A procedure for urban water supply system planning under climate variability and deep uncertainty conditions

Procedimento para o planejamento de sistemas de abastecimento de água urbana sob condições de variabilidade climática e incerteza profunda

ABSTRACT

This study developed a structured, easy-to-understand methodology for evaluating urban water supply system alternatives subject to climate variability and deep uncertainties. The water balance in all alternatives was computed from 2021 to 2060 using the Water Evaluation And Planning System (WEAP) platform and incorporating decision making under deep uncertainty (DMDU) techniques. A scenario ensemble of climate conditions and deep uncertainty factors was defined as a state of the world (SOW) for the comparison of performance among the alternatives. The procedure stablished system objectives (minimizing service failure, energy consumption and reservoir failures) and criteria to guide the evaluation among the alternatives. The methodology was applied to the urban water supply system (UWSS) in the Federal District of Brazil, revealing that one specific supply option demonstrated the best overall performance from a multi-objective perspective. However, if the goal is strictly to minimize the intensity of service failures, a different supply configuration emerges as the more effective choice. The methodology allows the selection of the condition that best meet the objectives and priorities of the water utility by choosing specific objective. It was possible to illustrate the tradeoffs that arise from the selection of specific objectives during the evaluation.

Keywords:
Deep uncertainties; Urban water supply system; Decision support system; WEAP

RESUMO

Este estudo desenvolveu uma metodologia estruturada e de fácil compreensão para avaliação de alternativas de sistemas de abastecimento de água urbanos sujeitas à variabilidade climática e incertezas profundas. O balanço hídrico em todas as alternativas foi calculado de 2021 a 2060 usando a plataforma Water Evaluation And Planning System (WEAP), incorporando técnicas de tomada de decisão sob incerteza profunda. Um conjunto de cenários de condições climáticas e fatores de incerteza profunda foi definido como um estado do mundo para a comparação do desempenho entre as alternativas. O procedimento estabeleceu objetivos do sistema (minimização de falhas de atendimento, consumo de energia e falhas de reservatórios) e critérios para a avaliação entre as alternativas. A metodologia foi aplicada ao Sistema de Abastecimento de Àgua Urbano no Distrito Federal do Brasil, revelando que uma alternativa de abastecimento demonstrou um melhor desempenho geral sob uma perspectiva multiobjetivo. No entanto, se o objetivo for apenas minimizar a intensidade das falhas de atendimento, uma configuração de abastecimento diferente se mostra mais eficaz. A metodologia permite a seleção da condição que melhor atende aos objetivos e prioridades da concessionária de água ao escolher um objetivo específico. Foi possível ilustrar as compensações que surgem da seleção de objetivos específicos durante a avaliação.

Palavras-chave:
Incertezas profundas; Sistema urbano de abastecimento de água; Sistema de apoio à decisão; WEAP

INTRODUCTION

The planning of urban water supply systems (UWSS) requires considering the complexity involved in long-term forecasts of climate and other socioeconomic variables a with significant degree of uncertainty. In a long-term horizon, it is hard to estimate the future water availability, nor the growth of water demand and some economic variables such as water tariff, discount rates and operational costs.

Deep uncertainties are defined as those that are difficult to handle and for which there is no clear consensus on their representation (Lempert et al., 2003).

Recently, many studies have evolved toward the incorporation of DU factors into well-characterized uncertainties for the definition of scenarios ensembles (state of the worlds) for which alternative of system planning and managements are tested (Basheer et al., 2021, 2023; Kingsborough et al., 2016).

Adaptive management is a systematic process of continuous improvement of management policies and practices, based on learning from the outcomes of implemented strategies. For this management approach to be successful, it is essential to combine various adaptive actions within complex systems over time. These systems are characterized as networks of components that interact with each other, typically in a non-linear manner. In a self-organized way, these systems evolve in a manner that is neither completely regular nor entirely random (Sayama, 2015).

When it comes to the analysis of multiple scenarios, especially over very long periods, there are parameters that can be extrapolated through hydrological or statistical models, known as well-characterized uncertainties, and others where there is no consensus on their evolution over time, known as deep uncertainties (Lempert et al., 2003). By better understanding the different dimensions of uncertainties and their potential impacts on relevant policy issues, there is a prioritization of research and development activities to support decision-making (Walker et al., 2003). In adaptive planning, as new information is incorporated into the future planning cycle, uncertainty analysis must be updated (Kingsborough et al., 2016).

There are several methods for addressing deep uncertainties in decision-making planning (Marchau et al., 2019). These decisions rely on anticipating changes. Among them, the most notable are Robust Decision Making (RDM), Multiobjective Robust Decision Making (MORDM), Dynamic Adaptive Planning (DAP), Dynamic Adaptive Policy Pathways (DAPP), Info-Gap Decision Theory (IG), and Decision Scaling (DS).

The Decision Scaling method stands out, which, according to Brown et al. (2019), is an effective approach for dealing with uncertainties, particularly climate change. Through simulations of various possible climate scenarios, it allows for the sensitivity assessment of available water systems.

The Corumbá System was incorporated into the UWSS of the Federal District of Brazil (FDB) in April 2022, with the promise of ensuring the region's water security for several years. However, despite this incorporation, questions remain about its optimal integration with the systems already in operation. A more in-depth assessment is necessary to evaluate the objectives and synergies between supply alternatives, which will be able to ensure water security for the region in the best possible way.

The intake of this system is from Lake Corumbá IV, located on the Corumbá River and part of the reservoir system of the Parana River basin (Corumbá Concessões S. A., 2024). The intake point is located about 20 km from the boundaries of the FDB, with an approximate geometric drop of 240 meters.

There are studies on future scenarios for the FDB UWSS involving the concept of deep uncertainties, with the evaluation of optimal moments for short and medium-term actions to ensure the region's water security (Giacomazzo, 2020; Araújo, 2023).

In this context, there arises the possibility of using a multi-objective analysis methodology with the construction of multiple scenarios, where well-characterized uncertainties, such as flow series (Borgomeo et al., 2015), combine with deep uncertainties, which are difficult to predict (Kwakkel et al., 2016). Thus, we can evaluate management portfolios where actions are sequenced, minimizing negative impacts and balancing reservoirs levels with adequate service to the population, aiming to achieve the best long-term results.

This study proposes a methodology to assist in operational planning, enabling better utilization of water resources and existing infrastructures.

METHODOLOGY

The methodology presented is divided into four sequential stages, as shown in Figure 1.

Figure 1
Diagram for the research methodology.

The main ideas of the methodology revolve around in Problem Formulation, where the Service Areas are delineated, and the objectives to be evaluated in the study's results are defined. Following this, the Definition of Supply Alternatives and Scenario Development, the scenarios will be established by defining the uncertainty limits and the supply alternatives. Then, it enables the Simulation in WEAP for processing the results. Subsequently, in Evaluation of the Robustness of Supply Alternatives, the objectives will be assessed using a multi-objective approach, leading to the selection of the most favorable alternative.

CASE STUDY

The proposed methodology was applied to the UWSS of the FDB including the contribution of the Corumbá System which had its initial operation in 2022. The Corumbá System is located in the state of Goiás, outside the FDB borders.

The water supply system of the FDB is comprised of five major interconnected subsystems named the Descoberto, Torto/Maria/Paranoá Sobradinho/Planaltina, São Sebastião, and more recently, the Corumbá System. The population in the FDB is distributed in 35 Administrative Regions (AR), whose physical boundaries define the jurisdiction of government action for administrative decentralization and coordination of public services. These ARs are served by these systems and other isolated subsystems (Brazlândia and rural subsystems).

The Descoberto, Torto/Santa Maria/Paranoá, Sobradinho/Planaltina, and São Sebastião systems are linked by pipelines that transfer treated water among them to supply water to some Administrative Areas based on short term decisions without specific operational rules. These water transfers operate as complementary water source and do not fully meet the demand of any Administrative Area. The most important water transfer among these systems is known as the “Reversible Pipeline,” which, through supply network maneuvers, enables transferring water between the Descoberto and Torto/Santa Maria/Paranoá systems in both directions.

Considering the interconnections among some AR, there is the possibility of being served by more than one UWSS and this choice has different impacts on operational costs. The present study analyses the choices of integration among the main subsystems of the FDB UWSS including the Corumbá System. The largest water treatment plants have been selected to be analyzed together, along with the population that is entirely dependent on them.

Service areas delimitation

The study area was divided in Service Areas defined as a region whose geographical broundaries mostly coincide with the water distribution network limits. The largest and most relevant sources of water for the Service Areas are reservoirs linked to water treatment plants. The Descoberto reservoir is linked to the water treatment plant of Descoberto (ETA.RDE.001). The Santa Maria/Torto reservoir is linked to the water treatment plant of Brasilia (ETA.BSB.001). The Paranoá reservoir is linked to the water treatment plant in Lago Norte (ETA.LNT.001). And the Corumba reservoir is connected to the water treatment plant in Corumba (ETA.COR.001).

Each service area represents the set of administrative regions (ARs) supplied entirely or predominantly by a specific subsystem. The population living in each Administrative Area (Companhia de Planejamento do Distrito Federal, 2022) are the basis for the definition of each total water supply demand in each Service Area. The following Service Areas were defined:

• Service Area 1: AR - Guará;

• Service Area 2: AR - Candangolândia and AR - Núcleo Bandeirante;

• Service Area 3: AR - Recanto das Emas, AR - Gama, AR - Parkway, AR -Santa Maria and AR - Riacho Fundo II;

• Exclusive Service Area 1: AR - Taguatinga, AR - Ceilândia, AR - Samambaia, AR - Riacho Fundo I, AR - Águas Claras, AR - Vicente Pires, AR - Arniqueira and AR - Sol Nascente/Pôr do Sol. These AR included in the Exclusive Service Area 1 are supplied by the Descoberto Subsystem;

• Exclusive Service Area 2: AR - Brasília, AR - Cruzeiro, AR - Lago Sul, AR - Lago Norte, AR - Sudoeste/Octogonal, AR - Varjão, AR - SIA, SCIA, AR - Jardim Botânico, AR - Paranoá, and AR - Itapoã. These RA receive water from Torto/Santa Maria/Paranoá Subsystem;

• Exclusive Service Area 3: AR - Planaltina, AR - Sobradinho I, AR - Sobradinho II and AR - Fercal. These AR get water from the Sobradinho/Planaltina Subsystem.

Figure 2 shows the link representation among the Service Areas and specific Water Treatment Plants:

Figure 2
Integration between service areas and their respective water source and water treatment plants.

The key point of the decision related to the integration among the systems is the selection of Service Areas that would be supplied either by ETA.BSB.001 or ETA.RDE.001, one at a time. There are three water supply alternatives (WSA) to be evaluated upon the simulation of the system as a whole. They are illustrated in Figure 3.

Figure 3
Description of the Water Supply Alternative considered in this study.

As shown in Figure 3, in Water Supply Alternative 1, ETA.BSB.001 supplies Service Area 1, while ETA.RDE.001 supplies Service Area 2. In Water Supply Alternative 2, both Service Areas 1 and 2 are supplied by ETA.RDE.001. In Water Supply Alternative 3, the supply is reversed, with ETA.BSB.001 serving both Service Areas 1 and 2.

The integration of the ETA.COR.001 with the production of the FDB began in 2022, to supplement Service Area 3.

Sampling the state of the world (SOW) - the ensemble of scenarios

In order to evaluate the long run performance of WSA, a model of the water supply system was simulated using an ensemble of multiple scenarios based on different demand growth rates and reduction of network real losses. These parameters represent deep uncertainty factors that have significant influence in the long run planning and management of the system as a whole The inclusion of deep uncertainty factors in the analysis of long run planning and management of water resources systems allows the perception of risks embedded in different decisions.

For each WSA, these deep uncertainty factors were incorporated by setting fixed values within defined limits. The authors combined demand growth rates (1,0%, 1,2%, 1,4%, 1,6%, 1,8% and 2,0%) and network water loss reductions (0%, 2%, 4%, 6%, 8% and 10%), as shown in Table 1.

Table 1
The variability of deep uncertainty factors in the sampling of scenario ensemble.

Scenario	Supply	Demand Growth	Water Loss Reduction
1 - 36	Alternative 1	1% to 2%	0% to 10%
37 - 72	Alternative 2	1% to 2%	0% to 10%
73 - 108	Alternative 3	1% to 2%	0% to 10%

The scenarios of each WSA resulted in 36 combinations of demand growth rate and network water loss reduction. Once there WSA were defined, the total ensemble resulted in 108 scenarios of system configuration. Each scenario was simulated using an ensemble of 1,000 40-year synthetic generated inflows and evaporation series into all reservoirs. Synthetic records were generated using the method proposed by Kirsch et al. (2013) and developed by Araújo (2023) for the FDB. This overall sampling procedure is illustrated in Figure 4.

Figure 4
Representation of the Urban Water Supply System in the Federal District of Brazil and the Sampling of the State of the World – the scenario ensemble.

Demand definition

In order to determine the water demand for each service area, the population survey and the average flow produced between 2017 and 2020 for these regions were related, as presented in Table 2.

Table 2
Water demand of the biggest systems.

	Descoberto System	Torto/ Santa Maria/ Paranoá System
Average Consumption 2017-2022 (L/s)1	3,993.95	2,619.58
Population (2020)2	1,734,144	773,150
Demand (m3/hab/ano)	72.63	106.85

• ¹
  Information provided by the water production department of Companhia de Saneamento Ambiental do Distrito Federal (2023);
• ²
  Companhia de Planejamento do Distrito Federal (2022).

In addition to the large water treatment plants, these systems consist of other smaller treatment units. Since only 4 Water Treatment Plants (WTP) are being evaluated, the population of each service area was reduced in proportion to the contribution of each WTP within its respective production system, reflecting only the population served exclusively by the WTP, as shown in Table 3.

Table 3
Distribution of the Population in the Service Areas.

Service Area	Total Population (2020)	Population Covered (2020)
Exclusive Service Area 1	1,188,855	1,141,301
Exclusive Service Area 2	632,590	588,309
Service Area 1	140,560	130,721
Service Area 2	40,357	38,743
Service Area 3	504,932	484,735

This study considered that an increase in the water demand will proportionally impact the withdrawal in the respective water source (reservoirs).

Production capacity

The production capacity of each WTP was defined based on design project of the plant and limitations determined by water use regulation and permits. Since the WT does not consistently operate at the maximum installed infrastructure capacity, the effective production capacity was calculated based on the historical production records of the units, as shown in Table 4.

Table 4
Effective Capacity of Production.

Water Plant	Production Capacity	k1	Effective Capacity
ETA.RDE.001	4,300L/s	1.1	3,835L/s
ETA.COR.001	1,700L/s	1.1	1,545L/s
ETA.BSB.001	2,800L/s	1.4	1,985L/s
ETA.LNT.001	700L/s	1.6	448L/s

Service Failure Intensity

When simulating each scenario, the intensity of unmet service area demands was quantified annually, referred to as the service failure intensity, to assess the extent to which the demands of the service areas cannot be met. For the entire simulation period, the total demands and the annual volumes not delivered to the service areas were accumulated. Using the ratio between these values, the annual service failure intensity for each scenario was calculated. To compare the behavior of each scenario, the average of these annual objectives was computed, as shown in Equation 1, where smaller values represent better water distribution service to end consumers.

$I_{Failure}=\frac{1}{40}{\displaystyle \sum }_{t=1}^{40}\frac{U{D}_t}{D_t}$ (1)

where, I_Failure is the Service Demand Failure Intensity, UD is the unmet demand in m³, D is the volume demanded by the service areas in m³ and t is the simulation period in years.

Energy consumption

The Energy Consumption objective represents the amount of electrical energy required for water treatment and distribution in each scenario. This objective evaluates the efficiency and sustainability of the UWSS by quantifying the energy usage associated with the operation of the water treatment plants and the pumping stations.

This included energy consumption data from Water Pumping Stations to Water Treatment Plants, as well as units contained within the treatment complexes. The units related to each WTP were grouped to summarize the consumption for each group, enabling a comprehensive analysis of energy usage across the different scenarios.

For the baseline energy and volume produced, the year 2022 was chosen due to the consistency of the electricity bill values and because the Corumbá System became operational in September of that year. Consequently, for Corumbá Plant, only 4 months of data were considered. The year 2022 was selected to get the data of energy and water volume produced in all systems.

The electrical energy consumption, in kWh, and the water volume produced, in m³, by each treatment unit were summed to calculate the consumption-to-production ratio for each water plant, as indicated in Table 5.

Table 5
Relative Energy Consumption per Cubic Meters.

Unit	Volume (m3)	Consumption (kWh)	kWh/m3
ETA.BSB.001	65,032,417.00	43,762,930.00	0.67
ETA.COR.001	4,203,007.14	4,839,927.34	1.15
ETA.LNT.001	17,998,665.54	12,222,554.00	0.68
ETA.RDE.001	127,416,231.35	123,333,680.00	0.97

The equation that combines the energy consumption of each water plant is represented in Equation 2.

$C_{Total}={\displaystyle \sum }_{t=1}^{40}(0.97\times V_{RDE,t}+0.67\times V_{BSB,t}+0.68\times V_{LNT,t}+1.15\times V_{COR,t})$ (2)

where, C_Total is the energy consumption, in kWh, to all the simulation period; V is the volume produced by each water plant, in m³, and t is the simulation time in years.

Reservoir failure

At each timestep of the simulation, as withdrawals are made from the reservoirs, there are changes in the elevation of the reservoirs and consequently changes in the available volume for operation. Therefore, the number of months during the entire simulation period in which the established operational limit was reached was also quantified, as shown in Equations 3 and 4. The function in Equation 3 counts the number of times the reservoir elevation reaches the operational elevation, in each reservoir.

$F_{R_i}={\displaystyle \sum }_{t=1}^{n}I\left({\text{E}}_{R_{i,t}}\le {\text{OE}}_{R_i}\right)$ (3)

$F_{total}={\displaystyle \sum }_{i=1}^m{F}_{R_i}$ (4)

where, E_Ri is the elevation of a reservoir R_i, in meters; OE_Ri is the operational elevation of reservoir R_i, in meters; F_Ri is the number of failures for reservoir R_i; F_Total is the number of failures across all reservoirs; n is the simulation time, in months and i is the reservoir index.

RESULTS AND DISCUSSIONS

The WSAs were compared based on the multi-objective analysis of the results from the WEAP simulation across 108,000 scenarios, which allocated water to service areas over 40 simulated years, presented in sequence.

Multi-objective analysis

With the compilation of results from the three objectives and the establishment of service level ranges, scenarios with the best combined results were selected.

The objectives for each simulation were plotted on a parallel coordinates plot, illustrated in Figure 5, to assess the range and relationships between each of them. Since each scenario was tested under 1,000 different weather conditions, a total of 108,000 simulations were represented in the graph. The aim is to minimize each objective.

Figure 5
Parallel Coordinates Objectives Plot.

The results of each objective can be observed in Figure 6. It allows the visualization of result concentration ranges, mainly differentiated by the service failure intensity and energy consumption. In item (1), (2) and (3) we can observe the results of Alternatives 1, 2 and 3, respectively.

Figure 6
Parallel Coordinates Objectives Plots for each Water Supply Alternative (1, 2 and 3) and for all alternatives that met predefined criteria (4).

In order to evaluate the WSA, one could investigate the number of scenarios each alternative perform in accordance with some predetermined criteria. In this studay, we defined these criteria as the averages of the three objectives as shown in Equations 5, 6 and 7.

$T_{Failure}<\frac{1}{108000}{\displaystyle \sum }_{n=1}^{108000}{I}_{Failure,n}$ (5)

$T_{Energy}<\frac{1}{108000}{\displaystyle \sum }_{n=1}^{108000}{C}_{Total,n}$ (6)

$T_{Reservoir}<\frac{1}{108000}{\displaystyle \sum }_{n=1}^{108000}{F}_{Total,n}$ (7)

where, T_Failure is the threshold for the failure intensity in service; T_Energy is the threshold for energy consumption; T_Reservoir is the threshold for the reservoir failure and n is the number of simulations (see Table 6).

Table 6
Objective threshold of the performance criteria.

Objective Threshold (average)
Service Failure Intensity	Energy Consumption	Reservoir Failure
0.124	8.40x109	3.45

This allowed for the selection of the set of alternatives whose simulations produced objective results below the calculated performance criteria. Item (4) in Figure 6 shows the three alternatives and their corresponding simulation sets that were aligned with the established criteria.

Based on these limits, it was possible to discretize and count the number of simulations that met the established criteria and stratify them for each set of scenarios with the same percentage reduction in losses, as observed in Table 7. The ratio of the number of simulations to the total simulated for each loss reduction range is shown to illustrate the proportion in relation to the whole.

Table 7
Number of Simulations Classified Within the 3 indicators, by water loss reduction.

Water Loss Reduction	Alternative 1		Alternative 2		Alternative 3
	Simulations	Ratio	Simulations	Ratio	Simulations	Ratio
0%	709	2%	0	0%	810	2%
2%	1276	4%	0	0%	1605	4%
4%	1782	5%	0	0%	1680	5%
6%	2244	6%	224	1%	2483	7%
8%	2916	8%	466	1%	2575	7%
10%	3023	8%	875	2%	3399	9%

In Table 8, the discretization was based on the demand growth percentage of reduction.

Table 8
Number of Simulations Classified Within the 3 indicators, by demand growth.

Demand Growth	Alternative 1		Alternative 2		Alternative 3
	Simulations	Ratio	Simulations	Ratio	Simulations	Ratio
1.0%	5066	14%	1135	3%	5403	15%
1.2%	3554	10%	414	1%	4112	11%
1.4%	2255	6%	16	0%	2312	6%
1.6%	1075	3%	0	0%	725	2%
1.8%	0	0%	0	0%	0	0%
2.0%	0	0%	0	0%	0	0%

In all alternatives, with or without loss reduction, supply Alternative 3 seems to be the most favorable according to the criteria listed, with Alternative 1 being the second-best option. Supply Alternative 2 starts appearing from a 6% reduction in real losses.

Now, when the simulation results are subjected only to the Service Failure Intensity criterion, there is a reversal in the favorable ranking of supply alternatives. Looking at the results, Supply Alternative 2 becomes the most favorable, as it presents a Service Failure Intensity objective of 0.103, compared to 0.129 and 0.138 for Alternatives 1 and 3, respectively. Therefore, depending on the goals of the sanitation company, another supply condition may be chosen, emphasizing the choice to reduce the impact on consumers.

By breaking down the results for each Service Area (SA) of each Alternative, it is possible to visualize the behavior of the Service Failure Intensity objective (see Table 9). Table 9 shows that Service Area 1 is the most affected in Supply Alternative 1. In Alternative 2, Exclusive Service Areas (ESA) 1 and 2 have the highest failure rates. In Alternative 3, Service Areas 1 and 2 are the most impacted. A common point in these two conditions is that in Alternatives 1 and 3, SA 1 is exclusively dependent on the ETA.BSB.001 and becomes the most affected SA in these alternatives.

Table 9
Service Failure Intensity for each Service Area.

Service Area	Alternative 1	Alternative 2	Alternative 3
SA1	0.350	0.097	0.373
SA2	0.047	0.094	0.371
SA3	0.047	0.082	0.041
ESA1	0.049	0.097	0.042
ESA2	0.229	0.123	0.250
ESA3	0.227	0.122	0.248

What the results shows in Alternatives 1 and 3 is that ETA.RDE.001 has more capacity than ETA.BSB.001 in serving the set of service areas dependent on each of them. This situation could be different if ETA.COR.001 had not come into operation in 2022.

In Figure 7, each blue dot represents the simulation of a scenario subjected to 1,000 variations of climate series, where the center of the dot represents the 40-year average of the service failure intensity, and the diameter represents the standard deviation of the intensities among the 1,000 climate variation series simulated for each scenario.

Figure 7
Alternatives Service Failure Intensity Diagram.

When visualizing the Service Failure Intensity objective in Figure 7, it shows that a 0.2% reduction in the expected annual demand growth results in a more significant reduction in service failure intensity than a 2% reduction in losses. This means that the impact of reducing demand is more effective than reducing losses.

Another way to assess the behavior of each alternative is by plotting the average annual intensity of supply failure across the set of simulations it encompasses. This approach highlights the most critical and optimistic scenarios, as seen in Figures 8 to 10. Especially in Figure 9, Alternative 2, there is a period where all simulations show no intensity of service failure. This period is between 2022, when ETA.COR.001 began operation, and 2026.

Figure 8
Service Failure Intensity of scenarios in Alternative 1.

Figure 10
Service Failure Intensity of scenarios in Alternative 3.

Figure 9
Service Failure Intensity of scenarios in Alternative 2.

CONCLUSIONS

This paper proposed a methdology to evaluate alternatives of water supply system planning and operation under climate variability and deep uncertainty conditions. The procedure was applied to the water supply system in the Federal District of Brazil assisting in the selection of pre-defined alternatives, according to specific objectives and criteria preferred by the decision-makers.

It was possible to demonstrate the dependency of Service Area 1 on ETA.BSB.001, emphasizing that without the incorporation of any new infrastructure, the system will place great demand on this Water Treatment Plant and should be avoided to reduce the intensities of service failure.

It was also possible to illustrate that the reduction of energy costs has a tradeoff with the reduction in the intensity of service demand failure in the service areas. It was also highlighted the importance of water conservation campaigns, given that water demand reduction had a more substantial effect on the system reliability.

The difference observed in choosing an alternative, depending on the objectives analyzed, reinforces the relevance of the methodology to consider multi-objective evaluation balancing different objectives according to the preference of the decision makers.

ACKNOWLEDGEMENTS

I would like to thank the Environmental Sanitation Company of the Federal District (Caesb) for the opportunity to develop this study through the Graduate Program in Environmental Technology and Water Resources (PTARH) at the University of Brasília (UnB).

REFERENCES

• Araújo, B. M. (2023). Explorando abordagens adaptativas resistentes a incertezas profundas: uma aplicação ao Distrito Federal (Dissertação de mestrado). Departamento de Engenharia Civil e Ambiental, Universidade de Brasília, Brasília, DF.
• Basheer, M., Nechifor, V., Calzadilla, A., Siddig, K., Etichia, M., Whittington, D., Hulme, D., & Harou, J. J. (2021). Collaborative management of the Grand Ethiopian Renaissance Dam increases economic benefits and resilience. Nature Communications, 12(1), 5622. http://doi.org/10.1038/s41467-021-25877-w
  » http://doi.org/10.1038/s41467-021-25877-w
• Basheer, M., Nechifor, V., Calzadilla, A., Gebrechorkos, S., Pritchard, D., Forsythe, N., Gonzalez, J. M., Sheffield, J., Fowler, H. J., & Harou, J. J. (2023). Negotiating Nile infrastructure management should consider climate change uncertainties. Nature Climate Change, 13(1), 1-3. http://doi.org/10.1038/s41558-022-01557-5
  » http://doi.org/10.1038/s41558-022-01557-5
• Borgomeo, E., Farmer, C. L., & Hall, J. W. (2015). Numerical rivers: a synthetic streamflow generator for water resources vulnerability assessments. Water Resources Research, 51(7), 5382-5405. http://doi.org/10.1002/2014WR016827
  » http://doi.org/10.1002/2014WR016827
• Brown, C., Steinschneider, S., Ray, P., Wi, S., Basdekas, L., & Yates, D. (2019). Decision Scaling (DS): decision support for climate change. In V. Marchau, W. Walker, P. Bloemen & S. Popper (Eds.), Decision making under deep uncertainty: from theory to practice (pp. 255-287). Cham: Springer. http://doi.org/10.1007/978-3-030-05252-2_12
  » http://doi.org/10.1007/978-3-030-05252-2_12
• Corumbá Concessões S. A. – CCSA. (2024). Informações técnicas. Retrieved in 2024, June 29, from https://www.corumbaconcessoes.com.br/uhe-corumba-iv/informacoes-tecnicas/
  » https://www.corumbaconcessoes.com.br/uhe-corumba-iv/informacoes-tecnicas/
• Companhia de Saneamento Ambiental do Distrito Federal – Caesb. (2023). Águas Claras, DF. Retrieved in 2024, June 29, from https://www.caesb.df.gov.br/
  » https://www.caesb.df.gov.br/
• Companhia de Planejamento do Distrito Federal – CODEPLAN. (2022). Retrieved in 2024, June 29, from https://www.codeplan.df.gov.br/
  » https://www.codeplan.df.gov.br/
• Giacomazzo, A. P. (2020). Análise de portfólios de planejamento e gestão do sistema de abastecimento urbano do Distrito Federal no contexto de incertezas profundas (Dissertação de mestrado). Departamento de Engenharia Civil e Ambiental, Universidade de Brasília, Brasília, DF.
• Kingsborough, A., Borgomeo, E., & Hall, J. W. (2016). Adaptation pathways in practice: mapping options and trade-offs for London’s water resources. Sustainable Cities and Society, 27, 386-397. http://doi.org/10.1016/j.scs.2016.08.013
  » http://doi.org/10.1016/j.scs.2016.08.013
• Kirsch, B. R., Characklis, G. W., & Zeff, H. B. (2013). Evaluating the impact of alternative hydro-climate scenarios on transfer agreements: practical improvement for generating synthetic streamflows. Journal of Water Resources Planning and Management, 139(4), 396-406.
• Kwakkel, J. H., Walker, W. E., & Haasnoot, M. (2016). Coping with the wickedness of public policy problems: approaches for decision making under deep uncertainty. Journal of Water Resources Planning and Management, 142(3), 1-5. http://doi.org/10.1061/(ASCE)WR.1943-5452.0000626
  » http://doi.org/10.1061/(ASCE)WR.1943-5452.0000626
• Lempert, R. J., Popper, S. W., & Bankes, S. C. (2003). Shaping the next one hundred years: new methods for quantitative, long-term policy analysis. Santa Monica, CA: RAND Corporation. http://doi.org/10.7249/MR1626
  » http://doi.org/10.7249/MR1626
• Marchau, V. A. W. J., Walker, W. E., Bloemen, P. J. T. M., & Popper, S. W. (2019). Decision making under deep uncertainty Cham: Springer. http://doi.org/10.1007/978-3-030-05252-2
  » http://doi.org/10.1007/978-3-030-05252-2
• Sayama, H. (2015). Introduction to the modeling and analysis of complex systems Geneseo: Open SUNY Textbooks.
• Walker, W. E., Harremoës, P., Rotmans, J., van der Sluijs, J. P., van Asselt, M. B. A., Janssen, P., & Krayer von Krauss, M. P. (2003). Defining uncertainty: A conceptual basis for uncertainty management in modelbased decision support. Integrated Assessment, 4(1), 5-17. http://doi.org/10.1076/iaij.4.1.5.16466
  » http://doi.org/10.1076/iaij.4.1.5.16466

Edited by

• Editor-in-Chief:
  Adilson Pinheiro

• Associated Editor:
  Iran Eduardo Lima Neto

Publication Dates

• Publication in this collection
  17 Feb 2025
• Date of issue
  2025

History

• Received
  23 Aug 2024
• Reviewed
  19 Oct 2024
• Accepted
  19 Nov 2024