Citation
Evaluating and enhancing urban wastewater system sustainability

Material Information

Title:
Evaluating and enhancing urban wastewater system sustainability
Creator:
Pitterle, Mark Thomas
Publication Date:
Language:
English
Physical Description:
xvii, 191 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Sewage disposal plants -- Energy consumption -- Colorado -- Denver Metropolitan Area ( lcsh )
Greenhouse gases -- Colorado -- Denver Metropolitan Area ( lcsh )
Sustainable urban development -- Colorado -- Denver Metropolitan Area ( lcsh )
Greenhouse gases ( fast )
Sewage disposal plants -- Energy consumption ( fast )
Sustainable urban development ( fast )
Colorado -- Denver Metropolitan Area ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 178-189).
General Note:
Department of Civil Engineering
Statement of Responsibility:
by Mark Thomas Pitterle.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
437263071 ( OCLC )
ocn437263071
Classification:
LD1193.E53 2009d P57 ( lcc )

Full Text
EVALUATING AND ENHANCING URBAN WASTEWATER SYSTEM
SUSTAINABILITY
By
Mark Thomas Pitterle
B.S., Earth Science, Pennsylvania State University, 2000
M.S., Environmental Engineering, Virginia Polytechnic and State University, 2004
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy (Ph.D.)
College of Engineering and Applied Science
Civil Engineering
May 2009


by Mark Thomas Pitterle, 2009
All rights reserved


This thesis for the Doctor of Philosophy
degree by
Mark Thomas Pitterle
has been approved
by
Anu Ramaswami

Greg Cronin
Zoo\
Date


Pitterle, Mark T, (Ph.D., College of Engineering and Applied Sciences)
Evaluating and Enhancing Urban Wastewater System Sustainability
Thesis directed by Professor Anu Ramaswami
ABSTRACT
A hybrid life cycle assessment methodology is developed to measure life cycle
energy use and greenhouse gas (GHG) emissions from a Wastewater Treatment Plant
(WWTP). The method allows inclusion of operating energy (including
sludge/biosolids disposal), embodied energy particularly of WWTP infrastructure,
and inclusion of avoided impacts, such as avoided fertilizer use following land
application of biosolids. This methodology is the first known full system Life Cycle
Assessment (LCA) for US WWTPs examining the impact from the energy supply to
infrastructure materials sector to WWTP operations to end of life of biosolids. This
full system method allows comparison across WWTPs of different types, for
example, comparison of septic systems with no operating energy use with
conventional plants.
The method was applied to six Front Range, CO plants, namely Metro Wastewater
Reclamation District (MWRD), Longmont WWTP, Louisville WWTP, a mobile
park, single family septic with AdvanTEX, and a single family septic system. Results
show Annualized Life Cycle Primary Energy Intensities associated with end use of
energy, ALCPElEnd-use, ranging from 26.6 to 61.2 GJ/million gallon (MG) treated,
with the largest conventional WWTP (MWRD) being most efficient (26.6 GJ/MG
treated) except for the septic system (which only uses transport energy every 4 years
for pumped septic solids). Inclusion of embodied energy impacts increased the gross
total ALCPEI from 12.7% to 96.9%, with septic systems having the largest increase
(65.2% to 96.9%). Avoided energy impacts were significant for MWRD (30.6% of
gross total), followed by Longmont WWTP (10.8%) and Louisville WWTP (4.2%),


while the smaller wastewater (WW) systems did not realize any avoided energy
impacts.
The total annualized GHG emissions intensities for the entire WWTP ranged from
1.73 to 13.3 mt CO2 equivalents/MG treated (excluding avoided impacts), with the
largest WWTPs performing best. Inclusion of avoided impacts shifted total GHG
emissions intensities for the entire WWTP to 0.50 to 13.3 mt C02e/MG treated,
where the smallest WWTPs did not realize avoided impacts.
Consumption data for end-use of electricity and natural gas are consistent with
literature values normalized on a per MG treated basis (Literature: 5.0 to 15.0 GJ/MG
treated vs. This Study: 7.6 to 17.8 GJ/MG treated, excluding septic only systems
which consume no electricity and natural gas). The expanded system wide LCA
results support and enhance literature findings where larger WWTPs have lower
overall life cycle impacts on a per MG treated basis when compared to single family
systems; however, further studies are needed to draw statistically significant
conclusions across scale.
The model is most sensitive to consumption data pertaining to electricity, biogas
Combined Heat & Power (CHP), and infrastructure, and also to GHG emission
factors for electricity, natural gas, infrastructure, and nitrogen/methane
transformations, respectively. Biogas CHP, electricity energy efficiency, and avoided
water and fertilizer are promising pathways that yield the largest net reduction for a
WWTP to reduce its GHG footprint.
This abstract accurately represents the content of the candidates thesis. I recommend
its publication.
Signed
Anu Ramaswami


DEDICATION
This thesis is dedicated to my parents, whose love and guidance have taught me that
anything in this world is attainable, even that which is perceived impossible.
In particular, I dedicate this to my father, whose encouragement and wisdom is the
only reason why I have been able to see this through. He has been calling me Doc
for years, even before I even started my Ph.D. As if he knew in advance where my
pursuits would one day take me, he has reserved the exclusive right to call me Doc.
May his wisdom be carried on for generations to come.


ACKNOWLEDGEMENT
I recognize the contributions towards many of the insights contained within this thesis
to my advisor, Anu Ramaswami. I also wish to thank members of my committee for
their valuable input. Collectively, my committee has truly helped the final product to
be a defendable contribution towards the future of sustainability in the wastewater
industry. I wish to thank my industry mentor, Dale Gabel, whose input towards
industry perspectives was invaluable. Of great importance, I wish to thank
individuals from the wastewater treatment systems that provided data used in this
work without your help, this work would have been truly impossible.
I also wish to thank all of the GAANN students I worked with through the years and
cant wait to bring our work and dreams together in business after school. Also, to
Karen Kronoveter, whose efforts in the lab always brought back hope in the face of
adversity. Also to Helen Frey who single-handedly runs the entire College of
Engineering. And a special thanks to my life-long friend, Heidi Zellie, who has been
the light at the end of the tunnel with all of her unending support and advice along the
way. Lastly, I want to thank the dean, Dr. Renjeng Su, who has given critical advice
along the way and has smoothed over any potential speedbumps. Dr. Su has truly
been the catalyst to bring CU Denvers Engineering Department towards continual
national recognition.
This work has been supported by grants from US Department of Education: Graduate
Assistance in Areas of National Need (GAANN) and the Colorado Commission of
Higher Education (CCHE).


TABLE OF CONTENTS
Tables.........................................................................xii
Figures........................................................................xiv
Acronyms.......................................................................xvi
1. Urban Water Systems & Sustainability........................................1
1.1 The Challenge of Sustainable Urban Development..........................1
1.2 Urban Water, Wastewater & Sustainability................................2
1.3 Toward More Sustainable Wastewater Treatment Systems....................7
1.3.1 GHG Protocol for WWTPs................................................8
1.3.2 Integrating LCA into WWTP Sustainability Protocol...................8
1.4 Specific Project Objectives..............................................11
1.5 Major Contributions......................................................12
2. Overview of LCA and GHG Accounting..........................................13
2.1 Measuring Progress Towards Sustainability................................13
2.2 Metrics for Evaluating the Sustainability of Wastewater Systems........13
2.2.1 Overview of Sustainable Indicators for WWT to Date.................14
2.3 GHG Accounting Protocol for Businesses..................................15
2.3.1 WRI Implications for WWTPs..........................................15
2.4 Life Cycle Assessment...................................................17
2.4.1 Main Steps in LCA...................................................18
2.4.2 LCA Methods.........................................................20
2.4.2.1 Mass/Energy LCA..................................................20
2.4.2.2 Economic Input-Output LCA.......................................20
2.4.2.3 Hybrid LCA......................................................21
3. WWTP Hybrid LCA Method Development..........................................23
3.1 Sustainable Indicators...................................................24
3.1.1 Water Cycle Indicators...............................................27
3.1.1.1 Global Indicators................................................27
3.1.1.2 Regional Indicators.............................................27
3.2 Energy & GWP Calculation Methods Overview...............................31
3.2.1 Life Cycle Primary Energy vs. End-use Energy........................34
3.3 Calculating Life Cycle Primary Energy and Life Cycle GWP................37
3.3.1 Life Cycle Primary Energy...........................................37
3.3.2 Life Cycle Global Warming Potential.................................38
3.4 Annualizing and Normalizing Life Cycle Data.............................39
3.5 Energy and GHG Intensity Metrics........................................41
vm


3.6 Model Input Parameters...................................................42
3.7 Calculation of Intensity Parameters......................................43
3.7.1 End-use Energy........................................................44
3.7.1.1 End-use Consumption Parameters...................................44
3.7.1.2 Grid Specific Electricity Life Cycle Calculation Techniques.....45
3.7.1.3 PEI and EF for Natural Gas......................................50
3.7.1.4 Solids Handling..................................................50
3.7.1.5 Biogas Electricity and Heat Recovery Life Cycle Calculation
Techniques........................................................53
3.7.1.6 IPCC GHG Emissions...............................................55
3.7.1.6.1 Methane Emissions.......................................55
3.7.1.6.2 N2O Emissions...........................................56
3.7.2 On- & Off-site Embodied Primary Energy..............................60
3.7.2.1 PEI and EF for Chemicals........................................61
3.7.2.2 On- & Off-site Infrastructure and Maintenance EIO LCA.........62
3.7.2.3 Off-site Infrastructure Mass LCA..............................64
3.7.3 Avoided Energy.......................................................66
3.7.3.1 Avoided Biogas Electricity and Heat Recovery.....................67
3.7.3.2 Avoided Fertilizer...............................................68
3.7.3.3 Avoided Water....................................................73
3.7.3.3.1 Direct Water Reuse......................................74
3.7.3.3.2 Treated Water Reuse.....................................76
3.7.4 BOD GHG Emissions....................................................81
3.7.5 Summary of Intensity Parameters and Benchmarks......................82
4. Energy and GHG Impacts of a WWTP: MWRD as a Case Study......................85
4.1 MWRD Consumption Parameters..............................................85
4.2 On- & Off-site End-use Energy...........................................87
4.2.1 On- & Off-site Electricity............................................87
4.2.2 On- & Off-site Natural Gas...........................................88
4.2.2.1 Off-site Combined Electricity and Natural Gas Use...............89
4.2.3 Solids Handling......................................................90
4.2.4 Biogas Electricity and Heat Recovery.................................90
4.2.5 IPCC Greenhouse Gas Emissions........................................91
4.3 On- & Off-site Embodied Energy...........................................93
4.3.1 Chemicals.............................................................93
4.3.1.1 Chemical Manufacturing Embodied Energy...........................93
4.3.1.2 Chemical Transportation Embodied Energy..........................93
4.3.1.3 Embodied Chemical ALCPE and ALCGWP..............................94
4.3.2 Infrastructure Embodied Energy.......................................95
IX


5.6.2.3 Biogas Electricity Consumption Benchmarking..................162
5.6.3 Natural Gas Use Benchmarking....................................163
5.6.4 Combined Electricity and Natural Gas Use Benchmarking............163
5.6.5 Solids Handling Benchmarking....................................164
5.6.6 Avoided Fertilizer..............................................164
5.7 Summary of Energy Consumption Data for Six CO WWTPs..................165
6. GHG Mitigation Strategies for WWTPs.....................................167
6.1 Renewable Energy....................................................170
6.2 Energy Efficiency...................................................170
6.3 Water Efficiency....................................................172
6.4 Chemical vs. UV Disinfection........................................173
6.5 Capturing Effluent Potential Energy via Small-scale Hydroelectricity.174
6.6 Strategies for WWTPs to Reduce N2O Emissions........................175
6.6.1 Suggestions for Future N2O EF Development.......................175
6.7 MWRD Maximum GHG Mitigation Implications............................176
Bibliography...............................................................178
Appendix A.................................................................190
Appendix B.................................................................191
xi


5.6.2.3 Biogas Electricity Consumption Benchmarking..................162
5.6.3 Natural Gas Use Benchmarking.....................................163
5.6.4 Combined Electricity and Natural Gas Use Benchmarking........163
5.6.5 Solids Handling Benchmarking.....................................164
5.6.6 Avoided Fertilizer...............................................164
5.7 Summary of Energy Consumption Data for Six CO WWTPs..............165
6. GHG Mitigation Strategies for WWTPs......................................167
6.1 Renewable Energy.....................................................170
6.2 Energy Efficiency....................................................170
6.3 Water Efficiency.....................................................172
6.4 Chemical vs. UV Disinfection.........................................173
6.5 Capturing Effluent Potential Energy via Small-scale Hydroelectricity.174
6.6 Strategies for WWTPs to Reduce N2O Emissions.........................175
6.6.1 Suggestions for Future N2O EF Development........................175
6.7 MWRD Maximum GHG Mitigation Implications.............................176
BIBLIOGRAPHY................................................................178
Appendix A..................................................................190
Appendix B..................................................................191
xi


List of Tables
Tables
1: Review of LCA Studies of WWTPs................................................10
2: Quantitative Metrics for US WWTP Sustainability...............................25
3: US Population with Access to Improved Drinking Water and Sanitation...........27
4: Proposed Sustainable Water Indicators.........................................30
5: WRI Scope Requirements and LCA Boundaries.....................................32
6: End-use, Primary, and Life Cycle Primary Energy Comparison....................35
7: Renewable Electricity Primary Energy Quantification...........................36
8: Energy and GHG Intensity Metrics..............................................42
9: WW LCA Consumption Input Parameters...........................................43
10: Major Variables for Calculating PElEiectricity................................48
11: PElEiectricity Benchmarked for US and CO Averages.............................49
12: Major Variables for Calculating Emission Factors for Electricity..............49
13: EFfiiectricity Benchmarked to Literature......................................50
14: PEI and EF for Common Transport Modes.........................................52
15: UK Single Truck PEI and EF by Gross Vehicle Weight............................53
16: IPCC N2O Process EF Uncertainty Range vs. Literature & MWRD...................59
17: PEI and GWP for Common WWT Chemicals..........................................62
18: EIO-LCA Infrastructure PEIs and EFs...........................................63
19: Linear Mass for Given Pipe Diameter and Material..............................65
20: PEIs and EFs for Various Pipe Types...........................................66
21: Weighted Average of Fertilizer Use for PEI & EF Calculations..................71
22: Avoided Fertilizer PEIs and EFs...............................................71
23: Water Supply Energy Use by Type (Arpke and Hutzler, 2006).....................76
24: DWRP Streamlined ALCPE and ALGWP Summary......................................78
25: EUEIs, PEIs, and EFs for Regional WTPs and WRPs...............................80
26: Summary of Primary Energy Intensity Parameters and Benchmarks.................83
27: Summary of Emission Factor Intensity Parameters and Benchmarks................84
28: MWRD Data Collection Categories and Sources...................................86
29: Qualitative Data Quality Criteria.............................................87
30: MWRD ALCPE for On- & Off-site Electricity 2002-2007.........................88
31: MWRD ALCGWP for On- & Off-site Electricity 2002-2007........................88
32: MWRD ALCPE for On- & Off-Site Natural Gas Use.................................89
33: MWRD ALCGWP for On- & Off-site Natural Gas Use................................89
34: MWRD ALCPE and ALCGWP for Solids Handling.....................................90
xii


35: MWRD ALCPE for Biogas Electricity and Heat Recovery.......................91
36: IPCC Nitrogen Effluent GHG Emissions: Uncertainty Implications............92
37: Denitrification Removal and IPCC ALGWP....................................92
38: MWRD 2006 WW ALCPE and ALCGWP for Chemical Manufacturing..................93
39: MWRD 2006 Chemical Transport LCPE and LCGWP...............................94
40: MWRD Chemical ALCPE and ALGWP.............................................94
41: Longmont WWTP Summary of Methods for Annualizing Expenditures.............104
42: MWRD Summary of Methods for Annualizing Expenditures......................109
43: Truncation Error for Various Life Expectancies............................110
44: MWRD On-site Infrastructure ALCPE and ALCGWP (1988-2007).................. 112
45: MWRD Combined On- & Off-site Infrastructure Expenditure Results...........113
46: MWRD Annualized Off-site Infrastructure ALCPE and ALCGWP..................114
47: Annualized Off-site Infrastructure (1988-2007): EIO vs. Mass LCA...........115
48: Avoided ALCPE & ALCGWP for Biogas Electricity & Heat Recovery.............116
49: MWRD 2006 Avoided Fertilizer ALCPE and ALCGWP..............................117
50: MWRD Avoided Fertilizer ALCPE and ALGWP...................................117
51: MWRD Annual Water Reuse...................................................118
52: MWRDs BOD-related CO2 Emissions..........................................118
53: MWRD 2006 ALCPEI and ALCGWPI Summary......................................120
54: MWRD 2006 Annualized Intensity Overview...................................123
55: Scale Variations of Six Participating Front Range WWT Systems..............132
56: Life Cycle Sustainable Indicators Comparing Six WWT Systems................134
57: Biogas Electricity and Heat Recovery Potentials...........................162
58: CO WWTP Consumption Benchmarking..........................................166
59: Potential GHG Benefits for Selected GHG Mitigation Strategies.............169
60: Estimated GHG Benefits for Unexplored GHG Mitigation Strategies...........169
61: LCPE and LCGWP for Chemical vs. UV Disinfection...........................174
62: MWRD GHG Mitigation Potential.............................................177
Xlll


List of Figures
Figures
1: Four Evaluatives of Sustainable Infrastructure Development...................2
2: Typical Household Water Usage................................................4
3: Life Cycle Inventory Stages of LCA (EPA, 2006)............................. 18
4: Annual Renewable Water Supply per Capita....................................28
5: Contemporary Population Relative to Relative Water Demand = 0.4.............29
6: Hybrid LCA WWT system boundaries and components.............................33
7: US Electricity Grid Mix (NREL, 2007: Year 2000 data)........................46
8: Illustration of CHP Based on WWTP Combined Heat & Power Energy Flows:
Assuming r|pp.bi0gas= 0.25 (EPAa, 2007).................................54
9: Nitrogen EF Literature Review.............................................69
10: Phosphorus EF Literature Review 56.........................................70
11: Potassium EF Literature Review56...........................................70
12: Denver Water Recycling Plant ALCGWPI Summary...............................78
13: MWRD Off-site Energy (Electricity + Natural Gas) Use, GJ...................90
14: Longmont WWTP Annual Capital Expenditures in 1997$ from 1961 to Present 97
15: Longmont WWTP Annualized Capital Expenditures..............................99
16: Longmont WWTP Annualized Capital Expenditures Clean Water Act
Close-up................................................................99
17: Longmont WWTP Single Year Annualized Capital Expenditures Truncation
Error..................................................................100
18: Longmont WWTP 10-yr Moving Average of Annualized Expenditures............101
19: Longmont WWTP 20-yr Moving Average of Annualized Expenditures............102
20: Longmont WWTP 30-yr Moving Average of Annualized Expenditures............102
21: Longmont WWTP 40-yr Moving Average of Annualized Expenditures............102
22: Longmont WWTP Average Annualized Capital Expenditures Truncation
Error..................................................................103
23: MWRD Annual Capital Expenditures in 1997$ from 1964-2007................. 105
24: MWRD On- vs. Off-site Infrastructure Expenditures, 1988-2007............. 105
25: MWRD Annualized Capital Expenditures......................................106
26: MWRD Total Capital Expenditures Clean Water Act Close-up................106
27: MWRD Single-year Annualized Capital Expenditure Truncation Error.........107
28: MWRD 20-yr Moving Average of Annualized Expenditures......................108
29: MWRD 30-yr Moving Average of Annualized Expenditures......................108
30: MWRD 40-yr Moving Average of Annualized Expenditures......................108
xiv


31: MWRD Average Annualized Capital Expenditure Truncation Error.............109
32: Relative Contribution of MWRD Infrastructure Sectors on Embodied
Primary Energy and GWP (1988-2007).................................... 112
33: MWRD 2006 ALCPE Summary, GJ/AMG..........................................121
34: MWRD 2006 ALCGWP Summary, mt C02e/AMG....................................121
35: MWRD 2006 ALCPE with Annual Variability..................................123
36: MWRD 2006 ALCGWP with Annual Variability.................................124
37: MWRD Consumption Sensitivity: 10% Increase in Consumption................126
38: MWRD Emission Factor Sensitivity: 10% Increase in EFs....................128
39: Governing Uncertainty in MWRDs GHG Footprint............................129
40: BOD Removal Efficiency: BOD Removed per Acre Treatment System............138
41: Annualized Infrastructure Costs/AMG......................................139
42: Annualized Life Cycle Costs/AMG..........................................140
43: Water Efficiency: Volumetric Wastewater Generation per Capita............142
44: End-use Energy Intensity (EUEI) & Energy Efficiency......................146
45: LCPEIwwtp-2006 for Six WWT Systems.......................................146
46: ALCPEI for On-site Infrastructure........................................148
47: ALCPEI for On- & Off-site Infrastructure.................................149
48: ALCGWPIwwtp-2006 for Six WWT Systems.....................................152
49: Solids Transport Consumption for Six WWT Systems.........................153
50: Combined IPCC GWP for Six WWTPs..........................................154
51: Chemical Consumption Inputs for Six WWT Systems..........................155
52: Annualized Avoided Fertilizer GWP........................................156
53: WWTP Electricity Use Benchmarks by Type..................................160
54: NY and Cal. WWTP Survey Electricity Use, per gallon basis................161
55: NY and Cal. WWTP Survey Electricity Use, per mt BODr basis...............161
56: Electricity and Natural Gas Use: NY WWTPs vs Current Study...............164
57: GHG Mitigation Strategies for WWTPs......................................168
58: Typical Activated Sludge Energy Use by Major Process (PG&E, 2007)........ 171
xv


List of Acronyms
Acronym_____________Description
ALCGWP Annualized Life Cycle Global Warming Potential
ALCPE Annualized Life Cycle Primary Energy
AWWA American Water Works Association
BOD Biological Oxygen Demand
BODr Biological Oxygen Demand removed
CAS Conventional Activated Sludge
CCHE Colorado Commission of Higher
CCX Chicago Climate Exchange
CEC California Energy Commission
CHP Combined Heat & Power
COD Chemical Oxygen Demand
CPI Consumer Price Index
DAP Diammonium Phosphate
DO Dissolved Oxygen
DWRP Denver Water Recycling Plant
EE Energy Efficiency
EF Emission Factor
EIA Energy Information Administration
EIO Economic Input-Output
EMS Energy Management System
EPA US Environmental Protection Agency
EPACT Energy Policy Act
EU European Union
ECX European Climate Exchange
EUE End-use Energy
EUEI End-use Energy Intensity
GANN Graduate Assistance in Areas of National Need
GASB Government Accounting Standards Board
GHG Greenhouse Gas
GJe GJ of electricity
GJth GJ of thermal energy
GVW Gross Vehicle Weight
GWP Global Warming Potential
HVAC Heating, Ventilation, and Cooling
IPCC Intergovernmental Panel on Climate Control
ISO International Standardization Organization
K Potassium
kWeh kWh of electricity
LCA Life Cycle Assessment
LCC Life Cycle Cost
xvi


List of Acronyms cont,
Acronym_____________Description
LCGWP Life Cycle Global Warming Potential
LCGWPI Life Cycle Global Wanning Potential Intensity
LCI Life Cycle Inventory
LCPE Life Cycle Primary Energy
LCPEI Life Cycle Primary Energy Intensity
MAP Monoammonium Phosphate
MG Million Gallons
MGD Million Gallons per Day
MWRD Metro Wastewater Reclamation District
N Nitrogen
NAICS North American Industry Classification System
NFRV Nutrient Fertilizer Replacement Value
NREL National Renewable Energy Laboratory
NYSERDA New York State Energy Research and Development Authority
O&M Operation & Maintenance
P Phosphorus
PEI Primary Energy Intensity
PG&E Pacific Gas & Electric Company
PV Photovoltaics
RECs Renewable Energy Certificates
REP Request for Proposals
SI Sustainable Indicator
SIC Standard Industrial Classification
SRT Solids Retention Time
TEAM Tool for Environmental Analysis and Management
TN Total Nitrogen
T^biosolids Annual Total Nitrogen in the biosolids
TNeff Annual Total Nitrogen in the effluent
TNjnf Annual Total Nitrogen in the influent
TSS Total Suspended Solids
VFDs Variable Frequency Drives
WERF Water Environment Research Foundation
WRI World Resource Institute
WTP Water Treatment Plant
WRP Water Reuse Plant
ww Wastewater
WWT Wastewater Treatment
WWTP Wastewater Treatment Plant
XVII


1. Urban Water Systems & Sustainability
1.1 The Challenge of Sustainable Urban Development
Strained by population growth, urban areas are becoming increasingly overburdened
by rising resource demand, often falling short of providing basic urban services.
Global population could reach nearly 11 billion by 2050 (UN, 2001). Based upon UN
statistics, more than 50% of the total world population, and about 75% of the
population of the developed world, currently reside in cities. These urban populations
are growing at an unprecedented rate over the past 2 decades, with most rapid growth
occurring in developing nations at an annual rate in excess of 10% forcing city
planners to have to plan for a doubling of the entire citys infrastructure every 7 years
- a daunting task. In the US, alarming urban growth rates, such as 191% growth in
Douglas County in the 1990s and a projected population increase of 65% in the next
thirty years for Colorados urban areas, are placing increasing strain on the already
low water availability of the American west (OEDIT, 2003; US Census Bureau,
2000). As population continues to increase, cities are increasingly strained to divvy
up a shrinking supply of finite resources amongst a larger population. Thus, both
nationally and internationally, urban sustainability is essential for global sustainable
development.
Sustainable development, broadly defined as Development which meets the needs of
the present without compromising the ability of future generations to meet their own
needs (Brundtland, 1987), addresses performance, economic, environmental and
social aspects of human development (Figure 1). For urban systems, sustainable
infrastructure is an important component of urban sustainability. Urban infrastructure
1


refers to the engineered systems that provide services pertaining to water, wastewater,
energy and transport (of human goods and information) within an urban area
(Ramaswami, 2005). Sustainable urban infrastructure systems are therefore those that
provide these services in a manner that meets the long-term needs of the urban
residents while also being protective of the environment and ecosystems. The focus
of this thesis is on evaluating and enhancing the sustainability of urban wastewater
systems in Colorado.
Figure 1: Four Evaluatives of Sustainable Infrastructure Development
1.2 Urban Water, Wastewater & Sustainability
Urban wastewater systems are closely related to the following aspects of
sustainability:
Water
Energy
Nutrients
Infrastructure
2


Each of these aspects is described below with focus on current Wastewater Treatment
(WWT) practices.
Water Use and Efficiency: According to the UN, more than half of the world's
population will reach a critical water shortage by the year 2025 (Gleick, 1996).
Strains on the worlds water resources are only feared to be getting worse, as more
than 2 billion people in India and China adopt western styles of living, transitioning
from their current public water demand of 22 and 88 gal/capita-day, respectively, to
the near 180 gal/capita-day US average (UNSD, 2005; USGS, 2005). Water demand
in the semi-arid southwestern US is continuously increasing and in many cases well-
above the US average public supply usage (180 gal/capita-day), exemplified by
Colorados 240 gal/capita-day public water demand (USGS, 2005).
Water use patterns and wastewater collection/treatment systems presently in use in
the US are quite inefficient. Current practice treats nearly all water to potable levels
when, in fact, the majority of water usage in the US is for non-potable needs. For
example, non-potable uses for agriculture and industry account for 84% of U.S. water
usage (USGS, 2005). Meanwhile, only 30% of municipal water use is for potable
needs.1 This totals to over 96% of water usage in the US is for non-potable needs.
Furthermore, potable water often only functions as a transport medium, where a
concentrated waste source (e.g. human excreta) is diluted in potable water,
transported large distances, only to be treated to near potable water standards prior to
discharge. Significant cost, energy, resources, and environmental impact are
associated with these inefficient system design considerations. Figure 2 shows that
typical water use in a US household is composed of about 70% greywater (household
1 Value includes water consumed for faucets, other domestic, and leaks (Steinfeld and Del Porto,
2004). Actual potable requirement is likely to be significantly less.
3


shower/sink water) and 30% blackwater (household toilet water) (Steinfeld and Del
Porto, 2004). The commingling of blackwater and greywater considerably
complicates treatment doubling the volume of wastewater and more than doubling
the energy inputs required for treatment (EPA, 2002).
!B Toilets
Clothes Washer
Showers
Faucets
Leaks
B Other Domestic
Baths
Dish Washers
[e]nergy represents the largest
controllable cost of providing water or wastewater services to the public, and thus
demands our primary attention (CEC, 2006). In California alone, over $500 million
annually are spent for electricity used to treat water and wastewater (CEC, 2006). An
already strained global energy market is expected to witness a consumption increase
of 60% between 1997 and 2020, principally via fossil fuels (IEA, 2000). The
dependence of the water industry on the energy industry, and vice versa, show us that
the only sustainable solutions will require viewing the two resources as
interconnected sub-systems in the larger global system emphasizing that the lack of
one can lead to a lack of all.
Other
Domestic
2%
Baths
2%
i
Faucets
15%
Showers
18%
Dish Washers
1%
Toilets
26%
Clothes
Washer
23%
Figure 2: Typical Household Water Usage
Source: Steinfeld and Del Porto, 2000
Wastewater Treatment & Energy: Currently,
4


Rather than requiring extensive amounts of energy to treat the waste, Wastewater
Treatment Plants (WWTPs) in the US have the potential to be net-energy producing,
which has already been demonstrated by several plants in the US that produce twice
as much energy as they consume (AWWARF, 2007). Further, the centralized
anaerobic digesters of Sweden currently supply 45% of the countrys natural gas
needs for transportation. Theoretically, biogas could supply 20% of all vehicle fuel
needs for nearly all nations (Boesel, 2005; Persson, 2006). Such an effort in the US
would greatly reduce our reliance on foreign oil, while also greatly reducing our
nations greenhouse gas (GHG) emissions. In fact, a recent US Environmental
Protection Agency (EPA) study showed that if existing US WWTPs with anaerobic
digestion (and > 5 MGD2 influent flow) were to use Combined Heat & Power (CHP:
e.g., on-site cogeneration), an equivalent of 2.3 million mt CC^e would be offset.
This is equivalent to the emissions of approximately 430,000 cars (EPAa, 2007).
However, currently, only 20 of Colorados 4950 WWTPs use anaerobic digesters
(Personal communication: Jones, 2005) and very little of the biogas energy is
recovered. Only two WWTPs in CO use biogas CHP (EPAa, 2007). Others recover
some of the energy for heating, but often flare the remainder. For example,
Longmont WWTP uses some of the gas for heating the digesters, while flaring the
remaining 70% of total biogas energy content (Current work in progress). Nine of the
twenty WWTPs meet the EPA criteria of >5 MGD and if CHP was utilized, an
estimated 12,700 cars could be offset just by these nine CO WWTPs alone
(Calculated based on EPAa, 2007). The treatment process alternative to anaerobic
digestion is typically aerobic digestion, which requires additional aeration process
energy, produces 1/3 more sludge, and has no energy co-product production (Grady
et al. 1999). Clearly, the implications across the state could be significant if a larger
2 MGD = Million Gallons per Day
5


number of plants utilized the energy producing process alternative: anaerobic
digestion with CHP.
Nutrient Recycling: WWTPs also have the potential to reduce our GHG footprint
through natural nutrient cycling in the environment. In 1998, only 60% of the
biosolids produced in the US were beneficially reused (EPA, 1999). Benefits from
avoided fertilizer use were not seen for nearly 2.8 million dry tons of material. Life
cycle studies have shown that the recycling on nutrients can be an energy favorable
process, while simultaneously avoiding the creation of air emissions (e.g.,
incineration) or the unnecessary use of landfill space (Lundin et al., 2000; Wilsenach
and van Loosdrecht, 2003).
Wastewater Infrastructure: Current WW infrastructure is also energy intensive.
Common constant gravity sewers are large, often 4 to 6 ft in diameter, and designed
to never be more than Vi full. The large pipe size combined with the need to maintain
constant gravity (3 fps velocity) creates considerable environmental impact during
excavation and installation. For example, a centralized sewer installation in the Toms
Creek Basin area of the town of Blacksburg, Va. would have to cross Toms Creek
over 30 times in a 12.5 mile section, causing considerable damage to stream wildlife
and ecology (TCB Sewer Project, 2004). Also, gravity sewers must utilize energy-
intensive lifting stations when topographical barriers prevent constant gravity flow.
Further, most operations infrastructure is outdated, and electricity efficiency
improvements exceeding 50% exist, such as converting to fine bubble aeration
(AWWARF, 2007; Malcolm Pimie, 2006). And as a nation, we are starting to realize
the economic burdens of such a material and energy intensive system design an
estimated 1 trillion dollars will be spent in the next 20 years to replace existing,
failing water and wastewater infrastructure (EPA, 1997).
6


In summary, as can be seen from the above discussion, while WWTPs offer much
benefit to public health (EPA, 1997), there are many opportunities to improve upon
inefficiencies in current WWT practice. Above all, energy expenditures and water
conservation and reuse are most critical. WWTPs must adopt methods to carefully
reconsider energy expenditure, and to consider whether WWT can be a positive
energy producer. Furthermore, savings of nutrient recovery, possible with biosolids
land application, must be realized to minimize energy intensive fertilizer
production/application to agricultural lands.
To date, no study has quantified full system WWTP environmental impacts from
Energy -> Infrastructure -> WWTP -> Avoided Fertilizer.
1.3 Toward More Sustainable Wastewater Treatment Systems
The development of life cycle-based, business-integrated quantitative sustainability
metrics for WWTPs is needed to compare WWTPs of different types and scale.
The Rocky Mountain Institutes Valuing Decentralized Wastewater Technologies,
as well as Ocean Arks Methods for Comparison of Wastewater Treatment Options
have provided tremendous qualitative comparisons across scale (e.g., 100+ MGD to
100 gal/day plants) (RMI, 2004; Kirk et al., 2005). To critically evaluate progress
towards more sustainable WWT, qualitative approaches must be converted to
quantitative metrics that are updated annually to measure progress towards
sustainability. No quantitative sustainability metrics have been developed for US
WWTPs. However, GHG emissions from water and wastewater utilities have
received much attention in California and the US (CEC, 2006; EPAb, 2007).
7


1.3.1 GHG Protocol for WWTPs
The increasing threat of global warming, along with on the horizon adoption of US
GHG accounting for US businesses, necessitates development of a protocol for
WWTP GHG accounting. The World Resource Institute (WRI) has developed a
protocol for business GHG accounting, but this protocol has not been yet applied to
WWTPs (WRI, 2004). No established methodology exists for US WWTP GHG
accounting that, in accordance with WRI protocol, integrates life cycle impacts of a
WWTP from cradle to grave (e.g., Energy Infrastructure WWTP Avoided
Fertilizer).
1.3.2 Integrating LCA into WWTP Sustainability Protocol
Most wastewater Life Cycle Assessment (LCA) studies (Table 1) have focused on
unit operations (e.g., sludge handling, etc.) and all lack components of a full WWTP
LCA (e.g., Energy -> Infrastructure -> WWTP -> Avoided Fertilizer).
A review of existing LCA wastewater treatment system studies (Table 1) shows that
no one has performed a comprehensive LCA that examines all three categories (e.g.,
end-use energy, embodied energy, and avoided energy) for an existing real-world
treatment system. Many studies have focused on specific processes (e.g., solids
handling, activated sludge configurations, etc.), while excluding key life cycle
consumption use within these specific processes (Palme et al., 2005; Murray et ah,
2008; Houillon and Jolliet, 2005; Munoz et al.., 2007; Vidal et ah, 2002; Jeschar et
ah, 1995). Meanwhile, other studies have looked at the entire WWTP system, but all
lack a complete WWTP system (e.g., Energy -> Infrastructure -> WWTP -> Avoided
Fertilizer) (See WWTP Systems in Table 1). For example, many of these studies do
not address infrastructure, collection system, chemicals, and/or avoided impacts.
Some of these studies calculate only GHG impacts (Friedrich et ah, 2007; Hospido et
ah, 2008; Ortiz et ah, 2007; Machado et ah, 2007; Lassaux et ah, 2007), while others
8


also include energy use impacts (Zhang and Wilson, 2000; Dixon et al., 2003; Remy
and Ruhland, 2006). Remy and Ruhland (2006) is the only known existing study that
includes aspects of all three categories in the system boundaries, but the analysis was
not only theoretical, it also did not report results in terms of end-use, embodied, and
avoided energy and GHG emissions. Further, these studies (Table 1) have poor
transferability due to their lack of identification of major assumptions or key emission
factors (i.e., no identification of major emission factors, Is upstream fuel processing
included?, etc.), while also using scientifically disputed end-point LCA analyses, such
as Eco-Indicators.
9


Table 1: Review of LCA Studies of WWTPs3
Reference Limitations GWP
Sludge Disposal Only
Palme et al., 2005 No sludge treatment, infrastructure C02 only
Murray et al., 2008 (EIO-LCA3 4) No disposal, infrastructure GWP
Houillon and Jolliet, 2005 No sludge treatment, infrastructure GWP
Process Specific
Micro-pollutant removal Munoz et al., 2007 No indirect energy GWP
Activated sludge configs. No infrastructure, off-site energy, GWP
Vidal et al., 2002 chemicals, and avoided energy
Pipe Infrastructure
Jeschar et al. (1995) Pipe materials only GWP
WWTP systems
GHG Emissions Only (No Energy Use Results) + Eco-indicators/other indicators
Friedrich et al., 2007 GWP
Hospido et al., 2008 Theoretical only No infrastructure, avoided energy GWP
Ortiz et al., 2007 No collection system, chemicals C02 only
Machado et al., 2007 No collection system, chemicals C02 only
Lassaux et al., 2007 Limited chemicals, No avoided water GWP
GWP + Primary Energy
Zhang and Wilson, 2000 No indirect energy, collection system, avoided fertilizer, avoided water None
Dixon et al., 2003 Theoretical only No collection system, avoided fertilizer, avoided water C02 only
Remy and Ruhland, 2006 Theoretical only No avoided water GWP
A full system US WWTP LCA is needed to establish WWTP primary energy use and
GHG emission baselines and to evaluate GHG mitigation alternatives.
3 GWP = Global Wanning Potential. If GWP is assessed, it is noted in the GWP column.
4 EIO = Economic Input-Output
10


1.4 Specific Project Objectives
This work will outline in detail the methodology used to perform a life cycle
assessment (LCA) and Life Cycle Costing (LCC) of a US WWTP, highlighting
primary energy use and greenhouse gas emissions, and specific policy options to
reduce life cycle impacts. Intensity factors (e.g., primary energy, GHG) used to
calculate life cycle impacts will be benchmarked with literature. Results from six
Front Range Colorado WWT systems will be reported, as well as effects of scale,
infrastructure impacts, and avoided fertilizer use implications. Results will be
compared using developed quantitative sustainability metrics that span across all life
cycle phases of a WWTP. GHG mitigation options will be quantified, thus providing
the pathway for WWTPs to reduce their GHG footprint and potentially enabling the
WWTP to be a net negative GHG emitter.
The specific dissertation objectives are:
I. Method Development
a. Develop methodology for full Life Cycle Assessment (LCA) of a WWTP
from Energy -> Materials & Infrastructure WWTP -> Ag/Land Use
b. Integrate the LCA method with business practices of GHG accounting (WRI
Protocol)
c. Benchmarking LCA results with past literature
II. Analysis
a. Case study of six WWTPs
i. Examining impact of scale
III. Policy
a. Pathways for GHG mitigation
11


1.5 Major Contributions
To the best of my knowledge, the major and original contributions of this work are:
> Development of US WWTP LCA methodology for the full system WWTP
LCA (e.g., Energy -> Infrastructure -> WWTP -> Avoided Fertilizer) with
real-world data
> Development of a GHG accounting protocol in line with current WRI
business protocol that incorporates LCA data
> First listing of Sustainable Indicators (Sis) and use of end-use, embodied, and
avoided energy categories for US WWTP LCA
> First US Quantification of WWTP metrics such as:
End-use Energy Intensity (EUEI)
Life Cycle Primary Energy Intensity (LCPEI)
Life Cycle Global Warming Potential Intensity (LCGWPI)
GHG footprint
> First established baseline of full system WWTP primary energy and GHG
LCA impacts, comparing economic and environmental life cycle WWTP Sis
across scale
In addition, this work can be used as a template for future studies, since existing
studies lack transparency. There is little to no benchmarking of emission factors or
primary energy intensities in literature. Improved transparency is essential for
standardization of LCA of WWTPs.
12


2. Overview of LCA and GHG Accounting
2.1 Measuring Progress Towards Sustainability
In 1987, the term sustainable development was introduced by Gro Brundtland as
Development which meets the needs of the present without compromising the ability
of future generations to meet their own needs (Brundtland, 1987). While this
definition captures the essence of sustainable development, the vagueness of the term
leaves progress toward sustainable development difficult to measure. Consequently,
metrics, termed Sustainable Indicators (Sis) are applied along with Life Cycle
Assessment (LCA) thinking, to measure progress.
Without a systematic and quantitative method to gauge and steer impacts, the
long-term sustainability of basic modem day necessities: food, clothing, water,
shelter, energy, and sanitation infrastructure systems are quickly put into jeopardy.
Developing a scientifically-based framework that measures progress of dominant
parameters towards system sustainability is the first critical step in making progress
toward sustainable development goals.
2.2 Metrics for Evaluating the Sustainability of Wastewater Systems
Common among all Sis is an effort to define metrics that capture the main aspects of
sustainable development. The Sustainable Seattle project showed that these Sis must
meet four important criteria: 1) Must be fundamental to long-term economic, social,
and environmental health, 2) Can be understood and accepted by the community, 3)
Must have interest and appeal for use by local media, and 4) Must be statistically
measurable (Sustainable Seattle, 1998). However, Sustainable Seattle did not
specifically quantify the sustainability of water systems.
13


Evaluating the sustainability of WWTPs from a life cycle perspective is becoming
increasingly more valued. The Guiding Principles of the American Water Works
Association (AWWA) includes a Commitment to Sustainability that focuses on
assuring that water is managed for the greatest good of people and the environment
and that all segments of society have a voice in the process (AWWA, 2005). The
Water Environment Research Foundation (WERF) has recently disseminated a
Request for Proposals (RFP) for calculation of the GHG life cycle impacts of
WWTPs and evaluating the implications of infrastructure on WWT (WERF, 2008).
Sis that quantify these different dimensions of sustainability are needed and must be
demonstrated and applied to specific US regions and cities. Sis can be used to
characterize the current state of the wastewater system in a city, and to evaluate
alternative pathways and technologies of the future.
2.2.1 Overview of Sustainable Indicators for WWT to Date
In Europe, an extensive list of Sustainable Indicators (Sis) has been developed for the
domestic wastewater industry by Balkema et al. (1998) and Lundin (1999). These Sis
evaluate the wastewater industry on functional, economical, environmental, social,
and cultural levels (Figure 1). Although extensive, it is not clear how these indicators
may be successfully adopted in other parts of the world, particularly the US. In
addition, many of these Sis are difficult to quantify (e.g., cultural acceptance, etc.),
while simultaneously failing to tie all impacts into an itemized, common unit of
measurement (e.g., primary energy or GHG emissions for electricity, natural gas, etc.
consumption). Currently, WWTPs typically do not meter specific unit processes, but
as metering and data collection improve, WWTPs are going to desire sustainability
comparison between infrastructure choices (e.g., chemical disinfection vs. UV;
aerobic digestion vs. anaerobic digestion, etc.). Sis developed to date lack flexibility
14


to be further compartmentalized to include process specific comparisons (e.g.,
primary sedimentation, aeration basin, anaerobic digestion, etc.).
Developing process specific comparisons of WWTP life cycle impacts across scale
will be essential as WW businesses are required to perform GHG inventories.
2.3 GHG Accounting Protocol for Businesses
The World Resource Institute has created an internationally accepted GHG
accounting protocol for businesses (WRI, 2004). WRIs policy neutral GHG protocol
aims to provide a simple, accurate, cost-effective, annually updated protocol for
businesses to monitor and reduce GHG emissions through time. However, this
protocol has not yet been applied to WWTPs in the US.
2.3.1 WRI Implications for WWTPs
Three scopes are included in WRI GHG Protocol. Transitioning from Scope I (e.g.,
basic, on-site emissions) towards Scope III (e.g., upstream fuel and material
processing emissions can be included) represents more holistic, life cycle-based
emissions inclusion. WRIs Scope I includes direct end-use energy (e.g., electricity,
natural gas, and transport fuels), while Scope II adds indirect electricity emissions.
Scope III is optional, but can add as many LCA-based indirect emissions as possible
including waste treatment (e.g., landfills, etc.), fuel processing, electricity
transmission line loss, materials energy, and avoided energy. According to the WRI
GHG Protocol, the designation of organizational boundaries can have significant
impact on whether emissions are categorized as Scope I or Scope III (WRI, 2006).5
5 In more detail, the WRI GHG Protocol uses two distinct approaches equity share (e.g., financial
input into current operation) and the control approach (e.g., financial or operational control) to
15


If a WWTP has complete financial or operational control over an operation, then the
GHG emissions are categorized as Scope I, and all remaining non-electricity
emissions are Scope III. However, if they have a joint share or partnership in various
processes, and do not control a majority financial or operational control, then the
WWTP would not include these emissions in Scope I.
Overall, the implications for WWTPs are that they must have financial or operational
control if the WWTP wants to account for their emissions as Scope I. Currently, this
rule has little impact on WWTP GHG inventories, since most WWTPs have financial
and operational control over all typical Scope I emissions (e.g., electricity, natural
gas, transport fuel, etc.). As a result, all other embodied and avoided materials (e.g.,
chemicals, piping, waste, avoided fertilizer, etc.) are automatically deemed as Scope
III. However, the WRI methodology is slowly integrating a more holistic LCA
approach to GHG accounting, as demonstrated by recent inclusion of Scope II
upstream electricity emissions. Scope II emissions are included since they consume
the electricity in equipment that the WWTP owns. Thus, it is possible that ultimately
the WRI protocol will include embodied energy for materials that they use in their
own operations. If this transition one day occurs, it will be critical for WWTPs to
own their solids handling operations so that they can realize benefits of avoided
fertilizer in Scope I.
establish organizational boundaries to consolidate GHG emissions. Companies can choose which
method they use. If a company has full financial control of all of its operations, then the financial and
operational boundaries are identical and the results of equity share and the control approach are
equivalent all emissions are allocated to Scope I. For equity share, Scope I GHG emissions are
allocated based on percent financial input into the current operation, where financial inputs include
economic expenditures for infrastructure, land, and operations & maintenance. For equity share,
financial input into the current operation holds legal precedence over actual ownership; however, the
two are typically self-aligned. For the control approach, if a company controls the majority of
financial or operational decisions, then that company allocates 100% of GHG emissions to Scope I. If
it does not control majority financial or operational decisions, then that company accounts for 0% of
GHG emissions as Scope I and then has the option to include these emissions in Scope III.
16


Nevertheless, the methodology to perform up to and including Scope III using WRIs
GHG Protocol for WWTPs has not been developed. Scope III inclusions yield a
more holistic inventory and a small set of most relevant Scope III inclusions are
recommended by the US EPA Climate Leaders Program (WRI, 2006). However,
LCA data are needed to inform Scope III.
2.4 Life Cycle Assessment
Life Cycle Assessment (LCA) is an objective process that quantifies the cradle-to-
grave life cycle impacts associated with a given product, process, or activity by
tracking and quantifying energy and materials inputs/outputs and their associated
emissions through a mass balance approach. LCA enables one to quantitatively
compare the sustainability of one product/process against another. When combined
with Sis, LCA serves as a common tool to enable a quantitative comparison of
WWTPs of different sizes and scales.
Environmental Life Cycle Assessments (LCAs) provide a total systems view of a
given product or process that can then be used to scientifically compare similar
products or processes, based on their environmental impact. In a Life Cycle
Inventory (LCI), material and energy inputs from resource extraction (the cradle) to
the end of life (the grave, such as landfilling or recycling) are quantified in a
systematic manner. All environmental inputs and outputs of a product/process
throughout production, usage, and disposal are quantified (Figure 3). For example,
material balances could include water, consumables (renewable vs. non-renewable),
chemicals, and waste input and output mass balances. The end result is a full systems
accounting of the environmental impact of a given product or process.
17


Inputs
Quip uts
R*w
Materials
Energy
Raw Materials Acquiatton |
I -
]
Manufacturing
~~r~
U so / Reuse / Maintenance
--.m
Recycle / WasteManagement |
Atmotsphenc
Emissions
Waterborne
Wastes
Solid
Wastes
Coproducts
Other
Releases
System Boundary
Exhibit 11. Lie Cycle Stages (Source: EPA. 199 3>
Figure 3: Life Cycle Inventory Stages of LCA (EPA, 2006)
There is no known work that assesses the sustainability of various WWTPs of
different sizes and scales throughout the US using a LCA-based approach to
specifically quantify the contribution of infrastructure on life cycle energy use and
GHG emissions. Such an analysis would enable city planners to select the most
sustainable option of WWT for their respective town/city. Existing regulations in
California, for example, require GHG accounting for new construction, but no known
information exists in literature on how much variation exists in existing WWTP
installations and how the system boundary affects results.
2.4.1 Main Steps in LCA
There are four steps to conducting a LCA:
1) Goal Definition and Scoping
2) Inventory
3) Impact Analysis
4) Interpretation
18


Goal definition and scoping establishes the project boundaries and the most
appropriate functional unit to meet project goals. In this project, all results are
expressed in terms of the functional unit of a volume of wastewater treated. The
system boundaries span from raw material extraction for all the energy and materials
used in a WWTP to the end-of-life land disposal of biosolids (and avoided fertilizer
use). In the Life Cycle Inventory (LCI) step, one quantifies all material and energy
flows associated with a product/process system, including greenhouse gas (GHG),
toxic emissions, etc. During the impact analysis step, one attempts to put meaning to
the LCI output numbers. Impact analysis can involve either mid-point or end-point
analysis. Midpoint assessment is a problem-orientated, scientifically-based approach
that reports LCI results within impact categories (e.g., raw amount of emissions that
goes to a specific media X tons of mercury per year or Y tons of GHG emissions
per year). Although scientifically-based, it is very difficult for decision-makers to
discern the impact of X number of tons of mercury emitted annually. Endpoint
analysis, on the other hand, is a damage-orientated method that takes these raw
numbers and converts them to a number that is more readily understood by the public,
such as expected mortalities per year. Given the complexity of natural systems, there
is much uncertainty in accurately quantifying the impact of a process on humans and
consequently the use of endpoint assessment can be controversial.
In this report, the focus will be on steps 1 and 2 above, and results will be presented
from a midpoint assessment perspective, i.e., in terms of raw numbers representing
environmental emissions and energy & water use. Endpoint assessment use in this
report will be limited and only used to shed light on a midpoint result.
19


2.4.2 LCA Methods
There are three main approaches to conducting environmental LCA:
1) Mass Energy LCA
2) Economic Input-Output LCA
3) Hybrid LCA, combining numbers 1 and 2 above
Each is described further below.
2.4.2.1 Mass/Energy LCA
In a mass-energy LCA, material and energy inputs/outputs along with their associated
impacts and emissions are tracked and quantified throughout the life cycle of a
product/process. Impacts include global warming potential, air, water, and soil
emissions, as well as toxic chemical releases. The methodology for mass/energy
LCAs is highly standardized, based upon International Standardization Organization
(ISO) 14040 series.
Mass-energy LCAs are the most commonly used LCA methodology and have been
used to evaluate the environmental impact of a variety of products such as the US
cement industry (Hanle, 2005) and an LCA of biodegradable plastic bags (James and
Grant, 2005). While this approach is suited well to examine the environmental
impacts of well-known products, detailed manufacturing data must be available to
expand the analysis to include the full embodied energy and environmental impact of
large-scale infrastructure used in WWTPs.
2.4.2.2 Economic Input-Output LCA
Economic Input-Output (EIO) LCA, developed at Carnegie Melon following the
methodology Leonteif (1970), uses monetary expenditures to estimate environmental
impacts of manufacturing a product. Monetary exchanges between over 500+ sectors
20


of the US economy are integrated into the analyses and linked to energy and
environmental impacts via national-level toxics release inventories and energy
consumption databases linked to economic data. Both direct as well as indirect
monetary exchanges that occur during manufacturing of a product are thus linked to
environmental impact. For example, laying a reinforced concrete pavement has direct
environmental impact from direct material use (e.g., manufacture of concrete, rebar,
aggregate, etc.), but also indirect impacts from use of infrastructures (e.g., concrete,
rebar, and aggregate) used to transport the needed reinforced concrete materials.
These interrelationships are often hard to quantify in a typical process-based LCA,
but can be quantified via EIO through the monetary exchanges between various
sectors of the economy and the associated aggregated environmental impact.
EIO represents the simplest approach to overcome the detailed materials data required
for a comprehensive upstream mass-energy LCA (Module 1). EIO LCA has been
applied to a wide variety of applications including passenger car comparisons (Cobas-
Flores et al., 1998) and the selection of various electronic components (Cobas et al.,
1995). If economic expenditures data are available, EIO will provide a full LCI in a
quick and simple manner. The simplicity, however, does cause some loss in accuracy
of the results, since EIO is based on a nationally aggregated database, where similar
processes/products are aggregated together. Thus, a more general result is obtained
that highlights major differences in expenditures on various sectors. Furthermore,
EIO only assesses the environmental impact of manufacturing goods/materials and
does not address the operations and maintenance phase of a products life cycle.
2.4.2.3 Hybrid LCA
A new approach to LCA, termed hybrid LCA, is emerging that attempts to combine
the benefits of mass-energy LCA for product/process operation (e.g., more specific
and accurate) with EIO LCA (e.g., easier to perform, less data and labor-intensive) to
21


capture nearly all of the upstream manufacturing (Lenzen, 2002). A hybrid LCA uses
EIO to simplify complex, multi-stage upstream manufacturing of the infrastructure
that is installed on-site at a WWTP, while relying on process-based LCA to capture
the operational use phase. The combined hybrid LCA approach can prevent
truncating of upstream manufacturing processes, which has been shown to create
errors as high as 50% (Lewandowska and Foltynowicz, 2004). Hybrid LCA use is
becoming increasingly popular (Horvath and Pacca, 2003) and has, for example, been
applied to evaluate the impacts of primary vs. rechargeable batteries (Lankey and
McMichael, 2000) as well as having been applied to Denvers city-scale GHG
inventory (Ramaswami et al., 2008).
Typically in LCA of buses, cars, etc, the operational and maintenance life cycle
phases represent 90% of the total life cycle impacts of a given process, while the
manufacturing phase impact is of the order of 10% (Beer et al., 2000). Nevertheless,
the considerably different infrastructure requirements of centralized vs. decentralized
systems merits further investigation of the applicability of this result to WWTPs.
Historical upstream mass/energy data for infrastructure inputs are difficult, if not
impossible, to gather while it is easier to find historical economic infrastructure
expenditures data from which a hybrid ElO-mass-energy LCA can be constructed.
Consequently, to capture both aspects of a WWTPs life cycle (i.e., infrastructure
installation, and operations), a hybrid LCA method will be used, coupling process-
based mass-energy LCA to address WWTP operations, and Economic Input-Output
(EIO) LCA to address the embodied energy of infrastructure. The methodology for
hybrid LCA development for WWTPs is described in Chapter 3.
22


3. WWTP Hybrid LCA Method Development
The methodology used to calculate annualized life cycle energy and GHG emissions
is detailed in this chapter. The intent is to begin from a more global view of how life
cycle energy and GHG emissions are calculated, and then gradually transition to a
more micro view of the details behind life cycle calculations.
The specific goals of this chapter are:
1) to introduce LCA-based Sustainable Indicators (Sis) developed for US
WWTPs
2) to develop a model for full system WWTP LCA for energy use and GHG
emissions calculations from Energy -> Materials & Infrastructure -> WWTP
-> Ag/Land Use
3) to discuss how life cycle energy and GHG emissions are annualized on a per
year basis
4) to detail input parameter estimation and benchmarking for energy intensity
and GHG intensity
The goal of this work is to develop and apply a transparent, benchmarked US WWTP
GHG accounting methodology that integrates LCA with WRI Scope 1, 2, and 3 and
can serve as a stepping stone towards standardization of WWTP LCA.
23


3.1 Sustainable Indicators
In an analogous manner to the work of Balkema et al. (1998), the Sustainable
Indicators (Sis) used in this work to quantify the sustainability of US WWTPs are
composed of four categories:
A. Performance and Management: BOD removal, Permit violations, etc.
B. Economic: Cost/gal treated, Capital investment
C. Environmental Impact: Plant footprint, Greenhouse Gas, Solids handling, etc.
D. Resource Efficiency: Water, Energy, Material, Nutrients
These Sis (Table 2) are the first comprehensive US list applied to WWTPs and
represent a more holistic approach than previous European works (e.g., Balkema et
al., 1998; Lundin, 1999, etc.).
Additional metrics proposed for the water cycle are highlighted in the following
section.
24


Table 2; Quantitative Metrics for US WWTP Sustainability
Quantitative indicators that represent various aspects of WWTP performance WWTP Data Units
A. Performance & Management Indicators
1. Annual BOD removal % removal
2. Annual TSS removal % removal
3. # of permit violations events/yr (days/yr)
4. Capacity Factor % hr on-line/yr
5. Work-related accidents # incidents/yr
6. BOD loading lb BOD/day-acre WWT system
B. Economic Indicators
7. $0&M per AMG $/AMG
8. $ Infrastructure/AMG Annualized $/AMG
9. $ Infrastructure/MGDcao Annualized $/MGDcaDacitv
10. % capital utilization % MGD treated/MGDcaDacitv
11. Capital removal eff. $ infrastructure/kg BOD removal
C. Resource Efficiency Indicators
Water Cycle
13. Water efficiency, i.e. volume WW generated/capita gal WW/capita-day
14. Water infiltration and inflow annual est. gal I&I/AMG
15. % Process water use MG process water use/AMG
16. % water recycled/reused gal reused/AMG
17. Relative water reuse MG reuse/MG water consumption
18. % local aquifer recharge gal GW recharge/total gal discharge
19. WW watershed return MG watershed discharge/AMG
Energy Cycle
20. EUEI (GJ + GJ fuel)/AMG (GJ PE/AMG)
21. ALCPEIwwtp GJ/AMG
22. % renewables of EUEI GJ renewable/GJ
23. % biogas flared on-site GJ flared/GJ produced
Material Cycle
24. Off-site piping infrastructure lb material/customer
25. Infrastructure emb. energy GJ embodied/MGD capacity
Nutrient Cycle
26. Nutrient Recycling kg fertilizer/AMG
25


Table 2: Quantitative Metrics for US WWTP Sustainability (Cont)
Quantitative indicators that w represent various ^ js aspects of WWTP ^ performance Units
D. Environmental Impact Indicators
27. Annualized GHG Emissions by Category for All WW Operations
a. On-site Electricity kg C02e/AMG (% gross total)
b. Off-site Electricity kg C02e/AMG (% gross total)
c. On-site Natural Gas kg C02e/AMG (% gross total)
(Off-site Natural Gas) kg C02e/AMG (% gross total)
d. Biosolids Transport kg C02e/AMG (% gross total)
e. Chemicals kg C02e/AMG (% gross total)
f. On-site infrastructure kg C02e/AMG (% gross total)
g. Off-site infrastructure kg C02e/AMG (% gross total)
h. On-site maintenance kg C02e/AMG (% gross total)
(Off-site maintenance) kg C02e/AMG (% gross total)
i. Digester Electricity kg C02e/AMG (% gross total)
j. Digester Heat Rx kg C02e/AMG (% gross total)
k. Avoided Fertilizer kg C02e/AMG (% gross total)
1. Avoided Water kg C02e/AMG (% gross total)
m. IPCC GWP emissions kg C02e/AMG (% gross total)
n. BOD emissions kg C02e/AMG (% gross total)
28. ALCGWPIwwtp kg C02e/AMG
29. On-site plant footprint acre required/AMG
30. Chemical inputs kg chemicals used/AMG
31. Biosolids transport wet mt-km/AMG
32. Landfill waste wet lb solids/gal treated
LEGEND: ALCGWPwwtp = WWTP annualized life cycle GWP intensity; ALCPEIWwtp = WWTP
annualized life cycle primary energy intensity; AMG = annualized million gallons treated; BOD =
Biological Oxygen Demand; Emb. = Embodied; EUEI = End-use Energy Intensity; GJe = GJ of
electricity; GW = groundwater; I&I = Infiltration & Inflow; IPCC = Intergovernmental Panel on
Climate Control; O&M = Operations & Maintenance; PE = Primary Energy; TSS = Total Suspended
Solids; WWT = Wastewater Treatment; WWTP = Wastewater Treatment Plant
NOTES: BOD emissions have a net zero impact on GWP (IPCC, 2006)
All $US reported in year 1997 equivalents
26


3.1.1 Water Cycle Indicators
Because water is the central component to wastewater treatment, particular attention
was paid to developing water cycle indicators. Literature indicators commonly
address availability, but few indicators have been developed that enable simple life
cycle quantification. Further, many of literature-reported water cycle indicators are
regional in scale (e.g., regional water availability) and cannot be applied to WWTPs.
In addition, many of these indicators have regional variability and the values of input
parameters are not available in the literature. Consequently, many of the indicators
presented below are not quantified in this study, but are rather recommended as a
research need for future studies.
3.1.1.1 Global Indicators
The UN uses the percentage of the national populations that have improved drinking
water and sanitation as an important water cycle social indicator. For the US, the
percentages are very high as shown in Table 3.
Table 3: US Population with Access to Improved Drinking Water and
Sanitation (UN, 2006)_________________________________
% of US Population
Population has access to; Total Urban Rural
Improved Drinking Water 99% 100% 94%
Improved Sanitation 100% 100% 99%
3.1.1.2 Regional Indicators
Many literature-reported water cycle indicators require knowledge of regional
characteristics. For example, annual renewable total water supply (e.g., residential,
industry, commercial, and agriculture) per person can shed some insight on water
resources. The World Resource Institute defines water stress as having less than 1700
m3/person-year of renewable water supply (WRI, 2000). Only south central Colorado
is experiencing water stress in terms of renewable water supply (500-1000 m /person-
27


yr). Front Range Colorado has 4000-10,000 m3/person-year available for renewable
water supply.
Figure 4: Annual Renewable Water Supply per Capita
Colorado outlined on map. Source: World Resource Institute, 2000
Vorosmarty et al. (2000) combines renewable water supply with use through the
relative water demand indicator. Relative water demand is defined as the ratio of
water use to total water availability [(water consumed/(surface water discharge +
sustained groundwater yield)]. When comparing regional water availability, this
metric assures that people do not consume more water than is renewably available.
Vorosmarty et al. (2000) defined a value above 0.4 as severe water scarcity. Figure 5
shows that the Front Range, Colorado region (Colorado is outlined in figure) is in a
severe water scarcity situation.
Using Colorado total annual water use data from the USGS (e.g., residential, industry,
commercial, and agriculture) and CO Census data, the total use of 4047 m3/per-yr
(2,929 gal/person-day) was calculated for the state of Colorado (USGS, 2005; US
Census Bureau, 2000). This indicates that average Colorado usage falls within Front
Range, CO renewable water supply of 4000-10,000 m3/per-yr. However, using the
range of 4000-10,000 m3/per-yr, Colorados relative water demand range of 0.56-1.40
can be calculated. Thus, even Front Range, Colorado, which has above average
28


renewable water availability, has severe water scarcity in terms of the 0.4 relative
water demand threshold set by Vorosmarty et al. (2000).
Figure 5: Contemporary Population Relative to Relative Water Demand = 0.4
Colorado outlined on map. Denotes 1000s of people above or below severe Relative Water
Demand threshold of 0.4. Source: Vorosmarty et al., 2000
Table 4 highlights potential sustainable water use indicators. Some of these
indicators are needed to facilitate quantification of life cycle studies (e.g., ratio of
treated/untreated water use, etc.), while others are essential to quantify supply vs. use
characteristics. Values were calculated for indicators with sufficient data found in the
literature. In particular, a total of seven indicators were applied to WWTPs in this
study and ranges are reported in Table 4. Values for individual WWTPs in this study
can be found in Chapters 4 and 5.
29


Table 4: Proposed Sustainable Water Indicators
Water Cycle Metric Description Units
Metrics Detailed in This Section for Front Range, Colorado
Consumption6 Total water use per capita 4047 m3 use/person-yr
Availability7 Annual renewable water supply (all sources) 4000-10,000 m3 supply/person-yr
Withdrawal Relative water demand (Ratio of Withdrawal to Availability) 0.56-1.4 m3 use/m3 supply
Metrics Calculated for Wastewater Treatment Plants in This Study (Ch. 4 & 5)
Wastewater Generation Wastewater generation per capita 69-140 m3 WW treated/person-yr
Aquifer Recharge Ratio of aquifer to total water (aquifer+surface water) discharge 0-1 m3 aquifer recharge/m3 total discharge
Process water use8 Ratio of process water use to wastewater treated 0.0026 m3 process water/m3 WW treated
Water Reuse Ratio of WW reused to wastewater treated 0-0.12 m3 WW reused/m3 WW treated
Relative Water Reuse9 Ratio of WW reuse per capita to consumption per capita 0-0.056 m3 reuse/m3 consumption
WWTP I&I Ratio of infiltration and inflow to wastewater treated 0.11-0.25 m3I&I/m3WW treated
Wastewater Watershed Ratio of wastewater discharged in local 1 m3 WW watershed
Return watershed to total wastewater treated discharge/m3 WW treated
Other Regional Metrics Requiring More Data
Reliability Ratio of local to imported water supply m3 local/m3 imported
Watershed Return Ratio of water returned to the local watershed (aquifer+surface water) to total water use m3 watershed retum/m3 total water use
Surface Water Ratio of total surface water discharge to m3 surface discharge/m3
Replenishment total stream flow stream flow
Indoor Water Use Ratio of wastewater generated to total treated water use m3 WW treated/m3 treated water use
Type of Use Ratio of groundwater use to freshwater use m3 groundwater use/m3 freshwater use
Treated Water Ratio of treated surface water/total surface water use m3 treated surface use/m3 total surface use
Groundwater Treatment Ratio of groundwater treated/total groundwater use m3 treated groundwater use/m3 total groundwater use
6 Calculated fromUSGS, 2005; US Census Bureau, 2000
7 Reported in WRI, 2000
8 Data only available for MWRD, which uses 119.5 MG/yr of treated water for on-site processes, while
treating 46,458 MG/yr.
9 MWRD reuses 9.2 gal/person-day of water. Denver Water treats 164 gal/capita-day of potable
drinking water.
30


3.2 Energy & GWP Calculation Methods Overview
Energy use and Global Warming Potential (GWP) data are organized according to
three major categories and system boundaries:
1) End-use Energy (EEnd-use)
a. Operational primary energy (electricity, natural gas, etc.) for on-
and off-site collection and treatment including all upstream fuel
processing and losses
b. Solids handling transport including all upstream fuel processing
and losses
2) Embodied Energy of Materials (EEmbodied)
a. Embodied energy of chemicals including raw material extraction,
manufacture, and transport to WWT facility
b. Embodied energy of on- and off-site infrastructure and
maintenance using EIO-LCA
3) Avoided Energy (EAvoided)
a. Avoided energy use from water reuse including embodied
materials and upstream fuel processing and losses
b. Avoided fertilizer use including raw material extraction,
production, and transport to fields
c. Avoided energy (e.g., biogas electricity production and heat
recovery, etc.), including all upstream fuel processing and losses
By organizing data into these three major data categories (Table 5), results are readily
compared to existing GHG accounting reporting standards, such as the World
Resources Institutes (WRIs) GHG Protocol (WRI, 2004). Table 5 shows that the
LCA boundaries used in this work represent the most holistic LCA possible. Scope I
and II are termed an inventory, while Scope III enables one to determine the overall
31


footprint (Ramaswami et al., 2008). Thus, the analysis performed in this work is the
first GHG footprint computation for a US WWTP. The analysis yields life cycle
primary energy (LCPE) and life cycle global warming potential (LCGWP), as
detailed next.
Table 5: WRI Scope Requirements and LCA Boundaries
Scopes
Scope I and II
+ Waste
Scope III
Requirement WRI GHG Emissions Boundaries
Required
Highly
Recommended
by EPA Climate
Leaders
Protocol
......................... ................"i
I Scope I: Direct end-use energy ;
l Scope II: Indirect electricity + Waste ;
Scope III:
Can add any of the following indirects using LCA:!
; Upstream fuel processing
I Direct materials processing ;
Direct avoided energy |
Transmission line loss !
LCA Boundary
for this Study
More specifically, the hybrid LCA boundaries of a WWTP used in this study are
shown in Figure 6. End-use energy, used in this study, represents the typical WRI
Scope I and II boundaries. The addition of waste to Scope I and II is required by
WRI Protocol, while this study also includes additional Scope III embodied and
avoided energy categories, as well as upstream fuel processing and transmission loss.
32


EIOLCA
Figure 6: Hybrid LCA WWT system boundaries and components
Hybrid LCA results will focus on:
Energy Use
GHG Emissions
Economics
Comparison across scale
33


3.2.1 Life Cycle Primary Energy vs. End-use Energy
Before launching into calculation of LCPE and LCGWP, it is important to discuss the
difference between end-use energy and life cycle primary energy. Cradle-to-grave
life cycle assessments calculate the raw upstream material and energy use from
resource extraction to end-of-life (EPA, 2006). This study performs a near complete
cradle-to-grave LCA, but excludes the highly variable ultimate demolition of the
WWTP. The number of GJ of electricity (GJe) consumed for treatment activities plus
the thermal heat equivalent energy (GJth) of fuels used on-site for heating and
transportation (e.g., natural gas, gasoline, diesel, etc.) is called the End-use Energy
(EUE). End-use energy grossly underestimates the true life cycle energy use since it
does not include the raw fuel energy to make electricity, as well as material
production and its associated transportation. For electricity, about 3 GJth (GJth = GJ
of thermal heat energy)10 of raw fuel energy is required to make 1 GJe of electricity
(GJe = GJ of electricity). When electricitys raw fuel energy and material production
energy is included, this is termed primary energy, or the heat equivalent energy of
fossil and renewable fuels consumed for transport, heating, materials production, as
well as the amount of these fuels used for electricity generation (GJ) (EERE, 2008).
When upstream fuel processing, power plant construction, and transmission losses are
additionally included, it is termed the life cycle primary energy (Table 5).
10 GJ* used interchangeably with GJ throughout this work, as all energy units are reported as primary
energy (unless otherwise noted). For primary energy, 1 GJ is the same as 1 GJ*, as both describe the
raw heat energy used in a particular process; heat and energy are equivalent and mutually transferable.
34


Table 6: End-use, Primary, and Life Cycle Primary Energy Comparison
Energy Type Electricity GJe Site Fuel Use GJ Electricity Fuel Energy GJth Materials Production Energy GJ Upstream Fuel Processing and Losses GJ
End-use X X
Primary X X X X
Life Cycle Primary X X X X X
While life cycle primary energy is more inclusive than end-use energy, both are
important to make conclusions about system efficiency. The total life cycle primary
energy use can be compared to the end-use energy, providing insight as to how well
the quality of energy was matched to its specific use.
Energy quality deals with the Second Law of Thermodynamics: energy systems have
a tendency to increase in entropy, which implies that the inherent energy quality of a
source (e.g., electricity highest quality; low temperature heat lowest quality)
should be best matched to its intended energy quality use (e.g., low temperature heat
used for low temperature heat, etc.) to minimize entropy losses. Although electricity
can be directly converted back into heat with 100% efficiency (and is thus the highest
quality of energy), large amounts of heat energy go into making electricity in the first
place, and thus this would be a very poor use of energy quality.
According to Amory Lovins of the Rocky Mountain Institute, a thermodynamically
sound energy policy would rely increasingly on more efficient uses of renewable
energy that minimize entropy, while matching scale and quality to end use.
Mismatching high energy quality for low energy quality purposes leads to poor
system efficiency and large entropy losses. In Amory Lovins words, using high-
35


quality electric power for purposes such as cooking and space heating is like cutting
butter with a chainsaw (Edinger, 1999).
Of particular interest to the above definition of primary energy, renewable energy is
included into primary energy calculations. For renewable liquid fuels, raw fuel heat
equivalents are used (EOE, 2007). For renewable electricity, photovoltaics (PV),
wind, and hydroelectric are treated differently when compared to heat-derived
biomass and geothermal. In the US, the two methods used by the US Energy
Information Administration (EIA) and the National Renewable Energy Laboratory
(NREL) calculate renewable electricity primary energy calculation differently (EOE,
2007), as shown:
Table 7: Renewable Electricity Primary Energy Quantification
Renewable Electricity Source Method
EIA, 2007
Wind, Solar, Hydroelectric NREL, 2007
________________________________This work
EIA, 2007
Biomass, Geothermal11 NREL, 2007
This work
tipp Used
Fossil Fuel Equivalent -33%
100%
100%____________________
Actual US Avg. Heat Rates11 12
-25%
33% Assumed
Actual US Avg. Heat Rates12
-25%
The EIA approach essentially calculates the primary energy used if all energy sources
were fossil fuel; meanwhile, NREL assumes a 33% efficiency for biomass and
geothermal efficiencies. The approach in this work is to adopt the approach of NREL
(e.g., 100% efficiency for wind, PV, and hydroelectric), but to replace the assumed
33% efficiencies with actual US average efficiencies for nuclear, biomass, and
geothermal electricity generation. Lastly, all energy sources incorporate upstream
11 NREL and the EIA treat nuclear power in a similar manner, using average plant heat rates
12 Heat rates are simply the heat energy (GJ*) required to produce 1 GJ of electricity (GJe), having
units of GJ,h/ GJe. GJe = GJ of electricity; GJth = GJ of heat energy.
36


fuel processing and losses. The underlying details behind life cycle primary energy
and GWP calculations are described further in the following section.
3.3 Calculating Life Cycle Primary Energy and Life Cycle GWP
3.3.1 Life Cycle Primary Energy
The summation of the primary energy used in each energy use category is defined as
the total life cycle primary energy use, LCPE^ai (Units: GJ), as shown:
LCPETotai LCPEEnd_use +LCPEEmbodied -LCPEAvoided
where: LCPEjotai = Total life cycle primary energy use for entire WWT system from
Energy -> Materials & Infrastructure -> WWTP -> Ag/Land Use (GJ)
LCPEEnd-use = Life cycle primary end-use energy use (GJ)
LCPEEmbodied = Life cycle primary embodied energy use (GJ)
LCPEAvoided = Life cycle primary avoided energy use (GJ)
Calculation of the total life cycle primary energy (LCPEx0tai) entails using Primary
Energy Intensity (PEI) factors. The primary energy intensity for a given unit of
consumption (e.g., electricity use, pipe or chemical use, etc.) is the cumulative
amount of primary energy used to deliver this product to the treatment plant,
including all upstream energy use during mining/manufacturing and embodied energy
in materials (e.g., petroleum products used to make plastics). The expanded form of
Equation (1) is then:
LCPEfotal S[Xt E End-use^ PEI End-usei l+£[Af Embodied PEI Embodiedj 1
EX E Avoided h* PEl Avoided k l]
k k (2)
where: n = number of data years
i = number of energy end uses (typically four: electricity, natural gas,
transport fuel, and biogas)
j = number of different materials types (includes infrastructure materials and
chemicals)
37


k = number of avoided energy categories (includes avoided fertilizer, water,
and energy, such as biogas)
^End-use= Total end-use energy for each end-use energy type (GJe, GJth)
PElEnd-use= Primary energy intensity for each end-use energy type
(GJ/GJe, GJ/GJth)
MEmbodied = Total mass of materials used for each infrastructure or material
type (mt)
PEIsmbodied = Primary energy intensity for each material type
(GJ/mt material, GJ/$1 capital expenditure)
EAvoided = Avoided energy use for each avoided type (GJ)
PE I Avoided = Primary energy intensity for each avoided material (GJ/MG water
avoided, GJ/GJe avoided, GJ/mt fertilizer avoided)13
3.3.2 Life Cycle Global Warming Potential
The global warming potential (GWP) reports greenhouse gas emissions in tonnage of
carbon dioxide equivalents (e.g., mt-CC^e, where mt = 1 metric tonne and CC^e =
carbon dioxide global warming potential equivalents CO2). The Intergovernmental
Panel on Climate Change (IPCC) defines the 100-year GWP as (2007):
GWP-C02e =Ca+C//4x25 +N2Ox298 (3>
where: GWP-CC^e = Global Warming Potential, reported as mass of carbon dioxide
equivalents
CO2 = mass of carbon dioxide
CH4 = mass of methane
N2O = mass of nitrous oxide
It is important to recognize that biological oxygen demand (BOD) related CO2
emissions are not included in IPCC Protocol, since these emissions are biogenic in
origin and thus carbon-neutral (IPCC, 2007).
Similar to Equation (1), the total life cycle GHG emissions in Global Warming
Potential (GWP) equivalents, LCGWPxotai, (Units: mt-C02e) is defined as the
summation of the GHG emissions from each LCA category, as shown below:
13 MG = Million Gallons
38


LCGWPmai = LCGWPEnd_use + LCGWPEmbodied-LCGWPAvoided (4)
where: LCGWPxotai = Total life cycle GWP for entire WWT system from Energy ->
Materials & Infrastructure -> WWTP -> Ag/Land Use (mt-C02e)
LCGWPEnd.use = Life cycle primary end-use GWP (mt-C02e)
LCGWPEmbodied = Life cycle primary embodied GWP (mt-C02e)
LCGWPAvoided = Life cycle primary avoided GWP (mt-CC^e)
Similarly, the total life cycle global warming potential can be defined in more detail
using an Emission Factor (EF) for unit of consumption (e.g., electricity production,
pipe, chemical production, etc.). The emission factor for a given energy process
(e.g., electricity production, road transport, etc.) or material generation process (e.g.,
pipe or chemical production, etc.) is the cumulative amount of GHGs-CC^e emitted to
deliver this process/service, including all upstream energy use during fuel processing.
The expanded form of Equation (4) is then:
LCGWPTotal = S[ Xt E End-use t *EF End-use { 1+ Embodiedy* EF Embodied j 1
E Avoided],* EF Avoided], l]
* k k (5)
where: EFEnd-use= Emission factor for each end-use energy type
EFEmbodied = Emission factor for each material type
EFAv0ided = Emission factor for each avoided material
3.4 Annualizing and Normalizing Life Cycle Data
LCA studies are always normalized according to the functional unit. The functional
unit is the intended output or goal of an LCA product/process and provides a basis
for comparison with other studies. In the case of wastewater treatment, the intended
goal of the process is to treat wastewater, and as a result, the functional unit in this
study is one million gallons of treated wastewater (hereon referred to as MG treated).
39


In an ideal scenario, where all data dating back to plant inception are available, the
LCPEjotai would be normalized to the MGrotai (i.e., the total MG treated over the
WWTPs life):
LCPEI,
WWTP
ALCPEoperations + ALCPE Infrastructure ALCPE 4 voided
AMG
LCPETotal S LCPEOperations + LCPEInfrastructure LCPEAvoided
L CPEl wwtp = j^q
'Total
m*AMG
(6a)
(6b)
where: LCPEIwwtp = Life cycle primary energy intensity for the WWTP, GJ/MG
m = life span of the wastewater treatment plant, yr
AMG = annualized million gallons treated, MG/yr
ALCPEjotai = total annualized life cycle primary energy, GJ/yr
ALCPEoperations = annualized life cycle primary energy for end-use energy
and chemicals, GJ/yr
ALCPEjnfrastructure = annualized life cycle primary energy for infrastructure, GJ
ALCPEAvoided = annualized life cycle primary energy for avoided energy &
materials, GJ/yr
The major drawbacks of this approach are:
1. AMG & Operations historic data not available
2. PEIs, EFs have changed over 40 years
As a result, a better approximation is annualizing LCPEI:
ALCPE Operations ALCPE Avoided
LCPEI
current year
WWTP
LCPEI\
current year
WWTP
AMG
A J (~'ppcurrent year AT CPF'"'""" + AT CPF
riT-'A 1-j Operations ^Avoided Infrastructure
current year
,current year
LCPE Infrastructure
Ttl* AMG current year
(7a)
(7b)
AMG
,current year
40


Similarly, for GWP intensity:
LCGWPI
current year
WWTP
ALGWP Operations ~ ALGWP Avoided
AMG
current year
LCGWP Infrastruc ture
m* AMG current year
(8)
where: LCGWPIwwtp = Life cycle global warming potential intensity for the
WWTP, GJ/MG
Unless otherwise noted, results will be reported in the form of Equations (7) and (8).
Next we will examine the primary data gathered to better understand the individual
LCA data categories.
3.5 Energy and GHG Intensity Metrics
Four useful intensity metrics can be determined from the above methodology. Table
8 depicts the four metrics and their significance. End-use Energy Intensity (EUEI)
is defined as the sum of on- and off-site site electricity use (GJe) and the heat energy
of fuels used for heating and transportation (GJfueis). When compared with the life
cycle primary energy for end-uses, information can be gathered as to how well the
quality of energy was matched to its end-use.
41


Table 8: Energy and GHG Intensity Metrics
Intensity Metric Description Units
EUEI Energy use in operations phase (WWTP) (GJe + GJfoeO/AMG
EUEI LCPEI End_use End-use efficiency: How is quality of energy matched to end-use? GJeuei/GJpe14
LCPEIwwtp WWTP primary energy footprint (Energy->Inf./Mat.->WWTP->Ag. Land) GJpe/AMG
LCGWPIwwtp WWTP GHG footprint (Energy-Mnf./Mat.->WWTP->Ag. Land) mt CChe/AMG
3.6 Model Input Parameters
As seen from Equations (1) to (8), the total LCPE and LCGHG are computed using
the master Equations (9) and (10), respectively, shown below:
LCPE = y [ £ MFA i > \pek\ m
t.j.k : :
LCGWP = v {: E/MFA > do)
i.j.k r 1
Consumption Intensity
Parameters Parameters
where: E/MFA = Energy or material flow analysis consumption input
Consumption and intensity parameters are described for each of the three energy use
categories below. Consumption input parameters are shown in Table 9, including
three energy use categories, general characteristics (primarily used for normalization),
and wastewater performance characteristics. Annual on- & off-site capital
14 PE = Primary Energy
42


expenditures are collected dating back as far back as possible, ideally since initial
plant construction.
Table 9: WW LCA Consumption Input Parameters
Primary Data Gathered Primary Data Gathered
General data Wastewater Characteristics
Million gallons (MG) treated/yr Number of customers Million gallon/day (MGD) capacity Plant footprint WW characteristics: BOD5, TSS, TN (mg/L) Annual permit violations Volume recharged to aquifer Estimated MG I&I/yr
Service area
Unit operation downtime Embodied Energy
Chemicals
End-Use Energy Annual mass of chemical, by type
On-site Operating Energy % active fraction
Annual kWhe used from grid Distance from manufacturer -> WWTP
Annual Btu natural gas used Transport method (e.g., rail, truck, etc.)
Annual electricity from digester gas Capital Expenditures
Annual heat recovery from digester gas Annual on- & off-site capital expenditures
Off-site Operating Energy Annual maintenance capital expenditures
Electricity supplier grid mix
Annual kWh used Avoided Energy
Annual Btu natural gas used Avoided Biogas
Solids Handline Biogas electricity (kWhe), heat recovery (Btu)
Annual dry mass biosolids generated, %TS Avoided Fertilizer Use
Distance and mass fractions of biosolids % TN, TP, TK of biosolids
Land application rate Avoided Water
Other spreading/processing fuel use Annual MG water reused
3.7 Calculation of Intensity Parameters
The following sections detail the calculations performed for the PEIs and EFs used
and their associated benchmarks. Since US LCA databases are extremely rare, most
US LCA studies rely on European databases for life cycle inventory calculation.
Existing US databases, such as NREL and SimaPros Franklin Database (e.g., LCA
software), have sparse entries and very few of these are relevant to wastewater LCA
(NREL, 2007; SimaPro, 2006).
43


As a result, extensive background work was performed to utilize and create US-based
databases where possible. Sources such as EIO-LCA, NREL, the US Department of
Energys Energy Information Administration (EIA), EPA, etc. were used to create
US-relevant PEIs and EFs. When US data were unavailable, PEIs and EFs defaulted
to European LCA databases from two LCA software, namely TEAM (Tool for
Environmental Analysis and Management) and SimaPro.
When available, PEIs and EFs used in this study were benchmarked to values
reported in the literature.
3.7.1 End-use Energy
3.7.1.1 End-use Consumption Parameters
Five end-use consumption parameters were collected from the sources shown (Table
9):
Electricity utility power bills
Natural Gas utility bills
Biogas Electricity electricity metered directly on-site
Biogas Heat Recovery heat recovery estimated based on known heat
exchanger efficiency
Solids Transport biosolids annual reports
Building upon Equation (2), we calculate life cycle primary energy (including
upstream fuel processing and losses) associated with five end uses of energy, namely
electricity, natural gas, transportation fuel use, biogas electricity, and biogas heat
recovery. End-use energy consumption for WWTP operations, Ej, is obtained from
utility billing data, detailing annual natural gas and electricity totals. When available,
end-use energy data (Ej, i = 1.. .5, etc.) are collected separately for all on- and off-site
44


WWT activities (Table 9). The life cycle primary energy use for each end-use energy
type (e.g., electricity, natural gas, etc.) is summed to determine the total life cycle
primary end-use energy, LCPEEnd.use (Equation (2)). A similar process is repeated
for life cycle end-use energy global warming potential, LCGWPEnd_use (Equation
(5))-
3.7.1.2 Grid Specific Electricity Life Cycle Calculation Techniques
The electricity grid mix of the supplier of electricity (e.g., Xcel, Longmont Power,
etc.), dictates the magnitude of the PEIs and EFs for life cycle end-use electricity
calculations. When a grid mix is dominated by coal, more primary energy will go
into making one kWh of electricity, thus emitting a larger amount of GHGs. Thus, a
coal-rich grid mix will have a larger PEI and EF, when compared to a renewable-rich
grid mix.
Calculation of a composite GJ of electricity (GJe) is adapted from Kim and Dale
(2005), which is based on US ELA (Energy Information Administration) average
thermal efficiencies at U.S. power plants. Typical LCA calculations only account for
the direct primary energy, or the raw fuel energy (GJth) that goes into making one
GJe with added line losses (WRI, 2004). A more complete life cycle approach
accounts for the indirect energy that goes into making a composite GJe, including
both upstream fuel processing and power plant construction (Meier, 2002; Kim and
Dale, 2005). As shown in Table 6 above, the life cycle primary energy is the sum of
both the energy use up to the power plant (including line losses) plus the energy use
upstream of the power plant encompassing fuel processing and embodied energy of
infrastructure, captured in Rpayback-
Calculation of the life cycle primary energy intensity (PEIeiectncity) for a composite GJe
with line loss, can be summarized as:
45


Y.LCPE Electricity ^ Fp(l + F,) Fp
PEI Electricity ~ rp , ) = X ' "
L t End-use electricity e ^ Tjpp Rpayback
where: p = Total number of component energy sources in the grid mix (e.g., coal,
natural gas, etc.)
Fp = Fraction of component p in grid mix (e.g., coal, natural gas, etc.), GJe
F| = Electricity line loss factor, 8.036% for Colorado15
Ppp = US average power plant efficiency, GJth/GJe
LCPEjriectricity = Life cycle primary energy for electricity use, GJ
EEnd-use electricity = Electrical energy consumed by the WWTP, GJe
(11)
Values for Fp are shown in Figure 7, using the 2000 US average electricity grid mix
as an example.
Hydroelectric Wind
Geotnermal______ 7 1% |~0 1%___________Solar (Photovoltaic)
a Bituminous
Subbituminous
Lignite Coal
Oil (Residual Fuel
Oil)
Natural Gas
a Other Fossil
Nuclear (U02)
Biomass/Wood
Geothermal
Hydroelectric
Wind
a Solar (Photovoltaic)
Figure 7: US Electricity Grid Mix (NREL, 2007: Year 2000 data)
The energy payback ratio, Rpayback, incorporates embodied energy of power plant
infrastructure plus upstream fuel processing energy, defined as (Meier, 2002):
15 CO is part of the Western Electricity Utility, having a collective line loss factor of 8.036%, based on
US EPA 2000 e-GRID data (NREL, 20007)
46


R Payback
LCEElectric-out (12)
LCPE [ndlrect
where: Rpayback = life cycle upstream energy payback ratio, GJe/GJth
LCEEieCtric-out= Life cycle electric output of power plant p (e.g., coal, natural
gas, etc.), GJ over entire life
LCPE indirect = Life cycle indirect primary energy of power plant p (e.g., coal,
natural gas, etc.), GJ over entire life
The values used to calculate PElEiectricity are shown in Table 10, using Xcels
electricity grid mix as an example. For Xcels grid mix (Table 10), Rpayback is highest
for hydroelectric, followed by biomass, and conventional coal. No PElEiectricity
benchmark exists for Xcels grid mix; however, the PElEiectricity is consistent with the
literature range of 3.28 to 3.38 for average CO and US electricity grid mixes (EIA,
2001).16
16 Xcels grid mix is nearly 40% natural gas, compared to 22.4% for COs average; thus, Xcels grid
mix should have a lower PEI when compared to CO.
47


Table 10: Major Variables for Calculating PElElectricity
Energy Source F 17 rp GJe Hpp GJe/GJth Direct Only PElElectricity fOP a Composite GJe18 GJth/GJe Rpayback GJe/GJth PElElectricity for a Composite GJef818 GJth/GJe
Bituminous 53.38% 32.9%19 1.76 112 1.81
Subbituminous 3.41% 30.6%19 0.12 ll20 0.124
Natural Gas 39.53% 38.3%19 1.12 4.120 1.22
Hydroelectric 2.05% 100%21 0.02 16.922 0.02
Biomass 1.64% 22.0%23 0.03 14.920 0.03
Totals 100% 3.00 | PElElectricity 3.20
To more clearly address model uncertainty for PEIEiectncity, modeled values for CO
and the US were compared to literature. Typically, literature reported PEIs for
electricity do not include upstream fuel processing. Further, as discussed above, the
ELA assumes a fossil fuel equivalent thermal electricity generation efficiency, r|pp, for
renewables (-33%). Thus, to benchmark the model to literature values, upstream fuel
processing was excluded and qpp for renewables was assumed equivalent to a
weighted average of fossil fuel thermal efficiencies (-33%). Comparing the model
used in this study with the above changes to literature reported values, the percent
variation ranged from 0.88% to 2.24% (Table 11).24 As can be seen from Table 11,
the model is extremely consistent with EIA reported PEIs for Electricity. Upstream
17 Chapman, 2006
18 Includes transmission line loss = 8.036% (NREL, 2007)
19 Kim and Dale, 2005
20 Meier, 2002
21 NREL, 2007
22 White and Kulcinski, 1998
23 US EPA eGRID model reports nearly 100% of the biomass contribution of the EERE, 2008
24 In 2000, the US had an average transmission loss of 9.91% (NREL, 2007).
48


fuel processing represents an additional 3.54% to 3.88% increase for CO and US
average PEIs, respectively.
Table 11; PElEiectridtv Benchmarked for US and CO Averages
Grid Mix Model Value Literature Value Percent Standard
GJth/GJe GJth/GJe Variation Deviation
Colorado Avg 3.21 3.28 2.24% 0.0507
USAvg 3.35 3.38 0.88% 0.0208
Similarly, the values used to calculate the emission factor for electricity, EFpjectricity,
are shown in Table 12.
Table 12; Major Variables for Calculating Emission Factors for Electricity
Energy Source F25 rp GJe Direct Only GHG Emission Rate for EFEiectricity for a Component p Composite GJe26 mt C02e/GJe mt C02e/GJe Life Cycle Primary Energy GHG Emission Rate for EFEiectricity for a Component p Composite GJe27 mt C02e/GJe mt C02e/GJe
Bituminous 53.4% 0.284 0.151 0.289 0.167
Subbituminous 3.4% 0.284 0.00966 0.289 0.0107
Natural Gas 39.5% 0.141 0.0683 0.173 0.0740
Hydroelectric 2.1% N/A N/A 0.00500 0.000111
Biomass 1.6% N/A N/A 0.00917 0.000162
Totals 100% 0.217 EFEiectricity 0.252
EFs for electricity benchmarked to literature reported values are shown in Table 13.
These values do not include upstream fuel processing. Upstream fuel processing adds
an additional 7.29%, 13.4%, and 14.6% for Xcel, Colorado, and the US, respectively.
25 Numbers may not add up to 100% due to rounding (Chapman, 2006)
26 Includes transmission line loss = 8.036% (NREL, 2007)
27 Includes transmission line loss = 8.036% (NREL, 2007)
49


Grid Mix Model Value Literature Value
mt CChe/GJe mt C02e/GJe
Percent
Variation
Standard
Deviation
Xcel
Colorado Avg
US Avg
0.217
0.244
0.178
0.221
0.244
0.170
1.93%
0.04%
4.94%
0.00297
0.0000610
0.00623
3.7.1.3 PEI and EF for Natural Gas
When available, natural gas consumption data are collected separately for on- and
off-site natural gas use by the WWT system. The PEI and EF for natural gas was
derived from NRELs LCI database. The separate consumption usage values are then
applied to Equation (2) and (5) with the PEI and EF to calculate the life cycle
impacts. Fugitive losses can be significant representing 1.4% of total natural gas
production, with literature reported ranges of 1-4% (Spath and Mann, 2000).
3.7.1.4 Solids Handling
Transportation impacts are modeled by computing the primary energy used to
transport one ton of material for one mile. In the US, this method is commonly
referred to as the ton-mile method.
When using the ton-mile method, there are two methods used for calculation
(Mangmeechai and Matthews, 2007):
1) multiply the distance of each transport by the mass (in tons) to get ton-
miles for individual shipments. Sum all ton-miles for individual
shipments to get total ton-miles.
50


2) compute the average haul distance (total miles/total number of shipments)
and multiply this average haul distance by the total number of tons
transported.
The two methods lead to an identical result, and method selection is based upon data
availability. The latter approach is used in this study for all transportation
calculations, since average haul distance was the only data available. This approach
assumes a one-way transport with full cargo. In the case of solids handling, the truck
travels with a full payload on the way there, but returns empty. To account for the
decreased payload usage on the empty return trip, the TEAM LCA Software Users
Manual suggests to use the above approach (e.g., multiply the one-way average haul
distance by the total number of tons transported) and multiply this total by 5/3 (3/3
for full payload on the way there, 2/3 for the return empty trip), to account for the
increased fuel efficiency on the return empty trip (TEAM, 1999).
The PEItransport and EFtransport are derived from NRELs US life cycle inventory (LCI)
database, which are based upon US transportation averages for various transportation
modes (Table 14) (NREL, 2007). The annual consumption data (e.g., average haul
distance & total tons transported) are then applied to Equation (2) and (5) with the
PEI and EF to calculate the life cycle impacts.
In this study, only locomotive and truck transport were used. Benchmarks for the PEI
and EF for locomotive transport are higher than NRELs; however, benchmarked data
include vehicle manufacture and maintenance (i.e., 25% increase), while NRELs
module does not (NNFCC, 2007; NREL, 2007). Details of the calculation of NRELs
Solids transport calculations assume a round trip, presuming the train/truck/etc. travels with full
cargo to the disposal site, and returns to the WWT facility with empty cargo. Since trucks typically
return to the WWT facility after dumping at the disposal site, this is an appropriate assumption.
51


single unit truck module were unavailable, and consequently a European dataset
(United Kingdom) was examined in more detail (Table 15).
UK single truck data show a PEI range of 1.002-4.019 MJ/mt-km (vs. NRELs single
truck = 2.26 MJ/mt-km) shown in Table 15 (NNFCC, 2007). This suggests that
NRELs single truck module uses a weighted average of truck use in the US. This
finding is further supported by the fact that nearly all truck usage in the US is either
less than 9 mt Gross Vehicle Weight (GVW) or between 25-36 mt GVW (e.g., a
weighted average using Table 15 values would result approximately with NRELs
2.26 MJ/mt-km single truck value). However, solids transport typically involves use
of a tractor semi-trailer (e.g., a payload greater than 10 mt/load). MWRDs typical
payload is on the order of 16-18 mt. Consequently, NRELs single truck dataset was
not used, and a surrogate UK dataset was used, using the 26 mt GVW with a payload
of 17.2 mt (Table 15).
Table 14: PEI and EF for Common Transport Modes
PEI EF
PEItransport Benchmark EFtransPort Benchmark
Transport Mode__________________MJ/mt-km___________________kg CCLe/mt-km
Ocean Freighter 0.19 0.140 0.014 0.011
Locomotive 0.25 0.464 0.019 0.032
Barge 0.37 0.538 0.028 0.040
Cargo Plane29 0.79 N/A 0.053 N/A
Combination Truck 1.05 N/A 0.081 N/A
Single Unit Truck 2.26 See below 0.173 See below
29 Cargo Plane database only includes C02-related GHG emissions. Additional GHG releases, such as
N20 and CH4, would likely increase GWP (kg C02e/mt-km) by an additional 1%.
52


Table 15: UK Single Truck PEI and EF by Gross Vehicle Weight
Single Truck Transport Gross Vehicle Weight, mt Payload mt PEItransport MJ/mt-km EFtransport kg C02e/mt-km
3.5 1.8 4.019 0.327
5.5 3 3.122 0.234
7.5 3.2 3.819 0.286
18 11.2 1.376 0.103
26 17.2 1.326 0.0995
32 21 1.263 0.0947
40 25.5 1.002 0.0751
44 28.5 1.032 0.0772
3.7.1.5 Biogas Electricity and Heat Recovery Life Cycle Calculation Techniques
When anaerobic digestion is used in WWT, gas is produced as a by-product of
anaerobic decomposition of sludge. This gas, often termed biogas, has a typical
composition of 40-60% methane, with the remainder being carbon dioxide and traces
of water vapor, hydrogen, and hydrogen sulfide. This gas can be used directly on-site
for heat or electricity production.
In many instances, raw digester gas can be used to create electricity. Since efforts to
scrub digester gas, if any, are aggregated into the total electricity and infrastructure
used on-site at the WWTP, no effort is made to add additional energy used for gas
scrubbing to this LCA (for it would be double counting).
Since biogas is produced from a renewable and biogenic energy source on-site,
biogas primary energy appears in both operations and avoided (negative values) life
cycle categories. In the case of primary energy, the two directly cancel each other
out, as it should since there is no net flow of energy outside the system. This
methodology is employed so that the end-use energy used to treat a gallon of
53


wastewater includes biogas derived energy; consequently, these end-use values can
be directly compared to other WWTPs that do not employ anaerobic digestion.
When biogas is used in Combined Heat & Power (CHP) turbines, both electricity and
waste heat can be recovered for maximum energy use efficiency. For MWRDs CHP
system, the power plant efficiency of biogas to electricity, tiPP-biogas, from 2001-2007
ranged from 22.5 to 25.5%. For illustrative purposes, assuming tiPP.biogas is 25%, the
ideal scenario for heat recovery is shown (Figure 8). Figure 8 shows an ideal scenario
with an overall system efficiency, ijchp System, (electricity + heat recovery) of 80%,
which is the maximum system efficiency for existing CHP systems at WWTPs
(EPAa, 2007).
In many instances, the heat exchanger and distribution losses can be higher. As a
lower bound, an overall system efficiency of 50% is used, as this is typically the
definition used by states in the US to define minimal CHP efficiencies (EPA, 2008).
100 GJ
biogas
25 GJe
Electricity
PEIuj0gas_EiectrjCjty 25 1.0
25
T|CHP System 25 _+ 55 0.8
100
Heat
75 GJ
67.5 GJ
60.75 GJ
Heat Recovered
~55 GJ
Combustion Heat Exchanger Distribution
Losses 10% Losses 10% Losses 10%
PEIuiogas-Heat Rx 25 1.36
55
Figure 8: Illustration of CHP Based on WWTP Combined Heat & Power Energy
Flows: Assuming Tipp-wog., = 0.25 (EPAa, 2007)
54


For PEI calculation, it is assumed that the PElBi0gas-Eiectncity =1.0 and all remaining
heat energy is allocated to the heat recovery primary energy intensity, PElBi0gas-Heat rx-
A heat recovery efficiency of 72.9% is assumed throughout this study, reflecting the
combined losses of combustion, heat exchanger, and distribution shown in Figure 8.
This approach is consistent with WRI Protocol for CHP (WRI, 2006). Further, this
method avoids double counting.
Since emissions associated with the combustion of biogas are biogenic in origin and
thus carbon neutral, the EF for biogas, EFBiogas, is zero (IPCC, 2007).
3.7.1.6 IPCC GHG Emissions
3.7.1.6.1 Methane Emissions
The IPCC methodology for GHG accounting on the national scale incorporates
fugitive methane losses and nitrous oxide releases into GWP calculations. As
mentioned above, BOD emissions are not included, as they are biogenic in origin and
thus carbon neutral (IPCC, 2007). Discourse with WWTP personnel that utilize
anaerobic digestion (Rogowski, 2006; Anderson, 2006), revealed fugitive methane
losses from anaerobic digestion would be negligible. Further, as defined by the
IPCC, the default methane emission factor for a well managed and not overloaded
(demonstrated by the daily flow vs. design capacity) centralized aerobic treatment
plant is zero, as is the case with MWRD, Longmont WWTP, Louisville WWTP, and
the Mobile Park.
For septic-based systems (e.g., AdvanTex and Septic), since lA of the BOD is trapped
in the septic system, it is presumed half of all incoming BOD can be converted to
methane (IPCC, 2007). Using IPCC methodology and assuming BOD removed from
55


the septic tank during every 4-10 yr for septic tank pumping is negligible on a life
cycle basis, this translates into 232.69 kg CC^e/person-yr (IPCC, 2007).
Overall, IPCC methane emissions for MWRD, Longmont WWTP, Louisville
WWTP, and the Mobile Park were assumed negligible, while the IPCC methane
emission factor of 232.69 kg C02e/person-yr were applied to the septic-based systems
in this study (i.e. AdvanTex and Septic).
3.7.1.6.2 N2O Emissions
There are two sources of N2O emissions for WWTPs: 1) nitrogen in the effluent, and
2) nitrification & denitrification processes. In recent reports, the US EPA and IPCC
both acknowledge the large uncertainty in the emission factors for wastewater related
N2O emissions (EPAb, 2007; IPCC, 2007). Large variability exists in IPCC emission
factors, where the low to high uncertainty range is a manageable factor of four
increase for nitrification/denitrification, but nitrogen in the effluent has a factor of
500 increase from low to high uncertainty boundaries. The implications for N2O EFs
for nitrogen in the effluent and nitrification/denitrification are discussed next,
respectively.
N2O from Nitrogen in the Effluent
When nitrogen is released in the effluent of WWTPs and into surrounding surface
water bodies, a portion of this nitrogen forms the very strong GHG N2O (e.g., 298
times more powerful than CO2). For effluent nitrogen N2O, large uncertainty exists
largely because the effluent travels through a wide range of conditions as it makes its
way towards the ocean modeling the range in conditions and interplay of other
water bodies is complex and regionally variable, leading to large uncertainty. The
IPCCs default emission factor is 0.005 kg N2O formed per kg of N released in the
56


effluent, but has a reported range of 0.0005 to 0.25 kg N20-N/kg Neff a range
difference of a factor of 500 (IPCC, 2007). Essentially, IPCCs default EF means
0.5% of effluent nitrogen results in N2O formation, as a nominal emission factor.
Crutzen et al. (2007) shows that 2-4% of the nitrogen in runoff is converted to N2O
from the point of entering the stream forward. Barton and Atwater (2002) reported
studies having a range of estimates with a lower bound of 0.2% and an upper bound
of 2-4% of TN in wastewater effluent discharged into a surface water body, and
ultimately forming N2O. As a result of literature studies, the uncertainty range
reported by the IPCC (e.g., 0.0005 to 0.25 kg N20-N/kg Nefr) was refined to recent
literature findings, having a range of 0.002 to 0.04 kg N20-N/kg Neff. The IPCCs EF
of 0.005 kg N20-N/kg Neff falls within the range and was used, but recent literature
findings suggest that the true EF for US WWTP effluent should be 0.02-0.04 kg N2O-
N/kg Neff if one considers the long-term fate and transport to the ocean and beyond
(Barton and Atwater, 2002; Crutzen et al., 2007). Lastly, the implications of nitrogen
WW releases to groundwater and ultimate N2O formation are not reported in the
literature. Consequently, septic-based systems were conservatively estimated as if
100% of the effluent was discharged to the surface. Thus, all WW systems in this
study were treated identically for N2O formation from nitrogen in the effluent (EF =
0.005 kg N20-N/kg Neff, with an error range of 0.02-0.04 kg N20-N/kg Neff).
N2O Releases from Nitrification and Denitrification
WWTPs also have direct process-related N2O emissions from intentional and
unintentional nitrification and denitrification. Intentional nitrification and
denitrification hereon referred to as Biological Nutrient Removal (BNR) is a WW
process, where ammonia, nitrate, nitrates, and organic nitrogen in the WW are
ultimately converted to N2 gas. Microbes also convert some of the nitrogen in the
WW to the GHG N20.
57


Nitrification and denitrification related N2O emissions are usually very minor in
magnitude when compared to nitrogen effluent releases and are optionally suggested
for IPCC calculations. The IPCC uses an EF of 3.2 g N20/person-year (with a range
of uncertainty of 2-8 g N20/person-year) for advanced WWTPs with BNR; however
this EF is derived from Czipiel et al. (2002) who studied a single centralized WWTP
without intentional nitrification or denitrification in Durham, NH. Further, it appears
that the uncertainty boundaries were based on expert judgment not literature-based
scientific studies (IPCC, 2007). Recent work by the EPA and WRI has looked at the
differences between plants with or without nitrification and denitrification more
critically (CARB et al., 2008; EPA, 2007a). The EPA and WRI Local Government
Operations Protocol use an EF of 7 g N20/person-year for plants with nitrification
and denitrification (from Scheele and Dorn, 2002) and an EF of 3.2 g N20/person-
year for plants without intentional nitrification and denitrification (from Czipiel et al.,
1995, 2002). Interestingly, Scheele and Dorn noted significant diurnal and seasonal
variation in N2O emission rates, which is currently unaddressed (2002).
Recent field and laboratory tests also help to readdress uncertainty boundaries
(Skikowski, 2007; Barton and Atwater, 2002; Park, 2000; Scheele and Dorn, 2002;
Paramsivam, 2008). Barton and Atwater looked at 26 literature papers addressing
N2O emissions from WWTPs. Large variability was found between bench-scale
laboratory data and plant-size, real world data. Bench-scale tests reported N2O
generation as high as 40% of TN in the influent, TNjnf, with the majority of studies
reporting similar rates for nitrification (2.3-16% of influent ammonium-N) and
denitrification (~8% of influent NO3-N) (Barton and Atwater, 2002). Meanwhile,
literature data at existing full-scale WWTPs have shown much lower emissions with a
range of 0.001-0.0025% of TNjnf for CAS WWTPs with no nitrification and
denitrification and 0.01-0.08% for WWTPs with Biological Nutrient Removal (BNR)
(i.e., has nitrification and denitrification) (Table 16). Barton and Atwater (2002)
58


estimate IPCC EFs (i.e., 2-8 g N20/person-year) to represent a range of 0.0016-
0.0063% of TNjnf, which correlates well to literature values for CAS without BNR.
Table 16 also compares the use of IPCC EFs to actual percentages of TNmf for
MWRD. Interestingly, applying IPCC EFs (based on WWTPs with no nitrification
and denitrification) to MWRD results in an order of magnitude increase of %TNjnf,
when compared to literature CAS without BNR (Table 16). In other words, MWRD
N2O emissions (for CAS with no BNR) appear to be overestimated by over an order
of magnitude on a %TNinf basis.
Table 16; IPCC N2O Process EF Uncertainty Range vs. Literature & MWRD
IPCC Low Default IPCC High Units
Uncertainty Value Uncertainty___________
IPCC EF Nit/Denit 2 3.2 8 g N20/per-yr
Calculated IPCC % of TNinf30 0.0016% 0.0025% 0.0063% % Of TNinf
MWRD Using IPCC EFs 0.022% 0.034% 0.086% % Of TNjnf
Literature N/D BNR 0.01% 0.08% % Of TNjnf
Literature CAS No N/D 0.001% 0.0025% % of TNjnf
The EPA, WRI, and IPCC do not provide any indication of how to handle plants with
nitrification but without denitrification (EPAb, 2007; CARB et al., 2008; IPCC,
2006). Park et al. (2000) showed that in a well-operated plant where Dissolved
Oxygen (DO) concentrations are kept above 0.5 mg/L, nitrification-related N2O
emissions were reduced by 95%. Because WWTPs with nitrification analyzed in this
study (e.g., MWRD, Longmont WWTP, and Louisville WWTP) have maintained
DO>0.5 mg/L for all processes, default IPCC EFs for CAS without BNR was used
(3.2 g/person-day). For these three WWTPs that have industrially and/or
commercially discharged proteins, a 1.25 multiplier was used (IPCC, 2007). This all
translates to a default value of 1192 mt CO2e/106 people served-yr, with a range of
745-2980 mt CO2e/106 people served-yr for advanced WWTPs with nitrification and
30 Barton and Atwater, 2002
59


no denitrification that receive some industrial TN load (e.g., MWRD, Longmont
WWTP, and Louisville WWTP).31 Because of variations in literature reporting (e.g.,
based on %TNjnf) with IPCC EFs (e.g., based on per person-yr), the IPCC uncertainty
range was used in uncertainty calculations. However, literature suggests that an EF
of 7 g/person-day for WWTPs with BNR (e.g., 0.0057% of TNinf for IPCC, but
literature estimates 0.01-0.08% of TNjnf) is likely underestimated by over an order of
magnitude (Barton and Atwater, 2002). Nevertheless, all WWTPs in this study did
not have intentional denitrification. MWRD, Longmont WWTP, Louisville WWTP,
and the Mobile Park all used the IPCC EF of 3.2 g N20/person-yr, with MWRD,
Longmont WWTP, and Louisville WWTP using the additional 1.25 industrial
multiplier.
3.7.2 On- & Off-site Embodied Primary Energy
Four embodied energy consumption parameters were collected from the sources
shown (Table 9):
Mass of chemicals used in WWTP operations purchase receipts
$US 1997 expenditures on on-site infrastructure annual expenditures
$US 1997 expenditures OR material use in off-site infrastructure annual
expenditures
$US 1997 expenditures on maintenance capital annual expenditures
EIO LCA is based upon dollar expenditures in the year 1997, thus to use the database,
all expenditures must be converted to $1997.
31 The Mobile Park, AdvanTex, and Septic systems do not include any nitrification or denitrification
and thus the IPCC nitrification/denitrification emissions are zero.
60


Annual totals of on-site capital expenditures are recorded by year and converted to
equivalent 1997 US$ $1997anniiai, using the Consumer Price Index (CPI) for all
urban consumers, 1982-1984 base period (USDOL, 2007). To convert the equivalent
value of $1 spent in a desired year to equivalent $1997, one uses the following
formula:
$1997 =
CPL
1997
CPIExpendituie Year
Spent In Expenditure Year
(13)
For example, the CPI in 2007 was 207.342, while the CPI in 1997 was 160.5. $1
spent in 2007 has the following equivalent value in 1997:
$1997=S^*$1=$0-77 (l4)
Historical annual total expenditures are then converted to $1997 and these values are
summed and normalized over the number of data years, as detailed above.
The PEI and EF estimations for embodied energy are described next.
3.7.2.1 PEI and EF for Chemicals
The life cycle impacts of on-site chemical use are composed of the material and
energy used in manufacturing a chemical (including ultimate raw resource extraction)
and the shipment of the chemical to the WW facility.
The active mass of chemicals (e.g., the manufacturer reported fraction of the gross
mass) used at the WWTP relates to chemical manufacturing, while the gross chemical
mass relates to transportation. For chemical manufacturing, the energy used and
GHG emitted by the inactive fraction of the chemicals is assumed negligible in this
LCA.
61


The PEI and EF used for chemical manufacturing in this study are found in (Table
17). The PEI and EF used for chemical transportation in this study are identical to
those used for solids transport (Table 14 and Table 15).
Table 17: PEI and GWP for Common WWT Chemicals
PEI
PEIchemicaiBenchrnark EF chemical
mt
Chemical Used Database/Module Used GJ/mt C02e/mt
Calcium hypochlorite TEAM, Sodium Chlorate32 2.42 N/A 0.15
Chlorine, gas TEAM, Chlorine Gas33 16.94 19.234 0.99
Ferric chloride SimaPro, Iron Sulfate35 2.50 N/A 0.12
Hydrochloric acid TEAM, Hydrochloric Acid 54.88 N/A 3.29
Nitrogen, gas TEAM, Nitrogen Gas 5.56 N/A 0.33
Oxygen, gas TEAM, Oxygen Gas 5.56 N/A 0.33
Polymer SimaPro, Organic Chemical ETH S36 69.29 N/A 1.90
Sulfur dioxide gas TEAM. Sulfur; SimaPro Sulfur->S0237 52.77 N/A 0.34
3.7.2.2 On- & Off-site Infrastructure and Maintenance EIO LCA
Most LCAs ignore the impacts of infrastructure due to their complexity in accurately
accounting for all materials used for a typical mass-LCA. As explained above, a
hybrid-LCA is being performed to incorporate the impacts of infrastructure using
EIO-LCA.
32 Sodium chlorate used as a surrogate for calcium hypochlorite
33 Values listed include allocation by mass fraction for co-products. Cl2 is 46.4% of total mass of co-
products created. Thus, original total primary energy of 36.5 GJ/mt all products created, is multiplied
by 46.4% to obtain 16.94 GJ/mt chlorine created.
34 Worrell et al., 2000
35 Iron sulfate used as a surrogate for iron chloride
36 A general organic chemical module was used as a surrogate for proprietary polymer production
37 A combined module was created, by linking a TEAM sulfur resource extraction module with a
SimaPro sulfur dioxide production (from sulfur) module. Combined module includes raw sulfur
resource extraction through sulfur dioxide production.
62


EIO-LCA is based upon monetary expenditures in NAICS (North American Industry
Classification System) sectors of the US economy, which has replaced the older SIC
(Standard Industrial Classification) codes identification. Over 500 sectors of the US
economy, identified by NAICS sectors, are tied into environmental release industries
used to model the environmental impacts of expenditures in each economy sector.
The PEI and EF for capital expenditures are derived from the EIO-LCA database
(Table 18) (CMUGDI, 2008). Please note there is no possible way to benchmark
EIO-LCA PEIs and EFs, since they are, by definition, already benchmarked to the US
economy.
Table 18: EIO-LCA Infrastructure PEIs and EFs
Infrastructure Element Evaluated NAICS Sector P ^Infrastructure GJ/$1000 US 1997 EF Infrastructure mtC02e/$1000 US 1997
On-site 230210: Construction: Manufacturing and Industrial 7.63 0.588
Off-site 230240: Water, sewer, pipeline construction 11.5 0.899
On-site Maintenance 230340: Other maintenance and repair construction 10.6 0.821
Off-site Maintenance 230340: Other maintenance and repair construction 10.6 0.821
When off-site infrastructure expenditures are available, the above approach is used to
calculate life cycle impacts from the collection system; however, when unavailable,
the impacts of off-site infrastructure are approximated by materials accounting of the
collection system piping. In the special case of MWRD, both a mass-LCA and EIO-
LCA approach were taken to compare the range in the two calculation methods.
63


3.7.2.3 Off-site Infrastructure Mass LCA
Off-site infrastructure includes all capital used in the transmission system from all
customer house/business effluent to the WWTP, including, but not limited to,
collection piping, pumps/dosing systems, lift stations, meters.
The mass-LCA approach for off-site infrastructure only includes embodied piping
materials for the collection system. It does not include lift station infrastructure, nor
does it include transport or excavation energy used to install the pipes. Thus, the
mass-LCA approach to off-site infrastructure underestimates life cycle impacts. Pipe
costs can range from 4%-50% of total pipe installation costs, where larger pipes at
shallower depths represent the higher end (Lamb, 1995). Consequently, mass-LCA
off-site infrastructure results could be underestimated by a factor of 2 to 25.
To calculate life cycle impacts, first materials accounting of the piping infrastructure
installed in the collection system must be performed. Off-site infrastructure
consumption data include the lengths, diameters, and pipe type for all piping used in
the entire collection system. The total mass of off-site pipe for each material is
calculated using Table 19.
64


Table 19: Linear Mass for Given Pipe Diameter and Material38
Nominal Size inches PVCP38 39 VCP40 Linear NRCP41 Mass, Ib/I RCP42 near ft DIP43 PETP44 HDPEP45
6 3.7 21 17 (15) 18 (3.7) 4.97
8 5.6 30 27 (37) 24 (5.6) 8.43
10 8.0 44 37 (60) 30 (7.9) 13.09
12 10.5 54 50 90 39 (10.5) 18.41
14 12.5 (87) (69) (111) 47 (12.4) 22.2
15 (14.7) 90 80 120 (51) (14.6) (26.0)
16 16.3 (110) (90) (138) 57 (16.2) 28.99
18 20.6 141 110 160 66 (20.4) 36.7
20 24.2 (164) (140) (196) 78 (24.0) 45.3
21 (26.5) 168 160 210 (78) (26.3) (50)
24 33.7 230 200 270 93 (33.4) 65.24
27 (41.8) 290 390 310 (110) (41.5) (82)
30 (50.8) 333 450 360 123 (50.4) 101.9
33 (60.7) 415 520 (420) (147) (60.2) (123.3)
36 (71.5) 445 580 (479) 163 (71.0) 146.8
39 (83.2) 540 (645) (541) (188) (82.6) (172.5)
42 (95.8) 603 (709) (606) 206 (95.1) (200.2)
48 (123.6) (773) (837) (745) 261 (122.7) (261.9)
54 (155.1) (956) (965) (895) 325 (153.9) (332.0)
60 (190.1) (1158) (1093) (1056) 371 (188.7) (410.5)
64 (215.5) (1303) (1178) (1170) 410 (213.9) (467.5)
70 (256.5) (1535) (1306) (1351) (480) (254.6) (559.9)
78 (316.7) (1874) (1477) (1609) (577) (314.4) (696.2)
38 Values in parentheses are interpolated
39 PVC = Polyvinyl Chloride. Based on SCH 40 PVC pipe (Harvel. 2007).
40 VCP = Vitrified Clay Pipe. Based on Gladding Mcbean VCP, Standard Bell & Spigot (Gladding
Mcbean, 2007).
41 NRCP = Non-reinforced Concrete Pipe. Based on Class 1 NRCP, Bell and Spigot Joint, 150 lb
concrete/ft3 (ACPA, 2007).
42 RCP = Reinforced Concrete Pipe. Based on Wall A RCP, Bell and Spigot Joint, 150 lb concrete/ft3
(ACPA, 2007).
43 DIP = Ductile Iron Pipe. Based on minimum pressure class DIP and standard cement-mortar lining
(DIPRA, 2006).
44 PETP = Polyethylene Terephthalate Pipe, made with Cast in Place technology. No standard
reference found. Wall thickness assumed identical to PVC. Linear weight adjusted by relative density
compared to PVC pipe. Density of PET = 1370 kg/m3. Density of PVC = 1380 kg/m3. (Wikipedia,
2007)
45 HDPEP = High Density Polyethylene Pipe. Based on Class 160 DR11, PolyPipe PW Pipe.
(PolyPipe, 2007).
65


The PEIs and EFs reported in literature are found in Table 20.
Table 20: PEIs and EFs for Various Pipe Types
PEI EF
Pipe PEI Used US Lit. Range World Lit. Range EF Used World Lit. Range
Material GJ/mt GJ/mt GJ/mt mt C02e/mt mt C02e/mt
PVC 67.846 60.5-79.147 50.3-79.148 2.2038 2.20-4.8649
VCP 7.0350 6.02-7.0351 2.92-7.0352 0.414i 0.25-0.4153
NRCP 1.8354 0.95-6.2955 0.52-6.2956 0.2346 0.17-0.2357 58
RCP 3.4238 2.85-3.4259 N/A 0.3350 0.25-0.3351
DIP 19.5542 N/A N/A 1.4345 N/A
Cast Iron N/A 34.6-50.760 N/A N/A N/A
Steel 5LP N/A 35.3-51.161 3.2239 2.75-3.22i3
HDPE 89.339 98.1662 89.3-98.1663 2.5035 N/A
NOTES: Based upon 8" diameter pipe, converted to mass based on standard linear
mass by pipe type (Rosenkranz, 1979).
3.7.3 Avoided Energy
Four avoided energy consumption parameters were collected from the sources shown
(Table 9):
46 TEAM, 2006
47 Friedrich et al., 2007; Rosenkranz, 1979; SPI, 1994
48 Friedrich et al., 2007; Rosenkranz, 1979; SPI, 1994; Claus, 1995; TEAM, 2006; Hammond and
Jones, 2008
49 Friedrich et al., 2007; Claus, 1995; TEAM, 2006; Hammond and Jones, 2008
50 Friedrich et al., 2007
51 Friedrich et al., 2007; Rosenkranz, 1979; SPI, 1994
52 Friedrich et al., 2007; Rosenkranz, 1979; SPI, 1994; Claus, 1995
53 Friedrich et al., 2007; Claus, 1995
54 Reiner, 2007
55 Friedrich et al., 2007; Rosenkranz, 1979; Reiner, 2007; BG, 2007
56 Friedrich et al., 2007; Rosenkranz, 1979; Reiner, 2007; BG, 2007; TEAM, 2006
57 Reiner, 2007; TEAM, 2006
58 Calculated based on steel requirement of 5 lb steel/ft3 for reinforced concrete (NIST, 2007),
amounting to 3.23% of RCP is steel, by mass. Uses Reiner (2007) for concrete PEI and EF. Uses
TEAM (2006) for steel PEI and EF.
59 Calculated based on PEI and EF for concrete from Reiner (2007) and Friedrich et al. (2007). Uses
TEAM (2006) for steel PEI and EF, assuming 3.23% of RCP is steel, by mass.
60 Rosenkranz, 1979; SPI, 1994
61 TEAM, 2006; Hammond and Jones, 2008
62 SPI, 1994
63 TEAM, 2006; SPI, 1994
66


Avoided Biogas Electricity Plant meters
Avoided Biogas Heat Recovery Estimated based on known heat exchanger
efficiency
Avoided Fertilizer Biosolids annual reports
Avoided Water Annual wastewater characteristics report
These consumption parameters are combined with the PEI and EF for avoided energy
to calculate life cycle impacts. The PEIs and EFs for avoided energy are discussed
next.
3.7.3.1 Avoided Biogas Electricity and Heat Recovery
Biogas electricity production and heat recovery energy is internal to the overall LCA
system boundary. As a result, the avoided energy benefits are equivalent to the
benefits used in End-use Energy. PEIs used in the End-use Energy section are
identical to PEIs used for avoided energy, only the PEI for avoided energy has a
negative value. Thus, the End-use Energy for biogas exactly cancels out the Avoided
Energy for biogas.
There currently is no methodology universally adopted within the US or
internationally for GHG allocation in CHP systems. However, an approach called
output-based emissions with double benchmarking, adopted by the IEA, EPA, and
several US states, provides incentives for CHP, especially CHP derived from a
biogenic fuel (EPAa, 2007; IEA, 2008). Since typical practice is to purchase
electricity and natural gas (for heating) separately, this approach gives credit for the
displaced emissions coming from separate generators. As an example, on-site biogas
electricity production at MWRD displaces the use of grid electricity from Xcel
Energy. Thus, the EF for biogas electricity is simply the negative value of Xcel
electricity production (shown in the End-use Energy section). Similarly, biogas heat
67


recovery displaces the purchase of natural gas from Tri-Gen, thus the EF is the
negative value of Tri-Gen natural gas production and combustion (shown in the End-
use Energy section).
3.7.3.2 Avoided Fertilizer
When biosolids are beneficially reused for compost or direct land application, less
fertilizer is required to be amended to the soil to meet appropriate crop-specific
agronomic rates. Thus, application of biosolids avoids fertilizer production. As a
result, avoided fertilizer benefits serve as a credit towards life cycle primary energy
and GWP calculations. Reviews of US fertilizer use and literature reported PEIs and
EFs were performed to more accurately characterize PEIs and EFs used for avoided
fertilizer calculation. Additional uncertainty exists when comparing the nutrient
fertilizer replacement value of biosolids, as well as nitrogen run-off and ultimate N2O
formation.
Review of US Fertilizer Use and Literature Reported PEIs and EFs
In 2006, single nutrient fertilizers were the majority (62.0%) of Nitrogen (N),
Phosphorus (P), and Potassium (K) fertilizer consumption in the US (i.e. N, P, and K
fertilizers were purchased separately) (USDA, 2007). US nitrogen fertilizer use is
21.7% urea, followed by ammonia (17.1%), ammonium (8.8%), with other nitrogen
inputs such as nitrogen solution and sodium nitrate make up the majority (52.4%).
US phosphorus fertilizer use is predominantly in the form of nitrogen phosphates
(diammonium phosphate = 36.4%, other nitrogen phosphates = 59.7%).
Superphosphates represent only 2.2% of US phosphorous fertilizer use. Potassium
chloride represents 85.9% of potassium fertilizer use in the US in 2006 (USDA,
2007).
68


Predominantly European literature reported EFs for nitrogen, phosphorus, and
potassium fertilizers are shown in Figure 9, Figure 10, and Figure 11, respectively.
The EFs shown are based on single nutrient fertilizers, except for diammonium
phosphate (DAP) and monoammonium phosphate (MAP). Due to the predominance
of nitrogen phosphates in phosphorus fertilizer use in the US, mixed nitrogen
phosphate fertilizers were included in the phosphorus EF literature review, allocated
between nitrogen and phosphorus proportionally by mass composition. Clearly, the
common use of superphosphates PEIs and EFs in LCA literature is inappropriate for
the US (e.g., only 2.2% of total phosphorus).
10
8
Ammonia Ammonium Urea Total NAvg
Figure 9: Nitrogen EF Literature Review 64
64 Wood and Cowie, 2004; TEAM, 2006; NNFCC, 2007; Worrell et al., 2000; Energetics, 2000; EPAb,
2007
69


6
5
E
DAP MAP Superphosphates Total P Avg
I-------------------------------------------------------------------
Figure 10: Phosphorus EF Literature Review
[MAP = monoammonium phosphate, DAP = diammonium phosphate]
3.00
2.50
1.50
o
o
1.00
E
0.50
0.00
Potash
Figure 11: Potassium EF Literature Review56
Due to the complexity of matching limited literature reported EFs to an incomplete
listing of US fertilizer use by type, avoided fertilizer EFs used in this study were
calculated by using a weighted averaged of the EFs shown in the figures above (Table
21). For unknown other categories, the average emission factors (e.g., N fertilizer
average, P fertilizer average) were used. Given the limited dataset for PEIs for
fertilizer, an average of all literature reported values was used (i.e., a weighted
70


average could not be performed). The resultant PEIs and EFs for fertilizer use are
shown in Table 22. The number of references and the standard deviations are
reported for each fertilizer type below. As can be seen, a large standard deviation
exists. Improved standardization of literature studies and collection of US life cycle
inventory data are needed to decrease the uncertainty in avoided fertilizer life cycle
quantification.
Table 21: Weighted Average of Fertilizer Use for PEI & EF Calculations
Fertilizer Nutrient Fertilizer Type Mass % of US Fertilizer Use, by Nutrient Type (e.g., N, P, K) Fertilizer PEI & EF Used
Nitrogen Urea 21.71% Urea
Ammonia 17.06% Ammonia
Ammonium 8.82% Ammonium
Other 52.40% Total N Avg
Phosphorus DAP 36.38% DAP
Other nitrogen phosphates 59.74% MAP
Superphosphates 2.22% SSP, TSP
Other 1.66% Total P Avg
Potassium KC1 85.93% Potash
Other 14.07% Potash
Table 22: Avoided Fertilizer PEIs and EFs
Fertilizer Nutrient Num. of PEI References, n PEI GJ/mt-N,P,K Num. of EF References, n EF mt C02e/mt-N,P,K
N, Nitrogen 6 -63.7 +/-12.3 22 -4.57 +/-1.57
P, Phosphorus 4 -67 +/-71.0 14 -1.25 +/-0.83
K, Potassium 4 -14.5 +/-14.6 2 -1.29 +/-1.17
Nutrient Fertilizer Replacement Value
Further uncertainty exists when comparing nutrient availability of biosolids with
chemical fertilizers across time scales. On an annual basis, farmers must additionally
apply chemical fertilizers to meet nutrient demands not sufficiently met by the
71


biosolids alone. In the US, biosolids application rates are typically based on N
demands of the crop in use. This application rate typically incorporates a Nutrient
Fertilizer Replacement Value (NFRV). A nutrient fertilizer replacement value
(NFRV) is defined as the kg/ha of N, P, or K fertilizer needed to have the equivalent
plant yield as 100 kg-N,P,K/ha in the biosolids (Petersen, 2003). Petersen (2003)
showed the nitrogen NFRV of biosolids is 49-68%, when compared to chemical
fertilizers. Coker and Carlton-Smith (1986) showed the phosphorus NFRV of
biosolids is 60%, when compared to superphosphate. Elliott and OConnor (2007)
suggest that phosphorus NFRV varies significantly due to treatment technologies
used: 99% NFRV for biological nutrient removal biosolids, 49% NFRV for most
digester biosolids, and only 5% NFRV for Fe or A1 heat dried and/or alkaline
stabilized biosolids.
On the contrary, it can be argued that as biosolids decompose further, ultimately (i.e.,
several years+) all of the nutrients contained in the biosolids will be available for
plant growth. Further, biosolids can be viewed as a slow release fertilizer, where the
increased organic matter of biosolids reduces nutrient erosion, and thus an increased
amount of the nutrients are ultimately available to plants, when compared to near
organic matter free chemical fertilizers.
Since it can be argued that all of the nutrients in the biosolids are ultimately available
to the plants, this study does not look at nutrient availability when calculating avoided
fertilizer benefits. If the slow release fertilizer benefits of biosolids can be calculated
annually, farmers could reduce fertilizer application rates in subsequent years (i.e.,
years 2+) and thus reduce fertilizer application rates, enabling the total nutrient value
of biosolids to be realized. Additional studies are needed to address this uncertainty
for WWTP LCA calculations.
72


Nitrogen Runoff and N20-related Emissions
Further uncertainty exists when comparing the runoff of nitrogen and ultimate N2O
formation for biosolids versus chemical fertilizers. The IPCC uses a value of mass
percent-N of 1 % for the amount of nitrogen fertilizer that is converted to N2O during
runoff on the agricultural field. Crutzen et al. (2007) shows that when runoff is
extended beyond the agricultural field and into adjacent water bodies, this figure
increases to 3-5%.
Elliott and OConnor (2007) postulate that the added organic matter in biosolids
reduces nitrogen runoff, when compared to chemical fertilizers. Thus, when
compared to direct fertilizer application, nitrogen runoff for biosolids should be less
than that of chemical fertilizers. Therefore, nitrogen runoff and N20-related
emissions are not included in this study, since, if anything, biosolids would have an
additional benefit when compared to chemical fertilizer. Additional studies are
needed to address biosolids nitrogen runoff and to confirm that biosolids are indeed
superior at reducing N20-related nitrogen runoff emissions.
3.7.3.3 Avoided Water
WWTPs can reuse water in two major ways:
Direct water reuse (no additional treatment beyond the WWTP) WW
effluent directly applied at a golf course, parks, etc. for irrigation
Treated water reuse WW effluent to water recycling treatment plant to
industry, etc.
In both instances, application of avoided water benefits is site specific, depending
upon what water source was displaced:
1. Raw, untreated surface water
2. Treated surface water (i.e., Water Treatment Plant = WTP)
73


3. Untreated groundwater
4. Treated groundwater
Based on literature reported electricity use only, energy usage for various types of
water, as no LCA-based study was found for the various types of water supply.
Operational energy use reports lower energy use and associated GHG emissions for
untreated groundwater (0.605 kWh/1000 gal), when compared to surface water
treatment (1.4 kWh/1000 gal treated) (Arpke and Hutzler, 2006).
Consequently, the additional energy inputs after leaving the WWTP (e.g., pumping,
embodied infrastructure, water reuse treatment plant, etc.) must be compared to
energy inputs that would have gone into the displaced water source (e.g., raw surface
water, groundwater, treated groundwater, and treated surface water).
3.7.3.3.1 Direct Water Reuse
Direct water reuse simply involves transporting WW effluent, without further
treatment, to its reuse location, such as a municipal park, golf course, etc. Additional
pumping energy and embodied infrastructure for the transmission system termed
water reuse delivery primary energy goes into providing direct water reuse. As
discussed above, the volume of wastewater reused for beneficial purposes displaces
either: 1) untreated surface water, 2) untreated groundwater, 3) treated groundwater,
or 4) treated surface water. Ideally, the net benefit of water reuse delivery primary
energy minus the displaced water source primary energy would be calculated to
determine avoided water use benefits.65
65 The WWTP may or may not own the operation (e.g., pumping and transmission system impacts) of
bringing the water to the reuse location and thus may not be able to claim the benefit.
74


Louisville WWTP was the only plant in this study that had direct water reuse, but the
reuse water displaced untreated surface water. Further, reuse distribution
infrastructure impacts amounted to less than 0.3% of Louisville WWTP LCPEI and
less than 0.14% of LCGWPI. Consequently, it is estimated that the pumping and
infrastructure embodied energy to provide the reuse water from Louisville WWTP is
similar or larger in magnitude than the pumping and embodied infrastructure energy
to bring untreated surface water to the reuse location. As a result, water reuse has no
net benefit for Louisville WWTP in terms of avoided energy and GHG emissions.
Extending the LCA boundary further could address whether the untreated surface
water had embodied energy to get it into the existing drainage basin. For example, in
the Denver metro area, water is brought from the other side of the continental divide
(i.e., outside Denver metro drainage basin) and extensive pumping and infrastructure
embodied energy goes into bringing the water to the Denver region. Consequently, it
is possible that untreated surface water does have a reasonable embodied energy
impact. Further studies are recommended to address extending the system boundaries
beyond the WTP and incorporate water acquisition energy.
The use of avoided impacts should only occur when there is certainty to the
application. Unless the site-specific variations described above are known, avoided
water impacts should not be calculated. Literature WTP data provides only
operational electricity energy use, thus ignoring natural gas, solids transport, and
infrastructure impacts (Table 23). Further, values shown in Table 23 are not
representative of US averages, as only a handful of data points were included.
Although not applicable to the WWTPs in this study, if direct water reuse displaced
conventional surface water treatment, the net benefits would be 16.1 GJ/AMG and
75


1.27 mt C02e/AMG (Table 23). Additional studies are needed to add other LCA-
based inputs into the electricity only water supply PEIs and EFs shown in Table 23.
Scope III credit for reuse of wastewater depends upon site specific factors such as the
water supply displaced and the energy invested in the reuse operations. For example,
as shown below, water treatment energy costs can vary from zero (for gravity fed
untreated surface water) to 9.0 GJe/AMG for conventional water treatment plants. No
energy credit would apply as in the case of Louisville when untreated water is
displaced by reused waste water (however a water credit may apply). If direct water
reuse displaced conventional surface water treatment, the net end-use energy benefits
would be as much 9.0 GJe/AMG. This establishes the range of credit that may be
given for wastewater reuse. When wastewater is treated prior to reuse, additional
LCA is needed, which is described next.
Table 23: Water Supply Energy Use by Type (Arpke and Hutzler, 2006)
Avoided Water Source
Gravity fed, untreated
surface water
Untreated groundwater66
Disinfected groundwater
Conventional WTP
Average EUE Use [Range]
kWeh/1000 gal (GJe/AMG)
0(0)
0.605 (2.18)
1.82 (6.55)
1.4 [0.42-2.5] (5.04)[1.5-9.0]
3.7.3.3.2 Treated Water Reuse
Treated water reuse involves the WW effluent being sent to a water recycling plant,
where treatment energy inputs are used to bring the WW typically to drinking water
standards. In this study, MWRD was the only plant whose WW effluent was sent to a
Water Reuse Plant (WRP). Typically, the WW plant does not own the water
66 Assumes 150 ft average well depth (Arpke and Hutzler, 2006). Actual value will vary according to
well depth.
76


recycling collection, treatment, and transmission nor control the operations therein,
and consequently does not include avoided impacts within its avoided water LCA.
For example, MWRD does not include avoided water reuse for the following reasons:
1) Metro is required by law to send all reused water to the water recycling plant and
consequently, Denver Waters Recycling Plant was not a cost-effective build nor are
its relative life cycle impacts anticipated to be similar to industry averages, 2) Limited
data suggests the impacts of a water recycling plant are on the same order as raw
water treatment plants, and in certain instances, the water reuse plant can use more
energy than the water treatment plant, and 3) Detailed below, Denver Waters
Recycling Plant (DWRP) actually consumes more end-use energy (EUEI = 55.5
GJ/AMG) than typical (raw) water treatment plants (e.g., 2.01 to 5.04 GJ/AMG).
Because avoided water and water reuse is truly a site-specific analysis, a streamlined
LCA was performed to calculate the impacts of DWRP in 2008. With the ultimate
goal of comparing ultimately to compare the relative impacts of Water Treatment
Plants (WTPs) and Water Reuse Plants (WRPs). Electricity, natural gas, vehicle
fleet, and sludge handling used mass-energy LCA, while chemicals, maintenance, and
on- & off-site infrastructure used EIO LCA. Infrastructure is normalized over a 40 yr
life expectancy. Results are shown in Table 24 and Figure 12. DWRP is only
treating 5.5 MGD on average, representing less than 20% of its current 30 MGD
capacity, while being expandable to 45 MGD capacity. The transmission system is
not yet complete and consequently results are shifted on a per gallon treated basis. It
is anticipated that these results could be reduced to nearly a factor of 20 once the
transmission system is complete. Future work should update this study once the
transmission is complete and the recycle plant is functioning closer to design
capacity.
77


Table 24: DVVRP Streamlined ALCPE and ALGWP Summary
Data Year Energy and MFA Input Annualized Expenditure PEI GJ/$1997 EF mt C02e/$1997 ALCPE GJ/yr ALGWP mt C02e/yr
2008 Chemicals $535,000 $1997/yr 0.0361 0.00212 1,4397 846
2008 Maintenance $240,000 $1997/yr 0.0106 0.000821 1,896 147
2008 Solids Handling $200,000 $1997/yr 0.0168 0.00212 2,505 316
2004 On-site Infrastructure $1,525,000 $1997/yr 0.00763 0.000588 9,886 762
2004 Off-site Infrastructure $2,325,000 $1997/yr 0.0115 0.000899 22,718 1,776
GJ/MFA mt C02e/MFA
Annual Use units units GJ/yr mt C02e/yr
2008 Electricity 1,092,000 GJ/yr 3.197 0.252 164,922 13,011
2008 Natural Gas 409,500 GJe/yr 1.220 0.068 72,188 4,003
2008 Vehicle Fleet 2,090 gal/yr 0.131 0.010 274 21
2008 Sludge Handling 2,190 mt-km/yr 0.00133 0.00010 3 0
LCCqwrp = $4,090,000 $1997/yr EUEI 55
LCPEIdwrp = 143 GJ/AMG Operating energy only totals 237,387 17,036
LCGWPIdwrr = 10.3 mt C02e/AMG Annual Totals 288,513 20,861
Figure 12: Denver Water Recycling Plant ALCGWPI Summary
78


Comparing these results to other regional Front Range WTPs and Water Reuse Plants
(WRPs) provides a good indication of relative life cycle impacts. Specifically, Table
25 compares Broomfield WTP and WRP, as well as DWRP, Denver Waters
combined WTPs, and Fort Collins WTP. Broomfield WTP and WRP, as well as
DWRP were all compared using the same PEIs and EFs. Fort Collins WTPs used
other PEIs and EFs, particularly the Platter River Power Authority electricity PEI and
EF, in accordance with Local Government Operations Protocol Scope I and II
boundaries. Because of the use of different PEIs and EFs, comparisons are limited to
relative comparisons, and absolute conclusions cannot be made.
Table 25 shows that the relative impacts of WTPs and WRPs are relatively similar,
and in some cases WRPs have a larger relative impact than WTPs (e.g., DWRP).
Consequently, life cycle benefits for WRPs are considered a wash when compared
to WTPs. Thus, if a WWTP sends its WW effluent to a WRP, this has a similar
impact to WTPs and consequently, no benefits are given to WRPs in this study.
79


Table 25; EUEIs, PEIs, and EFs for Regional WTPs and WRPs
Water Treatment Treatment Volume EUEI PEIwtp EFwtp LCA Boundaries
Plants MGD GJ/AMG GJ/AMG mt C02e/AMG
Denver Water 187.6 2.01 n/a n/a Direct electricity use only
Fort Collins67 25.0 n/a n/a 0.47 WRI Scope I&II
Broomfield68 6.0 2.53 113 6.52 Footprint using PEIs and EFs from this study
Literature69 n/a 5.04 n/a n/a Direct electricity use only
Water Reuse Plants Treatment Volume (Capacity) MGD (MGD) EUEI GJ/AMG PEIwrp GJ/AMG EFwrp mt C02e/AMG LCA Boundaries
Denver Water 5.5 (30) 55.5 143 10.3 Direct electricity use only
Broomfield70 2.2 2.44 29.4 2.18 Footprint using PEIs and EFs from this study
Literature71 26.1 8.7 64.4 3.87 Footprint using external PEIs and EFs
Consequently, the inclusion of Denver Waters Recycling Plant would actually
increase the impacts of MWRD under the Avoided Water category. Since MWRD
is required to provide the wastewater effluent for recycle by law, but does not own the
operation of transmitting it to DWRP, the impacts of DWRP are not included in this
analysis. As the comparison between regional and literature WTPs and WRPs PEIs
and EFs, WRPs tend to have similar life cycle impacts as WTPs. Comparisons are
extremely site specific, and as the above analysis of DWRP showed, MWRD would
realize a net increase in life cycle impacts if DWRP impacts were included.
To be conservative, at this time, no credit is given for wastewater reuse from an
energy and GHG standpoint.
67 ClimateWise, 2009
68 Knight et al., 2009
69 Arpke and Hutzler, 2006
70 Knight et al., 2009
71 Stokes and Horvath, 2009
80


3.7.4 BOD GHG Emissions
For illustrative purposes only, the biological oxygen demand (BOD) GHG emissions
were calculated. However, it should be noted that since BOD-emissions are biogenic
in origin, BOD emissions do not have a net impact on GWP and thus are carbon
neutral (IPCC, 2007).
BOD emissions were approximated using the generalized equation for biomass
growth of domestic wastewater, utilizing half-reactions (Grady et al., 1999):
R = Rd-fe*Ra-fs*Rc
where: R = overall stoichiometric equation
Rd = half-reaction for electron donor
Ra = half-reaction for electron acceptor
Rc = half reaction for cell material
fe = fraction of the electron donor used for energy
fs = fraction of the electron donor used for cell synthesis
And:
fs + fe = 1.0
For aerobic growth of heterotrophs using domestic wastewater as a substrate and
ammonia nitrogen as the nitrogen source, fs = true growth yield = 0.71 mg biomass
COD formed/mg substrate COD used (Grady et ah, 1999).
The summation of each half reaction is shown below, yields our final balanced
equation72:
Rd
0.02 C10H19O3N + 0.36 H20 = 0.18 C02 +0.02 NH4+ +0.02 HCOf +1 H+ +1 e
-fe*Ra
72 C10H19O3N and CsH702N used for the composition of domestic wastewater and microbial biomass,
respectively (Grady et al., 1999).
81


-0.29 [0.5 H20 = 0.25 02 + 1 H+ + 1 e-]
-fs*Rc
-0.71 [0.05 C5H702N + 0.45 H20 = 0.2 C02 + 0.05 HC03' + 0.05 NH4+ + 1 H++ 1 e]
On a molar basis, the balanced stoichiometric equation for aerobic heterotrophic
growth on domestic wastewater with ammonia as a nitrogen source is:
C10H19O3N+ 3.63 02 + 0.78 HC03"+ 0.78 NH4+ = 1.78C5H702N + 5.23 H20 + 1.90 C02
On a mass basis, the balanced stoichiometric equation for aerobic heterotrophic
growth on domestic wastewater with ammonia as a nitrogen source is (NOTE: On a
mass basis, the charges are no longer balanced, but the sum of the stoichiometric
coefficients for the reactants equals the sum of the products):
C10H19O3N + 0.58 02 + 0.24 HC03' + 0.09 NH4+ = 1.00 C5H702N + 0.47 H20 + 0.42 C02
For domestic wastewater, the biodegradable fraction of the chemical oxygen demand,
CODbo, can be related to the 5-day BOD, BOD5, as (Grady et al., 1999):
CODbo ~ 1.71 BOD5
Using the above balanced equation, carbon dioxide emissions are calculated.
3.7.5 Summary of Intensity Parameters and Benchmarks
The summary of the PEIs and EFs used in this study, as well as literature benchmarks
are shown in Table 26 and Table 27, respectively.
82


Table 26; Summary of Primary Energy Intensity Parameters and Benchmarks
Sector or End-Use p£| PEI Data Source PEI used PEI Benchmark Units
End-use Energy
On-site Electricity Kim and Dale, 2005 3.20 3.28 GJth/GJe
Off-site Electricity Kim and Dale, 2005 3.20 3.28 GJth/GJe
On-site Natural Gas NREL, 2007 1.22 1.08-1,2273 GJin/GJout
Off-site Natural Gas NREL, 2007 1.22 1.08-1.22 GJin/GJout
Solids Transport NREL, 2007 0.00133 See discussion above GJ/mt-km
Digester Electricity NREL, 2007 1 N/A GJth/GJe
Digester Heat Rx WRI, 2006 1.43 N/A GJin/GJout
Embodied Energy
Chemicals
Chemicals (Material) TEAM; SimaPro, 2006 31.2 See discussion above GJ/mt
Chemicals (Transport) NREL, 2007 0.00052 See discussion above GJ/mt-km
On-site Infrastructure CMUDGI, 2008 7.63 N/A GJ/$1000
Off-site Infrastructure Multiple -->Aggregate 5.2 See discussion above GJ/mt pipe
Maintenance CMUDGI, 2008 10.60 N/A GJ/$1000
Avoided Energy
Digester Electricity Kim and Dale, 2005 -1 3.28 GJth/GJe
Digester Heat Rx NREL, 2007 -1.43 N/A GJin/GJout
Avoided Fertilizer
N TEAM, 2006 -63.7 -40.7 to -76.6 GJ/mt-N
P TEAM, 2006 -66.7 -12.8 to-163.5 GJ/mt-P
K TEAM, 2006 -14.5 -4.6 to -35.7 GJ/mt-K
Avoided Water Lundie et al., 2005 -16.1 -4.83 to-19.2 GJ/MG
73 Literature reports 1.08 GJin/GJout without fugitive losses (TEAM, 2006), while 1.22 GJin/GJout
includes fugitive losses (Spath and Mann, 2000).
83