United States Environmental Protection Agency
Office of Atmospheric Programs (6207J)
Washington, DC 20005
EPA-430-R-13-011
September 2013
Global Mitigation of Non-CO2
Greenhouse Gases: 2010-2030
-------
How to obtain copies
You can electronically download this document from the U.S. EPA's Web site at
.
For further information
The results presented in this report can be downloaded in spreadsheet format from the U.S. EPA's Web
site at . For
additional information, contact Shaun Ragnauth, (202) 343-9142, ragnauth.shaun@epa.gov, U.S.
Environmental Protection Agency.
Peer-reviewed document
This report has undergone an external peer review consistent with the guidelines of the U.S. EPA Peer
Review Policy. Comments were received from experts in the private sector, academia, nongovernmental
organizations, and other government agencies. See the Acknowledgements section for a list of
reviewers. A copy of the EPA Peer Review guidelines can be downloaded from the following web page at
.
-------
Acknowledgements
This report was prepared under a contract between the U.S. Environmental Protection
Agency (USEPA) and RTI International. Shaun Ragnauth edited and directed completion
of the report. Lead authors include Jeffry Petrusa of RTI, Shaun Ragnauth of USEPA,
Jameel Alsalam of USEPA, David Godwin of USEPA, Jared Creason of USEPA, Jia Li of
USEPA, and Robert Beach of RTI. We thank USEPA reviewers Reid Harvey, Allison Costa,
Kurt Roos, Tom Wirth, Jefferson Cole, Carey Bylin, Suzanne Waltzer, Tom Frankiewicz,
Chris Godlove, Sally Rand, Kirsten Cappel, Jayne Somers, Melissa Weitz, and Elisa Rim.
Other significant contributors and co-authors include Sara Ohrel of USEPA, William Salas
of Applied GeoSolutions, Changsheng Li of University of New Hampshire, Pete Ingraham
of Applied GeoSolutions, Stephen Ogle of Colorado State University, and
Benjamin Henderson of International Food Policy Research Institute.
The staff at RTI assisted in compiling and finalizing the report. The staff at RTI and
IGF Consulting prepared many of the individual analyses. Special recognition goes to
Jeffrey Petrusa at RTI and Marian Van Pelt at IGF Consulting for their significant
contributions on this project.
We also thank the following external reviewers: Stuart Day (CSIRO Energy Technology),
Clark Talkington (Advanced Resources International), Jon Kelefant (Advanced Resources
International), Mike Godec (Advanced Resources International), Ronald Collings (Ruby
Canyon Engineering), Pam Lacey (American Gas Association), Erica Bowman
(America's Natural Gas Alliance), Richard Meyer (American Gas Association), Karin Ritter
(American Petroleum Institute), Lisa Beal (Interstate Natural Gas Association of America),
Alex MacPherson (USEPA), David McCabe (Clean Air Task Force), Maureen Hardwick
(International Pharmaceutical Aerosol Consortium), Paul Ashford (Caleb Group),
Debra Reinhart (National Science Foundation), Morton Barlaz (North Carolina State
University), Chris Bayless (World Aluminum), Jerry Marks, Kenneth Martchek (Alcoa),
Lee Bray (U.S. Geological Survey), Silvio Stangherlin (ABB), Archie McCulloch,
Deborah Kramer (U.S. Geological Survey), Roger Desaulniers (Polycontrols), Kurt Werner
(3M), Cynthia Murphy (University of Texas), Lambert Kuijpers (Technical University ECfS),
Robert Wilkins (Danfoss), Bill Neuffer (USEPA), David Hind (Orica), Lorraine Gershman
(American Chemistry Council), Mausami Desai (USEPA), Nathan Frank (USEPA),
Himanshu Pathak (Indian Agricultural Research Institute), Kristen Johnson (Washington
State University), Ron Sass (Rice University), Phil Robertson (Michigan State University),
Arvin Mosier (USDA/ARS), Keith Smith (University of Edinburgh) Steven Smith (Pacific
Northwest National Laboratory), Steven Rose (Electric Power Research Institute),
Francesco Tubiello (UN Food and Agriculture Organization).
Although these individuals participated in the review of this analysis, their efforts do not
constitute an endorsement of the report's results or of any USEPA policies and programs.
-------
CONTENTS
Contents
Section Page
Executive Summary ES-1
I. Technical Summary
I. Technical Summary 1-1
I.I Overview 1-1
1.2 Non-CO2 Greenhouse Gases 1-1
1.2.1 Methane (CH4) 1-2
1.2.2 Nitrous Oxide (N2O) 1-3
1.2.3 F-Gases Gases 1-3
1.2.4 Use of GWPs in this Report 1-4
1.3 Methodology 1-4
1.3.1 Baseline Emissions for Non-CO2 Greenhouse Gases 1-5
1.3.2 Mitigation Option Analysis Methodology 1-7
1.3.3 Marginal Abatement Cost Curves 1-13
1.3.4 Methodological Enhancements from Analysis 1-15
1.4 Aggregate Results 1-15
1.4.1 Baselines 1-16
1.4.2 Global MACs 1-19
1.5 Limitations and Uncertainties 1-20
1.5.1 Exclusion of Transaction Costs 1-21
1.5.2 Static Approach to Abatement Assessment 1-21
1.5.3 Limited Use of Regional Data 1-21
1.5.4 Exclusion of Indirect Emissions Reductions 1-21
References 1-22
II. Energy
11.1. Coal Mining 11-1
II.l.l Sector Summary II-l
II.1.2 Methane Emissions from Coal Mining II-3
II.1.2.1 Activity Data and Related Assumptions II-4
II.1.2.2 Emissions Estimates and Related Assumptions II-5
II.1.3 Abatement Measures and Engineering Cost Analysis II-6
II.1.3.1 Methane Recovery System from Degasification/Drainage Systems II-8
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CONTENTS
H.l.3.2 Degasification for Utilization in Energy Production II-9
II.1.3.3 Degasification for On-site Utilization—Process Heat II-ll
II.1.3.4 Combustion through Flaring 11-12
II.1.3.5 VAM Oxidation 11-12
II.1.3.6 Evaluation of Future Mitigation Option and Trends 11-13
II.1.4 Marginal Abatement Costs Analysis 11-13
II.1.4.1 Methodological Approach 11-13
II.1.4.2 Assessment of Sectoral Trends 11-14
II.1.4.3 Definition of Model Facilities for the Analysis 11-15
II.1.4.4 Estimating Abatement Project Costs and Benefits 11-15
II.1.4.5 MAC Analysis Results 11-16
II.1.4.6 Uncertainties and Limitations 11-17
References 11-19
11.2. Oil and Natural Gas Systems 11-21
II.2.1 Sector Summary 11-21
II.2.2 Methane Emissions: Oil and Natural Gas Systems 11-22
II.2.2.1 Activity Data or Important Sectoral or Regional Trends and Related
Assumptions 11-24
II.2.2.2 Emissions Estimates and Related Assumptions 11-25
II.2.3 Abatement Measures and Engineering Cost Analysis 11-26
II.2.3.1 Oil and Natural Gas Production 11-26
II.2.3.2 Gas Processing and Transmission Segments 11-29
II.2.3.3 Gas Distribution Segment 11-34
II.2.4 Marginal Abatement Costs Analysis 11-35
II.2.4.1 Methodological Approach 11-35
II.2.4.2 MAC Analysis Results 11-40
II.2.4.3 Uncertainties and Limitations 11-42
References 11-43
III. Waste
111.1. Landfill Sector 111-1
III.1.1 Sector Summary III-l
III.1.2 Methane Emissions from Landfills III-3
III.1.2.1 Activity Data or Important Sectoral or Regional Trends and Related
Assumptions III-4
III.1.2.2 Emissions Estimates and Related Assumptions III-5
III.1.3 Abatement Measures and Engineering Cost Analysis III-6
IV GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CONTENTS
III.1.3.1 Landfill CH4 Mitigation Technologies III-7
III.1.3.2 Diversion Alternatives 111-10
III.1.4 Marginal Abatement Costs Analysis 111-15
III.1.4.1 Methodological Approach 111-16
III.1.4.2 MAC Analysis Results 111-22
III.1.4.3 Uncertainties and Limitations 111-23
References 111-24
III.2. Wastewater 111-27
III.2.1 Sector Summary 111-27
III.2.2 GHG Emissions from Wastewater 111-29
III.2.2.1 CH4 Emissions from Domestic and Industrial Wastewater 111-29
III.2.2.2 N2O Emissions from Domestic Waste water—Human Sewage 111-30
III.2.2.3 Emissions Estimates and Related Assumptions 111-30
III.2.3 Abatement Measures and Engineering Cost Analysis 111-32
III.2.3.1 Overview of Abatement Measures 111-34
III.2.3.2 CH4 Mitigation Technology for Existing Decentralized Treatment 111-35
III.2.3.3 CH4 Mitigation Technology for Existing Collection System without
Treatment 111-37
III.2.3.4 CH4 Mitigation Technology for Existing Centralized Aerobic WWTPs 111-38
III.2.4 Marginal Abatement Costs Analysis 111-40
III.2.4.1 Methodological Approach 111-40
III.2.4.2 MAC Analysis Results 111-42
III.2.4.3 Uncertainties and Limitations 111-44
References 111-45
IV. Industrial Processes
IV.1. Nitric and Adipic Acid Production IV-1
IV.1.1 Sector Summary IV-1
IV.1.2 N2O Emissions from Nitric and Adipic Acid Production IV-2
IV.1.2.1 Nitric Acid Production and Emission Factors IV-2
IV.1.2.2 Adipic Acid Production and Emission Factors IV-4
IV.1.2.3 Emissions Estimates and Related Assumptions IV-5
IV.1.3 Abatement Measures and Engineering Cost Analysis IV-6
IV.1.3.1 Adipic Add-N2O Abatement Methods IV-7
IV.1.3.2 Nitric Acid—Primary Abatement Measures IV-7
IV.1.3.3 Nitric Acid—Secondary Abatement Measures IV-7
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CONTENTS
IV.1.3.4 Nitric Acid—Tertiary Abatement Measure: Direct Catalytic
Decomposition IV-8
IV.1.3.5 Nitric Acid—Tertiary Abatement Measure: Non-selective Catalytic
Reduction (NSCR) IV-9
IV.1.4 Marginal Abatement Costs Analysis IV-9
IV.1.4.1 Methodological Approach IV-9
IV.1.4.2 MAC Analysis Results IV-13
IV.1.4.3 Uncertainties and Limitations IV-15
References IV-16
IV.2. HFC Emissions from Refrigeration and Air Conditioning IV-19
IV.2.1 Sector Summary IV-19
IV.2.2 Emissions from Refrigeration and Air Conditioning IV-20
IV.2.2.1 Activity Data or Important Sectoral or Regional Trends IV-22
IV.2.2.2 Emission Estimates and Related Assumptions IV-23
IV.2.3 Abatement Measures and Engineering Cost Analysis IV-23
IV.2.3.1 Enhanced HFC-134a in New MVACs IV-25
IV.2.3.2 HFO-1234yf in New MVACs IV-25
IV.2.3.3 Enhanced HFO-1234yf in New MVACs IV-25
IV.2.3.4 Distributed Systems in New Large Retail Food IV-26
IV.2.3.5 HFC Secondary Loop and/or Cascade Systems in New Large Retail Food IV-26
IV.2.3.6 NHs or HCs Secondary Loop and/or Cascade Systems in New Large
Retail Food IV-27
IV.2.3.7 CO2 Transcritical Systems in New Large Retail Food IV-27
IV.2.3.8 Retrofits of R-404A in Large Retail Food IV-28
IV.2.3.9 HCs in New Small Retail Food Refrigeration Systems IV-28
IV.2.3.10 HCs in New Window AC and Dehumidifiers IV-29
IV.2.3.11 R-32 in New Unitary AC Equipment and PTAC/PTHP IV-29
IV.2.3.12 MCHX in New Unitary AC Equipment IV-30
IV.2.3.13 R-32 with MCHX in New Unitary AC Equipment IV-30
IV.2.3.14 MCHX in New Positive Displacement Chillers IV-30
IV.2.3.15 NHs or CC>2 in New IPR and Cold Storage Systems IV-31
IV.2.3.16 Refrigerant Recovery at Disposal for All Existing Equipment Types IV-31
IV.2.3.17 Refrigerant Recovery at Servicing for Existing Small Equipment IV-32
IV.2.3.18 Leak Repair for Existing Large Equipment IV-32
IV.2.3.19 HCs in New Domestic Refrigeration Systems IV-33
IV.2.3.20 CC>2 in Transport Refrigeration IV-33
IV.2.3.21 Low-GWP Refrigerants and Blends IV-33
VI GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CONTENTS
IV.2.4 Engineering Cost Data Summary IV-33
IV.2.5 Marginal Abatement Costs Analysis IV-35
IV.2.5.1 Methodological Approach IV-36
IV.2.5.2 Assessment of Technical Effectiveness IV-36
IV.2.5.3 Estimating Abatement Project Costs and Benefits IV-39
IV.2.5.4 MAC Analysis Results IV-41
IV.2.6 Uncertainties and Limitations IV-42
References IV-44
IV.3. HFC Emissions from Solvent Use IV-49
IV.3.1 Sector Summary IV-49
IV.3.2 Emissions from Solvents IV-50
IV.3.2.1 Activity Data or Important Sectoral or Regional Trends IV-51
IV.3.2.2 Emission Estimates and Related Assumptions IV-51
IV.3.3 Abatement Measures and Engineering Cost Analysis IV-52
IV.3.3.1 HFC to HFE IV-53
IV.3.3.2 Retrofit IV-53
IV.3.3.3 Not-in-Kind Aqueous IV-54
IV.3.3.4 Not-in-Kind Semi-aqueous IV-54
IV.3.3.5 Low-GWP Alternatives IV-54
IV.3.3.6 Engineering Cost Data Summary IV-55
IV.3.4 Marginal Abatement Costs Analysis IV-55
IV.3.4.1 Methodological Approach IV-56
IV.3.4.2 Assessment of Technical Effectiveness IV-56
IV.3.4.3 Estimating Abatement Project Costs and Benefits IV-56
IV.3.4.4 MAC Analysis Results IV-57
IV.3.5 Uncertainties and Limitations IV-58
References IV-59
IV.4. HFC Emissions from Foams Manufacturing IV-61
IV.4.1 Sector Summary IV-61
IV.4.2 Emissions from Foams IV-62
IV.4.2.1 Activity Data, Important Sectoral or Regional Trends and Related
Assumptions IV-64
IV.4.2.2 Emission Estimates and Related Assumptions IV-64
IV.4.3 Abatement Measures and Engineering Cost Analysis IV-64
IV.4.3.1 HCs in PU Appliances IV-66
IV.4.3.2 HCs in Commercial Refrigeration IV-66
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES VII
-------
CONTENTS
IV.4.3.3 HC in Polyurethane Spray Foams IV-67
IV.4.3.4 CO2 in Polyurethane Spray Foams IV-67
IV.4.3.5 LCD/Alcohol in XPS Boardstock IV-67
IV.4.3.6 HFC-134a to HCs in PU One-Component Foam IV-68
IV.4.3.7 HFC-152a to HCs in PU One-Component Foam IV-68
IV.4.3.8 HCs in PU Continuous and Discontinuous Foams IV-68
IV.4.3.9 Manual Blowing Agent Recovery from Appliances at End of Life (EOL) IV-69
IV.4.3.10 Fully Automated Blowing Agent Recovery from Appliances at EOL IV-69
IV.4.3.11 Solstice Liquid Blowing Agent in PU Foams IV-69
IV.4.3.12 Solstice Gas Blowing Agent in XPS Foam and One-Component Foam IV-70
IV.4.3.13 Methyl Formate in PU and XPS Foams IV-71
IV.4.4 Engineering Cost Data Summary IV-71
IV.4.5 Marginal Abatement Cost Analysis IV-72
IV.4.5.1 Methodological Approach IV-72
IV.4.5.2 Assessment of Technical Effectiveness IV-72
IV.4.5.3 Estimating Abatement Project Costs and Benefits IV-73
IV.4.5.4 MAC Analysis Results IV-74
IV.4.6 Uncertainties and Limitations IV-74
References IV-76
IV.5. HFC Emissions from Aerosol Product Use IV-79
IV.5.1 Sector Summary IV-79
IV.5.2 Emissions from Aerosol Product Use IV-80
IV.5.2.1 Activity Data or Important Sectoral or Regional Trends IV-81
IV.5.2.2 Emission Estimates and Related Assumptions IV-81
IV.5.3 Abatement Measures and Engineering Cost Analysis IV-81
IV.5.3.1 Hydrocarbons IV-82
IV.5.3.2 Not-in-Kind IV-83
IV.5.3.3 HFO-1234ze IV-83
IV.5.3.4 HFC-134a to HFC-152a IV-84
IV.5.3.5 Dry Powder Inhalers IV-84
IV.5.3.6 Engineering Cost Data Summary IV-85
IV.5.4 Marginal Abatement Costs Analysis IV-85
IV.5.4.1 Methodological Approach IV-85
IV.5.4.2 Assessment of Technical Effectiveness IV-85
IV.5.4.3 Estimating Abatement Project Costs and Benefits IV-86
IV.5.4.4 MAC Analysis Results IV-87
VIII GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CONTENTS
IV.5.5 Uncertainties and Limitations IV-88
References IV-89
IV.6. HFC and RFC Emissions from Fire Protection IV-91
IV.6.1 Sector Summary IV-91
IV.6.2 Emissions from Fire Protection IV-92
IV.6.2.1 Activity Data or Important Sectoral or Regional Trends IV-93
IV.6.2.2 Emission Estimates and Related Assumptions IV-93
IV.6.3 Abatement Measures and Engineering Cost Analysis IV-94
IV.6.3.1 FK-5-1-12 in New Class A Total Flooding Applications IV-95
IV.6.3.2 Inert Gas Systems in New Class A Total Flooding Applications IV-96
IV.6.3.3 Water Mist Systems in New Class B Total Flooding Applications IV-96
IV.6.4 Engineering Cost Data Summary IV-97
IV.6.5 Marginal Abatement Cost Analysis IV-97
IV.6.5.1 Methodological Approach IV-97
IV.6.5.2 Assessment of Technical Effectiveness IV-98
IV.6.5.3 Estimating Abatement Project Costs and Benefits IV-99
IV.6.5.4 MAC Analysis Results IV-99
IV.6.6 Uncertainties and Limitations IV-99
References IV-101
IV.7. RFC Emissions from Primary Aluminum Production IV-103
IV.7.1 Sector Summary IV-103
IV.7.2 Emissions from Primary Aluminum Production IV-105
IV.7.2.1 Activity Data and Important Trends IV-106
IV.7.2.2 Emission Estimates and Related Assumptions IV-107
IV.7.3 Abatement Measures and Engineering Cost Analysis IV-108
IV.7.3.1 Minor Retrofit IV-109
IV.7.3.2 Major Retrofit IV-109
IV.7.3.3 Engineering Cost Data Summary IV-110
IV.7.4 Marginal Abatement Costs Analysis IV-110
IV.7.4.1 Methodological Approach IV-111
IV.7.4.2 Definition of Model Facilities IV-111
IV.7.4.3 Assessment of Technical Effectiveness IV-111
IV.7.4.4 Estimating Abatement Project Costs and Benefits IV-112
IV.7.4.5 MAC Analysis Results IV-113
IV.7.4 Uncertainties and Limitations IV-114
References IV-115
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IX
-------
CONTENTS
IV.8. HFC-23 Emissions from HCFC-22 Production IV-117
IV.8.1 Sector Summary IV-117
IV.8.1.1 Emissions from HCFC-22 Production IV-119
IV.8.1.2 Activity Data or Important Sectoral or Regional Trends IV-119
IV.8.1.3 Emission Estimates and Related Assumptions IV-120
IV.8.2 Abatement Measures and Engineering Cost Analysis IV-122
IV.8.2.1 Thermal Oxidation IV-123
IV.8.2.2 Evaluation of Future Mitigation Options and Trends IV-123
IV.8.3 Marginal Abatement Costs Analysis IV-124
IV.8.3.1 Methodological Approach IV-124
IV.8.3.2 Assessment of Technical Effectiveness IV-124
IV.8.3.3 Estimating Abatement Project Costs and Benefits IV-125
IV.8.3.4 MAC Analysis Results IV-125
IV.8.4 Uncertainties and Limitations IV-127
References IV-128
IV.9. F-GHG Emissions from Semiconductor Manufacturing IV-129
IV.9.1 Sector Summary IV-129
IV.9.2 Emissions from Semiconductor Manufacturing IV-131
IV.9.2.1 Activity Data or Important Sectoral or Regional Trends IV-131
IV.9.2.2 Emissions Estimates and Related Assumptions IV-133
IV.9.3 Abatement Measures and Engineering Cost Analysis IV-134
IV.9.3.1 Thermal Abatement IV-135
IV.9.3.2 Catalytic Abatement IV-136
IV.9.3.3 Plasma Abatement IV-136
IV.9.3.4 NFs Remote Chamber Clean IV-137
IV.9.3.5 Gas Replacement IV-138
IV.9.3.6 Process Optimization IV-138
IV.9.4 Marginal Abatement Cost Analysis IV-139
IV.9.4.1 Methodological Approach IV-139
IV.9.4.2 Definition of Model Facilities IV-140
IV.9.4.3 Assessment of Technical Effectiveness IV-140
IV.9.4.4 Estimating Abatement Project Costs and Benefits IV-142
IV.9.4.5 MAC Analysis Results IV-143
IV.9.5 Uncertainties and Limitations IV-144
References IV-145
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CONTENTS
IV.10.SF6 Emissions from Electric Power Systems IV-147
IV.10.1 Sector Summary IV-147
IV.10.2 SFe Emissions from Electric Power Systems IV-149
IV.10.2.1 Activity Data or Important Sectoral or Regional Trends IV-150
IV.10.2.2 Emission Estimates and Related Assumptions IV-152
IV.10.3 Abatement Measures and Engineering Cost Analysis IV-153
IV.10.3.1 SFe Recycling IV-153
IV.10.3.2 Leak Detection and Leak Repair (LDAR) IV-155
IV.10.3.3 Equipment Refurbishment IV-155
IV.10.3.4 Improved SF6 Handling IV-156
IV.10.4 Marginal Abatement Costs Analysis IV-157
IV.10.4.1 Methodological Approach IV-157
IV.10.4.2 Definition of EPS Model Facilities IV-157
IV.10.4.3 Parameters Used to Estimate Technical Effectiveness IV-159
IV.10.4.4 Estimating Abatement Project Costs and Benefits IV-159
IV.10.4.5 MAC Analysis Results IV-160
IV.10.5 Uncertainties and Limitations IV-162
References IV-163
IV.11.SF6 Emissions from Magnesium Production IV-165
IV.11.1 Sector Summary IV-165
IV.11.2 SFe Emissions from Magnesium Manufacturing IV-166
IV.11.2.1 Activity Data or Important Sectoral or Regional Trends IV-167
IV.11.2.2 Emission Estimates and Related Assumptions IV-168
IV.11.3 Abatement Measures and Engineering Cost Analysis IV-169
IV.11.3.1 Replacement with Alternative Cover Gas—Sulfur Dioxide (SCh) IV-169
IV.11.3.2 Replacement with Alternative Cover Gas-HFC 134a IV-170
IV.11.3.3 Replacement with Alternative Cover Gas-Novec™ 612 IV-170
IV.11.3.4 Summary of Mitigation Technology Costs and Characteristics IV-170
IV.11.4 Marginal Abatement Costs Analysis IV-171
IV.11.4.1 Methodological Approach IV-171
IV.11.4.2 Model Facilities Defined IV-171
IV.11.4.3 Assessment of Technical Effectiveness IV-172
IV.11.4.4 Estimating Abatement Project Costs and Benefits IV-172
IV.11.4.5 MAC Analysis Results IV-172
IV.11.5 Uncertainties and Limitations IV-174
References IV-175
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES XI
-------
CONTENTS
IV.12. Emissions from Photovoltaic Cell Manufacturing IV-177
IV.12.1 Sector Summary IV-177
IV.12.2 Emissions from Photovoltaic Cell Manufacturing IV-178
IV.12.2.1 Activity Data and Important Sectoral/Regional Trends IV-180
IV.12.2.2 Emissions Estimates and Related Assumptions IV-180
IV.12.3 Abatement Measures and Engineering Cost Analysis IV-181
IV.12.3.1 Thermal Abatement IV-182
IV.12.3.2 Catalytic Abatement IV-183
IV.12.3.3 Plasma Abatement IV-183
IV.12.3.4 NFs Remote Chamber Clean IV-183
IV.12.3.5 Summary of Mitigation Technology Costs and Characteristics IV-184
IV.12.4 Marginal Abatement Costs Analysis IV-184
IV.12.4.1 Methodological Approach IV-184
IV.12.4.2 Definition of Model Facility IV-184
IV.12.4.3 Assessment of Technical Effectiveness IV-185
IV.12.4.4 Estimating the Break-Even Price of Abatement Measures IV-185
IV.12.4.5 MAC Analysis Results IV-186
IV.12.5 Uncertainties and Limitations IV-187
References IV-189
IV.13.PFC Emissions from Flat Panel Display Manufacturing IV-191
IV.13.1 Sector Summary IV-191
IV.13.2 Emissions from Flat Panel Display Manufacturing IV-192
IV.13.2.1 Activity Data or Important Sectoral or Regional Trends and Related
Assumptions IV-193
IV.13.2.2 Emissions Estimates and Related Assumptions IV-194
IV.13.3 Abatement Measures and Engineering Cost Analysis IV-194
IV.13.3.1 Central Abatement IV-195
IV.13.3.2 Thermal Abatement IV-195
IV.13.3.3 Catalytic Abatement IV-195
IV.13.3.4 Plasma Abatement IV-196
IV.13.3.5 NFs Remote Chamber Clean IV-196
IV.13.3.6 Gas Replacement IV-196
IV.13.3.7 Summary of Mitigation Technology Costs and Characteristics IV-197
IV.13.4 Marginal Abatement Costs Analysis IV-197
IV.13.4.1 Methodological Approach IV-197
IV.13.4.2 Facility Definition IV-197
IV.13.4.3 Estimating the Technical Effectiveness Parameter IV-197
XII GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CONTENTS
IV.13.4.4 Estimating Break-Even Prices IV-198
IV.13.4.5 MAC Analysis Results IV-199
IV.13.5 Uncertainties and Limitations IV-200
References IV-201
V. Agriculture
V.1. Non-Rice Croplands V-1
V.I.I Sector Summary V-1
V.1.2 Emissions from Non-Rice Croplands V-4
V.l.2.1 Methodology V-4
V.1.3 Abatement Measures and Engineering Cost Analysis V-6
V.I.3.1 Mitigation Technologies V-6
V.1.4 Marginal Abatement Costs Analysis V-ll
V.l.4.1 Estimate Abatement Measure Costs and Benefits V-ll
V.l.4.2 MAC Analysis Results V-ll
V.1.5 Sensitivity Analysis V-13
V.1.5 Uncertainties and Limitations V-15
References V-17
V.2. Rice Cultivation V-19
V.2.1 Sector Summary V-19
V.2.2 CH4 and N2O Emissions and Changes in Soil Carbon from Rice Cultivation V-21
V.2.2.1 Activity Data or Important Sectoral or Regional Trends and Related
Assumptions V-22
V.2.2.2 Emissions Estimates and Related Assumptions V-32
V.2.3 Abatement Measures and Engineering Cost Analysis V-32
V.2.4 Marginal Abatement Costs Analysis V-36
V.2.4.1 MAC Analysis Results V-36
V.2.5 Sensitivity Analyses V-39
V.2.5. Uncertainties and Limitations V-41
References V-42
V.3. Livestock V-43
V.3.1 Sector Summary V-43
V.3.2 CH4 and N2O Emissions from Livestock Management V-46
V.3.2.1 CH4 Emissions from Enteric Fermentation V-46
V.3.2.2 CH4 and N2O Emissions from Manure Management V-46
V.3.2.3 Baseline CH4 and N2O Emissions Estimates V-47
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES XIII
-------
CONTENTS
V.3.3 Abatement Measures and Engineering Cost Analysis V-50
V.3.3.1 Enteric Fermentation CH4 Mitigation Technologies V-50
V.3.3.2 Manure Management CH4 Mitigation Technologies V-53
V.3.4 Marginal Abatement Costs Analysis V-58
V.3.4.1 Development of Disaggregated Baseline Livestock Populations V-58
V.3.4.4 MAC Analysis Results V-63
V.3.5 Sensitivity Analyses V-65
V.3.4.5. Uncertainties and Limitations V-71
References V-73
Appendixes
A: Tables A-l
B: Coal Mining B-l
C: Natural Gas and Oil Systems C-l
D: Refrigeration and Air Conditioning D-l
E: Solvent Use E-l
F: Foams Manufacturing F-l
G: Aerosol Product Use G-l
H: Fire Protection H-l
I: Primary Aluminum Production 1-1
J: HCFC-22 Production J-l
K: Electric Power Systems K-l
L: Magnesium Manufacturing L-l
M: Photovoltaic Cell Manufacturing M-l
N: Flat Panel Display Manufacturing N-l
O: Description of the Input Data Used in DAYCENT Simulations O-l
XIV GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CONTENTS
List of Pictures
Number
Section I
I-l Contribution of Anthropogenic Emissions of Greenhouse Gases to the Enhanced
Greenhouse Effect from Preindustrial to Present (measured in watts/meter2) 1-2
1-2 Illustrative Non-CCh Marginal Abatement Curve 1-13
1-3 Percentage Share of Global Non-CCfe Emissions3 by Type of Gas in 2010 1-16
1-4 Non-CCh Global Emissions Forecast to 2030 by Greenhouse Gas 1-17
1-5 Global Emissions by Major Sector for all Non-CCh Greenhouse Gases 1-18
1-6 Projected World Emissions Baseline for Non-CCh Greenhouse Gases, Including
Top Emitting Regions 1-18
1-7 Global 2030 MACs by Non-CO2 Greenhouse Gas 1-19
1-8 Global 2030 MACs for Non-CCh Greenhouse Gases by Major Emitting Regions 1-20
1-9 Global 2030 MACs for Non-CCh Greenhouse Gases by Major Emitting Regions 1-20
Section II
1-1 CH4 Emissions from Coal Mining: 2000-2030 II-l
1-2 Global Abatement Potential in Coal Mining: 2010, 2020, and 2030 II-2
1-3 Flow Chart of the Coal Mining Sector MAC Modeling Approach 11-14
1-4 Marginal Abatement Cost Curve for Top 5 Emitters and Rest of World in 2030 11-17
2-1 Emissions Projections for the Oil and Natural Gas Systems Sector: 2000-2030 11-21
2-2 Global Abatement Potential in Oil and Natural Gas Systems: 2010, 2020, and 2030 11-22
2-3 Segments of Oil and Natural Gas Systems 11-23
2-4 Global Natural Gas Production: 2015-2035 11-24
2-5 Diagram of BAU Emissions for the United States Oil and Natural Gas System 11-37
2-6 Marginal Abatement Cost Curves for Top 5 Emitters in 2030 11-41
Section III
1-1 Emissions Projections for the Landfill Sector: 2000-2030 III-l
1-2 Global Abatement Potential in Landfill Sector: 2010, 2020, and 2030 III-2
1-3 Conceptual Model for Estimating Mitigation Potential in the MSW Landfill Sector 111-17
1-4 Marginal Abatement Cost Curve for Top 5 Emitters in 2030 111-23
2-1 CH4 Emissions from Wastewater: 2000-2030 111-27
2-2 N2O Emissions from Domestic Wastewater: 2000-2030 111-28
2-3 Global MAC for Wastewater: 2010, 2020, and 2030 111-29
2-4 Sanitation Ladder for Improvements to Wastewater Treatment 111-33
2-5 Five Existing Scenarios Evaluated for Given Wastewater Discharge Pathways
Based on Technology Level, Treatment Alternative, and Collection Method 111-33
2-6 Mitigation Technology Approach for Developing Countries with Decentralized
Treatment 111-35
2-7 Mitigation Technology Approach for Developing Countries with Decentralized
Treatment 111-38
2-8 Mitigation Technology Approach for Countries with Existing Centralized WWTPs 111-39
2-9 Domestic Wastewater MAC Analysis Flow Chart 111-40
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
XV
-------
CONTENTS
2-10 Share of Wastewater Q-k Emissions to Domestic and Industrial Sources (Avg.
2002-2007) 111-41
2-11 Marginal Abatement Cost Curve for Top 5 Emitters in 2020 111-43
Section IV
1-1 N2O Emissions from Nitric and Adipic Acid: 2000-2030 IV-1
1-2 Global MAC for Nitric and Adipic Acid: 2010, 2020, and 2030 IV-2
1-3 Adipic Acid Production Capacity by Country: 2010 IV-4
1-4 Operational Adipic Acid Production Facilities in 2010 by Share of Global Capacity IV-11
1-5 Marginal Abatement Cost Curve for Top 5 Emitters in 2030 IV-14
2-1 HFC Emissions from Refrigeration and AC: 2000-2030 (MtCChe) IV-19
2-2 Global Abatement Potential in Refrigeration and AC: 2010, 2020, and 2030 IV-20
2-3 Global HFC Emissions in 2020 by Application Type (% of GWP-Weighted
Emissions) IV-21
2-4 Marginal Abatement Cost Curves for Top Five Emitters in 2030 IV-42
3-1 HFC and PFC Emissions from Solvent Use: 2000-2030 (MtCChe) IV-49
3-2 Global Abatement Potential in Solvent Use: 2010, 2020, and 2030 IV-50
3-3 Global HFC Emissions in 2020 by Degreaser Type (% of GWP-Weighted Emissions).... IV-51
3-4 Marginal Abatement Cost Curves for Top Five Emitters in 2030 IV-58
4-1 HFC Emissions from Foams Manufacturing: 2000-2030 (MtCChe) IV-61
4-2 Global Abatement Potential in Foams Manufacturing: 2010, 2020, and 2030 IV-62
4-3 Global HFC Emissions in 2020 by Application Type (% of GWP-Weighted
Emissions) IV-63
4-4 Marginal Abatement Cost Curves for Top Five Foam Emitters in 2030 1V-75
5-1 HFC Emissions from Aerosol Product Use: 2000-2030 (MtCChe) IV-79
5-2 Global Abatement Potential in Aerosol Product Use: 2010, 2020, and 2030 IV-80
5-3 Global HFC Emissions in 2020 by Facility Type (% of GWP-Weighted Emissions) IV-81
5-4 Marginal Abatement Cost Curves for Top Five Emitters in 2030 IV-88
6-1 HFC and PFC Emissions from Fire Protection: 2000-2030 (MtCChe) IV-91
6-2 Global Abatement Potential in Fire Protection: 2010, 2020, and 2030 IV-92
6-3 Global HFC and PFC Emissions in 2020 (% of GWP-Weighted Emissions) IV-93
6-4 Marginal Abatement Cost Curves for Top Five Emitters in 2030 IV-100
7-1 PFC Emissions from Primary Aluminum Production: 2000-2030 (MtCChe) IV-104
7-2 Global Abatement Potential in Primary Aluminum Production: 2010, 2020, and
2030 IV-105
7-3 Global PFC Emissions in 2020 by Facility Type (% of GWP-Weighted Emissions) IV-106
7-4 Marginal Abatement Cost Curves for Top Five Emitters in 2030 IV-114
8-1 HCF-23 Emissions from HCFC-22 Production: 2000-2030 (MtCChe) IV-117
8-2 Global Abatement Potential in HCFC-22 Production: 2010, 2020, and 2030 IV-118
8-3 Global HFC-23 Emissions in 2020 by Facility Type (% of GWP-Weighted
Emissions) IV-120
8-4 Marginal Abatement Cost Curves for Countries with Abatement Potential in 2030 IV-126
9-1 Projected Baseline Emissions from Semiconductor Manufacturing: 2000-2030
(MtCChe) IV-129
9-2 Global Abatement Potential in Semiconductor Manufacturing: 2010, 2020, and 2030 .. IV-130
9-3 Global F-GHG Emissions in 2020 by Fab Type and Process (% of GWP-Weighted
Emissions) IV-132
XVI GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CONTENTS
9-4 Marginal Abatement Cost Curves for Top Five Emitters in 2030 IV-143
10-1 SFe Emissions from Electric Power Systems: 2000-2030 (MtCChe) IV-147
10-2 Global Abatement Potential in Electric Power Systems: 2010, 2020, and 2030 IV-148
10-3 Percentage of Global SFe Emissions in 2020 by Emission Stream (% of GWP-
Weighted Emissions) IV-149
10-4 Global SFe Emissions in 2020 by Facility Type (% of GWP-Weighted Emissions) IV-151
10-5 Distribution of 2010 Emission Rates Reported through USEPA's Voluntary
Partnership IV-158
10-6 Marginal Abatement Cost Curves for Top Five and Rest of World Emitters in 2030.... IV-161
11-1 SFe Emissions from Magnesium Production: 2000-2030 (MtCChe) IV-165
11-2 Global Abatement Potential in Magnesium Manufacturing: 2010, 2020, and 2030 IV-166
11-3 Global SFe Emissions in 2020 by Facility Type (% of GWP-Weighted Emissions) IV-167
11-4 Marginal Abatement Cost Curves for Top Five Emitters in 2030 IV-174
12-1 F-GHG Emissions from PV Cell Manufacturing: 2000-2030 (MtCChe) IV-177
12-2 Global Abatement Potential in PV Cell Manufacturing: 2010, 2020, and 2030 IV-178
12-3 Global F-GHG Emissions in 2020 by Facility Type (% of GWP-Weighted Emissions).. IV-179
12-4 Marginal Abatement Cost Curves for Top Five Emitters and Rest of World in 2030.... IV-187
13-1 F-GHG Emissions from FPD Manufacturing: 2000-2030 (MtCChe) IV-191
13-2 Global Abatement Potential in FPD Manufacturing: 2010, 2020, and 2030 IV-192
13-3 Global F-GHG Emissions in 2020 by Facility Type (% of GWP-Weighted Emissions).. IV-193
13-4 Marginal Abatement Cost Curves by Emitting County in 2030 IV-199
Section V
1-1 Global Baseline Emissions from Non-Rice Croplands by GHG: 2010-2030 V-2
1-2 Baseline Net GHG Emissions from Non-Rice Croplands, Top Five Emitting
Countries V-3
1-3 Global MAC Curve for Net GHG Reductions from Non-Rice Cropland Soils V-4
1-4 Marginal Abatement Cost Curve for Top-Five Emitting Countries in 2010 and 2030 V-12
1-5 Global Abatement Potential in Non-rice Croplands Management: 2010, 2020, and
2030 (Includes "Optimal N Fertilization" Strategy) V-14
1-6 Marginal Abatement Cost Curve for Top 5 Emitters in 2010, 2030 (Includes
"Optimal N Fertilization" Strategy) V-15
2-1 Net GHG Emissions Projections for Rice Cultivation: 2000-2030 V-19
2-2 Global Abatement Potential in Rice Cultivation with Production Equal to Baseline
Levels: 2010, 2020, and 2030 V-21
2-3 DNDC Rice Cropland Area Sown, Top 5 countries, by Type and Water
Management V-23
2-4 Marginal Abatement Cost Curve for Top 5 Emitters in 2030, Baseline Production
Case V-39
2-5 Marginal Abatement Cost Curve, Baseline Area Case V-40
2-6 Marginal Abatement Cost Curve for Top 5 Emitters in 2030, Baseline Area Case V-40
3-1 Total Net GHG Emissions and Projections for the Livestock Sector: 2000-2030 V-43
3-2 CH4 Emissions Projections for the Livestock Sector: 2010-2030 V-44
3-3 N20 Emissions Projections from the Livestock Sector: 2010-2030 V-44
3-4 Global Abatement Potential in Livestock Management: 2010, 2020, and 2030 V-45
3-5 Marginal Abatement Cost Curve for Top 5 Emitters in 2030 (Baseline Production
Case) V-64
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES XVII
-------
CONTENTS
3-6 Global Net GHG Livestock Emissions Reduction Potential by Mitigation Option
(Baseline Production Case) V-65
3-7 Global Abatement Potential in Livestock Management, Baseline Number of
Animals : 2010, 2020, and 2030 V-66
3-8 Global Abatement Potential in Livestock Management, Baseline Production with
No Antimethanogen: 2010, 2020, and 2030 V-67
3-9 Global Abatement Potential in Livestock Management, Baseline Number of
Animals with No Antimethanogen: 2010, 2020, and 2030 V-69
3-10 Global Beef Production under Baseline and Mitigation Options, Assuming Full
Adoption of Individual Options and Holding the Number of Animals Constant V-70
3-11 Global Production of Milk from Dairy Cattle Under Baseline and Mitigation
Options, Assuming Full Adoption of Individual Options and Holding the Number
of Animals Constant V-71
XVIII GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CONTENTS
List of Tables
Number
Section I
I-l Global Non-CO2 Greenhouse Gas (GHG) Emissions for 2010 (MtCChe) by Source
and Gas Type 1-2
1-2 Global Warming Potentials 1-4
1-3 Calculation of Potential Emission Reduction for an Abatement Option 1-9
1-4 Financial Assumptions in Breakeven Price Calculations for Abatement Options 1-11
1-5 International Economic Adjustment Factors for Selected Countries 1-12
Section II
1-1 IPCC Suggested Underground Emissions Factors for Selected Countries in
m3/tonne Coal Produced II-5
1-2 Projected Emissions from Coal Mine CH4 by Country and Region: 2010 to 2030
(MtCO2e) II-6
1-3 Summary of Abatement Measures for Coal Mines II-7
1-4 Factors Used to Estimate Abatement Potential in Coal Mines II-7
1-5 Example Break-Even Price Calculation for Coal Mine Abatement Measures 11-15
1-6 Abatement Potential by Region at Selected Break-Even Prices ($/tCO2e) in 2030 11-16
2-1 Emissions Source from Oil and Natural Gas Systems 11-23
2-2 Projected Baseline CH4 Emissions for Oil and Natural Gas Systems by
Country/Region: 2010-2030 (MtCO2e) 11-25
2-3 Abatement Measures Applied in Oil and Gas Production Segments 11-27
2-4 Abatement Measures for the Natural Gas Processing Segment 11-30
2-5 Abatement Measures for the Natural Gas Transmission Segment 11-31
2-6 Abatement Measures for the Distribution Segment 11-34
2-7 International Statistics on Key Activity Drivers: 2010 11-36
2-8 Example Break-Even Price Calculation based on 2010 MAC for the United States 11-39
2-9 Abatement Potential by Region at Selected Break-Even Prices in 2030 11-41
Section III
1-1 Projected Baseline Emissions for MSW Landfills by Country: 2010-2030 (MtCO2e) III-5
1-2 Summary of the Engineering and Cost Assumptions for Abatement Measures at
Landfills III-6
1-3 Electricity Generation Technology Costs III-8
1-4 Model Facilities Share of BAU Emissions: 2010-2030 111-18
1-5 Model Facility Assumptions for International LFG Mitigation Options 111-19
1-6 CH4 Generation Factors by Country 111-20
1-7 Example Break-Even Prices for MSW Landfill Technology Options 111-21
1-8 Break-Even Prices of Waste Diversion Options 111-21
1-9 Abatement Potential by Region at Selected Break-Even Prices in 2030 (MtCO2e) 111-22
2-1 Projected CH4 Baseline Emissions from Wastewater: 2010-2030 (MtCO2e) 111-31
2-2 Projected N2O Baseline Emissions from Human: 2010-2030 (MtCO2e) 111-31
2-3 Abatement Measures for the Wastewater Sector 111-34
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
XIX
-------
CONTENTS
2-4 Example Break-Even Prices for Wastewater Abatement Measures in 2030 for the
United States 111-42
2-5 Abatement Potential by Region at Selected Break-Even Prices in 2030 111-43
Section IV
1-1 IPCC Emissions Factors for Nitric Acid Production IV-3
1-2 Projected N2O Baseline Emissions from Nitric and Adipic Acid Production: 2010-
2030 IV-5
1-3 Abatement Measures for Nitric and Adipic Acid Production IV-6
1-4 Adipic Acid-Producing Countries' Share of Baseline Emissions IV-12
1-5 Model Nitric Acid Facilities Assumptions IV-13
1-6 Example Break-Even Prices for N2O Abatement Measures IV-13
1-7 Abatement Potential by Region at Selected Break-Even Prices in 2030 IV-14
2-1 Projected Baseline Emissions from Refrigeration and AC: 2010 to 2030 (MtCChe) IV-23
2-2 Refrigeration and AC Abatement Options IV-24
2-3 Engineering Cost Data on a Facility Basis IV-34
2-4 Technical Effectiveness Summary IV-36
2-5 Example Break-Even Prices for Abatement Measures in Refrigeration and AC IV-39
2-6 Abatement Potential by Country/Region at Selected Break-Even Prices in 2030
(MtCChe) IV-41
3-1 Projected Baseline Emissions from Solvent Use: 2010-2030 (MtCChe) IV-52
3-2 Solvent Use Abatement Options IV-52
3-3 Engineering Cost Data on a Facility Basis IV-55
3-4 Technical Effectiveness Summary IV-56
3-5 Example Break-Even Prices for Abatement Measures in Solvent Use IV-57
3-6 Abatement Potential by Country/Region at Selected Break-Even Prices in 2030
(MtCO2e) IV-57
4-1 Projected Baseline Emissions from Foams Manufacturing: 2010-2030 (MtQIhe) IV-65
4-2 Foams Manufacturing Abatement Options IV-65
4-3 Engineering Cost Data on a Facility Basis IV-71
4-4 Technical Effectiveness Summary IV-73
4-5 Example Break-Even Prices for Abatement Measures in Foams Manufacturing IV-74
4-6 Abatement Potential by Country/Region at Selected Break-Even Prices in 2030
(MtCO2e) IV-75
5-1 Projected Baseline Emissions from Aerosol Product Use: 2010-2030 (MtCO2e) IV-82
5-2 Aerosol Product Use Abatement Options IV-83
5-3 Engineering Cost Data on a Facility Basis IV-85
5-4 Technical Effectiveness Summary IV-86
5-5 Example Break-Even Prices for Abatement Measures in Aerosol Product Use IV-86
5-6 Abatement Potential by Country/Region at Selected Break-Even Prices in 2030
(MtCO2e) IV-87
6-1 Projected Baseline Emissions from Fire Protection: 2010 to 2030 (MtCO2e) IV-94
6-2 Fire Protection Abatement Options IV-95
6-3 Engineering Cost Data on a Facility Basis IV-98
6-4 Technical Effectiveness Summary IV-98
6-5 Example Break-Even Prices for Abatement Measures in Fire Protection IV-99
XX GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CONTENTS
6-6 Abatement Potential by Country/Region at Selected Break-Even Prices in 2030
(MtCChe) IV-100
7-1 Projected Baseline Emissions from Primary Aluminum Production: 2010-2030
(MtCChe) IV-108
7-2 Primary Aluminum Production Abatement Options IV-109
7-3 Engineering Cost Data on a Facility Basis IV-111
7-4 Description of Primary Aluminum Production Facilities IV-112
7-5 Technical Effectiveness Summary IV-112
7-6 Example Break-Even Prices for Abatement Measures in Primary Aluminum
Production IV-113
7-7 Abatement Potential by Country/Region at Selected Break-Even Prices in 2030
(MtCChe) IV-113
8-1 Projected Baseline Emissions from HCFC-22 Production: 2010-2030 (MtCChe) IV-120
8-2 Distribution of HCF-23 Emissions by Location and Facility Type: 2010-2030 IV-121
8-3 HCFC-22 Production Abatement Options IV-122
8-4 Engineering Cost Data on a Facility Basis IV-122
8-5 Technical Effectiveness Summary IV-125
8-6 Example Break-Even Prices for Abatement Measures in HCFC-22 Production IV-125
8-7 Abatement Potential by Country/Region at Selected Break-Even Prices in 2030
(MtCO2e) IV-126
9-1 Projected Baseline Emissions from Semiconductor Manufacturing: 2010-2030
(MtCO2e) IV-133
9-2 Semiconductor Manufacturing Abatement Options IV-134
9-3 Engineering Cost Data on a Facility Basis IV-135
9-4 Annual Cost per Tool for Thermal Abatement Systems IV-136
9-5 Capital Costs per CVD Chamber for Making a Facility NFs Ready IV-138
9-6 Percentage of Annual Emissions by Process and Fab Type IV-141
9-7 Technical Effectiveness Summary for New Fabs (Constant Over Time) IV-141
9-8 Technical Effectiveness Summary for Old Fabs (in 2020) IV-141
9-9 Example Break-Even Prices for Abatement Measures in Semiconductor
Manufacturing IV-142
9-10 Abatement Potential by Country/Region at Selected Break-Even Prices in 2030
(MtCO2e) IV-143
10-1 Projected Baseline Emissions from Electric Power Systems: 2010-2030 (MtCO2e) IV-152
10-2 EPS Abatement Options IV-153
10-3 Engineering Cost Data on a Facility Basis IV-154
10-4 EPSs System Country Mapping IV-159
10-5 Technical Effectiveness Summary IV-160
10-6 Example Break-Even Prices for Abatement Measures in EPSs IV-160
10-7: Abatement Potential by Country/Region at Selected Break-Even Prices in 2030
(MtCO2e) IV-161
11-1 Projected Baseline Emissions from Magnesium Production: 2010-2030 (MtCO2e) IV-168
11-2 Magnesium Production Abatement Options IV-169
11-3 Engineering Cost Data on a Facility Basis IV-171
11-4 Technical Effectiveness Summary IV-172
11-5 Example Break-Even Prices for Abatement Measures in Magnesium Manufacturing.. IV-173
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES XXI
-------
CONTENTS
11-6 Abatement Potential by Country/Region at Selected Break-Even Prices in 2030
(MtCChe) IV-173
12-1 Projected Baseline Emissions from PV Cell Manufacturing: 2010-2030 (MtCChe) IV-182
12-2 PV Cell Manufacturing Abatement Options IV-182
12-3 Engineering Cost Data on a Facility Basis IV-184
12-4 Technical Effectiveness Summary IV-185
12-5 Example Break-Even Prices for Abatement Measures in PV Cell Manufacturing IV-186
12-6 Abatement Potential by Country/Region at Selected Break-Even Prices in 2030
(MtCChe) IV-186
13-1 Projected Baseline Emissions from FPD Manufacturing: 2010-2030 (MtCChe) IV-194
13-2 FPD Manufacturing Abatement Options IV-195
13-3 Engineering Cost Data on a Facility Basis IV-197
13-4 Technical Effectiveness Summary IV-198
13-5 Example Break-Even Prices for Abatement Measures in FPD Manufacturing IV-198
13-6 Abatement Potential by Country/Region at Selected Break-Even Prices in 2030
(MtCO2e) IV-199
Section V
1-1 Projected Net GHG Baseline Emissions from Non-Rice Croplands by Country:
2010-2030 (MtCO2e) V-3
1-2 DAYCENT Base Mean Yields, and Differences from Mean Yield for Mitigation
Strategies, by Year (Metric tons of Grain per Hectare) V-9
1-3 Abatement Potential at Selected Break-Even Prices in 2030 (No "Optimal
Fertilization "Scenario) V-ll
1-4 Global Total Abatement Potential from Cropland Soils by Measure (MtCO2e)
("Optimal N Fertilization" Strategy excluded) V-13
1-5 Global Total Abatement Potential from Cropland Soils by Measure (MtCO2e)
(Includes "Optimal N Fertilization" Strategy) V-14
2-1 Baseline CH4, N2O, and Soil Carbon Estimates for Rice Cropland for 2010, 2020 and
2030 by Region V-20
2-2 Baseline yields for 2010, 2020 and 2030 for selected countries (kg/ha) V-24
2-3 Baseline production for 2010, 2020 and 2030 for selected countries (metric tonnes) V-24
2-4 DNDC Average N Fertilizer Application Rate by Country and Rice Production
Type V-25
2-5 Distribution of Baseline Water Management for Irrigated Rice by Country (%) V-28
2-6 Alternative Rice Management Scenarios Simulated using DNDC V-33
2-7 Rice Management Techniques V-35
2-8 Abatement Potential by Region at Selected Break-Even Prices in 2030 (MtCO2e) V-37
2-9 Distribution of Net GHG Reductions across Mitigation Options, Baseline
Production Case V-38
3-1 Projected Baseline Emissions from Livestock Management: 2010-2030 (MtCO2e) V-48
3-2 Projected Baseline Emissions from Enteric Fermentation: 2010-2030 (MtCO2e) V-48
3-3 Projected Baseline Emissions from Manure Management: 2010-2030 (MtCO2e) V-49
3-4 Abatement Measures for Enteric Fermentation Q-k V-51
3-5 Abatement Measures for Manure Management V-54
3-6 Projected Global Livestock Populations by Species V-58
3-7 Regional Livestock Populations by Species, 2010 and 2030 V-60
XXII GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CONTENTS
3-8 Livestock Distribution by Intensity and Livestock Production System for India,
2010 (% of animals by species) V-62
3-9 Abatement Potential by Region at Selected Break-Even Prices in 2030 (MtCChe),
Baseline Production Case V-63
3-10 MAC Results and Differences from Constant Production Case for Baseline Number
of Animals Scenario V-66
3-11 MAC Results and Differences from Constant Production Case for No
Antimethanogen Scenario V-68
3-12 MAC Results and Differences from Constant Production Case for Combined
Baseline Number of Animals and No Antimethanogen Case V-69
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES XXIII
-------
EXECUTIVE SUMMARY
Executive Summar
reenhouse gases other than carbon dioxide (CCh) play an important role in the effort to
understand and address global climate change. Non-carbon dioxide (non-CCh) greenhouse
gases include methane (Q-k), nitrous oxide (N2O), and a number of high global warming
potential or fluorinated gases. The non-CCh greenhouse gases are more potent than CCh (per
unit weight) at trapping heat within the atmosphere and, once emitted, can remain in the atmosphere for
either shorter or longer periods of time than CCh. Approximately 30% of the anthropogenic greenhouse
effect since preindustrial times can be attributed to these non-CCh greenhouse gases (Intergovernmental
Panel for Climate Change [IPCC], 2001b); approximately 25% of GWP-weighted greenhouse gas
emissions in the year 2005 comprise the non-CCh greenhouse gases (U.S. Environmental Protection
Agency [USEPA], 2012).
Greenhouse gases are the primary driver of climate change, which can lead to hotter, longer heat
waves that threaten the health of the sick, poor, or elderly; increases in ground-level ozone pollution
linked to asthma and other respiratory illnesses; and other threats to human health and welfare. In some
cases, reducing non-CCh emissions can have a more rapid effect on the climate and be more cost-effective
than reducing CCh emissions. Given the important role that mitigation of non-CCh greenhouse gases can
play in climate strategies, there is a dear need for an improved understanding of the mitigation potential
for non-CCh sources, as well as for the incorporation of non-CCh greenhouse gas mitigation in climate
economic analyses. This report is a follow-on to the 2006 EPA report Global Mitigation of Non-COi
Greenhouse Gases and illustrates the abatement potential of non-CCh greenhouse gases through a
comprehensive global analysis and resulting data set of marginal abatement cost (MAC) curves.
The report provides a comprehensive global analysis and resulting data set of MACs that illustrate
the abatement potential of non-CCh greenhouse gases by sector and by region. This analysis incorporates
updated mitigation technologies, costs, and emissions baselines with an updated modeling approach. The
results of the analysis are MAC curves that reflect aggregated break-even prices for implementing
mitigation options in a given sector and region with more detail than available in the previous report.
This assessment of mitigation potential is unique because it is comprehensive across all non-CCh
greenhouse gases, across all emitting sectors of the economy, and across all regions of the world. The
MAC curves allow for improved understanding of the mitigation potential for non-CCh sources, as well
as inclusion of non-CCh greenhouse gas mitigation in economic modeling of multigas mitigation
strategies.
The basic methodology—a bottom-up, engineering cost approach—is the same as the methodology
followed in the 2006 report. Building on the baseline non-CCh emissions projections from the USEPA's
Global Anthropogenic Non-COi Greenhouse Gas Emissions: 1990-2030 (2012), this analysis applies mitigation
options to the emissions baseline in each sector. The technical abatement potential and cost are calculated
for each mitigation option across all the emitting greenhouse gas sectors. The average break-even price is
calculated for the estimated abatement potential for each mitigation option. The options are then ordered
in ascending order of break-even price (cost) and plotted against abatement potential. The resulting MAC
is a stepwise function; each point on the curve represents the break-even price point for a discrete
mitigation option (or defined bundle of mitigation strategies) and the associated abatement potential.
This report makes no explicit assumption about policies that would be required to facilitate and generate
adoption of mitigation options. Therefore, this report provides estimates of technical mitigation potential.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES ES-1
-------
EXECUTIVE SUMMARY
The results of this analysis are MAC curves that reflect the prices for implementing mitigation
options in a given sector and region. This report provides improved data to better understand the
mitigation potential for non-CCh sources and allows for inclusion of non-CCh greenhouse gas mitigation
approaches in economic modeling of multigas mitigation strategies. The MAC data sets can be
downloaded in spreadsheet format from the USEPA Web site at
http://www.epa.gov/dimatechange/EPAadivities/economics/nonco2mi tigation.html.
Mitigation of Non-CO2 Gases Can Play an Important Role in Climate Strategies. Worldwide, the
potential for cost-effective non-CO2 greenhouse gas abatement is significant. Figure ES-1 shows the global
total aggregate MAC for the year 2030. Without a price signal (i.e., at $0/tCChe), the global mitigation
potential is greater than 1,800 million metric tons of CO2 equivalent (MtCChe), or 12% of the baseline
emissions (refer to Section 1.3.3 of this report for a more detailed explanation of unrealized mitigation
potential in the MACs). As the break-even price rises, the mitigation potential grows. Significant
mitigation opportunities could be realized in the lower range of break-even prices. The global mitigation
potential at a price of $10/tCChe is greater than 3,000 MtCChe, or 20% of the baseline emissions, and
greater than 2,400 MtCChe or 24% of the baseline emissions at $20/tCChe. In the higher range of break-
even prices, the MAC becomes steeper, and less mitigation potential exists for each additional increase in
price.
Figure ES-1: Global Total Aggregate MAC for Non-C02 Greenhouse Gases in 2030
1,500 2,000 2,500 3,000 3,500 4,000 4,500
Non-CO2 Reductions (MtCO2e)
Globally, the Sectors with the Greatest Potential for Mitigation of Non-CO2 Greenhouse Gases are
the Energy and Agriculture Sectors. Figure ES-2 shows the global MACs by economic sector in 2030. At a
break-even price of $5/tCChe, the potential to reduce of non-CCh greenhouse gases is greater than 1,190
MtCChe in the energy sector and approximately 1,080 MtCChe in the industrial process sector. At a
break-even price of $30/tCO2e, the potential increases to approximately 1,475 MtCChe in the industrial
sector, nearly 1,400 MtCChe in the energy sector, and 500 and 332 MtCChe in the agriculture and waste
sectors, respectively.
ES-2
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
EXECUTIVE SUMMARY
Methane Mitigation has the Largest Potential across All the Non-CO2 Greenhouse Gases.
Figure ES-3 shows the global MACs by greenhouse gas type for 2030. At or below $0/tCO2e, the potential
for CH4 mitigation is greater than 1,000 MtCChe. The potential for reducing CH4 emissions grows to over
2,000 MtCChe as the break-even price rises from $0 to $30/tCO2e, while less than that of CH4, N2O, and F-
gases exhibit significant mitigation potential at or below SO/tCChe.
Figure ES-2: Global 2030 MACs for Non-C02 Greenhouse Gases by Major Sector
Waste
Agriculture
Energy
Industrial
800 1,000 1,200 1,400
1,600 1,800
Non-CO2 Reductions (MtCO2e)
•Nitrous Oxide
F-Gases
•Methane
Figure ES-3: Global 2030 MACs by Non-C02 Greenhouse Gas Type
$100
$80
$60
$40
$20
$0
-$20
-$40
2,000
2,500
Non-CO2 Reductions (MtCO2e)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
ES-3
-------
EXECUTIVE SUMMARY
Major Emitting Regions of the World Offer Large Potential Mitigation Opportunities. Figure ES-4
shows the global MACs by region for 2030. The United States and China are the top two contributors to
global mitigation potential with cost-effective mitigation of 260 and 200 MtCChe, respectively. The largest
sources of mitigation potential in these regions are oil/gas, refrigeration/ac, livestock, and coal. The EU,
India, and Brazil represent significant mitigation potential as well. At a break-even price of $30/tCO2e the
five largest emitting countries represent 46% of the global abatement potential.
Figure ES-4: Global 2030 MACs for Non-CC-2 Greenhouse Gases by Major Emitting Regions
Brazil
India
EU27
United States
China
Rest of World
2,500
3,000
Non-CO2 Reductions (MtCO2e)
The aggregate MACs by economic sector, greenhouse gas type, and region highlight the importance
of including non-CO2 greenhouse gases in the analysis of multigas climate strategies. The MACs illustrate
that a significant portion of this emissions reduction potential can be realized at zero or low carbon
prices. The mitigation potential in each economic sector is examined in greater detail in this report.
Intergovernmental Panel on Climate Change (IPCC). 2001b. Technical Summary: A Report Accepted by
Working Group I of the IPCC but not approved in detail. A product resulting from The Third Assessment
Report of Working Group I of the Intergovernmental Panel on Climate Change, January 2001.
Available at: www.ipcc.ch.
U.S. Environmental Protection Agency (USEPA). (2012a). Global Anthropogenic Non-COi Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA. Available at:
http://www.epa.gov/climatechange/economics/international.html.
ES-4
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
TECHNICAL SUMMARY
I. Technical Summa
1.1 Overview
he objective of this peer reviewed technical report is to provide a comprehensive and
consistent data set on global mitigation of noncarbon dioxide (non-CCh) greenhouse gases by
sector and by region. Mitigating emissions of non-CCh greenhouse gases can be relatively
inexpensive compared with mitigating CCh emissions. Thus, attention continues to focus on
incorporating international non-CCh greenhouse gas mitigation options into climate economic analyses.
This requires a large data collection effort and expert analysis of available technologies and opportunities
for greenhouse gas reductions across diverse regions and sectors.
This report is an update to the 2006 EPA report, Global Mitigation of Non-COi Greenhouse Gases, and
incorporates an updated modeling approach and new data on mitigation technologies, costs, and
emissions baselines. The basic methodology—a bottom-up, engineering cost approach—is the same as
was followed in the 2006 report, with some enhancements (as described in Section 1.3.4 of this report).
The results of this analysis are marginal abatement cost (MAC) curves. The end result of this report is a
set of marginal abatement curves (MACs) that allow for improved understanding of the mitigation
potential for non-CCh sources, as well as inclusion of non-CCh greenhouse gas mitigation in economic
modeling. The MAC data sets can be downloaded in spreadsheet format from the USEPA's Web site at
http://www.epa.gov/climatechange/EPAactivities/econornics/nonco2rni tigation.html.
Non-CO2 Greenhouse Gases
Greenhouse gases other than CCh play an important role in the effort to understand and address
global climate change. The non-CCh gases include methane (Q-k), nitrous oxide (N2O), and a number of
high global warming potential or fluorinated gases. The non-CCh greenhouse gases are more potent than
CCh (per unit weight) at trapping heat within the atmosphere and, once emitted, can remain in the
atmosphere for either shorter or longer periods of time than CCh. Figure 1-1 shows that these non-CCh
greenhouse gases are responsible for approximately 30 percent of the enhanced, anthropogenic
greenhouse effect since preindustrial times.
Table 1-1 shows the global total greenhouse gas emissions for the year 2010, broken down by sector
and by greenhouse gas type. The non-CCh gases constitute 28 percent of the global total greenhouse gas
emissions.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 1-1
-------
TECHNICAL SUMMARY
Figure 1-1:
Contribution of Anthropogenic Emissions of Greenhouse Gases to the Enhanced
Greenhouse Effect from Preindustrial to Present (measured in watts/meter2)
High-GWP Gases
0.7%
Source: Intergovernmental Panel on Climate Change (IPCC). 2007. Climate Change 2007: The Physical Science Basis. Contribution of
Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Solomon, S., D. Qin, M. Manning, Z.
Chen, M. Marquis, K.B. Averyt, M. Tignor and HI. Miller (eds.) Cambridge University Press, Cambridge, United Kingdom and New York, NY,
USA.
Table 1-1: Global Non-C02 Greenhouse Gas (GHG) Emissions for 2010 (MtC02e) by Source and Gas Type
Sectors
Agriculture
Energy
Industry
Waste
Global Total
Percentage of Global
Total Non-CC-2 GHGs
CH4
3,102
2,991
83
1,374
7,549
66%
N20
2,897
54
118
97
3,166
28%
F-Gases
—
—
672
—
672
6%
Global Total
Non-CC-2
Emissions
5,999
3,044
873
1,471
11,387
Percentage of
Global Total
Non-CC-2 GHGs
53%
27%
8%
13%
Source: USEPA. 2012. Global Anthropogenic Non-COi Greenhouse Gas Emissions: 1990-2030. EPA 430-S-12-006. USEPA: Washington
D.C. Available at: http://www.epa.aov/climatechange/EPAactivities/economics/nonco2proiections.html
1.2.1 Methane (CH4)
CH4 is about 21 times more powerful at warming the atmosphere than CCh over a 100-year period
(IPCC, 1996). In addition, CH/s chemical lifetime in the atmosphere is approximately 12 years, compared
with approximately 100 years for CCh. These two factors make Q-k a candidate for mitigating global
warming in the near term (i.e., within the next 25 years or so) or in the time frame during which
atmospheric concentrations of Q-k could respond to mitigation actions.
1-2
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
TECHNICAL SUMMARY
CH4 is emitted from a variety of manmade sources, including landfills, oil and natural gas systems,
agricultural activities, coal mining, stationary and mobile combustion, wastewater treatment, and certain
industrial processes. Q-k is also a primary constituent of natural gas and an important energy source. As
a result, efforts to prevent or capture and use Q-k emissions can provide significant energy, economic,
and environmental benefits.
1.2.2 Nitrous Oxide (N2O)
N2O is a dear, colorless gas with a slightly sweet odor. Because of its long atmospheric lifetime
(approximately 120 years) and heat-trapping effects—about 310 times more powerful than CCh on a per-
molecule basis—N2O is an important greenhouse gas.
N2O has both natural and manmade sources and is removed from the atmosphere mainly by
photolysis (i.e., breakdown by sunlight) in the stratosphere. In the United States, the main manmade
sources of N2O are agricultural soil management, livestock waste management, mobile and stationary
fossil fuel combustion, adipic acid production, and nitric acid production. N2O is also produced naturally
from a variety of biological sources in soil and water.
1.2.3 F-Gases Gases
There are three major groups or types of F-Gases gases: hydrofluorocarbons (HFCs),
perfluorocarbons (PFCs), and sulfur hexafluoride (SFe). These compounds are the most potent
greenhouse gases because of their large heat-trapping capacity and, in the cases of SFe and the PFCs, their
extremely long atmospheric lifetimes. Because some of these gases, once emitted, can remain in the
atmosphere for centuries, their accumulation is essentially irreversible. F-Gases gases are emitted from a
broad range of industrial sources; most of these gases have few (if any) natural sources.
HFCs
HFCs are manmade chemicals, many of which have been developed as alternatives to ozone-
depleting substances (ODSs) for industrial, commercial, and consumer products. The GWPs of HFCs
range from 140 (HFC-152a) to 11,700 (HFC-23). The atmospheric lifetime for HFCs varies from just over a
year (HFC-152a) to 260 years (HFC-23). Most of the commercially used HFCs have atmospheric lifetimes
of less than 15 years (for example, HFC-134a, which is used in automobile air-conditioning and
refrigeration, has an atmospheric lifetime of 14 years).
PFCs
Primary aluminum production, semiconductor manufacturing and flat panel display manufacturing
are the largest known manmade sources of tetrafluoromethane (CF4), and hexafluoroethane (CiFe). PFCs
are also relatively minor substitutes for ODSs. Over a 100-year period, CF4 and C2pe are, respectively,
6,500 and 9,200 times more effective than CO2 at trapping heat in the atmosphere.
Sulfur Hexaflouride (SF6)
The GWP of SFe is 23,900, making it the most potent greenhouse gas evaluated by IPCC. SFe is a
colorless, odorless, nontoxic, nonflammable gas with excellent dielectric properties. It is used (1) for
insulation and current interruption in electric power transmission and distribution equipment; (2) to
protect molten magnesium from oxidation and potentially violent burning in the magnesium industry;
(3) to create circuitry patterns and to dean vapor deposition chambers during manufacture of
semiconductors and flat panel displays; and (4) for a variety of smaller uses, induding uses as a tracer gas
and as a filler for sound-insulated windows.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 1-3
-------
TECHNICAL SUMMARY
Like the PFCs, SFe is very long lived, so all manmade sources contribute directly to its accumulation
in the atmosphere. Measurements of SFe show that its global average concentration increased by about 7
percent per year during the 1980s and 1990s, from less than 1 ppt in 1980 to almost 4 ppt in the late 1990s
(IPCC, 2001a).
1.2.4 Use of GWPs in this Report
The GWP compares the relative ability of each greenhouse gas to trap heat in the atmosphere during
a certain time frame. Per IPCC (1996) guidelines, CCh is the reference gas and thus has a GWP of 1. Based
on a time frame of 100 years, the GWP of CH4 is 21 and the GWP of N2O is 310. Table 1-2 lists all GWPs
used in this report to convert the non-CCh emissions into CCh-equivalent units. This report uses GWPs
from the 1996 IPCC Second Assessment Report (rather than the 2001 Third Assessment Report) because
these are the values specified by greenhouse gas reporting guidelines under the United Nations
Framework Convention on Climate Change.
Table 1-2: Global Warming Potentials
Gas
Carbon dioxide (C02)
Methane (CH4)
Nitrous oxide (N20)
HFC-23
HFC-32
HFC-125
HFC-134a
HFC-143a
HFC-152a
HFC-227ea
HFC-236fa
HFC-4310mee
CF4
C2F6
GWP*
1
21
310
11,700
650
2,800
1,300
3,800
140
2,900
6,300
1,300
6,500
9,200
C4Fio 7,000
C6Fi4 7,400
SF6
23,900
Source: IPCC, 1996.
a 100 year time horizon.
1.3 Methodology
This section describes the basic methodology used in this report to analyze potential emissions and
abatement of non-CCh greenhouse gases. The analysis builds on the approach presented in the 2006
Global Mitigation of Non-CCh Greenhouse Gases report (USEPA, 2006a). For the current analysis several
enhancements were made for the MAC analysis and these will be highlighted in the discussion that
follows. Primary enhancements include:
1-4 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
TECHNICAL SUMMARY
• Updating baseline emissions projections
• Disaggregating mitigation potential and costs to the country level for 195 countries
• Updating reduction efficiencies for individual measures by country
• Updating capital and operation and maintenance (O&M) costs for individual measures
• Segmenting O&M costs into labor, materials and energy components
• Developing international adjustments factors used to construct country specific abatement costs
and benefits
• Updating crop process model simulations of changes in crop yields and emissions associated
with rice cultivation and cropland soil management
MAC curves are constructed for each region and sector by estimating the carbon price at which the
present value benefits and costs for each mitigation option equilibrates. The methodology produces a
stepwise curve, where each point reflects the average price and reduction potential if a mitigation
technology were applied across the sector within a given region. In conjunction with appropriate baseline
and projected emissions for a given sector the results are expressed in terms of absolute reductions of
carbon dioxide equivalents (MtCChe).This section describes the components of our methodology.
First, we establish the baseline emissions for each sector as described in Section 1.3.1. Section 1.3.2
presents the methodology used to evaluate mitigation options, which involves calculating the abatement
potential and the breakeven price for each option. Lastly, we describe the construction of the MACs in
Section 1.3.3. Some sectors deviate from this methodology depending on specific circumstances, which are
briefly mentioned here and described in more detail in the sector-specific chapters.
The results of the analysis are presented as MACs by region and by sector and generally focus on the
2010 to 2030 time frame. Emissions abatement in the MACs is shown as both absolute emissions
reductions and as percentage reductions from the baseline. Non-CCh emissions sources analyzed in this
report are
• coal mining;
• oil and natural gas systems;
• solid waste management;
• wastewater;
• specialized industrial processes; and
• agriculture.
1.3.1 Baseline Emissions for Non-CO2 Greenhouse Gases
For consistency across regions and sectors the MAC Report analysis primarily uses the EPA report,
Global Anthropogenic Non-COi Greenhouse Gas Emissions: 1990-2030 for baseline emissions and projections.
The Global Emissions Report (GER) was published in December of 2012, and uses a combination of
country-prepared, publicly-available reports (UNFCCC National Communications) and IPCC Tier 1
methodologies to fill in missing or unavailable data. The basis for the U.S. historical emissions in the GER
is the U.S. Inventory of Greenhouse Gases and Sinks published in April of 2011. The methods used to
estimate and project non-CCh emissions in USEPA (2012) are briefly summarized here. In some cases,
particularly for agricultural emissions, it was necessary to develop separate baselines from which to
assess the mitigation analyses. For the agricultural sector, the baseline emissions used in this report were
based on crop process model simulations and livestock population data combined with projected crop
areas and livestock populations, respectively, from the International Food Policy Research Institute
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 1-5
-------
TECHNICAL SUMMARY
International Model for Policy Analysis of Agricultural Commodities and Trade (IMPACT) model. These
deviations are also explained in more detail in this report.
The preferred approach for estimating historical and projected emissions is to use country-prepared,
publicly-available reports. EPA applied an overarching methodology to estimate emissions across all
sectors, and deviations to this methodology are discussed in each of the source-specific methodology
sections of USEPA (2012). The following summary of the general methodology used to estimate global
non-CO2 emissions is replicated from the USEPA (2012) report.
Historical Emissions
For Annex I Countries (Al), the UNFCCC flexible query system (UNFCCC, 2012) provides emission
estimates for Al countries from Common Reporting Format (CRF) files, submitted with annual national
inventories. The full or partial time series of source disaggregated data is available for Al countries from
1990 through 2007. The time series is complete for the majority of sources; however there are gaps in the
time series for some countries and categories and data for missing years were supplemented. The
methodology used by each source to interpolate, backcast, or forecast depends on the availability of CRF
data and the distribution of that data over time. In general, the following methodology was applied to
interpolate, backcast, or forecast data:
• When two years are reported such that a year requiring an estimate (e.g., 1995) occurred between
the reported years (e.g., 1993 and 1997), EPA interpolates the missing estimate (1995) using
reported estimates.
• EPA backcasted or forecasted emission estimates to complete the historical series for 1990, 1995,
2000, and 2005 on a source by source basis. For each source, EPA used growth rates for available
activity data believed to best correlate with emissions (e.g., production, consumption). If either 1)
more than one type of activity data should be used, 2) the emission factor will vary over time, or
3) the relationship between the activity data and emissions is not linear (i.e., exponential), then
EPA used Tier 1 growth rates. This involves estimating emissions for 1990, 1995, 2000, and 2005
using a Tier 1 approach, then using the rate of growth of this emission estimate to backcast and
forecast the country-reported emissions.
• If a country-reported an estimate for an individual source for one year, but reported aggregate
estimates for other years, EPA disaggregated the estimates using the percent contribution of the
individual source in the latest reported year.
For Non-Annex I countries historical emissions data were available in the UNFCCC flexible query
system as well, but generally these reported data do not constitute a full time series. The methodology for
interpolating or backcasting missing historical data used by each source will follow the same general
guidelines outlined in the earlier in this section. Because the data for non-Al countries from the
UNFCCC flexible query system do not generally have a complete time series, it is likely that non-Al
sources will rely more heavily on Tier 1 calculated growth rates or activity data growth rates for
backcasting and forecasting emissions between 1990 and 2005.
Projected Emissions
Emission projections by source and country were obtained from National Communications (NCs)
reports. For Al countries, this refers to the Fifth NCs currently being released. For non-Al countries, EPA
reviewed the most recent NCs submitted to the UNFCCC.
If an NC had projections for a sector but not a source, EPA used the relative proportion of emissions
for the latest year of historical emissions to disaggregate projected emissions for a source. For example, if
France projected CH4 emissions from agriculture to 2030 but does specify what portion is from manure
1-6 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
TECHNICAL SUMMARY
management, EPA took the proportion of emissions that manure contributes to agriculture Q-k emissions
in France's 2007 GHG Inventory, assume this proportion remains constant for 2030, and apply this to the
2030 agriculture estimate.
If projections for a sector are not available from a NC, EPA used activity data drivers or Tier 1 growth
rates, specific to each source. The specific methodology followed by each source category is outlined in
each sector's methodology description.
For most countries, emissions and projections are not available for the sources of F-GHGs. Therefore,
EPA estimates F-GHG emissions and projections using detailed source methodologies described in
USEPA (2012).
Baseline Emissions for Agriculture
Although USEPA (2012) contains estimates of baseline emissions for agricultural sources, alternative
baselines were developed for the purposes of the mitigation report. The primary rationale was to ensure
consistency in the area, number of livestock head, production, and price projections used across the entire
agricultural sector. Projections provided by IFPRI from their IMPACT model of global agricultural
markets were used to adjust values for agricultural activities and associated emissions over time. In
addition, detailed process-based models—Daily Century (DAYCENT) for croplands and DeNitrification-
DeComposition (DNDC) for rice cultivation—were used for both the baseline emissions estimates and the
greenhouse gas implications of mitigation options, thus allowing for a dear identification of baseline
management conditions and consistent estimates of changes to those conditions through mitigation
activities. Emissions obtained using these detailed simulation models differ from those obtained in
USEPA (2012), which relied upon IPCC default emissions factors. For emissions associated with livestock,
the mitigation analysis in this report relies on projections similar to those used in USEPA (2012), but with
some differences due to the adjustments made for consistency with IFPRI IMPACT projections across all
agricultural sectors. The baseline emissions were also disaggregated by livestock production system and
intensity using data provided by the United Nations Food and Agriculture Organization (FAO). Further
details about the emissions baselines estimated by the DAYCENT and DNDC models, and their
relationship to USEPA (2012) estimates, are provided in Section V Agriculture of this report.
1.3.2 Mitigation Option Analysis Methodology
Mitigation options represented in the MACs of this report are applied to the baselines described in
Section IV.1.3.1. The mitigation option analysis throughout this report was conducted using a common
methodology and framework. This section outlines the basic methodology. The sector-specific chapters
describe the mitigation estimation methods in greater detail, including any necessary deviations from the
basic methodology.
The abatement analysis for all non-CCh gases for agriculture, coal mines, natural gas systems, oil
systems, landfills, wastewater treatment, and nitric and adipic acid production are based on USEPA ,
2006 and improve upon DeAngelo et al. (2006), Beach et al. (2008), Delhotal et al. (2006), and Ottinger et
al. (2006). These studies provided estimates of potential CH4 and N2O emissions reductions from major
emitting sectors and quantified costs and benefits of these reductions.
Given the detailed data available for U.S. sectors, the USEPA's U.S. analysis uses representative
facility estimates but then applies the estimates to a highly disaggregated and detailed set of emissions
sources for all the major sectors and subsectors. For example, the USEPA analysis of the natural gas sector
is based on more than 100 emissions sources in that industry, including gas well equipment, pipeline
compressors and equipment, and system upsets. Thus, the USEPA analysis provides significant detail at
the sector and subsector levels.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 1-7
-------
TECHNICAL SUMMARY
The analysis generally begins with developing sector level model facilities or units to which
mitigation options are applied. In many cases the model facilities, abatement costs and mitigation
potential are based on detailed US and EU inventory estimates, and then extrapolate to "model" facilities
for other countries. For some sectors, such as wastewater, landfills, and selected industrial sectors,
additional detail on international abatement options and costs are available and are incorporated into the
model.
A scaling factor is used to reconcile inventory data with the GER baseline emissions data. For the F-
Gases abatement analysis, natural gas and oil, and landfills sectors it is assumed that some mitigation
technologies are adopted to meet future regulations or voluntary industry reduction targets. Therefore,
some mitigation options are accounted for in the baseline emissions. If an option is assumed to be
adopted in the baseline, it is not included when generating the MAC. In addition, expert judgment
determines market shares for mitigation technologies competing for the same set of emissions (when
multiple options are available that are substitutes for each other).
The agricultural sector's emissions abatement analysis improves upon previous studies supported by
the USEPA (USEPA, 2006; DeAngelo et al., 2006; Beach et al., 2008) that generated MACs by major world
region for cropland N2O, livestock enteric CH4, manure management CH4, and rice cultivation CH4. The
most significant change in this report is the use of updated versions of the biophysical, process-based
models used in previous studies (i.e., DAYCENT and DNDC) applied at a more disaggregated spatial
scale to better capture the net greenhouse gas and yield effects and to capture the spatial and temporal
variability of those effects for the cropland and rice emissions baseline and mitigation scenarios. Use of
these process-based models is intended to show broad spatial and temporal baseline trends and broad
changes when mitigation scenarios are introduced, rather than to show definitive absolute emissions
numbers for specific locations. In addition, baseline emissions estimates have been updated and a larger
number of mitigation options are now assessed, particularly for rice cultivation (e.g., increased emphasis
on options that reduce N2O as well as CH4). Considerably greater disaggregation of the baseline by
production system has been incorporated to improve our ability to characterize technical applicability for
different types of livestock and cropping systems. More detailed results are provided for rice cultivation
under deepwater, upland, rainfed, and irrigated conditions, with separate calculations for alternative
irrigated water management strategies and for livestock management based on livestock production
system and management intensity.
Technical Characteristics of Abatement Options
The non-CO2 abatement options evaluated in this report are compiled from the studies mentioned
above, as well as from the literature relevant for each sector. For each region, either the entire set of
sector-specific options or the subset of options determined to be applicable is applied. Options are
omitted from individual regions on a case-by-case basis, using either expert knowledge of the region or
technical and physical factors (e.g., appropriate climate conditions). In addition, the share or extent of
applicability of an option within different regions may vary based on these conditions.
The selective omission of options represents a static view of the region's socioeconomic conditions. In
some instances the reduction efficiency of an option improves over time reflecting anticipated technology
advances. However, the applicability of options is held constant over time. Ideally, more detailed
information on country-specific conditions, technologies, and experiences will be available in the future,
which will enable more rigorous analyses of abatement option availability over time in each region. In
addition, the average technical lifetime of an option (in years), determined using expert knowledge of the
technology or recent literature, is held constant over time and across regions.
1-8 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
TECHNICAL SUMMARY
Table 1-3 summarizes how the potential emission reduction is calculated for each of the available
abatement options. First the technical effectiveness of each option is calculated by multiplying the options
technical applicability by its market share by its reduction efficiency. This yields the percentage of
baseline emissions that can be reduced at the national or regional level by a given option. This is then
applied to the Emissions stream (MtCChe) to which the option is applied to yield the emissions
reductions for the mitigation option.
Table 1-3: Calculation of Potential Emission Reduction for an Abatement Option
Technical
Applicability
(%)
Percentage of
total baseline
emissions from
a particular
emissions
source to which
a given option
can be
potentially
applied.
Market
X Share3 (%)
Percentage of
technically
applicable
baseline
emissions to
which a given
option is
applied;
avoids double
counting
among
competing
options
Reduction
Efficiency
X (%)
Percentage of
technically
achievable
emissions
abatement
for an option
after it is
applied to a
given
emissions
stream
Technical
Effectiveness
Technical
Effectiveness
(%)
Percentage of
baseline
emissions that
can be
reduced at the
national or
regional level by
a given
option.
^^M
Baseline Unit
X Emissions
(MtC02e)
Emissions
stream to
which the
option is
applied
|^^|
Unit
= Emission
Reduction
(MtC02e)
Unit
emission
reductions
a Implied market share non competing options (i.e., only one options is applicable for an emissions streams) is assumed to add to 100 percent
Technical applicability accounts for the portion of emissions from a facility or region that a mitigation
option could feasibly reduce based on its application. For example, if an option applies only to the
underground portion of emissions from coal mining, then the technical applicability for the option would
be the percentage of emissions from underground mining relative to total emissions from coal mining.
The implied market share of an option is a mathematical adjustment for other qualitative factors that
may influence the effectiveness or adoption of a mitigation option. For certain energy, waste, and
agriculture sectors, it was outside the scope of this analysis to account for adoption feasibility, such as
social acceptance and alternative permutations in the sequencing of adoption. For example, if n
competing (overlapping) mitigation options are available for a single emissions stream, the implied
market share of each of the n overlapping options is equal to l/n. This avoids cumulative reductions of
greater than 100 percent across options. Given the lack of region-specific data for determining the relative
level of diffusion among options that could compete for the same emissions stream, we applied this
conservative adjustment. An example of overlapping options is the sequencing of cropland mitigation
options, where the adoption of one option (e.g., conversion to no tillage) affects the effectiveness of
subsequent options (e.g., reduced fertilizer applications). While this describes the basic application of the
implied adoption rate in the energy, waste, and agriculture sectors, this factor is informed by expert
insight into the potential market penetration over time in the industrial processes sector. For sectors such
as landfills, where market share assumptions are available, customized shares that sum to one are used
instead of l/n.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
1-9
-------
TECHNICAL SUMMARY
When nonoverlapping options are applied, they affect 100 percent of baseline emissions from the
relevant source. Examples of two nonoverlapping options in the natural gas system are inspection and
maintenance of compressors and replacement of distribution pipes. These options are applied
independently to different parts of the sector and do not compete for the same emissions stream.
The reduction efficiency of a mitigation option is the percentage reduction achieved with adoption.
The reduction efficiency is applied to the relevant baseline emissions as defined by technical applicability
and adoption effectiveness. Most abatement options, when adopted, reduce an emissions stream less than
100 percent. If multiple options are available for the same component, the total reduction for that
component is less than 100 percent.
Once the technical effectiveness of an option is calculated as described above, this percentages
multiplied by the baseline emissions for each sector and region to calculate the absolute amount of
emissions reduced by employing the option. The absolute amount of baseline emissions reduced by an
option in a given year is expressed in million metric tons of CCh equivalent (MtCCheq).1
If the options are assumed to be technically feasible in a given region, the options are assumed to be
implemented immediately, Furthermore, once options are adopted, they are assumed to remain in place
for the duration of the analysis, and an option's parameters are not changed over its lifetime.
Economic Characteristics of Abatement Options
Each abatement option is characterized in terms of its costs and benefits per an abated unit of gas
(tCCheq or tons of emitted gas [e.g., tQ-k]). The benefits include a carbon value/price expressed as
$/tCO-2e. The carbon price at which an option's benefits equal the costs is referred to as the option's
breakeven price.
For each mitigation option, the carbon price (P) at which that option becomes economically viable is
calculated using the equation below (i.e., where the present value of the benefits of the option equals the
present value of the costs of implementing the option). A present value analysis of each option is used to
determine breakeven abatement costs in a given region. Breakeven calculations are independent of the
year the mitigation option is implemented but are contingent on the life expectancy of the option. The net
present value calculation solves for breakeven price P, by equating the present value of the benefits with
the present value of the costs of the mitigation option. More specifically,
Y F(l - TRXP • ER + fi) + TB\ _ Y T(l - 7fi)fiCl
Net Present Value Benefits Net Present Value Costs
\ __ I I
where
P = the breakeven price of the option ($/tCChe);
ER = the emissions reduction achieved by the technology (MtCChe);
R = the revenue generated from energy production (scaled based on regional energy prices) or
sales of by-products of abatement (e.g., compost) or change in agricultural commodity
prices ($);
T = the option lifetime (years);
DR = the selected discount rate i
1 One MtCCteeq equals 1 teragram of CCte equivalent (TgCCteeq); 1 metric ton = 1,000 kg = 1.102 short tons = 2,205 Ibs.
1-10 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
TECHNICAL SUMMARY
CC = the one-time capital cost of the option ($);
RC = the recurring (O&M) cost of the option (portions of which may be scaled based on regional
labor and materials costs) ($/year);
TR = the tax rate (%); and
TB = annual tax benefit of depreciation = f—J • TR.
Assuming that the emissions reduction ER, the recurring costs RC, and the revenue generated R do
not change on an annual basis, then we can rearrange this equation to solve for the breakeven price P of
the option for a given year:
CC RC R CC TR
P = 1 + —- —-
-TV} FK VT l ER ER ER'T (I-™)
i t\) en Zj£=l/"-^ _|_ nrAt
Costs include capital or one-time costs and operation and maintenance (O&M) or recurring costs.
Most of the agricultural sector options, such as changes in management practices, do not have applicable
capital costs, with the exception of anaerobic digesters for manure management.
Benefits or revenues from employing an abatement option can include (1) the intrinsic value of the
recovered gas (e.g., the value of Q-k either as natural gas or as electricity/heat, the value of HFC-134a as a
refrigerant), (2) nongreenhouse gas benefits of abatement options (e.g., compost or digestate for waste
diversion options, increases in crop yields), and (3) the value of abating the gas given a greenhouse gas
price in terms of dollars per tCCh eq ($/tCCheq) or dollars per metric ton of gas (e.g., $/tCH4,
$/tHFC-134a). In most cases, there are two price signals for the abatement of Q-k: one price based on
CH/s value as energy (because natural gas is 95 percent Q-k) and one price based on CH4's value as a
greenhouse gas. All cost and benefit values are expressed in constant year 2010 U.S. dollars. This analysis
is conducted using a 10 percent discount rate and a 40 percent tax rate. For quick reference, Table 1-4 lists
the basic financial assumptions used throughout this report.
Table 1-4: Financial Assumptions in Breakeven Price Calculations for Abatement Options
Economic Parameter Assumption
Discount Rate 10%
Tax Rate 40%
Constant Year Dollars 2010$
International Adjustment Factors
Costs and benefits of abatement options are adjusted to reflect regional prices. Wages and prices will
vary by country. Hence recurring O&M costs are segmented into labor, energy and materials costs.
Material costs components range from materials and supplies in the in the industrial and energy sectors,
to fertilizer costs in the agricultural sectors - all of which are likely to vary by region. One-time capital
costs are assumed to relatively stable across regions and not adjusted from country to country.
For some options data were available on the relative cost shares between labor, energy and materials.
For instance, in coal mining, different technologies have different cost shares which were developed
based on expert judgment. For options without detailed cost breakouts, the shares are generally assigned
evenly as 33% each to labor, energy, and materials. For the agricultural sector, labor, energy, water and
other input costs are calculated from their shares of agricultural production costs based on social
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 1-11
-------
TECHNICAL SUMMARY
accounting matrix (SAM) data from the Global Trade Analysis Project (GTAP) v8 database and
agricultural wage data from the International Food Policy Research Institute (IFPRI).
In regions where there is a lack of detailed revenue (benefits) data, revenues are scaled based on the
ratio between average prices of natural gas (when Q-k is abated and sold as natural gas) or of electricity
(when Q-k is used to generate electricity or heat) in a given region and in the United States. Similarly,
revenues from non-Q-k benefits of abatement options are scaled based on the ratio between the GDPs per
capita in a given region and in the United States. In the agricultural sector, changes in revenue occur as a
change in either crop yield or livestock productivity. Data on changes in crop yield or livestock
productivity are combined with data on regional producer prices for the relevant agricultural commodity
to calculate revenue changes.
Table 1-5 lists the international economic adjustment factors for selected countries. Using publically
available data on country-specific wage rates and energy prices, along with input from previous MAC
analysis, indices reflecting each country's wage rates and prices relative to the United States were created.
Adjustment Factors were created for labor, natural gas, electricity, coal and material costs. When data
was not available for a country, the country was either mapped to a similar country (with data) or
previously developed EMF factors were used.
Table 1-5: International Economic Adjustment Factors for Selected Countries
Country
Afghanistan
Brazil
Congo
China
India
Madagascar
Mexico
Norway
Poland
Russian Federation
Switzerland
United States
Uzbekistan
|E]*T*j^^l
0.02
0.24
0.19
0.04
0.03
0.19
0.12
1.80
0.26
0.12
1.35
1.00
0.12
0.75
1.30
1.06
0.62
0.67
1.06
1.04
1.62
0.98
0.19
1.62
1.00
0.19
Electricity"
1.30
1.60
0.34
0.63
1.69
0.34
1.42
0.77
1.19
0.56
1.41
1.00
0.38
Coalb
0.89
0.76
0.37
0.68
0.69
0.37
0.94
2.57
1.25
0.67
2.04
1.00
0.19
Materials0
0.01
0.13
0.05
0.07
0.02
0.01
0.20
1.61
0.24
0.15
1.30
1.00
0.02
aWage data was obtained primarily from U.S Bureau of Labor Statistics's International Labor Comparisons (BLS, 2010) and augmented with
(BLS, 2010b), (BLS, 2010c) and (FSSS.2010).
bEnergy Prices were obtained from ElA's International Energy Statistics (EIA, 201 Ob).
"Material factors were based on GDP/Capita statistics obtained from UNCTAD Statistical Database (UNCTAD, 2012).
Note that breakeven price calculations for this analysis do not include transaction or monitoring and
reporting costs, because there are no explicit assumptions in this report about policies that would
encourage and facilitate adoption of the mitigation options. Refer to Section 1.5 for a more complete
discussion of the limitations of this analysis.
1-12 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
TECHNICAL SUMMARY
1.3.3 Marginal Abatement Cost Curves
MACs are used to show the amount of emissions reduction potential at varying carbon price levels.
In theory, a MAC illustrates the cost of abating each additional ton of emissions. Figure 1-2 shows an
illustrative MAC. The x-axis shows the amount of emissions abatement in MtCCheq, and the y-axis shows
the breakeven price in $/tCCheq required to achieve the level of abatement. Therefore, moving along the
curve from left to right, the lowest cost abatement options are adopted first.
Figure 1-2: Illustrative Non-C02 Marginal Abatement Curve
Value of C02
Equivalent
($/tC02eq)
Market Price
$0/tC02eq
Total Abatement Potential
Energy/Commodity
Prices
Abated GHG Emissions (MtC02eq)
The curve becomes vertical at the point of maximum total abatement potential, which is the sum of
all technically feasible abatement options in a sector or region. At this point no additional price signals
from GHG credit markets could motivate emissions reductions; any additional emissions reductions
(shifting the vertical axis to the right) are due to increased energy efficiencies, conservation of production
materials, or both.
The points on the MAC that appear at or below the zero cost line ($0/tCCheq) illustrate potentially
profitable mitigation options. These "below-the-line" amounts represent mitigation options that are
already cost-effective given the costs and benefits considered (and are sometimes referred to as "no-
regret" options) yet have not been implemented. However, there may be nonmonetary barriers that are
preventing their adoption.
The MACs in this report are constructed from bottom-up average breakeven price calculations. The
average breakeven price is calculated for the estimated abatement potential for each mitigation option
(see Section I.). The options are then ordered in ascending order of breakeven price (cost) and plotted
against abatement potential. The resulting MAC is a stepwise function, rather than a smooth curve, as
seen in the illustrative MAC (Figure 1-2), because each point on the curve represents the breakeven price
point for a discrete mitigation option (or defined bundle of mitigation strategies).
Conceptually, marginal costs are the incremental costs of an additional unit of abatement. However,
the abatement cost curves developed here reflect the incremental costs of adopting the next cost-effective
mitigation option. We estimated the costs and benefits associated with all or nothing adoption of each
well-defined mitigation practice. We did not estimate the marginal costs of incremental changes within
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
1-13
-------
TECHNICAL SUMMARY
each practice (e.g., the net cost associated with an incremental change in paddy rice irrigation). Instead,
the MACs developed in this report reflect the average net cost of each option for the achieved reduction -
hence the noncontinuous, stepwise nature of the curve.
In the energy and waste sectors, representative facilities facing varied mitigation costs employ
mitigation technologies based on the lowest average breakeven option price. In calculating the abatement
potential, options are evaluated according to whether they are complements or substitutes. If a group of
options are complements (or independent of one another), the implied market shares are all equal to one.
If options are substitutes for each other, then market shares that sum to one are used to distribute
adoption across the available options (see table 1.3). In some instances, the lowest price option is selected
for each representative facility. When limited information is available, the market share is evenly
distributed (1/n) across all viable options. In this way, the implied adoption rate for each technology is
estimated.
In the industrial processes sector, mitigation options are applied to representative facilities, in order
of lowest average breakeven price to highest average breakeven price. Each option is applied to a portion
of the baseline emissions based on the implied adoption rate (the market share factor, as described in
Section 1.3.2.2), which, in the industrial sector, is informed by expert insight into potential adoption rates
of various mitigation technologies.
In the agriculture sector, mitigation options are applied to the portion of emissions where they are
technically applicable (e.g., anaerobic digesters are assumed to be applicable only in intensively managed
dairy and hog production systems). The implied market share for competing options is based purely on
the number of available migration options (n) that are applicable to a given subset of emissions and that
reduce emissions2 (1/n), where each option is applied to an equal portion of the cropland base or
livestock population and, thus regional baseline emissions, for each region over time. Given the existence
of nonprice and implementation factors that influence market share and the lack of accurate and detailed
information regarding these qualitative characteristics, we assume an even distribution of options across
the relevant baseline for the agriculture sector. This approach allows options to share a portion of market
penetration, regardless of their cost-effectiveness, rather than allowing only the least-cost option to
completely dominate the market. Our methodology is more conservative than if we had assumed only
price factors exist, thus allowing the least-cost option to penetrate the sector by 100 percent.
The MACs represent the average economic potential of mitigation technologies in that sector, because
it is assumed that if a mitigation technology is technically feasible in a given region, then it is
implemented according to the relevant economic conditions. Therefore, the MACs do not represent the
market potential or the social acceptance of a technology. The models used in the analysis are static (i.e.,
they do not represent adoption of mitigation technologies over time). This analysis assumes partial
equilibrium conditions that do not represent economic feedbacks from the input or output markets. This
analysis makes no assumptions regarding a policy environment that might encourage the
implementation of mitigation options. Additional discussion of some key limitations of the methodology
is provided in Section 1.5.
2 Some agricultural mitigation options may increase emissions under certain conditions depending on baseline
regional management and soil, climate, and other considerations. In addition, there are many mitigation options that
increase emissions per head of livestock or per hectare of land, but reduce emissions intensity per unit of output.
Thus, agricultural MACs are calculated both assuming constant production and constant area/head of livestock to
present a range of potential mitigation. The options that provide net emissions reductions may differ between these
alternative methods of MAC generation.
1-14 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
TECHNICAL SUMMARY
The end result of this analysis is a tabular data set for the MACs by sector, gas, and region, which are
presented in Appendix A.3 Sectoral MACs are aggregated by gas and by region to create global MACs,
which are presented in Section 1.4.2.
1.3.4 Methodological Enhancements from Analysis
This report builds on a study previously conducted by the USEPA for Stanford's EMF-21 and the
USEPA (2006) report. The EMF-21 focused specifically on multigas strategies and the incorporation of
non-CO2 greenhouse gas data sets into economic models. Although this analysis is built largely on the
previous USEPA analysis for the EMF-21, we have made several key enhancements.
New mitigation options have been added to the analysis for coal mining, agriculture, natural gas and
oil systems sectors. This report also presents MAC curves for the domestic wastewater sector, flat panel
display production, and photovoltaic cell production, which were not available in the previous report.
For industrial sources of fluorinated gases, the emissions baselines have been updated since the EMF-
21 analysis. In addition, the MACs for aluminum manufacturing and electrical power systems have been
enhanced with additional data.
The emissions baselines in the ODS substitute sector have also been enhanced. The EMF-21 ODS
substitute baseline was an average between baselines derived by the USEPA and ECOFYS. For this
report, the USEPA has generated an updated baseline. Assumptions in the ODS substitute sector, such as
the market penetration potential of various mitigation options, have been updated from the EMF-21
analysis based on the input of industry experts.
In the agricultural sector, the previous methodology is improved on for this analysis by using
updated versions of the biophysical, process-based models DAYCENT and DNDC that are utilized at a
more spatially disaggregated level and with a more disaggregated set of baseline management types to
which these options can be applied. These models capture the net greenhouse gas effects of the cropland
and rice baseline emissions and mitigation options, and they reflect the heterogeneous emissions and
yield effects of adopting mitigation practices. In addition, new agricultural mitigation options are now
assessed, and more detailed results are provided for alternative baseline crop and livestock management
practices.
Global total non-CCfe greenhouse gas baseline emissions in 2010 are estimated at 11,389 MtCChe, and
projected to increase 33% by 2030, totaling 15,157 MtCChe. Non-CCh anthropogenic emissions come from
four major emitting sectors: the energy, waste management, industrial processes, and agricultural
industries. China, United States, Russia, India and Brazil are the 5 largest country emitters and account
for 40% of total emissions.
This section presents the projected baseline emissions for non-CCh anthropogenic greenhouse gases
and provides a global overview of the MAC analysis results by sector and top emitting countries and
regions from 2010-2030. The gases represented in the analysis are CH4, N20, and F-Gases4, which are
3 Tables are presented that provide the percentage abatement for a series of breakeven prices. The MAC data are
presented as tables so that exact values can be determined for use in modeling activities.
4 F-Gases include fluorinated gases used as substitutes for Ozone Depleting Substances (ODS) and High-GWP gases
from industrial processes (PFC, HFC-23, SFe).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 1-15
-------
TECHNICAL SUMMARY
emitted from four major sectors: the agricultural, energy, waste, and industrial processes industries.
China, the United States, the European Union, Brazil, and Russia are the world's five largest emitting
countries as of 2010, accounting for 46 percent of total non-CCh emissions.
The data are aggregated in this chapter and provide a summary of all emitting sources and non-CCh
greenhouse gases. The individual chapters are organized by source and present the full details of these
analyses. For a complete set of mitigation potential by sector, gas, and region, refer to Appendix A.
Baseline projections presented in this section come from the Global Anthropogenic Non-COi Greenhouse
Gas Emissions: 1990-2030 (USEPA, 2012). Since its publication there have been some minor revisions to the
baseline projections for the industrial processes photovoltaic (PV) and flat panel display (FPD)
manufacturing. The totals presented in this report will differ slightly from the projections in the 2012
report.
1.4.1 Baselines
By Non-CO2 Greenhouse Gas
Figure 1-3 illustrates the relative share of each non-CCh greenhouse gas that comprises the global
baseline emissions total. CH4 represents the largest share of emissions worldwide, accounting for
approximately 66% of the total non-CCh greenhouse gas emissions in 2010, while N2O and F-Gases
account for the 28 percent and 6 percent, respectively.
Figure 1-3: Percentage Share of Global Non-C02 Emissions3 by Type of Gas in 2010
World Total = 11,389 MtCO2e
Source: USEPA, 2012.
a C02 equivalency based on 100-year GWP.
Figure 1-4 presents the projected baseline emissions by greenhouse gas for 2010, 2020, and 2030. F-
Gases represent the most significant change in baseline emissions. Accordingly to Figure 1-4, high GWPs
are to increase nearly 300 percent between 2010 and 2030. CH4 and N2O observe a more modest increase
at an average decadal rate of roughly 10 percent. As a result, F-Gases are projected to gain 15 percent of
the total share of non-CCh greenhouse gas emission by 2030, up from 4 percent in 2010
1-16
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
TECHNICAL SUMMARY
Figure 1-4: Non-C02 Global Emissions Forecast to 2030 by Greenhouse Gas
High GWPs
IN2O
ICH4
2010
2030
Source: USEPA, 2012.
By Major Emitting Sectors and Countries
The sources of non-CCh emissions are categorized into four major emissions sectors: energy, waste,
industrial processes, and agriculture. Figures 1-5 and 1-6 provide the projected global baseline emission
for 2010, 2020, 2030, by major emissions sector and by major emitting region, respectively. The agriculture
sector includes soil and manure management, rice cultivation, enteric fermentation, and other
nonindustrial sources such as biomass burning. Emissions sources categorized in the energy sector
include coal mining activities, natural gas transmission and distribution, and gas and oil production. The
waste sector includes municipal solid waste management, as well as human sewage and other types of
wastewater treatment. The industrial processes sector includes a wide range of activities, such as
semiconductor manufacturing, primary aluminum production, and electricity transmission and
distribution.
Agriculture is the primary source of non-CCh greenhouse gas emissions, accounting for 45 percent of
the total 2010 baseline. Energy holds the second largest share of non-CCh emissions, representing 23
percent of the baseline. The waste and industrial processes sectors represent 11 and 7 percent,
respectively. This trend will change through 2030, however, as emissions from the industrial processes
sector is projected to increase by more than double, and will therefore produce more non-CCh emissions
that waste by 2030.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
1-17
-------
TECHNICAL SUMMARY
Figure 1-5: Global Emissions by Major Sector for all Non-C02 Greenhouse Gases
I Waste
Industrial Processes
I Energy
I Agriculture
2010
2030
Source: USEPA, 2012.
Figure I-6:
Projected World Emissions Baseline for Non-C02 Greenhouse Gases, Including Top Emitting
Regions
Brazil
Russia
India
Central &S. America
Europe
I United States
I Asia
China
I Africa
I Rest of World
2010
2020
Year
2030
Source: USEPA, 2012.
Figure 1-6 shows the projected emissions baselines for the world, as well as the largest emitting
countries. The largest non-CCh emitting countries are typically characterized as mature, highly
industrialized countries or countries with significant agricultural industries. In 2010, the top five emitting
countries - China, the United States, EU-15, Brazil, and Russia - account for 44% of the world's total non-
1-18
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
TECHNICAL SUMMARY
CO2 emissions. Although 2010's top five emitting countries is projected to change during the next 20
years, their relative contribution to the world baseline will likely remain constant through 2030.
1.4.2 Global MACs
The MAC analysis methodology described in Section 1.3 of this report develop bottom-up projections
of potential reduction in non-CCh emissions in terms of the break-even price ($/tCChe). The emission
reduction potential is constrained by the limitations of the technologies considered in the analysis, as well
as regional and geographical applicability. In this report, MACs are developed for each major source by
sector and country. The resulting series of MACs are aggregated up across sectors, gases and regions. The
MACs indicate the potential reduction in non-CCh gas emissions for a given breakeven price. Figure 1-7
presented the results from the MAC analysis for 2030 by major economic sector. Figure 1-8 presents
aggregate MACs by greenhouse gas type for 2030. Figure 1-9 presents the 2030 MACs for the world's
largest non-CCh greenhouse gas emitting regions.
Figure 1-7: Global 2030 MACs by Non-C02 Greenhouse Gas
o
u
Waste
Agriculture
Energy
Industrial
800 1,000 1,200 1,400 1,600 1,800
Non-CO2 Reductions (MtCO2e)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
1-19
-------
TECHNICAL SUMMARY
Figure 1-8: Global 2030 MACs for Non-C02 Greenhouse Gases by Major Emitting Regions
$100
$80
$60
«„, $40
O
u
^ $20
$0
-$20
-$40
•Nitrous Oxide
F-Gases
•Methane
Non-CO2 Reductions (MtCO2e)
2,000
2,500
Figure 1-9: Global 2030 MACs for Non-CC-2 Greenhouse Gases by Major Emitting Regions
Brazil
India
EU27
United States
China
Rest of World
Non-CO2 Reductions (MtCO2e)
2,500
3,000
1.5
Limitations and Uncertainties
The results of this analysis cover the major emitting regions, emissions sources, and abatement
options; we discuss a few limitations of this analysis briefly below.
1-20
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
TECHNICAL SUMMARY
1.5.1 Exclusion of Transaction Costs
Ongoing work in the area of mitigation costs continues focus on including transactions costs. As
discussed in the 2006 version of this report, Lawrence Berkeley National Laboratory (LBNL), Assessing
Transaction Costs of Project-based Greenhouse Gas Emissions Trading (Antinori and Sathaye 2007), which
reported that transactions costs range between $0.03 per metric ton of carbon dioxide for large projects to
$4.05 per ton of carbon dioxide for smaller projects, with a weighted average of $0.36 per tonne of carbon
dioxide for a suite of projects considered. More recent MAC work by others (Rose, et al. 2013) estimated
the unit cost of an abatement project increases by an average of 15% when transaction costs are included.
Transaction costs vary significantly, contingent on the size of the project, the applicable mitigation
technology, and other factors. Given the lack of comprehensive data, this analysis continues to exclude
transaction costs from the analysis.
1.5.2 Static Approach to Abatement Assessment
This analysis does not account for the technological change in such option characteristics as
availability, reduction efficiency, applicability, and costs. For example, the same sets of options are
applied in 2010 and 2030 and an option's parameters are not changed over its lifetime. This current
limitation likely underestimates abatement potential because technologies generally improve over time
and costs fall. The introduction of a dynamic approach to assessing regional abatement potentials
requires additional assumptions about rates of technological progress and better baseline projections,
that, once incorporated into this analysis, will yield a better representation of how MACs change over
space and time. Developing more dynamic MACs to capture the impacts of technological change should
be included in any future MAC development.
1.5.3 Limited Use of Regional Data
The analytic framework used in this study is flexible enough to incorporate regional differences in all
the characteristics of abatement options. However, a lack of country-specific data led to a reliance on
expert judgment, as noted in the sector-specific chapters. This expert judgment was obtained from
source-level technical experts in government and industry with knowledge of project-level technologies,
costs, and specific regional conditions. Applicability of abatement options, for example, is reliant on
expert judgment, because the makeup of the current infrastructure in a given country in a given sector is
uncertain. A much greater use of data originating from local experts and organizations is recommended
for the follow-up research of CH4 abatement in countries outside the United States and EU. Incorporating
more regional data could also enhance the range of emissions sources and mitigation options addressed
in this analysis.
1.5.4 Exclusion of Indirect Emissions Reductions
This analysis does not account for indirect emissions reductions, which can result from either the
substitution of electricity from the grid, with electricity produced on-site from recovered CH4, or from the
substitution of natural gas in pipelines with recovered CH4. Calculation of such indirect reductions
requires additional assumptions about the carbon intensity of electricity in different regions. In the U.S.
landfill sector, indirect reductions generally augment emissions reductions by about 15 percent. In the
agricultural sector, although some mitigation options primarily target a single gas, implementation of the
mitigation options will have multiple greenhouse gas effects, most of which are reflected in the
agricultural results.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 1-21
-------
TECHNICAL SUMMARY
References
Antinori, C. and J. Sathaye 2007. Assessing Transaction Costs of Project-based Greenhouse Gas Emissions
Trading. Lawrence Berkeley National Laboratory (LBNL). Report No. LBNL-57315.
Intergovernmental Panel on Climate Change (IPCC). 2007. Climate Change 2007: The Physical Science Basis.
Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change. Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L.
Miller (eds.) Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.
Intergovernmental Panel for Climate Change (IPPC). 1996. IPCC Guidelines for National Greenhouse Gas
Inventories. Three volumes: Vol. 1, Reporting Instructions; Vol. 2, Workbook; Vol. 3, Reference
Manual. Paris, France: Intergovernmental Panel on Climate Change, United Nations Environmental
Programme, Organization for Economic Co-Operation and Development, International Energy
Agency.
Rose, S.K., R. Beach, K. Calvin, B. McCarl, J. Petrusa, B. Sohngen, R. Youngman, A. Diamant, F. de la
Chesnaye, J. Edmonds, R. Rosenzweig, and M. Wise, 2013. Estimating Global Greenhouse Gas Emissions
Offset Supplies: Accounting for Investment Risks and Other Market Realties. Electric Power Research
Institute [EPRI], Palo Alto, CA. Report 1025510.
Russian Federation Federal State Statistics Service (FSSS), 2010. Main Indicators: Living Standards. Table
7.7: Average Monthly Nominal Accrued Wages of Employees of Organisations by Kinds of Economic
Activities. 2010. http://www.gks.ru/bgd/regl/blO 12/IssWWW.exe/stg/d01/07-07.htm
United Nations Conference on Trade and Development (UNCTAD). 2012. National Accounts: Nominal and
Real GDP, total and per capita, annual 1970-2011. UNCTAD Statistical Database. Available at:
http://unctadstat.unctad.org.
U.S. Bureau of Labor Statistics (BLS), 2010a. International Comparisons of Hourly Compensation Costs in
Manufacturing, 1996-2008. Table 8: Production Workers: Hourly Compensation Costs in U.S. Dollars
in Manufacturing, 33 countries or areas and selected economic groups, 1975-2008; and Table 2: All
Employees: Hourly Compensation Costs in U.S. Dollars in Manufacturing, 33 countries or areas and
selected economic groups, 1996-2008 Released August 26, 2010. http://www.bls.gov/fls/.
U.S. Bureau of Labor Statistics (BLS), 2010b. "Labor costs in India's organized manufacturing sector."
Table 2: Hourly compensation costs in India's organized manufacturing sector, 1999-2005. May 2010.
http://www.bls.gov/opub/mlr/2010/05/artlfull.pdf.
U.S. Bureau of Labor Statistics (BLS), 2010c. "Manufacturing in China." Table 1: Hourly compensation
costs of manufacturing workers in China, 2002-2008." October 27, 2010.
http://www.bls.gov/fls/china.htmtfdata comparability.
U.S. Environmental Protection Agency (USEPA). 2012. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA.
U.S. Department of Energy's Energy Information Administration (EIA), 2010a. International Energy
Statistics. Electricity Prices for Industry. June 10, 2010.
http://www.eia.gov/emeu/international/elecprii.html
U.S. Department of Energy's Energy Information Administration (EIA), 2010b. International Energy
Statistics. Natural Gas Prices for Industry. June 10, 2010.
http://www.eia.gov/emeu/international/ngasprii.html.
1-22 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
-------
COALMINING
11.1.1
Sector Summary
(oal mining is a significant source of anthropogenic GHG emissions. Coal is an important
energy resource in many of the world's economies, used for energy generation or as a
feedstock in industrial production processes. Extracting this energy resource through
underground and surface mining releases methane (Cl-k) stored in the coal bed and surrounding geologic
strata. The U.S. Energy Information Administration's (USEIA's) (2011) most recent international energy
outlook projects a 39% increase in coal production between 2010 and 2035, reflecting continued economic
and industrial development of the world's emerging economies. In the absence of widespread adoption
of abatement measures by the coal mining sector, expanding coal production to meet growing energy
demands will subsequently lead to increases in anthropogenic emissions.
Worldwide, the coal mining industry liberated more than 589 million metric tons of carbon dioxide
equivalents (MtCO2e), which accounted for 8% of total anthropogenic Q-k1 emissions in 2010. The top 5
emitting countries of China, the United States, Russia, Australia, and Ukraine account for more than 80%
of coal mining CH4 emissions. Figure 1-1 summarizes the business-as-usual (BAU) emission baselines for
the coal mining sector. By 2030, emissions levels are projected to more than double the levels in 2000. The
most rapid period of emissions growth occurred in the first decade of this century. More measured
growth is projected beyond 2010. Between 2010 and 2030, emissions are projected to grow by 33%.
Currently, China represents over 50% of global emissions. China's share of global emissions is projected
to increase to 55% by 2030.
Figure 1-1: CH4 Emissions from Coal Mining: 2000-2030
784
2000
2010 2020
Year
2030
Source: U.S. Environmental Protection Agency (USEPA), 2012.
1 Global CH4 in 2010 = 7,549.2 MtCO2e (see Table A-2 in USEPA, 2012)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
COALMINING
Capture for use or destruction are two alternative abatement measures that can mitigate
emissions associated with underground mining. For mines that are able to utilize the recovered gas, the
captured methane represents a potential revenue stream that may offset a portion of the cost of
implementing the abatement measure. Specifically, three categories of abatement measures are
considered: (1) gas recovery for energy end uses; (2) combustion through flaring; and (3) ventilation air
methane (VAM) recovery and destruction through thermal or catalytic oxidation, where low
concentrations of Q-k present in ventilation air exhaust flows are oxidized.
Global abatement potential that is technologically achievable from underground coal mining based
on the abatement measures considered is approximately 60% of total annual emissions in 2030. Marginal
abatement cost (MAC) curve results are presented in Figure 1-2 for 2010, 2020, and 2030. Maximum
abatement potential in the coal mining sector is 400 and 468 MtCO2e in 2020 and 2030 respectively.
Figure 1-2: Global Abatement Potential in Coal Mining: 2010, 2020, and 2030
o
u
•2010
•2020
•2030
500
Non-CO2 Reduction (MtCO2e)
While maximum abatement could only be achieved at higher carbon prices, the MAC results suggest
that significant opportunities for CH4 reductions in the coal mining sector at carbon prices at or below
$10. Furthermore there are approximately 78 MtCO2e of reductions that are cost-effective at currently
projected energy prices. These reductions are sometimes referred to as no-regret options.
The following section offers a brief explanation of how CH4 is emitted from coal mines, followed by a
discussion of projected trends in international baseline emissions. Section II.1.3 characterizes possible
abatement technologies, outlining their technical specifications, costs and possible benefits, and potential
in selected countries. The final section of this chapter discusses emissions reductions that occur following
the implementation of each abatement technology and how these reductions are reflected in the MACs.
11-2
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
COALMINING
11.1.2 Methane Emissions from Coal Mining
Methane is produced during a natural process that converts organic material into coal. Methane is
stored in the coal through a physical process referred to as sorption. Sorbed methane is condensed within
the matrix of the coal as long as the hydrostatic pressure is maintained, but during the mining process,
the pressure drops and the gas will begin to desorb and flow into the mine's workings. Methane is also
stored in the free spaces of the coal strata and migrates to the mine workings. Many factors affect the
quantity of Q-k released, including the gas content of the coal, the permeability and porosity of the coal
seams, the method of mining used, and the production capacity of the mining operation. The
concentration of methane present in the coal seam depends on several factors but generally increases with
depth. There are four major sources of CH4 emissions in the coal sector including underground mines,
surface mines, post-mining processing, and abandoned mines. Underground mining is the largest single
source of emissions in the sector.
Underground Mines. High concentrations of Q-k in underground coal mines is a safety hazard. Mines
are ventilated by use of large fans which are capable of moving large volumes of air through the active
workings. Air is drawn across the working face, where coal is being extracted, and exhausted to the
atmosphere. This is often adequate to maintain safe levels of methane in the mine workings.
In especially gassy mines, the ventilation system may be supplemented by degasification systems, to
ensure adequate evacuation of methane from the mine to ensure safe working conditions. Degasification
systems are necessary to ensure safe operations in highly gas prone underground mines that are
susceptible to gas outbursts and high methane emissions encountered at the mining face. The primary
methods to reduce emissions at the mining face include pre-mine drainage systems that reduce the
methane pressure in the coal seam, thereby reducing both the total volume of methane emitted at the
mining face and the rate at which it is emitted and post-mining boreholes which drain methane from the
collapsed and fractured zone (gob) behind the mining face. These reduce the concentration of methane,
especially near the active mining coal face.
Degasification systems consist of a network of boreholes drilled from the surface, or within the mine
for the purpose of removing Q-k before, during, or after mining. These wells extract coal mine methane
from the coal seam at relatively high concentrations (30% to 90%). Concentrations vary depending on the
type of coal mined and the degasification technique used. In contrast, underground mine ventilation
systems emit large quantities of very dilute methane (typically less than 1% methane), known as
"ventilation air methane" or VAM.
Traditionally, Q-k extracted from the mine is released or vented into the atmosphere. It is possible to
mitigate underground mine methane emissions, especially from degasification systems, by capturing the
gas and either flaring it or recovering and using it for energy. In the case of VAM, the relatively low Q-k
concentration makes it more challenging both technically and economically to mitigate it or recover
energy from it.
Surface Mines. Surface mining is a technique used to extract coal from shallow depths at or below the
Earth's surface. Because the hydrostatic pressure at shallow depths is lower, the in situ CH4 content is not
as high at surface mines as at underground mines. CH4 emissions from surface mines (expressed as
volume of CH4 per mass of coal mined) are typically less than from underground mines. As the overlying
soil and rock is removed and the coal exposed, CH4 is emitted directly into the atmosphere. Both because
of its lower methane contents and because surface mining is only applicable in certain geographic
regions, surface mines may contribute only a small fraction of a country's overall emissions. For example,
in the United States in 2009, surface mining accounted for over 60% of coal production, while only
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-3
-------
COALMINING
accounting for 18% of Q-k emissions from coal mining (USEPA, 2011b). In China, there is very little
surface mining whereas in India almost all coal production is from surface mines. The only technically
feasible abatement measures available to surface mining are pre-mine methane drainage in advance of
mining, similar to coal bed methane (CBM) recovery operations (USEPA, 2008a), or horizontal boreholes
into a high wall where the operation starts as a surface mine but eventually the drift requires the
operation to become an underground mine. Given the limited contribution surface mines make to
national baseline emissions, this analysis did not consider any abatement measures for surface mining.
Post-mining Operations. Following the mining operations, a series of operations, called post-mining
operations, constitutes a third source of Q-k emissions. Not all Q-k gas is released from coal during the
process of coal breakage that takes place during extraction and transport to the surface at mining
operations; some emissions occur during the processing, storage, and transport of coal as the coal
continues to de-gas. The rate of post-mining emissions depends on the rank of coal and the way it is
handled. The highest rate of emissions occurs when coal is crushed, sized, and dried for industrial and
utility uses. Given the limited contribution of post-mining emissions to national baseline emissions and
the limited technical options to abate these emissions from rail cars or storage piles, this analysis does not
consider any abatement measures for post-mining operations.
Abandoned Mines. Abandoned mines are another source of CH4 emissions. Emissions are released
through old wells, ventilation shafts, and cracks and fissures in overlying strata. In some cases, the CH4
from these mines has been captured and used as a source of natural gas or to generate electricity. The
2006 Intergovernmental Panel on Climate Change (IPCC) guidelines provide a separate methodology for
reporting emissions from abandoned coal mines. Hence, emissions from this source are excluded from
this analysis and are not included in the baseline estimates. Although there are abatement options for
recovering and using methane from abandoned mines, these options were not examined in this analysis.
In summary, the majority of the CH4 emitted from coal mining operations comes from gassy
underground mines via ventilation systems and degasification systems. Smaller, but still significant,
amounts of Q-k are emitted from surface mining and post-mining operations and from abandoned mines.
Future levels of Q-k emissions from coal mining, however, will be primarily determined by the
management of CH4 gas at active underground mines.
11.1.2.1 Activity Data and Related Assumptions
Globally, coal production is expected to increase by 39% from 2010 to 2035, growing at an average
annual rate of 1.8%. Future baseline CH4 emissions estimates are directly related to projections of future
levels of coal production. Projected coal production is based on global trends in the demand and supply
of coal, which are particularly influenced by the global mix of electricity generation sources. China and
India are expected to account for 72% of the increase in global production as they try to meet their
demand with domestically produced coal (USEIA, 2011).
Three quarters of the world's recoverable coal reserves are located in five countries: the United States
(27%), Russia (18%), China (13%), Australia (9%), and India (7%) (USEIA, 2011). Because global coal
consumption is projected to increase over the next several decades, it is also expected that these five
countries will produce the majority of coal to meet the demand. Efforts in recent years by China to
modernize its coal mining operations are allowing coal to be mined at greater depths and at lower cost.
This, combined with a tremendous demand for coal-generated electricity, has contributed to substantial
increases in CHk emissions.
Emissions factors for coal mining vary depending on the type of coal being mined, the depth at
which the mining face is located, and how much coal is being produced in a given year. These factors also
11-4 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
COALMINING
vary across countries and time. Emissions factors are estimated for each country and are based on the
methodologies detailed in the 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The IPCC
guidelines provide a methodology for countries developing emissions factors based on the availability
and certainty of emissions data.2 Table 1-1 reports IPCC Tier 1 emissions factors for underground mines
based on Q-k intensity and coal seam depth unadjusted for any Q-k utilization or flaring.
Table 1-1: IPCC Suggested Underground Emissions Factors for Selected Countries in ms/tonne Coal
Produced
Tier 1— Cm Emissions Factor
Low (< 200m)
Average
High (> 400m)
Emissions Factor (m3/tonne)
10
18
25
Emissions Factor3 (tCC^e/tonne)
0.14
0.26
0.36
Source: IPCC, 2006. Chapter 4: Fugitive Emissions in 2006 IPCC Guidelines for National Greenhouse Gas Inventories. Vol. 2. Energy.
Available at: http://www.ipcc-nggip.iges.or.ip/public/2006gl/vol2.html
a Conversion factor of 1 m3 = 0.0143 tC02e
Improvements made in mining technology throughout the last 30 years have resulted in the ability to
extract coal from increasingly greater depths. Developing countries' adoption of advanced mining
technology has allowed countries such as China and Russia to reach deeper into their existing coal
reserves. As noted earlier, the volume of Q-k in the coal seam may increase at greater depths because of
increasing hydrostatic pressure. Thus, it is expected that the Q-k emission factors will increase as
technology allows large coal-producing countries to mine deeper, gassier coal seams.
1.1.2.2
Emissions Estimates and Related Assumptions
This section briefly discusses the historical and projected emissions trends and presents the baseline
emissions used in the MAC analysis.3
Historical Emissions Estimates
Global Q-k emissions from coal mining increased by 14% between 1990 and 2010. Key factors that
contributed to the emissions growth over this time period include overall increases in coal production as
well as technological improvements that have enabled coal mining at increased depths. For additional
detail on historical emissions estimates we refer the reader to USEPA's Global Emissions Report (2012).
Projected Emissions Estimates
Absent the widespread adoption of abatement technologies, worldwide global Q-k emissions from
coal mining will continue to increase at an accelerated rate. Over the next 20 years, emissions are
expected to grow at an average annual rate of 1.5%, compared with 0.07% between 1990 and 2010. The
projected increase is driven by a number of factors, including continued mining technology advances and
increasing demand for coal for electricity production over the same period. Large, increasingly developed
countries, such as China and India, are expected to experience high levels of economic growth. Economic
Emissions factors for underground mines, the largest source of CH4 emissions from coal mining, are the same as
those described in the Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories, and emissions factors
for surface mines, post-mining, and abandoned mines are all based on the IPCC's 2006 guidelines.
3 For more detail on baseline development and estimation methodology, the authors refer the reader to the USEPA's
Global Emissions Projection Report available at: vyvyyy.epa.gov/climatechange/economics/international.html.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-5
-------
COALMINING
growth in these countries will be the biggest driver of future Q-k emissions from coal mining. Increasing
rates of technological adoption and modernization of mining operations will allow developing countries
to mine deeper and more effectively and, in turn, produce more Q-k emissions. Table 1-2 presents
baseline emissions projections by country and region from 2010 to 2030.
Table 1-2: Projected Emissions from Coal Mine CH4 by Country and Region: 2010 to 2030 (MtC02e)
Country/Region
2010
2015
2020
2025
2030
CAGR*
(2010-2030)
Top 5 Emitting Countries
China
United States
Russia
Australia
Ukraine
296
67
49
27
30
321
70
51
30
31
354
70
51
31
31
397
73
50
34
31
436
78
51
37
31
2.0%
0.7%
0.3%
1.5%
0.3%
Rest of World (ROW) by Region"
Africa
Central & South America
Middle East
Europe
Eurasia
Asia
North America
World Total
10
9
0
30
22
45
3
589
11
11
0
29
23
49
3
630
11
13
0
29
23
55
3
671
12
14
0
29
23
61
3
725
12
16
0
29
24
67
3
784
1.1%
3.0%
3.5%
-0.2%
0.3%
2.0%
-0.7%
1.4%
aCAGR = Compound Annual Growth Rate
bROW by Region excludes emissions from top 5 countries.
Source: USEPA, 2012
11.1.3 Abatement Measures and Engineering Cost Analysis
This analysis considers five abatement measures classified into three technology categories that
include recovery for pipeline injection, power generation or use as a process fuel/on-site heating, flaring,
and catalytic or thermal oxidation of VAM. It should be noted that mitigation of gas from degasification
systems and ventilation systems are independent of each other. Abatement measures in the coal mining
sector consist of one or more of the following primary components: (1) a drainage and recovery system
(where applicable) to remove methane from the coal seam pre-mining or from the gob area post mining,
(2) the end-use application for the gas recovered from the drainage system (where applicable), and (3) the
ventilation air methane recovery or mitigation system (where applicable).
Costs are derived from USEPA's Coalbed Methane Outreach Program (CMOP) project Cash Flow
Model (USEPA, 2011b) and applied to a representative population of underground mines. Table 1-3
summarizes the average total installed capital costs and annual operations and maintenance (O&M) costs
for each abatement measure.
11-6 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
COALMINING
Table 1-3: Summary of Abatement Measures for Coal Mines
Abatement Measure
I Installed Total Annual Technical
Capital Cost3 O&M Cost Technical Effectiveness15
(million USD) (million USD) Lifetime (Years) (%)
Energy End Uses
Pipeline injection
On-site electricity generation
On-site use for process heat
8.4
23.0
2.8
2.4
2.6
1.2
15
15
15
21%
28%
28%
Excess Gas Flaring
Enclosed flare system
2.3
1.5
15
28%
Mitigation of VAM
VAM oxidation
8.0
1.3
15
19-68%
a Capital costs include costs of both recovery and abatement equipment requirements.
b Abatement potential expresses the maximum potential emission reductions at a facility level.
This section describes the abatement measures and associated costs of the methane recovery and
abatement in the coal mining sector. Each technology is briefly characterized followed by a discussion of
costs, potential benefits, technical effectiveness, and applicability assumptions used to estimate the
abatement potential.
Technical effectiveness factors are calculated by considering a number of technological efficiency and
applicability factors. Table 1-4 presents these factors for each abatement measure. These include the
technical effectiveness of the recovery system and reduction efficiency of the utilization or destruction
technology. Technical effectiveness, represented by [E] in Table 1-4, of any option at the mine level is
equal to the product of the facility applicability, recovery efficiency, technical feasibility, and reduction
efficiency factors.
Table 1-4: Factors Used to Estimate Abatement Potential in Coal Mines
Facility
Applicability
Abatement Measure [A]
Recovery
Efficiency
[B]
Technical
Feasibility
[C]
Reduction
Efficiency
Technical
Effectiveness
[E]
Energy End Uses: Drained Gas
Pipeline injection
On-site electricity generation
On-site direct use
38%
38%
38%
75%
75%
75%
100%
100%
100%
75%
98%
98%
21%
28%
28%
Mitigation only: Drained gas
Enclosed flare system
38%
75%
100%
98%
28%
Oxidation of VAM
VAM oxidation
62%
25% - 90%
77%
98%
19-68%
Technical Effectiveness [A] x [B] x [C] x [D] = [E]
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
II-7
-------
COALMINING
Facility applicability [A] represents the share of total mine-level methane emissions that are available
for abatement through degasification and VAM. Approximately one-third4 of total mine emissions can be
recovered through degasification (also commonly referred to as drainage), while the majority of mine
emissions are released at low concentrations in the ventilation air referred to as VAM.
Recovery efficiency [B] relates to the collection system (see Section II.1.3.1) itself and reflects what
may be recovered through the drainage wells or ventilation exhaust systems. Only a fraction of the total
drained CH4 may be effectively used or destroyed because of natural variances in the volume and
concentration of methane collected. With respect to VAM oxidation, for this analysis recovery efficiency
is set at 25% in 2010 and grows to 90% by 2030.
Technical feasibility [C] relates to the physical or technical limitations of the technologies. It is
technically feasible to safely combust mine gas with CH4 concentrations greater than 30% for drained gas
or 0.25% for VAM. A value of 77% for VAM represents the fraction of exhaust vents with methane
concentrations high enough (>0.25%) to allow for oxidation.5 Finally, the reduction efficiency [D] is the
factor that describes the destruction efficiency of each end use or combustion technology. For pipeline
injection, the reduction efficiency represents the methane losses that occur during transport from the
mine to the point of sale into a natural gas pipeline.
11.1.3.1 Methane Recovery System from Degasification/Drainage Systems
High-quality CH4 is recovered from coal seams by drilling vertical wells from the surface up to 10
years in advance of a mining operation or drilling in-mine horizontal boreholes several months or years
before mining. Most mine operators exercise just-in-time management of gate road development;
subsequently, horizontal cross-panel boreholes are installed and drain gas for 6 months or less (USEPA,
2011b).
The components of the capital and annual costs for the drainage wells are outlined as given in
USEPA's CMOP Cash Flow Model documentation (USEPA, 2011b). The recovery system includes the
equipment required for drainage wells, gas gathering lines, and delivery systems for coal mine methane
(CMM). The recovery system is included in the costs of all abatement measures with the exception of
VAM oxidation.6 These costs are additive to the costs associated with each abatement measure.
• Capital Cost: The capital costs for a drainage system are a function of the recovered gas flow rate.
Equipment requirements include construction of the drainage well(s), a wellhead blower, a
satellite compressor station, and gathering pipelines that connect the compressors to the methane
end-use technology. The total installed capital costs will vary by location and gas flow rate. For
example, assuming a 600 Mcf/day volume of CMM gas (with a CH4 concentration of 90%), we
estimate the capital costs would be $850,000. See Appendix B for additional detail on equipment
cost assumptions.
4 The proportion of mine QHk emissions recoverable through degasification systems can vary from 0% to 70%
depending on the gassiness of the mine. This analysis uses 38%, which represents an average for gassy mines.
5 This value may be a high estimate based on anecdotal evidence from field testing experience. For example, the
number of mines in Asia that meets a threshold for application of available and field-tested VAM abatement systems
is much lower.
6 A recovery system is not required for VAM oxidation because it relies on the mine's existing ventilation system that
would be installed before mining operations commence.
11-8 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
COALMINING
• Annual Operating and Maintenance (O&M) Costs: The annual costs are required to maintain
the drainage system equated to approximately $2.2/Mcf per year. These costs include the ongoing
installation of gob wells and the gathering system piping that connects the wells to satellite
compressors. In keeping with the example mine of 600 Mcf/day, the annual O&M costs associated
with the recovery system would be approximately $475,000.
• Recovery Efficiency: Recovery efficiency is assumed to be 75%.
11.1.3.2 Degasification for Utilization in Energy Production
This category of abatement measures includes (1) recovery for pipeline injection and (2) recovery for
electricity generation. Both options require a recovery system in place to extract the methane gas from the
coal seams. Which technology is most cost-effective will be determined by a combination of regional
energy prices and the capital equipment requirements.
Degasification for Pipeline Injection
Natural gas companies may purchase Q-k recovered from coal mines. CH4 suitable for sale into
natural gas pipelines must have a concentration of at least 96% and contain no more than 4%
concentration of noncombustible gases with a maximum of 4% carbon dioxide or nitrogen and 1 ppm
oxygen. Although Q-k from coal mines requires water removal, it is typically free of hydrogen sulfide
and other impurities found in natural gas. Hence, little to no additional treatment and processing are
necessary to meet the requirements for pipeline injection. In some cases, high-quality CH4 also can be
obtained from gob wells.
Premining degas wells are the preferred recovery method for producing pipeline quality Q-k from
coal seams because the recovered methane is not contaminated with ventilation air from the working
areas of the mine.
Gob wells, in contrast, generally do not produce pipeline-quality gas because the methane is
frequently mixed with ventilation air. Gob gas CH4 concentrations can range from 30% to over 90%. It is
possible to upgrade gob gas for pipeline quality although blending with pre-mine drained gas and/or
oxygen removal may be necessary, adding to the cost of gas processing. However, it is possible to
maintain a higher and more consistent gas quality through careful monitoring and adjustment of the
vacuum pressure in gob wells as has been demonstrated in the United States (USEPA, 2008b).
The viability of a pipeline project is affected by several key factors. First is a coal mine's proximity to
a commercial pipeline. The cost of constructing a pipeline to connect to a commercial pipeline can vary
greatly depending on distance. Secondly, and more importantly, the terrain will affect the viability of a
commercial pipeline project. Many mining areas are located in hilly and mountainous regions, thus
increasing the difficulty and cost of constructing both gathering lines and pipeline to connect to the
commercial natural gas pipeline (USEPA, 2008b). Finally, disposal of water produced from vertical wells
may be a significant factor in determining the economic viability of a pipeline project.
• Initial Capital Cost: The per facility installed capital costs for pipeline injection of gob gas, as
described in USEPA (2011b), include the installation of a pressure swing adsorption system to
remove nitrogen and carbon dioxide down to a 4% inert level. The utilization cost is a function of
both the inlet gas flow rate and methane concentration and includes the cost of dehydration and
compression necessary to process the gas and boost the sales gas to pressure for injection in a
natural gas transmission pipeline. While there may be a range of pressures at which pipelines
operate, this analysis assumes an operating pressure of 900 psig. This option also includes the
installation of a catalytic oxygen removal system and a pipeline to connect the mine's gas
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-9
-------
COALMINING
processing system to a natural gas pipeline. Pipeline costs estimated for this analysis are adjusted
based on mine proximity to commercial pipeline but do not attempt to account for variations in
terrain across countries.
• Annual O&M Costs: The annual costs include costs of recovery system and cost of gas treatment
and compression required for injection into commercial natural gas pipelines.
• Annual Benefits: Revenues from this option are the gas sales based on the volume of gas
produced and the market price of natural gas.
• Technical Effectiveness: The analysis assumes a technical effectiveness of 21%. As shown in
Table 1-4, this considers a recovery efficiency of 75% (reflects the loss of 25% of the gas cannot be
used in pipeline injection because the methane concentration is too low to process to pipeline
specifications) and destruction efficiency of 75% to account for losses during transport to point of
sale and injection into a commercial natural gas pipeline.
• Technical Lifetime: 15 years
Degasification for Electricity Generation
Drained methane can be used to fire internal combustion (1C) engines that drive generators to make
electricity for sale to the local power grid (USEPA, 2011b). The quality of methane required for use in
power generation can be less than that required for pipeline injection. Internal combustion (1C) engine
generators can generate electricity using gas that has heat content as low as 300 Btu/cf or about 30%
methane. Mines can use electricity generated from recovered methane to meet their own on-site
electricity requirements and can also sell electricity generated in excess of on-site needs to utilities
(USEPA, 2008b).
Coal mining is a very energy-intensive industry that could realize significant cost savings by
generating its own power. Nearly all equipment used in underground mining runs on electricity,
including mining machines, conveyor belts, ventilation fans, and elevators. While most of the equipment
used in mining operations is used 250 days a year for two shifts per day, ventilation systems are required
to run continuously year round. These systems require large amounts of energy, up to 60% of a mine's
total electricity usage. Total electricity demand can exceed 24 kWh per ton of coal produced (USEPA,
2008b).
• Capital Cost: The cost for this option includes the cost of gas processing to remove gas
contaminants (primarily water vapor and solid particles), the electricity generation equipment,
and power grid connection equipment. Costs are assumed to be $l,300/kW. Assuming a 2 MW
facility and a capacity factor of 90%, total installed capital costs of electricity generation would be
$2.7 million. Total installed capital costs for this abatement measure would be $4.5 million, which
includes the $850,000 for recovery, assuming 20% owner's costs and 5% contingencies.
• Annual O&M Costs: The annual costs include $0.02/kWh for the engine-gen set in addition to
the $2.2/Mcf cost of the recovery system. Assuming a 600 Mcf/d flow and 90% capacity total
O&M costs would be approximately $0.8 million USD.
• Annual Benefits: Revenues in the form of power sales at market electricity prices. A 2 MW
capacity generation facility with a 90% capacity factor would be expected to generate
11-10 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
COALMINING
approximately 16,000 MWh of electricity. Assuming an energy price of $0.075/kWh7, this project
would generate $1.2 million in revenue from electricity sales.
• Technical Effectiveness: The analysis assumes a technical effectiveness of 28%, assuming a
recovery efficiency of 75% and destruction efficiency of 98% in the energy generation unit.
• Technical Lifetime: 15 years
11.1.3.3 Degasification for On-site Utilization—Process Heat
This category of abatement measures includes (1) recovery for use in the boiler for supporting in-
mine heating and (2) recovery for coal drying.
Mine Boilers
Drained methane can be used to fuel on-site boilers that provide space or water heat to mine facilities.
This analysis assumes that existing boilers will be retrofitted to burn methane and that the drained
methane is of sufficient quality to fuel the mine's boiler and no additional gas processing is required.
• Capital Cost: The costs for this option are primarily associated with the capital cost to retrofit the
mine boiler to fire drained gas. The analysis assumes a 8.1 MMBTU/hr8 average boiler heat load
and a retrofit cost of $7,500/MMBTU/hr. Assuming the mine boiler fuel demand was 10 Mcf/hr,
total installed capital costs for this abatement measure would be $635,000, which includes
$382,000 for the recovery system, $122,000 for boiler retrofit, and an additional 20% in owner's
costs and 5% for contingencies.
• Annual O&M Costs: The annual costs are the $2.4/Mcf to operate the recovery system. Assuming
a 240 Mcf/d flow and 90% capacity, the total O&M costs would be approximately $213,000 USD.
• Annual Benefits: Benefits are the energy costs offset by using the drained methane gas as a
substitute fuel source (offsetting coal consumption). Revenues associated with this project will be
the quantity of coal replaced at the mine mouth coal market price ($/MMBTU).
• Technical Effectiveness: The analysis assumes a recovery efficiency of 75% and destruction
efficiency of 98% when combusted in mine boiler.
• Technical Lifetime: 15 years
Coal Drying
Another on-site direct use application for drained CMM is to use it as a fuel in thermal coal drying
operations at coal preparation plants co-located near an underground mine. The existing coal drying
process can be retrofitted to burn methane as a supplemental fuel in addition to burning coal. Similar to
the mine boiler option, we assumed the CMM will not require further processing to serve as fuel to the
dryer.
• Capital Cost: The cost of converting the dryer to burn CMM was assumed to be the same as the
cost of converting the boiler firing system [$7,500/MMBtu/hr]. The analysis assumed an average
coal drying rate of 380 tons/hr (USEPA, 1998). Assuming the average coal dryer heating
7 The actual price utilities would be willing to pay will vary depending on market and regulatory environment
within the specific country. In the absence of any additional market incentives, purchasers would likely only pay the
price of generation in the range of $0.025/kWhr in the United States.
8 MMBTU/hr = Million British Thermal Units per hour
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-11
-------
COALMINING
requirement is 228 MMBTU/hr,9 CMM gas with a lower heating value of 991 BTU/cf, the total
installed capital costs for this abatement measure would be $635,000, which includes $382,000 for
the recovery system, $122,000 for boiler retrofit, and an additional 20% in owner's costs and 5%
for contingencies.
• Annual O&M Costs: The annual costs to operate the recovery system are assumed to be
$2.4/Mcf. Assuming a 240 Mcf/d flow and 90% capacity factor, total O&M costs would be
approximately $213,000 USD.
• Annual Benefits: Benefits are the energy costs offset by using the drained methane gas as a
substitute fuel source (offsetting coal consumption). Revenues associated with this project will be
the quantity of coal replaced based on assumed energy content (MMBTU/ton) at the mine mouth
coal market price ($/MMBTU).
• Technical Effectiveness: The analysis uses a technical effectiveness of 28%, which assumes a
recovery efficiency of 75% and destruction efficiency of 98% when combusted in mine boiler.
• Technical Lifetime: 15 years
11.1.3.4 Combustion through Flaring
After recovering methane using the drainage well technique, mines can choose to flare methane of
greater than 30% concentration (USEPA, 2011a). Flare systems considered include an open flare and
enclosed combustion system, which consists of a mounted burner where the flame is exposed (open) or
the flame is enclosed in a stack. The costs of the flaring system consist of firing equipment and
monitoring and metering equipment to validate methane destruction levels.
• Capital Cost: The cost of installing a flare system to burn CMM was assumed to be $280/Mcf/day.
Assuming an average daily flow rate of 600 Mcf gas, the total installed capital costs for this
abatement measure would be $1,239,000, which includes $850,000 for the recovery system,
$134,000 for the flare system, and an additional 20% in owner's costs and 5% for contingencies.
• Annual O&M Costs: The annual costs to operate the recovery system are assumed to be
$2.4/Mcf. Assuming a 600 Mcf/d flow and 90% capacity factor, total O&M costs would be
approximately $489,000 USD.
• Annual Benefits: There are no revenues associated with this option.
• Technical Effectiveness: The analysis uses a technical effectiveness of 28%, which assumes a
recovery efficiency of 75% and destruction efficiency of 98% when combusted in flaring system.
• Technical Lifetime: 15 years
1.1.3.5
VAM Oxidation
Oxidation technologies (both thermal and catalytic) have the potential to use Q-k emitted from coal
mine ventilation air. Extremely low Q-k concentration levels (typically below 1%) make it economically
infeasible to sell this gas to a pipeline. However, thermal oxidizers can combust the VAM converting it to
FLO and CO2. VAM oxidation is technically feasible at CFk concentrations between 0.25% and 1.25%. For
mines with lower VAM concentrations, a supplemental gas is required to bring the concentration above
the 0.25% concentration limit.
' MMBTU/hr = Million British Thermal Units per hour
11-12
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
COALMINING
• Capital Cost: Abatement measure costs include the ductwork required to collect VAM exhaust
from the mine's ventilation system at the surface vents, the design and installation of a thermal
oxidizer unit, and any supporting auxiliary equipment. The total installed capital cost of the
VAM oxidizer unit is $23 per unit of recoverable ventilation air flow measured in cubic feet per
minute [cfm]. Assuming the recoverable ventilation air flow rate of 100 Mcfm and a Q-k
concentration of 0.2%, capital costs would be $2.3 million USD (=100,000 cfm X $23/cfm). The total
installed capital costs for this abatement measure would be approximately $2.8 million after
assuming allowances of 20% in owner's costs and 5% for contingencies.
• Annual O&M Costs: Annual operating costs include costs to maintain the oxidizer unit, the
electricity required to operate the oxidizer blowers (0.075 kWh/mcf), and the periodic relocation
costs of moving equipment to the new location of a mine ventilation shaft (every 5 years at a cost
of $4/cfm). Assuming a 100 Mcfm flow rate, total O&M costs would be approximately $462,000
USD, where VAM blower electricity costs account for 55% of the annual costs, while oxidizer
O&M costs represent 28%, and annualized relocation costs make up the balance.
• Annual Benefits: Although low-grade heat can be captured from the VAM oxidizer, little
economic benefit can be obtained from it and only under special site-specific conditions; for this
reason, we assume there are no energy-related benefits for this abatement measure.
• Technical Effectiveness: The analysis assumes a technical effectiveness of between 19% and 68%,
which assumes a recovery efficiency of 25% (in 2010) to 90% (in 2030) and destruction efficiency
of 98% in a VAM oxidation unit.
• Technical Lifetime: 15 years
11.1.3.6 Evaluation of Future Mitigation Option and Trends
Based on our review of existing abatement measures, technology improvements have the potential to
reduce the costs of VAM oxidation technology. Despite its abatement potential, VAM oxidation is the
measure with the highest abatement costs largely due to three key factors that include: (1) the equipment
itself is large and costly; (2) the lack of a revenue source; and (3) only a handful of technologies have been
demonstrated at a commercial scale and as such economies of scale in production have not been realized.
The development of an international carbon market like the United Nations Framework Convention on
Climate Change's (UNFCCC's) Clean Development Mechanism (CDM) or the European Trading System
(ETS) would provide a source of revenue from the sale of carbon reduction credits. In addition,
improvements in design and catalysts have the potential to reduce the cost of oxidation over time. All
other abatement measures described in this chapter are assumed to be mature technologies.
11.1.4 Marginal Abatement Costs Analysis
This section describes the methodological approach to the international assessment of CMM
abatement measures. Here we describe the modeling approach applied to the sector and highlight the
unique facets of the modeling approach that are required to align with the general modeling framework
described in the technical summary of this report.
11.1.4.1 Methodological Approach
The analysis seeks to characterize the cost of abatement in the coal mining sector by developing
project cost estimates for a series of representative mines that represent the population of active
underground coal mines in a reference location, which is the United States. Abatement measures are
applied to the technically applicable stream of emissions (degasification, or ventilation air streams). The
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-13
-------
COALMINING
MAC model calculates break-even prices for each representative coal mine based on the facility
characteristics that include annual methane liberation, presence of existing degasification operations,
mine ventilation air flow rate, and VAM concentration. Figure 1-3 illustrates the flow of emissions and
the country and technology factors that determine the abatement potential.
Figure 1-3: Flow Chart of the Coal Mining Sector MAC Modeling Approach
tfitl-el
The MAC model internationalizes the abatement measures' project costs by applying country-specific
factors to adjust individual components of the technology costs and expected benefits (i.e., capital, labor,
energy and materials) to transpose costs from a United States context to the international country of
interest. The MAC model then applies the general break-even cost calculation using the internationalized
costs and benefits to develop country-specific abatement cost estimates.
1.1.4.2
Assessment of Sectoral Trends
Abatement potential estimated in this report is limited to the subset of emissions from underground
coal mining activities. No abatement measures are considered for surface mining, abandoned mines, or
post-mining operations. The analysis assumed that the majority of emissions from the coal mining sector
come from underground coal mining activity. As a result, a significant proportion of the BAU emissions
projected (see Table 1-3 above) are available for abatement via the measures discussed in this chapter.
This analysis considers country-specific data when available to adjust the basic assumption that between
70% and 98% of emissions are available for abatement (i.e., the quantity of emissions from underground
mining activities). For countries for which no other information was available, expert judgment was used
to assess the quantity of emissions eligible for abatement.
11-14
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
COALMINING
1.1.4.3
Definition of Model Facilities for the Analysis
A population of representative underground coal mines was developed using publicly available data
for the U.S. active underground coal mines. The dataset included detailed information on over 100 active
mines with average methane liberation rates greater than 0.1 million cubic feet per day. Information was
also available on the methane concentration in mine ventilation air.
The international population of facilities is defined through a representative dataset of underground
mines with the accompanying mine-specific characterizations. This includes the gassiness of the mine and
the concentration of methane present in the mine's ventilation air.
1.1.4.4
Estimating Abatement Project Costs and Benefits
Mine characteristics for each mine in the facility database were used to estimate abatement project
costs and benefits. Applying the costs described in Section II.1.3, the CMOP project Cash Flow Model
provided outputs including initial capital investment, annual recurring costs for operation and
maintenance, and the quantity of energy saved or offset. The costs and benefits data are then used as
inputs in the MAC model. The cost functions used in the CMOP model are assumed to be representative
of typical projects in the United States. Please refer to the CMOP model documentation for additional
details on how costs are calculated.
Table 1-5 provides an example of how the break-even prices are calculated for each abatement
measure. Project costs and benefits calculated for each coal mine are used in the calculation that solve for
the break-even price that sets the project's benefits equal to its costs.
The break-even prices presented in Table 1-5 represent weighted average break-even prices weighted
by total annual methane liberated across the population of coal mines used for this analysis. Each coal
mine will have its own break-even price based on the amount of methane recovered. Break-even prices
will be higher for less gassy coal mines and lower (potentially negative) for most gassy mines. Complete
international MAC results are presented in Section II.1.4.5.
Table 1-5: Example Break-Even Price Calculation for Coal Mine Abatement Measures
Annualized
Capital Net Annual Tax Benefit of Break-Even
Costs'5 Cost3 Depreciation Priceb
($/tC02e) ($/tC02e) ($/tC02e) ($/tC02e)
Abatement Option
Reduced
Emissions
(tC02e)
Energy End Uses
Pipeline Injection
Electricity Generation
On-Site Direct Use
Excess Gas Flaring
Enclosed Flare System
Combustion of VAM
VAM Oxidation
99,629
130,338
249,175
298,333
46,430
$18.5
$38.7
$2.5
$1.7
$37.5
-$19.5
-$33.0
-$2.8
$5.0
$28.0
$3.76
$7.84
$0.50
$0.35
$7.61
-$4.69
-$2.18
-$0.85
$6.33
$57.91
a Assumes tax rate = 40%; discount rate = 10%, technical lifetime = 15 years
b AEO 2010 Energy prices; dry natural gas ($/Mcf); electricity $/kWh); and coal ($/ton)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-15
-------
COALMINING
1.1.4.5
MAC Analysis Results
Global abatement potential in 2020 and 2030 is 400 and 468 MtCO2e, respectively. Nearly 16% of the
reduction can be achieved by implementing currently available technologies that are cost-effective at
projected energy prices. If an additional emission reduction incentive (e.g., tax incentive, subsidy, or
tradable emission reduction credit) above the zero break-even price were available to coal mine
operators, then additional emission reductions could be cost-effectively achieved. The results of the MAC
analysis are presented in Table 1-6 and Figure 1-4 by major country and regional grouping at select break-
even prices in 2030.
Table 1-6: Abatement Potential by Region at Selected Break-Even Prices ($/tC02e) in 2030
Country/Region
Break-Even Price ($/tC02e)
-10 -5 0 5 10 15 20 30 50
100
100+
Top 5 Emitting Countries
Australia
China
Russia
Ukraine
United States
29.8
3.8
3.4
1.9
32.2
4.5
4.2
3.6
3.3
43.9
6.3
4.4
4.5
15.9
204.7
24.8
15.1
23.8
17.3
219.6
26.6
16.0
25.4
18.0
264.2
31.5
16.4
27.4
19.1
265.3
31.6
17.3
28.5
22.4
266.5
31.8
19.5
33.5
22.6
267.2
31.8
19.6
34.0
22.7
269.6
32.1
19.6
34.2
22.9
269.6
32.1
19.7
34.5
Rest of Region
Africa
Central and South America
Middle East
Europe
Eurasia
Asia
North America
World Total
1.0
1.6
0.0
2.1
0.0
6.5
0.2
48.5
1.0
1.6
0.0
2.4
0.0
6.8
0.2
58.6
1.1
1.9
0.0
2.6
1.3
8.2
0.3
77.7
6.0
6.7
0.1
10.8
8.5
30.9
1.3
348.7
6.6
7.8
0.1
11.9
9.6
34.0
1.4
376.2
7.5
8.0
0.1
12.4
11.0
37.4
1.4
435.3
7.5
8.3
0.1
13.0
11.0
38.9
1.5
442.2
7.5
9.8
0.2
15.1
11.1
41.9
1.7
461.1
7.6
9.9
0.2
15.2
11.1
42.1
1.8
463.0
7.7
9.9
0.2
15.2
11.2
42.3
1.8
466.4
7.7
10.0
0.2
15.4
11.2
42.4
1.8
467.6
11-16
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
COALMINING
Figure 1-4: Marginal Abatement Cost Curve for Top 5 Emitters and Rest of World in 2030
$60 -,
$50
300
•Australia
Russia
•China
•Ukraine
•United States
•Rest of World
Non-CO2 Reduction (MtCO2e)
11.1.4.6
Uncertainties and Limitations
Several key limitations in current data availability constrain the accuracy of this analysis. Successfully
addressing these issues would improve development of the MACs and predictions of their behavior as a
function of time. Some of these limitations include the following.
• Accurate Distribution of Mine Type for Each Country. Extrapolating from available information
about individual mines to project fugitive emissions at a national level implies that the available
data are representative of the country's coal production not already included in the existing
database. A more accurate distribution of representative mines would improve the accuracy of
the cost estimates and the shape of each MAC. These data would include mines of all sizes,
emissions factors, and production levels. This lack of information becomes increasingly
problematic when evaluating a country such as China, where the majority of mines are small and
private mines are not represented in currently available data sources.
• Country-Specific Tax and Discount Rates. In this analysis, a single tax rate was applied to mines
in all countries to calculate the annual benefits of each technology. Similarly, the discount rate
may vary by country. Improving the level of country-specific detail will help analysts more
accurately quantify benefits and break-even prices.
• Improved Information on Public Infrastructure. A more detailed understanding of each
country's natural gas infrastructure would improve the estimates of costs associated with
transporting CH4 from a coal mine to the pipeline. Countries with little infrastructure will have a
much higher transportation cost associated with degasification and enhanced degasification
technologies.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-17
-------
COALMINING
Concentrations for VAM in International Mines. The effectiveness and applicability of VAM
technology depends on VAM concentration and mine-specific coal production rates. Improved
data on the VAM concentration levels for individual mines would enhance the accuracy of cost
estimates. This information would also help to more accurately identify the minimum threshold
concentration levels that make VAM oxidation an economically viable option.
11-18 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
COALMINING
References
Global Methane Initiative (GMI). 2010. Coal Mine Methane Country Profiles. This document provides
individual profiles for 37 countries that covers coal production and coal mine methane markets.
Available at: http://www.globalmethane.org/tools-resources/coal overview.aspx.
Mine Safety and Health Administration (MSHA). January 7, 2010. 2008 Mine Data. Personal
communications with Chad Rancher at MSHA. Coal Mine Data, methmine08nma.xls.
Somers, J. M and L. Schultz. 2009."Coal Mine Methane Ventilation Air Methane: New Mitigation
Technologies." Available at: http://www.epa.gov/cmop/resources/vam.html.
U.S. Energy Information Administration (USEIA). 2011. International Energy Outlook 2011. Coal Overview.
DOE/EIA-0484(2011). Washington, DC: EIA. Available at: www.eia.gov/forecasts/ieo/.
U.S. Environmental Protection Agency (USEPA). 2012. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA.
U.S. Environmental Protection Agency (USEPA). 2012. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA.
U.S. Environmental Protection Agency (USEPA). 2011a. Inventory of U.S. Greenhouse Gas Emissions and
Sinks: 1990-2009. Annex 3 Methodological Descriptions for Additional Source or Sink Categories. USEPA
M30-R-11-005. Washington, DC: USEPA. Obtained on May 5, 2011, at:
http://epa.gov/climatechange/emissions/usinventoryreport.html.
U.S. Environmental Protection Agency (USEPA). 2011b. User's Manual for Coal Mine Methane Project Cash
Flow Model (Version 2). Obtained January 25, 2010, at: http://www.epa.gov/
cmop/docs/cashflow user guide.pdf.
U.S. Environmental Protection Agency (USEPA). Coal Mine Methane Project Cash Flow Model (Version 2).
Obtained on January 25, 2010, at: http://www.epa.gov/cmop/resources/index.html.
U.S. Environmental Protection Agency (USEPA). July 2008a. U.S. Surface Coal Mine Methane Recovery
Project Opportunities. USEPA M30-R-08-001. U.S. EPA Coalbed Methane Outreach Program. Obtained
on January 7, 2011, at: http://www.epa.gov/cmop/docs/cmm recovery opps surface.pdf.
U.S. Environmental Protection Agency (USEPA). September 2008b. (Revised January 2009). Identifying
Opportunities for Methane Recovery at U.S. Coal Mines: Profiles of Selected Gassy Underground Coal Mines
2002-2006. EPA 430-K-04-003. U.S. EPA Coalbed Methane Outreach Program. Obtained January 7,
2010, at http://www.epa.gov/cmop/docs/profiles 2008 final.pdf.
U.S. Environmental Protection Agency (USEPA). 2004. Identifying Opportunities for Methane Recovery at
U.S. Coal Mines: Profiles of Selected Gassy Underground Coal Mines 1999-2003. EPA 430-K-04-003.
Washington, DC: Office of Air and Radiation. Available at:
http://www.epa.gov/cmop/docs/profiles 2003 final.pdf.
U.S. Environmental Protection Agency (USEPA). November 1998. Use of Coal Mine Methane in Coal
Dryers. (US)EPA Coalbed Methane Outreach Program Technical Options Series. USEPA #6202J.
Washington, DC: USEPA. Obtained on January 7, 2011, at: http://www.epa.gov/cmop/docs/
016red.pdf.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-19
-------
OIL AND NATURAL GAS SYSTEMS
11.2. Oil and Natural Gas Svstems
1.2.1
Sector Summary
(il and natural gas (ONG) systems are a leading source of anthropogenic Q-k emissions,
emitting 1,677 MtCO2e or 23% of total global CH4 emissions in 2010 (USEPA, 2012a). Russia,
the United States, Iraq, Kuwait, and Uzbekistan accounted for more than half of the world's
CH4 emissions in this sector in 2010. Figure 2-1 presents the business-as-usual baseline projections for the
ONG sector between 2000 and 2030.
Figure 2-1: Emissions Projections for the Oil and Natural Gas Systems Sector: 2000-2030
2,113
Uzbekistan
i Kuwait
i Iraq
i United States
i Russia
ROW
2000
2010
2020
2030
Year
Source: U.S. Environmental Protection Agency (USEPA), 2012a
ONG system emissions are projected to grow 26% between 2010 and 2030 with Brazil and Iraq
experiencing the highest rate of growth at 128% and 100%, respectively, over the same time period.
A number of abatement measures are available to mitigate Q-k losses from activities associated with
or directly from the operation of equipment components common across the ONG system segments of
production, processing, transmission, and distribution. These abatement options in the ONG system
segments generally fall into three categories: equipment modifications/upgrades; changes in operational
practices, including direct inspection and maintenance (DI&M); and installation of new equipment. The
abatement measures may be applied to components and equipment used in ONG operations, including
compressors/engines, dehydrators, pneumatics/controls, pipelines, storage tanks, wells, and other
processes and equipment commonly used in some or all of the ONG system segments. The global
abatement potential associated with the suite of abatement measures applicable for ONG systems is
illustrated in the marginal abatement cost (MAC) curves for 2010, 2020, and 2030 presented in Figure 2-2.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-21
-------
OIL AND NATURAL GAS SYSTEMS
Figure 2-2: Global Abatement Potential in Oil and Natural Gas Systems: 2010, 2020, and 2030
•2010
•2020
•2030
1,200
Non-C02 Reduction (MtC02e)
Note: Figure 2-2 does not show the entire MAC curve, an additional 10% of abatement potential is available at prices > $60/tC02e.
Global abatement potential in the ONG sector is 60% of the sector emissions in 2010, or 997 MtCO2e.
The abatement potential increases over time, growing to 1,103 and 1,218 MtCO2e in 2020 and 2030
respectively (representing 58% of each years' BAU emissions). Nearly 70% of the abatement potential is
achievable at a carbon prices below $5. In addition, over 61% of abatement (747 MtCO2e in 2030) is cost-
effective at current energy prices (i.e. a carbon price < $0/tCO2e).
The following section briefly explains Q-k emissions from ONG systems. This is followed by
international Q-k emissions projections. Subsequent sections characterize the abatement technologies and
present the costs and potential benefits. Finally, this chapter concludes with a discussion of the MAC
analysis and the regional results.
I 2 2
I • ^m • ^m
Methane Emissions: Oil and Natural Gas Systems
CH4 is the principal component of natural gas.1 Fugitive CH4 is emitted through activities and
components associated with the natural gas production, processing, transmission, and distribution. Oil
production and processing upstream of oil refineries can also emit CH4 in significant quantities through
1 CH4 concentrations typically increase as the natural gas moves from production to distribution. Typically QHk
concentrations in non-associated gas are assumed to be 80% at production, increasing to 87% in processing, and 95%
in transmission and distribution. Associated gas typically has a lower concentration (between 65 and 75%) depending
on the presence of other hydrocarbons in the gas mix.
11-22
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
OIL AND NATURAL GAS SYSTEMS
routine venting, flaring, and other fugitive sources associated with the production, transmission,
upgrading, and refining of crude oil and distribution of crude oil products (IPCC, 2006). Figure 2-3
identifies the facilities and equipment associated with the ONG system segments.
Figure 2-3: Segments of Oil and Natural Gas Systems
Production & Processing
1. Drilling and Well Completion
2. Producing Wells
3. Gathering Lines
4. Gathering and Boosting Stations
5. Gas Processing Plant
Natural Gas
Transmission & Storage
6. Transmission CompressorStations
7. Transmission Pipeline
8. Underground Storage
r f-r ~s n^^aB^^H~^Hh jp^sm^muiB^^*aaji
iob
Distribution
9. Distribution Mains
10. Regulators and Meters for:
a. City Gate
b. Large Volume Customers
c. Residential Customers
d. Commercial Customer
Source: Adapted from American Gas Association (AGA) and Natural Gas STAR Program.
Table 2-1 provides examples of the typical facilities and equipment that comprise ONG systems.
Fugitive CH4 emissions result from equipment leaks, system upsets, process venting, and deliberate
flaring at oil and gas production fields, natural gas processing facilities, natural gas transmission lines
and compressor stations, natural gas storage facilities, and natural gas distribution lines (USEPA, 2012a).
Table 2-1: Emissions Source from Oil and Natural Gas Systems
Segment Facility
Production Wells, central gathering facilities
Equipment at the Facility
Separators, pneumatic devices, chemical
injection pumps, dehydrators, compressors,
heaters, meters, pipelines, liquid storage
tanks
Processing Gas plants
Vessels, dehydrators, compressors, acid gas
removal (AGR) units, heaters, pneumatic
devices
Transmission Transmission pipeline networks, compressor stations, meter
and storage and pressure-regulating stations, underground
injection/withdrawal facilities, liquefied natural gas (LNG)
facilities
Vessels, compressors, pipelines,
meters/pressure regulators, pneumatic
devices, dehydrators, heaters
Distribution Main and service pipeline networks, meter and pressure-
regulating stations
Pipelines, meters, pressure regulators,
pneumatic devices, customer meters
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
II-23
-------
OIL AND NATURAL GAS SYSTEMS
11.2.2.1 Activity Data or Important Sectoral or Regional Trends and Related
Assumptions
Emissions from ONG systems are closely correlated with the quantity of ONG produced and
consumed. Globally, production and consumption of natural gas are expected to increase in both the near
term and long term. Between 2008 and 2035, natural gas supplies are expected to increase by almost 60
trillion cubic feet, or roughly 1.6% per year (EIA [U.S. Energy Information Administration], 2011a). The
majority of production growth is projected to occur in non-Organisation for Economic Co-operation and
Development (OECD) countries most notably, in the Middle East, Asia, and Africa regions, where
production growth rates average 2.8, 2.5, and 2.4% per year, respectively. Figure 2-4 presents projected
global gas production by major region from 2008 to 2035. Growth in natural gas production from non-
OECD countries between 2015 and 2035 is projected to be twice the growth in production from OECD
countries. Expanded production in non-OECD countries is expected to exceed regional demand allowing
for net exports to OECD countries.
Figure 2-4: Global Natural Gas Production: 2015-2035
200
169
158
I Non-OECD
I OECD
2008
2015
2020
2025
2030
2035
Source: U.S. Energy Information Administration (EIA). (2011a). International Energy Outlook 2011. Table G1. World total natural gas
production by region, Reference case, 2008-2035.
Another trend in the international gas market is the increased production of unconventional gas
resources (i.e., tight gas, shale gas, and coalbed methane). Preliminary international estimates suggest
that the quantity of "technically recoverable shale gas resources" is equal to all existing proven natural
gas reserves worldwide (EIA, 2011b). Although the unconventional gas resources have not been fully
assessed, energy experts are projecting significant increases in production over 2035 time horizon. The
most notable increases are expected in the United States, Canada, and China, where unconventional gas
is expected to account for 47, 51, and 72% of domestic production, respectively, in 2035 (EIA, 2011a).
Technology advancements in horizontal drilling and hydraulic fracturing have enabled the United States
to tap into its vast unconventional gas resources. Emerging research on extraction techniques from shale
gas formations suggests there are different emissions profiles compared with conventional gas
production.
11-24
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
OIL AND NATURAL GAS SYSTEMS
1.2.2.2
Emissions Estimates and Related Assumptions
This section briefly discusses the historical and projected emission trends globally and presents the
baseline emissions used in the MAC analysis.2
Historical Emissions Estimates
Emissions from ONG systems globally grew by 31% between 1990 and 2010 with an average annual
growth rate of 1.4%. Key factors that contributed to the growth in emissions include expansions in ONG
production and increases in natural gas consumption.
Projected Emissions Estimates
Worldwide Q-k emissions from ONG systems are projected to increase by 26% between 2010 and
2030 (an average annual rate of 1.2%), slightly lower than in early years (1990-2010). By 2030, the top 5
emitting countries are projected to account for 55% of global emissions in this sector. Although Russia
and the United States remain the largest emitters in this sector, their relative share of the world's
emissions is expected to fall slightly as the ONG industry in Africa and the Middle East expands in future
years. Table 2-2 presents the projected baseline Q-k emissions for the top 5 emitting countries and
remaining country groups by world region.
Table 2-2: Projected Baseline ChU Emissions for Oil and Natural Gas Systems by Country/Region: 2010-
2030 (MtC02e)
Country/Region
2010
2015
2020
2025
2030
CAGR*
(2010-2030)
Top 5 Emitting Countries
Russia
United States
Iraq
Kuwait
Uzbekistan
332.0
247.8
94.1
106.0
84.7
341.9
258.3
109.2
106.9
95.8
382.8
281.6
157.8
103.8
102.7
401.8
307.2
172.9
108.2
104.9
417.9
313.1
187.9
115.9
107.3
1.2%
1.2%
3.5%
0.4%
1.2%
Rest of Regions
Africa
Central and South America
Middle East
Europe
Eurasia
Asia
North America
World Total
274.5
58.0
131.5
42.8
90.8
128.4
86.6
1,677.3
292.2
59.7
138.6
41.6
104.8
141.1
88.2
1,778.3
291.1
67.7
131.4
40.9
112.0
149.7
90.2
1,911.8
302.9
75.8
136.1
41.3
115.7
156.7
97.2
2,020.6
315.1
84.2
137.9
42.4
120.4
164.8
106.1
2,112.9
0.7%
1.9%
0.2%
0.0%
1.4%
1.3%
1.0%
1.2%
aCAGR = Compound Annual Growth Rate
Source: USEPA, 2012a.
For more detail on baseline development and estimation methodology, we refer the reader to the USEPA's Global
Emissions Projection Report available at: http://www.epa.gov/climatechange/economics/international.html. Note that
national emissions inventories are often recalculated when new data become available. The Inventory of U.S.
Emissions and Sinks (the source of United States emissions estimate presented in this report) has been updated since
this analysis was conducted, and the revised 2010 value for oil and gas methane emissions is 174 MtCCfee.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-25
-------
OIL AND NATURAL GAS SYSTEMS
11.2.3 Abatement Measures and Engineering Cost Analysis
Within the four segments of ONG systems, a number of abatement measures can be applied to
mitigate Q-k losses from activities associated with or directly from the operation of equipment and
components. The abatement measures, such as inspection and maintenance programs for leaks or
equipment retrofits or modifications, may be applied to ONG processes and equipment, including
compressors/engines, dehydrators, pneumatics/controls, pipelines, storage tanks, and wells.
Abatement measures available to mitigate Q-k losses from activities associated with or directly from
the operation of equipment components common across the ONG system segments of production,
processing, transmission, and distribution. These abatement options in the ONG system segments
generally fall into three categories: equipment modifications/upgrades; changes in operational and
maintenance practices including DI&M; and installation of new equipment. ONG industry-related
voluntary programs such as the Global Methane Initiative (GMI) and USEPA's Natural Gas STAR
Program, which are aimed at identifying cost-effective Q-k emission reduction opportunities, have
developed a well-documented catalog of potential Cl-k abatement measures that are applicable across the
segments of the ONG system. Abatement measures documented by the USEPA's Natural Gas STAR
Program serve as the basis for estimating the costs of abatement measures used in this analysis. It is
important to note that although abatement measures identified by the Natural Gas STAR Program are
cited as cost-effective based on Industry Partner-reported experiences, the abatement measure's cost-
effectiveness is determined by the component's emissions rate and the value of energy recovered. This
analysis uses average emission factors when estimating the break-even prices for each measure. In many
cases, these average emission rates are lower than the case study examples cited in the Natural Gas STAR
Program's documentation. As a result, abatement measures cited as cost-effective by the Natural Gas
STAR Program's Partners may not necessarily be the lowest cost options in the MAC analysis.
This section discusses the abatement measures considered for this analysis and presents the costs,
benefits, technical applicability, reductions efficiency, and the expected technology lifetime of each
measure. The abatement measures presented in Tables 2.3 through 2.6 provide an overview of the options
considered in each segment of the oil and gas sector. A more complete list of the abatement measures
included in the Oil and Gas Sector MAC model is provided as Appendix D to this chapter.
11.2.3.1 Oil and Natural Gas Production
The production segment of the ONG system consists of wells, compressors, dehydrators, pneumatic
devices, chemical injection pumps, heaters, meters, pipeline, liquid storage tanks, and central
gathering/storage facilities. Table 2-3 presents the list of abatement measures applied to the production
segment of ONG systems. In addition, this section characterizes two important abatement measures
considered in the production segment: reduced emissions from hydraulically fractured gas well
completions and installation of vapor recovery units (VRUs) on crude oil storage tanks.
Reduced Emissions for Hydraulically Fractured Natural Gas Well Completions
Reduced emissions completion (REC) is a method designed to capture 90% of the gas that would
otherwise be flared or vented during new well construction and workovers on existing wells that are
hydraulically fractured. Equipment includes a sand trap, separator, and a gathering line to route gas to
sales pipelines or reserve tanks. Depending on the well field operations and frequency of well
completions, it may be more cost-effective to rent rather than purchase capital equipment (USEPA,
2011a). The use of RECs will result in increased sales of recovered gas. Furthermore, condensate may also
II-26 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
OIL AND NATURAL GAS SYSTEMS
Table 2-3: Abatement Measures Applied in Oil and Gas Production Segments
Total
Installed Annual
Capital Cost O&M
($2008) ($2008)
Abatement Measure
Component
Time Technical
Horizon Effectiveness3
Directed Inspection & Maintenance at
Gas Production Facilities
Installing Surge Vessels for Capturing
Blowdown Vents
Installing Electronic Starters on
Production Field Compressors
Directed Inspection & Maintenance at
Gas Production Facilities
Install Flash Tank Separators on
dehydrators
Optimize glycol circulation rates in
dehydrators
Installing Catalytic Converters on Gas
Fueled Engines and Turbines
Installing Plunger Lift Systems in Gas
Wells
Replace Gas-Assisted Glycol Pumps
with Electric Pumps
Directed Inspection & Maintenance at
Gas Production Facilities
Installing Plunger Lift Systems in Gas
Wells
Directed Inspection & Maintenance on
Offshore Oil Platforms
Flaring Instead of Venting on Offshore
Oil Platforms
Installing Vapor Recovery Units on
Storage Tanks
Using Pipeline Pump-Down Techniques
to Lower Gas Line Pressure Before
Maintenance
Directed Inspection & Maintenance at
Gas Production Facilities
Convert Gas Pneumatic Controls to
Instrument Air
Replacing High-bleed Pneumatic
Devices in the Natural Gas Industry
Chemical Injection —
Pumps
Compressor BD 158,940
Compressor 2,649
Starts
Deepwater Gas —
Platforms
Dehydrator Vents 6,540
Dehydrator Vents —
Gas Engines - 7,924
Exhaust Vented
Gas Well 5,646
Workovers
Kimray Pumps 2,788
Non-associated —
Gas Wells
Non-associated 5,646
Gas Wells
Offshore —
Platforms,
Deepwater oil,
fugitive, vented
and combusted
Offshore 165,888,859
Platforms,
Shallow water Oil,
fugitive, vented
and combusted
Oil Tanks 473,783
Pipeline BD —
Pipeline Leaks —
Pneumatic Device 72,31 1
Vents
Pneumatic Device 165
Vents
6,675
28,078
5,849
50,000
—
15
4,374
(13,855)
1,949
817
(13,855)
50,000
4,976,666
161,507
1,352
82
24,321
—
1
15
10
1
5
1
10
5
10
1
5
1
15
15
1
1
10
10
40%
50%
75%
95%
30% to 60%
33% to 67%
56%
80%
100%
95%
80%
43%
98%
58%
90%
60%
50% to 90%
8% to 17%
(continued)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
II-27
-------
OIL AND NATURAL GAS SYSTEMS
Table 2-3: Abatement Measures Applied in Oil and Gas Production Segments (continued)
Abatement Measure
Component
Total
Installed Annual
Capital Cost O&M
($2008) ($2008)
Time Technical
Horizon Effectiveness3
Directed Inspection & Maintenance at
Gas Production Facilities
Reduced Emission Completions for
Hydraulically Fractured Natural Gas
Wells
Reduced Emission Completions for
Hydraulically Fractured Natural Gas
Wells
Installing Surge Vessels for Capturing
Blowdown Vents
Installing Plunger Lift Systems in Gas
Wells
Shallow water
Gas Platforms
Unconventional
Gas Well
Completions
Unconventional
Gas Well
Workovers
Vessel BD
Well Clean Ups
(LP Gas Wells)
- 33,333
- 30,038
- 30,039
158,940 28,078
5,646 (13,855)
1
1
1
15
5
95%
90%
90%
50%
40%
a Technical effectiveness reflects the percentage reduction achievable from implementing the abatement measure considering the presence of
complementary options. Technical effectiveness is the product of three separate factors—the reduction efficiency, technical applicability, and
market penetration.
b Lower technical effectiveness is due to limited applicability at LP gas wells.
be sold, generating additional revenue. The actual savings generated from these sales also depends on the
market price of gas and gas liquids. Although hydraulically fractured natural gas well completions are
currently limited to the United States and Canada, the analysis assumes that this technology will be
adopted by other countries over time.
• Capital cost: This analysis assumes that natural gas producers rent the REC equipment from a
third-party service provider hence there are no initial capital costs. If a well operator were to be
purchase equipment, the capital cost of the equipment would be approximately $500,000 or more
depending on the complexity of the REC set-up (USEPA, 2011a).
• Annual operation and maintenance (O&M) cost: Cost of implementing this abatement measure
represents the incremental cost of REC to recover the gas over the traditional well completion
cost. The equipment rental costs range between $700 and $6,500/per day (equivalent to $815 to
$7,568 in 2008 dollars). Completions typically take between 3 and 10 days. This analysis assumes
7 days for well clean-up and completions at a cost of $30,000 in 2008 dollars.
• Annual benefits: Revenues may be derived from gas sales from avoided venting/flaring
operations. Additional benefits could come from the sale of recovered natural gas condensate. In
the United States, an average of 34 barrels of condensate are recovered during each completion or
recompletion (USEPA, 2011b). Although the value of the recovered gas condensate would be
determined by the gas composition, based on an assumed price of $70 per barrel (bbl), the
recovered gas condensate would contribute an additional $2,400 in revenues per completion or
recompletion.
• Applicability: This technique applies to hydraulically fractured gas well completions and
workovers.
• Technical Effectiveness: This analysis assumes a technical effectiveness is 54% which is the
product of the 90% reduction efficiency and a technical applicability of 60% and market
penetration of 100%.
11-28
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
OIL AND NATURAL GAS SYSTEMS
• Technical lifetime: 1 implementation event per year per hydraulically fractured well.
Install Vapor Recovery Units (VRUs)
Crude oil and condensate storage tanks are widely used to stabilize the flow of oil or condensate
between wells and transportation sites. Inside these tanks, light hydrocarbons (often with a heavy
concentration of methane) dissolved in the crude oil or condensate flash out of solution and collect
between the liquid and the roof of the tank. As liquid levels fluctuate, vapors are often vented into the
atmosphere. VRUs can capture 95% of these light hydrocarbon vapor emissions (USEPA, 2006a). The
recovered vapors can be sold or used on site as fuel.
• Capital costs: Capital costs range from $40,000 to $120,000 (equivalent to $50,000 to $140,000 in
2008 dollars), depending on the capacity of the unit (between 25 and 500 Mcf per day), sales line
pressure, number of tanks, size and type of compressor, and the degree of automation.
Installation costs range from 50 to 100% of the capital equipment cost and vary depending on the
location of the tanks and the size of the VRU required.
• Annual costs: Incremental annual O&M costs are about 15% of initial cost. The annual costs are
determined by the capacity of the VRU, as well as the location (weather), electricity costs, and the
type of oil produced.
• Annual benefits: VRUs can reduce the hydrocarbon vapor emissions of hydrocarbon liquid
storage tanks by about 95%. The vapors that are recovered can be used in several different ways.
They can be used on site as fuel (where their value is equal to the price of the fuel they displaced).
Alternatively, the vapors can be piped to a natural gas gathering pipeline or to a processing plant
that separates the natural gas liquids and the methane and sells them separately. Because the
recovered vapors generally have a higher Btu content than pipeline quality natural gas, the
vapors are more valuable and sell for a higher price on an energy content basis.
• Applicability: Applied to crude oil and condensate storage tanks
• Technical Effectiveness: The technical effectiveness of this option is 58% based on a reduction
efficiency of 95%, technical applicability of 61%, and a market penetration of 100%.
• Technical Lifetime: 15 years
For detailed discussion of other options available to the ONG production segments, we refer the
reader to USEPA's Natural Gas STAR Program website.
11.2.3.2 Gas Processing and Transmission Segments
The processing segment of the natural gas system consists of gas plant facilities that incorporate the
use of vessels, dehydrators, compressors, acid gas removal (AGR) units, heaters, and pneumatic devices.
The transmission segment consists of transmission pipeline networks, compressor stations, and meter
and pressure-regulating stations. Table 2-4 and Table 2-5 present the list of abatement measures applied
to the gas processing and gas transmissions segment of a natural gas system. Similar to the previous
section, this section briefly characterizes four important abatement measures considered in the gas
processing and transmission segment.
Directed Inspection & Maintenance (DI&M) on Processing Plants and Booster Stations
DI&M is a cost-effective approach to reduce methane emissions from leaking components throughout
the oil and natural gas industry including at natural gas processing plants. The activities include a four-
part process that identifies, prioritizes, and implements the most cost-effective emissions reductions. Step
1 of the process is to identify and measure the leaks using leak detection and measurement techniques.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-29
-------
OIL AND NATURAL GAS SYSTEMS
Table 2-4: Abatement Measures for the Natural Gas Processing Segment
Abatement Measure
Total
Installed Annual
Capital Cost O&M Time Technical
($2008) ($2008) Horizon Effectiveness3
Installing Surge Vessels for Capturing
Slowdown Vents
Directed Inspection & Maintenance
at Processing Plants and Booster
Stations - Compressors
Directed Inspection & Maintenance
at Processing Plants and Booster
Stations - Compressors
Replacing Wet Seals with Dry Seals
in Centrifugal Compressors
Installing Catalytic Converters on
Gas Fueled Engines and Turbines
Replace Gas-Assisted Glycol Pumps
with Electric Pumps
Directed Inspection & Maintenance
at Processing Plants and Booster
Stations
Directed Inspection & Maintenance
at Processing Plants and Booster
Stations - Compressors
Early replacement of Reciprocating
Compressor Rod Packing Rings
Fuel Gas Retrofit for BD valve - Take
Recip. Compressors Offline
Reciprocating Compressor Rod
Packing (Static-Pac)
Blowdowns/Venting 158,940 28,078
Centrifugal — 15,581
Compressors (dry
seals)
Centrifugal — 6,131
Compressors (wet
seals)
Centrifugal 380,804 (102,803)
Compressors (wet
seals)
Gas Engines - 7,924 4,374
Exhaust Vented
Kimray Pumps 2,788 1,949
Plants — 10,134
Recip. Compressors — 6,131
Recip. Compressors 7,800 0
Recip. Compressors 2,365 —
Recip. Compressors 5,696 —
15
1
1
5
10
10
5
1
5
5
5
50%
12%
12%
66%
56%
100%
95%
10%
1%
21%
0%
a Technical effectiveness reflects the percentage reduction achievable from implementing the abatement measure considering the presence of
complementary options. Technical effectiveness is the product of three separate factors: the reduction efficiency, technical applicability, and
market penetration.
Steps 2 and 3 are to assess the measurements to determine which leaks are most cost-effective to repair by
comparing the value of the natural gas lost through leakage to the overall cost of repair. Lastly, in Step 4 a
survey plan is developed for future DI&M to focus efforts on those sources most likely to be leaking and
reduce the cost of subsequent programs. Although the initial expense of the survey can be relatively high,
it was found that the costs can be recovered in the first year through reductions in gas leakage. USEPA
(2003a) documentation suggests the initial baseline survey cost is typically between $1 and $2 per
component on average. Depending on their size, typical processing facilities may have between 14,000
and 55,000 components. Subsequent follow-up surveys are found to cost significantly less compared with
the initial survey, because they are more targeted to the components that are most likely to leak and the
most beneficial to repair.
11-30
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
OIL AND NATURAL GAS SYSTEMS
Table 2-5: Abatement Measures for the Natural Gas Transmission Segment
Abatement Measure
Component
Total
Installed Annual
Capital Cost O&M Time Technical
($2008) ($2008) Horizon Effectiveness3
Directed Inspection and Maintenance
at Compressor Stations -
Compressors
Replacing Wet Seals with Dry Seals in
Centrifugal Compressors
Install Flash Tank Separators on
dehydrators
Optimize glycol circulation rates in
dehydrators
Installing Catalytic Converters on Gas
Fueled Engines and Turbines
Directed Inspection and Maintenance
at Gate Stations and Surface Facilities
Directed Inspection and Maintenance
on Transmission Pipelines
Using Pipeline Pump-Down
Techniques to Lower Gas Line
Pressure Before Maintenance
Convert Gas Pneumatic Controls to
Instrument Air
Replacing High-bleed Pneumatic
Devices in the Natural Gas Industry
Directed Inspection and Maintenance
at Compressor Stations -
Compressors
Early replacement of Reciprocating
Compressor Rod Packing Rings
Early replacement of Reciprocating
Compressor Rod Packing Rings and
Rods
Fuel Gas Retrofit for BD valve - Take
Recip. Compressors Offline
Reciprocating Compressor Rod
Packing (Static-Pac)
Installing Surge Vessels for Capturing
Slowdown Vents
Directed Inspection and Maintenance
at Compressor Stations
Directed Inspection and Maintenance
at Gas Storage Wells
Centrifugal
Compressors (dry
seals)
Centrifugal
Compressors (wet
seals)
Dehydrator Vents
Dehydrator vents
Engine/Turbine Exhaust
Vented
M&R (Trans. Co.
Interconnect)
Pipeline Leaks
Pipeline venting
Pneumatic Devices
Pneumatic Devices
Recip Compressor
Recip Compressor
Recip Compressor
Recip Compressor
Recip Compressor
Station venting
Stations
Wells (Storage)
- 15,581
380,804 (102,803)
9,504 -
- 15
7,924 4,374
- 1,741
- 41
- 1,352
72,311 24,321
165 -
- 15,581
7,800 -
41,068 -
2,365 -
5,696 -
158,940 28,078
- 1,398
- 651
1
5
5
1
10
1
1
1
10
10
1
5
5
5
5
15
1
1
13% to 14%
71% to 77%
67%
67%
56%
72%
60%
90%
50% to 90%
8% to 17%
10% to 12%
1%
1%to74%
36% to 39%
6% to 9%
50%
85%
95%
a Technical effectiveness reflects the percentage reduction achievable from implementing the abatement measure considering the presence of
complementary options. Technical effectiveness is the product of three separate factors: the reduction efficiency, technical applicability, and
market penetration.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-31
-------
OIL AND NATURAL GAS SYSTEMS
DI&M analysis parameters include:
• Capital costs: There is no capital costs associated with this option.
• Annual costs: Initial survey design and leak detection, measurement, and repair. This analysis
assumes a $1 to $2 cost per component for leak detection and repair. The analysis assumes an
average processing plant has approximately 14,000 components.
• Benefits: Gas savings from emission reductions.
• Applicability: Applicable to gas processing, gas gathering and booster stations, gas storage wells,
gate stations and surface facilities, and transmission compressor stations.
• Technical Effectiveness: This analysis assumes a technical effectiveness of 95% based on a 95%
reduction efficiency, a 100% technical applicability factor, and a 100% market penetration factor.
• Technical Lifetime: This analysis assumes a 1 year technical lifetime.
Identify and Replace or Retrofit High-Bleed Pneumatic Devices
Pneumatic devices are widely used as controllers and monitors in the production sector, pressure
regulators and valve controllers in the processing sector, and actuators and regulators in the transmission
sector of the natural gas industry. When driven by natural gas, pneumatic devices release or bleed natural
gas into the atmosphere and thus are a leading source of methane emissions in the natural gas industry.
Replacing high-bleed devices with low-bleed devices and installing low-bleed retrofit kits on operating
devices can reduce emissions by between 50 and 90% (USEPA, 2006b).
• Capital costs: Capital costs are the main component of replacement and retrofitting and vary
greatly among the options. Multiple options can be employed at once to reduce gas bleed. Some
typical options include replacing high-bleed level and pressure controllers with low-bleed
controllers, reducing supply pressure, and repairing leaks. This analysis assumes the capital cost
to be $165, which represents the incremental cost between a high bleed device and a low bleed
device (USEPA, 2011b).
• Annual costs: Some improved maintenance costs are recurring. Maintenance costs are small
relative to the cost of equipment. Replacing and retrofitting devices can potentially reduce annual
maintenance costs. For this analysis the incremental operation and maintenance (O&M) cost is
assumed to be $0.
• Benefits: Revenue from gas savings of reduced methane leakage. Reductions in methane
emissions range from 45 to 260 Mcf per device annually depending on the device and application.
• Applicability: Applicable for high to moderate bleed pneumatic devices in the gas transmission
segments.
• Technical Effectiveness: Technical effectiveness for this option ranges between 8% and 16%
depending on the gas bleed rate. This analysis assumes a reduction efficiency of 9% (low bleed),
23% (medium bleed) and 25% (high bleed). The technical applicability of 50%, 75%, and 90% for
low-, medium-, and high-bleed devices, respectively. Market penetration rate is assumed to be
100% for all devices.
• Technical Lifetime: 10 years
Reducing Methane Emissions from Compressor Rod Packing Systems
In natural gas compressors, the packing systems are used to maintain a tight seal around the piston
rod, preventing unwanted gas leakage while allowing the rod to move freely (USEPA, 2006). Leak rates
depend on the fit, alignment of the packing parts, and wear. New packing systems installed on smooth,
11-32 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
OIL AND NATURAL GAS SYSTEMS
well-aligned compressor rods can be expected to leak as little as 11.5 scfh. Leak rates increase as the
system ages because of wear on the packing rings and piston rod. Regularly monitoring and replacing
these systems can result in cost savings and emissions reductions. This abatement measure is applied to
compressors in the gas processing and transmission segments of the natural gas system. Packing systems
comprise flexible rings that are secured around the compressor shaft. Packing cups hold the rings in
place, and a nose gasket reduces leaks around the packing cups. Conventional packing rings have a life
expectancy of about 3 to 5 years, but when the packing breaks down, leaks tend to increase so
dramatically that it may be desirable to replace packing rings even more frequently. A new, well-
functioning system could leak as little as 11 standard cubic feet per hour (scfh), compared with worn
compressor rod packing systems that have leak rates as high as 900 scfh (USEPA, 2006c).
• Capital costs: Replacement compressor rod packing systems range from $7,800 per unit to
replace the packing rings to over $41,000 for replacement of the piston rods and packing rings.
• Annual costs: There are no annual costs for these options.
• Benefits: Revenue from gas savings of reduced methane leakage.
• Applicability: Applies to reciprocating compressors located at processing plants and compressor
stations in the transmission segment
• Technical Effectiveness: Technical effectiveness for this option is 1.5%, based on a reduction
efficiency of 10% a technical applicability of 15%, and a market penetration of 100%.
• Technical Lifetime: 5 years.
Replacing Wet Seals with Dry Seals in Centrifugal Compressors
Centrifugal compressors are used in the production, processing, and transmission of natural gas. The
seals located on the rotating shafts to reduce methane leakage have traditionally been "wet" (oil) seals.
Replacement of wet seals with dry seals leads to substantially reduced emissions and operating costs. The
dry seals are the only piece of capital equipment required and may be installed during a scheduled
downtime. The lifetime of dry seals may be double that of wet seals, and they also emit significantly
lower emissions. It has been estimated that the wet seals may pay for themselves in as little as 11 months
(USEPA, 2006d). Other benefits include lower electricity requirements and maintenance costs and
increased operating efficiency of the compressor and pipeline, which may also lead to higher gas sales.
• Capital costs: This analysis assumes a capital cost of $381,000 in 2008 dollars for wet seal
replacement on a compressor with a shaft beam size of 6 inches. Cost of dry seals ($15,200 per
shaft inch) represents 48% of initial capital costs; equipment testing services (-0.5% of equipment
cost); engineering, procurement, and construction (EPC) services were assumed to be 100% of
equipment and testing costs.
• Annual costs: O&M costs of dry seals are expected to be less than O&M costs for wet seals
because of reduced electricity requirements, increased operating efficiency of the compressor,
increased reliability of the compressor, and potentially lower maintenance costs. Hence,
incremental recurring costs are assumed to equal a cost savings of just over $100,000 each year.
These incremental cost savings are added to the annual benefits resulting from increased gas
sales.
• Benefits: Revenue from gas savings of reduced methane leakage. Other annual cost savings due
to lower operation and maintenance costs are captured in the annual costs.
• Applicability: Applies to centrifugal compressors located at gas processing plants and
compressor stations in the transmission segment.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-33
-------
OIL AND NATURAL GAS SYSTEMS
• Technical Effectiveness: Technical effectiveness for this option is 66%. This value is based on a
reduction efficiency of 85%, a technical applicability of 78% and a market potential of 100%.
• Technical Lifetime: 5 years
11.2.3.3 Gas Distribution Segment
The distribution segment consists of main and service pipeline networks, meter and pressure-
regulating stations, pneumatic devices, and customer meters. Table 2-6 presents the list of abatement
measures applied to the distribution segment of a natural gas system. DI&M activities' cost and benefit
components are discussed below.
Table 2-6: Abatement Measures for the Distribution Segment
Abatement Measure
Component
Total
Installed
Capital Cost
($2008)
Annual
O&M
($2008)
Time Technical
Horizon Effectiveness3
Directed Inspection and Maintenance
at Gate Stations and Surface Facilities
Replace Cast Iron Pipeline
Replace Unprotected Steel Pipeline
Replace Unprotected Steel Service
Lines
M&R<100
Mains— Cast Iron
Mains-
Unprotected steel
Services-
Unprotected steel
—
373,633
373,633
418,023
1,604
182
182
311
1
5
5
5
30% to 80%
95%
95%
95%
a Technical effectiveness reflects the percentage reduction achievable from implementing the abatement measure considering the presence of
complementary options. Technical effectiveness is the product of three separate factors: the reduction efficiency, technical applicability, and
market penetration.
DI&M at Gate Stations and Surface Facilities
Leaking meters, pipes, valves, flanges, fittings, open-ended lines, and pneumatic controllers at gate
stations and surface facilities are a significant source of methane emissions. DI&M is a proven and cost-
effective way to detect, measure, prioritize, and repair equipment leaks to reduce methane emissions and
achieve gas savings (USEPA, 2003b). To implement DI&M, first, a baseline survey identifies and
quantifies leaks at gate stations and surface facilities. The results of this survey are then used to direct
repairs toward the components that were identified as being most susceptible to leaking and the most
profitable to repair. Then, the results of the initial survey are used to guide subsequent inspections and
maintenance.
• Capital costs: There are no capital costs associated with this option.
• Annual costs: The costs associated with starting a DI&M program are the cost of labor and
equipment for identifying leaking components and estimating the mass leak rate of those
components; the labor cost for recording survey information; the labor cost of pinpointing
leaking components that are cost-effective to repair; the cost of parts, labor, and equipment
downtime to fix the leaks; and the cost of labor for developing a plan that directs future
inspection and maintenance. Costs differ depending on the type of screening and measurement
equipment used and the characteristics of the staff who conduct the surveys and repairs.
Maintenance and repair are ongoing, so most costs are recurring. Annual costs vary depending
on the frequency and comprehensiveness of the surveys and repairs. Over time, the scope and
frequency of the surveys can be fine-tuned, as patterns emerge. This analysis assumes an average
annual cost of $1,600 in 2008 dollars.
11-34
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
OIL AND NATURAL GAS SYSTEMS
• Benefits: Gas savings and methane emissions reductions vary widely depending on the number
of stations involved in the DI&M program and how long the program has been operating.
• Applicability: Applies to components inside gate stations and surface facilities.
• Technical Effectiveness 30% to 80%; higher efficiency for facilities handling higher volumes of
natural gas. Technical effectiveness measure assumes reduction efficiency between 30% and 80%,
technical applicability of 90% to 100% and a market penetration of 100%.
• Technical Lifetime: 5 years
11.2.4 Marginal Abatement Costs Analysis
This section discusses the methodological approach used to conduct the international MAC analysis
in the ONG sector.
11.2.4.1 Methodological Approach
The MAC analysis approach consists of four sequential steps. Step 1 was to assess the sectoral trends,
which entailed reviewing recent international energy statistics for oil and gas. The second step was to
develop source-level emission estimates that could be used to build different model ONG systems. These
model systems reflect country-specific variations in production process techniques, level of maintenance,
and vintage of the existing infrastructure. Step 3 was to estimate country-specific abatement costs and
benefits based on the relative cost factors for labor, energy, and non-energy inputs. Step three was to
compute the break-even prices for each country-specific abatement measure. Finally, the MAC model
computes the abatement potential as a cumulative reduction for each measure assuming full (system-
wide) implementation. Sorting the break-even prices lowest to highest, the incremental reductions are
cumulated to construct the MAC curve presented in Section II.2.4.2.
Assessment of Sectoral Trends
The objective in assessing the sectoral trends is to understand how emissions differ across countries
and how they vary over time. This not only considers aggregate growth or decline in emissions but also
any potential shift in sector emissions across the oil and gas segments. To this end, we reviewed the
current international gas and oil industry activity data for 2010. Statistics reviewed included gross natural
gas production, oil production, LNG imports, and gas processing throughput (EIA, 2011; Oil & Gas
Journal, 2011). In the absence of real infrastructure data, these statistics provide insights on the relative
importance of segments internationally. Table 2-7 presents these key statistics for the 10 largest emitting
countries in 2010.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-35
-------
OIL AND NATURAL GAS SYSTEMS
Table 2-7: International Statistics on Key Activity Drivers: 2010
Country
2010 Emissions
(MtC02e)
Dry Natural Gas Crude Oil
Production3 Production15
(Bcf/year) (Mbbl/day)
Gas Processing
Plant Gas
Throughput0 Transmission
(MMcfd) Pipelines'1 (km)
Russia
United States
Kuwait
Iraq
Angola
Uzbekistan
Libya
Canada
Iran
Venezuela
332.0
247.8
106.0
94.1
84.9
84.7
77.4
53.3
47.2
30.2
22,965
26,858
422
596
364
2,123
1,069
6,695
7,774
2,510
10,146
9,688
2,450
2,408
1,988
105
1,789
3,483
4,252
2,375
926
45,808
1,034
1,550
137
NA
2,567
29,154
10,509
3,555
160,952
548,665
269
3,365
-
10,253
-
75,835
20,725
5,347
a EIA. International Energy Statistics: Gross Natural Gas Production.
b EIA. International Energy Statistics: Total Oil Supply.
c Oil & Gas Journal [OGJ|. June 6,2011. Worldwide Processing Survey.
d CIA. 2011. The World Factbook.
e EIA. 2012. Country analysis Brief-Uzbekistan. Available at: http://www.eia.gov/countries/cab.cfm?fips=UZ
Although differences in annual production and throughput provide some indication of the size of a
country's ONG system, considerations of age and the condition of the infrastructure are major factors in
determining the rate of source-level emissions and in turn the abatement potential associated with each
abatement measure. In general, countries with aging infrastructure will have "leakier" components and
in turn have a greater abatement potential. Conversely, countries with newly developed infrastructure
will have less abatement potential.
Another important trend to consider is the expansion of unconventional gas (shale gas) production.
The growth in unconventional gas production (e.g., the United States, Canada, and China) is likely to
result in an increased frequency of hydraulically fractured gas well completions and related workovers.
In the absence of any regulatory or voluntary actions to reduce emissions from these sources, this trend
suggests that the gas production segment will represent an even greater proportion of these nations'
baseline emissions over time.
Defining International Model Facilities for the Analysis
For this analysis, we developed model ONG systems for each segment based initially on the USEPA
ONG system emissions inventory. Scaling factors were developed based on country-specific activity
factors developed from the international statistics illustrated in Table 2-6. Where reliable data were
available, international adjustments were made to reflect specific country systems. For countries for
which data were not available, this analysis assumed the gas and oil system was similar to that in the
United States in terms of the distribution of emissions (total BAU emissions for each country is exogenous
to the MAC model obtained from USEPA, 2012a). The relative international factor was multiplied by the
percentage share of U.S. gas and oil CH4 emissions inventory at the segment/component source level
(e.g., compressors, valves, connections, pneumatic devices). The resulting "Technical Applicability" (TA)
factor is used to allocate a fraction of the national baseline emissions to each component source in the
ONG inventory (e.g., wells, tanks, compressors, valves).
II-36
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
OIL AND NATURAL GAS SYSTEMS
Multiplying the TA factor by the baseline emissions yields the subset of emissions available for
reductions from each component source and abatement measure. The TA factor comprises two parts. The
U.S. 2010 GHG emissions inventory serves as the basis for the distribution of emissions across the
constituent components (see USEPA, 2012b, Annex 3). The second component of the TA factor is the
country- and segment-specific relative activity factor (e.g., total oil production, gross natural gas
production).
Regulatory Considerations for the U.S. Natural Gas and Oil System
Special considerations were made for the United States to reflect the New Source Performance
Standards (NSPS) regulation that was in effect starting in year 2012. This regulation will affect the
production and processing segments of the ONG system by requiring the use of abatement measures
included in this analysis to control for volatile organic compound (VOC) emissions from major new and
modified emitting sources in the United States. This mitigation is no longer considered additional and
thus should be removed from the U.S. domestic MAC curve. For the purposes of this analysis, we have
removed emissions sources covered under the NSPS regulation and any subsequent abatement potential
that would have come from these sources. Figure 2-5 illustrates the key elements that led us to the
resulting distribution of emissions.
Figure 2-5: Diagram of BAD Emissions for the United States Oil and Natural Gas System
Residual
from NSP
11% of P
Baseline
i
-—- -
missions
5 sources
rejected
t>2015)
~— • -_
residual emissions
from NSPS controlled sources
Non-NSP Sources
US Natural Gas and Oil Systems Inventory in 2010
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-37
-------
OIL AND NATURAL GAS SYSTEMS
To capture the impact of the NSPS regulation for model years 2015 and beyond, we needed to
estimate the relative distribution of emissions that will be associated with controlled and uncontrolled
sources. We start with the 2010 U.S. inventory (USEPA, 2012), which is the basis of our analysis. Next, we
identify all components in the inventory subject to the NSPS in the production and processing segments.
For the NSPS controlled sources, we applied controls to these components using the reduction
efficiency (%) for each abatement measure from the MAC model to estimate the level of abatement
achieved by the NSPS rule. In Figure 2-5, controlled emissions equals the sum of reductions achieved
across the NSPS sources. The residual emissions from NSPS sources are assumed to be included in the
projected baselines (2015 to 2030).
For the purposes of this analysis, we estimated residual emissions from controlled sources were 11%
of projected emissions, while emissions from uncontrolled sources were 89% of projected emissions. We
applied these shares to the baseline projections for years > 2015.
This approach assumes a fixed distribution of emission over time in the MAC model. We recognize
the limitations of this assumption and would ideally like to apply a trend to the shares for model years
beyond 2015. Unfortunately, at the time of writing this report, data to develop this trend were not
available. Any future work related to the U.S. MAC curve should consider developing a more dynamic
trend that more accurately estimates the level of NSPS-controlled emissions and the subsequent
distribution of emissions in the baseline projections over time.
Based on the analysis described here, the United States' abatement potential presented in the MAC
modeling results can be summarized in the following expression:
Abatement Potential (USA) = y Uncontrolled Emissionsit * Technical Effectiveness^ t
where:
Technical Effectiveness^ = Reduction Efficiencyy,t * Tech Applicability^ * Market Penetration^t
i = Uncontrolled emissions source
j = Abatement technology
t = Modeled year
Estimate Abatement Project Costs and Benefits
Turning to the abatement measures discussed in Section II.2.3, the analysis begins with technology
costs for the United States as reported in the USEPA Lesson Learned documentation. We applied the
Nelson-Farrar3 Oil Field and Refinery Operation cost indices to convert from reported-year costs to 2008
dollars (USD) for capital and O&M costs, respectively. Next, we applied the country-specific relative
price factors for labor, energy, and nonenergy components of annual costs and benefits. This final step
yielded country-specific costs and benefits used to compute the break-even price for each abatement
measure. Abatement measure costs and technical efficiencies were applied to estimate the break-even
prices. Table 2-7 presents the break-even prices for selected ONG system abatement measures for the
United States in 2010. For this analysis, we used the abatement measure costs, revenue, and reduction
efficiency as described in Section II.2.3 to estimate the break-even price for each abatement measure. A
complete list of ONG system abatement measures is presented in Appendix C.
1 Nelson-Farrar Annual Cost Indices are available in the first issue of each quarter of the Oil and Gas Journal.
11-38 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
OIL AND NATURAL GAS SYSTEMS
The first step is to estimate the reduced emissions on a per unit basis for each technology. This value
is calculated by multiplying the abatement measure's technical efficiency by the annual emissions per
unit of the component or process to which the abatement measure is being applied. The resulting annual
reduced emissions serve as the denominator in the break-even price calculation.
In Table 2-8 we present abatement cost and revenues per metric ton of CCh equivalent (tCO2e)
reduced for the abatement measures with the largest national emissions reductions. Costs include the
annualized total installed capital cost and annual O&M costs. Offsetting these costs are the annual
revenue in terms of gas savings and the tax benefit of depreciation. The break-even prices reported in
Table 2-8 are calculated by subtracting the annual revenues from the annualized costs.
Table 2-8: Example Break-Even Price Calculation based on 2010 MAC for the United States
Abatement Measure
System
Component/
Process
Reduced
Emissions
per Unit
(tC02e)
Annualized
Capital
Costs
($/tC02e)
Annual
Cost
($/tC02e)
Annual
Revenue
($/tC02e)
Tax Benefit
of
Depreciation
($/tC02e)
Break-
Even
Price
($/tC02e)
National
Incremental
Reductions
(MtC02e)
Production
Convert Gas Pneumatic
Controls to Instrument Air
Reduced Emission
Completions for Hydraulically
Fractured Natural Gas Wells
Replacing High-bleed
Pneumatic Devices in the
Natural Gas Industry
Pneumatic Device
Vents
Unconventional Gas
Well Completions
Pneumatic Device
Vents
71.0
2,703.96
9.7
$335.68
$0.00
$7.38
$441.41
$11.11
$0.00
$10.01
$10.01
$10.01
$82.50
$0.00
$1.81
$684.58
$1.10
-$4.44
15.29
8.82
2.30
Processing
Directed Inspection &
Maintenance at Processing
Plants and Booster Stations
Fuel Gas Retrofit for BD
valve -Take Recip.
Compressors Offline
Replacing Wet Seals with Dry
Seals in Centrifugal
Compressors
Plants
Recip. Compressors
Centrifugal
Compressors (wet
seals)
1,109.0
351.9
5,000.8
$0.00
$2.96
$33.48
$9.14
$0.00
-$20.56
$10.01
$10.01
$10.01
$0.00
$0.90
$10.15
-$0.87
-$7.95
-$7.24
0.50
1.34
2.53
Transmission
Convert Gas Pneumatic
Controls to Instrument Air
Directed Inspection and
Maintenance at Compressor
Stations
Fuel Gas Retrofit for BD
valve -Take Recip.
Compressors Offline
Pneumatic Devices
Stations
Recip Compressor
89.9
3,655.9
1,014.8
$2,898.32
$0.00
$1.07
$3,811.28
$0.41
$0.00
$10.01
$10.01
$10.01
$712.36
$0.00
$0.32
$5,987.24
-$9.60
-$9.26
2.88
6.61
5.65
Distribution
Directed Inspection and
Maintenance at Gate Stations
and Surface Facilities
Directed Inspection and
Maintenance at Gate Stations
and Surface Facilities
Replace Cast Iron Pipeline
MSR >300
MSR 100-300
Mains— Cast Iron
511.6
220.2
91.7
$0.00
$0.00
$1,790.73
$3.40
$7.90
$1.99
$10.01
$10.01
$10.01
$0.00
$0.00
$543.06
-$6.60
-$2.10
$1,239.65
1.58
2.48
2.54
Note: Break-even price assumes a 10% discount rate and a 40% tax rate. Annual energy benefits are based on a natural gas price of $4/Mcf
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
II-39
-------
OIL AND NATURAL GAS SYSTEMS
From Table 2-8, the annualized capital cost are calculated using the total installed capital costs
discussed in Section II.2.3 and expressed in the following equation:
Total Capital Cost
Annualized Costs = —
- TR) • £[=1 —
Where:
ER = Annual reduced emissions per unit (e.g. compressor, well, dehydrator, etc.)
TR = Tax rate
T = Technology lifetime in years
DR = Discount rate
Annual O&M costs and expected revenues are calculated using the following equations.
International variation in break-even prices is achieved by using regionally adjusted prices for energy
labor and materials when computing the country specific annual costs and benefits.
Annual O&M Costs
Annual O&M Costs =
C t\
Annual Revenues
Annual Revenues =
£j t\
The tax benefit of depreciation is calculated for each option using the following equation:
Total Capital Costs TR
Tax Benefit of Depreciation =
ER-T (1 -
Finally, the break-even price is calculated by subtracting the benefits from the costs as shown in the
equation below.
Break- even Price
= Annualized Capital Cost + Annual O&M Cost — Annual Revenues
— Tax Benefit of Depreciation
11.2.4.2 MAC Analysis Results
As highlighted at the beginning of this chapter, global abatement potential related to ONG systems
equates to approximately 58% of total annual emissions. MAC curve results are presented in Table 2-9
and Figure 2-6. Maximum abatement potential for ONG systems is 1,218 MtCO2e in 2030. For the year
2030, the results suggest that 842 MtCO2e or 40% of CH4 reductions in the ONG sector can be achieved at
carbon prices less than or equal to $5/tCO2e. Furthermore over 35% of the 2030 emission reductions (747
MtCO2e) are cost-effective at current energy prices (carbon prices < $0/tCO2e). However, because natural
gas prices vary greatly by region, the break-even price and quantity of cost effective reductions varies by
country.
11-40 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
OIL AND NATURAL GAS SYSTEMS
Table 2-9: Abatement Potential by Region at Selected Break-Even Prices in 2030
Country/Region
-10
EJB
•EH
^•_n~R^v
m
^"ji\yiwpiti
Top 5 Emitting Countries
Iraq
Kuwait
Russia
United States
Uzbekistan
69.2
37.3
37.3
79.1
9.9
71.9
44.0
37.3
81.0
9.9
73.9
46.7
138.3
84.4
35.4
76.0
47.6
185.7
93.3
47.6
80.3
47.7
187.0
93.4
47.9
80.9
50.1
189.6
98.5
48.5
80.9
50.7
205.3
98.8
52.4
85.1
51.1
205.8
99.2
52.6
85.2
55.2
219.5
105.5
57.8
89.9
58.6
232.8
109.7
59.6
110.0
73.5
266.9
140.5
68.3
Rest of Region
Africa
Central and South
America
Middle East
Europe
Eurasia
Asia
North America
World Total
116.3
31.8
43.2
15.0
22.9
40.4
29.9
532.2
124.0
32.7
51.6
16.0
23.3
55.8
32.0
579.3
124.1
32.8
53.3
16.3
42.6
59.2
39.3
746.5
129.4
34.2
55.3
16.8
51.0
62.7
42.7
842.3
135.8
36.2
57.8
17.2
52.3
69.4
43.3
868.1
136.2
36.3
58.5
17.4
53.5
71.0
44.4
884.9
141.4
37.4
58.7
18.0
56.9
72.1
44.9
917.5
142.2
38.3
60.7
18.2
57.0
73.6
45.2
928.9
145.0
38.4
61.2
18.8
61.6
75.1
46.5
969.8
149.3
40.6
63.6
19.8
64.2
76.7
47.7
1,012.4
178.8
49.4
78.4
26.1
76.7
90.6
59.6
1,218.6
Figure 2-6: Marginal Abatement Cost Curves for Top 5 Emitters in 2030
«*, $20
o
o
Non-C02 Reduction (MtC02e)
•Iraq
Russia
•Kuwait
•United States
Uzbekistan
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-41
-------
OIL AND NATURAL GAS SYSTEMS
The MAC illustrates the cumulative abatement achievable at incrementally higher carbon prices. At
extremely high break-even prices (> $500/tCO2e), the MAC becomes inelastic or unresponsive. The point
at which the MAC becomes unresponsive to any price change can also be considered the technical
potential associated with the suite of abatement measures considered. Thus, it can be inferred that
additional reductions beyond approximately 58% of the projected baseline in 2030 would be unlikely
without additional policy incentives or technology improvements.
Economies of scale have an impact on the cost-effectiveness of the abatement options. Hence,
abatement measures may have a lower break-even price when applied to facilities with higher Q-k
emission rates and higher break-even price at facilities with a lower emissions rate.
11.2.4.3 Uncertainties and Limitations
Several key areas of uncertainty constrain the accuracy of this analysis. Addressing these
uncertainties would improve the development of the MACs and predictions of their behavior as a
function of time. Two primary limitations are discussed below.
Improved information on the distribution of emissions in international baselines. This analysis
relies on historical activity factors to adjust the distribution of U.S. baseline emissions to develop
projections by country. Improvements to information on how gas and oil baselines are changing over
time and across segments would improve the accuracy of abatement potential estimates.
Complete information on current abatement technologies used in the gas and oil industry
internationally. Additional information on the current and planned implementation of abatement
measures internationally would improve the international estimates of abatement.
11-42 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
OIL AND NATURAL GAS SYSTEMS
References
Intergovernmental Panel on Climate Change (IPCC). (2006). 2006 IPCC Guidelines for National Greenhouse
Gas Inventories: Volume 2 Energy. Chapter 4 Fugitive Emissions. Obtained from: http://www.ipcc-
nggip.iges.or.jp/public/2006gl/vol2.html.
U.S. Energy Information Administration (EIA). (2011a). International Energy Outlook 2011. DOE/EIA—
0484(2011). Washington, DC: EIA. Obtained from: http://www.eia.gov/forecasts/ieo/nat gas.cfm.
U.S. Energy Information Administration (EIA). (2011b). World Shale Gas Resources: An Initial Assessment of
U Regions Outside the United States. Washington, DC: EIA. Obtained from:
http://www.eia.gov/analysis/studies/worldshalegas/.
U.S. Energy Information Administration (EIA). (2011c). International Energy Statistics: Gross Natural Gas
Production. Washington, DC: EIA. Obtained from:
http://tonto.eia.doe.gov/cfapps/ipdbproject/IEDIndex3.cfm?tid=3&pid=3&aid=l.
U.S. Energy Information Administration (EIA). (2011d). International Energy Statistics: Total Oil Supply.
IEA: Washington D.C. Obtained from:
http://www.eia.gov/cf apps/ipdbproject/IEDIndex3.cfm?tid=5&pid=53&aid=l.
U.S. Energy Information Administration (EIA). (2012). Country Analysis Brief - Uzbekistan. Washington,
DC: EIA. Obtained from: http://www.eia.gov/countries/country-data.cfm?fips=UZ.
U.S. Environmental Protection Agency and Gas Research Institute (USEPA/GRI). (1996). Methane
Emissions from the Natural Gas Industry Volume 2: Technical Report. EPA-600/R-96-080b. Washington,
DC: USEPA.
U.S. Environmental Protection Agency (USEPA). (2003a, October). Directed Inspection and Maintenance at
Gas Processing Plants and Booster Stations. Lessons Learned from Natural Gas STAR Partners.
Washington, DC: USEPA.
U.S. Environmental Protection Agency (USEPA). (2003b, October). Directed Inspection and Maintenance at
Gate Stations and Surface Facilities. Lessons Learned from Natural Gas STAR Partners. Washington,
DC: USEPA.
U.S. Environmental Protection Agency (USEPA). (2006a, October). Installing Vapor Recovery Units on
Storage Tanks. Lessons Learned From Natural Gas STAR Partners. Washington, DC: USEPA.
U.S. Environmental Protection Agency (USEPA). (2006b, October). Options for Reducing Methane Emissions
from Pneumatic Devices in the Natural Gas Industry. Lessons Learned From Natural Gas STAR Partners.
Washington, DC: USEPA.
U.S. Environmental Protection Agency (USEPA). (2006c, October). Reducing Methane Emissions from
Compressor Rod Packing Systems. Lessons Learned From Natural Gas STAR Partners. Washington, DC:
USEPA.
U.S. Environmental Protection Agency (USEPA). (2006d, October). Replacing Wet Seals with Dry Seals in
Centrifugal Compressors. Lessons Learned from Natural Gas STAR Partners. Washington, DC: USEPA.
U.S. Environmental Protection Agency (USEPA). (2010). Greenhouse Gas Emissions Reporting from the
Petroleum and Natural Gas Industry. Background Technical Support Document. Washington, DC:
USEPA. Obtained from: http://www.epa.gov/climatechange/ernissions/downloadslO/Subpart-
W TSD.pdf.
U.S. Environmental Protection Agency (USEPA). (2011a). Reduced Emission Completions for Hydraulically
Fractured Natural Gas Wells. Lessons Learned from Natural Gas STAR Partners. Washington, DC:
USEPA.
U.S. Environmental Protection Agency (USEPA). (2011b). Oil and Natural Gas Sector: Standards of
Performance for Crude Oil and Natural Gas Production, Transmission, and Distribution—Background
Technical Support Document for Proposed Standards. EPA-453/R-11-002. Washington, DC: USEPA.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-43
-------
OIL AND NATURAL GAS SYSTEMS
U.S. Environmental Protection Agency (USEPA). (2012a). Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA. Obtained from:
http://www.epa.gov/climatechange/econornics/international.html.
U.S. Environmental Protection Agency (USEPA). (2012b). U.S. Greenhouse Gas Inventory Report: Inventory
of U.S. Greenhouse Gas Emissions and Sinks: 1990-2010. EPA 430-R-12-001. Washington, DC: USEPA.
Obtained from: http://epa.gov/climatechange/emissions/usinventoryreport.html.
U.S. Environmental Protection Agency (USEPA). (2012c). 2011 Annex 3: Methodological Descriptions for
Additional Source or Sink Categories. Sections 3.4 and 3.5. In 2022 U.S. Greenhouse Gas Inventory
Report: Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2009. EPA 430-R-12-001.
Washington, DC: USEPA. Obtained from:
http://epa.gov/climatechange/emissions/usinventoryreport.html.
Natural Gas Star Program Lessons Learned Documents Referenced in Model
U.S. Environmental Protection Agency (USEPA). October 2006. Reducing Methane Emissions from
Compressor Rod Packing Systems: Lessons Learned from Natural Gas Star Partners. Washington,
DC: USEPA. Obtained April 2, 2013 from http://www.epa.gov/gasstar/tools/recommended.htrnl.
U.S. Environmental Protection Agency (USEPA). October 2006. Replacing Wet Seals with Dry Seals in
Centrifugal Compressors: Lessons Learned from Natural Gas Star Partners. Washington, DC: USEPA.
Obtained on April 2, 2013 from http://www.epa.gov/gasstar/tools/recommended.html.
U.S. Environmental Protection Agency (USEPA). October 2006. Reducing Emissions When Taking
Compressors Offline: Lessons Learned from Natural Gas Star Partners. Washington, DC: USEPA.
Obtained on April 2, 2013 from http://www.epa.gov/gasstar/tools/recommended.html.
U.S. Environmental Protection Agency (USEPA). October 2006. Replacing Glycol Dehydrators with
Desiccant Dehydrators: Lessons Learned from Natural Gas Star Partners. Washington, DC: USEPA.
Obtained on April 2, 2013 from http://www.epa.gov/gasstar/tools/recommended.html.
U.S. Environmental Protection Agency (USEPA). October 2006. Optimize Glycol Circulation and Install
Flash Tank Separators in Dehydrators: Lessons Learned from Natural Gas Star Partners. Washington,
DC: USEPA. Obtained on April 2, 2013 from http://www.epa.gov/gasstar/tools/recommended.html.
U.S. Environmental Protection Agency (USEPA). October 2003. Directed Inspection and Maintenance at
Compressor Stations: Lessons Learned from Natural Gas Star Partners. Washington, DC: USEPA.
Obtained on April 2, 2013 from http://www.epa.gov/gasstar/tools/recommended.html.
U.S. Environmental Protection Agency (USEPA). October 2003. Directed Inspection and Maintenance at
Gate Stations and Surface Facilities: Lessons Learned from Natural Gas Star Partners. Washington,
DC: USEPA. Obtained on April 2, 2013 from http://www.epa.gov/gasstar/tools/recommended.html.
U.S. Environmental Protection Agency (USEPA). October 2003. Directed Inspection and Maintenance at
Gas Processing Plants and Booster Stations: Lessons Learned from Natural Gas Star Partners.
Washington, DC: USEPA. Obtained on April 2, 2013 from
http://www.epa.gov/gasstar/tools/recommended.html.
U.S. Environmental Protection Agency (USEPA). October 2006. Replacing Gas-Assisted Glycol Pumps
with Electric Pumps: Lessons Learned from Natural Gas Star Partners. Washington, DC: USEPA.
Obtained on April 2, 2013 from http://www.epa.gov/gasstar/tools/recornmended.html.
U.S. Environmental Protection Agency (USEPA). October 2006. Using Pipeline Pump-Down Techniques
to Lower Gas Line Pressure Before Maintenance: Lessons Learned from Natural Gas Star Partners.
Washington, DC: USEPA. Obtained on April 2, 2013 from
http://www.epa.gov/gasstar/tools/recommended.html.
U.S. Environmental Protection Agency (USEPA). October 2006. Composite Wrap for Non-Leaking
Pipeline Defects: Lessons Learned from Natural Gas Star Partners. Washington, DC: USEPA.
Obtained on April 2, 2013 from http://www.epa.gov/gasstar/tools/recommended.html.
11-44 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
OIL AND NATURAL GAS SYSTEMS
U.S. Environmental Protection Agency (USEPA). October 2006. Using Hot Taps for In Service Pipeline
Connections: Lessons Learned from Natural Gas Star Partners. Washington, DC: USEPA. Obtained
on April 2, 2013 from http://www.epa.gov/gasstar/tools/recommended.html.
U.S. Environmental Protection Agency (USEPA). October 2006. Convert Gas Pneumatic Controls to
Instrument Air: Lessons Learned from Natural Gas Star Partners. Washington, DC: USEPA. Obtained
on April 2, 2013 from http://www.epa.gov/gasstar/tools/recommended.html.
U.S. Environmental Protection Agency (USEPA). October 2006. Options for Reducing Methane Emissions
From Pneumatic Devices in the Natural Gas Industry: Lessons Learned from Natural Gas Star
Partners. Washington, DC: USEPA. Obtained on April 2, 2013 from
http://www.epa.gov/gasstar/tools/recommended.html.
U.S. Environmental Protection Agency (USEPA). (2006a, October). Installing Vapor Recovery Units on
Storage Tanks. Lessons Learned From Natural Gas STAR Partners. Washington, DC: USEPA.
Obtained from http://www.epa.gov/gasstar/tools/recommended.html.
U.S. Environmental Protection Agency (USEPA). 2011. Reduced Emission Completions for Hydraulically
Fractured Natural Gas Wells: Lessons Learned from Natural Gas Star Partners. Washington, DC:
USEPA. Obtained on April 2, 2013 from http://www.epa.gov/gasstar/tools/recornmended.htrnl.
U.S. Environmental Protection Agency (USEPA). October 2006. Installing Plunger Lift Systems in Gas
Wells: Lessons Learned from Natural Gas Star Partners. Washington, DC: USEPA. Obtained on April
2, 2013 from http://www.epa.gov/gasstar/tools/recommended.html.
U.S. Environmental Protection Agency (USEPA). 2011. Options for Removing Accumulated Fluid and
Improving Flow in Gas Wells: Lessons Learned from Natural Gas Star Partners. Washington, DC:
USEPA. Obtained on April 2, 2013 from http://www.epa.gov/gasstar/tools/recommended.html.
U.S. Environmental Protection Agency (USEPA). 2011. Install Electric Motor Starters: Partner Reported
Opportunities (PROs) for Reducing Methane Emissions. PRO Fact Sheet No. 105. Washington, DC:
USEPA. Obtained on April 2, 2013 from http://www.epa.gov/gasstar/tools/recommended.html.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-45
-------
I. Waste Sector
-------
LANDFILLS
1.1. Landfill Sector
1.1.1 Sector Summary
andfills produce methane (Q-k) in combination with other landfill gases (LFGs) through the
1 natural process of bacterial decomposition of organic waste under anaerobic conditions. LFG is
generated over a period of several decades, with gas flows usually beginning 1 to 2 years after
the waste is put in place. Q-k makes up approximately 50% of LFG. The remaining 50% is carbon dioxide
(CCfe) mixed with small quantities of other gases, including volatile organic compounds (VOCs). The
amount of CH4 generated by landfills per country is determined by a number of factors that include
population size, the quantity of waste disposed of per capita, composition of the waste disposed of, and
the waste management practices applied at the landfill. Changes in these key factors drive projected
trends in CH4 emissions. For a number of countries, LFG is one of the largest anthropogenic sources of
CH4 emissions. Despite efforts to control large landfill emissions, the landfill sector remains a significant
source of CH4 emissions because of increasing waste streams in developed countries. In developing
countries, the shift toward sanitary landfills and increased use of abatement measures is a key driver
toward CH4 mitigation.
In 2010, global CH4 emissions from landfills accounted for approximately 850 MtCC^e. Emissions
from landfills are moderately concentrated in several countries. Over 50% of emissions in 2010 come from
just ten countries. Figure 1-1 displays the business-as-usual (BAU) emissions for the landfill sector and
identifies the top five emitting countries. Landfill emissions are projected to grow 13% between 2010 and
2030. In 2030, emissions from landfills represent 10% of the global total CH4 from all sources.
Figure 1-1: Emissions Projections for the Landfill Sector: 2000-2030
959
i Malaysia
i Russia
China
i Mexico
United States
ROW
2000
2010 2020
Year
2030
Source: U.S. Environmental Protection Agency (USEPA), 2012
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LANDFILLS
Several abatement measures are available to control landfill Q-k emissions and they are commonly
grouped into three major categories: (1) collection and flaring, (2) LFG utilization systems (LFG capture
for energy use), and (3) enhanced waste diversion practices (e.g., recycling and reuse programs).
Although flaring is currently the most common abatement measure, energy recovery options may be
more cost-effective. Similarly, under favorable market conditions, recycling and reuse or composting
alternatives may provide additional means for reducing emissions from landfills. Note that options may
not be mutually exclusive in that recycling can reduce the quantity of methane generated, which, in turn,
will affect the economics of utilization systems.
Global abatement potential in the solid waste landfill sector is estimated to be approximately 589
MtCO2e of total annual emissions in 2030, or 61% of the baseline emissions. The marginal abatement cost
(MAC) curve results are presented below in Figure 1-2. These curves suggest that there are significant
opportunities for CH4 reductions in the landfill sector at carbon prices below $20. Furthermore there are
approximately 70 to 80 MtCO2e of reductions that are cost-effective (no regret options) at current energy
prices.
Figure 1-2: Global Abatement Potential in Landfill Sector: 2010, 2020, and 2030
•2010
•2020
•2030
Non-C02 Reduction (MtC02e)
The following section briefly explains CH4 emissions from landfills. This is followed by the
international baseline CH4 emissions projections from landfills. Subsequent sections characterize the
abatement technologies and present the costs and potential benefits. The chapter concludes with a
discussion of the MAC analysis and the regional results.
II-2
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LANDFILLS
1.1.2 Methane Emissions from Landfills
This section discusses the characteristics of landfills and how these characteristics affect
emissions. In this section, we also describe historical and projected trends that influence baseline
emissions from municipal solid waste (MSW) landfills. By volume, LFG is about half CH4 and half CCh.
Typically, LFG also contains small amounts of nitrogen, oxygen, and hydrogen; less than 1% non-Q-k
volatile organic compounds (NMVOCs); and trace amounts of inorganic compounds. The amount and
rate of Q-k generation depend on the quantity and composition of the landfilled material, as well as the
site design and resulting physical conditions inside the fill.
Organic waste is initially decomposed by aerobic bacteria after being landfilled. When the oxygen in
the landfill cell (section of a landfill) is depleted, the remaining waste is broken down by anaerobic
bacteria through decomposition. Fermentation creates gases and short-chain organic compounds that
form the substrates, which provide for the growth of methanogenic bacteria, which in turn generates a
biogas consisting of approximately 50% CCh and 49% CH4, by volume. Measurable gas volumes are
generally available between 1 or 2 years after the waste is landfilled and continue to be generated for 10
to 60 years.
The amount and rate of Q-k production over time at a landfill depends on five key characteristics of
the landfill material and surrounding environment:
• Quantity of Organic Material: The quantity of organic material, such as paper, food, and yard
waste, is crucial to sustaining CH4-producing microorganisms. The Q-k production capacity of a
landfill is directly proportional to its quantity of organic waste. CH4 generation increases as the
waste disposal site continues to receive waste and then gradually declines after the site stops
receiving waste.
• Nutrients: CH4-generating bacteria need nitrogen, phosphorus, sulfur, potassium, sodium, and
calcium for cell growth. These nutrients are derived primarily from the waste placed in the
landfill.
• Moisture Content: The bacteria need water for cell growth and metabolic reactions to convert
cellulose to CH4. Landfills receive water from incoming waste, water produced by
decomposition, surface water infiltration (precipitation), groundwater infiltration (in unlined
landfills). In general, CH4 generation occurs at slower rates in arid climates than in nonarid
climates.
• Temperature: Warm temperatures in a landfill speed the growth of CH4-produting bacteria. The
temperature of waste in the landfill depends on landfill depth, the number of layers covering the
landfill, and the regional climate.
• pH: CH4 is produced in a neutral acidic environment (dose to pH 7.0). The pH of most landfills is
between 6.8 and 7.2. Above pH 8.0, CH4 production is negligible.
The methodology for estimating CH4 emissions from municipal solid waste landfills in this analysis is
based on the first order decay model (Intergovernmental Panel on Climate Change [IPCC], 2006).
The key characteristics described above can vary considerably across the different types and features
of the waste disposal site, and this, in turn, influences landfill CH4 generation. This analysis considers
abatement measures' impacts on three model facilities representing the solid waste management
alternatives with different levels of methane generating capacity. The following are the model facilities
considered:
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 111-3
-------
LANDFILLS
Open dump sites: defined as solid waste disposal facilities where the waste is left uncompacted and
without cover. The waste in open dump sites is relatively shallow, therefore promoting aerobic
biodegradation. This model facility is particularly relevant to developing countries where solid waste
management practices are not well established. These facilities generate relatively small amounts of
methane and for this, and safety reasons, have more limited applicability of mitigation technologies,
which are less effectiveness where applicable.
Basic landfills (also referred to as managed dump sites): defined as solid waste disposal facilities
where the waste is compacted and covered but do not have additional engineered systems. These
facilities generate methane and in some cases can be modified to support an oxidation system and/or a
gas collection and flaring or energy recovery system. However, the collection efficiency1 may not be as
efficient with a capture efficiency of approximately 75%. These facilities represent the baseline in most
developing countries.
Engineered sanitary landfills: defined as facilities that include not only waste compaction and cover
but they also are designed and constructed with gas and leachate collection systems. The higher degree of
engineering at these facilities generally allows for more efficient gas collection and control than basic
sanitary landfills. Engineered landfills typically have a collection efficiency of around 85%. These facilities
represent the majority of baseline emissions in major industrialized countries.
III.1.2.1 Activity Data or Important Sectoral or Regional Trends and Related
Assumptions
This section discusses the historical and projected activity factor data that determine Q-k generation
at solid waste disposal sites and policies set to improve waste management practices. Historical and
projected changes in population and household income are used as indicators of changes in the quantity
and type of consumption, which are directly linked to the quantity and type of waste generated by
countries.
For developing and emerging economies, the projected baseline emissions reflect assumptions about
population growth, economic growth, and changes in waste management practices over time in (USEPA,
2012). Continued growth in population along with increased household income and improvements in
waste management practices will result in the growth of both waste generated and waste disposed of in
managed and engineered landfills.
For developed countries with stable or declining growth in population and income, consumption is
assumed to result in only small increases in emissions over time. Developed countries are also assumed
to increasingly engage in waste diversion practices (e.g., recycling and composting) that divert
biodegradable waste from landfills, ultimately changing the composition of landfilled waste and
lowering the annual methane generated over time.
1 Collection efficiency refers to the amount of methane generated in the landfill that is captured by the collection
system. In contrast, the reduction efficiency refers to the share of collected methane that is destroyed. For example
flare have a reduction efficiency of approximately 98%.
111-4 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LANDFILLS
III.1.2.2 Emissions Estimates and Related Assumptions
This section briefly discusses the historical and projected emission trends globally and presents the
baseline emissions used in the MAC analysis.2
Historical Emissions Estimates
Emissions from landfills were estimated to have grown by 13% between 1990 and 2010. Key factors
that contribute to the growth in landfill emissions include population growth, growth in personal income,
increased industrialization, and improvements in waste management practices (USEPA, 2012).
Projected Emissions Estimates
Worldwide Q-k emissions from landfills are expected to increase at an average long run annual rate
of 0.6% (USEPA, 2012). Although some of the largest economies in the world continue to emit significant
quantities of Q-k, developing and emerging economies are projected to account for majority of growth in
CH4 emissions. Table 1-1 presents the projected baseline Q-k emissions for the top five emitting countries
and remaining country groups by world region.
Table 1-1: Projected Base
Country
line Emission
2010
s for MSW La
2015
ndfills by Coi
2020
jntry: 201 0-2030
MtC02e)
CAGR
2030 (2010-2030)
Top 5 Emitting Countries
China
Malaysia
Mexico
Russia
United States
47.1
29.9
56.4
47.2
129.7
48.2
32.5
59.5
46.1
128.4
49.0
35.1
62.5
44.8
127.7
49.4
37.8
65.2
43.4
128.0
49.3 0.2%
40.3 1.5%
67.7 0.9%
42.1 -0.6%
128.0 -0.1%
Rest of Regions
Africa
Central & South America
Middle East
Europe
Eurasia
Asia
North America
World Total
101.2
71.4
67.3
87.2
55.8
133.2
20.3
846.7
106.5
74.2
72.3
92.4
58.6
135.1
21.9
875.6
111.9
76.8
77.1
96.8
61.5
138.4
23.3
905.0
117.3
79.1
81.7
100.9
64.3
141.5
24.8
933.3
122.4 1.0%
81.1 0.6%
86.1 1.2%
104.6 0.9%
66.8 0.9%
144.4 0.4%
26.5 1.3%
959.4 0.6%
aCAGR = Compound Annual Growth Rate
Source: USEPA. 2012.
2 For more detail on baseline development and estimation methodology, we refer the reader to the USEPA's Global
Emissions Projection Report available at: http://www.epa.gov/climatechange/economics/international.html.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES III-5
-------
LANDFILLS
The United States is the largest emitter of landfill CH4, accounting for over twice the emissions of the
second largest emitter, Mexico. Although emissions from the top 4 emitters observed in 2010 are
projected to remain relatively constant, emissions from developing regions including Africa, non-
Organisation for Economic Co-Operation and Development (OECD) Asia, and the Middle East are all
projected to have annual grow rates of greater than 1%. This trend reflects higher population growth
rates, changing consumption patterns, and improved waste management systems among developing
nations.
III.1.3 Abatement Measures and Engineering Cost Analysis
This analysis considers two types of abatement measures: mitigation technologies and diversion
alternatives (see Table 1-2). It is important to note the distinction between these two approaches to
emission reductions. Mitigation technologies represent add-on technologies that can be applied to one or
more landfill types (i.e., open dump, basic landfill, engineered landfill) intended to capture and destroy
the Q-k generated at the facility. Diverting organic waste from the landfill for alternative uses is the
second approach to reduce the quantity of LFG generated at existing landfills. As noted previously, these
measures are not mutually exclusive. By changing the composition of waste that is landfilled, diversion
options lower the methane-generating potential of remaining waste that is landfilled. Diversion
alternatives are covered in this analysis but are distinguished from landfill-based mitigation technologies.
This section discusses the abatement measures considered for this analysis. Each technology is briefly
characterized followed by a discussion of abatement measures' implementation costs, potential benefits,
and system design assumptions used in the MAC analysis.
Table 1-2: Summary of the Engineering and Cost Assumptions for Abatement Measures at Landfills
Abatement Option
Total Installed
Capital Cost
(millions 2010 USD)
Annual
O&M Cost
(millions 2010 USD)
Time
Horizon
(Years)
Reduction
Efficiency
(%)a
LFG Mitigation Options
LFG collect and flaring system
LFG for electricity generation
Internal combustion engine
Gas turbine (> 3 MW)
Micro-turbine (< 1 MW)
Combined heat and power production
Direct gas use
Enhanced oxidation systems
1.7
6.3
5.6
4.1
7.9
2.6
5.4
0.3
0.8
0.6
0.1
0.8
0.5
0.0
15
15
15
15
15
15
50
85%
85%
85%
85%
85%
85%
85%
44%
Waste Diversion Options
Composting
Anaerobic digestion
Mechanical biological treatment
Paper recycling
Waste to energy
1.8
16.9
15.4
34.9
165.7
0.7
1.7
1.8
8.9
8.0
15
20
20
20
20
95%
95%
95%
95%
100%
a Reduction efficiency reflects the abatement measures ability to mitigation/avoid methane generation. However this does not reflect the total
mitigation potential.
III-6
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LANDFILLS
III.1.3.1 Landfill CH4 Mitigation Technologies
This section characterizes the mitigation technologies that can be applied to landfills to reduce
emissions. Mitigation options considered for this analysis include collection of LFG for flaring, collection
for electricity production, collection for direct use, and enhanced oxidation systems.
LFG Collection and Flaring
Most basic landfills and engineered landfills have (or are applicable for) LFG collection systems for
both public health and facility safety concerns to prevent high concentrations of LFG in the fill. These
systems prevent the migration of CH4 to on-site structures and adjacent property and prevent the release
of non-CH4 organic compounds (NMOCs) to the atmosphere. Wells and gathering lines may be
constructed in advance or installed after waste has been landfilled. LFG collection usually begins after a
portion of a landfill is closed. Collection systems are configured either as vertical wells (which are most
common), horizontal trenches (which are primarily used for deeper landfills and landfill cells that are
actively being filled), or a combination of the two. Trenches or wellheads are connected to lateral piping
that transports the LFG to a collection header. Typically there is a collection system monitor installed to
allow operators to adjust the gas flow (USEPA, 2010).
Flares ignite and burn LFG. Large landfills have historically collected CH4 and flared the gas.3 Flare
designs include open and enclosed flares. Enclosed flares are more expensive but provide greater control
of combustion conditions, allow for stack testing, reduce light and noise nuisances, and might have
higher combustion efficiencies (USEPA, 2010).
• Capital Cost: Capital cost includes the construction of wells, wellheads, and laying of gathering
lines that make up the collection system, as well as the flare system with monitoring and control
systems. Costs were derived from the USEPA Landfill Methane Outreach Program (LMOP)
Project Cost Estimation Model. The capital costs assume one well per acre installed at an average
installation cost of $150/ft. Installation of the wellheads and gathering lines is approximately
$17,000 per acre. Installed cost of the knockout blower and flare system is based on open flares
with the maximum expected flow of LFG per minute ($963/maximum cubic feet per meter [cfm]).
• Annual Operation and Maintenance (O&M) Cost: Typical annual O&M costs for collection
systems are $2,250 per well and $4,500 per flare. Electricity costs to operate the blower for a 600
cfm active gas collection system average $44,500 per year4 (USEPA, 2010), assuming an electricity
price of 7 cents/kWh and consumption rate of 0.002 kWh per ft3.
Annual Benefits: No economic benefits (energy production) are associated with this option.
Applicability: This option applies to all basic landfills and engineered landfills.
Technical Efficiency: This analysis assumes a collection efficiency of 75% for basic landfills and
of 85% for engineered landfills and a flaring efficiency of 98%.
Technical Lifetime: 15 years
3 Flares are typically a required component of energy recovery projects. In energy recovery projects, the flare system
is used to control LFG emissions during energy generation startups and downtime and may also be used to control
excess gas production.
4 For this analysis we assume an electricity price of 7.5 cents/kWh and an energy consumption rate of 0.002 kWh/ft3.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 111-7
-------
LANDFILLS
LFG Collection for Electricity Generation
Converting LFG to electricity offers a potentially cost-effective way to use the gas being generated by
the landfill. Often, revenue from the sale of energy produced can provide a cash flow that more than
offsets the implementation costs of this option. This option requires a LFG collection and flare system as
described earlier in this section, as well as the electricity generation system. Components of the electricity
generation system include the equipment for generating energy (e.g., internal combustion engine, gas
turbine, or microturbine) and the interconnections for transmitting electricity produced to the energy
grid.
LFG is extracted from landfills using a series of vertical or horizontal wells and a blower (or vacuum)
system. This system directs the collected gas to a central point, where it can be processed and treated
depending on the ultimate use of the gas. LFG treatment removes moisture and other contaminants (e.g.,
siloxanes) that may disrupt the energy generation equipment (USEPA, 2010). Treatment requirements
depend on the end-use application.
This analysis considers four alternative technologies under this abatement measure that include
internal combustion engine, gas turbine, micro-turbine, and combined heat and power (CHP) approach.
Table 1-3 summarizes the typical costs for the alternative electricity-generating technologies.
• Capital Cost: Capital cost includes the costs of the collection and flare system discussed and the
treatment system, energy generation equipment, and interconnection equipment for selling
electricity to the power grid. Costs were derived from the USEPA LMOP Project Cost Estimation
Model, which is available at USEPA's LMOP web page. Costs ranged from $1,400 to $5,500 per
Kwh (see Table 1-3).
• Annual O&M Cost: Typical annual O&M costs for energy generation systems are between $130
and $380 per kilowatt of capacity.
• Annual Benefits: Annual revenues are derived from the sale of electricity.
• Applicability: This option applies to all basic landfills and engineered landfills.
• Technical Efficiency: This analysis assumes a collection efficiency of 75% for basic landfills and
85% for engineered landfills and combustion efficiency of 98%.
• Technical Lifetime: 15 years
Table 1-3: Electricity Generation Technology Costs
Technology
Internal combustion engine (> 0.8 MW)
Small 1C engine (< 1 MW)
Gas turbine (> 3 MW)
Microturbine (< 1 MW)
CHP with 1C engine (< 1 MW)
Capital Cost
(2010 $/kW)
$1,700
$2,300
$1,400
$5,500
$2,300
Annual O&M Costs
(2010 $/kW)
$180
$210
$130
$380
$210
Source: USEPA 2010. U.S. Environmental Protection Agency (USEPA). September 2010. Project Development Handbook. Chapter 3. Project
Technology Options. Landfill Methane Outreach Program. Obtained from: http://www.epa.aov/lmop/publications-tools/tfone.
Note: Costs include the cost of the basic treatment system typically required with each type of technology.
III-S
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LANDFILLS
LFG Collection for Direct Use
Direct use provides an alternative use of LFG with minimal treatment. Under this option, LFG
collected at the landfill is pumped to a nearby (< 5 miles) end user. The gas delivered can serve as a
medium-BTU fuel for boiler or drying operations, kiln operations, and cement and asphalt production.5
Although little condensate removal and filtration is needed, combustion equipment might need slight
modifications to run with LFG (USEPA, 2010). However these modification costs are not considered part
of the technology costs.
There is no cost-effective way to store LFG, so ideally the LFG consumer has a steady annual gas
demand compatible with the landfill's gas flow. If a landfill does not have adequate flow, the LFG can be
used to power only a portion of the machinery or mixed with other fuels. The cost for a gas compression
and treatment system includes compression, moisture removal, and filtration equipment necessary for
transporting and using the gas.
• Capital Cost:6 The capital costs for direct use include the equipment and installation cost of a
skid-mounted filter, compressor, and dehydrator, and the cost to construct a gas pipeline to carry
the gas to a nearby (< 5 miles) end user(s). Filter, compressor, and dehydrator costs are scaled to
the project's expected minimum LFG flow and equal to approximately $300 per cfm. Pipeline
construction costs are assumed to be $320,000 per mile.
• Annual Cost: Annual O&M costs include the cost of electricity and maintenance of the filters,
compressors, and dehydrators. The electricity costs are calculated by multiplying electricity price
times the energy required to power the equipment and transmit gas to end users, assuming a
system power demand of 0.002 kWh/ft3. Non energy-related O&M costs are scaled to LFG project
volumes assuming a cost of $0.0014/ft3.
• Benefits: Annual revenue accrues to the project through the sale of LFG to an end user at an
assumed price that is 80% of the current natural gas price; the discounted price reflects the lower
BTU content of the gas. There may also be local or national policies such as tax incentives, loans,
and grants available to landfill operators to incentivize LFG utilization.
• Applicability: This option is available to all basic landfills and engineered landfills.
• Technical Efficiency: This analysis assumes a collection efficiency of 75% for basic landfills and
85% for engineered landfills and an end-use combustion efficiency of 98%.
• Technical Lifetime: 15 years
Enhanced Oxidation Systems
Enhanced oxidation systems are considered mitigation technologies that exploit the propensity of
some naturally occurring bacteria to oxidize CH4.7 By providing optimum conditions for microbial
5 Other direct use applications include use in infrared heaters, greenhouses, artisan studios, leachate evaporation,
and biofuel production.
6 It is important to note that direct use of LFG may require equipment modifications at the end-user site to handle the
lower BTU content of LFG or additional treatment systems to improve the energy content; these costs are not
considered part of this abatement measure's project costs. Including these costs would increase project costs by more
than $200,000 (USEPA, 2010).
7 Oxidation of methane entails mixing the gas (CIHk) with oxygen (02) and converting the CH4 to CO2 and water
(H20).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 111-9
-------
LANDFILLS
habitation and efficiently routing landfill gases to where they are cultivated, a number of bio-based
systems, such as temporary or long-term biocovers, passively or actively vented biofilters, and
biowindows, have been developed that can alone, or with gas collection, mitigate landfill CH4 emissions.
The previous non-CCh mitigation report (USEPA, 2006) evaluated the use of a biocover consisting of a
clay cap topped by a soil cover.
• Capital Cost: Capital costs are the incremental costs of enhanced oxidation systems above the
traditional day/soil cover. These costs assume an incremental cost of $6 million for 100 acres of
cover. The cost of designing and constructing the biocover assumes $3/yd3 for earth moving, a
compost price of $5/tonne,8 and an average cover depth of 3 feet.
Annual O&M Cost: The O&M cost is assumed to be less than 0.1% of installed capital costs.
Annual Benefits: No revenues are associated with this option.
Applicability: This option applies to basic landfills and engineered landfills.
Technical Efficiency: This option analysis assumes a reduction efficiency of 44% of the
remaining 15% of methane not collected by LFG collection system (Weitz, 2011).
Technical Lifetime: 50 years
III.1.3.2 Diversion Alternatives
Diversion alternatives redirect biodegradable components of the waste stream from the landfill for
reuse through recycling or conversion to a value-add product (e.g. energy or compost). Diverting organic
waste components such as yard waste, paper, and food waste lowers the amount of methane generated at
the landfill. These measures derive benefits through the sale of recyclables (both organic and non-
organic), electricity, and cost savings in avoided tipping fees. Although these options were considered in
the previous mitigation report (USEPA, 2006), all diversion options were not included in the final
mitigation estimates reported. The following diversion alternatives were considered for this analysis:
• composting
• anaerobic digestion (AD) for electricity production from gas
• mechanical biological treatment (MET)
• paper recycling
• waste to energy
Composting
Composting consists of the aerobic digestion of the fermentable organic fraction of MSW to produce a
reusable product. In the presence of oxygen, microorganisms decompose the biodegradable organic
matter to form compost, which contains nutrients and trace elements, and is used in agriculture as soil
conditioner. The composting process emits a gas basically formed by CCh and l-hO, while traces of (VOCs
are also present. This analysis considers three types of composting processes—windrow composting,
aerated static pile (ASP) composting, and in-vessel composting—but cost and emissions data were only
obtained for windrow composting because it is the most common type.
Windrow composting processes occur in the open, usually in long rows of triangular cross-sections,
these being turned periodically to introduce air into the process. The material received by the composters
is processed, formed into a windrow, turned (using portable diesel-powered equipment), and screened
! The compost price assumes a weight by volume of 0.32 tonnes/yd3 (DST Model Documentation).
11-10 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LANDFILLS
prior to sale. A typical facility will accept both green material and wood waste from residential curbside
programs and an increasing number of composting facilities are beginning to accept food scraps from
residential curbside programs, as well as from dedicated commercial routes or large generators.
Windrow composting processes may have Q-k emissions from anaerobic decomposition and nitrous
oxides (N2O) emissions from NOx denitrification during the latest composting stages. The IPCC (2006)
provides representative Q-k emissions of 4 to 10 g/Kg of waste (dry weight) and N2O emissions of 0.3 to
0.6 g/kg waste (dry weight).
• Capital Costs: Capital cost includes the purchase of land and equipment, site preparation and
facility construction equal to $1.8 million (2010 USD). Capital costs were obtained from the
composting process model documentation of the Municipal Solid Waste Decision Support Tool
(MSW DST) (MSW DST Documentation), which presents this cost for 100 tons/day facilities
producing marketable high-quality compost products as opposed to nonmarketable, low-quality
compost product (e.g., used as landfill cover).
• Annual Cost: The O&M cost of the windrow composting facility includes the labor, overhead,
fuel, electricity, and equipment maintenance costs.9 This analysis assumes an O&M cost of
$19/tonne-yr (obtained from the composting process model documentation of the MSW DST
(MSW DST Documentation).
• Annual Benefits: Revenue from compost is from sales and cost savings from avoided landfilling.
The composting process is not perfectly efficient, and this analysis assumes that 80% of the
incoming organic waste is converted to marketable compost product. A compost price of
$5/tonne10 was used to estimate the revenue from compost sales. A tipping fee of $29/tonne is
used to estimate the costs savings of avoided landfilling.
• Applicability: This option applies to yard and food components of the waste stream.
• Technical Efficiency: This analysis assumes reduction efficiency of 95%, which represents the
avoided methane potential.
• Technical Lifetime: 15 years
Anaerobic Digestion (AD)
AD is a complex biological process that uses anaerobic microorganisms to hydrolyze complex
organics to simple monomers and hence to volatile fatty acids; the volatile fatty acids are converted to
CH4 and CCh in the biogasification step. The biogas can be recovered and used to generate energy.
Existing AD facilities are most commonly located at wastewater treatment plants, but the process is
equally applicable for solid waste. A few of these facilities supplement their operations with other types
of organic waste.
Solid waste AD facilities come in different shapes and sizes. Most digesters have vertical tanks, but
some are horizontal. AD mechanisms vary considerably, and a number of patented processes exist.
Processes may operate at high or low solids content, operate at mesophilic or thermophilic temperatures,
be one- or two-stage systems, and be continuous or batch processes. The process could also differ
9 This analysis assumes that no precomposting screening will take place. Therefore, there will not be organics rejects
from the process needing disposal at a landfill facility, which is consistent with the data provided for high quality
compost production in the composting process model documentation of the MSW DST (MSW DST Documentation).
10 Represents the lower end price $15 to 34/yard3 assuming a 0.35 tonne/yard3. Prices reported in Recycle.cc's
December 2011 newsletter. Obtained at: http://vyvyyy.recycle.ee/compostprices.pdf
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 111-11
-------
LANDFILLS
according to the type of product produced, so some processes only produce electricity, others produce
combined electricity and heat, and some produce gas upgraded for use as vehicle fuel. This analysis
considers AD that produces electricity using a gas engine, which is the most common product. A small
amount of Q-k may be released as fugitive emissions during the digestion process. This analysis assumes
CH4 emissions of 1 to 2 g/kg of waste (dry weight) as reported in IPCC (2006).
• Capital Costs: The plant's capital cost includes the cost of land, the digesters, the gas engine, and
air pollution control and monitoring devices. The capital cost for this analysis is $472/design
tonne was considered in this analysis and obtained from Eunomia (2008), which describes this
cost for facilities of 20,000 to 30,000 tonnes/yr in the United Kingdom (UK).
• Annual Cost: The O&M cost of the AD facility includes the labor, overhead, fuel, electricity, and
maintenance cost. An O&M cost of $55/tonne yr1 (reported as £35 GBP/tonne) was considered in
this analysis and obtained from Eunomia (2008), which presents costs typical of UK facilities. This
analysis assumes that no predigestion screening will take place and that the digested solids are
not commercialized. Therefore, there will be no organics rejects from the process needing
disposal at a landfill facility.
• Annual Benefits: Revenue from the sale of electricity generated with the biogas is sold to an end
user. The biogas recovery from the digestion process is not perfectly efficient and assumed to be
75% of total value, and the biogas composition is assumed 60/40% Q-k/CCh according to
Eunomia (2008). Similarly, the efficiency of the biogas conversion to electricity in the gas engine is
assumed to be 37% as reported by Eunomia (2008). The electricity produced per tonne of waste
can be then estimated according to the Q-k yield (2,781 ft3 Q-k/wet ton) of the incoming waste.
The market price of electricity is used to estimate the revenues.
• Applicability: This option assumes removal of wood, paper, and food waste.
• Technical Efficiency: This analysis assumes a capture efficiency of 75% and a reduction
efficiency of 95%.
• Technical Lifetime: 20 years
Mechanical Biological Treatment (MBT)
MET can be defined as the processing or conversion of solid waste with biologically degradable
components via a combination of mechanical and other physical processes (for example, cutting or
crushing, sorting) with biological processes (aerobic composting, anaerobic digestion). The primary
objective is to reduce the mass and the volume of the waste. A secondary objective is a lower
environmental impact of the waste after its deposition (i.e., low emissions of landfill gas, small amounts
of leachate, and a reduced settlement of the landfill body). Furthermore, MBT includes the separation of
useful waste components for industrial reuse, such as metals, plastics, and refuse-derived fuel (RDF).
There are three main types of biological treatment processes: (1) an aerobic stabilization system in
which the stabilized output is assumed to be sent to a landfill or used for land remediation/recovery
projects, (2) an aerobic biodrying system producing an RDF with the reject stream sent to a landfill (after
undergoing an aerobic stabilization process), and (3) systems combining aerobic and anaerobic
treatments in which the anaerobic process is used to produce biogas, followed by an aerobic process that
produces a stabilized output that can be sent to a landfill. Because of the similarities that can be found
between Option (1) and composting, and Option (3) and AD, this analysis focuses on Option (2) in which
the RDF is destined for energy generation.
To produce RDF, both windrow and box systems are applied. In box systems, the waste is treated
aerobically for only 1 week but with high aeration rates. The result is a dried material with a slightly
11-12 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LANDFILLS
reduced organic content. Only the most easily degradable compounds are metabolized so that the loss of
caloric value is low. The dry material can be fractionated very easily, because adhesive substances were
eliminated in the bio-process. Iron and nonferrous metals, as well as glass and minerals, are separated for
material recovery. The remaining material has a calorific value of 15 to 18 MJ/kg, mainly due to the high
content of plastics, wood, and paper. It can be used as a substitute for fossil fuels in power stations and
cement kilns and in the production of process gases. Similar to the composting process, there is a small
level of fugitive Q-k emissions that accompany the aerobic degradation process as well as some N2O
emissions from NOx denitrification during the curing stages of the stabilization process. Representative
Q-k emissions of 0.01 kg/tonne of waste and N2O emissions of 0.02 kg/tonne of waste were obtained from
Eunomia (2008).
• Capital Costs: The plant's capital cost includes the cost of land, facility, equipment, and air
pollution control and monitoring devices. The analysis assumes a capital cost of $15 million
based on reported facility costs of $244/design tonne (reported as £150 British pounds/tonne) was
used for this analysis and obtained from Eunomia (2008). Costs are reported for a 60,000 tonne/yr
facility in the UK.
• Annual O&M Costs: The O&M cost of the MBT facility is $2 million in 2010. This cost includes
the labor, overhead, taxes, administration, insurance, indirect costs, energy, and maintenance
costs. It does not include residues disposal. A 2007 annual O&M cost of $22/tonne (reported as
£13 British pounds/tonne) was considered in this analysis and obtained from Eunomia (2008),
which presents costs typical of UK facilities.
• Annual Benefits: Annual revenues from the sale of RDF and recydables that are produced from
the MBT process are sold to an end user (i.e., cement kilns or coal-fired utility). According to
Eunomia (2008), RDF is produced at a typical rate of 0.48 tonne/tonne of waste. Eunomia (2008)
also reports that 1 tonne of RDF can be assumed to replace 0.90 tonne of coal used to fuel a
cement kiln and 0.38 tonne of coal for power generation. The market coal price of $40/tonne is
used to estimate the revenues. Similarly, Eunomia (2008) reports an 80% recovery rate for ferrous
metals, 70% recovery rate for nonferrous metals, and 70% recovery rate for glass. Sale prices of
$352/tonne for ferrous metals (USGS, 2012), $l,881/tonnen for nonferrous metals, and $25/tonne
for glass were used to estimate the revenues from recydables sale.
Applicability: This option applies to all landfill types
Technical Efficiency: This analysis assumes a reduction efficiency of 95%.
Technical Lifetime: 20 years
Paper Recycling
Recycling typically consists of two major processes: the separation process at a material recovery
facility (MRF) and the remanufacturing process where recydables are used to produce new produds. For
consistency with other mitigation option induded in this report, the costing component of this analysis
only considers the separation process. The different types of MRFs vary according to the type of waste
they receive and the destination of the recydables (e.g., mixed waste MRF, commingled recydables MRF,
presorted recydables MRF, co-collection MRFs, and front-end MRFs to other waste diversion alternatives
such as composting). Because it is the most common, this analysis considers a mixed waste MRF.
11
Price obtained from MetalPrices.com at http://vyyyyy.metalprices.com/FreeSite/metals/al scrap/al scrap.asp#Tables.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 111-13
-------
LANDFILLS
Under the mixed waste MRF design, mixed waste is typically collected at curbside and dumped on a
tipping floor at the MRF. It is then loaded onto a conveyer by using a front-end loader or Bobcat. This
conveyer feeds a bag opening station because most waste is collected in bags. Undesirable items in the
waste (e.g., white goods, bulky items) are removed from the mixed waste before and after the bag
opening point in the MRF. Bags can be opened either manually or mechanically, and this analysis
considers mechanical bag opening. Loose waste from the bag opening operation is then conveyed into an
elevated and enclosed sorting room where the recyclables are recovered. Newsprint, old corrugated
cardboard, and other paper can be picked from the mixed waste as individual components. Because other
paper components are present in small quantities and are likely to be wet and contaminated, they can
only be recovered as mixed paper. Metal cans remain in the refuse on the conveyer at the end of the sort
room. Separation of aluminum cans can be manual or automated, and this analysis assumes manual
separation. Ferrous metal is assumed to be recovered by a magnet.
Apart from power consumption, no residual greenhouse gas (GHG) emissions are assumed, and the
MRF facility costs are divided into three components: capital cost, O&M cost, and revenue from
recyclables sale.
• Capital Costs. The capital cost for this option is $35 million in (2010 USD). The capital cost
consists of construction, engineering, and equipment costs. It assumes a handling capacity of
100,000 tonnes of waste per year. This analysis relies on a $297/tonne of annual capacity (2006
prices), which is an average of reported capital costs from CalRecycle (2009) for similar sized
facilities.
• O&M Cost. The O&M cost of the MRF facility includes wages, overhead, equipment and
building maintenance, and utilities. An O&M cost of $66/tonne of annual waste capacity before
residue disposal, based on reported operating costs used in CalRecycle (2009) report. The cost of
disposal of the MRF rejects can be estimated assuming an MRF separation efficiency of 55% of the
incoming organic waste and that the rejects are sent to a regular landfill with a tipping fee of
$29/tonne, which represents a U.S. national average tipping fee obtained from Municipal Solid
Waste Facility Directory (Chartwell, 2004).
• Annual Benefits: Annual benefits come from the sale of recyclables and decreased waste. The
recyclables that are separated at the MRF are sold to an end user (e.g., a remanufacturing facility)
sometimes through brokers. The 55% separation efficiency and recyclables sale prices were used
to estimate the revenues from recyclables sale. The following prices were used in the analysis:
mixed paper12—$140/tonne; scrap metals13—$l,307/tonne; and scrap glass—$25/tonne. Tonnage
sold for reuse avoids landfilling costs. Annual cost savings are equal to tonnage sold for reuse
times the tipping fee of $29/tonne.
Applicability: This option applies to the entire waste stream.
Technical Efficiency: This analysis assumes a reduction efficiency of 95% of potential methane.
Technical Lifetime: 20 years
12 Prices were obtained from: http://vyyyyy.recycle.ee/freepapr.htm.
13 Assumes a weighted average price of aluminum can scrap and ferrous metal scrap prices. The aluminum can scrap
price was obtained from http://www.metalprices.com/. The ferrous metal price was obtained from 2012 USGS
Mineral Commodities Summary: Iron & Steel Scrap at:
http://minerals.usgs.gov/minerals/pubs/commodity/iron & steel scrap/.
11-14 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LANDFILLS
Waste-to-Energy (WTE)
WTE is a combustion process; thus, its main emissions include CCh, CO, NOx, and non-Q-k volatile
organic compounds (NMVOCs). Municipal waste is incinerated to reduce its volume to save landfill costs
and recover energy from its combustion either for heating and/or electricity generation. The two most
widely used and technically proven incineration technologies are mass-burn incineration and modular
incineration. Fluidized-bed incineration has been employed to a lesser extent, although its use has been
expanding and experience with this relatively new technology has increased. RDF production and
incineration have also been used, primarily in Europe, but the number of successful cases is limited. This
analysis considers WTE using mass-burn incineration and electricity recovery, which is the most common
WTE design. Representative Q-k emissions of 0.2 to 60 kg/Gg of waste (wet weight) and N2O emissions
of 41 to 56 g/ton of waste (wet weight) were obtained from IPCC (2006). WTE facility costs are divided
into three components: capital cost, O&M cost, and revenue from electricity generation.
• Capital Costs. The plant's capital cost of $165 million includes the facility design engineering and
construction. Capital equipment includes the cost of land, incinerators, ash handling system,
turbine, and air pollution control and monitoring devices. Costs assume $829/tonne of design
capacity. This cost was derived from Eunomia (2008), which describes this cost for a 200,000
tonne/yr facility in the UK.
• O&M Cost. The annual O&M cost of the WTE facility is $8 million, approximately 4% of installed
capital costs. Annual costs include labor, overhead, taxes, administration, insurance, indirect
costs, auxiliary fuel cost, electricity cost, and maintenance cost. It does not include the cost for
disposing of the combustion residue and spray dryer residue. Cost is based on annual O&M cost
of $41/tonne/yr. Annual avoided landfilling is also included as a cost savings. The cost of disposal
of the fly and bottom ash from the incineration process assumes an estimated 15% of the
incoming organic waste will be converted to ash (MSW DST Documentation). No reuse of the
bottom ash (e.g., in construction projects) is assumed and the bottom and fly ash will be mixed
and sent to a landfill. Both the avoided landfilling costs and residual waste landfilling costs
assume a tipping fee of $29/tonne.
• Annual Benefits: Annual revenue from electricity sales. Electricity that is generated by
recovering heat from combusting waste is sold to an end user. The recovery of the heat is not
perfectly efficient. This inefficiency is represented by the heat rate of the plant, reported as 18,000
(BTU/kWh) in the WTE process model documentation of the MSW DST (MSW DST
Documentation). The electricity produced per tonne of waste can then be estimated according to
the heat value of the waste incinerated (4,750 BTU/tonne of waste). The market price of electricity
is used to estimate the revenues.
Applicability: This option applies to entire waste stream.
Technical Efficiency: This analysis assumes reduction efficiency of 100%.
Technical Lifetime: 20 years
1.1.4 Marginal Abatement Costs Analysis
The MAC analysis assimilates the abatement measures' technology costs, expected benefits, and
emission reductions presented in above to compute the net cost/benefit of abatement for each project.
Similar to the approach used in other non-CO2 sectors of this report, we compute a break-even price for
each abatement project (abatement measure by facility type). Project break-even prices are then weighted
by emission abatement potential to construct MAC curves illustrate the technical, net GHG mitigation
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 111-15
-------
LANDFILLS
potential at specific break-even prices for 2010 to 2030. MAC curves are produced for 195 countries using
country specific parameters, such as wage rates and energy prices.
This section describes the general modeling approach applied in the landfill sector as well as the
approach used to define the international facility populations and the assessment of sectoral trends.
These factors serve as additional inputs to the MAC analysis that adjust the abatement project costs,
benefits, and the technical abatement potential in each country.
III.1.4.1 Methodological Approach
The overarching modeling framework applied in the landfill sector is captured in two basic steps. The
first is to calculate the break-even price for each mitigation measure for each facility type by country. The
second is to determine the country-level abatement potential.
The break-even price, as defined in the technical summary to this report, estimates the costs and
expected benefits of each technology based on the characteristics of the model facility and relative
international prices (equipment, labor, and energy).
Country abatement potential reflects the number of abatement measures available and technical
effectiveness of each option. Figure 1-3 illustrates the conceptual modeling for estimating the abatement
potential in the landfill sector.
The MAC model uses a three-step approach to allocating a fraction of the BAU emissions to each
facility and technology considered. The model starts by splitting the BAU emissions out to our three
landfill types (open dump, basic landfill, engineered landfill). Next the model uniformly distributes BAU
emissions by the number of abatement measures considered. Finally, the model estimates abatement
potential by multiplying the BAU emissions (indexed by facility type and technology) by each
technology's technical effectiveness. Summing over all abatement measures and facility type indicates
this product yields a country's abatement potential.
It is important to note that depending on the scenario considered in the model, diversion options may
or may not be included. As shown in Figure 1-3, if diversion options are considered, BAU emissions
(indexed by facility type) are uniformly distributed by the total number of technologies (N = 12). If
diversion options are omitted, BAU emissions are distributed by the number of landfill-based mitigation
technologies (N = 7).
Assessment of Sectoral Trends
Underlying the general modeling approach, the MAC analysis also incorporated additional
international considerations to capture shifts in the share of BAU emissions allocated to the three model
landfill types defined earlier in Section III.1.2 (i.e., open dump, basic landfill, and engineered landfill).
Table 1-4 presents the facility share of BAU emissions over time. In the United States and the EU, we
assumed advanced waste management practices were already in place. Reflecting this assumption, we
assumed zero emissions coming from open dumps in these countries and assumed all emissions come
from basic and engineered landfills. Given the existing level of infrastructure in place there is very little
change in the assumed distribution over the 20-year modeling horizon.
11-16 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LANDFILLS
Figure 1-3: Conceptual Model for Estimating Mitigation Potential in the MSW Landfill Sector
For emerging economies and developing countries the analysis assumes a greater share of emissions
is represented by open dumps in 2010. Over the next 20 years, this distribution is projected to shift away
from open dumps as countries begin to adopt advanced waste management practices with greater shares
of total waste going to basic sanitary and engineered landfills. These shares were developed using expert
judgment after reviewing existing literature on waste disposal trends and abatement opportunities
provided through various studies by the World Bank, USEPA's LMOP program, and the Global Methane
Imitative (GMI).
Define Model Facilities for the Analysis
Seeking to improve the specificity of the break-even prices calculated for each country, this analysis
developed an international population of model facilities. This step of the analysis consisted of defining
the characteristics of the model facilities specific to countries and regions. The characteristics of interest
included the
• average annual waste acceptance rates by facility type,
• average waste depth by facility,
• decay constant (k) based on climate and moisture content in waste landfilled, and
• potential Q-k generation capacity (Lo) of the typical waste managed in a given model facility.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-17
-------
LANDFILLS
Table 1-4: Model Facilities Share of BAD Emissions: 2010-2030
Country/Region
China
Brazil
Mexico
Russia
Ukraine
Australia
Canada
Japan
Turkey
United States
India
South Korea
EU-27
Africa
Central & South
America
Middle East
Eurasia
Asia
Dump
Sites
20%
10%
10%
20%
20%
10%
10%
10%
20%
0%
20%
10%
0%
40%
10%
20%
20%
20%
2010
Basic
LF
60%
60%
60%
40%
40%
30%
30%
30%
40%
20%
60%
30%
20%
40%
60%
60%
60%
60%
Engineered Dump
LF Sites
20%
30%
30%
40%
40%
60%
60%
60%
40%
80%
20%
60%
80%
20%
30%
20%
20%
20%
10%
10%
10%
20%
20%
10%
10%
0%
20%
0%
10%
0%
0%
30%
10%
10%
10%
10%
2020
Basic
LF
60%
50%
50%
40%
40%
30%
30%
30%
40%
20%
60%
30%
20%
40%
50%
60%
60%
60%
Engineered Dump
LF Sites
30%
40%
40%
40%
40%
60%
60%
70%
40%
80%
30%
70%
80%
30%
40%
30%
30%
30%
10%
0%
0%
10%
10%
0%
0%
0%
10%
0%
10%
0%
0%
20%
0%
10%
10%
10%
2030
Basic
LF
50%
50%
50%
40%
40%
30%
30%
20%
40%
10%
50%
20%
10%
40%
70%
60%
60%
60%
Engineered
LF
40%
50%
50%
50%
50%
70%
70%
80%
50%
90%
40%
80%
90%
40%
30%
30%
30%
30%
Source: Based on expert judgment in consultation with World Bank (2010) and USEPA (2009,2011).
Various data sources were consulted to define the characteristics of the model facilities in the
different countries and regions, and a proxy country approach was used when data were not found for a
given country. Under this approach, countries for which no data were available were paired with a
representative proxy country based on similarities in socioeconomic and technology development trends
that are closely correlated with a country's waste composition. Furthermore, waste composition is the
only parameter that affects both Lo (Q-k generation rate) and k constant (decay rate), two key factors used
to estimate gas generation from the model facilities.
To ensure project costs and benefits were comparable, we assumed annual waste acceptance rates
(WAR) were fixed at 100,000 tonnes/yr, and the average depth of waste was assumed to be between 25
and 50 feet. Open dumps have shallower waste depths sprawling over large areas. In contrast, basic and
engineered landfills concentrate the disposed waste over a smaller area and at increased depths of
between 40 and 50 feet. Facility methane recovery (also referred to as capture efficiency), also varies by
landfill type and range from 10% for open dumps to 85% for engineered landfills. Table 1-5 summarizes
the standardized model facility assumptions.
11-18
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LANDFILLS
Table 1-5: Model Facility Assumptions for International LFG Mitigation Options
Facility Type
Engineered landfill
No. Years
Open
15
Annual WAR Project Design Waste Depth Facility CH4
(tonnes/yr) Acreage (ft) Recovery
100,000
40
50
85%
Basic landfill
15
100,000
50
40
75%
Open dump
15
100,000
80
25
10%
To improve the heterogeneity in the break-even options across countries, we developed a dataset of
country-specific data of Lo (methane generation potential) and k constant (decay rate) values, the two key
parameters in the first order decay model, which is used to estimate landfill gas generation. Both
parameters were calculated based on the composition of the waste being landfilled, which is determined
by the country-specific sotioeconomic conditions, consumption patterns, and waste management
practices. Therefore, the methane generation results and, consequently, the amount of methane
potentially mitigated by each landfill gas control measure are driven by the waste composition, which is
related to consumption patterns and sotioeconomic conditions. We grouped the countries according to
the following logic:
First, we identified the decay constant (k) and Q-k generation potential of waste (Lo) for 16 countries
that included at least 1 country within the each major region (Africa, Asia, Caribbean/Central & South
America, Eurasia, Europe, Middle East, and North America). This information was obtained from a
number of sources, including international studies conducted by the World Bank, USEPA's voluntary
program, the MSW Decision Support Tool (DST), and other peer-reviewed literature.
Second, we then used expert judgment, taking into consideration trends of sotioeconomic and
technological development to associate countries with other countries for which we have methane
generation data (e.g., we have methane generation data for Jordan and considered that Algeria, Egypt,
and South Africa have similar sotioeconomic and technological conditions). Alternatively, we have
methane generation data for Guinea, but we think that the socioeconomic and technological conditions in
Egypt, Algeria, and South Africa are closer to those in Jordan than to those in Guinea.
Table 1-6 presents the data used to characterize the model facilities for specific countries identified for
this analysis.
The international assessment of other OECD countries assumes waste management practices and
landfill designs similar to those in the United States. For this reason, we leverage the existing United
States-based landfill population, scaling the landfill size and emissions to meet projected baselines. For all
non-OECD countries for which we had no data, we developed three model facilities to represent the
allocation of waste to each type of waste management facility (i.e., engineered landfill, sanitary landfill,
and open dump). Each facility type was assumed to have similar characteristics in terms of capacity,
average depth of waste in place, and annual waste acceptance rates.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-19
-------
LANDFILLS
Table 1-6: Cm Generation
Country
Guinea
China
India
Japan
Nepal
Pakistan
Philippines
Argentina
Belize
Colombia
Nicaragua
Panama
Bosnia and Herzegovina
Ukraine
Jordan
United States
Factors by Country
Region1
Africa
Asia
Asia
Asia
Asia
Asia
Asia
CCSA
CCSA
CCSA
CCSA
CCSA
Eurasia
Eurasia
Middle East
North America
k Constant
(1/yr)
0.18
0.11
0.11
0.11
0.04
0.11
0.18
0.11
0.12
0.11
0.11
0.11
0.06
0.06
0.02
0.04
Lo
(ft3/shortton)
4,690
1,532
3,988
4,620
6,890
3,193
1,922
4,122
2,499
2,948
2,627
3,236
4,295
4,886
5,984
3,055
Data Source
WB
LMOP
Zhuetal. (2007)
WB
WB
WB
MSW DST
WB
MSW DST
LMOP
MSW DST
MSW DST
WB
LMOP
WB
LMOP
1CCSA = Central & South America
Sources: WB—World Bank Studies by Country; LMOP—USEPA's LMOP country-specific landfill gas models; MSW DST—decision support
model; and Zhu et al. (2007) "Improving municipal solid waste management in India."
Estimate Abatement Project Costs and Benefits
This analysis leveraged the USEPA LFG to energy project costs model to estimate abatement project
costs and benefits for the landfill-based mitigation technologies (with the exception of enhanced
oxidation). Key model facility characteristics discussed above were used as inputs to estimate the project
costs across countries. For waste diversion alternatives, we assumed that waste was diverted from
landfills and sent to alternative facilities for separation and reuse. Any residual waste from these facilities
is then sent to a landfill for final disposal. Model facilities reflect the recycling or reuse facility's annual
waste processing capacity as described in Section III.1.3.2.
Table 1-7 and Table 1-8 provide example break-even prices for model landfills and diversion facilities
using U.S. parameters and costs.
11-20
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LANDFILLS
Table 1-7: Example Break-Even Prices for MSW Landfill Technology Options
Annualized
Reduced Capital Annual Annual
Emissions Costs Cost Revenue
(tC02e) ($/tC02e) ($/tC02e) ($/tC02e)
Option by Landfill Type
Annual Tax
Benefit of Break-
Depreciation Even Price
($/tC02e) ($/tC02e)
Open Dump
Direct use
Combined heat and power
Engine
Microturbine
Turbine
Flare
7,475
7,475
7,475
7,475
7,475
7,475
$50
$86
$55
$54
$57
$38
$28
$31
$30
$31
$29
$27
$11
$10
$10
$7
$8
$0
$10
$17
$11
$11
$12
$8
$57
$89
$64
$67
$66
$58
Basic Landfill
Direct use
Combined heat and power
Engine
Microturbine
Turbine
Flare
56,061
56,061
56,061
56,061
56,061
56,061
$6
$17
$12
$11
$13
$4
$4
$6
$6
$4
$5
$3
$11
$10
$10
$7
$8
$0
$1
$4
$2
$2
$3
$1
-$2
$10
$6
$6
$7
$6
Engineered Landfill
Direct use
Combined heat and power
Engine
Microturbine
Turbine
Flare
63,536
63,536
63,536
63,536
63,536
63,536
$5
$16
$11
$10
$12
$3
$4
$6
$5
$3
$4
$2
$11
$10
$10
$7
$8
$0
$1
$3
$2
$2
$3
$1
-$4
$8
$4
$4
$6
$5
Note: Based on USA ChU generation parameters: Lo = 3,204 and k = 0.04. Assuming model landfill standardized size assumptions from
Table 1 -5. Break-even price is calculated using a discount rate of 10% and a tax rate of 40% and assumes energy prices of $3.2/Mcf and
$0.07/kWh for gas and electricity.
Table 1-8: Break-Even Prices of Waste Diversion Options
Waste Diversion Options
Annualized Annual Tax
Reduced Capital Annual Annual Benefit of Break-
Emissions Costs Cost Revenue Depreciation Even Price
(tC02e) ($/tC02e) ($/tC02e) ($/tC02e) ($/tC02e) ($/tC02e)
Composting
Anaerobic digestion
Mechanical biological treatment
Paper recycling
Waste to energy
Enhanced oxidation systems
5,222
4,658
18,605
6,164
55,816
10,483
$119
$1,626
$414
$1,613
$2,247
$143
$121
$360
$68
$1,249
$142
$1
$185
$330
$263
$1,028
$284
$0
$24
$277
$70
$275
$383
$11
$31
$1,380
$148
$1,559
$1,722
$132
Note: Assuming model sizes as described in Section 111.1.3. Present values calculated using a discount rate of 10% and a tax rate of 40%.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-21
-------
LANDFILLS
1.1.4.2
MAC Analysis Results
The MAC curve results are presented in Table 1-9 and Figure 1-4 by major emitting country and rest
of regional country groups. The MAC curves illustrate the increase in abatement achievable at higher
carbon prices. In 2030, the MAC curves show that approximately 589 MtCO2e, or 61% of global baseline
CH4 emissions from landfills, can be abated by adopting mitigation and avoidance options presented in
Section III.1.3.
Approximately 112 MtCO2e, or 19% of global abatement potential has a break-even price of zero or
less. These mitigation options are sometimes referred to as "no regret" options because the benefit cost
analysis implies that they would have a positive return. However, as discussed previously, there may be
transaction costs not captured in this analysis that are currently limiting their adoption.
At break-even prices between $20/tCO2e to $50/tCO2e most countries MAC curves become non
responsive (vertical). This is because there are few options within this break-even range. Between
$50/tCO2e to $100/tCO2e an additional 20% of abatement potential becomes economically viable. And, at
break-even prices (> $100/t CO2e) the remaining set of emission reduction options are economically
viable, but at extremely higher prices. The point at which the MAC becomes unresponsive to any price
change can also be considered the full technical potential associated with the suite of abatement measures
considered. Thus, it can be inferred that additional reductions beyond approximately 60% of the
projected baseline in 2030 would be unlikely without additional policy incentives or technology
improvements.
Table 1-9: Abatement Potential by Region at Selected Break-Even Prices in 2030 (MtC02e)
Country/Region
-10
-5
Break-Even Price ($/tC02e)
5 10 15 20 30
BTjljH
KTJTJ9
Top 5 Emitting Countries
China
Malaysia
Mexico
Russia
United States
3.7
1.3
2.4
7.8
1.5
2.4
20.1
1.3
1.7
2.4
20.1
8.2
7.1
9.0
20.1
11.4
10.3
11.7
3.5
26.1
15.1
11.1
17.3
3.5
26.1
15.1
12.1
17.3
3.5
26.1
15.1
13.6
17.3
3.5
26.1
15.3
14.3
23.8
8.7
34.8
20.7
14.5
37.0
19.6
53.0
32.0
14.6
Rest of Region
Africa
Central and South America
Middle East
Europe
Eurasia
Asia
North America
World Total
5.6
1.6
3.5
22.8
2.9
1.5
43.0
5.6
4.4
4.0
36.0
6.7
1.6
70.0
5.6
8.7
7.5
49.0
11.8
4.2
112.4
9.7
8.8
20.1
70.9
15.8
8.7
171.9
26.1
8.8
20.8
82.6
17.5
10.8
217.4
31.3
9.1
23.2
86.7
1.6
19.1
13.6
252.2
42.7
16.3
25.0
91.8
2.3
29.7
13.8
295.7
42.7
16.4
26.3
92.7
2.3
30.1
14.0
300.1
43.2
16.4
27.0
93.1
2.3
30.1
14.1
302.8
60.3
27.6
36.2
98.8
5.8
48.7
15.0
394.9
95.4
50.9
55.7
110.4
12.9
86.3
21.5
589.4
II-22
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LANDFILLS
Figure 1-4: Marginal Abatement Cost Curve for Top 5 Emitters in 2030
Non-C02 Reduction (MtC02e
•China
Mexico
•Malaysia
•Russia
United States
30.0
1.1.4.3
Uncertainties and Limitations
Uncertainty and limitations persist despite attempts to incorporate all publicly available information.
Additional country-specific detailed information would improve the accuracy of the MAC projections.
• Energy prices are negotiated on a case-by-case basis and may not be as high as the wholesale
price used in the analysis.
• National/regional or local policies for permitting projects may differ; also incentives such as tax
credits, grants, loans and other financial incentives for LFG projects differ across states.
• Additional data characterizing specific landfills are necessary for a more accurate financial
analysis of each technology or specific project at a specific site. Costs can vary depending on the
depth area, waste composition, and annual waste in place.
Efforts to reduce landfilling (e.g., recycling, composting) can also reduce CH4 emissions and will have
an effect on the most appropriate type of project and its cost-effectiveness at a given landfill. In general,
additional country specific information would be useful in determining which abatement measures
would be most likely to be adopted over time.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
II-23
-------
LANDFILLS
References
CalRecycle. 2009. Life Cycle Assessment and Economic Analysis of Organic Waste Management and
Greenhouse Gas Reduction Options. Obtained March 8, 201, at:
http://www.calrecy cle.ca.gov/climate/Events/LifeCycle/2009/default.htm.
Chartwell Information Publisher Services. 2004. Chartwell Municipal Solid Waste Facility Directory. Solid
Waste Market Intelligence, Analysis, and Strategy Services.
Eunomia. 2008. Development of Marginal Abatement Cost Curves for the Waste Sector. United Kingdom:
Committee on Climate Change. Defra and Environment Agency. Available at:
http://www.theccc.org.uk/pdfs/Eunomia%20Waste%20MACCs%20Report%20Final.pdf
Huber-Humer, M., J. Gerber, and H. Hilber. 2008. "Biotic Systems to Mitigate Landfill Methane
Emissions." Waste Management & Research 26: 33-46.
Intergovernmental Panel on Climate Change (IPCC). 2006. 2006 IPCC Guidelines for National Greenhouse
Gas Inventories: Reference Manual (Volume 5-Waste). Available at: http://www.ipcc-
nggip.iges.or.jp/public/2006gl/index.html. As obtained on March 4, 2011.
RTI International. Municipal Solid Waste Decision Support Tool. MSW DST Documentation. Research
Triangle Park, NC: RTI International. Obtained at: https://mswdst.rti.org/.
U.S. Environmental Protection Agency (USEPA). 2010. Landfill Gas Energy Cost Model. Washington, DC:
USEPA, Landfill Methane Outreach Program (LMOP). Obtained at:
http://www.epa.gov/lmop/publications-tools/index.html.
U.S. Environmental Protection Agency (USEPA). 2006. Global Mitigation of Non-COi Greenhouse Gases.
Washington, DC: USEPA. EPA 430-R-06-005. Obtained on February 18, 2011, at:
http://www.epa.gov/climatechange/economics/international.html.
U.S. Environmental Protection Agency (USEPA).2008. Background Information Document for Updating AP42
Section 2.4 for Estimating Emissions from Municipal Solid Waste Landfills. EPA/600/R-08-116.
Washington, DC: USEPA. Available at: http://www.epa.gov/ttn/chief/ap42/ch02/draft/db02s04.pdf.
U.S. Environmental Protection Agency (USEPA). 2009. Municipal Solid Waste in the United States: 2009
Facts and Figures. Washington, DC: USEPA, Office of Solid Waste and Emergency Response. Obtained
on March 3, 2011, at: http://www.epa.gov/osw/nonhaz/municipal/msw99.htm.
U.S. Environmental Protection Agency (USEPA). September 2010. Project Development Handbook. Chapter
3. Project Technology Options. Landfill Methane Outreach Program. Available at:
http://www.epa.gov/lmop/publications-tools/tfone
U.S. Environmental Protection Agency (USEPA). 2011. Draft Inventory of U.S. Greenhouse Gas Emissions and
Sinks 1990-2011. Washington, DC: USEPA, Office of Solid Waste and Emergency Response. Obtained
on March 7, 2011, at: http://www.epa.gov/climatechange/emissions/usinventoryreport.html.
U.S. Environmental Protection Agency (USEPA). 2012. Global Anthropogenic Non-COi Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA.
U.S. Geologic Survey (USGS). 2012. Mineral Commodities Summary: Iron & Steel Scrap. Available at:
http://minerals.usgs.gov/minerals/pubs/commodity/iron & steel scrap/.
Unnikrishnan, S., and A. Singh. 2010. "Energy Recovery in Solid Waste Management through CDM in
India and Other Countries." Resources, Conservation and Recycling 54(10): 630-640.
doi:16/j.resconrec.2009.11.003
Van Haaren, R., N. Themelis, and N. Goldstein. 2010. "17th Annual BioCycle Nationwide Survey: The
State of Garbage in America." BioCycle October.
Weitz, K. 2011. "Updated Research on Methane Oxidation in Landfills." Technical memorandum
providing an update to previous (March 2006) research on methane oxidation rates in landfills.
Prepared for Rachel Schmeltz, EPA, Climate Change Division. January.
World Bank Group. 2010. 2009—World Development Indicators: Table 2.1 Population Dynamics. Obtained on
March 8, 2011, at: http://data.worldbank.org/data-catalog/world-development-indicators.
11-24 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LANDFILLS
Zhu, D., P.U. Asnani, C. Zurbriigg, S. Anapolsky and S. Marti. 2007. "Improving Municipal Solid Waste
Management in India." Washington DC: The World Bank.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 111-25
-------
WASTEWATER
1.2. Waste water
1.2.1
Sector Summary
Iomestic and industrial wastewater treatment activities can result in deliberate venting and
fugitive emissions of methane (Q-k). In addition, domestic wastewater is also a source of
nitrous oxide (N2O) emissions. Q-k is produced when the organic material present in the
wastewater flows decomposes under anaerobic conditions. Although most developed
countries rely on centralized aerobic wastewater treatment systems, which limit the level of Q-k
generated, less developed countries often rely on a broader suite of wastewater treatment technologies
with a significant proportion of wastewater flows handled by anaerobic systems such as septic tanks,
latrines, open sewers, and lagoons (USEPA, 2012a).
Worldwide Q-k from wastewater accounted for more than 500 MtCO2e in 2010. Wastewater is the
fifth largest source of anthropogenic Q-k emissions, contributing approximately 4% of total global Q-k
emissions in 2010. China, Nigeria, Mexico, India, and the United States, combined account for 60% of the
world's Q-k emissions from wastewater (see Figure 2-1). Global Q-k emissions from wastewater are
expected to grow by approximately 19% between 2010 and 2030.
United States
India
Mexico
Nigeria
China
Rest of World
Source: U.S. Environmental Protection Agency (USEPA). 2012a.
N2O emissions from human sewage are a second significant source of GHG emissions within the
wastewater sector, contributing an additional 2% of global N2O emissions in 2010. Figure 2-2 illustrates
the growth in N2O emissions out to 2030 for the wastewater sector. China, the United States, Brazil,
Russia, and India are projected to be the five largest emitters of N2O in 2030, representing 36% of total
N2O emissions in the wastewater sector. Growth in N2O emissions between 2010 and 2030 is expected to
be 16%, slightly lower than the projected growth in Q-k emissions over the same time period.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-27
-------
WASTEWATER
India
Russia
Brazil
United States
China
Rest of World
Source: U.S. Environmental Protection Agency (USEPA). 201 2a.
Global abatement potential1 of CH4 in wastewater treatment is 138 and 218 MtCO2e in 2020 and 2030,
respectively.2 These corresponding sectoral MAC curves are shown in Figure 2-3. As the marginal
abatement cost (MAC) curves show, high-cost mitigation measures in the wastewater treatment sector
constrain the level of abatement achievable at lower carbon prices (less than $30 tCC^e-1) to less than 5%
of CH4 emissions in 2030. Maximum abatement potential (218 MtCO2e) is 36% of total Q-k emissions in
the wastewater sector in 2030.
The following section provides a brief explanation of sector activity, how Q-k and N2O emissions are
generated, and projected emissions from wastewater from 2010 to 2030. Subsequent sections characterize
the abatement measures available to the wastewater sector and present the costs of their implementation
and operation. The chapter concludes with a discussion of the MAC analysis approach unique to this
sector and presents the regional MAC results.
1 This analysis only assesses abatement measures to reduce CHi emissions. Mitigation potentials reported in this
chapter do not consider potential reductions in N2O emissions, because of limited information on abatement measure
costs.
2 Vertical axis is scaled to limited range of prices between $0 and $800/tCO2e. This scale was chosen because it shows
sufficient detail in the MAC curves at lower break-even prices. Only 45% of the total abatement is visible in the figure
simply due to the price limits chosen for the vertical axis when reporting the data.
11-28
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
WASTEWATER
20
40 60 80
Non-CO2 Reduction (MtCO2e)
100
120
1.2.2
GHG Emissions from Wastewater
This section discusses how Q-k and N2O emissions are produced in wastewater treatment and
disposal activities and the current projections of baseline emissions between 2010 and 2030.
1.2.2.1
CH4 Emissions from Domestic and Industrial Wastewater
CH4 is emitted during the handling and treatment of domestic and industrial wastewater.
Wastewater Q-k emissions are produced through the anaerobic decomposition of organic material
present in the wastewater. Three key factors that determine the Q-k generation potential are the quantity
of degradable organic material present in the wastewater, the temperature, and the type of treatment
system used (Intergovernmental Panel on Climate Change [IPCC], 2006). The organic content of
wastewater is typically expressed in terms of either biochemical oxygen demand (BOD) or chemical
oxygen demand (COD) (IPCC, 2006; USEPA, 2012a). Q-k generation is positively related to temperature
so that higher temperatures result in a great amount of Q-k produced. The third key factor that
determines Q-k generation is the type of treatment system used and more specifically the amount of
decomposition occurring under anaerobic conditions which is positively related the quantity of Q-k
generated.
Types of centralized systems that can result in Q-k emissions include 1) aerobic systems that are
either improperly operated or designed to have periods of anaerobic activity and 2) anaerobic lagoons
(USEPA, 2012b). Most developed countries currently use centralized aerobic wastewater treatment
facilities with dosed anaerobic sludge digester systems to process municipal and industrial wastewater,
minimizing Q-k emissions.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-29
-------
WASTEWATER
The IPCC guidelines for national greenhouse gas reporting identifies five major industrial
wastewater sources for Q-k emissions, which include pulp and paper manufacturing, meat and poultry
processing (slaughterhouses), alcohol/beer and starch production, organic chemicals production, and
other drink and food processing (e.g., dairy products, vegetable oil, fruits and vegetables, canneries, juice
making) (IPCC, 2006). The significance of Q-k emissions from the various industrial sources will depend
on the concentration of degradable organics present in the wastewater flow, volume of wastewater
generated, the quantity of wastewater treated in anaerobic treatment systems (e.g., anaerobic lagoons).
III. 2.2.2 N2O Emissions from Domestic Wastewater — Human Sewage
N2O is produced during both the nitrification and denitrification of urea, ammonia, and proteins.
These waste materials are converted to nitrate (NOs) via nitrification, an aerobic process converting
ammonia-nitrogen to nitrate. Denitrification occurs under anoxic conditions (without free oxygen) and
involves the biological conversion of nitrate into dinitrogen gas (N2). N2O can be an intermediate product
of both processes but is more often associated with denitrification (Sheehle and Doom, 2002).
III. 2.2.3 Emissions Estimates and Related Assumptions
This section discusses the historical and projected baseline emissions for the wastewater sector.3
Historical emissions are characterized as those emissions released between 1990 and 2010. Projected
emissions estimates cover the 20-year period starting in 2010 and ending in 2030.
Historical Emissions Estimates
Between 1990 and 2005, Q-k and N2O emissions from wastewater increased by 20% from a combined
total of 421 MtCO2e in 1990 to 505 MtCO2e in 2005. The primary driver of both CH4 and N2O emissions
associated with wastewater is population growth. Country-level Q-k emissions are particularly sensitive
to population growth in countries that rely heavily on anaerobic treatment systems such as septic tanks,
latrines, open sewers, and lagoons for wastewater treatment (USEPA, 2012a).
The share each countries total emissions that is attributed to domestic versus industrial wastewater
sources is determined by the level of industrial activity and types of domestic wastewater treatment
systems employed. In developing countries, domestic wastewater sources account for the majority if not
all of Q-k emissions from wastewater. In countries with industrial wastewater sources, the contribution
of industrial wastewater emissions will depend on the level of production and the commodity produced
(e.g. paper, sugar, alcoholic beverages, and processed meat/poultry/fish). Based on the UNFCCC's
national reporting inventory database of GHG emissions, only a small number of developed countries
have historically reported Q-k emission from Industrial sources. For these 24 countries reporting
industrial and domestic Q-k emissions the share of emissions reported for industrial wastewater ranged
from less than 2% to nearly 70% of total Q-k emissions from all wastewater sources. Section III.2.4
discusses these distributions of emissions to domestic and industrial sources in more detail.
Projected Emissions Estimates
Worldwide Q-k emissions are projected to increase by approximately 19% (97 MtCC^e) between 2010
and 2030. N2O emissions are projected to increase by a similar proportion, up 16% (14 MtCO2e) over the
same time period. Tables 2-1 and 2-2 present the Q-k and N2O emissions projections for the wastewater
sector.
3 For more detail on baseline development and estimation methodology, we refer the reader to the USEPA's Global
Emissions Projection Report available at: http://vyyyyy.epa.gov\ climatechange\ economics\ international.html.
111-30 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
WASTEWATER
Table 2-1: Projected CH4 Baseline Emissions from Wastewater: 2010-2030 (MtC02e)
Country ^^^|
2010
2015
2020
2025
2030
CAGR*
(2010-2030)
Top 5 Emitting Countries
China
Nigeria
Mexico
India
United States
132
56
48
42
25
135
62
51
45
26
137
67
54
47
27
138
73
56
50
29
138
78
58
52
30
0.2%
1.7%
0.9%
1.1%
0.9%
Rest of Regions
Africa
Central & South America
Middle East
Europe
Eurasia
Asia
North America
World Total
27
47
22
19
26
68
0
512
29
50
23
19
25
72
0
539
32
53
25
20
25
76
0
565
35
56
26
20
24
81
0
588
38
59
28
20
23
84
0
609
1.9%
1.1%
1.2%
0.2%
-0.5%
1.1%
0.7%
0.9%
a CAGR = Compound Annual Growth Rate
Source: U.S. Environmental Protection Agency (USEPA). 2012a.
Table 2-2: Projected N20 Baseline Emissions from Human: 2010-2030 (MtC02e)
Country
2010
2015
2020
2025
2030
CAGR
(2010-2030)
Top 5 Emitting Countries
China
United States
Brazil
Russia
India
17
5
5
4
3
17
5
5
4
3
17
6
5
4
3
17
6
5
4
3
17
6
6
4
4
0.2%
0.9%
0.9%
1.1%
-0.6%
Rest of Region
Africa
Asia
Central & South America
Eurasia
Europe
Middle East
North America
World Total
11
5
4
14
3
12
3
86
13
5
5
14
3
13
3
90
14
6
5
14
3
14
3
93
15
6
5
14
3
14
3
97
17
6
6
14
3
15
3
100
1.8%
1.0%
1.0%
0.1%
0.0%
1.3%
0.9%
0.7%
Source: U.S. Environmental Protection Agency (USEPA). 2012a.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-31
-------
WASTEWATER
As shown in Table 2-1, Africa and the Middle East are two regions projected to experience significant
growth in Q-k emissions over the next 20 years, increasing by 50% and 33%, respectively. Q-k emissions
growth in Asia and the Central and South American regions is also expected to be significant, growing by
25% over the same time period.
N2O emissions are expected to grow by similar proportions across all regions with the exception of
Eurasia, where emissions are expected to remain relatively unchanged over the next 20 years. The
primary driver of this trend is Russia's 11% drop in N2O emissions between 2010 and 2030. Despite this
decline, Russia still ranks as one of the top five emitters in 2030.
III.2.3 Abatement Measures and Engineering Cost Analysis
This section characterizes the abatement measures considered for the wastewater sector. This analysis
focused on domestic wastewater treatment and implementation of abatement measures aimed at
reducing Q-k emissions, which can be mitigated through investment in infrastructure and/or equipment.
Conversely, there are no proven and reliable technologies for mitigation of N2O emissions. Mitigation
steps to limit N2O emissions from wastewater treatment are operational, and include careful control of
dissolved oxygen levels during treatment, controlling the biological waste load-to-nitrogen ratio, and
limiting operating system upsets. These measures require technical expertise and experience rather than
an engineered solution, thus they fall outside the scope of an engineered cost analysis.
It is important to couch the discussion of greenhouse abatement measures for municipal wastewater
in the appropriate context. In practice, changes to wastewater management strategies in developing
countries are unlikely to be driven by the mitigation of greenhouse gases. Factors such as economic
resources, population density, government, and technical capabilities are all important in determining
both the current state and the potential for improvement to a country's wastewater sanitation services.
Figure 2-4 is an illustration of the sanitation ladder, which relates the level of available wastewater
sanitation to the population and cost for treatment. The transition from a latrine to a sewer/wastewater
treatment plant (WWTP)/anaerobic digester can increase the operation and maintenance wastewater
treatment cost per person by a factor 20 (Guchte and Vandeweerd, 2004). This does not account for the
capital cost that would be required in such large scale projects.
The reader should bear in mind throughout the analysis that the wastewater sanitation technology is
likely to be fixed by these external factors, and improvements in technology will be driven by the
population's desire/capacity for improved sanitation and hygiene, with any improvements to greenhouse
gas emissions a secondary result of the change. Thus, although abatement measures are presented in this
chapter, they should not be considered to be a viable control measure that could be implemented for the
sole purpose of reducing a country's GHG emissions, but rather a byproduct of a country's position on
the sanitation ladder.
This analysis considers abatement measures that may be applied to one of five existing wastewater
treatment systems currently being utilized in a given country. Scenarios 1 and 2 correspond to the upper
half of the sanitation ladder, while scenarios 3 through 5 correspond to the lower half the sanitation
ladder. The five baseline scenarios for the existing status quo are presented in Figure 2-5. In actuality,
there are many more than five baseline technology scenarios that may be utilized throughout the world,
within a country, or even within a municipality. For example, a population may utilize aerobic or
anaerobic ditches for centralized treatment of wastewater, which could be viewed as an intermediate
option between scenarios 1 and 2 in Figure 2-5. These baseline scenarios are not meant to be an
exhaustive list of the actual existing treatment technologies employed worldwide, but rather an attempt
111-32 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
WASTEWATER
Figure 2-4: Sanitation Ladder for Impr
More developed
countries "
Less developed
countries ~
oven
-
- —
-
^-
lents to Wastewater Treatment
Sewer t WWTP +
anaerobic digester
Sewer + WWfP
Open sewer
Septic tank
Lairine
No treatment
1
^
s.
"5
ii;
Q
'4-1
fn
u
c
W
u
Figure 2-5: Five Existing Scenarios Evaluated for Given Wastewater Discharge Pathways Based on
Technology Level, Treatment Alternative, and Collection Method
No centralized
collection
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-33
-------
WASTEWATER
to broadly categorize and quantify technologies that represent the major classes of treatment technologies
employed throughout the world.
Discharge pathways 1 and 2 (in Figure 2-5) assume the existence of a collection system for all
wastewater generated and are grouped according to the final disposal/treatment approach. Pathways 3, 4,
and 5 are the scenarios for which no existing centralized treatment exists and the waste is treated on site
with latrines or septic tanks. For each of the five pathways and corresponding treatment systems, a
mitigation approach is evaluated for CH4 reduction. The analysis considers three abatement measures
that include both mitigation technologies as well as complete shifts in wastewater management, that is, a
jump up the sanitation ladder.
It is important to note the distinction between the two types of abatement measures. Mitigation
technologies represent add-on technologies that can be applied to existing wastewater treatment systems
(such as an anaerobic digester with cogeneration) intended to capture and destroy the CH4 generated at
the facility. The second type of abatement measure represents a shift away from an existing anaerobic
wastewater treatment approach to an aerobic system which in turn will reduce the volume of CH4
generated during the treatment process. This shift in wastewater treatment approaches will require the
construction of a new facility that fundamentally changes the existing wastewater management approach.
This approach usually requires construction of new infrastructure and, therefore, will require significant
capital investment. As demonstrated in the cost analysis, the construction and operation and maintenance
cost per person is dependent on the population density of the region. For a collection system, more rural
areas require the more material (per person) to be used to build a system to collect and transport the
waste.
1.2.3.1
Overview of Abatement Measures
This section discusses the abatement measures considered for this analysis. Each technology is briefly
characterized and followed by a discussion of abatement measures' implementation costs, potential
benefits, and system design assumptions used in the MAC analysis. Table 2-3 compares the three
abatement alternatives for an example population of 400,000 people, population density of 3,000/km2, and
wastewater generation rate of 340 L/person/day.
Table 2-3: Abatement Measures for the Wastewater Sector
Abatement Option
Anaerobic biomass digester with ChU
collection and cogen.
Aerobic wastewater treatment plant
(WWTP)
Centralized wastewater collection
(+ aerobic WWTP)
Total Installed Annual O&M Time
Capital Cost
(201 01 06 USD)
21.1
Cost Horizon
(201 01 06 USD) (Years)
5.0
20
Technical
Efficiency
60-80%
97.2 4.7 20 60-80%
55.9(153.1)
1.6(6.3)
50
60-80%
For this analysis, abatement measures are assigned based by on the existing wastewater treatment
system pathway in place. For example, a population considering the addition of an anaerobic biomass
digester will already have an existing collection system and aerobic WWTP in place. There is no
technology selection in the current analysis because we have identified one abatement measure for each
type of treatment system. The following subsections characterize each of the three abatement measures
and the assumptions regarding applicability and costs. In reality, feasible mitigation measures will vary
III-34
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
WASTEWATER
due to the wide range of wastewater disposal options currently employed in each country and the
external factors that govern a country's ability to transition from one technology to another. In addition,
as discuss above regarding the baseline scenarios, there are dozens of wastewater technology options
available to a population; this discussion highlights three major categories that represent shifts in water
management or add-on technology.
1.2.3.2
CH4 Mitigation Technology for Existing Decentralized Treatment
This section characterizes the reduction in Q-k emissions by adding a collection system and
centralized treatment facility in developing countries where the current practice is decentralized
wastewater treatment. As shown in Figure 2-6, this approach necessitates two large-scale capital
investments: the construction of a sewerage system for centralized collection and the construction of an
anaerobic WWTP.
Figure 2-6: Mitigation Technology Approach for Developing Countries with Decentralized Treatment
Domestic
wastewater
No centralized
collection
Mitigation Technology
Treated
Septic tank Latrine
Other system
Centralized
Aerobic WWTP
Wastewater Collection System—New Construction
For areas of the developing world without centralized wastewater treatment, latrines and/or septic
tanks are typically used to dispose of domestic wastewater. In both of these cases, the organic matter in
the wastewater will undergo anaerobic degradation to produce CH4. The construction and
implementation of a collection system and subsequent treatment at a centralized facility would
significantly reduce CH4 formation because transporting wastewater through sewers promotes aerobic
conditions and reduces the fraction of organic content that undergoes anaerobic digestion.
The design and size of a wastewater collection system depend on the population served, the service
area size, and water use characteristics of the population. Wastewater collection systems link all
household and commercial discharges through underground piping, conveying the water to either a
centralized treatment facility or directly to an outfall point where it is released into the environment.
Pipelines can vary from 6 inches in diameter to concrete-lined tunnels up to 30 feet in diameter.
Collection systems are built with a gradient so gravity can facilitate the water flow; where there are large
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-35
-------
WASTEWATER
distances that must be covered, periodic pump stations (also called lift stations) are sometimes used to
pump the sewage to a higher elevation and again allow gravity to transport the sewage. Sewage pumps
are typically centrifugal pumps with open impellers, designed to have a wide opening to prevent the raw
sewage from dogging the pump. This scenario evaluates the impact of installing a sewer collection
system without a centralized treatment facility.
• Capital Cost: The cost estimation model Water and Wastewater Treatment Technologies
Appropriate for Reuse (WAWTTAR) (Finney and Gearheart, 2004) was used to determine the
capital cost of the sewer construction. The model is used by engineers, planners, decision makers,
and financiers to estimate the costs of making improvements to wastewater treatment systems
while minimizing impacts to water resources. The capital cost curve for wastewater collection
systems is based on the population density: Capital Cost ($MM/km2) = 360.54 * DP-0844, where DP
is population density in (persons/km2).
• Annual Operation and Maintenance (O&M) Cost: Annual O&M costs for collection systems
were scaled from the capital cost and assumed to be a factor of 0.028 * initial capital cost, which
for this case gives the following cost curve, based on population density: O&M Cost ($MM/km2) =
10.095 x DP-°844.
• Annual Benefits: No benefits are associated with this option.
• Applicability: This option applies to all scenarios having no existing centralized collection
system.
• Technical Efficiency: This analysis assumes an initial collection efficiency of 60%, which
increases by 10% each year, due to an assumed improvement in technical efficiency.
• Technical Lifetime: 50 years
Aerobic WWTP—New Construction
Contaminants in wastewater are removed using a variety of physical, chemical, and biological
methods. A WWTP typically comprises many unit operations from each of these broad categories.
Wastewater treatment technologies are also divided into stages of treatment, each of which comprises one
or more individual treatment processes. A brief summary of each of these classifications is as follows:
• Fretreatment: This stage involves the removal of wastewater constituents. These constituents can
include rags, sticks, floatables, grit, and grease that may cause maintenance or operational
problems with the treatment operations, processes, and ancillary systems. Screening methods are
employed here, suing bars, rods, grates, or wire meshes.
• Primary treatment: This stage focuses on the removal of a portion of the total suspended solids
(TSS) and organic matter from the wastewater. Primary treatment is a physical unit process in
which the sewage flows into large tanks, known as primary clarifiers or primary settling tanks. A
settling tank is constructed of concrete and designed so that the residence time of the wastewater
is such that the flow slows down enough so that readily settleable particles are collected at the
bottom of the tank.
• Secondary treatment: This stage focuses on the removal of biodegradable organic matter (in
solution or suspension) and TSS by aerobic or anaerobic biological treatment. Disinfection is also
typically included in the definition of conventional secondary treatment. Secondary treatment is
a biological process that cultivates and uses a consortium of microorganisms to degrade the
organic wastes and reduce nutrient levels in wastewater. Secondary treatment can either be
aerobic (with oxygen) or anaerobic (without oxygen). By far, the most common approach used in
WWTPs is the activated sludge process. This process is an aerobic suspended-growth system
111-36 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
WASTEWATER
containing a biomass that is maintained with oxygen and is capable of stabilizing organic matter
found in wastewater. During the activated sludge process, the effluent flows into a concrete tank
where air or oxygen is bubbled through the wastewater to encourage microbial degradation of
the organic material. The treated effluent flows to a secondary settling tank, where it is separated
from the biomass. Most of the biomass collected at the bottom of the settling tank is removed for
further dewatering and stabilization before final disposal. A small fraction of the biomass is
recycled back into the bioreactor to maintain the population. It is important to monitor proper
control of oxygen levels, pH, and the amount of sludge recycled back into the reactor to ensure
that proper treatment levels of the wastewater are maintained.
• Tertiary treatment: This stage involves the removal of residual suspended solids (after secondary
treatment), usually by granular medium filtration or microscreens. Disinfection is also typically a
part of tertiary treatment. Nutrient removal is often included in this stage.
The cost breakdown for this mitigation approach is as follows:
• Capital Cost: Capital costs were estimated using EPA cost curves detailing the construction costs
of publicly owned wastewater treatment facilities (USEPA 1980). The costs curves in this report
are based on actual winning bids for treatment plans, which include detailed equipment and
materials requirements, including labor, amortization, land, concrete, pumps, pipes, power,
haulage, chemicals, and design fees. All cost curves were updated to year 2010 dollars. The cost
curve is based on the flow rate of the WWTP: Capital Cost ($MM) = 0.0174 x Qo /3/ where Q is the
flow rate in m3/day.
• Annual Operation and Maintenance (O&M) Cost: Typical annual O&M costs of an aerobic
WWTP are due to electricity used to provide aeration and operation equipment, labor to operate
the plant, chemicals, and equipment replacement. EPA cost curves (updated to 2010 dollars)
provides the following cost curve for an aerobic WWTP, based on the flow rate: 0.0002 x Qossi/
• Annual Benefits: None.
• Applicability: This option applies to all conditions when new WWTPs are constructed.
• Technical Efficiency: This analysis assumes an initial collection efficiency of 60%, which
increases by 10% each year.
• Technical Lifetime: 20 years.
III. 2.3.3 CH4 Mitigation Technology for Existing Collection System without
Treatment
This section characterizes the reduction in Q-k emissions for the existing condition of a centralized
collection system without a treatment facility. Figure 2-7 illustrates the step change in technical capability,
which in this case necessitates the construction of a new anaerobic WWTP.
As noted above, contaminants in wastewater are removed via a variety of physical, chemical, and
biological methods. An anaerobic WWTP typically comprises many unit operations divided into stages of
treatment: pretreatment, primary treatment, secondary treatment, and tertiary treatment.
The cost breakdown for this mitigation approach is identical to that above and is as follows:
• Capital Cost: Capital costs were estimated using EPA cost curves detailing the construction costs
of publicly owned wastewater treatment facilities. The cost curve is based on the flow rate of the
WWTP: Capital Cost ($MM) = 0.0174 x Qo^ where Q is the flow rate in m3/day.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 111-37
-------
WASTEWATER
Figure 2-7: Mitigation Technology Approach for Developing Countries with Decentralized Treatment
Centralized
Aerobic WWTP
Annual Operation and Maintenance (O&M) Cost: Typical annual O&M costs of an aerobic
WWTP are due to electricity used to provide aeration and operation equipment, labor to operate
the plant, chemicals, and equipment replacement. Capdetworks v2.5 was used to estimate O&M
costs. Capdetworks is a planning level tool that enables the user to evaluate the costs associated
with individual treatment units or entire systems. The costs are based on detailed equipment and
materials database that utilizes published cost indices, including labor, amortization, and energy
requirements. Capdetworks provides the following cost curve for an aerobic WWTP, based on
the flow rate: O&M cost ($MM) = 0.0002 x Qo.ssi/.
Annual Benefits: None.
Applicability: This option applies to all conditions when new WWTPs are constructed.
Technical Efficiency: This analysis assumes an initial collection efficiency of 60%, which
increases by 10% each year.
Technical Lifetime: 20 years.
1.2.3.4
CH4 Mitigation Technology for Existing Centralized Aerobic WWTPs
This section characterizes the reduction in Q-k emissions from adding an activated sludge digester
for Q-k collection and energy generation. This option is only applicable to existing centralized aerobic
WWTPs primarily found in developed countries (Figure 2-8).
111-38
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
WASTEWATER
Figure 2-8: Mitigation Technology Approach for Countries with Existing Centralized WWTPs
Domestic
wastewater
Centralized
collection
T
Treated
Aerobic
WWTP
Domestic
wastewater
Treated
Aerobic WWTP
w/sludge digestion +
CH4 collection
Anaerobic Biomass Digester with CH4 Collection
The top of the technology ladder evaluated assumes an existing centralized WWTP is used to treat all
wastewater generated in the region. The significant quantity of biomass generated during the
decomposition of the sewage is a major operational component of WWTP operation. Typical approaches
to sludge handling include dewatering to reduce the overall volume and further water reduction in open-
air drying beds. The sludge is rich in organic matter and has the potential to produce high amounts of
CH4 during degradation. Anaerobic digestion is an additional sludge-handling step that can be employed
to further reduce the sludge volume; it is a process that involves the decomposition of this organic
material in an oxygen-free environment to produce and collect Q-k. Anaerobic digesters are large
covered tanks that are heated to optimize the methane-generating process. The tanks typically employ a
mixing mechanism to ensure uniform conditions throughout the tank and are designed with headspace
to collect the gas generated, which is typically a mix of 60 to 70% Q-k and the 30 to 40% CCh, along with
trace gases. The remaining solid material is nutrient rich and is a suitable fertilizer for land application.
The heat from the flared gas can be used to heat the digester, lowering the overall energy requirements of
the system. Alternatively, the gas can be used to produce electricity with a turbine.
• Capital Cost: Costs were derived from EPA process cost curves for new construction of an
anaerobic digester. The capital cost covers the construction of the tank with heater and cover and
includes concrete, all equipment, process piping and steel required for digester construction.
Costs were derived from CapdetWorks. The cost curve is based on the flow rate of the WWTP:
Capital Cost ($MM) = 0.0004 x Qo^ where Q is the flow rate in m3/day.
• Annual Operation and Maintenance (O&M) Cost: Typical annual O&M costs for collection
systems are based on CapdetWorks. CapdetWorks provides the following cost curve for aerobic
WWTP, based on the flow rate: O&M cost ($MM) = 0.00042 x (30.7939.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-39
-------
WASTEWATER
• Annual Benefits: Stabilized sludge can be land applied as fertilizer. The cogeneration option
provides electricity. Flared gas can be used elsewhere at the plant to reduce overall energy
requirements.
• Applicability: This option applies to all existing WWTP types.
• Technical Efficiency: This analysis assumes an initial collection efficiency of 60%, which
increases by 10% each year.
• Technical Lifetime: 20 years
III.2.4 Marginal Abatement Costs Analysis
This section describes the methodological approach to the international assessment of Q-k abatement
measures for wastewater treatment systems.
1.2.4.1
Methodological Approach
The MAC analysis is based on project costs developed for a set of model facilities based on the
technical and economic parameters discussed in Section III.2.3. Similar to the steps taken in other sectors,
we developed an inventory of facilities that are representative of existing facilities. Next, we applied the
abatement costs reported above to calculate the break-even prices for each option and wastewater
treatment scenario. Finally, the model estimates the mitigation potential based on the country-specific
share of emissions attributed to each wastewater treatment scenario. Figure 2-9 shows the organization of
the domestic wastewater MAC model. The country-specific distributions are based on analysis conducted
by USEPA (2012a).
Figure 2-9: Domestic Wastewater MAC Analysis Flow Chart
111-40
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
WASTEWATER
Assessment of Sectoral Trends
The first step in the analysis is to assess the level of baseline emissions attributable to domestic versus
industrial wastewater sources. The analysis allocates, when information is available, a percentage of
annual emissions to domestic wastewater treatment. For each country, the remaining share of emissions
is allocated to industrial wastewater treatment.
Shares allocated to each source (domestic/industrial) were based on historical emissions data
obtained from the United Nations Framework Convention on Climate Change's (UNFCCC's) GHG
emissions reporting database. Data were limited to 24 Annex I countries accounting for 15% of emissions
in 2010. For these 24 countries, we calculated a 5-year average share of CH4 emissions attributable to
domestic sources based on emissions reported between 2002 and 2007. For all other countries, because of
a lack of data, we assumed emissions projections are wholly attributable to domestic wastewater
treatment systems to be consistent with USEPA (2012a) projections methodology. Figure 2-10 presents the
average share of emissions attributed to domestic and industrial sources by country.
Figure 2-10: Share of Wastewater ChU Emissions to Domestic and Industrial Sources (Avg. 2002-2007)
Australia
Bulgaria
Croatia
Czech Republic
Finland
Greece
Hungary
Iceland
Ireland
Italy
Japan p 7%
Latvia
Netherlands 4%
New Zealand
Norway 0%
Poland ^H 19%
Portugal
Romania
Russian Federation
Slovakia 2%
Slovenia
Spain
Ukraine 2%
United States of America
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
• 6.B.I - Industrial Wastewater • 6.B.2.1 - Domestic & Commercial (w/o human sewage)
Source: United Nations Framework Convention on Climate Change (UNFCCC). Flexible Data Queries. Online Database. Available at:
http://unfccc.int/di/FlexibleQueries/Event.do?event=hideProjection.
The analysis also leverages estimated changes in wastewater disposal activity along each wastewater
treatment pathway discussed earlier in this chapter. This data was obtained from previous USEPA
analysis used to developed international wastewater projections. Trends in wastewater disposal activity
are determined by population projections, distribution of population between rural and urban settings,
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-41
-------
WASTEWATER
population density, and wastewater flow rates per person. These parameters are used to estimate
country- and technology-specific abatement project costs.
Other trends applied for this analysis include increasing the technical applicability factor and
technical effectiveness factor. The technical applicability factor is assumed to increase at 1% per year
between 2010 and 2030. The technical effectiveness factor increases at a similar rate, growing from 60% to
80% over the 20-year time period. These assumptions are based on expert judgment and intended to
reflect increases in both the adoption of improved sanitation systems and improvements through
learning best management practices for the alternative treatment systems that reduced Q-k emissions.
Estimate Abatement Project Costs and Benefits
Project costs were estimated based on the cost functions defined in Section III.2.3. Country-specific
demographic information on wastewater flow rates and population density was used to estimate the
initial capital costs for each population segment. Table 2-4 provides example abatement measure cost
estimates for the United States and the corresponding break-even prices associated with each option.
Table 2-4: Example Break-Even Prices for Wastewater Abatement Measures in 2030 for the United States
Present Present Present
Value of Value of Value of Tax
Annual After-Tax Benefit of Break-Even
Cost Benefits Depreciation Price
($/tC02e) ($/tC02e) ($/tC02e) ($/tC02e)
Abatement Option
Installed
Reduced Capital
Emissions Costs
(tC02e) ($/tC02e)
Rural
Septic to aerobic WWTP
Latrine to aerobic WWTP
Open sewer to aerobic WWTP
Anaerobic sludge digester with
co-gen
Urban
Septic to aerobic WWTP
Latrine to aerobic WWTP
Open sewer to aerobic WWTP
WWTP— add-on anaerobic
sludge digester with co-gen
6,493,070
288,581
—
57,716
1,082,178
—
—
2,056,139
$51,771
$25,886
—
$6,929
$10,936
—
—
$5,206
$3,080
$1,540
—
$533
$2,251
—
—
$255
$179
$179
—
$154
$179
—
—
$154
$4,106
$2,053
—
$1,180
$867
—
—
$886
$5,100
$2,541
$720
$1,224
—
—
$519
Note: Break-even price was calculated using a 10% discount rate and 40% tax rate.
1.2.4.2
MAC Analysis Results
The global abatement potential of Q-k emissions in wastewater treatment is 36% of total annual
emissions by 2030. Table 2-5 and Figure 2-11 present the MAC curve results for 2030 showing a
cumulative reduction potential of 218 MtCO2e. The top five emitters contribute approximately 58% of
total abatement potential.
Significant initial capital costs combined with no direct monetary benefits limits the abatement
potential achieved at lower break-even prices. As shown in Table 2-5, less than 20% of the total abatement
potential is realized at prices below $50/tCO2e in 2030. These results do not reflect human health benefits
or other positive externalities that accompany improvements in wastewater sanitation. If these additional
social benefits were included, it would result in higher levels of abatement achievable at lower break-
even prices.
111-42
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
WASTEWATER
Table 2-5: Abatement Potential by Region at Selected Break-Even Prices in 2030
Country/Region
-10
Br
-505
eak-Evc
10
m Price ($/tC02e)
15 20 30 50
100
100+
Top 5 Emitting Countries
China
Indonesia
Mexico
Nigeria
United States
—
—
—
—
—
—
—
—
1.4
—
—
—
—
1.4
—
—
—
—
1.4
—
—
—
—
5.2
—
2.9
1.3
—
5.2
—
2.9
1.3
—
5.2
—
11.1
4.4
—
5.2
—
11.1
4.4
3.8
5.2
—
13.0
5.0
3.9
8.0
—
49.7
18.8
20.9
21.9
14.3
Rest of Region
Africa
Asia
Central and South America
Eurasia
Europe
Middle East
North America
World Total
0.2
—
—
—
—
—
—
0.2
0.5
0.7
—
—
—
0.0
—
2.6
0.7
0.8
0.0
—
—
0.5
—
3.4
0.9
0.9
0.0
—
—
0.8
—
4.0
0.9
2.0
0.0
—
—
1.2
—
9.3
1.4
2.6
0.0
0.1
—
2.0
—
15.5
1.5
2.6
0.0
0.1
—
3.5
—
17.1
1.6
3.3
0.2
0.3
—
6.3
—
32.4
1.8
5.7
1.3
0.7
0.0
7.0
—
41.0
3.2
6.8
1.8
1.0
5.5
12.9
—
61.2
10.8
21.1
9.9
8.9
10.9
30.9
0.2
218.3
Figure 2-11: Marginal Abatement Cost Curve for Top 5 Emitters in 2020
$800
$700
5 10 15
Non-CO2 Reduction (MtCO2e)
20
•China
Mexico
•Indonesia
•Nigeria
United States
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-43
-------
WASTEWATER
III.2.4.3 Uncertainties and Limitations
The 2006 version of this report did not explicitly model any abatement measures. This analysis makes
an initial attempt at estimating the abatement potential that could be achieved in the wastewater sector.
The previous report identified two major factors preventing the modeling of abatement in this sector. The
first was data limitations on the type of treatment systems currently employed in each country. The
second was the overriding economic and social factors influencing wastewater treatment practices and
investment throughout the world.
The analysis presented in this chapter attempts to address the data limitations issue by estimating the
quantities of wastewater treated in a number of alternative treatment systems. For simplification
purposes, we have exogenously assigned abatement measures to specific existing wastewater treatment
systems. Ideally, one would have significantly more data on existing treatment pathway types to support
the incorporation of substitutable abatement measures when the investment decision is driven by cost
minimization under country- and system-specific conditions.
The investment in large-scale public infrastructure required to improve wastewater treatment
systems would not be determined solely by the carbon price associated with Q-k emissions reductions.
The public health benefits of such large-scale sanitation infrastructure projects greatly outweigh the
potential benefits provided through any carbon market mechanism. However, the analysis presented
here estimates the level of abatement that is technically achievable and the marginal costs of supplying
reductions through these technologies, ignoring other potential positive externalities derived from
putting these systems in place.
Finally, this chapter does not consider the potential impact of abatement measures applied to
industrial wastewater treatment systems. The authors acknowledge that CH4 emissions from industrial
sources can be significant, and in some countries industrial wastewater emissions may represent more
than half of total emissions associate with wastewater. However, data limitations, specifically information
on the types of treatment systems employed in specific industries and correspondingly the abatement
measures available to those systems is needed to estimate the abatement potential from industrial
sources. International partnerships like the Global Methane Initiative (GMI) have begun to assess the
level of CH4 emissions available for recovery and use. Any future attempt to model abatement potential
from the industrial wastewater sector would also require additional detail on the relative contribution of
CH4 emissions coming from domestic versus industrial wastewater sources.
111-44 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
WASTEWATER
References
Bogner, ]., R. Pipatti, S. Hashimoto, C. Diaz, K. Mareckova, L. Diaz, P. Kjeldsen, S. Monni, A. Faaij, Q.
Gao, T. Zhang, M. Abdelrafie Ahmed, R.T.M. Sutamihardja, and R. Gregory. 2008. "Mitigation of
Global Greenhouse Gas Emissions from Waste: Conclusions and Strategies from the
Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report. Working Group III
(Mitigation)." Waste Management Research 26:11.
Buchanan, J.R. 2010. Performance and Cost of Decentralized Unit Processes. Publication No. DEC2R08.
Alexandria, VA: Water Environment Research Foundation (WERF). Obtained at:
http://ndwrcdp.werf.org/documents/DEC2R08/DEC2R08web.pdf.
Finney, B., and R. Gearheart. 2004. WAWTTAR: Water and Wastewater Treatment Technologies
Appropriate for Reuse. Obtained at: http://firehole.humboldt.edu/wawttar/wawttar.html.
Intergovernmental Panel on Climate Change (IPCC). 2006. 2006 IPCC Guidelines for National Greenhouse
Gas Inventories: Volume 5: Waste; Chapter 6: Wastewater Treatment and Discharge. Obtained at:
http://www.ipcc-nggip.iges.or.Jp/public/2006gl/pdf/5 Volume5/V5 6 Ch6 Wastewater.pdf.
Scheehle, E.A., and M.R.J. Doom. 2002. "Improvements to the U.S. Wastewater Methane and Nitrous
Oxide Emissions Estimates." Working paper. USEPA: Washington, DC.
U.S. Environmental Protection Agency (USEPA). 2012b. Inventory of U.S. Greenhouse Gas Emissions and
Sinks 1990-2011. Washington, DC: USEPA, Office of Solid Waste and Emergency Response. Obtained
at: http://www.epa.gov/climatechange/emissions/usinventoryreport.html.
U.S. Environmental Protection Agency (USEPA). 2012a. Global Anthropogenic Non-COi Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA.
U.S. Environmental Protection Agency (USEPA). 2009. Municipal Wastewater Treatment Sector: Options for
Methane Emission Mitigation (Draft). Methane to Markets Administrative Support Group. Obtained at:
http://www.globalmethane.org/documents/events steer 20090910 scoping.pdf.
U.S. Environmental Protection Agency (USEPA). 1980. Construction Costs for Municipal Wastewater Plants:
1973-1978. Document EPA/430/9-80-003.
van de Guchte, C., and V. Vandeweerd. 2004. Targeting Sanitation. Our Planet, the Magazine of the United
Nations Environment Programme (UNEP). Issue: Water, Sanitation, People. 14(4):19-21.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 111-45
-------
IV. Industrial Processes Sector
-------
NITRIC AND ADIPIC ACID PRODUCTION
IV. 1. Nitric and Adi Die Acid Productior
IV.1.1 Sector Summary
he production of nitric and adipic acid results in significant nitrous oxide (N2O) emissions as a
by-product. Nitric and adipic acid are commonly used as feedstocks in the manufacture of a
variety of commercial products, particularly fertilizers and synthetic fibers (USEPA, 2012a).
Combined global emissions of N2O from nitric and adipic acid production are shown in Figure 1-1.
Globally, emissions have declined by about 13% (17 MtCO2e) over the past decade; however, over the
next 20 years—2010 to 2030—emissions are expected to increase steadily (-20%) growing by 24 MtCO2e
over the time period. This trend is largely due to increased demand for fertilizer (nitric acid is an input)
and increased demand for synthetic fibers (adipic acid is an input). In 2030, the United States, South
Korea, Brazil, China, and Ukraine are expected to be five largest emitters of N2O from nitric and adipic
acid production.
Figure 1-1: N20 Emissions from Nitric and Adipic Acid: 2000-2030
142
Ukraine
I China
Brazil
I South Korea
I United States
I Rest of World
2000
2010
2020
2030
Year
Source: USEPA, 2012a.
Over the coming decades, increased demand for adipic acid in Asia is expected to contribute to
higher N2O emissions from adipic acid production, while abatement control technologies employed at
adipic acid production facilities in the 1990s in the United States, Canada, and some countries of the EU
are expected to reduce N2O emissions (USEPA, 2012a). Overall, increased global demand for adipic acid
is expected to have the effect of higher annual N2O emissions resulting from adipic acid production. In
addition, global N2O emissions from nitric acid are expected to increase as demand for nitrogen-based
fertilizer increases. Although concerns about nutrient run-off have caused some countries to reduce their
demand for nitrogen-based fertilizer, growing world demand for agricultural commodities is expected to
have the effect of increasing nitric acid production and, consequently, N2O emissions.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-1
-------
NITRIC AND ADIPIC ACID PRODUCTION
Global abatement potential of N2O in nitric acid and adipic acid production is 98 and 115 MtCO2e in
2020 and 2030, respectively. These results are depicted in the sectoral marginal abatement cost (MAC)
curves in Figure 1-2. As the MAC curves show, roughly 45% of the maximum abatement potential in each
year is achievable at relatively low carbon prices (between $2 and $10 tCO2e-1).
Figure 1-2: Global MAC for Nitric and Adipic Acid: 2010, 2020, and 2030
•2010
•2020
•2030
140
Non-CO2 Reduction (MtCO2e)
The following section provides a brief explanation of the manufacturing processes that result in the
formation of N2O emissions. Next we discuss the projected emissions from these processes out to 2030.
Subsequent sections characterize the abatement measures and present the cost of implementation and
operation for each. The chapter concludes with a discussion of the MAC analysis approach unique to this
sector and the regional results.
IV.1.2 N2O Emissions from Nitric and Adipic Acid Production
N2O emissions are closely correlated with the production of nitric and adipic acid. The following
section discusses global production activity data, typical emissions factors, and baseline emissions
estimates of N2O from nitric and adipic acid production. The MAC analysis presented here starts by
assuming the projected emissions presented in USEPA's 2012 Global Emissions Report (2012a). This
analysis then derives industry activity from the USEPA emissions projections based on current industrial
activity.
IV.1.2.1
Nitric Acid Production and Emission Factors
Ammonium nitrate production represents the largest demand market for nitric acid, with the
majority of nitric acid being consumed by ammonium nitrate producers. The demand for ammonium
IV-2
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
NITRIC AND ADIPIC ACID PRODUCTION
nitrate products, especially fertilizer, is the main driver of nitric acid demand. Nitric acid production
levels closely follow trends in fertilizer demand (Mainhardt and Kruger, 2000). Trends in fertilizer
demand vary widely across different regions of the world. For example, in Western Europe, because of
concerns over nutrient runoff and nitrates in the water supply, use of nitrogen-based fertilizer has been
significantly reduced in the past couple of decades (USEPA, 2012a). Despite this trend, the European
Fertilizer Manufacturers Association (EFMA) is predicting modest growth in demand of 1.3% annually
over this decade. In other parts of the world, a continued desire to secure domestic fertilizer production
capacity to supplant reliance on imports in combination with expansions in capacity for large fertilizer
consuming countries are expected to result in a net increase in nitrogen-based fertilizer production
capacity between 2010 and 2015 (IFA, 2011). Globally, over the next several decades, increases in food
consumption and demand for agricultural products will continue to put upward pressure on fertilizer
demand, which, in turn, is expected to increase the demand and consumption of nitric acid.
The actual number of nitric acid production plants globally is unknown. Previous reports cited by the
Intergovernmental Panel on Climate Change (IPCC) have suggested the number to be between 255 and
600. More recent estimates suggest that between 500 and 600 plants were in operation in 2010 (Kollmus
and Lazarus, 2010). The actual number is uncertain because many nitric acid plants are often part of
larger integrated chemical facilities that manufacture products using nitric acid—in the production of a
wide range of chemical products such as fertilizer and explosives (Kollmus and Lazarus, 2010; Mainhardt
and Kruger, 2000).
As noted earlier, global nitric acid production is expected to increase over time. Projections in nitric
acid production levels by country are not publicly available.
The IPCC reports that N2O emissions factors for nitric acid production also remain relatively
uncertain, because of a lack of information on manufacturing processes and emissions controls. The
emissions factor is estimated, based on the average amount of N2O generated per unit of nitric acid
produced, combined with the type of technology employed at a plant. The IPCC uses a default range of 2
to 9 kilograms N2O per ton of nitric acid produced. As a result, emissions factors for nitric acid
production plants may vary significantly based on the operating pressure of the plant, the type of
nitrogen oxide (NOx) control technology (if any) deployed on the plant, and whether N2O abatement has
been implemented. As shown in Table 1-1, N2O emission rates increase as the plant operating pressure
increases. Furthermore, non-selective catalytic reduction (NSCR) is very effective at destroying N2O,
whereas other technologies used to control NOx emissions, such as selective catalytic reduction (SCR) and
extended absorption, do not reduce N2O emissions.
Table 1-1: IPCC Emissions Factors for Nitric Acid Production
N20 Emissions Factor
Production Process (relating to 100% pure acid)
Plants with NSCRa (all processes)
Plants with process-integrated or tailgas N2
-------
NITRIC AND ADIPIC ACID PRODUCTION
worldwide do not employ NSCR technology, which makes it more likely that the default range of
potential emissions factors provided by the IPCC underestimates the true emissions baselines (Mainhardt
and Kruger, 2000). Therefore, the uncertainties associated with these emission factors may be higher than
listed in Table 1-1. However, no uncertainty assessments other than the IPCC's have been published, so
without more information, this analysis relies on the published values above.
IV.1.2.2
Adipic Acid Production and Emission Factors
Adipic acid is used primarily in the production of synthetic fibers, predominantly as a precursor for
the production on nylon 6,6 (The Chemical Company [CC], 2010a). As a result, production of adipic acid
is closely correlated with the world's nylon production. Worldwide, the largest single use of adipic acid is
in carpet manufacturing, accounting for 30% of the market (USEPA, 2012a; Chemical Week, 2007).
Global demand for adipic acid is expected to increase over the next few decades, particularly in Asia,
driven primarily by the growth in demand for synthetic fibers (i.e., nylon 6,6), particularly for use in
carpet manufacturing. Nylon 6,6 accounts for approximately 90% of adipic acid demand; demand for
nylon is a strong indicator of future adipic acid demand (CC, 2010b). Future production of adipic acid is
expected to closely following the demand trend for synthetic fibers. Figure 1-3 shows the share of adipic
acid production capacity by country in 2010.
Figure 1-3: Adipic Acid Production Capacity by Country: 2010
South Korea, 5%
Japan, 5%
Singapore, 4%
Brazil, 3%
Italy, 3%
Ukraine, 2%
India, 0%
Global capacity in 2010 was approximately 3 million metric tons, concentrated in the United States
(30%), the European Union (29%), and China (22%) (Schneider, Lazarus, and Kollmuss, 2010; USEPA,
2012b). In the same year, global production of adipic acid was approximately 2.5 million metric tons (CC,
2010a). Plants in located in Canada and the United Kingdom were recently shut down and two of four
facilities in the United States were idle in 2009 and assumed to remain so in 2010 (USEPA, 2012b).
The IPCC provides a default emissions factor of 300 kilograms ± 10% N2O per ton of adipic acid
produced (IPCC, 2006). This emissions factor assumes that no N2O control system is in place.
Additionally, this factor should be used only when national total data are available and plant-level data
are deemed unreliable. This factor was developed using laboratory experiments measuring the
IV-4
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
NITRIC AND ADIPIC ACID PRODUCTION
reactionary stoichiometry for N2O generation during the production of adipic acid (Mainhardt and
Kruger, 2000). This emissions factor has been supported by some selected measurement at industrial
plants. The IPCC recommends using plant-specific data for those plants with abatement controls already
in place and reliable plant-level data (IPCC, 2006).
IV.1.2.3 Emissions Estimates and Related Assumptions
This section discusses the historical and projected baseline emissions from the production of nitric
and adipic acid.1 Historical emissions are characterized as those emissions released from the 1990 to 2010
period and projected emissions estimates cover the period from 2010 to 2030.
Historical Emissions Estimates
Between 1990 and 2005, N2O emissions from nitric and adipic acid production decreased by 37%
down from 200 MtCO2e in 1990 to 126 MtCO2e in 2005. Key factors that influence emissions are the
demand for final products that include intermediates of nitric and adipic acid such as carpet and
fertilizers. Although demand for and production of nitric and adipic acid increased over the 1990 to 2005
time period, N2O emissions actually decreased over the period. The reductions in N2O emissions over
this period were mostly due to the installation of abatement technologies in the adipic acid industry
(USEPA, 2012a).
Projected Emissions Estimates
Table 1-2 lists the combined projected N2O baseline emissions from nitric and adipic acid production
by country/region and year. Worldwide N2O emissions are projected to increase by approximately 21%
(24 MtCO2e) between 2010 and 2030. South Korea, Canada, and Brazil are expected to experience the
largest percentage increase in baseline emissions over the 2010 to 2030 period, with increases of 93, 77,
and 68%, respectively. The United States is expected to have the second largest absolute increase (8
MtCO2e) in emissions, which represents a 28% increase from 2010, while South Korea is expected to have
the largest absolute increase (11 MtCO2e).
IV.1.3 Abatement Measures and Engineering Cost Analysis
This analysis considered four abatement measures applied to the chemical processes used to produce
nitric and adipic acid to reduce the quantity of nitrous oxide (N2O) emissions released during the
production process. Thermal destruction is the abatement measure applied to the adipic acid production
process. The three remaining measures were applied to the nitric acid production process.
Nitric acid facilities have the option of using specially designed catalysts to control N2O emissions.
The location of catalyst placement within the nitric acid production process determines the catalyst
design, composition, and terminology. Abatement measures applicable to nitric acid are characterized by
where in the production process they are implemented. These options include primary abatement,
secondary abatement, and tertiary abatement. Primary abatement measures occur within the ammonia
burner, preventing the formation of N2O. Secondary abatement measures such as homogeneous thermal
decomposition and catalytic decomposition are installed at an intermediate point in the production
process, removing the N2O formed through ammonia oxidation. Tertiary abatement measures, such as
catalytic decomposition and NSCR units are applied to the tail gas streams at the end of the nitric acid
1 For more detail on baseline development and estimation methodology, we refer the reader to the USEPA's Global
Emissions Projection Report. Available at: http://www.epa.gov\ climatechange \ economics \ international.html.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-5
-------
NITRIC AND ADIPIC ACID PRODUCTION
Table 1-2: Projected N20 Baseline Emissions from Nitric and Adipic Acid Production: 2010-2030
Country
2010
2015
2020
2025
2030
CAGR
(2010-2030)
Top 5 Emitting Countries
United States
South Korea
Brazil
China
Ukraine
29
12
8
7
6
30
14
9
8
7
32
16
11
9
7
35
19
12
10
7
37
23
14
10
7
1.2%
3.4%
2.6%
1.6%
0.4%
Rest of Region
Africa
Central & South America
Middle East
Europe
Eurasia
Asia
North America
World Total
3
1
1
32
7
9
2
118
3
2
1
25
7
10
3
118
3
2
1
25
7
10
3
127
4
2
1
25
7
11
3
136
3
2
1
29
7
12
4
147
0.1%
0.4%
0.5%
-0.5%
0.2%
1.3%
2.3%
1.1%
aCAGR = Compound Annual Growth Rate
b Mexico is the only country included under North America, as Canada and the United States are reported individually above.
Source: USEPA, 2012a.
production process. The implementation of one technology over another is driven largely by facility
design constraints and/or cost considerations. The high operating costs of NSCR units and improvement
in modern facility design will drive most future abatement projects to adopt secondary or tertiary
catalysts over NSCR units.
This section briefly characterizes each abatement measure and the supporting technical assumptions
that were used to compute the breakeven prices. Table 1-3 summarized the costs and technical
assumptions for the four abatement measures. Abatement measure costs were derived from a variety of
sources reporting in euros and dollars over a number of base years. For consistency, we assumed a fixed
exchange rate of 1.32 (USD/EUR), and the Chemical Engineering Plant Cost Index (CEPI) was used to
adjust costs for inflation. Consistent with other sectors evaluated in this study, the costs of abatement
developed for this analysis exclude capital and O&M costs attributable to monitoring, reporting, and
verification (MRV) activities.
IV.1.3.1
Adipic Acid—N2O Abatement Methods
Adipic acid facilities typically direct the flue gas to a reductive furnace in a thermal destruction
process to reduce nitric oxide (NOx) emissions. Thermal destruction is the combustion of off-gases
(including N2O) in the presence of methane. The combustion process converts N2O to nitrogen, resulting
primarily in emissions of NO and some residual N2O (Ecofys, Fraunhofer ISIR, and Oko-Institute, 2009).
Facilities may also employee a catalytic decomposition method to abatement N2O generated. The EU
Emissions Trading System [ETS] and CDM methodologies for this abatement measure suggest that heat
generated from the decomposition of N2O can be used to produce process steam for utilization in local
processes, substituting for more expensive steam generated using fossil fuel alone. For this analysis, we
assume the abatement measures' conversion of N2O to nitrogen technical effectiveness is 95%. Costs
presented below are based on a catalytic decomposition method.
IV-6
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
NITRIC AND ADIPIC ACID PRODUCTION
Table 1-3: Abatement Measures for Nitric and Adipic Acid Production
010 USD
Total Annual Benefits
Installed Annual Time Average
Capital O&M Horizon Technical Non- Reductions
Cost Cost (Years) Efficiency,% Energy energy (tC02e/yr)
Adipic Acid Production3
Thermal/catalytic decomposition
11.4
2.2
20
96%
- 0.3 4,206,218
Nitric Acid Production11
Secondary Abatement-
Catalytic decomposition in the burner
Tertiary Abatement-
Direct catalytic decomposition
Tertiary Abatement—
Non-selective catalytic reduction unit
1.3
2.3
4.0
0.4
0.2
2.1
15
15
20
85%
95%
95%
- - 779,571
- - 871,286
- - 871,286
a Based on adipic acid plant capacity of 200 metric tons of adipic acid per day.
b Based on nitric acid plant capacity of 1,000 tHNOs/day.
• Capital Costs: Initial capital costs are $156 per metric ton of production capacity in 2010 dollars.
This cost includes the costs of engineering design and process modifications in addition to
equipment and installation costs. Assuming a plant with capacity of 200 tonnes adipic acid
production per day, the initial capital cost would be approximately $11.4 million (2010 USD).
• Annual O&M Costs: Annual costs total $38 per metric ton of production in 2010 dollars, which
includes the costs of annual energy requirements and system maintenance. Assuming a plant
with capacity of 200 tonnes and a utilization factor of 80%, the annual operating cost would be
$2.2 million (2010 USD). Catalyst consumption represents 60% of the annual costs.
• Annual Benefits: Steam produced through the decomposition of N2O under this abatement
measure can offset steam generated using more expensive energy sources providing a fuel cost
savings. These annual benefits can equal up to 60% of operating costs (Ecofys et al., 2009). This
analysis assumes a more conservative estimate of 16% of operating costs or $5.6 per metric ton of
adipic acid production based on CDM project documentation.
• Applicability: This option applies to adipic acid production facilities that do not currently
control N2O emissions. Based on recent analysis (Schneider et al., 2010), only 9 of the 23
operational facilities in 2010 had unabated N2O emissions.
• Technical Effectiveness: This analysis assumes a 95% efficiency converting N2O into nitrogen
and water.
Technical Lifetime: 20 years
IV.1.3.2
Nitric Acid—Primary Abatement Measures
This group of abatement measures can be applied at the ammonia oxidation stage of the nitric acid
production process. Perez-Ramirez (2003) identified three alternative approaches categorized as primary
abatement options: optimized oxidation, modification of the Pt-Rh gauzes, and oxide-based combustion
catalysts. All three technologies prevent the formation of N2O in the ammonia burner and would require
making adjustments to the ammonia oxidation process and/or catalyst (Perez-Ramirez, 2003). Although
the primary abatement technology options are technically feasible, they are not modeled in this analysis
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-7
-------
NITRIC AND ADIPIC ACID PRODUCTION
because of a lack of technology cost data and the fact that the alternative options discussed below achieve
greater abatement and are better defined.
IV.1.3.3 Nitric Acid—Secondary Abatement Measures
Secondary abatement measures remove N2O immediately following the ammonia oxidation stage,
between the ammonia converter and the absorption column (Perez-Ramirez, 2003). Abatement measures
include thermal decomposition, catalytic decomposition inside or immediately following the ammonia
burner. Thermal decomposition, developed by Norsk Hydro in the 1990s, is better suited for inclusion in
new plants, because it requires redesigning the reaction chamber immediately following the ammonia
burner. This design change can increase the capital cost of a new plant by 5 to 6% but has no impact on
operating costs (Perez-Ramirez, 2003). The catalytic decomposition option is better suited for retrofitting
and can be incorporated as an add-on technology at minimal cost. For this analysis, the catalytic
decomposition costs were used as the representative costs of the secondary abatement option.
• Capital Costs: Capital costs include the purchase and installation of the catalyst and any
technical modifications made to the production process. This analysis assumes a capital cost of
$3.5/tonne of HNOs production capacity2 and a plant capacity of 1,000 tHNOs/day. Using these
assumptions, the initial capital costs would equal $1.3 million (2010 USD).
• Annual O&M Costs: Annual costs include catalyst replacement and recycling of spent catalyst,
replacement of spare catalyst, and loss of production due to catalyst disruptions. This analysis
assumes an annual cost of $1.3/tHNOs produced and a plant utilization rate of 90% (Perez-
Ramirez, 2003). Following the plant example of a 1,000 tHNOs/day, the annual cost would be $0.4
million (2010 USD).
Annual Benefits: No benefits are associated with this option.
Applicability: This option is applicable to all existing nitric acid plants.
Technical Effectiveness: This analysis assumes an 80% efficiency converting N2O into nitrogen
and water.
Technical Lifetime: 20 years
IV.1.3.4 Nitric Acid—Tertiary Abatement Measure: Direct Catalytic Decomposition
Tertiary abatement measures are located after the absorption tower where tail gas leaving the
absorption column is treated to destroy N2O (Perez-Ramirez, 2003). Similar to earlier abatement
measures, this measure reduces the N2O into nitrogen and oxygen, through thermal or catalytic
decomposition.
• Capital Costs: Capital costs include the purchase and installation of the catalyst and any
technical modifications made to the production process. This analysis assumes a capital cost of
$6.3/tonne of HNOs production capacity3 and a plant capacity of 1,000 tHNOs/day. Using these
assumptions, the initial capital costs would equal $2.3 million (2010 USD).
2 Based on costs of € 0.25/tHNO3 reported in 2008 euros (EC, 2008) scaled to 2010 USD using the Chemical
Engineering Plant Cost Index (CEPI) and an exchange rate of 1.32 (USD/EUR).
3 Based on costs of € 0.5/ tHNOs reported in 2008 euros (EC, 2008) scaled to 2010 USD using the Chemical
Engineering Plant Cost Index (CEPI) and an exchange rate of 1.32 (USD/EUR).
IV-8 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
NITRIC AND ADIPIC ACID PRODUCTION
• Annual O&M Costs: Annual costs include catalyst replacement and recycling of spent catalyst;
replacement of spare catalyst; loss of production due to catalyst disruptions or the lowering of
the process pressure. This analysis assumes an annual cost of $0.6/tHNO3 produced and a plant
utilization rate of 90% (Perez-Ramirez, 2003). Following the plant example of a 1,000 tHNOs/day,
the annual cost would be $0.2 million (2010 USD).
• Annual Benefits: Minor benefits are associated with this option. Decomposition is an exothermic
process, so a small amount of heat could be recovered from the process and converted to steam.
However, the costs of the equipment needed to recover the heat and convert it to steam could
outweigh the benefit. The ability to accrue benefits would also be limited by the amount of space
available to add the equipment.
• Applicability: This option is applicable to most existing nitric acid plants but is highly dependent
on site-specific factors, such as age of the facility and the footprint of the facility. Tertiary
abatement measures may require additional space and additional equipment.
• Technical Effectiveness: The analysis assumes a 82% efficiency converting N2O into nitrogen
and water.
• Technical Lifetime: 20 years
IV.1.3.5 Nitric Acid—Tertiary Abatement Measure: Non-selective Catalytic
Reduction (NSCR)
One specialized type of tertiary catalyst is an NSCR system. The NSCR typically costs more than
other types of tertiary catalysts because it requires a reagent fuel, such as natural gas, propane, butane, or
hydrogen, to reduce NOx and N2O over a catalyst. If an NSCR system is used at a nitric acid plant that is
collocated with other chemical processes, the costs of these reagent fuels may be lessened. For example, if
ammonia is produced near the nitric acid production plant, the waste gas stream from ammonia
production is a hydrogen-rich gas stream that could be used as the reagent fuel for an NSCR.
• Capital Costs: Capital costs include the purchase and installation of the NSCR unit and catalyst.
This analysis assumes a capital cost of $12.6/tonne of HNOs production capacity based on
$8.2/tHNC>3 reported in 1991 USD (USEPA, 1991) scaled to 2010 USD using the Chemical
Engineering Plant Cost Index (CEPI). Assuming a plant capacity of 1,000 tHNOs/day, the initial
capital cost would equal $4.6 million (2010 USD).
• Annual O&M Costs: Annual costs total $8.8/tHNOs produced. Annual costs include the cost of
reagent fuel, labor, maintenance, and other fixed costs for capital recovery and insurance. Total
annual cost for the example plant would be $2.9 million per year (2010 USD).
• Annual Benefits: Energy benefits are associated with this option. The NSCR reaction is
exothermic, which means that the reaction generates heat. This heat can be recovered and
converted into steam for use as an energy source.
• Applicability: This option is applicable to all nitric acid production facilities without existing
tertiary abatement measures. Although it is theoretically possible to employ multiple abatement
measures, the likelihood of multiple retrofitted abatement measures operating together in an
efficient manner is very low.
• Technical Effectiveness: The analysis assumes 90% efficiency converting N2O into nitrogen and
water.
• Technical Lifetime: 20 years
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-9
-------
NITRIC AND ADIPIC ACID PRODUCTION
IV.1.4 Marginal Abatement Costs Analysis
This section describes the methodological approach to the N2O abatement measures for nitric and
adipic acid production facilities.
IV.1.4.1 Methodological Approach
The MAC analysis is based on project costs developed for a set of model facilities based on the
abatement measure costs discussed in Section IV.1.3. Similar to the steps taken in other sectors, we
developed an inventory of facilities that are representative of existing facilities. Next we applied the
abatement costs to calculate the break-even prices for each option and applicable facility pair. Finally, the
model estimates the mitigation potential based on the country-specific share of emissions attributed to
nitric versus adipic acid production. This analysis takes the N2O emissions projections (given) and
allocates emissions based on production process to derive the model facility inventories.
Adipic acid facilities are defined through a detailed inventory of the 23 production facilities
worldwide operating in 2010. While no comprehensive inventory was available for nitric acid plants, it is
believed that there are roughly 500 to 600 nitric acid plants globally (Kollmuss and Lazarus, 2010).
Instead, we developed a series of 4 model nitric acid production units based on plant characteristics
obtained from a detailed inventory of 67 nitric acid plants that varied in age and production processes.4
Adipic Acid—Facility inventory
The first step in the analysis was to determine the allocation of projected emissions to nitric and
adipic acid production by country. For example, in the United States, adipic acid production accounted
for approximately 15% of total baseline emissions in 2010, while the majority of emissions were attributed
to nitric acid production. Once the share of baseline emissions is determined, the MAC model can assess
the abatement potential on the technically applicable pool of emissions available for abatement.
To estimate the technically applicable share of emissions, we developed a detailed inventory of
operational adipic acid plants in 2010. Adipic acid plants were used as the starting point because the
number of international adipic acid plants is small (<30 globally), supported by recent literature
providing detail on existing plants in 2010 obtained from Schneider, Lazarus, and Kollmuss (2010). The
detailed inventory includes 23 adipic acid production facilities operating in 11 countries totaling
approximately 3,000 kt of production capacity.5 Schneider and co-authors also identified the N2O
abatement technologies and plant utilization factors. Figure 1-4 summarizes the global adipic acid
production capacity breakdown by country, and facility counts are reported in parenthesis after the
country labels. The bottom-up inventory was used to estimate N2O emission from adipic acid production
by country.
4 While there are a number of different processes employed at nitric acid production facilities, single-pressure plants
are much more common in the United States. Based on information gathered, most nitric acid production plants were
constructed to maximize the yield from stages two and three of the production process and, therefore, operate at high
pressures.
5 Major changes to previously reported adipic acid inventories (Mainhardt and Kruger, 2000; Organisation for
Economic Co-operation and Development [OECD], 2004) includes the opening of 5 new plants in China between
2008 and 2009; and the closure of two plants located in Canada and the United Kingdom. In addition, a fourth plant
located in the United States was idle between 2008 and 2009 and assumed to continue to idle in 2010 (USEPA, 2012b).
IV-10 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
NITRIC AND ADIPIC ACID PRODUCTION
Figure 1-4: Operational Adipic Acid Production Facilities in 2010 by Share of Global Capacity
8
t>0
E
0.
United States (3)
China (7)
Germany (3)
France (1)
South Korea (1)
Japan (2)
Singapore (1)
Brazil (1)
Italy (1)
Ukraine (2)
India (1)
^•2%
0%
1%
• with N2O Abatement
• Unabated
2%
0% 5% 10% 15% 20% 25%
Share of Global Production Capacity
30%
35%
Source: Adapted from Schneider et al. 2010.
Note: Facility counts are listed in parentheses beside country name.
Although 11 countries currently produce adipic acid, only 4 countries (China, Ukraine, Japan, and
India) have operational facilities that are known to have no N2O emission controls in place. As the figure
shows, all but 15% of the adipic acid capacity has N2O abatement controls in place. The 15% of capacity
with no N2O abatement controls is represented by the nine smallest facilities in the industry located in
China (5), Ukraine (2), Japan (1), and India (1).
In the 1990s, most of the adipic acid producers in developed countries voluntarily adopted N2O
abatement measures (Schneider et al., 2010; EcoFys, 2009, USEPA, 2012b). In 2005, with the establishment
of the CDM methodology for crediting N2O abatement projects at adipic acid plants, producers in
developing countries began to adopt N2O abatement measures. Schneider and co-authors point out that
although the CDM methodology was effective in achieving N2O reductions in developing countries, it
was limited to facilities that were in operation prior to 2005.
Since 2005, much of the growth in adipic acid production capacity has been in China, with five plants
coming online between 2008 and 2009 (Schneider et al., 2010). Future growth is also projected to be highly
concentrated in Asia (Global Industry Analysts Inc. [GIA], 2010). China alone is expected to see its
capacity more than double in the near term with five new adipic acid plants between 2011 and 2013
(Zhao, 2011).
Only 15% of global capacity continues to operate with no known N2O abatement. China and Ukraine
account for over 95% of the capacity with unabated N2O emissions. In China, the five plants operating
without abatement controls account for two-thirds of the country's total adipic acid capacity. For this
analysis, we assume that future abatement potential is limited to the nine plants identified as having no
known N2O abatement measure in place.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-11
-------
NITRIC AND ADIPIC ACID PRODUCTION
Although no information was available on specific plant utilization rates, we assume utilization rates
of 60% for all non-CDM facilities, 85% for CDM facilities,6 and 45% for non-CDM facilities in other parts
of Asia (Schneider et al., 2011). Combining plant capacities and corresponding utilization rates yields a
total adipic acid production in 2010 of 1.84 million metric tons.
Next we estimate net emissions for each country by applying the IPCC emissions factor of 300 kg
N2O per metric ton of adipic acid produced to the plant production estimated above. Net emissions
estimated account for existing abatement activity assuming a control efficiency of 96%. This analysis
yields net emissions by country totaling 103,800 tonnes of N2O (32.2 MtCO2e) in 2010.
We assume the net emissions calculated for each country represents adipic acid's representative share
of total projected baseline emissions (see Table 1-2). Table 1-4 provides the percentages used to breakout
the N2O emissions baseline to adipic acid.
The analysis assumes that N2O emissions from adipic acid production account for the percentage of
total sectoral baseline listed in Table 1-4. We attribute the balance of baseline emissions to nitric acid
production.
Table 1-4: Adipic Acid-Producing Countries' Share of Baseline Emissions3
Country
Brazil
China"
France
Germany
India
Italy
Japanb
Singapore
South Korea
Ukraine13
United States
Other Countries
Share of N20 Baseline, %
Adipic Acid
5
36
30
21
1
27
36
25
5
36
15
0
Nitric Acid
95
64
70
79
99
73
64
75
95
64
85
100
a For China, Japan, and Ukraine, the more detailed inventory-based estimate of emissions developed for this analysis yielded emission values
greater than the total baseline projections for 2010. Hence, we defaulted back to percentages assumed for the previous report (36%).
b China, Japan, and Ukraine percentages used are from EMF 21 MAC model (USEPA, 2006).
Model Facility Description for Nitric Acid
While it is believed that there are roughly 500 to 600 nitric acid plants globally (Kollmuss and
Lazarus, 2010), no comprehensive inventory was available for nitric acid plants. Instead, we developed a
series of 4 model nitric acid production units based on plant characteristics obtained from a detailed
inventory of 67 nitric acid plants that varied in age and production processes. We organize the model
facilities based on production capacity. All four facility types are assumed have an uncontrolled
' Facilities located in Brazil, China, and South Korea.
IV-12
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
NITRIC AND ADIPIC ACID PRODUCTION
emissions factor of 8.5 kg N2O per tHNOs produced7 (IPCC, 2006). Table 1-5 summarizes the model
facilities for nitric acid production by capacity and resulting annual N2O emissions.
Table 1-5: Model Nitric Acid Facilities Assumptions
Model Plants
Small
Medium
Large
Modern plant
Production
(tHNOs/yr)
30,600
113,333
226,667
340,000
Annual N20 Emissions
(uncontrolled)
(tN20)
261
968
1,936
2,904
Estimate Abatement Project Costs and Benefits
Abatement measure costs and technical efficiencies were applied to each of the model facilities to
estimate the break-even prices. Based on facility characteristics, we estimate the abatement project costs
and calculated the break-even prices. Table 1-6 summarizes the break-even price calculation for nitric and
adipic acid production facilities.
Table 1-6: Example Break-Even Prices for N20 Abatement Measures
Annualized
Reduced Capital Annual Annual Tax Benefit of Break-
Emissions Costs Cost Revenue Depreciation Even Price
Abatement Option (MtC02e) ($/tC02e) ($/tC02e) ($/tC02e) ($/tC02e) ($/tC02e)
Adipic Acid Production
Thermal destruction
9.2
0.22
1.15
1.12
0.04
0.21
Nitric Acid Production
Secondary Abatement-
Catalytic decomposition in the
burner
Tertiary Abatement—
Tailgas catalytic decomposition
Tertiary Abatement—
NSCRunit
0.8
0.9
0.9
0.4
0.5
1.0
0.6
0.2
3.3
0.0
0.0
0.0
0.1
0.1
0.2
0.86
0.67
4.19
Note: Break-even price assumes 10% discount rate and 40% tax rate.
Thermal destruction based on adipic acid production capacity of 75 kt yr1 and nitric acid options based on 328kt HMOs yr1 production capacity.
IV.1.4.2
MAC Analysis Results
Global abatement potential of N2O emissions in nitric and adipic acid production is 78% of annual
emissions. The majority of abatement potential is associated with nitric acid production because of the
high degree of abatement already occurring at adipic acid facilities. Table 1-7 and Figure 1-5 present the
MAC curve results for 2030 showing a cumulative reduction potential of 111 MtCO2e.
7 The default emissions factor for the high pressure process is 9 kg N2O per ton of nitric acid; the default emissions
factor for medium pressure processes is 7 kg N2O per ton of nitric acid produced.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-13
-------
NITRIC AND ADIPIC ACID PRODUCTION
Table 1-7: Abatement Potential by Region at Selected Break-Even Prices in 2030
Country/Region
-10
-505
10
en Prici
15
5 ($/tCC
20
26)
30 50
100
100+
Top 5 Emitting Countries
Brazil
China
South Korea
Ukraine
United States
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
2.8
2.6
4.6
3.0
4.7
7.6
4.1
12.4
4.8
10.3
8.5
7.8
13.9
4.8
20.0
8.5
7.8
13.9
5.7
20.0
11.3
7.8
18.4
5.7
23.4
11.3
7.8
18.4
5.7
26.5
11.3
7.8
18.4
5.7
26.5
11.3
7.8
18.4
5.7
26.5
Rest of Region
Africa
Central and South America
Middle East
Europe
Eurasia
Asia
North America
World Total
—
—
—
—
—
—
—
0.0
—
—
—
—
—
—
—
0.0
—
—
—
—
—
—
—
0.0
1.4
0.3
0.2
4.3
2.9
2.3
0.6
29.6
2.0
0.9
0.5
12.2
4.6
5.5
1.3
66.3
2.7
1.0
0.6
16.4
6.0
6.6
2.3
90.8
2.7
1.1
0.6
18.9
6.0
7.5
2.4
95.1
2.7
1.3
0.8
20.8
6.1
8.4
2.8
109.6
2.7
1.3
0.8
23.2
6.1
8.8
3.1
115.8
2.7
1.3
0.8
23.2
6.1
8.8
3.1
115.8
2.7
1.3
0.8
23.2
6.1
8.8
3.1
115.8
Figure 1-5: Marginal Abatement Cost Curve for Top 5 Emitters in 2030
Non-CO2 Reduction (MtCO2e)
•Brazil
South Korea
•China
•Ukraine
•United States
IV-14
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
NITRIC AND ADIPIC ACID PRODUCTION
A majority of abatement potential spans over 30 nitric acid-producing countries, and only a small
fraction of abatement potential associated with adipic acid production is limited to adipic acid plants in
China, Ukraine, and India. Total reduction potential is achieved at break-even prices between $0 and
$50/tCO2e.
IV.1.4.3 Uncertainties and Limitations
This analysis leverages new data from public sources to improve on the facility-level detail used in
developing abatement project costs. In addition, we have incorporated a comprehensive international
inventory of current adipic acid production facilities. However, additional date and detail would
improve our abatement potential estimates.
• Abatement technology utilization rates: Active CDM and Joint Implementation (JI) abatement
projects in this sector have reported N2O reduction efficiencies and utilization rates significantly
higher than the default assumptions provided by the IPCC.
• Technology applicability across various nitric acid production processes and better
understanding of how cost for abatement measures would vary with each process.
• Improved estimates of regional changes in production over the next 20 to 30 years. For example
expected increases in Chinese adipic acid production capacity out to 2015, assuming no
abatement measures are installed would have significant impacts on both emission projections
and abatement potential in some countries.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-15
-------
NITRIC AND ADIPIC ACID PRODUCTION
The Chemical Company (CC). 2010a. "About Adipic Acid." Jamestown, RI: The Chemical Company.
The Chemical Company (CC). 2010b. "Adipic Acid Overview." Jamestown, RI: The Chemical Company.
Ecofys, Fraunhofer ISIR (Institute for Systems and Innovation Research), and the Oko-Institute. 2009.
Methodology for the Free Allocation of Emission Allowances in the EU ETS Post 2012 (pp. 6-20, and 42-48).
European Commission. Obtained at: http://ec.europa.eu/clima/policies/ets/benchmarking/docs/
bm study-chemicals en.pdf.
Intergovernmental Panel on Climate Change (IPCC). 2006. 2006 IPCC Guidelines for National Greenhouse
Gas Inventories: Volume 3 Industrial Processes and Product Use. Obtained at: http://www.ipcc-
nggip.iges.or.jp/public/2006gl/pdf/3 Volume3/V3 3 Ch3 Chemical Industry.pdf.
International Fertilizer Industry Association (IFA). 2011. Fertilizer Outlook 2011 - 2015. 79th IFA Annual
Conference. Montreal (Canada), 23-25 May 2011.
Global Industry Analysts Inc. (GIA) 2011. Global Adipic Acid Market to Reach 5.5 Billion Pounds by 2015,
According to a New Report by Global Industry Analysts, Inc. Obtained on March 11, 2011 at:
http://www.prweb.corn/releases/2011/l/prweb8043623.htm.
Heffer, P., & Prud'homme, M. (2011). Fertilizer Outlook 2011-2015. (A/11/69) Presented at the 79th IFA
Annual Conference, Montreal, Canada: International Fertilizer Industry Association (IFA). Obtained
on April 2, 2012 at: http://www.fertilizer.org/ifa/HomePage/LIBRARY/Publication-
database.html/Fertilizer-Outlook-2011-2015.html.
Kollmuss, A., and M. Lazarus. 2010. "Industrial N2O Projects Under the CDM: The Case of Nitric Acid
Production" (Working Paper No. WP-US-1007). Somerville, MA: Stockholm Environment Institute
(SEI). Obtained at: http://www.sei-international.org/mediamanager/documents/
Publications/Climate/sei-nitricacid-9nov2010.pdf.
Mainhardt, H. and D. Kruger. 2000. "N2O Emissions from Adipic Acid and Nitric Acid Production." Good
Practice and Uncertainty Management in National Greenhouse Gas Inventories. Montreal:
Intergovernmental Panel on Climate Change, National Greenhouse Gas Inventories Programme.
Obtained at: http://www.ipcc-nggip.iges.or.Jp/public/gp/bgp/3 2 Adipic Acid Nitric Acid
Production.pdf.
Organisation of Economic Cooperation and Development (OECD). 2004. SIDS Initial Assessment Report
For SIAM 18 Adipic Acid. Paris, France. Obtained at: http://www.inchem.org/
documents/sids/sids/124049.pdf.
Perez-Ramirez, J., Kapteijn, F., Schoffel, K., & Moulijn, J. A. 2003. "Formation and control of N2O in nitric
acid production: Where do we stand today?" Applied Catalysis B: Environmental, 44(2), 117-151.
doi:10.1016/S0926-3373(03)00026-2.
Schmidt, D.V. and M. von Velsen-Zerweck. 2010. "N2O emissions from nitric acid plants in CDM and JI
projects." Nitrogen+Syngas, 307,43 -47.
Schneider, L., M. Lazarus, and A. Kollmuss. 2010. "Industrial N2O Projects Under the CDM: Adipic
Acid—A Case of Carbon Leakage?" (Working Paper No. WP-US-1006). Somerville, MA: Stockholm
Environment Institute (SEI). Obtained at: http://ec.europa.eu/clima/consultations/0004/
unregistered/cdm watch 2 en.pdf.U.S. Environmental Protection Agency (USEPA). 1991. Alternative
Control Techniques Document—Nitric and Adipic Acid Manufacturing Plants. EPA-450/3-91-026. USEPA:
Research Triangle Park, NC. Obtained at: http://www.epa.gov/ttn/catc/dirl/nitric.pdf.
U.S. Environmental Protection Agency (USEPA). 2011. Inventory of U.S. Greenhouse Gas Emissions and
Sinks: 1990-2009. USEPA: Washington DC. As obtained on March 15, 2011. Obtained at:
http://epa.gov/climatechange/emissions/usinventoryreport.html.
U.S. Environmental Protection Agency (USEPA). 2012a. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. USEPA: Washington DC. Obtained at:
http://www.epa.gov/climatechange/EPAactivities/economics/nonco2projections.html.
IV-16 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
NITRIC AND ADIPIC ACID PRODUCTION
U.S. Environmental Protection Agency (USEPA). 2012b. Inventory of U.S. Greenhouse Gas Emissions and
Sinks: 1990-2010. USEPA: Washington DC. As obtained on March 19, 2012. Obtained at:
http://epa.gov/climatechange/ernissions/usinventoryreport.html.
Zhao, D. November 7, 2011. "China Adipic Acid Tumbles 36% on Weak Demand from PU Sector."
Market news article from ICIS.com. Obtained March 28, 2012, at:
http://www.icis.com/Articles/2011/ll/07/ 9505810/china-adipic-acid-tumbles-36-on-weak-demand-
from-pu-sector.html.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-17
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
IV.2. HFC Emissions from Refrigeration
and Air Conditioning
IV.2.1 Sector Summary
number of hydrofluorocarbons (HFCs) are used in refrigeration and air conditioning (AC)
systems and are emitted to the atmosphere during equipment operation, repair, and
disposal, unless recovered, recycled and ultimately destroyed. The most common HFCs
include HFC-134a, R-404A, R-410A, R-407C, and R-507A.1 In response to the ozone depleting substance
(ODS) phaseout, equipment is being retrofitted or replaced to use HFC-based substitutes or intermediate
substitutes (e.g., hydrochlorofluorocarbons [HCFCs]) that will eventually need to be replaced by HFCs or
other non-ozone-depleting alternatives. Greenhouse gas (GHG) emissions from the refrigeration/AC
sector (excluding chlorofluorocarbons (CFCs) and HCFCs) were estimated at roughly 349 million metric
tons of carbon dioxide equivalent (MtCO2e) in 2010. By 2020, emissions from this sector are expected to
reach 733 million MtCO2e, as shown in Figure 2-1. A majority of the growth will result from increased use
of HFCs in developing countries.
Figure 2-1: HFC Emissions from Refrigeration and AC: 2000-2030 (MtC02e)
1,596
Japan
I Russia
South Korea
I United States
I China
ROW
2000
2010 2020
Year
2030
Source: U.S. Environmental Protection Agency (USEPA), 2012.
This analysis reviews options to reduce emissions from the refrigeration/AC sector by using low-
global warming potential (GWP) refrigerants, low-emission technologies, and improved practices to
properly recover refrigerant at equipment servicing and disposal.
1 R-404A, R-410A, R-407C, and R-507A refrigerant blends are composed of HFCs. Specifically, R-404A is 44% by
weight HFC-125, 52% HFC-143a, and 4% HFC-134a,. R-410A is 50% HFC-32 and 50% HFC-134a. R-407C is 23% HFC-
32, 25% HFC-125, and 52% HFC-134a. R-507A (also called R-507) is 50% HFC-125 and 50% HFC-143a.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-19
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
Global abatement potential from refrigeration and AC in 2030 as evaluated in this analysis equates to
approximately 70% of total annual emissions from refrigeration and AC end-uses and 28% of total
emissions from ODS substitutes. In the near-term, abatement opportunities within refrigeration and AC
are partially limited because many of the abatement options identified apply only to newly manufactured
equipment and are thus limited by the turnover rate of the existing refrigeration and AC stock. In
addition, this analysis does not explore new equipment abatement options for all refrigeration and AC
equipment types, although such options may exist. Marginal abatement cost (MAC) curve results are
presented in Figure 2-2. Maximum abatement potential of the options in the refrigeration and AC sector
explored in this analysis is 540 MtCO2e in 2030. There are 317 MtCO2e of emissions reductions available
in 2030 that are cost-effective at currently estimated prices.
Figure 2-2: Global Abatement Potential in Refrigeration and AC: 2010, 2020, and 2030
$60
$50
$40
$30
« $20
O
u
$10
$0
-$10
-$20
-$30
•2010
•2020
•2030
1,000
Non-CO2 Reduction (MtCO2e)
IV.2.2 Emissions from Refrigeration and Air Conditioning
HFC emissions from refrigeration/AC occur during equipment manufacturing; as equipment is filled
with coolant; during use, as a result of component failure, leaks, and purges; during servicing; and at the
time of disposal, if the remaining refrigerant charge is not properly recovered. The use of refrigeration
and AC equipment also generates indirect emissions of GHGs (primarily carbon dioxide) from the
generation of power required to operate the equipment. HFC emissions can be reduced by adopting
alternative technologies (that either reduce HFC leakage or substitute the refrigerant for a low/no GWP
option) and by improving technician practices for equipment maintenance/servicing and disposal.
IV-20
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
For the purpose of this analysis, the sector is characterized by six major end-use types, presented in
descending order of 2020 GWP-weighted HFC emissions internationally (see Figure 2-3):
• residential and commercial AC, including window units and dehumidifiers, large and small unitary
air conditioners (including both ducted and non-ducted split systems), centrifugal and positive
displacement chillers, and packaged terminal air conditioners and packaged terminal heat pumps
(PTAC/PTHP), used to regulate the temperature and reduce humidity in homes, apartment
buildings, offices, hotels, shopping centers, and other large buildings, as well as in specialty
applications such as ships, submarines, nuclear power plants, and other industrial applications;
• retail food refrigeration, including small commercial refrigerators/freezers; medium-sized
condensing units found in convenience stores, restaurants, and other food service establishments;
and large systems found in supermarkets;
• motor vehicle air-conditioning (MVAC) used in cars, trucks, and buses;
• refrigerated transport, including refrigerated vans/trucks, containers, ship holds, truck trailers,
railway freight cars, and other shipping containers;
• industrial process refrigeration (IPR) and cold storage warehouses, including complex refrigeration
systems used in the food/beverage production, chemical, petrochemical, pharmaceutical, oil and
gas, metallurgical, and other industries as well as refrigeration systems used to cool meat,
produce, dairy products, and other perishable goods that are in storage; and
• household refrigerators and freezers used primarily in residential buildings.
Figure 2-3: Global HFC Emissions in 2020 by Application Type (% of GWP-Weighted Emissions)
IPR, Cold
Storage, 6%
Household
Refrigerators, 1%
Refrigerated
Transport, 8%
Small Retail
Food, <1%
Medium Retail
Food, 2%
Centrifugal
Chillers, 1%
Large Unitary
AC, 2%
Window Units &
Dehumidifiers, 1%
PTAC/PTHP, <1%
PD Chillers, <1%
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-21
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
For the purpose of this analysis, the sector considers nine facilities and/or applications, as defined
below.
• MVAC system used in a typical passenger car; for cost modeling purposes, this system is
characterized as having a charge size of 0.77 kg of R-134a.
• Large retail food refrigeration system used in a typical supermarket (assumed 60,000 sq. ft.); for
cost modeling purposes, this facility is characterized as having a charge size of 1,633 kg of R-
404A.
• Small retail food equipment (e.g., stand-alone systems) typically used in supermarkets and
convenience stores; for cost modeling purposes, this equipment is characterized as having a
charge size of 0.5 kg of 90% R-134a and 10% R-404A (based on the average HFC refrigerant types
currently installed in the U.S. market).
• Window AC unit or dehumidifier; for cost modeling purposes, this unit is characterized as
having a charge size of 0.4 kg of R-410A.
• Unitary AC system or PTAC/PTHP; for cost modeling purposes, this system is characterized as
having a charge size of 8 kg of R-410A.
• Positive displacement chiller (i.e., screw or scroll chiller); for cost modeling purposes, this
equipment is characterized as having a charge size of 270 kg of 33% R-410A, 33% R-407C, and
33% R-134a (based on the average HFC refrigerant types currently installed in the U.S. market).
• IPR/cold storage system;2 for cost modeling purposes, this system is characterized as having a
charge size of 2,000 kg using 25% R-507A, 25% R-404A, 25% R-134a, and 25% R-410A (based on
the average HFC refrigerant types currently installed in the U.S. market).
• Typical auto disposal yard using a recovery device to recover refrigerant from MVACs; for cost
modeling purposes, this facility is characterized as recovering refrigerant from about 425 MVACs
per year (with an average of 0.13 kg of R-134a recoverable per MVAC).
• Typical auto service shop using a recovery/recycling device to service MVACs; for cost modeling
purposes, this facility is characterized as recovering refrigerant from about 150 MVACs per year
(with an average 0.29 kg of R-134a recoverable per MVAC).
For modeling purposes, data typical for U.S. systems/equipment are used. Certain cost assumptions,
such as labor rates, energy prices and capital costs, are adjusted for other regions.3 Otherwise, it is
assumed that the costs and reductions achieved in the United States can be scaled and are representative
of the costs and reductions in other regions.
IV.2.2.1 Activity Data or Important Sectoral or Regional Trends
Refrigeration/AC consumption, which is estimated using USEPA's Vintaging Model for the United
States, is used to represent activity data. Consumption is assumed to scale with country gross domestic
product (GDP). Regional differences are applied to other developed countries in the European Union
2 Abatement options for these types of equipment apply to only those facilities using HFCs. Many such facilities
currently use ammonia and hence are not evaluated for further emission reductions in this analysis.
3 In developing countries, it is assumed that capital costs are 10% higher, fuel prices are 30% higher, electricity costs
are 66% higher, and labor costs are 20% lower than those relative to the United States.
IV-22 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
(EU) to reflect higher consumption of low-GWP refrigerants in new passenger MVAC systems, domestic
refrigerators, and large supermarket systems. For example, HCs have begun to penetrate the EU market
in vending machines and other small retail food equipment (e.g., reach-in cases). Adjustments were also
made to account for differences in the rates of ODS phaseout relative to the U.S. substitution scenarios.
IV.2.2.2
Emission Estimates and Related Assumptions
Global emissions from refrigeration/AC were estimated at 349 MtCO2 in 2010, projected to grow to
733 MtCO2 by 2020 and 1,596 MtCO2 by 2030. Table 2-1 presents the projected emissions from
refrigeration/AC between 2010 and 2030. Growth in emissions is driven largely by GDP. Globally, HFC
emissions from refrigeration/AC have been growing also because of the phaseout of ODS under the
Montreal Protocol. Growth has also been driven by increased demand for air conditioning equipment,
especially in emerging economies. Because of regulations associated with HFC-based refrigerants and/ or
growing public pressure to transition away from such refrigerants, many developed countries have
transitioned/are transitioning from ODS to natural refrigerants or other low-GWP alternatives in some
end-uses. More detail on how HFC consumption and reduction potential of options are modeled is
contained in Appendix D to this chapter.
Table 2-1: Projected Baseline Emissions from Refrigeration and AC: 2010 to 2030 (MtCC^e)
Country/Region
2010
2015
2020
2025
2030
CAGR*
(2010-2030)
Top 5 Emitting Countries
China
United States
South Korea
Russia
Japan
45.8
114.0
17.0
16.0
30.9
82.7
162.1
24.5
23.2
38.7
151.2
213.9
35.7
33.4
48.0
328.8
282.5
61.1
56.0
62.0
534.1
316.7
79.7
72.0
66.7
13.1%
5.2%
8.0%
7.8%
3.9%
Rest of Regions
Africa
Central and South America
Middle East
Europe
Eurasia
Asia
North America
World Total
12.5
16.2
13.3
46.3
2.0
20.6
14.8
349.3
18.7
23.4
19.8
59.3
2.9
32.2
20.4
507.9
28.1
34.3
29.6
74.6
4.2
52.7
27.4
733.1
49.3
59.5
52.3
99.9
7.0
101.3
40.4
1,199.9
64.9
78.2
69.9
109.7
9.0
146.9
48.1
1,596.1
8.6%
8.2%
8.6%
4.4%
7.9%
10.3%
6.1%
7.9%
a CAGR = Compound Annual Growth Rate.
Source: U.S. Environmental Protection Agency (USEPA), 2012.
IV.2.3 Abatement Measures and Engineering Cost Analysis
For the purpose of evaluating the cost of reducing HFC emissions from the refrigeration/AC sector,
this analysis considers reduction costs for applying 14 new technologies and using three types of
improved technician practices. Table 2-2 summarizes the technology and practice options reviewed and
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-23
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
the types of equipment that are assumed to adopt such options, and the associated system type
definitions (i.e., the equipment characteristics assumed in order to develop the option costs).
Table 2-2: Refrigeration and AC Abatement Options
Enhanced HFC-134a in MVACs
HFO-1234yf in MVACs
Enhanced HFO-1234yf in MVACs
Distributed systems in large retail food
HFC secondary loop and/or cascade systems in
large retail food
NHs or HC secondary loop and/or cascade
systems in large retail food
C02 Transcritical systems in large retail food
Retrofits of R-404A systems in large retail food
HCs in small retail food systems
HC in window units and dehumidifiers
R-32 in unitary AC and PTAC/PTHP
MCHX in small and medium AC systems
R-32 with MCHX in unitary AC
MCHX in large AC systems
NHs or C02 in large refrigeration systems
Refrigerant recovery at disposal
Refrigerant recovery at servicing
Leak repair
Reduction
Efficiency
50%
99.7%
99.8%
80%
94.6%
100%
100%
46%
100%
100%
75%
37.5%
84.5%
37.5%
100%
85%
95%
40%
Applicability
New MVACs in light-duty vehicles
New MVACs in light-duty vehicles
New MVACs in light-duty vehicles
New large retail food refrigeration systems
New large retail food refrigeration systems
New large retail food refrigeration systems
New large retail food refrigeration systems
Existing large retail food refrigeration systems
New small retail food refrigeration systems
New window AC units and dehumidifiers
New unitary AC equipment and PTAC/PTHP
New unitary AC equipment
New unitary AC equipment
New positive displacement chillers
New I PR and cold storage systems
All existing refrigeration/AC reaching disposal
All small equipment (i.e., MVACs, small unitary AC)
All existing large equipment (i.e., large retail food, IPR,
cold storage, and chillers)
A number of these technology options have already begun penetrating certain markets, particularly
in the EU, Japan, and several other developed countries. For example, using HCs in small retail food
equipment is increasingly common for new equipment sold in the EU and Japanese markets. Likewise,
use of alternative refrigerants in passenger MVAC systems has begun in the EU in response to Directive
2006/40/EC (the MAC Directive), which requires replacing HFC-134a with a refrigerant having a GWP of
less than 150 in new model vehicles beginning in 2011 and in all new vehicles by 2017.4 Alternative
refrigerants are also increasingly being used in large supermarkets across Northern Europe and to a
smaller extent in the United States, Canada, Australia, and other developed countries. The options of
increased refrigerant recovery at service and disposal, as well as more rigorous leak repair for large
4 Due to supply problems of the refrigerant originally chosen by the MVAC industry (i.e., HFO-1234yf), the EU
granted a 2-year dispensation to the auto industry. Additionally, some automobile OEMs have recently announced
that they plan to continue to use HFC-134a refrigerant while they further investigate low-GWP options (EurActiv,
2013). Other OEMs have stated that they have not changed their plans to introduce HFO-1234yf and in fact some of
these systems are already in operation today (RAC Magazine, 2013).
IV-24
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
equipment, can bring significant emission savings, especially in developing countries, where they are less
likely to be practiced in the baseline.
Each of the abatement options is described further in the sections below. Additional options
considered but not yet included in the analysis are described in Sections IV.2.2.18 through IV.2.2.20.
Several other options, not mentioned below, are also potentially available but have not been included in
this analysis. All costs are presented in 2010 dollars based on the Consumer Price Index (U.S. Department
of Labor, 2011).
IV.2.3.1 Enhanced HFC-134a in New MVACs
This option reduces annual leak rates of HFC-134a MVAC systems by 50% through better system
components, including improved system sealing, lower permeation hoses, improved fittings, and higher
evaporator temperatures (USEPA and NHTSA, 2011). Enhanced HFC-134a systems are additionally
assumed to reduce fuel consumption by an estimated 42% through improved component efficiency,
improved refrigerant cycle controls, and reduced reheat of the cooled air (USEPA and NHTSA, 2011).
This option is applicable to a newly manufactured MVAC system in light duty vehicles5 in all
countries except the EU, where it is assumed to penetrate in the baseline. The one-time cost is estimated
at roughly $73 per MVAC system—assumed to be 10% greater in developing countries. These costs are
offset by annual savings that result from reduced fuel and refrigerant consumption (a combined savings
of approximately $38 per system). In developing countries, fuel prices are assumed to be 30% greater,
resulting in a combined savings of approximately $48 per system.
IV.2.3.2 HFO-1234yf in New MVACs
HFO-1234yf has a GWP of only four and performs similarly to HFC-134a, making the use of current
MVAC system designs with minimal changes feasible. HFO-1234yf is, however, slightly flammable
(designated 2L flammability in Addendum h to ANSI-ASHRAE Standard 34-2010), which may
necessitate certain safety mitigation strategies. This option has already begun penetrating the EU and U.S.
markets in a couple models (Refrigeration and Air Conditioning Magazine, 2013). This option is assumed to
be as efficient as conventional HFC-134a MVAC systems (Oko-Recherche et al., 2011; Koban, 2009).
This option is applicable to a newly manufactured MVAC system in a light-duty vehicle in all
countries except the EU, where it is assumed to penetrate the baseline market. The one-time capital cost is
estimated at approximately $59 per MVAC system, resulting from incremental refrigerant costs and
hardware changes the latter of which is assumed to be 10% greater in developing countries. Annual costs
are assumed to be approximately $8 per system associated with incremental refrigerant replacement
costs.
IV.2.3.3 Enhanced HFO-1234yf in New MVACs
As a newly developed technology, HFO-1234yf MVAC systems may cost more than those currently
containing HFC-134a. Further, a lower global production of the chemical, combined with the additional
processes needed to produce it, is expected to lead to an initially high price for the chemical, but this may
decrease as production increases. Similar to the Enhanced HFC-134a option, this option explores HFO-
' This category includes cars, pick-up trucks, minivans, and sport utility vehicles.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-25
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
1234yf for use in new MVAC systems using improved system components to allow for 50% reduced
refrigerant leakage and efficiency improvements of 42%.
This option is applied to a newly manufactured MVAC system in light-duty vehicles in all countries.
The one-time cost is estimated at roughly $100 per MVAC system, resulting from incremental refrigerant
costs and better system components—the latter of which is assumed to be 10% greater in developing
countries. The assumed incremental cost of the refrigerant is lower for this option than the original HFO-
1234yf option on the basis that, over time with mass production of the chemical and systems to use it, the
price will drop. Given this as well as the assumed lower leak rate of this option, annual costs are assumed
to be only approximately $2 per MVAC system. These costs are offset by annual savings that result from
reduced fuel consumption, equaling roughly $37 per system. In developing countries, fuel prices are
assumed to be 30% greater, resulting in a savings of almost $48 per system.
IV.2.3.4 Distributed Systems in New Large Retail Food
A distributed system consists of multiple compressors that are distributed throughout a retail food
store (e.g., a supermarket), near the display cases they serve, and are connected by a water loop to a
chiller or other type of equipment that rejects heat (e.g., a cooling tower) that is located on the roof or
elsewhere outside the store. Because distributed systems have smaller refrigeration units distributed
among the refrigerated and frozen food display cases, refrigerant charges for distributed systems can be
smaller than the refrigerant charge used in a comparable traditional centralized direct expansion (DX)
system. The reduction in original charge size of the system will reduce HFC consumption (at first fill) and
reduce potential emissions at the end of the equipment's life. Additionally, because of the placement of
the units, a distributed system can require less refrigerant tubing and fittings than DX systems, thereby
reducing total HFC leaks during the useful lifetime of the equipment to an estimated 80% relative to
conventional systems. However, distributed systems are estimated to be 5% less efficient than
conventional HFC centralized DX systems (IPCC, 2005). This technology is already being implemented
today in many developed countries.
This abatement option is applied to a newly manufactured large retail food system in a large
supermarket. In developed countries, one-time costs are estimated to be 5% more expensive than
conventional HFC centralized DX systems (IPCC, 2005), equivalent to an incremental cost of about $9,100
per supermarket; these costs are estimated to be 10% greater in developing countries. Annual costs are
estimated at nearly $3,700 per supermarket in developed countries due to higher electricity consumption;
these annual electricity costs are estimated to be 66% greater in developing countries. At the same time,
annual refrigerant savings (due to reduced leakage) are realized, totaling nearly $1,800 per supermarket.
IV.2.3.5 HFC Secondary Loop and/or Cascade Systems in New Large Retail Food
Secondary loop systems use two fluids: a primary refrigerant and a secondary fluid. The secondary
fluid is cooled by the primary refrigerant in the machine room and then pumped throughout the store to
remove heat from the display equipment. In supermarkets, secondary loop systems are also sometimes
used in combination with a cascade system. Cascade designs consist of two independent refrigeration
systems that share a common cascade heat exchanger. The heat exchanger acts as the low temperature
refrigerant condenser and serves as the high temperature refrigerant evaporator. Each component of a
cascade design uses a different refrigerant that is most suitable for the given temperature range, with CO2
commonly used in the low temperature circuit and an HFC used as the refrigerant at the medium
temperature phase (RTOC, 2011). Because the HFC refrigerant is contained in the machine room in a
secondary loop system and is not required for use in the low temperature circuit of a cascade design,
these systems require a significantly lower refrigerant charge and have lower leakage rates, resulting in
IV-26 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
approximately 90% less annual leakage. While historically these systems were less efficient than
conventional DX systems, today these systems are found to be just as efficient as conventional DX
systems, if not more so, due to simplified piping, newly designed circulating pumps, and fewer
components (Wang et al., 2010; DelVentura, et al., 2007; SuperValu, 2012; WalMart, 2006; Hinde, Zha, and
Lan, 2009).
This abatement option is applied to a newly manufactured large retail food system in a large
supermarket. The one-time cost in developed countries is estimated to be 17.5% more expensive than
conventional DX systems (IPCC, 2005), equivalent to an incremental cost of nearly $32,000 per
supermarket; this capital cost is estimated to be 10% greater in developing countries. Annual savings
associated with reduced refrigerant leakage are estimated to equal almost $2,000 per supermarket. These
systems are assumed to be equally as efficient as DX systems, so no costs or savings are associated with
annual energy consumption.
IV.2.3.6 NH3or HCs Secondary Loop and/or Cascade Systems in New Large Retail
Food
Similar to the HFC secondary loop and/or cascade option, in this system a secondary fluid is cooled
by a primary refrigerant in the machine room and then pumped throughout the store to remove heat
from the display equipment. In some cases, the secondary loop system is also used in combination with a
cascade design, which does not rely on any HFCs in the low temperature circuit. For this abatement
option, the primary refrigerant is assumed to be ammonia (NHs) or HCs, which have a GWP that is
negligible. Ammonia is not used in conventional supermarket refrigeration systems because such a
system could expose consumers to toxic and slightly flammable refrigerant. Similarly, HCs are not used
due to their high flammability. However, using a secondary loop allows the primary refrigerant to be
isolated to a mechanical room with controlled access to only those with specific training. Because
ammonia/HC secondary loop systems avoid running the primary refrigerant through miles of piping to
and from food storage cases, they have lower leakage rates than conventional DX systems and operate at
reduced charges. In addition, these systems are conservatively assumed to be 5% more energy efficient
than conventional DX systems, with some supermarkets reporting actual efficiency gains of 0.5% to 35%
(Wang et al., 2010; SuperValu, 2012; Hydrocarbonconversions.com, 2011).
This abatement option is applied to a newly manufactured large retail food system in a large
supermarket. The one-time cost in developed countries is estimated to be 25% greater than conventional
DX systems (IPCC, 2005), equivalent to an incremental cost of roughly $45,600 per supermarket; this
capital cost is estimated to be 10% greater in developing countries. Annual savings are estimated at
roughly $5,900 due to both reduced energy consumption and refrigerant savings (due to avoided HFC
refrigerant leaks). In developing countries, where electricity rates are assumed to be higher, annual
savings are assumed to total more than $8,300.
IV.2.3.7 CO2 Transcritical Systems in New Large Retail Food
This option eliminates the use of HFCs in large retail food refrigeration systems through the use of
CO2 as the primary refrigerant in a transcritical cycle. CO2 transcritical systems are similar to traditional
centralized DX designs but must operate at high pressures to accommodate the low critical temperature
of CC>2 (GTZ Proklima, 2008). As a result, special controls and component specifications must be
incorporated into the system design, which often result in higher upfront costs (USEIA, 2012).
Additionally, CO2 transcritical systems operate most efficiently in cooler climates, performing an
estimated 5% to 10% more efficiently than DX systems using an HFC refrigerant in regions with an
average annual temperature below 50°F (Supermarket News, 2012). At the same time, due to a possible
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-27
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
energy penalty, the use of CO2 transcritical systems in warmer climates is currently considered less
viable. Specifically, use of these systems is most widely accepted in areas where the maximum ambient
temperature is frequently below 88°F (ACHR News, 2010), or where the average annual temperature is
lower than 59°F (Hill PHOENIX, 2012). Today, it is estimated that over 1,300 CO2 transcritical systems are
currently in operation in Europe with installations as far south as Italy and Spain in addition to a handful
of systems that have been installed in Canada (Shecco, 2012). Plans are also currently underway to deploy
the technology in supermarkets in the United States.
This abatement option is assumed to be applied to a newly manufactured large retail food system in
large supermarkets in cooler climates. One-time costs in developed countries are estimated to be 17.5%
more expensive than conventional HFC centralized DX systems (Australian Green Cooling Council, 2008;
R744.com, 2012), equivalent to an incremental cost of nearly $32,000 per supermarket; these capital costs
are estimated to be 10% greater in developing countries. Annual savings are estimated at about $5,900 per
supermarket, which result from both refrigerant savings (due to avoided HFC refrigerant leaks) that total
approximately $2,200 per supermarket, and energy savings (due to increased efficiency), which total
approximately $3,700 per supermarket. In developing countries, where electricity rates are assumed to be
higher, annual savings are assumed to total more than $8,300.
IV.2.3.8 Retrofits of R-404A in Large Retail Food
Retail food refrigeration systems containing R-404A, which have high charge sizes and annual leak
rates, can be retrofitted with lower-GWP refrigerants, such as R-407A (with a GWP of 1,770), to reduce
their annual climate impact. While some system retrofits will require little to no change to achieve the
desired operational characteristics, others may need additional modifications, such as changing the
orifice or TXV size to achieve the same efficiency. If proper system evaluation is performed and
considerations are taken to ensure the continued reliability of the system, retrofitting can lead to system
improvements as a result of recommissioning the equipment (e.g., due to properly setting up controls
and system operating valves, which may have wandered from set-point due to lack of maintenance)
(ACHR News, 2012). However, because such changes may have occurred during remodeling, when the
refrigerant retrofit is assumed to occur, no change in energy efficiency due to the new refrigerant is
assumed.
To perform a system retrofit, the entire system must be shut down and checks should be made for
leaks throughout the system. Solenoid sealings must then be changed as well as the filter driers. After the
entire system is evacuated, the replacement fluid is deposited into the system.
For cost modeling purposes, it is assumed that retrofits are performed on large retail food systems at
about half-way through its useful lifetime (i.e., 7 years) at the same time retrofits or remodeling of cases
are underway. The original R-404A charge size of 1,633 kg is assumed to be replaced with an equivalent
amount of R-407A. The procedure is assumed to require 10 hours of a service technician's time (5 hours
for the medium temperature system and 5 hours for the low temperature system), all of which is assumed
to cost a total of roughly $500 in developed countries and $100 in developing countries (based on
technician labor rates). Since the composition of R-404A and R-407A are similar, the cost of the refrigerant
is assumed to be the same. Therefore, no annual costs or savings are assumed for this option.
IV.2.3.9 HCs in New Small Retail Food Refrigeration Systems
For small retail food equipment, this option explores the replacement of HFC-134a and R-404A with
HCs. HCs, such as butane and propane, have negligible GWPs. Although safety issues associated with
HC use in small equipment previously presented a barrier to their use, these issues can be addressed,
making them a viable alternative to HFCs. International standards exist to evaluate and mitigate such
IV-28 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
safety issues. For example, although R-290 (propane) is flammable, it has been successfully implemented
in some self-contained hermetic systems.
This option is applied to a newly manufactured small retail food refrigeration system (e.g., stand-
alone equipment). No one-time costs are estimated because this option is cost neutral (Unilever, 2008). An
annual savings of less than $1 per system is estimated to result from avoided HFC refrigerant costs.
IV.2.3.10 HCs in New Window AC and Dehumidifiers
R-410A is widely used in window AC units and dehumidifiers, brought about by regulations phasing
out HCFC-22, which was previously used. By replacing R-410A with HCs, such as propane (R-290),
which is assumed to have a negligible GWP, significant emissions can be avoided. Two Chinese AC
manufacturers have already commercialized room AC units using R-290. The R-290 AC designs achieve
lower refrigerant charge sizes than is currently required by international standard (IEC 60335-2-240) and
include additional safety features, such as a special compressor design and refrigerant leak alarm system.
Mass production of the R-290 units started in 2009, for initial sale in Europe and China (GTZ-Proklima,
2009).
For cost modeling purposes, this option is applied to newly manufactured window AC units and
dehumidifiers. This option is conservatively assumed to have no one-time costs even though there is
indication that R-290 AC units can be produced more cheaply than R-410A units as a result of the better
heat transfer properties and lower pressure drop of R-290, which allows for the use of narrower tubes in
the condenser and evaporator (GTZ-Proklima, 2009). Annual savings are estimated based on the reduced
cost of HC refrigerant replacement compared with R-410A—estimated to result in a savings of
approximately $0.33 per unit.
IV.2.3.11 R-32 in New Unitary AC Equipment and PTAC/PTHP
In this option, R-32,6 a mildly flammable (category 2L in ANSI/ASHRAE Standard 34-2010)
refrigerant with a GWP of 650, is used in new unitary AC equipment and PTAC/PTHP to replace R-410A,
which has a GWP of 1,725. R-32 performs with a reduced charge volume ratio of 66% compared to
R-410A (Xu et al., 2012). This reduced charge volume results in a 75% reduction of the direct global
warming impact compared to the R-410A system. It is also reportedly 2% to 3% more energy efficient
than R-410A (Pham and Sachs, 2010). The equipment used also has the potential to be downsized by up
to 15%, which can decrease one-time costs by reducing the amount of materials used. R-32 AC products
are already available in Japan and are to be launched in India in February 2013 (Daikin, 2012; Stanga,
2012). Manufacturers in Algeria, China, Thailand, and Indonesia also plan to transition to R-32 AC
systems (Stanga, 2012).
For cost modeling purposes, this option is applied to a newly manufactured unitary AC system (e.g.,
residential, small commercial and large commercial unitary AC) and PTAC/PTHP. The option is
conservatively assumed to result in a one-time cost savings of approximately $30 per system, due to the
reduced quantity of refrigerant required and lower cost of the alternative refrigerant. Additional savings
may be realized through reduced material costs; however, there may also be costs associated with
6 R-32 is chosen here due to the availability of data. Other options are under development or being applied. For
example, both Godrej in India and Gree in China are producing units with HC refrigerants (Godrej, 2012; Gree, 2012).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-29
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
designing a system to use a mildly flammable refrigerant. Annual savings are associated with reduced
refrigerant replacement costs, estimated at approximately $2.6 per system. Annual energy savings are
also likely to be associated with this option but are not quantified in this analysis.
IV.2.3.12 MCHX in New Unitary AC Equipment
This option explores the use of microchannel heat exchangers (MCHX) in unitary AC equipment
using R-410A. MCHXs are a modification of conventional heat exchangers, which transfer heat in AC and
refrigeration systems (e.g., for the rejection of heat from indoor cooled spaces to the outside ambient
space). Because MCHXs transfer heat through a series of small tubes instead of a single or multiple large-
diameter tubes, systems using them require between 35% and 40% less refrigerant to operate than those
using conventional heat exchangers. Likewise, if average leak rates remain the same,7 the actual amount
of refrigerant emitted would be less. In addition, MCHX systems perform better and are more energy
efficient than conventional systems. They also require smaller components, which results in reduced
quantities of metals and other materials required per unit, although potential savings in material costs are
not analyzed here. MCHXs are already used widely by multiple manufacturers in the automotive
industry and in certain models of screw and scroll chillers.
For cost modeling purposes, this option is applied to a newly manufactured unitary AC system (e.g.,
residential, small commercial, and large commercial unitary AC). One-time savings equal to roughly $27
are assumed due to the smaller refrigerant charge. No annual costs are assumed for this option. The
annual savings associated with avoided refrigerant losses is estimated at approximately $2.30 per system.
IV.2.3.13 R-32 with MCHX in New Unitary AC Equipment
Similar to the option described above, this option explores the use of MCHX in unitary AC
equipment but with R-32 refrigerant (with a GWP of 650) in place of R-410A (with a GWP of 1,725). The
use of the MCHX results in a refrigerant charge reduction of between 35% and 40% compared with
conventional heat exchangers, while the use of R-32 refrigerant allows a further charge size reduction of
66% compared with R-410A. Combined, the reduced charge volume and GWP lead to a lower global
warming impact of approximately 85%.
For cost modeling purposes, this option is applied to a newly manufactured unitary AC system (e.g.,
residential, small commercial, and large commercial unitary AC). One-time savings equal to roughly $46
are assumed as a result of the smaller refrigerant charge and lower cost of the alternative refrigerant. No
annual costs are assumed for this option. The annual savings associated with avoided refrigerant losses is
estimated at approximately $3.90 per system.
IV.2.3.14 MCHX in New Positive Displacement Chillers
This option is assumed to be applicable in screw and scroll chillers. As explained above, MCHXs are
a modification of conventional heat exchangers, which transfer heat in AC and refrigeration systems (e.g.,
for the rejection of heat from indoor cooled spaces to the outside ambient space). Because MCHXs
transfer heat through a series of small tubes instead of a single or multiple large-diameter tubes, systems
7 For example, if average leak rates are dominated by failures or service errors that lead to a catastrophic (100%) loss,
and the MCHX system has the same reliability, then average leak rates would be the same.
IV-30 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
using them require between 35% and 40% less refrigerant than those using conventional heat exchangers.
In addition, MCHX systems perform better and may be more energy efficient than conventional systems.
MCHXs are already used widely by multiple manufacturers in the automotive industry and in certain
models of screw and scroll chillers.
For cost modeling purposes, this option is applied to a newly manufactured positive displacement
chiller. One-time savings equal to nearly $900 are assumed due to the smaller refrigerant charge. No
annual costs are assumed for this option. The annual savings associated with avoided refrigerant losses is
estimated at approximately $50 per chiller.
IV.2.3.15 NH3 or CO2in New IPR and Cold Storage Systems
This abatement option is assumed to be applicable to cold storage and industrial process refrigeration
systems. Although NHs refrigeration systems are already common in refrigerated spaces over 200,000 sq.
ft., additional penetration of NHs systems is possible in facilities that are less than 200,000 sq. ft. but
greater than 50,000 sq. ft. In addition, modern NHs absorption refrigeration units are compact,
lightweight, efficient, economical, and safe, which has made more applications possible. Improved
technologies have also expanded the technical feasibility of using CO2 systems. CO2 systems are being
used in low-temperature refrigeration (-30°C to -56°C), while ammonia/CO2 systems can be used for
higher temperature refrigeration (-35°C to -54°C). The lower temperature for both systems is limited
primarily by the -56°C triple point of CO2 being used on the low side. The choice between these systems
is primarily due to outdoor temperatures; in colder climates, a CO2 system is both energy efficient and
simpler, while in hotter climates a cascade system may be needed to maintain energy efficiency. In
Europe and the United States, storage and production facilities have been built with ammonia/CO2
cascade systems. These systems are estimated to be 2% to 20% more energy efficient compared with their
HFC counterparts (Gooseff and Horton, 2008).
For cost modeling purposes, this option is applied to a newly constructed IPR/cold storage
refrigeration system/facility. The incremental one-time cost is estimated at approximately $210,700 per
system in developed countries (Gooseff and Horton, 2008), assumed to be 10% more in developing
countries. The annual savings of approximately $50,300 per system is associated with lower refrigerant
replacement costs and reduced energy consumption of 11%; annual electricity cost savings are assumed
to be 66% greater for developing countries, resulting in annual savings of approximately $83,100.
IV.2.3.16 Refrigerant Recovery at Disposal for All Existing Equipment Types
Some level of refrigerant recovery at equipment disposal already occurs in the baseline of developed
and developing countries, because it is illegal to vent HFCs when equipment is discarded in the U.S. and
elsewhere. However, this option explores more widespread, thorough efforts to recover refrigerant at
disposal across all equipment types. The approach involves using a refrigerant recovery device that
transfers refrigerant into an external storage container prior to disposal of the equipment. Once the
recovery process is complete, the refrigerant contained in the storage container may be cleaned by using
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-31
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
recycling devices, sent to a reclamation facility to be purified,8 or destroyed using approved technologies
(e.g., incineration).
For cost modeling purposes, this option is applied to an auto dismantling facility assumed to use a
single refrigerant recovery device that meets SAE J2788 standards to perform MVAC recovery jobs. The
incremental one-time cost is estimated at approximately $2,025 per facility for the purchase of a
refrigerant recovery device in developed countries (ICF, 2008); this cost is estimated to be 10% greater in
developing countries. Annual costs are estimated at roughly $1,100 per auto dismantling facility for
technician labor time and the purchase of new filters for the recovery device (ICF, 2008). In developing
countries, technician labor costs are assumed to be one-fifth the cost of that in developed countries;
therefore, annual costs are assumed to be about $240. The annual savings is estimated at about $440 per
auto dismantling facility, based on the value of the recovered refrigerant for reclamation/reuse.
IV.2.3.17 Refrigerant Recovery at Servicing for Existing Small Equipment
Similar to disposal recovery, this option assumes more widespread and thorough refrigerant
recovery practices while servicing HFC refrigeration/AC systems. Because it is assumed that significant
refrigerant is already recovered during servicing of large equipment, this abatement option is only
applied to MVAC and small unitary AC systems.
For cost modeling purposes, this option is applied to a U.S. auto servicing facility assumed to
perform MVAC servicing jobs using a recovery/recycling (recharge) device designed to meet the SAE
J2788 standard. The incremental one-time cost is estimated at approximately $4,050 per servicing facility
for the purchase of a refrigerant recovery device in developed countries (ICF, 2008); this cost is estimated
to be 10% greater in developing countries. The annual cost is estimated at roughly $870 per auto servicing
facility in developed countries for technician labor time and the purchase of new filters for the recovery
device (ICF, 2008); in developing countries, technician labor costs are assumed to be one-fifth the cost of
that in developed countries; therefore, the annual cost is assumed to be nearly $194. The annual savings is
estimated at roughly $350 per auto servicing facility, based on the value of the recovered refrigerant for
reclamation/reuse.
IV.2.3.18 Leak Repair for Existing Large Equipment
This abatement option is assumed to be applicable to large retail food, cold storage and industrial
refrigeration, and positive displacement chiller systems. Some level of leak repair activity is already
practiced in the baseline, but this option explores additional efforts to repair leaks.
For cost modeling purposes, this option is applied to large supermarkets requiring significant small
repairs (e.g., maintenance of the purge system or replacement of a gasket or O-ring). A one-time cost of
approximately $1,870 is estimated per supermarket for parts and labor needed to perform the repair in
developed countries (USEPA, 1998); in developing countries, this cost is estimated to be 10% greater. The
annual savings associated with avoided refrigerant replacement is estimated at $1,470 per supermarket.
8 Recycling cleans and reclamation purifies recovered refrigerant; reclamation is more thorough and involves
repeated precision distillation, filtering, and contaminant removal. Recycling is used for on-site servicing of MVACs
and other equipment, whereas reclamation requires sending the refrigerant off-site to a reclaimer.
IV-32 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
IV.2.3.19 HCs in New Domestic Refrigeration Systems
HFC-134a may be replaced with HCs in household refrigerators. HCs, such as butane and propane,
have very low GWPs of 4.0 and 3.3, respectively. The main disadvantage of HCs is that they are
flammable, but engineering design changes and safety features in manufacturing plants have been
successfully implemented to overcome these challenges. In 2009, roughly 40% of new household
refrigerators/freezers produced globally contained HCs, with more than 400 million HC units in use
worldwide (Greenpeace, 2009). This option is not quantitatively assessed in this analysis but will be
considered for future updates.
IV.2.3.20 CO2 in Transport Refrigeration
CO2 is currently being used in trucks in cryogenic (open-loop) systems and is also viable for use in
refrigerated ships and intermodal containers using a compressor system (Carrier, 2013; Environmental
Leader, 2010). However, more sophisticated refrigeration cycles are needed for CO2 systems to match the
efficiency of equivalent HFC units under high ambient temperature operation. The cycle operation is
often transcritical, which results in compressor discharge pressures up to five times higher than
conventional HFC systems. Therefore, entirely new parts, design approaches, test procedures, service
training, etc. are needed to design, build and operate a trans-critical CO2 system (TEAP, 2011). Due to a
lack of readily available cost information on this alternative, this option is not quantitatively assessed in
this analysis.
IV.2.3.21 Low-GWP Refrigerants and Blends
Significant research and development (R&D) efforts are underway to identify feasible alternatives for
high-GWP HFC refrigerants in multiple subsectors. For instance, the U.S. EPA's Significant New
Alternatives Policy (SNAP) program has found - 1233zd(E) (also called trans-l-chloro-3,3,3-trifluoroprop-
1-ene) and HFO-1234ze(E) acceptable for use in chillers, used mainly for comfort AC in large residential
and commercial buildings, including facilities with data processing and communication centers. Potential
alternatives in numerous refrigeration and AC uses include ammonia, hydrocarbons, CO2, water, HFC-
32, and new low-GWP refrigerants such as HFO-1234yf, HFO-1234ze, -1233zd(E), and blends containing
HFOs. The Air-Conditioning, Heating and Refrigeration Institute (AHRI) launched the Low-GWP
Alternative Refrigerants Evaluation Program to evaluate low-GWP alternatives to HCFC-22, HFC-134a,
R-404A, R-407C, and R-410A in various product types (AHRI, 2013). Some 40 chemicals were identified,
including refrigerants or blends with no, low, and high flammability. The GWPs of these products also
varied, from zero to about 1,300. Some of these blends are under intense development and testing and are
most commonly known by trade names, including DR-5, DR-7, DR-33, L-20, L-40, L-41, N-13, N-40, and
XP-10. Because of a lack of readily available cost information on this alternative, these options are not
quantitatively assessed in this analysis.
IV.2.4 Engineering Cost Data Summary
Table 2-3 presents the engineering cost data for each mitigation option outlined above, including all
cost parameters.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-33
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
Table 2-3: Engineering Cost Data on a Facility Basis
Abatement Option/Facility Type
Capital
Project Cost
Lifetime (2010
(years) USD)
Annual Annual Abatement
Revenue O&M Costs Amount
(2010 USD) (2010 USD) (tC02e)a
Enhanced HFC-134a
MVAC— U.S./Other Developed, New
MVAC-EU, New
MVAC— Developing, New
12.0
12.0
12.0
73.2
73.2
80.5
37.9
37.9
48.1
- 0.1
- 0.1
- 0.1
HFO-1234yf
MVAC— U.S./Other Developed, New
MVAC-EU, New
MVAC— Developing, New
12.0
12.0
12.0
59.1
59.1
60.4
—
—
—
7.9 0.2
7.9 0.2
7.9 0.2
Enhanced HFO-1234yf
MVAC— U.S./Other Developed, New
MVAC-EU, New
MVAC— Developing, New
12.0
12.0
12.0
101.7
101.7
109.0
37.4
37.4
47.5
2.0 0.2
2.0 0.2
2.0 0.2
Distributed Systems
Large Retail Food— U.S./Other Developed, New
Large Retail Food— EU, New
Large Retail Food— Developing, New
15.0
15.0
15.0
9,117.3
9,117.3
10,029.1
1,763.7
1,763.7
1,763.7
3,684.0 656.6
3,684.0 656.6
6,139.3 656.6
HFC Secondary Loop and/or Cascade Systems
Large Retail Food— U.S./Other Developed, New
Large Retail Food— EU, New
Large Retail Food— Developing, New
15.0
15.0
15.0
31,910.6
31,910.6
35,101.7
1,984.1
1,984.1
1,984.1
- 784.9
- 784.9
- 784.9
NHs or HC Secondary Loop and/or Cascade Systems
Large Retail Food— U.S./Other Developed, New
Large Retail Food— EU, New
Large Retail Food— Developing, New
15.0
15.0
15.0
45,586.6
45,586.6
50,145.3
5,888.6
5,888.6
8,343.9
- 834.0
- 834.0
- 834.0
C02 Transcritical Systems
Large Retail Food— U.S./Other Developed, New
Large Retail Food— EU, New
Large Retail Food— Developing, New
15.0
15.0
15.0
31,910.6
31,910.6
35,101.7
5,888.6
5,888.6
8,343.9
- 834.0
- 834.0
- 834.0
Retrofits of R-404A
Large Retail Food— U.S./Other Developed, Existing
Large Retail Food— EU, Existing
Large Retail Food— Developing, Existing
8.0
8.0
8.0
500.0
500.0
100.0
—
—
—
- 417.1
- 417.1
- 417.1
HCs
Small Retail Food— U.S./Other Developed, New
Small Retail Food— EU, New
Small Retail Food— Developing, New
20.0
20.0
20.0
—
—
—
0.3
0.3
0.3
- 0.1
- 0.1
- 0.1
(continued)
IV-34
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
Table 2-3: Engineering Cost Data on a Facility Basis (continued)
Capital
Project Cost
Lifetime (2010
Abatement Option/Facility Type (years) USD)
Annual Annual Abatement
Revenue O&M Costs Amount
(2010 USD) (2010 USD) (tC02e)a
HCs
Window Units/Dehumidifiers— U.S./Other Developed,
New
Window Units/Dehumidifiers— EU, New
Window Units/Dehumidifiers— Developing, New
11.5
11.5
11.5
—
—
—
0.3
0.3
0.3
- 0.1
- 0.1
- 0.1
R-32
Unitary AC and PTAC/PTHP— Developed, New
Unitary AC and PTAC/PTHP— Developing, New
15.0
15.0
-29.8
-29.8
2.6
2.6
- 1.2
- 1.2
MCHX
Unitary AC— Developed, New
Unitary AC— Developing, New
15.0
15.0
-27.0
-27.0
2.3
2.3
- 0.8
- 0.8
R-32 with MCHX
Unitary AC— Developed, New
Unitary AC— Developing, New
15.0
15.0
-45.7
-45.7
3.9
3.9
- 1.3
- 1.3
MCHX
Positive Displacement Chiller— Developed, New
Positive Displacement Chiller— Developing, New
20.0
20.0
-877.5
-877.5
52.7
52.7
- 11.3
- 11.3
NH3 or C02
I PR/Cold Storage— Developed, New
I PR/Cold Storage— Developing, New
25.0
25.0
210,659.6
231,725.6
50,228.1
83,121.0
- 258.8
- 258.8
Recovery at Disposal
Auto Disposal Yard— U.S./Other Developed
Auto Disposal Yard— EU
Auto Disposal Yard— Developing
7.0
7.0
7.0
2,025.6
2,025.6
2,228.1
443.3
443.3
443.3
1,083.8 72.0
1,083.8 72.0
237.0 72.0
Recovery at Servicing
Auto servicing station— U.S./Other Developed
Auto servicing station— EU
Auto servicing station— Developing
7.0
7.0
7.0
4,051.1
4,051.1
4,456.3
351.1
351.1
351.1
869.5 57.1
869.5 57.1
194.2 57.1
Leak Repair
Large Retail Food— U.S./Other Developed, Existing
Large Retail Food— EU, Existing
Large Retail Food— Developing, Existing
5.0
5.0
5.0
1,872.9
1,872.9
2,060.2
1,469.7
1,469.7
1,469.7
- 532.4
- 532.4
- 532.4
a Emission reductions shown include only reductions associated with MFCs; they do not include indirect (C02) emissions associated with
differences in energy consumption.
IV.2.5 Marginal Abatement Costs Analysis
This section describes the methodological approach to the assessment of international abatement
measures for refrigeration and air conditioning.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-35
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
IV.2.5.1
Methodological Approach
The analysis is based on the above representative project costs for model facilities in the United
States, developed countries, and developing countries. We applied the costs to calculate the break-even
prices for each appropriate option for each country. The model estimates the mitigation potential based
on the percentage of the total ODS substitutes' baseline (that is, the HFC emissions from the traditional
ODS industries) attributable to each representative type of equipment and the technical effectiveness for
each technology in each facility or system.
IV.2.5.2
Assessment of Technical Effectiveness
The analysis also developed a technical effectiveness parameter, defined as the percentage reductions
achievable by each technology/region/equipment type combination. Estimating this parameter requires
making a number of assumptions regarding the distribution of emissions by facility in addition to
process-specific estimates of technical applicability and market penetration. Market penetration rates
vary over time as systems are upgraded; market penetration is a modeled value that accounts for a
number of elements, such as market choice, the turnover rate to replace existing banks of equipment that
use HFCs, and the lifetime of refrigeration and air-conditioning equipment. Technical effectiveness
figures do not account for indirect GHG impacts (i.e., increases or decreases in electricity or fuel
consumption), which are accounted for in the cost analysis.9 Table 2-4 summarizes these assumptions and
presents technical effectiveness parameters used in the MAC model.
Table 2-4: Technical Effectiveness Summary
Facility/Abatement Option
Technical
Applicability
Market
Penetration
Rate (2030)*
Reduction
Efficiency
Technical
Effectiveness
(2030)"
MVACs - U.S./Other Developed
Enhanced HFC-134a
HFO-1234yf
Enhanced HFO-1234yf
New systems
0%
0%
100%
50%
99.7%
99.8%
0%
36%
64%
MVACs - EU
Enhanced HFC-134a
HFO-1234yf
Enhanced HFO-1234yf
New systems
0%
0%
100%
50%
99.7%
99.8%
0%
0%
32%
(continued)
9 Indirect GHG emissions are not accounted for in the technical effectiveness calculations so that the analysis can
show purely ODS substitute (e.g., HFC) emission reductions achievable. While it is recognized that indirect GHG
emissions can be significant, the incremental differences of the options considered here compared to traditional HFC
systems are expected to be relatively small. Such differences, to the extent data is available on such, are accounted for
in the cost analyses.
IV-36
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
Table 2-4: Technical Effectiveness Summary (continued)
Facility/Abatement Option
Market
Technical Penetration
Applicability Rate (2030)*
Reduction
Efficiency
Technical
Effectiveness
(2030)"
MVACs -Developing
Enhanced HFC-134a
HFO-1234yf
Enhanced HFO-1234yf
New systems
0%
100%
0%
50%
99.7%
99.8%
4%
44%
0%
Large Retail Food - U.S./Other Developed
Distributed systems
HFC secondary loop and/or cascade systems
NHs or HC secondary loop and/or cascade systems
C02 transcritical systems
Retrofits of R-404A
New systems
R-404A systems at or
beyond average age
10%
50%
20%
20%
100%
80%
95%
100%
100%
46%
17%
28%
13%
19%
1%
Large Retail Food - ED
Distributed systems
HFC secondary loop and/or cascade systems
NHs or HC secondary loop and/or cascade systems
C02 transcritical systems
Retrofits of R-404A
New systems
R-404A systems at or
beyond average age
0%
0%
35%
60%
100%
80%
95%
100%
100%
46%
10%
16%
21%
47%
1%
Large Retail Food - Developing
Distributed systems
HFC secondary loop and/or cascade systems
NHs or HC secondary loop and/or cascade systems
C02 transcritical systems
Retrofits of R-404A
New systems
R-404A systems at or
beyond average age
30%
37%
13%
20%
100%
80%
95%
100%
100%
46%
15%
16%
4%
8%
3%
Small Retail Food - U.S./Other Developed
HCs
New systems
100%
100%
68%
Small Retail Food - EU
HCs
New systems
100%
100%
62%
Small Retail Food - Developing
HCs
New systems
100%
100%
27%
Window AC Units and Dehumidifiers - U.S./Other Developed
HCs
New systems
34%
100%
3%
Window AC Units and Dehumidifiers - EU
HCs
New systems
50%
100%
3%
(continued)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-37
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
Table 2-4: Technical Effectiveness Summary (continued)
Facility/Abatement Option
Technical
Applicability
Market
Penetration
Rate (2030)a
Reduction
Efficiency
Technical
Effectiveness
(2030)"
Window AC Units and Dehumidifiers - Developing
HCs
New systems
50%
100%
3%
PTAC/PTHP - Developed
R-32
New systems
100%
75%
22%
PTAC/PTHP - Developing
R-32
New systems
50%
75%
6%
Unitary AC - Developed
R-32
MCHX
R-32 with MCHX
New systems
0%
0%
100%
75%
37.5%
84.5%
27%
11%
23%
Unitary AC - Developing
R-32
MCHX
R-32 with MCHX
New systems
50%
50%
0%
75%
37.5%
84.5%
8%
12%
0%
Large AC: PD Chillers - Developed
MCHX
New systems
100%
37.5%
20%
Large AC: PD Chillers - Developing
MCHX
New systems
100%
37.5%
19%
IPR & Cold Storage - Developed
NH3 or C02
New systems
40%
100%
21%
IPR & Cold Storage - Developing
NH3 or C02
New systems
20%
100%
5%
Cross-Cutting Practice Options - U.S/Other Developed
Refrigerant recovery at disposal
Refrigerant recovery at servicing (small equipment)
Leak repair (large equipment)
Existing equipment
100%
40%
100%
85%
95%
40%
39%
16%
64%
Cross-Cutting Practice Options - EU
Refrigerant recovery at disposal
Refrigerant recovery at servicing (small equipment)
Leak repair (large equipment)
Existing equipment
100%
40%
100%
85%
95%
40%
65%
15%
84%
Cross-Cutting Practice Options - Developing
Refrigerant recovery at disposal
Refrigerant recovery at servicing (small equipment)
Leak repair (large equipment)
Existing equipment
100%
40%
100%
85%
95%
40%
35%
20%
42%
a Market penetration assumptions for this analysis vary over time, and the technical effectiveness values are based on the cumulative market
penetration rates assumed until that point. More information on the market penetration assumptions is provided in Appendix D to this chapter.
b Technical effectiveness figures represent the percentage of baseline emissions from the relevant facility type that can be abated in 2030;
figures do not account for indirect greenhouse gas impacts (i.e., increases or decreases in electricity or fuel consumption), which are accounted
for in the cost analysis.
IV-38
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
IV.2.5.3
Estimating Abatement Project Costs and Benefits
Table 2-5 provides an example of how the break-even prices are calculated for each abatement
measure. Project costs and benefits are calculated for model facilities in developed and developing
countries and are used in the calculation that solves for the break-even price that sets the project's
benefits equal to its costs. The previous section describes the assumptions used to estimate different costs
for developed and developing countries. Additional details on the analyses can be found in Appendix D
to this chapter.
Table 2-5: Example Break-Even Prices for Abatement Measures in Refrigeration and AC
Abatement Option/Facility Type
Reduced Capital
Emissions Costs
(tC02e) ($/tC02e)
x Benefit
Annual of Break-Even
Cost Depreciation Price
($/tC02e) ($/tC02e) ($/tC02e)
Enhanced HFC-134a
MVAC— U.S./Other Developed, New
MVAC-EU, New
MVAC— Developing, New
0.1
0.1
0.1
198.7
198.7
218.6
-420.8
-420.8
-533.8
45.1
45.1
49.6
-267.2
-267.2
-364.8
HFO-1234yf
MVAC— U.S./Other Developed, New
MVAC-EU, New
MVAC— Developing, New
0.2
0.2
0.2
67.3
67.3
68.7
36.8
36.8
36.8
15.3
15.3
15.6
88.8
88.8
89.9
Enhanced HFO-1234yf
MVAC— U.S./Other Developed, New
MVAC-EU, New
MVAC— Developing, New
0.2
0.2
0.2
115.5
115.5
123.9
-164.2
-164.2
-211.5
26.2
26.2
28.1
-74.9
-74.9
-115.8
Distributed systems
Large Retail Food— U.S./Other Developed, New
Large Retail Food— EU, New
Large Retail Food— Developing, New
656.6
656.6
656.6
3.0
3.0
3.3
2.9
2.9
6.7
0.6
0.6
0.7
5.4
5.4
9.3
HFC secondary loop and/or cascade systems
Large Retail Food— U.S./Other Developed, New
Large Retail Food— EU, New
Large Retail Food— Developing, New
784.9
784.9
784.9
8.9
8.9
9.8
-2.5
-2.5
-2.5
1.8
1.8
2.0
4.6
4.6
5.3
NHs or HC secondary loop and/or cascade
systems
Large Retail Food— U.S./Other Developed, New
Large Retail Food— EU, New
Large Retail Food— Developing, New
834.0
834.0
834.0
12.0
12.0
13.2
-7.1
-7.1
-10.0
2.4
2.4
2.7
2.5
2.5
0.5
C02 Transcritical systems
Large Retail Food— U.S./Other Developed, New
Large Retail Food— EU, New
Large Retail Food— Developing, New
834.0
834.0
834.0
8.4
8.4
9.2
-7.1
-7.1
-10.0
1.7
1.7
1.9
-0.4
-0.4
-2.7
(continued)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-39
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
Table 2-5: Example Break-Even Prices for Abatement Measures in Refrigeration and AC (continued)
Abatement Option/Facility Type
Reduced Capital Annual of Break-Even
Emissions Costs Cost Depreciation Price
(tC02e) ($/tC02e) ($/tC02e) ($/tC02e) ($/tC02e)
Retrofits of R-404A
Large Retail Food— U.S./Other Developed,
Existing
Large Retail Food— EU, Existing
Large Retail Food— Developing, Existing
417.1
417.1
417.1
0.4
0.4
0.1
—
—
—
0.1
0.1
0.0
0.3
0.3
0.1
HCs
Small Retail Food— U.S. /Other Developed, New
Small Retail Food— EU, New
Small Retail Food— Developing, New
0.1
0.1
0.1
—
—
—
-3.5
-3.5
-3.5
—
—
—
-3.5
-3.5
-3.5
HCs
Window Units/Dehumidifiers— U.S./Other
Developed, New
Window Units/Dehumidifiers— EU, New
Window Units/ Dehumidifiers— Developing, New
0.1
0.1
0.1
—
—
—
-2.6
-2.6
-2.6
—
—
—
-2.6
-2.6
-2.6
R-32
Unitary AC and PTAC/PTHP— Developed, New
Unitary AC and PTAC/PTHP— Developing, New
1.2
1.2
-5.6
-5.6
-2.2
-2.2
-1.1
-1.1
-6.7
-6.7
MCHX
Unitary AC— Developed, New
Unitary AC— Developing, New
0.8
0.8
-7.3
-7.3
-2.9
-2.9
-1.5
-1.5
-8.7
-8.7
R-410AtoR-32
Unitary AC— Developed
Unitary AC— Developing
1.2
1.2
-5.6
-5.6
-2.2
-2.2
-1.1
-1.1
-6.7
-6.7
R-32 with MCHX
Unitary AC— Developed
Unitary AC— Developing
1.3
1.3
-7.6
-7.6
-3.0
-3.0
-1.5
-1.5
-9.1
-9.1
MCHX
Positive Displacement Chiller— Developed, New
Positive Displacement Chiller— Developing, New
258.8
258.8
149.5
164.4
-194.1
-321.2
21.7
23.9
-66.3
-180.7
NH3 or C02
IPR/Cold Storage— Developed, New
IPR/Cold Storage— Developing, New
258.8
258.8
149.5
164.4
-194.1
-321.2
21.7
23.9
-66.3
-180.7
Recovery at disposal
Auto Disposal Yard— U.S./Other Developed
Auto Disposal Yard— EU
Auto Disposal Yard— Developing
72.0
72.0
72.0
9.6
9.6
10.6
8.9
8.9
-2.9
2.7
2.7
2.9
15.8
15.8
4.8
(continued)
IV-40
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
Table 2-5: Example Break-Even Prices for Abatement Measures in Refrigeration and AC (continued)
Abatement Option/Facility Type
Reduced Capital Annual of Break-Even
Emissions Costs Cost Depreciation Price
(tC02e) ($/tC02e) ($/tC02e) ($/tC02e) ($/tC02e)
Recovery at servicing
Auto servicing station— U.S./Other Developed
Auto servicing station— EU
Auto servicing station— Developing
57.1
57.1
57.1
24.3
24.3
26.7
9.1
9.1
-2.8
6.8
6.8
7.4
26.6
26.6
16.5
Leak Repair
Large Retail Food— U.S./Other Developed,
Existing
Large Retail Food— EU, Existing
Large Retail Food— Developing, Existing
532.4
532.4
532.4
1.5
1.5
1.7
-2.8
-2.8
-2.8
0.5
0.5
0.5
-1.7
-1.7
-1.6
IV.2.5.4
MAC Analysis Results
Global abatement potential in 2020 and 2030 is 208 and 994 MtCO2e, respectively. There are 479
MtCO2e of emissions reductions available in 2030 from implementing currently available technologies
that are cost-effective at projected costs. If an additional emissions reduction value (e.g., tax incentive,
subsidy, or tradable emissions reduction credit) above the zero break-even price were available to users
or manufacturers of refrigeration and AC systems, then additional emission reductions could be cost-
effective. The results of the MAC analysis are presented in Table 2-6 and Figure 2-4 by major country and
regional grouping at select break-even prices in 2030.
Table 2-6: Abatement Potential by Country/Region at Selected Break-Even Prices in 2030 (MtC02e)
Country/Region
^^H
^B^
^•_n~R^j
M
100
100+
Top 5 Emitting Countries
China
Japan
Russia
South Korea
United States
5.6
6.7
0.8
8.0
31.7
35.6
23.8
4.8
28.4
112.9
108.7
29.0
14.7
34.6
137.5
152.1
36.3
20.5
43.3
172.1
226.2
39.3
30.5
47.0
186.5
226.2
39.3
30.5
47.0
186.5
260.4
45.6
35.1
54.5
216.3
260.4
48.2
35.1
57.6
228.6
260.4
48.2
35.1
57.6
228.6
279.1
51.2
37.6
61.2
243.1
279.1
51.2
37.6
61.2
243.1
Rest of Region
Africa
Central and South America
Middle East
Europe
Eurasia
Asia
North America
Total
0.7
1.0
2.9
2.8
0.1
2.6
4.8
67.7
4.3
5.9
11.8
29.8
0.6
13.4
17.2
288.4
13.2
16.5
19.9
49.2
1.8
32.7
20.9
478.6
18.5
22.9
26.3
62.0
2.6
45.0
26.2
627.7
27.5
33.5
33.7
66.0
3.8
64.2
28.4
786.7
27.5
33.5
33.7
66.0
3.8
64.2
28.4
786.7
31.6
38.6
38.9
78.6
4.4
74.0
32.9
910.9
31.6
38.7
39.9
82.1
4.4
74.5
34.8
935.8
31.6
38.7
39.9
82.1
4.4
74.5
34.8
935.8
33.9
41.5
42.6
82.7
4.7
79.8
37.0
994.3
33.9
41.5
42.6
82.7
4.7
79.8
37.0
994.3
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-41
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
Figure 2-4: Marginal Abatement Cost Curves for Top Five Emitters in 2030
300
•China
Russia
•Japan
•South Korea
• United States
Non-CO2 Reduction (MtCO2e)
IV.2.6
Uncertainties and Limitations
The development of alternative refrigerants and technologies is quickly evolving in this sector, with
efficiencies increasing and costs decreasing as research and market share expand. Thus, the costs and
reduction efficiencies of the alternatives reviewed in this analysis are subject to change and likely very
conservative. Moreover, new options not quantified in this analysis are entering the market and will
continue to do so; additional options, such as HCs in domestic refrigerators, CO2 in transport
refrigeration, and low-GWP refrigerants for comfort cooling chillers, could be quantitatively considered
in future analyses.
In addition, the costs for the options explored in this analysis are highly variable, depending on the
types of systems reviewed. In particular, estimates of the amount of refrigerant recoverable from
equipment at service and disposal are highly uncertain and highly variable based on the type of
equipment. Recovery from large equipment is generally more cost-effective than for small equipment,
because the amount of refrigerant recoverable is greater and the relative amount of technician time
needed to perform the recovery is smaller. Similarly, because leak repair can be performed on many
different equipment types and can involve many different activities/tools, it is difficult to determine an
average cost of such repairs or the average emission reduction associated with them. This analysis,
therefore, relies on broad assumptions available in the published literature, which may not reflect specific
or even average values for the leak repair activities modeled.
Finally, it is assumed that numerous abatement options result in increased or decreased energy
consumption (e.g., enhanced HFO-1234yf or HFC-134a in MVACs, CC>2 transcritical large retail food
IV-42
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
refrigeration systems, distributed refrigeration systems, NHs or CO2 in new IPR and cold storage
systems). While the costs associated with the increase or decrease in energy consumption, which would
vary widely based on region as well as particular application, is quantified as part of this analysis, the
increase or decrease in CO2 emissions associated with this energy use is not quantified. To accurately
capture net emission reductions of these abatement options, emissions associated with the increase or
decrease in energy use should also be calculated.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-43
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
References
3M. 2008. Personal communication between John Owens of 3M and Emily Herzog of ICF International.
August 2008.
Air-Conditioning, Heating and Refrigeration Institute (AHRI). 2013. AHRI Low-GWP Alternative
Refrigerants Evaluation Program. Available at
http://www.ahrinet.org/ahri+low gwp+alternative+refrigerants+evaluation+program.aspx. Accessed
August 28, 2013.
Air Conditioning, Heating, Refrigeration News (ACHR News). 2012. Retrofitting with HCFs. December 3,
2012. Available at: http://digital.bnpmedia.com/publication/?i=136503.
Air Conditioning, Heating, Refrigeration News (ACHR News). 2010. CO2 in Supermarket Refrigeration.
September 6, 2010. Available at: http://www.achrnews.com/articles/co2-in-supermarket-refrigeration.
Arthur D. Little, Inc. (ADL). 2002. Global Comparative Analysis of HFC and Alternative Technologies for
Refrigeration, Air Conditioning, Foam, Solvent, Aerosol Propellant, and Fire Protection Applications. Final
Report to the Alliance for Responsible Atmospheric Policy. Reference Number 75966.
Automotive Engineering International (AEI). 2008. "Consensus Building on Refrigerant Type."
September 2008.
Australian Green Cooling Council. 2008. Transcritical CO2 (R744) Supermarket Refrigeration in Australia.
Available at: http://ec.europa.eu/clima/events/0007/australian green cooling en.pdf.
Carrier. November 11, 2008. Personal communication between William Walter and Richard Lord of
Carrier, and Brie Johnson of ICF International.
Carrier. 2013. NaturaLINE: Another Milestone in Innovation and Sustainability. Obtained June 12, 2013,
at:
http://\v\v\v.container.carrier.com/Carrier+Brand+Sites/Carrier+Transicold+Container/Side+Bar+Links
/Home/NaturaLINE.
Centre Ricerche Fiat. June 10, 2008. The Four Fiat Pandas Experiment - To evaluate different MAC systems.
Presentation given by C-Aa Malvicino, Head of Thermal Systems Dept., Centre Ricerche Fiat at the
SAE 2008 Alternate Refrigerant Systems Symposium, Scottsdale, AZ.
Daikin. September 27, 2012. News Release: World's First Commercialization of Air Conditioning Equipment
Using Next-Generation Refrigerant HFC32. Obtained February 14, 2013, at:
http://www.daikin.com/press/2012/120927/index.html.
DelVentura, R. et al., September 2007. "BOHN the Cold Standard: Secondary Loop Systems for the
Supermarket Industry." White Paper.
Koban, Mary (DuPont Fluoroproducts). 2009. "HFO-1234yf Low GWP Refrigerant LCCP Analysis."
Presented at SAE 2009 World Congress. April 2009.
Earthcare. October 2008. Personal communication between Nicholas Cox, Earthcare Products Ltd., and
Brie Johnson, ICF International.
Environmental Investigation Agency. 2012. HFC-Free Technologies Are Available in the US Market for the
Supermarket-Retail Refrigeration Sector. Available at http://eia-global.org/PDF/USSUPERMARKET
REPORT.pdf.
Environmental Leader: Environmental and Energy Management News. December 1, 2010. "Carrier
Unveils Natural Refrigerated Transport Solution." Obtained January 25, 2013, at:
http://www.environmentalleader.com/2010/12/01/carrier-unveils-natural-refrigerated-transport-
solution/.
EurActiv. January 23, 2013. "Daimler Faces Legal Action Threat over Green Air Conditioning." Obtained
February 7, 2013, at: http://www.euractiv.corn/dimate-environment/daimler-slammed-return-mac-
direc-news-516741.
IV-44 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
Godrej. 2012. "HC Technology for Room Air Conditioners." Presented by Dilip Rajadhyaksha at
Advancing Ozone and Climate Protection Technologies: Next Steps, Bangkok, Thailand. July 21-22,
2012.
Gooseff, A., and J. Horton. 2008. "Ammonia as the Sustainable Refrigerant: An Ammonia-Halocarbon
Comparison." Presented at 30th Annual Meeting International Institute of Ammonia Refrigeration,
Colorado Springs, CO. March 16-19, 2008.
Gree Electronic Appliances. 2012. "Application of HC Refrigerant in Room Air Conditioners." Presented
at Advancing Ozone and Climate Protection Technologies: Next Steps, Bangkok, Thailand. July 21-22,
2012.
Greenpeace. 2009. Cool Technologies: Working Without HFCs: Examples of HFC-Free Cooling Technologies in
Various Industrial Sectors. Prepared by J. Mate of Greenpeace International with C. Papathanasopoulos
of Greenpeace USA. Available at: http://www.greenpeace.org/usa/
assets/binaries/cool-technology-report-2009.
GTZ Proklima. 2008. Natural Refrigerants: Sustainable Ozone- and Climate-Friendly Alternatives to HCFCs.
Deutsche Gesellschaft fur Technische Zusammenarbeit (GTZ) GmbH. Eschborn, Germany. July 2008.
Available at: http://epa.gov/greenchill/downloads/en-gtz-proklima-natural-refrigerants.pdf.
GTZ Proklima. 2009. Gree Electric R290 Air Conditioner. Presented on behalf of Gree by Dr. Volkmar
Hasse, GTZ-Proklima International, at the Joint West Asia and South Asia Network Meeting. May 10,
2009. Available at: http://www.hydrocarbons21.com/files/news/gree-ac-unit-hydrocarbons.pdf.
Hill PHOENIX. 2012. "Advances in CO2 Supermarket Refrigeration in North America." Presented by S.
Martin, Hill PHOENIX, at the ATMOsphere 2012 conference, Washington, DC. Available at:
http://www.atmo.org/media.presentation.php?id=139.
Hinde, D. Zha, S. and Lan, L ASHRAE. February 2009. "Carbon Dioxide in North American
Supermarkets." ASHRAE Journal 51.
Hydrocarbonconversions.com. November 15, 2011. 200 Lidl stores to feature R290 technology by 2012.
Available at: http://www.hydrocarbonconversions.com/200-lidl-stores-to-feature-r290-technology-by-
20127.
ICF International. 2008. Regulatory Impact Analysis (RIA) of Current Section 609 Requirements and Potential
Updates: Draft Report. Submitted to the U.S. EPA. Washington, DC: ICF International.
Intergovernmental Panel on Climate Change (IPCC). 2005. IPCC/TEAP Special Report on Safeguarding the
Ozone Layer and the Global Climate System: Issues Related to Hydrofluorocarbons and Perfluorocarbons. In B.
Metz, L. Kuijpers, S. Solomon, S.O. Andersen, O. Davidson, J. Pons, et al. (Eds.). Prepared by
Working Groups I and III of the IPCC, and the Technical and Economic Assessment Panel. New
York, NY: Cambridge University Press.
Oko-Recherche, Oko-Institute, HEAT International, Danish Technological Institute, Re-phridge,
Karlsruhe University of Applied Sciences, Estonia Environmental Research Centre, and Ammonia
Partnership. 2011. Preparatory study for a review of regulation (EC) No 842/2006 on certain fluorinated
greenhouse gases. Prepared for the European Commission. September 2011.
Pham, H. and H. Sachs. 2010. Next Generation Refrigerants: Standards and Climate Policy Implications of
Engineering Constraints. ACEEE Summer Study on Energy Efficiency in Buildings. Available at:
http://www.aceee.org/files/proceedings/2010/data/papers/1933.pdf.
R744.com. 2012. Cascade CO2 could be winning option in emerging economies - CCAC Forum, Part 118.
December 18, 2012. Available at http://www.r744.com/news/view/3755.
Refrigeration, Air Conditioning and Heat Pumps Technical Options Committee (RTOC). 2011. 2020
Assessment. UNEP Nairobi. Available at: http://ozone.unep.org/teap/Reports/RTOC/RTOC-
Assessment-report-2010.pdf.
Refrigeration and Air Conditioning (RAC) Magazine. 2013. "US car giant GM publicly backs HFO-1234yf."
January 7, 2013. Obtained February 7, 2013, at: http://wTv\v.racplus.com/ne\vs/us-car-giant-gm-
publicly-backs-1234yf/8640815.artide?blocktitle=Latest-news&contentID=2332.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-45
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
Rugh, J., and V. Hovland. July 17, 2003. National and World Fuel Savings and carbon dioxide Emission
Reductions by Increasing Vehicle Air Conditioning COP. Presented by John Rugh and Valerie
Hovland of the National Renewable Energy Laboratory at the SAE 2003 Automotive Alternate
Refrigerant Systems Symposium in Phoenix, AZ. Society of Automotive Engineers.
Shecco. 2012. "2012: Natural Refrigerants Market Growth for Europe." Available at:
http://guide.shecco.com/.
Stanga, M. December 13, 2012. Correspondence: Cost ofR-32for unitary AC equipment.
Supermarket News. 2012. "Refrigeration Systems Chillin' with Carbon Dioxide." Available at:
http://supermarketnews. com/technology /refrigeration-systems-chillin-carbon-
dioxide?page=lshowtext.cfm?t=ptb0810.
SuperValu. June 12, 2012. SuperValu's Development of Next Generation Refrigeration Systems: Application of
Ammonia, CO2 and Propane. Heath, R. and J. Armer. Presented at ATMOsphere America the Business
Case: Natural Refrigerants.
Technology and Economic Assessment Panel (TEAP). 2012. Decision XXIII19 Task Force Report: Additional
Information on Alternatives to Ozone-Depleting Substances.
Technology and Economic Assessment Panel (TEAP). 2011. UNEP 2010 Report of the Refrigeration, Air
Conditioning and Heat Pumps Technical Options Committee: 2010 Assessment. February 2011. ISBN 978-
9966-20-002-0. Available at http://ozone.unep.org/teap/Reports/RTOC/RTOC-Assessment-report-
2010.pdf.
Unilever. 2008. "Unilever Ice Cream Cabinets: Conversion to Natural Refrigerants." Case Study
disseminated at the EC International Technical Meeting on HCFC Phase-Out. Montreal, Canada,
April 5-6, 2008. Available at:
http://ec.europa.eu/environment/ozone/pdf/unilever hfc icecream cabinets.pdf.
U.S. Bureau of Labor Statistics (BLS). 2007. Occupational Employment and Wages, May 2006: 49-3023
Automotive Service Technicians and Mechanics. Available at: http://www.bls.gov/oes/current/
oes493023.htm.
U.S. Bureau of Labor Statistics (BLS). 2011. Consumer Price Index. Obtained May 2011 at:
http://www.bls.gov/cpi/.
U.S. Energy Information Administration (USEIA). 2011. Annual Energy Review: 2010 industrial (real)
Electricity Cost. Table 8.10: Average Retail Prices of Electricity, 1960-2010. Available at
http://www.eia.gov/totalenergy/data/annual/showtext.cfm?t=ptb0810. Accessed October 19, 2011.
U.S. Energy Information Administration (USEIA). 2008. Retail Motor Gasoline Prices in Selected
Countries, 1990-2007. Obtained October 2008, at:
http://www.eia.gov/countries/prices/gasolinewithtax.cfm.
U.S. Environmental Protection Agency (USEPA). May 1998. Draft Regulatory Impact Analysis: The
Substitutes Recycling Rule. Prepared by ICF Incorporated for U.S. Environmental Protection Agency.
Washington, DC: ICF International.
U.S. Environmental Protection Agency (USEPA). June 2001. U.S. High GWP Gas Emissions 1990-2010:
Inventories, Projections, and Opportunities for Reductions. EPA #000-F-97-000. Washington, DC: Office of
Air and Radiation, U.S. Environmental Protection Agency.
U.S. Environmental Protection Agency (USEPA). June 2006. Global Mitigation ofNon-CO2 Greenhouse Gases.
EPA #430-R-06-005. Washington, DC: Office of Atmospheric Programs, U.S. Environmental
Protection Agency.
U.S. Environmental Protection Agency (USEPA). October 2009. Marginal Abatement Cost Curve Analysis for
Reduction of HFCs in Traditional Ozone Depleting Substance (ODS) End-Use Applications: Draft Report.
Prepared by ICF International for U.S. Environmental Protection Agency.
U.S. Environmental Protection Agency (USEPA). (2012a). Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA. Obtained from:
http://www.epa.gov/dimatechange/econornics/international.html.
IV-46 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM REFRIGERATION AND AIR CONDITIONING
U.S. Environmental Protection Agency and National Highway Traffic Safety Administration (USEPA and
NHTSA). 2011. Draft Joint Technical Support Document: Proposed Rulemaking for 2017-2025 Light-Duty
Vehicle Greenhouse Gas Emission Standards and Corporate Average Fuel Economy Standards.
WalMart. September 2006. DOW Case Study: Secondary Loop Refrigeration and DOWFROST™ Fluid
Contribute to Wal-Mart "Green Initiative."
Wang, K., M. Eisele, Y. Hwang, and R. Radermacher. October 2010. Review of secondary loop
refrigeration systems. International Journal of Refrigeration 33 (2), 212-234.
Xu, X., Y. Hwang, and R. Radermacher. 2012. "Performance Measurement of R32 in Vapor Injection Heat
Pump System." Presented at the International Refrigeration and Air Conditioning Conference at
Purdue, July 16-19, 2012.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-47
-------
HFC EMISSIONS FROM SOLVENT USE
IV.3. HFC Emissions from Solvent Use
IV.3.1 Sector Summary
istorically, chlorofluorocarbons (CFCs) (in particular CFC-113), methyl chloroform, and, to a
lesser extent, carbon tetrachloride were used as the predominant solvent cleaning agents.
Hydrofluorocarbons (HFCs), hydrofluoroethers (HFEs), perfluorocarbons (PFCs), and
aqueous and semi-aqueous not-in-kind (NIK) solvents have since replaced these historical solvents, with
HFC emissions currently dominating the global warming potential (GWP)-weighted emissions from the
solvents sector.
Greenhouse gas emissions from the solvents sector (excluding CFCs and HCFCs) were estimated at
roughly 5 million metric tons of carbon dioxide equivalent (MtCO2e.) in 2010. By 2030, emissions from
this sector are expected to more than double, reaching over 10 MtCO2e. A majority of the growth will
result from increased use of HFCs in developing countries. Figure 3-1 presents the HFC and PFC baseline
emissions from solvent use between 2000 and 2030.
Figure 3-1: HFC and PFC Emissions from Solvent Use: 2000-2030 (MtC02e)
10
South Korea
I Russia
Japan
I United States
I China
ROW
2000
2010
2020
2030
Year
Source: U.S. Environmental Protection Agency (USEPA), 2012a
Four abatement options were identified for the solvent sector: (1) replacement of HFCs with HFEs,
(2) retrofitting of equipment, (3) transition to NIK aqueous, and (4) transition to NIK semi-aqueous.
The global abatement potential is equal to approximately 58.9% of total annual emissions from the
solvent sector and 0.3% of total annual emissions from all sectors that use ODS substitutes. These results
are partly due to the assumed adoption of HFEs which, although they have a relatively lower GWP than
HFCs, still result in emissions of greenhouse gases. In the same way, the adoption of equipment retrofits,
another abatement option, still results in emissions of greenhouse gases. Finally, it is assumed that due to
the performance limitations of the available alternatives, in the absence of policy measures, a portion of
the market will not make the transition away from HFCs.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-49
-------
HFC EMISSIONS FROM SOLVENT USE
Marginal abatement cost (MAC) curve results are presented in Figure 3-2. Maximum abatement
potential in the solvents sector is 5.7 MtCO2e in 2030. There are 4.8 MtCO2e of emissions reductions in
2030 that are cost-effective (i.e., $0/tCO2e or lower abatement cost) at currently projected energy prices.
Figure 3-2: Global Abatement Potential in Solvent Use: 2010, 2020, and 2030
8
$0
-$30
f—-
2010
2020
2030
Non-CO2 Reduction (MtCO2e)
IV.3.2
Emissions from Solvents
Although solvents are primarily an emissive use, emissions from solvent applications are not equal to
the amount of solvent consumed in a year because a portion of used solvent remains in the liquid phase
and is not emitted as gas during use. However, as the solvent is continuously reused through a distilling
and cleaning process or through recycling, it is assumed that eventually approximately 90% of the solvent
consumed in a given year is emitted, while 10% of solvent is disposed of with the sludge that remains.
For the purpose of this analysis, the sector is characterized by precision cleaning applications and
electronics cleaning applications. Precision cleaning requires a high level of cleanliness to ensure the
satisfactory performance of the product being cleaned, and electronics cleaning is defined as a process
that removes contaminants, primarily solder flux residues, from electronics or circuit boards. To develop
the cost analysis, the model vapor degreaser is assumed to be 10 square feet in size, uses HFC-4310mee as
a solvent, and emits 250 to 500 pounds of solvent annually, depending on whether the equipment has
been retrofitted. Figure 3-3 presents the global distribution of HFC and PFC emissions from solvent use in
2020 by degreaser type.
IV-50
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM SOLVENT USE
Figure 3-3: Global HFC Emissions in 2020 by Degreaser Type (% of GWP-Weighted Emissions)
Non-retrofitted
Electronic
Cleaning, 16% Retrofitted
Precision
Cleaning, 36%
Retrofitted
Electronic
Cleaning, 27%
Non-retrofitted
Precision
Cleaning, 21%
IV.3.2.1
Activity Data or Important Sectoral or Regional Trends
Solvent consumption, which is estimated using USEPA's Vintaging Model for the United States, is
used to represent activity data. Solvent consumption is scaled according to country gross domestic
product. Solvent emissions are directly correlated with solvent consumption; it is assumed that almost all
(90%) of the solvent consumed in a given year is emitted. There are no regional differences in assumed
emissions rates.
In developed countries, retrofits are assumed to have already been fully adopted, and in developing
countries all equipment is assumed to remain nonretrofitted. In addition, although NIK replacement
alternatives and HFE solvent applications currently exist worldwide, the baseline emissions considered
here only covers that portion of the market still using HFCs and PFCs. Hence, for the purposes of
transitioning away from the high GWP solvents in this analysis, there is no technology adoption of the
NIK and HFE solvents in the baseline.
IV.3.2.2
Emission Estimates and Related Assumptions
Global emissions of HFCs from the solvents sector were 5 MtCO2e in 2010, growing to 10 MtCO2e in
2030. Table 3-1 presents the projected emissions for solvents use by country and regions between 2010
and 2030. All emissions are the result of HFC-4310mee consumption.1 Emissions are projected to grow
significantly as developing country economies grow and demand for such solvents grows. Emissions
were estimated based on assumptions about initial market size of the sector, the specific transitions away
from CFCs and other ODSs in terms of timing and alternative solvent used, charge sizes, and leak rates,
using the Vintaging Model.
1 PFC solvent use in precision cleaning end-uses is assumed to discontinue such that no emissions of PFCs are
projected beyond 2010 from this sector.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-51
-------
HFC EMISSIONS FROM SOLVENT USE
Table 3-1: Projected Baseline Emissions from Solvent Use: 2010-2030 (MtC02e)
Coun^/Region
2010
2015
2020
2025
2030
CAGR*
(2010-2030)
Top 5 Emitting Countries
China
United States
Japan
Russia
South Korea
0.7
1.3
0.9
0.4
0.2
1.0
1.5
1.0
0.4
0.2
1.4
1.6
1.0
0.5
0.2
2.0
1.8
1.1
0.7
0.3
2.7
2.0
1.2
0.8
0.3
6.9%
2.0%
1.2%
4.2%
3.5%
Rest of Regions
Africa
Central & South America
Middle East
Europe
Eurasia
Asia
North America
World Total
0.1
0.2
0.0
1.1
0.0
0.2
0.2
5.2
0.1
0.2
0.0
1.1
0.0
0.3
0.2
6.0
0.1
0.2
0.0
1.2
0.0
0.3
0.2
7.0
0.1
0.3
0.1
1.3
0.0
0.4
0.3
8.2
0.2
0.3
0.1
1.4
0.0
0.5
0.3
9.7
3.9%
3.3%
3.8%
1.2%
4.2%
4.2%
2.5%
3.1%
aCAGR = Compound Annual Growth Rate
Source: USEPA, 2012a
IV.3.3 Abatement Measures and Engineering Cost Analysis
A total of four abatement options were identified and analyzed for the solvent sector: (1) replacement
of HFCs with HFEs, (2) retrofitting of equipment, (3) transition to NIK aqueous, and (4) transition to NIK
semi-aqueous. Table 3-2 provides a technology overview of each abatement option.
Low-GWP alternatives for use in solvent applications are still emerging onto the market—
perfluorobutyl iodide and Solstice 1233zd(E) are two such alternatives. The introduction of these
substances involves regulatory approvals (such as the Significant New Alternatives Policy (SNAP)
program evaluation process in the United States) followed by entry into the market and acceptance by
users. These alternatives are discussed qualitatively in this chapter under "12.3.5 Low-GWP
Alternatives."
Table 3-2: Solvent Use Abatement Options
Abatement Option
HFCtoHFE
Retrofit
NIK aqueous
NIK semi-aqueous
Reduction Efficiency
76.4%
50%
100%
100%
Applicability
All facilities
Nonretrofitted facilities
Electronics cleaning
Electronics cleaning
IV-52
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM SOLVENT USE
IV.3.3.1 HFCtoHFE
This option, which is applicable to all facilities in the baseline, examines the replacement of HFC-
4310mee with lower GWP HFE solvents. Although other low-GWP chemicals may be feasible, HFE-7100
and HFE-7200 are used as proxies for this abatement option because they display material compatibility
properties similar to HFCs, a prime factor that has led to their success in the market. To model emission
reductions, this option assumes that the degreaser transitions to the use of 75% HFE-7100 and 25% HFE-
7200.2 For the purpose of this analysis, the 100-year GWP of alternative solvents is calculated as the
weighted average of 75% HFE-7100, with a GWP of 390, and 25% HFE-7200, with a GWP of 55, for a GWP
of 306.25. The GWP of the solvent being replaced, HFC-4310mee, is 1,300;3 thus, this option has a
reduction efficiency of 76.4%.
Costs associated with the conversion to HFE solvents are assumed to be negligible because of
similarities in key chemical properties of HFE solvents and HFC-4310mee, as well as similar pricing
structures.
IV.3.3.2 Retrofit
This abatement option is applicable to nonretrofitted facilities using solvents for the purpose of
precision cleaning and electronics cleaning. Retrofits, including engineering control changes (e.g.,
increased freeboard height, installation of freeboard chillers, and use of automatic hoists), improved
containment, and implementation of other abatement technologies can reduce emissions of HFCs used in
solvent cleaning. Retrofitting a vapor degreaser, combined with proper operation and maintenance, can
reduce solvent emissions from 46% to as much as 70%, depending on the specific retrofit methods chosen
(Durkee, 1997). For example, installing a freeboard refrigeration device, sometimes referred to as a chiller
(i.e., a set of secondary coils mounted in the freeboard), and maintaining a freeboard ratio of 1.0 to
minimize diffusional solvent losses, can reduce emissions by 46%, while installing heating coils to
produce superheated vapor along with installing a chiller can reduce emissions by 70%. For the purpose
of this analysis, the reduction efficiency of the retrofit option is assumed to equal 50%.
In the United States, many enterprises have bought new equipment or retrofitted aging equipment
into compliance with the National Emissions Standard for Hazardous Air Pollutants (NESHAP), which
limits emissions from degreasers using traditional chlorinated solvents such as trichloroethylene.
Fluorinated solvents such as HFCs are not covered by this regulation; nonetheless, a number of
companies using HFCs and other nonchlorinated solvents have adopted NESHAP-compliant solvent
cleaning machines because of the associated economic, occupational, and environmental benefits (i.e.,
reduced emissions) (Durkee, 1997). Likewise, many European countries have imposed stringent
environmental and safety regulations that require the lowest level of emissions attainable by solvent
degreasing equipment. Retrofit techniques were either already implemented or simply not required if the
user had purchased new emission-tight vapor degreasers. Consequently, this analysis assumes that end
2 In actuality, a facility would choose one of the two HFEs for adoption; however, for modeling purposes this
assumption was used to reflect the market presence of the two HFEs.
3 Although the GWP value for HFC-4310mee was taken from the IPCC Second Assessment Report (1996), the report did
not provide GWP values for either HFE. Consequently, this analysis uses the GWP values listed in the IPCC Third
Assessment Report (2001) for both HFEs.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-53
-------
HFC EMISSIONS FROM SOLVENT USE
users in the United States and developed countries have already adopted these controls and that any
emissions from these facilities cannot be further abated by this option. In contrast, solvent users in non-
Annex I (i.e., developing) countries are assumed to not have retrofitted their equipment but may consider
the equipment retrofit option, because updating their equipment may be preferred over investing in
entirely new units.
Costs associated with adopting improved equipment and cleaning processes using existing solvents
(retrofit) are incurred during the retrofit process and are estimated at $24,500 per degreaser. Annual
savings of $4,500 are also realized through the avoided consumption of HFC that results from a reduction
in emissions.
IV.3.3.3 Not-in-Kind Aqueous
This abatement option is applicable only to facilities that use solvents for the purposes of electronics
cleaning. This option replaces HFC-containing systems used for electronic cleaning end uses with an
aqueous cleaning process. In the aqueous process, a water-based cleaning solution is used as the primary
solvent and is usually combined with a detergent to remove contaminants. Because all HFCs are replaced
with a solvent that does not have a GWP, the reduction efficiency of this option is 100%.
Costs associated with adopting an aqueous NIK replacement alternative are assumed to be $50,000
for the initial investment and $7,400 annually due to energy and water consumption costs. However,
annual savings are also assumed to result from not using an HFC-based cleaner; savings are estimated to
range from $6,700 to $11,200 depending on whether the solvent-based cleaning system had been
retrofitted, which will significantly offset annual costs.
IV.3.3.4 Not-in-Kind Semi-aqueous
This abatement option is applicable only to facilities that use solvents for the purposes of electronics
cleaning. This option replaces HFC-containing systems used for electronic cleaning end uses with a semi-
aqueous cleaning process. In the semi-aqueous process, the cleaning solution is an organic solvent that is
blended with a surfactant, making it water soluble. An example of a solvent/surfactant blend is a
terpene/water combination blended with glycol ethers. Because all HFCs are replaced with solvents that
have no GWP, the reduction efficiency is 100%.
Costs associated with adopting a semi-aqueous NIK replacement alternative are assumed to be
$55,000 for the initial investment and $9,100 annually due to energy and water consumption costs.4
Annual savings are also assumed to result from not using an HFC-based cleaner; savings are estimated to
range from $6,700 to $11,200 depending on whether the solvent-based cleaning system had been
retrofitted.
IV.3.3.5 Low-GWP Alternatives
Two low-GWP alternatives, perfluorobutyl iodide (PFBI) and Solstice 1233zd(E), are also emerging
options. Both substances are new alternatives that may potentially abate HFC and HFE emissions in
4 Although these costs are higher than the NIK aqueous abatement option, it is assumed that the semi-aqueous
option will nonetheless be adopted in some facilities, for example where the NIK aqueous option might not be
effective for the particular cleaning required.
IV-54 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM SOLVENT USE
solvent cleaning; however, it is too soon to determine reasonable market penetration and costs associated
with the transition to such options.
Solstice 1233zd(E) is a hydrochlorofluoro-olefin and is also referred to as trans-l-chloro-3,3,3-
trifluoroprop-1-ene. Solstice 1233zd(E) is part of a new class of solvents specifically designed with a low
atmospheric lifetime, ODP, and GWP, making it a candidate to replace high GWP HFCs and low or
moderate GWP HFE solvents, as well as saturated HCFCs in solvent cleaning applications (UNEP, 2012).
It has a GWP of 4.7 to 7 (USEPA, 2012b) and is nonflammable, making it an attractive option for some
markets. The United States is completing its evaluation of whether it can be considered acceptable for use
in electronics, precision, and metals cleaning.
PFBI also has a low GWP of 5. In 2012, PFBI was listed by the US Significant New Alternatives Policy
(SNAP) program as acceptable for use in electronics, metal, and precision cleaning (USEPA, 2012c). This
substance may be feasible for the cleaning oxygen systems in the aerospace industry as a potential
replacement for HCFC-225ca/cb because of its good cleaning performance (Mitchell and Lowrey, 2012);
however, it is unclear the extent to which this solvent will be used in place of HCFCs as well as HFC and
HFE solvents.
Given the low GWP of these and similar options under development, we could expect emission
reductions to be similar to the NIK aqueous and semi-aqueous options under the same market
penetration assumptions. However, because these chemicals can also compete with HFEs with mid-range
GWPs and could avoid the energy and water consumption barriers seen with the NIK options, market
penetration may be further or faster than the options analyzed here.
IV.3.3.6
Engineering Cost Data Summary
Table 3-3 presents the engineering cost data for each mitigation option outlined above, including all
cost parameters necessary to calculate the break-even price. For more detailed costs see Appendix E.
Table 3-3: Engineering Cost Data on a Facility Basis
Abatement
Option
HFC to HFE
Retrofit
NIK aqueous
NIK semi-
aqueous
Facility Type
Retrofitted
Nonretrofitted
Nonretrofitted
Electronic/Retrofitted
Electronic/Nonretrofitted
Electronic/Retrofitted
Electronic/Nonretrofitted
Project
Lifetime
(years)
15
15
15
15
Capital
Cost
(2010 USD)
$24,500
$50,000
$55,000
Annual
Revenue
(2010 USD)
$4,500
$6,700
$11,200
$6,700
$11,200
Annual O&M
Costs
(2010 USD)
—
<(-? Ann
-------
HFC EMISSIONS FROM SOLVENT USE
IV.3.4.1 Methodological Approach
The analysis is based on the above representative project costs for model facilities. We applied the
costs to calculate the break-even prices for each appropriate option for each country. The model estimates
the mitigation potential based on the percentage of the total ODS substitutes' baseline attributable to each
representative facility and the technical effectiveness for each technology in each facility.
IV.3.4.2 Assessment of Technical Effectiveness
The analysis also developed a technical effectiveness parameter, defined as the percentage reductions
achievable by each technology/facility type combination. Estimating this parameter requires making a
number of assumptions regarding the distribution of emissions from model facilities in addition to
process-specific estimates of technical applicability and market penetration. Market penetration rates
vary over time as systems are upgraded and the options are applied in the future. Table 3-4 summarizes
these assumptions and presents technical effectiveness parameters used in the MAC model.
Table 3-4: Technical Effectiveness Summary
Abatement Option
Precision— retrofitted
Technical
Applicability
60%
Market
Penetration Rate
(2030)
100%
Reduction
Efficiency
76%
Technical
Effectiveness
46%
HFC to HFE
Precision— nonretrofitted
HFC to HFE
Retrofit
60%
100%
80%
20%
76%
50%
37%
10%
Electronics— retrofitted
HFC to HFE
Aqueous
Semi-aqueous
100%
100%
100%
80%
10%
10%
76%
100%
100%
61%
10%
10%
Electronics— nonretrofitted
HFC to HFE
Retrofit
Aqueous
Semi-aqueous
100%
100%
100%
100%
40%
20%
6%
6%
76%
50%
100%
100%
31%
10%
6%
6%
IV.3.4.3 Estimating Abatement Project Costs and Benefits
Table 3-5 provides an example of how the break-even prices are calculated for each abatement
measure. Project costs and benefits are calculated for model facilities in developed and developing
countries and are used in the calculation that solves for the break-even price that sets the project's
monetary benefits equal to its costs. The previous section describes the assumptions used to estimate
costs. The HFC to HFE option is available at no cost and represents 4.3 MtCO2e of reductions in 2030.
The break-even prices presented in Table 3-5 represent model facilities. Actual prices vary by country
because of the scaling of costs and benefits by international price factors.
IV-56 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM SOLVENT USE
Table 3-5: Example Break-Even Prices for Abatement Measures in Solvent Use
Abatement Option
Reduced
Emissions
(tC02e)
Annualized
Capital Costs
($/tC02e)
Net Annual
Cost
($/tC02e)
Tax Benefit of
Depreciation
($/tC02e)
Break-Even
Price
($/tC02e)
Precision— retrofitted
HFC to HFE
113
—
—
—
—
Precision— nonretrofitted
HFC to HFE
Retrofit
136
147
—
36.4
—
-30.5
—
7.4
—
-1.5
Electronics— retrofitted
HFC to HFE
Aqueous
Semi-aqueous
113
147
147
—
74.3
81.8
—
4.7
16.3
—
15.1
16.6
—
64.0
81.4
Electronics— nonretrofitted
HFC to HFE
Retrofit
Aqueous
Semi-aqueous
136
147
295
295
—
36.4
37.2
40.9
—
-30.5
-12.9
-7.1
—
7.4
7.5
8.3
—
-1.5
16.7
25.5
IV.3.4.4 MAC Analysis Results
Global abatement potential in 2020 and 2030 is 3.0 and 5.7 MtCO2e, respectively. There are 4.8
MtCO2e of reductions in 2030 resulting from implementing currently available technologies that are cost-
effective at projected energy prices. The results of the MAC analysis are presented in Table 3-6 and
Figure 3-4 by major country and regional grouping at select break-even prices in 2030.
Table 3-6: Abatement Potential by Country/Region at Selected Break-Even Pr
Country/Region
•
ices in
2030 (IV
Break-Even Price ($/tC02e)
5 10 15 20 30 50
tC02e)
100
100+
Top 5 Emitting Countries
China
Japan
Russia
South Korea
United States
- - 1
- - 0
- - 0
- - 0
— — 1
2
6
4
2
1
1
0
0
0
1
2
6
4
2
1
1.2
0.6
0.4
0.2
1.1
1.2
0.6
0.4
0.2
1.1
1.3
0.6
0.4
0.2
1.1
1.4
0.6
0.4
0.2
1.1
1.4
0.6
0.4
0.2
1.1
1.4
0.8
0.4
0.2
1.3
1.4
0.8
0.4
0.2
1.3
Rest of Region
Africa
Central and South America
Middle East
Europe
Eurasia
Asia
North America
World Total
- - 0
- - 0
- - 0
- - 0
- - 0
- - 0
- - 0
1
1
0
7
0
2
2
- - 4.8
0
0
0
0
0
0
0
1
1
0
7
0
2
2
4.8
0.1
0.1
0.0
0.7
0.0
0.2
0.2
4.8
0.1
0.1
0.0
0.7
0.0
0.2
0.2
4.8
0.1
0.1
0.0
0.7
0.0
0.2
0.2
5.0
0.1
0.2
0.0
0.7
0.0
0.2
0.2
5.1
0.1
0.2
0.0
0.7
0.0
0.2
0.2
5.1
0.1
0.2
0.0
0.9
0.0
0.3
0.2
5.7
0.1
0.2
0.0
0.9
0.0
0.3
0.2
5.7
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-57
-------
HFC EMISSIONS FROM SOLVENT USE
O
u
$0
0.5
1.0
1.5
•China
Russia
•Japan
•South Korea
United States
Non-CO2 Reduction (MtCO2e)
IV.3.5
Uncertainties and Limitations
This analysis assumes that all developed countries have already adopted retrofitted equipment while
all developing nations are still using nonretrofitted equipment. This is a very simplistic assumption that
may not adequately reflect regional differences in the adoption of retrofitted equipment. Additionally, the
reductions associated with adopting retrofitted equipment are based on older sources that may no longer
be applicable to the current market. Further research in this area is needed to refine both baseline
estimates and the reduction potential associated with retrofits.
Another area of uncertainty in this analysis is related to how costs for the mitigation technologies
may vary internationally. The analysis is currently limited due to the lack of region-specific cost
information.
Also, it is assumed that the aqueous and semi-aqueous abatement options result in increased energy
consumption (3M, 2008); however, the increase in CO2 emissions associated with this energy use is not
quantified as part of this analysis. To accurately capture net emission reductions of these abatement
options, emissions associated with the increased energy use should also be calculated.
Finally, low-GWP alternatives for use in solvent applications are still emerging onto the market and
could potentially replace HFCs and HFEs, further reducing projected emissions once adopted. This
analysis does not project further abatement that can occur because of these alternatives as information on
their potential uptake by the market and associated transition costs is unknown at this time.
IV-58
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM SOLVENT USE
References
3M. 2008. "Advantages of Solvent vs. Aqueous Cleaning." Electronic Market Materials Division.
Available at: http://solutions.3ni.coni/wps/portal/3M/en US/Novec/Home/Product Information/
Engineered Fluids/.
3M Performance Materials. October 27, 2003. Written correspondence between industry technical expert
John G. Owens, P.E., of 3M Performance Materials and Mollie Averyt and Marian Martin Van Pelt of
ICF Consulting.
Brulin & Company, Inc. October 2008. Personal communication between Mike Beeks of Brulin &
Company, Inc. and Emily Herzog and Mollie Averyt of ICF International.
Crest Ultrasonics. September/October 2008. Personal communication between Eric Larson of Crest
Ultrasonics and Emily Herzog of ICF International.
Durkee, J.B. 1997. "Chlorinated Solvents NESHAP—Results to Date, Recommendations and
Conclusions." Presented at the International Conference on Ozone Layer Protection Technologies in
Baltimore, MD, November 12-13,1997.
ICF Consulting. October 7, 2003. Personal communication between solvent industry experts and William
Kenyon of ICF Consulting.
Intergovernmental Panel on Climate Change (IPCC). 1996. Climate Change 1995: The Science of Climate
Change, Contribution of Working Group I to the Second Assessment Report of the Intergovernmental
Panel on Climate Change. J.T. Houghton, L.G. Meira Filho, B.A. Callander, N. Harris, A. Kattenberg,
and K. Maskell (ed.). Cambridge, United Kingdom: Cambridge University Press.
Intergovernmental Panel on Climate Change (IPCC). 2001. Climate Change 2001: The Scientific Basis.
Contribution of Working Group I to the Third Assessment of the Intergovernmental Panel on Climate
Change. J.T. Houghton, Y. Ding, D.J. Griggs, M. Noguer, P.J. van der Linden, X. Dai, K. Maskell, and
C.A. Johnson (eds.). Cambridge, United Kingdom and New York: Cambridge University Press, 881
pp.
Mitchell, M. and N. Lowrey. The Search for Nonflammable Solvent Alternatives for Cleaning Aerosopace
Oxygen Systems. Presented at the 2012 International Workshop on Environment and Alternative
Energy. Greenbelt, MD. December 4-7, 2012.
United Nations Environment Programme (UNEP). 2012. Report of the Technology and Economic Assessment
Panel: Decision XXIII/9 Task Force Report Additional Information on Alternatives to Ozone Depleting
Substances. Available at: http://conf.montreal-protocol.org/meeting/mop/mop-24/presession/
Background%20Documents/teap-task-force-XXIII-9-report-may 2012.pdf.
U.S. Environmental Protection Agency (USEPA). (2012a). Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA. Obtained from:
http://www.epa.gov/climatechange/economics/international.html.
U.S. Environmental Protection Agency (USEPA). 2012b. Protection of Stratospheric Ozone: Determination 27
for Significant New Alternatives Policy (SNAP) Program. 77 FR 47768-47779. Available at:
http://www.gpo.gov/fdsys/pkg/FR-2012-08-10/pdf/2012-19688.pdf.
U.S. Environmental Protection Agency (USEPA). 2012c. Substitutes in Non-aerosol Solvent Cleaning Under
SNAP as of August 10, 2012. Available at: http://www.epa.gov/ozone/snap/solvents/
solvents.pdf.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-59
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
IV.4. HFC Emissions from Foams Manufacturin
cam is used as insulation in a range of equipment and products, including refrigerated
appliances, reefers and other refrigerated transport systems, in buildings (e.g., walls, roofs,
floors) and pipes, and to produce other products, such as steering wheels, furniture, and shoes,
for example. A wide variety of foam types are used for these applications, which have historically been
produced with blowing agents that are ozone-depleting substances (i.e., chlorofluorocarbons (CFCs) and
hydrochlorofluorocarbons [HCFCs]), but since the ODS phaseout under the Montreal Protocol, HFCs
have commonly been adopted in their place—primarily HFC-134a, HFC-152a, HFC-245fa, and HFC-
365mfc. Greenhouse gas emissions from the foams sector (excluding CFCs and HCFCs) were estimated at
roughly 22 million metric tons of carbon dioxide (CO2) equivalent (MtCO2e) in 2010. By 2020, emissions
from this sector are expected to reach over 52 MtCC^e, as shown in Figure 4-1.
Figure 4-1: HFC Emissions from Foams Manufacturing: 2000-2030 (MtCC^e)
100 -
90 -
80 -
70 -
60 -
50 -
40 -
30 -
20 -
10 -
0
92
22
I Italy
I France
I Germany
I Japan
I United States
ROW
2000
2010 2020
Year
2030
Source: USEPA, 2012a.
This analysis reviews options to reduce emissions from the foams sector by using low-global
warming potential (GWP) foam-blowing agents in lieu of HFCs in new equipment/products and by
recovering and destroying foam-blowing agents from household refrigerators at the end of the
equipment's life.
Global abatement potential from the options reviewed equates to approximately 40.3% of total
annual foam sector emissions and 21.9% of total emissions from ODS substitutes in 2030. While many
options have been analyzed that can completely replace the HFC blowing agent in foams, abatement in
the foams sector is limited by the lifetime of the installed base of foam products; all abatement
opportunities analyzed replace the blowing agent in newly manufactured foams only, or destroy the
blowing agent only at the foam natural end of life. Marginal abatement cost (MAC) curve results are
presented in Figure 4-2. Maximum abatement potential in the foams sector is 37.0 MtCC^e in 2030. There
are 27 MtCC^e of cost-effective emissions reductions in 2030, representing 29.4% of the foams baseline,
based on the assumptions presented in this analysis. No reductions are available in 2010 as a result of the
assumption that options did not start to penetrate the market until 2011.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-61
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
Figure 4-2: Global Abatement Potential in Foams Manufacturing: 2010, 2020, and 2030
8
I
•2010
•2020
•2030
30 35 40
Non-CO2 Reduction (MtCO2e)
IV.4.2
Emissions from Foams
Although there are two main types of foams—open cell and dosed cell—HFCs are primarily used in
closed-cell foam applications for their physical and performance properties.1 HFC blowing agents are
emitted during product/equipment manufacture, use, disposal, and even following disposal (e.g., in
landfills) if the foam substance is not specially treated. For the purpose of evaluating the cost of reducing
HFC emissions from this sector, this analysis considers emissions from the following closed-cell foam
applications: polyurethane (PU) appliance foam, PU commercial refrigeration foam, extruded
polystyrene (XPS) boardstock foam, PU continuous and discontinuous panel foam, PU one-component
foam, and PU spray foam. The relative GWP-weighted emission shares of these applications in 2020 are
shown in Figure 4-3.
1 Open cell foams experience significant blowing agent leakage due to the cell structure.
IV-62
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
Figure 4-3: Global HFC Emissions in 2020 by Application Type (% of GWP-Weighted Emissions)
PU One-
Component,
PU Continuous/
Discontinuous,
1%
PU Commercial
Refrigeration,
3%
For the purpose of this analysis, the sector considers eight facilities and/or applications, as defined
below.
• PU appliance foam manufacturing facility using HFC blowing agent: characterized as a typical
manufacturing facility that produces 550,000 refrigerators per year and consumes nearly 537,000
kg of HFC-245fa blowing agent annually.
• PU commercial refrigeration foam manufacturing facility using HFC blowing agent:
characterized as a typical manufacturing facility that produces 50,000 commercial units per year
and consumes 70,000 kg of HFC-245fa blowing agent annually.
• PU spray foam contractor using HFC-245fa/CO2 blowing agent: characterized as a typical PU
spray foam contractor that uses nearly 58,000 kg of HFC-245fa/CO2 PU spray foam annually.
• One-component foam manufacturing facility using HFC-134a or HFC-152a blowing agent:
characterized as a typical facility that produces one-component foam and uses over 130,000 kg
per year of HFC blowing agent.
• XPS boardstock production facility using HFC-134a/CO2 blowing agent: characterized as a
typical facility that creates approximately 1,000,000 board feet of XPS boardstock per year across
10 lines using nearly 7,100 kg of HFC-134a and CO2 blowing agent.
• PU continuous and discontinuous foams manufacturing facility using HFC-134a blowing agent:
characterized as a typical manufacturing facility that uses 453,000 kg of HFC-134a per year.
• Appliance demanufacturing facility using manual blowing agent recovery: characterized as a
typical demanufacturing facility that manually processes 125,000 disposed domestic refrigerators
per year.
• Appliance demanufacturing facility using fully automated blowing agent recovery: characterized
as a typical demanufacturing facility that processes 200,000 domestic refrigerators per year using
fully automated equipment.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-63
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
For modeling purposes, data typical for facilities in the United States are used. Certain cost assumptions,
such as capital costs, are adjusted for other regions.2 Otherwise, it is assumed that the costs and
reductions achieved in the modeled facilities could be scaled and would be representative of the costs
and reductions in other regions.
IV.4.2.1 Activity Data, Important Sectoral or Regional Trends and Related
Assumptions
Foam consumption for the United States is estimated using the U.S. Environmental Protection
Agency's (USEPA's) Vintaging Model. This consumption and associated emissions are assumed to scale
with country gross domestic product (GDP), with several regional adjustments made to account for
differences in HFC foam consumption based on data provided in FTOC (2010). Specifically, in the
European Union, HFC consumption for XPS boardstock and commercial refrigeration is assumed to be
lower than in the United States, because of a faster transition to low-GWP alternatives, while no HFC
consumption is assumed in the PU appliance subsector (because the transition away from HFCs is
already complete). Similarly, in developing countries, no HFC consumption is assumed in PU appliance,
commercial refrigeration, XPS boardstock, PU spray, and PU continuous and discontinuous foams,
because these subsectors are transitioning directly from ODS to non-HFC low-GWP alternatives.
Additionally, a reduced proportion of HFC consumption in PU one-component foam is assumed in
developing countries relative to the U.S. subsector (with consumption assumed only for HFC-134a, not
HFC-152a).
IV.4.2.2 Emission Estimates and Related Assumptions
Global HFC emissions from foams were estimated at 22 MtCO2e in 2010, projected to grow to 52
MtCO2eby 2020 and 92 MtCO2e by 2030. Growth in emissions is driven by GDP. Globally, HFC emissions
from foam production and use have been growing because of the phaseout of ODS under the Montreal
Protocol. Because of the costs associated with HFC-based foams, many countries have transitioned/are
transitioning from ODS to hydrocarbons or other non-HFC alternatives. Because of developing countries'
minimal use of HFCs, the growth in global emissions for the past decade has been driven by emissions
from developed countries. Consumption is modeled based on USEPA's Vintaging Model, with emissions
estimated based on assumed blowing agent loss rates at manufacture, during lifetime, and at disposal—
which vary by foam application and blowing agent type. Emissions for major countries and regions are
presented in Table 4-1.
IV.4.3 Abatement Measures and Engineering Cost Analysis
This analysis considers the costs of reducing foam emissions by (1) replacing HFCs with low-GWP
blowing agents in various types of foam manufacturing operations and (2) properly recovering and
disposing of foam contained in the equipment at the end of its life. Specifically, eight abatement options
were identified and analyzed for reducing emissions at product/equipment production by using
hydrocarbon (HC) or CO2 blowing agents in place of HFCs, and two options were identified for reducing
' In developing countries, it is assumed that capital costs are 10% higher than those in the United States.
IV-64 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
Table 4-1: Projected Baseline Emissions from Foams Manufacturing: 2010-2030 (MtCOje)
Country/Region
2010
2015
2020
2025
2030
CAGR*
(2010-2030)
Top 5 Emitting Countries
United States
Japan
Germany
France
Italy
6.1
7.6
1.7
1.3
1.2
8.7
10.0
2.4
1.7
1.6
16.4
15.0
4.4
3.2
2.9
23.0
19.7
5.9
4.3
3.9
30.5
25.5
7.7
5.6
5.1
8.4%
6.3%
7.7%
7.7%
7.7%
Rest of Regions
Africa
Central & South America
Middle East
Europe
Eurasia
Asia
North America
World Total
—
—
—
3.0
0.1
0.2
0.5
21.7
—
—
—
4.1
0.2
0.3
0.7
29.6
—
—
—
7.7
0.3
0.5
1.3
51.8
—
—
—
10.2
0.4
0.8
1.8
70.0
—
—
—
13.4
0.5
1.1
2.4
91.8
0.0%
0.0%
0.0%
7.7%
6.5%
9.4%
8.4%
7.5%
aCAGR = Compound Annual Growth Rate
Source: U.S. Environmental Protection Agency (USEPA), 2012a.
emissions at the end of the equipment's life by using various methods of foam recovery at the time of
appliance disposal. These options are described in the subsections below and summarized in Table 4-2.
Additional details of the calculations are provided in Appendix F to this chapter.
Additional options considered, but not yet included in the cost analysis, are examined in Sections
4.3.11 through 4.3.13, after the 10 options listed in Table 4-2 are discussed. These and other options not
mentioned are also potentially available but have not been included in this analysis due to data
availability and time.
Table 4-2: Foams Manufacturing Abatement Options
Abatement Option
Appliance: MFCs to HCs
Commercial refrigeration: MFCs to HCs
Spray: HFC245fa/C02 to HC
Spray: HFC245fa/C02 to C02
XPS: HFC134a/C02to LCD/alcohol
One component: HFC-134a to HCs
One component: HFC-152a to HCs
Continuous and discontinuous: HFC134ato HCs
Appliance EOL: Manual process
Appliance EOL: Fully automated process
Reduction
Efficiency
100%
100%
100%
100%
100%
100%
100%
100%
85%
95%
Applicability
New PU appliances
New PU commercial refrigeration units
New spray foam applications
New spray foam applications
New XPS boardstock foam applications
New PU one-component foam applications
New PU one-component foam applications
New PU continuous and discontinuous foam applications
Domestic refrigerators reaching end of life
Domestic refrigerators reaching end of life
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-65
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
IV.4.3.1 HCs in PU Appliances
This option replaces HFC-245fa used in PU appliance foam with HCs. HCs are inexpensive and have
near-zero direct GWPs. Technical issues exist with using HCs, including flammability, and lower
insulation performance (USEPA, 2009), but these can be overcome through proper safety controls and
engineering design. A significant advantage of hydrocarbons is that they can be easily blended to impact
a range of properties, such as thermal performance, cell gas pressure, and foam density, as well as cost
(TEAP, 2012). Approximately 50% of hydrocarbons used in appliances are based on cyclopentane and
isopentane, due to their low operating costs and good foam properties (UNEP, 2010; TEAP, 2012).
Cyclopentane has also been blended with isobutene to improve flow-ability and compressive strength, or
with methylal, to improve performance (UNEP, 2010). Flammability, however, may cause a high
incremental capital cost for facilities, which may be uneconomic for small or medium-sized enterprises;
otherwise, HCs have low operating costs (TEAP, 2012). Using HCs instead of HFC-245fa in PU appliance
foam eliminates HFC emissions at all stages of the product life cycle (i.e., at manufacturing, during
appliance use, and at appliance disposal).
This option applies to HFC-245fa in newly manufactured PU appliance foam. This option is not
applied in the European Union or in developing countries, because no HFC consumption is assumed in
this application in the baseline. One-time costs are estimated at approximately $4.8 million per facility for
safety modifications, installation/retrofit of high-pressure foam dispensers, installation of systems storage
tanks, pumps, and premixing stations, as well as training, trials, testing, and certification (TEAP, 2012;
UNEP, 2011a). Incremental annual costs associated with replacement formulations are estimated at
approximately $1.6 million (UNEP, 2011a). These annual costs are more than offset by annual savings of
nearly $4.4 million associated with lower blowing agent costs.
IV.4.3.2 HCs in Commercial Refrigeration
This option replaces HFC-245fa used in commercial refrigeration foam with HCs, namely
cyclopentane and cyclopentane/isopentane blends. HCs are inexpensive and have near-zero direct GWPs.
Technical issues exist with using hydrocarbons, such as flammability and lower insulation performance,
but these can be overcome through proper safety controls and engineering design. A significant
advantage of hydrocarbons is that they can be easily blended to impact a range of properties, such as
thermal performance, cell gas pressure, and foam density, as well as cost (TEAP, 2012). Use of
cyclopentane and cydo/iso blends in commercial refrigeration has a particularly good balance between
foam properties and density. Such HC blends are associated with low operating costs and are well-
proven (UNEP, 2010; TEAP, 2012). Flammability, however, may cause a high incremental capital cost for
facilities, which may be uneconomic for small or medium-sized enterprises; otherwise, hydrocarbons
have low operating costs (TEAP, 2012).
This option applies to HFC-245fa in newly manufactured PU commercial refrigeration foam. This
option is not applied in developing countries, because no HFC consumption is assumed in this
application in the baseline. One-time costs are estimated at about $1.26 million per facility associated with
safety modifications, installation/retrofit of high-pressure foam dispensers, installation of hydrocarbon
storage systems, pumps, and premixing stations, as well as, safety audits, trials, and training (TEAP,
2012; UNEP, 2011b). Incremental annual costs associated with replacement formulations are estimated at
nearly $105,000 (UNEP, 2011b). These annual costs are more than offset by annual savings of about
$602,000 associated with lower blowing agent costs.
IV-66 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
IV.4.3.3 HC in Polyurethane Spray Foams
This option replaces HFC-245fa/CO2 used in PU spray foam with HCs, specifically an 80/20 blend of
cyclopentane and isopentane. HCs are inexpensive and have near-zero direct GWPs. However, it should
be noted that technical issues exist with using hydrocarbons, including flammability, which may render
this alternative unsafe in some spray applications (TEAP, 2012). Using HCs instead of HFC-245fa/CO2 in
PU spray foam would eliminate HFC emissions during application (first year) and over the product
lifetime (USEPA, 2009).
This option applies to all HFC-245fa/CO2 in newly manufactured PU spray foam. This option is not
applied in developing countries, because no HFC consumption is assumed in this application in the
baseline. One-time costs are estimated at $15,700 per contractor for new formulations and equipment
(e.g., spray nozzles), while annual operating costs are estimated at $45,200 for fire retardant, worker
safety training, and cost increases from blowing agent density change. These annual costs are offset by
annual savings of $50,400 associated with lower blowing agent costs.
IV.4.3.4 CO2 in Polyurethane Spray Foams
This option replaces HFC-245fa/CO2 blends used in PU spray foam with CO2 (water). In the process
of using CO2 (water) in foams, first isocyanate and the polyol or polyamine react to form a polymer,
which forms a solid. Water is introduced and a chemical reaction between the water and polymeric
isocyanate produces CO^ which is used as a blowing agent. Using CO2 (water) instead of HFC-245fa in
PU spray foams eliminates HFC emissions during the production and application stages and over the
product lifetime (USEPA, 2009). CO2 is considered to have moderate foam properties (due to its high
thermal conductivity and high density), and requires greater thickness that leads to a cost penalty
compared to other options (TEAP, 2012). The use of CO2 in this application is most predominant in Japan,
with reported use also in North America and Spain (UNEP, 2010).
This option applies to all HFC-245fa/CO2 in newly manufactured PU spray foam. This option is not
applied in developing countries, because no HFC consumption is assumed in this application in the
baseline. One-time costs are estimated at $4,600 per contractor (for new formulations and minimal
equipment modifications), while annual operating costs are estimated at $60,700 (for fire retardant and
the cost increase from blowing agent density change). These annual costs are partly offset by annual
savings of $10,700 associated with lower agent costs.
IV.4.3.5 LCD/Alcohol in XPS Boardstock
This option replaces the HFC-134a and CO2-based blends used in extruded polystyrene (XPS)
boardstock foam with liquid CO2 (LCD)/alcohol. LCD is blended with other foam components under
pressure prior to the initiation of the chemical reaction. When decompressed, the CO2 expands, resulting
in froth foam that further expands with the additional release of CO2 from the water/isocyanate resin
reaction that forms the PU foam matrix. Difficulties encountered in using LCD include the limited
solubility of the chemical mixture, controlled decompression, and distribution of the unavoidable froth
(USEPA, 2009). Foams blown with CO2 may suffer from lower thermal performance, lower dimensional
stability, and higher density versus fluorocarbon-blown foams (USEPA, 2009). To overcome these
limitations, CO2 can be blended with HCs or HFCs (Williams et al., 1999; Honeywell, 2000; Alliance,
2001).
This option applies to all HFC-134a/CO2 blends in newly manufactured XPS boardstock foam. This
option is not applied in developing countries, because no HFC consumption is assumed in this
application in the baseline. This analysis assesses the costs for the foam producer to replace an HFC-
134a/CO2-based blend with LCD/alcohol in one of 10 production lines. One-time costs are estimated at
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-67
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
$5,856,000 per facility (for equipment, safety, and incineration considerations), while annual operating
costs are estimated at $915,000 (for labor, energy, and lost capacity). These annual costs are offset by
annual savings of $4,770,000 associated with lower costs for agent and polystyrene resin.
IV.4.3.6 HFC-134a to HCs in PL) One-Component Foam
This option replaces HFC-134a used in PU one-component foam with HCs, specifically a 50/50 blend
of propane and butane. HCs are inexpensive and have near-zero direct GWPs. Technical issues exist with
using HCs, such as flammability and lower insulation performance, but these can be overcome through
proper safety controls and engineering design (USEPA, 2009). A significant advantage of hydrocarbons is
that they can be easily blended to impact a range of properties, such as thermal performance, cell gas
pressure, and foam density, as well as cost (TEAP, 2012). Use of butane and propane in one component
foams is well proven and is associated with low operating costs (TEAP, 2012). Using HCs instead of HFC-
134a in PU one-component foams eliminates HFC emissions during manufacturing and over an assumed
product lifetime of 1 year (USEPA, 2009).
This option applies to all HFC-134a in newly manufactured PU one-component foam in both
developed and developing countries. One-time costs are estimated at $399,0003 per facility (for capital
costs and safety equipment), while annual operating costs are estimated at $342,000 (for fire retardant and
worker safety training). These annual costs are offset by annual savings of $859,000 associated with lower
agent costs.
IV.4.3.7 HFC-152a to HCs in PU One-Component Foam
Similar to the option above, this option replaces HFC-152a used in PU one-component foam with
HCs, specifically a 50/50 blend of propane and butane.
This option applies to all HFC-152a in newly manufactured PU one-component foam. This option is
not applied in developing countries, because no baseline HFC-152a consumption is assumed in this
application. One-time costs are estimated at $399,000 per facility (for capital costs and safety equipment),
while annual operating costs are estimated at $342,000 (for fire retardant and worker safety training).
These annual costs are offset by annual savings of $409,000 associated with lower agent costs.
IV.4.3.8 HCs in PU Continuous and Discontinuous Foams
This option replaces HFC-134a used in PU continuous and discontinuous panel foam with HCs. HCs
are inexpensive and have near-zero direct GWPs. Some technical issues exist with using HCs, such as
flammability and lower insulation performance, but these can be overcome through proper safety
controls and engineering design (USEPA, 2009). A significant advantage of hydrocarbons is that they can
be easily blended to impact a range of properties, such as thermal performance, cell gas pressure, and
foam density, as well as cost (TEAP, 2012). Using HCs instead of HFC-134a in PU continuous and
discontinuous panel foam eliminates HFC emissions during the manufacturing stage, during the foam's
assumed 50-year lifetime, and at time of product disposal (USEPA, 2009).
This option is assumed to be applicable to all HFC-134a in newly manufactured PU continuous and
discontinuous panel foam. This option is not applied in developing countries, because no HFC
' In developing countries, it is assumed that capital costs are 10% higher than those in the United States.
IV-68 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
consumption is assumed in this application in the baseline. One-time costs are estimated at $319,000 per
facility (for capital costs and safety equipment), while annual operating costs are estimated at $2,490,000
(for fire retardant, worker safety training, and changes in foam density). These annual costs are offset by
annual savings of $2,937,000 associated with lower agent costs.
IV.4.3.9 Manual Blowing Agent Recovery from Appliances at End of Life (EOL)
In most countries,4 foams contained in appliances and other products typically end up in landfills,
where the remaining blowing agent still contained in the insulation at the end of the product's life is
released during shredding and compaction or slowly over time (CARB, 2011). This option involves
manual labor to disassemble appliances and remove the foam in large pieces; the recovered foam is then
sealed in plastic bags to prevent further off-gassing of the blowing agent and subsequently sent for
incineration in a waste-to-energy plant. This practice is currently being used in the United States and
Canada, where voluntary programs and/or demand-side management programs are in place to
encourage the safe disposal of inefficient appliances (CARB, 2011).
This abatement option applies to existing domestic refrigerators reaching disposal. This option is not
applied in the European Union or in developing countries, because no baseline HFC consumption is
assumed in PU appliances (further, recovery of foam at appliance equipment EOL is assumed in the EU
baseline). One-time costs are estimated at $1 million per facility (for automated saws), while net annual
operating costs are estimated at $4,865,000 per facility for labor and handling costs (CARB, 2011).
IV.4.3.10 Fully Automated Blowing Agent Recovery from Appliances at EOL
Similar to the previous option, this option involves the recovery of foam at the end of the appliance's
life, but instead using fully automated appliance dismantling machines that separate all components,
including the foam-blowing agent. The blowing agent is then reconcentrated and sent to a destruction
facility approved to destroy ODS, while the remaining foam fluff is typically sent to a landfill. Fully
automated appliance recycling technologies can handle an estimated annual throughput of 150,000 to
250,000 units (CARB, 2011).
This abatement option applies to existing domestic refrigerators reaching disposal. This option is not
applied in the European Union or in developing countries, because no baseline HFC consumption is
assumed in this application. One-time costs are estimated at $5,000,000 per facility (for the fully
automated unit), while net annual operating costs are estimated at $6,130,000 per facility for labor,
handling, and electricity costs (CARB, 2011).
IV.4.3.11 Solstice Liquid Blowing Agent in PU Foams
Solstice Liquid Blowing Agent5 produced by Honeywell (also referred to as Solstice LBA, Solstice
1233zd(E), or Trans-l-chloro-3,3,3-trifluoroprop-l-ene) is under development/in early commercialization
4 Foam recovery from disposed appliances is already mandatory in a number of countries, including Japan and the
European Union.
5 Other unsaturated HFCs and HCFCs with low GWPs are being developed by DuPont and Arkema, among others,
for use in PU and other foam applications, which are likely to be commercialized in the coming years. Although these
compounds are not yet commercialized, they may be expected to have similar GWPs and applications as the Solstice
compounds described here.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-69
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
to replace HFC-134a and HFC-245fa used in a range of PU foam applications, including appliance foam,
commercial refrigeration foam, continuous and discontinuous panel foam, and spray foam (TEAP, 2012).
The blowing agent has been approved in the United States, Japan, European Union (up to 10MT), India,
and Asia Pacific, South America, Central America, Middle East, and Africa regions (Honeywell, 2012a).
The first commercial manufacture is expected to occur in 2013 (TEAP, 2012). Recently, Whirlpool
announced that by 2014, Solstice LBA will be used in all of the company's refrigerators produced in the
United States (Whirlpool, 2012). Solstice LBA has a GWP of 4.7 to 7 (USEPA, 2012b). The foam blowing
agent is also considered to be nonflammable (Honeywell, 2012a), which could limit capital costs to
convert from an HFC to this product. Solstice LBA exhibits relatively high performance and is considered
a drop-in alternative to HFCs, with no additional capital costs (TEAP, 2012). Any significant cost is
expected to be the result of the incremental cost of the blowing agent, which is expected to range from
$ll/kg to $17/kg (TEAP, 2012; Williams, 2013), but which may be lower, especially once production
volume increases. Recent evaluations showed up to a 4% improvement in energy efficiency compared to
HFC-245fa in domestic refrigerators (TEAP, 2012; Honeywell, 2012a), and a 5% to 10% energy
performance improvement compared to HFC-245fa in spray and panel foams (Honeywell, 2012a),
making it an attractive and likely option for markets where thermal insulation properties of the foam are
important, such as in domestic refrigerators. With its low GWP, Solstice 1233zd(E) could be used in lieu
of HC abatement options analyzed above for PU foams applications and achieve similar emission
reductions, but at different costs. Because it is nonflammable and of similar properties as the HFCs it
would replace, it would avoid some of the barriers due to pressure and flammability that exist with the
other options, such as in PU spray foam, and hence may penetrate those markets further or faster than the
options currently analyzed. This option is not quantitatively assessed in this analysis but may be added
as a future update.
IV.4.3.12 Solstice Gas Blowing Agent in XPS Foam and One-Component Foam
Solstice Gas Blowing Agent6 (also referred to as Solstice GBA, Solstice 1234ze(E), and HFO-1234ze)
can replace HFC-134a used in PU one-component foam and is under development/in early
commercialization to replace HFC-134a in XPS foam beginning in 2013. The blowing agent has been
commercial since 2008, with initial sales starting in the EU and Japan; approval for sales in the United
States was awarded in 2011 (Honeywell, 2012a). The GWP of Solstice GBA is <6 and it is considered to be
nonflammable up to temperatures of 28°C (82.4°F) (TEAP, 2012). In XPS foam, Solstice GBA has shown
good insulation performance (i.e., energy efficiency) and compressive strength and dimensional stability
similar to HFC-134a, and allows extrusion of thick foam (Honeywell, 2012a), making it an attractive and
likely option for markets where such properties of the foam are important, such as in XPS foam. Solstice
GBA can be used as a near drop-in replacement for HFC-134a, and can be handled, transported, and
stored in the same manner (Honeywell, 2012b). Transition to Solstice GBA is anticipated to require
negligible capital costs (TEAP, 2012; Williams, 2013). Any significant cost is expected to be the result of
the incremental cost associated with the blowing agent, which is estimated to range from $ll/kg to $17/kg
(TEAP, 2012; Williams, 2013) but which may be lower, especially once production volume increases. With
its low GWP, HFO-1234ze(E) could be used in lieu of LCD/alcohol in XPS foam and hydrocarbons in one-
6 Other unsaturated HFCs and HCFCs with low GWPs are being developed by DuPont and Arkema, among others,
in XPS, one-component, and other foam, which are likely to be commercialized in the coming years. These
compounds may be expected to have similar GWPs and applications as the Solstice compounds described here.
IV-70 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
component foam and achieve similar emission reductions, but at different costs. Because it is
nonflammable and of similar properties as the HFCs it would replace, it would avoid some of the barriers
due to controlling decompression and flammability that exist with the other options, and hence may
penetrate those markets further or faster than the options currently analyzed. This option is not
quantitatively assessed in this analysis but may be added as a future update.
IV.4.3.13 Methyl Formate in PU and XPS Foams
Methyl formate may replace HFCs in commercial refrigeration, continuous and discontinuous panels,
spray foam, transport refrigeration foam, and XPS foam. It is currently used in PU spray foams
internationally, including Africa, Asia, Americas, Australia, China, and Europe (Ecomate, 2012). Methyl
formate was approved as a "Generally Recognized as Safe" substance by the U.S. Food and Drug
Administration (Ecomate, 2012) and has a negligible GWP (TEAP, 2012). It may be blended with polyols
to produce non-flammable blends that reduce conversion costs (TEAP, 2012). In commercial refrigeration
and panel foam, corrosion protection is recommended and may require moderate incremental capital
costs (TEAP, 2012). This option is reportedly associated with a 10% increase in operating costs due to the
need for higher densities to address foam instability (UNEP, 2010). Generally, use of this alternative does
not require large capital changes to facilities (Ecomate, 2012). One producer of rigid foams for refrigerated
transport applications in Brazil completed conversion to methyl formate within 3 years and has since
reported an increase in the productivity of the lines and reduced operational costs compared to HFC-134a
(Crestani, 2012). Relative to hydrocarbon systems, methyl formate is safer to handle, and has lower
shipping, handling, and storage costs (Ecomate, 2012). The GWP of methyl formate is similar to that of
the options examined in this report; hence, emission reductions would be similar. Due to a lack of readily
available cost information on this alternative, this option is not quantitatively assessed in this analysis.
IV.4.4 Engineering Cost Data Summary
Table 4-3 presents the engineering cost data for each mitigation option outlined above, including all
cost parameters necessary to calculate the break-even price.
Table 4-3: Engineering Cost Data on a Facility Basis
Abatement Option
Appliance: HFCs to
HCs
Commercial
refrigeration: HFCs to
HCs
Spray: HFC245fa/C02
toHC
Spray: HFC245fa/C02
toC02
XPS:HFC134a/C02to
LCD/alcohol
One-component:
HFC-134atoHCs-
Developed
Facility Type
PU appliance foam
manufacturing facility
PU commercial refrigeration
foam manufacturing facility
PU spray foam contractor
PU spray foam contractor
XPS boardstock production
facility
One-component foam
manufacturing facility
Project
Lifetime
(Years)
25
15
25
25
25
25
Capital Cost
(2010 USD)
$4,831,000
$1,260,000
$15,700
$4,600
$5,856,000
$399,000
Annual
Revenue
(2010 USD)
$4,375,000
$602,000
$50,400
$10,700
$4,770,000
$859,000
Annual O&M
Costs
(2010 USD)
$1,610,000
$105,000
$45,200
$60,700
$915,000
$342,000
Abatement
Amount
(tC02e)
509,951
66,500
54,654
54,654
9,168
169,603
(continued)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-71
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
Table 4-3: Engineering Cost Data on a Facility Basis (continued)
Abatement Option
One-component:
HFC-134atoHCs-
Developing
One-component:
HFC-152atoHCs
Continuous and
discontinuous:
HFC134atoHCs
Appliance EOL:
Manual process
Appliance EOL: Fully
automated process
Facility Type
One-component foam
manufacturing facility
One-component foam
manufacturing facility
PU continuous and
discontinuous foams
manufacturing facility
Appliance demanufacturing
facility using manual
blowing agent recovery
Appliance demanufacturing
facility using fully automated
blowing agent recovery
Project
Lifetime
(Years)
25
25
25
25
25
Annual
Capital Cost Revenue
(2010 USD) (2010 USD)
$438,900 $859,000
$399,000 $409,000
$319,000 $2,937,000
$1,000,000 -
$5,000,000 -
Annual O&M
Costs
(2010 USD)
$342,000
$342,000
$2,490,000
$4,865,000
$6,130,000
Abatement
Amount
(tC02e)
169,603
18,265
588,900
99,380
177,716
IV.4.5 Marginal Abatement Cost Analysis
IV.4.5.1
Methodological Approach
The analysis is based on the above representative project costs for model facilities. We applied the
costs to calculate the break-even prices for each appropriate option and facility or operation. The model
estimates the mitigation potential based on the percentage of the total ODS substitutes' baseline (that is,
the HFC emissions from sectors that historically used ODSs) attributable to each representative
facility/operation and the technical effectiveness for each technology in each facility.
IV.4.5.2
Assessment of Technical Effectiveness
The analysis also developed a technical effectiveness parameter, defined as the percentage reductions
achievable by each technology/facility type combination. Estimating this parameter requires making a
number of assumptions regarding the distribution of emissions by facility in addition to process-specific
estimates of technical applicability and market penetration. Market penetration is a modeled value that
takes into account the market's willingness to adopt the option, the rate of uptake of the alternative into
new foams, and the lifetime of the existing foam base. Because foam lifetimes can be decades, replacing
the stock of foams with non-HFC blowing agents will take many years. The market penetration rate is
modeled to capture such time lapses. Technical effectiveness figures do not account for indirect
greenhouse gas impacts associated with changes in electricity consumption (e.g., for foam blowing
processes or for end-of-life appliance processing), which are accounted for in the cost analysis.7 Table 4-4
summarizes these assumptions and presents technical effectiveness parameters used in the MAC model.
7 Indirect greenhouse gas emissions are not accounted for in the technical effectiveness calculations so that the
analysis can show purely ODS substitute (i.e., HFC) emission reductions achievable. It is recognized that indirect
GHG emissions can be significant, and such differences, to the extent data is available on such, are accounted for in
the cost analyses.
IV-72
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
Table 4-4: Technical Effectiveness Summary
PeSatn
Facility/Abatement Option
Appliance: MFCs to HCs-U.S./Other
developed
Commercial refrigeration: MFCs to
HCs— Developed
Spray: HFC245fa/C02 to HC-Developed
Spray: HFC245fa/C02 to C02-Developed
XPS: HFC134a/C02to LCD/Alcohol-
U.S./Other developed
XPS: HFC134a/C02to LCD/Alcohol-EU
One-component: HFC-134a to HCs—
Developed & developing
One-component: HFC-152a to HCs—
Developed
Continuous and Discontinuous: HFC134a
to HCs— Developed
Appliance EOL: Manual process—
U.S./Other developed
Appliance EOL: Fully automated
process— U.S./Other developed
Technical Applicability Rate (2030)a
New PU appliances
New PU commercial
refrigeration units
New spray foam applications
New spray foam applications
New XPS boardstock foam
applications
New XPS boardstock foam
applications
New PU one-component foam
applications
New PU one-component foam
applications
New PU continuous and
discontinuous foam applications
Domestic refrigerators reaching
end of life
Domestic refrigerators reaching
end of life
100%
100%
30%
70%
75%
75%
100%
100%
100%
50%
20%
Technical
Reduction Effectiveness
Efficiency (2030)"
100%
100%
100%
100%
100%
100%
100%
100%
100%
85%
95%
37%
39%
9%
22%
66%
55%
94%
6%
49%
29%
12%
a Market penetration assumptions for this analysis vary over time, and the technical effectiveness values are based on the cumulative market
penetration rates assumed until that point. More information on the market penetration assumptions is provided in the appendix F to this
chapter.
b Technical effectiveness figures represent the percent of baseline emissions from the relevant facility type that can be abated in 2030; figures
do not account for indirect greenhouse gas impacts associated with increased electricity consumption (e.g., for foam blowing processes or for
end-of-life appliance processing), which are accounted for in the cost analysis.
IV.4.5.3
Estimating Abatement Project Costs and Benefits
Table 4-5 provides examples of the break-even prices calculated for each abatement measure. Project
costs and benefits are calculated for model facilities and are used in the calculation that solves for the
break-even price that sets the project's benefits equal to its costs. The previous section describes the
assumptions used to estimate costs for each technology for applicable facilities. Additional details on the
analyses can be found in Appendix F to this chapter.
The break-even prices presented in Table 4-5 represent model facilities. Actual prices vary by country
because of the scaling of costs and benefits by international price factors. Complete international MAC
results are presented in Section IV.4.5.4.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-73
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
Table 4-5: Example Break-Even Prices for Abatement Measures in Foams Manufacturing
Abatement Option
Tax Benefit
Reduced Annualized Net Annual of Break-
Emissions Capital Costs Cost Depreciation Even Price
(tC02e) ($/tC02e) ($/tC02e) ($/tC02e) ($/tC02e)
Appliance: MFCs to HCs
Commercial refrigeration: MFCs to HCs
Spray: HFC245fa/C02 to HC
Spray: HFC245fa/C02 to C02
XPS: HFC134a/C02to LCD/Alcohol
One-component: HFC-134a to HCs—
Developed
One-component: HFC-134a to HCs—
Developing
One-component: HFC-152a to HCs
Continuous and discontinuous: HFC134ato
HCs
Appliance EOL: Manual process
Appliance EOL: Fully automated process
509,951
66,500
54,654
54,654
9,168
169,603
169,603
18,265
588,900
99,380
177,716
1.7
4.2
0.1
0.0
117.3
0.4
0.5
4.0
0.1
1.8
5.2
-5.4
-7.5
-0.1
0.9
-420.5
-3.0
-3.0
-3.7
-0.8
49.0
34.5
0.3
0.8
0.0
0.0
17.0
0.1
0.1
0.6
0.0
0.3
0.8
-3.9
-4.2
-0.1
0.9
-320.2
-2.7
-2.6
-0.2
-0.7
50.5
38.9
IV.4.5.4
MAC Analysis Results
Global abatement potential in 2020 and 2030 is 13.7 and 37.0 MtCO2e, respectively. There are 27.0
MtCO2e of reductions available in 2030 resulting from implementing currently available technologies that
are economical at projected costs. If an additional emissions reduction value (e.g., tax incentive, subsidy,
or tradable emissions reduction credit) above the zero break-even price were available to
manufacturers/users of foams, then additional emission reductions could be cost-effective. The results of
the MAC analysis are presented in Table 4-6, which shows abatement potential by major country and
regional grouping at select break-even prices in 2030; Figure 4-4 illustrates the marginal abatement cost
curves of the top five emitting countries.
IV.4.6
Uncertainties and Limitations
Available data on costs for abatement technologies were not scaled to reflect potential differences in
the costs outside of the United States. Additional research may be required to determine actual variability
in costs across regions. Moreover, the development of alternative blowing agents is quickly evolving;
thus, new options may enter the market, which should be considered quantitatively in future analyses
(e.g., unsaturated fluorinated compounds and methyl formate).
IV-74
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
Table 4-6: Abatement Potential by Country/Region at Selected Break-Even Prices in 2030 (MtC02e
Top 5 Emitting Countries
France
Germany
Italy
Japan
United States
0.2
0.2
0.1
3.1
3.7
0.2
0.2
0.1
3.1
3.7
0.3
0.4
0.3
10.5
12.6
0.3
0.4
0.3
10.7
12.8
0.3
0.4
0.3
10.7
12.8
0.3
0.4
0.3
10.7
12.8
0.3
0.4
0.3
10.7
12.8
0.3
0.4
0.3
10.7
12.8
0.3
0.4
0.3
11.8
14.1
0.3
0.4
0.3
14.6
17.5
0.3
0.4
0.3
14.6
17.5
Rest of Region
Africa — — _________
Central and South America — — _________
Middle East — — _________
Europe
Eurasia
Asia
North America
World Total
0.4
-
0.1
0.3
8.1
0.4
-
0.1
0.3
8.1
1.1
0.5
0.5
1.0
27.0
1.2
0.5
0.5
1.0
27.7
1.2
0.5
0.5
1.0
27.7
1.2
0.5
0.5
1.0
27.7
1.2
0.5
0.5
1.0
27.7
1.2
0.5
0.5
1.0
27.7
1.2
0.5
0.5
1.1
30.2
1.3
0.5
0.6
1.4
37.0
1.3
0.5
0.6
1.4
37.0
Figure 4-4: Marginal Abatement Cost Curves for Top Five Foam Emitters in 2030
$50
o
(J
10
]__[
15
Non-CO2 Reduction (MtCO2e)
•France
20
•Germany
•Japan
•United States
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-75
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
References
Alliance for Responsible Atmospheric Policy. 2001. Review of EPA Draft Chapter 9 by Members of the
Alliance for Responsible Atmospheric Policy. Alliance. May 16, 2001.
Caleb. 2000. In response to Proposed Rulemaking 65 Fed. Reg. 42653, July llth 2000. Caleb Management
Services Limited, UK (p. 3).
CARB. September 2011. Lifecycle Analysis of High Global Warming Potential Greenhouse Gas Destruction.
Prepared by ICF International.
Crestani, Felipe (Folle/Bantech-Randon). 2012. Alternative Technologies in Rigid Foam for Transportation. The
Folk Conversion Case. Advancing Ozone and Climate Protection Technologies: Next Steps Conference.
July 20-21, 2012. Bangkok, Thailand. Summary available at
http://www.unep.org/ccac/Portals/24183/docs/Bangkok%20Technology%20Conference%20-
%20Report%20and%20Cover%20-%20FINAL.pdf.
Ecomate. 2012. Cost Effective Blowing Agent for Polyurethane Foam Applications. Advancing Ozone and
Climate Protection Technologies: Next Steps Conference. July 20-21, 2012. Bangkok, Thailand.
FTOC. 2010. 2020 Rigid and Flexible Foams Report. Available at: http://ozone.unep.org/teap/Reports/
FTOC/FTOC-2010-Assessment-Report.pdf.
Honeywell. 2000. Comments of Honeywell Inc. on U.S. Environmental Protection Agency Proposed
Listing of Certain HCFCs and Blends as "Unacceptable" Substitutes for HCFC-141b-65 Fed. Reg.
42653 (11 July 2000). Personal communication from Richard Ayers of Howrey, Simon, Arnold, and
White to Anhar Karimjee of EPA on September 11, 2000. Available from EPA's Foams Docket A-200-
18, Document IV-D-41.
Honeywell (Williams, D.). 2012a. Solstice Blowing Agents Low Climate Impact, High Energy Efficiency.
Advancing Ozone and Climate Protection Technologies: Next Steps Conference. July 20-21, 2012.
Bangkok, Thailand.
Honeywell. 2012b. Honeywell Low Global Warming Technologies: One-Component Foam—HFO-1234ze. 2004-
2012. Available at: http://www51.honeywell.com/srn/lgwp-uk/applications-n2/one-component-foam-
HFO-1234ze.html.
Technology and Economic Assessment Panel (TEAP). May 2012. Decision XXIII/0 Task Force Report:
Additional Information on Alternatives to Ozone-Depleting Substance. Volume II.
United Nations Environment Programme (UNEP). 2010. Guidance on the Process for Selecting Alternatives to
HCFCs in Foams Sourcebook on Technology Options for Safeguarding the Ozone Layer and the Global Climate
System. Prepared by Caleb Management Services.
United Nations Environment Programme (UNEP). 2011a. Project Proposal: China. China HCFC Phase-out
Management Plan. UNEP/OzL.Pro/ExCom/64/39. June 15, 2011. Available at:
http://www.multilateralfund.Org/MeetingsandDocuments/currentmeeting/64/English/l/6429.pdf.
United Nations Environment Programme (UNEP). 2011b. Project Proposal: Mexico. Mexico HCFC Phase-
out Management Plan. UNEP/OzL.Pro/ExCom/64/39. June 17, 2011. Available at:
http://www.multilateralfund.Org/MeetingsandDocuments/currentmeeting/64/English/l/6439.pdf.
U.S. Environmental Protection Agency (USEPA). June 2006. Global Mitigation ofNon-CO2 Greenhouse Gases.
U.S. EPA M30-R-06-005. Washington, DC: Office of Atmospheric Programs, U.S. Environmental
Protection Agency.
U.S. Environmental Protection Agency (USEPA). October 30, 2009. 2009 Marginal Abatement Cost Curve
Analysis for Reduction of HFCs in Traditional Ozone Depleting Substance (ODS) End-Use Applications:
Draft Report. Prepared by ICF International.
U.S. Environmental Protection Agency (USEPA). (2012a). Global Anthropogenic Non-COi Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA. Obtained from:
http://www.epa.gov/climatechange/economics/international.html.
IV-76 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM FOAMS MANUFACTURING
U.S. Environmental Protection Agency (USEPA). August 10, 2012b. Protection of Stratospheric Ozone:
Determination 27 for Significant New Alternatives Policy Program. Federal Register. Vol. 77, No. 155.
http://www.gpo.gov/fdsys/pkg/FR-2012-08-10/pdf/2012-19688.pdf.
Whirlpool. August 15, 2012. Whirlpool Corporation and Honeywell Introduce Most Environmentally
Responsible and Energy Efficient Insulation Available into U.S. Made Refrigerators. Available at:
http://honeywell.com/News/Pages/Whirlpool-Corp-and-Honeywell-Introduce-Most-
Environmentally-Responsible-and-Energy-Efficient-Insulation-Available-Refridge.aspx.
Williams, D. January 2013. Personal communication between Jenny Tanphanich, ICF International, and
Dave Williams, Honeywell.
Williams, D.J., M.C. Bogdan, and P.B. Logsdon. 1999. Optimizing Performance and Value: HFC-245fa and
Blends of HFC-245fa for Insulating Foams. Conference Proceedings from the Earth Technologies Forum
TF 1999, pp. 290-302.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-77
-------
HFC EMISSIONS FROM AEROSOL PRODUCT USE
IV.5. HFC Emissions from Aerosol Product Use
erosol propellant formulations containing HFCs are used in a variety of consumer products,
such as spray deodorants and hair sprays, and specialty aerosol uses, such as freeze spray
and dust removal products. Additionally, aerosol propellants containing HFCs are used in
pharmaceutical products, primarily metered dose inhalers (MDIs).
Global HFC emissions from the aerosols sector were estimated at roughly 45 million metric tons of
carbon dioxide equivalent (MtCC^e) in 2010 and are expected to climb to 146 MtCC^e by 2030 as shown
in Figure 5-1. A majority of the growth is attributed to an increase in the consumption of HFCs for aerosol
applications in developing countries.
Figure 5-1: HFC Emissions from Aerosol Product Use: 2000-2030 (MtC02e)
146
I Mexico
I Russia
India
I United States
I China
ROW
2000
2010
2020
2030
Year
Source: USEPA, 2012.
A variety of abatement measures are available to reduce emissions. For consumer aerosol products,
the options include transitioning to replacement propellants with very low global warming potentials
(GWPs) and converting to a not-in-kind (NIK) alternative, such as a stick, roller, or finger/trigger pumps.
Dry powder inhaler (DPI) technology is considered in this analysis as a replacement measure for MDIs.
The global abatement potential from aerosols is equal to approximately 66% of total annual emissions
from the aerosols sector and 5% of total annual emissions from all sectors that use ODS substitutes.
Potentially, nearly 100% abatement is possible for consumer aerosol products; whereas abatement is more
limited for MDIs due to medical reasons (e.g., DPIs are not suitable for cases of severe asthma or for
young children). Marginal abatement cost (MAC) curve results are presented in Figure 5-2. Maximum
abatement potential in the aerosols sector is 97 MtCO2e in 2030. There are 70 MtCO2e of emissions
reductions that are cost-effective at currently projected energy prices for 2030.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-79
-------
HFC EMISSIONS FROM AEROSOL PRODUCT USE
Figure 5-2: Global Abatement Potential in Aerosol Product Use: 2010, 2020, and 2030
8
-$30
7
H h
£
Non-CO2 Reduction (MtCO2e)
•2010
2020
•2030
/U
80
90
IV.5.2
Emissions from Aerosol Product Use
Aerosol propellants are completely emitted during product use. This analysis estimates emissions
from consumer aerosol products, such as spray deodorants and hair sprays, and specialty aerosol uses,
such as freeze spray and dust removal products, and, separately, emissions from MDIs. HFC-134a has
been introduced as an alternative propellant to chlorofluorocarbons (CFCs) in both MDIs and consumer
aerosol products; in addition, HFC-227ea is used in MDIs and HFC-152a is used in consumer aerosol
applications. This analysis characterizes these three categories, for the purposes of evaluating the cost of
reducing HFC emissions, as follows:
• a facility that produces 10 million consumer aerosol cans per year, with each can containing an
HFC-134a aerosol propellant charge of two ounces;
• a facility that produces 10 million consumer aerosol cans per year, with each can containing an
HFC-152a aerosol propellant charge of two ounces; and
• a single 200-dose MDI aerosol unit with a charge size of 15 grams that uses HFC-134a propellant.
The relative shares of these applications are displayed in Figure 5-3.
IV-80
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM AEROSOL PRODUCT USE
Figure 5-3: Global HFC Emissions in 2020 by Facility Type (% of GWP-Weighted Emissions)
IV.5.2.1 Activity Data or Important Sectoral or Regional Trends
HFC emissions from the aerosols sector represented 10% of total ODS substitute emissions in 2010.
Emissions were estimated based on estimated market size and charge size of aerosol products in the
United States, which was then proxied to other regions based on each country's gross domestic product.
Growth in aerosol products correlates with economic growth, and because of the global, commoditized
nature of this sector, there are no significant regional differences in the aerosol products themselves.
Aerosols are fully emissive, and the HFCs contained within the aerosol are assumed to be emitted within
a year of consumption. Many non-MDI aerosols already use alternative propellants, either hydrocarbons
(HCs) or manual pump mechanisms; thus, the baseline adoption of the reduction technologies is quite
high in the market.
IV.5.2.2
Emission Estimates and Related Assumptions
Global emissions of HFCs from the aerosols sector were estimated to be 45 MtCO2e in 2010, growing
to 146 MtCO2e in 2030, as shown in Table 5-1. The majority of emissions are HFC-134a, with lesser
amounts of HFC-152a and HFC-227ea. Emissions are projected to grow significantly as developing
countries' economies grow and demand for consumer aerosols grows. Emissions were estimated using
USEPA's Vintaging Model and assumptions about initial market size of the sector, the specific transitions
away from CFCs in terms of timing and alternative propellant used, charge sizes, and leak rates.
IV.5.3 Abatement Measures and Engineering Cost Analysis
A total of five abatement options were identified for the aerosols sector. For consumer aerosol
products, the options are transitioning to a replacement propellant—including HCs, HFO-1234ze, and
HFC-152a (for those products containing HFC-134a)—and converting to a NIK alternative, such as a stick,
roller, or finger/trigger pump. Costs are analyzed by looking at a model facility that uses HFC to fill 10
million aerosol cans a year. For MDIs, the abatement measure examined by this analysis is further use of
dry powder inhaler technology where suitable for the patient. Costs are analyzed based on a single DPI
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-81
-------
HFC EMISSIONS FROM AEROSOL PRODUCT USE
Table 5-1: Projected Baseline Emissions from Aerosol Product Use: 2010-2030 (MtC02e)
Country/Region
2010
2015
2020
2025
2030
CAGR*
(2010-2030)
Top 5 Emitting Countries
China
United States
India
Russia
Mexico
9.4
8.9
2.4
2.6
1.9
23.0
11.9
5.7
3.9
3.7
31.2
13.0
7.6
4.6
4.2
42.2
14.2
9.9
5.4
4.8
56.9
15.6
13.0
6.2
5.4
9.4%
2.8%
8.8%
4.4%
5.3%
Rest of Regions
Africa
Central and South America
Middle East
Europe
Eurasia
Asia
North America
World Total
3.0
1.3
1.8
9.5
0.4
4.0
0.2
45.5
6.3
2.6
3.7
14.6
0.5
7.7
0.4
84.2
7.6
3.0
4.4
15.1
0.6
8.7
0.4
100.4
8.9
3.5
5.1
15.6
0.7
9.9
0.5
120.7
10.4
4.0
6.0
16.0
0.8
11.2
0.5
145.8
6.3%
5.7%
6.2%
2.6%
4.4%
5.3%
4.6%
6.0%
a CAGR = Compound Annual Growth Rate
Source: U.S. Environmental Protection Agency (USEPA), 2011
compared to a single MDI, with estimated cost data that incorporates the cost associated with avoided
use of HFC-134a propellant, the increase in the cost of DPI treatment, the cost to market the new
treatment, and the cost to retrain the patients in using the DPI (Ecofys, 2000; Enviros, 2000). Table 5-2
summarizes the applicability of each abatement option to the aerosol emission categories. The subsequent
subsections describe each abatement option in more detail.
IV.5.3.1
Hydrocarbons
This option replaces HFC-134a or HFC-152a in non-MDI aerosols with an HC-based propellant. HC
aerosol propellants are usually mixtures of propane, butane, and isobutane.1 Their primary advantage
lies in their affordability; the price of HC propellants which range from one-third to one-half that of
HFCs. The main disadvantages of HC aerosol propellants are flammability concerns and, because they
are volatile organic compounds (VOCs), their contribution to ground-level ozone and smog. Despite
these concerns, HC aerosol propellants already hold a sizable share of the market and may be acceptable
for additional applications.
1 For calculation purposes, a GWP of 3.48 is used based on an average of the GWP of propane (GWP=3.3) and
isobutane (GWP=3.65).
IV-82
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM AEROSOL PRODUCT USE
Table 5-2: Aerosol Product Use Abatement Options
Reduction Efficiency
Consumer Consumer
Aerosol Aerosol Metered
Facility/ Facility/ Dose
Abatement Option HFC-134a HFC-152a Inhaler
Applicability
Consumer Aerosol Products
Hydrocarbon
Not-in-kind
HFO-1234ze
HFC-134atoHFC-152a
Pharmaceutical Aerosol Products
Dry powder inhalers
99.7%
100%
99.5%
89.2%
(MDIs)
NA
97.5%
100%
95.7%
NA
NA
NA
NA
100%
Consumer aerosol facility/HFC-134a/HFC-152a
Consumer aerosol facility/HFC-134a/HFC-152a
Consumer aerosol facility/HFC-134a/HFC-152a
Consumer aerosol facility/HFC-134a
Metered dose inhaler
Costs of converting filling facilities to accept HC propellants can range from $10,000 to potentially as
high as $1.2 million; the one-time cost varies based on the need for investments in new equipment and
the need to relocate to regions where the use of HCs is considered safe (Nardini, 2002). To accommodate
any flammable propellant, a company is required to build a storage tank to house the product. This tank
will need to be connected to the main facility through a plumbing system (Techspray, 2008; MicroCare,
2008). According to discussions with industry, the majority of companies would already have fire
insurance and other fire safety precautions intact; therefore, no significant additional costs would be
associated with housing a flammable chemical, and the increase in annual costs would be zero
(Techspray, 2008; MicroCare, 2008). Given that HCs (estimated at $l/lb) are lower cost than HFC-134a or
HFC-152a (estimated at $3/lb and $2/lb, respectively), the adoption of this abatement measure is expected
to result in an annual savings associated with gas purchases, ranging from $1 million to nearly $3 million.
IV.5.3.2
Not-in-Kind
NIK aerosol devices include finger/trigger pumps, powder formulations, sticks, rollers, brushes,
nebulizers, and bag-in-can/piston-can systems. These systems often prove to be a better and more cost-
effective option than HFC-propelled aerosols, particularly in areas where a unique HFC property is not
specifically needed. Because all of the HFC (either HFC-134a or HFC-152a) contained in the aerosol can is
replaced with a device that does not use any GHGs, the reduction efficiency of this option is 100%.
Significant variability exists in financial components of projects targeting NIK replacements for HFC-
containing aerosol products. This variability is attributable to the wide range of potential aerosol and NIK
product types. A one-time cost to make the conversion is estimated at $250,000; additionally, annual costs
of $500,000 are estimated to address higher material costs of the particular sticks, rollers, and pumps
being used (UNEP, 1999). An annual savings is expected, ranging from $2.3 million to $4.1 million, as a
result of eliminating the need for a HFC propellant.
IV.5.3.3
HFO-1234ze
HFO-1234ze has potential application both as a propellant and also as the active ingredient in aerosol
dusters. HFO-1234ze is nonflammable (at room temperature) and has physical properties that are very
similar to both the HFC-134a and HFC-152a. Hence, it may be used as a 'drop-in' replacement for HFC
propellants (MicroCare, 2011). The manufacturer of this chemical indicates that Europe and Japan have
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-83
-------
HFC EMISSIONS FROM AEROSOL PRODUCT USE
already begun to adopt HFO-1234ze, while interest is also rising in the United States because of
awareness of environmental sustainability (Honeywell, 2011a). A number of dusters using HFO-1234ze
are available today (Amazon, 2013; ITW Chemtronics, 2013; Miller Stephenson, 2013; Stanley Supply and
Services, 2013). A large scale production facility is being built in the United States with an expected
production of HFO-1234ze in late 2013 (Honeywell, 2011b). In the absence of regulations, adoption in
Europe and Japan is expected to grow continuously at a moderate rate (reaching a maximum of 15% to
20% of today's HFC volume); therefore, this option is expected to penetrate up to 15% of the non-MDI
HFC-134a market and up to 20% of the non-MDI HFC-152a market. In the United States, adoption of
HFO-1234ze is expected to follow a similar path, but with a later start. In developing countries, no
interest in HFO-1234ze is expected in the foreseeable future because of inexpensive options that are the
preferred solutions today.
For this analysis, a one-time cost of roughly $500,000 is assumed because of the need for bulk storage.
According to MicroCare (2011), although it is possible to use one ton cylinders and avoid the costs of
adding bulk storage, using ton cylinders is inefficient and adds to the unit cost of the HFO material.
Therefore, any facility using this material would almost certainly need to use bulk storage. This is likely a
conservative (high) one-time cost estimate, considering it is about the same capital cost considered in the
next section for a flammable propellant, whereas HFO-1234ze(E) is not flammable at room temperatures.
Because HFO-1234ze has a higher cost than the other HFCs (i.e., HFC-134a and HFC-152a), a facility
making the transition would incur a higher annual cost when adopting this propellant.
IV.5.3.4 HFC-134a to HFC-152a
This abatement measure examines replacing HFC-134a (with a GWP of 1,300) with HFC-152a (with a
GWP of 140). HFC-134a is assumed to represent 58% of non-MDI aerosols; therefore, this abatement
option is only applicable to 58% of the non-MDI aerosol model facilities. HFC-134a is the primary
nonflammable propellant in certain industrial products. HFC-152a possesses only moderate flammability
hazards and might, therefore, be acceptable for some applications that use HFC-134a, but it may present
problems for other applications.
Costs of converting filling facilities to accept HFC-152a may range from $500,000 to $600,000,
(Techspray, 2008; MicroCare, 2008). To accommodate HFC-152a (or any flammable propellant), a
company is required to build a storage tank to house the product. This tank will need to be connected to
the main facility through a plumbing system (Techspray, 2008). Aside from the costs associated with
building a storage house, no other significant expenses would be incurred. According to discussions with
industry, the majority of companies would already have fire insurance and other fire safety precautions
intact; therefore, no significant additional costs would be associated with housing a flammable chemical,
and the increase in annual costs would be zero (Techspray, 2008; MicroCare, 2008). The lower cost of
HFC-152a (compared with HFC-134a) results in an annual savings associated with gas purchases,
estimated at $1.8 million for a typical aerosol filling facility.
IV.5.3.5 Dry Powder Inhalers
Dry powder inhalers (DPIs) are a viable abatement measure for most anti-asthma drugs, although
they are not successful with all patients or all drugs. Micronised dry powder, that contains the drug
agent, is contained in the DPI, a non-pressurized delivery system, and is inhaled and deposited in the
lungs. They are suitable only in patients who are able to inhale robustly enough to transport the powder
to the lungs. DPIs are not suitable for persons with severe asthma or for young children. Unlike MDIs,
powdered drug particles contained in DPIs tend to aggregate and may cause problems in areas with hot
and humid climates. Other issues that doctors and patients consider when choosing a treatment device
IV-84 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM AEROSOL PRODUCT USE
include the patient's manual dexterity, ability to adapt to a new device, and perception of the
effectiveness of the medicine and taste of any added ingredients. Ultimately, these and other critical
patient care issues must be assessed by the doctor and patient in choosing whether a DPI, MDI or other
type of therapy is most appropriate (Price et al., 2004; UNEP, 2010). Where feasible, DPIs—which do not
contain GHGs—could be used in lieu of HFC-containing MDIs; hence, the reduction efficiency of this
option is 100%.
Costs incurred with using DPIs include the increased cost of DPI treatment, the cost to market the
new treatment, and the cost to retrain patients to use the DPI (Enviros, 2000). The cost of research and
development of new drugs and delivery mechanisms can also be significant; however, as both MDIs and
DPIs are available to the market today, and due to a lack of specific cost information, such expenses are
not considered in this analysis.
IV.5.3.6
Engineering Cost Data Summary
Table 5-3 presents the engineering cost data for each mitigation option outlined above, including all
cost parameters necessary to calculate the break-even price. For more detailed costs information see
Appendix G.
Table 5-3: Engineering Cost Data on a Facility Basis
Abatement
Option
Consumer Aerosol Products
Project Capital Annual Annual Abatement
Lifetime Cost Revenue O&M Costs Amount
(years) (2010 USD) (2010 USD) (2010 USD) (tC02e)
HCs
Consumer aerosol/HFC- 134a
Consumer aerosol/HFC- 152a
10
$325,000
$2,800,000
$1,000,000
735,130
77,410
Not-in-Kind
Consumer aerosol/HFC-134a
Consumer aerosol/HFC-152a
$4,100,000 737,100
10 $250,000 $500,000
$2,300,000 79,380
HFO-1234ze
Consumer aerosol/HFC-134a
Consumer aerosol/HFC-152a
10
$500,000
$1,400,000 730,312
$3,200,000 72,722
HFC-134ato
HFC-152a
Consumer aerosol/HFC-134a
10
$500,000 $1,800,000
586,889
Pharmaceutical Aerosol Products (MDIs)
DPIs
Metered dose inhaler
10
$700,000
1,300
IV.5.4 Marginal Abatement Costs Analysis
IV.5.4.1
Methodological Approach
The analysis is based on the above representative project costs for model facilities. We applied the
costs to calculate the break-even prices for each appropriate option for each country. The model estimates
the mitigation potential based on the percentage of the total ODS substitutes baseline attributable to each
representative facility and the technical effectiveness for each technology in each facility.
IV.5.4.2
Assessment of Technical Effectiveness
The analysis also developed a technical effectiveness parameter, defined as the percentage reductions
achievable by each technology/facility type combination. Market penetration rates vary over time as
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-85
-------
HFC EMISSIONS FROM AEROSOL PRODUCT USE
systems are upgraded in the future. Table 5-4 summarizes the assumptions regarding technical
applicability, market penetration, and technical effectiveness of each option.
Table 5-4: Technical Effectiveness Summary
r-'*Fl'MllLHlH*jM'IWll
Technical
Applicability
Market
Penetration Rate
(2030)
Reduction
Efficiency
Technical
Effectiveness
(2030)
Consumer Aerosol Products
HFC-134atoHC
HFC-134atoNIK
HFC-134atoHFO-1234ze
HFC-134atoHFC-152a
HFC-152atoHC
HFC-152atoNIK
HFC-152atoHFO-1234ze
Pharmaceutical Aerosol Products
DPI
58%
58%
58%
58%
42%
42%
42%
(MDIs)
100%
50%
20%
15%
15%
20%
60%
20%
20%
99.7%
100.0%
99.5%
89.2%
97.5%
100.0%
95.7%
100.0%
29.0%
11.6%
8.7%
7.8%
8.2%
25.1%
8.0%
20.0%
a Technical effectiveness (TE) is the product of TA*MP*RE.
IV.5.4.3
Estimating Abatement Project Costs and Benefits
Table 5-5 provides an example of how the break-even prices are calculated for each abatement
measure. Project costs and benefits are calculated for model facilities and are used in the calculation that
solves for the break-even price that sets each project's benefits equal to its costs. The previous section
describes the assumptions used to estimate different costs for different facilities.
The break-even prices presented in Table 5-5 represent model facilities. Actual prices vary by country
because of the scaling of costs and benefits by international price factors. Complete international MAC
results are presented in Section 5.4.4.
Table 5-5: Example Break-Even Prices for Abatement Measures in Aerosol Product Use
Abatement Option
Reduced
Emissions
(tC02e)
Annualized
Capital Costs
($/tC02e)
Net Annual
Cost ($/tC02e)
Tax Benefit of
Depreciation
($/tC02e)
Break-Even
Price
($/tC02e)
Consumer Aerosol Products
HFC-134atoHC
HFC-134atoNIK
HFC-134atoHFO-1234ze
HFC-134atoHFC-152a
HFC-152atoHC
HFC-152atoNIK
HFC-152atoHFO-1234ze
Pharmaceutical Aerosol Products
DPI
735,130
737,100
730,312
586,889
77,410
79,380
72,722
(MDIs)
1,300
0.1
0.1
0.2
0.2
1.1
0.9
1.9
0.0
-3.8
-4.9
1.9
-3.1
-12.9
-22.7
44.0
538.5
0.0
0.0
0.0
0.1
0.3
0.2
0.5
0.0
-3.7
-4.8
2.1
-2.9
-12.1
-22.0
45.4
538.5
IV-86
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM AEROSOL PRODUCT USE
IV.5.4.4
MAC Analysis Results
Global abatement potentials in 2020 and 2030 are 66 and 97 MtCO2e, respectively. There are 70
MtCO2e of reductions in projected baseline emissions in 2030 resulting from implementing currently
available technologies that are cost-effective at projected prices. If an additional emissions reduction
value (e.g., tax incentive, subsidy, or tradable emissions reduction credit) above the zero break-even price
were available to producers of aerosols, then additional emission reductions could be cost-effective. The
results of the MAC analysis are presented in Table 5-6 and Figure 5-4 by major country and regional
grouping at select break-even prices in 2030.
Table 5-6: Abatement Potential by Country/Region at Selected Break-Even Prices in 2030 (MtC02e)
Country/Region
-10
-5
Break-Even Price ($/tC02e)
5 10 15 20 30
Top 5 Emitting Countries
China
India
Mexico
Russia
United States
11.2
2.5
1.1
1.2
3.1
11.2
2.5
1.1
1.2
3.1
27.4
6.2
2.6
3.0
7.5
30.4
6.9
2.9
3.3
8.3
30.4
6.9
2.9
3.3
8.3
30.4
6.9
2.9
3.3
8.3
30.4
6.9
2.9
3.3
8.3
30.4
6.9
2.9
3.3
8.3
33.0
7.5
3.1
3.6
9.0
33.0
7.5
3.1
3.6
9.0
37.7
8.6
3.6
4.1
10.3
Rest of Region
Africa
Central and South America
Middle East
Europe
Eurasia
Asia
North America
Total
2.0
0.8
1.2
3.1
0.2
2.2
0.1
28.7
2.0
0.8
1.2
3.1
0.2
2.2
0.1
28.7
5.0
1.9
2.9
7.7
0.4
5.4
0.2
70.3
5.5
2.1
3.2
8.5
0.4
6.0
0.3
77.8
5.5
2.1
3.2
8.5
0.4
6.0
0.3
77.8
5.5
2.1
3.2
8.5
0.4
6.0
0.3
77.8
5.5
2.1
3.2
8.5
0.4
6.0
0.3
77.8
5.5
2.1
3.2
8.5
0.4
6.0
0.3
77.8
6.0
2.3
3.5
9.3
0.5
6.5
0.3
84.7
6.0
2.3
3.5
9.3
0.5
6.5
0.3
84.7
6.9
2.6
4.0
10.6
0.6
7.4
0.3
96.7
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-87
-------
HFC EMISSIONS FROM AEROSOL PRODUCT USE
Figure 5-4: Marginal Abatement Cost Curves for Top Five Emitters in 2030
•China
Mexico
•India
•Russia
•United States
30
35
Non-CO2 Reduction (MtCO2e)
IV.5.5
Uncertainties and Limitations
The significant areas of uncertainty in this analysis are in how costs for these mitigation technologies
may vary internationally. The analysis is currently limited in the lack of detail on cost assumptions,
which may not accurately represent the transition costs regionally. Additionally, the cost assumptions for
the transition to DPIs are based on a study released in 2000, which may not reflect the latest technical and
economic parameters. Finally, the general methodology used here projects increasing use of HFC aerosols
based on historical growth and expanding GDPs. Some market sectors may not expand that quickly; for
instance, HFC growth in MDIs is projected to grow from current use of 4,000 metric tons globally to
7,000-10,500 metric tons in 2015 (UNEP, 2010).
IV-88
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC EMISSIONS FROM AEROSOL PRODUCT USE
References
Amazon, 2013. Falcon DPSGRN Dust-Off ECO Duster, 5 oz Canister (Case of 12). Obtained August 21,
2013. Available at: http://www.amazon.com/industrial-scientific/dp/B009L9A2X6.
Ecofys. 2000. Abatement of Emissions of Other Greenhouse Gases: Engineered Chemicals. Prepared for the
International Energy Agency Greenhouse Gas Research and Design Programme. Ecofys.
Enviros March. 2000. Study on the Use of HFCs for Metered Dose Inhalers in the European Union.
Commissioned by the International Pharmaceutical Aerosol Consortium (IPAC). Enviros March.
Honeywell. November 2011a. E-mail communication between Alberto Malerba (Honeywell) and Pamela
Mathis (ICF International).
Honeywell. July 2011b. Press Release, "Honeywell To Invest $33 Million In Louisiana Facility." Obtained
in February 2013 at: http://honeywell.com/News/Pages/Honeywell-To-Invest-$33-Million-In-
Louisiana-Facility.aspx.
ITW Chemtronics, 2013. Typhoon Blast™ Duster. Obtained August 21, 2013. Available at:
http://www.chemtronics.com/products/product.asp?id=598.
Javadi, M.S., R. Sondergaard, O.J. Nielsen, M.D. Hurley, and T.J. Wellington. 2008. "Atmospheric
Chemistry of Trans-CF3CH=CHF: Products and Mechanisms of Hydroxyl Radical and Chlorine
Atom Initiated Oxidation." Atmospheric Chemistry and Physics 8:1069-1088. Available at:
http://www.atmos-chem-phys-discuss.net/8/1069/2008/acpd-8-1069-2008.pdf.
MicroCare Corporation. August 2008. Personal communication between Jay Tourigny, MicroCare
Corporation, and Emily Herzog, ICF International.
MicroCare Corporation. November 2011. E-mail communication between Jay Tourigny, MicroCare
Corporation, and Emily Herzog, ICF International.
Miller-Stephenson, 2013. Aero-Duster® product information. Obtained August 21, 2013. Available at:
http://www.miller-stephenson.eom/assets/l/Store%20Item/MS-222L.pdf.
Nardini, Geno. May 2002. Personal communication between Geno Nardini and Iliriana Mushkolaj of ICF
Consulting.
Price, D., E. Valovirta, and J. Fischer. 2004. "The Importance of Preserving Choice in Inhalation Therapy:
The CFC Transition and Beyond." Journal of Drug Assessment 7:45-61.
Stanley Supply & Services, 2013. Techspray Renew 1580-10S Eco-Duster, lOoz. Obtained August 21, 2013.
Available at: http://www.stanleysupplyservices.com/techspray-renew-1580-10s-eco-duster-
lOoz/p/477-260.
TechSpray. August 2008. Personal communication between Steve Cook of TechSpray and Emily Herzog
of ICF International.
United Nations Environment Programme (UNEP). 1999. 2998 Report of the Solvents, Coatings, andAdhesives
Technical Options Committee (STOC): 1998 Assessment. Nairobi, Kenya: UNEP Ozone Secretariat.
United Nations Environment Programme (UNEP). 2010. Report of the UNEP Medical Technical Options
Committee: 2010 Assessment. Nairobi, Kenya: UNEP Ozone Secretariat.
U.S. Environmental Protection Agency (USEPA). 2001. U.S. High GWP Gas Emissions 1990-2010:
Inventories, Projections, and Opportunities for Reductions. EPA #000-F-97-000. Washington, DC:
USEPA.
U.S. Environmental Protection Agency (USEPA). June 2011. U.S. High GWP Gas Emissions 1990-2010:
Inventories, Projections, and Opportunities for Reductions. U.S. EPA #000-F-97-000. Washington, DC:
Office of Air and Radiation, U.S. Environmental Protection Agency.
U.S. Environmental Protection Agency (USEPA). (2012). Global Anthropogenic Non-COi Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA. Obtained from:
http://www.epa.gov/dimatechange/econornics/international.htrnl.
X-rates.com. 2006. Available at: www.x-rates.com. Accessed January 2006.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-89
-------
HFC AND RFC EMISSIONS FROM FIRE PROTECTION
IV.6. HFC and RFC Emissions from Fire Protection
The fire protection sector emits hydrofluorocarbons (HFCs) and perfluorocarbons (PFCs) from
the use of total flooding fire protection systems in fire suppression. This sector also includes
portable (hand-held) fire extinguishers. Greenhouse gas emissions from the fire protection
sector (excluding halons and hydrochlorofluorocarbons) were estimated at roughly 21 million metric tons
of carbon dioxide equivalent (MtCO2e) in 2010, as shown in Figure 6-1. By 2030, emissions from this
sector are expected to reach over 59 MtCO2e. A majority of the growth will result from increased use of
HFCs in developing countries.
Figure 6-1: HFC and RFC Emissions from Fire Protection: 2000-2030 (MtC02e)
59
I Mexico
I Poland
I Japan
I China
I Australia
ROW
2000
2010
2020
2030
Year
Source: USEPA, 2012.
This analysis reviews options to reduce emissions from the fire protection sector by using
extinguishing agents with low global warming potential (GWP) in lieu of HFCs/PFCs in new total
flooding equipment, specifically, replacement of the HFC with inert gas, water mist, or the agent
FK-5-1-12.
The global abatement potential from the fire protection options reviewed is equal to approximately
11% of total annual emissions from the fire protection sector and 0.3% of total annual emissions from all
sectors that use ODS substitutes. The options identified to abate emissions in the fire protection sector
completely replace the fluorinated fire protection agents, however, they are assumed to only be
implemented when new systems are built. Thus, their effectiveness at reducing emissions will increase
over time as new systems are built, but it will take many years before the existing stock would be
replaced. Marginal abatement cost (MAC) curve results are presented in Figure 6-2. Maximum abatement
potential in the fire extinguishing sector is 6.4 MtCC^e in 2030. There are no emission reductions that are
cost-effective at prices estimated in this analysis. No reductions are available in 2010 as a result of the
assumption that the available technologies are only used for new installations and did not start to
penetrate the market until 2011.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-91
-------
HFC AND RFC EMISSIONS FROM FIRE PROTECTION
Figure 6-2: Global Abatement Potential in Fire Protection: 2010, 2020, and 2030
$50
o
u
-$30
?
•2010
2020
•2030
Non-CO2 Reduction (MtCO2e)
IV.6.2
Emissions from Fire Protection
Emissions from fire protection equipment occur from equipment leakage, accidental discharges, and
use during fire extinguishing. In general, fire protection systems have very low leak rates, except when
discharged during a fire event. The majority of emissions associated with fire extinguishing come from
the equipment's use in the total flooding market (see Figure 6-3). Portable extinguishers—used most
frequently in offices, manufacturing and retail facilities, aerospace/marine applications, and homes—also
use HFCs, but such use has been limited (Wickham, 2003).
For the purpose of this analysis, we considered the sector's two types of fire protection systems—
(1) portable fire extinguishers (i.e., streaming applications) and (2) total flooding applications. Typical
portable extinguishers contain 1.3 kg of HFC-236fa, which is assumed to be extinguished with a
frequency such that emissions are 3.5%/year. The typical flooding application uses HFC-227ea; because
discharge is infrequent, emissions are 2% per year.
For modeling purposes, data typical for facilities in the United States are used. Certain cost
assumptions, such as capital and electricity costs, are adjusted for developing countries.1 Otherwise, it is
assumed that the costs and reductions achieved in the modeled facilities could be scaled and would be
representative of the costs and reductions in other regions.
1 In developing countries, it is assumed that capital costs are 10% higher and electricity costs are two-thirds higher
relative to those in the United States.
IV-92
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC AND RFC EMISSIONS FROM FIRE PROTECTION
Figure 6-3: Global HFC and RFC Emissions in 2020 (% of GWP-Weighted Emissions)
B Flooding
Agents, 3%
IV.6.2.1
Activity Data or Important Sectoral or Regional Trends
The key activity data for fire extinguishing emissions is the consumption of the fire extinguishing
agent. Fire extinguishing agent consumption is assumed to occur in all countries and to scale with a
country's gross domestic product (GDP); in addition, because there are no regional differences in
emission rates, emissions also scale with GDP.
Globally, HFC and PFC emissions from fire protection have been growing because of the phaseout of
halon under the Montreal Protocol. Because developed countries phased out the use of halon earlier than
developing countries, the growth in global emissions for the past decade has been driven by emissions
from developed countries. However, because of the high GDP growth in developing countries, it is
anticipated that emissions will grow more quickly from developing countries in the future.
In total flooding applications that require "dean agents,"2 most developed countries have primarily
adopted HFCs as a replacement for Halon 1301, which is being phased out under the Montreal Protocol.
In developing countries, the adoption of HFCs in this application has been delayed by the slower
phaseout of halon but will increase over time.
IV.6.2.2
Emission Estimates and Related Assumptions
Global emissions from fire protection are currently estimated to be 21 MtCO2e and are projected to
grow to 43 MtCO2e in 2020. Growth in emissions is driven by GDP. The emissions assume a constant
'' "Clean agents" are gaseous extinguishing agents that are electrically nonconductive and leave no residue.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-93
-------
HFC AND RFC EMISSIONS FROM FIRE PROTECTION
annual release rate of 3.5% and 2% of consumption for streaming and flooding, respectively.3
Consumption was modeled using USEPA's Vintaging Model. Emissions for major countries and regions
are presented in Table 6-1.
Table 6-1: Projected Baseline Emissions from Fire Protection: 2010 to 2030 (MtC02e)
Country/Region
2010
2015
2020
2025
2030
CAGR*
(2010-2030)
Top 5 Emitting Countries
Australia
China
Japan
Poland
Mexico
2.9
2.2
1.3
1.7
1.1
4.0
3.2
2.2
2.5
1.7
4.8
4.0
3.1
3.3
2.2
5.4
4.6
4.0
4.0
2.6
6.0
5.2
4.7
4.5
3.0
3.8%
4.3%
6.7%
5.0%
5.1%
Rest of Region
Africa
Latin America
Middle East
Europe
Eurasia
Asia
North America
World Total
1.3
1.6
1.8
3.9
0.8
1.8
0.9
21.2
1.9
2.5
2.6
6.6
1.5
2.6
1.2
32.5
2.4
3.3
3.3
9.8
2.4
3.2
1.6
43.2
2.7
3.9
3.8
12.4
3.2
3.6
2.0
52.2
3.0
4.5
4.3
13.9
3.8
4.1
2.3
59.3
4.5%
5.1%
4.5%
6.6%
8.5%
4.2%
4.9%
5.3%
aCAGR = Compound Annual Growth Rate
Source: U.S. Environmental Protection Agency (USEPA), 2012.
IV.6.3 Abatement Measures and Engineering Cost Analysis
The alternatives to HFCs/PFCs in total flooding applications can be categorized as in-kind gaseous
agent alternatives (i.e., CO2, inert gases, fluorinated ketones) and "not-in-kind" alternatives (i.e.,
dispersed and condensed aerosol extinguishing systems, water sprinklers, water mist, foam, or inert gas
generators). Already, climate-friendly dean agents and new not-in-kind alternative technologies have
been introduced to the market.
This analysis reviews options to reduce emissions from the fire protection sector by using zero-GWP
or low-GWP extinguishing agents in lieu of HFCs/PFCs in new total flooding equipment. Specifically, this
analysis assesses alternative agents used in newly built total flooding systems to protect against Class A
3 In general, fire protection systems would be expected to have very low leak rates, except when discharged during a
fire event. For modeling purposes, however, total flooding systems are assumed to have an average annual leak rate
of 2% (see Appendix H to this chapter).
IV-94
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC AND RFC EMISSIONS FROM FIRE PROTECTION
surface fire hazards4 and newly built total flooding systems to protect against Class B fuel hazards5 in
large (^3,000 m3) marine applications. All costs are presented in 2010 dollars based on the Consumer
Price Index (U.S. Department of Labor, 2011).
Facilities/emissions for which no abatement options are considered in this analysis include existing
total flooding systems (used to protect against all fire hazards) and all new and existing portable
extinguishers. Existing flooding systems were not assessed because alternative fire protection agents
require larger space requirements, rendering system retrofit costs highly dependent on the particular
facility and possibly cost-prohibitive. Portable extinguishers were not assessed because emissions from
this source are small, and climate-friendly alternatives are already assumed to be used widely in the
baseline.
The analyzed facilities are assessed on a per-cubic meter of protected space basis, assuming an
average emission rate of 2% per year. Specifically, for Class A surface fire hazards, an average of 0.633 kg
of HFC-227ea is needed to protect each cubic meter of protected space; while 0.630 kg is required for large
Class B fire hazards (Wickham, 2003).
Three abatement options were considered for this analysis: in new Class A total flooding systems,
replacement with either FK-5-1-12 or inert gas, and in new Class B total flooding systems, replacement
with water mist. Each of these options is described below in Table 6-2.
Table 6-2: Fire Protection Abatement Options
^02 ij*]7
FK-5-1-12
Inert gas
Water mist
Applicable System Types
New Class A total flooding
New Class A total flooding
New Class B total flooding
Reduction Efficiency
100%
100%
100%
IV.6.3.1
FK-5-1-12 in New Class A Total Flooding Applications
FK-5-l-12-mmy2 (also known as l,l,l,2,2,4,5,5,5-nonafluoro-4-(trifluoromethyl)-3-pentanone, and
commonly referred to as FK-5-1-12) is a fluorinated ketone with an atmospheric lifetime of five days and
a 100-year GWP of approximately one (Kidde Fire Protection, 2011). This option examines use of FK-5-1-
12 in total flooding systems. The option is applicable in new Class A total flooding applications, replacing
HFCs (primarily HFC-227ea). Class A total flooding application end-uses represent an estimated 95% of
4 This analysis assumes that Class A fire hazards represent an estimated 95% of the total flooding sector. This
estimate is based on Wickham (2002), who estimates that over 90% of the Halon 1301 systems ever installed in the
United States were designed to protect against hazards where the anticipated fire type was primarily Class A in
nature and approximately 10% of the U.S. applications served by Halon 1301 had hazardous materials of the Class B
type. However, because much of the former Halon 1301 Class B applications have been replaced by non-HFC
alternatives (e.g., carbon dioxide), this analysis assumes that only 5% of HFC emissions from the total flooding sector
are from Class B applications and the remaining 95% are from Class A applications.
5 This analysis assumes that Class B fire hazards represent an estimated 5% of the total flooding sector, based on
Wickham (2002) (as explained in previous footnote).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-95
-------
HFC AND RFC EMISSIONS FROM FIRE PROTECTION
the total flooding sector; 6 the additional adoption of FK-5-1-12 is assumed to only occur when new
systems are installed because replacing installed systems may be cost prohibitive.
For cost modeling purposes, this option is assumed to replace the use of HFC-227ea. One-time costs
are estimated at $9.49 per cubic meter of protected space in developed countries; these costs are
associated with installation and equipment, as well as construction of additional floor space needed to
house large volumes of extinguishing agent. In developing countries, these capital costs are assumed to
be 10% higher to account for higher tariffs. Annual costs are estimated at $0.03 per cubic meter of
protected space for additional electricity needed to heat/cool the additional space as well as the higher
agent replacement costs.
IV.6.3.2 Inert Gas Systems in New Class A Total Flooding Applications
Inert gas systems extinguish fires using argon, nitrogen, or a blend of the two, sometimes
incorporating CC>2 as a third component (UNEP, 2001). Inert gas systems provide an equivalent level of
both fire protection and life safety/health protection in most Class A (ordinary combustible) fire hazards,
including electronics and telecommunications applications. Limitations of the inert gas systems include a
slower discharge time than that of HFC systems—60 seconds or more compared with 10 to 15 seconds
(Kucnerowicz-Polak, 2002)—and a larger volume of agent needed than in HFC systems to extinguish
fires. The weight-support structures and space needed for additional steel cylinders of gas may prohibit
the retrofit of many existing HFC systems, such as those on small ships and in other applications where
the system infrastructure is fixed.
This technology option is assumed to be applicable in new Class A application end uses, replacing
HFCs (primarily HFC-227ea). Class A total flooding application end uses represent an estimated 95% of
the total flooding sector;7 the additional adoption of inert gas systems is assumed to only occur when
new systems are installed because replacing installed systems may be cost prohibitive.
For cost modeling purposes, this option is assumed to replace the use of HFC-227ea. One-time costs
are $11.16 per cubic meter of protected space in developed countries; these costs are associated with
installation and equipment, as well as construction of additional floor space needed to house large
volumes of extinguishing agent. In developing countries, these capital costs are assumed to be 10% higher
to account for higher tariffs. Annual costs in developed countries are estimated at $0.17/cubic meter of
protected space for additional electricity needed to heat/cool the additional space; the electricity costs are
assumed to be two-thirds greater in developing countries. Annual costs are offset by annual savings
associated with lower agent replacement costs, estimated at $0.28/cubic meter of protected space.
IV.6.3.3 Water Mist Systems in New Class B Total Flooding Applications
Water mist systems use relatively small droplet sprays under low, medium, or high pressure to
extinguish fires. These systems use specially designed nozzles to produce much smaller droplets than are
produced by traditional water-spray systems or conventional sprinklers, so they use less water to
extinguish a fire (UNEP, 2001; Wickham, 2002). However, some barriers have impeded broad use of
water mist systems. First, these systems may be cost prohibitive in small spaces and have not proven
6 See footnote 4.
7 See footnote 4.
IV-96 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC AND RFC EMISSIONS FROM FIRE PROTECTION
effective in extinguishing small fires in large volume spaces (> 3,000 m3) (IMO, 2001; Wickham, 2002).
Additionally, because there is a nonlinear relationship between the volume of space and the amount of
water mist needed to extinguish a given fire and because this relationship (referred to as the "mechanism
of extinguishment") is not well understood, applications of water mist systems have been limited to those
where fire test protocols have been developed, based on empirically tested system performance. Other
market barriers for this option include additional space requirements for system storage compared with
conventional HFC-227ea systems. Water mist systems can provide equivalent fire protection and life
safety/health protection for Class B fuel hazards, where low temperature freezing is not a concern
(USEPA, 2004).
This technology option is assumed to be applicable in large (>3,000 m3), new Class B total flooding
application end uses, replacing HFCs (primarily HFC-227ea). This analysis assumes that systems
designed to protect against Class B fire hazards represent an estimated 5% of the total flooding sector;8
the additional adoption of water mist systems is assumed to only occur when new systems are installed
because replacing installed systems may be cost prohibitive.
For cost modeling purposes, this option is assumed to replace the use of HFC-227ea. One-time costs
are estimated at $13.14 per cubic meter of protected space in developed countries; these costs are
associated with installation and equipment, as well as construction of additional floor space needed to
house large volumes of extinguishing agent. In developing countries, these capital costs are assumed to
be 10% higher to account for higher tariffs. Annual costs in developed countries are estimated at
$0.36/cubic meter of protected space for additional electricity needed to heat/cool the additional space;
these electricity costs are assumed to be 81% greater in developing countries. Annual costs are partly
offset by annual savings associated with lower agent replacement costs, estimated at $0.28/cubic meter of
protected space.
IV.6.4 Engineering Cost Data Summary
Table 6-3 presents the engineering cost data for each mitigation option outlined above, including all
cost parameters necessary to calculate the break-even price.
IV.6.5 Marginal Abatement Cost Analysis
This section describes the methodological approach to the international assessment of abatement
measures for fire extinguishing.
IV.6.5.1 Methodological Approach
The analysis is based on the above representative project costs for model facilities in the developing
and developed world. We applied the costs to calculate the break-even prices for each appropriate option
for each country. The model estimates the mitigation potential based on the percentage of the total ODS
substitutes baseline attributable to each representative facility and the technical effectiveness for each
technology in each facility.
See footnote 3.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-97
-------
HFC AND RFC EMISSIONS FROM FIRE PROTECTION
Table 6-3: Engineering Cost Data on a Facility Basis
Annual Annual O&M Costs, (2010
Project Capital Cost (2010 USD) Revenue USD) Abatement
Abatement Facility Lifetime (2010 Amount
Option Type (years) Developed Developing USD) Developed Developing (tC02e)
FK-5-1-12
New Class
A total
flooding
Inert gas
Water mist
New Class
A total
flooding
Large, New
Class B
total
flooding
20
20
20
$9.49
$11.16
$13.14
$10.44
$12.28
$14.45
—
$0.28
$0.28
$0.03
$0.17
$0.36
$0.03
$0.31
$0.65
0.04
0.04
0.04
IV.6.5.2
Assessment of Technical Effectiveness
The analysis also developed a technical effectiveness parameter, defined as the percentage reductions
achievable by each technology/class/facility type combination. Estimating this parameter requires making
a number of assumptions regarding the distribution of emissions by Class (A or B), in addition to process-
specific estimates of technical applicability and market penetration. Market penetration rates vary over
time; the market penetration used in this calculation is a modeled value that represents the assumed rate
of penetration of the abatement option into fire protection systems over time, market willingness to adopt
the option, and the turnover rate of existing fire protection systems. Table 6-4 summarizes these
assumptions and presents technical effectiveness parameters used in the MAC model.
Table 6-4: Technical Effectiveness Summary
Facility Abatement Option
FK-5-1-12-Developed
FK-5-1-12— Developing
Inert gas systems— Developed
Inert gas systems— Developing
Water mist systems— Developed
Water mist systems— Developing
Technical Applicability
New Class A total flooding
New Class A total flooding
New Class A total flooding
New Class A total flooding
New Class B total flooding
New Class B total flooding
Market
Penetration
Rate (2030)=
40%
40%
30%
20%
75%
50%
Reduction
Efficiency
100%
100%
100%
100%
100%
100%
Technical
Effectiveness
(2030)"
31%
12%
20%
6%
3%
1%
a Market penetration assumptions for this analysis vary over time, and the technical effectiveness values are based on the cumulative market
penetration rates assumed until that point. More information on the market penetration assumptions is provided in Appendix H to this chapter.
b Technical effectiveness figures represent the percentage of baseline emissions from the relevant facility type that can be abated in 2030;
figures do not account for indirect greenhouse gas impacts associated with increased electricity consumption for heating/cooling of additional
space, which is accounted for in the cost analysis.
IV-98
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC AND RFC EMISSIONS FROM FIRE PROTECTION
IV.6.5.3
Estimating Abatement Project Costs and Benefits
Table 6-5 provides an example of how the break-even prices are calculated for each abatement
measure. Project costs and benefits are calculated for model facilities in developed and developing
countries and are used in the calculation that solves for the break-even price that sets the project's
benefits equal to its costs. The previous section describes the assumptions used to estimate different costs
for developed and developing countries. All options have positive break-even costs, because the costs of
building and maintaining additional space are not offset by any available savings associated with lower
agent replacement costs.
Table 6-5: Example Break-Even Prices for Abatement Measures in Fire Protection
Abatement Option
Reduced
Emissions
(tC02e)
Annualized
Capital Costs
($/tC02e)
Net Annual
Cost ($/tC02e)
Tax Benefit of
Depreciation
($/tC02e)
Break-Even
Price3 ($/tC02e)
Developed
FK-5-1-12
Inert gas systems
Water mist systems
0.04
0.04
0.04
50.6
59.6
70.4
0.8
-2.8
2.2
8.6
10.1
12.0
42.8
46.6
60.6
Developing
FK-5-1-12
Inert gas systems
Water Mist Systems
0.04
0.04
0.04
55.7
65.5
77.4
0.8
1.0
10.2
9.5
11.2
13.2
47.1
55.4
74.4
a Break-even price calculated using a tax rate of 40% and discount rate of 10%.
The break-even prices presented in Table 6-5 represent model facilities for developed and developing
countries. Actual prices vary by country because of the scaling of costs and benefits by international price
factors. Complete international MAC results are presented in Section IV.6.5.4.
IV.6.5.4
MAC Analysis Results
Global abatement potential in 2020 and 2030 is 0.7 and 6.4 MtCO2e, respectively. There are no
reductions in projected baseline emissions resulting from implementing currently available technologies
that are cost-effective at projected prices. If an additional emissions reduction value (e.g., tax incentive,
subsidy, or tradable emissions reduction credit) above the zero break-even price were available to users
or installers of fire extinguishing systems, then additional emission reductions could be cost-effective.
The results of the MAC analysis are presented in Table 6-6 and Figure 6-4 by major country and regional
grouping at select break-even prices in 2030.
IV.6.6
Uncertainties and Limitations
One area of uncertainty is how capital costs for these mitigation technologies may vary
internationally; cost estimates were developed for several countries when possible. In addition, it should
be noted that the global implementation of each option is based on information currently available and
expert opinion. Great uncertainty is associated with future projections of market behavior.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-99
-------
HFC AND RFC EMISSIONS FROM FIRE PROTECTION
Table 6-6: Abatement Potential by Country/Region at Selected Break-Even Prices in 2030 (MtC02e)
Country/Region
Break-Even Price ($/tC02e)
-10 -50 5 10 15 20 30 50
100
100+
Top 5 Emitting Countries
Australia
China
Japan
Mexico
Poland
________ 0.4
________ 0.6
________ 0.3
________ 0.1
________ 0.2
0.4
1.0
0.3
0.2
0.3
0.4
1.0
0.3
0.2
0.3
Rest of Region
Africa
Central and South America
Middle East
Europe
Eurasia
Asia
North America
Total
________ 0.4
________ 0.5
________ 0.4
________ 0.8
________ 0.5
________ 0.3
________ 0.1
________ 4.6
0.6
0.7
0.7
0.9
0.7
0.5
0.1
6.4
0.6
0.7
0.7
0.9
0.7
0.5
0.1
6.4
Figure 6-4: Marginal Abatement Cost Curves for Top Five Emitters in 2030
$50
-------
HFC AND RFC EMISSIONS FROM FIRE PROTECTION
References
ICF International. 2009. Opportunities for Combined Heat and Power in Data Center.
Prepared for Oak Ridge National Laboratory. March 2009. Available at:
https://wwwl .eere.energy.gov/manufacturing/datacenters/pdf s/chp data center
s.pdf.
International Maritime Organization (IMO). November 30, 2001. Performance Testing and Approval
Standards for Fire Safety Systems: Fire Test Protocols for Fire-Extinguishing Systems. Submitted by
Germany to the International Maritime Organization Subcommittee on Fire-Protection, 46th session,
Agenda item 12.
Kidde Fire Protection. 2011. Fire Suppression System Engineered for Use with 3M™ Novec™ 1230 Fire
Protection Fluid. Ref. 6351-5-05/11. Available at: http://www.kfp.co.uk/utcfs/ws-438/Assets/6351-
5 Novec System.pdf.
Kucnerowicz-Polak, B. March 28, 2002. "Halon Sector Update." Presented at the 19th Meeting of the
Ozone Operations Resource Group (OORG), The World Bank, Washington, DC.
R.S. Means Company, Inc. 2007. Means Square Foot Costs, 29th Annual Edition 2008. Kingston, MA: R.S.
Means Company, Inc. U.S. Energy Information Administration (EIA). 2008. "Average Retail Prices of
Electricity, 1960-2007." Available at: http://www.eia.doe.gov/emeu/aer/txt/ptb0810.html.
United Nations Environment Programme (UNEP). 2001. Standards and Codes of Practice to Eliminate
Dependency on Halons: Handbook of Good Practices in the Halon Sector. United Nations Publication ISBN
92-807-1988-1. UNEP Division of Technology, Industry and Economics (DTE) under the OzonAction
Programme under the Multilateral Fund for the Implementation of the Montreal Protocol, in
cooperation with The Fire Protection Research Foundation.
U.S. Department of Labor, Bureau of Labor Statistics. 2011. Consumer Price Index. Obtained May 2011 at
http://www.bls.gov/cpi/.
U.S. Energy Information Administration (EIA). October 19, 2011. "Average Retail
Prices of Electricity, 1960-2010." Annual Energy Review. Obtained December
2011 at: http://www.eia.gov/totalenergy/data/annual/showtext.cfm?t=ptb0810.
U.S. Environmental Protection Agency (USEPA). 2004. Analysis of Costs to Abate
International Ozone-Depleting Substance Substitute Emissions. EPA 40-R-04-006.
Washington, DC: U.S. Environmental Protection Agency. Available at:
http://www.epa.gov/ozone/snap/emissions/downloads/ODSsubstituteemissions.
pdf.
U.S. Environmental Protection Agency (USEPA). June 2006. Global Mitigation ofNon-CO2 Greenhouse Gases.
EPA #430-R-06-005. Washington, DC: Office of Atmospheric Programs, U.S. Environmental
Protection Agency.
U.S. Environmental Protection Agency (USEPA). October 2009. Marginal Abatement Cost Curve Analysis
for Reduction of HFCs in Traditional Ozone Depleting Substance (ODS) End-Use Applications: Draft
Report. Prepared by ICF International for U.S. Environmental Protection Agency.
U.S. Environmental Protection Agency (USEPA). (2012). Global Anthropogenic Non-COi Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA. Obtained from:
http://www.epa.gov/dimatechange/econornics/international.htrnl.
Werner, K. 2011. Personal communication between Kurt Werner of 3M and Pamela Mathis of ICF
International, October 14, 2011.
Wickham, R. 2002. Status of Industry Efforts to Replace Halon Fire Extinguishing Agents. Wickham
Associates. Stratham, New Hampshire. Available at: http://www.epa.gov/ozone/snap/fire/status.pdf.
Wickham, R. 2003. Review of the Use of Carbon Dioxide Total Flooding Fire Extinguishing Systems. Wickham
Associates. Stratham, New Hampshire. Available at: http://www.epa.gov/ozone/snap/fire/co2/
co2report2.pdf.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-101
-------
ALUMINUM
IV.7. RFC Emissions from Primarv Aluminum Productioi
missions of the perfluorocarbons (PFCs)—perfluoromethane (CF4) and perfluoroethane
(C2F6)—are generated during brief process upset conditions in the aluminum smelting process.
During the aluminum smelting process, when the alumina (A12O3) in the electrolytic bath falls
below critical levels required for electrolysis, rapid voltage increases occur. These voltage
excursions are termed "anode effects" (AEs). AEs produce CF4 and C2F6 emissions when carbon from the
anode, instead of reacting with alumina, as it does during normal operating conditions, combines with
fluorine from the dissociated molten cryolite bath. In general, the magnitude of emissions for a given
level of production depends on the frequency and duration of these AEs; the more frequent and long-
lasting the AEs, the greater the emissions.1
Global emissions of PFCs from primary aluminum production declined from 2000 to 2010, which is
likely due to a variety of factors including improvement in process performance of existing smelter
capacity, closure of high emitting facilities, and addition of low emitting new smelting capacity.
However, global emissions of PFCs from this sector are expected to increase from an estimated value of
26 million metric tons of carbon dioxide equivalent (MtCO2e) in 2010 to a projected value of 37 MtCO2e in
2030 (USEPA, 2012) (see Figure 2-1).2 This projected increase is largely the result of an anticipated
increase in demand for primary aluminum globally over the same period that is increasing at a higher
rate than the assumed decrease in PFC emissions intensities (see Emissions from Aluminum Production
below).
Five different electrolytic cell types are used to produce aluminum: Vertical Stud Soderberg (VSS),
Horizontal Stud Soderberg (HSS), Side-Worked Prebake (SWPB), Center-Worked Prebake (CWPB), and
Point Feed Prebake (PFPB).3 PFPB is considered the most technologically advanced process to produce
aluminum and all new greenfield smelters built in the world today utilize this technology. Existing, older
and higher PFC emitting PFPB systems can further improve their anode effect performance by
implementing management and work practices, as well as improved control software. Facilities using
VSS, HSS, SWPB, and CWPB cells can reduce emissions by retrofitting smelters with emission-reducing
technologies such as computer control systems and point-feeding systems, by shifting production to
PFPB technology, and by adopting management and work practices aimed at reducing PFC emissions.
1 It should be noted that over the last several years there has been the discovery and documentation of non-anode
effect (NAE) related emissions. USEPA has supported some of the most significant work on NAE emissions. These
emissions can be a significant, perhaps the major, source of PFC emissions in some smelter cells. It should also be
noted that NAE emissions and NAE abatement measures are not addressed in this report.
2 Please note that IAI (2011) publishes a historical value of 24.4 million metric tons of carbon dioxide equivalent
(MtCO2e) for 2010.
3 It should be noted that PFPB and CWPB are essentially same cell design, but with different alumina feed processes.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-103
-------
ALUMINUM
However, in practice, the greatest potential for reduction in PFC emissions is through addition of new
greenfield PFPB capacity in concert with the shutdown of existing high emitting facilities.4
Figure 7-1: PFC Emissions from Primary Aluminum Production: 2000-2030 (MtC02e)
37
Australia
iCanada
Russia
i United States
i China
Rest of World
2000
2010
2020
2030
Year
Source: U.S. Environmental Protection Agency (USEPA), 2012
Global abatement potential in the primary aluminum sector is 22 MtCO2e in 2030, which represents
approximately 58% of the projected baseline emissions. Figure 7-2 shows the global marginal abatement
cost (MAC) curves for 2010, 2020, and 2030.
4 More information on how global primary aluminum production according to cell type changed from 1990-2012 is
available in "Figure 2: Primary aluminium smelting technology mix, 1990-2012" in Results of the 2012 Anode Effect
Survey: Report on the Aluminium Industry's Global Perfiuorocarbon Gases Emissions Reduction Programme, International
Aluminium Institute, London, United Kingdom. Available online at: http://www.world-
aluminium.org/media/filer public/2013/08/20/2012 anode effect survey report.pdf.
IV-104
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
ALUMINUM
Figure 7-2: Global Abatement Potential in Primary Aluminum Production: 2010, 2020, and 2030
•2010
•2020
2030
Non-CO2 Reduction (MtCO2e)
IV.7.2 Emissions from Primary Aluminum Production
Emissions of the PFCs CF4 and C2F6 from primary aluminum production are estimated using a
variety of activity data (e.g., historical emissions, aluminum production, nameplate capacity), key growth
assumptions (e.g., production growth rate, country-specific trends), and emission factors. Calculations of
PFC emissions from this sector are based on historical and expected levels of aluminum production and
reported (i.e., International Aluminium Institute [IAI]) emission factors from historical experience.
Emissions factors vary by aluminum production technology (e.g., electrolytic cell type) and region (e.g.,
China, rest of the world). More information on the estimation methodology is available in the update to
the Global Emission Report (USEPA, 2012). Figure 7-3 shows the percentage of total PFC emissions
according to production technology type in 2020.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-105
-------
ALUMINUM
Figure 7-3: Global RFC Emissions in 2020 by Facility Type (% of GWP-Weighted Emissions)
CWPB, 3%^ HSS, 3%
For the purpose of developing the cost analysis, five different types of aluminum manufacturing
facilities were considered, based on the technology types—VSS, HSS, SWPB, CWPB, and PFPB. Each
model production facility analyzed produces 200,000 metric tons of aluminum annually.5
IV.7.2.1
Activity Data and Important Trends
The main activity data used to estimate emissions from primary aluminum production are historical
and projected, country-specific, annual, primary aluminum production estimates.
Historical, country-specific, primary aluminum production data for 2010 are compiled from data
provide by the U.S. Geological Survey (USGS) in Mineral Yearbook: Aluminum (USGS, 2011a). In 2010,
world primary aluminum production totaled approximately 40,800 thousand metric tons (USGS, 2011a).6
Projected, country-specific, primary aluminum production data for 2015, 2020, 2025, and 2030 are
estimated based on a combination of either applying the global aluminum production compounded
annual growth rate of 2.5% per year as reported by the Intergovernmental Panel on Climate Change
(IPCC) (Martchek, 2006) to the 2010 country-specific production estimate, or for certain countries, specific
5 It should be noted that the nameplate capacities of newly-built PFPB facilities around the world are typically larger
than 200,000 metric tons.
6 It should be noted that the world primary aluminum production total for 2010 was revised from the value available
at the time of the analysis for this report—40,800 thousand metric tons—to 41,200 thousand metric tons in 2011
Mineral Yearbook: Aluminum [Advance Release]. Available online at:
http://minerals.usgs.gov/minerals/pubs/commodity/aluminum/mybl-2011-alumi.pdf. Most of this increase was in
India and Norway.
IV-106
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
ALUMINUM
production projections provided in comments from the USGS (USGS, 2011b).7 For countries with newly
developed primary aluminum production (e.g., Qatar, Saudi Arabia) or newly re-commissioned primary
production (e.g., Nigeria), the production projections are based on expected production capacity in future
years. The country-specific aluminum production data described above is then disaggregated to cell
technology type using primary aluminum smelting technology mix information derived from lAI's 2020
Anode Effects Survey Report (IAI, 2011). For countries where future production is anticipated to be greater
than the nameplate capacity in 2010, the excess production is assigned to PFPB technology. All current
and future production in China is also assumed to be PFPB technology.
New primary aluminum production is increasingly taking place in emerging and developing regions,
including China and the Middle East, and other countries, including Iceland, where there is the
availability of long term, economically attractive power. New facilities open using new, less emissive
PFPB technologies. In addition, over time older facilities, which use older, more emissive technologies,
are likely to dose, especially if they do not have continued access to competitive power (and labor)
agreements. For example, China is anticipated to continue to be the major primary aluminum producer,
and now uses PFPB technology across all of its facilities.8 Other high-production countries use a mix of
technology types, and production in some of these countries is anticipated to grow, while production in
others is expected to remain constant. For example, after 2015, the United States is anticipated to no
longer be a top five producer of primary aluminum. Thus, emissions in countries with new production
are likely to grow more slowly than emissions from countries with existing production given similar
increases in the rate of primary aluminum produced.
IV.7.2.2 Emission Estimates and Related Assumptions
As previously mentioned, global emissions of PFCs from primary aluminum production are estimate
as 26 MtCO2e in 2010, and are projected to grow to 37 MtCO2e in 2030. Emissions from aluminum
production can generally be described in terms of historical emissions and projected emissions.
Historical, global, PFC emissions data for 1990 through 2010 are compiled from data provided by the IAI
in 2020 Anode Effects Survey Report (IAI, 2011). Projected, country-specific, PFC emissions data for 2010
through 2030 are estimated based on the product of the country-specific production by cell technology
type and the technology-specific mean emission factor for the respective year.
Table 7-1 shows the top countries and regional breakout of emissions of PFCs from primary
aluminum production from 2010 to 2030.
7 It should be noted that growth rates in primary aluminum production have exceeded this value over the past
decade—e.g., 5.3% this decade based on IAI statistics (see http://www.world-aluminium.org/statistics/primary-
aluminium-production/#data) — and aluminum industry leaders have publically estimated growth rates of at least 5%
in the foreseeable future. Growth rates in aluminum production have historically tracked GDP growth in developed
countries, and aluminum growth rates have been historically greater than GDP growth in developing countries based
on the need for aluminum in infrastructure and transportation development. In addition, production growth rates
are region specific, and consumption is not driven by an average (mean) global GDP growth. Rather, aluminum
demand is driven by fast growing economies, thus is more influenced by high growth GDP regions. Therefore, the
published value of 2.5% available at the time of the analysis for this report is likely too small, even taking into
account the current global slowdown. Out to 2020 and beyond, as China, India and other countries increasingly
urbanize, this growth rate will likely further accelerate.
8 It should be noted that while China has converted to PFPB technology, the country's cell designs and control
strategies result in PFC emission factors great than those for PFPB technology operated in the rest of world (ROW).
As a result, separate PFPB emission factors were applied for China versus ROW.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-107
-------
ALUMINUM
Table 7-1: Projected Baseline Emissions from Primary Aluminum Production: 2010-2030 (MtC02e)
Country/Region
2010
2015
2020
2025
2030
CAGR*
(2010-2030)
Top 5 Emitting Countries
China
United States
Russia
Canada
Australia
11.1
3.7
2.5
1.7
0.8
12.6
3.7
2.8
1.8
0.8
14.3
3.6
3.0
2.0
0.9
16.2
3.6
3.2
2.1
1.0
18.3
3.6
3.4
2.3
1.1
2.5%
-0.1%
1.6%
1.5%
1.5%
Rest of Regions
Africa
Central & South America
Middle East
Europe
Eurasia
Asia
1.2
0.8
1.2
1.8
0.3
0.9
1.2
0.8
1.8
1.9
0.3
1.0
1.3
0.9
1.9
2.1
0.4
1.1
1.4
1.0
2.0
2.3
0.4
1.2
1.4
1.1
2.0
2.4
0.4
1.3
1.1%
1.8%
2.6%
1.5%
1.8%
1.6%
North America _____ _
World Total
26.0
28.9
31.4
34.3
37.4
1.8%
aCAGR = Compound Annual Growth Rate
Source: USEPA, 2012
IV.7.3 Abatement Measures and Engineering Cost Analysis
PFC emission reductions can primarily be achieved by installing/upgrading process computer control
systems9 and installing alumina point-feed systems.10 The two abatement options considered for this
analysis are (1) a minor retrofit involving the upgrade of process computer control systems only and (2) a
major retrofit involving both the installation/upgrade of process computer control systems and the
installation of alumina point-feed systems.11 The installation of alumina point-feed systems is not
analyzed on its own, because it would be very unlikely that an aluminum production facility would
install alumina point-feed systems without also installing or upgrading process computer control
systems.12
9 Process computer control systems control the repositioning of carbon anodes as they are consumed and provide
greater control over raw material (alumina) feeding. All smelters operate with process control computers. The
upgrade would involve changes in the algorithms controlling feed and anode effect detection.
10 Point-feed systems allow more precise alumina feeding.
11 A major retrofit results in PFPB technology, which is the state-of-the-art technology in aluminum production.
Conversion to PFPB technology results in the most reliable increases in operational and production efficiency;
although the capital outlay for this option is significant. In addition, retrofit options are usually implemented after
extensive computer modeling and large-scale development work is conducted on test cells.
12 It should be noted that, as previously mentioned, existing, older, and higher PFC emitting systems can further
improve their anode effect performance by implementing management and work practices, as well as improved
control software.
IV-108
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
ALUMINUM
IV.7.3.1
Minor Retrofit
A minor retrofit involves the installation/upgrade of process computer control systems. Minor
retrofits can be performed at any facility type other than the state-of-the-art PFPB facilities. For the
purpose of the cost analysis, a minor retrofit has a lifetime of 10 years for VSS, HSS, and SWPB facility
types; 20 years at the CWPB facility; and 30 years at the PFPB facilities, based on expert judgment. The
lifetime of the minor retrofit at older facilities is shorter because the estimated remaining lifetime of the
facilities themselves is shorter.
• Capital Costs: Capital costs represent the costs associated with purchasing and installing the
process computer control systems at the aluminum production facilities. The capital costs,
obtained from International Energy Agency (IEA) (2000) and confirmed by Marks (2011a), range
from $6 million to $8 million (2010 USD), depending on the facility type.
• Annual O&M Costs: The annual O&M costs associated with the retrofits are strictly the
additional operating costs for the increased aluminum production. The additional operating costs
were assumed to equal the percentage increase in current efficiency multiplied by the capital
costs of the retrofit, which is the method used to estimate O&M costs by IEA (2000). These costs
range from approximately $60,000 to $120,000 (2010 USD), depending on the facility type.
• Annual Revenue: Based on expert judgment, it is assumed that the increased current efficiency
(aluminum production/unit of electricity) resulting from the retrofits would be used by the model
facilities to produce more aluminum with the same amount of electricity consumption as before
(rather than producing the previous levels of aluminum production and realizing the electricity
savings).13 Consequently, additional revenues of approximately $0.5 million to $1 million (2010
USD) are associated with the minor retrofit option, depending on the facility type.
• Technical Lifetime: The expected lifetime range is assumed to be 10 years for VSS, HSS, and
SWPB facility types. Longer lifetimes of 20 and 30 years are applied to CWPB and PFPB facility
types, respectively.
• Reduction Efficiency: The minor retrofits reduction efficiency varies by facility type
(see Table 7-2).
Table 7-2: Primary Aluminum Production Abatement Options
Abatement Option Reduction Efficiency
Minor retrofit
24%-55%
Applicability
All facility types other than state-of-the-art PFPB facilities
Major retrofit
77%-96%
VSS, HSS, SWPB facilities
IV.7.3.2
Major Retrofit
A major retrofit involves both the installation/upgrade of process computer control systems and the
installation of alumina point-feed systems. A major retrofit results in AE performance approaching that
of PFPB technology, which is the state-of-the-art technology in aluminum production. A major retrofit
also results in increases in operational and production efficiency, although the capital outlay for this
option is significant. In addition, retrofit options are usually implemented after extensive computer
modeling and large-scale development work are conducted on test cells. Major retrofits can be performed
13 Note that this is a simplifying assumption for the purpose of this analysis; any increase in production would be
driven by demand for aluminum, not specifically driven by a set level of electricity consumption.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-109
-------
ALUMINUM
for the older facility types (VSS, HSS, and SWPB). According to Marks (2011b), there is no opportunity for
conventional CWPBs to install point feeders because they already have "bar break" feed systems, which
have roughly the same anode effect performance as point feeders; and by definition, a PFPB model
facility has point-feeding systems, so there is no opportunity for additional application.
For the purpose of the cost analysis, a major retrofit has a lifetime of 10 years, based on expert
judgment.
• Capital Costs: Capital costs represent the costs associated with purchasing and installing the
process computer control systems and alumina point-feeding technologies at the aluminum
production facilities. The capital costs, obtained from IEA (2000) and confirmed by Marks (2011a),
range from $12 million to $90 million (2010 USD), depending on facility type.
• Annual Operation and Maintenance (O&M) Costs: The annual O&M costs associated with the
retrofits are strictly the additional operating costs for the increased aluminum production. The
additional operating costs were assumed to equal the percentage increase in current efficiency
multiplied by the capital costs of the retrofit, which is the method used to estimate O&M costs by
IEA (2000), and range from $350,000 to $3.4 million (2010 USD), depending on facility type.
• Annual Revenue: Based on expert judgment, it is assumed that the increased current efficiency
(aluminum production/unit of electricity) resulting from the retrofits would be used by the model
facilities to produce more aluminum with the same amount of electricity consumption as before
(rather than producing the previous levels of aluminum production and realizing the electricity
savings). Increased current efficiencies for each facility and retrofit are available from IEA (2000).
For a major retrofit, these revenues range from $1 million to $2 million (2010 USD), depending on
facility type.
• Technical Lifetime: The expected lifetime of this technology is 10 years.
• Reduction Efficiency: This analysis assumes a reduction efficiency of 78% for HSS, 96% for
SWPB, and 77% for VSS facilities.
IV.7.3.3 Engineering Cost Data Summary
Table 7-3 presents the engineering cost data for each abatement measure outlined above, including all
cost parameters necessary to calculate the break-even price.
IV.7.4 Marginal Abatement Costs Analysis
This section discusses the modeling approach and documents some additional assumptions used in
the MAC analysis for the primary aluminum production sector.
IV-110 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
ALUMINUM
Table 7-3: Engineering Cost Data on a Facility Basis
Abatement
Option
Minor retrofit
Major retrofit
Facility Type
vss
HSS
SWPB
CWPB
PFPB
VSS
HSS
SWPB
Project
Lifetime
(Years)
10
10
10
20
30
10
10
10
Capital Cost
(2010 USD)
$5,980,801
$5,980,801
$6,238,348
$7,125,452
$8,026,865
$84,546,778
$89,039,533
$11,804,213
Annual
Revenue
(2010 USD)
$1,019,402
$509,701
$764,552
$509,701
$509,701
$2,038,805
$1,019,402
$1,529,104
Annual O&M
Costs
(2010 USD)
$119,616
$59,808
$93,575
$71,255
$80,269
$3,381,871
$1,780,791
$354,126
Abatement
Amount
(tC02e)
21,277
112,894
41,900
83,800
129,607
112,894
41,900
83,800
IV.7.4.1
Methodological Approach
The MAC analysis applies the abatement measure costs discussed in the previous Section 7.3 of this
chapter at the five model facilities to calculate a break-even price for the applicable options at each
facility.
IV.7.4.2
Definition of Model Facilities
As mentioned at the beginning of this chapter, five different electrolytic cell types are used to
produce aluminum—VSS, HSS, SWPB, CWPB, and PFPB, which is considered the most technologically-
advanced process to produce aluminum.
A facility's performance may be represented by the mean or median (depending on the size of the
cohort of facilities and the range of performance) PFC emission factor—PFC emissions per unit
production (e.g., metric tons CO2e/metric ton Al)—for a particular cell technology type. However, in the
case of PFPB technology, the universe of facilities using this technology is further subdivided into state-
of-the-art (i.e., newer) PFPB facilities for which no abatement measures are applicable and other (i.e.,
older) PFPB facilities for which certain abatement measures are an option. The performance for state-of-
the-art (as opposed to other) PFPB technology is therefore better represented by a PFC emission factor
less than the average (i.e., the median). Table 7-4 presents a description of the model facilities considered
for this analysis.
IV.7.4.3
Assessment of Technical Effectiveness
To assess the abatement potential from each technology option, one additional parameter is needed,
which is termed the technical effectiveness. The technical effectiveness parameter determines the share of
emissions reductions attributed to each abatement measure. Similar to other industrial process sectors
covered in this report, the technical effectiveness parameter is defined as the percentage of emissions
reductions achievable by each technology/facility combination. Estimating this parameter requires
assumptions regarding the distribution of emissions by manufacturing process (i.e., VSS, HSS, SWPB,
CWPB, and PFPB) in addition to process-specific estimates of technical applicability and market
penetration. The technical applicability, market penetration, and reduction efficiency assumptions are
held constant for all model years. Table 7-5 reports the technical applicability parameters estimated of
each abatement measure/facility type combination. Table 7-5 also presents the market penetration,
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-111
-------
ALUMINUM
Table 7-4: Description of Primary Aluminum Production Facilities
Facility Type
vss
HSS
SWPB
CWPB
PFPB (state of the
art)
PFPB (other)
This facility uses VSS technology, with an average RFC emission factor of 1 .01 metric tons
C02e/metric ton Al. The production capacity of the facility is 200,000 metric tons per year.
This facility uses HSS technology, with an average RFC emission factor of 1 .07 metric tons
C02e/metric ton Al. The production capacity of the facility is 200,000 metric tons per year.
This facility uses SWPB technology, with an average RFC
C02e/metric ton Al. The production capacity of the facility
This facility uses CWPB technology, with an average RFC
C02e/metric ton Al. The production capacity of the facility
emission factor of 5.41 metric tons
s 200,000 metric tons per year.
emission factor of 0.51 metric tons
s 200,000 metric tons per year.
This facility uses state-of-the-art PFPB technology, with a median PFC emission factor of 0.23 metric
tons C02e/metric ton Al. The production capacity of the facility is 200,000 metric tons per year.
This facility uses other PFPB technology, with an average
C02e/metric ton Al. The production capacity of the facility
PFC emission factor of 0.51 metric tons
s 200,000 metric tons per year.3
a It should be noted that "state of the art" has been improving rapidly with respect to anode effect performance and the best PFPB facilities (top
10%) are performing at better than 0.06 metric tons C02e/metric ton Al. Median performance for all IAI non-Chinese producers is about 0.23
metric tons C02e/metric ton Al, while median Chinese PFPB performance is about 0.7 metric tons C02e/metric ton Al.
Table 7-5: Technical Effectiveness Summary
Abatement Option
Model Facility
Type
Technical
Applicability
Market
Penetration
Reduction
Efficiency
Technical
Effectiveness
Minor retrofit
VSS
HSS
SWPB
CWPB
PFPB
27%
100%
100%
100%
100%
100%
50%
50%
100%
100%
39%
39%
24%
55%
55%
11%
20%
12%
55%
55%
Major retrofit
VSS
HSS
SWPB
73%
100%
100%
100%
50%
50%
77%
78%
96%
56%
39%
48%
technical applicability, and reduction efficiency assumptions used to develop the abatement measures'
technical effectiveness parameter. Technical effectiveness is equal to the product of the technical
applicability, market penetration, and reduction efficiency.
Technical applicability factor for VSS is based on the assumption that roughly 27% of VSS capacity
already has point feeding (Marks, 2011b). This percentage is described in more detail in Appendix I.
IV.7.4.4
Estimating Abatement Project Costs and Benefits
The MAC model uses the estimated abatement project costs and benefits as described in
Section IV.7.3 to calculate the break-even price for each mitigation option/facility combination. Table 7-6
illustrates the break-even calculation for each abatement measure expressed in 2010 USD.
IV-112
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
ALUMINUM
Table 7-6: Example Break-Even Prices for Abatement Measures in Primary Aluminum Production
Abatement
Option
Model Facility
Type
Reduced
Emissions
(tC02e)
Annualized
Capital Costs
($/tC02e)
Net Annual
Cost
($/tC02e)
Tax Benefit of
Depreciation
($/tC02e)
Break-Even
Price
($/tC02e)
Minor retrofit
vss
HSS
SWPB
CWPB
PFPB
21,277
41,900
129,607
56,667
56,087
$76
$39
$13
$25
$25
-$42
-$11
-$5
-$8
-$8
$19
$10
$3
$4
$3
$4
$9
$2
$13
$14
Major retrofit
VSS
HSS
SWPB
112,894
83,800
518,429
$203
$288
$6
$12
$9
-$2
$50
$71
$2
$120
$113
$1
IV.7.4.5
MAC Analysis Results
The global abatement potential for PFC emissions in the primary aluminum production sector is 22
MtCC^e, which is approximately 58% of total projected emissions in 2030. Table 7-7 presents the
cumulative reductions achieved at selected break-even prices. Figure 7-4 shows the MAC curve for the
top five emitting countries and the rest of world.
Table 7-7: Abatement Potential by Country/Region at Selected Break-Even Prices in 2030 (MtC02e)
Break-Even Price ($/tC02e)
-10 -50 5 10 15 20 30
HnJH
^^B
^^B
Top 5 Emitting Countries
Australia
Canada
China
Russia
United States
— — —
— — —
- 0.3 1.7
- 0.1 0.3
— — —
0.1
0.21
8.6
1.6
0.34
0
0
1
2
8.6
1
0
6
4
0.1
1.1
8.6
1.6
1.7
0.5
1.1
8.6
1.6
1.7
0
1
5
1
8.6
1
1
6
7
0.5
1.1
10.6
2.0
1.7
0.5
1.1
10.6
2.0
1.7
0.6
1.3
10.6
2.0
2.1
Rest of Region
Africa
Central and South America
Middle East
Europe
Eurasia
Asia
- 0.0 0.3
- - 0.0
- - 0.0
- - 0.0
- 0.0 0.1
- - 0.0
0.7
0.2
0.2
0.2
0.2
0.1
0
0
0
0
0
0
7
5
4
6
2
6
0.7
0.5
0.8
0.8
0.2
0.6
0.7
0.5
0.9
0.9
0.2
0.6
0
0
0
1
0
0
7
5
9
1
2
6
0.8
0.5
0.9
1.1
0.2
0.6
0.8
0.6
1.0
1.3
0.2
0.7
0.8
0.6
1.2
1.4
0.2
0.7
North America __________ _
World Total
- 0.43 2.53
12.4
14.0
16.7
17.3
17.5
20.1
20.5
21.6
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-113
-------
ALUMINUM
Figure 7-4: Marginal Abatement Cost Curves for Top Five Emitters in 2030
DU
3U
oj On
N -?^U
O
y
J/f iu
$0
iu
zu
<3D
j
Australia
^^— China
Canada
Russia
^— United States
; 2 4 6 8 10 12
Non-CO2 Reduction (MtCO2e)
IV.7.4
Uncertainties and Limitations
The emission projections (i.e., baseline emissions) account for the historical reduction in the effective
emission factor (i.e., metric ton CO2e/metric ton Al) realized by facilities, but do not assume that
aluminum producers have conducted retrofits, or will continue to introduce technologies and practices
aimed at reducing PFC emissions. That said, the global primary aluminum industry through the IAI has a
voluntary PFC emission reduction goal of reduce emissions of PFCs per metric ton of aluminum by at
least 50% by 2020 as compared to 2006 (IAI, 2013). In addition, commissioning of new—less emissive-
facilities to meet global demand will also have the result of reducing the effective emission factor.
IV-114
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
ALUMINUM
Harbor Intelligence. 2009. "Interesting Trends in the Chinese Aluminum Market." Obtained at:
http://www.harboralurmnum.corn/archivos/China Aluminum Monitor (Tan 2009).pdf.
International Aluminium Institute (IAI). 2010. Results of the 2009 Anode Effects Survey: Report on the
Aluminum industry's Global Perfluorocarbon Gases Emissions Reduction Program. London, United
Kingdom. July 5. Obtained at: http://www.world-aluminium.org/cache/fl0000361.pdf.
International Aluminium Institute (IAI), 2011. Results of the 2010 Anode Effects Survey: Report on the
Aluminum Industry's Global Perfluorocarbon Gases Emissions Reduction Program. London, UK. August 24.
International Aluminium Institute (IAI), 2013. Results of the 2012 Anode Effects Survey: Report on the
Aluminum Industry's Global Perfluorocarbon Gases Emissions Reduction Program. London, UK.
International Energy Agency (IEA). 2000. Greenhouse Gas Emissions from the Aluminium Industry. The
IEA Greenhouse Gas R&D Program. Cheltenham, United Kingdom.
London Metals Exchange (LME). 2011. LME Official Prices for Aluminum. Obtained at:
http://www.lme.corn/alurninium.asp. Accessed on October 31, 2011.
Marks, J. 2006. Personal communication with Jerry Marks, IAI. as cited in USEPA. 2006. Global Mitigation
ofNon-CO2 Greenhouse Gases. Washington, DC: U.S. Environmental Protection Agency.
Marks, J. 2011a. Personal communication with Jerry Marks via e-mail. June 22.
Marks, J. 2011b. Personal communication with Jerry Marks via e-mail. July 14.
Martchek, K.J. 2006. Modeling More Sustainable Aluminum: Case Study. Int. J. LCA 11(1) 2006:4.
Obtained at: http://www.alcoa.com/sustainability/en/pdfs/KMartchek ITLCA 7772.pdf.
U.S. Environmental Protection Agency (USEPA). (2012). Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA. Obtained from:
http://www.epa.gov/climatechange/economics/international.html.
U.S. Geological Survey (USGS). 2011a. 2020 Mineral Yearbook: Aluminum [Advanced Release]. U.S.
Geological Survey. Reston, VA. September. Available at
http://minerals.usgs.gov/minerals/pubs/commodity/aluminum/.
U.S. Geological Survey (USGS). 2011b. Personal communication with E. Lee Bray, U.S. Geological Survey
Mineral Commodity Specialist.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-115
-------
HFC-23 EMISSIONS
IV.8. HFC-23 Emissions from HCFC-22 Production
rifluoromethane (HFC-23) is generated and emitted as a by-product during the production of
chlorodifluoromethane (HCFC-22). HCFC-22 is used both in emissive applications (primarily
air-conditioning and refrigeration) and as a feedstock for production of synthetic polymers.
Because HCFC-22 depletes stratospheric ozone, its production for dispersive uses is scheduled to be
phased out under the Montreal Protocol. However, feedstock production, a nondispersive use, is
permitted to continue indefinitely.
Global HCF-23 emissions are projected to more than double, growing from 127.9 million metric tons
of carbon dioxide equivalent (MtCO2e) in 2010 to 286.4 MtCO2e in 2030. Figure 8-1 shows the projected
changes in annual emissions of HFC-23 out to year 2030. China, India, and Mexico are projected to see the
largest increases in HFC-23 emissions primarily because of increased HCFC-22 production capacity in
these countries.
Figure 8-1: HCF-23 Emissions from HCFC-22 Production: 2000-2030 (MtC02e)
286
United States
I Russia
Mexico
I India
I China
ROW
2000
2010 2020
Year
2030
Source: U.S. Environmental Protection Agency (USEPA), 2012
The production of HCFC-22 in developed countries has decreased in the last decade, while growth of
annual HCFC-22 production in developing countries has grown substantially, driven primarily by the
demand for its use as feedstock in fluoropolymer manufacture (Montzka et al., 2010). All HCFC-22
producers in developed countries have implemented either process optimization and/or thermal
destruction to reduce HFC-23 emissions. In a few cases, HFC-23 is collected and used as a substitute for
Ozone Depleting Substances (ODSs), mainly in very-low temperature refrigeration and air-conditioning
systems.1 Several HCFC-22 production facilities in developing countries participate in the United Nations
1 Emissions from this use are quantified in the air conditioning and refrigeration chapters and are therefore not
included here.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-117
-------
HFC-23 EMISSIONS
Framework on Climate Change's Clean Development Mechanism (CDM) and through their destruction
of coproduced HFC-23, they are eligible for Certified Emission Reduction (CER) credits. Such projects
were approved beginning in 2003, and although currently 19 projects are approved, a large fraction of
facilities producing HCFC-22 in developing countries are not CDM participants, in part because not all
HCFC-22 facilities are eligible to earn credits under CDM. Current CDM rules state that HCFC-22
production facilities must have an operating history of at least three years between January 2000 and
December 2004 in order to be eligible for a project. A study published in 2010 reported that
approximately 57 percent of HCFC-22 were produced but not covered by existing CDM projects
(Montzka et al., 2010). In another assessment, approximately 43 production lines within 26 existing
HCFC-22 facilities were identified in Article 5 countries. There are about 23 production lines within 17
facilities in Article 5 countries with CDM Projects approved or awaiting approval (Hufford et al., 2012).
This analysis examines the costs to mitigate HFC-23 emissions from HCFC-22 production plants that
do not have incineration technology installed and the costs to mitigate HFC-23 emissions from those
facilities that have thermal destruction devices installed because of a CDM project but are assumed not to
choose to continue their operation after the CDM crediting period expires. There is uncertainty regarding
the future of HFC-23 CDM projects and compliance carbon markets in general; the assumptions chosen to
develop projected abatement potential in this analysis represent one potential scenario. A discussion
regarding the limitations of this analysis is presented in Section IV.8.4
Global mitigation potential of HFC-23 from HCFC-22 production in 2030 is 255 MtCO2e, roughly 89%
of the projected baseline emissions. Figure 8-2 presents the sector marginal abatement cost (MAC) curves
for 2010, 2020, and 2030. This analysis examines the abatement option employed by production facilities
to destroy HFC-23—installation and/or operation of thermal oxidation devices; as shown in the figure,
abatement can be achieved at a low break-even price between $0/tCO2e and $l/tCO2e.
Figure 8-2: Global Abatement Potential in HCFC-22 Production: 2010,2020, and 2030
$60
$50
$40
$30
£ $20
S $10
$0
-$10
-$20
-$30
•2010
•2020
•2030
50
100
150
200
250
300
Non-CO2 Reduction (MtCO2e)
IV-118
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC-23 EMISSIONS
IV.8.1.1 Emissions from HCFC-22 Production
In the production of HCFC-22, HFC-23 is separated as a vapor from the condensed HCFC-22, and
emissions occur through venting of HFC-23 to the atmosphere as an unwanted by-product.
For the purpose of evaluating the cost of reducing HFC-23 emissions from HCFC-22 production, this
analysis considers reduction costs for a typical HCFC-22 production facility, characterized as having a
production capacity of approximately 22,400 metric tons of HCFC-22 (the average production capacity of
all HCFC-22 production plants) and HCFC-22 production at 82% of that production capacity based on
production estimates (Will et al., 2004; 2008).
Additionally, this analysis considers several possibilities for the level of abatement technology used
at the typical HCFC-22 production facility, reflecting different levels of emissions. The analysis examines
four possible categories of facilities. The first two categories address current and historical levels of
emissions from current facilities:
1. Facilities with abatement controls in place already. This level of abatement is true for all production
facilities in the Annex I countries and facilities that have CDM projects. Since the start of Clean
Development Mechanism (CDM) projects for HCFC-22 production, there have been 19 CDM
projects; the majority of these projects are at HCFC-22 production facilities in China (11 in total),
followed by India (5), Argentina (1), Mexico (1), and the Republic of Korea (1).
2. Facilities with no abatement technology controls installed. Such facilities currently exist in China and
Venezuela.
The third and fourth facility categories assist in projecting future emissions from current and new
facilities:
1. New facilities entering the market. To meet future global demand of HCFC-22, the analysis
estimates new facilities to enter the market once projected production for a non-Annex country
exceeds current capacity of the facilities within the country. New facilities are characterized as
being built without control technology.
2. Facilities having previously participated in a CDM project, but not currently incinerating. When a CDM
crediting period is over and the CDM project is completed, this analysis assumes that the
incineration device installed as a result of the CDM project will not be kept in operation. The cost
assumptions for these facilities differ from those of a new uncontrolled facility in that no capital
costs will be needed to install the incinerator. This analysis assumes that all facilities participating
in CDM have completed their crediting periods by 2020.
IV.8.1.2 Activity Data or Important Sectoral or Regional Trends
The primary activity data for HFC-23 emissions from this sector are the level of HCFC-22 production
in the country and whether the production uses any HFC-23 abatement. A total of 20 countries produce
HCFC-22, and of this total, only 12 countries are assumed to continue to produce HCFC-22 between 2015
and 2030. Regionally, abatement of HFC-23 emissions is occurring in developed countries, and in
developing countries, abatement is driven by the CDM incentives for HFC-23 abatement. Thus, the most
significant regional trends are driven by assumptions about the extent to which abatement is occurring by
country and whether that abatement will continue in the future. Overall, global HCFC-22 production is
assumed to continue to grow at a modest rate to meet the demand of HCFC-22 use for feedstock in
fluoropolymer manufacturing, despite restrictions on HCFC-22 production for dispersive uses of
HCFC-22 in response to the controls of HCFC-22 consumption under the Montreal Protocol. Figure 8-3
shows the projected distribution of global HFC-23 emissions by facility type in 2020.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-119
-------
HFC-23 EMISSIONS
Figure 8-3: Global HFC-23 Emissions in 2020 by Facility Type (% of GWP-Weighted Emissions)
Residual
Emissions
controlled
facility), 79
Existing
Uncontrolled
Facility, 10%
New
Uncontrolled
Facility, 6%
IV.8.1.3 Emission Estimates and Related Assumptions
Emissions of HFC-23 from HCFC-22 production were estimated to be 127.9 MtCO2e in 2010, growing
to 286.4 MtCO2e in 2030. Table 8-1 presents the projected annual HFC-23 emissions between 2010 and
2030 for the top five emitting countries and rest of global regions.
Table 8-1: Projected Baseline Emissions from HCFC-22 Production: 2010-2030 (MtC02e)
Country/Region
2010
2015
2020
2025
2030
CAGRa
(2010-2030)
Top 5 Emitting Countries
China
India
Mexico
Russia
United States
62.1
29.1
10.2
9.3
11.8
70.0
37.3
13.1
6.9
10.6
132.2
75.2
26.5
4.6
10.8
142.3
80.9
28.5
5.9
7.9
147.0
83.6
29.4
7.5
6.0
4.4%
5.4%
5.4%
-1.1%
-3.3%
Rest of Region
Africa _____ _
Central & South America
2.5
2.8
3.7
4.0
4.1
2.6%
Middle East _____ _
Europe
1.0
1.0
1.1
1.4
1.8
2.9%
Eurasia _____ _
Asia
1.9
2.4
4.7
5.4
6.8
6.7%
North America _____ _
World Total
127.9
144.2
258.8
276.3
286.4
4.1%
aCAGR= Compound Annual Growth Rate
Source: USEPA, 2012
IV-120
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC-23 EMISSIONS
To estimate historical emissions of HFC-23, dispersive and feedstock HCFC-22 production levels
were developed and subsequently multiplied by a HFC-23/HCFC-22 coproduction ratio (i.e., tons of
HFC-23 emitted per ton of HCFC-22 produced). To account for thermal abatement technologies in the
baseline, the analysis used a lower HFC-23/HCFC-22 production ratio. Depending on how well the
process is optimized, these ratios can range from 1.4% to 4% (Rotherham, 2004; McCulloch and Lindley,
2007). The emission rate for Annex I countries was assumed to be 2% across the entire time series
(Montzka et al., 2010). The emission rate for non-Annex I countries and Russia was assumed to be 3%
from 1990 through 2005 (USEPA, 2006) and 2.9% from 2006 through 2007 (Miller et al., 2010). The lower
emission rate takes into account any HFC-23 emission offsets from CDM projects in these countries and
the Joint Implementation (JI) project at Russia's HCFC-22 plant in Perm. Where UNFCCC-reported HFC-
23 emission estimates were available through the UNFCCC flexible query system, these estimates were
used in place of estimates calculated using production data (UNFCCC, 2009).
HFC-23 emission projections were developed for Annex I countries including Germany, Japan, the
Netherlands, Russia, Spain, and the United States. For the United States, National Communications
projections of emissions were used for 2010 to 2020 (UNFCCC, 2009); emissions trends were used to
project HFC-23 emissions for the remainder of the time series (2025 through 2030). For all other Annex I
countries, the dispersive production and feedstock production portion of emissions were projected
separately to account for the decline in the production for dispersive purposes because of the phaseout
requirements of the Montreal Protocol. To project the feedstock production portion of HFC-23 emissions,
USEPA applied the 5% global growth rate of feedstock HCFC-22 production as reported in Montzka et al.
(2010).
HFC-23 emission projections were developed for non-Annex I countries including China, India,
Mexico, South Korea, and Venezuela. To do so, non-Annex I aggregate HCFC-22 production was
projected for both dispersive and feedstock production; the production was then disaggregated by
country using the percentage of each country's contribution to 2007 non-Annex I HCFC-22 production.
Each country's HCFC-22 projected production was then apportioned into four different model facilities
for each developing country, and two HFC-23/HCFC-22 coproduction ratios were applied to develop
emission estimates—to address the varying use of abatement technologies by facilities. Table 8-2 presents
the assumed distribution of baseline emissions by model facility and country/country group over time.
Table 8-2: Distribution of HCF-23 Emissions by Location and Facility Type: 2010-2030
Country/Group
Annex I
Annex I
Annex I
Annex I
Argentina
Argentina
Argentina
Argentina
China
China
China
China
Model Facility Type
Residual emissions
Non-CDM and uncontrolled facility
New uncontrolled facility
Post-CDM facility
Residual emissions
Non-CDM and uncontrolled facility
New uncontrolled facility
Post-CDM facility
Residual emissions
Non-CDM and uncontrolled facility
New uncontrolled facility
Post-CDM facility
2010
100%
0%
0%
0%
100%
0%
0%
0%
68%
32%
0%
0%
2015
100%
0%
0%
0%
100%
0%
0%
0%
68%
32%
0%
0%
2020
100%
0%
0%
0%
0%
0%
0%
100%
0%
17%
0%
83%
2025
100%
0%
0%
0%
0%
0%
0%
100%
0%
16%
6%
78%
2030
100%
0%
0%
0%
0%
0%
0%
100%
0%
15.9%
8.8%
75.2%
(continued)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-121
-------
HFC-23 EMISSIONS
Table 8-2: Distribution of HCF-23 Emissions by Location and Facility Type: 2010-2030 (continued)
Country/Group
India
India
India
India
Mexico
Mexico
Mexico
Mexico
South Korea
South Korea
South Korea
South Korea
Venezuela
Venezuela
Venezuela
Venezuela
Model Facility Type
Residual emissions
Non-CDM and uncontrolled facility
New uncontrolled facility
Post-CDM facility
Residual emissions
Non-CDM and uncontrolled facility
New uncontrolled facility
Post-CDM facility
Residual emissions
Non-CDM and uncontrolled facility
New uncontrolled facility
Post-CDM facility
Residual emissions
Non-CDM and uncontrolled facility
New uncontrolled facility
Post-CDM facility
2010
100%
0%
0%
0%
100%
0%
0%
0%
0%
0%
0%
100%
0%
100%
0%
0%
2015
78%
0%
22%
0%
78%
0%
22%
0%
0%
0%
11%
89%
0%
89%
11%
0%
2020
0%
0%
14%
86%
0%
0%
14%
86%
0%
0%
14%
86%
0%
86%
14%
0%
2025
0%
0%
20%
80%
0%
0%
20%
80%
0%
0%
20%
80%
0%
80%
20%
0%
2030
0%
0%
22.6%
77.4%
0%
0%
22.6%
77.4%
0%
0%
22.6%
77.4%
0%
77.4%
22.6%
0%
IV.8.2 Abatement Measures and Engineering Cost Analysis
One abatement option, thermal oxidation, is examined in this analysis of the HCFC-22 production
sector; Table 8-3 and Table 8-4 provide a technology overview of this abatement measure. For more
detailed information on the abatement measures considered for this sector see Appendix J.
Table 8-3: HCFC-22 Production Abatement Options
Abatement Option
Thermal Oxidation
Reduction Efficiency
95%
Applicability
Facilities with no abatement technology controls installed
New facilities entering the market
Facilities having previously participated in a COM project
Table 8-4: Engineering Cost Data on a Facility Basis
Facilities with no abatement
technology controls installed
New facilities entering the
market
Facilities having previously
participated in a COM project
Project
Lifetime
(years)
20
Capital Cost
(2010 USD)
$4,800,000
$3,700,000
$0
Annual
Revenue
(2010 USD)
$0
Annual O&M
Costs
(2010 USD)
$119,000
Abatement
Amount (tC02e)
5,932,661
IV-122
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC-23 EMISSIONS
IV.8.2.1 Thermal Oxidation
Thermal oxidation, the process of oxidizing HFC-23 to CCh, hydrogen fluoride, and water, is a
demonstrated technology for the destruction of halogenated organic compounds. For example,
destruction of more than 99% of HFC-23 can be achieved under optimal conditions (i.e., a relatively
concentrated HFC-23 vent stream with a low flow rate) (Rand et al., 1999). In practice, actual reductions
will be determined by the fraction of production time that the destruction device is actually operating.
Units may experience some downtime because of the extreme corrosivity of hydrogen fluoride and the
high temperatures required for complete destruction. This analysis assumes a reduction efficiency of
95%.2
The destruction of HFC-23 by thermal oxidation is assumed to be 100% applicable to all facilities, and
the analysis assumes a project lifetime of 20 years. Cost estimates for installing and operating a thermal
oxidizer are summarized below:3
• Capital Costs: The capital cost for a thermal oxidation system is estimated to be approximately
$4.8 million to install at an existing plant and $3.7 million to install as part of constructing a new
plant (Irrgang, 2011).
• Annual O&M Costs: O&M costs are approximately 2% to 3% of total capital costs (Irrgang, 2011).
This analysis assumes an annual cost that is 2.5% of total capital costs for facilities with no
abatement technology control installed and just over 3% of total capital costs for new facilities
that are entering the market.
• Annual Revenue: No annual savings or revenues are associated with the thermal oxidation
abatement option.4
• Technical Lifetime: 20 years
• Reduction Efficiency: Thermal oxidation technology is assumed to be 95% efficient in abating
HFC-23 emissions.
IV.8.2.2 Evaluation of Future Mitigation Options and Trends
This analysis evaluates how thermal oxidation can be applied to facilities that are current CDM
participants after the crediting period is over and the CDM project is completed. Because an incineration
device is already installed as a result of the CDM project, the costs to adopt the abatement measure relate
only to its annual operation. Facilities participating in CDM are assumed to have completed their
crediting periods by 2020.
This analysis also assumes that new facilities will enter the market to meet future global demand of
HCFC-22. New facilities are assumed to enter the market once projected production for a non-Annex I
country exceeds current plant capacities. According to industry, the costs of installing thermal oxidation
systems in new plants are generally less expensive than the cost of installation at existing plants. This
2 A representative of a company that manufactures thermal oxidation systems stated that new systems are built using
materials that better resist corrosion than the materials used in older systems. The representative indicated that such
new systems were likely to experience very limited downtime, considerably less than 5% (Rost, 2006).
3 Estimates developed for this analysis are based on communication with industry and best available industry
assessments; actual costs of some systems could differ from these estimates.
4 It should be noted that annual revenue is generated for participants of CDM projects; however, CDM projects are
not assumed to be covering further abatement of emissions in this analysis.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-123
-------
HFC-23 EMISSIONS
analysis uses a capital cost for new facilities that is approximately 23% less than the cost of installation at
existing facilities (Irrgang, 2011).
IV.8.3 Marginal Abatement Costs Analysis
This section discusses the modeling approach and documents some additional assumptions used in
the MAC analysis for HCFC-22 production sector.
IV.8.3.1 Methodological Approach
The MAC analysis applies the abatement measure costs discussed in the previous section of this
chapter at the three model facility types to calculate a break-even price for each option at each facility. As
mentioned earlier, this analysis developed four potential model facilities to model the mitigation
potential in this sector. These facilities included the following:
• Residual Emissions: These are facilities that have abatement controls in place already. All
facilities in the Annex I countries and facilities that have CDM projects (mitigation projects
funded by developed countries under the Kyoto Protocol) in the developing countries are
considered "residual emission facilities."
• Non-CDM and Uncontrolled Facility: Non-CDM facilities are existing facilities that are
uncontrolled. These facilities exist in China, South Korea, and Venezuela.
• New Uncontrolled Facility: New facilities are assumed to be uncontrolled when built. It is
assumed that a new facility enters the market once projected production exceeds current
capacity. In other words, the percentage of emissions from new facilities is 0% until projected
production exceeds capacity. It is assumed that new facilities will only be built in non-Annex I
countries.
• Post-CDM Facility: Similar to the "less mitigation scenario" of Miller et al. (2011), this analysis
assumes that the 12 CDM projects that opted for a 7-year crediting period (in China, South Korea,
Mexico, and Argentina) are not renewed after their first terms (note the remaining seven facilities
opted for a one-time fixed crediting period that cannot exceed 10 years). Please see Section IV.8.4
for a discussion on the uncertainty and limitations regarding this assumption. Under this
assumption, by 2020, all facilities previously controlled via CDM ("residual emission model
facility") are considered a "post-CDM" facility. It is assumed that the incineration device installed
(via a CDM project) will not be kept in operation once the CDM crediting period is over. This
analysis costs out mitigation from these facilities differently than a new uncontrolled facility by
taking into account that no capital costs will be needed to install the incinerator.
IV.8.3.2 Assessment of Technical Effectiveness
The analysis also developed a technical effectiveness parameter, defined as the percentage reductions
achievable by each technology/facility type combination. Market penetration rates vary over time as
systems are upgraded in the future. Table 8-5 summarizes the assumptions regarding technical
applicability, market penetration, and technical effectiveness of thermal oxidation for each facility type.
IV-124 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
HFC-23 EMISSIONS
Table 8-5: Technical Effectiveness Summary
Model Facility Type
Non-CDM and uncontrolled facility
New uncontrolled facility
Post-CDM facility
Technical
Applicability
100%
100%
100%
Market
Penetration
Rate (2030)
100%
100%
100%
Reduction
Efficiency
95%
95%
95%
Technical
Effectiveness
(2030)
95%
95%
95%
IV.8.3.3 Estimating Abatement Project Costs and Benefits
Abatement project costs discussed in the previous section were used to calculate the break-even price
for implementing the thermal oxidation technology at each facility type (excluding residual emission
facilities). Using the engineering cost data discussed earlier, Table 8-6 presents the example break-even
prices for each facility type.
Table 8-6: Example Break-Even Prices for Abatement Measures in HCFC-22 Production
Model Facility Type
Non-CDM and uncontrolled facility
New uncontrolled facility
Post-CDM facility
Reduced
Emissions
(tC02e)
5,932,661
5,932,661
5,932,661
Annualized
Capital Costs
($/tC02e)
0.16
0.12
0.00
Net Annual
Cost
($/tC02e)
0.02
0.02
0.02
Tax Benefit of
Depreciation
($/tC02e)
0.03
0.02
0.00
Break-Even
Price
($/tC02e)
0.15
0.12
0.02
IV.8.3.4
MAC Analysis Results
The global abatement potential for HFC-23 reductions in HCFC-22 production sector is 255 MtCO2e,
which is approximately 89% of projected emissions in 2030. Table 8-6 presents the cumulative reductions
achieved at selected break-even prices.
The results are driven largely by the designation of model facilities in different countries. For
example, the United States and Russia have zero mitigation potential because they are included in the
Annex I group of countries and were assumed to have 100% of their baseline emissions represented by
the residual emission model facility (see Table 8-2).
Figure 8-4 shows the corresponding MAC curves for the six countries with abatement potential in
2030, which include China, India, Mexico, South Korea, and Venezuela. Total abatement potential is
achieved at break-even prices between $0/tCO2e and $l/tCO2e in 2030, hence the "L" shape of the curves.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-125
-------
HFC-23 EMISSIONS
Table 8-7: Abatement Potential by Country/Region at Selected Break-Even Prices in 2030 (MtC02e)
Country/Region
Top 5 Emitting Countries
China
India
Mexico
South Korea
Venezuela
Rest of Region
Central and South America
Mirlrllp Fa
-------
HFC-23 EMISSIONS
IV.8.4 Uncertainties and Limitations
This analysis examines a scenario in which the current CDM projects, including those projects with
seven-year crediting periods, are completed by 2020.5 Whether project renewals will occur is uncertain; it
is also uncertain whether facilities would continue to abate even in the absence of CDM incentives.
Although the first seven-year crediting period for the South Korean plant in Ulsan, which ended in
December 2009, was recently renewed for another seven years by the CDM Executive Board in November
2011, the European Commission recommended in January 2011 that the EU cease the purchase of certified
emission reductions (CERs) derived from emission mitigation of HFC-23 production after May 2013
(Europa, 2012). In addition to this ban, which has been formally adopted, other countries, such as New
Zealand and Australia have announced that they will not accept CERs from HFC-23 destruction projects.
The projections in this analysis are limited to this scenario to examine mitigation costs in the absence
of continued CDM projects post-2020; this analysis has not attempted to examine emission projections
and MAC curves under a scenario where CDM projects are renewed post-2020.
' This scenario is similar to the "Less Mitigation" scenario as presented by Miller et al. (2011).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-127
-------
HFC-23 EMISSIONS
References
Chemical and Economics Handbook (CEH). 2001. Fluorocarbons CEH Marketing Research Report.
Chemical and Economics Handbook.
Europa. 2012. Emissions Trading: Commission Welcomes Vote to Ban Certain Industrial Gas Credits.
Press release. Obtained at: http://europa.eu/rapi d/pressReleasesAction.do?reference=IP/ll/56.
Accessed February 17, 2012.
Hufford, D., C. Newberg, D. Godwin, and E. Rim. 2012 "Benefits of Addressing HFCs under the Montreal
Protocol." ASHRAE/NIST Ref Conf.
Irrgang, G. May 19, 2011. Personal communication between Gene Irrgang of Selas Linde and Veronica
Kennedy of ICF International. Information provided by Mr. Rost to Ms. Ottinger in 2006 confirmed
by Mr. Irrgang and adjusted for inflation.
McCulloch, A. and A.A. Lindley. 2007. "Global Emissions of HFC-23 Estimated to Year 2015." Atmos.
Environ. 41:1560-1566.
Miller, B.R., and L.J.M. Kuijpers. 2011. "Projecting Future HFC-23 Emissions." Atmos. Chem. Phys.
Discuss., 11: 23081-23102.
Miller, B.R., M. Rigby, L.J.M. Kuijpers, P.B. Krummel, L.P. Steele, M. Leiste, P.J. Fraser, A. McCulloch, C.
Harth, P. Salameh, J. Miihle, R.F. Weiss, R.G. Prinn, R.H.J. Wang, S. O'Doherty, B.R. Greally, and P.G.
Simmonds. 2010. "HFC-23 (CHF3) Emission Trend Response to HCFC-22 (CHC1F2) Production and
Recent HFC-23 Emission Abatement Measures." Atmos. Chem. Phys. 10: 7875-7890.
Montzka, S.A., L. Kuijpers, M.O. Battle, M.A.K. Verhulst, E.S. Saltzman, and D.W. Fahey. 2010. "Recent
Increases in Global HFC-23 Emissions." Geophysical Research Letters. 37: L02808.
Rand, S., D. Ottinger, and M. Branscome. 1999. Opportunities for the Reduction of HFC-23 Emissions from the
Production of HCFC-22. IPCC/TEAP Joint Expert Meeting. Petten, Netherlands, May 26-28.
Rost, M. April 24, 2006. Personal communication between Marc Rost of T-thermal and Debora Ottinger of
the USEPA.
Rotherharn, D. 2004. Greenhouse Gas Emission Reduction Verification Audit for Dupont's Louisville Works
Freon®22 Plant, Final Report. ICF Consulting, Toronto, Canada, Obtained at:
http://cdm,unfccc.intipublicjnputs/inputamOOO ULetter DuponLAnnex2 03Tune04.pdf.
United Nations Environment Programme (UNEP). 2010. Data Access Centre. HCFC Production.
Obtained at: http://ozone.unep.org/ Data Reporting/Data Access/.
UNFCCC. 2012. United Nations Framework Convention on Climate Change Flexible GHG Data Queries.
Online Database Accessed: Spring 2012. Available at:
http://unfccc.int/di/FlexibleOueries/Setup.do.
U.S. Environmental Protection Agency (USEPA). 2006. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2020. Office of Atmospheric Programs: Climate Change Division. Obtained at:
http://www.epa.gov/ dimatechange/economics/international.html.
U.S. Environmental Protection Agency (USEPA). (2012). Global Anthropogenic Non-COi Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA. Obtained from:
http://www.epa.gov/climatechange/economics/international.html.
Will, R., A. Kishi, and S. Schlag. 2004. CEH Marketing Research Report: Fluorocarbons. Chemical
Economics Handbook. SRI Consulting, Menlo Park, California.
Will, R., and H. Mori. 2008. CEH Marketing Research Report: Fluorocarbons. Chemical Economics
Handbook. SRI Consulting, Menlo Park, CA.
IV-128 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SEMICONDUCTOR MANUFACTURING
IV.9. F-GHG Emissions from Semiconductor Manufacturin
IV.9.1 Sector Summary
T
he semiconductor manufacturing sector uses several fluorinated greenhouse gases (F-GHGs)
including sulfur hexafluoride (SF6), nitrogen trifluoride (NF3), perfluorcarbons (PFCs) such as
carbon tetrafluoride (CF4) and perfluoroethane (C2F6), and the hydrofluorocarbon HFC-23
during fabrication, a portion of which are eventually emitted to the atmosphere. In addition,
nitrous oxide (N2O) and several fluorinated heat transfer fluids (HTFs) are used in the sector, but
emissions from HTFs and N2O are not included in this analysis.
Between 2000 and 2010 the production levels in the semiconductor manufacturing industry have
rapidly grown, and the complexity of devices produced has advanced substantially. However, over this
time period F-GHG emissions from this sector have declined (see Figure 9-1). This reduction can be
attributed to ongoing mitigation efforts in response to voluntary emissions reduction goals set by the
World Semiconductor Council (WSC). For 2010, the WSC set a quantitative emissions target below the
baseline level, and for 2020 it has set an emissions rate target which will entail further implementation of
mitigation technologies.1
Figure 9-1: Projected Baseline Emissions from Semiconductor Manufacturing: 2000-2030 (MtC02e)
30 -i
22
South Korea
• Singapore
• Japan
• United States
• China
Rest of World
2000
2010 2020
Year
2030
Source: U.S. Environmental Protection Agency (USEPA), 2012
1 The emissions projection baseline used here is based on the projection presented in USEPA (2012). That analysis
was conducted before the details of the 2020 WSC voluntary commitment were available. The projection assumes a
continuation of meeting an absolute emissions-level goal through 2030. Further analysis is needed to estimate future
emissions as a result of the new normalized emission rate goal set by the WSC, which will depend on future
production levels. More information on the specific assumptions in the baseline are available in USEPA ( 2012), and
more information on the new WSC goal can be found at
http://vyvyvy.semiconductors.org/nevys/2013/05/28/nevys 2013/global semiconductor leaders reach agreement on pi
an to strengthen industry through international cooperation/
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-129
-------
SEMICONDUCTOR MANUFACTURING
Six mitigation technologies are considered which reduce F-GHG emissions from semiconductor
manufacturing: thermal abatement systems, catalytic abatement systems, plasmas abatement systems, the
NF3 remote chamber clean process, gas replacement, and process optimization. Significant
implementation of these technologies is included in the baseline projection as part of meeting global
voluntary targets. The following mitigation analysis is intended to characterize further reductions beyond
this level, meaning that reductions are fewer and more costly than reductions would be applied to an
uncontrolled baseline.
Mitigation costs and potentials for a particular facility depend on a variety of factors including the
processes and gases used, and emissions reduction technologies already in use. The analysis in this
chapter considers mitigation technologies applied to two stylized facilities: one representing a "new
facility" with relatively new semiconductor technologies and processes and which has already
implemented a suite of mitigation technologies, and one "old facility" which has relatively older
semiconductor technologies and processes and limited existing use of emissions reduction technologies.
Full details on the model facilities are in Section IV.9.4.2.
Global abatement potential of F-GHG emissions in semiconductor manufacturing is estimated to be
12 MtCO2e in 2010, 4.6 MtCO2e in 2020 and 4.2 MtCO2e in 2030. These abatement amounts correspond to
67%, 23%, and 20% respectively in 2010, 2020, and 2030. Figure 9-2 presents the sector marginal
abatement cost (MAC) curves for these three years. The relative availability of potential further
reductions below the baseline projections declines in later years because more mitigation technology is
already included in the baseline as part of meeting voluntary reduction goals. In 2030, less than 1% of the
technically feasible reductions could be supplied cost-effectively (at or below a zero break-even price),
and nearly 7% of those reductions would be achievable at a carbon price of $50/tCO2e.
Figure 9-2: Global Abatement Potential in Semiconductor Manufacturing: 2010,2020, and 2030
•10-
8 $300
•2010
2020
•2030
Non-CO2 Reduction (MtCO2e)
IV-130
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SEMICONDUCTOR MANUFACTURING
This chapter begins by providing a brief discussion of activities and sources of F-GHG emissions in
the semiconductor manufacturing process, and presents the projected emissions from 2010 to 2030. This is
followed by an overview of the abatement measures available to the sector for achieving reductions, their
technological parameters, and economic costs and benefits. The chapter concludes with a discussion of
the MAC analysis and regional MAC results.
IV.9.2 Emissions from Semiconductor Manufacturing
Semiconductor manufacturing emissions considered in this analysis result from two main types of
manufacturing processes used: the etching of substrates and the cleaning of chemical vapor deposition
(CVD) chambers. In addition to direct emissions of portions of F-GHGs that are not consumed in these
processes, by-product emissions of CF4 and other gases (e.g., C2F6) occur when a fraction of the gases
used in processes react to form other F-GHGs.
Other than etching and chamber cleaning, at least three other semiconductor manufacturing
processes result in greenhouse gas emissions. This includes the use of F-GHGs in cleaning wafers, the use
of nitrous oxide in chemical vapor deposition and other processes, and the use of fluorinated heat
transfer fluids. However, these emitting processes were not considered in this analysis because there is
very limited public information that would make estimating emissions from them feasible. In the future,
if more quantitative information is gathered on these three emissive uses of F-GHGs, they can be
considered in an updated analysis.
For the purpose of evaluating the cost of reducing F-GHG emissions from semiconductor
manufacturing, this analysis considers the apparent differences in emissions resulting from newer and
older manufacturing processes and mitigation practices; reduction costs for two typical fabrication
facilities (fabs), which were generally characterized based on fab capacity (i.e., the number of
manufacturing tools a typical fab may have); and the existing use of various mitigation technologies to
etch and clean emissions. The emissions breakdown illustrated in Figure 9-3 represent emissions from
these two types of fabs, further broken out by emission from etch and chamber clean processes. In 2020,
new facilities are expected to make up 30% of global emissions, with old facilities accounting for the
remainder. A description of the characteristics of the old and new fab considered for purposes of analysis
is contained in Section IV.9.4.2.
IV.9.2.1 Activity Data or Important Sectoral or Regional Trends
Several important industry trends drive changes in emissions and mitigation potential from
semiconductor manufacturing: 1) rapid production growth, 2) evolving manufacturing processes and
increasing complexity in devices produced, and 3) impacts of mitigation efforts resulting from voluntary
emissions reduction goals. These trends are described below.
Between 2001 and 2011, global semiconductor manufacturing, measured on the basis of total
manufactured layer area (TMLA), indicates a compound annual growth rate of approximately 10% per
year, which is higher than the silicon consumption growth rate of approximately 7% per year (VLSI, 2012
and WFF, 2012).2 Both silicon area and TMLA are metrics of semiconductor production; however the
difference in growth rate is driven by the increasing complexity of devices, as TMLA reflects
2 Silicon consumption was taken from VLSI, 2012. TMLA was derived in the EPA PFC Emissions Vintaging Model
using data from VLSI, 2012 and WFF, 2012.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-131
-------
SEMICONDUCTOR MANUFACTURING
Figure 9-3: Global F-GHG Emissions in 2020 by Fab Type and Process (% of GWP-Weighted Emissions)
the silicon wafer base layer plus all the metal interconnect layers and the silicon consumption reflects just
the base. More recently production growth has slowed. Between 2006 and 2011, annual production
growth is estimated to be about 5% on a TMLA basis, or 4% on a silicon area basis.
Etch and chamber-cleaning processes have evolved as semiconductor technologies have advanced
and understanding of the emission pathways associated with manufacturing has improved. As
technologies advanced, the semiconductor industry used larger wafer sizes to increase to increase chip
production (e.g., 150 mm to 200 mm to 300 mm). Fabs that produce semiconductors on smaller wafers, on
average, tend to be older and use manufacturing processes that result in a different breakdown of F-GHG
emissions from etch and dean processes as compared to newer fabs. Older fabs may emit approximately
80% of F-GHG emissions total from chamber-cleaning processes and about 20% of emissions from etch
processes. These percentages change to about 45%/55% clean/etch for newer fabs. This shift in the source
of emissions over time is a result of the following: 1) newer fabs generally are trending to NF3 remote-
clean technologies that result in lower emissions on a CO2e basis than traditional older C2F6- or C4F8-
based dean systems, 2) more technologically advanced etch processes have a significantly greater number
of steps, resulting in more F-GHG emissions, and 3) newer fabs can have less physical limitations on
using abatement. As a new generation of fabs come online using 450 mm wafers, it is expected they will
continue to use NF3 remote clean technologies, abatement, and more advanced etch processes.
The WSC set an absolute emissions redudion goal for 2010 and a further emissions rate reduction
goal for 2020. The 2010 WSC goal was an emission reduction of 10% relative to 1995 baseline F-GHG
emissions.3 This emission redudion goal was met in 2010. Achievement of the 2010 WSC emissions
redudion goal has occurred in the context of significantly increasing underlying manufacturing adivity.
A 2011 joint WSC statement reported that the participating industry assodations had achieved a
collective 32% redudion from their baselines while semiconductor industry produdion increased roughly
six times over the same time period (WSC, 2011).
In 2011, the WSC outlined a new voluntary F-GHG agreement for 2020 (WSC, 2011). This agreement
sets a normalized emission rate goal of 0.22 kgCChe/cm2, which is a 30% redudion from the 2010 WSC
aggregate baseline emissions rate (including China, which was not included in the 2010 goal) of 0.33
' Korea has a baseline year of 1998.
IV-132
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SEMICONDUCTOR MANUFACTURING
kgCChe/cm2. In the 2020 voluntary goal the WSC also strongly suggests the use of best practices at newly
built manufacturing facilities.4 EPA has not yet analyzed how this new agreement would relate to
absolute emission reductions in WSC countries, however achieving this goal would require significant
use of mitigation technologies. The baseline used for this analysis assumes continued production growth
and continued implementation of mitigation technologies; however it was estimated before the details of
the 2020 WSC goal were available. Further analysis is necessary to estimate expected semiconductor
manufacturing emission accounting for the 2020 WSC goal. Incorporating the new goal would likely
result in somewhat reduced projected emissions in 2020 and somewhat increased projected emissions for
2030.
IV.9.2.2
Emissions Estimates and Related Assumptions
Projected emissions are based on estimated production level and capacity (described above),
emissions rates drawn from voluntary reporting in the U.S., emissions as reported to the UNFCCC, and
achievement of voluntary goal levels.
The preferred activity data to determine semiconductor emission estimates are gas consumption data.
However, this information is not available globally. Instead, limited data on gas usage from the USEPA
Voluntary Semiconductor Partnership was used to calculate emissions in relation to production.
Therefore, emissions were estimated using this information, emission estimates from the UNFCCC, and
alternative activity data, which is production capacity.
As described above, this analysis takes into account voluntary mitigation activities in the various
WSC member countries and assumes that member countries maintain emissions at the goal level in the
future. Projected emissions for major countries and regions are presented in Table 9-1.
Table 9-1: Projected Baseline Emissions from Semiconductor Manufacturing: 2010-2030 (MtCC^e)
Country
2010
2015
2020
2025
2030
CAGRa
(2010-2030)
Top 5 Emitting Countries
United States
China
Japan
Singapore
South Korea
4.4
4.5
4.1
1.3
1.4
6.3
4.5
4.1
1.6
1.4
5.1
4.5
4.1
1.9
1.4
5.1
4.5
4.1
2.3
1.4
5.1
4.5
4.1
2.7
1.4
0.7%
0.0%
0.0%
3.9%
0.0%
Rest of Regions
Africa
0.02
0.03
0.03
0.04
0.05
5.0%
Central and South America _____ _
Middle East
Europe
Eurasia
Asia
North America
World Total
0.2
1.7
0.1
0.4
0.02
18.2
0.3
1.7
0.2
0.5
0.02
20.6
0.3
1.7
0.2
0.6
0.02
20.0
0.4
1.7
0.3
0.8
0.02
20.7
0.5
1.7
0.4
1.0
0.03
21.5
4.3%
0.0%
4.6%
5.4%
3.0%
0.8%
aCAGR= Compound Annual Growth Rate
Source: USEPA, 2012
4 Best practices, which will be continuously reviewed and updated by the WSC, can be found here:
http://vyvyvy.semiconductorcouncil.org/vysc/uploads/Final WSC Best Practice Guidance 26 Sept 201-2.pdf
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-133
-------
SEMICONDUCTOR MANUFACTURING
To estimate potential for further reductions beyond the voluntary control levels included in the
baseline emissions projection, emissions must be allocated between uncontrolled emissions and residual
emissions remaining after control measures have been implemented. Specific information on current use
of mitigation technologies is not available, so the degree of mitigation is inferred by considering two
stylized model facilities: one where almost full mitigation is used and one where almost no mitigation is
used, and from comparing emission rates for controlled and uncontrolled facilities as determined using
data reported through EPA's Greenhouse Gas Reporting Program (USEPA, 2013) and information
gathered during EPA's voluntary partnership with the semiconductor industry.
IV.9.3 Abatement Measures and Engineering Cost Analysis
Six mitigation technology options were considered for the semiconductor manufacturing sector:
thermal abatement, catalytic abatement, plasma abatement, NF3 remote chamber clean, gas replacement,
and process optimization.
• Thermal abatement: These point-of-use abatement systems, that use heat to destroy or remove F-
GHGs from effluent process streams, are connected directly to a manufacturing tool.
• Catalytic abatement: Tool effluent process streams are run through abatement systems with
catalysts (e.g., CuO, ZnO, A12O3) that destroy or remove F-GHGs.
• Plasma abatement: Plasma, in a point-of-use abatement system, is used to react with (thereby
destroying or removing) F-GHGs from the process effluent stream.
• NF3 remote chamber clean: Highly ionized NF3 is used to clean chemical vapor deposition
chambers. This process is very efficient (using -98% of the gas in a process) resulting in lower
emissions on a mass and CO2 basis than traditional in-situ chamber dean processes that use
approximately 20% to 50% of the gas in a process and have lower efficiencies (USEPA, 2010).
• Gas replacement: Higher global warming potential (GWP) gases are replaced with lower GWP
gases, and in some cases more efficient gases (e.g., C4F8 may replace C2F6 in a traditional
chamber-cleaning process).
• Process optimization: Processes are adjusted to become more efficient, using more gas within the
process, and thus resulting in lower emissions.
These technologies reduce emissions from either etch or chamber-cleaning processes or in some cases
both. Table 9-2 demonstrates the applicability of each mitigation technology to each process type. While
in reality some of these technologies can be stacked, or used together (e.g., a process can be optimized
and then abatement can be applied to that process), the cost and mitigation analysis does not model this
situation.
Table 9-2: Semiconductor Manufacturing Abatement Options
Fab/Emissions Type
Reduction Efficiency
Thermal
Abatement
95%
Catalytic
Abatement
99%
Plasma
Abatement
97%
NF3 Remote
Clean
95%
Gas
Replacement
77%
Process
Optimization
54%
New fab
Etch emissions
Clean emissions
X
X
X
X
X
Old fab
Etch emissions
Clean emissions
X
X
X
X
X
X
X
IV-134
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SEMICONDUCTOR MANUFACTURING
Table 9-3 presents a summary of the engineering cost data for each of the mitigation technologies.
Table 9-3: Engineering Cost Data on a Facility Basis
Abatement
Option
ject Lifetime
(years)
New Old
Capital Costs
(2010 USD)
Annual Costs
(2010 USD)
Abatement Amount
(tC02e)
Thermal 7 7
abatement
Catalytic 7 7
abatement
Plasma abatement 7 7
NFs remote clean 22 1 1
Gas replacement 22 1 1
Process 22 1 1
optimization*
$11,403,942 $5,701,971
$13,813,189 $6,906,594
$3,629,329 $1,814,664
$3,005,084 $9,200,867
n/a $1,180,000
n/a $109,440
$657,723 $328,862
$910,555 $455,277
$103,695 $51,848
$1,214,892 $3,374,861
n/a $64,231
n/a ($129,071)
10,497 52,375
n/a 11,851
n/a 11,612
1,166 41,002
n/a 29,911
n/a 20,976
Note: Values in parentheses denote negative costs.
"Values listed as capital costs for process optimization represent one-time labor costs, not cost of capital.
IV.9.3.1
Thermal Abatement
Thermal abatement systems can be used to abate emissions from both etching and CVD chamber-
cleaning processes by heating process effluent streams to high temperatures to remove or destroy
F-GHGs. The use of thermal abatement offers the benefit of not affecting the manufacturing process
(Applied Materials, 1999); however, the systems do require space that may not be available in sub-fabs,
particularly in older facilities. In addition, these systems require large amounts of cooling water, and the
system's use results in regulated NOX emissions. Thermal abatement systems are currently the most
widely used abatement system in the semiconductor industry.
The engineering cost estimates for this technology are as follows:
• Capital Costs: Thermal abatement system capital costs cover the cost of the abatement unit with
ducting and water recirculation ($157,000 per unit), hook-up costs ($35,550), and natural gas
infrastructure costs ($35,550) (Fthenakis, 2001; Burton, 2003). One unit is needed per tool at a
facility. The total facility capital cost ranges from $11.4 million for new fabs to $5.7 million for old
fabs.
• Annual Costs: Annual operating costs per manufacturing tool, as presented in Table 9-4, are the
same for both new and old fabs. Total annual costs (e.g., utilities) for a new fab are estimated to
be $658,000 and $329,000 for an old fab. The higher capital and annual costs for new fabs are
based on the fact that new fabs typically have larger manufacturing capacities (i.e., more tools)
(WFF, 2011). Annual costs per tool are summarized in Table 9-4. The per-tool cost is the same for
both new and old fabs.
• Annual Revenue: No financial benefits (e.g., cost savings) are associated with using this
mitigation technology without outside policy or other drivers.
• Reduction Efficiency: This analysis assumes a 95% reduction efficiency (Fthenakis, 2001; Beu,
2005; USEPA, 2009).
• Technical Lifetime: Based on expert judgment, it was estimated that the average lifetime of this
system, and other abatement systems discussed in this analysis, is 7 years.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-135
-------
SEMICONDUCTOR MANUFACTURING
Table 9-4: Annual Cost per Tool for Thermal Abatement Systems
Annual Cost Component
Water/waste water/maintenance
Consumables
Electricity
Natural gas
Cost (2010 USD)
$2,370
$5,330
$2,610
$2,840
Source: Burton, 2003.
IV.9.3.2 Catalytic Abatement
A catalytic abatement system uses a catalyst to destroy or remove F-GHG emissions from the
effluents of both plasma etching and CVD chamber-cleaning processes. This type of abatement is
applicable at most facilities, but there may be some space constraints as mentioned above for thermal
abatement systems. Additionally, because these systems are based on destruction via catalyst, they must
be process/stream specific to achieve the 99% emission reductions quoted in the literature and used in
this analysis (Fthenakis, 2001; Burton, 2003).
Because catalytic destruction systems operate at relatively low temperatures, their use results in little
or no NOX emissions, and the required amounts of water are also low. Because of the high cost of catalyst
replacement, these systems are the least widely used type of abatement (expert judgment).
Cost estimates for this technology are as follows:
• Capital Costs. Capital costs are associated with purchasing and installing the abatement systems
(Burton, 2003). One unit costs $217,010, and the installation costs $59,250, leading to estimated
costs of $6.9 million and $13.8 million for old and new fabs, respectively.
• Annual Costs. Facilities incur annual costs per tool for water ($3,790), waste chemicals ($60),
catalyst replacement ($12,580), and electricity ($1,780) (Burton, 2003). A new fab incurs annual
costs for catalytic abatement of $910,600, and an old fab incurs costs of $455,300.
• Annual Revenue. No cost savings are associated with this technology.
• Reduction Efficiency: The analysis assumes 99% reduction efficiency for catalytic abatement
(Fthenakis, 2001).
• Technical Lifetime: Seven years.
IV.9.3.3 Plasma Abatement
These systems, which use plasma to remove or destroy F-GHGs, are applicable to etch processes in
most facilities, with some physical space limitations. (These systems, though, are smaller in size
compared with thermal and catalytic systems.) Plasma abatement systems use a small plasma source that
effectively dissociates the F-GHG molecules that react with fragments of the additive gas (hydrogen (H2),
oxygen (O2), water (H2O), or methane (CH4) to produce low molecular weight by-products such as
hydrogen fluoride (HF) with little or no GWP. After disassociation, wet scrubbers can remove the
molecules. The presence of additive gas is necessary to prevent later downstream reformation of F-GHG
molecules (Motorola, 1998).
A plasma abatement system is needed on each tool chamber. The costs of plasma abatement systems
are developed using the following information:
IV-136 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SEMICONDUCTOR MANUFACTURING
• Capital Costs: The plasma abatement technology requires capital costs that cover the purchase
and installation of the system, which total $41,500 per chamber, equating to one-time costs of $3.6
million for new tabs and $1.8 million for old tabs (Fthenakis, 2001; Burton, 2003).
• Annual Costs: Facilities with plasma abatement systems are estimated to incur annual operation
costs of $1,190 per chamber, which includes general maintenance and use of the system. Total
annual facility costs are $103,700 for new tabs and $51,800 for old tabs, based on the assumption
that there are 3.5 chambers per etch tool and varying numbers of tools for new and old tabs
(Fthenakis, 2001; Burton, 2003).
• Annual Revenue: As with other abatement technologies, the use of plasma abatement systems
will not result in any cost savings.
• Reduction Efficiency: The emissions reduction efficiency of this option is estimated to be 97%
(Fthenakis, 2001; Hattori et al., 2006).
• Technical Lifetime: 7 years.
IV.9.3.4 NF3 Remote Chamber Clean
NF3 remote chamber clean is an alternative cleaning technology that offers the benefit of having a
particularly high (-98%) utilization rate of NF3 (IPCC, 2006), resulting in relatively low emissions
compared with traditional chamber cleans. NF3 remote clean systems dissociate NF3 using argon gas, and
converting the source gas to active F-atoms in the plasma upstream of the process chamber. These
electrically neutral atoms can selectively remove material in the chamber. The by-products of remote
dean include HF, fluorine (F2), and other gases, most of which are removed by facility acid scrubber
systems. The use of NF3 remote dean systems is much more prevalent in newer tabs because the
technology was not available when many older tabs were constructed.
Capital costs for NF3 remote clean systems will differ for new and old tabs because of the "readiness"
for NF3 remote dean installation. "Readiness" consists of having the current infrastructure (e.g., duct
work, hook-ups) for system installation. It was assumed that old tabs are do not have the current
infrastrudure to use NFS remote clean, whereas new tabs do. Therefore, the capital costs for old tabs
reflect the needed infrastrudure changes for the fab.
Cost assumptions indude the following:
• Capital Costs: Both facility types would incur capital costs for purchasing the NF3 remote system
and the additional necessary F2 scrub for use after the chamber deaning of the waste stream. The
costs for system purchase for a new fab are estimated to be $3 million. Old tabs are assumed to
not be "NF3 ready," or in other words, these facilities are not assumed to have the current
infrastrudure to handle the direct installation of NF3 remote systems. Therefore, old tabs also
incur capital costs, in addition to system costs, assodated with investments such as gas hook-ups
and necessary hardware such as manifolds and valves in addition to the costs of the systems
which are assumed to be already installed at new tabs. (These costs are detailed in Table 9-5.) The
old fab costs are estimated to be $9.2 million.
• Annual Costs: Facilities operating NF3 remote dean systems are subjed to annual costs that
include the purchase of larger volumes of gas (NF3 versus traditional chamber-deaning gases
such as C2F6), general maintenance, and the cost of F2 scrubs to remove the highly explosive gas
from the effluent. Remote clean requires a lot of NF3, so much so that NF3 purchases are
estimated to comprise anywhere from 25% to upward of 75% of annual fadlity gas consumption
(expert judgment). New fab costs annually for NF3 remote dean are estimated to be $1.2 million
and to be $3.4 million for old tabs (Burton, 2003).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-137
-------
SEMICONDUCTOR MANUFACTURING
Table 9-5: Capital Costs per CVD Chamber for Making a Facility NFs Ready
Activity
Labor/gas hookup
NFs manifold, valves, etc.
Toxic monitor
Stainless steel line (double walled)
Capital Cost (2010 USD)
$3,980
$16,591
$7,700
$10,310
Source: Burton, 2003
• Annual Revenue: No cost savings are assumed to be associated with this technology.
• Reduction Efficiency: The analysis assumes this technology offers a reduction of 95% of
emissions (Beu, 2005).
• Technical Lifetime: Once the remote dean systems are installed, they will last for the lifetime of a
facility. Based on information from the World Fab Forecast, the average remaining lifetime of a
facility is 11 years for an old fab and 22 years for a new fab.
IV.9.3.5 Gas Replacement
Gas replacement can be used to mitigate emissions from the traditional CVD chamber-cleaning
process. This method can be applied in most facilities and has already been used throughout the industry
in many instances. For this strategy, a lower GWP gas replaces a higher GWP gas. The most common
replacement seen is using C4F8 to replace C3F8 or C2F6. In addition, the replacement gas (C4F8 ) is often
used/consumed more efficiently during CVD chamber cleaning than the original gas C2F6 or C3F8, which,
combined with the differences in GWP, yields lower emissions.
As with most other technologies considered in this analysis, there are no associated cost savings.
• Capital Costs. Facilities replacing C2F6 or C3F8 with C4F8 face a capital expenditure that reflects
the aggregate cost of the C4F8gas hook-up and an engineer's time cost for implementation. Based
on the Clean Development Mechanism (CDM) number NM0317, the aggregated cost of
equipment, C4F8 gas hook-up, and an engineer's time for implementation and installation is
estimated to be $1.2 million for old tabs (the technology is not assumed to be used at new tabs).
• Annual Costs. Facilities face an annual cost that reflects the cost of replacing C2F6 or C3F8 with
the more expensive C4F8 The costs of these gases, taken from CDM NM0303, are $35 per
kilogram of C2F6, $26 per kilogram of C3F8, and $72 per kilogram of C4F8. Costs for old tabs were
estimated to be $64,230, which is based on an average amount of gas consumed per facility. Gas
consumption information was estimated based on USEPA Voluntary Partnership data, in which
facility age and gas consumption relationships were not distinguishable.
• Annual Revenue. No cost savings are associated with this technology.
• Reduction Efficiency: The analysis assumes a reduction efficiency of 77% for this mitigation
technology (CDM methods NM0289, NM303, NM0317, NM0335).
• Technical Lifetime: As with NF3 remote dean, once a gas is replaced, the "new" process will last
for the lifetime of a fab. Based on information from the World Fab Forecast, the average
remaining lifetime of a facility is 11 years for an old fab.
IV.9.3.6 Process Optimization
Process optimization is the redudion in GHG emissions from a process by modifying or adding to
the process recipe. Process optimization is considered to be only applicable for chamber deans because
IV-138 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SEMICONDUCTOR MANUFACTURING
these processes offer the opportunity for more flexibility than etch processes. Etch processes are typically
developed to optimize production yield, and they are only adjusted to increase this yield; a company
would not risk negatively impacting it (Beu, 2005; Fthenakis, 2001). Process gas optimizations for CVD
clean processes can be implemented because adjustments to these processes are much less precise than
etch processes. There is room to reduce emissions without affecting yield. Optimization of dean processes
to reduce emissions usually results in small production gains but sometimes can result in large increases
in efficiency.
Facilities optimizing processes incur labor costs of an estimated $109,440; it is assumed that old fabs
incur this cost, while new fabs do not implement this technology due to their assumed use of NFS remote
dean for the majority of dean processes.
Details of the cost estimates for this technology indude the following:
• One Time Labor Costs: Facilities' optimizing processes incur a one-time labor cost. Labor costs of
$43 per hour were used for a materials engineer in the semiconductor industry based on BLS
(2010) information and an estimated 2,560 hours of work, resulting in a total labor cost for each
model facility of $109,440.
• Annual Costs: No annual costs are assodated with process optimization for clean processes that
are outside of business-as-usual (BAU) annual facility costs.
• Annual Revenue: Because process optimization involves adjusting a process to perform more
effidently, the cost savings assodated with this option are due to a lowered amount of gas
required to be purchased. For simplicity, the process considered in this analysis for this option
was a C2F6 traditional chamber dean, and the related savings were estimated to be $129,070. As
for the gas replacement annual cost, this number varies depending on the size of the fadlity.
However, consumption information was only able to be estimated based on USEPA Voluntary
Partnership data, in which facility age and gas consumption relationships were not
distinguishable.
• Reduction Efficiency: Observed reduction effidendes for abatement of C2F6 in the literature
range from 10% to 56% (Beu, 2005) and as high as 75% (Fthenakis, 2001). For the purposes of this
analysis, an average reduction effidency of 54% was used, and we assumed the change in process
is permanent over the life of a facility.
• Technical Lifetime: As with NF3 remote dean, once a gas is replaced, the "new" process will last
for the lifetime of a fab. Based on information from the World Fab Forecast, the average
remaining lifetime of a facility is 11 years for an old fab.
IV.9.4 Marginal Abatement Cost Analysis
This sedion discusses the modeling approach and documents some additional assumptions used in
the MAC analysis for semiconductor manufaduring.
IV.9.4.1 Methodological Approach
The MAC analysis applies the abatement measure costs discussed in the previous section of this
chapter at two hypothetical fadlities to calculate a break-even price for each option at each facility (new
and old). This sedion presents detailed information on how each type of fab was defined in this analysis,
and detailed information on how costs were built out for each mitigation technology.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-139
-------
SEMICONDUCTOR MANUFACTURING
IV.9.4.2 Definition of Model Facilities
For this sector, two fab types were considered: an old fab and a new fab. The differences between
these two fabs are discussed in more detail below:
• Old: The old fab is intended to capture facilities that use smaller wafer sizes, such as 150 mm and
below, as well as some older 200 mm manufacturing facilities (i.e., fabs built before 2000). This
fab is expected to use less current manufacturing processes and produce less, in terms of silicon
area, was estimated for this analysis, based emissions data reported through the U.S. EPA
Greenhouse Gas Reporting Program, that a typical "old" fab breakdown of emissions on a CO2e
basis is approximately 20% etch emissions and 80% dean emissions. This is because older etch
processes involve less GHG-using steps and more commonly use traditional chamber deans with
gases such as C2F6 as opposed to remote chamber deaning processes. This fab is also expeded to
not use any abatement and only use minimal process optimization and gas switching. It was
assumed that an old fab has an average of 30 tools with 3.5 chambers per tool.5
• New: The new fab type encompasses facilities that use larger wafer, such as 300 mm wafers. It is
estimated, based again on emissions data reported through the U.S. EPA Greenhouse Gas
Reporting Program, that the total emission breakdown for the new fab is approximately 55% etch
emissions and 45% dean emissions. In contrast to the old fab, the new fab uses more recent etch
processes that have comparatively many more GHG-using steps and the fab has higher
produdion, in terms of silicon area.6 Another process shift seen in newer fabs is the trend toward
using NF3 remote chamber deans as opposed to traditional chamber deans, which results in
relatively lower cleaning emissions. New fabs are assumed to use NF3 remote clean mainly, and
have abatement on all etch processes and all in situ chamber cleaning processes. It was assumed
that new fab facility has about 50 tools with 3.5 chambers per tool.
The emission breakdowns are essential to this analysis, because some mitigation technologies are
applicable to either both or just one type of manufacturing process. One other important fador is facility
size. Newer fabs tend to have relatively larger produdion capacities than older fabs, and this difference
was taken into account in this analysis.
The fadlities used represent two clearly defined and distind types of fadlities. These defined fadlities
represent two existing scenarios, a better existing mitigation case and a worse existing mitigation case, for
semiconductor manufaduring fabs. Given the variety of mitigation options, there are fadlities that exist
that may be in the somewhere between the two scenarios modeled. For instance, some fabs may partially
abate emissions as opposed to using full abatement or no abatement. These fabs were not explidtly
considered in this analysis due to the uncertainty associated with developing assumptions about their
current mitigation practices.
IV.9.4.3 Assessment of Technical Effectiveness
The analysis also developed a technical effectiveness parameter, defined as the percentage reductions
achievable by each technology/process/fadlity type combination. Estimating this parameter requires
making a number of assumptions regarding the distribution of emissions by manufacturing process (etch
and dean) in addition to process-spedfic estimates of technical applicability and market penetration. The
5 CVD and etch tools generally vary between having three to four chambers.
6 Although newer etch processes are more efficient (i.e., gas utilization is higher) than older processes, the relative
number of GHG-using steps in more recent processes negates the potential benefit of higher utilization of gas when
considering overall facility etch emissions.
IV-140 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SEMICONDUCTOR MANUFACTURING
split of etch to dean emissions is held constant for all years. The technical applicability and market
penetration of mitigation technologies is held constant over time for new facilities as it is assumed these
facilities are addressing emissions as much as possible already. Whereas, technical applicability and
market penetration for old facilities varies over time as it is assumed that more action will need to be
taken by older facilities to meet stated voluntary reduction goals. Table 9-6 presents the assumed
distribution of annual facility-level emissions by process for each fab type.
Table 9-6:
Percentage of Annual Emissions by Process and Fab Type
'ercentage of Total Annual Emissions
Process
Etch
54%
22%
Clean
46%
78%
Total
100%
100%
Table 9-7 and Table 9-8 report the technical applicability, market penetration, and reduction
efficiency assumptions used to develop the abatement measures' technical effectiveness at new and old
fabs. The technical effectiveness is the weighted average of the abatement measures using the process
emissions presented in Table 9-6 for each process as the weight multiplied by the product of the technical
applicability, market penetration, and reduction efficiency.
Table 9-7: Technical Effectiveness Summary for New Fabs (Constant Over Time)
Etch (54%) Clean (46%)
Technical Market Technical Market Reduction Technical
Abatement Measure Applicability Penetration Applicability Penetration Efficiency Effectiveness
Thermal abatement
Catalytic abatement
Plasma abatement
NFs remote clean
Gas replacement
Process optimization
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
90%
0%
0%
10%
0%
0%
50%
0%
0%
50%
0%
0%
95%
99%
97%
95%
77%
54%
20%
0%
0%
2%
0%
0%
Table 9-8: Technical Effectiveness Summary for Old Fabs (in 2020)
Etch (20%) Clean (80%)
Technical Market Technical Market
Abatement Measure Applicability Penetration Applicability Penetration
Reduction
Efficiency
Technical
Effectiveness
Thermal abatement
Catalytic abatement
Plasma abatement
NFs remote clean
Gas replacement
Process optimization
50%
50%
50%
0%
0%
0%
90%
5%
5%
0%
0%
0%
50%
0%
0%
100%
10%
10%
15%
5%
0%
5%
40%
40%
95%
99%
97%
95%
77%
54%
15%
1%
1%
4%
2%
2%
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-141
-------
SEMICONDUCTOR MANUFACTURING
Technical applicability assumptions presented in Table 9-7 and Table 9-8 are intended to reflect the
space limitations or preexisting process performance issues that are likely to be found at a fraction of all
facilities, particularly old facilities, preventing the total implementation of the abatement measures.
Assumed market penetration rates are based on cost (lower cost options will penetrate the market more)
and expert knowledge of industry trends. For example, fabs tend to use thermal abatement more than the
other technologies in their etching processes. In addition, it was assumed that because most new fabs
already have NF3 remote systems in place less market share would go to gas replacement and process
optimization.
The technical effectiveness estimates are then multiplied by the share of total emissions for each
facility type to estimate the abatement potential achievable under each abatement measure. For the
purposes of this analysis we assume a 10/90 split in 2010 for the distribution of annual emissions coming
from new and old fabs. In future years, we assume based on expert judgment the share of total emissions
coming from new fabs increases by 10% each year to account for the use of best practices by WSC
members, old fabs closing and changes in wafer size/technology demands.
IV.9.4.4
Estimating Abatement Project Costs and Benefits
The MAC model uses the estimated abatement project costs and benefits as described in Section
IV.9.3 to calculate the break-even price for each mitigation option at both new and old fab facilities.
Table 9-9 illustrates the break-even calculation for each abatement measure expressed in 2010 USD.
Although new fabs have lower break-even prices for thermal, catalytic, and plasma abatement measures,
old fabs have lower break-even prices for NF3 remote dean, gas replacement, and process optimization
because of their relatively smaller size. Note that process optimization is the only negative break-even
price option because of its low one-time cost and relatively high annual cost savings.
Table 9-9: Example Break-Even Prices for Abatement Measures in Semiconductor Manufacturing
Abatement Option
Reduced
Emissions
(tC02e)
Annualized
Capital Costs
($/tC02e)
Net Annual
Cost
($/tC02e)
Tax Benefit of
Depreciation
($/tC02e)
Break-Even
Price3
($/tC02e)
New fabs
Thermal abatement
Catalytic abatement
Plasma abatement
NFs remote clean
Gas replacement
Process Optimization
41,199
8,096
11,833
22,551
6,240
1,648
85
526
94
114
36
13
14
101
8
242
10
-78
24
146
26
18
6
2
76
481
76
338
40
-68
Old fabs
Thermal abatement
Catalytic abatement
Plasma abatement
NFS remote clean
Gas replacement
Process optimization
22,802
4,575
3,293
37,768
9,753
3,864
103
620
226
75
31
7
17
119
19
107
7
-33
29
173
63
18
7
2
91
567
182
165
30
-28
a Break-even price calculated using a tax rate of 40% and discount rate of 10%.
IV-142
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SEMICONDUCTOR MANUFACTURING
IV.9.4.5
MAC Analysis Results
The global abatement potential for F-GHG reduction in the semiconductor manufacturing sector is
estimated to be 87% of total projected emissions in 2030. Table 9-10 presents the cumulative reductions
achieved at selected break-even prices. Figure 9-4 shows the MAC curve for the top five emitting
countries for this sector.
Table 9-10: Abatement Potential by Country/Region at Selected Break-Even Prices in 2030 (MtC02e)
Break-Even Price ($/tC02e)
Country /Region
-10 -5
mm
B
mm
mm
20
30
50
100
100+
Top 5 Emitting Countries
China
Japan
Singapore
South Korea
United States
- 0.0
— —
- 0.0
- 0.0
- 0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.0
0.0
0.0
0.0
0.1
0.0
0.1
0.0
0.1
0.3
0.1
0.1
0.0
0.1
0.3
0.1
0.1
0.0
0.1
0.3
0.1
0.1
0.1
0.1
0.3
0.3
0.2
0.1
0.3
0.4
0.3
0.2
0.1
0.3
1.0
0.8
0.5
0.3
0.9
Rest of Region
Africa
- 0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Central and South America —
Middle East
Europe
Eurasia
Asia
North America
World Total
- 0.0
- 0.0
- 0.0
- 0.0
- 0.0
- 0.1
0.0
0.0
0.0
0.0
0.0
0.2
0.0
0.0
0.0
0.0
0.0
0.2
0.0
0.0
0.0
0.0
0.0
0.4
0.0
0.0
0.0
0.0
0.0
0.7
0.0
0.0
0.0
0.0
0.0
0.7
0.0
0.0
0.0
0.1
0.0
0.8
0.0
0.1
0.0
0.1
0.0
1.4
0.0
0.1
0.0
0.1
0.0
1.6
0.1
0.3
0.1
0.2
0.0
4.2
Figure 9-4: Marginal Abatement Cost Curves for Top Five Emitters in 2030
Non-CO2 Reduction (MtCO2e)
Japan
•China
•Singapore
•South Korea
United States
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-143
-------
SEMICONDUCTOR MANUFACTURING
As stated earlier, early voluntary action by the semiconductor manufacturing industry resulted in a
dramatic decrease in the level of F-GHG emissions emitted in 2010 and later years as compared with
2000. However, emissions are expected to grow from the current level based on increased demand for
semiconductors over the next 20 years, and this may be particularly true in light of the new WSC
normalized emission rate goal The MAC analysis suggests that additional reductions from this sector are
costly. In the absence of any external climate policy drivers, major reductions in the semiconductor
manufacturing sector would require a significantly high carbon price (>$100/tCO2e) to incentivize
manufacturers to adopt additional mitigation options.
IV.9.5 Uncertainties and Limitations
A few key uncertainties exist with respect to the analysis for the semiconductor sector. The extent
of current abatement is unclear; there is no comprehensive published information on the extent
abatement systems are really in use in the industry. In addition, abatement system reduction efficiencies
assumed in this analysis are really only achievable if the systems are properly operated and maintained,
which may not always be the case. Also, abatement system reduction efficiencies may vary by gas (e.g.,
CF4 is harder to abate than other F-GHGs because of its relatively high thermo-stability, or bond
strengths). Finally, the pace at which the semiconductor manufacturing sector has advanced has been
historically very fast-paced. This continues to be true, but it cannot be certain that this will continue to be
true given the continued rising costs of advancement.
The limitations to this analysis are that it could not consider the full picture of emissions from
semiconductor manufacturing (e.g., heat transfer fluid emissions are not included), and that the new
WSC normalized emission rate goal was not known at the time of the analysis.
IV-144 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SEMICONDUCTOR MANUFACTURING
Applied Materials. 1999. Catalytic Abatement of PFC Emissions. Presented at Semicon Southwest 99: A
Partnership for PFC Emissions Reductions, October 18,1999, Austin, TX.
Beu, L. 2005. "Reduction of Perfluorocarbon (PFC) Emissions: 2005 State-of-the-Technology Report."
TT#0510469A-ENG. International SEMATECH Manufacturing Initiative (ISMI), Albany, New York.
December 2005. Obtained at: http://www.epa.gov/highgwp/semiconductor-
pfc/documents/final tt report.pdf.
Bureau of Labor Statistics. 2010. Overview of BLS Wage Data by Area and Occupation. Available at:
http://bls.gov/bls/blswage.htm
Burton, S. 2003. Personal communication with Brown of Motorola (2002) supplemented by personal
communication with Von Compel of BOC Edwards (2003), research of DuPont's Zyron Web site
(2003), and personal communication with Air Liquide regarding thermal destruction, NF3 remote
clean, and capture membrane unit costs.
Fthenakis, V. December 2001. Options for Abating Greenhouse Gases from Exhaust Streams. Brookhaven
National Laboratory. Obtained at: http://www.bnl.gov/isd/documents/23784.pdf.
Hattori et al., 2006. "Application of Atmospheric Plasma Abatement System for Exhausted Gas from
MEMS Etching Process." Presetned at the Institute of Electrical and Electronics Engineers
International Symposium on Semiconductor Manufacturing 2006, Tokyo, Japan, September 25-27,
2006.
Intergovernmental Panel on Climate Change (IPCC). 2006. 2006 IPCC Guidelines for National Greenhouse
Gas Inventories. The National Greenhouse Gas Inventories Programme (Volume 3, Chapter 6:
Electronics Industry Emissions), The Intergovernmental Panel on Climate Change, H.S. Eggleston, L.
Buendia, K. Miwa, T. Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
Motorola. October 18, 1998. Long-Term Evaluation of Litmus "Blue" Inductively-Coupled Plasma Device for
Point-of-Use PFC and HFC Abatement. Presented at Semicon Southwest 99: A Partnership for PFC
Emissions Reductions, Austin, TX.
U.S. Environmental Protection Agency (USEPA). 2009. Developing a Reliable Fluorinated Greenhouse
Gas (F-GHG) Destruction or Removal Efficiency (DRE) Measurement Method for Electronics
Manufacturing: A Cooperative Evaluation with IBM (EPA 430-R-10-004), Office of Air and Radiation
Office of Atmospheric Programs, Climate Change Division, U.S. Environmental Protection Agency,
Washington, DC. Available at: http://www.epa.gov/highgwp/semiconductor-
pfc/documents/ibm report.pdf.
U.S. Environmental Protection Agency (USEPA). 2010. Draft Emission Factors for Refined Semiconductor
Manufacturing Process Categories. Office of Air and Radiation Office of Atmospheric Programs,
Climate Change Division, U.S. Environmental Protection Agency, Washington, DC. Available in
docket EPA-HQ-OAR-2009-0927.
U.S. Environmental Protection Agency (USEPA). (2012). Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA. Obtained from:
http://www.epa.gov/climatechange/economics/international.html.
U.S. Environmental Protection Agency (USEPA). (2013). Data from the Greenhouse Gas Reporting
Program. 40 CFR Part 98. Washington, D.C. Available at: http://www.epa.gov/ghgreporting/ghgdata/
VLSI, 2012.VLSI Research, Inc. (2012) Worldwide Silicon Demand. August 2012.
WSC, 2011. Joint Statement of 15th Meeting of the World Semiconductor Council (WSC). Resulting from the
World Semiconductor Council in Fukuoka, Japan. May 26, 2011.
Bartos, Scott, Kshetry, Nina, and C. Shepherd Burton. 2007. Modelling China's Semiconductor Industry
Fluorinated Compound Emissions and Drafting a Roadmap for Climate Protection. Obtained at:
http://www.climatevision.gov/sectors/semi conductors/pdfs/ISESH_2007_Final.pdf
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-145
-------
SEMICONDUCTOR MANUFACTURING
Brown et. al., 2012. "Catalytic technology for PFC emissions control." Solid State Technology. Obtained
at: http://\v\v\v.electroiq.com/content/eiq-2/en/articles/sst/print/volunie-44/issue-7/features/emissions-
control/catalytic-technology-for-pfc-emissions-control.html. Accessed August 2012.
Molina, Wooldridge, and Molina. 1995. Atmospheric Geophysical Research Letters 22(13).
Other Sources Reviewed for this Analysis:
• Semiconductor Equipment and Materials Industry. 2011 World Fab Forecast, May 2011 Edition.
• Semiconductor Equipment and Materials Industry. 2012 World Fab Forecast, August 2012
Edition.
• WSC, 2012. Joint Statement of 16th Meeting of the World Semiconductor Council (WSC). Resulting from
the World Semiconductor Council in Saratoga Spring, New York, United States. May 24, 2012.
Clean Development Mechanism Proposed Methodologies Reviewed:
• NM0289: PFC gas emission reduction by gas replacement for CVD cleaning at 200 mm (8 inches)
process by Hynix Semiconductor Inc. (submitted September 2008)
• NM0303: PFC gas emissions reduction by gas replacement for CVD cleaning processes in
semiconductor processing operations (submitted April 2009)
• NM0317: Substitution of fluorinated compound (FC) gases for cleaning CVD reactors in the
semiconductor industry (submitted June 2009)
• NM0330: Substitution of fluorinated compound (FC) gases for cleaning CVD reactors in the
semiconductor industry (submitted December 2009)
• NM0332: PFCs emission reduction from installation of an abatement device in a semiconductor
manufacturing facility (submitted January 2010)
• NM0335: PFC emission reduction by gas replacement in the process of CVD cleaning in
semiconductor production (submitted February 2010)
IV-146 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS
IV.10. SFfi Emissions from Electric Power Svstems
IV.10.1 Sector Summary
lectric utilities use transmission and distribution equipment that contains sulfur hexafluoride
|(SF6). Equipment insulated with SF6 is most frequently found at electrical substations.
Emissions of SF6 occur as a result of leaking equipment and improper handling practices
during servicing and disposal.
Global SF6 emissions from electric power systems (EPSs) are expected to increase through 2030,
reaching 64 million metric tons of carbon dioxide equivalent (MtCC^e) (see Figure 10-1). In these
projections, China represents a significant share of total emissions by 2030. Brazil, India, South Korea, and
the rest of world increase their SF6 emissions marginally, while the United States experiences a decline
over the same time period.
Figure 10-1: SFe Emissions from Electric Power Systems: 2000-2030 (MtC02e)
64
South Korea
I Brazil
India
I United States
I China
ROW
2000
2010
2020
2030
Year
Source: U.S. Environmental Protection Agency (USEPA), 2012
The following technologies and handling practices can be implemented to reduce both causes of
emissions—leaking equipment and improper handling:
• Leak detection and leak repair (LDAR): Various monitoring and repair methods reduce gas
leakage from gaskets and faulty seals in equipment.
• Equipment refurbishment: Refurbishing old equipment reduces longer-term leakage problems
that cannot be addressed sufficiently by LDAR.
• SF6 recycling: Technicians transfer SF6 to special gas carts prior to maintenance or
decommissioning, reducing emissions that would otherwise result from the venting of SF6 to the
atmosphere.
• Improved SF6 handling: Employee training efforts that improve general handling practices of
SF6 to reduce and avoid instances such as accidentally venting the gas, using inappropriate
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-147
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS)
fittings to connect transfer hoses to cylinders or equipment, misplacing gas cylinders, and other
similar situations that result in handling losses.
Europe and Japan have largely adopted emission reduction measures to the greatest extent possible;
it is believed that few opportunities remain for reduction; data reported to the UNFCCC indicate a
downward trend of emissions within the last decade (UNFCCC, 2009). In the United States, SF6 recycling
is widely used, but there remains significant potential for reductions through other measures, particularly
improved SF6 handling. In the developing world, SF6 recycling is rarely conducted; therefore, there are
significant opportunities for reductions from increased SF6 recycling in addition to significant reduction
opportunities from improved SF6 handling (NCGC, 2010; NEPA, 2005). The most cost-effective
reductions can be achieved by improving general SF6 handling practices at EPSs in the developing world.
In these cases, the cost per ton is -$1.20/tCO2e. The most expensive emission reductions for the
developing world are from implementing LDAR at $1.98/tCO2e. Opportunities to reduce emissions in the
United States are more expensive, expected to range from -$0.20/tCO2e for improved SF6 handling to
$9.40/tCO2e for equipment refurbishment.
The manufacture of equipment for electrical transmission and distribution can also result in SF6
emissions, but this type of emission is not included in this assessment.
The global abatement potential in the EPS sector is 42.8 MtCO2e in 2030, which represents 67% of
projected baseline emissions. This represents the maximum level of reductions that are technically
achievable by applying the four abatement measures in the EPS sector. For example, leak detection and
leak repair is assumed to have a reduction efficiency of 50%, and is applied only to a the stream of
emissions that occur due to periodic leakage; other options have a greater reduction efficiency, but no
options are available to reduce 100% of emissions from all emission streams. Figure 10-2 presents the
global marginal abatement cost (MAC) curves charting the potential emission reductions in 2010, 2020,
and 2030.
Figure 10-2: Global Abatement Potential in Electric Power Systems: 2010, 2020, and 2030
$60
$50
$40
$30
•i, $20
O
u
5? $10
$0
-$10
-$20
-$30
•2010
•2020
•2030
10
15
20
25
30
35
40
45
Non-CO2 Reduction (MtCO2e)
IV-148
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS
In 2030, approximately 7.5 MtCO2e, or a 12% reduction in baseline emissions, is technically
achievable at a break-even price of $0/tCO2e. At $5/tCO2e, an additional 28 MtCO2e may be reduced,
equating to a cumulative reduction of 56% of the global emissions baseline. The remaining reductions of 7
MtCO2e are available at incrementally higher prices.
In the following sections of this chapter, we first characterize the source of SF6 emissions in the EPS
sector and the trends driving future emissions projections. Next, we discuss the projected baselines from
2010 to 2030. This is followed by a description of the abatement measures' engineering and cost
assumptions assumed for this analysis. Section IV.10.4 presents the additional assumptions used in the
MAC analysis unique to the EPS sector. The final section presents the MAC results in more detail and
discusses some of the uncertainties and limitations to the analysis.
IV.10.2 SF6 Emissions from Electric Power Systems
Emissions of SF6 from electrical equipment used in EPSs broadly occur through two routes:
equipment leakage and handling losses. Leakage losses can occur at gasket seals, flanges, and threaded
fittings and are generally larger in older equipment. Emissions from improper handling can include
intentional venting to the atmosphere or unintentional venting, such as transferring SF6 between
containers and equipment using improperly attached or improperly sized fittings. Figure 10-3 presents
the global distribution of SF6 emissions by emission stream assumed for this analysis. Leakage losses
correspond to periodic leakage from equipment (9%) and chronic leakage from equipment (23%).
Improper handling and venting losses correspond to venting gas during equipment maintenance and
disposal and improper handling. The break-out percentages are based on assumptions used to develop
the technical applicability of the options identified to mitigate these emission streams.
Figure 10-3: Percentage of Global SFe Emissions in 2020 by Emission Stream
(% of GWP-Weighted Emissions)
Residual
Emissions
6%
Improper
handling of SF6
38%
Chronic leakage
from
equipment
23%
Venting gas
during
equipment
maintenance
and disposal
24%
Periodic
leakage from
equipment
9%
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-149
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS)
The amount of SF6 gas that each piece of electrical transmission and distribution equipment can hold
when properly insulated is referred to as "nameplate capacity," which is measured in pounds or
kilograms of the gas. For the purpose of evaluating the cost of reducing SF6 emissions from EPSs, this
analysis considers reduction costs for a typical electric transmission and distribution system that uses SF6-
insulated electrical equipment totaling 100,000 pounds of nameplate capacity. The system includes a
variety of SF6-insulated electrical equipment (including circuit breakers, circuit switchers, and gas-
insulated substations), although the vast majority of SF6 is contained in high voltage circuit breakers.
Circuit breakers within the system are assumed to be produced by ABB, Alstom, HVB AE, Mitsubishi,
and Siemens, with an equal proportion of breakers from each manufacturer.
The analysis considers several possibilities for the maintenance and SF6 handling procedures used at
the typical electric transmission and distribution system, reflecting different levels of emissions. For the
purpose of this analysis, the three types of systems include the following:
• Residual emissions system: In Europe and Japan, abatement options are dose to fully
implemented. Therefore, a residual emissions system represents an EPS containing SF6-insulated
equipment located in Europe or Japan.
• Uncontrolled system: In contrast, abatement options have only been minimally applied or not
applied at all in most developing countries (Czerepuszko, 2011a; NCGC, 2010; NEPA, 2005;
Rothlisberger, 2011a). Therefore, the uncontrolled abatement system represents an EPS
containing SF6-insulated equipment located in a developing country, which for this analysis
would mean any country outside of Europe, Japan, and the United States.1
• Partially controlled system: Abatement options have been partially to fully applied in the United
States.2 The partially controlled system represents an EPS containing SF6-insulated equipment
located in the United States.
Figure 10-4 displays the breakdown of global emissions by system type as projected for 2020. For the
purpose of this analysis, the uncontrolled systems and partially uncontrolled systems are representative
of every system within their identifying regions. Therefore, the engineering cost results will not vary
among systems within the developing world or among systems within the United States.
IV.10.2.1 Activity Data or Important Sectoral or Regional Trends
The key activity data that drives SF6 emissions from EPSs is the amount of SF6-insulated electrical
equipment in use; this quantity is important for both leakage and handling losses.
However, data are not available on the total amount of SF6-insulated equipment currently in use or
historically in use at the country level. In the absence of such data, changes in the amount of SF6-insulated
electrical equipment in use (both historically and in the future) can be estimated from the historical and
projected changes in electricity demand at the regional level. This is because electricity demand is
correlated with the size of the electrical grid required to service that demand, and the size of the electrical
grid is correlated to the amount of SF6 consumed by utilities within the region. Thus, the key activity data
used to drive emissions is electricity demand. Other important activity data include the characteristics of
the equipment in use (such as age) to estimate leakage emissions and employee training and investments
in SF6 handling technologies to estimate handling emissions. In this analysis, these drivers are addressed
on a regional level.
1 Australia and New Zealand are considered to have uncontrolled systems, which may be one limitation to this
assumption.
2 This system is also assumed for Canada.
IV-150 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS
Figure 10-4: Global SFe Emissions in 2020 by Facility Type (% of GWP-Weighted Emissions)
Residual System
(Europe/Japan),
6%
artially
ntrolled
Abatement
System (U.S.),
22%
Uncontrolled
Abatement
System
(Developing),
73%
According to EIA (2009), electricity demand through 2030 is projected to grow two to three times
faster in developing countries than developed countries.
Leakage Emissions
Over the past 30 to 40 years, the amount of SF6 necessary for the operation of switchgear has declined
significantly and the tightness of equipment has improved. Such engineering design changes have
resulted in smaller leakage amounts and less frequent leakage over time (McCracken et al., 2000;
Rhiemeier et al., 2010). The average age of SF6-insulated equipment in developed countries (including
Europe, Japan, and the United States) is considerably older than in developing countries. Until recently,
developing countries were slow to adopt SF6-insulated equipment because of its relatively high cost
compared with other traditional technologies, such as oil-insulated circuit breakers (Rothlisberger,
2011b). Also, the electrical grid in developing countries has grown rapidly over the last decade with
economic growth, so the average age of all types of electrical equipment tends to be newer in developing
countries than in developed countries, whose electrical grid has historically grown at a more gradual
pace. The average SF6-insulated circuit breaker in developing countries, therefore, is assumed to leak less
than the average SF6-insulated circuit breaker in developed countries.
Handling Emissions
Employee training and investments in SF6 handling technologies (such as SF6 recovery carts) vary
widely among countries and regions. The use of equipment and accessories to properly handle SF6 is high
in developed countries (Rothlisberger, 2011a) yet low to nonexistent in at least some developing countries
(NCGC, 2010; NEPA, 2005). Employee training is perhaps strongest in Europe, where the European
Commission requires that personnel who handle SF6 receive formal training and certification (EC Reg.
No. 842/2006). In the United States, employee training has improved significantly since 1999 by
companies participating in the USEPA's voluntary SF6 Emission Reduction Partnership for Electric Power
Systems, but, in general, training is not as rigorous as in Europe, and it is uncertain what level of training
(if any) has been instituted by companies not part of the USEPA voluntary program. Employee training is
low to nonexistent in at least some developing countries (NEPA, 2005).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-151
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS)
IV. 10.2.2 Emission Estimates and Related Assumptions
Global SF6 emissions from EPSs in 2010 were estimated to be 44 MtCO2e, which represents a 10%
decrease from 1990 levels. This emissions decline, despite increases in the amount of SF6 in use over the
same time period, was based largely on improved management practices and the retirement of old leak-
prone equipment in the United States and EU. However, it is estimated that emissions have increased in
recent years because of the rapid increase in the amount of SF6-insulated equipment being used in the
developing world without the application of SF6 abatement technologies and practices. These emission
increases have been offset somewhat by the improved design of modern SF6-insulated equipment being
installed in the developing world. But from 2010 to 2030, global SFe emissions from EPSs are still
projected to increase 44% from 44 to 64 MtCO2e (see Table 10-1), driven largely by increases in emissions
from developing countries in Asia.
Table 10-1: Projected Baseline Emissions from Electric Power Systems: 2010-2030 (MtC02e)
Country
on
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS
region; as described above, changes in electricity demand correlate to changes in the electrical grid, which
correlate to changes in SF6 emissions.
IV.10.3 Abatement Measures and Engineering Cost Analysis
The four abatement options for this sector are SF6 recycling, LDAR, equipment refurbishment, and
improved SF6 handling. Replacing existing SF6-insulated equipment with newer equipment that holds
less SF6 and is more leak-tight is another possible abatement option; however, this mitigation practice is
assumed to already occur in the baseline. Given that the investment to replace a circuit breaker or other
SF6-containing equipment can be as high as a million to several million dollars, it is not examined in this
analysis for systems located in developing countries. All options are applicable to EPSs that are subject to
abatement (those outside of Europe and Japan). For the purpose of this analysis, four distinct emission
streams were analyzed for the sector, and each emission stream can only be abated by one of the
abatement options (the abatement options are not capable of abating emissions for any of the other
streams). Hence, the application of an abatement option to its unique emission stream does not affect the
applicability of any other options to their own emission streams. Table 10-2 shows the reduction
efficiency used for each abatement option.
Table 10-2: EPS Abatement Options
Abatement Option
SFe Recycling
LDAR
Equipment Refurbishment
Improved SFe Handling
Applicable System Types
Uncontrolled abatement system
Partially controlled system
Uncontrolled system
Partially controlled system
Uncontrolled system
Partially controlled system
Uncontrolled system
Partially controlled system
Reduction Efficiency
90%
50%
95%
90%
Table 10-3 presents the engineering cost data for each mitigation option outlined above, including all
cost parameters necessary to calculate the break-even price.
The characteristics, applicability, and key engineering cost results for each abatement option are
presented below. For additional information on these technologies see Appendix K.
IV.10.3.1 SF6 Recycling
This option involves transferring SF6 from electrical equipment into storage containers during
equipment servicing or decommissioning so that the SF6 can be reused. Recycling is conducted using an
SF6 reclamation cart (commonly referred to as a gas cart). The gas cart recovers the SF6 from the
equipment and purifies it for future use; the recovered and purified SF6 gas can then be stored within the
cart, in a separate storage container, or transferred back to the equipment for reuse. Proper recycling
techniques are documented in technical literature (CIGRE, 2005; IEC, 2008; IEEE, 2012). The alternative to
using a gas cart is venting the used SF6 into the atmosphere and then replacing it with fresh SF6. Venting
is typically performed in areas where environmental consideration is low because the cost of purchasing
new gas is often cheaper than purchasing gas carts and paying technicians to reclaim gas from
equipment.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-153
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS)
Table 10-3: Engineering Cost Data on a Facility Basis
Abatement Option
SFe Recycling
LDAR
Equipment
Refurbishment
Improved SFe
Handling
Project Lifetime
(Years)
-1C
c
on
1
Capital Cost
(2010 USD)
$479,560
$71,934
$95,420
$91,485
$126,069
$453,849
$13,526
$13,526
Annual Revenue
(2010 USD)
$67,994
$6,256
$12,592
$3,476
$9,570
$5,283
$90,659
$25,025
Annual O&M
Costs
(2010 USD)
$5,372
$19,937
$534
$6,339
—
—
$253
$2,508
Abatement
Amount (tC02e)
46,833
8,618
8,673
4,788
6,591
7,278
62,444
34,474
The SF6 recycling option addresses emissions that occur if SF6 contained inside equipment is vented
directly to the atmosphere, either because the equipment is undergoing a maintenance/repair activity
requiring removal of the gas or because the equipment is being decommissioned. Based on expert
judgment, SF6 vented to the atmosphere accounts for 30% of emissions from uncontrolled systems (in
developing countries) and 10% of emissions from partially controlled systems (in the United States).
SF6 recycling can reduce emissions by 4,320 pounds for the uncontrolled systems and 795 pounds for
the partially controlled systems. The lifetime of this abatement option is 15 years (Rothlisberger, 2011a).
Cost and revenue estimates for the SF6 recycling option are summarized below:
• Capital costs. The average total capital costs associated with the purchase of gas carts were
estimated to be about $480,000 for the uncontrolled system and $72,000 for the partially
controlled system. The cost per gas cart unit was the same for both systems at approximately
$96,000. Gas carts can range in cost from as low as $20,000 to as high as $175,000 depending on
their size (Rothlisberger, 2011a), and a mid-range gas cart size was assumed for both system
types in this analysis. The average capital costs for the partially controlled system are less than
the cost per unit, because U.S. systems have implemented SF6 recycling to a greater extent and
fewer gas carts are needed across U.S. systems: less than one per system.
• Annual operation and maintenance (O&M) costs. O&M costs were estimated to be $5,000 for the
uncontrolled system and $20,000 for the partially controlled system. The lower O&M costs for the
uncontrolled system were driven by the significantly lower labor cost in developing countries
relative to the United States.
• Annual revenue. Annual revenue, which was estimated based on the reduction of SF6 emissions
multiplied by the cost per pound of SF6 gas, was close to $68,000 for the uncontrolled system and
$6,000 for the partially controlled system. Annual revenues are significantly higher for the
uncontrolled system because the uncontrolled system has not implemented the option at all,
while the partially controlled system has implemented the option to 85% of its potential—
therefore the potential for reductions is greater. In addition, the cost of SF6 per pound varies
regionally and is relatively low in the United States (Rothlisberger, 2011a), so less money is saved
through reduced emissions.
• Technical Lifetime: The technical lifetime of this option is 15 years.
• Reduction Efficiency: The reduction efficiency for SF6 recycling is 90%.
IV-154
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS
IV. 10.3.2 Leak Detection and Leak Repair (LDAR)
LDAR is a two-step process. First, a leak detection technique is used to identify gas leaks from SF6-
insulated equipment. Leak detection methods vary and can involve simple techniques such as using soap
and water solutions to more sophisticated techniques such as those requiring cameras to visualize the
source of SF6 leaks by exploiting the strong infrared adsorption of SF6 for detection. Thermal imaging
cameras allow the detection of even minor leaks without the need to take equipment out of service. The
abatement option analyzed in this analysis is the use of a thermal imaging camera. Identified leaks are
typically repaired by applying a sealing material to the component that is leaking, although in some cases
the component needs to be replaced completely. The International Council on Large Electric Systems
(CIGRE), published "SF6 Tightness Guide" (Brochure No. 430) offers details on more specific methods for
leak detection and tightness procedures and test methods (CIGRE, 2010).
Emissions addressed by LDAR occur when a piece of equipment periodically develops a manageable
leak from a specific component such as a bushing flange gasket. Once the leak is repaired, the equipment
tends to last months to years without another major leak. SF6 emissions from periodic equipment leakage
account for 10% of emissions from both uncontrolled systems (in developing countries) and partially
controlled systems (in the United States) (Rothlisberger, 2011a; 2011b).
Emission reductions from LDAR were estimated to be 800 pounds for the uncontrolled system and
440 pounds for the partially controlled system. The lifetime of this abatement option is five years
(Czerepuszko, 2011a). Cost and revenue estimates for LDAR are summarized below:
• Capital costs. The capital costs associated with purchasing thermal imaging cameras were
estimated to be $95,000 for an uncontrolled abatement system and $91,000 for a partially
controlled system. The cost for a single thermal imaging camera was approximately $98,000
(Czerepuszko, 2011a).
• Annual O&M costs. O&M costs were estimated to be $540 for the uncontrolled system and
$6,300 for the partially controlled system. The lower O&M costs for the uncontrolled system were
driven by the significantly lower labor cost in developing countries relative to the United States.
• Annual revenue. Annual revenue, which was estimated based on the reduction of SF6 emissions
multiplied by the cost per pound of SF6 gas, was $12,600 for the uncontrolled system and $3,500
for the partially controlled system. Annual revenues are significantly higher for the uncontrolled
system primarily because it was assumed that the uncontrolled system has implemented the
option to a lesser extent than the partially controlled system—therefore the potential for
reductions is greater. In addition, because the cost of SF6 per pound varies regionally and costs
significantly less in the United States (Rothlisberger, 2011a), the cost of SF6 per pound is
significantly less for the partially controlled system relative to systems in other regions, and so
less money is saved through reduced emissions.
• Technical Lifetime: The technical lifetime of this option is five years.
• Reduction Efficiency: The reduction efficiency for LDAR is 50%.
IV.10.3.3 Equipment Refurbishment
Unlike LDAR, which tends to focus on small leaks on specific components such as a bushing flange
gasket, refurbishment addresses the need for a comprehensive repair from equipment that chronically
leaks large amounts of SF6 gas. Refurbishment is a process in which equipment is disassembled and
rebuilt (and possibly upgraded) using remachined, cleaned, and/or new components. The option is
focused mostly toward dual-pressure circuit breakers built before 1980, which hold large amounts of SF6
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-155
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS)
and were initially built with gasket material that corroded metal within the breaker (resulting in
numerous leaks over time).
Although the leaks can be temporarily repaired using the LDAR option (and often are), using LDAR
over time requires high servicing costs because of the extent of the LDAR required and the amount of gas
needed to replace the emitted gas. Using LDAR only for this category of equipment also increases the risk
for system reliability issues. Ultimately the equipment needs to be refurbished or replaced for these issues
to be solved.
Based on expert judgment, SF6 from chronically leaking equipment accounts for 20% of emissions
from uncontrolled systems (in developing countries) and 40% of emissions from partially controlled
systems (in the United States).
Emission reductions from equipment refurbishment were estimated to be 600 pounds for the
uncontrolled system and 670 pounds for the partially controlled system. The lifetime of this abatement
option was estimated to be 20 years based on the assumption that the average lifetime of new equipment
was 40 years, and the lifetime of refurbished equipment will be about half that of new equipment. Cost
and revenue estimates for equipment refurbishment are summarized below:
• Capital costs. The capital costs associated with equipment refurbishment were estimated to be
$125,000 for an uncontrolled system and $450,000 for a partially controlled system. The estimated
cost to replace a single 1,130-pound nameplate capacity circuit breaker was estimated to be
$143,000 (developed from McCracken et al. [2000]).
• Annual O&M costs. It was assumed that the equipment refurbishment is conducted off-site of
the system facility by the manufacturer and that there are no incremental O&M costs associated
with the equipment after it has been refurbished and returned to the EPS.
• Annual revenue. Annual revenue, which was estimated based on the reduction of SF6 emissions
multiplied by the cost per pound of SF6 gas, was $9,600 for the uncontrolled system and $5,300
for the partially controlled system. Annual revenues are significantly higher for the uncontrolled
system primarily because it was assumed that the uncontrolled system has implemented the
option to a lesser extent than the partially controlled system—therefore the potential for
reductions is greater. In addition, the cost of SF6 per pound is significantly less for the partially
controlled system, so less money is saved through reduced emissions.
• Technical Lifetime: The technical lifetime of this option is 20 years.
• Reduction Efficiency: The reduction efficiency for equipment refurbishment is 95%.
IV.10.3.4 Improved SF6 Handling
This option involves improving the procedures and techniques for handling SF6, especially when
maintenance is being performed on gas-insulated circuit breakers. Handling-related leaks can occur when
(1) inappropriate fittings are used to connect transfer hoses to cylinders or equipment; (2) SF6 is not
cleared from transfer hoses before the hoses are disconnected from cylinders/equipment; (3) gas cylinders
are not monitored/maintained because they have been misplaced or stored improperly; and (4) any time
SF6 is actidently vented by a technician. Improving SF6 handling involves both training technicians to
properly handle gas and purchasing adapter kits that ensure proper fittings are available for connecting
hoses to all gas-insulated equipment throughout the system.
SF6 emissions from handling-related leaks account for 40% of emissions from both uncontrolled
systems (in developing countries) and partially controlled systems (in the United States) (Rothlisberger,
2011a; 2011b).
IV-156 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS
Employee training and investments in SF6 handling technologies (such as SF6 recovery carts) are
measures that improve SF6 handling; several technical references are available with detailed guidance on
the proper techniques for recovering, disposing and other handling practices of SF6 gas (CIGRE, 2005;
IEC, 2008; IEEE, 2012). Emission reductions from improved SF6 handling were estimated to be 5,800
pounds for the uncontrolled system and 3,200 pounds for the partially controlled system. The lifetime of
this abatement option was one year, with training conducted on an annual basis (Rothlisberger, 2011a).
Cost and revenue estimates for the improved SF6 handling option are summarized below:
• Capital costs. The capital costs associated with improved SF6 handling were estimated to be
$13,500 for both the uncontrolled system and the partially controlled system. This capital cost
consists entirely of purchasing adapter kits, which were estimated to cost $1,350 each (middle of
cost range provided by Rothlisberger [2011a]).
• Annual O&M costs. O&M costs were estimated to be $250 for the uncontrolled system and
$2,500 for the partially controlled system. The lower O&M cost for the uncontrolled system was
driven by the significantly lower labor cost in developing countries relative to the United States.
• Annual revenue. Annual revenue, which was estimated based on the reduction of SF6 emissions
multiplied by the cost per pound of SF6 gas, was $91,000 for the uncontrolled system and $25,000
for the partially controlled system. Annual revenue was significantly higher for the uncontrolled
system primarily because it was assumed that the uncontrolled system has implemented the
option to a lesser extent than the partially controlled system—therefore the potential for
reductions is greater. In addition, the cost of SF6 per pound is significantly less for the partially
controlled system, so less money is saved through reduced emissions.
• Technical Lifetime: The technical lifetime of this option is 1 year.
• Reduction Efficiency: The reduction efficiency for improved SF6 handling is 90%.
IV.10.4 Marginal Abatement Costs Analysis
This section discusses the modeling approach and documents some additional assumptions used in
the MAC analysis for SF6 emissions reduction.
IV. 10.4.1 Methodological Approach
The MAC analysis applies the abatement measure costs discussed in the previous section for two
types of EPS systems, defined earlier as uncontrolled and partially controlled systems, to calculate a
break-even price for the options available for each EPS system.
IV.10.4.2 Definition of EPS Model Facilities
Key Characteristics of the Residual Emissions System
Facilities in Europe and Japan have been classified as residual emission systems. In these regions,
abatement options are close to fully implemented. The vast majority of the SF6 emissions that do occur
are considered residual emissions from occurrences such as catastrophic equipment failure and accidents
associated with gas handling (Rhiemeier et al., 2010). Japanese equipment designs and maintenance
practices are believed to be similar to those in Europe (Yokota et al., 2005). Country-reported data as
reported in UNFCCC inventory submissions for Europe and Japan show that SF6 emissions from electric
power systems have declined from 1990 through 2003. Emissions are expected to continue to decline in
these regions as utilities, through government-sponsored voluntary and mandatory programs,
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-157
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS)
implement reduction measures such as leak detection and repair and gas recycling practices (USEPA,
2012).
Key Characteristics of the Partially Controlled System
In the last decade, electric utilities have begun to voluntarily reduce SF6 emissions by applying the
abatement options included in this analysis (USEPA, 2011a). The baseline emission projections for the
partially controlled system were developed under the assumption that the adoption of abatement
technologies and practices in the United States will continue to grow into the future (as it has over the last
decade through voluntary efforts such as USEPA's SF6 Emission Reduction Partnership for Electric Power
Systems). For purposes of the engineering cost analysis, the emission rate for the partially controlled
system was 8.8%, which was the average U.S. emission rate in 2009 as estimated by USEPA (2011b). The
size of the partially controlled system was a typical medium-sized facility with 100,000 pounds of
installed SF6 nameplate capacity. The size of the system was chosen to yield realistic nominal abatement
and cost values because the size of a system in the United States does not significantly influence the
system's emission rate, baseline abatement levels, or other key characteristics.
Figure 10-5 presents 2010 emission rates from EPSs that participate in the USEPA voluntary
Partnership. Emission rates reported through the Partnership contributed to the average U.S. emission
rate of 8.8% used in this analysis (USEPA, 2011a).
Figure 10-5: Distribution of 2010 Emission Rates Reported through USEPA's Voluntary Partnership
100%
90%
0)
•§
o
+3
O
re
0)
E
o
0%
5%
10%
15%
20%
25%
30%
SFR Emission Rate
Source: USEPA, 2011 a
Key Characteristics of the Uncontrolled System
The baseline emission projections for developing countries were developed under the assumption
that the application of abatement technologies does not increase in the future. For purposes of the
engineering cost analysis, the assumed emission rate for the uncontrolled system was 16%, which is
approximately double the U.S. emission rate. The emission rate for the developing world is very
uncertain because a published emission rate based on actual measurements of emissions is not known.
The 16% emission rate was developed by considering a probable emission rate for an EPS in the United
States (for which emission rates are available) if that system had similar characteristics to the average
IV-158
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS
system in the developing world. The size of the uncontrolled system was a typical medium-sized system
with 100,000 pounds of installed SF6 nameplate capacity. The size of the system was chosen simply to
yield realistic nominal abatement and cost values because the size of a system in developing countries
does not significantly influence the system's emission rate, baseline abatement levels, or other key
characteristics.
Table 10-4 lists the countries or regions associated with each model facility system. The residual
emissions systems include Japan and Europe, which includes a majority of European Union member
countries, in additional to Norway and Switzerland. The United States and Canada represent partially
controlled systems, and the rest of the world reflects uncontrolled systems.
Table 10-4: EPSs System Country Mapping
Residual Emission Systems
Japan
Europe
Austria
Belgium
Bulgaria
Cyprus
Czech Republic
Denmark
Estonia
Finland
France
Germany
Greece
Hungary
Ireland
Italy
Latvia
Lithuania
Luxembourg
Netherlands
Norway
Poland
Partially Controlled Uncontrolled Systems
United States Rest of World
Canada
Portugal
Romania
Slovakia
Slovenia
Spain
Sweden
Switzerland
United Kingdom
IV.10.4.3
Parameters Used to Estimate Technical Effectiveness
The analysis also developed a technical effectiveness parameter, defined as the percentage reductions
achievable by each abatement measure/system type combination. Estimating this parameter requires
making a number of assumptions regarding estimates of technical applicability, market penetration, and
reduction efficiency. These assumptions are held constant for all model years. Table 10-5 presents the
technical applicability, market penetration, and reduction efficiency assumptions used to develop the
abatement measures' technical effectiveness.
IV.10.4.4 Estimating Abatement Project Costs and Benefits
The MAC model uses the estimated abatement project costs and benefits and technical lifetime as
described in Section IV.10.3 to calculate the break-even price for each mitigation option at each model
facility. Table 10-6 illustrates the break-even calculation for each abatement measure expressed in 2010
USD. Improved SF6 handling is the only options with a negative break-even price, also known as a "no-
regrets" option because the benefits of adopting the abatement measure outweigh the costs of
implementation. The remaining three options have break-even prices greater than $0/tCO2e.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-159
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS)
Table 10-5: Technical Effectiveness Summary
Abatement Option
Developing Countries
SFe Recycling
LDAR
Equipment Refurbishment
Improved SFe Handling
United States/Canada
SFe Recycling
LDAR
Equipment Refurbishment
Improved SFe Handling
• Technical
Applicability
30%
10%
20%
40%
10%
10%
40%
40%
Market
Penetration
100%
100%
20%
100%
100%
100%
20%
100%
Reduction
Efficiency
90%
50%
95%
90%
90%
50%
95%
90%
Technical
Effectiveness
27%
5%
4%
36%
9%
5%
8%
36%
Table 10-6: Example Break-Even Prices for Abatement Measures in EPSs
Abatement Option
r Reduced
Emissions
(tC02e)
Annualized
Capital Costs
($/tC02e)
Net Annual
Cost
($/tC02e)
Tax Benefit of
Depreciation
($/tC02e)
Break-Even
Price3
($/tC02e)
Developing Countries
SFe Recycling
LDAR
Equipment Refurbishment
Improved SFe Handling
46,833
8,673
6,591
62,444
$2.2
$4.8
$3.7
$0.4
-$1.3
-$1.4
-$1.5
-$1.4
$0.5
$1.5
$0.6
$0.1
$0.45
$1.98
$1.65
-$1.20
United States/Canada
SFe Recycling
LDAR
Equipment Refurbishment
Improved SFe Handling
8,618
4,788
7,278
34,474
$1.8
$8.4
$12.2
$0.7
$1.6
$0.6
-$0.7
-$0.7
$0.4
$2.5
$2.1
$0.3
$3.05
$6.45
$9.40
-$0.20
a Break-even prices were calculated using a tax rate of 40% and a discount rate of 10%.
IV.10.4.5 MAC Analysis Results
The global abatement potential for SF6 reductions in the EPS sector is estimated to be 43 MtCO2e,
which is 67% of total projected emissions in 2030. Table 10-7 presents the cumulative reductions achieved
at selected break-even prices. Figure 10-6 shows the MAC curve for the top five emitting countries in the
EPS sector. Over 83% of the maximum abatement potential is achieved at break-even prices below
$5/tCO2e in 2030.
IV-160
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS
Table 10-7: Abatement Potential by Country/Region at Selected Break-Even Prices in 2030 (MtC02e)
Country/Region
Break-Even Price ($/tC02
5 10 15 20
2)
30
^HHi^l
•filial
KuTin
Top 5 Emitting Countries
Brazil
China
India
South Korea
United States
— — —
— — —
— — —
- - 0.8
- - 3.7
1.7
16.6
2.7
1.4
3.7
1.7
16.6
2.7
1.4
3.7
1.7
16.6
2.7
1.4
3.7
1.7
16.6
2.7
1.4
3.7
1.7
17.9
2.7
1.4
3.7
1.9
18.9
2.9
1.5
4.6
2.0
18.9
3.0
1.5
4.6
2.0
18.9
3.0
1.5
5.9
Rest of Regions
Africa
Central and South America
Middle East
Europe
Eurasia
Asia
North America
World Total
- - 0.6
- - 0.4
- 0.0 0.7
- - 0.3
- - 0.1
- - 0.4
- - 0.5
- 0.0 7.5
1.8
1.0
1.9
0.6
0.7
3.0
0.9
35.9
1.8
1.1
2.0
0.6
0.7
3.0
0.9
36.0
1.8
1.1
2.0
0.6
0.7
3.0
0.9
36.0
1.8
1.1
2.0
0.6
0.7
3.0
0.9
36.0
2.0
1.1
2.0
0.6
0.7
3.0
0.9
37.6
2.0
1.2
2.2
0.6
0.8
3.4
1.0
41.0
2.0
1.2
2.2
0.7
0.8
3.4
1.0
41.5
2.0
1.2
2.2
0.7
0.8
3.4
1.0
42.8
a The World Total may not equal the sum of the country and region break-even prices due to small differences in rounding.
Figure 10-6: Marginal Abatement Cost Curves for Top Five and Rest of World Emitters in 2030
•Brazil
•China
•India
South Korea
•United States
•Rest of World
Non-CO2 Reduction (MtCO2e)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-161
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS)
IV.10.5 Uncertainties and Limitations
Despite a comprehensive literature review and correspondence with some of the most
knowledgeable representatives from the electric power transmission and distribution industry,
considerable uncertainty is associated with some of the engineering cost data used for this analysis.
Emission data account for the greatest area of uncertainty.
We are not aware of any published information on emission levels or rates in the developing world
that are based on actual measurements. Also, there is very limited information on the distribution of
emissions within a typical EPS because the system-level mass-balance approach (currently the standard
emissions monitoring method) does not track where or how emissions occur. The lack of reliable
continuous emission monitoring methods at specific points within EPSs also makes it difficult to
accurately monitor the reduction efficiencies associated with specific abatement options, so the reduction
efficiencies used for this analysis are based on judgments from industry experts rather than the studies
involving emissions monitoring. Much less uncertainty is associated with cost data because most cost
data were obtained directly from industry representatives that provide the equipment and services for
abating emissions.
IV-162 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS
References
Czerepuszko, P. May 13, 2011a. Documentation of meeting between Paul Czerepuszko of FLIR Systems
and Paul Stewart of ICF International.
Czerepuszko, P. May 23, 2011b. Documentation of meeting between Paul Czerepuszko of FLIR Systems
and Paul Stewart of ICF International.
International Council on Large Electric Systems (CIGRE). 2005. Guide for the Preparation of Customised
"Practical SF6 Handling Instructions." Brochure No. 276.
International Council on Large Electric Systems (CIGRE). 2010. SF6 Tightness Guide; Brochure No. 430.
International Electrotechnical Commission (IEC). 2008. High-voltage Switchgear and Controlgear - Part 303:
Use and Handling of Sulphur Hexafluoride (SF6). IEC Technical Report 62271-303.
Institute of Electrical and Electronics Engineers (IEEE). 2012. IEEE Guide for Sulphur Hexafluoride (SF6) Gas
Handling for High-Voltage (over 1000 Vac) Equipment. IEEE Std C37.122.3.
McCracken, G., R. Christiansen, and M. Turpin. 2000. The Environmental Benefits of Remanufacturing:
Beyond SF6 Emission Remediation. Obtained at: http://www.epa.gov/electricpower-
sf6/documents/confOO mccracken paper.pdf.
North China Grid Company (NCGC). 2010. SF6 Recycling Project of North China Grid—Project Design
Document. North China Grid Company. CDM Project 3707. Obtained at:
http://cdm.unfccc.int/Projects/Validation/DB/
MPHV4MMDLVU2IOE4ZP69TWKSRAT3HB/view.html.
National Electric Power Authority (NEPA). 2005. Reducing SF6 Emissions in High-Voltage
Transmission/Distribution Systems in Nigeria—Project Design Document. National Electric Power
Authority. NM0135. Obtained at:
http://cdm.unfccc.int/methodologies/PAmethodologies/pnm/byref/NM0135.
Rhiemeier, J., S. Wartmann, M. Pagnotta, N. Makowska, and X. Li. 2010. Update on Global SF6 Emissions
Trends from Electrical Equipment—Edition 1.1. Ecofys Emission Scenario Initiative on Sulphur
Hexafluoride for Electric Industry. Obtained at: http://www.ecofys.com/files/files/esi-
sf6 finalreport editionll 100701 vOl.pdf.
Rothlisberger, L. May 5, 2011a. Documentation of meeting between Lukas Rothlisberger of DILO and
Paul Stewart of ICF International.
Rothlisberger, L. June 2, 2011b. Documentation of meeting between Lukas Rothlisberger of DILO and
Paul Stewart of ICF International.
Smythe, K. December 1-3, 2004. Trends in SF6 Sales and End-Use Applications: 1961-2003. International
Conference on SF6 and the Environment: Emission Reduction Technologies, Scottsdale, AZ.
United Nations Frame work Convention on Climate Change (UNFCCC). 2009. United Nations
Framework Convention on Climate Change Flexible GHG Data Queries. Obtained at:
http://unfccc.int/di/FlexibleOueries/Setup.do.
U.S. Bureau of Labor Statistics (BLS). 2011. "International Comparisons of Hourly Compensation Costs in
Manufacturing." Press Release on December 21, 2011. Bureau of Labor Statistics, U.S. Department of
Labor. USDL-11-1778. Obtained at: http://www.bls.gov/news.release/pdf/ichcc.pdf.
U.S. Energy Information Administration (EIA). 2009. International Energy Outlook 2009. Energy
Information Administration, U.S. Department of Energy. Report* DOE/EIA-0484(2009). Obtained at:
http://www.eia.gov/oiaf/archive/ieo09/index.html.
U.S. Environmental Protection Agency (USEPA). (2011a). SF6 Emission Reduction Partnership for Electric
Power Systems—2010 Annual Report. U.S. Environmental Protection Agency. Obtained at:
http://www.epa.gov/electricpower-sf6/documents/sf6 2010 ann report.pdf.
U.S. Environmental Protection Agency (USEPA). (2011b). Inventory of U.S. Greenhouse Gas Emissions
and Sinks: 1990-2009. U.S. Environmental Protection Agency. USEPA M30-R-11-005. Obtained at:
http://epa.gov/climatechange/ernissions/usinventorvreport.html.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-163
-------
SF6 EMISSIONS FROM ELECTRIC POWER SYSTEMS)
U.S. Environmental Protection Agency (USEPA). (2012). Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA. Obtained from:
http://www.epa.gov/dimatechange/econornics/international.htrnl.
Yokota, T., K. Yokotsu, K., Kawakita, H. Yonezawa, T. Sakai, and T. Yamagiwa. 2005. Recent Practice for
Huge Reduction of SF6 Gas Emissions from GIS&GCB in Japan. Presented at the CIGRE SC A3 & B3
Joint Colloquium in Tokyo.
IV-164 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
MAGNESIUM PRODUCTION
IV.11.1 Sector Summary
Ihe magnesium metal production and casting industry uses SF6 as a cover gas to prevent the
spontaneous combustion of molten magnesium in the presence of air. Fugitive SF6 emissions
occur primarily during three magnesium manufacturing processes: primary production, die-
casting, and recycling-based production. Additional processes that may use SF6 include sand
and gravity casting, as well as wrought, anode, and permanent mold casting; however, these are not
included in this analysis.
Between 2000 and 2010, global SF6 emissions from magnesium manufacturing have decreased 50%,
from 10 million metric tons of carbon dioxide equivalent (MtCO2e) to 5 MtCO2e (USEPA, 2012). Over this
time period, magnesium production has increased, but growth has been offset by major initiatives to
phase out the use of SF6 in magnesium production in numerous countries. As Figure 11-1 shows, from
2010 to 2030, emissions from magnesium production are projected to stay in the range of approximately
5 MtCO2e (USEPA, 2012).
Figure 11-1: SFe Emissions from Magnesium Production: 2000-2030 (MtC02e)
12 i
Ukraine
I Israel
Kazakhstan
I Russia
I China
ROW
2000
2010
2020
2030
Year
ROW = Rest of World
Source: U.S. Environmental Protection Agency (USEPA), 2012.
Global abatement potential of SF6 in the magnesium manufacturing sector is 5 MtCO2e in 2030, which
is approximately 98% of the projected emissions. Figure 11-2 presents the sector marginal abatement cost
(MAC) curves for 2010, 2020, and 2030. Three potential options are available for reducing SF6 emissions
from magnesium production and processing operations. These emission abatement measures all include
substituting SF6 with an alternate cover gas: SO2, HFC-134a, or Novec™ 612.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-165
-------
MAGNESIUM MANUFACTURING
Figure 11-2: Global Abatement Potential in Magnesium Manufacturing: 2010, 2020, and 2030
$60
$50
$40
$30
«N $20
8
$10
$0
-$10
-$20
-$30
•2010
2020
•2030
Non-CO2 Reduction (MtCO2e)
This chapter follows a structure similar to previous chapters, starting with a description of the
industrial activity, facility types, and source of emissions, followed by a discussion of the projected
emissions out to 2030. Section IV.11.3 characterizes the abatement measures by providing a brief
description of each option and information on their costs and performance assumptions. The chapter
concludes with a discussion of regional MAC results.
IV.11.2 SF6 Emissions from Magnesium Manufacturing
Use of SF6 as a cover gas is the only source of emissions from magnesium production. Although
studies indicate some destruction of SF6 in its use as a cover gas (Bartos et al., 2003), the analysis
described here follows current Intergovernmental Panel on Climate Change (IPCC) guidelines (IPCC,
2006), which assumes that all SF6 used is emitted to the atmosphere. This analysis uses three model
facilities to define magnesium production across die casting, primary production, and reprocessing
(recycle/remelt) facilities. Global SF6 emissions from magnesium production by facility type are shown in
Figure 11-3. Model facilities are based on industry data from the United States, but apply to magnesium
facilities globally.
For the purpose of evaluating the cost of reducing SF6 emissions from magnesium production, this
analysis considers reduction costs for three typical magnesium production facilities—die casting,
recycle/remelt, and primary production, which were generally characterized based on facility-specific
case studies measuring average SF6 consumption, production capacity, and type. We characterize these
typical facilities as follows:
IV-166
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
MAGNESIUM PRODUCTION
Figure 11-3: Global SFe Emissions in 2020 by Facility Type (% of GWP-Weighted Emissions)
Recycle, 2%
• Die Casting Facility: This model facility represents a medium-sized die casting facility. The
facility is characterized based on real data from a case study (USEPA, 2011) where a given
abatement option was implemented in 2008. The facility produces 26,014 metric tons of
magnesium per year, and emits 0.17 kg of SF6 per metric ton of magnesium produced,
representing a total annual facility emission of 4,483 kg of SF6. Production and emissions data
from 2007 is used to define the pre-abatement emissions baseline (USEPA, 2011). Emissions data
was reported by the facility based on its consumption of SF6, assuming all SF6 used is emitted to
the atmosphere.
• Recycle/Remelt Facility: This model facility represents a medium-sized recycle facility. The
facility is characterized based on real data from a case study where a given abatement option was
implemented in 2008. The facility produces 18,453 metric tons of magnesium per year, and emits
1.09 kg of SF6 per metric ton of magnesium produced, representing a total annual facility
emission of 20,026 kg of SF6. Production and emissions data from 2007 is used to define the pre-
abatement emissions baseline (USEPA, 2011). Emissions data were reported by the facility based
on its consumption of SF6, assuming all SF6 used is emitted to the atmosphere.
• Primary Production Facility: Assumes the same characteristics as the die casting facility.
IV.11.2.1 Activity Data or Important Sectoral or Regional Trends
The primary activity data for this sector are the quantities of magnesium produced or processed.
Between 1990 and 2010, global SF6 emissions from magnesium manufacturing have decreased 58%, from
12 MtCO2e to 5 MtCO2e (USEPA, 2012). Over this time period, magnesium production has increased, but
this growth has been offset by major initiatives to phase out the use of SF6 in magnesium production in
numerous countries.
From 2010 to 2030, emissions from magnesium production are projected to stay in the range of
approximately 5 MtCC^e (USEPA, 2012). Emissions from Organisation for Economic Co-operation and
Development (OECD) countries decrease substantially in the short term because of facility closures in
North America and SF6 phaseout efforts (U.S. Geological Survey [USGS], 2011). As a result, the OECD
share of global SF6 emissions from magnesium manufacturing is projected to decrease from 40% in 2010
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-167
-------
MAGNESIUM MANUFACTURING
to 12% in 2030. Major SF6 phaseout efforts are driven by the USEPA's voluntary partnership in the United
States, and the regulatory directives in Japan and Europe.
SF6 emissions from magnesium manufacturing in non-OECD Asia are projected to increase
significantly between 2010 and 2030, increasing the region's global share of emissions from 20% to 44%.
Emissions in the non-OECD Europe and Eurasia region experience similar growth. The overall increase in
non-OECD Asia's share of global emissions results from an increase in Chinese primary magnesium
production and die casting fueled by local and foreign investment. China's emissions growth is driven by
their die casting operations as well as by the share of China's primary production (approximately 10%)
that is assumed to use SF6 as the cover gas mechanism. Emissions from Central and South America are
driven by production in Brazil. Brazil's emissions were estimated to have declined considerably since
implementation of a Clean Development Mechanism project after 2005 involving a switch to SO2 as the
cover gas (UNFCCC, 2010).
IV.11.2.2 Emission Estimates and Related Assumptions
Global emissions from the magnesium production sector were 5.13 MtCO2e in 2010, growing to 5.22
MtCO2e in 2030. Emission estimates for U.S facilities were based on magnesium production statistics and
specific emissions factors for each manufacturing process using data from the USEPA SF6 Emission
Reduction Partnership (USEPA, 2011) and USGS (2011). As per IPCC 2006 guidelines, it is assumed that
all SF6 used as a cover gas is emitted. Data used in this analysis on magnesium production and cover gas
use for a typical facility were taken from a case study on U.S. facilities (USEPA, 2011) and may vary for
facilities in other countries. Table 11-1 presents projected emissions between 2010 and 2030 by country
and region.
Table 11-1: Projected Baseline Emissions from Magnesium Production: 2010-2030 (MtC02e)
Country /Region
2010
2015
2020
2025
2030
CAGRa
(2010-2030)
Top 5 Emitting Countries
China
Russia
Kazakhstan
Israel
Ukraine
1.2
0.9
0.4
0.4
0.1
1.6
1.1
0.5
0.4
0.1
2.1
1.2
0.5
0.5
0.1
2.2
1.4
0.5
0.5
0.1
2.3
1.6
0.6
0.6
0.1
3.2%
2.6%
1.7%
1.7%
1.7%
Rest of Region
Africa _____ _
Central and South America
0.1
0.0
0.0
0.0
0.0
—
Middle East _____ _
Europe
0.2
0.0
0.0
0.0
0.0
-22.2%
Eurasia _____ _
Asia
North America
World Total
0.6
1.3
5.1
0.6
0.4
4.6
0.6
0.2
5.1
0.0
0.1
4.8
0.0
0.1
5.2
-20.8%
-13.3%
0.1%
aCAGR = Compound Annual Growth Rate
Source: USEPA, 2012
IV-168 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
MAGNESIUM PRODUCTION
IV.11.3 Abatement Measures and Engineering Cost Analysis
Three potential options are available for reducing SF6 emissions from magnesium production and
processing operations. These emission abatement measures all include replacing SF6 with an alternative
cover gas: SO2, HFC-134a, or Novec™ 612. Table 11-2 presents the reduction efficiency and applicability
for the three alternative cover gas options.
The remainder of this section provides an overview of each abatement option and details the cost and
reduction assumptions.
Table 11-2: Magnesium Production Abatement Options
Abatement Option
Alternative cover gas—Novec™ 612
Reduction Efficiency
100%
Applicability
Die casting
Recycle/remelt
Primary production
Alternative cover gas—HFC-134a
95%
Die casting
Recycle/remelt
Primary production
Alternative cover gas—S02
100%
Die casting
Recycle/remelt
Primary production
IV.11.3.1 Replacement with Alternative Cover Gas—Sulfur Dioxide (SO2)
Historically, SO2 has been used as a cover gas in magnesium production and processing activities.
However, because of toxitity, odor, and corrosivity concerns, SO2 use was discontinued in most
countries. Current SO2 technology research aims to improve process feed systems and control technology,
as well as to address the toxicity and odor issues with improved containment and pollution control
systems (Environment Canada, 1998). The use of SO2 has the potential to reduce SF6 emissions by 100%
because a complete replacement of the cover gas system is involved. Currently, SO2 is being used as a
cover gas; for example, it is used as a cover gas at one diecasting facility in Brazil (UNFCCC, 2010). This
option is assumed to be technically applicable to all three model facilities. The maximum market
penetration for this option is assumed to be 80% of the emissions of SF6 for recycle/remelt facilities, and
10% for both die casting and primary production facilities. The lifetime of this option is assumed to be 15
years.
Facilities implementing SO2 as an alternative cover gas incur capital costs related to the cost for new
piping, pollution control equipment, and safety equipment for workers. The total capital cost was
$490,781 for all three facility types. Facilities also incur annual costs (or generate annual cost savings)
based on the purchase price of the alternate cover gas. This option results in annual gas purchase costs of
$16,763 each for die casting and primary production facilities and an annual gas purchase cost of $74,833
for recycle/remelt facilities. SO2 is significantly less expensive than SF6, and the required gas replacement
ratio is 1:1, resulting in a significant net savings in material costs. Replacing SF6 with SO2 also results in
avoided costs of $131,633 each for both die casting and primary production model facilities and $588,018
for the recycle/remelt model facility associated with the purchase of SF6.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-169
-------
MAGNESIUM MANUFACTURING
IV.11.3.2 Replacement with Alternative Cover Gas—HFC 134a
Research has shown that candidate fluorinated compounds such as HFC-134a can be a cover gas
substitute for SF6 (Milbrath, 2002; Ricketts, 2002; Hillis, 2002). In addition, currently, HFC-134a is used as
a cover gas at two diecasting facilities in Israel (UNFCCC, 2008a, 2008b).While fluorinated gases have an
advantage over SO2 because they have potentially fewer associated health, safety, odor, and corrosive
impacts, some current fluorinated gas alternatives (including HFC-134a) still have global warming
potential (GWP). However, the GWP of HFC-134a is significantly less than that of SF6: thus, the GWP-
weighted cover gas emissions could be reduced by 95%. HFC-134a is assumed to be technically
applicable to all model facilities. The maximum market penetration for this option is assumed to be 45%
of the emissions of SF6 for die casting and primary production facilities, and 10% for recyde/remelt
facilities. The lifetime of this option is assumed to be 15 years.
Facilities implementing HFC-134a as an alternative cover gas do not incur up-front capital costs, as
use of HFC-134a is a simple drop-in option and does not require additional/new systems or training.
They incur annual costs (or generate annual cost savings) based on the purchase price of the alternate
cover gas. Use of HFC-134a results in annual gas purchase costs of $32,908 each for die casting and
primary production facilities and $147,005 for the recyde/remelt facility. HFC-134a is not only less
expensive than SF6, but additionally HFC-134a has a gas replacement ratio of 0.5:1, resulting in significant
net savings in material costs. Replacing SF6 with HFC-134a results in avoided costs of $131,633 each for
both die casting and primary production model facilities and $588,018 for the recycle/remelt fadlity
associated with the purchase of SF6.
IV.11.3.3 Replacement with Alternative Cover Gas—Novec™ 612
Research has shown that candidate fluorinated compounds such as Novec™ 612 can be a cover gas
substitute for SF6 (Milbrath, 2002; Ricketts, 2002; Hillis, 2002). Additionally, currently, Novec™612 is
being used at one die casting facility in the United States. The use of Novec™ 612 as an alternative cover
gas represents an advantage over SO2 because, like other fluorinated gases, Novec™ 612 has potentially
fewer assodated health, safety, odor, and corrosive impacts. Novec™ 612 is a zero GWP gas and
therefore has a reduction effidency of 100% compared with SF6. Novec™ 612 is assumed to be technically
applicable to all model facilities.
Facilities implementing Novec™ 612 as an alternative cover gas incur capital costs related to the
purchase of computerized mass flow control cabinets and piping material to dired the gas. The total
capital cost was $245,390 for the die casting fadlity, $33,128 for the recyde/remelt facility, and $496,916
for the primary produdion fadlity. Fadlities also incur annual costs (or generate annual cost savings)
based on the purchase price of the alternate cover gas. Use of Novec™612 results in annual gas purchase
costs of $60,754 for die casting and primary production facilities and $271,393 for the recyde/remelt
fadlity. However, because the replacement ratio of Novec™ 612 to SF6 is 0.3:1, significantly less Novec™
612 is required to process the same quantity of magnesium. These costs are offset by the avoided costs of
purchasing SF6, an annual cost savings of $131,633 for both die casting and primary production model
facilities and $588,018 for the recycle/remelt model fadlity.
IV.11.3.4 Summary of Mitigation Technology Costs and Characteristics
Table 11-3 presents all of the data needed to calculate the break-even price for the options analyzed.
All options have an assumed lifetime of 15 years.
IV-170 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
MAGNESIUM PRODUCTION
Table 11-3: Engineering Cost Data on a Facility Basis
Abatement
Option
S02
HFC-134a
Novec™ 612
Facility Type
Die casting
Recycle/remelt
Primary production
Die casting
Recycle/remelt
Primary production
Die casting
Recycle/remelt
Primary production
r Project
Lifetime
(Years)
15
15
15
15
15
15
15
15
15
Capital Costs
(2010 USD)
$490,781
$490,781
$490,781
—
—
—
$245,390
$33,128
496,916
Annual
Savings*
(2010 USD)
$131,633
$588,018
$131,633
$131,633
$588,018
$131,633
$131,633
$588,018
$131,633
Annual O&M
Costs (2010
USD)
$16,763
$74,883
$16,763
$32,908
$147,005
$32,908
$60,754
$271,393
$60,754
Abatement
Amount
(tC02e)
107,144
478,621
107,144
101,316
452,588
101,316
107,139
478,601
107,139
1 These numbers are not net annual savings.
IV.11.4 Marginal Abatement Costs Analysis
This section discusses the modeling approach and documents some additional assumptions used in
the MAC analysis for magnesium production.
IV. 11.4.1
Methodological Approach
The MAC analysis applies the abatement measure costs discussed in the previous section of this
chapter at three magnesium production facility types to calculate a break-even price for each option at
each facility (i.e., die casting, recycle/remelt, and primary production). This section presents detailed
information on how each type of facility was defined in this analysis and detailed information on how
costs were built out for each mitigation technology.
IV. 11.4.2
Model Facilities Defined
The break-even cost analysis is conducted on three model facility types defined as follows:
• Die casting facility—Represents medium-sized facility currently in production in the United
States where abatement option was implemented in 2008. Annual production is assumed to be 26
kilo tons. The annual SF6 usage rate was 0.17kg/ ton produced (based on data reported by facility
under USEPA SF6 Emission Reduction Partnership). This emission factor multiplied by the
annual production of magnesium yields annual facility emissions of approximately 4.4 tons of
SF6 (equal to 107,550 tCO2e).
• Recycle/Remelt facility—Represents medium-sized facility currently in production in the United
States where abatement option was implemented in 2008. Production data is from 2007, prior to
implementation of abatement option. Annual production is assumed to be 18 kilo tons. The
annual SF6 usage rate was 1.09 kg/ ton of magnesium produced (based on data reported by
facility under USEPA SF6 Emission Reduction Partnership). This emission factor multiplied by
the annual production yields annual facility emissions of approximately 20 tons of SF6.
• Primary production facility—Assumes similar characteristics as the die casting model facility.
Annual magnesium production is 26 kilotons and a SF6 usage rate of 0.17 kg/ton of production.
The model primary production facility annual emissions are 4.4 tons of SF6
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-171
-------
MAGNESIUM MANUFACTURING
IV. 11.4.3
Assessment of Technical Effectiveness
For this analysis, we developed a technical effectiveness parameter, defined as the percentage
reductions achievable by each technology/facility-type combination. Table 11-4 lists the assumptions
regarding technical applicability, market penetration, and technical effectiveness of each option.
Table 11-4: Technical Effectiveness Summary
Abatement Option
Technical
Applicability
Market
Penetration
Reduction
Efficiency
Technical
Effectiveness
Die Casting Facility
Alternative cover gas— Novec™ 612
Alternative cover gas— HFC-134a
Alternative cover gas— 862
100%
100%
100%
45%
45%
10%
100%
95%
100%
45%
43%
10%
Recycle/Remelt Facility
Alternative cover gas— Novec™ 612
Alternative cover gas— HFC-134a
Alternative cover gas— S02
100%
100%
100%
10%
10%
80%
100%
95%
100%
10%
9%
80%
Primary Production Facility
Alternative cover gas— Novec™ 612
Alternative cover gas— HFC-134a
Alternative cover gas— S02
100%
100%
100%
45%
45%
10%
100%
95%
100%
45%
43%
10%
We assume that all three abatement measures are technically applicable to all facility types, hence the
technical applicability factor of 100%. Market penetration rates were assumed based on expert judgment.
For example, for die casting facility experts believe that a facility would adopt both Novec™612 and
HFC-134a over SO2, with an equal chance of adopting either Novec™612 or HFC-134a. The same
assumptions were made for a primary production facility. For recyde/remelt facility, experts believed
that there would be a preference for SO2 over the other two alternative cover gases, with an equal chance
of adopting either Novec™612 or HFC-134a. Multiplying the technical applicability, market penetration,
and reduction efficiency for each alternative cover gas at each facility type produces the technical
effectiveness estimates for each facility type. These assumptions are held constant for all model years.
IV.11.4.4 Estimating Abatement Project Costs and Benefits
The MAC model uses the estimated abatement project costs and benefits as described in Section
IV.11.3 to calculate the break-even price for each mitigation option at each model facility. Table 11-5
illustrates the break-even calculation for each abatement measure expressed in 2010 USD. Die casting and
recycle facility types have negative break-even prices for all three abatement measures. The only positive
break-even price estimated was for alternate cover gas—Novec™ 612 when applied to the primary
production facility due to a higher initial capital cost compared with other facility types. The remaining
two abatement measures applied to the primary production facility have negative break-even prices.
IV.11.4.5 MAC Analysis Results
The global abatement potential for SF6 reductions in the magnesium manufacturing sector is
estimated to be 98% of total projected emissions in 2030. Table 11-6 presents the cumulative reductions
achieved at selected break-even prices. Figure 11-4 shows the MAC curve for the top five emitting
countries for this sector. Total abatement potential is achieved at break-even prices below $5/tCO2e in
2030.
IV-172
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
MAGNESIUM PRODUCTION
Table 11-5: Example Break-Even Prices for Abatement Measures in Magnesium Manufacturing
Annualized Net
Reduced Capital Annual
Emissions Costs Cost
Tax Benefit Break-
of Even
Depreciation Price
Abatement Option
(tC02e)
($/tC02e) ($/tC02e) ($/tC02e)
($/tC02e)
Die Casting Facility
Alternative cover gas— Novec™ 612
Alternative cover gas— HFC-134a
Alternative cover gas— S02
Recycle/Remelt Facility
Alternative cover gas— Novec™ 612
Alternative cover gas— HFC-134a
Alternative cover gas— 862
Primary Production Facility
Alternative cover gas— Novec™612
Alternative cover gas— HFC-134a
Alternative cover gas— S02
107,139
104,230
107,144
478,601
465,605
478,621
107,139
104,230
107,144
$0.50
$0.00
$1.00
$0.02
$0.00
$0.22
$1.02
$0.00
$1.00
-$0.66
-$0.95
-$1.07
-$0.66
-$0.95
-$1.07
-$0.66
-$0.95
-$1.07
$0.10
$0.00
$0.20
$0.00
$0.00
$0.05
$0.21
$0.00
$0.20
-$0.26
-$0.95
-$0.27
-$0.65
-$0.95
-$0.89
$0.15
-$0.95
-$0.27
Table 11-6: Abatement Potential by Country/Region at Selected Break-Even Prices in 2030 (MtC02e)
Country/Region
Break-
-10
Even Pr
-5
ce ($/tC
0
02e)
5
10 15 20 30 50
100
100+
Top 5 Emitting Countries
China
Israel
Kazakhstan
Russia
Ukraine
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
2.3
0.5
0.6
1.6
0.1
2.3
0.5
0.6
1.6
0.1
2.3
0.5
0.6
1.6
0.1
2.3
0.5
0.6
1.6
0.1
2.3
0.5
0.6
1.6
0.1
2.3
0.5
0.6
1.6
0.1
2.3
0.5
0.6
1.6
0.1
2.3
0.5
0.6
1.6
0.1
Rest of Region
Africa
Central and South America
Middle East
Europe
Eurasia
Asia
North America
World Total
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
0.0
—
0.0
—
0.0
0.1
5.1
—
0.0
—
0.0
—
0.0
0.1
5.1
—
0.0
—
0.0
—
0.0
0.1
5.1
—
0.0
—
0.0
—
0.0
0.1
5.1
—
0.0
—
0.0
—
0.0
0.1
5.1
—
0.0
—
0.0
—
0.0
0.1
5.1
—
0.0
—
0.0
—
0.0
0.1
5.1
—
0.0
—
0.0
—
0.0
0.1
5.1
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-173
-------
MAGNESIUM MANUFACTURING
Figure 11-4: Marginal Abatement Cost Curves for Top Five Emitters in 2030
8
+•*
vv
$o
-$30
0.5
1.0
1.5
2.0
2.5
•China
Kazakhstan
•Israel
•Russia
Ukraine
Non-CO2 Reduction (MtCO2e)
IV.11.5 Uncertainties and Limitations
As per IPCC guidelines (2006), this analysis assumes that all cover gas used is emitted during
magnesium production. However, any SF6 destruction that occurs during use would result in lower
emission estimates than currently assumed in this analysis. In addition, this analysis uses data available
from U.S. facilities that implemented the three abatement options available. Although the data are
representative of abatement costs for U.S. facilities, the data may not be equally applicable to facilities in
other countries. Finally, uncertainties are associated with the emission estimates, which are detailed in
the Global Emissions Report.
IV-174
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
MAGNESIUM PRODUCTION
References
Bartos, S., J. Marks, R. Kantamaneni, C. Laush. 2003. Measured SF6 Emissions from Magnesium Die
Casting Operations. Magnesium Technology 2003, Proceedings of The Minerals, Metals & Materials
Society (TMS) Conference, March 2003.
Environment Canada. 1998. Powering GHG Reductions through Technology Advancement. Clean Technology
Advancement Division, Environment Canada.
Hillis, J.E., 2002. The International Program to Identify Alternatives to SF6 in Magnesium Melt Protection.
Presented at the International Conference on SF6 and the Environment: Emissions Reduction
Technologies, November 21-22, San Diego, CA.
Intergovernmental Panel on Climate Change (IPCC). 2006. 2006 IPCC Guidelines for National
Greenhouse Gas Inventories. The National Greenhouse Gas Inventories Programme, The
Intergovernmental Panel on Climate Change, H.S. Eggleston, L. Buendia, K. Miwa, T. Ngara, and K.
Tanabe (eds.). Hayama, Kanagawa, Japan.
Meridian. 2011. Personal Communication between Charles Woodburn of Meridian Magnesium Products
and Neha Mukhi of ICF International. May 24, 2011.
Milbrath, D. 2002. Development of 3M Novec 612 Magnesium Protection Fluid as a Substitute for SF6 in the
Magnesium Industry. Presented at the International Conference on SF6 and the Environment:
Emissions Reduction Technologies, November 21-22, San Diego, CA.
Ricketts, N. 2002. Environmental Implications of Using HFC-134a as a Replacement for SF6 in the Magnesium
Industry. Presented at the International Conference on SF6 and the Environment: Emissions Reduction
Technologies, November 21-22, San Diego, CA.
United Nations Framework Convention on Climate Change (UNFCCC). 2008a. SF6 Switch at Ortal
Diecasting 1993 Ltd. Production. United Nations Framework Convention on Climate Change.
Obtained May 2010 at: http://cdm.unfccc.int/Projects/DB/TUEV-SUED1233931497.2/view.
United Nations Framework Convention on Climate Change (UNFCCC). 2008b. SF6 Switch at Dead Sea
Magnesium. United Nations Framework Convention on Climate Change. Obtained May 2010 at:
http://cdm.unfccc.int/Projects/DB/TUEV-SUED1235638608.46/view.
United Nations Framework Convention on Climate Change (UNFCCC). 2010. Conversion of SF6 to the
Alternative SO2 at RIMA Magnesium Production. United Nations Framework Convention on Climate
Change. Obtained May 2010 at: http://cdm.unfccc.int/Projects/DB/TUEV-SUED1239262577.48/view.
U.S. Environmental Protection Agency (USEPA). (2012). Global Anthropogenic Non-COi Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA. Obtained from:
http://www.epa.gov/climatechange/economics/international.html.
U.S. Environmental Protection Agency (USEPA). 2011. Case Study: U.S. Magnesium Industry Adopts
New Technology for Climate Protection. Washington, DC: USEPA. In process.
U.S. Geological Survey (USGS). 2011. Minerals Yearbook 2010: Magnesium. GPO Stock #024-004-02538-7,
Reston, VA.
Werner K., and D. Milbrath. 2011. Novec™ 612 as a Substitute for SF6 in Magnesium Processing:
Experience to Date in Varied Casting Operations. 3M.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-175
-------
PHOTOVOLTAIC CELL MANUFACTURING
IV.12. Emissions from Photovoltaic Cell Manufacturin
IV.12.1 Sector Summary
he photovoltaic (PV) cell manufacturing process can use multiple fluorinated greenhouse gases
(F-GHGs) during production, including nitrogen trifluoride (NF3) and the
perfluorocompounds (PFCs) carbon tetrafluoride (CF4) and perfluoroethane (C2F6,). During the
manufacture of PV cells some of the F-GHGs not used in processes are released to the atmosphere.
F-GHG emissions from PV cell manufacturing are estimated to be approximately 2.3 million metric
tons of carbon dioxide equivalents (MtCC^e) in 2010. As Figure 12-1 shows, by 2030, emissions from the
manufacturing of PV cells are expected to decrease to 1.9 MtCO2e. The baseline projections presented
here are updated relative to those presented in Global Anthropogenic Non-CO2 Greenhouse Gas Emissions:
1990 to 2030 (USEPA, 2012). The updates incorporate new market information which has resulted in
significantly lower emission estimates. The decrease in emissions can be attributed to lower expected
growth in production and lower fraction of production assumed to use F-GHGs. Emissions projections
for this sector are particularly uncertain to due limited information on emissions rates, use of fluorinated
gases, production growth rates, and policies to encourage renewable energy development.
Figure 12-1: F-GHG Emissions from PV Cell Manufacturing: 2000-2030 (MtC02e)
3 -i
2 -
2 -
O
u
1 -
1 -
2.29
2.11
1.87
0.02
• Malaysia
• Germany
• United States
• Japan
• China
Rest of World
2000
2010 2020
Year
2030
Source: Update of projections presented in USEPA, 2012.
Four mitigation technologies were considered in this analysis as options for reducing F-GHG
emissions from PV manufacturing: thermal abatement systems, catalytic abatement systems, plasmas
abatement systems, and the NF3 remote chamber clean process. Due to the lack of mitigation cost
information specific to PV production, data is drawn from experience reducing emissions from similar
processes in semiconductor manufacturing.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-177
-------
PHOTOVOLTAIC CELL MANUFACTURING
The marginal abatement cost (MAC) analysis estimates a global mitigation potential of 1.7 MtCO2e,
based on the four abatement measures. The abatement potential represents 90% of the projected
emissions in 2030. Figure 12-2 presents the global MAC curves for 2010, 2020, and 2030 for the PV
manufacturing sector.
Figure 12-2: Global Abatement Potential in PV Cell Manufacturing: 2010, 2020, and 2030
$900
$800
•2010
•2020
•2030
0.5
1.0 1.5
Non-CO2 Reduction (MtCO2e)
2.0
2.5
High capital costs and relatively low emissions reduction amounts result in relatively high break-
even prices for abatement measures in the PV manufacturing sector relative to other industrial process
sectors. As a result, break-even prices in this sector are all greater than $120/tCO2e.
The following sections of this chapter first describe the activities and sources of F-GHG emissions in
the PV manufacturing sector and present the projected emissions for 2010 to 2030. Subsequent sections
characterize the four abatement measures considered and present the engineering cost assumptions used
in the MAC analysis. This is followed by a discussion of the MAC modeling assumptions that were used
to estimate the global abatement potential. We conclude the chapter by presenting the MAC curves for a
select number of individual countries and discuss some of the major uncertainties and limitations.
IV.12.2 Emissions from Photovoltaic Cell Manufacturing
PV manufacturing may use F-GHGs, thereby resulting in F-GHG emissions, including CF4, C2F& and
NF3, from etching and chamber cleaning processes used during the manufacture of PV cells. Etching is
done on various substrates, including crystalline silicon, amorphous silicon, and other thin-films. CF4 and
C2F6 are used during the manufacture of some crystalline silicon (c-Si) PV cells, and NF3 is used during
the manufacture of amorphous silicon (a-Si) and tandem a-Si/nanocrystaline (nc) silicon PV cells.
However, not all poly-silicon manufacturing process use F-GHGs, this was taken into consideration in the
analysis. Processes for PV cells manufactured on other thin films do not require the use of GHGs;
therefore, these processes were not considered in this analysis.
IV-178
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
PHOTOVOLTAIC CELL MANUFACTURING
The emission estimates presented in Figure 12-1 represent a piece of the total life cycle emissions
associated with manufacturing PV cells. One motivation of using PV cells is the production of reliable
low carbon energy, so it is not only important to consider the life cycle GHG emissions but also to
consider the benefits from using PV cells versus traditional fossil generated power. The European
Photovoltaic Industry Association (EPIA) analyzed the life cycle of a PV cell (from material sourring,
through manufacturing, transportation, construction, operation, dismantling and to product collection
and recycling into account) and estimated that "The carbon footprint of PV systems —assuming a location
in southern Europe—ranges from 16 to 32 gCO2 eq. per kWh compared to between 300 and 1000 g CO2
eq. per kWh when produced from fossil fuels." EPIA also estimates that solar power will still have a
carbon footprint of 10 to 20 times less than traditional fossil-based power with carbon capture and
storage. While solar power is a good low-carbon alternative to fossil based power and over the lifetime of
the cell it has climate benefits over traditional power sources, there is still potential to make it even more
beneficial. According to EPIA "The carbon footprint of PV has decreased by approximately 50% in the
last 10 years thanks to performance improvements, raw material savings and manufacturing process
improvements" (EPIA, 2011). Implementing PFC and NF3 mitigation strategies offers an opportunity to
further decrease the carbon footprint of PV cells, particularly given the high potency of these gases.
To evaluate the cost of reducing F-GHG emissions from the manufacture of PV cells, this analysis
considers the emissions and reduction costs for a typical PV manufacturing facility, characterized based
on an average facility capacity and the applicability of various mitigation technologies to etch and clean
emissions. The facility has an average capacity of 80 megawatts (MW) with an estimated 25 tools with 3.5
chambers. The facility uses only three F-GHGs: CF4, C2F6, and NFs.1 Figure 12-3 shows the breakdown of
etch and clean emissions for the typical facility.
Figure 12-3: Global F-GHG Emissions in 2020 by Facility Type (% of GWP-Weighted Emissions)
1 Although these gases are used for different PV technologies, for simplicity in this analysis, one general facility
producing an unidentifiable PV technology was considered.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-179
-------
PHOTOVOLTAIC CELL MANUFACTURING
IV.12.2.1 Activity Data and Important Sectoral/Regional Trends
Several industry trends which will influence future emissions from this sector include the rate of
growth in PV panel production, the overall penetration of PV into the global energy market, and the
relative proportion of various PV technologies with different rates of F-gas usage or emissions rates.
Current market trends indicate an increase and then a dip in production, and therefore, increases
followed by decreases in annual emissions. Despite many slowing sectors because of the global economic
slowdown, the PV sector continued to exceed expected growth rates while other sectors saw slowdowns.
According to the Global Market Outlook for Photovoltaics Until 2016, the main reasons for the large historic
and continued growth in the PV sector are: 1. "Renewable energy has continued to prove itself to be a
mainstream energy source and a significant contributor to achieving energy, environmental and
economic goals"; 2. "Some countries have increased their focus on renewable energy standards in the
wake of the Fukushima nuclear disaster, requiring them to consider new policies that move the market in
this direction"; 3. "PV modules have undergone significant price decreases..."; and 4. "In some countries,
questions about the future of support-scheme levels has produced boom-and-bust cycles." Also, many
countries are just starting to tap into the installed potential for PV. These factors account for the fact that
there was large general growth across all regions of the world. However, in 2012, the Congressional
Research Service noted that "The creation of incentives for solar installations in several countries around
2004 led many companies to enter the PV industry. According to an estimate by Photon International,
more than 1,000 PV module manufacturers worldwide supplied the market in 2011. But with demand in
some countries declining and prices weak, the industry appears to have entered a phase of rapid
consolidation on a global basis" (CRS, 2012).
The world saw booms in PV cell manufacturing and installation. This growth however, led to an
oversupply of panels starting in 2011. Balancing out supply with demand for solar panels has and
resulted in factories to close, and will also result in a continued decrease in manufacturing (Forbes, 2012).
There is significant uncertainty in whether production levels will remain relatively constant in the future
or resume the rapid annual increases which the industry experienced prior to 2011.
The projections presented in this chapter assume that production levels will be sufficient to achieve
the cumulative installed capacities from the "New Policy" Scenario of World Energy Outlook 2012,
without accounting for installed capacity replacement. These assumptions result in annual production in
2030 decreased to 22.6 GW compared to 24.6 GW in 2010. While there is a growth in demand for solar
energy, reflected in forecasted growth in total global installed solar capacity from about 38 gigawatts
(GW) in 2010 to 491 GW in 2030 (IEA, 2012), this is expected to be met through already existing
uninstalled stock of solar panels and future annual production. Figure 12-1 presents the business as usual
emissions from 2000 to 2030 for the five largest emitting countries and the rest of world (ROW).
Uncertainty regarding future policies, panel prices, and PV technology improvement result in
particularly uncertain projections of future production and associated emissions.
IV.12.2.2 Emissions Estimates and Related Assumptions
Emissions resulting from PV manufacturing processes were estimated using projected PV cumulative
installed capacities from the New Policy Scenario of World Energy Outlook, 2012; capacity and efficiency
data from the DisplaySearch PV Database; an assumed solar constant of 1,000 watts per meter squared;
Intergovernmental Panel on Climate Change (IPCC) Tier 1 emission factors for PV manufacturing; and an
NF3 emission factor from Fthenakis et al. (2010) and an emission factor developed using sensitive process
information. There is little to no variation in manufacturing processes and practices or in emissions trends
on a regional basis.
IV-180 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
PHOTOVOLTAIC CELL MANUFACTURING
F-GHG emissions from PV manufacturing are estimated to be approximately 2.29 MtCO2e in 2010. By
2030, emissions from manufacturing PV cells are expected to decrease to 1.87 MtCO2e. This decrease can
be attributed to the decrease in the annual PV production for 2030 compared to 2010. The annual
production in 2030 is expected to decrease to 22.6 GW compared to 24.6 GW in 2010.
In December of 2012 EPA published the Global Anthropogenic Non-CO2 Greenhouse Gas Emissions:
1990-2030. The baseline emission estimates presented in this analysis have been revised, and the revised
estimates are used for the MAC analysis presented in this report. Like the previous estimates, since little
literature is available describing the mitigation activities in the PV cell manufacturing sector, and unlike
the semiconductor and FPD manufacturing sectors, the PV manufacturing sector has not set a voluntary
reduction goal, the revised estimates do not include assumptions about the use of mitigation technologies
for crystalline silicon type manufacturing. However, unlike the previous estimates, the baseline now
assumes that half of the production process for amorphous silicon (a-Si) use abatement based on new
literature published (Fthenakis et al., 2010). The baseline also assumes that half of the production process
for crystalline silicon (c-Si) technology uses and emits CF4 and C2F6 during manufacturing, as not all PV
processes use F-GHGs.
The projected revised emissions are estimated based on annual PV production, differentiated by type
of technology, country, and the emission factors for respective types. However, now the future annual PV
production is estimated from the projected cumulative installed capacity obtained from World Energy
Outlook, New Policy Scenario (IEA, 2012) as opposed to assumed growth rates. The total annual PV
production is then differentiated into various types by dividing the total according technology
proportions: 77% crystalline Silicon (c-Si), 12% amorphous Silicon (a-Si) and the rest Cadmium Telluride,
Copper Indium Gallium Selenide and other categories, which are from the DisplaySearch database
(DisplaySearch, 2009). Similarly, the capacity is apportioned by country based on the DisplaySearch
database (DisplaySearch, 2009). Like the previous baseline methodology, PV production capacity for each
country for historic years was extracted from DisplaySearch database (DisplaySearch, 2009). The area of
cell produced is estimated based on the cell efficiency, using new data obtained from IEA Solar
Photovoltaic Roadmap (IEA, 2010), and an assumed solar constant of 1000 W/m2.
The NF3 emission factor for a-Si has been updated in the revised baseline estimation methodology.
For the a-Si technology, it was assumed that all facilities use NF3 and out of those, 50% of the facilities
have abatement devices installed. The emission factor for abated facilities was derived from data
presented in literature looking at lifecyde NF3 emissions of PV cells (Fthenakis et al., 2010). The emission
factor for unabated facilities was developed using sensitive process information. Lastly, the revised
baseline assumes that only 50% of c-Si technology is assumed to use and emit F-GHGs during
manufacturing and none of the other technologies are assumed to use and emit F-GHGs.
IV.12.3 Abatement Measures and Engineering Cost Analysis
Four mitigation technology options were identified for the PV manufacturing sector: thermal
abatement, catalytic abatement, plasma abatement, and NF3 remote chamber clean.
• Thermal abatement: These point-of-use abatement systems, that use heat to destroy or remove F-
GHGs from effluent process streams, are connected directly to a manufacturing tool.
• Catalytic abatement: Tool effluent process streams are run through abatement systems with
catalysts (e.g., CuO, ZnO, A12O3) that destroy or remove F-GHGs.
• Plasma abatement: Plasma in a point-of-use abatement system is used to react (destroy or
remove) F-GHGs from the process effluent stream.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-181
-------
PHOTOVOLTAIC CELL MANUFACTURING
Table 12-1: Projected Baseline Emissions from PV Cell Manufacturing: 2010-2030 (MtC02e)
Country
2010
2015
2020
2025
2030
CAGRa
(2010-2030)
Top 5 Emitting Countries
China
Japan
United States
Germany
Malaysia
1.0
0.3
0.2
0.3
0.2
1.2
0.3
0.2
0.3
0.1
1.2
0.2
0.2
0.2
0.1
1.1
0.2
0.2
0.1
0.1
1.2
0.2
0.2
0.1
0.1
0.8%
-1.5%
0.1%
-5.6%
-5.0%
Rest of Region
Africa ______
Central and South America ______
Middle East
Europe
Eurasia
Asia
0.0
0.1
0.0
0.2
0.0
0.1
0.0
0.2
0.0
0.1
0.0
0.1
0.0
0.1
0.0
0.1
0.0
0.1
0.0
0.1
-5.0%
-5.0%
-5.0%
-5.0%
North America ______
World Total
2.3
2.4
2.1
1.9
1.9
-1.0%
aCAGR= Compound Annual Growth Rate
Source: Updated from projections presented in USEPA, 2012
• NF3 remote chamber clean: Highly ionized NF3 is used to dean chemical vapor deposition
chambers. This process is highly efficient (-98%), resulting in lower emissions on a mass and CO2
basis than traditional in-situ chamber clean processes with utilization efficiencies around 20% to
50% (IPCC, 2006).
These technologies reduce emissions from either etch or chamber clean processes, or in some cases both.
Table 12-2 presents the applicability and the reduction efficiency of each abatement measure. The next
sections describe each of these mitigation options and additional detail is provided in Appendix M.
Table 12-2: PV Cell Manufacturing Abatement Options
Abatement Option
Thermal abatement
Catalytic abatement
Plasma abatement
NFs remote chamber clean
Applicable 3s)
Etch and clean
Etch and clean
Etch
Clean
Reduction
Efficiency
95%
99%
97%
95%
Information Source
Fthenakis (2001), Beu (2005), and USEPA (2009)
Fthenakis (2001), Brown et al. (2012)
Fthenakis (2001), Hattori etal. (2006)
Beu (2005)
IV.12.3.1
Thermal Abatement
Thermal abatement systems can be used to abate F-GHG emissions from both etching and chamber-
cleaning processes. The use of thermal abatement offers the benefit of not affecting the manufacturing
process (Applied Materials, 1999); however, the systems do require space that may not be available in
IV-182
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
PHOTOVOLTAIC CELL MANUFACTURING
some facilities. In addition, these systems require large amounts of cooling water, and the use of the
systems result in regulated NOX emissions.
The total facility capital cost for installing thermal abatement systems is estimated to be $5.7 million.
This estimate includes costs for the systems, the necessary ducting, water recirculation and hook up, and
natural gas costs (Fthenakis, 2001; Burton, 2003). The annual operating cost is estimated to be $328,860 at
the facility level. No annual cost savings are associated with using this technology
IV.12.3.2 Catalytic Abatement
A catalytic abatement system uses a catalyst to destroy or remove F-GHG emissions from the
effluents of both plasma etching and chemical vapor deposition (CVD) chamber-cleaning processes. This
type of abatement is applicable at most facilities, but again there may be some space constraints as
mentioned for thermal abatement systems. Additionally, because these systems are based on destruction
via catalyst, they must be process/stream specific to achieve the 99% emission reductions quoted in the
literature and used in this analysis (Fthenakis, 2001; Burton, 2003).
Because catalytic destruction systems operate at relatively low temperatures, their use results in little
or no NOX emissions, and the required amounts of water are low as well. Because of the high cost of
catalyst replacement, these systems are the least widely used type of abatement (expert judgment).
The capital cost associated with the purchase and installation of the abatement systems is estimated
to be $6.9 million per fab (i.e., facility). To use catalytic abatement systems, facilities must factor in the
annual cost of resources such as water, waste chemicals, electricity, and catalyst replacements. To cover
these operating expenses, a facility manufacturing PV cells is estimated to incur an annual cost of
$455,280. As with other abatement technologies considered in this sector, the use of catalytic abatement
systems will not result in annual cost savings.
IV.12.3.3 Plasma Abatement
These systems are applicable to etch processes in most facilities, with some physical space limitations.
(These systems though are relatively smaller in size compared with thermal and catalytic systems.)
Plasma abatement systems use a small plasma source that effectively dissociates the F-GHG molecules
that react with fragments of the additive gas (hydrogen (H2), oxygen (O2), water (H2O), or methane
(CH4)) to produce low molecular weight by-products such as hydrogen fluoride (HF) with little or no
GWP. After disassociation, wet scrubbers can remove the molecules. The presence of additive gas is
necessary to prevent later downstream reformation of PFC molecules (Burton, 2003).
The capital cost for plasma abatement systems is estimated to be $1.8 million per facility (Burton,
2003), which covers the purchase and installation of plasma systems. Plasma abatement systems require
an annual operation cost of $1,190 per chamber, which includes general maintenance and use of the
systems. Total annual facility costs are $51,850 based on an assumed four chambers per tool and 25 tools
per facility. The use of plasma abatement systems will not result in annual cost savings.
IV.12.3.4 NF3 Remote Chamber Clean
NF3 remote chamber clean is an alternative cleaning technology that has the benefit of having a
particularly high (-98%) utilization rate of NF3 (IPCC, 2006), resulting in relatively low emissions
compared with traditional chamber cleans. (The stated utilization is based on utilizations for
semiconductor manufacturing; as a result of large gas flows in PV manufacturing, the actual utilization
may be lower.) NF3 remote clean systems dissociate NF3 using argon gas and converting the source gas to
active F-atoms in the plasma upstream of the process chamber. These electrically neutral atoms can
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-183
-------
PHOTOVOLTAIC CELL MANUFACTURING
selectively remove material in the chamber. The by-products of remote dean include HF, fluorine (F2),
and other gases, most of which are removed by facility acid scrubber systems.
It is assumed that PV facilities are not "NF3 ready"; in other words, these facilities do not have the
current infrastructure to handle the direct installation of NF3 remote systems because this is a relatively
new technology. Therefore, facilities incur capital costs, in addition to system costs, associated with items
such as gas hook ups and necessary hardware such as manifolds and values. The facility cost is estimated
to be $9.2 million. The annual facility cost for NF3 remote dean is estimated to be $3.4 million (Burton,
2003). These costs are associated with the purchase of larger volumes of gas (NF3 versus traditional
chamber dean gases), general maintenance, and F2 scrubs to remove the highly explosive gas from the
effluent. No annual cost savings are associated with using this technology.
IV.12.3.5 Summary of Mitigation Technology Costs and Characteristics
Table 12-3 summarizes the information used to estimate the break-even prices in the MAC analysis.
Table 12-3: Engineering Cost Data on a Facility Basis
Abatement Option
Thermal abatement
Catalytic abatement
Plasma abatement
NFs remote clean
Project
Lifetime
(Years)
7
7
7
25
Capital Costs
(2010 USD)
$5,701,971
$6,906,594
$1,814,664
$9,200,867
Annual Revenue
(2010 USD)
$0
$0
$0
$0
Annual O&M
Costs
(2010 USD)
$328,862
$455,277
$51,848
$3,374,861
Abatement
Amount
(tC02e)
13,625
14,199
1,398
14,427
IV.12.4 Marginal Abatement Costs Analysis
This sedion discusses the modeling approach and documents some additional assumptions used in
the MAC analysis for the PV manufacturing sedor.
IV. 12.4.1
Methodological Approach
The MAC analysis applies the abatement measure costs discussed in the previous section of this
chapter at two hypothetical fadlities to calculate a break-even price for each abatement measure. This
sedion presents how we defined the model fadlity used in this analysis and the technical effediveness
assumptions used to estimate the incremental reductions for each measure. This sedion also provides an
example of how the break-even prices were calculated for each option.
IV.12.4.2 Definition of Model Facility
The manufacture of PV uses F-GHGs depending on the particular substrate and process used in the
produdion. Substrates used in the industry include crystalline silicon, amorphous silicon, and other thin
films. F-GHGs are used during the manufacture of crystalline silicon (c-Si) PV cells, amorphous silicon (a-
Si), and tandem a-Si/nanocrystaline (nc) silicon PV cells. Other thin film PV technologies do not require
the use of F-GHGs. As with the other eledronics manufacturing sedors, emissions in this sector result
from two main types of manufacturing processes: etching substrates and cleaning CVD chambers.
Manufaduring processes and uses of GHGs across the industry are generally similar; therefore, only one
type of model facility was considered for this analysis.
IV-184
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
PHOTOVOLTAIC CELL MANUFACTURING
• Model: The model facility is a facility that represents a PV manufacturing facility of average
manufacturing capacity (DisplaySearch, 2009) of 80 MW with an estimated 25 tools with 3.5
chambers. The facility uses only three F-GHGs: CF4, C2F6, and NF3.2 The emissions breakdown
for a PV manufacturing facility is estimated to be 10% etch emissions and 90% clean emissions.
The model facility emission breakdown is essential to this analysis, because some mitigation
technologies are applicable to either both or just one type of manufacturing process.
IV.12.4.3
Assessment of Technical Effectiveness
Similar to the methods employed for analyzing abatement in the semiconductor manufacturing
sector, this analysis developed a technical effectiveness parameter, defined as the percentage reductions
achievable by each technology/process combination. Estimating this parameter required making a
number of assumptions regarding the distribution of emissions by manufacturing process (etch and
clean), in addition to process-specific estimates of technical applicability and market penetration. These
assumptions are held constant for all model years. Table 12-4 presents the technical applicability, market
penetration, and reduction efficiency assumptions used to develop the abatement measures' technical
effectiveness parameters.
Table 12-4: Technical Effectiveness Summary
Etch (10%)
Technical Market
Abatement Measure Applicability Penetration
Clean (90%)
Technical Market
Applicability Penetration
Reduction Technical
Efficiency Effectiveness
Thermal abatement
Catalytic abatement
Plasma abatement
NFs remote clean
85%
85%
85%
0%
65%
10%
25%
0%
85%
85%
0%
100%
20%
10%
0%
70%
95%
99%
97%
95%
20%
8%
2%
60%
The technical effectiveness is the weighted average of the abatement measures using the emissions
attributed to each process (i.e., 10% etching, and 90% cleaning) as the weight multiplied by the product of
the technical applicability, market penetration, and reduction efficiency for each abatement measure.
The technical effectiveness estimates are then multiplied by the share of total emissions to estimate
the abatement potential achievable under each abatement measure.
IV.12.4.4 Estimating the Break-Even Price of Abatement Measures
The MAC model uses the estimated abatement project costs and benefits as described in Section 12.3
to calculate the break-even price for each abatement measure. Table 12-5 illustrates the break-even
calculation for each abatement measure expressed in 2010 USD.
2 Although these gases are used for different PV technologies, for simplicity in this analysis, one general facility
producing an unidentifiable PV technology was considered.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-185
-------
PHOTOVOLTAIC CELL MANUFACTURING
Table 12-5: Example Break-Even Prices for Abatement Measures in PVCell Manufacturing
Abatement Option
Thermal abatement
Catalytic abatement
Plasma abatement
NFs remote clean
(tC02e)
13,625
14,199
1,398
14,427
Capital Costs
($/tC02e)
$143
$167
$444
$117
Cost
($/tC02e)
$24
$32
$37
$234
Tax Benefit of
Depreciation
($/tC02e)
$40
$46
$124
$17
Break-Even
Price3
($/tC02e)
$128
$152
$358
$334
a Break-even price calculated using a tax rate of 40% and discount rate of 10%.
As Table 12-5 shows, the high capital intensity of the abatement measures coupled with no annual
benefits results in break-even prices that are all well above $100/tCO2e. These significantly higher break-
even prices suggest that achieving emission reductions in the PV manufacturing sector will require
additional incentives or regulations to control F-GHG emissions.
IV.12.4.5 MAC Analysis Results
The global abatement potential for F-GHG reductions in the PV manufacturing sector is estimated to
be 1.7 MtCO2e, or 90% of total projected emissions in 2030. Table 12-6 presents the cumulative reductions
achieved at selected break-even prices for the top five emitting countries and the grouping of countries
that make up the rest of each region. Figure 12-4 shows the MAC curves for the top five emitting
countries and the rest of world for this sector.
Table 12-6: Abatement Potential by Country/Region at Selected Break-Even Prices in 2030 (MtCO^e)
Country/Region -10 -5
Break-Even Price ($/tC02e)
0 5 10 15 20 30 50 100 100+
Top 5 Emitting Countries
China — —
Germany — —
Japan — —
Malaysia — —
United States — —
________ 1.1
________ 0.1
________ 0.2
________ 0.1
________ 0.2
Rest of Region
Africa — —
--------0.0
Central and South America ___________
Middle East ___________
Europe — —
Eurasia — —
Asia — —
--------0.0
--------0.0
________ 0.1
North America ___________
World Total 0.0 0.0
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.7
IV-186
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
PHOTOVOLTAIC CELL MANUFACTURING
Figure 12-4: Marginal Abatement Cost Curves for Top Five Emitters and Rest of World in 2030
.,600
1,000
8
$600
$200
•China
Japan
•Germany
•Malaysia
•United States
+
+
+
0.0 0.2 0.4 0.6 0.8
Non-CO2 Reduction (MtCO2e)
1.0
1.2
As the results show, the abatement potential in the PV manufacturing sector can be significant.
Unfortunately, these reductions in the absence of additional regulatory or market incentives would only
be achievable at significantly high break-even prices (>$200/tCO2e).
IV.12.5 Uncertainties and Limitations
The PV industry is a relatively new manufacturing sector, with high levels of growth to meet
continually growing demands for renewable energy. There is no comprehensive published information
on the extent abatement systems are really in use in the industry. Assumed abatement system reduction
efficiencies are really only achievable if the systems are properly operated and maintained, which may
not always be the case (USEPA, 2008a and 2008b). Also, abatement system reduction efficiencies may
vary by gas (e.g., CF4 is harder to abate than other F-GHGs because of its thermo-stability). Additionally,
there are not known voluntary reduction initiatives for the PV sector. Because of this the model facility is
uncontrolled, information about the use of abatement in baseline emissions is highly uncertain.
Other reasons for uncertainties inherent to the baseline emission estimates include assumptions about
the portion of the PV manufacturing industry that uses F-GHGs, and the unpredictability in the growth
of the solar PV production capacity. The activity data used, the cumulative solar PV capacity, is modeled
with its own set of assumptions from the IEA and is framed by the fast-growing renewable energy sector.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-187
-------
PHOTOVOLTAIC CELL MANUFACTURING
As this is the foundation of future baseline emission estimates, it gives rise to uncertainties. Furthermore,
market dynamics will also contribute to the fluctuations with respect to facilities' utilization rates as well
as fractions of PV cells started and inventoried, all of which are assumed constant for purposes of
developing baseline emissions. Another limitation is that the baseline emission estimates do not take into
account the retiring of PV cells. It is assumes that any new manufacturing is done to meet the increase in
installed capacity and not to replace any replacement of capacity The emission estimates are hence
conservative as inclusion of this assumption would lead to slightly higher emissions for each year. The
use of Tier 1 emission factors to estimate emissions also gives rise uncertainty as it is the "least accurate
estimation method" (IPCC, 2006). The Tier 1 method gives an aggregate emission estimate for all of the F-
GHG using processes simultaneously, which introduces a higher level of uncertainty as the utilization
rates of gases differ between etch and chamber cleaning processes.
Lastly, due to the similarities between this sector and the semiconductor manufacturing sector, the
mitigation technologies considered for PV were those used in the semiconductor sector.
IV-188 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
PHOTOVOLTAIC CELL MANUFACTURING
Applied Materials. 1999. "Catalytic Abatement of PFC Emissions." Presented at Semicon Southwest 99:
Partnership for PFC Emissions Reductions, Austin, TX, October 18,1999.
Beu, L. 2005. "Reduction of Perfluorocarbon (PFC) Emissions: 2005 State-of-the-Technology Report."
TT#0510469A-ENG. International SEMATECH Manufacturing Initiative (ISMI), Albany, New York.
December 2005. Obtained at: http://www.epa.gov/highgwp/semiconductor-
pfc/documents/final tt report.pdf.
Brown et al., 2012. "Catalytic technology for PFC emissions control." Solid State Technology. Obtained at:
http://www.electroiq.com/content/eiq-2/en/articles/sst/print/volume-44/issue-7/features/emissions-
control/catalytic-technology-for-pfc-emissions-control.html. Accessed August 2012.
Burton, S. 2003. Personal communication with Brown of Motorola (2002) supplemented by personal
communication with Von Compel of BOC Edwards (2003), research of DuPont's Zyron Web site
(2003), and personal communication with Air Liquide regarding thermal destruction, NF3 remote
clean, and capture membrane unit costs.
CRS, 2012. "U.S. Solar Photovoltaic Manufacturing: Industry Trends, Global Competition, Federal
Support." Congressional Research Service. 7-5700. R42509. June 2012. Available at:
http://www.f as.org/sgp/crs/misc/R42509.pdf.
DisplaySearch. 2009. DisplaySearch Q4'09 Quarterly PV Cell Capacity Database & Trends Report.
DisplaySearch, LLC, Santa Clara, California, 2009.
EPIA, 2011. "EPIA Working Group Fact Sheet: Sustainability of Photovoltaic Systems, The Carbon
Footprint." European Photovoltaic Industry Association. March 2011. Available at:
http://www.epia.org/publications/factsheets.html.
Forbes, 2012. "Report: Solar Panel Supply Will Far Exceed Demand Beyond 2012". Forbes. Obtained at:
http://www.forbes.com/sites/uciliawang/2012/06/27/report-solar-panel-production-will-far-exceed-
demand-beyond-2012/.
Fthenakis, V. December 2001. Options for Abating Greenhouse Gases from Exhaust Streams. Brookhaven
National Laboratory, Upton, New York, 2001. Obtained at:
http://www.bnl.gov/isd/documents/23784.pdf.
Fthenakis, V, Clark. D. O., Moalem, M., and Chandler, P., "Life-Cycle Nitrogen Trifluoride Emissions
from Photovoltaics." Environ. Sci. Technol. 2010, vol 44, pp. 8750-8757.
Hattori et al., 2006. "Application of Atmospheric Plasma Abatement System for Exhausted Gas from
MEMS Etching Process." Presetned at the Institute of Electrical and Electronics Engineers
International Symposium on Semiconductor Manufacturing 2006, Tokyo, Japan, September 25-27,
2006.
International Energy Agency (IEA). 2010. Technology Roadmap: Solar photovoltaic energy - Foldout.
Available online at:
http://www.iea.org/publi cations/freepublications/publication/name,28528,en.html.
International Energy Agency (IEA). 2012. World Energy Outlook 2012 Available for purchase at:
http://www.worldenergyoutlook.org/publications/weo-2012/.
IPCC. 2006. 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse
Gas Inventories Programme (Volume 3, Chapter 6: Electronics Industry Emissions), The
Intergovernmental Panel on Climate Change, H.S. Eggleston, L. Buendia, K. Miwa, T. Ngara, and K.
Tanabe (eds.). Hayama, Kanagawa, Japan.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-189
-------
PHOTOVOLTAIC CELL MANUFACTURING
U.S. Environmental Protection Agency (USEPA). (2008a). Developing a Reliable Fluorinated Greenhouse
Gas (F-GHG) Destruction or Removal Efficiency (DRE) Measurement Method for Electronics
Manufacturing: A Cooperative Evaluation with NEC Electronics, Inc. (EPA 430-R-10-005). Office of
Air and Radiation Office of Atmospheric Programs, Climate Change Division, U.S. Environmental
Protection Agency, Washington, DC. Available at: http://www.epa.gov/highgwp/semiconductor-
pfc/documents/nec report.pdf.
U.S. Environmental Protection Agency (USEPA). (2008b). Developing a Reliable Fluorinated Greenhouse
Gas (F-GHG) Destruction or Removal Efficiency (DRE) Measurement Method for Electronics
Manufacturing: A Cooperative Evaluation with Qimonda (EPA 430-R-08-017). Office of Air and
Radiation Office of Atmospheric Programs, Climate Change Division, U.S. Environmental Protection
Agency, Washington, DC. Available at: http://www.epa.gov/highgwp/semiconductor-
pfc/documents/qimonda report.pdf.
U.S. Environmental Protection Agency (USEPA). 2009. Developing a Reliable Fluorinated Greenhouse Gas (F-
GHG) Destruction or Removal Efficiency (DRE) Measurement Method for Electronics Manufacturing: A
Cooperative Evaluation with IBM. (EPA 430-R-10-004). Washington, DC: Office of Air and Radiation
Office of Atmospheric Programs, Climate Change Division, U.S. Environmental Protection Agency.
Obtained at: http://www.epa.gov/highgwp/semiconductor-pfc/documents/ibm report.pdf.
U.S. Environmental Protection Agency (USEPA). 2012. Global Anthropogenic Non-COi Greenhouse Gas
Emissions: 1990-2030. Washington, DQUSEPA, OAR, Climate Change Division. EPA 430-R-12-006.
Obtained at:
http://www.epa.gov/climatechange/Downloads/EPAactivities/EPA Global NonCO2 Projections De
c2012.pdf.
IV-190 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
FLAT PANEL DISPLAY
IV.13.1 Sector Summary
lat panel display (FPD) manufacturing processes generate fluorinated greenhouse gas (F-GHG)
emissions including sulfur hexafluoride (SF6), nitrogen trifluoride (NF3), and carbon
tetrafluoride (CF4), which are used in etching and chamber-cleaning processes commonly used
in electronics manufacturing.
Global annual F-GHG emissions from FPD manufacturing are estimated to be approximately seven
and a half million metric tons of carbon dioxide equivalents (MtCC^e) in 2010, projected to grow to 12
MtCC^e by 2030.l Figure 13-1 shows the F-GHG emission projections by country from 2000 to 2030. The
growth in emissions is primarily driven by projected growth in manufacturing capacity in China. The
baseline projections presented here are updated relative to those presented in Global Anthropogenic Non-
CO2 Greenhouse Gas Emissions: 1990 to 2030 (USEPA, 2012). The updated projections include lower
expected production levels of Liquid Crystal Displays and increased mitigation, resulting in significantly
lower projected emissions. Emissions projections for this sector are particularly uncertain due to limited
information on emissions rates and reduction efficiencies, variable industry production growth rates, and
rapidly evolving FPD technologies.
Figure 13-1: F-GHG Emissions from FPD Manufacturing: 2000-2030 (MtC02e)
14 -
12 -
10 -
8
6
4 -
2 -
0
I Singapore
Japan
I South Korea
I China
2000
2010 2020
Year
2030
Source: U.S. Environmental Protection Agency (USEPA), 2012
1 The term flat panel display encompasses many technologies, such as liquid crystal displays, low temperature
polysilicon (LTPS) or transparent amorphous oxide semiconductor (TAOS), and active matrix light emitting diodes
(AMOLED). This analysis focuses on liquid crystal displays and does not include newer technologies such as
AMOLEDs.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-191
-------
FLAT PANEL DISPLAY
Six mitigation technology options were examined for this sector: central abatement, thermal
abatement, catalytic abatement, plasma abatement, NF3 remote chamber dean, and gas replacement.
Global abatement of F-GHG emissions in FPD manufacturing is estimated to be 9.3 MtCO2e in 2020
and 9.6 MtCO2e in 2030. The emissions reductions equate to 80% of the projected baseline emissions in
each year. While all abatement options identified for this sector reduce emissions significantly, none has
the ability to reduce or replace 100% of emissions from FPD manufacturing. Figure 13-2 presents the
global MAC curves for 2010, 2020, and 2030 for the FPD manufacturing sector.
Figure 13-2: Global Abatement Potential in FPD Manufacturing: 2010,2020, and 2030
$60
$50
$40
8 $30
$20
$10
$0
•2010
2020
•2030
234
Non-CO2 Reduction (MtCO2e)
The following sections of this chapter first describe the activities and sources of F-GHG emissions in
the FPD manufacturing sector and present the projected emissions for 2010 to 2030. Subsequent sections
characterize the six abatement measures considered and present the engineering costs assumptions used
in the MAC analysis. This is followed by a discussion of the MAC modeling assumptions used to
estimate the global abatement potential. We conclude the chapter by presenting the MAC curves for each
emitting country and briefly discuss some of the major uncertainties and limitations to our analysis.
IV.13.2 Emissions from Flat Panel Display Manufacturing
In FPD manufacturing, high global warming potential (GWP) greenhouse gases are used for chemical
vapor deposition (CVD) cleaning processes and plasma dry etching during the manufacture of arrays of
thin-film transistors on glass substrates, which switch pixels of liquid crystal displays and organic light-
emitting diode displays.
IV-192
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
FLAT PANEL DISPLAY
To evaluate the cost of reducing F-GHG emissions from FPD manufacturing, this analysis considers
the emissions and reduction costs for a typical FPD manufacturing facility (manufacturing processes and
uses of GHGs across the industry are generally similar), characterized based on an average gas
consumption, an assumed 25 tools per facility with approximately 3.5 chambers each, and the average
emissions of F-GHGs for etch and dean processes. Figure 13-3 shows the breakdown of etch and clean
emissions for the typical facility.
Figure 13-3: Global F-GHG Emissions in 2020 by Facility Type (% of GWP-Weighted Emissions)
IV.13.2.1 Activity Data or Important Sectoral or Regional Trends and Related
Assumptions
FPD facility-level emissions were modeled using estimated gas consumption data based on World
LCD Industry Cooperation Committee (WLICC) reported data and Intergovernmental Panel on Climate
Change (IPCC) Tier 2b emission factors. Because the WLICC does not fully represent all world
manufacturing capacity, global emissions were estimated using estimated global capacity, in terms of
area of substrate produced, from the DisplaySearch Equipment Database and IPCC Tier 1 emission
factors for FPD manufacturing as opposed to using WLICC gas consumption information.
Global emission estimates take into account the WLICC goal, which was agreed to by all three
member associations. The goal is to meet and maintain an aggregate 2010 F-GHG emission target of 10%
of projected business-as-usual 2010 emissions, or 3.01 MtCO2e (expressed in the goal as 0.82 million tons
of carbon equivalent). (This emission reduction goal was met in 2010; hence, mitigation technologies and
strategies were assumed to have penetrated the market to a certain extent already in WLICC Partner
countries Japan, Taiwan, and Korea.) The WLICC member associations are estimated to have 96% of the
world's FPD manufacturing capacity in 2010. It is assumed, therefore, that there has already been some
level of mitigation technology penetration to meet the stated goal in this sector in the baseline projections.
Current market trends indicate major growth for capacity in this sector shifting to China, and without a
reduction goal or mitigation measures, FPD emissions in China will drastically increase.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-193
-------
FLAT PANEL DISPLAY
IV.13.2.2 Emissions Estimates and Related Assumptions
The large majority of FPD manufacturing is in Asia, and production capacity continues to grow in
that region. As shown in Table 13-1, from 2010 to 2030, emissions from global FPD manufacturing are
expected to grow by a factor of 54 times, from 3 MtCO2e in 2010 to 162 MtCO2e in 2030. Much of this
growth occurs between 2015 and 2030. This increase is expected to be attributed to growth in production
and emissions in China for two reasons: (1) an expected increase in China's domestic demand for FPDs,
and much of this demand will be met through domestic production (DisplaySearch, 2010); and (2)
China's share of world emissions is projected to steeply increase partly because emissions for the WLICC
member countries are assumed to meet and maintain the reduction goal of 3.01 MtCO2e.
Table 13-1: Projected Baseline Emissions from FPD Manufacturing: 2010-2030 (MtC02e)
Country
China
South Korea
Japan
Singapore
World Total
2010
1.9
5.1
0.4
0.0
7.4
2015
3.8
5.4
0.4
0.0
9.5
2020
5.7
5.5
0.4
0.0
11.7
2025
5.8
5.6
0.5
0.1
11.9
2030
5.9
5.6
0.5
0.1
12.0
CAGR*
(2010-2030)
5.8%
0.5%
0.2%
4.4%
2.4%
aCAGR= Compound Annual Growth Rate
Source: USEPA, 2012
IV.13.3 Abatement Measures and Engineering Cost Analysis
Six mitigation technology options were identified for the FPD manufacturing sector: central
abatement, thermal abatement, catalytic abatement, plasma abatement, NF3 remote chamber clean, and
gas replacement.
• Central abatement: These are large-scale abatement systems that are generally located on the
roof of facilities and are applicable to etch emissions (SF6).
• Thermal abatement: These point-of-use (POU) abatement systems, that use heat to destroy or
remove F-GHGs from effluent process streams, are connected directly to a manufacturing tool.
• Catalytic abatement: Tool effluent process streams are run through POU abatement systems with
catalysts (e.g., CuO, ZnO, A12O3) that destroy or remove F-GHGs.
• Plasma abatement: Plasma in a POU abatement system is used to react (destroy or remove)
F-GHGs from the process effluent stream.
• NF3 remote chamber clean: Highly ionized NF3 is used to clean chemical vapor deposition
chambers. This process is highly efficient (using -98% of the gas in a process) resulting in lower
emissions on a mass and CO2 basis than traditional in-situ chamber clean processes which uses
approximately 20% to 50% of the gas in a process (IPCC, 2006).
• Gas replacement: Higher GWP gases are replaced with lower GWP gases.
Table 13-2 presents the reduction efficiency and the applicability of each mitigation technology to
emissions from a type of process (etch and/or dean). The next sections describe each of these mitigation
options in more detail.
IV-194
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
FLAT PANEL DISPLAY
Table 13-2: FPD Manufacturing Abatement Options
Abatement Option
Central abatement
Thermal abatement
Catalytic abatement
Plasma abatement
NFs remote chamber clean
Gas replacement
Applicable Process
Emissions Type(s)
Etch
Etch and clean
Etch and clean
Etch
Clean
Clean
Reduction
Efficiency
77%
95%
99%
97%
95%
77%
Information Source
COM project #3333
Fthenakis (2001), Beu (2005), and USEPA (2009)
Fthenakis (2001), Brown etal. (2012)
Fthenakis (2001), Hattori etal. (2006)
Beu (2005)
COM methods NM0289, NM303, NM0317, NM0335
IV.13.3.1
Central Abatement
Central abatement systems (CAS) have begun to be designed and used to handle the generally high
gas (SF6) flows used in FPD manufacturing because of the large size of the substrate being etched. A CAS
is a large-scale thermal oxidation abatement system that is located on the roof of FPD facilities, so there
are little expected space limitations for this technology. This technology has recently started to come
online and is only applicable to etch emissions. Two CDM projects (one from LG and one from Samsung)
in Korea have used this technology (CDM Project #3440 and CDM project #3333). Its use is, however,
limited throughout the rest of the industry because it is expensive and relatively new.
The capital cost for a CAS is $4.5 million. The annual operation and maintenance (O&M) cost, which
include items such as utilities and parts, is estimated to be $2.5 million for a facility. No revenues are
generated from using a CAS.
IV.13.3.2
Thermal Abatement
Thermal abatement systems can be used to abate emissions from both etching and CVD chamber-
cleaning processes. The use of thermal abatement offers the benefit of not affecting the manufacturing
process (Applied Materials, 1999); however, the systems do require space that may not be available in
some facilities. In addition, these systems require large amounts of cooling water, and the use of the
systems results in regulated NOX emissions.
The total facility capital cost for installing thermal abatement systems is estimated to be $5.7 million.
This estimate includes costs for the systems, the necessary ducting, water recirculation and hook up, and
natural gas costs (Fthenakis, 2001; Burton, 2003). The annual operating cost is estimated to be $328,860 at
the facility level. No annual savings are associated with using this technology.
IV.13.3.3 Catalytic Abatement
A catalytic abatement system is used to abate emissions from both etching and CVD chamber-
cleaning processes. This type of abatement is applicable at most facilities, but again there may be some
space constraints, as also discussed for thermal abatement systems, limiting the use of these systems in
the market. Another limitation to their use is high catalyst replacement costs.
The capital cost associated with purchasing and installing the abatement systems is estimated to be
$6.9 million per facility. To use catalytic abatement systems, facilities must factor in the annual cost of
resources such as water, waste chemicals, electricity, and catalyst replacements. To cover these operating
expenses, a facility manufacturing FPDs is estimated to incur an annual cost of $455,280. As with other
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-195
-------
FLAT PANEL DISPLAY
abatement technologies considered in this sector, the use of thermal abatement systems will not result in
annual savings.
IV.13.3.4 Plasma Abatement
Plasma abatement systems are assumed to be applicable to etch processes in most facilities, with
some physical space limitations. (These systems, however, are relatively smaller in size compared with
thermal and catalytic systems.) Plasma abatement systems use a small plasma source that effectively
dissociates the F-GHG molecules that react with fragments of the additive gas (hydrogen (H2), oxygen
(02), water (H2O), or methane (CH4)) to produce low molecular weight by-products such as hydrogen
fluoride (HF) with little or no global warming potential (GWP). After disassoriation, wet scrubbers can
remove the molecules. The presence of additive gas is necessary to prevent later downstream reformation
of F-GHG molecules (Motorola, 1998).
The capital cost for plasma abatement systems is estimated to be $1.8 million per facility (Fthenakis,
2001; Burton, 2003), which covers the purchase and installation of plasma systems. Plasma abatement
systems require an annual operation cost of $1,190 per chamber, which includes general maintenance and
use of the systems. The total annual facility cost is $51,850 based on an assumed number of tools per
facility and chambers per tool. The use of plasma abatement systems will not result in annual cost
savings.
IV.13.3.5 NF3 Remote Chamber Clean
NF3 remote chamber clean is an alternative cleaning technology that has the benefit of having a
particularly high utilization rate of NF3 (-98%; IPCC, 2006), resulting in relatively low emissions
compared with traditional chamber deans. (Note: The stated utilization is based on utilizations for
semiconductor manufacturing; as a result of large gas flows in FPD manufacturing the actual utilization
may be lower.) NF3 remote clean systems dissociate NF3 using argon gas, converting the source gas to
active F-atoms in the plasma, upstream of the process chamber. These electrically neutral atoms can
selectively remove material in the chamber. The by-products of remote dean include HF, fluorine (F2),
and other gases, most of which are removed by fadlity add scrubber systems. The use of NF3 remote
clean systems is much more prevalent in new facilities because the technology was not available when
many old fadlities were constructed.
It is assumed that FPD fadlities are not "NF3 ready"; in other words, these fadlities do not have the
current infrastrudure to handle the dired installation of NF3 remote systems because this is a relatively
new technology. Therefore, facilities incur capital costs, in addition to system costs, associated with items
such as gas hook ups and necessary hardware such as manifolds and values. The facility cost is estimated
to be $9.2 million. The annual fadlity cost for NF3 remote dean is estimated to be $3.3 million (Burton,
2003). This cost is assodated with the purchase of larger volumes of gas (NF3 versus traditional chamber
dean gases), general maintenance, and F2 scrubs to remove the highly explosive gas from the effluent. No
annual cost savings are assodated with using this technology.
IV.13.3.6 Gas Replacement
Gas replacement can be used to mitigate emissions from the traditional CVD chamber-deaning
process. Gas replacement can be applied in most fadlities and has already been used throughout the
industry in many instances. For this strategy, a lower GWP gas replaces a higher GWP gas. The most
common replacement is using NF3 to replace SF6.
Facilities repladng SF6 with NF3 incur an estimated capital cost of $1.2 million for items such as gas
hook ups and implementation. Annual savings for this option result from the lower cost of the
IV-196 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
FLAT PANEL DISPLAY
replacement gas and were estimated to be $34,400, based on the incremental cost of the gases and the
average amount of gas consumed per facility. Gas replacement has no operational costs.
IV.13.3.7 Summary of Mitigation Technology Costs and Characteristics
Table 13-3 summarizes the information used to estimate the break-even prices discussed in the
following section.
Table 13-3: Engineering Cost Data on a Facility Basis
Abatement
Option
Central abatement
Thermal abatement
Catalytic abatement
Plasma abatement
NFs remote clean
Gas replacement
Project
Lifetime
(Years)
15
7
7
7
21
21
Capital Costs
(2010 USD)
$4,487,400
$5,701,971
$6,906,594
$1,814,664
$9,200,867
$1,180,000
Annual
Revenue
(2010 USD)
$0
$0
$0
$0
$0
$34,400
Annual O&M Abatement
Costs Amount
(2010 USD) (tCOe)
$2,547,000
$328,862
$455,277
$51,847
$3,374,861
$0
13,889
62,587
65,222
14,941
56,496
22,797
IV.13.4 Marginal Abatement Costs Analysis
This section discusses the modeling approach and documents some additional assumptions used in
the MAC analysis for the FPD manufacturing sector.
IV. 13.4.1
Methodological Approach
The MAC analysis applies the abatement measure costs discussed in the previous section at the
model FPD manufacturing facility to calculate a break-even price for each abatement measure. This
section presents how we defined the model facility used in this analysis and the technical effectiveness
assumptions used to estimate the incremental reductions for each measure. This section also provides an
example of how the break-even prices were calculated for each option.
IV.13.4.2 Facility Definition
The typical facility considered in this analysis represents an average-sized FPD facility, with an
estimated 25 tools and 3.5 chambers per tool. Based on WLICC-reported emissions data, the facility is
assumed have an emissions breakdown of 23% etch emissions and 77% chamber clean emissions, which
assumes a certain level of abatement is used. The facility emission breakdown is essential to this analysis,
because some mitigation technologies are applicable to either both or just one type of manufacturing
process (see Table 13-2 above).
IV.13.4.3 Estimating the Technical Effectiveness Parameter
Similar to the methods employed for analyzing abatement in the semiconductor and photovoltaics
manufacturing sectors, this analysis developed a technical effectiveness parameter, defined as the
percentage reductions achievable by each technology/process combination. Estimating this parameter
requires assumptions regarding the distribution of emissions by manufacturing process (etch and clean),
in addition to process-specific assumptions on technical applicability and market penetration. These
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-197
-------
FLAT PANEL DISPLAY
assumptions are held constant for all model years in the MAC analysis. Table 13-4 presents the technical
applicability, market penetration, and reduction efficiency assumptions used to develop the abatement
measures' technical effectiveness parameters.
Table 13-4: Technical Effectiveness Summary
Etch (23%) Clean (77%)
Technical Market Technical Market Reduction Technical
Abatement Measure Applicability Penetration Applicability Penetration Efficiency Effectiveness
Central abatement
Thermal abatement
Catalytic abatement
Plasma abatement
NFs remote clean
Gas replacement
100%
85%
85%
85%
0%
0%
40%
30%
10%
20%
n/a
n/a
0%
85%
85%
0%
100%
50%
n/a
55%
15%
n/a
20%
10%
77%
95%
99%
97%
95%
77%
7%
40%
12%
4%
15%
3%
The technical effectiveness is a weighted average of the abatement measure's emissions reductions
when applied to each applicable process(es). The share of total emissions attributed to each process (i.e.,
23% etching and 77% cleaning) is the weight which is multiplied by the product of the technical
applicability, market penetration, and reduction efficiency for each abatement measure. The technical
effectiveness estimates are then multiplied by the facility annual emissions to estimate the abatement
potential achievable through each of the six abatement measures. Summing the technical effectiveness
across the six abatement measures yields the maximum level of emissions reductions that is technically
achievable.
IV.13.4.4 Estimating Break-Even Prices
The MAC model uses the estimated abatement project costs and benefits as described in Section 6.3 to
calculate the break-even price for each abatement measure. Table 13-5 illustrates the break-even
calculation for each abatement measure expressed in 2010 USD.
Table 13-5: Example Break-Even Prices for Abatement Measures in FPD Manufacturing
Abatement Option
Central Abatement System
Thermal abatement
Catalytic abatement
Plasma abatement
NFS remote clean
Gas replacement
Reduced
Emissions
(tC02e)
13,889
62,587
65,222
14,941
56,496
22,797
Annualized
Capital Costs
($/tC02e)
$71
$31
$36
$42
$31
$10
Net Annual
Cost
($/tC02e)
$183
$5
$7
$7
$60
$0
Tax Benefit of
Depreciation
($/tC02e)
$14
$9
$10
$12
$5
$2
Break-Even
Price
($/tC02e)
$240
$28
$33
$37
$86
$8
As Table 13-5 shows, having no annual benefits and high initial capital costs results in relatively
higher break-even prices. Break-even prices range between $8 and $240/tCC>2e based on the initial cost
assumptions (unadjusted for country-specific prices). Gas replacement is the cheapest of the six options
with a break-even price of $8/tCO2e. The CAS is the most expensive at $240/tCO2e. These costs are
IV-198
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
FLAT PANEL DISPLAY
relatively high compared to some other sectors examined in this report, however they can be better
understood in perspective of the high costs associated with this manufacturing industry. For example in
2004, Samsung and Sony invested more than $3 billion in capital costs for two new FPD production lines
(Ramstad, 2011).
IV.13.4.5 MAC Analysis Results
The global abatement potential for F-GHG emissions reductions in the FPD manufacturing sector is
estimated to be 9.6 MtCO2e, or 80% of total projected emissions in 2030. Table 13-6 presents the
cumulative reductions achieved at selected break-even prices for China, Japan, Singapore, and South
Korea. These are the only four countries with projected emissions in 2030 in the FPD manufacturing
sector. Figure 13-4 presents the resulting MAC curves for each country in 2030.
Table 13-6: Abatement Potential by Country/Region at Selected Break-Even Prices in 2030 (MtC02e)
Break-Even Price ($/tC02e)
Country
China
Japan
Singapore
South Korea
World Total
-10 -50 5 10
— — — — —
— — — — —
— — — — —
— — — — —
0.0 0.0 0.0 0.0 0.0
- - 2.4
— — —
— — —
— — —
0.0 0.0 2.4
2.5
—
—
2.2
4.7
2.8
0.2
0.0
2.4
5.3
4.7
0.4
0.0
4.5
9.6
Figure 13-4: Marginal Abatement Cost Curves by Emitting County in 2030
•China
Singapore
•Japan
•South Korea
1.0
2.0 3.0
Non-CO2 Reduction (MtCO2e)
4.0
5.0
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-199
-------
FLAT PANEL DISPLAY
As the results show, the abatement potential in the FPD manufacturing sector can be significant (80%
of sectoral emissions). Over 2.4 MtCO2e of F-GHG emissions could be reduced in China at a carbon price
of ~$25/tCO2e; these reductions alone represent 20% of global emissions in 2030.
IV.13.5 Uncertainties and Limitations
Because of the similarities between this sector and the semiconductor manufacturing sector, most
mitigation technologies (and hence cost estimates) are assumed to be the same as for the semiconductor
manufacturing sector (with the exception of the CAS technology). We made this assumption because of
the limited amount of public information on the extent to which various types of mitigation technologies
are being used in this sector.
IV-200 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
FLAT PANEL DISPLAY
References
Applied Materials. 1999. Catalytic Abatement of PFC Emissions. Presented at Semicon Southwest 99:
Partnership for PFC Emissions Reductions, October 18,1999, Austin, TX.
Bartos, S. 2010. Draft: A Review of WLICC's Progress Toward Reducing Potent GHG Emissions from the
Manufacture of Liquid Crystal Displays. Washington DC, U.S. Environmental Protection Agency.
Beu, L. December 2005. Reduction of Perfluorocarbon (PFC) Emissions: 2005 State-of-the-Technology Report,
TT#0510469A-ENG, International SEMATECH Manufacturing Initiative (ISMI).
Brown et. al., 2012. "Catalytic technology for PFC emissions control." Solid State Technology. Obtained
at: http://www.electroiq.com/content/eiq-2/en/articles/sst/print/volume-44/issue-7/features/emissions-
control/catalytic-technology-for-pfc-emissions-control.html. Accessed August 2012.
Burton, S. 2003. Personal communication with Brown of Motorola (2002) supplemented by personal
communication with Von Compel of BOC Edwards (2003), research of DuPont's Zyron Web site
(2003), and personal communication with Air Liquide regarding thermal destruction, NF3 Remote
Clean, and Capture Membrane unit costs.
COM Project #3440 and CDM project #3333. Project data available at:
http://cdm.unfccc.int/Registry/index.html
DisplaySearch. 2010. DisplaySearch Q4'09 Quarterly FPD Supply/Demand and Capital Spending Report.
DisplaySearch, LLC, Santa Clara, California, 2010.
Fthenakis, V. December 2001. Options for Abating Greenhouse Gases from Exhaust Streams. Brookhaven
National Laboratory, Upton, New York, 2001. Obtained at:
http://www.bnl.gov/isd/documents/23784.pdf.
Hattori et al., 2006. "Application of Atmospheric Plasma Abatement System for Exhausted Gas from
MEMS Etching Process." Presetned at the Institute of Electrical and Electronics Engineers
International Symposium on Semiconductor Manufacturing 2006, Tokyo, Japan, September 25-27,
2006.
Intergovernmental Panel on Climate Change (IPCC). 2006. 2006 IPCC Guidelines for National Greenhouse
Gas Inventories. The National Greenhouse Gas Inventories Programme (Volume 3, Chapter 6: Electronics
Industry Emissions), The Intergovernmental Panel on Climate Change, H.S. Eggleston, L. Buendia, K.
Miwa, T. Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
Motorola. October 18,1998. "Long-term Evaluation of Litmus "Blue" Inductively-Coupled Plasma Device
for Point-of-Use PFC and HFC Abatement." Presented at Semicon Southwest 99: A Partnership for
PFC Emissions Reductions, Austin, TX.
Prather, M., and J. Hsu. June 2008. "NF3, the Greenhouse Gas Missing from Kyoto." Geophysical Research
Letters 35.
Ramstad, E. December, 2011. "Samsung Electronics Ends LCD Venture With Sony." Wall Street Journal
US. Edition. Published on December 27, 2011.
Rothlisberger, L. May 5, 2011. Documentation of meeting between Lukas Rothlisberger of DILO and Paul
Stewart of ICF International.
U.S. Environmental Protection Agency (USEPA). 2009. Developing a Reliable Fluorinated Greenhouse Gas (F-
GHG) Destruction or Removal Efficiency (DRE) Measurement Method for Electronics Manufacturing: A
Cooperative Evaluation with IBM (EPA 430-R-10-004), Office of Air and Radiation Office of
Atmospheric Programs, Climate Change Division, U.S. Environmental Protection Agency,
Washington, DC. Obtained at: http://www.epa.gov/highgwp/semiconductor-
pfc/documents/ibm report.pdf.
U.S. Environmental Protection Agency (USEPA). 2012. Global Anthropogenic Non-COi Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington, DC: USEPA.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-201
-------
V. Aariculture Sector
-------
CROPLANDS
V.1. Non-Rice Croolands
V.1.1 Sector Summary
and management in croplands influences soil N2O emissions, Q-k fluxes, and soil organic
I carbon (C) stocks (and associated CCh fluxes to the atmosphere). Soil N2O emissions are
influenced by human activity, including synthetic nitrogen fertilization practices, application
of organic fertilizers such as manure, drainage of organic soils, cultivation of N-fixing crops, and
enhancement of N mineralization in soils through practices such as cultivation/management of native
grasslands and forests (Mosier et al., 1998; Smith et al., 2007). Globally, N2O emissions from agricultural
soils increased by about 19% between 1990 and 2010 While N2O emissions from all sources grew only 4%.
In 2010, soil N2O emissions account for approximately 56% of the global N2O emissions, up from 51% in
1990.l In contrast to soil N2O, where there are sizable annual fluxes that depend on human activity, soil
organic C stocks are assumed to be approximately in equilibrium.2
The marginal abatement cost curves presented in this chapter consider mitigation strategies that
apply to only a fraction of the total emissions from agriculture. Specifically, the following categories are
included:
• Direct and indirect emissions from mineral-based cropland soils processes
Synthetic and organic fertilization
- Residue N
- Mineralization and asymbiotic fixation, based on temperature and moisture, etc.
• Major crops supplemented by selected similar minor crops
Barley (plus rye)
Maize (plus green corn)
Sorghum
Soybeans (plus lentils, other beans)
Wheat (plus oats)
In addition, compared to the estimates typically developed for GHG inventories, the emissions
presented in this chapter will be lower because the following types of emissions are excluded due to data
and resource limitations:
• Drainage of organic soils.
• Grassland soils
• Other crops not mentioned above (e.g. vegetables)
• Restoration of degraded lands
• Burning of residues or biofuel
The focus is on emissions from major crops, which is consistent with our evaluation of mitigation
options that can be applied to mitigate emissions from these major crops in this chapter.
1 Global total N2O emissions were 3240.7 MtCCtee in 1990 and 3,519.6 MtCCtee in 2010. Agricultural soils total N2O
emissions were 1,658.1 MtCCtee in 1990 and 1,969.0 MtCO2e in 2010 (USEPA, 2012).
2 Major changes in soil C occurred when land was first cultivated, but changes associated with agricultural soil
management are approximately balanced at a global scale based on current management and land use change trends
(Smith etal, 2007).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-1
-------
CROPLANDS
For the period 2010—2030 a business-as-usual forecast was constructed using projected growth rates
in acreage, output, prices, yields, population, and GDP by the International Food Policy Research
Institute (IFPRI)'s International Model for Policy Analysis of Agricultural Commodities and Trade
(IMPACT) (Nelson et al., 2010). The IFPRI IMPACT model projections provide a set of prices consistent
with population and productivity assumptions for the MAC analysis.3'4
Figure 1-1 presents projected baseline N2O and CH4 emissions and changes in soil organic carbon
from non-rice cropland soils; As shown in Figure 20-1, N2O emissions from global non-rice cropland soils
are projected to be 506, 500 and 504 million metric tons of CCh equivalent (MtCChe) in 2010, 2020 and
2030, respectively.5 Non-rice cropland soils are a net sink for methane, sequestering approximately 38
MtCCfee of CH4 per year. The estimated net changes in soil organic carbon suggest that the carbon stock
changes are roughly balanced at the global scale.
Figure 1-1: Global Baseline Emissions from Non-Rice Croplands by GHG: 2010-2030
ICH4
IN2O
I Soil C
Year
3 The IMPACT outputs separated the world into 116 regions, with larger countries defined individually and smaller
countries combined into regions. A mapping was created between IMPACT regions and the 195 countries in this
analysis, using shares of country-level Non-Rice Croplands population in 2010 based on USEPA (2012) to
disaggregate regional projections from the IMPACT model to individual countries within each region.
4 The business as usual forecast excludes such potential drivers as deforestation, biofuels expansion and changes in
consumer preferences for meat.
5 The relative constant GHG emissions projected in the baseline are mainly driven by the DAYCENT modeling that
assumes the same management practices are applied throughout the study period as well as relatively small changes
in demand in the IMPACT model projections.
V-2
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CROPLANDS
Figure 1-2 presents the projected net GHG emissions (N2O and Q-k) from the top-five emitting
countries. The top 5 countries of China, India, the United States, Brazil and Argentina represent about
63% of global net emissions from cropland in 2010.
Figure 1-2: Baseline Net GHG Emissions from Non-Rice Croplands, Top Five Emitting Countries
474
460
472
I Argentina
I Brazil
India
lUnited_States
China
ROW
2010
2030
Note: ROW indicates Rest of the World
Table 1-1: Projected Net GHG Baseline Emissions from Non-Rice Croplands by Country: 2010-2030
(MtC02e)
Country
2010
2015
2020
2025
2030
(2010-2030)
Top 5 Emitting Countries
China
U.S.A
India
Brazil
Argentina
109
82
60
35
14
123
80
58
32
16
116
71
61
33
14
115
84
61
33
16
105
86
66
34
13
-0.2%
0.2%
0.5%
-0.2%
-0.2%
Rest of Regions
Asia
Africa
Europe
Middle East
Central & South America
Eurasia
North America
World Totals
31
31
62
4
13
18
15
474
26
26
56
9
14
14
15
470
27
30
59
7
15
15
14
460
27
28
63
9
15
15
16
482
27
29
60
10
15
13
14
472
-0.8%
-0.3%
-0.2%
4.2%
0.8%
-1.4%
-0.2%
0.0%
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-3
-------
CROPLANDS
Figure 1-3 presents the MAC curves for the global non-rice croplands, in 2010, 2020 and 2030. The
non-rice croplands MAC curves presented in this chapter are distinctive because they show less
abatement potential in 2030 than in 2010 - the 2030 curve is to the left or "inside" the 2020 and 2010
curves. This is due to the effect of soils becoming "saturated" with C and reaching a new equilibrium
within a few years of a management change. In other words, the 2020 mitigation estimate is the change
from the baseline emissions in 2020, for a management change started in 2010.
MAC analysis of the mitigation options described above suggests that at a relatively low carbon price
of $5 per ton of CCh equivalent ($/tCChe), net GHG abatement potential for global non-rice cropland soils
is approximately 65 MtCCfee, or about 13% of its baseline net emissions of 476 MtCChe in 2010. Mitigation
potential at $5/ tCChe reduces to 10% of the sector's baseline emissions in 2020 and 6% in 2030.
Figure 1 -3: Global MAC Curve for Net GHG Reductions from Non-Rice Cropland Soils
•2010
•2020
•2030
Non-CO2 Reduction (MtCO2e)
The following section offers a brief description of the model used. Section IV.20.3 presents selected
abatement technologies, their technical specifications, costs and potential benefits. Section IV.20.4
discusses the MAC analysis and estimated abatement potential and at global and regional levels. The
final section discusses uncertainties and limitations.
V.1.2
Emissions from Non-Rice Croplands
V.1.2.1
Methodology
The DAYCENT ecosystem model was used to estimate crop yields, N2O and Q-k emissions, and soil
C stocks in this analysis. DAYCENT is a process-based model (Parton et al., 1998; Del Grosso et al., 2001)
that simulates biogeochemical C and N fluxes between the atmosphere, vegetation, and soil by
V-4
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CROPLANDS
representing the influence of environmental conditions on these fluxes including soil characteristics and
weather patterns, crop and forage qualities, and management practices. DAYCENT utilizes the soil C
modeling framework developed in the Century model (Parton et al. 1987, 1988, 1994; Metherell et al.
1993), with refinement to simulate C dynamics at a daily time-step. Key processes simulated by
DAYCENT include crop production, organic matter formation and decomposition, soil water and
temperature regimes by layer, in addition to nitrification and denitrification processes. DAYCENT has
been evaluated in several studies (e.g., Del Grosso et al. 2002, 2005, 2009) and has also been recently
adopted by EPA to develop the soil C and soil estimates for the annual Inventory of U.S. Greenhouse Gas
Emissions and Sinks (EPA, 2013) submitted to the UNFCCC.
Crop yields, direct N2O and Q-k emissions, and soil organic C stock changes were simulated by
DAYCENT at a 0.5°grid resolution. Indirect N2O emissions6 were estimated simulated amounts of nitrate
leaching, N runoff in overland water flow, and NOx emissions from a site according to the DAYCENT
model7 combined with the IPCC default factors for indirect N2O emissions (De Klein et al., 2006). In order
to represent the longer term effect of cultivation on soil C, simulations started in 1700 after a simulation of
3000 years of native vegetation, which is a similar procedure to the methods applied in the US
Greenhouse Gas Inventory for agricultural soil C and N2O (USEPA, 2013).
For this study, a number of data sources were used to establish the business-as-usual baseline
conditions and simulate alternative management options for the global non-rice croplands. Weather data
were based on a dataset generated by the North American Carbon Program at a 0.5°resolution with daily
minimum and maximum temperatures and daily precipitation.8 The soils data were based on the FAO
Digitized Soil Map of the World (FAO 1996). Major cropland areas of the world were simulated according
to a global cropland map developed by Ramankutty et al. (2008), with grid cells with less than 5%
cropland area excluded in the analysis.
Native vegetation data are described in Cramer and Field (1999) and Melillo et al. (1993). Natural
vegetation was converted to cropland in the DAYCENT simulations at an approximate first year of
cultivation, based on historical records compiled by Ramankutty and Foley (1998) and Ramankutty et al.,
(2008).
Due to lack of global data availability, low input crop production with intensive tillage practices were
assumed prior to 1950, consistent with typical practices in that time period. From 1950 to 2010,
management was based on data including tillage and residue management, weeding practices, mineral N
fertilization, manure N amendments to soils, and irrigation. Crop planting and harvest dates were based
on Sacks et al. (2008). Crops were assumed to grow in monoculture due to insufficient data for
determining typical crop rotation practices from the global datasets. Maize and sorghum were double-
cropped in some regions based on Sacks et al. (2008). Model performance was evaluated by comparing
simulated crop yields to observed crop yields (Monfreda et al. 2008), and minor adjustments were made
to parameters in order to be reasonably consistent with the observed yields. More detail on the input data
and simulation framework is provided in Appendix O.
6 N2O emissions occurring with transport of N from one site to another where N2O emissions occur with N addition.
7 The same method as used in the US National Greenhouse Gas Inventory (USEPA, 2013).
8 The Multi-Scale Synthesis and Terrestrial Model Intercomparison Project (MsTMIP) developed consistent weather
data in order to "isolate, interpret, and address differences in process parameterizations among [terrestrial biospheric
models]" Source: http://nacp.ornl.gov/MsTMIP.shtml.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-5
-------
CROPLANDS
Global DAYCENT modeling was carried out for irrigated and non-irrigated production systems for
maize, wheat, barley, soybean and sorghum. Crop yields and GHG fluxes were simulated at the
0.5°resolution for periods 2000-2010 and 2011-2030 with five-year increments. A baseline scenario is
established for each crop production system assuming business-as-usual management practices described
above. Seven mitigation scenarios were then analyzed (see Section 3.4 below).
Emissions estimated by the DAYCENT model for major crop types (maize, wheat, barley, sorghum,
soybean and millet) were based on emissions per unit (m2) of physical area in each in each 0.5° x 0.5° grid
cell, and so were multiplied by an estimate of cropland area in each grid cell to compute total GHG
emissions. We approximated crop-specific areas using harvested area data. First, crop-specific harvested
areas for each 0.5° x 0.5° grid cell were estimated from Monfreda et al. (2008). For each grid cell where we
simulated double-cropping for maize or sorghum, we reduced maize or sorghum harvested area by 50%.
Next, harvested areas for analogous crops were added to areas of the major crop types (i.e., oats with
wheat, rye with barley, green corn with maize, and lentil, green bean, string bean, broad bean, cow pea,
chickpea and dry bean with soybeans) to increase the coverage of cropland area. The sums of harvested
areas fractions computed in this manner were less than total cropland areas (Ramankutty et al. 2008) for
all but 1.6% of grid cells. In the last step, total harvested area was scaled to match at the country scale
data on harvested areas reported in FAOSTAT. By including analogous crops and matching FAOSTAT
harvested areas, the cropland area simulated by DAYCENT was about 61% of the global non-rice
cropland areas reported by FAOSTAT.
Projected baseline emissions and crop production were then established for both irrigated and
rainfed production systems using simulated yields and GHG emissions rates from DAYCENT model and
adjusting with projected growth rates of these production systems by IFPRI's International Model for
Policy Analysis of Agricultural Commodities and Trade (IMPACT) model. In DAYCENT, crop
production areas were held constant at the 2010 level to obtain the biophysical effects of management
practice changes on crop yields and GHG fluxes. Projected acreage changes from IMPACT model reflect
socio-economic drivers such as population growth and technological changes to meet global food
demand (Nelson et al., 2010).
V.1.3 Abatement Measures and Engineering Cost Analysis
V.1.3.1 Mitigation Technologies
The mitigation options evaluated in this analysis were based on review of the literature to identify
the most promising options, while also taking data availability and potential for modeling within
DAYCENT into consideration. The mitigation options represent alternative management practices that
would alter crop yields and the associated GHG emissions, including adoption of no-till management,
split N fertilization applications, application of nitrification inhibitors, increased N fertilization (20%
increase over business-as-usual), decreased N fertilization (20% reduction from business-as-usual), and
100% crop residue incorporation.
The N management practices (split N fertilization, nitrification inhibitors, increased and decreased N
fertilization) influence N2O emissions in addition to soil organic C stocks due to reduced or enhanced C
inputs associated with the level of crop production. Smith et al. (2007) estimated that 89% of the overall
technical potential for mitigation of agricultural greenhouse gas emissions is associated with carbon
sequestration in soils. Although soil organic C stock fluxes are negligible in the baseline, there is
considerable opportunity to modify stocks in the future. Levels of soil organic matter and in particular
soil carbon both influence, and are influenced by cropland productivity. Other things being equal, higher
crop yields may increase soil C wherever more crop residue can be incorporated into the soil. Similarly,
V-6 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CROPLANDS
reducing crop residue removal would impact soil organic C stocks by changing the amount of C input to
the soil. Practices such as adoption of conservation tillage, restoration of degraded lands, improved water
and nutrient management, and cropping intensification can increase soil carbon by enhancing C inputs to
soils from greater crop production or decrease the losses of C from soils with lower decomposition rates
(Paustian et al. 1997; Six et al., 2000).
No-Till Adoption
All cultivation and field preparation events were removed except for seeding, which occurred
directly into the residue.
• Applicability: This option is available in all regions and all time periods
• Economic Applicability and Cost: There are reductions in labor costs associated with the
reduction in field preparation that are based on data from U.S. Department of Agriculture
(USDA) Agricultural Resource Management Survey (ARMS) data, which provides labor
estimates for conventional and conservation tillage on both irrigated and rain-fed land by major
crop. Conversion to no-till would require purchasing equipment for direct planting. However, if
this equipment is purchased in place of equipment used for traditional tillage, there may be little
incremental capital cost associated with no-till. Some crop budgets actually indicate lower capital
costs for no-till because of the need for fewer passes over the field, which lead to reduced
equipment depreciation. Thus, no incremental capital costs were assumed for no-till adoption.
• Additional Factors: In cases where yields change as a result, production is valued at the market
price. No tax or other benefits are included in this option.
Reduced Fertilization
This option reduced baseline fertilizer application levels by 20%.
• Applicability: This option is available in all regions and all time periods with nonzero baseline
fertilizer application levels.
• Economic Applicability and Cost: This option reduces operation costs by the value of fertilizer
withheld.
• Additional Factors: In cases where yields decrease as a result, the reduction in production is
valued at the market price. No tax or other benefits are included in this option.
Increased Fertilization
This option increased baseline fertilizer application levels by 20%.
• Applicability: This option is available in all regions and all time periods with nonzero baseline
fertilizer application levels.
• Economic Applicability and Cost: This option increases operation costs by the value of additional
fertilizer used.
• Additional Factors: In cases where yields increase as a result, production is valued at the market
price. No tax or other benefits are included in this option.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-7
-------
CROPLANDS
Split N Fertilization
Under this option, the baseline N application amount was applied in three separate and equal
amounts (planting day, 16 days after planting day, and 47 days after planting day) instead of once on
planting day.9
• Applicability: This option is available in all regions and all time periods with nonzero baseline
fertilizer application levels.
• Economic Applicability and Cost: This option was assumed to require 14% more labor to account
for additional passes over the fields to apply fertilizer multiple times rather than only once.
• Additional Factors: In cases where yields change as a result, production is valued at the market
price. No tax or other benefits are included in this option.
Nitrification Inhibitors
The baseline N application amount was applied once annually on date of planting. Nitrification
inhibitors were applied at time of fertilization, and reduced nitrification by 50% for 8 weeks10.
• Applicability: This option is available in all regions and all time periods with nonzero baseline
fertilizer application levels.
• Economic Applicability and Cost: The costs of this option include the cost of the nitrification
inhibitor, assumed to be $20 per hectare for the United States (Scharf et al., 2005) and scaled to
other regions.
• Additional Factors: In cases where yields change as a result, production is valued at the market
price. No tax or other benefits are included in this option.
100% Residue Incorporation
In this option, all crop residue was assumed to remain after harvest. This option serves to evaluate
how reducing removals would impact soil organic C stocks.
• Applicability: This option is available in all regions and all time periods
• Economic Applicability and Cost: No cost is associated with this option.
• Additional Factors: In cases where yields change as a result, production is valued at the market
price. No tax or other benefits are included in this option.
9 Following Del Grosso et al. (2009).
10 Following Del Grosso et al. (2009) and Branson et al. (1992).
V-8
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CROPLANDS
Table 1-2: DAYCENT Base Mean Yields, and Differences from Mean Yield for Mitigation Strategies, by
Year (Metric tons of Grain per Hectare)
2010
2015
2020
2025
2030
Maize
Base Yield
No-Till
Optimal N fertilization*
Split N Fertilization
100% Residue Inc.
Nitrification Inhibitors
Reduced Fertilization
Increased Fertilization
3.64
0
0
0
0.22
0
-0.05
0.04
3.64
-0.25
2.9
0.16
0.23
-0.01
-0.36
0.28
3.64
-0.17
3.05
0.17
0.24
-0.01
-0.39
0.29
3.59
-0.12
3.1
0.19
0.24
-0.01
-0.4
0.31
3.6
-0.07
3.08
0.18
0.24
-0.01
-0.4
0.31
Millet
Base Yield
No-Till
Optimal N fertilization*
Split N Fertilization
100% Residue Inc.
Nitrification Inhibitors
Reduced Fertilization
Increased Fertilization
1.16
0
0
0
0.09
0
-0.01
0.01
1.17
-0.09
2.38
0.09
0.08
0.02
-0.08
0.08
1.14
-0.07
2.59
0.09
0.09
0.03
-0.09
0.09
1.11
-0.05
2.55
0.09
0.09
0.03
-0.1
0.09
1.12
-0.03
2.61
0.08
0.08
0.03
-0.1
0.09
Sorghum
Base Yield
No-Till
Optimal N fertilization*
Split N Fertilization
100% Residue Inc.
Nitrification Inhibitors
Reduced Fertilization
Increased Fertilization
2.34
0
0
0
0.15
0
-0.03
0.03
2.34
-0.18
3.07
0.14
0.15
-0.02
-0.22
0.19
2.35
-0.13
3.27
0.14
0.17
-0.03
-0.25
0.22
2.33
-0.1
3.19
0.13
0.16
-0.02
-0.26
0.22
2.32
-0.06
3.25
0.14
0.17
-0.02
-0.27
0.23
Winter Wheat
Base Yield
No-Till
Optimal N fertilization*
Split N Fertilization
100% Residue Inc.
Nitrification Inhibitors
Reduced Fertilization
Increased Fertilization
2.94
0
0
0
0.1
0
-0.01
0
2.92
-0.13
1.55
0.09
0.11
0.03
-0.22
0.19
2.89
-0.11
1.82
0.1
0.12
0.04
-0.26
0.2
2.8
-0.07
1.87
0.11
0.13
0.04
-0.25
0.2
2.87
-0.05
1.78
0.11
0.12
0.05
-0.27
0.21
(continued)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-9
-------
CROPLANDS
Table 1-2: DAYCENT Base Mean Yields, and Differences from Mean Yield for Mitigation Strategies, by
Year (Metric tons of Grain per Hectare) (continued)
2010
2015
2020
2025
2030
^^^^^^^^^^^^^^^^^^^^^^^^^—
Spring Wheat
Base Yield
No-Till
Optimal N fertilization*
Split N Fertilization
100% Residue Inc.
Nitrification Inhibitors
Reduced Fertilization
Increased Fertilization
2.85
0
0
0
0.1
0
-0.03
0.02
2.94
-0.16
1.49
0.07
0.11
0.02
-0.2
0.14
2.92
-0.13
1.46
0.08
0.11
0.03
-0.22
0.15
2.85
-0.1
1.4
0.08
0.11
0.03
-0.21
0.14
2.83
-0.08
1.36
0.08
0.11
0.03
-0.21
0.14
Winter Barley
Base Yield
No-Till
Optimal N fertilization
Split N Fertilization
100% Residue Inc.
Nitrification Inhibitors
Reduced Fertilization
Increased Fertilization
3.55
0
0
0
0.35
0
0
0
3.59
-0.2
2.64
0.04
0.37
0.01
-0.34
0.31
3.58
-0.21
3.11
0.06
0.39
0.03
-0.39
0.35
3.5
-0.15
3.07
0.06
0.39
0.03
-0.41
0.36
3.57
-0.1
3
0.05
0.39
0.03
-0.43
0.38
Spring Barley
Base Yield
No-Till
Optimal N fertilization*
Split N Fertilization
100% Residue Inc.
Nitrification Inhibitors
Reduced Fertilization
Increased Fertilization
2.76
0
0
0
0.19
0
-0.04
0.04
2.83
-0.29
1.8
0.08
0.21
0.01
-0.28
0.24
2.79
-0.24
1.8
0.09
0.22
0.02
-0.31
0.26
2.77
-0.2
1.67
0.09
0.21
0.02
-0.31
0.25
2.77
-0.17
1.63
0.08
0.21
0.02
-0.32
0.25
Soybeans
Base Yield
No-Till
Optimal N fertilization*
Split N Fertilization
100% Residue Inc.
Nitrification Inhibitors
Reduced Fertilization
Increased Fertilization
2.9
0
0
0
0.02
0
0
0
2.95
-0.02
0.06
0
0.02
0
-0.01
0.01
2.94
-0.02
0.07
0
0.02
0.01
-0.01
0.01
2.92
-0.01
0.07
0
0.02
0.01
-0.01
0.01
2.92
-0.01
0.07
0
0.02
0.01
-0.01
0.01
*Note: Optimal N Fertilization, discussed below, is excluded from the main MAC analysis and presented for information only
V-10 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CROPLANDS
V.1.4 Marginal Abatement Costs Analysis
The MAC analysis assimilates the abatement measures' technology costs, expected benefits, and
emission reductions presented in Section X.3 to compute the cost of abatement for each measure. Similar
to the approach used in other non-CCfe sectors of this report, we compute a break-even price for each
abatement option for 195 countries to construct MAC curves illustrating the technical, net GHG
mitigation potential at specific break-even prices for 2010, 2020, and 2030.
This section describes the general modeling approach applied in this sector, which serve as additional
inputs to the MAC analysis that adjust the abatement project costs, benefits, and the technical abatement
potential in each country.
V. 1.4.1
Estimate Abatement Measure Costs and Benefits
As a general framework of the MAC analysis, the break-even price for each mitigation option is
calculated by setting total benefits (i.e., higher yields ) equal to total costs of a given mitigation option.
This framework, also referred to as the International Marginal Abatement Cost (IMAC) model, is
documented in USEPA (2006) and Beach et al. (2008).
V. 1.4.2
MAC Analysis Results
Global abatement potential in the Non-Rice Croplands sector equates to approximately 6 to 13% of its
total annual emissions between 2010 and 2030 at a relatively low carbon price of $5 per ton of CCh
equivalent ($/tCChe). Table 1-3 presents mitigation potential at selected break-even prices for 2030. GHG
mitigation and its cost-effectiveness vary significantly by country or region. Figure 1-4 displays the MAC
curve of the top-five emitting countries in 2010 and 2030.
Table 1-3: Abatement Potential at Selected Break-Even Prices in 2030 (No "Optimal Fertilization
"Scenario)
Country/Region
-10
Break-Even Price ($/tC02e)
5 10 15 20 30
^^H
•fffl
•fffVV
Top 5 Emitting Countries
China
U.S.A
India
Brazil
Argentina
11.2
5.4
2.5
0.2
0.6
11.2
5.4
2.9
0.2
0.6
11.2
5.5
3.1
0.2
0.6
11.2
5.5
3.1
0.2
0.6
11.3
5.5
3.6
0.2
0.7
11.3
5.5
3.6
0.2
0.7
11.3
5.5
3.6
0.2
0.7
12.1
8.7
3.6
0.2
0.7
12.1
8.7
4.0
0.2
0.7
12.1
8.8
4.0
0.2
0.7
12.8
10.9
5.3
2.1
1.0
Rest of Region
Africa
Asia
Central & South
America
Eurasia
Europe
Middle East
North America
World Total
1.7
1.5
0.3
0.2
3.0
0.2
0.6
27.4
1.9
1.6
0.4
0.2
3.0
0.3
0.6
28.3
2.1
1.8
0.4
0.2
3.4
0.8
0.6
30.0
2.2
1.9
0.5
0.2
3.5
0.8
0.6
30.4
2.2
1.9
0.6
0.3
3.6
0.8
0.8
31.5
2.2
2.0
0.8
0.3
3.6
0.8
0.8
31.8
2.2
2.0
0.8
0.3
3.8
1.3
0.8
32.4
2.3
2.0
0.8
0.4
4.1
1.3
0.9
37.2
2.3
2.3
0.9
1.7
4.3
1.4
0.9
39.6
2.8
2.5
1.1
2.3
6.0
1.4
1.0
43.0
3.9
3.0
1.8
2.7
8.7
1.7
1.9
55.8
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-11
-------
CROPLANDS
Figure 1-4: Marginal Abatement Cost Curve for Top-Five Emitting Countries in 2010 and 2030
(J
20.0
China 2010
— India 2010
USA 2010
— Brazil 2010
Argentina 2010
... China 2030
... USA2030
India 2030
... Brazil 2030
• —— Argentina 2030
-$30
Non-CO2 Reduction (MtCO2e)
Table 1-4 below presents a summary of estimated global total mitigation potential by mitigation
option. Overall the MAC analysis results suggest that No-till is the most effective strategy for GHG
mitigation in cropland soil management.11 This option accounts for approximately 70% of the total global
mitigation potential in 2010 and 43.7% in 2030. The second most significant mitigation option is reduced
fertilization, accounting for about 16% of the global total mitigation potential in 2010 and 40% in 2030.
Adoption of nitrification inhibitors and split fertilization may also make significant contributions to net
GHG reductions from cropland soil management.
11 As discussed above, mitigation potential from adoption of no-till practice is likely over-estimated with 100%
conventional tillage assumed in the business-as-usual baseline.
V-12
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CROPLANDS
Table 1-4: Global Total Abatement Potential from Cropland Soils by Measure (MtC02e) ("Optimal N
Fertilization Strategy excluded)
GHG Mitigation by Option (total all prices)
2010
Reduced Fertilization
Increased Fertilization
100% Residue Incorporation
Nitrification Inhibitors
Split N Fertilization
No-Till Adoption
Optimal N Fertilization
TOTAL
14.05
0.30
0.33
7.08
4.38
60.82
0.00
86.94
16%
0%
0%
8%
5%
70%
0%
100%
2020
18.09
0.03
0.18
6.46
3.14
42.47
0.00
70.37
26%
0%
0%
9%
4%
60%
0%
100%
2030
22.39
0.00
0.04
6.66
2.36
24.40
0.00
55.85
40.1%
0.0%
0.1%
11.9%
4.2%
43.7%
0.0%
100.0%
The relative mitigation potentials of no-till and reduced fertilization illustrate the difference between
dynamics of soil C and N2O and are worth a closer look. No-till dominates the mitigation potentials in the
early years, owing to its large effect on soil C. However, this dominance disappears over time as soils
become "saturated" with C. By 2030, the mitigation potential (limited to N2O) of reduced fertilization
nearly equals that of no-till. Over an even longer time scales, only the N2O flux remains as soils reach a
new equilibrium level of Soil C.
We tested the sensitivity of the results by adding an additional "Optimal N Fertilization" option,
which has substantial effects on global yields and emissions.
Optimal N fertilization
This option allows the model to maximize soil carbon through optimization of fertilizer inputs,
giving a "best case" result of the application of existing technology and crop patterns. Of course, baseline
levels vary widely from this optimum with some regions over-applying N and many under-applying N
relative to crop needs. This case shows what could be achieved if nutrient stress is removed at each time
step.
• Applicability: This option is available in all regions and all time periods
• Economic Applicability and Cost: Due to the large number of ways this option might be put in
practice, costs are limited to the change in N used.
• Additional factors: In cases where yields increase as a result, production is valued at the market
price. No tax or other benefits are included in this option.
This analysis resulted in the global MAC curve shown in Figure 1-5, and summarized in Table 1-5.
With Optimal N Fertilization included in the analysis, global mitigation increases from a maximum of 86
Mt to 129 Mt in 2010. Global mitigation in 2030 increases from a maximum of 56 Mt to about 86 Mt.
Overall the MAC analysis results suggest that optimal fertilization to achieve maximum crop yields is
potentially the single most significant source of GHG mitigation in cropland soil management. This
option accounts for approximately 44% of the total global mitigation potential in 2010 and 2030. The
second most significant mitigation option is no-till practice, accounting for about 39% of the global total
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-13
-------
CROPLANDS
mitigation potential.12 Reduction in N fertilizer application and adoption of nitrification inhibitors would
also make substantial contributions to net GHG reductions from cropland soil management.
Figure 1-5: Global Abatement Potential in Non-rice Croplands Management: 2010, 2020, and 2030
(Includes "Optimal N Fertilization" Strategy)
•2010
•2020
•2030
140
Non-CO2 Reduction (MtCO2e)
Table 1-5: Global Total Abatement Potential from Cropland Soils by Measure (MtC02e) (Includes "Optimal N
Fertilization" Strategy)
GHG Mitigation by Option (total all prices)
2010
Reduced Fertilization
Increased Fertilization
100% Residue Incorporation
Nitrification Inhibitors
Split N Fertilization
No-Till Adoption
Optimal N Fertilization
TOTAL
11.1
0.2
0.3
6.0
3.6
50.8
57.3
129.4
9%
0%
0%
5%
3%
39%
44%
100%
2020
0.0
0.0
0.1
5.6
2.7
35.4
42.2
86.1
14%
0%
0%
6%
3%
35%
42%
100%
2030
17.7
0.0
0.0
6.1
2.2
20.9
37.7
84.7
21%
0%
0%
7%
3%
25%
45%
100%
12 As discussed above, mitigation potential from adoption of no-till practice is likely over-estimated with 100%
conventional tillage assumed in the business-as-usual baseline.
V-14
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CROPLANDS
Figure 1-6 shows the effect on the top-5 countries. With "Optimal N Fertilization" included as a
strategy, China has the largest mitigation potential of any country and is also among the few countries
that have mitigation potential that increases over the 2010-2030 period. This appears to be related to
fertilizer use that is much higher than optimal.13 This suggests that N2O emissions may be reduced
without a yield, or soil C, penalty.
Figure 1-6: Marginal Abatement Cost Curve for Top 5 Emitters in 2010, 2030 (Includes "Optimal N
Fertilization" Strategy)
560
*oL
loi
lol
lol
o" I
< II
0
On
II
II
II
K
II
II
II
II
II
II
II
II
ii
1 -
1
n-1
i
i
i
! j
5.0 lOK)
•— '
i
(^ J
i'r^
P
China 2010
India 2010
USA 2010
Brazil 2010
Argentina 2010
... USA 2030
... India 2030
!5.0 30.0
... Brazil 2030
J — — — Argentina 2030
r1
r
Non-CO2 Reduction (MtCO2e)
V.1.5
Uncertainties and Limitations
Given the complexities of the global crop production sector, the estimated GHG mitigation potential
and marginal abatement cost curves are subject to a number of uncertainties and limitations:
• Optimistic assumptions on technology adoption. Mitigation technologies represent technical
potentials. The analysis assumes that if mitigation technology is considered feasible in a country
13 In the DAYCENT optimal fertilization scenario, where the model determined the optimal fertilizer rates, fertilizer
use typically decreased in China between 30 and 50% for major crops as compared to baseline levels. N2O emissions
also declined.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-15
-------
CROPLANDS
or region, it is fully adopted in 2010 and through the analysis period. Research suggests that
adoption of new technology in the agricultural sector is a gradual process and various factors
potentially slow the adoption of a new GHG-mitigating technology (e.g., farm characteristics,
access to information and capital, and cultural and institutional conditions). The mitigation
potential presented in this analysis should be viewed to represent the technical potential of the
mitigation options analyzed.
Availability and quality of data to represent the highly complex and heterogeneous crop
production systems of the world. Compared to the previous EPA marginal abatement cost curve
analysis (USEPA, 2006), there are major improvements in the datasets used to represent the
global crop production systems and the business-as-usual baseline conditions. However, data in
some areas, such as management practices which have significant influence on the GHG fluxes,
are not always available for all countries or regions. Approximations had to be made based on
limited literature or expert judgment. Moreover, collecting and developing regionally specific
cost estimates of emerging and/or not widely adopted management practices or mitigation
measures has been a challenge and in some cases global datasets had to be used.
Biophysical modeling uncertainties. The evaluation of simulated crop yields against observed
1 A
yields suggests that DAYCENT modeling performance varies by crop , leading to potential
biases in estimated GHG emissions. Model structure is found to be the largest contributor to
uncertainty in simulation results using DAYCENT, typically more than 75% of overall
uncertainty in estimates (Ogle et al. 2010, Del Grosso et al. 2010). Further model evaluation will
be carried out to understand potential model bias and prediction error using empirical based
procedure discussed in Ogle et al. (2007). In addition, soil carbon, which has a significant impact
on the net GHG emissions and mitigation potential from the sector, is particularly challenging to
simulate given the lack of monitoring data at the global scale. Sensitivity tests would be useful to
assess how alternative modeling approaches and assumptions may influence modeling results.
Potential interactions of multiple mitigation measures are not fully addressed in this analysis.
In this analysis, mitigation options are applied to independent segments of the crop production
systems to avoid double counting. In reality, multiple mitigation options can be applied, and
their order of adoption and potential interactions may affect the aggregate GHG mitigation.
Alternative approach should be investigated to provide more realistic representation of economic
applicability of potential mitigation measures.
14 Overall, simulated yields for maize agree reasonably well with observed yields; simulated average yields for
wheat, barley and sorghum are lower than observed yields; simulated average yields for soybean are above observed
yields.
V-16 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
CROPLANDS
References
Beach, R.H., DeAngelo, B., Rose, S., Salas, W., Li, C, DelGrosso, S. (2008). Mitigation potential and costs
for global agricultural greenhouse gas emissions. Agricultural Economics 38(2): 109-115.
Bronson, K.F., Mosier, A.R., Bishnoi, S.R., 1992. Nitrous oxide emissions in irrigated corn as affected by
nitrification inhibitors. Soil Sci. Soc. Am. J. 56,161-165.
Cramer, W., Field, C.B., 1999. Comparing global models of terrestrial net primary production (NPP):
overview and key results. Glob. Change Biol. 5 (Suppl. 1), 1-15.
De Klein, C., R.S.A. Novoa, S. Ogle, K.A. Smith, et al. 2006. Chapter 11: N2O emissions from managed
soil, and CCh emissions from lime and urea application. In 2006 IPCC guidelines for national greenhouse
gas inventories, Vol. 4: Agriculture, forestry and other land use, edited by S. Eggleston, L. Buendia, K.
Miwa, T. Ngara and K. Tanabe. Kanagawa, Japan: ICES.
Del Grosso, S.J., Parton, W.J., Mosier, A.R., Hartman, M.D., Brenner, J., Ojima, D.S., Schimel, D.S. 2001.
Simulated interaction of carbon dynamics and nitrogen trace gas fluxes using the DAYCENT model.
In: Schaffer, M., et al. (Eds.), Modeling Carbon and Nitrogen Dynamics for Soil Management, p. 303-
332, CRC Press, Boca Raton, Florida, USA.
Del Grosso, S.J., Ojima, D.S., Parton, W.J., Mosier, A.R., Peterson, G.A., Schimel, D.S., 2002. Simulated
effects of dryland cropping intensification on soil organic matter and greenhouse gas exchanges
using the DAYCENT ecosystem model. Environmental Pollution 116: S75-S83.
Del Grosso, S.J., A.R. Mosier, W.J. Parton and D.S. Ojima. 2005. DAYCENT model analysis of past and
contemporary soil N2O and net greenhouse gas flux for major crops in the USA. Soil Tillage and
Research 83:9-24.
Del Grosso, S.J., D.S. Ojima, W.J. Parton, E. Stehfest, M. Heistemann, B. DeAngelo, S. Rose. 2009. Global
Scale DAYCENT Model Analysis of Greenhouse Gas Mitigation Strategies for Cropped Soils. Global
and Planetary Change, 67, 44-50.
Del Grosso, S., S.M. Ogle, W.J. Parton, and F.J. Breidt. 2010. Estimating uncertainty in N2O emissions from
US cropland soils. Global Biogeochemical Cycles, 24, GB1009, doi:10.1029/2009GB003544.
FAO, 1996. The Digitized Soil Map of the World Including Derived Soil Properties, CDROM. Food and
Agriculture Organization, Rome.
Melillo, J.M., McGuire, A.D., Kicklighter, D.W., Moore III, B., Vorosmarty, C.J., Schloss, A.L. 1993. Global
climate change and terrestrial net primary production. Nature 363, 234-240.
Metherell, A.K., L.A. Harding, C.V. Cole, and W.J. Parton (1993) "CENTURY Soil Organic Matter Model
Environment." Agroecosystem version 4.0. Technical documentation, GPSR Tech. Report No. 4,
USDA/ARS, Ft. Collins, CO.
Monfreda, C., N. Ramankutty and J.A. Foley. (2008), "Farming the planet: 2. Geographic distribution of
crop areas, yields, physiological types, and net primary production in the year 2000", Global
Biogeochemical Cycles, Vol.22, GB1022, doi:10.1029/2007GB002947.
Mosier, A. R., J.M. Duxbury, J.R. Freney, O. Heinemeyer, K. Minami. 1998. Assessing and mitigating N2O
emissions from agricultural soils. Climatic Change 40:7-38.
Ogle, S.M., F.J. Breidt, M. Easter, S. Williams and K. Paustian. 2007. Empirically based uncertainty
associated with modeling carbon sequestration in soils. Ecological Modelling 205:453-463.
Ogle, S.M., F.J. Breidt, M. Easter, S. Williams, K. Killian, and K. Paustian. 2010. Scale and uncertainty in
modeled soil organic carbon stock changes for US croplands using a process-based model. Global
Change Biology 16:810-820.
Ogle, S.M., A. Swan and K. Paustian. 2012. No-till management impacts on crop productivity, carbon
input and soil carbon sequestration. Agriculture, Ecosystems and Environment 149:37-49.
Parton, W.J., D.S. Schimel, C.V. Cole, D.S. Ojima (1987) "Analysis of factors controlling soil organic matter
levels in Great Plains grasslands." Soil Science Society of America Journal 51:1173-1179.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-17
-------
CROPLANDS
Parton, W.J., J.W.B. Stewart, C.V. Cole. (1988) "Dynamics of C, N, P, and S in grassland soils: a model."
Biogeochemistry 5:109-131.
Parton, W.J., D.S. Ojima, C.V. Cole, and D.S. Schimel (1994) "A General Model for Soil Organic Matter
Dynamics: Sensitivity to litter chemistry, texture and management," in Quantitative Modeling of Soil
Forming Processes. Special Publication 39, Soil Science Society of America, Madison, WI, 147-167.
Parton, W.J., M.D. Hartman, D.S. Ojima, and D.S. Schimel (1998) "DAYCENT: Its Land Surface
Submodel: Description and Testing". Glob. Planet. Chang. 19: 35-48.
Paustian, K., H. P. Collins, and E. A. Paul. 1997. Management controls on soil carbon. Pages 15-49 in E. T.
E. Paul E.A., K. Paustian, and C.V. Cole, editor. Soil organic matter in temperate agroecosystems:
long-term experiments in North America. CRC Press, Boca Raton.
Robertson, G. P., E.A. Paul, and R.R. Harwood. 2000. Greenhouse gases in intensive agriculture:
contributions of individual gases to the radiative forcing of the atmosphere. Science 289:1922-1925.
Ramankutty, N. and J.A. Foley. 1998. Characterizing patterns of global land use: an analysis of global
croplands data. Global Biogeochemical Cycles 12(4), 667-685.
Ramankutty et al. 2008. "Farming the planet: 1. Geographic distribution of global agricultural lands in the
year 2000", Global Biogeochemical Cycles, Vol. 22, GB1003, doi:10.1029/2007GB002952.
Sacks, W.J., D. Deryng, J.A. Foley, and N. Ramankutty. 2010. Crop planting dates: an analysis of global
patterns. Global Ecology and Biogeography 19, 607-620. DOI: 10.1111/j.l466-8238.2010.00551.x.
Six, J., E. T. Elliott, and K. Paustian. 2000. Soil macroaggregate turnover and microaggregate formation: A
mechanism for C sequestration under no-tillage agriculture. Soil Biology and Biochemistry 32:2099-2103.
Smith, P., D. Martino, Z. Cai, D. Gwary, H. Janzen, P. Kumar, B. McCarl, S. Ogle, F. O'Mara, C. Rice, B.
Scholes, O. Sirotenko,2007: Agriculture. In Climate Change 2007: Mitigation. Contribution of
Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change [B. Metz, O.R. Davidson, P.R. Bosch, R. Dave, L.A. Meyer (eds)], Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA.
Nelson, G.C., M.W. Rosegrant, A. Palazzo, I. Gray, C. Ingersoll, R. Robertson, S. Tokgoz, T. Zhu, T.B.
Sulser, C. Ringler, S. Msangi, and L. You. (2010). Food Security, Farming, and Climate Change to
2050: Scenarios, Results, Policy Options. International Food Policy Research Institute: Washington,
DC.
Smith, W.N., R.L. Desjardins, B. Grant, C. Li, R. Lemke, P. Rochette, M.D. Corre, and D. Pennock. (2002).
Testing the DNDC model using N2O emissions at two experimental sites in Canada. Canada Journal of
Soil Science 82:365-374.
U.S. Environmental Protection Agency (USEPA). 2006. Global Mitigation ofNon-COi Greenhouse Gases. EPA
430-R-06-005. Washington DC.
U.S. Environmental Protection Agency (USEPA). 2012. Global Anthropogenic Non-COi Greenhouse Gas
Emissions: 1990-2030. EPA 430-R-12-006. Washington DC.
U.S. Environmental Protection Agency (USEPA). 2013. Inventory of U.S. Greenhouse Gas Emissions and
Sinks: 1990-2011. Washington, D.C.
World Resources Institute (WRI). 2012. Climate Analysis Indicators Tool (CAIT) database. Available at
http://www.wri.org/project/cait/.
V-18 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
RICE
V.2. Rice Cultivation
V.2.1
Sector Summary
kice cultivation is an important global source of methane (Cl-k) and nitrous oxide (N2O)
emissions. There are also changes in soil organic carbon (C) stocks and associated CCh fluxes.
When paddy fields are flooded, decomposition of organic material gradually depletes the
oxygen present in the soil and floodwater, causing anaerobic conditions in the soil. Anaerobic
decomposition of soil organic matter by methanogenic bacteria generates Q-k. Some of this Q-k is
dissolved in the floodwater, but the remainder is released to the atmosphere, primarily through the rice
plants themselves. Minor amounts of Q-k also escape from the soil via diffusion and bubbling through
the floodwaters. In addition, as with other crops, human activities influence soil N2O emissions through
addition of synthetic and organic nitrogen fertilizers and other practices and soil C stocks through
residue management as well as any practices that effect crop yields.
In 2010, the net global GHG emissions from rice cultivation were approximately 561 MTCO2e. The
top 5 emitting countries - India, Indonesia, Bangladesh, Vietnam, and China -accounted for 77% of the
global total net emissions. Figure 2-1 displays the baseline net global GHG emissions for the rice sector.
Net GHG emissions from rice cultivation are projected to grow by 33% to 750 MTCC^e between 2010 and
2030. There is a small amount of growth in emissions occurring in developing regions to meet the
demand for rice products from growing populations and higher incomes, but the biggest contributor to
the increase in net GHG emissions simulated between 2010 and 2030 is a reduction in the soil C sink over
time. In the Denitrification-Decomposition (DNDC) model, there are fairly large increases in soil C in the
initial periods in many countries as they have recently changed practices to incorporate more residues
into the soil. However, as soil C moves to a new equilibrium, the incremental changes in future years
become much smaller and offset a smaller portion of the non-CCh emissions.
Figure 2-1: Net GHG Emissions Projections for Rice Cultivation: 2000-2030
722
756
I China
I Vietnam
Bangladesh
I Indonesia
India
ROW
2010
2030
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-19
-------
RICE
Table 2-1 shows the baseline Q-k, N2O and soil carbon estimates for rice cropland by region. Rice
cultivation results in emissions of Q-k and N2O, and these are offset by storage of carbon in the soil. In
2010, GHG emissions from rice cultivation include 484.1 MTCO2e CH4 and 260.0 MTCO2e N2O, offset by
179.2 MTCC^e of c stored in the soil, for net non-CCfe emissions of 564.9.1 MTCC^e, or about 5.8 percent
of global non-CO2 emissions (EPA, 2012).
Table 2-1: Baseline CH4, N20, and Soil Carbon Estimates for Rice Cropland for 2010, 2020 and 2030 by
Region
Country/Region
CH4
2010
N20
SoilC
CH4
2020
N20
SoilC
CH4
2030
N20
SoilC
Top 5 Emitting Countries
India
Indonesia
China
Vietnam
Bangladesh
91.2
81.7
72.9
47
54.4
76.7
25.5
34.6
25.7
63
-50
2.2
-69.4
-4.8
-16
94
75.4
72.8
45.4
54.3
93.2
23.4
36.7
33.1
98.6
-27.5
-0.5
-31.3
-2.8
-8.5
89
70.7
66.5
44
54.5
94.1
22.1
35.9
34.5
112.4
-18.6
-1.3
-16.9
-1.8
-5
Rest of Region
Africa
Asia
Central & South
America
Eurasia
Europe
Middle East
North America
World Total
11.6
79.8
32.3
1
1.8
2.8
7.5
484.1
4.5
22.8
4.5
0.1
0.1
0.1
2.3
260
-3.8
-26.6
-5.1
-0.1
-1.4
-1.4
-2.8
-179.2
12.6
85.9
33.5
1.2
2.2
3.6
8.3
489.2
6.2
25.3
5.3
0.1
0.1
0.1
2.4
324.5
-2.7
-13.2
-3.2
0.1
-0.6
-0.6
-0.6
-91.6
13.4
85.9
33.4
1.3
2.3
3.9
8.1
472.9
7
25.9
5.6
0.1
0.1
0.1
2.6
340.5
-2.1
-8.5
-2.2
0
-0.4
-0.4
-0.4
-57.4
Global abatement potential in paddy rice cultivation systems equates to approximately 27% - 35% of
total annual net emissions. Marginal abatement cost (MAC) curve results are presented in Figure 2-2 for
2010, 2020, and 2030, assuming that production remains equal to baseline levels under the mitigation
scenarios. Maximum abatement potential in the rice sector is 199 MtCChe in 2010, 203 MtCChe in 2020 and
200 MtCChe in 2030.
Figure 2-2 also shows the finding that significant reductions are feasible even at a low values per ton
of carbon. For example, there are approximately 76 MtCChe of net GHG emission reductions that are cost-
effective in 2010 at a price of $5/ton, (13.5 % of the baseline estimate). In 2030, approximately 87 MtCCh of
reductions are feasible at a price of $5 per ton (11.5 % of the baseline estimate). These results suggest that
there are significant opportunities for net GHG reductions in the rice cultivation sector.
V-20
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
RICE
Figure 2-2: Global Abatement Potential in Rice Cultivation with Production Equal to Baseline Levels:
2010, 2020, and 2030
•2010
•2020
•2030
180
Non-CO2 Reduction (MtCO2e)
The following section offers a brief description of CH4 and N2O emissions as well as changes in soil
carbon stock from rice cultivation, and a discussion of projected trends in global baseline emissions. The
subsequent section presents possible abatement technologies, their technical specifications, costs and
potential benefits. The final section discusses the estimated abatement potential and MAC analysis at a
regional level.
V.2.2 CH4 and N2O Emissions and Changes in Soil Carbon from
Rice production is a major source of GHG emissions. Global, Tier-I datasets such as EPA's Global
Anthropogenic Non-CCfe GHG Emissions Report (EPA, 2012) show that agriculture is the biggest source
of Q-k emissions, and within agriculture, rice cultivation is the second largest source, behind enteric
fermentation.1 Rice cultivation accounted for 7% of global CH4 emissions in 2005 (USEPA, 2012). Rice
cultivation is also a significant source of N2O emissions but these are not included in most global datasets.
1 Global CH4emissions from agriculture were estimated at 3,035.4 MtCCtee (2005), about 45% of the global total of
6815.8 MtCCtee. Rice produced 500.9 MtCCtee and enteric fermentation produced 1,894.3 MtCCtee (USEPA, 2012,
Table 6).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-21
-------
RICE
In this section, we describe baseline emissions of CHi,, N2O, and soil carbon from rice cultivation as well
as crop production data and assumptions that support the analysis of mitigation potential.
Rice production systems can be classified as wetland rice (irrigated, rain-fed and deepwater) or
upland rice (Neue, 1993). Wetland rice is the largest category, and is responsible for large net Q-k
emissions.2 Aerobic decomposition of organic matter gradually depletes the oxygen present in the soil
and water, resulting in anaerobic conditions in the rice paddies. Methanogenic bacteria decompose soil
organic matter under anaerobic conditions in rice paddies, generating Q-k. Significant amounts of Q-k
are oxidized by aerobic methanotrophic bacteria into CCh in the soil. The remaining unoxidized Q-k is
released to the atmosphere through diffusion and ebullition and through roots and stems of rice plants.
Thus, unlike the non-paddy rice agricultural soils which are typically Q-k sinks, paddy rice cultivation is
a major source of Q-k emissions.
N2O is another significant component of net GHG emissions from rice cultivation. N2O is produced
through nitrification and denitrification from microbial activities under the anoxic condition. N2O
emissions occur directly from soils, and indirectly through volatilization of compounds such as NHs and
NOx and subsequent deposition as well as through leaching and runoff. Table 2-1 shows that in 2010,
while CH4 accounted for the largest share of emissions with 484.1 MtCChe (65% of non-CCh emissions
from rice cultivation), N2O contributed substantially, with 260.0 MtCChe (35%). Both dry and irrigated
rice are a source of N2O emissions.
Soil carbon stocks are not a non-CCh GHG but also have important implications for net GHG
emissions and are affected by non-CCh mitigation options so we estimate total emissions net of their
effect on soil C stocks in this report.
V.2.2.1 Activity Data or Important Sectoral or Regional Trends and Related
Assumptions
DA/DC Modeling of GHG Fluxes and Crop Yields
The Denitrification-Decomposition (DNDC) model was used to simulate production, crop yields and
greenhouse gas fluxes of global paddy rice under "business-as-usual" (BAU) condition and various
mitigation strategies. DNDC is a soil biogeochemical model that simulates the processes determining the
interactions among ecological drivers, soil environmental factors, and relevant biochemical or
geochemical reactions, which collectively determine the rates of trace gas production and consumption in
agricultural ecosystems (Li, 2001). Details of management (e.g., crop rotation, tillage, fertilization, manure
amendment, irrigation, weeding, and grazing) have been parameterized and linked to the various
biogeochemical processes (e.g., crop growth, litter production, soil water infiltration, decomposition,
nitrification, denitrification, fermentation) embedded in DNDC (e.g., Li et al., 2004; Li et al., 2006; Li, 2011;
Abdalla et al., 2011; Giltrap et al., 2011; Dai et al., 2012).3
DNDC predicts daily CH4, N2O and soil carbon fluxes from rice paddies through the growing and
fallow seasons as fields remain flooded or move between flooded and drained conditions during the
season.
Globally, about 2 percent of rice is grown in dry conditions and this production system is a net sink for CH4 (source:
DNDC estimates discussed below).
3 The paddy-rice version of DNDC has been validated for a number of countries and world regions and is used for
national trace gas inventory studies in North America, Europe, and Asia (e.g., Smith et al., 2002; Follador et al., 2011;
Leip et al., 2011; Li et al., 2002; Cai et al., 2003; Li et al., 2005).
V-22 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
RICE
For this study, a modified version of the DNDC 9.5 Globe database was used to simulate crop yields
and GHG fluxes from global paddy rice cultivation systems. The DNDC 9.5 global database contains
information on soil characteristics, crop planted area, and management conditions (fertilization,
irrigation, season, and tillage) on a 0.5 by 0.5 degree grid cell of the world. The database is used to
establish the initial conditions in the model in 2000. The model considers all paddy rice production
systems, including irrigated and rainfed rice, and single, double and mixed rice as well as deepwater and
upland cropping systems. For this study, baseline and mitigation scenario modeling is carried out for all
rice-producing countries in the world that produce a substantial quantity of rice.
The Food and Agriculture Organization (FAO) country-level statistics (FAOSTAT 2010) were used to
establish harvested area for rice. The total area was calculated for each country in the Globe database for
each type, and evenly distributed across all grid cells within a country in the absence of sub-national
information. Figure 2-3 shows the distribution of rice across major systems for the five largest producers
and an aggregate of the rest of the world.
Figure 2-3: DNDC Rice Cropland Area Sown, Top 5 countries, by Type and Water Management
Deepwater
Upland
I Irrigated - Dry Seeding
Irrigated - Midseason Drainage
I Irrigated - Continuous Flooding
I Rainfed
^>
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-23
-------
Table 2-2: Baseline yields for 2010, 2020 and 2030 for selected countries (kg/ha)
2010
Country/Region Irrigated
Top 5 Countries
China
India
Indonesia
Bangladesh
Vietnam
by Production
6,158.2
4,832.6
5,546.1
7,322.9
7,388.0
Rainfed
4,002.9
1,681.5
4,758.1
4,823.1
5,208.4
Deepwater
622.5
685.3
1,233.3
1,257.6
963.2
Upland
2,280.5
1,114.1
2,142.8
2,501.9
2,240.9
Irrigated
6,522.0
5,271.8
5,625.0
7,447.9
7,503.2
2020
Rainfed
4,193.0
1,745.2
4,756.5
4,732.3
5,156.1
Deepwater Upland
702.5 2,560.9
846.7 1,266.8
1,167.0 2,010.9
1,592.3 2,927.3
940.4 2,386.4
Irrigated
7,161.0
5,722.5
5,833.2
7,642.4
7,647.7
2030
Rainfed
4,583.2
1,849.3
4,859.4
4,766.4
5,222.3
Deepwater
814.9
993.5
1,171.8
1,857.9
960.5
Upland
2,920.1
1,409.6
1,961.6
3,196.2
2,513.0
•5
o
Table 2-3: Baseline production for 2010, 2020 and 2030 for selected countries (metric tonnes)
2010
Country/ Region Irrigated
Rainfed
Deepwater
Upland
Irrigated
2020
Rainfed
Deepwater
Upland
Irrigated
2030
Rainfed
Deepwater
Upland
Top 5 Countries by Production
China 185,106,646
India 128,759,438
Indonesia 44,515,927
Bangladesh 22,123,824
Vietnam 27,265,526
136,866
18,667,823
17,723,001
29,082,198
15,725,436
532
701,074
3,144
1,870,274
260,980
72,470
4,544,661
3,230,419
2,471,779
1,121,298
187,171,179
137,564,781
45,020,613
22,891,662
27,967,786
136,879
18,975,458
17,666,403
29,029,646
15,723,459
573
848,345
2,966
2,409,109
257,362
77,697
5,061,283
3,022,817
2,942,311
1,206,084
187,849,985
141,353,559
45,509,275
23,309,923
28,142,533
136,761
19,034,013
17,593,406
29,015,593
15,722,150
607
942,360
2,903
2,789,425
259,495
80,982
5,330,762
2,874,405
3,187,968
1,253,843
-------
RICE
The global meteorological data from the National Oceanic and Atmospheric Administration's
National Centers for Environmental Prediction climate reanalysis product were used to establish climate
data for 2010 in the model. The 2010 climate data were used for all model years. Planting and harvest
dates were matched approximately to local growing season. Tillage and flooding and drainage dates for
irrigated rice were established based on the planting dates.
Nitrogen fertilizer application rates were based on DNDC fertilizer use data, which is derived from
global data sources. Table 2-4 summarizes the assumed fertilizer use per hectare for rice by country.
Assumptions on the distribution of irrigated rice across water management regimes for each country
were developed based on Yan et al. (2009) (see Table 2-5).
Table 2-4: DNDC Average N Fertilizer Application Rate by Country and Rice Production Type
Planted Area-Weighted Mean Fertilizer N Rate
(kgN/ha)
Irrigated Ramfed Upland Deepwater
Country
Planted Area
Afghanistan
Angola
Argentina
Australia
Azerbaijan
Bangladesh
Belize
Benin
Bhutan
Bolivia
Brazil
Brunei
Bulgaria
Burkina-Faso
Burundi
Cambodia
Cameroon
Central-African-Republic
Chad
Chile
China
Colombia
Congo
Costa-Rica
Cote-dlvoire
Cuba
208,030
2,465
211,148
175,085
5,720
11,526,108
5,303
24,138
30,472
232,626
2,696,270
613
24,732
133,240
18,582
2,730,963
32,568
13,560
118,190
49,282
30,125,402
435,924
520,829
87,372
493,322
196,891
40
1
90
180
20
107
50
50
40
30
50
5
60
25
40
30
35
30
10
50
164
108
2
50
7
28
40
1
90
180
20
107
50
50
40
30
50
5
60
25
40
30
35
30
10
50
164
108
2
50
7
28
— —
— —
9 -
15 -
1 -
30 -
11 -
2 -
— —
1 -
15 -
— —
24 -
1 -
— —
— —
1 -
— —
2 -
30 -
23 -
18 -
— —
18 -
2 -
6 -
(continued)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-25
-------
RICE
Table 2-4: DNDC Average N Fertilizer Application Rate by Country and Rice Production Type (continued)
Country
(kgN/ha)
Planted Area Irrigated Rainfed Upland Deepwater
Dominican-Republic
Ecuador
Egypt
El-Salvador
Ethiopia
France
French-Guiana
Gabon
Ghana
Greece
Guatemala
Guinea
Guinea-Bissau
Guyana
Haiti
Honduras
Hungary
India
Indonesia
Iran
Iraq
Italy
Japan
Kazakhstan
Kenya
Korea-North
Korea-South
Kyrgyzstan
Laos
Liberia
Madagascar
Malawi
Malaysia
Mali
208,865
454,982
402,249
8,674
40
18,919
10,920
202
105,678
42,021
25,578
818,010
162,054
187,731
82,387
10,531
53,797
42,848,326
13,261,499
563,918
47,978
220,850
1,627,707
97,643
7,358
582,246
902,339
14,724
848,955
79,879
1,703,119
28,106
677,984
646,334
35
55
203
88
25
127
20
35
30
94
40
1
30
5
10
40
35
69
82
79
40
99
80
30
50
70
189
39
45
10
—
20
65
40
35
55
203
88
25
127
20
35
30
94
40
1
30
5
10
40
35
69
82
79
40
99
80
30
50
70
189
39
45
10
—
20
65
40
10
6
34
19
3
28
8
—
—
20
15
—
1
11
2
26
15
20
16
17
56
22
24
—
8
15
34
5
2
—
1
9
16
2
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
(continued)
V-26
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
RICE
Table 2-4: DNDC Average N Fertilizer Application Rate by Country and Rice Production Type (continued)
Country
(kgN/ha)
Planted Area Irrigated Rainfed Upland Deepwater
Mauritania
Mexico
Morocco
Mozambique
Myanmar
Nepal
Nicaragua
Niger
Nigeria
Pakistan
Panama
Paraguay
Peru
Philippines
Portugal
Romania
Russia
Rwanda
Senegal
Sierra-Leone
[Spain
Sri-Lanka
Sudan
Suriname
Switzerland
Tajikistan
Tanzania
Thailand
The-Gambia
Togo
Trinidad-Tobago
Turkey
Turkmenistan
Uganda
28,607
162,208
12,110
64,834
8,013,037
1,455,906
136,469
41,083
2,415,653
2,366,291
110,696
44,291
383,322
4,355,767
88,342
13,191
200,099
3,790
75,558
500,905
122,793
1,062,007
303
39,758
2,320
31,808
1,058,671
12,116,749
12,677
39,899
2,838
99,015
60,042
54,966
85
85
120
5
50
22
85
10
20
40
10
85
170
60
90
85
85
85
85
25
76
60
45
50
40
85
30
30
10
8
35
127
30
30
85
85
120
5
50
22
85
10
20
40
10
85
170
60
90
85
85
85
85
25
76
60
45
50
40
85
30
30
10
8
35
127
30
30
—
18
13
1
8
5
5
—
3
20
9
2
17
19
10
6
3
—
4
—
17
16
1
27
27
2
1
20
—
3
21
20
11
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
(continued)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-27
-------
RICE
Table 2-4: DNDC Average N Fertilizer Application Rate by Country and Rice Production Type (continued)
Country
(kgN/ha)
Planted Area Irrigated Rainfed Upland Deepwater
Ukraine
United-States
Uruguay
Uzbekistan
Venezuela
Vietnam
Zambia
Zimbabwe
29,078
1,444,924
174,987
36,221
295,441
7,481,119
13,872
176
85
139
151
90
85
120
12
15
85
139
151
90
85
120
12
15
3 -
19 -
11 -
30 -
16 -
29 -
4 -
8 -
Table 2-5: Distribution of Baseline Water Management for Irrigated Rice by Country (%)
Region
Afghanistan
Algeria
Angola
Argentina
Australia
Azerbaijan
Bangladesh
Belize
Benin
Bhutan
Bolivia
Brazil
Brunei
Bulgaria
Burkina-Faso
Burundi
Cameroon
Central-African-Republic
Chad
Chile
China
Colombia
Comoros
Continuous Flooding 1
100%
100%
100%
100%
100%
100%
20%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
20%
100%
100%
iiBBIBBBBff
0%
0%
0%
0%
0%
0%
80%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
80%
0%
0%
e Dry Seeding
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
V-28
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
RICE
Table 2-5: Distribution of Baseline Water Management for Irrigated Rice by Country (%) (continued)
Region
Congo
Costa-Rica
Cote-dlvoire
Cuba
Dominican-Republic
Ecuador
Egypt
El-Salvador
Ethiopia
Fiji
France
French-Guiana
Gabon
Ghana
Greece
Guatemala
Guinea
Guinea-Bissau
Guyana
Haiti
Honduras
Hungary
India
Indonesia
Iran
Iraq
Italy
Jamaica
Japan
Kazakhstan
Kenya
Korea-North
Korea-South
Kyrgyzstan
Liberia
Macedonia
Continuous Flooding 1
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
30%
43%
100%
100%
100%
100%
20%
100%
100%
100%
100%
100%
100%
100%
4ffi^^^9ffl^£
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
70%
57%
0%
0%
0%
0%
80%
0%
0%
0%
0%
0%
0%
0%
e Dry Seeding
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
(continued)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-29
-------
RICE
Table 2-5: Distribution of Baseline Water Management for Irrigated Rice by Country (%) (continued)
Region
Madagascar
Malawi
Malaysia
Mali
Mauritania
Mexico
Micronesia
monsoon Asia
Morocco
Mozambique
Nepal
Nicaragua
Niger
Nigeria
Pakistan
Panama
Papua-New-Guinea
Paraguay
Peru
Philippines
Portugal
Reunion
Romania
Russia
Rwanda
Senegal
Sierra-Leone
Solomon-Is
Somalia
South-Africa
Spain
Sri-Lanka
Sudan
Suriname
Swaziland
Tajikistan
Continuous Flooding 1
100%
100%
100%
100%
100%
100%
100%
43%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
4ffi^^^9ffl^£
0%
0%
0%
0%
0%
0%
0%
57%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
e Dry Seeding
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
V-30 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
RICE
Table 2-5: Distribution of Baseline Water Management for Irrigated Rice by Country (%) (continued)
Region
Tanzania
The-Gambia
Timor-Leste
Togo
Trinidad-Tobago
Turkey
Turkmenistan
Uganda
Ukraine
United-States-California
United-States-Mid_South
Uruguay
Uzbekistan
Venezuela
Vietnam
Zambia
Zimbabwe
Continuous Flooding R
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
0%
100%
100%
100%
100%
100%
100%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
e Dry Seeding
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
100%
0%
0%
0%
0%
0%
0%
Source: Van etal. (2009).
A baseline scenario is established for each country using DNDC 9.5, reflecting assumptions on water
management, fertilizer application, residue management and tillage practices described above. Rice
yields and GHG fluxes (Q-k, direct and indirect N2O and changes in soil organic carbon) were simulated
in DNDC model for each grid cell and results were aggregated at the country level for irrigated, rainfed,
deep water, and upland production systems for each scenario, in both mean annual rates per hectare and
mean annual national totals.4 Results were reported for 2010 and by 5-year increments through 2030.
Finally, results from DNDC were adjusted with projected acreage of these production systems by the
International Food Policy Research Institute (IFPRI)'s International Model for Policy Analysis of
Agricultural Commodities and Trade (IMPACT) model. In DNDC, rice production areas were held
constant at the 2010 level to obtain the biophysical effects of management practice changes on crop yields
and GHG fluxes. Projected acreage changes from IMPACT model reflect socio-economic drivers (such as
population growth) and technological changes to meet the global food demand (Nelson et al., 2010). The
IMPACT modeling projects that while global rice production would increase by 11 percent between 2010
and 2030, the total area dedicated to rice cultivation would decrease by 5 percent during the same period
due to productivity improvements.
4 The mean values were calculated using weighted averages; rice yields represent total annual yields of rice from all
production systems.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-31
-------
RICE
V.2.2.2 Emissions Estimates and Related Assumptions
This section briefly discusses the historical and projected emission trends from global paddy rice
cultivation and presents simulated baseline emissions projections.
Historical Emissions Estimates
According to the EPA Global Emissions Report (GER) (USEPA, 2012), total methane emissions from
global rice cultivation increased by 4.4% between 1990 and 2005, from 480 MtCChe to 501 MtCChe. Asia,
the predominant rice-producing region, accounted for over 80% of the total Q-k emissions in 2005. Africa
contributed another 10%, and the remaining methane emissions in this sector came from Central and
South America and other regions. The GER did not report historic N2O emissions and soil carbon stock
changes from the rice cultivation sector.
Projected Emissions Estimates
Worldwide Q-k and N2O emissions from rice cultivation are projected to have only modest increases
between 2010 and 2030. This is mainly because demand for rice products will remain relatively constant
while global food demand shifts towards more livestock and other more expensive food products with
higher incomes. The estimated total CH4 emissions from rice cultivation are 484.1 MtCChe in 2010, 482.2
MtCChe in 2020 and 472.9 MtCChe in 2030. The total estimated N2O emissions are 260.0 MtCChe in 2010,
324.5 MtCChe in 2020 and 340.5 MtCChe in 2030.
V.2.3 Abatement Measures and Engineering Cost Analysis
The mitigation options included in the analysis were based on review of the literature to identify the
most promising options, while also taking data availability and potential for modeling within DNDC into
consideration. For the purposes of developing MAC curves for this study, mitigation options that
increase net emissions of non-CCh GHG were excluded from the analysis.
Twenty-six mitigation scenarios were then analyzed using DNDC 9.55. The scenarios addressed
management techniques in various combinations hypothesized to reduce GHG emissions from rice
systems: water management regime (continuous flooding, mid-season drainage, dry seeding, alternate
wetting and drying, and switching to dryland rice production system), residue management (partial or
total residue incorporation), tillage, and various fertilizer management alternatives (ammonium sulfate in
place of urea, urea with nitrification inhibitor, slow release urea, 10% reduced fertilizer, 20% reduced
fertilizer, and 30% reduced fertilizer).
The water management system under which rice is produced is one of the most important factors
influencing Q-k emissions. Specifically, switching from continuous flooding of rice paddy fields to
draining flooded fields periodically during the growing season - a water conservation practice that is
increasingly adopted in the baseline to reduce water use - would significantly reduce Cl-k emissions.
Other practices (e.g., fertilizer applications, tillage practices and residue management) also alter the soil
conditions and hence affect crop yields and the soil carbon- and nitrogen-driving processes such as
5 Note that 38 different scenario names are reported in the outputs. Because water management practices are
assumed not to affect non-irrigated rice emissions, the simulation results for options combine d with continuous
flooding or midseason drainage are the same for non-irrigated rice. The analogous options that alter fertilizer and
other management practices but do not affect water management were identified as beginning with "base" rather
than "cf" or "md".
V-32 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
RICE
decomposition, nitrification and denitrification (Neue and Sass, 1994; Li et al., 2006). Due to the complex
interactions, changes in management practices would trigger changes in multiple GHG fluxes. For
instance, while drainage of rice fields during the growing season would significantly reduce Q-k
emissions, emissions of N2O actually increase (Zheng et al., 1997, 2000; Cai et al., 1999; Zou et al. 2007).
Rice mitigation options
The mitigation options included for rice water management system under which rice is produced is
one of the most important factors influencing Q-k emissions. Specifically, switching from continuous
flooding of rice paddy fields to draining flooded fields periodically during the growing season - a water
conservation practice that is increasingly adopted in the baseline to reduce water use - would
significantly reduce Q-k emissions. Other practices (e.g., fertilizer applications, tillage practices and
residue management) also alter the soil conditions and hence affect crop yields and soil carbon- and
nitrogen-driving processes.
There were 26 scenarios that were run using DNDC 9.5 (see Table 2-6). The scenarios addressed
management techniques in various combinations hypothesized to reduce GHG emissions from rice
systems: flood regime (continuous flooding [CF], mid-season drainage [MD], dry seeding [DS], alternate
wetting and drying [AWD], and switching to dryland (upland) rice), residue management (partial
removal or 100% incorporation), conventional tillage or no till, and various fertilizer alternatives
(conventional / urea, ammonium sulfate in place of urea, urea with nitrification inhibitor, slow release
urea, 10% reduced fertilizer, 20% reduced fertilizer, 30% reduced fertilizer, and DNDC optimization of
fertilizer application to maximize yields). Further definition of these assumptions is provided in Table 2-
7.
Table 2-6: Alternative Rice Management Scenarios Simulated using DNDC
Abbreviation
cf_r50
cf_r100
cf_r50_amsu
cf_r50_ninhib
cf_r50_slowrel
cf_r50_notill
cf_r50J70
cf_r50J90
cf_r50_auto
md_r50
Scenario
Continuous Flooding
Continuous Flooding, 100%
Residue Incorporation
Continuous Flooding,
Ammonium Sulphate
Fertilizer
Continuous Flooding,
Nitrification Inhibitor
Fertilizer
Continuous Flooding, Slow
Release Fertilizer
Continuous Flooding, No
Till
Continuous Flooding, 30%
Reduced Fertilizer
Continuous Flooding, 10%
Reduced Fertilizer
Continuous Flooding, Auto-
fertilization to maximize
yields
Mid-season Drainage
Flooding
CF
CF
CF
CF
CF
CF
CF
CF
CF
MD
Residue
Incorporation
50%
100%
50%
50%
50%
50%
50%
50%
50%
50%
Alternative
Management Fertilization
— conventional
— conventional
— ammonium sulfate
— nitrification inhibitor
— slow release
no till conventional
— 30% reduced
— 10% reduced
— Automatically adjusted
by DNDC to maximize
yields
— conventional
(continued)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-33
-------
RICE
Table 2-6: Alternative Rice Management Scenarios Simulated using DNDC (continued)
Abbreviation
md_r100
md_r50_amsu
md_r50_ninhib
md_r50_slowrel
md_r50_notill
md_r50J70
md_r50J90
md_r50_ds
md_r50_auto
awd_r50
awd_r50_ninhib
awd_r50_slowrel
base_r50_ds
base_r50J80_ds
dry_r50
dry_r50J80
Scenario
Mid-season Drainage
w/1 00% Residue
Incorporation
Mid-season Drainage,
Ammonium Sulphate
Fertilizer
Mid-season Drainage,
Nitrification Inhibitor
Fertilizer
Mid-season Drainage, Slow
Release Fertilizer
Mid-season Drainage, No
Till
Mid-season Drainage, 30%
Reduced Fertilizer
Mid-season Drainage, 10%
Reduced Fertilizer
Mid-season Drainage, Dry
Seeding
Mid-season Drainage,
Auto-fertilization to
maximize yields
Alternate Wetting & Drying
(AWD)
AWD w/Nitrification
Inhibitor
AWD w/Slow Release
Dry Seeding
Dry Seeding, 20%
Reduced Fertilizer
Dryland Rice
Dryland Rice, 20%
Reduced Fertilizer
Residue
Flooding Incorporation
MD
MD
MD
MD
MD
MD
MD
MDw/DS
MD
AWD
AWD
AWD
DS
DS
dryland rice
dryland rice
100%
50%
50%
50%
50%
50%
50%
50%
50%
50%
50%
50%
50%
50%
50%
50%
Management Fertilization
— conventional
— ammonium sulfate
— nitrification inhibitor
— slow release
no till conventional
— 30% reduced
— 10% reduced
— conventional
— Automatically adjusted
by DNDC to maximize
yields
— conventional
— nitrification inhibitor
— slow release
— conventional
— 20% reduced
— conventional
— 20% reduced
For non-irrigated rice, there is no difference between scenarios with alternative water management.
Thus, we refer to those scenarios for the non-irrigated rice with "base_" in front rather than "cf" or "md".
V-34
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
RICE
Table 2-7: Rice Management Techniques
Management
Technique
Rice flooding
Continuous
Flooding (CF)
Description
rice paddy is flooded on planting date and drained 10 days prior to harvest date - applies to both
irrigated and rainfed rice
Mid-season
drainage (MD)
rice paddy is drained twice during growing season for 8 days - final drainage is 10 days prior to
harvest date - applies only to irrigated rice
Alternate wetting rice paddy is initially flooded to 10 cm - water level is reduced at rate of -0.5 cm/day till to -5cm and
and drying (AWD) then reflooded at rate of 0.5 cm/day till to 10 cm - applies only to irrigated rice
Dryland rice
I irrigated and rainfed rice are swapped for dryland rice - no flooding occurs
Rice seeding
Direct seeding (DS) rice paddy is flooded 40 days after planting date and drained 10 days prior to harvest date - applies to
both irrigated and rainfed rice
Residue incorporation
50
-------
RICE
with conventional fertilizer (md_r50) rather than being compared to the baseline weighted average
emissions per ha. This is done to better represent the mitigation potential from adopting each mitigation
option on each baseline subcategory. As an example, an option such as cf_r50_amsu may result in
emissions reductions relative to cf_r50 but increases in emissions relative to md_r50 (and possibly the
weighted baseline emissions as well) in many countries. This is resulting from the change in water
management regime in moving from mid-season drainage to continuous flooding, whereas we are trying
to isolate the effects of changing fertilizer for a given baseline water management strategy in that
example.
• Capital Cost: None of the options were assumed to have any capital cost.
• Annual Operation and Maintenance (O&M) Cost: Changes in labor, fertilizer, and other inputs
associated with each option.
• Annual Benefits: Calculated based on changes in production associated with changes in yield,
valued at market prices.
• Applicability: All options applicable for a given cropping pattern were assumed available to all
acres in all countries. However, water management options (e.g., shifting from continuous
flooding to midseason drainage, etc.) are only applicable to irrigated systems. No water
management options are available for rainfed, deepwater, or upland rice
• Technical Efficiency: Determined by the DNDC Model for each country, production type, and
water management combination for each mitigation option.
• Technical Lifetime: Indefinite; there are no capital costs being included for which a lifetime must
be defined.
V.2.4 Marginal Abatement Costs Analysis
The MAC analysis assimilates the abatement measures' technology costs, expected benefits, and
emission reductions presented in Section X.3 to compute the cost of abatement for each measure. Similar
to the approach used in other non-CCfe sectors of this report, we compute a break-even price for each
abatement option for 195 countries to construct MAC curves illustrating the technical, net GHG
mitigation potential at specific break-even prices for 2010, 2020, and 2030.
V.2.4.1 MAC Analysis Results
The MAC analysis of the mitigation options described above suggests that net GHG abatement
potential for global paddy rice cultivation equates to approximately 6 percent of its total annual
emissions between 2010 and 2030 at a carbon price of $5 per ton of CCh equivalent ($/tCChe). In 2030, total
abatement potential in the sector is 21 MtCChe at no carbon price, 57 MtCChe at a carbon price of
$5/tCO2e, and 124 MtCChe at a carbon price of $20/tCO2e, representing 2%, 6% and 12% of the net GHG
emissions in the year, respectively. Figure 2-4 presents the MAC curves for the global rice cultivation, in
2010, 2020 and 2030. The estimated net GHG mitigation potential at various break-even prices for the top-
emitting countries and aggregate regions comprising the rest of the globe are presented in Table 2-8.
V-36 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
RICE
Table 2-8:
Abatement Potential by Region at Selected Break-Even Prices in 2030 (MtC02e)
Country/
Region
-10
I
•
mm
M
wsm
!•
El
100
100+
Top 5 Emitting Countries
India
Indonesia
Bangladesh
Vietnam
China
2.4
6.0
2.8
0.0
0.6
2.4
9.1
3.4
0.0
1.6
5.5
12.8
19.5
6.9
3.2
14.5
14.4
30.4
9.0
3.5
15.1
16.3
30.4
9.8
9.5
16.8
19.1
30.5
13.2
10.0
16.8
19.1
30.5
16.0
10.6
16.8
19.1
31.9
16.0
10.6
20.4
21.8
33.1
16.0
12.6
28.8
24.8
35.6
17.6
19.1
34.5
25.6
35.9
21.6
23.7
Rest of Region
Africa
Asia
Central &
South
America
Eurasia
Europe
Middle East
North
America
World Total
0.1
2.1
0.4
0.0
0.0
0.0
0.7
15.2
0.3
2.7
0.6
0.0
0.0
0.1
0.7
20.9
0.8
6.9
1.4
0.0
0.1
0.1
0.8
57.8
1.2
9.2
3.5
0.0
0.2
0.1
1.0
87.0
1.6
14.7
4.5
0.0
0.2
0.3
1.5
103.9
2.7
16.6
6.3
0.1
0.3
0.3
1.7
117.5
3.6
21.1
7.3
0.1
0.3
0.4
1.9
127.8
4.1
25.5
8.1
0.1
0.3
0.5
2.2
135.4
5.1
28.2
9.5
0.2
0.4
0.8
2.8
150.9
5.4
31.3
10.9
0.3
0.7
1.1
3.4
178.9
5.7
34.9
12.1
0.4
1.0
1.2
3.7
200.3
Mitigation potential and its cost-effectiveness vary significantly by country or region. At the regional
level, Asia (in particular South and Southeast Asia), Africa, Central and South America and the European
Union show the most significant potential for reducing GHG emissions from rice cultivation. For
instance, in 2030 mitigation potential in Asia is estimated to be 27 MtCChe with no carbon price and 34
MtCChe at a carbon price of $20/tCChe. Central and South America can achieve mitigation potential of 12
MtCChe in 2030 at no carbon price, and mitigation potential can increase to 22 MtCChe at a carbon price
of $20/tCChe. Figure 2-4 shows the MAC curves for the top-five emitting countries in 2030.
There are a large number of mitigation options included for rice cultivation and almost all provide
net GHG reductions. The overall distribution of GHG mitigation across mitigation options included in
this analysis is presented in Table 2-9. The options providing the largest quantify of GHG reductions are
the two that involve switching to dryland production, which significantly reduces or eliminates CH4
emissions. Those options do result in major reductions in yields, however. Other options that account for
large reductions include nitrification inhibitors in combination with midseason drainage or alternate
wetting and drying, along with switching to no-till, fertilizer reductions, and optimal fertilization options
on non-irrigated lands. The relative share of mitigation provided by different options varies across years
due to the dynamics of GHG emissions, especially for changes in soil C.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-37
-------
RICE
Table 2-9: Distribution of Net GHG Reductions across Miti
ation Options, Baseline Production Case
wn
base_r100
base_r50_amsu
base_r50_ninhib
base_r50_slowrel
base_r50_notill
base_r50_f70
base_r50J80
base_r50J90
base_r50_auto
base_r50_ds
base_r50J80_ds
cf_r100
cf_r50_amsu
cf_r50_ninhib
cf_r50_slowrel
cf_r50_notill
cf_r50J70
cf_r50J80
cf_r50J90
cf_r50_auto
md_r50
md_r100
md_r50_amsu
md_r50_ninhib
md_r50_slowrel
md_r50_notill
md_r50J70
md_r50J80
md_r50J90
md_r50_auto
md_r50_ds
awd_r50
awd_r50_ninhib
awd_r50_slowrel
dry_r50
dry_r50J80
TOTAL
1.74
2.23
4.86
1.37
4.42
6.59
4.49
2.30
5.89
0.95
1.01
0.11
1.26
2.22
2.24
0.04
0.46
0.34
0.19
0.40
5.08
6.36
6.47
18.32
7.93
3.12
6.14
5.98
5.61
4.80
1.34
5.27
19.70
8.41
25.35
25.74
198.73
0.35
1.86
4.38
0.36
15.39
12.77
8.81
4.52
10.55
0.61
0.66
0.00
1.50
2.57
1.96
0.01
0.47
0.35
0.19
0.23
5.52
3.76
6.97
20.02
7.43
3.10
6.66
6.49
6.10
5.13
1.01
4.85
19.11
7.53
15.00
17.00
203.23
0.36
1.57
4.64
0.27
17.84
13.12
8.96
4.57
11.46
0.57
0.62
0.00
1.51
2.61
1.87
0.01
0.46
0.35
0.19
0.18
5.59
3.75
6.92
20.40
7.48
2.93
7.07
6.80
6.30
5.33
1.01
4.31
17.95
7.08
13.23
13.00
200.33
V-38
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
RICE
Figure 2-4: Marginal Abatement Cost Curve for Top 5 Emitters in 2030, Baseline Production Case
.0 15.0 20.0 25.0 30.0 35.0
•India
China
•Indonesia
•Vietnam
Bangladesh
Non-CO2 Reduction (MtCO2e)
V.2.5
Sensitivity Analyses
In this section, we explore sensitivity analyses to examine the potential effects of alternative
assumptions on estimated mitigation potential. Because many of the mitigation options simulated impact
rice yields, the assumption of constant production implies a change in the area devoted to rice
production. There are options that increase productivity, but also many that decrease productivity. Thus,
land requirements may increase or decrease to maintain production at baseline levels, but overall the
package of mitigation options being considered tends to reduce yields. In this sensitivity analysis, we
hold the area of cultivated rice at the baseline area and recalculate the MACs.
Baseline Acreage
This section explores this relationship further by presenting an alternative scenario built around a
constraint on the number of acres, keeping the harvested area the same as estimated in the baseline.
As before, the MAC model only includes options that result in lower emissions. The result for area
held fixed at projected baseline area is shown in Figure 2-5. Generally speaking, emissions and emission
reduction potential are slightly higher although the effects vary by country. Overall, global maximum
potential mitigation is 320 MtCO2e, 60% higher than the global maximum potential mitigation of
200MtCO2e in the constant production case. Figure 2-6 shows the MAC for the top 5 rice producing
countries under the constant area case. China's MAC shows relatively little change under the assumption
of constant area, but the other countries show increased emissions mitigation potential to varying
degrees.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-39
-------
RICE
Figure 2-5: Marginal Abatement Cost Curve, Baseline Area Case
•2010
•2020
•2030
250
Non-CO2 Reduction (MtCO2e)
Figure 2-6: Marginal Abatement Cost Curve for Top 5 Emitters in 2030, Baseline Area Case
10.0 15.0 20.0 25.0 30.0 35.0
Non-CO2 Reduction (MtCO2e)
•India
China
•Indonesia
•Vietnam
Bangladesh
V-40
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
RICE
V.2.5. Uncertainties and Limitations
Given the complexities of the global rice sector, the estimated GHG mitigation potential and marginal
abatement cost curves are subject to a number of uncertainties and limitations:
• Availability and quality of data to represent the highly complex and heterogeneous rice
production systems of the world. Although there are major improvements in representing the
global rice production systems and the business-as-usual baseline conditions compared to the
previous EPA report (USEPA, 2006), data in some areas, such as management practices, are not
always available for all countries or regions and approximations must be made based on limited
literature or expert judgment. Moreover, collecting and developing consistent cost estimates of
emerging and/or not widely adopted mitigation measures is challenging.
• Biophysical modeling uncertainties, in particular with respect to soil organic carbon
simulations. The DNDC modeling of the business-as-usual baseline conditions and mitigation
scenarios was performed using a set of inputs and assumptions developed based on various
sources. The quality of input data ultimately affects the simulated results. Soil organic carbon,
which has a significant impact on the net GHG emissions from the sector, is particularly
challenging to simulate given the lack of monitoring data at the global scale. Sensitivity tests
would be useful to assess how alternative modeling approaches and assumptions may influence
modeling results.
• Optimistic assumptions on technology adoption. The analysis assumes that if mitigation
technology is considered feasible in a country or region, it is fully adopted in 2010 and through
the analysis period. Research suggests that adoption of new technology in the agricultural sector
is a gradual process and various factors potentially slow the adoption of a new GHG-mitigating
technology (e.g., farm characteristics, access to information and capital, and cultural and
institutional conditions). The mitigation potential presented in this analysis should be viewed to
represent the technical potential of the mitigation options analyzed.
• Potential interactions of multiple mitigation measures are not fully addressed in this analysis.
In this analysis, mitigation options are applied to independent segments of the rice production
systems to avoid double counting. In reality, multiple mitigation options can be applied, and
their order of adoption and potential interactions may affect the aggregate GHG mitigation
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-41
-------
RICE
Bates, ]., N. Brophy, M. Harfoot and J. Webb. (2009). Agriculture: methane and nitrous oxide. Sectoral
Emission Reduction Potentials and Economic Costs for Climate Change (SERPEC-CC). pp. 62.
Nelson, G.C., M.W. Rosegrant, A. Palazzo, I. Gray, C. Ingersoll, R. Robertson, S. Tokgoz, T. Zhu, T.B.
Sulser, C. Ringler, S. Msangi, and L. You. (2010). Food Security, Farming, and Climate Change to
2050: Scenarios, Results, Policy Options. International Food Policy Research Institute: Washington,
DC.
Neue, H. 1993. Methane emission from rice fields: Wetland rice fields may make a major contribution to
global warming. BioScience 43 (7): 466-73.
U.S. Environmental Protection Agency (USEPA) (2006). Global Mitigation of Non-CO2 Greenhouse
Gases. EPA 430-R-06-005. Washington DC.
USEPA (2012). Global Anthropogenic Non-CO2 Greenhouse Gas Emissions: 1990 - 2030. EPA 430-D-ll-
003. Washington DC.
World Resources Insititute (WRI). (2012). Climate Analysis Indicators Tool (CAIT) database. Available at
http://www.wri.org/proiect/cait/.
V-42 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LIVESTOCK
V.3. Livestock
V.3.1
Sector Summary
ivestock operations generate methane (Q-k) and nitrous oxide (N2O) emissions. The
(greenhouse gas (GHG) emissions mainly come from two sources, enteric fermentation and
manure management. Methane is produced as a by-product of the digestive process in animals
through a microbial fermentation process. The quantity of enteric fermentation Q-k emissions is
determined by the animal's digestive system, diet and management practices. Livestock manure
management can produce both Q-k and N2O emissions. Methane is produced when manure decomposes
under anaerobic conditions. The quantity of manure CH4 emissions is determined by the type of
treatment or storage facility, the ambient climate, and the composition of the manure. Manure N2O
emissions result from nitrification and denitrification of the nitrogen that is excreted in manure and urine.
In 2010, the global non-CCh GHG emissions from livestock operations were approximately 2,286
MtCCfee. Figure 3-1 presents projected total emissions for the top 5 emitting countries and the total for the
rest of the world.
Methane emissions predominate with 2,152 MtCChe emitted in 2010. Globally, livestock is the largest
source of Q-k emissions, contributing approximately 29% of total global Q-k emissions in 2010. As
shown in Figure 3-2, the top 5 emitting countries - India, China, Brazil, the United States, and Pakistan -
accounted for 44% of the sector's total CH4 emissions. Growth in CH4 emissions is expected to be about
20% between 2010 and 2030.
Figure 3-1: Total Net GHG Emissions and Projections for the Livestock Sector: 2000-2030
2,729
I Pakistan
I United States
Brazil
I China
India
ROW
2000
2010
2020
2030
Year
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-43
-------
LIVESTOCK
Figure 3-2: Cm Emissions Projections for the Livestock Sector: 2010-2030
1600
1400
1200
a. 1000
rf
8 800
^ 600
400
200
0 -
(D := (D « c >
^ N c a; ns >
~ IB u 55 15 DC
(D
4-*
'c
2010
(D := (D « c >
^ N c OJ (D >
~ M u 55 15 DC
(D
4-*
'c
2020
III!.
(D := (D « c >
^ N c OJ (D >
~ M u 55 15 DC
(D
4-*
'c
2030
• Manure CH4
• Enteric CH4
Nitrous oxide emissions from manure management are a second significant source of GHG emissions
within the livestock sector, contributing an additional 135 MtCChe. Livestock contributed approximately
4% of total global N2O emissions in 2010. As presented in Figure 3-3, China, India, the United States,
Brazil, and Pakistan together account for 63% of global N2O emissions from livestock operations in 2010.
N20 emissions are expected to grow about 16% between 2010 and 2030 to about 156 MtCChe, slightly
lower than the projected growth in Q-k emissions over the same time period.
Figure 3-3: M Emissions Projections from the Livestock Sector: 2010-2030
I India
I China
Brazil
I United States
I Pakistan
ROW
2010
2020
2030
V-44
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LIVESTOCK
Marginal abatement cost (MAC) curve results assuming the production of livestock products
remains constant at projected baseline levels are presented in Figure 3-4. Maximum abatement potential
in the livestock sector is 268 MtCChe in 2030, or about 9.8% of total GHG emissions in that year.1 These
results suggest that there are significant opportunities for GHG reductions in the livestock sector.
Approximately 86 MtCChe can be reduced in 2030 at no or low carbon prices below $5 per ton of CCh
equivalent.
Figure 3-4: Global Abatement Potential in Livestock Management: 2010, 2020, and 2030
$60
$50
$40
$30
« $20
8
$10
$0
-$10
-$20
-$30
•2010
2020
•2030
250
Non-CO2 Reduction (MtCO2e)
The following section offers a brief description of Q-k and N2O emissions from livestock operations,
and a discussion of projected trends in global baseline emissions. The subsequent section presents
possible abatement technologies, their technical specifications, costs and potential benefits. The final
section discusses the MAC analysis and estimated abatement potential at global and regional levels.
1 This analysis only assesses abatement measures that are designed to reduce QHk emissions. Mitigation options that
focus on potential reductions in N2O emissions are not included due to relatively small potential abatement potential
and limited information on abatement measures and costs. However, N2O emissions are affected by changes in
livestock productivity under our primary scenario with production held constant because the number of animals
required to produce a given quantity of livestock products, and their associated emissions, changes with
productivity.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-45
-------
LIVESTOCK
V.3.2 CH4 and N2O Emissions from Livestock Management
This section discusses how Q-k and N2O emissions are produced in livestock operations and the
current projections of baseline emissions between 2010 and 2030.
V.3.2.1 CH4 Emissions from Enteric Fermentation
Enteric fermentation produced about 1945 MtCChe of Q-k in 2010 and accounts for about 90% of the
total CH4 emissions from livestock. Methane is produced as a by-product of the digestive process in
animals. This microbial fermentation process produces Q-k that can be exhaled or excreted by the animal.
The quantity of Q-k produced through enteric fermentation depends largely on the animal's digestive
system, diet and management practices. Ruminant animals (e.g., cattle, buffalo, sheep, goats, and camels)
are the major sources of enteric Q-k emissions; nonruminant animals (e.g., swine, horses, mules) also
produce enteric Q-k emissions but at much lower rates compared to ruminant animals.
The quantity, quality and digestibility of feed significantly affect enteric Q-k emissions. The main
constituents of the diet - sugars, starch, fiber, protein and lipid - appear to have varying impacts on
methane emissions. In general, increased intake of starch and soluble sugars decreases rumen pH, which
suppresses methanogens, thus resulting in lower Q-k emissions. Lower feed quality such as higher
content of insoluble fiber leads to higher Q-k emissions. Provision of feed supplements, such as dietary
oils, is found to have an inhibitory effect on Q-k production in the rumen (Hristov et al., 2013).
Management practices that improve animal productivity, such as the usage of antibiotics and bovine
somatotropin (bST), often reduce Q-k emissions per unit of meat or milk even though these activities can
increase Q-k emissions per animal.
V.3.2.2 CH4 and N2O Emissions from Manure Management
Manure Management CH4 Emissions
Manure management produced about 206 MtCChe of Q-k in 2010, smaller than enteric fermentation
but still a significant global source of Q-k at about 3% of global total methane production.2 In livestock
waste management systems, Q-k is produced when manure decomposes under anaerobic conditions, for
example in lagoons, ponds or pits. The quantity of Q-k emitted from manure management operations is
determined by the type of treatment or storage facility, the ambient climate, and the composition of the
manure (USEPA, 2012). Higher ambient temperature and moisture conditions favor Q-k production.
Manure Management N2O Emissions
In addition to Q-k, livestock waste management produced about 135 MtCChe of N2O in 2010. Nitrous
oxide is produced from livestock waste through nitrification and denitrification. Nitrous oxide emissions
from livestock waste depend on the composition of the waste, the type of bacteria involved in the
decomposition process, and the oxygen and liquid content of waste (USEPA 2012). Nitrous oxide
generation is most likely to occur in dry manure handling systems.
2 Global CH4 emissions in 2010 totaled 7,549.2 MTCO2e (USEPA, 2012, Table A2)
V-46 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LIVESTOCK
V.3.2.3 Baseline CH4 and N2O Emissions Estimates
This section discusses the historical and projected baseline emissions for the livestock sector.
Historical emissions are characterized as those released between 1990 and 2005. Projected emissions cover
the 20-year period 2010 - 2030.3
Historical Emissions Estimates
Over the 1990 - 2005 period, total non-CCh GHG emissions from livestock operations increased by 4%
between 1990 and 2005, from 2,201 to 2,292 MtCChe (USEPA, 2012). This modest growth is caused by two
opposing trends: growth in Africa and Central and South America has been partially offset by the effects
of market restructuring in non-OECD Europe. Enteric fermentation Q-k emissions increased 7% between
1990 and 2005 while emissions of Q-k and N20 from livestock waste management decreased 9% between
1990 and 2005.
Projected Emissions Estimates
This analysis uses the 2005 country-level livestock population data from the Global Anthropogenic
Non-CCh Emissions Report ("GER") as a starting point (USEPA, 2012). However, for the period 2010 —
2030 an alternate business-as-usual forecast was constructed using livestock production and market price
projections generated by the International Food Policy Research Institute (IFPRI)'s International Model
for Policy Analysis of Agricultural Commodities and Trade (IMPACT) (Nelson et al., 2010) to derive
projected livestock populations. A key rationale for relying directly on these model outputs is that the
IFPRI IMPACT model projections provide a set of prices and global production patterns consistent with
their livestock population and productivity assumptions. Using these data improves the internal
consistency of the MAC analysis.4
Table 3-1 shows projected baseline emissions from livestock management for the top 5 emitting
countries and the rest of the world, divided into major regions.5 Global emissions from livestock
management are projected to grow at an average annual rate of 0.9%. In general, emissions are growing
much more rapidly in developing countries than in the developed world.
3 The year 2010, although historical, is the first year of the modeling forecast period.
4 The IMPACT outputs separated the world into 116 regions, with larger countries defined individually and smaller
countries combined into regions. A mapping was created between IMPACT regions and the 195 countries in this
analysis, using shares of country-level livestock population in 2010 based on USEPA (2012) to disaggregate regional
projections from the IMPACT model to individual countries within each region.
5 Regional totals exclude the top 5 emitting countries that are presented separately in the table.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-47
-------
LIVESTOCK
Table 3-1: Projected Baseline Emissions from Livestock Management: 2010-2030 (MtC02e)
Country
2010
2015
2020
2025
2030
CAGR
(2010-2030)
Top 5 Emitting Countries
India
China
Brazil
United States
Pakistan
300
242
235
174
80
311
253
247
179
89
322
262
248
181
99
333
271
247
184
110
344
278
246
186
122
0.7%
0.7%
0.2%
0.3%
2.1%
Rest of Regions
Asia
Africa
Europe
Middle East
Central & South America
Eurasia
North America
World Totals
259
293
257
28
227
118
74
2,286
283
320
257
30
245
120
77
2,411
307
343
257
32
258
121
80
2,512
335
369
257
35
271
124
83
2,619
367
395
257
38
284
126
85
2,729
1.8%
1.5%
0.0%
1.6%
1.1%
0.3%
0.8%
0.9%
Table 3-2 summarizes projected baseline emissions from enteric fermentation. Worldwide
emissions from enteric fermentation are projected to increase at an average annual rate of 0.9% between
2010 and 2030. The top five countries, India, Brazil, China, the United States, and Pakistan, combine for
about 44% of global totals in 2010, but the baseline projection has emissions from all of these countries
except Pakistan growing at a slower rate than the global average. Annualized growth rates in the top five
countries average 0.7%, lower than the average 0.9% growth projected in the rest of regions. By 2030, the
top five countries are the source of 42% of global enteric fermentation emissions.
Table 3-2: Projected Baseline Emissions from Enteric Fermentation: 2010-2030 (MtCC^e)
Country
2010
2015
2020
2025
2030
CAGR
(2010-2030)
Top 5 Emitting Countries
India
Brazil
China
United States
Pakistan
265
225
162
132
73
274
236
172
136
81
283
237
179
138
90
293
236
186
141
100
301
234
191
143
111
0.7%
0.2%
0.8%
0.4%
2.1%
(continued)
V-48
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LIVESTOCK
Table 3-2: Projected Baseline Emissions from Enteric Fermentation: 2010-2030 (MtC02e) (continued)
Country
2010
2015
2020
2025
2030
CAGR
(2010-2030)
Rest of Regions
Asia
Africa
Europe
Middle East
Central & South America
Eurasia
North America
World Totals
211
277
195
26
218
97
64
1,945
231
302
196
28
235
99
68
2,059
251
325
197
30
248
101
71
2,150
275
349
198
33
261
103
73
2,246
303
374
198
36
272
105
76
2,345
1.8%
1.5%
0.1%
1.6%
1.1%
0.4%
0.8%
0.9%
Similarly, worldwide emissions from manure management are projected to increase at an average
annual rate of 0.6% between 2010 and 2030, but that world average combines slower growth in the top-
emitting countries with faster growth in the rest of regions. In 2010, the top five countries combine to
account for 51% of global emissions from manure management. By 2030, these same five are projected to
account for just under 50% of global emissions, equivalent to annual growth of 0.4%. In the rest of
regions, global emissions grow at an average annual growth rate of 0.8%.
Table 3-3: Projected Baseline Emissions from Manure Management: 2010-2030 (MtCC^e)
•^^^^^1
Jj^^ffij^B
2010
2015
2020
2025
(2010-2030)
Top 5 Emitting Countries
China
United States
India
Brazil
France
79
43
35
10
8
81
43
37
10
8
83
43
39
11
8
85
43
40
11
8
87
43
42
12
8
0.5%
0.0%
0.9%
0.8%
-0.4%
Rest of Regions
Asia
Africa
Europe
Middle East
Central & South America
Eurasia
North America
World Totals
56
16
53
2
9
21
9
341
60
17
53
2
10
21
9
352
65
19
52
2
10
21
10
362
70
20
52
2
11
21
10
373
76
21
52
2
11
21
10
384
1.5%
1.4%
-0.1%
1.3%
1.2%
0.1%
0.3%
0.6%
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-49
-------
LIVESTOCK
V.3.3 Abatement Measures and Engineering Cost Analysis
A significant number of livestock GHG mitigation measures can be identified in the literature (e.g.,
Hristov et al., 2013; Archibeque et al. 2012; UNFCCC 2008, Whittle et al, 2013). However, developing
consistent and regional-specific cost estimates for emerging mitigation measures or options that are not
widely adopted has proven a challenging task. The measure cost data are scarce and often reflect
anecdotal experience reported in a specific country, region or livestock production system. Assumptions
have to be made to extrapolate the estimates in other countries, regions and production systems. This
review uncovered only a few studies where cost information was presented in addition to associated
emission reductions for a number of mitigation measures. Moreover, for some mitigation measures, such
as those that potentially reduce livestock enteric fermentation Q-k emissions, the literature varies on the
estimated magnitude of emissions reductions as well as the long-term mitigation effects and animal and
human health impacts.
Based on the availability and quality of mitigation measure cost and emission reduction efficiency
information, this analysis evaluates six mitigation options for enteric fermentation Q-k emissions and ten
options for manure management Q-k emissions. Each technology is briefly characterized followed by a
discussion of abatement measures' implementation costs, potential benefits, and system design
assumptions used in the MAC analysis.
V.3.3.1 Enteric Fermentation CH4 Mitigation Technologies
This section characterizes the mitigation technologies that can be applied to reduce enteric Q-k
emissions. Many of the currently available enteric fermentation mitigation options, summarized in Table
3-4, work indirectly by increasing animal growth rates and reducing time-to-finish (or increasing milk
production for dairy cows). The potential GHG mitigation estimated here depends on the assumption
that total production of meat or milk remains the same as in the baseline. Simply put, these strategies
work because increased productivity means fewer animals are required to produce the same amount of
meat or milk, and fewer animals mean reduced GHGs.
Unfortunately, many of the productivity enhancing options in this group are not without controversy
(Hristov et al., 2013; Grainger et al., 2010). Some, such as bST and antibiotics, have raised concerns
outside than their potential role in reducing GHGs. Most have greater than usual uncertainty about costs
and effectiveness, especially under long term use. For example, Whittier et al. (2013), in developing MAC
curves for Australia, assume that feed supplements (analogous to Improved Feed Conversion here) and
antimethanogen vaccines will become available by 2020 for some types of livestock operations. However,
ICF international, writing in a report prepared for the USD A, provides only a qualitative description of
enteric fermentation GHG mitigation options, excluding them from cost or break-even analysis because
"more research is needed to evaluate the potential GHG impacts of changes in diets, use of feed
additives, and breeding (ICF International, 2013, p 3.62)."
In what follows we present descriptions and economic information used to derive the MAC curves.
We examine the sensitivity of these results to productivity assumptions in Section V.3.5 which replaces
the assumption of constant production with an assumption of constant animal population and also
examines a no antimethanogen case.
V-50 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LIVESTOCK
Table 3-4: Abatement Measures for Enteric Fermentation
Abatement
Option
otal Installed
Capital Cost
(2010 USD)
(2010 USD)
Capital
Lifetime
(Years)
Reduction
Efficiency
(change in
emissions per
head)
Benefits
(Changes in
Livestock or
Energy
Revenue)
Improved Feed
Conversion
Antibiotics
bST
Propionate
Precursors
Antimethanogen
Intensive Grazing
0
0
0
0
0
0
25-295 per head
4-9 per head
123-300 per
head
40-1 20 per head
9-33 per head
-1 80 to +1 per
head
NA
NA
NA
NA
NA
NA
CH4: -39.4% to
+39.6%
CH4: -0.4% to -
6%
CH4: -0.2% to
+10.3%
CH4:-10%beef
cattle and sheep;
-25% dairy
animals
CH4:-10%
CH4: -13.3% beef
cattle; -15.5%
dairy cattle
0-79% increase
in animal yield
5% increase in
animal yield
12. 5% increase in
animal yield
5% increase in
animal yield
5% increase in
animal yield
-11. 2% reduction
in dairy cattle
yield
Improved Feed Conversion
This mitigation measure encompasses a number of management practices that would improve the
proportion of feed energy converted to final products. The practices include increased amount of grain
fed to livestock, and inclusion of dietary additives. This option is more effective in reducing emissions in
regions where baseline feed is of relatively low quality.
• Annual Cost: Typical annual costs for improving feed are between $2 and $295 per head for beef
and dairy cattle. No data were identified for other species. One of the primary costs for this
option, as well as most of the others below, is for additional labor costs necessary for
implementation. Differences in labor input share and labor costs per hour are also major reasons
for the wide variation in costs between regions and livestock production systems.
• Annual Benefits: Ration improvements result in an increase in yield (kg of meat or milk per
animal) between 0 and 79%. There is considerable variation in the productivity impacts,
primarily related to differences in baseline feed quality and productivity. Livestock raised in
countries with low quality feeds in the baseline tend to have much greater productivity benefits
from improved feed conversion than those in developed countries where feed conversion is
already highly efficient.
• Applicability: This option applies to beef and dairy cattle in areas where baseline livestock
growth rates and milk production are low, primarily developing regions including Africa. This
option is assumed to be available only for urban livestock production or intensively managed
livestock production and only applied in regions where the yield gains associated with the option
are greater than baseline yield increases (typically limited to regions that do not already feed
mixed rations).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-51
-------
LIVESTOCK
• Technical Efficiency: This analysis assumes a change in emissions per head between -39% and
+40%. Cases with increased emissions are excluded from the MAC analysis.6
Antibiotics
Feed antibiotics (e.g., monensin) to promote increased weight gain and reduce feed intake per metric
ton of meat produced.
• Annual Cost: Typical annual costs for providing antibiotics are between $4 and $9 per head for
beef cattle including the costs of antibiotics and increased labor costs for implementation. No
data exist for other species.
• Annual Benefits: Ration improvements result in an increase in yield of 5% kg/animal
• Applicability: This option applies to beef cattle in all regions, but is restricted to urban livestock
production and intensively managed livestock production.
• Technical Efficiency: This analysis assumes a reduction in emissions per head between 0% and
6%.
Bovine Somatrotropin (bST)
This measure administers bST to dairy cattle to increase milk production. Because of opposition to
the use of growth hormones like bST in many countries, this option was only applied in a subset of
countries.
• Annual Cost: Typical annual costs for purchasing and administering bST were estimated to be
between $123 and $300 per head for dairy cows. This cost is based on the cost of purchasing bST
and the additional labor costs required for administering.
• Annual Benefits: Using bST results in an average annual increase in yield (kg milk per head) of
12.5%
• Applicability: This option applies to dairy cows in all countries that currently approve the use of
bST or are likely to do so in the near future. This option is assumed to be available only for urban
or intensively managed livestock production.
• Technical Efficiency: This analysis assumes a reduction in emissions per head between 0% and
6%.
Propionate Precursors
This option involves administering propionate precursors (malate, fumarate) to animals on a daily
basis. Hydrogen produced in the rumen through fermentation can react to produce either Q-k or
propionate. By adding propionate precursors to animal feed, more hydrogen is used to produce
propionate and less Q-k is produced.
• Annual Cost: Typical annual costs for purchasing and administering propionate precursors are
between $40 and $120 per head for beef cattle, sheep, and dairy animals.
• Annual Benefits: Administering propionate precursors results in an increase in yield (kg of meat
or milk per animal) of 5%.
6 For the primary scenario where production is held constant, options that increase emissions per unit of output are
excluded from the MAC calculations. Thus, mitigation options that increase emissions per head are still included in
the MAC calculations if they increase productivity more than then they increase emissions, resulting in a reduction in
emissions intensity per unit of output.
V-52 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LIVESTOCK
• Applicability: This option applies to beef cattle, sheep, and dairy animals in all regions.
However, as with other options, it is only applied in urban and intensive livestock production
systems.
• Technical Efficiency: This analysis assumes a reduction in Q-k emissions per head of 10% for
beef cattle and sheep and a reduction of 25% for dairy animals.
Antimethanogen
Antimethanogen is a vaccine that can be administered to animals to suppress Q-k production in the
rumen. The vaccine is currently in infancy of development with limited information on emission
reduction efficiency and long-term mitigation effects and animal health impacts.
• Annual Cost: Typical annual costs for providing antimethanogens are between $9 and $33 per
head for purchasing and administering antimethanogens.
• Annual Benefits: Increases yields by 5% as more of the energy contained is feed is used by the
animals to produce for meat or milk rather than producing methane
• Applicability: This option applies to all ruminants in all regions, though again it is assumed that
only urban and intensively managed systems can adopt this option.
• Technical Efficiency: This analysis assumes a reduction in emissions per head of 10%.
Intensive Grazing
Improving nutrition through more intensive pasture management and cattle rotations to allow for
regrowth while decreasing reliance upon prepared rations.
• Annual Cost: Estimated reduction in yield of 11.2% for dairy cattle. Beef yields are assumed to
remain unchanged under this option.
• Annual Benefits: Estimated annual cost savings of between $0 and $180 per head for reduced
expenditures on feed.
• Applicability: This option applies only to beef and dairy cattle in developed regions and Latin
America. It was assumed to be available only in intensively managed systems within livestock
production system categories that receive relatively large amounts of annual rainfall such that
intensive grazing is feasible.
• Technical Efficiency: This analysis assumes a reduction in emissions per head of about 13-15%.
V.3.3.2 Manure Management CH4 Mitigation Technologies
Mitigation options for reducing Q-k from livestock manure focus on changes in manure management
practices that capture the Q-k to flare or use for energy production (see Table 3-5). There are fewer
options for reducing N2O emissions from manure because these emissions tend to result from
decomposition under aerobic conditions, such as from pasture, range, and paddock where manure is
much less concentrated and more difficult to manage.
This analysis includes both large capital-intensive digesters applied in developed regions and small-
scale digesters for developing regions. Revenues are generated from the use of captured Q-k for either
heat or electricity on the farm; these revenues are scaled to other regions based on an electricity price
index. Capital costs and O&M costs for digester systems are mainly based on the USEPA AgSTAR
program data and experience in the U.S. and the developing countries (USEPA, 2010; Roos, personal
communication 2012; Costa, personal communication 2012), supplemented by information from USDA
(2007, 2011). For the EU, technology cost and performance parameters are based on Bates et al. (2009). For
developing countries, the U.S. technology cost data are assumed for large digester systems with
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-53
-------
LIVESTOCK
adjustments made to represent O&M costs in the developing countries. Capital costs for small-scale
systems are based on USEPA (2006), which estimates the capital cost per 1,000 pounds liveweight.
Because liveweight tends to be much smaller in developing countries, the capital cost per animal
generally ends up being lower than in developed regions.
Table 3-5: Abatement Measures for Manure Management
Total
Installed Annual O&M
Capital Cost Cost Capital
Abatement Lifetime
Option (2010 USD) (2010 USD) (Years)
Reduction Benefits
Efficiency (Changes in
(change in Livestock or
[change in Livestock or Adjustments
emissions Energy Technical Across
per head) Revenue) Applicability Regions
Complete-mix Digester, Hogs
With Engine
Without
Engine
100 per head
(US)
61 per head
(US)
0.11 per head 20
(US)
0.07 per head 20
(US)
CH4: -85% $8 energy
revenue/
savings per
head (US)
CH4: -85% none
Hogs in
selected IPS
and
management
intensities
Hogs in
selected IPS
and
management
intensities
Labor costs,
labor share,
energy prices
Labor costs,
labor share
Complete-mix Digester, Dairy Cattle
With Engine 958 per head
(US)
Without
Engine
588 per head
(US)
3.35 per head 20
(US)
2.06 per head 20
(US)
CH4: -85% $65 energy
revenue/
savings per
head (US)
CH4: -85% none
Dairy cattle in
selected IPS
and
management
intensities
Dairy cattle in
selected IPS
and
management
intensities
Labor costs,
labor share,
energy prices
Labor costs,
labor share
Plug-flow Digester, Dairy Cattle
With Engine
Without
Engine
1288 per head
(US)
790 per head
(US)
2.3 20
8.9 20
CH4: -85% $65 energy
revenue/
savings per
head (US)
CH4: -85% none
Dairy cattle in
selected IPS
and
management
intensities
Dairy cattle in
selected IPS
and
management
intensities
Labor costs,
labor share,
energy prices
Labor costs,
labor share
(continued)
V-54
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LIVESTOCK
Table 3-5: Abatement Measures for Manure Management (continued)
Abatement
Option
Installed Annual O&M
Capital Cost Cost
(2010 USD) (2010 USD)
Capital
Lifetime
(Years)
ucnon Benefits
Efficiency (Changes in
(change in Livestock or Adjustments
emissions Energy Technical Across
per head) Revenue) Applicability Regions
Fixed-film Digester, Hogs
With Engine
Without
Engine
128 per head 0.1 5 per head 20
(US) (US)
102 per head 0.1 2 per head 20
(US) (US)
CH4: -85% $8 energy
revenue/
savings per
head (US)
CH4: -85% none
Hogs in Labor costs,
selected LPS labor share,
and energy prices
management
intensities
Hogs in Labor costs,
selected LPS labor share
and
management
intensities
Covered Lagoon, Large-Scale, Hogs
With Engine
Without
Engine
43 per head 0.1 3 per head 20
(US) (US)
25 per head 0.06 per head 20
(US) (US)
CH4: -85% $8 energy
revenue/
savings per
head (US)
CH4: -85% none
Hogs in Labor costs,
selected LPS labor share,
and energy prices
management
intensities
Hogs in Labor costs,
selected LPS labor share
and
management
intensities
Covered Lagoon, Large-Scale, Dairy Cattle
With Engine
Without
Engine
Dome
Digester,
Cooking Fuel
and Light
11 82 per head 3.43 per head 20
(US) (US)
773 per head 2.01 per head 20
(US) (US)
50 per 1000 1.25 per 1000 10
Ibs liveweight Ibs liveweight
CH4: -85% $65 energy
revenue/
savings per
head (US)
CH4: -85% none
CH4: -50% $7 energy
revenue/
savings per
head hogs,
$48 energy
revenue/
savings per
head dairy
cattle
Dairy cattle in Labor costs,
selected LPS labor share,
and energy prices
management
intensities
Dairy cattle in Labor costs,
selected LPS labor share
and
management
intensities
Hogs and Labor costs,
dairy cattle in labor share,
selected LPS energy prices
and
management
intensities in
developing
countries
(continued)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-55
-------
LIVESTOCK
Table 3-5: Abatement Measures for Manure Management (continued)
Installed Annual O&M
Capital Cost Cost
Abatement
Ootion (2010 USD) (2010 USD)
Capital
Lifetime
(Years)
ucnon Benefits
Efficiency (Changes in
(change in Livestock or Adjustments
emissions Energy Technical Across
per head) Revenue) Applicability Regions
Polyethylene
Bag Digester,
Cooking Fuel
and Light
Centralized
Digester
20 per 1000
Ibs liveweight
163 per head
average for
hogs across
the EU, 1007
per head
average for
dairy cattle
across the EU
0.5 per 1000 10
Ibs liveweight
0.07 per head 20
for hogs, 2.06
dairy cattle
ChU: -50% $7 energy
revenue/
savings per
head hogs,
$48 energy
revenue/
savings per
head dairy
cattle
ChU: -85% $7 energy
revenue/
savings per
head hogs,
$48 energy
revenue/
savings per
head dairy
cattle
Hogs and
dairy cattle in
selected IPS
and
management
intensities in
developing
countries
Hogs and
dairy cattle in
selected IPS
and
management
intensities in
the EU-27
region
Labor costs,
labor share,
energy prices
Labor costs,
labor share,
energy prices
Complete-mix Digester
These digesters are more common in warmer climates, where manure is flushed out of barns or pens
with water, lowering the solids' concentration to a level generally between 3 and 10%. Often there is a
mixing tank where the manure accumulates before entering the digester. These digesters make use of
gravity and pumps to move the manure through the system. They are often in the shape of a vertical
cylinder and made of steel or concrete with a gas-tight cover. These digesters are typically heated to
maintain a constant temperature and gas flow.
• Capital Cost: $61/$100 per head (swine), $588/$958 per head (cattle) depending on optional
engine
• Annual O&M Cost: Estimated $0.07-$0.11 per head (swine), $2.06/3.35 (cattle)
• Annual Benefits: $8 per head (swine), $65 per head (cattle) if equipped with an engine and used
to displace purchased power
• Applicability: This option applies only to swine and cattle managed in intensive production
systems in developed regions
• Technical Efficiency: This analysis assumes a reduction in emissions per head of about 85%.
• Capital Lifetime: 20 years
Plug-flow Digester
These digesters consist of long and relatively narrow heated tanks, often built below ground level,
with gas-tight covers. Plug-flow digesters are only used for dairy manure because they require higher
manure solids' content, around 11 to 13%. As with complete-mix digesters, they are maintained at
constant temperatures throughout the year to maintain constant gas production.
V-56
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LIVESTOCK
• Capital Cost: $790/$1288 per head
• Annual O&M Cost: Estimated $2.30 - $8.90 per head
• Annual Benefits: $65 per head if equipped with an engine and used to displace purchased power
• Applicability: This option applies only to dairy cattle in developed regions
• Technical Efficiency: This analysis assumes a reduction in emissions per head of about 85%.
• Capital Lifetime: 20 years
Fixed-film Digester
This digester option may be appropriate where concentrations of solids are very low, such as in
swine manure management situations where manure is very diluted with water. Fixed-film digesters
consist of a tank packed with inert media on which bacteria grow as a biofilm.
• Capital Cost: $102/$128 per head
• Annual O&M Cost: Estimated $0.06 - $0.13 per head
• Annual Benefits: $8 per head if equipped with an engine and used to displace purchased power
• Applicability: This option applies only to swine managed in intensive production systems in
developed regions
• Technical Efficiency: This analysis assumes a reduction in emissions per head of about 85%.
• Capital Lifetime: 20 years
Large-scale Covered Lagoon
Covered earthen lagoons are the simplest of the systems used in developed countries and generally
the least expensive, though there is quite a bit of variation in the systems that have been built. This
system is used with low manure solids' concentration (less than 3%) and can be used for swine or dairy
cattle. CH4 is captured by covering the lagoon where manure is stored with a floating cover and piping
the gas out to a flare or used on-farm. Because these digesters are not generally heated, the available gas
flow varies significantly over the course of the year.
• Capital Cost: $25/$43 per head (swine), $773/$l,182 (cattle)
• Annual O&M Cost: Estimated $0.06/$0.13 per head (swine), $2.01/$3.43 (cattle)
• Annual Benefits: $8 per head (swine), $65 per head (cattle) if equipped with an engine and used
to displace purchased power
• Applicability: This option applies only to swine and dairy cattle managed in intensive
production systems in developed regions
• Technical Efficiency: This analysis assumes a reduction in emissions per head of about 85%.
• Capital Lifetime: 20 years
Small-scale Dome Digester
These are small, unheated digesters used in some developing countries, including China and India. A
typical dome digester is a brick-lined cylinder sunk in the ground with a wall dividing the cylinder in
two with inlet and outlet ports connected to the bottom of the tank. Biogas generated is typically used by
the household for cooking and other household energy needs.
• Capital Cost: $50 per 1,000 Ibs liveweight
• Annual O&M Cost: Estimated $1.25 per 1,000 Ibs liveweight
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-57
-------
LIVESTOCK
• Annual Benefits: $7 per head (swine), $48 per head (cattle)
• Applicability: This option applies to swine and dairy cattle in developing regions
• Technical Efficiency: This analysis assumes a reduction in emissions per head of about 50%.
• Capital Lifetime: 10 years
Centralized Digester
Large centralized digesters where individual farmers transport their waste to in order for large scale
digestion and dispersion of capital costs.
• Capital Cost: $163 per head (swine), $1,007 per head (cattle)
• Annual O&M Cost: Estimated $0.07 per head (swine), $2.06 per head (cattle)
• Annual Benefits: Assumed to provide the same annual benefits per head of livestock as the large
individual systems described above.
• Applicability: This option applies only to swine and dairy cattle in intensively managed
production systems in EU-27 regions
• Technical Efficiency: This analysis assumes a reduction in emissions per head of about 85%.
• Capital Lifetime: 20 years
V.3.4
Marginal Abatement Costs Analysis
The MAC analysis assimilates the abatement measures' technology costs, expected benefits, and
emission reductions presented in Section X.3 to compute the cost of abatement for each measure. Similar
to the approach used in other non-CCfe sectors of this report, we compute a break-even price for each
abatement option for 195 countries to construct MAC curves illustrating the technical, net GHG
mitigation potential at specific break-even prices for 2010, 2020, and 2030.
This section describes the general modeling approach applied in this sector, which serve as additional
inputs to the MAC analysis that adjust the abatement project costs, benefits, and the technical abatement
potential in each country.
V.3.4.1
Development of Disaggregated Baseline Livestock Populations
Livestock population projections at a disaggregated level are a key component of estimating potential
emissions reductions from livestock production. Tables 3-6 and 3-7 present baseline projected livestock
populations by species at the global and regional levels, respectively. As noted earlier in this chapter,
these projections are based on country-level livestock population data from USEPA (2012), adjusted using
livestock production and market price projections from Nelson et al. (2010) to derive projected livestock
populations.
Table 3-6: Projected Global Livestock Populations by Species
Species
Asses
Mules
Buffalo
Camels
Cattle
2010
43,694,545
10,687,809
181,068,216
25,230,544
1,141,799,067
2015
44,710,040
9,719,699
190,207,386
27,116,465
1,233,755,944
2020
46,511,983
9,087,894
200,872,941
29,660,950
1,293,778,238
2025
49,232,861
8,688,065
213,277,930
33,095,191
1,348,359,726
2030
53,072,574
8,454,990
227,690,865
37,758,103
1,392,273,902
(continued)
V-58
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LIVESTOCK
Table 3-6: Projected Global Livestock Populations by Species (continued)
Species
Dairy cattle
Goats
Horses
Other camelids
Pigs
Sheep
Turkeys
Chickens
Ducks
Geese
l32m
247,195,753
882,119,170
58,864,443
6,926,082
947,222,554
1,126,923,912
488,712,578
18,934,787,428
1,156,375,916
365,742,348
^BH^I
248,770,901
947,475,133
59,669,740
7,090,544
963,684,813
1,264,771,843
506,073,755
20,500,590,776
1,288,661,778
404,547,438
^^QQB
250,894,992
1,035,241,803
61,198,242
7,260,388
981,443,858
1,421,729,708
524,421,101
22,251,209,335
1,437,928,802
447,801,893
2025
253,588,443
1,151,801,402
63,481,024
7,435,790
1,000,597,025
1,600,736,874
543,822,679
24,210,358,750
1,606,439,449
496,016,058
2030
256,874,692
1,306,127,535
66,580,631
7,616,931
1,021,251,228
1,805,223,246
564,352,297
26,405,046,832
1,796,773,159
549,759,182
The livestock populations were disaggregated into 14 categories of livestock production systems
(LPSs) based on the Gridded Livestock of the World (Robinson et al., 2011), along with an "UNKNOWN"
category that was added to account for cases where there were no data available to assign a livestock
species to an LPS:
• LGA - livestock only grassland arid and semiarid
• LGH - livestock only grassland humid and subhumid
• LGT - livestock only grassland highland temperate
• LGY - livestock only grassland hyper arid
• MIA - irrigated mixed crop-livestock systems arid and semiarid
• MIH - irrigated mixed crop-livestock systems humid and subhumid
• MIT - irrigated mixed crop-livestock systems highland temperate
• MIY - irrigated mixed crop-livestock systems hyper arid
• MRA - rainfed mixed crop-livestock systems arid and semiarid
• MRH - rainfed mixed crop-livestock systems humid and subhumid
• MRT - rainfed mixed crop-livestock systems highland temperate
• MRY - rainfed mixed crop-livestock systems hyper arid
• URBAN - built-up areas
• OTHER - other systems
• UNKNOWN - no data available to assign to LPS
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-59
-------
Table 3-7: Regional Livestock Populations by Species, 2010 and 2030
Asses
Mules
Buffalo
Camels
Cattle
Dairy Cattle
Goats
Horses
Pigs Sheep Turkeys
Chickens
Ducks
Geese
2010
AFRC
MIEA
CSAM
EURO
EURA
ASIA
NAAM
19,060,943
2,749,155
3,798,475
846,866
677,880
13,249,225
3,312,000
1,077,045
198,708
2,955,700
281,198
1,068
2,862,090
3,312,000
5,339,864
753,069
1,111,814
425,943
343,107
173,094,419
—
21,477,486
1,247,756
309,916,289
1,208
221,501
2,282,593
—
220,327,356
8,021,314
37,138,939
76,446,634
29,855,690
362,788,699
134,443,085
58,488,802
5,567,448
25,161,984
30,849,246
23,418,817
79,456,157
12,276,345
299,505,213
45,065,821
17,250,838
20,260,700
12,030,080
468,038,089
12,057,283
4,709,306
225,902
6,926,082
4,593,009
4,071 ,274
11,512,981
16,501,133
27,178,558 304,049,685 17,230,236
214,307 109,243,396 6,102,128
69,414,647 79,203,085 57,089,883
163,465,255 141,794,134 101,594,263
28,727,659 78,557,745 18,007,749
563,373,447 398,701 ,823 2,955,455
94,848,681 15,374,043 285,732,864
1 ,452,628,008
845,345,519
2,434,716,295
1 ,660,083,208
664,246,895
9,099,019,576
2,778,747,928
16,880,560
1 ,886,081
8,424,513
44,889,390
11,508,213
1,056,093,702
16,693,456
12,657,925
2,092,632
425,747
14,050,300
8,305,203
327,903,127
307,414
2030
AFRC
MIEA
CSAM
EURO
EURA
ASIA
NAAM
28,605,408
2,742,926
3,525,580
609,117
849,921
13,427,623
3,312,000
1,410,927
197,774
2,565,528
234,846
844
733,070
3,312,000
9,840,993
1,165,203
876,235
1,008,184
282,435
214,517,815
—
33,269,240
1,503,917
363,165,169
4,597
442,251
2,538,098
—
266,035,319
10,420,595
34,416,861
78,327,604
33,053,022
487,150,416
154,121,776
73,540,139
6,127,068
24,447,270
27,596,465
21,338,795
82,381,796
11,473,568
432,866,460
53,996,856
17,058,230
20,202,782
24,550,502
735,097,827
14,965,839
7,753,070
238,226
7,616,931
5,548,271
5,634,282
10,892,381
19,456,171
37,167,402 537,137,245 19,285,366
212,496 181,275,748 7,462,777
87,526,659 123,946,758 79,940,136
153,126,179 192,995,324 102,791,593
27,922,758 130,930,762 18,608,201
624,250,619614,909,005 3,991,403
91,045,115 24,028,403 332,272,820
1,916,766,477
993,216,019
3,293,431,862
1,706,164,516
700,410,055
14,449,006,323
3,346,051,579
21 ,622,904
2,251,984
11,572,205
44,614,600
12,039,543
1,683,405,389
21 ,266,535
15,777,661
2,547,450
620,925
13,888,742
8,669,961
507,862,027
392,416
m
en
Note: AFRC = Africa; MIEA = Middle East; CSAM = Central and South America; EURO = Europe; EURA = Eurasia; ASIA = Asia; NAAM = North America
-------
LIVESTOCK
The LPSs capture major combinations of livestock production systems of the world with respect to
land use type and climate. Livestock populations across livestock production systems were assigned for
pigs, goats, sheep, dairy cattle, and beef cattle based on the country-level data from Robinson et al. (2011).
Approximation was made for the distribution of selected species where LPS data were not available.
In addition to disaggregation by LPS, certain livestock species were further disaggregated into
production intensity categories. For pigs, data were provided by the Food and Agriculture Organization
(FAO) that separated country-level pig populations into three intensity categories for each LPS: intensive,
semi-intensive, and extensive. Those data were used to assign intensity levels to pig populations and this
distribution was used as a proxy for poultry production intensity in countries with both pig and poultry
production. For beef and dairy cattle, regional allocation of cattle across intensity categories in Robinson
et al. (2011) was used to assign intensity levels to each country located within that region. For other
species, all intensity levels were defined as unknown. As an example, Table 3-8 presents the assumed
distribution of livestock across livestock production systems and intensity classifications for India, the
largest emitter for the livestock production sector.
The detailed disaggregation of baseline populations allows for better definition of the technical
applicability of mitigation options. For instance, this study only applies large-scale digesters to intensive
dairy and hog production systems in each country. Intensive grazing is assumed to be applicable only to
relatively high productivity mixed crop-livestock systems that rely on irrigation or are in humid and
subhumid or temperate highland LPS designations. The use of a highly disaggregated baseline in this
study serves to define the share of emissions where mitigation options can potentially be applied.
Enteric fermentation and manure management emissions for each subset of livestock populations
were calculated using the IPCC default values consistent with those used in USEPA (2012). The one
exception is for enteric fermentation emissions in Africa, where relative emissions reported in Robinson
et al. (2011) were used to scale default IPCC emissions per head for different LPS categories.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-61
-------
Table 3-8: Livestock Distribution by Intensity and Livestock Production System for India, 2010 (% of animals by species)
Species
Asses
Mules
Buffalo
Camels
Cattle
Cattle
Cattle
Dairy Cattle
Dairy Cattle
Dairy Cattle
Goats
Horses
Pigs
Pigs
Pigs
Sheep
Chickens
Chickens
Chickens
Ducks
Ducks
Ducks
Intensity
unknown
unknown
unknown
unknown
intensive
extensive
unknown
intensive
extensive
unknown
unknown
unknown
intensive
extensive
semi-intensive
unknown
intensive
extensive
semi-intensive
intensive
extensive
semi-intensive
LGA
0.5%
0.5%
0.5%
0.5%
0.3%
0.2%
0.0%
0.2%
0.2%
0.0%
1.8%
0.5%
0.0%
0.0%
0.0%
3.4%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
LGH
0.1%
0.1%
0.1%
0.1%
0.1%
0.0%
0.0%
0.1%
0.0%
0.0%
0.0%
0.1%
0.5%
0.0%
0.1%
0.0%
0.5%
0.0%
0.1%
0.5%
0.0%
0.1%
LGT
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.1%
0.0%
0.0%
0.0%
0.1%
0.0%
0.0%
0.1%
0.0%
0.0%
LGY
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
MIA
31.7%
31.7%
31.7%
31.7%
17.7%
12.6%
1.4%
18.6%
13.3%
1.5%
34.6%
31.7%
13.8%
14.6%
3.3%
30.8%
13.8%
14.6%
3.3%
13.8%
14.6%
3.3%
MIH
6.6%
6.6%
6.6%
6.6%
3.7%
2.6%
0.3%
3.9%
2.8%
0.3%
7.0%
6.6%
3.0%
2.0%
0.7%
5.0%
3.0%
2.0%
0.7%
3.0%
2.0%
0.7%
MIT
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.3%
0.0%
0.1%
0.1%
0.3%
0.0%
0.1%
0.3%
0.0%
0.1%
MIY
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
MRA
40.2%
40.2%
40.2%
40.2%
22.4%
16.0%
1.8%
21.0%
15.0%
1.7%
39.0%
40.2%
7.1%
12.0%
1.7%
45.5%
7.1%
12.0%
1.7%
7.1%
12.0%
1.7%
MRH
11.0%
11.0%
11.0%
11.0%
6.2%
4.4%
0.5%
6.5%
4.6%
0.5%
7.9%
11.0%
16.9%
4.3%
4.0%
5.1%
16.9%
4.3%
4.0%
16.9%
4.3%
4.0%
MRT
1.3%
1.3%
1.3%
1.3%
0.7%
0.5%
0.1%
0.8%
0.5%
0.1%
1.3%
1.3%
5.2%
0.5%
1.2%
2.6%
5.2%
0.5%
1.2%
5.2%
0.5%
1.2%
MRY
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
URBAN
7.3%
7.3%
7.3%
7.3%
4.1%
2.9%
0.3%
4.1%
3.0%
0.3%
7.2%
7.3%
2.6%
4.1%
0.6%
6.8%
2.6%
4.1%
0.6%
2.6%
4.1%
0.6%
Other
1.2%
1.2%
1.2%
1.2%
0.7%
0.5%
0.1%
0.7%
0.5%
0.1%
1.3%
1.2%
0.5%
0.5%
0.1%
0.7%
0.5%
0.5%
0.1%
0.5%
0.5%
0.1%
m
en
-------
LIVESTOCK
V.3.4.4
MAC Analysis Results
As highlighted at the beginning of this chapter, global abatement potential in the livestock sector
equates to approximately 3% of its total annual emissions between 2010 and 2030 at no or a relatively low
carbon price of $5 per ton of CCh equivalent ($/tCChe). In 2030, total abatement potential in the livestock
sector is 70 MtCCfee at no carbon price, 86 MtCCfee at a carbon price of $5/tCChe, and 128 MtCCfee at a
carbon price of $20/tCChe, representing 2.6%, 3.2% and 4.7% of the total sector emissions, respectively.
Table 3-9 presents the estimated mitigation potential at various break-even prices for the top-five emitting
countries and rest of regional groups in 2030 under an assumption that livestock populations adjust to
maintain production at baseline levels when mitigation options result in changing productivity.
Table 3-9: Abatement Potential by Region at Selected Break-Even Prices in 2030 (MtC02e), Baseline
Production Case
Country/Region
-10
Break-Even Price ($/tC02e)
5 10 15 20 30 50
100
100+]
Top 5 Emitting Countries
India
China
Brazil
United States
Pakistan
7.6
5.6
4.6
0.3
0.7
7.6
5.6
4.6
0.3
1.7
11.7
6.2
4.8
4.1
2.5
14.0
6.2
6.7
8.7
2.5
14.5
10.4
7.1
8.7
2.5
14.5
10.4
9.9
13.2
2.7
16.4
14.6
10.2
13.2
2.7
22.4
24.1
10.6
15.5
3.1
24.7
32.6
12.5
24.0
3.1
25.0
35.5
13.2
37.5
4.4
27.4
38.3
13.6
43.2
5.6
Rest of Region
Africa
Asia
Central & South
America
Eurasia
Europe
Middle East
North America
World Total
8.7
12.3
5.8
1.2
6.2
1.5
1.3
55.7
9.3
13.6
6.4
1.3
6.4
1.5
1.7
60.0
11.8
18.1
7.8
1.4
10.7
1.6
2.5
83.3
12.3
21.2
8.9
1.5
11.0
1.6
2.5
97.2
12.6
24.8
10.4
1.6
11.3
1.6
2.6
108.1
12.9
26.3
11.1
1.6
12.4
1.6
3.8
120.5
13.1
30.2
12.6
1.6
15.5
1.7
4.4
136.3
13.3
35.0
13.1
1.6
16.4
1.7
5.0
161.6
13.6
38.1
14.2
1.8
20.9
1.7
6.2
193.3
14.0
40.4
14.8
2.0
29.7
1.7
9.2
227.5
14.6
45.5
15.2
2.7
50.6
1.9
10.0
268.6
Mitigation potential and its cost-effectiveness vary significantly by country or region. At the regional
level, Asia (in particular South and Southeast Asia), Africa, Central and South America and the European
Union show the most significant potential for reducing GHG emissions from livestock operations. For
instance, in 2030 mitigation potential in Asia is estimated to be 27 MtCChe with no carbon price and 34
MtCCfee at a carbon price of $20/tCChe. Central and South America can achieve mitigation potential of 12
MtCChe in 2030 at no carbon price, and mitigation potential can increase to 22 MtCChe at a carbon price
of $20/tCChe. Figure 3-5 shows the MAC curves for the top-five emitting countries in 2030.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-63
-------
LIVESTOCK
Figure 3-5: Marginal Abatement Cost Curve for Top 5 Emitters in 2030 (Baseline Production Case)
o
u
a, $20
$10
$0
-$10
-$20
-$30
•India
Brazil
• China
•United States
Pakistan
10.0 15.0 20.0 25.0 30.0 35.0
Non-CO2 Reduction (MtCO2e)
The MAC analysis also suggests that mitigation of enteric fermentation methane emissions presents
the most cost-effective mitigation opportunity for options evaluated in this report. Manure management
mitigation measures mostly require additional investments or financial incentives to achieve emissions
reductions. The most cost-effective mitigation options for the livestock sector (i.e., measures that
dominate the MAC curves at break-even carbon prices at or below $0/MtCChe) include:
• intensive grazing in East Asia (e.g., Japan, Korea and China) and Central and South America;
• BST administered to dairy cattle in developing regions;
• antimethanogens administered to sheep and goats as well as beef and dairy cattle;
• improved feed conversion efficiency of the cattle populations; and
• propionate precursors administered to beef and dairy cattle in developing regions
Figure 3-6 shows the distribution of mitigation potential across individual types of options at a global
scale based on total technical potential (regardless of price) calculated in the MAC analysis.
V-64
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LIVESTOCK
Figure 3-6. Global Net GHG Livestock Emissions Reduction Potential by Mitigation Option (Baseline
Production Case)
300
250
2030
Large digester
Small digester
Antibiotics
Propionate precursors
I Improved feed conversion
Intensive grazing
I Antimethanogen
ibST
V.3.5
Sensitivity Analyses
In this section, we explore sensitivity analyses to examine the potential effects on estimated
mitigation potential. Although many of the mitigation options examined are expected to increase
productivity and would therefore require fewer animals to produce the same amount of output, livestock
populations may not decrease accordingly. Due to increasing demand for livestock products and
potential reductions in the price of these products with higher productivity, the quantity of livestock
products demanded may increase. Thus, we examine an alternative scenario that holds the number of
livestock constant at the projected baseline populations. To the extent that productivity is increased by
adoption of the GHG mitigation options considered, this scenario will result in higher global production.
In addition, given mixed conclusions on the near-term prospects of antimethanogens, we also present
mitigation estimates developed excluding antimethanogens as an option.
Baseline Number of Animals
As noted above, many of the mitigation options in the baseline production case reduce the emissions
per unit of meat or milk but may increase the emissions per animal. This section explores this relationship
further by presenting an alternative scenario built around a constraint on the number of animals, keeping
the herd sizes the same as estimated in the baseline.
As before, the MAC model only includes options that result in lower emissions. But with the number
of animals held constant, those mitigation strategies that increase emissions per animal in a given region
are excluded in that region. The result is 15 to 39% lower mitigation potential as shown in Figure 3-7 and
Table 3-10.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-65
-------
LIVESTOCK
Figure 3-7: Global Abatement Potential in Livestock Management, Baseline Number of Animals : 2010,
2020, and 2030
•2010
•2020
•2030
250
Non-CO2 Reduction (MtCO2e)
Table 3-10: MAC Results and Differences from Constant Production Case for Baseline Number of Animals
Scenario
$/tC02e
Total Difference from Total Difference from Total Difference
Reduction Constant Reduction Constant Reduction from Constant
MTC02e Production (%) MTC02e Production(%) MTC02e Production (%)
0
5
10
15
20
25
30
35
40
45
50
49
58
65
79
84
86
90
93
96
98
102
-21%
-20%
-27%
-25%
-29%
-33%
-35%
-37%
-39%
-39%
-39%
54
61
68
78
88
91
97
98
101
103
108
-19%
-22%
-27%
-29%
-31%
-33%
-34%
-34%
-39%
-39%
-39%
60
65
73
81
87
97
101
106
109
113
118
-15%
-25%
-25%
-26%
-31%
-33%
-35%
-33%
-35%
-37%
-38%
V-66
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LIVESTOCK
No Antimethanogen
The science and policy literature varies in its treatment of antimethanogens. The Australian
government included them in their recent study (Whittle et al., 2013). However ICF International, in a
recent analysis for USD A, concludes that "more research is needed to evaluate the potential GHG impacts
of changes in diets, use of feed additives, and breeding." (ICF International, 2013) For comparison
purposes we estimated MAC curves as above except by assuming antimethanogens are unavailable in all
regions and time periods. Results are shown in Figure 3-8 and Table 3-11. Globally, the mitigation
potential in the livestock sector is reduced 16 to 31% in the scenario with no antimethanogens and
baseline production.
Figure 3-8: Global Abatement Potential in Livestock Management, Baseline Production with No
Antimethanogen: 2010, 2020, and 2030
•2010
•2020
•2030
180
Non-CO2 Reduction (MtCO2e)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-67
-------
LIVESTOCK
Table 3-11: MAC Results and Differences from Constant Production Case for No Antimethanogen Scenario
Total Difference Total Difference from Total Difference
Reduction from Constant Reduction Constant Reduction from Constant
$/tC02e MTC02e Production (%) MTC02e Production(%) MTC02e Production (%)
0
5
10
15
20
25
30
35
40
45
50
48
54
70
87
99
110
120
126
133
136
145
-28%
-31%
-25%
-21%
-22%
-19%
-18%
-15%
-19%
-20%
-18%
53
61
73
85
103
113
122
125
136
140
147
-20%
-22%
-21%
-23%
-19%
-18%
-16%
-16%
-17%
-17%
-17%
49
62
73
82
98
115
126
130
139
149
152
-31%
-28%
-25%
-26%
-23%
-20%
-19%
-18%
-17%
-17%
-20%
Combined Baseline Number of Animals and No Antimethanogen
Results for a combined scenario including both an assumption that the number of livestock under the
mitigation scenario remains equal to the baseline and no applicability of antimethanogens are presented
in Figure 3-9 and Table 3-12. Under this scenario, there is a reduction in mitigation potential of between
16 and 43% relative to the primary case where production of livestock products is assumed to remain
equal to baseline levels.
V-68
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LIVESTOCK
Figure 3-9: Global Abatement Potential in Livestock Management, Baseline Number of Animals with No
Antimethanogen: 2010, 2020, and 2030
8
•2010
2020
•2030
120
Non-CO2 Reduction (MtCO2e)
Table 3-12: MAC Results and Differences from Constant Production Case for Combined Baseline Number
of Animals and No Antimethanogen Case
$/tC02e
Total Difference from Total
Reduction Constant Reduction
MTC02e Production (%) MTC02e
Difference from
Constant
Production (%)
Total Difference
Reduction from Constant
MTC02e Production (%)
0
5
10
15
20
25
30
35
40
45
50
50
57
63
70
74
78
82
87
90
93
98
-20%
-21%
-30%
-34%
-38%
-39%
-41%
-41%
-42%
-42%
-42%
56
64
69
77
82
86
90
96
100
102
104
-16%
-18%
-26%
-31%
-36%
-37%
-39%
-35%
-39%
-40%
-41%
57
67
73
78
82
86
94
97
102
105
108
-18%
-22%
-26%
-30%
-35%
-41%
-40%
-39%
-39%
-41%
-43%
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-69
-------
LIVESTOCK
Change in Production of Livestock Products with Number of Animals Held at Baseline
Levels
For the scenario where livestock populations are kept at projected baseline levels, there will be changes in
production of livestock products due to changes in output per head for many options. Figures 3-10 and
3-11 show the change in global beef production and global milk production from dairy cattle estimated if
all production were to switch from baseline management into that option.
Figure 3-10: Global Beef Production under Baseline and Mitigation Options, Assuming Full Adoption of
Individual Options and Holding the Number of Animals Constant
120,000,000
100,000,000
80,000,000
V)
c
o
.a 60,000,000
tt
40,000,000
20,000,000
I Base
I Small Digesters
Large Digesters
I Improved Feed Conversion
Antibiotics
IbST
Propionate Precursors
Antimethanogen
Intensive Grazing
2010
2030
V-70
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LIVESTOCK
Figure 3-11: Global Production of Milk from Dairy Cattle Under Baseline and Mitigation Options,
Assuming Full Adoption of Individual Options and Holding the Number of Animals Constant
120,000,000
100,000,000
80,000,000
V)
o
.a 60,000,000
tt
40,000,000
20,000,000
I Base
I Small Digesters
Large Digesters
I Improved Feed Conversion
Antibiotics
IbST
Propionate Precursors
Antimethanogen
Intensive Grazing
2010
2030
V.3.4.5.
Uncertainties and Limitations
Given the complexities of the global livestock sector, the estimated GHG mitigation potential and
marginal abatement cost curves are subject to a number of uncertainties and limitations:
• Availability and quality of data to represent the highly complex and heterogeneous livestock
production systems of the world. Although there are major improvements in the characterization
of the business-as-usual baseline conditions since the previous EPA report (USEPA, 2006), data in
some areas, such as management practices, are not always available for all countries or regions
and approximations must be made based on limited literature or expert judgment.
• Availability of mitigation measure cost data and in some cases scientific understanding of
mitigation impacts. Collecting and developing consistent cost estimates of emerging mitigation
measures or options that are not widely adopted has proven to be challenging. Moreover,
scientific understanding of the mitigation effects and animal and human health impacts of some
mitigation measures is still limited. In addition, some mitigation measures, such as pasture
management options that lead to reductions in enteric Q-k emissions and enhancement in soil
carbon storage, would require a different analytical framework that is beyond the scope of this
study.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-71
-------
LIVESTOCK
Optimistic assumptions on technology adoption. The analysis assumes that if mitigation
technology is considered feasible in a country or region, it is fully adopted in 2010 and through
the analysis period. Research suggests that adoption of new technology in the agricultural sector
is a gradual process and various factors potentially inhibit the adoption of a new GHG-mitigating
technology (e.g., farm characteristics, access to information and capital, and cultural and
institutional conditions). Adoption of the various technologies and management practices (such
as supplementation) faces even greater challenges. The mitigation potential presented in this
analysis should be viewed to represent the technical potential of the mitigation options analyzed.
Potential market feedback from livestock productivity improvement. The analysis assumes
constant production level when evaluating mitigation potential of abatement measures. This
analysis does not, however, address the possibility of an emissions increase as a result of lower
costs per unit through such efficiency gains, which could in turn increase the quantity demanded.
Potential interactions of multiple mitigation measure. In this analysis, mitigation options are
applied to independent segments of the livestock populations to avoid double counting. In
reality, multiple mitigation options can be applied and their potential interactions may affect the
aggregate GHG mitigation. For example, various measures can improve feed conversion
efficiency (e.g., concentrate inclusion, dietary additives such as oils) and their effectiveness would
depend on the other measures implemented; measures that reduce Q-k emissions from manure
management (e.g., aeration) would likely increase N2O emissions; measures that improve feed
conversion efficiency would likely change N2O emissions in livestock manure; measures that
improve diet quality for grazing livestock would likely change GHG emissions from agricultural
soils. The interactive effects are not fully addressed in this analysis.
V-72 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
LIVESTOCK
References
Archibeque, S., Haugen-Kozyra, K., Johnson, K., Kebreab, E., & Powers-Schilling, W. (2012). Near-Term
Options for Reducing Greenhouse Gas Emissions from Livestock Systems in the United States: Beef,
Dairy, and Swine Production Systems. Nicholas Institute for Environmental Policy Solutions Report
NI R12-04, July 2012.
Bates, J., N. Brophy, M. Harfoot and J. Webb. (2009). Agriculture: methane and nitrous oxide. Sectoral
Emission Reduction Potentials and Economic Costs for Climate Change (SERPEC-CC). pp. 62.
Costa, A. (2012). Personal communication.
Grainger, C., R. Williams, R. J. Eckard, and M. C. Hannah. 2010. A high dose of monensin does not reduce
methane emissions of dairy cows offered pasture supplemented with grain. J. Dairy Sti. 93:5300-5308.
Hristov, A.N., J. Oh, C. Lee, R. Meinen, F. Montes, T. Ott, J. Firkins, A. Rotz, C. Dell, A. Adesogan, W.
Yang, J. Tricarico, E. Kebreab, G. Waghorn, J. Dijksta, and S. Oosting. 2013. Mitigation of greenhouse gas
emissions in livestock production -A review of technical options for non-CO2 emissions. Edited by Pierre J.
Gerber, Benjamin Henderson and Harinder P.S. Makkar. FAO Animal Production and Health Paper
No. 177. FAO: Rome, Italy.
ICF International, Greenhouse Gas Mitigation Options and Costs for Agricultural Land and Animal
Production in the United States. February 2013.
Nelson, G.C., M.W. Rosegrant, A. Palazzo, I. Gray, C. Ingersoll, R. Robertson, S. Tokgoz, T. Zhu, T.B.
Sulser, C. Ringler, S. Msangi, and L. You. (2010). Food Security, Farming, and Climate Change to
2050: Scenarios, Results, Policy Options. International Food Policy Research Institute: Washington,
DC.
Robinson, T.P., Thornton P.K., Franceschini, G., Kruska, R.L., Chiozza, F., Notenbaert, A., Cecchi, G.,
Herrero, M., Epprecht, M., Fritz, S., You, L., Conchedda, G. & See, L. (2011). Global livestock
production systems. Rome, Food and Agriculture Organization of the United Nations (FAO) and
International Livestock Research Institute (ILRI), 152 pp.
Roos, K. (2012). Personal communication.
U.S. Department of Agriculture, Economic Research Service (USDA, ERS). February 2011. Climate Change
Policy and the Adoption of Methane Digesters on Livestock Operations. ERR-111. U.S. Department of
Agriculture, Economic Research Service.
U.S. Department of Agriculture, Natural Resources Conservation Service (USDA, NRCS). October 2007.
Analysis of Energy Production Costs of Anaerobic Digestion Systems of U.S. Livestock Production Facilities.
Available at: http://www.agmrc.org/media/cms/manuredigesters FC5C31FOF7B78.pdf.
U.S. Environmental Protection Agency (USEPA) (2006). Global Mitigation of Non-CO2 Greenhouse
Gases. EPA 430-R-06-005. Washington DC.
USEPA (2010). Anaerobic Digestion Capital Costs for Dairy Farms. Washington, DC: U.S. Environmental
Protection Agency. Available at: http://www.epa.gov/AgSTAR/pdf/digester cost fs.pdf.
USEPA. (2012). Global Anthropogenic Non-CO2 Greenhouse Gas Emissions: 1990-2030. EPA 430-R-12-006.
Washington, DC: USEPA.
Whittle, L., Hug, B., White, S., Heyhoe, E., Harle, K., Mamun, E., & Ahammad, H. (2013). Costs and
Potential of Agricultural Emissions Abatement in Australia. Australian Bureau of Agricultural and
Resource Economics and Sciences. Technical Report 13.2, March 2013.
Williams, Y. J., S. Popovski, S. M. Rea, L. C. Skillman, A. F. Toovey, K. S. Northwood, and A. D. G.
Wright. 2009. A vaccine against rumen methanogens can alter the composition of archaeal
populations. Appl. Environ. Microbiol. 75:1860-1866.
World Resources Institute (WRI). (2012). Climate Analysis Indicators Tool (CAIT) database. Available at
http://www.wri.org/project/cait/.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-73
-------
-------
&ER&
United States Environmental Protection Agency
Office of Atmospheric Programs (6207J)
1200 Pennsylvania Ave., NW
Washington, DC 20460
------- |