MARIA VITÓRIA DA SILVA GARCIA 

 
 

 

IMPACTO DAS CONCEPÇÕES DO SISTEMA DE DRENAGEM NO 

METABOLISMO URBANO 

Impact of drainage system concepts in urban metabolism 

 

 

 

 

 

 

 

 

 

 
 

Bauru 
2023  



 

 

 

 

MARIA VITÓRIA DA SILVA GARCIA 

 
 

 

 

 IMPACTO DAS CONCEPÇÕES DO SISTEMA DE DRENAGEM NO 

METABOLISMO URBANO 

 

 

 

 

Texto de defesa de dissertação apresentada 

como requisito para a obtenção do título de 

Mestre em Engenharia Civil e Ambiental da 

Universidade Estadual Paulista “Júlio de 
Mesquita Filho”, Área de Concentração 
Saneamento. 

 

 

Orientador: Prof. Dr. Rodrigo Braga Moruzzi 

Coorientador: Prof. Dr. Kourosh Behzadian 

 

 

 

 

 
 
 

Bauru 
2023  



G216i

Garcia, Maria Vitória da Silva

    Impact of drainage system concepts in urban metabolism / Maria

Vitória da Silva Garcia. -- Bauru, 2023

    90 p. : il., tabs., fotos, mapas

    Dissertação (mestrado) - Universidade Estadual Paulista (Unesp),

Faculdade de Engenharia, Bauru

    Orientadora: Rodrigo Braga Moruzzi

    Coorientadora: Kourosh Behzadian

    1. sistema metabólico. 2. drenagem urbana. 3. tomada de decisão. 4.

SUDS. 5. performance. I. Título.

Sistema de geração automática de fichas catalográficas da Unesp. Biblioteca da Faculdade de

Engenharia, Bauru. Dados fornecidos pelo autor(a).

Essa ficha não pode ser modificada.



UNIVERSIDADE ESTADUAL PAULISTA

Câmpus de Bauru

ATA DA DEFESA PÚBLICA DA DISSERTAÇÃO DE MESTRADO DE MARIA VITÓRIA DA SILVA
GARCIA, DISCENTE DO PROGRAMA DE PÓS-GRADUAÇÃO EM ENGENHARIA CIVIL E
AMBIENTAL, DA FACULDADE DE ENGENHARIA - CÂMPUS DE BAURU.

Aos 14 dias do mês de abril do ano de 2023, às 09:00 horas, por meio de Videoconferência, realizou-

se a defesa de DISSERTAÇÃO DE MESTRADO de MARIA VITÓRIA DA SILVA GARCIA, intitulada

IMPACT OF DRAINAGE SYSTEM CONCEPTS IN URBAN METABOLISM. A Comissão

Examinadora foi constituida pelos seguintes membros: Prof. Dr. RODRIGO BRAGA MORUZZI

(Orientador(a) - Participação Virtual) do(a) Departamento de Engenharia Civil e Ambiental / FEB -

UNESP Bauru/SP, Prof. Dr. FABIANO TOMAZINI DA CONCEIÇÃO (Participação Virtual) do(a) IGCE /

UNESP/Rio Claro (SP), Prof. Dr. ALEXANDRE SILVEIRA (Participação Virtual) do(a) Instituto de

Ciência e Engenharia Ambiental / UNIVERSIDADE FEDERAL DE ALFENAS. Após a exposição pela

mestranda e arguição pelos membros da Comissão Examinadora que participaram do ato, de forma

presencial e/ou virtual, a discente recebeu o conceito final:_ _ _ _ _ _ _ _ _ _ _ _ _ . Nada mais

havendo, foi lavrada a presente ata, que após lida e aprovada, foi assinada pelo(a) Presidente(a) da

Comissão Examinadora.

Prof. Dr. RODRIGO BRAGA MORUZZI

Faculdade de Engenharia - Câmpus de Bauru -
Eng. Luiz Edmundo Carrijo Coube, 14-01, 17033360, Bauru - São Paulo

http://www.feb.unesp.br/posgrad_civilCNPJ: 48.031.918/0030-69.

APROVADO



 

 

 

 

MARIA VITÓRIA DA SILVA GARCIA 

 
 

 

 Impact of drainage system concepts in urban metabolism 

 

 

 

 

 

Dissertation defence presented as a 

requirement for obtaining the master’s in civil 
and environmental Engineering title at 

Universidade Estadual Paulista “Júlio de 
Mesquita Filho”, Concentration Area: 
Sanitation. 

 

 

Advisory: Prof. Ph.D. Rodrigo Braga Moruzzi 

Co-supervisor: Prof. Ph.D. Kourosh Behzadian 

 

 

 

 

 

Bauru 
2023 

  



 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DEDICATÓRIA 

Dedico esse trabalho aos meus pais Rosana e Claudinei (In memoriam) 

que sempre me apoiaram, com muito amor e gratidão.  



AGRADECIMENTOS 

Agradeço, primeiramente ao meu orientador Professor Rodrigo Braga 

Moruzzi, pelo apoio, conselhos e ensinamentos, sem os quais este trabalho 

seria impossível. 

Ao meu coorientador Professor Kourosh Behzadian, o qual aceitou 

também me ajudar e apoiar nesta jornada. 

Á Coordenação de aperfeiçoamento de Pessoal de Nível Superior (CAPES) 

pela concessão de bolsa a qual viabilizou este trabalho. 

Á minha família e amigos pelo apoio e compreensão. 

Aos meus companheiros de grupo de pesquisa, os quais trocamos 

experiências e juntos prestaram suporte e apoio durante o tempo de 

pesquisa. 

Á minha banca que juntos contribuíram para este trabalho. 

Á todo o departamento de Pós-graduação de Engenharia Civil e Ambiental 

da UNESP, pela oportunidade e confiança. 

E para todos aqueles que de alguma forma contribuíram para esta 

conquista, que mesmo sem estarem citados aqui merecem minha gratidão 

e meu agradecimento. 

Muito obrigada! 

 

 

 



i 
 

 

 

 

Resumo 

As mudanças climáticas estão afetando significativamente o meio ambiente, e 

atualmente são necessárias mudanças seja em comportamento ou ações governamentais, 

para reduzir esses impactos. Então paralelamente a isso é necessário a implantação de 

medidas em diversas áreas, inclusive e de extrema importância no planejamento, 

implantação e operação de águas urbanas, de maneira que durante essas fases sejam 

considerados o sistema como um todo. O que é explicado através do metabolismo da água 

que apresenta a transformação que ocorre com as águas no ambiente urbano e suas demandas 

e emissões. Nesse processo pode ser considerado a demanda de água de precipitação, 

superficial e subterrânea, a transformação, geração e transporte de efluentes, gases de efeito 

estufa, poluentes e outros. Indicadores os quais podem ser mensurados e considerado através 

de modelos específicos de metabolismo, como o Watermet². Os sistemas de drenagem 

urbana atualmente são comumente projetados conforme sua performance em determinados 

índices, seja em sua eficiência quanto seus impactos, porém estes últimos indicadores são 

escolhidos de diversas formas. Este trabalho tem o foco em mostrar como a performance do 

sistema de drenagem urbana impacta nos processos metabólicos urbanos, ou seja, por meio 

da análise da eficiência em função de um objetivo ou redução de impactos, à luz do fluxo 

natural. Para isso foi utilizado o software Watermet² onde após um teste de sensibilidade 

foram inseridos dados com o intuito de simular os diversos sistemas de drenagem urbana 

sustentável, o qual mostrou que praticamente todos as variáveis inseridas são importantes 

para o sistema. Revela-se que este software (Watermet²) pode ser aplicado pelos tomadores 

de decisão com o objetivo de promover a gestão e adaptação para um sistema de drenagem 

integrado, com baixo impacto em todo o sistema e melhorar a resiliência urbana às mudanças 

climáticas, pois este demonstrou-se ser sensível as alterações de concepções de tipos 

drenagem urbana.. Assim como também este software se mostra capaz de mensurar os 

indicadores de performance metabólica do sistema de maneira direta, tornando-o ideal para 

auxiliar juntamente a outras ferramentas de projeto. Com o software e o processo é possível 

uma padronização dos fundamentos e consequentemente os indicadores-chave de 

desempenho em uma análise de performance específica de drenagem urbana. Com este é 

possível também a determinação de uma melhor hipótese dentro das alternativas previstas, 

sendo neste estudo de caso e com os pesos determinados concluído como o pavimento 

permeável. 

 

Palavras-chave: sistema metabólico, drenagem urbana, tomada de decisão, SUDS, 

performance.  



ii 
 

 

 

 

 

 

Abstract 

 

Climate change is significantly affecting the environment, and we currently find 

ourselves in a time where changes are needed, whether in behaviour or governmental actions 

to reduce these impacts. So, in parallel, it is necessary to implement measures in diverse 

areas, including and of extreme importance in the planning, implementation, and operation 

of urban waters, so that during these phases the system is considered. What is explained 

through the water metabolism, presents the transformation that occurs with the urban 

environment waters and its demands and emissions. In this process, the demand for 

precipitation, surface and underground water, and the transformation, generation and 

transport of effluents, greenhouse gases, pollutants and others can be considered. These 

indicators can be measured and considered through specific models of metabolism, such as 

Watermet². Currently, urban drainage systems are commonly designed according to their 

performance in specific indices, whether in terms of their efficiency or their impacts. Still, 

these last indicators are chosen in unusual ways. This work focuses on showing how the 

performance of the urban drainage system impacts in the urban metabolic processes, that is, 

through the analysis of efficiency in terms of an objective or reduction of impacts, in light 

of the natural flow. For this, the Watermet² software was used, where, after a sensitivity test, 

the data was inserted to simulate the various sustainable urban drainage systems, which 

showed that all the inserted variables are important for the process. It is revealed that this 

software (Watermet²) can be applied by decision makers to promote management and 

adaptation to an integrated drainage system, with low impact on the entire system and that 

improves urban resilience to climate change, as this proved to be sensitive to conceptions 

changes in urban drainage types. As well as this software can measure the metabolic 

performance indicators of the system directly, making it ideal to help along with other design 

tools. With the software and the process, it is possible to standardize the fundamentals and, 

consequently, the key performance indicators in a specific performance analysis of urban 

drainage. With this method it is also possible to determine a better hypothesis among the 

alternatives, being in this case study and with the determined weights concluded as the 

permeable pavement. 

 

Keywords: metabolic system, urban drainage, decision making, SUDS, performance. 

 



iii 
 

 

 

 

 

 

Graphical Abstract 

 

 
  



iv 
 

 

Figure Index 

Figure 1 – Number of papers per year by subject about Metabolism and Urban water 

systems. ................................................................................................................................. 7 

Figure 2 - Common flux and hierarchy in urban metabolism. .............................................. 8 

Figure 3 - Main process of Watermet² software. ................................................................. 10 

Figure 4 - Main flows and storage in the software. ............................................................. 11 

Figure 5 - SUDS pillars. ...................................................................................................... 13 

Figure 6 - Decision making processes. ................................................................................ 13 

Figure 7 – Research summary. ............................................................................................ 15 

Figure 8 - Sensitivity test scheme. ....................................................................................... 18 

Figure 9 - Catchment localization map at São Paulo State.................................................. 24 

Figure 10 - Available public areas. ...................................................................................... 29 

Figure 11 – Example of wetland hypothesis. ...................................................................... 30 

Figure 12- Dry ponds scheme. ............................................................................................. 35 

Figure 13 - SA: Sewer System Balance in UWS. ............................................................... 40 

Figure 14 - SA: Energy in the UWS. ................................................................................... 40 

Figure 15 - SA: GHG emissions in the UWS. ..................................................................... 41 

Figure 16 - SA: Acidification in UWS. ............................................................................... 41 

Figure 17 - SA: Eutrophication in UWS. ............................................................................ 42 

Figure 18 - SA: Contaminant Load in UWS. ...................................................................... 42 

Figure 19 - SA: Sludge generation in the UWS. ................................................................. 43 

Figure 20 - Accumulated dimensionless impacts. ............................................................... 48 

Figure 21 - Subcatchment detailing. .................................................................................... 63 

Figure 22 – Water system scheme. ...................................................................................... 64 

Figure 23 – Study case form: Water supply topology. ........................................................ 68 

Figure 24 – Study case form: Water supply operation. ....................................................... 68 

Figure 25 – Study case form: Water supply Assets. ............................................................ 69 

Figure 26 - Study case form: Subcatchment topology. ....................................................... 70 

Figure 27 - Study case form: Subcatchment Specifications. ............................................... 70 

Figure 28 – Study case and conventional system: Subcatchment form assets in each local 

area....................................................................................................................................... 71 

Figure 29 - Study case form: Wastewater topology. ........................................................... 72 

Figure 30 - Study case form: Wastewater operation. .......................................................... 72 

Figure 31 - Study case and conventional system form: Subcatchment Assets. ................... 73 



v 
 

 

 

Table index 

Table 1 - Files database. ...................................................................................................... 16 

Table 2 - Weather Database................................................................................................. 17 

Table 3 – City data for Sensitivity analysis. ........................................................................ 19 

Table 4 – Input time series for Sensitivity analysis. ............................................................ 19 

Table 5 – Subcatchment parameters for Sensitivity analysis. ............................................. 20 

Table 6 – Input parameters - Water supply system for Sensitivity analysis........................ 20 

Table 7 – Input Wastewater parameters for Sensitivity analysis......................................... 21 

Table 8 – Input Water resource recovery parameters for Sensitivity analysis. ................... 22 

Table 9 – Input parameters Water supply system for Study case analysis. ......................... 25 

Table 10 – Subcatchment parameters for Study case analysis. ........................................... 27 

Table 11 – Input Wastewater parameters for Study case analysis. ..................................... 27 

Table 12 - Hypothesis nomenclature. .................................................................................. 28 

Table 13 – Constructed wetland data. ................................................................................. 31 

Table 14 – Storage tanks and rain barrels data. ................................................................... 32 

Table 15 - Infiltration trenches data. ................................................................................... 33 

Table 16 - Permeable pavement data. .................................................................................. 34 

Table 17 – Dry ponds data. .................................................................................................. 35 

Table 18 - Criteria weights. ................................................................................................. 36 

Table 19 - Constant results of KPIs in UWS. ...................................................................... 39 

Table 20 - Percentual of the principal components of the sensitivity data. ......................... 44 

Table 21 - Bauru catchment results. .................................................................................... 45 

Table 22 - Ranking of the hypotheses. ................................................................................ 49 

Table 23 – Sensitivity Results Simulations ......................................................................... 65 

Table 24 - Principal components variables.......................................................................... 66 

Table 25 – Scores of Principal Components in the KPIs. ................................................... 67 

Table 26 – Resume of Hypothesis implementation data for the influence variables. ......... 74 

Table 27 - Comparation alternatives resume. ...................................................................... 76 

Table 28 - Comparation alternatives resume. ...................................................................... 77 

 
  



vi 
 

 

 

List of Acronyms and Abbreviations 

COD  Chemical oxygen demand; 

CSO  Combined sewer overflow; 

DM   Distribution Mains; 

DSC   Drainage System Conduits;  

GHG  Greenhouse gas; 

IUWM Integrated urban water management; 

KPIs   Key performance indicators; 

PCA  Principal component analysis; 

PMB   Bauru City Hall; 

SA   Sensitivity analysis; 

SC   Subcatchment; 

StCa  Study Case; 

SCa   Water Supply Conduits; 

STO  Storm tank overflow; 

SUDs  Sustainable urban drainage systems; 

TM   Trunk Mains;  

UDM  Urban drainage models; 

UWS   Urban water system; 

WC   Waste Conduits; 

WTW   Water Treatment Works; 

WWTWs  Wastewater Treatment Works. 

 

 

  



vii 
 

 

 

 

 

Table of contents 

 

RESUMO .................................................................................................................................................... I 

ABSTRACT ............................................................................................................................................... II 

GRAPHICAL ABSTRACT ...................................................................................................................... III 

FIGURE INDEX ...................................................................................................................................... IV 

TABLE INDEX .......................................................................................................................................... V 

LIST OF ACRONYMS AND ABBREVIATIONS ................................................................................. VI 

TABLE OF CONTENTS ......................................................................................................................... VII 

1 INTRODUCTION ............................................................................................................................ 1 

2 OBJECTIVES .................................................................................................................................. 4 

2.1 SPECIFIC OBJECTIVES ................................................................................................................... 4 

3 LITERATURE REVIEW ................................................................................................................ 5 

3.1 URBAN WATER SYSTEM ............................................................................................................... 6 

3.2 URBAN METABOLISM .................................................................................................................... 7 

3.3 WATERMET² ..................................................................................................................................... 9 

3.3.1 Urban drainage ....................................................................................................................... 12 

4 JUSTIFICATION AND HYPHOTESIS ........................................................................................ 14 

5 MATERIALS AND METHODS .................................................................................................... 15 

5.1 MATERIALS .................................................................................................................................... 16 

5.1.1 Spatial data .............................................................................................................................. 16 

5.1.2 Metabolism software: WaterMet² ............................................................................................ 16 

5.2 METHOD.......................................................................................................................................... 18 

5.2.1 Sensitivity test .......................................................................................................................... 18 

5.2.2 Case of study ............................................................................................................................ 24 

5.2.3 Hypotheses/intervention options .............................................................................................. 28 

5.2.4 Watermet² Processing .............................................................................................................. 38 

6 RESULTS AND DISCUSSION ...................................................................................................... 39 

6.1 SENSITIVITY RESULTS ................................................................................................................ 39 



viii 
 

 

6.1.1 Normalized results ................................................................................................................... 39 

6.1.2 Principal component analysis .................................................................................................. 43 

6.2 STUDY CASE RESULTS ................................................................................................................ 44 

6.2.1 Bauru catchment Results ......................................................................................................... 44 

6.3 INTERVENTION RESULTS ........................................................................................................... 47 

6.3.1 Individual intervention Results ................................................................................................ 47 

6.3.2 Multivariate analysis ............................................................................................................... 48 

7 FINAL CONSIDERATIONS ......................................................................................................... 50 

8 RECOMMENDATIONS ................................................................................................................ 52 

9 REFERENCES ............................................................................................................................... 53 

APPENDIX A .......................................................................................................................................... 63 

APPENDIX B .......................................................................................................................................... 64 

APPENDIX C .......................................................................................................................................... 65 

APPENDIX D .......................................................................................................................................... 66 

APPENDIX E .......................................................................................................................................... 67 

APPENDIX F .......................................................................................................................................... 68 

APPENDIX G .......................................................................................................................................... 70 

APPENDIX H .......................................................................................................................................... 71 

APPENDIX I ........................................................................................................................................... 72 

APPENDIX J ........................................................................................................................................... 74 

APPENDIX K .......................................................................................................................................... 76 

APPENDIX L .......................................................................................................................................... 77 

 

 



1 
 

 

1 INTRODUCTION 

 

This work will focus on Urban and Regional Infrastructure studies and knowledge, 

inserted in the Environmental Sanitation great area. A holistic environmental analysis is 

fundamental when we are working with urban water systems, which allows for measuring 

the various external factors, strategies, or interventions that influence the conditions of the 

general system (VENKATESH et al., 2017). 

Although analysing the urban water system is the main objective, looking from 

different perspectives is necessary, because a massive use of resources is necessary in order 

to make changes to the set that can negatively impact the environment; with this becomes 

the needing to apply sustainability in long-term multi-objective (BEHZADIAN; KAPELAN, 

2015b). Which is also defended by The United Nations as essential for sustainable 

development and for an accelerated adaptation to climate change, applied as parks and urban 

agriculture to reduce floods and heat islands, for example (IPCC, 2022). 

Climate change has a significant impact on hydric resources, which can be seen in 

the water problems like overflows, inundations, and flooding episodes (UNESCO, 2020; 

KOURTIS; TSIHRINTZIS, 2021; ARNBJERG-NIELSEN; LEONARDSEN; MADSEN, 

2015). This can be a consequence that originates from many factors such as land use, urban 

growth, and the demand for water use, which highlights the need for reliable resource 

management practices. An example of these changes can be saw in Brazil between 2001 and 

2017, a period that suffered from droughts, floods, storms, and landslides, impacting 64.9 

million people, and causing US$ 34.6 billion in damages (UNESCO, 2020). 

With climate change increasing concern, the use of drainage systems to alleviate 

this problem is becoming more evident (KOURTIS; TSIHRINTZIS, 2021). Sustainable 

solutions are the main alternatives to solve these problems and promote urban resilience 

(ZHANG et al., 2021;KOURTIS; TSIHRINTZIS, 2021; LA ROSA; PAPPALARDO, 2020; 



2 
 

 

MCCLYMONT et al., 2020; ZISCHG et al., 2019). Integrating existing conventional 

systems installed in most cities with these sustainable devices is the best solution to solving 

these problems, bringing multifunctionality to the system (BROWDER et al., 2019; YANG; 

ZHANG, 2021); since “grey buildings” play a major role in meeting the needs of the 

population while contributing to climate change, the opposite of “green buildings” that can 

meet these demands while simultaneously bringing numerous benefits (BROWDER et al., 

2019).  

Different models, projections, risk analyses, and other statistical methods is 

commonly used to project the urban drainage devices, which can result in different 

conclusions, thus contributing to the influence and impact on the environment analysis. The 

use of the usual systems is not possible anymore, since an automatic adaptation to the new 

methods already occurs by the community (ARNBJERG-NIELSEN; LEONARDSEN; 

MADSEN, 2015); and for that, an unusual approach to performance analysis is necessary, 

aiming at the process, being necessary this holistic view to consider different paths  

(BEHZADIAN; KAPELAN, 2015b). 

For this reason, a dynamic metabolism model was developed, a sustainability 

assessment tool in the urban water department, that takes into account the changes in the 

environment caused by urbanism and its impact on several fronts such as water quality, 

adaptation and mitigation of climate changes, environmental life cycle assessment, and how 

to improve their operations within the water-energy-carbon nexus (VENKATESH; 

SÆGROV; BRATTEBØ, 2014). 

For changes in these aspects to be conducted, improvements are needed where 

stakeholders give more importance, which is related to economic, social, and sustainable 

dimensions; those which are required to improve performance (VENKATESH; SÆGROV; 

BRATTEBØ, 2014). 

Therefore, it is essential innovative studies related to the planning process, aiming 

to help the decision-maker with a focus on improving performance and standardizing the 

process simultaneously, to reduce the dependence on conservative methods. 

In the Brazilian scenario, few studies address issues related to the system multi-

objective performance, being found in works such as: (NOVAES, 2016) and (MENDONÇA; 

SOUZA, 2019), highlighting the need to deepen the issue in a national context. 



3 
 

 

Several pieces of research have been made in the international environment based 

on the decision between sustainable methods based on the performance of the system against 

different indicators, such as those carried out by YANG; ZHANG, (2021);  DOS SANTOS 

et al., (2021); SEYEDASHRAF; BOTTACIN-BUSOLIN; HAROU, (2021); 

MCCLYMONT et al., (2020); (LI, 2020); FOOMANI; MALEKMOHAMMADI, (2020); 

LA ROSA; PAPPALARDO, (2020); (ARIZA et al., 2019), among many others. These 

works focus on the system's performance, but the choice for these indicators is usually 

random, thus evidencing a gap in the consolidation of methods that include and delimit all 

these indicators.  

With this context and the urban drainage systems importance on diverse fronts, this 

work aims to analyse a fundament to complement the decision process during the planning 

phase of the systems or even their adaptation. This complementation is based on the 

processes of transformation and urban impact on the environment, the urban metabolism; so 

that it presents itself as a foundation in the performance analysis to include and support the 

various indicators commonly used. 

The next sections are structured so that the Objectives (2) of the research can be 

presented, in a General way and each stage of the Specific Objectives (2.1), followed by a 

Bibliographic Review (3) to present the General Context of the problem, as well as its 

Justification and Hypothesis (4), followed by the Materials and Methods (5) used for 

research, where the physical or digital Materials (5.1) and the Methods (5.2) applied to 

deepen the question are presented. Subsequently, the Results (6) obtained during the 

Sensitivity Analysis (5.2.1), System Influence Analysis (6.2) and their aforementioned 

Discussions will be presented. Therefore, Conclusions and Recommendations (7) are 

presented for this and will also be future studies. 

  



4 
 

 

 

2 OBJECTIVES 

The fundamental objective of this work is to analyse the impact of the different 

concepts/types of urban drainage systems in the urban water metabolism performance 

indicators and show that the fundaments, interactions, and techniques of urban metabolism 

can be used for differentiating and choose between them.  

2.1 SPECIFIC OBJECTIVES  

▪ Perform a sensitivity analysis for the variables that govern the process of urban 

metabolism. 

▪ Map the metabolism characteristics of study object. 

▪ Propose and evaluate drainage intervention options impacts over urban water 

metabolism. 

 

 

  



5 
 

 

 

3 LITERATURE REVIEW 

Climate change is already significantly affecting our environment, with periods of 

intense heat and cold, melting ice and affecting sea levels, among other countless impacts. 

As happened during the Covid-19 pandemic, where a large drop in Greenhouse gas (GHG) 

emissions was observed, showing a great dependence on fossil fuels, it was also possible to 

observe where efforts can be made to mitigate these impacts (WORLD ECONOMIC 

FORUM, 2022). 

The time for change is now, we have the tools and how to apply them, policies and 

regulations are being effective, reinforcing the attention that must be taken in this area. Only 

with this adaptation and with effective mitigation measures will sustainable development be 

possible (IPCC, 2022). 

These adaptations, when possible, can be carried out in different areas, and how the 

climate change causes even greater risks in the water sector, alternative measures must be 

applied simultaneously with conventional ones, thus having multiple fronts, such as hard 

approaches and soft approaches, thus helping in an efficient planning of cities. Climate 

change will significantly affect water, in its quantity, quality or even availability. Which will 

cause great difficulties for their management, significantly affecting access to sanitation 

(UNESCO, 2020). Therefore, efforts must be made in the planning area do try minimizing 

the impacts.  

An integrated analysis of the urban water systems management has many benefits. 

Nowadays there is a great diversity of available tools that can show the efficiency of the used 

methods, which are usually chosen according to the decision maker's objectives, making 

difficult the standardization of the impacts and their respective controls metrics (MOSLEH; 

NEGAHBAN-AZAR, 2021). 



6 
 

 

3.1 URBAN WATER SYSTEM 

Currently, a concept that is popularized among administrators and decision makers 

is the "Integrated urban water management" (IUWM), which analyses impacts on indicators 

considering the totality of urban water systems, applying integrated solutions, which 

consequently require integrated analyses (MOSLEH; NEGAHBAN-AZAR, 2021). 

In international context, there is many works with performance/integrated analysis 

in urban water systems as a subject, and if we pic the system performance as a specific topic 

there is a small number, and applying this to metabolism there is much less. 

According to research made in Scopus database, we can see the number of papers 

and the gap in the Metabolism and the urban water system area, even in their subtypes as the 

urban drainage system, keeping this in mind a new research focus in Metabolism and Urban 

Water Systems keywords, this time in Scopus, Science Direct and Web of Science database 

between the years of 2012 and 2022 for a deeper analysis. Also, research with urban drainage 

and performance and metabolism keywords was made in the last five years, described more 

in a later section, 3.3.1 item. 

In this search 224 documents were found in all databases described before. A first 

filter was made to eliminate the duplicated articles and other types of publications, resulting 

in 174 documents. After that, a title and abstract reading was made to exclude other areas 

papers, remaining only 45. This data was separated in groups: Metabolism fundaments 

foundation and WaterMet² creation, reduce ratio analysis (only with a few of key 

performance indicators or in a small proportion), case studies and planning & management, 

as shown in Figure 1.  



7 
 

 

Figure 1 – Number of papers per year by subject about Metabolism and Urban water 

systems. 

 

Source: Research data - Scopus, Science Direct and Web of Science. 

 

With these we can observe there is a constant publication of reduced ratio papers, 

which are those ones with only unique Key performance indicators (KPIs), or specific related 

to water supply systems or wastewater systems; the also happens with the urban planning & 

management subject. The rest was concentrated in the concept creation and some case 

studies focus on water management; after that only a few works was made mention 

metabolism in urban water systems, but usually for making reviews.  

3.2 URBAN METABOLISM 

The urban metabolism refers to the chemic and biochemic process necessary to 

maintain life, in a similarly way of the body definition of the word. The term in water system 

refers to a circulatory process which need of the “inputs” to future being transformed inside 

and generate some “output” and provide something to this system (VENKATESH; 

GOVINDARAJAN, 2014).  

The urban metabolism starts as a tool to analyse the sustainability, policies, and 

urban design (KENNEDY; PINCETL; BUNJE, 2011). With the method improvement the 

term was divided into two aspects, firstly related to energy and materials and second to the 

0

1

2

3

4

5

6

2013 2014 2015 2016 2017 2018 2019 2020 2021 2022

Number of papers per year by subject

Planning & management Reduced ratio

Integrated Urban water system Case studies

Metabolism Fundaments and WaterMet² creation



8 
 

 

mass fluxes, used in this work (KENNEDY; PINCETL; BUNJE, 2011; apud HUANG et al., 

2013).  

The urban water systems in general need a holistic analysis of their impacts, and 

this can be done through metabolic analysis, that means how much the urban system needs 

resources and how can transform the natural conditions of the environment, that could be in 

demand of investment and operating costs, materials and chemicals products, or change 

impacts, such as contaminants discharge and emissions in general (BEHZADIAN; 

KAPELAN, 2015a). 

The metabolism-based approach considers the mass stocks in the system and the 

inflows and outflows (of materials, energy, chemicals, emissions and wastes generation), 

considering the lifecycle assessment, and the urban water dynamics or the life cycle 

assessment (VENKATESH, 2013), the fluxes can been saw in Figure 2. The water 

metabolism approach behaves the same manner as described before but focus on urban water 

flow and their processes (VENKATESH; SÆGROV; BRATTEBØ, 2014).  

Figure 2 - Common flux and hierarchy in urban metabolism. 

 

Source: (RENOUF et al., 2017). 

In national context only a few studies related to urban metabolism were made 

(Scopus data), mainly related to water management (GALVÃO; MARINHO; MIRANDA, 

2017), Ambiental risks (FARIAS; MENDONÇA, 2022),  urban planning (MARINHO, 

2018; (MARINHO; GALVÃO; MIRANDA, 2020), urban drainage focuses on hydrologic 

performance (BEZERRA et al., 2020) and multicriterial performance (NOVAES, 2016). The 

last two considers the structure and hydrological performance of the system, and the last one 



9 
 

 

also considers Ambiental and social parameters too, but do not consider the contaminants, 

energy demand and GHG emissions.  

In international context, many works have been doing in urban metabolism area, 

the urban drainage in general and urban water systems has a lot of papers published related 

to performance, while metabolism in urban drainage has much less works than the 

metabolism in urban water systems. However, this number become zero when we observe 

that correlating the metabolism to performance. 

A recent review of urban systems performance analysis studied the tools and how 

they are being used to analyse urban water systems. Among these tools, two were highlighted 

as the most recent ones, the Watershed Management Optimization Support Tool (WMOST) 

and the Watermet² Model. The WMOST is focused on local and small basins and the 

Watermet², object of this work and detailed in 3.3 item, allows a metabolism analysis at the 

city and large basins levels (MOSLEH; NEGAHBAN-AZAR, 2021). 

There is a tool called Dynamic metabolism model (DMM), that refers to the 

dynamics of cities and their outputs inside the Urban metabolism. The mainly difference 

between the DMM and the Watermet² model, is that in the first, the simulation of the total 

annual water production and some factor application; when the Watermet² model consider 

other water and flows (BEHZADIAN; KAPELAN, 2015a). 

3.3 WATERMET² 

The WaterMet² it is a software that analyse the water mass-balance-based model 

that simulate the urban fluxes and with the inputs generated the sustainable performance 

outputs. The software only considers the dynamic of the operation of the system, due this 

installation and production have a minimal impact in the outputs. The software process is 

shown in the Figure 3 (BEHZADIAN; KAPELAN, 2015a). The software recognizes 

different kind of storages, water streams, treatments (K. BEHZADIAN et al., 2014). 



10 
 

 

Figure 3 - Main process of Watermet² software. 

 

Source: (BEHZADIAN; KAPELAN, 2015a). 

The Watermet² was created to measure the quantity and quality indicators based on 

the system's metabolic and sustainability performance plus other metrics. Its strength lies in 

its analysis of the integrated urban water system. (BEHZADIAN; KAPELAN, 2015b); it can 

analyse different flows such as water, energy, GHG emissions, pipe materials, chemicals, 

and pollutants; also, these fluxes can be aggregated temporally and spatially to the city 

subsystems (K. BEHZADIAN et al., 2014). 

Each flow depends of the spatial level of the analysis, which are Indoor area (Water 

consumption points), Local area (Water consumption points, Rainfall-runoff modelling, 

Rainwater harvesting tank and Grey water recycling tank), the last two in local area it’s also 

and the only ones generated in subcatchment area, and the System area (Water supply 

conduits, Trunk mains, Distribution mains, Combined/separate sewer networks, water and 

wastewater treatment works, Service reservoirs and Water resources) (BEHZADIAN; 

KAPELAN, 2015a; K. BEHZADIAN et al., 2014), as shown in Figure 4. 



11 
 

 

Figure 4 - Main flows and storage in the software. 

 

Source: (BEHZADIAN; KAPELAN, 2015a) 

The flows were applied to the spatial level and quantitative data of key performance 

indicators trough time are generated as results. The software does the simulation in a daily 

time step, since in small temporal analysis the results generated versus the necessary input 

is not efficient. When bigger the temporal resolution, most accurate it is the quantity in small 

spatial scales (BEHZADIAN; KAPELAN, 2015a).  

The steps to achieve these results are described in the software manual are: Add and 

describe the topology and characteristics of the components in the respectively forms, 

specify the constant values in the option form, run a simulation and extract the results (K. 

BEHZADIAN et al., 2014).  

The basic input data forms are necessary to run a simulation, which are the Water 

supply, subcatchment and wastewater forms, representing the structure, the Option form 



12 
 

 

representing the characteristics and the Time series form representing the time data (K. 

BEHZADIAN et al., 2014). Also, there is some calibration parameters that need to be fill 

for improve the model accurance, for water supply system based on daily, month and annual 

variation demand, and for wastewater system based on hydrological and structure values (K. 

BEHZADIAN et al., 2014).  

3.3.1 Urban drainage 

The results of adaptation lack of cities are showing up, many occurrences of 

flooding, overflows and others are more frequent. The concern about drainage devices is 

growing up, and with that comes the need for new approaches in urban systems, whether to 

promote resilience (KOURTIS; TSIHRINTZIS, 2021; OSENBERGER et al., 2021; 

PANOS; WOLFAND; HOGUE, 2021; MCCLYMONT et al., 2020) or improve their 

performance (SEYEDASHRAF; BOTTACIN-BUSOLIN; HAROU, 2021; YANG; 

ZHANG, 2021; FOOMANI; MALEKMOHAMMADI, 2020; LA ROSA; PAPPALARDO, 

2020).  

Recent studies that use the drainage system performance as a metric for the method 

selection, do it based mainly on: existing criteria in the literature of sustainable devices, 

diffuse pollution, environmental impact, quantity (hygienist concepts), professionals 

consulting or even randomly, as demonstrated in the works: (YANG; ZHANG, 2021), 

(SEYEDASHRAF; BOTTACIN-BUSOLIN; HAROU, 2021), (ARIZA et al., 2019), 

(WANG et al., 2017), (NOVAES, 2016). 

The terminology is extremely important in these evaluations, nowadays there is a 

high quantity of applicable drainage system terms used in the studies to improve the 

urbanization and the cities resilience, as green infrastructure, Low-impact development, 

water-sensitive urban design, best management practices, alternative techniques and 

sustainable urban drainage systems, being the last one the most recent (GEBEREMARIAM, 

2021) and the mainly term focus in this work.  

The planning of the sustainable urban drainage system (SUDs) is based on four 

pillars: Biodiversity, Quality, Softness/Amenity and Quantity, as shown in Figure 5, which 

are focused on increasing natural surfaces, improving water quality in water bodies, creating 

a system multifunctionality and approximating the natural flow regime, respectively 

(WOODS BALLARD et al., 2015).  



13 
 

 

Figure 5 - SUDS pillars. 

 
Source: (LÄHDE et al., 2019) 

The most frequent typologies of SUDS are the Green Spaces (Tree Pits, 

Raingardens, Green roofs, grassed swales, Bioretention ponds, Bioretention Cells), 

Constructed wetlands, Storage Tanks and Rain Barrels, Infiltration trenches and sand filters, 

Permeable pavements, infiltration basins and dry ponds (FERRANS et al., 2022); which 

attend to the sustainable pillars. 

The decision making between the possible approaches to the drainage system is 

usually made following the steps shown in Figure 6, which corresponds to decision processes 

and the enhancement to the optimal solution (FERRANS et al., 2022). As could be saw in 

the image, to perform a simulation in the urban drainage models first we have to propose 

initially a solution, raise the parameters of study to after that run a simulation at the model, 

make this process how many times is necessary and then apply the decision process based 

on the model outputs.  

Figure 6 - Decision making processes. 

 
Note: Urban drainage models (UDM). 

Source: (FERRANS et al., 2022) 



14 
 

 

 
 

4 JUSTIFICATION AND HYPHOTESIS 

In urban drainage the most frequency software or Drainage model used for all the 

questions related to the sustainable systems implementation, as the design decision criteria, 

water quantity and quality, are Storm Water Management Model (SWMM) and own-

developed models (FERRANS et al., 2022). The actual version of SWMM allow the user to 

design and simulated the green infrastructure such as rain barrels, permeable pavement, and 

infiltration trenches; also permits the implementation of Best Management Practices (NIAZI 

et al., 2017). The software used to calculate the water quantity and quality, also supporting 

some green infrastructure. As a result, the specific area of SWMM is calculate the quantity 

and quality of water, and their output is only the water contents, the software does not 

calculate the GHG emissions/flow, energy flow, acidification and eutrophication flow, some 

chemical flux, and costs (ROSSMAN; HUBER, 2016).  

The Watermet² model is a specific software for the metabolism analysis, and it is 

shown as a potential tool to analyse the KPIs, being capable of measure system flows (water 

flow, energy flow, GHG flow, acidification/eutrophication flow, chemical flux, pollutant 

flux, quantify the pipelines material flux), which permits the quantification of many of this 

indicators, such as the operational and maintenance costs, or any risks and intervention 

analysis for the components (K. BEHZADIAN et al., 2014). 

 The Watermet² is the mainly tool of this paper, pretending to observe the 

comportment of sustainable urban drainage system under a metabolic perspective, and how 

this could support the decision making between the SUDS types.  

During the bibliographic review process no papers related to metabolism 

performance of drainage systems (Scopus database) was found, and only a few studies of 

urban metabolism and drainage systems. But in the last one does not focus on the drainage 

system, only in the urban water system design and management, citing briefly the relation 

between them. Highlighting the importance in the in-depth on this subject. 



15 
 

 

 

5 MATERIALS AND METHODS 

Taking the errors and difficulties in the choosing methods process of urban drainage 

(MCCLYMONT, et al, 2020) and the challenges to achieve an urban resilience 

(MCCLYMONT, et al, 2020), this research presents a way of help the decision maker 

between drainage systems based on urban metabolism fundaments. 

When there is a decision to be made based on several criteria, it is necessary to 

perform a sensitivity analysis to measure the uncertainty of the outputs based on the inputs, 

which can be assessed by measuring the performance of an alternative, based on specific 

criteria (HADDAD; SANDERS, 2018), such as performed in this study. 

This evaluation was made through the construction of the normalized hypothesis 

model to make the sensitivity analysis, alternatives situations to apply in the study object, 

processing in the analysis software to generate the performance indicators and result 

analysis, resumed in Figure 7. 

Figure 7 – Research summary. 

 



16 
 

 

 

5.1 MATERIALS 

5.1.1 Spatial data 

For all the map’s elaboration for this work was used the software QGis 3.16.15, 

where shapefiles and data were inserted according to Table 1. 

Table 1 - Files database. 

DATA REFERENCES 
Background image Google Satellite [1] 

Zones (SEPLAN, 2022) 
Streets (IBGE, 2019) 

São Paulo State (IGC, 2015) 
Contour line (INPE, 2011) 

Drainage system assets  (PMB, 2019) 
Water supply system assets (AMPLA, 2016c) and (DAE; HIDROSAN, 2014a) 
Wastewater system assets (AMPLA, 2016a) and (DAE; HIDROSAN, 2014a) 

[1] Google images was obtained inside the QGis Software.  

5.1.2 Metabolism software: WaterMet²  

The input dataset necessary to run a simulation is the “Inflow time series”, “Weather 

data” and “Distribution network pipelines data”; also, is demanded to fill some forms inside 

the software with the characteristic’s local conditions, as described before in the Watermet² 

section.  

5.1.2.1 Inflows time series  

The inflow time series is represented by number of resources that is specified in 

Water Supply, since they are the quantity of water that are inserted in the system (K. 

BEHZADIAN et al., 2014). The values were made by the system characteristics, which is 

specified in the Method Section (5.2). 

5.1.2.2 Weather data 

To generate the performance indexes, it is necessary for the calculations insert a 

series of temporal data, which must include: Date, precipitation data (rain or snow) and type, 

minimum, mean and maximum temperature, average wind speed, hours of sunshine, mean 

relative humidity and vapour pressure (K. BEHZADIAN et al., 2014), which was obtained 

as presented in Table 2. 



17 
 

 

Table 2 - Weather Database. 

DATA REFERENCES 
Date (ALMAGRO et al., 2021) 

Precipitation “P” (mm) (ALMAGRO et al., 2021) 
Snow depth “Pn” (cm) (ALMAGRO et al., 2021) 
Precipitation type “Tp” Analysis [1]  

Mean Temperature “Tm” (ºC) (SILVEIRA et al., 2012) [2] 
Minimum Temperature “Tmin” (ºC) (ALMAGRO et al., 2021) 
Maximum Temperature “Tmax” (ºC) (ALMAGRO et al., 2021) 

Average wind speed “Vv” (m/s) (ALMAGRO et al., 2021) 
Hour of sunshine “hs” (h) (LUIZ JCB CARVALHO et al., 1985)[3] 

Mean relative humidity “h” (ALMAGRO et al., 2021) 
Vapour Pressure (hPa) (ALLEN et al., 1998) [4] 

[1] The precipitation type should be determined by Rain, Snow or No, as there is not snow data at the 
location (ALMAGRO et al., 2021; IPMET, 2022) this data was classified according to the rain episode, 
being named as Rain or No. 

[2] 𝑻𝒎 = 𝑻𝒎𝒊𝒏+𝑻𝒎𝒂𝒙𝟐   , where “Tmin” and “Tmax” values (ALMAGRO et al., 2021). 

[3] hs= 𝟐𝟏𝟓 × 𝐜𝐨𝐬−𝟏[−(𝐭𝐚𝐧 𝜽 × 𝐭𝐚𝐧 𝜹)], where “θ” is the local latitude in degrees and “δ” is represented 
by Equation 1: 𝛅 = 𝟐𝟑, 𝟒𝟒º × 𝐜𝐨𝐬[ 𝟐𝛑𝟑𝟔𝟓°( 𝐓 + 𝟏𝟎)]                                                                                                [1] 

With “T” the Julian day, ranging from 1 to 365-366 (BOCZKO, 1984 apud BEDAQUE; BRETONES, 
2016). 

[4] 𝒆𝒂 = 𝒉𝑹𝟏𝟎𝟎 × (𝟎, 𝟔𝟏𝟎𝟖 × 𝐞𝐱𝐩 𝟏𝟕,𝟐𝟕×𝑻𝒎𝒆𝒅𝑻𝒎𝒆𝒅+𝟐𝟑𝟕,𝟑), the vapour pressure can be determined as a function of the 

relative humidity associated with the Saturation Vapor Mean Pressure equation (ALLEN et al., 1998).  

With the previous values, an excel table was created to insert in Watermet² Software 

as a Weather Database for each the Sensitivity analysis and the KPIs processes for the case 

study.  

5.1.2.3 Distribution network pipelines data 

The Distribution network pipelines, also in the time series data works in the same 

manner as the inflow time series, represented by the characteristics of the pipelines 

distributed in the urban water system (UWS), only necessary when the rehabilitation is 

considered in the simulation (K. BEHZADIAN et al., 2014). The values were made by the 

system characteristics, which is specified in the Method Section (5.2). 



18 
 

 

5.2 METHOD 

5.2.1 Sensitivity test 

Bearing in mind that we must analyse the urban metabolism in an integrated way, 

looking to all urban water system, initially an unitary area was created with the objective 

that we can analyse in future the most influence urban characteristics on the KPIs.  

The sensitivity test consists in the characterization of these standards area to 

generate the KPIs in the Watermet² software and after a statistical analysis to define the 

variables that influence in the process.  

5.2.1.1 Normalized area 

The software requires the characteristics of the area, the occupancy, and others; for 

this step was used the structure shown in Figure 8.  

Figure 8 - Sensitivity test scheme. 

 

According to the population growth in large centres, the tendency is for added 

resources to be made as subterraneous systems, so the sensitivity analysis was performed 

with an underground system for supply. 

Due to data restrictions and local specificity, the sensitivity analysis was performed 

considering consumption data available in consolidated reference. For water treatment, 

simple filtration was considered, with pH control (Cal) and disinfection (Chlorine); which 

goes to the storage tanks using water-pumping for transport (LEMOS et al., 2013). 

In the city data was considered for the hypothetic catchment, the bigger basin 

between the small subcatchment as a default value, having an area of 2km² (DAEE, 2008). 

For the population number, was considered the São Paulo City that is the bigger city in Brazil 

(IBGE, 2010), which contains an urban density surround 7399 residents per Km² (IBGE, 

2022), also an aleatory city block was chosen in the same city (Coordinates: 23°32'18.8"S 



19 
 

 

46°37'43.7"W) for an estimative of properties, roofs, pavement, permeable and impermeable 

areas.  

How this analysis pretends simulate an ideal urban area, was considered a half of 

commercial activities and the rest of residential areas. All the default data chosen for this 

simulation was consulted for São Paulo State, considering that is the most urbanized area 

(IBGE, 2010), these values are resumed and shown by Table 3.  

Table 3 – City data for Sensitivity analysis.  

DATA REFERENCES VALUE 
Area (DAEE, 2008) 200 ha 

Population (IBGE, 2022) 14 797 residents 
Impermeable area Calculated 86,6 % 

Roof area Calculated 107,2 ha 
Permeable area (PMSP, 2002) 26,8 ha 
Pavement area  Calculated 66 ha 

The time series depends on the supply, demand, and specific characteristics, so was 

considered for this simulation:  

▪ Inflows time series: standard values according to the unit simulation. 

▪ Weather time series: the weather data is many variables as the place locate it, in this 

case was used the case study weather data. 

▪ Distribution network pipelines: standard values according to the unit simulation. 

Table 4 – Input time series for Sensitivity analysis. 

DATA  REFERENCES 
Inflows Not necessary¹ 
Weather Case study (5.1.2.2 item) 

Distribution network pipelines Not necessary² 
¹ The inflow time series represents the dataset of superficial water volume and is not necessary for 
groundwaters resources (K. BEHZADIAN et al., 2014). 
² Was not considered because was considered an old system, with no structure renovation.  

The structure used for this step was made by assuming that only exists one unit each 

phase of the water system and the data was assumed the default values, as the shown in Table 

4 for Input time series, Table 5 for subcatchment, Table 6 for Water supply system, Table 7 

for Wastewater and Table 8 for Water resource recovery. 

 



20 
 

 

Table 5 – Subcatchment parameters for Sensitivity analysis. 

DATA REFERENCES VALUE 

Local 
area 

Number of properties Google Images 
36 per block = 

3600 
Total area (ha) (DAEE, 2008) 200 

Indoor water demand (L/day per capita) (SNS, 2022) 176,0 

Industrial/Commercial demand (m³/day) 
(FECOMERCIOSP, 

2014) 
0,288 

Irrigation and other water demand (m³/day) (Figure 8) 04 

Frost tapping (m³/day) 
(ALMAGRO et al., 

2021) 
0 

Unregistered public use (L/day per capita) Assumed³ 0 
Occupancy 

(Population/properties) 
(K. BEHZADIAN et 

al., 2014) 
14797/3600 = 4,11 

Roof Area Proportion (%) Calculated. 53,6 
Pervious Area Proportion (%) Calculated. 13,4 

Pavement & Road Area Proportion (%) Calculated. 33,0 
Runoff Coefficient Calculated. 0,866 ≈ 0,9 

Infiltration Coefficient 
(JAWORSKA-SZULC, 

2009) 
0,25¹ 

Water demand specific No variation² 
¹ For this number selection was used the permeable urban areas. 
² Considering that in the sensitivity test was selected a unitary area, the variation throughout the day and 
months was ignored. 
³ This value was assumed as zero because the existing data it is integrated values. 
4 This value only exists when there is an irrigation use and others uses beside the indoor, industrial, and 
commercial demand.  

Table 6 – Input parameters - Water supply system for Sensitivity analysis. 

(Continue) 

DATA REFERENCES V ALUE 

Water 
Resource 

(WR) 

Water supply type (Table 4) Groundwater 
Capacity and Volume (m³) (K. BEHZADIAN et al., 2014) Not necessary. 

Water Loss (%) (K. BEHZADIAN et al., 2014) 
0% 

(groundwater) 
Electricity (kWh/m³) (SABESP, 2022) 0 

Fossil fuel (L/m³) (SABESP, 2022) 0 
Operational Cost (Euro/yr.) * * 

Diverted Flow (m³/day) Assumed¹ 0 

Water supply 
conduits (SC) 

Transmission capacity (m³/day) (NETTO, 1998) 1190 
Leakage Assumed² 0 

Electricity (kWh /m³) (SABESP, 2022) 0 
Fossil fuel (L/m³) (SABESP, 2022) 0 

Operational Cost (Euro/yr.) * * 

Water 
treatment 

works 
(WTWs) 

Treatment capacity (m³/day) (NETTO, 1998) 1190 
Water loss (% of treatment/day) (SABESP, 2022) 0,9 

Phys. Electricity (kWh /m³) (SABESP, 2022) 0 
Phys. Fossil (L/m³) (LEMOS et al., 2013) 0 

Phys. Operational Cost (Euro/yr.) * * 

Chemical electricity (kWh /m³) 
(GUANAIS; COHIM; 

MEDEIROS, 2017) 
3,22 

Chemical Fossil (L/m³) Assumed³ 0 

 



21 
 

 

(continuation) 

 

Chemical Operational Cost 
(Euro/Yr.) 

* * 

Average Chemical Cost (Euro/Yr.) * * 
Sludge generated (kg/m³) (LEMOS et al., 2013) 0,68 
Diverted flow (m³/day) Assumed¹ 0 

Chemicals (Calcium hydroxide) 
(GUANAIS; COHIM; 

MEDEIROS, 2017) 
0,019 

Chemicals (Chlorine) 
(GUANAIS; COHIM; 

MEDEIROS, 2017) 
0,012 

Trunk Mains 
(TM) 

Transmission capacity (m³/day) (SABESP, 2022) 1180 
Leakage Assumed¹ 0 

Electricity (kWh/m³) (SABESP, 2022) 0 
Fossil fuel (L/m³) (LEMOS et al., 2013) 0,00018 

Operational Cost (Euro/Yr.) * * 

Service 
Reservoir 

(SR) 

Storage capacity (m³) (NETTO, 1998) 891 
Initial volume (m³) (NETTO, 1998) 0 

Water loss (% of treatment/day) Assumed¹ 0 
Operational Cost (Euro/Yr.) * * 

Average Cost (Euro/Yr.) * * 
Diverted flow (m³/day) Assumed¹ 0 

Chemicals Assumed³ 0 

Distribution 
Mains (DM) 

Transmission capacity (m³/day) (SABESP, 2022) 1180 
Leakage (SABESP, 2022) 0,279 

Annual Rehabilitation (%) 0 0 
Electricity (kWh/m³) (SABESP, 2022) 0,83 

Fossil fuel (L/m³) (LEMOS et al., 2013) 0,0016 
Operational Cost (Euro/Yr.) * * 

* The cost was not considered in this study. 
¹ Was assumed diverted flow as zero, since there is not loose in these systems. 
² Was assumed Leakage as zero, since there is not loose in these systems. 
³ This value was assumed as zero because the existing data it is integrated values. 
 

Table 7 – Input Wastewater parameters for Sensitivity analysis. 

(continue) 

DATA REFERENCES  

Combined/ 
Sanitary 

Sewer 

CSO Structure Capacity (m³/day) (SABESP, 2022) 778 

Infiltration (%) 
(K. BEHZADIAN et al., 

2014) 
0 

Exfiltration (%) 
(K. BEHZADIAN et al., 

2014) 
0 

Rehabilitation (%) Assumed² 0 
Electricity (kWh/m³) Assumed¹ 0 

Fossil fuel (L/m³) (LEMOS et al., 2013) 0 
Operational Cost (Euro/Yr.) * * 

Storm sewer 

STO Structure Capacity (m³/day) (DAEE, 2008) 110 

Infiltration (%) 
(JAWORSKA-SZULC, 

2009) 
25 

Exfiltration (%) 
(JAWORSKA-SZULC, 

2009) 
75 

Rehabilitation (%) Assumed² 0 

 



22 
 

 

(continuation) 

 
Electricity (kWh/m³) Assumed³ 0 

Fossil fuel (L/m³) Assumed³ 0 
Operational Cost (Euro/Yr.) * * 

Wastewater 
treatment 

works 
(WTWs) 

Treatment capacity (m³/day) Assumed¹ 778 
Storage capacity (m³) Assumed² 0 
Electricity (kWh/m³) (SABESP, 2022) 0,44 

Fossil (L/m³) (LEMOS et al., 2013) 0 
Operational Cost (Euro/Yr.) * * 

Average Chemical Cost 
(Euro/Yr.) 

* * 

Other Specifications ² ² 
Receiving 

water Receiving water name RW1 

* The cost was not considered in this study. 
¹ This value was assumed as zero because the existing data it is integrated values. 
² Not considered for the sensitivity test. 
³ This value was assumed as zero because of the gravity simple system adopted.  
 

Table 8 – Input Water resource recovery parameters for Sensitivity analysis. 

DATA SITUATION 

Water recovery Allocation Allocation from Rainwater 
Harvesting 

Not considered in the sensitivity test. 

Decentralized water reuse 
Rainwater Harvesting tank Not considered in the sensitivity test. 
Greywater Harvesting tank Not considered in the sensitivity test. 

Centralized water reuse Local areas Not considered in the sensitivity test. 

5.2.1.2 Key performance indicators 

The result form to select and fill consists of five parts, spatial resolution, temporal 

resolution, KPIs, previous KPIs and KPIs for plot (K. BEHZADIAN et al., 2014). For the 

spatial resolution was always used the UWS, due to this work objective. The temporal 

resolution, which contains the selection of the graph time representation. The previous KPIs, 

place that was saved those generated the simulation before. The KPIs for plot, of the actual 

simulation.  

The KPIs available for plotting in the Watermet² can be adjusted for the users need 

(K. BEHZADIAN et al., 2014), in this case was choose one of each category with selected 

subcategories as the running simulation, which are Water balance, fraction of water demand 

delivered, sewer system balance, energy, GHG emission, Acidification, Eutrophication, 

Contaminant Load and Sludge Generation, represented by graphs and tables. For the plot 

form was choose the absolute value, to a better understanding of the impacts.  



23 
 

 

5.2.1.3 Statistical analysis 

Until this step was made a first fill with the normalized area described before and 

the graphs and tables necessary was generated, constant in the 6.1 item; then was made a 

system disturbance of 100% in diverse subcatchments and wastewater variables for this 

analysis (Irrigation and other water demand, Frost tapping, Industrial and Commercial 

demand, Rehabilitation, Electricity, Fossil Fuel, Treatment Capacity, Storage Capacity, 

Electricity and Fossil Fuel, Number of properties, Indoor water demand, occupancy, 

Infiltration coefficient and Exfiltration, Total area and the Transmission capacity), those 

changes have a response in the generated KPIs (Total of Water Demand, Total leakage, 

Fraction of Water Demand Delivered, Excess stormwater, Excess Wastewater, Total 

Combined Sewer Overflow, Treated outflow from Wastewater Treatment Works, Storm 

tank overflow, Total Energy, Total GHG emissions, Total Acidification, Total 

Eutrophication, Contaminants – Chemical oxygen demand (COD), Phosphorous and 

Nitrogen, and Total Sludge generation). The sensitivity analysis considers the accumulated 

values of these KPIs in a reduced planning horizon, to analyse the impacts in short term 

analysis.  

The method for this evaluation was chosen because of the extensive numbers of 

input variables in the system, so a Principal Component Analysis (PCA) was made to 

evaluate the significant ones.  

The principal Component Analysis is a method that reduce the number of variables 

or the data available for analysis preserving the maximum of correspondence about the 

dataset; this can be accomplished by defining direction called Principal Components were 

the variation in data are most frequent. The PCA create new variables (PCs) with the 

combination of the original variables that influence in the process/ experiment. (RINGNÉR, 

2008).  

The PAST software was chosen due to intuitive manipulation and basic input data 

type. PAST (Paleontological Statistics) software applies spreadsheet data entry and process 

to calculate statistics, which includes the PCA multivariate statistics; and create diverse types 

of graphs, histograms, and scatter plots (HAMMER et al., 2001).  



24 
 

 

5.2.2 Case of study   

To apply the method will be necessary to use it in an existing urban drainage system. 

The site defined for the study is in the municipality of Bauru, located in the watershed of the 

Bauru River; this being one of the main water bodies in receiving rainwater runoff and the 

discharge of entirety domestic effluents (SMMA, 2008). 

The selected area has a prominent level of impermeabilization and a deficiency in 

the drainage system, causing recurrent episodes of flooding and inundation (IPMET, 2022). 

The urban basin has this mouth at the intersection of Nations Units Avenue with Nuno de 

Assis Avenue, located at latitude and longitude coordinates of -22.312; -49,069, 

respectively; called “Córrego das Flores” or “Ribeirão das Flores”, the respective basin is 

shown in the Figure 9. 

Figure 9 - Catchment localization map at São Paulo State.  

 

The administration has established through Municipal Laws some guidelines for 

solving the drainage problems, with still in slow steps application, such as: implementation 

of land use policy, sustainably recover valley bottoms, containment works, administrative 

measures of water containment (control of land use by basin and control of works), channels 



25 
 

 

dredging, incentive to cisterns use and retention devices, as well as “zero impact” policies 

in new neighbourhoods and subdivisions, incentive to drainage pavement projects, 

conservation units and rainwater reuse in existing buildings, and especially the mandatory 

reuse of rainwater in new buildings over 300m² (PMB, 2017a). 

5.2.2.1 Input data 

All the area data was made as the subcatchment data, measure in the QGis map 

contents in the Appendix A, which refers to subcatchments, impermeable, roof, permeable 

and pavement area, and an estimation of population based on (IBGE, 2022), similarly to the 

described in Table 3. 

The time series was filled as follow: 

▪ Inflows time series: as the water supply system description. 
▪ Weather database: as described in 5.1.2.1 item. 
▪ Distribution network pipelines: optional when the rehabilitation is not 

considered. 

A resume of the study object system was inserted in the Appendix B. Which are 

described individually below.  

5.2.2.2 Water Supply system description 

The water supply system of the catchment has one abstraction of surface water and 

the rest of groundwater, stocking the water in diversify reservoir, the interconnecting 

between then and then supplying to the subcatchments (DAE; HIDROSAN, 2014b).  

Table 9 – Input parameters Water supply system for Study case analysis. 

(continue) 

DATA REFERENCES  

Water 
Resource 

(WR) 

Water supply type (DAE; HIDROSAN, 2014b) 

Capacity and Volume (m³) (AMPLA, 2016c) 

Water Loss (%) (AMPLA, 2016c) 

Electricity (kWh/m³) (DAE; HIDROSAN, 2014c) 

Fossil fuel (L/m³) (AMPLA, 2016c) 

Operational Cost (Euro/Yr.) * 

Diverted Flow (m³/day) Assumed¹ 

Water 
supply 

conduits 
(SC) 

Transmission capacity (m³/day) (DAE; HIDROSAN, 2014b) 

Leakage (DAE; HIDROSAN, 2014d) 

Electricity (kWh/m³) (DAE; HIDROSAN, 2014d) 

Fossil fuel (L/m³) (DAE; HIDROSAN, 2014d) 

Operational Cost (Euro/Yr.) * 



26 
 

 

(continuation) 

 Water 
treatment 

works 
(WTWs) 

Treatment capacity (m³/day) 
(DAE; HIDROSAN, 2014d) and (DAE; 

HIDROSAN, 2014c) 

Water loss (% of treatment/day) (AMPLA, 2016c) 

Water 
treatment 

works 
(WTWs) 

Phys. Electricity (kWh/m³) 
(DAE; HIDROSAN, 2014d) and (DAE; 

HIDROSAN, 2014c) 

Phys. Fossil (L/m³) 
(DAE; HIDROSAN, 2014d) and (DAE; 

HIDROSAN, 2014c) 

Phys. Operational Cost (Euro/yr.) * 

Chemical electricity (kWh/m³) Assumed³ 

Chemical Fossil (L/m³) Assumed³ 

Chemical Operational Cost (Euro/Yr.) * 

Average Chemical Cost (Euro/Yr.) * 

Sludge generated (kg/m³) 
(DAE; HIDROSAN, 2014a) and 

(KATAYAMA, 2012) 

Diverted flow (m³/day) (DAE; HIDROSAN, 2014a) 

Chemicals 
(DAE; HIDROSAN, 2014d) and (DAE; 

HIDROSAN, 2014c) 

Trunk Mains 
(TM) 

Transmission capacity (m³/day) (DAE; HIDROSAN, 2014c) 

Leakage (DAE; HIDROSAN, 2014c) 

Electricity (kWh/m³) (DAE; HIDROSAN, 2014c) 

Fossil fuel (L/m³) (DAE; HIDROSAN, 2014c) 

Operational Cost (Euro/Yr.) * 

Service 
Reservoir 

(SR) 

Storage capacity (m³) (DAE; HIDROSAN, 2014a) 

Initial volume (m³) (DAE; HIDROSAN, 2014a) 

Water loss (% of treatment/day) Assumed¹ 

Operational Cost (Euro/Yr.) * 

Average Cost (Euro/Yr.) * 

Diverted flow (m³/day) Assumed¹ 

Chemicals (DAE; HIDROSAN, 2014a) 

Distribution 
Mains (DM) 

Transmission capacity (m³/day) (SABESP, 2022)  

Leakage (AMPLA, 2016c) 

Annual Rehabilitation (%) (AMPLA, 2016c) 

Electricity (kWh/m³) (AMPLA, 2016c) 

Fossil fuel (L/m³) Assumed³ 

Operational Cost (Euro/Yr.) * 

* The cost was not considered in this study. 

¹ Was assumed diverted flow as zero since there is not loose in these systems. 

² Was assumed Leakage as zero since there is not loose in these systems. 
³ This value was assumed as zero because the existing data it is integrated values. 

 

 



27 
 

 

5.2.2.3 Subcatchment system description 

 

The area was classified as the APENDDIX A and divided in thirteen subcatchments 

which were allocated in local areas based in the city zones, named as “a, b, c, …”.  

Table 10 – Subcatchment parameters for Study case analysis. 

DATA REFERENCES 

Local 
area 

Number of properties QGis counting 
Total area (ha) QGis measure 

Indoor water demand (L/day per capita) (SNS, 2022) 
Industrial/Commercial demand (m³/day) (FECOMERCIOSP, 2014) 

Irrigation and other water demand (m³/day) As the zone areas 
Frost tapping (m³/day) (ALMAGRO et al., 2021) 

Unregistered public use (L/day per capita) Assumed² 
Occupancy 

(Population/properties) 
(K. BEHZADIAN et al., 2014) and (IBGE, 

2010) 
Roof Area Proportion (%) Calculated. 

Pervious Area Proportion (%) Calculated. and (PMB, 2008) 
Pavement & Road Area Proportion (%) Calculated. 

Runoff Coefficient (APOLINARIO et al., 2017) 
Infiltration Coefficient (JAWORSKA-SZULC, 2009) 
Water demand specific (AMPLA, 2016b) 

¹ For this number selection was used the permeable urban areas. 
² This value was assumed as zero because the existing data it is integrated values. 

5.2.2.4 Wastewater system description 

The wastewater system it is a separated one, with the indoor capitation and disposal 

by pipelines to the Wastewater treatment works, then return to water bodies (AMPLA, 

2016a).  

Table 11 – Input Wastewater parameters for Study case analysis. 
(continue) 

DATA REFERENCES  

Combined/ 
Sanitary Sewer 

CSO Structure Capacity (m³/day) (AMPLA, 2016a) 
Infiltration (%) (K. BEHZADIAN et al., 2014) 
Exfiltration (%) (K. BEHZADIAN et al., 2014) 

Rehabilitation (%) (AMPLA, 2016b) 
Electricity (kWh/m³) (AMPLA, 2016a) 

Fossil fuel (L/m³) (AMPLA, 2016a) 
Operational Cost (Euro/Yr.) * 

Storm sewer 

STO Structure Capacity (m³/day) (DAEE, 2008) 110 
Infiltration (%) (JAWORSKA-SZULC, 2009) 
Exfiltration (%) (JAWORSKA-SZULC, 2009) 

Rehabilitation (%) Assumed¹ 
Electricity (kWh/m³) Assumed² 

Fossil fuel (L/m³) Assumed² 
Operational Cost (Euro/Yr.) * 



28 
 

 

(continuation) 

Wastewater 
treatment works 

(WTWs) 

Treatment capacity (m³/day) (AMPLA, 2016a) 
Storage capacity (m³) (AMPLA, 2016a) 
Electricity (kWh/m³) (AMPLA, 2016a) 

Wastewater 
treatment 

works (WTWs) 

Fossil (L/m³) (AMPLA, 2016a) 
Operational Cost (Euro/Yr.) * 

Average Chemical Cost (Euro/Yr.) * 
Other Specifications ² 

Receiving 
water Receiving water name Adopted. 

* The cost was not considered in this study. 
¹ Not considered. 
² As the hypothesis scheme.  

 

5.2.3 Hypotheses/intervention options 

The intervention schematic profiles were built in the QGis 3.10 software (chosen 

because of the possibility of including layers and/or georeferenced points for sustainable 

urban drainage devices) and, when necessary, Google Earth® software. 

Will be made six hypotheses, being the first one the same as the case study, named 

as described in Table 12; and planning as described previously. 

Table 12 - Hypothesis nomenclature. 

HYPHOTESIS HYPHOTESIS NUMBER 
Conventional system HP1 

Constructed wetlands, green roofs, Bioretention ponds and 
Bioretention Cells 

HP2 

Storage Tanks and Rain Barrels (Decentralized system) HP3 
Infiltration trenches and sand filters HP4 

Permeable pavements HP5 
Green Spaces (Tree Pits, dry ponds, Raingardens, Grassed swales, 

and infiltration basin) HP6 

Storage Tanks and runoff reservoir (Centralized system) HP7 

For the Watermet² software hypothesis conditions, was firstly fill the fixed values, 

as shown in the APPENDIX F for water supply system, APPENDIX G for subcatchment 

data and APPENDIX I for wastewater values and then complete with the intervention 

characteristics, as described below.  

The conventional system simulation was filled according to the case study 

described before, since the study case model uses only conventional drainage methods. The 

detailed and complete filling of the form with the system numbers consists of the study case 

data, with fixed values, and the study case data, with variable values that is shown in the 



29 
 

 

APPENDIX G. This hypothesis was the reference to compare between the applications 

impacts of the alternatives. Also, all intervention profiles were applied as the available 

permeable public areas, as shown in Figure 10. 

Figure 10 - Available public areas. 

 



30 
 

 

5.2.3.1 Constructed wetlands, green roofs, Bioretention ponds and Bioretention 

Cells. 

The constructed wetlands are recommended for storm water treatment, mainly the 

free water surface flow wetlands (FLETCHER et al., 2020) shown in the Figure 11, which 

was chosen for this work. This type of wetland contains a specific type of vegetation that as 

the water passes through it is slowly filtered, through degradation and absorption of 

pollutants (FLETCHER et al., 2020). 

Figure 11 – Example of wetland hypothesis. 

 

Source: Adapted from (FLETCHER et al., 2020). 

Since that the constructed wetland for runoff waters need a considerable area for 

application because the necessity of a sedimentation basin and an outlet structure (LOPEZ; 

CAPRARA; UDA, 2018), in this work the application in the study area will happens only in 

park areas. The amount of 50% of the park areas will be transformed in constructed wetlands.  

The constructed wetland will have a similar structure to the model build for an 

external study, which has 1910m² and a depth of 2m (10 centimeters of free edge, 



31 
 

 

corresponding to a total of 3629m³ of treated runoff (LOPEZ; CAPRARA; UDA, 2018), 

resulting in the data shown in Table 13 for fulfillment. 

Table 13 – Constructed wetland data. 

L
oc

al
 A

re
a 

H2 (Constructed wetlands) 
W

et
la

nd
 a

re
a 

SC
 -

 R
oo

f 
A

re
a 

SC
 -

 P
er

vi
ou

s 
A

re
a 

SC
 -

 P
av

em
en

t 
&

 R
oa

d 
A

re
a 

SC
 -

 R
oo

f 
A

re
a 

pr
op

or
ti

on
 (

%
) 

SC
 -

 P
er

vi
ou

s 
A

re
a 

pr
op

or
ti

on
 (

%
) 

SC
 -

 P
av

em
en

t 
&

 R
oa

d 
A

re
a 

pr
op

or
ti

on
 (

%
) 

SC
 -

 R
un

of
f 

C
oe

ff
ic

ie
nt

 

W
W

 -
 T

re
at

m
en

t 
C

ap
ac

it
y 

(m
³/d

ay
) 

W
W

 -
 S

to
ra

ge
 C

ap
ac

it
y 

(m
³)

 

1 0.0000 0.0351 0.0067 0.0095 68.47 13.01 18.52 0.87 0 0 

2 0.0023 0.3595 0.0044 0.0932 72.33 8.93 18.75 0.90 4399 4399 

3 0.0000 0.0750 0.0083 0.0177 74.22 8.25 17.53 0.90 0 0 

4 0.0000 0.0000 0.0361 0.0322 0.00 52.83 47.17 0.63 0 0 

5 0.0000 0.0000 0.0370 0.0055 0.00 87.08 12.92 0.43 0 0 

6 0.0186 0.0000 0.0296 0.0209 0.00 58.64 41.36 0.60 35378 35378 

7 0.0000 0.3589 0.0412 0.0567 78.57 9.02 12.42 0.90 0 0 

8 0.0000 0.0657 0.0083 0.0137 74.90 9.52 15.59 0.89 
0 0 

0 0 

9 0.0000 0.2994 0.0375 0.0612 75.20 9.42 15.38 0.89 0 0 

10 0.0077 0.1809 0.0309 0.0528 68.38 11.67 19.94 0.88 14621 14621 

11 0.0000 0.0131 0.0024 0.0046 65.42 11.71 22.87 0.88 0 0 

12 0.0000 0.3678 0.0479 0.0752 74.92 9.75 15.33 0.89 
0 0 

0 0 

13 0.0000 0.1043 0.0122 0.0237 74.41 8.70 16.89 0.90 0 0 

14 0.0000 0.1043 0.0119 0.0251 73.82 8.40 17.78 0.90 0 0 

15 0.0024 0.2631 0.0355 0.0645 72.46 9.77 17.77 0.89 4567 4567 

16 0.0000 0.4088 0.0069 0.0929 80.38 1.37 18.26 0.94 0 0 

17 0.0000 0.0274 0.0030 0.0022 83.94 9.33 6.74 0.89 0 0 

18 0.0000 0.0644 0.0013 0.0142 80.70 1.58 17.72 0.94 0 0 

19 0.0000 0.0637 0.0071 0.0138 75.32 8.37 16.32 0.90 0 0 

20 0.0000 0.0682 0.0035 0.0138 79.82 4.04 16.14 0.93 0 0 

21 0.0000 0.0540 0.0041 0.0071 82.91 6.26 10.83 0.91 
0 0 

0 0 

22 0.0000 0.0434 0.0075 0.0167 64.16 11.09 24.74 0.88 0 0 

TOTAL 58965 58965 



32 
 

 

5.2.3.2 Storage Tanks and Rain Barrels 

The rain barrels were implemented in all the residences and commerce in the region. 

The idea goes in parallel with some government programs that becomes to show a progress 

in give an incentive to install cistern in the residences of semi-arid areas to supply the water 

basic necessity (ASA, 2022); and (SECRETARIA NACIONAL DE 

DESENVOLVIMENTO REGIONAL E URBANO, 2019), witch according to World Health 

Organization is 20 litters per capita (WHO, 2013 apud HERKENHOFF, 2020). The values 

of the data considering the storage described before is detailed in Table 14. 

Table 14 – Storage tanks and rain barrels data.  

Local 
Area 

H3 (Storage tanks and rain barrels) 
WW - STO Structure 

Capacity (m³/day) 
SC - Indoor water demand 

(L/day per capita) 
SC - Industrial/Commercial 

demand (m³/day) 
1 6.80 

The values of demand, even indoor and the industrial/commercial 
was automatic calculated by the Watermet² software in the Water 

Resource Recovery tab. 

2 58.31 

3 8.28 
4 4.85 
5 0.00 

6 
0.00 
0.00 

7 53.58 

8 2.41 

9 51.01 

10 15.89 

11 1.43 

12 57.58 

13 13.90 
14 14.05 

15 42.27 

16 66.12 
17 3.84 
18 8.18 
19 9.92 
20 6.03 

21 1.79 

22 1.86 
TOTAL 428.09  



33 
 

 

5.2.3.3 Infiltration trenches and sand filters  

For the infiltration trenches also called filter drains was considered an application 

adjacent to roads in their centre square and in the side bed, with one meter depth witch also 

help with the pollutant removal (WOODS BALLARD et al., 2015). The data with this 

application is shown in the Table 15. 

Table 15 - Infiltration trenches data. 

Local 
Area 

H4 (Infiltration trenches) 

Infiltration 
trenches 

area 

SC - Roof 
Area 

Proportion 
(%) 

SC - Pervious Area Proportion (%) 
SC - 

Pavement 
& Road 

Area 
Proportion 

(%) 

SC - 
Runoff 

Coefficient Green 
zones 

Trenches 
area 

Total 
pervious area 

1 0.002766 68.47 07.61 05.40 13.01 18.52 0.86 

2 0.0021033 72.33 08.97 00.42 09.39 18.28 0.89 

3 0 74.22 08.25 00.00 08.25 17.53 0.90 

4 0.0360625 00.00 00.00 52.83 52.83 47.17 0.55 

5 0.03704 00.00 00.00 87.08 87.08 12.92 0.30 

6 0.010991 00.00 73.74 21.76 95.51 04.49 0.34 

7 0.0013173 78.57 08.73 00.29 09.02 12.42 0.90 

8 0.0001708 74.90 09.32 00.19 09.52 15.59 0.89 

9 0 75.20 09.42 00.00 09.42 15.38 0.89 

10 0.0030851 68.38 13.42 01.17 14.58 17.04 0.86 

11 0.0008923 65.42 07.27 04.44 11.71 22.87 0.87 

12 0.005154 74.92 08.70 01.05 09.75 15.33 0.89 

13 0.0006061 74.41 08.27 00.43 08.70 16.89 0.90 

14 0.0002799 73.82 08.20 00.20 08.40 17.78 0.90 

15 0.0038256 72.46 09.38 01.05 10.43 17.11 0.89 

16 0.0069442 80.38 00.00 01.37 01.37 18.26 0.94 

17 0 83.94 09.33 00.00 09.33 06.74 0.89 

18 0.0012607 80.70 00.00 01.58 01.58 17.72 0.94 

19 0 75.32 08.37 00.00 08.37 16.32 0.90 

20 0.0034513 79.82 00.00 04.04 04.04 16.14 0.92 

21 0.003135 82.91 01.44 04.81 06.26 10.83 0.91 

22 0.0026782 64.16 07.13 03.96 11.09 24.74 0.88 



34 
 

 

5.2.3.4 Permeable pavements  

The permeable pavements chosen for this work was the porous asphalt with an 

application in the 50% rate of all the pavement. Even in some cases with the functionality of 

infiltration as flood attenuation, still has a positive impact, in this case was used the total 

infiltration type (WOODS BALLARD et al., 2015), with the details data in the Table 16. 

Table 16 - Permeable pavement data. 

Loc
al 

Are
a 

H4 (Permeable pavement) 

Permeable 
pavement 

area 

SC - 
Roof 
Area 

Proporti
on (%) 

SC - 
Pervious 

Area 
Proporti
on (%) 

SC - Pavement & Road Area Proportion 
(%) (permeable pavement/pavement/total 

SC - 
Runoff 

Coefficie
nt 

Permeable 
pavement 

Pavement 
& Road Total  

1 0.00474 68.47 13.01 9.26 9.26 18.52 0.84 

2 0.045435 72.33 9.39 9.14 9.14 18.28 0.86 

3 0.00886 74.22 8.25 8.77 8.77 17.53 0.87 

4 0.016100105 0.00 52.83 23.59 23.59 47.17 0.55 

5 0.002748864 0.00 87.08 6.46 6.46 12.92 0.40 

6 0.001134626 0.00 95.51 2.25 2.25 4.49 0.37 

7 0.028355 78.57 9.02 6.21 6.21 12.42 0.87 

8 0.006835 74.90 9.52 7.79 7.79 15.59 0.87 

9 0.030615 75.20 9.42 7.69 7.69 15.38 0.87 

10 0.022535 68.38 14.58 8.52 8.52 17.04 0.83 

11 0.0022985 65.42 11.71 11.44 11.44 22.87 0.84 

12 0.03762 74.92 9.75 7.66 7.66 15.33 0.86 

13 0.011835 74.41 8.70 8.44 8.44 16.89 0.87 

14 0.012555 73.82 8.40 8.89 8.89 17.78 0.87 

15 0.0310615 72.46 10.43 8.55 8.55 17.11 0.86 

16 0.04643 80.38 1.37 9.13 9.13 18.26 0.91 

17 0.0011015 83.94 9.33 3.37 3.37 6.74 0.88 

18 0.007075 80.70 1.58 8.86 8.86 17.72 0.91 

19 0.0069 75.32 8.37 8.16 8.16 16.32 0.87 

20 0.006895 79.82 4.04 8.07 8.07 16.14 0.90 

21 0.003527 82.91 6.26 5.42 5.42 10.83 0.89 

22 0.0083604 64.16 11.09 12.37 12.37 24.74 0.84 



35 
 

 

5.2.3.5 Green Spaces 

For the green spaces, all the available areas were used, and applied the scheme 

corresponding to the Figure 12. In this system the water will infiltrate, and this type of SUDs 

can be applied in many places due to variety of applicable surfaces materials (WOODS 

BALLARD et al., 2015). 

Figure 12- Dry ponds scheme. 

 
Source: (WOODS BALLARD et al., 2015). 

The data used in this hypothesis are shown in the Table 17. 

Table 17 – Dry ponds data. 
(continue) 

Local 
Area 

H4 (Infiltration trenches) 

Green areas 

SC - Roof 
Area 

Proportion 
(%) 

SC - Pervious Area 
Proportion (%) 

SC - 
Pavement & 
Road Area 
Proportion 

(%) 

SC - Runoff 
Coefficient Public 

green 
areas 

Private 
green 
areas 

Total 

1 0.002766 68.47 5.40 7.61 13.01 18.52 0.86 

2 0.0067343 72.33 1.35 8.04 9.39 18.28 0.89 

3 0 74.22 0.00 8.25 8.25 17.53 0.90 

4 0.0360625 0.00 52.83 0.00 52.83 47.17 0.55 

5 0.03704 0.00 87.08 0.00 87.08 12.92 0.30 

6 0.048231 0.00 95.51 0.00 95.51 4.49 0.23 

7 0.0013173 78.57 0.29 8.73 9.02 12.42 0.90 

8 0.00104641 74.90 1.19 8.32 9.52 15.59 0.89 

9 0.004258 75.20 1.07 8.36 9.42 15.38 0.89 

10 0.0184751 68.38 6.98 7.60 14.58 17.04 0.85 

11 0.0008923 65.42 4.44 7.27 11.71 22.87 0.87 

12 0.0070183 74.92 1.43 8.32 9.75 15.33 0.89 



36 
 

 

(continuation) 

13 0.0006061 74.41 0.43 8.27 8.70 16.89 0.90 

14 0.0002799 73.82 0.20 8.20 8.40 17.78 0.90 

15 0.0086326 72.46 2.38 8.05 10.43 17.11 0.88 

16 0.0069442 80.38 1.37 0.00 1.37 18.26 0.94 

17 0 83.94 0.00 9.33 9.33 6.74 0.89 

18 0.0012607 80.70 1.58 0.00 1.58 17.72 0.94 

19 0 75.32 0.00 8.37 8.37 16.32 0.90 

20 0.0034513 79.82 4.04 0.00 4.04 16.14 0.92 

21 0.0040761 82.91 6.26 0.00 6.26 10.83 0.90 

22 0.0026782 64.16 3.96 7.13 11.09 24.74 0.88 

 

5.2.3.6 Storage Tanks and runoff reservoir (Centralized system) 

Similarly, the runoff reservoir will used the structure proposed for storage tanks and 

rain barrels. Will be applied the value per capita for all residences and commerce volume of 

daily minimum demand as shown in the Table 14, and, as the centralized system usually has 

a considerable volume of water, will be used a water pump to back to the water system this 

water. The water pump used for this simulation was selected according to those ones used 

by the municipality for the correction of the systems, with a consume of 0,014kWh.m-3 

(PMB, 2019), with can be allocated for public non potable uses. 

5.2.3.7 Multivariate analysis 

The Absolute value generated by the Watermet² allow us to quantify the impact of 

the hypothesis in the KPIs, and, with these values use a multicriteria decision analysis 

(MCDA) to rank the better ones.  

The KPIs will be divided by the fluxes (equally) following the structure and weights 

described in the Figure 19. 

Table 18 - Criteria weights. 

Criteria Criteria weight Sub criteria Sub criteria weight 

Water Balance (m³) 12.50% 

TWD 4.17% 

Total leakage 4.16% 

FWDD 4.17% 

Sewer system balance (m³) 12.50% 
Excess stormwater 2.50% 

Excess Wastewater 2.50% 



37 
 

 

TCSO 2.50% 

TfrWWTW 2.50% 

Storm tank overflow 2.50% 

Energy (kWh) 12.50% Total Energy 12.50% 

GHG emissions (ton CO2-eq) 12.50% Total GHG emissions 12.50% 

Acidification (ton SO2-eq) 12.50% Total Acidification 12.50% 

Eutrophication (ton PO4-eq) 12.50% Total Eutrophication 12.50% 

Contaminants (kg) 12.50% 

COD 4.17% 

Phosphorous 4.17% 

Nitrogen 4.16% 

Sludge (ton) 12.50% Total Sludge generation 12.50% 

The compromise programming (CP) comes as a solution to bring the better situation 

between options to achieve an ideal one (YU; ZELENY, 1974). The method consists of the 

metric distance of the solutions propose for an ideal point, with the objective of achieve an 

ideal solution using the decision makers preference (CECAGNO et al., 2019; BEHZADIAN; 

KAPELAN, 2015b). Due to the progressive achievement of the objectives and considering 

that the ideal point is extremely difficult to reach, for this work will be used a nearest one, 

called goal point (CECAGNO et al., 2019). 

The discrepancy of values and units of the KPIS show a necessity to a 

standardization of the values, to future comparison and interrelation be facilized 

(CARVALHO, 2006). This normalization will be done according to the equation 1, that 

represents a direct proportion transforming the values between zero and one (CANALES, 

2009).  𝒇𝒊′(𝒙) = 𝒇𝒊𝒃𝒆𝒔𝒕(𝒙)− 𝒇𝒊(𝒙)𝒇𝒊𝒃𝒆𝒔𝒕(𝒙)−𝒇𝒊𝒘𝒐𝒓𝒔𝒕(𝒙)         ( 1 ) 

Where: fi’(x) is the standard calculated value of the “x” alternative in the “i” 

criteria; fi(x) is the calculated value of the “x” alternative in the “i” criteria; fibest(x) is the 

best calculated value of the “x” alternative in the “i” criteria; and fiworst(x) is the worst 

calculated value of the “x” alternative in the “i” criteria (CANALES, 2009). 

In general context the distance of the ideal point to the solution point, proposed can 

be described as (GERSHON; DUCKSTEIN, 1983):  

𝑳𝒑(𝒙) = [∑ ∝𝒊𝒑𝒏𝒊=𝟏 | 𝒇𝒊𝒃𝒆𝒔𝒕(𝒙)− 𝒇𝒊(𝒙)𝒇𝒊𝒃𝒆𝒔𝒕(𝒙)−𝒇𝒊𝒘𝒐𝒓𝒔𝒕(𝒙)|𝒑]𝟏𝒑
      ( 2 ) 



38 
 

 

Where: Lp(x) is the distance in compromise programming; ∝𝑖𝑝 the weights for each 

criterion; p is the weight of the maximum deviation of the mathematic function; and, the rest 

already described in the equation 1; the p value will be considered 2 because of the 

mathematic adjustments to consider the Euclidean distance (GERSHON; DUCKSTEIN, 

1983). 

The ranking with the best alternatives, considering an equal weight of the fluxes 

criteria will consider the Euclidean distance relative to the compromise programming.  

5.2.4 Watermet² Processing 

The Watermet² processing was made by firstly the completion of the conventional 

system values in the software and the key performance indicators (KPIs) generation, also 

used in the sensitivity analysis (5.2.1.2 item). After the KPI’s generation of the conventional 

system, the same was made for all the seven hypotheses. 

Each alternative must fill in the software the next gaps to differentiate between the 

urban drainage systems, those ones also used in the sensitivity test and shown as influence 

variable in the APPENDIX J, the table shows the Watermet² filling variables for each 

strategy. 

Even the sensitivity test and the case study analysis in this step was made by short-

term planning horizon, with the accumulate values over the short term (five years, (SMDU, 

2012)), using an annual time-step to the graphics construction for each KPI, and to compare 

between then only used the values in the 5th year.  

According to the objective of this study as shown the impact of changes in the urban 

drainage system in the water metabolism key performance indicators, for the time step plot 

was selected the “Urban water system” component, to considerate as an integrated system, 

and then generated with the absolute values. 

 

 

 

 



39 
 

 

 

6 RESULTS AND DISCUSSION 

6.1 SENSITIVITY RESULTS 

6.1.1 Normalized results 

The model brings the results through time, enable a short-, medium- and long-term 

analysis. Preliminarily the results of the first input to sensitivity evaluate it is shown below, 

the constant KPIs results are shown in Table 19.  

Table 19 - Constant results of KPIs in UWS. 

KEY PERFORMANCE INDICATORS VALUE UNIT 

Water Balance 
 

Total Water Demand 2604.3840 M³ 
Total delivered Water Demand 642.4110 M³ 
Total Potable Water Demand 2604.3840 M³ 

Total delivered Potable Water Demand 642.4110 M³ 
Total undelivered Potable Water Demand 1961.9730 M³ 

Total Leakage 248.5890 M³ 
Potable Domestic Water Demand 2.604.0936 M³ 
Potable Industrial Water Demand 0.2880 M³ 

Fraction of water 
demand delivered Fraction of Water Demand Delivered 0.2467 

- 

The software demonstrate that can show the results with the minimum of inputs, 

which prove that we can run a simulation in any size of UWS. 

The variables results are shown in the Figure 13 for Sewer system balance, Figure 

14 for Energy, Figure 15 for GHG emissions, Figure 16 for Acidification, Figure 17 for 

Eutrophication,  for Contaminant Load KPIs and for Sludge Generation.  

The Water Balance shown the structure capacity of the system, but how the 

configuration used for the sensitivity analysis inputs are assumed as the literature sizing, this 

number will be useful only in the future analysis; the same happens to Sewer System 

Balance.  



40 
 

 

Figure 13 - SA: Sewer System Balance in UWS. 

 

Source: Results generated by Watermet² model. 

The total energy consumption, shown in Figure 14, demonstrated an oscillation of 

usage surround 138000 kWh to 143400 kWh per month, mainly from the Electricity energy 

with only 0,8% of embodied energy, which comes from the chemical products used in the 

processes and 0,4% of Fossil fuel energy.  

Figure 14 - SA: Energy in the UWS. 

 

Source: Results generated by Watermet² model. 

The total of GHG emissions are represented in the Figure 15, has a similar behavior 

as the energy consume, where 96% of the GHG emitted is from electricity and embodied 

emissions, usually from chemicals used and pipelines materials.  



41 
 

 

Figure 15 - SA: GHG emissions in the UWS. 

 

Source: Results generated by Watermet² model. 

We can see that even for energy and GHG emissions, both has huge influence of 

the electricity consumed by the system, which highlight the specific point that need changes 

or improvement. The medium electricity demand of 140700 kWh and the emissions of 29.5-

ton CO2 eq per month, are equivalent to use of seven houses for one year and the equivalent 

of 26289 gallons of diesel consumed, respectively (EPA, 2022). 

The total acidification of the system is shown in the Figure 16, which oscillated 

surround 0,07 Ton SO2-eq per month. 

Figure 16 - SA: Acidification in UWS. 

 
Source: Results generated by Watermet² model. 

 



42 
 

 

The eutrophication is demonstrated in Figure 17, is most influenced by the ammonia 

(NH3) and Phosphorous.  

Figure 17 - SA: Eutrophication in UWS. 

 

Source: Results generated by Watermet² model. 

The contaminant load is demonstrated in Figure 18, which demonstrate that most 

part of the contaminants are generated by the sewer system in the normalized area.  

Figure 18 - SA: Contaminant Load in UWS. 

 

Source: Results generated by Watermet² model. 



43 
 

 

And the last KPIs, the sludge generation is shown in the Figure 19, that will be more 

detailed in future analysis.  

Figure 19 - SA: Sludge generation in the UWS. 

 
Source: Results generated by Watermet² model. 

In all of these graphics for the normalized area, we can observe a cyclicality in the 

values in annually marks, this can be explained by the relation between all these calculations 

with de subcatchment demand in the areas, which was assumed with no population variation 

due to the area been already highly occupied, and for better understanding and comparation. 

Based in the similarity of the results behavior inside the KPIs tables and graphics, 

the next analysis in this work will consider only the total values (Total Eutrophication, Total 

Acidification, Total GHG emissions, Total Energy consumed and Total of contaminants) 

and the single results (Sludge Generation), and in the multiples ones the relevant for this 

work was selected (Total water demand, Total leakage, Fraction of water delivered, Excess 

Storm water and Wastewater, Total combined sewer overflow (CSO), Treated overflow from 

WWTWs, and Storm Tank Overflow).  

6.1.2 Principal component analysis 

The short-term sensitivity simulation results are shown in the APPENDIX C, which 

corresponds the results for each KPIs under the disturbance application.  



44 
 

 

After the application in the PAST software, the summary table shown that the PC 

1, 2 and 3 corresponds to 98,2 % of the accumulate variance proportion of data, being the 

most influence ones. 

Table 20 - Percentual of the principal components of the sensitivity data. 

PC Eigenvalue % of Variance 

1 18.7824 78.2600 

2 3.2530 13.5540 

3 1.5343 6.3931 

4 0.3737 1.5570 

5 0.0546 0.2275 

6 0.0012 0.0050 

7 0.0007 0.0031 

8 0.0000 0.0000 

9 0.0000 0.0000 

10 0.0000 0.0000 

11 0.0000 0.0000 

12 0.0000 0.0000 

13 0.0000 0.0000 

14 0.0000 0.0000 

15 0.0000 0.0000 

The detailed influence of variable for each principal component is shown in the 

APPENDIX D, which demonstrate that the most influence variables for the process are the 

Irrigation and other water demand, Frost tapping, Industrial and Commercial demand, 

Rehabilitation, Electricity, Fossil Fuel, Treatment Capacity, Storage Capacity, Electricity 

and Fossil Fuel for PC1; Number of properties, Indoor water demand, occupancy, Infiltration 

coefficient and Exfiltration for PC2; and, Total area for PC3. The only variable that does not 

influence on the principal component and consequently not influence on KPIs is the 

Transmission capacity, which show that almost all the analyzed variables are important to 

the process. Therefore, for the study case analysis, all the variables were carefully filled 

according to the hypothesis.  

6.2 STUDY CASE RESULTS 

6.2.1 Bauru catchment Results 

The results obtained in the software for the study case strategy, that is the same as 

the conventional system hypothesis, considering that is the only system applied in the 



45 
 

 

catchment, can be seen in the Table 21. The table represents the absolute value of the key 

performance indicators in the intervention that will be used to comparison. For a validation 

of these results was made a comparation between some model results and external quantify 

information.  

Table 21 - Bauru catchment results. 

KPIs Conventional system results 

Water Balance (m³) 

Total Water Demand 16692487.659 

Total leakage 5109239.577 

FWDD 0.306 

Sewer system balance (m³) 

Excess stormwater 0.000 

Excess Wastewater 0.000 

TCSO 5498693.828 

TfrWWTWs 0.000 

Storm tank overflow 20295120.256 

Energy (kWh) Total Energy 24504978.191 

GHG emissions (ton CO2-eq) Total GHG emissions 5894.686 

Acidification (ton SO2-eq) Total Acidification 16.993 

Eutrophication (ton PO4-eq) Total Eutrophication 9698.500 

Contaminants (kg) 

COD 476982.785 

Phosphorous 3113041.417 

Nitrogen 1330832.838 

Sludge (ton) Total Sludge generation 38438.844 
Observations: 
FWDD – Fraction of water demand delivered. 
TCSO – Total of combined sewer overflow. 
TfrWWTWs – Treated outflow from wastewater treatment works. 
COD – Chemical oxygen demand. 

The administration of the urban water supply of the Bauru city, Water, and 

wastewater department (DAE) made a report with the 2013 data from the water supply and 

wastewater system. In the 2013 year the department produced an amount of 77380m³/day of 

the total of 190279m³/day consumed by the catchment, an amount of 0,40 of fraction of 

water demand delivered from the department. Also, according to the private and public map 

of wells from the document, the city’s suburb is more supplied by the public department and 

the center areas from particular fonts (not considered in this study), showing that this fraction 

of 0,4 it’s even smaller (DAE; HIDROSAN, 2014c), corresponding to the value of 0,306 of 

fraction of water demand delivered from public sources calculated in the Watermet² 

software.   

The DAE document brings the total of 190279m³/day of water demand, the 

equivalent to 0,50m³/day per capita, value compared with the 788,83 m³ per capita in five 



46 
 

 

years, equivalent to 0,44m³/day per capita, calculates by Watermet² validate the total od 

water demand, and consequently the water balance of the system.  

The values of water balance are inside the range of normal values, with the total 

demand of 0,44m³/day per capita, equivalent to 440l/day is considered a normal value due 

to the lost in Bauru’s urban water system are usually next to 50% of the produced water.  

According to the city sanitation plan, the intention it is in the future install a 

wastewater treatment w