity values can be found in [ 29 ]. The higher the water productivity the higher the economic sustainability of the BWA. 3. Budget limits, measured as the maximum expenditure in water investments. Due to the economic crisis in Spain, the IRWAs have limited their budget for investments. This may have an impact on the measures needed to achieve the objectives of the WFD. The lower the budget limits, the higher the economic sustainability of the IRWA. The environmental dimension is assessed on the basis of four indicators: 1. Water stress, measured as the ratio of the volume of water consumed and existing water resources in the basin. Water stress is an increasingly important phenomenon that causes deterioration of
Water 2016,8, 342 6 of 14 fresh water resources in terms of quantity (overexploited aquifers, dry rivers, and polluted lakes) and quality (eutrophication, organic matter pollution, and saline intrusion). It happens when water demand is greater than the available amount during a certain time or when it is restricted by its low quality for a time period. The lower the water stress, the higher the environmental sustainability of the IRWA. 2. Number of measures aimed at achieving environmental objectives. The main environmental objective established in the WFD is to achieve good status of water bodies. To do this, the IRWAs establish measures to prevent or mitigate the punctual and diffuse pollution and to involve hydrological and environmental restoration of the basin. The higher the number of measures aimed at achieving environmental objectives, the higher the environmental sustainability of the IRWA. 3. Efficiency measured as losses in distribution infrastructures. Once captured, the water must be transported to the point of purification, to then be stored in tanks from which the distribution infrastructures are supplied to the points of domestic, agricultural, or industrial supply, in which once used it is evacuated. The main technical problem of water distribution infrastructures is the volume of losses due to deterioration. The lower the losses in distribution infrastructures, the higher the environmental sustainability of the IRWA. 4. Recycled water volume in the total amount of water supplied. Reusing wastewater is an increasing practice in arid or semiarid countries, where water resources are scarce. The uses that can be given to recycled wastewater are many and varied: watering (crops, gardens, greenbelts, golf camps, etc.), industrial reuse (cooling, boiler feed), non-potable urban uses (greenery, fire extinction, sanitary, air conditioning, washing cars, cleaning streets, etc.), and others (aquaculture, livestock cleaning, snowmelt, construction, dust removal, etc.). The higher the recycled water volume, the higher the environmental sustainability of the IRWA. The social dimension is measured using three indicators: 1. Additional population served over the resident population in the basin. In addition to the local population in the basin, the population may increase during certain seasonal periods for different reasons: work, holidays, etc. This indicator measures the capacity of the basin to satisfy this additional water demand. The higher the additional population served, the higher the social sustainability of the IRWA. 2. Number of measures aimed at satisfying demands. Economic sectors require water (and other resources) to develop their economic activities. The IRWA provides a series of measures to be able to respond to this demand. The objectives of these measures are to increase the availability of resources through regulation and management infrastructures, encourage recycling, and increase water use efficiency. The higher the number of measures aimed at satisfying demands, the higher the social sustainability of the IRWA. 3. Employment relative to the volume of water supplied in the basin. This indicator refers to employment on activities that require water resources for their economic development. The higher the employment ratio, the higher the social sustainability of the IRWA. Finally, the governance dimension is assessed using three indicators: 1. Number of measures to improve governance. Governance allows addressing the problems of resource and territory management through an integrated and systematic way. Clark and Semmahasak [ 30 ] examine the introduction of adaptive governance to water management in Thailand. The analysis shows the significant role that the new approach may play in resolving underlying differences between stakeholders. The higher the number of measures to improve governance, the higher the governance sustainability of the IRWA. 2. Number of administrations involved in management, implementation and/or financing of measures. Besides the IRWAs, other administrations and institutions are also involved in the
Water 2016,8, 342 7 of 14 development, implementation, and financing of programs of measures. The higher the number of administrations, the higher the governance sustainability of the IRWA. 3. Number of initiatives to encourage active participation of the public. These initiatives encourage the transparency and participation of stakeholders in both the decision-making and the planning processes. Hedelin [ 31 ] analyses two criteria based on the concepts of participation and integration. She notes that these concepts work as well-established dimensions of both sustainable development and management. The higher the number of initiatives, the higher the governance sustainability of the IRWA. The values of these indicators for each IRWA have been assessed using the information included in the IRWA management plans [ 13 – 22 ], and can be found in the Supplementary Materials (Table S1). Considering the indicators mentioned above, two multicriteria decision-making methods were used to assess the sustainability of IRWAs. More concretely, AHP was used to get the importance of each dimension and each indicator in the sustainability of the IRWA, and afterwards TOPSIS allowed us to rank the IRWAs according to their sustainability. The AHP method was created by [ 32 ] as a structured but flexible technique for making decisions in a multicriteria context. This method is based on dealing with complex decision problems using a hierarchical structure. Figure 3shows the three-level structure considered for our case study. Water2016,8,3427of14 sustainabledevelopmentandmanagement.Thehigherthenumberofinitiatives,thehigherthe governancesustainabilityoftheIRWA. ThevaluesoftheseindicatorsforeachIRWAhavebeenassessedusingtheinformationincludedin theIRWAmanagementplans[13–22],andcanbefoundintheSupplementaryMaterials(TableS1). Consideringtheindicatorsmentionedabove,twomulticriteriadecision‐makingmethodswere usedtoassessthesustainabilityofIRWAs.Moreconcretely,AHPwasusedtogettheimportanceof eachdimensionandeachindicatorinthesustainabilityoftheIRWA,andafterwardsTOPSIS allowedustoranktheIRWAsaccordingtotheirsustainability. TheAHPmethodwascreatedby[32]asastructuredbutflexibletechniqueformaking decisionsinamulticriteriacontext.Thismethodisbasedondealingwithcomplexdecision problemsusingahierarchicalstructure.Figure3showsthethree‐levelstructureconsideredforour casestudy. Figure3.AHPstructure. Inthishierarchicalstructure,therelativeimportanceorweights(wk)ofeachcriterionor subcriterionhangingoneachnodeareobtainedfrompairwisecomparisonsbetweenthem.Inorder toperformthesepairwisecomparisons,a1–9scaleisused,asproposedby[33].Table3showsthe relativescoresandtheirinterpretation. Table3.Tableofrelativescores. ValueofajkScaleMeaning 1jandkareequallyimportant 3jisslightlymoreimportantthank 5jismoreimportantthank 7jisstronglymoreimportantthank 9jisabsolutelymoreimportantthank 2,4,6,8Middlevaluesoftheabove reciprocalajk=1/akj ScoresofthesecomparisonsareusedtobuildtheSaatymatrices(A=ajk),whichareemployedto determinethevectorofprioritiesorweights(w1,...wk,...wn).Althoughdifferentproceduresto estimatetheseweightshavebeenproposed,forthiscaseweselectedthesimplestone:thegeometric meanmethod[34]. TheAHPdecisiontechniquewasoriginallydesignedforindividualdecision‐makers,butwas promptlyextendedforgroupdecisions[34],suchasourcasestudy.Thus,inordertodeterminethe weightsattachedtoeachcriterionwehavetoconsiderthejudgmentsofagroupofpeople(p),each withhis/herownpairwisecomparisonmatrix(Ap=ajkp)anditsrelatedweights(wkp).Thisindividual informationissuitablytreatedinordertoobtainasynthesisofaggregatedweights(wk). IRWASustainability Economic Environmental Social Governance Costrecovery Waterproductiv. Budgetlimits Waterstress No.measures Efficiency Recycling Additionalpop. No.measures Employment No.measures No.administ. No.initiatives Globalobjective Criteria Indicators Figure 3. AHP structure. In this hierarchical structure, the relative importance or weights (w k ) of each criterion or subcriterion hanging on each node are obtained from pairwise comparisons between them. In order to perform these pairwise comparisons, a 1–9 scale is used, as proposed by [ 33 ]. Table 3shows the relative scores and their interpretation. Table 3. Table of relative scores. Value of ajk Scale Meaning 1jand kare equally important 3jis slightly more important than k 5jis more important than k 7jis strongly more important than k 9jis absolutely more important than k 2, 4, 6, 8 Middle values of the above reciprocal ajk = 1/akj Scores of these comparisons are used to build the Saaty matrices (A=a jk ), which are employed to determine the vector of priorities or weights (w 1 , ...w k , ...w n ). Although different procedures to estimate these weights have been proposed, for this case we selected the simplest one: the geometric mean method [34].
Water 2016,8, 342 8 of 14 The AHP decision technique was originally designed for individual decision-makers, but was promptly extended for group decisions [ 34 ], such as our case study. Thus, in order to determine the weights attached to each criterion we have to consider the judgments of a group of people (p), each with his/her own pairwise comparison matrix (A p =a jkp ) and its related weights (w kp ). This individual information is suitably treated in order to obtain a synthesis of aggregated weights (wk). For this purpose, Saaty et al. [ 35 , 36 ] suggest that group decision-making should be done by aggregating individual priorities using the geometric mean: wk“m cźp“m p“1wkp (1) For indicators weights, a panel of 25 experts in water management sustainability was consulted. The members of this panel have been selected on the basis of their experience in water management, their scientific and technical contribution to the analysis of water sustainability and their involvement in the development and implementation of river basin plans. In addition, experts have been also selected in order to cover different technical profiles, such as university lecturers, researchers in agricultural research centres, civil servants in charge of water policy implementation, environmental journalists, hydrogeologists, agronomists, economists, environmental organisations, and farmers. Before aggregating priority scores, the consistency of respondents’ pairwise choices was tested by means of the consistency ratio (CR) based on the eigenvalue method [ 37 ]. In this paper we consider only CR lower than 0.1 [ 38 ]. Taking into account this CR, the percentage of consistent experts was 72%. Once the weights of each dimension had been calculated, by considering the experts’ evaluations, another MCDA technique was applied in order to rank IRWAs according to their sustainability. To do that, TOPSIS was used. The principle behind the method is that the optimal alternative should have the shortest distance from the positive ideal solution and the furthest distance from the negative ideal solution. The positive and negative ideal solutions are artificial alternatives which are hypothesised by the decision-maker, based on the ideal solution for all criteria and the worst solution which possesses the most inferior decision variables. Assuming that every indicator has an increasing or decreasing scale, TOPSIS calculates the results by comparing Euclidean distances between the actual and the hypothesised alternatives. Generally, the TOPSIS approach consists of seven steps, as it is summarized below [39,40]. Step 1. Constructing the decision matrix D on the basis of the value of each indicator (F i ) by IRWA (Ai), where fij is the performance of the IRWA Aiwith respect to the indicator Fj. F1F2¨ ¨ ¨ Fj¨ ¨ ¨ Fn D“ A1 A2 . . . Ai . . . Am » — — — — — — — — — – f11 f12 ¨ ¨ ¨ f1j¨ ¨ ¨ f1n f21 f22 ¨ ¨ ¨ f2j¨ ¨ ¨ f2n . . .. . .¨ ¨ ¨ . . .¨ ¨ ¨ . . . fi1fi2¨ ¨ ¨ fij ¨ ¨ ¨ fin . . .. . .¨ ¨ ¨ . . .¨ ¨ ¨ . . . fm1fm2¨ ¨ ¨ fmj ¨ ¨ ¨ fmn fi ffi ffi ffi ffi ffi ffi ffi ffi ffi fl (2) Step 2. Normalizing the initial decision matrix to eliminate the effects of complex relations. The normalized value vij is calculated as: vij “fij dn ř j“1 f2 ij (3)
Water 2016,8, 342 9 of 14 Step 3. Calculating the weighted normalized decision matrix Rby using the weights w j obtained through the APH for each indicator. The weighted normalized value fij is calculated as: rij “vij ¨wj(4) Step 4. Determining the positive and negative ideal reference points: T`“ r` 1,r` 2, . . . , r` n(“ `maxirij ˇˇjPJ1˘,`minirij |jPJ2˘( (5) T´“ r´ 1,r´ 2, . . . , r´ n(“ `minirij ˇˇjPJ1˘,`maxirij |jPJ2˘( (6) where J 1 and J” are linked to the indicators with positive polarity (more is better) and the indicators with negative polarity (less is better), respectively. Step 5. Calculating the distances to the positive and negative ideal reference points using the Euclidean distance. The separation of each IRWA from the positive-ideal solution ( S` i ) and the separation of each IRWA from the negative-ideal solution (S´ i) is given by the expressions: S` i“g f f e n ÿ j“1 prij ´r` jq2(7) S´ i“g f f e n ÿ j“1 prij ´r´ jq2(8) Step 6. Calculating the relative closeness to the ideal solution for each IRWA (Ci): Ci“S´ i S` i`S´ i ,i“1, . . . , m(9) where C i is an index with values ranging between 0 and 1, where 0 corresponds to the worst possible performance of the IRWA and 1 to the best. Step 7. ranking the IRWA, according to the Civalues. 5. Results Table 4shows the results of the application of the AHP method. First, we can see the weights for the sustainability dimensions according to the preferences of the group of experts. The environmental dimension is playing the most important role in the whole sustainability (40%), followed by both the economic and social criteria (25%). The governance dimension is the least important for sustainability (11%) according to the panel of experts. Table 4. Normalised weights for dimensions/criteria and indicators. Dimensions Indicators Economic 0.246 Ratio of cost recovery 0.471 Water productivity 0.313 Budget limits 0.216 Environmental 0.402 Water stress 0.380 Number of measures of environmental objectives 0.358 Efficiency: losses in distribution infrastructures 0.133 Reused water 0.128 Social 0.246 Additional population served 0.236 Number of measures aimed at satisfying demands 0.394 Employment 0.370 Governance 0.106 Number of measures to improve governance 0.434 Number of administrations 0.247 Number of initiatives 0.319