ic- A;j-"";y
.n- ,. -if Atmvsp!i»>i< f-
Wn-,ninqton PC?0460
'oms (6207J) EPA 430-R-06-005
June 2006
Global Mitigation of Non-CO2
Greenhouse Gases
-------
How to obtain copies
You can electronically download this document from the
U.S. EPA's Web site at . To obtain additional copies of this
report, call the National Service Center for Environmental
Publications (NSCEP) at 1 - (800) 490-9198.
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 Christa Clapp,
(202) 343-9807, clapp.christa@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 Acknowledgments section
for a list of reviewers. A copy of the EPA Peer Review
guidelines can be downloaded from the following Web
page at .
-------
Global Mitigation of Non-CO2
Greenhouse Gases
June 2006
United States Environmental Protection Agency
Office of Atmospheric Programs (6207J)
1200 Pennsylvania Ave., NW
Washington, DC 20460
-------
Acknowledgments
This report was prepared under a contract between the U.S. Environmental Protection
Agency (USEPA) and RTI International (RTI). Casey Delhotal directed the preliminary
version of the report. Christa Clapp edited and directed completion of the report. Lead
authors include Mike Gallaher of RTI (Energy and Waste), Deborah Ottinger of USEPA
(Industrial Fluorinated Gases), Dave Godwin of USEPA (Ozone-Depleting Substitutes),
and Benjamin DeAngelo of USEPA (Agriculture). We thank USEPA reviewers Francisco
de la Chesnaye, Dina Kruger, Brian Guzzone, Reid Harvey, Kurt Roos, and Tom Wirth.
Other significant contributors and co-authors include Steven Rose of USEPA, Robert
Beach of RTI, Jochen Harnisch and Sina Wartmann of Ecofys, William Salas of Applied
GeoSolutions, Changsheng Li of University of New Hampshire, Stephen Del Grosso of
Colorado State University, and Timothy Sulser of International Food Policy Research
Institute.
The staff at RTI assisted in compiling and finalizing the report. The staff at ICF Consulting
and RTI prepared many of the individual analyses. Special recognition goes to Mike
Gallaher and Jeffrey Petrusa at RTI and Marian Van Pelt at ICF Consulting.
We also thank the following external reviewers: Paul Ashford (Caleb Group), Ward
Atkinson (SAE, retired), Dave Bateman (DuPont Fluoroproducts), Steven H. Bernhardt
(Honeywell Fluorine Products), Donald Bivens (DuPont Fluoroproducts), Eric Campbell
(DILO Company, Inc.), Nick Campbell (Arkema), Jim Crawford (The Trane Company),
David Crawley (Eurelectric), Hugh Crowther (McQuay International), William Dietrich
(York), Tony Digmanese (York), Maureen Hardwick (International Pharmaceutical Aerosol
Consortium), Jochen Harnisch (Ecofys), Susan Herrenbruck (Extruded Polystyrene .Foam
Association), Kenneth Hickman (York, retired), William Hill (General Motors), Mark
Hudgins (Environmental Control Systems, Inc.), Andy S. Kydes (Energy Information
Administration), Dick LaLumondier (NEMA), Stefan Lechtenbohmer (Wuppertal Institute
for Climate, Environment, Energy), Jan Lewandrowski (USDA Office of the Chief
Economist), Lin Erda (Chinese Academy of Agriculture Sciences), Jerry Marks (Jerry
Marks & Associates), Archie McCulloch (Marbury Technical Consulting and University
of Bristol, UK), Abid Merchant (DuPont), John Mutton (The Dow Chemical Company),
Enrique Otegui (Capiel), John Owens (3M), Friedrich Ploger (Siemens), G. Philip
Robertson (Michigan State University), J. Patrick Rynd (Owens Corning), Keith Smith
(University of Edinburgh), Pete Smith (University of Aberdeen), Eugene Smithart
(Danfoss Turbocor), Jerry Triplett (Partnership for Energy and Environmental Reform),
Tom Tripp (US Magnesium), Dan Verdonik (Hughes Associates, Inc.), William Walter
(Carrier Corporation), Thomas E. Werkema (Arkema), Kert Werner (3M), J. Jason West
(Princeton University), Robert Wickham (Wickham Associates), and Li Yue (Chinese
Academy of Agriculture Sciences). 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
Section Page
Executive Summary ES-1
I. Technical Summary
1.1 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 High-GWP Gases 1-4
1.2.3.1 HFCs 1-4
1.2.3.2 PFCs 1-4
1.2.3.3 Sulfur Hexaflouride (SF6) 1-4
1.2.4 Use of GWPs in this Report 1-5
1.3 Methodology 1-5
1.3.1 Baseline Emissions for Non-CO2 Greenhouse Gases 1-6
1.3.1.1 Baseline Emissions for Agriculture 1-7
1.3.1.2 Baseline Emissions for Fluorinated Gases 1-7
1.3.2 Mitigation Option Analysis Methodology 1-8
1.3.2.1 Technical Characteristics of Abatement Options .' 1-9
1.3.2.2 Economic Characteristics of Abatement Options 1-11
1.3.3 Marginal Abatement Curves 1-13
1.3.4 Methodological Enhancements from Energy Modeling Forum Study 1-15
1.4 Aggregate Results 1-16
1.4.1 Baselines 1-16
1.4.1.1 By Non-CO2 Greenhouse Gas 1-16
1.4.1.2 By Major Emitting Sectors and Countries 1-17
1.4.2 Global MACs 1-19
1.5 Limitations and Applications of MACs 1-21
1.5.1 Limitations and Uncertainties 1-21
1.5.1.1 Exclusion of Transaction Costs 1-22
1.5.1.2 Static Approach to Abatement Assessment 1-22
1.5.1.3 Limited Use of Regional Data 1-22
1.5.1.4 Exclusion of Indirect Emissions Reductions 1-22
VII
-------
Section Page
1.5.2 Practical Applications of MACs in Economic Models 1-22
1.6 References , 1-23
Appendix
A: Additional Information to Technical Summary A-l
II. Energy
11.1 Coal Mining Sector 11-1
II.l.l Introduction II-l
II.1.2 Baseline Emissions Estimates II-2
II.1.2.1 Activity Data II-3
II.1.2.2 Emissions Factors and Related Assumptions II-4
II.1.2.3 Emissions Estimates and Related Assumptions II-5
II.1.3 Cost of CH4 Emissions Reductions from Coal Mining II-6
II.1.3.1 Abatement Option Opportunities II-6
II.1.4 Results II-9
II.1.4.1 Data Tables and Graphs II-9
II. 1.4.2 Uncertainties and Limitations 11-11
II.1.5 Summary 11-12
II.1.6 References 11-12
11.2 Natural Gas Sector 11-15
II.2.1 Introduction 11-15
II.2.2 Baseline Emissions Estimates 11-17
II.2.2.1 Activity Data 11-17
11.22.2 Emissions Factors and Related Assumptions 11-19
II.2.2.3 Emissions Estimates and Related Assumptions 11-22
II.2.3 Cost of CH4 Emissions Reductions from Natural Gas Systems 11-23
II.2.3.1 Abatement Option Opportunities 11-24
II.2.4 Results 11-26
II.2.4.1 Data Tables and Graphs 11-26
II.2.5 Summary 11-30
II.2.6 References 11-30
11.3 Oil Sector 11-31
II.3.1 Introduction 11-31
VIII
-------
Section Page
II.3.1.1 Emissions from Production Field Operations 11-32
II.3.1.2 Emissions from Crude Oil Transportation 11-32
II.3.1.3 Emissions from Crude Oil Refining 11-32
II.3.1.4 Abatement Options 11-32
II.3.2 Baseline Emissions Estimates 11-33
II.3.2.1 Activity Factors 11-33
II.3.2.2 Emissions Factors and Related Assumptions 11-33
II.3.2.3 Emissions Estimates and Related Assumptions 11-37
II.3.3 The Cost of CH4 Emissions Reductions from Oil 11-37
II.3.3.1 Abatement Option Opportunities 11-37
II.3.4 Results 11-40
II.3.4.1 Data Tables and Graphs 11-40
II.3.5 Uncertainties and Limitations 11-40
II.3.6 Summary 11-42
II.3.7 References 11-42
Appendixes
B: Coal Mining Sector-Incorporating Technology Change to MAC Analysis B-l
C: Natural Gas Sector-Incorporating Technology Change to MAC Analysis C-l
D: Supporting Materials for Analysis of Oil Systems D-l
III. Waste
111.1 Landfill Sector 111-1
III. 1.1 Introduction III-l
III.1.2 Baseline Emissions Estimates III-2
III.1.2.1 Activity Data III-3
III.1.2.2 Emissions Factors and Related Assumptions III-3
III.1.2.3 Emissions Estimates and Related Assumptions III-4
III.1.3 Cost of Emissions Reductions from Landfills III-7
III.1.3.1 Abatement Option Opportunities III-7
III.1.4 Results III-9
III.1.4.1 Data Tables and Graphs III-9
III.1.4.2 Uncertainties and Limitations III-ll
III.1.5 Summary and Analysis 111-12
III.1.6 References Ill-12
IX
-------
Section Page
III.2 Wastewater Sector 111-13
III.2.1 Introduction Ill-13
III.2.1.1 Emissions from Wastewater Systems 111-14
III.2.2 Baseline Emissions Estimates 111-16
III.2.2.1 Activity Factors 111-17
III.2.2.2 Emissions Factors and Related Assumptions 111-18
III.2.2.3 Emissions Estimates and Related Assumptions 111-18
III.2.3 Emissions Reductions from Wastewater 111-20
III.2.3.1 Abatement Option Opportunities 111-20
III.2.3.2 Uncertainties and Limitations 111-22
III.2.4 Summary 111-22
III.2.5 References 111-22
Appendix
E: MSW Landfill Sector—Incorporating Technology Change to MAC Analysis E-l
IV. Industrial Processes
IV.1 N2O Emissions from Nitric and Adipic Acid Production , IV-1
IV.1.1 Introduction IV-1
IV.1.1.1 Nitric Acid '. IV-2
IV.1.1.2 Adipic Acid IV-2
IV.1.2 Baseline Emissions Estimates IV-2
IV.1.2.1 Activity Factors IV-2
IV.1.2.2 Emissions Factors and Related Assumptions IV-4
IV.1.2.3 Emissions Estimates and Related Assumptions TV-5
IV.1.3 • Cost of N2O Emissions Reductions from Industrial Processes IV-6
IV.1.3.1 Nitric Acid: N2O Abatement Option Opportunities IV-7
IV.1.3.2 Adipic Acid: N2O Abatement Option Opportunities IV-8
IV. 1.4 Results IV-8
IV.1.4.1 Data Tables and Graphs IV-8
IV.1.4.2 Uncertainties and Limitations IV-8
IV.1.5 Summary IV-12
IV.1.6 References IV-12
-------
Section Page
IV.2 HFC Emissions from Refrigeration and Air-Conditioning IV-15
IV.2.1 Introduction IV-15
IV.2.1.1 Household Refrigeration IV-15
IV.2.1.2 Motor Vehicle Air-Conditioning (MVAC) IV-16
IV.2.1.3 Chillers IV-16
IV.2.1.4 Retail Food Refrigeration IV-17
IV.2.1.5 Cold Storage Warehouses IV-17
IV.2.1.6 Refrigerated Transport IV-17
IV.2.1.7 Industrial Process Refrigeration IV-18
IV.2.1.8 Residential and Small Commercial Air-Conditioning and Heat Pumps IV-18
IV.2.2 Baseline Emissions Estimates IV-18
IV.2.2.1 Emissions Estimating Methodology IV-18
IV.2.2.2 Baseline Emissions IV-24
IV.2.3 Cost of HFC Emissions Reduction from Refrigeration and Air-Conditioning IV-25
IV.2.3.1 Description and Cost Analysis of Abatement Options IV-25
IV.2.3.2 Summary of Technical Applicability, Market Penetration, and Costs of
Abatement Options IV-45
IV.2.4 Results IV-45
IV.2.4.1 Data Tables and Graphs IV-46
IV.2.4.2 Uncertainties and Limitations IV-54
IV.2.5 Summary IV-54
IV.2.6 References IV-55
IV.3 HFC, HFE, and RFC Emissions from Solvents IV-59
IV.3.1 Introduction IV-59
IV.3.2 Baseline Emissions Estimates IV-60
IV.3.2.1 Emissions Estimating Methodology IV-60
IV.3.2.2 Baseline Emissions IV-61
IV.3.3 Cost of HFC, HFE, and PFC Emissions Reductions for Solvents IV-62
IV.3.3.1 Description and Cost Analysis of Abatement Options IV-62
IV.3.3.2 Summary of Technical Applicability, Market Penetration, and Costs of
Abatement Options TV-67
TV.3 A Results IV-68
IV.3.4.1 Data Tables and Graphs IV-68
IV.3.4.2 Uncertainties and Limitations IV-71
XI
-------
Section Page
IV.3.5 Summary IV-71
IV.3.6 References IV-72
IV.4 HFC Emissions from Foams IV-75
IV.4.1 Introduction IV-75
IV.4.2 Baseline Emissions Estimates IV-76
IV.4.2.1 Emissions Estimating Methodology IV-76
IV.4.2.2 Baseline Emissions IV-79
IV.4.3 Cost of HFC Emissions Reductions from Foams IV-81
IV.4.3.1 Abatement Options IV-81
IV.4.3.2 Description and Costs of Abatement Options IV-83
IV.4.3.3 Summary of Technical Applicability, Market Penetration, and Costs of
Abatement Options IV-90
IV.4.4 Results IV-91
IV.4.4.1 Data Tables and Graphs IV-91
IV.4.4.2 Uncertainties and Limitations IV-91
IV.4.5 Summary IV-99
IV.4.6 References IV-100
IV.5 HFC Emissions from Aerosols IV-103
IV.5.1 Introduction IV-103
IV.5.2 Baseline Emissions Estimates IV-103
IV.5.2.1 Emissions Estimating Methodology , IV-103
IV.5.2.2 Baseline Emissions IV-105
IV.5.3 Cost of HFC Emissions Reductions for Aerosols IV-105
IV.5.3.1 Description and Cost Analysis of Abatement Options IV-105
IV.5.3.2 Summary of Technical Applicability, Market Penetration, and Costs of
Abatement Options IV-110
IV.5.4 Results IV-111
IV.5.4.1 Data Tables and Graphs IV-111
IV.5.4.2 Uncertainties and Limitations IV-116
IV.5.5 Summary IV-116
IV.5.5.1 MDI Aerosols IV-117
IV.5.5.2 Non-MDI Aerosols IV-117
IV.5.6 References IV-117
XII
-------
Section Page
IV.6 HFC Emissions from Fire Extinguishing IV-119
IV.6.1 Introduction IV-119
IV.6.2 Baseline Emissions Estimates IV-121
IV.6.2.1 Emissions Estimating Methodology IV-121
IV.6.2.2 Baseline Emissions IV-122
IV.6.3 Cost of HFC Emissions Reductions from Fire Extinguishing IV-124
IV.6.3.1 Description and Cost Analysis of Abatement Options IV-124
IV.6.3.2 Summary of Technical Applicability, Market Penetration, and Costs of
Abatement Options IV-130
IV.6.4 Results IV-133
IV.6.4.1 Data Tables and Graphs IV-133
IV.6.4.2 Uncertainties and Limitations IV-136
IV.6.5 Summary IV-136
IV.6.6 References IV-136
IV.7 RFC Emissions from Aluminum Production IV-139
IV.7.1 Technology-Adoption Baseline IV-139
IV.7.2 No-Action Baseline IV-141
IV.7.3 Cost of PFC Emissions Reduction from Aluminum Production IV-142
IV.7.3.1 Abatement Options IV-142
IV.7.4 Results IV-146
IV.7.4.1 Data Tables and Graphs IV-146
IV.7.4.2 Global and Regional MACs and Analysis IV-149
IV.7.4.3 Uncertainties and Limitations IV-151
IV.7.5 References IV-152
IV.8 HFC-23 Emissions from HCFC-22 Production IV-155
IV.8.1 Source Description IV-155
IV.8.1.2 No-Action Baseline IV-156
IV.8.1.3 Technology-Adoption Baseline IV-158
IV.8.2 Cost of HFC-23 Reduction from HCFC-22 Production IV-159
IV.8.2.1 Abatement Options IV-159
IV.8.3 Results IV-162
IV.8.3.1 Data Tables and Graphs IV-162
IV.8.3.2 Global and Regional MACs and Analysis IV-165
IV.8.3.3 Uncertainties and Limitations IV-167
XIII
-------
Section Page
IV.8.4 References IV-168
IV.9 RFC and SF6 Emissions from Semiconductor Manufacturing IV-169
IV.9.1 Source Description IV-169
IV.9.1.1 Technology-Adoption Baseline IV-170
IV.9.1.2 No-Action Baseline IV-172
IV.9.2 Cost of PFC and SF6 Emissions Reduction from Semiconductor Manufacturing IV-173
IV.9.2.1 Abatement Options IV-173
IV.9.3 Results IV-179
IV.9.3.1 Data Tables and Graphs IV-179
IV.9.3.2 Global and Regional MACs and Analysis IV-179
IV.9.3.3 Uncertainties and Limitations IV-183
IV.9.4 References IV-185
IV.10 SF6 Emissions from Electric Power Systems IV-187
IV.10.1 Source Description IV-187
IV.10.1.1 Technology-Adoption Baseline IV-188
IV.10.1.2 No-Action Baseline IV-189
IV.10.2 Cost of SF6 Emissions Reduction from Electric Power Systems IV-190
IV.10.2.1 Abatement Options IV-190
IV.10.3 Results IV-195
IV.10.3.1 Data Tables and Graphs IV-195
IV.10.3.2 Global and Regional MACs and Analysis IV-198
IV.10.3.3 Uncertainties and Limitations IV-199
IV.10.4 References IV-202
IV.11 SF6 Emissions from Magnesium (Mg) Production IV-205
IV.11.1 Source Description IV-205
IV.11.1.1 Technology-Adoption Baseline IV-205
IV.11.1.2 No-Action Baseline IV-207
IV.11.2 Cost of SF6 Emissions Reduction from Mg Production and Processing Operations IV-208
IV.11.2.1 Abatement Options IV-208
IV.11.3 Results IV-210
IV.11.3.1 Data Tables and Graphs IV-210
IV.11.3.2 Global and Regional MACs and Analysis IV-213
IV.11.3.3 Uncertainties and Limitations IV-215
IV.11.4 References IV-216
XIV
-------
Section Page
Appendixes:
F: Cost and Emissions Reduction Analysis for Options to Abate International HFC
Emissions from Refrigeration and Air-Conditioning F-l
G: Cost and Emissions Reduction Analysis for Options to Abate International HFC,
HFE, and PFC Emissions from Solvents G-l
H: Cost and Emissions Reduction Analysis for Options to Abate International HFC
Emissions from Foams H-l
I: Cost and Emissions Reduction Analysis for Options to Abate International HFC
Emissions from Aerosols 1-1
J: Cost and Emissions Reduction Analysis for Options to Abate International HFC
Emissions from Fire Extinguishing J-l
K: Cost and Emissions Reduction Analysis for Options in Aluminum Production K-l
L: Cost and Emissions Reduction Analysis for Options in HCFC-22 Production L-l
M: Cost and Emissions Reduction Analysis for Options in Semiconductor
Manufacturing M-l
N: Cost and Emissions Reduction Analysis for Options in Electric Power Systems N-l
O: Cost and Emissions Reduction Analysis for Options in Magnesium Production O-l
V. Agriculture
V.1 Introduction and Background V-1
V.I.I Brief Points of Comparison with Other Non-CO2 Emissions Sectors V-2
V.1.2 Previous Estimates for EMF-21 and New Improvements V-2
V.2 Emissions Characterization, Baselines, and Mitigation Scenarios V-5
V.2.1 Croplands (N2O and Soil Carbon) V-5
V.2.1.1 Cropland N2O and Soil Carbon Emissions Characterization V-5
V.2.1.2 DAYCENT Baseline Estimates of Cropland N2O, Soil Carbon, and Yields V-6
V.2.1.3 Mitigation Options for Cropland N2O and Soil Carbon Emissions V-8
V.2.1.4 DAYCENT Results for Changes in Cropland N2O, Soil Carbon, and
Yields V-10
V.2.2 Rice (CH4, N2O, and Soil Carbon) V-12
V.2.2.1 Rice CH4, N2O, and Carbon Emissions Characterization V-12
V.2.2.2 DNDC Baseline Estimates of Rice CH4, N2O, Soil Carbon, and Yields V-13
V.2.2.3 Mitigation Options for Rice CH4, N2O, and Soil Carbon Emissions V-16
V.2.2.4 DNDC Estimates for Changes in Rice CH4, N2O, Soil Carbon, and Yields V-17
V.2.3 Livestock (CH4 and N2O) V-20
V.2.3.1 Livestock Enteric CH4 Emissions Characterization V-20
xv
-------
V.2.3.2 Livestock Manure CH4 and N2O Emissions Characterization V-20
V.2.3.3 The USEPA Baseline Estimates of Livestock Enteric CH4 Emissions V-21
V.2.3.4 The USEPA Baseline Estimates of Livestock Manure CH4 and N2O
Emissions V-21
V.2.3.5 Mitigation Options for Livestock Emissions V-21
V.2.3.6 Changes in Livestock CH4 and Productivity V-26
V.3 Results V-31
V.3.1 Estimating Average Costs and Constructing Abatement Curves V-31
V.3.2 Baselines, Mitigation Costs and MACs for Croplands V-32
V.3.3 Baselines, Mitigation Costs, and MACs for Rice Cultivation V-41
V.3.4 Baselines, Mitigation Costs, and MACs for Livestock Management V-45
V.3.5 Baselines, Mitigation Costs, and MACs for Total Agriculture V-55
V.3.6 Agricultural Commodity Market Impacts of Adopting Mitigation Options: Use of
the IMPACT Model V-57
V.4 Conclusions V-65
V.5 References V-69
Appendixes:
P: Summary of Non-CO2 Agricultural Mitigation Analysis Completed for EMF-21 P-l
Q: DAYCENT Model Description and Methods Q-l
R: DNDC Model Description and Methods R-l
S: Baseline Differences and Methods for This Mitigation Analysis S-l
T: IMPACT Commodity Price Data T-l
U: Detailed Data Tables U-l
XVI
-------
List of Figures
Number Page
Section I
2-1 Contribution of Anthropogenic Emissions of Greenhouse Gases to the Enhanced
Greenhouse Effect from Preindustrial to Present (measured in watts/meter2) 1-2
3-1 Illustrative Non-CO2 Marginal Abatement Curve 1-13
4-1 Percentage Share of Global Non-CO2 Emissions by Type of Gas in 2005 1-16
4-2 Non-CO2 Global Emissions Forecast to 2020 by Greenhouse Gas 1-17
4-3 Global Emissions by Major Sector for All Non-CO2 Greenhouse Gases 1-18
4-4 Projected World Emissions Baselines for Non-CO2 Greenhouse Gases, Including the
Top Emitting Regions 1-18
4-5 Global 2020 MACs for Non-CO2 Greenhouse Gases by Major Sector 1-19
4-6 Global 2020 MACs by Non-CO2 Greenhouse Gas Type 1-20
4-7 Global 2020 MACs for Non-CO2 Greenhouse Gases by Major Emitting Regions 1-20
Section II
1-1 CH4 Emissions from Coal Mining, by Country: 2000-2020 II-l
1-2 EMF MACs for Top Five Emitting Countries/Regions from Coal: 2020 11-11
2-1 CH4 from Natural Gas Systems by Country: 2000-2020 11-15
2-2 EMF MACs for Top Five Emitting Countries/Regions from Natural Gas: 2020 11-29
3-1 CH4 Emissions from Oil Production by Country: 2000-2020 11-31
3-2 EMF MACs for Top Five Emitting Countries/Regions from Oil: 2020 11-42
Section III
1-1 CH4 Emissions from Municipal Solid Waste by Country: 2000-2020 III-l
1-2 Components of CH4 Emissions from Landfills III-5
1-3 EMF MACs for Top Five Emitting Countries/Regions from Landfills: 2020 III-ll
2-1 CH4 Emissions from Wastewater by Country: 2000-2020 111-13
2-2 N2O Emissions from Wastewater by Country: 2000-2020 111-14
Section IV
1-1 N2O Emissions from Industrial Production by Country: 2000-2020 IV-1
1-2 EMF MACs for Top Five Emitting Country/Regions from Nitric Acid Production: 2020 IV-11
1-3 EMF MACs for Top Five Emitting Country/Regions from Adipic Acid Production: 2020.... IV-11
2-1 Baseline HFC Emissions from Refrigeration and Air-Conditioning by Region
(MtCO2eq) IV-27
2-2 2010 MAC for Refrigeration/Air-Conditioning, 10% Discount Rate, 40% Tax Rate IV-53
2-3 2020 MAC for Refrigeration/Air-Conditioning, 10% Discount Rate, 40% Tax Rate IV-54
3-1 Total Baseline HFC, PFC, and HFE Emissions Estimates from Solvents (MtCO2eq) IV-63
XVII
-------
Number Page
3-2 2010 MAC for Solvents, 10% Discount Rate, 40% Tax Rate IV-70
3-3 2020 MAC for Solvents, 10% Discount Rate, 40% Tax Rate IV-71
4-1 Total Baseline Emissions Estimates for Foams (MtCO2eq) IV-80
4-2 2010 MAC for Foams, 10% Discount Rate, 40% Tax Rate IV-99
4-3 2020 MAC for Foams, 10% Discount Rate, 40% Tax Rate IV-99
5-1 Total Baseline HFC Emissions Estimates from MDI Aerosols (MtCO2eq) IV-106
5-2 Total Baseline HFC Emissions Estimates from Non-MDI Aerosols (MtCO2eq) IV-107
5-3 2010 MAC for MDI Aerosols, 10% Discount Rate, 40% Tax Rate IV-114
5-4 2020 MAC for MDI Aerosols, 10% Discount Rate, 40% Tax Rate IV-115
5-5 2010 MAC for Non-MDI Aerosols, 10% Discount Rate, 40% Tax Rate IV-115
5-6 2020 MAC for Non-MDI Aerosols, 10% Discounl Rate, 40% Tax Rate IV-116
6-1 Baseline HFC Emissions from Fire Extinguishing by Region (MtCO2eq) IV-123
6-2 2010 MAC for Fire Extinguishing, 10% Discount Rate, 40% Tax Rate IV-135
6-3 2020 MAC for Fire Extinguishing, 10% Discount Rate, 40% Tax Rate IV-135
7-1 PFC Emissions from Aluminum Production Based on a Technology-Adoption
Scenario-1990-2020 (MtCO2eq) IV-141
7-2 PFC Emissions from Aluminum Production Based on a No-Action Scenario —1990-
2020 (MtCO2eq) IV-142
7-3 2010 and 2020 Global Technology-Adoption and No-Action MACs for Primary
Aluminum Production IV-150
7-4 2010 Regional Technology-Adoption MACs for Primary Aluminum Production IV-150
7-5 2020 Regional Technology-Adoption MACs for Primary Aluminum Production IV-151
8-1 HFC-23 Emissions from HCFC-22 Production Based on a No-Action Scenario -1990-
2020 (MtCO2eq) IV-157
8-2 HFC-23 Emissions from HCFC-22 Production Based on a Technology-Adoption
Scenario-1990-2020 (MtCO2eq) IV-158
8-3 2010 and 2020 Global Technology-Adoption and No-Action MACs for HCFC-22
Production IV-165
8-4 2010 Regional Technology-Adoption MACs IV-166
8-5 2020 Regional Technology-Adoption MACs IV-166
9-1 PFC Emissions from Semiconductor Manufacturing Based on a Technology-Adoption
Scenario-1990 through 2020 (MtCO2eq) IV-171
9-2 WSC and Non-WSC Countries' Contribution to Global PFC Emissions (MtCO2eq) IV-171
9-3 PFC Emissions from Semiconductor Manufacturing Based on a No-Action Scenario —
1990 through 2020 (MtCO2eq) IV-173
9-4 2010 Regional Technology-Adoption MACs for Semiconductor Manufacturing IV-184
9-5 2020 Regional Technology-Adoption MACs for Semiconductor Manufacturing IV-184
10-1 SF6 Emissions from Electric Power Systems on a Technology-Adoption Scenario —
1990-2020 (MtCO2eq) IV-189
10-2 SF6 Emissions from Electric Power Systems on a No-Action Scenario —1990-2020
(MtCO2eq) IV-190
10-3 2010 and 2020 Global Technology-Adoption and No-Action MACs for Electric Power
Systems IV-198
10-4 2010 Regional Technology-Adoption MACs for Electric Power Systems IV-200
XVIII
-------
Number Page
10-5 2020 Regional Technology-Adoption MACs for Electric Power Systems IV-200
11-1 SF5 Emissions from Mg Manufacturing Based on a Technology-Adoption Scenario —
1990-2020 (MtCO2eq) IV-207
11-2 SF6 Emissions from Mg Manufacturing Based on a No-Action Scenario —1990-2020
(MtC02eq) IV-208
11-3 2010 and 2020 Global Technology-Adoption and No-Action MACs for Mg Production IV-213
11-4 2010 Regional Technology-Adoption MACs IV-214
11-5 2020 Regional Technology-Adoption MACs IV-214
Section V
1-1 Global Cropland Yields for Baseline and Mitigation Options Estimated by DAYCENT,
2010 V-10
1-2 Global Net Greenhouse Gas (N2O and Soil Carbon) Cropland Emissions Estimated by
DAYCENT under Baseline and Mitigation Scenarios V-ll
1-3 Global Net Greenhouse Gas (CH4 and N2O) Livestock Emissions under Baseline and
Mitigation Scenarios, Assuming Full Adoption of Individual Options and Holding
Number of Animals Constant V-26
1-4 Global Net Greenhouse Gas (CH4 and N2O) Livestock Emissions under Baseline and
Mitigation Scenarios, Assuming Full Adoption of Individual Options and Holding
Production Constant V-27
1-5 Global Beef Production under Baseline and Mitigation Options, Assuming Full
Adoption of Individual Options and Holding the Number of Animals Constant V-28
1-6 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-28
1-7 Global MAC for Net Greenhouse Gas Emissions from Croplands, Holding Area
Constant, 2000-2020 V-37
1-8 Global MAC for Net Greenhouse Gas Emissions from Croplands, Holding Area
Constant, Allocating Adoption of Mitigation Strategies to the Three Most Effective
Options Only, 2000-2020 V-38
1-9 MAC for Net Greenhouse Gas Emissions from Cropland Management in the United
States, Holding Area Constant, 2000-2020 V-39
1-10 MAC for Net Greenhouse Gas Emissions from Cropland Management in the EU-15,
Holding Area Constant, 2000-2020 V-39
1-11 MAC for Net Greenhouse Gas Emissions from Cropland Management in the FSU,
Holding Area Constant, 2000-2020 V-40
1-12 MAC for Net Greenhouse Gas Emissions from Cropland Management in China,
Holding Area Constant, 2000-2020 V-40
1-13 Global MAC for Net Greenhouse Gas Emissions from Rice Cultivation, Holding Area
Constant, 2000-2020 V-43
1-14 MAC for Net Greenhouse Gas Emissions from Rice Cultivation in India, Holding Area
Constant, 2000-2020 V-44
1-15 MAC for Net Greenhouse Gas Emissions from Rice Cultivation in China, Holding Area
Constant, 2000-2020 V-44
XIX
-------
Number Page
1-16 Global MAC for Greenhouse Gas Emissions from Livestock Management, Holding
Number of Animals Constant, 2000-2020 V-50
1-17 Global MAC for Greenhouse Gas Emissions from Livestock Management, Holding
Production Constant, 2000-2020 V-50
1-18 MAC for Greenhouse Gas Emissions from Livestock Management in the United States,
Holding Number of Animals Constant, 2000-2020 V-51
1-19 MAC for Greenhouse Gas Emissions from Livestock Management in China, Holding
Number of Animals Constant, 2000-2020 V-51
1-20 MAC for Greenhouse Gas Emissions from Livestock Management in India, Holding
Number of Animals Constant, 2000-2020 V-52
1-21 MAC for Greenhouse Gas Emissions from Livestock Management in Brazil, Holding
Number of Animals Constant, 2000-2020 V-52
1-22 MAC for Greenhouse Gas Emissions from Livestock Management in the United States,
Holding Production Constant, 2000-2020 V-53
1-23 MAC for Greenhouse Gas Emissions from Livestock Management in China, Holding
Production Constant, 2000-2020 V-53
1-24 MAC for Greenhouse Gas Emissions from Livestock Management in India, Holding
Production Constant, 2000-2020 V-54
1-25 MAC for Greenhouse Gas Emissions from Livestock Management in Brazil, Holding
Production Constant, 2000-2020 V-54
1-26 Global MAC for Net Greenhouse Gas Emissions from Agriculture, Holding
Area/Animals Constant, 2000-2020 V-56
1-27 Global MAC for Net Greenhouse Gas Emissions from Agriculture, Holding Production
Constant, 2000-2020 V-57
1-28 Effect of Global Adoption of the Antimethanogen Vaccine Mitigation Option on World
Prices Using the IMPACT Model V-58
1-29 Effect of Global Adoption of the Antimethanogen Vaccine Mitigation Option on Global
Production Using the IMPACT Model V-59
1-30 Effect of Global Adoption of the Antimethanogen Vaccine Mitigation Option on Global
Number of Animals Using the IMPACT Model V-59
1-31 Effect of Global Adoption of the Shallow Flooding Mitigation Option on World Prices
Using the IMPACT Model V-61
1-32 Effect of Global Adoption of the Shallow Flooding Mitigation Option on Global
Production Using the IMPACT Model V-61
1-33 Effect of Global Adoption of the Shallow Flooding Mitigation Option on Global Rice
Area Using the IMPACT Model V-62
1-34 Net GHG Abatement under Global Adoption of the Antimethanogen Vaccine Option
with Number of Cattle Constant, Production Constant, and Market Adjustments Using
the IMPACT Model, 2010 V-62
1-35 Comparison of Net GHG Abatement from Rice Cultivation under Global Adoption of
the Shallow Flooding Mitigation Option with Area Constant, Production Constant, and
Market Adjustments Using the IMPACT Model, 2010 V-63
xx
-------
List of Tables
Number Page
Section I
2-1 Global Greenhouse Gas (GHG) Emissions for 2000 (MtCO2eq) 1-2
2-2 Global Warming Potentials 1-5
3-1 Abatement Potential Calculation for Mitigation Options 1-10
3-2 Financial Assumptions in Breakeven Price Calculations for Abatement Options 1-13
Section II
1-1 Historical Coal Mining Activity Data for Selected Countries (Million Metric Tons) II-4
1-2 IPCC Suggested Underground Emissions Factors for Selected Countries II-5
1-3 Historical Baseline Emissions for Coal Mine CH4 for Selected Countries (MtCO2eq) II-5
1-4 Projected Baseline Emissions for Coal Mine CH4 for Selected Countries (MtCO2eq) II-6
1-5 Summary of Average Abatement Costs and Benefits for U.S. Coal Mines (in 2000$) 11-7
1-6 Summary of Coal Mining Abatement Options Included in the Analysis II-9
1-7 Baseline Emissions by EMF Regional Grouping: 2000-2020 (MtCO2eq) 11-10
1-8 Coal Mining MACs for Countries Included in the Analysis 11-10
2-1 Natural Gas Industry Characterization 11-16
2-2 Natural Gas Production by Country and Region: 1980-2003 (Trillion Cubic Feet) 11-18
2-3 Natural Gas Consumption by Country and Region: 1980-2003 (Trillion Cubic Feet) 11-19
2-4 Projected Natural Gas Production by Country and Region: 2010-2025 (Trillion Cubic
Feet) 11-20
2-5 Projected Natural Gas Consumption by Country and Region: 2010-2025 (Trillion Cubic
Feet) 11-20
2-6 IPCC Estimated Emissions Factors from Natural Gas by Region 11-21
2-7 Baseline Emissions for Natural Gas Systems for Selected Countries: 1990-2000
(MtC02eq) 11-21
2-8 Projected Baseline Emissions for Natural Gas Systems for Selected Countries: 2005-
2020 (MtCO2eq) 11-23
2-9 Prevalence of Abatement Options by Infrastructure Component 11-23
2-10 Natural Gas MACs for Countries Included in the Analysis 11-27
2-11 Baseline Emissions by EMF Regional Grouping: 2000-2020 (MtCO2eq) 11-28
2-12 Natural Gas MACs for Countries Included in the Analysis 11-29
3-1 Oil Production by Country: 1990-2003 (MMbbl per Day) 11-34
3-2 Forecasted Oil Production for Selected Countries (MMbbl per Day, Unless Otherwise
Noted) 11-35
3-3 Forecasted Oil Consumption for Selected Countries (MMbbl per Day, Unless
Otherwise Noted) 11-36
3-4 IPCC Emissions Factors for Petroleum Systems in Select Regions 11-37
3-5 Baseline Emissions from Oil Production, by Country: 1990-2000 (MtCO2eq) 11-38
3-6 Projected Baseline Emissions from Oil Production by Country: 2005-2020 (MtCO2eq) II-38
3-7 Cost of Reducing CH4 Emissions from Oil 11-39
XXI
-------
Number Page
3-8 Percentage Abatement for CH4 for Selected Breakeven Price ($/rCO2eq): 2000 11-40
3-9 Baseline Emissions by EMF Regional Grouping: 2000-2020 (MtCO2eq) 11-41
3-10 Oil System MACs for Countries Included in the Analysis 11-41
Section III
1-1 CH4 Emissions from Municipal Solid Waste by Country: 1990-2000 (MtCO2eq) III-5
1-2 Projected Baseline CH4 Emissions from Municipal Solid Waste by Country: 2005-2020
(MtCO2eq) III-6
1-3 Components of Collection and Flaring and LFG Utilization Abatement Options III-7
1-4 Breakeven Prices of MSW Landfill Technology Options III-9
1-5 Baseline Emissions by EMF Regional Grouping: 2000-2020 (MtCO2eq) III-10
1-6 MSW Landfill MACs for Countries Included in the Analysis 111-10
2-1 CH4 Emissions from Wastewater by Country: 1990-2000 (MtCO2eq) 111-19
2-2 N2O Emissions from Wastewater by Country: 1990-2000 (MtCO2eq) 111-19
2-3 Projected Baseline CH4 Emissions from Wastewater by Country: 2005-2020 (MtCO2eq) 111-21
2-4 Projected Baseline N2O Emissions from Wastewater by Country: 2005-2020 (MtCO2eq) 111-21
Section IV
1-1 2003 Adipic Acid Production Capacity (Thousands of Metric Tons/Year) IV-3
1-2 IPCC Emissions Factors for Nitric Acid Production in Select Countries IV-4
1-3 N2O Emissions from Nitric and Adipic Acid Production: 1990-2000 (MtCO2eq) IV-5
1-4 Projected N2O Baseline Emissions from Nitric and Adipic Acid Production: 2005-2020
(MtCO2eq) IV-6
1-5 Cost of Reducing N2O Emissions from Industrial Processes IV-7
1-6 Projected N2O Emissions from Nitric Acid by Region: 2000-2020 (MtCO2eq) IV-9
1-7 Percentage Abatement for Nitric Acid for Selected Breakeven Prices ($/tCO2eq): 2010-
2020 IV-9
1-8 Projected N2O Emissions from Adipic Acid by Region: 2000-2020 (MtCO2eq) IV-10
1-9 Percentage Abatement for Adipic Acid for Selected Breakeven Prices ($/tCO2eq): 2010-
2020 IV-10
2-1 Reductions in Baseline Emissions in Non-U.S. Countries to Reflect Market Adjustments.... IV-21
2-2 Estimated Percentage of GWP-Weighted Refrigeration and Air-Conditioning HFC
Emissions Attributo MVACs in the United States IV-22
2-3 Percentage of Newly Manufactured Vehicles Assumed to Have Operational Air-
Conditioning Units in India IV-22
2-4 Percentage of Newly Manufactured Vehicles Assumed to Have Operational Air-
Conditioning Units in All Other Countries IV-23
2-5 Estimated Percentage of Refrigeration and Air-Conditioning HFC Emissions Attributo
MVACs IV-24
2-6 Distribution of Refrigeration- and Air-Conditioning-Sector HFC Emissions by End-
Use, Region, and Year (Percent) IV-26
2-7 Total Baseline HFC Emissions from Refrigeration and Air-Conditioning (MtCO2eq) IV-27
2-8 Assumptions on Duration and Applicability of Emissions Reduction Options IV-29
XXII
-------
Number Page
2-9 Summary of Assumptions for Leak Repair for Large Equipment IV-32
2-10 Summary of Assumptions for Recovery and Recycling from Small Equipment IV-33
2-11 Summary of Assumptions for Distributed Systems for New Stationary Equipment IV-37
2-12 Summary of Assumptions for HFC Secondary Loop Systems for New Stationary
Equipment IV-38
2-13 Summary of Assumptions for Ammonia Secondary Loop Systems for New Stationary
Equipment IV-39
2-14 Summary of Assumptions for Enhanced HFC-134a Systems for New MVACs IV-40
2-15 Summary of Assumptions for HFC-152a DX Systems in New MVACs IV-41
2-16 Summary of Assumptions for CO2 Systems in New MVACs IV-43
2-17 Summary of Technical Applicability of Abatement Options by Region (Percent) IV-47
2-18 Assumed Regional Market Penetration of Abatement Options into Newly
Manufactured Equipment, Expressed as a Percentage of Emissions from New
Equipment IV-48
2-19 Market Penetration of Abatement Options, Expressed as a Percentage of Total Sector
Emissions IV-49
2-20 Percentage of (Direct) Reduction Off Baseline Emissions of All Abatement Options by
Region IV-50
2-21 Summary of Abatement Option Cost Assumptions (2000$) IV-51
2-22 Country/Regional Emissions Reductions in 2010 and Breakeven Costs for
Refrigeration/Air-Conditioning at 10% Discount Rate, 40% Tax Rate ($/tCO2eq) IV-52
2-23 Country/Regional Emissions Reductions in 2020 and Breakeven Costs for
Refrigeration/Air-Conditioning at 10% Discount Rate, 40% Tax Rate ($/tCO2eq) IV-52
2-24 World Breakeven Costs and Emissions Reductions in 2020 for Refrigeration/Air-
Conditioning IV-53
3-1 General Overview of Solvent Technologies Used Globally IV-60
3-2 Total Baseline HFC, PFC, and HFE Emissions Estimates from Solvents (MtCO2eq) IV-62
3-3 Retrofit Techniques for Batch Vapor Cleaning Machine (Less than 13 Square Feet) IV-65
3-4 Technical Applicability and Incremental Maximum Market Penetration of Solvent
Options (Percent) IV-67
3-5 Emissions Reductions Off the Total Solvent Baseline (Percent) IV-68
3-6 Summary of Abatement Option Cost Assumptions IV-68
3-7 Country/Regional Emissions Reductions in 2010 and Breakeven Costs for Solvents at
10% Discount Rate, 40% Tax Rate ($/tCO2eq) IV-69
3-8 Country/Regional Emissions Reductions in 2020 and Breakeven Costs for Solvents at
10% Discount Rate, 40% Tax Rate ($/tCO2eq) IV-69
3-9 World Breakeven Costs and Emissions Reductions in 2020 for Solvents IV-70
4-1 The USEPA's Vintaging Model Emissions Profile for Foams' End-Uses IV-79
4-2 Baseline Emissions Estimates for Foams (MtCO2eq) IV-80
4-3 Reduction Efficiency of Foam Options (Percent) IV-90
4-4 Technical Applicability of Foam Options (Percent) IV-92
4-5 Incremental Maximum Market Penetration Expressed as a Percentage of New
Emissions for Which the Options Apply IV-93
4-6 Incremental Maximum Market Penetration Expressed as a Percentage of All Emissions IV-94
4-7 Emissions Reductions Off Total Foams Baseline (Percent) IV-95
XXIII
-------
Number Page
4-8 Summary of Abatement Option Cost Assumptions IV-96
4-9 Country/Regional Emissions Reductions in 2010 and Breakeven Costs for Foams at 10%
Discount Rate, 40% Tax Rate ($/tCO2eq) IV-97
4-10 Country/Regional Emissions Reductions in 2020 and Breakeven Costs for Foams at 10%
Discount Rate, 40% Tax Rate ($/tCO2eq) IV-97
4-11 World Breakeven Costs and Emissions Reductions in 2020 for Foams IV-98
5-1 Total Baseline HFC Emissions Estimates from MDI Aerosols (MtCO2eq) IV-106
5-2 Total Baseline HFC Emissions Estimates from Non-MDI Aerosols (MtCO2eq) IV-107
5-3 Technical Applicability and Incremental Maximum Market Penetration of Aerosol
Options (Percent) IV-110
5-4 Emissions Reductions Off the Total Applicable Aerosols Baseline (Percent) IV-110
5-5 Summary of Abatement Option Cost Assumptions IV-111
5-6 Country/Regional Emissions Reductions in 2010 and Breakeven Costs for MDI
Aerosols at 10% Discount Rate, 40% Tax Rate ($/tCO2eq) IV-112
5-7 Country/Regional Emissions Reductions in 2020 and Breakeven Costs for MDI
Aerosols at 10% Discount Rate, 40% Tax Rate ($/tCO2eq) IV-112
5-8 Country/Regional Emissions Reductions in 2010 and Breakeven Costs for Non-MDI
Aerosols at 10% Discount Rate, 40% Tax Rate ($/tCO2eq) IV-113
5-9 Country/Regional Emissions Reductions in 2020 and Breakeven Costs for Non-MDI
Aerosols at 10% Discount Rate, 40% Tax Rate ($/tCO2eq) IV-113
5-10 World Breakeven Costs and Emissions Reductions in 2020 for Aerosols IV-114
6-1 Total Baseline HFC Emissions from Fire Extinguishing (MtCO2eq) IV-123
6-2 Assumed Breakout of Total GWP-Weighted Baseline Fire-Extinguishing Emissions
(Percent) IV-124
6-3 Summary of Technical Applicability of Abatement Options (Percent) IV-130
6-4 Assumed Incremental Market Penetration of Abatement Options into Newly Installed
Class A or Class B Extinguishing Systems, Expressed as a Percentage of Emissions from
All New Equipment IV-131
6-5 Market Penetration of Abatement Options into Newly Installed Class A or Class B
Extinguishing Systems, Expressed as a Percentage of Total Sector Emissions IV-132
6-6 Percentage of Emissions Reductions Off Total Fire-Extinguishing Baseline IV-132
6-7 Summary of Abatement Option Cost Assumptions (2000$) IV-132
6-8 Country/Regional Emissions Reductions in 2010 and Breakeven Costs for Fire
Extinguishing at 10% Discount Rate, 40% Tax Rate ($/tCO2eq) IV-133
6-9 Country/Regional Emissions Reductions in 2020 and Breakeven Costs for Fire
Extinguishing at 10% Discount Rate, 40% Tax Rate ($/tCO2eq) IV-134
6-10 World Breakeven Costs and Emissions Reductions in 2020 for Fire Extinguishing IV-134
7-1 Total PFC Emissions from Aluminum Manufacturing (MtCO2eq)—No-Action Baseline... IV-140
7-2 Total PFC Emissions from Aluminum Manufacturing (MtCO2eq) —Technology-
Adoption Baseline IV-140
7-3 Reduction Efficiency Potential for Abatement Option by Cell Type (Percent) IV-143
7-4 Average Baseline Market Penetration of Complete Retrofits by Cell Type and Scenario
(Percent) IV-145
XXIV
-------
Number Page
7-5 Emissions Reductions in 2010 and Breakeven Costs ($/tCO2eq) for Aluminum
Production at 10% Discount Rate, 40% Tax Rate (MtCO2eq)-No-Action Baseline IV-146
7-6 Emissions Reductions in 2020 and Breakeven Costs ($/tCO2eq) for Aluminum
Production at 10% Discount Rate, 40% Tax Rate (MtCO2eq)-No-Action Baseline IV-147
7-7 Emissions Reductions in 2010 and Breakeven Costs ($/tCO2eq) for Aluminum
Production at 10% Discount Rate, 40% Tax Rate (MtCO2eq)-Technology-Adoption
Baseline IV-147
7-8 Emissions Reductions in 2020 and Breakeven Costs ($/tCO2eq) for Aluminum
Production at 10% Discount Rate, 40% Tax Rate (MtCO2eq)-Technology-Adoption
Baseline IV-148
7-9 Emissions Reduction and Costs in 2020—No-Action Baseline IV-148
7-10 Emissions Reduction and Costs in 2020—Technology-Adoption Baseline IV-149
8-1 Total HFC-23 Emissions from HCFC-22 Production (MtCO2eq)-No-Action Baseline IV-155
8-2 Total HFC-23 Emissions from HCFC-22 Production (MtCO2eq) — Technology-Adoption
Baseline IV-156
8-3 Baseline Market Penetration of Thermal Oxidation—No-Action Baseline IV-161
8-4 Baseline Market Penetration of Thermal Oxidation—Technology-Adoption Baseline IV-161
8-5 Emissions Reductions in 2010 and Breakeven Costs ($/tCO2eq) for HFC-23 Emissions
from HCFC-22 Production at 10% Discount Rate, 40% Tax Rate (MtCO2eq)-No-Action
Baseline IV-162
8-6 Emissions Reductions in 2020 and Breakeven Costs ($/tCO2eq) for HFC-23 Emissions
from HCFC-22 Production at 10% Discount Rate, 40% Tax Rate (MtCO2eq)-No-Action
Baseline IV-163
8-7 Emissions Reductions in 2010 and Breakeven Costs ($/tCO2eq) for HFC-23 Emissions
from HCFC-22 Production at 10% Discount Rate, 40% Tax Rate (MtCO2eq)-
Technology-Adoption Baseline IV-163
8-8 Emissions Reductions in 2020 and Breakeven Costs ($/tCO2eq) for HFC-23 Emissions
from HCFC-22 Production at 10% Discount Rate, 40% Tax Rate (MtCO2eq)-
Technology-Adoption Baseline IV-164
8-9 World Breakeven Costs and Emissions Reductions in 2020 —No-Action Baseline IV-164
8-10 World Breakeven Costs and Emissions Reductions in 2020 —Technology-Adoption
Baseline IV-164
9-1 Total PFC Emissions from Semiconductor Manufacturing (MtCO2eq) —No-Action
Baseline IV-169
9-2 Total PFC Emissions from Semiconductor Manufacturing (MtCO2eq)—Technology-
Adoption Baseline IV-170
9-3 Maximum Market Penetrations for WSC Countries in the No-Action Baseline (Percent)... IV-174
9-4 Maximum Market Penetrations for Non-WSC Countries in the No-Action Baseline
(Percent) IV-174
9-5 Baseline Market Penetrations for WSC Countries in the Technology-Adoption Baseline
(Percent) IV-175
9-6 Maximum Market Penetrations for WSC Countries in the Technology-Adoption
Baseline (Percent) IV-175
XXV
-------
Number Page
9-7 Baseline Market Penetrations for Non-WSC Countries in the Technology-Adoption
Baseline in 2020 (Percent) IV-175
9-8 Maximum Market Penetrations for Non-WSC Countries in the Technology-Adoption
Baseline (Percent) IV-176
9-9 Emissions Reductions in 2010 and Breakeven Costs ($/tCO2eq) at 10% Discount Rate,
40% Tax Rate (MtCO2eq)-No-Action Baseline IV-180
9-10 Emissions Reductions in 2020 and Breakeven Costs ($/tCO2eq) at 10% Discount Rate,
40% Tax Rate (MtCO2eq) - No-Action Baseline IV-180
9-11 Emissions Reductions in 2010 and Breakeven Costs ($/tCO2eq) at 10% Discount Rate,
40% Tax Rate (MtCO2eq)-Technology-Adoption Baseline IV-181
9-12 Emissions Reductions in 2020 and Breakeven Costs ($/tCO2eq) at 10% Discount Rate,
40% Tax Rate (MtCO2eq)-Technology-Adoption Baseline IV-181
9-13 Emissions Reduction and Costs in 2020—No-Action Baseline IV-182
9-14 Emissions Reduction and Costs in 2020—Technology-Adoption Baseline IV-182
10-1 Total SF6 Emissions from Electric Power Systems (MtCO2eq) — No-Action Baseline IV-187
10-2 Total SF6 Emissions from Electric Power Systems (MtCO2eq) —Technology-Adoption
Baseline IV-188
10-3 Emissions Reductions in 2010 and Breakeven Costs ($/tCO2eq) for Electric Power
Systems at a 10% Discount Rate, 40% Tax Rate (MtCO2eq)—No-Action Baseline IV-195
10-4 Emissions Reductions in 2020 and Breakeven Costs ($/tCO2eq) for Electric Power
Systems at 10% Discount Rate, 40% Tax Rate (MtCO2eq) —No-Action Baseline IV-196
10-5 Emissions Reductions in 2010 and Breakeven Costs ($/tCO2eq) for Electric Power
Systems at 10% Discount Rate, 40% Tax Rate (MtCO2eq)—Technology-Adoption
Baseline IV-196
10-6 Emissions Reductions in 2020 and Breakeven Costs ($/tCO2eq) for Electric Power
Systems at 10% Discount Rate,, 40% Tax Rate (MtCO2eq)-Technology-Adoption
Baseline IV-197
10-7 Emissions Reduction and Costs in 2020—No-Action Baseline IV-197
10-8 Emissions Reduction and Costs in 2020 —Technology-Adoption Baseline IV-198
11-1 Total SF6 Emissions from Mg Manufacturing (MtCO2eq) — No-Action Baseline IV-206
11-2 Total SF6 Emissions from Mg Manufacturing (MtCO2eq) —Technology-Adoption
Baseline IV-206
11-3 Emissions Reductions in 2010 and Breakeven Costs ($/tCO2eq) for Mg Production at
10% Discount Rate, 40% Tax Rate (MtCO2eq) —No-Action Baseline IV-210
11-4 Emissions Reductions in 2020 and Breakeven Costs ($/tCO2eq) for Mg Production at
10% Discount Rate, 40% Tax Rate (MtCO2eq)-No-Action Baseline IV-211
11-5 Emissions Reductions in 2010 and Breakeven Costs ($/tCO2eq) for Mig Production at
10% Discount Rate, 40% Tax Rate (MtCO2eq)—Technology-Adoption Baseline IV-211
11-6 Emissions Reductions in 2020 and Breakeven Costs ($/tCO2eq) for Mg Production at
10% Discount Rate, 40% Tax Rate (MtCO2eq)-Technology-Adoption Baseline IV-212
11-7 Emissions Reduction and Costs in 2020—No-Action Baseline IV-212
11-8 Emissions Reduction and Costs in 2020—Technology-Adoption Baseline IV-212
XXVI
-------
Number Page
Section V
1-1 DAYCENT N2O and Soil Carbon Estimates for 2000, 2010, and 2020 by Key Region
(MtCO2eq/yr) V-6
1-2 Cropland N2O and Soil Carbon Mitigation Options Run Through DAYCENT V-9
1-3 Rice-Only Baseline CH4/ N2O, and Soil Carbon Estimates for 2000, 2010, and 2020 by
Asian Region (Midpoints from DNDC in MtCO2eq/yr; Negative Carbon Numbers
Indicate Net Sequestration) V-15
1-4 Rice CH4, N2O, and Soil Carbon Mitigation Options Run Through DNDC V-17
1-5 DNDC Estimates of Net Greenhouse Gas Results for Baseline and Mitigation Scenarios
for China (Annual Averages in MtCO2eq/yr over 2000-2020) V-17
1-6 Changes from Baseline in Greenhouse Gas Emissions, Crop Yields, and Water
Consumption for China (Annual Averages over 2000-2020; Negative Numbers Indicate
Decreases Relative to the Baseline) V-18
1-7 Net Greenhouse Gas Results for Baseline and Mitigation Options for Other Asian
Countries (Annual Averages in MtCO2eq/yr over 2000-2020) V-19
1-8 Livestock Enteric Fermentation Greenhouse Gas Mitigation Options V-23
1-9 Livestock Manure Management Greenhouse Gas Mitigation Options V-25
1-10 Baseline Net GHG Emissions from Croplands from DAYCENT Estimates (MtCO2eq) V-33
1-11 Croplands Mitigation Option Detail for Key Regions V-34
1-12 Croplands: Percentage Reductions from Baselines at Different $/tCO2eq Prices V-37
1-13 Baseline Emissions from Rice Cultivation from DNDC Estimates (MtCO2eq) V-41
1-14 Rice Cultivation Mitigation Option Detail for Key Regions V-42
1-15 Rice Cultivation: Percentage Reductions from Baseline at Different $/tCO2eq Prices V-43
1-16 Baseline Emissions from Livestock Management from USEPA (2006) (MtCO2eq) V-45
1-17 Livestock Mitigation Option Detail for Key Regions V-46
1-18 Livestock Management: Percentage Reductions from Baselines at Different $/tCO2eq
Prices V-49
1-19 Baseline Emissions from All Agriculture Used in This Report (MtCO2eq) V-55
1-20 Total Agriculture: Percentage Reductions from Baseline at Different $/tCO2eq Prices V-56
XXVII
-------
XXVIII
-------
Executive Summary
-------
EXECUTIVE SUMMARY
ES-ii GLOBAL MITIGATION OF NON-CO, GREENHOUSE GASES
-------
EXECUTIVE SUMMARY
he mitigation of noncarbon dioxide (non-CO2) greenhouse gas emissions can be a relatively
inexpensive supplement to CO2-only mitigation strategies. The non-CO2 gases include
methane (CH4), nitrous oxide (N2O), and a number of high global warming potential (high-
GWP) or fluorinated gases. These gases trap more heat within the atmosphere than CO2 per unit weight.
Approximately 30 percent of the anthropogenic greenhouse effect since preindustrial times can be
attributed to these non-CO2 greenhouse gases (Intergovernmental Panel for Climate Change [IPCC],
2001b); approximately 24 percent of GWP-weighted greenhouse gas emissions in the year 2000 are
comprised of the non-CO2 greenhouse gases (de la Chesnaye et al., in press; U.S. Environmental
Protection Agency [USEPA], 2006).
Given the important role that mitigation of non-CO2 greenhouse gases can play in climate strategies,
there is a clear need for an improved understanding of the mitigation potential for non-CO2 sources, as
well as for the incorporation of non-CO2 greenhouse gas mitigation in climate economic analyses. This
report provides a comprehensive global analysis and resulting data set of marginal abatement curves
(MACs) that illustrate the abatement potential of non-CO2 greenhouse gases by sector and by region. This
assessment of mitigation potential is unique because it is comprehensive across all non-CO2 gases, across
all emitting sectors of the economy, and across ail regions of the world.
The analysis in this report is the latest refinement of the methodology on mitigation of various non-
CO2 gases, which has been underway since 1999. A significant contribution to the climate change
mitigation literature is Stanford University's Energy Modeling Forum Working Group 21 (EMF-21),
which focused on mitigation of multiple greenhouse gases and resulted in the publication of a special
issue of the Energy Journal (see Weyant and de la Chesnaye, in press). The specific non-CO2 mitigation
papers in the EMF-21 study include energy- and industry-related CH4 and N2O (Delhotal et al., in press);
agricultural-related CH4 and N2O (DeAngelo et al., in press); and industry-related fluorinated gases
(Ottinger et al., in press). Much of the original work comes from three previous USEPA studies for the
United States (2006, 2001, 1999) and a study conducted by the European Commission (EC) (2001) that
evaluated technologies and costs of CH4 abatement for European Union (EU) members from 1990 to 2010.
These studies provided estimates of potential CH4 and N2O emissions reductions from major emitting
sectors and quantified costs and benefits of these reductions.
Building on the baseline non-CO2 emissions projections from the USEPA's Global Anthropogenic Non-
CO2 Greenhouse Gas Emissions: 1990-2020 (2006), this analysis applies mitigation options to the emissions
baseline in each sector. Across all the emitting greenhouse gas sectors, for each mitigation option, the
technical abatement potential and cost are calculated. The MACs are determined by the series of
breakeven price calculations for the suite of available options for each sector and region. Each point along
the curve indicates the abatement potential given the economically feasible mitigation technologies at a
given breakeven price. 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.
The result of these efforts is a set of MACs that allow for improved understanding of the mitigation
potential for non-CO2 sources, as well as inclusion of non-CO2 greenhouse gas mitigation in economic
modeling. The MAC data sets can be downloaded in spreadsheet format from the USEPA Web site at
.
Highlights of this analysis include the following:
GLOBAL MITIGATION OF NON-CO, GREENHOUSE GASES ES-1
-------
EXECUTIVE SUMMARY
Mitigation of Non-CO2 Gases Can Play an Important Role in Climate Strategies. Worldwide, the
potential for "no-regret" non-CO2 greenhouse gas abatement is significant. Figure ES-1 shows the global
total aggregate MAC for the year 2020. Without a price signal (i.e., at $0/tCO2eq), the global mitigation
potential is greater than 600 million metric tons of CO2 equivalent (MtCO2eq), or 5 percent 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 breakeven price rises, the mitigation potential grows. Significant
mitigation opportunities could be realized in the lower range of breakeven prices. The global mitigation
potential at a price of $10/tCO2eq is greater than 2,000 MtCO2eq, or 15 percent of the baseline emissions,
and greater than 2,185 MtCO2eq or 17 percent of the baseline emissions at $20/tCO2eq. In the higher range
of breakeven 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 2020
i 1 1 1 r
500 1,000 1,500 2,000 2,500 3,000 3,500 4,000
Non-CO2 Reduction (MtCO2eq)
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 2020. At a
breakeven price of $30/tCO2eq, the potential for reduction of non-CO2 greenhouse gases is nearly 1,000
MtCO2eq in the energy sector, and approximately 600 MtCO2eq in the agriculture sector. While less than
that of the energy and agriculture sectors, mitigation potential in the waste and industrial processes
sectors can play an important role, particularly in the absence of a carbon price incentive.
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 2020. At or below $0/tCO2eq, the
potential for CH4 mitigation is approximately 500 MtCO2eq. The potential for reducing CH4 emissions
grows to nearly 1,800 MtCO2eq as the breakeven price rises from $0 to $30/tCO2eq. While less than that of
CH4, N2O and high-GWP gases exhibit significant mitigation potential at or below $0/tCO2eq.
ES-2
GLOBAL MITIGATION OF NON-CO, GREENHOUSE GASES
-------
EXECUTIVE SUMMARY
Figure ES-2: Global 2020 MACs for Non-C02 Greenhouse Gases by Major Sector
$60 -i
$50 -
$40 -
$30 -
$20 -
$10
$0
-$10 :
-$20 -
-$30
-$40 J
0>
w
O
O
Energy
Agriculture
Waste
Industrial Processes
250
500
750
1,000
1,250
1,500
Non-CO2 Reduction (MtCO2eq)
Figure ES-3: Global 2020 MACs by Non-C02 Greenhouse Gas Type
$60 -
$50 -
$40
$30
$20 -
$10 1
$0
CT
0)
CM
O
o
-$10 -
-$20 -
-$30
-$40 J
J
500
Methane
Nitrous Oxide
High-GWP
1,000
1,500
2,000
2,500
3,000
Non-CO2 Reduction (MtCO2eq)
GLOBAL MITIGATION OF NON-CO, 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 2020. China, the United States, EU, India, and Brazil are the
countries or regions that emit the most non-CO2 greenhouse gases. As the largest emitters, they also offer
important mitigation opportunities. These regions show significant mitigation potential in the lower
range of breakeven prices, with the MACs getting steeper in the higher range of breakeven prices as each
additional ton of emissions becomes more expensive to reduce.
Figure ES-4: Global 2020 MACs for Non-C02 Greenhouse Gases by Major Emitting Regions
$60 -I
$50 -
$40 -
$30 -
o- $20-
0 $10 -
4-1
^ $0 -
(
-$10 -
I'l A • ''"
j I j — * China ,f
| United States /
. • EU-15 /
-- ~ Brazil '
*• India
; * Rest of the world
;X-^
If ,, 250 500 750 1,000 1,250 1,500
ill
-$20 I,
-$30 -II
-$40 J'
Non-CO2 Reduction (MtCO2eq)
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.
ES-4
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
I. Technical Summary
-------
SECTION I — TECHNICAL SUMMARY
l-ii GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION I — TECHNICAL SUMMARY
1.1 Overview
he objective of this report is to provide a comprehensive and consistent data set on global
mitigation of noncarbon dioxide (non-CO2) greenhouse gases to facilitate multigas analysis of
climate change issues. Mitigating emissions of non-CO2 greenhouse gases can be relatively
inexpensive compared with mitigating CO2 emissions. Thus, attention has been focused on incorporating
international non-CO2 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 builds on a study previously conducted by the U.S. Environmental Protection Agency
(USEPA) for the Energy Modeling Forum, Working Group 21 (EMF-21). The Energy Modeling Forum
was established by Stanford University to explore energy and environmental issues through the
collaboration of diverse modeling teams from around the world. The EMF-21 focused specifically on
multigas strategies to address climate change and resulted in the publication of a special issue of the
Energy Journal (see Weyant and de la Chesnaye [in press]). The specific non-CO2 mitigation papers in the
EMF-21 study include energy- and industry-related methane (CH4) and nitrous oxide (N2O) (Delhotal et
al., in press), agricultural-related CH4and N2O (DeAngelo et al, in press), and industry-related
fluorinated gases (Ottinger et al., in press). Much of the original work comes from two previous USEPA
studies for the United States (USEPA, 2001, 1999) and a study conducted by the European Commission
(EC) (2001) that evaluated technologies and costs of CH4 abatement for EU members from 1990 to 2010.
Following the basic methodology of the EMF-21 study with some enhancements (as described in
Section 1.3.4 of this report), this report contains detailed analyses by economic sector and region for all
non-CO2 greenhouse gases over the period from 2000 to 2020. 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-CO2 sources, as well as inclusion of non-CO2 greenhouse gas mitigation in economic modeling. The
MAC data sets can be downloaded in spreadsheet format from the USEPA's Web site at
.
1.2 Non-COa Greenhouse Gases
reenhouse gases other than CO2 play an important role in the effort to understand and
address global climate change. The non-CO2 gases include CH4, N2O, and a number of high
global warming potential or fluorinated gases. The non-CO2 greenhouse gases are more
potent than CO2 (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 CO2. Figure 2-1 shows that these non-
CO2 greenhouse gases are responsible for approximately 30 percent of the enhanced, anthropogenic
greenhouse effect since preindustrial times.
Table 2-1 shows the global total greenhouse gas emissions for the year 2000, broken down by sector
and by greenhouse gas type. The non-CO2 gases constitute 24 percent of the global total greenhouse gas
emissions in 2000.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 1-1
-------
SECTION I — TECHNICAL SUMMARY
Figure 2-1:
Contribution of Anthropogenic Emissions of Greenhouse Gases to the Enhanced
Greenhouse Effect from Preindustrial to Present (measured in watts/meter2)
High-GWP
Gases
0.4%
Source: IPCC, 2001b. Note that gases regulated under the Montreal Protocol are excluded.
Table 2-1: Global Greenhouse Gas (GHG) Emissions for 2000 (MtC02eq)
Sectors
Energy
Agriculture
Industry
Waste
Global Total
Percentage of Global Total GHGs
C02
23,408
7,631
829
31,868
77%
CH4
1,646
3,113
6
1,255
6,021
15%
N20
237
2,616
155
106
3,114
8%
High-
GWP
380
380
1%
Global
Total
25,291
13,360
1,370
1,361
41,382
Percentage
of Global
Total GHGs
61%,
32%
3%
3%
Source: Adapted from de la Chesnaye et al., in press; USEPA, 2006.
1.2.1 Methane (CH4)
CH4 is about 21 times more powerful at warming the atmosphere than CO2 over a 100-year period.1
In addition, CH4's chemical lifetime in the atmosphere is approximately 12 years, compared with
approximately 100 years for CO2. These two factors make CH4 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 CH4 could respond to mitigation actions.
1 Per IPCC (1996) guidelines. The GWP of methane in the IPCC Third Assessment Report (2001a) is 23.
1-2
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION I —TECHNICAL SUMMARY
CH4 is emitted from a variety of manmade sources, including landfills, natural gas and petroleum
systems, agricultural activities, coal mining, stationary and mobile combustion, wastewater treatment,
and certain industrial processes. CH4 is also a primary constituent of natural gas and an important energy
source. As a result, efforts to prevent or capture and use CH4 emissions can provide significant energy,
economic, and environmental benefits.
The historical record, based on analysis of air bubbles trapped in glaciers, indicates that CH4 is more
abundant in the Earth's atmosphere now than at any time during the past 400,000 years (National
Research Council [NRC], 2001). Since 1750, global average atmospheric concentrations of CH4 have
increased 150 percent, from approximately 700 to 1,745 parts per billion by volume (ppbv)
(Intergovernmental Panel for Climate Change [IPCC], 2001a). Although CH4 concentrations have
continued to increase, the overall rate of CH4 growth during the past decade has slowed. In the late 1970s,
the growth rate was approximately 20 ppbv per year. In the 1980s, growth slowed to 9 to 13 ppbv per
year. From 1990 to 1998, CH4 saw variable growth between 0 and 13 ppbv per year (IPCC, 2001a). A
recent study by Dlugokencky et al. (2003) shows that atmospheric CH4 was at a steady state of 1,751 ppbv
between 1999 and 2002.
Once emitted, CH4 is removed from the atmosphere by a variety of processes, frequently called sinks.
The balance between CH4 emissions and CH4 removal processes ultimately determines atmospheric CH4
concentrations and determines the length of time CH4 emissions remain in the atmosphere. The dominant
sink is oxidation within the atmosphere by chemical reaction with hydroxyl radicals (OH). Methane
reacts with OH to produce alkyd radicals (CH3) and water in the tropospheric layer of the atmosphere.
Stratospheric oxidation also plays a minor role in removing CH4 from the atmosphere. Similar to
tropospheric oxidation, in stratospheric oxidation, minor amounts of CH4 are destroyed by reacting with
OH in the stratosphere. These two reactions account for almost 90 percent of CH4 removal (IPCC, 2001 c).
Other known sinks include microbial uptake of CH4 in soils and the reaction of CH4 with chlorine (Cl)
atoms in the marine boundary layer. It is estimated that these two sinks contribute 7 percent and less than
2 percent of total CH4 removal, respectively.
1.2.2 Nitrous Oxide (N2O)
N2O is a clear, 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 CO2 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. On a global basis, it is estimated that natural
sources account for over 60 percent of total N2O emissions (IPCC, 2001c).
Global atmospheric concentrations of N2O have increased from about 270 ppbv in 1750 to 314 ppbv
in 1998, which equates to a 16 percent increase. In the last 2 decades, atmospheric concentrations of N2O
continue to increase at a rate of 0.25 percent per year. There has been a significant multiyear variance in
observed growth of N2O concentrations, but the reasons for these trends are not fully understood yet
(IPCC, 2001b).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION I —TECHNICAL SUMMARY
1.2.3 High-GWP Gases
There are three major groups or types of high-GWP gases: hydrofluorocarbons (HFCs),
perfluorocarbons (PFCs), and sulfur hexafluoride (SFfl). These compounds are the most potent
greenhouse gases because of their large heat-trapping capacity and, in the cases of SF6 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. High-GWP gases are emitted from
a broad range of industrial sources; most of these gases have few (if any) natural sources.
1.2.3.1 MFCs
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 (HF'C-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).
The HFCs with the largest measured atmospheric abundances are (in order) HFC-23 (CHF3), HFC-
134a (CF3CH2F), and HFC-152a (CH3CHF2). The only significant emissions of HFCs before 1990 were
from HFC-23, which is generated as a by-product during the production of HCFC-22. Between 1978 and
1995, HFC-23 concentrations increased from 3 to 10 parts per trillion (ppt), and these concentrations
continue to rise. In 1990, HFCs other than HFC-23 were almost undetectable; today, global average
concentrations of HFC-134a have risen significantly to almost 10 ppt. HFC-134a has an atmospheric
lifetime of about 14 years and its abundance is expected to continue to rise in line with its increasing use
as a refrigerant around the world. HFC-152a has increased steadily to about 0.3 ppt in 2000; however, its
relatively short lifetime (1.4 years) has kept its atmospheric concentration below 1 ppt (IPCC, 2001 a).
1.2.3.2 PFCs
Primary aluminum production and semiconductor manufacture are the largest known manmade
sources of tetrafluoromethane (CF4) and hexafluoroethane (C2F6). PFCs are also relatively minor
substitutes for ODSs. Over a 100-year period, CF4 and C2F6 are, respectively, 6,500 and 9,200 times more
effective than CO2 at trapping heat in the atmosphere.
PFCs have extremely stable molecular structures and are largely immune to the chemical processes in
the lower atmosphere that break down most atmospheric pollutants. Not until the PFCs reach the
mesosphere, about 60 kilometers above Earth, are they destroyed by very high-energy ultraviolet rays
from the sun. This removal mechanism is extremely slow; as a result, PFCs accumulate in the atmosphere
and remain there for several thousand years. The estimated atmospheric lifetimes for CF4 and C2F6
emissions are 50,000 and 10,000 years, respectively. Measurements in 2000 estimated CF4 global
concentrations in the stratosphere at over 70 ppt. Recent relative rates of concentration increase for these
two important PFCs are 1.3 percent per year for CF4 and 3.2 percent per year for C2F6 (IPCC, 2001a).
1.2.3.3 Sulfur Hexaflouride (SF6)
The GWP of SF6 is 23,900, making it the most potent greenhouse gas evaluated by IPCC. SF6 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 clean vapor deposition chambers during manufacture of
1-4 GLOBAL MITIGATION OF NO'N-C02 GREENHOUSE GASES
-------
SECTION I — TECHNICAL SUMMARY
semiconductors and flat panel displays; and (4) for a variety of smaller uses, including uses as a tracer gas
and as a filler for sound-insulated windows.
Like the PFCs, SF6 is very long lived, so all manmade sources contribute directly to its accumulation in
the atmosphere. Measurements of SF6 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, CO2 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 2-2 lists all GWPs
used in this report to convert the non-CO2 emissions into CO2-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 2-2: Global Warming Potentials
Gas
Carbon dioxide (C02)
Methane (CH4)
Nitrous oxide (N20)
HFC-23
HFC-125
HFC-134a
HFC-143a
HFC-1523
HFC-227ea
HFC-236fe
HFC4310mee
CF4
CzF6
C/w
C6F14
SF6
GWP
1
21
310
11,700
2,800
1,300
3,800
140
2,900
6,300
1,300
6,500
9,200
7,000
7,400
23,900
1.3 Methodology
his section describes the basic methodology used in this report to analyze potential emissions
and abatement of non-CO2 greenhouse gases. In this analysis we construct MAC curves 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. This section describes the components of our methodology. First, we
establish the baseline emissions for each sector in Section 1.3.1. Then we describe the methodology used to
evaluate mitigation options in Section 1.3.2, which involves calculating the abatement potential and the
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 1-5
-------
SECTION I — TECHNICAL SUMMARY
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 or
within the 2000 to 2020 time frame. In some cases, sensitivities to the MACs are presented where the
discount rate, tax rate, and energy prices vary. Emissions abatement in the MACs is shown as both
absolute emissions reductions and as percentage reductions from the baseline. Non-CO2 emissions
sources analyzed in this report are coal mining; natural gas production, processing, transmission, and
distribution; oil production; solid waste management; wastewater; specialized industrial processes; and
agriculture.
1.3.1 Baseline Emissions for Non-CQ2 Greenhouse Gases
Current and projected (through 2020) emissions estimates are based primarily on emissions
projections from the USEPA's Global Anthropogenic Non-CO2 Greenhouse Gas Emissions: 1990-2020 (USEPA,
2006). The methods used to estimate and project non-CO2 emissions in USEPA (2006) are briefly
summarized here. In some cases, particularly for the fluorinated gas emissions and agricultural
emissions, it was necessary to develop separate baselines from which to assess the mitigation analyses.
These deviations are also explained in this report.
For Annex I countries,2 baseline (i.e., reference) projections are based largely on publicly available
reports produced by the countries themselves. The preferred sources for these reports are the National
Communications for the United Nations Framework Convention on Climate Change,3 which contain
current emissions rates and emissions projections through 2020. Estimates from the various countries
should be comparable because they rely on the same (or similar) IPCC methodologies and country-
specific activity data.
Estimates of historical and projected emissions for developing countries were based on national and
international reports. These emissions rates also reflect the most recent results of the USEPA study Global
Anthropogenic Non-CO2 Greenhouse Gas Emissions: 1990-2020 (USEPA, 2006). The preferred approach to
estimate emissions from developing countries is to use the latest published information for each country.
Some developing countries reported emissions estimates from 1990 or later in the latest National
Communications, in Asia Least-Cost Greenhouse Gas Abatement Strategy (ALGAS) (Asian Development Bank,
1998), or in a country-specific report. Preference is given to the latest published estimates from the
National Communications and ALGAS reports, including both historical and projected estimates.
When the emissions data from these references did not cover the entire historical or projected period
from 1990 to 2020, or in cases where no emissions data were reported, estimated emissions were obtained
using the following approaches:
1. For countries reporting estimates from 1990 to 2010 in 10-year intervals, a linear interpolation
was used to estimate values in 5-year increments.
2 Annex I countries are countries that are listed in Annex I to the United Nations Framework Convention on Climate
Change. A complete list of the Annex I countries is available at
.
3 The National Communications are available at .
1-6 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION I — TECHNICAL SUMMARY
2. For countries not reporting emissions through 2000, emissions growth rates were estimated based
on IPCC Tier I4 estimates for the country for 1990 through 2000. The growth rates were applied to
reported inventories since 1990 and used to estimate the remaining years through 2000.
Projections to 2020 are based on growth-rate projections applied to source-specific drivers for
each country, using the estimate for 2000 as the base year.
3. When no emissions data were available or when the data were insufficient, the USEPA
developed emissions estimates, projections, or both, using the default methodology presented in
the 1996 Revised IPCC Guidelines (IPCC, 1997) and the IPCC Good Practice Guidance (IPCC, 2000).
Baseline projections represent business-as-usual scenarios, where currently achieved reductions are
incorporated, but future mitigation actions are only included if either a well-established program or an
international sector agreement is in place. Thus, projections do not include planned climate change
source-level mitigation efforts, although they do include voluntary and nonclimate-based policies that
indirectly reduce greenhouse gases. For consistency, if a country's reported projections include planned
climate mitigation efforts, the reductions from those efforts were added back into the emissions
projections, where identified. If planned climate policy reductions could not be identified, a country's
emissions projections were estimated by continuing trends from previous years, as reported in historical
inventories.
Source-by-source and country-by-country explanations of how the projections were developed can be
found in the appendix to USEPA (2006).
1.3,1.1 Baseline Emissions for Agriculture
For the agricultural mitigation analysis, separate baseline emissions for croplands and rice cultivation
were developed and used, even though USEPA (2006) includes estimates for these sources. Process-based
models—DAYCENT for croplands and DeNitrofication-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 clear identification of baseline management conditions and consistent
estimates of changes to those conditions through mitigation activities. For emissions associated with
livestock, the mitigation analysis in this report relies on USEPA (2006) baseline estimates. Further details
about the emissions baselines estimated by the DAYCENT and DNDC models, and their relationship to
USEPA (2006) estimates, are provided in Section V Agriculture of this report.
1.3.1.2 Baseline Emissions for Fluorinated Gases
Baselines for the fluorinated gases are also based on Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2020 (USEPA, 2006). The 2006 USEPA analysis builds on the 2001 USEPA analysis to
develop country-by-country and industry-by-industry projections of emissions using projections of
activity data, emissions factors, or other data related to emissions. For the industrial sources, activity data
were multiplied by emissions factors to obtain emissions projections. For the substitutes for ODSs,
estimates of country-specific ODS consumption as reported under the Montreal Protocol were used in
conjunction with output from the USEPA's Vintaging Model to project emissions. Activity data and
activity growth projections were obtained from a variety of sources, including international industry
trade organizations and databases, U.S. government agencies, and international organizations. For all
industries, country-specific estimates of activity (or a factor related to activity) were available.
Information on emissions rates was generally less precise but was often available on a regional, if not
country-specific, basis.
4 Tier 1 refers to the emissions factor estimation methodology in the IPCC guidelines with the highest level of implied
accuracy in emissions estimation in a hierarchy of methodology tiers.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 1-7
-------
SECTION I — TECHNICAL SUMMARY
For industrial sources of fluorinated gases, this report presents international baselines and MACs for
five industrial sources of HFCs, PFCs, and SF6, including the production of aluminum, magnesium,
semiconductors, and HCFC-22, and the use of electrical equipment in electric power systems. For all five
of these sources, two sets of baselines and MACs are presented: the technology-adoption baseline, based
on the assumption that the industries will achieve their announced global emissions reduction goals for
the year, and the no-action baseline, based on the assumption that the industries' emissions rates will
remain constant. Detailed discussions of the methodology used to develop the baselines for each source
can be found in USEPA (2006).
In addition to the industrial sectors, this report also includes estimates of fluorinated gases that are
used as substitutes for ODSs. The USEPA's Vintaging Model and industry data were used to simulate the
aggregate impacts of the ODS phaseout on the use and emissions of various fluorocarbons and their
substitutes in the United States. Emissions estimates for non-U.S. countries incorporate estimates of the
consumption of ODSs by country, as provided by the United Nations Environment Programme (UNEP)
(1999). The estimates for the European Union (EU) were provided in aggregate, and each country's gross
domestic product (GDP) was used as a proxy to divide the consumption of the individual member nation
by the EU total. Estimates of country-specific ODS consumption, as reported under the Montreal
Protocol, were then used in conjunction with Vintaging Model output for each ODS-consuming sector. In
the absence of country-level data, preliminary estimates of emissions were calculated by assuming that
the transition from ODSs to HFCs and other substitutes follows the same general substitution patterns
internationally as observed in the United States. From this preliminary assumption, emissions estimates
were then tailored to individual countries or regions by applying adjustment factors to U.S. substitution
scenarios, based on relative differences in economic growth, rates of ODS phaseout, and the distribution
of ODS use across end-uses in each region or country, as explained in Section IV Industrial Processes in
this report.
1.3.2 Mitigation Option Analysis Methodology
Although non-CO2 emissions from each sector are estimated according to the available data and
issues important to that sector, the mitigation option analysis throughout this report was conducted using
a common methodology. 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. Mitigation options represented in the MACs of this report are applied to the baselines
described in Section 1.3.1.
The abatement analysis for all non-CO2 gases for agriculture, coal mines, natural gas systems, oil
systems, landfills, wastewater treatment, and nitric and adipic acid production are based on and improve
upon DeAngelo et al. (in press), Delhotal et at. (in press), and Ottinger et al. (in press); two previous
USEPA studies for the United States (USEPA, 2001, 1999); and a study conducted by the European
Commission (EC) (2001) that evaluated technologies and costs of CH4 abatement for EU members from
1990 to 2010. These studies provided estimates of potential CH4 and N2O emissions reductions from
major emitting sectors and quantified costs and benefits of these reductions.
The EC study evaluates the abatement potential and cost options at representative facilities or point
sources of emissions, such as waste digesters, and then extrapolates the results to a country and to the EU
level. Given the more detailed data available for U.S. estimates, the USEPA's U.S. analysis also 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
1-8 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION I — TECHNICAL SUMMARY
equipment, pipeline compressors and equipment, and system upsets. Thus, the EC analysis provides
more of a sector-average cost for individual abatement options at the country or EU level, while the
USEPA analysis provides more detail at the sector and subsector levels.
For this report, average U.S. abatement costs and benefits are estimated for each abatement option to
build a set of regional options and estimates comparable to that for the EU. Together, this new combined
set of abatement options is applied to all defined regions in the study, both the United States and the EU,
as well as to regions where data and detailed analyses are unavailable. The advantage of using the
"average" approach over the more detailed analyses for the United States and the EU is that the approach
incorporates the latest emissions estimated and compiled in USEPA (2006) and provides for a consistent
methodology throughout the analysis for all regions. It should be noted that mitigation estimates from
this "average" approach are more conservative than those reported in the USEPA and EC reports.
For the high-GWP abatement analysis, it is assumed that some mitigation technologies are adopted to
meet 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 penetration rates of mitigation technologies
competing for the same set of fluorinated gas emissions.
The agricultural sector's emissions abatement analysis improves upon a previous study supported by
the USEPA (DeAngelo et al., in press) that generated MACs by major world region for cropland N2O,
livestock enteric CH4, manure CH4, and rice CH4 for the year 2010. The most significant change in this
report is the use of biophysical, process-based models (i.e., DAYCENT and DNDC) 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.
Additional mitigation options are now assessed (e.g., slow-release fertilizers, nitrogen, (N)-inhibitors, and
no-till), and more detailed, less aggregated results are provided for individual crop types under both
irrigated and rainfed conditions. Improved agriculture MACs are generated for 2000, 2010, and 2020.
1.3.2.1 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 rate or extent of
penetration of an option into the market within different regions may vary based on these conditions. The
selective omission of options represents a static view of the region's socioeconomic conditions. 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. The average technical lifetime of an option (in years) is also determined using expert knowledge
of the technology or recent literature, as referenced in each section of this report.
Table 3-1 summarizes how the abatement potential is calculated for each of the available abatement
options. The total abatement potential of an option for each region is equal to an option's technical
applicability multiplied by its implied adoption rate multiplied by its reduction efficiency. Total baseline
emissions are summed from each of the emissions sources within each sector and each region. Each
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 1-9
-------
SECTION I — TECHNICAL SUMMARY
Table 3-1: Abatement Potential Calculation for Mitigation Options
Technical applicability
(%)
Percentage of total
baseline emissions from
a particular emissions
source to which a given
option can be potentially
applied.
Implied adoption rate
X (%)
Percentage of technically
applicable baseline
emissions to which a
given option is applied;
avoids double counting
among overlapping
options and fixes
penetration rate of options
relative to each other.3
Reduction efficiency
X (%)
Percentage of
technically achievable
emissions abatement
for an option after it is
applied to a given
emissions stream.
Abatement potential
(%)
Percentage of baseline
emissions that can be
reduced at the national or
regional level by a given
option. Product of technical
applicability, implied
adoption rate, and reduction
efficiency of the option.
a Implied adoption rate for nonoverlapping options (i.e., applicable to different emissions streams) is assumed to add to 100 percent of
technically applicable baseline emissions.
mitigation option reduces baseline emissions by the reduction efficiency percentage of the relevant
portion of the total baseline emissions, as defined by the technical applicability and implied adoption
rate.
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 adoption rate of an option is a mathematical adjustment for other qualitative factors that
may influence the effectiveness of a mitigation option. For the 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. The implied adoption rate of each of the n
overlapping options is equal to lln, which 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.
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. 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.
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.
Once the total abatement potential of an option is calculated as described above, the abatement
potential is multiplied by the baseline emissions for each sector and region to calculate the absolute
1-10
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION I — TECHNICAL SUMMARY
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 CO2 equivalent (MtCO2eq).5
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.
1.3.2.2 Economic Characteristics of Abatement Options
Each abatement option is characterized in terms of its costs and benefits per an abated unit of gas
(tCO2eq or tons of emitted gas [e.g., tCH4]).
For each mitigation option, the carbon price (P) at which that option becomes economically viable can
be calculated (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. However, in the
energy and waste sectors, sensitivities are conducted to examine the implication of time. 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,
(\-TR)(P-ER
(1 - TR)RC
(\ + DR)r
_
y
Net Present Value Benefits Net Present Value Costs ,3 .| \
where
P = the breakeven price of the option ($/tCO2eq);
ER = the emissions reduction achieved by the technology (MtCO2eq);
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
($);
T = the option lifetime (years);
DR = the selected discount rate (%);
CC = the one-time capital cost of the option ($);
RC = the recurring (O&M) cost of the option (portions of which may be scaled based on regional
labor costs) ($/year);
TR = the tax rate (%); and
TB = the tax break equal to the capital cost divided by the option lifetime, multiplied by the tax rate
($)•
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:
' One MtCO2eq equals 1 teragram of CO2 equivalent (TgCO2eq): 1 metric ton = 1,000 kg = 1.102 short tons = 2,205 Ibs.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 1-11
-------
SECTION I — TECHNICAL SUMMARY
_ cc
Costs include capital or one-time costs and operation and maintenance (O&M) or recurring costs.
Additionally, some one-time costs (where data are available) are subdivided into labor and equipment
components. Recurring costs may also be subdivided into labor costs, fertilizer costs, and other cost
components. Benefits or revenues from employing an abatement option can include (1) the intrinsic value
of the recovered gas (e.g., the value of CH4 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 tCO2eq ($/tCO2eq) or dollars per metric ton of gas (e.g.,, $/tCH4, $/tHFC-
134a). In most cases, there are two price signals for the abatement of CH4: one price based on CH4's value
as energy (because natural gas is 95 percent CH4) and one price based on CH4's value as a greenhouse
gas. All cost and benefit values are expressed in constant year 2000 U.S. dollars.
Costs and benefits of abatement options are adjusted based on energy and labor costs in
corresponding regions. If not otherwise available, the equipment component of fixed costs is not adjusted
and stays the same for all regions. Most of the agricultural sector options, such as changes in management
practice, do not have applicable capital costs, with the exception of anaerobic digesters for manure
management. In general, labor costs comprise the majority of O&M costs. Given this fact, we have used
labor costs as a proxy to adjust O&M costs across regions, as well as the labor component of the one-time
cost. Specifically, O&M costs for each region are estimated based on a ratio between the average regional
labor cost in manufacturing in that region and in the United States for U.S.-based options or the EU for
EU-based options. Regional labor costs in manufacturing are taken from World Bank data (2000). For the
agricultural sector, labor costs are calculated labor shares of agricultural production cosls from the Global
Trade Analysis Project (GTAP) and agricultural wage data from the International Food Policy Research
Institute (IFPRI).
Breakeven price calculations for this analysis do not include transaction 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.
In regions where there is a lack of detailed revenue data, revenues are scaled based on the ratio
between average prices of natural gas (when CH4 is abated and sold as natural gas) or of electricity (when
CH4 is used to generate electricity or heat) in a given region and in the United States or EU. Similarly,
revenues from non-CH4 benefits of abatement options are scaled based on the ratio between the GDPs
per capita in a given region and in the United States or EU. 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.
This analysis is conducted using a 10 percent discount rate and a 40 percent tax rate. In some sectors,
sensitivities on alternative discount and tax rates illustrate different social and industry perspectives.
Sensitivities with a social perspective use lower discount rates and a zero percent tax rate, while
sensitivities with an industry perspective assume higher discount rates and greater than zero tax rates.
For quick reference, Table 3-2 lists the basic financial assumptions used throughout this report. In
addition, because of the high sensitivity to energy prices, the analysis tests the MAC sensitivity to
1-12 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION I — TECHNICAL SUMMARY
Table 3-2: Financial Assumptions in Breakeven Price Calculations for Abatement Options
Variable
Discount rate
Tax rate
Year dollars
Assumption
10%
40%
2000$
changes in base energy price (from -50 percent to 200 percent) for both electricity and natural gas, where
this sensitivity test is relevant to the sector. The energy price assumptions are also included in the
TechTables.xls file in the appendices to the International Analysis of Methane and Nitrous Oxide Abatement
Opportunities: Report to Energy Modeling Forum, Working Group 21 on the USEPA's Web site
(USEPA, 2005).
1.3.3 Marginal Abatement Curves
MACs are used to show the amount of emissions reduction potential at varying price levels. In
theory, a MAC illustrates the cost of abating each additional ton of emissions. Figure 3-1 shows an
illustrative MAC. The x-axis shows the amount of emissions abatement in MtCO2eq, and the y-axis shows
the breakeven price in $/tCO2eq required to achieve the level of abatement. Therefore, moving along the
curve, the lowest cost abatement options are adopted first. The curve becomes vertical at the point of
maximum total abatement potential, which is the sum of abatement across all options in a sector or
region.
Figure 3-1: Illustrative Non-C02 Marginal Abatement Curve
Value of C02
Equivalent
($/tC02eq)
Market Price
$0/tC02eq
Total Abatement Potential
Energy/Commodity
Prices
Abated GHG Emissions (MtCO.eq)
In Figure 3-1, the commodity/energy market price is aligned to $0/tCO2eq since this price represents
the point at which no additional price signals exist from GHG credits to motivate emissions reductions;
all emissions reductions are due to increased energy efficiencies, conservation of production materials, or
both. As a value is placed on GHG reductions in terms of $/tCO2eq, these values are added to the
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
1-13
-------
SECTION I — TECHNICAL SUMMARY
commodity/energy market prices and allow for additional emissions reductions to clear the market. The
points on the MAC that appear at or below the zero cost line ($0/tCO2eq) illustrate this dual price-signal
market. 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 because of the existence of nonmonetary barriers.
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 1.3.2.2). 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 3-1), 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 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 (ER in Equations 3.1 and 3.2).
When data were not available to clearly identify marginal abatement roles ior mitigation technologies
because of either (a) the potential for abatement of the same share of baseline emissions, or (b)
sensitivities to the order of adoption, we employed the implied adoption rate (Table 3-1).
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 the^ are complements or substitutes. If a group of
options are complements (or independent of one another), the implied adoption rates are all equal to one.
If options are substitutes for each other, the lowest price option is selected for each representative facility;
in this way, the implied adoption rate for each technology is estimated.
In the industrial processes sector, mitigation options are applied to one representative facility, 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 lln factor, as described in
Section 1.3.2.1), 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 representative farms of each region based
on the lowest average breakeven price. The implied adoption rate is based purely on the number of
available migration options (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 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.,
1-14 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION I — TECHNICAL SUMMARY
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.
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.6 Sectoral MACs are aggregated by gas and by region to create global MACs,
which are presented in Section 1.4.
1.3.4 Methodological Enhancements from Energy Modeling Forum
Study
This report builds on a study previously conducted by the USEPA for Stanford's EMF-21. 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.
In the energy and waste sectors, new sensitivity cases illustrate the effect of technical change over
time. Introducing technical change by incorporating the rate of change of technical applicability can
potentially shift the MAC down and to the right on the graph, as abatement potential increases and net
costs decrease at a given carbon price.
For industrial sources of fluorinated gases, the emissions baselines have been updated since the EMF-
21 analysis. The analysis included one set of baseline emissions for industrial sources, while this report
presents two sets of baselines for aluminum, magnesium, and semiconductor manufacturing. One
baseline set assumes industry agreements establishing emissions reduction targets will be upheld, while
the other baseline set assumes that the industry agreement has no effect on the baseline emissions. 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 the
biophysical, process-based models DAYCENT and DNDC. 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 individual crop types.
Finally, the agricultural commodity market effects are explored with a global agricultural trade model
(IMPACT of the IFPRI).
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.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 1-15
-------
SECTION I — TECHNICAL SUMMARY
1.4 Aggregate Results
orldwide, 2005 total non-CO2 anthropogenic greenhouse gas baseline emissions are
estimated to be 10,278 MtCO2eq and are projected to increase by 27 percent to 13,013
MtCO2eq by 2020. These gases are emitted from four major emitting sectors: the energy,
waste management, industrial processes, and agricultural industries. China, India, the United States,
Brazil, and the European Union are the world's five largest emitters and account for approximately 76
percent of total non-CO2 emissions.
This section presents the forecasted baseline emissions and provides a global overview of the results
from the MAC analysis by sector and for the five largest emitting regions. The data represented in this
chapter are aggregated and provide a summary of all sources and non-CO2 greenhouse gases. The
individual chapters are organized by source and present the full details of these analyses. For a complete
data set of mitigation potential by sector, gas, and region, refer to Appendix A.
For the purposes of aggregation, the results from the "technology adoption" baseline were used from
industrial process subsectors with dual baselines. In the agriculture sector, the MAC data from the
"constant area" scenarios were used, while the baselines from Global Anthropogenic No«-CO2 Greenhouse
Gas Emissions: 1990-2020 (USEPA, 2006) were used for consistency across the sectors in aggregation.
1.4.1 Baselines
1.4.1.1 By Non-CO2 Greenhouse Gas
Figure 4-1 provides information on the relative share of each greenhouse gas that comprises the
global non-CO2 greenhouse gas baseline emissions total. CH4 represents the largest share of emissions
worldwide, accounting for approximately 61 percent of the total non-CO2 emissions in 2005, while N2O
and high-GWP gases accounted for 34 percent and 5 percent, respectively.
Figure 4-1: Percentage Share of Global Non-C02 Emissions3 by Type of Gas in 2005
World Total = 10,280 MtCO2eq
High-GWP 5%
Source: USEPA, 2006.
a C02 equivalency based on 100-year GWP.
1-16
GLOBAL MITIGATION OF NON-COj GREENHOUSE GASES
-------
SECTION I — TECHNICAL SUMMARY
Figure 4-2 presents the projected baseline emissions by greenhouse gas for 2000, 2010, and 2020. The
distribution of non-CO2 greenhouse gases is forecasted to remain relatively unchanged through 2020. The
most significant change is represented by a projected increase in the relative share of high-GWP gases
with respect to CH4 and N2O, growing from 5 percent to more than 7 percent of global non-CO2
emissions between 2005 and 2020.
Figure 4-2: Non-C02 Global Emissions Forecast to 2020 by Greenhouse Gas
o
o
14,000
12,000
10,000
8,000
6,000
4,000
2,000
0
• High-GW
D N20
• CH,
Source: USEPA, 2006.
1.4.1.2 By Major Emitting Sectors and Countries
The sources of non-CO2 emissions are categorized into four major emissions sectors: energy, waste,
industrial processes, and agriculture. Figures 4-3 and 4-4 provide the projected global emissions baseline
for 2000, 2010, and 2020, 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-CO2 emissions, accounting for 60 percent of the total 2010
baseline. Energy is the second largest emissions producer, representing 20 percent of the total baseline.
The waste sector represents 14 percent of the total baseline, and the industrial processes sector represents
7 percent.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
1-17
-------
SECTION I —TECHNICAL SUMMARY
Figure 4-3: Global Emissions by Major Sector for All Non-C02 Greenhouse Gases
ET
0>
CM
o
o
14,000
12,000
10,000
8,000
6,000
4,000
2,000
0
• Industrial Processes
• Waste
D Energy
| Agriculture
2000
Source: USEPA, 2006. Note that this mitigation analysis uses baseline emissions projections for croplands and rice (within agriculture) that
differ from USEPA (2006)
Figure 4-4: Projected World Emissions Baselines for Non-C02 Greenhouse Gases, Including the Top
Emitting Regions
14,000
12,000
10,000
o-
® 8,000
O
I 6,000
4,000
2,000
0
H India
D Brazil
HEU-15
• Latin America/Caribbean
n United States
• South &SE Asia
• Africa
D China
• Rest of the world
2000
2010
2020
Source: USEPA, 2006.
EU-15 = European Union.
1-18
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION I — TECHNICAL SUMMARY
Figure 4-4 shows the projected emissions baselines for the world, as well as the largest emitting
countries. The largest non-CO2 emitting countries are typically characterized as mature, highly
industrialized countries or countries with significant agricultural sectors. In 2005, the top five emitting
countries—China, the United States, EU-15, Brazil, and India—account for 44 percent of the world's total
non-CO2 emissions, and their relative contribution to the world baseline is projected to remain the same
during the next 15 years.
1.4.2 Global MACs
The MAC analysis methodology outlined in Section 1.3 of this report develops bottom-up projections
of potential reductions in non-CO2 emissions in terms of the breakeven price ($/tCO2eq). The emissions
reduction potential is constrained by technology limitations, as well as by 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-CO2 gas emissions for a given breakeven price. Figure 4-5 presents the results
from the MAC analysis for 2020 by major economic sector. Figure 4-6 presents aggregate MACs by
greenhouse gas type for 2020. Figure 4-7 presents the 2020 MACs for the world's largest non-CO2
greenhouse gas emitting regions.
Figure 4-5: Global 2020 MACs for Non-C02 Greenhouse Gases by Major Sector
Energy
Agriculture
•* Waste
= Industrial Processes
1,000
1,250
1,500
Non-CO2 Reduction (MtCO2eq)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
1-19
-------
SECTION I —TECHNICAL SUMMARY
Figure 4-6: Global 2020 MACs by Non-C02 Greenhouse Gas Type
$60-1
$50 -
$40 -
$30 -
a- $20-
u>
g $10 -
* $0-
(
-$10 -
-$20 -
-$30 -
-$40-
i » *
f
J
I | — ' — * Methane
I i — ' Nitrous Oxide
• J • High-GWP
J
r— 1 , '
^ p- | , | 1 !
ff' 500 1,000 1,500 2,000 2,500 3,000
I
Non-CO2 Reduction (MtCO2eq)
Figure 4-7: Global 2020 MACs for Non-C02 Greenhouse Gases by Major Emitting Regions
$60 1
$50 -
$40 -
$30 -
o- $20-
0)
0 $10 J
^ $0-
(
-$10-
M
J
r
. • "'*
T — * China w
| United States f
— * EU'15 J
— * Brazil j
— * India jJ
— « Rest of the world f
I | | ,_™._.™J
—r^250 500 750 1,000 1,250 1,500
-$20 I 1
-$30 1
-$40 -I''
Non-CO2 Reduction (MtCO2eq)
I-20
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION I — TECHNICAL SUMMARY
In the aggregate MACs by gas for the agriculture sector, the net greenhouse gas effects are
represented in the aggregate MACs by gas for both CH4 and N2O. While mitigating in the livestock and
rice sectors affects both N2O and CH4 emissions, the dominant effect is on CH4. Thus, for this analysis, the
net effect on CO2 equivalents is represented in the CH4 global aggregate MAC. Likewise, cropland soil
mitigation affects both N2O and CH4 emissions, but the net greenhouse gas effect is represented in the
global aggregate N2O MAC, because N2O is the dominant mitigation effect.
The 2020 global MACs by major sector (Figure 4-5) illustrate the breakeven mitigation potential for
each of the economic sectors. The greatest potential for cost-effective mitigation (i.e. employing mitigation
options that are economically feasible in the absence of a carbon price signal), is in the energy and
agriculture sectors. In the energy sector, it is estimated that a reduction of approximately 250 MtCO2eq is
possible at a zero-dollar breakeven price. The MACs also show that at higher emissions prices, such as
$20 or $30 per tCO2eq, the energy and agriculture sectors show the greatest potential for emissions
reduction. The industrial processes and waste sectors also show increased mitigation potential at higher
prices, but to a lesser degree. The more vertical slope of the MAC for the industrial sector shows that an
increase in the emissions price may not result in any further mitigation beyond a certain point.
Across all non-CO2 greenhouse gases, methane has the greatest mitigation potential, as shown in the
2020 MACs by greenhouse gas type (Figure 4-6). In the absence of a carbon price signal, methane
emissions could be reduced by nearly 500 MtCO2eq. Nitrous oxide and high-GWP gases also exhibit
significant cost-effective mitigation potential, although to a lesser extent than that of methane. As
breakeven prices rise, methane potential continues to grow, approaching a reduction potential of 1,800
MtCO2eq at a breakeven price of $30/tCO2eq.
The MACs by major emitting regions (Figure 4-7) exhibit China's large mitigation potential in 2020 at
higher breakeven prices. At $30/tCO2eq, China could potentially reduce non-CO2 emissions by up to
nearly 450 MtCO2eq, approximately three times the mitigation potential for the European Union. Both
China and the United States exhibit the largest potential for mitigation at higher breakeven prices. India
and Brazil also fall in the largest five emitting regions for non-CO2 greenhouse gases.
The aggregate MACs by economic sector, greenhouse gas type, and region highlight the potential for
including non-CO2 greenhouse gases in multigas strategy analysis. The MACs illustrate that a significant
portion of this mitigation potential can be realized at a zero cost and at low carbon prices. This report
examines the mitigation potential in each sector in greater detail. Sensitivity analysis on factors such as
discount rates, the rate of technical change, and the ratio of domestic to foreign inputs can be found in the
sector-specific chapters of this report.
1.5 Limitations and Applications of MACs
hile this global mitigation report has important implications for researchers and modelers,
it is important to understand not only the limitations of this analysis, but also the potential
for misapplication of the data in other analyses.
1.5.1 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.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 1-21
-------
SECTION I — TECHNICAL SUMMARY
1.5.1.1 Exclusion of Transaction Costs
Future work in the area of mitigation costs will focus on including transactions costs. Current work
still in draft by Lawrence Berkeley National Laboratory (LBNL), Transaction Costs of GHG Emissions
Reduction Projects: Preliminary Results (2003), estimates that transactions costs will add approximately $1
per ton of carbon to a project. However, the LBNL study is not comprehensive, because it considered only
two non-CC>2 projects. Transaction costs are likely to vary significantly, contingent on the size of the
project, the applicable mitigation technology, and other factors. Given the lack of comprehensive data,
this analysis does not include transaction costs.
1.5.1.2 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 2020 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.
1.5.1.3 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.
1.5.1.4 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.
1.5.2 Practical Applications of MACs in Economic Models
MAC data are presented in both percentage reduction and absolute reduction terms relative to the
baseline emissions. These data can also be downloaded in spreadsheet format from our Web site at
.
1-22 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION I — TECHNICAL SUMMARY
The MAC data are an important input into the economic modeling of global climate change. The
MACs can be applied in a variety of economic models to represent the potential emissions abatement of
non-CO2 greenhouse gases in each sector at a given carbon price.
While the results presented in this report can inform economic models, caution should be taken not to
apply the MAC data directly as offset curves. Offset curves are a supply curve of emissions permits that
could potentially be available in the market at a given carbon-price environment. However, a price signal
alone is not likely to bring about all of the mitigation opportunities available along the MACs presented
in this report. Other nonprice factors, such as social acceptance, tend to inhibit mitigation option
installation in many sectors. Because of the lack of quantitative data on nonprice factors determining
market penetration, we have represented the implied adoption rate of mitigation technologies in our
analysis with a mathematical distribution of technologies across the baseline emissions of a sector. Thus,
the MACs in our analyses do not represent a supply curve of emissions permits that would be available
for purchase, but rather the technical mitigation potential at a given carbon price.
In addition, caution should be taken when applying MACs for sectors that are dependent on energy
supply, because of the potential sensitivity of the MACs for these sectors to carbon prices. For example, a
positive carbon-price environment may result in reduction in coal use, which may reduce CH4 emissions.
This potential reduction in emissions would have occurred because of a decrease in use of the facility,
rather than the installation of a mitigation option in the facility.
This analysis focuses only on the mitigation of non-CO^ without considering the impacts of CO2
mitigation. It should be noted that the mitigation potential of non-CO2 greenhouse gas emissions
generated in the energy sector (e.g., coal mining) is inherently tied to the mitigation potential of CO2
emissions from the same sector. Any modeling of greenhouse gas mitigation in the energy sector should
consider the coeffects of any change in energy consumption in both non-CO2 and CO2 mitigation
potential.
1.6 References
Asian Development Bank. 1998. Asia Least-Cost Greenhouse Gas Abatement Strategy. Country Studies.
Manila, Philippines: Asian Development Bank, Global Environment Facility and the United Nations
Development Programme.
Committee on the Science of Climate Change, Division on Earth and Life Studies, National Research
Council (NRC). 2001. "Climate Change Science: An Analysis of Some Key Questions." Washington,
DC: National Academy Press. Available at .
de la Chesnaye, F.C., C. Delhotal, B. DeAngelo, D. Ottinger Schaefer, and D. Godwin. In press. "Past,
Present, and Future of Non-CO2 Gas Mitigation Analysis in Human-Induced Climate Change: An
Interdisciplinary Assessment." Cambridge University Press, Cambridge.
DeAngelo, B.J., F. de la Chesnaye, R.H. Beach, A. Sommer, and B.C. Murray. In press. "Methane and
Nitrous Oxide Mitigation in Agriculture." Energy Journal.
Delhotal, C., F. de la Chesnaye, A. Gardiner, J. Bates, and A. Sankovski. In press. "Estimating Potential
Reductions of Methane and Nitrous Oxide Emissions from Waste, Energy and Industry." Energy
Journal.
Dlugokencky, E.J., S. Houweling, L. Bruhwiler, K.A. Masarie, P.M. Lang, J.B. Miller, et al. 2003.
"Atmospheric Methane Levels Off: Temporary Pause Or New Steady State?" Geophysical Research
Letters 30:10.1029/2003GL018126.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 1-23
-------
SECTION I —TECHNICAL SUMMARY
European Commission (EC). 2001. "Economic Evaluation of Sectoral Emissions Reduction Objectives for
Climate Change." Brussels, Belgium: European Commission. Available at .
Intergovernmental Panel on 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, Organisation for Economic Co-Operation and Development, International Energy
Agency.
Intergovernmental Panel on Climate Change (IPPC). 1997. 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, Organisation for Economic Co-Operation and Development, International Energy
Agency.
Intergovernmental Panel on Climate Change (IPPC). 2000. "Good Practice Guidance and Uncertainty
Management in National Greenhouse Gas Inventories." Kanagawa, Japan: Intergovernmental Panel
on Climate Change.
Intergovernmental Panel on Climate Change (IPCC) 2001a. Summary for Policy Makers: A Report of Working
Group I of the Intergovernmental Panel on Climate Change. The Third Assessment Report of Working
Group I of the Intergovernmental Panel on Climate Change, Summary for Policy Makers approved in
Shanghai in January 2001. Available at .
Intergovernmental Panel on Climate Change (IPCC). 2()01b. Technical Summary: A Report Accepted by
Working Group I of the IPCC but not approved in detail. A product resulting irom The Third Assessment
Report of Working Group I of the Intergovernmental Panel on Climate Change, January 2001.
Available at .
Intergovernmental Panel on Climate Change (IPCC) 2001c. Climate Change 2001: The Scientific Basis.
Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on
Climate Change (IPCC), January 2001. Available at .
Ottinger Schaefer, D., D. Godwin, and J. Harnisch. In press. "Estimating Future Emissions and Potential
Reductions of HFCs, PFCs, and SF6." Energy Journal.
Sathaye, A.J., E. Smith, and M. Shelby. 2003. "Transaction Costs of GHG Emissions Reduction Projects:
Preliminary Results." Lawrence Berkeley National Laboratory.
United Nations Environment Programme (UNEP). 1999. The Implications to the Montreal Protocol of the
Inclusion of HFCs and PFCs in the Kyoto Protocol. UNEP HFC and PFC Task Force of the Technology
and Economic Assessment Panel (TEAP).
U.S. Environmental Protection Agency (USEPA). 1999. U.S. Methane Emissions 1990-2020: Inventories,
Projections, and Opportunities for Reductions. Washington, DC: USEPA, Office of Air and Radiation,
EPA 430-R-99-013.
U.S. Environmental Protection Agency (USEPA). 2001. "Addendum Update to U.S. Methane Emissions
1990-2020: Inventories, Projections, and Opportunities for Reductions." Washington, DC: USEPA.
Available at .
U.S. Environmental Protection Agency (USEPA). 2001b. Emissions and Projections of Non-CO2 Greenhouse
Gases for Developed Countries: 1990-2010. Washington, DC: USEPA. Available at
.
U.S. Environmental Protection Agency (USEPA). 2002. Emissions and Projections of Non-CO2 Greenhouse
Gases for Developing Countries: 1990-2020. (Draft). Washington, DC: USEPA. Available via e-mail or
from the EMF-21 Web site.
U.S. Environmental Protection Agency (USEPA). 2005. "International Analysis of Methane and Nitrous
Oxide Abatement Opportunities: Report to Energy Modeling Forum, Working Group 21." Stanford
University. Available on the USEPA Web site . As obtained on December 21, 2005.
1-24 GLOBAL MITIGATION OF IMON-C02 GREENHOUSE GASES
-------
SECTION I — TECHNICAL SUMMARY
U.S. Environmental Protection Agency (USEPA). 2006. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2020. Washington, DC: USEPA.
Weyant, J. and F. de la Chesnaye (eds.). In press. "Multigas Mitigation and Climate Change." Energy
Journal.
World Bank. 2000. "World Development Indicators, Table 2.6 Wages and Productivity." Washington, DC:
World Bank. Available at .
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 1-25
-------
SECTION I — TECHNICAL SUMMARY
1-26 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
Section I: Technical Summary Appendixes
Appendixes for this section are available for download from the USEPA's Web site at
http://www.epa.gov/nonco2/econ-inv/international.html.
-------
II. Energy
-------
SECTION II — ENERGY • PREFACE
Section II presents international emissions baselines and marginal abatement curves (MACs) for energy
sources. There are three chapters, each addressing an individual source from the coal mining, natural gas,
and oil sectors. These sources are associated with methane (CH4) emissions. MAC data are presented in
both percentage reduction and absolute reduction terms relative to the baseline emissions. These data can
be downloaded in spreadsheet format from the USEPA's Web site at
.
Section II—Energy chapters are organized as follows:
Methane (CH4)
II. 1 Coal Mining Sector
II.2 Natural Gas Sector
II.3 Oil Sector
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY « COAL MINING
11.1 Coal Mining Sector
orldwide, the coal mining industry liberated more than 377 million metric tons of carbon
dioxide equivalent (MtCO2eq), which accounted for 3.3 percent of total anthropogenic
methane (CH4) emissions in 2000. China, the United States, India, and Australia account
for more than 56 percent of coal mining CH4 emissions (Figure 1-1). Emissions are projected to grow 20
percent from 2000 to 2020, with China increasing its share of worldwide emissions from 31 percent to 42
percent.
Figure 1 -1: CH4 Emissions from Coal Mining, by Country: 2000-2020
H Australia
• India
• United States
DChina
• Rest of the world
Source: U.S. Environmental Protection Agency (USEPA), 2006.
11.1.1 Introduction
CH4 is produced during the process of converting organic matter to coal. The CH4 is stored in pockets
within a coal seam until it is released during coal mining operations. The largest source of emissions
occurs during mining. Although, some emissions occur during the processing, transport, and storage of
coal. Many factors affect the quantity of CH4 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 depth of a coal seam and the type of coal determine the amount of CH4 present
(or the gas content) in and around the coal seams. Deep coal seams generally have large amounts of CH4
because of greater overburden pressures. As a result, more than 90 percent of fugitive CH4 emissions
from the coal sector come from underground coal mining.
A high concentration of CH4 in underground coal mines is a safety hazard; the CH4 must be extracted
before mining operations can be undertaken. To maintain low levels of CH4 in the mine, degasification is
employed prior to mining and ventilation air systems are used during mining operations. Traditionally,
CH4 extracted from the mine is released or vented into the atmosphere. Abatement options have been
developed to mitigate these emissions.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-1
-------
SECTION II — ENERGY . COAL MINING
The three coal mining abatement options addressed in this chapter are (1) degasification, where holes
are drilled and CH4 is captured (not vented) before mining operations begin (or, in the case of gob gas
wells, during and after mining operations); (2) enhanced degasification, where advanced drilling
technologies are used and captured low-grade gas is purified; and (3) ventilation air methane (VAM)
abatement, where low concentrations of CH4 ventilation air exhaust flows are oxidized to generate heat
for process use and/or electricity generation.
The following discussion offers a brief explanation of how CH4 is emitted from coal mines, followed
by a discussion of international baseline emissions for CH4 from coal mining and projections for future
baseline emissions. Then, we characterize 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 marginal abatement curves (MACs).
11.1.2 Baseline Emissions Estimates
Baseline emissions estimates are calculated by developing activity factors and emissions factors per
unit of activity. The activity factor for coal mining's level of coal production and the emissions factor are
expressed in terms of the quantity of CH4 release per ton of coal produced.
CH4 and coal are created through a combination of biological and geological forces, where plant
biomass is converted to coal. CH4 is stored in natural wells and is also diffused inside the coal itself. CH4
is contained within the coal seam or strata layer by pressure surrounding the seam. When this pressure
drops because of natural erosion, faulting, and underground and surface mining, CH4 emissions occur.
CH4 emissions vary by type of coal mine and type of mining operation. Abandoned mines are also a
source of CH4 emissions.
Underground Mines. The quantity of CH4 present in a mine is determined significantly by the coal
depth. Geologic pressure increases with depth, trapping more CH4. Coal from underground mines tends
to have a higher carbon content, which is associated with a higher CH4 content.
Ventilation air systems are used in underground mines to maintain low concentration levels of CH4
during mining operations. CH4 is combustible at concentrations between 5 percent and 15 percent. As a
safety precaution, countries such as the United States require the use of ventilation systems in mines that
have any detectable levels of CH4. Ventilation systems maintain a CH4 concentration below 1 percent by
using large fans to inject fresh air from the surface into the mine, thereby lowering the in-mine CH4
concentration. This ventilation air is extracted from the mine and vented to the atmosphere through
ventilation shafts or bleeder shafts (see explanatory note 1). The vent air contains very low concentrations
of CH4 (typically below 1 percent).
Degasification systems consist of a network of vertical wells drilled from the surface or boreholes
drilled within the mine and gathering systems to pull the CH4 from the wells to the surface. These wells
extract large quantities of CH4 from the coal seam before and after mining operations. CH4 extracted by
degasification systems has higher concentrations (30 percent to 90 percent) than VAM. Concentrations
vary depending on the type of coal mined and the degasification technique used.
Surface Mines Surface mining is a technique used to extract coal from shallow depths below the
Earth's surface. Because the geologic pressure at shallow depths is much lower, there is insufficient
pressure to contain high concentrations of CH4, so CH4 content is generally also much lower (see
explanatory note 2). As the overlying surface is removed and the coal exposed, CH4 is emitted directly
into the atmosphere. Surface mines contribute only a small fraction of a country's overall emissions, and
11-2 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY • COAL MINING
surface mining is only applicable in certain geographic regions. For example, in the United States in 2003,
surface mining accounted for 67 percent of total domestic coal production. In countries such as China,
there is very little surface mining; coal seams are present only at greater depths.
Postmining Operations. The primary source of CH4 emissions in coal mining is the underground
production of coal. However, some emissions occur during processing, storage, and transport of coal. The
rate of emissions depends on the type 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.
Abandoned Mines. Abandoned mines are another source of CH4 emissions. Emissions are released
through old wells and ventilation shafts. In some cases, the CH4 from these mines has been captured and
used as a source of natural gas or to generate electricity. Currently, these emissions are not included in
the baseline estimates.
In summary, the majority of the CH4 emitted from coal mining comes from gassy underground mines
through ventilation and degasification systems. Future emissions levels and the potential for CH4
recovery and use will be determined by trends in the management of CH4 gas at such mines.
11.1.2.1 Activity Data
Historical Activity Data
Worldwide coal consumption has increased over time, except in Western Europe, Eastern Europe,
and the Former Soviet Union (FSU) (excluding the Russian Federation). Coal consumption decreased 30
percent in Western Europe and 40 percent in Eastern Europe and the FSU from 1990 to 2001. Table 1-1
reports coal mining activity for selected countries during the same period.
In the 1990s, the majority of China's coal mines were operated without modern mining techniques,
which usually include cutting equipment, hydraulic pumps, power roof supports, and automated
loading devices. In the past decade, in an effort to update their equipment, countries such as China have
begun to institute programs to modernize their coal mining operations, allowing them to mine at greater
depths. However, several countries experienced decreased demand for coal in the late 1990s, and in
response, these countries cut mining production until their surplus supply could be reduced. China
dramatically reduced its coal production between 1995 and 2000, and has spent the past 4 years
expanding its coal exports to reduce its surplus. Policies and market forces such as these counteract the
effects of modernization in mining operations and subsequently increase CH4 emissions.
Projected Activity Data
Estimated CH4 emissions baselines are directly related to coal production projections. Sixty percent of
the world's recoverable reserves are located in three regions: the United States (25 percent), FSU (23
percent), and China (12 percent) (USEIA, 2003). China is projected to have the largest increase in coal
projections because of rapid economic growth; the country is projected to almost double coal
consumption by 2025 (USEIA, 2004a).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-3
-------
SECTION II — ENERGY • COAL MINING
Table 1-1: Historical Coal Mining Activity Data for Selected Countries (Million Metric Tons)
Country
China
United States
India
Australia
Russian Federation
South Africa
Germany
Poland
Indonesia
Ukraine
Kazakhstan
Greece
Canada
Czech Republic
Turkey
Rest of the world
World Total
1990
1,190.4
1,029.1
247.6
225.8
NA
193.2
NA
237.1
11.6
NA
NA
57.2
75.3
NA
52.3
1,839.9
5,347.6
1995
1,537.0
1,033.0
320.6
266.5
270.9
227.3
274.2
221.2
45.4
94.6
93.1
63.6
82.7
82.6
60.6
370.0
5,096.0
2000
1,314.4
1,073.6
370.0
338.2
264.9
248.9
226.0
179.5
84.4
69.1
81.5
70.4
76.2
71.8
69.6
340.9
4,930.6
2001
1,458.7
1,127.7
385.4
362.9
2/3.4
250.8
227.1
180.3
102.0
68.0
93.0
73.1
77.6
72.9
68.3
354.9
5,225.3
2002
1,521.2
1,094.3
401.1
376.8
261.8
245.8
232.6
178.5
113.9
65.6
89.2
77.7
73.3
69.8
58.7
355.4
5,259.3
2003
1,635.0
1,069.5
403.1
373.4
294.0
263.8
229.1
177.8
132.4
63.5
86.4
75.3
. 68.5
70.4
53.1
356.4
5,406.3
Source: Energy Information Administration (USEIA), 2004a.
NA = data unavailable.
Note: Coal production values include anthracite, bituminous, and lignite coal types.
11.1.2.2 Emissions Factors and Related Assumptions
Historical Emissions Factors
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. In 2000, emissions
factors for 56 gassy mines in the United States ranged from 57 to 6,000 million cubic feet of CH4 per mine
annually. Emissions factors for 34 the Russian Federation gassy mines ranged from 17 to 3,200 million
cubic feet per mine. For China's 678 state-run mines, emissions factors ranged from 17 to 6,000 million
cubic feet per mine annually from coal production. While the range of emissions factors for the United
States and China is similar, China has significantly more mines with higher emissions factors. The
Intergovernmental Panel on Climate Change (IPCC) estimates average emissions factors by country.
Table 1-2 reports emissions factors for selected countries.
Projected Emissions Factors and Related Assumptions
Improvements made in mining technology throughout the last 20 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 India to reach deeper into their existing coalbed
reserves. As discussed earlier, the volume of CH4 in the coal seam increases at deeper depths because of
increasing geological pressure. Thus, CH4 emissions will rise as technology allows large coal-producing
countries to mine deeper.
11-4 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY • COAL MINING
Table 1 -2: IPCC Suggested Underground Emissions Factors for Selected Countries
Country
FSU
United States
Germany
United Kingdom
Poland
Czechoslovakia
Australia
Emissions Factor
(m3/ton)
17.8-22.2
11.0-15.3
22.4
15.3
6.8-12.0
23.9
15.6
Emissions Factor"
(tCOjeq/ton)
0.25-0.32
0.16-0.22
0.32
0.22
0.10-0.17
0.34
0.22
Source: IPCC, 1996. Adapted from Reference Manual Table 1-54.
FSU = Former Soviet Union.
a Conversion factor of 1 m3 = 0.0143 tC02eq = 35.31 ft3 x 0.00404 tC02eq
11.1.2.3 Emissions Estimates and Related Assumptions
Historical Emissions Estimates
Baseline emissions for Annex I countries are built using publicly available reports produced by the
countries themselves. IPCC's Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories
methodology was used to estimate emissions in each country, ensuring comparability across countries
(IPCC, 1996). The USEPA's baselines assume a "business-as-usual" scenario that does not include climate
change mitigation efforts or other national policies that may indirectly reduce the emissions of
greenhouse gases.
Table 1-3 reports countries with the largest historical CH4 baseline emissions for the years 1990, 1995,
and 2000. CH4 emissions declined worldwide between 1990 and 2000 at an average annual rate of about
10 percent.
Table 1-3: Historical Baseline Emissions for Coal Mine CH4 for Selected Countries (MtC02eq)
Country
China
United States
India
Australia
Russian Federation
Ukraine
North Korea
Poland
South Africa
United Kingdom
Germany
Kazakhstan
Colombia
Mexico
Czech Republic
Rest of the world
World Total
1990
126.1
81.9
10.9
15.8
60.9
55.3
25.3
16.8
6.7
18.3
25.8
24.9
1.9
1.5
7.6
37.2
516.7
1995
149.1
65.8
13.7
17.5
36.8
30.1
27.2
15.6
6.7
12.6
17.6
17.2
2.0
1.8
5.8
32.3
451.5
2000
117.6
56.2
15.8
19.6
29.0
.28.3
26.9
11.9
7.1
7.0
10.2
10.0
3.0
2.1
5.0
27.1
376.9
Source: USEPA, 2006.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES II-5
-------
SECTION II — ENERGY • COAL MINING
Projected Emissions Estimates
Without the introduction of abatement technologies, worldwide CH4 emissions from coal mining are
projected to increase in the next 20 years. This increase is paralleled by a projected increase in coal
consumption over the same period. At the same time, coal's share of overall energy consumption is
expected to steadily decrease as a result of technology advances in other energy markets, such as natural
gas, and renewed interest in nuclear energy.
Technology adoption and organizational restructuring will improve countries' abilities to produce
larger amounts of coal each year. Table 1-4 reports predicted CH4 baseline emissions for the largest coal-
producing countries in the world, assuming the absence of CH4 abatement technologies.
Table 1-4: Projected Baseline Emissions for Coal Mine CH4 for Selected Countries (MtC02eq)
Country
China
United States
India
Australia
Russian Federation
Ukraine
North Korea
Poland
South Africa
United Kingdom
Germany
Kazakhstan
Colombia
Mexico
Czech Republic
Rest of the world
World Total
2005
135.7
55.3
19.5
21.8
26.3
26.3
25.6
11.3
7.4
' 6.7
8.4
6.7
3.4
25
4.8
26.5
388.1
2010
153.8
5l!l
23.1
26.4
27.5
24.5
24.3
10.8
7.2
6,6
7.7
6.4
4.0
2.8
3.9
27.5
407.6
2015
171.8
46.4
28.4
28.2
26.9
23.8
23.1
10.3
7.1
6.4
7.1
6.1
4.7
3.3
3.1
28.9
425.6 ,
2020
189.9
46.4
33.6
29.7
26.3
23.2
21.S
9.8
7.4
6.2
5.9
5.8
•5.5
3,7
3.0
• 11.1
449.5
Source: USEPA, 2006.
11.1.3 Cost of CH4 Emissions Reductions from Coal Mining
The following is a discussion of the abatement technologies and their costs and benefits.
11.1.3.1 Abatement Option Opportunities
Three abatement opportunities currently available to the coal mining sector are
• degasification,
• enhanced degasification, and r
• oxidation of VAM.
11-6 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY • COAL MINING
Engineering costs for each abatement option are based on representative mine characteristics, such as
annual mine production, gassiness of the coal deposits, and CH4 concentration in ventilation flows.
Table 1-5 provides a summary of the one-time investment costs, annual operation and maintenance
(O&M) costs, and benefits from using the captured CH4 as an energy source for each of the three coal
mining abatement options included in the analysis.
Table 1-5: Summary of Average Abatement Costs and Benefits for U.S. Coal Mines (in 2000$)a
Average Costs/Benefits (Millions in 2000$)
Costs
One-Time Costs
Compressor capital
Gathering line capital
Processing capital
Ventilation capital
Miscellaneous capital
Annual Costs
Drilling capital
Drilling materials
Compressors energy (kWh)
Gathering lines labor ,
Processing materials
Ventilation operating costs
Miscellaneous labor
Annual After-Tax Benefits
CH4 sold or purchases offset
Depreciation Tax Benefits
Degasif ication
$1.00
$0.90
$0.04
N/A
$0.38
$0.50
$0.94
$0.33
$0.25
$0.13
N/A
$0,28
$0.97
$0.02
Enhanced
Degasificationb
$0.39
$0.20
$2.56
N/A
$0.14
$0.36
$0.31
$0.13
$0.96
$0.18
N/A
$0.12
$0.34
$0.24
VAMC
N/A
N/A
N/A
$18.64
N/A
N/A
N/A
N/A
N/A
N/A
$0.91
N/A
$2.78
$0.14
Source: Gallaher and Delhotal, 2005.
N/A = Not applicable.
3 Based on a population of 57 U.S. coal mines, accounting for 75 percent of the total liberated CH4 from U.S. coal production.
b Incremental costs and benefits in addition to degasification (Option 1).
c Underlying VAM costs are from Delhotal et al. (2005).
Degasification and Pipeline Injection
High-quality CH4 is recovered from coal seams by drilling vertical wells up to 10 years in advance of
a mining operation or drilling horizontal boreholes up to 1 year 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. Long horizontal boreholes are used by only a
few operators in the United States and Australia.
In some cases, high-quality CH4 also can be obtained from gob wells. Gob gas CH4 concentrations can
range from 50 percent to over 90 percent (USEPA, 1999). The gas recovered is injected into a natural gas
pipeline requiring virtually no purification in the initial stages of production, but necessitating treatment
over time to upgrade the gas to pipeline quality. Gob gas sales from a given location typically decline
over time because of declining levels of concentration. In the United States, of the CH4 recovered from
degasification (or gas drainage as it is often called) 57 percent can be directly used for pipeline injection
(USEPA, 1999).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-7
-------
SECTION II — ENERGY • COAL MINING
Cost Analysis
• Capital Costs. Capital costs include the one-time (upfront) costs of purchasing compressors,
gathering lines, dehydrators, and other miscellaneous capital such as safety equipment and
licenses. Table B-6 in Appendix B for this chapter offers a detailed description of the factors that
determine the required number of each capital component by mine.
• Annual Costs. These costs include materials and labor for drilling, moving gathering lines, and
maintaining the dehydrators. Drilling capital is also considered an annual cost because drilling is
conducted annually. Annual costs generally increase or decrease proportionally to the volume of
CH4 liberated at the individual mine. Table B-6 offers a detailed description of the factors that
determine these costs.
• Cost Savings. Cost savings result from the capture and reuse of natural gas. For basic
degasification, it is assumed that 57 percent of gas capture is suitable for injection into the natural
gas pipelines and hence can be sold directly into the system (USEPA, 1999).
Enhanced Degasification and Pipeline Injection
In enhanced degasification, CH4 is recovered in the same way as in degasification, using vertical
wells, horizontal boreholes, and gob wells. In addition, the mine invests in enrichment technologies such
as nitrogen removal units (NRUs) and dehydrators, used primarily to enhance medium-quality gob well
gas by removing impurities, allowing for larger quantities of CH4 to be captured and used. This option
also assumes tighter well spacing to increase recovery. The enrichment process and tighter spacing
improve recovery efficiency 20 percent more than the first option discussed above (USEPA, 1999). All
costs and benefits presented in Table 1-5 for enhanced degasification are incremental in that they
represent additional abatement costs and CH4 sales above and beyond the basic degasification.
Cost Analysis
• Capital Costs. Enhanced degasification requires the same capital equipment as the degasification
option. In addition, the enhanced option requires an NRU with an estimated average cost of
$200,000 per unit.
• Annual Costs. Similar to degasification, enhanced degasification's annual costs include materials
and labor for drilling, moving gathering lines, and maintaining the dehydrators. However,
annual drilling costs are higher for enhanced degasification because the wells are drilled at closer
intervals to one another. Costs vary proportionally to the amount of gas liberated.
• Cost Savings. It is assumed that 77 percent of the CH4 captured as part of enhanced
degasification can be injected into the natural gas pipeline system. There is a 21 percent increase
over the basic degasification mitigation option (incremental benefits) because gas processing
equipment facilitates nitrogen removal.
Oxidation of Ventilation Air Methane
Oxidation technologies (both thermal and catalytic) have the potential to use CH4 emitted from coal
mine ventilation air. It is not economically feasible to sell this gas to a pipeline because of its extremely
low CH4 concentration levels (typically below 1 percent). However, VAM can be oxidized to generate
CO2 and heat, which in turn may be used directly to heat water or to generate electricity. If oxidizer
technology were applied to all mine ventilation air with concentrations greater than 0.15 percent CH4,
approximately 97 percent of the CH4 from the ventilation air could be mitigated.
11-8 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY • COAL MINING
Cost Analysis
• Capital Costs. Capital costs for VAM oxidation are a function of the level of CH4 concentration in
the ventilation air and the ventilation air flow rate.
• Annual Costs. Annual costs consist primarily of the labor and electricity costs associated with
running the oxidizer. Both of these are proportional to coal production.
• Cost Savings. Heat generated by oxidation systems can be used to heat water (e.g., for steam or
district heating applications) or to generate electricity.
11.1.4 Results
This section presents the Energy Modeling Forum (EMF) Working Group 21 study's MAC analysis
results in tabular format.
11.1.4.1 Data Tables and Graphs
Table 1-6 presents the average breakeven price and the reduction in absolute and percentage terms
for the mitigation options discussed in Section II.1.3.1.
Table 1-6: Summary of Coal Mining Abatement Options Included in the Analysis
Technology
Breakeven
Cost
($ftC02eq)
Emissions
Reduction (%
from baseline)
Emissions Emissions
Reduction in Reduction in
2010(MtC02eq) 2020 (MtC02eq)
Assuming a 1 0% discount rate and a 40% tax rate
Degasification and pipeline injection
Enhanced degasification, gas
enrichment, and pipeline injection
Catalytic oxidation8 (United States)
Flaring
Degasification and power production— A
Degasification and power production— B
Degasification and power production— C
Catalytic oxidation (EU-15)
-$11.66
$2.40
$14.36
$2.47
-$2.09
$5.68
$19.80
$11.34
28%
10%
24%
1%
5%
9%
28%
18%
0.55
0.19
0.77
0.03
0.04
0.06
0.70
0.13
0.55
0.19
0.94
0.03
0.03
0.06
0.83
0.11
Source: USEPA, 2003 Adapted from Coal Sector technology tables in Appendix B of EMF report.
EU-15 = European Union.
Note: Some technologies are not present in all countries. See source for the individual technology's presence in various countries.
a Catalytic oxidation is considered a VAM technology.
The EMF regional baselines and MAC results of the EMF-21 study are presented in Tables 1-7 and 1-8
for 2010 and 2020 using the base energy price, a 10 percent discount rate, and a 40 percent tax rate. These
MACs represent percentage reductions in baseline emissions for individual regions/countries at selected
breakeven prices. Figure 1-2 provides MACs for the five EMF countries/regions with the largest estimated
emissions from coal mining in 2020.
The MACs presented in this section represent static abatement curves using breakeven prices built on
the assumption of fixed mitigation cost and aggregate countrywide natural gas statistics. Appendix B
presents more recent efforts to develop an alternative framework for conducting MAC analysis that
addresses the limitations of the EMF-21 MAC analysis.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-9
-------
SECTION II — ENERGY « COAL MINING
Table 1-7: Baseline Emissions by EMF Regional Grouping: 2000-2020 (MtC02eq)
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South &SE Asia
United States
World Total
2000
9.3
181.9
mo
1.3
117.6
24.3
22.5
15.8
0.8
2.1
61.7
123.5
29.0
31.7
56.2
376.9
2010
8.2
1735
26.6
1.1
153.8
23.4
19.6
23.1
0.7
2.8
56.8
120.2
27.5
29.8
51.1
407.6
2020
8.7
165.8
30,3
1.0
189.9
24,1
17.0
33.6
0.7
3.7
54,7
1154
26.3
28.4
46.4
449.5
Source: USEPA, 2006.
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Note: World Total does not equal the sum of the countries listed in this table because the regional groupings are a subset of the full EMF
regional grouping list. See Appendix A of this report for the full EMF list of countries by region.
Table 1-8: Coal Mining MACs for Countries Included in the Analysis
Country/Region
Africa
Annex!
Australia/New Zealand
Brazil
China
Eastern
EU-15
India
Japan
Mexico
Europe
Non-OECD Annex!
OECD
Russian
Souths
Federation
SEAsia
United States
World Total
$0
38.50%
34.81%
27.91%
0.00%
0.00%
34.16%
0.00%
0.00%
98.00%
28.50%
32.10%
35.40%
27.65%
28.15%
49.22%
16.66%
$15
85.53%
78.05%
83.05%
0.00%
84.45%
73.23%
41.11%
84.18%
98.00%
85.53%
84.80%
75.22%
84.29%
84.09%
85.97%
79.84%
2010
$30
85.53%
78.05%
83.05%
0.00%
84.45%
73.23%
41.11%
84.18%
98.00%
85.53%
84.80%
7542%
84.29%
84.09%
85.97%
79.84%
$45
85.53%
78.05%
83.05%
0.00%
84.45%
73.23%
41.11%
84.18%
98.00%
85.53%
84.80%
75.22%
84.29%
84.09%
85.97%
79.84%
$60
85.53%
78.05%
83.05%
0.00%
84.45%
73.23%
41.11%
84.18%
98.00%
85.53%
84.80%
7552%
84.29%
84.09%
85.97%
79.84%
2020
$0
38.50%
36.33%
27.91%
0.00%
0.00%
34.16%
0.00%
0.00%
98.00%
28.50%
39.21%
34.95%
27.65%
28.15%
49.22%
14.51%
$15
85.53%
81.45%
83.05%
0.00%
84.45%
73.23%
41.11%
84.18%
98.00%
85.53%
103.58%
74.26%
84.29%
84.09%
85.97%
79.81%
$30
85.53%
81.45%
83.05%
0.00%
84.45%
73.23%
41.11%
84.18%
98.00%
85.53%
103.58%
74.26%
84,29%
84.09%
8St7%
79.11%
$41
85,53%
81.45%
83.05%
0.00%
84.45%
73.23%
41.11%
84.18%
9S.OO%
85.53%
103.58%
74.26%
84.21%
84.01%
81,97%
79.81%
$60
85.53%
81.45%
83.05%
0,00%
84.45%
7a23%
41.11%
84.18%
98.00%
85.53%
103.58%
74.26%
84.29%
84.09%
S5.97%
TJJ1%
Source: USEPA, 2003.
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
11-10
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY « COAL MINING
Figure 1-2: EMF MACs for Top Five Emitting Countries/Regions from Coal: 2020
Australia/New Zealand
China
—• South & SE Asia
—~ United States
—> Russian Federation
O $20
50
100
150
200
250
Absolute Reduction (MtCO2eq)
Source: USEPA, 2003.
Note: Regional MACs were constructed using percentage reductions from USEPA (2003), with baselines from USEPA (2005).
11.1.4.2 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,
private mines that are not represented in currently available data sources.
• Country-Specific Tax and Discount Rates. In this analysis, a single tax rate is applied to mines in
all countries to calculate the annual benefits of each technology. In reality, however, tax rates
vary across countries, and in the case of state-run mines in China, taxes may not even be
applicable. Similarly, the discount rate may vary by country. Improving the level of country-
specific detail will help analysts more accurately quantify benefits and breakeven 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
11-11
-------
SECTION II — ENERGY • COAL MINING
• 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.
11.1.5 Summary
The methodology and data discussed in this section describe the MAC analysis conducted for the
coal mining sector by the EMF-21 study. MACs for 2010 and 2020 were estimated based on aggregated
industry data from each country or region. The MACs represent static estimates of potential CH4
mitigation from coal mines based on available information regarding infrastructure and country-reported
emissions estimates provided through the United Nation's Framework Convention on Climate Change
emissions inventory reports.
11.1.6 References
Delhotal, C, F. de la Chesnaye, A. Gardiner, J. Bates, and A. Sankovski. In press. "Estimating Potential
Reductions of Methane and Nitrous Oxide Emissions from Waste, Energy and Industry." Energy
Journal.
Gallaher, M., and K. C. Delhotal. 2005. "Modeling the Impact of Technical Change on Emissions
Abatement Investments in Developing Countries." Journal of Technology Transfer 30 1/2, 211-255.
Intergovernmental Panel on Climate Change (IPCC). 1996. Revised 1996 IPCC Guidelines for National
Greenhouse Gas Inventories: Reference Manual (Volume 3). Available at . As obtained on April 26, 2004.
U.S. Energy Information Administration (USEIA). 2003. International Energy Annual 2001. Table 2.5.
DOE/EIA-0219 (2001) Washington, DC: USEIA.
U.S. Energy Information Administration (USEIA). 2004a. International Energy Annual 2002. Table 7.5.
Washington, DC: USEIA.
U.S. Energy Information Administration (USEIA). 2004b. System for the Analysis of Global Energy Markets.
Washington, DC: USEIA.
U.S. Environmental Protection Agency (USEPA). 1999. U.S. Methane Emissions 1990-2020: Inventories,
Projections, and Opportunities for Reductions. EPA 430-R-99-013. Washington, DC: USEPA Office of Air
and Radiation.
U.S. Environmental Protection Agency (USEPA). 2003. International Analysis of Methane and Nitrous
Oxide Abatement Opportunities: Report to Energy Modeling Forum, Working Group 21.
Appendices. Washington, DC: USEPA. Available at . As obtained on September 27, 2004.
U.S. Environmental Protection Agency (USEPA). 2006. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2020. Washington, DC: USEPA.
11-12 GLOBAL MITIGATION OF NOIN-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY • COAL MINING
Explanatory Notes
1. Bleeder shafts are currently used in only a limited number of countries, including the United States
and the Russian Federation.
2. There are exceptions. In Kazakhstan, for example, the surface mines in Ekibastuz are very gassy and
prone to outbursts; this is the rare exception, though.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-13
-------
SECTION II — ENERGY • COAL MINING
11-14 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY • NATURAL GAS
11.2 Natural Gas Sector
atural gas systems are a leading source of anthropogenic CH4 emissions, accounting for
more than 970 MtCO2eq (USEPA, 2006). The USEPA estimates that natural gas systems
account for 8 percent of total global CH4 emissions. The Russian Federation, the United
States, Africa, and Mexico account for more than 43 percent of the world's CH4 emissions in the natural
gas sector (USEPA, 2006) (Figure 2-1).
Figure 2-1: CH4 from Natural Gas Systems by Country: 2000-2020
Mexico
Africa
United States
ussian Federation
Rest of the world
Source: USEPA, 2006.
Emissions are projected to increase 54 percent from 2005 to 2020, with Brazil and China having the
largest growth of 737 percent and 611 percent, respectively (USEPA, 2006). The two regions projected to
experience the largest growth in production are the Middle East and the developing countries of Latin
America.
11.2.1 Introduction
Natural gas systems include the production, processing, transportation and storage, and distribution
of natural gas. Table 2-1 identifies facilities and equipment associated with different segments of the
natural gas system.
During production, gas exit swells under pressure greater than 1,000 pounds per square inch (psi).
The gas is routed through dehydrators, where water and other liquids are removed, and then to small-
diameter gathering lines for transport to either processing plants or injection directly into transmission or
distribution pipelines. Processing plants further purify the gas by removing natural gas liquids, sulfur
compounds, particulates, and CO2. Impurities in the gas are extracted through a cooling process that
forces the impurities to condense into a liquid, which is then vaporized in a reboiler and vented into the
atmosphere.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-15
-------
SECTION II — ENERGY • NATURAL GAS
Table 2-1: Natural Gas Industry Characterization
Segment
Facility
Equipment at the Facility
Production Wells, central gathering facilities
Processing Gas plants
Transmission and Transmission pipeline networks, compressor
storage stations, meter and pressure-regulating stations,
underground injection/withdrawal facilities,
liquefied natural gas (LNG) facilities
Distribution Main and service pipeline networks, meter and
pressure-regulating stations
Wellheads, separators, pneumatic devices,
chemical injection pumps, dehydrators,
compressors, heaters, meters, pipelines
Vessels, dehydrators, compressors, acid gas
removal (AGR) units, heaters, pneumatic devices
Vessels, compressors, pipelines, meters/pressure
regulators, pneumatic devices, dehydrators,
heaters
Pipelines, meters, pressure regulators, pneumatic
devices, customer meters
Source: USEPA, 1996.
Processed gas, which is 95 percent CH4, is then injected into large-diameter transmission pipelines,
where it is compressed and transported to storage and distribution facilities. Storage stations are either
above- or belowground facilities and include compressor stations. Distribution companies reduce high-
pressure gas (averaging 300 psi to 600 psi) to pounds or even ounces per square inch for delivery to
homes, businesses, and industries.
CH4 emissions occur from normal operations in each of the four segments of the natural gas industry.
Equipment/pipeline leaks and venting activities are the primary sources of CH4 emissions in the natural
gas sector (USEPA, 1996). As the gas moves through system components under extreme pressure, CH4
can escape to the atmosphere through worn valves, flanges, pump seals, compressor seals, and joints or
connections in gathering pipelines. For example, in the production segment of the natural gas system,
emissions occur at the wellhead, during dehydration, and when the gas is compressed to be transported
from the wellhead site to a processing plant. CH4 emissions also occur during routine maintenance
throughout the natural gas system. For example, emissions from the transmission segment include
intentional blowdown or purge activities during maintenance and inspection.
Abatement options for the natural gas sector generally fall into three categories: equipment
changes/upgrades, changes in operational practices, and direct inspection and maintenance (DI&M).
Many abatement options are applicable across all four segments of the natural gas system described in
Table 2-1.
• Natural gas emissions from pneumatic control devices are one of the largest sources of CH4
emissions in the natural gas industry. Substituting compressed air for pressurized natural gas
throughout the natural gas system eliminates the constant bleed of natural gas to the atmosphere.
• Changing operational practices, such as using pumpdown techniques to remove product (i.e.,
natural gas) from sections of pipeline and the compressor during maintenance and repair,
reduces the volume of natural gas vented to the atmosphere when components are taken offline.
• Implementing Dl&M programs can eliminate as much as 80 percent of fugitive CH4 emissions
that result from equipment and pipeline leaks throughout the system.
The following sections discuss the activity data and emissions factors used to develop baseline
emissions, abatement options and their costs, and CH4 MACs for natural gas systems for selected
countries. The chapter concludes with sensitivity analyses on key assumptions and a discussion of
uncertainties and limitations.
11-16
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY • NATURAL GAS
1.2.2 Baseline Emissions Estimates
Annual emissions baselines for natural gas systems are calculated using activity factors, activity
factor drivers, and emissions factors. Each of these factors can be affected by variations in individual
countries' production process techniques, the intensity of maintenance schedules, and the age of the
natural gas system. Table C-7 (see Appendix C) lists the activity factors, emissions factors, and emissions
for sources in the United States.
11.2.2.1 Activity Data
Activity factors and activity factor drivers are used to estimate the population of equipment in each
segment of the natural gas system.
Activity Factors
Activity factors include both the physical number of units and the level of operation/activity of these
units. These factors inform the underlying population for each type of equipment present in a natural gas
system. Examples of activity factors include the number of compressors in the production segment, the
throughput across segments, miles of pipeline, number of blowdowns, and the total number of gas wells.
Activity factors are used in conjunction with emissions factors (discussed below) to calculate annual
baseline emissions. This report uses activity factors used to characterize the U.S. natural gas system in
1992 (USEPA, 1996).
Activity Factor Drivers
Activity factor drivers are used to adjust the activity factors from 1992 to reflect changes over time or
differences between countries. The primary drivers are changes in production and consumption levels,
but drivers can also include changes in the age or underlying technology of natural gas systems. Activity
factor drivers determine how the equipment population numbers fluctuate in response to expanding or
contracting natural gas markets. For example, the number of dehydrators in a natural gas system is
determined by the number of wells, which is driven by production levels. If production of natural gas
drops, the number of wells required decreases. This drives down the number of dehydrators in operation
(or the operating capacity of dehydrators in place), reducing the baseline emissions estimate.
Historical Activity Data
Historically, natural gas has been produced by developed countries, which have the technology base
and capital available to facilitate the development of natural gas industries. In 2001, the FSU and the
United States accounted for 33 percent of the world's natural gas production (91.1 trillion cubic feet)
(USEIA, 2005a). Table 2-2 reports natural gas production by country and region for 1980 through 2003.
During the past 20 years, natural gas consumption has increased (see Table 2-3). Developing
countries have experienced the largest increase in consumption in recent years, while industrialized
countries have experienced small but steady growth over the same period. Currently, developing
countries consume significantly less natural gas than developed countries; however, this trend is
projected to change in the next 5 to 10 years.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-17
-------
SECTION II — ENERGY • NATURAL GAS
Table 2-2: Natural Gas Production by Country and Region: 1980-2003 (Trillion Cubic Feet)
Country/Region
Canada
Mexico
United States
North America
Antarctica
Central and South America
Netherlands
Norway
United Kingdom
Western Europe
Russian Federation
Turkmenistan
Uzbekistan
Eastern Europe and FSU
Iran
Saudi Arabia
United Arab Emirates
Middle East
Algeria
Africa
Indonesia
Malaysia
Asia and Oceania
World Total
1980
2.76
0.90
19.40
23.06
0.00
1.23
3.40
0.92
1.32
7.46
NA
NA
NA
17.06
0.25
0.33
0.20
1.42
0.41
0.69
0.63
0.06
2.44
53.35
1990
3.85
0.90
17.81
22.56
0.00
2.01
2.69
0.98
1.75
7.24
NA
NA
NA
30.13
0.84
1.08
0.78
3.72
1.79
2.46
1.53
0.65
5.44
73.57
1995
5.60
0.96
18.60
25.16
0.00
2.58
2.98
1.08
2.67
8.80
21.01
1.14
1.70
25.93
1.25
1.34
1.11
4.99
2.05
3.01
2.24
1.02
7.50
77.96
2000
6.47
1.31
19.18
26.97
0.00
3.43
2.56
1.87
3.83
10.19
20.63
1.64
1.99
26.22
2.13
1.76
1.36
7.57
2.94
4.44
2.36
1.50
9.48
88.29
2001
6.60
1.30
19.62
27.51
0.00
3.65
2.75
1.95
3.69
10.27
20.51
1.70
2.23
26.48
2.33
1.90
1.39
7.98
2.79
4.63
2.34
1.66
9.92
90.45
2002
6.63
1.33
18.93
26.89
0.00
3.72
2.66
2.41
3.61
10.55
21.03
1.89
2.04
27.05
2.65
2.00
1.53
8.67
2.80
4.74
2.48
1.71
10.53
92.15
2003
6.45
1.49
19.04
26.98
0.00
4.20
2.58
2.59
3.63
10.62
21.77
2.08
2.03
28.00
2.79
2.12
1.58
9.12
2.91
5.07
2.62
1.89
11.19
95.18
Source: USEIA, 2005b.
FSU = Former Soviet Union; NA = Data unavailable.
Projected Activity Data
Production and consumption of natural gas are expected to increase in the near term, with
developing countries experiencing the largest percentage increases over the next 20 years. Table 2-4 and
Table 2-5 list projected natural gas production and consumption, respectively, by selected country and
region from 2010 to 2025. Annual growth in production in Central and South America and Africa is
expected to approach 5 percent. However, the United States, Eastern Europe, and the FSU are still
projected to account for more than 50 percent of world natural gas production in 2025 (USEIA, 2004).
Natural gas is projected to be the fastest growing source of primary energy over the next 20 years.
Consumption is expected to increase by more than 70 percent (average annual rate of 2.2 percent) from
2001 to 2025 (USEIA, 2005a). Developing countries will continue to experience the largest percentage
increases in demand.
11-18
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY « NATURAL GAS
Table 2-3: Natural Gas Consumption by Country and Region: 1980-2003 (Trillion Cubic Feet)
Country/Region
Canada
Mexico
United States
North America
Central and South America
France
Germany
Italy
Netherlands
United Kingdom
Western Europe
Russian Federation
Ukraine
Uzbekistan
Eastern Europe and FSU
Iran
Saudi Arabia
United Arab Emirates
Middle East
Africa
China
Indonesia
Japan
Asia and Oceania
World Total
1980
1.88
0.80
19.88
22.56
1.24
0.98
NA
0.97
1.49
1.70
8.66
NA
NA
NA
15.86
0.23
0.33
0.11
1.31
0.74
0.51
0.20
0.90
2J2
52.89
1990
2.38
0.92
19.17
22.47
2.02
1.00
NA
1.67
1.54
2.06
10.50
NA
NA
NA
27.83
0.84
1.08
0.66
3.60
1.35
0.49
0.55
1.85
5.61
73J7
1995
2.79
1.04
22.21
26.04
2.58
1.18
3.17
1.92
1.70
2.69
12.76
14.51
2.97
1.35
23.04
1.24
1.34
0.88
474
1.69
0.58
1.06
2.21
7.79
78.64
2000
2.95
1.40
23.33
27.68
3.30
1.40
3.10
2.50
1.73
3.37
15.13
14.13
2.78
1.51
22.80
2.22
1.76
1.11
6.82
2.04
0.93
1.08
2.84
10.43
88.21
2001
2.91
1.40
22.24
26.55
3.54
1.47
3.24
2.51
1.77
3.34
15.51
14.41
2.62
1.60
23.30
2.48
1.90
1.15
7.05
2.28
1.05
1.18
2.84
11.08
89.31
2002
3.06
1.50
23.01
27.57
3.56
1.59
3.20
2.49
1.76
3.31
15.87
14.57
2.78
1.64
23.68
2.80
2.00
1.29
7.63
2.45
1.13
1.20
2.94
11.76
92.51
2003
3.21
1.82
22.38
27.41
3.82
1.54
3.32
2.72
1.78
3.36
16.43
15.29
3.02
1.67
2497
2.79
2.12
1.34
7.86
2.55
1.18
1.23
3.05
12.46
95.50
Source: USEIA, 2005b.
FSU = Former Soviet Union; NA = Data unavailable.
II.2.2.2 Emissions Factors and Related Assumptions
Emissions factors in the natural gas sector are defined as the rate of CH4 emissions from a facility or
piece of equipment or from normal operations and routine maintenance. Estimated emissions factors are
used in conjunction with activity factors and activity factor drivers to generate baseline emissions
estimates by country. Table 2-6 reports estimated emissions factors by country, provided by IPCC's
Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories: Reference Manual. These emissions
factors represent the average estimated emissions factor across all segments of the natural gas system.
The system-level emissions factors in Table 2-6 are used to calculate country-specific baseline
emissions (see Section II.2.2.3) for countries outside the United States. For the United States, a more
detailed set of emissions factors is used to calculate baseline emissions. Appendix Table C-7 presents the
individual facility and equipment emissions factors estimated for the U.S. natural gas system, adapted
from the USEPA report Methane Emissions form the Natural Gas Industry (USEPA, 1996).
This section discusses the source of the emissions factors used to develop country-specific baseline
emissions.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-19
-------
SECTION II — ENERGY • NATURAL GAS
Table 2-4: Projected Natural Gas Production by Country and Region: 2010-2025 (Trillion Cubic Feet)
Country/Region
Canada
Mexico
United States
North America
Central and South America
Western Europe
Eastern Europe and FSU
Middle East
Africa
China
Asia
World Total
2010
7.6
1.5
20.5
29.6
5.5
9.0
31.0
9.8
8.1
1.6
12.5
105.5
2015
7.5
1.6
21.6
30.6
7.1
9.0
35.7
12.1
9.9
1.9
14.2
118.5
2020
7.1
1.9
23.8
32.8
8.6
8.9
40.4
15.6
11.9
2.3
16.3
134.5
Average Annual
Percentage Change,
2025 2001-2025
7.5
2.1
24.0
33.6
10.6
9.8
45.3
18.8
14.1
3.1
18.8
151.0
0.5
2.0
0.8
0.8
4.6
• -0.2
2.2
3.5
4.8
4.5
2.6
2.1
Source: USEIA, 2004.
FSU = Former Soviet Union
Table 2-5: Projected Natural Gas Consumption by Country and Region: 2010-2025 (Trillion Cubic Feet)
Country/Region
Canada
Mexico
United States
North America
Brazil
Other Central/South America
Western Europe
Russian Federation
Eastern Europe
FSU
Middle East
Africa
China
Emerging Asia
World Total
2010
3.9
1.8
25.6
31.3
0.9
3.8
17.3
16.2
4.0
. 25.6
10.6
3.1
2.6
10.6
111.4
2015
4.3
2.2
28.3
34.8
1.3
4.3
19.0
17.9
4.6
29.0
12.6
4.1
3.4
13.3
127.9
2020
4.6
2.6
30.4
37.6
1.7
4.8
20.4
19.5
5.2
31.0
14.5
4.9
4.2
16.3
141.6
Average Annual
Percentage Change,
2025 2001-2025
4.7
3.0
30.9
38.6
2.1
5.4
22.4
20.7
5.8
33.3
16.6
6.0
6.5
20.7
156.2
2.0
3.0
1.3
1.5
6.8
2.4
1.8
1.5
3.5
2.0
3.1
4.0
7.8
4.3
2.3
Source: USEIA, 2005c.
FSU = Former Soviet Union.
11-20
GLOBAL MITIGATION OF NOIM-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY • NATURAL GAS
Table 2-6: IPCC Estimated Emissions Factors from Natural Gas by Region
Emissions Factors by Industry Segment
(kg/petajoule)
Country/Region
Eastern Europe/FSUa
Other oil-exporting countries"
United States and Canada
Western Europe6
Rest of the world"
Production
392,800
67,795
71,905
20,900
67,795
Consumption
527,900
228,310
88,135
84,500
228,310
Source: IPCC, 1996. Adapted from Reference Manual Tables 1-60,1-61,1-62,1-63, and 1-64.
FSU = Former Soviet Union
a Includes Albania, Bulgaria, Czech and Slovak Republics, Hungary, Poland, Romania, and the former Yugoslavia,
b Includes Algeria, Nigeria, Venezuela, Indonesia, Iran, Iraq, Kuwait, Saudi Arabia, United Arab Emirates, Ecuador, and Mexico.
c Includes Austria, Belgium, Denmark, Faroe Islands, Finland, France, Germany, Gibraltar, Greece, Iceland, Ireland, Italy, Luxembourg, Malta,
the Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, and the United Kingdom.
d Includes Asia, Africa, Middle East, Oceania, and Latin America.
Historical Emissions Factors
The United States conducted a study to measure and estimate emissions factors for all components in
its national infrastructure (USEPA, 1996). This study measured or estimated emissions factors for more
than 100 pieces of natural gas equipment, such as gas wells, compressors, pipeline, and system upsets.
The study was conducted in 1992, and the emissions factors were revised and published in 1996.
Table C-7 (see Appendix C) lists the study's emissions factors by component and segment of the
infrastructure. These emissions factors are used to calculate the U.S. baseline emissions estimate (see
Table 2-7). For all other countries, IPCC systems emissions factors (Table 2-6) were used to develop
baseline emissions estimates.
Table 2-7: Baseline Emissions for Natural Gas Systems for Selected Countries: 1990-2000 (MtC02eq)
Country
Russian Federation
United States
Iran
Mexico
Ukraine
Turkmenistan
Nigeria
Venezuela
Turkey
India
United Arab Emirates
Uzbekistan
Indonesia
Canada
Argentina
Rest of the world
World Total
1990
334.3
143.9
19.4
22.7
78.3
19.5
12.5
29.8
19.9
8.0
18.9
27.2
31.3
25.4
8.0
132.0
931.0
1995
240.6
148.0
29.1
25.3
81.8
16.7
17.6
34.8
28.5
12.5
26.7
30.3
41.4
34.3
10.9
145.3
923.8
2000
165.3
145.7
34.6
37.4
86.9
24.3
37.8
37.7
38.7
15.8
33.2
34.8
42.1
37.3
14.9
186.0
972.4
Source. USEPA, 2006
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-21
-------
SECTION II — ENERGY • NATURAL GAS
Projected Emissions Factors and Related Assumptions
Over time, the USEPA estimates that the proportional growth in baseline CH4 globally will slow
relative to the growth in overall production and consumption. Emissions factors in mature natural gas
systems are projected to increase because of equipment age and fatigue. However, this increase will be
counterbalanced by rapidly expanding industries in developing countries that will employ state-of-the-
art technology when constructing natural gas infrastructures.
For example, China is in the early stages of developing a natural gas infrastructure. China's use of
state-of-the-art technology supplied by the United States, the European Union (EU), and Japan will result
in low emissions factors, and these low emissions factors will constrain the growth in China's national
baseline emissions over time.
11.2.2.3 Emissions Estimates and Related Assumptions
The USEPA estimates the emissions contribution of each segment in the natural gas system by
multiplying emissions factors (EF) by associated activity factors (AF) and then summing them, as shown
below:
Country Total Emissions = Production (EF * AF) + Processing (EF * AF) + (2.1)
Transport (EF * AF) + Storage (EE x AF) +
Distribution (EF x
From Equation (2.1), individual country baseline estimates using natural gas production and
consumption data are coupled with the IPCC system emissions factors presented in Table 2-6. This
section discusses the historical and projected changes in the baseline emissions estimates.
Historical Emissions Estimates
Baseline emissions are built using publicly available reports produced by the countries themselves.
IPCC guidelines and methods were used to estimate emissions in each country, ensuring comparability
across countries. Table 2-7 presents the countries with the largest historical CH4 baseline emissions for
1990, 1995, and 2000. CH4 emissions increased worldwide from 1990 to 2000 at an average annual rate of
3 percent.
Projected Emissions Estimates
Overall, world CH4 emissions are expected to increase during the next 20 years at an average annual
rate of 5.7 percent (USEPA, 2003a), reflecting a projected increase in natural gas use as a share of total
energy consumption. Table 2-8 presents the predicted CH4 baseline emissions for the largest emitting
countries in the global natural gas sector. Developing countries will experience the largest percentage
increases in emissions, which closely parallel expected increases in consumption and production of
natural gas. However, the level of technology employed in building new infrastructure will help
constrain baseline emissions for these countries.
H-22 GLOBAL Mil IGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY • NATURAL GAS
Table 2-8: Projected Baseline Emissions for Natural Gas Systems for Selected Countries: 2005-2020
(MtC02eq)
Country
Russian Federation
United States
Iran
Mexico
Ukraine
Turkmenistan
Nigeria
Venezuela
Turkey
India
United Arab Emirates
Uzbekistan
Indonesia
Canada
Argentina
Rest of the work)
World Total
2005
171.9
124.3
56.8
49.5
90.4
46.2
49.1
45.2
50.2
25.8
38.7
39.6
46.8
37.3
14.9
213.8
1,100.4
2010
178.6
138.6
74.0
64.0
93.9
72.1
59.2
50.7
56.6
35.7
47.4
44.3
48.0
38.2
16.7
253.4
1,271.5
2015
185.8
151.0
96.4 •
82.6
97.7
83.2
73.3
63.0
62.9
49.5
52.8
45.4
46.3
39.8
20.9
313.0
1,463.7
2020
193.1
164.8
125.3
111.4
101.5
93.9
89.4
84.8
75.5
61.4
59.7
46.8
45.2
41.1
28.1
373.8
1 ,695.8
Source. USEPA, 2006.
II.2.3 Cost of CH4 Emissions Reductions from Natural Gas Systems
Capital costs, annual costs, and annual benefits for individual abatement options are obtained from
the USEPA's economic cost model. The economic cost model incorporates activity and emissions factors
published by the USEPA and the Gas Research Institute (GRI) (USEPA, 1996). The USEPA's economic
cost model reports one-time capital costs, annual operating costs, and reduction efficiencies for 118
different abatement options applied across the four sectors: production, processing, transmission and
storage, and distribution. Options range from upgrading compressors and pipes to enhancing inspection
and detection techniques. The number of options by sector is presented in Table 2-9. Table C-8 (in
Appendix C) contains a brief description of the major categories of natural gas abatement options.
It should be noted that a large number of abatement options for the natural gas sector are substitutes
for each other. Thus, there may be several options for reducing emissions for a particular piece of
equipment, but only one may be selected. For example, DI&M of gas wells is substitutable with enhanced
DI&M. In developing the MACs, the model chooses between substitute options, selecting the option with
the lowest breakeven price.
Table 2-9: Prevalence of Abatement Options by Infrastructure Component
Infrastructure Component
Production
Processing
Transmission and storage
Distribution
Total
39
2
51
26
Source: USEPA, 2000.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES H-23
-------
SECTION II — ENERGY • NATURAL GAS
11.2.3.1 Abatement Option Opportunities
This section presents a general overview of the applicable abatement options for each segment of the
natural gas system, followed by a more detailed discussion of the costs and benefits of selected abatement
options. Engineering cost and benefit estimates represent equipment and operating costs in the United
States for 1999. Whereas some abatement options are unique to a specific segment of the natural gas
system, many are applicable in multiple segments.
Production Abatement Options
The production segment of the natural gas sector consists of wells, compressors, dehydrators,
pneumatic devices, chemical injection pumps, heaters, meters, pipeline, and central gathering facilities.
Abatement technologies associated with the production segment include
• catalytic converters for select well field engines and compressors,
• replacement of wet seals with dry seals in centrifugal compressors,
• direct/enhanced inspection and maintenance at production sites,
• flash tank separator installation in glycol dehydration systems,
• replacement of high-bleed pneumatic devices, and
• optimization of glycol recirculation rates.
One example of technology available to the production segment reduces glycol recirculation rates.
Producers use triethylene glycol (TEG) in dehydrators to remove water from the natural gas coming out
of the ground to meet pipeline quality standards. "Dry" TEG is combined with natural gas to remove
moisture content before the natural gas is sold into a pipeline. The "rich" TEG then enters a boiler, where
the foreign substances are evaporated and emitted into the atmosphere and the cycle repeats itself. The
rate at which this process occurs is directly proportional to the amount of CH4 emitted from glycol
dehydrators. Production fields become less productive over time, but the rate at which the TEG
recirculates is commonly based on the initial rate of production. As the well site matures, the TEG
circulation rate becomes oversized. Recirculation can be recalculated to achieve sufficient moisture
removal from the gas and minimize the release of CH4 from the system. The following are the cost
components for this abatement option:
• Capital Costs. This abatement option requires minimal or no additional equipment. However,
similar to inspection and maintenance programs, the option is labor intensive, with the
calculations and circulation adjustments conducted by engineering staff.
• Annual Costs. Annual costs primarily include the labor required to calculate new optimal
recirculation rates each year as the well site becomes less productive.
• Cost Savings/Benefits. More CH4 is brought to market for sale.
Processing Abatement Options
The processing segment consists of gas plant facilities that incorporate the use of vessels,
dehydrators, compressors, acid gas removal (AGR) units, heaters, and pneumatic devices. Abatement
technologies associated with the processing segment include
• fuel gas retrofit for reciprocating compressors,
• replacement of wet seals with dry seals in centrifugal compressors,
• conversion of gas pneumatic controls to instrument air, and
• DI&M at gas processing plants.
M-24 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY • NATURAL GAS
One example of abatement technology available to the processing segment converts gas pneumatic
controls to compressed instrument air systems. Processing plants use pneumatic control systems to
monitor various gas and liquid levels. As part of their normal operations, these devices release or bleed
CH4 into the atmosphere. Processing plants can substitute compressed air for natural gas within
pneumatic systems. The following are the cost components for this abatement option:
• Capital Costs. Capital costs include the purchase and installation of a compressor, dehydrator,
and volume tank—the major components of the instrument air system. Depending on the size of
the gas processing plant, capital costs are estimated to be between $4,500 and $35,000 for the
required capital equipment.
• Annual Costs. Annual costs include the annual energy, materials, and labor required to operate
and monitor the equipment used in the compressed instrument air system. Annual energy costs
are determined by the size of the compressor. Annual servicing costs range from $800 to $3,600
per year.
• Cost Savings/Benefits. By replacing natural gas with compressed instrument air, CH4 is no
longer being vented during normal operations. The benefit is the market value of CH4 abated.
Transmission Abatement Options
The transmission segment of a natural gas system consists of transmission pipeline networks,
compressor stations, and meter and pressure-regulating stations. The following are abatement
technologies available to the transmission segment:
• conversion of gas pneumatic controls to instrument air,
• use of pipeline pumpdown techniques to lower gas line pressure before maintenance,
• DI&M at compressor stations and surface facilities,
• replacement of wet seals with dry seals in centrifugal compressors, and
• replacement of compressor rod packing systems.
One example of the abatement options available to the transmissions segment is DI&M at compressor
stations. Compressor stations amplify pressure at several stages along a transmission natural gas pipeline
to combat pressure loss over long distances. Over time, compressors and other related components
become fatigued and may leak CH4. The DI&M program reduces CH4 emissions at compressor stations
by identifying leaks and focusing maintenance on the largest leaks. The following are the cost
components for this abatement option:
• Capital Costs. Capital costs include the cost of purchasing a leak detection device, which varies
widely depending on the type of device used. The cost of screening devices ranges from $1,000 to
$20,000. The cost of more sensitive sampling devices ranges from $1,000 to $10,000.
• Annual Costs. Annual costs include the cost of labor and materials to develop a maintenance
schedule and implement the survey and maintenance annually. Annual costs account for the
majority of costs associated with implementing this abatement option.
• Cost Savings/Benefits. Cost savings are approximately $3 per thousand cubic feet (Mcf) of CH4
recovered. The savings will depend on the intensity of the DI&M program and whether the leak,
once detected, is fixed. The average station leak rate is approximately 41,000 Mcf per year, and
the average annual cost savings is $88,000 at a gas price of $3 per Mcf (USEPA, 2003b).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-25
-------
SECTION II — ENERGY • NATURAL GAS
Distribution Abatement Options
The distribution segment consists of main and service pipeline networks, meter and pressure-
regulating stations, pneumatic devices, and customer meters. Abatement technologies available to the
distribution segment include the
• use of hot taps in service pipeline connections,
• DI&M at gate stations,
• use of composite wrap for nonleaking pipeline defects, and
• use of a pipeline pumpdown technique to lower gas line pressure before maintenance.
An example abatement option available to the distribution segment is the use of a pipeline
pumpdown technique when performing maintenance on segments of distribution pipeline. Operators
routinely reduce line pressure and discharge gas from a pipeline during maintenance and repair
activities. Using a pumpdown technique, which requires the use of inline and/or portable compressors to
depressurize the section of pipeline, operators can mitigate CH4 emissions. The following are the cost
components for this abatement option:
• Capital Costs. Capital costs include the one-time costs of purchasing a portable compressor. The
cost of this compressor varies by size and ranges from $500,000 (300 psi) to $3,000,000 (1,000 psi).
Installation and freight costs are determined by the size of the compressor purchased.
• Annual Costs. Annual costs include fuel/energy, maintenance, and labor costs. Average energy
costs vary based on the compressor's horsepower rating. Maintenance costs range from $4 to $9
per horsepower per month.
• Cost Savings. Cost savings will vary depending on the volume of gas available for recovery. The
volume of gas available is determined by the length of pipeline to be repaired and the flow rate of
gas during normal operations.
11.2.4 Results
This section presents the EMF-21 study's MAC results in tabular format.
11.2.4.1 Data Tables and Graphs
Table 2-10 presents the average breakeven price and the reduction in absolute and percentage terms
for the mitigation options discussed in Section II.2.3.1.
The EMF regional baselines and MAC results of the EMF-21 study are presented in Tables 2-11 and 2-
12 for 2010 and 2020 using a base energy price, a 10 percent discount rate, and a 40 percent tax rate. These
MACs represent static percentage reductions in baselines for individual regions/countries and represent
the official MACs used in climate change modeling. Figure 2-2 provides MACs for the five EMF
countries/regions with the largest estimated emissions for natural gas systems in 2020.
The MACs presented in this section represent static abatement curves using breakeven prices built on
the assumption of fixed mitigation cost and aggregate countrywide natural gas statistics. Appendix C to
this chapter presents more recent efforts to develop an alternative framework for conducting MAC
analysis that addresses the limitations of the EMF-21 MAC analysis.
11-26 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY • NATURAL GAS
Table 2-10: Natural Gas MACs for Countries Included in the Analysis
Technology
Emissions Emissions Emissions
Breakeven Reduction (% Reduction in Reduction in
Cost from 2010
baseline) (MtCOzeq)
Assuming a 10% discount rate and a 40% tax rate
P&T—compressors altering start-up procedure
Prod-D I&M (chemical inspection pumps)
Prod-D I&M (enhanced)
Prod-D I&M (offshore)
Prod-D t&M (onshore)
Prod-D I&M (pipeline leaks)
(production)
Installation of ffash tank separators (production)
(production)
Mft
• P&T-D I&M (compressor enhanced)
'
^
$113.36
-11522
$121.98
$836.05
$49,51
$55,82
$9,829.72
$85.47
$3,233.11
$178.81
-$25.03
-$12.22
$8,774.06
$36.75
-$mt?
-$27.55
$78,81
-$85.24
-$24,45
$10057
$2,863.14
compressors (P&T)
$7.57
$17SJt
$34.30
4%
0%
0%
0%
0%
0%
1%
2%
0%
0%
5%
4%
0%
3%
2%
0%
3%
0%
0%
2%
0%
0.21
0.01
0.01
0.01
0.01
0.01
0.07
0.00
0.09
0.00
0.00
0.01
0.23
0.20
0.00
0.16
0.01
0.16
0.01
0.02
8.00
om
0.00
MO
ttl
0.27
0.01
0.01
0.01
0.01
0.01
0.09
0,00
0.10
0.00
0.00
0.02
0.27
OJ3
0.00
0.20
0.10
0.01
0.2D
0.03
0.03
QM
om
0.13
(continued)
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-27
-------
SECTION II — ENERGY . NATURAL GAS
Table 2-10: Natural Gas MACs for Countries Included in the Analysis (continued)
Technology
Replace high-bleed pneumatic devices with
compressed air systems (P&T)
Replace high-bleed pneumatic devices with low-
bleed pneumatic devices (P&T)
Surge vessels for station/well venting (P&T)
D-D I&M (distribution)
D-D I&M (enhanced: distribution)
Electronic monitoring at large surface facilities (D)
Replacement of cast iron/unprotected steel pipeline
P)
Replacement of unprotected steel services (D)
Emissions
Breakeven Reduction (%
Cost from
($/tC02eq) baseline)
$88.69
-$12.22
$8,774.06
-$23.20
$21.02
$0.76
$19,347.78
$461,544.32
2%
2%
1%
2%
4%
5%
7%
3%
Emissions
Reduction in
2010
(MtC02eq)
0.09
0.08
0.08
0.12
0.22
0.27
0.34
0.14
Emissions
Reduction in
2020
ptCQaeq)
0.11
0.10
0.09
0.15
0.27
0.33
0.42
0.17
Source: USEPA, 2003a. Adapted from Natural Gas Sector technology tables in Appendix B.
D = Distribution; I&M = Inspection and maintenance; P = Production; T = Transmission.
Table 2-11: Baseline Emissions by EMF Regional Grouping: 2000-2020 (MtC02eq)
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan '
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
65.7
517.3
6.1
1.8
1.9
8.5
25.2
15.8
0.4
37.4
255.9
301.9
165.3
71.7
145.7
972.4
2010
95.7
556.9
9.6
6.9
5.8
12.2
25.4
35.7
0.4
64.0
277.0
349.8
178.6
85.5
138.6
1,271.5
2020
144.5
639.2
15.2
14.9
13.2
17.7
26.4
61.4
0.4
111.4
299.6
459.4
193.1
105.8
164.8
1,695.8
Source: USEPA, 2006.
EU-15 = European Union, OECD = Organisation for Economic Co-operation and Development.
Note: World Total does not equal the sum of the countries listed in this table because the regional groupings are a subset of the full EMF
regional grouping list. See Appendix A of this report for the full EMF list of countries by region.
11-28
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY • NATURAL GAS
Table 2-12: Natural Gas MACs for Countries Included in the Analysis
Percentage Reduction from Baseline (tCC^eq)
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China
Eastern
EU-15
India
Japan
Mexico
Europe
Non-OECD Annex 1
OECD
Russian
South &
Federation
SEAsia
United States
World Total
$0
20.38%
9.60%
14.44%
16.64%
17.05%
19.05%
11.58%
10.70%
28.05%
11.06%
6.26%
13.86%
3.75%
11.51%
14.52%
10.11%
$15
29.98%
24.21%
20.06%
25.42%
36.78%
25.84%
18.38%
28.15%
28.12%
23.15%
27.29%
20.73%
26.85%
29.75%
19.24%
24.98%
2010
$30
37.85%
31.68%
29.35%
36.87%
43.33%
34.03%
28.39%
36.44%
32.51%
37.02%
33.72%
29.85%
33.14%
37.75%
28.14%
32.95%
$45
43.62%
35.12%
36.94%
43.54%
44.11%
34.22%
29.01%
43.49%
46.17%
43.55%
35.50%
35.60%
35.11%
43.61%
35.47%
37.90%
$60
56.03%
\
50.46%
" 56.54%
57.79%
45.92%
48.71%
49.18%
58.74%
61.10%
57.62%
48.29%
53.75%
48.42%
56.22%
54.76%
53.36%
2020
$0
20.38%
8.98%
14.44%*
16.64%
17.05%
19.05%
11.58%
10.70%
28.05%
11.06%
6.09%
12.14%
3.75%
11.51%
14.52%
10.19%
$15
29.98%
22.63%
20.06%
25.42%
36.78%
25.84%
18.38%
28.15%
28.12%
23.15%
26.56%
18.17%
26.85%
29.75%
19.24%
25.25%
$30
37.85%
29.62%
29.35%
36.87%
43.33%
34.03%
28.39%
36.44%
32.51%
37.02%
32.81%
26.16%
33.14%
37.75%
28.14%
33.24%
$45
43.62%
32.83%
36.94%
43.54%
44.11%
34.22%
29.01%
43.49%
46.17%
43.55%
34.54%
31.20%
35.11%
43.61%
35.47%
38.40%
S60
56.03%
47.18%
56.54%
57.79%
45.92%
48.71%
49.18%
58.74%
61.10%
57.62%
46.99%
47.11%
48.42%
56.22%
54J6%
53.81%
Source: USEPA, 2003a.
EU-15 = European Union.
Figure 2-2:
$50 n
$40-
$30 -
O"
O $20-
O
*•»
t»
$10 -
(Cfl
(
-$10 -
EMF MACs for Top Five Emitting Countries/Regions from Natural Gas: 2020
y * ' - • ' ' ' ' ' '
f _J • • W 20" 30 40 50 60 70 80
Absolute Reduction (MtCO2eq)
Source: USEPA, 2003a.
Note: This table was constructed using percentage reductions from USEPA (2003), with baselines from USEPA (2005).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-29
-------
SECTION II — ENERGY • NATURAL GAS
11.2.5 Summary
The methodology and data discussed in this section describe the MAC analysis conducted for the
natural gas sector by the EMF-21 study. MACs for 2010 and 2020 were estimated based on aggregated
industry data from each country or region. The MACs represent static estimates of potential CH4
mitigation from natural gas systems based on available information regarding infrastructure and
country-reported emissions estimates provided through the United Nation's Framework Convention on
Climate Change emissions inventory reports.
11.2.6 References
Intergovernmental Panel on Climate Change (IPCC). 1996. Revised 1996 IPCC Guidelines for National
Greenhouse Gas Inventories: Reference Manual (Volume 3). Available at . As obtained on April 26, 2004.
U.S. Department of Energy (USDOE), Energy Information Administration (USEIA). 2004. International
Energy Outlook 2004. Table 11. DOE/EIA-0484(2004) Washington, DC: USEIA. Available at
.
U.S. Department of Energy (USDOE), Energy Information Administration (USEIA). 2005a. International
Energy Annual 2003. Table 2.4. Washington, DC: USEIA. Available at . As obtained on May 2, 2006.
U.S. Department of Energy (USDOE), Energy Information Administration (USEIA). 2005b. International
Energy Annual 2003. Table 1.3. Washington, DC: USEIA. Available at . As obtained on May 2, 2006.
U.S. Department of Energy (USDOE), Energy Information Administration (USEIA). 2005c. International
Energy Outlook 2005. DOE/EIA-0484(2005) Table A5. Washington, DC: USEIA. Available at
. As obtained on May 2, 2006.
U.S. Environmental Protection Agency (USEPA). 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). 2000. Spreadsheet Cost Model. Washington, DC: USEPA.
U.S. Environmental Protection Agency (USEPA). 2003a. International Analysis of Methane and Nitrous Oxide
Abatement Opportunities: Report to Energy Modeling Forum, Working Group 21. Appendices. Washington,
DC: USEPA. Available at . As obtained on
September 27, 2004.
U.S. Environmental Protection Agency (USEPA). 2003b. Lessons Learned From Natural Gas STAR Partners.
Washington, DC: USEPA. Available at . As obtained on
August 19, 2003.
U.S. Environmental Protection Agency (USEPA). 2006. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2020. Washington, DC: USEPA.
11-30 GLOBAL MIT IGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION I
-ENERGY. OIL
11.3 Oil Sector
orldwide CH4 emissions from oil production accounted for more than 57 MtCO2eq in 2000
(USEPA, 2006). Oil is the llth largest source of anthropogenic CH4 emissions globally. The
USEPA estimates that oil production contributed approximately 0.5 percent of total global
CH4 emissions in 2000 (USEPA, 2006). Combined, Mexico, Eastern Europe, the United States, and China
accounted for approximately 67 percent of the world's CH4 emissions from oil (Figure 3-1). Global CH4
emissions from oil are expected to grow by approximately 104 percent between 2005 and 2020.
Figure 3-1: CH4 Emissions from Oil Production by Country: 2000-2020
China
United States
Eastern Europe
exico
Rest of the world
Source: USEPA, 2006.
11.3.1 Introduction
Oil production begins by extracting crude oil either from underground production field wells
(onshore) or platform oil derricks (offshore). The process of extracting oil involves drilling a deep well to
access an oil reservoir underground. Once a well is drilled, compressors are used to pressurize the well,
allowing the crude oil to exit the well through the vertical shaft. The compressed oil is transported via
pipeline to a processing system and finally to a storage tank. Marine, rail, and truck tankers are the three
major forms of transportation used by the oil sector to move crude oil from the site of production to the
refinery. Pumping stations regulate the transfer of crude oil from storage tanks or pipelines onto
transport tankers.
CH4 emissions are associated with crude oil production, transportation, and refining operations.
These oil production segments release CH4 into the atmosphere as fugitive emissions, emissions from
operational upsets, and emissions from fuel combustion (USEPA, 2004). In the United States, the largest
emissions sources include high-bleed pneumatic devices, flaring, chemical injection pumps, and oil
wellheads for light crude (USEPA, 2004). Emissions from oil production fields accounted for more than 97
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-31
-------
SECTION II — ENERGY • OIL
percent of the total oil industry emissions. The remaining 3 percent was emitted from crude oil
refinement (2 percent) and transportation (1 percent) (USEPA, 2004).
11.3.1.1 Emissions from Production Field Operations
During production field operations, CH4 is released into the atmosphere via venting, accidental leaks,
and fuel combustion. The USEPA suggests that the majority of emissions come from oil wellheads,
storage tanks, and related field processing equipment such as compressors and chemical and injection
pumps. CH4 emissions from storage tanks, a dominant source of emissions, are created when the CH4
entrained in crude oil under high pressure volatilizes as the oil enters the tank where it is stored at
atmospheric pressure. Equipment leaks and vessel blowdowns during routine maintenance make up the
second largest share of emissions from oil systems. The remaining emissions from field operations are
associated with fugitive leaks and combustion through flares (USEPA, 2004).
Saudi Arabia and the United States were the two largest producers of oil in 2000, producing a
reported 9.2 and 8.1 million barrels of crude oil per day, respectively. However, they do not have the
largest CH4 emissions from oil. Onshore production of oil generates less CH4 emissions than offshore oil
operations, because CH4 produced onshore is more readily captured and transported for use. Oil
production in many of the Organization of the Petroleum Exporting Countries (OPEC) members,
including Saudi Arabia, consists primarily of onshore production operations. In contrast, a large share of
the oil production in Mexico comes from offshore platforms.
11.3.1.2 Emissions from Crude Oil Transportation
Venting activities in transport tanks and marine vessel loading operations account for the majority of
emissions in the transportation segment. Fugitive emissions from floating roof tanks account for the
remainder of oil transportation emissions in the United States (USEPA, 2004).
11.3.1.3 Emissions from Crude Oil Refining
Most of the CH4 entrained in crude oil has already escaped prior to the refining stage. Vented
emissions that occur during normal operations account for the majority of emissions from this sector.
Examples include refinery system blowdowns during routine maintenance and asphalt blowing. Fugitive
leaks and combustion emissions are also a source of emissions. Most fugitive emissions come from leaks
in a refinery's fuel gas system. Combustion emissions result from small amounts of unburned CH4 in
process heater stacks and from unburned CH4 in engine exhausts and flares (USEPA, 2004).
11.3.1.4 Abatement Options
Three abatement options are discussed for the oil sector: flaring, direct use, and reinjection of gas into
oil fields. The installation of a flaring system results in an estimated 98 percent reduction in fugitive
emissions but can be costly in an offshore environment because of technical, environmental, and safety
concerns. Direct use is applicable primarily to oil platforms, because CH4 captured onshore is typically
injected into the pipeline system (and is reflected in the baseline emissions). Reinjection of CH4 back into
the oil production field is an alternative to flaring or direct use and can enhance future oil recovery.
The following sections discuss the activity data and emissions factors used to develop baseline
emissions, abatement options and their costs, and CH4 MACs for oil production for selected countries.
The chapter concludes with sensitivity analysis of key assumptions and a discussion of uncertainties and
limitations.
II-32 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY • OIL
11.3.2 Baseline Emissions Estimates
Baseline emissions from the oil sector are composed of emissions from production field operations,
crude oil transportation, and crude oil refining. These emissions are classified either as fugitive emissions,
vented emissions from operations, or emissions from fuel combustion (USEPA, 2004).
A country's baseline emissions estimate is the product of activity factors and emissions factors. The
following section provides an overview of activity and emissions factors and concludes with a discussion
of historical and projected baselines by type of equipment used.
11.3.2.1 Activity Factors
Activity factors characterize a given industry's size, either as the number of units (e.g., number of
wells or miles of pipeline) or the flow through the units (million barrels [MMbbl] per day or year). The
United States tracks 70 different activity factors for the oil industry. Some of these activity factors change
annually in proportion to rates of crude oil production, transportation, and refinery runs, while others
change in proportion to the number of facilities such as oil wells and petroleum refineries (USEPA, 2004).
A detailed list of the activity factors related to production field operations, transportation, and refining is
provided in Appendix D to this chapter (see Table D-l).
IPCC recognizes that this level of detailed information is not readily available in every country and
therefore offers guidance on more aggregate activity factors that can be used to quantify the size of a
country's oil system. Generally, aggregate activity factors such as production and consumption of oil are
used.
Historical Activity Data
Oil production and consumption rates depend on economic conditions, global demand, and available
reserves. For the purposes of this report, historical activity data were taken from publicly available
reports, either from national communications or, when information was unavailable, from expert
judgment (USEIA, 2005a). Table 3-1 reports oil production for selected countries in MMbbl per day for
1990 to 2003.
Projected Activity Data
Oil production is projected to increase by approximately 43 percent during the next 20 years.
Table 3-2 and Table 3-3 list forecasted estimates for oil production and consumption between 2002 and
2025. In addition to OPEC countries continuing to expand production, Eastern European and some
developing countries are forecasted to experience large proportional growth. Countries from the FSU in
the Caspian Area are expected to experience the largest increase in production between 2002 and 2025,
expanding from 1.66 to 6.22 MMbbl per day. Developing countries in regions such as Africa and the
Middle East are also expected to expand production by 127 percent and 46 percent, respectively.
11.3.2.2 Emissions Factors and Related Assumptions
Emissions factors from oil production are defined as CH4 emissions rates by either equipment type or
operation. Equipment used in crude oil production includes wellheads, compressors, pipelines, storage
tanks, and pneumatic devices. The United States has conducted a detailed bottom-up analysis to estimate
average emissions factors by equipment or operation type. For countries or regions where this level of
detail is unavailable, the IPCC's 1996 Revised Guidelines Reference Manuel provides suggested approximate
average emissions factors for each segment of oil systems for various regions around the world.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-33
-------
SECTION II — ENERGY • OIL
Table 3-1: Oil Production by Country: 1990-2003 (MMbbl per Day)
Country
Saudi Arabia
United States
Russian Federation
Iran
Venezuela
Mexico
China
Norway
Canada
Iraq
United Arab Emirates
United Kingdom
Kuwait
Nigeria
Brazil
World Total
1990
7.0
9.0
N/A
3.1
2.3
3.0
2.8
1.8
2.0
2.1
2.3
1.9
1.2
1.8
0.8
65.5
1995
9.2
8.6
6.2
3.7
3.0
3.1
3.0
2.9
2.4
0.6
2.4
2.8
2.2
2.0
0.9
68.9
2000
9.5
8.1
6.7
3.8
3.4
3.4
3.2
3.3
2.7
2.6
2.6
2.5
2.2
2.2
1.5
75.9
2002
8.8
8.0
7.7
3.5
2.9
3.6
3.4
3.3
2.9
2.0
2.4
2.5
2.0
2.1
1.7
75.0
2003
10.1
7.8
8.5
3.8
2.6
3.8
3.4
3.3
3.0
1.3
2.7
2.3
2.3
2.2
1.8
77.7
Source: USEIA, 2005a. Adapted from Table G-1 in the International Energy Annual 2003.
Historical Emissions Factors
Historical emissions factors have remained relatively constant. Countries use the IPCC's emissions
factors cited in the 1996 Revised Guidelines to estimate annual emissions baselines each year from
publication of the Guidelines to the present. Table 3-4 lists aggregate emissions factors provided by the
IPCC for petroleum system production, transportation, and refinement. These emissions factors are based
on top-down estimates of emissions by industry segment. However, as mentioned earlier, the detailed
bottom-up approach taken by the United States may enable a more accurate estimate of baseline
emissions by country. The U.S. oil industry emissions factors (see Tables D-l, D-2, and D-3) are also
assumed to remain constant in the short term (USEPA, 2004).
IPCC and the United States report higher emissions factors in the production segment than in any
other segment of a petroleum system (IPCC, 1996; USEPA, 2004). In the United States, pneumatic devices
used in production field operations, flares, chemical injection pumps, and offshore platforms have the
highest emissions factors of any type of equipment or operation in the petroleum system.
Projected Emissions Factors
Projected emissions factors from oil are expected to follow historical trends. IPCC and the USEPA
predict only slight changes in their estimated emissions factors for the next 20 years.1 Although new
technology for equipment and operating procedures may improve in the future, current emissions factors
for equipment and operations will increase slightly because of equipment age and usage.
1 Emissions estimates do not necessarily reflect the IPCC emissions factors presented in Table 3-2.
11-34 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY • OIL
Table 3-2: Forecasted Oil Production for Selected Countries (MMbbl per Day, Unless Otherwise Noted)
Production
Conventional8
Industrialized Countries
United States
Canada
Mexico
Western Europeb
Japan
Australia and New Zealand
Total industrialized
Transitional Economies
FSU
Russian Federation
Caspian and other0
Eastern Europe"
Total transitional economies
Emertfng Economies
OPEC9
Asia
Middle East
North Africa
West Africa
Sou* America
Non-OPEC
China
Other Asia
Middle East'
Africa
South and Cental America
Total emerging economies
Total Production (Conventional)
Total Production (Unconventionai)9
Total Production
2002
9.3
2.1
3.6
6.9
0.2
0.8
22.9
11.2
9.6
1.6
0.2
11.4
1.4
19.0
3.0
2.0
2.9
3.0
2.4
1.9
2.9
3.8
42.3
76.6
1.5
78.1
2010
9.9
1.8
4.3
6.4
0.1
1.0
23.5
13.6
10.3
3.3
0.3
13.9
1.6
25.8
3.6
2.5
3.5
3.7
2.7
2.3
3.8
4.6
54.1
91.5
2.8
94.3
2015
9.7
1.7
4.6
6.0
0.1
0.9
23
15.3
10.8
4.5
0.4
15.7
1.5
27.9
3.9
2.7
4.0
3.6
2.8
2.5
4.9
5.5
59.3
98.0
4.9
102.9
2020
9.5
1.6
4.7
5.6
0.1
0.9
22.4
16.4
11.1
5.3
0.4
16.8
1.5
32.1
4.4
3.1
4.4
3.6
2.8"
2.6
5.5
6.0
66.0
105.2
5.5
110.7
2025
9.3
1.6
4.9
5.0
0.1
0.9
21.8
17.5
11.3
6.2
0.5
18.0
1.5
36.7
4.6
3.6
5.0
3.5
2.7
2.8
6.5
6.5
73.4
113.2
5.7
118.9
Source1 USEIA, 2005b. Adapted from the International Energy Outlook 2004. Table E4. World Oil Production by Region and Country,
Reference Case, 1990-2025.
FSU = Former Soviet Union.
Note: Totals may not equal sum of components because of independent rounding. Data for 2002 and 2003 are model results and may differ
slightly from official USEIA data reports.
a Includes production of crude oil (including lease condensates), natural gas, plant liquids, other hydrogen and hydrocarbons for refinery
feedstocks, alcohol and other sources, and refinery gains.
b Includes Austria, Belgium, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Luxembourg, Macedonia, the Netherlands,
Norway, Portugal, Spain, Sweden, Switzerland, and the United Kingdom.
c Includes Armenia, Azerbaijan, Belarus, Estonia, Georgia, Kazakhstan, Krygyzstan, Latvia, Lithuania, Moldova, Tajikistan, Turkmenistan,
Ukraine, and Uzbekistan.
d Includes Albania, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, Hungary, Macedonia, Poland, Romania, Serbia Montenegro,
Slovakia, and Slovenia.
e OPEC = Organization of Petroleum Exporting Countries. Includes Algeria, Indonesia, Iran, Iraq, Kuwait, Libya, Nigeria, Qatar, Saudi Arabia,
the United Arab Emirates, and Venezuela.
' Non-OPEC Middle East includes Bahrain, Cyprus, Israel, Jordan, Lebanon, Oman, Syria, Turkey, and Yemen.
9 Includes liquids produced from energy crops, natural gas, coal, oil sands, and shale. Includes both OPEC and non-OPEC producers in the
regional breakdown.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-35
-------
SECTION II — ENERGY . OIL
Table 3-3: Forecasted Oil Consumption for Selected Countries (MMbbl per Day, Unless Otherwise Noted)
Consumption
Mature Market Economies
United States
Canada
Mexico
Western Europe
Japan
Australia/New Zealand
Total mature market economies
Transitional Economies
FSU
Eastern Europe
Total transitional economies
Emerging Economies
China
India
South Korea
Other Asia
Middle East
Africa
South and Central America
Total emerging economies
Total Consumption
2002
19.7
2.1
2.0
13.8
5.3
1.0
43.9
4.1
1.4
5.5
5.2
2.2
2.2
5.6
5.7
2.7
5.2
28.7
78.2
2010
22.5
2.3
2.3
14.1
5.3
1.2
47.7
4.7
1.6
6.3
9.2
3.1
2.6
7.9
7.3
3.7
6.8
40.6
94.6
2015
24.2
2.5
2.5
14.3
5.4
1.3
50.1
4.9
1.8
6.7,
10.7
3.7
2.8
9.2
8.0
4.3
7.8
46.3
103.2
2020
25.8
2.5
2.8
14.4
5.4
1.4
52.2
5.2
1.9
7.2
12.3
4.2
2.9
10.4
8.6
4.6
8.5
51.6
111.0
2025
27.3
2<6
3.0
14.9
5.3
1.5
54.6
5.5
2.1
7.6
14.2
4.9
2.9
11.6
9.2
4.9
9.3
57.0
119.2
Source: USEIA, 2005b. Adapted from the International Energy Outlook 2004. Table A4. World Oil Consumption by Region, Reference Case,
1990-2025.
FSU = Former Soviet Union.
Note: Totals may not equal sum of components because of independent rounding. Data for 2002 and 2003 are model results and may differ
slightly from official USEIA data reports.
Table 3-4: IPCC Emissions Factors for Petroleum Systems in Select Regions
Petroleum System Industry Segments (kg/petajoule)
Production
Region
Western Europe
United States and Canada8
FSU, Central and Eastern Europe
Other oil-exporting countries
Rest of the world
Fugitive
Emissions
300-5,000
300-5,000
300-5,000
300-5,000
300-5,000
Venting and
Flaring
1,000-3,000
3,000-14,000*
—
—
—
Transportation
745
745
745
745
745
Storage
90-1,400
90-1,400
90-1,400
90-1,400
90-1,400
Refining
20-250
20-250
20-250
20-250
20-250
Source: IPCC, 1996. Adapted from Table 1-58 in 1996 Revised Guidelines Reference Manual.
FSU = Former Soviet Union.
a In the United States and Canada, venting and flaring emissions are based on total production of both oil and gas produced.
11-36
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY • OIL
11.3.2.3 Emissions Estimates and Related Assumptions
This section discusses the historical and projected baseline emissions from oil production.
Historical Emissions Estimates
Table 3-5 lists CH4 emissions by country from 1990 through 2000. Historically, a country's emissions
in the oil sector have correlated closely with oil production trends. Throughout the last decade, Mexico's
oil emissions have grown to be the largest of any country. By 2000, Mexico had surpassed Romania,
which experienced a sharp decline in baseline emissions over the same time period from 1990 through
2000.
Projected Emissions Estimates
As shown in Table 3-6, worldwide CH4 emissions from oil are expected to increase by more than
80 percent from 2005 to 2020. Countries projected to experience increased production are also projected to
have the largest growth in baseline emissions. Mexico and Brazil are projected to experience the largest
increases at 160 percent and 157 percent, respectively, in their baseline emissions between 2005 and 2020.
11.3.3 The Cost of CH4 Emissions Reductions from Oil
This section discusses opportunities for emissions reductions beyond existing baseline practices.
11.3.3.1 Abatement Option Opportunities
Three abatement options can be applied to the oil sector: flaring, direct use, and reinjection of gas into
oil fields for enhanced oil recovery. Table 3-7 summarizes the costs and emissions reductions associated
with each option.
Flaring in Place of Venting: Offshore and Onshore
The installation of a flaring system results in an estimated 98 percent reduction in fugitive emissions.
Implementation of a flare in an offshore environment is more expensive because of technical,
environmental, and safety concerns. For offshore application, total capital costs are estimated to be
approximately $818 per tCO2eq, and O&M costs are estimated to be approximately $25 per tCO2eq. This
abatement option has a technical lifetime of 15 years, yielding a breakeven price of approximately
$177per tCO2eq. For onshore sites, total capital costs are $34 per tCO2eq, and annual O&M costs are
approximately $1.10 per tCO2eq, yielding a breakeven price of $7 per tCO2eq. Capital costs are assumed
to be constant across countries, but O&M costs vary because of differences in labor costs across countries.
This option has no monetary benefits because the CH4 is combusted and vented as CO2 to the
atmosphere.
Direct Use of CH4
This abatement option applies primarily to offshore platforms and has an estimated reduction
efficiency of 90 percent. .In this abatement option, CH4 is used for consumption on oil platforms and/or
converted to liquefied natural gas. A 15-year lifetime is estimated for this abatement option. Total capital
costs for this abatement option are approximately $55 per tCO2eq. In the United States, O&M costs are
estimated at $1.10 per tCO2eq (O&M cost varies by country). Benefits for this abatement option are the
cost savings from substituting CH4 for alternative energy sources. For the United States, the breakeven
price for direct use of CH4 is $7 per tCO2eq.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 11-37
-------
SECTION II — ENERGY . OIL
Table 3-5: Baseline Emissions from Oil Production, by Country: 1990-2000 (MtC02eq)
Country
Mexico
Romania
China
United States
Nigeria
Iran
Kuwait
United Arab Emirates
Indonesia
Iraq
Ecuador
Canada
Bulgaria
Russian Federation
Lithuania
Rest of the world
World Total
1990
18.8
20.1
1.2
4.4
0.9
1.3
0.4
1.1
1.2
1.1
0.3
0.8
0.6
1.7
0.5
8.2
62.6
1995
19.3
11.4
1.4
4.1
1.0
1.6
0.8
1.2
1.0
0.3
0.3
0.8
0.7
0.9
0.5
8.2
53.5
2000
23.3
8.3
2.2
3.9
1.8
1.2
1.0
1.2
1.9
0.9
0.5
0.9
0.6
0.6
0.4
8.7
57.4
Source: USEPA, 2006.
Table 3-6: Projected Baseline Emissions from Oil Production by Country: 2005-2020 (MtC02eq)
Country
Mexico
Romania
China
United States
Nigeria
Iran
Kuwait
United Arab Emirates
Indonesia
Iraq
Ecuador
Canada
Bulgaria
Russian Federation
Lithuania
Rest of the world
World Total
2005
27.7
9.3
2.9
3-4
2.2
1.8
1.0
1.2
1.9
0.8
0.6
0.9
0.7
0.7
0.6
9.1
64.7
2010
38.7
12.0
4.4
3.7
2.7
2.2
1.3
1.2
1.8
0.9
0.6
0.9
0.8
0.8
0.6
10.1
82.9
2015
54.1
14.7
6.1
4.1
3.3
2.7
1.4
1.4
1.6
1.0
0.8
1.0
0.9
0.9
0.7
114
106.1
2020
71.9
17.3
6.5
4.5
4.1
3.6
1.8
1.7
1.4
1.3
1.1
1.0
1.0
1.0
0.8
12.9
131.8
Source: USEPA, 2006.
II-38
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II —ENERGY. OIL
Table 3-7: Cost of Reducing CH4 Emissions from Oil
Abatement
Technology Year
Flaring: offshore
2010
2020
Flaring: onshore
2010
2020
Direct use (offshore)
2010
2020
Reinjection (onshore)
2010
2020
Capital
Cost
(S/tCO^)
832.60
832.60
33.30
33.30
55.51
55.51
66.61
66.61
Annual
Cost
(SftCOzeq)
24.91
24.91
0.99
0.99
1.11
1.11
2.21
2.21
U.S. Emissions
Available for
Reduction
(MtCOzeq)"
0.52
0.53
0.52
0.53
0.52
0.53
0.52
0.53
U.S. Emissions
Reduction Reductions
Efficiency (MtCO^q)
98% 0.51
98% 0.52
98% 0.51
98% 0.52
90% 0.47
90% 0.48
95% 0.49
95% 0.50
Breakeven
Price
(tfCOzeq)"
$170.35
$170.35
$6.82
$6.82
$7.09
$7.09
$10.14
$10.14
a Based on 50 percent of CH4 emissions generated onshore and 50 percent offshore (USEPA, 2003).
b Based on 15-year lifetime
Reinjection of CH4
Reinjection of CH4 is an alternative to flaring or direct use. In this option, CH4 captured from oil field
operations is reinjected into the oil production field to enhance future oil recovery. Reinjection has an
estimated reduction efficiency of 95 percent and a technical lifetime of 15 years. Total capital costs for this
technology are approximately $67 per tCO2eq. Annual O&M costs are estimated to be $2.20 per tCO2eq in
the United States, but vary by country. Benefits associated with this option include an additional increase
in oil recovery and the mitigation of costs associated with flaring. The estimated breakeven price for the
United States is $10 per tCO2eq.
11.3.4 Results
This section presents the EMF-21 study's MAC results in tabular format and provides a graph of the
MACs for regions with the largest emissions.
11.3.4.1 Data Tables and Graphs
Percentages reported in Table 3-8 are from the report to the EMF provided by the USEPA (USEPA,
2003). It is estimated that there are no "no-regret" options for CH4 abatement in the oil sector.
At a breakeven price of $23 per tCO2eq, the average percentage abatement is 17 percent for the
United States and 38 percent for China, reflecting the high cost of offshore options. Technology changes
have not been incorporated into abatement potential for CH4 from the oil sector.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-39
-------
SECTION II — ENERGY • OIL
Table 3-8: Percentage Abatement for CH4 for Selected Breakeven Price ($/tC02eq): 2000
Technology
Breakeven
Cost
($/tC02eq)
Emissions
Reduction (%
from Baseline)
Emissions
Reduction in 2010
(MtC02eq)
Emissions
Reduction in 2020
(MtC02eq)
Assuming a 10% discount rate and 40% tax rate
Flaring instead of venting (offshore)
Flaring instead of venting (onshore)
Direct use
Reinjection
$575.81
$23.03
$22.22
$31.22
6%
3%
13%
8%
0.02
0.01
0.05
0.03
0.03
0.01
0.06
0.04
Source: USEPA, 2003. Adapted from Oil Sector technology tables in Appendix B to EMF report.
The EMF regional baselines and MAC results of the EMF-21 study are presented in Tables 3-9 and
3-10 for 2010 and 2020 using the base energy price, a 10 percent discount rate, and a 40 percent tax rate.
These MACs represent static percentage reductions in baselines for individual regions/countries and
represent the official MACs used in climate change modeling. Figure 3-2 provides MACs for the five EMF
countries/regions with the largest estimated emissions from the oil sector in 2020.
Table 3-9: Baseline Emissions by EMF Regional Grouping: 2000-2020 (MtC02eq)
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
3.4
17.6
0.1
0.3
2.2
10.0
1.0
0.2
0.0
23.3
10.9
30.1
0.6
2.5
3.9
57.4
2010
4.7
21.9
0.2
0.3
4.4
14.1
1.0
0.3
0.0
38.7
15.3
45.5
0.8
2.5
3.7
82.9
2020
7.4
28:8
0.3
0.5
6.5
19.7
1.1
0.4
0.0
71.9
21.1
79.8
1.0
2:3
4.5
131.8
Source: USEPA, 2006.
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Note: World Total does not equal the sum of the countries listed in this table because the regional groupings are a subset of the full EMF
regional grouping list. See Appendix A of this report for the full EMF list of countries by region.
11-40 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION II — ENERGY . OIL
Table 3-10: Oil System MACs for Countries Included in the Analysis
Percentage Reduction from Baseline (tCOjeq)
Country/Region
Africa
Annex!
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex i
OECD
Russian Federation
South &SE Asia
United States
World Total
$0
0.00%
0.06%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.12%
0.00%
0.00%
0.00%
0.00%
0.00%
0,00%
0.00%
$15
37.27%
21.48%
22.07%
26.69%
38.t7%
13.12%
11.71%
17.54%
0.22%
34,64%
31.07%
24.55%
33.98%
24.07%
17.67%
28.08%
2010
$30
37.27%
21.48%
22.07%
26.69%
38.17%
13.12%
11.71%
17.54%
0.22%
34.64%
31.67%
24.55%
33.98%
24.07%
17.67%
28.08%
$45
37.27%
21.48%
22.07%
26.69%
38.17%
13.12%
11.71%
17.54%
0.22%
34.64%
31.67%
24.55%
33.98%
24.07%
17.67%
28.08%
$60
46.05%
26.54%
27.26%
32.97%
47.15%
16.20%
14.47%
21.66%
0.27%
42.79%
39.12%
30.33%
41.97%
29.73%
21.83%
34.69%
2020
$0
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.12%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
$15
37.27%
20.16%
22.07%
26.69%
38.17%
13.12%
11.71%
17.54%
0.22%
34.64%
30.81%
22.75%
33.98%
24.07%
17.67%
28.96%
$30
37.27%
20.16%
22.07%
26.69%
38.17%
13.12%
11.71%
17.54%
0.22%
34.64%
30.81%
22.75%
33.98%
24.07%
17.67%
28.96%
$45
37.27%
20.16%
22.07%
26.69%
38.17%
13.12%
11.71%
17.54%
022%
34.64%
30.81%
22.75%
33.98%
24.07%
17.67%
28.96%
S60
46.05%
24.91%
27.26%
32.97%
47.15%
16.20%
14.47%
21.66%
0.27%
42.79%
38.06%
28.11%
41.97%
29.73%
21.83%
35.78%
Source: USEPA, 2003.
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Figure 3-2: EMF MACs for Top Five Emitting Countries/Regions from Oil: 2020
$50-,
$40 -
$30-
0)
O $20 \
$10-
$0
-$10 J
United States
Mexico
"• Eastern Europe
•- China
> Africa
10
15
20
25
30
35
Absolute Reduction (MtCO2eq)
Source: USEPA, 2003.
Note: Regional MACs were constructed using percentage reductions from USEPA (2003), with baselines from USEPA (2005).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
11-41
-------
SECTION II — ENERGY . OIL
11.3.5 Uncertainties and Limitations
Uncertainties and limitations persist despite attempts to incorporate all publicly available
international oil sector information. Limited information on the oil systems of developing countries
increases this uncertainty. Additional information would improve the accuracy of baseline emissions
projections:
• Improved Cost Data. Improved documentation of oil CH4 abatement options and their cost
components would make it easier to estimate baseline reductions, given some estimate of market
penetration.
• Improved Emissions Factor Data. Improved documentation of emissions factors for oil systems
of countries outside the United States would enhance the accuracy of international analysis of
CH4 emissions.
• Improved Abatement Option Data. Improved abatement option data are needed to identify true
abatement opportunities for oil systems. For example, although flares have long been thought of
as a potential abatement option, new research suggests that some amount of CH4 may be
escaping combustion at the site of the flare. Accurate information on emissions factors is
necessary before reduction efficiencies can be estimated.
11.3.6 Summary
The data discussed in this chapter demonstrate that oil is a significant source of greenhouse gas
emissions, but because of information limitations for some countries, a more thorough cost analysis is not
possible. Self-regulation by industry and changes in market structure may lead to reductions in emissions
baselines in the future. However, to truly understand the potential benefits of an abatement option in an
oil system and to estimate potential market penetration across countries, more information is needed.
11.3.7 References
Intergovernmental Panel on Climate Change (IPCC). 1996. Revised 1996 IPCC Guidelines for National
Greenhouse Gas Inventories: Reference Manual (Volume 3). Available at . As obtained on April 26, 2004.
U.S. Department of Energy (USDOE), U.S. Energy Information Administration (USEIA). 2005a.
International Energy Annual 2003. Washington, DC: USEIA. Available at . As obtained on May 3, 2006.
U.S. Department of Energy (USDOE), U.S. Energy Information Administration (USEIA). 2005b.
International Energy Outlook 2005. Washington, DC: USEIA. Report # DOE/EIA-0484. Available at
. As obtained on March 3, 2006.
U.S. Environmental Protection Agency (USEPA). 2003. International Analysis of Methane and Nitrous Oxide
Abatement Opportunities: Report to Energy Modeling Forum, Working Group 21. Appendices. Washington,
DC: USEPA. Available at . As obtained on
September 27, 2004.
U.S. Environmental Protection Agency (USEPA). 2004. Inventory of U.S. Greenhouse Gas Emissions and Sinks
1990-2002. Washington, DC: USEPA, Office of Solid Waste and Emergency Response. Available at
. As obtained on October 17, 2004.
U.S. Environmental Protection Agency (USEPA). 2006. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2020. Washington, DC: USEPA.
11-42 GLOBAL Mil 1GATION OF NON-C02 GREENHOUSE GASES
-------
Section II: Energy Sector Appendixes
Appendixes for this section are available for download from the USEPA's Web site at
http://www.epa.gov/nonco2/econ-inv/international.html.
-------
III. Waste
-------
SECTION III — WASTE • PREFACE
Section III presents international emissions baselines and marginal abatement curves (MACs) for waste
sources. There are two chapters, one addressing individual sources from the landfill sector and one
addressing sources from the wastewater sector. These sources include emissions of methane (CH4) and
nitrous oxide (N2O). MAC data are presented in both percentage reduction and absolute reduction terms
relative to the baseline emissions. These data can be downloaded in spreadsheet format from the
USEPA's Web site at .
Section III—Waste chapters are organized as follows:
Methane (CH4)
III.l Landfill Sector
Methane (CH4) and Nitrous Oxide (N2O)
111.2 Wastewater Sector
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION III — WASTE • LANDFILL
111.1 Landfill Sector
orldwide methane (CH4) from the landfilling of municipal solid waste (MSW) accounted
for over 730 million metric tons of carbon dioxide (MtCO2eq) equivalent in 2000 and
represented over 12 percent of total global CH4 emissions. The United States, Africa,
Eastern Europe, and China combined account for 42 percent of the world's CH4 emissions from landfills
(see Figure 1-1). Global CH4 emissions from landfills are expected to grow by 9 percent between 2005 and
2020. Most developed countries have regulations that will constrain and potentially reduce future growth
in CH4 emissions from landfills. However, areas of the world such as Eastern Europe and China are
projected to experience steady growth in landfill CH4 emissions because of improved waste management
practices diverting more MSW into managed landfills.
Figure 1-1: CH4 Emissions from Municipal Solid Waste by Country: 2000-2020
H China
• Eastern Europe
• Africa
D United States
• Rest of the world
2000
2020
Source: Environmental Protection Agency (USEPA), 2006.
III.1.1 Introduction
CH4 from landfills is produced in combination with other landfill gases (LFGs) through the natural
process of bacterial decomposition of organic waste under anaerobic conditions. The CH4 along with
other LFGs is generated over a period of several decades (usually beginning 1 to 2 years after the waste is
put in place). CH4 makes up approximately 50 percent of LFG, with the remaining 50 percent being CO2
mixed with small quantities of other gases. If landfill CH4 is not collected, it will escape to the
atmosphere.
The production of landfill CH4 gas depends on several key characteristics, including waste
composition, landfill design, and operating practices, as well as local climate conditions. Two factors that
will accelerate the rate of CH4 generation within a landfill are an increased share of organic waste (paper,
food scraps, brush) in the mix of MSW being landfilled and increased levels of moisture in the waste. In
addition, if the landfill has used a soil cover (daily cover, intermediate cover, or final cover) in its
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION III —WASTE • LANDFILL
operations, a portion of the CH4 will be oxidized as it passes through these soil layers and converted to
CO2. Many landfill management practices are regulated to control for health arid environmental concerns.
The U.S. federal government currently requires all landfills to monitor and control landfill gas
migration and requires larger landfills to collect and combust landfill gas to destroy the non-CH4 organic
compounds. Landfills with a design capacity greater than 2.5 million megagrams (or 2.8 million short
tons) are subject to the New Source Performance Standards and Guidelines (NSPS/EG) of the Clean Air
Act (USEPA, 1999a), referred to in this chapter as the "Landfill Rule." Similar regulations exist in the
European Union (EU-15) and other developed countries to control the CH4 emissions from large landfills.
However, in most developing countries, there are no regulations covering landfill CH4 emissions. Despite
efforts to control large landfill emissions, the landfill sector remains a significant source of CH4 emissions.
Abatement options include the capture of CH4 for flaring or energy production and enhanced waste
management practices to reduce waste disposal at landfills (such as recycling-and-reuse programs). CH4
recovery for energy use is another approach and is the focus of this report's marginal abatement curve
(MAC) analysis. Because of its low cost, flaring is the most commonly adopted abatement option;
however, this report also considers two energy recovery options as viable alternatives to flaring that may
provide greater financial incentive to landfill managers.
The following sections discuss the activity data and emissions factors used to develop baseline
emissions, abatement options and their costs, and CH4 MACs for the landfill sector. The chapter
concludes with a discussion of uncertainties and limitations. As an appendix to this analysis, we discuss
recent efforts to improve on the MAC methodology by incorporating technology change and by building
the MACs from a population of individual landfills.
111.1.2 Baseline Emissions Estimates
This section discusses the characteristics of landfills and how the characteristics affect CH4 emissions.
In this section, we also describe historical and projected trends that influence baseline emissions from
MSW landfills. In general, the quantity of CH4 generated is determined by four main factors:
• population
• quantity of waste disposed of per capita
• composition of waste disposed of
• type of waste disposal site (landfill versus open dump)
It is commonly accepted that waste generation grows approximately proportional to a country's
population. In addition, countries with higher gross domestic product (GDP) per capita typically generate
more waste per capita. The amount of waste generated per capita multiplied by the population
determines the amount of MSW available for disposal.
The composition of waste, which influences CH4 emissions rates, varies across countries. The level of
recycling or reuse of plastics, metals, organics, and other inorganic waste affects both the amount of
waste disposed of and the type of waste available to generate CH4. Generally, formal recycling-and-reuse
programs are incremental improvements employed by countries that already have sanitary landfills in
place. However, open dumps often have high levels of recovery of both organic and inorganic materials
from informal programs involving human activities and animal scavenging.
The type of waste disposal site also significantly influences CH4 generation. There are generally three
types of waste disposal sites—open dumps, controlled or managed dumps, and sanitary landfills. Open
dumps are characterized by open fills with loosely compacted waste layers. Managed dumps are similar
MI-2 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION III —WASTE • LANDFILL
to open dumps but are better organized and may have some level of controls in place. Open and
controlled dumps are not conducive to CH4 generation primarily because of aerobic conditions as well as
other factors such as shallow layers and unconsolidated disposal (i.e., waste disposed in different parts of
the same landfill site on different days). Sanitary landfills are sites designed and operated to accept MSW
and employ waste management practices, such as mechanical waste compacting and the use of liners,
daily cover, and a final cap (Intergovernmental Panel on Climate Change [IPCC], 1996). Developed
countries primarily employ sanitary landfills. In developing countries, there is a mix of open dumps (in
rural and some urban sites), managed dumps (mainly in larger townships), and sanitary landfills (in large
cities).
III.1.2.1 Activity Data
This section discusses the historical and projected activity factors that determine CH4 generation at
solid waste disposal sites and policies set to improve waste management practices.
Historic Activity Data
Industrialized countries traditionally have the highest per capita waste generation rates and have
accounted for the dominant share of global MSW production each year. Industrialized countries have
also been the first to adopt sanitary landfills, employing waste compaction, dirt covers, and final caps.
Sanitary landfills enable more waste to decay in an anaerobic environment, which ultimately leads to an
increase in CH4 production. However, industrialized countries have also led the way in adopting landfill
gas (LFG) regulations and LFG utilization projects.
Developing countries historically have high population growth rates but use open dumps for waste
disposal because of decentralized waste management programs and cost factors. Open dump waste
disposal sites often do not provide the anaerobic conditions necessary to produce large quantities of CH4.
Some developing countries may have managed dumps that could create the anaerobic conditions
required to generate CH4 emissions. When calculating a country's baseline emissions, it is important to
determine whether the country has any managed dumps. Additionally, economic growth in developing
countries may result in an increased migration from rural communities to larger urban settings. Larger
amounts of waste landfilled in the sanitary and managed dumps in these larger urban cities may
potentially increase the amount of CH4 generated.
Projected Activity Data
Globally, projections indicate that the amount of MSW being deposited into sanitary landfills is
expected to grow. Developing countries are expected to move away from open dumps toward more
sanitary landfills. The fraction of waste disposed of in landfills versus open dumps is expected to increase
at the rate of per capita GDP growth.
Industrialized countries are expected to increase the level of LFG regulation and LFG utilization
projects. These countries will also continue to improve or implement composting, recycling, and reuse
programs. For example, in the United States the fraction of waste generated that is landfilled has
decreased from 72 percent of all waste generated in 1989 to 56 percent of all waste generated in 2000
(USEPA, 2003b).
III.1.2.2 Emissions Factors and Related Assumptions
The emissions factors for sanitary landfills are defined as the CH4 generated per ton of waste
accumulated and are primarily determined by, but are not limited to, four factors: the type and age of the
waste buried in the landfill, the quantity and types of organic compounds in the waste, the moisture
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES III-3
-------
SECTION III —WASTE . LANDFILL
content of the waste, and temperature of the waste. Temperature and moisture levels are influenced by
the surrounding climate. CH4 emissions factors are significantly higher for sanitary landfills compared
with open dumps because of the presence of anaerobic conditions.
Historical Emissions Factors
Industrialized countries have only recently begun adopting waste management practices such as
recycle-and-reuse programs for organic materials. Before these programs were instituted, industrialized
MSW had a higher organic material composition, which resulted in higher emissions factors.
Developing countries' emissions factors for landfills have historically been lower than industrialized
countries because of the use of open dumps, which have shallow layers of rapidly decaying organic
matter under aerobic conditions, preventing the accumulation of CH4. In addition, open dumps make it
easy for both animal scavengers and human waste pickers to remove food and paper, effectively reducing
the amount of organic waste that would otherwise decay and ultimately generate CH4. Fires are also
common at open dump sites and can alter the composition of the MSW, reducing its ability to generate
CH4.
Projected Emissions Factors
Industrialized countries' emissions factors for landfills are projected to decrease. As these countries
continue improving their waste management practices, more of the organic waste will be taken out of the
MSW disposed of at landfills, thereby lowering the landfill's CH4 generation potential. One example is
the EU Landfill Directive, which has limited the amount of organic matter that can enter MSW facilities.
Additionally, steady economic growth and small or negative population growth may again lower
emissions factors for landfills in industrialized countries.
Emissions factors for developing countries' landfills will increase as these countries move away from
open dumps toward sanitary landfills. Sanitary landfills typically do not allow for scavengers to reduce
the organic composition of the MSW. This possibility, in combination with the lack of established
recycling programs, could lead to a dramatic increase in the emissions factors for these landfills.
III.1.2.3 Emissions Estimates and Related Assumptions
This section discusses the historical and projected baseline landfill emissions for both industrialized
and developing countries. Figure 1-2 summarizes the components of landfill baseline CH4 emissions,
where incremental landfill management improvements, such as increased recycling programs, are
accounted for through a reduction in the amount of waste accumulating at a landfill. This has a direct
effect on the quantity of CH4 generated at MSW landfills. In countries for which no emissions estimate
was available, the IPCC Tier 1 methodology was used to estimate baselines using IPCC default values.
For more detailed discussion of baseline emissions calculation methodology, see the USEPA's (2006)
Global Emissions Inventory Report.
Historical Emissions Estimates
Table 1-1 lists the historical baselines for the world's leading countries in CFI4 emissions from
landfills. The United States, by far the largest emitter of CH4 from landfills, experienced a decline in
baseline emissions as a result of the Landfill Rule and LFG utilization. Former Soviet countries of Eastern
Europe, such as the Ukraine and Poland, have experienced gradual increases as these newly independent
states begin to develop their waste management programs and a larger fraction of the MSW generated is
disposed of at managed landfills.
IH-4 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION III —WASTE • LANDFILL
Figure 1-2: Components of CH4 Emissions from Landfills
Total landfill CH4 emissions
equal
CH4 generated from MSW landfills
minus
CH4 recovered and flared or used for energy
minus
CH4 oxidized from MSW landfills
plus
Methane emissions from industrial waste sites
Source: USEPA, 1999b.
Table 1-1: CH4 Emissions from Municipal Solid Waste by Country. 1990-2000 (MtC02eq)
Country
United States
China
Mexico
Canada
o ~:~- r-_.i.....f,~..
Hussian rsoeration
Saudi Arabia
India
Brazil
Ukraine
Potent
South Africa
Turkey
Israel
Australia
Dem. Rep. of Congo (Kinshasa)
Rest of the work)
WorM Total
1990
172.2
40.4
26.o
18.5
37.8
12.5
10.7
13.0
14.2
16.1
14.1
8.2
6.6
7.5
5,0
358.7
781.4
1995
162.4
42.6
28.5
20.4 «
37.8
T4.4
12.2
14.5
14.5
15.9
15.2
8.9 '
7.8
8.3
5,9
360.4
781,7
2000
130.7
44.6
31.0
22.9
35.1
16.8
13.9
15.1
t2<1
17.0
18.3
S.7
8.8
S.O
f.4
341.6
730.3
Source: USEPA, 2006.
Historically, in developed countries, baseline CH4 emissions from landfills are decreasing because of
improved recovery technologies and mandated regulation to capture and control LFG (which includes
CH4) produced at the world's CH4-producing landfills. Many countries have instituted regulations that
require large landfills to install CH4 capture-and-flaring systems either for safety or environmental
concerns. For example, the United States enacted the Landfill Rule in 1996; the EU and the United
Kingdom have enacted similar legislation to limit LFG generation or require its collection and control.
The landfill rule requires landfill gas to be collected and combusted either through flaring or use at
landfills that have a design capacity greater than 2.5 million metric tons (Mt) and 2.5 million cubic meters
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 111-5
-------
SECTION III — WASTE . LANDFILL
(m3). This rule and similar rules in other developed countries have reduced the amount of CH4 in the
baseline estimates for each year after 1999.
Developing countries are increasing the fraction of waste disposed of at landfills as the amount of
waste generated increases with per capita GDP. However, as discussed earlier, open dumps have been
the primary method for waste disposal in developing countries, and because of the characteristics of these
landfills, they tend not to produce large amounts of CH4. Open dumps have kept CH4 baseline emissions
from landfills in developing countries low. However, very large open dumps and managed dumps can
be significant sources of CH4 emissions given sufficient conditions, such as depth, the amount of waste in
place, and the rate of waste accumulation annually.
Projected Emissions Estimates
Worldwide CH4 emissions from landfills are expected to decrease in industrialized countries and
increase in developing countries. Industrialized countries' baselines will continue to decline because of
expanding recycling-and-reuse programs, increased LFG regulation, and improved LFG recovery
technologies. Developing countries' baseline landfill emissions are expected to increase because of their
rapidly expanding populations —trending away from open dumps to sanitary landfills to improve health
conditions —and because of a lack of formal recycling programs in the near future. Formal recycling
programs typically follow the adoption of sanitary landfills. Table 1-2 lists the projected baseline
emissions for the world's top emitters over the period from 2005 to 2020 in MtCO2eq.
Table 1-2: Projected Baseline CH4 Emissions from Municipal Solid Waste by Country: 2005-2020 (MtC02eq)
Country
United States
China
Mexico
Canada
Russian Federation
Saudi Arabia
India
Brazil
Ukraine
Poland
South Africa
Turkey
Israel
Australia
Dem. Rep. of Congo (Kinshasa)
Rest of the world
World Total
2005
130.6
46.0
33.3
25.3
34.2
19.4
15.9
16.6
13.4
17.0
16.8
10.4
9.7
8.7
7.4
342.7
747.4
2010
125.4
47.5
35.5
27.7
33.2
22.1
17.1
17.5
14.7
17.0
16.6
11.0
10.6
9.4
8.6
346.7
760.6
2015
124.1
48.8
37.4
30.7
32.2
24.8
18.1
18.3,
16.4
17.0
16.4
11.6
11.3
10.6
9.8
360.5
788.1
2020
123.5
49.7 '
39.2
33.6
31.1
27.5
19.1
19.0
18.0
17.0
16.2
12.1
11.9
11.9
11.2
375.9
816.9
Source: USEPA, 2006.
IM-6 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION III — WASTE • LANDFILL
Developing nations are projected to experience only slight declines in baseline emissions through
government policies such as the Landfill Rule passed in the United States in 1996. As recovery techniques
improve, the number of landfills that can profit from the LFG recovery will increase, which will continue
to drive down the level of baseline emissions in developed as well as developing countries.
III.1.3 Cost of Emissions Reductions from Landfills
CH4 emissions from landfills can be reduced using two approaches:
• capture the CH4 and flare it or use it for energy and
• change waste management practices to reduce waste disposal at landfills by adding composting
and recycling-and-reuse programs.
CH4 recovery for flaring or energy is the most popular approach and is the focus of this report's cost
analysis. However, documented or expected changes in disposal rates due to composting and recycling
are accounted for in the baseline emissions estimates for each country.
III.1.3.1 Abatement Option Opportunities
Collection systems are present in most landfills as a mechanism to prevent migration of the gas to on-
site structures or away from the landfill to adjacent property and to prevent the release of non-CH4
organic compounds to the atmosphere. Following the collection of CH4, the landfill operator must make a
decision to flare, pump the gas to an end user for process heat, or generate electricity. Table 1-3 specifies
the components of the gas collection and flaring system and direct-use system.
Table 1-3: Components of Collection and Flaring and LFG Utilization Abatement Options
System Type of Equipment
Collection and flaring Wells
Wellheads and gathering pipeline system
Knockout, blower, and flare
Utilization (i.e., electricity production and direct use) Skid mounted filter
Compressor
Dehydrator unit
Pipeline
Turbine, engine, or boiler
Source: USEPA, 2003a.
The USEPA's LFG cost model estimates LFG generation, one-time capital costs, annual operation and
maintenance (O&M) fees, and the quantity of gas recoverable for flaring or utilization for individual
landfills. An expected technology lifetime of 15 years is used. This section discusses the one-time capital
and annual costs and the annual cost savings for the two most popular options: collection and flaring and
utilization. For a complete list of the technology options considered by the Economic Modeling Forum
(EMF) 21 study for the landfill sector, see Table 1-4 below.
Collection and Flaring
The presence of CH4 can be a public health concern, as well as a safety hazard at landfills if the
concentration builds up. For this reason, large landfills have historically removed the CH4 and then
combusted the gas through flaring. Gas is collected through vertical wells and a series of horizontal
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 111-7
-------
SECTION III — WASTE • LANDFILL
collectors typically installed following the closing of a landfill cell. Vertical wells are the most common
type of well, while horizontal collectors are used primarily for deeper landfills and landfill cells that are
actively being filled. Once captured, the gas is then channeled through a series of gathering lines to a
main collection header. The USEPA recommends that the collection system be designed so that an
operator can monitor and adjust the gas flow.
• Capital Costs. This abatement option requires the installation of vertical or horizontal wells;
wellheads and gathering pipeline system; and a knockout, blower, and flare system. The
USEPA's cost model estimates one well for every acre of landfill at a cost of $7,200 per well. The
gathering pipeline system's cost is determined by the number of wells at the landfill. The USEPA
estimates the cost for the collection system as a fixed cost of $19,000, plus a cost of $8,756 per well.
Finally, the cost of the knockout, blower, and flan? system is determined by the gas flow rate. For
example, if a landfill produced 1,000,000 cubic feet per day, the USEPA estimates the cost to be
approximately $200,000.
• Annual Costs. Annual costs include labor costs associated with monitoring the gas flows,
moving or maintaining gas collection systems, and maintaining the flare. Additionally, there is an
annual cost associated with the electricity used by blowers. Annual costs are typically 10 percent
of one-time capital costs.
• Cost Savings/Benefits. Increased environmental and public health benefits, as well as increases
in safety at the landfill site, are the primary benefits. The flaring system is an effective way of
reducing large quantities of CH4 emissions from landfills. Additional nonmarket benefits include
the reduction of volatile organic compounds (VOCs) and reduced odor.
LFG Utilization Systems
Components of a capture and utilization abatement option for the landfill sector include a landfill gas
collection system, utilization pumping system, or some mechanism such as a turbine for generating
energy through the combustion of landfill CH4 gas. 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. From this
point, the gas simply can be flared or used to generate electricity, or the gas can be pumped to an end-
user for process heat. Additional direct-use options, such as fuel to run leachate evaporators and liquid
natural gas production, also reduce CH4 emissions.
In addition, landfill CH4 gas can be transported and used in industrial processes, such as boilers,
drying operations, kiln operations, and cement and asphalt production. Gas collected from the landfill
can be piped directly to local industries where it is used as a replacement or supplementary fuel. The
ideal customers will have a steady, annual energy demand that will use a large percentage or all of the
landfill's gas flow.
• Capital Costs. Utilization systems may require the installation of a skid-mounted filter,
compressor, and dehydrator unit and mile(s) of pipeline to carry gas to the customer. Costs for
the skid-mounted filter, compressor, and dehydrator unit are based on the gas flow rate. For a
landfill with a gas flow rate of 1 million cubic feet per day, the USEPA estimates the installed
costs of the filter, compressor, and dehydrator to be approximately $180,000. The USEPA
estimates the installation cost for the pipeline is $264,000 per mile.
• Annual Costs. Annual costs are composed primarily of electricity usage by the compressor and
dehydrator unit. Estimated annual costs for O&M and electricity are $100,000.
HI-8 GLOBAL MIT IGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION III — WASTE • LANDFILL
Annual Savings/Benefits. Annual benefits are determined by the quantity of gas sold, the British
thermal unit (Btu) content of the landfill gas, and the current market price of natural gas. Given
the 2004 price of natural gas in the United States, annual benefits can be up to 10 times as great as
annual costs.
III.1.4 Results
This section presents the EMF-21 study's MAC results in tabular format.
III.1.4.1 Data Tables and Graphs
Table 1-4 presents the average breakeven price and the reduction in absolute and percentage terms
for the mitigation options discussed in Section IIL1.3.1. Table 1-5 presents the baseline emissions for
landfills by EMF regional grouping. Table 1-6 presents the percentage reduction in the baseline emissions
at specific breakeven prices, and Figure 1-3 provides MACs for the five EMF countries/regions with the
largest estimated emissions from MSW landfills in 2020.
Table 1-4: Breakeven Prices of MSW Landfill Technology Options
Technology
Breakeven
Cost
Emissions
Reduction (%
from baseline)
Emissions &
Reduction in 2010 Redu
(MtCOaeq) (1
liiiiitittfumMi
niocions
ctionin2020
Assuming a 10% discount rate and a 40% tax rate
Anaerobic digestion 1 (AD1)a
Anaerobic digestion 2 (AD2)b
Composting (C1)c
Composting (C2)d
Mechanical biological treatment
Heat production
Increased oxidation
U.S. direct gas use (profitable at
base price)
Electricity generation
Direct gas use (profitable above
base price)
Flaring
$36.03
$428.74
$243.45
$265.41
$362.94
-$16.70
$265.20
$0.90
$73.02
$8.09
$24.69
10%
10%
13%
12%
10%
9%
6%
10%
10%
10%
10%
0.16
0.16
0.45
0.43
Q.W
0,31
0.21
0.34
0.34
0.34
0.34
0.18
0.17
0.52
0,48
0.1€
4J&- ;;
$&•• '
om
o.st
0.39
0.3§
Source: USEPA, 2003c. Adapted from landfill technology tables in Appendix B.
a AD1 expedites the natural decomposition of organic material without oxygen by using a vessel that excludes oxygen and maintains the
temperature, moisture content, and pH close to their optimum values. Cfy can be used to produce heat and/or electricity.
b AD2 expedites the natural decomposition of organic material without oxygen by using a vessel that excludes oxygen in the same way as
AD1, but with additional income from compost.
0 C1 involves degradation of the organic matter under aerobic conditions. It requires separating organic matter from the waste stream.
Finished compost has a market value because it is used to enhance soil in horticulture/landscape and agricultural sites.
d C2 involves the degradation of organic matter under aerobic conditions and the separation of organic matter from the waste stream in the
same way as C1, but there are larger costs.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
111-9
-------
SECTION III —WASTE • LANDFILL
Table 1-5: Baseline Emissions by EMF Regional Grouping: 2000-2020 (MtC02eq)
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South &SE Asia
United States
World Total
2000
84.2
349.6
9.4
15.6
44.6
47.2
84.6
13.9
* 3.9
31.0
62.2
328.6
35.1
23.6
130.7
730.3
2010
101.1
315.7
11.0
17.5
47.5
49.7
46.3
17.1
3.1
35.5
65.1
297.0
33.2
27.9
125.4
760.6
2020
118.8
312,4
13.6
19,0
. 49.7
51.9
32.7
19.1
2.4
39.2
69.1
293.5
31.1
31.5
123.5
816.9
Source: USEPA, 2006.
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Table 1-6: MSW Landfill MACs for Countries Included in the Analysis
Percentage Reduction f
2010
Country/Region
Africa
Annex!
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
Ma
Japan
Mexico
Non-OECD Annex I
OECD
Russian Federation
South SSE Asia
United Steles
World Total
$0
20.71%
11.1«}t
7.00%
20.71%
10.00%
20.71%
7.00%
10.00%
31.50%
10.00%
10.00%
11.42%
10.00%
10.00%
10.00%
11.71%
$15
42.14%
38.89%
29.50%
42.14%
42.14%
42.14%
29.50%
52.86%
66.00%
42.14%
42.14%
38.42%
42.14%
42.14%
42.14%
40.54%
$30
52.86%
45.45%
46.50%
52.86%
52.86%
52.86%
46.50%
52.86%
66.00%
52.86%
5246%
44.53%
52.86%
52.86%
42.14%
48.95%
$45
52.86%
63.58%
46.50%
52.86%
52.86%
52.86%
46.50%
52.86%
66.00%
52.86%
52.86%
64.55%
52.86%
52.86%
80.71%
58.35%
$60
87.31%
88.25%
90.12%
87.31%
87.31%
87.31%
90.12%
87.31%
90.12%
87.31%
87.31%
88.37%
87.31%
87.31%
87.31%
87.81%
rom Baseline in tCC^eq
2020
$0
20.71%
11.54%
7.00%
20.71%
10.00%
20.71%
7.00%
10.00%
31.50%
10.00%
9.20%
11,91%
10.00%
10.00%
10.00%
11.82%
$15
42.14%
40,1f%
29.50%
42.14%
42.14%
42.14%
29.50%
52.86%
66.00%
42.14%
38.76%
40.05%
42.14%
42.14%
42.14%
40.68%
$30
52.88%
46.96%
46.50%
52.86%
52.86%
52.86^
46.50%
52,86%
66.00%
52,86%
48.61%
46.43%
52,86%
52,86%
42.14%
49.62%
$45
82.86%
tsm
46.50%
52.86%
52.86%
52.86%
4&SO%
52.86%
66.00%
52.86%
48.61%
67,31%
52.16%
52.86%
80.71%
56.84%
$80
87,31%
91,1S%
90*12%
87.31%
87.31%
87,31%
90.12%
87.31%
90.12%
87.31%
80.30%
92.14%
87.11%
«?.S1%
«7.31%
87.76%
Source: USEPA, 2003c.
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
111-10
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION III —WASTE • LANDFILL
Figure 1-3: EMF MACs for Top Five Emitting Countries/Regions from Landfills: 2020
$50 n
$40 -
$30 -
cr
Q)
0 $20 -
«
vt
$10 -
-------
SECTION III — WASTE • LANDFILL
* Adjusting Costs for Specific Domestic Situations. Currently, the technologies considered in this
report are available in the United States, Canada, and Western Europe for the costs reported.
However, countries other than these countries may be faced with higher costs because of
transportation and tariffs associated with purchasing the technology from abroad or could be
faced with lower costs due to domestic production of these technologies. Data on domestically
produced technologies, both costs and reduction efficiencies, are not available.
• Country-Specific Tax and Discount Rates. A single tax rate is applied to landfills and landfills in
all countries to calculate the annual benefits of each technology. Tax rates can vary across
countries and in the case of state-run mines and landfills in China, taxes may be less applicable.
Similarly the discount rate may vary by country Improving the level of country-specific detail
will help analysts more accurately calculate benefits and hence breakeven prices.
lit.1.5 Summary and Analysis
The methodology and data discussed in this section describe the MAC analysis conducted for the
landfill sector by the EMF-21 study. MACs for 2010 and 2020 were estimated based on aggregated
industry data from each country or region. The MACs represent estimates of potential CH4 mitigation
from landfills based on available information regarding MSW practices, infrastructure, climate, and
country reported emissions estimates provided through Ihe United Nation's Framework Convention on
Climate Change (FCCC) emissions inventory reports.
ill.1.6 References
Intergovernmental Panel on Climate Change (IPCC). 1996. Revised 1996 IPCC Guidelines for National
Greenhouse Gas Inventories: Reference Manual (Volume 3). Available at . As obtained on April 26, 2004.
U.S. Environmental Protection Agency (USEPA). 1999a. Final Plan for Municipal Solid Waste Landfills.
Available at . Obtained on May 24, 2004.
U.S. Environmental Protection Agency (USEPA). 1999b. U.S. Methane Emissions 1990-2020: Inventories,
Projections, and Opportunities for Reductions. Washington, DC: USEPA. Available at .
U.S. Environmental Protection Agency (USEPA). 2003a. Landfill Gas Energy Cost Model Version 1.2.
Washington, DC: USEPA, Landfill Methane Outreach Program.
U.S. Environmental Protection Agency (USEPA). 2003b. Municipal Solid Waste in the United States: 2001
Facts and Figures. EPA530-R-03-011. Washington, DC: USEPA, Office of Solid Waste and Emergency
Response.
U.S. Environmental Protection Agency (USEPA). 2003c. International Analysis of Methane and Nitrous Oxide
Abatement Opportunities: Report to Energy Modeling Forum, Working Group 21. Appendices. Washington,
DC: USEPA. Available at . As obtained on
September 27, 2004.
U.S. Environmental Protection Agency (USEPA). 2006. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2020. Washington, DC: USEPA.
MM 2 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION III — WASTE • WASTEWATER
1.2 Wastewater Sector
orldwide CH4 from wastewater accounted for more than 523 MtCO2eq in 2000.
Wastewater is the fifth largest source of anthropogenic CH4 emissions, contributing
approximately 9 percent of total global CH4 emissions in 2000. India, China, the United
States, and Indonesia combined account for 49 percent of the world's CH4 emissions from wastewater
(see Figure 2-1). Global CH4 emissions from wastewater are expected to grow by approximately 20
percent between 2005 and 2020.
Figure 2-1: CH4 Emissions from Wastewater by Country: 2000-2020
700 -i
• Indonesia
• United States
• China
D India
• Rest of the world
Source: USEPA, 2006.
Wastewater is also a significant source of nitrous oxide (N2O). Worldwide, N2O emissions from
wastewater accounted for approximately 91 MtCO2eq in 2000 (see explanatory note 1). Wastewater as a
source is the sixth largest contributor to N2O emissions, accounting for approximately 3 percent of N2O
emissions from all sources. Indonesia, the United States, India, and China accounted for approximately 50
percent of total N2O emissions from domestic wastewater in 2000 (see Figure 2-2). Global N2O emissions
from wastewater are expected to grow by approximately 13 percent between 2005 and 2020. This chapter
only discusses the mitigation options that may be available to control CH4 at wastewater treatment
plants. No formal MAC analysis is presented for this sector because data are insufficient on wastewater
systems' infrastructure and abatement technology costs.
III.2.1 Introduction
Wastewater from domestic (sewage) and industrial sources is typically moved through a wastewater
sewer system to a centralized wastewater management treatment center. At the treatment center, soluble
organic material, suspended solids, pathogenic organisms, and chemical contaminants are removed from
water using biological processes in which microorganisms consume the organic waste. This results in the
production of biomass sludge. The microorganisms can perform this biodegradation process in aerobic
and anaerobic environments, the former producing CO2 and the latter producing CH4.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
111-13
-------
SECTION III — WASTE • WASTEWATER
Figure 2-2: N20 Emissions from Wastewater by Country: 2000-2020
H Pakistan
• Brazil
• United States
D China
• Rest of the world
Source: USEPA, 2006.
Wastewater treatment plants (WWTP) may be located on-site or off-site. In the case of domestic
wastewater, septic tanks are an example of an on-site treatment plant for domestic wastewater, while a
centralized municipal WWTP is an example of an off-site facility. The USEPA estimates that 25 percent of
domestic wastewater is treated through on-site facilities such as septic tanks (USEPA, 2004). Centralized
WWTP requires that the wastewater be transported to the facility through a municipal sewer system.
III.2.1.1 Emissions from Wastewater Systems
CH4 is produced by decay of organic material in wastewater as it decomposes in anaerobic
environments. CH4 emissions from wastewater are determined by the amount of organic material
produced and the extent to which this material is allowed to decompose under anaerobic conditions. The
organic content of wastewater is typically expressed in terms of either biochemical oxygen demand
(BOD) or chemical oxygen demand (COD) (IPCC, 1996a). Most developed countries use centralized
aerobic wastewater treatment facilities with closed anaerobic sludge digester systems to process
municipal and industrial wastewater. Employment of these practices increases CH4 generation but
ultimately reduces baseline emissions.
N2O is produced during both the nitrification and denitrification of urea, ammonia, and proteins.
These waste materials are converted to nitrate (NO3) 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, 2001).
An overview of treatment methods, wastewater composition, and sources of CH4 emissions for
domestic and industrial wastewater systems is provided below, followed by a discussion of N2O
emissions.
111-14
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION III — WASTE • WASTEWATER
Domestic Wastewater
The process of treating domestic wastewater (sewage) involves three major phases. First, the
wastewater collected at a centralized WWTP goes through a primary treatment phase. During this phase,
large solids are removed through a filtration process where grit is removed and oxygen is added. Next,
the wastewater enters a primary clarifier that removes almost 95 percent of settleable solids. This process
takes approximately 30 minutes to an hour, and the initial biodegradation by microorganisms begins.
Primary sludge is separated from the effluent at this stage. During this process, wastewater is generally
aerated ensuring that the decomposition of the organic matter occurs in an aerobic environment.
Following the primary treatment, it is common to subject the remaining effluent to a secondary
treatment. During this phase, the effluent undergoes bio-oxidation through an aerobic process in which
aerobic microorganisms break down any remaining organic solids. In the secondary treatment, the
effluent is passed through a trickling filter or aeration basin for approximately 4 to 6 hours. Next, the
remaining effluent moves into a final clarifier where further biodegradation can occur. This secondary
treatment produces additional secondary sludge (biomass). Following the secondary treatment, the
effluent is released to a receiving stream.
The sludge (biomass) produced during the primary and secondary phases of treatment is then
combined and moved into an encapsulated silo-like digester where it undergoes an anaerobic
decomposition process using microorganisms that continue to break down the organics. The digester
comprises a holding tank, a gas capture system, and a heating element. Over a period of time (weeks),
microorganisms break down the large organic molecules in the feed sludge. Still smaller organisms
convert this organic material into CH4 and CC^- On average, 40 to 45 percent of feed sludge is converted
to CH4 and CO2 during the process. The CH4 produced is closely monitored for safety concerns and then
combusted either in the form of a flare or used to generate heat required during this process. The
remaining sludge is sent to landfills.
Industrial Wastewater
Industries producing large volumes of wastewater and industries with high organic COD wastewater
load are likely to have significant CH4 emissions. In the United States, the meat and poultry, pulp and
paper, and produce (i.e., fruits and vegetable) industries are the largest sources of industrial wastewater
and contain high organic COD. These industries are also considered CH4-emitting industries because
they employ either shallow lagoons or settling ponds in their treatment of wastewater, which promotes
anaerobic degradation.
The meat and poultry industry in the United States has been identified as a major source of CH4
emissions because of its extensive use of anaerobic lagoons in sequence to screening, fat traps, and
dissolved air flotation. It is estimated that 77 percent of all wastewater from the meat and poultry
industry degrades anaerobically (USEPA, 1997a).
Treatment of industrial wastewater from the pulp and paper industry is similar to the treatment of
municipal wastewater. Treatment in this industry generally includes neutralization, screening,
sedimentation, and flotation/hydrocycloning to remove solids. Anaerobic conditions are most likely to
occur during lagooning for storage, settling, and biological treatment (secondary treatment). During the
primary treatment phase, lagoons are aerated to reduce anaerobic activity. However, the size of these
lagoons makes it possible for zones of anaerobic degradation to take place. Approximately half of the
initial COD remains following the primary treatment. This remaining COD is passed into a secondary
treatment phase where anaerobic degradation is more likely to take place. The USEPA estimates that 25
percent of COD in secondary treatment lagoons degrades anaerobically (USEPA, 1997b).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 111-15
-------
SECTION III — WASTE • WASTEWATER
The fruit, vegetable, and juice-processing industries generate large amounts of wastewater. The
treatment of wastewater from these industries generally includes screening, coagulation/settling, and
biological treatment (lagooning), while effluent is typically discharged into municipal sewer system.
Anaerobic degradation can occur within the lagoons during biological treatment. In the United States it is
assumed that these lagoons are intended for aerobic operation, but during peak seasonal usage, anaerobic
conditions may occur. The USEPA estimates that approximately 5 percent of wastewater organics
degrade anaerobically (Sheehle and Doom, 2001).
N2O from Wastewater
The two most significant sources of N2O identified in the United States are emissions from
wastewater treatment processes and emissions from effluent discharge into aquatic environments. IPCC
assumes that nitrogen disposal associated with land disposal, subsurface disposal, and domestic
wastewater treatment are negligible as sources of N2O emissions. Generally countries use the IPCC
methodology (IPCC, 2000) for estimating national emissions from wastewater. However, current
methodologies do not allow for a complete estimate of N2O emissions. As a result, N2O baselines
reported in this chapter represent the human sewage component only; no methodology exists to estimate
N2O emissions from industrial wastewater.
The remainder of this chapter discusses the activity data and emissions factors used to develop
baseline emissions and CH4 MACs for wastewater systems. The chapter concludes with a discussion of
uncertainties and limitations.
111.2.2 Baseline Emissions Estimates
CH4 generation occurs as organic matter undergoes decomposition in anaerobic conditions.
However, CH4 generation varies widely depending on waste management techniques. Specifically
engineered environments can increase the CH4 generation rates.
The quantity of CH4 generated can be expressed in terms of several key activity and emissions
factors:
Domestic Wastewater
CH4 Generation = (POP) * (BOD) * (PAD) * (CH4P) (2.1)
where
POP = total population,
BOD = production of BOD per capita per year,
PAD = percentage of BOD anaerobically digested per year, and
CH4P = CH4 generation potential per kg of BOD.1
Industrial Wastewater
CH4 Generation = (IP) * (COD) * (PAD) * (CH4P) (2.2)
1 IPCC emissions factor of 0.6 kilogram CH4 per kilogram of BOD, cited in the USEPA's Inventory of U.S. Greenhouse
Gas Emissions and Sinks: 1990-2002.
HI-16 GLOBAL MIT IGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION III — WASTE • WASTEWATER
where
IP = industry production,
COD = production of COD per unit of output,
PAD = percentage of COD anaerobically digested per year, and
CH4P = CH4 generation potential per kg of COD.2
111.2.2.1 Activity Factors
Activity factors determine the quantity of wastewater produced and the intensity of organic content
(see explanatory note 2). Domestic wastewater production is related to the population size. The
population size, in conjunction with the level of organic waste present in the wastewater (BOD),
determines a country's CH4 generation potential. The per capita production of BOD may vary over time
or by country depending on a population's consumption preferences.
Industrial wastewater generation is based largely on the annual product output from major
wastewater-producing industries, including meat and poultry packing; pulp and paper manufacturing;
and vegetable, fruits, and juices processing. Differences in production processes and recycling practices
can influence the COD per unit of production in these industries.
N2O production is typically estimated using an activity factor of annual per capita protein
consumption (kilograms per year). However, it has been suggested that this factor alone underestimates
the actual amount of protein entering wastewater treatment systems. Food (waste) that is not consumed
is often washed down the drain using garbage disposals. In addition, laundry water can contribute to
nitrogen loadings. For these reasons, multipliers are commonly applied to the annual per capita protein
consumption activity factor to account for these other sources of nitrogen loading.
Historic Activity Data
Wastewater production is directly related to a country's domestic population and industrial
production of select industries. Population growth rates are traditionally higher in developing countries,
while more industrialized countries have recently tended to experience smaller increases in population
over time. Along with population growth, production of BOD per capita has also been growing, which
means that more organic material is present in wastewater. Increases in BOD per capita can result from
various economic improvements, which could lead to a change in the availability of food types and
consumption preferences.
Industrial growth rates and treatment practices differ by country. Whereas most developed and
developing countries have thriving meat and poultry and produce industries, differences exist in the local
regulation and treatment practices. Developing countries are more likely to employ lagoons or settling
ponds in their treatment of industrial waste, which promotes anaerobic degradation.
Projected Activity Data
Both domestic and industrial wastewater production are expected to increase in the future as
populations continue to grow and key industries continue to expand.
2 IPCC emissions factor of 0.25 kilogram CH4 per kilogram of COD, cited in the USEPA's Inventory of U.S. Greenhouse
Gas Emissions and Sinks: 1990-2002.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 111-17
-------
SECTION III — WASTE • WASTEWATER
III.2.2.2 Emissions Factors and Related Assumptions
The primary determinants of wastewater emissions factors are
• CH4 generation potential per unit of BOD or COD and
• the percentage of BOD or COD that degrades in anaerobic conditions.
CH4 generation potential per unit of BOD or COD is likely to remain constant because this is a
measure of chemical potential, not the result of varying preferences. However, wastewater management
practices vary across cities and countries, affecting the percentage of BOD or COD that degrades under
anaerobic conditions. Even for managed systems, differences in operations and maintenance can result in
unintended anaerobic conditions that lead to additional CH4 emissions.
Historical Emissions Factors
A CH4 generation factor of 0.6 kilogram CH4 per kilogram BOD is provided in the IPCC Good
Practice Guidance (IPCC, 2000) for domestic wastewater. This generation factor is also applied to the
pulp and paper and meat and poultry industries. A CH4 generation factor of 0.4 kg CH4 per kilogram
BOD is applied to the fruit, vegetable, and juice-processing industries. This generation factor represents
the potential CH4 generation from a given unit of BOD, assuming that a unit of BOD degrades under
anaerobic conditions.
Most developed countries have adopted municipal wastewater treatment practices that prevent the
formation of anaerobic conditions in managing and treating wastewater. Developing countries have
traditionally employed wastewater management practices that foster controlled anaerobic environments
where the CH4 is captured for flaring or direct use. Settling ponds that are open to the atmosphere are
typically aerated to promote the production or CO2 as opposed to CH4. However, in developing
countries, industries, such as the pulp and paper or meat and poultry, are less likely to have adopted
practices to prevent anaerobic degradation of COD in wastewater.
Projected Emissions Factors
Projected emissions factors from wastewater are expected to follow historic trends. The CH4
generation potential per unit of BOD will remain constant over time. Improvements to wastewater
management practices are projected to occur with increased GDP. These improvements may result in
decreased baseline emissions for developing countries. As developing countries adopt better
management practices, their baseline emissions will approach the baselines of developed countries with
established wastewater infrastructure already in place. Overall, reductions in CH4 emissions factors from
wastewater will occur because of improvements in wastewater management and treatment.
III.2.2.3 Emissions Estimates and Related Assumptions
This section discusses the historical and projected baseline emissions from wastewater. As shown in
Equations (2.1) and (2.2), the amount of CH4 generated each year from wastewater is determined by a
country's population, the per capita production of BOD or COD (in the industry), and the percentage of
BOD that degrades under anaerobic conditions.
Historical Emissions Estimates
Tables 2-1 and 2-2 provide emissions by country for CH4 and N2O. Historically, China and India have
the largest baseline CH4 and N2O emissions from wastewater. China and India are the two most
populous countries in the world with 1.3 and 1.1 billion people, respectively, in 2002 (World Bank, 2005).
Their large populations in highly concentrated urban areas, combined with limited infrastructure for
IH-18 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION III — WASTE • WASTEWATER
Table 2-1: CH4 Emissions from Wastewater by Country: 1990-2000 (MtC02eq)
Country
India
China
United States
Indonesia
Brazil
Pakistan
Bangladesh
Mexico
Nigeria
Philippines
Viet Nam
Iran
Turkey
Russian Federation
Ethiopia
Rest of the work)
World Total
1990
81.8
94.4
24.9
18.0
18.0
10.9
10.4
10.0
6.8
6.2
6.7
6.0
5.7
9.4
3.9
132.8
445.9
1995
89.7
99.7
29.9
19.5
19.3
12.2
11.7
11.0
7.9
7.0
7.4
6.6
6.3
9.4
4.5
141.7
483.8
2000
97.6
104.2
34.3
20.9
20.7
14.0
13.0
11.9
9.0
7.7
8.0
7.2
6.8
9.3
5.1
152.7
522.5
Source: USEPA, 2006.
Table 2-2: N20 Emissions from Wastewater by Country: 1990-2000 (MtC02eq)
Country
China
United States
Brazil
Pakistan
Indonesia
Russian Federation
India
Germany
Nigeria
Iran
Mexico
Bangladesh
Saudi Arabia
Viet Nam
Egypt
Rest of the world
World Total
1990
17.6
13.0
3.7
1.8
2.1
3.7
2.0
2,2
1.0
1.3
1.3
0.9
0.7
1.0
0.9
27.4
80.7
1995
18.5
14.2
3.7
2.0
2.3
3.6
2.2
2.2
1.1
1.4
1.4
1.1
0.8
1.1
1.0
28.4
85.1
2000
19.4
15.6
4.0
2.3
2.5
3.4
2.4
2.2
1.3
1.6
1.6
1.2
0.9
1.2
1.1
30.3
96.8
Source: USEPA, 2006.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 111-19
-------
SECTION III — WASTE • WASTEWATER
handling wastewater, result in substantial emissions. Similar conditions exist in Cambodia and Indonesia
where densely populated areas produce significant CH4 emissions.
Projected Emissions Estimates
Worldwide CH4 emissions from wastewater are expected to increase in both developed and
developing countries because of expanding populations and increases in GDP. Tables 2-3 and 2-4 list
projected baseline emissions by country for CH4 and N2O. India is projected to replace China as the
world's leading emitter of wastewater CH4. The World Bank projects India's average annual growth rate
in population of 1.2 percent over the next 10 years, while China's is projected to be 0.6 percent over the
same time period (World Bank, 2005). Although both countries' GDP is projected to increase over time,
the most influential factor in determining each country's baseline will be the extent to which these
countries improve their wastewater management practices.
III.2.3 Emissions Reductions from Wastewater
Components of abatement options for the wastewater sector include the incremental addition of CH4
mitigation equipment not already included in the initial construction of a municipal wastewater
treatment plant. This section discusses opportunities for emissions reductions beyond existing baseline
practices qualitatively but, because of data limitations, does not attempt to model MACs.
III.2.3.1 Abatement Option Opportunities
We describe two approaches to reducing CH4 emissions from wastewater following the
implementation of municipal infrastructure:
• improved wastewater treatment practices (domestic and industrial) and
• anaerobic digester with collection and flaring or cogeneration.
Improved wastewater treatment practices include reducing the amount of organic waste anaerobically
digested. This reduction can be achieved through improved aeration and/or the scaling back of the use of
stagnant settling lagoons. Costs for improving treatment practices vary widely based on the technology
applied and specific characteristics of the wastewater. Improvements to existing wastewater treatment
practices assume that infrastructure is already in place and that the cost of any improvements would
represent the incremental addition of technology as a capital improvement or increases in O&M costs.
Anaerobic digesters can be flared or the CH4 used for cogeneration to reduce CH4 emissions from
biomass or liquid effluents with high organic content. The IPCC estimates construction costs for
anaerobic digesters to be $0.1 to $3 million (IPCC, 1996b). This estimate includes the construction of a
collection system and either a flare or a utilization system. IPCC estimates annual O&M costs for this type
of system at between $10,000 and $100,000, assuming wastewater flows of 0.1 to 100 million gallons (400
to 0.4 x 106 m3) per day (IPCC, 1996b).
111-20 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION III — WASTE • WASTEWATER
Table 2-3: Projected Baseline CH4 Emissions from Wastewater by Country: 2005-2020 (MtC02eq)
Country
Ms
China
United States
Indonesia
Brazil
PMHW; - • •
Bangladesh
,
Nigeria
MM ' \ •'
lusty . . -_••! ; •;"
.":.v
/' '::'.':".. ,
jj^'^lalt^JL'' j 1^,'JteJE "-% ' ": '
' :WBnt jflff ^BlB ^JfwKJBf' ' ' '
•lSiSip^:':;!^">?-
2005
105,4
108,0
35.2
12.2-
22,0
n.9
14.5
12.8
10.3
84
8.5
7.7
7.3
t,0
5J
•; -. ' m*
••;•• " mi
2010
112.7.
111.7
36.1
23.5
23.2
18.0
15.S
13.0
11.6
9.2
8.0
8,2
7,7
- * *? '
.. '.''I* -.-i
•; • ';::"lWt%; •"
' •• .pini •"••".
tots
mi
115,3
37.0
24,7
24,4
20.2
17.4
14.4
13,1
9.8
».6
a,s
'at
; '41
;• -7.3 r
: -:^JM ' .. ;
^iit' ' ••
2020
ftl.0 •
118J
37.8
25.9
245
21.8
18J
15.1
14.6
10,3
tO.2
9.5
;. 8.5
W
•ft2-
'"••:'«.".. :l!l*'.
'-••-.Bi' '
Source: USEPA, 2006.
Table 2-4: Projected Baseline N20 Emissions from Wastewater by Country: 2005-2020 (MtC02eq)
Country
China
United States
Brazil
Pakistan
Indonesia
Russian Federation
India
Gennany
Nigeria
Iian
Mexico
Bangladesh
Saudi Arabia
v«*# ;, - /• .:.'
fygt ';-''• ' ,
R«ttofftew«« ;
^ftfiilMtiaitt^yf - *
Source: USEPA, 2006.
«y@5 jj^Np. , , wH9' ! '•'••'-• wimp' - '
20,1 20H'"'" ' "'$!&'. r,-:'"",;:^' '
4C T 4£ A ' 4^ '4 - ' - ' * * '" *li2&"^
la./ is.i-, !-_ , ._ ipj . . - . - f^j • . *
4,2 4J' •>'''.: '' }4,7''- '' ' -.; __ "."i*!.1 ^*
2,6 2i!,-v.:'-."- ' • .$*"•.- ;•".:' .• ':; P; ; -
2,6 *fe-v '-:' -. '"?*''••' r ';•;-.:'; i',;"^';' ' •
as &&-V. ; ': . '. %1 ; '^; '•!" " ' ;>: : '• • •»-;:/
OK ^ *? ~ a 9~Q ' ^ ~ : ^ A' t
«•*> <£./.." ', . , 'w-W . • • • ' " - •'•M'! :
25 2* • " 2.2 ;..' . , • j ;, • ; 'pSS1- *-"."•" '••
1.5 1.7" • --..:• /.fit ,;;•'..:. > •''•;.,:';' .:fe^/
1.7 1J| , ' .;; Sj*:'^' ';;l\;v-'';;ffe.:;:'' ".
1.7 1J ?'tJ,:-: ;.%':;V'0 ; ';MK':';. V
t.3 1.4, . . -. :- '-.;4$ f. '.",:: f: f1 '!||f-;'"':
11 " 11 ';"-- ' • '' ''•'•""'•'fcl "' -:- -''."- .'.;*• " l-^jf''^"' •
;• l ",-, ' , *^ ^ , ', ,-,\ • ,^; " ~ tVv^v;-^
1,3 M'?"! ; 'V -yfif :'•-"' -•;^>S;--.^ii--i>'
U t4 ;--,..•; "^::'/::i; 1*;;-\':j|:::\i; i :%P%4,
g|J ., ^i;%:"' •';•'- • i;^tt^^";j;;^.;?^;|^C^:': '
95.0 ' nl".";-1' ' •' •• %: "-.-'^11? ";"•'-" 'rrt-T"v- ^SS^-17: '
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
111-21
-------
SECTION III — WASTE • WASTEWATER
III.2.3.2 Uncertainties and Limitations
Uncertainty and limitations persist despite attempts to incorporate all publicly available information
on international wastewater sectors. Limited information on the wastewater systems of developing
countries increases this uncertainty. Additional information would improve the accuracy of baseline
emissions projections.
• BOD Production Rates: Improved information on specific population diets and consumption
habits would greatly improve the analyst's ability to calculate baseline emissions.
• Country-Specific Waste Management Practices: Improved documentation of wastewater
management practices would allow deviations from the normal assumption, allowing country-
by-country estimates of percentage of BOD undergoing anaerobic degradation.
• Improved Cost Data: Improved documentation of wastewater CH4 abatement options and their
cost components would improve the analyst's ability to estimate baseline reductions given some
estimate of market penetration.
III.2.4 Summary
The data discussed in this chapter demonstrate that wastewater is a significant source of greenhouse
gas emissions. However, policy approaches directly targeted at mitigating CH4 emissions from
wastewater are limited, and no specific abatement options are presented as part of the analysis in this
chapter. Several factors contribute to difficulties in developing MACs for wastewater abatement options.
The primary factor for determining emissions from the wastewater sector (in terms of CH4 emissions
per BOD) is the type of treatment system employed to manage the waste. Centralized, managed
treatment facilities can control anaerobic environments and have a greater potential to capture and use
CH4. Because most centralized systems automatically either flare or capture and use CH4 for safety
reasons, "add-on" abatement options do not exist. As a result, potential emissions reductions depend on
large-scale structural changes in waste management practices. In contrast, smaller decentralized systems
have less control over the share of aerobic versus anaerobic decomposition and have few feasible options
for capturing CH4.
At issue is that overriding economic and social factors influence wastewater treatment practices
throughout the world. The benefits of installing a wastewater system in a developing country for the
purpose of disease reduction greatly outweigh potential benefits associated with CH4 mitigation. This is
not to say that CH4 mitigation is not one of many factors to be potentially considered in selecting
wastewater treatment systems. However, because of the scope of the costs and benefits of the investment
decision, it would be misleading to imply that potential carbon prices (reflected in MACs) would be the
driving force behind investment decisions that influence CH4 emissions from wastewater.
III.2.5 References
Intergovernmental Panel on Climate Change (IPCC). 1996a. Revised 1996 IPCC Guidelines for National
Greenhouse Gas Inventories: Reference Manual (Volume 3). Available at . As obtained on April 26, 2004.
Intergovernmental Panel on Climate Change (IPCC). 1996b. Technologies, Policies, and Measures for
Mitigating Climate Change. Available at < http://www.gcrio.org/ipcc/techrepl/index.html>. As obtained
on February 25, 2004.
111-22 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION III — WASTE • WASTEWATER
Intergovernmental Panel on Climate Change (IPCC). 2000. Good Practice Guidance and Uncertainty
Management in National Greenhouse Gas Inventories. Intergovernmental Panel on Climate Change,
National Greenhouse Gas Inventories Programme, Montreal, IPCC-XVI/DOC. 10 (1.IV.2000).
Available at . As obtained on
January 10, 2005.
Scheehle, E.A., and M.R.J. Doom. 2001. "Improvements to the U.S. Wastewater Methane and Nitrous
Oxide Emissions Estimates." Working paper. Washington, DC.
U.S. Environmental Protection Agency (USEPA). 1997a. "Estimate of Global Greenhouse Gas Emissions
from Industrial Wastewater Treatment." Washington, DC: USEPA. EPA-600/R-97-091.
U.S. Environmental Protection Agency (USEPA). 1997b. Supplemental Technical Development Document for
Effluent Limitations Guidelines and Standards for the Pulp, Paper, and Paperboard Point Source Category.
EPA-821-R-97-001. Washington, DC: USEPA.
U.S. Environmental Protection Agency (USEPA). 2004. Inventory of U.S. Greenhouse Gas Emissions and Sinks
1990-2002. Washington, DC: USEPA, Office of Solid Waste and Emergency Response. Available at
. As obtained on October 17, 2004.
U.S. Environmental Protection Agency (USEPA). 2006. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2020. Washington, DC: USEPA.
World Bank Group. 2005. 2004 — World Development Indicators: Table 2.1 Population Dynamics. Available at
. As obtained on February 24, 2005.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES 111-23
-------
SECTION III — WASTE • WASTEWATER
Explanatory Notes
1. Assuming a global warming potential (GWP) value of 310.
2. The wastewater treatment practices that determine the share of BOD that degrades under anaerobic
conditions are included in the emissions factor discussion.
HI-24 GLOBAL MIT IGATION OF NON-C02 GREENHOUSE GASES
-------
Section III: Waste Sector Appendixes
Appendixes for this section are available for download from the USEPA's Web site at
http://www.epa.gov/nonco2/econ-inv/international.html.
-------
IV. Industrial Processes
-------
SECTION IV — INDUSTRIAL PROCESSES • PREFACE
This section presents international emissions baselines and marginal abatement curves (MACs) for 11
industrial sources. Each chapter in this section addresses one of these sources. These sources include
nitrous oxide (N2O) emitted during nitric and adipic acid production; fluorinated gases that are used as
substitutes for ozone-depleting substances (ODSs); and high-global warming potential (GWP) gases,
including hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), and sulfur hexafluoride (SF6) from
several industrial sources. MAC data are presented in both percentage reduction and absolute reduction
terms relative to the baseline emissions. These data can be downloaded in spreadsheet format from the
U.S. Environmental Protection Agency (USEPA) Web site at .
The Section IV—Industrial Processes chapters are organized as follows:
Nitric Oxide
TV.I N2O Emissions from Nitric and Adipic Acid Production
Fluorinated Gases Used as Substitutes for ODSs
IV.2 HFC Emissions from Refrigeration and Air-Conditioning
IV.3 HFC, HFE, and PFC Emissions from Solvents
IV.4 HFC Emissions from Foams
IV.5 HFC Emissions from Aerosols
IV.6 HFC Emissions from Fire Extinguishing
High-GWP Gases from Industrial Processes
IV.7 PFC Emissions from Aluminum Production
IV.8 HFC-23 Emissions from HCFC-22 Production
IV.9 PFC and SF6 Emissions from Semiconductor Manufacturing
IV.10 SF6 Emissions from Electric Power Systems
IV. 11 SF6 Emissions from Magnesium (Mg) Production
IV-ii GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • PREFACE
IV. Industrial Processes Overview
his section presents international emission baselines and MACs for twelve sources of various
greenhouse gases, including N2O, HFCs, PFCs, and SF6. These sources include production of
nitric and adipic acid, which emit N2O; production of aluminum, magnesium, semiconductors,
and HCFC-22, which emit PFCs, SF6, and HFCs; and use of electrical equipment in electric power
systems, which emits SF6. In addition to the industrial sectors, this section also includes emissions
estimates and MACs for fluorinated gases (generally HFCs) that are used as substitutes for ODSs.
While a single set of baseline emissions estimates is presented for most industrial processes covered
in this section, five subsectors have dual baselines and MACs. These processes are the production of
aluminum, semiconductors, Mg, and HCFC-22, and the use of electrical equipment. For all five of these
industries, clearly defined, industry-specific global or regional emissions reduction goals have been
announced. First, in response to concerns regarding the high GWPs and long lifetimes of their emissions,
the global aluminum, semiconductor, and Mg industries have committed to reduce future emissions by
substantial percentages. Second, users (and, in some cases, manufacturers) of electrical equipment in
Japan, Europe, and the United States have committed to reduce emissions in those countries and regions.
Finally, HCFC-22 producers in several developing countries have agreed to host mitigation projects
funded by developed countries under the Clean Development Mechanism (CDM) of the Kyoto Protocol.
The HFC-23 abatement projects considered in this analysis are either registered or are in the process of
being registered in the CDM pipeline. (HCFC-22 producers in developed countries are also continuing to
reduce emissions.) These global and regional emissions reduction goals are summarized in the table
below.
Table: Global and Regional Emissions Reduction Commitments
Global Industry Association,
Industry Region, or Country
Percentage of World
Production/Emissions in
2003
Goal
Semiconductor World Semiconductor Council
manufacturing
Mg production International Magnesium
Association
Aluminum
production
International Aluminum Institute
Electrical EU-25+3, Japan, and United
equipment (use) States
85%
80% of the magnesium industry is
outside of China; about 80% of
global SF6 emissions
70% (but goal applies to entire
industry)
40% of use emissions
HCFC-22
China, India, Korea, and Mexico 65% of emissions
Reduce fluorinated emissions to
90% of 1995 level by 2010
Phaseout SF6 use by 2011
Reduce PFCs/ton of aluminum
by 80% relative to 1990 levels
by 2010
Country-specific reductions from
2003 totaling 2.5 MtCC^q, or
15% of these countries' 2003
emissions from use
CDM projects totaling 55
MtC02eq,or63%ofthese
countries' 2010 emissions
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-iii
-------
SECTION IV — INDUSTRIAL PROCESSES • PREFACE
The first scenario presented in this report, called the "technology-adoption baseline," is based on the
assumption that these industries will achieve their announced global or regional emissions reduction
goals for the year 2010. The second scenario, called the "no-action baseline," is based on the assumption
that emissions rates will remain constant from the present onward in these industries.
The USEPA believes that actual future emissions are likely to be far closer to those envisioned in the
technology-adoption baseline than those envisioned in the no-action baseline. Since 1990, all five
industries have already made great progress in reducing their emissions rates, and research is continuing
into methods to further reduce those rates. Nevertheless, additional actions will be required to actually
realize additional reductions. These actions range from process optimization and chemical recycling to
chemical replacement. In some cases, the actions are estimated to carry net private costs; in others, net
private benefits.
The MACs for the technology-adoption baseline have been adjusted to reflect the implementation of
some options in the baseline. When an option is assumed to be adopted in the baseline, the emissions
stream to which that option is applied in the MAC is correspondingly decreased, so that options that are
fully implemented in the technology-adoption baseline are not present in the technology-adoption MAC
at all.
Depending on the context, either set of baselines and MACs may be of interest. For example, analysts
interested in the incremental costs of reducing emissions below the levels anticipated in current global
industry commitments can use the technology-adoption baseline and the associated MACs. On the other
hand, analysts interested in the future costs of achieving the currently planned industry reductions can
use the no-action baseline and the associated MACs. The difference between the two baselines is itself of
interest, demonstrating that the industry commitments are likely to avert very large emissions.
It should be noted that the USEPA modeled only those reduction efforts that had been clearly
announced and quantified on an industry-specific basis at the time this report was prepared. This means
that even in the technology-adoption baseline, significant reduction opportunities remain in 2010 and
2020, primarily in developing countries. This is particularly true for the HCFC-22 and electric power
system industries. In fact, there is a significant probability that many of these emissions will be averted
(e.g., by fuller implementation of COM or other reduction efforts). However, the precise extent of
additional reduction actions is uncertain. Thus, the technology-adoption baseline reflects only current,
quantitative, industry-specific goals.
Past emissions (1990 through 2000) for all five sources are identical under either scenario, but they are
provided with both scenarios to provide context for the divergent future trends.
Detailed discussions of the methodology used to develop the baselines for each source can be found
in the USEPA (2006) report Global Anthropogenic Non-CO2 Greenhouse Gas Emissions: 1990-2020.
IV-iv GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • NITRIC AND ADIPIC ACID PRODUCTION
IV.1 N2O Emissions from Nitric and Adipic Acid Production
orldwide N2O emissions from industrial sources account for more than 154 million metric
tons of carbon dioxide (CO2) equivalent (MtCO2eq) (USEPA, 2006). The USEPA estimates
that emissions from nitric and adipic acid production combined contributed
approximately 5 percent of total global N2O emissions in 2000 (USEPA, 2003). Nitric acid production
accounts for 67 percent of N2O emissions from industrial production, and adipic acid accounts for the
remaining 33 percent (USEPA, 2003).
Eastern Europe, the United States, China, and the European Union (EU-15) combined account for 79
percent of total N2O emissions from industrial production (Figure 1-1). The Intergovernmental Panel on
Climate Change (IPCC) reports that the number of nitric acid production plants worldwide is estimated
at 250 to 600. The United States is the primary producer of adipic acid, with four production sites alone,
accounting for approximately 40 percent of total adipic acid production worldwide (USEPA, 2001). Other
countries have at most one adipic acid plant (IPCC, 2000).
Figure 1-1: N20 Emissions from Industrial Production by Country: 2000-2020
• Eastern Europe
• United States
• China
QEU-15
• Rest of the world
2020
Source: USEPA, 2006.
EU-15 = European Union.
Global N2O emissions from industrial production sources are expected to grow by approximately 13
percent between 2005 and 2020 (USEPA, 2006), although the percentage distribution of emissions across
countries is projected to remain relatively unchanged.
IV.1.1 Introduction
The two major sources of anthropogenic N2O emissions from industry are production of nitric and
adipic acid. These dicarboxylic acids produce N2O as a by-product of the production process. N2O is then
emitted in the waste gas stream (USEPA, 2001).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-1
-------
SECTION IV — INDUSTRIAL PROCESSES « NITRIC AND ADIPIC ACID PRODUCTION
IV.1.1.1 Nitric Acid
Nitric acid is an inorganic compound, typically used to make synthetic commercial fertilizer. Nitric
acid is also used in the production of adipic acid, explosives, and metal etching and in the processing of
ferrous metals. Nitric acid is produced through catalytic oxidation of ammonia (CH4) at high
temperatures, which creates N2O as a reactionary by-product released from reactor vents into the
atmosphere (Mainhardt and Kruger, 2000). IPCC believes that nitric acid production now represents the
majority of N2O emissions from industrial process as a result of implementing abatement technologies at
adipic acid plants.
In the United States, the nitric acid industry controls for nitrogen oxides gases using a combination of
nonselective catalytic reduction (NSCR) and selective catalytic reduction (SCR) technologies (USEPA,
2004). The NSCR units destroy nitrogen oxides, but they also destroy N2O. However, NSCR is considered
costly and obsolete at modern plants. NSCR units were commonly installed in production facilities built
between 1971 and 1977 (USEPA, 2004). The USEPA reports that NSCR is currently used by approximately
20 percent of the U.S. nitric acid production plants; the majority of the industry uses SCR or extended
absorption, neither of which is known to reduce N2O (USEPA, 2004).
IV.1.1.2 Adipic Acid
Adipic acid is a white crystalline solid used primarily as a component in the production of nylon
(nylon 6/6). Adipic acid is also used in the manufacture of low-temperature synthetic lubricants, coatings,
plastics, polyurethane resins, and plasticizers and is used to give some imitation foods a "tangy" flavor.
Industrial sources report that by 2000, all major adipic acid production plants had implemented
abatement technologies and consequently have dramatically reduced N2O emissions from this source
(Mainhardt and Kruger, 2000).
Adipic acid is produced through a two-stage process during which N2O is generated in the second
stage. The first stage of manufacturing usually involves the oxidation of cyclohexane to form
cyclohexanone/cyclohexanol mixture. The second stage entails oxidizing this mixture with nitric acid to
produce adipic acid. N2O is produced as a by-product during the nitric acid oxidation stage and
potentially is emitted in the waste gas stream (USEPA, 2004). Emissions from this source vary depending
on the type of technologies and level of emissions controls employed by a specific facility.
IV.1.2 Baseline Emissions Estimates
N2O emissions correlate closely with the production of nitric and adipic acid. This section discusses
production activity, suggested emissions factors, and the resulting baseline emissions estimates based on
publicly available reports.
IV. 1.2.1 Activity Factors
Activity factors characterize the intensity of production in these industries, which, when combined
with emissions factors, result in an estimated baseline emission.
Historical Activity Data
Nitric Acid
Nitric acid production levels closely follow trends in fertilizer consumption, because of nitric acid's
role as a major component in fertilizer production (Mainhardt and Kruger, 2000). Trends in fertilizer
production vary widely across different regions of the world. For example, in Western Europe, because of
concerns over nutrient runoff, nitrogen-based fertilizer use has been scaled back. However, in regions
IV-2 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • NITRIC AND ADIPIC ACID PRODUCTION
where agriculture accounts for a larger share of the gross domestic product (GDP), such as Asia, South
America, and the Middle East, nitrogen-based fertilizer production capacity is increasing (Mainhardt and
Kruger, 2000).
The actual number of nitric acid production plants globally is unknown. Previous reports cited by the
IPCC have suggested the number to be between 250 and 600. This uncertainty is due to the fact that many
nitric acid plants are often part of larger facilities that manufacture products using nitric acid, such as
fertilizer and explosives facilities (Mainhardt and Kruger, 2000).
Adipic Acid
Adipic acid is used primarily in the production of nylon. As a result, production of adipic acid is
closely correlated with the world's nylon production. Global demand for engineering plastics has
increased over time, resulting in major expansion in production capacity in North America and Western
Europe and new facilities in the Asia Pacific region. In the United States, adipic acid production increased
by approximately 50 percent between 1990 and 2000 (USEPA, 2004).
Global capacity for adipic acid was approximately 2.8 million metric tons in 2003. Table 1-1 lists
estimated adipic acid production capacity in 2003 by country. Demand for adipic acid was estimated at
2.21 million metric tons for the same year (Chemical Week [CW], 2003). As a result of this oversupply in the
global market, many adipic acid facilities have been operating at an average rate of 85 percent of capacity.
Table 1-1:2003 Adipic Acid Production Capacity (Thousands of Metric Tons/Year)
Country
United States
Germany
France
United Kingdom
Canada
South Korea
China
Japan
Singapore
Brazil
Italy
Ukraine
World Total
Adipic Acid Capacity
1,002.0
408.0
320.0
220.0
170.0
135.0
127.0
122.0
114.0
80.0
70.0
56.0
2,824.0
Source: CW, 2003.
Projected Activity Data
Nitric Acid
Nitric acid production is expected to increase over time (Mainhardt and Kruger, 2000). The Global
Emissions Report, from which the emissions projections came, used data that did not report specific
country activity. Projected production data for nitric acid production were unavailable at the time of
publication of this report.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-3
-------
SECTION IV — INDUSTRIAL PROCESSES • NITRIC AND ADIPIC ACID PRODUCTION
Adipic Acid
Industrial demand for adipic acid is expected to continue to increase by approximately 2 percent per
year between 2003 and 2008 (CW, 2003). Nylon 6,6 accounts for approximately 70 percent of demand for
adipic acid. The demand for fiber-grade nylon 6,6 is projected to grow by 1 percent per year, whereas
engineering-grade nylon 6,6 is projected to grow by 4.5 percent per year. The dramatic growth in
engineering-grade nylon is a result of its increased use as a substitute for metal in under-the-hood
automotive applications (CW, 2003).
IV.1.2.2 Emissions Factors and Related Assumptions
Nitric Acid
The IPCC reports that N2O emissions factors for nitric acid production 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 type of technology employed at the plant. For example, NSCR is very effective
at destroying N2O, whereas alternative technologies such as SCR and extended absorption do not reduce
N2O emissions.
In the United States, a weighted average of 2 kilograms N2O per ton nitric acid is used for plants
using NSCR systems, and 9.5 kilograms N2O per ton nitric acid is used for plants not equipped with
NSCR. Table 1-2 lists the reported emissions factors by IPCC in the Revised 1996 Reference Manual.
Table 1-2: IPCC Emissions Factors for Nitric Acid Production in Select Countries
Country Nitric Acid Emissions Factors
United States 2.0-9.0"
Norway—modern, integrated plant ' < 2.0
Norway—atmospheric-pressure plant 4.0-5.0
Norway—medium-pressure plant 6.0-7,5
Japan 2.2-5.7
Source: IPCC, 1996.
a Emissions factors up to 19 kilograms per ton nitric acid have been reported for plants not equipped with NSCR technology.
The IPCC points out that potential emissions factors as high as 19.5 kilograms N2O per ton of nitric
acid have been estimated in previous reports. In addition, estimates of 80 percent of the nitric acid plants
worldwide do not employ NSCR technology, which makes it more likely that the default range of
potential emissions factors provided by the IPCC greatly underestimates the true emissions baselines
(Mainhardt and Kruger, 2000).
Adipic Acid
The IPCC provides countries with a default emissions factor of 300 kilograms N2O per ton of adipic
acid produced. This emissions factor assumes that no N2O control system is in place. This factor was
developed using laboratory experiments measuring the 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. IPCC recommends using plant-specific
data for those plants with abatement controls already in place (IPCC, 1996).
IV-4 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • NITRIC AND ADIPIC ACID PRODUCTION
IV.1.2.3 Emissions Estimates and Related Assumptions
This section discusses the historical and projected baseline emissions from the industrial process
sector for the production of nitric and adipic acid.
Historical Emissions Estimates
Table 1-3 lists historical N2O emissions by country. Worldwide N2O baseline emissions from nitric
and adipic acid production decreased by 28 percent between 1990 and 2000. The United Kingdom,
Germany, and Canada experienced the largest declines in baselines emissions, with 88 percent, 84
percent, and 77 percent declines, respectively, over the same 10-year period. However, countries such as
China, Japan, South Korea, and India saw baseline increases of 54, 29, 25, and 29 percent, respectively.
Table 1-3: N20 Emissions from Nitric and Adipic Acid Production: 1990-2000 (MtC02eq)
Country
China
United States
France
South Korea
Italy
Netherlands
Brazil
United Kingdom
Germany
Belgium
Japan
Poland
India
Bulgaria
Romania
Rest of the world
World Total
1990
19.6
33.1
24.1
5.7
6.7
7.6
2.5
29.3
23.5
3.9
7.4
5.0
2.4
2.3
8.9
41.4
223.4
1995
27.5
37.1
26.2
6.1
7.1
7.5
4.3
19.0
25.0
4.6
7.4
4.9
2.8
1.9
3.6
35.0
220.1
2000
30.1
25.6
11.5
7.1
7.8
7.1
5.0
6.3
5.5
4.6
4.2
4.3
3.0
1.3
2.9
27.5
154.0
Source: USEPA, 2006.
Projected Emissions Estimates
Table 1-4 lists combined projected N2O baseline emissions from nitric and adipic acid by country.
Worldwide total N2O emissions from nitric and adipic acid are projected to increase by approximately 16
percent between 2005 and 2020. The United States, South Korea, and Brazil are expected to experience the
largest increase in baseline emissions, with 28, 22, and 22 percent, respectively, between 2005 and 2020.
Nitric Acid
Emissions from nitric acid production are expected to increase by 13 percent between 2000 and 2020,
because of an expanding market for synthetic fertilizer (see explanatory note 1). Brazil, Mexico, and India
are projected to increase their N2O baseline emissions by 29, 25, and 22 percent, respectively, from nitric
acid production (USEPA, 2006).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-5
-------
SECTION IV — INDUSTRIAL PROCESSES • NITRIC AND ADIPIC ACID PRODUCTION
Table 1-4: Projected N20 Baseline Emissions from Nitric and Adipic Acid Production: 2005-2020 (MtC02eq)
Country
China
United States
India
France
Italy
Brazil
Netherlands
South Korea
United Kingdom
Germany
Belgium
Japan
Poland
Bulgaria
Ukraine
Rest of the world
World Total
2005
32.0
22.4
3.2
12.9
8.2
5.5
7.5
7.9
6.3
5.7
4.7
4.6
4.3
2.3
2.4
26.5
156.5
2010
34.1
23.9
3.4
14.3
8.6
6.1
7.7
8.7
6.3
5.9
4.9
4.6
4.3
2.7
2.4
26.7
164.6
2015
35.5
25.5
3.6
14.4
9.1
6.4
8.1
9.1
6.3
6.1
5.1
4.8
4.3
2.9
2.4
26.9
170.4
2020
37.0
27.2
3.8
14.5
9.6
6.7
8.3
9.6
6.3
6.2
5.2
5.0
4.3
3.4
2.4
27.2
176.6
Source: USEPA, 2006.
Adipic Acid
Emissions from adipic acid production are projected to increase by approximately 40 percent between
2000 and 2020, reflecting increased demand for engineering nylon (see explanatory note 1). Southeast
Asia, Brazil, and Mexico are projected to experience 45, 44, and 39 percent increases, respectively, in
baseline emissions of N2O.
IV.1.3 Cost of N2O Emissions Reductions from Industrial Processes
N2O emissions can be reduced by optimizing the catalytic oxidation of CH4 to nitrogen oxide or by
decomposing N2O either during the processing of nitric acid or in the tail gas. Currently, N2O reduction
technologies include extending the reaction process through thermal decomposition in the reaction
chamber, reducing N2O through catalytic reduction in the reaction chamber, using NSCR or SCR in the
upstream tail gas expander, or using SCR in the downstream tail gas expander (Smit, Gent, and van den
Brink, 2001). Each of the technologies has advantages and disadvantages, including the amount of
utilities required to run the technology, downtime at the plant for installation, consumption of the
reducing agent, and retrofit limitations at existing plants. Depending on the technology, reduction
efficiencies can range from 70 percent to 98 percent and costs can range from $0.52 to $9.30 per tCO2eq for
new installations and $0.86 to $9.48 per tCO2eq.
Abatement options for the nitric and adipic acid sectors at the time of the Energy Modeling Forum 21
(EMF-21) analysis were relatively limited. However, more recent innovations have proven effective
options for abating N2O at nitric acid production plants. The data presented in this report use an average
reduction and cost of NSCR and SCR technologies. Therefore, the reduction potential is at the high end of
the reduction range and the costs are on the lower end of the range. Table 1-5 summarizes cost and
emissions reductions for the abatement options included in the EMF-21 analysis (USEPA, 2003).
IV-6 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • NITRIC AND ADIPIC ACID PRODUCTION
Table 1-5: Cost of Reducing N20 Emissions from Industrial Processes
Technology
Emissions
Breakeven Reduction
Price (% from
($/tC02eq) baseline)8
Emissions
Reduction in Emissions
2010 Reduction in
(MtC02eq) 2020 (MtC02eq)
Assuming a 10% discount rate and
Nitric Acid Sector"
Grand Paroisse— high-temperature catalytic
reduction method
BASF— high-temperature catalytic reduction
method
Norsk Hydro— high-temperature catalytic
reduction method
HITK— high-temperature catalytic reduction
method
Krupp uhde— low-temperature catalytic
reduction method
ECN— low-temperature selective catalytic
reduction with propane addition
NSCRC
$2.59
$2.36
$1.99
$2.75
$2.92
$5.81
$4.03
6%
6%
7%
7%
7%
7%
6%
0.05
0.05
0.05
0.06
0.06
0.06
0.05
40% tax rate
0.05
0.05
0.06
0.06
0.06
0.06
0.05
Adipic Acid Sector*
Thermal destruction
$0.50
50%
0.21
0.24
Source: USEPA, 2003. Adapted from Nitric Acid and Adipic Acid Sector technology tables in Appendix B.
a Values represent average percentages across all EMF-21 countries/regions included in the analysis.
b Based on 10-year lifetime.
c Based on 20-year lifetime.
IV.1.3.1 Nitric Acid: N2O Abatement Option Opportunities
High-Temperature Catalytic Reduction Method
This N2O abatement option has several variations developed by different companies, all involving
the decomposition of N2O into nitrogen and oxygen using various catalysts. The average estimated
reduction efficiency is approximately 90 percent. Total capital costs for these abatement technologies
range from $2.18 to $3.27 per tCO2eq. Operating and maintenance (O&M) costs vary by country. In the
United States, O&M costs can range from $0.14 to $0.22 per tCO2eq. This abatement option has an
average technical lifetime of 10 years, yielding a breakeven price of approximately $0.82 per tCO2eq.
Low-Temperature Catalytic Reduction Method
Low-temperature catalytic reduction systems work similarly to high-temperature counterparts but do
not require heat to decompose the N2O. This abatement option has a reduction efficiency of 95 percent.
Some versions of this abatement option require propane be added to the gas stream before undergoing
the reaction process. Total capital cost for this option ranges from $3.27 to $3.55 per tCO2eq. In the United
States, O&M costs range from $0.27 to $1.91 per tCO2eq. This option has a technical lifetime of 10 years,
yielding a breakeven price of approximately $0.82 per tCO2eq.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-7
-------
SECTION IV — INDUSTRIAL PROCESSES . NITRIC AND ADIPIC ACID PRODUCTION
Nonselective Catalytic Reduction
NSCR uses a fuel and a catalyst to consume free oxygen in the tail gas, converting nitrogen oxides to
elemental nitrogen. The gas from the nitrogen oxides abatement is passed through a gas expander for
energy recovery, resulting in a reduction efficiency of 85 percent. The process requires additional fuel and
emits CO2. The total capital cost for this option is $6.27 per tCO2eq. In the United States, the O&M cost is
estimated at $0.16 per tCO2eq. NSCR has a technical lifetime of 20 years, yielding a breakeven price of
approximately $1.90 per tCO2eq.
IV.1.3.2 Adipic Acid: N2O Abatement Option Opportunities
Thermal Destruction
Thermal destruction is the destruction of off-gases in boilers using reducing flame burners with
premixed CH4 (or natural gas). The system eliminates between 98 percent and 99 percent of N2O and
operates from 95 percent to 99 percent of the time. Total capital costs for thermal destruction are $0.38 per
tCO2eq. In the United States, O&M costs are estimated to be approximately $0.16 per tCO2eq. This
abatement option has a technical lifetime of 20 years, yielding a breakeven price of approximately $0.27
per tCO2eq.
IV. 1.4 Results
This section presents the EMF-21's MAC analysis results.
IV.1.4.1 Data Tables and Graphs
The nitric and adipic baselines are presented in Tables 1-6 and 1-8. Tables 1-7 and 1-9 present
percentage reductions for different carbon prices ($/tCO2eq) from the emissions baselines for each sector.
Figures 1-2 and 1-3 present these results in graphical form. Significant abatement potential is estimated to
exist at $15 per tCO2eq. It is estimated that there are no "no-regret" options for N2O nitric or adipic acid
production. At a breakeven price of $15 per tCO2eq, the percentage abatement is 89 percent for nitric acid
and 96 percent for adipic acid, reflecting the relatively high technical potential and low abatement cost for
options in these industrial processes. Technology changes have not been incorporated in the abatement
potential for N2O emissions from industrial processes.
IV.1.4.2 Uncertainties and Limitations
Uncertainties and limitations persist despite attempts to incorporate all publicly available
information on international sectors. Limited information on the systems of developing countries
increases this uncertainty. Additional information would improve the accuracy of baseline emissions
projections.
Improved Cost Data
Improved documentation of N2O abatement options and their cost components would improve the
analyst's ability to estimate baseline reductions given some estimate of market penetration.
Improved Manufacturing Data for Nitric Acid
Currently, worldwide nitric acid production is very uncertain because of a lack of good production
estimates. In addition, improved data on the types of equipment generally employed by industries and
trends in technology adoption in each country would improve the analyst's ability to estimate baseline
emissions over time.
IV-8 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • NITRIC AND ADIPIC ACID PRODUCTION
Table 1-6: Projected N20 Emissions from Nitric Acid by Region: 2000-2020 (MtC02eq)
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
1.9
68.0
0.0
3.4
20.1
9.9
33.8
2.0
2.8
0.6
6.6
66.8
0.2
0.5
17.1
102.6
2010
1.9
68.5
0.0
4.0
22.1
9.4
36.2
2.2
3.0
0.7
6.5
68.4
0.2
0.5
15.5
107.0
2020
1.8
71.9
0.0
4.3
23.7
9.7
37.3
2.4
3.2
0.8
6.8
72.0
0.2
0.6
17.4
113.1
Source: USEPA, 2006.
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Table 1-7: Percentage Abatement for Nitric Acid for Selected Breakeven Prices ($/tC02eq): 2010-2020
2010
Country/Region
Africa
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Russian Federation
South & SE Asia
United States
World Total
$0
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
$15
0.00%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
0.00%
0.00%
88.94%
88.94%
$30
0.00%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
0.00%
0.00%
88.94%
88.94%
$45
0.00%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
0.00%
0.00%
88.94%
88.94%
$60
0.00%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
0.00%
0.00%
88.94%
88.94%
2020
$0
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
$15
0.00%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
0.00%
0.00%
88.94%
88.94%
$30
0.00%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
0.00%
0.00%
88.94%
88.94%
$45
0.00%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
0.00%
0.00%
88.94%
88.94%
$60
0.00%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
88.94%
0.00%
0.00%
88.94%
88.94%
Source: USEPA, 2003. Adapted from Nitric Acid Sector technology tables in Appendix B.
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-9
-------
SECTION IV — INDUSTRIAL PROCESSES • NITRIC AND ADIPIC ACID PRODUCTION
Table 1-8: Projected N20 Emissions from Adipic Acid by Region: 2000-2020 (MtC02eq)
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
1.0
34.1
0.0
1.7
10.0
5.0
16.9
1.0
1.4
0.3
3.3
33.5
0.1
0.2
8.6
51.4
2010
1.0
36.9
0,0
2.1
11.9
5.0
19.5
1.2
1.6
0.4
3.5
36.8
0.1
0.3
8.4
57.6
2020
1.0
40.3
0.0
2.4
13.3
5.4
20.9
1.4
1.8
0.4
3.8
40.4
0.1
0.3
9.8
63.5
Source: USEPA, 2006.
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Table 1-9: Percentage Abatement for Adipic Acid for Selected Breakeven Prices ($/tC02eq): 2010-2020
Country/Region
Africa
Australia/New
Zealand
Brazil
China
Eastern
EU-15
India
Japan
Mexico
Russian
South &
Europe
Federation
SE Asia
United States
World Total
$0
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
$15
0.00%
96.00%
96.00%
96.00%
96.00%
96.00%
0.00%
96.00%
0.00%
0.00%
96.00%
96.00%
96.00%
2010
$30
0.00%
96.00%
96.00%
96.00%
96.00%
96.00%
0.00%
96.00%
0.00%
0.00%
96.00%
96.00%
96.00%
$45
0.00%
96.00%
96.00%
96.00%
96.00%
96.00%
0.00%
96.00%
0.00%
0.00%
96.00%
96.00%
96.00%
$60
0.00%
96.00%
96.00%
96.00%
96.00%
96.00%
0.00%
96.00%
0.00%
0.00%
96.00%
96.00%
96.00%
$0
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
$15
0.00%
96.00%
96.00%
96.00%
96.00%
96.00%
0.00%
96.00%
0.00%
0.00%
96.00%
96.00%
96.00%
2020
$30
0.00%
96.00%
96.00%
96.00%
96.00%
96.00%
0.00%
96.00%
0.00%
0.00%
96.00%
96.00%
96.00%
$45
0.00%
96.00%
96.00%
96.00%
96.00%
96.00%
0.00%
96.00%
0.00%
0.00%
96.00%
96.00%
96.00%
$60
0.00%
96.00%
96.00%
96.00%
96.00%
96.00%
0.00%
96.00%
0.00%
0.00%
96.00%
96.00%
96.00%
Source: USEPA, 2003. Adapted from Nitric Acid Sector technology tables in Appendix B.
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
IV-10
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • NITRIC AND ADIPIC ACID PRODUCTION
Figure 1-2:
$1.5 -,
$1 2
o- $0.9 -
O
o
«» $0.6 -
$0.3 -
China
) 5 10 15 20 25 30
Absolute Reduction (MtCO2eq)
EU-15 = European Union.
Figure 1-3: EMF MACs for Top Five Emitting Country/Regions from Adipic Acid Production: 2020
!l i
$1.4 -
$1.1 -
$0.8 -
$ $0.5 -
O
0 $0.2 -
V*
-$0.1 -
-$0.4 -
-$0.7 -
-$1.0 -
i
• -
5
,
i i i i
10 15 20 25
— * EU-15
China
— • United States
- India
Absolute Reduction (MtCO2eq) — * Eastern Europe
EU-15 = European Union.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-11
-------
SECTION IV — INDUSTRIAL PROCESSES • NITRIC AND ADIPIC ACID PRODUCTION
Improved Emissions Factor Estimates
Current emissions factors are the result of laboratory experiments and only a few on-site facility
measurements. Additional facility measurements would greatly improve the accuracy of each country's
baseline emissions.
1V.1.5 Summary
Adipic acid producers in the United States have already adopted options to mitigate emissions of
N2O. Nitric and adipic acid production will continue to increase, correlating closely with the world's
demand for synthetic fertilizers and nylon. However, certain abatement options may mitigate significant
portions of a country's baseline if adopted by producers.
IV. 1.6 References
Chemical Week (CW). 2003. "Adipic Acid." Chemical Week. April 23, 2003. pg. 25.
Intergovernmental Panel on Climate Change (IPCC). 1996. Revised 1996 IPCC Guidelines for National
Greenhouse Gas Inventories: Reference Manual (Volume 3). Available at . As obtained on April 26, 2004.
Intergovernmental Panel on Climate Change (IPCC). 2000. Good Practice Guidance and Uncertainty
Management in National Greenhouse Gas Inventories. Intergovernmental Panel on Climate Change,
National Greenhouse Gas Inventories Programme, Montreal, IPCC-XVI/DOC. 10 (1.IV.2000).
Available at . As obtained on
January 10, 2005.
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. Intergovernmental Panel
on Climate Change, National Greenhouse Gas Inventories Programme, Montreal, IPCC-XVI/DOC. 10
(1.IV.2000). Available at .
Smit, A.W., M.M.C. Gent, and R.W. van den Brink. 2001. Market Analysis DeN20: Market Potential For
Reduction of N2O Emissions at Nitric Acid Facilities. Leiden, Netherlands: Jacobs Engineering
Nederland.
U.S. Environmental Protection Agency (USEPA). 2001. "U.S. Adipic Acid and Nitric Acid N2O Emissions
1990-2020: Inventories, Projections and Opportunities for Reductions." Washington, DC: USEPA.
U.S. Environmental Protection Agency (USEPA). 2003. International Analysis of Methane and Nitrous Oxide
Abatement Opportunities. Report to Energy Modeling Forum, Working Group 21. Appendices
"Nitrous Oxide Baselines." Washington, DC: USEPA. Available at . As obtained on March 25, 2005.
U.S. Environmental Protection Agency (USEPA). 2004. Inventory of U.S. Greenhouse Gas Emissions and Sinks
1990-2002. FRL-05-3794. Washington, DC: USEPA, Office of Solid Waste and Emergency Response.
Available at . As obtained on March 24, 2005.
U.S. Environmental Protection Agency (USEPA). 2006. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2020. Washington, DC: USEPA.
IV-12 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • NITRIC AND ADIPIC ACID PRODUCTION
Explanatory Notes
1. Separate emissions estimates for nitric and adipic acid were unavailable for 2005, thus projected
percentage changes are presented for 2000 to 2020. Note that individual percentage changes for nitric
and adipic acid are not comparable with the total percentage change of 16 percent, which is for 2005
to 2020.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-13
-------
SECTION IV — INDUSTRIAL PROCESSES • NITRIC AND ADIPIC ACID PRODUCTION
IV-14 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
IV.2 HFC Emissions from Refrigeration and Air-Conditioning
IV.2.1 Introduction
number of HFCs are used in refrigeration and air-conditioning systems and are emitted to
the atmosphere during equipment operation and repair. Specifically, emissions occur during
product and equipment manufacturing and servicing, and from disposal of equipment and
used refrigerant containers. Emissions also occur during equipment operation, as a result of component
failure, leaks, and purges. The use of refrigeration and air-conditioning equipment also generates indirect
emissions of greenhouse gases (primarily CO^) from the generation of power required to operate the
equipment. In some refrigeration and air-conditioning applications, these indirect emissions outweigh the
direct emissions. Therefore, energy efficiency has a major impact on the total greenhouse gas emissions of
an application. To the extent possible, both direct and indirect emissions were considered in the
refrigeration and air-conditioning analysis; however, options aimed solely at improving energy efficiency
rather than abating HFC emissions were not explored in detail. HFCs used in this sector have 100-year
GWPs that range from 140 to 11,700; the majority of HFCs used today in the refrigeration and air-
conditioning sector have GWPs from 1,300 (i.e., HFC-134a) to 3,300 (i.e., R-507A).
The refrigeration and air-conditioning sector includes eight major end-uses:
• household refrigeration,
• motor vehicle air-conditioning (MVAC),
• chillers,
• retail food refrigeration,
• cold storage warehouses,
• refrigerated transport,
• industrial process refrigeration, and
• residential and small commercial air-conditioning/heat pumps.
Each end-use is composed of a variety of equipment types that have historically used ODSs such as
chlorofluorocarbons (CFCs) or hydrochlorofluorocarbons (HCFCs). As the ODS phaseout is taking effect
under the Montreal Protocol, equipment is being retrofitted or replaced to use HFC-based substitutes or
intermediate substitutes (e.g., HCFCs) that will eventually need to be replaced by non-ozone-depleting
alternatives. HCFCs are beginning to be replaced with HFCs or other alternative refrigerants. The eight
major end-uses are explained in more detail below.
IV.2.1.1 Household Refrigeration
This end-use consists of household refrigerators and freezers. HFC-134a is the primary substitute for
CFC-12 in domestic refrigeration units in the United States and most developing countries, with
hydrocarbon (HC) refrigerant, especially isobutane (HC-600a), dominating much of the European market
and continuing to grow in market share. HC-600a is also gaining market share in Japan (Kuijpers, 2002).
The charge size of a typical household refrigeration unit in the United States has decreased over the past
15 years to about 0.17 kilograms for new HFC-134a units, with sizes even smaller in Europe.1 HC-600a
1 Differences in charge sizes are accounted for in the modeling methodology.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-15
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
systems are about 40 percent smaller than HFC-134a systems. The equipment has an expected lifetime of
20 years. This end-use is one of the largest in terms of the number of units in use; however, because the
charge sizes are small and the units are hermetically sealed (and, therefore, rarely require recharging),
emissions are relatively low. Thus, the potential for reducing emissions through leak repair is small. In
most Annex I countries, where regulations are in place that require the recovery of refrigerant from
appliances prior to disposal, the retirement of old refrigerators is not expected to result in significant
refrigerant emissions. Refrigerant emissions at disposal from developing countries, where refrigerant
recovery is not generally required, are expected to be greater. Emissions from the insulating foam in
household refrigerators and freezers are discussed in a separate chapter of this report.
IV.2.1.2 Motor Vehicle Air-Conditioning (MVAC)
This end-use includes the air-conditioning systems in motor vehicles (e.g., cars, trucks, and buses).
Currently, the quantity of refrigerant contained in a tvpical car air conditioner is approximately 1
kilogram—generally from 1 to 1.2 kilograms for vehicles containing CFC-12 systems, and an average of
approximately 0.8 kilograms for vehicles containing HFC-134a systems (Atkinson, 2000; European
Commission [EC], 2003) —although this varies by car and region (e.g., in Japan, the average amount is
about 0.5 kilograms). Because of concerns over the environmental impact of refrigerants, the average
charge size of MVACs—as well as associated leak rates—have been reduced over time; this trend is
expected to continue. The expected lifetime of MVACs is approximately 12 years. Refrigerant use in this
sector is significant because more than 700 million motor are vehicles registered globally (Ward's, 2001).
In developed countries, CFC-12 was used in MVACs until being phased out of new cars in 1992 through
1994. Since then, all air conditioners installed in new automobiles use HFC-134a refrigerant. HFC-134a is
also used as a retrofit chemical for existing CFC-12 systems (UNEP, 1998).
CFC-12 availability in developing countries and in some developed countries (e.g., the United States)
has resulted in its use for servicing older MVACs that were originally manufactured as CFC-12 systems.
A variety of refrigerant blends are approved for use in the United States by the USEPA as replacements
for CFC-12 in MVACs. However, these blends have not been endorsed by vehicle or system
manufacturers. Globally, these blends have captured only a small and declining share of the retrofit
market. Some conversions from CFC-12 to pure HCs have been done. However, this is illegal in the
United States, and such use in direct expansion systems not designed for a flammable refrigerant can
pose safety concerns and is not considered acceptable by much of the global MVAC industry. Climate
change concerns associated with the use of HFC-134a resulted in research into and development of other
MVAC alternatives. Possible alternatives to HFC-134a systems include transcritical CO2 systems,
hydrocarbons (e.g., in new secondary-loop systems), and HFC-152a systems, all of which are under study
and development (SAE, 2003a).
IV.2.1.3 Chillers
Chillers are used to regulate the temperature and reduce humidity in offices, hotels, shopping
centers, and other large buildings, as well as in specialty applications on ships, submarines, nuclear
power plants, and other industrial applications. The four primary types of chillers are centrifugal,
reciprocating, scroll, and screw, each of which is named for the type of compressor employed. Chillers
last longer than most air-conditioning and refrigeration equipment. The majority of operating chillers will
remain in service for more than 20 years, and some will last 30 years or more. A wide variety of chillers is
available, with cooling capacities from 7 kilowatts to over 30,000 kilowatts (RTOC, 2003). The charge size
of a chiller depends mostly on cooling capacity and ranges from less than 25 kilograms (reciprocating) to
over 2,000 kilograms (centrifugal). HCFC-123 has been the refrigerant of choice as a retrofit option for
newer CFC-11 units, and HFC-134a has been the refrigerant of choice as a retrofit option for newer CFC-
IV-16 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
12 units. The replacement market for CFC-12 high-pressure chillers and CFC-11 low-pressure chillers is
dominated by both HCFC-123 chillers and HFC-134a chillers in developed and developing countries.
Following phaseout of the production of HCFCs (in 2030 for developed countries and 2040 for
developing countries), recycled, recovered, and reclaimed HCFCs will continue to be used in most
countries. This trend is not the case, however, in the European Union (EU-25), where there are restrictions
on the use of HCFCs in new equipment, the production of HCFCs is not permitted beyond 2010, and
recycled HCFCs may not be reused beyond 2015. In the EU, HFC-134a will be an important option for
chillers, but because of its global warming impact, ammonia chillers are being used as an alternative in
some countries (Kuijpers, 2002).
Additionally, HFC-245fa is a potential refrigerant for new low-pressure chillers. However, for a
variety of reasons, the commercialization of this chiller technology is not likely to occur in the near future,
if at all. High-pressure chillers that currently use HCFC-22 will ultimately be replaced by several HFC
refrigerant blends and HFC-134a chillers. Likewise, existing CFC-114 chillers have been converted to
HFC-236fa or replaced with HFC-134a chillers, for use primarily in specialty applications (e.g., on ships
and submarines, and in nuclear power plants) (RTOC, 2003; IPCC/TEAP, 2005).
IV.2.1.4 Retail Food Refrigeration
Retail food refrigeration includes refrigerated equipment found in supermarkets, convenience stores,
restaurants, and other food service establishments. This equipment includes small refrigerators and
freezers, refrigerated display cases, walk-in coolers and freezers, and large parallel systems. Charge sizes
range from 6 to 1,800 kilograms, with a lifetime of about 15 years. Convenience stores and restaurants
typically use standalone refrigerators, freezers, and walk-in coolers. In contrast, supermarkets usually
employ large parallel systems that connect many display cases to a central compressor rack and
condensing unit by means of extensive piping. Because the connection piping can be miles long, these
systems contain very large refrigerant charges and often experience high leakage rates.
During the earlier phases of the CFC phaseout in developed countries, the use of HCFC-22 in retail
food refrigeration was expanded considerably. Retail food equipment is being retrofitted with HCFC-
based blends, although HFC blends are also used as a retrofit refrigerant. The HFC blend R-404A is the
preferred refrigerant in new retail food equipment in developed countries, while R-507A is also used
extensively in the market (Kuijpers, 2002). In developed countries, both distributed and centralized
systems that use HFCs, HCs, ammonia, and CO2 are being developed (both with and without secondary
loops) (Kuijpers, 2002).
IV.2.1.5 Cold Storage Warehouses
Cold storage warehouses are used to store meat, produce, dairy products, and other perishable
goods. The expected lifetime of a cold storage warehouse is 20 to 25 years, and although charge sizes vary
widely with system size and design, a rough average is about 4,000 kilograms. Warehouses in developed
countries have historically used CFC-12 and R-502 refrigerants and currently use HCFC-22, R-404A, and
R-507A. The latter two refrigerants are expected to replace HCFC-22 in new warehouses. Retrofits are
also possible; for example, existing CFC-12 cold storage warehouses can be retrofitted with R-401A, and
existing R-502 warehouses can be retrofitted with R-402A. Not all cold storage warehouses use
halocarbon refrigerants. Many facilities, for example, use ammonia in secondary loop brine systems.
IV.2.1.6 Refrigerated Transport
The refrigerated transport end-use includes refrigerated ship holds, truck trailers, railway freight
cars, refrigerated rigid vans/trucks, and other shipping containers. Although charge sizes vary greatly,
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-17
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
the average charge sizes are relatively small (7 to 8 kilograms). The expected lifetime of a refrigerated
transport system is 12 years. Trailers, railway cars, and shipping containers using CFC-substitute
refrigerants are commonly charged with HFC-134a, R-404A, and HCFC-22 (UNEP, 1999a). Ship holds, on
the other hand, rely on HCFC-22 (UNEP, 1999a) and ammonia. In addition to HFC-134a, R-404A can be
used in new equipment. Existing equipment can be retrofitted with R-401A, R-402A, R-404A, R-507A, and
other refrigerants. In addition, refrigerated transport equipment includes systems that operate based on
the evaporation and expansion of liquid CO2 or nitrogen.
IV.2.1.7 Industrial Process Refrigeration
Industrial process refrigeration includes complex, often custom-designed refrigeration systems used
in the chemical, petrochemical, food processing, pharmaceutical, oil and gas, and metallurgical
industries; in sports and leisure facilities; and in many other applications. Charge sizes typically range
from 650 to 9,100 kilograms, and the average lifetime is approximately 25 years. Ammonia, HCs, HCFC-
123, and HFC-134a are expected to be the most widely used substitute refrigerants for new equipment in
the near future (UNEP, 1999a). Upon completion of the HCFC phaseout, HFC-134a, R-404A, and R-507A
are expected to be the primary refrigerants used in this end-use.
IV.2.1.8 Residential and Small Commercial Air-Conditioning and Heat Pumps
Residential and small commercial air-conditioning (e.g., window units, unitary air conditioners, and
packaged terminal air conditioners) and heat pumps are another source of HFC emissions. Most of these
units are window and through-the-wall units, ducted central air conditioners, and nonducted split
systems. The charge sizes of the equipment in this sector range from 0.5 to 10 kilograms for residential
systems, and about 10 to 180 kilograms for commercial systems based on cooling capacity requirements.
The average lifetime of this type of equipment is 15 years. Residential and commercial air-conditioning
has been relying almost exclusively on HCFC-22 refrigerant. R-410A, R-407C, and HFC-134a are currently
used to replace HCFC-22 in some new equipment for most end-uses, and this trend is expected to
continue as HCFC-22 is phased out. In particular, R-410A is expected to dominate the U.S. residential
market in the future, whereas R-407C is expected to replace HCFC-22 in retrofit applications and some
new residential and commercial equipment. Other countries may experience different patterns of R-410A
and R-407C use.
IV.2.2 Baseline Emissions Estimates
IV.2.2.1 Emissions Estimating Methodology
Description of Methodology
Specific information on how the model calculates refrigeration and air-conditioning emissions is
described below.
The USEPA's Vintaging Model and industry data were used to simulate the aggregate impacts of the
ODS phaseout on the use and emissions of various fluorocarbons and their substitutes in the United
States. Emissions estimates for non-U.S. countries incorporated estimates of the consumption of ODSs by
country, as provided by the United Nations Environment Programme (UNEP, 1999b). The estimates for
EU-15 were provided in aggregate, and each country's gross domestic product (GDP) was used as a
proxy to divide the consumption of the individual member nations by the EU-15 total. Estimates of
country-specific ODS consumption, as reported under the Montreal Protocol, were then used in
conjunction with Vintaging Model output for each ODS-consuming sector. In the absence of country-level
data, preliminary estimates of emissions were calculated by assuming that the transition from ODSs to
IV-18 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
HFCs and other substitutes follows the same general substitution patterns internationally as observed in
the United States. From this preliminary assumption, emissions estimates were then tailored to
individual countries or regions by applying adjustment factors to U.S. substitution scenarios, based on
relative differences in (1) economic growth; (2) rates of ODS phaseout; and (3) the distribution of ODS use
across end-uses in each region or country, as explained below.
Emissions Equations
For refrigeration and air-conditioning products, emissions calculations were split into two categories:
emissions during equipment lifetime, which arise from annual leakage and service losses, and disposal
emissions, which occur at the time of discard. The first equation calculates the emissions from leakage
and service, and the second equation calculates the emissions resulting from disposal of the equipment.
These service, leakage, and disposal emissions were added to calculate the total emissions from
refrigeration and air-conditioning. As new technologies replace older ones, improvements in their
leakage, service, and disposal emissions rates were assumed to occur.
Emissions from any piece of equipment include both the amount of chemical leaked during
equipment operation and the amount emitted during service. Emissions from leakage and servicing can
be expressed as follows:
SJ = (la + ls) x QcH+1 for / = 1 -> k (2.1)
where
Es = Emissions from equipment serviced. Emissions in year j from normal leakage and
servicing of equipment.
la = Annual leakage rate. Average annual leakage rate during normal equipment operation,
expressed as a percentage of total chemical charge.
ls = Service leakage rate. Average annual leakage from equipment servicing, expressed as a
percentage of total chemical charge.
Qc = Quantity of chemical in new equipment. Total amount of a specific chemical used to
charge new equipment in a given year, by weight.
j = Year of emissions.
i = Counter. From 1 to lifetime (k).
k = Lifetime. The average lifetime of the equipment.
Note: It is recognized that leakage rates are not a function of the total system, but change with system
pressure and temperature. For instance, when equipment charges are diminished because of refrigerant
losses (i.e., leakage), system pressures are also reduced somewhat and the leakage rate changes. This
change becomes appreciable once the entire liquid refrigerant is gone. The average leakage rates used in
the equation above were intended to account for this effect. The rates also accounted for the range of
equipment types (from those that do not leak at all to those with high leaks) and service practices (i.e.,
proper refrigerant recovery and refrigerant venting).
Emissions also occur during equipment disposal. The disposal emissions equations assumed that a
certain percentage of the chemical charge will be emitted to the atmosphere when that vintage is
discarded. Disposal emissions are thus a function of the quantity of chemical contained in the retiring
equipment fleet and the proportion of chemical released at disposal:
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-1 9
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
Edj = QCj-M*[l-(rmxrc)] (2.2)
where
Ed = Emissions from equipment disposed. Emissions in year j from the disposal of equipment.
Qc = Quantity of chemical in new equipment. Total amount of a specific chemical used to
charge new equipment one lifetime (k) ago, by weight.
rm = Chemical remaining. Amount of chemical remaining in equipment at the time of disposal,
expressed as a percentage of total chemical charge.
re = Chemical recovery rate. Amount of chemical that is recovered just prior to disposal,
expressed as a percentage of chemical remaining at disposal (rm).
j = Year of emissions.
i = Counter. From 1 to lifetime (k).
k = Lifetime. The average lifetime of the equipment.
Finally, lifetime and disposal emissions were summed to provide an estimate of total emissions:
E} = ESJ + Edj (2.3)
where
E = Total emissions. Emissions from refrigeration and air-conditioning equipment in year j.
Es = Emissions from equipment serviced. Emissions in a given year from normal leakage and
servicing (recharging) of equipment.
Ed = Emissions from equipment disposed. Emissions in a given year from the disposal of
equipment.
j = Year of emissions.
Regional Variations and Adjustments
From the general methodology, the following regional assumptions were applied:
• Adjustment for Regulation (EC) No 2037/2000. Countries in the EU-15 were assumed to be in
full compliance with Regulation (EC) No 2037/2000, which stipulates that no new refrigeration
and air-conditioning equipment should be manufactured with HCFCs, as of January 1, 2002.2 The
European Commission (EC) regulation also bans the use of HCFCs for servicing equipment after
January 1, 2015. Compliance with these regulations will likely lead to increased use of HFCs to
replace HCFCs. These changes were assumed to correspond to increased emissions of 20 percent
in 2005, 15 percent in 2010, and 15 percent in 2020, relative to what the EU-15 baseline otherwise
would be. These relative emissions increases were determined by running a Vintaging Model
scenario where the uses of HCFCs were assumed to comply with the regulation. No adjustments
for Regulation (EC) No 2037/2000 were made to the 10 countries that joined the EU in March
2004, as this analysis was conducted prior to this date.
2 The ban was delayed until July 1, 2002, for fixed air-conditioning equipment with a cooling capacity of less than 100
kW and until January 1, 2004, for reversible air-conditioning/heat pump systems.
IV-20 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
• Recovery and Recycling Adjustments. For developing (i.e., non-Annex I) countries, countries
with economies in transition (CEITs), and Turkey, the emissions were increased by
approximately 20 percent over initial estimates to reflect the assumed low levels of recovery and
recycling of refrigerants from small end-uses (i.e., MVACs, commercial and residential air-
conditioning, refrigerated transport, and other appliances), relative to the United States. This
assumed increase in emissions from lower levels of recovery and recycling was based on an
analysis of a variety of scenarios using the Vintaging Model, where emissions were first projected
assuming an 80-percent baseline recovery rate to reflect the assumed status quo in developed
countries and then projected again assuming a 30-percent baseline recovery rate to reflect the
assumed status quo in developing countries. The GWP-weighted emissions in the latter low-
recovery scenario were determined to be approximately 20 percent higher than in the former
high-recovery scenario (ICF Consulting, 2002a).
• Market Adjustments. The baseline assumes that HC and ammonia refrigerants and other non-
HFC or low-emitting options will penetrate international markets more than the United States
market because of differences in safety standards; greater acceptance of non-HFC choices by
industry, end-users, regulators, and insurance companies; and increased public and regulatory
scrutiny to reduce HFC emissions. To reflect this penetration, baseline emissions estimates of
non-U.S. countries were reduced by the following amounts (Table 2-1).
Table 2-1: Reductions in Baseline Emissions in Non-U.S. Countries to Reflect Market Adjustments
Country/Region
EU-15
Japan
Non-EU-1 5 Europe
CEITs
Australia/New Zealand
All other countries
Percent
30a
30
20a
20
10
20
EU-15 = European Union; CEITs = countries with economies in transition.
a The new EC Directive on MVACs, which bans the use of HFC-134a in new vehicle models in 2011 and in all vehicles in 2017, was not
considered in developing these baseline emissions adjustments for EU countries, as the directive was not finalized at the time this analysis
was conducted.
These assumptions were based solely on qualitative information on current and future global
market penetration of low-GWP refrigerants, as well as low-emission technologies and practices.
For example, HC technology is believed to dominate the domestic refrigeration market in
Western Europe, particularly in Germany and Scandinavia. HC domestic refrigerators are
produced by major manufacturers in Germany, Denmark, Italy, Japan, United Kingdom, France,
Spain, and Sweden. Some of the largest manufacturers in China, India, Indonesia, Australia,
Korea, and Cuba are also producing domestic refrigerators that use HCs (Greenpeace, 2001;
Japan Times, 2002). To reflect this and many other trends, baseline emissions from non-U.S.
countries were adjusted downward, as shown above.
• Redistribution of Emissions by End-Use, Based on MVAC Analysis. Based on a variety of
available data on international motor vehicle sales, air-conditioning usage, and MVAC emissions,
a separate analysis was conducted to estimate total MVAC emissions by region. These MVAC
emissions estimates by region were then used to determine the relative share of refrigeration and
air-conditioning emissions attributable to MVACs and to reapportion emissions from all other
end-uses accordingly, relative to the end-use breakout calculated for the United States. The
methodology used to perform this analysis is explained in detail below.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-21
-------
SECTION IV— INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
MVAC Analysis
The Vintaging Model estimates MVAC emissions for the United States based on vehicle sales data,
assumptions on the percentage of vehicles with functional air-conditioning, and a projected growth rate
of 2.6 percent (based on sales data from 1970 through 2001). Table 2-2 presents the Vintaging Model's
estimated percentage of baseline refrigeration and air-conditioning emissions attributable to MVACs in
the United States from 2005 through 2020.
Table 2-2: Estimated Percentage of GWP-Weighted Refrigeration and Air-Conditioning HFC Emissions
Attributable to MVACs in the United States
Percent
2005
35.9
2010
27.6
2015
22.6
2020
19.9
However, because the market penetration of air-conditioning into vehicles is assumed to be different
in other countries and regions,3 and because MVACs are assumed to account for a different proportion of
total refrigeration and air-conditioning emissions in the United States compared with most other
developed and developing countries, this end-use has been modeled separately to achieve a higher
degree of accuracy in emissions estimates. To this end, for all countries for which data on MVACs or
historical vehicle sales were available, country-specific MVAC models were developed to estimate the
total number of MVACs in past, present, and future years. Ward's World Motor Vehicle Data (2001), the
Society of Indian Automobile Manufacturers (SIAM) (2005), and the China Association of Automobile
Manufacturers (2005) were used as data sources.
The remainder of this section describes the assumptions and data used to project the number of
MVACs by country and region. It should be noted that, while the MVAC industry is investigating new
refrigerants and other emissions reduction initiatives (see http://www.epa.gov/cppd/i-nac/), these actions
are not considered in the baseline estimates.
India
India's MVAC fleet estimates were developed based on (1) data on MVAC sales prior to 2004, from
SIAM (2005), (2) projected annual growth rates of new vehicle sales, and (3) projected annual growth
rates of air-conditioning penetration. Specifically, India's future vehicle fleet growth was assumed to be 8
percent per year,4 while air-conditioning penetration was assumed to increase linearly to reach 95 percent
in 2010.5 Beyond 2010, it was assumed that air-conditioning penetration will be maintained at 95 percent
because vehicle air-conditioning will become standard. The assumed air-conditioning market penetration
rates for India are summarized in Table 2-3.
Table 2-3: Percentage of Newly Manufactured Vehicles Assumed to Have Operational Air-Conditioning Units
in India
Percent
2005
92.5
2010
95
2015
95
2020
95
3 Except for Japan, which is assumed to have the same market penetration rate of MVACs into new vehicles as the
United States.
4 This growth rate was based on the annual growth rate of passenger vehicles (assumed to be linear) between 2000
and 2004, with the fleet size in 2000 based on Ward's (2001) and the fleet size in 2004 based on SIAM (2005).
5 Air-conditioning penetration was grown from 92 percent in 2004, based on data from SIAM (2005).
IV-22 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
China
MVAC estimates for China are based on data on Chinese production of vehicles with air-conditioning
from 1994 to 2004, provided by the China Association of Automobile Manufacturers (2005). Projections of
future MVACs in China were based on the assumed growth rate of India's vehicle market beyond 2005
(assumed to be 8 percent per year, as described above).6 The same assumptions were applied to Hong
Kong.
All Other Countries
For all countries other than the United States, Japan, India, China, and Hong Kong, the number of
operational MVACs was estimated based on (1) annual historical sales of passenger cars and light trucks,
as provided in Ward's (2001), and (2) estimates of the percentage of the vehicle fleet equipped with air-
conditioning, based on quantitative and qualitative data provided in EC (2003); Hill and Atkinson (2003);
OPROZ (2001); and Barbusse, Clodic, and Roumegoux (1998), as presented in Table 2-4.
Table 2-4: Percentage of Newly Manufactured Vehicles Assumed to Have Operational Air-Conditioning Units
in All Other Countries
Country/Region
All other Annex 1 countries
Latin America and Caribbean
All other non-Annex 1 countries, Russian
Federation, and Ukraine
2005
65.5
50.0
23.0
2010
70.0
55.0
28.0
2015
80.5
60.0
33.0
2020
95.0
65.0
38.0
As shown above, MVACs were assumed to increasingly penetrate the vehicle fleet over time. In the
developing countries that were modeled, this rate of increase was assumed to be 1 percent each year,
while in all other Annex I countries, the rate of increase was assumed to be more rapid, reaching 95
percent of the vehicle fleet in 2020 (EC, 2003; Hill and Atkinson, 2003).
Once the MVAC fleet was estimated by country/region, annual MVAC emissions were calculated
assuming annual average leak and service emissions of 10.9 percent.7 MVAC emissions at disposal were
assumed to be 42.5 percent of the original MVAC charge in developed countries and 69 percent in
developing countries (as a result of zero recovery assumed).8 All systems were assumed to use HFC-134a
refrigerant in the baseline. The new EC Directive on MVACs9 was not considered in the baseline
estimates, as this directive was not finalized at the time this analysis was conducted.
6 India's projected growth rate was selected for use in place of China's historical growth rate because China's
historical growth rate (of approximately 25%) was considered unrealistically high to maintain for 2.5 decades.
7 This emissions rate includes emissions released during routine equipment operation from leaks, as well as those
released during the servicing of equipment by both professionals and do-it-yourselfers.
8 This percentage (69 percent) is the implied loss at disposal given the assumption that twice the original MVAC
charge is emitted over the course of a vehicle's lifetime in developing countries.
In April 2006, the European Parliament adopted a legislative resolution on the joint text approved by the
Conciliation Committee for a directive of the European Parliament and of the Council relating to emissions from air
conditioning systems in motor vehicles and amending Council Directive 70/156/EEC. The directive places a ban on
the use of fluorinated gases with a GWP of more than 150 in new vehicle models planned from 2011 onwards, and in
all vehicles from 2017 onwards.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-23
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
Once MVAC emissions were estimated by country/region, the proportion of MVAC emissions as a
percentage of the total refrigeration and air-conditioning emissions (developed using the methodology
described above) was calculated. These percentages were then averaged by region. The average estimated
percentage of refrigeration and air-conditioning GWP-weighted emissions that are attributable to
MVACs by regional grouping are presented in Table 2-5.
Table 2-5: Estimated Percentage of Refrigeration and Air-Conditioning HFC Emissions Attributable to MVACs
Country/Region
United States and Japan
All other Annex 1 countries
China, Hong Kong, and India
Latin America and Caribbean
Russian Federation, Ukraine, and all other non-
Annex 1 countries
2005
35.9
46.9
41.3
14.2
3.8
2010
27.6
42.8
53.0
13.3
3.8
2015
22.6
31.8
62.0
12.6
5.4
2020
19.9
36.6
65.8
12.0
8.0
Based on the above percentage of sector baseline emissions assumed to come from MVACs for each
region, for lack of reliable data to suggest otherwise, the U.S. baseline emissions breakout by end-use was
used to proportionally redistribute the remaining emissions of a particular country/region. For example,
because MVACs contributed only 14.2 percent of total sector emissions in Latin American countries in
2005, the balance of emissions in Latin America was distributed across all other end-uses, in proportion to
the U.S. end-use breakout. The resulting subdivision of baseline GWP-weighled HFC emissions by end-
use and region are summarized in Table 2-6. These emissions subdivisions by end-use help determine the
maximum amount of emissions that can be avoided by any given abatement option, because each option
is applicable only to specific end-uses.
IV.2.2.2 Baseline Emissions
The amount of HFC emissions from MVAC units is expected to rise, because HFC-134a has been the
primary refrigerant used in the growing automobile industry, and because HFC-134a is the primary
refrigerant used to replace older CFC-12 systems. The baseline for MVACs assumes a mix of
professionally serviced systems and those serviced by people without recovery equipment. Because
commercial unitary and residential air-conditioning equipment has yet to transition fully into HFCs, the
emissions of HFCs from these end-uses in 2005 were estimated to be relatively insignificant, but will
increase substantially over time. Retail food systems are expected to (and in many cases, already have)
transition at least partially to HFC-134a and HFC-containing blends because of certain equipment
characteristics (such as their large number of fittings); such systems may have higher refrigerant
emissions rates. Cold storage systems also have large charge sizes, but their emissions relative to other
refrigeration and air-conditioning end-uses are not expected to increase significantly. HFC emissions
from chillers are relatively low as a result of the continued use of HCFC-123 in this application,10 as well
as the low leakage rates of new HFC-134a units. The baseline emissions projections assumed that the
recovery and recycling of refrigerants during service and disposal in Annex I countries will curtail
emissions across all end-uses.
10 Note that emissions of all CFC and HCFC refrigerants, including HCFC-123, were not included in the baseline
emissions estimates.
IV-24
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
The resulting baseline estimates of HFC emissions are summarized in Table 2-7 and Figure 2-1 in
million metric tons of carbon dioxide equivalents (MtCO2eq).
IV.2.3 Cost of HFC Emissions Reduction from Refrigeration and Air-
Conditioning
This section presents a cost analysis for achieving HFC emissions reductions from the emissions
baselines presented above. Each abatement option is described below, but only those options not
assumed to occur in the baseline and for which adequate cost data are available were included in the cost
analysis. To the extent possible, this analysis considered total equivalent warming impacts (TEWI)11 to
account for the climate and cost impacts of energy consumption (i.e., indirect emissions). Because of data
limitations, a full life cycle analysis was not possible. For example, the cost and emissions impacts
associated with (1) the manufacture of refrigerant and all system components, (2) the energy required for
reclamation, and (3) the recycling of all system components at the end of equipment life were not
assessed in this analysis.
The remainder of this section describes the economic assumptions for these abatement options.
IV.2.3.1 Description and Cost Analysis of Abatement Options
HFC emissions from refrigeration and air-conditioning equipment can be reduced through a variety
of practice and technology options. Many of the options considered in this report would require
voluntary action by the private sector or further government regulation. For example, national
governments can regulate maximum allowable leakage rates for refrigeration and air-conditioning
equipment and/or require the recovery of refrigerant and the proper disposal of nonreclaimable
refrigerant. Many Annex I countries have already implemented a variety of such regulatory actions to
reduce ODS emissions. Some of the most widely recognized options to reduce refrigerant emissions are
listed below (UNEP, 1998; UNEP, 1999a; Crawford, 1999; USEPA, 2001a).
Practice Options
• leak repair
• refrigerant recovery and recycling
• proper refrigerant disposal
• technician certification and HFC sales restriction
Alternative Refrigerant Options
• ammonia
• HCs
• CO2
• other low-GWP refrigerants
11 TEWI is the combined effects of direct greenhouse gas impacts (i.e., chemical emissions) and indirect greenhouse
gas impacts (i.e., energy-related CC>2 emissions).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-25
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDI HONING
Table 2-6: Distribution of Refrigeration- and Air-Conditioning-Sector HFC Emissions by End-Use, Region,
and Year (Percent)
End-Use
United
States and
Japan
All Other Latin
Annex 1 America and
Countries Caribbean
All Other Non-
China, Annex 1 Countries,
Hong Kong, Russian Federation,
and India and Ukraine
2005
Chillers
Retail food
Cold storage
Industrial process
Commercial air-conditioning
Residential air-conditioning
Refrigerated transport
Other appliances8
MVACs
3.2
39.0
1.2
4.6
1.1
0.6
14.0
0.5
35.9
2.7
32.3
1.0
3.8
0.9
0.5
11.6
0.4
46.9
4.3
52.2
1.6
6.1
1.4
0.8
18.8
0.6
14.2
3.0
35.7
1.1
4.2
1.0
0.6
12.8
0.4
41.3
4.8
58.4
1.8
6.8
1.6
0.9
21.0
0.7
3.8
2010
Chillers
Retail food
Cold storage
Industrial process
Commercial air-conditioning
Residential air-conditioning
Refrigerated transport
Other appliances8
MVACs
2.3
41.7
1.4
6.0
5.3
5.5
9.7
0.4
27.6
1.8
33.0
1.1
4.8
4.2
4.4
7.7
0.3
42.8
2.8
50.0
1.7
7.2
6.3
6.6
11.6
0.5
13.3
1.5
27.0
0.9
3.9
4.3
3.6
6.3
0.3
53.2
3.1
55.4
1.9
8.0
7.0
7.4
12.9
0.6
3.8
2015
Chillers
Retail food
Cold storage
Industrial process
Commercial air-conditioning
Residential air-conditioning
Refrigerated transport
Other appliances8
MVACs
1.8
41.2
1.4
6.4
8.8
9.7
7.2
1.0
22.6
1.6
36.3
1.2
5.6
7.8
8.5
6.3
0.9
31.8
2.0
46.5
1.6
7.2
10.0
10.9
8.1
1.1
12.6
0.9
20.2
0.7
3.1
4.3
4.7
3.5
0.5
62.0
2.2
50.3
1.7
7.8
10.8
11.8
8.7
1.2
5.4
2020
Chillers
Retail food
Cold storage
Industrial process
Commercial air-conditioning
Residential air-conditioning
Refrigerated transport
Other appliances8
MVACs
1.5
39.1
1.4
6.6
. 11.3
13.3
6.1
0.8
19.9
1.2
31.0
1.1
5.2
8.9
10.5
4.9
0.6
36.6
1.6
43.0
1.6
7.3
12.4
14.6
6.7
0.9
12.0
0.6
16.7
0.6
2.8
4.8
5.7
2.6
0.3
65.8
1.7
44.9
1.6
7.6
12.9
15.2
7.0
0.9
8.0
Note: Totals may not sum because of independent rounding.
a Other appliances include refrigerated appliances, dehumidifiers, and ice makers.
IV-26
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
Table 2-7: Total Baseline HFC Emissions from Refrigeration and Air-Conditioning (MtC02eq)
Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
2.8
95.1
1.3
1.5
4.1
0.9
13.3
0.5
16.4
1.4
1.8
98.5
1.3
2.9
58.0
117.0
2010
12.8
244.9
3.2
6.9
25.8
4.2
37.9
2.6
32.6
6.6
9.3
260.8
6.9
14.7
148.6
356.4
2020
20.4
414.4
5.6
12.0
61.7
7.3
58.4
5.4
45.1
11.2
17.3
441.4
13.4
28.1
264.6
627.3
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Figure 2-1: Baseline HFC Emissions from Refrigeration and Air-Conditioning by Region (MtC02eq)
700
1990
2000
2010
2020
Year
D Middle East
• Africa
• Non-EU FSU
E Latin America
• S&E Asia
D China/CPA
• OECD90+
CPA = Centrally Planned Asia; Non-EU FSU = non-European Union Former Soviet Union countries; OECD90+ = Organisation for Economic
Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-27
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
Technology Options
• distributed systems12 for stationary commercial refrigeration equipment
• secondary loop systems for stationary equipment, including HFC secondary loop systems and
ammonia secondary loop systems
• enhanced HFC-134a systems in MVACs
• HFC-152a refrigerant in MVACs (direct expansion or secondary loop systems)
• CO2 systems in MVACs
• oil-free compressors
• geothermal (in lieu of air-to-air) cooling systems
• desiccant cooling systems
• absorption systems
Table 2-8 summarizes the duration and applicability of the process and technology emissions
reduction options across all end-use applications considered in this analysis. The applicability of the
alternative refrigerant options depends on the technology used; hence, some options were explored in
more detail in the analysis of technology options. Consideration of distribution costs associated with the
technology options was not included in the analysis. All costs are presented in 2000 dollars.
The following section describes all of these options in greater detail and presents a cost analysis for
those options not assumed to occur in the baseline and for which adequate cost data were available. The
resulting emissions abatement potentials and costs of each option explored in the cost analysis are
summarized in Section IV.2.4. The technology options explored in this chapter do not include retrofit
costs and, therefore, were assumed to penetrate only the markets of new (not existing) equipment. New
equipment is defined as air-conditioning and refrigeration equipment manufactured in 2005 or later.
Detailed descriptions of the cost and emissions reduction analysis for each option can be found in
Appendix F for this chapter.
12 The term distributed system, as used in this report, refers to commercial refrigeration equipment used in retail
food and cold storage applications, although the term could also refer to equipment used in other applications, such
as residential and small commercial air-conditioning.
IV-28 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV— INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
o
"5.
O
_o
t>
3
T3
a>
E """
a> , o>
5 is |
35 -e
11
en LU
2
=
£
U
(O
.1
s
IS-3T
in *- j;
*S 3 S
^P ^*
§1^-
1
a
o
if
U
1
m \ m m
m \ m m
m \. m m
\ m . .
\ m m m
\ • • «
in r— Z Z
tice Options
repair
gerant recovery
er refrigerant disposal
inician certification
i JS ir ct K?
...
• S S
. . .
. . .
\ . . .
\ • • •
O) CD 0) CD
E E E E
Q. o- a. o_
'5 '5 •= "5
cr o~ o~ cr
O) CD CD CD
"5 "5 "S "S
•1 1 -I -1
•native Refrigerants
lonia
r low-GWP refrigerants
S fc CO tM 02
S E 0 O €
< < I O O
. . . .
s s \ •
. .
.
s s s.
s s s
"
c H c c c ccc He
tE E 1 E E!E EE
Q_ Q- Q_ O- CL Q. Q. Q. Q-
"n "5 "3 "3 "13 "3 "3 "^ "3 '3
§f oT SiT S^ S" S^S^S^ S^S"
1 .1 1 .1 1 1 1 .1 .1 1
"CD ^3 "CD "53 '53 "oS'co'SS 'S5'a)
-^ *- £• c en
f: c c as o c
1 S g § 'g o
ccT-9- '•§•='§£•— °- -i
o CT ^ 92 »— c c ^ 'co
! I- J I
lyi ti
1^5 co co LU x: ooo Q<
g
IS
o
CD
CL
CD
tf
. O
"C Q-
o o>
Q. *—
CD en
CO _>,
'«> CO
^- c=
CO CO
a
«-§
Ci CO CO CD
CO c c S
If f|
:S .
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-29
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
Practice Options
Four practice options are discussed in this section—leak repair, refrigerant recovery, proper
refrigerant disposal, and technician certification. Together with additional measures (including designing
and installing equipment to minimize HFC emissions), these practices are often considered standard
good practices and are identified in a number of different responsible use guides —such as that published
by the Alliance for Responsible Atmospheric Policy (ARAP) (see http://www.arap.org/
responsible.html) —and endorsed through voluntary industry partnerships, including those initiated by
the USEPA (see http://www.epa.gov/ozone/snap/emissions/index.html). However, this report assumes
that there are opportunities to further apply these options to reduce emissions from the baseline prepared
for this report.
Leak Repair for Large Equipment
Reducing leakage rates can significantly reduce HFC emissions, especially in systems such as chillers,
cold storage warehouses, and retail food systems that can leak large amounts of refrigerant. Although
some of the options available may be impractical for existing equipment, given the difficulty and expense
of retrofitting, there are still many options that are economically feasible. Some of the leak repair options
used in current industry practice include
• use of preventive maintenance, including scheduled inspection and repairs;
• monitoring of leaks using stationary leak monitors or other new technologies, such as early
warning signals,13 remote monitoring, and diagnostics;
• use of new, more durable gasket materials that provide tighter seals and absorb less refrigerant;
• augmentation of threaded joints with O-ring seals;
• augmentation or replacement of gaskets and O-ririgs with adhesive sealants;
• broader use and improvement of brazing techniques rather than threaded or snap fittings (e.g.,
use of sufficient silver content14 and use of dry nitrogen or other inert gas to avoid oxidation);
• focus on ensuring accessibility to field joints and use of isolation valves, which allows for greater
ease of repair;
• focus on proper securing to reduce vibration fractures in the pipe and connections from the
compressor and other moving parts of the system;
• repair or retrofit of high-emitting systems through targeted component upgrades;15 and
• performance of major modifications to the systems (USEPA, 1997; USEPA, 1998; Calm, 1999).16
13 Technologies in the final stages of development are expected to generate early warning signals at less than 5
percent charge loss in commercial refrigeration and air-conditioning systems (Gaslok, 2002).
14 For solder, a 15-percent silver content is recommended (USEPA, 1997).
15 This option may include replacing the purge unit or other component upgrades that typically require the removal
of refrigerant from the machine, 2 full days of two technicians' time, and several thousand dollars' worth of materials
(USEPA, 1998).
IV-30 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV— INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
As suggested by the above list, leak reduction options range from simple repairs to major system
upgrades. Even in countries where maximum allowable leakage rates are regulated by law, further leak
reduction improvements, such as the replacement or upgrade of a major system component, are still
possible. For example, preliminary data gathered from U.S. industry indicate that leakage rates for certain
types of existing equipment in the United States range from 8 to 40 percent, whereas achievable leakage
rates for new or modified equipment range from 4 to 15 percent. According to the Intergovernmental
Panel on Climate Change/Technology and Economic Assessment Panel (IPCC/TEAP), studies have
reported global annual refrigerant loss from supermarket refrigeration systems to range from 3.2 percent
in the Netherlands to 22 percent in the United States (IPCC/TEAP, 2005). For this same type of
equipment, the International Energy Agency (IEA) estimates that historical leakage rates have been 30
percent or higher, whereas newer systems can achieve leakage rates of approximately 15 percent or
slightly lower (IEA, 2003). Some newer retail food equipment has reached leakage rates of less than 10
percent (Crawford, 2002).
Since the lower-cost leak reduction options represent significant cost savings, this analysis assumes
that the leak reductions occur under the baseline. The cost analysis therefore focused only on the more
extensive and costly options. This option was assumed to be technically applicable17 to all equipment
with large charge sizes (i.e., chillers, retail food refrigeration, cold storage, and industrial process
refrigeration). This analysis assumed that 50 percent of emissions occur as a result of equipment leakage
during routine operation, while the other 50 percent of emissions are released during equipment
servicing and disposal. Thus, the maximum technical applicability of this option was assumed to be 50
percent of emissions from large equipment (see Table 2-9). Furthermore, this analysis assumed that leak
repair can reduce annual system leakage by 40 percent, using an example of a supermarket system that
leaks at 25 percent annually but only at 15 percent following repairs. The project lifetime was assumed to
be 1 year. Regional technical applicability for 2010 and 2020 and reduction efficiency are presented in
Table 2-9. Assumptions on maximum market penetration for each region and year are presented in
Table 2-19.
Refrigerant Recovery and Recycling from Small Equipment
Recovery and recycling of HFCs help to decrease HFC emissions during equipment service and
disposal. The approach involves the use of a refrigerant recovery device that transfers refrigerant into an
external storage container prior to servicing of the equipment. Once the recovery process and source
operations are complete, the refrigerant contained in the storage container may be recharged back into
the equipment, cleaned through the use of recycling devices, sent to a reclamation facility, to be purified,18
or disposed of through the use of incineration technologies. Refrigerant recovery may also be an
16 This option may include modifications that are not strictly leak repair, but would result in greatly reduced leakage
rates. For example, combining the installation of a new purge system, the replacement of flare joints, and other
containment options, or combining the replacement of gaskets and seals, replacement of the motor, and installation
of new refrigerant metering.
17 In this report, the terms "technically applicable" and "technical applicability" refer to the emissions to which an
option can theoretically be applied. The leak repair option was assumed to be technically applicable to all emissions
from leaks (but not servicing and disposal) from the four end-uses listed in Table 2-9.
18 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, and reclamation requires sending the refrigerant off-site to a reclaimer.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-31
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
Table 2-9: Summary of Assumptions for Leak Repair for Large Equipment
Country/Region
United States and Japan
Other Annex 1 countries
Latin America and Caribbean
China, Hong Kong, and India
Other non-Annex 1 countries, Russian
Federation, and Ukraine
Applicable
End-Uses3
Chillers
Retail food
Cold storage
Industrial process
Reduction
Efficiency3
40.0%
Technical Applicability13
2010 2020
25.7% 24.3%
20.3% 19.3%
30.8% 26.7%
16.7% 10.4%
34.2% 27.9%
a End-uses and reduction efficiency apply to all regions.
b Technical applicability is shown as a percentage of total refrigeration- and air-conditioning-sector emissions and equals 50 percent of total
refrigeration and air-conditioning emissions from chillers, retail food refrigeration, cold storage, and industrial process refrigeration. See
Section IV.2.4 for a more complete explanation of how technical applicability, reduction efficiency, and market penetration were used to
calculate emissions reductions associated with each option.
important way to reduce emissions from near-empty refrigerant containers (i.e., can heels). Refrigerant
recovery is assumed to be widely practiced in Annex I countries in the baseline, where the procedure is
typically required by law.
This analysis assesses only the recovery of refrigerant from small equipment (i.e., MVACs,
refrigerated transport, household and other small appliances, and unitary equipment) above that which is
already practiced (e.g., recovery due to regulations in many developed countries or for economic reasons)
at service and disposal. It is assumed that recovery from large equipment is already widely practiced in
the baseline19 because of the significant cost savings associated with recovery of large quantities of
refrigerant from this equipment. Because emissions reductions and costs vary by scenario and end-use,
emissions reductions and costs associated with four recovery scenarios were averaged to obtain one
breakeven cost. The four scenarios studied were recovery and recycling of refrigerant from (1) MVACs at
service, (2) MVACs at disposal, (3) small appliances at service, and (4) small appliances at disposal.
This analysis assumed that 50 percent of emissions are released during equipment servicing and
disposal, while the remaining 50 percent occur as a result of leakage during normal operations. Thus, the
technical applicability20 of this option is 50 percent of emissions from small equipment (see Table 2-10).
Furthermore, because in the United States small appliances are considered completely recovered when 90
percent of the refrigerant is removed from units with running compressors, or when 80 percent of the
refrigerant is removed from units with nonoperating compressors, this analysis assumed that the
reduction efficiency of this option is 85 percent (Contracting Business Interactive, 2003; USEPA, 1993).
The project lifetime is assumed to be 1 year. Regional technical applicability for 2010 and 2020 and
reduction efficiency are presented in Table 2-10. Recovery from small appliances and MVACs was
19 Although the Society of Automotive Engineers (SAE) has issued industry standards on equipment and technician
procedures that apply to MVACs and provide for on-site recovery and recycling of HFC-134a from MVAC systems
for reuse in the serviced system, recovery from these and other small systems is still not believed to be widely
practiced in most developing countries as a result of a lack of infrastructure (i.e., recovery and recycling equipment)
(World Bank, 2002).
20 In this report, the terms "technically applicable" and "technical applicability" refer to the emissions to which an
option can theoretically be applied. The refrigerant recovery and recycling option was assumed to be technically
applicable to all emissions during servicing and disposal (but not leaks) from the five end-uses listed in Table 2-10.
IV-32
GLOBAL MITIGATION OF NON-COj GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
Table 2-10: Summary of Assumptions for Recovery and Recycling from Small Equipment
Country/Region
United States and Japan
Other Annex I countries
Latin America and Caribbean
China, Hong Kong, and India
Other non-Annex I countries, Russian
Federation, and Ukraine
Applicable End-Uses3
MVAC
Refrigerated transport
Household and other small appliances
Commercial unitary air-conditioning
Residential air-conditioning
Reduction
Efficiency3
85.0%
Technical
Applicability13
2010 2020
24.3% 25.7%
29.7% 30.7%
19.2% 23.3%
33.3% 39.6%
15.8% 22.1%
a End-uses and reduction efficiency apply to all regions.
b Technical applicability is shown as a percentage of total refrigeration- and air-conditioning-sector emissions and equals 50 percent of total
refrigeration and air-conditioning emissions from MVACs, refrigerated transport, household and other small appliances, and commercial
unitary and residential air-conditioning.
assumed to be practiced at 80 percent in the baseline in developed countries and at 30 percent in the
baseline in developing countries. Assumptions on maximum market penetration for each region and year
are presented in Table 2-19.
Proper Refrigerant Disposal
One potential source of emissions from the refrigeration and air-conditioning sector is the accidental
or deliberate venting of refrigerant. The venting of refrigerant can be reduced by increasing the
reclamation of used refrigerant (discussed in more detail below) and properly disposing of refrigerant
that cannot be reclaimed (such as highly contaminated refrigerant or mixed refrigerant). Disposal costs
vary by country and region, as do transportation costs, storage costs, and access to refrigerant disposal
facilities (e.g., high-temperature incinerators that handle refrigerants). Global average ODS destruction
costs are estimated to vary between $1.70 and $2.60 per pound (approximately $4 to $6 per kilogram) (ICF
Consulting, 2002b). This option was not explored in the cost analysis as a result of the uncertainty
associated with access to disposal facilities and cost disparities within regions.
Technician Certification and HFC Sales Restriction
By ensuring that refrigeration and air-conditioning technicians receive training in proper refrigerant
handling, including recovery and recycling practices, or by restricting the sale of HFC refrigerants to
certified technicians only, refrigerant emissions can be reduced. In some countries, including the United
States, technicians must be certified in accordance with national regulations to purchase CFC and HCFC
refrigerants and service refrigeration and air-conditioning equipment. Restricting the use of HFC
refrigerants to certified technicians would similarly reduce emissions. To the extent that technician
certification and HFC sales restrictions are practiced today, these actions were included in the baseline;
additional implementation of these practices was not explored in this analysis due to uncertainty in cost
and emissions reductions.
Alternative Refrigerant Options
This section describes four alternative refrigerants: ammonia, hydrocarbons, carbon dioxide, and
other low-GWP refrigerants.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-33
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
Ammonia
Ammonia, primarily used in water-cooled chillers, has excellent thermodynamic properties and can
be used in many types of systems. Because ammonia has a strong odor, refrigerant leaks are easier to
detect, and because ammonia is lighter than air, dispersion is facilitated in the event of a release (UNEP,
1999a). However, ammonia must be used carefully because it is toxic and slightly flammable. Ammonia is
an explosion hazard at 16 to 25 percent in air, which creates a problem in confined spaces. Chillers that
use ammonia as a refrigerant are commercially available in Europe and elsewhere, and they have
efficiencies that are comparable to those of HFC-134a chillers in some instances. Building and fire codes,
however, restrict the use of ammonia in urban areas of the United States and in many other countries.
These safety concerns and institutional barriers effectively limit the potential for expanded use of
ammonia chillers (Sand, Fischer, and Baxter, 1997).
Whereas the use of ammonia within public spaces, such as supermarkets, is limited in some countries
by building codes and ordinances, ammonia is a potential alternative for supermarkets if safety concerns
can be adequately addressed through engineering design such as secondary loops and isolation. Indeed,
modern ammonia systems manufactured in the United States are fully contained, closed-loop systems
with fully integrated controls that regulate pressures throughout the system. Also, all systems are
required to have an emergency diffusion system and a series of safety relief valves to protect the system
and its pressure vessels from overpressurization and possible failure (ASHRAE, 2002). Systems with
ammonia are being built and used in Europe (Sand et al, 1997). However, the further use of ammonia as a
supermarket primary refrigerant may be unlikely in the near future in the United Kingdom and other
countries because of the capital costs and issues of compliance with standards and safety regulations
(Cooper, 1997). Ammonia would also be an option in some industrial process refrigeration and cold
storage applications, contingent upon addressing all of the relevant concerns regarding flammability and
toxicity. For example, ammonia is used in about 80 percent of current installations of large-size
refrigeration plants, as well as in many indirect commercial refrigeration systems (RTOC, 2003).
The chemical properties of ammonia make it incompatible with current designs of light residential
and commercial unitary air-conditioning systems, which use copper for the refrigerant tubing, in the heat
exchangers, and in other components. In the presence of water, ammonia cannot be used with copper or
zinc (UNEP, 1999a); however, ammonia can be used in aluminum and steel systems. Compatible
components would need to be developed to use ammonia. As a result of these technical and cost barriers,
as well as ammonia's flammability and toxicity, ammonia is considered an unlikely candidate for use in
commercial and residential unitary equipment (Sand et al., 1997).
Many of the existing uses of ammonia were included in the baseline analysis. One additional
option—using ammonia secondary loop systems in retail food and cold storage end-uses—is analyzed in
more detail in the section on "Technology Options" that follows this section on alternative refrigerant
options.
HCs
HCs have thermodynamic properties comparable to fluorocarbons that make them good refrigerants;
however, the high flammability of HCs causes safety concerns. Considering technical requirements alone,
there is potential for use of HCs in retail food refrigeration, refrigerated transport, household
refrigeration, residential air-conditioning, MVACs, and commercial unitary systems. Currently used
refrigerants include HC-600a, HC-290, and HC-1270 (UNEP, 1999a). In addition to good thermodynamic
properties, HCs have other advantages such as energy efficiencies comparable to fluorocarbons, zero
ozone depletion potential (ODP), and very low direct GWP.
IV-34 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
The primary disadvantage of HCs is their flammability, resulting in significant safety and liability
issues. These concerns cause increased costs for safety precautions in factories and can necessitate design
changes in every application, such as relocation of electrical components to reduce the likelihood of
accidents from potential leaks (Kruse, 1996; Paul, 1996). These concerns also entail additional hardware
costs for many applications (ADL, 1999; Crawford, 2000). HC refrigerant use is generally restricted by
U.S. safety codes, and with the exception of industrial refrigeration, the USEPA has not listed HCs as
acceptable substitutes to ODS refrigerants (per Section 612 of the Clean Air Act Amendments of 1990).
Even if systems that are designed to use HC refrigerants were listed, liability concerns would remain.
Systems using flammable refrigerants will require additional engineering and testing, development of
standards and service procedures, and training of manufacturing and service technicians before
commercialization.
HC domestic refrigerators have been available in Western Europe since the early 1990s, and have
now fully penetrated some of the new domestic refrigeration markets. HC domestic refrigerators are
available in Argentina, Australia, Brazil, China, Cuba, Germany, India, Indonesia, Japan, and elsewhere.
Similarly, HC refrigerants are available in other products, although little information is readily available
regarding their market success to date (Hydro Cool Online, 2002; Calor Gas Refrigeration Web site, 2004;
CARE Web site, 2004).
In addition, HCs have been used in MVACs for the last several years. Some have estimated that, in
certain parts of Australia, 280,000 vehicles contain HC refrigerants (Greenchill Web site, 2000), although
independent data have not been supplied to confirm this estimate. The use of HC refrigerants in direct
expansion systems not designed for a flammable refrigerant can pose safety concerns and is not
considered acceptable by much of the global MVAC industry. The SAE's Alternate Refrigerant
Cooperative Research Program has demonstrated a secondary loop system using HC refrigerant that
minimizes the possible release of flammable refrigerant into the passenger compartment (Hill and
Atkinson, 2003).
Proponents of HC systems claim that these systems bring numerous benefits, including increased
energy efficiency, lower refrigerant cost, lower capital cost, and less noise (HyChill Web site, 2004;
Greenchill Web site, 2000), but little independent research exists to confirm these claims. In many parts of
the world, however, safety issues, public perception, and manufacturer acceptance impede further
penetration of this option.
This analysis does not consider the use of HCs in household refrigeration because this option was
assumed to reach maximum market penetration in the baseline. In those regions where HCs have not
successfully penetrated markets (e.g., North America), the perceived risk and lack of acceptance of HC
refrigerants, which has prevented adoption to date, was assumed to continue to serve as a barrier in the
foreseeable future. The use of HCs in other refrigeration end-uses was not considered because of
uncertainty about costs and likely market penetration.
CO?
Another option is to use CO2 as a refrigerant. Prototype CO2 systems have been developed for
numerous types of systems, including MVACs, industrial processing, refrigerated transport, and retail
food systems. CO2 has zero OOP and a GWP of 1, and is claimed by its proponents to be advantageous
for use as a refrigerant. However, CO2 is associated with potential safety risks and other technical and
economic disadvantages. Above certain concentrations, exposure to CO2 may result in adverse health
consequences. At very high concentrations, even for short periods of time, CO2 affects the central nervous
system and is toxic. To protect against adverse health effects from workplace exposure, the Occupational
Safety and Health Administration (OSHA) recommended an 8-hour time-weighted average exposure
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-35
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
limit of 5,000 parts per million (ppm) (ACGIH, 1999). Also, CO2 systems operate at a high pressure, which
presents a potential hazard and may increase the cost of designing and purchasing equipment. In
addition, potential loss of operational efficiency and associated increases in energy use and indirect
emissions, refrigerant containment issues, long-term reliability, and compressor performance are other
potential problems (Environment Canada, 1998).
For this analysis, CO2 systems were evaluated only as options for MVACs. CO2 is being investigated
for other end-uses but, because research is still in the early stage and there is little information, those end-
uses were not explored in this analysis. The MVAC option is described in detail in the section on
"Technology Options."
Other Low-GWP Refrigerants
The use of other low-GWP refrigerants (e.g., HFC-152a with a GWP of 140) in place of higher-GWP
refrigerants (e.g., HFC-134a with a GWP of 1,300) is another option for reducing greenhouse gas
emissions. The use of HFC-152a in MVACs was explored in this cost analysis, as described in detail in the
"Technology Options" section.
Several other low-GWP refrigerants exist. For example, CO^ discussed above, has a GWP of 1. In
addition, HCFC-123 and HCFC-124, which are not considered alternatives to HFCs, have low direct
GWPs, but their use is complicated by other factors, including their contribution to stratospheric ozone
depletion. While some studies (e.g., Calm, Wuebbles, and Jain, 1999; Wuebbles and Calm, 1997; USEPA,
2002; RTOC, 2003) suggest that the extended use of HCFC-123 in large tonnage chillers may reduce direct
GWP-weighted refrigerant emissions, and in some instances may reduce overall greenhouse gas
emissions, this option was not examined here because full compliance with the current HCFC phaseout
schedule was assumed.
Technology Options
This section presents cost analyses for six alternative technology options, three of which apply to the
stationary equipment (distributed systems, HFC secondary loop systems, and ammonia secondary loop
systems), and three of which apply to mobile systems (enhanced HFC-134a, HFC-152a, and CO2). Oil-free
compressors, geothermal cooling systems, and desiccant cooling systems are also described qualitatively.
Distributed Systems for Stationary Commercial Refrigeration Equipment
A distributed system consists of multiple compressors that are distributed throughout a store, near
the display cases they serve, and are connected by a water loop to a single cooling unit that is located on
the roof or elsewhere outside the store. Refrigerant charges for distributed systems can be smaller than
the refrigerant charge used in a comparable traditional centralized direct expansion (DX) system.
Significant reductions in total global warming impact from current levels may be possible with
distributed systems that use HFC refrigerants (Sand et al., 1997).
Using HFC-distributed systems in lieu of HFC centralized DX systems in retail food settings offers
the potential to reduce HFC emissions. Distributed systems have smaller refrigeration units distributed
among the refrigerated and frozen food display cases, with each unit sending heat to a central water
cooling system. A distributed system would significantly reduce the refrigerant inventory—by an
estimated 75 percent—and minimize the length of refrigerant tubing and the number of fittings that are
installed in DX systems, thereby reducing HFCs leaks by an estimated 5 percent to 7 percent
(IPCC/TEAP, 2005).
This technology option is assumed to be applicable to the retail food and cold storage end-uses. The
project lifetime is assumed to be 15 years, and the emissions reduction efficiency is calculated to be 90
IV-36 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
percent. Regional technical applicability for 2010 and 2020 and reduction efficiency are presented in
Table 2-11. Assumptions on maximum market penetration for each region and year are presented in
Tables 2-18 and 2-19, expressed as a percentage of emissions from new equipment, and as a percentage of
emissions from all equipment (new and existing), respectively. Because the cost analysis for this option
does not address the costs to retrofit existing DX systems, this option is assumed to penetrate only new
retail food and cold storage installations (i.e., those installed in 2005 or beyond).
Table 2-11: Summary of Assumptions for Distributed Systems for New Stationary Equipment
Country/Region
United States and Japan
Other Annex I countries
Latin America and Caribbean
China, Hong Kong, and India
Other non-Annex I countries, Russian
Federation, and Ukraine
Applicable End-
Use Sector(s)a
Retail food
Cold storage
Reduction
Efficiency8
90.0%
Technical Applicability11
2010 2020
43.1% 40.6%
34.1% 32.1%
51.7% 44.5%
28.0% 17.3%
57.3% 46.6%
a End-uses and reduction efficiency apply to all regions.
b Technical applicability is shown as a percentage of total refrigeration and air-conditioning sector emissions and equals the percentage of total
refrigeration and air-conditioning emissions that are assumed to come from retail food and cold storage end-uses.
Secondary Loop Systems for Stationary Equipment
Secondary loop systems pump cold fluid to remove heat from equipment (e.g., refrigerated food
display cases) or areas to be cooled. The fluid, often a brine solution, passes through a heat exchanger to
be cooled by a refrigerant isolated from the equipment or areas cooled. These systems require a
significantly lower refrigerant charge, have lower leakage rates, and can allow the use of flammable or
toxic refrigerants.
Secondary loops may be used in commercial and industrial refrigeration applications, for example, to
cool supermarket display cases without circulating toxic or flammable refrigerants throughout the store
or to reduce the needed charge of HFC refrigerants. The primary disadvantages of the secondary loop
system are a loss of energy efficiency and higher capital costs. Potential benefits of secondary cooling
systems, however, include decreased charge sizes, decreased leakage rates, faster defrost, lower
maintenance needs, and longer shelf lives, which can result in significant cost savings over time (Bennett,
2000; Baxter, 2003; Faramarzi and Walker, 2003). Indeed, the reduction in size and leakage rate of the
refrigerant charge could result in a reduced global warming impact, even with the use of fluorocarbon
refrigerants. The use of zero-GWP refrigerants could result in even lower global warming impacts (Sand,
et al., 1997). Furthermore, secondary loop systems have improved temperature control compared with
conventional direct expansion systems, which can represent an important advantage in countries like the
United States, where recent regulations on temperature control for refrigerated products such as meat,
poultry, and fish have become more stringent. Moreover, recent technological improvements to
secondary cooling systems, such as high-efficiency evaporative condensers and display cases with high
temperature brines, have increased system efficiency (Baxter, 2003; Faramarzi and Walker, 2003). Two
types of secondary loop systems, for use in retail refrigeration and cold storage warehouses, are analyzed
in greater detail below.
Secondary loops could mitigate some but not all of the risks of using flammable refrigerants in
residential and commercial unitary end-uses. In addition, secondary loops have potential applications in
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-37
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
MVACs, discussed further in "HFC-152a Refrigerant in MVACs." Because of the lack of technical and
cost information on secondary loop systems in these other applications, they are not included as options
in this analysis.
HFC Secondary Loop Systems for Stationary Commercial Refrigeration Equipment
Designing new retail food and cold storage systems to operate using secondary loops with HFCs can
reduce HFC emissions. As discussed above, secondary loop systems circulate a secondary coolant or
brine from the central refrigeration system to the display cases (UNEP, 1999a; ADL, 1999). These systems
have lower leakage rates and operate at reduced charges. Additionally, pipes used in these systems are
now premanufactured and can be made of preinsulated plastic instead of copper. This design reduces
material costs and, by eliminating the need for brazing, allows for faster installation. In the United States,
installation costs have been reduced significantly in recent years. With continued research and
development, this technology is expected to soon be as cost-effective to purchase, install, and operate as
centralized DX systems (Bennett, 2000). This technology option is assumed to be applicable to the retail
food and cold storage end-use sectors, and is expected to reduce charge size by between 75 percent and
85 percent and bring annual leakage rates down to about 5 percent (1PCC/TEAP, 2005)—reducing direct
emissions from appropriate end-uses by approximately 93 percent (see calculation below). The project
lifetime is assumed to be 15 years. The regional technical applicabilities for 2010 and 2020 and the
reduction efficiencies are presented in Table 2-12. Assumptions on maximum market penetration for each
region and year are presented in Tables 2-18 and 2-19. Because the cost analysis for this option does not
address the costs to retrofit existing DX systems, this option is assumed to penetrate only new retail food
and cold storage installations (i.e., those installed in 2005 or beyond).
Table 2-12: Summary of Assumptions for HFC Secondary Loop Systems for New Stationary Equipment
Country/Region
United States and Japan
Other Annex I countries
Latin America and Caribbean
China, Hong Kong, and India
Other non-Annex I countries, Russian
Federation, and Ukraine
Applicable End-
Use Sector(s)a
Retail food
Cold storage
Reduction
Efficiency3
93.33%
Technical Applicability5
2010 2020
43.1% 40.6%
34.1% 32.1%
51.7% 44.5%
28.0% 17.3%
57.3% 46.6%
a End-uses and reduction efficiency apply to all regions.
b Technical applicability is shown as a percentage of total refrigeration and air-conditioning sector emissions and equals the percentage of total
refrigeration and air-conditioning emissions that are assumed to come from equipment in the retail food and cold storage end-uses.
Ammonia Secondary Loop Systems for Stationary Commercial Refrigeration Equipment
The use of ammonia is very common in some countries, while strongly restricted in others. For
example, for many decades ammonia has been used in almost all dairies, breweries, slaughterhouses, and
large freezing plants across Europe, while its use has been heavily regulated in North America (ACHR
News, 2000). Ammonia refrigeration has historically been used in large, low-temperature industrial
refrigeration, as well as in medium and large chillers, generally for food processing (Crawford, 1999).
However, the use of ammonia refrigerant is beginning to expand into retail food and smaller chillers in
some countries, particularly in the EU-15.
Because of ammonia's materials capability, toxicity, and flammability, major design modifications
would be required for the majority of traditional HFC systems. Furthermore, since different countries
IV-38
GLOBAL MITIGATION OF MON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
have different sets of building codes, fire codes, and other safety standards relating to the use of ammonia
in building equipment, some countries (e.g., the United States) would need to revise those codes to allow
for the expanded use of ammonia in new equipment types.
Ammonia can be used as the primary refrigerant in secondary loop systems in place of HFCs.
Because ammonia 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 centralized DX systems
and operate at reduced charges. In these types of systems, ammonia is kept out of public contact (e.g.,
outside of buildings), and nontoxic fluids are used as secondary coolants. Incremental one-time costs for
ammonia systems are assumed to include expenditures for equipment needed to ensure safety. The
annual operating costs also include net energy requirements, but, because of a lack of information, do not
cover costs associated with training technicians and development and updating of safety protocols to
handle more hazardous refrigerants, including ammonia. This technology option is assumed to be
applicable to the retail food and cold storage end-uses. The project lifetime is assumed to be 15 years. The
reduction efficiency of this option is 100 percent, as the ammonia completely replaces the HFC. Because
the cost analysis for this option does not address the costs to retrofit existing DX systems, this option is
assumed to be technically applicable in only new (i.e., those installed in 2005 or beyond) retail food and
cold storage installations.
Table 2-13 presents the reduction efficiency and regional technical applicabilities for 2010 and 2020.
Table 2-13: Summary of Assumptions for Ammonia Secondary Loop Systems for New Stationary Equipment
Country/Region
United States and Japan
Other Annex I countries
Latin America and Caribbean
China, Hong Kong, and India
Other non-Annex I countries, Russian
Federation, and Ukraine
Applicable End-
Use Sector(s)a
Retail food
Cold storage
Reduction
Efficiency9
100.0%
Technical Applicability11
2010 2020
43.1% 40.6%
34.1% 32.1%
51.7% 44.5%
28.0% 17.3%
57.3% 46.6%
a End-uses and reduction efficiency apply to all regions.
b Technical applicability is shown as a percentage of total refrigeration and air-conditioning sector emissions and equals the percentage of total
refrigeration and air-conditioning emissions that are assumed to come from equipment in the retail food and cold storage end-uses.
Ammonia systems are assumed to penetrate a greater percentage of non-U.S. markets as a result of
different safety standards and greater acceptance by industry, end-users, regulators, and insurance
companies in those countries. Assumptions on maximum market penetration for each region and year are
presented in Tables 2-18 and 2-19.
Enhanced HFC-134a Systems in MVACs
Various options exist to reduce emissions of HFC-134a in MVACs by reducing charge size, leakage
rates, or system efficiency (i.e., reducing system power consumption). Specifically, reducing the volume
of the system components, such as the condenser and refrigerant lines, can reduce charge size. Similarly,
leakage rates can be lowered and system efficiency improved by using better system components, such as
improved system sealing, lower permeation hoses, improved fittings, and higher evaporator
temperatures (Lundberg, 2002; Xu and Amin, 2000). Additional savings of indirect emissions can be
obtained by improving system efficiency, for example through the use of oil separators and externally
controlled swashplate compressors.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-39
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
Based on the latest science and industry estimates available when this analysis was performed,
enhanced HFC-134a systems can reduce baseline direct emissions by 50 percent (SAE, 2003a). This
technology is not expected to become commercial until after 2006 (SAE, 2003a). This analysis assumes a
project lifetime (i.e., MVAC lifetime) of 12 years. Regional technical applicabilities and the reduction
efficiency are presented in Table 2-14.
Table 2-14: Summary of Assumptions for Enhanced HFC-134a Systems for New MVACs
Country/Region
United States and Japan
Other Annex I countries
Latin America and Caribbean
China, Hong Kong, and India
Other non-Annex I countries, Russian
Federation, and Ukraine
Applicable End-
Use Sector(s)
MVACs
Reduction
Efficiency8
50.0%
Technical Applicability"
2010 2020
27.6% 19.9%
42.8% 36.6%
13.3% 12.0%
53.0% 65.8%
3.8% 8.0%
a Reduction efficiency applies to all regions and represents the reduction in direct emissions (compared with conventional HFC-134a systems)
as a result of reduced leakage.
b Technical applicability is shown as a percentage of total refrigeration and air-conditioning sector emissions and equals the percentage of total
refrigeration and air-conditioning sector emissions that are assumed to come from MVACs.
Acceptance of this substitute would likely vary by region, based on consumer and industry attitudes,
economic variables, and availability of competing options. Enhanced HFC-134a systems are expected to
become commercially available several years before other alternatives (e.g., CO2 and HFC-152a).
Therefore, this analysis assumes that, initially, enhanced HFC-134a systems will begin to penetrate the
markets of developed countries—with the exception of Europe, which is expected to move away from
HFC-134a use in MVACs in response to new EC legislation.21 In developed countries such as the United
States, Japan, and Canada, where the industry is resistant to switching from HFC-134a and/or regulations
phasing out the use of HFC-134a in MVACs do not exist, this option is assumed to gain the greatest
market penetration. In developing countries, capital cost is expected to prevent this option from
significantly penetrating the market before 2010; however, given the global market, these systems are
expected to gain market share by 2020. The cost analysis for this option does not include any costs
associated with retrofitting existing HFC-134a systems. Therefore, this option is assumed to penetrate
only new MVACs produced after 2004. Assumptions on maximum market penetration for each region
and year are presented in Tables 2-18 and 2-19.
HFC-152a Refrigerant in MVACs
Replacing HFC-134a refrigerant in MVACs with HFC-152a represents a significant opportunity to
reduce GWP-weighted HFC emissions, since the GWP of HFC-152a is 140, 89 percent less than that of
HFC-134a, whose GWP is 1,300. HFC-152a is a flammable refrigerant but is less flammable than HCs.
HFC-152a can be used in DX and secondary loop MVAC systems. Because there is still great uncertainty
associated with the future costs of HFC-152a secondary loop systems for MVACs, this cost analysis only
considers the DX option. Likewise, because there is still great uncertainty associated with future costs of
improved HFC-152a MVACs, only the conventional DX systems are considered in this cost analysis.
However, like the enhanced HFC-134a system discussed above, HFC-152a MVACs will use improved
21 According to the EC Directive, HFC-134a will be phased out from 2011 onward for new vehicle models and from
2017 for all new vehicles. The directive applies to gases with a GWP higher than 150 (EC, 2004).
IV-40
GLOBAL MITIGATION OF NOfJ-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
system components to further reduce refrigerant leakage rates and increase system efficiency (e.g.,
externally controlled variable displacement compressors).
In addition to direct emissions reductions associated with a lower GWP, HFC-152a DX systems in
MVACs also reduce indirect emissions by improving system efficiency by about 10 percent (SAE, 2003a).
This analysis assumes a project lifetime (i.e., MVAC lifetime) of 12 years. Regional technical
applicabilities and the reduction efficiency are presented in Table 2-15.
Table 2-15: Summary of Assumptions for HFC-152a DX Systems in New MVACs
Country/Region
United States and Japan
Other Annex I countries
Latin America and Caribbean
China, Hong Kong, and India
Other non-Annex I countries, Russian
Federation, and Ukraine
Applicable End-
Use Sector(s)
MVACs
Reduction
Efficiency8
89.0%
Technical Applicability13
2010 2020
27.6% 19.9%
42.8% 36.6%
13.3% 12.0%
53.0% 65.8%
3.8% 8.0%
a Reduction efficiency applies to all regions and represents the reduction in direct emissions (compared with conventional HFC-134a systems)
as a result of lower GWP.
b Technical applicability is shown as a percentage of total refrigeration and air-conditioning sector emissions and equals the percentage of total
refrigeration and air-conditioning sector emissions that are assumed to come from MVACs.
The use of HFC-152a DX systems in MVACs would not require any significant changes to existing
HFC-134a system components apart from a safety mitigation system (e.g., a refrigerant detector and a
valve to isolate the remaining charge from the passenger compartment), thereby rendering this option
easy to introduce into the market. Furthermore, compared with baseline HFC-134a systems, HFC-152a
systems are expected to be more efficient and may operate at reduced refrigerant charges and leakage
rates.22 However, because HFC-152a is a slightly flammable gas, safety systems are needed. Thus,
personnel training would be needed to enable the safe and effective recovery and recycling of refrigerant
at service and disposal, and additional safety systems to minimize the potential for large leaks into the
passenger compartment may be required. New fire-safe service equipment for refrigerant recovery and
charging and leak detection may also be required.
While the MVAC industry has demonstrated the use of HFC-152a in prototype DX (and secondary
loop) MVAC systems, the technology is still in the research and development phase. HFC-152a systems
are expected to become commercially available between 2006 and 2008 (SAE, 2003a). Once available, it is
assumed that, initially, HFC-152a systems will gain market share in developed countries, although use in
Europe will be tempered by conditions that may favor CO2 systems. Market penetration in developing
countries is expected to lag by about 5 years. Retrofitting HFC-134a systems to HFC-152a systems is not
considered technically or economically feasible, because it is assumed that additional safety systems to
reduce potential passenger exposure must be incorporated into the system. Thus, costs associated with
retrofit were not assessed, and this option is assumed to penetrate only new MVACs produced after 2004.
Assumptions on maximum market penetration for each region and year are presented in Tables 2-18 and
2-19.
22 Because these systems are still under development, this cost analysis does not consider the possible reduction in
charge and leakage rates, although efficiency improvement predictions based on SAE (2003a) are included.
GLOBAL MITIGATION OF NON-CO, GREENHOUSE GASES
IV-41
-------
SECTION iv — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
CO? in MVACs
Systems that use CO2 as the refrigerant in MVACs represent a potential opportunity for emissions
reduction. This technology uses a transcritical vapor cycle that differs from conventional MVAC systems
and requires innovative design and engineering. The arrangement of components in CO2 systems is
generally consistent with conventional systems; however, a suction line heat exchanger is added and a
low side accumulator is used (in place of a high side receiver, which is used in most conventional HFC-
134a systems). In addition, the individual system components are designed to reflect the extremely high
pressure levels of supercritical CO2 (about 2,000 pounds per square inch [psig]).
Because CO2 has a GWP of \, its use would virtually eliminate the climate impacts of direct
refrigerant emissions from MVACs. CO2 systems perform most efficiently in areas like northern Europe
that require air conditioners for cooling and other purposes, but generally have mild ambient
temperatures.23 In addition, heat pump technology for vehicles is under development (VDA, 2003), which
may allow CO2 systems to be used for supplemental heating of the passenger compartment (SAE, 2003a).
This technology may be an important function in cars with very efficient engines, where minimal waste
heat is available to warm the passenger compartment.
While CO2 has the advantage of being non-flammable, it is toxic. A short exposure to elevated levels
of CO2 can lead to dizziness, drowsiness, and even death (Lambertsen, 1971; Wong, 1992). In addition,
CO2 system operating pressure is 5 to 10 times that of HFC-134a; therefore, appropriate safety features
and new system and component designs are required before this option can be brought to market.
Furthermore, an internal heat exchanger, which would further cool the high-temperature CO2 from the
gas cooler and heat the low-temperature CO2 from the accumulator, would be needed to increase cooling
capacity and energy efficiency to acceptable levels. Also, in the event of a large leak, passengers could be
exposed to potentially dangerous levels of CO2; thus, it is assumed that safety systems designed to
minimize passenger exposure would be incorporated into the system design.
Several engineering constraints must still be overcome, including those associated with flexible lines,
increased system weight, and system leakage and leak detection methods. In addition, because these
systems will be designed and built differently than current MVACs and because the high pressure
presents additional risks, technicians will need to be trained on how to service and maintain these new
systems safely and correctly in order to prevent safety hazards and maintain system performance. New
service equipment for refrigerant charging and leak detection may also be required. Moreover, because of
the high pressure of these systems and toxicity concerns, MVAC servicing and maintenance would need
to be performed by skilled technicians, to prevent safety hazards and maintain system performance.
The efficiency gains associated with CO2 systems are between 20 and 25 percent (SAE, 2003a). In this
cost analysis, 22.5 percent is used for calculation purposes. While there are ongoing efforts to develop
improved CO2 systems for MVACs —which experts predict would exceed this 20 to 25 percent energy
efficiency gain—much uncertainty remains regarding the investment costs required to manufacture these
systems. Therefore, these improved CO2 systems are not considered further in this analysis. The assumed
project lifetime (i.e., MVAC lifetime) is 12 years. Regional technical applicabilities and the reduction
efficiency for the CO2 option are presented in Table 2-16.
23 Compared with other refrigerant technologies, prototype CO2 MVAC systems are not as efficient in warmer
climates. The MVAC industry is actively pursuing research and development activities to improve system efficiency
in warmer weather conditions (SAE, 2003b).
IV-42 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
Table 2-16: Summary of Assumptions for C02 Systems in New MVACs
Country/Region
United States and Japan
Other Annex I countries
Latin America and Caribbean
China, Hong Kong, and India
Other non-Annex I countries, Russian
Federation, and Ukraine
Applicable End-
Use Sector(s)
MVACs
Reduction
Efficiency3
100.0%
Technical Applicability19
2010 2020
27.6% 19.9%
42.8% 36.6%
13.3% 12.0%
53.0% 65.8%
3.8% 8.0%
a Reduction efficiency applies to all regions and represents the reduction in direct emissions (compared with conventional HFC-134a systems).
b Technical applicability is shown as a percentage of total refrigeration and air-conditioning sector emissions and equals the percentage of total
refrigeration and air-conditioning sector emissions that are assumed to come from MVACs.
CO2 systems may be available on the market in the next few years (SAE, 2003a). In light of the new
EC directive on MVACs, and because European manufacturers are most aggressively pursuing CO^, this
option is expected to become the dominant market player in this market. In other developed countries,
such as the United States, Australia, New Zealand, and Canada, the industry is not developing this
technology as aggressively, and it is assumed that this option will not be widely adopted in these markets
in the near future. Finally, because of the high capital costs associated with this option (see details below),
this technology is also not expected to be adopted in developing countries until later years, assuming a
projected global market shift to non-GWP alternatives. The project lifetime is assumed to be 12 years, and
assumptions on maximum market penetration for each region and year are presented in Tables 2-18 and
2-19. Retrofitting HFC-134a systems to CO2 is not considered technically or economically feasible because
of the high operating pressures and because it is assumed that additional safety systems to reduce
potential passenger exposure must be incorporated into the systems. Thus, costs to retrofit were not
assessed, and this option is assumed to penetrate only new MVACs produced after 2004.
Oil-Free Compressors
Oil-free compressors are available for chillers, industrial process applications, and other applications
where compressors are used. The elimination of oil in refrigeration and air-conditioning compressors has
been achieved through various innovative designs, including the incorporation of magnetic or hybrid
ceramic bearings (SKF, 2003; Smithart, 2003). In some systems, oil may decrease heat transfer and reduce
operating efficiency; therefore, removing oil may increase the ability to sustain system efficiency over the
life of the equipment. This reduction will lower indirect emissions of CO2 associated with electricity
production. Eliminating the use of oil in compressors can reduce the number of equipment components
(e.g., oil separators and sealing, fittings, and connections), allowing equipment to be made tighter,
resulting in lower leakage rates. In addition, oil-free compressors remove the need for oil changes and the
associated refrigerant emissions that may be experienced through the service practices used or from
refrigerant dissolved in the oil. However, this potential emissions reduction may be offset by an increased
frequency of compressor and bearing inspection or replacement (Digmanese, 2004), although an
increasing history of operation may prove that unnecessary. This option was not included in the cost
analysis because limited data were available.
Geothermal Cooling Systems
In some locations, geothermal cooling systems for residential and commercial spaces are popular and
economically sound as an alternative to conventional air-conditioning systems. Geothermal technology
transfers heat between the system and the earth and can provide both space heating and cooling. Though
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-43
-------
SECTION IV — INDUSTRIAL PROCESSES « REFRIGERATION AND AIR-CONDITIONING
installation costs for geothermal systems are typically 30 percent to 50 percent higher than for
conventional systems, annual costs are reduced by 20 percent to 40 percent because of increased energy
efficiency. Economic paybacks can accrue in as little as 3 to 5 years. Geothermal systems may save
homeowners 20 percent to 50 percent in cooling costs (Geoexchange, 2000; Rawlings, 2000). Because of a
lack of cost and market penetration data, this technology is not considered further in this analysis.
Desiccant Cooling Systems
Desiccant cooling is produced by removing moisture from an air stream using a desiccant and then
separately cooling the dry air. The desiccant is thermally regenerated, typically by burning natural gas or
by capturing excess heat. Desiccant cooling may replace the latent cooling done by some end-uses, such
as unitary systems. Integrated desiccant cooling systems that combine a desiccant system with a vapor
compression or other cooling system have been successfully installed in some commercial buildings
(Fisher, Tomlinson, and Hughes, 1994). However, current designs are used primarily in niche markets
that require precisely controlled humidity or low humidity levels, such as hospital operating rooms and
certain industrial processes. For desiccant-based systems to be considered widely feasible in the
commercial air-conditioning market, improvements in efficiency, cost, size, reliability, and life expectancy
must be made (Sand et al., 1997).
Desiccant systems have also been tried in MVAC systems, but were found technically and
economically infeasible. These systems require an intermittent source of heat; however, because new
automobiles produce very little waste heat, there is not enough heat for a desiccant system to function.
Desiccant systems may only be feasible where there is a large heat source, such as a large truck or bus
(Environment Canada, 1998). Furthermore, in order for desiccant air-conditioners to become viable
options for MVACs, the varying heat source must be controlled during normal driving conditions when
vehicle speed is continually changing. Current prototypes are large and heavy, and the systems have not
been shown to be cost-effective or durable enough to justify the initial investment (USEPA, 2001 a).
Because of the technical barriers and insufficient cost information associated with the feasibility of
this option, desiccant cooling systems were not explored further in this analysis.
Absorption Systems
Absorption systems refrigerate or cool using two fluids and some quantity of heat input, rather than
using electrical input. Specifically, absorption systems use a secondary fluid or absorbent to circulate the
refrigerant (Rafferty, 2003). These systems can be used in residential refrigeration and chiller applications
and, potentially, in heat pumps in residential and light commercial applications, as described below.
• Refrigeration Systems. In the late 1990s, more than 1 million of an estimated 62 million
refrigerators sold annually were thermally activated ammonia or water absorption systems (Sand
et al., 1997). The refrigerants used for absorption refrigeration have negligible GWPs. Absorption
refrigeration is commonly used in hotel rooms and for recreational vehicles because the process
operates quietly and can use bottled gas for energy. Absorption refrigerators are limited in size
because of design constraints. Through design improvements, the thermal coefficient of
performance (COP) of these refrigerators can be increased by as much as 50 percent from a COP
of 0.2 to 0.3 without degrading cooling capacity (Sand et al., 1997). However, the low efficiency of
absorption equipment means that the indirect emissions must be carefully analyzed. Inherent
design limitations make it unlikely that absorption refrigeration will become a significant
replacement for vapor compression refrigerators. Still, absorption refrigeration has great capacity
and operating attributes that permit the technology to fill niche markets (Sand et al., 1997).
iv-44 GLOBAL MITIGATION OF Ncw-co2 GREENHOUSE GASES
-------
SECTION IV— INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
• Chillers. Gas-fired (as opposed to electrically powered) absorption water chillers are sold in the
United States and Japan. These systems are used primarily where there is a relatively short
cooling season, where electricity costs (especially demand charges) are high, or where fairly high-
grade waste heat is available. Although absorption chillers are far less efficient than competitive
systems if waste heat is unavailable, the technology is feasible and, under some economic
circumstances, compares favorably with vapor compression chillers using fluorocarbon
refrigerants. Market success will be determined by factors such as the relative costs of natural gas
and electricity, peak load charges, and purchase costs. In addition, absorption chillers currently
have higher capital costs than vapor compression equipment, such that significant operating cost
savings would be necessary to make their purchase economically competitive.
• Heat Pumps. Research and development efforts are attempting to create absorption heat pumps
for heating and cooling in residential and light commercial applications. Several years ago in
Europe and the United States, generator absorber heat exchange (GAX) ammonia-water
absorption heat pumps were being developed and in Japan field test units had been built.
Absorption heat pumps could be used to reduce global warming impacts in areas where heating
load dominates, although the pumps would have the opposite effect in areas where cooling
dominates (Sand et al., 1997).
Because these options are either still under development or are primarily optimal in niche markets,
sufficient information was not available to include their costs and reduction potential in this analysis.
IV.2.3.2 Summary of Technical Applicability, Market Penetration, and Costs of
Abatement Options
Table 2-19 summarizes the percentage of total refrigeration and air-conditioning sector emissions that
may be technically abated by each of the options explored in this analysis, based on the percentage of
sector emissions from each end-use (which varies by region), as provided in Table 2-6. Market
penetration values for each abatement option were developed for each region, when possible, to best
reflect qualitative information available on region-specific realities and possible future action. The
commercial refrigeration and MVAC technology options explored in this chapter are assumed to
penetrate only new (not existing) equipment, where new equipment is defined as equipment
manufactured in 2005 or later. Table 2-18 presents the assumed maximum market penetration for the
technology options into equipment manufactured in 2005, 2010, 2015, and 2020. Table 2-19 presents the
final maximum penetration into the installed base of equipment, taking into account the percentage of
each market that is new (i.e., manufactured in 2005 or beyond) in all preceding years. Values from
Table 2-19 are multiplied by technical applicabilities (Table 2-17) and the reduction efficiency to generate
the percentage reduction off baseline emissions for each option, as presented in Table 2-20. The text box
provided in Section IV. 2.4provides further explanation on how the results (i.e., percentage reduction off
baseline emissions) are calculated.
IV.2.4 Results
Emissions reduction potential for abatement options varies by region based on assumed end-use
breakouts (provided in Table 2-6) and on qualitative information regarding current and future likelihood
of market penetration by region. The percentage reduction from the baseline associated with each
abatement option is calculated by multiplying the technical applicability (from Table 2-17) by both the
incremental maximum market penetration (from Table 2-18) and the reduction efficiency. For more
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-45
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
information on how emissions reductions are calculated for each option, please see the text box below,
which presents an illustrative example of the emissions reduction methodology.
Calculating Emissions Reductions for Each Abatement Option
The equation used to derive total emissions reductions off the baseline for each option is as follows:
Emissions Reduction = technical applicability x incremental maximum market penetration
(expressed as percentage of entire installed base) * reduction efficiency
The following table provides a sample calculation using the option of leak repair for large equipment
in the United States in 2020 as an example.
Sample Calculation of Emissions Reductions: Leak Repair for Large Equipment— United States (2020
Applicable End-
Uses
(Table 2-9)
Chillers
Retail food
Cold storage
Industrial
process
Total
Technical
Applicability3
(Based on Tables
2-6 and 2-9)
1.5x50%
39.1x50%
1.4x50%
6.6 x 50%
48.7 x 50%
Incremental
Maximum Market
Penetration
(Table 2-19)
5%
5%
5%
5%
x 5% x
Reduction
Efficiency
(Table 2-9)
40%
40%
40%
40%
40%
Percentage
Reduction from
2020 Baseline
(Table 2-20)
0.02
0.39
0.01
0.07
0.49"
a For each country/region, technical applicability varies based on the percentage ot sector emissions from applicable end-uses, as
provided in Table 2-6. Additionally, for the leak repair and refrigerant recovery and recycling options, only half of Ihe emissions from
applicable end-uses (i.e., large end-uses for leak repair and small end-uses for recovery and recycling) are assumed to be abatable;
for all other options, 100 percent of emissions from new (post-2004) equipment in applicable end-uses are assumed to be abatable.
b Total may not sum due to independent rounding.
Table 2-21 presents a summary of the cost assumptions used for the refrigeration/air-conditioning
options presented in the discussions above.
IV.2.4.1 Data Tables and Graphs
Tables 2-22 and 2-23 provide a summary of the potential emissions reductions at various breakeven
costs by country/region in 2010 and 2020, respectively. The costs to reduce 1 tCO2eq are presented at a 10
percent discount rate and 40 percent tax rate. Table 2-24 presents the potential emissions reduction
opportunities and associated annualized costs for the \vorld in 2020 ordered by increasing costs per
tCO2eq, using the highest cost in the region. Because many of the options analyzed affect indirect (CO2
from energy generation) emissions, the net (HFC + CO2) emissions reduced by each option are presented.
The direct (HFC) emissions reduced by the option and a cumulative total of direct emissions reduced, in
MtCO2eq and percentage of the regional refrigeration and air-conditioning baseline, are also presented.
Figures 2-2 and 2-3 present MACs for this sector at 10 percent discount rates and 40 percent tax rates in
2010 and 2020, respectively.
IV-46
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
c
o
co
Q.
_O
CO
DC
>.
CO
o
"a.
"c
CO
&
CO
•£
»*—
o
^^
~
!Q
CO
.0
~a.
Q.
TO
_O
"c
1
,^__
o
^^
CO
E
CO
CNI
CO
&
- c
o 2,
5 CO
OJ^
J,s
a
"^S
o
C
i
-9
^j
CO
^5
«
e
0202
9L02
0102
9002
0202
9L02
0102
9002
0202
910Z
0102
9002
0202
91-02
01-02
9002
9002
91-02
01-02
9002
o
S.
O
i
^_
CM
O
0>
00
ui
o
CO
CO
CM
CO
CM
CM
Oi
JS^
1-
^
8
CO
^
CM
S
cq
CO
m
CO
CO
8
rv
s'l-
1 Refrigerant re<
from small e
CD
CM
O
CO
CM
cS
C3
^
£
CO
CM
CO
8
^—
C\4
CO
CO
CD
^.
?M
CO
CJ
CM
CD
CD
O
in
CM
^.
CD
O5
CM
CO
ct
^>
in
CM
i^
in
CM
CD
L_
ILeak repair fo
equipment
cq
^
CM
in
cq
CO
S
in
5
S
in
CO
CO
in
^_
co1
m
^_
S
CO
CO
CO
CO
*~
o
o
CM
o
CO
CM
CO
CD
CD
^_
5
CM
CO
"D
CO
ft
CO
£
^_
CM
in
CO
En
CO
o
CO
in
5
5
f-;
un
CO
CO
in
^
CM
CO
in
£
^_
c^
CO
S
CO
'"
CD
8
o
CO
CM
CO
CO
CO
,_
CO
CM
0)
«
1 Distributed syi
CO
CO
.,_
c\i
m
CO
r-I
in
CO
§
in
5
i
in
CO
CO
m
^
c^
in
CO
^r-
CO
CO
CO
CO
CO
I--!
O
O
CM
CO
CO
CM
CO
CO
CO
^_
5
CM
§
HFC secondai
system
0
CO
,-J.
in
CO
CO
CO
CO
o
CM
CO
c\i
CO
CO
CM
^
CO
CD
CO
cq
CO
cq
o
3
CO
in
CD
o
S
o
CO
in
CO
o>
cri
CO
si
CO
EM
CD
C
O
Enhanced HF
MVACs
O
CO
^J.
in
oo
CO
oo
CO
o
CM
CO
CM
CO
CO
CM
'"
cq
cq
CO
CO
en
£
CO
in
CO
o
CM
CO
o
CO
in
CO
en
oi
CO
CM
CM
cq
CM
en
in
CO
CO
|
CO
S
1
o
CO
^_
iri
oo
CO
CO
CO
o
CM
CO
CM
CO
CO
CM
'I~
CD
CO
CO
CO
i
CO
in
CO
CD
S
o
CO
in
cq
en
oi
CO
§M
CO
EM
e»
c?
CO
1
d
i?
c/i
Q
CO
O)
O)
c:
c:
o
f—
8
"co
"O
CO
o
^
CD
1
'o
Ol
CO
1
CD
Q.
a Expressed as a
GLOBAL MITIGATION OF NON-C08 GREENHOUSE GASES
IV-47
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
0
8,
•i
0>
CO
CO
CO
"S
CO
CO
£
CL
£
c
E
Q.
LU
1
3
"o
CO
c
co
^
3
d>
1
CO
_o
Q.
o
c
O)
to
<
•5
_o
1 1
c E
CU Q.
if
O or
CO S
^™ 7*
g E
•2 v?
cu co
cc c
Assumed
Emissio
CO
CVI
CD
SI
l_
c ,_
08 •« £
8 S = E g °>
fit ^" *" ^? ^ »^
E "* "S C •*••
< § « * "1 |
» is if — z °
j '{5 JJ
O =>
d>
c
o
en .2
o "c
X ~
.1
0
35
CU
i 2
<•«.
•
ii
So
<
08
.JS-o
11
CO ™
< M
CO
~*
•«
i"
LU
*H
«
55
o>
c
0202
9L02
OL02
9002
0202
9L02
OL02
9002
0202
91-02
OL02
9002
0202
9L02
OL02
9002
0202
9L02
0102
9002
0202
9102
OU)2
9002
CO
g
'•s.
Practice 0|
Q
•z
Q
Q
Q
Z.
O
Q
^y
E
o
|l
£ Q.
§ §"
O> CO
LT "
Q
•z
Q
O
Q
Q
Q
-^
CU
•2 *-
Leak repair
equipmen
CO
o
B.
o
"o
i
CD
'"
0
^~
0
in
0
o
0
in
in
CO
o
in
in
CO
o
in
in
CO
"~
CD
in
CM
CO
m
CO
Q.
§
CO
•o
8
CD
Ammonia s
in
CM
0
CVJ
in
CO
in
CM
°
in
co
0
8
o
CM
O
O
8
o
CM
O
o
o
CO
CM
CD
8
CO
CM
in
CO
t
CO
Distributed
0
CM
CO
1~
a
CO
o
CO
CD
CO
m
CM
oo
1 —
CD
in
in
CM
CO
CD
m
m
CM
co
"~
CD
in
o
CM
in
T
CO
§-
_o
CO
HFC secon
system
0
"*
CD
CM
in
CD
0
CM
in
o
o
co
o
in
o
o
CD
CO
o
in
CD
o
o
CD
CD
CD
g
8
CD
CD
CO
-r
Enhanced 1
MVACs
CD
CM
.1_
O
O
o
CM
T —
O
CD
O
CO
O
CM
••-
CD
8
o
CM
••-
O
in
CM
m
"~
-
CD
8
CD
CM
,—
O
8
tr
HFC-152ai
m
1
0
o
in
•-
o
0
0
m
'-
0
o
m
••-
o
S
LO
CO
LO
CD
O
in
T—
O
CO
O
$
_c
C/)
c:
o
Cfj
'CD
c:
CD
Q.
"QJ
-s
CO
o>
E
CD
o
"c
0
to
o
0)
N
£1
"d
E
=3
CO
™
o
-Q.
O
o5
§.
s
"CO
"o
c:
o
I
'CD
c:
CD
ID
CO
E
CD
c:
"CD
en
S
cu
£
CO
s.
c:
"i5i
<
-^ CD
CO
'i s
CO CD
.
£= CD
CO C
E o
"-"i
o .2 ^
"O CO
LU CO LU Q
IV-48
GLOBAL MITIGATION OF NOM-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
fa
O
'co
CO
'E
LLJ
-
0
u
CO
C/J
^_
*s
^~
Q)
"c
^
*
CD
CO
to
I
(O
fl>
X
LLJ
co"
g
^^
O
r*
CD
0)
ta
o
.g
'CD
"55
a>
'cu
co
o>
01
CD
ja
C of ...
08 .2 e • S
Latin America i
Caribbean, Russ
Federation, Ukrai
& All Other Not
Annex 1 Countri
08
en
c
o
o>.2
c. -a
o c
i •—
of
.^
O
1 CO
^C "*••
fc_ ?•
0> 3
15
^y
OB
.£^•0
(O ®
< N
C d)
8.2
CO
-3
0
o>
Q
01
§
to
53
•o
'c
0202
9L02
0102
9002
0202
91-02
01-02
S002
0202
9102
0102
S002
0202
SW2
01-02
9002
0202
9102
OL02
•
S002
0202
91.02
01-02
9003
CO
c
_o
OL
o
CD
O
CD
8
CO
CD
CM
O
CO
m
CO
"*
CO
8
CO
O
CM
CO
in
CO
CD
o
CD
o
iri
CD
in
0
CD
0
CD
"
o
in
CO
iri
CD
CD
1 —
CO
CD
T —
o
in
CO
0
CD
1 —
o
CD
o
iri
Refrigerant recovery
from small
equipment1"
o
V—
co
CM
O
O
O
iri
CO
iri
•*—
CD
""
O
O
o
in
0
in
0
iri
CO
CO
CO
CO
CO
iri
CO
in
CO
CO
co
o
iri
CO
in
0
CO
CO
co
CD
iri
co
in
CO
CO
CD
CO
ILeak repair for large
equipment
Technology
Options
m
co
co
iri
cq
CM
co
CD
in
06
CO
iri
CO
CM
CO
o
CO
CD
CO
co
CM
CO
o
CO
o
cd
CO
CM
CO
CO
CO
o
cd
CO
CM
CO
co
^J-
oi
CO
2
CM
o
1 Ammonia secondary
loop
.1
CO
1 —
oi
in
CD
co
^~
oi
, —
^
m
CD
in
CO
CM
CM
CO
in
co
co
in
CO
CM
CM
CO
in
CO
CD
in
CO
CM
CD
CM
co
iri
CO
CD
en
CD
O)
5
in
CD
| Distributed system
,_:
T —
co
CO
CM
CO
in
CD
h-
^~
CO
cb
CM
CO
in
CD
in
CO
CD
CO
CM
CO
CD
in
CO
CD
CO
cvi
CO
CD
in
CO
CD
co
CM
CO
CO
en
^
o>
5
in
o
HFC secondary loop
system
en
O5
i —
co
o
co
O5
^~
^^
lv^
CO
CO
CD
in
£
CM
en
CM
co
o
o
o
in
CO
CM
S3
o
CD
T
co
co
CO
co
co
co
CO
CD
CD
CD
in
CM
o>
CM
CD
CD
O
CD
CO
Ij
co I5£
•c 5^
tu '"
in
CO
CD
eo
co
co
CD
iri
CO
CD
o
CD
o
CD
CM
CO
iri
CO
CO
CO
co
CM
CO
iri
CO
o
o
CD
co
CM
CM
CO
CD
CD
CD
CM
CO
^
iri
CO
CO
CO
o
& CO
to O
O ^
u_ ^
h~
CO
CD
CD
o
CD
CD
r--
co
CD
CD
C3
O
CD
en
^
CO
CD
O
CD
O5
^J
CO
CD
CD
CO
CO
CO
in
in
CM
CM
cq
CO
ca
o
CO
^
.,_•
CO
CD
O
CD
|
51
o
•wise noted.
>ing countries.
O) O
4^ (U
O >
« ^
c/3 *•*
01 t=
c
=3 "c
0 8
CD 3
Sri °
CD in z
cn S
CO O LLJ
_°2 O O)
-g 2 a>
^ CO "§
CJ CO TT
c: CD y
3 Total sector emissions i
b Shown percentage valu
c Europe is assumed to ir
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-49
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
Table 2-20: Percentage of (Direct)3 Reduction Off Baseline Emissions of All Abatement Options by Region
Practice Options
Refrigerant recovery
from small
equipment
Leak repair for large
equipment
Technology Options
Ammonia secondary
loop
Distributed system
HFC secondary loop
system
Enhanced HFC-134a
in MVACs
HFC-152a in MVACs
C02 in MVACs
Practice Options
Refrigerant recovery
from small
equipment
Leak repair for large
equipment
Technology Options
Ammonia secondary
loop
Distributed system
HFC secondary loop
system
Enhanced HFC-134a
in MVACs
HFC-152a in MVACs
C02 in MVACs
United States
in o in o
§f- T- CM
o o o
CM CM CM CM
1.1 2.1 2.1 3.3
0.3 0.3 0.5 0.5
0.1 0.6 1.8 3.8
0.2 1.6 3.8 6.5
0.2 1.6 3.9 6.8
0.0 1.4 3.3 4.8
0.0 0.1 1.1 2.9
0.0 0.1 0.4 1.0
All Other Annex 1
Countries
memo
ST- T- CM
o o o
CM CM CM CM
1.3 2.5 2.4 3.9
02 02 0.4 0.4
01 09 23 3.3
0.2 1.6 4.4 6.8
0.1 0.8 2.5 4.1
0.0 2.1 4.6 8.9
0.0 0.1 1.5 5.3
0.0 0.1 0.5 1.8
Europe13
in o m o
ST- T- CM
o o o
CM CM CM CM
1.3 2.5 2.4 3.9
0.2 0.2 0.4 0.4
0.1 0.9 2.3 3.3
0.2 1.6 4.4 6.8
0.1 0.8 2.5 4.1
0.0 0.0 0.0 0.0
0.0 0.1 1.2 4.2
0.0 1.6 7.2 18.5
China, Hong Kong, &
India
m o m o
§T- •«- CM
o o o
CM CM CM CM
4.8 8.5 12.8 16.8
04 0.7 0.6 06
01 0.7 1.2 15
0.2 1.0 1.8 2.5
0.2 0.8 1.3 1.9
0.0 0.3 2.2 6.6
0.0 0.0 0.1 3.2
0.0 0.0 0.2 1.1
Japan
in o in o
0 T- T- CM
CM CM 8 CM
1.1 2.1 2.1 3.3
0.3 0.3 0.5 0.5
0.1 1.1 2.6 4.2
0.2 2.1 5.0 8.6
0.1 1.1 28 51
0.0 1.4 3.3 4.8
0.0 0.1 1.1 2.9
0.0 0.1 0.4 1.0
Latin America &
Caribbean
m o m o
O T-
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
Table 2-21: Summary of Abatement Option Cost Assumptions (2000$)
Option
Refrigerant recovery
Distributed system
Secondary loop
Ammonia secondary loop
.
Leak repair
C02 for new MVACs
Enhanced HFC-134a in
MVACs
HFC-152a in MVACs
Time
Horizon
(Years)
1
15
15
15
1
12
12
12
Unit of Costs
Per recovery job
Per 60,000 ft2
supermarket
Per 60,000 ft2
supermarket
Per 60,000 ft2
supermarket
Per repair job
Per MVAC
Per MVAC
Per MVAC
U.S. One-
Time Cost
a
$7,200.00
$25,200.00
$36,000.00
$1,480.00°
$105.30
$42.12
$23.69
U.S.
Annual
Cost
$10.10
$2,796.19"
$5,592.38"
$5,592.38"
—
—
—
—
U.S.
Annual
Savings
$13.71
$3,559.94
$3,691.79
$3,955.49
$2,636.99
$18.35"
$21.38d
$7.92"
Net U.S.
Annual
Costs
-$3.61
-$763.75
$1,900.59
$1,636.89
-
$2,636.99
-$18.35
-$21.38
-$7.92
a The cost of a high-pressure recovery unit is assumed to be approximately $860, but all costs associated with this option, including capital
costs, are annualized and expressed in terms of cost per job.
b In all other countries, this annual cost was adjusted by average electricity prices (average of 1994-1999) based on USEIA (2000).
c Includes parts and labor to perform repair job.
d Annual U.S. costs savings are associated with gasoline and refrigerant savings. For all other countries, the annual saving associated with
gasoline in the United States is adjusted by the estimated amount of gasoline saved per vehicle per year (based on Rugh and Hovland
[2003]) and by average regional costs of unleaded gasoline in 2003 (based on USEIA [2005]). No adjustments are made to the savings
associated with refrigerant.
e Annual U.S. costs savings are associated with gasoline savings. For all other countries, this annual savings is adjusted by the estimated
amount of gasoline saved per vehicle per year (based on Rugh and Hovland [2003]) and by average regional costs of unleaded gasoline in
2003 (based on USEIA [2005]).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-51
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
Table 2-22: Country/Regional Emissions Reductions in 2010 and Breakeven Costs for Refrigeration/Air-
Conditioning at 10% Discount Rate, 40% Tax Rate ($/tC02eq)
2010
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China & Hong Kong
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.69
9.08
0.09
0.42
2.63
0.15
1.08
0.24
1.22
0.40
0.62
9.86
0.52
0.79
5.67
16.60
$15
1.04
17.51
0.20
0.75
3.03
0.27
2.25
0.29
1.91
0.72
0.90
19.32
0.74
1.57
11.44
29.20
$30
1.37
18.63
0.27
0.76
3.12
0.27
2.36
0.31
2.63
0.73
0.91
20.44
0.75
1.57
11.44
31.03
$45
1.37
18.63
0.27
0.76
3.12
0.27
2.36
0.31
2.63
0.73
0.91
20.44
0.75
1.57
11.44
31.03
$60
1.37 -
19.34
0.27
0.76
3.12
0.34
2.97
0.31
2.63
0.73
0.94
21.12
0.75
1.57
11.44
31.73
>$60
1.37
19.38
0.27
0.76
3.12 '
0.34
2.97
0.31
2.65
0.73
0.94
21.16
0.75
1.57
11.44
31.77
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Table 2-23: Country/Regional Emissions Reductions in 2020 and Breakeven Costs for Refrigeration/Air-
Conditioning at 10% Discount Rate, 40% Tax Rate ($/tCQ2eq)
2020
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China & Hong Kong
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
2.26
43.63
0.24
1.38
12.33
0.81
4.95
0.94
3.87
1.29
2.89
45.69
2.39
3.11
30.26
73.22
$15
4.06
109.62
1.03
3.19
14.41
1.66
12.48
1.18
9.03
2.99
4.49
117.04
3.60
7.56
78.05
161.70
$30
5.73
117.89
1.81
3.41
20.41
1.66
13-.22
1.79
13.22
3.19
4.74
125.65
3.76
7.89
78.05
181.11
$45
5.73
117.89
1.81
3.41
20.41
1.66
13.22
1.79
13.22
3.19
4.74
125.65
3.76
7.89
78.05
181.11
$60
5.73
130.65
1.81
3.41
20.41
2.96
24.03
1.79
13.22
3.19
5.25
137.90
3.76
7.89
78.05
193.94
>S60
5.76
131.50
1.91
3.43
21.09
2.96
24.03
1.85
13.66
3.22
5.28
138.79
3.78
7.93
78.05
195.80
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
IV-52
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
Table 2-24: World Breakeven Costs and Emissions Reductions in 2020 for Refrigeration/Air-Conditioning
Cost (2000$/tC02eq)
DR=10%, TR=40%
Reduction Option
Leak repair
Refrigerant recovery
Distributed system
Enhanced HFC-134a in
MVACs
HFC-152ainMVACs
Ammonia secondary
loop
HFC secondary loop
C02 for new MVACs
Low
-$4.10
-$2.62
-$1.08
-$175.92
-$27.59
$6.33
$4.81
$7.57
High
-$4.10
-$2.62
$9.99
$16.21
$18.18
$26.40
$26.70
$91.60
Direct
Emissions
Reduction3
(MtC02eq)
4.91
40.16
39.67
22.69
15.72
22.18
33.20
17.26
Indirect
Emissions
Reduction"
(MtC02eq)
0.00
0.00
-0.43
21.67
0.81
-2.71
-0.06
1.83
Reduction
from 2020
Baseline
(%)
0.8%
6.4%
6.3%
3.6%
2.5%
3.5%
5.3%
2.8%
Running
Sum of
Reductions
(MtC02eq)
4.91
45.07
84.74
107.44
123.16
145.34
178.54
195.80
Cumulative
Reduction
from 2020
Baseline
(%)
0.8%
7.2%
13.5%
17.1%
19.6%
23.2%
28.5%
31.2%
a Direct reductions refer to HFC emissions reductions (off the baseline).
b Indirect emissions impacts are those associated with energy consumption (not included in the baseline).
Figure 2-2: 2010 MAC for Refrigeration/Air-Conditioning, 10% Discount Rate, 40% Tax Rate
$100 -
$80 -
« $60 -
•| $40 -
•§ CT $20 -
g) 0>
w U
§ §. -$20 ^
1 -$40 -
m -$60 -
-$80 -
-$100 -I
r_ _>
r^——^i.
LJ 2
_-f
J
f
J
s y Japan
« Other OECD
EU-15 = European Union; OECD = The Organisation for Economic Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-53
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
Figure 2-3: 2020 MAC for Refrigeration/Air-Conditioning, 1 0% Discount Rate, 40% Tax Rate
$100 -i
$80-
« $60-
1 $40 -
II $2°-
DC O en
« 0 $°
§ § -$20 4
8 ~ -$40 -
ill -$60 -
-$80-
-$100 J
[~^d
,,—•"'
^ — •
f
LJ 10 ?J)
M
< i
, I
30 40 50 60
Cumulative Emissions Reductions (MtCO2eq)
70 80 90
— * China
Rest of the world
— • United States
••••-* EU-15
— *• Japan
— « Other OECD
EU-15 = European Union; OECD = The Organisation for Economic Co-operation and Development.
IV.2.4.2 Uncertainties and Limitations
This section focuses on the uncertainties and limitations of the cost estimates presented in this
analysis. One significant area of uncertainty is how capital costs for these mitigation technologies may
vary internationally. The analysis is currently limited by the lack of this specificity on region-specific cost
analysis estimates. In addition, the main uncertainties related to the following abatement options are
listed below.
Leak Repair for Large Equipment
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
emissions 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.
Refrigerant Recovery for Small Equipment
Estimates of the amount of refrigerant recoverable from MVACs and small appliances at service and
disposal are highly uncertain. This analysis uses the estimates provided in USEPA (1998).
Stationary Technology Options (Distributed, HFC Secondary Loop, and Ammonia Secondary
Loop Systems)
This analysis assumes that emissions savings equal to 56 percent of the original equipment charge are
realized at disposal in the distributed and HFC and ammonia secondary loop options; however, the
actual amount of charge emitted at disposal is uncertain.
IV.2.5 Summary
Baseline HFC emissions from refrigeration and air-conditioning are expected to grow significantly
between 2005 and 2020, as HFCs become used increasingly throughout the world to replace gases phased
IV-54
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
out under the Montreal Protocol. The highest percentage of emissions growth is expected to occur in
developing countries.
This analysis considers the costs and emissions reduction potential of eight practice and technology
emissions mitigation options: (1) leak repair for large equipment, (2) refrigerant recovery and recycling
from small equipment, (3) distributed system, (4) HFC secondary loop, (5) ammonia secondary loop,
(6) enhanced HFC-134a systems in MVACs, (7) HFC-152a systems in MVACs, and (8) CO2 systems in
MVACs. The costs and emissions reduction benefits of each option were compared for each region.
Increasing leak repair of large equipment and refrigerant recovery/recycling from small equipment
represent cost-effective options for reducing emissions from stationary equipment worldwide. For
MVACs, the enhanced HFC-134a option represents the most cost-effective alternative for reducing
emissions.
IV.2.6 References
American Conference of Governmental Industrial Hygienists, Inc. (ACGIH). 1999. Guide to Occupational
Exposure Values. Cincinnati, OH.
Arthur D. Little, Inc. (ADL). 1999. 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 49648.
American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE). 2002. Ammonia
as a Refrigerant: Position Document. Approved by ASHRAE Board of Directors 17 January. Available at
.
Atkinson, W. 2000. Review comments on draft report, U.S. High GWP Gas Emissions 1990-2010: Inventories,
Projections, and Opportunities for Reductions [Refrigeration and Air-Conditioning Chapter]. Sun Test
Engineering.
Barbusse, S., D. Clodic, and J.P. Roumegoux. October 1998. Mobile Air Conditioning; Measurement and
Simulation of Energy and Fuel Consumptions. Presented at the Earth Technologies Forum. The
Alliance for Responsible Atmospheric Policy.
Baxter, Van D. 2003. IEA Annex 26: Advanced Supermarket Refrigeration/Heat Recovery Systems. Final Report
Volume 1—Executive Summary. Based on information developed in Canada, Denmark, Sweden,
United Kingdom, United States (Operating Agent). Oak Ridge National Laboratory.
Bennett, C. 2000. Personal communication between C. Bennett, Senior Vice President of Althoff
Industries, Inc., and ICF Consulting. December 14, 2000.
Calm, J. 1999. Emissions and Environmental Impacts from Air-Conditioning and Refrigeration Systems.
Joint IPCC/TEAP Expert Meeting on Options for the Limitation of Emissions of HFCs and PFCs.
Calm, J.M., D.J. Wuebbles, and A.K. Jain. 1999. Impacts on Global Ozone and Climate from Use and
Emission of 2,2-Dichloro-l,l,l-trifluoroethane (HCFC-123). Journal of Climate Change 42, 439-474.
Calor Gas Refrigeration. 2004. Care Refrigerants Technical Information. Available at . As obtained on June 7, 2004.
CARE (BOC Refrigerants) 2004. CAREing for our world. Available at . As obtained on March 1, 2004.
China Association of Automobile Manufacturers. 2005. Workshop on Technology Cooperation for Next
Generation Mobile Air Conditioning, 3-4 March 2005, New Delhi, India.
Contracting Business Interactive. 2003. Refrigerant Recovery in Residential Systems. Available at
. As obtained on July 7,
2003.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-55
-------
SECTION IV — INDUSTRIAL PROCESSES . REFRIGERATION AND AIR-CONDITIONING
Cooper, P.J. 1997. Experience with Secondary Loop Refrigeration Systems in European Supermarkets.
Proceedings of the International Conference on Ozone Protection Technologies, pg. 511. The Alliance
for Responsible Atmospheric Policy. November.
Crawford, J. May 1999. Limiting the HFC Emissions of Chillers. Joint IPCC/TEAP Expert Meeting on
Options for the Limitation of Emissions of HFCs and PFCs held in the Netherlands.
Crawford, J. March 2000. Review comments on the draft report, U.S. High GWP Gas Emissions 1990-2010:
Inventories, Projections, and Opportunities for Reductions [Refrigeration and Air-Conditioning Chapter]. The
Trane Company.
Crawford, J. 2002. Refrigerant Options for Air Conditioning. Presented at the Earth Technologies Forum.
The Alliance for Responsible Atmospheric Policy. March 26, 2002.
Digmanese, T. 2004. Peer review comments on the USEPA Draft Report, DRAFT Analysis of International
Costs of Abating HFC Emissions from Refrigeration and Air-Conditioning. York International Corporation.
March 19, 2004.
European Commission (EC). 2003. How to Considerably Reduce Greenhouse Gas Emissions Due to Mobile Air
Conditioners Consultation paper from the European Commission Directorate-General Environment.
February 4, 2003.
EC (European Commission). 2004. Climate Change: Commission Welcomes Political Agreement In The Council
To Reduce Emissions Of Fluorinated Greenhouse Gases. Press release issued on October 14, 2004.
Available at . As obtained on January 16, 2004.
Environment Canada. 1998. Powering GHG Reductions Through Technology Advancement, pp. 185-188.
Environment Canada, Clean Technology Advancement Division.
Faramarzi, R. and D. Walker. 2003. Field Evaluation of Secondary Loop Refrigeration for Supermarkets.
Presented at the 2003 ASHRAE Winter Meeting in Chicago, IL on January 26, 2003. The American
Society of Heating, Refrigerating, and Air-Conditioning Engineers, Inc.
Fisher, S.K., J.J. Tomlinson, and P.J. Hughes. 1994. Energy and Global Warming Impacts of Not-in-Kind and
Next Generation CFC and HCFC Alternatives. Prepared for the Alternative Fluorocarbons
Environmental Acceptability Study and U.S. Department of Energy. Oak Ridge National Laboratory.
Gaslok. 2002. Gaslok Flyer. Submitted electronically to IC'F Consulting by David Peall. Gaslok. Available
at .
Geoexchange. 2000. Information on geothermal heat pumps. Available at .
Greenchill. March 18, 2000. Fire and Ice. Sydney Morning Herald. Available at .
Hydro Cool Online. 2002. Cool Technologies: Working Without HFCs. Updated June 2002. Available at
.
ICF Consulting. 2002a. Analysis on Combined Global Emission Estimates Scenarios. Deliverable submitted by
ICF Consulting to the USEPA that included a revised analysis of the estimated level of recycling in
other countries. Delivered to Casey Delhotal, Dave Godwin, and Debbie Ottinger of the USEPA
Office of Atmospheric Programs.
ICF Consulting. 2002b. ODS Destruction Report. Revised draft report submitted to Julius Banks of the
USEPA Global Programs Division.
IV-56 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
International Energy Agency (IEA). 2003. "IEA Annex 26: Advanced Supermarket Refrigeration/Heat
Recovery Systems, Final Report Volume 1—Executive Summary." Compiled by Van D. Baxter, Oak
Ridge National Laboratory.
Intergovernmental Panel on Climate Change/Technical and Economic Assessment Panel (IPCC/TEAP).
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 TEAP.
Cambridge University Press.
Japan Times. March 19, 2002. Hydrocarbon Fridges Hit Environment-Savvy Japan.
Kruse, H. 1996. The State of the Art of Hydrocarbon Technology in Household Refrigeration. Proceedings
of the International Conference on Ozone Protection Technologies, pp. 179-188. The Alliance for
Responsible Atmospheric Policy.
Kuijpers, L. March 28, 2002. Refrigeration Sector Update. Presented at the 19th Meeting of the Ozone
Operations Resource Group (OORG), The World Bank.
Lambertsen, C. J. 1971. Therapeutic Gases: Oxygen, Carbon Dioxide, and Helium. In J.R. DiPalma (Ed.),
Drill's Pharmacology in Medicine. New York: McGraw-Hill.
Lundberg, E. July 9-11, 2002. An Enhanced R-134a Climate System. Presented at the 2002 SAE
Automotive Alternative Refrigerant Systems Symposium in Scottsdale, AZ. Society of Automotive
Engineers.
OPROZ (Oficina Programa Ozono [Ozone Program Office]). February 2001. Report on the Supply and
Consumption ofCFCs and Alternatives in Argentina.
Paul, J. October 1996. A Fresh Look at Hydrocarbon Refrigeration: Experience and Outlook. Proceedings
of the International Conference on Ozone Protection Technologies, pp. 252-259. The Alliance for
Responsible Atmospheric Policy.
Rafferty, K.D. 2003. Absorption Refrigeration. Geo-Heat Center, Bulletin Vol. 19, No. I. Available at
.
Rawlings, P. 2000. Personal communication between P. Rawlings of the Geothermal Heat Pump
Consortium and ICF Consulting. December 8, 2000.
Refrigeration Technical Options Committee (RTOC). 2003. 2002 Report of the Refrigeration, Air Conditioning
and Heat Pumps Technical Options Committee: 2002 Assessment. Section 8.4.2.7.
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.
Sand, J.R., S.K. Fischer, and V.D. Baxter. 1997. Energy and Global Warming Impacts of HFC Refrigerants and
Emerging Technologies. Prepared for the Alternative Fluorocarbons Environmental Acceptability Study
and U.S. Department of Energy. Oak Ridge National Laboratory.
SKF. 2003. Hybrid bearings in oil-free air conditioning and refrigeration compressors. Evolution. SKF's
business and technology magazine. Available at .
Smithart, G. October 17, 2003. Peer review comments on the USEPA Draft Report, DRAFT Analysis of
International Costs of Abating HFC Emissions from Refrigeration and Air-Conditioning. Turbocor Inc.
Society of Automotive Engineers (SAE). July 14, 2003a. Alternative Refrigerants Assessment Workshop.
Presented at the 2003 Conference on Mobile Air Conditioning Technologies in Phoenix, AZ.
Society of Automotive Engineers (SAE). July 15, 2003b. SAE Alternate Refrigerant Cooperative Research
Project: Project Overview. Slide presentation given by Ward Atkinson at the 2003 Conference on
Mobile Air Conditioning Technologies in Phoenix, AZ.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-57
-------
SECTION IV — INDUSTRIAL PROCESSES • REFRIGERATION AND AIR-CONDITIONING
Society of Indian Automobile Manufacturers (SIAM). March 3-4, 2005. Growth and Projects of Automobile
Industry and Mobile Air Conditioning, Workshop on Technology Cooperation for Next-Generation
Mobile Air Conditioning (MAC). Presentation by Dilip Chenoy, Director General, SIAM, New Delhi,
India.
United Nations Environment Programme (UNEP). 1998. 1998 Report of the Technology and Economic
Assessment Panel (Pursuant to Article 6 of the Montreal Protocol).
United Nations Environment Programme (UNEP). October 1999a. Report of the TEAP HFC and PFC Task
Force.
United Nations Environment Programme (UNEP). October 1999b. Production and Consumption of Ozone
Depleting Substances 1986-1998.
U.S. Energy Information Administration (USEIA). 2000. Annual Energy Outlook 2000 (Electricity Prices for
Industry). Available at . As obtained on
April 2, 2002.
U.S. Energy Information Administration (USEIA). 2005. Annual Energy Review 2004 (Table 8.10 on Average
Retail Prices of Electricity, 1960-2004, and Table 11.8 on Retail Motor Gasoline Prices in Selected Countries,
1990-2004), Report No. DOE/EIA-0384(2004), August 2005. Available at . As obtained on September 6, 2005.
U.S. Environmental Protection Agency (USEPA). 1993. Protection of Stratospheric Ozone; Refrigerant
Recycling, Final Rule. Federal Register citation 58 FR 28660. USEPA. 14 May 1993. Available at
.
U.S. Environmental Protection Agency (USEPA). 1997. Options for Reducing Refrigerant Emissions from
Supermarket Systems. EPA-600/R-97-039. Prepared by Eugene F. Troy of ICF Consulting for USEPA.
U.S. Environmental Protection Agency (USEPA). 1998. Draft Regulatory Impact Analysis: The Substitutes
Recycling Rule. Prepared by ICF Incorporated for USEPA.
U.S. Environmental Protection Agency (USEPA). 2001a. U.S. High GWP Gas Emissions 1990-2010:
Inventories, Projections, and Opportunities for Reductions. EPA #000-F-97-000. U.S. Environmental
Protection Agency, Office of Air and Radiation.
U.S. Environmental Protection Agency (USEPA). 2002. Building Owners Save Money, Save the Earth: Replace
Your CFC Air Conditioning Chiller. EPA M30-F-02-026. U.S. Environmental Protection Agency, Global
Programs Division and Climate Protection Partnerships Division.
U.S. Environmental Protection Agency (USEPA). March 2, 2006. Mobile Air Conditioning Climate Protection
Partnership. Available at . Accessed on June 15, 2006.
VDA (Verband der Automobilindustrie [Association of the Automotive Industry]) 2003. Various
presentations at the Alternative Refrigerant Winter Meeting: Automotive Air Conditioning and Heat
Pump Systems in Saalfelden, Austria. 13-14 February. VDA, Frankfurt, Germany. Available at
.
Ward's World Motor Vehicle Data. 2001. ISBN Number 0-910589-79-8. Southfielclbv, MI.
Wong, K.L. 1992. Carbon Dioxide. 1992. Internal Report, Johnson Space Center Toxicology Group, National
Aeronautics and Space Administration, Houston, TX.
World Bank. 2002. CFC Markets in Latin America. Latin America and Caribbean Region Sustainable
Development Working Paper No. 14. Prepared by ICF Consulting for the World Bank.
Wuebbles, D.J., and J.M. Calm. 1997. An Environmental Rationale for Retention of Endangered
Chemicals. Science. 278:1090-1091.
Xu, J., and J. Amin. 2000. Development of Improved R134a Refrigerant System. Presented at the 2000
Society of Automotive Engineers Automotive Alternative Refrigerant Systems Symposium in
Scottsdale, AZ.
IV-58 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV— INDUSTRIAL PROCESSES » SOLVENTS
IV.3 HFC, HFE, and RFC Emissions from Solvents
IV.3.1 Introduction
DSs have been used as solvents in a wide range of cleaning applications, including precision,
electronics, and metal cleaning (UNEP, 1999a). CFCs (in particular CFC-113), methyl
chloroform (1,1,1-trichloroethane or TCA), and to a lesser extent, carbon tetrachloride (CC14),
were historically used as solvents in the United States. Similar usage occurred elsewhere, except in India
and China, where greater volumes of CC14 were consumed.
To comply with the requirements of the Montreal Protocol,1 many countries started using HCFCs,
and aqueous and semiaqueous not-in-kind (NIK) solvents, as substitutes for ODSs. For example, the
majority of metal cleaning end-users and some of the electronics and precision cleaning solvent end-users
have already transitioned to no-clean, semiaqueous cleaning, and aqueous cleaning alternative methods.
Many of the in-kind replacement solvents, including HFCs and PFCs, have also taken a share of the
substitute market because they have high reliability, excellent compatibility, good stability, low toxicity,
and selective solvency. These HFCs and PFCs have 100-year GWPs ranging from 890 to 7,4002 and
relatively low boiling points (50°C to 90°C) that contribute to their inadvertent release to the atmosphere.
The replacement solvent technologies used globally are summarized in Table 3-1.
HFC solvents include HFC-4310mee, HFC-365mfc, and HFC-245fa. Of these HFCs, HFC-4310mee is
the most common solvent cleaner replacement. HFC-365mfc is used as an additive to form solvent blends
with HFC-4310mee, helping to reduce the cost of these products (Micro Care, 2002). HFC-245fa is used in
the aerosol solvent industry (Honeywell, 2003). Heptafluorocyclopentane is another HFC that could be
used, although it is not yet used in significant amounts. Certain solvent applications, particularly
precision cleaning end-uses, will continue to use HCFCs, especially HCFC-225ca/cb (until the HCFC
phaseout takes place), and to a much lesser extent, PFCs and perfluoropolyethers (PFPEs).
This report analyzes three solvent end-uses: metal, precision, and electronics cleaning. Metal cleaning
involves the removal of contaminants such as oils, greases, and particulate matter from metal surfaces
during the production of metal parts and the maintenance and repair of equipment and machinery.
Electronics cleaning, or defluxing, consists mainly of removing flux residue that remains after a soldering
operation for printed circuit boards and other contamination-sensitive electronics applications. Precision
cleaning may apply to either electronic components or to metal surfaces and is characterized by products
that require a high level of cleanliness and generally have complex shapes, small clearances, and other
cleaning challenges (UNEP, 1999a). Examples of applications and products requiring precision cleaning
include disk drives, gyroscopes, medical devices, and optical components. Based on current
understanding of market trends, HFC emissions from the precision and electronics cleaning end-uses
dominate the GWP-weighted emissions from the solvents sector. The metal cleaning
1 Parties to the Montreal Protocol on Substances that Deplete the Ozone Layer (Montreal Protocol) agreed to phase out
consumption of all ODSs, including those used as solvents. In developed countries, the solvent industry has phased
out its use of Class I ODSs (in particular CFCs and 1,1,1-trichloroethane). Developing countries are scheduled to
phase out these substances between 2008 and 2010.
2 7,400 is the GWP of perfluorohexane (C6F14), and is used in this report for estimating purposes as the GWP for
PFC/PFPEs.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-59
-------
SECTION IV — INDUSTRIAL PROCESSES • SOLVENTS
Table 3-1: General Overview of Solvent Technologies Used Globally
Solvent Classes
Chlorinated solvents
HCFC solvents (HCFC-225 ca/cb and HCFC-1415)
HFC solvents (primarily HFC-4310mee)
RFC solvents
Hydrofluoroether (HFE) solvents
Hydrocarbons
Alcohol solvents
Brominated solvents
Methyl siloxanes
Metal
X
X
X
X
X
Electronics
X
X
X
X
X
X
X
X
X
Precision
X
X
X
X
X
X
X
X
X
Alternative Cleaning Technologies
Aqueous cleaning
Semiaqueous cleaning
No-clean processes
X
X
X
X
X
xa
X
X
a For electronics cleaning, no-clean processes include low-solids flux or paste and inert gas soldering.
end-use has primarily transitioned away from ODSs directly into alternatives or processes that do not use
high-GWP chemicals.
IV.3.2 Baseline Emissions Estimates
IV.3.2.1 Emissions Estimating Methodology
Description of Methodology
Specific information on how the model calculates solvent emissions is described below.
The USEPA uses a detailed Vintaging Model of ODS-containing equipment and products to estimate
the use and emissions of various ODS substitutes in the United States, including HFCs and PFCs.
Emissions baselines from non-U.S. countries were derived using country-specific ODS consumption
estimates as reported under the M ontreal Protocol, in conjunction with Vintaging Model output
for each ODS-consuming end-use sector. For sectors where detailed information was available, these data
were incorporated into country-specific versions of the Vintaging Model to customize emission estimates.
In the absence of country-level data, these preliminary estimates were calculated by assuming that the
transition from ODSs to HFCs and other substitutes follows the same general substitution patterns
internationally as observed in the United States. From this preliminary assumption, emissions estimates
were then tailored to individual countries or regions by applying adjustment factors to U.S. substitution
scenarios, based on relative differences in (1) economic growth, (2) rates of ODS phaseout, and (3) the
distribution of ODS use across end-uses in each region or country.
Emissions Equations
Generally, the emissions model assumes that some portion of used solvent remains in the liquid
phase and is not emitted as gas. Thus, emissions are considered incomplete and are set as a fraction of the
amount of solvent consumed in a year. For solvent applications, a fixed percentage of the new chemical
used in equipment is assumed to be emitted in that year, with the remainder of the used solvent reused
IV-60 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • SOLVENTS
or disposed of without being released to the atmosphere. The following equation calculates emissions
from solvent applications:
ErL*QCj (3.1)
where
EJ = Total emissions of a specific chemical in a given year j from use in solvent applications, by
weight.
L = The percentage of the total chemical that is lost to the atmosphere, assumed to be 90 percent.
QCJ = Total quantity of a specific chemical sold for use in solvent applications in the given year j, by
weight.
j = Year of emissions.
Many solvent users have added emissions control features to their equipment, resulting in lower
solvent consumption. Eventually, almost all of the solvent consumed in a given year is emitted, because
the solvent is continuously reused through a distilling and cleaning process or through recycling, while a
small amount of solvent is disposed with the sludge that remains. The model used for this analysis
assumes that 90 percent of the solvent consumed annually is emitted to the atmosphere.
Regional Variations and Adjustments
The following adjustment factor assumptions, specific to the solvent sector, were used to customize
the global emissions estimating methodology, described above, for solvents:
• PFC/PFPE solvents were assumed to be used in countries with significant annual output from the
electronics industry. Global PFC usage for solvent cleaning was geographically distributed using
the semiconductor industry as a proxy; specifically, data on the share of world silicon wafer starts
per month (8-inch equivalent) (SEMI International, 2003) were used. PFC/PFPE solvent use was
assumed to be discontinued by 2010 in the United States and by 2015 in other countries.
• Emissions in EU-15 countries were assumed to equal only 80 percent of the preliminary estimate
to reflect that NIK technology has taken a more significant market share in European countries
(ECCP, 2001). Consequently, the resulting EU emissions estimate was reduced by 20 percent.
• A 50-percent adjustment factor was applied to countries with CEITs, European countries that are
not members of the EU-15, and developing (non-Annex I) countries. For these countries, the
primary barriers to the transition from ODS solvents to fluorinated solvents has been the high
cost of HFC-4310mee and the lack of domestic production (UNEP, 1999a; UNEP, 1999b).
IV.3.2.2 Baseline Emissions
Table 3-2 presents total HFC, PFC, and HFE emissions estimates in MtCO2eq for the solvent sector. In
the United States, HFC-4310mee is responsible for the majority of the country's projected ODS substitute
solvent emissions, whereas PFC/PFPE emissions are assumed to decline linearly until they are
discontinued completely in 2010. U.S. emissions reflect the continued decline of PFC/PFPE consumption
as a result of restrictions enforced by the USEPA's Significant New Alternatives Policy Program, which
limits PFC and PFPE use to only those applications where these solvents have been deemed necessary to
meet performance or safety requirements. U.S. solvent end-users that have historically used PFC/PFPEs
are turning to other solvents, including HFC-4310mee.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-61
-------
SECTION IV — INDUSTRIAL PROCESSES • SOLVENTS
Table 3-2: Total Baseline HFC, PFC, and HFE Emissions Estimates from Solvents (MtC02eq)
Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex I
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
0.0
10.0
0.2
0.0
4.0
0.1
3.7
0.0
3.5
0.0
0.0
11.6
0.0
2.3
2.4
16.4
2010
0.0
5.4
0.1
0.1 .
1.4
0.0
2.1
0.0
1.4
0.0
0.0
5.9
0.0
0.8
1.7
7.7
2020
. 0.0
4.1
0.1
0.1
0.1
0.0
0.9
0.0
0.9
0.0
0.0
4.1
0.0
0.2
2.0
4.5
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development
Similarly, some PFC use for precision and electronics cleaning in countries outside the United States
was assumed to decline linearly until use is discontinued completely in 2015. Global PFC use for solvent
cleaning, as provided by industry expert opinion, was apportioned to non-U.S. countries using the global
distribution of the semiconductor market as a proxy for circuit board cleaning, a predominant electronics
cleaning end-use (3M Performance Materials, 2004; DuPont FluoroProducts, 2004; SEMI International,
2003).
Figure 3-1 displays total HFC, PFC, and HFE emission estimates for the solvent sector by region from
1990 to 2020.
IV.3.3 Cost of HFC, HFE, and PFC Emissions Reductions for Solvents
This section presents a cost analysis for achieving HFC, HFE, and PFC emissions reductions from the
emissions baselines presented in Table 3-2 above. All cost analyses for the solvent emissions reduction
options assume a 10-year project lifetime. Each abatement option is described below.
IV.3.3.1 Description and Cost Analysis of Abatement Options
Some HFC, HFE, and PFC emissions from the solvent sector can be eliminated or mitigated through
several technologies and practices. Emissions and use of these compounds can be reduced by retrofitting
equipment and improving containment of the solvents, introducing carbon adsorption technologies, and
replacing outdated equipment with more modern technologies. Additionally, NIK technologies and
processes already used in many solvent markets worldwide employ semiaqueous, aqueous, or no-clean
methods in place of solvents. Ongoing research continues to identify low-GWP alternatives, including
low-GWP HFCs and HFEs that could replace high-GWP PFCs and HFCs. Some alternative solvent
cleaning approaches use other organic solvents, including chlorinated solvents, alcohols, petroleum
distillates, and aliphatic solvents.
IV-62 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • SOLVENTS
Figure 3-1: Total Baseline HFC, PFC, and HFE Emissions Estimates from Solvents (MtC02eq)
D Middle East
• Africa
• Non-EU FSU
• Latin America
• S&E Asia
D China/CPA
• OECD90+
1990
2000
2010
2020
Year
CPA = Centrally Planned Asia; Non-EU FSU = non-European Union Former Soviet Union; OECD = Organisation for Economic Co-operation
and Development.
Flammable organic solvent alternatives, such as ketones, ethers, and alcohols, can also potentially
replace MFCs, HFEs, and PFCs. Because these alternatives are fairly aggressive and would have different
materials compatibility issues than the fluorinated solvents, and because limited technical and cost
information is available on fire suppression equipment, explosion-proof wiring, and other workplace
controls, these additional alternatives are not addressed further in this analysis.
Three potential mitigation options are identified and analyzed in this report:
• conversion to HFE solvents,
• improved equipment and cleaning processes using existing solvents (retrofit), and
• aqueous and semiaqueous NIK replacement alternatives.
The remainder of Section IV.3.3 describes each of these options in detail and provides a discussion of
associated cost and emissions reduction estimates. A detailed description of the cost and emissions
reduction analysis for each option can be found in the Appendix G for this chapter.
Conversion to HFE Solvents
HFC and PFC solvents can be replaced by alternative organic solvents with lower GWPs, which are
making headway in the market. These alternative solvents include low-GWP HFCs and HFEs,
hydrocarbons, alcohols, volatile methyl siloxanes, brominated solvents, and non-ODS chlorinated
solvents. For the purpose of this analysis, commercially available HFE-7100 and HFE-7200 are used as
proxies for the alternative solvent abatement option because they display material compatibility
properties similar to HFCs and PFCs, a prime factor that has led to their current success in the market.
Specifically, HFEs have replaced PFCs, CFC-113, 1,1,1-trichloroethane, HFCs, and HCFCs in certain
precision cleaning operations. Many solvent users have successfully transitioned from PFC solvents to
HFC-4310mee and HFEs in cleaning applications such as computer disk lubrication, particulate cleaning,
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-63
-------
SECTION IV — INDUSTRIAL PROCESSES • SOLVENTS
and cleaning of electronic assemblies after soldering. HFEs and azeotropes of HFEs are also viable
replacements for HFC-4310mee in certain precision and electronics cleaning operations.
Because PFCs are specific to a small portion of the global solvent market, and because they are likely
to be more expensive than HFCs, costs for this analysis are calculated based on a transition from HFC-
4310mee to HFEs, rather than from PFCs to HFEs. Additionally, many users are switching from PFCs to
HFC-4310mee. Since this transition is assumed to occur in the baseline, the transition is not quantified as
an option for further reductions. Therefore, PFC solvent users that switch directly to HFEs may
experience a cost savings compared with HFC solvent users switching to HFEs.
For the purpose of this analysis, the 100-year GWP of alternative solvents reflects the market presence
of two HFEs. HFE-7100, which has a GWP of 390, is assumed to represent 75 percent of the market, and
HFE-7200, which has a GWP of 55, is assumed to represent the remaining 25 percent. The GWP of the
solvent being replaced, HFC-4310mee, is 1,300.3 Because of the lower average GWP, this option has a
reduction efficiency of 76.4 percent (i.e., the difference of the GWP of HFC-4310mee and the weighted
average of the HFE GWPs, divided by the GWP of HFC-4310mee). This analysis assumes that the
technical applicability4 of this option is 81 percent of total solvent emissions for each region in 2005,
dropping to 79 percent from 2010 through 2020 (Table 3-4).
HFE solvents are gaining acceptance in U.S. industry because of their availability, safety, and
effectiveness (Salerno, 2001); however, some uncertainty exists regarding the likelihood and ease with
which HFC-4310mee users will convert to an HFE-alternative solvent because of application-specific
requirements (UNEP, 1999b). The incremental maximum market penetration of this option in the United
States is assumed to increase from 10 percent in 2005 to 60 percent in 2020, as shown in Table 3-4.
For all other countries, the incremental maximum market penetration is assumed to increase from 5
percent in 2005 to 25 percent in 2020, representing a slower adoption of this option and less reliance on
the use of fluorinated compounds compared with the assumed scenario for the United States (see
Table 3-4). This assumption is based on current market data, which indicates that HFE solvents are
available and being used in the same regions where HFC solvents are being used (3M Performance
Materials, 2003).
Improved Equipment and Cleaning Processes Using Existing Solvents (Retrofit)
HFCs, HFEs, and PFCs are more expensive than historically used solvents such as CFC-113 and
HCFC-141b. Attempts to reduce emissions, and hence save costs, have led to significant improvements in
degreasing, defluxing, and other cleaning equipment containment technologies. 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, HFEs, and PFCs used in solvent cleaning (UNEP, 1999a; ICF Consulting, 1992). For example, some
cleaning equipment that uses HFC solvents is being retrofitted with higher freeboard height and low-
temperature secondary cooling coils. It is also possible to keep emissions at a minimum by vising good
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. The GWPs of HFEs are still being studied; for instance, some analyses show
the GWP of HFE-7100 to be approximately 300 (3M Performance Materials, 2003).
4 In this report, the term "technically applicable" refers to the emissions to which an option can theoretically be
applied. Because HFEs can be substituted for HFCs and PFCs, HFEs are technically applicable to all HFC and PFC
solvent emissions, but they are not technically applicable to HFE baseline emissions. Other factors will affect the
application of HFCs and PFCs, and the market penetration assumed in this analysis.
IV-64 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV— INDUSTRIAL PROCESSES • SOLVENTS
handling practices, such as reducing systems' solvent drag-out losses by keeping the workload in the
vapor zone long enough to drain and dry any entrapped or remaining solvent (UNEP, 1999a; Petroferm,
2000). One can also minimize evaporative losses by improving the design of solvent bath enclosures and
vapor recovery condensing systems (March Consulting Group, 1998 and 1999).
As shown in Table 3-3, retrofitting a vapor degreaser with an open-top area of 13 square feet,
combined with proper operation and maintenance, can reduce solvent emissions by as much as 46 to 70
percent, 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 percent, while installing heating coils to produce superheated vapor along with installing
a chiller can reduce emissions by 70 percent. For the purpose of this analysis, the reduction efficiency of
the retrofit option is assumed to equal 70 percent, which can be achieved at a one-time cost of $16,800 (see
Table 3-3).
Table 3-3: Retrofit Techniques for Batch Vapor Cleaning Machine (Less than 13 Square Feet)
Retrofit Technique
Freeboard ratio of 1 .0, freeboard refrigeration device
Working mode cover, freeboard refrigeration device
Superheated vapor, freeboard refrigeration device
Reduction Efficiency (%)
46.0%
64.0%
70.0%
One-Time Cost (2000$)
$11,200
$15,800
$16,800
Source: Durkee, 1997.
Retrofits to vapor degreasing machines larger than 13 square feet cost more but can achieve emissions
reduction efficiencies as high as 85 percent. Furthermore, for larger operations where there is more than
one vapor degreaser, retrofit methods, such as installing a carbon adsorber, can be implemented to
capture solvent vapor from the air for the entire facility. The reduction efficiency of a carbon adsorber
combined with the installation of heating coils and chillers has been estimated at 88 percent for larger
(i.e., greater than 13 square feet) vapor degreasers (Durkee, 1997).
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
(Durkee, 1997). Consequently, end-users in the United States are not expected to benefit from this option
in the future. Thus, this analysis assumes that the incremental maximum market penetration will drop
from 5 percent in 2005, to zero in 2010 through 2020 (i.e., by 2010 and beyond, the solvent equipment in
use will either already be retrofitted or will not require retrofitting, and the resulting lower emissions are
already incorporated into the baseline). The resulting maximum market penetrations are shown in
Table 3-4.
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. Therefore, for non-U.S. Annex I countries, the maximum market penetration for this
option is also assumed to be 5 percent in 2005, dropping to zero by 2010.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-65
-------
SECTION IV — INDUSTRIAL PROCESSES « SOLVENTS
This analysis assumes that most solvent users in non-Annex I (i.e., developing) countries may
consider the equipment retrofit option, because updating their equipment may be preferred over
investing in entirely new units. Consequently, this region is assumed to adopt these techniques slowly,
such that 5 percent of the market will have adopted this option by 2005. Adoption is assumed to increase
at a slow, steady rate to 15 percent in 2020 (see Table 3-4).
Aqueous and Semiaqueous NIK Replacement Alternatives
In addition to the emissions reduction approaches that use a combination of improved equipment
and cleaning practices, NIK technology processes and solvent replacements can be used to substitute for
PFC-, HFC-, and HFE-containing systems. 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. In the
semiaqueous 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 (UNEP, 1999a). The reduction efficiency of NIK abatement options is assumed to be 100
percent because the HFC or PFC solvent is completely replaced by water and an organic solvent,
combinations of which have low to no CWP.
Many electronics, metal, and precision cleaning end-users have already switched to aqueous and
semiaqueous NIK cleaning methods. Both NIK processes have proven very successful for large-scale
metal cleaning, where equipment and wastewater treatment costs are of less concern because of the large
volumes processed (UNEP, 1999a). Aqueous cleaning technologies have been available and widely used
for over 25 years and have replaced many electronics cleaning solvent systems in developed countries
(Chaneski, 1997; UNEP, 1999a). Semiaqueous cleaning has also been available for years but has lost much
of its initial promise in many developed nations for the cleaning of electronic assemblies because of the
additional complexity and subsequent expense associated with the cleaning process, which includes more
steps than aqueous cleaning (UNEP, 1999a).
Because the NIK options are applicable to both the electronic and precision cleaning end-uses, the
NIK options are assumed to be applicable to 100 percent of high-GWP solvent emissions, resulting in a
technical applicability of 100 percent for all regions (see Table 3-4). The assumed market penetration,
however, is lower, as explained below.
Technical limitations of NIK technologies arising from issues such as substrate corrosion or
inadequate performance for applications with complex parts can lead to reduced market acceptability.
The U.S. incremental maximum market penetrations for these options are assumed to be smaller than in
other regions, to reflect the belief that the U.S. market will likely prefer fluorinated solvents such as HFCs
and HFEs (see Table 3-4). The market penetrations are also assumed to be smaller because most
operations that can use aqueous and semiaqueous technologies are doing so already. For non-U.S. Annex
I and non-Annex I regions, the maximum market penetrations of these two NIK options are assumed to
be similar to each other from 2005 to 2020. NIK alternatives are currently gaining market share in
European countries, a trend that is assumed to continue for this region (ECCP, 2001).
Some developing countries are also assumed to prefer NIK technologies because of their perceived
low costs. Aqueous cleaning is popular in China, for example, because of the small cost per kilogram of
the nonfluorinated cleaning chemicals used, despite newly introduced costs such as wastewater
treatment. Conversely, the availability of water, the costs associated with energy to dry the product, and
local wastewater treatment regulations can discourage companies in developing regions of the world
from considering this option (UNEP, 2003). For all regions, the semiaqueous option is assumed to have
slightly smaller market penetrations than the aqueous cleaning option.
IV-66 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • SOLVENTS
IV.3.3.2 Summary of Technical Applicability, Market Penetration, and Costs of
Abatement Options
Table 3-4 summarizes the technical applicability and the maximum market penetration of the solvent
options presented in the discussions above. By 2020, it is assumed that the NIK replacement option can be
applied to 15 percent of the baseline solvent emissions in the United States, 30 percent of the baseline
solvent emissions in the Annex I countries, and 30 percent of the baseline solvent emissions in non-Annex
I countries. By 2020, the retrofit option is assumed to be viable only in non-Annex I countries. In addition,
the conversion to HFE solvents option can be applied to the baseline HFC and PFC emissions, as shown
below.
Table 3-4: Technical Applicability and Incremental Maximum Market Penetration of Solvent Options
(Percent)3
s
co c
1 < s
CO <
'E o o
Technical
Applicability
Option (All Years)"
Retrofit 100% 100% 100%
Conversion to HFE 79- 79- 79-
solvents 81% 81% 81%
NIK replacements 100% 100% 100%
Semiaqueous 100% 100% 100%
Aqueous 100% 100% 100%
United States
Non-U.S. Annex I
Non-Annex 1
United States
Non-U.S. Annex 1
Non-Annex 1
3
1 1 a
g c/i g
u = <
Ice
coo
X
« §
1 < 2
co c/5 g
•§ =? f
•E § §
Market Penetration
2005
5% -5% 5%
10% 5% 5%
4% 8% 8%
1% 3% 3%
3% 5% 5%
2010
0% 0% 8%
30% 10% 10%
8% 15% 15%
3% 5% 5%
5% 10% 10%
2015
0% 0% 12%
45% 15% 15%
12% 23% 23%
4% 8% 8%
8% 15% 15%
2020
0% 0% 15%
60% 25% 25%
15% 30% 30%
5% 10% 10%
10% 20% 20%
a Assumed maximum market penetration of options is presented as a percentage of total sector emissions for which the options are technically
applicable. The baseline market penetration is assumed to be zero to assess the emissions reductions possible due to increased use of each
option.
b The percentage of total emissions represented by HFEs varies by year. The technical applicability is 81 percent in 2005, and 79 percent in
2010 through 2020.
To calculate the percentage of emissions reductions off the total solvent baseline for each abatement
option, the technical applicability (Table 3-4) is multiplied by the market penetration value (Table 3-4)
and by the reduction efficiency of the option. For example, to determine the percentage reduction off the
2020 baseline for the "conversion to HFE solvents" option in the United States, the following calculation
is performed:
Technical applicability x Market penetration in 2020 x Reduction efficiency =
79.0% x 60.0% x 76.4% = 36.2%
Thus, using the assumptions in this analysis, converting to an HFE solvent could reduce
approximately 36 percent of the U.S. emissions baseline in 2020. This figure, along with the other
emissions reduction potentials, is shown in Table 3-5.
Table 3-6 presents a summary of the cost assumptions used for the solvent options presented in the
discussions above.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-67
-------
SECTION IV — INDUSTRIAL PROCESSES « SOLVENTS
Table 3-5: Emissions Reductions Off the Total Solvent Baseline (Percent)
Option
Retrofit
Conversion to HFE solvents
NIK replacements
Semiaqueous
Aqueous
55
8 § -
3 < 8
co £
1 1 f
coo
2005
3.5% 3.5% 3.5%
6.2% 3.1% 3.1%
3.8% 7.5% 7.5%
1.3% 2.5% 2.5%
2.5% 5.0% 5.0%
8
CO C
o> c ~~
% < 0
CO CO g
111
coo
2010
0.0% 0.0% 5.6%
18.1% 6.0% 6.0%
7.5% 15.0% 15.0%
2.5% 5.0% 5.0%
5.0% 70.0% 70.0%
. 1
1 < s
« oi i
111
coo
2015
0.0% 0.0% 8.4%
27.2% 9.1% 9.1%
11.3% 22.5% 22.5%
3.8% 7.5% 7.5%
7.5% 15.0% 15.0%
X
CD
CO C
1*1
CO CO g
"§=?«?
.-£ c c
coo
2020
0.0% 0.0% 10.5%
36.2% 15.1% 15.1%
15.0% 30.0% 30.0%
5.0% 70.0% 70.0%
70.0% 20.0% 20.0%
Table 3-6: Summary of Abatement Option Cost Assumptions
Option
Retrofit
NIK aqueous
NIK
semiaqueous
HFC to HFE
Time
Horizon
(Years)
10
10
10
10
Unit of Costs
Per degreaser with an
open-top area 13 ft2
Per standard
degreaser unit
Per standard
degreaser unit
Per kilogram of
solvent
Base One-
Time Cost
(2000$)
$16,800
$80,000
$10,000
$0
Base
Annual
Cost
(2000$)
$0
$0
$0
$0
Base
Annual
Savings
(2000$)
$233,300
$0
$0
$0
Net Annual
Costs
(2000S/yr)
-$233,300
$0
$0
$0
IV.3.4 Results
IV.3.4.1 Data Tables and Graphs
Tables 3-7 and 3-8 provide a summary of the potential emissions reduction opportunities at
associated breakeven costs in 15-dollar increments at a 10 percent discount rate (DR) and 40 percent tax
rate (TR). As shown, in 2010 and 2020, emissions reduction opportunities become available for regions
such as Annex I and OECD at the lowest breakeven cost of $0/tCO2eq. For regions such as Mexico and the
Russian Federation, emissions reduction opportunities are not available because emissions from the
solvent sector are so minute for these regions. A world total emissions reduction of 1.83 MtCO2eq is
projected by 2010 and 2.20 MtCO2eq by 2020, both at a breakeven cost of $15/tCO2eq.
Table 3-9 presents the costs, in 2000$, to reduce 1 MtCO2eq for a discount rate scenario of 10 percent
and a tax rate of 40 percent. The results are ordered by increasing costs per tCO2eq. Also presented are
the emissions reduced by the option, in MtCO2eq and percentage of the solvents baseline, and cumulative
totals of these two figures.
IV-68
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES « SOLVENTS
Table 3-7: Country/Regional Emissions Reductions in 2010 and Breakeven Costs for Solvents at 10%
Discount Rate, 40% Tax Rate ($/tC02eq)
2010
Country/Region
Africa
Annex!
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.00
0.53
0.00
0.01
0.16
0.00
0.13
0.00
0.08
0.00
0.00
0.59
0.00
0.03
0.30
0.80
$15
0.01
1.21
0.02
0.01
0.37
0.01
0.45
0.00
0.29
0.00
0.00
1.36
0.00
0.07
0.43
1.83
$30
0.01
1.21
0.02
0.01
0.37
0.01
0.45
0.00
0.29
0.00
0.00
1.36
0.00
0.07
0.43
1.83
$45
0.01
1.21
0.02
0.01
0.37
0.01
0.45
0.00
0.29
0.00
0.00
1.36
0.00
0.07
0.43
1.83
$60
0.01
1.21
0.02
0.01
0.37
0.01
0.45
0.00
0.29
0.00
0.00
1.36
0.00
0.07
0.43
1.83
>$60
0.01
1.21
0.02
0.01
0.37
0.01
0.45
0.00
0.29
0.00
0.00
1.36
0.00
0.07
0.43
1.83
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Table 3-8: Country/Regional Emissions Reductions in 2020 and Breakeven Costs for Solvents at 10%
Discount Rate, 40% Tax Rate (S/tC02eq)
2020
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.01
1.05
0.01
0.02
0.03
0.00
0.14
0.00
0.13
0.00
0.00
1.07
0.00
0.02
0.74
1.16
$15
0.02
1.96
0.04
0.04
0.06
0.01
0.40
0.00
0.40
0.00
0.01
2.01
0.00
0.05
1.05
2.20
$30
0.02
1.96
0.04
0.04
0.06
0.01
0.40
0.00
0.40
0.00
0.01
2.01
0.00
0.05
1.05
2.20
$45
0.02
1.96
0.04
0.04
0.06
0.01
0.40
0.00
0.40
0.00
0.01
2.01
0.00
0.05
1.05
2.20
$60
0.02
1.96
0.04
0.04
0.06
0.01
0.40
0.00
0.40
0.00
0.01
2.01
0.00
0.05
1.05
2.20
>$60
0.02
1.96
0.04
0.04
0.06
0.01
0.40
0.00
0.40
0.00
0.01
2.01
0.00
0.05
1.05
2.20
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-69
-------
SECTION IV — INDUSTRIAL PROCESSES • SOLVENTS
Table 3-9: World Breakeven Costs and Emissions Reductions in 2020 for Solvents
Reduction
Option
Retrofit
HFCtoHFE
NIK semiaqueous
NIK aqueous
Cost(2000$/tC02eq)
10%DR,40%TR
-$50.75
$0.00
$0.67
$5.36
Emissions
Reduction of
Option
(MtCOzeq)
0.0454
1.11
0.35
0.70
Reduction
from 2020
Baseline (%)
1.0%
24.7%
7.7%
15.5%
Cumulative
Reductions
(MtC02eq)
0.05
1.16
1.51
2.20
Cumulative
Reduction
from 2020
Baseline (%)
1.0%
25.7%
33.4%
48.9%
Figures 3-2 and 3-3 display the solvent international marginal abatement curves (MACs) by region
for 2010 and 2020, respectively.
Figure 3-2:
^ $40 -
o>
N
O
9 $20 -
C
Q et*f\
~ $0
§ 0.
"8
* -$20 -
in
0
'55
.22 -$40 -
ill
-$60 -
2010 MAC for Solvents, 10% Discount Rate, 40% Tax Rate
Y * * • :f
!
SL i. ~/~zr ~~" ', ' ' ~T- -• - •
DO 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50
— * China
Rest of the world
— • United States
EU-15
* Japan
— «r Other OECD
Cumulative Emissions Reductions (MtCO2eq)
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
IV-70
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • SOLVENTS
Figure 3-3: 2020 MAC for Solvents, 10% Discount Rate, 40% Tax Rate
$60
§f $40
CM
o
§. $20-
$0 -
0
w
c
'•B
1
w -$20 -
o
'w
| -$40 -\
m
-$60
0.2
0.4
0.6
0.8
1.0
1.2
China
Rest of the world
•* United States
' EU-15
*• Japan
< Other OECD
Cumulative Emissions Reductions (MtCO2eq)
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
IV.3.4.2 Uncertainties and Limitations
This section focuses on the uncertainties and limitations associated with the cost estimates presented
in this analysis. One significant area of uncertainty is how capital costs for these mitigation technologies
may vary internationally. The analysis is currently limited by the lack of this specificity on region-specific
cost analysis estimates. In addition, the three abatement options identified in this analysis have the
following uncertainties.
Conversion to HFE Solvents
Short- and long-term cost savings may occur with this option; yet because of their uncertainty, this
analysis conservatively assumes no cost savings.
Improved Equipment and Cleaning Processes Using Existing Solvents (Retrofit)
The analysis does not realize any annual labor costs that may accompany the use of retrofitted
equipment. These incurred costs may include training and frequent, mandatory maintenance checks.
Aqueous and Semiaqueous NIK Replacement Alternatives
The major uncertainties regarding this option are the annual costs and cost savings. Because cost
savings, which may offset increased operating costs, are not quantified for this analysis, this analysis does
not assume annual costs or cost savings for this option.
IV.3.5 Summary
Baseline global HFC, HFE, and PFC emissions from solvents are estimated to decline from 16.4 to 4.5
MtCO2eq between 2000 and 2020. In 2020, Annex I countries are assumed to account for approximately 90
percent of global emissions, with U.S. emissions assumed to account for half of emissions from Annex I
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-71
-------
SECTION IV — INDUSTRIAL PROCESSES • SOLVENTS
countries (Table 3-2). Projected growth in emissions (between 2010 and 2020) is expected to occur only in
the United States, from 1.7 MtCO2eq in 2010 to 2.0 MtCO2eq in 2020.
This analysis considers three emissions mitigation options for solvent use: (1) adoption of alternative,
(HFE-7100 or HFE-7200) partially fluorinated solvents, (2) improved system design through retrofitting
solvent processes, and (3) conversion to NIK (aqueous and semiaqueous replacements). The costs and
emissions reduction benefits of each option were compared in each region (Tables 3-7 and 3-8). Globally,
retrofitting represents the most cost-effective option for reducing HFC, HFE, and PFC emissions from the
solvent sector, with a cost savings of $50.75 per tCO2eq at a 10 percent discount rate and 40 percent tax
rate. Converting to an HFE solvent is a cost-neutral option for all regions. By 2020, 2.20 MtCO2eq, or 49
percent of global baseline emissions from solvents, can be reduced at a cost under $10 per tCO2eq.
For all three options, costs per tCO2eq for each region are equivalent because available data on costs
for abatement technologies were not scaled to reflect potential differences in the costs internationally.
Actual costs for abatement options for specific countries may vary and subsequently affect these
estimates. Additional research is required to determine actual variability in costs across regions.
[V.3.6 References
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.
3M Performance Materials. September 2004. Personal communication and written correspondence
between industry technical expert John G. Owens, P.E., of 3M Performance Materials and Mollie
Averyt of ICF Consulting.
Chaneski, W. November 1997. "Competing Ideas: Aqueous Cleaning—The Cost-Friendly Solution."
Modern Machine Shop. Available at .
DuPont FluoroProducts. October 2004. Personal communication and written correspondence between
industry technical expert Abid Merchant of DuPont FluoroProducts and Mollie Averyt of ICF
Consulting.
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.
European Climate Change Program (ECCP). February 2001. "Annex I to the Final Report on European
Climate Change Programme Working Group Industry Work Item Fluorinated Gases: ECCP
Solvents." Position paper provided by European Fluorocarbon Technical Committee (EFTC).
Honeywell. 2003. "Genesolv® S-T: A New HFC-Trans Blend Based Solvent for Industrial Aerosol,
Specialty Cleaning, Flushing and Deposition." Honeywell Technical Bulletin. BJ-6108-3/03-XXXX.
Available online at .
ICF Consulting. March 12, 1992. Cost of Alternatives to CFC-113 and Methyl Chloroform Solvent
Cleaning for the Safe Alternatives Analysis.
Intergovernmental Panel on Climate Change (IPCC). 1996. Climate Change 2995, The Science of Climate
Change. Contribution of Working Group I to the Second Assessment Report of the Intergovernmental
Panel on Climate Change. Cambridge University Press.
Intergovernmental Panel on Climate Change Third Assessment Report (IPCC TAR). 2Q01.Climate Change
2001, The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the
Intergovernmental Panel on Climate Change. Cambrid ge University Press.
March Consulting Group. 1998. Opportunities to Minimize Emissions of Hydrofluorocarbons (HFCs) from the
European Union: Final Report. Manchester, England: March Consulting Group.
IV-72 GLOBAL MITIGATION OF NON C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • SOLVENTS
March Consulting Group. 1999. UK Emissions of HFCs, PFC, and SF6 and Potential Emission Reduction
Options: Final Report. Manchester, England: March Consulting Group.
Microcare Marketing Services, Vertrel®. 2002. "What is HFC-365 and What Does It Do?" Microcare
Marketing Services. Available at .
Petroferm. January 2000. "Solvent Loss Control." Petroferm Technical Bulletin. Available at
.
Salerno, C. January 2001. "The New Generation of Solvents: Developmental Challenges Inspire Creative
Solutions." CleanTech. Available at .
SEMI International. 2003. Strategic Marketing Associates' World Fab Watch Database (WFW). April
edition.
United Nations Environment Programme (UNEP). 2003. UNEP 2002 Report of the Solvents, Coatings, and
Adhesives Technical Options Committee (STOC): 2002 Assessment. Nairobi, Kenya: UNEP Ozone
Secretariat.
United Nations Environment Programme (UNEP). 1999a. 1998 Report of the Solvents, Coatings, and
Adhesives Technical Options Committee (STOC): 1998 Assessment. Nairobi, Kenya: UNEP Ozone
Secretariat.
United Nations Environment Programme (UNEP). 1999b. The Implications to the Montreal Protocol of the
Inclusion of HFCs and PFCs in the Kyoto Protocol. United States: UNEP HFC and PFC Task Force of the
Technology and Economic Assessment Panel (TEAP) and Nairobi, Kenya: UNEP Ozone Secretariat.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-73
-------
SECTION IV— INDUSTRIAL PROCESSES • SOLVENTS
IV-74 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV— INDUSTRIAL PROCESSES • FOAMS
IV.4 HFC Emissions from Foams
IV.4.1 Introduction
arious HFCs are currently being used as blowing agents during the manufacture of foams.
These high-GWP gases are substitutes for ODSs that were historically the primary blowing
agents in the foams industry. Parties to the Montreal Protocol on Substances that Deplete the
Ozone Layer are phasing out CFCs, and many are using HCFCs as interim substitutes. Developed and
developing countries are at different phases of replacing CFCs with alternatives. Developed regions, such
as the United States and EU-15, have banned the sale and distribution of some foam products
manufactured with HCFCs and have begun transitioning to HFC use in foams where HCs and other
alternatives are not already used. For example, Denmark, Austria, Finland, and Sweden phased out the
use of HCFCs for foam blowing on January 1, 2002, while in the United States, HCFC-141b has been
phased out but HCFC-22 is still being used.
Developing countries have only recently begun transitioning from CFC-11 to HCFCs and other
alternatives. The rate of conversion to HFCs may be limited by the current availability of other ODS
substitutes and by technical barriers and cost. For example, the main blowing agent alternatives for CFC-
11 in rigid polyurethane (PU) insulating foams are HCs, such as pentanes, and HCFCs. Applying
alternative (i.e., HFC) technologies may require the use of higher-density foam, which would result in
incremental operating cost increases CO2.
The most commonly used HFC blowing agents are HFC-134a, HFC-152a, HFC-245fa, and HFC-
365mfc in combination with HFC-227ea. These blowing agents can be released into the atmosphere
during the foam manufacturing process, during on-site foam application, while foams are in use, and
when foams are discarded. These agents have 100-year GWPs of 1,300, 140, 950, and 890 in combination
with 2900, respectively, and have replaced historically used ODS blowing agents, including CFCs and
HCFCs. Foams studied in this analysis include the following:
• PU appliance foams found in various commercial and domestic refrigerators, vending machines,
freezers, water heaters, picnic boxes, flasks, thermoware, and refrigerated containers (reefers). PU
foam is the main insulation material used in refrigerators and freezers. PU foam must provide
continuous and effective insulation to ensure the quality of the product stored inside; therefore,
insulation properties must be maintained in order to preserve the performance of the appliance.
Basic performance requirements of some appliances are universal (e.g., refrigerators and freezers
keep food cold and water heaters keep water warm); however, some markets have specific
requirements such as energy consumption limits.
• PU spray foams are found in roofing insulation, wall insulation, and for insulation of various
tank pipe and vessel applications. PU spray foam is used in both residential and commercial
buildings as well as refrigerated transport. The main application in this category is spray roofing
insulation.
• PU continuous and discontinuous panel foam is used for insulation of cold storage, entrance and
garage doors, insulated trucks, etc.
• PU one-component foams are used for insulation around windows and doors, framing around
pipes, cable holes, jointing insulating panels, and certain roof components. PU one-component
foam is a preferred insulation method for portable and "easy to administer" applications.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-75
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
• Extruded polystyrene (XPS) boardstock foam is used mainly for thermal insulation purposes in
buildings. Its primary uses include basement walls, exterior walls, cavity walls, and roofing. Its
resistance to water absorption makes it a prime selection for "below-grade" applications. Some
XPS boardstock foam types are used in protection of roads or airport runaways against frost
("geofoam applications").
IV.4.2 Baseline Emissions Estimates
IV.4.2.1 Emissions Estimating Methodology
The USEPA uses a detailed Vintaging Model of ODS-containing equipment and products to estimate
the use and emissions of various ODSs and ODS substitutes in the United States, including HFCs and
PFCs. Emissions baselines from non-U.S. countries were derived using country-specific ODS
consumption estimates, as reported under the M ontreal Protocolin conjunction with Vintaging
Model output for each ODS-consuming end-use. These data were incorporated into country-specific
versions of the Vintaging Model to customize emissions estimates. In the absence of country-level data,
these preliminary estimates were calculated by assuming that the transition from ODSs to HFCs and
PFCs follows the same general substitution patterns internationally as the patterns observed in the
United States. From this preliminary assumption, emissions estimates were then tailored to individual
countries or regions by applying adjustment factors to U.S. substitution scenarios based on relative
differences in economic growth, rates of ODS phaseout, and the distribution of ODS use across end-uses
in each region or country.
Emissions Equations
Foams are given emissions profiles depending on the foam type (open cell or closed cell). Open-cell
foams are assumed to be 100 percent emissive in the year of manufacture, as described in Equation (4.1)
below. Closed-cell foams are assumed to emit a porfion of their total HFC or PFC content upon
manufacture, a portion at a constant rate over the lifetime of the foam, a portion at disposal, and a portion
postdisposal, as described in Equations (4.2) through (4.6), below.1
Open-Cell Foam
(4.1)
where
EJ = Emissions. Total emissions of a specific chemical in year j used for open-cell foam blowing,
by weight.
Qc = Quantity of chemical. Total amount of a specific chemical used for open-cell foam blowing in
a given year, by weight.
j = Year of emission.
1 Emissions from foams may vary because of handling and disposal of the foam; shredding of foams may increase
emissions, while landfilling of foams may abate some emissions (Scheutz and Kjeldsen, 2002; Scheutz and Kjeldsen,
2003). Average annual emissions are assumed in the model, which may not fully account for the range of foam
handling and disposal practices.
IV-76 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
Closed-Cell Foam
Emissions from foams occur at many different stages, including manufacturing, lifetime, disposal,
and postdisposal.
Manufacturing emissions occur in the year of foam manufacture, and are calculated as presented in
Equation (4.2).
Enij = lm* QCJ (4.2)
where
Errij= Emissions from manufacturing. Total emissions of a specific chemical in year j due to
manufacturing losses, by weight.
1m = Loss rate. Percent of original blowing agent emitted during foam manufacture.
Qc = Quantity of chemical. Total amount of a specific chemical used to manufacture closed-cell
foams in a given year.
j = Year of emission.
Lifetime emissions occur annually from closed-cell foams throughout the lifetime of the foam, as
calculated using Equation (4.3).
EUJ = lu x lQc;-_w for i = I -» k (4.3)
where
EUJ = Emissions from lifetime losses. Total emissions of a specific chemical in year j due to lifetime
losses during use, by weight.
lu = Leak rate. Percentage of original blowing agent emitted during lifetime use.
Qc = Quantity of chemical. Total amount of a specific chemical used to manufacture closed-cell
foams in a given year.
k = Lifetime. Average lifetime of foam product.
i = Counter. Runs from 1 to lifetime (k).
j = Year of emission.
Disposal emissions occur in the year the foam is disposed, and are calculated as presented in Equation
(4.4).
Edrld*Qcf_k (4.4)
where
Edj = Emissions from disposal. Total emissions of a specific chemical in year j at disposal, by
weight.
Id = Loss rate. Percent of original blowing agent emitted at disposal.
Qc = Quantity of chemical. Total amount of a specific chemical used to manufacture closed-cell
foams in a given year.
k = Lifetime. Average lifetime of foam product.
j = Year of emission.
Postdisposal emissions occur in the years after the foam is disposed, and are assumed to occur while the
disposed foam is in a landfill. Currently, the only foam type assumed to have postdisposal emissions is
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-77
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
polyurethane appliance foam, which is expected to continue to emit for 32 years postdisposal, and is
calculated as presented in Equation (4.5).
Epj = Ip x LQcy-_m for m = k -4 k + 32 (4.5)
where
Epj = Emissions postdisposal. Total postdisposal emissions of a specific chemical in year j, by
weight.
Ip = Leak rate. Percent of original blowing agent emitted post disposal.
Qc = Quantity of chemical. Total amount of a specific chemical used in closed-cell foams in a given
year.
k = Lifetime. Average lifetime of foam product.
m = Counter. Runs from lifetime (k) to (k + 32).
j = Year of emission.
To calculate total emissions from foams in any given year, emissions from all foam stages must be
summed, as presented in Equation (4.6).
j = Em, + EUJ + Ed, + Epf (4.6)
where
Ej = Total emissions. Total emissions of a specific chemical in year j, by weight.
Emj= Emissions from manufacturing losses. Total emissions of a specific chemical in year j due to
manufacturing leaks, by weight.
EUJ = Emissions from lifetime losses. Total emissions of a specific chemical in year j due to lifetime
losses during use, by weight.
Edj = Emissions at disposal. Total emissions of a specific chemical in year j due to disposal, by
weight.
Epj = Emissions postdisposal. Total postdisposal emissions of a specific chemical in year j, by
weight.
The emissions profile for foams estimated by the Vintaging Model is presented in Table 4-1.
Regional Adjustments
Foam sector emissions were estimated by developing Vintaging Model scenarios that were
representative of country- or region-specific substitution and consumption patterns. To estimate baseline
emissions, current and projected characterizations of international total foams markets were used to
create country- or region-specific versions of the Vintaging Model. The market information was obtained
from Ashford (2004), based on research conducted on global foam markets. Scenarios were developed for
Japan, Europe (both EU-15 and non-EU-15 countries combined), other developed countries (excluding
Canada), CEITs, and China. Other non-Annex I countries are assumed not to transition to HFCs during
the scope of this analysis. Once the Vintaging Model scenarios had been run, the emissions were
disaggregated to a country-specific level based on estimated 1989 CFC consumption for foams developed
for this analysis. Emissions estimates were adjusted slightly to account for relative differences in
countries' economic growth compared to the United States (USDA, 2002; USEIA, 2001).
IV-78 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV— INDUSTRIAL PROCESSES • FOAMS
Table 4-1: USEPA's Vintaging Model Emissions Profile for Foams' End-Uses
Loss at
Manufacturing
Foams End-Use (Percent)
Flexible PU
Polyisocyanurate boardstock
Rigid PU integral skin
PU appliance
PU commercial refrigeration
PU spray
One component
PU slabstock and other
Phenolic
Polyolefin
XPS foam sheet
XPS boardstock
Sandwich panel
100.0%
6.0%
95.0%
4.0%
6.0%
15.0%
100.0%
37.5%
23.0%
95.0%
40.0%
25.0%
5.5%
Annual
Release Rate
(Percent)
0.000%
1.000%
2.500%
0.250%
0.250%
1.500%
0.000%
0.750%
0.875%
2.500%
2.000%
0.750%
0.500%
Release
Lifetime
(Years)
1
50
2
20
15
56
1
15
32
2
25
50
50
Loss at
Disposal
(Percent)
0.00%
44.00%
0.00%
27.30%a
90.25%
1.00%
0.00%
51.25%
49.00%
0.00%
0.00%
37.50%
69.50%
Total
Released
(Percent)
100.0%
100.0%
100.0%
36.3%b
100.0%
100.0%
100.0%
100.0%
100.0%
100.0%
90.0%
100.0%
100.0%
a Estimated as 30 percent of the blowing agent remaining in the foam at the time of disposal (Scheutz and Kjeldsen, 2002).
b Emissions from disposed of products may continue if not otherwise abated. For MFCs, this analysis assumes 2 percent of the total blowing
agent used will continue to be emitted every year after disposal
Emissions baselines for Canada were derived using country-specific ODS consumption estimates, as
reported under the Montreal Protocol, in conjunction with U.S. Vintaging Model output for each ODS-
consuming end-use sector. Preliminary estimates were calculated by assuming that the transition from
ODSs to HFCs and other substitutes follows the same general substitution patterns as observed in the
United States.
Newly Manufactured Foam Emissions Versus Existing Foam Emissions
Technology options explored in the foams chapter are only applicable to new (i.e., not existing)
foams. Therefore, the technical applicabilities2 of the technology options in this sector include only
emissions from relevant end-uses that are from newly manufactured foams, which are defined as foams
manufactured in 2005 or later.
IV.4.2.2 Baseline Emissions
Table 4-2 provides a summary of baseline HFC emissions for the United States, other Annex I
countries, non-Annex I countries3 and other groupings through 2020. Emissions estimates for HFCs from
the foam sector are presented in MtCO2eq. These results are shown also in Figure 4-1.
In this report, the term "technically applicable" refers to the emissions to which an option can theoretically be
applied.
3 This analysis assumes that China is the only non-Annex I country that would transition to HFCs during the scope of
this study.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-79
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
Table 4-2: Baseline Emissions Estimates for Foams (MtC02eq)
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
0.0
1.5
0.0
0.0
0.0
0.0
1.1
0.0
0.0
0.0
0.0
1.5
0.0
0.0
0.3
1.5
2010
0.0
15.4
0.1
0.0
0.0
0.0
5.9
0.0
3.3
0.0
0.0
15.3
0.0
0.0
5.7
15.4
2020
0.0
28.6
0.2
0.0
0.1
0.0
11.4
0.0
4.8
0.0
0.1
28.5
0.0
0.0
11.3
28.6
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Figure 4-1: Total Baseline Emissions Estimates for Foams (MtC02eq)
2000
2010
Year
Q Middle East
• Africa
• Non-EU FSU
H Latin America
• S&E Asia
n China/CPA
• OECD90+
2020
CPA = Centrally Planned Asia; Non-EU FSU = non-European Union Former Soviet Union countries; OECD90+ = Organisation for Economic
Co-operation and Development; S&E Asia = Southeast Asia.
IV-80
GLOBAL MITIGATION OF NOM-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
IV.4.3 Cost of HFC Emissions Reductions from Foams
This section presents a cost analysis of achieving HFC emissions reductions from the emissions
baseline presented above.
For chemical replacement options, costs were based on the incremental differences between using the
HFC and switching to an HFC alternative. Financial information considered in this analysis includes
capital costs, which account for equipment costs to modify existing plants and to maintain production
capacity; blowing agent costs, which address the difference between costs and the quantity of the HFC
and non-HFC alternative required; foam costs, which address changes in foam density, the amount of fire
retardant used, the quantity and type of polyol, etc.; costs associated with profit and productivity; testing,
training, or other costs associated with transitioning to non-HFC alternatives; and costs to produce a
thicker, denser foam to account for any energy efficiency differences.
In addition, industry has indicated that there will be additional conversion or "learning curve" costs,
which are short-term costs incurred from yield, rate, and density penalties associated with conversion
uncertainties, as well as technical support costs. Such costs are highly variable and are not addressed in
the analysis.
IV.4.3.1 Abatement Options
Specific opportunities to reduce HFC emissions from the foams that were analyzed for this report fall
into two basic categories: blowing agent replacement options and end-of-life handling options.
Blowing agent replacement options include the following:
• replacing HFC-134a, HFC-245fa, and HFC-365mfc/HFC-227ea with HCs in PU continuous and
discontinuous panel foam;
• replacing HFC-134a and HFC-152a with HCs in one-component foam;
• replacing HFC-134a and carbon dioxide (CO2)-based blowing agents with liquid CO2
(LCD)/alcohol in XPS boardstock foam;
• replacing HFC-245fa/CO2 (water) and HFC-365mfc/HFC-227ea with CO2 (water) in PU spray
foam;
• replacing HFC-245fa/CO2 (water) and HFC-365mfc/HFC-227ea with HCs in PU spray foam;
• replacing HFC-134a with HCs in PU appliance; and
• replacing HFC-245fa and HFC-365mfc/HFC-227ea with HCs in PU appliance foam.
End-of-life handling options include the following:
• PU appliance foam practice: automated process with foam grinding and landfilling and
• PU appliance foam practice: manual process with incineration.
All abatement option cost analyses assume a 25-year project lifetime.
Replacement Options
Each of the replacement options includes the use of non-HFC blowing agents such as HCs, water-
blown CO^ and LCD. These foam technologies are described below. Section IV.4.3.2 gives specific
analyses of the costs of applying these alternate blowing agents to particular foam types.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-81
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
Hydrocarbons
HCs such as propane, butane, isobutane, n-pentane, isopentane, cycloperitane, and isomers of hexane
are alternatives to HFCs in foam blowing appliances. HCs are inexpensive and have near-zero direct
GWPs, much lower than HFCs. However, key technical issues associated with the use of HCs exist:
• Flammability. Factory upgrades that among other things ensure the use of nonsparking
equipment and employee training are required when switching to HCs, to meet the necessary
safety precautions in manufacturing, storage, handling, transport, and customer use. Examples of
upgrades include a dedicated storage tank for the HC, premixers, adapted high-pressure
dispensers, suitable molds plus process exhaust, HC detectors, and appropriate classification of
electrical equipment. To reduce fire risks, some applications might also require the use of a larger
quantity of flame retardants or the use of a more expensive fire retardant.
• Volatile Organic Compounds (VOCs). Because HCs contribute to ground-level ozone and smog,
they tend to be highly regulated. In many places, including some parts of the United States, HCs
cannot be used without emissions controls. Implementation of these controls can lead to
significant increases in the costs of conversion.
• Performance. Some HCs yield only about 85 percent of the insulating value of HCFC-141b, HFC-
245fa, and HFC-365mfc/HFC-227ea. Producing a thicker foam can compensate for this energy
efficiency difference, but will increase the cost of production and possible application costs (e.g.,
longer fasteners for thicker foam board). This option might not be viable in fixed-thickness
applications, such as refrigerated trucks, or in applications where an R-value is prescribed by
code, such as in PU spray roofing insulation. Other performance considerations include
dimensional stability and solubility. Addressing these factors might require a more expensive
and more limited polyol formulation.
Costs of converting to HCs and addressing technical considerations can be significant, but vary
according to factory-specific needs. HCs can also be used to enhance octane ratings, making them
valuable for gasoline use and affecting their cost (Werkema, 2006). In spite of these issues, HCs are
currently used in some applications and are being considered in a wide variety of additional applications
(UNEP, 1998; Alliance, 2000; Alliance, 2001).
Liquid Carbon Dioxide
The basic principle by which LCD blowing agents operate is the expansion of LCD to a gaseous state.
LCD is blended with other foam components under pressure prior to initiating the chemical reaction.
When decompressed, the CO2 expands, resulting in froth foam, which further expands with the
additional release of CO2 from the water/isocyanate resin reaction that forms the PU foam matrix. LCD
might require formulation changes to more readily dissolve the CO2 and to prevent deactivation of PU
catalysts. When LCD is introduced at the head, often referred to as third stream, the metering equipment
can be quite complicated and, to date, unreliable. Difficulties encountered in using LCD include the
limited solubility of the chemical mixture, controlled decompression, and distribution of the unavoidable
froth (UNEP, 1998). Foams blown with CO2 may suffer from lower thermal conductivity, lower
dimensional stability, and higher density than HCFC-blown foams. To overcome these limitations, CO2
can be blended with HCs or HFCs (Williams et al, 1999; Honeywell, 2000; Alliance, 2001).
Water-Blown (In Situ) CO? (Water)
In this process, CO2 produced from a chemical reaction between water and polymeric isocyanate is
used as a blowing agent. During manufacture, no ODS or high-GWP gases are emitted, and there are
limited health and safety risks during processing. However, foams produced using CO2/water are subject
IV-82 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
to the same performance limitations discussed for LCD-blown foams: lower thermal conductivity, lower
dimensional stability, and higher density than HCFC- and HFC-blown foams. In some PU foam
applications, a major concern when using water-generated or LCD systems is the increased open-cell
content, which results in poorer waterproofing performance and poorer waterproofing quality of the final
product. Another consideration is that the polymeric isocyanate content must be increased, which cannot
be accommodated by some PU spray foam equipment. To overcome these limitations, CO2 can be
blended with HCs or HFCs (Williams et al., 1999; Honeywell, 2000; Alliance, 2001). In some other
applications (e.g., PU block), there can be problems with uncontrollable exotherms when using purely
CO2 (water) systems. CO2/water blowing agent is used in extruded polystyrene boardstock in markets
where thermal efficiency is not critical; however, in some applications, higher densities or lower
conversion may offset the low costs of CO2/water. In some cases, costs associated with overcoming
technical challenges are so high that CO2/water systems may be out of reach for many small and medium
enterprises (IPCC, 2004).
Although LCD and CO2 generated in situ have similar performance issues, the process limitations
associated with each differ. Compared to LCD, fewer mechanical modifications are required when using
in situ CO2, and the foam manufacturer or PU spray foam applicator can be more certain of the final CO2
content and overall foam properties (Alliance, 2001).
End-of-Life PU Appliance Foam Practices
There are several methods for disposal of PU foam, including landfilling and incineration, with or
without ODS recovery and recycling or destruction. Two of the methods are described below, followed in
Section IV.4.3.2 by specific analyses of the costs associated with each method.
Landfilling
Traditionally, most decommissioned foam products have ended up in landfills. Although the
regulations related to the location and management of landfills have improved considerably, there is still
concern about the rate of release of blowing agent from foam in the first weeks after entering the landfill
(UNEP, 2002b).
Incineration
Incineration of foams in municipal solid waste incinerators (MSWIs) or waste-to-energy plants is a
practical and highly competitive technique for destruction of PU foam. An advantage of this technique is
that the foam can be incinerated without separating the foam matrix from the blowing agent prior to
incineration, which lowers the cost and the risk of fugitive emissions (UNEP, 2002b).
IV.4.3.2 Description and Costs of Abatement Options
The following section describes all options in greater detail and presents a cost analysis for those
options for which adequate cost data are available. The abatement options to reduce HFC emissions from
the foam sector are presented by foam type: PU continuous and discontinuous panel foam, one-
component foam, XPS boardstock foams, PU spray foams, and PU appliance foams. The technology
options explored in this chapter are assumed to penetrate only the markets of new (i.e., not existing)
foams. The remainder of this section provides a description of the economic assumptions for these
abatement options. Throughout this discussion, we refer to Tables 4-4 and 4-5, which provide information
on the technical applicability and the incremental maximum market penetrations assumed for each
abatement option. These tables are discussed in greater detail in Section IV.4.3.3. A detailed description of
the cost and emissions reduction analysis for each option can be found in Appendix H for this chapter.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-83
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
PU Continuous and Discontinuous Panel Foam
The only abatement option that was considered for this category is replacing HFCs with HCs. This
cost analysis estimates the breakeven carbon price for a hypothetical contractor to replace HFCs with
HCs. In the base case scenario, the blowing agent constitutes 8.7 percent of the foam, by weight. In the
base case, 1,048,600 pounds of blowing agent are consumed (UNEP, 2002a); hence, 12,052,874 pounds of
foam are produced (1,048,600/8.7% = 12,052,874). The foams manufactured with the alternative are
assumed to compensate for lower insulating performance relative to HFC-blown foams by increasing the
thickness and density of the foam. Although this end-use uses HFC-134a, HFC-245fa, and HFC-
365mfc/HFC-227ea, the analysis performed was based on a PU continuous and discontinuous panel foam
contractor that uses HFC-134a. A contractor that uses HFC-245fa and HFC-365mfc/HFC-227ea would see
higher cost savings for this replacement option because these HFCs are more expensive than HFC-134a.
But, because HFC-245fa and HFC-365mfc/HFC-227ea have lower GWPs, the option would yield a lower
ton of carbon equivalent (tCO2eq) savings. This analysis is based on a hypothetical PU continuous and
discontinuous panel foam contractor that uses approximately 1 million pounds of HFC-134a per year
(ICF Consulting, 2004).
Cost factors that are addressed include the following:
• capital equipment costs such as costs of installing safety equipment including nonsparking
equipment,
• increased cost of foam components (e.g., polyols, additives),
• increased consumption of foam components to compensate for increased foam density,
• worker safety,
• increased use of fire retardant, and
• incremental differences in the costs of blowing agents and the quantity required.
This option is technically applicable4 to all emissions from the newly produced continuous and
discontinuous panel foams. The technical applicability of this option from 2005 to 2020 is presented in
Table 4-4. This analysis assumes that the incremental maximum market penetration of this option in the
newly produced continuous and discontinuous panel market that uses HFC-134a will be 70 percent for
the United States and 90 percent for the rest of the world by 2010, both rising to 100 percent by 2020 (see
Table 4-5). Because the HFC is replaced by a HC, the reduction efficiency is assumed to be 100 percent.
Assumptions specific to this substitution are presented in Appendix H for this chapter.
One-Component Foam
Two blowing agent replacement abatement options were considered for this end-use:
• replacing HFC-134a with propane/butane and
• replacing HFC-152a with propane/butane.
An analysis was performed based on a hypothetical one-component foam contractor that uses 288,000
pounds per year of HFC-134a or HFC-152a (ICF Consulting, 2004). In the base case, the blowing agent
constitutes 8.7 percent of the foam, by weight; hence, 3,310,345 pounds of foam is produced (288,000/8.7%
4 In this report, the term "technically applicable" refers to the emissions to which an option can theoretically be
applied. Because this option examines the replacement of HFC-134a with HCs in specific end-uses and cannot be
retroactively applied to HFC-134a foam that has already entered the market, the technical applicability is the
percentage of baseline foam emissions that are HFC-134a from continuous and discontinuous panels placed on the
market after 2004. Other factors will affect the market penetration of the option assumed in this analysis.
IV-84 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
= 3,310,345). This cost analysis estimates the breakeven carbon price for this hypothetical contractor to
replace HFC-134a or HFC-152a with HCs (for details, see Appendix H).
Costs addressed include the following:
• capital equipment costs,
• increased cost of foam components (e.g., polyols, additives),
• increased consumption of foam components to compensate for increased foam density,
• worker safety,
• increased use of fire retardant, and
• incremental differences in the costs of blowing agents and the quantity required.
Replacing HFC-134a with HCs for One-Component Foam
This option is technically applicable to all HFC-134a emissions from the newly produced one-
component foams. The technical applicability of this option from 2005 to 2020 is presented in Table 4-4.
This analysis assumes that the incremental maximum market penetration for this option in the newly
produced one-component market that uses HFC-134a would be 70 percent for the United States and 90
percent for the rest of the world in 2010, both increasing to 100 percent by 2020 (see Table 4-5); reduction
efficiency is assumed to be 100 percent. Assumptions specific to this substitution are presented in
Table H-4 in Appendix H.
One-Component: Replacing HFC-152awith HCs
This option is technically applicable to all HFC-152a emissions from the newly produced one-
component foams. The technical applicability of this option from 2005 to 2020 is presented in Table 4-4.
This analysis assumes that the incremental maximum market penetration for this option in the newly
produced one-component foam market that uses HFC-152a would be 70 percent for the United States and
90 percent for the rest of the world by 2010, both increasing to 100 percent by 2020 (see Table 4-5);
reduction efficiency is assumed to be 100 percent.
XPS Boards took Foams
One blowing agent replacement option was considered for this end-use:
Replacing HFC-134a and CO,-based Blends with CO, (LCD)/Alcohol for XPS Boardstock Foams
An analysis was performed based on a hypothetical producer that manufactures approximately 1
billion board feet (bd-ft) of foam per year, across 10 lines, using HFC-134a and CO2-based blends as the
blowing agent. Various base case inputs and assumptions are presented in Table H-6 in Appendix H.
This cost analysis estimates the breakeven carbon price for this hypothetical producer to replace an HFC-
134a and CO2-based blend with CO2/alcohol in one of the 10 lines. Using this alternative, the foam
manufactured is assumed to compensate for lower insulating performance relative to HFC-blown foams
by increasing the thickness of the foam in the application, where possible. Thus, incremental differences
in indirect emissions and costs associated with energy penalties are negligible.
Cost factors that are considered include the following:
• blowing agent costs,
• capital equipment costs,
• increased consumption of foam components to compensate for increased foam density,
• incremental differences in the costs of blowing agents and the quantity required, and
• costs of lost capacity.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-85
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
The baseline blowing agent for XPS boardstock is assumed to be an HFC-134a and CO2-based blend.
Although many XPS boardstock facilities currently use HCFCs, it is assumed that this use will be phased
out under the Montreal Protocol and, hence, baseline alternative emissions are calculated assuming the
phase-in of HFC-134a. This option is technically applicable5 to all emissions from newly produced XPS
boardstock foam; that is, one could theoretically use CO2 (LCD)/alcohol in any new XPS boardstock foam
produced. The technical applicability of this option (i.e., the percent of foam sector emissions calculated
as arising from new XPS boardstock foam) from 2005 to 2020, is presented in Table 4-4. The incremental
maximum market penetration of this option into the newly produced XPS foam market is assumed to be
0 percent for the United States through 2020; 70 percent in 2010, rising to 90 percent by 2020 in all other
developed countries and CEITs; and 70 percent in 2010, rising to 90 percent by 2020 for China (see
Table 4-5). The option completely eliminates emissions of HFC-134a, where applied, and, hence, has a
reduction efficiency of 100 percent. Assumptions specific to this substitution are explained below and are
presented in Appendix H for this chapter.
PU Spray Foams
Two blowing agent replacement options were considered for this end-use:
• replacing HFC-245fa/CO2 (water) and HFC-365mfc/HFC-227ea with CO2 (water) and
• replacing HFC-245fa/CO2 (water) and HFC-365mfc/HFC-227ea with cyclopentane/ isopentane.
An analysis was performed based on a hypothetical PU spray foam contractor that produces
approximately 127,000 pounds of foam per year using a 75/25 blend of HFC-245fa6 and CO2 (water) as a
blowing agent. The base case blowing agent constitutes approximately 10 percent of the foam, by weight.
Various base case inputs and assumptions are presented in Table H-8 in Appendix H. The foams
manufactured with the two alternatives are assumed to require an increase in thickness and density to
compensate for lower insulating performance relative to HFC-blown foams. Thus, there are no
incremental differences in indirect emissions and costs associated with energy penalties. Although both
HFC-245fa and HFC-365mfc/HFC-227ea are used in this end-use, this analysis was based on a PU spray
foam contractor that uses HFC-245fa. Cost factors that are addressed include the following:
• fire testing costs incurred by system houses for various formulations,
• sparking of roof top equipment units,
• capital equipment costs,
• employee training costs (HCs only),
• increased cost of foam components (e.g., polyols, additives),
• increased consumption of foam components to compensate for increased foam density,
• increased use of fire retardant, and
• incremental differences in the costs of blowing agents and the quantity required.
5 In this report, the term "technically applicable" refers to the emissions to which an option can theoretically be
applied. Because this option examines the replacement of HFC-134a with CO2 in only XPS foam and cannot be
retroactively applied to foam that has already entered the market, the technical applicability is the percentage of
baseline foam emissions that arises from HFC emissions from XPS foam placed on the market after 2004. Hence,
technical applicability, in this sense, refers to the percentage of foam sector emissions calculated as arising from post-
2004 XPS foam. Other factors will affect the market penetration of the option assumed in this analysis.
6 The EU-15 countries use a blend of HFC-365mfc and HFC-227ea in ratios of 93:7 or 87:13, while Japan uses a blend
of HFC-245fa and HFC-365mfc in ratios of 80:20 or 70:30. This report presents a cost analysis based on the 75/25 HFC-
245fa/CO2 blend and applies it globally as a representative estimate.
IV-86 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
Annual emissions reductions were determined based on the estimated amount of blowing agent
consumed by the hypothetical contractor and from the emissions profile used in the Vintaging Model (see
Table 4-1).
PU Spray: Replacing HFC-245fa/CO? (Water) and HFC-365mfc/HFC-227ea with CO,
(Water)
This option is technically applicable7 to all emissions from the newly manufactured spray
polyurethane foam market, but the assumed market penetration is tempered by the existence of another
feasible option (i.e., HCs). The technical applicability of this option as well as other options from 2005 to
2020 is presented in Table 4-4. This analysis assumes that the incremental maximum market penetration
for this option into the newly formulated polyurethane spray foam market is 5 percent for the United
States in 2010, and 8 percent for the rest of the world, both rising to 20 percent by 2020 (see Table 4-5); the
reduction efficiency is assumed to be 100 percent because the HFC blowing agent is completely replaced
(the GWP of CO2 is not included in the analysis). For cost estimating purposes, this option assumes that
the baseline blowing agent is a 75/25 blend of HFC-245fa and CO2.
PU Spray: Replacing HFC-245fa/CO, (Water) and HFC-365mfc/HFC-227ea with HCs
The difference in costs between this abatement option and replacing HFC-245fa/CO2 with CO2 is the
cost of training workers in handling, storing, and using HCs. For cost-estimating purposes, the baseline
blowing agent is assumed to be a 75/25 blend of HFC-245fa and CO2, while the alternative blowing agent
is assumed to be an 80/20 blend of cyclopentane and isopentane. The technical applicability of this option
from 2005 to 2020 is presented in Table 4-4. This analysis assumes that the incremental maximum market
penetration of this option in the newly produced PU spray foam market would be 10 percent for the
United States and 5 percent for the rest of the world in 2010, rising in later years to 30 percent in the
United States and 15 percent in the rest of the world (see Table 4-5); reduction efficiency is assumed to be
100 percent. There could be some safety and liability concerns associated with this substitution, which
could lead to reduced market penetration or increased cost of this option.
PU Appliance Foams: Replacement Options
Two blowing agent replacement abatement options were considered for this end-use:
• replacing HFC-134a with cyclopentane/isopentane and
• replacing HFC-245fa and HFC-365mfc/HFC-227ea and with cyclopentane/isopentane.
This scenario examines a hypothetical facility that manufactures approximately 536,000 refrigerators
and consumes about 1.68 million pounds (0.00076 Mt) of blowing agent annually. The blowing agent was
assumed to constitute approximately 12 percent of the foam. The costs of producing a refrigerator using
each blowing agent (e.g., HFC-134a, HFC-245fa, and cyclopentane/isopentane) were provided by the
refrigeration industry. Data have been aggregated to protect confidential business information. This
scenario was developed for a facility manufacturing large appliances typically used in the United States.
While other markets may use different-sized refrigerators, and hence per-appliance factors may differ,
this analysis assumes that the resulting cost per HFC emissions abated (dollars per tCO2eq) is
approximately the same. Factors considered in these data include the following:
7 In this report, the term "technically applicable" refers to the emissions to which an option can theoretically be
applied. Because this option examines the replacement of HFCs with CO2 in a specific end-use and cannot be
retroactively applied to foam that has already entered the market, the technical applicability is the percentage of
baseline foam emissions from PU spray foam placed on the market after 2004. Other factors will affect the market
penetration of the option assumed in this analysis.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-87
-------
SECTION IV — INDUSTRIAL PROCESSES « FOAMS
• capital costs to convert;
• blowing agent costs;
• foam costs, including density considerations;
• high-impact polystyrene (HIPS) and acrylonitrile-butadiene-styrene (ABS) liner costs;
• additional costs required to meet the U.S. 2001 National Appliance Energy Conservation Act
(NAECA) energy efficiency standards; and
• the energy gap between different blowing agents and energy consumption increase as a result of
the conversion.
HFC emissions reductions over time were derived from the emissions profile used in the Vintaging
Model (see Table 4-1). These emissions account for gases released from the manufacturing process,
annual release, disposal, and post disposal. Because the cost data are based on the assumption that the
refrigerators manufactured using various blowing agents meet the same energy-efficiency standards,
there are no incremental differences in indirect emissions and costs resulting from energy-efficiency
differences.
PU Appliance: Replacing HFC-134a with HCs
This option is technically applicable8 to all HFC-]34a emissions from newly manufactured PU
appliance foam. The technical applicability of this option from 2005 to 2020 is presented in Table 4-4. This
analysis assumes that the incremental maximum market penetration in 2010 for this option in the newly
manufactured appliance market that uses HFC-134a would be 25 percent for the United States and 85
percent for the rest of Annex I, rising to 70 percent and 90 percent, respectively, by 2020 (see Table 4-5).
Because the HFC is completely replaced, the reduction efficiency is 100 percent.
PU Appliance: Replacing HFC-245fa and HFC-365mfc/HFC-227ea with HCs
Although some manufacturers may use HFC-365mfc/HFC-227ea instead of HFC-245fa, this analysis
was performed based on the cost to replace HFC-245fa in PU appliance foams. This option is technically
applicable to all emissions from the newly produced PU appliance foams that use HFC-245fa and HFC-
365mfc/HFC-227ea. The technical applicability of this option from 2005 to 2020 is presented in Table 4-4.
This analysis assumes that the incremental maximum market penetration of this option into the newly
manufactured appliance market that uses HFC-245fa and HFC-365mfc/HFC-227ea is 15 percent for the
United States in 2010 rising to 50 percent by 2020. For all other countries, the market penetration in 2010
is 85 percent, rising to 90 percent by 2020 (see Table 4-5). Because the HFC is completely replaced, the
reduction efficiency is 100 percent.
PU Appliance: End-of-Life Options
In addition to the two blowing agent replacement options considered above, two end-of-life
abatement options were considered for this end-use:
• automated process with foam grinding, HFC adsorption, and foam landfilling in PU appliance
foam and
• manual process with foam incineration in PU appliance foam.
8 In this report, the term "technically applicable" refers to the emissions to which an option can theoretically be
applied. Because this option examines the replacement of HFC-134a with HCs in a specific end-use and cannot be
retroactively applied to foam that has already entered the market, the technical applicability is the percentage of
baseline foam emissions from appliance foam made with HFC-134a and placed on the market after 2004. Other
factors will affect the market penetration of the option assumed in this analysis.
IV-88 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
The baseline emissions are based on the assumption that the remainder of the blowing agent
contained in the appliance foam is released after the foam's end of life, as shown in Table 4-1. Different
technologies exist for abating end-of-life emissions in PU appliance foams. These technologies include
landfilling the foam after recovering the blowing agent (which could either be destroyed or reclaimed
and sold back to the market) and incinerating the foam (and the remaining blowing agent) in a Municipal
Solid Waste Incinerator (MSWI) or waste-to-energy plant. This analysis analyzes the landfilling after
recovering HFC and the MSWI options. This analysis assumes that when the HFC is recovered, it will still
have value and hence contribute revenue to the process. HFC-134a and HFC-245fa are used in PU
appliance foam in some locations. To account for the chemicals' different GWPs and costs, this analysis
assumes that half of the appliances processed use HFC-134a and the other half use HFC-245fa. Further
market research could refine this assumption.
Appendix H for this chapter presents cost estimates for each step involved in the removal or
destruction of HFC contained in the foam, either through MSWI or grinding/adsorption/landfilling. Costs
are presented in terms of dollars per refrigerator and in dollars per pound of HFC abated. This analysis
uses the best cost information available; however, the costs presented should be considered illustrative
rather than definitive. The analysis is done using the U.S. market as an example, recognizing that a U.S.
refrigerator/freezer is typically larger than those used in other parts of the world. The final results (i.e.,
cost per unit of emissions abated) are applied to other regions because it is felt that the relative costs and
emissions abated should scale roughly linearly to smaller appliances used elsewhere. All assumptions are
based on a side-by-side refrigerator model.
The following two basic methods of handling appliances to abate blowing agent emissions are
examined:
• Automated Process with Foam Grinding, HFC Adsorption, and Foam Landfilling. This method
involves purchasing a sophisticated system where the appliance is brought into the system
without much preparation. The system shreds the appliance and uses various techniques such as
magnets and eddy current to separate the metals, plastics, and foams. The blowing agent (and the
refrigerant) is collected by adsorption9 onto a carbon substrate. Typically, the adsorbed gases are
then incinerated, or they can be reclaimed and sold back into the market. These systems are
capital-intensive, costing $3,646,308 (JACO, 2004); however, once established, the manual labor is
reduced. This type of process is generally only cost effective if a high flow of appliances (i.e.,
hundreds of thousands per year) is achieved.
• Manual Process with Foam Incineration. This method uses mostly manual labor to evacuate and
recycle the refrigerant, drain and recycle the compressor oil, and disassemble the appliances,
recovering and recycling glass shelves, plastic interior parts, steel, aluminum and other valuable
metals. The foam is removed in large pieces, which can be quickly sealed in plastic bags to
prevent further off-gassing of the blowing agent, and sent for incineration.
Cost factors that are addressed include the following:
• collection and consolidation of appliances,
• transportation of appliances to processing/disassembly location,
• disassembly and processing of appliances,
Other methods of blowing agent recovery are possible. For instance, some plants use liquid nitrogen to mitigate
explosion potential with HC units. The nitrogen also serves to liquefy and collect the blowing agent.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-89
-------
SECTION IV— INDUSTRIAL PROCESSES • FOAMS
• transportation of foam to landfilling or incineration location, and
• landfilling or incineration of foam.
Assumptions common to both the automated process with landfilling and the manual process with
incineration abatement options are presented in Appendix H for this chapter.
PU Appliance: Automated Process with Foam Grinding. HFC Adsorption, and Foam
Landfillinq
The technical applicability of this option from 2005 through 2020 is presented in Table 4-4. This
analysis assumes that the incremental maximum market penetration of this option in the appliance foam
market in 2020 would be 10 percent in the United States, 95 percent in Europe and Japan, and 70 percent
in the rest of the developed world (see Table 4-5).
PU Appliance: Manual Process with Foam Incineration
The technical applicability of this option from 2005 through 2020 is presented in Table 4-4. This
analysis assumes that the incremental maximum market penetration of this option in the appliance foam
market in 2020 would be 30 percent in the United States and 10 percent in other developed countries
except for the EU-15 and Japan, where the option is assumed not to penetrate the market.
IV.4.3.3 Summary of Technical Applicability, Market Penetration, and Costs of
Abatement Options
Table 4-3 presents a summary of the assumed reduction efficiency, while Table 4-4 shows the
technical applicability of the abatement options. Technical applicability values are based on the
percentage of total foam emissions from each end-use and are derived from the baseline emissions
methodology described in Section IV.4.2.1. The blowing agent replacement options explored in this
chapter are assumed to penetrate only new (not existing) equipment, where "new" equipment is defined
as equipment manufactured in 2005 or later.
Table 4-3: Reduction Efficiency of Foam Options (Percent)
Option Reduction Efficiency
PU appliance: HFC-134a to HC 100.0
PU appliance: HFC-245fa and HFC 365mfc/HFC-227ea to HC 100.0
PU appliance: automated process with foam grinding and landfilling 90.0
PU appliance: manual process with incineration 90.6
PU spray: HFC-245fa/C02 (water) and HFC-365mfc/HFC-227ea to HC 100.0
PU spray: HFC-245fa/C02 (water) and HFC-365mfc/HFC-227ea to C02 (water) 100.0
XPS boardstock: HFC-134a/C02 to CCyalcohol 100.0
One-component: HFC-134a to HC 100.0
One-component: HFC-152a to HC 100.0
PU continuous and discontinuous panel foam: HFC-134a to HC 100.0
IV-90 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
A summary of the incremental maximum market penetrations assumed for the abatement options
considered is presented in Tables 4-5 and 4-6.
To calculate the percent of emissions reductions off the total foams baseline for each abatement
option, the percent of baseline emissions from Table 4-4 (i.e., technical applicability) is multiplied by the
market penetration values from Table 4-6 and reduction efficiencies from Table 4-3. For example, to
determine the percentage reduction off the 2020 baseline for replacing HFC-134a/CO2 (LCD) with CO2
(LCD)/alcohol in the XPS foam option in Japan, the following calculation is used:
Technical applicability * Incremental maximum market penetration x Reduction efficiency =
29% x 83% x 100% = 24%
Thus, using the assumptions in this analysis, applying this chemical replacement option could reduce
Japan's baseline emissions by approximately 24 percent in 2020. This figure, along with the other
emissions reduction potentials, is shown in Table 4-7.
Table 4-8 summarizes the cost assumptions used for the foam options presented in the discussions
above.
IV.4.4 Results
IV.4.4.1 Data Tables and Graphs
Table 4-9 and 4-10 provide a summary of the potential emissions reductions at various breakeven
costs by country/region in 2010 and 2020, respectively. The costs to reduce 1 tCO2eq are presented for a
discount rate of 10 percent and a tax rate of 40 percent.
Table 4-11 presents the costs, in 2000$, to reduce 1 tCO2eq for a discount rate scenario of 10 percent
and a tax rate of 40 percent, for all the options analyzed. The results are ordered by increasing costs per
tCO2eq. Also presented are the emissions reduced by the option, in MtCO2eq and percentage of the foam
baseline, and cumulative totals of these two figures. Figures 4-2 and 4-3 present MACs for this sector at 10
percent discount rates and 40 percent tax rates in 2010 and 2020, respectively.
IV.4.4.2 Uncertainties and Limitations
This section focuses on the uncertainties associated with the cost estimates presented in this report.
Conversion Costs
In general, industry has indicated that there will be additional conversion or "learning curve" costs,
which are short-term costs incurred as a result of yield, rate, and density penalties associated with
conversion uncertainties as well as technical support costs. Such costs are highly variable and are not
addressed in the analysis.
Capital and Annual Costs
A major uncertainty of this analysis is the cost of the abatement technologies. Currently, almost all of
the costs for abatement options are available only for the United States. Thus, all U.S. capital and annual
costs were applied internationally, other than the capital costs for the spray foam option, for which we
had country-specific information for Japan and EU-15. However, costs may be higher internationally due
to transportation and tariffs associated with purchasing the technology from abroad, or may be lower if
there is domestic production of these technologies. In addition, for some abatement options, U.S. capital
and annual costs were obtained in 2001; thus, these costs may be somewhat out of date.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-91
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
"5s
x !E
go.
II
o
0
(O
m
o
1
J— s
1
c
|l
=
c
S.
*
e s.
§ i
U .3
« "«
^^
•^_^
(A
O
*-
a.
O
E 8
£ 1
° J
•51 c
.^_ "H^
__ ^J
re
0
0202
9102
0102
9002
0202
9102
01-02
9002
0202
9102
OL02
9002
0202
9102
01-02
9002
0202
9102
01.02
9002
0202
9102
0102
9002
"5
a.
"re
I
•i c
a> .2
5 S.
re O
0
CD
O
O
CD
0
CD
0
O
O
O
O
0
CD
CD
O
O
0
O
CD
-* '
CM
CM
CO
O
CO
|
a.
o
0
o
0
o
CD
CD
CD
S
CO
CM
CO
CM
O
CM
^.
^.
Tl-
CM
CTJ
O
^n
a>
J2
^
CO
LO
PU appliance: HFC-245fa
and HFC-365mfc/HFC-
227ea to HC
0
CD
O
0
CD
0
CD
0
CM
CO
CM
CO
CM
O
CM
^
^.
-*
CM
O5
0
^:
0>
LO
LO
LO
CO
CM
PU appliance: automated
process with foam
grinding, HFC
adsorption, and foam
landfilling .
o
CD
0
O
CD
O
CD
O
O
CM
CO
CM
CO
CM
CM
<«•
^-
-d-
CM
CJ5
O
JI
03
LO
^
LO
CO
CM
PU appliance: manual
process with foam
incineration
o
0
o
o
CD
0
O
0
CO
CO
CO
CO
CO
£
CO
CO
CO
CO
oo
CM
O5
CO
CM
CM
LO
CM
OO
§
"
CO
CM
CO
PU spray: HFC-245fa and
HFC-365mfc/HFC-227e
toHC
o
o
o
CD
CD
0
O
0
CO
CO
CO
CO
CO
£
CO
CO
CO
co
CM
CTJ
CO
CM
CTJ
CM
LO
CM
CO
§
OO
CO
•>*•
CO
CM
LO
CO
X 1
IS 8
03 x 2
a.
CD
o o
o 8
o 8
o 8
~ Si
CO "^
•* cS
o 8
0 0
CD O
CD O
CD O
gj ^
CO '^"
CO *~
& -
LO ,-yJ
05 „
CM CO
CO Y-
CM Y-
co '-
S -
^ Y-
0 0
co co
XPS boardstock: HFC-134
to COz/alcohol
One-component: HFC-134
toHC
o
o
0
o
o
o
0
o
o
o
o
o
0
CD
0
o
Y-
-
Y-
••-
CD
CD
O
O
CO
SJ
6
LJ_
X
1
Ix1
i o
a
o
CD
O
CD
O
CD
0
CD
CD
O
CD
O
CD
O
O
O
O5
O
Y—
cn
oo
Y-
Y-
Y-
Y-
PU continuous and
discontinuous panel:
HFC-134atoHC
CO
C
.0
"E
CD
C
1
o
0
CD
E
CO
.0
g
Q>
t
C.
CD
Q.
03
%
•o
1
2
Q.
05
c:
g
"5.
o
"o
'Fi
CO
o
Note: Assumed technical appli
IV-92
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
a.
a.
CO
o
'a
O
Q)
+•*
.O
Ic
,0
CO
.0
'co
CO
UJ
1
"o
0>
en
CD
c
Q>
0
^
CD
CO
CD
1
CO
2?
Q.
X
UJ
_o
f?
1
a!
"3
*£
CO
E
i
"x
CD
ble 4-5: Incremental M
K-
"co1
x !E
go
58
11
o
o
w
UJ
o
I
L^
o
J~.S
•g
II
^J
5
c
2.
CO
g.
g
UJ
CO
s
CO
^3
.«
C
0202
9U)2
0102
9002
0202
9L02
0102
9002
0202
9102
0102
9002
0202
91-02
01-02
9002
0202
9102
0102
9002
0202
9102
0102
9002
o
••s.
0
o
CO
8
8
8
CD
CO
CD
S
s
g
o
m
oo
m
CO
o
g
in
oo
in
CO
g
o
m
oo
CO
o
8
in
CM
o
0
CO
1
8
«
Q.
CO"^
Q-
8
O
CO
in
CO
m
CO
o
g
in
oo
m
CO
8
8
8
8
o
CD
8
m
oo
8
g
CD
8
m
CO
in
8
in
o
U appliance: HFC-245fa
and HFC-365mfc/HFC-
227eatoHC
Q_
CD
CD
CD
O
CD
CD
CD
CD
o
CD
CD
CD
cS
0
CD
CD
m
CO
CD
0
CD
°
un
CD
0
U appliance: automated
process with foam
grinding, HFC adsorption,
and foam landfilling
Q-
CD
O
O
O
O
O
CD
O
O
CD
O
O
0
O
O
CD
CD
O
0
CD
8
CD
CM
CD
0
U appliance: manual
process with foam
incineration
Q_
un
0
in
CD
in
o
in
o
in
CD
in
CD
in
0
m
0
in
CD
m
o
S
CM
CD
m
LJ spray: HFC-245fa and
HFC-365mfc/HFC-227ea
toHC
Q_
CD
CM
in
CO
in
°
in
oo
in
o
CM
in
CO
in
S
in
oo
in
CD
CM
m
oo
in
CD
CM
CD
in
CD
U spray: HFC-245fa, or
HFC-365mfc/HFC-227ea
toC02
Q.
O
O} O
o in
r-~ co
o o
r-- co
o o
in co
CD
o in
r^ co
CD CD
r~ o
CD CD
m co
8 |
o in
CD O
f^ CO
o o
in co
° 8
o in
O CD
(^ CO
O O
in • co
8 1
CD in
r^ co
o o
r^ en
8 8
o 8
T —
0 §
0 ^
o 8
PSboardstock:HFC-134a
to CCValcohol
ne-component: HFC-134a
toHC
X O
CD
O
in
en
8
8
8
in
en
o
o
co
8
in
o
o
en
o
CO
8
CO
CD
CO
8
8
in
en
o
en
8
8
8
o
8
ne-component: HFC-152a
toHC
o
CD
O
in
8
8
8
un
o
o
CO
o
o
in
8
8
8
un
CD
CO
8
8
T —
in
O)
8
8
8
8
CD
8
J continuous and
discontinuous panel:
HFC-134atoHC
Q.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-93
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
to
O
'co
CO
LU
"o
CD
ca
c
CU
0
CO
Q.
ca
CO
to
"8
CO
CO
£
£1
X
LU
c
75
75
cu
a.
1
CD
E
3
E
"x
OB
^>
ble 4-6: Incremental
i2
15-
gO
< CO
o c
o
O
to
UJ
o
1
o
g> $
cu '^
£ O
0
<
c
CO
a.
CO
8.
Q
UJ
CA
O
1
•Q
£
"«2
OZOZ
9LQ2
OM)2
SOOZ
OZOZ
9L02
OL02
9002
OZOZ
9LQ2
OMJZ
SOOZ
OZOZ
9L02
OL02
SOOZ
OZOZ
9LQ2
OLOZ
SOOZ
OZOZ
SLOZ
OL02
SOOZ
o
••s.
0
o
0
o
CD
O
O
CD
0
O
O
0
0
O
CD
O
CD
O
O
CD
0
CO
CO
CM
O
2
03
CO
|
S
1
Is
Q.
O
0
O
O
O
O
CD
0
CO
CO
co
CO
CD
CO
in
CO
co
CO
CM
CO
CO
CO
in
in
8
Si
j:
o
jo O
m LL.
6 -g
-i- m r t
gSx
1^5
a.-1- a>
Q. "O 1^-
CO C CM
_ CO CM
Q.
CD
CD
O
CD
CD
CD
O
CD
0
CD
0
O
CD
O
CD
O
0
CD
CD
O
.,_
CD
O
0
cf
O
U appliance automated
process with foam
grinding, HFC adsorptii
and foam landfilling
Q.
o
CD
o
o
0
0
CD
o
0
0
O
O
0
0
O
0
CD
O
CD
O
CO
O
CD
O
U appliance manual
process with foam
incineration
Q.
0
o
o
CD
O
O
0
0
-
CO
^r
CD
O
CO
CO
CD
-
1"-
•«*•
O
CO
CM
in
CO
•*
of
cu
U spray HFC-245fa and
HFC-365mfc/HFC-227
toHC
Q.
O
0
o
o
0
0
CD
O
CD
CM
r--
m
S
CD
in
CO
in
^
CO
•*
3:
r-
•*
o
CO
CD
U spray HFC-245fa, or
HFC-365mfc/HFC-227
toC02
Q.
0 S
o g?
o g
o S
SB 8
0 g>
0 0
o S
0 0
0 0
CD O
CD O
S 8
oo in
co en
1^ O
CD en
en CD
-* CO
__ CD
^ °
co in
co en
i- CD
co en
§ S
o 8
o g
o {5
o o
"o 2
s ^t
PS boardstock: HFC-
134a/C02toC02/alcof
ine-component: HFC-13
toHC
x o
o
o
0
0
o
o
0
0
o
o
o
o
o
o
CD
o
8
05
g
g
8
s
g
o
CO
>ne-component: HFC-15
toHC
o
0
CD
o
o
CD
CD
CD
O
CD
CD
O
CD
CD
0
CD
CD
CM
CO
f2
CO
8
fe
fc
8
§
U continuous and
discontinuous panel:
HFC-134atoHC
Q-
IV-94
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
~
s
0>
0^
0)
S~
CO
m
to
1
u_
15
£
§
(A
O
'•§
•o
d)
oc
to
o
'
d> o d> goo
O^O C3CMO5O OOO
o' ^ o o" •*' ui o ooo
^D ^: O ^3 CO ^J" O ^ O ^5
CO ^J ^3 i"^ CXI ^" CO f^> g*^ f^
pPp pog-^p pop
^5 TI? ^3 ^^ ^^ C\l ^5 ^3 ^3 ^3
pPo ppeqp ppp
ci^fo oooo ooo
OlOO C3U1C3'I~^ OC3O
o'crica cacoirij^j ooo
P"=tP PPCNC' °?PP
CO CO CO O CNJ CO 3^ CO CO CO
Ot-o ooom01? cnoo
O CO C) O O -r-' J^J O" O" O
C3 "*"" ^3 ^3 ^3 lO " . OO ^3 ^^
pcqo p-^co1^: «*iqin
T~
pcop pT^cvi01"! oicpcj
co oo co o CM co ^ c\i co r —
CO Is— O CO CO T— ^_; CO CO IO
pojp pp^^ ^ini^
ciirico cioco^ cbocj
mo-?— •*T-CDCO co o co
•^coo ooiirio ooo'
ocvio oiricvio o oo"
ir>"a;p poqcop cppco
OOO C3COCOCO OOCO
o'oo o'cvioo oo'o"
CO •S "O T3 T3
^* m co ^~ ^ ^ i , i • •
c?>;^ *g jg <" i *" i fS OO "o3
T 'V'tsOpES §E ^LL j5^J> u_ u_ u_ c
L£ LLEo-§£ ^ «-2 cf^ cf1 5 5 5 1 S"^
1 ^STs^^U-* £cOsSOc)!2o'go-c5a5 m w™S
§^? o> 9? '> ^C c co ~^ Q 1 1 £ ^z 1 1 £ X •*-• f^ r— cz qp d ^-^ 3 ^ co
._ ^ c^ « w £?'!-=' .§ S | ^c§ « ^c§ m 1 0 -§ §- S l-o .1 1 ^
l:x ai o§:81§il:S.l |_6 ^ |_6 ^E^^I^IcrSliS
^ S ^ w X ^ o. 5)^ J ^ §..£ ^Xcvl^XcviW^O g^ ®^.^^X
Q.Q.Q. Q.Q.Q.X OOQ-
o
I
p
o
§
0
<=>
(O
O)
p
CM
s
o
g
in
CM
0
^
si
Is-
CM
o>
cS
in
o
CO
CM
CO
CM
in
Is-;
CO
CO
CO
i
CM
CM
CO
CM
m
CO
p
CD
CM
1
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-95
-------
SECTION IV— INDUSTRIAL PROCESSES • FOAMS
Table 4-8: Summary of Abatement Option Cost Assumptions
Option
Time Base
Horizon Base One- Annual
(years): Unit of Costs Time Cost Cost
Base
Annual Net Annual
Savings Cost
Continuous and
discontinuous panel:
HFC-134atoHC
One-component: HFC-
134atoHC
One-component: HFC-
152atoHC
XPS:HFC-134a/C02to
COz/Alcohol
PU spray: HFC-
245fa/C02toC02
PU spray: HFC-
245fa/C02toHC
PU appliance: HFC-134a
toHC
PU appliance: HFC-245fa
toHC
PU appliance: automated
process with foam
grinding, HFC
adsorption, and foam
landfilling
PU appliance: manual
process with foam
incineration
25 Per contractor8 $273,473 $2,175,424 $2,164,154
$11,270
25 Per contractor" $341,841 $292,711
25 Per contractor" $341,841 $292,711
$639,673 -$346,962
$342,779 -$50,068
25 Per $5,013,674 $774,711 $3,582,474 -$2,807,764
manufacturer*
25 Per contractor'1 $4,000 $54,264
$9,724 $44,541
25 Per contractor" $13,728e $39,560
$47,060
-$7,500
25 Per factory* $50,000,000
$0 $1,506,160 -$1,506,160
25 Per factory* $50,000,000 $11,202,400"
$0h $11,202,400
25 Per facility^
$3,646,308 $1,848,159 $481,809 $1,366,350
25 Per facility'
$182,315 $456,528 $48,239 $408,289
a Based on a hypothetical PU continuous and discontinuous panel foam contractor that produces approximately 12 million pounds of foam per
year.
" Based on a hypothetical one component foam contractor that produces approximately 3.3 million pounds of foam per year.
c Based on a hypothetical XPS boardstock foam manufacturer that produces about 1 billion board-feet of foam per year.
d Based on a hypothetical PU spray foam contractor that produces about 127,000 pounds of foam per year.
e Only U.S.-based costs are presented in this summary table. However, EU-15 and Japan costs are $21,251 and $31,536 and are applied
accordingly. U.S. costs are applied to all other countries.
* Based on a hypothetical factory that manufactures 536,000 refrigerators per year.
9 Based on a facility that is assumed to process about 100,000 refrigerators per year.
h Base annual savings are incorporated into the base annual costs.
1 Based on a facility that is assumed to process 10,000 refrigerators per year.
IV-96
GLOBAL MITIGATION OF NOM-C02 GREENHOUSE GASES
-------
SECTION IV— INDUSTRIAL PROCESSES • FOAMS
Table 4-9: Country/Regional Emissions Reductions in 2010 and Breakeven Costs for Foams at 10% Discount
Rate, 40% Tax Rate (S/tC02eq)
2010
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.00
2.06
0.00
0.00
0.02
0.00
1.06
0.00
0.77
0.00
0.03
2.03
0.02
0.00
0.19
2.08
$15
0.00
2.41
0.00
0.00
0.02
0.01
1.40
0.00
0.77
0.00
0.03
2.39
0.02
0.00
0.21
2.43
$30
0.00
2.41
0.00
0.00
0.02
0.01
1.40
0.00
0.77
0.00
0.03
2.39
0.02
0.00
0.21
2.43
$45
0.00
2.66
0.00
0.00
0.02
0.01
1.49
0.00
0.82
0.00
0.03
2.64
0.02
0.00
0.31
2.69
$60
0.00
2.66
0.00
0.00
0.02
0.01
1.49
0.00
0.82
0.00
0.03
2.64
0.02
0.00
0.31
2.69
>$60
0.00
3.40
0.02
0.00
0.02
0.01
1.95
0.00
0.92
0.00
0.03
3.38
0.02
0.00
0.39
3.43
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Table 4-10: Country/Regional Emissions Reductions in 2020 and Breakeven Costs for Foams at 10%
Discount Rate, 40% Tax Rate ($/tC02eq)
2020
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.00
4.57
0.01
0.00
0.05
0.01
2.00
0.00
1.37
0.00
0.05
4.52
0.04
0.00
1.10
4.62
$15
0.00
5.49
0.01
0.00
0.05
0.01
2.86
0.00
1.37
0.00
0.05
5.44
0.04
0.00
1.17
5.54
$30
0.00
5.49
0.01
0.00
0.05
0.01
2.86
0.00
1.37
0.00
0.05
5.44
0.04
0.00
1.17
5.54
$45
0.00
7.09
0.02
0.00
0.05
0.01
3.34
0.00
1.61
0.00
0.05
7.04
0.04
0.00
1.98
7.14
$60
0.00
7.09
0.02
0.00
0.05
0.01
3.34
0.00
1.61
0.00
0.05
7.04
0.04
0.00
1.98
7.14
>$60
0.00
8.75
0.05
0.00
0.05
0.01
4.17
0.00
1.77
0.00
0.05
8.71
0.04
0.00
2.47
8.81
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-97
-------
SECTION IV— INDUSTRIAL PROCESSES • FOAMS
Table 4-11: World Breakeven Costs and Emissions Reductions in 2020 for Foams
Cost
(2000S/tC02eq)
DR=10%,
TP^ Mt\A/
TR=40%
Reduction Option
XPS boardstock: HFC-134a/C02
(LCD)— based blends to C02
(LCD)/alcohol
PU Spray: HFC-245fa/C02
(water) to HC
PU one-component HFC-134a to
HC
PU one-component HFC-152a to
HC
PU continuous and
discontinuous: HFC-134a to
HC
PU Appliance: automated
process with foam grinding,
HFC adsorption, and foam
landfilling
PU Spray: HFC-245fa/C02
(water) to C02 (water)
PU appliance: HFC-134a to HC
PU appliance: manual process
with foam incineration
PU appliance: HFC-245fa to HC
Low
-$7.81
-$5.19
-$1.76
-$0.15
$0.86
$36.07
$41.84
$42.06
$82.54
$192.54
High
-$7.81
-$2.91
-$1.76
-$0.15
$0.86
$36.07
$41.84
$42.06
$82.54
$192.54
Mil
Emissions
Reduction
of Option
(MtC02eq)
2.49
1.59
0.48
0.06
0.92
0.01
1.42
0.17
0.04
1.62
Reduction
from 2020
Baseline
8.7%
5.5%
1.7%
0.2%
3.2%
0.0%
5.0%
0.6%
0.1%
5.7%
Running
Sum of
Reductions
(MtC02eq)
2.49
4.08
4.56
4.62
5.54
5.55
6.98
7.14
7.18
8.81
Cumulative
Reduction
from 2020
Baseline
8.7%
14.2%
15.9%
16.1%
19.3%
19.4%
24.4%
24.9%
25.1%
30.7%
Market Penetrations
Market penetrations of abatement technologies are based on published reports and discussions with
industry experts in the U.S. and EU-15. However, actual market penetrations of these technologies may
be different. For example, market penetration rates and hence emissions reduction potentials may be
higher in countries that are establishing climate policies.
PU Continuous and Discontinuous Panel Foam
For PU continuous and discontinuous panel foam, the cost analysis was performed based on a
contractor that uses HFC-134a; however, this end-use also uses HFC-245fa, and HFC-365mfc/HFC-227ea.
XPS Boardstock Foam
Capital and annual costs for abatement technologies of XPS boardstock foam were based on a
consensus reached among industry representatives or averages of different estimates provided by
different manufacturers. These averages may not reflect the full range of costs that might be experienced.
IV-98
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
Figure 4-2: 201 0 MAC for Foams, 1 0% Discount Rate, 40% Tax Rate
$250-
„ $200-
c
o
'•§ ^$150-
K 6 $100-
w O
C 4i
|e $50-
E
u $0-
0
C U'
0
-$50 J
i
—
>
0.5
4 «
— * China
Rest of the world
— • United States
-- EU-15
*• Japan
» Other OECD
1.0 1.5 2.0 2.5
Cumulative Emissions Reductions (MtCO2eq)
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Figure 4-3: 2020 MAC for Foams, 10% Discount Rate, 40% Tax Rate
$250 -f
o
$200 -
$150 -
11
^ O $100 -
IM
|e $50 H
E
"J $0
-$50
China
Rest of the world
•* United States
- EU-15
* Japan
« Other OECD
0.5 1.0 1.5 2.0 2.5 3.0 3.5
Cumulative Emissions Reductions (MtCO2eq)
4.0
4.5
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
IV.4.5 Summary
Baseline emissions of HFCs from foams are estimated to grow from 1.5 to 28.6 MtCO2eq between
2000 and 2020. In 2020, OECD countries are assumed to account for almost 100 percent of the emissions,
while U.S. emissions and EU-15 emissions are each assumed to account for about 40 percent of this total.
The largest emissions growth is expected in the United States, from 0.3 MtCO2eq in 2000 to 11.3 MtCO2eq
in 2020.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-99
-------
SECTION IV — INDUSTRIAL PROCESSES • FOAMS
This analysis considers the following eight replacement emissions mitigation options for PU spray,
PU appliance, XPS boardstock, PU continuous and discontinuous panel foam, and one-component foams,
as well as the following two end-of-life options for PU appliance foams:
• replacing HFC-134a, HFC-245fa, and HFC-365mfc/HFC-227ea with HCs in PU continuous and
discontinuous panel foam;
• replacing HFC-134a with HCs in one-component foam;
• replacing HFC-152a with HCs in one-component foam;
• replacing HFC-134a/CO2 (LCD) with CO2 (LCD)/alcohol in XPS boardstock foam;
• replacing HFC-245fa/CO2 (water) and HFC-365mfc/HFC-227ea with CO2 (water) in PU spray
foam;
• replacing HFC-245fa/CO2 (water) and HFC-365mfc/HFC-227ea with HCs in PU spray foam;
• replacing HFC-134a with HCs in PU appliance foam;
• replacing HFC-245fa and HFC 365mfc/HFC-227ea with HCs in PU appliance foam;
• end-of-life PU appliance foam practice: automated process with foam grinding, HFC adsorption,
and foam landfilling in PU appliance foam; and
• end-of-life PU appliance foam practice: manual process with foam incineration in PU appliance
foam.
The emissions reduction benefits of each option were compared in each region. For spray end-uses,
the costs associated with converting to alternative blowing agents differ between the United States, EU,
and Japan. The costs per tCO2eq of all other abatement options for these three regions are equivalent
because available data on costs for abatement technologies were not scaled to reflect potential differences
in the costs internationally. Additional research may be required to determine actual variability in costs
across regions. This analysis shows that there are several cost-effective options available at the 10 percent
discount rate and 40 percent tax rate that may be used to eliminate the use of HFCs and reduce HFC-
associated emissions from foams.
IV.4.6 References
Alliance for Responsible Atmospheric Policy. May 26, 2000. Comments of the Alliance for Responsible
Atmospheric Policy on Draft of "Cost and Emission Reduction Analysis of HFC Emissions from
Foams in the United States." Fax sent from Alliance to ICF Consulting.
Alliance for Responsible Atmospheric Policy. May 16, 2001. Review of the USEPA Draft Chapter 9 by
Members of the Alliance for Responsible Atmospheric Policy.
Ashford, Paul. April 8, 2004. Personal communication between ICF Consulting and Paul Ashford of Caleb
Group.
Honeywell. September 11, 2000. Comments of Honeywell Inc. on U.S. Environmental Protection Agency's
(USEPA's) proposed listing of certain HCFCs and blends as "unacceptable" substitutes for HCFC-
141b—65 Fed. Reg. 42653 (11 July 2000). Personal communication from Richard Ayres of Howrey,
Simon, Arnold, and White to Anhar Karimjee of the USEPA. Available from the USEPA's Foams
Docket A-200-18, Document IV-D-41.
ICF Consulting. March 2004. Personal communication with Bob Russell, ICF Consulting.
Intergovernmental Panel on Climate Change (IPCC). 2004. "Special Report on Ozone and Climate."
Second-Order-Draft. Technology and Economic Assessment Panel.
IV-100 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES « FOAMS
JACO Environmental. May 13, 2004. Personal Communication between Michael Dunham, Director,
Energy & Environmental Programs, JACO Environmental, Inc., and Colm Kenny of U.S.
Environmental Protection Agency (USEPA).
Scheutz, C, and P. Kjeldsen. April 2002. "Determination of the Fraction of Blowing Agent Released from
Refrigerator/Freezer Foam After Decommissioning the Product." Denmark: Environment and
Resources DTU, Technical University of Denmark.
Scheutz, C., and P. Kjeldsen. August 2003. "Attenuation of Alternative Blowing Agents in Landfills."
Denmark: Environment and Resources DTU, Technical University of Denmark.
United Nations Environment Programme (UNEP). 1998. "1998 Report of the Flexible and Rigid Foams
Technical Options Committee." UNEP.
United Nations Environment Programme (UNEP). 2002a. "Report of the Technology and Economic
Assessment Panel. Progress Report. Montreal Protocol on the Substances that Deplete the Ozone
Layer." UNEP.
United Nations Environment Programme (UNEP). 2002b. "Report of the Technology and Economic
Assessment Panel of the Montreal Protocol, Task Force on Destruction Technologies, Volume 3b,
April." UNEP.
U.S. Department of Agriculture (USDA). 2002. "Real GDP (2000 dollars) Historical. International
Macroeconomic Data Set. Available at .
U.S. Energy Information Administration (USEIA). 2001. "International Energy Outlook, Table 7.
Comparison of Economic Growth Rates by Region, 1997-2020." Washington: U.S. Department of
Energy, Energy Information Agency.
Werkema, T. April 2006. Comments provided by Tom Werkema to the USEPA.
Whirlpool. April 19, 2004. Personal communication between Robert W. Johnson of Whirlpool and Colm
Kenny of the USEPA.
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, pp. 290-302.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-101
-------
SECTION IV— INDUSTRIAL PROCESSES • FOAMS
IV-102 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • AEROSOLS
IV.5 HFC Emissions from Aerosols
IV.5.1 Introduction
erosol propellants are used in metered dose inhalers (MDIs), as well as a variety of consumer
products. Historically, the majority of aerosol applications have used CFCs as propellants;
however, efforts have been made to transition away from CFC propellants. As a result of
initiatives under the Montreal Protocol, many pharmaceutical companies that produce MDIs have
committed to develop alternatives to CFC-based MDIs. Furthermore, many consumer products, such as
spray deodorants and hair sprays, and specialty aerosol uses, such as freeze spray and dust removal
products, have successfully been reformulated with HC propellants or replaced with NIK substitutes
such as pump sprays or solid and roll-on deodorants. Such transitions occurred in the United States as far
back as 1977, when the country placed a ban on CFC propellants in non-MDI aerosols for nonessential
uses.
Various HFCs have also been introduced as alternative propellants in aerosol applications. These
MFCs include HFC-134a, HFC-152a, and HFC-227ea and are associated, respectively, with 100-year
GWPs of 1,300, 140, and 2,900. Aerosol HFCs are emitted from pharmaceutical products (primarily
MDIs)1 and consumer and industrial products (primarily specialty aerosols).
The pharmaceutical aerosol industry is actively working to develop HFC-propellant MDIs, a type of
inhaled therapy used to treat asthma and chronic obstructive pulmonary disease (COPD). The earliest
non-CFC substitute products used HFC-134a, but eventually the industry expects products to use HFC-
227ea as well. In addition to MDIs that use propellants, dry powder inhalers (DPIs) can be used as a
substitute for some MDIs. Because MDIs are medical devices, substitute propellants must meet far stricter
performance and toxicology specifications than required for most other products. In the United States, for
example, the Food and Drug Administration (FDA) must approve MDIs reformulated with an alternative
propellant before they can enter the market.
Chemical manufacturers are also marketing HFCs, especially HFC-152a and HFC-134a, as aerosol
propellants in consumer products, primarily for use in specialty applications. This is particularly true for
applications where flammability or volatile organic compound (VOC) emissions and their impact on
urban air quality are of concern. If HFC use is accelerated, public concern over these emissions may
increase. This concern will likely spur the aerosol industry to promote responsible use of these chemicals,
for instance, by implementing emissions abatement options examined in this report (UNEP, 1999).
IV.5.2 Baseline Emissions Estimates
IV.5.2.1 Emissions Estimating Methodology
Description of Methodology
Specific information on the emissions model used to calculate ODS substitute emissions from all
sectors calculates aerosol emissions is described below.
The USEPA uses a detailed Vintaging Model of ODS-containing equipment and products to estimate
the use and emissions of various ODS substitutes in the United States, including HFCs. Emissions
This analysis excludes non-MDI aerosols produced by the pharmaceutical industry such as bandage sprays.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-103
-------
SECTION IV — INDUSTRIAL PROCESSES • AEROSOLS
baselines from non-U.S. countries were derived using country-specific ODS consumption estimates as
reported under the M ontreal Protocolin conjunction with Vintaging Model output for each ODS-
consuming end-use sector. For sectors where detailed information was available, these data were
incorporated into country-specific versions of the Vintaging Model to customize emissions estimates. In
the absence of country-level data, these preliminary estimates were calculated by assuming that the
transition from ODSs to HFCs and other substitutes follows the same general substitution patterns
internationally as observed in the United States. From this preliminary assumption, emissions estimates
were then tailored to individual countries or regions by applying adjustment factors to U.S. substitution
scenarios, based on relative differences in economic growth, rates of ODS phaseout, and the distribution
of ODS use across end-uses in each region or country.
Emissions Equations
All HFCs used in aerosols are assumed to be emitted in the year of manufacture. Since there is
currently no aerosol recycling, all of the annual production of aerosol propellants is assumed to be
released to the atmosphere. The following equation describes the emissions from the aerosols sector:
£/ = Qcy, (5.1)
where
Ej = Total emissions of a specific chemical in a year j from use in aerosol products, by weight.
QCJ = Total quantity of a specific chemical contained in aerosol products sold in the year /, by
weight
j = Year of emissions
For aerosols, two separate baseline emissions were created; one baseline tracks HFC emissions from
the MDI industry, while the other estimates HFC emissions from consumer and specialty products (i.e.,
non-MDI aerosols).
Regional Adjustments
The adjustment factor assumptions used in the global aerosol emissions estimating methodology
include both economic and timing adjustment factors. The timing factors reflect that some nations are not
moving at the same pace away from using CFCs and toward using HFCs as other nations are. For all ODS
end-uses, by 2005, non-Annex I (i.e., developing) countries are assumed to be 75 percent through the CFC
transition, and by 2010, the CFC transition should be complete. These timing factors are partially offset by
generally higher growth rates in developing countries.
In addition, the methodology used to estimate global aerosol emissions includes an adjustment
specific to non-MDI aerosols. This adjustment was necessary because the ban on CFC use in aerosols
caused the United States to transition out of CFCs earlier than other countries. Therefore, the unweighted
U.S. consumption of non-MDI aerosol ODS substitutes (including a large market segment that
transitioned into NIK or HC substitutes) was used as a proxy for U.S. 1990 non-MDI ODS consumption.
For countries other than the United States, it was then assumed that 15 percent of the non-MDI aerosols
ODS consumption transitioned to HFCs, while the remainder is assumed to transition to NIK or HC
alternatives.
IV-104 GLOBAL MITIGATION OF NOM-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • AEROSOLS
IV.5.2.2 Baseline Emissions
Table 5-1 and Figure 5-1 display total HFC emissions estimates in million metric tons of carbon
dioxide equivalent (MtCO2eq) for the MDI aerosol sector, while Table 5-2 and Figure 5-2 represent the
non-MDI aerosols sector. Both HFC-134a and HFC-227ea are expected to be emitted from using MDIs in
the future as substitutes for CFCs. The MDI emissions baseline accounts for all emissions of HFC-227ea
from the aerosols sector. Non-MDI emissions are responsible for the majority of the HFC-134a emissions
from the aerosols sector (mainly for specialty applications) and all of the HFC-152a emissions (mostly
formulated consumer products).
IV.5.3 Cost of HFC Emissions Reductions for Aerosols
This section presents a cost analysis for achieving HFC emissions reductions from the emissions
baselines presented in Tables 5-1 and 5-2. The cost analysis for the MDI option assumes a 15-year project
lifetime; all cost analyses for the non-MDI emissions reduction options assume a 10-year project lifetime.
Each abatement option is described below.
IV.5.3.1 Description and Cost Analysis of Abatement Options
Four potential mitigation options are analyzed in this report. The first mitigation option has the
potential to abate emissions from the MDI baseline (Table 5-1), while the other three options have the
potential to abate emissions from the non-MDI baseline (Table 5-2). The options are as follows:
• MDI replacement with DPIs (DPI [MDI])
• non-MDI replacement with lower GWP HFCs (HFC-134a to HFC-152a [non-MDI])
• non-MDI replacement with NIK alternatives (HFC to NIK [non-MDI])
• non-MDI replacement with HC aerosol propellants (HFC to HC [non-MDI])
DPIs have been authorized as a substitute for some HFC-propellant MDIs. The non-MDI baseline
includes emissions from specialty aerosol uses such as tire inflators, electronics cleaning products, dust
removal, freeze spray, signaling devices, and mold release agents, as well as consumer products such as
hairsprays, mousse, deodorants and antiperspirants, household products, and spray paints (Arthur D.
Little, 1999). HFCs are currently used when flammability issues cannot easily be overcome, such as tire
inflators and air signaling horns that use HFC-134a to avoid potential explosivity associated with highly
flammable propellants like propane or butane (Arthur D. Little, 1999). HFC-152a has been used in dusters
since 1993 (UNEP, 1999), and as a replacement for HFC-134a in general aerosol applications. Converting
to HFC-152a in these applications is a reduction strategy that has had significant success thus far and is
expected to continue. The other options to reduce HFC emissions from non-MDI aerosol applications
addressed in this analysis include NIK replacement and HC aerosol propellants. Other options, such as
using carbon dioxide as a propellant, may also exist but have not been addressed in this analysis because
specific information is lacking.
The remainder of this section describes the economic assumptions for these four abatement options.
A detailed description of the cost and emissions reduction analysis for each option can be found in
Appendix I for this chapter.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-105
-------
SECTION IV — INDUSTRIAL PROCESSES • AEROSOLS
Table 5-1: Total Baseline HFC Emissions Estimates from MDI Aerosols (MtC02eq)
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex I
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
0.0
2.8
0.2
0.0
0.0
0.1
1.9
0.0
0.1
0.0
0.4
2.4
0.4
0.0
0.1
2.9
2010
0.4
8.5
0.3
0.1
0.7
0.3
2.8
0.2
0.9
0.3
1.3
7.6
1.1
0.3
2.7
11.0
2020
0.9
13.8
0.4
0.2
2.2
0.4
3.8
0.6
1.6
0.7
1.8
12.8
1.5
0.8
5.5
20.1
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Figure 5-1: Total Baseline HFC Emissions Estimates from MDI Aerosols (MtC02eq)
1990
2000
2010
2020
Year
n Middle East
• Africa
• Non-EU FSU
• Latin America
• S&E Asia
n China/CPA
• OECD90+
CPA = Centrally Planned Asia; Non-EU FSU = non-European Union Former Soviet Union countries; S&E Asia = Southeast Asia; OECD90+:
Organisation for Economic Co-operation and Development.
IV-106
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • AEROSOLS
Table 5-2: Total Baseline HFC Emissions Estimates from Non-MDI Aerosols (MtC02eq)
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
0.0
24.1
0.7
0.0
0.0
1.1
10.5
0.0
0.0
0.0
2.1
22.0
1.6
0.0
9.9
24.2
2010
0.0
32.6
0.9
0.0
0.0
2.9
11.8
0.0
0.0
0.0
5.5
27.1
4.2
0.0
12.1
32.7
2020
0.0
39.4
1.1
0.0
0.0
4.0
13.3
0.0
0.0
0.0
7.2
32.2
5.5
0.0
14.8
39.5
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Figure 5-2: Total Baseline HFC Emissions Estimates from Non-MDI Aerosols (MtC02eq)
0)
I/I
c
o
8
E
UJ
2000
2010
2020
Year
H Middle East
• Africa
• Non-EU FSU
B Latin America
• S&E Asia
n China/CPA
• OECD90+
CPA = Centrally Planned Asia; Non-EU FSU = non-European Union Former Soviet Union countries; S&E Asia = Southeast Asia; OECD90+:
Organisation for Economic Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-107
-------
SECTION IV — INDUSTRIAL PROCESSES • AEROSOLS
MDI: Replacement with Dry Powdered Inhalers (DPIs)
As MDIs transition away from CFC use, alternatives such as HFC propellants, DPIs, and oral
medications are being developed. Although HCs have replaced CFCs as propellants in many commercial
aerosols, they have been found to be unacceptable for use in MDIs (International Pharmaceutical Aerosol
Consortium [IPAC], 1999). Given the unique medical requirements for developing MDIs, and the fact that
the industry has been investing heavily in the development of HFC technologies, an aerosol replacement
for HFC-based MDIs is unlikely to be developed within the time frame of this analysis. Globally, the
number of HFC MDIs used has grown to more than 100 million in 2001 (UNEP, 2002). Rather than
developing new alternatives that use HFCs, some MDI patients may turn to DPIs, oral medication, or
other NIK alternatives. In 2001, the number of multidose DPIs used worldwide was estimated at 65
million (UNEP, 2002).2 This analysis examines the option of further replacing HFC-based MDIs with DPIs
because of its technical feasibility and demonstrated success in the MDI market.
DPIs are a viable option with most anti-asthma drugs, although they are not successful with all
patients or all drugs. Micronised dry powder can be inhaled and deposited in the lungs from DPIs as
with MDIs, but 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 (March Consulting Group, 1999; UNEP, 2002). Other issues that doctors and patients
consider when choosing a treatment device include the patient's manual dexterity, ability to adapt to a
new device, and perception of the effectiveness of the medicine, and the taste of any added ingredients
(Price et al., 2004). It is important to note that the choice of treatment, including the type of propellant
used in MDIs, is a medical decision involving the pharmaceutical industry, FDA or other regulatory
authority, and ultimately doctors and their patients. Doctors and their patients will be involved in
selecting the method of therapy, treatment regimen, and type of device(s) and active ingredients(s) that
will prove most effective for particular individuals (IPAC, 1999).
In 1998, DPI use was estimated to represent 17 percent of all inhaled medication (i.e., inhaler units)
worldwide and had increased to 27 percent by 2002 (UNEP, 2002). DPIs may represent a viable
alternative, as suggested by their increased use in Europe; for example, in Holland they account for more
than 65 percent of inhaled medication (UNEP, 2002). The use of newly available DPIs is on the rise in the
United States, where DPIs made up 14 percent of the total U.S. market share as of mid-2002 (UNEP, 2002).
There is also a trend toward developing a broad range of oral treatments that would be swallowed, rather
than inhaled and may be introduced over the next 10 to 20 years. These new medications may affect MDI
use, although they will not likely replace inhaled MDI therapy entirely.
This analysis assumes that DPIs are technically applicable3 to all HFC emissions from MDIs.
However, because of the limitations in their use for severe asthma patients and young children, and the
difficulties experienced in hot and humid climates, this analysis assumes a global incremental maximum
market penetration into the HFC-based MDI market of 0 percent in 2005, increasing up to 50 percent in
2020 (Table 5-3). DPIs do not use HFCs; hence, they have a 100 percent reduction efficiency. To the extent
2 Multiple-dose DPIs contain premeasured doses that provide treatment for a day or up to 1 month. Single-dose DPIs
are also available for which only one dose can be loaded at a time (UNEP, 2002).
3 In this report, the term "technically applicable" refers to the emissions to which an option can theoretically be
applied. Because DPIs can eliminate emissions from MDIs, they are technically applicable to all MDI emissions but
are not technically applicable to non-MDI emissions. Other factors will affect their application and the market
penetration assumed in this analysis.
IV-108 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • AEROSOLS
that health and technical concerns are adequately met, a transition in inhalation therapy away from
propellant MDIs and toward NIK alternatives may occur over the next 10 to 20 years. The rapidity at
which these changes will occur depends on product development cycles (generally about 10 years), cost-
effectiveness, and manufacturing capacity (UNEP, 1999).
Non-MDI: Replacement with Lower GWP MFCs
Replacing higher GWP HFCs, such as HFC-134a, with a lower GWP HFC, such as HFC-152a, has the
potential to greatly reduce emissions from the non-MDI aerosols sector. 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 (UNEP, 2002) but may present problems
for other applications. This analysis assumes that converting to HFC-152a is technically applicable to all
emissions of HFC-134a from the non-MDI baseline but is only adopted by some users. Non-MDI
emissions of HFC-134a are calculated by the Vintaging Model to be 83 percent of total GWP-weighted
non-MDI aerosol emissions. As shown in Table 5-3, the incremental maximum market penetration of this
alternative is assumed to increase from 10 percent in 2005 to 50 percent in 2020. Because HFC-152a has a
GWP of 140 (versus a GWP of 1,300 for HFC-134a), this substitution has an emissions reduction efficiency
of 89.2 percent (i.e., the difference of the GWPs divided by the GWP of HFC-134a).
Non-MDI: Replacement with NIK Alternatives
NIK aerosol replacements include finger/trigger pumps, powder formulations, sticks, rollers,
brushes, nebulizers, and bag-in-can/piston-can systems. These systems often prove to be better and more
cost-effective options than HFC-propelled aerosols, particularly in areas where a unique HFC property is
not specifically needed for a certain end-use. NIKs already occupy a sizable share of markets where they
were introduced during the initial CFC phaseout. Since NIK products have already assumed much of the
available non-MDI HFC aerosol market share, an incremental maximum market penetration of 5 percent
was assumed in 2005 and 10 percent for years 2010, 2015, and 2020 (see Table 5-3). The analysis assumes
that this option is technically applicable to all non-MDI emissions and has a reduction efficiency of 100
percent. The GWP of 538 was used to represent both HFCs being abated and was calculated using the
weighted average of the U.S. HFC-134a and HFC-152a baseline emissions.
Non-MDI: Replacement with Hydrocarbon Aerosol Propellants
HC aerosol propellants are usually mixtures of propane, butane, and isobutane. Their primary
advantage lies in their affordability; the price of HC propellants is less than one-tenth that of HFCs. The
main disadvantages of HC aerosol propellants are flammability and VOC emissions concerns. HCs
contribute to ground-level ozone and smog and therefore may be regulated in some areas. In applications
and markets where flammability and/or VOC emissions are less of a concern, HCs already hold a sizable
share. Since HC aerosol propellants have already penetrated a significant amount of the market, further
penetration is limited because of flammability and VOC concerns. Hence, this analysis assumes an
incremental maximum market penetration of 5 percent in 2005, expanding to 10 percent in later years.
The analysis also assumes that converting to HCs is technically applicable to all non-MDI emissions, but
that various factors including the flammability of HCs will limit the market penetration of this option.
The reduction efficiency of this abatement option is taken to be 100 percent, since the HFC is completely
replaced by an HC propellant with a very low GWP. The GWP of 538 was used to represent both HFCs
being abated and was calculated using the U.S. weighted average of the HFC-134a and HFC-152a
baseline emissions.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-109
-------
SECTION IV — INDUSTRIAL PROCESSES • AEROSOLS
IV.5.3.2 Summary of Technical Applicability, Market Penetration, and Costs of Abatement
Options
Table 5-3 summarizes the technical applicability and incremental maximum market penetration of
the aerosol options presented in the discussions above.
Table 5-3: Technical Applicability and Incremental Maximum Market Penetration of Aerosol Options
(Percent)3
Technical Applicability
Option (All Years)
DPI (MDI)b
HFC to HC (non-MDI)
HFC to NIK (non-MDI)
HFC-134a to HFC-152a (non-MDI)
100%
100%
100%
83%c
r Incremental Maximum Market Penetration
2005
0%
5%
5%
10%
2010
5%
10%
10%
25%
2015
20%
10%
10%
35%
2020
50%
10%
10%
50%
a
Assumed maximum market penetration of options is presented as a percentage of total sector emissions for which the options are applicable.
The baseline market penetration is assumed to be zero to assess the emissions reductions possible due to increased use of each option.
b Assumptions are separated by the line to reflect that the MDI option addresses different baseline emissions than the non-MDI options.
c Based on percentage of non-MDI aerosol emissions as determined by the Vintaging Model.
To calculate the percentage of emissions reductions off the applicable (i.e., MDI or non-MDI) aerosols
baseline for each abatement option, the technical applicability, is multiplied by the market penetration
value, and by the reduction efficiency of the option. For example, to determine the percentage reduction
off the 2020 baseline for the conversion of HFC-134a aerosols to HFC-152a, the following calculation is
performed:
Technical applicability x Market penetration in 2020 x Reduction efficiency
83% x 50% x 89.2% = 37.0%
Thus, using the assumptions in this analysis, converting from HFC-134a to HFC-152a could reduce
over one-third of the non-MDI emissions baseline in 2020. This value, along with the other emissions
reduction potentials, is shown in Table 5-4.
Table 5-4: Emissions Reductions Off the Total Applicable Aerosols Baseline (Percent)
Option
DPI (MDI)a
HFC to HC (non-MDI)
HFC to NIK (non-MDI)
HFC-134a to HFC-152a (non-MDI)
2005
0.0
5.0
5.0
7.4
2010
5.0
10.0
10.0
18.5
2015
20.0
10.0
10.0
25.9
2020
50.0
10.0
10.0
37.0
a
Calculated percentages are separated by the line to reflect that the MDI option addresses different baseline emissions than the non-MDI
options.
IV-110 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • AEROSOLS
Table 5-5 summarizes the cost assumptions used for the aerosol options presented in the discussion
above.
Table 5-5: Summary of Abatement Option Cost Assumptions
Option
DPI (MDI)
HFC to HC (Non-
MDI)
HFC to NIK
(Non-MDI)
HFC-134ato
HFC-1523 (Non-
MDI)
Time
Horizon
(Years)
15
10
10
10
Unit of Costs
Per metric ton of abated
substance
Per10,000,000cans/yr
requiring 2 oz. propellant
each
Perl 0,000,000 cans/yr
requiring 2 oz. propellant
each
Per 10,000,000 cans/yr
requiring 2 oz. propellant
each
Base One-
Time Cost
(2000$)
$0
$325,000
$250,000
$500,000
Base
Annual
Cost
(2000$)
$571,400
$0
$500,000
$0
Base
Annual
Savings
(2000$)
$0
$2,001,456
$2,343,458
$1,090,257
Net Annual
Costs
(2000S/yr)
$571,400
-$2,001,456
-$1,843,458
-$1,090,257
IV.5.4 Results
IV.5.4.1 Data Tables and Graphs
Tables 5-6 through 5-9 provide a summary of the potential emissions reduction opportunities at
associated breakeven costs in 15-dollar increments at a 10 percent discount rate (DR) and 40 percent tax
rate (TR). Tables 5-6 and 5-7 present the results for MDI Aerosols, for 2010 and 2020, respectively. As
shown, in 2010 and 2020, emissions reduction opportunities are only available at a breakeven cost greater
than 60 dollars per tCO2eq for all regions. A world total emissions reduction of 0.55 MtCO2eq is projected
by 2010 and 10.06 MtCO2eq by 2020, both at a breakeven cost greater than $60/tCO2eq.
Tables 5-8 and 5-9 present the results for the Non-MDI Aerosols sector for 2010 and 2020,
respectively. In contrast to the MDI aerosol sector, the non-MDI sector emissions reduction opportunities
are available at the lowest breakeven cost of $0/tCO2eq for several regions. A total emissions reduction of
12.6 MtCO2eq is projected by 2010 and 22.54 MtCO2eq by 2020, both at a breakeven cost below 1 dollar
per tCO2eq.
In Table 5-10, the costs, in 2000$, to reduce tCO2eq are presented for a discount rate scenario of 10
percent and a tax rate of 40 percent. Within the options that address non-MDI emissions, the results are
ordered by increasing costs per tCO2eq. Additionally, the emissions reduced by the option, in MtCO2eq
and percent of the aerosols (either MDI or non-MDI) baseline, as well as cumulative totals of these two
figures are presented.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-111
-------
SECTION IV — INDUSTRIAL PROCESSES • AEROSOLS
Table 5-6: Country/Regional Emissions Reductions in 2010 and Breakeven Costs for MDI Aerosols at 10%
Discount Rate, 40% Tax Rate ($/tC02eq)
2010
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
OECD Annex I
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$15
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$30
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$45
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$60
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
>$60
0.02
0.43
0.01
0.00
0.04
0.01
0.14
0.01
0.05
0.01
0.06
0.38
0.06
0.02
0.14
0.55
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Table 5-7: Country/Regional Emissions Reductions in 2020 and Breakeven Costs for MDI Aerosols at 10%
Discount Rate, 40% Tax Rate ($/tC02eq)
2020
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$15
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$30
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$45
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$60
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
>$60
0.43
6.91
0.18
0.08
1.08
0.22
1.88
0.29
0.82
0.34
0.89
6.40
0.77
0.39
2.74
10.06
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
IV-112 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • AEROSOLS
Table 5-8: Country/Regional Emissions Reductions in 2010 and Breakeven Costs for Non-MDI Aerosols at
10% Discount Rate, 40% Tax Rate ($/tC02eq)
2010
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.01
12.56
0.34
0.00
0.01
1.13
4.55
0.00
0.00
0.00
2.12
10.45
1.61
0.00
4.67
12.60
$15
0.01
12.56
0.34
0.00
0.01
1.13
4.55
0.00
0.00
0.00
2.12
10.45
1.61
0.00
4.67
12.60
$30
0.01
12.56
0.34
0.00
0.01
1.13
4.55
0.00
0.00
0.00
2.12
10.45
1.61
0.00
4.67
12.60
$45
0.01
12.56
0.34
0.00
0.01
1.13
4.55
0.00
0.00
0.00
2.12
10.45
1.61
0.00
4.67
12.60
$60
0.01
12.56
0.34
0.00
0.01
1.13
4.55
0.00
0.00
0.00
2.12
10.45
1.61
0.00
4.67
12.60
>S60
0.01
12.56
0.34
0.00
0.01
1.13
4.55
0.00
0.00
0.00
2.12
10.45
1.61
0.00
4.67
12.60
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Table 5-9: Country/Regional Emissions Reductions in 2020 and Breakeven Costs for Non-MDI Aerosols at
10% Discount Rate, 40% Tax Rate ($/tC02eq)
2020
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.01
22.46
0.61
0.00
0.03
2.25
7.60
0.01
0.00
0.01
4.12
18.36
3.11
0.01
8.43
22.54
$15
0.01
22.46
0.61
0.00
0.03
2.25
7.60
0.01
0.00
0.01
4.12
18.36
3.11
0.01
8.43
22.54
$30
0.01
22.46
0.61
0.00
0.03
2.25
7.60
0.01
0.00
0.01
4.12
18.36
3.11
0.01
8.43
22.54
$45
0.01
22.46
0.61
0.00
0.03
2.25
7.60
0.01
0.00
0.01
4.12
18.36
3.11
0.01
8.43
22.54
$60
0.01
22.46
0.61
0.00
0.03
2.25
7.60
0.01
0.00
0.01
4.12
18.36
3.11
0.01
8.43
22.54
>$60
0.01
22.46
0.61
0.00
0.03
2.25
7.60
0.01
0.00
0.01
4.12
18.36
3.11
0.01
8.43
22.54
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-113
-------
SECTION IV — INDUSTRIAL PROCESSES • AEROSOLS
Table 5-10: World Breakeven Costs and Emissions Reductions in 2020 for Aerosols3
Reduction
Option
DPI
HFCtoHC
HFC to NIK
HFC-134ato152a
Cost
(2000$/tC02eq)
DR=10%,TR=40%
$439.54
-$6.34
-$5.87
-$1.07
Emissions
Reduction of
Option
(MtC02eq)
10.06
3.95
3.95
14.64
Reduction
from 2020
Baseline (%)
50.0%
10.0%
10.0%
37.0%
Cumulative
Reductions
(MtC02eq)
10.06
3.95
7.91
22.54
Cumulative
Reduction from
2020 Baseline (%)
50.0%
10.0%
20.0%
57.0%
a Results are separated by the line to reflect that the MDI option addresses different baseline emissions than the non-MDI options.
Figures 5-3 and 5-4 display the MDI aerosol international marginal abatement curves by region for
2010 and 2020, respectively.
Figure 5-3: 2010 MAC for MDI Aerosols, 10% Discount Rate, 40% Tax Rate
0)
§
e
o
'1
1
DC
in
O
'55
I
UJ
$500 -
$450 -
$400 -
$350 -
$300 -
$250 -
$200 -
$150 -
$100 -
$50 -
$OH
t tl V
___. I LJ „ 1 j
— * China
Rest of the world
— • United States
— s EU-15
— *• Japan
— * Other OECD
i i i i i
0.00 0.05 0.10 0.15 0.20 0.25
Cumulative Emissions Reductions (MtCO2eq)
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
IV-114
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES « AEROSOLS
Figure 5-4: 2020 MAC for MDI Aerosols, 10% Discount Rate, 40% Tax Rate
$500 -
If $450 -
O $400 -
S $350
g $300 -
'•§ $250
•§ $200
w $150
.1 $100 -
M
| $50 -
_tlf
... L.,1,—I .
1
China
Rest of the world
• United States
= EU-15
*• Japan
« Other OECD
0.0
1.0 2.0 3.0 4.0
Cumulative Emissions Reductions (MtCO2eq)
5.0
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Figures 5-5 and 5-6 display the non-MDI aerosol international MACs by region for 2010 and 2020,
respectively.
Figure 5-5:
i
f $4~
8 $2-
S»
C
* -$2°^
3
& -$4 -
in
° -$6 -
w
'i -$8 -
111
-$10 -"
2010 MAC for Non-MDI Aerosols, 10% Discount Rate, 40% Tax Rate
; * ii •
I
,1 , , , ,
)0 O.do 1.00 1.50 2.00 [;.50 3.00 3.50 4.00 4.50 5.00
— * China
Rest of the world
... .,„ „ , , r~~ ~ — • United States
- EU-15
* Japan
* Other OECD
Cumulative Emissions Reductions (MtCO2eq)
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-115
-------
SECTION IV— INDUSTRIAL PROCESSES • AEROSOLS
Figure 5-6: 2020 MAC for Non-MDI Aerosols, 10% Discount Rate, 40% Tax Rate
0)
CM
O
0
$4 -
-------
SECTION IV — INDUSTRIAL PROCESSES • AEROSOLS
IV.5.5.1 MDI Aerosols
Global baseline HFC emissions from MDI aerosols are estimated to grow from 2.9 to 20.1 MtCO2eq
between 2000 and 2020. In 2020, the Annex I region is estimated to be responsible for approximately 69
percent of the baseline emissions and represents the highest emissions growth from the MDI baseline,
from 2.8 MtCO2eq in 2000 to 13.8 MtCO2eq in 2020 (see Table 5-1).
As Table 5-10 illustrates, converting from HFC MDIs to DPIs is not a cost-effective abatement
option—the estimated cost worldwide is more than $400 dollars per tCO2eq at a 10 percent discount rate
and 40 percent tax rate—although the option may be implemented for other reasons (e.g., preferred
delivery system by a pharmaceutical company). At this cost, this option could abate 50 percent of global
MDI emissions, or 10.06 MtCO2eq, annually by 2020. The costs per tCO2eq for each region are equivalent
because available data on costs for abatement technologies were not scaled to reflect potential differences
in the costs internationally. Additional research may be performed to determine actual variability in costs
across regions.
IV.5.5.2 Non-MDI Aerosols
Baseline HFC emissions from non-MDI aerosols are estimated to grow from 24.2 MtCO2eq to 39.5
MtCO2eq globally for the years 2000 through 2020. In 2020, the majority of emissions is from the Annex I
region, which represents the highest emissions growth from the non-MDI baseline, from 24.1 MtCO2eq in
2000 to 39.4 MtCO2eq in 2020 (see Table 5-2).
As shown in Table 5-10, the greatest emissions reduction opportunities in all of the regions analyzed
may result from converting to HC, at a cost savings of $6.34 per tCO2eq at a 10 percent discount rate and
40 percent tax rate. The other two options, converting to NIK and HFC-152a, also represent a cost savings
of $5.87 and $1.07 per tCO2eq under the same discount and tax rate scenario, respectively. Globally, 22.54
MtCO2eq or 57 percent of global emissions from non-MDI aerosols, can be reduced in 2020 at a cost below
$0.00 per tCO2eq. As with MDI aerosols, costs per tCO2eq for all regions are equivalent because available
data on costs for abatement technologies were not scaled to reflect potential differences in the costs
internationally. Additional research may be performed to determine actual variability in costs across
regions.
IV.5.6 References
Arthur D. Little, Inc. 1999. 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 49648. Arthur D. Little, Inc.
Diversified CPC. May 2006. Personal communication between Bill Frauenheim of Diversified CPC and
Mollie Averyt of ICF International.
Diversified CPC. June 2004. Personal communication between Bill Frauenheim of Diversified CPC and
Mollie Averyt of ICF Consulting.
Dupont. July 2000. Personal communication between John Lueszler of Dupont and ICF Consulting.
Dupont. September 2005. Personal communication between Linda Calvarese of Dupont and ICF
Consulting.
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 MFCs for Metered Dose Inhalers in the European Union.
Commissioned by the International Pharmaceutical Aerosol Consortium (IPAC). Eviros March.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-117
-------
SECTION IV — INDUSTRIAL PROCESSES • AEROSOLS
International Pharmaceutical Aerosol Consortium (IPAC). 1999. Ensuring Patient Care, 2nd Edition.
Available at .
March Consulting Group. 1999. UK Emissions of MFCs, PFCs, and SF6 and Potential Emissions Reduction
Options: Final Report. March Consulting Group.
Nardini, Geno. May 2002. Personal communication between Geno Nardini and Iliriana Mushkolaj of ICE
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.
United Nations Environment Programme (UNEP). 1999. "The Implications to the Montreal Protocol of
the Inclusion of HFCs and PFCs in the Kyoto Protocol." UNEP HFC and PFC Task Force of the
Technology and Economic Assessment Panel (TEAP)
United Nations Environment Programme (UNEP). 2002. "2002 Report of the Aerosols, Sterilants,
Miscellaneous Uses and Carbon Tetrachloride Technical Options Committee: 2002 Assessment."
UNEP HFC and PFC Task Force of the Technology and Economic Assessment Panel (TEAP).
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.
X-rates.com. 2006. Available at . Accessed January 2006.
IV-118 GLOBAL MITIGATION OF NOM-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
IV.6 HFC Emissions from Fire Extinguishing
IV.6.1 Introduction
ire-extinguishing applications can be divided into two categories: portable fire extinguishers
(e.g., streaming applications) that originally used halon 1211 and total flooding applications
that originally used halon 1301 or halon 2402 (USEPA, 2004; March Consulting Group, 1998,
1999). Historically, SF6, another high-GWP gas, was used in select fire-extinguishing systems, such as for
system discharge testing purposes by the U.S. Navy. For the most part, however, SF6 is no longer used in
any capacity in the fire-protection sector.
The principal greenhouse gases used in and emitted from the fire-extinguishing sector are
hydrofluorocarbons (HFCs) (HFC-227ea, HFC-236fa, and HFC-23) and blends containing
perfluoromethane (CF4). These gases have 100-year GWPs that range from 2,900 to 11,700 (IPCC, 1996).
These high-GWP gases are substitutes for halons, ozone-depleting substances (ODSs) that have been,
and in many countries still are, widely used in fire-extinguishing applications. Although halons were
produced in much lower volumes than other ODSs, they have extremely high ozone depletion potentials
(ODPs) because of the presence of bromine, which reacts more strongly with ozone than chlorine. Halons
have been used historically in fire-suppression and explosion-protection applications because they are
electrically nonconductive, dissipate rapidly without residue, are safe for limited human exposure, and
are extremely efficient in extinguishing most types of fires (USEPA, 1994).
Portable fire extinguishers are most frequently used in offices, manufacturing and retail facilities,
aerospace/marine applications, and homes. Market penetration of HFCs in this sector has been limited
and is unlikely to grow or even keep pace with the growth in portable extinguishers (Wickham, 2003a).
Perfluorocarbons (PFCs) have had a very small penetration in the portable fire extinguisher market. By
2020, only one HFC, HFC-236fa, is expected to be used to a limited extent as a halon replacement in small
segments of the portable extinguisher sector. Overall, portable applications represent a much smaller
share of total fire-extinguishing sector greenhouse gas emissions than do total flooding applications, and
the U.S. Environmental Protection Agency (USEPA) projects that their relative share of emissions will
decrease over time, based on cost reasons outlined in Wickham (2002).
The majority of HFC emissions associated with fire extinguishing come from its use as a replacement
for some halon 1301 applications in the total flooding market. Total flooding systems are usually used to
protect a variety of spaces, including the following:
• electronic and telecommunications equipment, such as tape storage areas, computer facilities,
telecommunications gear, medical facilities, control rooms in nuclear power plants, and air traffic
control towers;
• military applications, including aviation engine nacelles1 and dry bays, naval engine
compartments, and engine compartments and occupied crew spaces of ground combat vehicles;
• oil production facilities;
• flammable liquid storage areas;
• engine nacelles and cargo bays of commercial aircraft;
1 Nacelles are enclosed engine housings.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-119
-------
SECTION IV— INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
• cultural institutions and museums;
• records storage areas;
• bank vaults;
• warehouses; and
• special facilities, such as research laboratories and military facilities.
Halon 1301 was widely used in total flooding applications because of its unique features (Wickham,
2002).2 Halon 1301 is a clean agent, meaning that it does not leave residue on equipment or in the
protection enclosure after discharge. In addition, halon 1301 is safe for limited, acute human exposure at
the concentration used for fire extinguishing. It is also very effective at extinguishing fires and works well
over a broad temperature range. Because halon 1301 was inexpensive, and design and installation of
halon 1301 systems were relatively simple compared with other fire-extinguishing systems, these systems
reached almost all segments of the total flooding fire-extinguishing market.
The alternatives to halon 1301 in total flooding applications can be categorized as in-kind, gaseous
agent alternatives (i.e., halofluorocarbons, CO^ inert gases, fluorinated ketones) and NIK alternatives
(i.e., dispersed and condensed aerosol extinguishing systems, water sprinklers, water mist, foam,3 or inert
gas generators). In most Annex I countries, halofluorocarbon HFC-227ea has emerged as the primary
replacement for halon 1301 in total flooding applications that require clean agents. Other HFCs, such as
HFC-23, HFC-236fa, and HFC-125, have been evaluated and determined to be safe for limited, acute
human exposure but are used in smaller amounts as a result of environmental,4 technical, and economic
concerns. For example, use of HFC-125 has been limited to normally unoccupied specialty applications,
such as aviation engine nacelles. However, over the next 20 years, HFC-23 and HFC-125 are expected to
only gain a strong foothold in the Russian Federation, based on confidential information collected for this
report from members of UNEP's Halon Technical Options Committee (HTOC). A small number of
telecommunications facilities in Eastern Europe historically used PFCs (C3F8 and C4F10). In the United
States, PFC use in fire suppression is very limited and is expected to tail off—the U.S. manufacturer of
PFCs for fire suppression withdrew these agents from the market a number of years ago because of
concern about their high GWP. In addition, HCFCs have historically been used as halon 1301
replacements, particularly in Eastern and Southern Europe. Over time, the use of HCFCs and PFCs in
total flooding applications is expected to be phased out and replaced primarily with HFCs, in addition to
other alternatives.
Available in-kind, nonhalocarbon alternatives in total flooding applications include CO2 systems,
used primarily in marine and industrial applications; fluorinated ketones; and inert gas systems, which
contain nitrogen or argon or blends of these gases, sometimes incorporating CO2 as a third component.
Inert gas systems have become the dominant halon 1301 replacement in many parts of Europe, most
notably in northern European countries.
Available NIK alternatives and technologies include powdered aerosols, water sprinklers, water mist
systems, foams, and combinations of these systems, such as aerosols with a halocarbon agent or water
mist with a gaseous agent or with foam.
2 The Russian Federation is an exception; it has historically relied on halon 2402, not halon 1301.
3 Foams can be protein based or synthetic based. Some synthetic-based foams contain fluorocarbons.
4 Whereas HFC-125 has a GWP of 2,800, approximately the same as HFC-227ea (GWP of 2,900), the other gases have
much higher GWPs. The GWP of HFC-236fa is 6,300, and the GWP of HFC-23 is 11,700.
IV-120 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
IV.6.2 Baseline Emissions Estimates
IV. 6. 2.1 Emissions Estimating Methodology
Description of Methodology
Specific information on how the emissions model was used to calculate ODS substitute emissions
from all sectors producing fire-protection emissions is described below.
The USEPA uses a detailed Vintaging Model of ODS-containing equipment and products to estimate
the use and emissions of ODS substitutes (HFCs) in the United States. Emissions estimates for non-U.S.
countries are derived using country-specific ODS consumption estimates, as reported under the Montreal
Protocol, in conjunction with Vintaging Model output for the fire-extinguishing sector. For countries for
which sufficient data were available, country-specific Vintaging Models were developed.
This analysis first incorporates estimates of the consumption of ODSs by country, as provided by
UNEP (1999). Estimates for EU were provided in aggregate, and GDP was used as a proxy to distribute
consumption among the individual member nations.
Emissions Equations
This analysis assumes that total emissions from leakage, accidental discharges, and fire extinguishing,
in aggregate, equal a percentage of the total quantity of chemical in operation at a given time. For
modeling purposes, the fire-extinguishing agent is assumed to be released at a constant rate for an
average equipment lifetime.
E, = r x Qcj-i+1for i=lrk (6.1)
where
E = Emissions. Total emissions of a specific chemical in year j for fire-extinguishing
equipment, by weight.
r = Percentage released. Average annual percentage of the total chemical in operation that is
emitted to the atmosphere.
Qc = Quantity of chemical. Total amount of a specific chemical used in new fire-extinguishing
equipment one lifetime (k) ago (e.g., j - k + 1), by weight.
i = Counter. Runs from 1 to lifetime (k).
j = Year of emissions.
k = Lifetime. The average lifetime of the equipment.
Estimates used for the percentage released, r, and lifetime of equipment, k, can have a significant
effect on the resulting emissions estimates. For this analysis, the U.S. Vintaging Model assumed that the
percentage released, r, or the emissions factor, is 2 percent for both the total flooding sector (Verdonik
and Robin, 2004) and the streaming sector. These estimates were chosen to account, on average, for all
emissions from servicing, leaks, accidental/false discharges, system decommissioning, or intentional
discharges to extinguish fires. The U.S. Vintaging Model also assumes equipment lifetime, k, for
streaming and flooding applications to be 10 and 20 years, respectively.
Regional Variations/Adjustments
To estimate baseline emissions, information collected on current and projected market
characterizations of international total flooding sectors was used to create country-specific versions of the
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-121
-------
SECTION IV — INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
Vintaging Model (i.e., country-specific ODS substitution patterns). Information on the current and
projected relative market size of halon substitutes was obtained for Australia, Brazil, China, India, Japan,
the Russian Federation, and UK, as provided by HTOC members from those countries.5 General
information was also collected on Northern, Southern, and Eastern Europe. For all other non-U.S.
countries, baseline emissions information from some of these countries was used to adjust substitution
patterns, as described below:
• Eastern Europe was used as a proxy for the countries in FSU and CEITs except the Russian
Federation, where specific information was available.
• Australia was used as a proxy for New Zealand.
• Brazil was used as a proxy for countries in Latin America and the Caribbean.
• India was used as a proxy for all other developing countries.
For all other non-U.S. Annex I countries, the U.S. ODS substitution pattern was used as a proxy.6 In
addition, an adjustment factor was applied to EU countries to account for European Regulation 2037/2000
on Substances that Deplete the Ozone Layer, which mandated the decommissioning of all halon systems
and extinguishers in the EU by the end of 2003 (with the exception of those applications that are defined
as critical uses) (Europa, 2003). To reflect this, the methodology assumes that all halon systems in the EU
were decommissioned by 2004.7
IV.6.2.2 Baseline Emissions
The resulting baseline estimates of GWP-weighted HFC emissions developed for this report are
summarized in Table 6-1 and Figure 6-1. Baseline emissions estimates are presented in million metric tons
of carbon dioxide equivalents (MtCO2eq). The estimates of the global total flooding fire-protection market
developed for this report are generally consistent with those in the Intergovernmental Panel on Climate
Change/Technology and Economic Assessment Panel (IPCC/TEAP) (1999) report, which estimated that in
the late 1990s, between 20 percent and 22 percent of systems that would formerly have used halons used
HFCs, and that less than 1 percent used PFCs.
5 Fire-protection experts in these countries provided confidential information on the status of national halon
transition markets and average costs to install the substitute extinguishing systems in use (on a per volume of
protected space basis) for 2001 through 2020.
6 This analysis assumes that, of the new total flooding protection systems in which halons have been previously used
in the United States, the market is currently made up of approximately 16 percent HFC-227ea, less than 1 percent
HFC-23,10 percent inert gas, and 74 percent other NIK agents.
7 The use of halon in marine applications is unlikely to have met the 2004 phaseout deadline because these
applications are also governed by regulations issued by the International Maritime Organization (IMO) and because
many EU ships contained halon 1301 fire-suppression systems. However, because of a lack of available data on
emissions from marine-based fire-protection systems as a percentage of the total EU fire-extinguishing sector, this
analysis simply assumes full compliance with the EU regulation
IV-122 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
Table 6-1: Total Baseline HFC Emissions from Fire Extinguishing (MtC02eq)
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
0.1
1.1
0.0
o.o .
0.3
0.0
0.1
0.0
0.1
0.0
0.0
1.1
0.0
0.0
0.7
1.7
2010
0.3
3.8
0.1
0.0
2.0
0.0
1.2
0.0
0.6
0.0
0.1
4.1
0.1
0.3
1.6
7.4
2020
0.6
5.8
0.1
0.0
4.9
0.1
2.2
0.1
0.9
0.1
0.3
6.3
0.3
0.5
1.9
13.7
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development
Figure 6-1: Baseline HFC Emissions from Fire Extinguishing by Region (MtC02eq)
§
U)
O
'55
E
tu
m Middle East
n Africa
• Non-EU FSU
• Latin America
• S&E Asia
• China/CPA
n OECD90+
2000
2010
2020
Year
CPA = Centrally Planned Asia; Non-EU FSU = non-European Union Former Soviet Union countries; OECD90+ = Organisation for Economic
Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-123
-------
SECTION IV— INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
This analysis assumes that systems designed to protect against Class A surface fire hazards represent
an estimated 95 percent of the total flooding sector in all countries, and that the remaining 5 percent of
the applications are designed to protect against Class B fire hazards (flammable liquids and gases).8
According to projected global average emissions estimates, emissions from systems that protect against
Class A fire hazards will account for approximately 74 percent of the global total fire-extinguishing sector
in 2005, 79 percent in 2010, 85 percent in 2015, and 87 percent in 2020. Table 6-2 presents the estimated
global average breakout of total fire-sector HFC emissions by application, as estimated by the USEPA's
Vintaging Model. This assumed breakout was used to estimate the emissions reduction potential of the
abatement options applicable to Class A or Class B total flooding sectors for all regions.
Table 6-2: Assumed Breakout of Total GWP-Weighted Baseline Fire-Extinguishing Emissions (Percent)
Annex 1 and Non-Annex 1 Countries
Flooding
Class A emissions (95% total flooding)
Class B emissions (5% total flooding)
Streaming
Total
2005
78.0%
74.1%
3.9%
22.0%
100.0%
2010
83.1%
78.9%
4.2%
16.9%
100.0%
2015
89.5%
85.0%
4.5%
10.5%
100.0%
2020
92.0%
87.4%
4.6%
8.0%
100.0%
Note: Totals may not sum because of independent rounding.
IV.6.3 Cost of HFC Emissions Reductions from Fire Extinguishing
This section presents a cost analysis for achieving HFC emissions reductions from the baselines
presented in Table 6-1. Each abatement option is described below, but costs are analyzed for only those
options not assumed to occur in the baseline (or for which a larger market penetration than reflected in
the baseline is believed to be possible) and for which adequate cost data are available. All cost analyses
assume a 20-year project lifetime. To the extent possible, this analysis considered total equivalent
warming impacts (TEWI) to account for the cost and greenhouse gas-emissions impacts of energy
consumption (i.e., indirect emissions) associated with the heating/cooling of additional space needed to
house alternative agents. However, because of data limitations, a full life-cycle analysis was not possible.
For example, the cost and emissions impacts associated with manufacturing alternative agents and all
system components were not assessed in this analysis, although they may potentially be significant.
IV.6.3.1 Description and Cost Analysis of Abatement Options
Because streaming applications account for a relatively small proportion of fluorocarbon use (e.g.,
HFC-236fa) in fire extinguishing, this cost analysis focuses only on abatement options for the total
8 Wickham (2002) estimates that over 90 percent 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 that
approximately 10 percent 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 percent of HFC emissions from the total flooding sector are
from Class B applications, and that the remaining 95 percent are from Class A applications.
IV-124
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
flooding sector.9 In 2005, the majority of emissions from the fire-extinguishing sector are expected to have
resulted from leaks and discharges (both accidental and intended use to extinguish fires) from total
flooding applications. The options for reducing HFC emissions from the fire-protection sector include the
use of alternative fire-protection agents and the use of alternative technologies and practices. Eight
potential options are identified, but only the first three are explored further in the cost analysis:
• inert gas,
• water mist,
• fluorinated ketone (FK-5-1-12),
• carbon dioxide,
• recovery and reuse of HFCs,
• improved detection systems,
• fine aerosols, and
• inert gas generators.
As described further below, available alternatives to reduce emissions in the fire-protection sector
may not be technically or economically viable for all end-use applications. For example, military
applications often have specialized needs that do not exist in other end-use applications. Applications
that are space and/or weight constrained, such as marine and aviation applications, are more limited in
their choice of agents. Electronic and telecommunication applications, which represent the largest use of
HFCs in the total flooding sector, offer the greatest opportunities to apply alternatives, although some
economic penalties and technical challenges may exist.
The remainder of this section provides an overview of each abatement option—inert gas, water mist,
and fluorinated ketone —and presents the assumptions and results of cost analyses. For reasons discussed
further below, these options are assumed to be applicable only to new (not existing) total flooding
systems, where "new" is defined as systems installed in 2005 or later. All costs are presented in 2000
dollars. A detailed description of the cost and emissions reduction analysis for each option can be found
in Appendix J for this chapter.
Inert Gas Systems
Inert gas systems extinguish fires using argon, nitrogen, or a blend of the two, sometimes
incorporating CO2 as a third component (UNEP, 2001). Inert gas systems provide both fire protection and
life safety/health protection equivalent to HFCs in most Class A (ordinary combustible) fire hazards,
including electronics and telecommunications applications.
Although inert gas systems are effective at extinguishing fires, their discharge time is slower than
that of HFC systems —60 seconds or more compared with 10 to 15 seconds (Kucnerowicz-Polak, 2002). In
areas where a rapidly developing fire is likely, inert gas systems are not recommended (UNEP, 2001;
Kucnerowicz-Polak, 2002). Improved devices that recognize and extinguish fires before they have a
chance to spread may help alleviate these concerns. Another limitation is that inert gas systems need a
larger volume of agent to extinguish fires than do HFC systems. The weight-support structures and space
9 The USEPA estimates that more than 90 percent of the halon replacement market in the streaming sector currently
consists of NIK alternatives, while HFCs account for less than 5 percent of this market. By 2020, the USEPA projects
that HFCs will account for an even smaller portion of the halon replacement market in the streaming sector. It is
expected that the high cost of HFCs will ensure that they are used only where they are absolutely needed (i.e., in
areas where cleanliness is an absolute necessity) (Wickham, 2002).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-125
-------
SECTION IV — INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
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. Extra storage
space for cylinders may also mean extra space to heat and cool, which means more expense and energy
consumption.
This analysis assumes inert gas systems are technically applicable10 to emissions from total flooding
systems designed for Class A fires. Because of the additional space requirements associated with inert gas
systems, it is not assumed to be economically feasible to retrofit existing HFC Class A fire-extinguishing
systems to this option. This analysis therefore assumes that this option is applicable only to new Class A
applications (i.e., those installed in 2005 or later). Because of the additional space and weight
requirements and the slower discharge times, market penetration rates reflect the assumption that this
option cannot fully displace HFC use in new Class A total flooding applications. Furthermore, because
this option entails additional costs (see discussion below), market penetration in non-Annex I countries is
assumed to be 50 percent less than in Annex I countries for all years because of economic challenges faced
by developing countries. Table 6-4 (Annex I countries) arid Table 6-5 (non-Annex I countries) present the
assumed market penetration rates of inert gas systems into new systems and as a percentage of total
sector baseline emissions.
Water Mist Systems
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). Another benefit of water mist systems is that, in some
applications (e.g., marine applications), they can be brought into action faster than HFC systems because
there is less concern about applying water mist in situations where openings to the space are not all
closed—which in turn leads to reduced fire damage. In addition, unlike HFC systems, which are usually
limited to a single discharge of agent, most water mist systems have an unlimited water supply in land-
based operations, and at least 30 minutes of potable water discharge followed by an unlimited amount of
seawater for marine applications (Wickham, 2003b).
To date, water mist systems have been used in shipboard accommodation, storage and machinery
spaces, combustion turbine enclosures, flammable and combustible liquid machinery applications, and
light and ordinary hazard sprinkler applications (UNEP, 2001). 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. Systems designed to protect against Class B (flammable liquid) fire
hazards are estimated to account for approximately 5 percent of the HFC total flooding market in the
United States and were assumed to account for the same percentage in all non-U.S. countries (Wickham,
2002). Water mist systems have also found acceptance in Class A applications but as replacements for
water sprinklers, not HFCs. Therefore, this report does not consider water mist as an option for abating
HFC emissions from Class A applications.
Some barriers have impeded broad use of water mist systems. First, these systems have not proven
effective in extinguishing small fires in spaces with volumes greater than 2,000 cubic meters (1MO, 2001;
10 In this report, the term "technically applicable" refers to the emissions to which an option can be theoretically
applied. Because inert gas systems are assumed to be used only in Class A fire total flooding applications, the
technical applicability is 100 percent of the emissions associated with those types of systems. Other factors will affect
the application of the option, for example to new or existing systems, and the market penetration assumed in this
analysis.
IV-126 . GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV— INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
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. Therefore, new applications may require empirical performance testing prior to installing
such systems to ensure safety and obtain approval of the proper regulatory or standard-setting authority.
Currently, an IMO working group is studying this situation and considering proposals that suggest an
overhaul to the test methods and approval guidelines. Should IMO change its water mist requirements to
something more flexible regarding the extinguishing of small fires in these sized spaces, it will make a
difference in the future cost and, thus, market acceptance of water mist systems (Wickham, 2003b). In
addition, the use of additives —such as salts or foam or a combination of these systems with gaseous
agents—offers other options under investigation to improve system performance for specific applications.
Many researchers and industry experts believe that solutions to these market barriers are well within
reach (Wickham, 2002).
Other market barriers for this option include additional space requirements for system storage
compared with conventional HFC-227ea systems. Indeed, water mist systems require an estimated seven
times more space than HFC-227ea (Wickham, 2003b). In addition, water mist systems used in marine
applications are cost prohibitive for protecting spaces less than 3,000 cubic meters in size.11
This analysis assumes that water mist systems are technically applicable to the emissions from total
flooding systems designed to protect against Class B fires. Because of the additional space requirements
associated with this option, it is assumed that water mist systems could not feasibly replace any existing
HFC systems in Class B fire-protection applications and, therefore, are used only in new Class B total
flooding applications (i.e., those installed in 2005 or later).
This analysis assumes that the remaining technical constraints associated with water mist systems
will gradually be overcome and that by 2020, in Annex I countries, water mist systems will reach full
market penetration in all new Class B fire-suppression systems used to protect large spaces. Market
penetration estimates for non-Annex I countries are assumed to be 50 percent less than those for
developed countries, as a result of economic considerations. Table 6-3 and Table 6-4 present the
maximum market penetrations assumed for this option.
Fluorinated Ketone (FK-5-1-12)
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 up to 2 weeks
and a 100-year GWP of approximately 1 (ICF Consulting, 2003). This alternative received the USEPA's
Significant New Alternatives Policy (SNAP) approval as an acceptable replacement for halon 1301 in
flooding applications at the end of 2002, and for halon 1211 in nonresidential streaming applications in
early 2003.
Compared with HFC-227ea total flooding systems, FK-5-1-12 systems are associated with slight space
and weight penalties; when averaged across different-sized spaces, space penalties are on the order of 7
This cost information is based on water mist systems employed under the current IMO requirements for marine
systems, which are much more severe than the requirements for land-based systems. The use of water mist systems
in nonmarine applications appears to be more cost competitive with other alternatives (Wickham, 2003a).
GLOBAL MITIGATION OF NON-co2 GREENHOUSE GASES iv-127
-------
SECTION IV — INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
percent,12 and weight penalties are approximately 17 percent—which could make its use in confined
spaces (e.g., ships, aircraft) less attractive, although some marine installations have already been reported
(Werner, 2004a).13 Moreover, because of its cost (see cost analysis below) and its relatively recent entry14
into this market, the extent of future commercial use of this option is not known.
Although this option is not associated with major floor space penalties and appears not to suffer from
any significant technical barriers, it is only assumed to be technically applicable in new Class A flooding
applications (i.e., those installed in 2005 or later), because the cost analysis does not assess retrofit costs.
Table 6-4 and Table 6-5 present the incremental maximum market penetration assumptions for this
option, which project that this option will gain a foothold in the marketplace and after 2010 will out-
compete inert gas systems in new Class A total flooding applications. Because of the reasons outlined
above, this analysis conservatively assumes that market penetration will be low in early years, although
others project higher sales (Werner, 2004b). Market penetration is assumed to be greater in Annex I
countries than in non-Annex I countries because of economic considerations.
Carbon Dioxide
CO2 has been used for many decades in total flooding systems. Some of the hazards and equipment
that CO2 systems protect are flammable liquid materials; electrical hazards such as transformers,
switches, circuit breakers, rotating equipment, and electronic equipment; engines using gasoline and
other flammable liquid fuels; ordinary combustibles such as paper, wood, and textiles; and hazardous
solids (NFPA, 2000). Because of the lethal concentrations at which CO2 is required for use as a fire-
extinguishing agent, there have been concerns about incidences of deaths and injuries attributed to
exposure to this agent (USEPA, 2000; Wickham, 2003b). In 2003, the NFPA Technical Committee on
NFPA 12: Standard for Carbon Dioxide Fire Extinguishing Systems reviewed a proposal to change the
standard to prohibit use of these systems in normally occupied areas (Wickham, 2003b). The IMO's Safety
of Life at Sea (SOLAS) standard does not prohibit the use of CO2 in normally occupied areas but calls for
the use of suitable alarms and mandates against the use of automatic release of the fire-extinguishing
medium, as noted in Carbon Dioxide as a Fire Suppressant: Examining the Risks (USEPA, 2000). IMO has also
considered whether to prohibit use of CO2 systems in occupied areas as part of that organization's broad
review of the current performance testing requirements for all shipboard fire-extinguishing systems
(IMO, 2003; Wickham, 2003b).
As one of the oldest fire-extinguishing agents in use,, and as a more economical option than HFCs,
CO2 has developed its own niche market in narrow-use total flooding applications. Whereas CO2 could
and does replace some halon use where permitted by regulations, this analysis assumes that CO2 would
be selected as a first-choice replacement of halon, not as a second transition, after more costly HFCs. For
example, the majority of U.S. ship owners have shifted from halon 1301 to CO2 for mandatory engine
room protection for new ships (Wickham, 2002). For this reason, any use of CO2 is assumed to occur in
the baseline and not as an option to replace HFC systems. It is therefore not considered in the cost
analysis.
12 Smaller spaces actually have no footprint penalty, but larger spaces (approximately equal to or greater than 1,000
cubic meters) have space penalties of roughly 14 percent.
13 It has been reported that the space penalty is only associated with use in large systems, and that the weight penalty
has not proven to be an impediment (Werner, 2004b).
14 This agent is in the 2004 edition of the National Fire Protection Association (NFPA) Standard on Clean Agent Fire
Extinguishing Systems (NFPA, 2004) and has been accepted for future addition to the International Standards
Organization (ISO) International Standard on Gaseous Fire-Extinguishing Systems (Wickham, 2003a).
IV-128 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
Recovery and Reuse of MFCs
HFCs can be recovered for reuse at service and decommissioning. For several reasons, however, this
analysis does not incorporate this option into the cost analysis. First, responsible halon management
practices are assumed to be a standard convention in fire protection throughout the world.15 Second,
given the high costs of HFCs, there is a strong financial incentive for maximum recovery following the
decommissioning of large HFC systems. Most HFC systems —with lifetimes ranging from 10 to 20
years—have not yet reached the end of their useful lifetimes and, therefore, wide-scale system recovery
and recycling at decommissioning has not yet occurred; this analysis assumes that such practices will
occur in the baseline.
Improved Detection Systems
One effective way of reducing HFC emissions from the fire-extinguishing sector is to install improved
detection and control systems to prevent a false discharge (e.g., high-sensitivity smoke detection systems
that provide early warning to preempt the need for actual system discharge) or minimize the amount of
agent discharged to extinguish a fire.
Since advanced detection systems have been available for the last decade or so, this analysis assumes
that total flooding HFC systems have been and are being equipped with such systems internationally.
Because improved detection systems are assumed to be used in the baseline, this option is not considered
in the cost analysis.
Fine Aerosols
Aerosols are being developed for use as extinguishing agents in niche markets in the United States,
such as aerospace, marine, and some military applications. The NFPA has written a draft standard (NFPA
2010) for this agent (NFPA, 2003). It is possible that if fine aerosols are ever successfully brought to
market, it may be applicable in other end-uses (Wickham, 2002).
Because fine aerosols are not currently a viable commercial alternative to HFCs in fire protection, and
much uncertainty exists as to whether the associated technical and economic barriers will be overcome to
enable them to become a viable option, fine aerosols are not considered in the cost analysis.
Inert Gas Generators
Inert gas generators use a solid material that oxidizes rapidly, producing large quantities of CO2
and/or nitrogen. Although this technology has demonstrated space and weight requirements equivalent
to halon 1301, it has thus far been used only in specialized applications in the United States (e.g., dry bays
on military aircraft) (Wickham, 2002). Because of insufficient data on these systems and the uncertainty
associated with their applicability in other fire-extinguishing applications, this option is not considered in
the cost analysis.
15 Responsible use practices are currently being developed and endorsed worldwide. For example, the Halon
Recycling Corporation (HRC) published a Code of Practice for Halon Reclaiming Companies (HRC, n.d.). Because the
equipment and training needed to reclaim halons are also required to reclaim HFCs, the HRC Code of Practice
establishes the necessary infrastructure and sets the practice of reclamation as the norm for how business is done in
the fire-protection industry. Although the HRC is a U.S. association, its membership consists of multinational
corporations operating throughout the world. Similarly, the Halon Alternatives Research Corporation (HARC), the
USEPA, and other organizations have recently developed and endorsed the Voluntary Code of Practice for the Reduction
of Emissions of HFC and PFC Fire Protection Agents (VCOP) (Cortina, n.d.). This VCOP will also have international
reach because HARC members include multinational companies in the alternative agent manufacturing, equipment
manufacturing, and distribution sectors.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-129
-------
SECTION IV — INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
IV.6.3.2 Summary of Technical Applicability, Market Penetration, and Costs of
Abatement Options
Table 6-3 summarizes the technical applicability of each option, which is equal to the estimated global
average breakout of total fire sector HFC emissions for the application (i.e., total flooding, Class A or B)
addressed by the option. Technical applicability is used in conjunction with market penetration
assumptions to develop the emissions reduction potentials for each option, as explained further below.
Table 6-4 provides the assumptions on maximum market penetration into annual installations of total
flooding systems designed for the particular application (i.e., Class A or B fires) for each option in 2005,
2010, 2015, and 2020. Market penetrations were developed separately for Annex I and non-Annex I
countries to best reflect region-specific qualitative information and possible future action. Table 6-5
presents the final maximum penetration into the installed base of equipment, taking into account the
percentage of each applicable fire hazard market that is new (i.e., systems installed in 2005 or later) in all
preceding years. Values from Table 6-5 are multiplied by technical applicabilities from Table 6-3 to
generate the percentage reduction off baseline emissions, as presented in Table 6-6.
Table 6-3: Summary of Technical Applicability of Abatement Options (Percent)
Annex 1 and Non-Annex 1 Countries
Abatement Option
Inert gas (Class A flooding)
Water mist (Class B flooding)
FK-5-1 -12 (Class A flooding)
2005
74.1%
3.9%
74.1%
2010
78.9%
4.2%
78.9%
2015
85.0%
4.5%
85.0%
2020
87.4%
4.6%
87.4%
Note: Values are expressed as a percentage of total fire-extinguishing emissions.
To calculate the percentage of emissions reductions off the total fire-extinguishing baseline for each
abatement option, the technical applicability (from Table 6-3) was multiplied by the market penetration
values (from Table 6-5), given that the reduction efficiency is 100 percent for each option. For example, to
determine the percentage reduction off the 2020 baseline for FK-5-1-12 in the United States (or other
Annex I countries), the following calculation was used:
Technical applicability * Incremental maximum market penetration = (6.2)
87.4% x 23.1% ~ 20.2%
Thus, using the assumptions in this analysis, FK-5-1-12 could reduce approximately one fifth of the
Annex I 2020 emissions baseline. This figure, along with the other projected emissions reductions, is
shown in Table 6-6.
Table 6-7 summarizes the cost assumptions used for the fire options presented in the discussions
above.
IV-130 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
Table 6-4: Assumed Incremental Market Penetration of Abatement Options into Newly Installed Class A or
Class B Extinguishing Systems, Expressed as a Percentage of Emissions from All New
Equipment
Abatement
Option
Inert gas
(New Class A)
Water mist
(New Class B)
FK-5-1-12
(New Class A)
Annex 1 Countries
2005 2010 2015 2020
10% 20% 30% 30%
25% 50% 75% 100%
4% 20% 40% 50%
Non-Annex 1 Countries9
2005 2010 2015 2020
5% 10% 15% 15%
13% 25% 38% 50%
2% 10% 20% 25%
Considerations/Rationale
• Can displace MFCs in new Class A
applications
• Additional space and weight
requirements
• Slower discharge times
• Higher costs compared with baseline
HFC-227ea systems lead to lower
market penetration in developing
countries
• Can displace HFCs in new Class B
applications used to protect large
spaces
• Technical constraints (assumed to
be gradually overcome)
• Higher costs compared with baseline
HFC-227ea systems lead to lower
market penetration in developing
countries
• Can displace HFCs in new Class A
applications
• No major additional space
requirements
• Lowest up-front cost of all
alternatives considered in this
analysis
• Newer player on market compared
with inert gas and water mist
systems; will take time to gain
foothold in market
• Higher costs compared with baseline
HFC-227ea systems lead to lower
market penetration in developing
countries
To account for economic considerations, assumed market penetration values in developing countries are half of those assumed for
developed countries.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-131
-------
SECTION IV — INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
Table 6-5: Market Penetration of Abatement Options into Newly Installed Class A or Class B Extinguishing
Systems, Expressed as a Percentage of Total Sector Emissions
Abatement Option
Inert gas (Class A)
Water mist (Class B)
FK-5-1-12(ClassA)
2005
0.5%
1.3%
0.2%
Annex
2010
4.5%
11.3%
3.6%
1 Countries
2015
11.0%
27.5%
11.6%
2020
18.5%
50.0%
23.1%
2005
0.3%
0.7%
0.1%
Non-Annex
2010
2.3%
5.7%
1.8%
Countries
2015
5.5%
13.9%
5.8%
2020
9.3%
25.2%
11.6%
Note: Values are expressed as a percentage of technical applicability (i.e., both new and existing Class A or Class B emissions).
Table 6-6: Percentage of Emissions Reductions Off Total Fire-Extinguishing Baseline
Abatement
Option
Inert gas
Water mist
FK-5-1-12
2005
0.4%
0.0%
0.1%
Annex I
2010
3.6%
0.5%
2.8%
Countries
2015
9.3%
1.2%
9.9%
2020
16.2%
2.3%
20.2%
2005
0.2%
0.0%
0.1%
Non-Annex
2010
1.8%
0.2%
1.4%
1 Countries
2015
4.7%
0.6%
4.9%
2020
8.1%
1.2%
10.1%
Table 6-7: Summary of Abatement Option Cost Assumptions (2000$)
Option
Inert gases
Water mist
FK-5-1-12
Time Horizon
(Years)
20
20
20
U.S. One-Time
Cost
$9.07a
$10.89"
$7.50e
U.S. Annual
Costs
$0.18"
$0.38"
$0.50"
U.S. Annual
Savings
$0.35C
$0.35C
$0.35C
Net U.S.
Annual Costs
-$0.17
$0.03
$0.15
Note: All costs are per cubic meter of protected space.
a This one-time cost includes an incremental capital cost and an incremental construction cost (to build additional space). Incremental capital
costs were assumed to be 10 percent greater in non-Annex I (developing) countries than in the United States and 10 percent less in Japan.
In all other Annex I countries, capital costs were assumed to be the same as in the United States. No regional adjustments were made to
incremental construction costs.
b This cost is associated with additional heating and cooling costs. For all other countries, this annual cost was adjusted by average country-
specific electricity prices (average of 1994-1999) based on Annual Energy Outlook 2000 (Electricity Prices for Industry 1994-1999) (USEIA,
(2000).
c Annual savings were assumed to result from avoided HFC-227ea emissions and associated replacement costs. No adjustments were
assumed for other countries.
d This one-time cost includes an incremental capital cost and an incremental construction cost (to build additional space) Capital costs were
assumed to be the same in all other Annex I countries and 10 percent higher in all developing countries. No regional adjustments were made
to incremental construction costs.
6 This one-time cost includes an incremental capital cost and an incremental construction cost (to build additional space). No regional cost
adjustments were made.
IV-132
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
IV.6.4 Results
IV.6.4.1 Data Tables and Graphs
Table 6-8 (2010) and Table 6-9 (2020) provide a summary of the potential emissions reductions at
various breakeven costs by country/region. The costs to reduce 1 tCO2eq are presented for a 10 percent
discount rate and 40 percent tax rate. Table 6-10 presents the potential emissions reduction opportunities
and associated annualized costs for the world in 2020. The results are ordered by increasing costs per
tCO2eq, using the highest cost in the region under the 10 percent discount rate/40 percent tax rate.
Because many of the options analyzed affect indirect (CO2 from energy generation for heating/cooling)
emissions, the net (HFC + CO2) emissions reduced by each option are presented. The direct (HFC)
emissions reduced by the option and a cumulative total of emissions reduced, in MtCO2eq and
percentage of the regional fire-extinguishing baseline, are also presented.
Table 6-8: Country/Regional Emissions Reductions in 2010 and Breakeven Costs for Fire Extinguishing at
10% Discount Rate, 40% Tax Rate ($/tC02eq)
2010
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex I
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$15
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$30
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$45
0.00
0.22
0.00
0.00
0.07
0.00
0.07
0.00
0.02
0.00
0.01
0.23
0.01
0.01
0.11
0.33
$60
0.01
0.26
0.00
0.00
0..07
0.00
0.08
0.00
0.04
0.00
0.01
0.26
0.01
0.01
0.11
0.37
>$60
0.01
0.26
0.00
0.00
0.07
0.00
0.08
0.00
0.04
0.00
0.01
0.27
0.01
0.01
0.11
0.39
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-133
-------
SECTION IV — INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
Table 6-9: Country/Regional Emissions Reductions in 2020 and Breakeven Costs for Fire Extinguishing at
10% Discount Rate, 40% Tax Rate ($/tC02eq)
2020
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$15
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$30
0.00
0.00
o.ob
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$45
0.06
1.97
0.04
0.00
0.88
0.04
0.79
0.01
0.19
0.01
0.12
2.00
0.11
0.10
0.70
3.30
$60
0.11
2.19
0.04
0.00
0.88
0.04
0.81
0.02
0.34
0.01
0.13
2.22
0.12
0.10
0.74
3.62
>$60
0.12
2.25
0.04
0.00
0.94
• 0.04
0.84
0.02
0.36
0.01
0.13
2.29
0.12
0.11
0.74
3.77
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Table 6-10: World Breakeven Costs and Emissions Reductions in 2020 for Fire Extinguishing
Cost(2000$/tC02eq)
Reduction .
Option
FK-5-1-12
Inert gases
Water mist
DR = 107o,TR=40%
LOW
$37.26
$34.53
$48.16
High
$37.58
$48.85
$82.40
Direct
Emissions
Reduction3
(MtC02eq)
1.97
1.58
0.23
Indirect
Emissions
Impact*
(MtCOjeq)
0.00
-0.11
-0.04
%
Reduction
from 2020
Baseline
14.4%
11.5%
1.6%
Running
Sum of
Reductions
(MtC02eq)
1.97
3.55
3.77
Cum. %
Reduction
from 2020
Baseline
14.4%
25.9%
27.6%
a Direct reductions refer to HFC emissions reductions (off the baseline).
b Indirect emissions impacts are those associated with energy consumption (not included in the baseline).
IV-134
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
Figures 6-2 (2010) and 6-3 (2020) present MACs for this sector at 10 percent discount rates and 40
percent tax rates.
Figure 6-2: 2010 MAC for Fire Extinguishing, 10% Discount Rate, 40% Tax Rate
§f $100 -
§ $90-
S $80-
Emissions Reductions
^-"•NiGJ-tvOlOi^l
OOOOOOOO
o ,,,,,,,,
IIP |
i r >JJ r-
, u
^ — i _L- ••
— * China
Rest of the world
— • United States
— s EU-15
— + Japan
— c other OECD
0 0.0 0.0 0.1 0.1 0.1 0.1
Cumulative Emissions Reductions (MtCO2 eq.)
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Figure 6-3: 2020 MAC for Fire Extinguishing, 10% Discount Rate, 40% Tax Rate
$100 -|
i $90-
g $80 -
§ $70 -
£ $60 -
o
•g $50 -
| $40-
« $30 -
o $20 -
'55
.2 $10 -
— * China || T f x
Rest of the world I i
— • United States -• I
-EU-15 i
?
— > Japan A r'
^ c\iv\r\r r\^Lt~^r\ , '
~« utner ULL.U ) I
/ /
^ : : J _^^, — — - ~" ^
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Cumulative Emission Reductions (MtCO2eq.)
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-135
-------
SECTION IV— INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
IV.6.4.2 Uncertainties and Limitations
This section focuses on the uncertainties and limitations associated with the cost estimates presented
in this analysis. One area of uncertainty is in how capital costs for these mitigation technologies may vary
internationally, given that estimates were available onty for several countries, and only for two of the
three options assessed (water mist and inert gas). The analysis of the FK-5-1-12 option is currently limited
in the lack of this specificity on region-specific cost analysis estimates. In addition, it should be noted that
the global implementation of each option through 2020 is based on information currently available and
expert opinion. Great uncertainty is associated with future projections of market behavior.
IV.6.5 Summary
Baseline HFC emissions from fire extinguishing are estimated to grow between 2005 and 2020, with
the highest emissions growth expected to occur in non-Annex I countries. It is estimated that the vast
majority of these emissions will come from total flooding applications; only a minor amount will come
from streaming applications. Several alternatives to ozone-depleting halon 1301 for total flooding
applications exist, including gaseous alternatives such as HFCs, carbon dioxide, inert gases, and
fluorinated ketones, as well as NIK alternatives such as dispersed and condensed aerosol systems, water
sprinklers, water mist, foam, and inert gas generators.
This analysis reviewed these alternatives and analyzed in detail three mitigation options for total
flooding fire-extinguishing applications: substituting HFC systems used in new systems designed to
protect against Class A fire hazards with inert gas systems, substituting HFCs used in new systems
designed to protect against Class B fire hazards with water mist systems, and substituting HFC systems
used in new systems designed to protect against Class A fire hazards with FK-5-1-12 systems. Inert gas
and FK-5-1-12 systems may offer good opportunities to reduce emissions in total flooding applications
globally. Water mist systems also have the potential to reduce global emissions from this sector, but to a
lesser extent, because they are applicable only to Class B fire hazards.
This analysis demonstrates that there is a portfolio of alternatives to HFCs in the total flooding sector
that can be employed to reduce HFC use and associated emissions. These alternatives include FK-5-1-12,
inert gases, water mist, and other agents and systems discussed qualitatively in this report. The global
implementation of each option through 2020 is based on a "best-guess" scenario. With more data, these
forecasts can be improved.
IV.6.6 References
Cortina, TA. No date. Voluntary Code of Practice for HFC and PFC Fire Protection Agents. Available at
. Accessed on June 14,
2006.
Europa. 2003. Regulation (EC) No. 2037/2000 of the European Parliament and of the Council of 29 June
2000 on substances that deplete the ozone layer. Europa. Available at . Accessed on October 13, 2003.
Halon Recycling Corporation. No date. Code of Practice for Halon Reclaiming Companies. Available at
http://www.halon.org/pdfs/code.pdf>. Accessed on June 14, 2006.
ICF Consulting. September 10, 2003. Re-evaluation of a C--6 Oxyfluorocarbon (trade name Novec 1230) and
References. Memorandum delivered by ICF Consulting to Erin Birgfeld under USEPA Contract
Number 68-D-00-266, Work Assignment 2-05 Task 03.
IV-136 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
Intergovernmental Panel on Climate Change (IPCC). 1996. Climate Change 1995: The Science of Climate
Change. J.T. Houghton, L.G. Meira Filho, B.A. Callander, N. Harris, A. Kattenberg, and K. Maskell
(eds.). Cambridge, UK: Cambridge University Press.
Intergovernmental Panel on Climate Change/Technology and Economic Assessment Panel (IPCC/TEAP).
July 1999. Meeting Report of the Joint IPCC/TEAP Expert Meeting on Options for the Limitation of Emissions
of HFCs and PFCs. Report jointly sponsored by the IPCC Working Group III and the TEAP of the
Montreal Protocol (ECN-RX--99-029). Available at . Accessed on November 20, 2003.
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.
International Maritime Organization (IMO). October 10, 2003. Performance Testing and Approval Standards
for Fire Safety Systems—Report of the Correspondence Group. Submitted by the United States to the
International Maritime Organization Subcommittee on Fire Protection, 48th session, Agenda item 5.
Kucnerowicz-Polak, B. March 28, 2002. "Halon Sector Update." Presented at the 19th Meeting of the
Ozone Operations Resource Group (OORG), The World Bank, in Washington, DC.
March Consulting Group. September 1998. Opportunities to Minimize Emissions of Hydrofluorocarbons
(HFCs) from the European Union: Final Report.
March Consulting Group. January 1999. UK Emissions of HFCs, PFCs, and SF6 and Potential Emissions
Reduction Options: Final Report.
National Fire Protection Association (NFPA). 2000. NFPA 12: Standard on Carbon Dioxide Extinguishing
Systems, 2000 Edition. National Fire Protection Association.
National Fire Protection Association (NFPA). July 21, 2003. Proposed Draft of NFPA 2010: Standard for Fixed
Aerosol Fire Extinguishing Systems, 2005 Edition. Available at . Accessed on October 24, 2003.
National Fire Protection Association (NFPA). 2004. NFPA 2001: Standard on Clean Agent Fire Extinguishing
Systems.
United Nations Environment Programme (UNEP). October 1999. Production and Consumption of Ozone
Depleting Substances 1986-1998.
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. Energy Information Administration (USEIA). 2000. Annual Energy Outlook 2000 (Electricity Prices for
Industry 1994-1999). Available at .
Accessed on April 2, 2002.
U.S. Environmental Protection Agency (USEPA). 1994. SNAP Technical Background Document: Risk Screen
on the Use of Substitutes for Class I Ozone-Depleting Substances: Fire Suppression and Explosion Protection
(Halon Substitutes). Washington, DC: U.S. Environmental Protection Agency.
U.S. Environmental Protection Agency (USEPA). February 2000. Carbon Dioxide as a Fire Suppressant:
Examining the Risks. EPA 430-R-00-002. Washington, DC: U.S. Environmental Protection Agency,
Office of Air and Radiation.
U.S. Environmental Protection Agency (USEPA). April 2004. Inventory of U.S. Greenhouse Gas Emissions
and Sinks: 1990-2002. EPA 430-R-04-003. Washington, DC: U.S. Environmental Protection Agency,
Office of Atmospheric Programs.
Verdonik, Daniel P. and Mark L. Robin. 2004. Analysis of Emissions Data, Estimates, and Modeling of
Fire Protection Agents. Conference proceedings from the 15th Annual Earth Technologies Forum and
Mobile Air Conditioning Summit in Washington, DC. April 13-15.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-137
-------
SECTION IV— INDUSTRIAL PROCESSES • FIRE EXTINGUISHING
Werner, Kurt. 2004a. Expert review comments on the Draft Analysis of International Costs to Abate HFC PFC
Emissions from Fire Extinguishing. Comments received via email on January 26.
Werner, Kurt. 2004b. Expert review comments on the Draft Analysis of International Costs to Abate HFC PFC
Emissions from Fire Extinguishing. Comments received via e-mail May 20-21.
Wickham, Robert. 2002. Status of Industry Efforts to Replace Halon Fire Extinguishing Agents. Wickham
Associates. March 16. Available at .
Wickham, Robert. 2003a. Expert review comments on the Draft Analysis of International Costs to Abate HFC
PFC Emissions from Fire Extinguishing. Comments received in writing and by phone in October.
Wickham, Robert. 2003b. Review of the Use of Carbon Dioxide Total Flooding Fire Extinguishing Systems.
Wickham Associates, August 8. Available at .
IV-138 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • ALUMINUM PRODUCTION
IV.7 RFC Emissions from Aluminum Production
he primary aluminum production industry is currently the largest source of PFC emissions
globally. During the aluminum smelting process, when the alumina ore content of the
electrolytic bath falls below critical levels required for electrolysis, rapid voltage increases
occur. These are termed "anode effects" (AEs). Anode effects produce CF4 and C2F6 emissions when
carbon from the anode and fluorine from the dissociated molten cryolite bath combine. In general, the
magnitude of emissions for a given level of production depends on the frequency and duration of these
anode effects; the more frequent and long-lasting the anode effects, the greater the emissions. This report
presents two baselines for PFC emissions from primary aluminum production: the technology-adoption
baseline and the no-action baseline (Tables 7-1 and 7-2).
IV.7.1 Technology-Adoption Baseline
Under the technology-adoption baseline scenario, it is assumed that aluminum producers will
continue to introduce technologies and practices aimed at reducing PFC emissions. It is assumed that
under the technology-adoption scenario, global aluminum producers, in accordance with International
Aluminum Institute (IAI) PFC emissions reduction commitments, will reduce their PFC emissions
intensity (i.e., PFC emissions per ton of produced aluminum) by 80 percent from 1990 levels by 2010. This
reduction can be achieved by retrofitting smelters with emissions-reducing technologies such as
computer control systems and point feeding systems, by shifting production to Point-Feed Prebake
(PFPB) technology, and by adopting management and work practices aimed at reducing PFC emissions.
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
PFPB, which is considered the most technologically advanced process to produce aluminum. Although
PFPB systems can be improved through the implementation of management and work practices, as well
as improved control software, the analysis assumes that retrofit abatement options will occur only on
existing VSS, HSS, SWPB, and CWPB cells.
Figure 7-1 presents total PFC emissions from aluminum production under the technology-adoption
baseline scenario from 1990 through 2020. Between 1990 and 1995, global emissions declined from 98 to
61 MtCO2eq. This significant decline was the result of voluntary measures undertaken by global smelters
to reduce their AE minutes per cell day. These measures included incremental improvements in smelter
technologies and practices, and a shift in the share of SWPB-related production to more state-of-the-art
PFPB facilities. Although a continuation of this AE minute reduction trend occurred through 2000,
emissions reductions were offset by a 24 percent increase in global aluminum production between 1995
and 2000.
The declining global emissions levels through 2010 reflect the successful adoption of IAI emissions
reduction goals through both retrofits and a continued shift of production from VSS, HSS, and SWPB to
PFPB. From 2010 through 2020, the emissions intensity is assumed to remain constant; consequently,
emissions will be driven by increasing aluminum production. PFC emissions in OECD, as well as non-EU
Eastern Europe, non-EU FSU, China/CPA, and S&E Asia are projected to remain relatively constant from
2010 through 2020, due to slowing aluminum production growth. Trends in the United States and the
EU-25 reflect overall trends in the developed (OECD) countries. Africa, Latin America, and the Middle
East are projected to increase their share of global emissions from 2010 through 2020, due to strong
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-139
-------
SECTION IV — INDUSTRIAL PROCESSES • ALUMINUM PRODUCTION
Table 7-1: Total PFC Emissions from Aluminum Manufacturing (MtC02eq)—No-Action Baseline
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
5.6
37.3
3.5
3.9
5.2
2.9
8.1
0.8
0.0
0.1
9.2
28.1
7.5 .
0.2
9.0
57.8
2010
6.0
37.8
2.8
3.5
13.2
1.2
4.7
2.2
0.2
0.1
8.3
29.6
7.4
0.8
14.7
69.8
2020
8.6
38.3
2.8
4.7
13.5
1.2
4.7
2.4
0.2
0.1
8.2
30.2
7.3
0.8
14.7
77.1
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development
Table 7-2: Total PFC Emissions from Aluminum Manufacturing (MtC02eq)—Technology-Adoption Baseline
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
5.6
37.4
3.5
3.9
5.2
2.9
8.1
0.8
0.0
0.1
9.3
28.2
7.5
0.2
9.0
58.0
2010
4.0
18.9
2.5
2.4
6.5
0.8
3.5
1.2
0.1
0.1
3.9
15.0
3.3
0.6
4.6
39.1
2020
5.7
19.6
2.5
3.2
6.7
0.8
3.5
1.3
0.1
0.1
3.9
15.8
3.3
0.7
4.4
44.7
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development
IV-140 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • ALUMINUM PRODUCTION
Figure 7-1: RFC Emissions from Aluminum Production Based on a Technology-Adoption Scenario-
1990-2020 (MtC02eq)
• Middle East
P S&E Asia
• Latin America
• Africa
P Non-EU FSU
• Non-EU Eastern Europe
• China/CPA
• OECD90+
• Japan
P EU-25
• United States
1990
2000 2010
Year
2020
CPA = Centrally Planned Asia; EU-25 = European Union; Non-EU FSU = non-European Union Former Soviet Union countries; OECD90+ =
Organisation of Economic Co-operation and Development; S&E Asia = Southeast Asia.
growth in aluminum production. In 2020, China/CPA, Latin America, Africa, and the Middle East are
expected to collectively account for 50 percent of global emissions. In comparison, OECD is projected to
account for 36 percent of global emissions, down from 51 percent in 2000.
IV.7.2 No-Action Baseline
Under the no-action baseline scenario, it is assumed that aluminum producers will take no retrofit
actions to reduce their emissions below the levels of the late 1990s; as a result, emissions projections do
not reflect anticipated technology adoptions and/or the implementation of improved work and
management practices to reduce emissions. Figure 7-2 presents total PFC emissions from aluminum
production under the no-action baseline scenario from 1990 through 2020. The trends from 1990 through
2000 are the same as those in the technology-adoption baseline. From 2000 through 2020, no additional
abatement retrofits are assumed to occur; however, as in the technology-adoption baseline, it is assumed
that the global historical trend in the shift of production from SWPB to PFPB continues (IAI, 2000, 2005b).
Based on these assumptions, global emissions under this scenario rise to 77 MtCO2eq in 2020, a 33 percent
increase over 2000 levels. This is primarily driven by increasing global aluminum production.
In 1990, OECD emissions accounted for approximately 60 percent of global emissions; however, by
2020, this share is reduced to 40 percent in this scenario. This reduction is the result of relatively flat
aluminum production levels between 2000 and 2020, as cheaper aluminum from developing countries
enters the global marketplace. The primary sources of this cheaper aluminum are China/CPA, the Middle
East, Latin America, and Africa, which in 2020 are projected to have production levels approximately
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-141
-------
SECTION IV — INDUSTRIAL PROCESSES • ALUMINUM PRODUCTION
Figure 7-2: RFC Emissions from Aluminum Production Based on a No-Action Scenario—1990-2020
(MtC02eq)
• Middle East
n S&E Asia
• Latin America
• Africa
Q Non-EU FSU
• Non-EU Eastern Europe
• China/CPA
B OECD90+
B Japan
D EU-25
B United States
1990
2000
2010
2020
Year
CPA = Centrally Planned Asia; EU-25 = European Union; Non-EU FSU = non-European Union Former Soviet Union countries; OECD90+ =
Organisation of Economic Co-operation and Development; S&E Asia = Southeast Asia.
200 percent greater than their 2000 levels. In 2020, China/CPA is projected to account for 17.5 percent of
global emissions, compared with 3 percent in 1990 and 9 percent in 2000.
The EU-25 and the United States reflect the general OECD trend, except that between 2000 and 2005
there is an increase in U.S. emissions and a decrease in EU emissions. The decrease in EU emissions is
primarily the result of their transition from SWPB to PFPB technology. The increase in U.S. emissions is
an artifact of the baseline calculation methodology. Past U.S. emissions reflect reductions already
implemented by members of the USEPA's Voluntary Aluminum Industrial Partnership, but under this
scenario, future U.S. emissions (from 2005 forward) are projected to occur at a higher rate.
IV.7.3 Cost of PFC Emissions Reduction from Aluminum Production
1V.7.3.1 Abatement Options
The most direct and cost-efficient method to reduce PFC emissions and improve process efficiency is
to retrofit existing aluminum production technology. Two types of retrofit options can be implemented:
(1) installation or refinement of process computer control systems, and (2) the installation or conversion
to alumina point-feed systems. The installation of process computer controls can be considered a minor
retrofit, whereas the installation of alumina point-feed systems can be considered a major retrofit. These
two reduction technologies are not mutually exclusive, but additive. In fact, point-feed systems require
computer controls in order to be effective, although the reverse is not true.
In this analysis, these two options are assumed to be adopted in succession. At relatively low carbon
prices, computer controls are adopted by 100 percent of the market (i.e., 100 percent of a given cell
technology) and at higher carbon prices, alumina point feed systems are adopted by 100 percent of the
IV-142
GLOBAL MITIGATION OF NOM-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • ALUMINUM PRODUCTION
market. The costs and reductions of installing alumina point-feed systems are additive to those of
installing computer controls. (Of course, if carbon prices are high enough at the outset, the two options
will not be adopted successively in time, but simultaneously. In any event, the MACs provide an accurate
measure of the potential reductions at all carbon prices.)
Cosf and Reduction Assumptions
Cost assumptions for each mitigation option are based on information reported in Greenhouse Gas
Emissions from the Aluminum Industry (IEA, 2000). The remainder of Section IV.7.2.1 will provide an
overview of each abatement option and detail the cost assumptions used.
Table 7-3 provides a summary of the potential reduction opportunities associated with each
mitigation option. The reduction efficiencies of complete retrofits (i.e., retrofits including installation of
both computer controls and point-feed systems) were estimated by assuming that after implementation of
the complete retrofit, the cell will operate (and emit PFCs) as a PFPB cell. Consequently, the reduction
efficiencies were based on the differences between the PFC emissions rates (tCO2eq/t Al) of unabated
VSS, HSS, SWPB, and CWPB cells and the PFC emissions rate of PFPB cells. The emissions rates of
unabated VSS, HSS, SWPB, and CWPB cells were represented by the average global 1995 emissions rates
of those technologies, because the market penetration of minor and major retrofits was believed to be
small in 1995. The complete retrofit reduction efficiencies range from 41 to 93 percent, depending on cell
type.
Table 7-3: Reduction Efficiency Potential for Abatement Option by Cell Type (Percent)
Cell Technology Type
Abatement Option
Computer controls (minor retrofit )
Point-feed (major retrofit )
Complete retrofit (both)
VSS
35.5%
35.5%
71.0%
HSS
33.5%
33.5%
77.0%
SWPB
23%
70%
93%
CWPB
31%
10%
41%
The distribution of maximum reduction efficiencies (i.e., those associated with complete retrofits)
between minor and major retrofits was estimated based on communications with industry (Marks, 2006).
For VSS and HSS, it was assumed that reductions are evenly split (i.e., minor and major retrofits each
achieve 50 percent of the reductions of a complete retrofit). For SWPB, 25 percent of the total reduction
was assumed to result through implementation of the minor retrofit, with the remainder occurring
through the major retrofit. For CWPB, 75 percent of the total reduction was assumed to occur through
implementation of the minor retrofit, whereas 25 percent was assumed to occur through implementation
of the major retrofit.
Although PFPB systems can be further improved through the implementation of control software
(Marks, 2006), this analysis assumes that retrofit abatement options will occur only on existing VSS, HSS,
SWPB, and CWPB technologies.
Because the PFC abatement options are based on the retrofitting of existing cell technologies and not
on a major change in technology, the maximum market penetration available is assumed to be 100
percent of emissions from the VSS, HSS, SWPB, and CWPB cell type production lines. However, the
applicability of cell type-specific retrofit options to baseline emissions is dependent on the country-level
distribution of cell technologies.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-143
-------
SECTION IV — INDUSTRIAL PROCESSES • ALUMINUM PRODUCTION
Installation/Refinement of Computer Controls (Minor Retrofit)
The minor retrofit option includes the installation of process computer control systems or the
refinement of existing process control algorithms. Computer systems provide greater control over
alumina feeding, enable control of the repositioning of the anodes as they are consumed during
aluminum production, and enhance the ability to predict and suppress AEs. Consequently, computer
systems have the potential to increase productivity, lower energy costs, and reduce PFC emissions. The
minor retrofit option is assumed to have the potential to reduce PFC emissions factors (tCO2eq/t Al) by
between 23 percent and 36 percent, depending on the cell type (see Table 7-4).
Cost and Emissions Reduction Analysis
• Capital/Upfront Costs. For a typical facility that produces 200,000 tonnes of aluminum, capital
costs are assumed to range from $4.2 to $5.1 million, depending on the cell type (IEA, 2000).
• Annual Costs. Implementation of the minor retrofit option is assumed to produce an incremental
increase in operating costs between 2 and 3 percent per year depending on the cell technology
type. Operating costs are assumed to include costs associated with operation and maintenance
labor and with overhead and administrative costs. Country-specific operating costs are
determined by applying the incremental operating cost associated with this option to the regional
baseline operating costs developed for IEA (2000).
• Cost Savings. The reduction in AE minutes per cell day produced by the mitigation option is
assumed to result in a corresponding increase in aluminum production. This analysis assumes
that the cost of aluminum is $1,400 per metric ton. Additional cost savings from reduced
aluminum fluoride losses and energy consumption were estimated using assumptions detailed in
Estimating the Cost of an Anode Effect (USEPA, 2002).
Installation of Point-Feed Systems (Major Retrofit)
The major retrofit option includes only the installation of alumina point-feed systems. This option is
considered to take place in addition to the implementation of the minor retrofit. The benefits and costs
associated with the minor retrofit option are considered fully implemented and therefore are not
included in the analysis of this option. The implementation of this option results in improved cell
performance and increased PFC emissions reductions.
The alumina point-feed system allows alumina to be fed at shorter time intervals and at different
positions along the bath, compared with feeding techniques used by existing VSS, HSS, SWPB, and
CWPB cells. PFC emissions occur as alumina levels in the cell bath decline, typically below 2 percent by
weight of cell bath composition (Dugois, 1994), and the remaining fluorine-containing bath components
begin to undergo electrolysis. Since AEs can be terminated through the addition of more alumina, point
feeding will ensure that alumina is fed continuously into the central parts of the cell, where the bath area
is largest. Furthermore, point feeding also increases the cell current efficiency and consequently reduces
the cell electricity consumption. The major retrofit option is assumed to have the potential to reduce PFC
emissions factors by between 10 and 70 percent, depending on the technology type (see Table 7-4).
Cost and Emissions Reduction Analysis
• Capital/Upfront Costs. Capital costs for VSS and HSS cells are assumed to be approximately $39
million (IEA, 2000), whereas those for SWPB cells are assumed to be approximately $82 million
(Marks, 2006). For CWPB cells, capital costs are approximately $3.5 million (IEA, 2000).
• Annual Costs. Implementation of the major retrofit option is assumed to produce an incremental
increase in operating costs between 1 and 3 percent per year, depending on the cell technology
IV-144 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • ALUMINUM PRODUCTION
type. Operating costs are assumed to include costs associated with operation and maintenance
labor and with overhead and administrative costs (IEA, 2000).
• Cost Savings. Similar to the minor retrofit option, cost savings include benefits associated with
avoided aluminum production losses, reduced electricity consumption, and reduced aluminum
fluoride losses. For this analysis, it is assumed that the cost of aluminum is $1,400 per metric ton.
Assumptions used to estimate benefits associated with reduced energy consumption and fluoride
losses are detailed in Estimating the Cost of an Anode Effect (USEPA, 2002).
Industry experts indicate that if the computer control system is installed separately from the point-
feed system, particularly if it is installed several years earlier, the computer control system is likely to
require an update in its software to accommodate the point-feed system. The costs of such an update are
not included in this analysis. However, the USEPA believes that these costs are likely to be small relative
to the costs of the point-feed system itself.
Baseline Market Penetration of Options in No-Action and Technology-Adoption Baselines
The reductions achieved by each technology in each scenario are based not only on the reduction
efficiency and maximum market penetration of that technology but also on the share of the market that is
already claimed by the technology in the baseline of concern. For example, if a technology had already
achieved a 100 percent market penetration in the baseline of concern, no reductions from that technology
would be available in the MAC associated with that baseline.
In both the no-action and technology-adoption scenarios, there is some baseline market penetration
by the minor and major retrofit options. In the no-action scenario, plant operators outside of the United
States are assumed to have adopted complete retrofits to the extent required to achieve the 2000
emissions factor for each technology, which is significantly lower than the 1995 emissions factor.1 In the
technology-adoption scenario, plant operators are assumed to have adopted complete retrofits to the
extent required to achieve the 2010 IAI goal of reducing emissions intensities by 80 percent relative to the
1990 level. Table 7-4 shows the global average baseline market penetrations of complete retrofits in both
scenarios.
Table 7-4: Average3 Baseline Market Penetration of Complete Retrofits by Cell Type and Scenario (Percent)
Cell Technology Type
Scenario
No-action
Technology-adoption
vss
30%
98%
HSS
21%
75%
SWPB
16%
24%
CWPB
52%
94%
a These are global averages. Individual countries may have slightly larger or slightly smaller baseline market penetrations.
For the United States, the baseline market penetration of retrofits in the no-action scenario is assumed
to be zero. This is because the U.S. no-action baseline emissions are based on a 1990 emissions factor, and
few if any retrofits are believed to have been performed by 1990.
The assumption that complete retrofits were adopted in all cases is a simplification; in fact, it is likely
that some plant operators have adopted only minor retrofits. If this were explicitly modeled in the
analysis, the baseline market penetration of minor retrofits would grow, while that of major retrofits
1 Although the various types of cells (VSS, SWPB, etc.) become PFPB cells after implementation of a complete retrofit,
the IEA model, which was used as the basis for this analysis, continues to track converted cells under their old cell
technologies. Thus, it is reasonable to treat retrofits this way in this context.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-145
-------
SECTION IV — INDUSTRIAL PROCESSES • ALUMINUM PRODUCTION
would decline. Thus, this analysis may overestimate the reductions available from minor retrofits and
underestimate those available from major retrofits.
IV.7.4 Results
This section discusses the results from the MAC analysis for the world and for various regions for the
no-action and technology-adoption scenarios.
IV.7.4.1 Data Tables and Graphs
Tables 7-5 through 7-10 provide a summary of the potential emissions reduction opportunities and
associated costs for various regions in 2010 and 2020 under the no-action and technology-adoption
scenarios. The costs to reduce 1 tCO2eq are presented at a 10 percent discount rate and 40 percent tax rate.
Table 7-5: Emissions Reductions in 2010 and Breakeven Costs ($/tC02eq) for Aluminum Production at 10%
Discount Rate, 40% Tax Rate (MtC02eq)—No-Action Baseline
2010
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
1.17
3.88
0.31
0.29
2.84
0.22
0.00
0.45
0.00
0.00
1.91
1.97
1.83
0.14
1.14
9.50
$15
2.56
12.28
0.44
1.49
3.48
0.48
1.60
0.65
0.05
0.00
2.79.
9.49
2.42
0.18
6.18
21.77
$30
3.23
17.45
0.44
1.91
6.78
0.62
1.90
1.13
0.11
0.00
5.22
12.23
4.68
0.18
7.48
31.90
$45
3.23
17.45
0.44
1.91
6.78
0.62
1.90
1.13
0.11
0.00
5.22
12.23
4.68
0.18
7.48
31.90
$60
3.23
17.45
0.44
1.91
6.78
0.62
1.90
1.13
0.11
0.00
5.22
12.23
4.68
0.18
7.48
31.90
>$60
3.23
17.45
0.44
1.91
6.78
0.62
1.90
1.13
0.11
0.00
5.22
12.23
4.68
0.18
7.48
31.90
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development
IV-146 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • ALUMINUM PRODUCTION
Table 7-6: Emissions Reductions in 2020 and Breakeven Costs ($/tC02eq) for Aluminum Production at 10%
Discount Rate, 40% Tax Rate (MtC02eq)—No-Action Baseline
2020
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
1.72
3.87
0.31
0.36
2.74
0.22
0.00
0.45
0.00
0.00
1.85
2.02
1.77
0.15
1.14
10.16
$15
3.64
12.47
0.44
1.97
3.60
0.48
1.62
0.70
0.07
0.00
2.77
9.70
2.38
0.19
6.18
23.86
$30
4.65
17.80
0.44
2.62
6.99
0.61
1.94
1.22
0.15
0.00
5.17
12.63
4.61
0.19
7.48
34.86
$45
4.65
17.80
0.44
2.62
6.99
0.61
1.94
1.22
0.15
0.00
5.17
12.63
4.61
0.19
7.48
34.86
$60
4.65
17.80
0.44
2.62
6.99
0.61
1.94
1.22
0.15
0.00
5.17
12.63
4.61
0.19
7.48
34.86
>S60
4.65
17.80
0.44
2.62
6.99
0.61
1.94
1.22
0.15
0.00
5.17
12.63
4.61
0.19
7.48
34.86
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development
Table 7-7: Emissions Reductions in 2010 and Breakeven Costs ($/tC02eq) for Aluminum Production at 10%
Discount Rate, 40% Tax Rate (MtC02eq)—Technology-Adoption Baseline
2010
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.35
0.40
0.06
0.19
0.02
0.07
0.00
0.01
0.00
0.00
0.16
0.24
0.12
0.02
0.18
1.14
$15
1.40
2.40
0.15
0.81
0.22
0.27
0.87
0.07
0.02
0.00
0.56
1.84
0.36
0.02
0.64
5.31
$30
1.40
2.93
0.15
0.86
0.41
0.28
0.87
0.12
0.04
0.00
0.85
2.08
0.60
0.02
0.81
6.13
$45
1.40
2.93
0.15
0.86
0.41
0.28
0.87
0.12
0.04
0.00
0.85
2.08
0.60
0.02
0.81
6.13
$60
1.40
2.93
0.15
0.86
0.41
0.28
0.87
0.12
0.04
0.00
0.85
2.08
0.60
0.02
0.81
6.13
>$60
1.40
2.93
0.15
0.86
0.41
0.28
0.87
0.12
0.04
0.00
0.85
2.08
0.60
0.02
0.81
6.13
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-147
-------
SECTION IV — INDUSTRIAL PROCESSES • ALUMINUM PRODUCTION
Table 7-8: Emissions Reductions in 2020 and Breakeven Costs ($/tC02eq) for Aluminum Production at 10%
Discount Rate, 40% Tax Rate (MtC02eq)—Technology-Adoption Baseline
2020
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United,States
World Total
$0
0.49
0.40
0.06
0.24
0.02
0.07
0.00
0.02
0.00
0.00
0.15
0.24
0.11
0.02
0.18
1.34
$15
1.95
2.47
0.15
1.03
0.28
0.27
0.87
0.09
0.02
0.00
0.57
1.90
0.37
0.02
0.65
6.24
$30
1.95
3.06
0.15
1.10
0.54
0.27
0.87
0.15
0.05
0.00
0.87
2.20
0.61
0.02
0.81
7.24
$45
1.95
3.06
0.15
1.10
0.54
0.27
0.87
0.15
0.05
0.00
0.87
2.20
0.61
0.02
0.81
7.24
$60
1.95
3.06
0.15
1.10
0.54
0.27
0.87
0.15
0.05
0.00
0.87
2.20
0.61
0.02
0.81
7.24
>S60
1.95
3.06
0.15
1.10
0.54
0.27
0.87
0.15
0.05
0.00
0.87
2.20
0.61
0.02
0.81
7.24
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development
Table 7-9: Emissions Reduction and Costs in 2020—No-Action Baseline
Cost
(2000S/tC02eq)
DR=10%, TR=40%
Reduction Option
Computer controls: SWPB
Computer controls: VSS
Computer controls: HSS
Computer controls: CWPB
Point feed: SWPB
Point feed: CWPB
Point feed: HSS
Point feed: VSS
Low
-$2.44
-$5.75
$0.71
-$16.93
$6.27
-$9.35
$19.21
$20.37
High
$0.73
$0.75
$4.75
$6.13
$6.98
$14.17
$23.25
$26.88
Emissions
Reduction of
Option
(MtC02eq)
1.85
8.25
2.74
4.09
5.56
1.36
2.74
8.25
Reduction
from 2020
Baseline (%)
2.4%
10.7%
3.6%
5.3%
7.2%
1.8%
3.6%
10.7%
Running Sum
of Reductions
(MtCOzeq)
1.85
10.11
12.85
16.94
22.50
23.86
26.61
34.86
Cumulative
Reduction
from 2020
Baseline (%)
2.4%
13.1%
16.7%
22.0%
29.2%
31.0%
34.5%
45.2%
IV-148
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • ALUMINUM PRODUCTION
Table 7-10: Emissions Reduction and Costs in 2020—Technology-Adoption Baseline
Cost
(2000$/tC02eq)
DR=10%, TR=40%
Reduction Option
Computer controls: SWPB
Computer controls: VSS
Computer controls: HSS
Computer controls: CWPB
Point feed: SWPB
Point feed: CWPB
Point feed: HSS
Point feed: VSS
Low
-$2.44
-$5.75
$0.71
-$16.93
$6.27
-$9.35
$19.21
$20.37
High
$0.73
$0.75
$4.75
$6.13
$6.98
$14.17
$23.25
$26.88
Emissions
Reduction of
Option
(MtC02eq)
1.22
0.20
0.80
0.26
3.67
0.09
0.80
0.20
Reduction
from 2020
Baseline (%)
2.7%
0.5%
1.8%
0.6%
8.2%
0.2%
1.8%
0.5%
Running Sum
of Reductions
(MtC02eq)
1.22
1.43
2.23
2.48
6.16
6.24
7.04
7.24
Cumulative
Reduction
from 2020
Baseline (%)
2.7%
3.2%
5.0%
5.5%
13.8%
14.0%
15.7%
16.2%
IV.7.4.2 Global and Regional MACs and Analysis
This section discusses the results from the MAC analysis for the world and by region, including
China, Japan, the United States, the EU-15, other OECD, and the rest of the world.
Figure 7-3 presents the 2010 and 2020 global technology-adoption and no-action MACs for aluminum
production. The difference in abatable emissions between the technology-adoption and no-action MACs
reflects the impact of retrofits adopted globally to meet lAI's PFC emissions reduction goal. The
technology-adoption baseline reflects the IAI goal, which is to reduce the global PFC emissions intensity
to 80 percent below 1990 levels by 2010. In contrast, the no-action baseline and MACs assume that
aluminum producers will implement no retrofit actions beyond those necessary to achieve year 2000
emissions rates.
For the technology-adoption and no-action global MACs, operational and capital costs for
implementing retrofits, as well as the global PFC emissions intensities for smelter technologies, are
assumed to remain constant between 2010 and 2020. Consequently, for both MAC scenarios, changing
aluminum production levels represents the primary driver for MAC curve shifts between 2010 and 2020.
Most of this increased production is expected to occur in the other OECD and rest of the world regions
(specifically in Africa and Latin America). The shift to the right is greater in the no-action global MACs
because of the larger presence of non-retrofitted smelters in all global regions. That is, while the
technology-adoption MACs assume that all CWPB, SWPB, as well as a majority of VSS and HSS smelters,
have been retrofitted to PFPB, the no-action MACs assume that no changes have occurred since 2000.
Figures 7-4 and 7-5 present 2010 and 2020 regional technology-adoption MACs for China, Japan, the
United States, the EU-15, other OECD, and the rest of the world. The 2020 regional MACs reflect the
successful and continuing attainment of lAI's 2010 global PFC emissions intensity goal, which is expected
to be achieved by retrofitting Soderberg and SWPB smelters with computer control systems and point-
feeding systems. As a result, relatively limited emissions reductions are available in Japan, the United
States, China, the EU-15, and other OECD countries in 2010. Where reductions are available, they will
predominantly occur at those smelters that still utilize HSS and SWPB-based technologies. SWPB retrofits
to PFPB will generally occur before HSS. SWPB retrofit costs range between -$2.4 and $7/tCO2eq,
compared with $0.7 to $23/tCO2eq for HSS. In 2010, most VSS smelters are assumed to have undergone
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-149
-------
SECTION IV— INDUSTRIAL PROCESSES • ALUMINUM PRODUCTION
Figure 7-3: 2010 and 2020 Global Technology-Adoption and No-Action MACs for Primary Aluminum
Production
-$20
20
2010
2020
Technology-Adoption
No-Action
25
30
35
Cumulative Emissions Reductions (MtCO2eq)
40
Figure 7-4: 201 0 Regional Technology-Adoption MACs for Primary Aluminum Production
$60 -j
$50 -
i $40-
j «-
"* O $20 -
in O
.2 S; $10 -
(0
E"~ U jjj — i i i i i i i i
m oft) 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
-$10 Hf
-$20 -f
Cumulative Emissions Reductions (MtCO2eq)
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development
IV-150
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • ALUMINUM PRODUCTION
Figure 7-5: 2020 Regional Technology-Adoption MACs for Primary Aluminum Production
China
Rest of the world
• United States
s EU-15
+ Japan
- « Other OECD
—I 1 1 1 1 1 1 1 1 1
OJO 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Cumulative Emissions Reductions (MtCO2eq)
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development
retrofit through voluntary actions, and therefore only limited reductions are still available for this cell
type. For the remaining rest of the world countries, significant reductions in the MAC will occur in
Africa, Brazil, and the Russian Federation (i.e., these regions account for over 80 percent, approximately 3
MtCO2eq, of the rest of the world reductions). For both Africa and Brazil, significant SWPB-based
production is expected to continue to occur, and thus be available for retrofit to PFPB. In the Russian
Federation, VSS, HSS, and SWPB-based production is assumed to continue to occur.
As in 2010, the majority of emissions reductions in 2020 are expected to be available in the rest of the
world, specifically Africa, Brazil, and the Russian Federation (Figure 7-5). Again reductions are expected
to occur predominantly through the retrofit of HSS and SWPB smelters. On a global basis, approximately
70 percent of reductions will occur at SWPB smelters. Because SWPB retrofits are relatively inexpensive,
this means that most of the reductions (6.2 MtCO2eq) will be available for less than $7/tCO2eq. Another
0.9 MtCO2eq will be available between $7 and $24/tCO2eq, primarily from the major retrofit of HSS
smelters. (Major and minor HSS smelter retrofits account for approximately 22 percent of global
emissions reductions.) The majority of HSS retrofits are expected to occur in China and the Russian
Federation.
IV.7.4.3 Uncertainties and Limitations
Uncertainties and limitations persist despite attempts to incorporate all publicly available
information on international aluminum production. Some key areas of uncertainty within the aluminum
MAC modeling methodology are provided below.
Aluminum Production
A major source of uncertainty in the MACs is due to variation in aluminum production for all
countries. Aluminum production is highly variable, with operations coming on- and offline as market
forces fluctuate; thus, a simple measure of capacity is not always indicative of actual production,
especially for long-range projections. Also, production fluctuations between cell types within a given
country can significantly affect the emissions estimates, because different cell types have significantly
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-151
-------
SECTION IV — INDUSTRIAL PROCESSES • ALUMINUM PRODUCTION
different emissions rates. The USEPA modeled documented Soderberg and SWPB to PFPB technology
shifts (IAI, 2005a) in the no-action and technology-adoption baselines; however, recent environmental
factors, as well as increasing energy costs, have resulted in the shutdown of many Soderberg smelters
(Marks, 2006). Consequently, technology mix assumptions used in this analysis may not represent actual
technology mix conditions in global aluminum markets.
Baseline Market Penetration of Retrofits
The USEPA has modeled the baseline market penetration of retrofits in both the no-action and
technology-adoption scenarios on a cell-type- and countiy-specific basis. However, the information used
in the model (from IEA, 2000) is now several years old, and it may not reflect the actual adoption of
retrofits globally. Thus, the reductions available for the various cell types may be over or underestimated.
Cost Savings
Benefits associated with reduced energy consumption and fluoride losses were calculated using
assumptions detailed in Estimating the Cost of an Anode Effect (USEPA, 2002). However, this cost savings is
dependent on a number of factors, such as the type of power system installed at smelters (e.g., constant
power consumption systems, which are used by most aluminum smelting facilities; or constant potline
current or amperage systems), which can vary significantly depending on smelter technology-types, age,
and operational characteristics. Consequently, conservative assumptions were applied when estimating
potential cost savings. Generally, these savings were estimated to be on the order of 1 to 4 percent of total
realized cost savings, of which the primary contributor is avoided aluminum production losses.
Adjusting Costs for Specific Domestic Situations
Currently, the technologies considered in this report are widely available. However, individual
countries may be faced with higher costs from transportation and tariffs associated with purchasing the
technology from abroad or with lower costs from domestic production of these technologies. Data on
domestically produced and implemented retrofit technologies in individual countries are not available.
Emissions Reduction Effectiveness of Retrofit Technologies
The PEC emissions factor reductions used for the minor and major retrofits may vary significantly
depending on various operational conditions (e.g., cell conditions, plant operator effectiveness).
Additionally, as technologies and control software evolve, additional reduction opportunities are likely to
occur. For example, recently Alcan Pechiney reported improved software and feed systems that have the
potential to make substantial reductions in emissions on cells that are already considered to be high
performing relative to PEC emissions (Marks, 2006). Any deviation from the assumed emissions
reduction potential of the retrofits would have a direct impact on estimated emissions.
IV.7.5 References
Dugois, J.P. 1994. Anode Effect Control as a Means to Limit PFC Emissions from Electrolytic Cells. PFC
Workshop. London, United Kingdom: International Primary Aluminum Institute.
International Aluminum Institute (IAI). 2000. Perfluorocarbon Emissions Reduction Programme 1990- 2000.
Available at .
International Aluminum Institute (IAI). 2005a. The International Aluminum Institute's Report on the
Aluminum Industry's Global Perfluorocarbon Gas Emissions Reduction Programme— Results of the 2003
Anode Effect Survey. Available at .
International Aluminum Institute (IAI). 2005b. Aluminum for Future Generations Sustainability Update.
Available at < http://www.world-aluminium.org/iai/publications/documerits/update_2004.pdf>.
IV-152 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV— INDUSTRIAL PROCESSES • ALUMINUM PRODUCTION
International Energy Agency (IEA). 2000. Greenhouse Gas Emissions from the Aluminum Industry.
Cheltenham, United Kingdom: The International Energy Agency Greenhouse Gas Research &
Development Program.
Marks, J. 2006. Personal communication with Jerry Marks, IAI.
U.S. Environmental Protection Agency (USEPA). March 2002. Estimating the Cost of an Anode Effect.
Washington, DC: USEPA.
U.S. Environmental Protection Agency (USEPA). 2006. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2020. Washington, DC: USEPA.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-153
-------
SECTION IV — INDUSTRIAL PROCESSES • ALUMINUM PRODUCTION
IV-154 GLOBAL MITIGATION OF NOM-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • HCFC-22 PRODUCTION
IV.8 HFC-23 Emissions from HCFC-22 Production
IV.8.1 Source Description
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 nonfeedstock uses is scheduled to be
phased out under the Montreal Protocol. However, feedstock production is permitted to continue
indefinitely.
Nearly all producers in developed countries have implemented process optimization or thermal
destruction to reduce HFC-23 emissions. In a few cases, HFC-23 is collected and used as a substitute for
ODSs, mainly in very-low temperature refrigeration and air-conditioning systems. Emissions from this
use are quantified in the Air Conditioning and Refrigeration chapters and are therefore not included here.
HFC-23 exhibits the highest GWP of the HFCs—11,700 under a 100-year time horizon—with an
atmospheric lifetime of 264 years. Baseline emissions estimates under both a technology-adoption and a
no-action baseline scenario are presented in Tables 8-1 and 8-2.
Table 8-1: Total HFC-23 Emissions from HCFC-22 Production (MtC02eq)—No-Action Baseline
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
0.0
51.0
0.0
0.1
33.3
0.0
6.5
4.7
13.6
2.6
1.1
56.2
1.1
0.0
29.8
95.6
2010
0.0
29.6
0,0
0.2
70.2
0.0
1.8
8.0
0.8
4.0
0.7
38.6
0.7
0.0
26.3
118.0
2020
0.0
26.3
0.0
0.2
91.0
0.0
1.0
9.2
0.9
4.3
0.3
36.6
0.3
0.0
24.0
137.5
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-155
-------
SECTION IV — INDUSTRIAL PROCESSES • HCFC-22 PRODUCTION
Table 8-2: Total HFC-23 Emissions from HCFC-22 Production (MtC02eq)—Technology-Adoption Baseline
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex I
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
0.0
51.0
0.0
0.1
33.3
0.0
6.5
4.7
13.6
2.6
1.1
56.2
1.1
0.0
29.8
95.6
2010
0.0
11.4
0.0
0.2
27.0
0.0
0.6
1.1
0.8
0.3
0.7
15.4
0.7
0.0
9.3
44.7
2020
0.0
10.1
0.0
0.2
47.8
0.0
0.4
2.3
0.9
0.6
0.3
15.3
0.3
0.0
8.5
66.2
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
IV.8.1.2 No-Action Baseline
Under the no-action baseline scenario, it is assumed that HCFC-22 producers will take no further
action to reduce their emissions; as a result, emissions projections do not reflect anticipated technology
adoptions to reduce emissions. Under this scenario, world HFC-23 emissions from HCFC-22 production
are expected to grow by an additional 56 percent between 2000 and 2015, but emissions are expected to
decline between 2015 and 2020 as a result of the phaseout of nonfeedstock HCFC-22 production in
developing countries.
Figure 8-1 reveals a striking shift: the majority of emissions will come from China and other
developing countries rather than from the OECD countries. This is due to (1) a combination of increased
use of emissions controls and the phaseout of HCFC-22 under the Montreal Protocol in OECD countries
and (2) increased HCFC-22 production in China. (These drivers are discussed further below.) Thus, while
HFC-23 emissions from developed countries are expected to decline by more than 60 percent from 1990 to
2020 in the no-action baseline, HFC-23 emissions in the China/CPA region are expected to increase
dramatically. Southeast Asia and Latin America are also projected to show increasing emissions during
this period. In 1990, the three largest emitters for this source were the United States, Japan, and France,
which together accounted for more than two-thirds of all emissions. In 2020, the three largest emitters are
projected to be China, India, and the United States. These nations are anticipated to account for 90 percent
of all HFC-23 emissions, while China alone is expected to be the world's major HFC-23 emitter,
accounting for more than 65 percent of total emissions.
IV-156 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • HCFC-22 PRODUCTION
Figure 8-1: HFC-23 Emissions from HCFC-22 Production Based on a No-Action Scenario—1990-2020
(MtC02eq)
• OECD90+
• Non-EU FSU
n Africa
• Latin America
• S&E Asia
B China/CPA
• Japan
a EU-25
• United States
1990
2000
2010
2020
Year
CPA = Centrally Planned Asia; EU-25 = European Union; Non-EU FSU = non-European Union Former Soviet Union countries; S&E Asia =
Southeast Asia; OECD90+ = Organisation for Economic Co-operation and Development
In the developed world, HFC-23 emissions decreased between 1990 and 2000 because of process
optimization and thermal destruction, although there were increased emissions in the intervening years.
The United States and EU drove these trends in the developed world. Although emissions increased in
the EU-25 between 1990 and 1995 because of increased production of HCFC-22, a combination of process
optimization and thermal oxidation led to a sharp decline in EU emissions after 1995, resulting in a net
decrease in emissions of 67 percent for this region between 1990 and 2000. U.S. emissions also declined by
15 percent during the same period, despite a 35 percent increase in HCFC-22 production; however,
during that time period, U.S. emissions demonstrated two distinct trends. Between 1990 and 1995, U.S.
emissions declined by 23 percent because of a steady decline in the emissions rate of HFC-23 (i.e., the
amount of HFC-23 emitted per kilogram of HCFC-22 manufactured). However, between 1995 and 2000,
U.S. emissions increased because of increases in HCFC-22 production.1
As illustrated in Figure 8-1 under the no-action baseline, HFC-23 emissions in developed countries
are predicted to continue to decrease through 2020 as a result of (1) Japan's implementation of either
thermal abatement or HFC-23 capture (for use) for 100 percent of its production beginning in 2005
(JICOP, 2006), (2) 100 percent implementation of thermal abatement in all EU countries except Spain by
2010, (3) closure of the HCFC-22 production plant in Greece in 2006, and (4) the HCFC-22 production
phaseout scheduled under the Montreal Protocol.
1 The apparent increase in U.S. emissions between 2000 and 2005 is an artifact of the method used to estimate U.S.
emissions in the no-action baseline. Under this approach, the U.S. emissions factor was assumed to revert to its
relatively high 1990 level in 2005, despite reductions in earlier years.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-157
-------
SECTION IV — INDUSTRIAL PROCESSES • HCFC-22 PRODUCTION
In the developing world, particularly China, emissions are increasing quickly because of a rapid
increase in the production of HCFC-22. This production is meeting growing demand for unitary air-
conditioning, commercial refrigeration, and substitutes for CFCs, which are currently being phased out in
developing countries under the Montreal Protocol (UNEP, 2003). Under the no-action baseline, the
increase in HFC-23 emissions is expected to continue through 2015, when HCFC-22 itself will begin to be
phased out by developing countries for most end-uses under the Montreal Protocol.
IV.8.1.3 Technology-Adoption Baseline
Under the technology-adoption baseline scenario, it is assumed that HCFC-22 producers will
introduce technologies and practices aimed at reducing HFC-23 emissions. Under this scenario, global
HFC-23 emissions from HCFC-22 production are expected to decline by 35 percent between 2000 and
2020. These trends are mainly a result of the expected implementation of Clean Development Mechanism
(COM) projects in China, India, Korea, and Mexico, as well as implementation of thermal oxidation in
Spain and the HCFC-22 production phaseout scheduled under the Montreal Protocol.
As seen in Figure 8-2, the most striking trend apparent in the technology-adoption baseline is the
dramatic decline in emissions from China (and thus for the world, since by 2005 China accounts for the
majority of emissions) between 2005 and 2010, followed by an increase in emissions from 2010 to 2015, at
which point emissions again decline. The first dip in this zigzag pattern is caused by the implementation
of CDM projects in China. Abatement is assumed to begin in 2010, decreasing emissions. However, while
abatement (in absolute terms) is held constant through 2015 and 2020, emissions grow between 2010 and
2015 as a result of the increase in production of HCFC-22 in China (discussed in Section 8.1.2). The
increase in HFC-23 emissions is then reversed after 2015, when HCFC-22 itself will begin to be phased out
by developing countries for most end-uses.
Figure 8-2: HFC-23 Emissions from HCFC-22 Production Based on a Technology-Adoption Scenario-
1990-2020 (MtC02eq)
120
• OECD90+
• Nori-EU FSU
D Africa
• Latin America
• S&E Asia
H China/CPA
• Japan
DEU-25
• United States
2000
2010
2020
Year
CPA = Centrally Planned Asia; EU-25 = European Union; Non-EU FSU = non-European Union Former Soviet Union countries; S&E Asia =
Southeast Asia; OECD90+ = Organisation for Economic Co-operation and Development.
IV-158
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • HCFC-22 PRODUCTION
Emissions in OECD countries are expected to decline by 80 percent between 1995 and 2015. As
Figure 8-2 shows, the majority of these emissions shift to China and other developing countries. This is
due to (1) a combination of increased use of emissions controls and the phaseout of HCFC-22 in OECD
countries and (2) increased HCFC-22 production in China. Thus, while HFC-23 emissions from
developed countries are expected to decline by more than 80 percent from 1990 to 2020, HFC-23
emissions in the China/CPA region are expected to increase dramatically, despite the adoption of
abatement measures under the COM. Southeast Asia and Latin America are also projected to show
increasing emissions through this period.
Global emissions in 1990 to 2000 follow the same trends as in the no-action baseline. As illustrated in
Figure 8-2, HFC-23 emissions in developed countries as compared with the no-action baseline are
predicted to decrease from 2010 through 2020, mainly as a result of the U.S. implementation of thermal
abatement.
IV.8.2 Cost of HFC-23 Reduction from HCFC-22 Production
IV.8.2.1 Abatement Options
Historically, the majority of HFC-23 emissions have been vented to the atmosphere. However, two
options have been identified as technically viable measures to reduce HFC-23 emissions from HCFC-22
production (IPCC, 2001):
• manufacturing process optimization and
• destruction of HFC-23 by thermal oxidation.
Process Optimization
Process optimization and modifications to production equipment can both optimize HCFC-22
production and reduce HFC-23 emissions. Process optimization is relatively inexpensive and is likely to
be most effective in reducing the emissions from plants that are generating HFC-23 at a rate of 3 percent
to 4 percent. Process optimization has been demonstrated to reduce emissions of fully optimized plants to
below 2 percent of HCFC-22 production. This analysis assumes that all plants in developed countries
have already implemented some optimization, resulting in HFC-23 emissions reductions. These plants
may achieve further reductions through additional process optimization, but these reductions are likely
to be more modest (Rand et al., 1999). Therefore, this option is not explicitly included as a mitigation
option in this MAC analysis.
Thermal Oxidation
Thermal oxidation, the process of oxidizing HFC-23 to CO^ HF, and water, is a demonstrated
technology for the destruction of halogenated organic compounds. For example, destruction of more than
99 percent 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 HF and the high temperatures required for
complete destruction. This analysis assumes a reduction efficiency of 95 percent.2
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 percent (Rost, 2006).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-159
-------
SECTION IV — INDUSTRIAL PROCESSES • HCFC-22 PRODUCTION
Although typical incinerators that burn only HFC-23 produce 6 pounds of CO2 for every 1 pound of
HFC-23 burned, almost all of the CO2 produced is prevented from entering the atmosphere by scrubbers
in the smoke stack. This reduction in CO2 emissions occurs while scrubbing to remove HF from the waste
stream (Oldach, 2000).
Cost and Reduction Assumptions
Cost estimates for thermal oxidation include the following assumptions:
• In the United States, total installed capital costs for a thermal oxidation system are assumed to be
approximately $3.4 million per plant in new plants (Rost, 2006) and $4.4 million per plant in
existing plants3 (Werling, 2006), with total annual operating costs of $334,928 per year (Lehman,
2002). These capital and annual costs are assumed for the United States and the rest of the world,
with the exception of the EU.
• In the EU, the total installed capital costs for a thermal oxidation system were estimated at
$2,834,447 million per plant, with total annual operating costs of $188,963 per year (Harnisch et
al., 2000).
Cost estimates for such systems are based upon the best available industry assessments; actual costs
of some systems could differ from these estimates.
Reduction estimates for thermal oxidation include the following assumptions:
• Based on international HCFC-22 production capacity data from the Chemical Economics
Handbook (CEH) (2001), the typical HCFC-22 plant outside of the EU was assumed to produce
20,000 tons of HCFC-22 annually. In the EU, plants were assumed to produce 10,000 tons of
HCFC-22 annually (Harnisch et al., 2000).
• Plants were assumed to emit HFC-23 at a rate of 2 percent of HCFC-22 production.
• As noted above, thermal oxidation was assumed to destroy 95 percent of HFC-23 emissions at
plants where it was applied.
Baseline Market Penetration of Thermal Oxidation
The maximum potential market penetration of this option is 100 percent. Thus, the abatement
potential of the option for any given year and region depends on the difference between the baseline
market penetration and 100 percent. Tables 8-3 and 8-4 present the baseline market penetration of
thermal abatement for 2010 and 2020 for the no-action and technology-adoption baseline scenarios.
The no-action scenario accounts only for the level of implementation of thermal oxidation at the time
this report was written. It does not account for additional implementation of thermal oxidation in the
future. (For the United States, the no-action scenario actually assumes that current abatement ceases.) In
contrast, the technology-adoption scenario accounts for additional implementation in the future. Most
additional thermal oxidation is assumed to be installed in developing countries as they conduct
mitigation projects, funded by developed countries under the CDM. The absolute level of abatement (in
MtCO2eq) for these projects is assumed to remain constant through 2020. Additional thermal oxidation is
also modeled for Spain, where the owner of the sole HCFC-22 plant has announced plans to install
thermal oxidation by 2010 (Campbell, 2006). These estimates are discussed in more detail in the USEPA
report Global Anthropogenic Non-CC>2 Greenhouse Gas Emissions: 1990-2020 (2006).
3 Ralph Werling and Kurt Werner of 3M estimated that the costs of installing thermal oxidation systems in existing
plants were 20 percent to 40 percent greater than the costs of installing the systems in new plants. This analysis
assumes that it costs 30 percent more to install a thermal oxidation system in an existing plant than in a new plant.
IV-160 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • HCFC-22 PRODUCTION
Table 8-3: Baseline Market Penetration of Thermal Oxidation—No-Action Baseline
Country
France
Germany
Italy
Netherlands
Japan
Russian Federation
Spain
United Kingdom
United States
India
Brazil
Mexico
Venezuela
China
Korea, Republic of
2010
100%
100%
100%
100%
100%
0%
0%
100%
0%
0%
0%
0%
0%
0%
0%
2020
100%
100%
100%
100%
100%
0%
0%
100%
0%
0%
0%
0%
0%
0%
0%
Table 8-4: Baseline Market Penetration of Thermal Oxidation—Technology-Adoption Baseline
Country
France
Germany
Italy
Netherlands
Japan
Russian Federation
Spain
United Kingdom
United States
India
Brazil
Mexico
Venezuela
China
Korea, Republic of
2010
100%
100%
100%
100%
100%
0%
100%
100%
65%
90%
0%
99%
0%
65%
26%
2020
100%
100%
100%
100%
100%
0%
100%
100%
65%
78%
0%
91%
0%
50%
23%
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-161
-------
SECTION IV— INDUSTRIAL PROCESSES • HCFC-22 PRODUCTION
Estimating Emissions from New Plants
The analysis also differentiates between emissions coming from new plants and those coming from
existing plants as different costs are associated with the abatement of these two sets of emissions. To
calculate emissions from new plants, it is assumed that all emissions growth after 2010 comes from new
plants. Since only developing countries will experience emissions growth after 2010, all new plants are
assumed to be built in developing countries.
IV.8.3 Results
This section discusses the result from the MAC analysis for the world and several regions for the no-
action and technology-adoption scenarios.
1V.8.3.1 Data Tables and Graphs
Based on the trends described above, the USEPA developed MACs for the world and several regions.
Tables 8-5 through 8-8 summarize the potential emissions reduction opportunities and associated costs
for the world and several regions in 2010 and 2020 for the no-action and technology-adoption baselines.
The costs to reduce 1 tCO2eq are presented for a discount rate of 10 percent and a tax rate of 40 percent.
Table 8-5: Emissions Reductions in 2010 and Breakeven Costs ($/tC02eq) for HFC-23 Emissions from
HCFC-22 Production at 10% Discount Rate, 40% Tax Rate (MtC02eq)—No-Action Baseline
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$15
0.00
26.80
0.00
0.19
66.69
0.00
1.14
7.58
0.00
3.80
0.66
35.41
0.66
0.00
25.00
110.78
$30
0.00
26.80
0.00
0.19
66.69
0.00
1.14
7.58
0.00
3.80
0.66
35.41
0.66
0.00
25.00
110.78
2010
$45
0.00
26.80
0.00
0.19
66.69
0.00
1.14
7.58
0.00
3.80
0.66
35.41
0.66
0.00
25.00
110.78
$60
0.00
26.80
0.00
0.19
66.69
0.00
1.14
7.58
0.00
3.80
0.66
35.41
0.66
0.00
25.00
110.78
>$60
0.00
26.80
0.00
0.19
66.69
0.00
1.14
7.58
0.00
3.80
0.66
35.41
0.66
0.00
25.00
110.78
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
IV-162 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • HCFC-22 PRODUCTION
Table 8-6: Emissions Reductions in 2020 and Breakeven Costs ($/tC02eq) for HFC-23 Emissions from
HCFC-22 Production at 10% Discount Rate, 40% Tax Rate (MtC02eq)—No-Action Baseline
2020
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$15
0.00
23.82
0.00
0.20
86.43
0.00
0.67
8.72
0.00
4.10
0.30
33.62
0.30
0.00
22.84
129.51
$30
0.00
23.82
0.00
0.20
86.43
0.00
0.67
8.72
0.00
4.10
0.30
33.62
0.30
0.00
22.84
129.51
$45
0.00
23.82
0.00
0.20
86.43
0.00
0.67
8.72
0.00
4.10
0.30
33.62
0.30
0.00
22.84
129.51
$60
0.00
23.82
0.00
0.20
86.43
0.00
0.67
8.72
0.00
4.10
0.30
33.62
0.30
0.00
22.84
129.51
>$60
0.00
23.82
0.00
0.20
86.43
0.00
0.67
8.72
0.00
4.10
0.30
33.62
0.30
0.00
22.84
129.51
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Table 8-7: Emissions Reductions in 2010 and Breakeven Costs ($/tC02eq) for HFC-23 Emissions from
HCFC-22 Production at 10% Discount Rate, 40% Tax Rate (MtC02eq)—Technology-Adoption
Baseline
2010
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$15
0.00
8.72
0.00
0.19
23.50
0.00
0.00
0.74
0.00
0.06
0.66
12.18
0.66
0.00
8.06
37.52
$30
0.00
8.72
0.00
0.19
23.50
0.00
0.00
0.74
0.00
0.06
0.66
12.18
0.66
0.00
8.06
37.52
$45
0.00
8.72
0.00
0.19
23.50
0.00
0.00
0.74
0.00
0.06
0.66
12.18
0.66
0.00
8.06
37.52
$60
0.00
8.72
0.00
0.19
23.50
0.00
0.00
0.74
0.00
0.06
0.66
12.18
0.66
0.00
8.06
37.52
>$60
0.00
8.72
0.00
0.19
23.50
0.00
0.00
0.74
0.00
0.06
0.66
12.18
0.66
0.00
8.06
37.52
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-163
-------
SECTION IV— INDUSTRIAL PROCESSES • HCFC-22 PRODUCTION
Table 8-8: Emissions Reductions in 2020 and Breakeven Costs ($/tC02eq) for HFC-23 Emissions from
HCFC-22 Production at 10% Discount Rate, 40% Tax Rate (MtC02eq)—Technology-Adoption
Baseline
J
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex I
OECD
Russian Federation
South & SE Asia
United States
World Total
2020
$0
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$15
0.00
7.66
0.00
0.20
43.24
0.00
0.00
1.89
0.00
0.35
0.30
12.32
0.30
0.00
7.36
58.19
$30
0.00
7.66
0.00
0.20
43.24
0.00
0.00
1.89
0.00
0.35
0.30
12.32
0.30
0.00
7.36
58.19
$45
0.00
7.66
0.00
0.20
43.24
0.00
0.00
1.89
0.00
0.35
0.30
12.32
0.30
0.00
7.36
58.19
$60
0.00
7.66
0.00
0.20
43.24
0.00
0.00
1.89
0.00
0'.35
0.30
12.32
0.30
0.00
7.36
58.19
>$60
0.00
7.66
0.00
0.20
43.24
0.00
0.00
1.89
0.00
0.35
0.30
12.32
0.30
0.00
7.36
58.19
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Table 8-9: World Breakeven Costs and Emissions Reductions in 2020—No-Action Baseline
Reduction Option
Thermal oxidation (new plants)
Thermal oxidation (existing plants)
Cost
(2000S/tC02eq)
DR=10%,
TR=40%
Low High
$0.23 $0.23
$0.28 $0.35
Emissions
Reduction of
Option
(MtC02eq)
21.72
107.80
Reduction
from 2020
Baseline
(%)
15.8%
78.4%
Running
Sum of
Reductions
(MtC02eq)
21.72
129.51
Cumulative
Reduction
from 2020
Baseline
(%)
15.8%
94.2%
Table 8-10: World Breakeven Costs and Emissions Reductions in 2020—Technology-Adoption Baseline
Reduction Option
Thermal oxidation (new plants)
Thermal oxidation (existing plants)
Cost
(2000S/tC02eq)
DR=10%,
TR=40%
Low High
$0.23 $0.23
$0.28 $0.35
Emissions
Reduction of
Option
(MtC02eq)
20.49
37.70
Reduction
from 2020
Baseline
(%)
31.0%
57.0%
Running
Sum of
Reductions
(MtC02eq)
20.49
58.19
Cumulative
Reduction
from 2020
Baseline
(%)
31.0%
87.9%
IV-164
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV— INDUSTRIAL PROCESSES . HCFC-22 PRODUCTION
IV.8.3.2 Global and Regional MACs and Analysis
This section discusses the results from the MAC analysis of the world and selected countries and
regions, including China, Japan, the United States, the EU-15, other OECD, and the rest of the world.
Figure 8-3 presents the 2010 and 2020 global technology-adoption and no-action MACs for HCFC-22
production. As shown in Figure 8-3, the MACs include different cost points depending on the scenario
and year. The 2020 no-action MAC includes all three cost points: $0.35/tCO2eq for thermal oxidation in
the EU-15, $0.28/tCO2eq for thermal oxidation at all existing plants in all other regions, and $0.23/tCO2eq
for thermal oxidation at new plants, which are assumed to exist only in developing countries in 2020. The
2010 MACs omit the cost point for thermal oxidation at new plants because no new plants are assumed to
be built until after 2010. Similarly, the technology-adoption MACs exclude the cost point for thermal
oxidation in the EU because the EU is assumed to have fully implemented thermal oxidation in the
baseline in the technology-adoption scenario. Costs (in terms of $/tCO2eq) are slightly higher in EU-15
than in other parts of the world because this analysis uses EU-specific values for capital costs and average
emissions per facility, which together result in a slightly higher calculated cost per tCO2eq.
Figure 8-3: 2010 and 2020 Global Technology-Adoption and No-Action MACs for HCFC-22 Production
$0.40 -i
ST $0-38-
§ $0.36 -
£. $0.34 -
tn
§ $0.32 -
o $0.30 -
« $0.28 -
£ $0.26 -
I $0.24-
,§ $0.22 -
$0.20 -
C
)
2010
2020
25 50 75
2010
T
2020
echnology-Adoption
o-Action
100 125 150
Cumulative Emissions Reductions (MtCO2eq)
As shown in Figure 8-3, in the no-action scenario, HCFC-22 production offers global emissions
reductions of about 111 MtCO2eq and 130 MtCO2eq in 2010 and 2020, respectively. The 17 percent
increase in emissions reductions between 2010 and 2020 is a result of baseline emissions increases in
developing countries, mainly China, between 2010 and 2020. Option costs are not assumed to vary
between 2010 and 2020; therefore, the additional emissions abatable in 2020 shifts the 2020 MAC slightly
to the right compared to the 2010 MAC.
In the technology-adoption scenario, HCFC-22 production offers global emissions reductions of about
38 MtCO2eq and 58 MtCO2eq in 2010 and 2020, respectively. Available reductions are smaller than in the
no-action MAC because more emissions are reduced in the technology-adoption baseline. The 55 percent
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-165
-------
SECTION IV— INDUSTRIAL PROCESSES • HCFC-22 PRODUCTION
increase in emissions reductions between 2010 and 2020 is, again, a result of baseline emissions increases
in developing countries, mainly China, between 2010 and 2020. Figures 8-4 and 8-5 present regional
MACs for 2010 and 2020 under the technology-adoption scenario.
Figure 8-4: 201 0 Regional Technology-Adoption MACs
^ $0.40 J
o $0.38 -
8 $0.36 -
«;
£. $0.34 -
(0
§ $0.32 -
« $0.30 -
| $0.28 -
» $0.26 -
•1 $0.24 -
w
1 $0.22 -
LLI
— * China
Rest of the world
— • United States
- EU-15
— — > Japan
* Other OECD
0 5 10 15 20 25
Cumulative Emissions Reductions (MtCO2eq)
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Figure 8-5: 2020 Regional Technology-Adoption MACs
f $0.36 -
y $0.34 -
7 $0.32 -
c
~ $0.30 -
u
| $0.28 -
{£.
g $0.26 -
o
g $0.24 -
• tfl 99
i ¥ <
i
i
' — * China
Rest of the world
• United States
= EU-15
— > Japan
* Other OECD
i
I
«
0 10 20 30 40 50
Cumulative Emissions Reductions (MtC02eq)
EU-15 = European Union; OECD = Organisation ol Economic Co-operation and Development.
IV-166
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • HCFC-22 PRODUCTION
Large emissions reductions are available in China even in the technology-adoption scenario. These
result from China's large expected production of HCFC-22, relatively high emissions rate (3 kg HFC-
23/100kg HCFC-22), and incomplete adoption of thermal oxidation in the baseline. China's expected 2010
HFC-23 emissions make up more than 60 percent of total global emissions in that year, and China's
reductions make up 63 percent of the total global reductions. (These percentages differ slightly because
some of the emissions from other regions are residual emissions that cannot be reduced further.) U.S.
reductions make up much of the remainder of available reductions because, like China, the United States
is a large producer of HCFC-22 (accounting for about 20 percent of global production in 2010 and 2020)
and is also assumed not to fully implement thermal abatement in the baseline. Emissions and reductions
from other regions are expected to be smaller because of (1) a smaller growth rate for HCFC-22
production (this growth rate is actually negative in most developed countries because of their ongoing
phaseout of most uses of HCFC-22 under the Montreal Protocol) and (2) the widespread use of abatement
technologies in developed countries.
Large emissions reductions are also available for 2020 in China both in existing and new plants.
China's expected 2020 HFC-23 emissions make-up 72 percent of total global emissions in that year, and
China's reductions make up 74 percent of the total global reductions.
IV.8.3.3 Uncertainties and Limitations
This section focuses on the uncertainties associated with the cost estimates presented in this report.
Uncertainties regarding emissions estimates are discussed in the USEPA report Global Anthropogenic Non-
CO2 Greenhouse Gas Emissions: 1990-2020 (USEPA, 2006).
There is some uncertainty associated with the costs of thermal oxidation. Currently, costs are
available for three versions of thermal oxidation based on data from the United States and the EU,
respectively. For these two regions, the estimated breakeven prices are expected to be reasonably robust,
an expectation that is supported by the fact that the breakeven prices for the United States and EU
versions of the technology are quite similar despite significant differences in capital costs and reductions.
Outside of the United States and EU, costs and breakeven prices are less certain. U.S. capital costs and
annual costs were applied to all countries outside of the EU. However, these countries may be faced with
higher costs from transportation and tariffs associated with purchasing the technology from abroad, or
with lower costs if there is domestic production of these technologies.
The estimated cost per tCO2eq is very sensitive to the assumed HCFC-22 production level and HFC-
23 emissions rate of plants where thermal oxidation is assumed to be installed. The capital cost
information used in this analysis was for an oxidation system with a 7 to 10 million Btu capacity. This
capacity is large enough to oxidize HFC-23 emissions from the largest plant in the world, which has a
production capacity of 100,000 tons. However, because most plants have capacities closer to 20,000 tons,
this analysis uses that production as the basis for the cost estimates. This may overestimate the
cost/tCO2eq at larger plants and underestimate it at smaller plants. Similarly, this analysis conservatively
uses a 2 percent emissions factor in its reduction estimates; plants with higher emissions factors rates
reduce more emissions by installing thermal abatement.
Future production levels, emissions rates, and abatement levels are particularly uncertain. Future
policies (e.g., under the Montreal Protocol) could affect total production of HCFC-22 and therefore
emissions of HFC-23. Changing emissions rates may also have a significant impact on emissions. In the
technology-adoption baseline, the USEPA assumed that currently identified COM projects will be
implemented in China, India, Korea, and Mexico. However, even after implementation of these projects,
significant reduction opportunities remain, both in these countries and elsewhere. There is a significant
probability that many of these emissions will be averted, either through COM or other mechanisms. In
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-167
-------
SECTION IV — INDUSTRIAL PROCESSES • HCFC-22 PRODUCTION
this case, HFC-23 emissions will be lower than projected in the technology-adoption baseline. Such a
decrease in emissions would also decrease the reductions available in the technology-adoption MACs.
IV.8.4 References
CEH. 2001. Fluorvcarbons, CEH Marketing Research Report, 2001. Available from the Chemical Economics
Handbook—SRI International.
Harnisch, J., and C.A. Hendriks. 2000. Economic Evaluation of Emission Reductions ofHFCs, PFCs and SF6 in
Europe: Special Report, a contribution to the study "Economic Evaluation of Sectoral Emission
Reduction Objectives for Climate Change" on behalf of the Directorate General Environment of the
Commission of the European Union, Prepared by Ecofys Energy and Environment, Brussels.
Intergovernmental Panel on Climate Change (IPCC). 2001. Climate Change 2001, The Scientific Basis.
Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on
Climate Change. Cambridge University Press.
JICOP. May 9, 2006. E-mails to Deborah Ottinger Schaefer of the USEPA from Mr. Shigehiro Uemura of
JICOP.
Lehman, G. 2002. Personal communication with Gail Lehman, General Council, Fluorine Products,
Corporate Law Department, Honeywell International Inc.
Oldach, R. 2000. Phone conversation between Robert Oldach, senior engineer with DuPont's Research
Group, and Carrie Smith on August 10, 2000.
Rand, S., D. Ottinger, and M. Branscome. May 26-28, 1999. Opportunities for the Reduction of HFC-23
Emissions from the Production of HCFC-22. IPCC/TEAP Joint Expert Meeting. Petten, Netherlands.
Rost, M. April 24, 2006. Personal communication between Marc Rost of T-thermal and Debora Ottinger of
the USEPA.
United Nations Environment Programme (UNPE). 2003. Report of the Technology and Economic Assessment
Panel. HCFC Task Force Report.
U.S. Environmental Protection Agency (USEPA). 2006. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2020. Washington, DC: USEPA.
Werling, R. April 24, 2006. Telephone conversation between Deborah Ottinger Schaefer of the LFSEPA and
Ralph Werling and Kurt Werner of 3M.
IV-168 GLOBAL MITIGATION OF NON C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
IV.9 RFC and SF6 Emissions from Semiconductor
Manufacturing
IV.9.1 Source Description
he semiconductor industry currently uses several fluorinated compounds (CF4/ C2F6/ C3F8/
C4F8, HFC-23, NF3, and SF6) during the fabrication process.1 A fraction of each of these gases is
emitted during two frequently used manufacturing process steps: the plasma etching of thin
films and the cleaning of chemical-vapor-deposition chambers.2 In addition, by-product emissions of CF4
result when a fraction of the heavier gases consumed is converted during the manufacturing process or
when F-atoms produced in a plasma react with the carbon present in certain low-dielectric strength films
for CF4. Total PFC emissions from this source vary by process and device type.3 Tables 9-1 and 9-2
present estimates of historical and forecasted semiconductor manufacturing PFC emissions for 1990
through 2020 under two different scenarios.
Table 9-1: Total PFC Emissions from Semiconductor Manufacturing (MtC02eq)—No-Action Baseline
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex I
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
0.1
17.0
,0.0
0.0
0.8
0.2
1.9
0.2
7.4
0.1
0.8
19.6
0.8
0.6
6.4
27.4
2010
0.1
47.2
0.0
0.0
10.7
0.9
5.3
0.6
11.0
0.0
1.5
59.1
1.4
5.7
28.2
99.2
2020
0.2
74.3
0.1
0.0
37.5
1.5
8.5
1.0
15.3
0.0
2.3
98.1
2.3
28.3
46.1
231.9
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
1 The chemical compound CHF3 is more commonly referred to as HFC-23; thus, the latter term is used here.
2 Very small amounts of CF4 are emitted during a process step called ashing or photoresist stripping. Because
emissions from this process are considered very small, they are not included.
3 Note that although the term "PFC" (strictly referring to only perfluorocarbon compounds) does not include all of
the fluorinated compounds emitted from this source, the semiconductor industry commonly refers to the mix of
fluorinated compounds as PFCs; this report adopts the same convention.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-169
-------
SECTION IV— INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
Table 9-2: Total PFC Emissions from Semiconductor Manufacturing (MtC02eq)—Technology-Adoption
Baseline
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex I
OECD
Russian Federation
South & SE Asia8
United States
World Total
2000
0.1
17.0
0.0
0.0
0.8
0.2
1.9
0.2
7.4
0.1
0.8
19.6
0.8
0.6
6.4
27.4
2010
0.1
12.9
0.0
0.0
10.7
0.9
1.2
0.6
3.7
0.0
1.5
13.8
1.4
5.7
5.5
36.9
2020
0.1
10.7
0.0
0.0
7.1
0.6
1.2
0.4
3.7
0.0
1.0
12.1
1.0
3.8
4.1
28.3
a Note that the region South and Southeast Asia (South & SE Asia) in the table above includes different countries than South and East Asia
(S&E Asia) as defined in the Global Anthropogenic Non-C02 Emissions: 1990-2020 (USEPA, 2006) and in Figure 9-1. South and East Asia
in Figure 9-1 includes the major semiconductor manufacturing regions of Taiwan and South Korea, while South and Southeast Asia excludes
these regions.
IV.9.1.1 Technology-Adoption Baseline
The technology-adoption baseline incorporates those reductions that have resulted or are anticipated
to result from international voluntary climate commitments. In April 1999, the semiconductor
manufacturing industry set an aggressive target to reduce PFC emissions. The World Semiconductor
Council (WSC) then agreed to reduce PFC emissions to 10 percent below 1995 levels by the year 2010.4
Because WSC members then accounted for production of over 90 percent of the world's semiconductors,
the goal is expected to have dramatic effects in decreasing emissions over time, which would widen the
gap over time between emissions forecasts shown under the two scenarios presented in Figure 9-1 and
Figure 9-2 (note that the scales are different in the two graphs).
OECD and Asia (including China/CPA and South and East Asia) regions are expected to account for
the vast majority of production, and therefore also the emissions, throughout the time horizon studied.
The highest-emitting countries worldwide in 2000 were Japan, the United States, Taiwan, South Korea,
and Germany. By 2010 and through 2020, the highest-emitting country worldwide is expected to be
China, followed by the United States, Japan, South Korea, Singapore,5 and Malaysia. The appearance of
4 The base year for South Korea is 1997.
5 This reflects the top emitting countries in 2020, in descending order of emissions; in 2010, Singapore has greater
emissions than South Korea.
IV-170 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
Figure 9-1: RFC Emissions from Semiconductor Manufacturing Based on a Technology-Adoption
Scenario—1990 through 2020 (MtC02eq)
-------
SECTION IV — INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
China, Singapore, and Malaysia6 among the top emitting countries reflects a geographic shift in
production such that the majority of future growth takes place in these countries. This reflects an industry
trend toward outsourcing production to dedicated manufacturing firms, called foundries, concentrated in
these countries.
Global emissions are estimated to have grown at a compound annual growth rate of 11 percent per
year through the year 2000. Following the introduction of voluntary commitments and resulting
mitigation efforts, however, a noticeable shift in direction is expected to occur under the technology-
adoption scenario. As shown in Figure 9-1, the overall trend in OECD emissions is reflected in the
emissions from the United States, the European Union (EU-25), and Japan. These regions, where most
manufacturers are WSC members, are expected to achieve the WSC goal collectively by 2010. In the long
run, even countries whose manufacturers have not adopted the WSC goal, such as China, Singapore, and
Malaysia—countries not part of the WSC, are assumed to reduce their emissions rates as new, lower-
emitting, more productive manufacturing equipment enters the global market. This expectation accounts
for the reduction in emissions from China and South and East Asia between 2010 and 2020.
Figure 9-2 shows the relative distribution of global emissions under the technology-adoption scenario
between WSC and non-WSC members and illustrates these trends even more clearly. Note that emissions
from WSC countries peaked in 2000.
IV.9.1.2 No-Action Baseline
The no-action scenario estimates emissions that would result from normal industry activity with no
emissions control measures, voluntary or regulation driven. This trajectory can be considered an upper
bound and can serve as a reference level to which the alternative technology-adoption scenario emissions
can be compared. The difference between these two emissions sets represents the emissions reductions
achieved by semiconductor manufacturers as they implement emissions control technologies or other
mitigation measures.
Figure 9-3 shows the relative distribution of global emissions under the no-action scenario. As in the
technology-adoption scenario, the OECD and Asia regions are expected to remain the largest emitters
throughout the time horizon studied; emissions from these two regions (including OECD90+, China/CPA,
and South and East Asia) combined are expected to make up 98 percent of global emissions in 2020.
Historical trends are the same as those presented for the technology-adoption baseline, including the
11 percent per year annual growth through 2000. However, in the no-action baseline, this high annual
growth continues virtually unabated through 2010 and is particularly pronounced in Asia beyond 2010.
In these countries, most notably China, Singapore, and Malaysia, semiconductor manufacturing is
assumed to increase significantly, as discussed above in the no-action baseline, contributing to higher
emissions over the time horizon presented. Beyond 2010, the growth rate is assumed to decline by one-
half, reflecting slower growth in demand for semiconductors. Nevertheless, global emissions are expected
to continue to climb substantially, reaching 232 MtCO2eq by 2020.
' As of May 2006, China, Singapore, and Malaysia have not joined the WSC.
IV-172 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
Figure 9-3: RFC Emissions from Semiconductor Manufacturing Based on a No-Action Scenario—1990
through 2020 (MtC02eq)
250
D Africa
• Middle East
• Latin America
DNon-EU FSU
• China/CPA
• S&E Asia
B Japan
• EU-25
D United States
• OECD90+
1990
2000
2010
2020
Year
CPA = Centrally Planned Asia; EU-25 = European Union; S&E Asia = South and East Asia; Non-EU FSU = non-European Union Former Soviet
Union countries; OECD90+ = Organisation for Economic Co-operation and Development.
IV.9.2 Cost of PFC and SF6 Emissions Reduction from Semiconductor
Manufacturing
IV.9.2.1 Abatement Options
Overview of Options and Analysis
This analysis considers six different emissions reduction technologies applicable to semiconductor
manufacturing. These options are shown in Tables 9-3 through 9-8 below. The five main characteristics
that determine the reductions achieved by each technology in each scenario are (1) whether the
technology is applicable to plasma etch processes, chemical vapor deposition (CVD) chamber cleaning
processes, or both; (2) the maximum share of the etch or clean market that is assumed to be claimed by
the technology relative to a baseline with no preexisting emissions controls; (3) the share of the etch or
clean market that is already claimed by the technology in the baseline of concern; (4) the reduction
efficiency of the technology; that is, the percentage by which the technology reduces the emissions stream
to which it is applied; and (5) for non-WSC countries, the year, because only a fraction of the full
reductions are assumed to be available to these countries in 2010. Of these characteristics, (4) and
sometimes (1) affect the cost per tCO2eq of the technology's reductions, while all five characteristics affect
the size and shape of the aggregate MACs.
In general, technologies applicable to CVD chamber cleaning achieve larger reductions than those
applicable only to etch processes because CVD chamber cleaning is estimated to account for 80 percent of
the emissions from semiconductor manufacturing, while etch is estimated to account for 20 percent. The
maximum share of the etch or clean market has been estimated for each technology by region and year
based on that technology's cost-effectiveness and applicability, industry trends, and expert judgment. The
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-173
-------
SECTION IV — INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
maximum shares for the no-action scenario are shown below in Tables 9-3 and 9-4. In the no-action
scenario, none of the technologies is assumed to be implemented in the baseline; thus, the market
penetrations provided in Tables 9-3 and 9-4 are percentages of the total emissions in the no-action
scenario for that year. That is, the percentages in Tables 9-3 and 9-4 correspond to the maximum shares
described in (2) above.
Table 9-3: Maximum Market Penetrations for WSC Countries in the No-Action Baseline (Percent)3
Option
Thermal destruction
Catalytic decomposition
Capture/recovery
Plasma abatement
NF3 remote clean
C3F8 replacement
Plasma Etching Process
5%
5%
15%
70%
0%
0%
CVD Chamber Cleaning Process
5%
5%
15%
0%
70%
5%
a Assumed market penetration of technology, presented as a percentage of no-action baseline emissions that result from etching or CVD
chamber cleaning, respectively.
Table 9-4: Maximum Market Penetrations for Non-WSC Countries in the No-Action Baseline (Percent)3
Option
Thermal destruction
Catalytic decomposition
Capture/recovery
Plasma abatement
NF3 remote clean
C3F8 replacement
Plasma Etching
Process
2010
3%
3%
2%
30%
0%
0%
CVD Chamber
Cleaning
Process
3%
3%
2%
0%
30%
2%
Plasma Etching
Process
CVD Chamber
Cleaning
Process
2020
7%
8%
5%
75%
0%
0%
7%
8%
5%
0%
75%
5%
a Assumed market penetration of technology, presented as a percentage of no-action baseline emissions that result from etching or CVD
chamber cleaning, respectively.
In the technology-adoption scenario, semiconductor manufacturers are assumed to implement
reduction technologies in the baseline to the extent necessary to achieve the WSC goal. Tables 9-5 and 9-7
provide the baseline market penetrations of the various technologies in the technology-adoption baseline
for WSC countries and non-WSC countries, respectively. To estimate the emissions reductions remaining
in the technology-adoption MACs after implementing the technologies shown in Tables 9-5 and 9-7, the
shares in Tables 9-5 and 9-7 are subtracted from the corresponding shares in Tables 9-3 and 9-4. The
resulting percentages are then recast in terms of the emissions that remain unabated in the technology-
adoption baseline. (These are different from the total emissions because the technology-adoption baseline
includes residual emissions from emissions streams to which technologies have already been applied.)
For example, to obtain the market share for remote clean in the WSC in the 2010 technology-adoption
baseline, the 57 percent in Table 9-5 is subtracted from the 70 percent in Table 9-3, and the difference is
then divided by 14 percent, the sum of the remaining, unused shares for the technologies applicable to
CVD chamber cleaning. These results are shown in Tables 9-6 (for WSC countries) and 9-8 (for non-WSC
countries).
IV-174
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
Table 9-5: Baseline Market Penetrations for WSC Countries in the Technology-Adoption Baseline (Percent)3
Option
Thermal destruction
Catalytic decomposition
Capture/recovery
Plasma abatement
NF3 remote clean
C3F8 replacement
Plasma Etching
Process
2010
4%
5%
15%
57%
0%
0%
CVD Chamber
Cleaning
Process
5%
5%
15%
0%
57%
4%
Plasma Etching
Process
CVD Chamber
Cleaning
Process
2020
4.8%
5.0%
15.0%
67.2%
0.0%
0.0%
5.0%
5.0%
15.0%
0.0%
67.2%
4.8%
a Assumed market penetration of technology, presented as a percentage of no-action baseline emissions that result from etching or CVD
chamber cleaning, respectively. -
Table 9-6: Maximum Market Penetrations for WSC Countries in the Technology-Adoption Baseline (Percent)3
Option
Thermal destruction
Catalytic decomposition
Capture/recovery
Plasma abatement
NF3 remote clean
C3F8 replacement
Plasma Etching
Process
2010
5%
0%
0%
69%
0%
0%
CVD Chamber
Cleaning
Process
0%
0%
0%
0%
93%
7%
Plasma Etching
Process
CVD Chamber
Cleaning
Process
2020
3%
0%
0%
35%
0%
0%
0%
0%
0%
0%
93%
7%
a Assumed market penetration of technology, presented as a percentage of the technology-adoption baseline emissions from etching or CVD
chamber cleaning that remain available for abatement.
Table 9-7: Baseline Market Penetrations for Non-WSC Countries in the Technology-Adoption Baseline in
2020 (Percent)3
Option
Thermal destruction
Catalytic decomposition
Capture/recovery
Plasma abatement
NF3 remote clean
C3F8 replacement
Plasma Etching Process
6%
8%
5%
62%
0%
0%
CVD Chamber Cleaning Process
7%
8%
5%
0%
62%
4%
Assumed market penetration of technology, presented as a percentage of no-action baseline emissions that result from etching or CVD
chamber cleaning, respectively.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-175
-------
SECTION IV — INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
Table 9-8: Maximum Market Penetrations for Non-WSC Countries in the Technology-Adoption Baseline
(Percent)3
Option
Thermal destruction
Catalytic decomposition
Capture/recovery
Plasma abatement
NF3 remote clean
C3F8 replacement
Plasma Etching
Process
2010
3%
3%
2%
30%
0%
0%
CVD Chamber
Cleaning
Process
3%
3%
2%
0%
30%
2%
Plasma Etching
Process
CVD Chamber
Cleaning
Process
2020
6%
0%
0%
68%
0%
0%
0%
0%
0%
0%
94%
6%
a Assumed market penetration of technology, presented as a percentage of the technology-adoption baseline emissions from etching or CVD
chamber cleaning that remain available for abatement.
For WSC countries, the full reductions from each technology are assumed to be available in 2010, as
shown in Table 9-3. For non-WSC countries, only 40 percent of the full reductions are assumed to be
available in 2010, but this percentage grows to 100 percent in 2020, as shown in Table 9-4.
NF3 Remote Clean Technology
The NF3 Remote Clean system is used to abate emissions from the chemical vapor deposition (CVD)
chamber cleaning process and is assumed to be applicable to all fabrication facilities. As noted above,
CVD chamber cleaning emissions are reported to constitute approximately 80 percent of all
semiconductor emissions. The system dissociates 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 Clean include HF, F^ arid other gases, of
which all but F2 are removed by facility acid scrubber systems.
This analysis assumes that the emissions reduction efficiency of this option is 95 percent. The
assumed maximum market penetrations of this option for WSC member countries and non-WSC
countries in the no-action baseline and technology-adoption baseline are presented in Tables 9-3 through
9-8.
Cost and Emissions Reduction Analysis
• Capital/Upfront Costs. Facilities moving to an NF3 Remote Clean system are assumed to face a
purchase and installation capital cost of $59,900 per chamber (Burton, 2003a).
• Annual Costs. Facilities operating NF3 Remote Clean systems are assumed to pay annual fees of
$11,000 per chamber for a preventative maintenance kit (Burton, 2003a) and to incur additional
costs equal to the difference in price between NFS and C2F6. Accounting for the amount of gases
used and their relative prices, an annual cost of $3,800 per chamber is assumed (Burton, 2003a).
Therefore, net annual costs are assumed to total $14,800 per chamber.
• Cost Savings. Facilities that install NFS Remote Clean systems achieve chamber-cleaning times
that are 30 to 50 percent faster than baseline C2F6 cleaning times (International SEMATECH,
1999) and decrease the number of cleanings between wafer passes. The end result is an increase
in the time devoted to the actual manufacturing portion of the process, which allows high-
utilization facilities to recoup their capital costs in an estimated 9 months or less. Because of this
IV-176 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
process improvement, assuming a 9-month capital return, it can be calculated that facilities
receive a cost savings of one and one-third times the capital cost, or $79,867 per chamber, on an
annual basis. (Burton, 2003b).
C3F8 Replacement
C3F8 replacement is used to abate emissions from the CVD chamber cleaning process and is assumed
to be applicable to all fabrication facilities. The C3F8 simply replaces C2F6, which reduces emissions
because C2F6 has a 100-year global warming potential (GWP) of 9,200, whereas C3F8 has a 100-year GWP
of 7,000 and an atmospheric lifetime that is less than one-third that of C2F6 (IPCC, 1996). In addition, C3F8
is more efficiently used/consumed during CVD chamber cleaning than C2F6 (and produces about the
same amount of CF4 during cleaning), which, combined with the differences in GWP, yields an assumed
emissions reduction efficiency of 85 percent. The assumed maximum market penetrations for WSC and
non-WSC countries under the no-action and technology-adoption scenarios are presented in Tables 9-3
through 9-8.
Cost and Emissions Reduction Analysis
• Capital/Upfront Costs. Because the C3F8 simply replaces the C2F6, it is assumed that facilities do
not incur any capital costs (Burton, 2003a).
• Annual Costs. The cost of C3F8 is assumed to equal the cost of C2F6, so the replacement results in
no annual costs (Burton, 2003a).
• Cost Savings. It is assumed that no cost savings are associated with this technology.
Point-of-Use Plasma Abatement
The Point-of-Use Plasma Abatement system is used to abate emissions from the plasma etching
process and is assumed to be applicable to all fabrication facilities. Plasma etching emissions constitute 20
percent of all semiconductor emissions. The system uses a small plasma source that effectively dissociates
the PFC molecules that react with fragments of the additive gas—H2, O^ H2O, or CH4—in order to
produce low-molecular-weight by-products such as 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 (Motorola, 1998). The evaluations performed to date indicate
no apparent interference with the etch process.
This analysis assumes that the emissions reduction efficiency of this option is 95 percent. The
assumed maximum market penetrations for WSC and non-WSC countries under the no-action and
technology-adoption scenarios are presented in Tables 9-3 through 9-8.
Cost and Emissions Reduction Analysis
• Capital/Upfront Costs. It is assumed that plasma abatement technology requires capital costs of
$35,000 per etching chamber, which covers the purchase and installation of the system (Burton,
2003a).
• Annual Costs. Facilities with plasma abatement technology are assumed to incur an annual
$1,000 operational expense per etch chamber (Burton, 2003a).
• Cost Savings. It is assumed that there are no cost savings associated with this technology.
Capture/Recovery
The capture/recovery membrane is used to abate emissions from both the plasma etching and CVD
chamber cleaning processes and is assumed to be applicable to all fabrication facilities. The
capture/recovery membrane separates unreacted and/or process-generated PFCs from other gases for
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-177
-------
SECTION IV — INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
further processing. The treatment process allows for the possibility of some reuse of the captured PFC gas
(Mocella, 1998). These capture/recovery systems can either reprocess the PFC for reuse or they can
concentrate the gas for subsequent off-site disposal. Because reprocessing inevitably produces PFC gas
that is less pure than virgin PFCs, semiconductor process engineers have little or no interest in reusing
the gas for fear of the possible process-harming impurities (Burton, 2003b). The lack of interest in PFC
reuse for semiconductor manufacturing combines with the lack of market for reprocessed PFC gas
outside the industry to make destruction highly attractive (Mocella, 1998; Burton, 2003b). Although a few
companies have installed pilot PFC capture/recovery systems, this technology is reported to be
unattractive if NF3 cleaning systems are used, because such cleaning processes do riot leave sufficient
PFCs in the stream to make gas recovery economically viable. In general, removal efficiencies for C276,
CF4, SF6, and C3F8 are in the high 90s, whereas CHF3 and NF3 removal efficiencies fall between 50 percent
and 60 percent.
This analysis assumes that the overall emissions reduction efficiency of this option is 96 percent
(International SEMATECH, 1999). The assumed maximum market penetrations for WSC and non-WSC
countries under the no-action and technology-adoption scenarios are presented in Tables 9-3 through 9-8.
Cost and Emissions Reduction Analysis
• Capital/Upfront Costs. Because the equipment is leased, capital costs associated with a
capture/recovery membrane include only installation and structural changes for preparing the
facility and its individual chambers for the membrane system. This analysis assumes that total
capital costs are $1,105,000 per facility, assuming that a standard facility has 200 chambers with a
4-to-l ratio of etch chambers to CVD chambers (Burton, 2003a).
• Annual Costs. Facilities are assumed to lease the equipment and an operator for an annual cost
of $300,000 (Burton, 2003a). Additionally, they are assumed to incur annual utility charges, which
encompass gas destruction, water, electricity, and all other costs, of $60,000—for a total annual
cost of $360,000 per facility.
• Cost Savings. It is assumed that there are no cost savings associated with this technology.
Catalytic Decomposition System
The catalytic decomposition system is used to abate emissions from both the plasma etching and
CVD chamber cleaning processes and is assumed to be applicable to all fabrication facilities. Catalytic
decomposition systems are installed in the process after the turbo pump, which dilutes the exhaust
stream prior to feeding it through the scrubber and emitting the scrubbed gases into the atmosphere.
Consequently, there is no back flow into the etching tool itself that could adversely affect the performance
of the etching tool. Because catalytic destruction systems operate at low temperatures, they also produce
little or no NOX emissions and they demand low volumes of water. Although the technology is applicable
at all fabrication facilities, off-the-shelf systems must be stream- or process-specification-specific, built to
reflect a certain minimum concentration and flow of PFC within the exhaust stream.
This analysis assumes that the emissions reduction efficiency of this option is 99 percent
(International SEMATECH, 1999). The assumed maximum market penetrations for WSC and non-WSC
countries under the no-action and technology-adoption scenarios are presented in Tables 9-3 through 9-8.
Cost and Emissions Reduction Analysis
• Capital/Upfront Costs. The purchase and installation capital costs associated with a catalytic
decomposition system are assumed to total $250,000 per every four etching chambers (Burton,
2003a).
IV-178 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
• Annual Costs. It is assumed that facilities incur annual costs totaling $19,750 per every four
etching chambers (Burton, 2003a). These costs are assumed to cover annual waste discharge
treatments, catalyst replacements, and utility charges.
• Cost Savings. It is assumed that no cost savings are associated with this technology.
Thermal Destruction/Thermal Processing Units (TPU)
The thermal destruction system is used to abate emissions from both the plasma etching and CVD
chamber cleaning processes and is assumed to be applicable to all fabrication facilities. Thermal
destruction technology is advantageous because it does not affect the manufacturing process (Applied
Materials, 1999). However, the combustion devices use significant amounts of cooling water, which
requires treatment as industrial wastewater. Finally, thermal oxidation may also produce NOX emissions,
which are regulated air pollutants.
This analysis assumes that the emissions reduction efficiency of this option is 97 percent. The increase
in other greenhouse gas emissions, both from the process-related burning of natural gas and from the
electricity demand, may reduce the efficiency of this option (Burton, 2003a). Future analysis could be
conducted to quantify the net reduction efficiency, which is expected to be closer to 90 percent (Burton,
2003a). The assumed maximum market penetrations for WSC and non-WSC countries under the no-
action and technology-adoption scenarios are presented in Tables 9-3 through 9-6.
Cost and Emissions Reduction Analysis
• Capital/Upfront Costs. Thermal decomposition systems are assumed to require capital costs
totaling $189,850 per every four etching chambers, which covers the purchase of the system,
installation, natural gas costs, and the installation of a water circulation unit (Burton, 2003a).
• Annual Costs. It is assumed that facilities incur annual costs of $11,100 per every four etching
chambers to cover system maintenance, waste disposal, and input purchases (Burton, 2003a).
• Cost Savings. It is assumed that no cost savings are associated with this technology.
IV.9.3 Results
IV.9.3.1 Data Tables and Graphs
Tables 9-9 through 9-12 provide a summary of semiconductor manufacturing emissions reductions at
a 10 percent discount rate and 40 percent tax rate by cost per metric ton of carbon dioxide equivalent
(tCO2eq) for various countries/regions of the world in 2010 and 2020 under the no-action and technology-
adoption scenarios. Table 9-13 and 9-14 provide a breakdown of the costs associated and the global
emissions reductions associated with implementing each abatement option under the two baseline
scenarios in 2020.
IV.9.3.2 Global and Regional MACs and Analysis
Global Trends
This section discusses the results from the MAC analysis of the world and selected countries/regions.
In the technology-adoption scenario, which is based on the assumption that World Semiconductor
Council manufacturers meet their global goal of reducing emissions to 90 percent of 1995 levels by 2010,
worldwide emissions reductions of up to 18.3 MtCO2eq are available in 2010 at a cost below $25/tCO2eq
In 2020, global reductions of 14.5 MtCO2eq are available at the same cost. In both years, significant
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-179
-------
SECTION IV — INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
Table 9-9: Emissions Reductions in 2010 and Breakeven Costs ($/tC02eq) at 10% Discount Rate, 40% Tax
Rate (MtC02eq)—No-Action Baseline
2010
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.0
25.9
0.0
0.0
2.6
0.3
3.0
0.2
6.2
0.0
0.4
33.1
0.4
1.4
16.0
49.3
$15
0.0
32.4
0.0
0.0
2.8
0.3
3.7
0.2
7.8
0.0
0.4
41.5
0.4
1.5
20.0
61.0
$30
0.1
40.7
0.0
0.0
3.7
0.4
4.7
0.2
9.8
0.0
0.5
52.1
0.5
2.0
25.2
76.9
$45
0.1
43.0
0.0
0.0
4.0
0.4
5.0
0.2
10.3
0.0
0.6
55.0
0.5
2.1
26.6
81.5
$60
0.1
43.0
0.0 .
0.0
4.0
0.4
5.0
0.2
10.3
0.0
0.6
55.0
0.5
2.1
26.6
81.5
>$60
0.1
43.0
0.0
0.0
4.0
0.4
5.0
0.2
10.3
0.0
0.6
55.0
0.5
2.1
26.6
81.5
EU-15 = European Union; OECD90+ = Organisation for Economic Co-operation and Development.
Table 9-10: Emissions Reductions in 2020 and Breakeven Costs ($/tC02eq) at 10% Discount Rate, 40% Tax
Rate (MtC02eq)—No-Action Baseline
2020
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.1
42.2
0.0
0.0
22.6
0.9
4.8
0.6
8.7
0.0
1.4
55.6
1.4
17.1
26.1
134.0
$15
0.1
52.5
0.0
0.0
24.4
1.0
6.0
0.6
10.9
0.0
1.5
69.6
1.5
18.4
32.7
160.4
$30
0.2
66.1
0.1
0.0
32.3
1.3
7.6
0.9
13.7
0.0
2.0
87.4
2.0
24.4
41.1
204.6
$45
0.2
69.9
0.1
0.0
35.3
1.4
8.0
0.9
14.4
0.0
2.2
92.4
2.2
26.6
43.4
218.2
$60
0.2
69.9
0.1
0.0
35.3
1.4
8.0
0.9
14.4
0.0
2.2
92.4
2.2
26.6
43.4
218.2
>$60
0.2
69.9
0.1
0.0
35.3
1.4
8.0
0.9
14.4
0.0
2.2
92.4
2.2
26.6
43.4
218.2
EU-15 = European Union; OECD90+ = Organisation for Economic Co-operation and Development.
IV-180 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV— INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
Table 9-11: Emissions Reductions in 2010 and Breakeven Costs ($/tC02eq) at 10% Discount Rate, 40% Tax
Rate (MtC02eq)—Technology-Adoption Baseline
2010
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.0
6.7
0.0
0.0
2.6
0.2
0.7
0.2
2.2
0.0
0.4
7.7
0.4
1.4
3.2
14.4
$15
0.0
6.8
0.0
0.0
2.8
0.2
0.7
0.2
2.2
0.0
0.4
7.8
0.4
1.5
3.2
14.8
$30
0.1
8.1
0.0
0.0
3.7
0.3
0.8
0.2
2.6
0.0
0.5
9.2
0.5
2.0
3.8
18.3
$45
0.1
8.2
0.0
0.0
4.0
0.3
0.8
0.2
2.6
0.0
0.6
9.3
0.5
2.1
3.8
19.0
$60
0.1
8.2
0.0
0.0
4.0
0.3
0.8
0.2
2.6
0.0
0.6
9.3
0.5
2.1
3.8
19.0
>S60
0.1
8.2
0.0
0.0
4.0
0.3
0.8
0.2
2.6
0.0
0.6
9.3
0.5
2.1
3.8
19.0
EU-15 = European Union; OECD90+ = Organisation for Economic Co-operation and Development.
Table 9-12: Emissions Reductions in 2020 and Breakeven Costs ($/tC02eq) at 10% Discount Rate, 40% Tax
Rate (MtC02eq)—Technology-Adoption Baseline
2020
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.1
3.7
0.0
0.0
4.2
0.4
0.4
0.2
1.1
0.0
0.6
3.8
0.6
2.2
1.2
12.5
$15
0.1
3.7
0.0
0.0
4.2
0.4
0.4
0.2
1.1
0.0
0.6
3.8
0.6
2.2
1.2
12.5
$30
0.1
4.1
0.0
0.0
5.0
0.4
0.4
0.3
1.2
0.0
0.7
4.2
0.7
2.7
1.3
14.5
$45
0.1
4.1
0.0
0.0
5.0
0.4
0.4
0.3
1.2
0.0
0.7
4.2
0.7
2.7
1.3
14.5
$60
0.1
4.1
0.0
0.0
5.0
0.4
0.4
0.3
1.2
0.0
0.7
4.2
0.7
2.7
1.3
14.5
>S60
0.1
4.1
0.0
0.0
5.0
0.4
0.4
0.3
1.2
0.0
0.7
4.2
0.7
2.7
1.3
14.5
EU-15 = European Union; OECD90+ = Organisation for Economic Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-181
-------
SECTION IV — INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
Table 9-13: Emissions Reduction and Costs in 2020—No-Action Baseline
Cost(2000S/tC02eq)
DR=10%,TR=40%
Reduction Option
Remote clean
C3F8 replacement
Capture/recovery
(membrane)
Plasma abatement (etch)
Thermal abatement
Catalytic abatement
Low
-$67.06
$0.00
$4.96
$16.83
$24.34
$33.17
High
-$67.06
$0.00
$4.96
$16.83
$24.34
$33.17
Emissions Reduction
Reduction from 2020
of Option Baseline
(MtC02eq) (%)
126.1
7.9
26.4
31.5
12.7
13.7
54.4%
3.4%
11.4%
13.6%
5.5%
5.9%
Running
Sum of
Reductions
(MtC02eq)
126.1
134.0
160.4
191.9
204.6
218.2
Cumulative
Reduction
from 2020
Baseline (%)
54.4%
57.8%
69.2%
82.8%
88.2%
94.1%
Table 9-14: Emissions Reduction and Costs in 2020—Technology-Adoption Baseline
Cost (2000$/tC02eq)
DR=10%,TR=40%
Reduction Option
Remote clean
C3F8 replacement
Capture/recovery
(membrane)
Plasma abatement (etch)
Thermal abatement
Catalytic abatement
Low
-$67.06
$0.00
$4.96
$16.83
$24.34
$33.17
High
-$67.06
$0.00
$4.96
$16.83
$24.34
$33.17
Emissions Reduction
Reduction from 2020
of Option Baseline
(MtC02eq) (%)
13.6
0.8
0.4
2.8
0.7
0.7
41.7%
2.5%
0.0%
6.3%
0.6%
0.0%
Running
Sum of
Reductions
(MtC02eq)
11.8
12.5
12.5
14.3
14.5
14.5
Cumulative
Reduction
from 2020
Baseline (%)
41.7%
44.2%
44.2%
50.5%
51.1%
51.1%
reductions can be achieved at breakeven costs less than or equal to $0/tCO2eq: 14.4 MtCO2eq in 2010 and
12.5 MtCO2eq in 2020, through implementation of the Remote Clean and C3F8 Replacement options. The
reductions available in 2020 are smaller than those available in 2010 because both the WSC countries and
especially the non-WSC countries are assumed to increase their implementation of reduction technologies
between 2010 and 2020. These reduction efforts outpace production growth, leading to a decline in
emissions in the technology-adoption baseline.
In the no-action MAC, under which no mitigation efforts are expected to have been implemented in
the baseline, available reductions are significantly higher, rising to 81.5 MtCO2eq in 2010 and 218.2
MtCO2eq in 2020. Of these reductions, over half (49.3 and 134.0 MtCO2eq, respectively) can be achieved at
$/tCO2eq values less than or equal to $0/tCO2eq. In the no-action scenario, available reductions rise in
proportion to increased semiconductor production.
The semiconductor manufacturing industry is treated by this analysis as a global market; the costs of
mitigation are therefore not expected to differ among manufacturing countries. Thus, each global MAC
curve has just six cost points.
IV-182
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV— INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
Regional Trends in the Technology-Adoption Scenario
Figures 9-4 and 9-5 present 2010 and 2020 regional MACs for China, Japan, the United States, the EU-
15, other OECD, and Rest of World for the technology-adoption scenario. China, followed by the United
States, accounts for the largest available reductions of any single country in 2010. Chinese emissions are
driven by China's significant share of global production and especially by the assumption that China, like
other countries outside the WSC, does not implement reduction technologies until after 2010 in the
technology-adoption scenario. Similarly, U.S. emissions are driven by the assumptions that the United
States accounts for 25 percent of global production in 2010 (the largest share of any country in that year)
and that the U.S. manufacturers that have not committed to the WSC goal (representing approximately 20
percent of U.S. production) will not meet it in 2010. The aggregate Rest of World region shows the largest
quantity of emissions reductions available in 2010, largely because this region includes Taiwan, South
Korea, Singapore, and Malaysia, which collectively account for approximately 30 percent of global
semiconductor production in 2010.
In 2020, the emissions reductions available in China grow slightly,-whereas those available in the Rest
of World region remain fairly constant, and those available in the other regions decrease. Although
baseline emissions from Japan, the United States, the EU-15, and other OECD countries change little
between 2010 and 2020 in the technology-adoption scenario, the reductions available to these countries
decrease significantly because more of these reductions are assumed to be implemented in the baseline.
This can be seen by comparing the 2010 and 2020 baseline market penetrations in Table 9-5 above.
Although Chinese baseline emissions decline from 2010 to 2020, potential reductions grow because
the market penetration of the abatement technologies increases as technologies become fully available in
countries outside the WSC. This trend is shown in Table 9-4. For the rest of the world region, which is
composed of both WSC members (Taiwan and South Korea) and non-WSC members (Malaysia and
Singapore), the decreases in reductions in the WSC counteract the increases in reductions elsewhere,
keeping available emissions reductions relatively constant from 2010 to 2020.
IV.9.3.3 Uncertainties and Limitations
The costs and savings presented in the above section are specific to individual technologies that
represent potential emissions mitigation options for the semiconductor manufacturing industry. The
assumptions that form the basis for these figures rely upon expert review of several options that were
believed to be favored by industry at the time of review. Discussions with industry scientists and analysts
contributed to capital and operating cost figures and were conducted in mid-2003.
Considering the rapid growth that characterizes the semiconductor manufacturing industry, it is
possible that both the relative and absolute costs of some options have changed over the last 3 years. For
the most part, the USEPA believes that recent changes will not change the relative ranking of the options
in the market. However, capture/recovery may be an exception to this. Capture/recovery, while
technologically feasible, also appears to present PFC cost and manufacturing risks that managers and
process engineers appear unwilling to take. Qualitatively, this suggests that more catalytic and thermal
decomposition abatement technology might be adopted than indicated in the tables and discussions in
this section. Further research might provide information for updating both the maximum market
penetrations and the baseline market penetrations that were assumed to apply in the technology-
adoption baseline (see Tables 9-2 through 9-8).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-183
-------
SECTION IV — INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
Figure 9-4: 2010 Regional Technology-Adoption MACs for Semiconductor Manufacturing
~ $40
d)
Japan
< Other OECD
Cumulative Emissions Reductions (MtCO2eq)
EU-15 = European Union; OECD90+ = Organisation for Economic Co-operation and Development.
Figure 9-5: 2020 Regional Technology-Adoption MACs for Semiconductor Manufacturing
8 $20-
0
to
o
I*"
w -$40
o
.<2 -$60 -
LU
-$80 J
0
0.5 1.
1.5 2.0 2.5 3.0 3.5
4.0 4.5 5.0 5.5 6.0
China
Rest of the world
—• United States
—~ EU-15
—> Japan
—« Other OECD
Cumulative Emissions Reductions (MtCO2eq)
EU-15 = European Union; OECD90+ = Organisation for Economic Co-operation and Development.
Capital and operating costs in the United States were assumed to apply to all semiconductor
manufacturing countries. This simple assumption is supported by the fact that the semiconductor
manufacturing industry represents a global market with relatively few international suppliers of
equipment and technology. Because fabrication facilities worldwide likely purchase equipment from the
same few suppliers, it is assumed that their costs remain the same. This approach is therefore justified as
a simplifying assumption, but it does not address any fab- or country-specific cost factors that may have
IV-184
GLOBAL MITIGATION OF NON C02 GREENHOUSE GASES
-------
SECTION IV— INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
effects on costs, such as energy prices and labor costs. Further research might provide justification for
country- or region-specific costs or scaling factors.
The MACs in this analysis (in terms of reduction percentages) were developed on a regional basis.
That is, within the WSC, countries were assumed to have identical percentages of their baselines available
for abatement in any given year and scenario. Similarly, countries outside of the WSC were assumed to
have identical .percentages of their baselines available for abatement, although these percentages were
different from those of the WSC countries. By distinguishing between the WSC and non-WSC countries,
this analysis accounts for much of the variation among countries in their emissions patterns and
reduction opportunities. However, even within the WSC, some countries are expected to make deeper
reductions than others to meet the WSC goal. This is because production has shifted from some countries,
such as Japan, to others, such as Taiwan, since the mid- to late 1990s. Thus, in the technology-adoption
baseline, this analysis may slightly underestimate the reduction opportunities that remain in Japan and
overestimate those that remain in Taiwan.
IV.9.4 References
Applied Materials. 1999. Catalytic Abatement of PFC Emissions. Presented at Semicon Southwest 99: A
Partnership for PFC Emissions Reductions, October 18,1999, Austin, TX.
Burton, S. 2003a. 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.
Burton, S. May 2003b. Personal communication with 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. Cambridge University Press.
International SEMATECH. October 18, 1999. Motorola Evaluation of the Applied Science and Technology, Inc.
(ASTex) ASTRON Technology for perfluorocompound (PFC) Emissions Reductions on the Applied Materials
DxL Chemical Vapor Deposition (CVD) Chamber. Presented at Semicon Southwest 99: A Partnership for
PFC Emissions Reductions, Austin, TX.
Mocella, M.T. 1998. PFC Recovery: Issues, Technologies, and Considerations for Post-Recovery Processing.
Dupont Fluoroproducts, Zyron Electron Gases Group. (Available on the Internet at
http://www.dupont.com/zyron/techinfo/monterey98.html).
Molina, Wooldridge, and Molina. 1995. Atmospheric Geophysical Research Letters. 22(13).
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). 2006. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2020. Washington, DC: USEPA.
World Fab Watch. 2002. Semiconductor Equipment and Material International, 2002.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-185
-------
SECTION IV — INDUSTRIAL PROCESSES • SEMICONDUCTOR MANUFACTURING
IV-186 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
IV.10 SF6 Emissions from Electric Power Systems
IV.10.1 Source Description
ulfur hexafluoride (SF6) is a colorless, odorless, nontoxic, and nonflammable gas with a GWP
that is 23,900 times that of CO2 during a 100-year time horizon and an atmospheric lifetime of
3,200 years (USEPA, 2005). SF6 is used as both an arc-quenching and insulating medium in
electrical transmission and distribution equipment. Several factors affect SF6 emissions from electrical
equipment, including the type and age of SF6-containing equipment and the handling and maintenance
protocols used by electric utilities. Historically, approximately 20 percent of total global SF6 sales have
gone to electric power systems, where the SF6 is believed to have been used primarily to replace emitted
SF6. Approximately 60 percent of global sales have gone to manufacturers of electrical equipment, where
the SF6 is believed to have been mostly banked in new equipment (Smythe, 2004).
SF6 emissions from electrical equipment used in transmission and distribution systems occur through
leakage and handling losses. Leakage losses can occur at gasket seals, flanges, and threaded fittings and
are generally larger in older equipment. Handling emissions occur when equipment is opened for
servicing, SF6 gas analysis, or disposal. Baseline emissions estimates under both a Technology-Adoption
and a no-action baseline scenario are presented in Table 10-1.
Table 10-1: Total SF6 Emissions from Electric Power Systems (MtC02eq)—No-Action Baseline
Country/Region
Africa
Annex!
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
0.5
20.5
0.3
0.5
1.8
0.4
1.9
0.7
0.5
0.3
1.5
19.5
1.1
0.6
15.0
26.8
2010
1.8
27.3
0.8
1.6
9.0
0.8
1.9
2.4
0.4
1.0
3.9
25.8
2.9
2.4
17.6
52.3
2020
2.5
28.6
0.8 -
2.4
14.9
0.8
1.9
3.5
0.4
1.6
3.9
28.0
2.9
3.2
18.9
65.8
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
GLOBAL MITIGATION OF NON-CO^ GREENHOUSE GASES
IV-187
-------
SECTION IV— INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
IV.10.1.1 Technology-Adoption Baseline
As shown in Table 10-2, global emissions from electric power systems are believed to have fallen
significantly between 1990 and 2000, based on SF6 sales to utilities and estimated equipment retirements.
This decline was due to a significant increase in the cost of SF6 gas in the mid-1990s, which motivated
electric utilities to implement better management practices to reduce their use of SF6. However, sales of
SF6 increased by more than 37 percent between 2000 and 2003, reversing the trend (Smythe, 2004). In
addition, equipment retirements (based on a 40-year equipment lifetime) are estimated to have more than
doubled between 2000 and 2003. Together, these two trends resulted in an estimated 55 percent increase
in global emissions between 2000 and 2003, creating emissions levels similar to those observed in 1990.
Table 10-2: Total SF6 Emissions from Electric Power Systems (MtC02eq)—Technology-Adoption Baseline
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
0.5
20.5
0.3
0.5
1.8
0.4
1.9
0.7
0.5
0.3
1.5
19.5
1.1
0.6
15.0
26.8
2010
1.8
21.8
0.8
1.6
9.0
0.7
1.4
2.4
0.3
1.0
3.9
20.3
2.9
2.4
12.8
46.8
2020
2.5
20.3
0.8
2.4
14.9
0.7
0.9
3.5
0.3
1.6
3.9
19.8
2.9
3.2
11.8
57.5
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
These global trends are reflected in the trends of the individual regions except for the United States;
EU-25;1 Norway, Switzerland, and Iceland (EU-25+3); and Japan. For the United States, emissions
estimates for 1990 through 2003 are taken from the Inventory of U.S. Greenhouse Gas Emissions and Sinks:
1990-2003 (USEPA, 2005). For EU-25+3, emissions estimates for 1990 through 2020 were obtained from
Reductions of SF6 Emissions from High and Medium Voltage Electrical Equipment in Europe (Ecofys, 2005). For
Japan, emissions estimates for 1990 through 2010 were obtained from Recent Practice for Huge Reduction of
SF6 Gas Emissions from GIS&GCB in Japan (Yokota et al., 2005), as well as personal communications with T.
Yokota (2006). These studies show declining emissions in these regions through 2003.
As illustrated in Figure 10-1, beyond 2005, emissions in developed countries are expected either to
remain steady or to decline. Emissions in non-EU Eastern Europe and non-EU FSU are expected to
remain relatively constant through 2020. Because the electric grids in these countries are mature and well
developed, it is assumed that there will be no additional growth of emissions from their electric
transmission and distribution systems. Any system growth is expected to be offset by decreases in the
equipment's average SF6 capacity and emissions rate as new, small, leak-tight equipment gradually
IV-188 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
Figure 10-1: SF6 Emissions from Electric Power Systems on a Technology-Adoption Scenario—1990-
2020 (MtC02eq)
§"
cv
o
o
*rf
5
v>
in
v>
E
HI
70
60 -
50
40
30
20
10
• Middle East
D Africa
• Non-EUFSU
• Latin America
DNon-EU Eastern Europe
• S&EAsia
• China/CPA
• Japan
DEU-25
• United States
1990
2000
2010
2020
Year
CPA = Centrally Planned Asia; EU-25 = European Union; Non-EU FSU = non-European Union Former Soviet Union countries; S&E Asia =
Southeast Asia; OECD90+ = The Organisation for Economic Co-operation and Development.
replace old, large, leaky equipment. In the United States, EU-25+3, and Japan, emissions are expected to
continue to decline as utilities, through government-sponsored voluntary and mandatory programs,
implement reduction measures such as leak detection and repair and gas recycling practices.
In contrast, emissions from developing regions (i.e., Latin America, South and East Asia, Middle East,
Africa and China/CPA) are expected to continue growing during the next 15 years. In these regions, it is
assumed that SF6-containing equipment has been installed relatively recently and that all equipment is
new. Consequently, as infrastructure expands to meet the demands of growing populations and
economies, emissions are estimated to grow at a rate proportional to country- or region-specific net
electricity consumption (USEIA, 2002). This growth drives global emissions growth, resulting in
worldwide emissions of 57 MtCO2eq in 2020. By 2020, Latin America, South and East Asia, the Middle
East, Africa, and China/CPA are expected to account for 63 percent of total emissions, versus
approximately 10 percent in 1990. OECD is projected to account for only 29 percent of global emissions in
2020, versus approximately 82 percent in 1990.
IV.10.1.2 No-Action Baseline
As illustrated in Figure 10-2, baseline emissions for the period 1990 through 2000 follow the same
trajectory as those under the technology-adoption scenario, with both baselines diverging after 2003.
Assumptions and emissions estimates for developing regions (i.e., Latin America, South, and East Asia,
Middle East, Africa and China/CPA) are the same as discussed under the technology-adoption baseline.
For the United States, Japan, EU-25, and EU-25+3, it is assumed that no additional voluntary measures are
adopted after 2003. For the United States, EU-25+3, and Japan, emissions are expected to increase from
2003 levels, with system growth being the driver in the EU and Japan. The marked increase in U.S.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-189
-------
SECTION IV— INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
Figure 10-2: SF6 Emissions from Electric Power Systems on a No-Action Scenario-
(MtC02eq)
-1990-2020
• Middle East
D Africa
• Non-EU FSU
• Latin America
D Non-EU Eastern Europe
• S&EAsia
• China/CPA
BOECD90+
•Japan
DEU-25
• United States
1990
2000
2010
2020
Year
CPA = Centrally Planned Asia; EU-25 = European Union; Non-EU FSU = non-European Union Former Soviet Union countries; S&E Asia =
Southeast Asia; OECD90+ = The Organisation for Economic Co-operation and Development.
emissions after 2000 is an artifact of the method used to estimate U.S. emissions in the no-action scenario.
Under this approach, the U.S. emissions factor was assumed to revert to its relatively high 1999 level in
2005, despite reductions in earlier years.
The assumption that the United States, EU-25+3, and Japan will pursue no additional voluntary
measures after 2003 increases their contribution to world emissions in 2020. Unlike the technology-
adoption baseline, where OECD accounts for only 29 percent of emissions in 2020, in the no-action
baseline, OECD accounts for 38 percent. In contrast, the contribution of developing regions, such as Latin
America, South and East Asia, the Middle East, Africa, and China/CPA decreases to 55 percent of total
2020 emissions in the no-action scenario, versus 63 percent under the technology-adoption scenario.
IV.10.2 Cost of SF6 Emissions Reduction from Electric Power
Systems
IV.10.2.1 Abatement Options
SF6 emissions during use of electrical equipment can occur either during the maintenance and
disposal of equipment or during the operation of equipment because of the failure of mechanical seals or
breaks in gas-insulated equipment enclosures.
For all countries except the EU-25, EU-25+3, and Japan, this analysis models three potential
abatement options for reducing SF6 emissions from electric power systems:
• SF6 recycling,
• leak detection and repair (LDAR), and
IV-190
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV— INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
• equipment refurbishment.
For the EU-25+3 and Japan, this analysis models the abatement options identified in Ecofys (2005),
because the baseline emissions for Europe (in both the no-action and technology-adoption scenarios) are
based on data presented in Ecofys (2005). Using the Ecofys options therefore maintains consistency with
the assumptions used in that report to estimate current and future emissions (i.e., current and future
levels of implementation of reduction options). These options are applied to Japan, because Japan is
believed to have implemented reduction options to approximately the same extent as Europe. The
following options are identified in Ecofys:
• awareness, including training, monitoring, and labeling;
• evacuation of equipment;
• repair or replacement; and
• decommissioning infrastructure.
Although Ecofys (2005) identifies four options, they are, for the most part, similar in nature to those
analyzed for countries other than the EU-25+3 and Japan in this study. For example, evacuation of
equipment is similar to the SF6 recycling option in that both address the recovery of SF6 from closed
pressure equipment. Decommissioning infrastructure also includes recovery of gas, in this case from
retiring equipment. Repair and replacement includes activities similar to those included in both the
LDAR and refurbishment options. As for awareness, some of the associated training costs and emissions
reductions are accounted for within the SF6 recycling option. The remainder of Section IV.10.3 provides
an overview of each abatement option and details the associated emissions and cost assumptions.
Abatement Options—For All Countries Except EU-25+3 and Japan
Emissions Available for Abatement
For most of the other sectors in this analysis, the quantity of emissions that can be abated through the
applicable abatement options is estimated directly based on the activity level and the fraction of the
emitting activity that remains uncontrolled in the baseline. For this sector, however, the analysis begins
with an emissions estimate rather than an activity level, making the estimate of uncontrolled emissions
somewhat more complicated. To develop the estimate of currently uncontrolled emissions, the technical
applicability, current market penetration, and reduction efficiency of the three abatement options are
estimated and applied to the hypothetical emissions that would result if emissions were not controlled at
all. Using this approach, it is possible to estimate the fraction of current emissions that consist of residual
emissions from the options as they are implemented in the baseline, as well as the fractions of current
emissions that can still be abated by the three options. For equipment whose emissions are not controlled,
33 percent of emissions are estimated to occur during operation as a result of leaks, and 67 percent are
estimated to occur during maintenance and disposal as a result of failure to recycle. This apportionment
is inferred from O'Connell et al. (2002), who report that leakage losses account for between 0.5 percent
and 1 percent per year, and handling losses account for between 1 percent and 2 percent per year. Based
on discussions with electric utilities and manufacturers of SF6 recycling equipment, this analysis assumes
that recycling, LDAR, and refurbishment options are currently applied to 80 percent of electrical
equipment. In addition, the analysis assumes that those utilities that currently recycle SF6 recover 80
percent of the gas each time. That is, 80 percent of the gas enclosed in electrical equipment is assumed to
be removed as the enclosure pressure drops from operational conditions to zero pounds per square inch
(psig). The abatement option described below assumes that, in the presence of a carbon price, an
additional 15 percent of the SF6 will be recovered—95 percent overall—which requires pulling a vacuum
on the equipment. This is within the technical capability of the equipment but is relatively time
consuming. Together, all of these assumptions lead to the conclusion that 49 percent of the baseline
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-191
-------
SECTION IV — INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
emissions remain available to abatement through recycling, 7 percent remains available to abatement
through LDAR, and 1 percent remains available to abatement through refurbishment.
SFg Recycling
For equipment whose emissions are not controlled, 67 percent of emissions are estimated to occur
during equipment servicing and disposal. This estimate is based on information reported by the
International Council on Large Electric Systems (CIGRE) (O'Connell et al., 2002), which indicates that
leakage and handling losses are on the order of between 0.5 percent and 1 percent per year and 1 percent
to 2 percent per year, respectively. Recycling gas cart systems typically withdraw, purify, and return the
SF6 gas to the gas-insulated equipment. Recycling equipment vendors state that utilities that use the
equipment typically recover about 80 percent of the gas held in high-voltage equipment, although
recycling equipment is theoretically capable of capturing almost 100 percent. Typically, utilities end
recovery early because the current price of the SF6 does not justify spending the additional time required
to recover it fully. In other words, it would take as much time to recover the final 20 percent of the gas as
it takes to recover the first 80 percent (by mass), because the density of the gas declines during the
recovery process. Consequently, it is assumed that 80 percent recovery is the current standard industry
practice.
The use of recycling equipment is considered a relatively straightforward option for conservative gas-
handling practices, and gas cart ownership and use have increased significantly worldwide (O'Connell et
al., 2002; Ellerton, 1997). Communications with gas cart manufacturers have also indicated that the
majority of electric utilities in North and South America use recycling equipment (ICF, 2001). This
analysis assumes that the current and future market penetration of recycling equipment in the baseline is
80 percent in both developed and developing countries.
In the presence of a carbon price, this analysis assumes that utilities that currently recover SF6 will
recover it more deeply, recovering 95 percent of the gas rather than the current 80 percent, and the
analysis assumes that the 20 percent of utilities that do not currently recover SF6 will begin recovering it
to the 95 percent level. Based on these assumptions, approximately 39 percent of the emissions reductions
for recycling are achieved through deeper recovery (going from 80 percent to 95 percent), while 61
percent of these reductions are achieved by increasing the market share of recycling.
Cost and Emissions Reduction Analysis
• Capital/Upfront Costs. The capital cost for smaller-capacity recycling equipment is between
$5,000 and $50,000 per unit and for higher-capacity units between $50,000 and $130,000 per unit
(ICF, 2001). Because older, larger electrical equipment is being replaced with newer, smaller
volume electrical equipment in developed countries, and because developing countries, which
have newer electrical T&D networks, use smaller electrical equipment, an average cost of $25,500
per unit was assumed. Total country-specific capital costs were calculated by estimating the
number of recycling gas cart units required to ensure 100 percent market penetration. In the
United States, it is estimated that between 750 and 1,000 recycling gas carts are in use, based on
communications with gas cart manufacturers (ICF, 2001). Based on the assumption that this
number represents an 80 percent market penetration, the number of recycling units required to
establish a 100 percent U.S. market penetration was estimated to be between 190 and 250 units.
For other countries, the number of recycling units that can still be implemented, based on an
existing 80 percent market penetration scenario, was estimated as the product of the number of
recycling units required to achieve 100 percent market penetration in the United States and the
ratio of country-specific net electricity consumption to U.S. net electricity consumption (USEIA,
2002). '
IV-192 GLOBAL MITIGATION OF NOM-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
• Annual Costs. Equipment recycling rates range from 50 to 500 pounds SF6 per hour (Dilo, 2003;
Cryoquip, 2003). Actual recovery speeds depend on recovery pump equipment, operating
pressure, and connecting equipment. As a conservative estimate, and considering that most SF6
users will require smaller-grade equipment because of smaller charge sizes, the average recovery
rate for removing gas held under positive pressure (i.e., 80 percent by mass) is assumed to be 100
pounds SF6 per hour. However, once a vacuum is drawn, the average recovery rate falls below
this level. Based on the assumption that 100 pounds per hour represents the average recovery
rate for recovering the first 80 percent of gas (achieving zero psig), average recovery rates were
calculated for recovery of the gas from its initial pressure to a vacuum (95 percent recovery),
which is applicable to the utilities that are not currently conducting recycling (i.e., 20 percent of
the market) and recovery of the gas from zero psig to a vacuum (15 percent), which is applicable
to the 80 percent of the market that currently only achieves 80 percent recovery. These average
recovery rates were 64 and 22 pounds per hour, respectively. The marginal labor time required
for recycling the gas is equal to the total gas recycled in the country multiplied by the estimated
emissions that can be reduced by increasing the market penetration of recycling (61 percent) and
increasing the depth of recovery from 80 percent to 95 percent (39 percent), and dividing by the
corresponding average recycling rate (64 and 22 pounds per hour, respectively). Associated labor
costs were estimated for a two-person crew and assumed an hourly labor rate of $50 per hour. To
account for additional labor time spent for training and setting up/tearing down recycling
equipment, conservative multipliers of 1.02 (i.e., assuming 2 percent [1 week] of annual labor
time spent conducting training) and 1.5 were also applied.
• Cost Savings. It is assumed that all SF6 recycled is a cost savings, because the facility's SF6
purchase and consumption rate will decrease. For this analysis, it is assumed that the cost of SF6
is $7 per pound.
Leak Detection and Repair
LDAR abatement options aim to identify and reduce the SF6 leakage that occurs from gas-insulated
equipment. For equipment whose emissions are not controlled, 33 percent of emissions are estimated to
occur during equipment operation, with 30 percent controllable through LDAR. SF6 leak detection is
accomplished through various techniques, including "sniffing" for gas with SF6 gas sensors and using
laser-based remote sensing technology (McRae, 2000). Similar to SF6 recycling, the current market
penetration of this option is assumed to be 80 percent of SF6 use. LDAR measures are assumed to have a
reduction efficiency of 50 percent.
Cost and Emissions Reduction Analysis
• Capital/Upfront Costs. Leak detection equipment costs vary depending on the type of
instrument used. For simple screening devices, costs are believed to be minimal (i.e., less than
$2,000 per unit). For expensive items, such as the laser-based imaging system, it is assumed that
facilities will lease equipment or contract private companies to provide leak detection services.
Hence, LDAR mitigation options will have no capital costs.
• Annual Costs. It is assumed that the average leak size is 25 pounds SF6 per leak per year and that
the average time to detect and repair a leak is 2 and 8 hours, respectively (ICF, 2000). The
marginal labor time required for LDAR is equal to the total gas emitted from this source divided
by the average leak size and multiplied by the average time to detect and service a leak.
Associated labor costs assume LDAR requires a one-person and two-person crew, respectively,
and that the hourly labor rate is $50 per hour.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-193
-------
SECTION IV— INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
• Cost Savings. It is assumed that all SF6 saved during leak detection and maintenance activities
represents a cost savings, because the facility SF6 purchase and consumption rate will decrease.
For this analysis, it is assumed that the cost of SFe, is $7 per pound.
Equipment Refurbishment
Unlike LDAR-based repairs, which tend to focus on small leaks on specific components, such as a
bushing flange gasket, refurbishment addresses the need, when leakage losses are large, for a
comprehensive repair. Refurbishment is a process in which equipment is disassembled and rebuilt (and
possibly upgraded) using remachined, cleaned, and/or new components. Generally, equipment
refurbishment represents the cheaper of two possible options: 1) equipment replacement, which for a
large breaker (362 kV) can be on the order of $300,000 to $400,000; and 2) refurbishment, which may cost
around $100,000 (McCracken et al, 2000).
It is assumed that 33 percent of uncontrolled emissions occur during equipment operation; of this
total, 3 percent is assumed to be controllable through refurbishment. Similar to the other options,
refurbishment is assumed to have a current market penetration of 80 percent. Refurbishment measures
are assumed to have a reduction efficiency of 95 percent.
Cost and Emissions Reduction Analysis
• Capital/Upfront Costs. The cost of labor and parts to undertake the refurbishment of a circuit
breaker with a nameplate capacity of 1,500 pounds and a leak rate of 20 percent is assumed to be
approximately $100,000 (McCracken et al., 2000). This includes equipment disassembly, rebuild,
testing, and installation.
• Annual Costs. During refurbishment, equipment is completely remanufactured. As a result, this
option is considered a one-time activity, with no annual costs for the lifetime of the project period
(i.e., 15 years).
• Cost Savings. It is assumed that all SF6 saved during refurbishment activities represents a cost
savings, because the facility SF6 purchase and consumption rate will decrease. For this analysis, it
is assumed that the cost of SF6 is $7 per pound.
Abatement Options—EU-25+3 and Japan Only
For each of the following options, cost and emissions reduction potential assumptions are detailed in
Ecofys (2005). Although these options were developed for EU-25+3 countries, they have also been applied
to Japan, because Japanese equipment designs and maintenance practices are believed to be similar to
those in the EU-25+3 (Ecofys, 2005; Yokota et al., 2005).
• Awareness, Including Training, Monitoring, and Labeling. Awareness includes costs to
implement training programs for SF6 gas handling during equipment top-up and maintenance.
Awareness also includes costs to implement SF6 gas management systems, where SF6 inventories,
purchases, and gas use are monitored.
• Evacuation of Equipment. Evacuation includes costs associated with attaining a higher level of
SF6 recovery from closed-pressure equipment (i.e , drawing evacuation pressure from 50 millibar
[mbar] down to 20 mbar).
• Repair or Replacement. Ecofys (2005) assumes that 3 percent of closed-pressure systems leak
more than 2.5 percent per year greater than their design leak rates. Costs are based on 90 percent
of equipment being repaired, with the remainder replaced.
• Decommissioning Infrastructure. This includes costs to develop infrastructure to handle end-of-
life treatment (both gas and equipment material) of SF6 electrical equipment.
IV-194 GLOBAL MITIGATION OF NOM-COo GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
IV.10.3 Results
This section discusses the results from the MAC analysis for the world and several regions, for both
the no-action and technology-adoption scenarios.
IV.10.3.1 Data Tables and Graphs
Based on the trends described above, the USEPA developed MACs for several regions. Tables 10-3
through 10-8 provide a summary of the potential emissions reduction opportunities and associated costs
for these regions in 2010 and 2020 for the no-action and technology-adoption baselines. The costs to
reduce 1 tCO2eq are presented for a discount rate of 10 percent and a tax rate of 40 percent.
Table 10-3: Emissions Reductions in 2010 and Breakeven Costs ($/tC02eq) for Electric Power Systems at a
10% Discount Rate, 40% Tax Rate (MtC02eq)—No-Action Baseline
2010
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
1.00
12.23
0.41
0.79
4.99
0.29
0.00
1.34
0.00
0.49
2.15
11.25
1.60
1.22
8.57
25.53
$15
1.03
14.99
0.48
0.93
5.13
0.41
0.66
1.38
0.15
0.57
2.22
14.14
1.64
1.37
10.05
29.24
$30
1.03
15.00
0.48
0.93
5.13
0.41
0.67
1.38
0.15
0.57
2.22
14.15
1.64
1.37
10.05
29.26
$45
1.03
15.00
0.48
0.93
5.13
0.41
0.67
1.38
0.15
0.57
2.22
14.15
1.64
1.37
10.05
29.26
$60
1.03
15.04
0.48
0.93
5.13
0.41
0.70
1.38
0.16
0.57
2.22
14.19
1.64
1.37
10.05
29.30
>S60
1.03
15.04
0.48
0.93
5.13
0.41
0.70
1.38
0.16
0.57
2.22
14.19
1.64
1.37
10.05
29.30
EU-15 = European Union; OECD = The Organisation for Economic Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-195
-------
SECTION IV — INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
Table 10-4: Emissions Reductions in 2020 and Breakeven Costs ($/tC02eq) for Electric Power Systems at
10% Discount Rate, 40% Tax Rate (MtC02eq)—No-Action Baseline
2020
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
1.40
12.85
0.41
1.18
8.29
0.29
0.00
1.92
0.00
0.76
2.15
12.33
1.60
1.63
9.18
32.69
$15
1.45
16.11
0.48
1.38
8.52
0.45
0.95
1.97
0.21
0.89
2.22
15.80
1.64
1.82
10.78
37.32
$30
1.45
16.12
0.48
1.38
8.52
0.45
0.95
1.97
0.21
0.89
2.22
15.80
1.64
1.82
10.78
37.33
$45
1.45
16.12
0.48
1.38
8.52
0.45
0.95
1.97
0.21
0.89
2.22
15.80
1.64
1.82
10.78
37.33
$60
1.45
16.16
0.48
1.38
8.52
0.46
0.98
1.97
0.22
0.89
2.22
15.84
1.64
1.82
10.78
37.36
>$60
1.45
16.16
0.48
1.38
8.52
0.46
0.98
1.97
0.22
0.89
2.22
15.84
1.64
1.82
10.78
37.36
EU-15 = European Union; OECD = The Organisation for Economic Co-operation and Development.
Table 10-5: Emissions Reductions in 2010 and Breakeven Costs ($/tC02eq) for Electric Power Systems at
10% Discount Rate, 40% Tax Rate (MtC02eq)—-Technology-Adoption Baseline
2010
Country/Region
Africa -
Annex I
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
1.00
8.15
0.41
0.79
4.99
0.29
0.00
1.34
0.00
0.49
2.15
7.17
1.60
1.22
4.48
21.44
$15
1.03
9.24
0.48
0.93
5.13
0.30
0.00
1.38
0.00
0.57
2.20
8.41
1.64
1.37
5.25
23.50
$30
1.03
9.24
0.48
0.93
5.13
0.30
0.00
1.38
0.00
0.57
2.20
8.41
1.64
1.37
5.25
23.50
$45
1.03
9.24
0.48
0.93
5.13
0.30
0.00
1.38
0.00
0.57
2.20
8.41
1.64
1.37
5.25
23.50
$60
1.03
9.24
0.48
0.93
5.13
0.30
0.00
1.38
0.00
0.57
2.20
8.41
1.64
1.37
5.25
23.50
>$60
1.03
9.24
0.48
0.93
5.13
0.30
0.00
1.38
0.00
0.57
2.20
8.41
1.64
1.37
5.25
23.50
EU-15 = European Union; OECD = The Organisation for Economic Co-operation and Development.
IV-196 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
Table 10-6: Emissions Reductions in 2020 and Breakeven Costs ($/tC02eq) for Electric Power Systems at
10% Discount Rate, 40% Tax Rate (MtC02eq)—Technology-Adoption Baseline
2020
Country/Region
Africa
Annex I
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
1.40
3.67
0.41
1.18
8.29
0.29
0.00
1.92
0.00
0.76
2.15
3.15
1.60
1.63
0.00
23.51
$15
1.45
7.68
0.48
1.38
8.52
0.30
0.00
1.97
0.00
0.89
2.20
7.38
1.64
1.82
3.69
28.88
$30
1.45
7.68
0.48
1.38
8.52
0.30
0.00
1.97
0.00
0.89
2.20
7.38
1.64
1.82
3.69
28.88
$45
1.45
7.68
0.48
1.38
8.52
0.30
0.00
1.97
0.00
0.89
2.20
7.38
1.64
1.82
3.69
28.88
$60
1.45
7.68
0.48
1.38
8.52
0.30
0.00
1.97
0.00
0.89
2.20
7.38
1.64
1.82
3.69
28.88
>$60
1.45
7.68
0.48
1.38
8.52
0.30
0.00
1.97
0.00
0.89
2.20
7.38
1.64
1.82
3.69
28.88
EU-15 = European Union; OECD = The Organisation for Economic Co-operation and Development.
Table 10-7: Emissions Reduction and Costs in 2020—No-Action Baseline
Cost(2000$/tC02eq)
DR=10%, TR=40%
Reduction Option
Recycling
Decommissioning
Awareness/training
Leak detection
Refurbishment
Evacuation
Repair and replacement
Low
-$0.61
$1.47
$2.04
-$0.56
$5.01
$27.28
$45.51
High
-$0.09
$1.47
$2.04
$2.68
$5.01
$27.28
$45.51
Emissions
Reduction of
Option
(MtC02eq)
30.65
1.04
0.32
4.38
0.93
0.01
0.04
Reduction
from 2020
Baseline
(%)
46.6%
1.6%
0.5%
6.7%
1.4%
0.0%
0.1%
Running
Sum of
Reductions
(MtC02eq)
30.65
31.69
32.01
36.39
37.32
37.33
37.36
Cumulative
Reduction
from 2020
Baseline
(%)
46.6%
48.2%
48.7%
55.3%
56.7%
56.8%
56.8%
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-197
-------
SECTION IV— INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
Table 10-8: Emissions Reduction and Costs in 2020—Technology-Adoption Baseline
Cost(2000S/tC02eq)
DR=10%,TR=40%
Reduction Option
Recycling
Decommissioning
Awareness/training
Leak detection
Refurbishment
Evacuation
Repair and replacement
Low
-$0.61
$1.47
$2.04
-$0.56
$5.01
$27.28
$45.51
High
$0.10
$1.47
$2.04
$2.68
$5.01
$27.28
$45.51
Emissions
Reduction of
Option
(MtC02eq)
24.61
0.00
0.00
3.17
1.10
0.00
0.00
Reduction
from 2020
Baseline
(%)
42.8%
0.0%
0.0%
5.5%
1.9%
0.0%
0.0%
Running
Sum of
Reductions
(MtC02eq)
24.61
24.61
24.61
27.78
28.88
28.88
28.88
Cumulative
Reduction
from 2020
Baseline
(%)
42.8%
42.8%
42.8%
48.3%
50.2%
50.2%
50.2%
IV.10.3.2 Global and Regional MACs and Analysis
This section discusses the results from the MAC analysis of the world and selected countries and
regions, including China, Japan, the United States, the EU-15, other OECD, and the rest of the world.
Figure 10-3 presents the 2010 and 2020 global technology-adoption and no-action MACs for electric
power systems. For the no-action MACs, significant reductions (comprising more than 48 percent of the
baseline emissions from this sector) are achievable below $0/tCO2eq in 2010 and 2020. These reductions
occur through the implementation of SF6 recycling and LDAR options in most countries, except EU-25+3
and Japan. For the latter countries, decommissioning and awareness/training reduce approximately 2
percent of global baseline emissions, with both options being implemented below $2/tCO2eq.
Figure 10-3: 2010 and 2020 Global Technology-Adoption and No-Action MACs for Electric Power
Systems
$5 -i
I $4~
O
33 $3 -
« $2 -
o
1 $1 ^
•o
-------
SECTION IV— INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
In the technology-adoption MACs, no reductions are available in EU-25+3 and Japan in 2010 and
2020, because mitigation technologies are assumed to be fully implemented in the baselines of these
countries. For the remaining countries, more than 40 percent of baseline emissions can be reduced for less
than $0/tCO2eq through SF6 recycling and LDAR measures. An additional 10 percent of baseline
emissions can be reduced through $5/tCO2eq. Most of these reductions result from SF6 recycling and
LDAR in developed regions, such as the United States, and the implementation of refurbishment in all
countries. The rightward shift in technology-adoption and no-action MACs between 2010 and 2020
reflects increasing emissions from electric grid infrastructure growth, specifically in developing country
regions, such as China, Latin America, and Africa. In developed countries, such as the United States, EU-
25+3, and Japan, voluntary and mandatory emissions reduction programs reduce both baseline emissions
and the reductions available in the technology-adoption MACs. (As noted above, emissions reductions in
the technology-adoption baseline actually exhaust the available reductions in Europe and Japan, leaving
no reductions in the technology-adoption MACs for these regions.) In the no-action MACs, however, it is
assumed that additional voluntary measures will not be implemented after 2003; consequently, U.S.
emissions increase while emissions from the EU-25+3 and Japan remain constant between 2010 and 2020,
reflecting the stabilization of European and Japanese SF6 banks.
Figures 10-4 and 10-5 present 2010 and 2020 technology-adoption MACs for China, Japan, the United
States, the EU-15, other OECD, and the rest of the world. These figures show the regional contributions to
the global trends described above. As noted above, EU-25+3 and Japan offer no emissions reduction
opportunities in 2010 or 2020, because it is assumed that all mitigation measures, such as
awareness/training, decommissioning, evacuation, and repair and replacement, are implemented in their
baselines. For all other countries, SF6 recycling offers the opportunity for significant emissions reductions
at low carbon cost (i.e., less than or equal to $0.10/tCO2eq). LDAR offers reductions at somewhat higher
costs ranging from about -$0.56/tCC>2eq to $2.68/tCO2eq. LDAR costs have a large labor component
compared with the recycling option. Consequently, abatement costs are higher in countries where labor
costs are high, such as the United States, Australia, and Canada, but are very low in other regions, such as
China, Latin America, and Africa. The pronounced "elbows" in the curves for most regions are indicative
of the lower emissions reduction potential and higher abatement costs offered by LDAR and
refurbishment, compared with recycling.
In 2020, the MACs for all countries and regions have a similar profile to 2010, where SF6 recycling is
followed by LDAR and refurbishment. However, for some countries and regions, there are minor shifts in
the curves, specifically those for developing economies, such as China, other OECD, and the rest of the
world. This shift reflects the potential for increased emissions reductions as electric transmission and
distribution grids expand and associated SF6 emissions increase to accommodate growing commercial
and residential energy needs. In comparison, for the United States, the 2020 MAC shifts left as emissions
available for abatement decrease because of the continuing success of domestic voluntary programs.
IV.10.3.3 Uncertainties and Limitations
In developing these estimates of emissions, reductions, and costs, the USEPA made use of multiple
international data sets and IPCC guidance on estimating emissions from this source. Nevertheless, this
analysis is subject to a number of uncertainties that affect both global and country-specific estimates of
emissions, reductions, and costs, particularly estimates for regions other than the United States, Japan,
and the EU-25+3.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-199
-------
SECTION IV — INDUSTRIAL PROCESSES « ELECTRIC POWER SYSTEMS
Figure 10-4: 2010 Regional Technology-Adoption MACs for Electric Power Systems
$5
$4-
$3 -
(0
c
o
9= $2 -
•o o"
53 $1
ga
2~ $0-
w
(/)
E
UJ
-$1 i)
-$2 -
r
China
Rest of the world
•• United States
* EU-15
*• Japan
« Other OECD
10
12
Cumulative Emissions Reductions (MtC02eq)
EU-15 = European Union; OECD = The Organisation for Economic Co-operation and Development.
Figure 1 0-5: 2020 Regional Technology-Adoption MACs for Electric Power Systems
$5^
$4 -
(0
.1 $3 -
4-1
| 9= $2-
Q 0)
*0 $1-
WO
0 <£ SO-
OT
1 -$1 <
m -$2J
-$3 -
L
_T
/
T
/
— * China
Rest of the world
— • United States
— * EU-15
— > Japan
— « Other OECD
..__—>
) I 4 b B
Cumulative Emissions Reduc
JT
— — i — i 1
10 12 14
tions (MtCO2eq)
EU-15 = European Union; OECD = The Organisation for Economic Co-operation and Development.
SF6 Emissions from Electric Power Systems
Although emissions from the United States, EU-25+3, and Japan are based on bottom-up evaluations
of emissions rates and SF6 banks in equipment, remaining country-specific estimates are based on
apportioning RAND survey data (Smythe, 2004) using net electricity consumption statistics. The
relationship between emissions and electricity consumption varies between regions and over time,
particularly as countries begin to adopt emission-reducing practices and technologies. Additional
uncertainties associated with this approach are described in detail in the USEPA (2006) report Global
Anthropogenic Non-CO2 Greenhouse Gas Emissions: 1990-2020. These uncertainties affect both the global
total and the country-by-country apportionment of that total.
IV-200
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
Emissions Apportionment/Market Penetration/Reduction Efficiency for All Regions, Except
EU-25+3 and Japan
• This analysis is based on the assumption that 67 percent of emissions are attributable to the
failure to recycle during maintenance and disposal, while 33 percent of emissions are due to
equipment leakage. The basis for this assumption is a study performed by CIGRE and reported
by O'Connell et al. (2002). There is very limited information on the apportionment of emissions
between handling and leakage losses; consequently, this is a potential source of uncertainty.
• This analysis assumes emissions are due to leakage and failure to recycle only; however, losses
due to improper handling of SF6 gas may also contribute to baseline emissions. Because of limited
information, this potential emissions source is not fully addressed in this analysis.
• Estimates of current market penetration are based on communications with U.S.-based industry
experts. These estimates are assumed to apply to all global regions; however, it is possible that
the current 80 percent penetration estimate may be too high for some regions, especially
developing countries. For example, it is possible that developing countries may not have the
resources to implement recycling and LDAR to the levels assumed in this analysis. If market
penetration is lower than assumed, the potential for emissions reduction from these countries
will be higher.
• For LDAR, reduction efficiency is assumed to be 50 percent; this is an average number that
accounts for varying degrees of LDAR success. However, it is technically possible, with current
practices (e.g., laser leak detection and new sealant technology) to achieve reductions that are
closer to 100 percent.
• For refurbishment, reduction efficiency is assumed to be 95 percent. As with LDAR, it is
technically feasible to achieve reductions of 100 percent; however, the current estimate assumes
varying degrees of refurbishment success.
Emissions Apportionment/Market Penetration/Reduction Efficiency for EU-25+3 and Japan
Only
This analysis uses cost and emissions reduction potentials developed for EU-25+3 countries (Ecofys,
2005). The same costs and emissions reduction potentials have been applied to Japan on the assumption
that Japan and Europe have similar electric transmission and distribution infrastructures and
maintenance practices. Because there are likely to be some differences in the general age and type of
equipment used, as well as in the current level of implementation of abatement options, this assumption
is a potential source for uncertainty in the MACs. Additional uncertainties associated with the specific
options used for EU-25+3 and Japan are detailed in Ecofys (2005).
Estimation of Annual Cost
• For recycling and LDAR, marginal labor cost estimates depend on average recycling rates (64 and
22 pounds per hour) and the average size of a leak (25 pounds per year per leak), respectively.
Both of these estimates may vary significantly and are, consequently, a source of uncertainty in
the MACs. In developing recycling costs, it is assumed that 61 percent of emissions are abatable
through increased market penetration, while 31 percent are abatable through "deeper" recovery
(i.e., 80 percent to 95 percent). The apportionment of reductions to these two types of recovery is
important because it affects the average recovery rate, and therefore labor costs, assumed in the
analysis. However, this apportionment depends on the market penetration, reduction efficiency,
and theoretical applicability assumed for recycling, all of which are subject to uncertainty.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-201
-------
SECTION IV— INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
• For refurbishment, costs are based on a circuit breaker with a nameplate capacity of 1,500 pounds
and an emissions rate of 20 percent; however, depending on the equipment (location, age, and
type), these values may vary significantly, adding a source of uncertainty in the MACs.
• Data on leak rates are limited, but a recent USEPA study looking at new equipment leak rates
may shed some light on this emissions source.
• Adjusting Costs for Specific Domestic Situations: The annual and capital costs associated with
implementing recycling and LDAR options are based on U.S. information. Although adjustments
for annual costs are included to account for differing country-specific labor costs, there remains a
potential source of uncertainty associated with recycling capital costs. Specifically, other
countries may be faced with higher costs from transportation and tariffs associated with
purchasing the technology abroad, or they may be faced with lower costs from domestic
production of these technologies. Also, it is assumed that LDAR capital costs are minimal;
however, repair costs can range from $10 to $100,000. Consequently, current MACs may
underestimate the dollars per tCO2eq associated with LDAR.
Country-Specific Tax and Discount Rates
A single tax rate is applied to the electric power sector in all countries to calculate the annual benefits
of each technology. Tax rates can vary across countries and, in the case of developing countries, taxes
may be less applicable. Similarly, the discount rate may vary by country. Improving the level of country-
specific detail will help analysts more accurately calculate benefits and hence breakeven prices.
IV. 10.4 References
Cryoquip. 2003. SF6 Gas Recycling Carts: Cart Selection. Available at .
Dilo. 2003. SF6 Recovery Equipment. Available at .
Ecofys. 2005. Reductions of SF6 Emissions from High and Medium Voltage Electrical Equipment in Europe. Final
Report to Capiel.
Ellerton, K. 1997. Recent Developments and the Outlook for Global SF6. AlliedSignal, Inc.
ICF. 2000. Personal Communication with Jam es D . M cC reary . Am erican Electric Pow er.
ICF. 2001. Personal Communication with Eric Campbell. Dilo Company, Inc.
McCracken, G.A., R. Christiansen, and M. Turpin. 2000. The Environmental Benefits of Remanufacturing:
Beyond SF6 Emissions Reduction. International Conference on SF6 and the Environment: Emissions
Reduction Technologies, November 2-3, San Diego, CA.
McRae, T. 2000. GasVue and the Magnesium Industry: Advanced SF6 Leak Detection. Presented at the
Conference on SF6 and the Environment: Emissions Reduction Strategies, November 1 -2, San Diego,
CA.
O'Connell, P., F. Heil, I. Henriot, G. Mauthe, H. Morrison, L. Neimeyer, M. Pittroff, R. Probst, and J.P.
Tailebois. 2002. SF6 in the Electric Industry, Status 2000. CIGRE.
Smythe, K. December 1-3, 2004. Trends in SF6 Sales and End-Use Applications: 1961-2003. Presented at the
International Conference on SF6 and the Environment: Emissions Reduction Technologies, Scottsdale,
AZ.
U.S. Energy Information Administration (USEIA). 2002. International Energy Outlook 2002. Washington,
DC: U.S. Department of Energy, Energy Information Administration, Office of Integrated Analysis
and Forecasting.
U.S. Environmental Protection Agency (USEPA). 2005. Inventory of U.S. Greenhouse Gas Emissions and
Sinks: 1990-2003. Washington, DC: USEPA.
U.S. Environmental Protection Agency (USEPA). 2006. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2020. Washington, DC: USEPA.
IV-202 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
Yokota, T., K. Yokotsu, K., Kawakita, H, Yonezawa, T. Sakai, and T. Yamagiwa. 2005. Recent Practice for
Huge Reduction ofSF6 Gas Emissions from GIS&GCB in Japan. Presented at the CIGRE SC A3 & B3 Joint
Colloquium in Tokyo.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-203
-------
SECTION IV — INDUSTRIAL PROCESSES • ELECTRIC POWER SYSTEMS
IV-204 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • MAGNESIUM PRODUCTION
IV.11 SF6 Emissions from Magnesium (Mg) Production
IV.11.1 Source Description
he Mg metal production and casting industry uses SF6 as a cover gas to prevent the
spontaneous combustion of molten Mg in the presence of air. The industry originally adopted
SF6 to replace SO2 as the primary cover gas. Although recent studies indicate some destruction
of SF6 in its use as a cover gas (Bartos et al., 2003), this analysis follows current IPCC guidelines (IPCC,
2000), which assume that all SF6 used is emitted into the atmosphere. Fugitive SF6 emissions may occur in
various phases of magnesium manufacture and casting, such as primary production, die-casting, and
recycling-based production. Additional processes that may use SF6 include sand and gravity casting;
however, these are assumed to be minor sources and are not included in the analysis. Baseline emissions
estimates under both a technology-adoption and a no-action baseline scenario are presented in
Tables 11-1 and 11-2.
IV.11.1.1 Technology-Adoption Baseline
Under the Technology-Adoption Baseline scenario, it is assumed that Mg producers and processors
outside of China will introduce technologies and practices aimed at reducing SF6 emissions. Specific
technologies include alternative cover gases, such as Novec 612 (a patented fluoroketone produced by
3M) and HFC-134a, and better containment and pollution control systems, which enable the use of SO2
without the industrial hygiene and odor problems of the past. Under this scenario, International
Magnesium Association (IMA) members, who account for 80 percent of the global Mg industry outside of
China (IMA, 2003), will meet a target of eliminating the use of SF6 by 2011.
Figure 11-1 presents total SF6 emissions from the Mg industry under the technology-adoption
scenario from 1990 through 2020. As shown in the graph, total emissions from the Mg industry remained
fairly constant through the mid-1990s, but then fell sharply to 9 MtCO2eq in 2000. The drop in global
emissions between 1995 and 2000 is the result of both facility closures in the United States and global
reductions in SF6 usage through more efficient operational practices. The latter is a response to increasing
SF6 gas prices and a growing environmental awareness of its high GWP. Additional plant closings have
been reported in Norway, Canada, and Japan, which have added to the decline in the share of global
emissions generated by OECD90+ through 2020. This lost production has been primarily absorbed by
China, which has dominated the foreign market with low-cost exports.
From 2000 through 2010, the steep decline in global SF6 emissions is attributable to the adoption of
alternative cover-gases; either SO2 or Novec 612 and HFC-134a. By the end of 2010, in accordance with
the IMA goal, all countries except China are assumed to have eliminated the use of SF6 from Mg
production and casting operations.
For China, it is assumed that some primary production and all casting facilities will use SF6 to
produce high-quality magnesium and products for the world market. Because Chinese producers and
processors are not IMA members and have not committed to the IMA emissions reduction goal, their SF6
use is assumed to continue through 2020. Consequently, from 2010 through 2020, the increase in global
emissions from 4 to 5 MtCO2eq will be driven entirely by China, whose emissions are expected to
increase from 2 to 4 MtCO2eq. In 2020, the China/CPA share of global emissions is expected to be 77
percent, compared with 0.3 percent in 1990. OECD's share of global emissions is projected to decrease
from 77 percent in 1990 to 21 percent in 2020 because of adoption of the IMA goal and reduction in
production capacity. In 2020, U.S. emissions account for a majority of OECD emissions. These emissions
are due to some U.S. casting and recycling firms that have not committed to phaseout use of SF6 (USEPA,
2005).
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-205
-------
SECTION IV — INDUSTRIAL PROCESSES • MAGNESIUM PRODUCTION
Table 11-1: Total SF6 Emissions from Mg Manufacturing (MtC02eq)—No-Action Baseline
Country/Region
Africa
Annex!
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
' 0.0
7.4
0.0
0.2
0.2
0.0
0.5
0.0
0.2
0.0
0.9
6.5
0.9
0.0
3.2
8.8
2010
0.0
8.3
0.0
0.5
1.7
0.0
1.0
0.0
0.1
0.0
1.5
6.8
1.2
0.0
4.6
12.1
2020
0.0
11.3
0.0 -
0.7
3.7
0.0
1.6
0.0
0.2
0.0
1.8
9.5
1.5
0.0
6.4
18.1
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Table 11-2: Total SF6 Emissions from Mg Manufacturing (MtC02eq)—Technology-Adoption Baseline
Country/Region
Africa
Annex!
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
0.0
7.4
0.0
0.2
0.2
0.0
0.5
0.0
0.2
0.0
0.9
6.5
0.9
0.0
3.2
8.8
2010
0.0
1.7
0.0
0.1
1.7
0.0
0.1
0.0
0.0
0.0
0.2
1.5
0.1
0.0
1.2
3.6
2020
0.0
1.1
0.0
0.0
3.7
0.0
0.0
0.0
0.0
0.0
0.0
1.0
0.0
0.0
1.0
4.8
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
IV-206 GLOBAL MITIGATION OF NOIM-C02 GREENHOUSE GASES
-------
SECTION IV— INDUSTRIAL PROCESSES • MAGNESIUM PRODUCTION
Figure 11-1: SF6 Emissions from Mg Manufacturing Based on a Technology-Adoption Scenario—1990-
2020 (MtC02eq)
• S&E Asia
• Latin America
D Middle East
• China/CPA
• Non-EUFSU
• OECD90+
• Japan
D EU-25
• United States
2000
2010
2020
Year
CPA = Centrally Planned Asia; EU-25 = European Union; Non-EU FSU = non-European Union Former Soviet'Union countries; OECD90+ =
Organisation for Economic Co-operation and Development; S&E Asia = Southeast Asia.
IV.11.1.2 No-Action Baseline
Under the no-action baseline scenario, it is assumed that Mg producers will take no action to reduce
their emissions; as a result, emissions projections do not reflect anticipated technology adoptions and/or
preventive maintenance steps taken to reduce emissions.
Figure 11-2 presents total SF6 emissions from Mg production under the no-action scenario from 1990
through 2020. The trends from 1990 through 2000 are the same as those discussed in the technology-
adoption baseline. From 2000 through 2020, global emissions in this scenario double to 19 MtCO2 as the
industry experiences strong growth, particularly in the die-casting and recycling segments. China/CPA
registers particularly significant emissions growth between 1990 and 2020, increasing its global share of
emissions from 0.3 percent in 1990 to approximately 21 percent in 2020. For OECD, emissions in this
scenario are assumed to continue to drop between 2000 and 2005 because of facility closures in Canada
stemming from pricing pressure from Chinese imports. However, by 2020, OECD emissions are expected
to return to 1990 levels as production levels increase. Because global emissions will increase by more than
50 percent during this period, the OECD share of global emissions will fall from 77 percent in 1990 to 53
percent in 2020.
Increasing Chinese primary production and die-casting is being fueled by local and foreign
investment, which has driven the overall increase in China/CPA's share of global emissions. This trend in
emissions growth is amplified by the assumption that 10 percent of their primary production utilized SF6
as the cover-gas mechanism.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-207
-------
SECTION IV — INDUSTRIAL PROCESSES • MAGNESIUM PRODUCTION
Figure 11-2: SF6 Emissions from Mg Manufacturing Based on a No-Action Scenario—1990-2020
(MtC02eq)
• S&E Asia
• Latin America
m Middle East
• China/CPA
• Non-EU FSU
• OECD90+
• Japan
DEU-25
• United States
1990
2000
2010
2020
Year
CPA = Centrally Planned Asia; EU-25 = European Union; Non-EU FSU = non-E:uropean Union Former Soviet Union countries; OECD90+ =
Organisation for Economic Co-operation and Development; S&E Asia = Southeast Asia.
IV.11.2 Cost of SF6 Emissions Reduction from Mg Production and
Processing Operations
IV.11.2.1 Abatement Options
Two potential abatement options are available for reducing SF6 emissions from Mg production and
processing operations. These technical measures are
• replacement with alternate cover gas — SO2 and
• replacement with alternate cover gas—fluorinated gases.
The remainder of Section IV. 11.2 provides an overview of each abatement option and details the cost
and reduction assumptions.
Replacement with Alternate Cover Gas—Sulfur Dioxide
In the past, SO2 has been used as a cover gas in Mg production and processing activities. However,
because of toxicity, 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 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 percent, because a
complete replacement of the cover gas system is involved. It is assumed to be applicable, with limited
exceptions (e.g., direct-chill casting), to all global Mg producers and processors. For all countries except
China, the maximum market penetration for this option is assumed to be 50 percent of the emissions of
IV-208
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • MAGNESIUM PRODUCTION
SF6 in the technology-adoption baseline. For China, where SO2 replacement is the only option to be
implemented, maximum market penetration is assumed to be 100 percent.
Cost and Emissions Reduction Analysis
• Capital/Upfront Costs. The assumed country-level capital spending requirements for the
implementation of an SO2 system were based on cost information developed for the Canadian
Mg production and processing system (Environment Canada, 1998). For Canada, with two
primary producers and five casters, the costs to install SO2 cover gas systems with containment
and pollution control systems were estimated at approximately $10 million. These capital costs
were divided by estimated Canadian emissions from Mg production and processing in 2000. The
resulting capital cost per tCO2eq, $5.73 per tCO2eq, was then applied to the rest of the world.
• Annual Costs. Because of potential employee turnover, it is assumed that costs associated with
worker training will occur on an annual basis at a cost of $50,000 per facility. Assuming seven
facilities were operating in Canada in 2000, total annual costs of $350,000 were divided by
estimated Canadian emissions from Mg production and processing in 2000, adjusted to account
for the variation of labor rates across countries and applied to the rest of the world.
• Cost Savings. The price of SO2 is significantly lower than that of SF6. Using industry-based
estimates, the installation of SO2 technology will reduce gas purchase costs by approximately 90
percent (ICF, 1998). It is assumed that all primary producers, casters, and recyclers use similar
amounts of SF6.
Replacement with Alternate Cover Gas—Fluorinated Gases
Research has yielded a number of candidate fluorinated compounds, such as Novec 612 and HFC-
134a, as cover gas substitutes for SF6 (Mibrath, 2002; Rickets, 2002; Hillis, 2002). While fluorinated gases
have an advantage over SO2 because they have potentially fewer associated health, odor, and corrosive
impacts, some current candidate gases still have GWPs. However, these GWPs are well below that of SF6,
and within a few years, fluorinated gases will likely provide a functional replacement for SF6. It is
estimated that where this technology is implemented, GWP-weighted cover gas emissions could be
reduced by 95 percent to essentially 100 percent. (This analysis uses an average reduction of 97 percent.)
This option is assumed to be applicable to all countries that produce and/or process Mg, except China,
where only SO2 replacement is assumed to be implemented. For those countries in which the option is
implemented, the maximum market penetration is assumed to be 50 percent of the baseline emissions of
SF6 in the technology-adoption scenario.
Cost and Emissions Reduction Analysis
• Capital/Upfront Costs. The capital costs associated with adopting the various fluorinated gases
are likely to vary, with some gases requiring relatively little retrofitting of the SF6 cover gas
system and others requiring more. This analysis conservatively assumes that, on average, the
capital costs of replacing SF6 with fluorinated gases will be the same as the capital costs of
replacing SF6 with SOj, which include costs for developing a special distributed feed system and
process and implementing pollution control systems.
• Annual Costs. Costs are conservatively assumed to be similar to those required for SO2
replacement.
• Cost Savings. It is assumed that alternate fluorinated gases will cost the same as SF6; hence, no
cost savings will be realized with a switch to this technology.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-209
-------
SECTION IV — INDUSTRIAL PROCESSES • MAGNESIUM PRODUCTION
IV.11.3 Results
This section discusses the result from the MAC analysis for the world and various regions for the no-
action and technology-adoption scenarios.
IV.11.3.1 Data Tables and Graphs
Tables 11-3 through 11-8 provide a summary of the potential emissions reduction opportunities and
associated costs for the world and various regions in 2010 and 2020 for the no-action and technology-
adoption baselines. The costs to reduce 1 tCO2eq are presented for a discount rate of 10 percent and a tax
rate of 40 percent.
Table 11-3: Emissions Reductions in 2010 and Breakeven Costs ($/tC02eq) for Mg Production at 10%
Discount Rate, 40% Tax Rate (MtC02eq)—No-Action Baseline
2010
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$15
0.00
6.13
0.00
0.47
1.65
0.00
1.01
0.00
0.11
0.00
1.43
6.70
1.20
0.00
4.50
11.94
$30
0.00
8.13
0.00
0.47
1.65
0.00
1.01
0.00
0.11
0.00
1.43
6.70
1.20
0.00
4.50
11.94
$45
0.00
8.13
0.00
0.47
1.65
0.00
1.01
0.00
0.11
0.00
1.43
6.70
1.20
0.00
4.50
11.94
$60
0.00
8.13
0.00
0.47
1.65
0.00
1.01
0.00
0.11
0.00
1.43
6.70
1.20
0.00
4.50
11.94
>$60
0.00
8.13
0.00
0.47
1-.65
0.00
1.01
0.00
0.11
0.00
1.43
- 6.70
1.20 •
0.00
4.50
11.94
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
IV-210 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • MAGNESIUM PRODUCTION
Table 11-4: Emissions Reductions in 2020 and Breakeven Costs (S/tC02eq) for Mg Production at 10%
Discount Rate, 40% Tax Rate (MtC02eq)—No-Action Baseline
2020
Country/Region
Africa
Annex!
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$15
0.00
11.13
0.00
0.70
3.74
0.00
1.54
0.00
0.18
0.00
1.78
9.35
1.51
0.00
6.26
17.85
$30
0.00
11.13
0.00
0.70
3.74
0.00
1.54
0.00
0.18
0.00
1.78
9.35
1.51
0.00
6.26
17.85
$45
0.00
11.13
0.00
0.70
3.74
0.00
1.54
0.00
0.18
0.00
1.78
9.35
1.51
0.00
6.26
17.85
$60
0.00
11.13
0.00
0.70
3.74
0.00
1.54
0.00
0.18
0.00
1.78
9.35
1.51
0.00
6.26
17.85
>$60
0.00
11.13
0.00
0.70
3.74
0.00
1.54
0.00
0.18
0.00
1.78
9.35
1.51
0.00
6.26
17.85
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Table 11-5: Emissions Reductions in 2010 and Breakeven Costs ($/tC02eq) for Mg Production at 10%
Discount Rate, 40% Tax Rate (MtC02eq)—Technology-Adoption Baseline
2010
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$15
0.00
1.54
0.00
0.05
1.65
0.00
0.10
0.00
0.01
0.00
0.14
1.39
0.12
0.00
1.17
3.40
$30
0.00
1.54
0.00
0.05
1.65
0.00
0.10
0.00
0.01
0.00
0.14
1.39
0.12
0.00
1.17
3.40
$45
0.00
1.54
0.00
0.05
1.65
0.00
0.10
0.00
0.01
0.00
0.14
1.39
0.12
0.00
1.17
3.40 ,
$60
0.00
1.54
0.00
0.05
•1.65
0.00
0.10
0.00
0.01
0.00
0.14
1.39
0.12
0.00
1.17
3.40
>$60
0.00
1.54
0.00
0.05
1.65
0.00
0.10
0.00
0.01
0.00
0.14
1.39
0.12
0.00
1.17
3.40
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-211
-------
SECTION IV — INDUSTRIAL PROCESSES • MAGNESIUM PRODUCTION
Table 11-6: Emissions Reductions in 2020 and Breakeven Costs ($/tC02eq) for Mg Production at
Discount Rate, 40% Tax Rate (MtC02eq)—Technology-Adoption Baseline
10%
2020
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South &SE Asia
United States
World Total
$0
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
$15
0.00
0.90
0.00
0.00
3.74
0.00
0.00
0.00
0.00
0.00
0.00
0.90
0.00
0.00
0.90
4.63
$30
0.00
0.90
0.00
0.00
3.74
0.00
0.00
0.00
0.00
0.00
0.00
0.90
0.00
0.00
0.90
4.63
$45
0.00
0.90
0.00
0.00
3.74
0.00
0.00
0.00
0.00
0.00
0.00
0.90
0.00
0.00
0,90
4.63
$60
0.00
0.90
0.00
0.00
3.74
0.00
0.00
0.00
0.00
0.00
0.00
0.90
0.00
0.00
0.90
4J3
>$60
0.00
0.90
0.00
0.00
3.74
0.00
0.00
0.00
0.00
0.00
0.00
0.90
0.00
0.00
0.90
4.63
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Table 11-7: Emissions Reduction and Costs in. 2020—No-Action Baseline
Reduction Option
S02 replacement
Fluorinated covergas
Cost(2000$/tC02eq)
DR=10%,TR=40%
Low High
$0.53 $0.79
$1.21 $1.48
Emissions
Reduction of
Option
(MtCOzeq)
10.90
6.95
Reduction
from 2020
Baseline (%)
60.3%
38.5%
Running
Sum of
Reductions
(MtC02eq)
10.90
17.85
Cumulative
•" Reduction
from 2020
Baseline (%)
«0,3%
98,8%
Table 11-8: Emissions Reduction and Costs in 2020—Technology-Adoption Baseline
Cost(2000$/tC02eq)
DR=10%,TR=40%
Reduction Option
Low
High
Emissions
Reduction of
Option
(MtC02eq)
Reduction
from 2020
Running
• Sum of
Reductions
(MtCOjeq)
S02 replacement $0.53 $0.79 4.19
Fluorinated eovergas $1.21 $1.48 0.44
86.6%
9.2%
4.19
4.63
9S.8%
IV-212
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • MAGNESIUM PRODUCTION
IV.11.3.2 Global and Regional MACs and Analysis
This section discusses the results from the MAC analysis of the world and selected countries and
regions, including China, Japan, the United States, EU-15, other OECD, and the rest of the world.
Figure 11-3 presents the 2010 and 2020 global technology-adoption and no-action MACs for Mg
production and processing. The technology-adoption MACs reflect the successful achievement of the
IMA goal to eliminate SF6 use by 2011. Although it is expected that the majority of SF6 use will be
eliminated by 2011, some use is assumed to continue to occur in the United States and China in 2020. The
rightward shift in technology-adoption MACs between 2010 and 2020 primarily reflects the continuing
growth of Chinese primary production and die-casting and their associated use of SF6 as a cover gas.
Figure 11-3: 2010 and 2020 Global Technology-Adoption and No-Action MACs for Mg Production
Emissions Reductions ($/tCO2eq)
D 0 -»• ^ M
3 Ol O O1 O
D O O O O
2010
r
0
p-
r
2010 2020
2020 -j . — r
"L<*
-1- t_ 1 A J i-
• I ecnnology-Adoption
— ^— No-Action
5 10 15 20
Cumulative Emissions Reductions (MtCO2eq)
In contrast, the no-action MACs assume that no technology adoptions and/or preventive maintenance
steps are taken to reduce emissions. Because the SF6 emissions intensities (i.e., SF6 emissions per unit of
Mg production) for primary, secondary, and casting production are assumed to remain constant between
2010 and 2020, increasing emissions (reflected by the rightward shift in the 2020 MAC) are driven by
continuing positive production growth in all regions, particularly the United States and China.
Figures 11-4 and 11-5 present 2010 and 2020 regional technology-adoption MACs for China, Japan,
the United States, EU-15, other OECD, and the rest of the world. The 2010 MAC represents potential
emissions reductions and associated costs for global Mg producers and processors a year before the
assumed successful adoption of the IMA goal. Consequently, the small reductions associated with EU-15
countries, Japan, other OECD (e.g., Norway and Canada), and the rest of the world (e.g., Brazil, Taiwan,
Israel, the Russian Federation, Ukraine, and Kazakhstan) reflect this assumption. In China and the United
States, some Mg casters and recyclers have not committed to the IMA goal; consequently, significant
emissions reductions are still achievable through implementation of abatement options. In China, SO2 is
considered the only abatement option, while in the United States, both SO2 and alternate fluorinated
gases are assumed to be available. In all regions, SO2 is estimated to be less expensive than fluorinated
gases such as Novec 612 and HFC-134a.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
IV-213
-------
SECTION IV— INDUSTRIAL PROCESSES • MAGNESIUM PRODUCTION
Figure 1 1 -4: 201 0 Regional Technology-Adoption MACs
-5: $2.00 -,
0>
CM
O
S $1.50 -
Emissions Reductions
/»•«/>•&»
3 P r*1
3 tn b
3 O O
0
M 1 f
| — •* China
,--A' Rest of the world
j — • United States
•• - EU-15
• — »• Japan
_J — « other OFCD
I
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
Cumulative Emissions Reductions (MtC02eq)
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Figure 11-5: 2020 Regional Technology-Adoption MACs
—* China
Rest of the world
—• United States
— = EU-15
—*• Japan
—« Other OECD
E
UJ $0.00
0.5
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Cumulative Emissions Reductions (MtC02eq)
5.0 5.5
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
In 2020, for the technology-adoption MACs, reductions are only available for the United States and
China. For all other Mg producing countries, it is assumed that the IMA goal has been successfully
achieved. In the United States, although abatement costs are the same as in 2010, the curve has shifted to
the left. This reflects the fact that primary production and a majority of casting companies in the United
States have agreed to meet the IMA goal; consequently, the only emissions available for abatement in
2020 are from the remaining casting and recycling firms that have not committed to phaseout SF6 use. In
China, carbon costs for SO2 abatement are assumed to be the same as in 2010 ($0.53/tCO2eq); however, the
IV-214
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION IV — INDUSTRIAL PROCESSES • MAGNESIUM PRODUCTION
MAC has shifted to the right because of increased use of SF6 during primary production and casting in
the baseline, which is projected to grow significantly to meet local demand.
IV.11.3.3 Uncertainties and Limitations
Uncertainties and limitations persist despite attempts to incorporate all publicly available
information on international Mg production and processing. Additional information would improve the
accuracy of the MACs.
Mg Production and Processing
Although historical primary production statistics are available for most countries, a source of
uncertainty is the limited information on historical country-specific secondary and die-casting production
and Chinese primary production. There is also limited information on country-specific SF6 usage rates
(i.e., emissions factors); these rates are likely to vary by facility and country because of differing
operational practices and manufacturing processes. Additionally, the current methodology assumes that
all SF6 gas used is emitted into the atmosphere; however, recent studies (Bartos et al., 2003; USEPA, 2004)
indicate that some SF6 degradation may occur (e.g., 10 percent) during magnesium melt protection. As a
result, current SF6 emissions factors may be higher than actual emissions rates. Improved country-specific
production-level data and growth projections and SF6 usage rates would improve the baseline emissions
estimates to which the abatement options are applied.
IMA SF6 Reduction Goal
In the technology-adoption baseline, emissions projections assume attainment of the IMA goal, which
is the elimination by 2011 of SF6 use in Mg production and processing operations in all countries except
China and the United States where some processors have not committed to the goal. Because there is
limited information on country-specific implementation of alternate cover gas practices, at the current
time, it is unclear whether this goal will be achieved. Improved information on country-specific practices
would improve modeling of the IMA goal.
Implementation of Alternate Cover Gases
For most countries, the MACs assume a 50/50 percent market penetration of SO2 and fluorinated
(Novec 612 and HFC-134a) cover gases. Because of limited information, it is unclear if this is an accurate
representation of actual future industry application of these technologies. However, because the
reduction efficiencies for the abatement options are high (i.e., 100 percent for SO2 and an average of 97
percent for fluorinated gases), the uncertainty in the reduction achieved will be low. Recent USEPA
studies have noted that both Novec 612 and HFC-134a undergo significant degradation while providing
melt protection; consequently, reduction efficiencies may be even higher than those currently used (i.e.,
99.9 percent) (USEPA, 2004).
Costs of Alternative Fluorinated Cover Gases
The capital costs associated with adopting the various fluorinated gases are likely to vary, with some
gases requiring relatively little retrofitting of the SF6 cover gas system and others requiring more. This
analysis conservatively assumes that, on average, the capital costs of replacing SF6 with fluorinated gases
will be the same as the capital costs of replacing SF6 with SO^ which includes costs for developing a
special distributed feed system and process and implementing pollution control systems. Not all
fluorinated gases are likely to require such systems. Thus, in some cases, this analysis probably
overestimates the costs of adopting alternative fluorinated gases.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES IV-215
-------
SECTION IV — INDUSTRIAL PROCESSES « MAGNESIUM PRODUCTION
Adjusting Costs for Specific Domestic Situations
The annual and capital costs associated with implementing the SC>2 abatement option are based on
information provided in the Powering GHG Reduction through Technology Advancements report by
Environment Canada (1998). Although adjustments for annual costs are provided to account for differing
country-specific labor costs, no adjustments for capital costs are applied. This is a potential source of
uncertainty because countries other than Canada may be faced with higher costs due to transportation
and tariffs associated with purchasing the technology from abroad or could be faced with lower costs due
to domestic production of these technologies.
Country-Specific Tax and Discount Rates
A single tax rate is applied to the Mg sector in all countries to calculate the annual benefits of each
technology; however, tax rates can vary across countries. Similarly, the discount rate may vary by
country. Improving the level of country-specific detail will help analysts more accurately calculate
benefits and hence breakeven prices.
IV.11.4 References
Bartos S., J. Marks, R. Kantamaneni, and C. Laush. 2003. "Measured SF6 Emissions from Magnesium Die
Casting Operations." In Howard A. Kaplan (ed.) Magnesium Technology pp. 23-27. Warrendale, PA:
Minerals, Metals, & Materials Society.
Environment Canada. 1998. Powering GHG Reductions through Technology Advancement. Clean Technology
Advancement Division, Environment Canada.
Hillis, J. 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.
ICF. 1998. Personal communication from Norsk Hydro.
Intergovernmental Panel on Climate Change (IPCC). 2000. "Good Practice Guidance and Uncertainty
Management in National Greenhouse Gas Inventories." Section 3.3. Available at .
International Magnesium Association (IMA). 2003. Available at .
Mibrath, D. 2002. Development of 3M Novec 612 Magnesium Protection Fluid as a Substitute for SF6 Over
Molten Magnesium. 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.
U.S. Environmental Protection Agency (USEPA). 2004. Characterization of Cover Gas Emissions from U.S.
Magnesium Casting. EPA 430-R-04-004. Washington, DC: U.S. Environmental Protection Agency,
Office of Air and Radiation.
U.S. Environmental Protection Agency (USEPA). 2006. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2020. Washington, DC: USEPA.
IV-216 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
Section IV: Industrial Processes Sector Appendixes
Appendixes for this section are available for download from the USEPA's Web site at
http://www.epa.gov/nonco2/econ-inv/international.html.
-------
V. Agriculture
-------
SECTION V — AGRICULTURE • PREFACE
Section V presents international emission baselines and marginal abatement curves (MACs) for all
significant agricultural non-CO2 sources. There are subsections that address emissions and mitigation
options for croplands, rice cultivation, and livestock. These sources are associated with methane (CH4)
and nitrous oxide (N2O) emissions, as well as soil carbon. MAC data are focused on percentage reduction
values from baseline emissions. These data can be downloaded in spreadsheet format from the USEPA's
Web site at .
Section V—Agriculture is organized as follows:
V.I Introduction and Background
V.2. Emissions Characterization, Baselines, and Mitigation Scenarios
V.2.1 Croplands (N2O and soil carbon)
V.2.2 Rice (CH4, N20 and soil carbon)
V.2.3 Livestock (CH4 and N20)
V.3 Results
V.3.1 Estimating Average Costs and Constructing Abatement Curves
V.3.2 Croplands
V.3.3 Rice
V.3.4 Livestock
V.3.5 Total Agriculture
V.3.6 Agricultural Commodity Market Impacts
V.4 Conclusions
V-ii GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • INTRODUCTION AND BACKGROUND
V.1 Introduction and Background
Agricultural activities currently generate the largest share, 63 percent, of the world's anthropogenic
non-carbon dioxide (non-CO2) emissions (84 percent of nitrous oxide [N2O] and 52 percent of methane
[CH4]), and make up roughly 15 percent of all anthropogenic greenhouse gas emissions (U.S.
Environmental Protection Agency [USEPA], 2006; Prentice et al., 2001).1 Agricultural greenhouse gas
emissions are projected to increase significantly over the next 20 years, especially in Asia, Latin America,
and Africa, because of increased demand for agricultural products as a result of population growth;
rising per capita caloric intake; and changing diet preferences, such as an increased consumption of meat
and dairy products over grains and vegetables (see Food and Agriculture Organization [FAO], 2002).
Agricultural soil N2O emissions are projected to increase 37 percent by 2020 compared with 2000 levels,
enteric livestock CH4 emissions are projected to increase 30 percent, manure CH4 and N2O to increase 24
percent,2 and rice CH4 to increase 22 percent (USEPA, 2006).
The agricultural sector presents unique challenges to developing greenhouse gas mitigation cost
estimates at regional and international scales. First, there is a high degree of spatial and temporal
heterogeneity in biophysical and management conditions and thus in resulting greenhouse gas emissions,
which are rarely directly monitored. This fact makes it challenging to extrapolate the greenhouse gas and
cost implications of farm-level mitigation analyses, the scale at which much of the literature on this
subject is found. Any large-scale mitigation analysis should emphasize the broad trends and direction
and magnitude of changes, which requires some trade-off in accuracy for very small spatial (e.g.,
individual farms) and temporal (e.g., days and seasons) scales. Second, there is a paucity of regional cost
data from which one can estimate the implications of implementing greenhouse gas mitigation practices,
in terms of changes in inputs, revenue, and labor. Third, estimating the expected level of adoption of the
mitigation options in response to financial incentives (e.g., carbon price) or, alternatively, in response to
extension services with greenhouse gas reduction objectives, is difficult given the information and
cultural barriers to adoption in different regions.
Nevertheless, agricultural net greenhouse gas and non-CO2 mitigation analyses have been developed
for several countries and the world. Some analyses include a relatively comprehensive set of greenhouse
gas mitigation options with a dynamic economic and biophysical representation of the agricultural and
forest sectors (see USEPA [2005a] for the United States). Others target individual agricultural emissions
sources with static, engineering mitigation estimation methods (see Kroeze and Mosier [1999] for global
cropland N2O and enteric CH4 emissions and Reimer and Freund [1999] for global rice emissions; see also
Table 3.27 in Moomaw et al. [2001] of IPCC Working Group III). The USEPA supported the development
of global mitigation estimates for cropland N2O, livestock enteric and manure CH4, and rice CH4
(DeAngelo et al., 2006) that were then incorporated into the Energy Modeling Forum-21 (EMF-21) study
of global multigas mitigation options (van Vuuren et al. [2006]). This report improves on the agricultural
analysis conducted for EMF-21 in a number of areas.
1 This value compares the International Panel on Climate Change (IPCC) (Prentice et al., 2001) estimate of gross
annual CO2 emissions from fossil fuel combustion, cement manufacturing, and land-use change with the USEPA
(2006) estimate of all anthropogenic non-CO2 emissions. Fossil fuel CO2 emissions associated with agriculture (e.g.,
on-farm equipment, fertilizer production) are not assigned to the agricultural sector in this estimate.
2 The estimated increase of manure CH4 and N2O emissions represents a joint estimate based on CO2 equivalent
units using global warming potentials (GWPs) from the IPCC Second Assessment Report.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-1
-------
SECTION V — AGRICULTURE • INTRODUCTION AND BACKGROUND
V.1.1 Brief Points of Comparison with Other Non-CO2 Emissions
Sectors
A few points about how baseline assumptions and methods for agriculture compare with those in
other non-CO2 sectors of this report are in order. First, baseline emissions projections used in this section
are not entirely consistent with baseline projections developed by the USEPA (2006). This is the case for
cropland N2O and rice CH4 emissions, where separate baseline emissions projections are developed with
process-based models. These models, described later in this section, are also used for the mitigation
scenarios, so that assumptions and all underlying activity data on both the baseline and mitigation sides
of the equation are consistent.
Second, although the focus of this report is on the non-CO2 greenhouse gases, soil carbon is included
in the agricultural analysis. Including soil carbon is important for agriculture because it provides a more
comprehensive picture of the net greenhouse gas effects of mitigation options that primarily target N2O
and CH4.
Third, the agricultural analysis, like other sectors in this report, presents marginal abatement curves
(MACs) by region, for the 2000 to 2020 period, showing the technical, net greenhouse gas mitigation
potential at various levels of U.S. dollars (USD) per tonne of CO2 equivalent ($/tCO2eq),, representing the
breakeven price of each mitigation option. The approach used to estimate the technical mitigation
potential is similar to that used in other sectors —a bottom-up, engineering approach. However, the
agricultural analysis illustrates the sensitivity of the mitigation estimates to potential economic market
feedbacks as a result of adopting the mitigation options (i.e., showing the effects of simultaneous changes
in crop yields, livestock productivity, commodity prices, cropland area, livestock herd size, and
emissions).
V.1.2 Previous Estimates for EMF-21 and New Improvements
Previously, the USEPA helped produce a non-CO2 mitigation analysis for world agriculture
(DeAngelo et al., 2006) to assist climate-economic and integrated assessment modelers who participated
in the EMF-21 study represent the agricultural sector. The study generated MACs by major world regions
for cropland N2O, livestock enteric CH4, manure CH4, and rice CH4 for 2010. This analysis used a static,
engineering approach by relying on literature sources to identify the non-CO2 reductions associated with
each mitigation option, extrapolating those results beyond their original scale of analysis (farm, region, or
nation) to other world regions, estimating regionally specific changes in input costs with FAOSTAT and
other data sources, and adjusting the extent to which each mitigation option applied to different regions.
Summary results of this previous analysis and how they compare with the current analysis are presented
in Appendix P.
The current analysis uses new approaches to improve on the previous EMF-21 study in a number of
areas. First, biophysical, process-based models (DAYCENT and DNDC) are used to better capture the net
greenhouse gas and yield effects of the cropland and rice emissions baseline and mitigation scenarios.
The previous analysis estimated the single, dominant gas effects only (e.g., no N2O or soil carbon effects
for rice CH4 mitigation practices). Furthermore, process-based models better reflect the heterogeneous
emissions and yield effects over space and time of adopting mitigation practices, whereas the previous
analysis usually assumed a uniform percentage change in emissions and/or yields across regions. The
process-based models also ensure greater consistency in underlying assumptions and activity data
between baseline and mitigation scenarios. For example, when emissions projections are estimated with
IPCC Tier I default methodologies, it is not always possible to identify what underlying management
V-2 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • INTRODUCTION AND BACKGROUND
practices are taking place, which in turn makes it difficult to ascertain if the chosen mitigation options are
indeed additional to baseline management practices.
New mitigation options are assessed (e.g., slow-release fertilizers, nitrogen inhibitors, and no till),
and more detailed, less aggregated results are provided for individual crop types (e.g., maize, wheat, and
soybeans) under both irrigated and rain-fed conditions. Lastly, sensitivity experiments using a global
agricultural trade model (IMPACT of the International Food Policy Research Institute [IFPRI]) are
conducted to assess the agricultural commodity market effects of adopting the mitigation options.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-3
-------
SECTION V — AGRICULTURE • INTRODUCTION AND BACKGROUND
V-4 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V —AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
V.2 Emissions Characterization, Baselines, and Mitigation
Scenarios
V.2.1 Croplands (N2O and Soil Carbon)
V.2.1.1 Cropland N2O and Soil Carbon Emissions Characterization
N2O is typically the dominant greenhouse gas source from agricultural systems and is produced
naturally in soils through the processes of nitrification and denitrification. These are soil microbial
processes whereby ammonium (NH3) is reduced to nitrate (NO3) under aerobic or oxygen-rich conditions
(nitrification), and nitrate is reduced to molecular nitrogen (N2) under anaerobic or oxygen-poor
conditions (denitrification). A number of activities add nitrogen to soils, thereby increasing the amount
available for nitrification and denitrification, and ultimately the amount of N2O emitted to the
atmosphere. Activities may add nitrogen to soils either directly or indirectly. Direct additions occur
through nitrogen fertilizer use, application of managed livestock manure and sewage sludge, production
of nitrogen-fixing crops and forages, retention of crop residues, and cultivation of histosols (i.e., soils with
high organic-matter content, also known as organic soils). Indirect emissions occur through volatilization
and subsequent atmospheric deposition of applied nitrogen, as well as through surface runoff and
leaching of applied nitrogen into groundwater and surface water.
Other soil management activities, such as irrigation, drainage, tillage practices, and fallowing of land,
can also affect fluxes of N2O, as well as soil carbon and fossil fuel CO2 emissions. Fossil fuel CO2
emissions can be generated on-farm by agricultural equipment and off-farm or upstream through the
energy-intensive production of fertilizers.3 These fossil fuel CO2 emissions are not included in this study;
thus some net emissions reduction benefits of the mitigation options are likely to be underestimated in
this report.4
Agricultural soil carbon emissions and/or sequestration tend to be less dominant than N2O emissions
in terms of the net greenhouse gas picture under baseline conditions; however, enhancing soil carbon
sequestration represents a significant greenhouse gas mitigation option, potentially more viable than N2O
reductions (see USEPA [2005a]). Croplands often emit CO2 as a result of conventional tillage practices
and other soil disturbances. This occurs when soils containing organic matter that would otherwise be
protected by vegetative cover are exposed to the air through tillage disturbances and become susceptible
to decomposition. Conservation tillage — defined in the United States as any tillage system that maintains
at least 30 percent of ground covered by crop residue after planting (Conservation Technology
Information Center [CTIC], 1994)—eliminates one or several practices associated with conventional
tillage, such as turning soils over with a moldboard plow and mixing soils with a disc plow (Lai et al.,
3 Under IPCC greenhouse gas inventory reporting guidelines and in the annual Inventory of U.S. Greenhouse Gas
Emissions and Sinks reported by the USEPA, these fossil fuel CO2 emissions are reported as energy-sector not
agricultural-sector emissions.
4 In the USEPA (2005a), the FASOM-GHG model of U.S. forestry and agriculture shows that on-farm and upstream
fossil fuel CO2 emissions associated with crop production are roughly 40 percent of the size of the joint CH4 and N2O
emissions in agriculture, on a CO2-equivalent basis. The DAYCENT modelers for this report assumed that for every
unit of nitrogen fertilizer applied, 0.8 units of CO2 were generated from fertilizer manufacturing, though these
numbers were intentionally excluded from this report to maintain consistency across emissions categories.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-5
-------
SECTION V —AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
1998). Conservation tillage, including no-till, allows crop residues to remain on the soil surface as
protection against erosion.
Lastly, seasonal temperature and precipitation changes, as well as regional climate variability,
influence rates of both soil N2O and carbon emissions.
V.2.1.2 DAYCENT Baseline Estimates of Cropland N2O, Soil Carbon, and Yields
The DAYCENT model is used to estimate baseline and mitigation scenario emissions of N2O and soil
carbon for a significant share of the world's nonrice croplands. (DAYCENT, rather than IPCC default
values, is also the tool now used to estimate the majority of agricultural soil N2O emissions for the annual
Inventory of U.S. Greenhouse Gas Emissions and Sinks, reported by the USEPA). Use of DAYCENT offers the
advantage of a consistent methodology and tool across regions and between the baseline and mitigation
scenarios. The DAYCENT model and emissions baseline methodologies are described briefly here, and
further details are provided in Appendix Q.
The DAYCENT model (Del Grosso et al, 2001; Parton et al., 1998) is a process-based model that
simulates crop growth, soil organic-matter decomposition, greenhouse gas fluxes, nitrogen deposited by
grazing animals, and other biogeochemical processes using daily climate data, land management
information, and soil physical properties. N2O emissions estimated by DAYCENT account for nitrogen
additions, crop type, irrigation, and other factors and capture both direct (through fertilizer applications)
and indirect (through volatilization and leaching) N2O emissions.
Global baseline N2O emissions for this report are estimated from DAYCENT to be 799 MtCO2eq in
2000, 795 MtCO2eq in 2010, and 859 MtCO2eq in 2020. With the net effects of soil carbon, global net
greenhouse gas estimates are 839 MtCO2eq, 830 MtCO2eq, and 893 MtCO2eq for 2000, 2010, and 2020,
respectively. These estimates represent the mean of decadal averages (e.g., 1996 to 2005 mean for reported
year 2000). Emissions estimates for individual key countries and regions are provided in Table 1-1 (see
Section V.l.3.2 for additional baseline data).
Table 1-1: DAYCENT N20 and Soil Carbon Estimates for 2000,2010, and 2020 by Key Region (MtC02eq/yr)
Region
United States
EU-15
Eastern Europe
FSU
Mexico
Brazil
India
China
2000
N20 Soil Carbon
164 3
95 -4
37 2
187 26
14 . 1
28 0
69 -3
84 7
2010
N20 Soil Carbon
176 2
98 -5
37 2
127 30
16 1
30 0
74 -4
95 3
2020
N20 Soil Carbon
197 3
' 107 -6
39 2
126 36
17 >0
30 0
78 -5
105 -1
EU-15 = European Union; FSU = Former Soviet Union.
Note: Negative numbers indicate net sequestration.
As described below, the cropland coverage for the DAYCENT simulations is incomplete. Therefore,
the baseline estimates from DAYCENT are intended to serve as the foundation from which to assess the
general implications of mitigation scenarios. The DAYCENT baseline estimates are not intended to serve
as independent national and global inventory estimates, which can be found elsewhere in the literature
(USEPA 2006; USEPA 2005b; Robertson 2004; IPCC 2001; Ehhalt et al., 2001; Mosier et al., 1998).
V-6
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE « EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
Appendix S describes how the DAYCENT N2O emissions baseline estimates compare with the USEPA's
(2006) more comprehensive estimates.
DAYCENT explicitly simulates the major crop types only: maize, spring and winter wheat, and
soybeans. However, analogous crops are added to these major crop types (e.g., rye, barely, and oats with
wheat; millet and sorghum with maize) to increase the coverage of cropland area and to capture a higher
portion of nitrogenous fertilizer applications. Grazing-land emissions are not included, and emissions
due to residue burning are not included. For these reasons, the DAYCENT baseline N2O estimates are
generally lower than other published inventory studies for national and world total N2O emissions.
DAYCENT simulations for maize, wheat, and soybean areas are run under both irrigated and rain-
fed conditions.5 The relative portion of maize, wheat, and soybean areas under irrigated and rain-fed
conditions are provided by IFPRI and vary by region and over time.
Underlying data for the emissions estimates include
global data sets of weather, soils, cropland area, and native
vegetation, mapped to an approximate 2° x 2° resolution.
Daily weather data (i.e., precipitation and maximum and
minimum temperatures) from the National Oceanic and
Atmospheric Administration (NOAA) National Centers for
Environmental Prediction for the 1991 to 2000 period are used
under both baseline and mitigation options; therefore, there is
no explicit assumption about anthropogenic climate change
for the 2000 to 2020 period. Soils data include texture
percentages of clay, sand, and silt and come from
FAO/UNESCO (1996). Histosols, a source of N2O, are not
included in the simulations. Cropland distribution and the
fractional area of specific crops are taken from both the
International Geosphere Biosphere Programme (IGBP) land
cover classification (Belward et al., 1999; Belward, 1996) and
the Global Land Cover (GLC) data set of Leff et al. (2004).
Cropland area is assumed to remain constant over time in
the DAYCENT simulations under both baseline and
mitigation options (the same assumption is held for rice areas
with the DNDC simulations). The subject of changing area in
response to market feedbacks due to the implementation of
mitigation options is discussed in Section V.I.3.6.
In addition to simulating N2O emissions from mineral cropland soils, a DAYCENT simulation was
performed for those same areas as though they were covered by native vegetation and never cultivated
(using potential vegetation from Cramer et al. [1999] and Melillo et al. [1993]), so that anthropogenic
emissions are isolated from natural background emissions. Therefore, all reported emissions estimates
Box 1-1: DAYCENT Estimates of U.S.
Agricultural N20 Emissions in This
Report versus Inventory
The annual Inventory of Greenhouse Gas
Emissions and Sinks, reported by the
USEPA, now uses the DAYCENT
model to estimate the majority of
agricultural soil N2O emissions.
Though DAYCENT is used in this
report and provides estimates for U.S.
agricultural soil N2O emissions under
baseline and mitigation scenarios, the
U.S. estimates in this report are not the
same as the U.S. estimates in the
Inventory. This is because the
Inventory uses input data specific to
the United States, while the input data
used by DAYCENT in this report come
from global data sets to provide as
much consistency as possible across
regions, including the United States.
5 Rain-fed conditions mean that the crop receives no extra water in addition to rainfall and the resultant water stored
in the soil. To simulate irrigation, extra water is added, if necessary, to bring soil water content to field capacity once
per week for 20 weeks during the growing season. This minimizes or eliminates plant water stress and is an
assumption consistent with the fact that farmers typically irrigate only when necessary because irrigation requires
resources.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-7
-------
SECTION V — AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
under both baseline and mitigation scenarios from croplands represent anthropogenic emissions
separated from natural background emissions.
Synthetic nitrogenous fertilization rates are based on globally uniform relationships with current and
projected production of the major crop types by region. Current and historic nitrogenous fertilization
data for each region are from FAOSTAT (2004b) and the International Fertilizer Industry Association
(IFA) (2002). Current nitrogenous fertilization rates (kg/hectare) were assumed to be the same as 1998
levels, and projected rates were scaled from this base. Projected fertilization rates for wheat and maize
were taken from regression equations based on crop production from FAO (2000); for soybeans, an
analogous regression equation was developed using a combination of FAOSTAT (2004b) and IF A (2002)
(see Appendix Q). Regionally specific projections of crop yields from IFPRI for 2010 and 2020 are used
with these equations to derive future fertilization rates. Crop area is assumed to remain constant for the
DAYCENT simulations. Increases in N2O out to 2020 for most regions (see Table 1-1) are therefore the
result of increasing rates of fertilization based on yield projections. All baseline fertilizer applications for
all regions are assumed to be administered in one application.
Organic-matter fertilizer additions are assumed to be a function of animal numbers by region.
Historical trends in organic fertilizer use were calculated from animal numbers reported by FAOSTAT
(2004a), using IPCC default factors concerning region-specific average nitrogen excretion per animal, and
the percentage of nitrogen distributed among waste management practices (see Appendix Q for IPCC
default factors). Projections of manure nitrogen are taken from underlying activity data used for USEPA
(2006).
Nonspatial data (such as planting date and fertilizer application rates) were assigned as point values
for each region or country and were assumed to be the same within each region. Global maps of 2° * 2°
resolution for baseline N2O emissions estimated by DAYCENT for areas of wheat, maize, and soybeans
are presented in Appendix Q (under rain-fed conditions only).
V.2.1.3 Mitigation Options for Cropland N2O and Soil Carbon Emissions
Mitigation options for croplands have been identified that could decrease N2O emissions, often the
result of applying fertilizer that exceeds crop demand, while maintaining yields (e.g., Mosier et al., 2002).
Mitigation options are chosen with this goal in mind. Options are listed in Table 1-2. The soil N2O
mitigation options involve either more efficient (or simply reduced) application of nitrogen-based
fertilizers (e.g., adding nitrification inhibitors; using split fertilization; reducing baseline nitrogen
fertilization by 10, 20, or 30 percent) or adoption of no-till cultivation methods. Because the focus of this
report is on the non-CO2 greenhouse gas emissions, additional options that might increase soil carbon
(e.g., reduced fallow periods, different cropping mix) are not considered.
These mitigation options are simulated by DAYCENT and resulting crop yields (of wheat, maize, and
soybeans), and emissions effects are compared with the DAYCENT baseline, as described above. Though
all mitigation options are represented in the final MACs, DAYCENT simulates only one mitigation option
at a time, assuming that each mitigation option is implemented on all croplands in 2000 and continuously
until 2020. No mitigation options are implemented simultaneously on the same croplands, or on different
portions of the croplands, within DAYCENT.
6 This approach of isolating anthropogenic emissions from natural background emissions is also used when the
DAYCENT model is applied to estimate anthropogenic N2O emissions from agricultural soils for the Inventory of U.S.
Greenhouse Gas Emissions and Sinks, reported by the USEPA.
V-8 GLOBAL MIT IGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
Table 1-2: Cropland N20 and Soil Carbon Mitigation Options Run Through DAYCENT
Mitigation
Option
Split fertilization
Simple fertilization
reduction— 10
percent
Simple fertilization
reduction— 20
percent
Simple fertilization
reduction— 30
percent
Nitrification inhibitor
No-till
Description
Application of same amount of nitrogen fertilizer as in baseline but divided
into three smaller increments during crop uptake period to better match
nitrogen application with crop demand and reduce nitrogen availability for
leaching, nitrification, denitrification, and volatilization.
Reduction of nitrogen-based fertilizer from one-time baseline application of
10 percent.
Reduction of nitrogen-based fertilizer from one-time baseline application of
20 percent.
Reduction of nitrogen-based fertilizer from one-time baseline application of
30 percent.
Reduces conversion of ammonium to N03, which slows the immediate
availability of nitrate (nitrate is water soluble). The inhibition of nitrification
reduces nitrogen loss and increases overall plant uptake.
Conversion from conventional tillage to no till, where soils are disturbed
less and more crop residue is retained.
Greenhouse Gas
Effects
N20, some soil carbon
N20, some soil carbon
N20, some soil carbon
N20, seme soil carbon
N20, some soil carbon
Soil carbon, some N20
As in the baseline scenario, each DAYCENT mitigation simulation is run according to the relative
portions of maize, wheat, and soybean areas under either irrigated or rain-fed management.
A number of mitigation options are found to increase net greenhouse gas emissions relative to the
baseline depending on crop, management, region, and time period. These options are removed to
estimate and construct the abatement curves. The number of options that increase net emissions grows
from the 2000 to the 2010 to the 2020 period. All of these options occur on either wheat or maize
croplands, are spread over most regions of the world, and predominantly involve reducing baseline
nitrogen fertilizers. The primary reason why decreasing nitrogen fertilizer use leads to an increase in net
GHG in some regions is a decrease in soil carbon —due to lower plant growth from the fertilizer
reductions and hence less residue returning to the soil—which more than compensates for the lower N2O
emissions. A small number of Asian regions experience an increase in emissions for the split-fertilization
option, which can occur if more frequent (but smaller) fertilizer applications coincide with rainy periods;
however, the timing of the applications for this option was assumed to be uniform across regions. In
practice, farmers would time fertilizer applications based on their local weather conditions and on plant
growth stages. In addition, some of the no-till scenarios in Western Europe increase net emissions; this is
primarily because no till allows for greater soil water content and enhances denitrification to produce
N2O emissions.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-9
-------
SECTION V —AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
V.2.1.4 DAYCENT Results for Changes in Cropland N2O, Soil Carbon, and Yields
Figure 1-1 summarizes total global production of the major crops of wheat, maize, and soybeans
under baseline and mitigation scenarios, holding area constant. DAYCENT simulations were performed
for both irrigated and nonirrigated conditions. In every case, production is higher with irrigation, as
expected. All three of the options that were effective in reducing emissions (i.e., nitrification inhibitors,
split-fertilization, and conversion to no till) simultaneously increased production. The three options that
involved reduced fertilization, on the other hand, resulted in substantial reductions in production.
Figure 1-1: Global Cropland Yields for Baseline and Mitigation Options Estimated by DAYCENT, 201 Oa
to
to
(0
§ 2,100 -
Baseline and Mitigation Options
Note: This figure shows total global production of wheat, maize, and soybeans—the three major crops modeled using
DAYCENT—simulated under the baseline and mitigation options.
a Ninhib—addition of nitrification inhibitor (irrigated and nonirrigated).
Split—split fertilization, dividing fertilizer applications into three smaller increments.
Red70—reduction of nitrogen-based fertilizer to 70 percent of baseline.
RedSO—reduction of nitrogen-based fertilizer to 80 percent of baseline.
Red90—reduction of nitrogen-based fertilizer to 90 percent of baseline.
Reduced nitrogen fertilizer leads to reduced yields because plant growth rates, and hence crop yields,
are highly sensitive to nutrient supply in DAYCENT. That is, DAYCENT assumes that plant growth is
limited by nutrient availability, as well as by water and temperature. Nitrification inhibitors and split
fertilizer caused the largest increase in yields because both of these options maintain higher nitrogen
availability for plants. Nitrification inhibitors keep more nitrogen in the root zone for two reasons: less
nitrogen is lost from the soil as nitrogen gas and, because the conversion of ammonium (NH4) to NOs is
inhibited, less NO3 is leached below the root zone. Split-nitrogen application increases plant-available
nitrogen, because nitrogen supply is more synchronized with plant-nitrogen demand. Higher plant-
nitrogen uptake also reduces nitrogen losses from nitrification, denitrification, and leaching, although to a
lesser extent than the nitrification inhibitor, for the reasons discussed above.
As shown in Figure 1-2, DAYCENT simulations for corn, soy, and wheat suggest that using
nitrification inhibitors and no-till cultivation lead to the largest reduction in net greenhouse gas emissions
at the global scale. Surprisingly, reduced nitrogen fertilizer leads to net emissions similar to the baseline
scenario. The decrease in crop production associated with reduced fertilizer applications leads to reduced
V-10
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V —AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
Figure 1-2: Global Net Greenhouse Gas (N20 and Soil Carbon) Cropland Emissions Estimated by
DAYCENT under Baseline and Mitigation Scenarios3
1,000
900
cr
0)
CM
o
o
O)
Baseline Ninhib Split Red70 RedSO Red90
Baseline and Mitigation Options
NT
12000
12010 D2020
a Ninhib—addition of nitrification inhibitor.
Split—split fertilization, dividing fertilizer applications into three smaller increments.
Red70—reduction of nitrogen-based fertilizer to 70 percent of baseline.
RedSO—reduction of nitrogen-based fertilizer to 80 percent of baseline.
Red90—reduction of nitrogen-based fertilizer to 90 percent of baseline.
NT—conversion from conventional tillage to no-till.
soil inputs and, hence, reduced soil carbon. Observations show that soil carbon levels are sensitive to
changes in crop residue inputs (e.g., Peterson et al. [1998]); however, the degree to which a particular soil
responds to changes in crop residue inputs depends on many factors, such as the history of land-use
management. The soil carbon reduction offsets the reduced N2O emissions to varying degrees. The net
emissions for the three different fertilizer reduction amounts (10, 20, and 30 percent) are similar. This
suggests that, at the global scale, the amount of soil carbon lost and the amount of N2O reduced respond
roughly linearly and equally to fertilizer inputs.
An additional consideration regarding potential trade-offs between N2O emissions and soil carbon is
that N2O reductions are long lasting, whereas soil carbon accumulation is reversible through future
changes in management. The reversibility of soil carbon accumulation is not accounted for in this
analysis, because all changes in management are assumed to occur immediately and continuously
through to 2020.
The nitrification inhibitor leads to the largest reduction in net emissions because it directly decreases
emissions from nitrification. Split fertilization also leads to significant net reductions, but it is possible
that the DAYCENT simulation is underestimating the mitigation potential of this option. This is because
the three separate fertilizer applications are occurring on the same day, regardless of the timing and
amount of rainfall represented in the model. If heavy rains happen to fall a couple of weeks after the
second fertilizer application, N2O may be higher than if all of the fertilizer were applied when the crop
was planted.
No-till cultivation leads to a large reduction in net emissions primarily because of increased carbon
storage in soil and surface residue, although this option also decreases N2O by a small amount at the
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-11
-------
SECTION V — AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES AND MITIGATION SCENARIOS
global scale. However, DAYCENT does project higher N2O emissions under no till for certain locations,
which is consistent with data showing higher N2O emissions under no till, particularly in humid
environments (Smith and Conen, 2004). Because the majority of the reduced greenhouse gas emissions
from the no-till option are from carbon storage, it should be noted that this benefit is transitory, because
the capacity of soils to store carbon is finite and most soils are likely to reach equilibrium within
approximately 50 years of land management change. Also, if cultivation intensity is increased in the
future, then much of the carbon that was stored is likely to be respired and returned to the atmosphere.
However, the reduced N2O emissions associated with applying nitrification inhibitors and split fertilizer
are irreversible and likely to persist indefinitely.
To estimate the breakeven $/tCO2eq of these mitigation options, DAYCENT results for the change
(from baseline) in net emissions, yields, and fertilizer applications are used in combination with crop and
fertilizer price information, as well as assumptions about other input costs and labor changes. The change
in crop revenue is estimated with DAYCENT changes in yields and IFPRI's current and projected region-
specific baseline prices for wheat, maize, and soybeans. Likewise, IFPRI price information for change in
fertilizer costs is used. Appendix T provides details on IFPRI's commodity prices.
No capital costs are assumed for any cropland mitigation options.7 The nitrification inhibitor option
is assumed to incur an additional input cost of $20 per hectare (Scharf et al., 2005), which is scaled from
the United States to other regions. Only two of the options are assumed to incur labor changes. Split
fertilization is assumed to require an increase in labor and no till a decrease in labor.8 These percentage
changes in labor are assumed to be uniform across regions; however, data from (he Global Trade
Analysis Project (GTAP, 2005) database are used to calculate the share of output attributable to labor costs
by crop for each region. This share is used to calculate baseline labor costs per hectare from the estimated
value of output per hectare by crop and by region. Labor rates are taken from IFPRI's IMPACT model to
calculate the implied number of labor hours per hectare consistent with the labor cost per hectare, as a
validity check on the labor costs being estimated. Section V.l.3.1 provides additional information on how
these individual parameters are used to estimate costs.
V.2.2 Rice (CH4. N2O, and Soil Carbon)
V.2.2.1 Rice CH4> N2O, and Carbon Emissions Characterization
Most rice in Asia and the rest of the world is grown in flooded paddy fields (less than 10 percent of
the rice in Asia is grown in upland conditions). When fields are flooded, decomposition of organic
material gradually depletes the oxygen present in the soil and floodwater, causing anaerobic conditions
7 No-till options would require purchasing no-till equipment for direct planting. However, if this equipment is
purchased in place of equipment used for traditional tillage, there may be little incremental capital costs 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 leads to reduced equipment depreciation. Thus, no incremental capital costs for the no-till
option are assumed.
8 Split fertilization is assumed to require 14 percent more labor, assuming one additional pass over the fields, where,
for this purpose, seven passes per year are assumed in the baseline (i.e., for tilling, planting, fertilizing, applying
herbicide, applying pesticide, and harvesting, some of which may not be done on all fields but may require more
than one pass on some farms). The biophysical modeling in DAYCENT assumes a one-time fertilizer application in
the baseline and two applications with split fertilization. No till is assumed to decrease labor requirements based on a
U.S. Department of Agriculture (USDA) Agricultural Resource Management Survey (ARMS) survey, which provides
labor estimates for conventional and conservation tillage on both irrigated and rain-fed land by major crop.
V-12 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V —AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
in the soil to develop. Anaerobic decomposition of soil organic matter by methanogenic bacteria
generates CH4. Varying amounts up to 90 percent of the CH4 is oxidized by aerobic methanotrophic
bacteria in the soil (Kriiger et al., 2002; Holzapfel-Pschorn et al., 1985; Sass et al., 1990). Some of the CH4 is
also leached away as dissolved CH4 in floodwater that percolates from the field. The remaining
unoxidized CH4 is transported from the soil to the atmosphere, primarily through the rice plants
themselves. Minor amounts of CH4 also escape from the soil via diffusion and bubbling through
floodwaters.
The water management system under which rice is grown is therefore one of the most important
factors affecting CH4 emissions. The amount of available carbon susceptible to decomposition is also
critical. Some flooded fields are drained periodically during the growing season, either intentionally or
accidentally. If water is drained and soils are allowed to dry sufficiently, CFLt emissions decrease or stop
entirely. This is due to soil aeration, which not only causes existing soil CH4 to oxidize but also inhibits
further CH4 production in soils.
Field measurements in China indicate, however, that midseason drainage, while significantly
reducing CH4/ actually increases N2O emissions (Zheng et al., 1997, 2000; Cai et al., 1999). One of the key
processes controlling CH4 and N2O production/consumption in paddy soils is the reduction potential
(Eh) dynamics. Methane and N2O are produced during different stages of soil redox potential
fluctuations.
In addition to water management, other practices (e.g., tillage, fertilization, manure amendments)
will alter the soil environmental conditions (e.g., temperature, moisture, pH) and hence affect the soil
carbon- and nitrogen-driving processes such as decomposition, nitrification, and denitrification. The
changes in the soil biogeochemical processes will finally affect the availability of soil nitrogen and water
to the crops and hence alter the crop yields. Because crop residue is the major source of soil organic
carbon, the change in crop yield and litter will redefine the soil organic-matter balance, which is one of
the most important factors determining the CH4, soil CO^ and N2O emissions (Li et al., 2006).
Soil temperature is also known to be an important factor regulating the activity of methanogenic
bacteria and, therefore, the rate of CH4 production.
V.2.2.2 DNDC Baseline Estimates of Rice CH4, N2O, Soil Carbon, and Yields
The DNDC model, in particular the paddy-rice version of the model (DNDC 8.6; Li et al., 2004; Li et
al., 2002; Cai et al., 2003; Zhang et al., 2002), was used to estimate baseline and mitigation scenario
emissions of CH4, N2O, and soil carbon, as well as yield and water resource changes, for Asian rice
systems. Greenhouse gas emissions from non-Asian rice systems, which represent about 10 percent of the
world's total rice area (Wassmann et al., 2000), are excluded, primarily because data for these areas were
not available at the time of the DNDC modeling. The DNDC model and emissions baseline
methodologies are briefly described here, and further details are provided in Appendix R. Appendix S
summarizes differences between the baseline rice GHG emissions used in this analysis and USEPA
(2006).
DNDC is a soil biogeochemical model that simulates both aerobic and anaerobic soil conditions and
estimates crop yields based on a detailed crop physiology-phenology model. It is designed for assessing
the impact of different management strategies on short-term and long-term soil organic carbon dynamics
and emissions of CH4, N2O, nitric oxide (NO), and NH3 from both upland and wetland agricultural
ecosystems. DNDC requires data on soils (e.g., pH, soil carbon, bulk density, and soil texture), rice
cropping areas and systems (e.g., single rice, double rice, rice rotated with upland crops), climate, and
management practices (e.g., fertilizer use, planting and harvesting dates, tillage, water use). DNDC runs
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-13
-------
SECTION V — AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
on a daily time step and can therefore capture temporal, as well as spatial, heterogeneity in emissions
processes.
DNDC has been tested against several CH4 and N2O flux data sets for wetland rice systems in
different regions of the world, and overall results indicate that DNDC is capable of estimating the
seasonal patterns and magnitudes of CH4 and N2O fluxes. In some cases (less than 20 percent of the sites
tested), there were discrepancies between modeled and observed patterns of CH4 and N2O fluxes. In
these cases, minor modifications to capture unique local management conditions, rice varieties, and
anaerobic processes resulted in good estimates of greenhouse gas emissions from all rice systems tested
(Li et al, 2006; Cai et al, 2003; Zhang et al., 2002; see also Appendix R).
DNDC simulates rice growth and yield by tracking heat (i.e., daily temperature) accumulation, water
availability, and nitrogen availability. If there is any stress in heat, water, or nitrogen detected by DNDC,
the yield will be reduced accordingly. The impacts of farming practices on yield are modeled based on
the effects of the practices on water and nitrogen (it is assumed the practices have little effect on heat
flux).
China is the core focus of the rice component of this study because China contains roughly 20 percent
of the world's rice paddies and generates 31 percent of the world's rice production (FAOSTAT, 2004a);
furthermore, previous DNDC modeling efforts had already collected a detailed database for Chinese rice
systems at the county scale. This Chinese rice component of the analysis is described in Li et al. (2006) and
briefly summarized here.
Table 1-3 contains DNDC estimates of rice emissions for China, individual water basin regions within
China, and other Asian countries (see Section V.l.3.3 for additional baseline summary information).
Methane emissions tend to increase over time because of soil carbon accumulation. N2O emissions tend to
decline, also because of the soil carbon accumulation, coupled with an assumed constant rate of
fertilization (which increases total denitrification).
Data on rice cropping systems, soils, climate, water management, residue management, fertilizer, and
optimum yield profiles are incorporated into DNDC for each of the approximately 2,500 Chinese
counties. County data are aggregated to water basin regions within China. Maximum and minimum
values of soil texture, pH, bulk density, and soil organic, carbon content are derived for each county.
These factors are used to determine the most sensitive factors to estimate uncertainty in emissions
estimates within each county. Based on sensitivity tests (Li et al., 2004), the most sensitive factors for CH4
and N2O emissions from rice paddies are soil texture and soil organic carbon. By varying soil texture and
soil organic carbon over the ranges reported in the county-scale database, a range of CH4 and N2O
emissions for each cropping system in each county is estimated. All emissions estimates from DNDC in
this study represent the midpoints of those ranges.
There are 11 different crop rotations, including single rice, double rice, rice-winter wheat, rice-
rapeseed, and rice-rice-vegetable. The area occupied by each rotation in each county is quantified by
combining the county-scale statistical database of crop-sown areas with a Landsat land-cover map for
mainland China (Frolking et al., 2002). Total rice area is assumed to remain fixed over the 2000 to 2020
period under both baseline and mitigation scenarios, though regional changes in rice area are certainly
expected to occur; the subject of changing rice area in response to implementing the mitigation options is
discussed in Section V.I.3.6.
V-14 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V —AGRICULTURE « EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
Table 1-3: Rice-Only Baseline CH4, N20, and Soil Carbon Estimates for 2000,2010, and 2020 by Asian
Region (Midpoints from DNDC in MtC02eq/yr; Negative Carbon Numbers Indicate Net
Sequestration)
Region
China
Huaihe
Haihe
Huanghe
Changjian
Songliao
Inland
Southwest
ZhuJiang
Southeast
Bangladesh
India
Indonesia"
Philippines
Thailand
Vietnam
CH4
211
41
3
2
87
23
1
1
33
19
41
103
131
58
66
45
2000
N20
199
23
2
1
104
8
0
2
38
20
4
5
36
7
6
6
Soil
Carbon
-25
-4
-1
-1
-18
6
-0
1
-2
-6
13
19
257
36
70
40
2010"
CH4
217
43
3
2
90
24
1
1
34
20
45
111
139
60
69
73
N20
132
18
1
1
65
6
0
1
25
15
2
10
5
2
3
3
Soil
Carbon
^8
-6
-1
-0
-24
-2
-0
-1
-10
-5
-1
-9
78
7
14
7
2020"
CH4
223
44
4
2
92
24
1
1
35
20
47
117
150
64
73
80
N20
114
15
1
1
56
6
0
1
21
13
2
15
4
2
3
3
Soil
Carbon
-35
-4
-0
-0
-18
-1
-0
-0
-8
-4
-2
.-10
65
6
12
6
a Average of 2006-2010.
b Average of 2016-2020.
c Indonesia has exceptionally large baseline decreases in soil carbon because it is starting from a very high initial soil carbon content (about 7
percent).
Total emissions are estimated from total sown area, including all rice systems that capture more than
one rice crop (i.e., double rice) for multiple growing seasons over the course of a year. Rice yields in
DNDC, however, are estimated from single-rice systems only and are assumed to be representative of
other types of rice systems (i.e., double rice and rice-winter wheat).9
Daily weather data (i.e., maximum and minimum air temperatures and precipitation) for 1990 from
610 weather stations in China were acquired from the National Center for Atmospheric Research
(http://dss.ucar.edu/datasets/ds485.0/). Climate data for 1990 are used for baseline and mitigation
scenarios; thus, as with the DAYCENT modeling runs, there is no explicit assumption about
anthropogenic climate change out to 2020. Climate, biophysical, and management conditions are
assumed to be the same within each county but vary across counties.
Midseason drainage is assumed to be a baseline management practice for a fixed percentage (80
percent) of Chinese paddies currently and out to 2020; the remaining 20 percent is assumed to be under
continuous flooding. Shen et al. (1998) estimate that 80 percent of Chinese rice systems have made the
9 There are plans to modify the DNDC model so that yields for multiple types of rice cropping systems, in addition to
single-rice systems, can be tracked separately.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-15
-------
SECTION V — AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
conversion from continuous flooding to midseason drainage, a practice that decreases CH4 emissions
because it decreases the period over which anaerobic conditions occur. Under midseason drainage, three
drainage events are assumed to be carried out for each rice-growing season.
Optimal Chinese rice yields in DNDC were set to increase by 1 percent per year over the 2000 to 2020
period to match yield projections from IFPRI's IMPACT model. As a result, realized yields in DNDC do
increase over time but fall short of the prescribed optimal yields due to nutrient limitations, primarily
insufficient nitrogen availability.
Fertilizer applications are assumed to be 140 kgN/hectare (70 kgN of urea and 70 kgN of ammonium
bicarbonate) for each rice-growing season. These rates remain fixed over the 2000 to 2020 period. Rice
straw (1,000 kg-C) is also amended at the beginning of each rice-growing season. No manure is applied.
Less detailed DNDC analyses are carried out for other Asian regions. These emissions analyses are
not intended to serve as national inventory studies but rather to provide a basis from which to assess the
effects of the different mitigation options. DNDC is run for individual sites under both rain-fed and
irrigated conditions in Bangladesh, India, Indonesia, Japan, the Philippines, Thailand, and Vietnam. For
each site, soil and climate data were compiled from several sources (Global Soil Data Task Group, 2000;
Kistler et al., 2001; Webb et al., 2000). To estimate national-level emissions, the simplified assumption is
made that the site-level conditions are representative of the entire country. Therefore, net emissions (CH4,
N2O, soil carbon) rates per hectare from these test sites are multiplied by the number of hectares under
either irrigated or rain-fed conditions in each country, according to data from the International Rice
Research Institute (IRRI). These areas also remain fixed over the 2000 to 2020 period. Like the DNDC
simulations in China, optimal yield projections out to 2020 from IFPRI (see Appendix R) are used to allow
annual baseline yields in DNDC to increase at different rates in different countries.
Midseason drainage is currently not widely practiced outside of China; for this reason, the'dominant
baseline management condition assumed in these other Asian regions is continuous flooding under either
irrigated or rain-fed conditions. Fertilization types and rates are assumed to be the same as in China.
DNDC simulations were not carried out for Malaysia, Myanmar, South Korea, and other Southeast Asian
countries, but nationally averaged emissions and yield results from DNDC in neighboring regions are
used as proxies (see Appendix R).
V.2.2.3 Mitigation Options for Rice CH4, N2O, and Soil Carbon Emissions
The mitigation options chosen for rice emissions have been identified as viable options in the
literature (e.g., Wassmann et al. [2000]; Van der Gon et al. [2001]). Table 1-4 lists these mitigation options,
which include changes in water management that reduce the time over which flooding conditions occur
(to reduce anaerobic conditions), use of alternative fertilizers and changes in the timing of organic
amendments (to inhibit methanogenesis), or switching from flooded to upland rice to eliminate anaerobic
conditions.
Unlike China, most other Asian countries have larger fractions of rice areas under rain-fed
management conditions. Mitigation options requiring a change in water management are not simulated
on rain-fed areas because these systems are water limited and rely only on precipitation. Mitigation
options involving fertilizer management and conversion to upland rice are assessed on all rice areas.
All mitigation options are intended primarily to reduce baseline CH4 emissions, but N2O emissions
and soil carbon are affected as well. Emissions reductions represented in the final cost estimates represent
these net greenhouse gas effects. The mitigation options are simulated by DNDC, and resulting rice crop
yields and emissions effects are compared with the DNDC baseline, as described above. Although all
V-16 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
Table 1-4: Rice CH4, N20, and Soil Carbon Mitigation Options Run Through DNDC
Mitigation Option
Description
Greenhouse Gas Effects
Full midseason
drainage
Shallow flooding
Off-season straw
Ammonium sulfate
Slow-release
fertilizer
Upland rice
In China, shift from 80 percent to 100-percent adoption of
midseason drainage. In rest of Asia, conversion from 0 percent to
100 percent. Rice fields are dried three times within a growing
season and surface water layer is 5 to 10 cm for remaining, flooded
period. Not applied on rain-fed areas.
Assumes rice paddies are marginally covered by flood water, with
the water table fluctuating 5 to 10 cm above and below soil surface.
Not applied on rain-fed areas.
Shifting straw amendment from in-season to off-season can reduce
availability of dissolved organic carbon and; thus, methanogens.
Assumes rice straw is applied 2 months before rather than at
beginning of rice-growing season.
Baseline fertilizers, urea, and ammonium bicarbonate, replaced
with 140 kg/hectare of ammonium sulfate. Sulfate additions to soil
can elevate reduction potential, which suppresses CH4 production.
Nitrogen is slowly released from coated or tablet fertilizer over a
30-day period following application. Applied in the same amount
and at the same time as in baseline case. Increases fertilizer-use
efficiency.
Assumes upland rice replaces existing paddy rice areas and that
fields do not receive any flood water.
CH4, N20, soil carbon
Same
Same
Same
Same
Same
mitigation options are represented in the final MACs, DNDC simulates only one mitigation option at a
time, assuming that each mitigation option is implemented on all rice lands in 2000 and continuously
until 2020. No mitigation options are implemented simultaneously on the same rice lands or on different
portions of the rice lands within DNDC.
Unlike the options with DAYCENT, no options that were found to increase net emissions relative to
baseline are removed from the rice portion of the analysis, because these net emissions increases were
generally small or temporary (i.e., occurring only in the later years of the analysis).
V.2.2.4 DNDC Estimates for Changes in Rice CH4, N2O, Soil Carbon, and Yields
Results here provide the most detail for China because that is the country for which the most detailed
DNDC modeling runs were carried out. Table 1-5 provides net greenhouse gas results aggregated to the
Chinese national level for the baseline and mitigation scenarios, averaged over the entire 2000 to 2020
period. The midpoint estimates from DNDC are those carried forward in the MAC calculations.
Table 1-5: DNDC Estimates of Net Greenhouse Gas Results for Baseline and Mitigation Scenarios for China
(Annual Averages in MtC02eq/yr over 2000-2020)
Estimate
Midpoint*
High estimate
Low estimate
Baseline
315
484
146
Midseason
Drainage
296
445
148
Shallow
Flooding
140
232
47
Off-Season
Straw
298
468
128
Ammonium
Sulfate
235
379
90
Slow-Release
Fertilizer
326
454
199
Upland
Rice
41
71
11
Source: Li et al., 2006.
a The high, rmd, and low estimates are the results of most sensitive factor (MSF) estimates carried out with the DNDC model.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-17
-------
SECTION V —AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
Table 1-6 shows individual greenhouse gas changes from the baseline, as well as the net greenhouse
gas and yield changes, at the Chinese national level on a per-hectare basis; change in water-use
requirements are also shown but are not used in the final cost estimates because water is not a priced
commodity in these rice systems.
Table 1-6: Changes from Baseline in Greenhouse Gas Emissions, Crop Yields, and Water Consumption for
China (Annual Averages over 2000-2020; Negative Numbers Indicate Decreases Relative to the
Baseline)
Management Option
Midseason drainage
Shallow flooding
Off-season straw
Ammonium sulfate
Slow-release fertilizer
Upland rice
CH4
(kgC02eq/ha)
-2,411
-7,402
-663
-367
287
-11,794
N20
(kgCOjeq/ha)
1,283
-2,440
^40
-3,668
727
-3,018
C02
(kgCCVha)
-1
591
21
-85
-191
239
Net Greenhouse Gas
(kgCOzeq/ha)
-1,129
-9,251
-682
-4,120
823
-14,573
Yield
(kgC/ha)
81
134
43
28
131
-381
Water
(mm/yr)
-9
-248
0
0
0
-566
Source: Li et al., 2006.
As described in Li et al. (2006), despite large-scale adoption of midseason drainage, there is still large
technical potential for additional CH4 reductions from Chinese rice paddies (e.g., over 60 percent
reductions are achieved in the shallow flooding scenario over 2000 to 2020). However, management
changes that reduce CH4 emissions simultaneously affect N2O emissions and soil carbon dynamics such
that the net greenhouse gas effects should be considered. Midseason drainage, for example, is an effective
CH4 reduction strategy but can significantly increase N2O emissions. Ammonium sulfate reduces CH4 by
a small amount but significantly reduces N2O; these low CH4 reductions are largely due to the fact that
mid-season drainage rather than continuous flooding is the baseline practice (conditions under which
sulfate is less effective at reducing CH4), whereas more significant N2O reductions occur because
ammonium sulfate is less susceptible to volatilization than the urea it is replacing (because all of the
nitrogen is already in the ammonium form).
In terms of net greenhouse gas technical mitigation potential only, the most effective mitigation
option appears to be shallow flooding, followed by ammonium sulfate, full midseason drainage
adoption, and off-season straw amendments; the slow-release fertilizer scenario enhances soil carbon but
increases the other gases and thus does not reduce net greenhouse gas emissions compared with the
baseline. The upland rice scenario, where it is assumed that existing rice fields receive no flood water, is
simulated in DNDC for China and is found to decrease net greenhouse gas emissions by about 87
percent.
The relative order of mitigation across scenarios remains the same even when the proportions of
midseason drainage vary (Li and Salas, 2005), suggesting that these results may apply to other regions
where midseason drainage has not been widely adopted. Appendix R contains information about the
time dynamics of these net greenhouse gas changes for each scenario for 2000 to 2020.
Most mitigation options, including slow-release fertilizer, increase rice yields compared with the
baseline. In general, rice yields vary directly with nitrogen availability, assuming no heat stress and
sufficient water resources: higher nitrogen availability leads to higher yields. Relative to continuous
flooding, midseason drainage or shallow flooding elevates soil aeration and hence accelerates
decomposition, which produces more inorganic nitrogen and increases nitrogen availability. Slow-release
V-18
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
fertilizer improves the fertilizer use efficiency by extending the period of nitrogen availability for the
plants, effectively increasing total nitrogen availability. Similar to the slow-release fertilizer, ammonium
sulfate has relatively low solubility compared with baseline fertilizer, urea, and ammonium bicarbonate
and thus is less susceptible to leaching. Although theoretically off-season straw may not have a direct
effect on yield, it is assumed that early incorporation of straw favored its decomposition because of the
high reduction potential conditions before flooding. This higher decomposition rate enhanced yields
caused by increased soil nitrogen availability when the rice was transplanted. The yield difference
between upland rice and paddy rice is in the rice's genetic characteristics. The current upland rice has
genetically low yield. This situation may change if new strains of upland rice are developed. In summary,
management practices that increase nitrogen availability (through increased decomposition or better
synchronization with plant needs) will typically increase rice yields.
Shifting to full midseason drainage and shallow flooding are also water-saving practices because they
significantly decrease water consumption (i.e., evapotranspiration), whereas the other mitigation options,
involving only changes in fertilization or straw amendment, have almost no effect on water consumption.
Table 1-7 shows net greenhouse gas results under each mitigation option compared with the baseline
for the other Asian countries. The pattern of results observed in China is similar for these other countries.
The most effective mitigation options in terms of net greenhouse gas reductions involve a change in
water management; the options involving a change in fertilization management are less effective. Slow-
release fertilizer is also a particularly poor greenhouse gas reduction strategy in these other Asian
countries, often leading to no net greenhouse gas reductions compared with the baseline.
Table 1-7: Net Greenhouse Gas Results for Baseline and Mitigation Options for Other Asian Countries
(Annual Averages in MtC02eq/yr over 2000-2020)
Country
Bangladesh
India
Indonesia
Japan
Philippines
Thailand
Vietnam
Baseline
47
113
237
29
72
91
84
Midseason
Drainage
23
60
139
15
40
79
71
Shallow
Flooding
8
23
142
6
16
74
59
Off-Season
Straw
32
79
193
22
54
65
51
Ammonium
Sulfate
43
101
223
26
65
85
78
Slow-
Pft In etc A
Release
Fertilizer
47
.115
238
29
73
91
84
Upland
Rice
21
41
190
10
26
72
43
To estimate the breakeven $/tCO2eq of these mitigation options, DNDC results for the change from
baseline in net emissions, yields, and fertilizer applications are used in combination with rice crop and
fertilizer price information, as well as assumptions about other input costs and labor changes. Change in
crop revenue is estimated with DNDC changes in yields and IFPRI's changes in current and projected
region-specific baseline producer prices for rice (see Appendix T). For the ammonium sulfate option, the
additional input cost is the extra cost of ammonium sulfate compared with urea and ammonium
bicarbonate, based on FAO prices. For the slow-release fertilizer option, the additional input cost is
assumed to be $20 per hectare for all regions, based on the cost of using Agrotain, a urease inhibitor
thought to be an appropriate proxy.
No one-time capital costs are assumed for any of the rice mitigation options. Three of the options (i.e.,
midseason drainage, shallow flooding under irrigated conditions, and off-season straw amendments) are
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-19
-------
SECTION V —AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
assumed to require additional labor compared with baseline practices. The percentage increase in labor
for these three options is assumed to be uniform across all regions and is estimated by assuming
percentage changes in preharvest labor based on data from Moser and Barrett (2002) for systems of rice
intensification as a rough proxy. Preharvest labor is assumed to account for 75 percent of total labor in all
regions. GTAP data on value of inputs used in rice production are used to calculate the share of output
value attributable to rice labor costs for each region. This share is used to calculate baseline labor costs per
hectare from the estimated value of output per hectare by region. Labor rates are taken from IFPRI's
IMPACT model to calculate the implied number of labor hours per hectare, consistent with the labor cost
per hectare, as a validity check on the labor costs being estimated. Section V.I.3.1 provides additional
information on how these individual parameters are used to estimate costs.
V.2.3 Livestock (CH4 and N2O)
V.2.3.1 Livestock Enteric CH4 Emissions Characterization
Methane is produced as part of the normal digestive process in animals. During digestion, microbes
present in an animal's digestive system ferment food consumed by the animal, This microbial
fermentation process is referred to as enteric fermentation and produces CH4 as a by-product, which can
be exhaled or eructated by the animal. The amount of CH4 produced and excreted by art animal depends
primarily on the animal's digestive system and the amount and type of feed it consumes.
Ruminant animals (e.g., cattle, buffalo, sheep, goats, and camels) are the major emitters of CH4
because of their unique digestive systems. Ruminants possess a rumen, or large fore-stomach, in which
microbial fermentation breaks down coarse plant material for digestion. Nonruminant domesticated
animals (e.g., swine, horses, mules) also produce CH4 emissions through enteric fermentation, although
this microbial fermentation occurs in the large intestine, where the capacity to produce CH4 is lower
(USEPA, 2005b).
An animal's feed quality and feed intake also affect CH4 emissions. In general, lower feed quality or
higher feed intake lead to higher CH4 emissions. Feed intake is positively related to animal size, growth
rate, and production (e.g., milk production, wool growth, pregnancy, or work). Therefore, feed intake
varies among animal types, as well as among different management practices for individual animal
types.
Because CH4 emissions represent an economic loss to the farmer—where feed is converted to CH4
rather than to product output—viable mitigation options can entail feed efficiency improvements to
reduce CH4 emissions per unit of beef or milk. However, these mitigation options can actually increase
CH4 per animal.
V.2.3.2 Livestock Manure CH4 and N2O Emissions Characterization
The management of livestock manure can produce both CH4 and N2O emissions. Methane is
produced by the anaerobic decomposition of manure. Nitrous oxide is produced through the nitrification
and denitrification of the inorganic nitrogen derived from livestock manure and urine.
When livestock and poultry manure is stored or treated in systems that promote anaerobic conditions
(e.g., as a liquid or slurry in lagoons, ponds, tanks, or pits), the decomposition of materials in the manure
tends to produce CH4. When manure is handled as a solid (e.g., in stacks or pits) or deposited on pasture,
range, or paddock lands, it tends to decompose aerobically and produce little or no CH4 (USEPA, 2005b).
Ambient temperature and manure storage or residency time also significantly affects the amount of
CH4 produced because of influences on the growth of the bacteria responsible for CH4 formation. For
V-20 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V —AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
example, CH4 production generally increases with rising temperature and residency storage time
(USEPA, 2005b). Also, for nonliquid-based manure systems, moist conditions (which are a function of
rainfall and humidity) favor CH4 production. Although the majority of manure is handled as a solid,
producing little CH4, the general trend in manure management, particularly for large dairy and swine
producers in the United States and other industrialized countries, is one of increasing use of liquid
systems.
The composition of the manure also affects the amount of CH4 produced. Manure composition varies
by animal type and diet. In general, the greater the energy content of the feed, the greater the potential for
CH4 emissions. For example, feedlot cattle fed a high-energy grain diet generate manure with a high CH4-
producing capacity, whereas range cattle fed a low-energy diet of forage material produce manure with
about half the CH4-producing potential (USEPA, 2005b). However, some higher-energy feeds also are
more digestible than lower quality forages, which can result in less overall waste excreted from the
animal.
A small portion of the total nitrogen excreted in manure and urine is expected to convert to N2O. The
production of N2O from livestock manure depends on the composition of the manure and urine, the type
of bacteria involved in the process, and the amount of oxygen and liquid in the manure system (USEPA,
2005b). For N2O emissions to occur, the manure must first be handled aerobically where NH3 or organic
nitrogen is converted to nitrates and nitrites (nitrification) and then handled anaerobically, where the
nitrates and nitrites are reduced to nitrogen gas (N2), with intermediate production of N2O (i.e.,
denitrification) (Groffman et al., 2000). These emissions are most likely to occur in dry manure handling
systems that have aerobic conditions but that also contain pockets of anaerobic conditions, such as rain
events.
V.2.3.3 The USEPA Baseline Estimates of Livestock Enteric CH4 Emissions
Baseline emissions of and activity data for livestock enteric CH4 are taken directly from USEPA
(2006). Enteric CH4 emissions from livestock are estimated to be the second largest source of global
agricultural non-CO2. In 2000, global enteric CH4 emissions were estimated to be 85,648 Gg or 1,799
MtCO2eq and are projected to increase more than 30 percent by 2020 to 111,633 Gg or 2,344 MtCO2eq (a
32 percent increase relative to 1990). Livestock enteric CH4 accounted for 32 percent of global agricultural
non-CO2 emissions in 2000. In the United States, enteric CH4 accounts for 27 percent of agricultural non-
CO2 and less than 2 percent of all greenhouse gas emissions (USEPA, 2005b).
V.2.3.4 The USEPA Baseline Estimates of Livestock Manure CH4 and N2O Emissions
Baseline emissions of and activity data for livestock manure CH4 and N2O are taken directly from
USEPA (2006). The joint CH4 and N2O emissions from livestock manure are estimated to be the fourth
largest source of global agricultural non-CO2 emissions. In 2000, livestock manure emissions were
estimated to be 421 MtCO2eq, or 10,732 Gg of CH4 and 632 Gg of N2O, and are projected to increase 24
percent by 2020 to 523 MtCO2eq, or 12,832 Gg of CH4 and 818 Gg of N2O (a 21 percent increase relative to
1990). Livestock manure emissions accounted for less than 8 percent of global agricultural non-CO2
emissions in 2000. In the United States, joint CH4 and N2O emissions from livestock manure account for
13 percent of agricultural non-CO2 emissions and less than 1 percent of all greenhouse gas emissions
(USEPA, 2005b).
V.2.3.5 Mitigation Options for Livestock Emissions
Non-CO2 greenhouse gas emissions from livestock can be reduced primarily through either reducing
CH4 emissions that occur during the normal digestive process (i.e., enteric fermentation) or capturing
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-21
-------
SECTION V — AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES AND MITIGATION SCENARIOS
CH4 emitted by livestock manure. There is also potential to affect N2O emissions from manure
management, either indirectly through options that target CH4, but also through mitigation options that
primarily target N2O. However, limited quantitative information is available on the co-effects of CH4
mitigation options on N2O or the cost or emissions reductions associated with options focused on N2O.
Thus, the focus here is on options designed to reduce CH4 but to account for changes in N2O emissions
for those options that change livestock populations by assuming a change in N2O emissions
proportionate to the change in livestock population.
V.2.3.5,1 Mitigation Options for Livestock Enteric CH4
The enteric CH4 mitigation options fall into four general categories: (1) improvements to food
conversion efficiency by increasing energy content and digestibility of feed, (2) increased animal
productivity through the use of natural or synthetic compounds that enhance animal growth and/or
lactation (e.g., bovine somatotropin [bST], antibiotics), (3) feed supplementation to combat nutrient
deficiencies that prevent animals from optimally using the potential energy available in their feed, and (4)
changes in herd management (e.g., use of intensive grazing).10 Some of these proposed options for enteric
fermentation may actually increase net greenhouse gas emissions per animal but lead to an even larger
increase in productivity. Thus, emissions per unit of product (e.g., meat, milk, or work) decline, and
mitigation at the national or regional level requires a sufficient reduction in the number of animals to
more than offset the increase in emissions per animal. To capture this issue, two separate estimates of
mitigation potential and costs were developed assuming both a constant number of animals and constant
production. The static, engineering approach does not allow for simultaneous adjustment in both number
of animals and production; however, sensitivity experiments at the end of this section using the global
agricultural commodity market model, IMPACT, allow for these dynamic feedbacks to occur.
Table 1-8 summarizes the enteric fermentation options. Most of these options could also be applied to
other livestock species (e.g., buffalo, sheep, goats), but no data were available on the emissions reductions
or productivity effects that would be expected for those species.
V.2.3.5.2 Mitigation Options for Livestock Manure CH4 and N2O
All manure CH4 mitigation options involve the capture and use of CH4 through anaerobic digesters.
Anaerobic digesters are currently in limited use on large-scale livestock operations in developed regions,
often primarily as a means of treating and stabilizing waste and controlling odor, but the CH4 that is
captured is also used as an energy source. Small-scale, ambient temperature digesters are also being used
in developing regions, such as China, India, and Vietnam, for household energy generation. The
feasibility of digesters depends in part on climate. There are a large number of different types of digesters
that can be used, with some being more appropriate for certain climates or livestock species than others.
Another important characteristic of digester systems is whether they include engines for electricity
generation. Systems generating electricity can potentially create savings by offsetting farm purchases of
electricity or even selling the electricity. Systems that do not include electricity generation generally use
the heat generated for on-farm use to offset purchases of heating fuels.
10 Emissions can also be mitigated through other methods, such as improving genetic characteristics, feeding
compounds that inhibit rumen methane formation, improving reproduction efficiency, and controlling disease better.
However, data are currently insufficient to include estimates for these options.
V-22 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
Table 1-8: Livestock Enteric Fermentation Greenhouse Gas Mitigation Options
Mitigation Option
Description
Greenhouse Gas
Effects8
Improved feed
conversion
Antibiotics
Bovine somatotropin
(bST)
Propionate precursors
Antimethanogen
Intensive grazing
Increase the amount of grain fed to livestock to increase the CH4, some N20
proportion of feed energy being converted to milk, meat, or work
instead of animal maintenance. This option tends to increase
emissions per animal but reduce emissions per unit output. It is
more effective in reducing emissions per unit of production in
regions where baseline feed is of relatively low quality. This option
is applied to both beef and dairy cattle in all regions, although it
was excluded from the MACs for some developed regions where it
resulted in slightly higher GHG emissions.
Administer antibiotics (e.g., monensin) to beef cattle to promote CH4, some N20
faster weight gain, which reduces time to maturity and CH4
production per kilogram of weight gain. This option is applied in all
regions.
Administer bST to dairy cattle to increase milk production. In many CH4, some N20
cases, this option increases CH4 emissions per animal but
typically increases milk production sufficiently to lower emissions
per kilogram of milk. Because of opposition to the use of bST in
many countries, this option was only applied in selected countries
that currently approve of the use of bST or are likely to approve its
use by 2010.
Involves administering propionate precursors to animals on a daily CH4, some N20
basis. Hydrogen produced in the rumen through fermentation can
react to produce either CH4 or propionate. By adding propionate
precursors to animal feed, more hydrogen is used to produce
propionate and less CH4 is produced. This option is applied to
both beef and dairy cattle in all regions.
Vaccine in development by Commonwealth Scientific and CH4, some N20
Industrial Research Organization (CSIRO) that can be
administered to animals and will suppress CH4 production in the
rumen. This option is applied to beef and dairy cattle, sheep, and
goats in all regions.
Moving to a more management-intensive grazing system where CH4, some N20
cattle are frequently rotated between pastures to allow recently
grazed pastures time to regrow and to provide cattle with more
nutritious pasture grazing that will permit replacement of more
feed grains. This option may actually reduce animal yields but will
decrease emissions by an even larger percentage. This option is
applied to beef and dairy cattle in developed regions and Latin
America.
For this analysis, effects on N20 are estimated only for the scenarios where production is held constant and there is a change in livestock
population.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-23
-------
SECTION V —AGRICULTURE « EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
The types of digesters are aggregated based on a categorization of representative systems provided
by the USEPA's AgStar program (Table 1-9). For most regions, swine and diary cattle account for the
majority of greenhouse gas emissions from manure, largely because their manure is often managed in
liquid systems under anaerobic conditions. Although CH4 emissions from manure could potentially be
captured from additional species (e.g., beef cattle, buffalo,, sheep, goats), these species typically account
for much smaller shares of emissions and are often managed on pasture much of the year with solid
manure handling. Manure from livestock on pasture does not produce much CH4 because it decomposes
under aerobic conditions, resulting in little to no emissions. Based on IPCC default factors, the CH4
emissions factor is actually higher for digesters than for dry manure management systems such as
pasture.
Complete-mix, plug-flow, fixed-film, and large-scale covered lagoon digesters are applied only in the
United States, EU-15, Japan, Australia, other developed countries, Eastern Europe, Central Asia, FSU,
China, South Korea, and other East Asian regions, based on climate, environmental regulations, capital
costs, and other considerations. The dome, polyethylene bag, small-scale covered lagoon, and flexible-bag
digesters are applied in all other world regions. China and other parts of East Asia are the only regions
where digesters in both of these groups are applied. Applicability of these options was further refined by
allocating the share of baseline emissions to swine and dairy cattle and applying emissions reductions
only to those portions of the emissions stream. The share of livestock manure emissions due to dairy and
swine is based on USEPA (2006), which relies on both IPCC inventory default methodologies and
individual country greenhouse gas inventory reports.
CH4 reduction efficiencies are assumed to be 85 percent from baseline for the complete-mix, plug-
flow, fixed-film, and large-scale covered lagoon digesters based on the difference between IPCC default
emissions factors for anaerobic manure management, where CH4 is released into the atmosphere, and
digesters. For the smaller-scale digesters applied in developing countries, the reduction efficiency is
assumed to be 50 percent from baseline, where baseline emissions are much lower because of a different
distribution of manure management practices and the likelihood of less efficient CH4 capture.
Capital costs are taken from the USEPA's AgStar program (Roos, personal communication, 2005),
which estimates the capital cost per 1,000 pounds of liveweight. These values are combined with the 1996
IPCC guideline values for average liveweight for different species in regions around the world to
generate estimates of the capital cost per animal. Because liveweight per animal tends lo be much smaller
in developing countries, the capital cost per animal ends up being lower than in developed regions. This
cost is annualized assuming that the large-scale digesters have an expected useful lifetime of 20 years and
the small-scale digesters have an expected useful lifetime of 10 years.
GTAP data on labor cost shares by region for livestock production and IMPACT data on regional
agricultural wage rates are used to calculate the baseline labor hours per animal and change in hours to
verify the reasonableness of these assumptions, as described above for cropland soil management and
rice cultivation. For large-scale digesters, labor requirements for swine farms are assumed to increase by 2
percent for options without engines and 4 percent for those with engines. For dairy farms, labor
requirements are assumed to increase by 0.5 percent for options without engines and 1 percent for those
with engines. The percentage increase in labor is smaller, because dairy farming is already much more
labor intensive and requires much more labor per animal in the baseline. The increase in labor for dairy
farms is calculated by assuming 200 hours per year in the United States for digester operation, repairs,
management, and typical farm size of 800 cows with 50 hours of labor per head per year in the baseline.
For hog farms, it is again assumed that a digester will add about 200 hours of labor per year, but
assuming an average of about 5,000 hogs per farm per year, that assumption could potentially add a
V-24 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
Table 1-9: Livestock Manure Management Greenhouse Gas Mitigation Options
Mitigation Option
Description
Greenhouse Gas
Effects3
Complete-mix
digester
Plug-flow digester
Fixed-film digester
Covered lagoon
digester, large-scale
light
Polyethylene bag
digester, cooking
fuel and light
Covered lagoon,
small-scale, for
cooking fuel, light,
fuel and light
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 percent and 10 percent.
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 constant gas flow.
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 percent to 13 percent. As with
complete-mix digesters, they are maintained at constant temperatures
throughout the year to maintain consistent gas production.
This digester option may be appropriate where concentrations of
solids are very low, such as in 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.
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
percent) 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 tie 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.
These are small-scale, unheated digesters used in some developing
nations, 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.
This small-scale unheated digester is in use in a variety of developing
countries. The digester essentially consists of a hole dug in the
ground and covered with a plastic bag, with an area for input of
manure and a pipe with a valve for biogas produced. Biogas
generated is typically used by the household for cooking and other
household energy needs.
This is smaller-scale and much cheaper version of the covered lagoon
above, used to generate biogas for household use. Some of these
digesters may produce enough energy for shaft power, in addition to
household cooking and other uses.
In developing countries where the biogas is generated and collected
within a plastic bag.
CH4
CH4
CH4
CH4
CH4
CH4
CH4
CH«
a Unlike options for reducing emissions from enteric fermentation, none of the options included for manure management are expected to affect
yields and there are no effects on N20 currently being estimated.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-25
-------
SECTION V —AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
digester and about 2 hours of labor per hog annually in the baseline. Options with engines and
accompanying electricity generation and transfer equipment were assumed to require twice as much
labor as those that produce heat only. For the small-scale digesters in developing nations, labor was
assumed to increase by 2.5 percent for options without engines and 5 percent for those with engines that
rely on manure from either swine or dairy cattle. Many of these digesters are used on operations with
only a few animals and often combine manure from multiple species with other household wastes.
None of these manure management options are expected to change livestock yields. Revenues (or
cost savings) are generated from using captured CH4 (essentially natural gas) for either heat or electricity
on the farm. Revenues are scaled for other non-U.S. regions based on a U.S. Energy Information Agency
(USEIA) electricity price index (2003).
Breakeven prices ($/tCO2eq) for these mitigation options are calculated by region and by species
using the emissions reductions from baseline, annualized capital costs, changes in labor costs, and energy
savings or revenue. Section V.l.3.1 provides additional information on how these individual parameters
are used to estimate costs.
V.2.3.6 Changes in Livestock CH4 and Productivity
Figure 1-3 shows total livestock emissions (enteric fermentation and manure management) associated
with the baseline and mitigation scenarios, assuming a constant number of animals, where each option is
assumed to be applied to 100 percent of the appropriate livestock species (see Tables 1-8 and 1-9), and
appropriate regions. Similarly, Figure 1-4 shows the relative emissions under baseline and the mitigation
scenarios, assuming constant production. Because percentage emissions reductions are assumed to be the
same across all large-scale digesters and across all small-scale digesters, the individual manure
management mitigation options identified in Table 1-9 are aggregated here.
Figure 1-3: Global Net Greenhouse Gas (CH4 and N20) Livestock Emissions under Baseline and
Mitigation Scenarios, Assuming Full Adoption of Individual Options and Holding Number of
Animals Constant
3,500
2000
2010
Year
2020
I Baseline
IbST
I Intensive grazing
• Improved feed conversion D Antibiotics
• Propionate precursors DAntimethanogen
• Large-scale digesters D Small-scale digesters
V-26
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
Figure 1-4: Global Net Greenhouse Gas (CH4 and N20) Livestock Emissions under Baseline and
Mitigation Scenarios, Assuming Full Adoption of Individual Options and Holding Production
Constant
3,500
2000
2010
Year
2020
I Baseline
IbST
I Intensive grazing
• Improved feed conversion D Antibiotics
• Propionate precursors D Antimethanogen
• Large-scale digesters D Small-scale digesters
Because some of the enteric fermentation options increase yields in many regions, holding production
of meat and milk constant means that there will be fewer animals needed for production. Thus, emissions
tend to fall more under the constant production assumption than the constant animals assumption. For
instance, improved feed conversion has such large positive yield impacts that emissions under full
adoption of that option are about 250 MtCO2eq less annually when holding production constant and
reducing the number of livestock, than when the number of animals is held constant. Section V.I.3.6
presents results that reflect adjustments for market impacts where there are simultaneous changes in the
number of animals and in production. In practice, a combination of mitigation options would be adopted
and no single option would be adopted for all production. In Section V.1.3.4, MACs are presented that
assume partial adoption of each of the available options in each region. However, Figures 1-3 and 1-4 are
included to give a sense of the relative emissions that would occur under different production scenarios.
Holding the number of animals constant, emissions are lowest under intensive grazing, followed by
propionate precursors, antimethanogen, large-scale digesters, improved feed conversion, small-scale
digesters, antibiotics, and bST. However, when production is held constant, the order differs, with
improved feed conversion followed by propionate precursors, antimethanogen, intensive grazing, large-
scale digesters, antibiotics, small-scale digesters, and bST.
As mentioned above, an important component of these options that will influence net emissions is the
change in yield corresponding to each of these options. For some options (e.g., improved feed
conversion), the yield effects vary substantially across regions because of widely differing baseline
conditions. To summarize the primary overall yield effects, Figures 1-5 and 1-6 show the difference in
global production of beef and milk from dairy cattle under full adoption of each of these options, holding
the number of animals constant.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE-GASES
V-27
-------
SECTION V —AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
Figure 1-5:
Global Beef Production under Baseline and Mitigation Options, Assuming Full Adoption of
Individual Options and Holding the Number of Animals Constant
100
2000
2010
Year
2020
I Baseline
I Propionate precursors
I Large-scale digesters
ffl Improved feed conversion D Antibiotics
n Antimethanogen D Intensive grazing
D Small-scale digesters
Figure 1-6:
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
900
o
2000
2010
Year
2020
I Baseline
I Propionate precursors
I Large-scale digesters
• Improved feed conversion D bST
HI Antimethanogen O Intensive grazing
D Small-scale digesters
V-28
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
For both beef and milk from dairy cattle, the largest yield increases are for improved feed conversion
in 2000. Yield increases relative to baseline are smaller in future years, largely because productivity is
improving rapidly under the baseline.11 Improved feed conversion remains the option with the largest
average global yield improvement for beef production in all years. Antibiotics, propionate precursors,
and antimethanogen all increase beef yield by a similar amount, while intensive grazing, large-scale
digesters, and small-scale digesters are assumed to have no impact on beef yield. For milk yield,
propionate precursors and antimethanogen lead to similar increases in yield, which become greater than
improved feed conversion by 2010. The use of bST leads to a small increase in yield, which would be
larger were it not for assumptions that many regions will not adopt this option. Large-scale and small-
scale digesters are assumed to have no impact on milk yield. Intensive grazing has a negative effect on
milk yield.
11 The percentage increase in yield attributable to the mitigation option for improved feed conversion is calculated
for each year by netting out the percentage increase in baseline yield projected by IMPACT. This is done to reflect
improved practices expected to be adopted in the baseline and avoid double-counting improvements in yield.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-29
-------
SECTION V — AGRICULTURE • EMISSIONS CHARACTERIZATION, BASELINES, AND MITIGATION SCENARIOS
V-30 GLOBAL MITIGATION OF MON-C02 GREENHOUSE GASES
-------
SECTION V —AGRICULTURE • RESULTS
V.3 Results
V.3.1 Estimating Average Costs and Constructing Abatement Curves
The methods used to estimate the average cost of each mitigation option and construct the MACs in
the agricultural sector follow the general methodology described in Section I. This section shows
additional baseline emissions data (as described in Section V.I.2), the average costs of the mitigation
options for key countries, and the MACs for key countries and world totals.
The average cost for each mitigation option represents the present-value breakeven price, expressed
as 2000$ t/CO2eq, where total benefits equal total costs. The $/tCO2eq is estimated according to
(PxER)(l-TR)+R(l-TR)
(l + DR)'
RC(\-TR)
(l + DR)' _
(1.1)
where
P = The breakeven price of the option in dollars per metric ton of CO2 equivalent ($/tCO2eq).
ER = The emissions reduction achieved by the technology (MtCO2eq).
R = The revenue generated from energy production (scaled based on regional energy prices) or
change in agricultural commodity prices ($).
T = The option lifetime (years).
DR = The selected discount rate (10%).
CC = The one-time capital cost of the option ($).
RC = The recurring (operation and maintenance [O&M]) cost of the option (portions of which may
be scaled based on regional labor costs) ($/year).
TR = The tax rate (40%).
TB = The tax break equal to the capital cost divided by the option lifetime, multiplied by the tax rate
($)-
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 as follows:
CC RC R CC TR
(1.2)
(l-TR)ER^-
I
ER ER ER-T (1-77?)
The cost estimate takes into account greenhouse gas reductions, revenue effects (e.g., positive or
negative changes in yield), any required capital costs (e.g., anaerobic digesters), labor requirements, and
changes in other input costs (e.g., increase or decrease in fertilizer applications), all relative to baseline
conditions. Section V.1.2 above describes the individual mitigation options; the methods for estimating
their associated effects on greenhouse gas emissions and yields; and the assumptions used for other
input, capital, and labor costs.
MACs showing greenhouse gas reductions, in terms of percentage reductions from the baseline in
2010 and 2020, are estimated for key regions and world totals. Emphasis is placed on percentage
reductions from the baseline, rather than on absolute emissions reduction numbers, because the overall
GLOBAL MITIGATION OF MON-C02 GREENHOUSE GASES
V-31
-------
SECTION V — AGRICULTURE « RESULTS
trends are most important to convey for this analysis. Furthermore, the baselines for croplands and rice
cultivation are not comprehensive greenhouse gas inventory estimates so that the percentage changes
from these baselines are viewed as more transferable to other kinds of analyses.
To construct the MAC, lines are essentially drawn to connect the points representing the average cost
of each mitigation option (where the X axis is MtCeq mitigated and the Y axis is S/tCC^eq). The following
general factors are estimated to ensure the MAC can represent simultaneous adoption of all mitigation
options. First, a technical applicability fraction is estimated to ensure the mitigation option is applied only
to the correct portion of baseline emissions. For example, options to reduce dairy cattle emissions can
only be applied to the fraction of livestock emissions attributable to dairy cattle. Second, an implied
adoption rate is estimated by segmenting the applicable baseline emissions into uniform fractions based
on the number of mitigation options. For example, if 10 mitigation options could technically be applied to
reduce cropland N2O emissions, then each mitigation option is assumed to apply to only 10 percent of
baseline emissions. This is a simplistic method to avoid double counting among mitigation options, but
unfortunately it does not allow lower-cost options to out-compete higher-cost options. Other factors in
addition to cost (e.g., adoption feasibility and implementation barriers) can of course determine the extent
to which one mitigation option is adopted over another. E5ecause such factors are not included, this static
approach of allowing each mitigation option to be applied equally is viewed as a conservative approach
to estimate the technical mitigation potential.
All mitigation options are assumed to be implemented immediately (i.e., in the first data year, 2000),
but only for appropriate regions, and are assumed to remain in place continuously until 2020. Therefore,
the MACs presented in 2010 represent the emissions reductions and associated costs that occur in year
2010, assuming that all mitigation options have been implemented since 2000. The emissions reductions
represented in 2010 are estimated relative to the 2010 emissions baseline under the assumption that no
mitigation options have been implemented since 2000.
Two general approaches are used to calculate all MACs in the agricultural sector. The first approach
keeps cropland area, rice area, and livestock populations constant over time, allowing total production to
change as yields per hectare and productivity per animal change as a result of the mitigation options. The
biophysical modeling in DAYCENT and DNDC also holds land area constant over time. The second
approach holds crop production, rice production, and livestock production (e.g., production of milk and
beef) constant over time, allowing land area and animal populations to change (postprocess). In this case,
land area is changed for each region by scaling the revised per-hectare yield numbers to maintain the
same regional crop or rice production as in the baseline. Livestock population numbers are changed in a
similar way. This latter approach is particularly important for the livestock sector, because many
proposed enteric fermentation mitigation options actually increase CH4 emissions per animal but decrease
CH4 emissions per unit product. Results are shown for both approaches.
V.3.2 Baselines, Mitigation Costs and MACs for Croplands
Table 1-10 presents the baseline net GHG emissions (N2O and soil carbon) from croplands
management by region by year used in this analysis. These are the values to which all estimated
percentage reductions in croplands emissions were applied.
V-32 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V —AGRICULTURE • RESULTS
Table 1-10: Baseline Net GHG Emissions from Croplands from DAYCENT Estimates (MtC02eq)
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex I
OECD
Russian Federation
South &SE Asia
United States
World Total
2000
29
508
13
27
91
U8
91
66
0
14
171
313
171
25
167
839
2010
32
484
17
30
97
39
93
69
0
16
123
338
123
26
179
830
2020
36
S21
17
30
104
41
101
73
0
28
124
373
124
28
200
893
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Note: These emissions include only croplands used for wheat, maize, or soybean production.
Note: Combinations of countries included in regions available from DAYCENT are not identical to those included in regions presented in this
report, but were aggregated to approximate these regions as closely as possible
Table 1-11 shows information on the yield effects, emissions reductions, and costs associated with
individual croplands mitigation options for the United States, EU-15, Brazil, China, and India.
Table 1-12 provides estimates of the percentage reduction in net GHG emissions (relative to the
croplands baseline used in this analysis) that could potentially be achieved at prices between $0/tCO2eq
and $60/tCO2eq for both 2010 and 2020 in major regions around the world.
Figure 1-7 shows the globally aggregated MAC for cropland greenhouse gas mitigation for 2000,
2010, and 2020, in terms of percentage emissions reductions from baseline over the applicable carbon
price range. With no price signal (i.e., at $0/tCO2eq), approximately 15 percent of cropland net GHG (N2O
and soil carbon) can be mitigated. More than 190 million tCO2eq (about 22 percent to 23 percent of
baseline emissions, depending on which year is analyzed) are mitigated at less than $45/tCO2eq in 2010
and 2020, but costs begin to rise rapidly beyond that point. Mitigation levels do not substantially increase
at higher prices, given the mitigation options considered here.
Negative costs result from options with cost savings because of lower applications of fertilizers while
maintaining yields, whereas high-cost options are those where revenues decline as yields decline in
response to suboptimal fertilizer applications. Negative cost options are consistent with previous studies,
finding large potential agricultural mitigation from "no-regret" options. The fact that farmers are not
adopting options that seemingly would increase profitability indicates that this analysis may not capture
some costs barriers to adoption exist, such as increased variability of profits or complexity of
management requirements.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-33
-------
SECTION V —AGRICULTURE • RESULTS
do -«d odd dod
9 *
d d "-1 o
o o o
ss
< ? <
z < z
in z *t
ui cxi
CM
I I I
odd
CO f—
op co co
cvi^ooociocicioooo o o c> o cs o o cj o ci
c\i»~'dddo°ddddddd d o o d d d d d d d
CO QO IO CO
CO IA
S
en in in o ci co oj o d o
TTT TTTT
S3 !o "• -• '*- — »- ^ ° *"
odododddocpd
co •*- c» So 53 p o T—• S S eo p So p
cjdcJdddRidi^T-^ddd d
T ?
o o o o
S
?!
a.
O
Sf
09 ^ SD ^^ CD 10 03 Pi ^J* O) o cjo QP
<£>T-^COCOC3OQ6cOO>o6T-:O<£>
*
CD
O)
03
•o
JS
Q.
2
o
I
f
S
o
z z z z
V-34
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • RESULTS
o o «-
o o o o
f i °P
o o o"
si £ I §
o »- o o e\i o o
il
CA
O
"Si
cc
S
o
Q.
O
o
OOOQOOOT-OCMOO
g* 8 8! Sj fe
ca ci c? o o
T
5O i-
T^ *~_
es c\i ei
-------
SECTION V — AGRICULTURE • RESULTS
jjf
i* §
DC J!
Slfi
CO
3^ 3^ 3^* ^' 3^ 3? 3^ 3^ 3^ 3^ 5_ S^~ ^? S^- 3~-
OCOT-OOOOi-OOOOOOOOOOOOOOOOOOOOOO-
' CD tO CNI O
T^ O O O
O> CO ^f O> ^^ ^^ to GO O}
ocjoodeiood
dodo dddd dddd
d d T- d d d
i O) i
5^- ^^ 3^- 3^ 3^~
00 ••» CO CO i-. 1-; -r- CVI T-; T-. ^OCO-^cgf-.Ol^-;
ddddddddoodddi^^ddd
T3
a>
(A
O
n
cc
><
>
Q_
O
U.I
V-36
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • RESULTS
Table 1-12: Croplands: Percentage Reductions from Baselines at Different $/tC02eq Prices
Country/Region
Africa
Annex 1
Australia/New
Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
11.1%
20.6%
21.2%
5.3%
6.4%
14.6%
11.9%
6.2%
11.8%
10.8%
28.3%
18.0%
28.3%
8.1%
21.7%
15.4%
$15
12.8%
23.2%
21.2%
5.3%
6.4%
18.8%
12.7%
11.4%
11.8%
14.3%
28.3%
21.4%
28.3%
8.3%
25.9%
17.6%
2010
$30
13.9%
29.7%
24.9%
13.3%
6.7%
21.0%
13.0%
11.4%
11.8%
23.4%
47.8%
23.8%
47.8%
9.6%
28.5%
22.0%
$45
14.5%
30.2%
34.7%
13.3%
10.1%
21.5%
13.7%
12.0%
11.8%
23.4%
47.9%
24.5%
47.9%
13.5%
28.5%
23.1%
$60
14.5%
30.9%
34.7%
13.3%
12.7%
24.1%
15.5%
12.4%
12.5%
23.4%
48.3%
25.0%
48.3%
14.4%
28.5%
24.0%
2020
$0
10.6%
19.6%
21.9%
4.5%
5.8%
13.5%
10.8%
5.8%
11.0%
10.5%
28.0%
17.0%
28.0%
8.3%
20.3%
14.6%
$15
13.5%
20.7%
21.9%
4.5%
6.3%
17.9%
10.8%
11.5%
11.0%
23.2%
28.0%
18.7%
28.0%
8.4%
21.0%
16.2%
$30
13.6%
24.2%
26.1%
12.3%
7.3%
20.8%
11.4%
11.5%
11.0%
23.2%
31.7%
22.0%
31.7%
11.0%
26.5%
18.8%
$45
14.0%
28.6%
36.1%
12.4%
10.5%
20.8%
11.7%
11.5%
11.6%
23.2%
47.5%
22.9%
47.5%
14.0%
26.5%
22.0%
$60
14.2%
29.2%
36.1%
12.4%
12.5%
20.8%
13.8%
11.5%
11.6%
23.2%
47.9%
23.5%
47.9%
14.3%
26.5%
22.7%
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Figure 1-7: Global MAC for Net Greenhouse Gas Emissions from Croplands, Holding Area Constant,
2000-2020
200
150 -
100 ^
50 -
-50 J
5%
15% 20% 25% 30%
Percentage Reduction in Net GHG Emissions
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-37
-------
SECTION V — AGRICULTURE • RESULTS
Figure 1-8 shows the sensitivity of the global cropland MAC when only the three options that were
most effective at mitigating net GHGs (nitrogen inhibitors, split fertilization, and no till) are applied. The
excluded options (reducing baseline levels of nitrogen fertilizer by varying amounts) had little impact on
net emissions at the global scale. As expected, the MAC with only the three most effective options—
where these options are each essentially applied to one-third of the cropland base—shows greater
mitigation.
Figure 1-8:
Global MAC for Net Greenhouse Gas Emissions from Croplands, Holding Area Constant,
Allocating Adoption of Mitigation Strategies to the Three Most Effective Options Only, 2000-
2020
200
150 -
0
-50
5%
15% 20% 25% 30%
Percentage Reduction in Net GHG Emissions
Figures 1-9, 1-10, 1-11, and 1-12 show the MACs for four major regions of the world—the United
States, EU-15, FSU, and China. Each of those figures show simulated mitigation potential assuming equal
adoption of the six mitigation options included for croplands. If mitigation strategies were limited to the
three most effective options for emissions reductions, excluding the fertilizer reduction options that have
little to no impact on net greenhouse gas in our DAYCENT model runs, total mitigation potential would
increase. This is analogous to the change at the global level, observed in Figures 1-7 and 1-8. Percentage
emissions reductions vary substantially, from less than 15 percent for China up to almost 50 percent in
FSU. Among these four regions, the FSU has the largest emissions reduction potential, followed closely
by the United States: both have emissions reductions above 50 MtCO2eq at less than $50/tCO2eq. EU-15
and China have less than one-third of the emissions reductions of the FSU or United States, with potential
reductions of 10 MtCO2eq to 15 MtCO2eq at $50/tCO2eq.
V-38
GLOBAL MITIGATION OF MON-C02 GREENHOUSE GASES
-------
SECTION V —AGRICULTURE • RESULTS
Figure 1-9: MAC for Net Greenhouse Gas Emissions from Cropland Management in the United States,
Holding Area Constant, 2000-2020
200 -i
150 -
§• 100 l
o
O
50 -
/o 5% 10% 15%
25% 30% 35%
Percentage Reduction in Net GHG Emissions
Figure 1-10: MAC for Net Greenhouse Gas Emissions from Cropland Management in the EU-15, Holding
Area Constant, 2000-2020
200 -i
150 -
O
O
50 -
15%
Percentage Reduction in Net GHG Emissions
20%
GLOBAL MITIGATION OF MON-C02 GREENHOUSE GASES
V-39
-------
SECTION V — AGRICULTURE « RESULTS
Figure 1-11: MAC for Net Greenhouse Gas Emissions from Cropland Management in the FSU, Holding
Area Constant, 2000-2020
200
150 -
O
o
50
-50 J
•2000
2010
•2020
%~ 20% 30% 40% 50% 60%
Percentage Reduction in Net GHG Emissions
Figure 1-12: MAC for Net Greenhouse Gas Emissions from Cropland Management in China, Holding Area
Constant, 2000-2020
200 i
150 -
O
O
100 \
50 -I
-50 J
Yo 2% 4%
T r
8% 10% 12% 14% 16%
Percentage Reduction in Net GHG Emissions
V-40
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V —AGRICULTURE . RESULTS
V.3.3 Baselines, Mitigation Costs, and MACs for Rice Cultivation
Table 1-13 presents estimates of baseline net GHG emissions from rice cultivation by region by year
used in this analysis. These are the values to which all estimated percentage reductions in emissions were
applied.
Table 1-13: Baseline Emissions from Rice Cultivation from DNDC Estimates (MtC02eq)
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECO
Russian Federation
South & SE Asia
United States
World Total
2000
—
45
—
—
385
—
—
127
45
—
—
63
—
929
—
1,504
2010
—
28
—
—
301
—
—
111
28
—
—
43
—
583
—
1,038
2020
—
27
—
—
302
—
—
122
27
—
—
43
—
594
—
1,062
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development
Table 1-14 presents information on the yield effects, emissions reductions, and costs associated with
individual rice cultivation mitigation options for China and India.
Table 1-15 provides estimates of the percentage reduction in net GHG emissions (relative to the rice
cultivation baseline used in this analysis) that could potentially be achieved at prices between $0/tCO2eq
and $60/tCO2eq for both 2010 and 2020 in major regions around the world.
Figure 1-13 shows the MACs estimated for 2000, 2010, and 2020. This outward shift in the curve
reflects changes in baseline emissions, commodity prices, labor rates, and other factors over time. Total
global mitigation for rice CH4 is estimated to be around 3 percent at negative or zero cost and about 13
percent at $45/tCO2eq in 2000. After that level, costs rise very rapidly. By 2010, global mitigation is
estimated to have increased to about 11 percent at negative or zero cost and 24 percent at $45/tCO2eq.
Between 2010 and 2020, there is little change in the MAC throughout most of its range.
Figures 1-14 and 1-15 display the MACs for the key rice-producing regions of India and China,
respectively. In both regions, the percentage emissions reduction is higher in 2010 than in 2000, but the
curve shifts inward in 2020. This is largely due to substantial changes in the baseline emissions over time
that are changing the reductions in net GHG relative to baseline conditions available. For instance,
baseline emissions from China are projected to decline substantially over time, leaving fewer emissions to
be abated in future years. DNDC simulations project baseline emissions from rice cultivation in China to
fall by 21.5 percent between 2000 and 2020, from 384.9 MtCO2eq in 2000 to 302.1 MtCO2eq by 2020.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • RESULTS
Table 1-14: Rice Cultivation Mitigation Option Detail for Key Regions
China
Option Labels
Midseason drainage—
rice(rf)
Midseason drainage—
rice(irri)
Shallow flooding—
rice(rf)
Shallow flooding—
rice(irri)
Offseason straw—
rice(rf)
Offseason straw—
rice(irri)
Sulfate fertilizer—
rice(rf)
Sulfate fertilizer—
rice(irri)
Slow-release
fertilizers— rice(rf)
Slow-release
fertilizer— rice(irri)
Switch to upland
rice— rice(rf)
Switch to upland
rice— rice(irri)
Breakeven
Cost
($tCOzeq)
NA
3.2%
NA
5.5%
NA
2.1%
NA
1.8%
NA
5.4%
NA
-15.0%
Emission
Reduction
(absolute,
MtC02eq)
NA
-$11.4
NA
-$1.9
NA
-$3.6
NA
-$2.9
NA
$27.9
NA
$10.1
Emission
Reduction
(1%from
baseline)
NA
3.5
NA
28.6
NA
2.2
NA
13.1
NA
-2.3
NA
45.1
Change in
Output
Yield (%
from
baseline)
NA
1.2%
NA
9.5%
NA
0.7%
NA
4.4%
NA
-0.8%
NA
15.0%
India
Breakeven
Cost
(StCOzeq)
NA
0.7%
NA
0.3%
0.0%
0.0%
0.0%
-0.3%
0.0%
-0.3%
6.1%
-32.9%
Emission
Reduction
(absolute,
MtCOjeq)
NA
4
NA
7
81
9
89
19
248
-319
21
62
Emission
Reduction
(1%from
baseline)
NA
8.9
NA
14.9
0.3
5.7
0.3
2.0
0.3
-0.3
-3.9
15.7
Change in
Output
Yield (%
from
. baseline)
NA
8.0%
NA
13.4%
0.3%
5.2%
0.3%
1.8%
0.2%
-0.2%
-3.5%
14.1%
NA = Data unavailable.
V-42
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • RESULTS
Table 1-15: Rice Cultivation: Percentage Reductions from Baseline at Different $/tC02eq Prices
Country/Region
Africa
Annex 1
Australia/New
Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South &SE Asia
United States
World Total
$0
NA
1.6%
NA
NA
15.8%
NA
NA
-0.2%
1.6%
NA
NA
4.0%
NA
10.4%
NA
10.5%
$15
NA
24.1%
NA
NA
30.8%
NA
NA
26.4%
24.1%
NA
NA
19.5%
NA
16.6%
NA
21.9%
2010
$30
NA
24.1%
NA
NA
30.0%
NA
NA
24.7%
24.1%
NA
NA
24.8%
NA
16.8%
NA
21.8%
$45
NA
24.1%
NA
NA
30.0%
NA
NA
24.7%
24.1%
NA
NA
24.8%
NA
20.7%
NA
24.0%
$60
NA
24.1%
NA
NA
30.0%
NA
NA
24.7%
24.1%
NA
NA
24.8%
NA
22.3%
NA
24.9%
2020
$0
NA
1.6%
NA
NA
13.1%
NA
NA
-0.3%
1.6%
NA
NA
4.4%
NA
12.1%
NA
10.7%
$15
NA
24.4%
NA
NA
26.3%
NA
NA
25.9%
24.4%
NA
NA
25.0%
NA
19.1%
NA
22.1%
$30
NA
24.4%
NA
NA
27.0%
NA
NA
25.9%
24.4%
NA
NA
25.0%
NA
19.1%
NA
22.4%
$45
NA
24.4%
NA
NA
27.0%
NA
NA
25.9%
24.4%
NA
NA
25.0%
NA
22.7%
NA
24.4%
$60
NA
24.4%
NA
NA
27.0%
NA
NA
25.9%
24.4%
NA
NA
25.0%
NA
22.7%
NA
24.4%
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development; NA = Data unavailable.
Figure 1-13: Global MAC for Net Greenhouse Gas Emissions from Rice Cultivation, Holding Area
Constant, 2000-2020
200
10% 15% 20% 25% 30%
Percentage Reduction in Net GHG Emissions
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-43
-------
SECTION V — AGRICULTURE « RESULTS
Figure 1-14: MAC for Net Greenhouse Gas Emissions from Rice Cultivation in India, Holding Area
Constant, 2000-2020
200 -,
150 -
or 100
o
o
50
-2000
2010
•2020
40%
Percentage Reduction in Net GHG Emissions
50%
Figure 1-15: MAC for Net Greenhouse Gas Emissions from Rice Cultivation in China, Holding Area
Constant, 2000-2020
200
150 -
§r 100
O
O
50 -I
5% 10% 15% 20%, 25% 30% 35%
Percentage Reduction in Net GHG Emissions
V-44
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • RESULTS
V.3.4 Baselines, Mitigation Costs, and MACs for Livestock
Management
Table 1-16 presents estimates of baseline net GHG emissions from livestock enteric fermentation and
manure management by region and by year. These are the values to which all estimated percentage
reductions in emissions were applied.
Table 1-17 presents information on the yield effects, emissions reductions, and costs associated with
individual livestock management mitigation options for the USA, EU-15, Brazil, China, and India.
Table 1-16: Baseline Emissions from Livestock Management from USEPA (2006) (MtC02eq)
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
271
704
91
222
313
48
222
224
20
43
111
642
66
187
171
2,220
2010
332
718
93
263
392
54
203
260
21
50
131
644
78
232
173
2,548
2020
395
748
94
297
470
58
202
286
22
57
150
663
91
276
171
2,867
EU-15 = European Union; OECD = Organisation tor Economic Co-operation and Development.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-45
-------
SECTION V — AGRICULTURE • RESULTS
id,
ss
•^woo^i-^
,_
T
^i
O4 OJ
O O
fp < <
T
T-OO-»-O-*-OO-»— O^-O
o o o i-' ca i-' T-' o T-' T-'
^-_ o> •** o> -^ -«r os «a- os-^-
^ ro T-1 co T-^ r-^ co ^ m ^
tlO CO
^ 2
-t— o o o o o o o o o o oo
11
O Cji O O t-" O »-' O CD
'
O C\J O OJ
^ D 3
10 h- i^ m t^ urir-
CJ W -r-" Cvl -r-' CVJ -r-'
;1
Q.
o
£ ^
1*1-
o o o o c>
-------
SECTION V — AGRICULTURE • RESULTS
1
03
o
'5
O)
oc
a
0>
o
Q.
o
o
'a
cr
CO
1
•«•
o o ooo o o o oo
CM 04 *-.*-.*-. *-. *»-: *-. *":'*•:
O CD OOO O O O O O
8
o o ooo o o o oo
m W
z z z
z zz
zzz
J
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
VA7
-------
SECTION V —AGRICULTURE • RESULTS
S
dddoddddd«c<<<«c<<<<<<<<t£>tt>CV|C3OOJ CO CO
CSJOOOTTO^ido^^^^^W
£ S? S« S?
p fa p in
o
o
o
'•i,
O)
V-48
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • RESULTS
Table 1-18 provides estimates of the percentage reduction in net GHG emissions (relative to the
livestock emissions baseline used in this analysis) that could potentially be achieved at prices between
$0/tCO2eq and $60/tCO2eq for both 2010 and 2020 in major regions around the world.
Table 1-18: Livestock Management: Percentage Reductions from Baselines at Different $/tC02eq Prices
2010
Country/Region
Africa.
Annex 1
Australia/New
Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
$0
0.7%
5.0%
4.1%
2.9%
2.0%
2.7%
6.3%
1.2%
4.0%
3.3%
2.5%
5.4%
2.4%
3.9%
6.4%
3.0%
$15
2.1%
6.9%
4.3%
4.5%
3.4%
2.7%
10.1%
2.1%
4.0%
3.3%
2.5%
7.4%
2.4%
5.3%
9.4%
4.4%
$30
2.6%
10.1%
6.8%
4.9%
3.7%
2.7%
13.0%
2.5%
4.4%
3.3%
2.5%
11.1%
2.4%
6.1%
17.2%
5.6%
$45
3.5%
11.3%
7.5%
4.9%
3.7%
2.7%
13.0%
2.5%
4.4%
3.4%
2.5%
12.5%
2.4%
6.2%
21.4%
6.1%
$60
3.6%
12.5%
8.4%
6.5%
4.1%
2.7%
16.9%
2.5%
4.4%
3.4%
2.5%
13.8%
2.4%
6.6%
21.4%
6.8%
2020
$0
0.5%
4.9%
4.2%
2.9%
2.0%
2.7%
6.4%
1.2%
4.1%
3.3%
2.4%
5.3%
2.4%
3.5%
6.3%
2.9%
$15
2.1%
7.4%
4.6%
4.5%
3.4%
2.7%
10.3%
2.5%
4.1%
3.3%
2.4%
8.1%
2.4%
4.6%
11.8%
4.4%
$30
2.6%
10.3%
7.2%
4.9%
3.7%
2.7%
12.2%
2.5%
4.1%
3.3%
2.4%
11.4%
2.4%
6.0%
19.8%
5.5%
$45
3.5%
11.9%
7.7%
4.9%
3.7%
2.7%
15.2%
2.5%
4.4%
3.4%
2.4%
13.3%
2.4%
6.0%
23.0%
6.1%
$60
3.6%
12.7%
8.7%
6.5%
4.0%
3.5%
17.1%
2.5%
4.4%
3.4%
3.0%
14.0%
2.9%
6.5%
23.0%
6.7%
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Total global mitigation for livestock management in 2020, holding the number of animals constant, is
estimated to be 3 percent at negative or zero cost, reaching about 7 percent at $60/tCO2eq (Figure 1-16).
Figure 1-17 shows the global MAC, holding production constant. The percentage of baseline emissions
mitigated at $60/tCO2eq increases from just under 7 percent with a constant number of animals to over 10
percent with constant production. If other greenhouse gas benefits were included (e.g., soil carbon
increases, cropland N2O reductions for less feed), the estimates of greenhouse gas mitigation would be
higher, but no model was identified to allow estimation of multigas impacts for livestock analogous to
the DNDC and DAYCENT models used for cropland management and rice cultivation.
Figures 1-18, 1-19, 1-20, and 1-21 show the MACs for mitigation of greenhouse gas emissions from
livestock management for the United States, China, India, and Brazil, respectively, holding the number of
animals constant. Among these four regions, the United States has the greatest potential for relatively
low-cost reductions in emissions in this sector, followed by China, Brazil, and India.
Because some options for mitigating emissions from enteric fermentation rely on improvements in
yield resulting in fewer animals to achieve emissions reductions, assumptions about changes in livestock
populations and production are important to examine for this sector. Thus, MACs are presented for the
United States, China, India, and Brazil, assuming that production remains constant (Figures 1-22, 1-23,
1-24, and 1-25) to show the impact of this assumption. The differences between the two sets of graphs
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-49
-------
SECTION V — AGRICULTURE • RESULTS
Figure 1-16: Global MAC for Greenhouse Gas Emissions from Livestock Management, Holding Number
of Animals Constant, 2000-2020
200 i
150
O
o
-50 J
-2000
2010
•2020
4% 6% 8% 10% 12% 14%
Percentage Reduction in Net GHG Emissions
Figure 1-17: Global MAC for Greenhouse Gas Emissions from Livestock Management, Holding
Production Constant, 2000-2020
-50 J
15%
Percentage Reduction in Net GHG Emissions
20%
V-50
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • RESULTS
Figure 1-18: MAC for Greenhouse Gas Emissions from Livestock Management in the United States,
Holding Number of Animals Constant, 2000-2020
200 n
-50 J
20%
Percentage Reduction in Net GHG Emissions
25%
Figure 1-19: MAC for Greenhouse Gas Emissions from Livestock Management in China, Holding Number
of Animals Constant, 2000-2020
200 -i
8%
Percentage Reduction in Net GHG Emissions
10%
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-51
-------
SECTION V — AGRICULTURE • RESULTS
Figure 1-20:
MAC for Greenhouse Gas Emissions from Livestock Management in India, Holding Number
of Animals Constant, 2000-2020
200 -,
150 -
100
CM
o
50 -
0
-50 J
•2000
2010
•2020
Vo 1% :>% 3% 4% 5% 6% 7% 8% 9% 10%
Percentage Reduction in Net GHG Emissions
Figure 1-21: MAC for Greenhouse Gas Emissions from Livestock Management in Brazil, Holding Number
of Animals Constant, 2000-2020
200
8%
Percentage Reduction in Net GHG Emissions
10%
V-52
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • RESULTS
Figure 1-22: MAC for Greenhouse Gas Emissions from Livestock Management in the United States,
Holding Production Constant, 2000-2020
25%
Percentage Reduction in Net GHG Emissions
30%
Figure 1-23: MAC for Greenhouse Gas Emissions from Livestock Management in China, Holding
Production Constant, 2000-2020
200
150 -
100 -j
o
O
-50 J
6%X8% 10% 12% 14% 16%
Percentage Reduction in Net GHG Emissions
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-53
-------
SECTION V — AGRICULTURE • RESULTS
Figure 1-24: MAC for Greenhouse Gas Emissions from Livestock Management in India, Holding
Production Constant, 2000-2020
200 -i
Percentage Reduction in Net GHG Emissions
-2000
2010
•2020
12% 14%
Figure 1-25: MAC for Greenhouse Gas Emissions from Livestock Management in Brazil, Holding
Production Constant, 2000-2020
200 i
150 -
o
o
Percentage Reduction in Net GHG Emissions
12% 14%
V-54
GLOBAL MITIGATION OF NON-CO;, GREENHOUSE GASES
-------
SECTION V — AGRICULTURE « RESULTS
reveal the importance of this assumption, because total mitigation is substantially larger at any given
price when production is assumed to remain constant. For those options that increase yields, constant
production can be maintained while reducing the number of livestock by an amount corresponding to the
increase in productivity. These reductions in the number of livestock can have a sizable impact on
emissions. With an assumption that the number of animals remains constant, emissions tend to fall less
because the reductions in emissions only come from the change in emissions per animal with no change
in emissions due to a change in population. Some mitigation options even increase net greenhouse gas
emissions per animal but increase productivity by an even greater proportion, leading to lower emissions
per unit of output. If a constant number of animals is assumed, however, then these options will lead to
an increase in emissions.
V.3.5 Baselines, Mitigation Costs, and MACs for Total Agriculture
Table 1-19 presents estimates of baseline net GHG emissions from agriculture, aggregated across
croplands management, rice cultivation, and livestock management by region by year used in this
analysis. These are the values to which all estimated percentage reductions in emissions were applied.
Table 1-19: Baseline Emissions from All Agriculture Used in This Report (MtC02eq)
Country/Region
Africa
Annex 1
Australia/New Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South & SE Asia
United States
World Total
2000
301
1,258
104
249
789
86
313
417
65
57
282
1,018
237
1,141
338
4,563
2010
'364
1,230
109
292
791
93
296
441
49
67
254
1,026
201
842
351
4,417
2020
431
1,297
111
327
876
99
303
480
50
85
274
1,080
215
898
370
4,822
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Note: These emissions reflect the baseline emissions used in calculating agricultural mitigation.
Table 1-20 provides estimates of the percentage reduction in net GHG emissions (relative to the
aggregated agricultural emissions baseline used in this analysis) that could potentially be achieved at
prices between $0/tCO2eq and $60/tCO2eq for both 2010 and 2020 in major regions around the world.
Figures 1-26 and 1-27 present MACs for global agriculture, aggregated across croplands
management, rice cultivation, and livestock management, for assumptions of constant area and number
of animals and constant production, respectively.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-55
-------
SECTION V — AGRICULTURE • RESULTS
Table 1-20: Total Agriculture: Percentage Reductions from Baseline at Different $/tC02eq Prices
Country/Region
Africa
Annex 1
Australia/New
Zealand
Brazil
China
Eastern Europe
EU-15
India
Japan
Mexico
Non-OECD Annex 1
OECD
Russian Federation
South &SE Asia
United States
World Total
$0
1.6%
11.1%
6.7%
3.2%
7.8%
7.7%
8.1%
1.6%
2.7%
5.2%
15.0%
9.5%
18.2%
8.5%
14.2%
7.1%
$15
3.1%
13.7%
6.9%
4.5%
14.2%
9.5%
10.9%
9.7%
15.5%
6.0%
15.0%
12.5%
18.2%
13.3%
17.8%
11.0%
2010
$30
3.6%
18.1%
9.5%
5.8%
14.1%
10.4%
13.0%
9.5%
15.6%
8.3%
24.4%
15.8%
30.1%
13.7%
22.9%
12.5%
$45
4.5%
19.1%
11.6%
5.8%
14.5%
10.7%
13.3%
9.6%
15.6%
8.3%
24.5%
16.9%
30.2%
16.4%
25.0%
13.5%
$60
4.5%
20.0%
12.4%
7.2%
15.0%
11.7%
16.4%
9.7%
15.6%
8.3%
24.7%
17.9%
30.4%
17.7%
25.0%
14.3%
2020
$0
1.4%
10.8%
6.9%
3.1%
6.3%
7.2%
7.9%
1.5%
2.8%
5.0%
14.0%
9.3%
17.2%
9.2%
13.8%
6.7%
$15
3.0%
13.1%
7.3%
4.5%
11.6%
9.0%
10.5%
9.3%
15.5%
8.0%
14.0%
12.4%
17.2%
14.1%
16.8%
10.4%
$30
3.5%
16.2%
10.2%
5.6%
12.1%
10.3%
12.0%
9.3%
15.5%
8.0%
15.7%
15.6%
19.3%
14.6%
23.4%
11.6%
$45
4.4%
18.9%
12.1%
5.6%
12.5%
10.3%
14.0%
9.3%
15.7%
8.0%
22.9%
17.0%
28.4%
17.0%
24.9%
13.0%
$60
4.4%
19.6%
12.9%
7.0%
12.9%
10.7%
16.0%
9.3%
15.7%
8.0%
23.3%
17.7%
28.9%
17.2%
24.9%
13.4%
EU-15 = European Union; OECD = Organisation for Economic Co-operation and Development.
Figure 1-26: Global MAC for Net Greenhouse Gas Emissions from Agriculture, Holding Area/Animals
Constant, 2000-2020
200 ->
15%
Percentage Reduction in Net GHG Emissions
20%
V-56
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • RESULTS
Figure 1-27: Global MAC for Net Greenhouse Gas Emissions from Agriculture, Holding Production
Constant, 2000-2020
200 n
-50 J
20%
Percentage Reduction in Net GHG Emissions
25%
V.3.6 Agricultural Commodity Market Impacts of Adopting Mitigation
Options: Use of the IMPACT Model
Many of the mitigation options considered for the agricultural sector have substantial impacts on
agricultural productivity and/or the cost of production. As a result, any significant adoption of these
options is expected to shift the market supply curves and move agricultural commodity markets to new
equilibrium points. Increases in productivity will have positive supply shifts and will tend to increase
market equilibrium quantity and reduce market prices, whereas increases in production costs will have
the opposite effect, reducing market quantity and putting upward pressure on market prices. Thus,
market equilibrium price and quantity could be either higher or lower than current and projected
baseline levels, depending on the net effects of the mitigation options. Not only do these market-level
impacts affect the cost of greenhouse gas mitigation, they also affect total mitigation, because emissions
are generally tied to the quantity of output produced. For instance, there may be an option that reduces
net greenhouse gas emissions per hectare of cropland but through broader market effects then leads to an
increase in equilibrium cropland area. With a large enough increase in area, total emissions from
adopting the option may actually increase because additional emissions from the larger area more than
offset the reduction in emissions per hectare.
To examine the magnitude of these market-level effects in agricultural markets, IFPRI's IMPACT
model is used. IMPACT models world supply and demand for agricultural products for 36 world
regions, as well as trade between regions. The model is capable of incorporating shifts in supply and/or
demand in one or more commodities in one or more regions and then solving for a new global
equilibrium in all commodities.
For cropland and rice cultivation options, estimated percentage changes in yield (kg/hectare) and
percentage changes in production costs per unit of output ($/metric ton) for each mitigation option were
provided as inputs to the IMPACT model. Similarly, percentage changes in milk or meat production per
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-57
-------
SECTION V — AGRICULTURE • RESULTS
animal (kg/head) and production costs per unit of output ($/metric ton) estimated for each enteric
fermentation option were provided as inputs to IMPACT. Manure management greenhouse gas
mitigation options were assumed to have no impact on livestock productivity and were not run through
the IMPACT model.
Applying the percentage changes in productivity and costs implied by the mitigation options to the
baseline levels of these variables in the IMPACT model, the model moves to a new equilibrium. The
values for key variables in the baseline and mitigation scenario are then compared to determine the
incremental impacts of adopting the mitigation option. Mitigation options are run through the model
independently from other options, but each option is applied simultaneously to all regions where that
option was believed to be feasible.
Two illustrative examples are presented to show the market adjustments being captured in the
IMPACT model and the influence of those adjustments on the abatement curves. The first examines the
effects of global adoption of the antimethanogen vaccine option on beef, dairy, and sheep and goat meat
markets relative to baseline values. As shown in Figure 1-28, world prices are reduced for all three of the
primary livestock categories that may adopt this mitigation option, with reductions ranging from about 4
percent to 9 percent. Figure 1-29 presents the effects on global production, where production increases by
2 percent to 4 percent for each product. In Figure 1-30, the impacts on global animal numbers are shown.
The IMPACT model estimates reductions in the number of animals of approximately 0.5 percent to 2
percent for each livestock category included. These changes in prices, production, and animal numbers
are attributable to the increase in productivity associated with this option. Although there are costs of
purchasing and administering the vaccine, the increase in productivity more than offsets these costs in
most regions, leading to more production from fewer animals. The resulting positive supply shift
decreases equilibrium market price.
Figure 1-28: Effect of Global Adoption of the Antimethanogen Vaccine Mitigation Option on World Prices
Using the IMPACT Model
3,500 -]
3,000 -
2,500 -
2,000 -
1,500 -
1,000 -
500 -
0
-*•
2000
2010
Year
2020
- Beef Baseline
•Dairy Vaccine
Beef Vaccine
-Sheep and Goats Baseline
•Dairy Baseline
•Sheep and Goats Vaccine
V-58
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • RESULTS
Figure 1-29: Effect of Global Adoption of the Antimethanogen Vaccine Mitigation Option on Global
Production Using the IMPACT Model
800,000 -,
700,000 -
600,000 -
w 500,000 -
c
° 400,000 -
o
o
° 300,000 -
200,000 -
100,000 -
0
Jfc
2000
2010
Year
2020
-Beef Baseline
-Dairy Baseline
-Sheep and Goat Meat Baseline
Beef Vaccine
-Dairy Vaccine
•Sheep and Goat Meat Vaccine
ra
-= 600,000 -
o
o
o
*-~ 400,000 -
200,000 -
0
Figure 1-30: Effect of Global Adoption of the Antimethanogen Vaccine Mitigation Option on Global
Number of Animals Using the IMPACT Model
1,200,000 j-
1,000,000 -
800,000 -
2000
2010
Year
2020
-Beef Baseline
-Dairy Vaccine
Beef Vaccine
-Sheep and Goats Baseline
-Dairy Baseline
-Sheep and Goats Vaccine
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-59
-------
SECTION V — AGRICULTURE • RESULTS
The second example simulates the effects of global adoption of the shallow flooding mitigation
option on rice markets relative to baseline values. The shallow flooding option is expected to result in
both reduced emissions and higher productivity. As shown in Figure 1-31, the global price of rice is
reduced, with reductions ranging from about 3 percent to 10 percent. Figure 1-32 presents the effects on
global rice production, where production increases by 1 percent to 2.5 percent. In Figure 1-33, the impact
on global rice area is shown, with the IMPACT model estimating acreage reductions between 0.3 to 1
percent. Similar to the antimethanogen vaccine option examined above, this op>tion results in an increase
in productivity, which results in more production from less area. The positive supply shift leads to a
decrease in equilibrium market price.
These market-level changes in area, production, and price are then incorporated into the MAC to
examine the sensitivity of the MACs to inclusion of market effects. These effects are potentially very
important because many options will have an effect on equilibrium prices and quantities if widely
adopted, which will affect the cost of mitigation, as well as total mitigation achieved. However, the
engineering approach used in this report is unable to capture feedbacks from changing market
conditions. For comparison purposes, the MAC curves wiihout market adjustments are recalculated with
full adoption of a single option being analyzed to be consistent with the MACs calculated using IMPACT
model results.
Figure 1-34 compares net GHG abatement under global adoption of the antimethanogen vaccine
calculated three different ways: 1) with the number of animals held constant, 2) with production of the
relevant commodities held constant, and 3) allowing both number of animals and production to vary, as
well as reflecting other market adjustments simulated using the IMPACT model. Mitigation with
production held constant is much greater than with the number of animals held constant because this is
an option that increases productivity. Thus, the same production level can be achieved with fewer
animals, which provides additional emissions reductions on top of the reduction in emissions per animal
associated with the option. Incorporating market adjustments results in a price decrease, which leads to
reduced incentives for investment and production in the livestock sector than if price were constant.
Thus, the increase in production is smaller than for the constant number of animals case. Because there is
an increase in production under the market adjustments scenario, the reduction in number of animals is
smaller than for the constant production case. In addition,, there are shifts in production regions and trade
patterns, though the changes are relatively small for this particular option.
Similarly, Figure 1-35 compares net GHG abatement under global adoption of the shallow flooding
mitigation option: 1) with rice area held constant, 2) with rice production held constant, and 3) allowing
both area and production to vary, as well as reflecting other market adjustments simulated by the
IMPACT model. Mitigation is similar with area held constant and production held constant for this
option because the yield changes are relatively small, causing less differentiation between calculation
method. The global curve with market adjustments is similar to the area constant and production
constant curves other than at very low and very high prices. However, there are differences in the regions
estimated to provide mitigation at different price levels depending on the abatement cost calculation
method. Because there are larger differences in estimated yield changes between regions, there are more
substantial shifts in production regions and trade patterns than for the antimethanogen vaccine option
considered above.
V-60 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V —AGRICULTURE • RESULTS
Figure 1-31: Effect of Global Adoption of the Shallow Flooding Mitigation Option on World Prices Using
the IMPACT Model
250 -,
200 -
150 -
100 -
50 -
0
2000
2010
Year
2020
•Rice Baseline
Rice SF
Figure 1-32:
Effect of Global Adoption of the Shallow Flooding Mitigation Option on Global Production
Using the IMPACT Model
600,000
500,000
w 400,000
o
o 300,000
o
o
*" 200,000
100,000
0
2000
2010
Year
2020
•Rice Baseline
Rice SF
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-61
-------
SECTION V — AGRICULTURE • RESULTS
Figure 1-33: Effect of Global Adoption of the Shallow Flooding Mitigation Option on Global Rice Area
Using the IMPACT Model
154,000 -1
153,000 -
152,000 -
151,000 -
J2 150,000 -
§ 149,000 -
148,000 -
147,000 -
146,000 -
145,000
2000
2010
Year
2020
•Rice Baseline
Rice SF
Figure 1-34: Net GHG Abatement under Global Adoption of the Antimethanogen Vaccine Option with
Number of Cattle Constant, Production Constant, and Market Adjustments Using the
IMPACT Model, 2010
200
175
150
125
g- 100
CM
8 75
S 50
25
0
-50 -
8%
10%
12%
Percentage Reduction in Net GHG Relative to Baseline
-# Animals Constant
Production Constant
•# Animals+Production+Other Market Adjustments
V-62
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • RESULTS
Figure 1-35: Comparison of Net GHG Abatement from Rice Cultivation under Global Adoption of the
Shallow Flooding Mitigation Option with Area Constant, Production Constant, and Market
Adjustments Using the IMPACT Model, 2010
100
75 -
50 -
8 25 ^
0 -
0
(25)-
(50)-
10% 20%
30%
40%
50%
60%
Percentage Reduction in Net GHG Relative to Baseline
•Area Constant
Production Constant
•Area+Production+Other Market Adjustments
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
V-63
-------
SECTION V — AGRICULTURE • RESULTS
V-64 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • CONCLUSIONS
V.4 Conclusions
The agricultural sector generates the largest share of global non-CO2 greenhouse gas emissions and a
significant share of total global greenhouse gas emissions. Emissions in this sector are projected to
increase significantly over the foreseeable future, especially in developing countries. Mitigation options to
abate agricultural non-CO2 emissions have been identified in the literature for most significant sources.
This report uses a number of data sources, analytic tools, and modeling approaches to compile and
estimate baseline emissions of the most significant world agricultural non-CO2 sources and to estimate
the costs of the technical mitigation potential for key regions and world totals. This report makes no
explicit assumption about the policy mechanisms that might be required for adopting the mitigation
options in different regions.
Cost estimates are feasible for most greenhouse gas mitigation options in the agricultural sector. This
report uses an engineering bottom-up approach to estimate the $/tCO2eq of each mitigation option in
different regions, over different time periods, by including most key elements that affect cost: net
greenhouse gas effects, changes in crop yield or livestock productivity, regionally specific agricultural
commodity prices, capital and input costs required to implement the mitigation option, and changes in
labor requirements. The quality of data and tools used for these input parameters varies by region and by
agricultural emissions category.
At the globally aggregated scale (including all regions and all non-CO2 sources), the technical
potential to mitigate non-CO2 greenhouse gases from agriculture appears significant but not
overwhelming in percentage terms. At roughly zero costs (where the benefits of implementing the
mitigation options compensate for the costs of doing so), approximately 10 percent of global agricultural
non-CO2 emissions can be mitigated. At very high costs around $150/tCO2eq, the technical mitigation
potential at the global scale approaches 20 percent of baseline emissions. These estimated percentage
reductions are larger than the previous analysis supported by the USEPA and carried out for EMF-21, but
these two studies are very different; this report uses more recent baseline scenarios for livestock
emissions and uses completely different approaches to estimate the baseline and mitigation scenarios
from croplands and rice cultivation.
For individual emissions categories, the magnitude of net greenhouse gas mitigation potential
appears significant to modest depending on region and time frame. For global cropland N2O and soil
carbon emissions, approximately 15 percent of baseline emissions can be mitigated at zero costs. At costs
above $50/tCO2eq, the percentage reduction approaches 25 percent. Nitrification inhibitors and no-till
appear the most viable mitigation options, according to the simulations with the DAYCENT model, with
regard to net emissions reduction potential and yield effects. The options where baseline nitrogen
fertilizers are simply reduced result in net emissions similar to baseline levels.
For Asian rice systems (representing about 90 percent of world rice emissions), close to 15 percent of
net baseline emissions (CH4, N2O and soil carbon) can be mitigated at zero costs in years 2010 and 2020.
As costs approach $100/tCO2eq, over 25 percent of net baseline emissions can be mitigated. Shallow
flooding, ammonium sulfate, and full conversion to midseason drainage appear the most viable
mitigation options, according to the simulations with the DNDC model, with regard to net emissions
mitigation potential and yield effects. Upland rice almost eliminates emissions but has adverse yield
effects. Shallow flooding has the additional benefit of water conservation, though water as an unpriced
commodity in this context does not factor into the cost estimates.
For global livestock emissions, approximately 7 percent can be mitigated at zero costs assuming a
constant number of animals, whereas roughly 9 percent can be mitigated assuming a constant level of
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-65
-------
SECTION V — AGRICULTURE • CONCLUSIONS
production. As costs approach $125/tCO2eq, the technical mitigation potential reaches 16 and 18 percent
of baseline emissions, when, respectively, constant number of animals and constant production are
assumed. There are somewhat surprisingly few large differences among the mitigation options in terms
of their non-CO2 mitigation potential at the global scale relative to baseline levels.
Many mitigation options have negative costs and it is difficult to assess whether important costs have
been omitted or if barriers to adoption exist that are not accounted for. High-cost options tend to be those
that are either not very effective at reducing net greenhouse gases or that have adverse yield and
productivity effects. Adoption barriers have not been explicitly addressed (all mitigation options
considered technically feasible in a given region are assumed to be adopted in data year 2000).
Accounting for adoption barriers to gain a more complete picture of greenhouse gas mitigation potential
is an important area for future research.
Consideration of net greenhouse gas effects (CH4, N2O, and soil carbon) is particularly important in
the agricultural sector to evaluate the effectiveness of different mitigation options. This is especially true
for croplands and rice cultivation. In croplands, options considered a priori to be good candidates for
reducing soil N2O emissions (e.g., reducing baseline nitrogen application rates) led to no significant net
greenhouse gas emissions reductions relative to the baseline because of offsetting soil carbon effects.
Likewise with rice cultivation, some options are good CH4-reducing strategies but increase N2O (e.g.,
midseason drainage), while others have little effect on CH4 but are good N2O-reducing strategies (e.g.,
use of ammonium sulfate under certain conditions). The inclusion of fossil fuel CO2 emissions associated
with either on-farm practices or off-farm production processes, such as fertilizer production, is an
additional net greenhouse gas consideration that was not included here.
The long-lasting benefits of N2O and CH4 reductions relative to the potentially reversible benefits of
soil carbon sequestration deserve attention. In this report, there is no reversal of soil carbon sequestration
due to, say, adoption of no till, because this practice is assumed to be adopted immediately and
continuously through to 2020.
Estimating mitigation potential of agricultural non-CO2 emissions is challenging at the international
scale given the high degree of heterogeneity in management and biophysical conditions. Use of process
models like DAYCENT (for cropland emissions) and DNDC (for rice cultivation emissions) in this report
can help capture this variability and improve confidence in net greenhouse gas and yield estimates. Use
of these models at such large scales is intended to show the general trends between baseline and
mitigation scenarios, while adequately capturing heterogeneous effects. These models are not intended to
match small-scale (e.g., farm scale) conditions over the entire regions to which they are being applied.
Livestock baseline emissions are taken from USEPA (2006), which for many regions relies on IPCC
Tier I default methodologies. Mitigation studies found in the literature are applied to those baselines.
This approach raises two key issues that need to be addressed in future work: 1) the identification of
more detailed baseline management activities so that there is more certainty about the implications of
adopting the mitigation options and 2) the suitability of applying mitigation options to regions outside of
an original case study area.
Adoption of the mitigation options in this report would lead to agricultural commodity effects and
therefore would affect baseline commodity prices, crop area, production levels of crops, and livestock
populations and livestock products. These changes in turn change greenhouse gas levels and thus the
effectiveness of the mitigation options. The agricultural market sensitivity experiments with the IMPACT
model show this to be the case. The core mitigation estimates in this report use a static, engineering
approach that is unable to capture these market dynamics. For this reason, cost estimates and marginal
abatement curves are shown using either constant area (livestock population) or constant crop (livestock)
V-66 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V —AGRICULTURE • CONCLUSIONS
production. For the livestock sector, where many mitigation options actually increase emissions per
animal, it is particularly important to show the implications of both approaches; using constant livestock
production, as expected, leads to greater emissions reduction estimates. Fuller representation of these
market feedbacks for future global agricultural mitigation analyses will be important.
Additional research is also necessary to identify which mitigation options are best suited for different
regions and subregions and what kind of adoption barriers different mitigation options may face. This
report excludes some options from being applied in certain regions, but further refinement is required.
Finally, anthropogenic climate change is not considered in this report for the 2000 to 2020 period but
could affect future agricultural baseline emissions and thus mitigation potential. Anthropogenic climate
change will become a more important issue for analyses that look beyond 2020.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-67
-------
SECTION V — AGRICULTURE • CONCLUSIONS
V-68 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE • REFERENCES
V.5 References
AEA Technology Environment. 1998. Options to Reduce Nitrous Oxide Emissions. Report prepared for the
European Commission.
Babu, Y.J., C. Li, S. Frolking, D.R. Nayak, A. Datta, and T.K. Adhya. 2005. "Modelling of Methane
Emissions from Rice-Based Productions Systems in India with the Denitrification and Decomposition
Model: Filed Validation and Sensitivity Analysis." Current Science 89 11,1-6.
Bates, J. 2001. Economic Evaluation of Emissions Reductions of Nitrous Oxides and Methane in Agriculture in the
EU: Bottom-Up Analysis. Contribution to a Study for DG Environment, European Commission by
Ecofys Energy and Environment, AEA Technology Environment and National Technical University
of Athens.
Belward, A.S. 1996. The IGBP-DIS Global 1 Km Land Cover Dataset (Discover) Proposal and Implementation
Plans: IGBP-DIS Working Paper No. 13. Toulouse, France.
Belward, A.S., J.E. Estes, and K.D. Kline. 1999. "The IGBP-DIS 1-Km Land-Cover Data Set DISCover: A
Project Overview." Photogrammetric Engineering and Remote Sensing 65 9,1013-1020.
Bouwman, A.F., G. van Drecht, and K.W. van der Hoek. 2005. "Global and Regional Surface Nitrogen
Balances in Intensive Agricultural Production Systems for the Periode 1970-2030." Pedosphere 15 2,
137-55.
Cai, Z., S. Sawamoto, C. Li, G. Kang, }. Boonjawat, A. Mosier, and R. Wassmann. 2003. "Field Validation
of the DNDC Model for Greenhouse Gas Emissions in East Asian Cropping Systems." Global
Biogeochemical Cycles 17 4, doi:10.1029/2003GB002046.
Cai, Z.C., G.X. Xing, G.Y. Shen, H. Xu, X.Y. Yan, and H. Tsuruta. 1999. "Measurements of CH4 and N2O
Emissions from Rice Paddies in Fengqiu, China." Soil Science Plant Nutrition 45,1-13.
Cramer, W., D.W. Kicklighter, A. Bondeau, B. Moore, G. Churkina, B. Nemry, A. Ruimy, and A.L.
Schloss. 1999. "Intercomparison, the Participants of the Potsdam NPP Model. Comparing Global
Models of Terrestrial Net Primary Productivity (NPP): Overview and Key Results." Global Change
Biology Supplement 1, 5 4,1-15.
Conservation Technology Information Center (CTIC). 1994. 2994 National Crop Residue Management
Survey. West Lafayette, IN: Conservation Technology Information Center.
DeAngelo, B.J., F. de la Chesnaye, R.H. Beach, A. Sommer and B.C. Murray. In press. "Methane and
Nitrous Oxide Mitigation in Agriculture." Energy Journal.
Del Grosso, S.J., W.J. Parton, A.R. Mosier, M.D. Hartman, L. Brenner, D.S. Ojima, and D.S. Schimel. 2001.
Simulated Interaction of Carbon Dynamics and Nitrogen Trace Gas Fluxes Using the DAYCENT
Model. In M. Schaffer, et al. (Eds.), Modeling Carbon and Nitrogen Dynamics for Soil Management, p. 303-
332. Boca Raton, FL: CRC Press.
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.
Ehhalt, D., M. Prather, et al. 2001. Atmospheric Chemistry and Greenhouse Gases. In J.T. Houghton et al.
(eds.) Climate Change 2001: The Scientific Basis, Contribution of Working Group I to the Third
Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press.
Food and Agriculture Organization (FAO). 2000. fertilizer Requirements for 2015 and 2030. Rome, Italy: UN
Food and Agriculture Organization.
Food and Agriculture Organization (FAO). 2002. World Agriculture: Towards 2015/2030. Rome, Italy: UN
Food and Agriculture Organization.
Food and Agriculture Organization/United Nations Educational, Scientific, and Cultural Organization
(FAO/UNESCO). 1996. Digital Soil Map of the World and Derived Soil Properties; Derived from the
FAO/UNESCO Soil Map of the World, Original Scale 1:5,000,000. Rome, Italy: UN Food and Agriculture
Organization.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-69
-------
SECTION V — AGRICULTURE • REFERENCES
FAOSTAT. 2004a. FAO Production Yearbook 2003, v. 57. In: FAO Statistics Series (FAO). No. 177. Rome,
Italy: UN Food and Agriculture Organization.
FAOSTAT. 2004b. FAO Fertilizer Yearbook 2003, v. 53 In: FAO Statistics Series (FAO). No. 183. Rome,
Italy: UN Food and Agriculture Organization.
Frolking, S., J. Qiu, S. Boles, X. Xiao, J. Liu, Y. Zhuang, C. Li, and X. Qin. 2002. "Combining Remote
Sensing and Ground Census Data to Develop New Maps of the Distribution of Rice Agriculture in
China." Global Biogeochemical Cycles 16 4, doi:10.1029/2001GB001425.
Gerbens, S. 1998. Cost-Effectiveness of Methane Emissions Reduction from Enteric Fermentation of Cattle and
Buffalo. Draft Report.
Global Soil Data Task Group. 2000. Global Gridded Surfaces of Selected Soil Characteristics, International
Geosphere-Biosphere Programme —Data and Information System Data Set. Oak Ridge National Laboratory
Distributed Active Archive Center. Available at .
Global Trade Analysis Project (GTAP). 2005. GTAP6 Data Base. Available at
.
Groffman, P.M., R. Brumme, K. Butterbach-Bahl, K.E. Dobbie, A.R. Mosier, D. Ojima, H. Papen, W.J.
Parton, K.A. Smith, and C. Wagner-Riddle. 2000. "Evaluating Annual Nitrous Oxide Fluxes at the
Ecosystem Scale." Global Biogeochemcial Cycles 14 4,1061-1070.
Holzapfel-Pschorn, A., R. Conrad, and W. Seiler. 1985. "Production, Oxidation, and Emissions of
Methane in Rice Paddies." FEMS Microbiology Ecology 31,149-158.
International Fertilizer Industry Association (IFA). 2002. Fertilizer Use by Crop, 5th edition. Available at
.
Johnson, D., H.W. Phetteplace, A.F. Seidl, U.A. Schneider, and B.A. McCarl. 2003a. Selected Variations in
Management of U.S. Dairy Production Systems: Implications for Whole Farm Greenhouse Gas
Emissions and Economic Returns. In Proceedings of the 3rd International Methane and Nitrous Oxide
Mitigation Conference, November 17-21, Beijing, China.
Johnson, D., H.W. Phetteplace, A.F. Seidl, U.A. Schneider, and B.A. McCarl. 2003b. Management
Variations for U.S. Beef Production Systems: Effects on Greenhouse Gas Emissions and Profitability.
In Proceedings of the 3rd International Methane and Nitrous Oxide Mitigation Conference, November 17-21,
Beijing, China.
Kistler, R., E. Kalnay, W. Collins, S. Saha, G. White, J. Woolen, M. Chelliah, W. Ebisuzaki, M. Kanamitsu,
V. Kousky, H. van den Dool, R. Jenne, and M. Fiorino. 2001. "The NCEP-NCAR 50-Year Reanalysis:
Monthly Means CD-ROM and Documentation." Bulletin of the American Meteorological Society 82, 247-
267.
Kroeze, C., and A. Mosier. 1999. New Estimates for Emissions of Nitrous Oxides. In Second International
Symposium on Non-CO2 Greenhouse Gases. J. van Ham, A.P.M. Baede, L.A. Meyer, and R. Ybema (eds.).
Dordrecht, The Netherlands: Kluwer Academic Publishers.
Kriiger, M., G. Eller, R. Conrad, and P. Frenzel. 2001. "Seasonal Variation in Pathways of CH4 Production
and in CH4 Oxidation in Rice Fields Determined by Stable Carbon Isotopes and Specific Inhibitors."
Global Change Biology 8 3, 265-280.
Lai, R., J.M. Kimble, R.F. Follett, and C.V. Cole. 1998. The Potential of U.S. Cropland to Sequester Carbon and
Mitigate the Greenhouse Effect. Boca Raton, FL: Lewis Publishers.
Leff, B., N. Ramankutty, and J.A. Foley. 2004. "Geographic Distribution of Major Crops across the
World." Global Biogeochemical Cycles 18 1, GB1009 10.1029/2003GB002108.
Li, C., and W. Salas. 2005. Update Results for DNDC Dynamics Modeling Runs of Net GWP from Rice
Cultivation in China from 2000 to 2020 and Site Level Results for India and Thailand. Memorandum to
Benjamin DeAngelo, USEPA. April 26.
Li, C., A. Mosier, R. Wassmann, Z. Cai, X. Zheng, Y. Huang, H. Tsuruta, J. Boonjawat, and R. Lantin. 2004.
"Modeling Greenhouse Gas Emissions from Rice-Based Production Systems: Sensitivity and
Upscaling." Global Biogeochemical Cycles 18, GB1043, doi:10.1019/2003GB002045.
V-70 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
SECTION V — AGRICULTURE « REFERENCES
Li, C, W. Salas B. DeAngelo, and S. Rose. In press. "Assessing Alternatives for Mitigating Net
Greenhouse Gas Emissions and Increasing Yields from Rice Production in China Over the Next 20
Years." Journal of Environmental Quality.
Li, C., J. Qiu, S. Frolking, X. Xiao, W. Salas, B. Moore III, S. Boles, Y. Huang, and R. Sass. 2002. "Reduced
Methane Emissions from Large-Scale Changes in Water Management in China's Rice Paddies During
1980-2000." Geophysical Research Letter 29 20, doi:10.1029/2002GL015370.
Li, Q.K. 1992. Water Management of Paddy Soils. In Q. Li (ed.), Paddy Soils of China. Beijing, China:
Science Press, pp. 1-545 (in Chinese).
Melillo, J.M., A.D. McGuire, D.W. Kicklighter, B. Moore III, C.J. Vorosmarty, and A.L. Schloss. 1993.
"Global Climate Change and Terrestrial Net Primary Productivity Production." Nature 363, 234-240.
Moomaw, W.R., J.R. Moreira, K. Blok, et al. 2001. Technological and Economic Potential of Greenhouse
Gas Emissions Reduction. In Climate Change 2001: Mitigation, Intergovernmental Panel on Climate
Change, Working Group III. Cambridge University Press.
Moser, C.M., and C.B. Barrett. 2002. The System of Rice Intensification in Practice: Explaining Low Farmer
Adoption and High Disadoption in Madagascar. Cornell University Working Paper. Ithaca, NY: Cornell
University.
Mosier, A., C. Kroeze, C. Nevison, O. Oenema, S. Seitzinger, and O. van Cleemput. 1998. "Closing the
Global N2O Budget: Nitrous Oxide Emissions through the Agricultural Nitrogen Cycle —
OECD/IPCC/IEA Phase II Development of IPCC Guidelines for National Greenhouse Gas Inventory
Methodology." Nutrient Cycling in Agroecosystems 52, 225-248.
Mosier, A.R., J.W. Doran, and J.R. Freney. 2002. "Managing Soil Denitrification." Journal of Water and Soil
Conservation 57 6, 505-513.
Parton, W.J., M.D. Hartman, D.S. Ojima, and D.S. Schimel. 1998. "DAYCENT: Its Land Surface Submodel:
Description and Testing." Global and Planetary Change 19, 35-48.
Pathak, H., C. Li, and R. Wassmann. 2005. "Greenhouse Gas Emissions from Indian Rice Fields:
Calibration and Upscaling Using the DNDC Model." Biogeosciences 2, 113-123.
Peterson, G.A., A.D. Halvorson, J.L. Havlin, O.R. Jones, D.J. Lyon, and D.L. Tanaka. 1998. "Reduced
Tillage and Increasing Cropping Intensity in the Great Plains Conserves Soil C." Soil & Tillage
Research 47 3, 207-218.
Prentice, I.C., G.A. Farquhar, MJ.R. Fasham, et al. 2001. The Carbon Cycle and Atmospheric Carbon
Dioxide. In Climate Change 2001: The Scientific Basis, Intergovernmental Panel on Climate Change, Working
Group I. Cambridge University Press.
Riemer, P., and P. Freund. 1999. Technologies for Reducing Methane Emissions. IEA Greenhouse Gas
Programme.
Robertson, G.P. 2004. Abatement of Nitrous Oxide, Methane, and the Other Non-CO2 Greenhouse Gases:
The Need for a Systems Approach. In C.B. Field and M.R. Raupach (eds.), pp. 493-506, The Global
Carbon Cycle. Washington, DC: Island Press.
Roos. Personal communication. 2005.
Sass, R.L., P.M. Fisher, P.A. Harcombe, and F.T. Turner. 1990. "Methane Production and Emissions in a
Texas Rice Field." Global Biogeochemical Cycles 4, 47-68.
Scharf, P., L. Mueller, and J. Medeiros. 2005. Making Urea Work in No-Till. Grant progress report, Missouri
Agricultural Experiment Station. Available at .
Scheehle, E., and D. Kruger. In press. "Global Anthropogenic Methane and Nitrous Oxide Emissions."
Energy Journal.
Shen, Z.R., X.L. Yang, and Y.S. Pei. 1998. Enhancing Researches on Elevating Efficiency of Water Use in
Chinese Agriculture. In Z.R. Shen and R.Q. Su (eds.), Strategies Against Water Crisis in Chinese
Agriculture. Beijing, China: Chinese Agricultural Science and Technology Press, pp. 1-267. (in
Chinese).
Siebert et al. Personal communication.
GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES V-71
-------
SECTION V — AGRICULTURE « REFERENCES
Smith, K.A., and F. Conen. 2004. "Impacts of Land Management on Fluxes of Trace Greenhouse Gases."
Soil Use and Management 20, 255-263.
Stehfest et al. Personal communication.
U.S. Energy Information Administration (USEIA). 2003. Electricity Prices for Industry. Available at
.
U.S. Environmental Protection Agency (USEPA). 2003. Current Status of Farm-Scale Digesters. Ag STAR
Digest, EPA-430-F-02-028. Washington, DC: USEPA.
U.S. Environmental Protection Agency (USEPA). 2005a. Greenhouse Gas Mitigation Potential in U.S. Forestry
and Agriculture. EPA 430-R-05-006. Washington, DC: USEPA. Available at
.
U.S. Environmental Protection Agency (USEPA). 2005b. Inventory of U.S. Greenhouse Gas Emissions and
Sinks: 1990-2003. EPA 430-R-05-003. Washington, DC: USEPA.
U.S. Environmental Protection Agency (USEPA). 2006. Global Anthropogenic Non-CO2 Greenhouse Gas
Emissions: 1990-2020. Washington, DC: USEPA.
Van der Gon, H.A.D., P.M. Van Bodegom, R. Wassmann, R.S. Lantin, and T.M. Metra-Corton. 2001.
"Sulfate-Containing Amendments to Reduce Methane Emissions from Rice Fields: Mechanisms,
Effectiveness and Costs." Mitigation and Adaptation Strategies for Global Change 6, 71-89.
Van Vuuren, D., }. Weyant, and F. de la Chesnaye. In press. "Multigas Scenarios to Stabilise Radiative
Forcing." Energy Journal.
Wassmann, R., R.S. Lantin, H.U. Neue, L.V. Buendia, T.M. Corton, and Y. Lu. 2000. "Characterization of
Methane Emissions from Rice Fields in Asia. III. Mitigation Options and Future Research Needs."
Nutrient Cycling in Agroecosystems 58, 23-36.
Webb, R.W., C.E. Rosenzweig, and E.R. Levine. 2000. Global Soil Texture and Derived Water-Holding
Capacities Data Set. Oak Ridge National Laboratory Distributed Active Archive Center. Available at
.
Zhang, Y., C. Li, C.C. Trettin, H. Li, and G. Sun. 2002. "An Integrated Model of Soil, Hydrology and
Vegetation for Carbon Dynamics in Wetland Ecosystems." Global Biogeochemical Cycles
10.1029/2001GB001838.
Zheng X., M. Wang, Y. Wang, R. Shen, J. Gou, J. Li, J. Jin, and L. Li. 2000. "Impacts of Soil Moisture on
Nitrous Oxide Emissions from Croplands: A Case Study on the Rice-Based Agro-Ecosystem in
Southeast China." Chemosphere-Global Change Science 2, 207-224.
Zheng, X.H., M.X. Wang, Y.S. Wang, R.X. Shen, X.J. Shangguan, J. Heyer, M. Kogge, H. Papen, J.S. Jin,
and L.T. Li. 1997. "CH4 and N2O Emissions from Rice Paddies in Southeast China." Chinese Journal of
Atmosphere Science 21, 167-174.
V-72 GLOBAL MITIGATION OF NON-C02 GREENHOUSE GASES
-------
Section V: Agriculture Sector Appendixes
Appendixes for this section are available for download from the USEPA's Web site at
http://vvwvv.epa. go v/nonco2/econ-inv/international. html.
-------
SgSi
£°s'|
rs TJ /->
CQ 5 % (/)
T?" ^ -^ <-+
O =J > D3
-34g 3 CD
01-0 W
0« w m
N> g:' =J" <
o ° CD =;•
O) <
O CD
.
O
CD
^ o _
1^ 5T
< 03 —
3 7
w 3
"oJ CD"
ro o
o ?-
CD
D
O
------- |