Toward Sustainable Sanitationcfsites1.uts.edu.au/find/isf/publications/mckibbin2008...SANITATION A...
145
University of Technology, Sydney Faculty of Engineering TOWARD SUSTAINABLE SANITATION A least cost planning approach for assessing alternative sanitation futures by John McKibbin Student Number: 10050969 Project Number: S07‐098 Major: Civil and Environmental Engineering Supervisor: Dr Prasanthi Hagare (UTS Faculty of Engineering) Industry co‐supervisor: Dr Juliet Willetts (UTS Institute for Sustainable Futures) A 12 Credit Point Project submitted in partial fulfilment of the requirement of a Bachelor of Engineering 20 June 2008
Transcript of Toward Sustainable Sanitationcfsites1.uts.edu.au/find/isf/publications/mckibbin2008...SANITATION A...
University of Technology, Sydney
Faculty of Engineering
TOWARD SUSTAINABLE
SANITATION A least cost planning approach for assessing alternative sanitation
futures
by
John McKibbin
Student Number: 10050969
Project Number: S07‐098
Major: Civil and Environmental Engineering
Supervisor: Dr Prasanthi Hagare (UTS Faculty of Engineering)
Industry co‐supervisor: Dr Juliet Willetts (UTS Institute for Sustainable Futures)
A 12 Credit Point Project submitted in partial fulfilment of the requirement of a
Bachelor of Engineering
20 June 2008
ii
Statement of Originality
I declare that the text, theories, concepts, methods and results contained in this
thesis are my own work, except where specifically attributed to another source. Any
help that I have received in my research work and the preparation of this thesis has
been acknowledged accordingly.
Signed,
John McKibbin
iii
Abstract
The challenges of extending sanitation to our rapidly sprawling cities have
prompted a rethink in the way we manage our waste. In industrial countries,
conventional centralised, large‐scale wastewater systems have been the subject of
renewed scrutiny in light of rising costs and increasingly apparent resource
constraints. In developing countries faced with limited financial and institutional
capacity, the value of investing in and maintaining conventional sanitation solutions
is also under serious question.
This research aims to compare the cost‐effectiveness of a broad suite of sanitation
options by building upon a best‐practice costing framework and a sophisticated
water resource management tool, both developed by the UTS Institute for
Sustainable Futures together with water service providers across Australia. This
involved developing a detailed forecast of water and nutrient flows associated with
the various activities comprising the urban system under study. The forecast was
then used as a baseline for assessing the impact of a series of capacity
augmentation and demand management options for addressing the objectives of
the system. The associated life‐cycle costs to each stakeholder were then identified
as a basis for assessing the preferred course of action.
The new application required a comprehensive review and extension of the existing
model with key developments including an integrated resource forecast of water
and nutrients, a spatial method for developing demographic projections, and a
means of assessing the financial costs borne by all stakeholders.
The extended tool was demonstrated and validated through a case study
implementation based on data and assumptions for the city of Melbourne. The
central findings highlight how demand management and alternative capacity
options including distributed and ecological sanitation can provide cost‐effective
alternatives to central treatment and sewerage augmentations.
The methods and tools are recommended as a valuable tool in driving the uptake of
a broader mix of responses toward more cost‐effective and ecologically sustainable
sanitation futures.
iv
Acknowledgements
Firstly I would like to thank my supervisors Dr Prasanthi Hagare of the UTS Faculty
of Engineering and Dr Juliet Willetts of the UTS Institute for Sustainable Futures. I
first approached Prasanthi with my ambitious research question in May 2007 and
upon reflection she recognized far earlier than I the magnitude of the project and
yet courageously volunteered her oversight and ongoing support. To Juliet I owe my
gratitude for trusting in my ability from the beginning and extending me as a
researcher through her critical review of my work.
For helpful advice and direction during the early stages of my project I would like to
thank Dana Cordell, for imparting her extensive knowledge regarding the important
role of phosphorus and material flow analysis; also Dena Fam, for contributing her
understanding of the cultural acceptability and design of ecological sanitation
systems; and Kurt Forrester, for providing his advice on end use modeling strategies
and computer coding.
For sharing their thought provoking research and providing peer review on the
assessment methodology I would like to acknowledge the delegates at the IWA
Specialised Conference on Small Water and Wastewater Systems including
specifically Dr. Günter Langergraber (BOKU University), Prof. Goen Ho (Murdoch
University), Dr Anna Norström (CIT Urban Water Management), Dr. Norbert
Weissenbacher (BOKU University), Martin Wafler (Seecon International GmbH) and
Prof. Leigh Davison (Southern Cross University). Special thanks also to my hosts and
guides during my sanitation field study in India, including Tency Baetens (Auroville
Centre for Scientific Research), Susmita Sinha (BORDA International), and S.
Vishwanath (Rainwater Club).
For peer review and insightful comments on developments to the supporting tool I
would like to thank my colleagues at the UTS Institute for Sustainable Futures
including Prof. Stuart White, Dr Simon Fane, Andrea Turner, Alexander Kazaglis and
Dr Kumi Abersuriya.
Final thanks go to my parents and my partner for enduring me this past 12 months!
v
Contents
ABSTRACT ........................................................................................................................ III
ACKNOWLEDGEMENTS ....................................................................................................... IV
CONTENTS ....................................................................................................................... V
LIST OF FIGURES AND TABLES .............................................................................................. VII
LIST OF ACRONYMS AND ABBREVIATIONS ................................................................................. X
1 INTRODUCTION ................................................................................................. 1
1.1 CONTEXT .......................................................................................................... 1
1.2 OBJECTIVES ...................................................................................................... 1
1.3 SCOPE ............................................................................................................. 2
1.4 REPORT STRUCTURE ............................................................................................ 2
2 CHALLENGES, RESPONSES AND WAYS FORWARD ............................................. 3
2.1 THE RISE OF SANITATION ...................................................................................... 3
2.2 THE CHALLENGES OF SANITATION ........................................................................... 5
2.2.1 MAXIMISING RESOURCE EFFICIENCY ........................................................................... 5
2.2.2 MAXIMISING COST‐EFFECTIVENESS .......................................................................... 11
2.3 NEW SANITATION RESPONSES .............................................................................. 17
2.3.1 DISTRIBUTED SANITATION: MATCHING SYSTEM SCALE TO URBAN DENSITY ........................ 17
2.3.2 ECOLOGICAL SANITATION: CLOSING THE RESOURCE LOOP ............................................. 19
2.4 NEW DECISION‐MAKING APPROACHES ................................................................... 21
2.4.1 MATERIAL FLOW ANALYSIS: OPTIMISING RESOURCE FLOWS .......................................... 21
2.4.2 LEAST COST PLANNING: OPTIMISING FINANCIAL FLOWS ................................................ 26
3 DEVELOPMENTS TO THE METHOD AND SUPPORTING TOOLS .......................... 33
3.1 ADAPTING THE METHOD .................................................................................... 33
3.2 EXTENDING THE SUPPORTING TOOL ....................................................................... 35
3.2.1 ESTABLISHING THE FRAMEWORK: NUTRIENT AND GREENHOUSE VALUATION ..................... 35
3.2.2 IDENTIFYING THE SYSTEM: MULTI‐RESOURCE ACCOUNTING ........................................... 39
3.2.3 SPECIFYING THE BASE CASE: SPATIAL CONTROL OF REGIONS .......................................... 49
3.2.4 IDENTIFYING THE OPTIONS: NEW OPTION MODELS ...................................................... 52
3.2.5 ANALYSING THE COSTS: FINANCIAL FLOW ANALYSIS ..................................................... 56
3.2.6 ANALYSING UNCERTAINTY: STOCHASTIC ANALYSIS ....................................................... 58
vi
4 A CASE STUDY IMPLEMENTATION ................................................................... 61
4.1 ESTABLISHING THE FRAMEWORK .......................................................................... 62
4.1.1 DEFINING THE OBJECTIVES ..................................................................................... 62
4.1.2 DEFINING THE ECONOMIC CRITERIA ......................................................................... 62
4.1.3 DEFINING THE TREATMENT OF EXTERNALITIES ............................................................ 63
4.2 IDENTIFYING THE SYSTEM ................................................................................... 64
4.2.1 DEFINING THE BOUNDARIES ................................................................................... 64
4.2.2 IDENTIFYING THE KEY COMPONENTS AND THEIR EXCHANGES ......................................... 64
4.2.3 QUANTIFYING KEY FLOWS ...................................................................................... 66
4.3 SPECIFYING THE BASE CASE ................................................................................. 67
4.3.1 PROJECTING BASELINE DEMOGRAPHY ....................................................................... 67
4.3.2 PROJECTING BASELINE RESOURCE FLOWS .................................................................. 70
4.4 IDENTIFYING THE OPTIONS .................................................................................. 88
4.4.1 DISTRIBUTED SYSTEM ........................................................................................... 88
4.4.2 ECOLOGICAL SYSTEM ............................................................................................ 93
4.5 ANALYSING THE COSTS ....................................................................................... 98
4.5.1 DISTRIBUTED ALTERNATIVE .................................................................................... 98
4.5.2 ECOLOGICAL ALTERNATIVE ................................................................................... 100
4.5.3 ASSESSING THE LEAST COST ALTERNATIVE ............................................................... 103
5 CONCLUSIONS AND RECOMMENDATIONS .................................................... 104
6 REFERENCES .................................................................................................. 107
APPENDIX A BASELINE ASSUMPTIONS AND DESCRIPTIONS REPORT ................ 112
APPENDIX B STOCK MODEL BUILDER INTERFACE AND CODE............................ 119
APPENDIX C EXAMPLE STOCHASTIC ANALYSIS MACRO .................................... 122
APPENDIX D IWA CONFERENCE PAPER ............................................................ 123
APPENDIX E FIELD STUDY PHOTOGRAPHS ........................................................ 129
vii
List of Figures and Tables
Figures
FIGURE 2‐1 – THEMATIC MAP OF GLOBAL BLUE WATER APPROPRIATION (SMAKHTIN, REVENGA & DÖLL 2004) ............ 6
FIGURE 2‐2 – THEMATIC MAP DEPICTING THE PROJECTED INCREASE IN CONSUMPTIVE WATER ASSOCIATED WITH MEETING
THE MDG FOR HUNGER (ROCKSTRÖM ET AL. 2005) ................................................................................ 7
FIGURE 2‐3 ‐ HISTORICAL SOURCES OF PHOSPHATE FERTILIZERS (CORDELL, DRANGERT & WHITE SUBMITTED) .............. 9
FIGURE 2‐4 ‐ GLOBAL MASS FLOW DIAGRAM OF PHOSPHORUS (UNEP 2005) ...................................................... 10
FIGURE 2‐5 ‐ THE ECONOMIES AND DISECONOMIES OF SANITATION SYSTEM SCALE (CLARK 1997) ............................ 12
FIGURE 2‐6 – OPTIMUM SCALE OF URBAN WATER SYSTEMS (CLARK 1997) .......................................................... 13
FIGURE 2‐7 ‐ WORLD ACCESS TO IMPROVED SANITATION (UNICEF & WHO 2004) ............................................. 15
FIGURE 2‐8 ‐ PROCESS FLOW DIAGRAM OF A DRY ECOLOGICAL SANITATION SYSTEM (ROCKSTRÖM ET AL. 2005) .......... 19
FIGURE 2‐9 ‐ SYSTEM MODEL OF HOUSEHOLD WATER BALANCE (GUMBO 2005) ................................................... 23
FIGURE 2‐10 ‐ SYSTEM MODEL OF HOUSEHOLD PHOSPHORUS BALANCE (GUMBO 2005) ........................................ 24
FIGURE 2‐11 – PERSONAL WATER AND NUTRIENT INPUTS AND OUTPUTS (GUMBO 2005) ...................................... 25
FIGURE 2‐12 – SCREENSHOT FROM ISDP MODEL DEMONSTRATING THE SUPPLY DEMAND FORECAST ......................... 30
FIGURE 2‐13 – SCREENSHOT FROM ISDP MODEL DEMONSTRATING THE IMPACT OF OPTIONS .................................. 31
FIGURE 2‐14 – SCREENSHOT FROM ISDP MODEL DEMONSTRATING THE RANKING OF OPTIONS BASED ON A STEP CHART 32
FIGURE 3‐1 ‐ PROJECTION OF GREENHOUSE GAS EQUIVALENT PERMIT PRICES FOR AUSTRALIA (MMA 2006) ............. 37
FIGURE 3‐2 ‐ PROJECTION OF RETAIL ENERGY PRICES BY CUSTOMER TYPE ............................................................. 38
FIGURE 3‐3 ‐ CONCEPTUAL MODEL OF THE SANITATION SYSTEM ......................................................................... 40
FIGURE 3‐4 ‐ EXCEL SCREENSHOT DESCRIBING TYPICAL STOCK MODEL ARRAY ........................................................ 43
FIGURE 3‐5 ‐ STOCK MODEL OUTPUT DESCRIBING CLOTHESWASHER STOCK SHARE BY APPLIANCE TYPE ....................... 45
FIGURE 3‐6 ‐ STOCK MODEL OUTPUT DESCRIBING TOILET STOCK SHARE BY APPLIANCE TYPE ...................................... 46
FIGURE 3‐7 ‐ STOCK MODEL OUTPUT DESCRIBING SHOWER STOCK SHARE BY APPLIANCE TYPE ................................... 47
FIGURE 3‐8 ‐ AVERAGE HOURLY PROFILE OF WATER DEMAND PER PROPERTY BY END USE (ROBERTS 2005) ................ 48
FIGURE 3‐9 ‐ SCREENSHOT FROM MAPINFO INTERFACE DEMONSTRATING THE SPATIAL SELECTION OF CENSUS DATA .... 50
FIGURE 3‐10 ‐ SCREENSHOT OF MAPINFO INTERFACE DEMONSTRATING THE SPATIAL SELECTION OF PROJECTION DATA . 51
FIGURE 3‐11 ‐ MODEL OUTPUT DEMONSTRATING THE MATCHING CLUSTER SUPPLY TO DEMAND .............................. 53
FIGURE 3‐12 – SCREENSHOT OF MODEL SHOWING FINANCIAL FLOW ANALYSIS OF OPTIONS ..................................... 57
FIGURE 4‐1 – CONCEPTUAL MODEL FOR FLOWS OF WATER AND PHOSPHORUS IN THE URBAN SYSTEM ........................ 65
FIGURE 4‐2 ‐ TIME SERIES OF POPULATION FOR THE STUDY AREA ........................................................................ 68
FIGURE 4‐3 ‐ TIME SERIES OF DWELLINGS FOR THE STUDY REGION ...................................................................... 69
FIGURE 4‐4 ‐ PROJECTION FOR WATER FLOWS ASSOCIATED WITH BATHING ACTIVITIES BY APPLIANCE ......................... 70
FIGURE 4‐5 ‐ PROJECTION FOR WATER FLOWS ASSOCIATED WITH CLOTHESWASHING ACTIVITIES BY APPLIANCE ............ 72
FIGURE 4‐6 ‐ PROJECTION FOR WATER FLOWS ASSOCIATED WITH DISHWASHING ACTIVITIES BY APPLIANCE .................. 73
FIGURE 4‐7 ‐ PROJECTION FOR WATER FLOWS ASSOCIATED WITH TOILET FLUSHING ................................................ 74
FIGURE 4‐8 ‐ PROJECTION OF WATER FLOWS ASSOCIATED WITH NON‐RESIDENTIAL COMPONENTS ............................. 75
FIGURE 4‐9 ‐ PROJECTION OF BASELINE SEWAGE FLOW BY END USE ..................................................................... 76
viii
FIGURE 4‐10 – SHARE OF SEWAGE FLOW BY END USE ....................................................................................... 77
FIGURE 4‐11 ‐ PROJECTION OF PHOSPHORUS FLOWS ASSOCIATED WITH TOILET FLUSHING ....................................... 78
FIGURE 4‐12 ‐ PROJECTION OF PHOSPHORUS FLOWS ASSOCIATED WITH CLOTHESWASHING DETERGENTS .................... 79
FIGURE 4‐13 ‐ PROJECTION OF PHOSPHORUS FLOWS ASSOCIATED WITH DISHWASHING DETERGENTS ......................... 80
FIGURE 4‐14 – PROJECTION OF PHOSPHORUS FLOWS ASSOCIATED WITH TOILET FLUSHING IN NON‐RESIDENTIAL PREMISES
.................................................................................................................................................... 81
FIGURE 4‐15 ‐ PROJECTION OF BASELINE PHOSPHORUS FLOW BY END USE ............................................................ 82
FIGURE 4‐16 ‐ SHARE OF PHOSPHORUS FLOW BY END USE ................................................................................ 83
FIGURE 4‐17 ‐ PROJECTION OF NITROGEN FLOWS ASSOCIATED WITH TOILET FLUSHING IN RESIDENTIAL DWELLINGS ...... 84
FIGURE 4‐18 – PROJECTION OF NITROGEN FLOWS ASSOCIATED WITH TOILET FLUSHING IN NON‐RESIDENTIAL PREMISES . 85
FIGURE 4‐19 ‐ PROJECTION OF BASELINE NITROGEN FLOW BY END USE ................................................................ 86
FIGURE 4‐20 – SHARE OF NITROGEN FLOW BY END USE .................................................................................... 87
FIGURE 4‐21 ‐ COLLECTION COMPONENTS: INTERCEPTOR TANK, EFFLUENT PUMP AND PRESSURISED SEWER (ORENCO
SYSTEMS 2007) .............................................................................................................................. 88
FIGURE 4‐22 ‐ TREATMENT COMPONENTS: MODULAR RECIRCULATING TEXTILE FILTERS (ORENCO SYSTEMS 2007) ...... 89
FIGURE 4‐23 ‐ PROJECTED OFFSET SEWAGE FLOW TO WESTERN TREATMENT PLANT .............................................. 90
FIGURE 4‐24 ‐ PROJECTED REDUCTION IN EFFLUENT PHOSPHORUS FROM WESTERN TREATMENT PLANT .................... 91
FIGURE 4‐25 – COLLECTION COMPONENTS: URINE DIVERTING PEDESTAL AND WATERLESS URINAL (KVARNSTRÖM ET AL.
2006) ........................................................................................................................................... 94
FIGURE 4‐26 ‐ TREATMENT COMPONENTS: ROTARY COMPOST BIN AND URINE STORAGE TANK (KVARNSTRÖM ET AL.
2006) ........................................................................................................................................... 94
FIGURE 4‐27 ‐ REUSE COMPONENTS: SMALL AND LARGE‐SCALE APPLICATION OF URINE FOR AGRICULTURAL REUSE
(KVARNSTRÖM ET AL. 2006) ............................................................................................................. 95
FIGURE 4‐28 – GREYWATER REUSE COMPONENTS: DIVERTER STORAGE, PUMP AND DRIPLINE SYSTEM (ECO‐CARE, 2008)
.................................................................................................................................................... 95
FIGURE 4‐29‐ FINANCIAL IMPACT OF DISTRIBUTED SANITATION OPTION ............................................................... 98
FIGURE 4‐30‐ FINANCIAL IMPACT OF ECOLOGICAL SANITATION OPTION ............................................................. 100
FIGURE 4‐31 ‐ ECONOMIC LEVELISED COSTS OF ALTERNATIVE SANITATION GROWTH SERVICING OPTIONS (BASELINE
TRANSFERS EXCLUDED) .................................................................................................................... 103
ix
Tables
TABLE 2‐1 ‐ TYPICAL COMPOSITION OF SANITATION STREAMS (ADAPTED FROM VINNERÅS ET AL. 2006) ...................... 5
TABLE 2‐2 ‐ CAPITAL AND OPERATING COSTS OF VARIOUS CENTRALISED WASTEWATER TREATMENT PLANTS (FOESS ET AL.
1998) ........................................................................................................................................... 11
TABLE 2‐3 ‐ ESTIMATED REPLACEMENT COST OF URBAN WASTEWATER SYSTEMS IN SELECTED COUNTRIES (MAURER,
ROTHENBERGER & LARSEN 2005) ...................................................................................................... 14
TABLE 2‐4 ‐ ESTIMATED SANITATION TARGET POPULATIONS TO MEET MDG BY 2015 BY UN REGION (ROCKSTRÖM ET AL.
2005) ........................................................................................................................................... 16
TABLE 2‐5 ‐ ESTIMATED HOUSEHOLD UNIT COSTS FOR URBAN AND RURAL SANITATION PROGRAMS (ROCKSTRÖM ET AL.
2005) ........................................................................................................................................... 16
x
List of acronyms and abbreviations
ABS Australian Bureau of Statistics
FAO United Nations Food and Agriculture Organisation
IRP Integrated resource planning
iSDP integrated supply‐demand planning
LCP Least cost planning
MFA Material flow analysis
MMA McLennan Magasanik & Associates
NPV Net present value
SFA Substance flow analysis
UNEP United Nations Environment Programme
WHO World Health Organisation
WSAA Water Services Association of Australia
Introd
uctio
n
1
1 Introduction
1.1 Context Sanitation can provide a system of balance. It can provide a system of resource balance
in returning the materials withdrawn from our environment; and a system of
ecological balance in preserving the health of our water and soils, and therefore
ourselves.
The premise of this research is that the dominant sanitation paradigm currently
induces a state of imbalance. By flushing excreta and water to rivers and oceans,
wastewater systems not only waste a valuable resource in its embodied water,
nutrients and organic material; they waste another, as receiving waters are
increasingly impacted with pathogens, eutrophying nutrients and other contaminants.
The dominant paradigm is also expensive. While industrial nations struggle with the
rapidly increasing costs associated with extending and sustaining conventional
centralised, large‐scale wastewater systems, developing nations are striving to
replicate those same systems at great financial and human cost.
A more sophisticated sanitation response is therefore called upon if the promise of
universal and sustainable sanitation is ever to be realised.
1.2 Objectives The intent of the project is to build upon the integrated resource planning framework
and supporting tools to compare alternative futures for cost‐effectively providing
sanitation. In so doing, this research seeks to reveal the conditions under which
sustainable sanitation programs may be technically and financially viable. The
objectives are therefore
1. To identify the existing and emerging challenges of sanitation planning and
some potential responses toward meeting them.
2. To extend the least cost planning method and supporting tool for economically
assessing alternative sanitation futures; and
3. To undertake a case study implementation of the method and supporting tool
to both validate the process and demonstrate its capabilities.
Introd
uctio
n
2
1.3 Scope The study will therefore involve a review of the key global challenges associated with
human waste sanitation systems, some broad responses from which to draw solutions,
a rigorous review and extension of the least cost planning method and supporting
tools and an illustrative demonstration of the method based on two alternative future
scenarios.
The study does not engage with the unique challenges of solid waste management, nor
the challenges associated with the emission of toxics, air or other waste streams.
Similarly the case study implementation does not engage with a detailed review of the
various available sanitation technologies or seek to provide a broad strategic review.
1.4 Report structure The structure of this report constitutes four parts:
• The first chapter (§2) describes a literature review introducing the history of
sanitation (§2.1), its existing and emerging challenges (§2.2), some potential
responses toward meeting those challenges (§2.3), and some approaches
toward their assessment (§2.4).
• The second chapter (§3) introduces the method applied within this dissertation
(§3.1) and describes the revision and extension of the supporting tool (§3.2).
• The third chapter (§4) demonstrates the method and supporting tool by
assessing alternative sanitation futures in the city of Melbourne, Australia.
• Finally the fourth chapter (§5) concludes the research with a critique of the
method and tools and provides recommendations upon further research into
their broader application.
To lay the context for this dissertation the following section begins by introducing the
history of sanitation.
Challenges, respo
nses and
ways forw
ard
3
2 Challenges, responses and ways forward
2.1 The rise of sanitation An understanding of the context of past technological decisions with regard to
sanitation provides a useful perspective upon the present challenges.
The history of sanitation is closely bound to the history of cities. That is, the need for
more sophisticated human waste management has been principally driven by the
perceived aesthetic and health impacts associated with dense, urban settlements.
Dry conservancy therefore arose as a first response to this displacement of urban
dwellers from agricultural land. The system implied collecting dry human excreta in a
cess pit or similar for regular transfer to the surrounding farmland by cart. Excreta
were typically applied to the soil fresh, and there was often provision for the separate
collection of urine for such purposes as tanning hides (Mumford 1961).
As urban settlements became increasingly populous and dense, this sanitary task
became increasingly challenging. While historic cities such as Edo (i.e. the forerunner
of Tokyo) developed increasingly sophisticated means of dry conservancy applying
composting processes (Narain 2002), ancient cities including Mohenjo‐Daro and Rome
began the first experiments with removing excreta by hydraulic carriage using
channels and sewers. However such systems were never broadly applied, with
coverage being limited to several public institutions and rich tenements, and dry
conservancy remained the dominant means of sanitation until relatively recent times
(Mumford 1961).
Beder (1993) traces the genesis of the hydraulic carriage paradigm to a forty year
period following the sanitary crisis of 19th Century London. At this time hydraulic
carriage was advocated on the basis that by enabling the rapidly removal of organic
waste from the household, the incidence of putrefaction would be reduced and
therefore the consequent 'miasma' or disease producing gases (Sewage and Health
Board, 1875 cited in Beder 1993). The benefits of centralised control were also
advocated publicly (Corfield, 1871, p118 cited in Beder 1993).
Challenges, respo
nses and
ways forw
ard
4
Dry conservancy, by contrast, was defended on the grounds that the solid component
of human waste contained the major fraction of nutrients for fertilising crops and was
the principal cause of waterway pollution (Burke, 1873, p21; Waring, 1889, p365 cited
in Beder 1993). However such arguments were only popularly voiced following the
recognition of the failings of existing sewerage systems and were made in an context
of plentiful water and fertiliser substitutes, an apparently abundant pollutant sink and
little cultural awareness of aquatic ecosystem health (Beder 1993).
The growing experience with water‐carriage systems lead to a recognition that
removing suspended solids and clarifying effluent did not prevent the effluent from
putrefying and causing a nuisance when discharged (Sewage and Health Board, 1877,
p9 cited in Beder 1993).
Staged treatment was therefore developed in response to the perceived failings of
wastewater treatment. An initial 'preliminary' treatment by chemical treatment, plain
sedimentation or septic tank treatment was only considered satisfactory unless
followed by a second stage of treatment, which was usually some form of filtration
(Royal Commission, 1908, p18 cited in Beder 1993).
Thus the Royal Commission in 1908 marked the firmament of a sanitation paradigm
marked by hydraulic carriage by gravity sewer and staged treatment comprising a
series of unit operations and processes that was subsequently permeated and adapted
across the world.
However it is important to note that the debate between hydraulic carriage and dry
conservancy was highly contentious, and was based upon an ignorance of the
mechanism of disease transmission, biological oxygen demand or eutrophication.
Nonetheless the subsequent level of investment, engineering experience, cultural
norms, and expectations each served to ‘lock‐in’ those decisions.
However for a number of reasons this study asserts that such technological inertia may
be overcome based on the force of evidence indicating the limitations of the current
paradigm, which is the subject of the following section.
Challenges, respo
nses and
ways forw
ard
5
2.2 The challenges of sanitation The current sanitation paradigm is now the subject to two key challenges: maximising
resource efficiency in light of the increasingly apparent scarcity of water, energy and
nutrients, and maximising cost‐effectiveness in light of rising costs and a global
sanitation deficit. The following section seeks to investigate the underlying drivers of
this situation as a foundation toward understanding its resolution.
2.2.1 Maximising resource efficiency
Conventional sanitation systems may be interpreted as a conveyance of three major
waste streams to oceans and rivers: those are urine, faeces, and water. Water may be
further subdivided into blackwater (i.e. that water associated with toilet flushing), and
greywater (i.e. that water associated with clothes washing, bathing and other uses).
These waste streams each include varying intensities of water, organic matter, and
nutrients (primarily nitrogen, phosphorus and potassium) as summarised in Table 2‐1.
Table 2‐1 ‐ Typical composition of sanitation streams (adapted from Vinnerås et al. 2006)
Parameter Unit Urine Faeces Greywater* Biodegradables
Wet mass Kg/person.a 550 2% 51 0.1% 36500 98% 80.3 0.2%
Dry mass Kg/person.a 21 26% 11 14% 20 25% 27.5 35%
Nitrogen g/person.a 4000 71% 550 10% 500 9% 550 10%
Phosphorus g/person.a 365 43% 183 22% 190 23% 104 12%
Potassium g/person.a 1000 55% 365 20% 365 20% 82 5%
*Note: blackwater is not accounted in these metrics
Note that faeces, and its associated pathogenic risk, only constitute approximately 50
kg per person per year, however this component contaminates over 36,000 litres of
water when mixed as a wastewater stream. Increasing attention has been directed
towards the sustainability of such a conveyance of resources owing to a number of
pressures.
Challenges, respo
nses and
ways forw
ard
6
A growing water scarcity crisis
Over 1.4 billion of the world’s population are currently defined as living in a situation
of acute water stress (Smakhtin, Revenga & Döll 2004). That is, 1.4 billion people are
living in hydrological catchments assessed to be currently over‐allocated, leaving
insufficient water to satisfy environmental flow requirements. The principle driver for
the situation is irrigated agriculture, which comprises the majority share of
consumptive water use.
Figure 2‐1 shows a thematic map of blue water appropriation relative to
environmental capacity as assessed by Smakhtin et al (2004).
Figure 2‐1 – Thematic map of global blue water appropriation (Smakhtin, Revenga & Döll 2004)
The map indicates those over‐appropriated areas in yellow and red. Note the red areas
are concentrated upon Northern Africa, Western and Southern Asia, and the Western
United States of America. The area marked yellow in Australia is the Murray‐Darling
catchment, which is a substantially over‐allocated irrigated agriculture region.
Challenges, respo
nses and
ways forw
ard
7
However recent analysis undertaken by Rockström et al (2005) provides a fuller picture
of the emerging water scarcity crisis. The analysis involved assessing the necessary
consumptive water requirements associated with meeting the United Nations
Millennium Development Goal for hunger: a united undertaking to reduce the global
undernourished population by half. The results depicted as a thematic map are shown
in Figure 2‐2.
Figure 2‐2 – Thematic map depicting the projected increase in consumptive water associated with meeting the MDG for hunger (Rockström et al. 2005)
The thematic map indicates an increase in water demand of between 80 to greater
than 120% in developing countries across Southern and South‐Eastern Asia and Sub‐
Saharan Africa.
The water scarcity crisis is therefore set to become more acute as global population
growth and agricultural development combine toward an additional 50% of current
global water demand by 2015 (Rockström et al. 2005). The authors conclude a new
agricultural revolution of a scale greater than that of the 1960s and 70s, which must be
predominantly fed by alternative sources to traditional blue water diversions and
storages.
Given such a context, the discharge of scarce water reserves associated with
conventional wastewater systems is questionable. However there is another
dimension that entwines sanitation with food security that is the subject of the
following section.
Challenges, respo
nses and
ways forw
ard
8
An interrupted nutrient cycle
An adequate and balanced supply of elements necessary for life, provided through the
ecological processes of nutrient cycling, underpins all other ecological services.
However the cycling of several key elements; chiefly carbon, phosphorus, nitrogen and
sulphur; have been substantially altered over the past two centuries leading to
significant consequences to human and ecological health and productivity (UNEP
2005).
A leading cause of this alteration is associated with the sanitary revolution of the late
19th century. By redirecting the excreta and its embodied nutrients from terrestrial to
aquatic ecosystems, such systems effectively ‘opened the nutrient loop’.
The impacts of disposing excess nutrient loads to aquatic systems are both diverse and
well recognised. These include shifts in the composition of bloom‐forming algae and
biomass, oxygen depletion, increased fish and shellfish mortality, aesthetic and water
treatment problems (UNEP 2005). This has lead to significant investments in nutrient
removal technologies (discussed in Section 2.2.2), together with an increased focus
upon the nutrient loads associated with detergents.
However, as alluded above, the diversion of nutrients associated with conventional
sanitation has broader implications for resource sustainability owing to their key role
in driving agricultural yields.
Challenges, respo
nses and
ways forw
ard
9
Figure 2‐3 depicts the historical sources of phosphorus fertilisers from the year 1800 to
2007.
Figure 2‐3 ‐ Historical sources of phosphate fertilizers (Cordell, Drangert & White submitted)
The chart disaggregates total fertiliser inputs by its constituent sources, whether they
be derived from manure, human excreta, guano or mineral phosphates. While prior to
the twentieth century the dominant source of phosphorus fertilisers was drawn from
manure and to a smaller extent human excreta and guano, the data specifically
highlight the degree to which artificial fertilisers have since driven agricultural
productivity growth.
The implication is that the sanitary revolution of the early twentieth century was
principally enabled by a parallel agricultural revolution, which obviated the need to
restore phosphorus to the land. However, as with water, the productivity gains
associated with the past agricultural revolution may be shortlived.
Challenges, respo
nses and
ways forw
ard
10
Figure 2‐4 depicts an analysis of the global mass flow of phosphorus.
Figure 2‐4 ‐ Global mass flow diagram of phosphorus (UNEP 2005)
The model indicates both a significant accumulation of phosphorus in soils and a
dramatically enhanced phosphorus flux comparative to natural levels.
Significantly, the model highlights the inherently unsustainable condition of the
modified phosphorus cycle. As discussed above phosphate rocks constitute the
principal driver of the total phosphorus flux. However the key point to note is the
dotted line at the bottom of the chart, which denotes that the return of phosphorus is
associated with geological time scales. There is therefore a net depletion of
phosphorus resources.
Similar to the current peak oil discourse, increasing attention has recently been
directed toward the potential implications of phosphate depletion (Rosmarin 2004).
Although projections upon the remaining stocks of phosphate rocks vary considerably
between 50 to 120 years (Steen 1998), the unarguable fact is that the discharge of
phosphates today reduces the potential productivity of soils and therefore Earth’s
future capacity to sustain life. And, unlike oil, phosphorus has no substitute (USGS
2008; Rosmarin 2004).
Challenges, respo
nses and
ways forw
ard
11
2.2.2 Maximising costeffectiveness
The principal costs associated with centralised wastewater systems include those costs
associated with treatment including capital construction and ongoing labour, chemical,
and sludge handling costs; and those costs associated with collection, including the
extensive initial outlay of sewerage, and operational costs associated with pumping
and maintenance. The magnitude and justification for such costs has recently been the
subject of increasing scrutiny in response to a number of pressures, detailed below.
Rising treatment costs
The addition of nutrient removal technology as a response to the eutrophication of
aquatic ecosystems has resulted in significant additional costs. These comprise capital
cost increases associated with construction of large nitrification tanks and additional
operating costs associated with the electricity, chemical inputs and sludge handling.
These increases have combined toward a doubling of life‐cycle costs on average (Bohn
1997; Nolting & Dahlem 1997 cited in Maurer, Rothenberger & Larsen 2005). Table 2‐2
depicts the observed capital and operational costs for a large sample of centralised
wastewater treatment plants in the United States.
Table 2‐2 ‐ Capital and operating costs of various centralised wastewater treatment plants (Foess et al. 1998)
System Type Cost System capacity (gallons / day)
4,000 10,000 25,000 50,000 100,000
Secondary Capital ($) 183,000 223,000 303,000 461,000 671,000
Operational ($/a) 22,000 26,500 39,200 52,100 78,000
3‐stage Tertiary Capital ($) 291,000 333,000 441,000 627,000 913,000
Operational ($/a) 35,900 41,900 56,400 76,200 115,900
4‐stage Tertiary Capital ($) 336,000 368,000 475,000 666,000 968,000
Operational ($/a) 52,500 57,600 73,800 95,900 132,300
Note that the improved quality that tertiary treatment affords is associated with
incremental increases in capital costs of between 36% to as much as 83%, while
incremental increases in operational costs are 49% to 139%.
Note also that the additional costs, which are predominantly associated with chemical
and sludge handling, do not offer the same economies of scale afforded to secondary
treatment systems. That is, they increase pro‐rata with increased sewage flows.
Challenges, respo
nses and
ways forw
ard
12
Rising collection costs
Traditional sanitary engineering theory has assumed centralised wastewater systems
inherently operate under an economy of scale. This implies that the marginal cost of
providing additional service connections to a centralised system reduces proportionate
to system scale. However recent studies have revealed a marginal cost minima is
reached following which marginal connection costs increase with system scale (Clark
1997).
Figure 2‐5 demonstrates a regression of capital and operational costs per service
connection in the city of Adelaide, Australia.
Figure 2‐5 ‐ The economies and diseconomies of sanitation system scale (Clark 1997)
The left chart shows the annualised capital costs per service connection associated
with capital and collection components. Treatment capital costs per connection exhibit
a significant economy of scale to 100 service connections, levelling off to at
approximately 10,000 service connections. Simultaneously collection capital costs per
service connection rise to become the dominant share of costs beyond approximately
100 service connections, and continue to rise beyond 10,000 service connections, at
which point constituting 82% of overall costs per service connection.
The chart on the right indicates the annual operating costs per service connection once
again apportioned to treatment and collection components. Treatment operational
costs per service connection dominate and decrease considerably with scale at a
decreasing rate, while collection operational costs are largely constant on a per
connection basis.
Challenges, respo
nses and
ways forw
ard
13
Combining these two components yielded the total system cost per service in Figure
2‐6.
Figure 2‐6 – Optimum scale of urban water systems (Clark 1997)
Density ‘x1’ indicates the characteristic densities for Adelaide city, which indicates a
negligible economy of scale from 600 to 1 million service connections of 4%. Also, as
the assumed density reduces by one fifth a clear minima emerges at approximately
100 service connections, while a five‐fold increase in densities shifts the total
annualised cost minimum down and to the right.
The implication being that there is little financial incentive for increased system scale
in urban conditions beyond approximately 600 service connections; while in areas of
lower density (e.g. peri‐urban areas) there is considerable financial incentive toward
more decentralised systems.
Note this analysis does not account for the significant capital and operational savings
in treatment inherent in a shift toward more distributed wastewater management
models (further discussed in Section 2.3.1).
Challenges, respo
nses and
ways forw
ard
14
Recent international research by Maurer et al (2005) suggests the maintenance and
renewal of sewerage infrastructure may have been under‐estimated globally. In fact,
the estimates predict an emerging renewal cost that is comparable or greater than the
initial capital outlay, and are shown in Table 2‐3.
Table 2‐3 ‐ Estimated replacement cost of urban wastewater systems in selected countries (Maurer, Rothenberger & Larsen 2005)
Country Population
(106)
Served Sewer/WWTP Total
(US$/Person)
Sewer
(US$/Person)
WWTP
(US$/Person)
Denmark 5.4 0.87 0.77 5300
Switzerland 7.3 0.96 0.95 4400 3650 750
Austria 8.2 0.75 0.74 4800 3900 900
Italy 58.0 0.77 0.63 3900 3200 700
UK 60.1 0.98 0.87 3700
France 60.2 0.81 0.77 2600
Germany 82.4 0.92 0.87 2600 1850 750
USA 275.3 1700
The data suggests that sewerage systems across the Western world are at the cusp of
a significant period of renewal that will serve to instil in decision‐makers of the true
life‐cycle costs of centralised wastewater systems. This is serving as a significant
opportunity in re‐thinking the scale and form of sanitation systems.
Meanwhile the developing nations are faced with an entirely different financial
challenge, which is the subject of the following section.
Challenges, respo
nses and
ways forw
ard
15
A global financial deficit in sanitation provision
A global perspective provides clarity upon the scale of the sanitation challenge. The
WHO / UNICEF Joint Monitoring Program estimates 2.6 billion people in the world lack
access to the most basic sanitation service, the vast majority of which live in Asia and
sub‐Saharan Africa as shown in Figure 2‐7 (UNICEF & WHO 2004).
Figure 2‐7 ‐ World access to improved sanitation (UNICEF & WHO 2004)
However the picture worsens if a more stringent, western definition of adequate
sanitation, with less than one third of the world’s population served by flushing toilets
and sewerage systems (UNDP 2006, p. 112).
The scale of installations required toward resolving these inequalities are considerable.
According to projections undertaken by Rockström et al (2005), over 1.7 billion urban
and rural household toilets will be required in order to meet the coming Millenium
Development Goal of halving the deficit in sanitation access by 2015 (United Nations
2000).
Challenges, respo
nses and
ways forw
ard
16
Table 2‐4 ‐ Estimated sanitation target populations to meet MDG by 2015 by UN Region (Rockström et al. 2005)
UN Region Urban sanitation Urban Population (millions) Rural Population (millions)
East Asia 247.9 147.8
Eurasia 7.5 16.2
Latin America & Caribbean 114.8 25.2
North Africa 27.6 17.8
Oceania 0.8 2.7
South‐East Asia 89.7 60.6
Southern Asia 189.5 380.9
Sub‐Saharan Africa 158.4 199.4
West Asia 44.5 22.8
Total 880.6 873.5
Financial analysis of the necessary investments toward extending sanitation access to
these populations was also undertaken by Rockström et al (2005), which are shown in
Table 2‐5.
Table 2‐5 ‐ Estimated household unit costs for urban and rural sanitation programs (Rockström et al. 2005)
UN Region Urban sanitation Urban household cost
(US$/hh)
Rural household cost (US$/hh)
East Asia 650 50
Eurasia 725 55
Latin America & Caribbean 1000 70
North Africa 900 65
Oceania 875 65
South‐East Asia 800 60
Southern Asia 440 40
Sub‐Saharan Africa 350 35
West Asia 1200 80
Weighted Average 774 46
The study was based upon an assessment of the unit household cost of appropriate
options to provide equivalent health and environmental protection to both urban and
rural households. Urban households therefore required significantly higher outlays to
provide for reticulation systems. The results indicate that urban sanitation programs
will cost 17 times rural installations on a per household basis, owing to the need for
reticulation services. The implication is that the financial sanitation challenge will
primarily be an urban one.
Challenges, respo
nses and
ways forw
ard
17
2.3 New sanitation responses The challenges identified in Chapter 2.2 clearly call upon a reconsideration of existing
and emerging sanitation options. This chapter introduces two key responses:
distributed sanitation, a reconsideration of scale; and ecological sanitation, a
reconsideration of process.
2.3.1 Distributed sanitation: matching system scale to urban density
There is a growing recognition that the infrastructure systems of the future will be
markedly different from the centralised systems that dominate today, and that
decentralised systems at the dwelling, cluster and neighbourhood scales will have an
increasing role. The distributed infrastructure approach is a response to this
recognition and implies an acceptance of a broader range of system scales to match
the social, economic and ecological conditions.
According to Berry et al (2004) the principal drivers for this shift in approach include:
• System limitations: the capacity of systems subject to increasingly apparent
ecological constraints is currently being outstripped by growing demand for
water, food, sanitation, energy, and transportation services, requiring more
sophisticated resource application (e.g. water‐efficient appliances) and
localised, synergistic supply systems (e.g. co‐generation).
• Increasing risks: the pace of change in demographic, social, technological and
economic conditions is a source of new levels of uncertainty, requiring more
responsive and scaleable service delivery models (e.g. modular sanitation).
Challenges, respo
nses and
ways forw
ard
18
Decentralised sanitation systems in particular have been increasingly recognised as a
cost‐effective alternative particularly in low density developments. This shift has been
driven by a recognition of a number of benefits, most comprehensively catalogued by
Pinkham et al (2004). These include:
• Improved resource efficiency: By avoiding the capital and operational expenses
of large re‐distribution networks, the distributed systems may provide cost‐
effective and efficient reuse of water and nutrients at the dwelling and
neighbourhood scale, yielding dual benefits in avoided water, fertiliser and
tertiary treatment costs.
• Avoided infrastructure: Where appropriate, the application of a more
decentralised system may yield avoided diseconomies of scale in collection
systems such as large diameter pipes and pumping, which often comprise the
majority of capital costs.
• Incremental investment: The outlay of distributed infrastructure can often be
more effectively scaled to match the capacity and treatment needs of the
community needs as they arise. This provides dual benefits in terms of reduced
financing costs and capacity to forecast risk.
Such systems have been enabled by new wastewater technologies including modular
textile filter package treatment and settled sewerage systems (Orenco Systems 2007;
Otis & Mara 1985; West 2003; White 2004).
Distributed systems therefore present significant opportunity to yield cost savings and
bringing the challenge of universal sanitation within reach, however in order to
address impending resource depletion and a changing climate, an entirely new
sanitation paradigm is called upon.
Challenges, respo
nses and
ways forw
ard
19
2.3.2 Ecological sanitation: closing the resource loop
Ecological sanitation (or EcoSan) is an ecosystems approach toward designing human
waste systems capable of sustaining or restoring the health and productivity of both
the people and the land upon which they depend (Esrey et al. 2000).
According to Esray et al (2000) the chief objectives of EcoSan are to:
• promote health and prevent disease;
• conserve water and protect water quality; and
• recycle nutrients and organics and preserve land fertility.
The process of ecological sanitation is interpreted to comprise three main stages
(Rockström et al. 2005).
• Containment: source separation is promoted to maximise the reuse potential
of the various waste streams and to ensure hazardous excreta are isolated.
• Sanitisation: hazardous excreta are sanitised by a combination of dessication,
raised pH, raised temperature, or microbiological competition.
• Recycling: the sanitised waste streams are restored to the land for productive
reuse.
Figure 2‐8 depicts an idealised process flow diagram of an ecological sanitation system.
Figure 2‐8 ‐ Process flow diagram of a dry ecological sanitation system (Rockström et al. 2005)
Challenges, respo
nses and
ways forw
ard
20
A urine diversion toilet is applied to maintain urine and faecal excreta streams
separately: the urine, which is sterile and contains the greater share of nutrients, is
diverted directly to agriculture as a liquid fertiliser; while the faeces, which are
potentially pathogenic and contain the greater share of organic material, are either
dessicated or composted and then diverted to a second composting process and
applied in agriculture as a soil conditioner. The agricultural yield is returned to the
kitchen thus completing the nutrient cycle. Greywater from baths and laundry,
together with the pre‐treated greywater from the kitchen is treated in a constructed
wetland and returned to the local surface and groundwater system. These storages,
together with rainwater, provide a local water source for the household, thus
completing the water cycle.
While interpretations vary as to whether water‐borne systems are strictly EcoSan
(Esrey et al. 2000; Kvarnström et al. 2006; Mara et al. 2007; Rockström et al. 2005), a
number of water‐borne systems also present some promise for resource oriented
sanitation.
Water‐borne ecological sanitation or wet conservancy systems apply anaerobic
digestion of combined ‘brownwater’ together with any other available biodegradable
waste to yield treated high nutrient effluent and biogas for energy. The settled solids
may be intermittently removed and co‐composted (FAO & CMS 1997). Given sufficient
organic input fuel and the right operating conditions, these systems capture a greater
degree of influent nitrogen for beneficial reuse than dry systems and present an
opportunity for a net reduction in greenhouse gases.
While not a rigorous inventory of options, this review demonstrates that more
ecologically sustainable sanitation futures are technically viable. However new
approaches are called upon to demonstrate that such options are economically cost‐
effective, which is the subject of the following section.
Challenges, respo
nses and
ways forw
ard
21
2.4 New decisionmaking approaches New methods and tools are called upon to aid decision‐making in favour of sustainable
sanitation. The following section introduces two key methods toward understanding
the challenges identified in Chapter 2.2, material flow analysis, a method of optimising
resource fluxes and least cost planning, a method of optimising costs.
2.4.1 Material flow analysis: optimising resource flows
Fundamental to comprehending the challenges of ecological sanitation, material flow
analysis (MFA) or substance flow analysis (SFA) is a process toward understanding the
exchanges of materials (and energy) between an economy and its ecology. The process
is summarised below (adapted from Brunner & Rechberger 2003).
1. Select the substance: choose a set of appropriate indicator
elements or compounds that may be traced through the
system in a unique, identical form;
2. Define the system: determine the spatial and temporal
boundaries consistent with the scope of the project;
3. Identify the relevant flows, stocks and processes: trace the
key flows of goods within and across the system boundaries;
4. Determine the mass flows, stocks and concentrations: assign
substance intensities to the flows of goods;
5. Assess the total material flows and stocks: aggregate the
substance fluxes to form a net flows and stocks of the
substance;
6. Analyse uncertainty: undertake sensitivity analysis to
quantify the uncertainty of the method; and
7. Document the results: report the methods and results
transparently.
Challenges, respo
nses and
ways forw
ard
22
Resource intensities: drawing parallels with end use forecasting
While by convention, material flow analyses are limited to an analysis of a single
substance (e.g. water) (Brunner & Rechberger 2003), the methods developed may be
applied toward analysing the urban metabolism holistically. This may be achieved by
modelling the behaviour and appliance stock associated with end uses (e.g. bathing,
dishwashing or clotheswashing) and assigning resource intensities (e.g. water, energy
or nutrients) to those activities as described by Baumann, Boland & Hanemann (1997).
The resulting end use formula implies the following:
Equation 1 – The basis of end use analysis (originally adapted from Baumann, Boland & Hanemann 1997)
))Re(%((1Re1
Type
n
source
n
TypeTypeTypeActivity nsitysourceInteApplianceUsagecUnitDemographiFlow ∑∑
==
×××=
Practical application: a material flow analysis of water and phosphorus
Although a number of studies have attempted to characterise the material flows
associated with sanitation systems (Belevi 2002; Jacobs & Haarhoff 2004; Motangero,
Nguyen & Belevi 2004; Tangsubkul, Moore & Waite 2005), an rigorous investigation
into the sustainability of a micro‐catchment in Harare, Zimbabwe, provides a useful
practical reference (Gumbo 2005).
The first stage of the study involved the identification of the flows of water and
nutrient within and across the system boundary. The result was a series of models
describing the interactions of three sub‐systems: a natural water balance, the
household nutrient and water balance, and an agriculture nutrient balance. The
system models describing the household nutrient and water balance have been
included as Figure 2‐9 and Figure 2‐10.
Challenges, respo
nses and
ways forw
ard
23
a, b, c, d – water use fractions determined through observation (-)
e – fraction of grey water diverted onto land for garden irrigation purposes (-)
Ec – water excreted by the human population mainly through respiration and perspiration (Wc – Wy)
Eg – evaporation from drying-out of laundry material and kitchen utensils
f – fraction of grey water evaporating either directly from laundry or after dripping to the ground surface (-)
h - fraction of storm water entering the foul sewer system conveyed as part of municipal sewage (-)
Q – discharge (Q = Qs + Qg)
Sa – storage in the atmosphere
Ti – transpiration arising from garden irrigation using municipal water
W – municipal water supply normalised to the catchment area
Wb – brown water generated from household activity related to toilet flushing after defecation
Wc – municipal water consumed by population either directly or contained in ingested food products
Wg – grey water generated from activities related to nourishing and cleaning (kitchen, bathroom and laundry)
Wi – municipal water used for garden irrigation (Wi = aW + eWg)
Wms – municipal sewage water, which is a combination of yellow, black, and proportion of grey and storm water
conveyed through a pipe to a sewage treatment plant (Wms = (1-e-f)Wg + Wb + Wy + hQ)
Ws – foul sewage or ‘black water’ which is a combination of yellow, brown, and a proportion of grey water (Ws =
Wms – hQ). This corresponds to dry weather flow.
Wy – volume of yellow water excreted by an equivalent adult population per month normalised to the micro-
catchment area (A) Figure 2‐9 ‐ System model of household water balance (Gumbo 2005)
Challenges, respo
nses and
ways forw
ard
24
e – proportion of grey water diverted onto land for garden irrigation purposes (-)
P – phosphorus flux as P in kg/month
Pb – brown water P-flux emanating from toilet flushing of human faecal material
Pfb – food and beverage P-flux reaching the household subsystem i.e. imported from outside microcatchment
and some produced within
Pg – grey water P-flux emanating from activities related to nourishing and cleaning, taken as equal to Psd
Ple - leaching P-flux due to percolation and groundwater flow
Pms – municipal sewage P-flux, which is a combination of yellow, black, and proportion of grey and storm water P-
fluxes conveyed through a pipe to a sewage treatment plant (Pms = (1-e)Pg + Pb + Py + h(Psr + Ple)
Ps – foul sewage or ‘black water’ P-flux which is a combination of yellow, brown, and proportion of greywater P-
fluxes (Ps = Pms – h(Psr + Ple). This corresponds to the P-flux during dry weather
Psd – soap and detergent P-flux reaching the household subsystem
Psr - surface runoff P-flux dissolved in storm water (elaborated in the next section)
Psw – organic solid waste P-flux derived from household activities and local vegetation growth and dieoff derived
from the agricultural subsystem
Py – yellow water P-flux derived from urinary excretion
q′ – proportion of organic solid waste which is either deliberately composted or is uncollected and end up being
manure on agricultural land
Figure 2‐10 ‐ System model of household phosphorus balance (Gumbo 2005)
Challenges, respo
nses and
ways forw
ard
25
The analysis then involved the measurement of water and nutrient fluxes per capita to
build up aggregate resource flows, also summarised below.
Figure 2‐11 – Personal water and nutrient inputs and outputs (Gumbo 2005)
Such models provide a useful means toward understanding the flows of water and
nutrients in an urban system. However to inform planning decisions the model must
also be capable of optimising the system financially. An appropriate costing approach
is therefore called upon, which is the subject of the following section.
Challenges, respo
nses and
ways forw
ard
26
2.4.2 Least cost planning: optimising financial flows
Least cost planning (LCP) is a method of assessing the least cost suite of options for
meeting a set of measurable objectives on the basis of their relative incremental costs.
A four‐stage model describing the process is summarised below (Mitchell et al. 2007).
1. Framing the study
Define the objectives
Describe the system
Adopt a specific cost perspective
Define key economic parameters
Determine the treatment of externalities
2. Characterising the study
Develop a water balance model
Specify the base case
Define a broad range of options and develop alternative
system configurations
3. Identifying and specifying costs
Specify the costs to include
Specify avoided costs and benefits to include
Specify and quantify externalities
4. Analysing and reporting incremental costs
Compare options using discounted cash flow analysis
Consider uncertainty; conduct sensitivity analysis
Document the analysis
Challenges, respo
nses and
ways forw
ard
27
The benefits of this analysis include (Mitchell et al. 2007):
• a balanced comparison of the complete set of options is enabled by
outcome‐based objectives;
• the full impact of projects is accounted by full life‐cycle costing and
intentionally defined system boundaries;
• economically optimal and financially viable solutions are revealed by
analysis of both stakeholder and societal cost perspectives;
• externalities and uncertainties are transparently treated; and
• stakeholders are actively engaged in all stages of the process.
Mitchell et al (2007) summarise the following core principles that form the foundation
of the method.
Economic and financial cost perspectives: An economic cost perspective entails
broadening the boundaries of the financial analysis to the whole of society. Transfer
payments (e.g. customer service fees) therefore cancel out while externalities are
accounted where possible. Economic cost‐effectiveness is therefore the most
appropriate driver toward providing efficient resource management decisions, while
financial cost‐effectiveness from all cost perspectives must be preserved by
renegotiation to ensure financial sustainability.
outcome vs output driven approach: Myriad options for achieving the same outcome
are revealed by reframing the challenge as supplying a service, rather than a
commodity. In so doing, more cost‐effective solutions may be drawn. Three corollaries
of this new approach include:
• dissociating scale of infrastructure with scale of demand
• focusing on the outcome rather than the capacity or volume
• matching the product to the service (i.e. the water quality cascade)
Systems thinking: consider all relevant elements of a system and be conscious of
connections and emergence.
Challenges, respo
nses and
ways forw
ard
28
Life cycle costs: account for the full life of assets including the aquisition, installation,
operation, maintenance, refurbishment, decommission, and replacement.
Incremental cost: account for costs and avoided costs relative to a base case. This will
include avoided or delayed capital expenditures and avoided operating costs. Use
levelised incremental cost where possible to reflect changing yield over time (i.e. NPV
of costs / NPV of yield)
Externalities: account for impacts that are beyond the actual or avoided costs to key
stakeholders. Appropriate responses include the following:
• Direct monetisation: use surrogate markets to attribute proxy values to
externality e.g. GHGe accounting
• Goals and limits: ensure all options meet a minimum standard or objective e.g.
nutrient loading
• Qualitative assessment: stakeholders rank suite of options against defined
criteria using deliberative process e.g.
Time value of money: account for the time value of money by reporting the time in
which costs will fall and making comparisons on the basis of discounted cash flow
analysis.
Risk, uncertainty, accuracy and precision: note uncertainties, map their associated risk
profiles and levels of uncertainty and then quantify and explicitly manage risks
Transparent reporting: document the analysis in a manner consistent with its easy
interpretation and replication.
Financial intensities: drawing parallels with material flow analysis
Least cost planning may be interpreted as an extension of material flow analysis and
end use modelling toward optimising both resource and financial efficiency. This is
achieved by assigning a series of financial intensities (or costs) to each of the
interventions and using this as a basis for prioritising those options yielding the lowest
cost per resource supplied or avoided. The most appropriate measure for this
comparison is the total (or economic) resource levelised cost (Fane, Robinson & White
2003), which is equivalent to the present value of costs divided by the present value of
Challenges, respo
nses and
ways forw
ard
29
additional or avoided resource capacity. Following an assessment of the economic
least cost alternative, the impact upon each stakeholder may in turn be assessed and
managed accordingly (e.g. by negotiating the transfer payments between
stakeholders).
Practical application: an integrated supplydemand planning model for water
Itself a tool to facilitate LCP, the integrated Supply‐Demand Planning model (iSDP)
model was developed by the Institute for Sustainable Futures in collaboration with
water utilities across Australia as an integrated water demand forecasting and options
analysis tool.
The model is structured around a central database drawing from a suite of
spreadsheets (or data sources), including:
• Scenarios, which link each region to its end use group and associated options,
and contain the generic study parameters including the discount rate;
• Regions, which contain data characterising the study area, including the
population, number of dwellings or properties, and the baseline system yield
(i.e. a time series of water supply capacity) associated with the study area;
• End uses, which may constitute whole sectors (e.g. the industrial sector), or
specific activities (e.g. residential bathing), and serve to calculate the
component baseline demand based upon a series of model assumptions and
analytical algorithms; and
• Options, which may constitute supply‐side interventions (e.g. an additional
reservoir) or demand‐side interventions (e.g. a showerhead program), and
serve to calculate the augmented or avoided system capacity and the
associated costs.
Challenges, respo
nses and
ways forw
ard
30
The model has facility for each of the assumptions embodied within the datasources to
be accounted and traceable by providing an assumptions reporting fields associated
with each assumption parameter within the database and an in‐built reference library
for ease of access by the reviewer.
The data sources then feed into a sophisticated demand forecasting and options
assessment engine, which effectively aggregates each of the end use components to
project a time series of baseline water demand in the study region. This may then be
compared to the baseline system yield to infer the time period beyond which demand
exceeds supply. Figure 2‐12 depicts a screenshot of a comparative forecast, which is a
typical output for undertaking such an analysis.
Figure 2‐12 – Screenshot from iSDP model demonstrating the supply demand forecast
As may be seen in this hypothetical example, the yield from the supply system reduces
with time, owing to the impact of climate change upon storages; whereas baseline
demand is projected to increase to outstrip supply by the year 2020.
Challenges, respo
nses and
ways forw
ard
31
This projection may then be modified by the addition of various suites of options which
serve to either lower the demand projection or raise the system yield projection.
Figure 2‐13 shows a typical example.
Figure 2‐13 – Screenshot from iSDP model demonstrating the impact of options
Note the system yield jumps considerably in the year 2012, potentially owing to a
supply augmentation such as a new reservoir or seawater desalination plant. Similarly
demand management options have been applied to reduce the demand forecast
considerably. The combined impact is an assessment that the new options will provide
sufficient capacity to meet demand beyond the year 2030.
Challenges, respo
nses and
ways forw
ard
32
The model also provides facility for options ranking and analysis based on a series of
cost parameters, which are based on the modeller’s input. For example, a step chart
may be provided, which ranks the costs associated with each option and positions
them along the x‐axis based upon their cumulative system capacity as shown in Figure
2‐14.
Figure 2‐14 – Screenshot from iSDP model demonstrating the ranking of options based on a step chart
This output provides excellent data on both the yield of each option and its associated
unit cost per kilolitre of capacity avoided or augmented. The area below the curve to
the left of a given yield is therefore equivalent to the present value of the least cost
alternative.
The iSDP model therefore provides a valuable tool towards undertaking
supply/demand forecasting and options analysis.
0
200
400
600
800
1000
1200
0 20,000 40,000 60,000 80,000 100,000 120,000
Tota
l Res
ourc
e Le
velis
ed C
ost (
c/kL
)
Yield [6/30/2020]
OPTION STEP DATA CHART
Develop
men
ts to
the metho
d and supp
ortin
g tools
33
3 Developments to the method and supporting tools
Having reviewed the current planning challenges, and identified the responses,
analytical methods and tools available toward their resolution, this chapter describes
the key developments this project has contributed to sanitation planning and the
wider integrated resource planning methodology and its associated tool, the iSDP
model.
3.1 Adapting the method The assessment methodology described by Mitchell et al and identified in section 2.3
was considered to be the most robust basis for the assessment process. The method
applied in this dissertation is therefore a generalised adaption of this method, and is
summarised as follows:
1. Establish the framework: Define the objectives of the analysis and specify the
economic criteria by which the alternatives will be assessed
2. Identify the system: Define the system boundaries, identify the components,
and model their exchanges within and across the system boundaries
3. Specify the base case: Inventory the suite of options available toward meeting
the objectives and estimate their incremental resource and financial impacts
4. Identify the options: Inventory the suite of options available and estimate their
incremental resource and financial impact
5. Specify the alternatives: Configure the options within the resource balance to
identify a suite of alternative sets of options that meet the study objectives
6. Analyse the costs: Apply discounted cash flow analysis to each alternative and
assess both the economic cost and the financial cost borne by each stakeholder
7. Analyse uncertainty: Undertake analyses to quantify uncertainties in the
resource balance, cost estimates and cost analysis
8. Evaluate the results: Review and document the process and outcomes of the
analysis to ensure the findings are transparent and accountable
Develop
men
ts to
the metho
d and supp
ortin
g tools
34
Peer review
The adapted process, including a detailed proposal of how the analysis could be
applied toward sanitation planning was presented for review as a presentation at the
International Water Association Specialised Conference on Small Water and
Wastewater Systems in Coimbatore, India, in February 2008. The paper, which was
subsequently included in the conference proceedings, has been included as Appendix
D.
In order to facilitate an open discussion of the research, the author arranged for a
seminar to be held with a group of delegates to discuss the challenges of assessing the
costs of distributed and ecological sanitation systems. While the research received a
positive response, the attendees discussed some key challenges of moving the project
forward. These included:
• Deciding upon a suitable boundary for the analysis
• Establishing a cost associated with both the discharge of nutrients to rivers and
the associated depletion of finite resources.
• Addressing the fundamentally different institutional arrangements associated
with distributed systems.
These research questions among others have been addressed in Section 3.2.
In addition to the conference proceedings, several field studies were conducted with
advocates of alternative sanitation systems. The findings have been documented as
Appendix E.
Develop
men
ts to
the metho
d and supp
ortin
g tools
35
3.2 Extending the supporting tool Having reviewed the core concepts and processes underlying the assessment process
presented by Mitchell et al, several key challenges emerged in extending the approach
to sanitation planning, these included:
1. How can nutrient discharges and greenhouse emissions be accounted within the
cost analysis?
2. How can both the hydraulic and nutrient load of sanitation systems be
accounted?
3. How can the boundaries of wastewater treatment systems be accurately
defined?
4. How can the incremental resource impact and costs of sanitation options be
modelled?
5. How can the financial impact of options to various stakeholders be accounted
transparently?
6. How can the inherent uncertainty associated with forecasting be accounted?
This section describes the extensive process of review and extension of the existing
iSDP model toward assessing alternative sanitation futures.
3.2.1 Establishing the framework: nutrient and greenhouse valuation
How can nutrient discharges and greenhouse emissions be accounted within the cost
analysis?
Accounting for nutrient discharges
The discharge and loss of nutrients is a significant externality for consideration in the
assessment of alternative sanitation systems as described in Section 2.2.1. Mitchell et
al (2007) present three paths toward accounting for externalities:
• Direct monetization, whereby a surrogate market is applied to produce a proxy
value
• Goals and limits, whereby any alternatives applied must be deemed to satisfy
criteria, either defined in terms of regulatory limits or sustainability targets.
Develop
men
ts to
the metho
d and supp
ortin
g tools
36
• Qualitative assessment, whereby appropriately deliberative or participatory
processes are applied to rank options on the basis of a set of agreed criteria
For the purpose of this analysis it was preferable to include (or internalise) the
externality within the cost analysis if possible, thereby incorporating the costs within
the key decision‐making parameter of the process. However this gave rise to the
challenge of accounting for a social and environmental cost that only typically becomes
tangible once the damage has occurred (therefore requiring a costly clean‐up project).
Following a review of the literature, a key regulation was discovered describing a
system of load‐based licensing in NSW whereby a direct financial incentive for
pollution minimization is encoded within the pollution licensing regime (Protection of
the Environment Operations (General) Regulation 1998). By applying the algorithms
and tables included in the legislation within the decision‐making tool, the economic
cost of nutrient disposal was internalized as either a financial cost to the polluting
industry within NSW, or as an equivalent economic cost in other areas yet to receive
load‐based pollution charging‐ which is dependent upon the assumption that the
current legislation captures the full economic cost.
Equation 2 ‐ Algorithm for calculating the cost of nutrient discharge to water
If AL < FRT
10000÷×××+×= CZPWPFUALNoAFUAFUFee
If AL > FRT
10000)2( ÷×××−+×= CZPWPFUFRTALNoAFUAFUFee
Where AFU = Administrative Fee Unit
AL = Assessable load
FRT = Fee rate threshold
PFU = Pollutant Fee Unit
PW = Pollutant weighting
CZ = Critical zone factor
Develop
men
ts to
the metho
d and supp
ortin
g tools
37
Accounting for greenhouse gas emissions
Greenhouse gas emissions have thus far been accounted within end use models as
either an external account in terms of tonnes carbon dioxide equivalent, or as an
externality in the form of a proxy greenhouse market price. However a more accurate
costing projection was deemed appropriate in light of the recent commitment by the
Australian Federal Government to establish a greenhouse gas emission trading
scheme.
Although Federal Government modelling for the establishment of the scheme is still in
progress, a consortium of Australian state governments represented by the National
Emissions Trading Taskforce recently commissioned an investigation into the effects of
such a scheme (Allen Consulting Group 2006). As part of this research, extensive
modelling was undertaken to establish the economic impacts of the proposed scheme
to both retailers and utilities (MMA 2006). One outcome of this modelling was a
projection of the value of greenhouse gas emission permits shown in Figure 3‐1.
Figure 3‐1 ‐ Projection of greenhouse gas equivalent permit prices for Australia (MMA 2006)
This projection was applied to cost all direct greenhouse gas emissions accounted
within the model.
$0
$5
$10
$15
$20
$25
$30
$35
30/6/2010
30/6/2011
30/6/2012
30/6/2013
30/6/2014
30/6/2015
30/6/2016
30/6/2017
30/6/2018
30/6/2019
30/6/2020
30/6/2021
30/6/2022
30/6/2023
30/6/2024
30/6/2025
30/6/2026
30/6/2027
30/6/2028
30/6/2029
30/6/2030
GHG emission
permit price [$
/ton
ne CO2‐e]
Time [date]
Develop
men
ts to
the metho
d and supp
ortin
g tools
38
In order to account for indirect emissions associated with electricity consumption, it
was necessary to translate the permit prices projection into equivalent projected
increases in residential and industrial retail prices. As part of the same modelling,
MMA created future scenarios projecting the shift in share of alternative generation
options. This was used as a basis for projecting the emission intensity of electricity
generation, and consequently the equivalent increase in wholesale electricity prices.
Based upon the assumptions of MMA, this increase was converted into changes in
retail prices borne by residential and industrial customers, shown in Figure 3‐2.
Figure 3‐2 ‐ Projection of retail energy prices by customer type
The projections indicate industrial customers likely stand to suffer the greater
proportionate increase in costs, with industrial prices currently only marginally higher
than wholesale prices.
As this projection accounted for the costs associated with an emissions trading
scheme, this projection of retail electricity costs was assumed to have internalised the
indirect economic cost associated with the emission of greenhouse gas emissions.
$‐
$0.05
$0.10
$0.15
$0.20
$0.25
$0.30
30/6/2010
30/6/2011
30/6/2012
30/6/2013
30/6/2014
30/6/2015
30/6/2016
30/6/2017
30/6/2018
30/6/2019
30/6/2020
30/6/2021
30/6/2022
30/6/2023
30/6/2024
30/6/2025
30/6/2026
30/6/2027
30/6/2028
30/6/2029
30/6/2030
Electricity cost [$
/kWh]
Time [date]
Residential retail price Industrial retail price
Develop
men
ts to
the metho
d and supp
ortin
g tools
39
3.2.2 Identifying the system: multiresource accounting
How can both the hydraulic and nutrient load of sanitation systems be accounted?
In order to enhance and extend the iSDP model such that it was capable of accurately
accounting for the key resource flows in sanitation systems, the research project made
several contributions to the existing iSDP end use model, which formed the basis for
the system analysis. These were:
• Identifying a relevant conceptual model for analysing the sanitation
system;
• Restructuring toward service‐based end uses;
• Reviewing and revising the model assumptions;
• Indexing all region‐specific data sets; and
• Revising the appliance stock models.
• Quantifying dynamic peaking of sewage flows
Identifying a conceptual model
The first stage of adapting the model for analysing the sanitation system involved
identifying a conceptual model characterising the key resource flows. System models
developed by Gumbo (2005) were adapted for this purpose to provide the conceptual
system model provided in.
Develop
men
ts to
the metho
d and supp
ortin
g tools
40
Figure 3‐3 ‐ Conceptual model of the sanitation system
Restructuring toward servicebased end uses
The pre‐existing end use model constituted the following end uses to characterise
indoor demand:
• Clotheswashing machines
• Dishwashing machines
• Showers
• Toilets
• Indoor miscellaneous (baths, hand basins, kitchen sinks, laundry troughs and
toilets leakage)
The miscellaneous components therefore constituted a large component of demand
that previous option studies had largely ignored, while the structure failed to project
key appliance interactions.
Return Options
Wastewater System (Centralised / Distributed)
PFaeces + PUrine + WGreywater + PGreywater (+ WBlackwater) WWastewater + PWastewater (+
PSoilConditioner)
Municipal Solid Waste System
PBiowaste PLandfill (+ PSoilConditioner)
Urine diverting / composting systems
PFaeces + PFoodWaste (+ PUrine) PSoilConditioner + PLosses
Losses
‐ PLandfill
‐ WWastewater + PWastewater
‐ WAgriculturalRunoff + PAgriculturalRunoff
Injections
‐ Wimported
‐ PMineralFertiliser
ServicesInputs: Water +Food + Soaps & Detergents
Activities: Bathing + Clotheswashing + Dishwashing + Toilet
Outputs: Faeces + Urine + Greywater + Solid Waste (+ WBlackwater)
Water Balance: WReticulated (+ WHarvested) WServices (– DM) (WBlackwater) + WGreywater + WOutdoor
Phosphorus Balance: PFood (+ PDetergent) PServices PFaeces + PUrine +PGreywater +PFoodWaste
Supply options
(not considered in this study)
Water supply (dams / recycling etc)
Wcollected (+ Wreclaimed) + Wimported Wreticulated + Wagriculatural
Food supply (conventional/ low‐tillage farming etc)
WAgricultural + PMineralFertiliser (+ PSoilConditioner + PUrine) PFood + PSoilStock + PAgriculturalRunoff
Non‐revenue water
Exfiltration
Infiltration / inflow
Develop
men
ts to
the metho
d and supp
ortin
g tools
41
As indicated in the conceptual diagram above, a key development that emerged from
the literature review was a new focus upon ultimate resource services (e.g. bathing),
rather than appliances (e.g. baths). This implied recognising that using a bath is often
interchangeable with showering, and using a dishwashing machine is interchangeable
with manual dishwashing in a kitchen sink.
The end uses constituting the iSDP model were therefore subjected to a significant
restructure to align along the following water‐related activities.
• Clotheswashing: constituting clotheswashing machines and troughs
• Dishwashing: constituting dishwashing machines and kitchen sinks
• Bathing: constituting showers and baths
• Toilet: constituting toilets flushing and leakage
Each of the activity‐based end uses applied algorithms to relate the use of its
consistent appliances. For instance, Roberts (2004) identified an important relationship
between the age profile of the study population and their associated frequencies of
using a bath or a shower. That is, the age of 12 years was observed to be a key
threshold beyond which water users shift from relatively infrequent use of a bath to
daily use of a shower. Similarly the Melbourne appliance usage survey identified the
relative frequency of manual dishwashing activities in households with and without a
dishwashing machine.
The outcome of this revised focus upon activities therefore allows the modeller to
identify the impact of key relationships that were previously ignored. For example, the
increased ownership of dishwashers, though a minor end use, was projected to
significantly impact upon the much more significant sink consumption, while the aging
population was found to significantly increase bathing demand over time, with an
associated shift in the use of showers as a replacement for baths.
Develop
men
ts to
the metho
d and supp
ortin
g tools
42
Revising and extending the model assumptions
Following a review of the latest revision of the iSDP model, it became apparent that
the majority of assumptions within the model with relation to water flows had not
been documented, while those documented assumptions were no longer current. A
thorough literature review was therefore necessary to identify the best available data.
Key sources adapted for this study included a rigorous appliance stock and usage
pattern survey conducted by Yarra Valley Water in Melbourne (Roberts 2004), a
subsequent end use measurement study (Roberts 2005), the most recent Australian
census and survey information (ABS 2007b), and a state‐wide market survey of
appliance sales and mean performance (GFK 2006). During this revision process each
assumption was documented using the model interface and all references were
attached as soft copies within the model, thus providing a traceable means of
supporting assumptions consistent with the transparency concepts outlined by
Mitchell et al.
Also, where region‐specific data was available this information was collated in full
within the model. The resulting data tables, with each row representing a state of
Australia, were indexed within the model based upon a single parameter drawn from
the central sheet defining the study region (i.e. if state is NSW, State = 1 etc).
The next stage involved extending the existing model toward accounting for nutrients‐
an additional resource stream. A completely new suite of parameters were therefore
incorporated into the model describing the various nutrient streams associated with
excreta, soaps and detergents. Key sources included an algorithm for excreta
associated nutrients (Jönsson & Vinnerås 2003) and an investigation of the
composition of detergents (Patterson 2007) (also referred in 0). These parameters
were then grouped within fields capable of being drawn by the model as a forecast.
In addition, several new parameters were added to characterise the augmented or
avoided nutrient capacity of the system. It is worthy to note that this development
provides significant scope for subsequent integration of resource models (further
discussed in Section 5).
The resulting models therefore constitute a series of end use assumptions and
algorithms (which are documented in full as 0) that are considered to be the most
Develop
men
ts to
the metho
d and supp
ortin
g tools
43
comprehensive and thoroughly documented water end use data available in Australia
as of June 2007. In addition, it is anticipated that the innovative indexing structure will
facilitate the use of this model across Australia for considering both water and
sanitation strategies.
Revising the appliance stock models
Another significant revision involved the overhaul of all existing appliance stock
models, that is, those models simulating the sale and replacement of cohorts of
appliance stock toward modelling the shift toward newer appliance types (e.g. efficient
showerheads). This was important to the study in order to address a key research
question of how resource efficiency and source controls such as efficient showerheads
impact upon the capacity and cost of wastewater systems.
Figure 3‐4 depicts the typical structure of such a stock model array.
Figure 3‐4 ‐ Excel screenshot describing typical stock model array
The array has two dimensions, the horizontal dimension (or x‐axis), which represents
time, and the vertical dimension (or y‐axis), representing a series of annual stock
cohorts. The intersection of equal years in the array therefore represents the total
number of appliance stock purchased in that year. The cells to its right represent the
decay of that cohort over time, which is based upon either a log‐normal or normal
Cohorts of stock Decay over time
Develop
men
ts to
the metho
d and supp
ortin
g tools
44
distribution. The expired stock from each cohort is replaced by cohorts in later years,
resulting in an effective simulation of the change in appliance stock over time.
In the pre‐existing iSDP model, two end uses had applied cohort stock models: those
were clotheswashing machines and dishwashing machines. The time series of these
models were previously restricted to launch in the year 1990, owing to perceived data
gaps. This provided a bias in the stock model as, owing to the structure of such models,
the sales in the launch year must be equal to the current annual sales plus all pre‐
existing stock. A realistic mix of stock ages in the forecast is therefore only achieved if
the first cohort is sufficiently decayed by this time as a proportion of the total stock
mix.
The first stage of resolving this issue involved seeking out suitable demographic data
characterising historical demographic changes in Australia. A recently released
collation of historical data collections was applied toward this purpose (ABS 2006).
Having done so, the next stage involved developing an Excel add‐in tool to rebuild the
stock models.1 The code and interface were structured to provide functionality for user
input on key variables including the lifetime of stock, its standard deviation, the
commencement and completion years, the model type (i.e. either total stock or
component), and the function type (i.e. either a log‐norm or normal decay of stock).
The interface and code have been included as Appendix B.
The inputs for the new tool were then set to begin the model in the year 1960. This
year was chosen to allow the stock models sufficient time to create an adequate mix of
ages of appliance stock and, since some appliances such as toilets can last up to 40
years, a hindcast period of at least this duration was necessary. The new tool was then
executed to generate new stock models for both dishwashing and clotheswashing
machines with the extended hindcast period. The outputs of the clotheswasher stock
model are shown in Figure 3‐5.
1 This was achieved with the assistance of Kurt Forrester, who contributed the structure of the looping
code
Develop
men
ts to
the metho
d and supp
ortin
g tools
45
Figure 3‐5 ‐ Stock model output describing clotheswasher stock share by appliance type
The output indicates the impact of the current surge in front‐loading clotheswashing
machine sales upon the total stock mix is subject to a lag.
The next stage involved applying similar cohort stock model toward analysing the
change in toilet stock toward progressively efficient toilet models. This was deemed
necessary owing to an apparent inconsistency in the internal logic of the existing
model, which was effectively over‐estimating the replacement rate of toilet stock. Data
gaps were overcome using a calibration process which effectively modified the missing
parameters (appliance lifetime and standard deviation) to provide outputs consistent
with existing survey data (ABS 2007b).
The outputs of this model, shown in Figure 3‐6, demonstrate the basis of this
calibration.
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
60.0%
70.0%
80.0%
90.0%
100.0%
1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040
Prop
ortion
of total stock [%
]
Time [financial year ending]
Front loaders Top loaders
Front loaders (ABS 2005) Top loaders (ABS 2005)
Develop
men
ts to
the metho
d and supp
ortin
g tools
46
Figure 3‐6 ‐ Stock model output describing toilet stock share by appliance type
The algorithm is based upon demographic data for the study region, assumed stock
numbers per dwelling, and sales share information from 1980 to 2007. The unknown
parameters, the mean and standard deviation of the appliance lifetime, are calibrated
to match the stock mix outputs with existing survey data for the proportion of stock
that is ‘single flush’. These same parameters are then applied for the other toilet stock
cohorts, based upon the explicit assumption that the mean rate of replacement of
toilet stock is constant.
A similar process was then applied toward analysing the change in shower stock. This
provided the extra benefit of providing new outputs modelling the number of new
showerheads purchased each year. This implementation of the model presented an
entirely different data gap: that is, a characterisation of the change in the mix of
showerhead sales over time. To overcome this data gap, the shift was assumed to be
characterised by an sigmoid (or ‘S‐curve’), with the shape parameters fitted to the
survey outputs (ABS 2007b) using the same calibration tool applied towards the toilet
stock model. The outputs for the shower stock model are shown in Figure 3‐7.
0%10%20%30%40%50%60%70%80%90%
100%
Prop
ortion
of total stock [%
]
Time [date]
Single Flush Dual Flush 11/6L Dual Flush 9/4.5L
Dual Flush 6/3L Dual Flush 4.5/3L Single Flush (ABS 2007)
Develop
men
ts to
the metho
d and supp
ortin
g tools
47
Figure 3‐7 ‐ Stock model output describing shower stock share by appliance type
The outcome is a series of mechanistic models that serve to provide the best‐available
simulation of the shifts appliance stock within the region. This measure provided a
rigorous foundation for establishing the impact of baseline population and dwelling
growth upon the appliance stock mix, which is the subject of the following section.
Quantifying dynamic peaking of sewage flows
The flows of water within the wastewater stream are subject to significant diurnal and
daily dynamic effects (Friedler & Butler 1996). An investigation into the relationship
between the hydraulic flows of the various end uses and their consequent effect upon
hourly peak hydraulic capacity was therefore warranted.
In order to best apply the model assumptions, data was therefore required to
characterise the hourly time series demand associated with each end use. The best
means of obtaining such information is based upon high resolution metering of water
demand, as recently undertaken by Yarra Valley Water in Melbourne (Roberts 2005).
Figure 3‐8 shows an analysis of average hourly water demand per property by end use
for a summer logging period.
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Prop
ortion
of total stock [%
]
Year ending
Normal Efficient Efficient (adj. ABS 2007)
Develop
men
ts to
the metho
d and supp
ortin
g tools
48
Figure 3‐8 ‐ Average hourly profile of water demand per property by end use (Roberts 2005)
The peak hourly dry weather sewage flow was then inferred by dividing the hourly
flow rate for each end use during the peak sewage period (i.e. 8‐9am) by the average
daily flow for each end use. This resulted in a series of hourly peak factors associated
with each appliance, which could be multiplied by each end use flow to infer the
hourly peak dry weather flow.
Note that this analysis does not provide an accurate basis for assessing hourly peak
sewer flows as it fails to account for inflow & infiltration. Such an analysis would
require a more detailed understanding of the hydraulic flows associated with system
inflow, which are complex and still poorly understood (Howe et al. 2005). A sewage
network model recently developed by Melbourne Water may shed light on this subject
(Corbett & Chan 2007).
The analysis also does not account for the peak attenuation associated with increasing
populations samples (Tchobanoglous, Burton & Stensel 1991, pp. 149‐153). Further
research is therefore warranted.
Develop
men
ts to
the metho
d and supp
ortin
g tools
49
3.2.3 Specifying the base case: spatial control of regions
How can the boundaries of wastewater treatment catchments be accurately defined?
As identified by Mitchell et al (2007), the boundaries of the system under study must
be intentionally drawn to capture all major interactions with the existing system, often
in the form of avoided costs. It was judged that the most appropriate system boundary
for a wastewater system would often be defined by the wastewater catchment of each
centralised treatment plant. This unique boundary provided the challenge of providing
a suitable baseline demographic time series of population and dwellings for the study
area (i.e. the components of the iSDP model region sheet) as the boundary will most
often fail to overlap with census data collection or projection regions.
A means of achieving an accurate estimate of current population, dwelling and
dwelling mix was therefore first sought to provide a suitable launching point for the
demand forecast and hindcast. Since such detailed data was unavailable for the
specific study area, a spatially referenced data source was therefore sought to provide
a basis for subsequent analysis. This data was drawn from the current Census Basics
dataset (ABS 2002), which is a series of spatially referenced files based upon the most
resolved statistical unit (i.e. the statistical collector district).
The spatially referenced files were then analysed using MapInfo software, which is a
geographical information system analysis tool. A typical screenshot from the software
has been included as Figure 3‐9. On the right shows the map containing the spatial
boundaries for each collector district that has been zoomed in on Melbourne. To its
left is a table containing the full Census data for each record (i.e. the statistical
collector district). The boundaries of the wastewater catchment were overlayed upon
this map and a geo‐database query executed to aggregate all those records with
centroids falling within the boundaries of the catchment.
The output is an aggregated set of population and dwellings by type for the study area
that may be inputted to the region sheet of the model.
Develop
men
ts to
the metho
d and supp
ortin
g tools
50
Figure 3‐9 ‐ Screenshot from MapInfo interface demonstrating the spatial selection of Census data
Having established an accurate launch year, the next stage involves developing a
demographic projection of that data over the duration of the assessment, which in this
case was 50 years. Projections of population and dwellings were therefore sourced
from modelling from the various state planning authorities, provided as spreadsheet
time series for each statistical local area (NSW Planning 2007; Vic Planning 2007). This
data was then linked to the corresponding spatial files containing the statistical local
area boundaries, which were sourced from the Australian Bureau of Statistics (2007a).
This enabled a similar process to that described above; however, as the analysis was
undertaken upon relatively coarse regions, the absolute numbers were only used to
establish growth factors for application to the more precise Census estimate. A
screenshot of this analysis is included as Figure 3‐10.
Develop
men
ts to
the metho
d and supp
ortin
g tools
51
Figure 3‐10 ‐ Screenshot of MapInfo interface demonstrating the spatial selection of projection data
Finally it was necessary to hindcast the time series to 1960 to provide sufficient
launching time for the stock models (as described above). This was achieved by
application of a the work of a recent project collating historical censuses as electronic
spreadsheet files (ABS 2006). As the time series was that for the entire city it was
necessary to assume the study region had a similar historical growth pattern. The
historical growth rates were therefore hindcast from the current state established
above.
The result of the method is therefore the best available time series of population and
dwellings for the study area from the year 1960 to 2050. The method and database
produced may be applied to any region of similar scale in Australia and is therefore a
valuable contribution to integrated resource planning and the end use forecasting
approach.
Develop
men
ts to
the metho
d and supp
ortin
g tools
52
3.2.4 Identifying the options: new option models
How can the incremental resource impact and costs of sanitation options be modelled?
The project involved the development of three new option models: these were
• distributed sanitation‐ a small‐bore settled sewerage system (see 4.4.1);
• ecological sanitation‐ a system comprising urine diverting co‐composting toilets
and greywater reuse (See Section 4.4.2); and
The distributed sanitation option
The sanitation model presented a number of unique challenges compared to those
developed previously, principal of which involved developing a means for projecting
the avoided resources and financial flows associated with providing alternative
capacity to the sanitation system.
Instead of simply applying a unit rate for incremental avoided flows it was deemed
necessary to project both the baseline and the option component in parallel. This
involved drawing all flows of sewage, phosphorus and nitrogen from the baseline
engine. A proportion of those inputs were then diverted to the option resulting in two
streams: that proportion diverted to the distributed system, and that remaining
proportion to the central system. The diverted stream is then used as a basis for
estimating the necessary treatment capacity. Sufficient treatment plants were
assumed to be installed to provide capacity for average estimated flows three years
hence, based on an assumed modular capacity. Note no hourly peaking factor was
deemed necessary owing to the daily storage provided by the interceptor tanks. In
addition, owing to the fused connections and avoided manholes of the effluent sewer,
no inflow or infiltration was assigned. The output is shown in Figure 3‐11.
Develop
men
ts to
the metho
d and supp
ortin
g tools
53
Figure 3‐11 ‐ Model output demonstrating the matching cluster supply to demand
Note the ability of the option to closely respond to growth in demand as it occurs. This
benefit is further discussed in Section 4.4.1.
Having projected both the flows to the distributed sanitation system and the central
system a final step involved modifying the nutrient flows to reflect the reduction in
total nutrients associated with the treatment process. Prior performance efficiencies
were deemed to be sufficient for this purpose.
The incremental avoided resource flow was then inferred from the difference in
discharges between the baseline and the option scenario, thus providing a basis for
calculating the associated sewage discharge fee based on the algorithm described
above.
‐
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
30/6/2008
30/6/2010
30/6/2012
30/6/2014
30/6/2016
30/6/2018
30/6/2020
30/6/2022
30/6/2024
30/6/2026
30/6/2028
30/6/2030
30/6/2032
30/6/2034
30/6/2036
30/6/2038
30/6/2040
30/6/2042
30/6/2044
30/6/2046
30/6/2048
30/6/2050
Cluster sup
ply/de
man
d [M
L/a]
Time [date]
Capacity demand Capacity supply
Develop
men
ts to
the metho
d and supp
ortin
g tools
54
The ecological sanitation option
In addition to the model extensions for distributed sanitation, a key challenge for the
ecological sanitation option involved estimating the value of any recovered nutrients.
This was achieved by calculating an equivalent dose of artificial fertiliser per unit of
nutrient based on NSW Department of Primary Industries figures (Rose 2004), and
using the current local price of that fertiliser per tonne. Single superphosphate
fertiliser was applied as a proxy for phosphorus, while urea was adopted as a proxy for
nitrogen. The calculations were as follows:
Equation 3 – Formula for calculating the equivalent market value of plant nutrients
Therefore, assuming single superphosphate at 8.8% P, costing $262 per tonne:
Recycled P = $0.262 / 0.088
= $2.98 / kg
Assuming urea at 46% N, costing $410 per tonne:
Recycled N = $0.41 / 0.46
= $0.89 / kg
Note this calculation assumes fertiliser costs will remain fixed over time, however as
discussed in the literature review, prices are likely to rise, consistent with depletion of
oil, gas and phosphorus and greenhouse gas emission reductions. Such an analysis was
outside the scope of this study.
Another important underlying assumption for this proxy valuation is that urine‐based
fertilisers will be an acceptable substitute for synthetic fertilisers in the market. Such
substitution will be dependent upon acceptance within the agricultural industry,
however the recent popularity of organic farming is encouraging.
Develop
men
ts to
the metho
d and supp
ortin
g tools
55
Another important challenge was developing an estimate of the costs associated with
the cartage of urine and compost, and the avoided costs associated with diverting
organic waste from the municipal solid waste stream. This challenge was enhanced by
a deficit in experience with implementing large‐scale ecological sanitation programs at
the municipal level.
Wang et al (1998) undertook a similar study investigating the relative costs of
providing an additional organic waste collection for an inner Melbourne municipality of
approximately 186,000 households. The authors applied a sophisticated stochastic
simulation of the operational costs and environmental impacts associated with
municipal solid waste collections under different assumptions. Some of the key
outputs of this study were a unit financial cost and greenhouse burden per tonne of
organic and municipal solid waste.
The unit rates for the organic waste stream were therefore adopted as a suitable proxy
for the collection costs of a composted faecal and kitchen waste collection service. The
unit rates for the municipal solid waste collection service were applied within the
baseline as a potential avoided cost.
Note that the avoided costs associated with municipal solid waste collection may be
underestimated as the diversion of biodegradable wastes from the waste stream may
ultimately obviate the need for weekly waste collections.
For costing the urine collection stream, an estimate was drawn from a study into the
feasibility of a urine diverting composting toilet program, also in Melbourne (GHD
2003). This estimate should be taken was taken with some trepidation as the basis for
the assumption was not adequately documented in the report. Further research to
validate this cost is therefore warranted.
Develop
men
ts to
the metho
d and supp
ortin
g tools
56
3.2.5 Analysing the costs: financial flow analysis
How can the financial impact of options to various stakeholders be accounted?
As identified by Mitchell et al (2007, pp. 12‐13), it is important to account for both the
financial costs experienced by each key stakeholder, in addition to the economic costs
borne by society as a whole.
A key concept that emerged from the literature review was that of financial intensities.
That is, in addition to modelling the flows of resources through a system, by assigning
resource intensities to the various activities or end uses (i.e. the infrastructural
system), the flows of finance may be modelled by assigning financial intensities to each
of the options (i.e. the institutional system). Such flows are assigned to each of the
various stakeholders and may occur within the institutional system boundary, as
transfer payments between stakeholders, or across the institutional system boundary,
as true costs.
While previous revisions of the iSDP model addressed this requirement to a limited
extent, by specifying whether the cost was attributed to a stakeholder or society, there
was no functionality to balance transfer payments from one stakeholder to another.
The user therefore was forced to add two parameters, the fee borne by one
stakeholder, and the reception of that fee by the other stakeholder. The incremental
impact of transfer payments between stakeholders was therefore largely ignored.
The financial flow analysis concept was therefore applied to each option model. This
was achieved by developing an innovate structure to assign financial flows to
stakeholders as illustrated in Figure 3‐12.
Develop
men
ts to
the metho
d and supp
ortin
g tools
57
Figure 3‐12 – Screenshot of model showing financial flow analysis of options
Referring to the figure above, the lower rows represent the various financial flows
projected in the option. As in all iSDP sheets, the Column A represents the name of
each parameter. Column B contains a code used to assign that financial flow to a
stakeholder. That is, if C is the first letter, the cost is borne by the customer. If U is the
last letter, that cost is received by the utility as a transfer payment. If the last letter is
absent, the financial flow is transmitted beyond the system as a true economic cost or
financial loss from the institutional system. Above these rows, in rows 5 to 8 are the
various financial costs considered in this analysis. The time series applies an algorithm
to obey the logic above and is shown below.
=SUMIF($B$12:$B$22,"C*",C$12:C$22)‐SUMIF($B$12:$B$22,"*C",C$12:C$22)
Column B for these rows contains the present value of costs borne by the stakeholder.
The row above, row 3, contains the economic cost, which aggregates all financial costs
to effectively cancel out all transfer payments to reveal the true economic cost of the
option.
After first assessing the suite of options providing least economic cost (i.e. the least
cost to society as a whole), this revision allowed the model to be used to modify and
assess the balance of transfer payments associated with an option. The innovation
therefore allows the model to be used as a basis for renegotiating transfer payments
to ensure financial incentives direct all stakeholders to the most economically efficient
course of action.
Develop
men
ts to
the metho
d and supp
ortin
g tools
58
3.2.6 Analysing uncertainty: stochastic analysis
How can the inherent uncertainty associated with forecasting be accounted?
Options with which we have relatively little experience will have an inherently higher
degree of uncertainty than more traditional options. Similarly demand‐side options or
options of relatively smaller scale will each typically have a higher degree of
uncertainty than large‐scale supply‐side options (Mitchell et al. 2007, p. 24).
Uncertainty must therefore be explicitly accounted and managed in order to provide a
defensible platform for any subsequent decisions. According to Mitchell et al (2007, p.
25), this may be addressed by two principal measures, including:
• sensitivity analyses, which essentially involves assessing the relative sensitivity
of the model outputs to a proportional change in the input parameters.
• monte carlo simulation, which essentially involves assigning a probability
density function to each key input parameter and assessing the uncertainty of
the model outputs by repeatedly generating random outputs.
While previous iSDP models have applied sensitivity analyses to identify sensitive input
parameters, no monte carlo simulations have thus far been undertaken. Such an
analysis would prove useful to identify the inherent range of certainty associated with
the each baseline component or option, and therefore their respective forecasts.
Although owing to time constraints such functionality was not realised within the
model, this project sought to identify how such an analysis could be achieved.
The first stage of the conversion would involve replacing each primary input parameter
within the model with one of the functions defined below:
Equation 4 ‐ Excel formula for a uniformly distributed random variable
min
Equation 5 = Excel formula for normally distributed random variable
, ,
That is, each primary input of known mean and variance would be replaced by an excel
function to generate a new random variable (i.e. based on a specified probability
density function) for each execution.
Develop
men
ts to
the metho
d and supp
ortin
g tools
59
A macro would then iteratively recalculate each datasource spreadsheet and output a
list of random results as a spreadsheet. A simple visual basic macro to achieve this
purpose has been included as Appendix B.
The list could then be analysed to calculate the confidence interval on the model
outputs, based on the formula below.
Equation 6 ‐ Upper and lower control limits (Wittwer 2004)
95% 1.96 1.96√
where mu and s are the mean and standard deviation of the outputs, and
n is the number of iterations of the calculation.
The output of this analysis would be a forecast with three time series: the mean
forecast and its upper and lower control limits. Such a forecast could be generated to
characterise the uncertainty in both the baseline forecast (i.e. the baseline supply‐
demand balance) and the comparative forecast (i.e. the relative impact of options).
Develop
men
ts to
the metho
d and supp
ortin
g tools
60
Peer review
The key developments described above were reviewed during a seminar held at the
Institute for Sustainable Futures on the 7th May 2008. Attendees included Professor
Stuart White, an internationally respected expert in the application of end use
forecasting and least cost water planning; Andrea Turner, a water engineer with over
20 years experience in both water and sanitation engineering and project director for
the ACT iSDP model; Alexander Kazaglis, the consultant principally responsible for
developing the iSDP model for Melbourne; and Dr Simon Fane, who investigated the
assessment of distributed sanitation systems as part of his doctoral thesis.
The seminar served both to validate the research methodology, and to highlight
potential areas of further investigation. Some key issues discussed included:
• The importance of a distinction between green field or infill development
• The complexity of assigning an hourly peaking factor
• The issue of salt in wastewater and its impact upon reuse potential
• Apparent disagreement between the model outputs and a comparable material
flow analysis for Sydney
• How could the system be designed for say, a regulatory goal of 4X average wet
weather flow etc
• Whether or not a treatment plant has P removal will affect cost savings‐ need
to be clear on if this included in base case or not
• The significance of nitrogen removal as an avoided cost
• Opportunities for synergy with water and energy planning
Having extended the methodology and tools towards analysing the sanitation system,
it was necessary to validate and demonstrate the method and supporting tools by way
of a case study example, which is the subject of the following chapter.
A case stud
y im
plem
entatio
n
61
4 A case study implementation
The following sections describe how the framework is applied by process of a
hypothetical example. A generic overview of the actions associated with each stage
has been provided, followed by a description of how the assessment was applied to
the case study context.
The Australian city of Melbourne was chosen for this analysis owing to both the rigour
of several pre‐existing stock usage and end use measurement surveys (Roberts 2004,
2005), and the availability of detailed forecasts for both population and dwellings (Vic
Planning 2007). The case study therefore provided opportunity to test the baseline
resource projections of the model against actual measurements of influent flow and
nutrient concentrations (Melbourne Water Corporation 1998; 1999; 2000; 2001; 2002;
2003; 2004; 2005; 2006; 2007).
Although a far broader suite of options are recommended for a thorough strategic
review, several options have been tested to demonstrate what may be achieved.
These include both a distributed and ecological sanitation system model for green‐field
service growth and low‐phosphate detergent program for reducing effluent
discharges.
Finally several implications and recommendations are discussed that are based on the
model outputs. Note that the model and options are based on publicly available data
alone and any assessments made are therefore illustrative rather than conclusive.
A case stud
y im
plem
entatio
n
62
4.1 Establishing the framework Define the objectives of the analysis and specify the economic criteria by which the
alternatives will be assessed
4.1.1 Defining the objectives
The drivers and constraints of the study are first clearly defined as a statement of
objectives.
The objective of this study is to establish an economically cost‐effective strategy for
servicing new developments on Melbourne’s fringe with a high quality sanitation
service while reducing nutrient flows from the Western System.
4.1.2 Defining the economic criteria
A common metric by which alternatives may be assessed is then defined in response to
the drivers of the analysis. This involves specifying the included costs, all stakeholder
perspectives to be accounted, an appropriate discount rate and common period for the
net present calculation.
The costs accounted in this study include:
• all real capital and operational costs borne by the customer, the utility
(wholesale and retail providers have been combined for simplicity), and any
program partners; and
• all major transfer payments identified as sewer, water and energy rates
• all major externalities identified as water pollution associated with sewage
discharge and greenhouse gas emissions (accounted as defined below)
The study has adopted the economic levelised dwelling cost as the appropriate metric
for ranking sanitation growth servicing options. This is defined as the present value of
aggregated costs and avoided costs associated with the alternative borne by all
stakeholders, divided by the present value of dwellings serviced over the assumed 50
year life‐cycle of the program. The weighted average cost of capital borne by the water
utility has been selected as a suitable basis for the discount rate (Mitchell et al. 2007),
which Strategic Financial Consulting Group estimate to be 6.4% (SFG 2007).
A case stud
y im
plem
entatio
n
63
4.1.3 Defining the treatment of externalities
All significant actual or avoided societal costs that are not reflected in market
exchanges are then assigned an appropriate assessment response. Externalities are
internalised within the costs using shadow pricing, embedded within the study
objectives, or accounted externally using a quantitative or qualitative assessment.
For the purposes of this study:
• impacts to receiving waters associated with the discharge of phosphorus and
nitrogen have been accounted as an equivalent load‐based pollution license
charge as defined by NSW State Government (Protection of the Environment
Operations (General) Regulation 1998) taken as a societal cost;
• impacts associated with indirect greenhouse gas emissions have been assumed
to have been accounted within the best available projection of energy costs
given a carbon emissions trading scheme as projected by the Australian
National Emissions Trading Taskforce (MMA 2006),
• impacts associated with direct greenhouse gas emissions have been accounted
as a best available forecasted cost of an equivalent emission permit, as defined
above (MMA 2006),
• other social and environmental impacts associated with the various options
have been assessed qualitatively and discussed within this report.
A case stud
y im
plem
entatio
n
64
4.2 Identifying the system Define the system boundaries, identify the components, and model their exchanges
within and across the system boundaries
4.2.1 Defining the boundaries
The boundaries of the system are first drawn to include the full scope of impacts
associated with the project. This implies an area inclusive of both any new development
area and any existing system components upon which the development may rely.
The case study boundary has been drawn inclusive of the existing and future
catchment boundary of the Western System, with all green‐field growth considered to
provide alternative capacity to the central system. Note a more comprehensive
boundary would be drawn inclusive of the Eastern System as the two systems have the
capacity for approximately 10% capacity transfer (Melbourne Water Corporation
2007).
4.2.2 Identifying the key components and their exchanges
A conceptual model of the elements of the system and their interactions was then
developed.
The system model accounts for three key resource flows:
• water, owing to its role as a constraint upon the capacity of treatment plants;
• nutrients, for which nitrogen and phosphorus are accounted owing to their key
role in both agriculture and the eutrophication of aquatic ecosystems; and
• energy, owing to its key role in greenhouse gas emissions.
The key activities to which these flows have been attributed include:
• bathing – showers, baths and basins;
• clotheswashing – clotheswashing machines and troughs;
• dishwashing –dishwashing machines and sinks;
• toilet – toilet flushing and leakage;
• non‐residential – activities associated with commercial, industrial and
recreational activities; and
• system gain – infiltration and inflow to the sewerage system.
A case stud
y im
plem
entatio
n
65
The activities were then placed within the context of the entire system as described by
Figure 4‐1. A similar model was developed by Gumbo (2005).
Figure 4‐1 – Conceptual model for flows of water and phosphorus in the urban system
Return Options
Wastewater System (Centralised / Distributed)
PFaeces + PUrine + WGreywater + PGreywater (+ WBlackwater) WWastewater + PWastewater (+ PSoilConditioner)
Municipal Solid Waste System
PBiowaste PLandfill (+ PSoilConditioner)
Urine diverting / composting systems
PFaeces + PFoodWaste (+ PUrine) PSoilConditioner + PLosses
Losses ‐ PLandfill
‐ WWastewater + PWastewater
‐ W l l ff +
Injections ‐ Wimported
‐ PMineralFertiliser
Services Inputs: Water +Food + Soaps & Detergents
Activities: Bathing + Clotheswashing + Dishwashing + Toilet
Outputs: Faeces + Urine + Greywater + Solid Waste (+ WBlackwater)
Water Balance: WReticulated (+ WHarvested) WServices (– DM) (WBlackwater) + WGreywater + WOutdoor
Phosphorus Balance:
PFood (+ PDetergent) PServices PFaeces + PUrine +PGreywater +PFoodWaste
Supply options
(not considered in this study)
Water supply (dams / recycling etc)
Wcollected (+ Wreclaimed) + Wimported Wreticulated + Wagriculatural
Food supply (conventional/ low‐tillage farming etc)
WAgricultural + PMineralFertiliser (+ PSoilConditioner + PUrine) PFood + PSoilStock + PAgriculturalRunoff
Non‐revenue water Exfiltration
Infiltration / inflow
A case stud
y im
plem
entatio
n
66
4.2.3 Quantifying key flows
Data describing the behaviour or usage associated with each of the activities, the type
and number of appliances applied, and the corresponding resources consumed is then
collected and reviewed.
Key sources applied in this study included an extensive survey of household appliance
stock and usage patterns and an end use measurement study both conducted by Yarra
Valley Water (Roberts 2004, 2005), state‐wide appliance sales and efficiency data (GFK
2006), and state‐wide water and energy appliance ownership surveys (ABS 2007b). The
key references and assumptions associated with each end use are described below:
• Bathing – based upon usage and water intensity data derived from a residential
end use measurement study in Melbourne (Roberts 2005), and state‐wide
showerhead ownership data (ABS 2007b);
• Clotheswashing – based upon usage data derived from a residential end use
measurement study in Melbourne (Roberts 2005), state‐wide appliance sales
and average appliance water water intensity data (GFK2006), and an Australian
detergent phosphorus intensity study (Patterson 2007) weighted to detergent
market share surveys (AC Nielsen 2007);
• Dishwashing – based upon usage data derived from a residential end use
measurement study in Melbourne (Roberts 2005), state‐wide sales and average
appliance water intensity (GFK 2006), and an Australian detergent phosphorus
intensity study (Patterson 2007) weighted to detergent market share surveys
(AC Nielsen 2007);
• Toilets – based upon usage and water intensity data from an end use
measurement study of Melbourne (Roberts 2005), state‐wide dual flush toilet
ownership data (ABS 2007b), nutrient intensity data derived from nation‐wide
dietary intake statistics (FAO 2007; Jönsson & Vinnerås 2003); and
• Non‐residential – based upon prior studies indicating approximately 10% of
baseline demand attributed to non‐residential components, deemed
proportional to population.
These data sets were then entered into the appropriate model fields. A list of model
components, their assumptions and references is provided as Appendix A.
A case stud
y im
plem
entatio
n
67
4.3 Specifying the base case Project the resource and financial flows associated with a ‘business as usual’ scenario
In order to later assess the incremental cost of an alternative, a baseline projection of
resource and financial flows is first developed.
4.3.1 Projecting baseline demography
The first stage of this process involved defining the historical and future demographic
changes within the study region to form the basis for the resource and financial
projections.
Current population, dwelling and lot mix data was therefore collected to provide an
accurate launching point for the demand forecast and hindcast. This data was drawn
from the most current spatially referenced data sets provided by the Australian Bureau
of Statistics (2002). This spatial data set, which is comprised of statistical collector
district units, was aggregated for the study areas.
The analysis was undertaken using MapInfo GIS software. This involved first overlaying
the spatial boundaries of the study regions upon the collector district‐level spatial
boundaries. A query was then executed specifying the software to select all those
collector districts with their centroid lying within the study area boundary. The
selected tables were then directly exported to the region data sources.
Historical demographic data was then required to establish an accurate age mix of
appliance stock for the commencement of the end use forecast. These were drawn
from a historical data collection project which provided a population times series for
the Melbourne Metropolitan Region (ABS 2006). This was converted to a time series of
hindcast growth factors to effectively apportion an equivalent hindcast growth to the
study area to that of the entire Melbourne region.
Finally future demographic projections were developed to form the basis for the
resource and financial projections. This data was drawn from the state planning
authority statistical local area projections (Vic Planning 2007), which were then
assessed by first fixing each of the SLA projections to spatial boundaries (ABS 2007a).
This enabled a similar process to that described above, however given the aggregation
was conducted using a larger component region (and was therefore of a lower
A case stud
y im
plem
entatio
n
68
resolution), this analysis was only applied to define a time series of population and
dwelling forecast growth factors similar to that applied for the hindcast. No data was
available post 2031, however given the degree of discounting inherent in the analysis
beyond this time period, a linear extrapolation was judged to be adequate.
The population and dwelling projection time series were then apportioned to either
single residential (i.e. a relatively homogenous group comprising detached dwellings)
or multi residential (i.e. a relatively heterogeneous group comprising terrace houses,
units, and other dwellings). This was undertaken based upon the census data that was
available.
Additionally, half of population and dwelling growth was apportioned as green‐field
development based upon Vic Planning projections for Melbourne assuming pro‐rate
rates for the Western System (Vic Planning 2004).
The outputs from this demographic analysis are shown in Figure 4‐4 and Figure 4‐3.
Figure 4‐2 ‐ Time series of population for the study area
0
500,000
1,000,000
1,500,000
2,000,000
2,500,000
30/6/1960
30/6/1963
30/6/1966
30/6/1969
30/6/1972
30/6/1975
30/6/1978
30/6/1981
30/6/1984
30/6/1987
30/6/1990
30/6/1993
30/6/1996
30/6/1999
30/6/2002
30/6/2005
30/6/2008
30/6/2011
30/6/2014
30/6/2017
30/6/2020
30/6/2023
30/6/2026
30/6/2029
30/6/2032
30/6/2035
30/6/2038
30/6/2041
30/6/2044
30/6/2047
30/6/2050
Popu
lation
Time [Year]
Single Multi
A case stud
y im
plem
entatio
n
69
Figure 4‐3 ‐ Time series of dwellings for the study region
As reflected in the relative difference in slope of the two charts, the projections reflect
a trend toward lower mean occupancy of dwellings in the study region, as illustrated in
the chart below.
‐
100,000
200,000
300,000
400,000
500,000
600,000
700,000
800,000
900,000
1,000,000
30/6/1960
30/6/1963
30/6/1966
30/6/1969
30/6/1972
30/6/1975
30/6/1978
30/6/1981
30/6/1984
30/6/1987
30/6/1990
30/6/1993
30/6/1996
30/6/1999
30/6/2002
30/6/2005
30/6/2008
30/6/2011
30/6/2014
30/6/2017
30/6/2020
30/6/2023
30/6/2026
30/6/2029
30/6/2032
30/6/2035
30/6/2038
30/6/2041
30/6/2044
30/6/2047
30/6/2050
Dwellin
gs
Time [Years]
Single Multi
0.000
0.500
1.000
1.500
2.000
2.500
3.000
3.500
30/6/1960
30/6/1963
30/6/1966
30/6/1969
30/6/1972
30/6/1975
30/6/1978
30/6/1981
30/6/1984
30/6/1987
30/6/1990
30/6/1993
30/6/1996
30/6/1999
30/6/2002
30/6/2005
30/6/2008
30/6/2011
30/6/2014
30/6/2017
30/6/2020
30/6/2023
30/6/2026
30/6/2029
30/6/2032
30/6/2035
30/6/2038
30/6/2041
30/6/2044
30/6/2047
30/6/2050
Popu
lation
Time [Years]
Single Multi
A case stud
y im
plem
entatio
n
70
4.3.2 Projecting baseline resource flows
Having established the baseline demographic projection for the study region, the next
stage involves projecting the baseline resource flows associated with the system under
study.
This is undertaken by executing an ‘update and transfer’ of the iSDP model, which
effectively uploads the demographic data to a central database and distributes that
data to each of the baseline components. A component forecast was then executed
representing the resource flows associated with each activity and its constituent
appliances. Each of these components forecasts are dependent upon the assumptions
established in Section 4.2, the impact of which has been fully described in the sections
below. The component forecasts were then re‐uploaded to the central database and
aggregated to form the various baseline forecasts.
Sewage water flows
The following charts demonstrate the model projections for the water flows being
drawn from the study region. Figure 4‐4 firstly shows the model projection for water
flows associated with bathing activities in residential dwellings, which include the use
of showers, baths and hand basins.
Figure 4‐4 ‐ Projection for water flows associated with bathing activities by appliance
‐
5,000
10,000
15,000
20,000
25,000
30,000
35,000
40,000
45,000
50,000
Water Dem
and [M
L/a]
Time [Year]
Shower Bath Basin
A case stud
y im
plem
entatio
n
71
The projection assumes shower water to be proportional to:
• population, which is trended to increase;
• frequency of showering, which assumes a slight increase owing to aging of the
population profile; duration of showering, which is assumed constant; and
• the water intensity per showerhead type, which is assumed to reduce owing to
a shift toward efficient showerheads.
The projection assumes bath water to be proportional to:
• population, which is trended to increase;
• frequency of bathing, which assumes a slight decrease owing to aging of the
population profile; and
• bath filled volumes, which assume an increase owing to aging of the
demographic profile
The model assumes hand basin water to be proportional to
• population;
• frequency and duration of use, which are both assumed constant; and
• flow rates, also assumed constant.
The resulting baseline component sewage forecast shows a discontinuity with
demographic growth predominantly reflecting an assumed baseline shift toward
efficient showerheads.
A case stud
y im
plem
entatio
n
72
Figure 4‐5 shows the model projection for water flows associated with clotheswashing
activities in residential dwellings, which include the use of clotheswashing machines
and laundry troughs.
Figure 4‐5 ‐ Projection for water flows associated with clotheswashing activities by appliance
The projection assumes that clotheswashing machine water is proportional to:
• the stock of machines, which is trended to increase with population growth,
reduced occupancy, and a slight increase in ownership levels;
• frequency of machine washing per machine, which is assumed to decrease with
reduced occupancy; and
• the water intensity per activity by appliance type, which is assumed to decrease
with a strong shift toward front‐loading clotheswashing machines.
The projection assumes that laundry trough water is proportional to
• the number of households, and
• inversely proportional to the ownership of clotheswashing machines
The resulting baseline component sewage forecast therefore shows a dramatic
reduction in both total and per capita water flows associated with clotheswashing
activities associated with a shift toward front‐loading washing machines.
‐
5,000
10,000
15,000
20,000
25,000
30,000
30/6/1960
30/6/1964
30/6/1968
30/6/1972
30/6/1976
30/6/1980
30/6/1984
30/6/1988
30/6/1992
30/6/1996
30/6/2000
30/6/2004
30/6/2008
30/6/2012
30/6/2016
30/6/2020
30/6/2024
30/6/2028
30/6/2032
30/6/2036
30/6/2040
30/6/2044
30/6/2048
Water Dem
and [M
L/a]
Time [Year]
Clotheswashers Troughs
A case stud
y im
plem
entatio
n
73
Figure 4‐6 shows the model projection for water flows associated with dishwashing
activities in residential dwellings, which include the use of dishwashing machines and
kitchen sinks.
Figure 4‐6 ‐ Projection for water flows associated with dishwashing activities by appliance
The projection is based upon the assumption that dishwashing machine water is
proportional to:
• the stock of machines, which is trended to increase with population growth,
reduced occupancy, and an increase in ownership levels;
• frequency of machine washing per machine, which is assumed to decrease with
reduced occupancy (though increasing per capita); and
• the water intensity per activity, which is assumed to approach an asymptote.
The projection assumes kitchen sink water to be proportional to:
• the number of dwellings, which is trended to increase;
• frequency of hand washing, which is projected to decrease with increased
dishwashing machine ownership; and
• consumption per hand washing activity, which is assumed constant.
‐
2,000
4,000
6,000
8,000
10,000
12,000
30/6/1960
30/6/1964
30/6/1968
30/6/1972
30/6/1976
30/6/1980
30/6/1984
30/6/1988
30/6/1992
30/6/1996
30/6/2000
30/6/2004
30/6/2008
30/6/2012
30/6/2016
30/6/2020
30/6/2024
30/6/2028
30/6/2032
30/6/2036
30/6/2040
30/6/2044
30/6/2048
Water flow
[ML/a]
Time [Year]
Dishwashers Sinks
A case stud
y im
plem
entatio
n
74
Based on these assumptions the uptake of dishwashing machines results in a net
decrease in household dishwashing water predominantly owing to the reduced
associated frequency of sink use. Also, the model assigns half of all kitchen sink use to
be flow related, the other half is related to capacity, which is consistent with
international research (Friedler & Butler 1996).
Figure 4‐7 shows the model projection for water flows associated with toilet flushing in
residential dwellings.
Figure 4‐7 ‐ Projection for water flows associated with toilet flushing
The projection is based upon the assumption that toilet flushing water is proportional
to:
• the population of the study area, which is projected to increase;
• the frequency of flushing at home, which is assumed constant; and
• the water intensity per stock type; which is projected to decrease significantly
with the shift toward dual flush toilet models of reduced mean flush volume.
The latter term is dominant with a significant reduction in per capita and total water
use. Note the model indicates around half of these savings are likely to already have
been reflected in water flows from the first year of projection.
‐
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
18,000
20,000
30/6/1960
30/6/1964
30/6/1968
30/6/1972
30/6/1976
30/6/1980
30/6/1984
30/6/1988
30/6/1992
30/6/1996
30/6/2000
30/6/2004
30/6/2008
30/6/2012
30/6/2016
30/6/2020
30/6/2024
30/6/2028
30/6/2032
30/6/2036
30/6/2040
30/6/2044
30/6/2048
Water flow
[ML / a]
Time [Year]
A case stud
y im
plem
entatio
n
75
Figure 4‐7 shows the model projection for water flows associated with non‐residential
components.
Figure 4‐8 ‐ Projection of water flows associated with non‐residential components
The projection is based upon the assumption that non‐residential water is consistent
with historically observed sewage flows assumed to increase pro‐rata with population.
This rate of growth is likely to be overestimated and warrants further investigation. A
review of customer databases should be conducted together with consultations and
measurements undertaken with major non‐residential customers (see
recommendations in Chapter 5).
‐
10,000
20,000
30,000
40,000
50,000
60,000
30/6/1997
30/6/1999
30/6/2001
30/6/2003
30/6/2005
30/6/2007
30/6/2009
30/6/2011
30/6/2013
30/6/2015
30/6/2017
30/6/2019
30/6/2021
30/6/2023
30/6/2025
30/6/2027
30/6/2029
30/6/2031
30/6/2033
30/6/2035
30/6/2037
30/6/2039
30/6/2041
30/6/2043
30/6/2045
30/6/2047
30/6/2049
Water flow
[ML / a]
Time [date]
A case stud
y im
plem
entatio
n
76
Figure 4‐9 shows the aggregated and stacked end use components as a baseline dry
weather sewage flow plotted against recorded influent flows at the wastewater
treatment plant (Melbourne Water Corporation 1998; 1999; 2000; 2001; 2002; 2003;
2004; 2005; 2006; 2007).
Figure 4‐9 ‐ Projection of baseline sewage flow by end use
The figure indicates that, based on the model assumptions, residential sewage flows
will decrease both in total and on a per capita basis, however this is offset by a slight
increase in total non‐residential demand.
Note the disparity between the modelled and recorded flows may be predominantly
attributed to wet weather inflows, which drop significantly over recent years owing to
unusually low rainfall.
0
20,000
40,000
60,000
80,000
100,000
120,000
140,000
160,000
180,000
200,000
1/01
/1998
1/01
/2001
1/01
/2004
1/01
/2007
1/01
/2010
1/01
/2013
1/01
/2016
1/01
/2019
1/01
/2022
1/01
/2025
1/01
/2028
1/01
/2031
1/01
/2034
1/01
/2037
1/01
/2040
1/01
/2043
1/01
/2046
1/01
/2049
Baselin
e sewage flo
ws [M
L/a]
Time [date]
Bathing Clotheswashing Dishwashing
Toilets Non‐residential Sewage flow (recorded)
A case stud
y im
plem
entatio
n
77
Figure 4‐10 shows the share of overall baseline sewage water flows by end use for the
financial year ending 2008, the first year of projection. The infiltration component
represents the residual in the year 2000, which is was assumed to represent the most
normal rainfall year, factored pro‐rate with the growth of dwelling numbers.
Figure 4‐10 – Share of sewage flow by end use
The figure indicates that a large component of the sewage stream (34%) is associated
with infiltration and inflow, that is, the sewage flow associated with the collection
system itself.
Also worthy of note is the relatively small share of toilet flushing, which is contrary to
similar international studies (Almeida, Butler & Friedler 1999; Butler 1993). This is a
reflection of the significant share of dual flush toilets of comparative water efficiency,
the result being a consequent reduction in end use share.
Bathing 22%
Clotheswashing 13%
Dishwashing 4%
Toilet 7%
Non‐residential 20%
Inflow & Infiltration 34%
A case stud
y im
plem
entatio
n
78
Sewage phosphorus flows
The following charts demonstrate the model projections for phosphorus flows being
drawn from the study region. Figure 4‐11 firstly shows the model projection for
phosphorus flows associated with toilet flushing in residential dwellings.
Figure 4‐11 ‐ Projection of phosphorus flows associated with toilet flushing
This projection is based upon the assumption that toilet flushing phosphorus is
proportional to:
• population, which is projected to increase; and
• dietary protein intake, proportionate share of vegetal protein, and time spent
at home, which in the absence of any further data have been assumed constant
over time.
The dietary protein intake in Australia is twice that recommended by Nutrition
Australia (2008). A shift toward such levels would therefore result in a 50% reduction
in phosphorus entering the wastewater stream. Also, based upon algorithms
developed by Jönsson & Vinnerås (2003), a shift toward a vegetarian diet in the
population would lead to 50% increase in the phosphorus load in the system, while, for
interest, an entirely carnivorous diet would lead to a 25% reduction based on an equal
dietary protein intake.
‐
200,000
400,000
600,000
800,000
1,000,000
1,200,000
1,400,000
1,600,000
30/6/1960
30/6/1964
30/6/1968
30/6/1972
30/6/1976
30/6/1980
30/6/1984
30/6/1988
30/6/1992
30/6/1996
30/6/2000
30/6/2004
30/6/2008
30/6/2012
30/6/2016
30/6/2020
30/6/2024
30/6/2028
30/6/2032
30/6/2036
30/6/2040
30/6/2044
30/6/2048
Phosph
orus flow
[kg/a]
Time [Year]
Urine Faeces
A case stud
y im
plem
entatio
n
79
Figure 4‐12 shows the model projection for phosphorus flows associated with
clotheswashing activities in residential dwellings, which predominantly occur as a
result of detergent builders in machine washing
Figure 4‐12 ‐ Projection of phosphorus flows associated with clotheswashing detergents
The projection is based upon the assumption that clotheswashing phosphate flows are
proportional to:
• the stock of machines, which is trended to increase with population growth,
reduced occupancy, and a slight increase in ownership levels;
• frequency of machine washing per machine, which is assumed to decrease with
reduced occupancy; and
• the phosphate intensity per activity by appliance type, which is assumed to
decrease marginally with a shift toward front‐loading clotheswashing machines.
The model therefore predicts phosphate inputs attributed to clotheswashing activities
to grow in total while reducing slightly on a per capita basis.
‐
100,000
200,000
300,000
400,000
500,000
600,000
700,000
800,000
900,000
30/6/1960
30/6/1964
30/6/1968
30/6/1972
30/6/1976
30/6/1980
30/6/1984
30/6/1988
30/6/1992
30/6/1996
30/6/2000
30/6/2004
30/6/2008
30/6/2012
30/6/2016
30/6/2020
30/6/2024
30/6/2028
30/6/2032
30/6/2036
30/6/2040
30/6/2044
30/6/2048
Phosph
orus flow
[kg/a]
Time [Year]
A case stud
y im
plem
entatio
n
80
Figure 4‐13 shows the model projection for phosphorus flows associated with
dishwashing activities in residential dwellings, which are solely attributed to
dishwashing machines (Patterson 2007).
Figure 4‐13 ‐ Projection of phosphorus flows associated with dishwashing detergents
The projection is based upon the assumption that dishwashing machine phosphorus is
proportional to:
• the stock of machines, which is trended to increase with population growth,
reduced occupancy, and an increase in ownership levels;
• frequency of machine washing per machine, which is assumed to decrease with
reduced occupancy (though increasing per capita); and
• the phosphorus intensity per activity, which is assumed constant.
The data presented by Patterson (2007) suggests there is significant variability in the
phosphate intensity of dishwashing detergents, with one sample exhibiting a ten‐fold
increase on average levels. This finding warrants further research to develop a
thorough inventory of the phosphate concentrations and market share of common
detergent products as a basis for developing a more rigorous weighted average
phosphate load.
‐
20,000
40,000
60,000
80,000
100,000
120,000
140,000
160,000
180,000
30/6/1960
30/6/1964
30/6/1968
30/6/1972
30/6/1976
30/6/1980
30/6/1984
30/6/1988
30/6/1992
30/6/1996
30/6/2000
30/6/2004
30/6/2008
30/6/2012
30/6/2016
30/6/2020
30/6/2024
30/6/2028
30/6/2032
30/6/2036
30/6/2040
30/6/2044
30/6/2048
Phosph
orus flow
[kg/a]
Time [Year]
A case stud
y im
plem
entatio
n
81
Figure 4‐18 shows the model projection for phosphorus flows associated with toilet
flushing in non‐ residential premises.
Figure 4‐14 – Projection of phosphorus flows associated with toilet flushing in non‐residential premises
The model assumes phosphorus flows associated with toilet flushing in non‐residential
dwellings to be proportional to population, dietary intake, and time spent outside the
home. Note that the proportions are slightly shifted toward urine derived phosphates
owing to the ‘at home’ assumptions favouring defecating at home.
0
100,000
200,000
300,000
400,000
500,000
600,000
30/6/1997
30/6/1999
30/6/2001
30/6/2003
30/6/2005
30/6/2007
30/6/2009
30/6/2011
30/6/2013
30/6/2015
30/6/2017
30/6/2019
30/6/2021
30/6/2023
30/6/2025
30/6/2027
30/6/2029
30/6/2031
30/6/2033
30/6/2035
30/6/2037
30/6/2039
30/6/2041
30/6/2043
30/6/2045
30/6/2047
30/6/2049
Phosph
orus flow
[kg/a]
Time [Year]
Urine Faeces
A case stud
y im
plem
entatio
n
82
Figure 4‐15 shows the aggregated and stacked end use components as baseline
phosphorus flows plotted against measurements of total phosphorus present in
wastewater treatment plant influent (Melbourne Water Corporation 1998; 1999; 2000;
2001; 2002; 2003; 2004; 2005; 2006; 2007).
Figure 4‐15 ‐ Projection of baseline phosphorus flow by end use
The chart indicates the model projections lie within the uncertainty range of the
measurements recorded by Melbourne Water Corporation.
The projection projects a growth in the baseline phosphorus flows within the system
that is approximately consistent with population growth, and, based upon the above
sewage modelling, it may be inferred that concentrations of influent phosphates to the
wastewater treatment plant may increase with reduced proportionate hydraulic flows.
0
500,000
1,000,000
1,500,000
2,000,000
2,500,000
3,000,000
1/01
/1998
1/01
/2001
1/01
/2004
1/01
/2007
1/01
/2010
1/01
/2013
1/01
/2016
1/01
/2019
1/01
/2022
1/01
/2025
1/01
/2028
1/01
/2031
1/01
/2034
1/01
/2037
1/01
/2040
1/01
/2043
1/01
/2046
1/01
/2049
Baselin
e ph
osph
orus flow
[kg/a]
Time [Year ending]
Residential Clotheswashing Residential Dishwashing Residential toilet
Non‐residential toilet Phosphorus flow (reported)
A case stud
y im
plem
entatio
n
83
Figure 4‐16 shows the shares of overall baseline phosphorus flow by end use for the
financial year ending 2008, the first year of projection.
Figure 4‐16 ‐ Share of phosphorus flow by end use
The component contributions of phosphorus by end use indicate approximately 60% of
phosphorus flows in wastewater systems are associated with excreta, which is slightly
greater than the 50% asserted by Butler and Davies (2004, p. 67).
Clotheswashing 34%
Dishwashing 5%
Residential toilets 39%
Non Residential toilets 22%
A case stud
y im
plem
entatio
n
84
Baseline nitrogen flows
The following charts demonstrate the model projections for nitrogen flows within the
study region, of which excreta have been assumed as the sole component. This
assumption is supported by Butler and Davies (2004, p. 67) who attribute 94% of
nitrogen to excreta. The residual, which is predominantly ammonia in cleaning
products, has not been accounted in this study.
Figure 4‐17 shows the model projection for nitrogen flows associated with toilet
flushing in residential dwellings.
Figure 4‐17 ‐ Projection of nitrogen flows associated with toilet flushing in residential dwellings
The model assumes nitrogen flows associated with toilet flushing in residential
dwellings to be proportional to:
• population, which is projected to increase and
• dietary intake and time spent at home, both assumed constant with time.
Note the relative shares for nitrogen attributed to urine and faeces is in disagreement
with those provided by Butler and Davies (2004, p. 67), but are consistent with the
most current empirical research (Gumbo 2005; Jönsson & Vinnerås 2003; Otterpohl
2001).
‐
1,000,000
2,000,000
3,000,000
4,000,000
5,000,000
6,000,000
7,000,000
8,000,000
9,000,000
10,000,000
30/6/1960
30/6/1964
30/6/1968
30/6/1972
30/6/1976
30/6/1980
30/6/1984
30/6/1988
30/6/1992
30/6/1996
30/6/2000
30/6/2004
30/6/2008
30/6/2012
30/6/2016
30/6/2020
30/6/2024
30/6/2028
30/6/2032
30/6/2036
30/6/2040
30/6/2044
30/6/2048
Nitrogen flo
w [k
g/a]
Time [Year]
Urine Faeces
A case stud
y im
plem
entatio
n
85
Figure 4‐18 shows the model projection for nitrogen flows associated with toilet
flushing in non‐ residential premises.
Figure 4‐18 – Projection of nitrogen flows associated with toilet flushing in non‐residential premises
The model assumes nitrogen flows associated with toilet flushing in non‐residential
premises to be proportional to:
• population, which is projected to increase; and
• dietary intake, and time spent outside the home, each assumed constant.
Note that owing to alternative ‘at home’ assumptions for excreta, the bulk of nitrogen
is owing to urine.
0
1,000,000
2,000,000
3,000,000
4,000,000
5,000,000
6,000,000
7,000,000
30/6/1997
30/6/1999
30/6/2001
30/6/2003
30/6/2005
30/6/2007
30/6/2009
30/6/2011
30/6/2013
30/6/2015
30/6/2017
30/6/2019
30/6/2021
30/6/2023
30/6/2025
30/6/2027
30/6/2029
30/6/2031
30/6/2033
30/6/2035
30/6/2037
30/6/2039
30/6/2041
30/6/2043
30/6/2045
30/6/2047
30/6/2049
Nitrogen flo
w [k
g/a]
Time [Year]
Urine Faeces
A case stud
y im
plem
entatio
n
86
Figure 4‐19 shows the aggregated and stacked end use components as baseline
nitrogen flows plotted against measurements of total nitrogen present in the
wastewater treatment plant influent (Melbourne Water Corporation 1998; 1999; 2000;
2001; 2002; 2003; 2004; 2005; 2006; 2007).
Figure 4‐19 ‐ Projection of baseline nitrogen flow by end use
The chart indicates the model projections lie within the uncertainty range of the
measurements recorded by Melbourne Water Corporation.
Nitrogen is subject to a number of chemical processes within the sewer stream, the
most dominant of which is the conversion of urea in urine to ammonia in both aerobic
and anaerobic conditions (Butler & Davies 2004). Such conversions warrant further
research to drawing any distinct connections between end use loads and wastewater
treatment plant influent nitrogen compositions.
0
2,000,000
4,000,000
6,000,000
8,000,000
10,000,000
12,000,000
14,000,000
16,000,000
1/01
/1998
1/01
/2001
1/01
/2004
1/01
/2007
1/01
/2010
1/01
/2013
1/01
/2016
1/01
/2019
1/01
/2022
1/01
/2025
1/01
/2028
1/01
/2031
1/01
/2034
1/01
/2037
1/01
/2040
1/01
/2043
1/01
/2046
1/01
/2049
Baselin
e nitrogen
flow
[kg/a]
Time [Year ending]
Residential toilets Non residential toilets Nitrogen flow (reported)
A case stud
y im
plem
entatio
n
87
Figure 4‐20 shows the shares of overall baseline nitrogen flow by end use for the
financial year ending 2008, the first year of projection.
Figure 4‐20 – Share of nitrogen flow by end use
The baseline mix of end use shares is simply a reflection of the ‘at home’ factors
specifying a 40 hour working week. This means that based on the model assumptions
the share will shift on weekends, while the total nitrogen inflow will remain constant
on a daily time step. An hourly time step would likely reveal a morning peak, however
no empirical studies appear to have supported or refuted such an assertion.
Conclusions and recommendations
The resource projections presented above validate well with the empirical
measurements undertaken by Melbourne Water Corporation and are considered to be
the best currently available end use disaggregation of dry weather sewage flows.
However further research to characterise peak wet weather flows is considered
strategic.
Residential toilets 59%
Non residential toilets 41%
A case stud
y im
plem
entatio
n
88
4.4 Identifying the options Inventory the broad suite of options available toward meeting the study objectives and
estimate their incremental resource and financial impacts
4.4.1 Distributed system
Small bore settled sewer systems are designed to receive only the liquid portion of
household wastewater, thus significantly simplifying wastewater collection and
treatment (Otis & Mara 1985).
The system applied in this option is based upon that developed by Orenco Systems
(2007). Combined wastewater from the house is transmitted to a household septic or
interceptor tank, which provides daily peak storage and separates the dominant share
of suspended solids for intermittent removal and disposal. The filtered wastewater is
then either pumped or conveyed by gravity to a low diameter pressurised sewer
network. Such networks have no minimum falls, thus avoiding the expense of deep
excavations or manholes, and apply watertight fittings, thus circumventing inflow,
infiltration and exfiltration in most circumstances. Figure 4‐21 depicts the typical
configuration of the collection components.
Figure 4‐21 ‐ Collection components: interceptor tank, effluent pump and pressurised sewer (Orenco Systems 2007)
The filtered effluent is treated by modular wastewater treatment systems, which
consist of recirculating sand or textile filters and UV disinfection. Such systems are
fitted with telemetry controls and therefore require a reduced level of supervision and
maintenance relative to centralised treatment plants.
A case stud
y im
plem
entatio
n
89
Figure 4‐22 shows a typical distributed treatment system.
Figure 4‐22 ‐ Treatment components: modular recirculating textile filters (Orenco Systems 2007)
The treated and disinfected effluent is rendered to a quality suitable for low‐risk
irrigation purposes.
Resource impact
The projected resource impact of the option has been based upon the following
assumptions:
• The penetration of the option is assumed to include all greenfield development
in the study region, which is assumed to approximately equal 50% of projected
new dwellings (Vic Planning 2007).
• All associated baseline sewage, phosphorus and nitrogen flows are diverted
from the central sewerage system.
• Infiltration flows apportioned to the affected components are assumed to be
negligible.
• Sufficient treatment plants are installed to provide capacity for average
estimated flows three years hence, with plant modular capacity equal to 60kL/d
(22ML/a). Note no peaking factor has been assigned owing to daily storage of
interceptor tanks.
When compared to centralised systems that often require designs to meet projected
demand several decades hence, this adaptive mechanism may provide significant
A case stud
y im
plem
entatio
n
90
benefits in terms of reduced exposure to uncertainty in forecasting future capacity
demand. This benefit, though important, is beyond the scope of this project.
Figure 4‐23 depicts the impact of this alternative capacity upon the influent sewage
flows at the Western Treatment Plant.
Figure 4‐23 ‐ Projected offset sewage flow to Western Treatment Plant
This output is designed to provide scope for assessing any avoided treatment plant
capacity upgrades. As may be seen the option has the effect of levelling the baseline
sewage projection. Although falling outside the scope of this study, a similar analysis
would be useful for sub‐components of the wastewater system (e.g. to project trunk
sewer flows and capacity).
0
50,000
100,000
150,000
200,000
250,000
1998 2002 2006 2010 2014 2018 2022 2026 2030 2034 2038 2042 2046 2050
influ
ent sew
age [M
L/a]
Year End June
Baseline Demand Forecast Distributed sanitation option impact
A case stud
y im
plem
entatio
n
91
In addition, Figure 4‐24 provides the ability for the decision‐maker to assess the impact
of the option in reducing the discharge of phosphorus to the receiving environment.
Figure 4‐24 ‐ Projected reduction in effluent phosphorus from Western Treatment Plant
A similar output is available for nitrogen flow reductions.
Financial impact
The financial impact of the option is based on the following assumptions:
• Capital costs assigned to each dwelling total $5100 and include an interceptor
tank, effluent filter and pump ($4500) and pressurized piping ($600, 30m @
$20/m) all inclusive (Holt & James 2006; Mitchell et al. 2007).
• Capital costs for each 60kL/d capacity treatment system total $500,000 and
include AdvanTex Treatment Pods (Orenco Systems 2007), UV disinfection and
on‐site sub‐surface irrigation system (Holt & James 2006).
• Operational costs assigned to each dwelling total $94/a and include inspection
($10/a: 1hr @ $40/h every 4 years), maintenance ($34), and administrative
overheads ($24/a: apportioned 0.5% of $5,600/a) and electricity (~$22: 110
kWh/a @ approx. $0.20/kWh (White 2004 adjusted for local units rates).
• Operational costs assigned to each treatment plant include regular inspections
($2,160/a: 4.5h/month @ $40/h), maintenance ($720/a), miscellaneous
‐
200,000
400,000
600,000
800,000
1,000,000
1,200,000
1,400,000
30/6/2008
30/6/2010
30/6/2012
30/6/2014
30/6/2016
30/6/2018
30/6/2020
30/6/2022
30/6/2024
30/6/2026
30/6/2028
30/6/2030
30/6/2032
30/6/2034
30/6/2036
30/6/2038
30/6/2040
30/6/2042
30/6/2044
30/6/2046
30/6/2048
30/6/2050
Phosph
orus discharged [kg/a]
Time [date]
A case stud
y im
plem
entatio
n
92
repaires ($1,119), monitoring ($2,140/a), administration ($2,800/a:
apportioned 50% of $5,600), and electricity (approx $1,200/a: 17MWh/a @
$0.07/kWh (White 2004 adjusted for local units rates).
As a first pass, the following avoided costs were included:
• Marginal capital costs associated with new reticulation sewerage (excluding
trunk sewers) ($5,500 per substituted dwelling) (Mitchell et al. 2007).
• Marginal operational costs associated with additional energy, sludge disposal
and chemicals associated with pumping and treatment ($100/a per substituted
dwelling) (GHD 2003; confirmed by Mitchell et al. 2007)
A case stud
y im
plem
entatio
n
93
4.4.2 Ecological system
Overview
Dry toilets collect, store and sanitise human excreta within the premises without the
addition of water for flushing (Del Porto & Steinfeld 1999). Such toilets may be
designed to a high aesthetic standard while permitting a minimal obligation for
homeowner intervention as demonstrated in a recent feasibility study for their
introduction in Melbourne (GHD 2003). The removal of faeces and urine reduces the
task of managing the remaining greywater considerably, and may be treated most
cost‐effectively by natural processes (Del Porto & Steinfeld 1999).
The system applied in this case study is a high standard urine diverting dry system
similar to that recently applied in Dong Sheng, China (Kvarnström et al. 2006). The
system transmits faeces down a vertical chute to a rotary compost storage bin located
in the basement, while a urine diverting pedestal collects the separate urine for
transmission by 75mm gravity pipe to a 1.3kL urine storage tank, also located in the
basement.
The separate management of urine and faeces was preferred primarily owing to the
significantly reduced volume of potentially pathogenic faecal waste and reduced
incidence of anaerobic conditions developing in the pile which is the most common
failure mode of composting systems (Kvarnström et al. 2006). In addition, the separate
collection of urine significantly reduces the required addition of carbonaceous material
to offset the heavy nitrogen load of the urine. The systems also facilitate the safe reuse
of urine (which has an inherently lower pathogenic risk), while minimising off‐gassing
of nitrogen as ammonia, which causes odour and reduces the value of the fertiliser
(Kvarnström et al. 2006).
A case stud
y im
plem
entatio
n
94
Figure 4‐25 depicts a urine‐diverting pedestal and waterless urinal. Note the two holes
of the pedestal.
Figure 4‐25 – Collection components: urine diverting pedestal and waterless urinal (Kvarnström et al. 2006)
Faeces are stored for at least 18 months where they are sanitised and reduced by two
thirds by co‐composting with household biodegradable waste. Similarly the potentially
contaminated urine is sanitised during a minimum 6 month storage owing to the
alkaline condition of the urine (WHO 2005). The faeces are aerated and any odour
extracted using a fan placed in the basement, thus providing a significant benefit
relative to water flush toilets. Figure 4‐26 depicts typical examples of both a rotary
compost bin and urine tank for medium density developments.
Figure 4‐26 ‐ Treatment components: rotary compost bin and urine storage tank (Kvarnström et al. 2006)
The urine and faeces may be applied safely on‐site or intermittently collected by a
waste collection contractor and transferred for subsequent post‐treatment for
agricultural land.
A case stud
y im
plem
entatio
n
95
Figure 4‐27 depicts typical examples of both a small‐scale and large‐scale application
systems.
Figure 4‐27 ‐ Reuse components: small and large‐scale application of urine for agricultural reuse (Kvarnström et al. 2006)
Greywater from bathrooms, laundries and kitchens is pre‐treated using a small sump
and filter to remove oil, grease, lint and other fine particles, thus providing facility for
on‐site reuse in sub‐surface irrigation. If unused within 24 hours, the greywater is
pumped by low diameter pressurised pipe to a central sub‐surface irrigation area such
as a community garden or parkland.
Figure 4‐28 depicts two key components of the system: an on‐site diverter pump and
dripline irrigation system for on‐site reuse.
Figure 4‐28 – Greywater reuse components: diverter storage, pump and dripline system (Eco‐care, 2008)
A case stud
y im
plem
entatio
n
96
Resource impact
The projected resource impact of the option is based upon the following assumptions:
• The penetration is assumed to include all greenfield development in the study
region, which is assumed to approximately equal 50% of projected new
dwellings (Vic Planning 2004).
• All associated baseline sewage, phosphorus and nitrogen flows are diverted
from the sewerage system.
• The products collected from each household are equivalent to
427Litres/person.annum as urine (1.5L per day x 365 days x 0.78) and
41kg/person.annum as compost (((50kg faeces + 10kg toilet paper)*0.9 + 70kg
kitchen waste)*0.33)
• The diverted nutrients flows in recycled kitchen waste have not been
accounted.
The flow outputs are therefore identical to those identified in the section above.
Financial impact
The projected financial impact of the option is based upon the following assumptions:
• Capital costs assigned to each dwelling total $5,270 and include rotary compost
bin, ventilation fan and chute ($3,170), 1kL polypropylene urine tank ($600),
greywater treatment, storage and pump ($900), and assigned greywater piping
layed on contour ($600: $20 x 30m), borne by the utility.
• Operational costs assigned to each dwelling total $47/a and include inspections
($40: 1h/a x $40/h), and energy associated with the ventilation fan (350kWh/a
x $0.2/kWh), borne by the utility.
• Operational costs associated with cartage of urine by rigid vacuum liquid
removal truck ($50/kL) (GHD 2003) and compost by rigid solid waste removal
truck including processing at existing composting facility ($120/tonne) (Wang
2001), borne by the solid waste provider
• A transfer payment of $58 per connected dwelling is borne by the utility to the
solid waste management service provider.
A case stud
y im
plem
entatio
n
97
• The greenhouse gas emissions associated with the cartage of compost have not
been accounted owing to time constraints. One potential means of quantifying
this externality is described by Wang (2001).
• Similarly the recovered value of recycled greywater has not been accounted.
The avoided costs have been taken as in the section above, excluding any avoided or
deferred capital infrastructure associated with centralised transfer and treatment.
• Marginal capital costs associated with new reticulation sewerage (excluding
trunk sewers) ($5,500 per substituted dwelling) (Mitchell et al. 2007), borne by
the utility
• Marginal operational costs associated with additional energy, sludge disposal
and chemicals associated with pumping and treatment ($100/a per substituted
dwelling) (GHD 2003; confirmed by Mitchell et al. 2007), borne by the utility
• Marginal avoided collection costs of organic waste removed from the municipal
solid waste stream have been taken as $60/tonne, including collection costs
($25/tonne) and landfill disposal ($35/tonne) (Wang, Richardson & Roddick
1998 inflated to present values), borne by the partner
A case stud
y im
plem
entatio
n
98
4.5 Analysing the costs Apply discounted cash flow analysis to each alternative and assess both the economic
cost and the financial cost borne by each stakeholder
4.5.1 Distributed alternative
Figure 4‐29 depicts the time series of incremental costs associated with the option
Figure 4‐29‐ Financial impact of distributed sanitation option
For the utility, the dominant incremental cost is associated with additional capital
investment for treatment, which has been assumed as a sunk cost in the baseline.
For society, there is a slight marginal avoided cost associated with the avoided
discharge of nutrients resulting in a present value saving of $600,000. This negligible
saving is associated with the assumption that the disposal of nutrients is not critical as
the receiving water is classed as estuarine. To provide an indication of the sensitivity of
this parameter, assuming a receiving environment classified as an enclosed water body
the incremental saving totals $11.7 million. Based upon the model assumptions this
avoided cost is therefore not a dominant factor even given restrictive conditions.
The aggregated present value economic cost of the option is projected to be $168
million, which translates as a 46% premium on all green‐field developments sites
‐$2,000,000
$0
$2,000,000
$4,000,000
$6,000,000
$8,000,000
$10,000,000
$12,000,000
$14,000,000
$16,000,00030
/6/2008
30/6/2010
30/6/2012
30/6/2014
30/6/2016
30/6/2018
30/6/2020
30/6/2022
30/6/2024
30/6/2026
30/6/2028
30/6/2030
30/6/2032
30/6/2034
30/6/2036
30/6/2038
30/6/2040
30/6/2042
30/6/2044
30/6/2046
30/6/2048
30/6/2050
Fina
ncial flow [$
/a]
Time [date]
Economy Utility Society
A case stud
y im
plem
entatio
n
99
assuming there are no additional capital costs associated with centralised transfer or
treatment.
Although the definitive analysis of such an assertion lies outside the scope of this
project, it is not unreasonable to project the delay or avoidance of capital
augmentations as a result of the option. By way of example, the Northern Sewerage
Project is flagged ostensibly to address new development in the northern peri‐urban
fringe of the Western System, and has been commissioned at a value of $650m spread
over two years (Melbourne Water Corporation 2008).
Based upon the model assumptions, a project delay of only five years would result in
an overall economic present value saving of $56m, whereas an avoidance of the
project would result in an economic present saving of $519m. Transfer and treatment
constraints are therefore key drivers of the cost analysis and warrant further
investigation.
Qualitative aspects
Cultural appropriateness
Experience in the United States suggests that there is little adjustment associated with
shifting a community from on‐site or centralised wastewater management to cluster‐
based systems if a centralised service provision approach is adopted (West 2003;
White 2004).
Human and ecological health
The public health risk of centrally managed distributed sanitation systems is assessed
to be equal or less than that associated with centralised wastewater management
given appropriate precautions are undertaken in irrigating with recycled wastewater.
A case stud
y im
plem
entatio
n
100
4.5.2 Ecological alternative
Figure 4‐30 depicts the time series of incremental costs associated with the ecological
sanitation alternative.
Figure 4‐30‐ Financial impact of ecological sanitation option
For the utility, there is a slight incremental avoided costs in the short term associated
with a slightly reduced capital investment per dwelling connected. This projected
saving is eroded with time owing to the incremental operational cost associated with
cartage, leading to a slight present value financial saving of $7 million. For the
municipal solid waste service provider, the avoided costs result in a net present value
financial saving of $400,000 over the life of the program. For society, the avoided costs
associated with reducing the discharge of effluent is again negligible however the
recovered nutrients lead to a net present value saving of $12.8 million. Such avoided
costs may be internalised over time by the utility as a market for human fertilisers is
established.
The aggregated stakeholder costs result in an economic present value saving of $20
million, which is approximately 5% of assumed baseline cost.
This projection fails to capture a number of additional avoided costs. Firstly and
principally, the projection has not accounted for additional transfer and treatment
‐$3,000,000
‐$2,500,000
‐$2,000,000
‐$1,500,000
‐$1,000,000
‐$500,000
$0
$500,000
30/6/2008
30/6/2010
30/6/2012
30/6/2014
30/6/2016
30/6/2018
30/6/2020
30/6/2022
30/6/2024
30/6/2026
30/6/2028
30/6/2030
30/6/2032
30/6/2034
30/6/2036
30/6/2038
30/6/2040
30/6/2042
30/6/2044
30/6/2046
30/6/2048
30/6/2050
Fina
ncial flow [$
/a]
Time [date]
Economy Utility Partner Society
A case stud
y im
plem
entatio
n
101
infrastructure as described in the section above. Secondly, as the system has been
specifically designed to enable the on‐site use of both greywater and compost, and
such actions reduce cartage fees considerably.
Qualitative aspects
Human and ecological health considerations
The on‐site use of composted faeces, urine and greywater is associated with a degree
of public health risk. The minimum storage time of composted faeces is a fundamental
barrier to the transmission of disease. Measures have been recommended to ensure
this is achieved: namely the option has recommended a rotary batch, rather than a
continuous process system. However the centralised collection of composted waste
was deemed to present an unacceptable risk of contamination, therefore post
treatment by active composting has been recommended and costed. Although this
additional barrier does not prevent ineffective treatment for reuse at the household
level, the reduced scale of reuse presents a significantly reduced risk level (WHO
2005).
For a similar reason the WHO does not specify a minimum storage time for on‐site use
of urine, as person‐to‐person transmission at the household level implies a significantly
higher risk. Therefore the requisite urine storage time for on‐site reuse may therefore
be significantly reduced along with the costs (though the full 6 month storage capacity
has been costed here). The WHO recommends a minimum storage time of 6 months
prior to the large‐scale agricultural reuse of urine. Swedish experience suggests this
may be cost‐effectively ensured by providing additional storage at the farm‐site,
though any implied additional costs have not been accounted in this study.
Finally, the option does not recommend any treatment or disinfection of greywater
beyond pre‐treatment for grease and fines, which may clog pumps, distribution and
receiving soils. This is based on current research suggesting the risks implied by
greywater reuse have been over‐estimated in the past, owing to the inappropriate use
of bacterial indicators as biomarkers for faecal contamination (WHO 2005, p. 14).
Nonetheless the option recommends sub‐surface irrigation, appropriate markings and
instructions be applied, consistent with the existing regulations. Another consideration
is the threat of excess nutrient burden upon soils and potential groundwater
A case stud
y im
plem
entatio
n
102
contamination. For this reason low‐phosphate detergents should be applied and locally
appropriate minimum irrigation areas should be specified by authorities. There is also
significant benefit associated with promoting the mixed household application of
greywater and urine to maintain optimal carbon to nitrogen levels for plant uptake
(Del Porto & Steinfeld 1999). Finally, a maximum storage time has also been adopted
as greywater is characterised by rapid biological oxygen demand and continued
storage beyond 24 hours usually results in oxygen depletion, anaerobic conditions and
its associated odour problems (Del Porto & Steinfeld 1999).
Cultural appropriateness
While there have been some limited studies with positive results (GHD 2003; Pollard,
Kohlenberg & Davison 1997), as yet there have been no large‐scale Australian studies
seeking to identify the attitudes and preconceptions toward purchasing or occupying
dwellings with composting toilets. There is therefore considerable uncertainty as to
whether the general Australian public possess the ability to make the necessary
adjustments to operate ecological sanitation systems.
Nonetheless the weight of international studies indicate that both a strong focus upon
the dissemination of user information and voluntary will are defining enabling factors
for successful, large‐scale urban ecological sanitation projects (Cordover & Knuth 2005;
Wegelin‐Schuringa 2000). The measure would therefore benefit from an active
awareness and mobilisation program encouraging and informing the responsible use
of treated human excreta‐ note that such a program has not been accounted in the
costs represented above. Also, the systems have been designed to provide facility for
both on‐site use and centralised management, allowing the user to decide the degree
of responsibility they are willing to assume.
A case stud
y im
plem
entatio
n
103
4.5.3 Assessing the least cost alternative
Having estimated the costs for each option, the least cost alternative to society is
assessed by discounting the economic costs upon all stakeholders (i.e. ignoring transfer
payments) to present value terms.
Consistent with the objective of the study, Figure 4‐31 provides the model output for
the levelised dwelling cost of the sanitation growth options, which is calculated by
discounting all incremental costs and dividing them by the discounted number of
dwellings serviced each year.
Figure 4‐31 ‐ Economic levelised costs of alternative sanitation growth servicing options (baseline transfers excluded)
The chart indicates the incremental levelised cost of $34 per dwelling associated with
the distributed sanitation option, which may be taken as marginal considering the
scale of investment; while the ecological sanitation option yields a levelised saving of
$685 per dwelling.
Given that the results do not account for any avoided costs associated with transfer
and treatment capital costs (i.e. they are assumed as sunk costs), this is an exceptional
result and implies significant savings may be realised if the full avoided costs are
accounted. The implication is that both distributed and ecological sanitation options
are likely to be cost‐effective substitutes to the baseline extension of the centralised
network.
‐$800 ‐$700 ‐$600 ‐$500 ‐$400 ‐$300 ‐$200 ‐$100 $0 $100
Ecological Sanitation
Distributed Sanitation
Economic levelised marginal dwelling cost [NPV(Economic cost)/NPV(Dwellings serviced)]
Conclusion
s and recommen
datio
ns
104
5 Conclusions and recommendations
The sanitation challenges identified within this study call for a reconsideration of the
dominant paradigm. Centralised wastewater systems in its current form has been
shown to be both prohibitively expensive and ecologically unsustainable, in emitting
unacceptable flows of water and nutrients.
The study has identified two key responses toward addressing the challenge, calling for
both a reconsideration of system scale, and a closed‐loop resource management
approach.
The least cost planning method and its supporting tool have both been
comprehensively reviewed and extended to provide a robust means of assessing
alternative sanitation futures. Key developments include an integrated resource
forecast of water and nutrients, a spatial method for developing demographic
projections, and a means of assessing the financial costs borne by all stakeholders.
The method and supporting tools were validated by empirical measurements in the
study area, and enabled a projection of the resource and financial implications of two
alternative sanitation futures.
However the study has a number of key limitations. As highlighted in the study the
means by which flows associated with non‐residential components were projected
could be significantly improved. In addition to large relatively constant demand
customers, baseline non‐residential efficiencies are considered to be likely, however in
the absence of supporting data the demand component was projected pro‐rata with
population. The projections of hydraulic flows into the future are therefore likely to be
overestimated.
Also, by failing to simulate the hourly peak wet weather flows and other dynamic
effects, the analysis failed to adequately capture the full avoided costs associated with
the alternatives. However nonetheless the analysis of the two future scenarios
revealed both to be cost‐effective, which was a surprising result. If the model
assumptions prove to be adequate, the implication would seem that a shift toward
more ecologically sustainable sanitation would indeed be a cost‐effective course of
action.
Conclusion
s and recommen
datio
ns
105
Finally, owing to time constraints, the analysis of the inherent uncertainty of the case
study projections has not been addressed. This is however a fundamentally important
dimension of the method and should be pursued further.
Despite these limitations, the study has demonstrated the benefits of extending the
least cost planning method across sectors and marks a significant third step as the
method has been transmitted first from energy planning, then water, and now
sanitation. The seemingly disparate sectors have been revealed during this path to be
characterised by essentially the same fundamental logic. A logical fourth step is
therefore to integrate these sectors into a coherent, truly integrated resource planning
framework.
Recommendations
A study of this scope inevitably results in a broad range of recommendations worthy of
future exploration, however the following measures are considered to present
significant opportunities.
In order to determine a more rigorous basis upon which to project non‐residential
sewage flows, further research is warranted to establish a process by which the
customer database may be applied, supported by consultations with major non‐
residential customers.
Further investigations are also warranted to investigate dynamic peak and wet
weather effects of wastewater systems as a means of informing an understanding of
future network capacity thresholds. Future research could include undertaking
hydraulic modelling of the sewer network to characterise both peaking factors for dry
weather flows, and the peak inflows associated with both existing and future sewerage
components.
Also, in order to establish a basis for analysing uncertainty in the model outputs,
further work is necessary to facilitate the application of stochastic inputs in the
supporting tool.
These measures combined should then form a sound basis for a broad‐based strategic
policy review undertaken in collaboration with a forward‐thinking water service
provider.
Conclusion
s and recommen
datio
ns
106
Finally the feasibility of integrating the least cost planning models of water, energy,
and sanitation presents a significant and strategic opportunity that should be explored
further. This is likely to be a significant undertaking, however this study has
demonstrated the value of the information arising from such an integration.
References
107
6 References
AC Nielsen 2007, Nielsen Convenience Report, viewed 5 May 2008 <http://www.c‐store.com.au/industry/viewer/pdfs/01.pdf>.
Allen Consulting Group 2006, The Economic Impacts of a National Emissions Trading Scheme, final report prepared for the National Emissions Trading Taskforce.
Almeida, M.C., Butler, D. & Friedler, E. 1999, 'At‐source domestic wastewater quality', Urban Water, vol. 1, no. 1, pp. 49‐55.
Australian Bureau of Statistics (ABS) 2002, 2001 Census Basics, Australian Bureau of Statistics. Australian Bureau of Statistics (ABS) 2006, Australian Historical Population Statistics, Australian
Bureau of Statistics, viewed 6 May 2008 <http://www.abs.gov.au/AUSSTATS/[email protected]/log?openagent&3105065001_table18.xls&3105.0.65.001&Data%20Cubes&23F533BC3E26D892CA2571760022856F&0&2006&23.05.2006&Latest>.
Australian Bureau of Statistics (ABS) 2007a, Australian Standard Geographical Classification Digital Boundaries, cat. no. 1259.0.30.001, Australian Bureau of Statistics, viewed 6 May 2008 <http://www.abs.gov.au/AUSSTATS/[email protected]/DetailsPage/1259.0.30.0012007?OpenDocument>.
Australian Bureau of Statistics (ABS) 2007b, Environmental issues : people's views and practices, Cat. no. 4602.0, Australian Bureau of Statistics, viewed 6 May 2008 <http://www.abs.gov.au/AUSSTATS/[email protected]/Lookup/4602.0Main+Features1Mar%202007?OpenDocument>.
Baumann, D.D., Boland, J. & Hanemann, W.M. 1997, Urban water demand management and planning, McGraw‐Hill, New York.
Beder, S. 1993, 'Pipelines and Paradigms: the development of sewerage engineering', Australian Civil Engineering Transactions, vol. CE35, no. 1, pp. 79‐85.
Belevi, H. 2002, 'Material flow analysis as a strategic planning tool for regional waste water and solid waste management', paper presented to the Globale Zukunft: Kreislaufwirtschaftskonzepte im kommunalen Abwasser‐ und Fäkalienmanagement, Munich, Germany.
Berry, T., Campbell, S., Riedy, C. & White, S. 2004, A New Distributed Infrastructure and Services Market: assessing Australia's future market potential, Institute for Sustainable Futures, Sydney.
Brunner, P.H. & Rechberger, H. 2003, Practical handbook of material flow analysis, CRC/Lewis, Boca Raton, FL.
Butler, D. 1993, 'The influence of dwelling occupancy and day of the week on domestic appliance wastewater discharges', Building and Environment, vol. 28, no. 1, pp. 73‐79.
Butler, D. & Davies, J.W. 2004, Urban drainage, 2nd edn, Spon Press, New York. Clark, R. 1997, Optimum Scale for Urban Water Systems, South Australian Department of
Environment and Natural Resources. Corbett, T. & Chan, A. 2007, The Melbourne Water sewer network model, Wallingford
Software, viewed 5 May 2008 <http://www.wallingfordsoftware.com/news/fullarticle.asp?id=671>.
Cordell, D., Drangert, J.‐O. & White, S. submitted, 'The Story of Phosphorus: Global food security and food for thought', submitted to Global Environmental Change.
Cordover, A. & Knuth, B. 2005, 'User satisfaction in large‐scale, urban dry sanitation programs in Mexico', Urban Water, vol. 2, no. 4, pp. 227‐243.
Del Porto, D. & Steinfeld, C. 1999, The composting toilet system book : A practical guide to choosing, planning and maintaining composting toilet systems, a water‐saving, pollution‐preventing wastewater solution, Center for Ecological Pollution Prevention, Concord, Mass.
References
108
Esrey, S.A., Andersson, I., Hillers, A. & Sawyer, R. 2000, Closing the loop: ecological sanitation for food security, UNDP, SIDA, Water and Sanitation Programme, Thrasher Research Fund and PAHO.
Fane, S., Robinson, J. & White, S. 2003, 'The use of levelised cost in comparing supply and demand side options', Water Science & Technology: Water Supply [Water Sci. Technol. Water Supply], vol. 3, no. 3, pp. 185‐192.
Foess, G., Steinbrecher, P., Williams, K. & Garrett, G. 1998, 'Cost and Performance Evaluation of BNR Processes', Florida Water Resources Journal, pp. 11‐16.
Food and Agriculture Organisation (FAO) 2007, Food Security Statistics, <http://www.fao.org/faostat/foodsecurity/Files/DietFoodGroupsProtein_en.xls>.
Food and Agriculture Organisation (FAO) & Consolidated Management Services (CMS) 1997, Biogas technology: A training manual for extension, Consolidated Management Services Nepal, <http://www.fao.org/waicent/faoinfo/sustdev/EGdirect/EGre0021.htm>.
Friedler, E. & Butler, D. 1996, 'Quantifying the inherent uncertainty in the quantity and quality of domestic wastewater', Water Science & Technology, vol. 33, no. 2, p. 13.
GHD 2003, Composting Toilet Demonstration Feasibility Study, prepared for the Victorian Department of Sustainability and Environment.
Growth from Knowledge (GFK) 2006, Greening White Goods: a report into the energy efficiency trends of major household appliances in australia from 1993 ‐ 2005 (detailed output tables), Australian Government Department of the Environment, Water, Heritage and the Arts, <http://www.energyrating.gov.au/library/pubs/200606‐gfkapp05.pdf>.
Gumbo, B. 2005, 'Short‐cutting the phosphorus cycle in urban ecosystems', PhD thesis, Delft University of Technology.
Holt, P. & James, E. 2006, Wastewater reuse in the Urban Environment: selection of technologies, prepared by Ecological Engineering for Landcom Corporation, <http://www.landcom.com.au/downloads/uploaded/Wastewater%20reuse%20technology%20report_links2_d960.pdf>.
Howe, C., Jones, R.N., Maheepala, S. & Rhodes, B. 2005, Implications of Potential Climate Change for Melbourne’s Water Resources, a collaborative project between Melbourne Water and CSIRO Urban Water and Climate Impact Groups.
Jacobs, H.E. & Haarhoff, J. 2004, 'Structure and data requirements of an end‐use model for residential water demand and return flow', Water SA, vol. 30, no. 3, p. 12.
Jasinki, S.M. 2008, Phosphate Rock Mineral Commodity Summary, United States Geological Survey, viewed 10 February 2008 <http://minerals.usgs.gov/minerals/pubs/commodity/phosphate_rock/mcs‐2008‐phosp.pdf>.
Jönsson, H. & Vinnerås, B. 2003, 'Adapting the nutrient content of urine and faeces in different countries using FAO and Swedish data', 2nd international symposium on ecological sanitation, Luebeck, Germany, <http://www2.gtz.de/ecosan/download/ecosan‐Symposium‐Luebeck‐session‐f.pdf>.
Kvarnström, E., Emilsson, K., Stintzing, A.R., Johansson, M., Jönsson, H., Petersens, E.a., Schönning, C., Christensen, J., Hellström, D., Qvarnström, L., Ridderstolpe, P. & Drangert, J.‐O. 2006, Urine Diversion: one step towards sustainable sanitation, EcoSanRes, SEI, Stockholm.
Mara, D., Drangert, J.O., Anh, N.V., Tonderski, A., Gulyas, H. & Tonderski, K. 2007, 'Selection of sustainable sanitation arrangements', Water Policy, vol. 9, no. 3, pp. 305‐318.
Maurer, M., Rothenberger, D. & Larsen, T. 2005, 'Decentralised wastewater treatment technologies from a national perspective: at what cost are they competitive?' Water Science & Technology: Water Supply [Water Sci. Technol. Water Supply], vol. 5, no. 6, pp. 145‐154.
McLennan Magasanik Associates (MMA) 2006, Impacts of a National Emissions Trading Scheme on Australia’s Electricity Markets, report prepared for the National Emissions Trading Taskforce.
References
109
Melbourne Water Corporation 1998, Annual Environment and Public Health Report 1997‐98, 2008 <http://www.melbournewater.com.au/content/library/publications/reports_archive/performance_reports/1997‐1998_Environment_and_Public_Health_Report.pdf>.
Melbourne Water Corporation 1999, Environment and Community Obligation Report 1998‐99, 2008 <http://www.melbournewater.com.au/content/library/publications/reports_archive/performance_reports/1998‐1999_Environment_and_Community_Obligation_Report.pdf>.
Melbourne Water Corporation 2000, 1999/2000 Environment Report, 2008 <http://www.melbournewater.com.au/content/library/publications/reports_archive/performance_reports/1999‐2000_Environment_Report.pdf>.
Melbourne Water Corporation 2001, 2000/01 Environment Review, <http://www.melbwater.com.au/content/library/publications/reports_archive/annual_reports/2000‐2001_Environment_Review.pdf>.
Melbourne Water Corporation 2002, Environment Review 2001/02, <http://www.melbournewater.com.au/content/library/publications/reports_archive/annual_reports/2001‐2002_Environment_Review.pdf>.
Melbourne Water Corporation 2003, Water Resources and the Environment 2002/03, <http://www.melbournewater.com.au/content/library/publications/reports_archive/annual_reports/2002‐2003_Water_Resources_and_the_Environment_Report.pdf>.
Melbourne Water Corporation 2004, Social and Environment Report 2003/04, <http://www.melbournewater.com.au/content/library/publications/reports_archive/annual_reports/2003‐2004_Social_and_Environment_Report.pdf>.
Melbourne Water Corporation 2005, Social and Environment Report 2004‐05, <http://www.melbournewater.com.au/content/library/publications/reports_archive/annual_reports/2004‐2005_Social_and_Environment_Report.pdf>.
Melbourne Water Corporation 2006, Social and Environment Data 2005/06, <http://www.melbournewater.com.au/applications/social_environment_report_2005_06/content/environmental_performance/western_treatment_plant_inputs_and_discharges.asp>.
Melbourne Water Corporation 2007, Social and Environment Data 2006/07, <http://www.melbournewater.com.au/applications/annual_report_2007/envdata/env_wesplant.htm>.
Melbourne Water Corporation 2008, About the Northern Sewerage Project, viewed May 4 2008 <http://www.nsp.net.au/content/about_the_project.asp>.
Mitchell, C., Fane, S., Willetts, J., Plant, R. & Kazaglis, A. 2007, Costing for Sustainable Outcomes in Urban Water Systems ‐ A Guidebook, CRC for Water Quality and Treatment, Research Report 35.
Motangero, A., Nguyen, T. & Belevi, H. 2004, 'Material Flow Analysis as a Tool for Environmental Sanitation Planning in Viet Tri, Vietnam', paper presented to the 30th WEDC Conference, Vientienne, Lao PDR.
Mumford, L. 1961, The city in history : its origins, its transformations, and its prospects, [1st edn, Harcourt, Brace & World, New York.
Narain, S. 2002, 'A Tale of Two Cities', Down to Earth, <http://www.ecosan.nl/content/download/542/4177/file/Down%20to%20Earth%20flush%20toilet%20ecologically%20mindless.pdf>.
New South Wales Department of Planning (NSW Planning) 2007, NSW SLA Population Projections, 2005 Release ‐ LGA Summary, Version 1.0, NSW Department of Planning, <http://www.transport.nsw.gov.au/tdc/local‐pop.html>.
Nutrition Australia 2008, How much protein?, viewed 15 May 2008 <http://www.nutritionaustralia.org/food_facts/faq/protein_day_faq.asp>.
Orenco Systems 2007, Effluent Sewers, <http://www.orenco.com/pdfs/ABR‐EFS‐1rev23.pdf>. Otis, R. & Mara, D. 1985, The Design of Small Bore Sewers, The World Bank, Washington.
References
110
Otterpohl, R. 2001, 'The colours of New Sanitation ‐ Black, Brown, Yellow, Grey ‐ Examples of High‐, Medium‐ and Low‐Tech Solutions for urban and rural areas', Water 21.
Patterson 2007, Laundry products research, Lanfax Laboratories, viewed 12 April 2008 <http://www.lanfaxlabs.com.au/papers/Na‐P‐graph‐2005.pdf>.
Pinkham, R., Hurley, E., Watkins, K., Lovins, A., Magliaro, J., Etnier, C. & Nelson, V. 2004, Valuing Decentralized Wastewater Technologies: A Catalog of Benefits, Costs, and Economic Analysis Techniques, Rocky Mountain Institute.
Pollard, R., Kohlenberg, A. & Davison, L. 1997, 'Effectiveness and User Acceptance of Composting Toilet Technology in Lismore, NSW', Environmental Technologies for Wastewater Management, UNEP, Murdoch University, Perth.
Protection of the Environment Operations (General) Regulation 1998. Roberts, P. 2004, Appliance Stock and Usage Pattern Survey, Yarra Valley Water, viewed 5 May
2008 <http://www.yvw.com.au/NR/rdonlyres/4CCF0B18‐4E7F‐4F0B‐AD17‐98A63F503606/0/ASUPReportFinal.pdf>.
Roberts, P. 2005, Residential End Use Measurement Study, Yarra Valley Water, viewed 5 May 2008 <http://www.yvw.com.au/NR/rdonlyres/983166CA‐B506‐4C08‐8B95‐5EDD378CFB53/0/REUMSReportFinal.pdf>.
Rockström, J., Göran, N.A., Falkenmark, M., Lannerstad, M., Rosemarin, A., Caldwell, I., Arvidson, A. & Nordström, M. 2005, Sustainable Pathways to Attain the Millenium Development Goals: Assessing the Key Role of Water, Energy and Sanitation, Stockholm Environment Institute, Stockholm International Water Institute.
Rose, C. 2004, Fertiliser Calculations, Department of Primary Industries, <http://www.dpi.nsw.gov.au/__data/assets/pdf_file/0004/166153/fertiliser‐calculations.pdf>.
Rosmarin, A. 2004, 'Nitrogen and Phosphorus: at war with two elements', Down to Earth, no. 30 June, pp. 27‐35.
Smakhtin, V., Revenga, C. & Döll, P. 2004, Taking into Account Environmental Water Requirements in Global‐scale Water Resources Assessments, International Water Management Institute, Colombo.
Steen, I. 1998, 'Phosphorus availability in the 21st century Management of a non‐renewable resource', Phosphorus & Potassium, no. 217.
Strategic Finance Group (SFG) 2007, Cost of Capital for Metropolitan Water Businesses, report prepared for City West Water, Melbourne Water, Southeast Water & Yarra Valley Water.
Tangsubkul, N., Moore, S. & Waite, T.D. 2005, 'Incorporating phosphorus management considerations into wastewater management practice'.
Tchobanoglous, G., Burton, F.L. & Stensel, H.D. 1991, Wastewater engineering : treatment, disposal, and reuse / Metcalf & Eddy, Inc., 3rd ed. / revised by George Tchobanoglous, Franklin L. Burton edn, McGraw‐Hill, New York :.
UNICEF & WHO 2004, Meeting the MDG Drinking Water and Sanitation Target: a Mid‐Term Assessment of Progress, WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation, New York and Geneva.
United Nations 2000, United Nations Millennium Declaration, UN General Assembly, 55 edn. United Nations Development Program (UNDP) 2006, Beyond scarcity: power, poverty and the
global water crisis, Human Development Report 2006, New York. United Nations Environment Programme (UNEP) 2005, Millennium Ecosystems Assessment,
United Nations Environment Programme. Victorian Department of Planning and Community Development (Vic Planning) 2004, Victoria
in Future 2004 Outline Report, Victorian Department of Planning and Community Development, <http://www.dse.vic.gov.au/DSE/dsenres.nsf/LinkView/8B232276A311D5F1CA256EF6001BCFE706C7DF80826B65674A256DEA002C0DCA>.
Victorian Department of Planning and Community Development (Vic Planning) 2007, Victoria in Future 2004 Projections, Victorian Department of Planning and Community
References
111
Development, <http://www.dse.vic.gov.au/DSE/dsenres.nsf/LinkView/B9023E3BAACA5A6ACA256EF60019E55806C7DF80826B65674A256DEA002C0DCA>.
Vinnerås, B., Palmquist, H., Balmér, P. & Jönsson, H. 2006, 'The characteristics of household wastewater and biodegradable solid waste ‐ a proposal for new Swedish design values', Urban Water Journal, vol. 3, no. 1, pp. 3 ‐ 11.
Wang, F.S. 2001, 'Deterministic and Stochastic Simulations for Solid Waste Collection Systems ‐ A SWIM Approach', Environmental Modeling & Assessment, vol. 6, no. 4, p. 11.
Wang, F.S., Richardson, A.J. & Roddick, F.A. 1998, 'SWIM ‐ Modelling Approach and Its Applications', paper presented to the Systems engineering models for waste management, Göteborg, Sweden, 25‐26 February.
Wegelin‐Schuringa, M. 2000, 'Public Awareness and Mobilisation for Ecosanitation', paper presented to the International Symposium on Ecological Sanitation, Bonn, Germany, 30‐31 October.
West, S.M. 2003, Innovative On‐site and Decentralised Sewage Treatment Reuse and Management Systems in Northern Europe & the USA: report of a study tour ‐ February to November 2000.
White, K. 2004, 'Centralised and Decentralised Wastewater Management Alternatives: Functional Models and Infrastructure Costs', paper presented to the Enviro 06, Melbourne, 9‐11 May 2006.
Wittwer, J.W. 2004, Monte Carlo Simulation in Excel: a practical guide, Vertex42.com, viewed May 12 2008 <http://www.vertex42.com/ExcelArticles/mc/>.
World Health Organization (WHO) 2005, Guidelines for the safe use of wastewater, excreta and greywater, WHO / UNEP / FAO, Geneva.
112
Appendix A Baseline assumptions and descriptions report
The following table is a model output containing the assumptions and descriptions associated
with each of the components of the baseline resource projection.
Components in Western Wastewater Catchment
Description Notes Assumptions
Res Dishwashing End use model baseline
component for dishwashing
activities
calculates demand
associated with residential
dishwasher use based upon
normal appliance stock
model, exponential decay
consumption model, and
FPLE ownership model
10 [Lifetime of appliance - used in stock modelling. as Years] based on sales data and information from
manufacturers (ECU 1995)
0.25 [Standard deviation - used in stock modelling.] based on sales data and information from
manufacturers (ECU 1995)
23.7957029620511 [Water consumed by each dishwashing activity using kitchen sinks as litres / activity]
based upon average filled capacity 25L filled to an average 0.54% filled volume (Roberts 2004) with the
remaining rinsing component calibrated against a historical end use study in Perth (MWA 1985). The
output estimates 50:50 split for filling:flowing activity types, which is consistent with the findings of a study
in the UK (Friedler & Butler 1996)
[Mean water consumption per wash for dishwashers as L / use] based upon exponential decay model with
asymptote assumed at 14L/activity. The historical consumption data is based on the figures presented in
Greening White Goods study for period 1993-2005 (GfK 2005)
[Proportion of dwellings with dishwashers as %] based upon FPLE model, ownership saturation assumed
to occur at 65% (guesstimate)
[Number of operational dishwashers within residential dwellings as appliances] based upon dishwasher
ownership model multiplied by total residential dwellings
[Mean frequency of washes per single residential dwelling each year as uses / household.year] based
upon an linear interpolation of per capita usage study undertaken for Melbourne (Roberts 2005) multiplied
by the occupancy ratio for single residential dwellings
[Mean frequency of washes per multi residential dwelling each year as uses / household.year] based upon
an linear interpolation of per capita usage study undertaken for Melbourne (Roberts 2005) multiplied by
the occupancy ratio for multi residential dwellings
5.9 [Mean frequency of sink handwashing per household per week for dwellings with dishwashers as
activities / dwelling.week] based upon stock usage survey in Melbourne (Roberts 2004)
10.1 [Mean frequency of sink handwashing per household per week for dwellings without dishwashers as
activities / dwelling.week] based upon stock usage survey in Melbourne (Roberts 2004)
114
Res Clotheswashing End use model baseline
component for
clotheswashing activities
calculates demand
associated with residential
clotheswasher use based on
lognormal stock model,
historical and forecasted
consumption (assumed to
plateau in 2007), frequency
model (linear to occupancy
ratio), and FPLE sales
forecast model
10 [Lifetime of appliance - used in stock modelling. as Years] based on sales data and information from
manufacturers (ECU 1995)
0.5 [Standard deviation - used in stock modelling.] based on sales data and information from
manufacturers (ECU 1995)
[Water consumed by laundry trough per household per day as litres/hh.day] based upon study conducted
in Perth finding 22 L/hh.day, adjusted to be inversely proportional to clotheswashing machine ownership in
that sample year (MWA 1985)
[Sales of top loading clotheswashers as appliances] based on ownership and lifetime assumptions.
[Sales of front loading clotheswashers as appliances] based upon ownership and lifetime assumptions
[Mean frequency of washes per multi residential dwelling per year as uses / household.year] based upon
an linear interpolation of per capita usage study undertaken for Melbourne (Roberts 2005) multiplied by
the occupancy ratio for multi residential dwellings
[Mean frequency of washes per single residential dwelling per year as uses / household.year] based upon
an linear interpolation of per capita usage study undertaken for Melbourne (Roberts 2005) multiplied by
the occupancy ratio for single residential dwellings
[Water consumed by top-loading clotheswashers per wash as litres / use] based upon linear decay model
with asymptote assumed at 115L/use The historical consumption data is based on the figures presented in
Greening White Goods study for period 1993-2005 (GfK 2005)
[Water consumed by front-loading clotheswashers per wash as litres / use] based upon linear decay
model with asymptote assumed at 70L/use The historical consumption data is based on the figures
presented in Greening White Goods study for period 1993-2005 (GfK 2005)
[Proportion of total clotheswasher stock that are front loaders as %] based on stock and penetration data
for Front Loaders sourced from ABS (1987) survey, EES (1999) & ABS (2002). REF: ABS (1987)
Domestic Water Use NSW July 1987, Australian Bureau of Statistics; ABS (2002) Domestic Water Use
NSW October 2002, Australian Bureau of Statistics.
[Proportion of total clotheswasher stock that are top loaders as %] proportion based upon stock and
penetration data for Top Loaders sourced from ABS (1987) survey, EES (1999) & ABS (2002).
[Number of operational front-loading clotheswashers within residential dwellings as appliances] Stock and
penetration data from ABS (1987) survey and EES (1999).
[Number of top-loading clotheswashers within residential dwellings as appliances] Stock and penetration
data from ABS (1987) survey and EES (1999).
[Proportion of dwellings with operational clotheswashers as %] based upon linear interpolation of ABS
115
data (ABS 2005 4602.0 March) to 2005 and assumed market saturation (i.e. constant ownership) after
2005
[Total number of clotheswashers of all types as appliances] based upon modelled penetration multiplied
by total number of residential dwellings
3.58600609209819 [Phosphorus consumed each wash by front loading washing machines in residential
dwellings as g/activity] based upon measurement of phosphorus load of common detergents under
recommended dosage (Patterson 2007), weighted based upon national detergent market shares (ACN
2007)
4.33200143343487 [Phosphorus consumed each wash by top loading washing machines in residential
dwellings as g/activity] based upon measurement of phosphorus load of common detergents under
recommended dosage (Patterson 2007), weighted based upon national detergent market shares (ACN
2007)
Res Bathing End use model baseline
component for bathing
activities
calculates demand
associated with residential
shower use based on a 2-
component stock model and
constant assumed flowrates,
duration and frequency;
calculates demand
associated with residential
bath use based on average
bath volume, age split (>12
years), and their respective
fill volumes, frequencies
13 [Lifetime of appliance - used in stock modelling. as Years] based on sales data and information from
manufacturers (ECU 1995)
0.5 [Standard deviation - used in stock modelling.] based on sales data and information from
manufacturers (ECU 1995)
112.5 [Mean water consumed each bath as litres / bath] based on average bath capacity and fill volumes
from an appliance usage and stock survey of Melbourne (Roberts 2004) finding average capacity of 200L,
with fill volumes of 35% and 60% for residents under and over the age of 12 years- adjusted for the local
age distribution
24.44 [Frequency of taking a bath per person per year as baths / person.year] based upon frequencies
drawn from appliance usage and stock survey of Melbourne (Roberts 2004) finding an average 2 and 0.2
baths per week for residents under and over the age of 12 respectively, adjusted for the local age
distribution.
4.9 [Typical operating flow rate for residential hand basins as litres / minute] based upon appliance stock
and usage pattern surveys in Melbourne (Roberts 2004)
[Proportion of toilet stock that are inefficient models as %] based on a sigmoid penetration model fitted to
ABS state-wide efficient showerhead penetration figures (deflated to account for overreporting). Assumes
the total stock penetration is equal to the total usage of that model
[Proportion of toilet stock that are efficient models as %] based on a sigmoid penetration model fitted to
ABS state-wide efficient showerhead penetration figures (deflated to account for overreporting). Assumes
the total stock penetration is equal to the total usage of that model
116
[Weighted stock average of showerheads per dwelling as units / dwelling] based upon modelling
undertaken by Wilkenfeld (unpublished)
0.85 [Mean frequency of showers per person per day as Showers / person.day] based upon Yarra Valley
end use measurement study (Roberts 2005), response deemed more accurate than 2004 and 1999 study
10.5 [Mean shower flow rate for inefficient models adjusted for throttle-back as Litres / minute] based upon
end use measurement survey of Melbourne (Roberts 2005)
7.6 [Mean shower flow rate for efficient models adjusted for throttle-back as Litres / minute] based upon
end use measurement survey of Melbourne (Roberts 2005)
7.1 [Mean duration of showers as Minutes / shower] based on two studies, each independently finding a
similar mean in Perth (Loh & Coghlin 2003) and Melbourne (Roberts 2005). Both studies also found no
observed difference in shower duration between those with normal and efficient showerheads.
[Annual percentage of total showerhead sales that are efficient models as %] based upon an assumed
sigmoid uptake, calibrated to adjusted efficient showerhead penetration data (ABS 2007).
5.5 [Frequency of hand basin activities as activities / day] based upon appliance stock and usage pattern
surveys in Melbourne (Roberts 2004)
0.33 [Typical duration of hand basin activity as minutes / activity] based upon appliance stock and usage
pattern surveys in Melbourne (Roberts 2004)
Res Toilet End use model baseline
componenet for toilet
flushing
calculates demand
associated with residential
toilet flushes based upon
fitted lognormal stock model,
frequency and consumption
estimates
30.0261416947322 [Lifetime of appliance - used in stock modelling. as Years] calibrated to observed
decay of single flush toilet stock
9.9 [Mean water consumption each toilet flush for single flush models as litres / flush] based on flush
volume from an end use measurement study in Melbourne (Roberts 2005)
7.186 [Mean water consumption each flush for 11/6L dual flush models as litres / flush] based on cistern
volumes from end use measurement study in Melbourne (Roberts 2005) and full flush frequency factors
inferred from AC Nielsen study yielding 0.33 full / half flush for larger cisterns
[Mean number of toilets within residential households as toilets / household] based upon two domestic
water use studies in Perth over a 18 year period (MWA 1985, Loh & Coghlin 2003) extrapolated and
interpolated linearly
0.351962380455425 [Standard deviation - used in stock modelling.] calibrated to observed decay of single
flush toilet stock
5.687 [Mean water consumption each flush for 9/4.5L dual flush models as litres / flush] based on flush
volumes from end use measurement study in Melbourne (Roberts 2005) and full flush frequency factors
inferred from AC Nielsen study yielding 0.33 full / half flush for larger flush volumes
117
4.15 [Mean water consumption each flush for 6/3L dual flush models as litres / flush] based on flush
volumes from end use measurement study in Melbourne (Roberts 2005) and full flush frequency factors
inferred from AC Nielsen study yielding 0.5 full / half flush for small flush volumes
[Proportion of dwellings with single flush toilets as %] based on cohort stock model
1.09575 based upon ISF research (1998)
[Proportion of toilet stock that are 11L/6L dual flush models as %] based on cohort stock model
[Proportion of toilet stock that are 6/3L dual flush models as %] based on cohort stock model
[Proportion of toilet stock that are 9/4.5L dual flush models as %] based on cohort stock model
[Proportion of toilet stock that are 4/3L dual flush models as %] based on cohort stock model
3.75 [Mean water consumption each flush for 4/3L dual flush models as litres / flush] based on flush
volumes from end use measurement study in Melbourne (Roberts 2005) and full flush frequency factors
inferred from AC Nielsen study yielding 0.5 full / half flush for small flush volumes
3.8 [Mean frequency of flushes within residential dwellings per person per day as flushes / person.day]
based upon end use measurement study in Melbourne (Roberts 2005)
0.71455098 [Phosphorus excreted at home each day per person as g/person.day] based upon national
dietary protein intake data (FAO 2007) apportioned to phosphorus as 1.1% of dietary protein, of which
55% is urine (Jönsson & Vinnerås 2003), adjusted by 78% for 'at-home' toilet use assuming approximately
40 working week and constant metabolic rate of excretion
0.6745761 [Phosphorus excreted in faeces at home each day per person as kg/a] based upon national
dietary protein intake data (FAO 2007) apportioned to phosphorus as 1.1% of dietary protein, of which
45% is faecal (Jönsson & Vinnerås 2003), adjusted by 90% for 'at-home' toilet use
9.73947 [Nitrogen excreted as urine at home each day per person as kg/a] based upon national dietary
protein intake data (FAO 2007) apportioned to nitrogen as 13% of dietary protein, of which 85% is urine
(Jönsson & Vinnerås 2003), adjusted by 78% for 'at-home' toilet use assuming approximately 40 working
week and constant metabolic rate of excretion
1.98315 [Nitrogen excreted as faeces at home each day per person as kg/a] based upon national dietary
protein intake data (FAO 2007) apportioned to nitrogen as 13% of dietary protein, of which 15% is urine
(Jönsson & Vinnerås 2003), adjusted by 90% for 'at-home' toilet use
Appendix B Stock model builder interface and code
The key components of the interface referred within the code include the following
• txtStartYear and txtEndYear the text box for the user to input the start and
end year of the stock model
• optPrimaryInput and optSecondaryInput the option buttons to specify
whether the required stock model is to be a primary or total type, or a
secondary or component type
• optLognorm and optNormdist the option buttons to specify whether the
stock model is based upon a normal or lognormal decay rate
The code has been included below.
120
Private Sub cmdBuildModel_Click() 'Open template file Workbooks.Add Template:=ThisWorkbook.Path & "\Stock.xltm" ' Validate user input If IsNumeric(txtStartYear.Text) And IsNumeric(txtEndYear.Text) Then ' Draws inputs from interface Start_Year = txtStartYear.Value End_Year = txtEndYear.Value Else MsgBox ("The input in the text boxes is invalid. Please enter a year") End End If Length = End_Year ‐ Start_Year + 1 With Worksheets("Stock") .Range(.Cells(6, 3), _ .Cells(6 + Length, 3 + Length)).NumberFormat = "_(* #,##0_);_(* (#,##0);_(* ""‐""_);_(@_)" .Range(.Cells(6, 3), _ .Cells(6 + Length ‐ 1, 3 + Length ‐ 1)).Value = 0 End With For i = 1 To Length ' Writes the date across the top of the worksheet Worksheets("Stock").Cells(1, i + 2).Value = "30/6/" & Start_Year ‐ 1 + i Worksheets("Stock").Cells(i + 5, 1).Value = "30/6/" & Start_Year ‐ 1 + i Next i ' Inserts the appliance lifetime decay function If optLognorm = True Then For j = 1 To Length ‐ 1 For i = 1 To Length ‐ j Sale_Year = RowCol_Absolute(i + 2) Age_Year = RowCol_Absolute(i + 2 + j) Worksheets("Stock").Cells(i + 5, i + 2 + j).Formula = _ "=(1‐LOGNORMDIST(YEAR(" & Age_Year & "$1)‐YEAR($" & Sale_Year & "$1),LN(Lifetime),StDev))*$" & Sale_Year & "$" & i + 5 Next i Next j Else For j = 1 To Length ‐ 1 For i = 1 To Length ‐ j
121
Sale_Year = RowCol_Absolute(i + 2) Age_Year = RowCol_Absolute(i + 2 + j) Worksheets("Stock").Cells(i + 5, i + 2 + j).Formula = _ "=(NORMSDIST((Lifetime+YEAR($" & Sale_Year & "$1)‐YEAR(" & Age_Year & "$1))*StDev))*$" & Sale_Year & "$" & i + 5 Next i Next j End If If optPrimaryInput = True Then Worksheets("Stock").Cells(6, 3).Formula = "=$C$3" Worksheets("Stock").Cells(4, 3).Formula = "=C6" For i = 2 To Length ' Inserts the sales row Worksheets("Stock").Cells(4, i + 2).Formula = "=" & RowCol_Absolute(i + 2) & i + 5 ' Inserts the sumation for estimating the in year sales Worksheets("Stock").Cells(i + 5, i + 2).Formula = _ "=" & RowCol_Absolute(i + 2) & "$3 ‐ SUM(" & RowCol_Absolute(i + 2) & "$6:" & RowCol_Absolute(i + 2) & i + 4 & ")" Next i Else Worksheets("Stock").Cells(6, 3).Formula = "=$C$4" Worksheets("Stock").Cells(3, 3).Formula = "=SUM(C6:C" & Length + 5 & ")" For i = 1 To Length ‐ 1 ' Inserts the total stock row Worksheets("Stock").Cells(3, i + 3).Formula = _ "=" & "SUM(" & RowCol_Absolute(i + 3) & 6 & ":" & RowCol_Absolute(i + 3) & Length + 5 & ")" ' Inserts the sumation for estimating the in year sales Worksheets("Stock").Cells(i + 6, i + 3).Formula = _ "=$" & RowCol_Absolute(i + 3) & "$4" Next i End If Application.Calculation = xlAutomatic End Sub
122
Appendix C Example stochastic analysis macro
Private Sub cmd_RunSimulation_Click()
Dim MaxIterations As Long
Dim IterationNum As Long
‘ clear results sheet
Worksheets("Results").Range("A7:D65536").Clear
‘ draw user input defining the necessary number of iterations
MaxIterations = Range("NumSimulations")
‘ loop for the specified number of iterations
For IterationNum = 1 To MaxIterations
‘ re‐calculate the worksheet
Worksheets("Data").Calculate
‘ output the result to the results spreadsheet
Worksheets("Results").Cells(IterationNum + 5, 1).Value = _
Int(IterationNum)
Worksheets("Results").Cells(IterationNum + 5, 2).Value = _
Worksheets("Data").Range("Result1")
Worksheets("Results").Cells(IterationNum + 5, 3).Value = _
Worksheets("Data").Range("Result2")
‘ and so on
Next IterationNum
End Sub
Appendix D IWA conference paper
The following paper was presented to the International Water Association Specialised Conference on Small and Decentralised Wastewater Systems, held in Coimbatore, India in February 2008.
8th IWA Specialized Conference on Small Water and Wastewater Systems (SWWS) and 2nd IWA Specialized Conference on Decentralised Water andWastewater International Network (DEWSIN), Coimbatore, India - February 6th to 9th, 2008 232
Valuing Sustainable Sanitation: the economic assessment ofalternative sanitation programsJ. L. McKibbin*, J. Willetts*, K. White** and P. Hagare***
* Institute for Sustainable Futures, University of Technology, Sydney, 235 Jones St, Broadway, Australia(Email: [email protected]; [email protected])** Department of Civil Engineering, University of South Alabama, EGCB 280 Mobile, AL 36688, US(Email: [email protected])*** Faculty of Engineering, University of Technology, Sydney, 15 Broadway, Ultimo, 2007, Australia(Email: [email protected])
Abstract This paper describes how an innovative costing method may be applied to compare thecost-effectiveness of alternative sanitation programs. The method is suggested as a tool capable offairly comparing a broad mix of responses to sanitation challenges including capacity demandmanagement and the full range of system scales.Keywords Cost-effectiveness analysis; material flow analysis; decentralised wastewater systems
INTRODUCTIONThe challenge of sustaining sanitation to rapidly sprawling cities has prompted a rethink in the waywe manage our waste. In industrial countries, conventional centralised, large-scale wastewatersystems have been the subject of renewed scrutiny in light of rising collection and treatment costs.In developing countries faced with limited financial and institutional capacity, the value ofinvesting and maintaining conventional western sanitation solutions is also under serious question(Newman 2001). A revolution in sanitation is under way through reconsideration of existingtechnologies such as dry composting, and the arise of new technologies including urine diversion,pressurised effluent carriage and modular small-scale treatment which offer new opportunities fordistributed systems with infrastructure scaled at the on-site and cluster scale (Pinkham et al. 2004).
A number of recent studies have sought to quantify the benefit of these new approaches andtechnologies (Kazaglis & Kraemer 2006; Pinkham et al. 2004; White 2004), however a holistic andrigorous cost-effectiveness analysis from both societal and stakeholder cost perspectives is yet tobe achieved. Such a study is important in demonstrating the value of these new models to decision-makers and driving their uptake by the mainstream.
This paper describes how a new costing method developed by the Institute for Sustainable Futuresin collaboration with five Australian water authorities (Mitchell et al. 2007) may be used toeffectively demonstrate in which situations the benefits of decentralised systems render them cost-effective alternatives to conventional centralised systems.
The method is a form of cost-effectiveness analysis, which involves developing alternative sets ofoptions to meet a set of agreed study objectives and assessing each alternative on the basis of itsincremental life cycle cost. The benefits of this method include:
- a balanced comparison of all available options is enabled by outcome-based objectives;- the full impact of options is accounted by intentionally defined system boundaries and
thorough life cycle costing;- both economically optimal and financially viable solutions are revealed by analysis of
stakeholder and societal cost perspectives; and- externalities and uncertainties are transparently treated.
The following section summarises our recent experience of how the method may be applied tocompare decentralised and centralised sanitation options.
METHOD1. Establish the analytical framework: Define the objectives of the analysis and specify theeconomic criteria by which the alternatives will be assessed including an appropriate treatment ofexternalitiesThe drivers and constraints of the study would ideally be decided on by stakeholders affected bythe costing study, and therefore may require a water authority to initiate others’ participation onthis decision. Potential objectives may include complying with minimum environmental standards,reducing the disposal of nutrients to the receiving river system or conserving water or nutrients.The economic criteria are then formed to ensure a balanced comparison of all options is possible.For assessing sanitation systems, a key decision will be the time scale of the net present valuecalculation, as this will control whether the entire life of the assets is accounted. Stakeholdersshould also agree upon an appropriate discount rate that will reflect the time value of money andthe cost of capital. An appropriate response is then defined for all significant actual or avoidedsocietal costs that are not reflected in market exchanges (e.g. eutrophication of rivers). Externalitiesmay be internalised within the costs using shadow pricing, embedded within the study objectives,or accounted externally using a quantitative or qualitative assessment.
2. Identify the system: Inventory the key components of the system, identify their interactions andmodel them within a resource balance.Material flow analysis forms the foundation of the method as the best means of understanding aproblem that is principally concerned with the management of resources (Brunner & Baccini1992). This implies identifying the inputs, stocks and outputs of each system component andlinking them within a model- a “resource balance model”. For the purpose of this study watervolumes have been the principal substance under review owing to their role in constrainingtreatment plant capacity. Recent research has also enabled us to quantify the key phosphorus flows(i.e. within excreta and detergents) that represent a key indicator of eutrophication and nutrientdepletion (Esrey et al. 2000; Gumbo 2005; Rockström et al. 2005; Tangsubkul, Moore & Waite2005). Gumbo (2005) has developed a useful series of water and phosphorus models for thispurpose during a study of a micro-catchment in the city of Harare, Zimbabwe (see below).
Figure 1. Material flow analyses for water and phosphorus in urban systems (Gumbo 2005)
When considering a sanitation system, the boundaries of the system are first carefully drawn tomake clear which impacts associated with a given option have been included or excluded. Forcentralised sanitation projects, the boundaries should preferably be drawn inclusive of both the newdevelopment and any existing system components upon which the development may rely. Theanalysis is then applied by quantifying the waste flow and its constituent components fromhouseholds (e.g. by estimating the wastewater generated per capita) and relating this to the capacityrequired by the sanitation system and the subsequent constraints associated with the receivingenvironment.
3. Specify the base case: Define and model the system configuration that a conventional approachwould imply in the futureIn order to later assess the incremental cost of an alternative, it is useful to first predict all futureactions associated with following the conventional, centralised approach. In the case of arequirement for commission of a new centralised treatment plant, any staged augmentations asdesign flows increase, and all operation and maintenance of the system over the life of thecomponents are included. In the case of extending an existing system, the base case will ofteninvolve extensive capacity upgrades to the collection system (e.g. trunk drainage) or treatmentsystem. It is critical to include all such costs so that later when the base case is compared toalternatives (eg a decentralised system), it is possible to demonstrate which of these costs may beavoidable.
4. Identify the options and specify alternatives to the base case: Inventory the broad suite ofoptions available toward meeting the study objectives and estimate their costsThe options may be modelled within the “resource balance model” to determine any base caseinvestments that may be avoided or delayed. For example, a delay in the staging of existingcentralised treatment capacity will have a significant impact upon the incremental cost of thealternative. Spreadsheet option models for both a hypothetical centralised base case and adecentralised alternative are presented below.
Figure 2. Spreadsheet output of annual capacity and demand for two alternative systems over time:the centralised base case (left) and the clustered alternative (right). Note the ability of the cluster-based system to respond to changes in demand as they arise.
In addition to considering smaller-scale or decentralised sanitation systems, capacity demandmanagement options including water conserving devices and source separation may be considered.
5. Analyse the costs: Apply discounted cash flow analysis to each of these alternativesThe key innovation of the method is the extension of material flow analysis toward analysingfinancial flows from alternative cost perspectives. As the costs associated with each option areestimated, careful attention is paid to which stakeholder the cost burden will fall (e.g. thehomeowner, the utility, the developer etc). In addition, all transfer payments (e.g. sewer fees,developer contributions) are accounted separately. Having estimated the costs for each option, theleast cost alternative to society is assessed by discounting the economic costs upon all stakeholders(i.e. ignoring transfer payments) to present value terms. The impact of the least cost alternativeupon each stakeholder’s financial viability may then in turn be assessed by including all costburdens and transfer payments from and to each stakeholder. This may then form the basis for re-negotiating transfer payments to ensure the avoided costs (and therefore the incentives towardeconomic efficiency) are shared equitably.
CONCLUSIONSThe assessment method presented within this paper presents a valuable tool to assessing thecontexts under which decentralised systems are most suitable. The ongoing application of theapproach will be important in driving the uptake of a broader range of system scales and achievingsustainable sanitation outcomes both within developed and developing nations.
REFERENCESBrunner, P.H. & Baccini, P. 1992, 'Regional material management and environmental protection',
Waste Management & Research, vol. 10, no. 2, pp. 203-212.Esrey, S.A., Andersson, I., Hillers, A. & Sawyer, R. 2000, Closing the loop: ecological sanitation
for food security, UNDP, SIDA, Water and Sanitation Programme, Thrasher Research Fundand PAHO.
Gumbo, B. 2005, 'Short-cutting the phosphorus cycle in urban ecosystems', PhD thesis, DelftUniversity of Technology.
Kazaglis, A. & Kraemer, P. 2006, 'Sanitation success stories in India and implications for urbansanitation planning', paper presented to the 32nd WEDC International Conference, Water,Engineering and Development Centre, Loughborough, UK, November 2006.
Mitchell, C., Fane, S., Willetts, J., Plant, R. & Kazaglis, A. 2007, Costing for SustainableOutcomes in Urban Water Systems - A Guidebook.
Newman, P. 2001, 'Sustainable urban water systems in rich and poor cities - steps towards a newapproach', Water Security for the 21st Century-Innovative Approaches. pp. 93-99. WaterScience & Technology [Water Sci. Technol.]. Vol. 43, vol. 43, p. 93.
Pinkham, R., Hurley, E., Watkins, K., Lovins, A., Magliaro, J., Etnier, C. & Nelson, V. 2004,Valuing Decentralized Wastewater Technologies: A Catalog of Benefits, Costs, andEconomic Analysis Techniques, Rocky Mountain Institute.
Rockström, J., Göran, N.A., Falkenmark, M., Lannerstad, M., Rosemarin, A., Caldwell, I.,Arvidson, A. & Nordström, M. 2005, Sustainable Pathways to Attain the MilleniumDevelopment Goals: Assessing the Key Role of Water, Energy and Sanitation, StockholmEnvironment Institute, Stockholm International Water Institute.
Tangsubkul, N., Moore, S. & Waite, T.D. 2005, 'Incorporating phosphorus managementconsiderations into wastewater management practice'.
White, K. 2004, 'Centralised and Decentralised Wastewater Management Alternatives: FunctionalModels and Infrastructure Costs', paper presented to the Enviro 06, Melbourne, 9-11 May2006.
Appendix E Field study photographs
The following photographs were taken during a field study of sanitation systems produced and
installed within India’s Southern regions. The photographs were taken principally at three sites,
depicted below.
Maps courtesy of Google™
Site B, C
Site A
Conference location
130
Site A: Centre for Scientific Research, Auroville
The Centre for Scientific Research in Auroville an NGO located nearby Pondicherry with particular
expertise in renewable energy systems, low energy construction, and ecological sanitation.
These photographs depict a facility used for the production of ferrocement urine diverting
squatting pans. Note the three holes for urine, faeces and washwater. Each pan contains two sets
of holes for two sets of chambers used interchangeably to ensure minimum storage times are
applied.
These pans are produced to scale at this facility and provided at production cost to communities.
The remaining components such as the compost pits are constructed in‐situ.
131
Site B: Friends of Camphill, Bangalore
The system installed is a biogas settler, followed by aerobic treatment in a baffle reactor for
subsequent reuse as irrigation water for the kitchen garden. The system and gardens are managed
by the residents of the school.
These women are preparing an afternoon meal of chapathi (i.e. a flat Indian bread) cooked with
biogas derived from the biogas settler system.
132
Site C: Rainwater Club, Bangalore
Rainwater Club is a public advocacy group for rainwater harvesting and ecological sanitation. The
office is situated in the home of Mr and Mrs Vishwanath, the founders of the organisation, who
have established their home as a demonstration of possibilities.
This squatting pan is produced using fibreglass by Eco‐solutions, a firm in Kerala in India’s South.
The toilets are typically operated as dessicating toilets, whereby a cup of ash is added after
defecation to reduce the moisture content of the faeces to below 20% while raising the pH, thus
destroying any pathogens.
These miniature wetland barrels are located on the roof of the dwelling and receive the greywater
from both the laundry and shower.
133
The figure on the left depicts a rice paddy, also situated on the roof of the dwelling, which is
irrigated with treated greywater. The paddy produces the majority of the needs for a family of
three. To the right is the kitchen water treatment system. Kitchen water in India is associated with
considerable loads of grease.
These Bangalore locals are tasked with collecting and source separating household refuse for
subsequent recycling. A considerable proportion of India's waste management is conducted
informally and results in extremely efficient recovery of waste however the benefit may be offset
by potential health impacts.
134
A disused lot in Bangalore transformed into an informal open landfill. Every now and then the pile is
ignited to keep the volume under control, which produces a constant haze over the city
Placed sporadically through the city, these bins formalise the practice of open, uncontrolled
incineration of waste.
135
Around the hill stations there is a growing unease with the volume of plastics accumulating in the
mountain streams. This sign is telling as the community appears to recognise that the ultimate
problem is the persistence of plastic waste, rather than inadequate sanitation practices.
Traditional Indian waste habits are to throw kitchen waste out the window where they would
biodegrade in place. With the introduction of persistent wastes such as plastic food packaging, the
cultural inertia of this practice has been difficult to shift.
