A Decision-making Framework for Hybrid Resource Recovery

University of South Florida University of South Florida

Scholar Commons Scholar Commons

Graduate Theses and Dissertations Graduate School

June 2019

A Decision-making Framework for Hybrid Resource Recovery A Decision-making Framework for Hybrid Resource Recovery

Oriented Wastewater Systems Oriented Wastewater Systems

Nader Rezaei University of South Florida, [email protected]

Follow this and additional works at: https://scholarcommons.usf.edu/etd

Part of the Environmental Engineering Commons, Other Environmental Sciences Commons, and the

Water Resource Management Commons

Scholar Commons Citation Scholar Commons Citation Rezaei, Nader, "A Decision-making Framework for Hybrid Resource Recovery Oriented Wastewater Systems" (2019). Graduate Theses and Dissertations. https://scholarcommons.usf.edu/etd/7907

This Dissertation is brought to you for free and open access by the Graduate School at Scholar Commons. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Scholar Commons. For more information, please contact [email protected].

http://scholarcommons.usf.edu/

http://scholarcommons.usf.edu/

https://scholarcommons.usf.edu/

https://scholarcommons.usf.edu/etd

https://scholarcommons.usf.edu/grad

https://scholarcommons.usf.edu/etd?utm_source=scholarcommons.usf.edu%2Fetd%2F7907&utm_medium=PDF&utm_campaign=PDFCoverPages

http://network.bepress.com/hgg/discipline/254?utm_source=scholarcommons.usf.edu%2Fetd%2F7907&utm_medium=PDF&utm_campaign=PDFCoverPages



mailto:[email protected]

A Decision-making Framework for Hybrid Resource Recovery Oriented Wastewater Systems

by

Nader Rezaei

A dissertation submitted in partial fulfillment of the requirements for the degree of

Doctor of Philosophy in Environmental Engineering Department of Civil and Environmental Engineering

College of Engineering University of South Florida

Major Professor: Qiong Zhang, Ph.D. Mahmood Nachabe, Ph.D. E. Christian Wells, Ph.D.

Nancy Diaz-Elsayed, Ph.D. Hadi Charkhgard, Ph.D.

Date of Approval: June 21, 2019

Keywords: Life cycle assessment, Life cycle cost analysis, Water reclamation, Optimization, System modeling

Copyright © 2019, Nader Rezaei

DEDICATION

This dissertation is dedicated to my parents, Zarrin Navi and Bijan Rezaei, for their

tremendous love, patience, and support throughout the years of my Ph.D. study.

ACKNOWLEDGEMENTS

I would like to thank my advisor, Dr. Qiong Zhang for her guidance, support, and

mentorship throughout my time at the University of South Florida (USF). I would also like to thank

my committee members, Dr. Mahmood Nachabe, Dr. Christian Wells, Dr. Nancy Diaz-Elsayed,

and Dr. Hadi Charkhgard for their advice and feedback on my research.

This material is based upon work supported by the U.S. National Science Foundation

CAREER Award (No. 1454559). Any opinions, findings, and conclusions or recommendations

expressed in this material are those of the author and do not necessarily reflect the views of the

National Science Foundation.

i

TABLE OF CONTENTS

LIST OF TABLES iii

LIST OF FIGURES vi

ABSTRACT viii

CHAPTER 1: INTRODUCTION 1 1.1 Background and Significance 1 1.2 Scope of Research 13

CHAPTER 2: A MULTI-CRITERIA SUSTAINABILITY ASSESSMENT OF WATER REUSE APPLICATIONS 17

2.1 Introduction 17 2.2 Materials and Methods 21

2.2.1 Study Area 21 2.2.2 Scenario Generation and Design 24 2.2.3 Indicator Description and Quantification 29

2.2.3.1 Economic Indicator 29 2.2.3.2 Environmental Indicators 30 2.2.3.3 Social Indicator 32

2.2.4 Scenario Evaluation 32 2.2.5 Location and Treatment Analysis for DPR 34

2.3 Results and Discussion 35 2.3.1 Tradeoffs for Water Reuse Management 35 2.3.2 Decentralized VS. Centralized Reuse and Treatment 39 2.3.3 Multi-Criteria Decision-Making 40 2.3.4 Sensitivity Analysis for DPR 43 2.3.5 Limitations and Future Work 44

2.4 Conclusion 45

CHAPTER 3: A MULTI-OBJECTIVE OPTIMIZATION MODEL FOR WATER RECLAMATION PLANNING 47


3.2.1 Deterministic Optimization Model 52 3.2.2 Case Study Background 55 3.2.3 Scope and Input Data 57 3.2.4 Degree of Decentralization and Marginal Benefit Calculations 59 3.2.5 Sensitivity Analysis 60

ii

3.3 Results and Discussion 60 3.3.1 Decentralization of Treatment 66 3.3.2 Selection of Treatment Technology 67 3.3.3 Sensitivity Analysis 69 3.3.4 Consideration of Social Indicators 73 3.3.5 Limitations 77

3.4 Conclusion 78

CHAPTER 4: IMPACTS OF EXTERNAL VARIABLES ON THE DESIGN OF RESOURCE RECOVERY WASTEWATER SYSTEMS 80


4.2.1 Deterministic Optimization Model 90 4.2.2 Hypothetical Scenarios 91 4.2.3 Scope and Input Data 94 4.2.4 Degree of Decentralization and Marginal Benefit Calculations 96

4.3 Results and Discussion 96 4.3.1 Impacts of Population Density 105 4.3.2 Impacts of Elevation Variation 106 4.3.3 Limitations 108

4.4 Conclusion 110

CHAPTER 5: CONCLUSIONS AND RECOMMENDATIONS 112 5.1 Summary 112 5.2 Research Limitations and Future Opportunities 117

5.2.1 Input Data 117 5.2.2 Sustainability Indicators 118 5.2.3 Improvement of the Developed Models and Frameworks 119

REFERENCES 121

APPENDIX A: NOMENCLATURE 137

APPENDIX B: SUPPLEMENTARY MATERIAL FOR CHAPTER 2 139

APPENDIX C: SUPPLEMENTARY MATERIAL FOR CHAPTER 3 154

APPENDIX D: SUPPLEMENTARY MATERIAL FOR CHAPTER 4 184

APPENDIX E: COPYRIGHT PERMISSION 195

ABOUT THE AUTHOR END PAGE

iii

LIST OF TABLES

Table 1.1 Results of the literature review 6

Table 2.1 Summary of information related to each scenario in this study 28

Table 2.2 Results for the regret-based model and the calculated regret score for each scenario 41

Table 3.1 Review of previously developed optimization models related to the study area 50

Table 3.2 Constraints associated with the optimization model 55

Table 3.3 Impacts of changes in the energy requirement for water transfer on the optimal solution with the highest marginal benefit (costs, environmental footprint, degree of decentralization, and degree of treatment) 71

Table 3.4 Impacts of changes in population density on the optimal solution with the highest marginal benefit (costs, environmental footprint, degree of decentralization, and degree of treatment) 73

Table 3.5 Demographic, economic, housing, and environmental health data related to the candidate sites for additional social considerations (Florida Brownfields Redevelopment Atlas, 2019) 76

Table 4.1 Limitations of the previous studies on evaluation of water reclamation system 88

Table B.1 Effluent water quality from Glendale water reclamation facility/wetland and water quality standards associated with each reuse scenario 140

Table B.2 Scope of line items included in calculation of capital and O&M costs for each scenario 141

Table B.3 Additional information used for design of scenario 1 142




iv




Table B.10 Total capital cost, O&M costs and production benefit for scenario 1 149

Table B.11 Total capital cost, O&M costs for scenario 2 150






Table B.17 Results for calculation of economic indicators for each reuse scenario 153

Table C.1 Information regarding the population clusters in the study area 155

Table C.2 Information related to the candidate locations for the new wastewater treatment facility 155

Table C.3 Information related to the types of treatment technology used in the model 156

Table C.4 Input information related to the treatment technologies selected for the model 156

Table C.5 Other input parameters used for the optimization model 157

Table C.6 Results obtained from solving the first stage of the multi-objective optimization model 158

Table C.7 Economic and environmental footprints associated with the optimal solutions 171

Table C.8 The detailed information regarding the least expensive solution (solution 1), the most environmentally friendly solution (solution 50), and the solution with the highest marginal benefit (solution 24) 172

Table C.9 Results obtained from the sensitivity analysis 173

v

Table D.1 Information related to the types of treatment technology used in the model 184

Table D.2 Input information related to the treatment technologies selected for the model 184

Table D.3 Other input parameters used for the optimization model 185

Table D.4 Results of the optimization model for scenario Low P Low E 186

Table D.5 Results of the optimization model for scenario Medium P Low E 187

Table D.6 Results of the optimization model for scenario High P Low E 188

Table D.7 Results of the optimization model for scenario Low P Medium E 189

Table D.8 Results of the optimization model for scenario Medium P Medium E 190

Table D.9 Results of the optimization model for scenario High P Medium E 191

Table D.10 Results of the optimization model for scenario Low P High E 192

Table D.11 Results of the optimization model for scenario Medium P High E 193

Table D.12 Results of the optimization model for scenario High P High E 194

vi

LIST OF FIGURES

Figure 1.1 Conventional reverse logistics (a) compared to its application for integrated wastewater management (b) 13

Figure 1.2 The integrated wastewater systems management in this study 16

Figure 1.3 The summary of two major approaches for decision-making framework development 16

Figure 2.1 Summary of the current water, wastewater and reclaimed water cycle in the City of Lakeland, Florida 22

Figure 2.2 Overview of the scenarios considered in the study 29

Figure 2.3 Annualized specific net present value (ASNPV) and value of resource recovery (VRR) for different reuse scenarios, based on a design life time of 33 years 36

Figure 2.4 Environmental impacts (carbon footprint [CFP] and eutrophication [EU]) associated with different reuse scenarios 37

Figure 2.5 Results for the regret-based model and the calculated regret score for each scenario 42

Figure 2.6 Location and treatment analysis for direct potable reuse (DPR) scenario 44

Figure 3.1 Similarities and differences between reverse logistics in (a) production activities, (b) municipal solid waste management, and (c) water systems 49

Figure 3.2 Hillsborough County boundary, County’s water service areas, the County's major cities, and the candidate locations for implementation and operation of WWTPs in the Southcentral water service area 57

Figure 3.3 Optimal solutions for the multi-objective optimization model for the study area 61

Figure 3.4 Configuration of the least expensive solution (solution 1), the most environmentally friendly solution (solution 50), and the solution with the highest marginal benefit (solution 24) 62

vii

Figure 3.5 Comparison of two selected treatment capacity in (a) medium-scale (2.0 MGD) and (b) large-scale (10.0 MGD) for the designed treatment trains in the study 64

Figure 3.6 Breakdown of the contributors to the (a) first and (b) second objective function 65

Figure 4.1 Nine different hypothetical scenarios generated for this study 93

Figure 4.2 The optimal solutions for the multi-objective optimization model for the hypothetical scenarios 98

Figure 4.3 The information regarding the optimal solutions with highest marginal benefit for each scenario 100

Figure 4.4 Information regarding the treatment trains in small, medium, and large scale, designed for this study 101

Figure 4.5 The breakdowns of the costs (the first objective function) associated with the optimal solutions with the highest marginal benefit 103

Figure 4.6 The breakdowns of the environmental impacts (the second objective function) associated with the optimal solutions with the highest marginal benefit 104

Figure B.1 The location of main water and wastewater infrastructure in the city of Lakeland, Florida 139

Figure B.2 Location of golf courses for scenario 1 142

Figure B.3 Location of strawberry irrigation lands for scenario 2 143

Figure B.4 Location of injection wells for scenario 3 144

Figure B.5 Location of water treatment plant for scenario 4 145

Figure B.6 Additional treatment trains designed for scenario 4 using IT3PR toolbox 145

Figure B.7 Location and pipeline required for implementation of each scenario 147

Figure B.8 Location of five decentralized medium-scale WWTPs for scenario 7 148

Figure C.1 Hillsborough County water service areas, current WWTPs location, and candidate locations for the new water reclamation facilities 154

Figure C.2 Environmental marginal benefits associated with the optimal solutions 172

viii

ABSTRACT

Water shortage, water contamination, and the emerging challenges in sustainable water

resources management (e.g., the likely impacts of climate change and population growth)

necessitate adopting a reverse logistics approach, which is the process of moving wastewater from

its typical final destination back to the water supply chain for reuse purposes. This practice not

only reduces the negative impacts of wastewater on the environment, but also provides an

alternative to withdrawal from natural water resources, forming a closed-loop water supply chain.

However, the design of such a supply chain requires an appropriate sustainability assessment,

which simultaneously accounts for economic, environmental, and social dimensions. The overall

aim of this work was therefore to contribute to the literature by evaluating the impacts of water

reclamation and reuse according to the triple-bottom-line sustainability indicators (i.e., economic,

environmental, and social) and to develop frameworks and mathematical models to help decision-

makers, stakeholders, and officials with the design of sustainable water reclamation and reuse

systems. The applicability of the developed frameworks and models was examined using real case

studies and hypothetical scenario analyses. This enactment also revealed the tradeoffs and

thresholds associated with the design of sustainable water reclamation and reuse systems.

To conquer the mentioned goal, the research was conducted in three major sections. The

first part of the research was outlined to design possible scenarios for water reuse based on water

reuse guidelines and evaluate the different types of end-use based on the three major dimensions

of sustainability (i.e., economic, environmental and social aspects), simultaneously. The different

reuse types considered include unrestricted urban reuse, agricultural reuse, indirect potable reuse

ix

(IPR), direct potable reuse (DPR), distributed unrestricted urban reuse, as well as some degree of

decentralization of treatment plants for distributed unrestricted urban reuse. The tradeoff

investigation and decision-making framework were demonstrated in a case study and a regret-

based model was adopted as the support tool for multi-criteria decision-making. This study

revealed that although increasing the degree of treatment for water reuse required implementation

of advanced treatment options and it increased the implementation, operation, and maintenance

(O&M) costs of the design, it increased the value of resource recovery significantly, such that it

can offset the capital and O&M costs associated with the treatment and distribution for DPR.

Improving the reclaimed water quality also reduced the environmental footprint (eutrophication)

to almost 50% for DPR compared to the other reuse scenarios. This study revealed that the distance

between the water reclamation facility and the end use plays a significant role in economic and

environmental (carbon footprint) indicators.

In the second part of this research, a multi-objective optimization model was developed to

minimize the costs and environmental footprint (greenhouse gas emissions), and maximize social

benefits (value of resource recovery) of the water reclamation systems by locating the treatment

facility, allocating treatment capacity, selecting treatment technology, and allocating customers

(final reclaimed water users). The expansion of the water reclamation system in Hillsborough

County, Florida was evaluated to illustrate the use of the model. The impacts of population density

and topography (elevation variation) of the water service area on the model outputs were also

investigated. Although the centralization of treatment facilities takes advantage of the economies

of scale, the results revealed that simultaneous consideration of economic and environmental

indicators favored decentralization of treatment facilities in the study area. This was mainly due to

the significant decrease in water transfer requirements, especially in less populous areas.

x

Moreover, the results revealed that contribution of population density to the optimal degree of

decentralization of treatment facilities was significant.

In the last part of this work, hypothetical scenarios for a water service area were generated

to evaluate the impacts of external variables on the design of water reclamation and reuse systems.

Although the conducted sensitivity analyses in the previous sections revealed the tradeoffs and

thresholds associated with the design of water reclamation systems, the concept of a hypothetical

study helped with the elimination of case-specific factors and local conditions that could possibly

influenced the outcomes. These factors, which were specific to the case studies (e.g., the location

of candidate sites for implementation of water reclamation facilities and special population

distribution patters) made barriers to the conclusions and hurdled the interpretation of findings.

Two major factors, which were found to be significant among the factors influencing the design

of water systems (i.e., elevation variation and population density), were selected for the evaluation.

Accordingly, three different topographies (i.e., flat region, medium elevation variation, and hilly)

and three types of population density (i.e., low, medium, and high) were considered for the design

of hypothetical cases and the previous model developed in the second section was modified and

used to evaluate the impacts. The results revealed that although decentralization of water

reclamation facilities decreases the costs and environmental impacts associated with water transfer

phase (i.e., wastewater collection and reclaimed water distribution), there were tradeoffs between

the impacts of decentralization of treatment plants and the benefits from economies of scale for

treatment. The results showed that when the population density is high and there is moderate to

high elevation variations in the water service area, decentralization of treatment facilities is the

beneficiary option. However, if the population density is low, economies of scale for treatment

becomes more influential and lower degrees of decentralization of treatment facilities is preferred.

1

CHAPTER 1: INTRODUCTION

1.1 Background and Significance

Currently, there are more than 4 billion people (two-thirds of the world’s population)

experiencing moderate to severe water scarcity (Mekonnen and Hoekstra, 2016). The emerging

global challenges such as increasing demand due to population growth and industrialization, water

scarcity, contamination of natural water bodies, and the likely impacts of climate change

(Zimmerman et al., 2008) have made serious hurdles to overcome for sustainable management of

water and wastewater systems around the globe. In this regard, integrated management of

wastewater systems that facilitates and promotes resource recovery has become a vital solution to

the sustainable design of water systems (Cornejo et al., 2016). Traditionally, the main objective of

wastewater treatment was to protect the environment and public health by removal of contaminants

prior to releasing the treated wastewater back to the receiving water bodies (Gallego et al., 2008;

Hospido et al., 2004). In the past century, wastewater has been considered only as waste with

regulated treatment and subsequent discharge requirements. The traditional approach for

wastewater treatment primarily relies on centralized systems (Tchobanoglous and Leverenz,

2013), which take advantage of economies of scale for treatment and better operation condition

and quality control due to centralized monitoring and management (Maurer et al., 2005). Although

these centralized treatment systems reduce the negative impacts of wastewater on natural

ecosystems and the environment (Morera et al., 2016), their performance can be criticized due to

high energy and chemical requirements during the treatment and water transfer phases (Godin et

al., 2012).

2

Recently, the sustainability of centralized wastewater systems are facing some other

challenges such as higher requirements associated with water transfer phase (i.e., wastewater

collection and reclaimed water distribution) (Maurer et al., 2005), presence of emerging

contaminants and pharmaceuticals in the produced wastewater (Tidåker et al., 2007), potential for

eutrophication of the receiving water bodies due to the high concentration of nutrients in the

effluent, diseconomies of treatment phase due to wastewater dilution (Ho, 2005), high energy and

water requirements during the operation phase (Bakir, 2001; Means, 2004), and high vulnerability

to terrorism and natural disasters (Wilderer and Schreff, 2000). The mentioned external stressors

(e.g., water shortage and population growth), the internal stressors (e.g., aging infrastructure)

(Copeland and Tiemann, 2010) and increasing the energy cost (Means, 2004), as well as the

emerging criticisms on the sustainability of traditional approaches for water and wastewater

systems management (Copeland and Tiemann, 2010), necessitate adopting new approaches to the

management of wastewater systems, as a vital component of an integrated water network, that

emphasizes resource recovery (e.g., water, energy, and nutrient) over treatment (Capodaglio and

G., 2017; Guest et al., 2009; Larsen et al., 2009). This not only offsets some portion of the required

energy for the treatment phase, but also provides an alternative to the withdrawal from natural

water resources and increases the sustainability of water systems by closing the water supply chain

(Anastas, 2012).

The empirical concept of integrated water systems management by practitioners was rooted

in 1997 at the first global water conference in Argentina (Iacob, 2013). The concept was accepted

worldwide and a strategic development was taken after the 1992 World Summit on Sustainable

Development in Rio de Janeiro. As a part of the integrated water resources management within the

broader governance, water system management requires local application of the integrated water

3

system management, which aims to improve the sustainability of such systems by providing access

to water in the local level. This integrated urban water system includes water withdrawal from

natural water resources (e.g., lakes, groundwater, rivers), water treatment, water distribution

network, water consumption by end users (e.g. drinking water, urban irrigation, agriculture,

landscaping, recreational, and industrial), wastewater collection and transfer system (sewer

network), wastewater treatment plants (WWTPs), and reclaimed water distribution system. A

higher level of integrated water system management involves the connection and communication

between internal departments, communities, stakeholders, coordination structures, and

government administrations (Iacob, 2013). Considering the complexity of water and wastewater

networks, this novel approach for water systems management must include the design,

transformation, and management of different components of the system and integrate them into an

optimized configuration that renovates traditional wastewater networks to resource recovery

systems and creates a closed-loop water supply chain. Accordingly, many believe that the hybrid

form of wastewater infrastructure is the trend of future wastewater management (Daigger, 2009;

Daigger and Crawford, 2007; Tchobanoglous and Leverenz, 2013).

Closed-loop supply chain design is receiving growing attention for solving production and

demand problems in a variety of production activities (Ramezani et al., 2013). Considering the

fact that water can be considered as a product, the same notion can be applied to water and

wastewater systems. However, one primary challenge in understanding such a closed-loop water

supply networks can be lack of decision support tools, planning models, and established design

frameworks that help decision-makers with finding the most sustainable solutions for the design

and management of such systems under a given context (Guest et al., 2009). “Such methodology

or frameworks must include environmental, economic, and social dimensions; span spatial and

4

temporal scales; be flexible to consider different geographic and cultural contexts; be able to

involve broad stakeholders; and resolve tradeoffs in decision-making. To create this framework, a

system approach is needed, and how the problem is framed is critical. Learning from reverse

logistics in the discipline of systems engineering, an innovative framework can be created to

formulate and solve integrated wastewater management problems in environmental engineering.”

(Zhang, 2015).

During the last decade, the mentioned emerging challenges in water systems management

such as water shortage, increasing water demand, and water resources contamination, have

motivated the researchers to evaluate the sustainability of water systems for improvement

purposes. Some studies focus on different types of technology used for resource recovery to

evaluate the environmental impacts associated with them (El-Shafai et al., 2007; Forrest et al.,

2008; Funamizu et al., 2001; Gaius-obaseki, 2010; Leverenz et al., 2011; Liu et al., 2004; Martí et

al., 2010; Nouri et al., 2006; Novotny, 2010; Rabaey et al., 2003; US EPA, 2007; Van der Bruggen,

2010; Verstraete et al., 2009; Voltolina et al., 2005; Wang et al., 2008; Wilkie and Mulbry, 2002).

Some other studies evaluate the sustainability of water systems with a focus on water reclamation,

water reuse, as well as source separation technologies (Berndtsson, 2006; Larsen et al., 2009). In

recent decades, researchers have also taken advantages of Life Cycle Assessment (LCA) and Life

Cycle Cost Analysis (LCCA) tools to determine the impacts associated with water treatment, water

distribution, or wastewater treatment for reuse purposes in order to find the most sustainable

solutions during the decision-making processes (Björklund et al., 2001; Brown et al., 2010; Hong

et al., 2009; Hospido et al., 2005; Houillon and Jolliet, 2005; Lundin et al., 2004; Lyons et al.,

2009a; Maurer et al., 2003; Meneses et al., 2010; Nakakubo et al., 2012; Ortiz et al., 2007; Pant et

al., 2011; Pasqualino et al., 2011; Peters and Lundie, 2001; Sander and Murthy, 2010; Stokes and

5

Horvath, 2006; Suh and Rousseaux, 2002; Svanström et al., 2005; Tillman et al., 1998). Some

studies also considered the risk and public perception in their evaluation (Hamilton et al., 2006;

Hartley, 2006; Nancarrow et al., 2008; Po et al., 2003; Soller and Nellor, 2011; Verbyla et al.,

2013).

Amores et al. (2013) evaluated the environmental impacts associated with using the

reclaimed water for non-potable purposes such as irrigation in Spain. They showed that the reuse

scenarios reduced the freshwater consumption due to the net water savings, however, it did not

make any significant improvement in the environmental impacts and it was manily due to the extra

resources required for the tertiary treatment. Pasqualino et al. (2011) studied the environmental

profile of four wastewater treatment plants for different water reuse scenarios and they revealed

that using the reclaimed water for potable purposes preserved freshwater resources, but also

resulted in higher environmental impacts due to the additional processes required for advanced

treatment. Muñoz et al. (2009) designed four bench-scale treatment systems to evaluate the

environmental impacts of wastewater treatment for reuse via irrigation. The results showed that

wastewater reuse for irrigation with any of the studied tertiary treatment options had lower

ecotoxicity impacts compared to those scenarios without tertiary treatment. Meneses et al. (2010)

used LCA methods to evaluate the environmental advantages and disadvantages of reclaimed

water use for non-potable applications. The results showed that replacing desalinated water with

reclaimed water for non-potable purposes was beneficial when there was a freshwater scarcity.

Other studies analyzed the environmental impacts of urban water systems and mainly

focused on the treatment technologies (e.g., Beavis and Lundie, 2003; Lim and Park, 2009; Metcalf

et al., 2007). The study conducted by Lim and Park (2009) revealed that as the degree of treatment

increases, the cost and the negative environmental impacts associated with the treatment increases,

6

although they offset a portion of the need for withdrawal from freshwater resources. Although

several studies have been conducted on evaluation of different elements involved in the design of

wastewater systems (e.g., treatment technology, reuse alternatives, and alternative water

resources), lack of a systematic holistic evaluation framework for integrated water and wastewater

systems design and management is evidenced. Table 1.1 shows the results of the literature review

and a summary of the most relevant studies in this regard.

Table 1.1 Results of the literature review

Year Focus Method Study

2001 - 2011

Different types of technology used for resource recovery to

evaluate the associated environmental impacts

LCA, LCCA

Forrest et al., 2008; Wang et al., 2008; El-Shafai et al., 2007; Liu et al., 2004; Marti et al., 2010; Gaiusobaseki, 2010; Van der Bruggen, 2010; Verstraete et al., 2009; Voltolina et al., 2005; Funamizu et al., 2001; Leverenz et al., 2011; Novotny, 2010; Wilkie and Mulbry, 2002; Rabaey et al, 2003; Nouri et al., 2006; U.S. EPA, 2007

2006 - 2009

Evaluation of the sustainability of a water system with a focus on water reclamation, water

reuse, as well as source separation technologies

LCA Larsen et. al., 2009; Berndtsson, 2006

1998 - 2012

The impacts resulting from water treatment, water

distribution, or wastewater treatment for different types of

reuse

LCA, LCCA

Björklund et. al., 2001; Hospido et al., 2005; Suh and Rousseaux, 2002; Meneses et al., 2010; Hong et al., 2009; Houillon and Jolliet, 2005; Tillman et al., 1998; Ortiz et al., 2007; Houillon and Jolliet, 2005; Peters and Lundie, 2001; Pant et al., 2011; Pasqualino et al., 2011; Lundin et al., 2004; Maurer et al., 2003; Sander and Murthy, 2010; Stokes and Horvath, 2006; Svanström et al., 2005; Nakakubo et al., 2012; Brown et al., 2010; Lyons et al., 2009

2003 - 2013 Evaluation of risk and public perception

Hartley, 2006; Nancarrow et al., 2008; Verbyla et al., 2013; Po et al., 2003; Soller and Nellor, 2011; Hamilton et al., 2006; Soller and Nellor, 2011

2009 - 2013 Non-potable and potable reuse types

Amores et al., 2013; Munoz et al., 2009; Pasqualino et al., 2011; Meneses et al., 2010

2003 - 2008

Environmental impacts of urban water systems that

mainly focus on treatment technologies

Beavis and Lundie, 2003; Metcalf et al., 2007; Lim and Park, 2008

2001 The impacts of product recovery on logistics networks

Fleischmann et al., 2001

2004 - 2011 The reverse logistic approach that leads to the formation of a closed-loop water supply chain

Levine and Asano, 2004; Pereira et al., 2011; Habibi et al., 2017

7

Table 1.1 (Continued)

Year Focus Method Study

2017 The cost and environmental

impacts of the municipal solid waste management system

Mixed-integer programming formulation

Lyeme et al., 2017

2018 Designing a reverse logistic network for product recovery

Mixed integer linear programming model John et al., 2018

2010

Selection of a long-term strategy regarding the optimal flow and facility locations in a reverse logistic network design

Two-stage stochastic programming model Kara and Onut, 2010

2012 - 2017 Reverse logistic production network design

Mixed-integer and stochastic

programming models

Lieckens and Vandaele, 2012; Toso and Ahem, 2014; Lieckens et al., 2013; Fattahi and Govindan, 2017; Jeihoonian et al., 2016; Nakatani et al., 2017

2017 Decision-making support tool

for allocation of water resources

Linear integer programming model Abdulbaki et al., 2017

2016 Optimization of the water sources and water distribution

Multi-objective programming model Al-Zahrani et al., 2016

2008 Several possible scenarios

applied to a realistic hypothetical water system

System dynamic simulation Chung et al., 2008

2012 - 2015 Different technologies for the secondary treatment EDSS Molinos-Senante et al., 2012; Molinos-Senante

et al., 2015

2016 Most economical option for water source/supply

Cost management model Ruiz-Rosa et al., 2016

Traditional supply network design primarily relies on forward logistics and manufacturing

the final products using the raw materials. The reverse logistic networks, also known as a backward

or recovery network, include the process of returning the used products to the collection and repair

centers for remanufacturing and reuse purposes. The same notion can be applied to water networks:

wastewater can be diverted back to decentralized, satellite, or centralized wastewater treatment

systems, treated to be qualified for distribution and reuse, and conveyed to the customers for reuse

purposes. As it was mentioned before, considering the mentioned emerging challenges in water

systems management, reverse logistic approach for wastewater management not only helps with

reducing the negative impacts of wastewater on the environment, but also provides an alternative

to the withdrawal from natural resources by recovering valuable resources from the wastewater

(e.g., water, energy, and nutrients) and releases the pressure from natural water resources by

formation of a closed-loop water supply chain (Levine and Asano, 2004; Pereira et al., 2011). A

8

study conducted by Fleischmann et al. (2001) analyzed the impacts of product recovery on logistic

networks. They showed that the impacts of product recovery (e.g., economic factors,

environmental impacts, and regulatory compliance) are context-dependent and require an

individual comprehensive approach to evaluate the implementation of such systems in any type of

production activity. The increasing importance of sustainability of water systems necessitate

adoption of such a reverse logistics and product recovery approach, which is moving the treated

wastewater from its typical destination for some reuse purposes.

In the past century, researchers have focused on the application of reverse logistics in urban

planning and have developed models capable of improving the performance of the supply chain

networks by locating the facilities and allocating the system’s components to each other (Farhad

Habibi et al., 2017). Lyeme et al. (2016) developed a mixed-integer programming model to

minimize the cost and environmental impacts of the municipal solid waste management system in

Dar es Salaam city by finding the best location for implementation of the facilities (e.g., recycling

facilities, landfills, and incineration plants) as well as the waste flow allocation between the

facilities. The results showed a significant reduction in the greenhouse gas emissions and the

amount of waste to landfill in the new design (55.2% and 76%, respectively) compared to the

current system. John et al. (2018) designed a reverse logistic network for remanufacturing and

product recovery. They developed a mixed integer linear programming model, with consideration

of material recycling, component repair, and product remanufacturing. The economic benefit was

considered as the model’s objective. The results revealed that with allocation of parts return,

recovery, and reuse in the refrigerator production industry, total benefit can increase up to 23.2%

compared to a base case scenario. Kara and Onut (2010) also developed a two-stage stochastic

9

programming model with an application in the paper industry to select a long-term strategy

regarding the optimal flow and facility locations in a reverse logistic network design problem.

Some other studies developed mixed-integer and stochastic programming models for the

design of reverse logistic production networks to make a decision on: collection facilities

(Jeihoonian et al., 2016a, 2016b; Lieckens and Vandaele, 2012); the location, capacity, and flow

between nodes of remanufacturing facilities (Lieckens et al., 2013); collection and

remanufacturing facilities with discrete time intervals taken into consideration (Toso and Alem,

2014); rate of production and return of the used products (Fattahi and Govindan, 2017); and

optimal flows (Nakatani et al., 2017). The application of reverse logistics in water systems is also

receiving more attention during the last decade. Abdulbaki et al. (2017) presented a linear integer

programming model as a decision-making support tool for allocation of water resources (e.g.,

seawater, groundwater, and wastewater), as well as selection of the optimal treatment train based

on economic and environmental indicators. The model was implemented in a coastal city with a

total population of 250,000 inhabitants and the results revealed that the environmental impacts in

terms of greenhouse gas emissions and global warming potential had limited contribution in the

decision process compared to the economic factors. Al-Zahrani et al. (2016) developed a multi-

objective programming model to optimize the water sources, water distribution, and the final end

uses in Riyadh, Saudi Arabia. The water sources consisted of desalinated water, groundwater, and

treated wastewater, and the final uses consisted of industrial, domestic, and agricultural end uses.

The model was developed based on the minimization of groundwater conservation, water

desalination, and overall costs, as well as maximization of total wastewater reuse. The results show

that using the optimal solution for the water system in Riyadh, the groundwater extraction is

reduced from 2,859 Mm3 to 2,162 Mm3 and reuse of wastewater becomes equal to the total

10

produced domestic wastewater. This optimal allocation of groundwater also satisfied the water

demand in the area until 2050.

Chung et al. (2008) used a system dynamic simulation to develop a model for evaluation

of several possible scenarios applied to a realistic hypothetical water system. For each scenario,

water balances, water quality, construction costs, and operation and maintenance costs were

estimated. The results revealed that when population density is low, implementation of

decentralized wastewater treatment systems inside the smaller residential clusters is economic

compared to a centralized system. A study conducted by Molinos-Senante et al. (2012) assessed

nine different technologies for the secondary treatment considering both economic and

environmental factors using an environmental decision-making support system (EDSS). Later on

Molinos-Senante et al. (2015) used an analytical network process as a different solution for solving

the formulated problem in their previous research. The results showed that extensive technologies

(i.e., low-cost technologies such as a sand filter), pond systems, and constructed wetlands were the

most preferred alternatives for small residential areas, while the technologies with extended

aeration were the least preferred options. On the other hand, a cost management model was

developed by Ruiz-Rosa et al. (2016) to determine the economic option for water resource/supply

among the available alternatives consisting of surface water, desalination, groundwater, and

wastewater treatment for reuse. The results revealed that in case of having a modern plant with a

total capacity of 20,000 m3/day for desalination, water reclamation reduces the water system’s

costs by up to 75%, while using surface water and groundwater reduces the costs up to 20% and

11%, respectively. Although application of reverse logistics shows successful results for

production activities and urban planning (e.g., municipal solid waste management), considering

the similarities between those systems management and water systems management, it was

11

evidenced that only few studies have been conducted on different aspects of such application in

water systems.

A comprehensive literature review on the topic reveals lack of an integrated holistic

approach for wastewater systems management with a concentration on resource recovery as well

as consideration of the triple-bottom-line sustainability matrix (i.e., economic, environmental, and

social) for the evaluation and decision-making process. Although several studies on evaluate

different aspects of water infrastructures (e.g., wastewater treatment technologies, water resources

allocation, and water distribution) based on specific criteria (e.g., economic, health, and social

acceptance), the limitation of focusing on one aspect and/or criterion can mislead finding the most

sustainable solution, holistically, and criticize the decision-making process. For instance, reliance

on the indicators that exclusively report water consumption or withdrawal from natural resources

can mask sustainability intimation of the urban water system and management decisions including

water quality, water transfer, and infrastructure design (Owens, 2001). Aside from developing

models and methodologies for decision-making, the applicability of multi-criteria analysis for

management and decision-making requires consideration of social dimensions and broad

stakeholder engagement, along with the economic and environmental indicators, in accordance

with the triple-bottom-line sustainability model (Little et al., 2016; Seidel et al., 2014).

Moreover, adoption of reverse logistic strategies for the design of product supply chains

has shown successful results in a variety of production activities as well as waste management

systems. This not only increases the revenue by decreasing the needs for raw material and

associated manufacturing costs, but also reduces the environmental footprints of the product by

lowering the energy and material requirements. The same notion can be applied for the design of

water systems by consideration of the extracted water from natural water bodies as the raw

12

material, water treatment facilities as the manufacturing facilities, and treated water as the final

product (see Figure 1.1). The produced wastewater is generally considered as waste with end life

treatment and subsequent discharge requirements, but with consideration of wastewater as a used

product that can be collected (sewer system), remanufactured (WWTPs), and reused (reclaimed

water), the application of reverse logistics for the formation of a closed-loop water supply chain is

possible. Although adoption of reverse logistics for water systems management can present

decision-makers with many challenges due to complexity of treatment trains, emerging new

contaminants such as pharmaceuticals, public health involvement, severe water shortage, and rapid

increase in demand, it is vital to consider returning the treated wastewater into the water supply

chain, as an alternative for freshwater withdrawn. However, the literature review on the topic

revealed lack of a planning and design framework to evaluate and identify the most sustainable

application for water reuse and sustainable solutions for water systems management with focus on

resource recovery from wastewater. Similar to the other production systems, the industry sector in

water systems management mainly focuses on economic aspects of their technology and products

(e.g. selling price and energy consumption), governments mainly focus on regulatory aspects,

water quality, and meeting the projected demands, and environmental engineers focus on

environmental aspects of the system (e.g., GHG emissions and eutrophication potential). Hence, a

systematic approach to consider all of those parameters by integration of those parties with

stakeholders and officials in the design and management of water systems was the other gap that

this research was outlined to address.

13

Figure 1.1 Conventional reverse logistics (a) compared to its application for integrated wastewater management (b).

1.2 Scope of Research

To address the mentioned knowledge gaps in the previous section, the overall goal of this

research was to develop multi-criteria decision-making frameworks, by conducting multi-criteria

analyses of wastewater systems and developing decision-making support tools for wastewater

systems management, with an emphasis on resource recovery. This study provides decision-

makers with useful decision support tools to find the most sustainable solutions for urban water

systems management, incorporating the three major dimensions of sustainability (i.e., economic,

environmental, and social impacts), while adopting an integrated approach for the design.

To achieve the research goal, three major objectives were considered to direct the study:

1) Investigating the sustainability impacts and the tradeoffs associated with selection among the

water reuse alternatives (e.g., agricultural, urban, IPR, and DPR), as well as developing a decision-

making framework to find the most sustainable solution for selecting among reuse alternatives

under a given context; 2) Developing a multi-objective optimization model that assists the

decision-makers with finding the most sustainable solutions for the design of a water reuse system

(e.g., decision on treatment technology, facilities location and capacity, degree of decentralization,

14

and customer allocation) under a given context; and 3) Evaluating the impacts of external variables

(e.g., topography of the water service area and population density) on the design of hybrid water

reclamation and reuse systems, with a focus on the degree of decentralization of wastewater

treatment plants, treatment technology, and the water reuse system’s optimal configuration.

Regarding the goal of this research, it was hypothesized that increasing the reclaimed water

quality (degree of treatment) increases the costs and the environmental impacts (e.g., carbon

footprint) associated with the wastewater systems, and it is manily due to the higher energy

requirements associated with advanced treatment options. Moreover, the higher water quality

would probably offset a large portion of the overall environmental impacts by lowering the other

environmental footprints such as eutrophication potential and it also increases the value of resource

recovery (water in this case). Increasing the degree of decentralization increases the capital costs

significantly, but it was hypothesized that decentralization of treatment facilities decreases the

costs associated with water transfer (i.e., wastewater collection and reclaimed water distribution).

The decrease in water transfer costs may offset the contribution of capital costs to the annual net

present value in long-term planning. This also reduces the environmental impacts such as carbon

footprint by lowering the energy requirements for wastewater collection and reclaimed water

distribution. It was also hypothesized that the distance between water reclamation and reuse plays

a significant role in the environmental and economic impacts of the design. Increasing the distance

between facilities increases the capital costs associated with the implementation of water transfer

facilities and the energy requirements for water transfer and it highly influences the final economic

and environmental (e.g., carbon footprint) impacts associated with the design. Moreover, selection

of treatment technologies with lower capital and O&M costs reduces the final economic and

environmental impacts significantly, but to a relatively smaller extent. Finally, in terms of the

15

impacts of external variables, it was hypothezied that higher population density and lower

elevation variation (lower head-loss) necessitate implementation of more centralized water

reclamation facilities, while in areas with lower population density and/or higher elevation

fluctuation, decentralization of wastewater treatment plants and reuse locations would lower the

economic and environmental impacts (e.g., carbon footprint) of the design.

This study quantified and incorporated all three major dimensions of sustainability

according to the triple-bottom-line sustainabiliy model (i.e., economic, environmental, and social)

to develop a comprehensive multi-criteria framework for the evaluation of water reuse alternatives

and a multi-objective optimization model for finding the optimal solutions for the design of

resource recovery systems. To achieve the research goal, this study adopted an integrated approach

towards wastewater systems management, as well as quantification of the environmental impacts,

cost analysis, and stakeholder engagement during the design processes. Figure 1.2 shows the

integrated water management in this study and Figure 1.3 shows the summary of two approaches

adopted for the decision-making frameworks and tools developments. The information related to

treatment trains (e.g., treatment technology, scale of implementation, effluent water quality,

requlatory, and reuse alternatives) was mainly adopted from the study conducted by Diaz-Elsayed

et al. (2019).

16

Figure 1.2 The integrated wastewater systems management in this study.

Figure 1.3 The summary of two major approaches for decision-making framework development. Abbreviations: IPR: indirect potable reuse; DPR: direct potable reuse; GHG: greenhouse gas.

17

CHAPTER 2: A MULTI-CRITERIA SUSTAINABILITY ASSESSMENT OF WATER

REUSE APPLICATIONS1

2.1 Introduction

The increasing demand, scarcity, and contamination of water resources, accompanied by

the likely impacts of climate change, have made complex challenges for sustainable water and

wastewater management, demonstrating the need for the integrated management of wastewater

systems that facilitates and promotes resource recovery (Zheng et al., 2016). Traditionally, the

main function of a wastewater treatment plant was defined as the removal of contaminants to safely

release it back to natural water bodies (Gallego et al., 2008; Hospido et al., 2004). The traditional

approach for wastewater management primarily relies on centralized treatment systems and

reduces the negative impacts of wastewater on the environment and natural ecosystems (Morera

et al., 2016). However, this is achieved at the expense of high energy and chemical consumption

by these treatment plants (Godin et al., 2012). In order to maintain and improve the sustainability

of current systems, a paradigm shift must occur in wastewater management that emphasizes

1 Chapter 2 has been published: Rezaei, N., Diaz-Elsayed, N., Mohebbi, S., Xie, X., & Zhang, Q. (2019). A multi-criteria sustainability assessment of water reuse applications: a case study in Lakeland, Florida. Environmental Science: Water Research & Technology, 5(1), 102-118. Attributions: I conducted the study and led the authorship of the manuscript. Co-author Nancy Diaz-Elsayed contributed assistance in conducting the study, manuscript preparation, and review. Co-authors Shima Mohebbi and Qiong Zhang contributed guidance on the design for reuse scenarios, formulation of analysis sections (regret-based model and sensitivity analysis), and assistance in manuscript preparation and review. Xiongfei Xie provided us with the city of Lakeland’s water pipeline model. The model was used to calculate the water distribution requirements in two of the designed reuse scenarios.

18

resource recovery (e.g., water, energy, and nutrients) over treatment (Capodaglio et al., 2017). This

paradigm shift not only offsets some portion of required energy for treatment, but also reduces the

need for freshwater withdrawals by supplementing the water supply chain with reclaimed water.

Supply chain network design is receiving growing attention for solving production and

demand problems in a variety of research fields (Ramezani et al., 2013). Traditional supply chain

designs rely primarily on forward networks to manufacture products using raw materials. The

reverse logistics network, also known as a backward or recovery network, is the process of

returning used products to the collection and repair centers in order to be remanufactured and

become qualified for reuse. The same notion can be applied to water production: wastewater can

be diverted back to decentralized, satellite, or centralized wastewater treatment systems such that

it is treated to a water quality level that permits water reclamation. A study conducted by

Fleischmann et al. (2001) analyzed the impacts of product recovery on logistics networks. They

showed that the product recovery impacts such as economic benefits, environmentally conscious

customers and regulations, are context-dependent and require an individually comprehensive

approach for redesigning any type of industrial production activity in an integral way.

One primary challenge in realizing such a closed-loop water system can be the lack of a

planning and design framework to evaluate and identify the most sustainable application for

reclaimed water. During the last decade, the emerging challenges in water systems such as water

shortage, increasing water demand, and water pollution, have motivated researchers to evaluate

and improve the sustainability of water systems by focusing on water reclamation and reuse. There

have also been several Life Cycle Assessment (LCA) studies, as a standard method (International

Organization for Standardization (ISO), 2006a, 2006b), in recent decades to determine the impacts

resulting from water treatment, water distribution, and/or wastewater treatment for reclaimed water

19

use. In combination or parallel with LCA, multi-criteria analysis has been widely used to evaluate

the available alternatives according to a defined set of measurable criteria (Figueira et al., 2005;

Kolios et al., 2016). These approaches are broadly used to help decision-makers choose the most

appropriate solutions in achieving particular goals according to the evaluation criteria. However,

lack of environmental dimensions in the evaluation criteria for decision-making has led to

tremendous problems in the past century (e.g. fog, acid rain, and red tide), necessitating a transition

in allocation of the evaluation criteria for decision-making. The transition needs to provide the

insights with respect to economic, environmental, and social impacts, amongst which tradeoffs

may arise, to be supported by the decision-makers in both private and public sector. In addition,

decision-makers may have to deal with unknowns and uncertainties, which are characteristics of

investing in new designs and models (Linder and Williander, 2015). The bottom line is that the

methodology and the trend of assessment are highly influenced by the decision-making

framework, which is selected initially based on the case-specific parameters and the study’s goal

(Guarini et al., 2018).

Amores et al. (2013) evaluated the environmental impacts of reclaimed water use for non-

potable purposes such as irrigation in Spain. They showed that this scenario reduces the freshwater

consumption due to net water savings, but it did not make a significant improvement to the

environmental impacts due to the additional resources required for tertiary treatment. Pasqualino

et al. (2011) studied the environmental profile of four wastewater treatment plants for different

water reuse scenarios and revealed that using the reclaimed water for potable purposes not only

preserves freshwater resources, but also result in higher environmental impacts due to the

additional required treatment processes. Muñoz et al. (2009) designed four bench-scale treatment

systems to evaluate the environmental impacts of wastewater treatment for reuse via irrigation.

20

The results showed that wastewater reuse for irrigation with any of the studied tertiary treatment

systems had lower ecotoxicity impacts than those without tertiary treatment. Meneses et al. (2010)

used LCA methods to evaluate the environmental advantages and disadvantages of reclaimed

water use for non-potable applications. The results showed that replacing desalinated water with

reclaimed water for non-potable purposes is beneficial when there is a scarcity of freshwater.

Other studies analyzed the environmental impacts of urban water systems that mainly focus

on treatment technologies (e.g., Beavis and Lundie, 2003; Lim and Park, 2009; Metcalf et al.,

2007). The study conducted by Lim and Park (2009) revealed that as the degree of treatment

increases, the cost and the negative environmental impacts associated with the treatment increases,

although they offset a portion of the freshwater needed. There are also few studies on application

of multi-criteria analysis in the design and evaluation of water systems. Ren and Liang (2017)

developed a multi-attribute decision analysis, with economic, environmental, and society-politic

evaluation criteria, to assess the sustainability of four treatment processes for water reclamation in

China. The results revealed that with the selected weighting strategy, anaerobic single-ditch

oxidation obtained the best score among the treatment technologies; however, the selection was

highly dependent on the weighting strategy. Benedetti et al. (2010) developed a Monte Carlo

simulation and multi-criteria analysis to achieve the optimal configuration in the operation phase

of a wastewater treatment plant in Belgium. The results revealed a significant improvement in

terms of economic (total costs and operation costs) and environmental (total nitrogen) impact

assessments. They also showed that the anoxic fraction of reactor volume and the volume of

primary clarifier played a significant role in system’s performance. Flores-Alsina et al. (2008) also

developed a multi-criteria analysis to evaluate the operation of six wastewater treatment plants

under uncertainty, using a Monte Carlo simulation. The evaluation criteria consisted of

21

environmental, economic, legal, and technical aspects. The results revealed that multi-criteria

analysis not only enhanced the performance of treatment systems, but also reduced the risk of

system failure. Nonetheless, no prior studies evaluated treatment requirements and different types

of water reuse applications in a holistic (i.e., economic, environmental and social) sustainability

assessment. Therefore, the goal of this part of the study was to evaluate the tradeoff between

reclaimed water quality and corresponding costs, environmental impacts and social benefits for

different types of water reuse applications. This tradeoff analysis paired with a regret-based model

can help decision-makers identify the degree of treatment needed to produce reclaimed water as

well as the type of reuse applications to initiate.

2.2 Materials and Methods

In this part of the study, a multi-criteria analysis framework was developed and used to

compare the water reuse alternatives in terms of economic, environmental, and social impacts. The

study was conducted in the City of Lakeland, Florida, while the water service area is experiencing

a rapid growth in terms of population. The methodology used in this study is described in this

section.

2.2.1 Study Area

The tradeoff evaluation for different types of reclaimed water applications was conducted

for the City of Lakeland, which is located on the western side of Polk County, Florida. The city

is within the Southwest Florida Water Management District (SWFWMD) boundary (REISS

Engineering, 2009), and has a total population of 106,420 and a population growth rate of 9.3%

(United States Census Bureau, 2016). Figure 2.1 shows the summary of current water, wastewater,

and reclaimed water systems in the City of Lakeland and a map showing the location of the primary

water and wastewater infrastructure can be found in the Appendix B (Figure B.1). The source

22

water for the city’s water supply is groundwater withdrawn from the Floridian aquifer using 19

wells, and the water is conveyed to two water treatment facilities via an 8.74 mile pipeline (City

of Lakeland, 2017). T.B. Williams is the larger water treatment facility with a design capacity of

51 mgd located in the west-central part of the city and C.W. Combee is the smaller plant with a

design capacity of 8 mgd located in the northern part of the city. The water distribution system

incorporates a service pipeline with approximately 998 miles of total length to deliver the treated

water to more than 54,000 active customers (City of Lakeland, 2017). Based on the city’s report,

water use is characterized as residential (65%), commercial and industrial (26.3%), aesthetic and

recreational (2.3%), fire flow (0.3%), and the remaining portion was unaccounted for.

Figure 2.1 Summary of the current water, wastewater and reclaimed water cycle in the City of Lakeland, Florida. The water usage is shown in percentage and the design capacity/operation capacity for the plants is shown in MGD.

The city’s sewer collection system covers approximately 40,000 square miles of service

area and encompasses 50 miles of forced sewer and 300 miles of gravity mains. The system is

23

being used to convey raw wastewater to two wastewater treatment plants (City of Lakeland, 2017).

The Glendale WWTP is the larger treatment facility with a design capacity of 13.7 mgd located in

the southern part of the city and the Northside plant is the smaller plant with a design capacity of

8 mgd, covering the northern part of Lakeland (REISS Engineering, 2009). Both wastewater

treatment plants consist of primary treatment and secondary treatment (conventional activated

sludge [CAS]) followed by disinfection (chlorination). The City of Lakeland’s current reclaimed

water infrastructure provides 5.11 mgd of reclaimed water to the McIntosh power generation

facility where the water is used as cooling make-up water. The other portion of treated wastewater

effluent receives further treatment in the Lakeland artificial wetlands. From there, the water is

pumped by the TECO power generation plant.

Although Lakeland’s water system is suitable for present-day water demand and treatment

requirements, the City of Lakeland is undergoing rapid growth in the southwest and northeast

regions of the service area, which makes it challenging to satisfy future water demand. The amount

of water that the City of Lakeland can withdraw from the Floridian aquifer has been limited to an

annual average daily demand (AADD) of 35.03 mgd and a monthly average maximum of 42.04

mgd. The city’s water use permit is issued by SWFWMD and is valid through December 16, 2028

(REISS Engineering, 2009). Since the service area and the population in the City of Lakeland are

growing quickly, it has been predicted that in 2026 the city will have a population of approximately

242,000 and a water demand projection of 35.03 mgd. Based on the city’s existing permit and

current water system capacity, meeting the water demand will be challenging in a few years.

Different types of water reuse options, which can satisfy the future water demand projection, were

designed, evaluated and compared based on economic and environmental criteria. Ultimately, a

decision-making tool that can be used by stakeholders to evaluate the tradeoffs between water

24

reuse types, degree of treatment and sustainability constraints was also introduced. The effluent

from the Glendale water reclamation facility and Lakeland’s artificial wetland were considered for

reuse scenarios, or as the influent for the additional treatment, when needed. The effluent water

quality reports were obtained from the facilities, which were reported based on an annual average

basis (2017). More information regarding the water quality and water quality requirements (reuse

standards) used for the design of additional treatments can be found in Appendix B (Table B.1).

2.2.2 Scenario Generation and Design

A supply chain network that contains a forward and backward network is known as a

closed-loop supply chain network (Ramezani et al., 2013). US EPA (2012) guidelines for water

reuse was used to design seven scenarios that can potentially improve the sustainability of the

current water network in the City of Lakeland and meet future demand. The alternatives in this

study consisted of: 1) urban reuse (unrestricted), 2) agricultural reuse (food crops), 3) indirect

potable reuse (IPR), 4) direct potable reuse (DPR), 5) distributed unrestricted urban reuse, 6)

centralized treatment for distributed unrestricted urban reuse and 7) decentralized treatment for

distributed unrestricted urban reuse (US EPA, 2012). The last two scenarios were designed to also

further evaluate the impacts of a degree of decentralization of treatment plants to the water systems.

For most reuse types, there are US guidelines, regulations and quality standards that the reclaimed

water has to meet. These guidelines were primarily based on the US EPA and Florida Department

of Environmental Protection (FDEP) for water reuse in the state of Florida (Florida DEP, 2016;

US EPA, 2012). Although US EPA water reuse guidelines lacks the quality requirements and

regulatory for DPR, it is recommended that water quality should meet the drinking water quality

for this reuse scenario. Additional treatment processes were added to the Glendale WWTP’s

existing treatment train when the effluent’s water quality did not meet the quality requirements for

25

water reuse (i.e., scenario 3 and scenario 4, see Table B.1 in Appendix B). Specifically, the

WateReuse Treatment Train Toolbox IT3PR and the guideline manual developed by WateReuse

Research Foundation (Trussell et al., 2015) were used for these scenarios. The WateReuse

Treatment Train Toolbox IT3PR considers US EPA water quality requirements in its database for

the design of additional treatments.

First, the best location for implementation of each reuse scenario was identified based on

various considerations such as available lands with the minimum distance from the reclaimed

water production’s location, land price in the City of Lakeland, the stakeholders and the city

officials’ preferences and the US EPA guidelines (e.g., requirement for the minimum water travel

distance between injection point and extraction wells for IPR). In the next step, considering the

amount of available reclaimed water for each scenario, reclaimed water quality at different points

of generation and the quality requirements, the best facility for providing the water needed for each

reuse design was selected (i.e., Glendale water reclamation facility or Lakeland’s artificial

wetland). The major pipelines were designed (i.e., diameter and length) to convey the reclaimed

water from the source of generation to the reuse scenario’s location; they accounted for the

required water flow rate and the expected water velocity. For the minor pipelines, the same

approach was adopted and the junctions and fittings were selected based on the space limitations

(where needed).

To calculate the pumping power required for each scenario (major and minor pumps), the

Darcy-Weisbach Pressure and Head Loss Equation was used. To obtain the Reynolds number,

Darcy’s friction factor, skin friction coefficients and pressure drops for pipe fittings, the Moody

diagram and Fundamentals of Engineering Reference Handbook were used (Moody, 1944;

NCEES, 2013). For the selection of the pumps, pipeline materials, pipeline fittings and the other

26

equipment needed for designing each scenario, the process equipment cost estimation manual (Loh

et al., 2002) and the McMaster-CARR website and manuals were used. For the calculation of the

pipelines’ length needed for reuse scenarios 5, 6, and 7, which require extensive pipelines for

unrestricted decentralized urban reuse, as well as for the energy requirements for reclaimed water

distribution, Bentley WaterGEMS CONNECT Software Edition [10.00.00.50] was used. The GIS

data and the water network and sewer system files were obtained from the City of Lakeland’s

Water Utilities Department.

The first reuse scenario (unrestricted urban reuse) evaluated the use of reclaimed water for

the irrigation of golf courses. With a total of 1,103 golf courses and 524 golf communities, golf in

the state of Florida is a critical industry contributing to the state economy (SRI International, 2015).

On average, irrigation of each golf course in Florida requires 0.26 mgd of water (Florida DEP,

2016). In this scenario, 2.83 mgd of reclaimed water was taken from the Glendale WWTP’s pond

and conveyed to 10 different golf courses around the City of Lakeland using 12-3/4” O.D. pipelines

with a total length of 30.26 miles. Since the water quality of Glendale WWTP’s effluent met the

requirement for the irrigation of golf courses, no additional treatment was needed.

Scenario 2 considered agricultural water reuse for irrigating strawberries – one of Florida’s

major food crops. Four major pipelines (12-3/4” O.D.) conveyed 4.6 mgd to 170 acres of farmland

over a total length of 18,406 ft. No additional wastewater treatment was required for this scenario

(Jeong et al., 2016) and drip irrigation was assumed for dispersal.

For scenario 3 (IPR), 2.83 mgd of reclaimed water was taken from the artificial wetlands

and was injected into two 750-ft injection wells (1.5 mgd capacity each). Ultraviolet (UV)

disinfection was added to the treatment train to meet the total number of fecal coliforms

27

requirement (COTTON et al., 2001; US EPA, 2003), and the reclaimed water was conveyed over

11.68 miles by a major pipeline (24” O.D.) from the wetlands to the injection site.

In direct potable reuse, reclaimed water serves as the influent for water treatment plants.

Although this type of reuse is rare, it has been receiving more attention during the last decade.

Regulations and guidelines for this type of reuse are non-existent in the U.S.; however, drinking

water quality standards are recommended (US EPA, 2012). For scenario 4, the reclaimed water

was conveyed 7.98 miles by a major pipeline (24” O.D.) from the artificial wetlands to the T.B.

Williams water treatment facility, which had the available capacity to receive the extra influent.

Additional filtration and disinfection processes were added to the treatment train to satisfy drinking

water quality guidelines (see Table 2.1 and Figure A7 in the supplementary material). Figures

showing the location and pipeline required to implement each scenario can be found in Appendix

B (see Figures B.3, B.4, B.5, B.6, and B.8).

In reuse scenario 5, a total of 2.83 mgd of treated wastewater from Glendale WWTP was

distributed using an extensive “purple” pipeline for non-potable urban reuse purposes such as

backyard irrigation, landscaping, and carwashes.

As it was mentioned before, the last two scenarios were designed to also evaluate the

impacts of some degree of decentralization for wastewater treatment plants. In scenario 6, one

centralized medium-scale WWTPs with a capacity of 3.00 mgd was designed to treat 2.83 mgd of

household wastewater. The reclaimed water was distributed using an extensive purple pipeline for

non-potable urban reuse. In scenario 7, the City of Lakeland was divided into five different clusters

and five decentralized medium-scale WWTPs with a capacity of 0.7 mgd were designed to treat

2.83 mgd of household wastewaters in total (see Figure A9 in the supplementary material). The

reclaimed water was distributed using an extensive purple pipeline, again for non-potable urban

28

reuse. Construction data from existing and decommissioned WWTPs in the City of Lakeland were

used to model the centralized as well as the five decentralized plants. Details about this and other

scenarios (e.g., the location of the WWTPs, pipelines, etc.) can be found in Appendix B (Tables

B.3-B.9).

Table 2.1 Summary of information related to each scenario in this study

Scenario Description Recommended treatment

Additional treatment required

Pipeline required

Pumping requireme

nt

Energy consumption by additional

treatment

Nitrogen and Phosphorus

concentration in the effluent

Scenario 1 Urban reuse Secondary treatment-Filtration-

Disinfection

- 30.26 mi 12-3/4”

O.D.

48,000 KWh/day

0 KWh/day 15.01 (mg TN/l)

5.7 (mg TP/l)

Scenario 2 Agricultural reuse

Secondary treatment-Filtration-

Disinfection

- 3.49 mi 12-3/4”

O.D.

16,000 KWh/day


5.7 (mg TP/l)

Scenario 3 Indirect potable reuse


Disinfection -Multiple barriers for pathogen and organics removal

(Advanced)

UV disinfection

11.68 mi 24” O.D.

32,486 KWh/day

298 KWh/day

1.54 (mg TN/l) 4.1 (mg TP/l)

Scenario 4 Direct potable reuse

No defined standard

Ultra-filtration-UV/H2O2-additional

Chlorination

7.98 mi 24” O.D.

31,937 KWh/day

2,678 KWh/day

1.0 (mg TN/l) 4.1 (mg TP/l)

Scenario 5 Distributed urban reuse


Disinfection

- 569.17 mi Varying diameter

35,635 KWh/day


5.7 (mg TP/l)

Scenario 6 Centralized treatment for distributed urban reuse


Disinfection

1 medium-scale CAS

system

569.17 mi Varying diameter

35,635 KWh/day

5,818 KWh/day

15.01 (mg TN/l)

5.7 (mg TP/l)

Scenario 7 Decentralized treatment

for distributed urban reuse


Disinfection

5 medium-scale CAS

systems

569.17 mi Varying diameter

19,599 KWh/day

7,263 KWh/day

15.01 (mg TN/l)

5.7 (mg TP/l)

Figure 2.2 shows the overview of the scenarios considered in the study and the summary

of information related to each scenario can be seen in Table 2.1.

29

Figure 2.2 Overview of the scenarios considered in the study. Abbreviations: UV: ultraviolet; UF: ultra-filtration; WTP: water treatment plant; WWTP: wastewater treatment plant; Cl: chlorination; NPR: non-potable reuse; IPR: indirect potable reuse; DPR: direct potable reuse.

2.2.3 Indicator Description and Quantification

In order to evaluate different feasible scenarios and provide a decision-making support tool

for stakeholders, multi-criteria evaluation was used. The criteria selected in this study consisted of

an economic indicator, environmental impacts and the value of resource recovery (VRR) as social

impacts.

2.2.3.1 Economic Indicator

In this study, capital costs and operation and maintenance (O&M) costs were considered

for each design. For the added treatment processes, the capital costs included land purchase,

pipelines, pumps, construction of pipelines and wells, and equipment and materials. The O&M

costs included pumping energy, pipeline maintenance, labor, chemicals, overhead and

management, energy consumed for the added treatment processes, repairs and material

consumption. Data were mainly collected from stakeholders, the primary power companies in the

state of Florida (TECO and Duke Energy) and engineering handbook manuals (e.g., NCEES,

2013). The data used to calculate capital and O&M costs for each scenario can be found in

Appendix B (Table B.2 and Tables B.10-B.16). The cost data obtained from the City of Lakeland

30

are converted to 2017 dollars using Unites States historical cost indexes (RSMeans, 2017) to

estimate the costs associated with the new design scenarios.

In order to combine capital and O&M costs for all the scenarios, annualized specific net

present value (ASNPV) was calculated (Maurer, 2009). First, the net present value (NPV) was

calculated, which consisted of the present value of capital and O&M expenditures. The O&M

expenses (CO&M) for each year (n = 1, 2, 3, …, 33) were converted to present values (PV) and the

annualized specific net present value (ASNPV) was calculated using equation 2.1 for an average

interest rate, i, of 5%, lifespan, Tp, of 33 years, and demand (Pt) at time t for each component.

More details about the cost calculations can be found in Appendix B (Equations B.1-B.4 and Table

B.17).

𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 =𝑁𝑁𝑁𝑁𝑁𝑁Capital+ ∑ 𝐶𝐶𝑂𝑂&𝑀𝑀 1

(1+i)𝑛𝑛 331

1𝑇𝑇𝑃𝑃

∫ 𝑁𝑁𝑡𝑡.𝑑𝑑𝑑𝑑𝑇𝑇𝑃𝑃0

(2.1)

2.2.3.2 Environmental Indicators

Environmental footprints of the designs are becoming increasingly important in the

construction of new infrastructures due to increasing environmental awareness (Du et al., 2011;

Phillips et al., 2013; Qi and Chang, 2013; Sinha et al., 2016). Carbon footprint and eutrophication

were used as environmental indicators in this study.

Carbon footprint (CFP) is an abstract environmental sustainability indicator (ESI) to

globally characterize the impact on climate change (Qi and Chang, 2013). It is an estimate of total

greenhouse gas (GHG) emissions from a defined activity over a specific time frame or over the

product/project’s life cycle, typically expressed as carbon dioxide equivalents (CO2-eq). Carbon

footprint is highly influenced by the electricity consumption of the processes (Byrne et al., 2017).

Since previous LCA studies have revealed that CFP in water and wastewater industries is

dominated by the electricity consumption during the processes (Loubet et al., 2014; Pintilie et al.,

31

2016), electricity consumption by the pumps and processes was selected to calculate CFP for this

case. In this study, greenhouse gas equivalencies for electricity consumption were calculated based

on eGRID data (US EPA, 2017). Electricity consumption data were collected from the individual

treatment plants in the City of Lakeland. Additionally, the pumping electricity was estimated based

on the types of pumps assumed for each scenario and engineering handbooks (NCEES, 2013).

Water eutrophication (EU) refers to the nutrient enrichment (nitrogen and phosphorus) of

aquatic environments and is becoming one of the biggest challenges in aquatic environmental

protection around the world (Heisler et al., 2008). Since the degree of eutrophication is largely

determined by the magnitude of external nitrogen (N) and phosphorus (P) loads (Valiela et al.,

2016), the concentration of those elements in the final reclaimed water was considered for this

environmental indicator expressed as PO4-equivalent. Depending on the level of treatment and the

source of reclaimed water used for each scenario, the concentration of these two elements and the

corresponding environmental impacts varied for each design. Moreover, for urban reuse (golf

course irrigation), agricultural reuse (strawberry irrigation) and distributed unrestricted urban reuse

(e.g., lawn irrigation), since nutrient uptake by the plants offsets a portion of eutrophication

potential of the reclaimed water, it was included in the calculation of the eutrophication potential

associated with these reuse scenarios. For agricultural reuse, drip irrigation was assumed for

dispersal and the design of the irrigation system (plants, irrigation lands, and water requirement)

for the calculation of nutrient uptake, were based on the studies of strawberry production in the

state of Florida (e.g., Peres et al., 2011). For calculation of nutrient uptake by golf course grass,

strawberry plant and lawn irrigation, the required data was obtained from previous studies (i.e.,

Kumar and Dey, 2011; Palmer et al., 2014; Vanhoutte et al., 2017). As a rough estimation, 12%,

9% and 10% nutrient uptake from the reclaimed water for grass surface irrigation, strawberry drip

32

irrigation and non-potable urban reuse (~80% for lawn irrigation) was assumed, respectively.

Water quality information was obtained mainly from stakeholders, the water and wastewater

treatment plants’ water quality data sheets, the artificial wetlands’ influent and effluent water

quality data and the water quality reports from the City of Lakeland.

2.2.3.3 Social Indicator

The value of resource recovery (the willingness to pay) was used as the social indicator for

the evaluation of each scenario. The value of resource recovery was collected from Polk County

and Hillsborough County’s reclaimed water prices (Hillsborough County, 2017; Polk County,

2017), considering the fact that as the value of the recovered resource increases, the willingness to

pay by the reclaimed water end users increases. For urban reuse, the monthly flat rate of the

reclaimed water for irrigation purposes (based on a 12" pipeline) was used. For agricultural reuse,

the selling price of reclaimed water to the farmers in the State of Florida was used. For IPR and

DPR, the price of drinking water was used for calculating the value of the reclaimed water,

considering the price deduction due to the additional processes (water extraction, conveyance and

treatment for IPR and water treatment for DPR) needed in these reuse scenarios before the water

became qualified to be sold to the customers. The data related to costs for water treatment was

obtained from the T.B. Williams water treatment facility in the City of Lakeland. Finally, for

distributed unrestricted urban reuse, the monthly charge for the reclaimed water network (purple

pipeline) in Hillsborough County was used as the value of resource recovery (Hillsborough

County, 2017a).

2.2.4 Scenario Evaluation

According to the technical literature on multi-criteria assessment and decision-making,

there is a variety of evaluation methods with application in different situations. However, selection

33

of the most appropriate method for an specific problem and field of application has not been

investigated previously (Guarini et al., 2018). Although there are advantages and disadvantages

associated with each assessment method, the selection depends on the case-specific parameters in

the case study (e.g., number of evaluation elements, typology of indicators, expected solutions,

type of decision-making problem, and solution approach) and the decision-makers preferences. In

this study, in order to evaluate each reuse scenario and investigate the tradeoffs, a regret-based

model was used based on the Minimax Regret criterion. The Minimax Regret model, also known

as the savage model, is an approach to decision-making under uncertainty. For instance, when the

likelihood of the possible outcomes is not known with sufficient precision to use the classical

expected value criteria, the regret-based model can be used as a support tool for the decision-

making process (Loulou and Kanudia, 1999). Moreover, when there is a discrete number of

choices, such as different possible real world scenarios, the minimax regret strategy is a useful tool

for a risk-neutral decision-making. Minimax regret model also provides decision-makers with the

ability to normalize the evaluation criteria when there is magnitude disorder and uncertainty. This

technique minimizes the risk of making the wrong decision in selection among possible

alternatives. Although there is a variety of alternatives for decision-making and a comparison to

other models can be made, it was outside of the scope of this study. In this study, a symmetric

formulation was obtained for a decision-making problem stated in terms of a specific constraint to

minimize (negative) or maximize (positive) impacts. If 𝐴𝐴𝑖𝑖,𝑗𝑗 is defined as the performance of

strategy 𝑖𝑖 ∈ 𝐴𝐴 (reuse scenario) for indicator 𝑗𝑗 ∈ 𝐹𝐹 (defined criteria and constraints), the regret (𝑅𝑅𝑖𝑖,𝑗𝑗)

is defined as the difference between the impact incurred and the optimum achievable (Loulou and

Kanudia, 1999), i.e.:

𝑅𝑅𝑖𝑖,𝑗𝑗 = �𝑜𝑜𝑜𝑜𝑜𝑜𝑖𝑖∈𝑆𝑆

(𝐴𝐴𝑖𝑖,𝑗𝑗) − 𝐴𝐴𝑖𝑖,𝑗𝑗�; where S: scenarios and F: indicators (2.2)

34

The optimum achievable is the optimum value (maximum or minimum) in each impact

category across reuse alternatives. In order to make the comparison across indicators, the

normalized regret scores (NR) can be calculated by:

𝐴𝐴𝑅𝑅𝑖𝑖,𝑗𝑗 = 𝑅𝑅𝑖𝑖,𝑗𝑗max𝑖𝑖∈𝑆𝑆

�𝑅𝑅𝑖𝑖,𝑗𝑗� (2.3)

and the final regret score (R�) for each scenario can be calculated by assigning weighting factors,

𝑤𝑤𝑗𝑗, for each indicator:

𝑅𝑅�𝑖𝑖 = ∑ �𝑤𝑤𝑗𝑗 ∙ 𝐴𝐴𝑅𝑅𝑖𝑖,𝑗𝑗�𝑗𝑗 ; Where ∑ 𝑤𝑤𝑗𝑗𝑗𝑗 = 1 (2.4)

The results were reported based on individual indicators and a multi-criteria analysis; in

the latter case, weighting schemes were assigned such that equal weighting was applied to each

indicator (the base case), as well as weighting schemes that were cost-centered and

environmentally-centered. The weighting factors for cost- and environmentally-centered results

were based on stakeholder preferences, where cost-centered assigned 55% weight for the economic

indicator and 15% for the other indicators and environmentally-centered assigned 35% weight for

each environmental indicator and 15% for the remaining indicators.

2.2.5 Location and Treatment Analysis for DPR

In this study, the minimum treatment requirement for DPR was considered. In other cases,

DPR can include more extensive treatment due to lower reclaimed water quality and/or higher

water quality requirements, which result in higher impacts. Moreover, this reuse scenario usually

receives less interest from stakeholders due to the complexity of treatment processes and some

other challenges such as social acceptance. In this scenario, the reuse location is also highly

restricted by the location of water treatment facilities in the area and it reduces the flexibility of

the end-use location for DPR. Hence, a sensitivity analysis was conducted to evaluate the impact

35

of increasing the distance to the end use location, in addition to increasing the ASNPV to

accommodate additional treatment requirements. In both instances, the variable in question was

increased in increments of 10% and the resulting regret scores (for the base case) were evaluated.

2.3 Results and Discussion

In this part of the study, different water reuse alternatives were designed to fill the gap

between available water resources and projected water demand in the City of Lakeland, Florida.

A multi-criteria analysis framework was developed to compare the water reuse alternatives and

provide the insights to the factors with the highest impacts. Moreover, a sensitivity analysis of

parameters that had a significant contribution to the impact categories was conducted.

2.3.1 Tradeoffs for Water Reuse Management

Based on the results of this study, it was evident that there were tradeoffs between the

degree of treatment for water reuse, water reuse type and location, and the economic,

environmental and social impacts of the reuse scenarios. For instance, the urban reuse and

agricultural reuse scenarios had the same treatment scheme, but the longer distance to the point of

urban reuse resulted in a much higher ASNPV (1,667 vs. 413 $/MG) as is shown in Figure 2.3.

36

Figure 2.3 Annualized specific net present value (ASNPV) and value of resource recovery (VRR) for different reuse scenarios, based on a design life time of 33 years. Abbreviations: IPR: indirect potable reuse; DPR: direct potable reuse; D: distributed; CT: centralized treatment; DT: decentralized treatment; MG: million gallon.

Moreover, although the scenarios had similar eutrophication impacts because of the

similarities in water quality and nutrient uptake, the carbon footprint was much higher for urban

reuse than agricultural reuse (8,684 vs. 1,781 kg CO2-eq/MG) because of higher energy

requirements for reclaimed water transfer and distribution. Agricultural reuse not only had lower

ASNPV compared to urban reuse, it also obtained a higher VRR due to the higher value of

reclaimed water for this reuse type. Since the selling price of the reclaimed water to the farmers

for agricultural purposes was much higher than the selling price for urban reuse, with the same

degree of treatment, agricultural reuse had a higher value of resource recovery, as much as $1,394

higher per million gallons of reclaimed water, compared to the urban reuse ($173/MG). Although

agricultural reuse was the most preferable option across most indicators (i.e., ASNPV, VRR and

carbon footprint), this reuse scenario had the highest eutrophication (see Figure 2.4) among all the

37

scenarios, which was mainly due to the high level of nutrients remaining in the reclaimed water

for irrigation purposes (Metcalf et al., 2007).

Figure 2.4 Environmental impacts (carbon footprint [CFP] and eutrophication [EU]) associated with different reuse scenarios. Abbreviations: IPR: indirect potable reuse; DPR: direct potable reuse; D: distributed; CT: centralized treatment; DT: decentralized treatment; MG: million gallon.

Primary and secondary treatment (conventional activated sludge (CAS) in this case) plays

a significant role in the cost of the treatment trains and it was common among all scenarios for

water reuse due to the minimum water quality requirements. Hence, the cost evaluation excluded

the common processes and only included the processes that were different for different reuse

scenarios. The results revealed that the implementation and operation of additional treatment

processes was not a significant contributor to the economic indicator compared to the capital and

O&M costs associated with the distribution of the reclaimed water (e.g., pipeline construction,

reclaimed water pumping). On the other hand, as the reclaimed water quality increases, the value

of resource recovery increases accordingly and the environmental impacts of water reclamation

(eutrophication) decreases due to greater nutrient removal. As it can be seen in Figure 2.4, although

improving the reclaimed water quality from urban reuse to IPR and DPR had little impact on

38

ASNPV (considering the costs associated with the water conveyance), it resulted in a significant

increase to the VRR (173 vs. 3,500 $/MG for urban reuse and IPR, respectively). As the result also

showed, increasing the degree of treatment after CAS from agricultural reuse to IPR and DPR did

not increase the carbon footprint significantly, due to the low energy requirements of the additional

treatment processes (i.e. ultra-filtration, UV disinfection and additional chlorination). Most of the

previous studies have also shown that the operation phase in treatment process and water transfer

are responsible for approximately 40% and 50% of GHG emissions associated with water systems,

respectively. Wastewater treatment and disposal (reclaimed water quality) were also the significant

contributors (~91%) to the freshwater eutrophication potential (e.g., Amores et al., 2013;

Barjoveanu et al., 2014; Lemos et al., 2013; Opher and Friedler, 2016; Risch et al., 2015; Slagstad

and Brattebø, 2014).

As Figure 2.4 also shows, distributed urban reuse (scenario 5) increased the ASNPV

significantly. Distributed urban reuse for non-potable purposes (e.g., lawn irrigation and

carwashes) required an extensive pipeline for distribution of the reclaimed water to the households

(purple pipeline) and it increased the capital costs associated with this scenario and the ASNPV

accordingly. Although distributed urban reuse had the highest ASNPV among all reuse scenarios,

this type of reuse reduces the cost associated with withdrawal, treatment and distribution of water

to the distributed end users (households) by replacing the potable water with the reclaimed water

for non-potable purposes, to a greater level than other reuse scenarios. These considerations were

outside the scope of this study since the amount of water offset was similar across scenarios. The

summary of different costs associated with each scenario and more details about the capital costs,

O&M costs and the value of resource recovery for reuse scenarios, can be found in Appendix B

(Table B17).

39

2.3.2 Decentralized VS. Centralized Reuse and Treatment

As it was mentioned before, two scenarios were designed to evaluate the impacts of some

degree of decentralization for the water systems. The results for these reuse scenarios can be seen

with the last two scenarios in Figure 2.4 and Figure 2.5. For both reuse scenarios, ASNPV

increased significantly due to the extensive pipeline requirements for distributed urban reuse.

Accordingly, these reuse scenarios obtained the highest carbon footprint among the different

scenarios, which is mainly due to the high electricity consumption by the major pumps for

distribution of reclaimed water to the final customers. Previous LCA studies have also revealed

that the collection and distribution of wastewater and reclaimed water, compared to the other steps

in the process, consume the highest amount of electricity in urban water and wastewater

infrastructure (Lyons et al., 2009). The higher degree of decentralization in scenario 7 resulted in

higher ASNPV due to the need for multiple medium-scale wastewater treatment plants and higher

O&M costs (per unit volume of wastewater) associated with them; however, the costs and energy

requirements for distribution of the reclaimed water to the final users (households) and associated

CFP were reduced significantly for this reuse scenario (see Tables B.8, B.9, B.15 and B.16 in

Appendix B). In addition, increasing the degree of decentralization has some advantages such as

more flexibility in operation, reliability and better management in case of natural disasters or

terrorist events (Daigger, 2009). Therefore, the tradeoffs have to be carefully evaluated for the

given context when considering the degree of decentralization. Some of the previous literatures

have also shown that decentralization of wastewater treatment facilities improves the

environmental and economic impacts associated with water systems (e.g., Chung et al., 2008;

Gardels et al., 2011; Glick and Guggemos, 2013; Lam et al., 2015), while other studies revealed

that centralized systems show better performances (e.g., Matos et al., 2014; Shehabi et al., 2012;

40

Thibodeau et al., 2014). Some believe that the decision on decentralization of the plants strongly

depends on local conditions (e.g., population density) and a framework is required to evaluate the

study area and make the final decision (Chung et al., 2008; Lehtoranta et al., 2014).

2.3.3 Multi-Criteria Decision-Making

The results for the regret-based analysis are shown in Figure 2.5 and Table 2.2. This table

shows the normalized regret score (NR) for each reuse scenario within each criterion and the final

regret score (R�) based on different weighting strategies. Based on the definition of the regret-based

model, the reuse scenarios with regret scores closer to zero obtained better values for the

corresponding criteria.

The preferred scenario, with respect to the normalized regret score, changed as different

individual impacts were considered. For example, agricultural reuse had the lowest normalized

regret score for the economic (NR_ASNPV) and carbon footprint indicators (NR_CFP) (see Table

2.2), although there is only a small difference between the agricultural reuse scenario and the urban

reuse, IPR and DPR scenarios in the case of the economic indicator. The lower regret scores could

be attributed to the lower infrastructure requirements for water transfer pipelines and treatment

(i.e., agricultural reuse, urban reuse, or IPR). Accordingly, the scenarios that required more water

transfer and distribution (as was the case with distributed reuse) had a significantly higher

NR_CFP. This was due to the higher consumption of pumping energy for reclaimed water

distribution. Interestingly enough, however, the second most preferred option for the carbon

footprint indicator (NR_CFP) was the implementation of decentralized treatment plants with

distributed urban reuse (Scenario 7). The savings in energy consumption from the local distribution

of reclaimed water were enough to lead to significant reductions in this indicator relative to all

centralized treatment options (excluding the most preferred option, agricultural reuse). Since the

41

water distribution infrastructure and pumping energy had a significant influence on the preferred

scenario, sensitivity to the distance to the end user and the type of terrain (hilly versus flat) are

expected. Moreover, the better reclaimed water quality for IPR and DPR resulted in significantly

lower social (NR_VRR) and environmental (NR_EU) impacts.

Table 2.2 Results for the regret-based model and the calculated regret score for each scenario. Abbreviations: IPR: indirect potable reuse; DPR: direct potable reuse; D: distributed; CT: centralized treatment; DT: decentralized treatment; ANPV: annualized net present value; CFP: carbon footprint; EU: eutrophication; VRR: value of resource recovery

42

Figure 2.5 Results for the regret-based model and the calculated regret score for each scenario.

From Table 2.2, it is evident that when the weighting strategy transitioned from the base

case to cost-centered, scenarios with a shorter distance between reclaimed water production and

end use locations, and/or lower complexity in design implementation and treatment, obtained

better final regret scores. Although increasing the distance from agricultural reuse to IPR and DPR

increased the ASNPV and CFP significantly, the lower environmental impact (EU) and the higher

social indicator (VRR) decreased the final regret scores (both cost- and environmentally-centered)

associated with these two scenarios. Moreover, changing the weighting strategy to

environmentally-centered improved the final regret score of scenarios with higher reclaimed water

quality (IPR and DPR). Accordingly, DPR obtained the best cumulative regret score across the

three weighting strategies. The sensitivity to the distance of the treatment plant and treatment costs

for the DPR scenario will be examined further in Section 3.4.

The results also revealed that the additional treatment needed after CAS results in a

relatively small increase in the economic indicator due to the simplicity of the design and the low-

cost treatment processes. However, the additional treatment increased the VRR significantly

43

(enough to offset all the capital and O&M costs associated with the reuse scenarios). Currently, the

major driver for implementation of DPR is severe drought due to the lack of enough regulations

and guidelines for DPR and the social acceptance concerns. This study showed that DPR for the

studied area is one of the best alternatives for supplementing water supply, based on different

dimensions of sustainability.

2.3.4 Sensitivity Analysis for DPR

Although DPR obtained the best regret score among reuse scenarios, increasing the

distance between the water reclamation facility and water treatment location, as well as increasing

the complexity of the additional treatment requirements had a significant influence on the regret

score of this reuse scenario. These two parameters not only affected the final capital and O&M

costs (ASNPV), they also affected the CFP associated with this reuse type.

Among different reuse scenarios, the selection of reuse location for DPR is highly restricted

by the location of water treatment plants and the flexibility of reuse location is usually much higher

for other reuse types. As Figure 2.7 shows, if the distance between water reclamation and water

treatment plant increases by 6.17 miles, DPR will not be the best reuse scenario based on the base

case regret score and IPR will become the best reuse type. Moreover, in some cases (for instance

when the quality requirements for DPR are higher and/or the reclaimed water has lower quality),

the treatment trains for DPR become more complex and it increases the associated cost with the

additional treatments significantly. As it can be seen in Figure 2.6, if the ASNPV associated with

the additional treatment processes increases form 1,712 $/MG to 26,809 $/MG, IPR will be a better

option than DPR. If the ASNPV of the additional treatment increases to $43,869/MG, agricultural

reuse will also obtain a better base case regret score than DPR. Although a 6.17 miles increase in

the distance between water reclamation and water treatment facilities is possible, a 26,809 $/MG

44

increase in ASNPV for additional treatments doesn’t seem realistic. According to the City of San

Diego’s report, in case of implementing an additional advanced water purification facility for IPR

and DPR, consisting of membrane filtration, reverse osmosis, UV disinfection, and advanced

oxidation, the ASNPV does not exceed $4,010/MG (City of San Diego, 2013).

Figure 2.6 Location and treatment analysis for direct potable reuse (DPR) scenario.

2.3.5 Limitations and Future Work

One limitation of this study is the treatment process considered for DPR. For this scenario,

only a few additional treatment processes were added after secondary treatment and treatment by

artificial wetlands (i.e., ultra-filtration, UV/H2O2, and chlorination). DPR treatment can include

more extensive treatment, which would result in different (likely higher) impacts. Accordingly,

future work can consider a sustainability evaluation of existing DPR treatment trains. Further

investigations can be conducted to evaluate the influence of the degree of decentralization on water

reuse options. The last two scenarios offered insight about decentralizing treatment to some extent,

however, the analysis does not reflect the full spectrum of decentralization that can be considered

45

(e.g., at the household- or building-level to large-scale WWT). Moreover, the effects of

decentralization of water reuse and wastewater treatment on the economic and environmental

impacts of the entire water system (e.g., including the freshwater withdrawn, water treatment and

its distribution) was outside of the scope of this study.

Although most of the data used for the design of reuse scenarios was obtained from the

previous construction projects in Polk County and the practical feedback from the City of

Lakeland’s officials, there were assumptions when the real data was missing (e.g. additional

treatments for DPR). However, the conducted sensitivity analyses addressed some aspects of the

uncertainty by showing robustness of the recommended solutions. An uncertainty analysis could

be conducted to further address this limitation, which was outside the scope of this study.

2.4 Conclusion

This study presented a multi-criteria evaluation of the sustainability of water reuse

scenarios, in which the City of Lakeland in Florida was used as a case study to design the city’s

integrated water system. The results of this study revealed that the distance between the water

reclamation facility and the end use played a significant role in economic and environmental

indicators. Increasing the average distance from 0.9 miles to 6.5 miles, with the same degree of

treatment for urban reuse and agricultural reuse, increased the CFP from 1,781 kg CO2-eq/MG to

8,684 kg CO2-eq/MG, while it increased the ASNPV from $413 to $1,667 respectively. The higher

reclaimed water quality required an increase in the complexity of the treatment processes, and

consequently increased the economic impact (ASNPV) and CFP. Higher water quality, however,

improved the EU of water reuse as well as the value of resource recovery significantly, and it

increased the final regret score. The higher value of resource recovery could also offset all the

capital and O&M costs associated with the treatment and distribution for DPR in the case study.

46

Considering this fact, DPR obtained the best regret score among the five alternatives, but the lack

of existing regulations and guidelines for its implementation, high water quality requirements, as

well as challenges with social acceptance, led stakeholders and officials to lose interest in this

water reuse scenario. Moreover, the sensitivity analysis revealed that if the distance between water

reclamation and water treatment plants increased by 6.17 miles, or the ASNPV associated with the

additional treatment requirements increased by 25,097 $/MG, DPR would not be the best reuse

scenario. Agricultural reuse obtained the best score in terms of both the individual economic and

environmental impact (i.e., CFP). Due to its ease of implementation, less complexity in design and

more flexibility in the end-use locations, this scenario received more attention from stakeholders.

Although the results of this study are case-specific, the factors that impact the sustainability

indicators, the tradeoff analysis, as well as the proposed regret-based decision making approach

can be applied for water reuse scenario analysis in other cases. This study also revealed that multi-

criteria assessments could provide decision-makers and officials with powerful tools in the

decision-making process. The results of this study showed the importance and influence of

bringing environmental and social aspects into account, in addition to adopting different weighting

strategies that depends on the stakeholders’ preferences.

47

CHAPTER 3: A MULTI-OBJECTIVE OPTIMIZATION MODEL FOR WATER

RECLAMATION PLANNING2

3.1 Introduction

Water management is facing a diverse set of stressors from both demand (e.g., population

growth, urban development, industrialization) and supply (e.g., climate change, water source

contamination). To tackle such challenges, alternative water sources, such as treated wastewater,

are investigated to alleviate pressure on local natural water supplies (US EPA, 2012). Historically,

the primary goal of wastewater treatment plants (WWTPs) was to protect the environment and

public health by reducing and removing pollutants (e.g., organic materials, pathogens, and

nutrients) (Audrey D. Levine and Asano, 2004). However, due to the emerging challenges of water

management along with the potential economic and environmental benefits associated with

resource recovery (e.g., water, energy, and nutrients) (Diaz-Elsayed et al., 2019) reclaimed water

infrastructure has become a vital component of sustainable water system management. The

presence of water reclamation facilities, alongside other facilities in water systems, such as water

extraction and treatment systems, necessitates adoption of a reverse logistics and product recovery

approach.

2 Chapter 3 has been submitted to the Journal of Cleaner Production and is in the review process for publication. Attributions: I conducted the study and led the authorship of the manuscript. Co-authors Alvaro Sierra-Altamiranda and Hadi Charkhgard contributed assistance in developing the optimization model. Co-author Nancy Diaz-Elsayed contributed assistance in data analysis for the results section and assistance in manuscript review. Co-author Qiong Zhang contributed guidance on conducting the study and assistance in manuscript preparation and review.

48

Reverse logistics considers moving used products from disposal to reuse purposes,

therefore forming a closed-loop supply chain. The field is receiving growing attention for solving

supply-demand problems in a variety of production activities such as product manufacturing, solid

waste management and recycling, and water supply networks (Fleischmann et al., 2001). For

design and evaluation of such closed-loop systems, mathematical models have been developed to

find the optimal configurations and operational conditions (F. Habibi et al., 2017).

While the focus of previous models in the design of manufacturing networks is on

maximizing the system’s profit and minimizing the associated costs, consideration of

environmental and social aspects in this field is less common. Some studies developed mixed-

integer and stochastic programming models to design reverse product manufacturing networks;

for example, selecting the location and capacity of (used) product collection centers (Jeihoonian

et al., 2016a, 2016b; Lieckens and Vandaele, 2012); the location and capacity of remanufacturing

facilities, and the flows between the network facilities (Lieckens et al., 2013); the rate of product

recovery, material recycling, and component repair, the allocation of products to be

remanufactured, and the selection of location and capacity for the associated facilities (John et al.,

2018b); the rate of production and return of used products (Fattahi and Govindan, 2017); and the

location of facilities in the supply chain network (i.e., production, collection, and recycling centers)

with optimized capacities and flows (S Soner Kara and Onut, 2010; Nakatani et al., 2017) (see

Figure 3.1a).

49

Figure 3.1 Similarities and differences between reverse logistics in (a) production activities, (b) municipal solid waste management, and (c) water systems.

In the past decade, reverse logistics has been extended from product supply chain

management to environmental management (e.g., municipal solid waste management (see Figure

3.1b)) (Habibi et al., 2017). Lyeme et al. (2016) developed a mixed-integer programming

formulation to minimize the cost and greenhouse gas (GHG) emissions of the municipal solid

waste management system in Dar es Salaam City, Tanzania. The model was capable of finding the

optimal location for the recycling plants, landfill, and incineration plants, as well as the waste flow

allocation between them. Toso and Alem (2014) also developed a single-objective optimization

model to minimize the cost associated with the solid waste management and recycling system in

Sorocaba, Brazil via the facility location-allocation decisions. The proposed model successfully

designed a waste management and recycling system that fell within the city’s financial budget.

Considering the similarities between production activities, municipal solid waste

management, and water management, it is evident that reverse logistics can be applied to water

supply systems and lead to the formation of a closed-loop water supply chain (Figure 3.1c). The

50

adoption of the supply chain design in water management, however, is limited and most studies

focused on technology selection as shown in Table 3.1. Abdulbaki et al. (2017) considered both

treatment capacity and flowrate in addition to treatment technology for water resources

management; the model incorporated costs and global warming potential with a normalization

scheme. Chung et al. (2008) used a system dynamics (SD) model to evaluate several possible

scenarios applied to a hypothetical water system. The model considered costs and water quality

for the selection of the treatment facilities’ location and the allocation of flow between the facilities

and customers. In wastewater management, Ruiz-Rosa et al. (2016) developed a single objective

model to minimize the costs associated with the design and operation of the wastewater treatment

system. The model was capable of evaluating the water reuse scenario vs. other existing water

resources (i.e., surface water, groundwater, and desalination).

Table 3.1 Review of previously developed optimization models related to the study area

Study

Number of objective(s) Sustainability indicators Decision variables

Single objective

Multi objective Economic Environmental Social Location Technology Capacity Flow

Water supply/treatment

Abdulbaki et al. (2017) X X X X X X

Al-Zahrani et al. (2016)

X X X X X

X X

Chung et al. (2008) X

X X

X

X

Wastewater treatment/reuse

Hreiz et al. (2015)

X X X

X

Kim et al. (2015)

X X X

X

Molinos-Senante et al. (2015)

X X X

X

Ruiz-Rosa et al. (2016) X

X

X

Zhang et al. (2014)

X X X

X

Proposed model X X X X X X X X

As Table 3.1 shows, when multiple sustainability indicators are considered, researchers

have developed either multi-objective or single-objective models. Although multi-objective

models exhibit greater complexity in both the formulation and solution phases, they provide

decision-makers with the ability to consider the tradeoffs between sustainability indicators as they

51

make a final selection among the optimal solutions. Several multi-objective models have been

developed that incorporate costs and environmental impacts in the selection of treatment options

for wastewater systems management (i.e., Hreiz et al., 2015; Kim et al., 2015; Molinos-Senante et

al., 2015; Zhang et al., 2014). However, reliance on solely the economic or environmental aspects

can mask sustainably optimal solutions and mislead management decisions (Owens, 2001). Only

the study by Al-Zahrani et al. (2016) incorporated the three dimensions of sustainability (i.e., costs,

environmental, and social). The model considered capacity and flowrate for different alternative

water resources and was formulated as a multi-objective minimization problem. However, the

model was not designed for selection of water treatment technology. The lack of including the

social dimension could be due to the complexity in quantifying social impacts (Mani et al., 2016).

While the primary focus of the studies concerning wastewater system management is the

selection of the treatment technology, the models developed thus far have not simultaneously

selected the optimal location of the WWTP(s), plant capacity, and flow to and from end users,

which are important design parameters. Moreover, the mentioned emerging challenges in water

demand and supply and increasing importance of sustainability in the design of such systems

(Little et al., 2016; Seidel et al., 2014), highlight the need for a model that accounts for the

economic, environmental, and social dimensions, concurrently. The goal of this part of the study

was, therefore, to develop a multi-objective optimization model that assists decision-makers with

site and technology selection, capacity allocation, and customer distribution for the design of

sustainable wastewater systems to facilitate water reclamation.


In this part of the study, a multi-objective optimization model was developed for

wastewater system management. The model includes the wastewater collection system, water

52

reclamation facilities (wastewater treatment plants), reclaimed water distribution network, and

population clusters. The raw wastewater generated from population clusters is collected by the

sewer network and transferred to the reclamation facilities. After wastewater treatment, some

portion of the reclaimed water is distributed back to the population clusters via the reclaimed water

distribution network for non-potable reuse purposes such as landscape irrigation.

The multi-objective optimization model was used to determine characteristics of the

wastewater treatment facility (location, technology implemented, design capacity, and average

flowrate), service area (number of WWTPs), sewer network (flow from each population cluster to

each WWTP), and reclaimed water distribution network (flow of reclaimed water from each

WWTP to each population cluster). The model minimized the total costs (implementation and

operation) and the GHG emissions associated with the operation phase and maximized the value

of resource recovery (correlated to the effluent water quality) to capture the social impact of the

design.

3.2.1 Deterministic Optimization Model

The proposed mathematical formulation consisted of two objective functions – the

definition of each parameter can be found in the nomenclature (Appendix A). The first objective

function (equation 3.1) minimized the costs associated with the implementation, operation, and

maintenance of the wastewater system and maximized the benefits associated with water

reclamation. This objective function included the capital costs associated with the implementation

of the WWTPs (CC), the overall operation and maintenance (O&M) costs (e.g., energy, labor,

chemicals) associated with the WWTPs (OC), the costs associated with wastewater collection

(CTW), and the costs associated with reclaimed water distribution (CTR). The last term (SRW) in

the first objective function was considered as a social indicator accounting for the value of

53

reclaimed water that depended on effluent water quality. Due to complexity in measurement and

quantification of social impacts, researchers have considered variety of indicators to include this

aspect into their assessment studies, such as working conditions for employees (Palme et al., 2005),

job and study opportunities (Akhoundi and Nazif, 2018), the ability to reducing diarrheal illness

and deaths (Ren et al., 2013), public acceptance (Lamichhane and Babcock, 2013; Mainali et al.,

2013; Rygaard et al., 2014), the service providers’ past experience (Sharma et al., 2009), and

operation expertise requirement (Tjandraatmadja et al., 2013). Since the success of water

reclamation and reuse in practice highly depends on public acceptance, the value of resource

recovery (VRR) (Rezaei et al., 2019), or willingness to pay for the reclaimed water, was considered

as the social benefits of the system in this study. The value of reclaimed water was incorporated

in the first objective function with a negative sign to maximize the social benefits of the system.

Based on the assumptions, only a portion (α) of the wastewater generated by customers in the

clusters will be returned and sold to them as reclaimed water.

Minimize 𝑍𝑍𝐶𝐶 = ∑ ∑ ∑ (𝐶𝐶𝐶𝐶𝑘𝑘𝑑𝑑𝜇𝜇𝐴𝐴𝑟𝑟𝑞𝑞𝑟𝑟𝑘𝑘/𝑛𝑛)𝜔𝜔𝑘𝑘𝑑𝑑𝑑𝑑𝑘𝑘𝑟𝑟 + ∑ ∑ ∑ 𝑂𝑂𝐶𝐶𝑑𝑑𝑑𝑑𝑘𝑘𝑟𝑟 𝜇𝜇𝐴𝐴𝑟𝑟𝑞𝑞𝑟𝑟𝑘𝑘 +

∑ ∑ (𝐶𝐶𝐶𝐶𝐶𝐶)𝐴𝐴𝐶𝐶𝐶𝐶𝑟𝑟𝑘𝑘𝐿𝐿𝑟𝑟𝑘𝑘𝑘𝑘𝑟𝑟 + ∑ ∑ (𝐶𝐶𝐶𝐶𝑅𝑅)𝐴𝐴𝑅𝑅𝐶𝐶𝑘𝑘𝑟𝑟𝐿𝐿𝑘𝑘𝑟𝑟 𝑟𝑟 − ∑ ∑ ∑ 𝐴𝐴𝑅𝑅𝐶𝐶𝑑𝑑𝛼𝛼𝜇𝜇𝐴𝐴𝑟𝑟𝑞𝑞𝑟𝑟𝑘𝑘𝑑𝑑𝑘𝑘𝑟𝑟𝑘𝑘 (3.1)

The second objective function (equation 3.2) minimized the GHG emissions of the

wastewater system, as the environmental footprint associated with the design. GHG emissions is

an abstract environmental sustainability indicator (ESI) used to characterize the impact on global

climate change (Qi and Chang, 2013). In this study, GHG emissions were calculated using

emission rate data from eGRID (Brander et al., 2011; US EPA, 2017) and the energy consumed

during the O&M phase, which has been found to be the dominant contributor in the water and

wastewater industries (Loubet et al., 2014; Pintilie et al., 2016). Energy consumed for the operation

of wastewater treatment plants (GP), wastewater transport from population clusters to reclamation

54

facilities (GTW), and reclaimed water conveyance from reclamation facilities to customers (GTR)

was accounted for.

Minimize 𝑍𝑍𝐸𝐸 = ∑ ∑ ∑ 𝐺𝐺𝐴𝐴𝑑𝑑𝜇𝜇𝐴𝐴𝑟𝑟𝑞𝑞𝑟𝑟𝑘𝑘𝜔𝜔𝑘𝑘𝑑𝑑𝑑𝑑𝑘𝑘𝑟𝑟 + ∑ ∑ (𝐺𝐺𝐶𝐶𝐶𝐶)𝐴𝐴𝐶𝐶𝐶𝐶𝑟𝑟𝑘𝑘𝐿𝐿𝑟𝑟𝑘𝑘𝑘𝑘𝑟𝑟 + ∑ ∑ (𝐺𝐺𝐶𝐶𝑅𝑅)𝐴𝐴𝑅𝑅𝐶𝐶𝑘𝑘𝑟𝑟𝐿𝐿𝑘𝑘𝑟𝑟𝑟𝑟𝑘𝑘

(3.2)

The first objective function had a multi-linear term obtained from the product of two

variables (qrk and ωkt). This multi-linear term made the model a non-linear optimization model. To

linearize the model, a McCormick envelope (relaxation of the bi-linear term using a new variable

and the addition of 3 constraints for each created variable) was used. Since the McCormick

envelope is a relaxation, the result is an approximation; however, considering that the variables qrk

and ωkt are binary, the approximation was exact for this case (Boland et al., 2015). The constraints

are summarized in Table 2. The optimization model was coded in C++ software and run using an

Intel® Core™ i7-4790 CPU @ 3.60GHz × 8 with 7.7 GiB of memory and Ubuntu 16.04 LTS

operative system. The optimizer used for the solution of the model was CPLEX 12.7 and the

Triangle Splitting Method was used for the multi-objective optimization algorithm (Boland et al.,

2015).

55

Table 3.2 Constraints associated with the optimization model

Subject to:* 1)** ∑ ∑ 𝐺𝐺𝐴𝐴𝑑𝑑𝐶𝐶𝐴𝐴𝐴𝐴𝑘𝑘𝑑𝑑𝜔𝜔𝑘𝑘𝑑𝑑𝑑𝑑𝑘𝑘 + ∑ ∑ (𝐺𝐺𝐶𝐶𝐶𝐶)𝐴𝐴𝐶𝐶𝐶𝐶𝑟𝑟𝑘𝑘𝐿𝐿𝑟𝑟𝑘𝑘𝑘𝑘𝑟𝑟 + ∑ ∑ (𝐺𝐺𝐶𝐶𝑅𝑅)𝐴𝐴𝑅𝑅𝐶𝐶𝑘𝑘𝑟𝑟𝐿𝐿𝑘𝑘𝑟𝑟𝑟𝑟𝑘𝑘 ≤ 𝑀𝑀𝐺𝐺

Limits the GHG emissions of the wastewater treatment plants, the wastewater collection and transfer network, and the reclaimed water distribution system, in case of an existing regulation and/or user’s preference that limits emission from the entire system.

2) 𝑞𝑞𝑟𝑟𝑘𝑘 ≤ ∑ 𝜔𝜔𝑘𝑘𝑑𝑑𝑑𝑑 ∀ 𝑟𝑟, 𝑘𝑘 Ensures that wastewater from a population cluster is not transported to a facility that does not exist in candidate location k.

3) ∑ 𝑞𝑞𝑟𝑟𝑘𝑘𝑘𝑘 = 1 ∀ 𝑟𝑟 Guarantees that produced wastewater in each population cluster will be allocated and transported to at least one existing wastewater treatment plant.

4) ∑ 𝜇𝜇𝐴𝐴𝑟𝑟𝑞𝑞𝑟𝑟𝑘𝑘𝑟𝑟 ≤ ∑ 𝐶𝐶𝐶𝐶𝑜𝑜𝑘𝑘𝑑𝑑𝜔𝜔𝑘𝑘𝑑𝑑𝑑𝑑 ∀ 𝑘𝑘 Maintains the capacities of the WWTP with technology and scale t established at candidate location k (i.e., the total wastewater flow entering each reclamation facility cannot exceed the plant’s design capacity).

5) ∑ 𝜔𝜔𝑘𝑘𝑑𝑑𝑑𝑑 ≤ 1 ∀ 𝑘𝑘 Ensures that there is no more than one constructed facility at each candidate location.

6) 𝜇𝜇𝐴𝐴𝑟𝑟𝑞𝑞𝑟𝑟𝑘𝑘 ≤ 𝐴𝐴𝐶𝐶𝐶𝐶𝑟𝑟𝑘𝑘 ∀ 𝑟𝑟, 𝑘𝑘 Ensures that the wastewater treatment plant constructed at candidate location k has enough capacity to receive the amount of wastewater required to be transferred from population cluster r.

7) 𝛼𝛼(𝜇𝜇𝐴𝐴𝑟𝑟𝑞𝑞𝑟𝑟𝑘𝑘) ≤ 𝐴𝐴𝑅𝑅𝐶𝐶𝑘𝑘𝑟𝑟 ∀ 𝑘𝑘, r Ensures that the amount of reclaimed water required to be transferred to the customer cluster r is produced in the reclamation facility at candidate location k.

8) 𝜔𝜔𝑘𝑘𝑑𝑑 , 𝑞𝑞𝑟𝑟𝑘𝑘 ∈ {1, 0} Defines the type of different decision variables used in the model. Note that variables related to the facilities’ location, technology, and related population clusters are binary.

9) 𝐴𝐴𝐶𝐶𝐶𝐶𝑟𝑟𝑘𝑘 , 𝐴𝐴𝑅𝑅𝐶𝐶𝑘𝑘𝑟𝑟 ≥ 0&𝐼𝐼𝑛𝑛𝑜𝑜𝐼𝐼𝐼𝐼𝐼𝐼𝑟𝑟 ∀ 𝑘𝑘, 𝑟𝑟, 𝑜𝑜 Defines the type of different decision variables used in the model. Note that variables defining the flow rates are integer.

10)*** ∑ 𝐴𝐴𝐶𝐶𝐶𝐶𝑟𝑟𝑘𝑘𝑟𝑟 ≤ 16580103.6 𝑘𝑘 = 1 Restricts the capacity of the Falkenburg water reclamation facility (limited land availability).

11)*** ∑ 𝐴𝐴𝐶𝐶𝐶𝐶𝑟𝑟𝑘𝑘𝑟𝑟 ≤ 13816753.0 𝑘𝑘 = 3 Restricts the capacity of the South County water reclamation facility (limited land availability).

* The definition of each parameter can be found in the nomenclature (Appendix A). ** The constraint does not apply to the study area in the case study. *** The last two constraints are specifically related to the case study in this article.

3.2.2 Case Study Background

Hillsborough County in Florida was considered for this study. It has a population of

1,376,238 with an overall growth rate of 12%; it is the fourth most populous county in Florida and

the most populous county outside of the Miami Metropolitan Area (United States Census Bureau,

2017). The County has three water service areas consisting of the Northwest region, Southcentral

56

region, and the City of Tampa (see Figure 3.2). The County owns and operates four WWTPs in

the Northwest service area with a total treatment capacity of 27.7 MGD (see Figure C.1 in

Appendix C). Hillsborough County Public Utilities Department is planning to consolidate

operation of the majority of wastewater treatment plants in the Northwest water service area by

abandoning three smaller WWTPs due to aging infrastructure and the expansion of the Northwest

Regional Water Reclamation Facility (Hillsborough County, 2017b). The City of Tampa has one

advanced wastewater treatment facility (Howard F. Curren Advanced Wastewater Treatment

Plant) with a total design capacity of 96 MGD and average daily flow rate of 55 million gallons

(City of Tampa, 2018).

In the Southcentral water service area, the County owns three water reclamation facilities

with a total capacity of 34 MGD. It is forecasted that in 2025, wastewater production in the

Southcentral service area will exceed the current total capacity of wastewater treatment plants.

Hence, the County is looking for the optimal location and treatment technology for a new

wastewater treatment facility (Hillsborough County, 2017b). As shown in Figure 3.2, the

Southcentral water service area has 19 major population clusters. The County also has purple

pipelines implemented in this service area for non-potable reuse to residential and commercial

customers.

57

Figure 3.2 Hillsborough County boundary, County’s water service areas, the County's major cities, and the candidate locations for implementation and operation of WWTPs in the Southcentral water service area.

3.2.3 Scope and Input Data

The focus of this study was on the Southcentral water service area in Hillsborough County,

Florida, from the generation of raw wastewater in the population clusters to treatment and the

distribution of reclaimed water to end users. The data regarding the population clusters in the

county and population growth for each cluster was obtained from US Gazeteer (2017) and is

available in Table C.1 in Appendix C. The candidate locations for the new facilities were

determined using the information provided by County officials and feedback from stakeholders.

The model calculates the distance between the nodes (candidate locations for implementation of

WWTPs and customer clusters) using the latitude and longitude for each node (see Table C.2 in

Appendix C). Since the wastewater collection system (sewer network) and reclaimed water

distribution network (also known as purple pipeline) were already implemented in the study area,

the shortest pipeline path in the main sewer/purple line between population clusters and

58

reclamation facilities was considered to calculate the requirements for water transfer in the study

area (pumping requirements and maintenance of the pipeline networks).

The options for treatment technologies were selected based on the new requirements for

secondary wastewater treatment in the State of Florida (State of Florida, 2018), and consisted of

Bardenpho (a dissolved oxygen bioreactor) followed by filtration, a membrane bioreactor (MBR)

followed by micro-filtration (MF), MBR followed by reverse osmosis (RO), MBR followed by

ultra-filtration (UF), or conventional activated sludge (CAS) followed by granular activated carbon

(GAC) and filtration (see Table C.3 in Appendix C). Small, medium, and large-scale systems were

considered for each technology based on the classification suggested by Diaz-Elsayed et al. (2019).

All of the selected technologies meet the treatment requirements in the State of Florida, and

treatment technologies with higher effluent water quality were also considered to evaluate the

impacts of added value associated with reclaimed water on the model outputs.

Information needed to calculate the capital costs, O&M costs, and energy requirements

associated with each treatment train at each scale, were obtained from County reports and literature

data (e.g., Abdelrasoul et al., 2016; Cashman et al., 2018; Chang et al., 2008; Cornejo et al., 2016;

Gabarrell et al., 2012; Guo et al., 2014; Holloway et al., 2016; Marufuzzaman et al., 2015; Ortiz

et al., 2007; Tang et al., 2018; Tarnacki et al., 2012; Zhou et al., 2011) (see Table C.4 in Appendix

C). The pumping electricity was estimated based on previous projects in the county and

engineering handbooks (NCEES, 2013) (see Table C.5 in Appendix C). The value of resource

recovery (the willingness to pay) was based on the average effluent water quality for each

treatment technology and Polk County and Hillsborough County’s reclaimed water prices

(Hillsborough County, 2017a; Polk County, 2017; Rezaei et al., 2019). The effluent water quality

for each treatment train was compared to the water quality requirements for non-potable and

59

potable reuses. The price for drinking water and the price of reclaimed water for non-potable

purposes are available in Hillsborough and Polk County. The required additional treatments to

achieve the quality requirements for non-potable urban and potable reuses were considered to

estimate and the price of reclaimed water with different qualities for the model. Specifically, IT3PR

Integrated Treatment Train Toolbox was used for the estimation of required additional treatments

for potable reuse (Trussell et al., 2015). The detail information regarding the value of resource

recovery for each treatment train at each scale can be found in Table C.4 in Appendix C.

3.2.4 Degree of Decentralization and Marginal Benefit Calculations

In order to quantify the degree of decentralization of treatment facilities, the normalized

sample variance of the WWTPs’ design capacity was used; see equation 3.3:

𝐴𝐴𝐴𝐴2 =∑ (𝑋𝑋𝑖𝑖− 𝑋𝑋𝑎𝑎𝑎𝑎𝑎𝑎)2𝑛𝑛𝑖𝑖=1(𝑁𝑁−1) (𝑇𝑇𝐶𝐶)2

(3.3)

where NS2 is the normalized sample variance of treatment capacities; N is the total number of

candidate locations for new and existing WWTPs; X is the capacity of WWTPs implemented in

the candidate locations, which can be zero or greater; Xavg is the average of the WWTP capacities;

and TC is the total required treatment capacity in the water service area. The normalized sample

variance represents how the total required treatment capacity is distributed across the candidate

locations for implementation of water reclamation facilities. A higher sample variance would

represent more centralized systems (a less distributed pattern for the total required treatment

capacity), while a lower sample variance would represent a higher degree of decentralization of

treatment facilities.

Since the model used two objective functions, equation 3.4 was used to identify the highest

achievable benefit with the lowest increase in associated costs, the marginal benefit, ME (i):

60

𝑀𝑀𝐸𝐸 (𝑖𝑖) = − Δ (GHG emissions)Δ (Total costs)

= − 𝑍𝑍𝐸𝐸 (𝑖𝑖)− 𝑍𝑍𝐸𝐸 (𝑏𝑏𝑎𝑎𝑏𝑏𝑏𝑏𝑏𝑏𝑖𝑖𝑛𝑛𝑏𝑏)

𝑍𝑍𝐶𝐶 (𝑖𝑖)− 𝑍𝑍𝐶𝐶 (𝑏𝑏𝑎𝑎𝑏𝑏𝑏𝑏𝑏𝑏𝑖𝑖𝑛𝑛𝑏𝑏) (3.4)

where i is an optimal solution; ZE (i) and ZC (i) are the environmental and economic impacts in ton

CO2-eq/year and M$/year, respectively; and baseline represents the solution with the minimum

costs and maximum GHG emissions.

3.2.5 Sensitivity Analysis

To evaluate the influence of input parameters on the optimal solutions, a sensitivity

analysis was conducted. In cases where long distance water transfer and/or high elevation variation

in the water service area exists, the costs and environmental footprints associated with wastewater

collection and reclaimed water distribution have been found to dominate the costs and

environmental footprints associated with the operation of WWTPs (Byrne et al., 2017; Lyons et

al., 2009; Rezaei et al., 2019). Moreover, the decision to decentralize plants strongly depends on

population density and other local conditions such as the topography of the area (Chung et al.,

2008; Lehtoranta et al., 2014). Accordingly, a sensitivity analysis was conducted to evaluate the

effect of 1) the energy requirements for water transfer, which affect the water transfer costs and

GHG emissions and 2) the population density on the configuration of water reclamation system.

Although the study area was flat, the energy requirement for water transfer was increased in

increments of 10% to represent a more hilly topography, considering the fact that higher elevation

variation only impacts the wastewater collection and reclaimed water distribution requirements

(Kavvada et al., 2016a) in the developed model. Changes in the population density were also

evaluated in increments of ±10%.


As it can be seen in Figure 3.3, solving the multi-objective optimization model resulted in

50 optimal solutions for the study area. The model generated a spreadsheet for each optimal

61

solution, representing the recommended reclamation facilities in each candidate location; the

technology, scale, and operation capacity at each facility; the allocation of customers to each

reclamation facility; and decision variables regarding the amount of wastewater and reclaimed

water that needs to be collected, transferred, and sold to customers. The details regarding the

optimal solutions can be found in Appendix C (Tables C.6 and C.7). Since the model was multi-

objective, decision-makers should make the final selection among the optimal points based on the

tradeoff for cost and environmental footprint.

Figure 3.3 Optimal solutions for the multi-objective optimization model for the study area.

As Figure 3.3 shows, the environmental footprint associated with the design can be

decreased by changing the system’s configuration towards more costly options (from optimal

solution 1 to 50). Moreover, increasing the investment costs by as much as 47 M$/year from

solution 1 to solution 24, decreased the environmental footprint significantly (6,332 metric ton

CO2-eq/year), however, increasing the costs from solution 24 to solution 50 did not make a

significant improvement in the environmental impact associated with the design. Solution 24

provided the greatest marginal benefit in this study (see Figure S2 in the supplementary material).

62

The configuration of the least expensive solution (solution 1), the most environmentally

friendly solution (solution 50), and the solution with the highest marginal benefit (solution 24) are

shown in Figure 3.4 and the detailed information regarding those solutions can be found in Table

C.8 in Appendix C.

Figure 3.4 Configuration of the least expensive solution (solution 1), the most environmentally friendly solution (solution 50), and the solution with the highest marginal benefit (solution 24). Abbreviations: CAS: conventional activated sludge; GAC: granular activated carbon; MBR: membrane bioreactor; MF: micro-filtration; MGD: million gallons per day.

63

As it can be seen in Figure 3.4, for the least expensive optimal solution, the model selected

eight candidate locations for implementation of wastewater treatment facilities. The sample

variance of treatment capacities was 0.01343 and the treatment technology of CAS+GAC was

selected in all of the candidate locations because it has lowest capital and O&M costs (see Figure

3.5). However, the selection of CAS+GAC, which has the highest GHG emissions among the

candidates for implementation of treatment technology, resulted in the highest environmental

impact associated with the least expensive optimal solution.

The model achieved the highest marginal benefit (solution 24) by selecting MBR+MF for

the treatment technologies in the eight selected candidate locations in solution 1 while keeping the

same degree of decentralization for implementation of WWTPs. As it can be seen in Figure 5,

although the treatment costs of MBR+MF is slightly higher than CAS+GAC, the GHG emissions

associated with this treatment technology is much lower and it decreased the overall environmental

impact of the system significantly. The optimal solution with the lowest environmental impact

(solution 50) had a lower degree of decentralization of WWTPs (sample variance of 0.02599)

compared to solutions 1 and 24, and Bardenpho technology, which has the lowest GHG emissions

among the candidates for implementation of treatment technology, was selected in all of the

candidate locations. There was not a significant reduction in the environmental impact of the

integrated wastewater system in comparing solution 24 to solution 50, although costs increased

significantly.

64

Figure 3.5 Comparison of two selected treatment capacity in (a) medium-scale (2.0 MGD) and (b) large-scale (10.0 MGD) for the designed treatment trains in the study. Abbreviations: CAS: conventional activated sludge; GAC: granular activated carbon; MBR: membrane bioreactor; RO: reverse osmosis; MF: micro-filtration; UF: ultra-filtration; GWP: global warming potential (greenhouse gas emissions); MGD: million gallons per day.

Since the wastewater collection and reclaimed water distribution networks already existed

in the study area, the costs and environmental footprint associated with installation of pipelines in

the water service area were not considered in this case study. However, the costs associated with

the required maintenance of the existing pipelines during the 33 years lifetime were considered in

the inputs to the model. Hence, the costs associated with wastewater collection and reclaimed

water distribution (energy requirements, implementation of pumping stations, and maintenance of

65

pipelines in the first objective function) and environmental impacts associated with the

implementation and operation of WWTPs (in the second objective function) were the major

contributors to the design of the optimal solutions in this case study. Figures 3.6a and 3.6b show

the breakdowns of the contributors to the first and second objective function in the study area.

Figure 3.6 Breakdown of the contributors to the (a) first and (b) second objective function. Abbreviations: ANPV: Annualized net present value; VRR: value of resource recovery; WW: wastewater; RW: reclaimed water; O&M: operation and maintenance; M$: million dollars; GWP: global warming potential (greenhouse gas emissions).

66

3.3.1 Decentralization of Treatment

With regards to the scale of implementation of the treatment facilities, the objective

functions tended towards opposing solutions. Minimizing the costs and environmental footprint

associated with the treatment phase would leverage economies of scale for treatment and establish

larger reclamation facilities, which have lower treatment costs and GHG emissions per unit volume

of wastewater, closer to populous areas. Furthermore, minimizing the costs and environmental

footprint associated with water transfer phase (wastewater collection and reclaimed water

distribution) would tend to decentralize treatment facilities to lower the water transfer

requirements (pumping energy and pipeline costs). As it can be seen in Figure 3.2 and Figure 3.4,

the population in Hillsborough County is distributed in a relatively large area in the Southcentral

region. This distributed population pattern along with the long distances between customers and

facilities, increases the costs and GHG emissions associated with water transfer significantly; in

fact, it overshadows the economies of scale for treatment facilities in this water service area.

Ultimately, the domination of the costs associated with wastewater collection and reclaimed water

distribution in the first objective function (see Figure 3.6a) and the simultaneous consideration of

economic and environmental impacts favored decentralization of treatment facilities for all of the

optimal solutions in the study area.

Although several previous studies have also shown that decentralization of treatment

facilities improve the costs and environmental footprints associated with the wastewater systems

(e.g., Chung et al., 2008; Kavvada et al., 2016), implementation of larger-scale centralized systems

is more favored by municipalities and counties. The stakeholders in this study indicated some

challenges associated with the implementation of decentralized treatment systems consistently.

The mentioned challenges included complication in management and monitoring, siting concerns

67

of prospective neighboring properties, operation permits and regulatory compliance, public

opposition, land availability, availability of certified operators and laboratories, reduction in the

ability to reclaim water in some cases (e.g., large-scale agricultural reuse), losing economies of

scale, and low acquisition logistics.

3.3.2 Selection of Treatment Technology

As it was mentioned before, all of the treatment technologies selected for this study meet

the effluent water quality requirements for non-potable urban reuse. However, the treatment

technologies with higher implementation, operation, and maintenance costs typically offer a higher

effluent water quality (e.g., MBR+RO and MBR+UF) and it increases the value of resource

recovery (the willingness to pay for the reclaimed water) (see Figure 3.5). Moreover, these

technologies usually consume more energy for operation, which consequently increases the

environmental footprint associated with the operation phase. The treatment technologies with

lower investment and operational costs (e.g., Bardenpho, MBR+MF) offer a lower removal

efficiency and lower effluent water quality, accordingly. These technologies also consume less

energy for treatment, and it reduces the environmental impact associated with the integrated

wastewater system, however, the lower effluent water quality decreases the value of resource

recovery. Among the selected treatment technologies, CAS+GAC offers the lowest

implementation, operation, and maintenance costs, however, it has the highest environmental

footprint associated with the treatment phase. It is mainly due to the high energy consumption by

the aeration pumps for the CAS phase and high frequency and energy intensity of the backwash

processes and regeneration of material between the treatment phases for the GAC phase.

Although the first two terms in the first objective function favored technologies with lower

implementation and operation costs for treatment, the third term, which maximizes the value of

68

resource recovery, favored treatment options with higher effluent water quality. Since the value of

resource recovery (effluent water quality) did not contribute to the second objective function and

the GHG emissions associated with treatment phase was the major contributor to this objective

function (see Figure 6b), the second objective function favored the selection of treatment

technologies with the minimum associated energy requirements (GHG emissions) for

implementation. As it was discussed in the previous section, due to the domination of water

transfer costs in the first objective function, the model kept almost the same pattern for

decentralization of treatment capacities among the optimal solutions. However, due to the

domination of environmental footprint associated with treatment phase in the second objective

function (see Figure 3.6b), the model optimized the total costs and environmental impact by

selecting alternative treatment technologies in the candidate locations (see Figure 3.3 and Table

C.8 in Appendix C). Consequently, replacing the CAS+GAC treatment technology with MBR+MF

from Solution 1 to Solution 24 resulted in a small increase in the cost of the integrated wastewater

system, however, the environmental footprint dropped significantly (by ~96%). Replacing

MBR+MF with Bardenpho (from solution 24 to solution 50) slightly reduced the environmental

impact, however, the capital and O&M costs increased by ~50%. As it can be seen in Figure 6a,

the value of resource recovery for Solution 50 is similar as that for Solution 24 and could not

compensate the increase of treatment costs in the first objective function. Therefore, the advanced

treatment options with relatively higher treatment costs and GHG emissions (i.e., MBR+RO and

MBR+UF) were not selected in this case study, although these technologies resulted in higher

effluent water quality consequently higher value of resource recovery.

69

3.3.3 Sensitivity Analysis

A variation in the energy requirements for water transfer was modeled to investigate its

impact. This resulted in the changes to the selection of treatment technologies and, consequently,

the greenhouse gas emissions of the integrated wastewater system. It also increased the overall

costs; however, no changes were observed in the degree of decentralization. In fact, due to the

domination of the water transfer requirements, the model kept the same decentralized

configuration across the scenarios to lower the overall costs by reducing the distances between

WWTPs and customer clusters. The first objective function opted for the selection of the least

expensive treatment technology (i.e., CAS+GAC), although it had the highest environmental

impact. This technology was selected across all of the candidate locations, as the economic

solution for all scenarios that varied the energy requirements in water transfer. The second

objective function selected the treatment technology with the lowest GHG emissions (i.e.,

Bardenpho) in the most environmentally friendly solutions across the scenarios.

Although the number of optimal solutions varied widely across the sensitivity analysis,

interestingly enough, the solution with the highest marginal benefit incorporated MBR+MF or

CAS+GAC in all of the recommended WWTPs (see Table 3.3). Table 3.3 shows the change in the

configuration of solutions with the highest marginal benefit with the change in the energy

requirements for water transfer from the base case scenario (CG). When the increase in the energy

requirements for water transfer reached 80% (scenario CG+80) the model selected CAS+GAC for

three of the candidate locations that were receiving 49% of the total generated wastewater. Below

this threshold, the first objective function was driving the model solutions by selecting a less

expensive treatment option (i.e., CAS+GAC vs. MBR+MF) to offset a portion of the costs incurred

due to increase in the water transfer requirements. Thereafter (>80% increase in energy

70

requirements for water transfer from the base case scenario), the costs of water transfer

significantly increased in the first objective function and the selection of a less expensive treatment

option did not make a significant contribution in lowering the overall costs. Moreover, while the

environmental impact in the scenario CG+80 was significantly higher than the base case scenario

(more than 1,061%) due to the selection of CAS+GAC for treatment of almost half of the total

generated wastewater, the second objective function became the model’s driver for selection of

treatment technology in the candidate locations. Hence, the model replaced the treatment

technology with lower environmental impact and higher associated costs (i.e., MBR+MF vs.

CAS+GAC) in all of the candidate locations (see Table C.9 in Appendix C).

The bottom line is that the conflict between the objective functions highlights the essence

of multi-objective optimization models, which are capable of simultaneously accounting for

different sustainability indicators, for the design and management of water systems. This not only

helps decision-makers with finding the tradeoffs between the sustainability indicators and selecting

among the optimal solutions accordingly, but also helps with identifying the thresholds, which

influence the optimal configurations. For instance, it was evident that selection of the most

sustainable treatment options was highly influenced by the energy requirements for water transfer

when it reached a certain level (80% increase from the base case scenario in this study). However,

the tradeoffs and thresholds are highly dependent on the case-specific parameters, local conditions,

and the limitations associated with the case study as discussed in Section 3.4.

71

Table 3.3 Impacts of changes in the energy requirement for water transfer on the optimal solution with the highest marginal benefit (costs, environmental footprint, degree of decentralization, and degree of treatment). Abbreviations: MBR: membrane bioreactor; MF: micro-filtration; CAS: conventional activated sludge; GAC: granular activated carbon; ZC: costs associated with the entire wastewater system; ZE: greenhouse gas emissions of the wastewater system

Unlike energy requirements for water transfer, population density had a high influence in

the configuration of optimal solutions (in terms of both decentralization of the plants and treatment

technology). Specifically, when the population density changed, the model minimized the overall

costs and environmental impacts by changing the assignation of treatment capacities to the

candidate locations (i.e., changing the degree of decentralization), as well as changing the

treatment options. Similarly, the least expensive treatment technology (CAS+GAC) was selected

as the treatment option for the economic solution and the treatment technology with the lowest

GHG emissions (i.e., Bardenpho) for the environmentally friendly solution in all of the candidate

locations across the scenarios. The model also selected a combination of MBR+MF and

72

CAS+GAC in all of the solutions with the highest marginal benefit. In general, a higher population

in water service area results in higher costs and energy requirements for collection, treatment, and

distribution, and it increases the overall costs and environmental impact associated with the water

system significantly. When the population density was low, the model selected MBR+MF in all

of the candidate locations similar as the base case scenario. When the population density slightly

increased from the base case scenario (up to 20%), the model replaced the treatment technology

with a less expensive option (i.e., CAS+GAC vs. MBR+MF) in one of the candidate locations to

offset a portion of the cost incurred and it increased the environmental footprint of the design.

Thereafter, the significant increase in the GHG emissions due to the increase in population density

was driving the model (the second objective function) to select the treatment option with a lower

environmental impact for all of the candidate locations (i.e., MBR+MF) (see Table 3.4 and Table

C.9 in Appendix C).

As Table 3.5 also shows, the population density influenced the assignation of treatment

capacities to the WWTP candidate locations (degree of decentralization of treatment facilities) in

the study area. When the population density decreased in the sensitivity analysis, the model’s

preference was to implement treatment facilities closer to the residential areas (more decentralized

systems) as demonstrated by the lower normalized sample variance. The trend towards

decentralized systems can primarily be attributed to the larger influence of water transfer costs (vs.

the benefits from economies of scale for treatment). Consequently, when the population density

increased, the model opted to take advantage of economies of scale (more centralized

configurations) to offset some portion of the increase in overall costs resulting from a higher

population in the water service area. Previous studies have also shown that when the population

73

density is low, decentralization of treatment facilities was preferred over centralized treatment

(Chung et al., 2008a; Lehtoranta et al., 2014).

Table 3.4 Impacts of changes in population density on the optimal solution with the highest marginal benefit (costs, environmental footprint, degree of decentralization, and degree of treatment). Abbreviations: MBR: membrane bioreactor; MF: micro-filtration; CAS: conventional activated sludge; GAC: granular activated carbon; ZC: costs associated with the entire wastewater system; ZE: greenhouse gas emissions of the wastewater system

3.3.4 Consideration of Social Indicators

Although the value of resource recovery or the willingness to pay for the reclaimed water

by the final consumers was considered as the social indicator in this study, other social factors can

also be considered and incorporated into the input data to the optimization model or in the selection

among the optimal solutions. For instance, the proximity of the reclamation facilities to the

residential areas can be considered in the selection and allocation of candidate locations for

implementation of new wastewater treatment plants.

74

In this regard, several models have been previously developed that are capable of assisting

the stakeholders and decision-makers with the incorporation of a variety of social indicators into

the developed optimization tool. One social consideration can be the decrease in the properties’

value around the implemented wastewater treatment facilities, especially for the lower income

communities, who are more vulnerable to the change in their properties’ value. For instance, in

this case study, Florida Brownfields Redevelopment Atlas (Florida Brownfields Redevelopment

Atlas, 2019) can be used to estimate the number of residents surrounding the candidate locations

for implementation of treatment facilities, the average households’ income in each location, the

average property value in the area, and the median age of the residents (see Table 3.5). Decision-

maker can use this information to further reduce the negative social impacts of the design for the

reclamation system, which are challenging to be quantified and incorporated in the model’s

formulation (Byrne et al., 2017). This practice can be made by incorporating the mentioned factors

into the input data to the model (i.e., selection of candidate sites), defining additional constraints

for the optimization formulation, systematic approach for the selection among the final optimal

solutions, or a combination of the mentioned methods. Another social consideration can be

preference to the areas that have less vulnerability to the environmental impacts for the

implementation of new reclamation facilities and/or allocation of customers to the candidate sites.

These considerations can be incorporated into the model by allocation of candidate locations (input

data), selection among the optimal solutions, or both. For instance, as it can be seen in Table 3.5,

location 2 in this study has relatively lower environmental health vulnerability in terms of the total

wastewater discharge, hazardous waste proximity, and superfund proximity, and the median

households’ income is the highest among the selected locations. However, location 5 is already

undergoing a dramatic environmental situation in terms of wastewater discharge and superfund

75

proximity, and the median households’ income in this location is the lowest. For other cases in the

United States, United States Census Bureau website and EPA's Environmental Justice Screening

and Mapping (EJSCREEN) Tool (US EPA, 2018) can be used to collect and quantify the required

information.

76

Table 3.5 Demographic, economic, housing, and environmental health data related to the candidate sites for additional social considerations (Florida Brownfields Redevelopment Atlas, 2019)

Demographic and Housing Information Economic Information Superfund Proximity

(From EPA's EJSCREEN Tool)

Hazardous Waste Proximity (From EPA's

EJSCREEN Tool)

Wastewater Discharge (From EPA's

EJSCREEN Tool)

Candidate Site Population (2016)

Households (2016)

Median Age

Median Household

Income (2016)

Mean Property Values

(Based on FL DOR Cadastral,

2018)

US Percentile

State Percentile

US Percentile

State Percentile

US Percentile

State Percentile

Falkenburg 2,653 44 33 $41,000 $1,728,585 99.02% 97.00% 85.07% 91.00% 92.67% 96.00%

Valrico 3,703 1,179 37 $82,702 $228,327 12.68% 10.00% 47.11% 33.00% 9.53% 3.00%

South County 4,293 1,236 31 $51,900 $185,400 93.79% 86.00% 70.08% 69.00% 97.94% 99.00%

Southeast County Landfill 1,635 536 37 $71,806 $298,589 53.42% 44.00% 57.11% 46.00% 18.47% 6.00%

Mosaic (Hillsborough) 1,266 465 34 $26,487 $544,162 91.31% 81.00% 83.51% 89.00% 94.61% 98.00%

South County Solid Waste Facility 1,348 425 28 $32,218 $431,741 85.31% 73.00% 79.08% 84.00% 94.26% 97.00%

14519 Balm Riverview Rd, Riverview

10,641 3,268 31 $81,824 $164,885 69.85% 56.00% 63.08% 55.00% 76.10% 88.00%

15110 Balm Wimauma Rd, Wimauma

1,252 405 41 $59,513 $257,890 44.19% 37.00% 54.85% 43.00% 76.10% 88.00%

16410 Balm Wimauma Rd, Wimauma

1,252 405 41 $59,513 $257,890 44.19% 37.00% 54.85% 43.00% 76.10% 88.00%

77

3.3.5 Limitations

This case study was limited in the number of candidate locations and treatment options

considered for the implementation of wastewater treatment facilities in the study area. The limited

number of locations and treatment technologies made it difficult to find the threshold for the

decentralization of treatment facilities and the selection of treatment alternatives when the

topography or population density of the area changed. However, the concept of normalized sample

variance helped to investigate the changes in the optimal solutions’ configuration (in terms of

assignation of treatment capacities to the candidate locations) with changes in the local conditions.

Further investigations can be conducted to evaluate the optimal configurations when there are more

degrees of freedom in the model, e.g., more candidate locations and treatment options to choose

from.

Moreover, the implemented sewer system for the collection of raw wastewater and the

distribution network of the reclaimed water for non-potable reuse purposes was specific to

Hillsborough County. Although the model was capable of specifying the reuse applications and

locations, the existing purple pipelines for urban reuse limited the reuse application and highly

influenced the design of the integrated wastewater system. In this study, the value of resource

recovery, which depended on the effluent water quality, was considered as the social benefit

associated with the design; however, the recovery of additional resources such as energy and

nutrients, and the consideration of other social indicators (e.g., the proximity of the treatment

facilities to the residential areas) were not evaluated. Finally, the GHG emissions (global warming

potential) was considered as the environmental impact of integrated water reclamation systems in

the developed model; however, water reclamation and reuse offsets a portion of the GHG

emissions associated with the entire water system by reducing the needs for freshwater withdrawal,

78

treatment, and distribution. These considerations were outside the scope of the life cycle stage in

this study and further investigations can be conducted to evaluate the impacts of these indicators

to the outcomes of the model.

3.4 Conclusion

In this part of the study, a multi-objective optimization model for wastewater systems

management with a focus on water reclamation was proposed, in which data pertaining to

Hillsborough County, Florida was used to design for the expansion of the county’s wastewater

system. The proposed model minimized the costs and GHG emissions associated with the entire

wastewater system and maximized the value of reclaimed water. The mathematical model

optimized the design of the integrated wastewater system by selecting the best location for

implementing the reclamation facilities, allocating treatment capacities, selecting treatment

technologies, allocating the end users, and finding the optimal degree of decentralization of the

treatment facilities. The results revealed that the model favored the decentralization of treatment

facilities, which was primarily due to the dominant nature of costs associated with long-distance

water transfer. A sensitivity analysis was also performed to provide insight into the factors

influencing the optimal configuration of water reclamation systems. The results of this study

revealed that finding sustainable solutions to the management of water and wastewater systems

(e.g., decisions pertaining to the effluent water quality, location of facilities, and the

decentralization of treatment facilities), highly depends on case-specific parameters such as the

population density and topography of the water service area. This study highlights the need for

comprehensive sustainability assessments that simultaneously account for economic,

environmental, and social dimensions and the usefulness of multi-objective optimization tools in

79

this regard that allow the decision-makers and stakeholders to weigh the tradeoff between the

objectives and the design options.

80

CHAPTER 4: IMPACTS OF EXTERNAL VARIABLES ON THE DESIGN OF

RESOURCE RECOVERY WASTEWATER SYSTEMS

4.1 Introduction

Water shortage and water resources contamination ensued from population growth,

industrialization, urbanization, and the likely impacts of climate change, necessitate adoption of

new approaches to the management of urban water resources. In this regard, developed countries

are moving to increase the resiliency and reliability of their water systems by acquiring

conservation policies as well as resource alternatives for water supply such as desalination,

stormwater harvesting, and water reclamation (Boulware, 2013). Conventional approaches to the

management of wastewater imply collection and conveyance of generated wastewater via sewer

networks, treatment in centralized wastewater treatment plants (WWTP), and discharge of treated

wastewater to the environment. In the United States, approximately 32 billion gallons (120 million

m3) of treated wastewater is discharged to the environment every day (National Research Council,

2012). The potential for recovering valuable resources from wastewater (e.g., energy, and

nutrients) and reusing some portion of the treated wastewater, as an alternative to withdrawal from

natural water resources, makes the WWTPs vital components for addressing the emerging

challenges in the sustainable management of water systems. While potable reuses (i.e., direct and

indirect potable reuses) require advanced treatment options to ensure the water is safe for drinking

purposes and it increases the costs of treatment significantly, non-potable reuse (NPR), which is

using the treated wastewater for non-potable purposes such as irrigation, landscaping, and toilet

flushing, is receiving more attention from municipalities and stakeholders (Diaz-Elsayed et al.,

81

2019). Although NPR reduces the pressure on local clean water resources, has lower treatment

requirements, and fewer regulatory barriers than potable reuse, it requires a separate reclaimed

water distribution network and has a lower value of resource recovery (final reclaimed water

value), compared to the potable reuse options. Given the large associated costs and energy

requirements of water systems due to the expensive and energy-intensive facilities required for

wastewater collection, wastewater treatment, and reclaimed water conveyance to the final users,

water systems have been shown to be responsible for the majority of electricity consumption and

greenhouse gas (GHG) emissions in urban areas. For instance, a study showed that the major user

of California's total energy is the water sector, accounting for almost 19% of the total state

electricity use in 2001 (Navigant Consulting Inc., 2006). The US EPA reported that the energy

expenditure for providing water services in the United States is approximately $4 billion, annually

(US EPA, 2012). Many believe that the energy requirement in the water sector in developed

countries is increasing by 20%-50% in the next 20 years and it is mainly due to population growth,

stringent water quality requirements, and increase in contamination levels (Bounds and Denn,

2017; International Energy Agency (IEA), 2014; Stokes and Horvath, 2010).

Due to the mentioned challenges, energy and GHG emissions (direct and indirect) have

become critical parameters in the design of sustainable water and wastewater systems (Dreizin,

2006; Rygaard et al., 2011). Hence, researchers in the water area are showing growing interest in

developing decision-making frameworks and optimization models to evaluate the alternatives for

reducing the costs, energy requirements, and environmental footprints associated with the design

of water systems that facilitate resource recovery. Among those alternatives, reducing the energy

intensity of water transfer networks by decentralization of treatment facilities, design of more

efficient treatment options with lower energy requirements, selection of the optimum degree of

82

treatment, and the design for sustainable water reuse alternatives are receiving more attention

during the last decade. Decentralization of wastewater treatment plants for NPR purposes has been

suggested to be a preferable option compared to the traditional large centralized systems in some

cases (Daigger, 2009). It is mainly due to lowering the conveyance needs (Gikas and

Tchobanoglous, 2009) and reducing the costs and environmental impacts of water and wastewater

systems, accordingly (Hering et al., 2013). Several studies have evaluated the impact of

decentralizing treatment facilities on the costs and environmental footprint associated with the

water reuse system. However, in the majority of cases decentralization of treatment facilities leads

to losing the economies of scale associated with the treatment phase. Smaller scale treatment

options not only have higher associated costs of implementation and operation per unit volume of

wastewater, but also have higher energy requirements for treatment (resulting in higher

environmental impacts associated with treatment phase) and lower effluent water quality. In some

cases, replacing the treatment technology with a less expensive treatment option can be an

alternative to make up for losing the economies of scale in case of decentralization of treatment

facilities, however, less expensive treatment technologies typically have lower effluent water

quality and lower value of resource recovery accordingly. Several previous studies have shown

that there are tradeoffs to overcome for the design of sustainable water and wastewater systems in

this regard (i.e., decentralization of treatment facilities, degree of treatment, selection of treatment

technology for implementation, and the selected sustainability criteria). However, those tradeoffs

are most likely dependent on the sustainability criteria considered during the evaluation phase, as

well as the case-specific factors and local conditions. Accordingly, the final decisions rely on local

conditions such as elevation variation and population density in the water service area and the

decision-maker’s preferences on the sustainability indicators associated with the final design.

83

Gardels et al. (2011) determined the economic feasibility and environmental sustainability

of five different types of decentralized treatment and reuse design, using a cost analysis and life

cycle assessment approach. The reuse alternatives were designed for four regions in the United

States and they included onsite treatments and non-potable reuse purposes (e.g., toilet flushing).

The results showed that there were tradeoffs between the total costs and overall environmental

impacts associated with each design, while graywater reuse with no treatment for irrigation and a

community level water reclamation and distribution for non-potable reuse purposes obtained the

best results in terms of economic feasibility and environmental sustainability. Glick and Guggemos

(2013) used both environmental life cycle assessment and life cycle cost analysis to evaluate the

long-term effects of decentralization of wastewater treatment plants in northern Colorado. The

results showed that decentralization of the treatment facilities reduced the global warming

potential, energy consumption, hazardous waste, and associated costs significantly by reducing the

sewer line’s investment costs and reducing the needs for long distance water transfer. Later on, a

study conducted by Lam et al. (2015) revealed that onsite treatment and reuse enhanced the

performance and environmental footprints of a water system in Tianjin, China and it was mainly

due to the elimination of the need for mineral fertilizer in the studied case. The results showed that

source-separating and onsite treatment had the best performance in terms of the selected

environmental impact categories.

Lehtoranta et al. (2014) studied the tradeoffs between the reduction of local emissions and

the increase in outside (overall) life cycle impacts of six onsite treatment plants for Finnish

countryside in Finland, where the houses were not connected to the main sewer lines due to

excessive distances. The results showed that package plants and dry toilets in combination with

graywater separation and treatment systems were the best alternatives for the studied area.

84

However, they concluded that the optimal onsite system selection and design strongly depend on

local conditions, and appropriate decision-making framework is needed for making the final

decisions. Another study conducted by Lemos et al. (2013) evaluated the life cycle impacts

associated with the entire water system in the municipality of Aveiro in Portugal. The results

revealed that water transfer and treatment were responsible for the majority of the impact

categories due to intensive energy requirements. Wastewater treatment and disposal also had the

highest impact on eutrophication and ecotoxicity. Matos et al. (2014) compared two types of

wastewater treatment and reuse systems in Portugal, including a centralized treatment and reuse

(i.e., large-scale irrigation) system and a decentralized (onsite) graywater treatment for onsite reuse

applications. The comparison was based on energy consumption, water quality, and CO2

emissions. The results showed that centralized system had higher environmental impacts due to

the need for a higher degree of treatment and onsite systems consumed 11.8% - 37.5% less energy

for serving the same number of inhabitants compared to a centralized system. A study conducted

by Shehabi et al. (2012) compared one centralized and one decentralized wastewater treatment

system in California in terms of air pollution resulted from the system’s operation, greenhouse gas

emissions, and total energy consumption. The results showed that the centralized system had lower

environmental impacts due to the scale economies for treatment of the same volume of wastewater.

However, in case of wastewater reuse in the decentralized system, this system had a lower energy

burden and emissions. The centralized system also reduced the greenhouse gas emissions by

flaring generated methane during the treatment process. Chung et al. (2008) proposed a system

dynamic model to evaluate the effect of decentralization of wastewater treatment facilities in terms

of costs, water quality, and water balances. The model was applied to a realistic hypothetical water

system for a 20-years planning period. The results revealed that when the population density was

85

low, decentralization of treatment plants was preferred. However, the consideration of energy and

the associated environmental impacts such as GHG emissions was outside the scope of their

assessment study.

Some other studies have assessed the corresponding costs and environmental impacts of

alternative water resources (rainwater harvesting, stormwater capture, and water reuse) (Bichai et

al., 2015; Cornejo et al., 2016; Newman et al., 2014). The results revealed that the outcomes are

extremely dependent on the case-specific factors and local condition. Although there have been

several case studies evaluating the impacts of design criteria on the selected sustainability

indicators, many believe that spatial mathematical models and systematic decision-making

frameworks capable of accounting for site-specific parameters must be used to estimate the costs,

energy requirements, and GHG emissions, and to reveal the overall tradeoff associated with the

design of such resource recovery systems (Kavvada et al., 2016). In this regard, Eggimann et al.

(2015) developed a two-step techno-economic spatial model to find the cost-optimal degree of

centralization of treatment plants based on the shortest-path-finding. However, the model did not

consider wastewater treatment and reuse. Rezaei et al. (2019) developed a multi-objective

decision-making framework for selection among different water reuse alternatives in the city of

Lakeland, Florida. The framework accounted for costs, environmental impacts, and effluent water

quality (value of resource recovery). However, the framework was not capable of finding the

optimal location and treatment technology for the treatment facilities. Guo et al. (2014) tried to

find the relationships defining the economies of scale for the costs associated with treatment

processes. The results showed that the costs of wastewater treatment are highly dependent on both

selected technology and scale of the treatment facility. Guo and Englehardt (2015) conducted a

hypothetical modeling study to assess the costs of direct potable reuse (DPR) with respect to

86

system size. They concluded that DPR for an urban area would be cost competitive vs. a traditional

wastewater system, only in medium-scale facilities (>10,000 capita per reuse facility). Some

studies have developed single- and multi-objective optimization models to select among treatment

technologies and design for a sustainable treatment train in terms of costs and environmental

impacts (e.g., Abdulbaki et al., 2017; Hreiz et al., 2015; Kim et al., 2015; Molinos-Senante et al.,

2015; Ruiz-Rosa et al., 2016; Zhang et al., 2014). However, the developed models did not consider

the impacts of degree of decentralization on the design for treatment train, optimal locations for

implementation, and the impacts of effluent water quality.

The few studies, which have focused on decentralization of treatment facilities, have

showed varied results due to the high dependency of the studies on the location and specific

technologies assessed (Baresel et al., 2015; Hendrickson et al., 2015; Rezaei et al., 2019; Shehabi

et al., 2012). Some studies have shown that decentralization of treatment facilities improves the

environmental performance of the design (e.g., Gardels, 2011; Lam et al., 2015; Matos et sl., 2014),

while other studies have shown that conventional centralized treatment options are

environmentally preferred (e.g., Thibodeau et al., 2014). The previous studies in the field have

provided the context and have made the important groundwork, however, it is evidence that

optimization of water reclamation and reuse systems has not been researched thoroughly. All of

the findings from the prior studies are highly influenced by site-specific and local conditions as

well as system scale (Kavvada et al., 2016) and none of the studies provides the optimal sustainable

solutions in terms of degree of decentralization, treatment technology, facilities location, and final

users (customers) allocation, simultaneously, with respect to the variation of local-specific

parameters, which have shown to have the highest influence on the outcomes. Moreover, the

limitation of focusing on only one or two sustainability indicator in the selected models for the

87

evaluation phase (i.e., solely economic in majority of cases, and environmental in few studies)

have masked the optimal solutions and the design tradeoffs according to the triple-bottom-line

sustainability matrix, which simultaneously accounts for economic, environmental, and social

dimensions, and it could criticize the outcomes. Table 4.1 shows the most relevant studies in this

regard and the local condition limitations related to findings from each study.

88

Table 4.1 Limitations of the previous studies on evaluation of water reclamation system

Study Modeling approach Sustainability indicators Decision variables

Model restrictions Single

objective Multi-

objective Assessment modeling Economic Environmental Social Decentralization

of treatment Treatment technology

Reuse allocation

Gardels et al. (2011) X X X X X Low population density and Low elevation variation

Glick and Guggemos (2013) X X X X High elevation variation Lam et al. (2015) X X X X High population density Lehtoranta et al. (2014) X X X

Medium population density and Low elevation variation

Lemos et al. (2013) X X X X Lack of consideration of elevation and population density

Matos et al. (2014) X X X Lack of consideration of elevation and population density

Shehabi et al. (2012) X X X Medium elevation variation and population pattern

Chung et al. (2008) X X X X No elevation variation consideration

Bichai et al., 2015 X X X X X No elevation variation and specific population pattern

Cornejo et al., 2016 X X X X X No elevation variation and specific population pattern

Newman et al., 2014 X X X X X No elevation variation and specific population pattern

Baresel et al. (2015) X X X X X No population consideration Hendrickson et al. (2015) X X X X Study conducted in one building

Kavvada et al. (2016) X X X X X Low population density and Low elevation variation

Abdulbaki et al. (2017) X X X X Specific population pattern Al-Zahrani et al. (2016) X X X X X X Specific population pattern

Hreiz et al. (2015) X X X X Specific population pattern and no elevation variation

Kim et al. (2015) X X X X Specific population pattern Molinos-Senante et al. (2015) X X X X

No elevation variation and specific population pattern

Ruiz-Rosa et al. (2016) X X X X Specific population pattern Zhang et al. (2014) X X X X Specific population pattern

Rezaei et al. (2019) X X X X X X X Low elevation variation and specific population pattern

Rezaei et al. (2019)

X

X X X X X X Low elevation variation and specific population pattern

89

The goal of this part of the study is therefore to evaluate how the external variables in the

water service area (i.e., topography of the water service area and population density) impact the

optimal design configuration of hybrid water reclamation and reuse systems in terms of degree of

decentralization of wastewater treatment plants, selection of treatment technology, and effluent

water quality (value of resource recovery). The previously developed multi-objective optimization

model, which is capable of simultaneously accounting for all major sustainability indexes, is

modified and used for the evaluation process. Hypothetical scenarios for water service area are

designed to generically evaluate the impacts of local conditions (i.e., topography and population

density) on the optimal design of water reclamation systems, while eliminating the impacts of other

case-specific factors that can mask and criticize the outcomes (e.g., a particular population

distribution pattern in the water service area) and make the findings and tradeoffs case-dependent.


In this part of the study, some hypothetical scenarios for the design and implementation of

wastewater system, with a focus on water reclamation and reuse, were generated to evaluate the

impacts of external variable and local conditions (i.e., topography of the water service area and

population density) on the optimal configuration associated with the design of such systems. The

multi-objective optimization model developed in the previous part of the research was modified

and used to find the optimal solutions associated with each scenario. The hypothetical scenarios

consisted of the wastewater collection system, water reclamation facilities (wastewater treatment

plants), reclaimed water distribution network, and population clusters. The raw wastewater

generated from population clusters is collected by the sewer network and transferred to the

reclamation facilities. After treatment of the produced wastewater in the area, some portion of the

reclaimed water (determined by the model) is distributed back to the population clusters via the

90

reclaimed water distribution network for non-potable reuse purposes such as backyard and

landscape irrigation.

The developed multi-objective optimization model was used to determine characteristics

of the wastewater treatment facility (i.e., location, implemented treatment technology, design

capacity, and average flowrates), service area (i.e., number of WWTPs and the connection of

customers to the facilities), sewer network (i.e., flow from each population cluster to each WWTP),

and reclaimed water distribution network (i.e., flow of reclaimed water from each WWTP to each

population cluster). The model minimized the costs and the GHG emissions associated with the

design and maximized the value of resource recovery (correlated to the effluent water quality) to

capture the selected social impact of the design.

4.2.1 Deterministic Optimization Model

As it was mentioned before, the multi-objective optimization model developed in Chapter

4 was modified and used for this part of the study. The proposed mathematical formulation

consisted of two objective functions. The first objective function minimized the costs associated

with the implementation, operation, and maintenance of the wastewater system and maximized the

benefits associated with water reclamation. The second objective function minimized the GHG

emissions of the wastewater system, as the selected environmental footprint associated with the

design. The design for the treatment trains and system scale were modified based on the selected

water quality and population spectrum in the hypothetical scenarios, the Pythagorean Theorem

was incorporated into the model for calculation of distances between the nodes in the hypothetical

water service areas, the case-specific constraints were eliminated from the model, and the OOES

and Criterion Space Search Algorithms were used for solving the multi-objective optimization

problems (Sierra-Altamiranda and Charkhgard, 2018a, 2018b) in this study. The detailed

91

information regarding the modifications of the model’s inputs are discussed in section 4.2.2 and

section 4.2.3.

4.2.2 Hypothetical Scenarios

As it was mentioned before, the goal of this part of the study was to evaluate the impacts

of external variables on the tradeoffs associated with the optimal design of resource recovery

systems. The external variables, which previously identified to have a higher influence on the

optimal configuration of water and wastewater systems, consisted of topography of the water

service area and population density. To achieve this and to examine a wider spectrum of the inputs

to the model while eliminating the impacts of other local-specific conditions (e.g., location of

candidate sites for the implementation of treatment facilities), some hypothetical scenarios for

different types of water service area were generated. The hypothetical scenarios represent a water

service area consisting of population clusters, candidate sites for implementation of treatment

facilities, sewer network, purple pipelines for distribution of reclaimed water for non-potable

purposes (e.g., backyard irrigation and landscaping), and pumping stations. The hypothetical

scenarios represented three types of population density (i.e., low, medium, and high) and three

types of topography (i.e., flat, medium elevation variation, and hilly). Figure 4.1 shows the nine

hypothetical scenarios generated for this study. The hypothetical scenarios consisted of low

population density with low elevation variation (LPLE), medium population density with low

elevation variation (MPLE), low population density with high elevation variation (LPHE), low

population density with medium elevation variation (LPME), medium population density with

medium elevation variation (MPME), high population density with medium elevation variation

(HPME), low population density with high elevation variation (LPHE), medium population density

with high elevation variation (MPHE), and high population density with high elevation variation

92

(HPHE). The water service area in each scenario was divided into 36 grids, in which the produced

wastewater from the grids can be transferred to 25 candidate sites for treatment. The population

was also assumed to be distributed equally over the water service area to further eliminate the

impacts of case-specific local conditions.

93

Figure 4.1 Nine different hypothetical scenarios generated for this study.

94

4.2.3 Scope and Input Data

The focus of this hypothetical study was from the generation of raw wastewater in

population clusters to treatment and distribution of reclaimed water to the end users. Three types

of population density were selected for the design of hypothetical cases based on the population

density pattern in Orange County, California and the New York metropolitan area (US Census

Bureau, 2018). Moreover, three types of topography (elevation variation) were also considered

based on the average elevation variation in Orange County, California (Orange County Public

Works, 2018). The population densities consisted of low (100 capita/mi2), medium (3,000

capita/mi2), and high (25,000 capita/mi2) and the topographies consisted of flat (no elevation

variation), medium elevation variation (an average of 15 m/mile variation), and hilly topography

(an average of 87 m/mile elevation variation). The candidate locations for the new facilities were

distributed equally in 25 different sites within the water service area, serving a total of 36

population clusters. The model calculates the distance (shortest path) between the nodes (candidate

locations for implementation of WWTPs and customer clusters) using the Pythagorean theorem in

the hypothetical water service areas. The shortest pipeline path in the main sewer/purple line

between population clusters and reclamation facilities was considered to calculate the requirements

for water transfer between the nodes in the hypothetical water service areas (e.g., pumping

requirements, and maintenance of the networks).

The options for treatment technologies were selected based on the study conducted by

Diaz-Elsayed et al. (2019) and recommendations from the stakeholders in Hillsborough County,

Florida. The treatment trains were designed based on the new requirements for secondary

wastewater treatment (State of Florida, 2018; US EPA, 2012) and consisted of Bardenpho (a

dissolved oxygen bioreactor) followed by filtration, a membrane bioreactor (MBR) followed by

95

micro-filtration (MF), MBR followed by ultra-filtration (UF), MBR followed by reverse osmosis

(RO), or rotating biological contactor (RBC) followed by tertiary filtration (TF) (see Table D.1 in

Appendix D for more information regarding the treatment trains). Small, medium, and large-scale

systems were considered for each technology based on the classification suggested by Diaz-

Elsayed et al. (2019). All of the selected technologies are capable of meeting the effluent water

quality requirements in the State of Florida and State of California for non-potable urban reuse,

and treatment technologies with higher effluent water quality were also considered to evaluate the

impacts of added value associated with reclaimed water on the model outputs.

Information needed to calculate the capital costs of implementation, operation and

maintenance costs, and energy requirements associated with each treatment train at each scale,

were obtained from County reports and literature data (e.g., Abdelrasoul et al., 2016; Cashman et

al., 2018; Chang et al., 2008; Cornejo et al., 2016; Gabarrell et al., 2012; Guo et al., 2014;

Holloway et al., 2016; Marufuzzaman et al., 2015; Ortiz et al., 2007; Tang et al., 2018; Tarnacki

et al., 2012; Zhou et al., 2011) (see Table D.2 in Appendix D). The pumping electricity and pipe

requirements were estimated based on previous projects in the counties and engineering handbooks

(NCEES, 2013) (see Table D.3 in Appendix D). The value of resource recovery (the willingness

to pay) was based on the average effluent water quality for each treatment technology and Polk

County and Hillsborough County’s reclaimed water prices (Hillsborough County, 2017a; Polk

County, 2017; Rezaei et al., 2019). The effluent water quality for each treatment train was

compared to the water quality requirements for non-potable and potable reuses. The price for

drinking water and the price of reclaimed water for non-potable purposes are available in

Hillsborough and Polk County. The required additional treatments to achieve the quality

requirements for non-potable urban and potable reuses were considered to estimate and the price

96

of reclaimed water with different qualities for the model. Specifically, IT3PR Integrated Treatment

Train Toolbox was used for the estimation of required additional treatments for potable reuse

(Trussell et al., 2015). The detailed information regarding the value of resource recovery for each

treatment train at each scale can be found in Table D.3 in Appendix D.

4.2.4 Degree of Decentralization and Marginal Benefit Calculations

Similar to the previous part of the study (section 3.2.4), in order to quantify the degree of

decentralization of treatment facilities, the concept of normalized sample variance of the WWTPs’

design capacity was used, and since the model in this study also used two objective functions, the

concept of marginal benefits was used to identify the highest achievable benefit with the lowest

increase in associated costs. Accordingly, the optimal solution with the highest environmental

marginal benefit was selected for further investigations in each hypothetical scenario.


Figure 4.2 shows the results of the multi-objective optimization model for the nine

generated hypothetical scenarios for the water service area. As it can be seen in this figure, solving

the optimization model resulted in a different number of optimal solutions for each scenario,

however, the concept of marginal benefit was used to select one solution among the optimal

solutions for each scenario. As Figure 4.2 shows, higher elevation variation in the water service

area resulted in lower number of optimal solutions generated by the model and it is mainly due to

the lower flexibility of the model in minimizing the overall costs and environmental impacts and

maximizing the VRR by changing the configuration of the design (e.g., changing the selected

treatment technologies in the candidate sites) when the water transfer requirements were higher.

The details regarding the optimal solutions with the highest marginal benefit can be found in

Appendix D (Tables D.4-D.12). These tables represent the recommended reclamation facilities in

97

each candidate site; the selected treatment technology, scale, and operation capacity at each

facility; the allocation of customers to each reclamation facility; and decision variables regarding

the amount of wastewater and reclaimed water that needs to be collected, transferred, and sold to

customers.

98

Figure 4.2 The optimal solutions for the multi-objective optimization model for the hypothetical scenarios.

99

As it can be seen in Figure 4.2, the environmental footprint of the design can decrease by

selecting more costly design options and decision-makers should make the final selection among

the optimal points based on the tradeoffs for the costs and environmental footprint. However, since

the candidate sites for implementation of treatment facilities and the population clusters were

distributed evenly within the hypothetical water service areas, the optimal solutions followed

almost a linear pattern in all of the scenarios (R2 ranges between 0.77 and 0.98). Accordingly,

when the population density was lower, the optimal solutions followed a more linearized pattern

and it was mainly due to the lower degrees of changes in the overall costs and environmental

impacts caused by changing the design configurations when the population density was lower in

the water service area. In this regard, since the comparison of solutions becomes more challenging

for selection, the concept of marginal benefit was a useful tool for the selection of one solution

among all the optimal solutions for the further analyses. Figure 4.3 shows the overall information

(i.e., total associated costs and environmental impacts, effluent water quality, and degree of

decentralization of treatment capacities in the candidate sites) regarding the optimal solution with

the highest marginal benefit associated with each scenario.

As Tables C.4-C.12 represent, the model’s preference was to send the maximum amount

of reclaimed water produced (90% of the total produced wastewater) back to the population cluster

for non-potable urban reuse purposes in all of the scenarios. Moreover, when the elevation

variation increased, the total cost and environmental impacts per capita associated with the design

of water reclamation system in the study area increased significantly, and it is mainly due to the

increase in the energy requirements for water transfer (wastewater collection and reclaimed water

distribution) in the areas with higher elevation variation. However, increasing the population

100

density did not make a significant change to the overall costs and environmental impacts (per

capita) of the design. The results are discussed in a greater depth in section 4.3.1 and section 4.3.2.

Figure 4.3 The information regarding the optimal solutions with highest marginal benefit for each scenario. Abbreviation: NS2: normalized sample variance of treatment capacities and VRR: value of resource recovery.

Figure 4.4 also shows the information regarding the treatment trains designed for the

implementation in the candidate sites in different scales. As it can be seen, treatment technologies

in larger scales offer lower capital and operation costs per unit volume of wastewater, and it is

typically due to the economies of scale associated with larger scale treatment systems. Moreover,

treatment technologies in larger scales have a slightly better effluent water quality compared to

smaller scale systems and it is mainly due to the better operational conditions associated with

biological systems in larger scales. Moreover, larger treatment facilities usually have lower energy

requirements for treatment of a unit volume of wastewater and it decreases the environmental

impacts associated with these systems, accordingly. As Figure 4.4 also shows, incorporation of

expensive treatment options such as UF and RO to the treatment train improves the effluent water

quality significantly and it increases the VRR, however, these treatment options have higher

energy requirements for operation and it increases the environmental impacts of such advanced

treatment options.

101

Figure 4.4 Information regarding the treatment trains in small, medium, and large scale, designed for this study. Abbreviations: ASNPV: annualized specific net present value; VRR: value of resource recovery; GWP: global warming potential; MBR: membrane bioreactor; MF: micro-filtration; UF: ultra-filtration; RO: reverse osmosis; RBC: rotating biological contactor; and TF: tertiary filtration.

Figure 4.5 and Figure 4.6 show the breakdowns of the costs (the first objective function)

and environmental impact (the second objective function) associated with the optimal solutions

with the highest marginal benefits in each scenario. As it can be seen in Figure 4.5, when the

population density is low, the capital costs of the treatment phase is the dominant factor in the first

objective function (scenarios LPLE, MPLE, and HPLE). However, when the population density is

medium or high, the costs associated with water transfer (i.e., wastewater collection and reclaimed

water distribution) is the dominant factor in the first objective function (scenarios MPME, HPME,

102

MPHE, and HPHE), except for scenarios with low elevation variation (i.e., scenarios LPME and

LPHE). As Figure 4.6 also shows, when the elevation variation in the water service area is low, the

GHG emissions associated with wastewater treatment is the significant contributor in the second

objective function (scenarios LPLE, MPLE, and HPLE). However, the GHG emissions associated

with water transfer phase became the major contributor to the second objective function when

there was medium or high elevation variation in the water service area (i.e., scenarios LPME,

MPME, HPME, LPHE, MPHE, and HPHE).

103

Figure 4.5 The breakdowns of the costs (the first objective function) associated with the optimal solutions with the highest marginal benefit. Abbreviations: WWTP: wastewater treatment plant; WW: wastewater; RW: reclaimed water; VRR: value of resource recovery; P: population density; E: elevation variation.

104

Figure 4.6 The breakdowns of the environmental impacts (the second objective function) associated with the optimal solutions with the highest marginal benefit. Abbreviations: WW: wastewater; RW: reclaimed water; GWP: global warming potential; P: population density; E: elevation variation.

105

4.3.1 Impacts of Population Density

As Figure 4.3 shows, population density had a high influence on both optimal degree of

decentralization of treatment facilities and optimal degree of treatment (selection of technology

for treatment). As it can be seen in Figure 4.3a, when the service area was flat, increasing the

population density did not make a significant change to the degree of decentralization of treatment

facilities. However, changing the population density when there was a medium or high elevation

variation in the water service area had a notable influence on the assignation of treatment capacities

to the candidate sites for implementation of WWTPs. In fact, when the elevation variation in the

service area was medium, increasing the population density from low to medium resulted in

implementation of more centralized configurations for treatment (~127% increase in NS2), and it

was mainly due to the domination of economies of scale for treatment when the population density

was low in scenario LPME (see Figure 4.5d). Interestingly enough, further increase in population

density from medium to high, when there was a medium elevation variation in the area, resulted

in more decentralization of treatment capacities in the water service area (~50% decrease in NS2

from scenario MPME). It was mainly because the water transfer requirements became the dominant

factor (vs. economies of scale for treatment) when the population density increased in scenario

MPME in both objective functions (see Figure 4.5e and Figure 4.6e). As it can also be seen in

Figure 4.3, when there was a high elevation variation in the water service area, increasing the

population density resulted in more decentralized pattern for implementation of treatment

facilities. While the majority of previous studies have been conducted in flat regions, with

consideration of only low and medium population density in their evaluations, it had been shown

that higher population density favors more centralized patterns for treatment (e.g., Chung et al.,

2008). Although this study showed the same results in scenarios LPLE and MPLE, it was evidenced

106

that when the elevation variation in the water service area was high, increasing the population

density favors more decentralized patterns for treatment. It was mainly due to the higher economic

and environmental benefits that the model could take by decentralization of treatment facilities, to

lower the impacts resulted from the increase in the population density in the water service area,

while the water transfer requirements was higher due to the higher elevation variation in the area.

Unlike the impacts on degree of decentralization of treatment capacities, the increase in

population density had the same impact on degree of treatment in all of the three types of elevation

variation considered in this study (i.e., low, medium, and high). In fact, when the population

density increased, the model favored implementation of treatment technologies with higher

effluent water quality to increase the value of resource recovery. As Figure 4.5 shows, when

population density increased, the significance of VRR in the first objective function increased

significantly (more population density represents more wastewater produced in the water service

area). Hence, when the population density increased, the model replaced the treatment technology

in some of the candidate locations with a more expensive treatment option that had higher

associated effluent water quality (i.e., MBR+MF vs. Bardenpho).

4.3.2 Impacts of Elevation Variation

As Figure 4.3 shows, elevation variation also had a high influence on the optimal degree

of decentralization of treatment facilities in the hypothetical scenarios. In fact, when the population

density was low, increasing the elevation variation from the flat case to medium resulted in

selection of a more decentralized pattern (~71% decrease in NS2) for the treatment capacities

within the water service area. It was mainly due to the increase in the contribution of water transfer

requirements in the first and second objective functions vs. the economies of scale for treatment

(see Figures 4.5a and 4.5d and Figures 4.6a and 4.6d). At this point (scenario LPME), while the

107

increase in the contribution of the water transfer phase to the second objective function was mainly

due to the increase in elevation variation in the area, the contribution of GHG emissions associated

with the treatment phase was significant due to the higher degree of decentralization in scenario

LPME (see Figure 4.6d). Hence, increasing the elevation variation from medium to high resulted

in implementation of larger treatment facilities (lower GHG emissions) in some of the candidate

sites and replacing the treatment technology with a treatment option with lower environmental

impact (i.e., Bardenpho vs. MBR+MF) in the corresponding sites to even further reduce the overall

GHG emissions of the design. Hence, the degree of decentralization of treatment facilities and the

GHG emissions associated with treatment phase decreased significantly (~184% increase in NS2)

(see Figure 4.3a and Figure 4.6g). It can be concluded that when the population density and

elevation variation are low, decentralization of treatment facilities improves the costs and

environmental impacts of the water system. However, when the elevation variation reaches to a

higher level, the design should take advantages of both economies of scale and decentralization of

treatment facilities in the water service area to achieve the optimal level of the corresponding

impacts.

Interestingly enough, when the population density was high, the model followed the same

pattern as the low population density scenarios to optimize the costs and environmental impact in

the scenarios with different elevation variation. However, when the population density was at the

medium level, the model took a different approach to offset the impacts resulted from the increase

in the water transfer requirements. In fact, the model’s preference was to implement larger

treatment facilities in some of the candidate sites (NS2 increased by 37% and 20% from flat to

medium elevation variation and from medium elevation variation to hilly, respectively) to improve

the economies of scale for treatment phase. Although several previous case studies show that

108

decentralization of treatment capacities are favorable when there is a high elevation variation in

the water service area, the results of this study revealed that the decision on decentralization is

highly influenced by the population density and topography of the area. Higher population

densities increase the impacts of economies of scale associated with the treatment phase and higher

degrees of elevation variation improves the contribution of water transfer requirements in the final

configuration. Eventually, the decision on the optimal degree of decentralization of treatment

capacities and tradeoffs vary based on the external variables integrated with the water service area.

Unlike population density, elevation variation did not have a significant influence on the optimal

degree of treatment and selection of treatment technology, accordingly (see Figure 4.3b).

The bottom line is that the tradeoffs associated with the main design characteristics of the

water systems (i.e., decentralization of treatment and treatment technology) are highly influenced

by the situation in the water service area and it is not possible to adopt a one-size-fits-all approach

for the sustainable design of such systems without bringing the other external variables into

account. The results of this study revealed that although some extent of decentralization of

treatment capacities was preferred in all of the water service areas due to the dominant nature of

the costs and environmental impacts associated with water transfer vs. economies of scale for

treatment, the optimal degrees of decentralizations and treatment options, as well as the associated

tradeoffs are highly dependent on the external variables in the water service area.

4.3.3 Limitations

One of the limitations of this study was the incorporation of social impacts into the model.

Previously, due to the complexity in measurement and quantification of the social impacts,

researchers have considered a variety of indicators to include the social aspect in their assessment.

The indicators consisted of job and study opportunities (Akhoundi and Nazif, 2018), working

109

conditions for employees (Palme et al., 2005), public acceptance (Lamichhane and Babcock, 2013;

Mainali et al., 2013; Rygaard et al., 2014), the ability to reduce diarrheal illness and deaths (Ren

et al., 2013), operation expertise requirement (Tjandraatmadja et al., 2013), and the service

providers’ past experience (Sharma et al., 2009). As it was mentioned before, since the success of

water reclamation and reuse in practice highly depends on public acceptance, the value of resource

recovery (VRR) (Rezaei et al., 2019), or willingness to pay for the reclaimed water, was considered

as the social benefits of the system in this study. Although the value of resource recovery (the

willingness to pay for the reclaimed water) was considered as the social impacts of the design, the

consideration of other social considerations (e.g., proximity to the residential areas, lack of enough

expert operators, and regulatory compliance) was outside the scope of the study. Although those

considerations can be incorporated in the inputs to the model (e.g., selection of appropriate

candidate sites for implementation of reclamation facilities), those social considerations can be

quantified using social data collection techniques (e.g., conducting surveys and questionnaires)

and incorporated to the model by advancing the model to a three objective optimization model.

Moreover, although decentralization of wastewater treatment facilities has been shown to

be economically and environmentally beneficial in the majority of cases, there are some other

barriers to the implementation of such systems by municipalities. For instance, regulatory

compliance and policies such as a minimum required distance between the location of wastewater

treatment facilities and the residential areas can be a hurdle to overcome for the design and

implementation of decentralized wastewater systems. The decision on decentralization of

reclamation facilities is also highly influenced by the emerging challenges to the management of

water and wastewater systems, such as the negative impacts of climate change and aging

infrastructure. For instance, the increasing frequency of natural disasters due to the likely impacts

110

of climate change, and the higher vulnerability of centralized systems to those natural recurrences

and terrorism (Daigger, 2009), can be another motivation for implementation of decentralized

systems by officials and decision-makers. Moreover, aging infrastructures and failure to meet the

required effluent water quality in the aged treatment facilities, as well as the increase in the costs

associated with maintenance of the sewer lines, necessitates adoption of new approaches that are

more reliable and resilient compared with the traditional centralized management. The mentioned

factors and their potential impacts on the decision-making process should be considered and

incorporated in the future by conducting additional evaluation studies such as risk and resiliency

assessments.

4.4 Conclusion

In this study, the previously developed multi-objective optimization model in Chapter 3

was modified and used to evaluate the impacts of external variables on the design and management

of wastewater systems that promote and facilitate water recovery and reuse. Two external

variables, which were previously identified as the factors with higher influences (i.e., topography

of the area and population density), were considered to design nine hypothetical scenarios for the

water service area. To capture the impacts of topography of the area, three types of elevation

variation (i.e., flat region, medium elevation variation, and hilly) was considered in the design of

hypothetical scenarios, in which three types of population density (i.e., low, medium, and high)

were incorporated in each topography. The concept of hypothetical scenarios helped with the

elimination of the impacts of other case-specific parameters and local conditions that can criticize

the interpretation of final findings (e.g., particular population distribution pattern), to provide more

independent insights to the studied factors in the previous sections of this research. The

mathematical model optimized the design of the integrated water reclamation system by selecting

111

the candidate sites for implementing the reclamation facilities, allocating treatment capacities to

the selected sites, selecting treatment technologies, finding the optimal degree of decentralization

of the treatment facilities, and allocating the end users. The results revealed that decentralization

of treatment facilities in the water service area was favored in terms of costs and environmental

impacts of the design for all of the hypothetical scenarios. However, there was a tradeoff between

the reduction in the costs and GHG emissions resulted from decentralization of WWTPs and the

benefits from economies of scale for treatment. The results showed that the decision on the optimal

degree of decentralization and the tradeoffs highly depend on the population density and

topography of the water service area and the model’s preferences biased toward different solutions

for different scenarios. Moreover, it was made evident that selection of advanced treatment options

with higher associated costs and GHG emissions was not favored in any of the scenarios. The

decision on selection of treatment option was also highly influenced by the population density in

the water service area.

112

CHAPTER 5: CONCLUSIONS AND RECOMMENDATIONS

5.1 Summary

The overall goal of this study was to develop a framework for integrated wastewater system

management that facilitates resource recovery over treatment and to contribute to the literature by

focusing on the design, assessment, and management of a closed-loop water supply chain through

a holistic sustainability approach. To achieve this, this research was outlined to develop decision-

making tools, assessment frameworks, and mathematical optimization models for the design,

management, and evaluation of water reclamation systems. The developed models and frameworks

were sketched based on the triple-bottom-line sustainability matrix, which simultaneously

accounted for economic, environmental, and social impacts of the design. The present work not

only provides decision-makers, officials, stakeholders, and researchers with some useful decision

support tools for the design and evaluation of water reclamation systems, but also investigates the

insights to the factors that have a high influence on the sustainability of such resource recovery

systems. This research was initially established to examine the following hypotheses regarding the

design of resource recovery oriented wastewater networks:

• Hypothesis 1: Increasing the reclaimed water quality (degree of treatment)

increases the costs and the environmental impacts (e.g. carbon footprint) of water

systems, due to the needs for implementation of tertiary or advanced treatment train

and higher energy requirements. Moreover, the higher water quality would

probably offset a large portion of the overall environmental impacts by lowering

113

the other footprints such as eutrophication and it also increases the value of resource

recovery (water in this case).

• Hypothesis 2: Increasing the degree of decentralization increases the capital costs

significantly, but it decreases the costs associated with the wastewater collection

and reclaimed water distribution (O&M costs). The decrease in O&M costs may

offset the contribution of capital costs to the annual net present value in long-term

planning. It also reduces the environmental impacts such as carbon footprint by

lowering the energy requirements for water transfer (i.e., wastewater collection and

reclaimed water distribution).

• Hypothesis 3: The distance between water reclamation and reuse plays a significant

role in the environmental and economic impacts of the design. Increasing the

distance between facilities increases the capital costs associated with the

implementation of water transfer facilities and the energy requirements for water

transfer and it highly affects the final economic and environmental (e.g., carbon

footprint) impacts. Moreover, selection of treatment technologies with lower

capital and O&M costs reduces the final economic and environmental impacts

significantly, however, to a relatively smaller extent.

• Hypothesis 4: Higher population density and lower elevation variation (lower head-

loss) necessitate moving towards more centralized water reclamation facilities,

while in areas with lower population density and/or higher elevation fluctuation,

decentralization of wastewater treatment plants and reuse locations would lower

the economic and environmental impacts such as carbon footprint.

114

To evaluate the impacts of reuse alternatives, required degree of treatment, reuse location,

effluent water quality, and decentralization of treatment facilities for specific reuse options, a

decision-making and evaluation framework that relied on a regret-based model was developed

(Chapter 2). The most common types of reuse alternatives (i.e., agricultural reuse, non-potable

distributed urban reuse, unrestricted urban reuse, indirect potable reuse, and direct potable reuse)

were considered in the evaluation study, as well as some degree of decentralization of treatment

facilities for distributed urban reuse (non-potable). The framework was developed based on the

triple-bottom-line sustainability matrix (i.e., economic, environmental, and social) and provided

the insights to the factors, which have higher influence to the design of water reuse applications.

The results of this part of the study revealed that the distance between the water reclamation facility

and the end use location played a significant role in economic and environmental indicators.

Increasing the average distance from 0.9 miles to 6.5 miles, with the same degree of treatment for

urban reuse and agricultural reuse, increased the environmental impact (carbon footprint) from

1,781 kg CO2-eq/MG to 8,684 kg CO2-eq/MG, while it increased the annualized specific net

present value from $413 to $1,667, respectively. The higher reclaimed water quality required

implementation of advanced treatment options in the treatment train, and consequently increased

the economic and environmental impacts. Higher water quality, however, improved the

eutrophication of water reuse as well as the value of resource recovery significantly, and it

decreased the final regret score. The higher value of resource recovery could also offset all the

capital and O&M costs associated with the treatment and distribution for direct potable reuse in

the case study. Considering this fact, this reuse option obtained the best regret score among the

five studied reuse alternatives. In this case, decentralization of treatment facilities, however, did

115

not make improvement in the total costs associated with the design, although it reduced the carbon

footprint by reducing the energy requirements for water transfer.

The results of this part of the study supported the first hypothesis. It was made evident that

increasing the reclaimed water quality in IPR and DPR, increased the carbon footprint of the

design, and it was mainly due to the need for implementation and operation of additional treatment.

However, the higher effluent water quality in these scenarios reduced the eutrophication potential,

and the increase in the value of resource recovery in DPR could offset the total increase in the

costs resulted from implementation and operation of the additional required treatments.

Nevertheless, the second hypothesis was rejected in the case study of this part of the research.

Although decentralization of treatment facilities reduced the O&M costs associated with the sewer

system (wastewater collection) and distribution network (reclaimed water distribution), it

increased the costs associated with the implementation of reclamation facilities (WWTPs)

significantly, and the decrease in O&M costs could not offset the increase in the capital costs of

implementation of decentralized WWTPs. Hence, decentralization of treatment facilities increased

the overall costs, although it reduced the total energy consumption and the overall global warming

potential of the system, accordingly. The third hypothesis was also supported in this part of the

study. The results of Chapter 2 revealed the significant contribution of the water transfer phase,

which is highly influenced by the distance between water reclamation facilities and final end users

(reclaimed water customers), in both economic and environmental indicators.

In Chapter 3, a multi-objective optimization model, which simultaneously incorporated the

economic, environmental, and social indicators, was developed to further help with the design and

evaluation of integrated water reclamation and reuse systems. The model was capable of locating

the wastewater treatment facilities, allocating treatment capacity to the facilities, selecting

116

treatment technology for implementation in the selected locations, allocating the reuse alternative,

and allocating the customers, as the final users of the produced reclaimed water. Actual data

pertaining to the Southcentral water service area in Hillsborough County, Florida was used to

design for the expansion of the water reclamation system in the corresponding water service area.

The impacts of population density and topography of the water service area (elevation variation)

on the model outputs were also investigated using a sensitivity analysis approach on water transfer

requirements and population density of the water service area. Although centralization of treatment

facilities improves the economies of scale, the results revealed that simultaneous consideration of

economic and environmental indicators favored decentralization of treatment facilities in the study

area, and it was mainly due to the significant decrease in the requirements for water transfer,

especially in less populous areas. Moreover, population density was the primary contributor to the

optimal degree of decentralization of treatment facilities.

Application of the developed frameworks and mathematical models in the two studied

areas provided useful information regarding the design of integrated water and wastewater systems

with a focus on water reclamation and reuse, as well as the insights to the parameters influencing

the sustainable design of such resource recovery systems in the studied areas. However, the

limitations resulted from the case-specific parameters and local conditions associated with those

real-case studies made some barriers to the understanding of final optimal solutions for the design

of integrated water reclamation systems. Hence, in Chapter 4, nine hypothetical scenarios for the

water service area was generated to further evaluate the impacts of parameters that were found to

have a higher influence on the design of water systems (i.e., topography of the water service area

and population density). The concept of hypothetical scenarios helped with eliminating the impacts

of other case-specific factors while proving the impacts of changes in elevation variation and

117

population density in the water service area. Three types of topography (i.e., flat, medium elevation

variation, and hilly), as well as three types of population density (i.e., low, medium, and high),

were considered for the design of hypothetical scenarios. The results revealed the tradeoffs

between reducing the impacts by decentralization of treatment facilities (lowering water transfer

requirements) and the benefits from economies of scale for treatment and the decision on the scale

of implementation of the treatment facilities is highly influenced by the topography and population

density of the water service area. It was also made evident that population density has a high impact

on the selection of treatment technology for the treatment facilities. Although Chapter 3 supported

the fourth hypothesis, extensively, the results of Chapter 4 showed that this hypothesis could not

be generalized to all of the cases and the results are highly influenced by the case-specific

parameters and local conditions.

5.2 Research Limitations and Future Opportunities

5.2.1 Input Data

In this study, a tremendous amount of data was used for the design and evaluation of the

integrated water and wastewater systems in different parts of the research. The information and

data consisted of (not limited to) pipeline material, pumping requirements, labor, data related to

construction and operation of WWTPs, information for the design of treatment trains, effluent

water qualities, construction of facilities, implementation of pipe networks, electricity

consumption by the processes, emissions, construction of injection wells, and cost/benefit data.

Since the information and data needed to conduct the research varied with regard to their nature

and strain, different resources (e.g., engineering handbooks, academic publications, county

reports, facility data sheets, and stakeholders) were used to collect the required data and

information for conducting the different parts of the study. Using the different resources for the

118

input data resulted in losing the uniformity and the quality of the input information to the developed

frameworks and mathematical models, consequently. However, for the majority of parts, the

information was collected from the local resources (e.g., studies conducted in the same area as the

case studies in this research, local facilities, and related officials and stakeholders) to keep the

negative impacts of non-uniform data at a minimum. For further investigations and future studies,

it is important to collect the input information from related resources; therefore, the conclusions

will be made authentically.

5.2.2 Sustainability Indicators

As it was mentioned before, this study was outlined based on the triple-bottom-line

sustainability matrix, which simultaneously accounts for economic, environmental, and social

factors. The concept of annualized specific net present value was used to normalize the costs (i.e.,

capital and O&M) associated with the design, however, the quantification of environmental and

social impacts was found to be more challenging. To address the environmental impacts of the

design, global warming and eutrophication potential was considered in this study. Although these

indicators are the most commonly used factors for quantification and evaluation of environmental

impacts of the design in LCA studies, other sustainability indicators (e.g., freshwater augmentation

and fossil fuel depletion) can be incorporated in the developed mathematical models and

frameworks. Although consideration of those sustainability indicators was outside the scope of the

current study, the corresponding factors can be easily quantified, normalized, and incorporated

into the developed frameworks and mathematical models in this research.

Furthermore, the major limitation of this study was the incorporation of social aspects into

the developed frameworks and models. As it was mentioned before, due to the complexity of

quantification of social indicators, researchers show little to no interest in incorporating such

119

aspects into their assessments. In this research, since the success of water reclamation and design

of a closed-loop water supply chain is highly dependent on public acceptance, value of resource

recovery or willingness to pay for the reclaimed water by the customers was considered as the

social impact of the design. However, other social considerations (e.g., proximity of WWTPs to

the residential areas, expert availability for operation of the facilities, work opportunities, social

acceptance for reuse alternatives, and regulatory compliance) in the mathematical models was

outside the scope of this study. Although this study tried to incorporate those social factors by

receiving feedbacks from the officials and stakeholders in the study areas through arrangement of

regular meetings and online correspondences, the direct quantification and consideration of those

parameters in the models’ formulation was lacking. While the developed models and frameworks

are capable of considering those indicators in their structure, further investigations are needed to

quantify and include those social factors in the formulations. Conducting surveys and

questionnaires among stakeholders, officials, facility operators, and residential areas can be an

effective option to collect and quantify those information. The input data can be used as an

independent indicator in the developed framework in Chapter 2 and as the third objective function

in Chapter 3 and Chapter 4. It should be noted that if more than one factor as the social indicators

are incorporated, the data must be unified using one of the standard normalization techniques such

as Z-Score or Min-Max method.

5.2.3 Improvement of the Developed Models and Frameworks

The developed model in Chapter 3 and Chapter 4 can be coupled with a Shortest Path

model such as Bellman Ford's Algorithm and Dijkstra’s Shortest Path Algorithm to incorporate

and investigate the impacts of pipeline requirements (e.g., material, length, diameter, and

implementation) to the model’s outputs in a greater depth. Although the linear shortest path was

120

considered in the current study for calculation of the average requirements for water transfer

between facilities and customers using the data from the previous projects in the studied areas,

coupling the model with a shortest path algorithm can provide outputs that are more accurate in

this regard.

Moreover, water was the only resource that was considered for recovery from wastewater

in the developed models and frameworks in the current study. However, there are other valuable

resources, which can be recovered from wastewater during the treatment process (e.g., energy and

nutrients). While recovery of these resources could further offset a portion of the costs and

environmental impacts associated with water reclamation systems, consideration of the influence

of energy and nutrient recovery, as the other most common recovery techniques in this regard, on

the sustainability assessment of such systems was outside the scope of this study. Further

investigations can incorporate recovery of these resources into the developed models and

frameworks.

As it was also discussed in sections 3.3.4 and 4.3.3, there is a variety of social

considerations that can be incorporated into the developed optimization models. Some of the

factors can be considered in the input data to the model (e.g., selection of candidate locations for

implementation of treatment facilities) and some others can be considered in selection among the

optimal solutions (e.g., selection of solutions that implement treatment facilities in the lower

vulnerable residential areas). Furthermore, some of the emerging policies such as the minimum

required distance between reclamation facilities and residential areas, and emerging requirements

for effluent water quality (treatment technology) can be incorporated by defining some additional

constraints in the formulation part of the model.

121

REFERENCES

Abdelrasoul, A., Doan, H., Lohi, A., Cheng, C.-H., 2016. Prediction of power consumption and performance in ultrafiltration of simulated latex effluent using non-uniform pore sized membranes. Korean J. Chem. Eng. 33, 1014–1027. https://doi.org/10.1007/s11814-015-0206-8

Abdulbaki, D., Al-Hindi, M., Yassine, A., Abou Najm, M., 2017. An optimization model for the allocation of water resources. J. Clean. Prod. 164, 994–1006. https://doi.org/10.1016/J.JCLEPRO.2017.07.024

Akhoundi, A., Nazif, S., 2018. Sustainability assessment of wastewater reuse alternatives using the evidential reasoning approach. J. Clean. Prod. https://doi.org/10.1016/J.JCLEPRO.2018.05.220

Al-Zahrani, M., Musa, A., Chowdhury, S., 2016. Multi-objective optimization model for water resource management: a case study for Riyadh, Saudi Arabia. Environ. Dev. Sustain. 18, 777–798. https://doi.org/10.1007/s10668-015-9677-3

Amores, M.J., Meneses, M., Pasqualino, J., Antón, A., Castells, F., 2013. Environmental assessment of urban water cycle on Mediterranean conditions by LCA approach. J. Clean. Prod. 43, 84–92. https://doi.org/10.1016/J.JCLEPRO.2012.12.033

Anastas, P.T., 2012. Fundamental Changes to EPA’s Research Enterprise: The Path Forward. Environ. Sci. Technol. 46, 580–586. https://doi.org/10.1021/es203881e

Bakir, H.A., 2001. Sustainable wastewater management for small communities in the Middle East and North Africa. J. Environ. Manage. 61, 319–328. https://doi.org/10.1006/JEMA.2000.0414

Baresel, C., Dahlgren, L., Almemark, M., Lazic, A., 2015. Municipal wastewater reclamation for non-potable reuse – environmental assessments based on pilot-plant studies and system modelling. Water Sci. Technol. 72, 1635–1643. https://doi.org/10.2166/wst.2015.373

Barjoveanu, G., Comandaru, I.M., Rodriguez-Garcia, G., Hospido, A., Teodosiu, C., 2014. Evaluation of water services system through LCA. A case study for Iasi City, Romania. Int. J. Life Cycle Assess. 19, 449–462. https://doi.org/10.1007/s11367-013-0635-8

Beavis, P., Lundie, S., 2003. Integrated environmental assessment of tertiary and residuals treatment - LCA in the wastewater industry. Water Sci. Technol. 47, 109–116. https://doi.org/10.2166/wst.2003.0678

Benedetti, L., De Baets, B., Nopens, I., Vanrolleghem, P.A., 2010. Multi-criteria analysis of wastewater treatment plant design and control scenarios under uncertainty. Environ. Model. Softw. 25, 616–621. https://doi.org/10.1016/J.ENVSOFT.2009.06.003

122

Berndtsson, J.C., 2006. Experiences from the implementation of a urine separation system: Goals, planning, reality. Build. Environ. 41, 427–437. https://doi.org/10.1016/J.BUILDENV.2005.02.006

Bichai, F., Ryan, H., Fitzgerald, C., Williams, K., Abdelmoteleb, A., Brotchie, R., Komatsu, R., 2015. Understanding the role of alternative water supply in an urban water security strategy: an analytical framework for decision-making. Urban Water J. 12, 175–189. https://doi.org/10.1080/1573062X.2014.895844

Björklund, J., Geber, U., Rydberg, T., 2001. Emergy analysis of municipal wastewater treatment and generation of electricity by digestion of sewage sludge. Resour. Conserv. Recycl. 31, 293–316. https://doi.org/10.1016/S0921-3449(00)00091-4

Boland, N., Charkhgard, H., Savelsbergh, M., 2015. A Criterion Space Search Algorithm for Biobjective Mixed Integer Programming: The Triangle Splitting Method. INFORMS J. Comput. 27, 597–618. https://doi.org/10.1287/ijoc.2015.0646

Boulware, E.W.B., 2013. Alternative water sources and wastewater management. McGraw-Hill.

Bounds, T.R., Denn, G., 2017. Decentralized Design Considerations and Life Cycle Costs.

Brander, M., Davis, G., Sood, A., Wylie, C., Haughton, A., Lovell, J., 2011. Electricity-specific emission factors for grid electricity, ecometrica.

Brown, S., Beecher, N., Carpenter, A., 2010. Calculator Tool for Determining Greenhouse Gas Emissions for Biosolids Processing and End Use. Environ. Sci. Technol. 44, 9509–9515. https://doi.org/10.1021/es101210k

Byrne, D.M., Lohman, H.A.C., Cook, S.M., Peters, G.M., Guest, J.S., 2017. Life cycle assessment (LCA) of urban water infrastructure: emerging approaches to balance objectives and inform comprehensive decision-making. Environ. Sci. Water Res. Technol. 3, 1002–1014. https://doi.org/10.1039/C7EW00175D

Capodaglio, A., Capodaglio, G., A., 2017. Integrated, Decentralized Wastewater Management for Resource Recovery in Rural and Peri-Urban Areas. Resources 6, 22. https://doi.org/10.3390/resources6020022

Cashman, S., Ma, X., Mosley, J., Garland, J., Crone, B., Xue, X., 2018. Energy and greenhouse gas life cycle assessment and cost analysis of aerobic and anaerobic membrane bioreactor systems: Influence of scale, population density, climate, and methane recovery. Bioresour. Technol. 254, 56–66. https://doi.org/10.1016/J.BIORTECH.2018.01.060

Chang, Y., Reardon, D.J., Kwan, P., Boyd, G., Brant, J., Rakness, K.L., Furukawa, D., 2008. Evaluation of Dynamic Energy Consumption of Advanced Water and Wastewater Treatment Technologies, AWWA Research Foundation & California Energy Commission. Denver.

Chung, G., Lansey, K., Blowers, P., Brooks, P., Ela, W., Stewart, S., Wilson, P., 2008. A general water supply planning model: Evaluation of decentralized treatment. Environ. Model. Softw. 23, 893–905. https://doi.org/10.1016/J.ENVSOFT.2007.10.002

123

City of Lakeland, 2017. Water transmission and distribution [WWW Document]. URL https://www.lakelandgov.net/departments/water-utilities/water-transmission-distribution/ (accessed 9.20.17).

City of San Diego, 2013. Advanced Water Purification Facility Study Report [WWW Document]. URL https://www.sandiego.gov/sites/default/files/legacy/water/purewater/pdf/projectreports/awpfstudyreport.pdf (accessed 3.18.18).

City of Tampa, 2018. Howard F. Curren Advanced Wastewater Treatment Plant [WWW Document]. URL https://www.tampagov.net/wastewater/info/advanced-wastewater-treatment-plant

Copeland, C., Tiemann, M., 2010. Water Infrastructure Needs and Investment: Review and Analysis of Key Issues.

Cornejo, P.K., Zhang, Q., Mihelcic, J.R., 2016. How Does Scale of Implementation Impact the Environmental Sustainability of Wastewater Treatment Integrated with Resource Recovery? Environ. Sci. Technol. 50, 6680–6689. https://doi.org/10.1021/acs.est.5b05055

Cotton, C.A., Owen, D.M., Cline, G.C., Brodeur, T.P., 2001. UV disinfection costs for inactivating “Cryptosporidium.” J. Am. Water Works Assoc. 93, 82–94.

Daigger, G.T., 2009. Evolving Urban Water and Residuals Management Paradigms: Water Reclamation and Reuse, Decentralization, and Resource Recovery. Water Environ. Res. 81, 809–823. https://doi.org/10.2175/106143009X425898

Daigger, G.T., Crawford, G. V, 2007. Enhancing water system security and sustainability by incorporating centralized and decentralized water reclamation and reuse into urban water management systems, J. Environ. Eng. Manage.

Diaz-Elsayed, N., Rezaei, N., Guo, T., Mohebbi, S., Zhang, Q., 2019. Wastewater-based resource recovery technologies across scale: A review. Resour. Conserv. Recycl. 145, 94–112. https://doi.org/10.1016/J.RESCONREC.2018.12.035

Dreizin, Y., 2006. Ashkelon seawater desalination project — off-taker’s self costs, supplied water costs, total costs and benefits. Desalination 190, 104–116. https://doi.org/10.1016/J.DESAL.2005.08.006

Du, L.N., Li, Y., Chen, X.Y., Yang, J.X., 2011. Effect of eutrophication on molluscan community composition in the Lake Dianchi (China, Yunnan). Limnologica 41, 213–219. https://doi.org/10.1016/J.LIMNO.2010.09.006

Eggimann, S., Truffer, B., Maurer, M., 2015. To connect or not to connect? Modelling the optimal degree of centralisation for wastewater infrastructures. Water Res. 84, 218–231. https://doi.org/10.1016/J.WATRES.2015.07.004

El-Shafai, S.A., El-Gohary, F.A., Nasr, F.A., Peter van der Steen, N., Gijzen, H.J., 2007. Nutrient recovery from domestic wastewater using a UASB-duckweed ponds system. Bioresour. Technol. 98, 798–807. https://doi.org/10.1016/J.BIORTECH.2006.03.011

124

Fattahi, M., Govindan, K., 2017. Integrated forward/reverse logistics network design under uncertainty with pricing for collection of used products. Ann. Oper. Res. 253, 193–225. https://doi.org/10.1007/s10479-016-2347-5

Figueira, J., Greco, S., Ehrgott, M., 2005. Multiple criteria decision analysis: state of the art surveys.

Fleischmann, M., Beullens, P., Bloemhof-Ruwaard, J.M., Wassenhove, L.N., 2001. The impact of product recovery on logisticsnetwork design. Prod. Oper. Manag. 10, 156–173. https://doi.org/10.1111/j.1937-5956.2001.tb00076.x

Flores-Alsina, X., Rodríguez-Roda, I., Sin, G., Gernaey, K. V., 2008. Multi-criteria evaluation of wastewater treatment plant control strategies under uncertainty. Water Res. 42, 4485–4497. https://doi.org/10.1016/J.WATRES.2008.05.029

Florida Brownfields Redevelopment Atlas, 2019. Florida Brownfields Redevelopment Atlas [WWW Document]. URL https://www.usf.edu/arts-sciences/departments/anthropology/research/brownfields.aspx (accessed 6.5.19).

Florida DEP, 2016. 2016 Reuse Inventory Report [WWW Document]. URL https://floridadep.gov/water/domestic-wastewater/documents/2016-reuse-inventory-report (accessed 9.20.17).

Forrest, A.L., Fattah, K.P., Mavinic, D.S., Koch, F.A., 2008. Optimizing Struvite Production for Phosphate Recovery in WWTP. J. Environ. Eng. 134, 395–402. https://doi.org/10.1061/(ASCE)0733-9372(2008)134:5(395)

Funamizu, N., Iida, M., Sakakura, Y., Takakuwa, T., 2001. Reuse of heat energy in wastewater: implementation examples in Japan. Water Sci. Technol. 43, 277–285. https://doi.org/10.2166/wst.2001.0640

Gabarrell, X., Font, M., Vicent, T., Caminal, G., Sarrà, M., Blánquez, P., 2012. A comparative life cycle assessment of two treatment technologies for the Grey Lanaset G textile dye: biodegradation by Trametes versicolor and granular activated carbon adsorption. Int. J. Life Cycle Assess. 17, 613–624. https://doi.org/10.1007/s11367-012-0385-z

Gaius-obaseki, T., 2010. International Journal of Environmental Sciences, International Journal of Environmental Sciences. Integrated Pub. Association.

Gallego, A., Hospido, A., Moreira, M.T., Feijoo, G., 2008. Environmental performance of wastewater treatment plants for small populations. Resour. Conserv. Recycl. 52, 931–940. https://doi.org/10.1016/J.RESCONREC.2008.02.001

Gardels, D., Stansbury, J., Killion, S., Zhang, T., Neal, J., Alahmad, M., Berryman, C., Lau, S., Li, H., Schwer, A., Shi, J., Shen, Z., 2011. Economic Input-Output Life Cycle Assessment of Water Reuse Strategies in Residential Buildings, in: World Environmental and Water Resources Congress 2011. American Society of Civil Engineers, Reston, VA, pp. 1652–1662. https://doi.org/10.1061/41173(414)172

Gikas, P., Tchobanoglous, G., 2009. The role of satellite and decentralized strategies in water resources management. J. Environ. Manage. 90, 144–152. https://doi.org/10.1016/J.JENVMAN.2007.08.016

125

Glick, S., Guggemos, A.A., 2013. Rethinking Wastewater-Treatment Infrastructure: Case Study Using Life-Cycle Cost and Life-Cycle Assessment to Highlight Sustainability Considerations. J. Constr. Eng. Manag. 139, A5013002. https://doi.org/10.1061/(ASCE)CO.1943-7862.0000762

Godin, D., Bouchard, C., Vanrolleghem, P.A., 2012. Net environmental benefit: introducing a new LCA approach on wastewater treatment systems. Water Sci. Technol. 65, 1624. https://doi.org/10.2166/wst.2012.056

Guarini, M.R., Battisti, F., Chiovitti, A., 2018. A Methodology for the Selection of Multi-Criteria Decision Analysis Methods in Real Estate and Land Management Processes. Sustainability 10.

Guest, J.S., Skerlos, S.J., Barnard, J.L., Beck, M.B., Daigger, G.T., Hilger, H., Jackson, S.J., Karvazy, K., Kelly, L., Macpherson, L., Mihelcic, J.R., Pramanik, A., Raskin, L., Van Loosdrecht, M.C.M., Yeh, D., Love, N.G., 2009. A New Planning and Design Paradigm to Achieve Sustainable Resource Recovery from Wastewater. Environ. Sci. Technol. 43, 6126–6130. https://doi.org/10.1021/es9010515

Guo, T., Englehardt, J., Wu, T., 2014. Review of cost versus scale: water and wastewater treatment and reuse processes. Water Sci. Technol. 69, 223 LP – 234.

Guo, T., Englehardt, J.D., 2015. Principles for scaling of distributed direct potable water reuse systems: A modeling study. Water Res. 75, 146–163. https://doi.org/10.1016/J.WATRES.2015.02.033

Habibi, Farhad, Asadi, E., Sadjadi, S.J., 2017. Developing a location-inventory-routing model using METRIC approach in inventory policy. Uncertain Supply Chain Manag. 337–358. https://doi.org/10.5267/j.uscm.2017.4.003

Hamilton, A.J., Stagnitti, F.S., Premier, R., Boland, A.-M., 2006. Is the risk of illness through consuming vegetables irrigated with reclaimed wastewater different for different population groups? Water Sci. Technol. 54, 379–386. https://doi.org/10.2166/wst.2006.714

Hartley, T.W., 2006. Public perception and participation in water reuse. Desalination 187, 115–126. https://doi.org/10.1016/J.DESAL.2005.04.072

Heisler, J., Glibert, P.M., Burkholder, J.M., Anderson, D.M., Cochlan, W., Dennison, W.C., Dortch, Q., Gobler, C.J., Heil, C.A., Humphries, E., Lewitus, A., Magnien, R., Marshall, H.G., Sellner, K., Stockwell, D.A., Stoecker, D.K., Suddleson, M., 2008. Eutrophication and harmful algal blooms: A scientific consensus. Harmful Algae 8, 3–13. https://doi.org/10.1016/J.HAL.2008.08.006

Hendrickson, T.P., Nguyen, M.T., Sukardi, M., Miot, A., Horvath, A., Nelson, K.L., 2015. Life-Cycle Energy Use and Greenhouse Gas Emissions of a Building-Scale Wastewater Treatment and Nonpotable Reuse System. Environ. Sci. Technol. 49, 10303–10311. https://doi.org/10.1021/acs.est.5b01677

Hering, J.G., Waite, T.D., Luthy, R.G., Drewes, J.E., Sedlak, D.L., 2013. A Changing Framework for Urban Water Systems. Environ. Sci. Technol. 47, 10721–10726. https://doi.org/10.1021/es4007096

126

Hillsborough County, 2017a. Schedule of Rates [WWW Document]. URL http://www.hillsboroughcounty.org/library/hillsborough/media-center/documents/public-utilities/water-rate-schedule.pdf (accessed 9.25.17).

Hillsborough County, 2017b. Hillsborough County Biosolids Master Plan.

Ho, G., 2005. Technology for sustainability: the role of onsite, small and community scale technology. Water Sci. Technol. 51, 15–20. https://doi.org/10.2166/wst.2005.0346

Holloway, R.W., Miller-Robbie, L., Patel, M., Stokes, J.R., Munakata-Marr, J., Dadakis, J., Cath, T.Y., 2016. Life-cycle assessment of two potable water reuse technologies: MF/RO/UV–AOP treatment and hybrid osmotic membrane bioreactors. J. Memb. Sci. 507, 165–178. https://doi.org/10.1016/J.MEMSCI.2016.01.045

Hong, Jinglan, Hong, Jingmin, Otaki, M., Jolliet, O., 2009. Environmental and economic life cycle assessment for sewage sludge treatment processes in Japan. Waste Manag. 29, 696–703. https://doi.org/10.1016/J.WASMAN.2008.03.026

Hospido, A., Moreira, M.T., Fernández-Couto, M., Feijoo, G., 2004. Environmental performance of a municipal wastewater treatment plant. Int. J. Life Cycle Assess. 9, 261. https://doi.org/10.1007/BF02978602

Hospido, A., Moreira, T., Martín, M., Rigola, M., Feijoo, G., 2005. Environmental Evaluation of Different Treatment Processes for Sludge from Urban Wastewater Treatments: Anaerobic Digestion versus Thermal Processes (10 pp). Int. J. Life Cycle Assess. 10, 336–345. https://doi.org/10.1065/lca2005.05.210

Houillon, G., Jolliet, O., 2005. Life cycle assessment of processes for the treatment of wastewater urban sludge: energy and global warming analysis. J. Clean. Prod. 13, 287–299. https://doi.org/10.1016/J.JCLEPRO.2004.02.022

Hreiz, R., Roche, N., Benyahia, B., Latifi, M.A., 2015. Multi-objective Optimization of Small-size Wastewater Treatment Plants Operation. Comput. Aided Chem. Eng. 37, 2495–2500. https://doi.org/10.1016/B978-0-444-63576-1.50110-2

Iacob, V.-S., 2013. The Wastewater – A Problem of Integrated Urban Water Management. Procedia Econ. Financ. 6, 436–443. https://doi.org/10.1016/S2212-5671(13)00160-3

International Energy Agency (IEA), 2014. World Energy Outlook. Paris, France.

International Organization for Standardization (ISO), 2006a. Environmental Management: Life Cycle Assessment; Principals and Framework [WWW Document]. URL https://www.iso.org/standard/37456.html (accessed 9.27.17).

International Organization for Standardization (ISO), 2006b. Environmental Management: Life Cycle Assessment; Requirements and Guidelines [WWW Document]. URL https://www.iso.org/standard/38498.html (accessed 9.27.17).

Jeihoonian, M., Kazemi Zanjani, M., Gendreau, M., 2016a. Accelerating Benders decomposition for closed-loop supply chain network design: Case of used durable products with different quality levels. Eur. J. Oper. Res. 251, 830–845. https://doi.org/10.1016/J.EJOR.2015.12.052

127

Jeihoonian, M., Zanjani, M.K., Gendreau, M., 2016b. A stochastic programming approach for closed-loop supply chain network design under uncertain quality status. 6th International Conference on Information Systems, Logistics and Supply Chain, Bordeaux, France.

Jeong, H., Kim, H., Jang, T., 2016. Irrigation Water Quality Standards for Indirect Wastewater Reuse in Agriculture: A Contribution toward Sustainable Wastewater Reuse in South Korea. Water 8, 169. https://doi.org/10.3390/w8040169

John, S.T., Sridharan, R., Ram Kumar, P.N., Krishnamoorthy, M., 2018. Multi-period reverse logistics network design for used refrigerators. Appl. Math. Model. 54, 311–331. https://doi.org/10.1016/J.APM.2017.09.053

Kara, S. Soner, Onut, S., 2010. A stochastic optimization approach for paper recycling reverse logistics network design under uncertainty. Int. J. Environ. Sci. Technol. 7, 717–730. https://doi.org/10.1007/BF03326181

Kavvada, O., Horvath, A., Stokes-Draut, J.R., Hendrickson, T.P., Eisenstein, W.A., Nelson, K.L., 2016. Assessing Location and Scale of Urban Nonpotable Water Reuse Systems for Life-Cycle Energy Consumption and Greenhouse Gas Emissions. Environ. Sci. Technol. 50, 13184–13194. https://doi.org/10.1021/acs.est.6b02386

Kim, D., Bowen, J.D., Ozelkan, E.C., 2015. Optimization of wastewater treatment plant operation for greenhouse gas mitigation. J. Environ. Manage. 163, 39–48. https://doi.org/10.1016/J.JENVMAN.2015.07.005

Kolios, A., Mytilinou, V., Lozano-Minguez, E., Salonitis, K., 2016. A Comparative Study of Multiple-Criteria Decision-Making Methods under Stochastic Inputs. Energies 9.

Kumar, S., Dey, P., 2011. Effects of different mulches and irrigation methods on root growth, nutrient uptake, water-use efficiency and yield of strawberry. Sci. Hortic. (Amsterdam). 127, 318–324. https://doi.org/10.1016/J.SCIENTA.2010.10.023

Lam, L., Kurisu, K., Hanaki, K., 2015. Comparative environmental impacts of source-separation systems for domestic wastewater management in rural China. J. Clean. Prod. 104, 185–198. https://doi.org/10.1016/J.JCLEPRO.2015.04.126

Lamichhane, K.M., Babcock, R.W., 2013. Survey of attitudes and perceptions of urine-diverting toilets and human waste recycling in Hawaii. Sci. Total Environ. 443, 749–756. https://doi.org/10.1016/J.SCITOTENV.2012.11.039

Larsen, T. V, Alder, A.C., Eggen, R.I.L., Maurer, M., Lienert, J., 2009. Source separation: Will we see a paradigm shift in wastewater handling? Environ. Sci. Technol. 43, DOI 10.1021/es803001r.

Lehtoranta, S., Vilpas, R., Mattila, T.J., 2014. Comparison of carbon footprints and eutrophication impacts of rural on-site wastewater treatment plants in Finland. J. Clean. Prod. 65, 439–446. https://doi.org/10.1016/J.JCLEPRO.2013.08.024

Lemos, D., Dias, A.C., Gabarrell, X., Arroja, L., 2013. Environmental assessment of an urban water system. J. Clean. Prod. 54, 157–165. https://doi.org/10.1016/J.JCLEPRO.2013.04.029

128

Leverenz, H.L., Tchobanoglous, G., Asano, T., 2011. Direct potable reuse: a future imperative. J. Water Reuse Desalin. 1, 2–10. https://doi.org/10.2166/wrd.2011.000

Levine, Audrey D., Asano, T., 2004. Peer Reviewed: Recovering Sustainable Water from Wastewater. Environ. Sci. Technol. 38, 201A-208A. https://doi.org/10.1021/es040504n

Lieckens, K., Vandaele, N., 2012. Multi-level reverse logistics network design under uncertainty. Int. J. Prod. Res. 50, 23–40. https://doi.org/10.1080/00207543.2011.571442

Lieckens, K.T., Colen, P.J., Lambrecht, M.R., 2013. Optimization of a stochastic remanufacturing network with an exchange option. Decis. Support Syst. 54, 1548–1557. https://doi.org/10.1016/J.DSS.2012.05.057

Lim, S.-R., Park, J.M., 2009. Environmental impact minimization of a total wastewater treatment network system from a life cycle perspective. J. Environ. Manage. 90, 1454–1462. https://doi.org/10.1016/J.JENVMAN.2008.09.005

Linder, M., Williander, M., 2015. Circular Business Model Innovation: Inherent Uncertainties. Bus. Strateg. Environ. 26, 182–196. https://doi.org/10.1002/bse.1906

Little, J.C., Hester, E.T., Carey, C.C., 2016. Assessing and Enhancing Environmental Sustainability: A Conceptual Review. Environ. Sci. Technol. 50, 6830–6845. https://doi.org/10.1021/acs.est.6b00298

Liu, H., Ramnarayanan, R., Logan, B.E., 2004. Production of Electricity during Wastewater Treatment Using a Single Chamber Microbial Fuel Cell. Environ. Sci. Technol. 38, 2281–2285. https://doi.org/10.1021/es034923g

Loh, H.P., Lyons, J., White, C.W., 2002. Process Equipment Cost Estimation, Final Report. Pittsburgh, PA, and Morgantown, WV (United States). https://doi.org/10.2172/797810

Loubet, P., Roux, P., Loiseau, E., Bellon-Maurel, V., 2014. Life cycle assessments of urban water systems: A comparative analysis of selected peer-reviewed literature. Water Res. 67, 187–202. https://doi.org/10.1016/J.WATRES.2014.08.048

Loulou, R., Kanudia, A., 1999. Minimax regret strategies for greenhouse gas abatement: methodology and application. Oper. Res. Lett. 25, 219–230. https://doi.org/10.1016/S0167-6377(99)00049-8

Lundin, M., Olofsson, M., Pettersson, G.., Zetterlund, H., 2004. Environmental and economic assessment of sewage sludge handling options. Resour. Conserv. Recycl. 41, 255–278. https://doi.org/10.1016/J.RESCONREC.2003.10.006

Lyeme, Halidi A, Mushi, A., Nkansah-Gyekye, Y., 2016. Multi-objective Optimization Model Formulation for Solid Waste Management: A Case of Dar es Salaam - Tanzania. Unpubl. Manuscr. 2017, 1–15.

Lyons, E., Zhang, P., Benn, T., Sharif, F., Li, K., Crittenden, J., Costanza, M., Chen, Y.S., 2009. Life cycle assessment of three water supply systems: importation, reclamation and desalination. Water Sci. Technol. Water Supply 9, 439–448. https://doi.org/10.2166/ws.2009.534

129

Mainali, B., Pham, T.T.N., Ngo, H.H., Guo, W., Miechel, C., O’Halloran, K., Muthukaruppan, M., Listowski, A., 2013. Vision and perception of community on the use of recycled water for household laundry: A case study in Australia. Sci. Total Environ. 463–464, 657–666. https://doi.org/10.1016/J.SCITOTENV.2013.06.008

Mani, V., Agarwal, R., Gunasekaran, A., Papadopoulos, T., Dubey, R., Childe, S.J., 2016. Social sustainability in the supply chain: Construct development and measurement validation. Ecol. Indic. 71, 270–279. https://doi.org/10.1016/j.ecolind.2016.07.007

Martí, N., Pastor, L., Bouzas, A., Ferrer, J., Seco, A., 2010. Phosphorus recovery by struvite crystallization in WWTPs: Influence of the sludge treatment line operation. Water Res. 44, 2371–2379. https://doi.org/10.1016/J.WATRES.2009.12.043

Marufuzzaman, M., Ekşioğlu, S.D., Hernandez, R., 2015. Truck versus pipeline transportation cost analysis of wastewater sludge. Transp. Res. Part A Policy Pract. 74, 14–30. https://doi.org/10.1016/J.TRA.2015.02.001

Matos, C., Pereira, S., Amorim, E.V., Bentes, I., Briga-Sá, A., 2014. Wastewater and greywater reuse on irrigation in centralized and decentralized systems — An integrated approach on water quality, energy consumption and CO2 emissions. Sci. Total Environ. 493, 463–471. https://doi.org/10.1016/J.SCITOTENV.2014.05.129

Maurer, M., 2009. Specific net present value: An improved method for assessing modularisation costs in water services with growing demand. Water Res. 43, 2121–2130. https://doi.org/10.1016/J.WATRES.2009.02.008

Maurer, M., Rothenberger, D., Larsen, T.A., 2005. Decentralised wastewater treatment technologies from a national perspective: at what cost are they competitive? Water Sci. Technol. Water Supply 5, 145–154. https://doi.org/10.2166/ws.2005.0059

Maurer, M., Schwegler, P., Larsen, T.A., 2003. Nutrients in urine: energetic aspects of removal and recovery. Water Sci. Technol. 48, 37–46. https://doi.org/10.2166/wst.2003.0011

Means, E., 2004. Water and wastewater industry energy efficiency: a research roadmap.

Mekonnen, M.M., Hoekstra, A.Y., 2016. Four billion people facing severe water scarcity. Sci. Adv. 2. https://doi.org/10.1126/sciadv.1500323

Meneses, M., Pasqualino, J.C., Castells, F., 2010. Environmental assessment of urban wastewater reuse: Treatment alternatives and applications. Chemosphere 81, 266–272. https://doi.org/10.1016/J.CHEMOSPHERE.2010.05.053

Metcalf, E., Asano, T., Burton, F., Leverenz, H., Tsuchihashi, R., Tchobanoglous, G., 2007. Water Reuse: Issues, Technologies, and Applications. Mc-Graw Hill. NewYork, USA.

Molinos-Senante, M., Garrido-Baserba, M., Reif, R., Hernández-Sancho, F., Poch, M., 2012. Assessment of wastewater treatment plant design for small communities: Environmental and economic aspects. Sci. Total Environ. 427–428, 11–18. https://doi.org/10.1016/J.SCITOTENV.2012.04.023

130

Molinos-Senante, M., Gómez, T., Caballero, R., Hernández-Sancho, F., Sala-Garrido, R., 2015. Assessment of wastewater treatment alternatives for small communities: An analytic network process approach. Sci. Total Environ. 532, 676–687. https://doi.org/10.1016/J.SCITOTENV.2015.06.059

Moody, F. L., 1944. Friction factors for pipe flow. Trans. ASME 66, 671–684.

Morera, S., Corominas, L., Poch, M., Aldaya, M.M., Comas, J., 2016. Water footprint assessment in wastewater treatment plants. J. Clean. Prod. 112, 4741–4748. https://doi.org/10.1016/J.JCLEPRO.2015.05.102

Muñoz, I., Rodríguez, A., Rosal, R., Fernández-Alba, A.R., 2009. Life Cycle Assessment of urban wastewater reuse with ozonation as tertiary treatment: A focus on toxicity-related impacts. Sci. Total Environ. 407, 1245–1256. https://doi.org/10.1016/J.SCITOTENV.2008.09.029

Nakakubo, T., Tokai, A., Ohno, K., 2012. Comparative assessment of technological systems for recycling sludge and food waste aimed at greenhouse gas emissions reduction and phosphorus recovery. J. Clean. Prod. 32, 157–172. https://doi.org/10.1016/J.JCLEPRO.2012.03.026

Nakatani, J., Konno, K., Moriguchi, Y., 2017. Variability-based optimal design for robust plastic recycling systems. Resour. Conserv. Recycl. 116, 53–60. https://doi.org/10.1016/J.RESCONREC.2016.09.020

Nancarrow, B.E., Leviston, Z., Po, M., Porter, N.B., Tucker, D.I., 2008. What drives communities’ decisions and behaviours in the reuse of wastewater. Water Sci. Technol. 57, 485–491. https://doi.org/10.2166/wst.2008.160

National Research Council, 2012. Water reuse: potential for expanding the nation’s water supply through reuse of municipal wastewater. National Academies Press.

Navigant Consulting Inc., 2006. Refining Estimates of Water‐Related Energy Use in California. California Energy Commission, PIER Industrial/Agricultural/Water End Use Energy Efficiency Program.

NCEES, 2013. Fundamentals of Engineering Reference Handbook, 9.1. ed.

Newman, J.P., Dandy, G.C., Maier, H.R., 2014. Multiobjective optimization of cluster-scale urban water systems investigating alternative water sources and level of decentralization. Water Resour. Res. 50, 7915–7938. https://doi.org/10.1002/2013WR015233

Nouri, J., Naddafi, K., Nabizadeh, R., Jafarinia, M., 2006. Energy recovery from wastewater treatment plant. Asian J. Water, Environ. Pollut. 4, 145–149.

Novotny, V., 2010. Urban Water and Energy Use From Current US Use to Cities of the Future. Cities of the Future/Urban River Restoration. Water Environment Federation, in: Water Environment Federation. pp. 118–140.

Opher, T., Friedler, E., 2016. Comparative LCA of decentralized wastewater treatment alternatives for non-potable urban reuse. J. Environ. Manage. 182, 464–476. https://doi.org/10.1016/J.JENVMAN.2016.07.080

131

Orange County Public Works [WWW Document], 2018. URL http://www.ocpublicworks.com/survey/products/maps

Ortiz, M., Raluy, R.G., Serra, L., 2007. Life cycle assessment of water treatment technologies: wastewater and water-reuse in a small town. Desalination 204, 121–131. https://doi.org/10.1016/J.DESAL.2006.04.026

Owens, J. W., 2001. Water Resources in Life-Cycle Impact Assessment: Considerations in Choosing Category Indicators. J. Ind. Ecol. 5, 37–54. https://doi.org/10.1162/10881980152830123

Palme, U., Lundin, M., Tillman, A.-M., Molander, S., 2005. Sustainable development indicators for wastewater systems – researchers and indicator users in a co-operative case study. Resour. Conserv. Recycl. 43, 293–311. https://doi.org/10.1016/J.RESCONREC.2004.06.006

Palmer, I.E., Gehl, R.J., Ranney, T.G., Touchell, D., George, N., 2014. Biomass yield, nitrogen response, and nutrient uptake of perennial bioenergy grasses in North Carolina. Biomass and Bioenergy 63, 218–228. https://doi.org/10.1016/J.BIOMBIOE.2014.02.016

Pant, D., Singh, A., Van Bogaert, G., Gallego, Y.A., Diels, L., Vanbroekhoven, K., 2011. An introduction to the life cycle assessment (LCA) of bioelectrochemical systems (BES) for sustainable energy and product generation: Relevance and key aspects. Renew. Sustain. Energy Rev. 15, 1305–1313. https://doi.org/10.1016/J.RSER.2010.10.005

Pasqualino, J.C., Meneses, M., Castells, F., 2011. Life Cycle Assessment of Urban Wastewater Reclamation and Reuse Alternatives. J. Ind. Ecol. 15, 49–63. https://doi.org/10.1111/j.1530-9290.2010.00293.x

Pereira, B.F.F., He, Z.L., Silva, M.S., Herpin, U., Nogueira, S.F., Montes, C.R., Melfi, A.J., 2011. Reclaimed wastewater: Impact on soil–plant system under tropical conditions. J. Hazard. Mater. 192, 54–61. https://doi.org/10.1016/J.JHAZMAT.2011.04.095

Peres, N.A., Price, J.F., Stall, W.M., Chandler, C.K., Olson, S.M., Taylor, T.G., Smith, S.A., Simonne, E.H., 2011. Strawberry production in Florida. pp. 375–382.

Peters, G.M., Lundie, S., 2001. Life-Cycle Assessment of Biosolids Processing Options. J. Ind. Ecol. 5, 103–121. https://doi.org/10.1162/10881980152830169

Phillips, G., Lyche-Solheim, A., Skjelbred, B., Mischke, U., Drakare, S., Free, G., Järvinen, M., de Hoyos, C., Morabito, G., Poikane, S., Carvalho, L., 2013. A phytoplankton trophic index to assess the status of lakes for the Water Framework Directive. Hydrobiologia 704, 75–95. https://doi.org/10.1007/s10750-012-1390-8

Pintilie, L., Torres, C.M., Teodosiu, C., Castells, F., 2016. Urban wastewater reclamation for industrial reuse: An LCA case study. J. Clean. Prod. 139, 1–14. https://doi.org/10.1016/J.JCLEPRO.2016.07.209

Po, M., Kaercher, J.D., Nancarrow, B.E., 2003. Literature review of factors influencing public perceptions of water reuse.

132

Polk County, 2017. Utilities Rates and Fees [WWW Document]. URL https://www.polk-county.net/utilities/utilities-rates-and-fees (accessed 9.25.17).

Qi, C., Chang, N. Bin, 2013. Integrated carbon footprint and cost evaluation of a drinking water infrastructure system for screening expansion alternativesq. J. Clean. Prod. 60, 170–181. https://doi.org/10.1016/j.jclepro.2012.10.032

Rabaey, K., Lissens, G., Siciliano, S.D., Verstraete, W., 2003. A microbial fuel cell capable of converting glucose to electricity at high rate and efficiency. Biotechnol. Lett. 25, 1531–1535. https://doi.org/10.1023/A:1025484009367

Ramezani, M., Bashiri, M., Tavakkoli-Moghaddam, R., 2013. A new multi-objective stochastic model for a forward/reverse logistic network design with responsiveness and quality level. Appl. Math. Model. 37, 328–344. https://doi.org/10.1016/J.APM.2012.02.032

REISS Engineering, 2009. Polk County comprehensive water supply plan [WWW Document]. URL http://www.reisscivilengineering.com/filerepository/ClientDocuments/Polk-CountySWFWMD-and-SWFMD_Polk-County-Comprehensive-Water-Supply-Plan.pdf (accessed 9.19.17).

Ren, D., Colosi, L.M., Smith, J.A., 2013. Evaluating the Sustainability of Ceramic Filters for Point-of-Use Drinking Water Treatment. Environ. Sci. Technol. 47, 11206–11213. https://doi.org/10.1021/es4026084

Ren, J., Liang, H., 2017. Multi-criteria group decision-making based sustainability measurement of wastewater treatment processes. Environ. Impact Assess. Rev. 65, 91–99. https://doi.org/10.1016/J.EIAR.2017.04.008

Rezaei, N., Diaz-Elsayed, N., Mohebbi, S., Xie, X., Zhang, Q., 2019. A multi-criteria sustainability assessment of water reuse applications: a case study in Lakeland, Florida. Environ. Sci. Water Res. Technol. 5, 102–118. https://doi.org/10.1039/C8EW00336J

Risch, E., Gutierrez, O., Roux, P., Boutin, C., Corominas, L., 2015. Life cycle assessment of urban wastewater systems: Quantifying the relative contribution of sewer systems. Water Res. 77, 35–48. https://doi.org/10.1016/J.WATRES.2015.03.006

RSMeans, 2017. Historical Cost Indexes [WWW Document]. URL https://www.rsmeansonline.com/references/unit/refpdf/hci.pdf (accessed 9.27.17).

Ruiz-Rosa, I., García-Rodríguez, F.J., Mendoza-Jiménez, J., 2016. Development and application of a cost management model for wastewater treatment and reuse processes. J. Clean. Prod. 113, 299–310. https://doi.org/10.1016/J.JCLEPRO.2015.12.044

Rygaard, M., Binning, P.J., Albrechtsen, H.-J., 2011. Increasing urban water self-sufficiency: New era, new challenges. J. Environ. Manage. 92, 185–194. https://doi.org/10.1016/J.JENVMAN.2010.09.009

Rygaard, M., Godskesen, B., Jørgensen, C., Hoffmann, B., 2014. Holistic assessment of a secondary water supply for a new development in Copenhagen, Denmark. Sci. Total Environ. 497–498, 430–439. https://doi.org/10.1016/J.SCITOTENV.2014.07.078

133

Sander, K., Murthy, G.S., 2010. Life cycle analysis of algae biodiesel. Int. J. Life Cycle Assess. 15, 704–714. https://doi.org/10.1007/s11367-010-0194-1

Seidel, C., Ghosh, A., Tang, G., Hubbs, S.A., Raucher, R., Crawford-Brown, D., 2014. Executive Summary: Identifying Meaningful Opportunities for Drinking Water Health Risk Reduction in the United States, Water Research Foundation.

Sharma, A.K., Grant, A.L., Grant, T., Pamminger, F., Opray, L., 2009. Environmental and Economic Assessment of Urban Water Services for a Greenfield Development. Environ. Eng. Sci. 26, 921–934. https://doi.org/10.1089/ees.2008.0063

Shehabi, A., Stokes, J.R., Horvath, A., 2012. Energy and air emission implications of a decentralized wastewater system. Environ. Res. Lett. 7, 024007. https://doi.org/10.1088/1748-9326/7/2/024007

Sierra-Altamiranda, A., Charkhgard, H., 2018a. A New Exact Algorithm to Optimize a Linear Function Over the Set of Efficient Solutions for Bi-objective Mixed Integer Linear Programs. optimization-online.org.

Sierra-Altamiranda, A., Charkhgard, H., 2018b. OOESAlgorithm. jl: A julia package for optimizing a linear function over the set of efficient solutions for bi-objective mixed integer linear programming. optimization-online.org.

Sinha, R., Lennartsson, M., Frostell, B., 2016. Environmental footprint assessment of building structures: A comparative study. Build. Environ. 104, 162–171. https://doi.org/10.1016/J.BUILDENV.2016.05.012

Slagstad, H., Brattebø, H., 2014. Life cycle assessment of the water and wastewater system in Trondheim, Norway – A case study. Urban Water J. 11, 323–334. https://doi.org/10.1080/1573062X.2013.795232

Soller, J.A., Nellor, M.H., 2011. Development and Application of Tools to Assess and Understand the Relative Risks of Regulated Chemicals in Indirect Potable Reuse Project: the Montebello Forebay Groundwater Recharge Project. Alexandria, VA.

SRI International, 2015. The Florida golf economy [WWW Document]. URL https://golf2020.com/wp-content/uploads/2017/11/fl_golf_full_rpt_sri_final1_16_15.pdf (accessed 9.19.17).

State of Florida, 2018. Sewage disposal facilities; advanced and secondary waste treatment [WWW Document]. URL http://www.leg.state.fl.us/statutes/index.cfm?mode=View Statutes&SubMenu=1&App_mode=Display_Statute&Search_String=403.086&URL=0400-0499/0403/Sections/0403.086.html (accessed 5.25.18).

Stokes, J., Horvath, A., 2006. Life Cycle Energy Assessment of Alternative Water Supply Systems (9 pp). Int. J. Life Cycle Assess. 11, 335–343. https://doi.org/10.1065/lca2005.06.214

Stokes, J.R., Horvath, A., 2010. Supply-chain environmental effects of wastewater utilities. Environ. Res. Lett. 5, 014015. https://doi.org/10.1088/1748-9326/5/1/014015

134

Suh, Y.-J., Rousseaux, P., 2002. An LCA of alternative wastewater sludge treatment scenarios. Resour. Conserv. Recycl. 35, 191–200. https://doi.org/10.1016/S0921-3449(01)00120-3

Svanström, M., Fröling, M., Olofsson, M., Lundin, M., 2005. Environmental assessment of supercritical water oxidation and other sewage sludge handling options. Waste Manag. Res. 23, 356–366. https://doi.org/10.1177/0734242X05054324

Tang, X., Pronk, W., Ding, A., Cheng, X., Wang, J., Xie, B., Li, G., Liang, H., 2018. Coupling GAC to ultra-low-pressure filtration to modify the biofouling layer and bio-community: Flux enhancement and water quality improvement. Chem. Eng. J. 333, 289–299. https://doi.org/10.1016/J.CEJ.2017.09.111

Tarnacki, K., Meneses, M., Melin, T., van Medevoort, J., Jansen, A., 2012. Environmental assessment of desalination processes: Reverse osmosis and Memstill®. Desalination 296, 69–80. https://doi.org/10.1016/J.DESAL.2012.04.009

Tchobanoglous, G., Leverenz, H., 2013. The rationale for decentralization of wastewater infrastructure, in: Larsen, T., Udert, K., Lienert, J. (Eds.), Source Separation and Decentralization for Wastewater Management. IWA Publishing, London, UK, pp. 101–115.

Thibodeau, C., Monette, F., Bulle, C., Glaus, M., 2014. Comparison of black water source-separation and conventional sanitation systems using life cycle assessment. J. Clean. Prod. 67, 45–57. https://doi.org/10.1016/J.JCLEPRO.2013.12.012

Tidåker, P., Sjöberg, C., Jönsson, H., 2007. Local recycling of plant nutrients from small-scale wastewater systems to farmland—A Swedish scenario study. Resour. Conserv. Recycl. 49, 388–405. https://doi.org/10.1016/J.RESCONREC.2006.05.004

Tillman, A.-M., Svingby, M., Lundström, H., 1998. Life cycle assessment of municipal waste water systems. Int. J. Life Cycle Assess. 3, 145–157. https://doi.org/10.1007/BF02978823

Tjandraatmadja, G., Sharma, A.K., Grant, T., Pamminger, F., 2013. A Decision Support Methodology for Integrated Urban Water Management in Remote Settlements. Water Resour. Manag. 27, 433–449. https://doi.org/10.1007/s11269-012-0195-x

Toso, E.A. V., Alem, D., 2014. Effective location models for sorting recyclables in public management. Eur. J. Oper. Res. 234, 839–860. https://doi.org/10.1016/J.EJOR.2013.10.035

Trussell, R., Trussell, S., Salveson, A., Steinle-Darling, E., He, Q., Snyder, S., Gerrity, D., 2015. quivalency of Advanced Treatment Trains for Potable Reuse: User Manual for Treatment Train Toolbox. Alexandria, Virginia.

United States Census Bureau, 2017. Quick Facts selected: Hillsborough County, Florida [WWW Document]. URL https://www.census.gov/quickfacts/fact/table/hillsboroughcountyflorida/PST120216 (accessed 3.3.18).

United States Census Bureau, 2016. Quick Facts selected: Lakeland city, Florida [WWW Document]. URL https://www.census.gov/quickfacts/fact/table/lakelandcityflorida/PST045216 (accessed 9.20.17).

135

United States Census Bureau [WWW Document], 2018. URL https://www.census.gov/ (accessed 4.15.18)

US EPA, 2017. Emissions & Generation Resource Integrated Database (eGRID) [WWW Document]. URL https://www.epa.gov/energy/emissions-generation-resource-integrated-database-egrid (accessed 4.15.18).

US EPA, 2018. Environmental Justice Screening and Mapping Tool [WWW Document]. URL https:// www.epa.gov/ejscreen (accessed 6.10.19).

US EPA, 2012. Guidelines for Water Reuse [WWW Document]. URL https://nepis.epa.gov/Adobe/PDF/P100FS7K.pdf (accessed 8.15.17).

US EPA, 2007. Opportunities for and Benefits of Combined Heat and Power at Wastewater Treatment Facilities.

US EPA, 2003. Ultraviolet Disinfection Guidance Manual [WWW Document]. URL https://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=901T0000.TXT (accessed 9.15.17).

US Gazeteer, 2017. Incorporated Cities, Towns & Census Designated Places in Hillsborough County [WWW Document]. URL https://florida.hometownlocator.com/counties/cities,cfips,057,c,hillsborough.cfm (accessed 2.25.18).

Valiela, I., Owens, C., Elmstrom, E., Lloret, J., 2016. Eutrophication of Cape Cod estuaries: Effect of decadal changes in global-driven atmospheric and local-scale wastewater nutrient loads. Mar. Pollut. Bull. 110, 309–315. https://doi.org/10.1016/J.MARPOLBUL.2016.06.047

Van der Bruggen, B., 2010. Chapter 3 The Global Water Recycling Situation. Sustain. Sci. Eng. 2, 41–62. https://doi.org/10.1016/S1871-2711(09)00203-7

Vanhoutte, B., Schenkels, L., Ceusters, J., De Proft, M.P., 2017. Water and nutrient uptake in Vriesea cultivars: Trichomes vs. Roots. Environ. Exp. Bot. 136, 21–30. https://doi.org/10.1016/J.ENVEXPBOT.2017.01.003

Verbyla, M.E., Oakley, S.M., Mihelcic, J.R., 2013. Wastewater Infrastructure for Small Cities in an Urbanizing World: Integrating Protection of Human Health and the Environment with Resource Recovery and Food Security. Environ. Sci. Technol. 47, 3598–3605. https://doi.org/10.1021/es3050955

Verstraete, W., Van de Caveye, P., Diamantis, V., 2009. Maximum use of resources present in domestic “used water.” Bioresour. Technol. 100, 5537–5545. https://doi.org/10.1016/J.BIORTECH.2009.05.047

Voltolina, D., Gómez-Villa, H., Correa, G., 2005. Nitrogen removal and recycling by Scenedesmus obliquus in semicontinuous cultures using artificial wastewater and a simulated light and temperature cycle. Bioresour. Technol. 96, 359–362. https://doi.org/10.1016/J.BIORTECH.2004.04.004

136

Wang, H., Brown, S.L., Magesan, G.N., Slade, A.H., Quintern, M., Clinton, P.W., Payn, T.W., 2008. Technological options for the management of biosolids. Environ. Sci. Pollut. Res. - Int. 15, 308–317. https://doi.org/10.1007/s11356-008-0012-5

Wilderer, P.A., Schreff, D., 2000. Decentralized and centralized wastewater management: a challenge for technology developers. Water Sci. Technol. 41, 1–8. https://doi.org/10.2166/wst.2000.0001

Wilkie, A.C., Mulbry, W.W., 2002. Recovery of dairy manure nutrients by benthic freshwater algae. Bioresour. Technol. 84, 81–91. https://doi.org/10.1016/S0960-8524(02)00003-2

Zhang, Q., 2015. CAREER: Envisioning Integrated Wastewater Management through the Lens of Reverse Logistics.

Zhang, R., Xie, W.-M., Yu, H.-Q., Li, W.-W., 2014. Optimizing municipal wastewater treatment plants using an improved multi-objective optimization method. Bioresour. Technol. 157, 161–165. https://doi.org/10.1016/J.BIORTECH.2014.01.103

Zheng, J., Egger, C., Lienert, J., 2016. A scenario-based MCDA framework for wastewater infrastructure planning under uncertainty. J. Environ. Manage. 183, 895–908. https://doi.org/10.1016/J.JENVMAN.2016.09.027

Zhou, J., Chang, V.W.-C., Fane, A.G., 2011. Environmental life cycle assessment of reverse osmosis desalination: The influence of different life cycle impact assessment methods on the characterization results. Desalination 283, 227–236. https://doi.org/10.1016/J.DESAL.2011.04.066

Zimmerman, J.B., Mihelcic, J.R., Smith, J., 2008. Global stressors on water quality and quantity. Environ. Sci. Technol. 42, 4247.

137

APPENDIX A: NOMENCLATURE

ANPV Annualized net present value MMF Multi-media filtration

ARTkr The amount of reclaimed water to be transport from WWTP at candidate location k to customer cluster r (m3/year)

MSWM Municipal solid waste management

ASNPV Annualized specific net present value N Nitrogen

AWTrk The amount of wastewater to be transport from customer cluster r to WWTP at candidate location k (m3/year)

n Total life time for constructed facilities (year)

BAC Biological activated carbon NPR Non-potable reuse

Capkt Capacity of WWTP k with technology/scale t (m3/year)

NPV Net present value

CAS Conventional activated sludge NR Normalized regret score

CCkt Total fixed cost of construction of a WWTP with technology/scale t at candidate location k (per plant's capacity) ($/(m3/year))

NS2 Normalized sample variance

CFP Carbon footprint O&M Operation and maintenance

CPA First constant for calculation of the costs associated with the implementation and maintenance of pipelines ($/mi/(m3/year))

OCt Annual operation and maintenance cost of WWTP with technology/scale t ($/m3)

CPB Second constant for calculation of the costs associated with the implementation and maintenance of pipelines ($/mi)

P Phosphorus

CPU Central processing unit Pr Number of customers in cluster r

CTR Unit transportation cost of reclaimed water ($/mi/m3)

Pt Water demand

CTW Unit transportation cost of wastewater ($/mi/m3)

PV Present value

DPR Direct potable reuse qrk A zero/one variable that equals 1 if cluster r and WWTP constructed at candidate location k are related, 0 otherwise

EDSS Environmental decision-making support system

R Regret score

EPA Environmental protection agency r Set of Clusters (population centers)

ESI Environmental sustainability indicator Ṝ Final regret score

EU Eutrophication RBC Rotating biological contactor

FDEP Florida Department of Environmental Protection

RO Reverse osmosis

FV Future value RO Reverse osmosis

GAC Granular activated carbon S2 Sample variance

GHG Greenhouse gas SF Sand filtration

GiB Gibibyte SRWt Selling price of reclaimed water produced by a WWTP with technology/scale t ($/m3)

138

GPt Amount of greenhouse gas emission from process of each m3 of wastewater in a WWTP with technology/scale t (gr CO2-eq/m3)

t Set of technologies/scales for WWTPs

GST Ground storage tank TF Tertiary filtration

GTR Amount of greenhouse gas emission for reclaimed water transfer (gr CO2-eq/mi/m3)

TN Total nitrogen

GTW Amount of greenhouse gas emission for wastewater transfer (gr CO2-eq/mi/m3)

Tp Planning horizon

i Annual discount rate TP Total phosphorus

IPR Indirect potable reuse UF Ultra-filtration

ISO International Organization for Standardization

US united States

IX Ion exchange US EPA

united States environmental protection agency

k Set of candidate sites for WWTPs UV Ultra violet

KWh Kilowatt hour VRR Value of resource recovery

LCA Life cycle assessment w Weighting factor

LCCA Life cycle cost analysis WHO World Health Organization

Lir, Lri Distance between customer cluster r and WWTP in candidate location k (mi)

WTP Water treatment plant

LTS Long-term support system WW Wastewater

M$ Million dollar WWTP Wastewater treatment plant

MBF Membrane filtration ZC Costs associated with the entire wastewater system (M$/year)

MBR Membrane bioreactor ZE GHG emission of the wastewater system (ton CO2-eq/year)

ME Environmental marginal benefit (ton CO2-eq/M$)

α The percentage of reclaimed water to be sold to the customers (%)

MF Micro-filtration μ Amount of recyclable wastewater generated by each customer (m3/capita/year)

MG Maximum greenhouse gas emission capacity imposed on each WWTP system (gr CO2-eq/year)

ωkt A zero/one variable that equals 1 if a WWTP with technology/scale t is constructed at candidate location k, 0 otherwise

MGD Million gallon per day

139

APPENDIX B: SUPPLEMENTARY MATERIAL FOR CHAPTER 2

Figure B.1 The location of main water and wastewater infrastructure in the city of Lakeland, Florida.

140

Table B.1 Effluent water quality from Glendale water reclamation facility/wetland and water quality standards associated with each reuse scenario. Abbreviations: BOD: biological oxygen demand; CBOD: Carbonaceous Biochemical Oxygen Demand; TRC: total residue chlorine; TSS: total suspended solids; TN: total nitrogen; TOC: total organic carbon; TP: total phosphorus; DO: dissolved oxygen; TDS: total dissolved solids (* Does not meet the water quality standards)

Reuse Type Urban Agricultural Indirect potable Direct potable Distributed Urban Distributed Urban (new plant)

Parameter Standards US EPA

Water quality

Standards US EPA

Water quality

Standards US EPA

Water Qualit

y

Standards US EPA

Water Qualit

y

Standards US EPA

Water Qualit

y

Standards US EPA

Water Quality

Treatment

Secondary treatment-Filtration-Disinfection


Secondary treatment-Filtration-Disinfection -Multiple barriers for pathogen and organics removal (Advanced)

-



pH 6.0 - 9.0 6.88 6.0 - 9.0 6.88 6.5 - 8.5 7.36 6.5 - 8.5 7.36 6.0 - 9.0 6.88 6.0 - 9.0 - BOD 10 mg/l - 10 mg/l - - - 2 mg/l - 10 mg/l - 10 mg/l 303mg/l

CBOD5 20 mg/l 1.82 (total-mg/l) 20 mg/l

1.82 (total-mg/l)

20 mg/l 3.13

(total-mg/l)

3 mg/l 3.13* (total-mg/l)

20 mg/l 1.82

(total-mg/l)

20 mg/l 668 mg/l

(COD)

TRC 1 mg/l (min) - 1 mg/l

(min) - 1 mg/l (min) - 0.2 - 2 mg/l (min) - 1 mg/l (min) - 1 mg/l (min) -

TSS 5 mg/l 4.90 mg/l 5 mg/l 4.90 mg/l 5 mg/l 3.61 mg/l - 3.61

mg/l 5 mg/l 4.90 mg/l 5 mg/l 309

mg/l Turbidity 2 - 2.5 NTU - 2 - 2.5 NTU - 2 - 2.5 NTU - 0.3 NTU - 2 - 2.5 NTU - 2 - 2.5 NTU

Fecal Coliform 25/100 ml (max) 1/100 ml 25/100 ml

(max) 1/100 ml 4/100 ml (total/max) 61/100* ml

0/100 ml (no

detectible)

61/100* ml

25/100 ml (max)

1/100 ml

25/100 ml (max) -

TN - 15.01 mg/l - 15.01 mg/l 10 mg/l 1.54

mg/l

1 mg/l as nitrite and 10 mg/l as

nitrate

1.54* mg/l - 15.01

mg/l - 28.3 mg/l

TOC - - - - 2 - 3 mg/l - 3 mg/l - - - - -

TP - 5.7 mg/l - 5.7 mg/l - 4.1 mg/l - 4.1

mg/l - 5.7 mg/l - 8.6 mg/l

DO - 6.63 mg/l - 6.63 mg/l - 6.64 mg/l - 6.64

mg/l - 6.63 mg/l - 6.3 mg/l

Conductivity - 1331 umhos/cm -

1331 umhos/c

m -

1130 umhos

/cm -

1130 umhos

/cm -

1331 umhos

/cm - -

TDS - 722 mg/l - 722 mg/l - 607 mg/l 500 mg/l 607*

mg/l - 722 mg/l - 3 mg/l

(TSS)

141

Table B.2 Scope of line items included in calculation of capital and O&M costs for each scenario

Urban Agricultural Indirect potable Direct potable Distributed Urban

Distributed Urban (new plant)

Parameter ($/MG)

Capital O&M Capital O&M Capital O&M Capital O&M Capital O&M Capital O&M

Major pipelines material * * * * * *

Construction of major pipelines * * * * * *

Major pumps * * * * * *

Construction of the design * * * * * *

Drilling and construction of injection well

*

UF system construction

*

UV/H2O2 system construction

* *

Additional chlorination system construction

*

Land needed for the additional treatment trains

*

Energy consumption for major pumps

* * * * * *

Labor * * * * * * Overhead & Management

* * * * * *

Energy consumption for the additional treatment

* *

Chemical cost for additional treatments

* *

Additional treatment trains operation and monitoring

* *

142

Figure B.2 Location of golf courses for scenario 1.

Table B.3 Additional information used for design of scenario 1

Requirements for urban reuse (inventory list) Parameter Quantity Unit Major information for the design Total number of golf courses in Florida 1448 Course Average size of a golf course 137 acre Irrigated part of a golf course 65 % Water requirement 0.26 - 0.55 mgd/course Available water 2.83 MGD Water covering capacity 10 Course Nitrogen uptake by the grass 15.38 Kg/acre-year Phosphorus uptake by the grass 4.86 Kg/acre-year

Data used for this design Fixed assessed costs for design Available golf course 10 Course Water demand for available courses ~ 2.83 MGD Average distance to the Glendale WWTP 6.50 Mile Total major pipes length 30.26 Mile Major pipes price (12-3/4” O.D., API, FBE coat) 10.75 $/ft Number of major pipelines/pumps 6 N/A Price for major pumps 62,500 $/major pump Construction of 12” pipeline on land 300,000 $/mile O&M costs for design Energy consumption for major pumps 8,000 KWh/day Electricity rate for industrial use in Florida 0.0804 $/KWh Labor, overhead and management fee 40 $/hr Labor, overhead and management time 8 hr/day

143

Figure B.3 Location of strawberry irrigation lands for scenario 2.


Requirements for strawberry irrigation (inventory list) Parameter Quantity Unit Major information for strawberry irrigation Irrigation repetition 3 1/year Agricultural season period 100 days Distance between plants 14 in Plant populations 19,000 1/acre Water requirement 27155.84 gal/acre-day Nitrogen uptake by the plant 35.85 Kg/acre-year Phosphorus uptake by the plant 8.03 Kg/acre-year

Data used for this design Fixed assessed costs for design Available land ~ 170 acre Water demand for available land ~ 4.6 MGD Average distance to the lands 0.9 Mile Total major pipes length 18405.76 ft Major pipes price (12-3/4” O.D., API, FBE coat) 10.75 $/ft Construction of 12” pipeline on land 300,000 $/mile Number of major pipelines/pumps 4 N/A Price for major pumps 21,000 $/major pump O&M costs for design Energy consumption for major pumps 4,000 KWh/day

144

Figure B.4 Location of injection wells for scenario 3.


Requirements for IPR (inventory list) Parameter Quantity Unit Major information for the design Available water 2.83 MGD

Data used for this design Fixed assessed costs for design Average distance between the wetland and wells 8.32 Mile Total major pipes length 11.68 Mile Major pipes price (24” OD X .500” wall, spiral, new surplus) 38.00 $/ft Number of major pumps 4 N/A Price for major pumps 62,500 $/major pump Construction of 24” pipeline on land 780,000 $/mile Drilling and construction of 750 feet injection well 386,209.12 $/well Total number of injection wells 2 Well UV Disinfection design (4.8 MGD) 845,107.91 $/unit O&M costs for design Energy consumption for major pumps 8,000 KWh/day Electricity rate for industrial use in Florida 0.0804 $/KWh Labor, overhead and management fee 40 $/hr Labor, overhead and management time 16 hr/day Well operation and monitoring 268,106 $/year UV Disinfection operation (2.83 MGD) 19,884.89 $/year

145

Figure B.5 Location of water treatment plant for scenario 4.

Figure B.6 Additional treatment trains designed for scenario 4 using IT3PR toolbox.

146


Requirements for DPR (inventory list) Parameter Quantity Unit Major information for the design Available water 2.83 MGD

Data used for this design Fixed assessed costs for design Average distance between the wetland and the water treatment plant 7.98 Mile Total major pipes length 7.98 Mile Major pipes price (24” OD X .500” wall, spiral, new surplus) 38.00 $/ft Number of major pumps 4 N/A Price for major pumps 62,500 $/major pump Construction of 24” pipeline on land 780,000 $/mile UF system construction for 2.83 MGD 7,075,000 $/unit UV/H2O2 system construction for 2.83 MGD 16,046,500 $/unit Additional chlorination system for 2.83 MGD 804,000 $/unit Land Price for the additional treatment trains 90,000 $/acre Land needed for the additional treatment trains 2 acre O&M costs for design Energy consumption for major pumps 8,000 KWh/day Energy consumption for the additional treatment train (UF) 2,678 KWh/day Energy consumption for the additional treatment train (UV/H2O2) 175.46 KWh/day Energy consumption for the additional treatment train (chlorination) 238.26 KWh/day Electricity rate for industrial use in Florida 0.0804 $/KWh Chemical cost for UF 119.457 $/day Chemical cost for additional chlorination 119.457 $/day Labor, overhead and management fee 40 $/hr Labor, overhead and management time 32 hr/day Additional treatment trains operation and monitoring 94,947.61 $/year

147

Figure B.7 Location and pipeline required for implementation of each scenario. The current constructed pipelines are used for the design of scenarios 5, 6 and 7.


Requirements for urban reuse – Decentralized (inventory list) Parameter Quantity Unit Major information for the design Type of WWTPs Conventional AS N/A

Data used for this design Distribution network material (total) 3005193 ft Total number of households 39,376 Household O&M costs for design Electricity rate for industrial use in Florida 0.0804 $/KWh Labor, overhead and management fee 40 $/hr Labor, overhead and management time 32 hr/day Energy consumption for distribution network 35,635.1336 KWh/day Benefits of the design Reclaimed water rate (Monthly) 21.29 $/properties <= 1 acre Reclaimed water rate (Monthly) 12.20 $/each additional acre

148

Figure B.8 Location of five decentralized medium-scale WWTPs for scenario 7.


Requirements for urban reuse – Decentralized (inventory list) Parameter Quantity Unit Major information for the design Total number of WWTPs 1 WWTP Capacity of each WWTP 3.0 MGD Land need for each plant 6.15 acre Type of WWTPs Conventional AS N/A

Data used for this design Total number of WWTPs 1 WWTP Capacity of the WWTP 3.0 MGD Land needed for the WWTP 6.15 acre Average land purchase price 68,500 $/acre Total fixed assessed cost for the WWTP 32,192,583 $/WWTP Distribution network material (total) 3005193 ft Total number of households 39,376 Household O&M costs for design Energy consumption for the WWTP 5817.86 KWh/day Electricity rate for industrial use in Florida 0.0804 $/KWh Chemical consumption for the WWTP 172028.43 $/year Labor, overhead and management fee 40 $/hr Labor, overhead and management time 75 hr/day Energy consumption for distribution network (WWTP) 19599.448 KWh/day Benefits of the design Reclaimed water rate (Monthly) 21.29 $/properties <= 1 acre Reclaimed water rate (Monthly) 12.20 $/each additional acre

149


Requirements for urban reuse – Decentralized (inventory list) Parameter Quantity Unit Major information for the design Total number of medium scale WWTP 5 WWTP Capacity of each WWTP 0.7 MGD Land need for each plant 1.46 acre Type of WWTPs Conventional AS N/A

Data used for this design Total number of medium scale WWTP 5 WWTP Capacity of each WWTP 0.7 MGD Land needed for each WWTP 1.46 acre Average land purchase price 68,500 $/acre Total fixed assessed cost for each WWTP 20,128,389 $/WWTP Distribution network material (North) 489498 ft Distribution network material (Central East) 744359 ft Distribution network material (Central West) 483646 ft Distribution network material (South East) 959379 ft Distribution network material (South West) 328311 ft Total number of households 39,376 Household O&M costs for design Energy consumption for each WWTP 1452.5 KWh/day Electricity rate for industrial use in Florida 0.0804 $/KWh Chemical consumption for each WWTP 38,228.54 $/year Labor, overhead and management fee 40 $/hr Labor, overhead and management time 32 hr/day Energy consumption for distribution network (each WWTP) 3919.8647 KWh/day Benefits of the design Reclaimed water rate (Monthly) 21.29 $/properties <= 1 acre Reclaimed water rate (Monthly) 12.20 $/each additional acre

Table B.10 Total capital cost, O&M costs and production benefit for scenario 1

Parameter Quantity Unit Unit price Total price Fixed assessed costs for design

Total major pipes length 159772.8 ft $ 10.75 $ 1,717,557.6 Major pumps 6 N/A $ 62,500 $ 375,000 Construction of major pipelines (12”) 3.49 mile $ 300,000 $ 1,047,000

O&M costs for design Energy consumption for major pumps 48,000 KWh/day $ 0.0804 $ 3,859.2 /day Labor, overhead and management 8 hr/day $ 40 $ 320 /day

150

Table B.11 Total capital cost, O&M costs for scenario 2


Major pipeline 18405.76 ft $ 10.75 $ 197,861.92 Major pumps 4 N/A $ 21,000 $ 84,000 Construction of major pipelines 3.49 mile $ 300,000 $ 1,047,000

O&M costs for design Energy consumption for major pumps 16,000 KWh/day $ 0.0804 $ 1,286.4 /day Labor (1 person/20 acre, 8 hours/day) 9.2 hr/day $ 8.10/hr $ 74.52 /day Overhead & Management 1375.59 $/acre $ 1375.59/acre $ 233.850.3 /day



Total major pipes length 61,670.4 ft $ 38.00 $ 2,343,475.2 Major pumps 4 N/A $ 62,500 $ 250,000 Construction of major pipelines (12”) 11.68 mile $ 780,000 $ 9,110,400 Drilling and construction of 750 feet injection well 2 well $ 386,209.12 $ 772,418.24 UV Disinfection design (4.8 MGD) 1 unit $ 845,107.91 $ 845,107.91

O&M costs for design Energy consumption for major pumps 32,486 KWh/day $ 0.0804 $ 2,611.87 /day Energy consumption for the additional treatment train (UV) 175.46 KWh/day $ 0.0804 $ 14.107 /day Labor, overhead and management 16 hr/day $ 40 $ 640 /day Well operation and monitoring 1 $/year $ 268,106 $ 734.54 /day UV Disinfection operation (2.83 MGD) 1 $/year $ 19,884.89 $ 54.48 /day



Total major pipes length 42,134.4 ft $ 38.00 $ 1,601,107.2 Major pumps 4 N/A $ 62,500 $ 250,000 Construction of major pipelines (12”) 7.98 mile $ 780,000 $ 6,224,400 UF system construction for 2.83 MGD 1 unit $ 7,075,000 $ 7,075,000 UV/H2O2 system construction for 2.83 MGD 1 unit $ 16,046,500 $ 16,046,500 Additional chlorination system for 2.83 MGD 1 unit $ 804,000 $ 804,000 Land Price for the additional treatment trains 2 acre $ 90,000 $ 180,000

O&M costs for design Energy consumption for major pumps 31,937 KWh/day $ 0.0804 $ 2,567.73 /day Energy consumption for additional treatment train (UF) 2,264 KWh/day $ 0.0804 $ 182.026 /day Energy consumption for additional treatment train (UV/H2O2) 175.46 KWh/day $ 0.0804 $ 14.107 /day Energy consumption for the additional treatment train 238.26 KWh/day $ 0.0804 $ 19.156 /day Labor, overhead and management 32 hr/day $ 40 $ 1280 /day Chemical cost for UF N/A 1/day $ 119.457 $ 119.457 /day Chemical cost for additional chlorination N/A 1/day $ 119.457 $ 119.457 /day Additional treatment trains operation and monitoring 1 1/year $ 3,588,825 $ 260.130 /day

151



Distribution network material (total) 3005193 ft N/A $239,015,861.50 Distribution network Construction (total) 489498 ft N/A $406,300,686.90 Major pumps 7 N/A $ 62,500 $ 437,500

O&M costs for design Labor, overhead and management for the network 80 hr/day $ 40 $ 6,400 /day Energy consumption for distribution network 35,635.1336 KWh/day $ 0.0804 $ 2,865.06 /day



Total fixed assessed cost for WWTP 1 WWTP $ 32,192,583 $ 32,192,583 Distribution network material (total) 3005193 ft N/A $239,015,861.50 Distribution network Construction (total) 489498 ft N/A $406,300,686.90 Major pumps 7 N/A $ 62,500 $ 437,500

O&M costs for design Energy consumption for WWTP 5817.86 KWh/day $ 0.0804 $ 467.756 /day Chemical consumption for WWTP 1 172028.73 $/day $ 471.311 /day Labor, overhead and management for WWTP 75 hr/day $ 40 $ 3000 /day Energy consumption for distribution network 35,635.1336 KWh/day $ 0.0804 $ 2,865.06 /day



Total fixed assessed cost for WWTPs 5 WWTP $ 20,128,389 $ 100,641,945 Distribution network material cost (North) 489498 ft N/A $ 38,931,870.99 Distribution network material cost (Central East) 744359 ft N/A $ 59,202,057.12 Distribution network material cost (Central West) 483646 ft N/A $ 38,466,436.38 Distribution network material cost (South East) 959379 ft N/A $ 76,303,518.01 Distribution network material cost (South West) 328311 ft N/A $ 26,111,979 Distribution network Construction (North) 489498 ft N/A $ 66,179,900.47 Distribution network Construction (Central East) 744359 ft N/A $ 100,636,988.4 Distribution network Construction (Central West) 483646 ft N/A $ 65,388,712.81 Distribution network Construction (South East) 959379 ft N/A $ 129,707,591.7 Distribution network Construction (South West) 328311 ft N/A $ 44,387,493.52 Major pumps 5 N/A $ 62,500 $ 312,500

O&M costs for design Energy consumption for WWTPs 7,262.5 KWh/day $ 0.0804 $ 583.905 /day Chemical consumption for WWTPs 5 104.736 $/WWTP/day $ 523.679 /day Labor, overhead and management for WWTPs 160 hr/day $ 40 $ 6,400 /day Energy consumption for distribution network 19,599.448 KWh/day $ 0.0804 $ 1,575.7956 /day

152

𝐴𝐴𝐴𝐴 = 𝐶𝐶𝑂𝑂&𝑀𝑀 1(1+i)𝑛𝑛

(B.1)

𝐴𝐴𝐴𝐴𝐴𝐴 = 𝐴𝐴𝐴𝐴𝐴𝐴Capital + ∑ 𝐴𝐴𝐴𝐴𝐴𝐴O&M331 (B.2)

𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 = 𝑖𝑖 (𝑁𝑁𝑁𝑁𝑁𝑁)

1− 1(1+i)𝐿𝐿

(B.3)

𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 = 𝑁𝑁𝑁𝑁𝑁𝑁1𝑇𝑇𝑃𝑃

∫ 𝑁𝑁𝑡𝑡.𝑑𝑑𝑑𝑑𝑇𝑇𝑃𝑃0

(B.4)

where PV = present value; CO&M = O&M cost; NPV = net present value; ANPV = annualized net

present value; ASNPV = annualized specific net present value; i = annual discount rate; n = each

O&M year; TP = planning horizon; Pt = demand at time t; and L = number of years in entire lifetime

of the design.

153

Table B.17 Results for calculation of economic indicators for each reuse scenario. Abbreviations: PV: present value; NPV: net present value; ANPV: annualized net present value; ASNPV: annualized specific net present value; i: annual discount rate; n: each O&M year

Scenario 1 Scenario 2 Scenario 3 Scenario 4 Scenario 5 Scenario 6 Scenario 7 Capital ($) 3,139,558 1,328,862 13,321,401 32,181,007 645,754,048 677,946,631 746,270,99

3 O&M ($/day) 4,179 1,671 4,002 4,567 9,265 6,804 9,083

NPV ($) 27,549,974 11,090,225 36,695,754 58,857,313 699,870,620 717,689,064 799,326,384 ANPV ($) 1,721,599 693,029 2,293,119 3,677,996 43,734,946 44,848,421 49,949,941

ASNPV ($/MG) 1,667 413 2,220 3,561 42,340 43,418 48,357

n i = 0.05 PV PV PV PV PV PV PV 1 0.952380952 1,452,769.5 580,941.0 1,391,108.9 1,587,622.4 3,220,711.3 2,365,244.1 3,157,555.

9 2 0.907029478 1,383,590.0 553,277.1 1,324,865.6 1,512,021.4 3,067,344.1 2,252,613.5 3,007,196.1 3 0.863837599 1,317,704.8 526,930.6 1,261,776.7 1,440,020.3 2,921,280.1 2,145,346.2 2,863,996.3 4 0.822702475 1,254,956.9 501,838.6 1,201,692.1 1,371,447.9 2,782,171.5 2,043,186.8 2,727,615.5 5 0.783526166 1,195,197.1 477,941.6 1,144,468.7 1,306,140.9 2,649,687.2 1,945,892.2 2,597,729.1 6 0.746215397 1,138,282.9 455,182.4 1,089,970.2 1,243,943.7 2,523,511.6 1,853,230.7 2,474,027.7 7 0.71068133 1,084,079.0 433,507.1 1,038,066.8 1,184,708.3 2,403,344.4 1,764,981.6 2,356,216.8 8 0.676839362 1,032,456.2 412,863.9 988,635.1 1,128,293.6 2,288,899.4 1,680,934.9 2,244,016.0 9 0.644608916 983,291.6 393,203.7 941,557.2 1,074,565.3 2,179,904.2 1,600,890.3 2,137,158.1 10 0.613913254 936,468.2 374,479.7 896,721.2 1,023,395.6 2,076,099.2 1,524,657.5 2,035,388.7 11 0.584679289 891,874.5 356,647.4 854,020.2 974,662.4 1,977,237.4 1,452,054.7 1,938,465.4 12 0.556837418 849,404.3 339,664.1 813,352.5 928,250.0 1,883,083.2 1,382,909.3 1,846,157.5 13 0.530321351 808,956.4 323,489.7 774,621.5 884,047.6 1,793,412.6 1,317,056.4 1,758,245.3 14 0.505067953 770,434.7 308,085.4 737,734.7 841,950.1 1,708,012.0 1,254,339.5 1,674,519.3 15 0.481017098 733,747.3 293,414.7 702,604.5 801,857.2 1,626,678.1 1,194,609.0 1,594,780.3 16 0.458111522 698,807.0 279,442.5 669,147.1 763,673.5 1,549,217.2 1,137,722.9 1,518,838.4 17 0.436296688 665,530.5 266,135.7 637,283.0 727,308.1 1,475,445.0 1,083,545.6 1,446,512.7 18 0.415520655 633,838.5 253,462.6 606,936.2 692,674.4 1,405,185.7 1,031,948.2 1,377,631.2 19 0.395733957 603,655.7 241,393.0 578,034.5 659,689.9 1,338,272.1 982,807.8 1,312,029.7 20 0.376889483 574,910.2 229,898.1 550,509.0 628,276.1 1,274,544.8 936,007.4 1,249,552.1 21 0.358942365 547,533.6 218,950.5 524,294.3 598,358.2 1,213,852.2 891,435.6 1,190,049.6 22 0.341849871 521,460.5 208,524.3 499,327.9 569,864.9 1,156,049.7 848,986.3 1,133,380.6 23 0.325571306 496,629.1 198,594.6 475,550.4 542,728.5 1,100,999.8 808,558.4 1,079,410.1 24 0.31006791 472,980.1 189,137.7 452,905.1 516,884.3 1,048,571.2 770,055.6 1,028,009.6 25 0.295302772 450,457.2 180,131.1 431,338.2 492,270.8 998,639.2 733,386.3 979,056.8

26 0.281240735 429,006.9 171,553.5 410,798.3 468,829.3 951,085.0 698,463.2 932,435.0

27 0.267848319 408,578.0 163,384.3 391,236.5 446,504.1 905,795.2 665,203.0 888,033.3

28 0.255093637 389,121.9 155,604.1 372,606.2 425,242.0 862,662.1 633,526.7 845,746.0

29 0.242946321 370,592.3 148,194.3 354,863.0 404,992.4 821,583.0 603,358.7 805,472.4

30 0.231377449 352,945.0 141,137.5 337,964.8 385,707.0 782,460.0 574,627.4 767,116.6

31 0.220359475 336,138.1 134,416.6 321,871.2 367,340.0 745,200.0 547,264.2 730,587.2

32 0.209866167 320,131.5 128,015.8 306,544.0 349,847.6 709,714.3 521,204.0 695,797.4

33 0.19987254 304,887.2 121,919.9 291,946.7 333,188.2 675,918.3 496,384.7 662,664.2

154

APPENDIX C: SUPPLEMENTARY MATERIAL FOR CHAPTER 3

Figure C.1 Hillsborough County water service areas, current WWTPs location, and candidate locations for the new water reclamation facilities.

155

Table C.1 Information regarding the population clusters in the study area

Cluster (r) Name Population (2017)

Population Growth Rate

(2017)

Projected Population in 33 Years

Latitude Longitude

1 Apollo Beach 20,149 0.0509 103,700 27.7630 -82.4033

2 Balm 3,968 0.1482 379,447 27.7530 -82.2895

3 Bloomingdale 24,100 0.0082 31,554 27.8767 -82.2614

4 Brandon 113,677 0.0130 174,094 27.9360 -82.2993

5 Dover 4,149 0.0158 6,960 27.9868 -82.2311

6 East Lake-Orient Park CDP 25,565 0.0162 43,447 28.0044 -82.3647

7 Fish Hawk 20,930 0.0561 126,776 27.8479 -82.2178

8 Gibsonton 16,587 0.0213 33,253 27.8285 -82.3791

9 Mango 12,142 0.0098 16,752 27.9901 -82.3083

10 Palm River-Clair Mel CDP 23,848 0.0175 42,275 27.9245 -82.3794

11 Plant City 39,087 0.0163 66,643 28.0144 -82.1203

12 Progress Village 8,220 0.0599 56,055 27.8856 -82.3643

13 Riverview 89,017 0.0316 248,512 27.8252 -82.3046

14 Ruskin 24,266 0.0485 115,813 27.7077 -82.4227

15 Seffner 8,297 0.0126 12,542 28.0016 -82.2741

16 Sun City Center 24,900 0.0361 80,252 27.7144 -82.3558

17 Thonotosassa 13,964 0.0142 22,238 28.0423 -82.2962

18 Valrico 38,711 0.0118 57,011 27.9191 -82.2313

19 Wimauma 8,615 0.0425 34,022 27.6749 -82.3132

Table C.2 Information related to the candidate locations for the new wastewater treatment facility

Location's type

k Location's name Capacity (MGD)

Latitude Longitude

Current WWTP

1 Falkenburg 12 27.952972 -82.340869

2 Valrico 12 27.957529 -82.229382

3 South County 10 27.722524 -82.383904

New candidate location

4 Southeast County Landfill N/A 27.762715 -82.198667

5 Mosaic (Hillsborough) N/A 27.863415 -82.386560

6 South County Solid Waste Facility N/A 27.801543 -82.382078

7 14519 BALM RIVERVIEW RD, RIVERVIEW

N/A 27.773225 -82.279631

8 15110 BALM WIMAUMA RD, WIMAUMA N/A 27.756540 -82.256282

9 16410 BALM WIMAUMA RD, WIMAUMA N/A 27.723188 -82.261762

156

Table C.3 Information related to the types of treatment technology used in the model

Treatment process Pretreatment Primary

treatment Secondary treatment Tertiary Treatment

Bardenpho Screening Grit removal N/A

Dissolved air bioreactor

Secondary clarifier Filtration UV

Disinfection

MBR+MF Screening Grit removal

Primary clarifier

Membrane bioreactor N/A Micro filtration UV

Disinfection

MBR+UF Screening Grit removal

Primary clarifier

Membrane bioreactor N/A Ultra filtration UV

Disinfection

MBR+RO Screening Grit removal

Primary clarifier

Membrane bioreactor N/A Reverse osmosis UV

Disinfection

CAS+GAC Screening Grit removal

Primary clarifier

Aeration basin

Secondary clarifier

Granular activated carbon/Filtration

UV Disinfection

Table C.4 Input information related to the treatment technologies selected for the model

Technology Scale t Capkt (m3/year)

Energy consumption (KWh/m3)

CCkt ($/(m3/year))

OCt ($/m3)

SRWt ($/m3)

GPt (kg CO2-eq/m3)

Bardenpho

Small 1 4,745 4.668 58.4398 1.9203 0.0457 3.30

Medium 2 5,584,865 1.528 8.5201 0.5880 0.1321 1.08

Large 3 282,801,270 0.552 5.4282 0.5283 0.4140 0.39

MBR+MF

Small 4 4,745 2.475 65.0096 0.8331 0.0457 1.75

Medium 5 5,584,865 1.881 9.7224 0.2562 0.1321 1.33

Large 6 282,801,270 1.768 4.4855 0.1870 0.4140 1.25

MBR+UF

Small 7 4,745 4.648 76.5810 1.8901 0.8321 3.29

Medium 8 5,584,865 4.054 12.2471 0.5787 0.9246 2.87

Large 9 282,801,270 3.941 8.2035 0.5200 0.9246 2.79

MBR+RO

Small 10 4,745 5.417 175.1432 4.2000 0.8321 3.83

Medium 11 5,584,865 4.116 25.5347 1.0642 0.9246 2.91

Large 12 282,801,270 4.017 16.2682 0.6872 0.9246 2.84

CAS+GAC

Small 13 4,745 3.793 19.8931 2.8174 0.0457 33.52

Medium 14 5,584,865 3.758 1.9930 0.2273 0.1321 33.21

Large 15 282,801,270 3.724 1.8250 0.0122 0.3963 32.91

157

Table C.5 Other input parameters used for the optimization model

Parameter Unit Value

μ m3/capita-year 118.824

GTW kg CO2-eq/mi-m3 0.042556

GTR kg CO2-eq/mi-m3 0.032735

MG kg CO2-eq/year 1.00E+50

CTW $/mi-m3 0.1736

CTR $/mi-m3 0.1389

Wastewater collection KWh/mi-m3 0.083

Reclaimed water distribution KWh/mi-m3 0.064

α Ratio 0.9

n Year 33

The wastewater system design problem for the study area had two 19×9×1 integer variables

(AWTrk and ARTrk), one 19×9×1 binary variable (qrk), and one 1×9×15 binary variable (ωkt), for a

total of 648 variables. The model had 11 constraints, in which three of them were 19×9×1

constraints, two others were 1×9×1 constraints, one was 9×1×1 constraint, and five were 1×1×1

constraints, for a total of 545 constraints. The McCormick envelope required adding a new variable

and three additional constraints for the defined variable. The new variable was 19×9×15 (a total

of 2,565 additional variables).

158

Table C.6 Results obtained from solving the first stage of the multi-objective optimization model

Optimal solution Location Location Technology Servicing

Clusters

Receiving Capacity (MGD)

Sending Capacity (MGD)

Solution_1

1 Falkenburg CAS+GAC 6, 9, 10, 17 10.7 9.7

2 Valrico CAS+GAC 3, 4, 5, 11, 15, 18 30.0 27.0

3 South County CAS+GAC 14 10.0 9.0

4 Southeast County Landfill CAS+GAC 7 10.9 9.8

5 Mosaic (Hillsborough) CAS+GAC 12 4.8 4.3

6 South County Solid Waste Facility CAS+GAC 1, 8 11.8 10.6

7 14519 BALM RIVERVIEW RD, RIVERVIEW CAS+GAC 2, 13 54.0 48.6

8 15110 BALM WIMAUMA RD, WIMAUMA Not Installed - 0.0 0.0

9 16410 BALM WIMAUMA RD, WIMAUMA CAS+GAC 16, 19 9.8 8.8

Solution_2


2 Valrico CAS+GAC 3, 4, 5, 11, 15, 18 30.0 27.0



5 Mosaic (Hillsborough) Bardenpho 12 4.8 4.3





Solution_3


2 Valrico CAS+GAC 3, 4, 5, 11, 15, 18 30.0 27.0



5 Mosaic (Hillsborough) MBR+MF 12 4.8 4.3

6 South County Solid Waste Facility MBR+MF 1, 8 11.8 10.6




Solution_4


2 Valrico CAS+GAC 3, 4, 5, 11, 15, 18 30.0 27.0

3 South County MBR+MF 14 10.0 9.0






9 16410 BALM WIMAUMA RD, WIMAUMA MBR+MF 16, 19 9.8 8.8

159

Table C.6 (Continued)

Solution_5


2 Valrico CAS+GAC 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_6


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_7

1 Falkenburg MBR+MF 6, 9, 10, 17 10.7 9.7

2 Valrico CAS+GAC 3, 4, 5, 11, 15, 18 30.0 27.0


4 Southeast County Landfill MBR+MF 7 10.9 9.8






Solution_8


2 Valrico CAS+GAC 3, 4, 5, 11, 15, 18 30.0 27.0








160


Solution_9


2 Valrico CAS+GAC 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_10


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_11


2 Valrico CAS+GAC 3, 4, 5, 11, 15, 18 30.0 27.0





7 14519 BALM RIVERVIEW RD, RIVERVIEW MBR+MF 2, 13 54.0 48.6



Solution_12


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








161


Solution_13


2 Valrico CAS+GAC 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_14


2 Valrico CAS+GAC 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_15


2 Valrico CAS+GAC 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_16


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








162


Solution_17


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_18


2 Valrico CAS+GAC 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_19


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_20


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








163


Solution_21


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_22


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_23


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_24


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








164


Solution_25

1 Falkenburg Bardenpho 6, 9, 10, 17 10.7 9.7

2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_26


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_27


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0


4 Southeast County Landfill Bardenpho 7 10.9 9.8


6 South County Solid Waste Facility Bardenpho 1, 8 11.8 10.6




Solution_28


2 Valrico Bardenpho 3, 4, 5, 11, 15, 18 30.0 27.0








165


Solution_29










Solution_30


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0

3 South County Bardenpho 14 10.0 9.0






9 16410 BALM WIMAUMA RD, WIMAUMA Bardenpho 16, 19 9.8 8.8

Solution_31


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0





7 14519 BALM RIVERVIEW RD, RIVERVIEW Bardenpho 2, 13 54.0 48.6



Solution_32


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








166


Solution_33


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_34


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_35


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_36


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








167


Solution_37


2 Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








Solution_38










Solution_39










Solution_40










168


Solution_41










Solution_42







7 14519 BALM RIVERVIEW RD, RIVERVIEW Bardenpho 2, 13, 16 60.9 54.8


9 16410 BALM WIMAUMA RD, WIMAUMA Bardenpho 19 2.9 2.6

Solution_43










Solution_44

1 Falkenburg Bardenpho 6 3.7 3.4

2 Valrico Bardenpho 3, 4, 5, 9, 11, 15, 17, 18 33.4 30.0



5 Mosaic (Hillsborough) Bardenpho 10, 12 8.5 7.6





169


Solution_45


2 Valrico Bardenpho 4, 5, 11, 15, 18 27.3 24.6








Solution_46


2 Valrico Bardenpho 4, 5, 11, 15, 18 27.3 24.6








Solution_47


2 Valrico Bardenpho 3, 4, 5, 9, 11, 15, 17, 18 33.4 30.0








Solution_48


2 Valrico Bardenpho 4, 5, 9, 11, 15, 17, 18 30.6 27.6








170


Solution_49






6 South County Solid Waste Facility Bardenpho 1, 8, 12 16.6 14.9




Solution_50







7 14519 BALM RIVERVIEW RD, RIVERVIEW Bardenpho 2, 7, 13, 16 71.8 64.6



171

Table C.7 Economic and environmental footprints associated with the optimal solutions

Optimal solution

ZC (M$/year)

ZE (metric ton CO2-eq/year)

Optimal solution

ZC (M$/year)


Sol_1 153.5 6,628.29 Sol_26 204.1 292.32

Sol_2 156.2 6,412.20 Sol_27 207.0 289.51

Sol_3 159.0 5,888.17 Sol_28 211.3 285.27

Sol_4 160.0 5,745.99 Sol_29 211.3 285.27

Sol_5 162.3 5,444.07 Sol_30 214.2 282.35

Sol_6 165.0 5,075.80 Sol_31 219.7 276.87

Sol_7 167.8 4,694.66 Sol_32 222.3 274.31

Sol_8 171.1 4,256.46 Sol_33 222.3 274.31

Sol_9 171.1 4,256.46 Sol_34 224.7 271.97

Sol_10 173.8 3,882.28 Sol_35 224.7 271.97

Sol_11 176.8 3,480.18 Sol_36 227.0 269.66

Sol_12 177.4 3,397.14 Sol_37 227.0 269.66

Sol_13 180.4 2,994.04 Sol_38 229.2 267.44

Sol_14 180.4 2,994.04 Sol_39 231.8 264.92

Sol_15 183.1 2,636.90 Sol_40 231.8 264.92

Sol_16 186.5 2,181.66 Sol_41 234.1 262.61

Sol_17 186.5 2,181.66 Sol_42 240.8 262.44

Sol_18 188.9 1,848.46 Sol_43 243.2 260.10

Sol_19 189.4 1,785.50 Sol_44 256.9 259.85

Sol_20 193.9 1,178.27 Sol_45 260.3 259.29

Sol_21 195.3 989.15 Sol_46 261.4 258.16

Sol_22 198.4 733.04 Sol_47 266.0 257.33

Sol_23 198.9 510.92 Sol_48 284.2 255.39

Sol_24 200.5 295.97 Sol_49 291.1 253.78

Sol_25 204.1 292.32 Sol_50 298.6 253.64

172

Table C.8 The detailed information regarding the least expensive solution (solution 1), the most environmentally friendly solution (solution 50), and the solution with the highest marginal benefit (solution 24)

Optimal Solution Loc. Tech. Scale

Receiving Capacity (MGD)

Sending Capacity (MGD)

NS2 Servicing Clusters

Sustainability Impacts

ZC (M$/year)


Solution 1 1 CAS +GAC Large 10.7 9.7 0.01343 6, 9, 10, 17 153.5 6,628.3 2 CAS +GAC Large 30.0 27.0 3, 4, 5, 11,

15, 18 3 CAS +GAC Large 10.0 9.0 14 4 CAS +GAC Large 10.9 9.8 7 5 CAS +GAC Large 4.8 4.3 12 6 CAS +GAC Large 11.8 10.6 1, 8 7 CAS +GAC Large 54.0 48.6 2, 13 8 Not Installed N/A 0.0 0.0 - 9 CAS +GAC Large 9.8 8.8 16, 19

Solution_50 1 Bardenpho Medium 3.7 3.4 0.02599 6 298.6 253.6 2 Bardenpho Large 30.6 27.6 4, 5, 9, 11,

15, 17, 18 3 Bardenpho Large 10.0 9.0 14 4 Bardenpho Medium 2.7 2.4 3 5 Bardenpho Medium 3.6 3.3 10 6 Bardenpho Large 13.7 12.4 1, 12 7 Bardenpho Large 71.8 64.6 2, 7, 13, 16 8 Bardenpho Medium 2.9 2.6 8 9 Bardenpho Medium 2.9 2.6 19

Solution 24 1 MBR+MF Large 10.7 9.7 0.01343 6, 9, 10, 17 200.5 296.0 2 MBR+MF Large 30.0 27.0 3, 4, 5, 11,

15, 18 3 MBR+MF Large 10.0 9.0 14 4 MBR+MF Large 10.9 9.8 7 5 MBR+MF Large 4.8 4.3 12 6 MBR+MF Large 11.8 10.6 1, 8 7 MBR+MF Large 54.0 48.6 2, 13 8 Not Installed N/A 0.0 0.0 - 9 MBR+MF Large 9.8 8.8 16, 19

Figure C.2 Environmental marginal benefits associated with the optimal solutions.

173

Table C.9 Results obtained from the sensitivity analysis

Scen.

Solution with

highest marginal benefit

Candidate location Treatment technology Servicing clusters

Receiving Cap.

(MGD)

Sending Cap.

(MGD)

Base Sol_24

Falkenburg MBR+MF 6, 9, 10, 17 10.7 9.7

Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0

South County MBR+MF 14 10.0 9.0

Southeast County Landfill MBR+MF 7 10.9 9.8

Mosaic (Hillsborough) MBR+MF 12 4.8 4.3

South County Solid Waste Facility MBR+MF 1, 8 11.8 10.6

14519 BALM RIVERVIEW RD, RIVERVIEW MBR+MF 2, 13 54.0 48.6

15110 BALM WIMAUMA RD, WIMAUMA Not Installed - 0.0 0.0

16410 BALM WIMAUMA RD, WIMAUMA MBR+MF 16, 19 9.8 8.8

CG+10 Sol_22


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0



Mosaic (Hillsborough) CAS+GAC 12 4.8 4.3





CG+20 Sol_22


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








CG+30 Sol_23


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








174


CG+40 Sol_20

Falkenburg CAS+GAC 6, 9, 10, 17 10.7 9.7

Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








CG+50 Sol_21


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








CG+60 Sol_20


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








CG+70 Sol_22


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








175


CG+80 Sol_3


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0





14519 BALM RIVERVIEW RD, RIVERVIEW CAS+GAC 2, 13 54.0 48.6



CG+90 Sol_23


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








CG+100 Sol_21


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








CG+110 Sol_23


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








176


CG+120 Sol_24


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








CG+130 Sol_21


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








CG+140 Sol_24


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








CG+150 Sol_21


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








177


CG+160 Sol_21


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








CG+170 Sol_21


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








CG+180 Sol_22


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








CG+190 Sol_21


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








178


CG+200 Sol_24


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








P-90 Sol_25

Falkenburg MBR+MF 4, 6, 9, 10, 17 2.6 2.3

Valrico MBR+MF 3, 5, 11, 15, 18 1.5 1.4

South County MBR+MF 14, 16 1.7 1.5






16410 BALM WIMAUMA RD, WIMAUMA MBR+MF 19 0.3 0.3

P-80 Sol_24


Valrico MBR+MF 3, 5, 11, 15, 18 3.0 2.7








P-70 Sol_24


Valrico MBR+MF 3, 5, 11, 15, 18 4.5 4.1








179


P-60 Sol_24


Valrico MBR+MF 3, 5, 11, 15, 18 6.0 5.4








P-50 Sol_16


Valrico MBR+MF 3, 5, 11, 15, 17, 18 8.5 7.6








P-40 Sol_19

Falkenburg MBR+MF 4, 6 11.2 10.1

Valrico MBR+MF 3, 5, 9, 11, 15, 17, 18 11.0 9.9



Mosaic (Hillsborough) MBR+MF 10, 12 5.1 4.6





P-30 Sol_23


Valrico MBR+MF 3, 5, 6, 11, 15, 17, 18 14.5 13.0








180


P-20 Sol_23


Valrico MBR+MF 3, 4, 5, 11, 15, 18 24.0 21.6








P-10 Sol_23


Valrico MBR+MF 3, 4, 5, 11, 15, 18 27.0 24.3








Base Sol_24


Valrico MBR+MF 3, 4, 5, 11, 15, 18 30.0 27.0








P+10 Sol_23


Valrico MBR+MF 3, 4, 5, 11, 15, 18 33.0 29.7

South County CAS+GAC 16 7.6 6.8



South County Solid Waste Facility MBR+MF 1, 8, 14 23.9 21.5




181


P+20 Sol_24

Falkenburg MBR+MF 6, 9, 10 10.6 9.5

Valrico MBR+MF 3, 4, 5, 11, 15, 17, 18 38.3 34.5







16410 BALM WIMAUMA RD, WIMAUMA CAS+GAC 19 3.5 3.2

P+30 Sol_25

Falkenburg MBR+MF 6, 9, 10 11.5 10.3

Valrico MBR+MF 3, 4, 5, 11, 15, 17, 18 41.5 37.3








P+40 Sol_23


Valrico MBR+MF 3, 4, 5, 9, 11, 15, 17, 18 46.7 42.0








P+50 Sol_25


Valrico MBR+MF 3, 4, 5, 9, 11, 15, 17, 18 50.0 45.0

South County Not Installed - 0.0 0.0







182


P+60 Sol_25


Valrico MBR+MF 3, 4, 5, 9, 11, 15, 17, 18 53.4 48.0








P+70 Sol_23

Falkenburg CAS+GAC 6, 9 8.8 7.9

Valrico MBR+MF 3, 4, 5, 11, 15, 17, 18 54.2 48.8








P+80 Sol_23


Valrico MBR+MF 3, 4, 5, 11, 15, 17, 18 57.4 51.7








P+90 Sol_23


Valrico MBR+MF 3, 4, 5, 11, 15, 17, 18 60.6 54.6








183


P+100 Sol_23


Valrico MBR+MF 3, 4, 5, 11, 15, 17, 18 63.8 57.4








184

APPENDIX D: SUPPLEMENTARY MATERIAL FOR CHAPTER 4

Table D.1 Information related to the types of treatment technology used in the model

Treatment process Pretreatment Primary

treatment Secondary treatment Tertiary Treatment

Bardenpho Screening Grit removal N/A Dissolved air

bioreactor Secondary clarifier Filtration UV

Disinfection

MBR+MF Screening Grit removal

Primary clarifier

Membrane bioreactor N/A Micro filtration UV

Disinfection

MBR+UF Screening Grit removal

Primary clarifier

Membrane bioreactor N/A Ultra filtration UV

Disinfection

MBR+RO Screening Grit removal

Primary clarifier

Membrane bioreactor N/A Reverse osmosis UV

Disinfection RBC+TF Screening Grit

removal Primary clarifier

Rotating biological contactor

Secondary clarifier

Tertiary filtration Chlorination

Table D.2 Input information related to the treatment technologies selected for the model

Technology Scale t Capkt (m3/year)

Energy consumption (KWh/m3)

CCkt ($/(m3/year))

OCt ($/m3)

SRWt ($/m3)

GPt (kg CO2-eq/m3)

Bardenpho

Small 1 12,000 29.220 1.920 0.046 3.300 12,000

Medium 2 6,000,000 8.520 0.588 0.132 1.080 6,000,000

Large 3 107,000,000 4.343 0.528 0.414 0.390 107,000,000

MBR+MF

Small 4 12,000 32.505 0.833 0.046 1.750 12,000

Medium 5 6,000,000 9.722 0.256 0.132 1.330 6,000,000

Large 6 107,000,000 3.588 0.187 0.414 1.250 107,000,000

MBR+UF

Small 7 12,000 38.291 1.890 0.832 3.290 12,000

Medium 8 6,000,000 12.247 0.579 0.925 2.870 6,000,000

Large 9 107,000,000 6.563 0.520 0.925 2.790 107,000,000

MBR+RO

Small 10 12,000 87.572 4.200 0.832 3.830 12,000

Medium 11 6,000,000 25.535 1.064 0.925 2.910 6,000,000

Large 12 107,000,000 13.015 0.687 0.925 2.840 107,000,000

CAS+GAC

Small 13 12,000 35.517 0.811 0.046 2.012 12,000

Medium 14 6,000,000 10.203 0.231 0.132 1.549 6,000,000

Large 15 107,000,000 4.823 0.132 0.414 1.322 107,000,000

185

Table D.3 Other input parameters used for the optimization model

Parameter Unit Value

μ m3/capita-year 118.824

MG kg CO2-eq/year 1.00E+50

GTW (flat) kg CO2-eq/mi-m3 0.042556

GTR (flat) kg CO2-eq/mi-m3 0.032735

CTW (flat) $/mi-m3 0.1736

CTR (flat) $/mi-m3 0.1389

Wastewater collection (flat) KWh/mi-m3 0.083

Reclaimed water distribution (flat) KWh/mi-m3 0.064 GTW (medium elevation variation) kg CO2-eq/mi-m3

0.82686308 GTR (medium elevation variation) kg CO2-eq/mi-m3

0.63604105 CTW (medium elevation variation) $/mi-m3

3.373048 CTR (medium elevation variation) $/mi-m3

2.698827 Wastewater collection (medium elevation variation) KWh/mi-m3 1.61269 Reclaimed water distribution (medium elevation variation) KWh/mi-m3 1.24352 GTW (hilly topography) kg CO2-eq/mi-m3

3.88238388 GTR (hilly topography) kg CO2-eq/mi-m3

2.98641405 CTW (hilly topography) $/mi-m3

15.837528 CTR (hilly topography) $/mi-m3

12.671847 Wastewater collection (hilly topography) KWh/mi-m3 7.57209 Reclaimed water distribution (hilly topography) KWh/mi-m3 5.83872 α Ratio 0.9

n Year 33

CPA $/mi/(m3/year) 0.0224

CPB $/mi 919,267.78

186

Table D.4 Results of the optimization model for scenario Low P Low E

Solution Location Technology Scale Servicing Clusters

Receiving Capacity (m3/day)

Sending Capacity (m3/day)

ZC (M$/year)

ZE (ton CO2/year)

Marginal

1 MBR+MF Medium 1, 7, 8 98 88

3.5645982 392.9913789

2 MBR+MF Small 2 33 29

3 MBR+MF Medium 3, 4, 9 98 88

4 Bardenpho Medium 5, 11 65 59

5 Bardenpho Small 6 33 29

6 - - - - -

7 - - - - -

8 - - - - -


10 MBR+MF Medium 12, 17, 18 98 88





15 MBR+MF Medium 23, 24 65 59


17 Bardenpho Medium 21, 26, 27 98 88


19 - - - - -



22 MBR+MF Medium 32, 33 65 59


24 MBR+MF Medium 29, 34, 35 98 88


187

Table D.5 Results of the optimization model for scenario Medium P Low E




ZC (M$/year)

ZE (ton CO2/year)

Marginal

1 MBR+MF Medium 1, 7 1,953 1,758

106.92963 12096.25643

2 MBR+MF Medium 2 977 879

3 MBR+MF Medium 3, 10 1,953 1,758

4 MBR+MF Medium 4, 5, 11 2,930 2,637

5 Bardenpho Medium 6, 12 1,953 1,758

6 Bardenpho Medium 8, 13, 14 2,930 2,637

7 - - - - -

8 MBR+MF Medium 9, 16 1,953 1,758

9 Bardenpho Medium 17 977 879





14 MBR+MF Medium 22, 23 1,953 1,758

15 - - - - -

16 - - - - -

17 - - - - -



20 MBR+MF Medium 24, 29 1,953 1,758


22 MBR+MF Medium 26, 32, 33 2,930 2,637



25 MBR+MF Medium 30, 36 1,953 1,758

188

Table D.6 Results of the optimization model for scenario High P Low E




ZC (M$/year)

ZE (ton CO2/year)

Marginal

1 Bardenpho Large 1, 2, 7 24,416 21,974

891.28392 93138.00646

2 Bardenpho Medium 3 8,139 7,325

3 MBR+MF Medium 4, 9 16,277 14,650


5 Bardenpho Large 5, 6, 12 24,416 21,974

6 MBR+MF Medium 13 8,139 7,325



9 MBR+MF Medium 11, 17 16,277 14,650


11 MBR+MF Medium 14 8,139 7,325

12 MBR+MF Medium 20 8,139 7,325

13 - - - -

14 MBR+MF Medium 23 8,139 7,325

15 - - - -

16 MBR+MF Medium 19, 25 16,277 14,650

17 - - - -


19 - - - -

20 MBR+MF Large 24, 29, 30 24,416 21,974


22 MBR+MF Large 26, 32, 33 24,416 21,974


24 MBR+MF Medium 28, 35 16,277 14,650


189

Table D.7 Results of the optimization model for scenario Low P Medium E




ZC (M$/year)

ZE (ton CO2/year)

Marginal

1 MBR+MF Medium 1, 7 65 59

5.2332999 651.4604706

2 MBR+MF Medium 2, 9 65 59











13 - - - - -

14 - - - - -




18 - - - - -

19 Bardenpho Large 22 33 29







190

Table D.8 Results of the optimization model for scenario Medium P Medium E




ZC (M$/year)

ZE (ton CO2/year)

Marginal


157.04462 17703.63531

2 - - - - -


4 - - - - -







11 - - - - -



14 - - - - -






20 - - - - -


22 - - - - -


24 - - - - -


191

Table D.9 Results of the optimization model for scenario High P Medium E




ZC (M$/year)

ZE (ton CO2/year)

Marginal

1 Bardenpho Medium 1 8138.63014 7324.7671

1308.705 147530.2647


3 Bardenpho Medium 3, 4 16277.2603 14649.534





8 Bardenpho Large 9, 10, 15 24415.8904 21974.301

9 - - - - -








17 - - - - -

18 Bardenpho Large 21, 27, 28 24415.8904 21974.301


20 - - - - -






192

Table D.10 Results of the optimization model for scenario Low P High E




ZC (M$/year)

ZE (ton CO2/year)

Marginal

1 Bardenpho Medium 1, 2, 7, 8 130 117

11.744384 2160.0698

2 - - - - -




6 - - - - -

7 - - - - -

8 - - - - -








16 - - - - -


18 - - - - -


20 - - - - -



23 - - - - -



193

Table D.11 Results of the optimization model for scenario Medium P High E




ZC (M$/year)

ZE (ton CO2/year)

Marginal

1 Bardenpho Medium 1, 2, 7, 8 3,907 3,516

352.31617 64798.40379






7 - - - - -

8 - - - - -

9 - - - - -

10 - - - - -

11 - - - - -









20 - - - - -


22 - - - - -




194

Table D.12 Results of the optimization model for scenario High P High E




ZC (M$/year)

ZE (ton CO2/year)

Marginal

1 MBR+MF Medium 1, 2 16,277 14,650

2934.677 588526.0171


3 MBR+MF Medium 3, 4 16,277 14,650

4 MBR+MF Medium 5 8,139 7,325

5 Bardenpho Large 6, 11, 12 24,416 21,974


7 MBR+MF Medium 14 8,139 7,325

8 MBR+MF Medium 10, 15, 16 24,416 21,974





13 MBR+MF Medium 22 8,139 7,325



16 MBR+MF Medium 19, 26 16,277 14,650

17 MBR+MF Medium 21 8,139 7,325

18 - - - - -

19 MBR+MF Medium 28, 29 16,277 14,650

20 - - - - -

21 MBR+MF Large 25, 31, 32 24,416 21,974



24 MBR+MF Medium 34 8,139 7,325

25 Bardenpho Large 30, 35, 36 24,416 21,974

195

APPENDIX E: COPYRIGHT PERMISSION

Chapter 2 of this dissertation was reproduced from Ref. “A multi-criteria sustainability

assessment of water reuse applications: a case study in Lakeland, Florida” (DOI:

10.1039/C8EW00336J) with permission from the Royal Society of Chemistry.

Available at: https://pubs.rsc.org/en/content/articlehtml/2019/ew/c8ew00336j

From: RSC1 (shared) <[email protected]> Date: Tue, Apr 30, 2019 at 10:37 Subject: Request Permissions To: Nader Rezaei <[email protected]> Dear Nader, Please use the text below. The Royal Society of Chemistry (RSC) hereby grants permission for the use of your paper(s) specified below in the printed and microfilm version of your thesis. You may also make available the PDF version of your paper(s) that the RSC sent to the corresponding author(s) of your paper(s) upon publication of the paper(s) in the following ways: in your thesis via any website that your university may have for the deposition of theses, via your university’s Intranet or via your own personal website. We are however unable to grant you permission to include the PDF version of the paper(s) on its own in your institutional repository. The Royal Society of Chemistry is a signatory to the STM Guidelines on Permissions (available on request). Please note that if the material specified below or any part of it appears with credit or acknowledgement to a third party then you must also secure permission from that third party before reproducing that material. Please ensure that the thesis states the following: Reproduced by permission of The Royal Society of Chemistry and includes a link to the paper on the Royal Society of Chemistry’s website. Please ensure that your co-authors are aware that you are including the paper in your thesis. Kind regards, Anna Cooksey Publishing Assistant, Customer Services Royal Society of Chemistry T: +44 (0) 1223 432176 | www.rsc.org

https://pubs.rsc.org/en/content/articlehtml/2019/ew/c8ew00336j

http://www.rsc.org/

ABOUT THE AUTHOR

Nader Rezaei graduated with a B.S. degree in Chemical Engineering from Sharif

University of Technology (SUT) in 2009. He obtained his M.S. degree in Chemical Engineering,

with a concentration on Environmental Engineering, from SUT in 2011. Nader started his PhD

career in Environmental Engineering program at the University of South Florida (USF) in 2017

and worked as a Research Assistant in Civil & Environmental Engineering department at USF.

His primary research interests include life cycle assessment (LCA); life cycle cost analysis

(LCCA); integrated water and wastewater system design and management; sustainability

assessment; treatment train design; water quality; resource conservation; water reclamation; water

supply chain design; and resource recovery.