A new class of conceptual simulation tools, as a complement to physically based models, is becoming available to simulate the whole water cycle in urban areas for strategic planning, often involving the allocation of a great amount of financial resources. These simulation tools are required to estimate the impact of the today's decisions on the system performance over the next decades and to compare and rank different intervention strategies. To achieve this, this paper aims to build the metabolism-based modelling of a real water supply system using the recently developed WaterMet2 model in order to evaluate long-term performance metrics for possible intervention strategies. This metabolism-based approach was demonstrated for evaluation of the water supply system of Reggio Emilia, Italy, which is one of the demonstration case studies in the EU TRUST (TRansitions to the Urban water Services of Tomorrow) project. Based on the strains imposed by pressing challenges (here population growth) two intervention strategies were analysed. The results obtained show that the built and calibrated WaterMet2 model allows a broader understanding of the impacts of alternative intervention strategies taking into account multidimensional aspects of the sustainability beside conventional service performance.

INTRODUCTION

Urban water systems (UWS) play an important role in the great sustainability challenge of reconstructing cities (Kennedy et al. 2011). An efficient future plan for sustainable use of water and other resources in a UWS needs to take into account their long-term impact on other flows in the UWS such as materials, energy and costs, in addition to the ability to meet the service goals of water supply. Some recently developed tools which have striven to attain this aim are Aquacycle developed by Mitchell et al. (2001) as a water balance model, urban water optioneering tool (UWOT) by Makropoulos et al. (2008) as a sustainable water management tool for selection of combinations of water-saving technologies, urban volume and quality (UVQ) by Mitchell & Diaper (2010) as a further modified version of Aquacycle to include contaminant and energy flow, and CWB by Mackay & Last (2010) as a city water balance model. However, none of these models can be considered as a holistic systemic perspective for: (i) analysing the main resource flows and their impacts on the future performance of UWS; (ii) examining intervention strategies for long-term planning in UWS. This approach can be envisaged through metabolism based modelling. Metabolism in UWS refers to the fluxes and conversion processes related to all kinds of water flows, materials and energy in the UWS, which are necessary to fulfil expected functions (Venkatesh & Brattebø 2011). Accounting for the metabolism dynamics enables decision makers to identify the critical components which have a major impact on different sustainability dimensions. This will also allow investigation of various intervention strategies which concurrently save water, energy, chemicals and costs and also minimize negative environmental impacts. The metabolism concept calls for a systematic engineering approach to the examination, understanding and improvement of urban water services, and it offers the possibility of a well-structured quantitative analysis of how the key system characteristics are interrelated. All this approach is currently encapsulated in a recently developed concept of the WaterMet2 model which simulates metabolism based performance of UWS over a long-term planning horizon (Behzadian & Kapelan 2015b). The performance indicators calculated in WaterMet2 can also be used to support risk-based indicators which are useful for a more comprehensive analysis in UWS (Ugarelli et al. 2014a, 2014b). WaterMet2 has been developed through TRansitions to the Urban water Services of Tomorrow (TRUST), a 4 year research project funded by the European Union (www.trust-i.net/). The ambition of TRUST is to deliver co-produced knowledge to enable water utilities to achieve a sustainable future without compromising service quality. The work presented in this paper is one of the products of TRUST aimed at delivering this ambition. The paper presents the application of the WaterMet2 model to the urban water supply system of the city of Reggio Emilia (Italy), managed by IREN Spa. This system, serving 170,000 inhabitants, is one of the demonstration case studies of the TRUST project. Therefore, this study aims to: (i) demonstrate the application of the above metabolism concept using the case study; (ii) contribute to the evaluation of the level of sustainability in the water supply system of Reggio Emilia; (iii) highlight the use of WaterMet2 as a provider of additional information, in respect to the traditional simulation tools, to the selection process of the most appropriate intervention strategies to overcome the future strain and challenges of the water supply system. The paper is organized as follows: the main features of the WaterMet2 model used in this study are first introduced in the next section, followed by a description of the case study of Reggio Emilia and its characteristics. Next, the built WaterMet2 model for Reggio Emilia case is illustrated. The results of the simulated model are then presented and discussed. Finally, conclusions are drawn and some recommendations for further research are made.

URBAN WATER METABOLISM MODEL: WATERMET2

WaterMet2 has been developed to calculate indicators for assessment of sustainability performance in UWS (Behzadian et al. 2013). WaterMet2 is a conceptual, simulation type, mass-balance-based, integrated UWS model which quantifies metabolism-related key performance of UWS with focus on sustainability-related issues over a long-term planning horizon (Behzadian & Kapelan 2015b). WaterMet2 tracks down a number of metabolism based fluxes within the operating phase of the UWS by using a range of input mass fluxes (e.g. water inflow, energy and chemicals used). This, in turn, will enable WaterMet2 to calculate metabolism related indicators in the UWS including principal water-related flows (e.g. water demand and supply), environmental-related fluxes (e.g. greenhouse gas (GHG) emissions), cost flows and so on. The WaterMet2 model is able to simulate the main UWS components. WaterMet2 is a distributed model which allows the user to define any arbitrary number of each type of the UWS components (e.g. water mains, service reservoirs (SR) and sub-catchment (SC)). Moreover, WaterMet2 can support various types of water demand profiles and water recycling options. WaterMet2 is also able to simulate rainwater harvesting and grey water recycling schemes in the UWS. WaterMet2 can also support the flows of chemical consumption, sludge and resource recovery for treatment purposes. All this, in turn, provides flexibility for WaterMet2 to simply model various fundamental intervention strategies and thus enables a decision support system to evaluate them over a long-term planning horizon. This kind of flexibility may not be easily achieved through other physically based models due mainly to the extensive and detailed data required for modelling new intervention options and the limited number of calculated performance indicators related to the sustainability framework in those models. These interventions can include either new UWS components for new developments or technological improvements in the UWS components (e.g. energy-efficient or water-efficient equipment). Further details of WateMet2 modelling processes and assumptions can be found in Behzadian et al. (2013) and Behzadian & Kapelan (2015b).

TRUST APPROACH TO WATER SYSTEM SUSTAINABILITY

The TRUST approach to obtaining a better level of sustainability in the UWS can be summarized as follows: (a) assessment of the current sustainability level; (b) setting of the sustainability target at the strategic horizon; (c) definition of a set of suitable interventions to match the targets in agreement with stakeholders expectations; (d) use of the available tools (e.g. metabolism model, risk manager, decision support, etc.) to find the best combination of interventions (roadmap); (e) implementation of the roadmap; (f) evaluation of sustainability results at intermediate time horizons; (g) fine-tuning the interventions to be done; (h) adjusting of the sustainability targets; and (i) selection of additional interventions if needed (Di Federico et al. 2014).

WaterMet2 is used here as a conceptual model and tool which is able to calculate several metrics related to the abovementioned sustainability criteria in the strategic long-term planning horizon within the sustainability framework adopted in the TRUST project (Alegre et al. 2012). Also note that the definition of the sustainability is extended here by including two additional domains, to take into account the assets and the governance characterizing the UWS. In the TRUST perspective, the main goal to be achieved at the strategic horizon is being more sustainable than now. Therefore, all the interventions (structural and non-structural ones) have to be aimed at this purpose. Objectives and metrics are established for a given UWS through a participate process that involves, among others, the water company, water authorities, stakeholders, research actors and several practitioners. The road map process for the UWS of Reggio Emilia and its main outcomes have been described in Di Federico et al. (2014).

CASE STUDY

The sustainability performance assessment of intervention strategies in WaterMet2 is demonstrated in the water supply system of Reggio Emilia city. Reggio Emilia lies in the Po plain in the North of Italy in the western part of the Emilia Romagna region. Its territory has deeply changed in the last decades becoming densely built, with a very high population growth rate due to immigration. It is also home to intensive agricultural practices, livestock farming and a number of medium and small industries, many of them devoted to mechatronic and food transformation. The area is currently facing important challenges, from both a social and an environmental point of view. From the water cycle perspective, the water system operates in a context of financial resources, water and energy scarcity. The application in the Reggio Emilia water system of analysis tools provided by TRUST, as the City profile and the Self-Assessment Tool (Hein et al. 2012; Van Leeuwen et al. 2013), show good hydraulic and water loss control performances. In fact, a very effective water loss control policy, conducted by the water utility since the late 1990s, reduced the real losses to under 10% of the inlet volume and the analysed hydraulic performance indicators for energy consumption, burst events, service interruptions documented the high service level of the system (Iren 2014). On this basis, further reductions of water losses are considered by the water company economically and technically unfeasible. The per-capita water consumption is aligned to the Water Protection Plan of the Emilia Romagna Region recommendations, and water consumption reduction could be achieved only by applying non-technical, but social solutions, i.e. willingness to pay analysis, user behaviour analysis, etc. (Pahl-Wostl 2002; Shove & Walker 2010). Being out of the water utility control, this study does not include social solutions within the alternatives proposed. Hence, only a few limited options are possible for the water company to improve the sustainability performance and thus to cope with external scenarios threatening the UWS, such as the expected reduction of the groundwater resource availability, up to it becoming insufficient to satisfy the forecasted increased water demand. The interventions included in this study relate to the water supply system, more than the water demand management.

Therefore, some interventions aimed at reducing the pressure on the fresh water sources are advisable. The road-mapping process of the TRUST project (Di Federico et al. 2014) was carried out in close collaboration with the Water Company and stakeholders. As a result, they highlighted two main concerns which have been selected by the water company to be addressed here: (1) the reduction of water withdrawals from groundwater and (2) the service reliability/resilience due to the tree shaped water supply system layout. Both of them can be exacerbated by the water scarcity foreseen in the next decades. Additionally, the consistently increasing population projected (Figure 1) by the Italian National Statistics Institute (ISTAT 2011) would lead to a reduction of the water volume per-capita available for consumption. These concerns lead to the following strategic sustainability objectives: (i) to reduce water withdrawals from groundwater; (ii) to improve efficiency of water consumption; (iii) to improve reliability of the water service (Di Federico et al. 2014).
Figure 1

Annual rates of population growth derived for the Emilia Romagna Region (ISTAT 2011).

Figure 1

Annual rates of population growth derived for the Emilia Romagna Region (ISTAT 2011).

According to the water utility, three intervention strategies have been identified as possible alternatives to improve the sustainability of the water supply system: (1) the Business As Usual (BAU) state (intervention strategy 1); (2) the construction of a new pipeline, located in the northern part of the water supply system, which connects an existing water source (R4 in Figure 2) to Reggio Emilia, of the water distribution system to be completed in 3 years (intervention strategy 2); (3) promotion of water consumption in a long-term planning horizon by adding water reuse as an alternative non-traditional water source to allow reduction of the current pressure on groundwater resources (intervention strategy 3). Note that these intervention strategies suggested by the Water Company and experts represent different perspectives for improving the UWS reliability and resilience (Behzadian & Kapelan 2015a). Other intervention strategies can be proposed for this case study, such as rainwater harvesting or grey water reuse but they are derived relative to water demand or supply interventions. Intervention strategy 2 will contribute to the improvement of the service reliability and therefore the system resilience by adding the redundancy of the water sources available for the city. Also note that the newly connected water source is located in an area with redundant water resources and hence it has no impact on the water available in that area. Also, the analysis conducted by WaterMet2 here only considers the functions in the long-term operation (i.e. use) phase of the UWS and the environmental impact of infrastructure activities such as construction, installation and demolition is not analysed due to the insignificant environmental impacts (Behzadian & Kapelan 2015a). The criteria analysed here are those quantitative metrics which can be calculated by WaterMet2. They include technical and environmental objectives including ratio of delivered water demand, energy and GHG emissions over the planning horizon. Other criteria such as social criteria are excluded as they usually need to be quantified by other qualitative methods such as analytical hierarchy analysis.
Figure 2

Sketch of the water supply system of Reggio Emilia represented as SC1 area. Adapted from Di Federico et al. (2014); R = water resource, WTW = water treatment works, SR = service reservoir, SC = sub-catchment, LC = local area.

Figure 2

Sketch of the water supply system of Reggio Emilia represented as SC1 area. Adapted from Di Federico et al. (2014); R = water resource, WTW = water treatment works, SR = service reservoir, SC = sub-catchment, LC = local area.

To demonstrate the capability of the metabolism approach for providing a useful insight into future planning for the decision makers, this paper analyses and compares the sustainability performance of the water supply system for the first two intervention strategies over a 30 year planning horizon.

BUILDING THE WATERMET2 MODEL

The geographical area of the City is represented in WaterMet2 as a collection of SC, each of which is specified by one or more local areas (LC) according with the local physical characteristics of the area and the typology of the water users. The WaterMet2 model was built for the existing water supply system of Reggio Emilia (Figure 2), which includes the city itself (SC1) and the neighbouring sub-catchments Roncocesi (SC2 in the figures) and Rivalta (SC3). Four well fields (R1, R2, R3 and R5) provide water to Reggio Emilia and Rivalta, while R4 currently provides water to Roncocesi (Figure 2). As water abstracted from these water sources has a high level of quality, only disinfection is needed to prepare suitable drinking water. In a few cases, a preliminary filtration process is necessary to remove particle components and minerals (R4 and partially R2). Therefore, water treatment works (WTWs) processes are limited only to these two basic functions. The SC Reggio Emilia, represented in WaterMet2 by two local areas (LC1 and LC4), accommodates totally ∼151,000 inhabitants; Roncocesi has one local area (LC2) with 85,000 equivalent inhabitants and finally Rivalta is linked to one local area (LC3) with 3,500 equivalent inhabitants. As mentioned above, WTW4 does not contribute to feed LC1. For the purposes of this paper, monthly variations of the water demand were considered by means of average monthly coefficients given in Table 1. Note that Italian water systems including the one for Reggio Emilia needs to follow a typical monthly water demand pattern such as the one reported in the table. Hence, we used the monthly peal factors of water demand proposed by Arredi (1990) as typical values for this study. However, accurate site-specific data may be required for a further reliable analysis.

Table 1

Adopted monthly peak factors for water demand for civil usages (Arredi 1990)

MonthJan.Feb.Mar.Apr.MayJun.Jul.Aug.Sep.Oct.Nov.Dec.
Coefficients of monthly variations 0.85 0.85 0.90 0.90 1.00 1.15 1.25 1.25 1.10 1.00 0.90 0.85 
MonthJan.Feb.Mar.Apr.MayJun.Jul.Aug.Sep.Oct.Nov.Dec.
Coefficients of monthly variations 0.85 0.85 0.90 0.90 1.00 1.15 1.25 1.25 1.10 1.00 0.90 0.85 

The real conceptualized scheme of the water supply system can be represented in Figure 3(a) according to a sketch of the case study shown in Figure 2. This conceptual model is represented as a collection of storage/link types in the water supply system, i.e. water resource (R), WTW, SR, SC, local area (LC) and blue and green links between the storage units representing supply conduits, trunk mains and distribution mains. As WaterMet2 only adopts the connection between different element types, and not any connection from one type to each other (i.e. here, SR to SR), this scheme was modified further as shown in Figure 3(b) to adapt to the conceptual model used in WaterMet2. Note that this modification was done such that logical links between source (i.e. water resource) and sink (i.e. SC) are preserved. Following this approach, the physical connection required for delivering water from R4 to LC1 in Reggio Emilia (in intervention strategy 2) is represented in these figures by a green water main (link L12). For each of these links and storages, the energy and cost required per unit volume of delivered water is specified in the model. The energy comprises the total amount of energy required for treating or transferring (e.g. pumping) water between elements. Likewise, the fixed and variable costs, associated with operation and maintenance, are specified in the model. Among the plethora of data necessary to build the metabolism model, the capacity of each element of the system needs to be specified. The water capacity of each element is given in WaterMet2 as the maximum water volume per day which each element is able to transfer/treat/store. These capacities can be specified with the aid of physically based models. In this study, the capacity of elements was estimated by using the physical characteristics and corresponding flow rates calculated by the EPANET model of the case study and allowable flow velocities (Rossman 2000). Thus, capacities of water mains were estimated by assuming the maximum flow velocity of pipeline (1.8 m/s) as the limiting factor and taking into account trunks connecting more than one SR (i.e. L6 in Figure 3(a)). WTWs and water supply conduits were assumed to be large enough to treat the entire water inflow volume. Also no limitation was assumed here for groundwater abstraction over the planning horizon. Finally, the water flows in the built WaterMet2 model are calibrated with the results obtained from the corresponding EPANET model.
Figure 3

Conceptualized scheme of the water supply system for (a) the water system in Figure 1 and (b) further modified for the resulting WaterMet2 model.

Figure 3

Conceptualized scheme of the water supply system for (a) the water system in Figure 1 and (b) further modified for the resulting WaterMet2 model.

Water allocation coefficients between the components in the WaterMet2 model for the BAU (strategy 1) are given in Table 2. In this strategy, SC 1 (Reggio Emilia) receives water from the three upstream service reservoirs (SR1, SR3 and SR6) and hence the relevant water allocation coefficient from SR2 is zero in the table. However, by connecting SR2 to Reggio Emilia in strategy 2, it is assumed that the water main L12 (Figure 3) supplies 25% of the total demand of Reggio Emilia. Hence, the new water allocation coefficients in strategy 2 are: 0.452 for Reggio Emilia from SR1, 0.250 for Reggio Emilia from SR2, 0.226 for Reggio Emilia from SR3 and 0.072 for Reggio Emilia from SR6.

Table 2

Coefficients of water allocation between components in the WaterMet2 model for the BAU

Connected componentsWater allocation coefficientConnected componentsWater allocation coefficientConnected componentsWater allocation coefficient
SR1 from WTW1 0.378 SR3 from WTW3 0.162 SC1 from SR1 0.603 
SR1 from WTW2 0.460 SR4 from WTW1 0.378 SC1 from SR2 0.000 
SR1 from WTW3 0.162 SR4 from WTW2 0.460 SC1 from SR3 0.301 
SR2 from WTW1 0.347 SR4 from WTW3 0.162 SC1 from SR6 0.096 
SR2 from WTW4 0.653 SR5 from WTW4 1.000 SC2 from SR2 0.300 
SR3 from WTW1 0.378 SR6 from WTW5 1.000 SC2 from SR5 0.700 
SR3 from WTW2 0.460   SC3 from SR4 1.000 
Connected componentsWater allocation coefficientConnected componentsWater allocation coefficientConnected componentsWater allocation coefficient
SR1 from WTW1 0.378 SR3 from WTW3 0.162 SC1 from SR1 0.603 
SR1 from WTW2 0.460 SR4 from WTW1 0.378 SC1 from SR2 0.000 
SR1 from WTW3 0.162 SR4 from WTW2 0.460 SC1 from SR3 0.301 
SR2 from WTW1 0.347 SR4 from WTW3 0.162 SC1 from SR6 0.096 
SR2 from WTW4 0.653 SR5 from WTW4 1.000 SC2 from SR2 0.300 
SR3 from WTW1 0.378 SR6 from WTW5 1.000 SC2 from SR5 0.700 
SR3 from WTW2 0.460   SC3 from SR4 1.000 

RESULTS AND DISCUSSION

The built WaterMet2 model was used to simulate the two aforementioned intervention strategies (i.e. strategy 1 and 2) under three scenarios of future population growth depicted in Figure 1 for a 30 year planning horizon. Figure 4 shows some of the calculated performance metrics for the BAU (intervention strategy 1) over the planning horizon. More specifically, variations of total water demand, fraction of water demand delivered, energy and GHG emissions for the entire water system (Reggio Emilia, Roncocesi and Rivalta) are shown in Figure 4(a)4(d), respectively. Note that the leakage is included in the total water demand. Also energy required and corresponding GHG emissions result from consumed electricity, fossil fuel and embedded energy, especially from chemicals used in the UWS components. As can be seen, both energy and GHG emissions increase over the planning horizon according to increasing water demand. It should also be noted that the fluctuations of these indicators for extreme population growth (minimum and maximum) compared to the average values, are up to around 5% of the average values during the later years of the planning horizon.
Figure 4

Performance indicators of the simulation for the BAU (strategy 1) for (a) total water demand, (b) fraction of water demand delivered, (c) total energy and (d) total GHG emissions.

Figure 4

Performance indicators of the simulation for the BAU (strategy 1) for (a) total water demand, (b) fraction of water demand delivered, (c) total energy and (d) total GHG emissions.

Figure 5(a)5(b) show, respectively, the total water demand and fraction of water demand delivered for strategy 2; as expected, both strategy 1 and 2 perform well in terms of total water demand (Figures 4(a) and 5(a)) and for fractional demand delivered (Figures 4(b) and 5(b)). In fact, water resources are considered unlimited here for the two strategies as water abstraction for urban consumption in the case study is far smaller than other water withdrawals from these resources and hence it is less likely to have a high impact on the results. For a more accurate analysis the estimation of the real water availability needs to be introduced in the model.
Figure 5

The total water demand and fractional demand delivered for strategy 2 are shown in (a) and (b), respectively. The differences between strategy 1 and strategy 2 in terms of total energy and total GHG emissions are shown in (c) and (d), respectively.

Figure 5

The total water demand and fractional demand delivered for strategy 2 are shown in (a) and (b), respectively. The differences between strategy 1 and strategy 2 in terms of total energy and total GHG emissions are shown in (c) and (d), respectively.

Figure 5(c)5(d) show the difference over time between strategies 1 and 2 in terms of energy and GHG emissions for the entire water system (Reggio Emilia, Roncocesi and Rivalta). As can be seen, the trend of the discrepancy between the performance indicators of the two strategies is increasing over the planning horizon. Also note that the impact of different population growth scenarios on the total energy is almost negligible, while the same fluctuations for GHG emissions are widening up to 0.5 ton CO2-eq compared to the average values during the later years of the planning horizon.

It is seen that the intervention strategy 2 outperforms the intervention strategy 1 with respect to the two selected metrics (GHG emission and total energy); this is so despite the fact that implementing strategy 2 entails a more powerful pumping station in R4, as the overall hydraulic configuration is energetically more favourable for strategy 2. More specifically, Table 3 reports the cumulative differences between strategy 1 and strategy 2 in terms of three selected performance indicators (GHG emissions, total energy and cost) at some specific time horizons (i.e. year 1, 10, 20 and 30). Positive values indicate the benefit of adopting strategy 2 with respect to 1.

Table 3

Cumulative differences between strategy 1 and strategy 2 in terms of energy, GHG and costs at different time horizons

YearTotal energy (KWh)GHG emissions (Ton CO2-eq.)Total operational cost (€)
30,180 6.3 1,512 
10 346,598 72.8 17,369 
20 649,582 136.4 32,552 
30 1,002,976 210.6 50,261 
YearTotal energy (KWh)GHG emissions (Ton CO2-eq.)Total operational cost (€)
30,180 6.3 1,512 
10 346,598 72.8 17,369 
20 649,582 136.4 32,552 
30 1,002,976 210.6 50,261 

The few metrics calculated by the WaterMet2 model in the water supply system of Reggio Emilia show the capability of this kind of analysis, which can provide useful information for decision makers when analysing a number of different intervention strategies. In other words, having had the conceptual model of the water system along with future water demands and future objectives allows the evaluation of numerous intervention strategies and finally the ranking of them in order to prioritize the most suitable ones.

In this study, two intervention strategies, both compatible with the hydraulic constraints, have been compared with two selected metrics. The results show that strategy 2 weakly outperforms strategy 1 in terms of energy and GHG emissions. Moreover, strategy 2 may be more preferred as it provides an additional water source for the Reggio Emilia sub-catchment (SC1). This issue needs to be further investigated by a comprehensive risk analysis between these two strategies to reveal the effects of redundant capacity provided to the Reggio Emilia SC.

It should be noted that the intervention strategies evaluated here are not for the purpose of hydraulic/mechanical performances only but they can effect all the domains of sustainability. As a result, reduction of operational costs will impact the economic domain, the GHG emissions level will affect the environmental domain, while the improvement of reliability will impact the social aspects. All these obtained impacts on the infrastructure are within the governance domain. The set of differences between the correspondent metrics calculated for intervention strategies in respect to the BAU scenario, provide a measure of the sustainability improvement.

CONCLUSIONS

The paper describes the application of a metabolism model, WaterMet2, to the water supply system of Reggio Emilia. WaterMet2 allows evaluation of the level of sustainability of existing UWS and the impact of possible intervention options; hence it improves understanding of how decisions can contribute to meeting sustainability targets. Furthermore, adopting a comprehensive approach to the UWS, as when applying simplified mass balance analysis, requires the adoption of a number of simplifications for carrying out management steps. However, such an approach turns out to be a powerful tool for long-term strategic analysis and evaluation of intervention options under different scenarios of change. The analysis shown would be difficult to complete using the physically based model (e.g. EPANET) as this is data-intensive/demanding and not necessary at the strategic level of planning. The performance metrics (i.e. selection criteria) analysed are limited only to a number of key technical and environmental metrics that are of paramount importance to the Water Company and the public. Therefore, other criteria such as social impacts are not considered here as these need to be quantified by other tools and techniques outside the WaterMet2 model. The results obtained show that the built and calibrated WaterMet2 model allows a broader understanding of the impacts of alternative intervention strategies taking into account multidimensional aspects of the sustainability beside conventional service performance. This, in turn, can assist water authorities to verify the suitability of selected intervention strategies and to make more informed strategic decisions. Although the results could compare some aspects of the sustainability performance in the two analysed strategies, they cannot, at the current stage of work, be used to make any real decisions. To obtain a robust solution, the current model still needs to be further developed, tested and evaluated for multiple performance indicators including risk type ones as well as different intervention strategies. Finally, incorporating the analysis into a comprehensive Multi Criteria Decision Analysis will improve the comparison of the impact of scenarios of change on more dimensions of sustainability than the few included in this example, and support even better decision makers.

ACKNOWLEDGEMENTS

The authors wish to acknowledge IREN Spa, for providing the data about the Reggio Emilia water supply system; C. Ziveri, P. Pedrazzoli, M. Cingi and F. Ferretti for their kind support and for the fruitful suggestions and discussions about the water supply system functioning and for their review of the paper. This work was carried out as part of the ‘TRansition to Urban water Services of Tomorrow’ (TRUST) project. The authors wish to acknowledge the European Commission for funding the TRUST project in the 7th Framework Programme under Grant Agreement No. 265122.

REFERENCES

REFERENCES
Alegre
H.
Cabrera
E.
Jr
Hein
A.
Brattebø
H.
2012
Framework for Sustainability Assessment of UWCS and development of a self-assessment tool, TRUST, Deliverable D3.1.1, WA 3
.
Arredi
F.
1990
Costruzioni Idrauliche, UTET, Torino
.
Behzadian
K.
Kapelan
Z.
2015b
Advantages of integrated and sustainability based assessment for metabolism based strategic planning of urban water systems
.
Sci. Total Environ.
527
,
220
231
.
Behzadian
K.
Kapelan
Z.
Venkatesh
G.
Brattebø
H.
Sægrov
S.
Rozos
E.
Makropoulos
C.
Ugarelli
R.
Milina
J.
Hem
L.
2013
Urban water system metabolism assessment using WaterMet2 model. In: 12th International Conference on Computing and Control for the Water Industry, CCWI2013, vol. 15
.
Di Federico
V.
Makropoulos
C.
Monteiro
A.
Liserra
T.
Baki
S.
Galvão
A.
2014
The TRUST approach for the Transition to Sustainability of Urban Water Services: the water scarcity cluster
. In:
Intelligent Distribution for Efficient Affordable Supplies
,
Bologna
,
22–24 October 2014
.
Hein
A.
Neskovic
M.
Hochstrat
R.
Smith
H.
2012
Roadmap guideline: A manual to organise transition planning in Urban Water Cycle Systems, TRUST, Deliverable D13.1, WA 1
.
Iren
A. G.
2014
Acquedotti: dati tecnici e risultanze analitiche relative all'anno 2013, Internal Report
.
ISTAT
2011
Previsioni demografiche – Anni 2011-2065 - dati pre-Censimento 2011 di fonte anagrafica. http://dati.istat.it/Index.aspx?DataSetCode=DCIS_PREVDEM
.
Kennedy
C.
Pincetl
S.
Bunje
P.
2011
The study of urban metabolism and its applications to urban planning and design
.
Environ. Pollut.
159
(
8
),
1965
1973
.
Mackay
R.
Last
E.
2010
SWITCH City water balance: a scoping model for integrated urban water management
.
Rev. Environ. Sci. Bio/Technol.
9
(
4
),
291
296
.
Makropoulos
C. K.
Natsis
K.
Liu
S.
Mittas
K.
Butler
D.
2008
Decision support for sustainable option selection in integrated urban water management
.
Environ. Modell. Software
23
(
12
),
1448
1460
.
Mitchell
V. G.
Diaper
C.
2010
UVQ User Manual: (urban water balance and contaminant balance analysis tool), Version 1.2, CMIT Report No. 2005-282. CSIRO
.
Mitchell
V. G.
Mein
R. G.
McMahon
T. A.
2001
Modelling the urban water cycle
.
Environ. Modell. Software
16
(
7
),
615
629
.
Rossman
L. A.
2000
Epanet2 User's Manual
.
US Environmental Protection Agency
,
Cincinnati
,
USA
.
Ugarelli
R.
Almeida
M. C.
Liserra
T.
Smeets
P.
2014a
Metabolism risk-controlled model, demonstrated by model for the city of Oslo, TRUST, Deliverable D32.1, WA3
.
Ugarelli
R.
Almeida
M. C.
Behzadian
K.
Liserra
T.
Smeets
P.
Kapelan
Z.
2014b
Sustainability risk based assessment of the integrated urban water system: a case study of Oslo
. In:
Proc. of the 11th International Conference on Hydroinformatics – HIC 2014
,
New York City
,
USA
.