1 EPA 430-R-16-002
2
3
4
5
6 DRAFT Inventory of U.S. Greenhouse Gas
7 Emissions and Sinks:
8 1990-2014
9
10
11
12
13
14 FEBRUARY 22,2016
15
16
17
18
19
20
21 U.S. Environmental Protection Agency
22 1200 Pennsylvania Ave., N.W.
23 Washington, DC 20460
24 U.S.A.
25
26
27
-------
1
2
3 HOW TO OBTAIN COPIES
4 You can electronically download this document on the U.S. EPA's homepage at
5 . To request free copies of this report, call
6 the National Service Center for Environmental Publications (NSCEP) at (800) 490-9198, or visit the web site above
7 and click on "order online" after selecting an edition.
8 All data tables of this document are available for the full time series 1990 through 2014, inclusive, at the internet site
9 mentioned above.
10
11 FOR FURTHER INFORMATION
12 Contact Mr. Leif Hockstad, Environmental Protection Agency, (202) 343-9432, hockstad.leif@epa.gov.
13 Or Ms. Melissa Weitz, Environmental Protection Agency, (202) 343-9897, weitz.melissa@epa.gov.
14 For more information regarding climate change and greenhouse gas emissions, see the EPA web site at
15 .
16
17 Released for printing: April 15, 2016
18
19
20
-------
Acknowledgments
2 The Environmental Protection Agency would like to acknowledge the many individual and organizational
3 contributors to this document, without whose efforts this report would not be complete. Although the complete list
4 of researchers, government employees, and consultants who have provided technical and editorial support is too
5 long to list here, EPA's Office of Atmospheric Programs would like to thank some key contributors and reviewers
6 whose work has significantly improved this year's report.
7 Work on emissions from fuel combustion was led by Leif Hockstad. Susan Burke and Amy Bunker directed the
8 work on mobile combustion and transportation. Work on industrial processes and product use emissions was led by
9 Mausami Desai. Work on fugitive methane emissions from the energy sector was directed by Melissa Weitz and
10 CateHight. Calculations for the waste sector were led by Rachel Schmeltz. Tom Wirth directed work on the
11 Agriculture and the Land Use, Land-Use Change, and Forestry chapters. Work on emissions of HFCs, PFCs, SF6,
12 and NF3 was directed by Deborah Ottinger and Dave Godwin.
13 Within the EPA, other Offices also contributed data, analysis, and technical review for this report. The Office of
14 Transportation and Air Quality and the Office of Air Quality Planning and Standards provided analysis and review
15 for several of the source categories addressed in this report. The Office of Solid Waste and the Office of Research
16 and Development also contributed analysis and research.
17 The Energy Information Administration and the Department of Energy contributed invaluable data and analysis on
18 numerous energy-related topics. The U.S. Forest Service prepared the forest carbon inventory, and the Department
19 of Agriculture's Agricultural Research Service and the Natural Resource Ecology Laboratory at Colorado State
20 University contributed leading research on nitrous oxide and carbon fluxes from soils.
21 Other government agencies have contributed data as well, including the U.S. Geological Survey, the Federal
22 Highway Administration, the Department of Transportation, the Bureau of Transportation Statistics, the Department
23 of Commerce, the National Agricultural Statistics Service, the Federal Aviation Administration, and the Department
24 of Defense.
25 We would also like to thank Marian Martin Van Pelt and the full Inventory team at ICF International including
26 Leslie Chinery, Randy Freed, Diana Pape, Robert Lanza, Lauren Marti, Mollie Averyt, Mark Flugge, Larry
27 O'Rourke, Deborah Harris, Dean Gouveia, Jonathan Cohen, Alexander Lataille, Andrew Pettit, Sabrina Andrews,
28 Marybeth Riley-Gilbert, Sarah Kolansky, David Towle, Bikash Acharya, Bobby Renz, Rebecca Ferenchiak, Kasey
29 Knoell, Cory Jemison, Kevin Kurkul, Matt Lichtash, Krisztina Pjeczka, Cecilia Bremner, and Gabrielle Jette for
30 synthesizing this report and preparing many of the individual analyses. Eastern Research Group, RTI International,
31 Raven Ridge Resources, and Ruby Canyon Engineering Inc. also provided significant analytical support.
32
-------
-------
Preface
2 The United States Environmental Protection Agency (EPA) prepares the official U.S. Inventory of Greenhouse Gas
3 Emissions and Sinks to comply with existing commitments under the United Nations Framework Convention on
4 Climate Change (UNFCCC). Under decision 3/CP.5 of the UNFCCC Conference of the Parties, national
5 inventories for UNFCCC Annex I parties should be provided to the UNFCCC Secretariat each year by April 15.
6 In an effort to engage the public and researchers across the country, the EPA has instituted an annual public review
7 and comment process for this document. The availability of the draft document is announced via Federal Register
8 Notice and is posted on the EPA web site. Copies are also mailed upon request. The public comment period is
9 generally limited to 30 days; however, comments received after the closure of the public comment period are
10 accepted and considered for the next edition of this annual report.
11
-------
-------
Table of Contents
2 ACKNOWLEDGMENTS I
3 PREFACE Ill
4 TABLE OF CONTENTS V
5 LIST OF TABLES, FIGURES, AND BOXES VIM
6 EXECUTIVE SUMMARY ES-1
7 ES.l. Background Information ES-2
8 ES.2. Recent Trends in U.S. Greenhouse Gas Emissions and Sinks ES-4
9 ES.3. Overview of Sector Emissions and Trends ES-17
10 ES.4. Other Information ES-22
11 1. INTRODUCTION 1-1
12 1.1 Background Information 1-2
13 1.2 National Inventory Arrangements 1-10
14 1.3 Inventory Process 1-13
15 1.4 Methodology andData Sources 1-14
16 1.5 Key Categories 1-15
17 1.6 Quality Assurance and Quality Control (QA/QC) 1-19
18 1.7 Uncertainty Analysis of Emission Estimates 1-21
19 1.8 Completeness 1-22
20 1.9 Organization of Report 1-22
21 2. TRENDS IN GREENHOUSE GAS EMISSIONS 2-1
22 2.1 Recent Trends in U.S. Greenhouse Gas Emissions and Sinks 2-1
23 2.2 Emissions by Economic Sector 2-23
24 2.3 Indirect Greenhouse Gas Emissions (CO, NOX, NMVOCs, and SO2) 2-34
25 3. ENERGY 3-1
26 3.1 Fossil Fuel Combustion (IPCC Source Category 1A) 3-4
27 3.2 Carbon Emitted from Non-Energy Uses of Fossil Fuels (IPCC Source Category 1A) 3-39
28 3.3 Incineration of Waste (IPCC Source Category lAla) 3-46
29 3.4 Coal Mining (IPCC Source Category IB la) (TO BE UPDATED) 3-50
-------
1 3.5 Abandoned Underground Coal Mines (IPCC Source Category IBla) (TO BE UPDATED) 3-54
2 3.6 Petroleum Systems (IPCC Source Category lB2a) 3-58
3 3.7 Natural Gas Systems (IPCC Source Category lB2b) 3-66
4 3.8 Energy Sources of Indirect Greenhouse Gas Emissions 3-77
5 3.9 International Bunker Fuels (IPCC Source Category 1: Memo Items) 3-78
6 3.10 WoodBiomass andEthanol Consumption (IPCC Source Category 1A) 3-83
7 4. INDUSTRIAL PROCESSES AND PRODUCT USE 4-1
8 4.1 Cement Production (IPCC Source Category 2A1) 4-6
9 4.2 Lime Production (IPCC Source Category 2A2) 4-9
10 4.3 Glass Production (IPCC Source Category 2A3) 4-14
11 4.4 Other Process Uses of Carbonates (IPCC Source Category 2A4) 4-17
12 4.5 Ammonia Production (IPCC Source Category 2B1) 4-20
13 4.6 Urea Consumption for Non-Agricultural Purposes 4-24
14 4.7 Nitric Acid Production (IPCC Source Category 2B2) 4-26
15 4.8 Adipic Acid Production (IPCC Source Category 2B3) 4-29
16 4.9 Silicon Carbide Production and Consumption (IPCC Source Category 2B5) 4-32
17 4.10 Titanium Dioxide Production (IPCC Source Category 2B6) 4-35
18 4.11 Soda Ash Production and Consumption (IPCC Source Category 2B7) 4-38
19 4.12 Petrochemical Production (IPCC Source Category 2B8) 4-41
20 4.13 HCFC-22 Production (IPCC Source Category 2B9a) (TO BE UPDATED) 4-46
21 4.14 CarbonDioxide Consumption (IPCC Source Category 2B10) 4-49
22 4.15 Phosphoric Acid Production (IPCC Source Category 2B10) 4-52
23 4.16 Iron and Steel Production (IPCC Source Category 2C1) and Metallurgical Coke Production 4-55
24 4.17 Ferroalloy Production (IPCC Source Category 2C2) 4-65
25 4.18 Aluminum Production (IPCC Source Category 2C3) (TO BE UPDATED) 4-68
26 4.19 Magnesium Production and Processing (IPCC Source Category 2C4) (TO BE UPDATED) 4-74
27 4.20 Lead Production (IPCC Source Category 2C5) 4-79
28 4.21 Zinc Production (IPCC Source Category 2C6) 4-81
29 4.22 Semiconductor Manufacture (IPCC Source Category 2E1) (TO BE UPDATED) 4-86
30 4.23 Substitution of Ozone Depleting Substances (IPCC Source Category 2F) 4-96
31 4.24 Electrical Transmission and Distribution (IPCC Source Category 2G1) (TO BE UPDATED) 4-103
32 4.25 Nitrous Oxide from Product Uses (IPCC Source Category 2G3) 4-111
33 4.26 Industrial Processes and Product Use Sources of Indirect Greenhouse Gases 4-113
34 5. AGRICULTURE 5-1
35 5.1 Enteric Fermentation (IPCC Source Category 3A) 5-2
36 5.2 Manure Management (IPCC Source Category 3B) 5-8
37 5.3 Rice Cultivation (IPCC Source Category 3C) (TO BE UPDATED) 5-15
vi DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 5.4 Agricultural Soil Management (IPCC Source Category 3D) 5-22
2 5.5 Field Burning of Agricultural Residues (IPCC Source Category 3F) 5-35
3 6. LAND USE, LAND-USE CHANGE, AND FORESTRY 6-1
4 6.1 Representation of the U.S. LandBase 6-5
5 6.2 Forest Land Remaining Forest Land 6-18
6 6.3 Land Converted to Forest Land (IPCC Source Category 4A2) (TO BE UPDATED) 6-34
7 6.4 Cropland Remaining Cropland (IPCC Source Category 4B1) 6-34
8 6.5 Land Converted to Cropland (IPCC Source Category 4B2) 6-47
9 6.6 Grassland Remaining Grassland (IPCC Source Category 4C1) 6-51
10 6.7 Land Converted to Grassland (IPCC Source Category 4C2) 6-56
11 6.8 Wetlands Remaining Wetlands (IPCC Source Category 4D1) 6-60
12 6.9 Land Converted to Wetlands (IPCC Source Category 4D1) (TO BE UPDATED) 6-66
13 6.10 Settlements Remaining Settlements 6-66
14 6.11 Land Converted to Settlements (IPCC Source Category 4E2) 6-74
15 6.12 Other (IPCC Source Category 4H) 6-74
16 7. WASTE 7-1
17 7.1 Landfills (IPCC Source Category 5A1) 7-3
18 7.2 Wastewater Treatment (IPCC Source Category 5D) (TO BE UPDATED) 7-16
19 7.3 Composting (IPCC Source Category 5B1) 7-30
20 7.4 Waste Incineration (IPCC Source Category 5C1) 7-34
21 7.5 Waste Sources of Indirect Greenhouse Gases 7-34
22 8. OTHER 8-1
23 9. RECALCULATIONS AND IMPROVEMENTS 9-1
24 10. REFERENCES 10-1
25
26
27
VII
-------
i List of Tables, Figures, and Boxes
2 Tables
3 Table ES-1: Global Warming Potentials (100-Year Time Horizon) Used in this Report ES-3
4 Table ES-2: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks (MMT CO2 Eq.) ES-5
5 Table ES-3: CO2 Emissions from Fossil Fuel Combustion by Fuel Consuming End-Use Sector (MMT CO2 Eq.).ES-
6 11
7 Table ES-4: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks by Chapter/IPCC Sector (MMT CO2 Eq.)
8 ES-17
9 Table ES-5: Emissions and Removals (Flux) from Land Use, Land-Use Change, and Forestry (MMT CO2 Eq.) ...ES-
10 21
11 Table ES-6: U.S. Greenhouse Gas Emissions Allocated to Economic Sectors (MMT CO2 Eq.) ES-23
12 Table ES-7: U.S Greenhouse Gas Emissions by Economic Sector with Electricity-Related Emissions Distributed
13 (MMTCO2Eq.) ES-24
14 Table ES-8: Recent Trends in Various U.S. Data (Index 1990 = 100) ES-25
15 Table 1-1: Global Atmospheric Concentration, Rate of Concentration Change, and Atmospheric Lifetime of
16 Selected Greenhouse Gases 1-4
17 Table 1-2: Global Warming Potentials and Atmospheric Lifetimes (Years) Used in this Report 1-8
18 Table 1-3: Comparison of 100-Year GWP values 1-9
19 Table 1-4: Key Categories for the United States (1990-2014) 1-16
20 Table 1-5: Estimated Overall Inventory Quantitative Uncertainty (MMT CO2 Eq. and Percent) (TO BE UPDATED)
21 1-21
22 Table 1-6: IPCC Sector Descriptions 1-22
23 Table 1-7: List of Annexes 1-24
24 Table 2-1: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks (MMT CO2 Eq.) 2-4
25 Table 2-2: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks (kt) 2-7
26 Table 2-3: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks by Chapter/IPCC Sector (MMT CO2 Eq.) 2-
27 9
28 Table 2-4: Emissions from Energy (MMT CO2 Eq.) 2-12
29 Table 2-5: CO2 Emissions from Fossil Fuel Combustion by End-Use Sector (MMT CO2 Eq.) 2-13
30 Table 2-6: Emissions from Industrial Processes and Product Use (MMT CO2Eq.) 2-16
31 Table 2-7: Emissions from Agriculture (MMT CO2 Eq.) 2-19
32 Table 2-8: Emissions and Removals (Flux) from Land Use, Land-Use Change, and Forestry (MMT CO2 Eq.)... 2-20
33 Table 2-9: Emissions from Waste (MMT CO2 Eq.) 2-22
34 Table 2-10: U.S. Greenhouse Gas Emissions Allocated to Economic Sectors (MMT CO2 Eq. and Percent of Total in
35 2014) 2-24
36 Table 2-11: Electricity Generation-Related Greenhouse Gas Emissions (MMT CO2 Eq.) 2-26
37 Table 2-12: U.S. Greenhouse Gas Emissions by Economic Sector and Gas with Electricity-Related Emissions
38 Distributed (MMT CO2Eq.) and Percent of Total in 2014 2-28
viii DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 2-13: Transportation-Related Greenhouse Gas Emissions (MMT CO2 Eq.) 2-30
2 Table 2-14: Recent Trends in Various U.S. Data (Index 1990 = 100) 2-33
3 Table 2-15: Emissions of NOX, CO, NMVOCs, and SO2 (kt) 2-35
4 Table 3-1: CO2, CH4, andN2O Emissions fromEnergy (MMT CO2 Eq.) 3-2
5 Table 3-2: CO2, CH4, and N2O Emissions from Energy (kt) 3-3
6 Table 3-3: CO2, CH4, and N2O Emissions from Fossil Fuel Combustion (MMT CO2 Eq.) 3-5
7 Table 3-4: CO2, CH4, and N2O Emissions from Fossil Fuel Combustion (kt) 3-5
8 Table 3-5: CO2 Emissions from Fossil Fuel Combustion by Fuel Type and Sector (MMT CO2 Eq.) 3-5
9 Table 3-6: Annual Change in CO2 Emissions and Total 2014 Emissions fromFossil Fuel Combustion for Selected
10 Fuels and Sectors (MMT CO2 Eq. and Percent) 3-6
11 Table 3-7: CO2, CH4, andN2O Emissions fromFossil Fuel Combustion by Sector (MMT CO2 Eq.) 3-10
12 Table 3-8: CO2, CH4, and N2O Emissions from Fossil Fuel Combustion by End-Use Sector (MMT CO2 Eq.).... 3-11
13 Table 3-9: CO2 Emissions from Stationary Fossil Fuel Combustion (MMT CO2 Eq.) 3-12
14 Table 3-10: CH4 Emissions from Stationary Combustion (MMT CO2 Eq.) 3-13
15 Table 3-11: N2O Emissions from Stationary Combustion (MMT CO2 Eq.) 3-13
16 Table 3-12: CO2 Emissions from Fossil Fuel Combustion in Transportation End-Use Sector (MMT CO2 Eq.)... 3-19
17 Table 3-13: CH4 Emissions fromMobile Combustion (MMT CO2 Eq.) 3-22
18 Table 3-14: N2O Emissions fromMobile Combustion (MMT CO2 Eq.) 3-23
19 Table 3-15: Carbon Intensity from Direct Fossil Fuel Combustion by Sector (MMT CO2 Eq./QBtu) 3-29
20 Table 3-16: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Energy-related Fossil Fuel
21 Combustion by Fuel Type and Sector (MMT CO2Eq. and Percent) 3-31
22 Table 3-17: Approach 2 Quantitative Uncertainty Estimates for CH4 and N2O Emissions from Energy-Related
23 Stationary Combustion, Including Biomass (MMT CO2Eq. and Percent) 3-35
24 Table 3-18: Approach 2 Quantitative Uncertainty Estimates for CH4 and N2O Emissions from Mobile Sources
25 (MMT CO2 Eq. and Percent) 3-38
26 Table 3-19: CO2 Emissions from Non-Energy Use Fossil Fuel Consumption (MMT CO2 Eq. and percent) 3-40
27 Table 3-20: Adjusted Consumption of Fossil Fuels for Non-Energy Uses (TBtu) 3-41
28 Table 3-21: 2014 Adjusted Non-Energy Use Fossil Fuel Consumption, Storage, and Emissions 3-41
29 Table 3-22: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Non-Energy Uses of Fossil
30 Fuels (MMT CO2Eq. andPercent) 3-43
31 Table 3 -23: Approach 2 Quantitative Uncertainty Estimates for Storage Factors of Non-Energy Uses of Fossil Fuels
32 (Percent) 3-43
3 3 Table 3 -24: CO2, CH4, and N2O Emissions from the Incineration of Waste (MMT CO2 Eq.) 3-47
34 Table 3-25: CO2, CH4, and N2O Emissions from the Incineration of Waste (kt) 3-47
35 Table 3-26: Municipal Solid Waste Generation (Metric Tons) and Percent Combusted (BioCycle data set) 3-48
36 Table 3-27: Approach 2 Quantitative Uncertainty Estimates for CO2 and N2O from the Incineration of Waste (MMT
37 CO2 Eq. and Percent) 3-49
38 Table 3-28: CH4 Emissions from Coal Mining (MMT CO2 Eq.) 3-51
39 Table 3-29: CH4 Emissions from Coal Mining (kt) 3-51
IX
-------
1 Table 3-30: Coal Production (kt) 3-52
2 Table 3-31: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Coal Mining (MMT CO2 Eq.
3 and Percent) 3-53
4 Table 3-32: CH4 Emissions from Abandoned Coal Mines (MMT CO2 Eq.) 3-55
5 Table 3-33: CH4 Emissions from Abandoned Coal Mines (kt) 3-55
6 Table 3-34: Number of Gassy Abandoned Mines Present in U.S. Basins, grouped by Class according to Post-
7 Abandonment State 3-57
8 Table 3-35: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Abandoned Underground Coal
9 Mines (MMT CO2 Eq. and Percent) 3-58
10 Table 3-36: CH4 Emissions from Petroleum Systems (MMT CO2Eq.) 3-60
11 Table 3-37: CH4 Emissions from Petroleum Systems (kt) 3-60
12 Table 3-38: CO2 Emissions from Petroleum Systems (MMT CO2Eq.) 3-60
13 Table 3-39: CO2 Emissions from Petroleum Systems (kt) 3-61
14 Table 3-40: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Petroleum Systems (MMT
15 CO2 Eq. and Percent) 3-63
16 Table 3-41: Potential Emissions from CO2 Capture and Transport (MMT CO2 Eq.) 3-65
17 Table 3-42: Potential Emissions from CO2 Capture and Transport (kt) 3-65
18 Table 3-43: CH4 Emissions from Natural Gas Systems (MMT CO2 Eq.)a 3-68
19 Table 3-44: CH4 Emissions from Natural Gas Systems (kt)a 3-68
20 Table 3 -45: Non-combustion CO2 Emissions from Natural Gas Systems (MMT CO2 Eq.) 3-69
21 Table 3-46: Non-combustion CO2 Emissions from Natural Gas Systems (kt) 3-69
22 Table 3-47: Approach 2 Quantitative Uncertainty Estimates for CH4 and Non-energy CO2 Emissions from Natural
23 Gas Systems (MMT CO2 Eq. and Percent) 3-72
24 Table 3-48: NOX, CO, and NMVOC Emissions from Energy-Related Activities (kt) 3-77
25 Table 3-49: CO2, CH4, and N2O Emissions from International Bunker Fuels (MMT CO2Eq.) 3-79
26 Table 3-50: CO2, CH4 and N2O Emissions from International Bunker Fuels (kt) 3-79
27 Table 3-51: Aviation CO2 and N2O Emissions for International Transport (MMT CO2 Eq.) 3 -80
28 Table 3 -52: Aviation Jet Fuel Consumption for International Transport (Million Gallons) 3-81
29 Table 3-53: Marine Fuel Consumption for International Transport (Million Gallons) 3-81
30 Table 3-54: CO2 Emissions from Wood Consumption by End-Use Sector (MMT CO2 Eq.) 3-83
31 Table 3-55: CO2 Emissions from Wood Consumption by End-Use Sector (kt) 3-83
32 Table 3-56: CO2 Emissions fromEthanol Consumption (MMT CO2 Eq.) 3-84
33 Table 3-57: CO2 Emissions fromEthanol Consumption (kt) 3-84
34 Table 3-58: Woody Biomass Consumption by Sector (Trillion Btu) 3-84
35 Table 3-59: Ethanol Consumption by Sector (Trillion Btu) 3-85
36 Table 4-1: Emissions from Industrial Processes and Product Use (MMT CO2Eq.) 4-3
37 Table 4-2: Emissions from Industrial Processes and Product Use (kt) 4-4
38 Table 4-3: CO2 Emissions from Cement Production (MMT CO2 Eq. and kt) 4-7
39 Table 4-4: Clinker Production (kt) 4-8
x DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 4-5: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Cement Production (MMT CO2
2 Eq. and Percent) 4-9
3 Table 4-6: CO2 Emissions from Lime Production (MMT CO2 Eq. and kt) 4-10
4 Table 4-7: Potential, Recovered, and Net CO2 Emissions from Lime Production (kt) 4-10
5 Table 4-8: High-Calcium- and Dolomitic-Quicklime, High-Calcium- and Dolomitic-Hydrated, and Dead-Burned-
6 Dolomite Lime Production (kt) 4-11
7 Table 4-9: Adjusted Lime Production (kt) 4-11
8 Table 4-10: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Lime Production (MMT CO2
9 Eq. and Percent) 4-13
10 Table 4-11: CO2 Emissions from Glass Production (MMT CO2 Eq. and kt) 4-14
11 Table 4-12: Limestone, Dolomite, and Soda Ash Consumption Used in Glass Production (kt) 4-15
12 Table 4-13: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Glass Production (MMT CO2
13 Eq. and Percent) 4-16
14 Table 4-14: CO2 Emissions from Other Process Uses of Carbonates (MMT CO2Eq.) 4-17
15 Table 4-15: CO2 Emissions from Other Process Uses of Carbonates (kt) 4-18
16 Table 4-16: Limestone andDolomite Consumption (kt) 4-19
17 Table 4-17: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Other Process Uses of
18 Carbonates (MMT CO2 Eq. and Percent) 4-20
19 Table 4-18: CO2 Emissions from Ammonia Production (MMT CO2 Eq.) 4-21
20 Table 4-19: CO2 Emissions from Ammonia Production (kt) 4-21
21 Table 4-20: Ammonia Production and Urea Production (kt) 4-22
22 Table 4-21: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Ammonia Production (MMT
23 CO2 Eq. and Percent) 4-23
24 Table 4-22: CO2 Emissions from Urea Consumption for Non-Agricultural Purposes (MMT CO2 Eq.) 4-24
25 Table 4-23: CO2 Emissions from Urea Consumption for Non-Agricultural Purposes (kt) 4-24
26 Table 4-24: Urea Production, Urea Applied as Fertilizer, Urea Imports, and Urea Exports (kt) 4-25
27 Table 4-25: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Urea Consumption for Non-
28 Agricultural Purposes (MMT CO2 Eq. and Percent) 4-25
29 Table 4-26: N2O Emissions from Nitric Acid Production (MMT CO2 Eq. and kt N2O) 4-27
30 Table 4-27: Nitric Acid Production (kt) 4-28
31 Table 4-28: Approach 2 Quantitative Uncertainty Estimates for N2O Emissions from Nitric Acid Production (MMT
32 CO2 Eq. and Percent) 4-29
33 Table 4-29: N2O Emissions from Adipic Acid Production (MMT CO2 Eq. and kt N2O) 4-30
34 Table 4-30: Adipic Acid Production (kt) 4-31
35 Table 4-31: Approach 2 Quantitative Uncertainty Estimates for N2O Emissions from Adipic Acid Production
36 (MMT CO2 Eq. and Percent) 4-32
37 Table 4-32: CO2 and QL; Emissions from Silicon Carbide Production and Consumption (MMT CO2 Eq.) 4-33
3 8 Table 4-33: CO2 and CH4 Emissions from Silicon Carbide Production and Consumption (kt) 4-33
39 Table 4-34: Production and Consumption of Silicon Carbide (Metric Tons) 4-34
XI
-------
1 Table 4-35: Approach 2 Quantitative Uncertainty Estimates for CH4 and CO2 Emissions from Silicon Carbide
2 Production and Consumption (MMT CO2Eq. and Percent) 4-34
3 Table 4-36: CO2 Emissions from Titanium Dioxide (MMT CO2 Eq. and kt) 4-36
4 Table 4-37: Titanium Dioxide Production (kt) 4-36
5 Table 4-38: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Titanium Dioxide Production
6 (MMT CO2 Eq. and Percent) 4-37
7 Table 4-39: CO2 Emissions from Soda Ash Production and Consumption Not Associated with Glass Manufacturing
8 (MMTCO2Eq.) 4-39
9 Table 4-40: CO2 Emissions from Soda Ash Production and Consumption Not Associated with Glass Manufacturing
10 (kt) 4-39
11 Table 4-41: Soda Ash Production and Consumption Not Associated with Glass Manufacturing (kt) 4-40
12 Table 4-42: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Soda Ash Production and
13 Consumption (MMT CO2 Eq. and Percent) 4-41
14 Table 4-43: CO2 and CH4 Emissions from Petrochemical Production (MMT CO2 Eq.) 4-43
15 Table 4-44: CO2 and CH4 Emissions from Petrochemical Production (kt) 4-43
16 Table 4-45: Production of Selected Petrochemicals (kt) 4-45
17 Table 4-46: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Petrochemical Production and
18 CO2 Emissions from Carbon Black Production (MMT CO2 Eq. and Percent) 4-45
19 Table 4-47: HFC-23 Emissions from HCFC-22 Production (MMT CO2 Eq. and kt HFC-23) 4-47
20 Table 4-48: HCFC-22 Production (kt) 4-48
21 Table 4-49: Approach 2 Quantitative Uncertainty Estimates for HFC-23 Emissions from HCFC-22 Production
22 (MMT CO2 Eq. and Percent) 4-48
23 Table 4-50: CO2 Emissions from CO2 Consumption (MMT CO2 Eq. and kt) 4-49
24 Table 4-51: CO2 Production (ktCO2) and the Percent Used for Non-EOR Applications 4-51
25 Table 4-52: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from CO2 Consumption (MMT CO2
26 Eq. and Percent) 4-51
27 Table 4-53: CO2 Emissions from Phosphoric Acid Production (MMT CO2 Eq. and kt) 4-52
28 Table 4-54: Phosphate Rock Domestic Consumption, Exports, and Imports (kt) 4-53
29 Table 4-55: Chemical Composition of Phosphate Rock (Percent by weight) 4-54
30 Table 4-56: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Phosphoric Acid Production
31 (MMT CO2 Eq. and Percent) 4-55
32 Table 4-57: CO2 Emissions from Metallurgical Coke Production (MMT CO2 Eq.) 4-56
33 Table 4-58: CO2 Emissions from Metallurgical Coke Production (kt) 4-57
34 Table 4-59: CO2 Emissions from Iron and Steel Production (MMT CO2 Eq.) 4-57
35 Table 4-60: CO2 Emissions from Iron and Steel Production (kt) 4-57
36 Table 4-61: CH4 Emissions from Iron and Steel Production (MMT CO2 Eq.) 4-57
37 Table 4-62: CH4 Emissions from Iron and Steel Production (kt) 4-58
38 Table 4-63: Material Carbon Contents for Metallurgical Coke Production 4-59
39 Table 4-64: Production and Consumption Data for the Calculation of CO2 and CH4 Emissions from Metallurgical
40 Coke Production (Thousand Metric Tons) 4-59
xii DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 4-65: Production and Consumption Data for the Calculation of CO2 Emissions from Metallurgical Coke
2 Production (Million ft3) 4-60
3 Table 4-66: CO2 Emission Factors for Sinter Production and Direct Reduced Iron Production 4-60
4 Table 4-67: Material Carbon Contents for Iron and Steel Production 4-60
5 Table 4-68: CH4 Emission Factors for Sinter and Pig Iron Production 4-61
6 Table 4-69: Production and Consumption Data for the Calculation of CO2 and CH4 Emissions from Iron and Steel
7 Production (Thousand Metric Tons) 4-62
8 Table 4-70: Production and Consumption Data for the Calculation of CO2 Emissions from Iron and Steel Production
9 (Million ft3 unless otherwise specified) 4-62
10 Table 4-71: Approach 2 Quantitative Uncertainty Estimates for CO2 and CH4 Emissions from Iron and Steel
11 Production and Metallurgical Coke Production (MMT CO2Eq. and Percent) 4-63
12 Table 4-72: CO2 and CH4 Emissions from Ferroalloy Production (MMT CO2 Eq.) 4-65
13 Table 4-73: CO2 and CH4 Emissions from Ferroalloy Production (kt) 4-66
14 Table 4-74: Production of Ferroalloys (Metric Tons) 4-67
15 Table 4-75: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Ferroalloy Production (MMT
16 CO2 Eq. and Percent) 4-68
17 Table 4-76: CO2 Emissions from Aluminum Production (MMT CO2 Eq. and kt) 4-69
18 Table 4-77: PFC Emissions from Aluminum Production (MMT CO2Eq.) 4-69
19 Table 4-78: PFC Emissions from Aluminum Production (kt) 4-70
20 Table 4-79: Production of Primary Aluminum (kt) 4-72
21 Table 4-80: Approach 2 Quantitative Uncertainty Estimates for CO2 and PFC Emissions from Aluminum
22 Production (MMT CO2 Eq. and Percent) 4-73
23 Table 4-81: SFe, HFC-134a, FK 5-1-12 and CO2 Emissions from Magnesium Production and Processing (MMT
24 CO2Eq.) 4-74
25 Table 4-82: SF6, HFC-134a, FK 5-1-12 and CO2 Emissions from Magnesium Production and Processing (kt)... 4-74
26 Table 4-83: SF6 Emission Factors (kg SF6 per metric ton of magnesium) 4-76
27 Table 4-84: Approach 2 Quantitative Uncertainty Estimates for SF6, HFC-134a and CO2 Emissions from
28 Magnesium Production and Processing (MMT CO2Eq. and Percent) 4-78
29 Table 4-85: CO2 Emissions from Lead Production (MMT CO2 Eq. and kt) 4-79
30 Table 4-86: Lead Production (Metric Tons) 4-80
31 Table 4-87: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Lead Production (MMT CO2
32 Eq. and Percent) 4-81
33 Table 4-88: Zinc Production (Metric Tons) 4-83
34 Table 4-89:CO2 Emissions from Zinc Production (MMT CO2Eq. and kt) 4-83
35 Table 4-90: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Zinc Production (MMT CO2
36 Eq. and Percent) 4-85
37 Table 4-91: PFC, HFC, SF6, NF3, and N2O Emissions from Semiconductor Manufacture (MMT CO2 Eq.) 4-87
3 8 Table 4-92: PFC, HFC, SF6, NF3, and N2O Emissions from Semiconductor Manufacture (kt) 4-87
39 Table 4-93: Approach 2 Quantitative Uncertainty Estimates for HFC, PFC, SF6, NF3 and N2O Emissions from
40 Semiconductor Manufacture (MMT CO2Eq. andPercent) 4-94
41 Table 4-94: Emissions of HFCs and PFCs from ODS Substitutes (MMT CO2 Eq.) 4-96
xiii
-------
1 Table 4-95: Emissions of HFCs and PFCs from ODS Substitution (Metric Tons) 4-96
2 Table 4-96: Emissions of HFCs and PFCs from ODS Substitutes (MMT CO2 Eq.) by Sector 4-97
3 Table 4-97: Approach 2 Quantitative Uncertainty Estimates for HFC and PFC Emissions from ODS Substitutes
4 (MMT CO2 Eq. and Percent) 4-99
5 Table 4-98: U.S. HFC Consumption (MMT CO2 Eq.) 4-101
6 Table 4-99: Averaged U.S. HFC Demand (MMT CO2 Eq.) 4-101
7 Table 4-100: SF6 Emissions from Electric Power Systems and Electrical Equipment Manufacturers (MMT CO2 Eq.)
8 4-103
9 Table 4-101: SF6 Emissions from Electric Power Systems and Electrical Equipment Manufacturers (kt) 4-103
10 Table 4-102: Transmission Mile Coverage and Regression Coefficients for Large and Non-Large Utilities, Percent
11 4-107
12 Table 4-103: Approach 2 Quantitative Uncertainty Estimates for SF6 Emissions from Electrical Transmission and
13 Distribution (MMT CO2 Eq. and Percent) 4-108
14 Table 4-104: 2013 Potential and Actual Emissions of HFCs, PFCs, SF6, and NF3 from Selected Sources (MMT CO2
15 Eq.) 4-110
16 Table 4-105: N2O Production (kt) 4-111
17 Table 4-106: N2O Emissions from N2O Product Usage (MMT CO2 Eq. and kt) 4-112
18 Table 4-107: Approach 2 Quantitative Uncertainty Estimates for N2O Emissions from N2O Product Usage (MMT
19 CO2 Eq. and Percent) 4-113
20 Table 4-108: NOX, CO, and NMVOC Emissions from Industrial Processes and Product Use (kt) 4-114
21 Table 5-1: Emissions from Agriculture (MMT CO2Eq.) 5-2
22 Table 5-2: Emissions from Agriculture (kt) 5-2
23 Table 5-3: CH4 Emissions from Enteric Fermentation (MMT CO2 Eq.) 5-3
24 Table 5-4: CH4 Emissions from Enteric Fermentation (kt) 5-3
25 Table 5-5: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Enteric Fermentation (MMT
26 CO2 Eq. and Percent) 5-6
27 Table 5-6: CH4 andN2O Emissions from Manure Management (MMT CO2Eq.) 5-9
28 Table 5-7: CH4 and N2O Emissions fromManure Management (kt) 5-10
29 Table 5-8: Approach 2 Quantitative Uncertainty Estimates for CH4 and N2O (Direct and Indirect) Emissions from
30 Manure Management (MMT CO2Eq. and Percent) 5-13
31 Table 5-9: 2006 IPCC Implied Emission Factor Default Values Compared with Calculated Values for CH4 from
32 Manure Management (kg/head/year) 5-14
33 Table 5-10: CH4 Emissions from Rice Cultivation (MMT CO2 Eq.) 5-16
34 Table 5-11: CH4 Emissions from Rice Cultivation (kt) 5-17
35 Table 5-12: Rice Area Harvested (Hectare) 5-18
36 Table 5-13: Ratooned Area as Percent of Primary Growth Area 5-18
37 Table 5-14: Non-USDA Data Sources for Rice Harvest Information 5-19
38 Table 5-15: Non-California Seasonal Emission Factors (kg CHVhectare/season) 5-19
39 Table 5-16: California Emission Factors (kg CHVhectare/year or season) 5-20
xiv DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 5-17: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Rice Cultivation (MMT CO2
2 Eq. and Percent) 5-21
3 Table 5-18: N2O Emissions from Agricultural Soils (MMT CO2 Eq.) 5-24
4 Table 5-19: N2O Emissions from Agricultural Soils (kt) 5-24
5 Table 5-20: Direct N2O Emissions from Agricultural Soils by Land Use Type and N Input Type (MMT CO2 Eq.) 5-
6 24
7 Table 5-21: Indirect N2O Emissions from Agricultural Soils (MMT CO2Eq.) 5-25
8 Table 5-22: Quantitative Uncertainty Estimates of N2O Emissions from Agricultural Soil Management in 2014
9 (MMT CO2 Eq. and Percent) 5-32
10 Table 5-23: CH4 and N2O Emissions from Field Burning of Agricultural Residues (MMT CO2 Eq.) 5-35
11 Table 5-24: CH4, N2O, CO, and NOX Emissions from Field Burning of Agricultural Residues (kt) 5-35
12 Table 5-25: Agricultural Crop Production (kt of Product) 5-38
13 Table 5-26: U.S. Average Percent Crop Area Burned by Crop (Percent) 5-38
14 Table 5-27: Key Assumptions for Estimating Emissions from Field Burning of Agricultural Residues 5-38
15 Table 5-28: Greenhouse Gas Emission Ratios and Conversion Factors 5-38
16 Table 5-29: Approach 2 Quantitative Uncertainty Estimates for CH4 and N2O Emissions from Field Burning of
17 Agricultural Residues (MMT CO2Eq. and Percent) 5-39
18 Table 6-1: Emissions and Removals (Flux) from Land Use, Land-Use Change, and Forestry by Land Use and Land-
19 Use Change Category (MMT CO2 Eq.) 6-2
20 Table 6-2: Emissions and Removals (Flux) from Land Use, Land-Use Change, and Forestry by Gas (MMT CO2
21 Eq.) 6-3
22 Table 6-3: Emissions and Removals (Flux) from Land Use, Land-Use Change, and Forestry by Gas (kt) 6-4
23 Table 6-4: Managed and Unmanaged Land Area by Land-Use Categories for All 50 States (Thousands of Hectares)
24 6-6
25 Table 6-5: Land Use and Land-Use Change for the U.S. Managed Land Base for All 50 States (Thousands of
26 Hectares) 6-6
27 Table 6-6: Data Sources Used to Determine Land Use and Land Area for the Conterminous United States, Hawaii,
28 and Alaska 6-11
29 Table 6-7: Total Land Area (Hectares) by Land-Use Category for U.S. Territories 6-17
30 Table 6-8: Net CO2 Flux from Forest Pools in Forest Land Remaining Forest Land and Harvested Wood Pools.
31 (MMTCO2Eq.) 6-21
32 Table 6-9: Net C Flux from Forest Pools in Forest Land Remaining Forest Land and Harvested Wood Pools (MMT
33 C) 6-21
34 Table 6-10: Forest area (1,000 ha) and C Stocks in Forest Land Remaining Forest Land and Harvested Wood Pools
35 (MMTC) 6-22
36 Table 6-11: Estimates of CO2 (MMT/yr) Emissions from Forest Fires in the Conterminous 48 States and Alaska3. 6-
37 24
38 Table 6-12: Quantitative Uncertainty Estimates for Net CO2 Flux from Forest Land Remaining Forest Land:
39 Changes in Forest C Stocks (MMT CO2Eq. and Percent) 6-28
40 Table 6-13: Estimated Non-CO2 Emissions from Forest Fires (MMT CO2Eq.) for U.S. Forests 6-30
41 Table 6-14: Estimated Non-CO2 Emissions from Forest Fires (kt) for U.S. Forests 6-30
xv
-------
1 Table 6-15: Quantitative Uncertainty Estimates of Non-CO2 Emissions from Forest Fires in Forest Land Remaining
1 Forest Land (MMT CO2Eq. and Percent) 6-31
3 Table 6-16: N2O Fluxes from Soils in Forest Land Remaining Forest Land (MMT CO2 Eq. and kt N2O) 6-32
4 Table 6-17: Quantitative Uncertainty Estimates of N2O Fluxes from Soils in Forest Land Remaining Forest Land
5 (MMT CO2 Eq. and Percent) 6-33
6 Table 6-18: Net CO2 Flux from Soil C Stock Changes in Cropland Remaining Cropland (MMT CO2 Eq.) 6-35
7 Table 6-19: Net CO2 Flux from Soil C Stock Changes in Cropland Remaining Cropland (MMT C) 6-36
8 Table 6-20: Approach 2 Quantitative Uncertainty Estimates for Soil C Stock Changes occurring within Cropland
9 Remaining Cropland (MMT CO2Eq. and Percent) 6-40
10 Table 6-21: Emissions from Liming (MMT CO2 Eq.) 6-42
11 Table 6-22: Emissions from Liming (MMT C) 6-42
12 Table 6-23: Applied Minerals (MMT) 6-43
13 Table 6-24: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Liming (MMT CO2 Eq. and
14 Percent) 6-44
15 Table 6-25: CO2 Emissions from Urea Fertilization (MMT CO2 Eq.) 6-44
16 Table 6-26: CO2 Emissions from Urea Fertilization (MMT C) 6-45
17 Table 6-27: Applied Urea (MMT) 6-46
18 Table 6-28: Quantitative Uncertainty Estimates for CO2 Emissions from Urea Fertilization (MMT CO2 Eq. and
19 Percent) 6-46
20 Table 6-29: Net CO2 Flux from Soil C Stock Changes in Land Converted to Cropland by Land Use Change
21 Category (MMT CO2 Eq.) 6-47
22 Table 6-30: Net CO2 Flux from Soil C Stock Changes in Land Converted to Cropland (MMT C) 6-48
23 Table 6-31: Approach 2 Quantitative Uncertainty Estimates for Soil C Stock Changes occurring within Land
24 Converted to Cropland (MMT CO2 Eq. and Percent) 6-50
25 Table 6-32: Net CO2 Flux from Soil C Stock Changes in Grassland Remaining Grassland (MMT CO2 Eq.) 6-52
26 Table 6-33: Net CO2 Flux from Soil C Stock Changes in Grassland Remaining Grassland (MMT C) 6-52
27 Table 6-34: Approach 2 Quantitative Uncertainty Estimates for C Stock Changes Occurring Within Grassland
28 Remaining Grassland (MMT CO2Eq. and Percent) 6-55
29 Table 6-35: Net CO2 Flux from Soil C Stock Changes for Land Converted to Grassland (MMT CO2 Eq.) 6-56
30 Table 6-36: Net CO2 Flux from Soil C Stock Changes for Land Converted to Grassland (MMT C) 6-57
31 Table 6-37: Approach 2 Quantitative Uncertainty Estimates for Soil C Stock Changes occurring within Land
32 Converted to Grassland (MMT CO2 Eq. and Percent) 6-59
33 Table 6-38: Emissions from PeatlandsRemaining Peatlands (MMT CO2 Eq.) 6-61
34 Table 6-39: Emissions from Peatlands Remaining Peatlands (kt) 6-61
35 Table 6-40: Peat Production of Lower 48 States (kt) 6-63
36 Table 6-41: Peat Production of Alaska (Thousand Cubic Meters) 6-63
37 Table 6-42: Approach 2 Quantitative Uncertainty Estimates for CO2, CEL, and N2O Emissions from Peatlands
38 Remaining Peatlands (MMT CO2Eq. and Percent) 6-64
39 Table 6-43: Net C Flux from Urban Trees (MMT CO2 Eq. and MMT C) 6-67
40 Table 6-44: Annual C Sequestration (Metric Tons C/yr), Tree Cover (Percent), and Annual C Sequestration per
41 Area of Tree Cover (kg C/m2-yr) for 50 states plus the District of Columbia (2014) 6-69
xvi DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 6-45: Approach 2 Quantitative Uncertainty Estimates for Net C Flux from Changes in C Stocks in Urban
2 Trees (MMT CO2 Eq. and Percent) 6-71
3 Table 6-46: N2O Fluxes from Soils in Settlements Remaining Settlements (MMT CO2 Eq. and kt N2O) 6-72
4 Table 6-47: Quantitative Uncertainty Estimates of N2O Emissions from Soils in Settlements Remaining Settlements
5 (MMT CO2 Eq. and Percent) 6-73
6 Table 6-48: Net Changes in Yard Trimming and Food Scrap Carbon Stocks in Landfills (MMT CO2 Eq.) 6-75
7 Table 6-49: Net Changes in Yard Trimming and Food Scrap Carbon Stocks in Landfills (MMT C) 6-75
8 Table 6-50: Moisture Contents, C Storage Factors (Proportions of Initial C Sequestered), Initial C Contents, and
9 Decay Rates for Yard Trimmings and Food Scraps in Landfills 6-77
10 Table 6-51: C Stocks in Yard Trimmings and Food Scraps in Landfills (MMT C) 6-78
11 Table 6-52: Approach 2 Quantitative Uncertainty Estimates for CO2 Flux from Yard Trimmings and Food Scraps in
12 Landfills (MMT CO2 Eq. and Percent) 6-78
13 Table 7-1: Emissions from Waste (MMT CO2 Eq.) 7-2
14 Table 7-2: Emissions from Waste (kt) 7-2
15 Table 7-3: CH4 Emissions from Landfills (MMT CO2 Eq.) 7-5
16 Table 7-4: CH4 Emissions from Landfills (kt) 7-5
17 Table 7-5: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Landfills (MMT CO2 Eq. and
18 Percent) 7-10
19 Table 7-6: Materials Discarded in the Municipal Waste Stream by Waste Type from 1990 to 2013 (Percent) 7-14
20 Table 7-7: CH4 and N2O Emissions from Domestic and Industrial Wastewater Treatment (MMT CO2 Eq.) 7-17
21 Table 7-8: CH4 and N2O Emissions from Domestic and Industrial Wastewater Treatment (kt) 7-17
22 Table 7-9: U.S. Population (Millions) and Domestic Wastewater BOD5 Produced (kt) 7-19
23 Table 7-10: Domestic Wastewater CH4 Emissions from Septic and Centralized Systems (2013) 7-20
24 Table 7-11: Industrial Wastewater CH4 Emissions by Sector (2013) 7-20
25 Table 7-12: U.S. Pulp and Paper, Meat, Poultry, Vegetables, Fruits and Juices, Ethanol, and Petroleum Refining
26 Production (MMT) 7-20
27 Table 7-13: Variables Used to Calculate Percent Wastewater Treated Anaerobically by Industry (percent) 7-22
28 Table 7-14: Wastewater Flow (m3/ton) and BOD Production (g/L) for U.S. Vegetables, Fruits, and Juices Production
29 7-23
30 Table 7-15: U.S. Population (Millions), Population Served by Biological Denitrification (Millions), Fraction of
31 Population Served by Wastewater Treatment (percent), Available Protein (kg/person-year), Protein Consumed
32 (kg/person-year), and Nitrogen Removed with Sludge (kt-N/year) 7-26
33 Table 7-16: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Wastewater Treatment (MMT
34 CO2 Eq. and Percent) 7-27
35 Table 7-17: CH4 and N2O Emissions from Composting (MMT CO2 Eq.) 7-31
36 Table 7-18: CH4 and N2O Emissions from Composting (kt) 7-31
37 Table 7-19: U.S. Waste Composted (kt) 7-33
38 Table 7-20: Approach 1 Quantitative Uncertainty Estimates for Emissions from Composting (MMT CO2 Eq. and
39 Percent) 7-33
40 Table 7-21: Emissions of NOX, CO, and NMVOC from Waste (kt) 7-35
41 Table 9-1: Revisions to U.S. Greenhouse Gas Emissions (MMT CO2 Eq.) 9-4
xvii
-------
1 Table 9-2: Revisions to Total Net Flux from Land Use, Land-Use Change, and Forestry (MMT CO2 Eq.) 9-6
2 Figures
3 Figure ES-1: U.S. Greenhouse Gas Emissions by Gas (MMT CO2Eq.) ES-4
4 Figure ES-2: Annual Percent Change in U.S. Greenhouse Gas Emissions ES-5
5 Figure ES-3: Annual Greenhouse Gas Emissions Relative to 1990(1990=0) ES-5
6 Figure ES-4: 2014 Greenhouse Gas Emissions by Gas (Percentages based on MMT CO2Eq.) ES-8
7 Figure ES-5: 2014 Sources of CO2 Emissions ES-9
8 Figure ES-6: 2014 CO2 Emissions from Fossil Fuel Combustion by Sector and Fuel Type (MMT CO2 Eq.) ES-10
9 Figure ES-7: 2014 End-Use Sector Emissions of CO2 from Fossil Fuel Combustion (MMT CO2Eq.) ES-10
10 Figure ES-8: 2014 Sources of CH4 Emissions (MMT CO2Eq.) ES-13
11 Figure ES-9: 2014 Sources of N2O Emissions (MMT CO2Eq.) ES-15
12 Figure ES-10: 2014 Sources of HFCs, PFCs, SF6, and NF3 Emissions (MMT CO2 Eq.) ES-16
13 Figure ES-11: U.S. Greenhouse Gas Emissions and Sinks by Chapter/IPCC Sector (MMT CO2 Eq.) ES-17
14 Figure ES-12: 2014 U.S. Energy Consumption by Energy Source ES-19
15 Figure ES-13: Emissions Allocated to Economic Sectors (MMT CO2Eq.) ES-23
16 Figure ES-14: Emissions with Electricity Distributed to Economic Sectors (MMT CO2Eq.) ES-25
17 Figure ES-15: U.S. Greenhouse Gas Emissions Per Capita and Per Dollar of Gross Domestic Product ES-26
18 Figure ES-16: 2014 Key Categories (MMT CO2 Eq.) ES-27
19 Figure 1-1
20 Figure 1-2
21 Figure 2-1
22 Figure 2-2
23 Figure 2-3
24 Figure 2-4
25 Figure 2-5
26 Figure 2-6
27 Figure 2-7
28 Figure 2-8
29 Figure 2-9
National Inventory Arrangements Diagram 1-11
U.S.QA/QC Plan Summary 1-20
U.S. Greenhouse Gas Emissions by Gas (MMT CO2 Eq.) 2-1
Annual Percent Change in U.S. Greenhouse Gas Emissions 2-2
Cumulative Change in Annual U.S. Greenhouse Gas Emissions Relative to 1990 (MMT CO2 Eq.).... 2-2
U.S. Greenhouse Gas Emissions and Sinks by Chapter/IPCC Sector (MMT CO2 Eq.) 2-9
2014 Energy Chapter Greenhouse Gas Sources (MMT CO2 Eq.) 2-11
2014 U.S. Fossil Carbon Flows (MMT CO2Eq.) 2-12
2014 CO2 Emissions from Fossil Fuel Combustion by Sector and Fuel Type (MMT CO2Eq.) 2-14
2014 End-Use Sector Emissions of CO2 from Fossil Fuel Combustion (MMT CO2 Eq.) 2-14
2014 Industrial Processes and Product Use Chapter Greenhouse Gas Sources (MMT CO2 Eq.) 2-16
30 Figure 2-10: 2014 Agriculture Chapter Greenhouse Gas Sources (MMT CO2 Eq.) 2-18
31 Figure 2-11: 2014 Waste Chapter Greenhouse Gas Sources (MMT CO2Eq.) 2-22
32 Figure 2-12: Emissions Allocated to Economic Sectors (MMT CO2 Eq.) 2-24
33 Figure 2-13: Emissions with Electricity Distributed to Economic Sectors (MMT CO2Eq.) 2-27
34 Figure 2-14: U.S. Greenhouse Gas Emissions Per Capita and Per Dollar of Gross Domestic Product 2-34
35 Figure 3-1: 2014 Energy Chapter Greenhouse Gas Sources (MMT CO2Eq.) 3-1
36 Figure 3-2: 2014 U.S. Fossil Carbon Flows (MMT CO2Eq.) 3-2
xviii DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Figure 3-3: 2014 U.S. Energy Consumption by Energy Source (Percent) 3-7
2 Figure 3-4: U.S. Energy Consumption (Quadrillion Btu) 3-7
3 Figure 3-5: 2014 CO2 Emissions from Fossil Fuel Combustion by Sector and Fuel Type (MMT CO2Eq.) 3-8
4 Figure 3-6: Annual Deviations from Normal Heating Degree Days for the United States (1950-2014) 3-9
5 Figure 3-7: Annual Deviations from Normal Cooling Degree Days for the United States (1950-2014) 3-9
6 Figure 3-8: Nuclear, Hydroelectric, and Wind Power Plant Capacity Factors in the United States (1990-2014).. 3-10
7 Figure 3-9: Electricity Generation Retail Sales by End-Use Sector (Billion kWh) 3-14
8 Figure 3-10: Industrial Production Indices (Index 2007= 100) 3-16
9 Figure 3-11: Sales-Weighted Fuel Economy of New Passenger Cars and Light-Duty Trucks, 1990-2014
10 (miles/gallon) 3-19
11 Figure 3-12: Sales of New Passenger Cars and Light-Duty Trucks, 1990-2014 (Percent) 3-19
12 Figure 3-13: Mobile Source CH4 and N2O Emissions (MMT CO2 Eq.) 3-22
13 Figure 3-14: U.S. Energy Consumption and Energy-Related CO2 Emissions Per Capita and Per Dollar GDP 3-29
14 Figure 4-1: 2014 Industrial Processes and Product Use Chapter Greenhouse Gas Sources (MMT CO2Eq.) 4-2
15 Figure 5-1: 2014 Agriculture Chapter Greenhouse Gas Emission Sources (MMT CO2Eq.) 5-1
16 Figure 5-2: Sources and Pathways of N that Result in N2O Emissions from Agricultural Soil Management 5-23
17 Figure 5-3: Crops, Annual Direct N2O Emissions Estimated Using the Tier 3 DAYCENT Model, 1990-2014 (MMT
18 CO2 Eq./year) (TO BE UPDATED) 5-25
19 Figure 5-4: Grasslands, Annual Direct N2O Emissions Estimated Using the Tier 3 DAYCENT Model, 1990-2014
20 (MMT CO2 Eq./year) (TO BE UPDATED) 5-25
21 Figure 5-5: Crops, Average Annual N Losses Leading to Indirect N2O Emissions Estimated Using the Tier 3
22 DAYCENT Model, 1990-2014 (ktN/year) (TO BE UPDATED) 5-25
23 Figure 5-6: Grasslands, Average Annual N Losses Leading to Indirect N2O Emissions Estimated Using the Tier 3
24 DAYCENT Model, 1990-2014 (ktN/year) (TO BE UPDATED) 5-26
25 Figure 5-7: Comparison of Measured Emissions at Field Sites and Modeled Emissions Using the DAYCENT
26 Simulation Model and IPCC Tier 1 Approach 5-33
27 Figure 6-1: Percent of Total Land Area for Each State in the General Land-Use Categories for 2014 (TO BE
28 UPDATED) 6-7
29 Figure 6-2: Changes in Forest Area by Region for Forest Land Remaining Forest Land in the conterminous United
30 States and coastal Alaska (1990-2014) 6-20
31 Figure 6-3: Forest Ecosystem C Stock Change by Region in Forest Land Remaining Forest Land in the
32 conterminous U.S. and coastal Alaska (1990-2014) 6-23
33 Figure 6-4: Estimated Net Annual Changes in C Stocks for Major C Pools in Forest Land Remaining Forest Land in
34 the Conterminous U.S. and Coastal Alaska (MMT C/year) 6-23
35 Figure 6-5: Total Net Annual CO2 Flux for Mineral Soils under Agricultural Management within States, 2014,
36 Cropland Remaining Cropland (TO BE UPDATED) 6-36
37 Figure 6-6: Total Net Annual CO2 Flux for Organic Soils under Agricultural Management within States, 2014,
38 Cropland Remaining Cropland (TO BE UPDATED) 6-36
39 Figure 6-7: Total Net Annual CO2 Flux for Mineral Soils under Agricultural Management within States, 2014, Land
40 Converted to Cropland (TO BE UPDATED) 6-48
41 Figure 6-8: Total Net Annual CO2 Flux for Organic Soils under Agricultural Management within States, 2014, Land
42 Converted to Cropland (TO BE UPDATED) 6-48
xix
-------
1 Figure 6-9: Total Net Annual CO2 Flux for Mineral Soils under Agricultural Management within States, 2014,
2 Grassland Remaining Grassland (TO BE UPDATED) 6-52
3 Figure 6-10: Total Net Annual CO2 Flux for Organic Soils under Agricultural Management within States, 2014,
4 Grassland Remaining Grassland (TO BE UPDATED) 6-52
5 Figure 6-11: Total Net Annual CO2 Flux for Mineral Soils under Agricultural Management within States, 2014,
6 Land Converted to Grassland (TO BE UPDATED) 6-57
7 Figure 6-12: Total Net Annual CO2 Flux for Organic Soils under Agricultural Management within States, 2014,
8 Land Converted to Grassland (TO BE UPDATED) 6-57
9 Figure 7-1: 2014 Waste Chapter Greenhouse Gas Sources (MMT CO2 Eq.) 7-1
10 Figure 7-2: Management of Municipal Solid Waste in the United States, 2013 7-13
11 Figure 7-3: MSW Management Trends from 1990 to 2013 (Million Tons) 7-13
12 Figure 7-4: Percent of Recovered Degradable Materials from 1990 to 2013 (Percent) 7-15
13 Figure 7-5: CH4 and N2O Emitted from Composting Operations Between 1990 and 2014 (kt) 7-32
14 Boxes
15 BoxES-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Sinks 1
16 BoxES-2: Recent Trends in Various U.S. Greenhouse Gas Emissions-Related Data 25
17 BoxES- 3: Recalculations of Inventory Estimates 28
18 Box 1-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Sinks 1-2
19 Box 1-2: The WCC Fifth Assessment Report and Global Warming Potentials 1-9
20 Box 1-3 :IPCC Reference Approach 1-15
21 Box 2-1: Methodology for Aggregating Emissions by Economic Sector 2-31
22 Box 2-2: Recent Trends in Various U.S. Greenhouse Gas Emissions-Related Data 2-33
23 Box2-3: Sources and Effects of Sulfur Dioxide 2-35
24 Box 3-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Sinks 3-3
25 Box 3-2: Energy Data from the Greenhouse Gas Reporting Program 3-4
26 Box 3 -3: Weather and Non-Fossil Energy Effects on CO2 from Fossil Fuel Combustion Trends 3-8
27 Box 3-4: Uses of Greenhouse Gas Reporting Program Data and Improvements in Reporting Emissions from
28 Industrial Sector Fossil Fuel Combustion 3-27
29 Box 3-5: Carbon Intensity of U.S. Energy Consumption 3-28
30 Box 3-6: Reporting of Lubricants, Waxes, and Asphalt and Road Oil Product Use in Energy Sector 3-45
31 Box 3-7: Carbon Dioxide Transport, Injection, and Geological Storage (TO BE UPDATED) 3-65
32 Box 4-1: Industrial Processes Data from EPA's Greenhouse Gas Reporting Program 4-6
33 Box 4-2: Potential Emission Estimates of HFCs, PFCs, SF6, andNF3 4-109
34 Box 5-1: Comparison of the U.S. Inventory Seasonal Emission Factors and IPCC (1996) Default Emission Factors
35 5-19
36 Box 5-2: Tier 1 vs. Tier 3 Approach for Estimating N2O Emissions 5-27
37 Box 5-3: Comparison of Tier 2 U.S. Inventory Approach and IPCC (2006) Default Approach 5-36
38 Box 6-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Sinks 6-4
xx DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Box 6-2: Preliminary Estimates of Land Use in U.S. Territories 6-17
2 Box 6-3: CCh Emissions from Forest Fires 6-24
3 Box 6-4: Tier 3 Approach for Soil C Stocks Compared to Tier 1 or 2 Approaches 6-38
4 Box 6-5: Comparison of the Tier 2 U. S. Inventory Approach and IPCC (2006) Default Approach 6-42
5 Box 6-6: Progress on Inclusion of Managed Coastal Wetlands in the U.S. Greenhouse Gas Inventory 6-65
6 Box 7-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Sinks 7-1
7 Box 7-2: Waste Data from the Greenhouse Gas Reporting Program 7-3
8 Box 7-3: Nationwide Municipal Solid Waste Data Sources 7-12
9 Box 7-4: Overview of the Waste Sector 7-13
10 Box 7-5: Description of a Modern, Managed Landfill 7-15
11 Box 7-6: Biogenic Wastes in Landfills 7-16
12
XXI
-------
-------
24
25
Executive Summary
2 An emissions inventory that identifies and quantifies a country's primary anthropogenic1 sources and sinks of
3 greenhouse gases is essential for addressing climate change. This inventory adheres to both (1) a comprehensive
4 and detailed set of methodologies for estimating sources and sinks of anthropogenic greenhouse gases, and (2) a
5 common and consistent mechanism that enables Parties to the United Nations Framework Convention on Climate
6 Change (UNFCCC) to compare the relative contribution of different emission sources and greenhouse gases to
7 climate change.
8 In 1992, the United States signed and ratified the UNFCCC. As stated in Article 2 of the UNFCCC, "The ultimate
9 objective of this Convention and any related legal instruments that the Conference of the Parties may adopt is to
10 achieve, in accordance with the relevant provisions of the Convention, stabilization of greenhouse gas
11 concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the
12 climate system. Such a level should be achieved within a time-frame sufficient to allow ecosystems to adapt
13 naturally to climate change, to ensure that food production is not threatened and to enable economic development to
14 proceed in a sustainable manner."2
15 Parties to the Convention, by ratifying, "shall develop, periodically update, publish and make available... national
16 inventories of anthropogenic emissions by sources and removals by sinks of all greenhouse gases not controlled by
17 the Montreal Protocol, using comparable methodologies..."3 The United States views this report as an opportunity
18 to fulfill these commitments.
19 This chapter summarizes the latest information on U.S. anthropogenic greenhouse gas emission trends from 1990
20 through 2014. To ensure that the U.S. emissions inventory is comparable to those of other UNFCCC Parties, the
21 estimates presented here were calculated using methodologies consistent with those recommended in the 2006
22 Intergovernmental Panel on Climate Change (IPCC) Guidelines for National Greenhouse Gas Inventories (IPCC
23 2006). The structure of this report is consistent with the UNFCCC guidelines for inventory reporting.4
Box ES-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Sinks
26 In following the UNFCCC requirement under Article 4.1 to develop and submit national greenhouse gas emissions
27 inventories, the emissions and sinks presented in this report are organized by source and sink categories and
28 calculated using internationally-accepted methods provided by the IPCC.5 Additionally, the calculated emissions
The term "anthropogenic," in this context, refers to greenhouse gas emissions and removals that are a direct result of human
activities or are the result of natural processes that have been affected by human activities (IPCC 2006).
2 Article 2 of the Framework Convention on Climate Change published by the UNEP/WMO Information Unit on Climate
Change. See .
3 Article 4(l)(a) of the United Nations Framework Convention on Climate Change (also identified in Article 12). Subsequent
decisions by the Conference of the Parties elaborated the role of Annex I Parties in preparing national inventories. See
.
4 See < http://unfccc.int/resource/docs/2013/copl9/eng/10a03.pdf>.
5 See < http://www.ipcc-nggip.iges.or.jp/public/index.html>.
Executive Summary ES-1
-------
1 and sinks in a given year for the United States are presented in a common manner in line with the UNFCCC
2 reporting guidelines for the reporting of inventories under this international agreement.6 The use of consistent
3 methods to calculate emissions and sinks by all nations providing their inventories to the UNFCCC ensures that
4 these reports are comparable. In this regard, U.S. emissions and sinks reported in this Inventory report are
5 comparable to emissions and sinks reported by other countries. The manner that emissions and sinks are provided in
6 this Inventory is one of many ways U.S. emissions and sinks could be examined; this Inventory report presents
7 emissions and sinks in a common format consistent with how countries are to report inventories under the
8 UNFCCC. The report itself follows this standardized format, and provides an explanation of the IPCC methods
9 used to calculate emissions and sinks, and the manner in which those calculations are conducted.
10 On October 30, 2009, the U.S. Environmental Protection Agency (EPA) published a rule for the mandatory
11 reporting of greenhouse gases from large greenhouse gas emissions sources in the United States. Implementation of
12 40 CFR Part 98 is referred to as the Greenhouse Gas Reporting Program (GHGRP). 40 CFR part 98 applies to direct
13 greenhouse gas emitters, fossil fuel suppliers, industrial gas suppliers, and facilities that inject CO2 underground for
14 sequestration or other reasons.7 Reporting is at the facility level, except for certain suppliers of fossil fuels and
15 industrial greenhouse gases. The GHGRP dataset and the data presented in this Inventory report are complementary
16 and, as indicated in the respective methodological and planned improvements sections in this report's chapters, EPA
17 is using the data, as applicable, to improve the national estimates presented in this Inventory.
18
19
ES.l. Background Information
27
20 Greenhouse gases trap heat and make the planet warmer. The most important greenhouse gases directly emitted by
21 humans include CO2, CH4, N2O, and several other fluorine-containing halogenated substances. Although the direct
22 greenhouse gases CO2, CH4, and N2O occur naturally in the atmosphere, human activities have changed their
23 atmospheric concentrations. From the pre-industrial era (i.e., ending about 1750) to 2014, concentrations of these
24 greenhouse gases have increased globally by 43, 152, and 20 percent, respectively (IPCC 2013 and NOAA/ESRL
25 2016). This annual report estimates the total national greenhouse gas emissions and removals associated with
26 human activities across the United States.
Global Warming Potentials
28 Gases in the atmosphere can contribute to climate change both directly and indirectly. Direct effects occur when the
29 gas itself absorbs radiation. Indirect radiative forcing occurs when chemical transformations of the substance
30 produce other greenhouse gases, when a gas influences the atmospheric lifetimes of other gases, and/or when a gas
31 affects atmospheric processes that alter the radiative balance of the earth (e.g., affect cloud formation or albedo).8
32 The IPCC developed the Global Warming Potential (GWP) concept to compare the ability of each greenhouse gas to
33 trap heat in the atmosphere relative to another gas.
34 The GWP of a greenhouse gas is defined as the ratio of the time-integrated radiative forcing from the instantaneous
35 release of 1 kilogram (kg) of a trace substance relative to that of 1 kg of a reference gas (IPCC 2013). Direct
36 radiative effects occur when the gas itself is a greenhouse gas. The reference gas used is CO2, and therefore GWP-
37 weighted emissions are measured in million metric tons of CO2 equivalent (MMT CO2 Eq.).9'10 All gases in this
6 See
7 See < http://www.epa.gov/ghgreporting > and .
8 Albedo is a measure of the Earth's reflectivity, and is defined as the fraction of the total solar radiation incident on a body that
is reflected by it.
9 Carbon comprises 12/44ths of carbon dioxide by weight.
10 One teragram is equal to 1012 grams or one million metric tons.
ES-2 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Executive Summary are presented in units of MMT CO2 Eq. Emissions by gas in unweighted mass tons are provided
2 in the Trends chapter of this report.
3 UNFCCC reporting guidelines for national inventories require the use of GWP values from the IPCC Fourth
4 Assessment Report (AR4) (IPCC 2007).11 To comply with international reporting standards under the UNFCCC,
5 official emission estimates are reported by the United States using AR4 GWP values, which have replaced the
6 previously required use of SAR GWP values in the U.S. Inventory. All estimates are provided throughout the report
7 in both CC>2 equivalents and unweighted units. A comparison of emission values using the AR4 GWP values versus
8 the IPCC Second Assessment Report (SAR) (IPCC 1996), IPCC Third Assessment Report (TAR) (IPCC 2001), and
9 the IPCC Fifth Assessment Report (AR5) (IPCC 2013) GWP values can be found in Chapter 1 and, in more detail,
10 inAnnex6.1 of this report. The GWP values used in this report are listed below in Table ES-1.
11
12 Table ES-1: Global Warming Potentials (100-Year Time Horizon) Used in this Report
13
14
Gas
CO2
CH4a
N2O
HFC-23
HFC-32
HFC-125
HFC-134a
HFC-143a
HFC-152a
HFC-227ea
HFC-236fa
HFC-4310mee
CF4
C2F6
C4Fio
C6Fi4
SF6
NF3
GWP
1
25
298
14,800
675
3,500
1,430
4,470
124
3,220
9,810
1,640
7,390
12,200
8,860
9,300
22,800
17,200
Source: IPCC (2007)
a The CH4 GWP includes the direct
effects and those indirect effects due
to the production of tropospheric
ozone and stratospheric water vapor.
The indirect effect due to production
of CO2 is not included.
11 See < http://unfccc.int/resource/docs/2013/copl9/eng/10a03.pdf:
Executive Summary ES-3
-------
j
4
5
6
7
8
9
10
11
12
13
ES.2. Recent Trends in U.S. Greenhouse Gas
Emissions and Sinks
In 2014, total U.S. greenhouse gas emissions were 6,872.6 MMT, or million metric tons, CCh Eq. Total U.S.
emissions have increased by 7.7 percent from 1990 to 2014, and emissions increased from 2013 to 2014 by 0.9
percent (61.5 MMT CCh Eq.). Additionally, relatively cool winter conditions led to an increase in fuels for the
residential and commercial sectors for heating. In 2014 there also was an increase in industrial production across
multiple sectors resulting in slight increases in industrial sector emissions. Lastly, transportation emissions increased
as a result of a small increase in vehicle miles traveled (VMT) and fuel use across on-road transportation modes.
Since 1990, U. S. emissions have increased at an average annual rate of 0.3 percent. Figure ES-1 through Figure ES-
3 illustrate the overall trends in total U.S. emissions by gas, annual changes, and absolute change since 1990.
Table ES-2 provides a detailed summary of U.S. greenhouse gas emissions and sinks for 1990 through 2014.
Figure ES-1: U.S. Greenhouse Gas Emissions by Gas (MMT COz Eq.)
i MFCs, PFCs, SFfi and NF3
• Methane
i Nitrous Oxide
• Carbon Dioxide
6,816 • 6,888 g66S 68116Ł73
14
15
ES-4 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Figure ES-2: Annual Percent Change in U.S. Greenhouse Gas Emissions
4% -i
2%
3.2%
2.9%
2.2%
.9%
J
4
Figure ES-3: Annual Greenhouse Gas Emissions Relative to 1990 (1990=0)
1,048 1099
492
5
6
7
O i-\ (N m TT
1—11—11—11—11—1
(N (N (N (N fM
Table ES-2: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks (MMT COz Eq.)
Gas/Source
1990
2005
2010
2011
2012
2013
2014
C02
Fossil Fuel Combustion
Electricity Generation
Transportation
Industrial
Residential
Commercial
U.S. Territories
Non-Energy Use of Fuels
Iron and Steel Production &
Metallurgical Coke Production
Cement Production
Natural Gas Systems*
Petrochemical Production
Lime Production
Other Process Uses of Carbonates
Ammonia Production
Incineration of Waste
Petroleum Systems'5
5,124.0
4,740.7
1,820.8
1,493.8
842.5
338.3
217.4
27.9
118.1
6,132.6
5,747.1
2,400.9
1,887.0
828.0
357.8
223.5
49.9
138.9
5,698.2
5,358.3
2,258.4
1,728.3
775.5
334.6
220.1
41.4
114.1
5,361.0
5,024.7
2,022.2
1,696.8
782.9
282.5
196.7
43.6
105.6
5,564.3
5,208.7
2,039.3
1,737.4
814.2
345.1
231.6
41.0
114.3
55.4
38.8
37.8
26.5
14.1
12.1
9.4
9.4
6.0
Executive Summary ES-5
-------
Urea Fertilization
Carbon Dioxide Consumption
Liming
Urea Consumption for Non-
Agricultural Purposes
Aluminum Production
Soda Ash Production and
Consumption
Ferroalloy Production
Titanium Dioxide Production
Glass Production
Phosphoric Acid Production
Zinc Production
Peatlands Remaining Peatlands
Lead Production
Silicon Carbide Production and
Consumption
Magnesium Production and
Processing
LULUCF Total NetFluof
Wood Biomass and Ethanol
Consumption11
International Bunker Fuels'
CH4
Landfills
Enteric Fermentation
Natural Gas Systems3
Coal Mining
Manure Management
Petroleum Systems'5
Wastewater Treatment
Rice Cultivation
Stationary Combustion
Forest Fires
Abandoned Underground Coal
Mines
Composting
Mobile Combustion
Field Burning of Agricultural
Residues
Petrochemical Production
Ferroalloy Production
Silicon Carbide Production and
Consumption
Iron and Steel Production &
Metallurgical Coke Production
Peatlands Remaining Peatlands
Incineration of Waste
International Bunker Fuels"
N2O
Agricultural Soil Management
Stationary Combustion
Manure Management
Mobile Combustion
Nitric Acid Production
Adipic Acid Production
Wastewater Treatment
Forest Fires
2.4
1.5
4.7
2.S
2.21
12
1.5
1.5
0.6
1.1
0.4
1.4
43
37
4 if
3.0
!'8
L9
13
1.0
11
0.6
0.2
3.8
4.4
4.8
4.7
2.7
2.7
1.7
1.8
1.5
1.1
1.2
1.0
0.5
4.1
4.1
3.9
4.0
3.3
2.7
1.7
1.7
1.3
1.2
1.3
0.9
0.5
4.2
4.0
6.0
4.4
3.4
2.8
1.9
1.5
1.2
1.1
1.5
0.8
0.5
4.3
4.2
3.9
4.2
3.3
2.8
1.8
1.7
1.3
1.1
1.4
0.8
0.5
4.5
4.5
4.1
4.0
3.3
2.8
1.9
1.8
1.3
1.1
1.0
0.8
0.5
0.2
0.2
0.2
0.2
265.1
117.0
720.8
176.3
171.3
268.1
111.7
711.8
176.9
168.9
0.1
419.4
322.9
21.4
17.5
20.0
10.5
5.5
4.9
7.3
0.1
411.0
318.4
22.9
17.5
18.2
10.7
4.0
4.9
4.8
0.2
(680.8) (682.4) (685.8)
267.7
105.8
703.8
173.5
166.7
154.4
66.5
63.7
23.3
15.2
12.2
6.6
11.1
6.2
1.9
2.2
0.3
0.1
286.3
99.8
704.0
176.7
165.5
157.4
64.6
61.4
25.2
15.0
12.2
8.0
7.3
6.2
2.0
2.1
0.3
0.1
293.7
103.2
707.9
181.8
164.3
157.4
64.6
61.2
25.2
15.0
12.2
8.1
7.3
6.2
2.1
2.0
0.3
0.1
0.1
411.4
318.5
23.4
17.5
16.3
10.9
5.4
4.9
4.8
ES-6 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
N2O from Product Uses 4.2 • 4.2
Settlement Soils 1.4! 2.3
Composting 0.3B 1.7
Forest Soils 0.11 0.5
Incineration of Waste O.sB 0.4
Semiconductor Manufacture + 0.1
Field Burning of Agricultural
Residues 0.11 0.1
Peatlands Remaining Peatlands + +|
International Bunker Fuels" 0.9U 1.0
HFCs, PFCs, SF6 and NF3 102.0 154.4
HFCs 46.6 133.3
Substitution of Ozone Depleting
Substancesf O.sl 113.0
HCFC-22 Production 46.1 20.0
Semiconductor Manufacture 0.21 0.2
Magnesium Production and
Processing 0.0 • 0.0
PFCs 24.3 6.6
Aluminum Production 21.5 3.4
Semiconductor Manufacture 2.8H 3.2
SF6 31.1 14.0
Electrical Transmission and
Distribution 25.4 10.6
Magnesium Production and
4.2
2.4
1.6
0.5
0.3
0.1
4.2
2.5
1.7
0.5
0.3
0.2
4.2
2.5
1.7
0.5
0.3
0.2
4.2
2.4
1.8
0.5
0.3
0.2
4.2
2.4
1.8
0.5
0.3
0.2
0.1
0.1
0.1
0.1
7.0
5.7
5.1
0.1
1.0
176.2
161.7
153.5
8.0
0.2
+
4.4
1.9
2.6
9.5
1.0
183.6
166.1
157.1
8.8
0.2
+
6.9
3.5
3.4
10.0
0.9
181.4
167.1
161.4
5.5
0.2
+
6.0
2.9
3.0
7.7
0.9
182.9
169.6
165.3
4.1
0.2
0.1
5.8
3.0
2.9
6.9
0.9
189.1
175.8
171.4
4.1
0.2
0.1
5.8
3.0
2.9
6.9
5.1
Processing
Semiconductor Manufacture
NF3
Semiconductor Manufacture
Total Emissions
LULUCF Emissions^
LULUCF Total Net Fluxc
LULUCF Sector Total"
Net Emissions (Sources and Sinks)
+
6,380.8
15.0
(704.2)
(689.1)
5,676.6
»l
0.5|
qjm
7,428.8
28.2
(636.1)
(607.9)
6,792.6
2.1
0.4
0.5
0.5
7,010.5
17.8
(683.2)
(665.3)
6,327.3
2.8
0.4
0.7
0.7
6,887.8
22.9
(683.6)
(660.7)
6,204.2
1.6
0.4
0.6
0.6
6,665.7
32.3
(680.8)
(648.5)
5,984.9
1.4
0.4
0.6
0.6
6,811.2
24.1
(682.4)
(658.3)
6,128.8
1.4
0.4
0.6
0.6
6,872.6
24.6
(685.8)
(661.3)
6,186.8
+ Does not exceed 0.05 MMT CO2 Eq.
a The values in this table for Natural Gas Systems are presented from the previous Inventory and do not reflect updates to emission
estimates for this category. See Section 3.7, Natural Gas Systems of the Energy chapter for more information. Gray highlighting was
added on 2/24 for clarification.
b The values in this table for Petroleum Systems are presented from the previous Inventory and do not reflect updates to emission
estimates for this category. See Section 3.6, Petroleum Systems of the Energy chapter for more information. Gray highlighting was
added on 2/24 for clarification.
c Total net flux from LULUCF is only included in the Net Emissions total. Net flux from LULUCF includes the positive C sequestration
reported for Forest Land Remaining Forest Land, Land Converted to Forest Land, Cropland Remaining Cropland, Land Converted to
Grassland, Settlements Remaining Settlements, and Other Land plus the loss in C sequestration reported for Land Converted to
Cropland and Grassland Remaining Grassland. Refer to Table ES-5 for a breakout of emissions and removals for Land Use, Land-Use
Change, and Forestry by gas and source category.
d Emissions from Wood Biomass and Ethanol Consumption are not included specifically in summing Energy sector totals. Net carbon
fluxes from changes in biogenic carbon reservoirs are accounted for in the estimates for Land Use, Land-Use Change, and Forestry.
e Emissions from International Bunker Fuels are not included in totals.
f Small amounts of PFC emissions also result from this source.
8 LULUCF emissions include the CO2, CFL., and N2O emissions reported for Forest Fires, Forest Soils, Liming, Urea Fertilization,
Settlement Soils, and Peatlands Remaining Peatlands.
h The LULUCF Sector Total is the sum of positive emissions (i.e., sources) of greenhouse gases to the atmosphere plus removals of CO2
(i.e., sinks or negative emissions) from the atmosphere.
Note: Totals may not sum due to independent rounding. Parentheses indicate negative values or sequestration.
1 Figure ES-4 illustrates the relative contribution of the direct greenhouse gases to total U.S. emissions in 2014. The
2 primary greenhouse gas emitted by human activities in the United States was CCh, representing approximately 81.0
3 percent of total greenhouse gas emissions. The largest source of €62, and of overall greenhouse gas emissions, was
4 fossil fuel combustion. CH4 emissions, which have decreased by 5.0 percent since 1990, resulted primarily from
Executive Summary ES-7
-------
1 decomposition of wastes in landfills, enteric fermentation associated with domestic livestock, and natural gas
2 systems. Agricultural soil management, manure management, mobile source fuel combustion and stationary fuel
3 combustion were the major sources of N2O emissions. Ozone depleting substance substitute emissions and
4 emissions of HFC-23 during the production of HCFC-22 were the primary contributors to aggregate
5 hydrofluorocarbon (HFC) emissions. Perfluorocarbon (PFC) emissions resulted as a byproduct of primary
6 aluminum production and from semiconductor manufacturing, electrical transmission and distribution systems
7 accounted for most sulfur hexafluoride (SF6) emissions, and semiconductor manufacturing is the only source of NF3
8 emissions.
9 Figure ES-4: 2014 Greenhouse Gas Emissions by Gas (Percentages based on MMT COz Eq.)
MFCs, PFCs,
SF6 and NF3
Subtotal
2.8%
23
10
11 Overall, from 1990 to 2014, total emissions of CO2 increased by 440.2 MMT CO2 Eq. (8.6 percent), while total
12 emissions of CH4 decreased by 37.4 MMT CO2Eq. (5.0 percent), andN2O increased by 1.9 MMT CO2Eq. (0.5
13 percent). During the same period, aggregate weighted emissions of HFCs, PFCs, SF6 and NF3 rose by 87.1 MMT
14 CO2 Eq. (85.4 percent). From 1990 to 2014, HFCs increased by 129.2 MMT CO2 Eq. (277.3 percent), PFCs
15 decreased by 18.4 MMT CO2 Eq. (76.0 percent), SF6 decreased by 24.1 MMT CO2 Eq. (77.7 percent), and NF3
16 increased by 0.5 MMT CO2 Eq. (1,070.1 percent). Despite being emitted in smaller quantities relative to the other
17 principal greenhouse gases, emissions of HFCs, PFCs, SF6 and NF3 are significant because many of these gases
18 have extremely high global warming potentials and, in the cases of PFCs and SF6, long atmospheric lifetimes.
19 Conversely, U.S. greenhouse gas emissions were partly offset by carbon sequestration in forests, trees in urban
20 areas, agricultural soils, and landfilled yard trimmings and food scraps, which, in aggregate, offset 10.0 percent of
21 total emissions in 2014. The following sections describe each gas's contribution to total U.S. greenhouse gas
22 emissions in more detail.
Carbon Dioxide Emissions
24 The global carbon cycle is made up of large carbon flows and reservoirs. Billions of tons of carbon in the form of
25 CO2 are absorbed by oceans and living biomass (i.e., sinks) and are emitted to the atmosphere annually through
26 natural processes (i.e., sources). When in equilibrium, carbon fluxes among these various reservoirs are roughly
27 balanced.12 Since the Industrial Revolution (i.e., about 1750), global atmospheric concentrations of CO2 have risen
28 approximately 43 percent (IPCC 2013 and NOAA/ESRL 2016), principally due to the combustion of fossil fuels.
29 Within the United States, fossil fuel combustion accounted for 93.6 percent of CO2 emissions in 2014. Globally,
30 approximately 32,310 MMT of CO2 were added to the atmosphere through the combustion of fossil fuels in 2012, of
12 The term "flux" is used to describe the net emissions of greenhouse gases to the atmosphere accounting for both the emissions
of CO2 to and the removals of CCh from the atmosphere. Removal of CCh from the atmosphere is also referred to as "carbon
sequestration."
ES-8 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1
2
o
J
4
9
10
11
12
13
14
15
16
17
18
19
20
21
22
which the United States accounted for about 16 percent.13 Changes in land use and forestry practices can also emit
CO2 (e.g., through conversion of forest land to agricultural or urban use) or can act as a sink for CCh (e.g., through
net additions to forest biomass). Although fossil fuel combustion is the greatest source of €62 emissions, there are
25 additional sources of €62 emissions (Figure ES-5).
Figure ES-5: 2014 Sources of COz Emissions
Fossil Fuel Combustion
Non-Energy Use of Fuels
Iron and Steel Prod. & Metallurgical Coke Prod.
Cement Production
Natural Gas Systems3
Petrochemical Production
Lime Production
Ottier Process Uses of Carbonates
Ammonia Production
Incineration of Waste
Petroleum Systems"
Urea Fertilization
Carbon Dioxide Consumption
Liming
Urea Consumption for Non-Agricultural Purposes
Aluminum Production
Soda Ash Production and Consumption
Ferroalloy Production
Titanium Dioxide Production
Glass Production
Phosphoric Acid Production
Zinc Production
Peadands Remaining Peatlands
Lead Production
Silicon Carbide Production and Consumption
Magnesium Production and Processing
5,209
CO2 as a Portion
of all Emissions
25 50 75 100
MMT CO2 Eq.
125
150
Note: Fossil Fuel Combustion includes electricity generation, which also includes emissions of less than 0.05 MMT CCh Eq.
from geothermal-based generation.
a The value in this figure for Natural Gas Systems is presented from the previous Inventory and does not reflect updates to
emission estimates for this category. See Section 3.7, Natural Gas Systems of the Energy chapter for more information.
b The value in this figure for Petroleum Systems is presented from the previous Inventory and does not reflect updates to
emission estimates for this category. See Section 3.6, Petroleum Systems of the Energy chapter for more information.
As the largest source of U.S. greenhouse gas emissions, €62 from fossil fuel combustion has accounted for
approximately 76 percent of GWP-weighted emissions since 1990, and is approximately 76 percent of total GWP-
weighted emissions in 2014. Emissions of €62 from fossil fuel combustion increased at an average annual rate of
0.4 percent from 1990 to 2014. The fundamental factors influencing this trend include (1) a generally growing
domestic economy over the last 25 years, (2) an overall growth in emissions from electricity generation and
transportation activities, along with (3) a general decline in the carbon intensity of fuels combusted for energy in
recent years by most sectors of the economy. Between 1990 and 2014, €62 emissions from fossil fuel combustion
increased from 4,740.7 MMT CO2 Eq. to 5,208.7 MMT CO2 Eq., a 9.9 percent total increase over the twenty-five-
year period. From 2013 to 2014, these emissions increased by 51.1 MMT €62 Eq. (1.0 percent).
Global CO2 emissions from fossil fuel combustion were taken from Energy Information Administration International Energy
Statistics 2013 < http://tonto.eia.doe.gov/cfapps/ipdbproject/IEDIndex3.cfm> (EIA 2013).
Executive Summary ES-9
-------
1
2
o
6
4
5
6
7
8
9
10
11
12
13
14
15
Historically, changes in emissions from fossil fuel combustion have been the dominant factor affecting U.S.
emission trends. Changes in CCh emissions from fossil fuel combustion are influenced by many long-term and
short-term factors, including population and economic growth, energy price fluctuations, technological changes,
energy fuel choices, and seasonal temperatures. In the short term, the overall consumption of fossil fuels in the
United States fluctuates primarily in response to changes in general economic conditions, energy prices, weather,
and the availability of non-fossil alternatives. For example, in a year with increased consumption of goods and
services, low fuel prices, severe summer and winter weather conditions, nuclear plant closures, and lower
precipitation feeding hydroelectric dams, there would likely be proportionally greater fossil fuel consumption than a
year with poor economic performance, high fuel prices, mild temperatures, and increased output from nuclear and
hydroelectric plants. In the long term, energy consumption patterns respond to changes that affect the scale of
consumption (e.g., population, number of cars, and size of houses), the efficiency with which energy is used in
equipment (e.g., cars, power plants, steel mills, and light bulbs) and behavioral choices (e.g., walking, bicycling, or
telecommuting to work instead of driving).
Figure ES-6:
COz Eq.)
2014 COz Emissions from Fossil Fuel Combustion by Sector and Fuel Type (MMT
19
2,500
2,000 -
Relative Contribution
by Fuel Type
2,039
1,737
16
17
18
Figure ES-7: 2014 End-Use Sector Emissions of COz from Fossil Fuel Combustion (MMT COz
Eq.)
2,000
i From Direct Fossil Fuel Combustion
1,741
From Electricity Consumption
ES-10 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 The five major fuel consuming sectors contributing to CO2 emissions from fossil fuel combustion are electricity
2 generation, transportation, industrial, residential, and commercial. CC>2 emissions are produced by the electricity
3 generation sector as they consume fossil fuel to provide electricity to one of the other four sectors, or "end-use"
4 sectors. For the discussion below, electricity generation emissions have been distributed to each end-use sector on
5 the basis of each sector's share of aggregate electricity consumption. This method of distributing emissions assumes
6 that each end-use sector consumes electricity that is generated from the national average mix of fuels according to
7 their carbon intensity. Emissions from electricity generation are also addressed separately after the end-use sectors
8 have been discussed.
9 Note that emissions from U.S. Territories are calculated separately due to a lack of specific consumption data for the
10 individual end-use sectors. Figure ES-6, Figure ES-7, and Table ES-3 summarize CCh emissions from fossil fuel
11 combustion by end-use sector.
12 Table ES-3: COz Emissions from Fossil Fuel Combustion by Fuel Consuming End-Use Sector
13 (MMT COz Eq.)
End-Use Sector
Transportation
Combustion
Electricity
Industrial
Combustion
Electricity
Residential
Combustion
Electricity
Commercial
Combustion
Electricity
U.S. Territories3
Total
Electricity Generation
1990
1,496.8
1,493.8
3.ol
1,529.2
842. 5 1
686.vl
931.41
338.3 1
593.o!
755.4 1
217.41
538.0
27.9
4,740.7
1,820.8
2005
1,891.8
1,887.0 1
4.7l
1,564.6 1
828.0
736.6 1
1,214.1
357.8 1
856.31
1,026.8
223.5
803.3
49.9
5,747.1
2,400.9
1 2010
1,732.7
1,728.3
4.5
1,416.5
775.5
641.0
1,174.6
334.6
840.0
993.0
220.1
772.9
41.4
5,358.3
2,258.4
2011
1,711.9
1,707.6
4.3
1,398.0
773.3
624.7
1,117.5
326.8
790.7
958.8
220.7
738.0
41.5
5,227.7
2,157.7
2012
1,700.6
1,696.8
3.9
1,375.7
782.9
592.8
1,007.8
282.5
725.3
897.0
196.7
700.3
43.6
5,024.7
2,022.2
2013
1,717.0
1,713.0
4.0
1,407.0
812.2
594.7
1,064.6
329.7
734.9
925.5
221.0
704.5
43.5
5,157.6
2,038.1
2014
1,741.5
1,737.4
4.1
1,407.8
814.2
593.6
1,080.4
345.1
735.2
938.1
231.6
706.5
41.0
5,208.7
2,039.3
Note: Totals may not sum due to independent rounding. Combustion-related emissions from electricity
generation are allocated based on aggregate national electricity consumption by each end-use sector.
a Fuel consumption by U.S. Territories (i.e., American Samoa, Guam, Puerto Rico, U.S. Virgin Islands, Wake
Island, and other U.S. Pacific Islands) is included in this report.
14 Transportation End-Use Sector. When electricity-related emissions are distributed to economic end-use sectors,
15 transportation activities accounted for 33.4 percent of U.S. €62 emissions from fossil fuel combustion in 2014. The
16 largest sources of transportation CC>2 emissions in 2014 were passenger cars (42.5 percent), medium- and heavy-
17 duty trucks (23.1 percent), light-duty trucks, which include sport utility vehicles, pickup trucks, and minivans (17.8
18 percent), commercial aircraft (6.6 percent), pipelines (2.7 percent), rail (2.6 percent), and ships and boats (1.6
19 percent). Annex 3.2 presents the total emissions from all transportation and mobile sources, including €62, CH4,
20 N2O, and MFCs.
21 In terms of the overall trend, from 1990 to 2014, total transportation €62 emissions rose by 16 percent due, in large
22 part, to increased demand for travel as fleet wide light-duty vehicle fuel economy was relatively stable (average new
23 vehicle fuel economy declined slowly from 1990 through 2004 and then increased more rapidly from 2005 through
24 2014). The number of vehicle miles traveled by light-duty motor vehicles (passenger cars and light-duty trucks)
25 increased 37 percent from 1990 to 2014, as a result of a confluence of factors including population growth,
26 economic growth, urban sprawl, and low fuel prices during the beginning of this period. Almost all of the energy
27 consumed for transportation was supplied by petroleum-based products, with more than half being related to
28 gasoline consumption in automobiles and other highway vehicles. Other fuel uses, especially diesel fuel for freight
29 trucks and jet fuel for aircraft, accounted for the remainder.
30 Industrial End-Use Sector. Industrial CO2 emissions, resulting both directly from the combustion of fossil fuels and
31 indirectly from the generation of electricity that is consumed by industry, accounted for 27 percent of €62 from
32 fossil fuel combustion in 2014. Approximately 58 percent of these emissions resulted from direct fossil fuel
Executive Summary ES-11
-------
1 combustion to produce steam and/or heat for industrial processes. The remaining emissions resulted from
2 consuming electricity for motors, electric furnaces, ovens, lighting, and other applications. In contrast to the other
3 end-use sectors, emissions from industry have steadily declined since 1990. This decline is due to structural changes
4 in the U.S. economy (i.e., shifts from a manufacturing-based to a service-based economy), fuel switching, and
5 efficiency improvements.
6 Residential and Commercial End-Use Sectors. The residential and commercial end-use sectors accounted for 21
7 and 18 percent, respectively, of CC>2 emissions from fossil fuel combustion in 2014. Both sectors relied heavily on
8 electricity for meeting energy demands, with 68 and 75 percent, respectively, of their emissions attributable to
9 electricity consumption for lighting, heating, cooling, and operating appliances. The remaining emissions were due
10 to the consumption of natural gas and petroleum for heating and cooking. Emissions from the residential and
11 commercial end-use sectors have increased by 16 percent and 24 percent since 1990, respectively, due to increasing
12 electricity consumption for lighting, heating, air conditioning, and operating appliances.
13 Electricity Generation. The United States relies on electricity to meet a significant portion of its energy demands.
14 Electricity generators consumed 34 percent of total U.S. energy uses from fossil fuels and emitted 39 percent of the
15 CO2 from fossil fuel combustion in 2014. The type of fuel combusted by electricity generators has a significant
16 effect on their emissions. For example, some electricity is generated through non-fossil fuel options such as nuclear,
17 hydroelectric, or geothermal energy. Including all electricity generation modes, generators relied on coal for
18 approximately 39 percent of their total energy requirements in 2014.14 In addition, the coal used by electricity
19 generators accounted for 93 percent of all coal consumed for energy in the United States in 2014.15 Recently a
20 decrease in the carbon intensity of fuels consumed to generate electricity has occurred due to a decrease in coal
21 consumption, and increased natural gas consumption and other generation sources. Including all electricity
22 generation modes, electricity generators used natural gas for approximately 27 percent of their total energy
23 requirements in 2014.16 Across the time series, changes in electricity demand and the carbon intensity of fuels used
24 for electricity generation have a significant impact on CC>2 emissions.
25 Other significant CCh trends included the following:
26 • Carbon dioxide emissions from non-energy use of fossil fuels have decreased by 3.8 MMT CCh Eq. (3.2
27 percent) from 1990 through 2014. Emissions from non-energy uses of fossil fuels were 114.3 MMT CCh
28 Eq. in 2014, which constituted 2.1 percent of total national CCh emissions, approximately the same
29 proportion as in 1990.
30 • Carbon dioxide emissions from iron and steel production and metallurgical coke production have declined
31 by 44.3 MMT CO2 Eq. (44.5 percent) from 1990 through 2014, due to restructuring of the industry,
32 technological improvements, and increased scrap steel utilization.
33 • Carbon dioxide emissions from ammonia production (9.4 MMT CO2 Eq. in 2014) decreased by 3.6 MMT
34 CO2 Eq. (27.7 percent) since 1990. Ammonia production relies on natural gas as both a feedstock and a
35 fuel, and as such, market fluctuations and volatility in natural gas prices affect the production of ammonia.
36 • Carbon sequestration from Land Use, Land-Use Change, and Forestry decreased by 18.4 MMT CCh Eq.
37 (2.6 percent) from 1990 through 2014. This increase was primarily due to a decrease in the rate of net C
38 accumulation in agricultural soil carbon stocks. Annual carbon accumulation in landfilled yard trimmings
39 and food scraps slowed over this period, while the rate of carbon accumulation in urban trees increased.
Box ES- 2: Use of Ambient Measurements Systems for Validation of Emission Inventories
41 In following the UNFCCC requirement under Article 4.1 to develop and submit national greenhouse gas emission
42 inventories, the emissions and sinks presented in this report are organized by source and sink categories and
43 calculated using internationally-accepted methods provided by the IPCC.17 Several recent studies have measured
44 emissions at the national or regional level (e.g., Petron 2012, Miller et al. 2013) with results that differ from EPA's
See < http://www.eia.gov/energyexplained/index.cfm?page=electricity_in_the_united_states >.
15 See Table 6.2 Coal Consumption by Sector of EIA 2015a.
16 See < http://www.eia.gov/energyexplained/index.cfm?page=electricity_in_the_united_states >.
17 See < http://www.ipcc-nggip.iges.or.jp/public/index.html>.
ES-12 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1
2
o
6
4
5
6
7
8
9
10
11
estimate of emissions. A recent study (Brandt et al. 2014) reviewed technical literature on methane emissions and
estimated methane emissions from all anthropogenic sources (e.g., livestock, oil and gas, waste emissions) to be
greater than EPA's estimate. EPA has engaged with researchers on how remote sensing, ambient measurement, and
inverse modeling techniques for greenhouse gas emissions could assist in improving the understanding of inventory
estimates. An area of particular interest in EPA's outreach efforts is how these data can be used in a manner
consistent with this Inventory report's transparency on its calculation methodologies, and the ability of these
techniques to attribute emissions and removals from remote sensing to anthropogenic sources, as defined by the
IPCC for this report, versus natural sources and sinks. In working with the research community on ambient
measurement and remote sensing techniques to improve national greenhouse gas inventories, EPA relies upon
guidance from the IPCC on the use of measurements and modeling to validate emission inventories.18
12
13
14
15
16
17
18
19
20
21
22
23
24
Methane Emissions
Methane (CH4) is 25 times as effective as CCh at trapping heat in the atmosphere (IPCC 2007). Over the last two
hundred and fifty years, the concentration of CH4 in the atmosphere increased by 152 percent (IPCC 2013 and
CDIAC 2014). Anthropogenic sources of CH4 include natural gas and petroleum systems, agricultural activities,
landfills, coal mining, wastewater treatment, stationary and mobile combustion, and certain industrial processes (see
Figure ES-8).
Figure ES-8: 2014 Sources of CH4 Emissions (MMT COz Eq.)
Landfills
Enteric Fermentation
Natural Gas Systems3
Coal Mining
Manure Management
Petroleum Systems13
Wastewater Treatment
Rice Cultivation
Stationary Combustion
Forest Rres
Abandoned Underground Coal Mines
Composting
Mobile Combustion
Field Burning of Agricultural Residues
Petrochemical Production
Ferroalloy Production
Silicon Carbide Production and Consumption
Iron and Steel Prod. & Metallurgical Coke Prod.
Wetlands Remaining Wetlands
Incineration of Waste
CH4 as a Portion
of all Emissions
10.3%
0
25
50
75 100
MMT CO2 Eq.
125
150
175
a The value in this figure for Natural Gas Systems is presented from the previous Inventory and does not reflect updates to
emission estimates for this category. See Section 3.7, Natural Gas Systems of the Energy chapter for more information.
b The value in this figure for Petroleum Systems is presented from the previous Inventory and does not reflect updates to
emission estimates for this category. See Section 3.6, Petroleum Systems of the Energy chapter for more information.
18
See.
Executive Summary ES-13
-------
1 Some significant trends in U.S. emissions of CH4 include the following:
2 • Landfills are the largest anthropogenic source of CH4 emissions in the United States (181.8 MMT CO2
3 Eq.), accounting for 25.7 percent of total CH4 emissions in 2014. From 1990 to 2014, CH4 emissions from
4 landfills decreased by 2.6 MMT CO2 Eq. (1.4 percent), with small increases occurring in some interim
5 years. This downward trend in overall emissions can be attributed to a 21 percent reduction in the amount
6 of decomposable materials (i.e., paper and paperboard, food scraps, and yard trimmings) discarded in MSW
7 landfills over the time series (EPA 2015b) and an increase in the amount of landfill gas collected and
8 combusted (i.e., used for energy or flared),19 which has more than offset the additional CH4 emissions
9 resulting from an increase in the amount of municipal solid waste landfilled.
10 • Enteric fermentation is the second largest anthropogenic source of CH4 emissions in the United States. In
11 2014, enteric fermentation CH4 emissions were 164.3 MMT CO2 Eq. (23.2 percent of total CH4 emissions),
12 which represents an increase of 0.1 MMT CO2 Eq. (0.1 percent) since 1990. This increase in emissions
13 from 1990 to 2014 in enteric generally follows the increasing trends in cattle populations. From 1990 to
14 1995 emissions increased and then generally decreased from 1996 to 2004, mainly due to fluctuations in
15 beef cattle populations and increased digestibility of feed for feedlot cattle. Emissions increased from 2005
16 to 2007, as both dairy and beef populations underwent increases and the literature for dairy cow diets
17 indicated a trend toward a decrease in feed digestibility for those years. Emissions decreased again from
18 2008 to 2014 as beef cattle populations again decreased.
19 • Substantial new data are available on natural gas and petroleum systems from EPA's Greenhouse Gas
20 Reporting Program (GHGPJ3) and a number of new studies. EPA is evaluating approaches for
21 incorporating this new data into its emission estimates. In Section 3.7 and Section 3.6 of the Energy
22 chapter, updated draft estimates of CH4 emissions for the year 2013 are presented for Natural Gas Systems
23 and Petroleum Systems, respectively, to provide reviewers of the public review draft an indication of the
24 sector-wide emission estimates resulting from the combined changes under consideration. EPA is
25 continuing to evaluate stakeholder feedback on the updates under consideration. For the final Inventory, the
26 2013 estimates presented in this section will be refined, and a full time series of emission estimates will be
27 developed based on feedback received through the earlier stakeholder reviews of the memos and through
28 this public review period. The details of the revisions under consideration for this year's Inventory, and key
29 questions for stakeholder feedback are available in segment-level memoranda at
30 http://www3.epa.gov/climatechange/ghgemissions/usinventorvreport/natural-gas-svstems.html.
31 • Methane emissions from manure management increased by 64.7 percent since 1990, from 37.2 MMT CO2
32 Eq. in 1990 to 61.2 MMT CO2 Eq. in 2014. The majority of this increase was from swine and dairy cow
33 manure, since the general trend in manure management is one of increasing use of liquid systems, which
34 tends to produce greater CH4 emissions. The increase in liquid systems is the combined result of a shift to
35 larger facilities, and to facilities in the West and Southwest, all of which tend to use liquid systems. Also,
36 new regulations limiting the application of manure nutrients have shifted manure management practices at
37 smaller dairies from daily spread to manure managed and stored on site.
38
Nitrous Oxide Emissions
39 Nitrous oxide is produced by biological processes that occur in soil and water and by a variety of anthropogenic
40 activities in the agricultural, energy-related, industrial, and waste management fields. While total N2O emissions are
41 much lower than CO2 emissions, N2O is approximately 300 times more powerful than CO2 at trapping heat in the
42 atmosphere (IPCC 2007). Since 1750, the global atmospheric concentration of N2O has risen by approximately 20
43 percent (IPCC 2013 and CDIAC 2014). The main anthropogenic activities producing N2O in the United States are
44 agricultural soil management, stationary fuel combustion, fuel combustion in motor vehicles, manure management
45 and nitric acid production (see Figure ES-9).
Carbon dioxide emissions from landfills are not included specifically in summing waste sector totals. Net carbon fluxes from
changes in biogenic carbon reservoirs are accounted for in the estimates for Land Use, Land-Use Change, and Forestry.
ES-14 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Figure ES-9: 2014 Sources of NzO Emissions (MMT COz Eq.)
3
4
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Agricultural Soil Management
Stationary Combustion
Manure Management
Mobile Combustion
Nitric Acid Production
Adipic Acid Production
Wastewater Treatment
Forest Fires
N2O from Product Uses
Settlement Soils
Composting
Forest Soils
Incineration of Waste
Semiconductor Manufacture
Field Burning of Agricultural Residues
Peatiands Remaining Peatlands
319
N2O as a Portion
of all Emissions
10 15
MMT CO2 Eq.
20
25
Some significant trends in U.S. emissions of N2O include the following:
• Agricultural soils accounted for approximately 77.4 percent of N2O emissions and 4.6 percent of total
emissions in the United States in 2014. Estimated emissions from this source in 2014 were 318.5 MMT
CO2 Eq. Annual N2O emissions from agricultural soils fluctuated between 1990 and 2014, although overall
emissions were 5.1 percent higher in 2014 than in 1990. Year-to-year fluctuations are largely a reflection
of annual variation in weather patterns, synthetic fertilizer use, and crop production.
• Nitrous oxide emissions from stationary combustion increased 11.5 MMT CO2 Eq. (96.4 percent) from
1990 through 2014. Nitrous oxide emissions from this source increased primarily as a result of an increase
in the number of coal fluidized bed boilers in the electric power sector.
• In 2014, total N2O emissions from manure management were estimated to be 17.5 MMT CO2 Eq.;
emissions were 14.0 MMT CO2 Eq. in 1990. These values include both direct and indirect N2O emissions
from manure management. Nitrous oxide emissions have remained fairly steady since 1990. Small
changes in N2O emissions from individual animal groups exhibit the same trends as the animal group
populations, with the overall net effect that N2O emissions showed a 24.9 percent increase from 1990 to
2014 and a 0.1 percent decrease from 2013 through 2014. Overall shifts toward liquid systems have driven
down the emissions per unit of nitrogen excreted.
• Nitrous oxide emissions from mobile combustion decreased 24.9 MMT CO2 Eq. (60.4 percent) from 1990
through 2014. Nitrous oxide emissions from this source decreased primarily as a result of N2O national
emissions control standards and emissions control technologies for on-road vehicles.
• Nitrous oxide emissions from adipic acid production were 5.4 MMT CO2 Eq. in 2014, and have decreased
significantly since 1990 due to both the widespread installation of pollution control measures in the late
1990s and plant idling in the late 2000s. Emissions from adipic acid production have decreased by 64.2
percent since 1990 and by 67.8 percent since a peak in 1995.
Executive Summary ES-15
-------
i HFC, RFC, SF6, and NF3 Emissions
2 HFCs and PFCs are families of synthetic chemicals that are used as alternatives to Ozone Depleting Substances,
3 which are being phased out under the Montreal Protocol and Clean Air Act Amendments of 1990. HFCs and PFCs
4 do not deplete the stratospheric ozone layer, and are therefore acceptable alternatives under the Montreal Protocol.
5 These compounds, however, along with SF6 and NF3, are potent greenhouse gases. In addition to having high global
6 warming potentials, SF6 and PFCs have extremely long atmospheric lifetimes, resulting in their essentially
7 irreversible accumulation in the atmosphere once emitted. Sulfur hexafluoride is the most potent greenhouse gas the
8 IPCC has evaluated (IPCC 2013).
9 Other emissive sources of these gases include HCFC-22 production, electrical transmission and distribution systems,
10 semiconductor manufacturing, aluminum production, and magnesium production and processing (see Figure ES-10).
11 Figure ES-10: 2014 Sources of HFCs, PFCs, SFe, and NFs Emissions (MMT COz Eq.)
Substitution of Ozone Depleting Substances | ^f ^| | 171
Electrical Transmission and Distribution
HFCs, PFCs, SF6 and NF3 as a Portion
of all Emissions
HCFC-22 Production _
_ ^ ,«
Semiconductor Manufacture
Aluminum Production
Magnesium Production and Processing
'
10 20
j2 MMTCO2Eq.
13
14 Some significant trends in U.S. HFC, PFC, SF6, and NF3 emissions include the following:
15 • Emissions resulting from the substitution of ozone depleting substances (ODS) (e.g., CFCs) have been
16 consistently increasing, from small amounts in 1990 to 171.4 MMT CO2 Eq. in 2014. This increase was in
17 large part the result of efforts to phase out CFCs and other ODSs in the United States. In the short term,
18 this trend is expected to continue, and will likely continue over the next decade as HCFCs, which are
19 interim substitutes in many applications, are themselves phased-out under the provisions of the
20 Copenhagen Amendments to the Montreal Protocol.
21 • GWP-weighted PFC, HFC, SF6, and NF3 emissions from semiconductor manufacture have increased by 12
22 percent from 1990 to 2014, due to industrial growth and the adoption of emissions reduction technologies.
23 Within that time span, emissions peaked in 1999, the initial year of EPA's PFC Reduction / Climate
24 Partnership for the Semiconductor Industry, but have since declined to 4.0 MMT CO2 Eq. in 2014 (a 56
25 percent decrease relative to 1999).
26 • SF6 emissions from electric power transmission and distribution systems decreased by 79.9 percent (20.3
27 MMT CO2 Eq.) from 1990 to 2014. There are two potential causes for this decrease: (1) a sharp increase in
28 the price of SF6 during the 1990s and (2) a growing awareness of the environmental impact of SF6
29 emissions through programs such as EPA's SFe Emission Reduction Partnership for Electric Power
30 Systems.
ES-16 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
PFC emissions from aluminum production decreased by 86.2 percent (18.5 MMT CO2 Eq.) from 1990 to
2014. This decline is due both to reductions in domestic aluminum production and to actions taken by
aluminum smelting companies to reduce the frequency and duration of anode effects.
4 ES.3. Overview of Sector Emissions and Trends
5 In accordance with the UNFCCC decision to set the 2006IPCC Guidelines for National Greenhouse Gas
6 Inventories (IPCC 2006) as the standard for Annex I countries at the Nineteenth Conference of the Parties
7 (UNFCCC 2014), Figure ES-11 and Table ES-4 aggregate emissions and sinks by these chapters. Emissions of all
8 gases can be summed from each source category from IPCC guidance. Over the twenty-five-year period of 1990 to
9 2014, total emissions in the Energy, Industrial Processes and Product Use, Agriculture, and Waste sectors grew by
10 388.9 MMT CO2 Eq. (7.3 percent), 47.7 MMT CO2 Eq. (14.0 percent), 44.2 MMT CO2 Eq. (8.3 percent), and 1.5
11 MMT CO2 Eq. (0.7 percent), respectively. Over the same period, estimates of net C sequestration in the Land Use,
12 Land-Use Change, and Forestry (LULUCF) sector (magnitude of emissions plus CO2 removals from all LULUCF
13 source categories) decreased by 27.9 MMT CO2 Eq. (4.0 percent).
14 Figure ES-11: U.S. Greenhouse Gas Emissions and Sinks by Chapter/IPCC Sector (MMT COz
15 Eq.)
16
O"
LJJ
o
(J
Industrial Processes and
Product Uses
Waste
LULUCF (emissions)
Land Use, Land-Use Change an
(500)
(1,000)
(1,500) J
17
18
19
Table ES-4: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks by Chapter/IPCC
Sector (MMT COz Eq.)
Chapter/IPCC Sector
1990
2005
2010
2011
2012
2013
2014
Energy
Fossil Fuel Combustion
Natural Gas Systems*
Non-Energy Use of Fuels
Coal Mining
Stationary Combustion
5,699.0
5,227.7
194.8
108.5
71.2
28.4
5,481.3
5,024.7
189.2
105.6
66.5
28.0
5,637.4
5,157.6
195.2
121.7
64.6
30.9
5,679.8
5,208.7
195.2
114.3
64.6
31.5
Executive Summary ES-17
-------
Petroleum Systems'5
Mobile Combustion
Incineration of Waste
Abandoned Underground Coal Mines
Industrial Processes and Product Use
Substitution of Ozone Depleting
Substances
Iron and Steel Production &
Metallurgical Coke Production
Cement Production
Petrochemical Production
Lime Production
Other Process Uses of Carbonates
Nitric Acid Production
Ammonia Production
Aluminum Production
Adipic Acid Production
Electrical Transmission and
Distribution
Carbon Dioxide Consumption
N2O from Product Uses
Semiconductor Manufacture
HCFC-22 Production
Urea Consumption for Non-
Agricultural Purposes
Soda Ash Production and
Consumption
Ferroalloy Production
Titanium Dioxide Production
Magnesium Production and
Processing
Glass Production
Phosphoric Acid Production
Zinc Production
Lead Production
Silicon Carbide Production and
Consumption
Agriculture
Agricultural Soil Management
Enteric Fermentation
Manure Management
Rice Cultivation
Field Burning of Agricultural
Residues
Land Use, Land-Use Change, and
Forestry (Emissions)
Forest Fires
Urea Fertilization
Liming
Settlement Soils
Peatlands Remaining Peatlands
Forest Soils
Waste
Landfills
Wastewater Treatment
Composting
Total Emissions
LULUCF Total Net Fluxc
Forest Land Remaining Forest Landd
Cropland Remaining Cropland
Land Converted to Cropland0
Grassland Remaining Grassland0
36.0 28.4 25.5
46.9B 37.1 1 25.9
8.4
7.2
340.9
0.3
99.7
33.3
21.8
11.7
4.9
12.1
13.0
28.3
15.2
25.4
1.5
4.2
3.6
46.1
3.8
2.8
2.2
1.2
5.2
1.5
1.5
0.6
0.5
0.4
529.8
302.9
164.2
51.1
11.3
0.3
15.0
5.4
2.4
4.7
1.4
1.1
0.1
204.1
184.4
19.0
12.8
6.6
367.6
113.0
66.6
45.9
27.5
14.6
6.3
11.3
9.2
7.6
7.1
10.6
1.4
4.2
4.7
20.0
3.7
3.0
1.4
1.8
2.7
1.9
1.3
1.0
0.6
0.2
552.9
11.4
6.6
365.2
153.5
55.7
31.3
27.3
13.4
9.6
11.5
9.2
4.6
4.2
7.0
4.4
4.2
3.8
8.0
4.7
2.7
1.7
1.8
2.1
1.5
1.1
1.2
0.5
0.2
582.3
296.7 320.4
168.9 171.3
72.9 78.1
14.21 12.2
0.3M 0.4
28.2 17.8
16.5 5.4
3.5U 3.8
4.3M 4.8
2.3M 2.4
l.ll 1.0
0.5M 0.5
211.1 200.0
187.3 176.3
20.2 20.2
0.7 3_5_B 3.5
6,380.8 7,428.8 7,010.5
(704.2) (636.1) (683.2)
(576.0) (532.4) (585.0)
(43.2)| (16.5) (4.7)
22.8 • 14.6 15.6
(12.9)| 2.9| 2.6
26.4
24.7
10.9
6.4
382.2
157.1
59.9
32.0
26.4
14.0
9.3
10.9
9.3
6.8
10.2
6.8
4.1
4.2
4.9
8.8
4.0
2.7
1.7
1.7
2.8
1.3
1.2
1.3
0.5
0.2
583.3
322.9
168.9
78.9
12.2
0.4
22.9
11.0
4.1
3.9
2.5
0.9
0.5
200.5
176.9
20.1
3.5
6,887.8
(683.6)
(578.1)
(20.0)
14.2
11.3
28.3
22.2
10.7
6.2
371.4
161.4
54.2
35.1
26.5
13.7
8.0
10.5
9.4
6.4
5.5
5.7
4.0
4.2
4.5
5.5
4.4
2.8
1.9
1.5
1.7
1.2
1.1
1.5
0.5
0.2
583.4
322.9
166.7
81.2
12.2
0.4
32.3
18.3
4.2
6.0
2.5
0.8
0.5
197.2
173.5
20.0
3.7
6,665.7
(680.8)
(576.7)
(18.7)
14.5
11.7
20.3
9.7
6.2
373.8
165.3
52.2
36.1
26.5
14.0
10.4
10.7
10.0
6.2
4.0
5.1
4.2
4.2
4.2
4.1
4.2
2.8
1.8
1.7
1.5
1.3
1.1
1.4
0.5
0.2
575.4
318.4
165.5
78.9
12.2
0.4
24.1
12.2
4.3
3.9
2.4
0.8
0.5
200.5
176.7
20.0
3.9
6,811.2
(682.4)
(580.1)
(16.8)
14.8
11.9
31.2
18.4
9.7
6.2
388.6
171.4
55.4
38.8
26.6
14.1
12.1
10.9
9.4
6.2
5.4
5.1
4.5
4.2
4.2
4.1
4.0
2.8
1.9
1.8
1.5
1.3
1.1
1.0
0.5
0.2
574.1
318.5
164.3
78.7
12.2
0.4
24.6
12.2
4.5
4.1
2.4
0.8
0.5
205.6
181.8
20.0
3.9
6,872.6
(685.8)
(583.4)
(16.0)
14.7
11.9
ES-18 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Land Converted to Grassland
Settlements Remaining Settlements
Other: Landfilled Yard Trimmings
and Food Scraps
Net Emissions (Sources and Sinks)
(8-5)1
(60.4)1
(26.0)
5,676.6
(12.8)1
(80.5)1
(11.4)
6,792.6
(12.3)
(86.1)
(13.2)
6,327.3
(11.0)
(87.3)
(12.7)
6,204.2
(10.9)
(88.4)
(12.2)
5,984.9
(10.9)
(89.5)
(11.7)
6,128.8
(10.9)
(90.6)
(11.6)
6,186.8
1 The values in this table for Natural Gas Systems are presented from the previous Inventory and do not reflect updates to emission estimates
for this category. See Section 3.7, Natural Gas Systems of the Energy chapter for more information. Gray highlighting was added on 2/24 for
clarification.
b The values in this table for Petroleum Systems are presented from the previous Inventory and do not reflect updates to emission estimates for
this category. See Section 3.6, Petroleum Systems of the Energy chapter for more information. Gray highlighting was added on 2/24 for
clarification.
c Total net flux from LULUCF is only included in the Net Emissions total. Net flux from LULUCF includes the positive C sequestration
reported for Forest Land Remaining Forest Land, Land Converted to Forest Land, Cropland Remaining Cropland, Land Converted to
Grassland, Settlements Remaining Settlements, and Other Land plus the loss in C sequestration reported for Land Converted to Cropland and
Grassland Remaining Grassland. Refer to Table ES-5 for a breakout of emissions and removals for Land Use, Land-Use Change, and
Forestry by gas and source category.
''Includes the effects of net additions to stocks of carbon stored in forest ecosystem pools and harvested wood products.
Note: Totals may not sum due to independent rounding. Parentheses indicate negative values or sequestration.
i Energy
2 The Energy chapter contains emissions of all greenhouse gases resulting from stationary and mobile energy
3 activities including fuel combustion and fugitive fuel emissions. Energy-related activities, primarily fossil fuel
4 combustion, accounted for the vast majority of U.S. CC>2 emissions for the period of 1990 through 2014. In 2014,
5 approximately 82 percent of the energy consumed in the United States (on a Btu basis) was produced through the
6 combustion of fossil fuels. The remaining 18 percent came from other energy sources such as hydropower, biomass,
7 nuclear, wind, and solar energy (see Figure ES-12). Energy-related activities are also responsible for CH4 and N2O
8 emissions (37 percent and 10 percent of total U.S. emissions of each gas, respectively). Overall, emission sources in
9 the Energy chapter account for a combined 82.6 percent of total U.S. greenhouse gas emissions in 2014.
10 Figure ES-12: 2014 U.S. Energy Consumption by Energy Source
Nuclear Electric
Power
8.5%
Renewable
Energy
9.8%
11
12 Industrial Processes and Product Use
13 The Industrial Processes and Product Use (IPPU) chapter includes greenhouse gas emissions occurring from
14 industrial processes and from the use of greenhouse gases in products.
15 Greenhouse gas emissions are produced as the by-products of many non-energy-related industrial activities. For
16 example, industrial processes can chemically transform raw materials, which often release waste gases such as €62,
17 CH4, and N2O. These processes include iron and steel production and metallurgical coke production, cement
18 production, ammonia production, urea consumption, lime production, other process uses of carbonates (e.g., flux
Executive Summary ES-19
-------
1 stone, flue gas desurfurization, and glass manufacturing), soda ash production and consumption, titanium dioxide
2 production, phosphoric acid production, ferroalloy production, CO2 consumption, silicon carbide production and
3 consumption, aluminum production, petrochemical production, nitric acid production, adipic acid production, lead
4 production, zinc production, and N2O from product uses. Industrial processes also release HFCs, PFCs, SF6, and
5 NF3. In addition to their use as ODS substitutes, HFCs, PFCs, SF6, NF3, and other fluorinated compounds are
6 employed and emitted by a number of other industrial sources in the United States. These industries include
7 aluminum production, HCFC-22 production, semiconductor manufacture, electric power transmission and
8 distribution, and magnesium metal production and processing. Overall, emission sources in the Industrial Process
9 and Product Use chapter account for 5.7 percent of U.S. greenhouse gas emissions in 2014.
10 Agriculture
11 The Agriculture chapter contains anthropogenic emissions from agricultural activities (except fuel combustion,
12 which is addressed in the Energy chapter, and agricultural CO2 fluxes, which are addressed in the Land Use, Land-
13 Use Change, and Forestry chapter). Agricultural activities contribute directly to emissions of greenhouse gases
14 through a variety of processes, including the following source categories: enteric fermentation in domestic livestock,
15 livestock manure management, rice cultivation, agricultural soil management, and field burning of agricultural
16 residues. CH4 and N2O were the primary greenhouse gases emitted by agricultural activities. CH4 emissions from
17 enteric fermentation and manure management represented 23.2 percent and 8.6 percent of total CH4 emissions from
18 anthropogenic activities, respectively, in 2014. Agricultural soil management activities such as fertilizer application
19 and other cropping practices were the largest source of U.S. N2O emissions in 2014, accounting for 77.4 percent. In
20 2014, emission sources accounted for in the Agricultural chapters were responsible for 8.4 percent of total U.S.
21 greenhouse gas emissions.
22 Land Use, Land-Use Change, and Forestry
23 The Land Use, Land-Use Change, and Forestry chapter contains emissions of CH4 and N2O, and emissions and
24 removals of CO2 from forest management, other land-use activities, and land-use change. Forest management
25 practices, tree planting in urban areas, the management of agricultural soils, and the landfilling of yard trimmings
26 and food scraps resulted in a net removal of CO2 (sequestration of C) in the United States. Forests (including
27 vegetation, soils, and harvested wood) accounted for 85 percent of total 2014 CO2 removals, urban trees accounted
28 for 13 percent, mineral and organic soil carbon stock changes accounted for less than 0.5 percent, and landfilled yard
29 trimmings and food scraps accounted for 1.7 percent of the total CO2 removals in 2014. The net forest sequestration
30 is a result of net forest growth and increasing forest area, as well as a net accumulation of carbon stocks in harvested
31 wood pools. The net sequestration in urban forests is a result of net tree growth in these areas. In agricultural soils,
32 mineral and organic soils sequester approximately 1.8 times as much C as is emitted from these soils through liming
33 and urea fertilization. The mineral soil C sequestration is largely due to the conversion of cropland to permanent
34 pastures and hay production, a reduction in summer fallow areas in semi-arid areas, an increase in the adoption of
35 conservation tillage practices, and an increase in the amounts of organic fertilizers (i.e., manure and sewage sludge)
36 applied to agriculture lands. The landfilled yard trimmings and food scraps net sequestration is due to the long-term
37 accumulation of yard trimming carbon and food scraps in landfills.
38 Land use, land-use change, and forestry activities in 2014 resulted in a C sequestration (i.e., net CO2 removals) of
39 685.8 MMT CO2 Eq. (Table ES-5). 20 This represents an offset of 10.0 percent of total (i.e., gross) greenhouse gas
40 emissions in 2014. Emissions from land use, land-use change, and forestry activities in 2014 represent 0.4 percent
41 of total greenhouse gas emissions.21 Between 1990 and 2014, total land use, land-use change, and forestry C
42 sequestration decreased by 2.6 percent, primarily due to a decrease in the rate of net C accumulation in agricultural
2" Net flux from LULUCF includes the positive C sequestration reported for Forest Land Remaining Forest Land, Land
Converted to Forest Land, Cropland Remaining Cropland, Land Converted to Grassland, Settlements Remaining Settlements,
and Other Land plus the loss in C sequestration reported for Land Converted to Cropland and Grassland Remaining Grassland.
LULUCF emissions include the CCh, CH4, andN2O emissions reported for Forest Fires, Forest Soils, Liming, Urea
Fertilization, Settlement Soils, and Peatlands Remaining Peatlands.
ES-20 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 soil carbon stocks. Annual C accumulation in landfilled yard trimmings and food scraps slowed over this period,
2 while the rate of annual C accumulation increased in urban trees.
3 CO2 removals are presented in Table ES-5 along with CO2, CH4, and N2O emissions for Land Use, Land-Use
4 Change, and Forestry source categories. 22 Liming and urea fertilization in 2014 resulted in CO2 emissions of 8.7
5 MMT CO2 Eq. (8,653 kt). Lands undergoing peat extraction (i.e., Peatlands Remaining Peatlands) resulted in CO2
6 emissions of 0.8 MMT CO2 Eq. (842 kt) and CH4 and N2O emissions of less than 0.05 MMT CO2 Eq. each. The
7 application of synthetic fertilizers to forest soils in 2014 resulted in N2O emissions of 0.5 MMT CO2 Eq. (2 kt).
8 N2O emissions from fertilizer application to forest soils have increased by 455 percent since 1990, but still account
9 for a relatively small portion of overall emissions. Additionally, N2O emissions from fertilizer application to
10 settlement soils in 2014 accounted for 2.4 MMT CO2 Eq. (8 kt). This represents an increase of 78 percent since
11 1990. Forest fires in 2014 resulted in CH4 emissions of 7.3 MMT CO2Eq. (294 kt), and in N2O emissions of 4.8
12 MMTCO2Eq. (16 kt).
13 Table ES-5: Emissions and Removals (Flux) from Land Use, Land-Use Change, and Forestry
14 (MMT COz Eq.)
~Gas/Land-Use Category 1990 2005 2010 2011 2012 2013 2014~
Net CO2 Flux3 (704.2)
Forest Land Remaining Forest Landb (576.0)
Land Converted to Forest Land
Cropland Remaining Cropland (43.2)
Land Converted to Cropland3 22.8
Grassland Remaining Grassland* (12.9)
Land Converted to Grassland (8.5)
Settlements Remaining Settlements0 (60.4)
Other: Landfilled Yard Lrimmings and
Food Scraps (26.0)
C02 8.1
Cropland Remaining Cropland: Liming 4.7
Cropland Remaining Cropland: Urea
Fertilization 2.4
Wetlands Remaining Wetlands:
Peatlands Remaining Peatlands 1.1
CH4 3.3
Forest Land Remaining Forest Land:
Forest Fires 3.3
Wetlands Remaining Wetlands:
Peatlands Remaining Peatlands +
N20 3.6
Forest Land Remaining Forest Land:
Forest Fires 2.2
Forest Land Remaining Forest Land:
Forest Soils'1 0.1
Settlements Remaining Settlements:
Settlement Soils6 1.4
Wetlands Remaining Wetlands:
Peatlands Remaining Peatlands +_
(636.1)
(532.4)
(683.2)
(585.0)
(4.7)
15.6
2.6
(12.3)
(86.1)
(683.6) (680.8) (682.4) (685.8)
(578.1) (576.7) (580.1) (583.4)
(20.0)
14.2
11.3
(11.0)
(87.3)
(18.7)
14.5
11.7
(10.9)
(88.4)
(16.8)
14.8
11.9
(10.9)
(89.5)
(16.0)
14.7
11.9
(10.9)
(90.6)
(12.7) (12.2) (11.7) (11.6)
8.9 11.0 9.0 9.5
3.9 6.0 3.9 4.1
3.5
1.1
9.9
9.9
9.3
6.5
0.5
2.3
+
3.8
1.0
3.3
3.3
5.0
2.2
0.5
2.4
4.1
0.9
6.6
6.6
7.3
4.4
0.5
2.5
4.2
0.8
11.1
11.1
10.3
7.3
0.5
2.5
4.3
0.8
7.3
7.3
7.7
4.8
0.5
2.4
4.5
0.8
7.4
7.3
7.7
4.8
0.5
2.4
LULUCF Emissions'
LULUCF Total Net Flux"
15.0
(704.2)
28.2
(636.1)
17.8
(683.2)
22.9
(683.6)
32.3
(680.8)
24.1
(682.4)
24.6
(685.8)
LULUCF Sector TotaP
(689.1)
(607.9)
(665.3) (660.7) (648.5) (658.3) (661.3)
Note: Lotals may not sum due to independent rounding. Parentheses indicate net sequestration.
+ Does not exceed 0.05 MMT CO2 Eq.
a Lotal net flux from LULUCF is only included in the Net Emissions total. Net flux from LULUCF includes the positive C
sequestration reported for Forest Land Remaining Forest Land, Land Converted to Forest Land, Cropland Remaining
Cropland, Land Converted to Grassland, Settlements Remaining Settlements, and Other Land plus the loss in C
sequestration reported for Land Converted to Cropland and Grassland Remaining Grassland.
b Includes the effects of net additions to stocks of carbon stored in forest ecosystem pools and harvested wood products.
22 Estimates from Land Converted to Forest Land are currently under development.
Executive Summary ES-21
-------
0 Estimates include C stock changes on both Settlements Remaining Settlements and Land Converted to Settlements.
d Estimates include emissions from N fertilizer additions on both Forest Land Remaining Forest Land, and Land Converted
to Forest Land, but not from land-use conversion.
e Estimates include emissions from N fertilizer additions on both Settlements Remaining Settlements, and Land Converted to
Settlements, but not from land-use conversion.
f LULUCF emissions include the CCh, CELi, andN2O emissions reported for Forest Fires, Forest Soils, Liming, Urea
Fertilization, Settlement Soils, and Peatlands Remaining Peatlands.
g The LULUCF Sector Total is the sum of positive emissions (i.e., sources) of greenhouse gases to the atmosphere plus
removals of CCh (i.e., sinks or negative emissions) from the atmosphere.
i Waste
2 The Waste chapter contains emissions from waste management activities (except incineration of waste, which is
3 addressed in the Energy chapter). Landfills were the largest source of anthropogenic greenhouse gas emissions in
4 the Waste chapter, accounting for 88.4 percent of this chapter's emissions, and 25.7 percent of total U.S. CH4
5 emissions.23 Additionally, wastewater treatment accounts for 9.7 percent of Waste emissions, 2.1 percent of U.S.
6 CH4 emissions, and 1.2 percent of U.S. N2O emissions. Emissions of CH4 and N2O from composting are also
7 accounted for in this chapter, generating emissions of 2.1 MMT CCh Eq. and 1.8MMT CC>2 Eq., respectively.
8 Overall, emission sources accounted for in the Waste chapter generated 3.0 percent of total U. S. greenhouse gas
9 emissions in 2014.
10
11
ES.4. Other Information
Emissions by Economic Sector
12 Throughout the Inventory of U.S. Greenhouse Gas Emissions and Sinks report, emission estimates are grouped into
13 five sectors (i.e., chapters) defined by the IPCC: Energy; Industrial Processes and Product Use; Agriculture; Land
14 Use, Land-Use Change, and Forestry; and Waste. While it is important to use this characterization for consistency
15 with UNFCCC reporting guidelines, it is also useful to allocate emissions into more commonly used sectoral
16 categories. This section reports emissions by the following economic sectors: Residential, Commercial, Industry,
17 Transportation, Electricity Generation, Agriculture, and U.S. Territories.
18 Table ES-6 summarizes emissions from each of these economic sectors, and Figure ES-13 shows the trend in
19 emissions by sector from 1990 to 2014.
Landfills also store carbon, due to incomplete degradation of organic materials such as harvest wood products, yard
trimmings, and food scraps, as described in the Land-Use, Land-Use Change, and Forestry chapter of the Inventory report.
ES-22 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Figure ES-13: Emissions Allocated to Economic Sectors (MMT COz Eq.)
2,500
2,000
S 1-500
1,000 -
500 -
Electric
Power Industry
Transportation
Industry
Agriculture
• Commercial (Red)
1 Residential (Blue)
Table ES-6: U.S. Greenhouse Gas Emissions Allocated to Economic Sectors (MMT COz Eq.)
Implied Sectors
Electric Power Industry
Transportation
Industry
Agriculture
Commercial
Residential
U.S. Territories
Total Emissions
LULUCF Emissions3
LULUCF Total Net Fluxb
LULUCF Sector Total0
Net Emissions (Sources and Sinks)
1990
1,864.8
1,551.3
1,586.9
574.9
422.9
346.3
33.7
6,380.8
15.0
(704.2)
(689.1)
5,676.6
• 2005
2,443
2,012
1,460
626
453
372
58
7,428
.9
.9
.6
.8
.6
.8
.2
.8
28.2
(636.
1)
(607.9)
6,792
.6
2010
2,300.5
1,839.7
1,354.9
646.6
459.9
363.6
45.3
7,010.5
17.8
(683.2)
(665.3)
6,327.3
2
1
1
6
2011
,198.1
,811.3
,353.9
654.2
464.7
360.1
45.4
,887.8
22.9
(683.6)
(660.7)
6
,204.2
2012
2,060.8
1,791.6
1,339.5
665.3
440.0
320.9
47.6
6,665.7
32.3
(680.8)
(648.5)
5,984.9
2013
2,077.7
1,800.5
1,392.1
648.0
470.2
375.1
47.5
6,811.2
24.1
(682.4)
(658.3)
6,128.8
2014
2,080.2
1,820.3
1,395.5
648.0
487.8
396.1
44.7
6,872.6
24.6
(685.8)
(661.3)
6,186.8
1 LULUCF emissions include the CCh, CELi, andN2O emissions reported for Forest Fires, Forest Soils, Liming, Urea
Fertilization, Settlement Soils, and Peatlands Remaining Peatlands.
b Total net flux from LULUCF is only included in the Net Emissions total. Net flux from LULUCF includes the positive C
sequestration reported for Forest Land Remaining Forest Land, Land Converted to Forest Land, Cropland Remaining
Cropland, Land Converted to Grassland, Settlements Remaining Settlements, and Other Land plus the loss in C
sequestration reported for Land Converted to Cropland and Grassland Remaining Grassland. Refer to Table ES-5 for a
breakout of emissions and removals for Land Use, Land-Use Change, and Forestry by gas and source category.
c The LULUCF Sector Total is the sum of positive emissions (i.e., sources) of greenhouse gases to the atmosphere plus
removals of CCh (i.e., sinks or negative emissions) from the atmosphere.
Note: Totals may not sum due to independent rounding. Parentheses indicate negative values or sequestration.
4 Using this categorization, emissions from electricity generation accounted for the largest portion (30 percent) of
5 U.S. greenhouse gas emissions in 2014. Transportation activities, in aggregate, accounted for the second largest
6 portion (26 percent), while emissions from industry accounted for the third largest portion (20 percent) of U.S.
7 greenhouse gas emissions in 2014. In contrast to electricity generation and transportation, emissions from industry
8 have in general declined over the past decade. The long-term decline in these emissions has been due to structural
9 changes in the U.S. economy (i.e., shifts from a manufacturing-based to a service-based economy), fuel switching,
10 and energy efficiency improvements. The remaining 23 percent of U.S. greenhouse gas emissions were contributed
Executive Summary ES-23
-------
1 by, in order of magnitude, the agriculture, commercial, and residential sectors, plus emissions from U.S. Territories.
2 Activities related to agriculture accounted for 9 percent of U.S. emissions; unlike other economic sectors,
3 agricultural sector emissions were dominated by N2O emissions from agricultural soil management and CH4
4 emissions from enteric fermentation. The commercial and residential sectors accounted for 7 percent and 6 percent
5 of emissions, respectively, and U.S. Territories accounted for 1 percent of emissions; emissions from these sectors
6 primarily consisted of CC>2 emissions from fossil fuel combustion. CC>2 was also emitted and sequestered by a
7 variety of activities related to forest management practices, tree planting in urban areas, the management of
8 agricultural soils, and landfilling of yard trimmings.
9 Electricity is ultimately consumed in the economic sectors described above. Table ES-7 presents greenhouse gas
10 emissions from economic sectors with emissions related to electricity generation distributed into end-use categories
11 (i.e., emissions from electricity generation are allocated to the economic sectors in which the electricity is
12 consumed). To distribute electricity emissions among end-use sectors, emissions from the source categories
13 assigned to electricity generation were allocated to the residential, commercial, industry, transportation, and
14 agriculture economic sectors according to retail sales of electricity.24 These source categories include CCh from
15 fossil fuel combustion and the use of limestone and dolomite for flue gas desulfurization, €62 and N2O from
16 incineration of waste, CH4 and N2O from stationary sources, and SF6 from electrical transmission and distribution
17 systems.
18 When emissions from electricity are distributed among these sectors, industrial activities and transportation account
19 for the largest shares of U.S. greenhouse gas emissions (28 percent and 27 percent, respectively) in 2014. The
20 residential and commercial sectors contributed the next largest shares of total U.S. greenhouse gas emissions in
21 2014. Emissions from these sectors increase substantially when emissions from electricity are included, due to their
22 relatively large share of electricity consumption (e.g., lighting, appliances, etc.). In all sectors except agriculture,
23 CO2 accounts for more than 78 percent of greenhouse gas emissions, primarily from the combustion of fossil fuels.
24 Figure ES-14 shows the trend in these emissions by sector from 1990 to 2014.
25 Table ES-7: U.S Greenhouse Gas Emissions by Economic Sector with Electricity-Related
26 Emissions Distributed (MMT COz Eq.)
Implied Sectors
Industry
Transportation
Commercial
Residential
Agriculture
U.S. Territories
Total Emissions
LULUCF Emissions3
LULUCF Total Net Fluxb
LULUCF Sector Total0
Net Emissions (Sources and Sinks)
1990
2,228.9
1 ,554.4B
973.9B
953.6
636.3B
33.7 1
6,380.8
15.0
(704.2m
(689.1)
5,676.6
2005
2,146.3
2,017.7
1,271.2
1,244 .4(
690.9B
58.2
7,428.8
28.2
(636.1)B
(607.9)
6,792.6
2010
1,939.4
1,844.3
1,247.2
1,219.3
715.0
45.3
7,010.5
17.8
(683.2)
(665.3)
6,327.3
2011
1,924.9
1,815.7
1,216.6
1,165.6
719.6
45.4
6,887.8
22.9
(683.6)
(660.7)
6,204.2
2012
1,881.4
1,795.5
1,153.7
1,060.1
727.4
47.6
6,665.7
32.3
(680.8)
(648.5)
5,984.9
2013
1,936.4
1,804.6
1,188.4
1,124.2
710.1
47.5
6,811.2
24.1
(682.4)
(658.3)
6,128.8
2014
1,939.4
1,824.4
1,208.5
1,146.1
709.5
44.7
6,872.6
24.6
(685.8)
(661.3)
6,186.8
Note: Emissions from electricity generation are allocated based on aggregate electricity consumption in each end-use sector.
a LULUCF emissions include the CCh, CELi, andN2O emissions reported for Forest Fires, Forest Soils, Liming, Urea
Fertilization, Settlement Soils, and Peatlands Remaining Peatlands.
b Total net flux from LULUCF is only included in the Net Emissions total. Net flux from LULUCF includes the positive C
sequestration reported for Forest Land Remaining Forest Land, Land Converted to Forest Land, Cropland Remaining Cropland,
Land Converted to Grassland, Settlements Remaining Settlements, and Other Land plus the loss in C sequestration reported for
Land Converted to Cropland and Grassland Remaining Grassland. Refer to Table ES-5 for a breakout of emissions and
removals for Land Use, Land-Use Change, and Forestry by gas and source category.
c The LULUCF Sector Total is the sum of positive emissions (i.e., sources) of greenhouse gases to the atmosphere plus removals
of CO2 (i.e., sinks or negative emissions) from the atmosphere.
Note: Totals may not sum due to independent rounding. Parentheses indicate negative values or sequestration.
Emissions were not distributed to U.S. territories, since the electricity generation sector only includes emissions related to the
generation of electricity in the 50 states and the District of Columbia.
ES-24 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Figure ES-14: Emissions with Electricity Distributed to Economic Sectors (MMT COz Eq.)
2
2,500
2,000
CT 1,500
LJJ
8
1,000
500
Industry (Green)
Transportation
(Purple)
Commercial (Red)
Residential (Blue)
Agriculture
1-1 IN m T LO
i
O O O O O O O
Box ES- 2: Recent Trends in Various U.S. Greenhouse Gas Emissions-Related Data
4 Total emissions can be compared to other economic and social indices to highlight changes over time. These
5 comparisons include: (1) emissions per unit of aggregate energy consumption, because energy-related activities are
6 the largest sources of emissions; (2) emissions per unit of fossil fuel consumption, because almost all energy-related
7 emissions involve the combustion of fossil fuels; (3) emissions per unit of electricity consumption, because the
8 electric power industry—utilities and non-utilities combined—was the largest source of U.S. greenhouse gas
9 emissions in 2014; (4) emissions per unit of total gross domestic product as a measure of national economic activity;
10 and (5) emissions per capita.
11 Table ES-8 provides data on various statistics related to U.S. greenhouse gas emissions normalized to 1990 as a
12 baseline year. Greenhouse gas emissions in the United States have grown at an average annual rate of 0.3 percent
13 since 1990. Since 1990, this rate is slightly slower than that for total energy and for fossil fuel consumption, and
14 much slower than that for electricity consumption, overall gross domestic product and national population (see
15 Figure ES-15).
16 Table ES-8: Recent Trends in Various U.S. Data (Index 1990 = 100)
Variable
1990
2005
2010 2011 2012
Avg. Annual
2013 2014 Growth Rate
Greenhouse Gas Emissions*
Energy Consumption15
Fossil Fuel Consumption15
Electricity Consumption15
GDPC
Population*1
a GWP-weighted values
b Energy content-weighted values (EIA 2015a)
0 Gross Domestic Product in chained 2009 dollars (BEA 2015)
d U.S. Census Bureau (2015)
0.3%
0.7%
0.5%
1.4%
2.5%
1.0%
Executive Summary ES-25
-------
1 Figure ES-15: U.S. Greenhouse Gas Emissions Per Capita and Per Dollar of Gross Domestic
2 Product
3
4
ii
o
175 -
165 -
155 -
145 -
135 -
125 -
115 -
105 -
95 -
85 -
75 -
65 -
55
Real GDP
Population
Emissions
per capita
Emissions
per $GDP
g
Source: BEA (2015), U.S. Census Bureau (2015), and emission estimates in this report.
Key Categories
7 The 2006IPCC Guidelines (IPCC 2006) defines a key category as a "[category] that is prioritized within the
8 national inventory system because its estimate has a significant influence on a country's total inventory of
9 greenhouse gases in terms of the absolute level, the trend, or the uncertainty in emissions and removals."25 By
10 definition, key categories are sources or sinks that have the greatest contribution to the absolute overall level of
11 national emissions in any of the years covered by the time series. In addition, when an entire time series of emission
12 estimates is prepared, a thorough investigation of key categories must also account for the influence of trends of
13 individual source and sink categories. Finally, a qualitative evaluation of key categories should be performed, in
14 order to capture any key categories that were not identified in either of the quantitative analyses.
15 Figure ES-16 presents 2014 emission estimates for the key categories as defined by a level analysis (i.e., the
16 contribution of each source or sink category to the total inventory level). The UNFCCC reporting guidelines request
17 that key category analyses be reported at an appropriate level of disaggregation, which may lead to source and sink
18 category names which differ from those used elsewhere in the Inventory report. For more information regarding key
19 categories, see Section 1.6 - Key Categories and Annex 1.
See Chapter 4 "Methodological Choice and Identification of Key Categories" in IPCC (2006).
ES-26 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Figure ES-16: 2014 Key Categories (MMTCOz Eq.) TO BE UPDATED
CO2 Emissions from Stationary Combustion - Coal - Elec. Gen.
CO2 Emissions from Mobile Combustion: Road
CO2 Emissions from Stationary Combustion - Gas - Industrial
CO2 Emissions from Stationary Combustion - Gas - Sec. Gen.
CO2 Emissions from Stationary Combustion - Gas - Residential
C02 Emissions from Stationary Combustion - Oil - Industrial
Direct N20 Emissions from Agricultural Soil Management
C02 Emissions from Stationary Combustion - Gas - Commercial
CH4 Emissions from Landfills
Emissions from Substitutes for Ozone Depleting Substances
CH4 Emissions from Enteric Fermentation
CH4 Emissions from Natural Gas Systems'
CO2 Emissions from Mobile Combustion: Aviation
CO2 Emissions from Non-Energy Use of Fuels
CO2 Emissions from Mobile Combustion: Other
CO2 Emissions from Stationary Combustion - Coal - Industrial
CO2 Emissions from Stationary Combustion - Oil - Residential
Fugitive Emissions from Coal Mining
CH4 Emissions from Manure Management
Indirect N2O Emissions from Applied Nitrogen
CO2 Emissions from Iron & Steel Prod. & Metallurgical Coke Prod.
CO2 Emissions from Cement Production
CO2 Emissions from Stationary Combustion - Oil - Commercial
CO2 Emissions from Natural Gas Systems"
C02 Emissions from Stationary Combustion - Oil - U.S. Territories
CO2 Emissions from Mobile Combustion: Marine
CH4 Emissions from Petroleum Systems"
Non-CO2 Emissions from Stationary Combustion - Bee Gen.
CH4 Emissions from Rice Cultivation
Non-CO2 Emissions from Stationary Combustion - Residential
Key Categories as a Portion of All Emissions
200 400 600 800 1,000 1,200 1,400 1,600 1,800
MMTCO2Eq.
j
4
5
6
7
10
11
12
13
14
15
16
17
18
19
Note: For a complete discussion of the key category analysis, see Annex 1. Blue bars indicate either an Approach 1, or Approach
1 and Approach 2 level assessment key category. Gray bars indicate solely an Approach 2 level assessment key category.
a The value in this figure for Natural Gas Systems is presented from the previous Inventory and does not reflect updates to
emission estimates for this category. See Section 3.7, Natural Gas Systems of the Energy chapter for more information.
b The value in this figure for Petroleum Systems is presented from the previous Inventory and does not reflect updates to
emission estimates for this category. See Section 3.6, Petroleum Systems of the Energy chapter for more information.
Quality Assurance and Quality Control (QA/QC)
The United States seeks to continually improve the quality, transparency, and credibility of the Inventory of U.S.
Greenhouse Gas Emissions and Sinks. To assist in these efforts, the United States implemented a systematic
approach to QA/QC. While QA/QC has always been an integral part of the U.S. national system for inventory
development, the procedures followed for the current inventory have been formalized in accordance with the
QA/QC plan and the UNFCCC reporting guidelines.
Uncertainty Analysis of Emission Estimates
Uncertainty estimates are an essential element of a complete inventory of greenhouse gas emissions and removals.
Some of the current estimates, such as those for CCh emissions from energy-related activities and cement
processing, are considered to have low uncertainties. For some other categories of emissions, however, a lack of
data or an incomplete understanding of how emissions are generated increases the uncertainty associated with the
Executive Summary ES-27
-------
1 estimates presented. Acquiring a better understanding of the uncertainty associated with inventory estimates is an
2 important step in helping to prioritize future work and improve the overall quality of the Inventory. Recognizing the
3 benefit of conducting an uncertainty analysis, the UNFCCC reporting guidelines follow the recommendations of the
4 2006IPCC Guidelines (IPCC 2006) and require that countries provide single estimates of uncertainty for source and
5 sink categories.
6 Currently, a qualitative discussion of uncertainty is presented for all source and sink categories. Within the
7 discussion of each emission source, specific factors affecting the uncertainty surrounding the estimates are
8 discussed. Most sources also contain a quantitative uncertainty assessment, in accordance with UNFCCC reporting
9 guidelines.
10
Box ES- 3: Recalculations of Inventory Estimates
12 Each year, emission and sink estimates are recalculated and revised for all years in the Inventory of U.S. Greenhouse
13 Gas Emissions and Sinks, as attempts are made to improve both the analyses themselves, through the use of better
14 methods or data, and the overall usefulness of the report. In this effort, the United States follows the 2006 IPCC
15 Guidelines (IPCC 2006), which states, "Both methodological changes and refinements over time are an essential
16 part of improving inventory quality. It is good practice to change or refine methods" when: available data have
17 changed; the previously used method is not consistent with the IPCC guidelines for that category; a category has
18 become key; the previously used method is insufficient to reflect mitigation activities in a transparent manner; the
19 capacity for inventory preparation has increased; new inventory methods become available; and for correction of
20 errors." In general, recalculations are made to the U.S. greenhouse gas emission estimates either to incorporate new
21 methodologies or, most commonly, to update recent historical data.
22 In each Inventory report, the results of all methodology changes and historical data updates are presented in the
23 "Recalculations and Improvements" chapter; detailed descriptions of each recalculation are contained within each
24 source's description contained in the report, if applicable. In general, when methodological changes have been
25 implemented, the entire time series (in the case of the most recent Inventory report, 1990 through 2013) has been
26 recalculated to reflect the change, per the 2006 IPCC Guidelines (IPCC 2006). Changes in historical data are
27 generally the result of changes in statistical data supplied by other agencies. References for the data are provided for
28 additional information.
29
ES-28 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1. Introduction
2 This report presents estimates by the United States government of U.S. anthropogenic greenhouse gas emissions and
3 sinks for the years 1990 through 2014. A summary of these estimates is provided in Table 2-1 and Table 2-2 by gas
4 and source category in the Trends in Greenhouse Gas Emissions chapter. The emission estimates in these tables are
5 presented on both a full molecular mass basis and on a Global Warming Potential (GWP) weighted basisl in order to
6 show the relative contribution of each gas to global average radiative forcing. This report also discusses the
7 methods and data used to calculate these emission estimates.
8 In 1992, the United States signed and ratified the United Nations Framework Convention on Climate Change
9 (UNFCCC). As stated in Article 2 of the UNFCCC, "The ultimate objective of this Convention and any related
10 legal instruments that the Conference of the Parties may adopt is to achieve, in accordance with the relevant
11 provisions of the Convention, stabilization of greenhouse gas concentrations in the atmosphere at a level that would
12 prevent dangerous anthropogenic interference with the climate system. Such a level should be achieved within a
13 time-frame sufficient to allow ecosystems to adapt naturally to climate change, to ensure that food production is not
14 threatened and to enable economic development to proceed in a sustainable manner."2'3
15 Parties to the Convention, by ratifying, "shall develop, periodically update, publish and make available... national
16 inventories of anthropogenic emissions by sources and removals by sinks of all greenhouse gases not controlled by
17 the Montreal Protocol, using comparable methodologies..."4 The United States views this report as an opportunity
18 to fulfill these commitments under the UNFCCC.
19 In 1988, preceding the creation of the UNFCCC, the World Meteorological Organization (WMO) and the United
20 Nations Environment Programme (UNEP) jointly established the Intergovernmental Panel on Climate Change
21 (IPCC). The role of the IPCC is to assess on a comprehensive, objective, open and transparent basis the scientific,
22 technical and socio-economic information relevant to understanding the scientific basis of risk of human-induced
23 climate change, its potential impacts and options for adaptation and mitigation (IPCC 2003). Under Working Group
24 1 of the IPCC, nearly 140 scientists and national experts from more than thirty countries collaborated in the creation
25 of the Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC/UNEP/OECD/IEA 1997) to
26 ensure that the emission inventories submitted to the UNFCCC are consistent and comparable between nations.
27 The IPCC Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories and the
28 IPCC Good Practice Guidance for Land Use, Land-Use Change, and Forestry further expanded upon the
29 methodologies in the Revised 1996 IPCC Guidelines. In 2006, the IPCC accepted the 2006 Guidelines for National
30 Greenhouse Gas Inventories at its Twenty-Fifth Session (Mauritius, April 2006). The 2006 IPCC Guidelines built
1 More information provided in "Global Warming Potentials" section of this chapter on the use of IPCC Fourth Assessment
Report (AR4) GWP values.
2 The term "anthropogenic," in this context, refers to greenhouse gas emissions and removals that are a direct result of human
activities or are the result of natural processes that have been affected by human activities (IPCC 2006).
3 Article 2 of the Framework Convention on Climate Change published by the UNEP/WMO Information Unit on Climate
Change. See . (UNEP/WMO 2000)
4 Article 4(l)(a) of the United Nations Framework Convention on Climate Change (also identified in Article 12). Subsequent
decisions by the Conference of the Parties elaborated the role of Annex I Parties in preparing national inventories. See
.
Introduction 1-1
-------
1 upon the previous bodies of work and include new sources and gases "... as well as updates to the previously
2 published methods whenever scientific and technical knowledge have improved since the previous guidelines were
3 issued. The UNFCCC adopted the 2006IPCC Guidelines as the standard methodological approach for Annex I
4 countries at the Nineteenth Conference of the Parties (Warsaw, November 11-23, 2013). This report presents
5 information in accordance with these guidelines.
6 Overall, this Inventory of anthropogenic greenhouse gas emissions and sinks provides a common and consistent
7 mechanism through which Parties to the UNFCCC can estimate emissions and compare the relative contribution of
8 individual sources, gases, and nations to climate change. The Inventory provides a national estimate of sources and
9 sinks for the United States, including all states and U.S. Territories.5 The structure of this report is consistent with
10 the current UNFCCC Guidelines on Annual Inventories (UNFCCC 2014).
11
12
Box 1-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Sinks
13 In following the UNFCCC requirement under Article 4.1 to develop and submit national greenhouse gas emission
14 inventories, the emissions and sinks presented in this report are organized by source and sink categories and
15 calculated using internationally-accepted methods provided by the IPCC.6 Additionally, the calculated emissions
16 and sinks in a given year for the United States are presented in a common manner in line with the UNFCCC
17 reporting guidelines for the reporting of inventories under this international agreement.7 The use of consistent
18 methods to calculate emissions and sinks by all nations providing their inventories to the UNFCCC ensures that
19 these reports are comparable. In this regard, U.S. emissions and sinks reported in this Inventory report are
20 comparable to emissions and sinks reported by other countries. The manner that emissions and sinks are provided in
21 this Inventory is one of many ways U.S. emissions and sinks could be examined; this Inventory report presents
22 emissions and sinks in a common format consistent with how countries are to report inventories under the
23 UNFCCC. Emissions and sinks provided in this inventory do not preclude alternative examinations, but rather this
24 inventory report presents emissions and sinks in a common format consistent with how countries are to report
25 inventories under the UNFCCC. The report itself follows this standardized format, and provides an explanation of
26 the IPCC methods used to calculate emissions and sinks, and the manner in which those calculations are conducted.
27 On October 30, 2009, the U.S. Environmental Protection Agency (EPA) published a rule for the mandatory
28 reporting of greenhouse gases from large greenhouse gas emissions sources in the United States. Implementation of
29 40 CFR Part 98 is referred to as the Greenhouse Gas Reporting Program (GHGRP). 40 CFR Part 98 applies to direct
30 greenhouse gas emitters, fossil fuel suppliers, industrial gas suppliers, and facilities that inject CO2 underground for
31 sequestration or other reasons.8 Reporting is at the facility level, except for certain suppliers of fossil fuels and
32 industrial greenhouse gases. The GHGRP dataset and the data presented in this Inventory report are complementary
33 and, as indicated in the respective planned improvements sections in this report's chapters, EPA is analyzing the
34 data for use, as applicable, to improve the national estimates presented in this Inventory.
35
36
37
1.1 Background Information
Science
38 For over the past 200 years, the burning of fossil fuels such as coal and oil, deforestation, land-use changes, and
39 other sources have caused the concentrations of heat-trapping "greenhouse gases" to increase significantly in our
U.S. Territories include American Samoa, Guam, Puerto Rico, U.S. Virgin Islands, Wake Island, and other U.S. Pacific Islands.
See .
7 See .
8 See and .
1-2 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 atmosphere (NOAA 2014). These gases in the atmosphere absorb some of the energy being radiated from the
2 surface of the Earth and then re-radiate this energy with some returning to the Earth's surface, essentially acting like
3 a blanket that makes the Earth's surface warmer than it would be otherwise.
4 Greenhouse gases are necessary to life as we know it. Without greenhouse gases in the atmosphere, the planet's
5 surface would be about 60°F cooler than present (EPA 2009). Carbon dioxide is also necessary for plant growth.
6 With emissions from biological and geological sources, there is a natural level of greenhouse gases that is
7 maintained in the atmosphere. But, as the concentrations of these gases continue to increase in from man-made
8 sources, the Earth's temperature is climbing above past levels. The Earth's averaged land and ocean surface
9 temperature has increased by about 1.2 to 1.9°F since 1880. The last three decades have each been the warmest
10 decade successively at the Earth's surface since 1850 (IPCC 2013). Most of the warming in recent decades is very
11 likely the result of human activities. Other aspects of the climate are also changing such as rainfall patterns, snow
12 and ice cover, and sea level.
13 If greenhouse gases continue to increase, climate models predict that the average temperature at the Earth's surface
14 is likely to increase from 0.5 to 8.6°F above 1986 through 2005 levels by the end of this century, depending on
15 future emissions (IPCC 2013). Scientists are certain that human activities are changing the composition of the
16 atmosphere, and that increasing the concentration of greenhouse gases will change the planet's climate. However,
17 they are not sure by how much it will change, at what rate it will change, or what the exact effects will be.9
18
Greenhouse Gases
19 Although the Earth's atmosphere consists mainly of oxygen and nitrogen, neither plays a significant role in
20 enhancing the greenhouse effect because both are essentially transparent to terrestrial radiation. The greenhouse
21 effect is primarily a function of the concentration of water vapor, carbon dioxide (CCh), methane (CH4), nitrous
22 oxide (N2O), and other trace gases in the atmosphere that absorb the terrestrial radiation leaving the surface of the
23 Earth (IPCC 2013). Changes in the atmospheric concentrations of these greenhouse gases can alter the balance of
24 energy transfers between the space and the earth system.10 A gauge of these changes is called radiative forcing,
25 which is a measure of the influence a perturbation has in altering the balance of incoming and outgoing energy in the
26 Earth-atmosphere system (IPCC 2013). Holding everything else constant, increases in greenhouse gas
27 concentrations in the atmosphere will produce positive radiative forcing (i.e., a net increase in the absorption of
28 energy by the Earth).
29 Human activities are continuing to affect the Earth's energy budget by changing the emissions and
30 resulting atmospheric concentrations ofradiatively important gases and aerosols and by changing land
31 surface properties (IPCC 2013).
32 Naturally occurring greenhouse gases include water vapor, CC>2, CH4, N2O, and ozone (Os). Several classes of
33 halogenated substances that contain fluorine, chlorine, or bromine are also greenhouse gases, but they are, for the
34 most part, solely a product of industrial activities. Chlorofluorocarbons (CFCs) and hydrochlorofluorocarbons
35 (HCFCs) are halocarbons that contain chlorine, while halocarbons that contain bromine are referred to as
36 bromofluorocarbons (i.e., halons). As stratospheric ozone depleting substances, CFCs, HCFCs, and halons are
37 covered under the Montreal Protocol on Substances that Deplete the Ozone Layer. The UNFCCC defers to this
38 earlier international treaty. Consequently, Parties to the UNFCCC are not required to include these gases in national
39 greenhouse gas inventories.11 Some other fluorine-containing halogenated substances—hydrofluorocarbons (HFCs),
40 perfluorocarbons (PFCs), sulfur hexafluoride (SF6), and nitrogen trifluoride (NF3)—do not deplete stratospheric
41 ozone but are potent greenhouse gases. These latter substances are addressed by the UNFCCC and accounted for in
42 national greenhouse gas inventories.
9 For more information see .
I" For more on the science of climate change, see NRC (2012).
Emissions estimates of CFCs, HCFCs, halons and other ozone-depleting substances are included in this document for
informational purposes.
Introduction 1-3
-------
1 There are also several other substances that influence the global radiation budget but are short-lived and therefore
2 not well-mixed. These substances include carbon monoxide (CO), nitrogen dioxide (NCh), sulfur dioxide (SCh), and
3 tropospheric (ground level) ozone (Os). Tropospheric ozone is formed by two precursor pollutants, volatile organic
4 compounds (VOCs) and nitrogen oxides (NOX) in the presence of ultraviolet light (sunlight).
5 Aerosols are extremely small particles or liquid droplets suspended in the Earth's atmosphere that are often
6 composed of sulfur compounds, carbonaceous combustion products (e.g., black carbon), crustal materials (e.g., dust)
7 and other human induced pollutants. They can affect the absorptive characteristics of the atmosphere (e.g.,
8 scattering incoming sunlight away from the Earth's surface, or, in the case of black carbon, absorb sunlight) and can
9 play a role in affecting cloud formation and lifetime affecting the radiative forcing of clouds and precipitation
10 patterns. Comparatively, however, while the understanding of aerosols has increased in recent years, they still
11 account for the largest contribution to uncertainty estimates in global energy budgets (IPCC 2013).
12 Carbon dioxide, methane, and nitrous oxide are continuously emitted to and removed from the atmosphere by
13 natural processes on Earth. Anthropogenic activities, however, can cause additional quantities of these and other
14 greenhouse gases to be emitted or sequestered, thereby changing their global average atmospheric concentrations.
15 Natural activities such as respiration by plants or animals and seasonal cycles of plant growth and decay are
16 examples of processes that only cycle carbon or nitrogen between the atmosphere and organic biomass. Such
17 processes, except when directly or indirectly perturbed out of equilibrium by anthropogenic activities, generally do
18 not alter average atmospheric greenhouse gas concentrations over decadal timeframes. Climatic changes resulting
19 from anthropogenic activities, however, could have positive or negative feedback effects on these natural systems.
20 Atmospheric concentrations of these gases, along with their rates of growth and atmospheric lifetimes, are presented
21 in Table 1-1.
22 Table 1-1: Global Atmospheric Concentration, Rate of Concentration Change, and
23 Atmospheric Lifetime of Selected Greenhouse Gases
Atmospheric Variable
Pre-industrial atmospheric
concentration
Atmospheric concentration
Rate of concentration change
Atmospheric lifetime (years)
CO2
280 ppm
401 ppms
2.3 ppm/yr
See footnote*
CH4
0.700 ppm
1.823ppma
0.005 ppm/yrb
12.4e
N2O
0.270 ppm
0.327ppma
0.26%/yr
121e
SF6
Oppt
7.9ppta
Linear0
3,200
CF4
40ppt
79 pptf
Linear0
50,000
Source: Pre-industrial atmospheric concentrations, atmospheric lifetime, and rate of concentration changes for CELi, N2O, SFe, and
CF4 are from IPCC (2013). The rate of concentration change for CCh is an average of the rates from 2011 through 2015 has
fluctuated between 1.7 to 3.0 ppm per year over this period (NOAA/ESRL 2016).
a The values presented are global annual average mole fractions (CDIAC 2015).
b The growth rate for atmospheric CtLt decreased from over 10 ppb/yr in the 1980s to nearly zero in the early 2000s; recently, the
growth rate has been about 5 ppb/year.
c IPCC (2007) identifies the rate of concentration change for SFe and CF4 as linear.
d For a given amount of carbon dioxide emitted, some fraction of the atmospheric increase in concentration is quickly absorbed by
the oceans and terrestrial vegetation, some fraction of the atmospheric increase will only slowly decrease over a number of years,
and a small portion of the increase will remain for many centuries or more.
e This lifetime has been defined as an "adjustment time" that takes into account the indirect effect of the gas on its own residence
time.
f The 2011 CF4 global mean atmospheric concentration is from the Advanced Global Atmospheric Gases Experiment (IPCC
2013).
g The atmospheric CCh concentration is the 2015 annual average at the Mauna Loa, HI station (NOAA/ESRL 2016).
24 A brief description of each greenhouse gas, its sources, and its role in the atmosphere is given below. The following
25 section then explains the concept of GWPs, which are assigned to individual gases as a measure of their relative
26 average global radiative forcing effect.
27 Water Vapor (H2O). Water vapor is the largest contributor to the natural greenhouse effect. Water vapor is
28 fundamentally different from other greenhouse gases in that it can condense and rain out when it reaches high
29 concentrations, and the total amount of water vapor in the atmosphere is in part a function of the Earth's
30 temperature. While some human activities such as evaporation from irrigated crops or power plant cooling release
31 water vapor into the air, this has been determined to have a negligible effect on climate (IPCC 2013). The lifetime of
32 water vapor in the troposphere is on the order of 10 days. Water vapor can also contribute to cloud formation, and
1-4 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 clouds can have both wanning and cooling effects by either trapping or reflecting heat. Because of the relationship
2 between water vapor levels and temperature, water vapor and clouds serve as a feedback to climate change, such
3 that for any given increase in other greenhouse gases, the total warming is greater than would happen in the absence
4 of water vapor. Aircraft contrails, which consist of water vapor and other substances, are aviation-induced clouds
5 with the same radiative forcing effects as high-altitude cirrus clouds (IPCC 1999).
6 Carbon Dioxide (CO2). In nature, carbon is cycled between various atmospheric, oceanic, land biotic, marine biotic,
7 and mineral reservoirs. The largest fluxes occur between the atmosphere and terrestrial biota, and between the
8 atmosphere and surface water of the oceans. In the atmosphere, carbon predominantly exists in its oxidized form as
9 CO2. Atmospheric CCh is part of this global carbon cycle, and therefore its fate is a complex function of
10 geochemical and biological processes. Carbon dioxide concentrations in the atmosphere increased from
11 approximately 280 parts per million by volume (ppmv) in pre-industrial times to 401 ppmvin2015, a 43 percent
12 increase (IPCC 2013 and NOAA/ESRL 2016).12-13 The IPCC definitively states that "the increase of CO2 ... is
13 caused by anthropogenic emissions from the use of fossil fuel as a source of energy and from land use and land use
14 changes, in particular agriculture" (IPCC 2013). The predominant source of anthropogenic CC>2 emissions is the
15 combustion of fossil fuels. Forest clearing, other biomass burning, and some non-energy production processes (e.g.,
16 cement production) also emit notable quantities of CO2. In its Fifth Assessment Report, the IPCC stated "it is
17 extremely likely that more than half of the observed increase in global average surface temperature from 1951 to
18 2010 was caused by the anthropogenic increase in greenhouse gas concentrations and other anthropogenic forcings
19 together," of which CO2is the most important (IPCC 2013).
20 Methane (CH4). Methane is primarily produced through anaerobic decomposition of organic matter in biological
21 systems. Agricultural processes such as wetland rice cultivation, enteric fermentation in animals, and the
22 decomposition of animal wastes emit CH4, as does the decomposition of municipal solid wastes. Methane is also
23 emitted during the production and distribution of natural gas and petroleum, and is released as a by-product of coal
24 mining and incomplete fossil fuel combustion. Atmospheric concentrations of CH4 have increased by about 152
25 percent since 1750, from a pre-industrial value of about 700 ppb to 1,762- 1,893 ppb in 2012,14 although the rate of
26 increase decreased to near zero in the early 2000s, and has recently increased again to about 5 ppb/year. The IPCC
27 has estimated that slightly more than half of the current CH4 flux to the atmosphere is anthropogenic, from human
28 activities such as agriculture, fossil fuel use, and waste disposal (IPCC 2007).
29 Methane is primarily removed from the atmosphere through a reaction with the hydroxyl radical (OH) and is
30 ultimately converted to CO2. Minor removal processes also include reaction with chlorine in the marine boundary
31 layer, a soil sink, and stratospheric reactions. Increasing emissions of CH4 reduce the concentration of OH, a
32 feedback that increases the atmospheric lifetime of CH4 (IPCC 2013). Methane's reactions in the atmosphere also
3 3 lead to production of tropospheric ozone and stratospheric water vapor, both of which also contribute to climate
34 change.
35 Nitrous Oxide (N2O). Anthropogenic sources of N2O emissions include agricultural soils, especially production of
36 nitrogen-fixing crops and forages, the use of synthetic and manure fertilizers, and manure deposition by livestock;
37 fossil fuel combustion, especially from mobile combustion; adipic (nylon) and nitric acid production; wastewater
38 treatment and waste incineration; and biomass burning. The atmospheric concentration of N2O has increased by 20
39 percent since 1750, from a pre-industrial value of about 270 ppb to 324-326 ppb in 2012,15 a concentration that has
40 not been exceeded during the last thousand years. Nitrous oxide is primarily removed from the atmosphere by the
41 photolytic action of sunlight in the stratosphere (IPCC 2007).
The pre-industrial period is considered as the time preceding the year 1750 (IPCC 2013).
13 Carbon dioxide concentrations during the last 1,000 years of the pre-industrial period (i.e., 750-1750), a time of relative
climate stability, fluctuated by about +10 ppmv around 280 ppmv (IPCC 2013).
14 The range is the annual arithmetic averages from a mid-latitude Northern-Hemisphere site and a mid-latitude Southern-
Hemisphere site for October 2012 through September 2013 (CDIAC 2014).
The range is the annual arithmetic averages from a mid-latitude Northern-Hemisphere site and a mid-latitude Southern-
Hemisphere site for October 2012 through September 2013 (CDIAC 2014).
Introduction 1-5
-------
1 Ozone (Os). Ozone is present in both the upper stratosphere,16 where it shields the Earth from harmful levels of
2 ultraviolet radiation, and at lower concentrations in the troposphere,17 where it is the main component of
3 anthropogenic photochemical "smog." During the last two decades, emissions of anthropogenic chlorine and
4 bromine-containing halocarbons, such as CFCs, have depleted stratospheric ozone concentrations. This loss of
5 ozone in the stratosphere has resulted in negative radiative forcing, representing an indirect effect of anthropogenic
6 emissions of chlorine and bromine compounds (IPCC 2013). The depletion of stratospheric ozone and its radiative
7 forcing was expected to reach a maximum in about 2000 before starting to recover.
8 The past increase in tropospheric ozone, which is also a greenhouse gas, is estimated to provide the fourth largest
9 increase in direct radiative forcing since the pre-industrial era, behind CC>2, black carbon, and CH4. Tropospheric
10 ozone is produced from complex chemical reactions of volatile organic compounds (including methane) mixing with
11 NOX in the presence of sunlight. The tropospheric concentrations of ozone and these other pollutants are short-lived
12 and, therefore, spatially variable (IPCC 2013).
13 Halocarbons, Perfluorocarbons, Sulfur Hexafluoride, and Nitrogen Triflouride. Halocarbons are, for the most part,
14 man-made chemicals that have both direct and indirect radiative forcing effects. Halocarbons that contain chlorine
15 (CFCs, HCFCs, methyl chloroform, and carbon tetrachloride) and bromine (halons, methyl bromide, and
16 hydrobromofluorocarbons HFCs) result in stratospheric ozone depletion and are therefore controlled under the
17 Montreal Protocol on Substances that Deplete the Ozone Layer. Although CFCs and HCFCs include potent global
18 warming gases, their net radiative forcing effect on the atmosphere is reduced because they cause stratospheric
19 ozone depletion, which itself is an important greenhouse gas in addition to shielding the Earth from harmful levels
20 of ultraviolet radiation. Under the Montreal Protocol, the United States phased out the production and importation
21 ofhalonsby 1994 and of CFCs by 1996. Under the Copenhagen Amendments to the Protocol, a cap was placed on
22 the production and importation of HCFCs by non-Article 5!8 countries beginning in 1996, and then followed by a
23 complete phase-out by the year 2030. While ozone depleting gases covered under the Montreal Protocol and its
24 Amendments are not covered by the UNFCCC, they are reported in this inventory under Annex 6.2 of this report for
25 informational purposes.
26 HFCs, PFCs, SF6, and NF3 are not ozone depleting substances, and therefore are not covered under the Montreal
27 Protocol. They are, however, powerful greenhouse gases. HFCs are primarily used as replacements for ozone
28 depleting substances but also emitted as a by-product of the HCFC-22 manufacturing process. Currently, they have
29 a small aggregate radiative forcing impact, but it is anticipated that their contribution to overall radiative forcing will
30 increase (IPCC 2013). PFCs, SF6, and NF3 are predominantly emitted from various industrial processes including
31 aluminum smelting, semiconductor manufacturing, electric power transmission and distribution, and magnesium
32 casting. Currently, the radiative forcing impact of PFCs, SF6, and NF3 is also small, but they have a significant
33 growth rate, extremely long atmospheric lifetimes, and are strong absorbers of infrared radiation, and therefore have
34 the potential to influence climate far into the future (IPCC 2013).
35 Carbon Monoxide. Carbon monoxide has an indirect radiative forcing effect by elevating concentrations of CH4 and
36 tropospheric ozone through chemical reactions with other atmospheric constituents (e.g., the hydroxyl radical, OH)
37 that would otherwise assist in destroying CH4 and tropospheric ozone. Carbon monoxide is created when carbon-
38 containing fuels are burned incompletely. Through natural processes in the atmosphere, it is eventually oxidized to
39 CO2. Carbon monoxide concentrations are both short-lived in the atmosphere and spatially variable.
16 The stratosphere is the layer from the troposphere up to roughly 50 kilometers. In the lower regions the temperature is nearly
constant but in the upper layer the temperature increases rapidly because of sunlight absorption by the ozone layer. The ozone-
layer is the part of the stratosphere from 19 kilometers up to 48 kilometers where the concentration of ozone reaches up to 10
parts per million.
The troposphere is the layer from the ground up to 11 kilometers near the poles and up to 16 kilometers in equatorial regions
(i.e., the lowest layer of the atmosphere where people live). It contains roughly 80 percent of the mass of all gases in the
atmosphere and is the site for most weather processes, including most of the water vapor and clouds.
Article 5 of the Montreal Protocol covers several groups of countries, especially developing countries, with low consumption
rates of ozone depleting substances. Developing countries with per capita consumption of less than 0.3 kg of certain ozone
depleting substances (weighted by their ozone depleting potential) receive financial assistance and a grace period often
additional years in the phase-out of ozone depleting substances.
1-6 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Nitrogen Oxides (NOX). The primary climate change effects of nitrogen oxides (i.e., NO and NO2) are indirect and
2 result from their role in promoting the formation of ozone in the troposphere, are a precursor to nitrate particles (i.e.,
3 aerosols) and, to a lesser degree, lower stratosphere, where they have positive radiative forcing effects.19
4 Additionally, NOX emissions are also likely to decrease CH4 concentrations, thus having a negative radiative forcing
5 effect (IPCC 2013). Nitrogen oxides are created from lightning, soil microbial activity, biomass burning (both
6 natural and anthropogenic fires) fuel combustion, and, in the stratosphere, from the photo-degradation of N2O.
7 Concentrations of NOX are both relatively short-lived in the atmosphere and spatially variable.
8 Nonmethane Volatile Organic Compounds (NMVOCs). Non-CH4 volatile organic compounds include substances
9 such as propane, butane, and ethane. These compounds participate, along with NOX, in the formation of
10 tropospheric ozone and other photochemical oxidants. NMVOCs are emitted primarily from transportation and
11 industrial processes, as well as biomass burning and non-industrial consumption of organic solvents. Concentrations
12 of NMVOCs tend to be both short-lived in the atmosphere and spatially variable.
13 Aerosols. Aerosols are extremely small particles or liquid droplets found in the atmosphere that are either directly
14 emitted into or are created through chemical reactions in the Earth's atmosphere. Aerosols or their chemical
15 precursors can be emitted by natural events such as dust storms and volcanic activity, or by anthropogenic processes
16 such as fuel combustion and biomass burning. Various categories of aerosols exist, including naturally produced
17 aerosols such as soil dust, sea salt, biogenic aerosols, sulfates, nitrates, and volcanic aerosols, and anthropogenically
18 manufactured aerosols such as industrial dust and carbonaceous20 aerosols (e.g., black carbon, organic carbon) from
19 transportation, coal combustion, cement manufacturing, waste incineration, and biomass burning. Aerosols can be
20 removed from the atmosphere relatively rapidly by precipitation or through more complex processes under dry
21 conditions.
22 Aerosols affect radiative forcing differently than greenhouse gases. Their radiative effects occur through direct and
23 indirect mechanisms: directly by scattering and absorbing solar radiation (and to a lesser extent scattering,
24 absorption, and emission of terrestrial radiation); and indirectly by increasing cloud droplets and ice crystals that
25 modify the formation, precipitation efficiency, and radiative properties of clouds (IPCC 2013). Despite advances in
26 understanding of cloud-aerosol interactions, the contribution of aerosols to radiative forcing are difficult to quantify
27 because aerosols generally have short atmospheric lifetimes, and have number concentrations, size distributions, and
28 compositions that vary regionally, spatially, and temporally (IPCC 2013).
29 The net effect of aerosols on the Earth's radiative forcing is believed to be negative (i.e., net cooling effect on the
30 climate). In fact, "despite the large uncertainty ranges on aerosol forcing, there is high confidence that aerosols have
31 offset a substantial portion of GHG forcing" (IPCC 2013).21 Although because they remain in the atmosphere for
32 only days to weeks, their concentrations respond rapidly to changes in emissions.22 Not all aerosols have a cooling
33 effect. Current research suggests that another constituent of aerosols, black carbon, has a positive radiative forcing
34 by heating the Earth's atmosphere and causing surface warming when deposited on ice and snow (IPCC 2013).
3 5 Black carbon also influences cloud development, but the direction and magnitude of this forcing is an area of active
36 research.
37 Global Warming Potentials
38 A global warming potential is a quantified measure of the globally averaged relative radiative forcing impacts of a
39 particular greenhouse gas (see Table 1-2). It is defined as the ratio of the time-integrated radiative forcing from the
40 instantaneous release of 1 kilogram (kg) of a trace substance relative to that of 1 kg of a reference gas (IPCC 2007).
41 Direct radiative effects occur when the gas itself absorbs radiation. Indirect radiative forcing occurs when chemical
19 NOX emissions injected higher in the stratosphere, primarily from fuel combustion emissions from high altitude supersonic
aircraft, can lead to stratospheric ozone depletion.
20 Carbonaceous aerosols are aerosols that are comprised mainly of organic substances and forms of black carbon (or soot)
(IPCC 2013).
21 The IPCC (2013) defines high confidence as an indication of strong scientific evidence and agreement in this statement.
Volcanic activity can inject significant quantities of aerosol producing sulfur dioxide and other sulfur compounds into the
stratosphere, which can result in a longer negative forcing effect (i.e., a few years) (IPCC 2013).
Introduction 1-7
-------
1 transformations involving the original gas produce a gas or gases that are greenhouse gases, or when a gas
2 influences other radiatively important processes such as the atmospheric lifetimes of other gases. The reference gas
3 used is COa, and therefore GWP-weighted emissions are measured in million metric tons of CCh equivalent (MMT
4 CO2 Eq.).23 The relationship between kilotons (kt) of a gas and MMT COa Eq. can be expressed as follows:
/ MMT \
5 MMT C02 Eq. = (kt of gas) X (GWP) X (^ QQQ fej
6 where,
7 MMT CO2 Eq. = Million metric tons of CO2 equivalent
8 kt = Kilotons (equivalent to a thousand metric tons)
9 GWP = Global warming potential
10 MMT = Million metric tons
11 GWP values allow for a comparison of the impacts of emissions and reductions of different gases. According to the
12 IPCC, GWPs typically have an uncertainty of ±35 percent. Parties to the UNFCCC have also agreed to use GWPs
13 based upon a 100-year time horizon, although other time horizon values are available.
14 ... the global warming potential values used by Parties included in Annex I to the Convention (Annex I
15 Parties) to calculate the carbon dioxide equivalence of anthropogenic emissions by sources and removals
16 by sinks of greenhouse gases shall be those listed in the column entitled "Global warming potential for
17 given time horizon " in table 2.14 of the errata to the contribution of Working Group I to the Fourth
18 Assessment Report of the Intergovernmental Panel on Climate Change, based on the effects of greenhouse
19 gases over a 100-year time horizon...24
20 Greenhouse gases with relatively long atmospheric lifetimes (e.g., CC>2, CH4, N2O, HFCs, PFCs, SF6, NF3) tend to
21 be evenly distributed throughout the atmosphere, and consequently global average concentrations can be
22 determined. The short-lived gases such as water vapor, carbon monoxide, tropospheric ozone, ozone precursors
23 (e.g., NOX, and NMVOCs), and tropospheric aerosols (e.g., SCh products and carbonaceous particles), however,
24 vary regionally, and consequently it is difficult to quantify their global radiative forcing impacts. Parties to the
25 UNFCCC have not agreed upon GWP values for these gases that are short-lived and spatially inhomogeneous in the
26 atmosphere.
27 Table 1-2: Global Warming Potentials and Atmospheric Lifetimes (Years) Used in this Report
Gas
C02
CH4a
N20
HFC-23
HFC-32
HFC-125
HFC-134a
HFC-143a
HFC-152a
HFC-227ea
HFC-236fa
HFC-4310mee
Atmospheric Lifetime
b
12
114
270
4.9
29
14
52
1.4
34.2
240
15.9
GWPC
1
25
298
14,800
675
3,500
1,430
4,470
124
3,220
9,810
1,640
Carbon comprises 12/44ths of carbon dioxide by weight.
Framework Convention on Climate Change; < http://unfccc.int/resource/docs/2013/copl9/eng/10a03.pdf>; 31 January 2014;
Report of the Conference of the Parties at its nineteenth session; held in Warsaw from 11 to 23 November 2013; Addendum; Part
two: Action taken by the Conference of the Parties at its nineteenth session; Decision 24/CP.19; Revision of the UNFCCC
reporting guidelines on annual inventories for Parties included in Annex I to the Convention; p. 2. (UNFCCC 2014)
1-8 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
CF4
C2F6
C4Fio
C6Fi4
SF6
NF3
50,000
10,000
2,600
3,200
3,200
740
7,390
12,200
8,860
9,300
22,800
17,200
Source: (IPCC 2007)
a The GWP of CH4 includes the direct effects and those indirect effects
due to the production of tropospheric ozone and stratospheric water
vapor. The indirect effect due to the production of CO2 is not included.
b For a given amount of carbon dioxide emitted, some fraction of the
atmospheric increase in concentration is quickly absorbed by the oceans
and terrestrial vegetation, some fraction of the atmospheric increase will
only slowly decrease over a number of years, and a small portion of the
increase will remain for many centuries or more.
0100-year time horizon.
Box 1-2: The IPCC Fifth Assessment Report and Global Warming Potentials
3 In 2014, the IPCC published its Fifth Assessment Report (AR5), which provided an updated and more
4 comprehensive scientific assessment of climate change. Within the AR5 report, the GWP values of several gases
5 were revised relative to previous IPCC reports, namely the IPCC Second Assessment Report (SAR) (IPCC 1996),
6 the IPCC Third Assessment Report (TAR) (IPCC 2001), and the IPCC Fourth Assessment Report (AR4) (IPCC
7 2007). Although the AR4 GWP values are used throughout this report, consistent with UNFCCC reporting
8 requirements, it is interesting to review the changes to the GWP values and the impact improved understanding has
9 on the total GWP-weighted emissions of the United States. In the AR5, the IPCC has applied an improved
10 calculation of CCh radiative forcing and an improved CCh response function in presenting updated GWP values.
11 Additionally, the atmospheric lifetimes of some gases have been recalculated, and updated background
12 concentrations were used. In addition, the values for radiative forcing and lifetimes have been recalculated for a
13 variety of halocarbons, and the indirect effects of methane on ozone have been adjusted to match more recent
14 science. Table 1-3 presents the new GWP values, relative to those presented in the AR4 and using the 100-year time
15 horizon common to UNFCCC reporting.
16 Table 1-3: Comparison of 100-Year GWP values
Gas
C02
CH4a
N20
HFC-23
HFC-32
HFC-125
HFC-134a
HFC-143a
HFC-152a
HFC-227ea
HFC-236fa
HFC-4310mee
CF4
C2F6
C4Fio
C6Fi4
SF6
NF3
SAR
1
21
310
11,700
650
2,800
1,300
3,800
140
2,900
6,300
1,300
6,500
9,200
7,000
7,400
23,900
NA
TAR
1
23
296
12,000
550
3,400
1,300
4,300
120
3,500
9,400
1,500
5,700
11,900
8,600
9,000
22,200
10,800
AR4
1
25
298
14,800
675
3,500
1,430
4,470
124
3,220
9,810
1,640
7,390
12,200
8,860
9,300
22,800
17,200
AR5b
1
28
265
12,400
677
3,170
1,300
4,800
138
3,350
8,060
1,650
6,630
11,100
9,200
7,910
23,500
16,100
Comparison to AR4
SAR
NC
(4)
12
(3,100)
(25)
(700)
(130)
(670)
16
(320)
(3,510)
(340)
(890)
(3,000)
(1,860)
(1,900)
1,100
NA
TAR
NC
(2)
(2)
(2,800)
(125)
(100)
(130)
(170)
(4)
280
(410)
(140)
(1,690)
(300)
(260)
(300)
(600)
(6,400)
AR5
NC
3
(33)
(2,400)
2
(330)
(130)
330
14
130
(1,750)
10
(760)
(1,100)
340
(1,390)
700
700
Introduction 1-9
-------
Source: (IPCC 2013, IPCC 2007, IPCC 2001, IPCC 1996)
NC (No Change)
NA (Not Applicable)
Note: Parentheses indicate negative values.
a The GWP of CH4 includes the direct effects and those indirect effects due to the production
of tropospheric ozone and stratospheric water vapor. The indirect effect due to the
production of CCh is not included.
b The GWPs presented here are the ones most consistent with the methodology used in the
AR4 report. The AR5 report has also calculated GWPs (not shown here) where climate-
carbon feedbacks have been included for the non-CCh gases in order to be consistent with the
approach used in calculating the CCh lifetime. Additionally, the AR5 reported separate values
for fossil versus biogenic methane in order to account for the CCh oxidation product.
1 To comply with international reporting standards under the UNFCCC, official emission estimates are reported by
2 the United States using AR4 GWP values, as required by the 2013 revision to the UNFCCC reporting guidelines for
3 national inventories.25 All estimates provided throughout this report are also presented in unweighted units. For
4 informational purposes, emission estimates that use GWPs from other IPCC Assessment Reports are presented in
5 detail in Annex 6.1 of this report.
7 1.2 National Inventory Arrangements
8 The U.S. Environmental Protection Agency (EPA), in cooperation with other U.S. government agencies, prepares
9 the Inventory of U.S. Greenhouse Gas Emissions and Sinks. A wide range of agencies and individuals are involved
10 in supplying data to, planning methodological approaches and improvements, reviewing, or preparing portions of the
11 U.S. Inventory—including federal and state government authorities, research and academic institutions, industry
12 associations, and private consultants.
13 Within EPA, the Office of Atmospheric Programs (OAP) is the lead office responsible for the emission calculations
14 provided in the Inventory, as well as the completion of the National Inventory Report and the Common Reporting
15 Format tables. EPA's Office of Transportation and Air Quality (OTAQ) is also involved in calculating emissions
16 for the Inventory. While the U.S. Department of State officially submits the annual Inventory to the UNFCCC,
17 EPA's OAP serves as the Inventory focal point for technical questions and comments on the U.S. Inventory. The
18 staff of OAP and OTAQ coordinates the annual methodological choice, activity data collection, and emission
19 calculations at the individual source category level. Within OAP, an inventory coordinator compiles the entire
20 Inventory into the proper reporting format for submission to the UNFCCC, and is responsible for the collection and
21 consistency of cross-cutting issues in the Inventory.
22 Several other government agencies contribute to the collection and analysis of the underlying activity data used in
23 the Inventory calculations. Formal relationships exist between EPA and other U.S. agencies that provide official
24 data for use in the Inventory. The U.S. Department of Energy's Energy Information Administration provides
25 national fuel consumption data and the U.S. Department of Defense provides military fuel consumption and bunker
26 fuels. Informal relationships also exist with other U.S. agencies to provide activity data for use in EPA's emission
27 calculations. These include: the U.S. Department of Agriculture, the U.S. Geological Survey, the Federal Highway
28 Administration, the Department of Transportation, the Bureau of Transportation Statistics, the Department of
29 Commerce, the National Agricultural Statistics Service, and the Federal Aviation Administration. Academic and
30 research centers also provide activity data and calculations to EPA, as well as individual companies participating in
31 voluntary outreach efforts with EPA. Finally, the U.S. Department of State officially submits the Inventory to the
32 UNFCCC each April. Figure 1-1 diagrams the National Inventory Arrangements.
25 See < http://unfccc.int/resource/docs/2013/copl9/eng/10a03.pdf>.
1-10 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1
2 Figure 1-1: National Inventory Arrangements Diagram
Inited States
Inventory Compilation
Emission Calculations
Data Collection
Energy
• Bureau of Oce
Inventory Submission
United Nations
Framework Convention on
Climate Change
U.S. Department of State
\ Energy Management
1 Federal Highway Administration
EPAGHGRP
U.S. Department of Defense
U.S. Department of Energy
U.S. Environmental
Protection Agency
Inventory Compiler
U.S. Environmental
Protection Agency
Other U.S.
Government
Agriculture and LULUCF
• Colorado State University
Industrial Processes and Product Use
• Alliance for Responsible Atmospheric Policy
• American Iron and Steel Institute
1 U.S. Aluminum Associa
• U.S. Bureau of Mines
nergency Response
arious publications and e
Introduction 1-11
-------
1.3 Inventory Process
2 EPA has a decentralized approach to preparing the annual U.S. Inventory, which consists of a National Inventory
3 Report (NIR) and Common Reporting Format (CRF) tables. The inventory coordinator at EPA is responsible for
4 compiling all emission estimates and ensuring consistency and quality throughout the NIR and CRF tables.
5 Emission calculations for individual sources are the responsibility of individual source leads, who are most familiar
6 with each source category and the unique characteristics of its emissions profile. The individual source leads
7 determine the most appropriate methodology and collect the best activity data to use in the emission calculations,
8 based upon their expertise in the source category, as well as coordinating with researchers and contractors familiar
9 with the sources. A multi-stage process for collecting information from the individual source leads and producing
10 the Inventory is undertaken annually to compile all information and data.
11 Methodology Development, Data Collection, and Emissions
12 and Sink Estimation
13 Source leads at EPA collect input data and, as necessary, evaluate or develop the estimation methodology for the
14 individual source categories. For most source categories, the methodology for the previous year is applied to the
15 new "current" year of the Inventory, and inventory analysts collect any new data or update data that have changed
16 from the previous year. If estimates for a new source category are being developed for the first time, or if the
17 methodology is changing for an existing source category (e.g., the United States is implementing a higher Tiered
18 approach for that source category), then the source category lead will develop a new methodology, gather the most
19 appropriate activity data and emission factors (or in some cases direct emission measurements) for the entire time
20 series, and conduct a special source-specific peer review process involving relevant experts from industry,
21 government, and universities.
22 Once the methodology is in place and the data are collected, the individual source leads calculate emissions and sink
23 estimates. The source leads then update or create the relevant text and accompanying annexes for the Inventory.
24 Source leads are also responsible for completing the relevant sectoral background tables of the Common Reporting
25 Format, conducting quality assurance and quality control (QA/QC) checks, and uncertainty analyses.
26 Summary Spreadsheet Compilation and Data Storage
27 The inventory coordinator at EPA collects the source categories' descriptive text and Annexes, and also aggregates
28 the emission estimates into a summary spreadsheet that links the individual source category spreadsheets together.
29 This summary sheet contains all of the essential data in one central location, in formats commonly used in the
30 Inventory document. In addition to the data from each source category, national trend and related data are also
31 gathered in the summary sheet for use in the Executive Summary, Introduction, and Recent Trends sections of the
32 Inventory report. Electronic copies of each year's summary spreadsheet, which contains all the emission and sink
3 3 estimates for the United States, are kept on a central server at EPA under the jurisdiction of the inventory
34 coordinator.
35 National Inventory Report Preparation
36 The NIR is compiled from the sections developed by each individual source lead. In addition, the inventory
37 coordinator prepares a brief overview of each chapter that summarizes the emissions from all sources discussed in
38 the chapters. The inventory coordinator then carries out a key category analysis for the Inventory, consistent with
39 the 2006IPCC Guidelines for National Greenhouse Gas Inventories, and in accordance with the reporting
40 requirements of the UNFCCC. Also at this time, the Introduction, Executive Summary, and Recent Trends sections
41 are drafted, to reflect the trends for the most recent year of the current Inventory. The analysis of trends necessitates
42 gathering supplemental data, including weather and temperature conditions, economic activity and gross domestic
43 product, population, atmospheric conditions, and the annual consumption of electricity, energy, and fossil fuels.
Introduction 1-13
-------
1 Changes in these data are used to explain the trends observed in greenhouse gas emissions in the United States.
2 Furthermore, specific factors that affect individual sectors are researched and discussed. Many of the factors that
3 affect emissions are included in the Inventory document as separate analyses or side discussions in boxes within the
4 text. Text boxes are also created to examine the data aggregated in different ways than in the remainder of the
5 document, such as a focus on transportation activities or emissions from electricity generation. The document is
6 prepared to match the specification of the UNFCCC reporting guidelines for National Inventory Reports.
7 Common Reporting Format Table Compilation
8 The CRF tables are compiled from individual tables completed by each individual source lead, which contain source
9 emissions and activity data. The inventory coordinator integrates the source data into the UNFCCC's "CRF
10 Reporter" for the United States, assuring consistency across all sectoral tables. The summary reports for emissions,
11 methods, and emission factors used, the overview tables for completeness and quality of estimates, the recalculation
12 tables, the notation key completion tables, and the emission trends tables are then completed by the inventory
13 coordinator. Internal automated quality checks on the CRF Reporter, as well as reviews by the source leads, are
14 completed for the entire time series of CRF tables before submission.
15
28
33
QA/QC and Uncertainty
16 QA/QC and uncertainty analyses are supervised by the QA/QC and Uncertainty coordinators, who have general
17 oversight over the implementation of the QA/QC plan and the overall uncertainty analysis for the Inventory (see
18 sections on QA/QC and Uncertainty, below). These coordinators work closely with the source leads to ensure that a
19 consistent QA/QC plan and uncertainty analysis is implemented across all inventory sources. The inventory QA/QC
20 plan, detailed in a following section, is consistent with the quality assurance procedures outlined by EPA and IPCC.
21 Expert and Public Review Periods
22 During the Expert Review period, a first draft of the document is sent to a select list of technical experts outside of
23 EPA. The purpose of the Expert Review is to encourage feedback on the methodological and data sources used in
24 the current Inventory, especially for sources which have experienced any changes since the previous Inventory.
25 Once comments are received and addressed, a second draft of the document is released for public review by
26 publishing a notice in the U.S. Federal Register and posting the document on the EPA Web site. The Public Review
27 period allows for a 30 day comment period and is open to the entire U.S. public.
Final Submittal to UNFCCC and Document Printing
29 After the final revisions to incorporate any comments from the Expert Review and Public Review periods, EPA
30 prepares the final National Inventory Report and the accompanying Common Reporting Format Reporter database.
31 The U.S. Department of State sends the official submission of the U.S. Inventory to the UNFCCC. The document is
32 then formatted and posted online, available for the public.l
1.4 Methodology and Data Sources
34 Emissions of greenhouse gases from various source and sink categories have been estimated using methodologies
35 that are consistent with the 2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2006). To the
36 extent possible, the present report relies on published activity and emission factor data. Depending on the emission
1 See
1-14 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 source category, activity data can include fuel consumption or deliveries, vehicle-miles traveled, raw material
2 processed, etc. Emission factors are factors that relate quantities of emissions to an activity.
3 The IPCC methodologies provided in the 2006IPCC Guidelines represent baseline methodologies for a variety of
4 source categories, and many of these methodologies continue to be improved and refined as new research and data
5 become available. This report uses the IPCC methodologies when applicable, and supplements them with other
6 available country-specific methodologies and data where possible. Choices made regarding the methodologies and
7 data sources used are provided in conjunction with the discussion of each source category in the main body of the
8 report. Complete documentation is provided in the annexes on the detailed methodologies and data sources utilized
9 in the calculation of each source category.
10
11
Box 1-3: IPCC Reference Approach
12 The UNFCCC reporting guidelines require countries to complete a "top-down" reference approach for estimating
13 CO2 emissions from fossil fuel combustion in addition to their "bottom-up" sectoral methodology. This estimation
14 method uses alternative methodologies and different data sources than those contained in that section of the Energy
15 chapter. The reference approach estimates fossil fuel consumption by adjusting national aggregate fuel production
16 data for imports, exports, and stock changes rather than relying on end-user consumption surveys (see Annex 4 of
17 this report). The reference approach assumes that once carbon-based fuels are brought into a national economy, they
18 are either saved in some way (e.g., stored in products, kept in fuel stocks, or left unoxidized in ash) or combusted,
19 and therefore the carbon in them is oxidized and released into the atmosphere. Accounting for actual consumption
20 of fuels at the sectoral or sub-national level is not required.
21
22
1.5 Key Categories
23 The 2006IPCC Guidelines (IPCC 2006) defines a key category as a "[category] that is prioritized within the
24 national inventory system because its estimate has a significant influence on a country's total inventory of
25 greenhouse gases in terms of the absolute level, the trend, or the uncertainty in emissions and removals."2 By
26 definition, key categories include those categories that have the greatest contribution to the absolute level of national
27 emissions. In addition, when an entire time series of emission and removal estimates is prepared, a thorough
28 investigation of key categories must also account for the influence of trends and uncertainties of individual source
29 and sink categories. This analysis culls out source and sink categories that diverge from the overall trend in national
30 emissions. Finally, a qualitative evaluation of key categories is performed to capture any categories that were not
31 identified in any of the quantitative analyses.
32 Approach 1, as defined in the 2006 IPCC Guidelines (IPCC 2006), was implemented to identify the key categories
33 for the United States. This analysis was performed twice; one analysis included sources and sinks from the Land
34 Use, Land-Use Change, and Forestry (LULUCF) sector, the other analysis did not include the LULUCF categories.
35 Following Approach 1, Approach 2, as defined in the 2006 IPCC Guidelines (IPCC 2006), was then implemented to
36 identify any additional key categories not already identified in Approach 1 assessment. This analysis, which includes
37 each source category's uncertainty assessments (or proxies) in its calculations, was also performed twice to include
38 or exclude LULUCF categories.
39 In addition to conducting Approach 1 and 2 level and trend assessments, a qualitative assessment of the source
40 categories, as described in the 2006 IPCC Guidelines (IPCC 2006), was conducted to capture any key categories that
41 were not identified by either quantitative method. One additional key category, international bunker fuels, was
42 identified using this qualitative assessment. International bunker fuels are fuels consumed for aviation or marine
43 international transport activities, and emissions from these fuels are reported separately from totals in accordance
See Chapter 4 "Methodological Choice and Identification of Key Categories" in IPCC (2006). See < http://www.ipcc-
nggip.iges.or.jp/public/2006gl/index.html>.
Introduction 1-15
-------
1 with IPCC guidelines. If these emissions were included in the totals, bunker fuels would qualify as a key category
2 according to Approach 1. The amount of uncertainty associated with estimation of emissions from international
3 bunker fuels also supports the qualification of this source category as key, because it would qualify bunker fuels as a
4 key category according to Approach 2. Table 1-4 presents the key categories for the United States (including and
5 excluding LULUCF categories) using emissions and uncertainty data in this report, and ranked according to their
6 sector and global warming potential-weighted emissions in 2014. The table also indicates the criteria used in
7 identifying these categories (i.e., level, trend, Approach 1, Approach 2, and/or qualitative assessments). Annex 1 of
8 this report provides additional information regarding the key categories in the United States and the methodologies
9 used to identify them.
10 Table 1-4: Key Categories for the United States (1990-2014)
IPCC Source
Categories
Gas
Approach 1
Level Trend Level Trend
Without Without With With
LULUCF LULUCF LULUCF LULUCF
Approach 2
Level Trend Level Trend
Without Without With With
LULUCF LULUCF LULUCF LULUCF
Quala
2014
Emissions
(MMT
CO2 Eq.)
Energy
CO2 Emissions from
Stationary
Combustion - Coal -
Electricity Generation
CO2 Emissions from
Mobile Combustion:
Road
CO2 Emissions from
Stationary
Combustion - Gas -
Industrial
CO2 Emissions from
Stationary
Combustion - Gas -
Electricity Generation
CO2 Emissions from
Stationary
Combustion - Gas -
Residential
CO2 Emissions from
Stationary
Combustion - Oil -
Industrial
CO2 Emissions from
Stationary
Combustion - Gas -
Commercial
CO2 Emissions from
Mobile Combustion:
Aviation
CO2 Emissions from
Non-Energy Use of
Fuels
CO2 Emissions from
Mobile Combustion:
Other
CO2 Emissions from
Stationary
Combustion - Coal -
Industrial
C02
CO2
C02
C02
CO2
CO2
C02
C02
C02
CO2
C02
272.9
189.2
150.1
114.3
1-16 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
CO2 Emissions from
Stationary
Combustion - Oil -
Residential
CO2 Emissions from
Stationary
Combustion - Oil -
Commercial
CO2 Emissions from
Natural Gas Systems
CO2 Emissions from
Stationary
Combustion - Oil -
U.S. Territories
CO2 Emissions from
Mobile Combustion:
Marine
CO2 Emissions from
Stationary
Combustion - Oil -
Electricity Generation
CO2 Emissions from
Stationary
Combustion - Coal -
Commercial
CO2 Emissions from
Stationary
Combustion - Gas -
U.S. Territories
CO2 Emissions from
Stationary
Combustion - Coal -
Residential
CH4 Emission
Natural Gas Systems
Fugitive Emissions
from Coal Mining
CH4 Emissions from
Petroleum Systems
Non-CO2 Emissions
from Stationary
Combustion -
Residential
Non-CO2 Emissions
from Stationary
Combustion -
Electricity Generation
N2O Emissions from
Mobile Combustion:
Road
Non-CO2 Emissions
from Stationary
Combustion -
Industrial
International Bunker
Fuelsb
C02
C02
C02
CO2
C02
CO2
CO2
C02
C02
CH4
CH4
CH4
CH4
N20
N2O
N20
Several
67.6
27.7
25.3
4.5
157.4
64.6
25.2
5.0
19.6
12.5
104.2
Industrial Processes and Product Use
CO2 Emissions from
Iron and Steel
Production &
Metallurgical Coke
Production
C02
55.4
Introduction 1-17
-------
CO2 Emissions from
Cement Production
CO2 Emissions from
Petrochemical
Production
CO2 Emissions from
Other Process Uses of
Carbonates
N2O Emissions from
Adipic Acid
Production
Emissions from
Substitutes for Ozone
Depleting Substances
SFe Emissions from
Electrical
Transmission and
Distribution
HFC-23 Emissions
from HCFC-22
Production
PFC Emissions from
Aluminum Production
HiGWP
HiGWP
HiGWP
HiGWP
Agriculture
CH4 Emissions from
Enteric Fermentation
CH4 Emissions from
Manure Management
CH4 Emissions from
Rice Cultivation
Direct N2O Emissions
from Agricultural Soil
Management
Indirect N2O
Emissions from
Applied Nitrogen
CH4
CH4
CH4
N2O
N20
164.3
61.2
12.2
261.3
57.2
Waste
CH4 Emissions from
Landfills
CH4
• • • •
• • • •
181.8
Land Use, Land-Use Change, and Forestry
CO2 Emissions from
Land Converted to
Cropland
CO2 Emissions from
Grassland Remaining
Grassland
CO2 Emissions from
Landfilled Yard
Trimmings and Food
Scraps
CO2 Emissions from
Cropland Remaining
Cropland
CO2 Emissions from
Urban Trees
CO2 Emissions from
Forest Land
Remaining Forest
Land
CH4 Emissions from
Forest Fires
CO2
C02
CO2
C02
C02
CO2
CH4
14.7
11.9
7.3
1-18 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
N2O Emissions from
Settlement Soils
N2O Emissions from
Forest Fires
N2O
N2O
Subtotal Without LULUCF
Total Emissions Without LULUCF
6,676.8
6,848.1
Percent of Total Without LULUCF
97%
Subtotal With LULUCF
Total Emissions With LULUCF
6,037.6
6,186.8
Percent of Total With LULUCF
98%
a Qualitative criteria.
b Emissions from this source not included in totals.
Note: Parentheses indicate negative values (or sequestration).
i 1.6 Quality Assurance and Quality Control
2 (QA/QC)
3 As part of efforts to achieve its stated goals for inventory quality, transparency, and credibility, the United States has
4 developed a quality assurance and quality control plan designed to check, document and improve the quality of its
5 inventory over time. QA/QC activities on the Inventory are undertaken within the framework of the U.S. QA/QC
6 plan, Quality Assurance/Quality Control and Uncertainty Management Plan for the U.S. Greenhouse Gas Inventory:
7 Procedures Manual for QA/QC and Uncertainty Analysis.
8 Key attributes of the QA/QC plan are summarized in Figure 1-2. These attributes include:
9 • Procedures and Forms: detailed and specific systems that serve to standardize the process of documenting
10 and archiving information, as well as to guide the implementation of QA/QC and the analysis of
11 uncertainty
12 • Implementation of Procedures: application of QA/QC procedures throughout the whole inventory
13 development process from initial data collection, through preparation of the emission estimates, to
14 publication of the Inventory
15 • Quality Assurance: expert and public reviews for both the inventory estimates and the Inventory report
16 (which is the primary vehicle for disseminating the results of the inventory development process)
17 • Quality Control: consideration of secondary data and source-specific checks (Tier 2 QC) in parallel and
18 coordination with the uncertainty assessment; the development of protocols and templates, which provides
19 for more structured communication and integration with the suppliers of secondary information
20 • Tier 1 (general) and Tier 2 (source-specific) Checks: quality controls and checks, as recommended by
21 IPCC Good Practice Guidance
22 • Record Keeping: provisions to track which procedures have been followed, the results of the QA/QC,
23 uncertainty analysis, and feedback mechanisms for corrective action based on the results of the
24 investigations which provide for continual data quality improvement and guided research efforts
25 • Multi-Year Implementation', a schedule for coordinating the application of QA/QC procedures across
26 multiple years
27 • Interaction and Coordination: promoting communication within the EPA, across Federal agencies and
28 departments, state government programs, and research institutions and consulting firms involved in
29 supplying data or preparing estimates for the Inventory. The QA/QC Management Plan itself is intended to
Introduction 1-19
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
be revised and reflect new information that becomes available as the program develops, methods are
improved, or additional supporting documents become necessary.
In addition, based on the national QA/QC plan for the Inventory, source-specific QA/QC plans have been developed
for a number of sources. These plans follow the procedures outlined in the national QA/QC plan, tailoring the
procedures to the specific text and spreadsheets of the individual sources. For each greenhouse gas emissions source
or sink included in this Inventory, a minimum of a Tier 1 QA/QC analysis has been undertaken. Where QA/QC
activities for a particular source go beyond the minimum Tier 1 level, further explanation is provided within the
respective source category text.
The quality control activities described in the U.S. QA/QC plan occur throughout the inventory process; QA/QC is
not separate from, but is an integral part of, preparing the Inventory. Quality control—in the form of both good
practices (such as documentation procedures) and checks on whether good practices and procedures are being
followed—is applied at every stage of inventory development and document preparation. In addition, quality
assurance occurs at two stages—an expert review and a public review. While both phases can significantly
contribute to inventory quality, the public review phase is also essential for promoting the openness of the inventory
development process and the transparency of the inventory data and methods.
The QA/QC plan guides the process of ensuring inventory quality by describing data and methodology checks,
developing processes governing peer review and public comments, and developing guidance on conducting an
analysis of the uncertainty surrounding the emission estimates. The QA/QC procedures also include feedback loops
and provide for corrective actions that are designed to improve the inventory estimates over time.
Figure 1-2: U.S. QA/QC Plan Summary
4-1
Ln
—
ru
C
<
>-
•M
c
(1)
>
_c
-M
I/)
~ro
c
<
U
a
§
1 Obtain data in electronic
format (if possible)
• Review spreadsheet
construction
Avoid hardwiring
Use data validation
Protect cells
1 Develop automatic
checkers for:
• Outliers, negative
values, or missing
data
Variabletypes
match values
• Time series
consistency
• Maintain trackingtab for
status of gathering
efforts
• Check input datafor
transcription errors
• Inspect automatic
checkers
• Identify spreadsheet
modifi cat ionsthat could
provide additional
QA/QC checks
•
1 Contact reports for non-
electronic communications
• Provide cell references for
primary data elements
• Obtain copies of all data
sources
• Listand location of any
working/external
spreadsheets
• Document assumptions
_
• Check citations in
spreadsheet andtextfor
accuracy and style
• Check reference docketfor
new citations
• Review documentation for
any data/ methodology
changes
•
• Clearly label parameters.
units, and conversion
factors
• Review spreadsheet
integrity
• Equations
• Units
• Inputs and output
• Develop automated
checkersfor:
• Input ranges
- Calculations
• Emission aggregation
_
• Reproduce calculations
• Review time series
consistency
• Review changes in
data/consistency with IPCC
methodology
• Common starting
versionsforeach
inventory year
• Utilize unalterable
•L summary tab foreach
^q source spreadsheetfor
linkingto a master
summary spreadsheet
• Follow strictversion
control procedures
• Document QA/QC
procedures
Data Gathering Data Documentation CalculatingEmissions Cross-Cutting
Coordination
21
1-20 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
i 1.7 Uncertainty Analysis of Emission Estimates
2 Uncertainty estimates are an essential element of a complete and transparent emissions inventory. Uncertainty
3 information is not intended to dispute the validity of the inventory estimates, but to help prioritize efforts to improve
4 the accuracy of future inventories and guide future decisions on methodological choice. While the U.S. Inventory
5 calculates its emission estimates with the highest possible accuracy, uncertainties are associated to a varying degree
6 with the development of emission estimates for any inventory. Some of the current estimates, such as those for €62
7 emissions from energy-related activities, are considered to have minimal uncertainty associated with them. For
8 some other categories of emissions, however, a lack of data or an incomplete understanding of how emissions are
9 generated increases the uncertainty surrounding the estimates presented. The UNFCCC reporting guidelines follow
10 the recommendation in the 2006IPCC Guidelines (IPCC 2006) and require that countries provide single point
11 estimates for each gas and emission or removal source category. Within the discussion of each emission source,
12 specific factors affecting the uncertainty associated with the estimates are discussed.
13 Additional research in the following areas could help reduce uncertainty in the U.S. Inventory:
14 • Incorporating excluded emission sources. Quantitative estimates for some of the sources and sinks of
15 greenhouse gas emissions are not available at this time. In particular, emissions from some land-use
16 activities and industrial processes are not included in the inventory either because data are incomplete or
17 because methodologies do not exist for estimating emissions from these source categories. See Annex 5 of
18 this report for a discussion of the sources of greenhouse gas emissions and sinks excluded from this report.
19 • Improving the accuracy of emission factors. Further research is needed in some cases to improve the
20 accuracy of emission factors used to calculate emissions from a variety of sources. For example, the
21 accuracy of current emission factors applied to CH4 and N2O emissions from stationary and mobile
22 combustion is highly uncertain.
23 • Collecting detailed activity data. Although methodologies exist for estimating emissions for some sources,
24 problems arise in obtaining activity data at a level of detail in which aggregate emission factors can be
25 applied. For example, the ability to estimate emissions of SF6 from electrical transmission and distribution
26 is limited due to a lack of activity data regarding national SF6 consumption or average equipment leak
27 rates.
28 The overall uncertainty estimate for total U.S. greenhouse gas emissions was developed using the IPCC Approach 2
29 uncertainty estimation methodology. Estimates of quantitative uncertainty for the total U.S. greenhouse gas
30 emissions are shown below, in Table 1-5.
31 The IPCC provides good practice guidance on two approaches—Approach 1 and Approach 2—to estimating
32 uncertainty for individual source categories. Approach 2 uncertainty analysis, employing the Monte Carlo
33 Stochastic Simulation technique, was applied wherever data and resources permitted; further explanation is provided
34 within the respective source category text and in Annex 7. Consistent with the 2006 IPCC Guidelines (IPCC 2006),
35 over a multi-year timeframe, the United States expects to continue to improve the uncertainty estimates presented in
36 this report.
37 Table 1-5: Estimated Overall Inventory Quantitative Uncertainty (MMT COz Eq. and Percent)
38 (TO BE UPDATED)
2013 Emission Uncertainty Range Relative to Emission Standard
Estimate3 Estimate1" Mean0 Deviation0
Gas (MMT CCh Eq.) (MMT CCh Eq.) (%) (MMT CCh Eq.)
• Lower Upper Lower Upper
Bound"1 Bound"1 Bound Bound
CO2
CH4e
N2Oe
PFC,HFC, SF6,andNF3e
5,505 5,400 5,766 -2% 5% 5,584
636 573 751 -10% 18% 656
355 320 445 -10% 25% 376
171 170 190 -1% 11% 180
95
45
32
5
Introduction 1-21
-------
Total 6,667 6,584 7,008 -1% 5% 6,795 110
Net Emissions (Sources and
Sinks) 5,785.5 5,613 6,220 -3% 8% 5,916 154
Notes:
a Emission estimates reported in this table correspond to emissions from only those source categories for which quantitative
uncertainty was performed this year. Thus the totals reported in this table exclude approximately 5.7 MMT CCh Eq. of
emissions for which quantitative uncertainty was not assessed. Hence, these emission estimates do not match the final total U.S.
greenhouse gas emission estimates presented in this Inventory.
b The lower and upper bounds for emission estimates correspond to a 95 percent confidence interval, with the lower bound
corresponding to 2.5th percentile and the upper bound corresponding to 97.5th percentile.
0 Mean value indicates the arithmetic average of the simulated emission estimates; standard deviation indicates the extent of
deviation of the simulated values from the mean.
d The lower and upper bound emission estimates for the sub-source categories do not sum to total emissions because the low and
high estimates for total emissions were calculated separately through simulations.
e The overall uncertainty estimates did not take into account the uncertainty in the GWP values for CH4, N2O and high GWP
gases used in the inventory emission calculations for 2013.
1 Emissions calculated for the U.S. Inventory reflect current best estimates; in some cases, however, estimates are
2 based on approximate methodologies, assumptions, and incomplete data. As new information becomes available in
3 the future, the United States will continue to improve and revise its emission estimates. See Annex 7 of this report
4 for further details on the U.S. process for estimating uncertainty associated with the emission estimates and for a
5 more detailed discussion of the limitations of the current analysis and plans for improvement. Annex 7 also includes
6 details on the uncertainty analysis performed for selected source categories.
16
1.8 Completeness
8 This report, along with its accompanying CRF tables, serves as a thorough assessment of the anthropogenic sources
9 and sinks of greenhouse gas emissions for the United States for the time series 1990 through 2014. Althoughthis
10 report is intended to be comprehensive, certain sources have been identified which were excluded from the estimates
11 presented for various reasons. Generally speaking, sources not accounted for in this inventory are excluded due to
12 data limitations or a lack of thorough understanding of the emission process. The United States is continually
13 working to improve upon the understanding of such sources and seeking to find the data required to estimate related
14 emissions. As such improvements are implemented, new emission sources are quantified and included in the
15 Inventory. For a complete list of sources not included, see Annex 5 of this report.
1.9 Organization of Report
17 In accordance with the revision of the UNFCCC reporting guidelines agreed to at the nineteenth Conference of the
18 Parties (UNFCCC 2014), this Inventory of U.S. Greenhouse Gas Emissions and Sinks is segregated into five sector-
19 specific chapters, listed below in Table 1-6. In addition, chapters on Trends in Greenhouse Gas Emissions and
20 Other information to be considered as part of the U.S. Inventory submission are included.
21 Table 1-6: IPCC Sector Descriptions
Chapter/EPCC Sector Activities Included
Energy Emissions of all greenhouse gases resulting from stationary and mobile energy
activities including fuel combustion and fugitive fuel emissions, and non-
energy use of fossil fuels.
Industrial Processes and Emissions resulting from industrial processes and product use of greenhouse
Product Use gases.
1-22 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Agriculture Anthropogenic emissions from agricultural activities except fuel combustion,
which is addressed under Energy.
Land Use, Land-Use Emissions and removals of CCh, CELi, and N2O from forest management, other
Change, and Forestry land-use activities, and land-use change.
Waste Emissions from waste management activities.
1 Within each chapter, emissions are identified by the anthropogenic activity that is the source or sink of the
2 greenhouse gas emissions being estimated (e.g., coal mining). Overall, the following organizational structure is
3 consistently applied throughout this report:
4 Chapter/IPCC Sector Overview of emission trends for each IPCC defined sector
5 Source C3t6gory: Description of source pathway and emission trends.
6 Methodology: Description of analytical methods employed to produce emission estimates and
7 identification of data references, primarily for activity data and emission factors.
8 Uncertainty and Timeseries Consistency: A discussion and quantification of the uncertainty in
9 emission estimates and a discussion of time-series consistency.
10 QA/QC and Verification: A discussion on steps taken to QA/QC and verify the emission
11 estimates, where beyond the overall U.S. QA/QC plan, and any key findings.
12 Recalculations: A discussion of any data or methodological changes that necessitate a
13 recalculation of previous years' emission estimates, and the impact of the recalculation on the
14 emission estimates, if applicable.
15 Planned Improvements: A discussion on any source-specific planned improvements, if
16 applicable.
17 Special attention is given to CCh from fossil fuel combustion relative to other sources because of its share of
18 emissions and its dominant influence on emission trends. For example, each energy consuming end-use sector (i.e.,
19 residential, commercial, industrial, and transportation), as well as the electricity generation sector, is described
20 individually. Additional information for certain source categories and other topics is also provided in several
21 Annexes listed in Table 1-7.
Introduction 1-23
-------
1 Table 1-7: List of Annexes
ANNEX 1 Key Category Analysis
ANNEX 2 Methodology and Data for Estimating CCh Emissions from Fossil Fuel Combustion
2.1. Methodology for Estimating Emissions of CCh from Fossil Fuel Combustion
2.2. Methodology for Estimating the Carbon Content of Fossil Fuels
2.3. Methodology for Estimating Carbon Emitted from Non-Energy Uses of Fossil Fuels
ANNEX 3 Methodological Descriptions for Additional Source or Sink Categories
3.1. Methodology for Estimating Emissions of CLU, N2O, and Indirect Greenhouse Gases from Stationary
Combustion
3.2. Methodology for Estimating Emissions of CH/t, N2O, and Indirect Greenhouse Gases from Mobile
Combustion and Methodology for and Supplemental Information on Transportation-Related Greenhouse Gas
Emissions
3.3. Methodology for Estimating Emissions from Commercial Aircraft Jet Fuel Consumption
3.4. Methodology for Estimating CH4 Emissions from Coal Mining (TO BE UPDATED)
3.5. Methodology for Estimating CH4 and CO2 Emissions from Petroleum Systems (TO BE UPDATED)
3.6. Methodology for Estimating CtLt Emissions from Natural Gas Systems (TO BE UPDATED)
3.7. Methodology for Estimating CO2 and N2O Emissions from Incineration of Waste
3.8. Methodology for Estimating Emissions from International Bunker Fuels used by the U.S. Military
3.9. Methodology for Estimating HFC and PFC Emissions from Substitution of Ozone Depleting Substances
3.10. Methodology for Estimating CEU Emissions from Enteric Fermentation
3.11. Methodology for Estimating CtLt and N2O Emissions from Manure Management
3.12. Methodology for Estimating N2O Emissions and Soil Organic C Stock Changes from Agricultural Soil
Management (Cropland and Grassland) (TO BE UPDATED)
3.13. Methodology for Estimating Net Carbon Stock Changes in Forest Lands Remaining Forest Lands (TO BE
UPDATED)
3.14. Methodology for Estimating CtLt Emissions from Landfills
ANNEX 4 IPCC Reference Approach for Estimating CO2 Emissions from Fossil Fuel Combustion
ANNEX 5 Assessment of the Sources and Sinks of Greenhouse Gas Emissions Not Included (TO BE UPDATED)
ANNEX 6 Additional Information
6.1. Global Warming Potential Values
6.2. Ozone Depleting Substance Emissions
6.3. Sulfur Dioxide Emissions
6.4. Complete List of Source Categories
6.5. Constants, Units, and Conversions
6.6. Abbreviations
6.7. Chemical Formulas
ANNEX 7 Uncertainty
7.1. Overview
7.2. Methodology and Results
7.3. Planned Improvements
ANNEX 8 QA/QC Procedures
8.1. Background
8.2. Purpose
8.3. Assessment Factors
1-24 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
4
5
6
7
8
9
10
11
12
13
14
15
2. Trends in Greenhouse Gas Emissions
2.1 Recent Trends in U.S. Greenhouse Gas
Emissions and Sinks
In 2014, total U.S. greenhouse gas emissions were 6,872.6 MMT or million metric tons CC>2 Eq. Total U.S.
emissions have increased by 7.7 percent from 1990 to 2014, and emissions increased from 2013 to 2014 by 0.9
percent (61.5 MMT CCh Eq.). The increase in CCh emissions from fossil fuel combustion was a result of multiple
factors, including: (1) colder winter conditions in the first quarter of 2014 resulting in an increased demand for
heating fuel in the residential and commercial sectors; (2) an increase in transportation emissions resulting from an
increase in vehicle miles traveled (VMT) and fuel use across on-road transportation modes; and (3) an increase in
industrial production across multiple sectors resulting in slight increases in industrial sector emissions. Since 1990,
U.S. emissions have increased at an average annual rate of 0.3 percent. Figure 2-1 through Figure 2-3 illustrate the
overall trend in total U.S. emissions by gas, annual changes, and absolute changes since 1990.
Figure 2-1: U.S. Greenhouse Gas Emissions by Gas (MMT COz Eq.)
• MFCs, PFCs, SF6 and NF3
• Methane
Nitrous Oxide
• Carbon Dioxide
C OT1
6]666 6RV] &8T3
rMm^-m^Di^ooo^Oi-ifMm^-m^Dr^ooc^Oi-i
CT^dC^^C^^C'iCTiOOOOOOOOOO1'—ii—ii—IT-i-i—I
o^a^c^aio^^aic^ooooooooooooooo
,Hi-Hi-ii-ii-ii-ii-ii-i(NrslrNlpJrNlrslrsl(Nrslrv]rslrslrvlpjR
Trends 2-1
-------
Figure 2-2: Annual Percent Change in U.S. Greenhouse Gas Emissions
2
o
6
4
5
4% -
2%
0%
3.2%
2.9%
2.2%
0.9%
Figure 2-3: Cumulative Change in Annual U.S. Greenhouse Gas Emissions Relative to 1990
(MMT COz Eq.)
1,048 1/099
s a a a a
7 Overall, from 1990 to 2014, total emissions of CO2 increased by 440.2 MMT CO2 Eq. (8.6 percent), while total
8 emissions of CH4 decreased by 37.4 MMT €62 Eq. (5.0 percent), and total emissions of N2O increased by 1.9 MMT
9 CO2 Eq. (0.5 percent). During the same period, aggregate weighted emissions of HFCs, PFCs, SF6, and NF3 rose by
10 87.1 MMT CO2 Eq. (85.4 percent). Despite being emitted in smaller quantities relative to the other principal
11 greenhouse gases, emissions of HFCs, PFCs, SF6, and NF3 are significant because many of them have extremely
12 high GWPs and, in the cases of PFCs SF6, and NF3, long atmospheric lifetimes. Conversely, U.S. greenhouse gas
13 emissions were partly offset by C sequestration in managed forests, trees in urban areas, agricultural soils, and
14 landfilled yard trimmings. These were estimated to offset 10.0 percent of total emissions in 2014.
15 As the largest contributor to U.S. greenhouse gas emissions, carbon dioxide (CCh) from fossil fuel combustion has
16 accounted for approximately 76 percent of global warming potential (GWP) weighted emissions for the entire time
17 series since 1990, from 74 percent of total GWP-weighted emissions in 1990 to 76 percent in 2014. Emissions from
18 this source category grew by 9.9 percent (468.0 MMT CO2 Eq.) from 1990 to 2014 and were responsible for most of
19 the increase in national emissions during this period. From 2013 to 2014, these emissions increased by 1.0 percent
20 (51.1 MMT CO2 Eq.). Historically, changes in emissions from fossil fuel combustion have been the dominant factor
21 affecting U.S. emission trends.
22 Changes in CO2 emissions from fossil fuel combustion are influenced by many long-term and short-term factors,
23 including population and economic growth, energy price fluctuations, technological changes, energy fuel choices,
24 and seasonal temperatures. On an annual basis, the overall consumption of fossil fuels in the United States
25 fluctuates primarily in response to changes in general economic conditions, energy prices, weather, and the
2-2 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 availability of non-fossil alternatives. For example, in a year with increased consumption of goods and services, low
2 fuel prices, severe summer and winter weather conditions, nuclear plant closures, and lower precipitation feeding
3 hydroelectric dams, there would likely be proportionally greater fossil fuel consumption than in a year with poor
4 economic performance, high fuel prices, mild temperatures, and increased output from nuclear and hydroelectric
5 plants.
6 In the longer-term, energy consumption patterns respond to changes that affect the scale of consumption (e.g.,
7 population, number of cars, and size of houses), the efficiency with which energy is used in equipment (e.g., cars,
8 power plants, steel mills, and light bulbs) and behavioral choices (e.g., walking, bicycling, or telecommuting to work
9 instead of driving).
10 Energy-related CC>2 emissions also depend on the type of fuel or energy consumed and its carbon (C) intensity.
11 Producing a unit of heat or electricity using natural gas instead of coal, for example, can reduce the €62 emissions
12 because of the lower C content of natural gas.
13 A brief discussion of the year to year variability in fuel combustion emissions is provided below, beginning with
14 2010.
15 From 2010 to 2011, CC>2 emissions from fossil fuel combustion decreased by 2.4 percent. This decrease is a result of
16 multiple factors including: (1) a decrease in the carbon intensity of fuels consumed to generate electricity due to a
17 decrease in coal consumption, with increased natural gas consumption and a significant increase in hydropower
18 used; (2) a decrease in transportation-related energy consumption due to higher fuel costs, improvements in fuel
19 efficiency, and a reduction in miles traveled; and (3) relatively mild winter conditions resulting in an overall
20 decrease in energy demand in most sectors. Changing fuel prices played a role in the decreasing emissions. A
21 significant increase in the price of motor gasoline in the transportation sector was a major factor leading to a
22 decrease in energy consumption by 1.2 percent. In addition, an increase in the price of coal and a concurrent
23 decrease in natural gas prices led to a 5.7 percent decrease and a 2.5 percent increase in fuel consumption of these
24 fuels by electric generators. This change in fuel prices also reduced the carbon intensity of fuels used to produce
25 electricity in 2011, further contributing to the decrease in fossil fuel combustion emissions.
26 From 2011 to 2012, CC>2 emissions from fossil fuel combustion decreased by 3.9 percent, with emissions from fossil
27 fuel combustion at their lowest level since 1994. This decrease from 2011 to 2012 is primarily a result of the
28 decrease in the carbon intensity of fuels used to generate electricity due to a slight increase in the price of coal, and a
29 significant decrease in the price of natural gas. The consumption of coal used to generate electricity decreased by
30 12.3 percent, while consumption of natural gas for electricity generation increased by 20.4 percent. Also, emissions
31 declined in the transportation sector largely due to a small increase in fuel efficiency across different transportation
32 modes and limited new demand for passenger transportation. In 2012, weather conditions remained fairly constant in
3 3 the summer and were much warmer in the winter compared to 2011, as cooling degree days increased by 1.7 percent
34 while heating degree days decreased 12.6 percent. This decrease in heating degree days resulted in a decreased
35 demand for heating fuel in the residential and commercial sector, which had a decrease in natural gas consumption
36 of 11.7 and 8.0 percent, respectively.
37 From 2012 to 2013, CC>2 emissions from fossil fuel combustion increased by 2.6 percent. This increase is primarily a
38 result of the increased energy consumption in the residential and commercial sectors, as heating degree days
39 increased 18.5 percent in 2013 as compared to 2012. The cooler weather led to an increase of 17.1 and 12.9 percent
40 direct use of fuels in the residential and commercial sectors, respectively. In addition, there was an increase of 1.5
41 and 0.8 percent in electricity consumption in the residential and commercial sectors, respectively, due to regions that
42 heat their homes with electricity. The consumption of natural gas used to generate electricity decreased by 9.8
43 percent due to an increase in the price of natural gas. Electric power plants shifted some consumption from natural
44 gas to coal, and as a result increased coal consumption to generate electricity by 4.0 percent. Lastly, industrial
45 production increased 1.9 percent from 2012 to 2013, resulting in an increase in the in CO2 emissions from fossil fuel
46 combustion from the industrial sector by 3.7 percent.
47 From 2013 to 2014, €62 emissions from fossil fuel combustion increased by 1.0 percent. This increase is primarily a
48 result of the increased energy consumption in the transportation, residential, and commercial sectors. In the
49 transportation sector, vehicle miles traveled increased by 1.3 percent resulting in increased fuel consumption across
50 on-road transportation modes. In the residential and commercial sectors, heating degree days increased 1.9 percent
51 in 2014 as compared to 2013, resulting in an increased demand in heating fuels for these sectors. The cooler weather
52 led to an increase of 4.5 and 4.9 percent in direct use of fuels in the residential and commercial sectors, respectively.
Trends 2-3
-------
1 In addition, there was an increase of 0.9 and 1.1 percent in electricity consumption in the residential and commercial
2 sectors, respectively, due to regions that heat their homes with electricity. There was also an increase in
3 transportation emissions resulting from an increase in vehicle miles traveled (VMT) and fuel use across on-road
4 transportation modes in 2014. Lastly, industrial production increased 3.7 percent from 2013 to 2014, resulting in a
5 slight increase in CCh emissions from fossil fuel combustion from the industrial sector by 0.2 percent.
6 Table 2-1 summarizes emissions and sinks from all U.S. anthropogenic sources in weighted units of MMT €62 Eq.,
7 while unweighted gas emissions and sinks in kilotons (kt) are provided in Table 2-2.
8 Table 2-1: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks (MMT COz Eq.)
Gas/Source
1990
2005
2010
2011
2012
2013
2014
C02
Fossil Fuel Combustion
Electricity Generation
Transportation
Industrial
Residential
Commercial
U.S. Territories
Non-Energy Use of Fuels
5
4
1
1
,124.0
,740.7
,820.8
,493.8
842.5
338.3
217.4
27.9
118.1
6,132.6
5,747.1
2,400.9
1
,887.0
828.0
357.8
223.5
49.9
138.9
5,698.2
5,358.3
2,258.4
1,728.3
775.5
334.6
220.1
41.4
114.1
5,568.6
5,227.7
2,157.7
1,707.6
773.3
326.8
220.7
41.5
108.5
5,361.0
5,024.7
2,022.2
1,696.8
782.9
282.5
196.7
43.6
105.6
5,513.2
5,157.6
2,038.1
1,713.0
812.2
329.7
221.0
43.5
121.7
5,564.3
5,208.7
2,039.3
1,737.4
814.2
345.1
231.6
41.0
114.3
Iron and Steel Production &
Metallurgical Coke Production
Cement Production
Natural Gas Systems'
Petrochemical Production
Lime Production
Other Process Uses of Carbonates
Ammonia Production
Incineration of Waste
Petroleum Systems'5
Urea Fertilization
Carbon Dioxide Consumption
Liming
Urea Consumption for Non-
Agricultural Purposes
Aluminum Production
Soda Ash Production and
Consumption
Ferroalloy Production
Litanium Dioxide Production
Glass Production
Phosphoric Acid Production
Zinc Production
Peatlands Remaining Peatlands
Lead Production
Silicon Carbide Production and
Consumption
Magnesium Production and
Processing
LULUCF Total Net Fluxc
Wood Biomass and Ethanol
Consumption11
International Bunker Fuels"
CH4
Landfills
Enteric Fermentation
Natural Gas Systems*
Coal Mining
2.8
22
1.2
15
1 5\
0.6
1.1
0.5
0.4|
3.7
4.1
s
L9
1.3
1.0
1.1
0.6
0.2
(704.2)
219.4
103.5
745.3
184.4
164.2
179.1
96.5
229.8
1135.4
187.3
4.7
2.7
2.7
1.7
1.8
1.5
1.1
1.2
1.0
0.5
0.2
4.0
3.3
2.7
1.7
1.7
1.3
1.2
1.3
0.9
0.5
0.2
4.4
3.4
2.8
1.9
1.5
1.2
1.1
1.5
0.8
0.5
0.2
52.2
36.1
37.8
26.4
14.0
10.4
10.0
9.4
6.0
4.3
4.2
3.9
4.2
3.3
2.8
1.8
1.7
1.3
1.1
1.4
0.8
0.5
0.2
229.8
113.1
735.4
187.3
168.9J
176.3
64.11
117.0 111.7 105.8
55.4
38.8
37.8
26.5
14.1
12.1
9.4
9.4
6.0
4.5
4.5
4.1
4.0
3.3
2.8
1.9
1.8
1.3
1.1
1.0
0.8
0.5
0.2
(683.2) (683.6) (680.8) (682.4) (685.8)
265.1 268.1 267.7 286.3 293.7
99.8 103.2
720.8
176.3
171.3
159.6
82.3
711.8
176.9
168.9
159.3
71.2
703.8
173.5
166.7
154.4
66.5
704.0
176.7
165.5
157.4
64.6
707.9
181.8
164.3
157.4
64.6
2-4 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Manure Management
Petroleum Systems'5
Wastewater Treatment
Rice Cultivation
Stationary Combustion
Forest Fires
Abandoned Underground Coal
Mines
Composting
Mobile Combustion
Field Burning of Agricultural
Residues
Petrochemical Production
Ferroalloy Production
Silicon Carbide Production and
Consumption
Iron and Steel Production &
Metallurgical Coke Production
Peatlands Remaining Peatlands
Incineration of Waste
International Bunker Fuels"
N20
Agricultural Soil Management
Stationary Combustion
Manure Management
Mobile Combustion
Nitric Acid Production
Adipic Acid Production
Wastewater Treatment
Forest Fires
N2O from Product Uses
Settlement Soils
Composting
Forest Soils
Incineration of Waste
Semiconductor Manufacture
Field Burning of Agricultural
Residues
Peatlands Remaining Peatlands
International Bunker Fuels"
HFCs,PFCs, SF6andNF3
HFCs
Substitution of Ozone Depleting
Substancesf
HCFC-22 Production
Semiconductor Manufacture
Magnesium Production and
Processing
PFCs
Aluminum Production
Semiconductor Manufacture
SF6
Electrical Transmission and
Distribution
Magnesium Production and
Processing
Semiconductor Manufacture
NF3
Semiconductor Manufacture
15.7
11.3
8.5
3.3
72|
0.4
56
d
0.21
409.5
302.9
11.9
14.0
41.2
12.1
15.21
4.2J
1.4
0.3
0.1
0.5
0.1
+
0.9\
102.0
46.6
56.3
23.5
15.91
14.21
7.4
9.9
6.6
1.9
2.7
d
0.1\
406.4
296.7
20.2
16. sl
34.41
11.3
7.1
4.3
65
4.2
2.3l
1 7I
0.5J
0.4
0.1
1.0
154.4
133.3
60.9
21.3
15.5
12.2
7.1
3.3
61.5
22.0
15.3
12.2
7.1
6.6
63.7
23.3
15.2
12.2
6.6
11.1
61.4
25.2
15.0
12.2
8.0
7.3
61.2
25.2
15.0
12.2
8.1
7.3
6.6
1.8
2.3
0.3
0.1
415.2
320.4
22.2
17.2
23.6
11.5
4.2
4.7
2.2
4.2
2.4
1.6
0.5
0.3
0.1
0.1
+
1.0
176.2
161.7
153.5
8.0
0.2
7.0
6.4
1.9
2.2
0.3
0.1
423.8
322.9
21.3
17.4
22.4
10.9
10.2
4.8
4.4
4.2
2.5
1.7
0.5
0.3
0.2
0.1
+
1.0
183.6
166.1
157.1
8.8
0.2
6.2
1.9
2.2
0.3
0.1
0.1
419.4
322.9
21.4
17.5
20.0
10.5
5.5
4.9
7.3
4.2
2.5
1.7
0.5
0.3
0.2
0.1
+
0.9
181.4
167.1
161.4
5.5
0.2
6.2
2.0
2.1
0.3
0.1
0.1
411.0
318.4
22.9
17.5
18.2
10.7
4.0
4.9
4.8
4.2
2.4
1.8
0.5
0.3
0.2
0.1
+
0.9
182.9
169.6
165.3
4.1
0.2
5.7
5.1
6.2
2.1
2.0
0.3
0.1
0.1
411.4
318.5
23.4
17.5
16.3
10.9
5.4
4.9
4.8
4.2
2.4
1.8
0.5
0.3
0.2
0.1
+
0.9
189.1
175.8
171.4
4.1
0.2
+
4.4
1.9
2.6
9.5
+
6.9
3.5
3.4
10.0
+
6.0
2.9
3.0
7.7
0.1
5.8
3.0
2.9
6.9
0.1
5.8
3.0
2.9
6.9
5.1
2.1
0.4
0.5
0.5
2.8
0.4
0.7
0.7
1.6
0.4
0.6
0.6
1.4
0.4
0.6
0.6
1.4
0.4
0.6
0.6
Trends 2-5
-------
Total Emissions
LULUCF Emissions?
LULUCF Total Net Fluxc
LULUCF Sector Total"
Net Emissions (Sources and Sinks)
6,380.8
15.0
(704.2)
(689.1)
5,676.6
7,428.8
28.2
(636.1)
(607.9)
6,792.6
7,010.5
17.8
(683.2)
(665.3)
6,327.3
6,887.8
22.9
(683.6)
(660.7)
6,204.2
6,665.7
32.3
(680.8)
(648.5)
5,984.9
6,811.2
24.1
(682.4)
(658.3)
6,128.8
6,872.6
24.6
(685.8)
(661.3)
6,186.8
+ Does not exceed 0.05 MMT CO2 Eq.
a The values in this table for Natural Gas Systems are presented from the previous Inventory and do not reflect
updates to emission estimates for this category. See Section 3.7, Natural Gas Systems of the Energy chapter for
more information. Gray highlighting was added on 2/24 for clarification.
b The values in this table for Petroleum Systems are presented from the previous Inventory and do not reflect
updates to emission estimates for this category. See Section 3.6, Petroleum Systems of the Energy chapter for more
information. Gray highlighting was added on 2/24 for clarification.
c Total net flux from LULUCF is only included in the Net Emissions total. Net flux from LULUCF includes the
positive C sequestration reported for Forest Land Remaining Forest Land, Land Converted to Forest Land,
Cropland Remaining Cropland, Land Converted to Grassland, Settlements Remaining Settlements, and Other Land
plus the loss in C sequestration reported for Land Converted to Cropland and Grassland Remaining Grassland.
Refer to Table 2-8 for a breakout of emissions and removals for Land Use, Land-Use Change, and Forestry by gas
and source category.
d Emissions from Wood Biomass and Ethanol Consumption are not included specifically in summing energy sector
totals. Net carbon fluxes from changes in biogenic carbon reservoirs are accounted for in the estimates for Land
Use, Land-Use Change, and Forestry.
e Emissions from International Bunker Fuels are not included in totals.
f Small amounts of PFC emissions also result from this source.
g LULUCF emissions include the CCh, CELi, andN2O emissions reported for Forest Fires, Forest Soils, Liming,
Urea Fertilization, Settlement Soils, and Peatlands Remaining Peatlands.
h The LULUCF Sector Total is the sum of positive emissions (i.e., sources) of greenhouse gases to the atmosphere
plus removals of CCh (i.e., sinks or negative emissions) from the atmosphere.
Note: Totals may not sum due to independent rounding. Parentheses indicate negative values or sequestration.
2-6 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Table 2-2: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks (kt)
Gas/Source
C02
Fossil Fuel Combustion
Electricity Generation
Transportation
Industrial
Residential
Commercial
U.S. Territories
Non-Energy Use of Fuels
Iron and Steel Production &
Metallurgical Coke
Production
Cement Production
Natural Gas Systems*
Petrochemical Production
Lime Production
Other Process Uses of
Carbonates
Ammonia Production
Incineration of Waste
Petroleum Systems'5
Urea Fertilization
Carbon Dioxide Consumption
Liming
Urea Consumption for Non-
Agricultural Purposes
Aluminum Production
Soda Ash Production and
Consumption
Ferroalloy Production
Titanium Dioxide Production
Glass Production
Phosphoric Acid Production
Zinc Production
Peatlands Remaining Peatlands
Lead Production
Silicon Carbide Production and
Consumption
Magnesium Production and
Processing
LULUCF Total NetFluxc
WoodBiomass and Ethanol
Consumption11
International Bunker Fuels"
CH4
Landfills
Enteric Fermentation
Natural Gas Systems*
Coal Mining
Manure Management
Petroleum Systems'5
Wastewater Treatment
Rice Cultivation
Stationary Combustion
Forest Fires
1990
5,124,039
4,740,671 1
1,820,818
1,493,758
842,473
338,347
217,393
27,882
118,114
99,669|
3 3, 278 1
37,645
21^609!
ll,70ol
4,907|
1 3,047|
7,972 1
4,445
2^417!
1,472|
4,667|
3,784l
6,831 1
2,822l
2,152l
l,195l
l,535l
l,529l
6321
l,055l
516M
;
i|
(704,191)
219,413
103,463
29,813
7,376 1
6,566
7,165
3,860
1,486|
1,261
6261
451
339
131
2005
6,132,598
5,747,142
2,400,874
1,887,033
827,999
357,834
223,480
49,923
138,876
66,543
45,910
29,995
27,380
14,552
6,339l
9,196 1
12,454
4,904
3,504
l,375l
,^,,-^y —
229,844
113,139
29,414
7,493 1
6,755
7,053
2,565
2,254!
•••
635H
567
295
397
2010
5,698,233
5,358,292
2,258,399
1,728,267
775,535
334,587
220,125
41,379
114,063
55,671
31,256
32,334
27,246
13,381
9,560
9,188
11,026
•U53
3,778
4,425
4,784
4,730
2,722
2,697
1,663
1,769
1,481
1,087
1,182
1,022
542
181
1
(683,157)
265,110
116,992
28,832
7,052
6,853
6,382
3,293
2,437
• 854
619
486
282
131
2011
5,568,569
5,227,690
2,157,688
1,707,631
773,312
326,808
220, 749
41,503
108,515
59,928
32,010
35,551
26,326
13,981
9,335
9,292
10,550
4.467
4,099
4,083
3,873
4,029
3,292
2,712
1,735
1,729
1,299
1,151
1,286
926
538
170
3
(683,557)
268,064
111,660
28,470
7,074
6,757
6,371
2,849
2,460
M
610
487
283
265
2012
5,360,975
5,024,685
2,022,181
1,696,752
782,929
282,540
196,714
43,569
105,617
54,229
35,051
34,764
26,464
13,715
8,022
9,377
10,362
5,060
4,225
4,019
5,978
4,449
3,439
2,763
1,903
1,528
1,248
1,093
1,486
812
527
158
2
(680,812)
267,730
105,805
28,153
6,942
6,670
6,176
2,658
2,548
• 931
606
488
264
443
2013
5,513,230
5,157,583
2,038,122
1,713,013
812,217
329,674
221,030
43,528
121,682
52,201
36,146
37,808
26,437
14,045
10,414
9,962
9,421
6,001
4,342
4,188
3,909
4,179
3,255
2,804
1,785
1,715
1,317
1,119
1,429
770
546
169
2
(682,365)
286,323
99,763
28,161
7,066
6,619
6,295
2,584
2,455
1,009
601
488
320
294
2014
5,564,285
5,208,654
2,039,321
1,737,410
814,197
345,145
231,590
40,991
114,333
55,355
38,755
37,808
26,509
14,125
12,077
9,436
9,421
6,001
4,514
4,471
4,139
4,007
3,255
2,827
1,914
1,755
1,341
1,095
956
842
518
173
2
(685,827)
293,729
103,201
28,315
7,271
6,572
6,295
2,584
2,447
1,009
601
488
323
294
Trends 2-7
-------
Abandoned Underground Coal
Mines 288 • 264
Composting I5l 15
Mobile Combustion 226 110
Field Burning of Agricultural
Residues 101
Petrochemical Production 9 H 31
Ferroalloy Production IP +1
Silicon Carbide Production and
Consumption
Iron and Steel Production &
Metallurgical Coke
Production 11 11
Peatlands Remaining Peatlands + +1
Incineration of Waste + +1
International Bunker Fuels" 7 5
N20 1,374 1,364
Agricultural Soil Management 1,017 996
Stationary Combustion 401 68
Manure Management 47H 55
Mobile Combustion 13 81 115
Nitric Acid Production 411 38
Adipic Acid Production 511 24
Wastewater Treatment III 15
Forest Fires 7H 22
N2O from Product Uses 141 14
Settlement Soils 5U 8
Composting 11 6
Forest Soils +B 2
Incineration of Waste 2 1
Semiconductor Manufacture + +
Field Burning of Agricultural
Residues + +
Peatlands Remaining Peatlands + +
International Bunker Fuels" 3U 3
HFCs, PFCs, SF6 and NF3 M M
HFCs M M
Substitution of Ozone
Depleting Substancesf M M
HCFC-22 Production 3 1
Semiconductor Manufacture +B +
Magnesium Production and
Processing 0 0
PFCs M M
Aluminum Production M M
Semiconductor Manufacture M M
Electrical Transmission and
Distribution 11 +
Magnesium Production and
Processing +B +
Semiconductor Manufacture + +
Semiconductor Manufacture + +
263
73
91
11
2
6
1,393
1,075
74
58
79
39
14
16
7
14
8
5
1
1
3
M
M
M
1
M
M
M
257
15
90
11
2
5
1,422
1,084
71
58
75
37
34
16
15
14
8
6
2
1
1
3
M
M
M
1
M
M
M
249
77
86
11
3
1
4
1,408
1,083
72
59
67
35
19
16
24
14
9
6
2
1
1
3
M
M
M
M
M
M
249
81
84
11
3
3
1,379
1,069
77
59
61
36
13
17
16
14
8
6
2
1
1
3
M
M
M
M
M
M
249
82
82
11
5
1
3
,381
,069
79
59
55
37
18
17
16
14
8
6
2
1
1
3
M
M
M
M
M
M
+ Does not exceed 0.5 kt.
M Mixture of multiple gases
a The values in this table for Natural Gas
estimates for this category. See Section
added on 2/24 for clarification.
Systems are presented from the previous Inventory and do not reflect updates to emission
3.7, Natural Gas Systems of the Energy chapter for more information. Gray highlighting was
2-8 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
b The values in this table for Petroleum Systems are presented from the previous Inventory and do not reflect updates to emission
estimates for this category. See Section 3.6, Petroleum Systems of the Energy chapter for more information. Gray highlighting was
added on 2/24 for clarification.
c Total net flux from LULUCF is only included in the Net Emissions total. Net flux from LULUCF includes the positive C sequestration
reported for Forest Land Remaining Forest Land, Land Converted to Forest Land, Cropland Remaining Cropland, Land Converted to
Grassland, Settlements Remaining Settlements, and Other Land plus the loss in C sequestration reported for Land Converted to Cropland
and Grassland Remaining Grassland. Refer to Table 2-8 for a breakout of emissions and removals for Land Use, Land-Use Change, and
Forestry by gas and source category.
d Emissions from Wood Biomass and Ethanol Consumption are not included specifically in summing energy sector totals. Net carbon
fluxes from changes in biogenic carbon reservoirs are accounted for in the estimates for Land Use, Land-Use Change, and Forestry.
e Emissions from International Bunker Fuels are not included in totals.
f Small amounts of PFC emissions also result from this source.
Note: Totals may not sum due to independent rounding. Parentheses indicate negative values or sequestration.
1 Emissions of all gases can be summed from each source category into a set of five sectors defined by the
2 Intergovernmental Panel on Climate Change (IPCC). Over the twenty five-year period of 1990 to 2014, total
3 emissions in the Energy, Industrial Processes and Product Use, Agriculture, and Waste sectors grew by 388.9 MMT
4 CO2 Eq. (7.3 percent), 47.7 MMT CO2 Eq. (14.0 percent), 44.2 MMT CO2 Eq. (8.3 percent), and 1.5 MMT CO2 Eq.
5 (0.7 percent), respectively. Over the same period, estimates of net C sequestration for the Land Use, Land-Use
6 Change, and Forestry sector (magnitude of emissions plus CO2 removals from all LULUCF source categories)
7 decreased by 27.9 MMT CO2 Eq. (4.0 percent).
8 Figure 2-4: U.S. Greenhouse Gas Emissions and Sinks by Chapter/IPCC Sector (MMT COz
9 Eq.)
10
Industrial Processes and
Product Use
LULUCF (emissions)
Energy
Use, Land-Use Change and Forestry (LULUCF) (removals
ddO^^
11
12
13
Table 2-3: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks by Chapter/IPCC
Sector (MMT COz Eq.)
Chapter/IPCC Sector
1990
2005
2010
2011
2012
2013
2014
Energy
Fossil Fuel Combustion
Natural Gas Systems1
Non-Energy Use of Fuels
Coal Mining
Stationary Combustion
Petroleum System:
5,290.9
4,740.7|
118.1
96.5
6,269.0
5,747.11
206.3
138.9
5,845.2
5,358.3
191.9
114.1
82.3
29.2
25.5
5,699.0
5,227.7
194.8
108.5
71.2
28.4
26.4
5,481.3
5,024.7
189.2
105.6
66.5
28.0
5,637.4
5,157.6
195.2
121.7
64.6
30.9
5,679.8
5,208.7
195.2
114.3
64.6
31.5
Trends 2-9
-------
Mobile Combustion 46.9 37.1
Incineration of Waste 8.4M 12.8
Abandoned Underground Coal Mines 7.21 6.6
Industrial Processes and Product Use 340.9 367.6
Substitution of Ozone Depleting
Substances O.sB 113.0
Iron and Steel Production &
Metallurgical Coke Production 99.7 66.6
Cement Production 33.3 45.9
Petrochemical Production 21.8 27.5
Lime Production 11.lU 14.6
Other Process Uses of Carbonates 4.91 6.3
Nitric Acid Production 12.1 11.3
Ammonia Production 13.0 9.2
Aluminum Production 28.3 7.6
Adipic Acid Production 15.2 7.1
Electrical Transmission and
Distribution 25.4 10.6
Carbon Dioxide Consumption 1-^1
N2O from Product Uses 4.21 4.2
Semiconductor Manufacture 3.6H 4.7
HCFC-22 Production 46.1 20.0
Urea Consumption for Non-
Agricultural Purpo ses 3.8 M 3.7
Soda Ash Production and
Consumption 2.8H 3.0
Ferroalloy Production 2.2H 1.4
Titanium Dioxide Production 1.21 1.8
Magnesium Production and
Processing ^-2H 2.7
Glass Production 1.51 1.9
Phosphoric Acid Production 1-^B
Zinc Production 0.6H 1.0
Lead Production O.sB 0.6
Silicon Carbide Production and
Consumption 0.4H 0.2
Agriculture 529.8 552.9
Agricultural Soil Management 302.9 296.7
0 0
Enteric Fermentation 164.2 168.9
Manure Management 51.1 72.9
Rice Cultivation 11.3 14.2
Field Burning of Agricultural
Residues O.sB 0.3
Land Use, Land-Use Change, and
25.9
11.4
6.6
365.2
153.5
4.7
0.2
582.3
320.4
171.3
78.1
12.2
0.4
24.7
10.9
6.4
382.2
157.1
4.0
0.2
583.3
322.9
168.9
78.9
12.2
0.4
22.2
10.7
6.2
371.4
161.4
4.4
0.2
583.4
322.9
166.7
81.2
12.2
0.4
20.3
9.7
6.2
373.8
165.3
4.2
0.2
575.4
318.4
165.5
78.9
12.2
0.4
18.4
9.7
6.2
388.6
171.4
55.7
31.3
27.3
13.4
9.6
11.5
9.2
4.6
4.2
59.9
32.0
26.4
14.0
9.3
10.9
9.3
6.8
10.2
54.2
35.1
26.5
13.7
8.0
10.5
9.4
6.4
5.5
52.2
36.1
26.5
14.0
10.4
10.7
10.0
6.2
4.0
55.4
38.8
26.6
14.1
12.1
10.9
9.4
6.2
5.4
7.0
4.4
4.2
3.8
8.0
6.8
4.1
4.2
4.9
8.8
5.7
4.0
4.2
4.5
5.5
5.1
4.2
4.2
4.2
4.1
5.1
4.5
4.2
4.2
4.1
4.0
2.7
1.7
1.8
2.1
1.5
1.1
1.2
0.5
2.7
1.7
1.7
2.8
1.3
1.2
1.3
0.5
2.8
1.9
1.5
1.7
1.2
1.1
1.5
0.5
2.8
1.8
1.7
1.5
1.3
1.1
1.4
0.5
2.8
1.9
1.8
1.5
1.3
1.1
1.0
0.5
0.2
574.1
318.5
164.3
78.7
12.2
0.4
Forestry (Emissions)
Forest Fires
Urea Fertilization
Liming
Settlement Soils
Peatlands Remaining Peatlands
Forest Soils
Waste
Landfills
Wastewater Treatment
Composting
Total Emissions
LULUCF Total Net Fluxc
Forest Land Remaining Forest Landd
15.0
5.4
2.4
4.7
i:i
204.1 1
184.41
19.ol
0.7
6,380.8
(704.2)
(576.0)
28.2
16. sl
3.5
1
211.1
187.3
20.2
3_5_B
7,428.8
(636.1)
(532.4)1
17.8
5.4
3.8
4.8
2.4
1.0
0.5
200.0
176.3
20.2
3.5
7,010.5
(683.2)
(585.0)
22.9
11.0
4.1
3.9
2.5
0.9
0.5
200.5
176.9
20.1
3.5
6,887.8
(683.6)
(578.1)
32.3
18.3
4.2
6.0
2.5
0.8
0.5
197.2
173.5
20.0
3.7
6,665.7
(680.8)
(576.7)
24.1
12.2
4.3
3.9
2.4
0.8
0.5
200.5
176.7
20.0
3.9
6,811.2
(682.4)
(580.1)
24.6
12.2
4.5
4.1
2.4
0.8
0.5
205.6
181.8
20.0
3.9
6,872.6
(685.8)
(583.4)
2-10 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Cropland Remaining Cropland
Land Converted to Cropland0
Grassland Remaining Grassland0
Land Converted to Grassland
Settlements Remaining Settlements
Other: Landfilled Yard Trimmings
and Food Scraps
Net Emissions (Sources and Sinks)
(43.2)1
22.8 •
(12.9)|
(8.5)1
(60.4)1
(26.0)
5,676.6
(16.5)1
14.61
2.9
(12.8)1
(80.5)1
(11.4)
6,792.6
(4.7)
15.6
2.6
(12.3)
(86.1)
(13.2)
6,327.3
(20.0)
14.2
11.3
(11.0)
(87.3)
(12.7)
6,204.2
(18.7)
14.5
11.7
(10.9)
(88.4)
(12.2)
5,984.9
(16.8)
14.8
11.9
(10.9)
(89.5)
(11.7)
6,128.8
(16.0)
14.7
11.9
(10.9)
(90.6)
(11.6)
6,186.8
1 The values in this table for Natural Gas Systems are presented from the previous Inventory and do not reflect updates to emission estimates
for this category. See Section 3.7, Natural Gas Systems of the Energy chapter for more information. Gray highlighting was added on 2/24
for clarification.
b The values in this table for Petroleum Systems are presented from the previous Inventory and do not reflect updates to emission estimates
for this category. See Section 3.6, Petroleum Systems of the Energy chapter for more information. Gray highlighting was added on 2/24 for
clarification.
c Total net flux from LULUCF is only included in the Net Emissions total. Net flux from LULUCF includes the positive C sequestration
reported for Forest Land Remaining Forest Land, Land Converted to Forest Land, Cropland Remaining Cropland, Land Converted to
Grassland, Settlements Remaining Settlements, and Other Land plus the loss in C sequestration reported for Land Converted to Cropland
and Grassland Remaining Grassland. Refer to Table 2-8 for a breakout of emissions and removals for Land Use, Land-Use Change, and
Forestry by gas and source category.
''Includes the effects of net additions to stocks of carbon stored in forest ecosystem pools and harvested wood products.
Note: Totals may not sum due to independent rounding. Parentheses indicate negative values or sequestration.
Energy
2
3
4
5
6
7
10
11
12
13
14
15
16
Energy-related activities, primarily fossil fuel combustion, accounted for the vast majority of U.S. CCh emissions for
the period of 1990 through 2014. In 2014, approximately 82 percent of the energy consumed in the United States
(on a Btu basis) was produced through the combustion of fossil fuels. The remaining 18 percent came from other
energy sources such as hydropower, biomass, nuclear, wind, and solar energy (see Figure 2-5 and Figure 2-6). A
discussion of specific trends related to CC>2 as well as other greenhouse gas emissions from energy consumption is
presented in the Energy chapter. Energy-related activities are also responsible for CH4 and N2O emissions (37
percent and 10 percent of total U.S. emissions of each gas, respectively). Table 2-4 presents greenhouse gas
emissions from the Energy chapter, by source and gas.
Figure 2-5: 2014 Energy Chapter Greenhouse Gas Sources (MMT COz Eq.)
Fossil Fuel Combustion
Natural Gas Systems'
Non-Energy Use of Fuels
Coal Mining
Stationary Combustion
Petroleum Systems"
Mobile Combustion
Incineration of Waste
Abandoned Underground Coal Mines
Energy as a Portion
of all Emissions
| 5,209
100 150
MMT CO2 Eq.
200
a The value in this figure for Natural Gas Systems is presented from the previous Inventory and does not reflect updates to
emission estimates for this category. See Section 3.7, Natural Gas Systems of the Energy chapter for more information.
b The value in this figure for Petroleum Systems is presented from the previous Inventory and does not reflect updates to
emission estimates for this category. See Section 3.6, Petroleum Systems of the Energy chapter for more information.
Trends 2-11
-------
Figure 2-6: 2014 U.S. Fossil Carbon Flows (MMT CO2 Eq.)
NEU Emissions 2
Fossil Fuel
Energy Exports
Coal Emissions
1,656
NEU Emissions 6
Natural Gas Emissions
1,432
NEU Emissions 59
NGI46 '
Coal I
Other! 89-
Fossi! Fuel
Non-Energy Consumption
Use Imports U.S,
19 Territories
42
3
4
Non-EnergyUse
Carbon Sequestered
Note: Totals may not sum due to independent rounding.
The "Balancing Item" above accounts for the statistical
imbalances and unknowns In tbe reported data sets combined
here.
NEU = Non-Energy Use
NG = Natural Gas
Table 2-4: Emissions from Energy (MMT COz Eq.)
Gas/Source
C02
Fossil Fuel Combustion
Electricity Generation
Transportation
Industrial
Residential
Commercial
U.S. Territories
Non-Energy Use of Fuels
Natural Gas Systems*
Incineration of Waste
Petroleum Systems'5
Biomass- Woodc
International Bunker Fuels'1
Biomass-Ethanolc
CH4
Natural Gas Systems*
Coal Mining
Petroleum Systems1"
Stationary Combustion
Abandoned Underground Coal
Mines
Mobile Combustion
Incineration of Waste
International Bunker Fuels'1
N20
Stationary Combustion
Mobile Combustion
Incineration of Waste
International Bunker Fuels'1
Total
1990
4,908.8
4,740.7
1,820.8
1,493.8
842.5
338.3
217.4
27.9
118.1
37.6
8.0
4.4
215.2
103.5
4.2
328.5
179.1
96.5 •
31.5
8.5
,..
5.6
+
0.2
53.6
11.9
41.2
0.5
0.9
5,290.9
2005
5,933.4
5,747.1
2,400.9M
1,887.0
828.0
357.8
223.5
49.9
138.9
30.0
12.5
4.9
206.9
113. /I
22.9
280.7
176.3
64.1
23.5
7.4
6.6
I
55.0
20.2
34.4
0.4
1.0
6,269.0
2010
5,519.9
5,358.3
2,258.4
1,728.3
775.5
334.6
220.1
41.4
114.1
32.3
11.0
4.2
192.5
117.0
72.6
279.2
159.6
82.3
21.3
7.1
6.6
2.3
+
0.1
46.1
22.2
23.6
0.3
1.0
5,845.2
2011
5,386.8
5,227.7
2,157.7
1,707.6
773.3
326.8
220.7
41.5
108.5
35.6
10.5
4.5
195.2
111.7
72.9
268.2
159.3
71.2
22.0
7.1
6.4
2.2
+
0.1
44.0
21.3
22.4
0.3
1.0
5,699.0
2012
5,180.5
5,024.7
2,022.2
1,696.8
782.9
282.5
196.7
43.6
105.6
34.8
10.4
5.1
194.9
105.8
72.8
259.1
154.4
66.5
23.3
6.6
6.2
2.2
+
0.1
41.7
21.4
20.0
0.3
0.9
5,481.3
2013
5,332.5
5,157.6
2,038.1
1,713.0
812.2
329.7
221.0
43.5
121.7
37.8
9.4
6.0
211.6
99.8
74.7
263.5
157.4
64.6
25.2
8.0
6.2
2.1
+
0.1
41.4
22.9
18.2
0.3
0.9
5,637.4
2014
5,376.2
5,208.7
2,039.3
1,737.4
814.2
345.1
231.6
41.0
114.3
37.8
9.4
6.0
277.7
103.2
76.1
263.5
157.4
64.6
25.2
8.1
6.2
2.0
+
0.1
40.0
23.4
16.3
0.3
0.9
5,679.8
2-12 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Note: Totals may not sum due to independent rounding.
+ Does not exceed 0.05 MMT CO2 Eq.
a The values in this table for Natural Gas Systems are presented from the previous Inventory and do not reflect updates to emission estimates
for this category. See Section 3.7, Natural Gas Systems of the Energy chapter for more information. Gray highlighting was added on 2/24 for
clarification.
b The values in this table for Petroleum Systems are presented from the previous Inventory and do not reflect updates to emission estimates for
this category. See Section 3.6, Petroleum Systems of the Energy chapter for more information. Gray highlighting was added on 2/24 for
clarification.
c Emissions from Wood Biomass and Ethanol Consumption are not included specifically in summing energy sector totals. Net carbon fluxes
from changes in biogenic carbon reservoirs are accounted for in the estimates for Land Use, Land-Use Change, and Forestry.
d Emissions from International Bunker Fuels are not included in totals.
1
2 Carbon dioxide emissions from fossil fuel combustion are presented in Table 2-5 based on the underlying U.S.
3 energy consumer data collected by EIA. Estimates of CC>2 emissions from fossil fuel combustion are calculated from
4 these EIA "end-use sectors" based on total consumption and appropriate fuel properties (any additional analysis and
5 refinement of the EIA data is further explained in the Energy chapter of this report). EIA's fuel consumption data for
6 the electric power sector comprises electricity-only and combined-heat-and-power (CHP) plants within the NAICS
7 22 category whose primary business is to sell electricity, or electricity and heat, to the public (nonutility power
8 producers can be included in this sector as long as they meet they electric power sector definition). EIA statistics for
9 the industrial sector include fossil fuel consumption that occurs in the fields of manufacturing, agriculture, mining,
10 and construction. EIA's fuel consumption data for the transportation sector consists of all vehicles whose primary
11 purpose is transporting people and/or goods from one physical location to another. EIA's fuel consumption data for
12 the industrial sector consists of all facilities and equipment used for producing, processing, or assembling goods
13 (EIA includes generators that produce electricity and/or useful thermal output primarily to support on-site industrial
14 activities in this sector). EIA's fuel consumption data for the residential sector consists of living quarters for private
15 households. EIA's fuel consumption data for the commercial sector consists of service-providing facilities and
16 equipment from private and public organizations and businesses (EIA includes generators that produce electricity
17 and/or useful thermal output primarily to support the activities at commercial establishments in this sector). Table
18 2-5 and Figure 2-7 summarize CC>2 emissions from fossil fuel combustion by end-use sector. Figure 2-8 further
19 describes the total emissions from fossil fuel combustion, separated by end-use sector, including CH4 and N2O in
20 addition to CO2.
21 Table 2-5: COz Emissions from Fossil Fuel Combustion by End-Use Sector (MMT COz Eq.)
End-Use Sector
Transportation
Combustion
Electricity
Industrial
Combustion
Electricity
Residential
Combustion
Electricity
Commercial
Combustion
Electricity
U.S. Territories3
Total
Electricity Generation
1990
1,496.8
1,493.8
3.ol
1,529.2
842.5
686.7B
931.41
338.3!
593.o|
755.41
217.41
538.ol
27.9
4,740.7
1,820.8
2005
1,891.8
1,887.0
4.7
1,564.6
828.0
736.61
1,214.1
357.81
856.3!
1,026.8
223.5!
803.31
49.9
5,747.1
2,400.9
2010
1,732.7
1,728.3
4.5
1,416.5
775.5
641.0
1,174.6
334.6
840.0
993.0
220.1
772.9
41.4
5,358.3
2,258.4
2011
1,711.9
1,707.6
4.3
1,398.0
773.3
624.7
1,117.5
326.8
790.7
958.8
220.7
738.0
41.5
5,227.7
2,157.7
2012
1,700.6
1,696.8
3.9
1,375.7
782.9
592.8
1,007.8
282.5
725.3
897.0
196.7
700.3
43.6
5,024.7
2,022.2
2013
1,717.0
1,713.0
4.0
1,407.0
812.2
594.7
1,064.6
329.7
734.9
925.5
221.0
704.5
43.5
5,157.6
2,038.1
2014
1,741.5
1,737.4
4.1
1,407.8
814.2
593.6
1,080.4
345.1
735.2
938.1
231.6
706.5
41.0
5,208.7
2,039.3
Note: Totals may not sum due to independent rounding. Combustion-related emissions from electricity
generation are allocated based on aggregate national electricity consumption by each end-use sector.
a Fuel consumption by U.S. Territories (i.e., American Samoa, Guam, Puerto Rico, U.S. Virgin Islands,
Wake Island, and other U.S. Pacific Islands) is included in this report.
Trends 2-13
-------
1 Figure 2-7: 2014 COz Emissions from Fossil Fuel Combustion by Sector and Fuel Type (MMT
2 COz Eq.)
2,500
2,000
Relative Contribution
by Fuel Type
2,039
1,737
Petroleum
• Coal
• Natural Gas
232
4 Figure 2-8: 2014 End-Use Sector Emissions of COz from Fossil Fuel Combustion (MMT COz
5 Eq.)
2,000
1,741
1,500
CT
LU
8 1,000 -
From Direct Fossil Fuel Combustion
From Electricity Consumption
7 The main driver of emissions in the Energy sector is CC>2 from fossil fuel combustion. Electricity generation is the
8 largest emitter of CC>2, and electricity generators consumed 34 percent of U.S. energy from fossil fuels and emitted
9 39 percent of the CCh from fossil fuel combustion in 2014. Electricity generation emissions can also be allocated to
10 the end-use sectors that are consuming that electricity, as presented in Table 2-5. The transportation end-use sector
11 accounted for 1,741.5 MMT CC>2 Eq. in 2014 or approximately 33 percent of total €62 emissions from fossil fuel
12 combustion. The industrial end-use sector accounted for 27 percent of CC>2 emissions from fossil fuel combustion.
13 The residential and commercial end-use sectors accounted for 21 and 18 percent, respectively, of €62 emissions
14 from fossil fuel combustion. Both of these end-use sectors were heavily reliant on electricity for meeting energy
15 needs, with electricity consumption for lighting, heating, air conditioning, and operating appliances contributing 68
16 and 75 percent of emissions from the residential and commercial end-use sectors, respectively. Significant trends in
17 emissions from energy source categories over the twenty five-year period from 1990 through 2014 included the
18 following:
2-14 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 • Total CO2 emissions from fossil fuel combustion increased from 4,740.7 MMT CO2 Eq. in 1990 to 5,208.7
2 MMT CO2 Eq. in 2014 - a 9.9 percent total increase over the twenty five-year period. From 2013 to 2014,
3 these emissions increased by 51.1 MMT CC>2 Eq. (1.0 percent).
4 • Substantial new data are available on natural gas and petroleum systems from subpart W of the EPA's
5 greenhouse gas reporting program (GHGRP) and a number of new studies. The EPA is evaluating
6 approaches for incorporating this new data into its emission estimates. In the Energy chapter, sections 3.7
7 and 3.6, updated draft estimates of CH4 emissions for the year 2013 are presented for Natural Gas Systems
8 and Petroleum Systems, respectively, to provide reviewers of the public review draft an indication of the
9 sector-wide emission estimates resulting from the combined changes under consideration. The EPA is
10 continuing to evaluate stakeholder feedback on the updates under consideration. For the final Inventory, the
11 2013 estimates presented in this section will be refined, and a full time series of emissions estimates will be
12 developed based on feedback received through the earlier stakeholder reviews of the memos and through
13 this public review period..
14 • Carbon dioxide emissions from non-energy use of fossil fuels decreased by 3.8 MMT CC>2 Eq. (3.2 percent)
15 from 1990 through 2014. Emissions from non-energy uses of fossil fuels were 114.3 MMT CC>2 Eq. in
16 2014, which constituted 2.1 percent of total national CCh emissions, approximately the same proportion as
17 in 1990.
18 • Nitrous oxide emissions from stationary combustion increased by 11.5 MMT CC>2 Eq. (96.4 percent) from
19 1990 through 2014. Nitrous oxide emissions from this source increased primarily as a result of an increase
20 in the number of coal fluidized bed boilers in the electric power sector.
21 • Nitrous oxide emissions from mobile combustion decreased 24.9 MMT CCh Eq. (60.4 percent) from 1990
22 through 2014. Nitrous oxide emissions from this source decreased primarily as a result of N2O national
23 emissions control standards and emissions control technologies for on-road vehicles.
24 • Carbon dioxide emissions from incineration of waste (9.4 MMT CC>2 Eq. in 2014) increased by 1.4 MMT
25 CO2 Eq. (18.2 percent) from 1990 through 2014, as the volume of plastics and other fossil carbon-
26 containing materials in municipal solid waste grew.
27 The increase in CC>2 emissions from fossil fuel combustion in 2014 was a result of multiple factors including: (1)
28 colder winter conditions in the first quarter of 2014 resulting in an increased demand for heating fuel in the
29 residential and commercial sectors; (2) an increase in industrial production across multiple sectors resulting in slight
30 increases in industrial sector emissions;1 and (3) an increase in transportation emissions resulting from an increase in
31 vehicle miles traveled (VMT) and fuel use across on-road transportation modes.
32
Industrial Processes and Product Use
33 The Industrial Processes and Product Use (IPPU) chapter includes greenhouse gas emissions occurring from
34 industrial processes and from the use of greenhouse gases in products.
35 Greenhouse gas emissions are produced as the by-products of many non-energy-related industrial activities. For
36 example, industrial processes can chemically transform raw materials, which often release waste gases such as CO2,
37 CH4, and N2O. These processes include iron and steel production and metallurgical coke production, cement
38 production, ammonia production, urea consumption, lime production, other process uses of carbonates (e.g., flux
39 stone, flue gas desulfurization, and glass manufacturing), soda ash production and consumption, titanium dioxide
40 production, phosphoric acid production, ferroalloy production, CO2 consumption, silicon carbide production and
41 consumption, aluminum production, petrochemical production, nitric acid production, adipic acid production, lead
42 production, zinc production, and N2O from product uses (see Figure 2-9). Industrial processes also release HFCs,
43 PFCs, SF6, and NF3. In addition to their use as ODS substitutes, HFCs, PFCs, SF6, NF3, and other fluorinated
44 compounds are employed and emitted by a number of other industrial sources in the United States. These industries
45 include aluminum production, HCFC-22 production, semiconductor manufacture, electric power transmission and
1 Further details on industrial sector combustion emissions are provided by EPA's GHGRP
(http://ghgdata.epa.gov/ghgp/main.do).
Trends 2-15
-------
1
2
3
4
distribution, and magnesium metal production and processing. Table 2-6 presents greenhouse gas emissions from
industrial processes by source category.
Figure 2-9: 2014 Industrial Processes and Product Use Chapter Greenhouse Gas Sources
(MMT COz Eq.)
Substitution of Ozone Depleting Substances
Iron and Steel Prod. & Metallurgical Coke Prod.
Cement Production
Petrochemical Production
Lime Production
Other Process Uses of Carbonates
Nitric Acid Production
Ammonia Production
Adipic Acid Production
Electrical Transmission and Distribution
Carbon Dioxide Consumption
N2O from Product Uses
Semiconductor Manufacture
HCFC-22 Production
Urea Consumption for Non-Agricultural Purposes
Soda Ash Production and Consumption
Ferroalloy Production
Titanium Dioxide Production
Magnesium Production and Processing
Glass Production
Phosphoric Acid Production
Zinc Production
Lead Production
Silicon Carbide Production and Consumption
5
6
171
Industrial Processes and Product Use as a Portion
of all Emissions
5.7%
r
'
< 0.5
10
20
30 40
MMT CO2 Eq.
50
,50
70
Table 2-6: Emissions from Industrial Processes and Product Use (MMT COz Eq.)
Gas/Source
1990
2005
2010 2011 2012 2013 2014
C02 207.1 190.3
Iron and Steel Production & Metallurgical Coke
Production 99.vB 66.5
Iron and Steel Production 97.2 64.5
Metallurgical Coke Production 2.5m 2.0
Cement Production 33.3 45.9
Petrochemical Production 21.6 27.4
Lime Production 11.?B 14.6
Other Process Uses of Carbonates 4.9B 6.3
Ammonia Production 13.0B 9.2
Carbon Dioxide Consumption 1.51 1.4
Urea Consumption for Non-Agricultural
Purposes 3.sB 3.7
Aluminum Production 6.sB 4.1
Soda Ash Production and Consumption 2.8 B 3.0
Ferroalloy Production 2.2B 1.4
Titanium Dioxide Production 1-^B 1
Glass Production 1.5 B 1.9
Phosphoric Acid Production 1.51 1.3
Zinc Production 0.6 B 1.0
168.8 172.9 169.5 171.7 178.6
55.7
53.6
2.1
31.3
27.2
13.4
9.6
9.2
4.4
59.9
58.5
1.4
32.0
26.3
14.0
9.3
9.3
4.1
54.2
53.7
0.5
35.1
26.5
13.7
8.0
9.4
4.0
52.2
50.4
1.8
36.1
26.4
14.0
10.4
10.0
4.2
55.4
53.4
1.9
38.8
26.5
14.1
12.1
9.4
4.5
4.7
2.7
2.7
1.7
1.8
1.5
1.1
1.2
4.0
3.3
2.7
1.7
1.7
1.3
1.2
1.3
4.4
3.4
2.8
1.9
1.5
1.2
1.1
1.5
4.2
3.3
2.8
1.8
1.7
1.3
1.1
1.4
4.0
3.3
2.8
1.9
1.8
1.3
1.1
1.0
2-16 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Lead Production 0.51 0.6
Silicon Carbide Production and Consumption 0.41 0.2
Magnesium Production and Processing +B +1
CH4 0.31 0.1
Petrochemical Production 0.2B 0.1
Ferroalloy Production +B +|
Silicon Carbide Production and Consumption +B +
Iron and Steel Production & Metallurgical Coke
Production +B +
Iron and Steel Production +1 +
Metallurgical Coke Production O.oU 0.0
N2O 31.61 22.8
Nitric Acid Production 12.1 11.3
Adipic Acid Production 15.2B 7.1
N2O from Product Uses 4.21 4.2
Semiconductor Manufacturing +B 0.1
HFCs 46.6 133.3
Substitution of Ozone Depleting Substances3 0.31 113.0
HCFC-22 Production 46.1 20.0
Semiconductor Manufacturing 0.2B 0.2
Magnesium Production and Processing O.oB 0.0
PFCs 24.3 6.6
Aluminum Production 21.5 3.4
Semiconductor Manufacturing 2.sB 3.2
SF6 31.1 14.0
Electrical Transmission and Distribution 25.4 10.6
Magnesium Production and Processing ^.2B 2.7
Semiconductor Manufacturing 0.5 B 0.7
NF3 +B 0.5
Semiconductor Manufacturing + 0.5
0.5
0.2
+
0.1
0.5
0.2
+
0.1
0.5
0.2
+
0.1
0.1
0.5
0.2
+
0.1
0.1
0.5
0.2
+
0.2
0.1
0.0
20.1
11.5
4.2
4.2
0.1
161.7
153.5
8.0
0.2
+
4.4
1.9
2.6
9.5
7.0
2.1
0.4
0.5
0.5
0.0
25.5
10.9
10.2
4.2
0.2
166.1
157.1
8.8
0.2
+
6.9
3.5
3.4
10.0
6.8
2.8
0.4
0.7
0.7
0.0
20.4
10.5
5.5
4.2
0.2
167.1
161.4
5.5
0.2
+
6.0
2.9
3.0
7.7
5.7
1.6
0.4
0.6
0.6
0.0
19.1
10.7
4.0
4.2
0.2
169.6
165.3
4.1
0.2
0.1
5.8
3.0
2.9
6.9
5.1
1.4
0.4
0.6
0.6
0.0
20.8
10.9
5.4
4.2
0.2
175.8
171.4
4.1
0.2
0.1
5.8
3.0
2.9
6.9
5.1
1.4
0.4
0.6
0.6
Total
340.9
367.6
365.2 382.2 371.4 373.8 388.6
+ Does not exceed 0.05 MMT CO2 Eq.
a Small amounts of PFC emissions also result from this source.
Note: Totals may not sum due to independent rounding.
1 Overall, emissions from the IPPU sector increased by 14.0 percent from 1990 to 2014. Significant trends in
2 emissions from IPPU source categories over the twenty five-year period from 1990 through 2014 included the
3 following:
4 • HFC emissions from ODS substitutes have been increasing from small amounts in 1990 to 171.4 MMT
5 CO2 Eq. in 2014. This increase was in large part the result of efforts to phase out CFCs and other ODSs in
6 the United States. In the short term, this trend is expected to continue, and will likely continue over the
7 next decade as HCFCs, which are interim substitutes in many applications, are themselves phased-out
8 under the provisions of the Copenhagen Amendments to the Montreal Protocol.
9 • Combined CC>2 and CH4 emissions from iron and steel production and metallurgical coke production
10 decreased by 6.0 percent to 55.4 MMT CO2 Eq. from 2013 to 2014, and have declined overall by 44.3
11 MMT CO2 Eq. (44.5 percent) from 1990 through 2014, due to restructuring of the industry, technological
12 improvements, and increased scrap steel utilization.
13 • Carbon dioxide emissions from ammonia production (9.4 MMT CO2 Eq. in 2014) decreased by 3.6 MMT
14 CO2 Eq. (27.7 percent) since 1990. Ammonia production relies on natural gas as both a feedstock and a
15 fuel, and as such, market fluctuations and volatility in natural gas prices affect the production of ammonia.
16 • Urea consumption for non-agricultural purposes (4.0 MMT CO2 Eq. in 2014) increased by 0.2 MMT CO2
17 Eq. (5.9 percent) since 1990. From 1990 to 2007, emissions increased by 30 percent to a peak of 4.9 MMT
18 CO2 Eq., before decreasing by 19 percent to 2014 levels.
Trends 2-17
-------
1
2
3
4
5
6
7
10
11
12
13
14
15
16
17
18
19
20
In 2014, N2O emissions from product uses constituted 1.0 percent of U.S. N2O emissions. From 1990 to
2014, emissions from this source category decreased by 0.4 percent, though slight increases occurred in
intermediate years.
Nitrous oxide emissions from adipic acid production were 5.4 MMT CC>2 Eq. in 2014, and have decreased
significantly since 1990 due to both the widespread installation of pollution control measures in the late
1990s and plant idling in the late 2000s. Emissions from adipic acid production have decreased by 64.2
percent since 1990 and by 67.8 percent since a peak in 1995.
PFC emissions from aluminum production decreased by 86.2 percent (18.5 MMT CO2 Eq.) from 1990 to
2014, due to both industry emission reduction efforts and lower domestic aluminum production.
Agriculture
Agricultural activities contribute directly to emissions of greenhouse gases through a variety of processes, including
the following source categories: enteric fermentation in domestic livestock, livestock manure management, rice
cultivation, agricultural soil management, and field burning of agricultural residues.
In 2014, agricultural activities were responsible for emissions of 574.1 MMT CCh Eq., or 8.4 percent of total U.S.
greenhouse gas emissions. Methane and nitrous oxide were the primary greenhouse gases emitted by agricultural
activities. Methane emissions from enteric fermentation and manure management represented about 23.2 percent
and 8.6 percent of total CH4 emissions from anthropogenic activities, respectively, in 2014. Agricultural soil
management activities, such as fertilizer use and other cropping practices, were the largest source of U.S. N2O
emissions in 2014, accounting for 77.4 percent.
Figure 2-10: 2014 Agriculture Chapter Greenhouse Gas Sources (MMT COz Eq.)
Agricultural Soil Management
Enteric Fermentation
Manure Management
Rice Cultivation
Field Burning of Agricultural Residues
319
Agriculture as a Portion of all Emissions
8.4%
< 0.5
50
21
100
MMT CO2 Eq.
150
2-18 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Table 2-7: Emissions from Agriculture (MMT COz Eq.)
Gas/Source
CH4
Enteric Fermentation
Manure Management
Rice Cultivation
Field Burning of Agricultural
Residues
N20
Agricultural Soil Management
Manure Management
Field Burning of Agricultural
Residues
Total
1990
212.8
164.2 1
37.2
:;
317.0
302.9 1
14.0
0.1
529.8
2005
239.6
168.9 I
56.3 1
14.21
0.2
313.3 1
296.7 1
16. 5 1
0.1
552.9
2010
244.7
171.3
60.9
12.2
0.3
337.7
320.4
17.2
0.1
582.3
2011
242.9
168.9
61.5
12.2
0.3
340.4
322.9
17.4
0.1
583.3
2012
242.9
166.7
63.7
12.2
0.3
340.5
322.9
17.5
0.1
583.4
2013
239.3
165.5
61.4
12.2
0.3
336.0
318.4
17.5
0.1
575.4
2014
238.0
164.3
61.2
12.2
0.3
336.1
318.5
17.5
0.1
574.1
Note: Totals may not sum due to independent rounding.
2 Some significant trends in U.S. emissions from Agriculture source categories include the following:
3 • Agricultural soils produced approximately 77.4 percent of N2O emissions in the United States in 2014.
4 Estimated emissions from this source in 2014 were 318.5 MMT CO2 Eq. Annual N2O emissions from
5 agricultural soils fluctuated between 1990 and 2014, although overall emissions were 5.1 percent higher in
6 2014 than in 1990. Year-to-year fluctuations are largely a reflection of annual variation in weather
7 patterns, synthetic fertilizer use, and crop production.
8 • Enteric fermentation is the second largest anthropogenic source of CH4 emissions in the United States. In
9 2014, enteric fermentation CH4 emissions were 164.3 MMT CO2 Eq. (23.2 percent of total CH4 emissions),
10 which represents an increase of 0.1 MMT CO2 Eq. (0.1 percent) since 1990. This increase in emissions
11 from 1990 to 2014 in enteric generally follows the increasing trends in cattle populations. From 1990 to
12 1995 emissions increased and then generally decreased from 1996 to 2004, mainly due to fluctuations in
13 beef cattle populations and increased digestibility of feed for feedlot cattle. Emissions increased from 2005
14 to 2007, as both dairy and beef populations underwent increases and the literature for dairy cow diets
15 indicated a trend toward a decrease in feed digestibility for those years. Emissions decreased again from
16 2008 to 2014 as beef cattle populations again decreased.
17 • Overall, emissions from manure management increased 53.8 percent between 1990 and 2014. This
18 encompassed an increase of 64.7 percent for CH4, from 37.2 MMT CO2 Eq. in 1990 to 61.2 MMT CO2 Eq.
19 in 2014; and an increase of 24.9 percent for N2O, from 14.0 MMT CO2 Eq. in 1990 to 17.5 MMT CO2 Eq.
20 in 2014. The majority of the increase observed in CH4 resulted from swine and dairy cow manure, where
21 emissions increased 44 and 118 percent, respectively, from 1990 to 2014. From 2013 to 2014, there was a
22 0.3 percent decrease in total CH4 emissions from manure management, mainly due to minor shifts in the
23 animal populations and the resultant effects on manure management system allocations.
24 Land Use, Land-Use Change, and Forestry
25 When humans alter the terrestrial biosphere through land use, changes in land use, and land management practices,
26 they also alter the background carbon fluxes between biomass, soils, and the atmosphere. Forest management
27 practices, tree planting in urban areas, the management of agricultural soils, and the landfilling of yard trimmings
28 and food scraps have resulted in a net removal of CO2 (sequestration of C) in the United States. Forests (including
29 vegetation, soils, and harvested wood) accounted for approximately 85 percent of total 2014 CO2 removals, urban
30 trees accounted for 13 percent, mineral and organic soil carbon stock changes accounted for less than 0.5 percent,
31 and landfilled yard trimmings and food scraps accounted for 1.7 percent of total CO2 removals in 2014. The net
32 forest sequestration is a result of net forest growth, increasing forest area, and a net accumulation of carbon stocks in
33 harvested wood pools. The net sequestration in urban forests is a result of net tree growth and increased urban forest
34 size. In agricultural soils, mineral and organic soils sequester approximately 1.8 times as much C as is emitted from
35 these soils through liming and urea fertilization. The mineral soil C sequestration is largely due to the conversion of
Trends 2-19
-------
1 cropland to hay production fields, the limited use of bare-summer fallow areas in semi-arid areas, and an increase in
2 the adoption of conservation tillage practices. The landfilled yard trimmings and food scraps net sequestration is
3 due to the long-term accumulation of yard trimming and food scraps carbon in landfills.
4 Land use, land-use change, and forestry activities in 2014 resulted in a C sequestration (i.e., net CO2 removals) of
5 685.8 MMT CO2 Eq. (Table 2-3).2 This represents an offset of approximately 10.0 percent of total (i.e., gross)
6 greenhouse gas emissions in 2014. Emissions from land use, land-use change, and forestry activities in 2014
7 represent 0.4 percent of total greenhouse gas emissions.3 Between 1990 and 2014, total land use, land-use change,
8 and forestry C sequestration decreased by 2.6 percent, primarily due to a decrease in the rate of net C accumulation
9 in agricultural soil carbon stocks.
10 Carbon dioxide removals are presented in Table 2-8 along with CO2, CH4, and N2O emissions for Land Use, Land-
11 Use Change, and Forestry source categories.4 Liming and urea fertilization resulted in CO2 emissions of 8.7 MMT
12 CO2 Eq. in 2014, an increase of about 22.2 percent relative to 1990. Lands undergoing peat extraction (i.e.,
13 Peatlands Remaining Peatlands) resulted in CO2 emissions of 0.8 MMT CO2Eq. and CH4 and N2O emissions of
14 less than 0.05 MMT CO2 Eq. each. N2O emissions from the application of synthetic fertilizers to forest soils have
15 increased from 0.1 MMT CO2 Eq. in 1990 to 0.5 MMT CO2 Eq. in 2014. Settlement soils in 2014 resulted in N2O
16 emissions of 2.4 MMT CO2Eq., a 78.4 percent increase relative to 1990. Emissions from forest fires in 2014
17 resulted in CH4 emissions of 7.3 MMT CO2 and inN2O emissions of 4.8 MMT CO2 (see Table 2-8).
18 Table 2-8: Emissions and Removals (Flux) from Land Use, Land-Use Change, and Forestry
19 (MMT COz Eq.)
Gas/Land-Use Category
1990
2005
2010
2011
2012
2013
2014
Net CO2 Flux3 (704.2)
Forest Land Remaining Forest Landb (576.0)
Cropland Remaining Cropland (43.2)
Land Converted to Cropland3 22.8
Grassland Remaining Grassland* (12.9)
Land Converted to Grassland (8.5)
Settlements Remaining Settlements0 (60.4)
Other: Landfilled Yard Lrimmings and
Food Scraps (26.0)
C02 8.1
Cropland Remaining Cropland: Liming 4.7
Cropland Remaining Cropland: Urea
Fertilization 2.4
Wetlands Remaining Wetlands:
Peatlands Remaining Peatlands 1.1
CH4 3.3
Forest Land Remaining Forest Land:
Forest Fires 3.3
Wetlands Remaining Wetlands:
Peatlands Remaining Peatlands +
N20 3.6
Forest Land Remaining Forest Land:
Forest Fires 2.2
Forest Land Remaining Forest Land:
Forest Soils'1 0.1
Settlements Remaining Settlements:
Settlement Soils6 1.4
Wetlands Remaining Wetlands: +
3.5
1.1
9.9
9.9
3.8
1.0
3.3
3.3
9.3 I 5.0
6.5 | 2.2
0.5 I 0.5
2.3 I 2.4
(683.6)
(578.1)
(20.0)
14.2
11.3
(11.0)
(87.3)
(12.7)
8.9
3.9
4.1
0.9
6.6
6.6
7.3
4.4
0.5
2.5
(680.8)
(576.7)
(18.7)
14.5
11.7
(10.9)
(88.4)
(12.2)
11.0
6.0
4.2
0.8
11.1
11.1
10.3
7.3
0.5
2.5
(682.4)
(580.1)
(16.8)
14.8
11.9
(10.9)
(89.5)
(11.7)
9.0
3.9
4.3
7.3
(685.8)
(583.4)
(16.0)
14.7
11.9
(10.9)
(90.6)
(11.6)
9.5
4.1
4.5
0.8
7.4
7.3
7.7 7.7
4.8 4.8
0.5 0.5
2.4 2.4
2 Net flux from LULUCF includes the positive C sequestration reported for Forest Land Remaining Forest Land, Land
Converted to Forest Land, Cropland Remaining Cropland, Land Converted to Grassland, Settlements Remaining Settlements,
and Other Land plus the loss in C sequestration reported for Land Converted to Cropland and Grassland Remaining Grassland.
3 LULUCF emissions include the CCh, CELi, andN2O emissions reported for Forest Fires, Forest Soils, Liming, Urea
Fertilization, Settlement Soils, and Peatlands Remaining Peatlands.
4 Estimates from Land Converted to Forest Land are currently under development.
2-20 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Peatlands Remaining Peatlands
1
LULUCF Emissions' 15.0 28.2 17.8 22.9 32.3 24.1 24.6
LULUCF Total Net Fluxa (704.2) (636.1) (683.2) (683.6) (680.8) (682.4) (685.8)
LULUCF Sector TotaP (689.1) (607.9) (665.3) (660.7) (648.5) (658.3) (661.3)
Note: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
+ Does not exceed 0.05 MMT CO2 Eq.
a Total net flux from LULUCF is only included in the Net Emissions total. Net flux from LULUCF includes the positive C
sequestration reported for Forest Land Remaining Forest Land, Land Converted to Forest Land, Cropland Remaining
Cropland, Land Converted to Grassland, Settlements Remaining Settlements, and Other Land plus the loss in C
sequestration reported for Land Converted to Cropland and Grassland Remaining Grassland.
b Includes the effects of net additions to stocks of carbon stored in forest ecosystem pools and harvested wood products.
c Estimates include C stock changes on both Settlements Remaining Settlements and Land Converted to Settlements.
d Estimates include emissions from N fertilizer additions on both Forest Land Remaining Forest Land, and Land Converted
to Forest Land, but not from land-use conversion.
e Estimates include emissions from N fertilizer additions on both Settlements Remaining Settlements, and Land Converted to
Settlements, but not from land-use conversion.
f LULUCF emissions include the CCh, CELi, andN2O emissions reported for Forest Fires, Forest Soils, Liming, Urea
Fertilization, Settlement Soils, and Peatlands Remaining Peatlands.
g The LULUCF Sector Total is the sum of positive emissions (i.e., sources) of greenhouse gases to the atmosphere plus
removals of CCh (i.e., sinks or negative emissions) from the atmosphere.
2 Other significant trends from 1990 to 2014 in emissions from land use, land-use change, and forestry source
3 categories include:
4 • Annual C sequestration by forest land (i.e., annual carbon stock accumulation in the five carbon pools) has
5 increased by approximately 1 percent. This is primarily due to increased forest management and the effects
6 of previous reforestation. The increase in intensive forest management resulted in higher growth rates and
7 higher biomass density. The tree planting and conservation efforts of the 1970s and 1980s continue to have
8 a significant impact on sequestration rates. Finally, the forested area in the United States increased over the
9 past twenty five years, although only at an average rate of 0.2 percent per year.
10 • Annual C sequestration by urban trees has increased by 50.0 percent over the period from 1990 to 2014.
11 This is primarily due to an increase in urbanized land area in the United States.
12 • Annual C sequestration in landfilled yard trimmings and food scraps has decreased by 55.4 percent since
13 1990. Food scrap generation has grown by 55 percent since 1990, and though the proportion of food scraps
14 discarded in landfills has decreased slightly from 82 percent in 1990 to 76 percent in 2014, the tonnage
15 disposed in landfills has increased considerably (by 45 percent). Overall, the decrease in the landfill
16 disposal rate of yard trimmings has more than compensated for the increase in food scrap disposal in
17 landfills.
is Waste
19 Waste management and treatment activities are sources of greenhouse gas emissions (see Figure 2-11). In 2014,
20 landfills were the largest source of U.S. anthropogenic CH4 emissions, accounting for 25.7 percent of total U.S. CH4
21 emissions.5 Additionally, wastewater treatment accounts for 9.7 percent of Waste emissions, 2.1 percent of U.S.
22 CH4 emissions, and 1.2 percent of N2O emissions. Emissions of CEU and N2O from composting grew from 1990 to
23 2014, and resulted in emissions of 3.9 MMT CO2 Eq. in 2014. A summary of greenhouse gas emissions from the
24 Waste chapter is presented in Table 2-9.
25
Landfills also store carbon, due to incomplete degradation of organic materials such as wood products and yard trimmings, as
described in the Land Use, Land-Use Change, and Forestry chapter.
Trends 2-21
-------
Figure 2-11: 2014 Waste Chapter Greenhouse Gas Sources (MMT COz Eq.)
4
5
Landfills
Wastewater Treatment
Composting
Waste as a Portion of all Emissions
182
0 20 40 60 80 100 120
MMT CO2 Eq.
Overall, in 2014, waste activities generated emissions of 205.6 MMT CC>2 Eq., or 3.0 percent of total U.S.
greenhouse gas emissions.
Table 2-9: Emissions from Waste (MMT COz Eq.)
Gas/Source
CH4
Landfills
Wastewater Treatment
Composting
N20
Wastewater Treatment
Composting
Total
1990
200.4
184.4
15.7
0.4
3.7
3.4
0.3
204.1
2005
205.1
187.31
15.9
1.9l
6.0
4.3
1.7
211.1
2010
193.6
176.3
15.5
1.8
6.4
4.7
1.6
200.0
2011
194.0
176.9
15.3
1.9
6.5
4.8
1.7
200.5
2012
190.6
173.5
15.2
1.9
6.6
4.9
1.7
197.2
2013
193.7
176.7
15.0
2.0
6.7
4.9
1.8
200.5
2014
198.9
181.8
15.0
2.1
6.8
4.9
1.8
205.6
Note: Totals may not sum due to independent rounding.
7 Some significant trends in U.S. emissions from waste source categories include the following:
8 • From 1990 to 2014, net CEU emissions from landfills decreased by 2.6 MMT CO2 Eq. (1.4 percent), with
9 small increases occurring in interim years. This slight downward trend in overall emissions is the result of
10 increases in the amount of landfill gas collected and combusted as well as reductions in the amount of
11 decomposable materials (i.e., paper and paperboard, food scraps, and yard trimmings) discarded in MSW
12 landfills over the time series,6 which has more than offset the additional CH4 emissions resulting from an
13 increase in the amount of municipal solid waste landfilled.
14 • Combined CH4 and N2O emissions from composting have generally increased since 1990, from 0.7 MMT
15 CO2 Eq. to 3.9 MMT CO2 Eq. in 2014, which represents slightly more than a five-fold increase over the
16 time series. The growth in composting since the 1990s is attributable to primarily two factors: (1) steady
17 growth in population and residential housing, and (2) the enactment of legislation by state and local
18 governments that discouraged the disposal of yard trimmings in landfills.
19 • From 1990 to 2014, CEU and N2O emissions from wastewater treatment decreased by 0.6 MMT CO2 Eq.
20 (4.0 percent) and increased by 1.6 MMT €62 Eq. (46.5 percent), respectively. Methane emissions from
The CO2 produced from combusted landfill CH4 at landfills is not counted in national inventories as it is considered part of the
natural C cycle of decomposition.
2-22 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 domestic wastewater treatment have decreased since 1999 due to decreasing percentages of wastewater
2 being treated in anaerobic systems, including reduced use of on-site septic systems and central anaerobic
3 treatment systems. Nitrous oxide emissions from wastewater treatment processes gradually increased
4 across the time series as a result of increasing U.S. population and protein consumption.
5 2.2 Emissions by Economic Sector
6 Throughout this report, emission estimates are grouped into five sectors (i.e., chapters) defined by the IPCC and
7 detailed above: Energy; Industrial Processes and Product Use; Agriculture; Land Use, Land-Use Change, and
8 Forestry; and Waste. While it is important to use this characterization for consistency with UNFCCC reporting
9 guidelines, it is also useful to allocate emissions into more commonly used sectoral categories. This section reports
10 emissions by the following U.S. economic sectors: residential, commercial, industry, transportation, electricity
11 generation, and agriculture, as well as U.S. Territories.
12 Using this categorization, emissions from electricity generation accounted for the largest portion (30 percent) of
13 U.S. greenhouse gas emissions in 2014. Transportation activities, in aggregate, accounted for the second largest
14 portion (26 percent). Emissions from industry accounted for about 20 percent of U.S. greenhouse gas emissions in
15 2014. In contrast to electricity generation and transportation, emissions from industry have in general declined over
16 the past decade. The long-term decline in these emissions has been due to structural changes in the U.S. economy
17 (i.e., shifts from a manufacturing-based to a service-based economy), fuel switching, and efficiency improvements.
18 The remaining 23 percent of U.S. greenhouse gas emissions were contributed by the residential, agriculture, and
19 commercial sectors, plus emissions from U.S. Territories. The residential sector accounted for 6 percent, and
20 primarily consisted of CC>2 emissions from fossil fuel combustion. Activities related to agriculture accounted for
21 roughly 9 percent of U.S. emissions; unlike other economic sectors, agricultural sector emissions were dominated by
22 N2O emissions from agricultural soil management and CEU emissions from enteric fermentation, rather than CO2
23 from fossil fuel combustion. The commercial sector accounted for roughly 7 percent of emissions, while U.S.
24 Territories accounted for less than 1 percent. Carbon dioxide was also emitted and sequestered (in the form of C) by
25 a variety of activities related to forest management practices, tree planting in urban areas, the management of
26 agricultural soils, and landfilling of yard trimmings.
27 Table 2-10 presents a detailed breakdown of emissions from each of these economic sectors by source category, as
28 they are defined in this report. Figure 2-12 shows the trend in emissions by sector from 1990 to 2014.
29
Trends 2-23
-------
Figure 2-12: Emissions Allocated to Economic Sectors (MMT COz Eq.)
2,000 -
B i'500 -
o
u
1-
| 1,000
500 -
_^ "
^~*^
^—*'*~*i^~*\
^S~~- — V/"X>. Electric
_^n*ll*~ ^"^ >w Power Industry
—"^ > —
^
__ — *""^ Transportation
""" - ~ %w Industry
N-X *"
^ Res.dentialfBlue)
o TH rsi m <
-------
N2O from Product Uses
Semiconductor Manufacture
HCFC-22 Production
Urea Consumption for Non-Agricultural
Purposes
Stationary Combustion
Soda Ash Production and Consumption
Ferroalloy Production
Titanium Dioxide Production
Magnesium Production and Processing
Mobile Combustion
Glass Production
Phosphoric Acid Production
Zinc Production
Lead Production
Silicon Carbide Production and
Consumption
Agriculture
N2O from Agricultural Soil Management
Enteric Fermentation
Manure Management
CO2 from Fossil Fuel Combustion
Rice Cultivation
CELi and N2O from Forest Fires
Urea Fertilization
Liming
CO2, CH4 and N2O from Managed
Peatlands
Mobile Combustion
N2O from Forest Soils
Field Burning of Agricultural Residues
Stationary Combustion
Commercial
CO2 from Fossil Fuel Combustion
Landfills
Substitution of Ozone Depleting
Substances
Wastewater Treatment
Human Sewage
Composting
Stationary Combustion
Residential
CO2 from Fossil Fuel Combustion
Substitution of Ozone Depleting
Substances
Stationary Combustion
Settlement Soil Fertilization
U.S. Territories
CO2 from Fossil Fuel Combustion
Non-Energy Use of Fuels
Stationary Combustion
Total Emissions
LULUCF Total Net Flux"
Forest Land Remaining Forest Land6
Land Converted to Forest Land
Cropland Remaining Cropland
Land Converted to Cropland"1
Grassland Remaining Grassland4
4.2
3.6
46.1 1
3.8
4.9
2.8
2.2
1.2
5.2
0.9
1.5
1.5
0.6
0.5
0.4l
574.91
302.9!
164.21
51. ll
31.0
11.3
5.4
2.4
4.7
1
0.3
0.1
0.3
+
422.91
217.4!
1X4.4!
+
15.7
3.4
0.7
1 4
346.3
338.3!
"
6.3
1.4
33.7
27.9!
5.7
0.1
6,380.8
(704.2)
(576.0)
(43.2)1
22.8
(12.9)|
4.2
4.7
20.0
3.7
4.6
3.0
1.9
1.3
1.0
0.6
47.4
14.2
16.5
::
0.5
0.5
0.3
17.6
15.9
4.3
3.5
1 4
372.8
357.8|
7.7
4.9
2.3
58.2
49.9
8.1
0.2
7,428.8
(636.1)
(532.4)
(16.5)1
14.61
2.9
4.2
3.8
8.0
4.7
3.9
2.7
1.7
1.8
2.1
1.4
1.5
1.1
1.2
0.5
0.2
646.6
320.4
171.3
78.1
48.2
12.2
5.4
3.8
4.8
1.0
0.5
0.5
0.4
+
459.9
220.1
176.3
38.5
15.5
4.7
3.5
1.4
363.6
334.6
21.8
4.8
2.4
45.3
41.4
3.7
0.2
7,010.5
(683.2)
(585.0)
(4.7)
15.6
2.6
4.2
4.9
8.8
4.0
3.9
2.7
1.7
1.7
2.8
1.4
1.3
1.2
1.3
0.5
0.2
654.2
322.9
168.9
78.9
49.9
12.2
11.0
4.1
3.9
0.9
0.5
0.5
0.4
+
464.7
220.7
176.9
42.1
15.3
4.8
3.5
1.4
360.1
326.8
25.9
4.9
2.5
45.4
41.5
3.7
0.2
6,887.8
(683.6)
(578.1)
(20.0)
14.2
11.3
4.2
4.5
5.5
4.4
3.9
2.8
1.9
1.5
1.7
1.4
1.2
1.1
1.5
0.5
0.2
665.3
322.9
166.7
81.2
51.4
12.2
18.3
4.2
6.0
0.8
0.6
0.5
0.4
+
440.0
196.7
173.5
44.9
15.2
4.9
3.7
1.2
320.9
282.5
31.4
4.5
2.5
47.6
43.6
3.8
0.2
6,665.7
(680.8)
(576.7)
(18.7)
14.5
11.7
4.2
4.2
4.1
4.2
3.9
2.8
1.8
1.7
1.5
1.5
1.3
1.1
1.4
0.5
0.2
648.0
318.4
165.5
78.9
50.4
12.2
12.2
4.3
3.9
0.8
0.6
0.5
0.4
+
470.2
221.0
176.7
47.4
15.0
4.9
3.9
1.3
375.1
329.7
37.0
5.9
2.4
47.5
43.5
3.8
0.2
6,811.2
(682.4)
(580.1)
(16.8)
14.8
11.9
4.2
4.2
4.1
4.0
3.9
2.8
1.9
1.8
1.5
1.5
1.3
1.1
1.0
0.5
0.2
648.0
318.5
164.3
78.7
51.2
12.2
12.2
4.5
4.1
0.8
0.6
0.5
0.4
+
487.8
231.6
181.8
49.2
15.0
4.9
3.9
1.4
396.1
345.1
42.6
6.0
2.4
44.7
41.0
3.5
0.2
6,872.6
(685.8)
(583.4)
(16.0)
14.7
11.9
0.1%
0.1%
0.1%
0.1%
0.1%
+
+
+
+
+
+
+
+
9.4%
4.6%
2.4%
1.1%
0.7%
0.2%
0.2%
0.1%
0.1%
+
+
+
+
7.1%
3.4%
2.6%
0.7%
0.2%
0.1%
0.1%
+
5.8%
5.0%
0.6%
0.1%
+
0.7%
0.6%
0.1%
+
100.0%
(10.0%)
(8.5%)
(0.2%)
0.2%
0.2%
Trends 2-25
-------
Land Converted to Grassland
Settlements Remaining Settlements
Other: Landfilled Yard Trimmings and
Food Scraps
Net Emissions (Sources and Sinks)
(8.5)1
(60.4)1
(26.0)
5,676.6
(12.8)
(80.5)
1 (H.4)
6,792.6
(12.3)
(86.1)
(13.2)
6,327.3
(11.0)
(87.3)
(12.7)
6,204.2
(10.9)
(88.4)
(12.2)
5,984.9
(10.9)
(89.5)
(11.7)
6,128.8
(10.9)
(90.6)
(11.6)
6,186.8
(0.2%)
(1.3%)
(0.2%)
90.0%
Note: Includes all emissions of CCh, CELi, N2O, HFCs, PFCs, SFe, and NFs. Parentheses indicate negative values or sequestration. Totals
may not sum due to independent rounding.
+ Does not exceed 0.05 MMT CO2 Eq. or 0.05 percent.
a Percent of total emissions for year 2014.
b The values in this table for Natural Gas Systems are presented from the previous Inventory and do not reflect updates to emission estimates
for this category. See Section 3.7, Natural Gas Systems of the Energy chapter for more information. Gray highlighting was added on 2/24 for
clarification.
c The values in this table for Petroleum Systems are presented from the previous Inventory and do not reflect updates to emission estimates for
this category. See Section 3.6, Petroleum Systems of the Energy chapter for more information. Gray highlighting was added on 2/24 for
clarification.
d Total net flux from LULUCF is only included in the Net Emissions total. Net flux from LULUCF includes the positive C sequestration
reported for Forest Land Remaining Forest Land, Land Converted to Forest Land, Cropland Remaining Cropland, Land Converted to
Grassland, Settlements Remaining Settlements, and Other Land plus the loss in C sequestration reported for Land Converted to Cropland
and Grassland Remaining Grassland. Refer to Table 2-8 for a breakout of emissions and removals for Land Use, Land-Use Change, and
Forestry by gas and source category.
e Includes the effects of net additions to stocks of carbon stored in forest ecosystem pools and harvested wood products.
i Emissions with Electricity Distributed to Economic Sectors
2 It can also be useful to view greenhouse gas emissions from economic sectors with emissions related to electricity
3 generation distributed into end-use categories (i.e., emissions from electricity generation are allocated to the
4 economic sectors in which the electricity is consumed). The generation, transmission, and distribution of electricity,
5 which is the largest economic sector in the United States, accounted for 30 percent of total U.S. greenhouse gas
6 emissions in 2014. Emissions increased by 12 percent since 1990, as electricity demand grew and fossil fuels
7 remained the dominant energy source for generation. Electricity generation-related emissions increased from 2013
8 to 2014 by 0.1 percent, primarily due to increased CC>2 emissions from fossil fuel combustion. Electricity sales to
9 the residential and commercial end-use sectors in 2014 increased approximately 0.9 percent and 1.1 percent,
10 respectively. The trend in the residential and commercial sectors can largely be attributed to colder more energy -
11 intensive winter conditions compared to 2013. Electricity sales to the industrial sector in 2014 increased by
12 approximately 1.2 percent. Overall, in 2014, the amount of electricity generated (in kWh) increased by 1.1 percent
13 from the previous year. Despite this increase in generation, CC>2 emissions from the electric power sector increased
14 by 0.1 percent as the consumption petroleum for electricity generation increased by 15.8 percent in 2014 and the
15 consumption of CCh-intensive coal and natural gas for electricity generation decreased by 0.1 and 0.2 percent,
16 respectively. Table 2-11 provides a detailed summary of emissions from electricity generation-related activities.
17 Table 2-11: Electricity Generation-Related Greenhouse Gas Emissions (MMT COz Eq.)
Gas/Fuel Type or Source
CO2
Fossil fuel combustion
Coal
Natural Gas
Petroleum
Geothermal
Incineration of Waste
Other Process Uses of
Carbonates
CH4
Stationary Sources (Elec
Gen)
Incineration of Waste
N2O
1990
1,831.2
1,820.8
1,547.6
175.3
97.5 1
0.4
8.0
2.5
0.3
,
7.8
I 2005
2,416.5
2,400.9 •
11,983.8
318.8
97.9M
0.4
12.5
3.2
0.5
1
2010
2,274.2
2,258.4
1,827.6
399.0
31.4
0.4
11.0
4.8
0.5
0.5
+
18.8
2011
2,172.9
2,157.7
1,722.7
408.8
25.8
0.4
10.5
4.7
0.4
0.4
+
17.9
2012
2,036.6
2,022.2
1,511.2
492.2
18.3
0.4
10.4
4.0
0.4
0.4
+
18.1
2013
2,052.8
2,038.1
1,571.3
444.0
22.4
0.4
9.4
5.2
0.4
0.4
+
19.4
2014
2,054.8
2,039.3
1,570.4
443.2
25.3
0.4
9.4
6.0
0.4
0.4
+
19.9
2-26 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Stationary Sources (Elec
Gen) 7.4
Incineration of Waste o 5
SF6 25.4
Electrical Transmission and
Distribution 25.4
II
10.6
18.5
0.3
7.0
7.0
17.6
0.3
6.8
17.8
0.3
5.7
5.7
19.1
0.3
5.1
5.1
19.6
0.3
5.1
5.1
Total
1,864.8
2.443.9
2,300.5 2,198.1 2,060.8 2,077.7 2,080.2
Note: Totals may not sum due to independent rounding.
a Includes only stationary combustion emissions related to the generation of electricity.
+ Does not exceed 0.05 MMT CO2 Eq.
1 To distribute electricity emissions among economic end-use sectors, emissions from the source categories assigned
2 to the electricity generation sector were allocated to the residential, commercial, industry, transportation, and
3 agriculture economic sectors according to each economic sector's share of retail sales of electricity consumption
4 (EIA 2015, Duffield 2006). These source categories include CCh from Fossil Fuel Combustion, CH4 and N2O from
5 Stationary Combustion, Incineration of Waste, Other Process Uses of Carbonates, and SF6 from Electrical
6 Transmission and Distribution Systems. Note that only 50 percent of the Other Process Uses of Carbonates
7 emissions were associated with electricity generation and distributed as described; the remainder of Other Process
8 Uses of Carbonates emissions were attributed to the industrial processes economic end-use sector.7
9 When emissions from electricity are distributed among these sectors, industrial activities account for the largest
10 share of total U.S. greenhouse gas emissions (28.2 percent), followed closely by emissions from transportation (26.5
11 percent). Emissions from the residential and commercial sectors also increase substantially when emissions from
12 electricity are included. In all sectors except agriculture, CO2 accounts for more than 78 percent of greenhouse gas
13 emissions, primarily from the combustion of fossil fuels.
14 Table 2-12 presents a detailed breakdown of emissions from each of these economic sectors, with emissions from
15 electricity generation distributed to them. Figure 2-13 shows the trend in these emissions by sector from 1990 to
16 2014.
17 Figure 2-13: Emissions with Electricity Distributed to Economic Sectors (MMT COz Eq.)
18
2,500
2,000
S 1'500
8
Z 1,000
500
Industry (Green)
Transportation
(Purple)
Commercial (Red)
Residential (Blue)
• Agriculture
^ >
Emissions were not distributed to U.S. Territories, since the electricity generation sector only includes emissions related to the
generation of electricity in the 50 states and the District of Columbia.
Trends 2-27
-------
1 Table 2-12: U.S. Greenhouse Gas Emissions by Economic Sector and Gas with Electricity-
2 Related Emissions Distributed (MMT COz Eq.) and Percent of Total in 2014
Sector/Gas
Industry
Direct Emissions
C02
CH4
N20
FIFCs, PFCs, SF6,
andNFs
Electricity-Related
C02
CH4
N20
SF6
Transportation
Direct Emissions
C02
CH4
N20
FIFCsb
Electricity-Related
CO2
CH4
N2O
SF6
Commercial
Direct Emissions
CO2
CH4
N2O
FIFCs
Electricity-Related
C02
CH4
N20
SF6
Residential
Direct Emissions
C02
CH4
N20
FIFCs
Electricity-Related
CO2
CH4
N2O
SF6
Agriculture
Direct Emissions
C02
CH4
N20
Electricity-Related
C02
CH4
N20
SF6
1990
2,228.9
1,586.9
1,158.7
316.6
35.4
76.3
642.0
638.3 1
-
0.8
1,554.4
1,551.3
1,505.6
5.4l
40.3
3.1
3.1
;
973.9
422.9
217.41
201.51
4.1
1
551.0
547.8 1
0,
2.3
0.7
953.6
346.3
338.31
5.2
2.4
0.3M
607.3
603.91
1"
0.8
636.3
574.9
39.2
216.21
319.5 1
61.31
61.0
03
0.1
2005
2,146.3
1,460.6
1,123.2
272.5
26.7
38.1
685.71
680.1 1
0.1
4.6
0.9
2,017.7
2,012.9
1,897.2
2.J
32.9
80.4
4.8
4.8
1
1,271.2
453.6
223.51
206.1 1
6.3
17.6
817.71
811 Q!
'
1,244.4
372.8
357.81
4.1
3.2
7.7
871.61
864. 5 1
1"
1.1
690.9
626.8
56.41
249.71
320.71
64. ll
63.5
0.4
0.1
2010
1,939.4
1,354.9
1,028.7
271.6
23.7
30.9
584.5
578.4
0.1
4.8
1.2
1,844.3
1,839.7
1,737.8
1.9
22.1
77.9
4.6
4.5
1,247.2
459.9
220.1
194.7
6.7
38.5
787.3
779.1
0.2
6.4
1.6
1,219.3
363.6
334.6
4.0
3.2
21.8
855.7
846.8
0.2
7.0
1.8
715.0
646.6
57.8
248.2
340.6
68.4
67.7
0.6
0.1
2011
1,924.9
1,353.9
1,027.4
260.6
29.1
36.8
571.0
565.0
0.1
4.7
1.2
1,815.7
1,811.3
1,716.6
1.8
20.9
72.0
4.3
4.3
1,216.6
464.7
220.7
195.0
6.8
42.1
751.9
744.0
0.1
6.1
1.6
1,165.6
360.1
326.8
4.0
3.3
25.9
805.5
797.1
0.2
6.6
1.7
719.6
654.2
58.8
249.7
345.6
65.4
64.7
0.5
0.1
2012
1,881.4
1,339.5
1,030.3
252.1
24.0
33.1
542.0
536.0
0.1
4.8
1.1
1,795.5
1,791.6
1,705.0
1.8
18.5
66.3
3.9
3.9
1,153.7
440.0
196.7
191.5
6.8
44.9
713.6
705.8
0.1
6.3
1.4
1,060.1
320.9
282.5
3.7
3.3
31.4
739.1
731.0
0.2
6.5
1.4
727.4
665.3
62.5
254.2
348.6
62.1
61.4
0.5
0.1
2013
1,936.4
1,392.1
1,081.2
255.2
22.7
32.9
544.2
537.7
0.1
5.1
1.4
1,804.6
1,800.5
1,721.8
1.7
16.6
60.5
4.1
4.0
1,188.4
470.2
221.0
194.8
7.0
47.4
718.1
709.5
0.1
6.7
1.8
1,124.2
375.1
329.7
5.0
3.4
37.0
749.2
740.1
0.2
7.0
1.9
710.1
648.0
59.4
246.9
341.7
62.1
61.3
0.6
0.2
2014
1,939.4
1,395.5
1,081.0
255.3
24.4
34.8
543.9
537.0
0.1
5.2
1.6
1,824.4
1,820.3
1,746.6
1.6
14.7
57.4
4.1
4.1
1,208.5
487.8
231.6
200.0
7.1
49.2
720.6
711.5
0.1
6.9
2.1
1,146.1
396.1
345.1
5.0
3.4
42.6
750.0
740.5
0.2
7.2
2.2
709.5
648.0
60.7
245.5
341.8
61.6
60.8
0.6
0.2
Percent3
28.2%
20.3%
15.7%
3.7%
0.4%
0.5%
7.9%
7.8%
+
0.1%
+
26.5%
26.5%
25.4%
0.2%
0.8%
0.1%
0.1%
17.6%
7.1%
3.4%
2.9%
0.1%
0.7%
10.5%
10.4%
0.1%
16.7%
5.8%
5.0%
0.1%
0.6%
10.9%
10.8%
0.0%
0.1%
+
10.3%
9.4%
0.9%
3.6%
5.0%
0.9%
0.9%
+
+
2-28 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
U.S. Territories 33.7 58.2 45.3 45.4 47.6 47.5 44.7 0.7%
Total 6,380.8 7,428.8 7,010.5 6,887.8 6,665.7 6,811.2 6,872.6 100.0%
Note: Emissions from electricity generation are allocated based on aggregate electricity consumption in each end-use
sector.
Totals may not sum due to independent rounding.
+ Does not exceed 0.05 MMT CCh Eq. or 0.05 percent.
a Percent of total emissions for year 2014.
b Includes primarily HFC-134a.
i Industry
2 The industry end-use sector includes CCh emissions from fossil fuel combustion from all manufacturing facilities, in
3 aggregate. This end-use sector also includes emissions that are produced as a byproduct of the non-energy-related
4 industrial process activities. The variety of activities producing these non-energy-related emissions includes
5 methane emissions from petroleum and natural gas systems, fugitive CH4 emissions from coal mining, by-product
6 CO2 emissions from cement manufacture, and HFC, PFC, SF6, and NF3 byproduct emissions from semiconductor
7 manufacture, to name a few. Since 1990, industrial sector emissions have declined. The decline has occurred both
8 in direct emissions and indirect emissions associated with electricity use. In theory, emissions from the industrial
9 end-use sector should be highly correlated with economic growth and industrial output, but heating of industrial
10 buildings and agricultural energy consumption are also affected by weather conditions. In addition, structural
11 changes within the U.S. economy that lead to shifts in industrial output away from energy-intensive manufacturing
12 products to less energy-intensive products (e.g., from steel to computer equipment) also have a significant effect on
13 industrial emissions.
14 Transportation
15 When electricity-related emissions are distributed to economic end-use sectors, transportation activities accounted
16 for 27 percent of U.S. greenhouse gas emissions in 2014. The largest sources of transportation greenhouse gases in
17 2014 were passenger cars (41.8 percent), freight trucks (22.8 percent), light-duty trucks, which include sport utility
18 vehicles, pickup trucks, and minivans (18.5 percent), commercial aircraft (6.4 percent), rail (2.7 percent), pipelines
19 (2.6 percent), and ships and boats (1.5 percent). These figures include direct CCh, CH4, and N2O emissions from
20 fossil fuel combustion used in transportation and emissions from non-energy use (i.e., lubricants) used in
21 transportation, as well as HFC emissions from mobile air conditioners and refrigerated transport allocated to these
22 vehicle types.
23 In terms of the overall trend, from 1990 to 2014, total transportation emissions rose by 17 percent due, in large part,
24 to increased demand for travel as fleetwide light-duty vehicle fuel economy was relatively stable (average new
25 vehicle fuel economy declined slowly from 1990 through 2004 and then increased more rapidly from 2005 through
26 2014). The number of vehicle miles traveled by light-duty motor vehicles (passenger cars and light-duty trucks)
27 increased 37 percent from 1990 to 2014, as a result of a confluence of factors including population growth,
28 economic growth, urban sprawl, and periods of low fuel prices. The decline in new light-duty vehicle fuel economy
29 between 1990 and 2004 reflected the increasing market share of light-duty trucks, which grew from about 30 percent
30 of new vehicle sales in 1990 to 48 percent in 2004. Starting in 2005, the rate of VMT growth slowed while average
31 new vehicle fuel economy began to increase. Average new vehicle fuel economy has improved almost every year
32 since 2005, and the truck share has decreased to about 41 percent of new vehicles in MY 2014 (EPA
33 2015a). Between 2013 and 2014, VMT increased by only 1.3 percent. Table 2-13 provides a detailed summary of
34 greenhouse gas emissions from transportation-related activities with electricity-related emissions included in the
35 totals.
36 From 2008 to 2009, CO2 emissions from the transportation end-use sector declined 4.2 percent. The decrease in
37 emissions could largely be attributed to decreased economic activity in 2009 and an associated decline in the
38 demand for transportation. Modes such as medium- and heavy-duty trucks were significantly impacted by the
39 decline in freight transport. After reaching a decadal low in 2012, CCh emissions from the transportation end-use
40 sector stabilized and grew slowly in 2013 and 2014 as the economic recovery gained strength.
Trends 2-29
-------
1 Almost all of the energy consumed for transportation was supplied by petroleum-based products, with more than
2 half being related to gasoline consumption in automobiles and other highway vehicles. Other fuel uses, especially
3 diesel fuel for freight trucks and jet fuel for aircraft, accounted for the remainder. The primary driver of
4 transportation-related emissions was CC>2 from fossil fuel combustion, which increased by 16 percent from 1990 to
5 2014. This rise in CCh emissions, combined with an increase in HFCs from close to zero emissions in 1990 to 57.4
6 MMT CO2 Eq. in 2014, led to an increase in overall emissions from transportation activities of 17 percent.
7 Table 2-13: Transportation-Related Greenhouse Gas Emissions (MMT COz Eq.)
Gas/Vehicle
Passenger Cars
CO2
CH4
N20
HFCs
Light-Duty Trucks
CO2
CH4
N20
HFCs
Medium- and Heavy-Duty
Trucks
CO2
CH4
N2O
HFCs
Buses
CO2
CH4
N20
HFCs
Motorcycles
CO2
CH4
N20
Commercial Aircraft3
CO2
CH4
N20
Other Aircraftb
CO2
CH4
N20
Ships and Boats0
CO2
CH4
N20
HFCs
Rail
CO2
CH4
N20
HFCs
Other Emissions from
Electricity Generation"1
Pipelines"
CO2
Lubricants
1990 2005 2010
656.7 708.9 783.6
629.3 660.1 742.0
3.2
24.1
+
335.6
321.1
1.7
12.8
0.0
231.1
230.1
0.3
0.7
0.0
8.4
8.4
+
+
0.0
1.8
1.7
+
+
110.9
109.9
0.0
1.0
78.3
77.5
0.1
0.7
44.9
44.3
+
0.6
0.0
39.0
38.5
0.1
0.3
0.0
0.1
36.0
36.0
11.8
l.ll 1.2
15.9 12.9
31.7 27.5
551.5 348.9
504.3 308.8
O.s| 0.4
13.2
33.3
409.3
395.4
0.1
1.1
12.7
12.1
11.8
+
+
0.3
1.7
1.6
+
+
134.0
132.7
0.0
1.2
59.7
59.1
0.1
5.5
34.2
400.0
385.6
0.1
1.1
13.2
15.8
15.3
+
0.1
0.4
3.7
3.6
+
+
114.4
113.3
0.0
1.0
40.4
40.1
+
O.sl 0.4
44.9 44.8
44.3
+
0.6
+
53.2
50.3
0.1
0.4
2.5
0.1
32.2
32.2
10.2
44.0
+
0.8
+
46.2
43.1
0.1
0.3
2.6
+
37.1
37.1
9.5
2011
774.3
736.9
1.2
12.3
23.9
332.0
294.8
0.4
5.0
31.7
398.2
383.9
0.1
1.0
13.3
16.8
16.2
+
0.1
0.4
3.6
3.6
+
+
115.7
114.6
0.0
1.1
34.2
33.9
+
0.3
46.4
45.5
+
0.8
+
47.8
44.7
0.1
0.3
2.6
+
37.8
37.8
9.0
2012
767.9
735.5
1.1
10.7
20.6
326.0
291.9
0.3
4.4
29.3
398.1
383.7
0.1
0.9
13.3
17.8
17.3
+
0.1
0.4
4.2
4.1
+
+
114.3
113.3
0.0
1.0
32.1
31.8
+
0.3
40.1
39.3
+
0.7
+
46.4
43.4
0.1
0.3
2.6
+
40.3
40.3
8.3
2013
763.2
735.5
1.0
9.4
17.3
323.4
292.5
0.3
3.9
26.7
404.6
390.3
0.1
0.9
13.3
18.0
17.5
+
0.1
0.4
4.0
3.9
+
+
115.4
114.3
0.0
1.1
34.7
34.4
+
0.3
39.4
38.7
+
0.7
+
47.2
44.2
0.1
0.3
2.6
+
45.9
45.9
8.8
2014
762.6
737.7
1.0
8.0
16.0
338.2
309.3
0.3
3.6
25.0
415.1
402.0
0.1
0.9
12.1
19.1
18.6
+
0.1
0.4
3.9
3.8
+
+
116.3
115.2
0.0
1.1
35.2
34.9
+
0.3
28.3
27.7
+
0.5
0.1
50.0
45.7
0.1
0.4
3.8
0.0
46.5
46.5
9.1
2-30 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
CO2
Total Transportation
11.8
1,554.4
10.2
2,017.7
9.5
1,844.3
9.0
1,815.7
1,795.5
1,804.6
9.1
1,824.4
International Bunker Fuel/
104.5
114.2
118.1
112.1
106.1
100.7
104.2
Note: Totals may not sum due to independent rounding. Passenger cars and light-duty trucks include vehicles typically used for
personal travel and less than 8,500 Ibs; medium- and heavy-duty trucks include vehicles larger than 8,500 Ibs. HFC emissions
primarily reflect HFC-134a.
+ Does not exceed 0.05 MMT CO2 Eq.
a Consists of emissions from jet fuel consumed by domestic operations of commercial aircraft (no bunkers).
b Consists of emissions from jet fuel and aviation gasoline consumption by general aviation and military aircraft.
c Fluctuations in emission estimates are associated with fluctuations in reported fuel consumption, and may reflect issues with data
sources.
d Other emissions from electricity generation are a result of waste incineration (as the majority of municipal solid waste is
combusted in "trash-to-steam" electricity generation plants), electrical transmission and distribution, and a portion of Other Process
Uses of Carbonates (from pollution control equipment installed in electricity generation plants).
e CO2 estimates reflect natural gas used to power pipelines, but not electricity. While the operation of pipelines produces CEU and
N2O, these emissions are not directly attributed to pipelines in the U.S. Inventory.
f Emissions from International Bunker Fuels include emissions from both civilian and military activities; these emissions are not
included in the transportation totals.
2
3
4
5
6
7
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Commercial
The commercial sector is heavily reliant on electricity for meeting energy needs, with electricity consumption for
lighting, heating, air conditioning, and operating appliances. The remaining emissions were largely due to the direct
consumption of natural gas and petroleum products, primarily for heating and cooking needs. Energy-related
emissions from the residential and commercial sectors have generally been increasing since 1990, and are often
correlated with short-term fluctuations in energy consumption caused by weather conditions, rather than prevailing
economic conditions. Landfills and wastewater treatment are included in this sector, with landfill emissions
decreasing since 1990 and wastewater treatment emissions decreasing slightly.
Residential
The residential sector is heavily reliant on electricity for meeting energy needs, with electricity consumption for
lighting, heating, air conditioning, and operating appliances. The remaining emissions were largely due to the direct
consumption of natural gas and petroleum products, primarily for heating and cooking needs. Emissions from the
residential sectors have generally been increasing since 1990, and are often correlated with short-term fluctuations in
energy consumption caused by weather conditions, rather than prevailing economic conditions. In the long-term,
this sector is also affected by population growth, regional migration trends, and changes in housing and building
attributes (e.g., size and insulation).
Agriculture
The agriculture end-use sector includes a variety of processes, including enteric fermentation in domestic livestock,
livestock manure management, and agricultural soil management. In 2014, agricultural soil management was the
largest source of N2O emissions, and enteric fermentation was the second largest source of CH4 emissions in the
United States. This sector also includes small amounts of CC>2 emissions from fossil fuel combustion by motorized
farm equipment like tractors. The agriculture sector is less reliant on electricity than the other sectors.
Box 2-1: Methodology for Aggregating Emissions by Economic Sector
In presenting the Economic Sectors in the annual Inventory of U.S. Greenhouse Gas Emissions and Sinks, the
Inventory expands upon the standard IPCC sectors common for UNFCCC reporting. Discussing greenhouse gas
emissions relevant to U.S.-specific sectors improves communication of the report's findings.
Trends 2-31
-------
1 In the Electricity Generation economic sector, CO2 emissions from the combustion of fossil fuels included in the
2 EIA electric utility fuel consuming sector are apportioned to this economic sector. Stationary combustion emissions
3 of CH4 and N2O are also based on the EIA electric utility sector. Additional sources include CO2, CH4, and N2O
4 from waste incineration, as the majority of municipal solid waste is combusted in "trash-to-steam" electricity
5 generation plants. The Electricity Generation economic sector also includes SF6 from Electrical Transmission and
6 Distribution, and a portion of CO2 from Other Process Uses of Carbonates (from pollution control equipment
7 installed in electricity generation plants).
8 In the Transportation economic sector, the CO2 emissions from the combustion of fossil fuels included in the EIA
9 transportation fuel consuming sector are apportioned to this economic sector (additional analyses and refinement of
10 the EIA data is further explained in the Energy chapter of this report). Additional emissions are apportioned from
11 the CH4 and N2O from Mobile Combustion, based on the EIA transportation sector. Substitutes of Ozone Depleting
12 Substances are apportioned based on their specific end-uses within the source category, with emissions from
13 transportation refrigeration/air-conditioning systems to this economic sector. Finally, CO2 emissions from Non-
14 Energy Uses of Fossil Fuels identified as lubricants for transportation vehicles are included in the Transportation
15 economic sector.
16 For the Industry economic sector, the CO2 emissions from the combustion of fossil fuels included in the EIA
17 industrial fuel consuming sector, minus the agricultural use of fuel explained below, are apportioned to this
18 economic sector. Stationary and mobile combustion emissions of CH4 and N2O are also based on the EIA industrial
19 sector, minus emissions apportioned to the Agriculture economic sector described below. Substitutes of Ozone
20 Depleting Substances are apportioned based on their specific end-uses within the source category, with most
21 emissions falling within the Industry economic sector (minus emissions from the other economic sectors).
22 Additionally, all process-related emissions from sources with methods considered within the IPCC Industrial
23 Process guidance have been apportioned to this economic sector. This includes the process-related emissions (i.e.,
24 emissions from the actual process to make the material, not from fuels to power the plant) from such activities as
25 Cement Production, Iron and Steel Production and Metallurgical Coke Production, and Ammonia Production.
26 Additionally, fugitive emissions from energy production sources, such as Natural Gas Systems, Coal Mining, and
27 Petroleum Systems are included in the Industry economic sector. A portion of CO2 from Other Process Uses of
28 Carbonates (from pollution control equipment installed in large industrial facilities) are also included in the Industry
29 economic sector. Finally, all remaining CO2 emissions from Non-Energy Uses of Fossil Fuels are assumed to be
30 industrial in nature (besides the lubricants for transportation vehicles specified above), and are attributed to the
31 Industry economic sector.
32 As agriculture equipment is included in EIA's industrial fuel consuming sector surveys, additional data is used to
33 extract the fuel used by agricultural equipment, to allow for accurate reporting in the Agriculture economic sector
34 from all sources of emissions, such as motorized farming equipment. Energy consumption estimates are obtained
35 from Department of Agriculture survey data, in combination with separate EIA fuel sales reports. This
36 supplementary data is used to apportion CO2 emissions from fossil fuel combustion, and CH4 and N2O emissions
37 from stationary and mobile combustion (all data is removed from the Industrial economic sector, to avoid double-
38 counting). The other emission sources included in this economic sector are intuitive for the agriculture sectors, such
39 as N2O emissions from Agricultural Soils, CH4 from Enteric Fermentation (i.e., exhalation from the digestive tracts
40 of domesticated animals), CH4 and N2O from Manure Management, CH4 from Rice Cultivation, CO2 emissions
41 from Liming and Urea Application, and CH4 and N2O from Forest Fires. N2O emissions from the Application of
42 Fertilizers to tree plantations (termed "forest land" by the IPCC) are also included in the Agriculture economic
43 sector.
44 The Residential economic sector includes the CO2 emissions from the combustion of fossil fuels reported for the
45 EIA residential sector. Stationary combustion emissions of CH4 and N2O are also based on the EIA residential fuel
46 consuming sector. Substitutes of Ozone Depleting Substances are apportioned based on their specific end-uses
47 within the source category, with emissions from residential air-conditioning systems to this economic sector. N2O
48 emissions from the Application of Fertilizers to developed land (termed "settlements" by the IPCC) are also
49 included in the Residential economic sector.
50 The Commercial economic sector includes the CO2 emissions from the combustion of fossil fuels reported in the
51 EIA commercial fuel consuming sector data. Stationary combustion emissions of CH4 and N2O are also based on the
52 EIA commercial sector. Substitutes of Ozone Depleting Substances are apportioned based on their specific end-uses
53 within the source category, with emissions from commercial refrigeration/air-conditioning systems to this economic
2-32 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 sector. Public works sources including direct CH4 from Landfills and CH4 and N2O from Wastewater Treatment and
2 Composting are included in this economic sector.
Box 2-2: Recent Trends in Various U.S. Greenhouse Gas Emissions-Related Data
5 Total emissions can be compared to other economic and social indices to highlight changes over time. These
6 comparisons include: (1) emissions per unit of aggregate energy consumption, because energy-related activities are
7 the largest sources of emissions; (2) emissions per unit of fossil fuel consumption, because almost all energy-related
8 emissions involve the combustion of fossil fuels; (3) emissions per unit of electricity consumption, because the
9 electric power industry—utilities and non-utilities combined—was the largest source of U.S. greenhouse gas
10 emissions in 2014; (4) emissions per unit of total gross domestic product as a measure of national economic activity;
11 or (5) emissions per capita.
12 Table 2-14 provides data on various statistics related to U.S. greenhouse gas emissions normalized to 1990 as a
13 baseline year. Greenhouse gas emissions in the United States have grown at an average annual rate of 0.3 percent
14 since 1990. Since 1990, this rate is slightly slower than that for total energy and for fossil fuel consumption, and
15 much slower than that for electricity consumption, overall gross domestic product and national population (see
16 Table 2-14).
17 Table 2-14: Recent Trends in Various U.S. Data (Index 1990 = 100)
Chapter/IPCC Sector
1990
2005
2010
2011
2012
2013
2014 Growth3
Greenhouse Gas Emissions15
Energy Consumption0
Fossil Fuel Consumption0
Electricity Consumption0
GDPd
Population6
a Average annual growth rate
b GWP-weighted values
0 Energy-content-weighted values (EIA 2015)
d Gross Domestic Product in chained 2009 dollars (BEA 2015)
e U.S. Census Bureau (2015)
108
115
110
137
168
125
104
112
107
135
171
126
107
116
110
136
174
126
108
117
111
138
178
127
0.3%
0.7%
0.5%
1.4%
2.5%
1.0%
18
Trends 2-33
-------
1 Figure 2-14: U.S. Greenhouse Gas Emissions Per Capita and Per Dollar of Gross Domestic
2 Product
4
5
ii
o
175
165
155 i
145
135
125
115
105
95
85
75
65
55
Real GDP
Population
Emissions
per capita
Emissions
per $GDP
9
Ol
Source: BEA (2015), U.S. Census Bureau (2015), and emission estimates in this report.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
2.3 Indirect Greenhouse Gas Emissions (CO,
NOx, NMVOCs, and SO2)
The reporting requirements of the UNFCCC8 request that information be provided on indirect greenhouse gases,
which include CO, NOX, NMVOCs, and SO2. These gases do not have a direct global warming effect, but indirectly
affect terrestrial radiation absorption by influencing the formation and destruction of tropospheric and stratospheric
ozone, or, in the case of SO2, by affecting the absorptive characteristics of the atmosphere. Additionally, some of
these gases may react with other chemical compounds in the atmosphere to form compounds that are greenhouse
gases. Carbon monoxide is produced when carbon-containing fuels are combusted incompletely. Nitrogen oxides
(i.e., NO and NO2) are created by lightning, fires, fossil fuel combustion, and in the stratosphere from N2O. Non-
CH4 volatile organic compounds—which include hundreds of organic compounds that participate in atmospheric
chemical reactions (i.e., propane, butane, xylene, toluene, ethane, and many others)—are emitted primarily from
transportation, industrial processes, and non-industrial consumption of organic solvents. In the United States, SO2 is
primarily emitted from coal combustion for electric power generation and the metals industry. Sulfur-containing
compounds emitted into the atmosphere tend to exert a negative radiative forcing (i.e., cooling) and therefore are
discussed separately.
One important indirect climate change effect of NMVOCs and NOX is their role as precursors for tropospheric ozone
formation. They can also alter the atmospheric lifetimes of other greenhouse gases. Another example of indirect
greenhouse gas formation into greenhouse gases is CO's interaction with the hydroxyl radical—the major
atmospheric sink for CH4 emissions—to form CO2. Therefore, increased atmospheric concentrations of CO limit
the number of hydroxyl molecules (OH) available to destroy CH4.
8 See < http://unfccc.int/resource/docs/2013/copl9/eng/10a03.pdf#page=2
2-34 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1
2
o
J
4
5
Since 1970, the United States has published estimates of emissions of CO, NOX, NMVOCs, and SO2 (EPA 2015),9
which are regulated under the Clean Air Act. Table 2-15 shows that fuel combustion accounts for the majority of
emissions of these indirect greenhouse gases. Industrial processes—such as the manufacture of chemical and allied
products, metals processing, and industrial uses of solvents—are also significant sources of CO, NOX, and
NMVOCs.
Table 2-15: Emissions of NOX, CO, NMVOCs, and SOz (kt)
Gas/Activity
NOx
Mobile Fossil Fuel Combustion
Stationary Fossil Fuel Combustion
Oil and Gas Activities
Industrial Processes and Product Use
Forest Fires
Waste Combustion
Agricultural Burning
Waste
CO
Mobile Fossil Fuel Combustion
Forest Fires
Stationary Fossil Fuel Combustion
Industrial Processes and Product Use
Waste Combustion
Oil and Gas Activities
Agricultural Burning
Waste
NMVOCs
Industrial Processes and Product Use
Mobile Fossil Fuel Combustion
Oil and Gas Activities
Stationary Fossil Fuel Combustion
Waste Combustion
Waste
Agricultural Burning
S02
Stationary Fossil Fuel Combustion
Industrial Processes and Product Use
Mobile Fossil Fuel Combustion
Oil and Gas Activities
Waste Combustion
Waste
Agricultural Burning
Source: (EPA 2015) except for estimates
NA (Not Available)
1990
21,770
10,862
10,023
139
592
64
82
6
+
132,271
119,360
2,300
5,000
4,129
978
302
201
1
20,930
7,638
10,932
554
912
222
673
NA
20,935
18,407
1,307
390
793
38
+
NA
from Field
2005
17,394
10,295
5,858
321
572
212
128
6
2
74,276
58,615
7,550
4,648
1,557
1,403
318
177
7
13,154
5,849
5,724
510
716
241
114
NA
13,196
11,541
831
180
619
25
1
NA
2010
12,607
7,290
4,092
545
472
121
77
7
1
50,984
39,475
4,323
4,103
1,280
1,084
487
229
5
11,641
4,133
4,591
2,205
576
92
44
NA
7,014
6,120
617
117
144
16
+
NA
2011
12
,629
7,294
3
58
38
,807
622
452
373
73
7
1
,845
,305
13,291
4
1
1
11
3
4
2
5
5
,170
,229
,003
610
232
5
,726
,929
,562
,517
599
81
38
NA
,877
,008
604
108
142
15
+
NA
2012
11,912
6,788
3,567
622
452
400
73
8
1
58,002
36,491
14,262
4,170
1,229
1,003
610
233
5
11,416
3,929
4,252
2,517
599
81
38
NA
4,711
3,859
604
108
125
15
+
NA
2013
11
6
3
,167
,283
,579
622
452
149
73
8
1
47,240
34,676
5
4
1
1
11
3
3
2
4
3
,310
,170
,229
,003
610
237
5
,107
,929
,942
,517
599
81
38
NA
,625
,790
604
108
108
15
+
NA
2014
10,605
5,777
3,522
622
452
149
73
8
1
45,424
32,861
5,310
4,169
1,229
1,003
610
238
5
10,796
3,928
3,632
2,517
599
81
39
NA
4,528
3,710
604
108
90
15
+
NA
Burning of Agricultural Residues.
Note: Totals may not sum due to independent rounding.
+ Does not exceed 0.5 kt.
Box 2-3: Sources and Effects of Sulfur Dioxide
8 Sulfur dioxide (802) emitted into the atmosphere through natural and anthropogenic processes affects the earth's
9 radiative budget through its photochemical transformation into sulfate aerosols that can (1) scatter radiation from the
10 sun back to space, thereby reducing the radiation reaching the earth's surface; (2) affect cloud formation; and (3)
NOX and CO emission estimates from Field Burning of Agricultural Residues were estimated separately, and therefore not
taken from EPA (2015).
Trends 2-35
-------
1 affect atmospheric chemical composition (e.g., by providing surfaces for heterogeneous chemical reactions). The
2 indirect effect of sulfur-derived aerosols on radiative forcing can be considered in two parts. The first indirect effect
3 is the aerosols' tendency to decrease water droplet size and increase water droplet concentration in the atmosphere.
4 The second indirect effect is the tendency of the reduction in cloud droplet size to affect precipitation by increasing
5 cloud lifetime and thickness. Although still highly uncertain, the radiative forcing estimates from both the first and
6 the second indirect effect are believed to be negative, as is the combined radiative forcing of the two (IPCC 2001).
7 However, because SC>2 is short-lived and unevenly distributed in the atmosphere, its radiative forcing impacts are
8 highly uncertain.
9 Sulfur dioxide is also a major contributor to the formation of regional haze, which can cause significant increases in
10 acute and chronic respiratory diseases. Once SC>2 is emitted, it is chemically transformed in the atmosphere and
11 returns to the earth as the primary source of acid rain. Because of these harmful effects, the United States has
12 regulated SCh emissions in the Clean Air Act.
13 Electricity generation is the largest anthropogenic source of SC>2 emissions in the United States, accounting for 64.0
14 percent in 2014. Coal combustion contributes nearly all of those emissions (approximately 92 percent). Sulfur
15 dioxide emissions have decreased in recent years, primarily as a result of electric power generators switching from
16 high-sulfur to low-sulfur coal and installing flue gas desulfurization equipment.
17
18
2-36 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
6
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
3. Energy
Energy-related activities were the primary sources of U.S. anthropogenic greenhouse gas emissions, accounting for
82.6 percent of total greenhouse gas emissions on a carbon dioxide (CCh) equivalent basis in 2014.1 This included
97, 37, and 10 percent of the nation's CCh, methane (CH4), and nitrous oxide (N2O) emissions, respectively.
Energy-related CC>2 emissions alone constituted 78.2 percent of national emissions from all sources on a CC>2
equivalent basis, while the non-CCh emissions from energy-related activities represented a much smaller portion of
total national emissions (4.4 percent collectively).
Emissions from fossil fuel combustion comprise the vast majority of energy-related emissions, with CCh being the
primary gas emitted (see Figure 3-1). Globally, approximately 32,310 MMT of CC>2 were added to the atmosphere
through the combustion of fossil fuels in 2012, of which the United States accounted for approximately 16 percent.2
Due to their relative importance, fossil fuel combustion-related CCh emissions are considered separately, and in
more detail than other energy-related emissions (see Figure 3-2). Fossil fuel combustion also emits CH4 and N2O.
Stationary combustion of fossil fuels was the second-largest source of N2O emissions in the United States and
mobile fossil fuel combustion was the fourth-largest source.
Figure 3-1: 2014 Energy Chapter Greenhouse Gas Sources (MMT COz Eq.)
Fossil Fuel Combustion
Natural Gas Systems"
Non-Energy Use of Fuels
Coal Mining
Stationary Combustion
Petroleum Systemsb
Mobile Combustion
Incineration of Waste
Abandoned Underground Coal Mines
Energy as a Portion
of all Emissions
50
100 150
MMT CO2 Eq.
200
a The value in this figure for Natural Gas Systems is presented from the previous Inventory and does not reflect updates to
emission estimates for this category. See Section 3.7 Natural Gas Systems in this chapter for more information.
b The value in this figure for Petroleum Systems is presented from the previous Inventory and does not reflect updates to
emission estimates for this category. See Section 3.6 Petroleum Systems in this chapter for more information.
1 Estimates are presented in units of million metric tons of carbon dioxide equivalent (MMT CCh Eq.), which weight each gas by
its global wanning potential, or GWP, value. See section on global warming potentials in the Executive Summary.
Global CO2 emissions from fossil fuel combustion were taken from Energy Information Administration International Energy
Statistics 2013 EIA (2013).
Energy 3-1
-------
Figure 3-2: 2014 U.S. Fossil Carbon Flows (MMT COz Eq.)
NEU Emissions 2
Fossil Fuel
Energy Export:
818
Coal Emissions
1,656
NEU Emissions 6
Natural Gas Emissions
1,432
Fossil Fuel
Non-Energy Consumption
Use Imports u-5
19 Territories
Non-Energy Use
Carbon Sequestered
Note: Totals may not sum due to independent rounding.
The 'Balancing Item" above accounts for the statistical
Imbalances and unknowns in trie reported data sets combined
here.
NEU = Non-Energy Use
NG = Natural Gas
4 Energy-related activities other than fuel combustion, such as the production, transmission, storage, and distribution
5 of fossil fuels, also emit greenhouse gases. These emissions consist primarily of fugitive CH4 from natural gas
6 systems, petroleum systems, and coal mining. Table 3-1 summarizes emissions from the Energy sector in units of
7 million metric tons of CCh equivalents (MMT CCh Eq.), while unweighted gas emissions in kilotons (kt) are
8 provided in Table 3-2. Overall, emissions due to energy-related activities were 5,679.8 MMT CCh Eq. in 2014,3 an
9 increase of 7.3 percent since 1990.
10 Table 3-1: COz, CH4, and N2O Emissions from Energy (MMT COz Eq.)
Gas/Source
CO2
Fossil Fuel Combustion
Electricity Generation
Transportation
Industrial
Residential
Commercial
U.S. Territories
Non-Energy Use of Fuels
Natural Gas Systems'5
Incineration of Waste
Petroleum Systems0
Biomass- Wood"
International Bunker Fuels"
Biomass-Ethanol"
CH4
Natural Gas Systems'5
Coal Mining
Petroleum Systems0
Stationary Combustion
1990
4,908.8
4,740.7
1,820.8
1,493.8 \
842.5
338.3
217.4
27.9
118.1
37.6
8.0 1
4.4
215.2
103.5
4.2
328.5
179.1
96.5
31.5
8.5
2005
5,933.4
5,747.1
2,400.9
1,887.0
828.0
357.8
223.5
49.9
\ 138.9 |
30.0
| 12.5
4.9
206.9
113.1
22.9
\ 280.7
176.3
| 64.1
23.5
7.4
2010
5,519.9
5,358.3
2,258.4
1,728.3
775.5
334.6
220.1
41.4
114.1
32.3
11.0
4.2
192.5
117.0
72.6
279.2
159.6
82.3
21.3
7.1
2011
5,386.8
5,227.7
2,157.7
1,707.6
773.3
326.8
220.7
41.5
108.5
35.6
10.5
4.5
195.2
111.7
72.9
268.2
159.3
71.2
22.0
7.1
2012
5,180.5
5,024.7
2,022.2
1,696.8
782.9
282.5
196.7
43.6
105.6
34.8
10.4
5.1
194.9
105.8
72.8
259.1
154.4
66.5
23.3
6.6
2013
5,332.5
5,157.6
2,038.1
1,713.0
812.2
329.7
221.0
43.5
121.7
37.8
9.4
6.0
211.6
99.8
74.7
263.5
157.4
64.6
25.2
8.0
2014
5,376.2
5,208.7
2,039.3
1,737.4
814.2
345.1
231.6
41.0
114.3
37.8
9.4
6.0
217.7
103.2
76.1
263.5
157.4
64.6
25.2
8.1
•^ Following the revised reporting requirements under the UNFCCC, this Inventory report presents CCh equivalent values based
on the IPCC Fourth Assessment Report (AR4) GWP values. See the Introduction chapter for more information.
3-2 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Abandoned Underground Coal
Mines
Mobile Combustion
Incineration of Waste
International Bunker Fuels"
N20
Stationary Combustion
Mobile Combustion
Incineration of Waste
International Bunker Fuels"
Total
7.2 1
I
53.6 1
11.9 1
41.2
0.5 1
0.9
5,290.9
6.6 1
2.7 1
+
0.1 \
55.0 1
20.2 1
34.4 1
0.4 1
1.0
6,269.0
6.6
2.3
+
0.1
46.1
22.2
23.6
0.3
1.0
5,845.2
6.4
2.2
+
0.1
44.0
21.3
22.4
0.3
1.0
5,699.0
6.2
2.2
+
0.1
41.7
21.4
20.0
0.3
0.9
5,481.3
6.2
2.1
+
0.1
41.4
22.9
18.2
0.3
0.9
5,637.4
6.2
2.0
+
0.1
40.0
23.4
16.3
0.3
0.9
5,679.8
+ Does not exceed 0.05 MMT CO2 Eq.
a These values are presented for informational purposes only, in line with IPCC methodological guidance and UNFCCC reporting obligations,
and are not included in the specific energy sector contribution to the totals, and are already accounted for elsewhere.
b The values in this table for Natural Gas Systems are presented from the previous Inventory and do not reflect updates to emission estimates for
this category. See Section 3.7 Natural Gas Systems in this chapter for more information. Gray highlighting was added on 2/24 for clarification.
c The values in this table for Petroleum Systems are presented from the previous Inventory and do not reflect updates to emission estimates for
this category. See Section 3.6 Petroleum Systems in this chapter for more information. Gray highlighting was added on 2/24 for clarification.
Note: Totals may not sum due to independent rounding. Gray highlight added 2/24 for clarification.
1 Table 3-2: COz, CH4, and NzO Emissions from Energy (kt)
Gas/Source
CO2
Fossil Fuel Combustion
Non-Energy Use of Fuels
Natural Gas Systems'5
Incineration of Waste
Petroleum Systems0
Biomass -Wood"
International Bunker Fuels"
Biomass - Ethanol"
CH4
Natural Gas Systems'5
Coal Mining
Petroleum Systems0
Stationary Combustion
Abandoned Underground
Coal Mines
Mobile Combustion
Incineration of Waste
International Bunker Fuels"
N20
Mobile Combustion
Stationary Combustion
Incineration of Waste
International Bunker Fuels"
1990 2005
4,908,847 5,933,371
4,740,671 5,747,142
118,114 | 138,876 |
37,645 29,995
7,972 12,454
4,445 4,904
215,186 1 206,901 \
103,463 113,139
4,227 \ 22,943 \
13,139 11,226
7,165 7,053
3,860 2,565
1,261 939
339 1 295 1
288 1 264
226 1 110
180 1 185
40 1 68
138 1 115
2 1 1 1
3 3
2010
5,519,868
5,358,292
114,063
32,334
11,026
4,153
192,462
116,992
72,647
11,166
6,382
3,293
854
282
263
91
+
6
155
74
79
1
3
2011
5,386,773
5,227,690
108,515
35,551
10,550
4,467
195,182
111,660
72,881
10,728
6,371
2,849
878
283
257
90
+
5
148
71
75
1
3
2012
5,180,488
5,024,685
105,617
34,764
10,362
5,060
194,903
105,805
72,827
10,365
6,176
2,658
931
264
249
86
+
4
140
72
67
1
3
2013
5,332,495
5,157,583
121,682
37,808
9,421
6,001
211,581
99,763
74,743
10,541
6,295
2,584
1,009
320
249
84
+
3
139
77
61
1
3
2014
5,376,218
5,208,654
114,333
37,808
9,421
6,001
217,654
103,201
76,075
10,542
6,295
2,584
1,009
323
249
82
+
3
134
79
55
1
3
+ Does not exceed 0.5 kt
" These values are presented for informational purposes only, in line with IPCC methodological guidance and UNFCCC reporting obligations, and are
not included in the specific energy sector contribution to the totals, and are already accounted for elsewhere.
b The values in this table for Natural Gas Systems are presented from the previous Inventory and do not reflect updates to emission estimates for this
category. See Section 3.7, Natural Gas Systems in this chapter for more information. Gray highlighting was added on 2/24 for clarification.
c The values in this table for Petroleum Systems are presented from the previous Inventory and do not reflect updates to emission estimates for this
category. See Section 3.6, Petroleum Systems in this chapter for more information. Gray highlighting was added on 2/24 for clarification.
Note: Totals may not sum due to independent rounding.
Box 3-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Sinks
3 In following the UNFCCC requirement under Article 4.1 to develop and submit national greenhouse gas emission
4 inventories, the emissions and sinks presented in this report and this chapter, are organized by source and sink
Energy 3-3
-------
1 categories and calculated using internationally-accepted methods provided by the Intergovernmental Panel on
2 Climate Change (IPCC). Additionally, the calculated emissions and sinks in a given year for the United States are
3 presented in a common manner in line with the UNFCCC reporting guidelines for the reporting of inventories under
4 this international agreement. The use of consistent methods to calculate emissions and sinks by all nations
5 providing their inventories to the UNFCCC ensures that these reports are comparable. In this regard, U.S. emissions
6 and sinks reported in this inventory report are comparable to emissions and sinks reported by other countries.
7 Emissions and sinks provided in this Inventory do not preclude alternative examinations, but rather, this Inventory
8 presents emissions and sinks in a common format consistent with how countries are to report Inventories under the
9 UNFCCC. The report itself, and this chapter, follows this standardized format, and provides an explanation of the
10 IPCC methods used to calculate emissions and sinks, and the manner in which those calculations are conducted.
11
12
35
36
37
Box 3-2: Energy Data from the Greenhouse Gas Reporting Program
13 On October 30, 2009, the U.S. Environmental Protection Agency (EPA) published a rule for the mandatory
14 reporting of greenhouse gases (GHG) from large GHG emissions sources in the United States. Implementation of 40
15 CFR Part 98 is referred to as the Greenhouse Gas Reporting Program (GHGRP). 40 CFR Part 98 applies to direct
16 greenhouse gas emitters, fossil fuel suppliers, industrial gas suppliers, and facilities that inject CO2 underground for
17 sequestration or other reasons. Reporting is at the facility level, except for certain suppliers of fossil fuels and
18 industrial greenhouse gases. 40 CFR part 98 requires reporting by 41 industrial categories. Data reporting by
19 affected facilities included the reporting of emissions from fuel combustion at that affected facility. In general, the
20 threshold for reporting is 25,000 metric tons or more of CO2 Eq. per year.
21 The GHGRP dataset and the data presented in this inventory report are complementary and, as indicated in the
22 respective planned improvements sections for source categories in this chapter, EPA is analyzing how to use
23 facility-level GHGRP data to improve the national estimates presented in this Inventory (see, also, Box 3-4). Most
24 methodologies used in EPA's GHGRP are consistent with IPCC, though for EPA's GHGRP, facilities collect
25 detailed information specific to their operations according to detailed measurement standards, which may differ with
26 the more aggregated data collected for the Inventory to estimate total, national U.S. emissions. It should be noted
27 that the definitions and provisions for reporting fuel types in EPA's GHGRP may differ from those used in the
28 Inventory in meeting the UNFCCC reporting guidelines. In line with the UNFCCC reporting guidelines, the
29 inventory report is a comprehensive accounting of all emissions from fuel types identified in the IPCC guidelines
30 and provides a separate reporting of emissions from biomass. Further information on the reporting categorizations in
31 EPA's GHGRP and specific data caveats associated with monitoring methods in EPA's GHGRP has been provided
32 on the GHGRP website.
33 EPA presents the data collected by its GHGRP through a data publication tool that allows data to be viewed in
34 several formats including maps, tables, charts and graphs for individual facilities or groups of facilities.
3.1 Fossil Fuel Combustion (IPCC Source
Category 1A)
38 Emissions from the combustion of fossil fuels for energy include the gases CO2, CH4, and N2O. Given that CO2 is
39 the primary gas emitted from fossil fuel combustion and represents the largest share of U.S. total emissions, CO2
40 emissions from fossil fuel combustion are discussed at the beginning of this section. Following that is a discussion
41 of emissions of all three gases from fossil fuel combustion presented by sectoral breakdowns. Methodologies for
42 estimating CO2 from fossil fuel combustion also differ from the estimation of CH4 and N2O emissions from
43 stationary combustion and mobile combustion. Thus, three separate descriptions of methodologies, uncertainties,
44 recalculations, and planned improvements are provided at the end of this section. Total CO2, CH4, and N2O
45 emissions from fossil fuel combustion are presented in Table 3-3 and Table 3-4.
3-4 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Table 3-3: COz, CH4, and NzO Emissions from Fossil Fuel Combustion (MMT COz Eq.)
Gas
CO2
CH4
N2O
Total
Table 3-4:
Gas
C02
CH4
N20
1990
4,740.7 1
14.1
53.1
4,807.9
COz, CH4,
1990
4,740,671
565
178 |
2005
5,747.1
54.6
5,811.9
2010
5,358.3
9.3
45.8
5,413.5
and NzO Emissions from
2005
5,747,142
405
183
2010
5,358,292
374
154
2011
5,227.7
9.3
43.8
5,280.8
Fossil Fuel
2011
5,227,690
373
147
2012
5,024.7
8.8
41.4
5,074.9
2013
5,157.6
10.1
41.2
5,208.8
2014
5,208.7
10.1
39.8
5,258.5
Combustion (kt)
2012
5,024,685
351
139
2013
5,157,583
403
138
2014
5,208,654
405
133
Note: Totals may not sum due to independent rounding
3 CO2 from Fossil Fuel Combustion
4 CO2 is the primary gas emitted from fossil fuel combustion and represents the largest share of U.S. total greenhouse
5 gas emissions. CO2 emissions from fossil fuel combustion are presented in Table 3-5. In 2014, CO2 emissions from
6 fossil fuel combustion increased by 1.0 percent relative to the previous year. The increase in €62 emissions from
7 fossil fuel combustion was a result of multiple factors, including: (1) colder winter conditions in the first quarter of
8 2014 resulting in an increased demand for heating fuel in the residential and commercial sectors; (2) an increase in
9 transportation emissions resulting from an increase in vehicle miles traveled (VMT) and fuel use across on-road
10 transportation modes; and (3) an increase in industrial production across multiple sectors resulting in slight increases
11 in industrial sector emissions.4 In 2014, CO2 emissions from fossil fuel combustion were 5,208.7 MMT CO2 Eq., or
12 9.9 percent above emissions in 1990 (see Table 3-5).5
13 Table 3-5: COz Emissions from Fossil Fuel Combustion by Fuel Type and Sector (MMT COz
14 Eq.)
Fuel/Sector
Coal
Residential
Commercial
Industrial
Transportation
Electricity Generation
U.S. Territories
Natural Gas
Residential
Commercial
Industrial
Transportation
Electricity Generation
U.S. Territories
Petroleum
Residential
1990
1,718.4
3.0
12.0 1
155.3
NE 1
1,547.6
0.6
1,000.3
238.0
142.1
408.9
36.0 1
175.3
NO 1
2,021.5
97.4
I 2005
2,112.3
0.8
9.3
115.3
NE
1,983.8
3.0
1,166.7
262.2 1
162.9
388.5 1
33.1
318.8 1
1.3
2,467.8
94.9
2010
1,927.7
0.0
6.6
90.1
NE
1,827.6
3.4
1,272.1
258.6
167.7
407.2
38.1
399.0
1.5
2,158.2
76.0
2011
1,813.9
0.0
5.8
82.0
NE
1,722.7
3.4
1,291.5
254.7
170.5
417.3
38.9
408.8
1.4
2,121.9
72.2
2012
1,592.8
0.0
4.1
74.1
NE
1,511.2
3.4
1,352.6
224.8
156.9
434.8
41.3
492.2
2.6
2,078.9
57.7
2013
1,654.4
0.0
3.9
75.7
NE
1,571.3
3.4
1,391.2
266.2
179.1
451.9
47.0
444.0
3.0
2,111.6
63.4
2014
1,653.7
0.0
4.5
75.3
NE
1,570.4
3.4
1,426.6
277.6
189.2
466.0
47.6
443.2
3.0
2,128.0
67.6
4 Further details on industrial sector combustion emissions are provided by EPA's GHGRP
.
5 An additional discussion of fossil fuel emission trends is presented in the Trends in U.S. Greenhouse Gas Emissions Chapter.
Energy 3-5
-------
Commercial
Industrial
Transportation
Electricity Generation
U.S. Territories
Geothermal3
Total
63.3 1
278.3
1,457.7
97.5
27.2
0.4
4,740.7
51.3
324.2 1
1,854.0
97.9
45.6
0.4
5,747.1
145.8
278.2
1,690.2
31.4
36.5
_ 0.4
5,358.3
44.5
274.0
1,668.8
25.8
36.7
0.4
5,227.7
35.7
274.1
1,655.4
18.3
37.6
0.4
5,024.7
38.0
284.6
1,666.0
22.4
37.1
0.4
5,157.6
37.9
272.9
1,689.8
25.3
34.6
0.4
5,208.7
+ Does not exceed 0.05 MMT CO2 Eq.
NE (Not estimated)
NO (Not occurring)
a Although not technically a fossil fuel, geothermal energy-related CCh emissions are included for reporting
purposes.
Note: Totals may not sum due to independent rounding.
1 Trends in CC>2 emissions from fossil fuel combustion are influenced by many long-term and short-term factors. On
2 a year-to-year basis, the overall demand for fossil fuels in the United States and other countries generally fluctuates
3 in response to changes in general economic conditions, energy prices, weather, and the availability of non-fossil
4 alternatives. For example, in a year with increased consumption of goods and services, low fuel prices, severe
5 summer and winter weather conditions, nuclear plant closures, and lower precipitation feeding hydroelectric dams,
6 there would likely be proportionally greater fossil fuel consumption than a year with poor economic performance,
7 high fuel prices, mild temperatures, and increased output from nuclear and hydroelectric plants.
8 Longer-term changes in energy consumption patterns, however, tend to be more a function of aggregate societal
9 trends that affect the scale of consumption (e.g., population, number of cars, size of houses, and number of houses),
10 the efficiency with which energy is used in equipment (e.g., cars, power plants, steel mills, and light bulbs), and
11 social planning and consumer behavior (e.g., walking, bicycling, or telecommuting to work instead of driving).
12 CO2 emissions also depend on the source of energy and its carbon (C) intensity. The amount of C in fuels varies
13 significantly by fuel type. For example, coal contains the highest amount of C per unit of useful energy. Petroleum
14 has roughly 75 percent of the C per unit of energy as coal, and natural gas has only about 55 percent.6 Table 3-6
15 shows annual changes in emissions during the last five years for coal, petroleum, and natural gas in selected sectors.
16 Table 3-6: Annual Change in COz Emissions and Total 2014 Emissions from Fossil Fuel
17 Combustion for Selected Fuels and Sectors (MMT COz Eq. and Percent)
Sector Fuel Type
Electricity Generation Coal
Electricity Generation Natural Gas
Electricity Generation Petroleum
Transportation* Petroleum
Residential Natural Gas
Commercial Natural Gas
Industrial Coal
Industrial Natural Gas
All Sectors" All Fuels"
2010 to 2011
-104.9 -5.7%
9.8 2.5%
-5.6 -17.8%
-21.4 -1.3%
-3.9 -1.5%
2.7 1.6%
-8.1 -9.0%
10.1 2.5%
-130.6 -2.4%
2011 to 2012
-211.5 -12.3%
83.5 20.4%
-7.5 -29.0%
-13.3 -0.8%
-29.8 -11.7%
-13.6 -8.0%
-7.9 -9.7%
17.5 4.2%
-203.0 -3.9%
2012 to 2013
60.1 4.0%
-48.3 -9.8%
4.1 22.3%
10.6 0.6%
41.4 18.4%
22.3 14.2%
1.7 2.3%
17.1 3.9%
132.9 2.6%
2013 to 2014
-0.9 -0.1%
-0.8 -0.2%
2.9 12.8%
23.8 1.4%
11.4 4.3%
10.0 5.6%
-0.4 -0.6%
14.2 3.1%
51.1 1.0%
Total 2014
1,570.4
443.2
25.3
1,689.8
277.6
189.2
75.3
466.0
5,208.7
a Excludes emissions from International Bunker Fuels.
b Includes fuels and sectors not shown in table.
18 In the United States, 82 percent of the energy consumed in 2014 was produced through the combustion of fossil
19 fuels such as coal, natural gas, and petroleum (see Figure 3-3 and Figure 3-4). The remaining portion was supplied
20 by nuclear electric power (8 percent) and by a variety of renewable energy sources (10 percent), primarily
' Based on national aggregate carbon content of all coal, natural gas, and petroleum fuels combusted in the United States.
3-6 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 hydroelectric power, wind energy and biofuels (EIA 2015a).7 Specifically, petroleum supplied the largest share of
2 domestic energy demands, accounting for 35 percent of total U.S. energy consumption in 2014. Natural gas and
3 coal followed in order of energy demand importance, accounting for approximately 28 percent and 19 percent of
4 total U.S. energy consumption, respectively. Petroleum was consumed primarily in the transportation end-use sector
5 and the vast majority of coal was used in electricity generation. Natural gas was broadly consumed in all end-use
6 sectors except transportation (see Figure 3-5) (EIA 2015a).
7 Figure 3-3: 2014 U.S. Energy Consumption by Energy Source (Percent)
Renewable
Energy
9.8%
Nuclear Electric
Power
8.5%
9 Figure 3-4: U.S. Energy Consumption (Quadrillion Btu)
120 -i
100 -
^^
3
2 8°
E 60
jS 40
20 H
0
Total Energy
Renewable & Nuclear
o i-t (M
* ^
a a
10
Renewable energy, as defined in EIA's energy statistics, includes the following energy sources: hydroelectric power,
geothermal energy, biofuels, solar energy, and wind energy.
Energy 3-7
-------
1 Figure 3-5: 2014 COz Emissions from Fossil Fuel Combustion by Sector and Fuel Type (MMT
2 COz Eq.)
2,500 1
2,000
1,500
1,000
500
0
Relative Contribution
by Fuel Type
Petroleum
• Coal
• Natural Gas
2,039
4 Fossil fuels are generally combusted for the purpose of producing energy for useful heat and work. During the
5 combustion process, the C stored in the fuels is oxidized and emitted as CCh and smaller amounts of other gases,
6 including CH4, CO, and NMVOCs.8 These other C containing non-CCh gases are emitted as a byproduct of
7 incomplete fuel combustion, but are, for the most part, eventually oxidized to CC>2 in the atmosphere. Therefore, it
8 is assumed all of the C in fossil fuels used to produce energy is eventually converted to atmospheric CCh.
Box 3-3: Weather and Non-Fossil Energy Effects on COz from Fossil Fuel Combustion Trends
11 In 2014, weather conditions, and a very cold first quarter of the year in particular, caused a significant increase in
12 energy demand for heating fuels and is reflected in the increased residential emissions during the early part of the
13 year (EIA 2015a). The United States in 2014 also experienced a cooler winter overall compared to 2013, as heating
14 degree days increased (1.9 percent). Cooling degree days decreased by 0.6 percent and despite this decrease in
15 cooling degree days, electricity demand to cool homes still increased slightly. Colder winter conditions compared to
16 2013 resulted in a significant increase in the amount of energy required for heating, and heating degree days in the
17 United States were 0.6 percent above normal for the first time since 2003 (see Figure 3-6). Summer conditions were
18 slightly cooler in 2014 compared to 2013, and summer temperatures were warmer than normal, with cooling degree
19 days 6.7 percent above normal (see Figure 3-7) (EIA 2015a).9
8 See the sections entitled Stationary Combustion and Mobile Combustion in this chapter for information on non-CCh gas
emissions from fossil fuel combustion.
9 Degree days are relative measurements of outdoor air temperature. Heating degree days are deviations of the mean daily
temperature below 65° F, while cooling degree days are deviations of the mean daily temperature above 65° F. Heating degree
days have a considerably greater effect on energy demand and related emissions than do cooling degree days. Excludes Alaska
and Hawaii. Normals are based on data from 1971 through 2000. The variation in these normals during this time period was +10
percent and +14 percent for heating and cooling degree days, respectively (99 percent confidence interval).
3-8 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Figure 3-6: Annual Deviations from Normal Heating Degree Days for the United States
2 (1950-2014)
E
20
-20 J
Normal
(4,524 Heating Degree Days)
O
-------
3
4
1 Figure 3-8: Nuclear, Hydroelectric, and Wind Power Plant Capacity Factors in the United
2 States (1990-2014)
100
90
80
1 6°
i2 50
1 40
<3 30
20
10
0
Wind
Fossil Fuel Combustion Emissions by Sector
6 In addition to the CC>2 emitted from fossil fuel combustion, CH4 and N2O are emitted from stationary and mobile
7 combustion as well. Table 3-7 provides an overview of the CCh, CH4, and N2O emissions from fossil fuel
8 combustion by sector.
9 Table 3-7: COz, CH4, and NzO Emissions from Fossil Fuel Combustion by Sector (MMT COz
10 Eq.)
End-Use Sector
Electricity Generation
C02
CH4
N20
Transportation
C02
CH4
N20
Industrial
C02
CH4
N20
Residential
C02
CH4
N2O
Commercial
CO2
CH4
N2O
U.S. Territories3
1990
1,828.5
1,820.8
0.3
7.4 1
1,540.6
1,493.8
5.6
41.2 1
847.4
842.5
1.8
3.1 1
344.6
338.3
5.2 1
1.0 1
218.8
217.4
1.0
0.4
28.0
2005
2,417.4
2,400.9
16.0 1
1,924.1
1,887.0
2.7
34.4 1
832.6
828.0
2.9 1
362.8
357.8
4.1 1
0.9
224.9
223.5
50.1 H
2010
2,277.4
2,258.4
0.5
18.5
1,754.2
1,728.3
2.3
23.6
779.5
775.5
1.5
2.5
339.4
334.6
4.0
0.8
221.5
220.1
1.1
0.3
41.6
2011
2,175.8
2,157.7
0.4
17.6
1,732.3
1,707.6
2.2
22.4
777.2
773.3
1.5
2.4
331.7
326.8
4.0
0.8
222.1
220.7
1.0
0.3
41.7
2012
2,040.4
2,022.2
0.4
17.8
1,718.9
1,696.8
2.2
20.0
786.9
782.9
1.5
2.4
287.0
282.5
3.7
0.7
197.9
196.7
0.9
0.3
43.7
2013
2,057.7
2,038.1
0.4
19.1
1,733.3
1,713.0
2.1
18.2
816.2
812.2
1.5
2.4
335.6
329.7
5.0
1.0
222.4
221.0
1.0
0.3
43.7
2014
2,059.4
2,039.3
0.4
19.6
1,755.8
1,737.4
2.0
16.3
818.1
814.2
1.5
2.4
351.1
345.1
5.0
1.0
233.0
231.6
1.1
0.3
41.2
3-10 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Total
4,807.9
5,811.9
5,413.5 5,280.8 5,074.9 5,208.8 5,258.5
Note: Totals may not sum due to independent rounding. Emissions from fossil fuel combustion by
electricity generation are allocated based on aggregate national electricity consumption by each
end-use sector.
a U.S. Territories are not apportioned by sector, and emissions are total greenhouse gas emissions
from all fuel combustion sources.
1 Other than CO2, gases emitted from stationary combustion include the greenhouse gases CH4 and N2O and the
2 indirect greenhouse gases NOX, CO, and NMVOCs.11 Methane and N2O emissions from stationary combustion
3 sources depend upon fuel characteristics, size and vintage, along with combustion technology, pollution control
4 equipment, ambient environmental conditions, and operation and maintenance practices. N2O emissions from
5 stationary combustion are closely related to air-fuel mixes and combustion temperatures, as well as the
6 characteristics of any pollution control equipment that is employed. Methane emissions from stationary combustion
7 are primarily a function of the CH4 content of the fuel and combustion efficiency.
8 Mobile combustion produces greenhouse gases other than CO2, including CH4, N2O, and indirect greenhouse gases
9 including NOX, CO, and NMVOCs. As with stationary combustion, N2O and NOX emissions from mobile
10 combustion are closely related to fuel characteristics, air-fuel mixes, combustion temperatures, and the use of
11 pollution control equipment. N2O from mobile sources, in particular, can be formed by the catalytic processes used
12 to control NOX, CO, and hydrocarbon emissions. Carbon monoxide emissions from mobile combustion are
13 significantly affected by combustion efficiency and the presence of post-combustion emission controls. Carbon
14 monoxide emissions are highest when air-fuel mixtures have less oxygen than required for complete combustion.
15 These emissions occur especially in idle, low speed, and cold start conditions. Methane and NMVOC emissions
16 from motor vehicles are a function of the CH4 content of the motor fuel, the amount of hydrocarbons passing
17 uncombusted through the engine, and any post-combustion control of hydrocarbon emissions (such as catalytic
18 converters).
19 An alternative method of presenting combustion emissions is to allocate emissions associated with electricity
20 generation to the sectors in which it is used. Four end-use sectors were defined: industrial, transportation,
21 residential, and commercial. In the table below, electricity generation emissions have been distributed to each end-
22 use sector based upon the sector's share of national electricity consumption, with the exception of CH4 and N2O
23 from transportation.12 Emissions from U.S. Territories are also calculated separately due to a lack of end-use-
24 specific consumption data. This method assumes that emissions from combustion sources are distributed across the
25 four end-use sectors based on the ratio of electricity consumption in that sector. The results of this alternative
26 method are presented in Table 3-8.
27 Table 3-8: COz, CH4, and NzO Emissions from Fossil Fuel Combustion by End-Use Sector
28 (MMT COz Eq.)
End-Use Sector
1990
2005
2010
2011
2012
2013
2014
Transportation
C02
CH4
N20
Industrial
C02
CH4
N20
Residential
C02
CH4
N20
1,722.8
1,700.6
2.2
20.1
1,385.0
1,375.7
1.6
7.7
1,018.8
1,007.8
3.9
7.1
1,737.4
1,717.0
2.1
18.2
1,416.6
1,407.0
1.6
8.0
1,077.5
1,064.6
5.1
7.9
1,759.9
1,741.5
2.0
16.4
1,417.5
1,407.8
1.6
8.1
1,093.6
1,080.4
5.2
8.1
11 Sulfur dioxide (SO2) emissions from stationary combustion are addressed in Annex 6.3.
Separate calculations were performed for transportation-related CH4 and N2O. The methodology used to calculate these
emissions are discussed in the mobile combustion section.
Energy 3-11
-------
Commercial
CO2
CH4
N2O
U.S. Territories3
759.1
755.4
1.1
2.5
28.0
11,033.7
1,026.8
1,000.9
993.0
1.2
6.6
41.6
966.3
958.8
1.2
6.3
41.7
904.5
897.0
1.1
6.4
43.7
933.6
925.5
1.2
6.9
43.7
946.4
938.1
1.2
7.1
41.2
Total
4,807.9
5,811.9
5,413.5 5,280.8 5,074.9 5,208.8 5,258.5
Note: Totals may not sum due to independent rounding. Emissions from fossil fuel combustion by
electricity generation are allocated based on aggregate national electricity consumption by each end-use
sector.
a U.S. Territories are not apportioned by sector, and emissions are total greenhouse gas emissions from all
fuel combustion sources.
i Stationary Combustion
2 The direct combustion of fuels by stationary sources in the electricity generation, industrial, commercial, and
3 residential sectors represent the greatest share of U.S. greenhouse gas emissions. Table 3-9 presents CO2 emissions
4 from fossil fuel combustion by stationary sources. The CO2 emitted is closely linked to the type of fuel being
5 combusted in each sector (see Methodology section of CO2 from Fossil Fuel Combustion). Other than CO2, gases
6 emitted from stationary combustion include the greenhouse gases CH4 and N2O. Table 3-10 and Table 3-11 present
7 CH4 and N2O emissions from the combustion of fuels in stationary sources.13 Methane and N2O emissions from
8 stationary combustion sources depend upon fuel characteristics, combustion technology, pollution control
9 equipment, ambient environmental conditions, and operation and maintenance practices. N2O emissions from
10 stationary combustion are closely related to air-fuel mixes and combustion temperatures, as well as the
11 characteristics of any pollution control equipment that is employed. Methane emissions from stationary combustion
12 are primarily a function of the CH4 content of the fuel and combustion efficiency. The CH4 and N2O emission
13 estimation methodology was revised in 2010 to utilize the facility-specific technology and fuel use data reported to
14 EPA's Acid Rain Program (see Methodology section for CH4 and N2O from stationary combustion). Please refer to
15 Table 3-7 for the corresponding presentation of all direct emission sources of fuel combustion.
16 Table 3-9: COz Emissions from Stationary Fossil Fuel Combustion (MMT COz Eq.)
Sector/Fuel Type
Electricity Generation
Coal
Natural Gas
Fuel Oil
Geothermal
Industrial
Coal
Natural Gas
Fuel Oil
Commercial
Coal
Natural Gas
Fuel Oil
Residential
Coal
Natural Gas
Fuel Oil
U.S. Territories
Since emission estimates for U.S.
1990
1,820.8
1,547.6
175.3
97.5
0.4 1
842.5
155.3
408.9
278.3
217.4
12.0 1
142.1
63.3
338.3
3.0 1
238.0
97.4
27.9 |
2005
2,400.9
1,983.8
318.8
97.9
0.4 1
828.0
115.3
388.5
324.2
223.5
9.3 1
162.9
51.3
357.8
0.8 1
262.2
94.9
49.9 |
2010
2,258.4
1,827.6
399.0
31.4
0.4
775.5
90.1
407.2
278.2
220.1
6.6
167.7
45.8
334.6
0.0
258.6
76.0
41.4
Territories cannot be disaggregated by
2011
2,157.7
1,722.7
408.8
25.8
0.4
773.3
82.0
417.3
274.0
220.7
5.8
170.5
44.5
326.8
0.0
254.7
72.2
41.5
gas in Table
2012
2,022.2
1,511.2
492.2
18.3
0.4
782.9
74.1
434.8
274.1
196.7
4.1
156.9
35.7
282.5
0.0
224.8
57.7
43.6
2013
2,038.1
1,571.3
444.0
22.4
0.4
812.2
75.7
451.9
284.6
221.0
3.9
179.1
38.0
329.7
0.0
266.2
63.4
43.5
3-10 and Table 3-11, the
2014
2,039.3
1,570.4
443.2
25.3
0.4
814.2
75.3
466.0
272.9
231.6
4.5
189.2
37.9
345.1
0.0
277.6
67.6
41.0
values for CH4
andN2O exclude U.S. territory emissions.
3-12 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Coal
Natural Gas
Fuel Oil
Total
0.6
NO
27.2
3,246.9
+ Does not exceed 0.05 MMT CO2 Eq.
NO: Not occurring
Table 3-10: ChU Emissions from
Sector/Fuel Type
Electric Power
Coal
Fuel Oil
Natural gas
Wood
Industrial
Coal
Fuel Oil
Natural gas
Wood
Commercial/Institutional
Coal
Fuel Oil
Natural gas
Wood
Residential
Coal
Fuel Oil
Natural Gas
Wood
U.S. Territories
Coal
Fuel Oil
Natural Gas
Wood
Total
1990
g
+
;
1.8 1
0.4 1
0.2 1
0.2 1
1.0 1
1.0 1
+ 1
0.2 1
S
5.2 1
0.2 1
0.3 1
0.5
4.1
+ 1
1
1
0.0 1
0.0
8.5
1 1
3,860.1
13.4 3.4
1.5 1.4
36.5 36.7
3.4
2.6
37.6
3.4 3.4
3.0 3.0
37.1 34.6
3,630.0 3,520.1 3,327.9 3,444.6 3,471.2
Stationary Combustion (MMT COz
2005 2010
0.5
0.3
+
0.1
1.7
0.3
0.2
0.2
1.0
1.1
+
0.2
0.4
0.5
4.1
0.1
0.3
0.6
3.1
0.1
+
0.1
+
0.
0
0
1.
0
0
0
0
1.
0
5
3
+
2
5
2
2
2
9
1
+
2
0.4
0.5
4.
0
0
0
3
0.
0
0.0 • 0
7.4 7.
0
0
3
6
1
1
+
1
+
0
1
2011
0.
0.
0.
1.
0.
0.
0.
0.
1.
0.
4
3
+
2
5
2
1
2
9
0
+
2
0.4
0.5
4.
0.
0.
0.
3.
0.
0.
0.
7.
0
0
3
6
2
1
+
1
+
0
1
2012
0.4
0.2
+
0.2
1.5
0.2
0.1
0.2
1.0
0.9
+
0.1
0.4
0.4
3.7
0.0
0.2
0.5
3.0
0.1
+
0.1
+
0.0
6.6
Eq.)
2013 2014
0.4
0.2
+
0.2
1.5
0.2
0.2
0.2
0.9
1.0
+
0.1
0.4
0.5
5.0
0.0
0.2
0.6
4.1
0.1
+
0.1
+
0.0
8.0
0
0
0
1
0
0
0
0
1
0
4
2
+
2
5
2
1
2
9
1
+
1
0.4
0.5
5
0
0
0
4
0
0
0
8
0
0
2
6
1
1
+
1
+
0
1
+ Does not exceed 0.05 MMT CO2 Eq.
Note: Totals may not sum due to independent rounding.
Table 3-11: NzO Emissions from
Sector/Fuel Type
Electricity Generation
Coal
Fuel Oil
Natural Gas
Wood
Industrial
Coal
Fuel Oil
Natural Gas
Wood
Commercial/Institutional
Coal
Fuel Oil
Natural Gas
1990
7.4
6.3 1
0.1
1.0
1
31
0.7 1
0.5
0.2
1.6 1
0.4 1
0.1
0.2
0.1
Stationary Combustion (MMT COz
• 2005
16.0
11.6
0.1
4.3
+
2.9
0.5
0.5
0.2
1.6
1
2010
18.5
12.5
+
5.9
+
2.5
0.4
0.4
0.2
1.4
0.3
0.0
0.1
0.1
2011
17.6
11.5
+
6.1
+
2.4
0.4
0.3
0.2
1.5
0.3
0.0
0.1
0.1
2012
17.8
10.2
+
7.5
+
2.4
0.4
0.3
0.2
1.5
0.3
0.0
0.1
0.1
Eq.)
2013
19.1
12.1
+
7.0
+
2.4
0.4
0.4
0.2
1.5
0.3
0.0
0.1
0.1
2014
19.6
12.4
+
7.2
+
2.4
0.4
0.3
0.2
1.5
0.3
0.0
0.1
0.1
Energy 3-13
-------
Wood
Residential
Coal
Fuel Oil
Natural Gas
Wood
U.S. Territories
Coal
Fuel Oil
Natural Gas
Wood
0.1
1.0
+
0.2
0.1
0.7
0.1
+
0.1
0.0
0.0
0.1
0.9
+
0.2
0.1
0.5
0.1
+
0.1
+
0.0
0.1
0.8
0.0
0.2
0.1
0.5
0.1
+
0.1
+
0.0
0.1
0.8
0.0
0.2
0.1
0.5
0.1
+
0.1
+
0.0
0.1
0.7
0.0
0.2
0.1
0.5
0.1
+
0.1
+
0.0
0.1
1.0
0.0
0.2
0.1
0.7
0.1
+
0.1
+
0.0
0.1
1.0
0.0
0.2
0.1
0.7
0.1
+
0.1
+
0.0
Total
11.9
20.2
22.2
21.3
21.4
+ Does not exceed 0.05 MMT CO2 Eq.
Note: Totals may not sum due to independent rounding.
22.9
23.4
i Electricity Generation
2 The process of generating electricity is the single largest source of CC>2 emissions in the United States, representing
3 37 percent of total CCh emissions from all CCh emissions sources across the United States. Methane and N2O
4 accounted for a small portion of emissions from electricity generation, representing less than 0.1 percent and 1.0
5 percent, respectively. Electricity generation also accounted for the largest share of CCh emissions from fossil fuel
6 combustion, approximately 39.2 percent in 2014. Methane and N2O from electricity generation represented 4.2 and
7 49.3 percent of total methane and N2O emissions from fossil fuel combustion in 2014, respectively. Electricity was
8 consumed primarily in the residential, commercial, and industrial end-use sectors for lighting, heating, electric
9 motors, appliances, electronics, and air conditioning (see Figure 3-9). Electricity generators, including those using
10 low-CO2 emitting technologies, relied on coal for approximately 39 percent of their total energy requirements in
11 2014. Recently an increase in the carbon intensity of fuels consumed to generate electricity has occurred due to an
12 increase in coal consumption, and decreased natural gas consumption and other generation sources. Total U.S.
13 electricity generators used natural gas for approximately 27 percent of their total energy requirements in 2014 (El A
14 2015b).
15 Figure 3-9: Electricity Generation Retail Sales by End-Use Sector (Billion kWh)
16
17
18
19
20
21
1,500
1,400
1,300 -
I 1,200
i 1,100
(0
1,000 -
900
800
Residential
Commercial
Industrial
gi-H
O^
oo
fNJ(N
o
rsl
oo
fM(N
o
rsl
The electric power industry includes all power producers, consisting of both regulated utilities and non-utilities (e.g.
independent power producers, qualifying co-generators, and other small power producers). For the underlying
energy data used in this chapter, the Energy Information Administration (EIA) places electric power generation into
three functional categories: the electric power sector, the commercial sector, and the industrial sector. The electric
power sector consists of electric utilities and independent power producers whose primary business is the production
3-14 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 of electricity, while the other sectors consist of those producers that indicate their primary business is something
2 other than the production of electricity.14
3 The industrial, residential, and commercial end-use sectors, as presented in Table 3-8, were reliant on electricity for
4 meeting energy needs. The residential and commercial end-use sectors were especially reliant on electricity
5 consumption for lighting, heating, air conditioning, and operating appliances. Electricity sales to the residential and
6 commercial end-use sectors in 2014 increased approximately 0.9 percent and 1.1 percent, respectively. The trend in
7 the residential and commercial sectors can largely be attributed to colder, more energy-intensive winter conditions
8 compared to 2013. Electricity sales to the industrial sector in 2014 increased approximately 1.2 percent. Overall, in
9 2014, the amount of electricity generated (in kWh) increased approximately 1.1 percent relative to the previous year,
10 while CO2 emissions from the electric power sector increased by 0.1 percent. The increase in CCh emissions, despite
11 the relatively larger increase in electricity generation was a result of a slight decrease in the consumption of coal and
12 natural gas for electricity generation by 0.1 percent and 0.2 percent, respectively, in 2014, and an increase in the
13 consumption of petroleum for electricity generation by 15.8 percent.
14 Industrial Sector
15 Industrial sector CCh, CH4, and N2O, emissions accounted for 16, 15, and 6 percent of CCh, CH4, and N2O,
16 emissions from fossil fuel combustion, respectively. COa, CH4, and N2O emissions resulted from the direct
17 consumption of fossil fuels for steam and process heat production.
18 The industrial sector, per the underlying energy consumption data from El A, includes activities such as
19 manufacturing, construction, mining, and agriculture. The largest of these activities in terms of energy consumption
20 is manufacturing, of which six industries—Petroleum Refineries, Chemicals, Paper, Primary Metals, Food, and
21 Nonmetallic Mineral Products—represent the vast majority of the energy use (EIA 2015a and EIA 2009b).
22 In theory, emissions from the industrial sector should be highly correlated with economic growth and industrial
23 output, but heating of industrial buildings and agricultural energy consumption are also affected by weather
24 conditions.15 In addition, structural changes within the U.S. economy that lead to shifts in industrial output away
25 from energy-intensive manufacturing products to less energy-intensive products (e.g., from steel to computer
26 equipment) also have a significant effect on industrial emissions.
27 From 2013 to 2014, total industrial production and manufacturing output increased by 3.7 percent (FRB 2015).
28 Over this period, output increased across production indices for Food, Petroleum Refineries, Chemicals, Primary
29 Metals, and Nonmetallic Mineral Products, and decreased slightly for Paper (see Figure 3-10). Through EPA's
30 Greenhouse Gas Reporting Program (GHGRP), industrial trends can be discerned from the overall EIA industrial
31 fuel consumption data used for these calculations. For example, from 2013 to 2014 the underlying EIA data showed
32 increased consumption of natural gas and a decrease in petroleum fuels in the industrial sector. EPA's GHGRP data
33 highlights that chemical manufacturing and nonmetallic mineral products were contributors to these trends.16
34
Utilities primarily generate power for the U.S. electric grid for sale to retail customers. Nonutilities produce electricity for
their own use, to sell to large consumers, or to sell on the wholesale electricity market (e.g., to utilities for distribution and resale
to customers).
Some commercial customers are large enough to obtain an industrial price for natural gas and/or electricity and are
consequently grouped with the industrial end-use sector in U.S. energy statistics. These misclassifications of large commercial
customers likely cause the industrial end-use sector to appear to be more sensitive to weather conditions.
Further details on industrial sector combustion emissions are provided by EPA's GHGRP
().
Energy 3-15
-------
Figure 3-10: Industrial Production Indices (Index 2007=100)
150
140
130
120
110
100
90
80
70
60
Total Industrial
Total excluding Computers, Communications
Equipment, and Semiconductors
Foods
150
140
130
120
110
100
90
SO
70
Stone, Clay & Glass
Products
Chemicals
CTiCTiOlClCTiClClCTiClCTiOOOOOOOOOOOOOOO
3 Despite the growth in industrial output (64 percent) and the overall U.S. economy (78 percent) from 1990 to 2014,
4 CO2 emissions from fossil fuel combustion in the industrial sector decreased by 3.4 percent over the same time
5 series. A number of factors are believed to have caused this disparity between growth in industrial output and
6 decrease in industrial emissions, including: (1) more rapid growth in output from less energy-intensive industries
7 relative to traditional manufacturing industries, and (2) energy-intensive industries such as steel are employing new
8 methods, such as electric arc furnaces, that are less carbon intensive than the older methods. In 2014, CO2, CH4, and
9 N2O emissions from fossil fuel combustion and electricity use within the industrial end-use sector totaled 1,417.5
10 MMT CO2 Eq., or approximately 0.1 percent above 2013 emissions.
11 Residential and Commercial Sectors
12 Residential and commercial sector CO2 emissions accounted for 7 and 4 percent of CO2 emissions from fossil fuel
13 combustion, CH4 emissions accounted for 49 and 11 percent of CH4 emissions from fossil fuel combustion, and N2O
14 emissions accounted for 2 and 1 percent of N2O emissions from fossil fuel combustion, respectively. Emissions
15 from these sectors were largely due to the direct consumption of natural gas and petroleum products, primarily for
16 heating and cooking needs. Coal consumption was a minor component of energy use in both of these end-use
17 sectors. In 2014, CO2, CH4, and N2O emissions from fossil fuel combustion and electricity use within the residential
18 and commercial end-use sectors were 1,093.6 MMT CO2 Eq. and 946.4 MMT CO2 Eq., respectively. Total CO2,
19 CH4, and N2O emissions from fossil fuel combustion and electricity use within the residential and commercial end-
20 use sectors increased by 1.5 and 1.4 percent from 2013 to 2014, respectively.
3-16 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Emissions from the residential and commercial sectors have generally been increasing since 1990, and are often
2 correlated with short-term fluctuations in energy consumption caused by weather conditions, rather than prevailing
3 economic conditions. In the long-term, both sectors are also affected by population growth, regional migration
4 trends, and changes in housing and building attributes (e.g., size and insulation).
5 In 2014, combustion emissions from natural gas consumption represent 80 and 82 percent of the direct fossil fuel
6 CO2 emissions from the residential and commercial sectors, respectively. Natural gas combustion CC>2 emissions
7 from the residential and commercial sectors in 2014 increased by 4.3 percent and 5.6 percent from 2013 levels,
8 respectively.
9 U.S. Territories
10 Emissions from U.S. Territories are based on the fuel consumption in American Samoa, Guam, Puerto Rico, U.S.
11 Virgin Islands, Wake Island, and other U.S. Pacific Islands. As described in the Methodology section for CCh from
12 fossil fuel combustion, this data is collected separately from the sectoral-level data available for the general
13 calculations. As sectoral information is not available for U.S. Territories, CCh, CH4, and N2O emissions are not
14 presented for U.S. Territories in the tables above, though the emissions will include some transportation and mobile
15 combustion sources.
16 Transportation Sector and Mobile Combustion
17 This discussion of transportation emissions follows the alternative method of presenting combustion emissions by
18 allocating emissions associated with electricity generation to the transportation end-use sector, as presented in Table
19 3-8. For direct emissions from transportation (i.e., not including emissions associated with the sector's electricity
20 consumption), please see Table 3-7.
21 Transportation End-Use Sector
22 The transportation end-use sector accounted for 1,759.9 MMT CCh Eq. in 2014, which represented 33 percent of
23 CO2 emissions, 20 percent of CEU emissions, and 41 percent of N2O emissions from fossil fuel combustion,
24 respectively.17 Fuel purchased in the United States for international aircraft and marine travel accounted for an
25 additional 104.2 MMT CCh Eq. in 2014; these emissions are recorded as international bunkers and are not included
26 in U.S. totals according to UNFCCC reporting protocols.
27 From 1990 to 2014, transportation emissions from fossil fuel combustion rose by 14 percent due, in large part, to
28 increased demand for travel with limited gains in fuel efficiency for much of this time period. The number of vehicle
29 miles traveled (VMT) by light-duty motor vehicles (passenger cars and light-duty trucks) increased 37 percent from
30 1990 to 2014, as a result of a confluence of factors including population growth, economic growth, urban sprawl,
31 and periods of low fuel prices.
32 From 2013 to 2014, CC>2 emissions from the transportation end-use sector increased by 1.4 percent.18 The increase
33 in emissions can largely be attributed to small increases in VMT and fuel use across many on-road transportation
34 modes. Commercial aircraft emissions have decreased 18 percent since 2007.19 Decreases in jet fuel emissions
35 (excluding bunkers) since 2007 are due in part to improved operational efficiency that results in more direct flight
36 routing, improvements in aircraft and engine technologies to reduce fuel burn and emissions, and the accelerated
37 retirement of older, less fuel efficient aircraft.
38 Almost all of the energy consumed for transportation was supplied by petroleum-based products, with more than
39 half being related to gasoline consumption in automobiles and other highway vehicles. Other fuel uses, especially
40 diesel fuel for freight trucks and jet fuel for aircraft, accounted for the remainder. The primary driver of
41 transportation-related emissions was CCh from fossil fuel combustion, which increased by 16 percent from 1990 to
17 Note that these totals include CCh, CELt and N2O emissions from some sources in the U.S. Territories (ships and boats,
recreational boats, non-transportation mobile sources) and CH4 and N2O emissions from transportation rail electricity.
18 Note that this value does not include lubricants.
19 Commercial aircraft, as modeled in FAA's AEDT, consists of passenger aircraft, cargo, and other chartered flights.
Energy 3-17
-------
1 2014. Annex 3.2 presents the total emissions from all transportation and mobile sources, including CO2, N2O, CH4,
2 and HFCs.
3 Transportation Fossil Fuel Combustion CO2 Emissions
4 Domestic transportation CO2 emissions increased by 16 percent (244.7 MMT CO2) between 1990 and 2014, an
5 annualized increase of 0.7 percent. Among domestic transportation sources, light-duty vehicles (including
6 passenger cars and light-duty trucks) represented 60 percent of CO2 emissions from fossil fuel combustion, medium-
7 and heavy-duty trucks and buses 24 percent, commercial aircraft 7 percent, and other sources 9 percent. See Table
8 3-12 for a detailed breakdown of transportation CO2 emissions by mode and fuel type.
9 Almost all of the energy consumed by the transportation sector is petroleum-based, including motor gasoline, diesel
10 fuel, jet fuel, and residual oil. Carbon dioxide emissions from the combustion of ethanol and biodiesel for
11 transportation purposes, along with the emissions associated with the agricultural and industrial processes involved
12 in the production of biofuel, are captured in other Inventory sectors.20 Ethanol consumption from the transportation
13 sector has increased from 0.7 billion gallons in 1990 to 12.9 billion gallons in 2014, while biodiesel consumption
14 has increased from 0.01 billion gallons in 2001 to 1.4 billion gallons in 2014. For further information, see the
15 section on biofuel consumption at the end of this chapter and Table A-93 in Annex 3.2.
16 Carbon dioxide emissions from passenger cars and light-duty trucks totaled 1,047.0 MMT CO2 in 2014, an increase
17 of 10 percent (96.6 MMT CO2) from 1990 due, in large part, to increased demand for travel as fleetwide light-duty
18 vehicle fuel economy was relatively stable (average new vehicle fuel economy declined slowly from 1990 through
19 2004 and then increased more rapidly from 2005 through 2014). Carbon dioxide emissions from passenger cars and
20 light-duty trucks peaked at 1,181.1 MMT CO2 in 2004, and since then have declined about 11 percent. The decline
21 in new light-duty vehicle fuel economy between 1990 and 2004 (Figure 3-11) reflected the increasing market share
22 of light-duty trucks, which grew from about 30 percent of new vehicle sales in 1990 to 48 percent in 2004. Starting
23 in 2005, the rate of VMT growth slowed while average new vehicle fuel economy began to increase. Average new
24 vehicle fuel economy has improved almost every year since 2005, and the truck share has decreased to about 41
25 percent of new vehicles in model year 2014 (EPA 2015a).
26 Medium- and heavy-duty truck CO2 emissions increased by 75 percent from 1990 to 2014. This increase was
27 largely due to a substantial growth in medium- and heavy-duty truck VMT, which increased by 94 percent between
28 1990 and2014.21 Carbon dioxide from the domestic operation of commercial aircraft increased by 5 percent (5.3
29 MMT CO2) from 1990 to 2014.22 Across all categories of aviation, excluding international bunkers, CO2 emissions
30 decreased by 20 percent (37.3 MMT CO2) between 1990 and 2014.23 This includes a 56 percent (19.6 MMT CO2)
31 decrease in CO2 emissions from domestic military operations.
32 Transportation sources also produce CH4 and N2O; these emissions are included in Table 3-13 and Table 3-14 and in
33 the "Mobile Combustion" Section. Annex 3.2 presents total emissions from all transportation and mobile sources,
34 including CO2, CH4, N2O, and HFCs.
35
Biofuel estimates are presented in the Energy chapter for informational purposes only, in line with IPCC methodological
guidance and UNFCCC reporting obligations. Net carbon fluxes from changes in biogenic carbon reservoirs in croplands are
accounted for in the estimates for Land Use, Land-Use Change, and Forestry (see Chapter 6). More information and additional
analyses on bio fuels are available at EPA's "Renewable Fuels: Regulations & Standards;" See
.
While FHWA data shows consistent growth in medium- and heavy-duty truck VMT over the 1990 to 2014 time period, part of
the growth reflects a method change for estimating VMT starting in 2007. This change in methodology in FHWA's VM-1 table
resulted in large changes in VMT by vehicle class, thus leading to a shift in VMT and emissions among on-road vehicle classes
in the 2007 to 2014 time period. During the time period prior to the method change (1990-2006), VMT for medium- and heavy-
duty trucks increased by 51 percent.
22 Commercial aircraft, as modeled in FAA's AEDT, consists of passenger aircraft, cargo, and other chartered flights.
Includes consumption of jet fuel and aviation gasoline. Does not include aircraft bunkers, which are not included in national
emission totals, in line with IPCC methodological guidance and UNFCCC reporting obligations.
3-18 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Figure 3-11: Sales-Weighted Fuel Economy of New Passenger Cars and Light-Duty Trucks,
2 1990-2014 (miles/gallon)
3
4
5
6
_o
"re
(D
25.0 n
24.5 -
24.0 -
23.5 -
23.0 -
22.5 -
22.0 -
21.5
21.0 -
20.5
20.0
19.5 -
19.0
18.5 -
18.0 -
,-
a
IN
LT)
8!
§IH fM n •
§ § S :
rM rM fM rM i
Model Year
rM
p p p p
rM rv] fM rM
p p
rM rvl
Source: EPA (2015)
Figure 3-12: Sales of New Passenger Cars and Light-Duty Trucks, 1990-2014 (Percent)
100%
8 75%
Ł 50%
25% H
0%
Passenger Cars
Light-Duty Trucks
o o
rM rM
o o o
rM rM rM
o o
rM rM
o o
rM rM
7
8
9
10
11
Source: EPA (2015)
Table 3-12: COz Emissions from Fossil Fuel Combustion in Transportation End-Use Sector
(MMTCO2Eq.)
Fuel/Vehicle Type
Gasoline0
Passenger Cars
Light-Duty Trucks
1990
983,
621
309
,5
.4
.1
2005
1,183.7
655.9
477.2
2010"
1,092.
738.
295.
,5
,2
,0
2011
1,068.8
732.8
280.4
2012
1,064.7
731.4
277.4
2013
1,065.6
731.4
277.7
2014
1,083.9
733.6
293.5
Energy 3-19
-------
Medium- and Heavy -Duty Trucksb
Buses
Motorcycles
Recreational Boats'1
Distillate Fuel Oil (Diesel) c''
Passenger Cars
Light-Duty Trucks
Medium- and Heavy -Duty Trucks'5
Buses
Rail
Recreational Boats
Ships and Other Boats'
International Bunker Fuels'1
Jet Fuel
Commercial Aircraft6
Military Aircraft
General Aviation Aircraft
International Bunker Fuels'1
International Bunker Fuels from
Commercial Aviation
Aviation Gasoline
General Aviation Aircraft
Residual Fuel Oil
Ships and Other Boats'
International Bunker Fuels'1
Natural Gas
Passenger Cars
Light-Duty Trucks
Buses
Pipelinef
LPG
Light-Duty Trucks
Medium- and Heavy -Duty Trucks'5
Buses
Electricity
Rail
EthanoW
Total
Total (Including Bunkers)"1
38.7 1
L7
12.2 1
262.9
7.9 1
11.5
190.5
8.0
35.5
2.0 1
7.5 1
11.7 1
184.2
109.9
35.0
39.4 1
38.0 1
30.0 1
"1
22.6 1
22.6
53.7
36.0
•
36.0 1
1.4 1
0.6 1
0.8 1
3.0 1
3.0
4.1
1,496.8
1,600.3
34.8 1
0.4 1
1.6
13.9 1
457.5
4.2 1
25.8
360.2
10.6
45.5
8.0 1
9.4
189.3
132.7
19.4
37.3
60.1
55.6 1
24 1
24 1
19.3 1
19.3 1
43.6 1
33.1 1
0.8 1
32.2 1
1.7 1
1.3 1
0.4 1
+ 1
4.7 1
4.7
22.4
1,891.8
2,004.9
42.3
0.7
3.6
12.6
422.0
3.7
12.5
342.7
13.5
38.6
3.6
7.4
9.5
151.5
113.3
13.6
24.6
61.0
57.4
1.9
1.9
20.4
20.4
46.5
38.1
1.1
37.1
1.8
1.3
0.6
4.5
4.5
77.J
1,732.7
1,849.7
38.9
0.7
3.6
12.4
430.0
4.1
13.0
344.4
14.4
40.4
3.6
10.1
7.9
146.6
114.6
11.6
20.4
64.8
61.7
1.9
1.9
19.4
19.4
38.9
38.9
1.1
37.8
2.1
1.5
0.6
4.3
4.3
77.5
1,711.9
1,823.6
38.7
0.8
4.1
12.3
427.5
4.1
12.9
344.4
15.4
39.5
3.7
7.5
6.8
143.4
113.3
12.1
18.0
64.5
61.4
1.7
1.7
15.8
15.8
34.5
41.3
1.0
40.3
2.3
1.6
0.7
3.9
3.9
77.5
1,700.6
1,806.4
39.5
0.8
3.9
12.3
433.9
4.1
12.9
350.0
15.5
40.1
3.7
7.5
5. 6
147.1
114.3
11.0
21.8
65.7
62.8
1.5
1.5
15.1
15.1
28.5
47.0
1.1
45.9
2.7
1.9
0.8
4.0
4.0
73.4
1,717.0
1,816.8
40.0
0.9
3.8
12.2
447.6
4.1
13.9
361.3
16.6
41.7
3.8
6.2
6.1
148.6
115.2
15.4
18.0
69.4
66.3
1.5
1.5
5.5
5.5
27.7
47.6
1.1
46.5
2.7
1.9
0.8
4.1
4.1
74.7
1,741.5
1,844.7
1 Note: This table does not include emissions from non-transportation mobile sources, such as agricultural equipment and
2 construction/mining equipment; it also does not include emissions associated with electricity consumption by pipelines or
3 lubricants used in transportation. In addition, this table does not include CO2 emissions from U.S. Territories, since these are
4 covered in a separate chapter of the Inventory.
5 Note: Totals may not sum due to independent rounding.
6 a In 2011 FHWA changed its methods for estimating vehicle miles traveled (VMT) and related data. These methodological
7 changes included how vehicles are classified, moving from a system based on body-type to one that is based on wheelbase.
8 These changes were first incorporated for the 2010 Inventory and apply to the 2007-14 time period. This resulted in large
9 changes in VMT and fuel consumption data by vehicle class, thus leading to a shift in emissions among on-road vehicle classes.
10 b Includes medium- and heavy-duty trucks over 8,500 Ibs.
11 ° Gasoline and diesel highway vehicle fuel consumption estimates are based on data from FHWA Highway Statistics Table VM-1
12 and MF-27 (FHWA 1996 through 2015). These fuel consumption estimates are combined with estimates of fuel shares by
13 vehicle type from DOE's TEDB Annex Tables A. 1 through A.6 (DOE 1993 through 2015). TEDB data for 2014 has not been
14 published yet, therefore 2013 data is used as a proxy.
15 d Official estimates exclude emissions from the combustion of both aviation and marine international bunker fuels; however,
16 estimates including international bunker fuel-related emissions are presented for informational purposes.
17 e Commercial aircraft, as modeled in FAA's AEDT, consists of passenger aircraft, cargo, and other chartered flights.
18 f Pipelines reflect CO2 emissions from natural gas powered pipelines transporting natural gas.
3-20 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 gEthanol estimates are presented for informational purposes only. See Section 3.10 of this chapter and the estimates in Land Use,
2 Land-Use Change, and Forestry (see Chapter 6), in line with IPCC methodological guidance and UNFCCC reporting
3 obligations, for more information on ethanol.
4 h In 2015, EPA incorporated the NONROAD2008 model into MOVES2014. This year's inventory uses the NONROAD
5 component of MOVES2014a for years 1999 through 2014. This update resulted in small changes (<2%) to the 1999-2013 time
6 series for NONROAD fuel consumption due to differences in the gasoline and diesel default fuel densities used within the
7 model iterations.
8 ' Note that updates to the distillate fuel oil heat content data from EIA for years 1993 through present resulted in changes to the
9 time series for energy consumption and emissions compared to the previous Inventory.
10 J Note that large year over year fluctuations in emission estimates partially reflect nature of data collection for these sources.
11 + Less than 0.05 MMT CO2 Eq.
12 Mobile Fossil Fuel Combustion CH4 andN2O Emissions
13 Mobile combustion includes emissions of CH4 and N2O from all transportation sources identified in the U.S.
14 Inventory with the exception of pipelines and electric locomotives;24 mobile sources also include non-transportation
15 sources such as construction/mining equipment, agricultural equipment, vehicles used off-road, and other sources
16 (e.g., snowmobiles, lawnmowers, etc.). 25 Annex 3.2 includes a summary of all emissions from both transportation
17 and mobile sources. Table 3-13 and Table 3-14 provide mobile fossil fuel CH4 and N2O emission estimates in MMT
18 CO2Eq.26
19 Mobile combustion was responsible for a small portion of national CH4 emissions (0.3 percent) but was the fourth
20 largest source of U.S. N2O emissions (4.0 percent). From 1990 to 2014, mobile source CH4 emissions declined by
21 64 percent, to 2.0 MMT CO2 Eq. (82 kt CH4), due largely to control technologies employed in on-road vehicles
22 since the mid-1990s to reduce CO, NOX, NMVOC, and CH4 emissions. Mobile source emissions of N2O decreased
23 by 60 percent, to 16.3 MMT CO2 Eq. (55 kt N2O). Earlier generation control technologies initially resulted in
24 higher N2O emissions, causing a 28 percent increase in N2O emissions from mobile sources between 1990 and 1997.
25 Improvements in later-generation emission control technologies have reduced N2O output, resulting in a 69 percent
26 decrease in mobile source N2O emissions from 1997 to 2014 (Figure 3-13). Overall, CH4 and N2O emissions were
27 predominantly from gasoline-fueled passenger cars and light-duty trucks.
28
24 Emissions of CH4 from natural gas systems are reported separately. More information on the methodology used to calculate
these emissions are included in this chapter and Annex 3.4.
25 See the methodology sub-sections of the CO2 from Fossil Fuel Combustion and CH4 and N2O from Mobile Combustion
sections of this chapter. Note thatN2O and CH4 emissions are reported using different categories than CO2. CO2 emissions are
reported by end-use sector (Transportation, Industrial, Commercial, Residential, U.S Territories), and generally adhere to a top-
down approach to estimating emissions. CO2 emissions from non-transportation sources (e.g., lawn and garden equipment, farm
equipment, construction equipment) are allocated to their respective end-use sector (i.e., construction equipment CO2 emissions
are included in the Commercial end-use sector instead of the Transportation end-use sector). CH4 and N2O emissions are
reported using the "Mobile Combustion" category, which includes non-transportation mobile sources. CH4 and N2O emissions
estimates are bottom-up estimates, based on total activity (fuel use, VMT) and emissions factors by source and technology type.
These reporting schemes are in accordance with IPCC guidance. For informational purposes only, CO2 emissions from non-
transportation mobile sources are presented separately from their overall end-use sector in Annex 3.2.
26 See Annex 3.2 for a complete time series of emission estimates for 1990 through 2014.
Energy 3-21
-------
2
3
4
Figure 3-13: Mobile Source CH4 and NzO Emissions (MMT COz Eq.)
60
50
. 40 -J
CT
LLJ
8
30 -
20 -
10
N-.O
CH4
gi-i
C^
§ S
Table 3-13: CH4 Emissions from Mobile Combustion (MMT COz Eq.)
Fuel Type/Vehicle Type3 1990 2005 2010 2011 2012 2013 2014
Gasoline On-Roadb
Passenger Cars
Light-Duty Trucks
Medium- and Heavy-Duty
Trucks and Buses
Motorcycles
Diesel On-Roadb
Passenger Cars
Light-Duty Trucks
Medium- and Heavy-Duty
Trucks and Buses
Alternative Fuel On-Roadg
Non-Road11
Ships and Boats
Railf
Aircraft
Agricultural Equipment0
Construction/Mining
Equipment4
Other6
5.2
3.2
1.7
0.3
+
+
+
+
+
+
0.4
+
0.1
0.1
0.1
0.1
0.1
2.2
1.2
0.8
0.1
+
+
+
+
+
+
0.5
+
0.1
0.1
0.2
0.1
0.1
1.7
1.2
0.4
0.1
+
+
+
+
+
+
0.5
+
0.1
+
0.2
0.1
0.1
1.6
1.2
0.4
0.1
+
+
+
+
+
+
0.5
+
0.1
+
0.2
0.1
0.1
1.5
1.1
0.3
0.1
+
+
+
+
+
+
0.6
+
0.1
+
0.2
0.1
0.1
1.5
1.0
0.3
0.1
+
+
+
+
+
+
0.6
+
0.1
+
0.2
0.1
0.1
1.4
1.0
0.3
0.1
+
+
+
+
+
+
0.6
+
0.1
+
0.2
0.1
0.1
Total
5.6
2.7
2.3
2.2
2.2
2.1
2.0
Note: Totals may not sum due to independent rounding.
Note: In 2011, FHWA changed its methods for estimating vehicle miles traveled (VMT) and related data. These
methodological changes included how vehicles are classified, moving from a system based on body-type to one that is
based on wheelbase. These changes were first incorporated for the 1990-2010 Inventory and apply to the 2007
through 2014 time period. This resulted in large changes in VMT and fuel consumption data by vehicle class, thus
leading to a shift in emissions among on-road vehicle classes.
a See Annex 3.2 for definitions of on-road vehicle types.
b Gasoline and diesel highway vehicle mileage are based on data from FHWA Highway Statistics Table VM-1
(FHWA 1996 through 2015). These mileage consumption estimates are combined with estimates of fuel shares by
vehicle type from DOE's TEDB Annex Tables A.I through A.6 (DOE 1993 through 2015). TEDB data for 2014
has not been published yet, therefore 2013 data is used as a proxy.
3-22 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
0 Includes equipment, such as tractors and combines, as well as fuel consumption from trucks that are used off-road in
agriculture.
d Includes equipment, such as cranes, dumpers, and excavators, as well as fuel consumption from trucks that are used
off-road in construction.
e "Other" includes snowmobiles and other recreational equipment, logging equipment, lawn and garden equipment,
railroad equipment, airport equipment, commercial equipment, and industrial equipment, as well as fuel
consumption from trucks that are used off-road for commercial/industrial purposes.
f Rail emissions do not include emissions from electric powered locomotives. Class II and Class III diesel
consumption data for 2014 is not available yet, therefore 2013 data is used as a proxy.
g In 2015, EIA changed its methods for estimating AFV fuel consumption. These methodological changes included
how vehicle counts are estimated, moving from estimates based on modeling to one that is based on survey data.
EIA now publishes data about fuel use and number of vehicles for only four types of AFV fleets: federal
government, state government, transit agencies, and fuel providers. These changes were first incorporated in this
year's inventory and apply to the 1990-2014 time period. This resulted in large reductions in AFV VMT, thus
leading to a shift in VMT to conventional on-road vehicle classes.
hln 2015, EPA incorporated the NONROAD2008 model into MOVES2014. This year's inventory uses the
NONROAD component of MOVES2014a for years 1999 through 2014. This update resulted in small changes
(<2%) to the 1999-2013 time series for NONROAD fuel consumption due to differences in the gasoline and diesel
default fuel densities used within the model iterations.
+ Less than 0.05 MMT CO2 Eq.
Table 3-14: NzO Emissions from Mobile Combustion (MMT COz Eq.)
Fuel Type/Vehicle Type3
Gasoline On-Roadb
Passenger Cars
Light-Duty Trucks
Medium- and Heavy-Duty
Trucks and Buses
Motorcycles
Diesel On-Roadb
Passenger Cars
Light-Duty Trucks
Medium- and Heavy-Duty
Trucks and Buses
Alternative Fuel On-Roadg
Non-Road11
Ships and Boats
Railf
Aircraft
Agricultural Equipment0
Construction/Mining
Equipment*
Other6
Total
1990 2005 2010
37.5 29.9 19.2
24.1 15.9( 12.9
12.8
0.5
+
0.2
+
+
0.2
+
3.5
0.6
0.3
1.7
0.2
0.3
0.4
13.2
0.8
+
0.3
+
+
0.3
+
4.1
0.6
0.3
1.8
0.4
0.5
5.5
0.8
+
0.4
+
+
0.4
+
4.0
0.8
0.3
1.4
0.4
0.6
0.6 0.6
41.2 34.4 23.6
2011
18.0
12.3
5.0
0.7
+
0.4
+
+
0.4
0.1
4.0
0.8
0.3
1.4
0.4
0.6
0.6
22.4
2012
15.7
10.7
4.4
0.6
+
0.4
+
+
0.4
0.1
3.9
0.7
0.3
1.3
0.4
0.6
0.6
20.0
2013
13.8
9.3
3.9
0.6
+
0.4
+
+
0.4
0.1
3.9
0.7
0.3
1.4
0.4
0.6
0.6
18.2
2014
12.1
7.9
3.6
0.5
+
0.4
+
+
0.4
0.1
3.8
0.5
0.3
1.4
0.4
0.6
0.6
16.3
Note: Totals may not sum due to independent rounding.
Note: In 2011, FLTWA changed its methods for estimating vehicle miles traveled (VMT) and related data. These
methodological changes included how vehicles are classified, moving from a system based on body type to one that is
based on wheelbase. These changes were first incorporated for the 1990-2010 Inventory and apply to the 2007 through
2014 time period. This resulted in large changes in VMT and fuel consumption data by vehicle class, thus leading to a
shift in emissions among on-road vehicle classes.
a See Annex 3.2 for definitions of on-road vehicle types.
b Gasoline and diesel highway vehicle mileage are based on data from FHWA Highway Statistics Table VM-1 (FHWA
1996 through 2015). These mileage consumption estimates are combined with estimates of fuel shares by vehicle type
from DOE's TEDB Annex Tables A. 1 through A.6 (DOE 1993 through 2015). TEDB data for 2014 has not been
published yet, therefore 2013 data is used as a proxy.
0 Includes equipment, such as tractors and combines, as well as fuel consumption from trucks that are used off-road in
agriculture.
Energy 3-23
-------
d Includes equipment, such as cranes, dumpers, and excavators, as well as fuel consumption from trucks that are used off-
road in construction.
e "Other" includes snowmobiles and other recreational equipment, logging equipment, lawn and garden equipment,
railroad equipment, airport equipment, commercial equipment, and industrial equipment, as well as fuel consumption
from trucks that are used off-road for commercial/industrial purposes.
f Rail emissions do not include emissions from electric powered locomotives. Class II and Class III diesel consumption
data for 2014 is not available yet, therefore 2013 data is used as a proxy.
g In 2015, EIA changed its methods for estimating AFV fuel consumption. These methodological changes included how
vehicle counts are estimated, moving from estimates based on modeling to one that is based on survey data. EIA now
publishes data about fuel use and number of vehicles for only four types of AFV fleets: federal government, state
government, transit agencies, and fuel providers. These changes were first incorporated in this year's inventory and
apply to the 1990-2014 time period. This resulted in large reductions in AFV VMT, thus leading to a shift in VMT to
conventional on-road vehicle classes.
hln 2015, EPA incorporated the NONROAD2008 model into MOVES2014. This year's inventory uses the NONROAD
component of MOVES2014a for years 1999 through 2014. This update resulted in small changes (<2%) to the 1999-
2013 time series for NONROAD fuel consumption due to differences in the gasoline and diesel default fuel densities
used within the model iterations.
1 Updates to the jet fuel heat content used in the mobile N2O emissions estimates for years 1990 through present resulted
in small changes to the time series emissions compared to the previous Inventory.
+ Less than 0.05 MMT CO2 Eq.
2 CO2 from Fossil Fuel Combustion
3 Methodology
4 The methodology used by the United States for estimating CC>2 emissions from fossil fuel combustion is
5 conceptually similar to the approach recommended by the IPCC for countries that intend to develop detailed,
6 sectoral-based emission estimates in line with a Tier 2 method in the 2006 IPCC Guidelines for National
7 Greenhouse Gas Inventories (IPCC 2006).27 The use of the most recently published calculation methodologies by
8 the IPCC, as contained in the 2006 IPCC Guidelines, is considered to improve the rigor and accuracy of this
9 Inventory and is fully in line with IPCC Good Practice Guidance. A detailed description of the U.S. methodology is
10 presented in Annex 2.1, and is characterized by the following steps:
11 1. Determine total fuel consumption by fuel type and sector. Total fossil fuel consumption for each year is
12 estimated by aggregating consumption data by end-use sector (e.g., commercial, industrial, etc.), primary
13 fuel type (e.g., coal, petroleum, gas), and secondary fuel category (e.g., motor gasoline, distillate fuel oil,
14 etc.). Fuel consumption data for the United States were obtained directly from the EIA of the U.S.
15 Department of Energy (DOE), primarily from the Monthly Energy Review and published supplemental
16 tables on petroleum product detail (EIA 2015a). The EIA does not include territories in its national energy
17 statistics, so fuel consumption data for territories were collected separately from EIA's International
18 Energy Statistics (EIA 2014) and Jacobs (2010).28
19
20 For consistency of reporting, the IPCC has recommended that countries report energy data using the
21 International Energy Agency (IEA) reporting convention and/or IEA data. Data in the IEA format are
22 presented "top down"—that is, energy consumption for fuel types and categories are estimated from energy
23 production data (accounting for imports, exports, stock changes, and losses). The resulting quantities are
24 referred to as "apparent consumption." The data collected in the United States by EIA on an annual basis
25 and used in this Inventory are predominantly from mid-stream or conversion energy consumers such as
26 refiners and electric power generators. These annual surveys are supplemented with end-use energy
27 consumption surveys, such as the Manufacturing Energy Consumption Survey, that are conducted on a
27 The IPCC Tier 3B methodology is used for estimating emissions from commercial aircraft.
Fuel consumption by U.S. territories (i.e., American Samoa, Guam, Puerto Rico, U.S. Virgin Islands, Wake Island, and other
U.S. Pacific Islands) is included in this report and contributed total emissions of 41.2 MMT CO2 Eq. in 2013.
3-24 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 periodic basis (every four years). These consumption data sets help inform the annual surveys to arrive at
2 the national total and sectoral breakdowns for that total.29
3 Also, note that U.S. fossil fuel energy statistics are generally presented using gross calorific values (GCV)
4 (i.e., higher heating values). Fuel consumption activity data presented here have not been adjusted to
5 correspond to international standards, which are to report energy statistics in terms of net calorific values
6 (NCV) (i.e., lower heating values).30
7 2. Subtract uses accounted for in the Industrial Processes and Product Use chapter. Portions of the fuel
8 consumption data for seven fuel categories—coking coal, distillate fuel, industrial other coal, petroleum
9 coke, natural gas, residual fuel oil, and other oil—were reallocated to the Industrial Processes and Product
10 Use chapter, as they were consumed during non-energy related industrial activity. To make these
11 adjustments, additional data were collected from AISI (2004 through 2013), Coffeyville (2014), U.S.
12 Census Bureau (2011), EIA (2015a), USGS (1991 through 2011), USGS (1994 through 2011), USGS
13 (1995, 1998, 2000 through 2002), USGS (2007), USGS (2009), USGS (2010), USGS (2011), USGS (1991
14 through 2010a), USGS (1991 through 2010b), USGS (2012a) and USGS (2012b).31
15
16 3. Adjust for conversion of fuels and exports of CC>2. Fossil fuel consumption estimates are adjusted
17 downward to exclude fuels created from other fossil fuels and exports of COa.32 Synthetic natural gas is
18 created from industrial coal, and is currently included in EIA statistics for both coal and natural gas.
19 Therefore, synthetic natural gas is subtracted from energy consumption statistics.33 Since October 2000,
20 the Dakota Gasification Plant has been exporting COa to Canada by pipeline. Since this CC>2 is not emitted
21 to the atmosphere in the United States, energy used to produce this CC>2 is subtracted from energy
22 consumption statistics. To make these adjustments, additional data for ethanol were collected from EIA
23 (2015), data for synthetic natural gas were collected from EIA (2014), and data for CCh exports were
24 collected from the Eastman Gasification Services Company (2011), Dakota Gasification Company (2006),
25 Fitzpatrick (2002), Erickson (2003), EIA (2008) and DOE (2012).
26
27 4. Adjust Sectoral Allocation of Distillate Fuel Oil and Motor Gasoline. EPA had conducted a separate
28 bottom-up analysis of transportation fuel consumption based on data from the Federal Highway
29 Administration that indicated that the amount of distillate and motor gasoline consumption allocated to the
30 transportation sector in the EIA statistics should be adjusted. Therefore, for these estimates, the
31 transportation sector's distillate fuel and motor gasoline consumption was adjusted to match the value
32 obtained from the bottom-up analysis. As the total distillate and motor gasoline consumption estimate from
33 EIA are considered to be accurate at the national level, the distillate and motor gasoline consumption totals
34 for the residential, commercial, and industrial sectors were adjusted proportionately. The data sources used
35 in the bottom-up analysis of transportation fuel consumption include AAR (2008 through 2015), Benson
36 (2002 through 2004), DOE (1993 through 2015), EIA (2007), EIA (1991 through 2015), EPA (2015a), and
37 FHWA (1996 through 2015).34
29 See IPCC Reference Approach for estimating CCh emissions from fossil fuel combustion in Annex 4 for a comparison of U.S.
estimates using top-down and bottom-up approaches.
30 A crude convention to convert between gross and net calorific values is to multiply the heat content of solid and liquid fossil
fuels by 0.95 and gaseous fuels by 0.9 to account for the water content of the fuels. Biomass-based fuels in U.S. energy statistics,
however, are generally presented using net calorific values.
3! See sections on Iron and Steel Production and Metallurgical Coke Production, Ammonia Production and Urea Consumption,
Petrochemical Production, Titanium Dioxide Production, Ferroalloy Production, Aluminum Production, and Silicon Carbide
Production and Consumption in the Industrial Processes and Product Use chapter.
32 Energy statistics from EIA (2015) are already adjusted downward to account for ethanol added to motor gasoline, and biogas
in natural gas.
33 These adjustments are explained in greater detail in Annex 2.1.
The source of highway vehicle VMT and fuel consumption is FHWA's VM-1 table. In 2011, FHWA changed its methods for
estimating data in the VM-1 table. These methodological changes included how vehicles are classified, moving from a system
based on body type to one that is based on wheelbase. These changes were first incorporated for the 1990-2010 Inventory and
apply to the 2007 to 2013 time period. This resulted in large changes in VMT and fuel consumption data by vehicle class, thus
Energy 3-25
-------
1
2 5. Adjust for fuels consumed for non-energy uses. U.S. aggregate energy statistics include consumption of
3 fossil fuels for non-energy purposes. These are fossil fuels that are manufactured into plastics, asphalt,
4 lubricants, or other products. Depending on the end-use, this can result in storage of some or all of the C
5 contained in the fuel for a period of time. As the emission pathways of C used for non-energy purposes are
6 vastly different than fuel combustion (since the C in these fuels ends up in products instead of being
7 combusted), these emissions are estimated separately in the Carbon Emitted and Stored in Products from
8 Non-Energy Uses of Fossil Fuels section in this chapter. Therefore, the amount of fuels used for non-
9 energy purposes was subtracted from total fuel consumption. Data on non-fuel consumption was provided
10 byEIA(2015a).
11
12 6. Subtract consumption of international bunker fuels. According to the UNFCCC reporting guidelines
13 emissions from international transport activities, or bunker fuels, should not be included in national totals.
14 U.S. energy consumption statistics include these bunker fuels (e.g., distillate fuel oil, residual fuel oil, and
15 jet fuel) as part of consumption by the transportation end-use sector, however, so emissions from
16 international transport activities were calculated separately following the same procedures used for
17 emissions from consumption of all fossil fuels (i.e., estimation of consumption, and determination of C
18 content).35 The Office of the Under Secretary of Defense (Installations and Environment) and the Defense
19 Logistics Agency Energy (DLA Energy) of the U.S. Department of Defense (DoD) (DLA Energy 2015)
20 supplied data on military jet fuel and marine fuel use. Commercial jet fuel use was obtained from FAA
21 (2016); residual and distillate fuel use for civilian marine bunkers was obtained from DOC (1991 through
22 2014) for 1990 through 2001 and 2007 through 2014, and DHS (2008) for 2003 through 2006.
23 Consumption of these fuels was subtracted from the corresponding fuels in the transportation end-use
24 sector. Estimates of international bunker fuel emissions for the United States are discussed in detail later in
25 the International Bunker Fuels section of this chapter.
26
27 7. Determine the total C content of fuels consumed. Total C was estimated by multiplying the amount of fuel
28 consumed by the amount of C in each fuel. This total C estimate defines the maximum amount of C that
29 could potentially be released to the atmosphere if all of the C in each fuel was converted to COa. The C
30 content coefficients used by the United States were obtained from EIA's Emissions of Greenhouse Gases in
31 the United States 2008 (EIA 2009a), and an EPA analysis of C content coefficients used in the GHGRP
32 (EPA 2010). A discussion of the methodology used to develop the C content coefficients are presented in
33 Annexes 2.1 and 2.2.
34
35 8. Estimate CO2 Emissions. Total CO2 emissions are the product of the adjusted energy consumption (from
36 the previous methodology steps 1 through 6), the C content of the fuels consumed, and the fraction of C
37 that is oxidized. The fraction oxidized was assumed to be 100 percent for petroleum, coal, and natural gas
38 based on guidance in IPCC (2006) (see Annex 2.1).
39
40 9. Allocate transportation emissions by vehicle type. This report provides a more detailed accounting of
41 emissions from transportation because it is such a large consumer of fossil fuels in the United States. For
42 fuel types other than jet fuel, fuel consumption data by vehicle type and transportation mode were used to
43 allocate emissions by fuel type calculated for the transportation end-use sector. Heat contents and densities
44 were obtained from EIA (2015a) and USAF (1998).36
45 • For on-road vehicles, annual estimates of combined motor gasoline and diesel fuel consumption by
46 vehicle category were obtained from FHWA (1996 through 2014); for each vehicle category, the
leading to a shift in emissions among on-road vehicle classes. For example, the category "Passenger Cars" has been replaced by
"Light-duty Vehicles-Short Wheelbase" and "Other 2 axle-4 Lire Vehicles" has been replaced by "Light-duty Vehicles, Long
Wheelbase." This change in vehicle classification has moved some smaller trucks and sport utility vehicles from the light truck
category to the passenger vehicle category in this emission Inventory. These changes are reflected in a large drop in light-truck
emissions between 2006 and 2007.
35 See International Bunker Fuels section in this chapter for a more detailed discussion.
For a more detailed description of the data sources used for the analysis of the transportation end use sector see the Mobile
Combustion (excluding CCh) and International Bunker Fuels sections of the Energy chapter, Annex 3.2, and Annex 3.8.
3-26 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 percent gasoline, diesel, and other (e.g., CNG, LPG) fuel consumption are estimated using data from
2 DOE (1993 through 2013).
3 • For non-road vehicles, activity data were obtained from AAR (2008 through 2015), APTA (2007
4 through 2015), APTA (2006), BEA (2015), Benson (2002 through 2004), DOE (1993 through 2015),
5 DLA Energy (2015), DOC (1991 through 2015), DOT (1991 through 2015), EIA (2009a), EIA
6 (2015a), EIA (2013), EIA (1991 through 2015), EPA (2015), and Gaffney (2007).
7 • For jet fuel used by aircraft, CO2 emissions from commercial aircraft were developed by the U. S.
8 Federal Aviation Administration (FAA) using a Tier 3B methodology, consistent with the 2006IPCC
9 Guidelines for National Greenhouse Gas Inventories (see Annex 3.3). Carbon dioxide emissions from
10 other aircraft were calculated directly based on reported consumption of fuel as reported by EIA.
11 Allocation to domestic military uses was made using DoD data (see Annex 3.8). General aviation jet
12 fuel consumption is calculated as the remainder of total jet fuel use (as determined by EIA) nets all
13 other jet fuel use as determined by FAA and DoD. For more information, see Annex 3.2.
14
15
16
Box 3-4: Uses of Greenhouse Gas Reporting Program Data and Improvements in Reporting Emissions from
Industrial Sector Fossil Fuel Combustion
17 As described in the calculation methodology, total fossil fuel consumption for each year is based on aggregated end-
18 use sector consumption published by the EIA. The availability of facility-level combustion emissions through
19 EPA's Greenhouse Gas Reporting Program (GHGRP) has provided an opportunity to better characterize the
20 industrial sector's energy consumption and emissions in the United States, through a disaggregation of EIA's
21 industrial sector fuel consumption data from select industries.
22 For EPA's GHGRP 2010, 2011, 2012, 2013, and 2014 reporting years, facility-level fossil fuel combustion
23 emissions reported through the GHGRP were categorized and distributed to specific industry types by utilizing
24 facility-reported NAICS codes (as published by the U.S. Census Bureau), and associated data available from EIA's
25 2010 Manufacturing Energy Consumption Survey (MECS). As noted previously in this report, the definitions and
26 provisions for reporting fuel types in EPA's GHGRP include some differences from the Inventory's use of EIA
27 national fuel statistics to meet the UNFCCC reporting guidelines. The IPCC has provided guidance on aligning
28 facility-level reported fuels and fuel types published in national energy statistics, which guided this exercise.37
29 This year's effort represents an attempt to align, reconcile, and coordinate the facility-level reporting of fossil fuel
30 combustion emissions under EPA's GHGRP with the national-level approach presented in this report. Consistent
31 with recommendations for reporting the Inventory to the UNFCCC, progress was made on certain fuel types for
32 specific industries and has been included in the Common Reporting Format (CRF) tables that are submitted to the
33 UNFCCC along with this report.38 However, a full mapping was not completed this year due to fuel category
34 differences between national statistics published by EIA and facility-level reported GHGRP data. Furthermore,
35 given that calendar year 2010 was the first year in which emissions data were reported to EPA's GHGRP, the
36 current Inventory's examination only focused on 2010, 2011, 2012, 2013, and 2014. For the current exercise, the
37 efforts in reconciling fuels focused on standard, common fuel types (e.g., natural gas, distillate fuel oil, etc.) where
38 the fuels in EIA's national statistics aligned well with facility-level GHGRP data. For these reasons, the current
39 information presented in the CRF tables should be viewed as an initial attempt at this exercise. Additional efforts
40 will be made for future inventory reports to improve the mapping of fuel types, and examine ways to reconcile and
41 coordinate any differences between facility-level data and national statistics. Additionally, in order to expand this
42 effort through the full time series presented in this report, further analyses will be conducted linking GHGRP
43 facility-level reporting with the information published by EIA in its MECS data, other available MECS survey
44 years, and any further informative sources of data. It is believed that the current analysis has led to improvements in
37 See Section 4 "Use of Facility-Level Data in Good Practice National Greenhouse Gas Inventories" of the IPCC meeting report,
and specifically the section on using facility-level data in conjunction with energy data, at .
38 See < http://www.epa.gov/climatechange/ghgemissions/usinventoryreport.html>.
Energy 3-27
-------
1 the presentation of data in the Inventory, but further work will be conducted, and future improvements will be
2 realized in subsequent Inventory reports.
3 Additionally, to assist in the disaggregation of industrial fuel consumption, EIA will now synthesize energy
4 consumption data using the same procedure as is used for the last historical (benchmark) year of the Annual Energy
5 Outlook (AEO). This procedure reorganizes the most recent data from the Manufacturing Energy Consumption
6 Survey (MECS) (conducted every four years) into the nominal data submission year using the same energy-
7 economy integrated model used to produce the AEO projections, the National Energy Modeling System (NEMS).
8 EIA believes this "nowcasting" technique provides an appropriate estimate of energy consumption for the CRF.
9 To address gaps in the time series, EIA performs a NEMS model projection, using the MECS baseline sub-sector
10 energy consumption. The NEMS model accounts for changes in factors that influence industrial sector energy
11 consumption, and has access to data which may be more recent than MECS, such as industrial sub-sector macro
12 industrial output (i.e., shipments) and fuel prices. By evaluating the impact of these factors on industrial subsector
13 energy consumption, NEMS can anticipate changes to the energy shares occurring post-MECS and can provide a
14 way to appropriately disaggregate the energy-related emissions data into the CRF.
15 While the fuel consumption values for the various manufacturing sub-sectors are not directly surveyed for all years,
16 they represent EIA's best estimate of historical consumption values for non-MECS years. Moreover, as an integral
17 part of each AEO publication, this synthetic data series is likely to be maintained consistent with all available EIA
18 and non-EIA data sources even as the underlying data sources evolve for both manufacturing and non-
19 manufacturing industries alike.
20 Other sectors' fuel consumption (commercial, residential, transportation) will be benchmarked with the latest
21 aggregate values from the Monthly Energy Review.39 EIA will work with EPA to back cast these values to 1990.
22
23
Box 3-5: Carbon Intensity of U.S. Energy Consumption
24 Fossil fuels are the dominant source of energy in the United States, and CO2 is the dominant greenhouse gas emitted
25 as a product from their combustion. Energy-related CO2 emissions are impacted by not only lower levels of energy
26 consumption but also by lowering the C intensity of the energy sources employed (e.g., fuel switching from coal to
27 natural gas). The amount of C emitted from the combustion of fossil fuels is dependent upon the C content of the
28 fuel and the fraction of that C that is oxidized. Fossil fuels vary in their average C content, ranging from about 53
29 MMT CO2 Eq./QBtu for natural gas to upwards of 95 MMT CO2 Eq./QBtu for coal and petroleum coke.40 In
30 general, the C content per unit of energy of fossil fuels is the highest for coal products, followed by petroleum, and
31 then natural gas. The overall C intensity of the U.S. economy is thus dependent upon the quantity and combination
32 of fuels and other energy sources employed to meet demand.
33 Table 3-15 provides a time series of the C intensity for each sector of the U.S. economy. The time series
34 incorporates only the energy consumed from the direct combustion of fossil fuels in each sector. For example, the C
35 intensity for the residential sector does not include the energy from or emissions related to the consumption of
36 electricity for lighting. Looking only at this direct consumption of fossil fuels, the residential sector exhibited the
37 lowest C intensity, which is related to the large percentage of its energy derived from natural gas for heating. The C
38 intensity of the commercial sector has predominantly declined since 1990 as commercial businesses shift away from
39 petroleum to natural gas. The industrial sector was more dependent on petroleum and coal than either the residential
40 or commercial sectors, and thus had higher C intensities over this period. The C intensity of the transportation
41 sector was closely related to the C content of petroleum products (e.g., motor gasoline and jet fuel, both around 70
42 MMT CO2 Eq./EJ), which were the primary sources of energy. Lastly, the electricity generation sector had the
43 highest C intensity due to its heavy reliance on coal for generating electricity.
39 See .
40 One exajoule (EJ) is equal to 1018 joules or 0.9478 QBtu.
3-28 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
4
5
6
7
8
9
10
11
12
13
14
15
Table 3-15: Carbon Intensity from Direct Fossil Fuel Combustion by Sector (MMT COz
Eq./QBtu)
Sector
1990
2005
Residential*
Commercial*
Industrial*
Transportation*
Electricity Generation15
U.S. Territories0
57.4
59.1
64.3
71.1
87.3
73.0
2011 2012 2013
2014
55.7
56.6
62.4
71.5
82.9
73.1
55.5
56.1
62.0
71.5
79.9
72.4
55.3
55.8
61.8
71.4
81.3
72.1
55.4
55.7
61.5
71.4
81.3
71.6
All Sectors0
73.0
73.5
72.4
72.0
70.9 70.9
70.7
* Does not include electricity or renewable energy consumption.
b Does not include electricity produced using nuclear or renewable energy.
c Does not include nuclear or renewable energy consumption.
Note: Excludes non-energy fuel use emissions and consumption.
Over the twenty-five-year period of 1990 through 2014, however, the C intensity of U.S. energy consumption has
been fairly constant, as the proportion of fossil fuels used by the individual sectors has not changed significantly.
Per capita energy consumption fluctuated little from 1990 to 2007, but in 2014 was approximately 8.5 percent below
levels in 1990 (see Figure 3-14). Due to a general shift from a manufacturing-based economy to a service-based
economy, as well as overall increases in efficiency, energy consumption and energy-related CO2 emissions per
dollar of gross domestic product (GDP) have both declined since 1990 (BEA 2015).
Figure 3-14: U.S. Energy Consumption and Energy-Related COz Emissions Per Capita and Per
Dollar GDP
COj/Energy
Consumption
Energy
Consumption/capita
Energy
Consumption/$GDP
(Red)
C intensity estimates were developed using nuclear and renewable energy data from EIA (2015a), EPA (2010a), and
fossil fuel consumption data as discussed above and presented in Annex 2.1.
16
17
18
19
20
21
Uncertainty and Time Series Consistency
For estimates of CCh from fossil fuel combustion, the amount of CCh emitted is directly related to the amount of
fuel consumed, the fraction of the fuel that is oxidized, and the carbon content of the fuel. Therefore, a careful
accounting of fossil fuel consumption by fuel type, average carbon contents of fossil fuels consumed, and
production of fossil fuel-based products with long-term carbon storage should yield an accurate estimate
emissions.
Energy 3-29
-------
1 Nevertheless, there are uncertainties in the consumption data, carbon content of fuels and products, and carbon
2 oxidation efficiencies. For example, given the same primary fuel type (e.g., coal, petroleum, or natural gas), the
3 amount of carbon contained in the fuel per unit of useful energy can vary. For the United States, however, the
4 impact of these uncertainties on overall CCh emission estimates is believed to be relatively small. See, for example,
5 Marland and Pippin (1990).
6 Although statistics of total fossil fuel and other energy consumption are relatively accurate, the allocation of this
7 consumption to individual end-use sectors (i.e., residential, commercial, industrial, and transportation) is less
8 certain. For example, for some fuels the sectoral allocations are based on price rates (i.e., tariffs), but a commercial
9 establishment may be able to negotiate an industrial rate or a small industrial establishment may end up paying an
10 industrial rate, leading to a misallocation of emissions. Also, the deregulation of the natural gas industry and the
11 more recent deregulation of the electric power industry have likely led to some minor problems in collecting
12 accurate energy statistics as firms in these industries have undergone significant restructuring.
13 To calculate the total CCh emission estimate from energy-related fossil fuel combustion, the amount of fuel used in
14 these non-energy production processes were subtracted from the total fossil fuel consumption. The amount of €62
15 emissions resulting from non-energy related fossil fuel use has been calculated separately and reported in the Carbon
16 Emitted from Non-Energy Uses of Fossil Fuels section of this report. These factors all contribute to the uncertainty
17 in the CCh estimates. Detailed discussions on the uncertainties associated with C emitted from Non-Energy Uses of
18 Fossil Fuels can be found within that section of this chapter.
19 Various sources of uncertainty surround the estimation of emissions from international bunker fuels, which are
20 subtracted from the U.S. totals (see the detailed discussions on these uncertainties provided in the International
21 Bunker Fuels section of this chapter). Another source of uncertainty is fuel consumption by U.S. Territories. The
22 United States does not collect energy statistics for its territories at the same level of detail as for the fifty states and
23 the District of Columbia. Therefore, estimating both emissions and bunker fuel consumption by these territories is
24 difficult.
25 Uncertainties in the emission estimates presented above also result from the data used to allocate CO2 emissions
26 from the transportation end-use sector to individual vehicle types and transport modes. In many cases, bottom-up
27 estimates of fuel consumption by vehicle type do not match aggregate fuel-type estimates from EIA. Further
28 research is planned to improve the allocation into detailed transportation end-use sector emissions.
29 The uncertainty analysis was performed by primary fuel type for each end-use sector, using the IPCC-recommended
30 Approach 2 uncertainty estimation methodology, Monte Carlo Stochastic Simulation technique, with @RISK
31 software. For this uncertainty estimation, the inventory estimation model for CO2 from fossil fuel combustion was
32 integrated with the relevant variables from the inventory estimation model for International Bunker Fuels, to
33 realistically characterize the interaction (or endogenous correlation) between the variables of these two models.
34 About 120 input variables were modeled for CO2 from energy-related Fossil Fuel Combustion (including about 10
35 for non-energy fuel consumption and about 20 for International Bunker Fuels).
36 In developing the uncertainty estimation model, uniform distributions were assumed for all activity-related input
37 variables and emission factors, based on the SAIC/EIA (2001) report.41 Triangular distributions were assigned for
38 the oxidization factors (or combustion efficiencies). The uncertainty ranges were assigned to the input variables
39 based on the data reported in SAIC/EIA (2001) and on conversations with various agency personnel.42
40 The uncertainty ranges for the activity-related input variables were typically asymmetric around their inventory
41 estimates; the uncertainty ranges for the emissions factors were symmetric. Bias (or systematic uncertainties)
41 SAIC/EIA (2001) characterizes the underlying probability density function for the input variables as a combination of uniform
and normal distributions (the former to represent the bias component and the latter to represent the random component).
However, for purposes of the current uncertainty analysis, it was determined that uniform distribution was more appropriate to
characterize the probability density function underlying each of these variables.
In the SAIC/EIA (2001) report, the quantitative uncertainty estimates were developed for each of the three major fossil fuels
used within each end-use sector; the variations within the sub-fuel types within each end-use sector were not modeled. However,
for purposes of assigning uncertainty estimates to the sub-fuel type categories within each end-use sector in the current
uncertainty analysis, SAIC/EIA (2001)-reported uncertainty estimates were extrapolated.
3-30 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 associated with these variables accounted for much of the uncertainties associated with these variables (S AIC/EIA
2 2001).43 For purposes of this uncertainty analysis, each input variable was simulated 10,000 times through Monte
3 Carlo sampling.
4 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 3-16. Fossil fuel
5 combustion CC>2 emissions in 2014 were estimated to be between 5,097.2 and 5,455.5 MMT CCh Eq. at a 95 percent
6 confidence level. This indicates a range of 2 percent below to 5 percent above the 2014 emission estimate of
7 5,208.7 MMT CO2 Eq.
8 Table 3-16: Approach 2 Quantitative Uncertainty Estimates for COz Emissions from Energy-
9 related Fossil Fuel Combustion by Fuel Type and Sector (MMT COz Eq. and Percent)
Fuel/Sector
Coal"
Residential
Commercial
Industrial
Transportation
Electricity Generation
U.S. Territories
Natural Gasb
Residential
Commercial
Industrial
Transportation
Electricity Generation
U.S. Territories
Petroleum1"
Residential
Commercial
Industrial
Transportation
Electric Utilities
U.S. Territories
Total (excluding Geothermal)b
Geothermal
Total (including Geothermal)b'c
2014 Emission Estimate
(MMT CO2 Eq.)
1,653.7
NE
4.5
75.3
NE
1,570.4
3.4
1,426.6
277.6
189.2
466.0
47.6
443.2
3.0
2,128.0
67.6
37.9
272.9
1,689.8
25.3
34.6
5,208.2
0.4
5,208.7
Uncertainty
Range Relative
to Emission
Estimate3
(MMT CO2 Eq.) (%)
Lower
Bound
1,595.9
NE
4.3
71.7
NE
1,509.3
3.0
1,412.3
270.0
183.8
451.9
46.3
430.3
2.6
1,992.7
63.8
35.9
218.4
1,576.0
24.1
32.0
5,096.8
NE
5,097.2
Upper
Bound
1,809.7
NE
5.2
87.1
NE
1,722.0
4.0
1,492.8
297.1
202.5
499.5
51.0
465.5
3.5
2,251.6
71.0
39.7
322.2
1,800.5
27.3
38.4
5,455.1
NE
5,455.5
Lower
Bound
-3%
NE
-5%
-5%
NE
-4%
-12%
-1%
-3%
-3%
-3%
-3%
-3%
-12%
-6%
-6%
-5%
-20%
-7%
-5%
-7%
-2%
NE
-2%
Upper
Bound
+9%
NE
+15%
+16%
NE
+10%
+19%
+5%
+7%
+7%
+7%
+7%
+5%
+17%
+6%
+5%
+5%
+18%
+7%
+8%
+11%
+5%
NE
+5%
NA (Not Applicable)
NE (Not Estimated)
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
b The low and high estimates for total emissions were calculated separately through simulations and, hence, the low and
high emission estimates for the sub-source categories do not sum to total emissions.
c Geothermal emissions added for reporting purposes, but an uncertainty analysis was not performed for CCh emissions
from geothermal production.
10 Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
11 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
12 above.
43 Although, in general, random uncertainties are the main focus of statistical uncertainty analysis, when the uncertainty
estimates are elicited from experts, their estimates include both random and systematic uncertainties. Hence, both these types of
uncertainties are represented in this uncertainty analysis.
Energy 3-31
-------
i QA/QC and Verification
2 A source-specific QA/QC plan for CCh from fossil fuel combustion was developed and implemented. This effort
3 included a Tier 1 analysis, as well as portions of a Tier 2 analysis. The Tier 2 procedures that were implemented
4 involved checks specifically focusing on the activity data and methodology used for estimating €62 emissions from
5 fossil fuel combustion in the United States. Emission totals for the different sectors and fuels were compared and
6 trends were investigated to determine whether any corrective actions were needed. Minor corrective actions were
7 taken.
8 Recalculations Discussion
9 The Energy Information Administration (EIA 2015a) updated energy consumption statistics across the time series
10 relative to the previous Inventory. One such revision is the historical coal and petroleum product consumption in the
11 industrial sector for the entire time series. In addition, EIA revised 2013 natural gas consumption in the
12 transportation sector and 2013 kerosene and Liquefied Petroleum Gas (LPG) consumption in the residential and
13 commercial sectors.
14 Kerosene consumption increased in the residential sector by 9 percent in 2013 and decreased by 14 and 25 percent in
15 the commercial and industrial sectors in 2013, respectively. Transportation sector distillate fuel consumption
16 decreased by 0.4 percent across the entire time series.
17 In early 2015, EIA revised the heat content used to calculate the energy of distillate fuel oil consumption.
18 Previously, a single constant factor (5.825 MMBtu/barrel) from EIA's Monthly Energy Review (MER) Table Al
19 was applied to the volumetric data. For the January 2015 release, this single constant factor in Table Al was
20 replaced with heat content factors for distillate fuel oil by sulfur content. Instead of using the factor(s) listed in
21 Table Al, EIA began to use an annually variable quantity-weighted factor (5.774 MMBtu/barrel for 2013) that was
22 added to Table A3. EIA notes that quantity-weighted averages of the sulfur-content categories of distillate fuel oil
23 are calculated by using heat content values shown in Table Al, and that these values exclude renewable diesel fuel
24 (including biodiesel) blended into distillate fuel oil.
25 Overall, these changes resulted in an average annual decrease of 1.1 MMT CO2 Eq. (less than 0.1 percent) in CO2
26 emissions from fossil fuel combustion for the period 1990 through 2013, relative to the previous report.
27 Planned Improvements
28 To reduce uncertainty of CO2 from fossil fuel combustion estimates, efforts will be taken to work with EIA and
29 other agencies to improve the quality of the U.S. Territories data. This improvement is not all-inclusive, and is part
30 of an ongoing analysis and efforts to continually improve the CO2 from fossil fuel combustion estimates. In
31 addition, further expert elicitation may be conducted to better quantify the total uncertainty associated with
32 emissions from this source.
33 The availability of facility-level combustion emissions through EPA's GHGRP will continue to be examined to help
34 better characterize the industrial sector's energy consumption in the United States, and further classify business
35 establishments according to industrial economic activity type. Most methodologies used in EPA's GHGRP are
36 consistent with IPCC, though for EPA's GHGRP, facilities collect detailed information specific to their operations
37 according to detailed measurement standards, which may differ with the more aggregated data collected for the
38 Inventory to estimate total, national U.S. emissions. In addition, and unlike the reporting requirements for this
39 chapter under the UNFCCC reporting guidelines, some facility-level fuel combustion emissions reported under the
40 GHGRP may also include industrial process emissions.44 In line with UNFCCC reporting guidelines, fuel
41 combustion emissions are included in this chapter, while process emissions are included in the Industrial Processes
42 and Product Use chapter of this report. In examining data from EPA's GHGRP that would be useful to improve the
43 emission estimates for the CO2 from fossil fuel combustion category, particular attention will also be made to ensure
44 time series consistency, as the facility-level reporting data from EPA's GHGRP are not available for all inventory
45 years as reported in this Inventory. Additional, analyses will be conducted to align reported facility-level fuel types
44 See .
3-32 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 and IPCC fuel types per the national energy statistics. Additional work will commence to ensure CO2 emissions
2 from biomass are separated in the facility-level reported data, and maintaining consistency with national energy
3 statistics provided by EIA. In implementing improvements and integration of data from EPA's GHGRP, the latest
4 guidance from the IPCC on the use of facility-level data in national inventories will continue to be relied upon.45
5 Another planned improvement is to develop improved estimates of domestic waterborne fuel consumption. The
6 inventory estimates for residual and distillate fuel used by ships and boats is based in part on data on bunker fuel use
7 from the U.S. Department of Commerce. Domestic fuel consumption is estimated by subtracting fuel sold for
8 international use from the total sold in the United States. It may be possible to more accurately estimate domestic
9 fuel use and emissions by using detailed data on marine ship activity. The feasibility of using domestic marine
10 activity data to improve the estimates is currently being investigated.
11 An additional potential improvement is to include CCh emissions from natural gas (LNG and CNG) use in medium-
12 and heavy-duty trucks, light trucks and passenger cars. Currently data from the Transportation Energy Data book is
13 used to allocate CCh emissions to vehicle categories. However, this data source only estimates natural gas use in
14 buses. We are currently investigating the use of alternative data sources from the EIA that would allow some of the
15 CChfrom natural gas consumption to be allocated to these other vehicle categories.
16 In addition, we are investigating an approach to account for CCh emissions from the use of urea-based additives in
17 catalytic converters for on-road vehicles between 2010 and 2014. The approach would utilize the MOVES model to
18 estimate fuel use by diesel vehicles with urea-based catalysts. The 2006 IPCC Guidelines estimates urea use
19 between one and three percent of diesel fuel used.
20
CH4and N2O from Stationary Combustion
21 Methodology
22 Methane and N2O emissions from stationary combustion were estimated by multiplying fossil fuel and wood
23 consumption data by emission factors (by sector and fuel type for industrial, residential, commercial, and U.S.
24 Territories; and by fuel and technology type for the electric power sector). Beginning with the current Inventory
25 report, the electric power sector utilizes a Tier 2 methodology, whereas all other sectors utilize a Tier 1
26 methodology. The activity data and emission factors used are described in the following subsections.
27 Industrial, Residential, Commercial, and U.S. Territories
28 National coal, natural gas, fuel oil, and wood consumption data were grouped by sector: industrial, commercial,
29 residential, and U.S. Territories. For the CH4 and N2O estimates, wood consumption data for the United States was
30 obtained from EIA's Monthly Energy Review (EIA 2015a). Fuel consumption data for coal, natural gas, and fuel oil
31 for the United States were also obtained from EIA's Monthly Energy Review and unpublished supplemental tables
32 on petroleum product detail (EIA 2015a). Because the United States does not include territories in its national
33 energy statistics, fuel consumption data for territories were provided separately by EIA's International Energy
34 Statistics (EIA 2014) and Jacobs (2010).46 Fuel consumption for the industrial sector was adjusted to subtract out
35 construction and agricultural use, which is reported under mobile sources.47 Construction and agricultural fuel use
36 was obtained from EPA (2014). Estimates for wood biomass consumption for fuel combustion do not include wood
37 wastes, liquors, municipal solid waste, tires, etc., that are reported as biomass by EIA. Tier 1 default emission
38 factors for these three end-use sectors were provided by the 2006 IPCC Guidelines for National Greenhouse Gas
45 See.
46 U.S. Territories data also include combustion from mobile activities because data to allocate territories' energy use were
unavailable. For this reason, CH4 and N2O emissions from combustion by U.S. Territories are only included in the stationary
combustion totals.
47 Though emissions from construction and farm use occur due to both stationary and mobile sources, detailed data was not
available to determine the magnitude from each. Currently, these emissions are assumed to be predominantly from mobile
sources.
Energy 3-33
-------
1 Inventories (IPCC 2006). U.S. Territories' emission factors were estimated using the U.S. emission factors for the
2 primary sector in which each fuel was combusted.
3 Electric Power Sector
4 The electric power sector now uses a Tier 2 emission estimation methodology as fuel consumption for the electricity
5 generation sector by control-technology type was obtained from EPA's Acid Rain Program Dataset (EPA 2015a).
6 This combustion technology- and fuel-use data was available by facility from 1996 to 2014 The Tier 2 emission
7 factors used were taken from the 2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2006),
8 which in turn are based on emission factors published by EPA.
9 Since there was a difference between the EPA (2015a) and EIA (2015a) total energy consumption estimates, the
10 remaining energy consumption from EIA (2015a) was apportioned to each combustion technology type and fuel
11 combination using a ratio of energy consumption by technology type from 1996 to 2014.
12 Energy consumption estimates were not available from 1990 to 1995 in the EPA (2015a) dataset, and as a result,
13 consumption was calculated using total electric power consumption from EIA (2015 a) and the ratio of combustion
14 technology and fuel types from EPA (2015a). The consumption estimates from 1990 to 1995 were estimated by
15 applying the 1996 consumption ratio by combustion technology type to the total EIA consumption for each year
16 from 1990 to 1995. Emissions were estimated by multiplying fossil fuel and wood consumption by technology- and
17 fuel-specific Tier 2 IPCC emission factors.
18 Lastly, there were significant differences between wood biomass consumption in the electric power sector between
19 the EPA (2015a) and EIA (2015a) datasets. The higher wood biomass consumption from EIA (2015a) in the electric
20 power sector was distributed to the residential, commercial, and industrial sectors according to their percent share of
21 wood biomass energy consumption calculated from EIA (2015a).
22 More detailed information on the methodology for calculating emissions from stationary combustion, including
23 emission factors and activity data, is provided in Annex 3.1.
24 Uncertainty and Time-Series Consistency
25 Methane emission estimates from stationary sources exhibit high uncertainty, primarily due to difficulties in
26 calculating emissions from wood combustion (i.e., fireplaces and wood stoves). The estimates of CH4 and N2O
27 emissions presented are based on broad indicators of emissions (i.e., fuel use multiplied by an aggregate emission
28 factor for different sectors), rather than specific emission processes (i.e., by combustion technology and type of
29 emission control).
30 An uncertainty analysis was performed by primary fuel type for each end-use sector, using the IPCC-recommended
31 Approach 2 uncertainty estimation methodology, Monte Carlo Stochastic Simulation technique, with @RISK
32 software.
33 The uncertainty estimation model for this source category was developed by integrating the CH4 and N2O stationary
34 source inventory estimation models with the model for CO2 from fossil fuel combustion to realistically characterize
35 the interaction (or endogenous correlation) between the variables of these three models. About 55 input variables
36 were simulated for the uncertainty analysis of this source category (about 20 from the CCh emissions from fossil
37 fuel combustion inventory estimation model and about 35 from the stationary source inventory models).
38 In developing the uncertainty estimation model, uniform distribution was assumed for all activity-related input
39 variables and N2O emission factors, based on the SAIC/EIA (2001) report.48 For these variables, the uncertainty
SAIC/EIA (2001) characterizes the underlying probability density function for the input variables as a combination of uniform
and normal distributions (the former distribution to represent the bias component and the latter to represent the random
component). However, for purposes of the current uncertainty analysis, it was determined that uniform distribution was more
appropriate to characterize the probability density function underlying each of these variables.
3-34 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 ranges were assigned to the input variables based on the data reported in SAIC/EIA (2001).49 However, the CH4
2 emission factors differ from those used by EIA. These factors and uncertainty ranges are based on IPCC default
3 uncertainty estimates (IPCC 2006).
4 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 3-17. Stationary
5 combustion CH4 emissions in 2014 (including biomass) were estimated to be between 4.6 and 20.6 MMT CO2 Eq. at
6 a 95 percent confidence level. This indicates a range of 42 percent below to 155 percent above the 2014 emission
7 estimate of 8.1 MMT CO2 Eq.50 Stationary combustion N2O emissions in 2014 (including biomass) were estimated
8 to be between 17.9 and 34.4 MMT CO2 Eq. at a 95 percent confidence level. This indicates a range of 24 percent
9 below to 47 percent above the 2014 emissions estimate of 23.4 MMT CO2 Eq.
10 Table 3-17: Approach 2 Quantitative Uncertainty Estimates for CH4 and NzO Emissions from
11 Energy-Related Stationary Combustion, Including Biomass (MMT COz Eq. and Percent)
2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
°UrCe aS (MMTCChEq.) (MMT CCh Eq.) (%)
Stationary Combustion
Stationary Combustion
CH4
N2O
8.1
23.4
Lower
Bound
4.6
17.9
Upper
Bound
20.6
34.4
Lower
Bound
-42%
-24%
Upper
Bound
+155%
+47%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
12 The uncertainties associated with the emission estimates of CH4 and N2O are greater than those associated with
13 estimates of CO2 from fossil fuel combustion, which mainly rely on the carbon content of the fuel combusted.
14 Uncertainties in both CH4 and N2O estimates are due to the fact that emissions are estimated based on emission
15 factors representing only a limited subset of combustion conditions. For the indirect greenhouse gases, uncertainties
16 are partly due to assumptions concerning combustion technology types, age of equipment, emission factors used,
17 and activity data projections.
18 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
19 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
20 above.
21 QA/QC and Verification
22 A source-specific QA/QC plan for stationary combustion was developed and implemented. This effort included a
23 Tier 1 analysis, as well as portions of a Tier 2 analysis. The Tier 2 procedures that were implemented involved
24 checks specifically focusing on the activity data and emission factor sources and methodology used for estimating
25 CH4, N2O, and the indirect greenhouse gases from stationary combustion in the United States. Emission totals for
26 the different sectors and fuels were compared and trends were investigated.
27 Recalculations Discussion
28 Methane and N2O emissions from stationary sources (excluding CO2) across the entire time series were revised due
29 revised data from EIA (2015a) and EPA (2015a) relative to the previous Inventory. The historical data changes
30 resulted in an average annual decrease of less than 0,1 MMT CO2 Eq. (less than 0.1 percent) in both CH4 and N2O
31 emissions from stationary combustion for the period 1990 through 2013.
49 In the SAIC/EIA (2001) report, the quantitative uncertainty estimates were developed for each of the three major fossil fuels
used within each end-use sector; the variations within the sub-fuel types within each end-use sector were not modeled. However,
for purposes of assigning uncertainty estimates to the sub-fuel type categories within each end-use sector in the current
uncertainty analysis, SAIC/EIA (2001)-reported uncertainty estimates were extrapolated.
The low emission estimates reported in this section have been rounded down to the nearest integer values and the high
emission estimates have been rounded up to the nearest integer values.
Energy 3-35
-------
i Planned Improvements
2 Several items are being evaluated to improve the CH4 and N2O emission estimates from stationary combustion and
3 to reduce uncertainty. Efforts will be taken to work with EIA and other agencies to improve the quality of the U.S.
4 Territories data. Because these data are not broken out by stationary and mobile uses, further research will be aimed
5 at trying to allocate consumption appropriately. In addition, the uncertainty of biomass emissions will be further
6 investigated since it was expected that the exclusion of biomass from the uncertainty estimates would reduce the
7 uncertainty; and in actuality the exclusion of biomass increases the uncertainty. These improvements are not all-
8 inclusive, but are part of an ongoing analysis and efforts to continually improve these stationary estimates.
9 Future improvements to the CH4 and N2O from Stationary Combustion category involve research into the
10 availability of CH4 and N2O from stationary combustion data, and analyzing data reported under EPA's GHGRP. In
11 examining data from EPA's GHGRP that would be useful to improve the emission estimates for CH4 and N2O from
12 Stationary Combustion category, particular attention will be made to ensure time series consistency, as the facility -
13 level reporting data from EPA's GHGRP are not available for all Inventory years as reported in this Inventory. In
14 implementing improvements and integration of data from EPA's GHGRP, the latest guidance from the IPCC on the
15 use of facility-level data in national inventories will be relied upon.51
16
CH4 and N2O from Mobile Combustion
17 Methodology
18 Estimates of CH4 and N2O emissions from mobile combustion were calculated by multiplying emission factors by
19 measures of activity for each fuel and vehicle type (e.g., light-duty gasoline trucks). Activity data included vehicle
20 miles traveled (VMT) for on-road vehicles and fuel consumption for non-road mobile sources. The activity data and
21 emission factors used are described in the subsections that follow. A complete discussion of the methodology used to
22 estimate CH4 and N2O emissions from mobile combustion and the emission factors used in the calculations is provided
23 in Annex 3.2.
24 On-Road Vehicles
25 Estimates of CH4 and N2O emissions from gasoline and diesel on-road vehicles are based on VMT and emission
26 factors by vehicle type, fuel type, model year, and emission control technology. Emission estimates for alternative
27 fuel vehicles (AFVs) are based on VMT and emission factors by vehicle and fuel type.52
28 Emission factors for gasoline and diesel on-road vehicles utilizing Tier 2 and Low Emission Vehicle (LEV)
29 technologies were developed by ICF (2006b); all other gasoline and diesel on-road vehicle emissions factors were
30 developed by ICF (2004). These factors were derived from EPA, California Air Resources Board (CARD) and
31 Environment Canada laboratory test results of different vehicle and control technology types. The EPA, CARD and
32 Environment Canada tests were designed following the Federal Test Procedure (FTP), which covers three separate
33 driving segments, since vehicles emit varying amounts of greenhouse gases depending on the driving segment.
34 These driving segments are: (1) a transient driving cycle that includes cold start and running emissions, (2) a cycle
35 that represents running emissions only, and (3) a transient driving cycle that includes hot start and running
36 emissions. For each test run, a bag was affixed to the tailpipe of the vehicle and the exhaust was collected; the
37 content of this bag was then analyzed to determine quantities of gases present. The emissions characteristics of
38 segment 2 were used to define running emissions, and subtracted from the total FTP emissions to determine start
39 emissions. These were then recombined based upon the ratio of start to running emissions for each vehicle class
40 from MOBILE6.2, an EPA emission factor model that predicts gram per mile emissions of CO2, CO, HC, NOX, and
41 PM from vehicles under various conditions, to approximate average driving characteristics.53
51 See.
52 Alternative fuel and advanced technology vehicles are those that can operate using a motor fuel other than gasoline or diesel.
This includes electric or other bi-fuel or dual-fuel vehicles that may be partially powered by gasoline or diesel.
53 Additional information regarding the model can be found online at .
3-36 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Emission factors for AFVs were developed by ICF (2006a) after examining Argonne National Laboratory's GREET
2 1.7-Transportation Fuel Cycle Model (ANL 2006) and Lipman and Delucchi (2002). These sources describe AFV
3 emission factors in terms of ratios to conventional vehicle emission factors. Ratios of AFV to conventional vehicle
4 emissions factors were then applied to estimated Tier 1 emissions factors from light-duty gasoline vehicles to
5 estimate light-duty AFVs. Emissions factors for heavy-duty AFVs were developed in relation to gasoline heavy-
6 duty vehicles. A complete discussion of the data source and methodology used to determine emission factors from
7 AFVs is provided in Annex 3.2.
8 Annual VMT data for 1990 through 2014 were obtained from the Federal Highway Administration's (FHWA)
9 Highway Performance Monitoring System database as reported in Highway Statistics (FHWA 1996 through
10 2015).54 VMT estimates were then allocated from FHWA's vehicle categories to fuel-specific vehicle categories
11 using the calculated shares of vehicle fuel use for each vehicle category by fuel type reported in DOE (1993 through
12 2015) and information on total motor vehicle fuel consumption by fuel type from FHWA (1996 through 2015).
13 VMT for AFVs were estimated based on Browning (2015). The age distributions of the U.S. vehicle fleet were
14 obtained from EPA (2015b, 2000), and the average annual age-specific vehicle mileage accumulation of U.S.
15 vehicles were obtained from EPA (2015b).
16 Control technology and standards data for on-road vehicles were obtained from EPA's Office of Transportation and
17 Air Quality (EPA 2007a, 2007b, 2000, 1998, and 1997) and Browning (2005). These technologies and standards are
18 defined in Annex 3.2, and were compiled from EPA (1994a, 1994b, 1998, 1999a) and IPCC (2006).
19 Non-Road Vehicles
20 To estimate emissions from non-road vehicles, fuel consumption data were employed as a measure of activity, and
21 multiplied by fuel-specific emission factors (in grams of N2Oand CH4 per kilogram of fuel consumed).55 Activity
22 data were obtained from AAR (2008 through 2015), APTA (2007 through 2015), APTA (2006), BEA (1991 through
23 2015), Benson (2002 through 2004), DHS (2008), DLA Energy (2015), DOC (1991 through 2015), DOE (1993
24 through 2015), DOT (1991 through 2015), EIA (2002, 2007, 2015a), EIA (2007 through 2015), EIA (1991 through
25 2015), EPA (2015b), Esser (2003 through 2004), FAA (2016), FHWA (1996 through 2015), Gaffney (2007), and
26 Whorton (2006 through 2014). Emission factors for non-road modes were taken from IPCC (2006) and Browning
27 (2009).
28 Uncertainty and Time-Series Consistency
29 A quantitative uncertainty analysis was conducted for the mobile source sector using the IPCC-recommended
30 Approach 2 uncertainty estimation methodology, Monte Carlo Stochastic Simulation technique, using @RISK
31 software. The uncertainty analysis was performed on 2014 estimates of CH4 and N2O emissions, incorporating
32 probability distribution functions associated with the major input variables. For the purposes of this analysis, the
33 uncertainty was modeled for the following four major sets of input variables: (1) VMT data, by on-road vehicle and
34 fuel type and (2) emission factor data, by on-road vehicle, fuel, and control technology type, (3) fuel consumption,
35 data, by non-road vehicle and equipment type, and (4) emission factor data, by non-road vehicle and equipment
36 type.
37 Uncertainty analyses were not conducted for NOX, CO, or NMVOC emissions. Emission factors for these gases
38 have been extensively researched since emissions of these gases from motor vehicles are regulated in the United
39 States, and the uncertainty in these emission estimates is believed to be relatively low. For more information, see
54 The source of VMT is FHWA's VM-1 table. In 2011, FHWA changed its methods for estimating data in the VM-1 table.
These methodological changes included how vehicles are classified, moving from a system based on body-type to one that is
based on wheelbase. These changes were first incorporated for the 2010 Inventory and apply to the 2007-14 time period. This
resulted in large changes in VMT by vehicle class, thus leading to a shift in emissions among on-road vehicle classes. For
example, the category "Passenger Cars" has been replaced by "Light-duty Vehicles-Short Wheelbase" and "Other 2 axle-4 Tire
Vehicles" has been replaced by "Light-duty Vehicles, Long Wheelbase." This change in vehicle classification has moved some
smaller trucks and sport utility vehicles from the light truck category to the passenger vehicle category in this emission inventory.
These changes are reflected in a large drop in light-truck emissions between 2006 and 2007.
The consumption of international bunker fuels is not included in these activity data, but is estimated separately under the
International Bunker Fuels source category.
Energy 3-37
-------
1 Section 3.7. However, a much higher level of uncertainty is associated with CH4 and N2O emission factors due to
2 limited emission test data, and because, unlike CO2 emissions, the emission pathways of CH4 and N2O are highly
3 complex.
4 Mobile combustion CH4 emissions from all mobile sources in 2014 were estimated to be between 1.8 and 2.4 MMT
5 CO2 Eq. at a 95 percent confidence level. This indicates a range of 12 percent below to 18 percent above the
6 corresponding 2014 emission estimate of 2.0 MMT CO2 Eq. Also at a 95 percent confidence level, mobile
7 combustion N2O emissions from mobile sources in 2014 were estimated to be between 15.7 and 20.7 MMT CO2
8 Eq., indicating a range of 4 percent below to 27 percent above the corresponding 2014 emission estimate of 16.3
9 MMT CO2 Eq.
10 Table 3-18: Approach 2 Quantitative Uncertainty Estimates for CH4 and NzO Emissions from
11 Mobile Sources (MMT COz Eq. and Percent)
2014 Emission Estimate3 Uncertainty Range Relative to Emission Estimate3
S°UrCe GaS (MMTCChEq.) (MMT CCh Eq.) (%)
Mobile Sources
Mobile Sources
CH4
N20
2.0
16.3
Lower
Bound
1.8
15.7
Upper
Bound
2.4
20.7
Lower
Bound
-12%
-4%
Upper
Bound
+18%
+27%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence
interval.
12 This uncertainty analysis is a continuation of a multi-year process for developing quantitative uncertainty estimates
13 for this source category using the IPCC Approach 2 uncertainty analysis. As a result, as new information becomes
14 available, uncertainty characterization of input variables may be improved and revised. For additional information
15 regarding uncertainty in emission estimates for CH4 and N2O please refer to the Uncertainty Annex.
16 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
17 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
18 above.
19 QA/QC and Verification
20 A source-specific QA/QC plan for mobile combustion was developed and implemented. This plan is based on the
21 IPCC-recommended QA/QC Plan. The specific plan used for mobile combustion was updated prior to collection and
22 analysis of this current year of data. This effort included a Tier 1 analysis, as well as portions of a Tier 2 analysis.
23 The Tier 2 procedures focused on the emission factor and activity data sources, as well as the methodology used for
24 estimating emissions. These procedures included a qualitative assessment of the emissions estimates to determine
25 whether they appear consistent with the most recent activity data and emission factors available. A comparison of
26 historical emissions between the current Inventory and the previous inventory was also conducted to ensure that the
27 changes in estimates were consistent with the changes in activity data and emission factors.
28 Recalculations Discussion
29 Decreases to CH4 and N2O emissions from mobile combustion are largely due to updates made to the Motor Vehicle
30 Emissions Simulator (MOVES 2014a) model that is used to estimate on-road gasoline vehicle distribution and
31 mileage across the time series. These changes are due to the updated MOVES age distributions for years 1999-2013
32 in this year's inventory. These changes in the age distribution increased the percentage of vehicles and VMT for
33 some vehicle types in newer model years that have better emissions control technology. For aircrafts, a weighted jet
34 fuel heat content was applied to the jet fuel N2O emissions calculation. The weighted factor accounts for the
35 different heat contents of jet fuels used in commercial aviation, general aviation and the military. This resulted in a
36 0.4 percent increase in the heat content and a similar increase in N2O emissions.
37 Estimates of alternative fuel vehicle mileage were also revised to reflect updates made to Energy Information
38 Administration (EIA) data on alternative fuel use and vehicle counts. The energy economy ratios (EERs) in the
39 alternative fuel vehicle analysis were also updated in this inventory. EERs are the ratio of the gasoline equivalent
3-38 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 fuel economy of a given technology to that of conventional gasoline or diesel vehicles. These were taken from the
2 Argonne National Laboratory's GREET model (ANL 2015). Most of the energy economy ratios were within 10
3 percent of their previous values. More significant changes occurred with Neighborhood Electric Vehicles (NEVs) (-
4 26 percent), Electric Vehicles (EVs) (17 percent), Fuel Cell Hydrogen (-15 percent), Neat Methanol Internal
5 Combustion Engines (ICEs) (12 percent), Neat Ethanol ICEs (25 percent), LPG ICEs (11 percent) and LPG Bi-fuel
6 (11 percent). Increases in EERs increase miles per gallon, estimated VMT, and emissions.
7 Overall, these changes resulted in an average annual decrease of 0.1 MMT CCh Eq. (4 percent) in CH4 emissions
8 and an average annual decrease of 1.4 MMT CC>2 Eq. (3 percent) in N2O emissions from mobile combustion for the
9 period 1990 through 2013, relative to the previous report.
10 Planned Improvements
11 While the data used for this report represent the most accurate information available, several areas have been
12 identified that could potentially be improved in the near term given available resources.
13 • Develop improved estimates of domestic waterborne fuel consumption. The inventory estimates for
14 residual and distillate fuel used by ships and boats is based in part on data on bunker fuel use from the U.S.
15 Department of Commerce. Domestic fuel consumption is estimated by subtracting fuel sold for
16 international use from the total sold in the United States. It may be possible to more accurately estimate
17 domestic fuel use and emissions by using detailed data on marine ship activity. The feasibility of using
18 domestic marine activity data to improve the estimates is currently being investigated. Additionally, the
19 feasibility of including data from a broader range of domestic and international sources for domestic bunker
20 fuels, including data from studies such as the Third IMO GHG Study 2014, is being considered.
21 • Continue to examine the use of EPA's MOVES model in the development of the inventory estimates,
22 including use for uncertainty analysis. Although the Inventory uses some of the underlying data from
23 MOVES, such as vehicle age distributions by model year, MOVES is not used directly in calculating
24 mobile source emissions. The use of MOVES will be further explored.
25
26
3.2 Carbon Emitted from Non-Energy Uses of
Fossil Fuels (IPCC Source Category 1A)
27 In addition to being combusted for energy, fossil fuels are also consumed for non-energy uses (NEU) in the United
28 States. The fuels used for these purposes are diverse, including natural gas, liquefied petroleum gases (LPG),
29 asphalt (a viscous liquid mixture of heavy crude oil distillates), petroleum coke (manufactured from heavy oil), and
30 coal (metallurgical) coke (manufactured from coking coal). The non-energy applications of these fuels are equally
31 diverse, including feedstocks for the manufacture of plastics, rubber, synthetic fibers and other materials; reducing
32 agents for the production of various metals and inorganic products; and non-energy products such as lubricants,
33 waxes, and asphalt (IPCC 2006).
34 CO2 emissions arise from non-energy uses via several pathways. Emissions may occur during the manufacture of a
35 product, as is the case in producing plastics or rubber from fuel-derived feedstocks. Additionally, emissions may
36 occur during the product's lifetime, such as during solvent use. Overall, throughout the time series and across all
37 uses, about 60 percent of the total C consumed for non-energy purposes was stored in products, and not released to
38 the atmosphere; the remaining 40 percent was emitted.
39 There are several areas in which non-energy uses of fossil fuels are closely related to other parts of this Inventory.
40 For example, some of the NEU products release CO2 at the end of their commercial life when they are combusted
41 after disposal; these emissions are reported separately within the Energy chapter in the Incineration of Waste source
42 category. In addition, there is some overlap between fossil fuels consumed for non-energy uses and the fossil-
43 derived CO2 emissions accounted for in the Industrial Processes and Product Use chapter, especially for fuels used
44 as reducing agents. To avoid double-counting, the "raw" non-energy fuel consumption data reported by EIA are
45 modified to account for these overlaps. There are also net exports of petrochemicals that are not completely
Energy 3-39
-------
1 accounted for in the EIA data, and the inventory calculations adjust for the effect of net exports on the mass of C in
2 non-energy applications.
3 As shown in Table 3-19, fossil fuel emissions in 2014 from the non-energy uses of fossil fuels were 114.3 MMT
4 CO2 Eq., which constituted approximately 2 percent of overall fossil fuel emissions. In 2014, the consumption of
5 fuels for non-energy uses (after the adjustments described above) was 4,571.6 TBtu, an increase of 8.4 percent since
6 1990 (see Table 3-20). About 55.9 MMT (205.1 MMT CO2 Eq.) of the C in these fuels was stored, while the
7 remaining 31.2 MMT C (114.3 MMT CO2 Eq.) was emitted.
8 Table 3-19: COz Emissions from Non-Energy Use Fossil Fuel Consumption (MMT COz Eq. and
9 percent)
10
Year
Potential Emissions
C Stored
Emissions as a % of Potential
Emissions
1990 •
312.1
194.oB
38%H
118.1
2005
377.5
238.6B
37%
138.9
2010
325.1
211.0
35%
114.1
2011
316.6
208.1
34%
108.5
2012
311.9
206.2
34%
105.6
2013
327.1
205.4
37%
121.7
2014
319.5
205.1
36%
114.3
Methodology
11 The first step in estimating C stored in products was to determine the aggregate quantity of fossil fuels consumed for
12 non-energy uses. The C content of these feedstock fuels is equivalent to potential emissions, or the product of
13 consumption and the fuel-specific C content values. Both the non-energy fuel consumption and C content data were
14 supplied by the El A (2013, 2015b) (see Annex 2.1). Consumption of natural gas, LPG, pentanes plus, naphthas,
15 other oils, and special naphtha were adjusted to account for net exports of these products that are not reflected in the
16 raw data from EIA. Consumption values for industrial coking coal, petroleum coke, other oils, and natural gas in
17 Table 3-20 and Table 3-21 have been adjusted to subtract non-energy uses that are included in the source categories
18 of the Industrial Processes and Product Use chapter.56'57 Consumption values were also adjusted to subtract net
19 exports of intermediary chemicals.
20 For the remaining non-energy uses, the quantity of C stored was estimated by multiplying the potential emissions by
21 a storage factor.
22 • For several fuel types—petrochemical feedstocks (including natural gas for non-fertilizer uses, LPG,
23 pentanes plus, naphthas, other oils, still gas, special naphtha, and industrial other coal), asphalt and road oil,
24 lubricants, and waxes—U.S. data on C stocks and flows were used to develop C storage factors, calculated
25 as the ratio of (a) the C stored by the fuel's non-energy products to (b) the total C content of the fuel
26 consumed. A lifecycle approach was used in the development of these factors in order to account for losses
27 in the production process and during use. Because losses associated with municipal solid waste
28 management are handled separately in this sector under the Incineration of Waste source category, the
29 storage factors do not account for losses at the disposal end of the life cycle.
30 • For industrial coking coal and distillate fuel oil, storage factors were taken from IPCC (2006), which in turn
31 draws from Marland and Rotty (1984).
56 These source categories include Iron and Steel Production, Lead Production, Zinc Production, Ammonia Manufacture, Carbon
Black Manufacture (included in Petrochemical Production), Litanium Dioxide Production, Ferroalloy Production, Silicon
Carbide Production, and Aluminum Production.
57 Some degree of double counting may occur between these estimates of non-energy use of fuels and process emissions from
petrochemical production presented in the Industrial Processes and Produce Use sector. Data integration is not feasible at this
time as feedstock data from EIA used to estimate non-energy uses of fuels are aggregated by fuel type, rather than disaggregated
by both fuel type and particular industries (e.g., petrochemical production) as currently collected through EPA's GHGRP and
used for the petrochemical production category.
3-40 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 • For the remaining fuel types (petroleum coke, miscellaneous products, and other petroleum), IPCC does not
2 provide guidance on storage factors, and assumptions were made based on the potential fate of C in the
3 respective NEU products.
4 Table 3-20: Adjusted Consumption of Fossil Fuels for Non-Energy Uses (TBtu)
Year
Industry
Industrial Coking Coal
Industrial Other Coal
1990
4,215.
2005
8 5,110.9
+ B 80.4
8.
Natural Gas to Chemical Plants 281.
Asphalt & Road Oil
LPG
Lubricants
Pentanes Plus
Naphtha (<401 ° F)
Other Oil (>401 ° F)
Still Gas
Petroleum Coke
Special Naphtha
Distillate Fuel Oil
Waxes
Miscellaneous Products
Transportation
Lubricants
U.S. Territories
Lubricants
Other Petroleum (Misc. Prod,
Total
+ Does not exceed 0.05 TBtu
NA (Not applicable)
1,170,
1,120,
186.
117.
326.
662.
36,
27,
100.
7.
33,
137.
176.
176.
86.
0.
.) 86,
4,478.
,2
,6
.2
.5
,3
,6
,3
1
.7
.2
,9
,0
.3
11.9
260.9
1,323.2
1,610.1
160.2
95.5
679.6
499.5
67.7
2010
4,572
64
10
298
877
1,834
149
75
474
433
147
.7
.8
.3
.7
.8
.0
.5
.3
.5
.2
.8
2011
4,470.2
60.8
10.3
297.1
859.5
1,865.7
141.8
26.4
469.4
368.2
163.6
2012
4,377.4
132.5
10.3
292.7
826.7
1,887.3
130.5
40.3
432.2
267.4
160.6
2013
4,621.4
119.6
10.3
297.0
783.3
2,062.9
138.1
45.4
498.8
209.1
166.7
2014
4,571.6
23.0
10.3
305.1
792.6
2,109.4
144.0
43.5
435.2
236.2
164.6
105.2 + + + + +
60.9
11.7
31.4
,8| 112.8
0
,0
,7
,7
.0
,5
151.3
151.3
121.9
4.6
H 117.3
5,384.1
25
5
17
158
141
141
.3
.8
.1
.7
.2
.2
56.4
1
^1 55
.0
.4
4,770.3
Table 3-21: 2014 Adjusted Non-Energy Use Fossil Fuel
Sector/Fuel Type
Industry
Industrial Coking Coal
Industrial Other Coal
Natural Gas to
Chemical Plants
Asphalt & Road Oil
LPG
Lubricants
Pentanes Plus
Naphtha (<401°F)
Other Oil (>401°F)
Still Gas
Petroleum Coke
Special Naphtha
Distillate Fuel Oil
Adjusted
Non-Energy
Use3
(TBtu)
4,571.6
23.0
10.3
305.1
792.6
2,109.4
144.0
43.5
435.2
236.2
164.6
+
104.4
5.8
Carbon
Content
Coefficient
(MMT
C/QBtu)
NA
31.00
25.82
14.46
20.55
17.06
20.20
19.10
18.55
20.17
17.51
27.85
19.74
20.17
Potential
Carbon
(MMTC)
83.3
0.7
0.3
4.4
16.3
36.0
2.9
0.8
8.1
4.8
2.9
+
2.1
0.1
21.8
5.8
15.1
164.7
133.9
133.9
56.7
1.0
55.7
4,660.9
14.1
5.8
15.3
161.6
123.2
123.2
58.1
1.0
57.1
4,558.7
96.6
5.8
16.5
171.2
130.4
130.4
57.4
1.0
56.4
4,809.2
Consumption, Storage, and
Storage
Factor
NA
0.04
0.65
0.65
1.00
0.65
0.09
0.65
0.65
0.65
0.65
0.04
0.65
0.04
Carbon
Stored
(MMTC)
55.6
0.1
0.2
2.9
16.2
23.6
0.3
0.5
5.3
3.1
1.9
+
1.3
0.1
Carbon
Emissions
(MMTC)
27.7
0.6
0.1
1.5
0.1
12.4
2.6
0.3
2.8
1.6
1.0
+
0.7
0.1
104.4
5.8
14.8
182.7
136.0
136.0
53.6
1.0
52.6
4,761.2
Emissions
Carbon
Emissions
(MMT
CO2
Eq.)
101.6
2.4
0.3
5.6
0.3
45.6
9.7
1.1
10.2
6.0
3.6
+
2.6
0.2
Energy 3-41
-------
Waxes
Miscellaneous Products
Transportation
Lubricants
U.S. Territories
Lubricants
Other Petroleum (Misc.
Prod.)
Total
14.8
182.7
136.0
136.0
53.6
1.0
52.6
4,761.2
19.80
20.31
NA
20.20
NA
20.20
20.00
0.3
3.7
2.7
2.7
1.1
+
1.1
87.1
0.58
0.04
NA
0.09
NA
0.09
0.04
0.2
0.0
0.3
0.3
0.1
+
0.1
55.9
0.1
3.7
2.5
2.5
1.0
+
0.9
31.2
0.5
13.6
9.1
9.1
3.5
0.1
3.5
114.3
+ Does not exceed 0.05 TBtu
NA (Not applicable)
a To avoid double counting, net exports have been deducted.
Note: Totals may not sum due to independent rounding.
1 Lastly, emissions were estimated by subtracting the C stored from the potential emissions (see Table 3-19). More
2 detail on the methodology for calculating storage and emissions from each of these sources is provided in Annex
3 2.3.
4 Where storage factors were calculated specifically for the United States, data were obtained on (1) products such as
5 asphalt, plastics, synthetic rubber, synthetic fibers, cleansers (soaps and detergents), pesticides, food additives,
6 antifreeze and deicers (glycols), and silicones; and (2) industrial releases including energy recovery, Toxics Release
7 Inventory (TRI) releases, hazardous waste incineration, and volatile organic compound, solvent, and non-
8 combustion CO emissions. Data were taken from a variety of industry sources, government reports, and expert
9 communications. Sources include EPA reports and databases such as compilations of air emission factors (EPA
10 2001), National Emissions Inventory (NEI) Air Pollutant Emissions Trends Data (EPA2015a), Toxics Release
11 Inventory, 1998 (2000b), Biennial Reporting System (EPA2004, 2009), Resource Conservation and Recovery Act
12 Information System (EPA 2013b, 2015b), pesticide sales and use estimates (EPA 1998, 1999, 2002, 2004, 2011),
13 and the Chemical Data Access Tool (EPA 2012); the EIA Manufacturer's Energy Consumption Survey (MECS)
14 (EIA 1994, 1997, 2001, 2005, 2010, 2013b, 2015b); the National Petrochemical & Refiners Association (NPRA
15 2002); the U.S. Bureau of the Census (1999, 2004, 2009); Bank of Canada (2012, 2013, 2014); Financial Planning
16 Association (2006); INEGI (2006); the United States International Trade Commission (1990-2015); Gosselin,
17 Smith, and Hodge (1984); EPA's Municipal Solid Waste (MSW) Facts and Figures (EPA 2013a; 2014a); the Rubber
18 Manufacturers' Association (RMA 2009, 2011, 2014); the International Institute of Synthetic Rubber Products
19 (IISRP 2000, 2003); the Fiber Economics Bureau (FEE 2001-2013); the EPA Chemical Data Access Tool (CDAT)
20 (EPA 2014b); and the American Chemistry Council (ACC 2003-2011, 2012, 2013, 2014a, 2014b, 2015). Specific
21 data sources are listed in full detail in Annex 2.3.
22 Uncertainty and Time-Series Consistency
23 An uncertainty analysis was conducted to quantify the uncertainty surrounding the estimates of emissions and
24 storage factors from non-energy uses. This analysis, performed using @RISK software and the IPCC-recommended
25 Approach 2 methodology (Monte Carlo Stochastic Simulation technique), provides for the specification of
26 probability density functions for key variables within a computational structure that mirrors the calculation of the
27 inventory estimate. The results presented below provide the 95 percent confidence interval, the range of values
28 within which emissions are likely to fall, for this source category.
29 As noted above, the non-energy use analysis is based on U.S.-specific storage factors for (1) feedstock materials
30 (natural gas, LPG, pentanes plus, naphthas, other oils, still gas, special naphthas, and other industrial coal), (2)
31 asphalt, (3) lubricants, and (4) waxes. For the remaining fuel types (the "other" category in Table 3-20 and Table
32 3-21), the storage factors were taken directly from the 2006 IPCC Guidelines for National Greenhouse Gas
33 Inventories, where available, and otherwise assumptions were made based on the potential fate of carbon in the
34 respective NEU products. To characterize uncertainty, five separate analyses were conducted, corresponding to
35 each of the five categories. In all cases, statistical analyses or expert judgments of uncertainty were not available
3-42 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 directly from the information sources for all the activity variables; thus, uncertainty estimates were determined using
2 assumptions based on source category knowledge.
3 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 3-22 (emissions) and Table
4 3-23 (storage factors). Carbon emitted from non-energy uses of fossil fuels in 2014 was estimated to be between
5 86.2 and 162.9 MMT CCh Eq. at a 95 percent confidence level. This indicates a range of 25 percent below to 42
6 percent above the 2014 emission estimate of 114.3 MMT CCh Eq. The uncertainty in the emission estimates is a
7 function of uncertainty in both the quantity of fuel used for non-energy purposes and the storage factor.
8 Table 3-22: Approach 2 Quantitative Uncertainty Estimates for COz Emissions from Non-
9 Energy Uses of Fossil Fuels (MMT COz Eq. and Percent)
Source
Feedstocks
Asphalt
Lubricants
Waxes
Other
Total
2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
aS (MMTCChEq.) (MMT CCh Eq.) (%)
C02
CO2
CO2
C02
C02
CO2
75.1
0.3
18.9
0.5
19.6
114.3
Lower
Bound
49.6
0.1
15.5
0.3
14.1
86.2
Upper
Bound
125.3
0.6
21.9
0.7
21.7
162.9
Lower
Bound
-34%
-57%
-18%
-28%
-28%
-25%
Upper
Bound
67%
117%
16%
63%
11%
42%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence
interval.
10 Table 3-23: Approach 2 Quantitative Uncertainty Estimates for Storage Factors of Non-
11 Energy Uses of Fossil Fuels (Percent)
12
13
14
15
16
17
18
19
20
Source
Feedstocks
Asphalt
Lubricants
Waxes
Other
2014 Storage Factor
GaS (%)
CO2
CO2
C02
C02
CO2
65%
99.6%
9%
58%
4%
Uncertainty Range Relative to Emission Estimate3
(%) (%, Relative)
Lower
Bound
52%
99.1%
4%
49%
4%
Upper
Bound
72%
99.8%
17%
70%
24%
Lower
Bound
-20%
-0.5%
-57%
-15%
-3%
Upper
Bound
10%
0.25%
88%
22%
479%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence
interval, as a percentage of the inventory value (also expressed in percent terms).
In Table 3-23, feedstocks and asphalt contribute least to overall storage factor uncertainty on a percentage basis.
Although the feedstocks category—the largest use category in terms of total carbon flows—appears to have tight
confidence limits, this is to some extent an artifact of the way the uncertainty analysis was structured. As discussed
in Annex 2.3, the storage factor for feedstocks is based on an analysis of six fates that result in long-term storage
(e.g., plastics production), and eleven that result in emissions (e.g., volatile organic compound emissions). Rather
than modeling the total uncertainty around all of these fate processes, the current analysis addresses only the storage
fates, and assumes that all C that is not stored is emitted. As the production statistics that drive the storage values
are relatively well-characterized, this approach yields a result that is probably biased toward understating
uncertainty.
21 As is the case with the other uncertainty analyses discussed throughout this document, the uncertainty results above
22 address only those factors that can be readily quantified. More details on the uncertainty analysis are provided in
23 Annex 2.3.
Energy 3-43
-------
1 Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
2 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
3 above.
22
QA/QC and Verification
5 A source-specific QA/QC plan for non-energy uses of fossil fuels was developed and implemented. This effort
6 included a Tier 1 analysis, as well as portions of a Tier 2 analysis for non-energy uses involving petrochemical
7 feedstocks and for imports and exports. The Tier 2 procedures that were implemented involved checks specifically
8 focusing on the activity data and methodology for estimating the fate of C (in terms of storage and emissions) across
9 the various end-uses of fossil C. Emission and storage totals for the different subcategories were compared, and
10 trends across the time series were analyzed to determine whether any corrective actions were needed. Corrective
11 actions were taken to rectify minor errors and to improve the transparency of the calculations, facilitating future
12 QA/QC.
13 For petrochemical import and export data, special attention was paid to NAICS numbers and titles to verify that
14 none had changed or been removed. Import and export totals were compared for 2011 as well as their trends across
15 the time series.
16 Petrochemical input data reported by EIA will continue to be investigated in an attempt to address an input/output
17 discrepancy in the NEU model. Since 2001, the C accounted for in the feedstocks C balance outputs (i.e., storage
18 plus emissions) exceeds C inputs. Prior to 2001, the C balance inputs exceed outputs. Starting in 2001 through
19 2009, outputs exceeded inputs. In 2010 and 2011, inputs exceeded outputs, and in 2012, outputs slightly exceeded
20 inputs. A portion of this discrepancy has been reduced (see Recalculations Discussion, below) and two strategies
21 have been developed to address the remaining portion (see Planned Improvements, below).
Recalculations Discussion
23 A number of updates to historical production values were included in the most recent Monthly Energy Review; these
24 have been populated throughout this document.
25 Planned Improvements - TO BE UPDATED
26 There are several improvements planned for the future:
27 • More accurate accounting of C in petrochemical feedstocks. EPA has worked with EIA to determine the
28 cause of input/output discrepancies in the C mass balance contained within the NEU model. In the future,
29 two strategies to reduce or eliminate this discrepancy will continue to be pursued. First, accounting of C in
30 imports and exports will be improved. The import/export adjustment methodology will be examined to
31 ensure that net exports of intermediaries such as ethylene and propylene are fully accounted for. Second,
32 reconsider the use of top-down C input calculation in estimating emissions will be reconsidered.
33 Alternative approaches that rely more substantially on the bottom-up C output calculation will be
34 considered instead.
35 • Response to potential changes in NEU input data. In 2013 EIA initiated implementation of new data
36 reporting definitions for Natural Gas Liquids (NGL) and Liquefied Petroleum Gases (LPG); the new
37 definitions may affect the characterization of the input data that EIA provides for the NEU model and may
38 therefore result in the need for changes to the NEU methodology. EIA also obtains and applies proprietary
39 data for LPG inputs that are not directly applied as NEU input data because the data are proprietary. The
40 potential use of the proprietary data (in an aggregated, non-proprietary form) as inputs to the NEU model
41 will be investigated with EIA.
42 • Improving the uncertainty analysis. Most of the input parameter distributions are based on professional
43 judgment rather than rigorous statistical characterizations of uncertainty.
44 • Better characterizing flows of fossil C. Additional fates may be researched, including the fossil C load in
45 organic chemical wastewaters, plasticizers, adhesives, films, paints, and coatings. There is also a need to
3-44 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 further clarify the treatment of fuel additives and backflows (especially methyl tert-butyl ether, MTBE).
2 • Reviewing the trends in fossil fuel consumption for non-energy uses. Annual consumption for several fuel
3 types is highly variable across the time series, including industrial coking coal and other petroleum
4 (miscellaneous products). A better understanding of these trends will be pursued to identify any
5 mischaracterized or misreported fuel consumption for non-energy uses. For example, "miscellaneous
6 products" category includes miscellaneous products that are not reported elsewhere in the EIA data set.
7 The EIA does not have firm data concerning the amounts of various products that are being reported in the
8 "miscellaneous products" category; however, EIA has indicated that recovered sulfur from petroleum and
9 natural gas processing, and potentially also C black feedstock could be reported in this category.
10 Recovered sulfur would not be reported in the NEU calculation or elsewhere in the Inventory.
11 • Updating the average C content of solvents was researched, since the entire time series depends on one
12 year's worth of solvent composition data. Unfortunately, the data on C emissions from solvents that were
13 readily available do not provide composition data for all categories of solvent emissions and also have
14 conflicting definitions for volatile organic compounds, the source of emissive C in solvents. Additional
15 sources of solvents data will be identified in order to update the C content assumptions.
16 • Updating the average C content of cleansers (soaps and detergents) was researched; although production
17 and consumption data for cleansers are published every 5 years by the Census Bureau, the composition (C
18 content) of cleansers has not been recently updated. Recently available composition data sources may
19 facilitate updating the average C content for this category.
20 • Revising the methodology for consumption, production, and C content of plastics was researched; because
21 of recent changes to the type of data publicly available for plastics, the NEU model for plastics applies data
22 obtained from personal communications. Potential revisions to the plastics methodology to account for the
23 recent changes in published data will be investigated.
24 • Although U.S.-specific storage factors have been developed for feedstocks, asphalt, lubricants, and waxes,
25 default values from IPCC are still used for two of the non-energy fuel types (industrial coking coal,
26 distillate oil), and broad assumptions are being used for miscellaneous products and other petroleum. Over
27 the long term, there are plans to improve these storage factors by analyzing C fate similar to those
28 described in Annex 2.3 or deferring to more updated default storage factors from IPCC where available.
29 • Reviewing the storage of carbon black across various sectors in the Inventory; in particular, the carbon
30 black abraded and stored in tires.
31
32
Box 3-6: Reporting of Lubricants, Waxes, and Asphalt and Road Oil Product Use in Energy Sector
33 The 2006 IPCC Guidelines provides methodological guidance to estimate emissions from the first use of fossil fuels
34 as a product for primary purposes other than combustion for energy purposes (including lubricants, paraffin waxes,
35 bitumen/asphalt, and solvents) under the Industrial Processes and Product Use (IPPU) sector.58 In this Inventory, C
36 storage and C emissions from product use of lubricants, waxes, and asphalt and road oil are reported under the
37 Energy sector in the Carbon Emitted from Non-Energy Uses of Fossil Fuels source category (IPCC Source Category
38 1A).59
39 The emissions are reported in the Energy sector, as opposed to the IPPU sector, to reflect national circumstances in
40 its choice of methodology and to increase transparency of this source category's unique country-specific data
41 sources and methodology. The country-specific methodology used for the Carbon Emitted from Non-Energy Uses of
42 Fossil Fuels source category is based on a carbon balance (i.e., C inputs-outputs) calculation of the aggregate
58 See Volume 3: Industrial Processes and Product Use, Chapter 5: Non-Energy Products from Fuels and Solvent Use of the
2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2006).
Non-methane volatile organic compound (NMVOC) emissions from solvent use are reported separately in the IPPU sector,
following Chapter 5 of the 2006 IPCC Guidelines.
Energy 3-45
-------
1 amount of fossil fuels used for non-energy uses, including inputs of lubricants, waxes, asphalt and road oil (see
2 section 3.2, Table 3-21). For those inputs, U.S. country-specific data on C stocks and flows are used to develop
3 carbon storage factors, which are calculated as the ratio of the C stored by the fossil fuel non-energy products to the
4 total C content of the fuel consumed, taking into account losses in the production process and during product use.60
5 The country-specific methodology to reflect national circumstances starts with the aggregate amount of fossil fuels
6 used for non-energy uses and applies a C balance calculation, breaking out the C emissions from non-energy use of
7 lubricants, waxes, and asphalt and road oil. Due to U.S. national circumstances, reporting these C emissions
8 separately under IPPU would involve making artificial adjustments to both the C inputs and C outputs of the non-
9 energy use C balance. These artificial adjustments would also result in the C emissions for lubricants, waxes, and
10 asphalt and road oil being reported under IPPU, while the C storage for lubricants, waxes, and asphalt and road oil
11 would be reported under Energy. To avoid presenting an incomplete C balance and a less transparent approach for
12 the Carbon Emitted from Non-Energy Uses of Fossil Fuels source category calculation, the entire calculation of C
13 storage and C emissions is therefore conducted in the Non-Energy Uses of Fossil Fuels category calculation
14 methodology, and both the C storage and C emissions for lubricants, waxes, and asphalt and road oil are reported
15 under the Energy sector.
16
17
18
3.3 Incineration of Waste (IPCC Source
Category lAla)
19 Incineration is used to manage about 7 to 19 percent of the solid wastes generated in the United States, depending on
20 the source of the estimate and the scope of materials included in the definition of solid waste (EPA 2000, Goldstein
21 and Madtes 2001, Kaufman et al. 2004, Simmons et al. 2006, van Haaren et al. 2010). In the context of this section,
22 waste includes all municipal solid waste (MSW) as well as scrap tires. In the United States, almost all incineration of
23 MSW occurs at waste-to-energy facilities or industrial facilities where useful energy is recovered, and thus
24 emissions from waste incineration are accounted for in the Energy chapter. Similarly, scrap tires are combusted for
25 energy recovery in industrial and utility boilers, pulp and paper mills, and cement kilns. Incineration of waste results
26 in conversion of the organic inputs to CCh. According to IPCC guidelines, when the CO2 emitted is of fossil origin,
27 it is counted as a net anthropogenic emission of CC>2 to the atmosphere. Thus, the emissions from waste incineration
28 are calculated by estimating the quantity of waste combusted and the fraction of the waste that is C derived from
29 fossil sources.
30 Most of the organic materials in municipal solid wastes are of biogenic origin (e.g., paper, yard trimmings), and
31 have their net C flows accounted for under the Land Use, Land-Use Change, and Forestry chapter. However, some
32 components—plastics, synthetic rubber, synthetic fibers, and carbon black in scrap tires—are of fossil origin.
33 Plastics in the U.S. waste stream are primarily in the form of containers, packaging, and durable goods. Rubber is
34 found in durable goods, such as carpets, and in non-durable goods, such as clothing and footwear. Fibers in
35 municipal solid wastes are predominantly from clothing and home furnishings. As noted above, scrap tires (which
36 contain synthetic rubber and carbon black) are also considered a "non-hazardous" waste and are included in the
37 waste incineration estimate, though waste disposal practices for tires differ from municipal solid waste. Estimates on
38 emissions from hazardous waste incineration can be found in Annex 2.3 and are accounted for as part of the C mass
3 9 balance for non-energy uses of fossil fuels.
40 Approximately 29.6 million metric tons of MSW were incinerated in the United States in 2013 (EPA 2015). Data for
41 the amount of MSW incinerated in 2014 were not available, so data for 2014 was assumed to be equal to data for
42 2013. CO2 emissions from incineration of waste rose 18 percent since 1990, to an estimated 9.4 MMT CO2 Eq.
43 (9,421 kt) in 2014, as the volume of scrap tires and other fossil C-containing materials in waste increased (see Table
44 3-24 and Table 3-25). Waste incineration is also a source of CH4 and N2O emissions (De Soete 1993, IPCC 2006).
Data and calculations for lubricants and waxes and asphalt and road oil are in Annex 2.3: Methodology and Data for
Estimating CCh Emissions from Fossil Fuel Combustion.
3-46 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 CH4 emissions from the incineration of waste were estimated to be less than 0.05 MMT CO2 Eq. (less than 0.5 kt
2 CEU) in 2014, and have not changed significantly since 1990. N2O emissions from the incineration of waste were
3 estimated to be 0.3 MMT CO2 Eq. (1 kt N2O) in 2014, and have not changed significantly since 1990.
4 Table 3-24: COz, CH4, and NzO Emissions from the Incineration of Waste (MMT COz Eq.)
Gas/Waste Product
C02
Plastics
Synthetic Rubber in Tires
Carbon Black in Tires
Synthetic Rubber in
MSW
Synthetic Fibers
CH4
N20
Total
a Set equal to 20 13 value.
Table 3-25: COz, CH4, and
Gas/Waste Product
CO2
Plastics
Synthetic Rubber in Tires
Carbon Black in Tires
Synthetic Rubber in MSW
Synthetic Fibers
CH4
N20
1990
8.0
5.6
0.3
0.4
0.9
0.8
+
0.5
8.4
NzO
1990
7,972
5,588
308
385
854
838
2
2005
12.5
6.9
1.6
2.0
0.8
1.2
+
0.4
12.8
Emissions
2005
12,454
6,919
1,599
1,958
765
1,212
1
from
2010
11.0
6.0
1.5
1.8
0.7
1.1
+
0.3
11.4
the
2010
11,026
5,969
1,461
1,783
701
1,112
1
2011
10.5
5.8
1.4
1.7
0.7
1.1
+
0.3
10.9
2012
10.4
5.7
1.3
1.5
0.7
1.1
+
0.3
10.7
2013
9
4
1
1
0
1
.4
.9
.2
.4
.7
.3
+
2014a
0.3
9
Incineration of Waste
2011
10,550
5,757
1,363
1,663
712
1,056
1
2012
10,362
5,709
1,262
1,537
705
1,149
1
2013
9,421
4,857
1,158
1,412
729
1,265
1
.7
9.4
4.9
1.2
1.4
0.7
1.3
+
0.3
9.7
(kt)
2014a
9,421
4,857
1,158
1,412
729
1,265
1
' Set equal to 2013 value.
Methodology
7 Emissions of CO2 from the incineration of waste include CO2 generated by the incineration of plastics, synthetic
8 fibers, and synthetic rubber in MSW, as well as the incineration of synthetic rubber and carbon black in scrap tires.
9 These emissions were estimated by multiplying the amount of each material incinerated by the C content of the
10 material and the fraction oxidized (98 percent). Plastics incinerated in municipal solid wastes were categorized into
11 seven plastic resin types, each material having a discrete C content. Similarly, synthetic rubber is categorized into
12 three product types, and synthetic fibers were categorized into four product types, each having a discrete C content.
13 Scrap tires contain several types of synthetic rubber, carbon black, and synthetic fibers. Each type of synthetic
14 rubber has a discrete C content, and carbon black is 100 percent C. Emissions of CO2 were calculated based on the
15 amount of scrap tires used for fuel and the synthetic rubber and carbon black content of scrap tires.
16 More detail on the methodology for calculating emissions from each of these waste incineration sources is provided
17 in Annex 3.7.
18 For each of the methods used to calculate CO2 emissions from the incineration of waste, data on the quantity of
19 product combusted and the C content of the product are needed. For plastics, synthetic rubber, and synthetic fibers in
20 MSW, the amount of specific materials discarded as municipal solid waste (i.e., the quantity generated minus the
21 quantity recycled) was taken from Municipal Solid Waste Generation, Recycling, and Disposal in the United States:
22 Facts and Figures (EPA 2000 through 2003, 2005 through 2014), Advancing Sustainable Materials Management:
23 Facts and Figures 2013: Assessing Trends in Material Generation, Recycling and Disposal in the United States
24 (EPA 2015) and detailed unpublished backup data for some years not shown in the reports (Schneider 2007). For
25 2014, the amount of MSW incinerated was assumed to be equal to that in 2013, due to the lack of available data.
26 The proportion of total waste discarded that is incinerated was derived from Shin (2014). Data on total waste
27 incinerated was not available for 2012 through 2014, so these values were assumed to equal to the 2011 value. For
Energy 3-47
-------
1 synthetic rubber and carbon black in scrap tires, information was obtained from U. S. Scrap Tire Management
2 Summary for 2005 through 2013 data (RMA 2014). Average C contents for the "Other" plastics category and
3 synthetic rubber in municipal solid wastes were calculated from 1998 and 2002 production statistics: C content for
4 1990 through 1998 is based on the 1998 value; C content for 1999 through 2001 is the average of 1998 and 2002
5 values; and C content for 2002 to date is based on the 2002 value. Carbon content for synthetic fibers was calculated
6 from 1999 production statistics. Information about scrap tire composition was taken from the Rubber
7 Manufacturers' Association internet site (RMA 2012a).
8 The assumption that 98 percent of organic C is oxidized (which applies to all waste incineration categories for CO2
9 emissions) was reported in EPA's life cycle analysis of greenhouse gas emissions and sinks from management of
10 solid waste (EPA 2006).
11 Incineration of waste, including MS W, also results in emissions of N2O and CH4. These emissions were calculated
12 as a function of the total estimated mass of waste incinerated and emission factors. As noted above, N2O and CH4
13 emissions are a function of total waste incinerated in each year; for 1990 through 2008, these data were derived from
14 the information published in BioCycle (van Haaren et al. 2010). Data for 2009 and 2010 were interpolated between
15 2008 and 2011 values. Data for 2011 were derived from Shin (2014). Data on total waste incinerated was not
16 available in the BioCycle data set for 2012 through2014, so these values were assumed to equal the 2011 Biocycle
17 data set value.
18 Table 3-26 provides data on municipal solid waste discarded and percentage combusted for the total waste
19 stream.The emission factors of N2O and CH4 emissions per quantity of municipal solid waste combusted are default
20 emission factors for the default continuously-fed stoker unit MSW incineration technology type and were taken from
21 the 2006IPCC Guidelines (IPCC 2006).
22 Table 3-26: Municipal Solid Waste Generation (Metric Tons) and Percent Combusted
23 (BioCycle data set)
24
Incinerated (% of
Year Waste Discarded Waste Incinerated Discards)
1990 235,733,657 30,632,057 13.0%
10.0%
2010
2011
2012
2013
2014
271,592,991
273,116,704
273,116,704a
273,116,704a
273,1 16,704a
22,714,122
20,756,870
20,756,870
20,756,870
20,756,870
8.0%
7.6%
7.6%
7.6%
7.6%
a Assumed equal to 2011 value.
Source: van Haaren et al. (2010)
Uncertainty and Time-Series Consistency
25 An Approach 2 Monte Carlo analysis was performed to determine the level of uncertainty surrounding the estimates
26 of CO2 emissions and N2O emissions from the incineration of waste (given the very low emissions for CH4, no
27 uncertainty estimate was derived). IPCC Approach 2 analysis allows the specification of probability density
28 functions for key variables within a computational structure that mirrors the calculation of the Inventory estimate.
29 Uncertainty estimates and distributions for waste generation variables (i.e., plastics, synthetic rubber, and textiles
30 generation) were obtained through a conversation with one of the authors of the Municipal Solid Waste in the
31 United States reports. Statistical analyses or expert judgments of uncertainty were not available directly from the
32 information sources for the other variables; thus, uncertainty estimates for these variables were determined using
33 assumptions based on source category knowledge and the known uncertainty estimates for the waste generation
34 variables.
35 The uncertainties in the waste incineration emission estimates arise from both the assumptions applied to the data
36 and from the quality of the data. Key factors include MSW incineration rate; fraction oxidized; missing data on
3-48 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 waste composition; average C content of waste components; assumptions on the synthetic/biogenic C ratio; and
2 combustion conditions affecting N2O emissions. The highest levels of uncertainty surround the variables that are
3 based on assumptions (e.g., percent of clothing and footwear composed of synthetic rubber); the lowest levels of
4 uncertainty surround variables that were determined by quantitative measurements (e.g., combustion efficiency, C
5 content of C black).
6 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 3-27. Waste incineration
7 CO2 emissions in 2014 were estimated to be between 8.5 and 11.5 MMT CCh Eq. at a 95 percent confidence level.
8 This indicates a range of 10 percent below to 14 percent above the 2014 emission estimate of 9.4 MMT CO2 Eq.
9 Also at a 95 percent confidence level, waste incineration N2O emissions in 2014 were estimated to be between 0.1
10 and 0.8 MMT CCh Eq. This indicates a range of 53 percent below to 163 percent above the 2014 emission estimate
11 of0.3MMTCO2Eq.
12 Table 3-27: Approach 2 Quantitative Uncertainty Estimates for COz and NzO from the
13 Incineration of Waste (MMT COz Eq. and Percent)
2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
Source Gas (MMT CCh Eq.) (MMT CCh Eq.) (%)
17
24
34
Incineration of Waste
Incineration of Waste
CO2
N20
9.4
0.3
Lower
Bound
8.5
0.1
Upper
Bound
11.5
0.8
Lower
Bound
-10%
-53%
Upper
Bound
+14%
+163%
a Range of emission estimates predicted by Monte Carlo Simulation for a 95 percent confidence interval.
14 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
15 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
16 above.
QA/QC and Verification
18 A source-specific QA/QC plan was implemented for incineration of waste. This effort included a Tier 1 analysis, as
19 well as portions of a Tier 2 analysis. The Tier 2 procedures that were implemented involved checks specifically
20 focusing on the activity data and specifically focused on the emission factor and activity data sources and
21 methodology used for estimating emissions from incineration of waste. Trends across the time series were analyzed
22 to determine whether any corrective actions were needed. Actions were taken to streamline the activity data
23 throughout the calculations on incineration of waste.
Recalculations Discussion
25 For the current Inventory, emission estimates for 2013 have been updated based on Advancing Sustainable
26 Materials Management: Facts and Figures 2013: Assessing Trends in Material Generation, Recycling and Disposal
27 in the United States (EPA 2015).
28 The data which calculates the percent incineration was updated in the current Inventory. Biocycle has not released a
29 new State of Garbage in America Report since 2010 (with 2008 data), which used to be a semi-annual publication
30 which publishes the results of the nation-wide MSW survey. The results of the survey have been published in Shin
31 2014.This provided updated incineration data for 2011, so the generation and incineration data for 2012 through
32 2014 are assumed equivalent to the 2011 values. The data for 2009 and 2010 were based on interpolations between
33 2008 and 2011.
Planned Improvements
35 The availability of facility-level waste incineration data through EPA's GHGPJ3 will be examined to help better
36 characterize waste incineration operations in the United States. This characterization could include future
37 improvements as to the operations involved in waste incineration for energy, whether in the power generation sector
38 or the industrial sector. Additional examinations will be necessary as, unlike the reporting requirements for this
Energy 3-49
-------
1 chapter under the UNFCCC reporting guidelines,61 some facility-level waste incineration emissions reported under
2 EPA's GHGRP may also include industrial process emissions. In line with UNFCCC reporting guidelines,
3 emissions for waste incineration with energy recovery are included in this chapter, while process emissions are
4 included in the Industrial Processes and Product Use chapter of this report. In examining data from EPA's GHGRP
5 that would be useful to improve the emission estimates for the waste incineration category, particular attention will
6 also be made to ensure time series consistency, as the facility-level reporting data from EPA's GHGRP are not
7 available for all inventory years as reported in this Inventory. Additionally, analyses will focus on ensuring CCh
8 emissions from the biomass component of waste are separated in the facility-level reported data, and on maintaining
9 consistency with national waste generation and fate statistics currently used to estimate total, national U.S.
10 greenhouse gas emissions. In implementing improvements and integration of data from EPA's GHGRP, the latest
11 guidance from the IPCC on the use of facility-level data in national inventories will be relied upon.62 GHGRP data
12 is available for MSW combustors, which contains information on the CO2, CH4, and N2O emissions from MSW
13 combustion, plus the fraction of the emissions that are biogenic. To calculate biogenic versus total CO2 emissions, a
14 default biogenic fraction of 0.6 is used. The biogenic fraction will be calculated using the current input data and
15 assumptions to verify the current MSW emission estimates.
16 If GHGRP data would not provide a more accurate estimate of the amount of solid waste combusted, new data
17 sources for the total MSW generated will be explored given that the data previously published semi-annually in
18 Biocycle (vanHaaren et al. 2010) has ceased to be published, according to the authors. Equivalent data was derived
19 from Shin (2014)for2011. A new methodology would be developed based on the available data within the annual
20 update of EPA's Advancing Sustainable Materials Management: Facts and Figures 2013: Assessing Trends in
21 Material Generation, Recycling and Disposal in the United States (EPA 2015). In developing the new
22 methodology, appropriate assumptions would need to be made to ensure that the MSW figures included all waste.
23 Additionally, the carbon content of the synthetic fiber will be updated based on each year's production mix.
24 Additional improvements will be conducted to improve the transparency in the current reporting of waste
25 incineration. Currently, hazardous industrial waste incineration is included within the overall calculations for the
26 Carbon Emitted from Non-Energy Uses of Fossil Fuels category. Waste incineration activities that do not include
27 energy recovery will be examined. Synthetic fibers within scrap tires are not included in this analysis and will be
28 explored for future inventories. The carbon content of fibers within scrap tires would be used to calculate the
29 associated incineration emissions. Updated fiber content data from the Fiber Economics Bureau will also be
30 explored.
31
32
3.4 Coal Mining (IPCC Source Category IBla)
(TO BE UPDATED)
33 Three types of coal mining-related activities release CH4 to the atmosphere: underground mining, surface mining,
34 and post-mining (i.e., coal-handling) activities. While surface mines account for the majority of U.S. coal
35 production (see Table 3-30), underground coal mines contribute the largest share of CH4 emissions (see Table 3-28
36 and Table 3-29) due to the higher CH4 content of coal in the deeper underground coal seams. In 2013, 395
37 underground coal mines and 637 surface mines were operating in the U.S. Also in 2013, the U.S. was the second
3 8 largest coal producer in the world (891 MMT), after China (3,561 MMT) and followed by India (613 MMT) (IEA
39 2014).
40 Underground mines liberate CH4 from ventilation systems and from degasification systems. Ventilation systems
41 pump air through the mine workings to dilute noxious gases and ensure worker safety; these systems can exhaust
42 significant amounts of CH4 to the atmosphere in low concentrations. Degasification systems are wells drilled from
43 the surface or boreholes drilled inside the mine that remove large, often highly-concentrated, volumes of CH4
61 See .
62 See.
3-50 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 before, during, or after mining. Some mines recover and use CH4 generated from ventilation and degasification
2 systems, thereby reducing emissions to the atmosphere.
3 Surface coal mines liberate CH4 as the overburden is removed and the coal is exposed to the atmosphere. Methane
4 emissions are normally a function of coal rank and depth. Surface coal mines typically produce lower rank coals and
5 remove less than 250 feet of overburden, thus the level of emissions is much lower than from underground mines.
6 In addition, CH4 is released during post-mining activities, as the coal is processed, transported and stored for use.
7 Total CH4 emissions in 2013 were estimated to be 64.6 MMT CC>2 Eq. (2,584 kt CH4), a decline of 33 percent since
8 1990 (see Table 3-28 and Table 3-29). Of this amount, underground mines accounted for approximately 71.6
9 percent, surface mines accounted for 15.0 percent, and post-mining emissions accounted for 13.4 percent.
10 Table 3-28: CH4 Emissions from Coal Mining (MMT COz Eq.)
12
20
Activity
UG Mining
Liberated
Recovered & Used
Surface Mining
Post-Mining (Under Ground)
Post-Mining (Surface)
Total
1990
74.2
80.8
(6.6)
10.8
9.2
2.3
96.5
2005
42.0
59.7
(17.7) 1
11.9
7.6
2.6
64.1
2009
59.2
78.7
(19.5)
11.5
6.7
2.5
79.9
2010
61.6
85.2
(23.6)
11.5
6.8
2.5
82.3
2011
50.2
71.0
(20.8)
11.6
6.9
2.5
71.2
2012
47.3
65.8
(18.5)
10.3
6.7
2.2
66.5
2013
46.2
65.8
(19.6)
9.7
6.6
2.1
64.6
Note: Totals may not sum due to independent rounding. Parentheses indicate negative values.
11 Table 3-29: CH4 Emissions from Coal Mining (kt)
Activity
UG Mining
Liberated
Recovered & Used
Surface Mining
Post-Mining (UG)
Post-Mining (Surface)
1990 1 | 2005 2009
2,968
3,234
(266)
430
368
93
1,682
2,390
(708)
475
306
103
2,367
3,149
(782)
461
267
100
2010
2,463
3,406
(943)
461
270
100
2011
2,008
2,839
(831)
465
276
101
2012
1,891
2,631
(740)
410
268
89
2013
1,849
2,633
(784)
388
263
84
Total 3,860 2,565 3,194 3,293 2,849 2,658 2,584
Note: Totals may not sum due to independent rounding. Parentheses indicate negative values.
Methodology
13 The methodology for estimating CH4 emissions from coal mining consists of two steps. The first step is to estimate
14 emissions from underground mines. There are two sources of underground mine emissions: ventilation systems and
15 degasification systems. These emissions are estimated on a mine-by-mine basis and then are summed to determine
16 total emissions. The second step of the analysis involves estimating CH4 emissions from surface mines and post-
17 mining activities. In contrast to the methodology for underground mines, which uses mine-specific data, the
18 methodology for estimating emissions from surface mines and post-mining activities consists of multiplying basin-
19 specific coal production by basin-specific gas content and an emission factor.
Step 1: Estimate CH4 Liberated and CH4 Emitted from Underground Mines
21 Underground mines generate CH4 from ventilation systems and from degasification systems. Some mines recover
22 and use the generated CH4, thereby reducing emissions to the atmosphere. Total CH4 emitted from underground
23 mines equals the CH4 liberated from ventilation systems, plus the CH4 liberated from degasification systems, minus
24 the CH4 recovered and used.
25 Step 1.1: Estimate CH4 Liberated from Ventilation Systems
Energy 3-51
-------
1 Because the U.S. Mine Safety and Health Administration (MSHA) samples CH4 emissions from ventilation systems
2 for all mines with detectable CH4 concentrations63 to ensure miner safety, these mine-by-mine measurements are
3 used to estimate CH4 emissions from ventilation systems. Since 2011, the EPA has also collected information on
4 ventilation emissions from underground coal mines liberating greater than 36,500,000 actual cubic feet of CH4 per
5 year (about 14,700 metric tons CC>2 Eq.) through its GHGRP (EPA 2014).64 Many of the underground coal mines
6 reporting to EPA's GHGRP use the quarterly CH4 emission data collected by MSHA. However, some mines use
7 their own measurements and samples, which are taken on a quarterly basis. The 2013 ventilation emissions were
8 calculated using the GHGRP data from the mines that take their own measurements and the MSHA data for all other
9 mines.
10 Step 1.2: Estimate CH4 Liberated from Degasification Systems
11 Some gassier underground mines also use degasification systems (e.g., wells or boreholes) to remove CH4 before,
12 during, or after mining. This CH4 can then be collected for use or vented to the atmosphere. Several data sets were
13 used to estimate the quantity of CH4 collected by each of the twenty-four mines using degasification systems in
14 2013. For Alabama mines that sold recovered CH4 to a pipeline, pipeline sales data published by state petroleum
15 and natural gas agencies were used to estimate degasification emissions. The well data was also used to estimate
16 CH4 collected from mined-through pre-drainage wells. For most other mines that either sold CH4 to a pipeline, used
17 CH4 on site, or vented CH4 from degasification systems, data on degasification emissions reported to the EPA's
18 GHGRP (EPA 2014) were used.
19 Step 1.3: Estimate CH4 Recovered from Degasification Systems and Utilized (Emissions
20 Avoided)
11 Finally, the amount of CH4 recovered by degasification and ventilation systems and then used (i.e., not vented) was
22 estimated. In 2013, fifteen active coal mines had CH4 recovery and use projects, of which thirteen mines sold the
23 recovered CH4 to a pipeline. One of the mines that sold gas to a pipeline also used CH4 to fuel a thermal coal dryer.
24 One mine used recovered CH4 for electrical power generation, and two other mines used recovered CH4 to heat mine
25 ventilation air or dry coal. Emissions avoided as a result of pipeline sales projects at Alabama and West Virginia
26 mines were estimated using gas sales data reported by the state agencies. For all other mines with pipeline sales or
27 used methane for electric power or heating, either the coal mine operators or project developers supplied information
28 regarding methane recovery or GHGRP data were used.
29 Step 2: Estimate ChU Emitted from Surface Mines and Post-Mining Activities
30 Mine-specific data were not available for estimating CH4 emissions from surface coal mines or for post-mining
31 activities. For surface mines, basin-specific coal production obtained from the Energy Information Administration's
32 Annual Coal Report (see Table 3-30) (EIA 2014) was multiplied by basin-specific gas contents and a 150 percent
33 emission factor (to account for CH4from over- and under-burden) to estimate CH4 emissions. The emission factor
34 was revised downward in 2012 from 200 percent, based on more recent studies in Canada and Australia (King 1994,
35 Saghafi 2013). The 150 percent emission factor was applied to all inventory years since 1990, retroactively. For
36 post-mining activities, basin-specific coal production was multiplied by basin-specific gas contents and a 32.5
37 percent emission factor for CH4 desorption during coal transportation and storage (Greedy 1993). Basin-specific in
38 situ gas content data was compiled from AAPG (1984) and USBM (1986). Beginning in 2006, revised data on in
39 situ CH4 content and emission factors have been taken from EPA (1996) and EPA (2005).
40 Table 3-30: Coal Production (kt)
Year
1990
2005
Underground
384,244
334,398
Surface
546,808
691,448
Total
931,052
1
1,025,846
63 MSHA records coal mine CH4 readings with concentrations of greater than 50 ppm (parts per million) CH4. Readings below
this threshold are considered non-detectable.
64 Underground coal mines report to EPA under Subpart FF of the program.
3-52 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
30
1 I
2009
2010
2011
2012
2013
301,241
305,862
313,529
310,608
309,546
671,475
676,177
684,807
610,307
581,270
1
972,716
982,039
998,337
920,915
890,815
Uncertainty and Time-Series Consistency
2 A quantitative uncertainty analysis was conducted for the coal mining source category using the IPCC-
3 recommended Approach 2 uncertainty estimation methodology. Because emission estimates from underground
4 ventilation systems were based on actual measurement data from MSHA or EPA's GHGRP, uncertainty is relatively
5 low. A degree of imprecision was introduced because the measurements used were not continuous but rather an
6 average of quarterly instantaneous readings. Additionally, the measurement equipment used can be expected to
7 have resulted in an average of 10 percent overestimation of annual CH4 emissions (Mutmansky & Wang 2000).
8 GHGRP data was used for a number of the mines beginning in 2013, however, the equipment uncertainty is applied
9 to both MSHA and GHGRP data.
10 Estimates of CH4 recovered by degasification systems are relatively certain for utilized CH4 because of the
11 availability of gas sales information. In addition, many coal mine operators provided information on mined-through
12 dates for pre-drainage wells. Many of the recovery estimates use data on wells within 100 feet of a mined area.
13 However, uncertainty exists concerning the radius of influence of each well. The number of wells counted, and thus
14 the avoided emissions, may vary if the drainage area is found to be larger or smaller than estimated.
15 Continuous CH4 monitoring is required of mines that report utilized methane on or off-site to EPA's GHGRP.
16 Beginning in 2013, use of GHGRP data for mines without publicly-available gas usage or sales records has reduced
17 the uncertainty from previous estimations. In addition, since 2012, GHGRP data has been used to estimate CH4
18 emissions from vented degasification wells, thus reducing the uncertainty associated with that subsource.
19 Compared to underground mines, there is considerably more uncertainty associated with surface mining and post-
20 mining emissions because of the difficulty in developing accurate emission factors from field measurements.
21 However, since underground emissions comprise the majority of total coal mining emissions, the uncertainty
22 associated with underground emissions is the primary factor that determines overall uncertainty. The results of the
23 Approach 2 quantitative uncertainty analysis are summarized in Table 3-31. Coal mining CH4 emissions in 2013
24 were estimated to be between 56.6 and 74.7 MMT CC>2 Eq. at a 95 percent confidence level. This indicates a range
25 of 12.4 percent below to 15.6 percent above the 2013 emission estimate of 64.6 MMT CC^Eq.
26 Table 3-31: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Coal
27 Mining (MMT COz Eq. and Percent)
„ „ 2013 Emission Estimate Uncertainty Range Relative to Emission Estimate3
source ixas (MMT CCh Eq.) (MMT CCh Eq.) (%)
Coal Mining CH4 64.6
Lower
Bound
56.6
Upper
Bound
74.7
Lower
Bound
-12.4%
Upper
Bound
+15.6%
1 Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
28 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
29 through 2013. Details on the emission trends through time are described in more detail in the Methodology section.
Recalculations Discussion
31 For the current Inventory, emission estimates have been revised to reflect the GWPs provided in the IPCC Fourth
32 Assessment Report (AR4) (IPCC 2007). AR4 GWP values differ slightly from those presented in the IPCC Second
33 Assessment Report (SAR) (IPCC 1996) (used in the previous inventories) which results in time-series recalculations
34 for most inventory sources. Under the most recent reporting guidelines (UNFCCC 2014), countries are required to
Energy 3-53
-------
1 report using the AR4 GWPs, which reflect an updated understanding of the atmospheric properties of each
2 greenhouse gas. The GWPs of CH4 and most fluorinated greenhouse gases have increased, leading to an overall
3 increase in CCh-equivalent emissions from CH4. The GWPs of N2O and SF6 have decreased, leading to a decrease in
4 CCh-equivalent emissions for these greenhouse gases. The AR4 GWPs have been applied across the entire time
5 series for consistency. For more information please see the Recalculations and Improvements Chapter.
6 Prior to the current Inventory, vented degasification emissions from underground coal mines were typically
7 estimated based on drainage efficiencies reported by either the mining company or MSHA. However, beginning in
8 2011, underground coal mines began reporting CH4 emissions from degasification systems to EPA under its
9 GHGRP, which requires degasification quantities to be measured weekly, thus offering a more accurate account than
10 previous methods. As a result, data reported to EPA's GHGRP in 2012 and 2013 were used to estimate vented
11 degasification volumes for those mines. GHGRP data was also used in 2013 for degas-used volumes at mines using
12 methane on-site or without available gas sales records. In addition, for forty-nine mines, the 2013 VAM emission
13 estimates included VAM data measured at least quarterly and reported to the GHGRP. Emissions avoided at mines
14 with VAM mitigation projects (2) were estimated based on emission reductions registered at the Climate Action
15 Reserve GHG Registry (CAR 2014).
16 Planned Improvements
17 Future improvements to the Coal Mining category will include continued analysis and integration into the national
18 inventory of the degasification quantities and ventilation emissions data reported by underground coal mines to
19 EPA's GHGRP. A higher reliance on the GHGRP will provide greater consistency and accuracy in future
20 inventories. MSHA data will serve as a quality assurance tool for validating GHGRP data. Reconciliation of the
21 GHGRP and Inventory data sets are still in progress. In implementing improvements and integrating data from
22 EPA's GHGRP, the latest guidance from the IPCC on the use of facility-level data in national inventories will be
23 relied upon (IPCC 2011).
24
25
3.5 Abandoned Underground Coal Mines (IPCC
Source Category IBla) (TO BE UPDATED)
26 Underground coal mines contribute the largest share of coal mine methane (CMM) emissions, with active
27 underground mines the leading source of underground emissions. However, mines also continue to release CH4
28 after closure. As mines mature and coal seams are mined through, mines are closed and abandoned. Many are
29 sealed and some flood through intrusion of groundwater or surface water into the void. Shafts or portals are
30 generally filled with gravel and capped with a concrete seal, while vent pipes and boreholes are plugged in a manner
31 similar to oil and gas wells. Some abandoned mines are vented to the atmosphere to prevent the buildup of CH4 that
32 may find its way to surface structures through overburden fractures. As work stops within the mines, CH4 liberation
33 decreases but it does not stop completely. Following an initial decline, abandoned mines can liberate CH4 at a near-
34 steady rate over an extended period of time, or, if flooded, produce gas for only a few years. The gas can migrate to
35 the surface through the conduits described above, particularly if they have not been sealed adequately. In addition,
36 diffuse emissions can occur when CH4 migrates to the surface through cracks and fissures in the strata overlying the
37 coal mine. The following factors influence abandoned mine emissions:
38 • Time since abandonment;
39 • Gas content and adsorption characteristics of coal;
40 • CH4 flow capacity of the mine;
41 • Mine flooding;
42 • Presence of vent holes; and
43 • Mine seals.
44
3-54 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Annual gross abandoned mine CH4 emissions ranged from 7.2 to 10.8 MMT CO2 Eq. from 1990 through 2013,
2 varying, in general, by less than 1 percent to approximately 19 percent from year to year. Fluctuations were due
3 mainly to the number of mines closed during a given year as well as the magnitude of the emissions from those
4 mines when active. Gross abandoned mine emissions peaked in 1996 (10.8 MMT CCh Eq.) due to the large number
5 of mine closures from 1994 to 1996 (72 gassy mines closed during the three-year period). In spite of this rapid rise,
6 abandoned mine emissions have been generally on the decline since 1996. Since 2002, there have been fewer than
7 twelve gassy mine closures each year. There were eight gassy mine closures in 2013. In 2013, gross abandoned
8 mine emissions decreased slightly to 8.8 MMT CCh Eq. (see Table 3-32 and Table 3-33). Gross emissions are
9 reduced by CH4 recovered and used at 37 mines, resulting in net emissions in 2013 of 6.2 MMT €62 Eq.
10 Table 3-32: CH4 Emissions from Abandoned Coal Mines (MMT COz Eq.)
Activity
Abandoned Underground Mines
Recovered & Used
Total
1990
7.2
+
7.2
2005
8.4
1.8 1
6.6
2009
9.9
3.6
6.4
2010
9.7
3.2
6.6
2011
9.3
2.9
6.4
2012
8.9
2.7
6.2
2013
8.8
2.6
6.2
11 + Does not exceed 0.05 MMT CO2 Eq.
12 Note: Totals may not sum due to independent rounding.
13
14 Table 3-33: CH4 Emissions from Abandoned Coal Mines (kt)
Activity 1990 2005 2009 2010 2011 2012 2013
Abandoned Underground Mines 288 334 398 389 373 358 353
Recovered & Used + 70 143 126 116 109 104
Total 288 264 254 263 257 249 249
15 Note: Totals may not sum due to independent rounding.
16 + Does not exceed 0.5 kt.
17 Methodology
18 Estimating CH4 emissions from an abandoned coal mine requires predicting the emissions of a mine from the time
19 of abandonment through the inventory year of interest. The flow of CH4 from the coal to the mine void is primarily
20 dependent on the mine's emissions when active and the extent to which the mine is flooded or sealed. The CH4
21 emission rate before abandonment reflects the gas content of the coal, rate of coal mining, and the flow capacity of
22 the mine in much the same way as the initial rate of a water-free conventional gas well reflects the gas content of the
23 producing formation and the flow capacity of the well. A well or a mine which produces gas from a coal seam and
24 the surrounding strata will produce less gas through time as the reservoir of gas is depleted. Depletion of a reservoir
25 will follow a predictable pattern depending on the interplay of a variety of natural physical conditions imposed on
26 the reservoir. The depletion of a reservoir is commonly modeled by mathematical equations and mapped as a type
27 curve. Type curves which are referred to as decline curves have been developed for abandoned coal mines. Existing
28 data on abandoned mine emissions through time, although sparse, appear to fit the hyperbolic type of decline curve
29 used in forecasting production from natural gas wells.
30 In order to estimate CH4 emissions over time for a given abandoned mine, it is necessary to apply a decline function,
31 initiated upon abandonment, to that mine. In the analysis, mines were grouped by coal basin with the assumption
32 that they will generally have the same initial pressures, permeability and isotherm. As CH4 leaves the system, the
33 reservoir pressure (Pr) declines as described by the isotherm's characteristics. The emission rate declines because
34 the mine pressure (Pw) is essentially constant at atmospheric pressure for a vented mine, and the productivity index
35 (PI), which is expressed as the flow rate per unit of pressure change, is essentially constant at the pressures of
36 interest (atmospheric to 30 psia). The CH4 flow rate is determined by the laws of gas flow through porous media,
37 such as Darcy's Law. A rate-time equation can be generated that can be used to predict future emissions. This
38 decline through time is hyperbolic in nature and can be empirically expressed as:
39 q = qt (1 + WV)(~1/&)
40 where,
41 q Gas flow rate at time t in million cubic feet per day (mmcfd)
Energy 3-55
-------
1 q; = Initial gas flow rate at time zero (to), mmcfd
2 b The hyperbolic exponent, dimensionless
3 D; = Initial decline rate, 1/yr
4 t = Elapsed time from to (years)
5 This equation is applied to mines of various initial emission rates that have similar initial pressures, permeability and
6 adsorption isotherms (EPA 2004).
7 The decline curves created to model the gas emission rate of coal mines must account for factors that decrease the
8 rate of emission after mining activities cease, such as sealing and flooding. Based on field measurement data, it was
9 assumed that most U.S. mines prone to flooding will become completely flooded within eight years and therefore no
10 longer have any measurable CH4 emissions. Based on this assumption, an average decline rate for flooded mines
11 was established by fitting a decline curve to emissions from field measurements. An exponential equation was
12 developed from emissions data measured at eight abandoned mines known to be filling with water located in two of
13 the five basins. Using a least squares, curve-fitting algorithm, emissions data were matched to the exponential
14 equation shown below. There was not enough data to establish basin-specific equations as was done with the
15 vented, non-flooding mines (EPA 2004).
16 q = q^'0^
17 where,
18 q = Gas flow rate at time t in mmcfd
19 qi = Initial gas flow rate at time zero (to), mmcfd
20 D = Decline rate, 1/yr
21 t Elapsed time from to (years)
22
23 Seals have an inhibiting effect on the rate of flow of CH4 into the atmosphere compared to the flow rate that would
24 exist if the mine had an open vent. The total volume emitted will be the same, but emissions will occur over a
25 longer period of time. The methodology, therefore, treats the emissions prediction from a sealed mine similarly to
26 the emissions prediction from a vented mine, but uses a lower initial rate depending on the degree of sealing. A
27 computational fluid dynamics simulator was used with the conceptual abandoned mine model to predict the decline
28 curve for inhibited flow. The percent sealed is defined as 100 x (1 - (initial emissions from sealed mine / emission
29 rate at abandonment prior to sealing)). Significant differences are seen between 50 percent, 80 percent and 95
30 percent closure. These decline curves were therefore used as the high, middle, and low values for emissions from
31 sealed mines (EPA 2004).
32 For active coal mines, those mines producing over 100 thousand cubic feet per day (mcfd) account for 98 percent of
33 all CH4 emissions. This same relationship is assumed for abandoned mines. It was determined that the 492
34 abandoned mines closed after 1972 produced emissions greater than 100 mcfd when active. Further, the status of
35 283 of the 492 mines (or 58 percent) is known to be either: 1) vented to the atmosphere; 2) sealed to some degree
36 (either earthen or concrete seals); or, 3) flooded (enough to inhibit CH4 flow to the atmosphere). The remaining 42
37 percent of the mines whose status is unknown were placed in one of these three categories by applying a probability
38 distribution analysis based on the known status of other mines located in the same coal basin (EPA 2004).
39
3-56 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 3-34: Number of Gassy Abandoned Mines Present in U.S. Basins, grouped by Class
2 according to Post-Abandonment State
Basin
Central Appl.
Illinois
Northern Appl.
Warrior Basin
Western Basins
Total
Sealed Vented Flooded Total Known Unknown Total Mines
33
32
42
0
27
134
25
3
22
0
3
53
48
14
16
16
2
96
106
49
80
16
32
283
137
27
36
0
9
209
243
76
116
16
41
492
4 Inputs to the decline equation require the average emission rate and the date of abandonment. Generally this data is
5 available for mines abandoned after 1971; however, such data are largely unknown for mines closed before 1972.
6 Information that is readily available, such as coal production by state and county, is helpful but does not provide
7 enough data to directly employ the methodology used to calculate emissions from mines abandoned before 1972. It
8 is assumed that pre-1972 mines are governed by the same physical, geologic, and hydrologic constraints that apply
9 to post-1971 mines; thus, their emissions may be characterized by the same decline curves.
10 During the 1970s, 78 percent of CH4 emissions from coal mining came from seventeen counties in seven states. In
11 addition, mine closure dates were obtained for two states, Colorado and Illinois, for the hundred year period
12 extending from 1900 through 1999. The data were used to establish a frequency of mine closure histogram (by
13 decade) and applied to the other five states with gassy mine closures. As a result, basin-specific decline curve
14 equations were applied to the 145 gassy coal mines estimated to have closed between 1920 and 1971 in the United
15 States, representing 78 percent of the emissions. State-specific, initial emission rates were used based on average
16 coal mine CH4 emissions rates during the 1970s (EPA 2004).
17 Abandoned mine emission estimates are based on all closed mines known to have active mine CH4 ventilation
18 emission rates greater than 100 mcfd at the time of abandonment. For example, for 1990 the analysis included 145
19 mines closed before 1972 and 258 mines closed between 1972 and 1990. Initial emission rates based on MSHA
20 reports, time of abandonment, and basin-specific decline curves influenced by a number of factors were used to
21 calculate annual emissions for each mine in the database. Coal mine degasification data are not available for years
22 prior to 1990, thus the initial emission rates used reflect ventilation emissions only for pre-1990 closures. CH4
23 degasification amounts were added to the quantity of CH4 vented to determine the total CH4 liberation rate for all
24 mines that closed between 1992 and 2013. Since the sample of gassy mines (with active mine emissions greater
25 than 100 mcfd) is assumed to account for 78 percent of the pre-1972 and 98 percent of the post-1971 abandoned
26 mine emissions, the modeled results were multiplied by 1.22 and 1.02 to account for all U.S. abandoned mine
27 emissions.
28 From 1993 through 2013, emission totals were downwardly adjusted to reflect abandoned mine CH4 emissions
29 avoided from those mines. The Inventory totals were not adjusted for abandoned mine reductions from 1990
30 through 1992 because no data was reported for abandoned coal mining CH4 recovery projects during that time.
31 Uncertainty and Time-Series Consistency
32 A quantitative uncertainty analysis was conducted to estimate the uncertainty surrounding the estimates of emissions
33 from abandoned underground coal mines. The uncertainty analysis described below provides for the specification of
34 probability density functions for key variables within a computational structure that mirrors the calculation of the
35 inventory estimate. The results provide the range within which, with 95 percent certainty, emissions from this
36 source category are likely to fall.
37 As discussed above, the parameters for which values must be estimated for each mine in order to predict its decline
38 curve are: 1) the coal's adsorption isotherm; 2) CH4 flow capacity as expressed by permeability; and 3) pressure at
39 abandonment. Because these parameters are not available for each mine, a methodological approach to estimating
40 emissions was used that generates a probability distribution of potential outcomes based on the most likely value and
41 the probable range of values for each parameter. The range of values is not meant to capture the extreme values, but
42 rather values that represent the highest and lowest quartile of the cumulative probability density function of each
43 parameter. Once the low, mid, and high values are selected, they are applied to a probability density function.
Energy 3-57
-------
1 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 3-35. Annual abandoned
2 coal mine CH4 emissions in 2013 were estimated to be between 5.0 and 7.7 MMT CCh Eq. at a 95 percent
3 confidence level. This indicates a range of 20 percent below to 24 percent above the 2013 emission estimate of 6.2
4 MMT CO2 Eq. One of the reasons for the relatively narrow range is that mine-specific data is available for use in
5 the methodology for mines closed after 1972. Emissions from mines closed prior to 1972 have the largest degree of
6 uncertainty because no mine-specific CH4 liberation rates exist.
7 Table 3-35: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from
8 Abandoned Underground Coal Mines (MMT COz Eq. and Percent)
^ „ 2013 Emission Estimate Uncertainty Range Relative to Emission Estimate3
6 (MMT CCh Eq.) (MMT CCh Eq.) (%)
Abandoned Underground ~ +24%
Coal Mines
9 a Range of emission estimates predicted by Monte Carlo Simulation for a 95 percent confidence interval.
10
11 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
12 through 2013. Details on the emission trends through time are described in more detail in the Methodology section,
13 above.
14
15
| Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
3.6 Petroleum Systems (IPCC Source Category
lB2a)
16 Substantial new data are available on natural gas and petroleum systems from the EPA's Greenhouse Gas Reporting
17 Program (GHGPJ3). The EPA is evaluating approaches for incorporating this new data into its emission estimates for
18 the Inventory of U.S. GHG Emissions and Sinks: 1990-2014.
19 The details of the revisions under consideration for the current Inventory, and key questions for stakeholder
20 feedback are available in segment-level memoranda at
21 http://www3.epa.gov/climatechange/ghgemissions/usinventorvreport/natural-gas-svstems.html. For petroleum
22 systems, revisions under consideration impact the production segment only.
23 Below, updated draft estimates of CH4 emissions for the year 2013 are presented to provide reviewers of the public
24 review draft an indication of the sector-wide emissions estimates resulting from changes under consideration.
25 Carbon dioxide emissions have not been updated in the public review draft, but will be updated in the final
26 Inventory.
27 The EPA is continuing to evaluate stakeholder feedback on the updates under consideration. For the final Inventory,
28 the 2013 estimates presented in this section will be refined, and a full time series of emissions estimates will be
29 developed based on feedback received through the earlier stakeholder reviews of the memos and through this public
30 review period.
31 Methane emissions from petroleum systems are primarily associated with onshore and offshore crude oil production,
32 transportation, and refining operations. During each of these activities, CH4 is released to the atmosphere as fugitive
33 emissions, vented emissions, emissions from operational upsets, and emissions from fuel combustion. Fugitive and
34 vented CCh emissions from petroleum systems are primarily associated with crude oil production and refining
35 operations but are negligible in transportation operations. Total CH4 emissions from petroleum systems in 2013
36 were 63.4 MMT CO2 Eq. (2,535 kt).
3-58 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Production Field Operations. The EPA is considering several revisions to the production segment. For information
2 on these revisions, and requests for stakeholder feedback please see the Production Segment memo.65 The
3 information presented here was calculated using the approaches under consideration in the memo, as discussed
4 below in the Recalculations Discussion section.
5 Production field operations account for over 98 percent of total CH4 emissions from petroleum systems. Vented CH4
6 from field operations account for approximately 92 percent of the net emissions from the production sector, fugitive
7 emissions are 5 percent, uncombusted CH4 emissions (i.e., unburned fuel) account for 4 percent, and process upset
8 emissions are approximately 0.1 percent. The most dominant sources of emissions, in order of magnitude, are
9 intermittent bleed pneumatic controllers, oil tanks, chemical injection pumps, high bleed pneumatic controllers,
10 shallow water offshore oil platforms, gas engines, oil wellheads (light crude services), and low bleed pneumatic
11 devices. These eight sources alone emit over 96 percent of the production field operations emissions. Offshore
12 platform emissions are a combination of fugitive, vented, and uncombusted fuel emissions from all equipment
13 housed on oil platforms producing oil and associated gas. Emissions from gas engines are due to unburned CH4 that
14 vents with the exhaust. Emissions from oil wellheads are due to fugitive losses. The remaining 4 percent of the
15 emissions are distributed among 26 additional activities within the four categories: vented, fugitive, combustion, and
16 process upset emissions.
17 Vented CCh associated with production field operations account for approximately 99 percent of the total
18 emissions from production field operations, while fugitive and process upsets together account for less than 1
19 percent of the emissions. The most dominant sources of vented CC>2 emissions are oil tanks, high bleed pneumatic
20 controllers, shallow water offshore oil platforms, low bleed pneumatic controllers, and oil wellheads (light crude
21 services). These five sources together account for slightly over 98 percent of the non-combustion CCh emissions
22 from production field operations, while the remaining 1 percent of the emissions is distributed among 24 additional
23 activities within the three categories: vented, fugitive, and process upsets. Note that CCh from associated gas flaring
24 is accounted in natural gas systems production emissions. CCh emissions from flaring for both natural gas and oil
25 were 16 MMT CO2 Eq. in 2013.
26 Crude Oil Transportation. Crude oil transportation activities account for approximately 0.7 percent of total CH4
27 emissions from the oil industry. Venting from tanks, truck loading, rail loading, and marine vessel loading
28 operations account for 82 percent of CH4 emissions from crude oil transportation. Fugitive emissions, almost
29 entirely from floating roof tanks, account for 14 percent of CH4 emissions from crude oil transportation. The
30 remaining 4 percent is distributed among two additional sources within the vented emissions category (i.e., pump
31 station maintenance and pipeline pigging). Emissions from pump engine drivers and heaters were not estimated due
32 to lack of data.
33 Since 1990, CH4 emissions from transportation have increased by almost 4 percent. However, because emissions
34 from crude oil transportation account for such a small percentage of the total emissions from the petroleum industry,
35 this has had little impact on the overall emissions. Methane emissions have increased by approximately 11 percent
36 from 2012 levels.
37 Crude Oil Refining. Crude oil refining processes and systems account for slightly above 3 percent of total CH4
38 emissions from the oil industry because most of the CH4 in crude oil is removed or escapes before the crude oil is
39 delivered to the refineries. There is an insignificant amount of CH4 in all refined products. Within refineries,
40 combustion emissions account for about 60 percent of the CH4 emissions, while vented and fugitive emissions
41 account for approximately 26 and 13 percent, respectively. Flare emissions are the primary combustion emissions
42 contributor. Refinery system blowdowns for maintenance and process vents are the primary venting contributors.
43 Most of the fugitive CH4 emissions from refineries are from equipment leaks and storage tanks.
44 CH4 emissions from refining of crude oil have increased by approximately 24 percent since 1990; however, similar
45 to the transportation subcategory, this increase has had little effect on the overall emissions of CH4. Since 1990, CH4
46 emissions have fluctuated between 27 and 34 kt.
65
http://www3.epa.gov/climatechange/ghgemissions/usinventoryreport/DRAFT_Proposed_Revision_to_Production_Segment_Emi
ssions_2-3-2016.pdf
Energy 3-59
-------
1 Flare emissions from crude oil refining accounts for 95 percent of the total CO2 emissions in petroleum systems.
2 Refinery CO2 emissions increased by 36 percent from 1990 to 2013.
3
4 Table 3-36: CH4 Emissions from Petroleum Systems (MMT COz Eq.)
Activity 1990 2005 2009 2010 2011 2012 2013 2014
Production Field Operations
(Potential) 63.1
Pneumatic controller venting* 37.8
Tank venting 7.9
Combustion & process upsets 2.8
Misc. venting & fugitives 13.1
Wellhead fugitives 1.5
Production Voluntary Reductions (0.8)
Production Field Operations
(Net) 62.4
Crude Oil Transportation 0.2
Refining 0.8
Total 63.4
Note: Totals may not sum due to independent rounding.
a Values presented in this table for pneumatic controllers are net emissions. The revised methodology for the
2016 Inventory incorporates GHGRP subpart W activity and net emissions data, and is detailed in the
Recalculations Discussion section.
5
6 Table 3-37: CH4 Emissions from Petroleum Systems (kt)
Activity 1990 2005 2009 2010 2011 2012 2013 2014
Production Field Operations
(Potential) 2,525
Pneumatic controller venting* 1,512
Tank venting 318
Combustion & process upsets 113
Misc. venting & fugitives 553
Wellhead fugitives 60
Production Voluntary Reductions (31)
Production Field Operations
(Net) 2,494
Crude Oil Transportation 7
Refining 34
Total 2,535
Note: Totals may not sum due to independent rounding.
a Values presented in this table for pneumatic controllers are net emissions. The revised methodology for the
2016 Inventory incorporates GHGRP subpart W activity and net emissions data, and is detailed in the
Recalculations Discussion section.
Table 3-38: COz Emissions from Petroleum Systems (MMT COz Eq.)
Activity
Production Field Operations
Pneumatic controller venting
Tank venting
Misc. venting & fugitives
Wellhead fugitives
Process upsets
Crude Refining
Total
1990
0.4
+
0.3
+
+
+
4.1
4.4
2005
0.3
+
0.2
+
+
+
4.6
4.9
2009
0.3
+
0.3
+
+
+
4.4
4.7
2010
0.3
+
0.3
+
+
+
3.8
4.2
2011
0.3
+
0.3
+
+
+
4.1
4.5
2012
0.4
+
0.4
+
+
+
4.7
5.1
2013 2014
0.5
+
0.4
+
+
+
5.5
6.0
3-60 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
+ Does not exceed 0.05 MMT CO2 Eq.
Note: Totals may not sum due to independent rounding.
1 Table 3-39: COz Emissions from Petroleum Systems (kt)
Activity
Production Field Operations
Pneumatic controller venting
Tank venting
Misc. venting & fugitives
Wellhead fugitives
Process upsets
Crude Refining
Total
1990
375
27
328
16
3
0.2
4,070
4,445
2005
285
23
246
13
3
0.1
4,620
4,904
2009
305
24
265
14
3
0.1
4,351
4,656
2010
317
24
276
14
3
0.2
3,836
4,153
2011
333
25
291
14
3
0.2
4,134
4,467
2012
394
26
351
14
3
0.2
4,666
5,060
2013
461
26
417
14
3
0.2
5,540
6,001
2014
Note: Totals may not sum due to independent rounding.
2 Methodology
3 EPA has made revisions to the methodology and data sources for many sources in the Inventory. Please see the
4 Recalculations Discussion section below.
5 The methodology for estimating CH4 emissions from petroleum systems is based on comprehensive studies of CH4
6 emissions from U.S. petroleum systems (GPJ/EPA 1996, EPA 1999) and EPA's GHGRP data. The 1996 and 1999
7 studies calculated emission estimates for 57 activities occurring in petroleum systems from the oil wellhead through
8 crude oil refining, including 33 activities for crude oil production field operations, 11 for crude oil transportation
9 activities, and 13 for refining operations. Annex 3.5 provides greater detail on the emission estimates for these 57
10 activities. The estimates of CH4 emissions from petroleum systems do not include emissions downstream of oil
11 refineries because these emissions are negligible.
12 Key references for activity data and emission factors are Drillinglnfo (2014), the Energy Information Administration
13 annual and monthly reports (EIA 1990 through 2014), (EIA 1995 through 2014a, 2014b, 2014c), "Methane
14 Emissions from the Natural Gas Industry by the Gas Research Institute and EPA" (EPA/GRI 1996a-d), "Estimates
15 of Methane Emissions from the U.S. Oil Industry" (EPA 1999), consensus of industry peer review panels,
16 BOEMRE and BOEM reports (BOEMRE 2004, BOEM 2011), analysis of BOEMRE data (EPA 2005, BOEMRE
17 2004), the Oil & Gas Journal (OGJ 2014a, 2013b), the Interstate Oil and Gas Compact Commission (IOGCC 2011),
18 the United States Army Corps of Engineers, (1995-2012), and the GHGRP (2010-2013).
19 The methodology for estimating CH4 emissions from the 46 oil industry activities (excluding refining activities)
20 employs emission factors initially developed by EPA (1999). Activity data for the years 1990 through 2013 were
21 collected from a wide variety of statistical resources. Emissions are estimated for each activity by multiplying
22 emission factors (e.g., emission rate per equipment item or per activity) by the corresponding activity data (e.g.,
23 equipment count or frequency of activity). EPA (1999) provides emission factors for all activities except those
24 related to offshore oil production and field storage tanks. For offshore oil production, two emission factors were
25 calculated using data collected for all federal offshore platforms (EPA 2015, BOEM 2014). One emission factor is
26 for oil platforms in shallow water, and one emission factor is for oil platforms in deep water. Emission factors are
27 held constant for the period 1990 through 2013. The number of platforms in shallow water and the number of
28 platforms in deep water are used as activity data and are taken from Bureau of Ocean Energy Management (BOEM)
29 (formerly Bureau of Ocean Energy Management, Regulation, and Enforcement [BOEMRE]) datasets (BOEM
30 201 la,b,c). For oil storage tanks, the emissions factor was calculated as the total emissions per barrel of crude
31 charge from E&P Tank data weighted by the distribution of produced crude oil gravities from the HPDI production
32 database (EPA 1999, HPDI 2011).
33 For some years, complete activity data were not available. In such cases, one of three approaches was employed.
34 Where appropriate, the activity data were calculated from related statistics using ratios developed for EPA (1996).
35 For example, EPA (1996) found that the number of heater treaters (a source of CH4 emissions) is related to both
36 number of producing wells and annual production. To estimate the activity data for heater treaters, reported statistics
Energy 3-61
-------
1 for wells and production were used, along with the ratios developed for EPA (1996). In other cases, the activity data
2 were held constant from 1990 through 2013 based on EPA (1999). Lastly, the previous year's data were used when
3 data for the current year were unavailable. The CH4 and CC>2 sources in the production sector share common activity
4 data. See Annex 3.5 for additional detail.
5 For petroleum refining activities, 2010 to 2013 emissions were directly obtained from EPA's GHGRP. All refineries
6 are required to report their CH4 and CCh emissions for all major activities since 2010. The national totals of these
7 emissions for each activity were used for the 2010 to 2013 emissions. The national emission totals for each activity
8 were also divided by refinery feed rates for those four Inventory years to develop average activity-specific emission
9 factors. These emission factors were used to estimate national emissions for each refinery activity from 1990 to
10 2009 based on national refinery feed rates for the respective Inventory year. (EPA 2015c).
11 The Inventory estimate for Petroleum Systems takes into account Natural Gas STAR reductions. Voluntary
12 reductions included in the Petroleum Sector calculations were those reported to Natural Gas STAR for the following
13 activities: artificial lift: gas lift; artificial lift: use compression; artificial lift: use pumping unit; consolidate crude oil
14 production and water storage tanks; lower heater-treater temperature; re-inject gas for enhanced oil recovery; re-
15 inject gas into crude; and route casinghead gas to vapor recovery unit or compressor. In addition, a portion of the
16 total Gas STAR reductions from pneumatic controllers in the production sector are applied to potential emissions in
17 the petroleum sector.
18 The methodology for estimating CC>2 emissions from petroleum systems combines vented, fugitive, and process
19 upset emissions sources from 29 activities for crude oil production field operations and three activities from
20 petroleum refining. For the production field operations, emissions are estimated for each activity by multiplying
21 emission factors by their corresponding activity data. The emission factors for €62 are generally estimated by
22 multiplying the CH4 emission factors by a conversion factor, which is the ratio of €62 content and CH4 content in
23 produced associated gas. One exception to this methodology are the emission factors for crude oil storage tanks,
24 which are obtained from E&P Tank simulation runs, and the emission factor for asphalt blowing, which was derived
25 using the methodology and sample data from API (2009). Other exceptions to this methodology are the three
26 petroleum refining activities (i.e., flares, asphalt blowing, and process vents); the CO2 emissions data for 2010 to
27 2013 were directly obtained from the GHGRP. The 2010 to 2013 CO2 emissions GHGRP data along with the
28 refinery feed data for 2010 to 2013 were used to derive CO2 emission factors (i.e., sum of activity emissions/sum of
29 refinery feed). The emission factors were then applied to the annual refinery feed to estimate CO2 emissions for
30 1990 to 2009.
31
Uncertainty and Time-Series Consistency
32 The most recent uncertainty analysis for the natural gas and petroleum systems emissions estimates in the Inventory
33 was conducted for the 1990-2009 Inventory that was released in 2011. Since the analysis was last conducted, several
34 of the methods used in the Inventory have changed, and industry practices and equipment have evolved. In addition,
35 new studies and other data sources such as those discussed in this memorandum offer improvement to understanding
36 and quantifying the uncertainty of some emission source estimates.
37 As updates to the Inventory data and methods are selected, the EPA will review information on uncertainty and
38 consider how the Inventory uncertainty assessment can be updated to reflect the new information.
39 A quantitative uncertainty analysis was conducted in 2010 to determine the level of uncertainty surrounding
40 estimates of emissions from petroleum systems using the IPCC-recommended Approach 2 methodology (Monte
41 Carlo Simulation technique). The @RISK software model was used to quantify the uncertainty associated with the
42 emission estimates using the 7 highest-emitting sources ("top 7 sources") for the year 1995. The @RISK analysis
43 provides for the specification of probability density functions for key variables within a computational structure that
44 mirrors the calculation of the Inventory estimate. The IPCC guidance notes that in using this method, "some
45 uncertainties that are not addressed by statistical means may exist, including those arising from omissions or double
46 counting, or other conceptual errors, or from incomplete understanding of the processes that may lead to
47 inaccuracies in estimates developed from models." As a result, the understanding of the uncertainty of emission
48 estimates for this category evolves and improves as the underlying methodologies and datasets improve.
49 The uncertainty analysis conducted in 2010 has not yet been updated for the 1990 through 2013 Inventory years;
50 instead, the uncertainty percentage ranges calculated previously were applied to 2013 emission estimates. The
3-62 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 majority of sources in the current Inventory were calculated using the same emission factors and activity data for
2 which PDFs were developed in the 1990 through 2009 uncertainty analysis. As explained in the Methodology
3 section above and the Recalculations Discussion below, several emission sources have undergone recent
4 methodology revisions, and the 2009 uncertainty ranges will not reflect the uncertainty associated with the recently
5 revised emission factors and activity data sources. Please see discussion on Planned Improvements.
6 The results presented below provide with 95 percent certainty the range within which emissions from this source
7 category are likely to fall for the year 2013, based on the previously conducted uncertainty assessment using the
8 recommended IPCC methodology. The heterogeneous nature of the petroleum industry makes it difficult to sample
9 facilities that are completely representative of the entire industry. Additionally, highly variable emission rates were
10 measured among many system components, making the calculated average emission rates uncertain. The results of
11 the Approach 2 quantitative uncertainty analysis are summarized in Table 3-40. Petroleum systems CH4 emissions
12 in 2013 were estimated to be between 19.2 and 62.8 MMT CC>2 Eq., while CCh emissions were estimated to be
13 between 4.6 and 14.9 MMT CC>2 Eq. at a 95 percent confidence level. This indicates a range of 24 percent below to
14 149 percent above the 2013 emission estimates of 25.2 and 6.0 MMT €62 Eq. for CH4 and CCh, respectively.
15 Table 3-40: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from
16 Petroleum Systems (MMT COz Eq. and Percent)
2013 Emission Estimate Uncertainty Range Relative to Emission Estimate3
aS (MMTCChEq.)" (MMTCChEq.)
20
29
Petroleum Systems
Petroleum Systems
CH4
CO2
25.2
6.0
Lower
Bound"
19.2
4.6
Upper
Bound"
62.8
14.9
Lower
Bound"
-24%
-24%
Upper
Bound"
149%
149%
a Range of 2013 relative uncertainty predicted by Monte Carlo Stochastic Simulation, based on 1995 base
year activity factors, for a 95 percent confidence interval.
b All reported values are rounded after calculation. As a result, lower and upper bounds may not be
duplicable from other rounded values as shown in table.
Note: Totals may not sum due to independent rounding
17 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
18 through 2013. Details on the emission trends through time are described in more detail in the Methodology section,
19 above.
QA/QC and Verification Discussion
21 The petroleum system emission estimates in the Inventory are continually being reviewed and assessed to determine
22 whether emission factors and activity factors accurately reflect current industry practices. A QA/QC analysis was
23 performed for data gathering and input, documentation, and calculation. QA/QC checks are consistently conducted
24 to minimize human error in the model calculations. EPA performs a thorough review of information associated with
25 new studies, GHGRP data, regulations, public webcasts, and the Natural Gas STAR Program to assess whether the
26 assumptions in the Inventory are consistent with current industry practices. In addition, EPA receives feedback
27 through annual expert and public review period. Feedback received is noted in the Recalculations and Planned
28 Improvement sections.
Recalculations Discussion
30 The EPA received information and data related to the emission estimates through the Inventory preparation process,
31 previous Inventories' formal public notice periods, GHGRP data, and new studies. The EPA carefully evaluated
32 relevant information available, and made revisions to the production segment methodology in the public review
33 draft—including revised equipment activity data as well as pneumatic controller activity and emissions data.
34 In February 2016, the EPA released a draft memo that discussed the changes under consideration and requested
3 5 stakeholder feedback on those changes.
Energy 3-63
-------
1 • Revisions under Consideration for Natural Gas and Petroleum Production Emissions66
2 The EPA is continuing to evaluate stakeholder feedback on the updates under consideration. For the final Inventory,
3 the 2013 estimates presented in this section will be refined, and a full time series of emissions estimates will be
4 developed based on feedback received through the earlier stakeholder reviews of information released in the
5 segment-specific memoranda, and through this public review period.
6 The combined impact of revisions to 2013 natural gas production segment emissions described below, compared to
7 the 1990-2013 Inventory, is an increase in petroleum system CH4 emissions of approximately 38 MMT CO2 Eq., or
8 151 percent.
9 Trend information has not yet been calculated, but it is expected that across the 1990-2013 time series, compared to
10 the previous (2015) Inventory, in the current (2016) Inventory, the total CH4 emissions estimate will increase, with
11 the largest increases in the estimate occurring in later years of the time series.
12 Production
13 This section references the Production Segment memo, available at
14 http://www3.epa.gov/climatechange/ghgemissions/usinventoryreport/DRAFT Proposed Revision to Production S
15 egment Emissions 2-3-2016.pdf.
16 Using newly available GHGRP activity data, the EPA has developed activity factors (i.e., counts per gas well) for
17 separators, headers, heater-treaters, pneumatic pumps, and pneumatic controllers. The EPA has applied these
18 updated activity factors (see Reporting Year (RY) 2014) values in Table 7 of the Production Segment memo) to
19 calculate emissions from these sources in this public review draft. The impact of using activity factors developed
20 from GHGRP data is an increase in calculated emissions. For the year 2013, the CH4 emissions increase due to use
21 of revised activity factors for maj or equipment and pneumatic pumps is approximately 5.8 MMT CCh Eq.
22 Using the GHGRP data, the EPA has also developed technology-specific activity data and emission factors for
23 pneumatic controllers. Reported data under EPA's GHGRP allow for development of emission factors specific to
24 bleed type (continuous high bleed, continuous low bleed, and intermittent bleed) and separation of activity data into
25 these categories. For the public review draft, EPA has used this separation of pneumatic controller counts by bleed
26 types, and emission factors developed from reported GHGRP data. See RY2013 values in Table 11 (activity data
27 for proportion of controllers in each category), and RY2011-2014 values in Table 10 (emission factors) of the
28 Production Segment memo. Comparing the updated 2013 estimate to the previous Inventory 2013 estimate, the
29 impact of using bleed type-specific emission factors and activity data developed from GHGRP data is an increase of
30 approximately 32.3 MMT CO2Eq.
31 The EPA is considering approaches to develop the Inventory time series (1990-2014) that will create consistency
32 between earlier years' estimates that generally rely on data from GRI/EPA 1996, and more recent years' estimates
33 that rely on GHGRP data. For years 1992 through 2010, the EPA is considering calculating equipment counts by
34 linearly interpolating between the data points of calculated national equipment counts in 1992 (using GRI/EPA) and
35 calculated national equipment counts in 2011 (using GHGRP).
36 The EPA's approach for revising the GHG Inventory methodology to incorporate GHGRP data obviates the need to
37 apply Gas STAR reductions data for certain sources in the production segment. EPA plans to carry forward reported
38 reductions for sources that are not being revised to use a net emission factor approach. There are also Gas STAR
39 reductions in the production segment that are not classified as applicable to specific emission sources ("other
40 voluntary reductions" are 0.8 MMT CCh Eq. CH4 in year 2013). Some portion of the "other voluntary reductions"
41 might apply to the emission sources for which the EPA is revising the methodology to use a net emission factor
42 approach. The EPA is investigating potential disaggregation of "other voluntary reductions." The EPA has retained
43 Gas STAR reductions classified as "other voluntary reductions" in year 2013 emission calculations for the public
44 review draft.
66 Available at: <
http://www3.epa.gov/climatechange/ghgemissions/usinventoryreport/DRAFT_Proposed_Revision_to_Production_Segment_Emi
ssions_2-3-2016.pdf>
3-64 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Planned Improvements
2 EPA will continue to refine the emission estimates to reflect the most robust information available, including
3 stakeholder feedback received on the Inventory improvement memos, and this public review draft.
Box 3-7: Carbon Dioxide Transport, Injection, and Geological Storage (TO BE UPDATED)
6 Carbon dioxide is produced, captured, transported, and used for Enhanced Oil Recovery (EOR) as well as
7 commercial and non-EOR industrial applications. This CC>2 is produced from both naturally-occurring CCh
8 reservoirs and from industrial sources such as natural gas processing plants and ammonia plants. In the Inventory,
9 emissions from naturally-produced CCh are estimated based on the application.
10 In the Inventory, CC>2 that is used in non-EOR industrial and commercial applications (e.g., food processing,
11 chemical production) is assumed to be emitted to the atmosphere during its industrial use. These emissions are
12 discussed in the Carbon Dioxide Consumption section. The naturally-occurring CO2 used in EOR operations is
13 assumed to be fully sequestered. Additionally, all anthropogenic CO2 emitted from natural gas processing and
14 ammonia plants is assumed to be emitted to the atmosphere, regardless of whether the CO2 is captured or not. These
15 emissions are currently included in the Natural Gas Systems and the Ammonia Production sections of the Inventory
16 report, respectively.
17 IPCC includes methodological guidance to estimate emissions from the capture, transport, injection, and geological
18 storage of CO2. The methodology is based on the principle that the carbon capture and storage system should be
19 handled in a complete and consistent manner across the entire Energy sector. The approach accounts for CO2
20 captured at natural and industrial sites as well as emissions from capture, transport, and use. For storage specifically,
21 a Tier 3 methodology is outlined for estimating and reporting emissions based on site-specific evaluations. However,
22 IPCC (IPCC 2006) notes that if a national regulatory process exists, emissions information available through that
23 process may support development of CO2 emissions estimates for geologic storage.
24 In the United States, facilities that conduct geologic sequestration of CO2 and all other facilities that inject CO2
25 underground, including facilities conducting EOR, are required to report greenhouse gas data annually to EPA
26 through its GHGRP. Facilities conducting geologic sequestration of CO2 are required to develop and implement an
27 EPA-approved site-specific monitoring, reporting and verification plan, and to report the amount of CO2 sequestered
28 using a mass balance approach. Data from this program will be evaluated closely and opportunities for improving
29 the emission estimates will be considered.
30 Preliminary estimates indicate that the amount of CO2 captured from industrial and natural sites is 46.2 MMT CO2
31 Eq. (46,198 kt) (see Table 3-41 and Table 3-42). Site-specific monitoring and reporting data for CO2 injection sites
32 (i.e., EOR operations) were not readily available, therefore, these estimates assume all CO2 is emitted. Values for
33 2011 to 2013 were proxied from 2010 data.
34 Table 3-41: Potential Emissions from COz Capture and Transport (MMT COz Eq.)
Stage
Acid Gas Removal Plants
Naturally Occurring CCh
Ammonia Production Plants
Pipelines Transporting CCh
Total
1990
4.8
20.8
+
25.6
2005
15.8
28.3
0.7
34.7
2009
7.0
39.7
0.6
+
47.3
2010
11.6
34.0
0.7
+
46.2
2011
11.6
34.0
0.7
+
46.2
2012
11.6
34.0
0.7
+
46.2
2013
11.6
34.0
0.7
+
46.2
+ Does not exceed 0.05 MMT CO2 Eq.
Note: Totals may not sum due to independent rounding.
35
Table 3-42: Potential Emissions from COz Capture and Transport (kt)
Stage
Acid Gas Removal Plants
Naturally Occurring CO2
Ammonia Production Plants
1990
4,832 1
20,811
+
2005
5,798
1 28,267 1
676
2009
7,035
39,725
580
2010
11,554
33,967
677
2011
11,554
33,967
677
2012
11,554
33,967
677
2013
11,554
33,967
677
Energy 3-65
-------
Pipelines Transporting CCh 8 7
Total 25,643 34,742 47,340 46,198 46,198 46,198 46,198
+ Does not exceed 0.5 kt.
Note: Totals do not include emissions from pipelines transporting CCh. Totals may not sum due to independent
rounding.
2 3.7 Natural Gas Systems (IPCC Source Category
3 lB2b)
4 Substantial new data are available on natural gas and petroleum systems from the EPA's Greenhouse Gas Reporting
5 Program (GHGRP) and a number of new studies. The EPA is evaluating approaches for incorporating this new data
6 into its emission estimates for the Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014.
1 The details of the revisions under consideration for this year's Inventory, and key questions for stakeholder feedback
8 are available in segment-level memoranda at
9 http://www3.epa.gov/climatechange/ghgemissions/usinventoryreport/natural-gas-systems.html.
10 Below, updated draft estimates of methane emissions for the year 2013 are presented to provide reviewers of the
11 public review draft an indication of the sector-wide emissions estimates resulting from the combined changes under
12 consideration. CCh emissions have not been updated in the public review draft, but will be updated in the final
13 Inventory.
14 The EPA is continuing to evaluate stakeholder feedback on the updates under consideration. For the final Inventory,
15 the 2013 estimates presented in this section will be refined, and a full time series of emissions estimates will be
16 developed based on feedback received through the earlier stakeholder reviews of the memos and through this public
17 review period.
18 The U.S. natural gas system encompasses hundreds of thousands of wells, hundreds of processing facilities, and
19 over a million miles of transmission and distribution pipelines. Overall, natural gas systems emitted approximately
20 169 MMT CO2 Eq. (6,745 kt) of CH4 in 2013.
21 CH4 and non-combustion CC>2 emissions from natural gas systems include those resulting from normal operations,
22 routine maintenance, and system upsets. Emissions from normal operations include: natural gas engine and turbine
23 uncombusted exhaust, bleed and discharge emissions from pneumatic controllers, and fugitive emissions from
24 system components. Routine maintenance emissions originate from pipelines, equipment, and wells during repair
25 and maintenance activities. Pressure surge relief systems and accidents can lead to system upset emissions. Below is
26 a characterization of the four major stages of the natural gas system. Each of the stages is described and the different
27 factors affecting CH4 and non-combustion CC>2 emissions are discussed.
28 Field Production (including Gathering and Boosting). In these initial stages, wells are used to withdraw raw gas
29 from underground formations, and emissions arise from the wells themselves, gathering pipelines, and well-site gas
30 treatment facilities such as dehydrators and separators. The EPA is considering several revisions to the production
31 (including gathering and boosting) segment. For information on these revisions, and requests for stakeholder
32 feedback please see the Production Segment memo67 and the Gathering and Boosting memo.68 The information
67 Available online at
68 Available online at
3-66 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 presented here was calculated using the approaches under consideration in the EPA memoranda, as discussed below
2 in the Recalculations Discussion section.
3 Emissions from gathering and boosting stations, pneumatic controllers, venting for liquids unloading, kimray
4 pumps, condensate tanks, pipeline leaks, and chemical injection pumps account for the majority of CH4 emissions in
5 2013. Emissions from production account for over 60 percent of CH4 emissions from natural gas systems in 2013.
6 Trend information for the revised production segment has not yet been calculated. Based on the methods under
7 consideration for revising the time series, it is expected that the recalculated time series will show an increasing
8 trend from 1990 for CH4 emissions from production and gathering and boosting. For example, for gathering and
9 boosting stations, under the approach under consideration, estimated CH4 emissions would nearly double from
10 1990-2013. Other key emissions sources in the production segment are calculated using gas well counts as an
11 activity data driver. Total national gas well counts doubled from 1990-2013. At the same time, changes in
12 technology and practices have also occurred, which in some cases, such as switching from high bleed to low bleed
13 pneumatic controllers, reduces the growth in emissions that would be expected based on well count trends alone.
14 Flaring emissions account for the majority of the non-combustion CCh emissions. CCh emissions from production
15 increased 63 percent from 1990 to 2013 due to increases in onshore and offshore flaring.
16 Processing. In this stage, natural gas liquids and various other constituents from the raw gas are removed, resulting
17 in "pipeline quality" gas, which is injected into the transmission system. The EPA has not yet assessed potential
18 revisions for this segment. Information presented here is from the previous (2015) Inventory.
19 Fugitive CH4 emissions from compressors, including compressor seals, are the primary emission source from this
20 stage. The majority of non-combustion CCh emissions come from acid gas removal units, which are designed to
21 remove CCh from natural gas. Processing plants account for 14 percent of CH4 emissions and 58 percent of non-
22 combustion CC>2 emissions from natural gas systems. CH4 emissions from processing increased by 6 percent from
23 1990 to 2013 as emissions from compressors increased as the quantity of gas produced increased.
24 CO2 emissions from processing decreased by 22 percent from 1990 to 2013, as a decrease in the quantity of gas
25 processed resulted in a decrease in acid gas removal emissions.
26 Transmission and Storage. Natural gas transmission involves high pressure, large diameter pipelines that transport
27 gas long distances from field production and processing areas to distribution systems or large volume customers
28 such as power plants or chemical plants. Compressor station facilities, which contain large reciprocating and turbine
29 compressors, are used to move the gas throughout the U.S. transmission system.
30 The EPA is considering several revisions to the transmission and storage segment. For information on these
31 revisions, and requests for stakeholder feedback please see the Transmission and Storage Segment memo.69 The
32 information presented here was calculated using the approaches under consideration in the EPA memorandum, as
3 3 discussed below in the Recalculations Discussion section.
34 Transmission station venting and fugitives, pipeline venting, uncombusted engine exhaust, and storage station
35 venting and fugitives account for the majority of the emissions from this stage. Natural gas is also injected and
36 stored in underground formations, or liquefied and stored in above ground tanks, during periods of low demand
37 (e.g., summer), and withdrawn, processed, and distributed during periods of high demand (e.g., winter).
38 Compressors and dehydrators are the primary contributors to emissions from these storage facilities. CH4 emissions
39 from the transmission and storage sector account for approximately 17 percent of emissions from natural gas
40 systems.
41 Trend information for transmission and storage has not yet been calculated. Based on the methods under
42 consideration for revising the time series, it is expected that the recalculated time series will show a decreasing trend
43 from 1990 for CH4 emissions from transmission and storage. Key industry changes impacting the trend include a
44 shift to larger and fewer centrifugal compressors from smaller reciprocating compressors, and a shift toward the
45 centrifugal compressors being equipped with dry seals rather than wet seals.
69 Available online at
Energy 3-67
-------
1 CO2 emissions from transmission and storage account for less than 1 percent of the non-combustion CO2 emissions
2 from natural gas systems. CC>2 emissions from transmission and storage have increased by 5 percent from 1990 to
3 2013 as the number of compressors has increased.
4 Distribution. Distribution pipelines take the high-pressure gas from the transmission system at "city gate" stations,
5 reduce the pressure and distribute the gas through primarily underground mains and service lines to individual end
6 users. There were 1,252,866 miles of distribution mains in 2013, an increase of nearly 310,000 miles since 1990
7 (PHMSA2014).
8 The EPA is considering several revisions to the distribution segment. For information on these revisions, and
9 requests for stakeholder feedback please see the Distribution Segment memo.70 The information presented here was
10 calculated using the approaches under consideration in the memorandum, as discussed below in the recalculations
11 discussion section.
12 Distribution system emissions, which account for 7 percent of CH4 emissions from natural gas systems, result
13 mainly from fugitive emissions from pipelines and stations.
14 Trend information for the distribution segment has not yet been calculated. Based on the methods under
15 consideration for updating the time series, it is expected that the recalculated time series will show a decreasing
16 trend from 1990 for CH4 emissions from distribution. Key industry changes impacting the trend include updates at
17 M&R stations and changes to lower emitting pipeline materials.
18 Distribution CO2 emissions in 2013 were 14 percent lower than 1990 levels (CO2 emission from this segment are
19 less than 0.1 MMTCCh Eq. across the time series).
20 Total CH4 emissions for the four major stages of natural gas systems are shown in MMT CCh Eq. (Table 3-43) and
21 kt (Table 3-44). For the final Inventory, more disaggregated information on potential emissions and emissions will
22 be available in Annex 3.6. See Methodology for Estimating CH4 and CCh Emissions from Natural Gas Systems.
23 Table 3-43: CH4 Emissions from Natural Gas Systems (MMT COz Eq.)a
Stage 1990
Production
Processing
Transmission and Storage
Distribution
Total
2005 2009 2010 2011 2012 2013
106
23
29
11
169
2014
a These values represent CH4 emitted to the atmosphere. CH4 that is captured, flared, or otherwise
controlled (and not emitted to the atmosphere) has been calculated and removed from emission totals.
Note: Totals may not sum due to independent rounding.
24 Table 3-44: CH4 Emissions from Natural Gas Systems (kt)a
~Stage 1990 2005 2009 2010 2011 2012 2013 2014
Production
Processing
Transmission and Storage
Distribution
4,230
906
1.151
458
Total 6,745
a These values represent CH4 emitted to the atmosphere. CH4 that is captured, flared, or otherwise
controlled (and not emitted to the atmosphere) has been calculated and removed from emission totals.
Note: Totals may not sum due to independent rounding.
70 Available online at
3-68 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 3-45: Non-combustion COz Emissions from Natural Gas Systems (MMT COz Eq.)
Stage
Field Production
Processing
Transmission and Storage
Distribution
Total
1990
9.8
27.8
+
2005
8.1
21.7
0.1
+
30.0
2009
10.9
21.2
0.1
+
32.2
2010
10.9
21.3
0.1
+
32.3
2011
14.0
21.5
0.1
+
35.6
2012
13.2
21.5
0.1
+
34.8
2013
15.9
21.8
0.1
+
37.8
2014
Note: Totals may not sum due to independent rounding.
+ Emissions are less than 0.1 MMT CCh Eq.
Table 3-46: Non-combustion COz Emissions from Natural Gas Systems (kt)
Stage
Field Production
Processing
Transmission and Storage
Distribution
Total
1990
9,775
27,763
62
46
37,645
2005
8,142
21,746
64
42
29,995
2009
10,906
21,188
65
41
32,201
2010
10,883
21,346
65
40
32,334
2011
13,980
21,466
65
40
35,551
2012
13,196
21,469
63
37
34,764
2013
15,947
21,757
65
40
37,808
2014
Note: Totals may not sum due to independent rounding.
3 Methodology
4 EPA has made revisions to the methodology and data sources for many sources in the Inventory. Please see the
5 Recalculations Discussion section below.
6 The methodology for natural gas emissions estimates presented in this Inventory involves the calculation of CH4 and
7 CO2 emissions for over 100 emissions sources, and then the summation of emissions for each natural gas sector
8 stage.
9 The calculation of emissions for each source of emissions in natural gas systems generally occurs in three steps:
10
11 Step 1. Calculate Potential Methane - Collect activity data on production and equipment in use and
12 apply emission factors (i.e., scf gas per unit or activity)
13 Step 2. Compile Reductions Data - Calculate the amount of the methane that is not emitted, using data on
14 voluntary action and regulations
15 Step 3. Calculate Net Emissions - Deduct methane that is not emitted from the total methane potential
16 estimates to develop net CH4 emissions, and calculate CCh emissions
17
18 This approach of calculating potential CH4 and then applying reductions data to calculate net emissions was used to
19 ensure an accurate time series that reflects real emission trends. As noted below, key data on emissions from many
20 sources are from a 1996 report containing data collected in 1992. Since the time of this study, practices and
21 technologies have changed. While this study still represents best available data for some emission sources, using
22 these emission factors alone to represent actual emissions without adjusting for emissions controls would in many
23 cases overestimate emissions. As updated emission factors reflecting changing practices are not available for most
24 sources, the 1992 emission factors continue to be used for many sources for all years of the Inventory, but they are
25 considered to be potential emissions factors, representing what emissions would be if practices and technologies had
26 not changed over time.
27 For the Inventory, the calculated potential emissions are adjusted using data on reductions reported to Natural Gas
28 STAR, and data on regulations that result in CH4 reductions. As more data become available, alternate approaches
29 may be considered. For example, new data on liquids unloading and on hydraulically fractured gas well completions
Energy 3-69
-------
1 and workovers enabled EPA to disaggregate or stratify these sources into distinct sub-categories based upon
2 different technology types, each with unique emission factors and activity data.
3 Step 1. Calculate Potential Methane—Collect activity data on production and equipment in use and apply
4 emission factors
5 In the first step, potential CH4 is calculated by multiplying activity data (such as miles of pipeline or number of
6 wells) by factors that relate that activity data to potential CH4. Potential CH4 is the amount of CH4 that would be
7 emitted in the absence of any control technology or mitigation activity. It is important to note that potential CH4
8 factors in most cases do not represent emitted CH4, and must be adjusted for any emissions-reducing technologies,
9 or practices, as appropriate. For more information, please see the Annex.
10 Potential Methane Factors
11 The primary basis for estimates of CH4 and non-combustion-related CCh emissions from the U.S. natural gas
12 industry is a detailed study by the Gas Research Institute and EPA (EPA/GRI 1996). The EPA/GRI study developed
13 over 80 CH4 emission factors to characterize emissions from the various components within the operating stages of
14 the U.S. natural gas system. The EPA/GRI study was based on a combination of process engineering studies,
15 collection of activity data and measurements at representative gas facilities conducted in the early 1990s. Methane
16 compositions from the Gas Technology Institute (GTI, formerly GRI) Unconventional Natural Gas and Gas
17 Composition Databases (GTI 2001) are adjusted year to year using gross production for oil and gas supply National
18 Energy Modeling System (NEMS) regions from the EIA. Therefore, emission factors may vary from year to year
19 due to slight changes in the CH4 composition for each NEMS oil and gas supply module region. The majority of
20 emission factors used in the Inventory were derived from the EPA/GRI study. The emission factors used to estimate
21 CH4 were also used to calculate non-combustion CCh emissions. Data from GTI 2001 were used to adapt the CH4
22 emission factors into non-combustion related CCh emission factors. Additional information about CCh content in
23 transmission quality natural gas was obtained from numerous U.S. transmission companies to help further develop
24 the non-combustion CCh emission factors.
25 Although the Inventory primarily uses EPA/GRI emission factors, updates were made to the emissions estimates for
26 several sources in recent Inventories. For liquids unloading, in the 2013 Inventory, the methodology was revised to
27 calculate national emissions through the use region-specific emission factors developed from well data collected in a
28 survey conducted by API/ANGA (API/ANGA 2012). In this methodology, the emission factors used for liquids
29 unloading are not potential factors, but are factors for actual emissions. For gas well completions and workovers
30 (refracturing) with hydraulic fracturing, in this Inventory, EPA used the 2011, 2012, and 2013 GHGRP Subpart W
31 data to stratify the emission sources into four different categories and developed CH4 emission factors for each
32 category. See the Recalculations Discussion below, and EPA memos "Inventory of U.S. Greenhouse Gas Emissions
33 and Sinks 1990-2013: Revision to Hydraulically Fractured Gas Well Completions and Workovers Estimate" and
34 "Updating GHG Inventory Estimate for Hydraulically Fractured Gas Well Completions and Workovers" for more
35 information on the methodology for this emission source (EPA 2013d and EPA 2015c).
36 In addition, in 2015, an update was made to the emission factors applied to offshore platforms. Previously, the
37 Inventory relied on the Bureau of Ocean Energy Management's (BOEM's) Gulf Offshore Activity Data System
38 (GOADS) year 2000 inventory to develop emission factors for offshore platforms; the methodology has been
39 updated to use more recent GOADS inventory data to develop emission factors. See the Recalculations Discussion
40 below, and EPA memo "Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2013: Revision to Offshore
41 Platforms Emissions Estimate" (EPA 2015b).
42 See Annex 3.6 for more detailed information on the methodology and data used to calculate CH4 and non-
43 combustion CO2 emissions from natural gas systems.
44 Updates to emission factors using GHGRP data for natural gas systems and other data continue to be evaluated.
45 Activity Data
46 Activity data were taken from the following sources: Drillinglnfo, Inc (Drillinglnfo 2014); American Gas
47 Association (AGA 1991-1998); Bureau of Ocean Energy Management, Regulation and Enforcement (previous
48 Minerals and Management Service) (BOEMRE 201 la, 201 Ib, 201 Ic, 201 Id); Natural Gas Liquids Reserves Report
49 (EIA 2005); Natural Gas Monthly (EIA 2014a, 2014b, 2014c); the Natural Gas STAR Program annual emissions
50 savings (EPA 2013c); Oil and Gas Journal (OGJ 1997-2014); Pipeline and Hazardous Materials Safety
3-70 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Administration (PHMSA 2014); Federal Energy Regulatory Commission (FERC 2014); Greenhouse Gas Reporting
2 Program (EPA 2014); other Energy Information Administration data and publications (EIA 2001, 2004, 2012, 2013,
3 2014). Data for estimating emissions from hydrocarbon production tanks were incorporated (EPA 1999). Coalbed
4 CH4 well activity factors were taken from the Wyoming Oil and Gas Conservation Commission (Wyoming 2014)
5 and the Alabama State Oil and Gas Board (Alabama 2014).
6 For a few sources, recent direct activity data are not available. For these sources, either 2012 data was used as proxy
7 for 2013 data, or a set of industry activity data drivers was developed and used to update activity data. Drivers
8 include statistics on gas production, number of wells, system throughput, miles of various kinds of pipe, and other
9 statistics that characterize the changes in the U.S. natural gas system infrastructure and operations. For example,
10 recent data on various types of field separation equipment in the production stage (i.e., heaters, separators, and
11 dehydrators) are unavailable. Each of these types of field separation equipment was determined to relate to the
12 number of non-associated gas wells. Using the number of each type of field separation equipment estimated by
13 GRI/EPA in 1992, and the number of non-associated gas wells in 1992, a factor was developed that is used to
14 estimate the number of each type of field separation equipment throughout the time series. More information on
15 activity data and drivers is available in Annex 3.6.
16 Step 2. Compile Reductions Data—Calculate the amount of the CH4 that is not emitted, using data on voluntary
17 action and regulations
18 The emissions calculated in Step 1 above represent potential emissions from an activity, and do not take into account
19 any use of technologies and practices that reduce emissions. To take into account use of such technologies, data,
20 where available, are collected on both regulatory and voluntary reductions. Regulatory actions reducing emissions
21 include National Emission Standards for Hazardous Air Pollutants (NESHAP) regulations for dehydrator vents and
22 condensate tanks. Voluntary reductions included in the Inventory are those reported to Natural Gas STAR for
23 activities such as replacing a high bleed pneumatic device with a low bleed device, and replacing wet seals with dry
24 seals at reciprocating compressors. For more information on these reductions, please see the Annex. The emission
25 estimates presented in Table 3-43 and Table 3-44 are the CH4 that is emitted to the atmosphere (i.e., net emissions),
26 not potential emissions without capture or flaring.
27 The Inventory includes impacts of the New Source Performance Standards (NSPS), which came into effect in
28 October 2012, for oil and gas (EPA 2013b). By separating gas well completions and workovers with hydraulic
29 fracturing into four categories and developing control technology-specific CH4 emission factors for each category,
30 EPA is implicitly accounting for NSPS reductions from hydraulically fractured gas wells. The NSPS also has VOC
31 reduction requirements for compressors, storage vessels, pneumatic controllers, and equipment leaks at processing
32 plants, which will also impact CH4 emissions in future Inventories.
33 Step 3. Calculate Net Emissions—Deduct CH4 that is not emitted from the total CH4 potential estimates to develop
34 net CH4 emissions, and calculate CO2 emissions
35 In the final step, emission reductions from voluntary and regulatory actions are deducted from the total calculated
36 potential emissions to estimate the net emissions that are presented in Table 3-43, and included in the Inventory
37 totals. Note that for liquids unloading, condensate tanks, gas well completions and workovers with hydraulic
38 fracturing, and centrifugal compressors, emissions are calculated directly using emission factors that vary by
39 technology and account for any control measures in place that reduce CH4 emissions. See Annex table A-17 for
40 more information on net emissions for specific sources.
41
Uncertainty and Time-Series Consistency
42 The most recent uncertainty analysis for the natural gas and petroleum systems emission estimates in the Inventory
43 was conducted for the 1990-2009 Inventory report that was released in 2011. Since the analysis was last conducted,
44 several of the methods used in the Inventory have changed, and industry practices and equipment have evolved. In
45 addition, new studies and other data sources such as those discussed in this memorandum offer improvement to
46 understanding and quantifying the uncertainty of some emission source estimates.
47 As updates to the greenhouse gas Inventory data and methods are selected, the EPA will review information on
48 uncertainty and consider how the greenhouse gas Inventory uncertainty assessment can be updated to reflect the new
49 information.
Energy 3-71
-------
1 A quantitative uncertainty analysis was conducted in 2010 to determine the level of uncertainty surrounding
2 estimates of emissions from natural gas systems using the IPCC-recommended Approach 2 methodology (Monte
3 Carlo Simulation technique). The @RISK software model was used to quantify the uncertainty associated with the
4 emissions estimates using the 12 highest-emitting sources ("top 12 sources") for the year 2009. The @RISK analysis
5 provides for the specification of probability density functions for key variables within a computational structure that
6 mirrors the calculation of the Inventory estimate. The IPCC guidance notes that in using this method, "some
7 uncertainties that are not addressed by statistical means may exist, including those arising from omissions or double
8 counting, or other conceptual errors, or from incomplete understanding of the processes that may lead to
9 inaccuracies in estimates developed from models." As a result, the understanding of the uncertainty of emissions
10 estimates for this category evolves and improves as the underlying methodologies and datasets improve.
11 The uncertainty analysis conducted in 2010 has not yet been updated for the 1990 through 2013 Inventory years;
12 instead, the uncertainty percentage ranges calculated previously were applied to 2013 emissions estimates. The
13 majority of sources in the current Inventory were calculated using the same emission factors and activity data for
14 which PDFs were developed in the 1990 through 2009 uncertainty analysis. As explained in the Methodology
15 section above and the recalculations discussion below, several emission sources have undergone recent methodology
16 revisions, and the 2009 uncertainty ranges will not reflect the uncertainty associated with the recently revised
17 emission factors and activity data sources. Please see discussion on Planned Improvements.
18 The results presented below provide with 95 percent certainty the range within which emissions from this source
19 category are likely to fall for the year 2013, based on the previously-conducted uncertainty assessment using the
20 recommended IPCC methodology. The heterogeneous nature of the natural gas industry makes it difficult to sample
21 facilities that are completely representative of the entire industry. Additionally, highly variable emission rates were
22 measured among many system components, making the calculated average emission rates uncertain. The results of
23 the Approach 2 quantitative uncertainty analysis are summarized in Table 3-47. Natural gas systems CH4 emissions
24 in 2013 were estimated to be between 127.5 and 187.3 MMT CCh Eq. at a 95 percent confidence level. Natural gas
25 systems non-energy CO2 emissions in 2013 were estimated to be between 30.6 and 49.1 MMT CO2 Eq. at 95
26 percent confidence level.
27 Table 3-47: Approach 2 Quantitative Uncertainty Estimates for CH4 and Non-energy COz
28 Emissions from Natural Gas Systems (MMT COz Eq. and Percent)
_ 2013 Emission Estimate Uncertainty Ranee Relative to Emission Estimate3
Source Gas
(MMT CO2 Eq.)b (MMT CCh Eq.) (%)
32
Natural Gas Systems
Natural Gas Systems0
CH4
C02
157.4
37.8
Lower
Bound"
127.5
30.6
Upper
Bound"
187.3
49.1
Lower
Bound"
-19%
-19%
Upper
Bound"
+30%
+30%
a Range of emission estimates predicted by Monte Carlo Simulation for a 95 percent confidence interval.
b All reported values are rounded after calculation. As a result, lower and upper bounds may not be duplicable from
other rounded values as shown in Table 3-44 and Table 3-45.
0 An uncertainty analysis for the non-energy CCh emissions was not performed. The relative uncertainty estimated
(expressed as a percent) from the CH4 uncertainty analysis was applied to the point estimate of non-energy CCh
emissions.
29 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
30 through 2013. Details on the emission trends through time are described in more detail in the Methodology section,
31 above.
QA/QC and Verification Discussion
33 The natural gas emission estimates in the Inventory are continually being reviewed and assessed to determine
34 whether emission factors and activity factors accurately reflect current industry practices. A QA/QC analysis was
35 performed for data gathering and input, documentation, and calculation. QA/QC checks are consistently conducted
36 to minimize human error in the model calculations. EPA performs a thorough review of information associated with
37 new studies, GHGRP data, regulations, public webcasts, and the Natural Gas STAR Program to assess whether the
38 assumptions in the Inventory are consistent with current industry practices. In addition, EPA receives feedback
3-72 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 through annual expert and public review periods. Feedback received is noted in the Recalculations and Planned
2 Improvement sections.
3
Recalculations Discussion
4 The EPA received information and data related to the emission estimates through the Inventory preparation process,
5 previous Inventories' formal public notice periods, GHGRP data, and new studies. The EPA carefully evaluated
6 relevant information available, and made several updates in the public review draft, including revisions to
7 production segment activity data, production segment pneumatic controller activity and emissions data, gathering
8 and boosting facility emissions, transmission and storage station activity and emissions data, distribution segment
9 emissions data for pipelines, distribution segment M&R station activity and emissions data, and distribution segment
10 customer meter emissions data.
11 From December 2015 through February 2016, the EPA released four draft memos that discussed the changes under
12 consideration and requested stakeholder feedback on those changes.
13 • Revisions under Consideration for Natural Gas and Petroleum Production Emissions71
14 • Revisions under Consideration for Gathering and Boosting Emissions72
15 • Revisions under Consideration for Transmission and Storage Emissions73
16 • Revisions under Consideration for Distribution Emissions74
17 The EPA is continuing to evaluate stakeholder feedback on the updates under consideration. For the final GHG
18 Inventory, the 2013 estimates presented in this section will be refined, and a full time series of emissions estimates
19 will be developed based on feedback received through the earlier stakeholder reviews of information released in the
20 segment-specific memoranda, and through this public review period. The combined impact of all revisions to 2013
21 natural gas production segment emissions in the public review draft, described below, compared to the 2015
22 Inventory, is an increase in CH4 emissions of approximately 10 MMT CO2 Eq., or 6 percent. Recalculations for the
23 production segment (including gathering and boosting facilities) resulted in a large increase in the 2013 CH4
24 emission estimate, from 47.0 MMT CO2 Eq. in the previous (2015) Inventory, to 105.7 MMT CO2 Eq. in the current
25 (2016) Inventory. Recalculations for the transmission and storage segment resulted in a large decrease in the 2013
26 CH4 emission estimate, from 54.4 MMT CO2 Eq. in the previous (2015) Inventory, to 28.8 MMT CO2 Eq. in the
27 current (2016) Inventory. Recalculations for the distribution segment also resulted in a large decrease in the 2013
28 CH4 emission estimate, from 33.3 MMT CO2 Eq. in the previous (2015) Inventory, to 10.2 MMT CO2 Eq. in the
29 current (2016) Inventory.
30 Time series information has not yet been calculated, but it is expected that across the 1990 through 2013 time series,
31 compared to the previous (2015) Inventory, in the current (2016) Inventory, the total CH4 emissions estimate will
32 increase, with the largest increases in the estimate occurring in later years of the time series.
33 Production
34 This section references the Production Segment memo, available at
35 http://www3.epa.gov/climatechange/ghgemissions/usinventoryreport/DRAFT Proposed Revision to Production S
36 egment Emissions 2-3-2016.pdf and the Gathering and Boosting memo, available at
71 Available online at:
72 Available online at:
73 Available online at:
74 Available online at: <
http://www3. epa.gov/climatechange/ghgemissions/usinventoryreport/Propo sed_Revisions_to_NG_Distribution_Segment_Emissi
ons.pdf>
Energy 3-73
-------
1 http://www3.epa.gov/climatechange/ghgemissions/usinventoryreport/DRAFT Proposed Revision to NG Gatherin
2 g-Boosting Segment Emissions 2-2-2016.pdf
3 Key data sources assessed for potential updates to the production segment include GHGRP and Marchese et al.
4 Using newly available GHGRP activity data, the EPA has developed activity factors (i.e., counts per gas well) for
5 compressors, separators, dehydrators, in-line heaters, meters/piping, pneumatic pumps, and pneumatic controllers.
6 The EPA has applied these updated activity factors (see Reporting Year (RY) 2014 values in Table 7 of the
7 Production Segment memo) to calculate emissions from these sources in this public review draft. The impact of
8 using activity factors developed from GHGRP data is an increase in emissions. Comparing the updated 2013
9 estimate to the previous Inventory 2013 estimate, the CH4 emissions increase due to use of revised activity factors
10 for compressors, separators, dehydrators, in-line heaters, meters/piping, and pneumatic pumps is approximately 3.3
11 MMTCO2Eq.
12
13 Using the GHGRP data, the EPA has also developed technology-specific activity data and emission factors for
14 pneumatic controllers. Reported data under EPA's GHGRP allow for development of emission factors specific to
15 bleed type (continuous high bleed, continuous low bleed, and intermittent bleed) and separation of activity data into
16 these categories. For the public review draft, EPA has used this separation of pneumatic controller counts by bleed
17 types, and emission factors developed from reported GHGRP data. See RY2013 values in Table 11 (activity data
18 for proportion of controllers in each category), and RY2011-2014 values in Table 10 (emission factors) of the
19 Production Segment memo. Comparing the updated 2013 estimate to the previous Inventory 2013 estimate, the
20 impact of using bleed type-specific emission factors and activity data developed from GHGRP data is an increase of
21 approximately 12.5 MMT CO2Eq.
22
23 The EPA is considering approaches to develop the greenhouse gas Inventory time series (1990-2014) for these
24 sources that will create consistency between earlier years' estimates that generally rely on data from GRI/EPA 1996,
25 and more recent years' estimates calculated with GHGRP data. Foryears 1992-2010, the EPA is considering
26 calculating well counts by interpolating the equipment counts per well from the data points of 1992 (GRI/EPA) and
27 2011 (GHGRP).
28
29 A 2015 study, Marchese et al. assessed CH4 emissions from an expanded universe of gathering stations than were
30 previously included in the Inventory. The Marchese et al. study analyzed emissions from five different types of
31 gathering stations: compression only; compression and dehydration; compression, dehydration, and acid gas
32 removal; dehydration only; and dehydration and acid gas removal. Previous Inventories estimated emissions from
33 only gathering compression stations. In the public review draft estimate presented here, the EPA has applied a
34 station-level emission factor and national activity estimates developed from the Marchese et al. data. See the
35 Gathering and Boosting memo for more information. The impact of using revised activity data and emission factors
36 for gathering stations cannot be straightforwardly determined based on the structure of previous GHG Inventories
37 (e.g., dehydrator emissions in previous Inventories are not differentiated between well pad and gathering facility
38 locations); however due to the activity data revision alone, production segment emissions greatly increase compared
39 to previous estimates. For the year 2013, the CH4 emissions from gathering facilities are 43.3 MMT CChEq. The
40 EPA is considering applying the emission factor to all years of the time series, and replacing current activity data
41 estimates with station counts based on the Marchese et al. estimate (scaled for earlier years based on national natural
42 gas marketed production).
43 The EPA's approach for revising the Inventory methodology to incorporate GHGRP data and Marchese et al. data
44 obviates the need to apply Gas STAR reductions data for certain sources in the production segment. EPA plans to
45 carry forward reported reductions for sources that are not being revised to use a net emission factor approach. There
46 are also significant Gas STAR reductions in the production segment that are not classified as applicable to specific
47 emission sources ("other voluntary reductions" are 16MMTCO2Eq. CH4 in year 2013). Some portion of the "other
48 voluntary reductions" might apply to the emission sources for which the EPA is revising the methodology to use a
49 net emission factor approach. The EPA is investigating potential disaggregation of "other voluntary reductions." The
50 EPA has retained Gas STAR reductions classified as "other voluntary reductions" in year 2013 emission
51 calculations for the public review draft.
3-74 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
i Transmission and Storage
2 This section references the Transmission and Storage memo, available at
3 http://www3.epa.gov/climatechange/ghgemissions/usinventorvreport/DRAFT Proposed Revisions to NGTransm
4 ission Storage Segment Emissions 2016-01-20.pdf.
5 Key data sources assessed for the transmission and storage segment updates include GHGRP and Zimmerle et al.
6 For transmission and storage non-compressor fugitive emissions in the public review draft, EPA has used Zimmerle
7 et al. data to develop the activity data for transmission stations ("Alternative") and EIA data on active storage fields,
8 along with the Zimmerle estimate of storage stations per storage field to develop storage station counts. See Table 3
9 of the Transmission and Storage memo. The EPA then applied emission factors from Zimmerle et al. to calculate
10 emissions for fugitives from these sources. See Table 4 of the Transmission and Storage memo.
11 The EPA is considering using interpolation between GRI/EPA emission factors in early years and Zimmerle et al.
12 emission factors in recent years, which will reflect net emissions across the Inventory time series. However, the
13 station fugitive emission factors in previous Inventories included station fugitives but not compressor fugitives, and
14 separate emission factors were applied for compressor emissions (including compressor fugitive and vented
15 sources). Because Zimmerle et al. grouped compressor fugitives with station fugitives, the two sets of emission
16 factors (GRI/EPA and Zimmerle et al.) cannot be directly compared. Therefore in this public review draft, the EPA
17 calculated total station-level emission factors for transmission and storage stations that include station and
18 compressor fugitive sources as well as compressor vented sources.
19 In this public review draft, the EPA incorporated Zimmerle et al. national population estimates of reciprocating and
20 centrifugal compressor activity data, along with the GHGRP proportions of centrifugal compressor seal types (wet
21 versus dry seals), and Zimmerle et al. emission factor data, in development of emission estimates for compressors in
22 transmission and storage. See Table 12 (reciprocating compressors EF from Zimmerle et al.), Table 14
23 (reciprocating and centrifugal compressor counts from Zimmerle et al.), Table 18 (split between wet and dry seal
24 from Subpart W), and Table 22 (Zimmerle et al. EF for centrifugal compressors), in the Transmission and Storage
25 memo.
26 For example, the EPA is considering approaches to calculate the Inventory time series (1990-2014) that will create
27 consistency between earlier years' and more recent years' estimates for compressors. For example, for all years
28 between 1992 and 2012, a linear correlation between 1992 and 2012 counts could be applied.
29 The overall impact of using revised emissions data and activity data from Zimmerle et al. and GHGRP is a decrease
30 in emissions for station fugitives and compressors. Comparing the updated 2013 estimate to the previous Inventory
31 2013 estimate, the CH4 emissions decrease due to use of revised emission factors and activity data for transmission
32 and storage station fugitives and compressor venting is approximately 21.7 MMT €62 Eq.
33 In the public review draft value presented here, the transmission and storage pneumatic controller values have been
34 calculated using the Zimmerle et al. data on controllers per station and emission factors. See Table 26, and Table 29
35 of the Transmission and Storage memo. The overall impact of using revised emissions data and activity data from
36 Zimmerle et al. or GHGRP would be a decrease in emissions from transmission station pneumatic controllers and
37 either a slight increase or decrease in emissions from storage station pneumatic controllers (depending on use of
38 Zimmerle et al. data or subpart GHGRP, respectively) for recent time series years. Comparing the updated 2013
39 estimate to the previous Inventory 2013 estimate, the CH4 emissions decrease due to use of revised emission factors
40 and activity data for transmission and storage station pneumatic controllers is 4.0 MMT €62 Eq. The EPA may
41 apply current Inventory AD and/or EF to early years of the time series, and assume a linear correlation to develop
42 year-specific AD and/or EFs for intermediate years, and apply the GHGRP or Zimmerle et al. AD and EF to recent
43 years of the Inventory time series.
44 The EPA's approach for revising the Inventory methodology to incorporate Zimmerle et al. and GHGRP data in the
45 transmission segment would result in net emissions being directly calculated for these sources in each time series
46 year. This obviates the need to apply Gas STAR reductions data for these sources. Previous Inventories have applied
47 Gas STAR reductions to other specific transmission and storage segment sources including compressor engine and
48 pipeline venting. EPA plans to carry forward reported reductions for these sources since they are not being revised
49 to use a net emission factor approach. There are also Gas STAR reductions in the transmission and storage segment
50 that are not classified as applicable to specific emission sources ("other voluntary reductions" are 3.6 MMT €62 Eq.
51 CH4 in year 2013). Some portion of the "other voluntary reductions" might apply to the emission sources for which
Energy 3-75
-------
1 the EPA is revising methodology to use a net emission factor approach. The EPA is investigating potential
2 disaggregation of "other voluntary reductions." The EPA has retained Gas STAR reductions classified as "other
3 voluntary reductions" in year 2013 emission calculations for the public review draft.
4 Distribution
5 This section references the Distribution memo, available at.
6 http://www3.epa.gov/climatechange/ghgemissions/usinventoryreport/Proposed Revisions to NG Distribution See
7 ment Emissions.pdf.
8 Key data reviewed for updates to the distribution segment include GHGRP, and Lamb et al.
9 For M&R stations, in this public review draft, the EPA used GHGRP reported activity data for counts of above
10 ground and below ground stations. The EPA scaled the GHGRP station counts to the national level based on the
11 miles of distribution pipeline main reported by GHGRP reporters, compared to the PHMSA national total miles of
12 distribution pipeline main. The EPA then applied the existing Inventory (from GRI) proportions of stations in the
13 various station inlet pressure categories to the scaled counts of above ground and below ground M&R stations, and
14 the station-level emission factors from Lamb et al. See Table 3 (GHGRP reported activity data), Table 2 (GHG
15 Inventory split between station types) and Table 4 (Lamb et al. emission factors) in the Distribution Memo.
16 Comparing the updated 2013 estimate to the previous Inventory 2013 estimate, the M&R stations CH4 emissions
17 decrease due to use of revised emission factors and activity data is approximately 13.6 MMT CC^Eq.
18 For pipeline leaks, in this public review draft, the EPA used the previous activity data sources for miles of pipeline
19 by material (PHMSA) and for leaks per mile (GRI), and Lamb et al. data on emissions per leak. See Table 7
20 (activity data on miles of pipeline), and Table 8 (leaks per mile, 2015 Inventory, leak rate, Lamb et al.) in the
21 Distribution Memo. Comparing the updated 2013 estimate to the previous Inventory 2013 estimate, the pipeline
22 leaks CH4 emissions decrease due to use of revised emission factors is approximately 10.5 MMT CO2 Eq.
23 For the time series, the EPA does not plan to revise existing estimates for M&R activity data for years before
24 GHGRP data are available (1990-2011). Using interpolation between GRI/EPA emission factors in early years and
25 Lamb et al. emission factors in recent years for M&R stations and pipeline leaks will reflect net emissions across the
26 Inventory time series without the need to incorporate Gas STAR reductions.
27 The EPA also considered recent data available to estimate emissions from customer meters. In this public review
28 draft, the EPA revised the emission factors for residential customer meters and commercial/industrial customer
29 meters. The EPA recalculated the residential customer meter emission factor by combining data from the 1996
30 GRI/EPA study (basis for existing greenhouse gas Inventory emission factor) with newer data from a GTI2009
31 study and Clearstone 2011 study. See Table 14 of the Distribution segment memo for the emission factor data for
32 residential meters. The EPA weighted emission factors developed in each study by the number of meters surveyed
33 in each study to develop the revised emission factor. In this public review draft, the EPA applied the GTI 2009
34 commercial customer meter emission factor to the total count of commercial and industrial meters in the greenhouse
35 gas Inventory. See Table 14 of the Distribution memo for the GTI emission factor for commercial meters. In
36 addition, the EPA used an updated data source for national customer meter counts (EIA data); previously, national
37 customer meter counts were scaled from a 1992 based year value but are now available directly for every year of the
38 time series from EIA. Comparing the updated 2013 estimate to the previous Inventory 2013 estimate, the customer
39 meters CH4 emissions increase due to use of revised emission factors and activity data is approximately 0.3 MMT
40 CChEq. The EPA is considering making this change (to use updated activity data and emission factors) for each
41 year of the time series.
42 For pipeline blowdowns and mishaps/dig-ins, the previous Inventories used base year 1992 distribution main and
43 service miles and scaled the value for non-1992 years using relative residential gas consumption. However, scaling
44 mileage based on residential gas consumption introduced volatility across the time series that does not likely
45 correlate to pipeline mileage trends (as gas consumption is affected by other factors such as equipment efficiency
46 and climate). The EPA used PHMSA data for the activity data to calculate the estimate in this public review draft.
47 See Table 18 for PHMSA AD. The overall impact of using the revised activity data for pipeline blowdowns and
48 mishaps/dig-ins is an increase in emissions. Comparing the updated 2013 estimate to the previous Inventory 2013
49 estimate, the pipeline blowdowns CH4 emissions increase due to use of revised activity data is approximately 0.04
50 MMT CChEq.; and for mishaps/dig-ins is approximately 0.6 MMT CChEq.
3-76 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1
2
o
6
4
5
6
7
8
9
10
11
12
13
The EPA's approach for revising Inventory methodology to incorporate Lamb et al. and subpart W data would result
in net emissions being directly calculated for M&R stations, pipeline leaks, and customer meters in each time series
year. This obviates the need to apply Gas STAR reductions data for these sources. Previous GHG Inventories have
applied Gas STAR reductions to mishaps/dig-ins. EPA plans to carry forward reported reductions for this source
since it is not being revised to use a net emission factor approach. There are also Gas STAR reductions in the
distribution segment that are not classified as applicable to specific emission sources ("other voluntary reductions"
are 1.0 MMT CCh Eq. CH4 in year 2013). Some portion of the "other voluntary reductions" might apply to the
emission sources for which the EPA is revising methodology to use a net emission factor approach. The EPA is
investigating potential disaggregation of "other voluntary reductions." The EPA has retained Gas STAR reductions
classified as "other voluntary reductions" in year 2013 emission calculations for the public review draft.
Planned Improvements
EPA will continue to refine the emission estimates to reflect the most robust information available, including
stakeholder feedback received on the greenhouse gas inventory improvement memos, and this public review draft.
14
15
3.8 Energy Sources of Indirect Greenhouse Gas
Emissions
16 In addition to the main greenhouse gases addressed above, many energy-related activities generate emissions of
17 indirect greenhouse gases. Total emissions of nitrogen oxides (NOX), carbon monoxide (CO), and non-CH4 volatile
18 organic compounds (NMVOCs) from energy-related activities from 1990 to 2014 are reported in Table 3-48.
19 Table 3-48: NOX, CO, and NMVOC Emissions from Energy-Related Activities (kt)
Gas/Source
NOx
Mobile Combustion
Stationary Combustion
Oil and Gas Activities
Waste Combustion
International Bunker Fuels"
CO
Mobile Combustion
Stationary Combustion
Waste Combustion
Oil and Gas Activities
International Bunker Fuels"
NMVOCs
Mobile Combustion
Oil and Gas Activities
Stationary Combustion
Waste Combustion
International Bunker Fuels"
1990
21,106
10,862
10,023
139 1
82 1
1,956
125,640
119,360
5,000
978
302 1
103 1
12,620
10,932
554
912 1
222 1
57 H
2005 2010
16,602 12,004
10,295 7,290
5,858 4,092
321 545
128 1 77
1,704 1,790
64,985 45,148
58,615 39,475
4,648 4,103
1,403 1,084
318 487
133 1 136
7,191 7,464
5,724 4,591
510 2,205
716 1 576
241 1 92
54 56
2011
11,796
7,294
3,807
622
73
1,553
44,088
38,305
4,170
1,003
610
137
7,759
4,562
2,517
599
81
51
2012
11,051
6,788
3,567
622
73
1,398
42,273
36,491
4,170
1,003
610
133
7,449
4,252
2,517
599
81
46
2013
10,557
6,283
3,579
622
73
1,139
40,459
34,676
4,170
1,003
610
129
7,139
3,942
2,517
599
81
41
2014
9,995
5,777
3,522
622
73
1,138
38,643
32,861
4,169
1,003
610
135
6,830
3,632
2,517
599
81
42
" These values are presented for informational purposes only and are not included in totals.
Note: Totals may not sum due to independent rounding.
20
21
22
Methodology
Emission estimates for 1990 through 2014 were obtained from data published on the National Emission Inventory
(NEI) Air Pollutant Emission Trends web site (EPA 2015), and disaggregated based on EPA (2003). Emission
Energy 3-77
-------
1 estimates for 2012, 2013, and 2014 for non-EGU and non-mobile sources are held constant from 2011 in EPA
2 (2015). Emissions were calculated either for individual categories or for many categories combined, using basic
3 activity data (e.g., the amount of raw material processed) as an indicator of emissions. National activity data were
4 collected for individual applications from various agencies.
5 Activity data were used in conjunction with emission factors, which together relate the quantity of emissions to the
6 activity. Emission factors are generally available from the EPA's Compilation of Air Pollutant Emission Factors,
7 AP-42 (EPA 1997). The EPA currently derives the overall emission control efficiency of a source category from a
8 variety of information sources, including published reports, the 1985 National Acid Precipitation and Assessment
9 Program emissions inventory, and other EPA databases.
10
16
17
Uncertainty and Time-Series Consistency
1 1 Uncertainties in these estimates are partly due to the accuracy of the emission factors used and accurate estimates of
12 activity data. A quantitative uncertainty analysis was not performed.
13 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
14 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
15 above.
3.9 International Bunker Fuels (IPCC Source
Category 1: Memo Items)
18 Emissions resulting from the combustion of fuels used for international transport activities, termed international
19 bunker fuels under the UNFCCC, are not included in national emission totals, but are reported separately based upon
20 location of fuel sales. The decision to report emissions from international bunker fuels separately, instead of
21 allocating them to a particular country, was made by the Intergovernmental Negotiating Committee in establishing
22 the Framework Convention on Climate Change.75 These decisions are reflected in the IPCC methodological
23 guidance, including the 2006 IPCC Guidelines, in which countries are requested to report emissions from ships or
24 aircraft that depart from their ports with fuel purchased within national boundaries and are engaged in international
25 transport separately from national totals (IPCC 2006).76
26 Two transport modes are addressed under the IPCC definition of international bunker fuels: aviation and marine.77
27 Greenhouse gases emitted from the combustion of international bunker fuels, like other fossil fuels, include CO2,
28 CH4 and N2O for marine transport modes, and CC>2 and N2O for aviation transport modes. Emissions from ground
29 transport activities—by road vehicles and trains—even when crossing international borders are allocated to the
30 country where the fuel was loaded into the vehicle and, therefore, are not counted as bunker fuel emissions.
31 The IPCC Guidelines distinguish between different modes of air traffic. Civil aviation comprises aircraft used for
32 the commercial transport of passengers and freight, military aviation comprises aircraft under the control of national
33 armed forces, and general aviation applies to recreational and small corporate aircraft. The IPCC Guidelines further
34 define international bunker fuel use from civil aviation as the fuel combusted for civil (e.g., commercial) aviation
35 purposes by aircraft arriving or departing on international flight segments. However, as mentioned above, and in
75 See report of the Intergovernmental Negotiating Committee for a Framework Convention on Climate Change on the work of
its ninth session, held at Geneva from 7 to 18 February 1994 (A/AC.237/55, annex I, para. Ic).
7^ Note that the definition of international bunker fuels used by the UNFCCC differs from that used by the International Civil
Aviation Organization.
77 Most emission related international aviation and marine regulations are under the rubric of the International Civil Aviation
Organization (ICAO) or the International Maritime Organization (IMO), which develop international codes, recommendations,
and conventions, such as the International Convention of the Prevention of Pollution from Ships (MARPOL).
3-78 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 keeping with the IPCC Guidelines, only the fuel purchased in the United States and used by aircraft taking-off (i.e.,
2 departing) from the United States are reported here. The standard fuel used for civil aviation is kerosene-type jet
3 fuel, while the typical fuel used for general aviation is aviation gasoline.78
4 Emissions of CC>2 from aircraft are essentially a function of fuel use. N2O emissions also depend upon engine
5 characteristics, flight conditions, and flight phase (i.e., take-off, climb, cruise, decent, and landing). Recent data
6 suggest that little or no CH4 is emitted by modern engines (Anderson et al. 2011), and as a result, CH4 emissions
7 from this category are considered zero. In jet engines, N2O is primarily produced by the oxidation of atmospheric
8 nitrogen, and the majority of emissions occur during the cruise phase. International marine bunkers comprise
9 emissions from fuels burned by ocean-going ships of all flags that are engaged in international transport. Ocean-
10 going ships are generally classified as cargo and passenger carrying, military (i.e., U.S. Navy), fishing, and
11 miscellaneous support ships (e.g., tugboats). For the purpose of estimating greenhouse gas emissions, international
12 bunker fuels are solely related to cargo and passenger carrying vessels, which is the largest of the four categories,
13 and military vessels. Two main types of fuels are used on sea-going vessels: distillate diesel fuel and residual fuel
14 oil. CO2 is the primary greenhouse gas emitted from marine shipping.
15 Overall, aggregate greenhouse gas emissions in 2014 from the combustion of international bunker fuels from both
16 aviation and marine activities were 104.2 MMT CC>2 Eq., or 0.3 percent below emissions in 1990 (see Table 3-49
17 and Table 3-50). Emissions from international flights and international shipping voyages departing from the United
18 States have increased by 82.2 percent and decreased by 48.4 percent, respectively, since 1990. The majority of these
19 emissions were in the form of CCh; however, small amounts of CH4 (from marine transport modes) and N2O were
20 also emitted.
21 Table 3-49: COz, CH4, and NzO Emissions from International Bunker Fuels (MMT COz Eq.)
Gas/Mode
CO2
Aviation
Commercial
Military
Marine
CH4
Aviation*
Marine
N2O
Aviation
Marine
Total
1990
103.5
38.0
30.0
8.1
65.4
0.2
0.0
0.2
0.9
0.4
0.5
104.5
2005
113.1
60.1
55.6
4.5
53.0
0.1
0.0
0.1
1.0
0.6
0.4
114.2
2010
117.0
61.0
57.4
3.6
56.0
0.1
0.0
0.1
1.0
0.6
0.4
118.1
2011
111.7
64.8
61.7
3.1
46.9
0.1
0.0
0.1
1.0
0.6
0.4
112.8
2012
105.8
64.5
61.4
3.1
41.3
0.1
0.0
0.1
0.9
0.6
0.3
106.8
2013
99.8
65.7
62.8
2.9
34.1
0.1
0.0
0.1
0.9
0.6
0.2
100.7
2014
103.2
69.4
66.3
3.1
33.8
0.1
0.0
0.1
0.9
0.7
0.2
104.2
Note: Totals may not sum due to independent rounding. Includes aircraft cruise altitude emissions.
aCH4 emissions from aviation are estimated to be zero.
22 Table 3-50: COz, CH4 and NzO Emissions from International Bunker Fuels (kt)
Gas/Mode
C02
Aviation
Marine
CH4
Aviation*
Marine
N2O
Aviation
Marine
78 Naphtha-type jet
1990
103,463
38,034
65,429
7 1
°l
7
3
1
2|
fuel was used
2005
113,139
60,125
53,014
5
°
5
3
2
1
in the past by
2010
116,992
160,967
56,025
6
o
6
3
2
1
the military in
2011
111,660
64,790
46,870
5
0
5
3
2
1
turbojet and
2012
105,805
64,524
41,281
4
0
4
3
2
1
2013
2014
99,763 103,201
65,664
34,099
3
0
3
3
2
1
69,411
33,791
3
0
3
3
2
1
turboprop aircraft engines.
Energy 3-79
-------
Note: Totals may not sum due to independent rounding. Includes aircraft cruise altitude emissions.
aCH4 emissions from aviation are estimated to be zero.
1 Table 3-51: Aviation COz and NzO Emissions for International Transport (MMT COz Eq.)
Aviation Mode
Commercial Aircraft
Military Aircraft
Total
1990
30.0
8.1
38.0
2005
55.6
4.5
60.1
2010
57.4
1 3.6
61.0
2011
61.7
3.1
64.8
2012
61.4
3.1
64.5
2013
62.8
2.9
65.7
2014
66.3
3.1
69.4
Note: Totals may not sum due to independent rounding. Includes aircraft cruise altitude emissions.
2 Methodology
3 Emissions of CO2 were estimated by applying C content and fraction oxidized factors to fuel consumption activity
4 data. This approach is analogous to that described under CO2 from Fossil Fuel Combustion. Carbon content and
5 fraction oxidized factors for jet fuel, distillate fuel oil, and residual fuel oil were taken directly from El A and are
6 presented in Annex 2.1, Annex 2.2, and Annex 3.8 of this Inventory. Density conversions were taken from Chevron
7 (2000), ASTM (1989), and USAF (1998). Heat content for distillate fuel oil and residual fuel oil were taken from
8 EIA (2015) and USAF (1998), and heat content for jet fuel was taken from EIA (2015). A complete description of
9 the methodology and a listing of the various factors employed can be found in Annex 2.1. See Annex 3.8 for a
10 specific discussion on the methodology used for estimating emissions from international bunker fuel use by the U.S.
11 military.
12 Emission estimates for CH4 and N2O were calculated by multiplying emission factors by measures of fuel
13 consumption by fuel type and mode. Emission factors used in the calculations of CH4 and N2O emissions were
14 obtained from the 2006IPCC Guidelines (IPCC 2006). For aircraft emissions, the following values, in units of
15 grams of pollutant per kilogram of fuel consumed (g/kg), were employed: 0.1 forN2O (IPCC 2006). For marine
16 vessels consuming either distillate diesel or residual fuel oil the following values (g/MJ), were employed: 0.32 for
17 CH4 and 0.08 for N2O. Activity data for aviation included solely jet fuel consumption statistics, while the marine
18 mode included both distillate diesel and residual fuel oil.
19 Activity data on domestic and international aircraft fuel consumption were developed by the U.S. Federal Aviation
20 Administration (FAA) using radar-informed data from the FAA Enhanced Traffic Management System (ETMS) for
21 1990, 2000 through 2014 as modeled with the Aviation Environmental Design Tool (AEDT). This bottom-up
22 approach is built from modeling dynamic aircraft performance for each flight occurring within an individual
23 calendar year. The analysis incorporates data on the aircraft type, date, flight identifier, departure time, arrival time,
24 departure airport, arrival airport, ground delay at each airport, and real-world flight trajectories. To generate results
25 for a given flight within AEDT, the radar-informed aircraft data is correlated with engine and aircraft performance
26 data to calculate fuel burn and exhaust emissions. Information on exhaust emissions for in-production aircraft
27 engines comes from the International Civil Aviation Organization (ICAO) Aircraft Engine Emissions Databank
28 (EDB). This bottom-up approach is in accordance with the Tier 3B method from the 2006 IPCC Guidelines (IPCC
29 2006).
30 International aviation CO2 estimates for 1990 and 2000 through 2014 are obtained from FAA's AEDT model (FAA
31 2016). The radar-informed method that was used to estimate CO2 emissions for commercial aircraft for 1990, and
32 2000 through 2014 is not possible for 1991 through 1999 because the radar data set is not available for years prior to
33 2000. FAA developed OAG schedule-informed inventories modeled with AEDT and great circle trajectories for
34 1990, 2000 and 2010. Because fuel consumption and CO2 emission estimates for years 1991 through 1999 are
35 unavailable, consumption estimates for these years were calculated using fuel consumption estimates from the
36 Bureau of Transportation Statistics (DOT 1991 through2013), adjusted based on 2000 through 2005 data.
37 Data on U.S. Department of Defense (DoD) aviation bunker fuels and total jet fuel consumed by the U.S. military
38 was supplied by the Office of the Under Secretary of Defense (Installations and Environment), DoD. Estimates of
39 the percentage of each Service's total operations that were international operations were developed by DoD.
40 Military aviation bunkers included international operations, operations conducted from naval vessels at sea, and
3-80 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 operations conducted from U.S. installations principally over international water in direct support of military
2 operations at sea. Military aviation bunker fuel emissions were estimated using military fuel and operations data
3 synthesized from unpublished data from DoD's Defense Logistics Agency Energy (DLA Energy 2015). Together,
4 the data allow the quantity of fuel used in military international operations to be estimated. Densities for each jet
5 fuel type were obtained from a report from the U.S. Air Force (USAF 1998). Final jet fuel consumption estimates
6 are presented in Table 3-52. See Annex 3.8 for additional discussion of military data.
7 Activity data on distillate diesel and residual fuel oil consumption by cargo or passenger carrying marine vessels
8 departing from U.S. ports were taken from unpublished data collected by the Foreign Trade Division of the U.S.
9 Department of Commerce's Bureau of the Census (DOC 2015) for 1990 through 2001, 2007 through 2014, and the
10 Department of Homeland Security's Bunker Report for 2003 through 2006 (DHS 2008). Fuel consumption data for
11 2002 was interpolated due to inconsistencies in reported fuel consumption data. Activity data on distillate diesel
12 consumption by military vessels departing from U.S. ports were provided by DLA Energy (2015). The total amount
13 of fuel provided to naval vessels was reduced by 21 percent to account for fuel used while the vessels were not-
14 underway (i.e., in port). Data on the percentage of steaming hours underway versus not-underway were provided by
15 the U.S. Navy. These fuel consumption estimates are presented in. Table 3-53.
16 Table 3-52: Aviation Jet Fuel Consumption for International Transport (Million Gallons)
Nationality 1990 2005
U.S. and Foreign Carriers 3,222 5,983 1
U.S. Military 862 462
Total 4,084 6,445
2010
6,173
1 367
6,540
Note: Totals may not sum due to independent rounding.
17 Table 3-53: Marine Fuel Consumption for International
Fuel Type 1990 2005
Residual Fuel Oil 4,781 3,881
Distillate Diesel Fuel & Other 617 444
U.S. Military Naval Fuels 522 471
Total 5,920 4,796
2010
4,141
476
1 448
5,065
2011
6,634
319
6,953
2012
6,604
321
6,925
2013
6,748
294
7,042
2014
7,126
318
7,445
Transport (Million Gallons)
2011
3,463
393
382
4,237
2012
3,069
280
381
3,730
2013
2,537
235
308
3,081
2014
2,466
261
331
3,058
Note: Totals may not sum due to independent rounding.
18 Uncertainty and Time-Series Consistency
19 Emission estimates related to the consumption of international bunker fuels are subject to the same uncertainties as
20 those from domestic aviation and marine mobile combustion emissions; however, additional uncertainties result
21 from the difficulty in collecting accurate fuel consumption activity data for international transport activities separate
22 from domestic transport activities.79 For example, smaller aircraft on shorter routes often carry sufficient fuel to
23 complete several flight segments without refueling in order to minimize time spent at the airport gate or take
24 advantage of lower fuel prices at particular airports. This practice, called tankering, when done on international
25 flights, complicates the use of fuel sales data for estimating bunker fuel emissions. Tankering is less common with
26 the type of large, long-range aircraft that make many international flights from the United States, however. Similar
27 practices occur in the marine shipping industry where fuel costs represent a significant portion of overall operating
28 costs and fuel prices vary from port to port, leading to some tankering from ports with low fuel costs.
29 Uncertainties exist with regard to the total fuel used by military aircraft and ships, and in the activity data on military
30 operations and training that were used to estimate percentages of total fuel use reported as bunker fuel emissions.
31 Total aircraft and ship fuel use estimates were developed from DoD records, which document fuel sold to the Navy
32 and Air Force from the Defense Logistics Agency. These data may slightly over or under estimate actual total fuel
79 See uncertainty discussions under Carbon Dioxide Emissions from Fossil Fuel Combustion.
Energy 3-81
-------
1 use in aircraft and ships because each Service may have procured fuel from, and/or may have sold to, traded with,
2 and/or given fuel to other ships, aircraft, governments, or other entities. There are uncertainties in aircraft operations
3 and training activity data. Estimates for the quantity of fuel actually used in Navy and Air Force flying activities
4 reported as bunker fuel emissions had to be estimated based on a combination of available data and expert judgment.
5 Estimates of marine bunker fuel emissions were based on Navy vessel steaming hour data, which reports fuel used
6 while underway and fuel used while not underway. This approach does not capture some voyages that would be
7 classified as domestic for a commercial vessel. Conversely, emissions from fuel used while not underway preceding
8 an international voyage are reported as domestic rather than international as would be done for a commercial vessel.
9 There is uncertainty associated with ground fuel estimates for 1997 through 2001. Small fuel quantities may have
10 been used in vehicles or equipment other than that which was assumed for each fuel type.
11 There are also uncertainties in fuel end-uses by fuel-type, emissions factors, fuel densities, diesel fuel sulfur content,
12 aircraft and vessel engine characteristics and fuel efficiencies, and the methodology used to back-calculate the data
13 set to 1990 using the original set from 1995. The data were adjusted for trends in fuel use based on a closely
14 correlating, but not matching, data set. All assumptions used to develop the estimate were based on process
15 knowledge, Department and military Service data, and expert judgments. The magnitude of the potential errors
16 related to the various uncertainties has not been calculated, but is believed to be small. The uncertainties associated
17 with future military bunker fuel emission estimates could be reduced through additional data collection.
18 Although aggregate fuel consumption data have been used to estimate emissions from aviation, the recommended
19 method for estimating emissions of gases other than CO2 in the 2006IPCC Guidelines (IPCC 2006) is to use data by
20 specific aircraft type, number of individual flights and, ideally, movement data to better differentiate between
21 domestic and international aviation and to facilitate estimating the effects of changes in technologies. The IPCC also
22 recommends that cruise altitude emissions be estimated separately using fuel consumption data, while landing and
23 take-off (LTO) cycle data be used to estimate near-ground level emissions of gases other than CCh.80
24 There is also concern regarding the reliability of the existing DOC (2015) data on marine vessel fuel consumption
25 reported at U.S. customs stations due to the significant degree of inter-annual variation.
26 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
27 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
28 above.
29
35
QA/QC and Verification
30 A source-specific QA/QC plan for international bunker fuels was developed and implemented. This effort included
31 a Tier 1 analysis, as well as portions of a Tier 2 analysis. The Tier 2 procedures that were implemented involved
32 checks specifically focusing on the activity data and emission factor sources and methodology used for estimating
33 CO2, CH4, and N2O from international bunker fuels in the United States. Emission totals for the different sectors and
34 fuels were compared and trends were investigated. No corrective actions were necessary.
Planned Improvements
36 The feasibility of including data from a broader range of domestic and international sources for bunker fuels,
37 including data from studies such as the Third IMO GHG Study 2014, is being considered.
80 U.S. aviation emission estimates for CO, NOX, and NMVOCs are reported by EPA's National Emission Inventory (NET) Air
Pollutant Emission Trends web site, and reported under the Mobile Combustion section. It should be noted that these estimates
are based solely upon LTO cycles and consequently only capture near ground-level emissions, which are more relevant for air
quality evaluations. These estimates also include both domestic and international flights. Therefore, estimates reported under the
Mobile Combustion section overestimate IPCC-defined domestic CO, NOX, and NMVOC emissions by including landing and
take-off (LTO) cycles by aircraft on international flights, but underestimate because they do not include emissions from aircraft
on domestic flight segments at cruising altitudes. The estimates in Mobile Combustion are also likely to include emissions from
ocean-going vessels departing from U.S. ports on international voyages.
3-82 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
i 3.10 Wood Biomass and Ethanol
2 Consumption (IPCC Source Category 1A]
3 The combustion of biomass fuels such as wood, charcoal, and wood waste and biomass-based fuels such as ethanol
4 generates CCh in addition to CH4 and N2O already covered in this chapter. In line with the reporting requirements
5 for inventories submitted under the UNFCCC, CCh emissions from biomass combustion have been estimated
6 separately from fossil fuel CC>2 emissions and are not directly included in the energy sector contributions to U.S.
7 totals. In accordance with IPCC methodological guidelines, any such emissions are calculated by accounting for net
8 carbon (C) fluxes from changes in biogenic C reservoirs in wooded or crop lands. For a more complete description
9 of this methodological approach, see the Land Use, Land-Use Change, and Forestry chapter (Chapter 6), which
10 accounts for the contribution of any resulting CC>2 emissions to U. S. totals within the Land Use, Land-Use Change,
11 and Forestry sector's approach.
12 In 2014, total CC>2 emissions from the burning of woody biomass in the industrial, residential, commercial, and
13 electricity generation sectors were approximately 217.7 MMT CCh Eq. (217,654 kt) (see Table 3-54 and Table
14 3-55). As the largest consumer of woody biomass, the industrial sector was responsible for 57.1 percent of the CCh
15 emissions from this source. The residential sector was the second largest emitter, constituting 27.5 percent of the
16 total, while the commercial and electricity generation sectors accounted for the remainder.
17 Table 3-54: COz Emissions from Wood Consumption by End-Use Sector (MMT COz Eq.)
End-Use Sector
Industrial
Residential
Commercial
Electricity Generation
Total
1990
135.3
59.8
6.8 1
13.3
215.2
2005
136.3
44.3
7.2
19.1
206.9
2010
119.5
45.4
7.4
20.2
192.5
Note: Totals may not sum due to independent rounding.
18 Table 3-55: COz Emissions from Wood Consumption by
End-Use Sector
Industrial
Residential
Commercial
Electricity Generation
Total
1990 •
135,348
59,808
6,779
13,252
215,186
2005
136,269
44,340
7,218
19,074
206,901
2010
119,537
45,371
7,385
20,169
192,462
2011
122.9
46.4
7.1
18.8
195.2
End-Use
2011
122,865
46,402
7,131
18,784
195,182
2012
125.7
43.3
6.3
19.6
194.9
Sector
2012
125,724
43,309
6,257
19,612
194,903
2013
123.1
59.8
7.2
21.4
211.6
(kt)
2013
123,149
59,808
7,235
21,389
211,581
2014
124.4
59.8
7.6
25.9
217.7
2014
124,369
59,808
7,569
25,908
217,654
Note: Totals may not sum due to independent rounding.
19 The transportation sector is responsible for most of the ethanol consumption in the United States. Ethanol is
20 currently produced primarily from corn grown in the Midwest, but it can be produced from a variety of biomass
21 feedstocks. Most ethanol for transportation use is blended with gasoline to create a 90 percent gasoline, 10 percent
22 by volume ethanol blend known as E-10 or gasohol.
23 In 2014, the United States consumed an estimated 1,111.3 trillion Btu of ethanol, and as a result, produced
24 approximately 76.1 MMT CO2 Eq. (76,075 kt) (see Table 3-56 and Table 3-57) of CO2 emissions. Ethanol
25 production and consumption has grown significantly since 1990 due to the favorable economics of blending ethanol
26 into gasoline and federal policies that have encouraged use of renewable fuels.
Energy 3-83
-------
Table 3-56: COz Emissions from Ethanol Consumption (MMT COz Eq.)
End-Use Sector
Transportation
Industrial
Commercial
Total
1990
1
+
4.2
2005
22.4
0.5 1
0.1
22.9
2010
71.3
1.1
0.2
72.6
2011
71.5
1.1
0.2
72.9
2012
71.5
1.1
0.2
72.8
2013
73.4
1.2
0.2
74.7
2014
74.7
1.2
0.2
76.1
+ Does not exceed 0.05 MMT CO2 Eq.
Note: Totals may not sum due to independent rounding.
Table 3-57: COz Emissions from Ethanol Consumption (kt)
End-Use Sector
Transportation*
Industrial
Commercial
Total
1990
4,136
56
34
4,227
2005
22,414 1
468 1
60
22,943
2010
71,287
1,134
226
72,647
2011
71,537
1,146
198
72,881
2012
71,510
1,142
175
72,827
2013
73,354
1,206
183
74,743
2014
74,661
1,227
186
76,075
a See Annex 3.2, Table A-92 for additional information on transportation consumption of these fuels.
Note: Totals may not sum due to independent rounding.
Methodology
4 Woody biomass emissions were estimated by applying two EIA gross heat contents (Lindstrom 2006) to U.S.
5 consumption data (EIA 2015) (see Table 3-58), provided in energy units for the industrial, residential, commercial,
6 and electric generation sectors. One heat content (16.95 MMBtu/MT wood and wood waste) was applied to the
7 industrial sector's consumption, while the other heat content (15.43 MMBtu/MT wood and wood waste) was applied
8 to the consumption data for the other sectors. An EIA emission factor of 0.434 MT C/MT wood (Lindstrom 2006)
9 was then applied to the resulting quantities of woody biomass to obtain CCh emission estimates. It was assumed
10 that the woody biomass contains black liquor and other wood wastes, has a moisture content of 12 percent, and is
11 converted into CC>2 with 100 percent efficiency. The emissions from ethanol consumption were calculated by
12 applying an emission factor of 18.67 MMT C/QBtu (EPA 2010) to U.S. ethanol consumption estimates that were
13 provided in energy units (EIA 2015) (see Table 3-59).
14 Table 3-58: Woody Biomass Consumption by Sector (Trillion Btu)
End-Use Sector
Industrial
Residential
Commercial
Electricity Generation
Total
1990
1,441.9 1
580.0
65.7
128.5 H
2,216.2
2005
1,451.7 1
430.0
70.0
185.0 1
2,136.7
1 2010
1,273.5
440.0
71.6
195.6
1,980.7
2011
1,308.9
450.0
69.2
182.2
2,010.2
2012
1,339.4
420.0
60.7
190.2
2,010.3
2013
1,312.0
580.0
70.2
207.4
2,169.5
2014
1,325.0
580.0
73.4
251.3
2,229.6
Note: Totals may not sum due to independent rounding.
3-84 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 3-59: Ethanol Consumption by Sector (Trillion Btu)
13
End-Use Sector
Transportation
Industrial
Commercial
Total
1990
60.4
0.8 1
0.5
61.7
2005
327.4
6.8 1
0.9 1
335.1
2010
1,041.4
16.6
3.3
1,061.2
2011
1,045.0
16.7
2.9
1,064.6
2012
1,044.6
16.7
2.6
1,063.8
2013
1,071.5
17.6
2.7
1,091.8
2014
1,090.6
17.9
2.7
1,111.3
Note: Totals may not sum due to independent rounding.
Uncertainty and Time-Series Consistency
3 It is assumed that the combustion efficiency for woody biomass is 100 percent, which is believed to be an
4 overestimate of the efficiency of wood combustion processes in the United States. Decreasing the combustion
5 efficiency would decrease emission estimates. Additionally, the heat content applied to the consumption of woody
6 biomass in the residential, commercial, and electric power sectors is unlikely to be a completely accurate
7 representation of the heat content for all the different types of woody biomass consumed within these sectors.
8 Emission estimates from ethanol production are more certain than estimates from woody biomass consumption due
9 to better activity data collection methods and uniform combustion techniques.
10
11
12 above.
Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
ahnvp
Recalculations Discussion
14 Wood consumption values for 2013 were revised relative to the previous Inventory based on updated information
15 fromEIA's Monthly Energy Review (EIA2015). These revisions of historical data for wood biomass consumption
16 resulted in an average annual increase in emissions from wood biomass consumption of 0.1 MMT CCh Eq. (less
17 than 0.1 percent) from 1990 through 2013. Ethanol consumption values remained constant relative to the previous
18 Inventory throughout the entire time-series.
19 Planned Improvements
20 The availability of facility-level combustion emissions through EPA's GHGRP will be examined to help better
21 characterize the industrial sector's energy consumption in the United States, and further classify business
22 establishments according to industrial economic activity type. Most methodologies used in EPA's GHGRP are
23 consistent with IPCC, though for EPA's GHGRP, facilities collect detailed information specific to their operations
24 according to detailed measurement standards, which may differ with the more aggregated data collected for the
25 Inventory to estimate total, national U.S. emissions. In addition, and unlike the reporting requirements for this
26 chapter under the UNFCCC reporting guidelines, some facility-level fuel combustion emissions reported under the
27 GHGRP may also include industrial process emissions.81 In line with UNFCCC reporting guidelines, fuel
28 combustion emissions are included in this chapter, while process emissions are included in the Industrial Processes
29 and Product Use chapter of this report. In examining data from EPA's GHGRP that would be useful to improve the
30 emission estimates for the CC>2 from biomass combustion category, particular attention will also be made to ensure
31 time series consistency, as the facility-level reporting data from EPA's GHGRP are not available for all inventory
32 years as reported in this Inventory. Additionally, analyses will focus on aligning reported facility-level fuel types
33 and IPCC fuel types per the national energy statistics, ensuring CO2 emissions from biomass are separated in the
34 facility-level reported data, and maintaining consistency with national energy statistics provided by EIA. In
35 implementing improvements and integration of data from EPA's GHGRP, the latest guidance from the IPCC on the
36 use of facility-level data in national inventories will be relied upon.82
81 See .
82 See.
Energy 3-85
-------
-------
i 4. Industrial Processes and Product Use
2 The Industrial Processes and Product Use (IPPU) chapter includes greenhouse gas emissions occurring from
3 industrial processes and from the use of greenhouse gases in products. The industrial processes and product use
4 categories included in this chapter are presented in Figure 4-1.
5 Greenhouse gas emissions are produced as the byproducts of various non-energy-related industrial activities. That
6 is, these emissions are produced either from an industrial process itself, and are not directly a result of energy
7 consumed during the process. For example, raw materials can be chemically or physically transformed from one
8 state to another. This transformation can result in the release of greenhouse gases such as carbon dioxide (CO2),
9 methane (CH4), and nitrous oxide (N2O). The processes included in this chapter include iron and steel production
10 and metallurgical coke production, cement production, lime production, other process uses of carbonates (e.g., flux
11 stone, flue gas desulfurization, and glass manufacturing), ammonia production and urea consumption, petrochemical
12 production, aluminum production, soda ash production and use, titanium dioxide production, CO2 consumption,
13 ferroalloy production, glass production, zinc production, phosphoric acid production, lead production, silicon
14 carbide production and consumption, nitric acid production, and adipic acid production.
15 In addition, greenhouse gases are often used in products or by end-consumers. These gases include industrial
16 sources of man-made compounds such as hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), sulfur hexafluoride
17 (SF6), nitrogen trifluoride (NF3), and nitrous oxide (N2O). The present contribution of HFCs, PFCs, SF6, and NF3
18 gases to the radiative forcing effect of all anthropogenic greenhouse gases is small; however, because of their
19 extremely long lifetimes, many of them will continue to accumulate in the atmosphere as long as emissions
20 continue. In addition, many of these gases have high global warming potentials; SF6 is the most potent greenhouse
21 gas the Intergovernmental Panel on Climate Change (IPCC) has evaluated. Use of HFCs is growing rapidly since
22 they are the primary substitutes for ozone depleting substances (ODSs), which are being phased-out under the
23 Montreal Protocol on Substances that Deplete the Ozone Layer. HFCs, PFCs, SF6, and NF3 are employed and
24 emitted by a number of other industrial sources in the United States such as aluminum production, HCFC-22
25 production, semiconductor manufacture, electric power transmission and distribution, and magnesium metal
26 production and processing. N2O is emitted by the production of adipic acid and nitric acid, semiconductor
27 manufacturing, end-consumers in product uses through the administration of anesthetics, and by industry as a
28 propellant in aerosol products.
29 In 2014, IPPU generated emissions of 388.6 million metric tons of CO2 equivalent (MMT CO2 Eq.), or 5.7 percent
30 of total U.S. greenhouse gas emissions. Carbon dioxide emissions from all industrial processes were 178.6 MMT
31 CO2 Eq. (178,572 kt CO2) in 2014, or 3.2 percent of total U.S. CO2 emissions. Methane emissions from industrial
32 processes resulted in emissions of approximately 0.2 MMT CO2 Eq. (32 kt CH4) in 2014, which was less than 1
33 percent of U.S. CH4 emissions. N2O emissions from IPPU were 20.8 MMT CO2 Eq. (70 kt N2O) in 2014, or 5.1
34 percent of total U.S. N2O emissions. In 2014 combined emissions of HFCs, PFCs, SF6, and NF3 totaled 189.1 MMT
35 CO2 Eq. Total emissions from IPPU in 2014 were 14.0 percent more than 1990 emissions. Indirect greenhouse gas
36 emissions also result from IPPU, and are presented in Table 4-108 in kilotons (kt).
Industrial Processes and Product Use 4-1
-------
1 Figure 4-1: 2014 Industrial Processes and Product Use Chapter Greenhouse Gas Sources
2 (MMT COz Eq.)
Substitution of Ozone Depleting Substances
Iron and Steel Prod. & Metallurgical Coke Prod.
Cement Production
Petrochemical Production
Lime Production
Other Process Uses of Carbonates
Nitric Acid Production
Ammonia Production
AdipicAcid Production
Electrical Transmission and Distribution
Carbon Dioxide Consumption
N2Ofrom Product Uses
Semiconductor Manufacture
HCFC-22 Production
Urea Consumption for Non-Agricultural Purposes
Soda Ash Production and Consumption
Ferroalloy Production
Titanium Dioxide Production
Magnesium Production and Processing
Glass Production
Phosphoric Acid Production
Zinc Production
Lead Production
Silicon Carbide Production and Consumption
Industrial Processes and Product
Use as a Portion of all Emissions
5.7%
171
10
20
30 40
MMT CO2 Eq.
50
60
70
4 The increase in overall IPPU emissions since 1990 reflects a range of emission trends among the emission sources.
5 Emissions resulting from most types of metal production have declined significantly since 1990, largely due to
6 production shifting to other countries, but also due to transitions to less-emissive methods of production (in the case
7 of iron and steel) and to improved practices (in the case of PFC emissions from aluminum production). Emissions
8 from mineral sources have either increased or not changed significantly since 1990 but largely track economic
9 cycles, while CO2 and CH4 emissions from chemical sources have either decreased or not changed significantly.
10 HFC emissions from the substitution of ozone depleting substances have increased drastically since 1990, while the
11 emission trends of HFCs, PFCs, SF6, and NF3 from other sources are mixed. N2O emissions from the production of
12 adipic and nitric acid have decreased, while N2O emissions from product uses has remained nearly constant over
13 time. Trends are explained further within each emission source category throughout the chapter.
14 Table 4-1 summarizes emissions for the IPPU chapter in MMT CO2 Eq. using IPCC Fourth Assessment Report
15 (AR4) GWP values, following the requirements of the revised UNFCCC reporting guidelines for national
16 inventories (IPCC 2007).l Unweighted native gas emissions in kt are also provided in Table 4-2. The source
17 descriptions that follow in the chapter are presented in the order as reported to the UNFCCC in the common
1 See .
4-2 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 reporting format tables, corresponding generally to: mineral products, chemical production, metal production, and
2 emissions from the uses of HFCs, PFCs, SF6, and NF3.
3 Table 4-1: Emissions from Industrial Processes and Product Use (MMT COz Eq.)
Gas/Source 1990 2005 2010 2011 2012 2013 2014
C02 207.1 190.3 168.8 172.9 169.5 171.7 178.6
Iron and Steel Production &
Metallurgical Coke Production 99.7 66.5 55.7 59.9 54.2 52.2 55.4
Iron and Steel Production 97.2 64.5 53.6 58.5 53.7 50.4 53.4
Metallurgical Coke Production 2.5 2.0 2.1 1.4 0.5 1.8 1.9
Cement Production 33.3 45.9 31.3 32.0 35.1 36.1 38.8
Petrochemical Production 21.6 27.4 27.2 26.3 26.5 26.4 26.5
Lime Production 11.7 14.6 13.4 14.0 13.7 14.0 14.1
Other Process Uses of Carbonates 4.9 6.3 9.6 9.3 8.0 10.4 12.1
Ammonia Production 13.0 9.2 9.2 9.3 9.4 10.0 9.4
Carbon Dioxide Consumption 1.5 1.4 4.4 4.1 4.0 4.2 4.5
Urea Consumption for Non-
Agricultural Purposes 3.8 3.7 4.7 4.0 4.4 4.2 4.0
Aluminum Production 6.8 4.1 2.7 3.3 3.4 3.3 3.3
Soda Ash Production and
Consumption 2.8 3.0 2.7 2.7 2.8 2.8 2.8
Ferroalloy Production 2.2 1.4 1.7 1.7 1.9 1.8 1.9
Titanium Dioxide Production 1.2 1.8 1.8 1.7 1.5 1.7 1.8
Glass Production 1.5 1.9 1.5 1.3 1.2 1.3 1.3
Phosphoric Acid Production 1.5 1.3 1.1 1.2 1.1 1.1 1.1
Zinc Production 0.6 1.0 1.2 1.3 1.5 1.4 1.0
Lead Production 0.5 0.6 0.5 0.5 0.5 0.5 0.5
Silicon Carbide Production and
Consumption 0.4 0.2 0.2 0.2 0.2 0.2 0.2
Magnesium Production and
Processing +1 +1 + + + + +
CH4 0.3 0.1 0.1 0.1 0.1 0.1 0.2
Petrochemical Production 0.2 0.1 + + 0.1 0.1 0.1
Ferroalloy Production
Silicon Carbide Production and
Consumption + | + | + +
Iron and Steel Production &
Metallurgical Coke Production + | + | + +
Iron and Steel Production
Metallurgical Coke Production 0.0 0.0 0.0 0.0 0.0 0.0 0.0
N2O 31.6 22.8 20.1 25.5 20.4 19.1 20.8
Nitric Acid Production 12.1 11.3 11.5 10.9 10.5 10.7 10.9
Adipic Acid Production 15.2 7.1 4.2 10.2 5.5 4.0 5.4
N20 from Product Uses 4.2 4.2 4.2 4.2 4.2 4.2 4.2
Semiconductor Manufacturing + I 0.1 0.1 0.2 0.2 0.2 0.2
HFCs 46.6 133.3 I 161.7 166.1 167.1 169.6 175.8
Substitution of Ozone Depleting
Substances* 0.3 113.0 153.5 157.1 161.4 165.3 171.4
HCFC-22 Production 46.1 20.0 8.0 8.8 5.5 4.1 4.1
Semiconductor Manufacture 0.2 0.2 0.2 0.2 0.2 0.2 0.2
Magnesium Production and
Processing 0.0 0.0 + + + 0.1 0.1
PFCs 24.3 6.6 4.4 6.9 6.0 5.8 5.8
Aluminum Production 21.5 3.4 1.9 3.5 2.9 3.0 3.0
Semiconductor Manufacture 2.8 3.2 2.6 3.4 3.0 2.9 2.9
SF6 31.1 14.0 9.5 10.0 7.7 6.9 6.9
Industrial Processes and Product Use 4-3
-------
Electrical Transmission and
Distribution
Magnesium Production and
Processing
Semiconductor Manufacture
NF3
Semiconductor Manufacture
25.4
10.6
2.7
0.7
0.5
0.5
7.0
2.1
0.4
0.5
0.5
0.7
0.7
5.7
1.6
0.4
0.6
0.6
5.1
1.4
0.4
0.6
0.6
5.1
1.4
0.4
0.6
0.6
Total
340.9
367.6
365.2
382.2
371.4
373.8
388.6
Note: Totals may not sum due to independent rounding.
+ Does not exceed 0.05 MMT CO2 Eq.
a Small amounts of PFC emissions also result from this source.
Table 4-2: Emissions from Industrial Processes and Product Use (kt)
Gas/Source
1990
2005
2010
2011
2012
2013
2014
CO2 207,054
Iron and Steel Production &
Metallurgical Coke Production 99,669
Iron and Steel Production 97,166
Metallurgical Coke Production 2,503
Cement Production 3 3,278
Petrochemical Production 21,609
Lime Production 11,700
Other Process Uses of Carbonates 4,907
Ammonia Production 13,047
Carbon Dioxide Consumption 1,472
Urea Consumption for Non-
Agricultural Purposes 3,784
Aluminum Production 6,831
Soda Ash Production and
Consumption 2,822
Ferroalloy Production 2,152
Titanium Dioxide Production 1,195
Glass Production 1,535
Phosphoric Acid Production 1,529
Zinc Production 632
Lead Production 516
Silicon Carbide Production and
Consumption 375
Magnesium Production and
Processing 1
CH4 12
Petrochemical Production 9
Ferroalloy Production 1
Silicon Carbide Production and
Consumption 1
Iron and Steel Production &
Metallurgical Coke Production 1
Iron and Steel Production 1
Metallurgical Coke Production 0
N2O 106
Nitric Acid Production 41
Adipic Acid Production 51
N2O from Product Uses 14
Semiconductor Manufacturing +
HFCs M
Substitution of Ozone Depleting
Substances* M
190,273
66,543
64,499
2,044
45,910
27,380
14,552
6,339
9,196
1,375
3,653
4,142
2,960
1,392
1,755
1,928
1,342
1,030
553
219
3
4
3
1
1
0
76
38
24
14
+
M
M
168,781 172,898 169,472 171,714 178,572
55,671
53,586
2,085
31,256
27,246
13,381
9,560
9,188
4,425
4,730
2,722
2,697
1,663
1,769
1,481
1,087
1,182
542
59,928
58,501
1,426
32,010
26,326
13,981
9,335
9,292
4,083
4,029
3,292
2,712
1,735
1,729
1,299
1,151
1,286
538
54,229
53,686
543
35,051
26,464
13,715
8,022
9,377
4,019
4,449
3,439
2,763
1,903
1,528
1,248
1,093
1,486
527
52,201
50,378
1,824
36,146
26,437
14,045
10,414
9,962
4,188
4,179
3,255
2,804
1,785
1,715
1,317
1,119
1,429
546
55,355
53,417
1,938
38,755
26,509
14,125
12,077
9,436
4,471
4,007
3,255
2,827
1,914
1,755
1,341
1,095
956
518
181
1
3
2
0
68
39
14
14
+
M
M
170
3
3
2
0
86
37
34
14
1
M
M
158
2
4
3
1
0
69
35
19
14
1
M
M
169
2
4
3
0
64
36
13
14
1
M
M
173
2
6
5
1
0
70
37
18
14
1
M
M
4-4 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
HCFC-22 Production 3 I 1 I 1 1 + + +
Semiconductor Manufacture + I + I + + + + +
Magnesium Production and
Processing 0 I 0 I + + + + +
PFCs M M M M M M M
Aluminum Production M M M M M M M
Semiconductor Manufacture M M M M M M M
SF6 2J1J+1 + + +
Electrical Transmission and
Distribution I I + I + + + + +
Magnesium Production and
Processing + I + I + + + + +
Semiconductor Manufacture + I + I + + + + +
NF3 + I + I + + + + +
Semiconductor Manufacture + I + I + + + + +
+ Does not exceed 0.5 kt.
M (Mixture of gases)
Note: Totals may not sum due to independent rounding.
a Small amounts of PFC emissions also result from this source.
1
2 The UNFCCC incorporated the 2006IPCC Guidelines for National Greenhouse Gas Inventories (2006IPCC
3 Guidelines) as the standard for Annex I countries at the Nineteenth Conference of the Parties (Warsaw, November
4 11-23, 2013). This chapter presents emission estimates calculated in accordance with the methodological guidance
5 provided in these guidelines.
QA/QC and Verification Procedures
7 For industrial processes and product use sources, a detailed QA/QC plan was developed and implemented. This plan
8 was based on the overall U.S. QA/QC plan, but was tailored to include specific procedures recommended for these
9 sources. Two types of checks were performed using this plan: (1) general, or Tier 1, procedures that focus on annual
10 procedures and checks to be used when gathering, maintaining, handling, documenting, checking, and archiving the
11 data, supporting documents, and files, and (2) source-category specific, or Tier 2, procedures that focus on checks of
12 the emission factors, activity data, and methodologies used for estimating emissions from the relevant industrial
13 process and product use sources. Examples of these procedures include checks to ensure that activity data and
14 emission estimates are consistent with historical trends; that, where possible, consistent and reputable data sources
15 are used across sources; that interpolation or extrapolation techniques are consistent across sources; and that
16 common datasets and factors are used where applicable. Tier 1 quality assurance and quality control procedures
17 have been performed for all industrial process and product use sources. Tier 2 procedures were performed for more
18 significant emission categories, consistent with the IPCC Good Practice Guidelines.
19 For most industrial process and product use categories, activity data is obtained through a survey of manufacturers
20 conducted by various organizations (specified within each source); the uncertainty of the activity data is a function
21 of the reliability of reported plant-level production data and is influenced by the completeness of the survey
22 response. The emission factors used are defaults from IPCC, derived using calculations that assume precise and
23 efficient chemical reactions, or were based upon empirical data in published references. As a result, uncertainties in
24 the emission coefficients can be attributed to, among other things, inefficiencies in the chemical reactions associated
25 with each production process or to the use of empirically-derived emission factors that are biased; therefore, they
26 may not represent U.S. national averages. Additional assumptions are described within each source.
27 The uncertainty analysis performed to quantify uncertainties associated with the 2014 emission estimates from
28 industrial processes and product use continues a multi-year process for developing credible quantitative uncertainty
29 estimates for these source categories using the IPCC Tier 2 approach. As the process continues, the type and the
30 characteristics of the actual probability density functions underlying the input variables are identified and better
31 characterized (resulting in development of more reliable inputs for the model, including accurate characterization of
32 correlation between variables), based primarily on expert judgment. Accordingly, the quantitative uncertainty
33 estimates reported in this section should be considered illustrative and as iterations of ongoing efforts to produce
34 accurate uncertainty estimates. The correlation among data used for estimating emissions for different sources can
Industrial Processes and Product Use 4-5
-------
1 influence the uncertainty analysis of each individual source. While the uncertainty analysis recognizes very
2 significant connections among sources, a more comprehensive approach that accounts for all linkages will be
3 identified as the uncertainty analysis moves forward.
Box 4-1: Industrial Processes Data from EPA's Greenhouse Gas Reporting Program
5 On October 30, 2009, the U. S. EPA published a rule requiring annual of greenhouse gas data from large GHG
6 emissions sources in the United States. Implementation of the rule, codified at 40 CFR part 98, is referred to as
7 EPA's Greenhouse Gas Reporting Program (GHGRP). The rule applies to direct greenhouse gas emitters, fossil fuel
8 suppliers, industrial gas suppliers, and facilities that inject CO2 underground for sequestration or other reasons and
9 requires reporting by sources or suppliers in 41 industrial categories. Annual reporting is at the facility level, except
10 for certain suppliers of fossil fuels and industrial greenhouse gases. In general, the threshold for reporting is 25,000
11 metric tons or more of CO2 Eq. per year, but reporting is required for all facilities in some industries. Calendar year
12 2010 was the first year for which data were reported for facilities subject to 40 CFR part 98, though some source
13 categories first reported data for calendar year 2011.
14 EPA's GHGRP dataset and the data presented in this Inventory report are complementary. EPA presents the data
15 collected by EPA's GHGRP through a data publication tool (ghgdata.epa.gov) that allows data to be viewed in
16 several formats, including maps, tables, charts, and graphs for individual facilities or groups of facilities. Most
17 methodologies used in EPA's GHGRP are consistent with IPCC, though for EPA's GHGRP, facilities collect
18 detailed information specific to their operations according to detailed measurement standards. This may differ from
19 the more aggregated data collected for the Inventory to estimate total, national U.S. emissions. It should be noted
20 that the definitions for source categories in the GHGRP may differ from those used in this Inventory in meeting the
21 UNFCCC reporting guidelines (IPCC 2011). In line with the UNFCCC reporting guidelines, the Inventory report is
22 a comprehensive accounting of all emissions from source categories identified in the IPCC guidelines. Further
23 information on the reporting categorizations in EPA's GHGRP and specific data caveats associated with monitoring
24 methods in EPA's GHGRP has been provided on the EPA's GHGRP website.
25 For certain source categories in this Inventory (e.g., nitric acid production and petrochemical production), EPA has
26 also integrated data values that have been calculated by aggregating GHGRP data that is considered confidential
27 business information (CBI) at the facility level. EPA, with industry engagement, has put forth criteria to confirm
28 that a given data aggregation shields underlying CBI from public disclosure. EPA is publishing only data values that
29 meet these aggregation criteria.2 Specific uses of aggregated facility-level data are described in the respective
30 methodological sections. For other source categories in this chapter, as indicated in the respective planned
31 improvements sections, EPA is continuing to analyze how facility-level GHGRP data may be used to improve the
32 national estimates presented in this Inventory, giving particular consideration to ensuring time series consistency and
33 completeness.
34
35
36
4.1 Cement Production (IPCC Source Category
2A1)
37 Cement production is an energy- and raw material-intensive process that results in the generation of CO2 from both
38 the energy consumed in making the cement and the chemical process itself. Emissions from fuels consumed for
39 energy purposes during the production of cement are accounted for in the Energy chapter.
40 During the cement production process, calcium carbonate (CaCOs) is heated in a cement kiln at a temperature of
41 about 1,450°C (2,700°F) to form lime (i.e., calcium oxide or CaO) and CO2 in a process known as calcination or
2 U.S. EPA Greenhouse Gas Reporting Program. Developments on Publication of Aggregated Greenhouse Gas Data, November
25, 2014. See
4-6 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 calcining. The quantity of CO2 emitted during cement production is directly proportional to the lime content of the
2 clinker. During calcination, each mole of limestone (CaCOs) heated in the clinker kiln forms one mole of lime
3 (CaO) and one mole of CO2:
4 CaC03 + heat -> CaO + C02
5 Next, the lime is combined with silica-containing materials to produce clinker (an intermediate product), with the
6 earlier byproduct CO2 being released to the atmosphere. The clinker is then allowed to cool, mixed with a small
7 amount of gypsum and potentially other materials (e.g., slag, etc.), and used to make Portland cement.3
8 CO2 emitted from the chemical process of cement production is the second largest source of industrial CO2
9 emissions in the United States. Cement is produced in 34 states and Puerto Rico. Texas, California, Missouri,
10 Florida, and Michigan were the five leading cement-producing States in 2014 and accounted for approximately 53
11 percent of total U.S. production (USGS 2015b). Clinker production in 2014 increased approximately 7 percent from
12 2013 levels. This increase can be attributed to an increase in spending in new residential construction and
13 nonresidential buildings. In 2014, U.S. clinker production totaled 74,946 kilotons (USGS 2015a). The resulting CO2
14 emissions were estimated to be 38.8 MMT CO2 Eq. (38,755 kt) (see Table 4-3).
15 Table 4-3: COz Emissions from Cement Production (MMT COz Eq. and kt)
Year MMT CCh Eq. kt
1990 33.3 33,278
2010
2011
2012
2013
2014
31.3
32.0
35.1
36.1
38.8
31,256
32,010
35,051
36,146
38,755
16 Greenhouse gas emissions from cement production increased every year from 1991 through 2006 (with the
17 exception of a slight decrease in 1997), but decreased in the following years until 2009. Emissions from cement
18 production were at their lowest levels in 2009 (2009 emissions are approximately 28 percent lower than 2008
19 emissions and 12 percent lower than 1990). Since 2010, emissions have increased slightly. In 2014, emissions from
20 cement production increased by 7 percent from 2013 levels.
21 Emissions since 1990 have increased by 16 percent. Emissions decreased significantly between 2008 and 2009, due
22 to the economic recession and associated decrease in demand for construction materials. Emissions increased
23 slightly from 2009 levels in 2010, and continued to gradually increase during the 2011 through 2014 time period due
24 to increasing consumption. Cement continues to be a critical component of the construction industry; therefore, the
25 availability of public and private construction funding, as well as overall economic conditions, have considerable
26 impact on the level of cement production.
27 Methodology
28 CO2 emissions were estimated using the Tier 2 methodology from the 2006IPCC Guidelines. The Tier 2
29 methodology was used because detailed and complete data (including weights and composition) for carbonate(s)
30 consumed in clinker production are not available, and thus a rigorous Tier 3 approach is impractical. Tier 2 specifies
31 the use of aggregated plant or national clinker production data and an emission factor, which is the product of the
32 average lime fraction for clinker of 65 percent and a constant reflecting the mass of CO2 released per unit of lime.
3 Approximately three percent of total clinker production is used to produce masonry cement, which is produced using
plasticizers (e.g., ground limestone, lime, etc.) and Portland cement (USGS 2011). Carbon dioxide emissions that result from the
production of lime used to create masonry cement are included in the Lime Manufacture source category.
Industrial Processes and Product Use 4-7
-------
1 The USGS mineral commodity expert for cement has confirmed that this is a reasonable assumption for the United
2 States (VanOss2013a). This calculation yields an emission factor of 0.51 tons of CCh per ton of clinker produced,
3 which was determined as follows:
4 EFciinker = 0.6460 CaO x [(44.01 g/mole C02) H- (56.08 g/mole CaO)] = 0.5070 tons C02/ton clinker
5 During clinker production, some of the clinker precursor materials remain in the kiln as non-calcinated, partially
6 calcinated, or fully calcinated cement kiln dust (CKD). The emissions attributable to the calcinated portion of the
7 CKD are not accounted for by the clinker emission factor. The IPCC recommends that these additional CKD CO2
8 emissions should be estimated as two percent of the CCh emissions calculated from clinker production (when data
9 on CKD generation are not available). Total cement production emissions were calculated by adding the emissions
10 from clinker production to the emissions assigned to CKD (IPCC 2006).
11 Furthermore, small amounts of impurities (i.e., not calcium carbonate) may exist in the raw limestone used to
12 produce clinker. The proportion of these impurities is generally minimal, although a small amount (1 to 2 percent)
13 of magnesium oxide (MgO) may be desirable as a flux. Per the IPCC Tier 2 methodology, a correction for
14 magnesium oxide is not used, since the amount of magnesium oxide from carbonate is likely very small and the
15 assumption of a 100 percent carbonate source of CaO already yields an overestimation of emissions (IPCC 2006).
16 The 1990 through 2012 activity data for clinker production (see Table 4-4) were obtained from USGS (Van Oss
17 2013b). Clinker production data for 2013 and 2014 were also obtained from USGS (USGS 2015a). The data were
18 compiled by USGS (to the nearest ton) through questionnaires sent to domestic clinker and cement manufacturing
19 plants, including the facilities in Puerto Rico.
20 Table 4-4: Clinker Production (kt)
21
Year Clinker
22
1990 64,355
2010 60,444
2011 61,903
2012 67,784
2013 69,901
2Q14 74,946
Note: Clinker production from 1990-2014 includes Puerto Rico. Data were obtained from USGS (Van Oss 2013a; USGS
2014), whose original data source was USGS and U.S. Bureau of Mines Minerals Yearbooks (2013 data obtained from
mineral industry surveys for cement in June 2014; 2014 data obtained from mineral industry surveys for cement in June
2015).
Uncertainty and Time-Series Consistency
23 The uncertainties contained in these estimates are primarily due to uncertainties in the lime content of clinker and in
24 the percentage of CKD recycled inside the cement kiln. Uncertainty is also associated with the assumption that all
25 calcium-containing raw materials are CaCOs, when a small percentage likely consists of other carbonate and non-
26 carbonate raw materials. The lime content of clinker varies from 60 to 67 percent; 65 percent is used as a
27 representative value (Van Oss 2013a). CKD loss can range from 1.5 to 8 percent depending upon plant
28 specifications. Additionally, some amount of CO2 is reabsorbed when the cement is used for construction. As
29 cement reacts with water, alkaline substances such as calcium hydroxide are formed. During this curing process,
30 these compounds may react with CO2 in the atmosphere to create calcium carbonate. This reaction only occurs in
31 roughly the outer 0.2 inches of surface area. Because the amount of CO2 reabsorbed is thought to be minimal, it was
32 not estimated.
33 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-5. Based on the
34 uncertainties associated with total U.S. clinker production, the CO2 emission factor for clinker production, and the
4-8 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 emission factor for additional CO2 emissions from CKD, 2014 CO2 emissions from cement production were
2 estimated to be between 36.5 and 41.1 MMT CO2 Eq. at the 95 percent confidence level. This confidence level
3 indicates a range of approximately 6 percent below and 6 percent above the emission estimate of 38.8 MMT CO2
4 Eq.
5 Table 4-5: Approach 2 Quantitative Uncertainty Estimates for COz Emissions from Cement
6 Production (MMT COz Eq. and Percent)
^ „ 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
°UrCe ^ (MMT CCh Eq.) (MMT CCh Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
Cement Production CCh 38.8 36.5 41.1 -6% +6%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
16
17
Planned Improvements
8 Future improvements involve continuing to evaluate and analyze data reported under EPA's GHGRP that would be
9 useful to improve the emission estimates for the Cement Production source category. Particular attention will be
10 made to ensure time series consistency of the emissions estimates presented in future Inventory reports, consistent
11 with IPCC and UNFCCC guidelines. This is required as facility-level reporting data from EPA's GHGRP, with the
12 program's initial requirements for reporting of emissions in calendar year 2010, are not available for all inventory
13 years (i.e., 1990 through 2009) as required for this Inventory. In implementing improvements and integration of data
14 from EPA's GHGRP, the latest guidance from the IPCC on the use of facility-level data in national inventories will
15 be relied upon.4
4.2 Lime Production (IPCC Source Category
2A2)
18 Lime is an important manufactured product with many industrial, chemical, and environmental applications. Lime
19 production involves three main processes: stone preparation, calcination, and hydration. Carbon dioxide is
20 generated during the calcination stage, when limestone—mostly calcium carbonate (CaCOs)—is roasted at high
21 temperatures in a kiln to produce CaO and CO2. The CO2 is given off as a gas and is normally emitted to the
22 atmosphere.
23 CaC03 -> CaO + C02
24 Some of the CO2 generated during the production process, however, is recovered at some facilities for use in sugar
25 refining and precipitated calcium carbonate (PCC) production.5 Emissions from fuels consumed for energy purposes
26 during the production of lime are accounted for in the Energy chapter.
27 For U.S. operations, the term "lime" actually refers to a variety of chemical compounds. These include calcium
28 oxide (CaO), or high-calcium quicklime; calcium hydroxide (Ca(OH)2), or hydrated lime; dolomitic quicklime
29 ([CaOMgO]); and dolomitic hydrate ([Ca(OH)2«MgO] or [Ca(OH)2«Mg(OH)2]).
30 The current lime market is approximately distributed across five end-use categories as follows: metallurgical uses,
31 38 percent; environmental uses, 31 percent; chemical and industrial uses, 22 percent; construction uses, 8 percent;
4 See
PCC is obtained from the reaction of CCh with calcium hydroxide. It is used as a filler and/or coating in the paper, food, and
plastic industries.
Industrial Processes and Product Use 4-9
-------
1 and refractory dolomite, 1 percent. The major uses are in steel making, flue gas desulfurization systems at coal-fired
2 electric power plants, construction, and water purification. Lime is also used as a CCh scrubber, and there has been
3 experimentation on the use of lime to capture €62 from electric power plants.
4 Lime production in the United States—including Puerto Rico— was reported to be 19,399 kilotons in 2014
5 (Corathers 2015). Principal lime producing states are Missouri, Alabama, Kentucky, Ohio, Texas (USGS 2014),
6 Nevada, and Pennsylvania.
7 U.S. lime production resulted in estimated net CC>2 emissions of 14.1 MMT CCh Eq. (14,125 kt) (see Table 4-6 and
8 Table 4-7). The trends in CC>2 emissions from lime production are directly proportional to trends in production,
9 which are described below.
10 Table 4-6: COz Emissions from Lime Production (MMT COz Eq. and kt)
Year MMT CCh Eq.
kt
1990
11.7
11,700
2010
2011
2012
2013
2014
13.4
14.0
13.7
14.0
14.1
13,381
13,981
13,715
14,045
14,125
11
13
14
15
16
17
18
19
20
21
Table 4-7: Potential, Recovered, and Net COz Emissions from Lime Production (kt)
Year
Potential
Recovered3 Net Emissions
1990
11,959
259
11,700
2010
2011
2012
2013
2014
13,776
14,389
14,188
14,513
14,630
395
407
473
467
505
13,381
13,981
13,715
14,045
14,125
1 For sugar refining and PCC production.
Note: Totals may not sum due to independent rounding.
12 In 2014, lime production was nearly the same as 2013 levels (increase of 1 percent) at 19,399 kilotons.
Methodology
To calculate emissions, the amounts of high-calcium and dolomitic lime produced were multiplied by their
respective emission factors using the Tier 2 approach from the 2006IPCC Guidelines (IPCC 2006). The emission
factor is the product of the stoichiometric ratio between CCh and CaO, and the average CaO and MgO content for
lime. The CaO and MgO content for lime is assumed to be 95 percent for both high-calcium and dolomitic lime
(IPCC 2006). The emission factors were calculated as follows:
For high-calcium lime:
[(44.01 g/mole C02) H- (56.08 g/mole CaO)] x (0.9500 CaO/lime) = 0.7455 g C02/g lime
For dolomitic lime:
4-10 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 [(88.02 g/mole C02) H- (96.39 g/mole CaO)] x (0.9500 CaO/lime) = 0.8675 g C02/g lime
2 Production was adjusted to remove the mass of chemically combined water found in hydrated lime, determined
3 according to the molecular weight ratios of H2O to (Ca(OH)2 and [Ca(OH)2«Mg(OH)2]) (IPCC 2006). These factors
4 set the chemically combined water content to 24.3 percent for high-calcium hydrated lime, and 27.2 percent for
5 dolomitic hydrated lime.
6 The 2006 IPCC Guidelines (Tier 2 method) also recommends accounting for emissions from lime kiln dust (LKD)
7 through application of a correction factor. LKD is a byproduct of the lime manufacturing process. LKD is a very
8 fine-grained material and is especially useful for applications requiring very small particle size. Most common LKD
9 applications include soil reclamation and agriculture. Currently, data on annual LKD production is not readily
10 available to develop a country specific correction factor. Lime emission estimates were multiplied by a factor of
11 1.02 to account for emissions from LKD (IPCC 2006).
12 Lime emission estimates were further adjusted to account for the amount of CO2 captured for use in on-site
13 processes. All the domestic lime facilities are required to report these data to EPA under its GHGRP. The total
14 national-level annual amount of CO2 captured for on-site process use was obtained from EPA's GHGRP (EPA
15 2015) based on reported facility level data for years 2010 through 2014. The amount of CO2 captured/recovered for
16 on-site process use is deducted from the total potential emissions (i.e., from lime production and LKD). The net lime
17 emissions are presented in Table 4-6 and Table 4-7. GHGRP data on CO2 removals (i.e., CO2 captured/recovered)
18 was available only for 2010 through 2014. Since GHGRP data are not available for 1990 through 2009, IPCC
19 "splicing" techniques were used as per the 2006 IPCC Guidelines on time series consistency (2006 IPCC
20 Guidelines, Volume 1, Chapter 5). The prior estimates for CO2 removal for 1990 through 2009 were adjusted based
21 on the "overlap" technique recommended by IPCC. Refer to the Recalculations Discussion section, below, for more
22 details.
23 Lime production data (by type, high-calcium- and dolomitic-quicklime, high-calcium- and dolomitic-hydrated, and
24 dead-burned dolomite) for 1990 through 2014 (see Table 4-8) were obtained from USGS (1992 through 2014,
25 Corathers 2015) and are compiled by USGS to the nearest ton. Natural hydraulic lime, which is produced from CaO
26 and hydraulic calcium silicates, is not manufactured in the United States (USGS 2011). Total lime production was
27 adjusted to account for the water content of hydrated lime by converting hydrate to oxide equivalent based on
28 recommendations from the IPCC, and is presented in Table 4-9 (IPCC 2006). The CaO and CaO'MgO contents of
29 lime were obtained from the IPCC (IPCC 2006). Since data for the individual lime types (high calcium and
30 dolomitic) were not provided prior to 1997, total lime production for 1990 through 1996 was calculated according to
31 the three year distribution from 1997 to 1999.
32 Table 4-8: High-Calcium- and Dolomitic-Quicklime, High-Calcium- and Dolomitic-Hydrated,
33 and Dead-Burned-Dolomite Lime Production (kt)
Year
High-Calcium
Quicklime
Dolomitic
Quicklime
High-Calcium
Hydrated
Dolomitic
Hydrated
Dead-Burned
Dolomite
1990
11,166
2,234
1,781
319
342
2005
14,100
2,990
2,220
474
200
2010
2011
2012
2013
2014
13,300
13,900
13,600
13,800
14,000
2,570
2,690
2,710
2,870
2,730
1,910
2,010
2,020
2,050
2,190
239
230
237
260
279
200
200
200
200
200
34 Table 4-9: Adjusted Lime Production (kt)
Year High-Calcium Dolomitic
1990 12,466 2,800
2005
15,721
3,522
Industrial Processes and Product Use 4-11
-------
2010 14,694 2,937
2011 15,367 3,051
2012 15,075 3,076
2013 15,297 3,252
2014 15,599 3,125
Note: Minus water content of hydrated
lime
i Uncertainty and Time Series Consistency
2 The uncertainties contained in these estimates can be attributed to slight differences in the chemical composition of
3 lime products and CC>2 recovery rates for on-site process use over the time series. Although the methodology
4 accounts for various formulations of lime, it does not account for the trace impurities found in lime, such as iron
5 oxide, alumina, and silica. Due to differences in the limestone used as a raw material, a rigid specification of lime
6 material is impossible. As a result, few plants produce lime with exactly the same properties.
7 In addition, a portion of the CC>2 emitted during lime production will actually be reabsorbed when the lime is
8 consumed, especially at captive lime production facilities. As noted above, lime has many different chemical,
9 industrial, environmental, and construction applications. In many processes, €62 reacts with the lime to create
10 calcium carbonate (e.g., water softening). Carbon dioxide reabsorption rates vary, however, depending on the
11 application. For example, 100 percent of the lime used to produce precipitated calcium carbonate reacts with CO 2;
12 whereas most of the lime used in steel making reacts with impurities such as silica, sulfur, and aluminum
13 compounds. Quantifying the amount of €62 that is reabsorbed would require a detailed accounting of lime use in
14 the United States and additional information about the associated processes where both the lime and byproduct €62
15 are "reused" are required to quantify the amount of €62 that is reabsorbed. Research conducted thus far has not
16 yielded the necessary information to quantify €62 reabsorption rates.6 However, some additional information on the
17 amount of €62 consumed on site at lime facilities has been obtained from EPA's GHGRP.
18 In some cases, lime is generated from calcium carbonate byproducts at pulp mills and water treatment plants.7 The
19 lime generated by these processes is included in the USGS data for commercial lime consumption. In the pulping
20 industry, mostly using the Kraft (sulfate) pulping process, lime is consumed in order to causticize a process liquor
21 (green liquor) composed of sodium carbonate and sodium sulfide. The green liquor results from the dilution of the
22 smelt created by combustion of the black liquor where biogenic C is present from the wood. Kraft mills recover the
23 calcium carbonate "mud" after the causticizing operation and calcine it back into lime—thereby generating €62—
24 for reuse in the pulping process. Although this re-generation of lime could be considered a lime manufacturing
25 process, the €62 emitted during this process is mostly biogenic in origin, and therefore is not included in the
26 industrial processes totals (Miner and Upton 2002). In accordance with IPCC methodological guidelines, any such
27 emissions are calculated by accounting for net carbon (C) fluxes from changes in biogenic C reservoirs in wooded or
28 crop lands (see the Land Use, Land-Use Change, and Forestry chapter).
29 In the case of water treatment plants, lime is used in the softening process. Some large water treatment plants may
30 recover their waste calcium carbonate and calcine it into quicklime for reuse in the softening process. Further
31 research is necessary to determine the degree to which lime recycling is practiced by water treatment plants in the
32 United States.
33 Another uncertainty is the assumption that calcination emissions for LKD are around 2 percent. The National Lime
34 Association (NLA) has commented that the estimates of emissions from LKD in the United States could be closer to
6 Representatives of the National Lime Association estimate that CCh reabsorption that occurs from the use of lime may offset as
much as a quarter of the CCh emissions from calcination (Males 2003).
7 Some carbide producers may also regenerate lime from their calcium hydroxide byproducts, which does not result in emissions
of CO2. In making calcium carbide, quicklime is mixed with coke and heated in electric furnaces. The regeneration of lime in
this process is done using a waste calcium hydroxide (hydrated lime) [CaC2 + 2H2O —> C2H2 + Ca(OH) 2], not calcium carbonate
[CaCOs]. Thus, the calcium hydroxide is heated in the kiln to simply expel the water [Ca(OH)2 + heat —»CaO + EhO] and no
CO2 is released.
4-12 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 6 percent. They also note that additional emissions (approximately 2 percent) may also be generated through
2 production of other byproducts/wastes (off-spec lime that is not recycled, scrubber sludge) at lime plants (Seeger
3 2013). There is limited data publicly available on LKD generation rates and also quantities, types of other
4 byproducts/wastes produced at lime facilities. Further research and data is needed to improve understanding of
5 additional calcination emissions to consider revising the current assumptions that are based on IPCC guidelines. In
6 preparing estimates for the current inventory, EPA initiated a dialogue with NLA to discuss data needs to generate a
7 country specific LKD factor and is reviewing the information provided by NLA.
8 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-10. Lime CCh emissions
9 for 2014 were estimated to be between 13.8 and 14.5 MMT CCh Eq. at the 95 percent confidence level. This
10 confidence level indicates a range of approximately 3 percent below and 3 percent above the emission estimate of
11 14.1 MMT CO2Eq.
12 Table 4-10: Approach 2 Quantitative Uncertainty Estimates for COz Emissions from Lime
13 Production (MMT COz Eq. and Percent)
,, „ 2014 Emission Estimate Uncertainty Ranee Relative to Emission Estimate3
Source Gas
(MMT CCh Eq.) (MMT CCh Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
Lime Production CCh 141 13.8 14.5 -3% +3%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
14 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
15 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
16 above.
17
Recalculations Discussion
18 Dead-burned dolomite production data for 2013 were updated relative to the previous Inventory based on the more
19 recent Minerals Yearbook: Lime 2013 [Advanced Release] (USGS 2014). This caused a slight decrease in 2013
20 emissions, by approximately 0.2 percent.
21 Planned Improvements
22 Future improvements involve evaluating recently obtained data to improve current assumptions associated with
23 emissions from production of LKD and other byproducts/wastes as discussed in the Uncertainty and Time Series
24 Consistency section per comments from the NLA. In response to comments, EPA met with NLA on April 7, 2015
25 to outline specific information required to apply IPCC methods to develop a country-specific correction factor to
26 more accurately estimate emissions from production of LKD. In response to this technical meeting, at the writing of
27 this report, NLA has compiled and shared historical emissions information reported by member facilities on an
28 annual basis under voluntary reporting initiatives over 2002 through 2011 associated with generation of total
29 calcined byproducts and LKD (LKD reporting only differentiated starting in 2010). This emissions information was
30 reported on a voluntary basis consistent with NLA's facility-level reporting protocol also recently provided.
31 Pending resources and data availability, historical CCh recovery rates at U.S. facilities producing lime will be
32 investigated to further evaluate results from use of overlap method to improve time series consistency.
Industrial Processes and Product Use 4-13
-------
i 4.3 Glass Production (IPCC Source Category
2 2A3)
3 Glass production is an energy and raw-material intensive process that results in the generation of CC>2 from both the
4 energy consumed in making glass and the glass process itself. Emissions from fuels consumed for energy purposes
5 during the production of glass are accounted for in the Energy sector.
6 Glass production employs a variety of raw materials in a glass-batch. These include formers, fluxes, stabilizers, and
7 sometimes colorants. The major raw materials (i.e., fluxes and stabilizers) which emit process-related CCh emissions
8 during the glass melting process are limestone, dolomite, and soda ash. The main former in all types of glass is silica
9 (SiCh). Other major formers in glass include feldspar and boric acid (i.e., borax). Fluxes are added to lower the
10 temperature at which the batch melts. Most commonly used flux materials are soda ash (sodium carbonate, Na2COs)
11 and potash (potassium carbonate, K2O). Stabilizers are used to make glass more chemically stable and to keep the
12 finished glass from dissolving and/or falling apart. Commonly used stabilizing agents in glass production are
13 limestone (CaCOs), dolomite (CaCOsMgCOs), alumina (A^Os), magnesia (MgO), barium carbonate (BaCOs),
14 strontium carbonate (SrCOs), lithium carbonate (I^COs), and zirconia (ZrCh) (OIT 2002). Glass makers also use a
15 certain amount of recycled scrap glass (cullet), which comes from in-house return of glassware broken in the process
16 or other glass spillage or retention such as recycling or cullet broker services.
17 The raw materials (primarily limestone, dolomite and soda ash) release CC>2 emissions in a complex high-
18 temperature chemical reaction during the glass melting process. This process is not directly comparable to the
19 calcination process used in lime manufacturing, cement manufacturing, and process uses of carbonates (i.e.,
20 limestone/dolomite use), but has the same net effect in terms of CC>2 emissions (IPCC 2006). The U.S. glass industry
21 can be divided into four main categories: containers, flat (window) glass, fiber glass, and specialty glass. The
22 majority of commercial glass produced is container and flat glass (EPA 2009). The United States is one of the major
23 global exporters of glass. Domestically, demand comes mainly from the construction, auto, bottling, and container
24 industries. There are over 1,500 companies that manufacture glass in the United States, with the largest being
25 Corning, Guardian Industries, Owens-Illinois, and PPG Industries.8
26 In 2014, 775 kilotons of limestone and 2,410 kilotons of soda ash were consumed for glass production (USGS
27 2015a, Willett 2015). Dolomite consumption data for glass manufacturing was reported to be zero for 2014. Use of
28 limestone and soda ash in glass production resulted in aggregate CO2 emissions of 1.3 MMT CO2 Eq. (1,341 kt) (see
29 Table 4-11). Overall, emissions have decreased 13 percent from 1990 through 2014.
30 Emissions from glass production have remained relatively constant over the time series with some fluctuations since
31 1990. In general, these fluctuations were related to the behavior of the export market and the U.S. economy.
32 Specifically, the extended downturn in residential and commercial construction and automotive industries between
33 2008 and 2010 resulted in reduced consumption of glass products, causing a drop in global demand for
34 limestone/dolomite and soda ash, and a corresponding decrease in emissions. Furthermore, the glass container sector
35 is one of the leading soda ash consuming sectors in the United States. Some commercial food and beverage package
36 manufacturers are shifting from glass containers towards lighter and more cost effective polyethylene terephthalate
37 (PET) based containers, putting downward pressure on domestic consumption of soda ash (USGS 1995 through
38 2014).
39 Table 4-11: COz Emissions from Glass Production (MMT COz Eq. and kt)
Year _ MMT CCh Eq.
1990 1.5 1,535
Excerpt from Glass & Glass Product Manufacturing Industry Profile, First Research. Available online at
.
4-14 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
2010
2011
2012
2013
2014
1.5
1.3
1.2
1.3
1.3
1,481
1,299
1,248
1,317
1,341
Methodology
Carbon dioxide emissions were calculated based on the 2006 IPCC Guidelines Tier 3 method by multiplying the
quantity of input carbonates (limestone, dolomite, and soda ash) by the carbonate-based emission factor (in metric
tons CCVmetric ton carbonate): limestone, 0.43971; dolomite, 0.47732; and soda ash, 0.41492.
Consumption data for 1990 through 2014 of limestone, dolomite, and soda ash used for glass manufacturing were
obtained from the USGS Minerals Yearbook: Crushed Stone Annual Report (1995 through 2015b), 2014
preliminary data from the USGS Crushed Stone Commodity Expert (Willett 2015), the USGS Minerals Yearbook:
Soda Ash Annual Report (1995 through 2014), USGS Mineral Industry Surveys for Soda Ash in January 2015
(USGS 2015a) and the U.S. Bureau of Mines (1991 and 1993a), which are reported to the nearest ton. During 1990
and 1992, the USGS did not conduct a detailed survey of limestone and dolomite consumption by end-use.
Consumption for 1990 was estimated by applying the 1991 percentages of total limestone and dolomite use
constituted by the individual limestone and dolomite uses to 1990 total use. Similarly, the 1992 consumption figures
were approximated by applying an average of the 1991 and 1993 percentages of total limestone and dolomite use
constituted by the individual limestone and dolomite uses to the 1992 total.
Additionally, each year the USGS withholds data on certain limestone and dolomite end-uses due to confidentiality
agreements regarding company proprietary data. For the purposes of this analysis, emissive end-uses that contained
withheld data were estimated using one of the following techniques: (1) the value for all the withheld data points for
limestone or dolomite use was distributed evenly to all withheld end-uses; or (2) the average percent of total
limestone or dolomite for the withheld end-use in the preceding and succeeding years.
There is a large quantity of limestone and dolomite reported to the USGS under the categories "unspecified-
reported" and "unspecified-estimated." A portion of this consumption is believed to be limestone or dolomite used
for glass manufacturing. The quantities listed under the "unspecified" categories were, therefore, allocated to glass
manufacturing according to the percent limestone or dolomite consumption for glass manufacturing end use for that
year.9
Based on the 2014 reported data, the estimated distribution of soda ash consumption for glass production compared
to total domestic soda ash consumption is 48 percent (USGS 2015a).
Table 4-12: Limestone, Dolomite, and Soda Ash Consumption Used in Glass Production (kt)
Activity
Limestone
Dolomite
Soda Ash
Total
1990
3
3
430
59
,177
,666
2005
3
4
920
541
,050
,511
2010
2
3
999
0
,510
,509
2011
614
0
2,480
3,094
2012
555
0
2,420
2,975
2013
693
0
2,440
3,133
2014
775
0
2,410
3,185
28 Uncertainty and Time-Series Consistency
29 The uncertainty levels presented in this section arise in part due to variations in the chemical composition of
30 limestone used in glass production. In addition to calcium carbonate, limestone may contain smaller amounts of
31 magnesia, silica, and sulfur, among other minerals (potassium carbonate, strontium carbonate and barium carbonate,
32 and dead burned dolomite). Similarly, the quality of the limestone (and mix of carbonates) used for glass
3 3 manufacturing will depend on the type of glass being manufactured.
9 This approach was recommended by USGS.
Industrial Processes and Product Use 4-15
-------
1 The estimates below also account for uncertainty associated with activity data. Large fluctuations in reported
2 consumption exist, reflecting year-to-year changes in the number of survey responders. The uncertainty resulting
3 from a shifting survey population is exacerbated by the gaps in the time series of reports. The accuracy of
4 distribution by end use is also uncertain because this value is reported by the manufacturer of the input carbonates
5 (limestone, dolomite & soda ash) and not the end user. For 2014, there has been no reported consumption of
6 dolomite for glass manufacturing. This data has been reported to USGS by dolomite manufacturers and not end-
7 users (i.e., glass manufacturers). There is a high uncertainty associated with this estimate, as dolomite is a major raw
8 material consumed in glass production. Additionally, there is significant inherent uncertainty associated with
9 estimating withheld data points for specific end uses of limestone and dolomite. The uncertainty of the estimates for
10 limestone and dolomite used in glass making is especially high; however, since glass making accounts for a small
11 percent of consumption, its contribution to the overall emissions estimate is low. Lastly, much of the limestone
12 consumed in the United States is reported as "other unspecified uses;" therefore, it is difficult to accurately allocate
13 this unspecified quantity to the correct end-uses. Further research is needed into alternate and more complete
14 sources of data on carbonate-based raw material consumption by the glass industry.
15 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-13. In 2014, glass
16 production CC>2 emissions were estimated to be between 1.3 and 1.4 MMT CCh Eq. at the 95 percent confidence
17 level. This indicates a range of approximately 4 percent below and 5 percent above the emission estimate of 1.3
18 MMTCO2Eq.
19 Table 4-13: Approach 2 Quantitative Uncertainty Estimates for COz Emissions from Glass
20 Production (MMT COz Eq. and Percent)
^ „ 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
6 (MMT CCh Eq.) (MMT CCh Eq.) (%)
24
31
Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
Glass Production CCh O O \_4 -4% +5%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
21 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
22 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
23 above.
Recalculations Discussion
25 Limestone and dolomite consumption data for 2013 were revised to reflect updated USGS data (USGS 2015b). This
26 change resulted in an increase of CC>2 emissions by approximately 14 percent. The preliminary data for 2013 was
27 obtained directly from the USGS Crushed Stone Commodity Expert (Willett 2014). In April 2015, USGS published
28 the 2013 Minerals Yearbook for Crushed Stone and the preliminary data were revised to reflect the latest USGS
29 published data. The published time series was reviewed to ensure time series consistency. Details on the emission
30 trends through time are described in more detail in the Methodology section, above.
Planned Improvements
32 Currently, only limestone and soda ash consumption data for glass manufacturing is publicly available. While
33 limestone and soda ash are the predominant carbonates used in glass manufacturing, there are other carbonates that
34 are also consumed for glass manufacturing, although in smaller quantities. Pending resources, future improvements
35 will include research into other sources of data for carbonate consumption by the glass industry.
36 Additionally, future improvements will also continuing to evaluate and analyze data reported under EPA's GHGPJ3
37 that would be useful to improve the emission estimates for the Glass Production source category. Particular attention
38 will be made to ensure time series consistency of the emissions estimates presented in future Inventory reports,
39 consistent with IPCC and UNFCCC guidelines. This is required as the facility-level reporting data from EPA's
40 GHGRP, with the program's initial requirements for reporting of emissions in calendar year 2010, are not available
41 for all inventory years (i.e., 1990 through 2009) as required for this Inventory. In addition, EPA's GHGPJ3 has an
42 emission threshold for reporting, so the data do not account for all glass production in the United States. In
4-16 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
implementing improvements and integration of data from EPA's GHGRP, the latest guidance from the IPCC on the
use of facility-level data in national inventories will be relied upon.10
4.4 Other Process Uses of Carbonates (IPCC
Source Category 2A4)
5 Limestone (CaCOs), dolomite (CaCOsMgCOs)11, and other carbonates such as soda ash, magnesite, and siderite are
6 basic materials used by a wide variety of industries, including construction, agriculture, chemical, metallurgy, glass
7 production, and environmental pollution control. This section addresses only limestone and dolomite use. For
8 industrial applications, carbonates such as limestone and dolomite are heated sufficiently enough to calcine the
9 material and generate CO2 as a byproduct.
10 CaC03 -> CaO + C02
11 MgC03 -> MgO + C02
12 Examples of such applications include limestone used as a flux or purifier in metallurgical furnaces, as a sorbent in
13 flue gas desulfurization (FGD) systems for utility and industrial plants, and as a raw material for the production of
14 glass, lime, and cement. Emissions from limestone and dolomite used in other process sectors such as cement, lime,
15 glass production, and iron and steel, are excluded from this section and reported under their respective source
16 categories (e.g., glass manufacturing IPCC Source Category 2A3.) Emission from soda ash consumption is reported
17 under respective categories (e.g., glass manufacturing IPCC Source Category 2A3 and soda ash production and
18 consumption IPCC Source Category 2B7). Emissions from fuels consumed for energy purposes during these
19 processes are accounted for in the Energy chapter.
20 Limestone is widely distributed throughout the world in deposits of varying sizes and degrees of purity. Large
21 deposits of limestone occur in nearly every state in the United States, and significant quantities are extracted for
22 industrial applications. The leading limestone producing States are Texas, Missouri, Florida, Ohio, and Pennsylvania,
23 which contribute 43 percent of the total U.S. output (USGS 2015). Similarly, dolomite deposits are also widespread
24 throughout the world. Dolomite deposits are found in the United States, Canada, Mexico, Europe, Africa, and Brazil.
25 In the United States, the leading dolomite producing states are Illinois, Pennsylvania, and New York, which
26 contribute 55 percent of the total U.S. output (USGS 2015).
27 In 2014, 25,085 kt of limestone and 3,359 kt of dolomite were consumed for these emissive applications, excluding
28 glass manufacturing (Willett 2015). Usage of limestone and dolomite resulted in aggregate CO2 emissions of 12.1
29 MMT CO2 Eq. (12,077 kt) (see Table 4-14 and Table 4-15). Overall, emissions have increased 146 percent from
30 1990 through 2014.
31 Table 4-14: COz Emissions from Other Process Uses of Carbonates (MMT COz Eq.)
Year Flux Stone
FGD
Magnesium
Production
Other
Miscellaneous
Uses
Total
1990
2.6
1.4
0.1
2010
2011
2012
10 See.
Limestone and dolomite are collectively referred to as limestone by the industry, and intermediate varieties are seldom
distinguished.
Industrial Processes and Product Use 4-17
-------
2013
2014
2.3
2.9
6.3
7.2
0.0
0.0
1.8
1.9
10.4
12.1
Notes: Totals may not sum due to independent rounding. "Other miscellaneous uses"
include chemical stone, mine dusting or acid water treatment, acid neutralization, and
sugar refining.
Table 4-15: COz Emissions from Other Process Uses of Carbonates (kt)
Year Flux Stone
FGD
Magnesium
Production
Other
Miscellaneous
Uses
Total
1990
2,592
1,432
64
4,907
2010
2011
2012
2013
2014
1,560
1,467
1,077
2,307
2,932
7,064
5,420
5,797
6,309
7,212
0
0
0
0
0
937
2,449
1,148
1,798
1,933
9,560
9,335
8,022
10,414
12,077
Methodology
3 Carbon dioxide emissions were calculated based on the 2006IPCC Guidelines Tier 2 method by multiplying the
4 quantity of limestone or dolomite consumed by the emission factor for limestone or dolomite calcination,
5 respectively, Table 2.1-limestone: 0.43971 metric ton CCh/metric ton carbonate, and dolomite: 0.47732 metric ton
6 CCVmetric ton carbonate.12 This methodology was used for flux stone, flue gas desulfurization systems, chemical
7 stone, mine dusting or acid water treatment, acid neutralization, and sugar refining. Flux stone used during the
8 production of iron and steel was deducted from the Other Process Uses of Carbonates estimate and attributed to the
9 Iron and Steel Production estimate. Similarly limestone and dolomite consumption for glass manufacturing, cement,
10 and lime manufacturing are excluded from this category and attributed to their respective categories.
11 Historically, the production of magnesium metal was the only other significant use of limestone and dolomite that
12 produced CC>2 emissions. At the end of 2001, the sole magnesium production plant operating in the United States
13 that produced magnesium metal using a dolomitic process that resulted in the release of CO2 emissions ceased its
14 operations (USGS 1995b through 2012; USGS 2013a).
15 Consumption data for 1990 through 2014 of limestone and dolomite used for flux stone, flue gas desulfurization
16 systems, chemical stone, mine dusting or acid water treatment, acid neutralization, and sugar refining (see Table
17 4-16) were obtained from the USGS Minerals Yearbook: Crushed Stone Annual Report (1995a through 2015),
18 preliminary data for 2014 from USGS Crushed Stone Commodity Expert (Willett 2015), and the U.S. Bureau of
19 Mines (1991 and 1993a), which are reported to the nearest ton. The production capacity data for 1990 through 2014
20 of dolomitic magnesium metal also came from the USGS (1995b through 2012, USGS 2013a) and the U.S. Bureau
21 of Mines (1990 through 1993b). During 1990 and 1992, the USGS did not conduct a detailed survey of limestone
22 and dolomite consumption by end-use. Consumption for 1990 was estimated by applying the 1991 percentages of
23 total limestone and dolomite use constituted by the individual limestone and dolomite uses to 1990 total use.
24 Similarly, the 1992 consumption figures were approximated by applying an average of the 1991 and 1993
25 percentages of total limestone and dolomite use constituted by the individual limestone and dolomite uses to the
26 1992 total.
27 Additionally, each year the USGS withholds data on certain limestone and dolomite end-uses due to confidentiality
28 agreements regarding company proprietary data. For the purposes of this analysis, emissive end-uses that contained
29 withheld data were estimated using one of the following techniques: (1) the value for all the withheld data points for
30 limestone or dolomite use was distributed evenly to all withheld end-uses; (2) the average percent of total limestone
12 IPCC 2006 Guidelines, Volume 3: Chapter 2
4-18 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1
2
3
4
5
6
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
or dolomite for the withheld end-use in the preceding and succeeding years; or (3) the average fraction of total
limestone or dolomite for the end-use over the entire time period.
There is a large quantity of crushed stone reported to the USGS under the category "unspecified uses." A portion of
this consumption is believed to be limestone or dolomite used for emissive end uses. The quantity listed for
"unspecified uses" was, therefore, allocated to each reported end-use according to each end-use's fraction of total
consumption in that year.13
Table 4-16: Limestone and Dolomite Consumption (kt)
Activity
Flux Stone
Limestone
Dolomite
FGD
Other Miscellaneous Uses
Total
1990
6,737
5, 8041
933M
3,258 •
1,835
11,830
2005
7,022
3,165
3,857
6,761
1,632
15,415
2010
4,440
1,921
2,520
16,064
2,121
22,626
2011
4,396
2,531
1,865
12,326
5,548
22,270
2012
3,666
3,108
559
13,185
2,610
19,461
2013
6,345
4,380
1,965
14,347
3,973
24,665
2014
7,648
4,304
3,344
16,402
4,395
28,444
Uncertainty and Time-Series Consistency
The uncertainty levels presented in this section account for uncertainty associated with activity data. Data on
limestone and dolomite consumption are collected by USGS through voluntary national surveys. USGS contacts the
mines (i.e., producers of various types of crushed stone) for annual sales data. Data on other carbonate consumption
are not readily available. The producers report the annual quantity sold to various end-users/industry types. USGS
estimates the historical response rate for the crushed stone survey to be approximately 70 percent, the rest is
estimated by USGS. Large fluctuations in reported consumption exist, reflecting year-to-year changes in the number
of survey responders. The uncertainty resulting from a shifting survey population is exacerbated by the gaps in the
time series of reports. The accuracy of distribution by end use is also uncertain because this value is reported by the
producer/mines and not the end user. Additionally, there is significant inherent uncertainty associated with
estimating withheld data points for specific end uses of limestone and dolomite. Lastly, much of the limestone
consumed in the United States is reported as "other unspecified uses;" therefore, it is difficult to accurately allocate
this unspecified quantity to the correct end-uses.
Uncertainty in the estimates also arises in part due to variations in the chemical composition of limestone. In
addition to calcium carbonate, limestone may contain smaller amounts of magnesia, silica, and sulfur, among other
minerals. The exact specifications for limestone or dolomite used as flux stone vary with the pyrometallurgical
process and the kind of ore processed.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-17. Other Process Uses of
Carbonates CC>2 emissions in 2014 were estimated to be between 10.7 and 14.0 MMT CCh Eq. at the 95 percent
confidence level. This indicates a range of approximately 12 percent below and 15 percent above the emission
estimate of 12.1 MMT CO2 Eq.
13 This approach was recommended by USGS, the data collection agency.
Industrial Processes and Product Use 4-19
-------
1 Table 4-17: Approach 2 Quantitative Uncertainty Estimates for COz Emissions from Other
2 Process Uses of Carbonates (MMT COz Eq. and Percent)
„ „ 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
rce (MMT CCh Eq.) (MMT CCh Eq.) (%)
15
16
Other Process
Uses of CO2 12.1
Carbonates
Lower
Bound
10.7
Upper
Bound
14.0
Lower
Bound
-12%
Upper
Bound
+15%
1 Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
3 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
4 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
5 above.
6 Recalculations Discussion
7 Limestone and dolomite consumption data, by end-use, for 2013 were updated relative to the previous Inventory
8 based on the recently published 2013 Minerals Yearbook: Crushed Stone. In the previous Inventory report (i.e.,
9 1990-2013), preliminary data were used for 2013 and updated for the current Inventory. In April 2015, USGS
10 published the 2013 Minerals Yearbook for Crushed Stone and the preliminary data were revised to reflect the latest
11 USGS published data. The published time series was reviewed to ensure time series consistency. This update caused
12 an increase in total limestone and dolomite consumption for emissive end uses in 2013 by approximately 120
13 percent. The revised 2013 emission estimate increased by approximately 135 percent relative to the previous report
14 due to this change.
4.5 Ammonia Production (IPCC Source
Category 2B1)
17 Emissions of CO2 occur during the production of synthetic ammonia, primarily through the use of natural gas,
18 petroleum coke, or naphtha as a feedstock. The natural gas-, naphtha-, and petroleum coke-based processes produce
19 CO2 and hydrogen (H2), the latter of which is used in the production of ammonia. The brine electrolysis process for
20 production of ammonia does not lead to process-based CO2 emissions. Emissions from fuels consumed for energy
21 purposes during the production of ammonia are accounted for in the Energy chapter.
22 In the United States, the majority of ammonia is produced using a natural gas feedstock; however one synthetic
23 ammonia production plant located in Kansas is producing ammonia from petroleum coke feedstock. In some U.S.
24 plants, some of the CO2 produced by the process is captured and used to produce urea rather than being emitted to
25 the atmosphere. There are approximately 13 companies operating 26 ammonia producing facilities in 17 states.
26 More than 56 percent of domestic ammonia production capacity is concentrated in the States of Louisiana (29
27 percent), Oklahoma (21 percent), and Texas (6 percent) (USGS 2015).
28 There are five principal process steps in synthetic ammonia production from natural gas feedstock. The primary
29 reforming step converts CH4 to CO2, carbon monoxide (CO), and H2 in the presence of a catalyst. Only 30 to 40
30 percent of the CH4 feedstock to the primary reformer is converted to CO and CO2 in this step of the process. The
31 secondary reforming step converts the remaining CH4 feedstock to CO and CO2. The CO in the process gas from
32 the secondary reforming step (representing approximately 15 percent of the process gas) is converted to CO2 in the
33 presence of a catalyst, water, and air in the shift conversion step. Carbon dioxide is removed from the process gas
34 by the shift conversion process, and the hydrogen gas is combined with the nitrogen (N2) gas in the process gas
35 during the ammonia synthesis step to produce ammonia. The CO2 is included in a waste gas stream with other
36 process impurities and is absorbed by a scrubber solution. In regenerating the scrubber solution, CO2 is released
37 from the solution.
4-20 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
29
1 The conversion process for conventional steam reforming of CH4, including the primary and secondary reforming
2 and the shift conversion processes, is approximately as follows:
3 0.88C7/4 +1.26Air + 1.24//20 -> 0.88C02 + N2 + 3H2
4 N2 + 3//2 -> 2N//3
5 To produce synthetic ammonia from petroleum coke, the petroleum coke is gasified and converted to CO2 and H2.
6 These gases are separated, and the H2 is used as a feedstock to the ammonia production process, where it is reacted
1 with N2 to form ammonia.
8 Not all of the CO2 produced during the production of ammonia is emitted directly to the atmosphere. Some of the
9 ammonia and some of the CO2 produced by the synthetic ammonia process are used as raw materials in the
10 production of urea [CO(NH2)2], which has a variety of agricultural and industrial applications.
11 The chemical reaction that produces urea is:
12 2NH3+ C02 -^NH2COONH4 -> CO(NH2}2 + H20
13 Only the CO2 emitted directly to the atmosphere from the synthetic ammonia production process is accounted for in
14 determining emissions from ammonia production. The CO2 that is captured during the ammonia production process
15 and used to produce urea does not contribute to the CO2 emission estimates for ammonia production presented in
16 this section. Instead, CO2 emissions resulting from the consumption of urea are attributed to the urea consumption
17 or urea application source category (under the assumption that the carbon stored in the urea during its manufacture is
18 released into the environment during its consumption or application). Emissions of CO2 resulting from agricultural
19 applications of urea are accounted for in the Cropland Remaining Cropland section of the Land Use, Land-Use
20 Change, and Forestry chapter. Emissions of CO2 resulting from non-agricultural applications of urea (e.g., use as a
21 feedstock in chemical production processes) are accounted for in the Urea Consumption for Non-Agricultural
22 Purposes section of this chapter.
23 Total emissions of CO2 from ammonia production in 2014 were 9.4 MMT CO2 Eq. (9,436 kt), and are summarized
24 in Table 4-18 and Table 4-19. Ammonia production relies on natural gas as both a feedstock and a fuel, and as such,
25 market fluctuations and volatility in natural gas prices affect the production of ammonia. Since 1990, emissions
26 from ammonia production have decreased by 28 percent. Emissions in 2014 have decreased by approximately 5
27 percent from the 2013 levels.
28 Table 4-18: COz Emissions from Ammonia Production (MMT COz Eq.)
Source
Ammonia Production
Total
Table 4-19:
Source
COz
1990
13.0
13.0
Emissions from
Ammonia Production
Total
1990
13,047
13,047
2005
9.2
9.2
Ammonia
2005
9,196
9,196
2010
9.2
9.2
2011
Production
• 2010
9,188
9,188
9.3
9.3
(kt)
2011
9,
9,
292
292
2012
9.4
9.4
2012
9
9
,377
,377
2013
10.0
10.0
2013
9,962
9,962
2014
9.4
9.4
2014
9,436
9,436
30 Methodology
31 Carbon dioxide emissions from production of synthetic ammonia from natural gas feedstock is based on the 2006
32 IPCC Guidelines (IPCC 2006) Tier 1 and 2 method. A country-specific emission factor is developed and applied to
33 national ammonia production to estimate emissions. The method uses a CO2 emission factor published by the
34 European Fertilizer Manufacturers Association (EFMA) that is based on natural gas-based ammonia production
35 technologies that are similar to those employed in the United States. This CO2 emission factor of 1.2 metric tons
36 CCVmetric ton NH3 (EFMA 2000a) is applied to the percent of total annual domestic ammonia production from
37 natural gas feedstock.
Industrial Processes and Product Use 4-21
-------
1 Emissions of CO2 from ammonia production are then adjusted to account for the use of some of the CO2 produced
2 from ammonia production as a raw material in the production of urea. The CC>2 emissions reported for ammonia
3 production are reduced by a factor of 0.733 multiplied by total annual domestic urea production. This corresponds
4 to a stoichiometric CCh/urea factor of 44/60, assuming complete conversion of NH3 and CCh to urea (IPCC 2006,
5 EFMA 2000b).
6 All synthetic ammonia production and subsequent urea production are assumed to be from the same process—
7 conventional catalytic reforming of natural gas feedstock, with the exception of ammonia production from
8 petroleum coke feedstock at one plant located in Kansas. Annual ammonia and urea production are shown in Table
9 4-20. The CCh emission factor for production of ammonia from petroleum coke is based on plant-specific data,
10 wherein all carbon contained in the petroleum coke feedstock that is not used for urea production is assumed to be
11 emitted to the atmosphere as CCh (Bark 2004). Ammonia and urea are assumed to be manufactured in the same
12 manufacturing complex, as both the raw materials needed for urea production are produced by the ammonia
13 production process. The CCh emission factor of 3.57 metric tons CCh/metric ton NH3 for the petroleum coke
14 feedstock process (Bark 2004) is applied to the percent of total annual domestic ammonia production from
15 petroleum coke feedstock.
16 The emission factor of 1.2 metric ton CCh/metric ton NH3 for production of ammonia from natural gas feedstock
17 was taken from the EFMA Best Available Techniques publication, Production of Ammonia (EFMA 2000a). The
18 EFMA reported an emission factor range of 1.15 to 1.30 metric ton CO2/metric ton NH3, with 1.2 metric ton
19 CCVmetric ton NH3 as a typical value (EFMA 2000a). Technologies (e.g., catalytic reforming process, etc.)
20 associated with this factor are found to closely resemble those employed in the United States for use of natural gas
21 as a feedstock. The EFMA reference also indicates that more than 99 percent of the CH4 feedstock to the catalytic
22 reforming process is ultimately converted to €62. As noted earlier, emissions from fuels consumed for energy
23 purposes during the production of ammonia are accounted for in the Energy chapter. The total ammonia production
24 data for 2011, 2012, 2013, and 2014 were obtained from American Chemistry Council (2015). Foryears before
25 2011, ammonia production data (See Table 4-20) were obtained from Coffeyville Resources (Coffeyville 2005,
26 2006, 2007a, 2007b, 2009, 2010, 2011, and 2012) and the Census Bureau of the U.S. Department of Commerce
27 (U.S. Census Bureau 1991 through 1994, 1998 through 2010) as reported in Current Industrial Reports Fertilizer
28 Materials and Related Products annual and quarterly reports. Urea-ammonia nitrate production from petroleum coke
29 for years through 2011 was obtained from Coffeyville Resources (Coffeyville 2005, 2006, 2007a, 2007b, 2009,
30 2010, 2011, and 2012), and from CVR Energy, Inc. Annual Report (CVR 2012 ,2014, and 2015) for 2012, 2013,
31 and 2014. Urea production data for 1990 through 2008 were obtained from the Minerals Yearbook: Nitrogen (USGS
32 1994 through 2009). Urea production data for 2009 through 2010 were obtained from the U.S. Census Bureau (U.S.
33 Census Bureau 2010 and 2011). The U.S. Census Bureau ceased collection of urea production statistics, and urea
34 production data for 2011, 2012, and 2013 were obtained from the Minerals Yearbook: Nitrogen (USGS 2014, 2015).
35 The urea production data for 2014 are not yet published and so 2013 data were used as proxies for 2014.
36 Table 4-20: Ammonia Production and Urea Production (kt)
37
., Ammonia Urea
Year
Production Production
1990 15,425 7,450
2010
2011
2012
2013
2014
10,084
10,325
10,305
10,930
10,515
5,122
5,430
5,220
5,480
5,480
Uncertainty and Time-Series Consistency
38 The uncertainties presented in this section are primarily due to how accurately the emission factor used represents an
39 average across all ammonia plants using natural gas feedstock. Uncertainties are also associated with ammonia
4-22 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 production estimates and the assumption that all ammonia production and subsequent urea production was from the
2 same process—conventional catalytic reforming of natural gas feedstock, with the exception of one ammonia
3 production plant located in Kansas that is manufacturing ammonia from petroleum coke feedstock. Uncertainty is
4 also associated with the representativeness of the emission factor used for the petroleum coke-based ammonia
5 process. It is also assumed that ammonia and urea are produced at collocated plants from the same natural gas raw
6 material.
7 Recovery of CC>2 from ammonia production plants for purposes other than urea production (e.g., commercial sale,
8 etc.) has not been considered in estimating the CC>2 emissions from ammonia production, as data concerning the
9 disposition of recovered CCh are not available. Such recovery may or may not affect the overall estimate of €62
10 emissions depending upon the end use to which the recovered €62 is applied. Further research is required to
11 determine whether byproduct €62 is being recovered from other ammonia production plants for application to end
12 uses that are not accounted for elsewhere.
13 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-21. Ammonia Production
14 CO2 emissions were estimated to be between 8.7 and 10.2 MMT €62 Eq. at the 95 percent confidence level. This
15 indicates a range of approximately 8 percent below and 8 percent above the emission estimate of 9.4 MMT €62 Eq.
16 Table 4-21: Approach 2 Quantitative Uncertainty Estimates for COz Emissions from
17 Ammonia Production (MMT COz Eq. and Percent)
„ „ 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
6 (MMT CCh Eq.) (MMT CCh Eq.) (%)
21
27
Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
Ammonia Production CCh 9.4 8/7 10.2 -8% +8%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
18 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
19 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
20 above.
Recalculations Discussion
22 Production estimates for urea production for 2013 were updated relative to the previous Inventory using information
23 obtained from the recent 2013 Minerals Yearbook: Nitrogen (USGS 2015). For the previous version of the Inventory
24 (i.e., 1990-2013), 2012 data was used as a proxy for 2013 as the 2013 data were not published prior to the previous
25 Inventory report. This update resulted in a slight decrease of emissions by approximately 2 percent for 2013 relative
26 to the previous report.
Planned Improvements
28 Future improvements involve continuing to evaluate and analyze data reported under EPA's GHGRP to improve the
29 emission estimates for the Ammonia Production source category. Particular attention will be made to ensure time
30 series consistency of the emissions estimates presented in future Inventory reports, consistent with IPCC and
31 UNFCCC guidelines. This is required as the facility-level reporting data from EPA's GHGRP, with the program's
32 initial requirements for reporting of emissions in calendar year 2010, are not available for all inventory years (i.e.,
33 1990 through 2009) as required for this Inventory. In implementing improvements and integration of data from
34 EPA's GHGRP, the latest guidance from the IPCC on the use of facility-level data in national inventories will be
35 relied upon.14 Specifically, the planned improvements include assessing data to update the emission factors to
36 include both fuel and feedstock CO2 emissions and incorporate CO2 capture and storage. Methodologies will also
37 be updated if additional ammonia production plants are found to use hydrocarbons other than natural gas for
38 ammonia production.
14 See.
Industrial Processes and Product Use 4-23
-------
i 4.6 Urea Consumption for Non-Agricultural
2 Purposes
3 Urea is produced using ammonia and CCh as raw materials. All urea produced in the United States is assumed to be
4 produced at ammonia production facilities where both ammonia and CCh are generated. There are approximately 20
5 of these facilities operating in the United States.
6 The chemical reaction that produces urea is:
7 2NH3+ C02 -^NH2COONH4 -> CO(NH2)2 + H20
8 This section accounts for CC>2 emissions associated with urea consumed exclusively for non-agricultural purposes.
9 Carbon dioxide emissions associated with urea consumed for fertilizer are accounted for in the Cropland Remaining
10 Cropland section of the Land Use, Land-Use Change, and Forestry chapter.
11 Urea is used as a nitrogenous fertilizer for agricultural applications and also in a variety of industrial applications.
12 The industrial applications of urea include its use in adhesives, binders, sealants, resins, fillers, analytical reagents,
13 catalysts, intermediates, solvents, dyestuffs, fragrances, deodorizers, flavoring agents, humectants and dehydrating
14 agents, formulation components, monomers, paint and coating additives, photosensitive agents, and surface
15 treatments agents. In addition, urea is used for abating NOX emissions from coal-fired power plants and diesel
16 transportation motors.
17 Emissions of CC>2 from urea consumed for non-agricultural purposes in 2014 were estimated to be 4.0 MMT CO2
18 Eq. (4,007 kt), and are summarized in Table 4-22 and Table 4-23. 2014 data on urea production data, urea exports
19 and imports are not yet published. 2013 data has been used as proxy for 2014. Net CO2 emissions from urea
20 consumption for non-agricultural purposes in 2014 have increased by approximately 6 percent from 1990.
21 Table 4-22: COz Emissions from Urea Consumption for Non-Agricultural Purposes (MMT COz
22 Eq.)
~Source 1990 2005 2010 2011 2012 2013 2014
Urea Consumption 3.8 3/7 4.7 4.0 4.4 4.2 4.0
Total 3.8 3.7 4.7 4.0 4.4 4.2 4.0
23 Table 4-23: COz Emissions from Urea Consumption for Non-Agricultural Purposes (kt)
Source 1990 2005 2010 2011 2012 2013 2014
Urea Consumption 3,784 3,653 4,730 4,029 4,449 4,179 4,007
Total 3,784 3,653 4,730 4,029 4,449 4,179 4,007
24 Methodology
25 Emissions of CC>2 resulting from urea consumption for non-agricultural purposes are estimated by multiplying the
26 amount of urea consumed in the United States for non-agricultural purposes by a factor representing the amount of
27 CO2 used as a raw material to produce the urea. This method is based on the assumption that all of the carbon in
28 urea is released into the environment as CO2 during use, and consistent with the 2006IPCC Guidelines (IPCC
29 2006).
30 The amount of urea consumed for non-agricultural purposes in the United States is estimated by deducting the
31 quantity of urea fertilizer applied to agricultural lands, which is obtained directly from the Land Use, Land-Use
32 Change, and Forestry chapter (see Table 6-25) and is reported in Table 4-24, from the total domestic supply of urea.
3 3 The domestic supply of urea is estimated based on the amount of urea produced plus the sum of net urea imports and
4-24 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 exports. A factor of 0.73 3 tons of CO2 per ton of urea consumed is then applied to the resulting supply of urea for
2 non-agricultural purposes to estimate CC>2 emissions from the amount of urea consumed for non-agricultural
3 purposes. The 0.733 tons of CCh per ton of urea emission factor is based on the stoichiometry of producing urea
4 from ammonia and CC>2. This corresponds to a stoichiometric CCh/urea factor of 44/60, assuming complete
5 conversion of NH3 and CO2 to urea (IPCC 2006, EFMA 2000).
6 Urea production data for 1990 through 2008 were obtained from the Minerals Yearbook: Nitrogen (USGS 1994
7 through 2009). Urea production data for 2009 through 2010 were obtained from the U.S. Census Bureau (2011).
8 The U.S. Census Bureau ceased collection of urea production statistics in 2011, therefore, urea production data for
9 2011, 2012, and 2013 were obtained from the Minerals Yearbook: Nitrogen (USGS 2014 through 2015). Urea
10 production data for 2014 are not yet publicly available and so 2013 data have been used as proxy.
11 Urea import data for 2014 are not yet publicly available and so 2013 data have been used as proxy. Urea import data
12 for 2013 were obtained from the Minerals Yearbook: Nitrogen (USGS 2015). Urea import data for 2011 and 2012
13 were taken from U.S. Fertilizer Import/Exports from USDA Economic Research Service Data Sets (U.S.
14 Department of Agriculture 2012). Urea import data for the previous years were obtained from the U.S. Census
15 Bureau Current Industrial Reports Fertilizer Materials and Related Products annual and quarterly reports for 1997
16 through 2010 (U.S. Census Bureau 1998 through 2011), The Fertilizer Institute (TFI 2002) for 1993 through 1996,
17 and the United States International Trade Commission Interactive Tariff and Trade Data Web (U.S. ITC 2002) for
18 1990 through 1992 (see Table 4-24). Urea export data for 2014 are not yet publicly available and so 2013 data have
19 been used as proxy. Urea export data for 2013 were obtained from the Minerals Yearbook: Nitrogen (USGS 2015).
20 Urea export data for 1990 through 2012 were taken from U.S. Fertilizer Import/Exports from USDA Economic
21 Research Service Data Sets (U.S. Department of Agriculture 2012).
22 Table 4-24: Urea Production, Urea Applied as Fertilizer, Urea Imports, and Urea Exports (kt)
Year
Urea
Production
Urea Applied
as Fertilizer
Urea
Imports
Urea
Exports
1990
7,450
3,296
1,860
2010
2011
2012
2013
2014
5,122
5,430
5,220
5,480
5,480
5,152
5,589
5,762
5,921
6,156
6,631
5,860
6,944
6,470
6,470
152
207
336
330
330
23 Uncertainty and Time-Series Consistency
24 There is limited publicly-available data on the quantities of urea produced and consumed for non-agricultural
25 purposes. Therefore, the amount of urea used for non-agricultural purposes is estimated based on a balance that
26 relies on estimates of urea production, urea imports, urea exports, and the amount of urea used as fertilizer. The
27 primary uncertainties associated with this source category are associated with the accuracy of these estimates as well
28 as the fact that each estimate is obtained from a different data source. Because urea production estimates are no
29 longer available from the USGS, there is additional uncertainty associated with urea produced beginning in 2011.
30 There is also uncertainty associated with the assumption that all of the carbon in urea is released into the
31 environment as CC>2 during use.
32 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-25. CCh emissions
33 associated with urea consumption for non-agricultural purposes were estimated to be between 3.5 and 4.5 MMT
34 CO2 Eq. at the 95 percent confidence level. This indicates a range of approximately 12 percent below and 12
35 percent above the emission estimate of 4.0 MMT CCh Eq.
36 Table 4-25: Approach 2 Quantitative Uncertainty Estimates for COz Emissions from Urea
37 Consumption for Non-Agricultural Purposes (MMT COz Eq. and Percent)
Source
Gas
2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
Industrial Processes and Product Use 4-25
-------
(MMT CCh Eg.) (MMT CCh Eg.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
Urea Consumption for
Non-Agricultural CO2 4.0 3.5 4.5 -12% +12%
Purposes
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
1 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
2 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
3 above.
4 Recalculations Discussion
5 Production estimates for urea production and estimates for urea exports and imports for 2013 were updated using
6 information obtained from the Minerals Yearbook: Nitrogen (USGS 2015). Also, the amount of urea consumed for
7 agricultural purposes in the United States for 2013 was revised based on the most recent data obtained from the
8 Land Use, Land-Use Change, and Forestry chapter (see Table 6-25). These updates resulted in a decrease in the
9 emission estimate relative to the previous report of approximately 10 percent in 2013.
10
11
4.7 Nitric Acid Production (IPCC Source
Category 2B2)
12 Nitrous oxide (N2O) is emitted during the production of nitric acid (HNOs), an inorganic compound used primarily
13 to make synthetic commercial fertilizers. It is also a major component in the production of adipic acid—a feedstock
14 for nylon—and explosives. Virtually all of the nitric acid produced in the United States is manufactured by the
15 high-temperature catalytic oxidation of ammonia (EPA 1998). There are two different nitric acid production
16 methods: weak nitric acid and high-strength nitric acid. The first method utilizes oxidation, condensation, and
17 absorption to produce nitric acid at concentrations between 30 and 70 percent nitric acid. High-strength acid (90
18 percent or greater nitric acid) can be produced from dehydrating, bleaching, condensing, and absorption of the weak
19 nitric acid. The basic process technology for producing nitric acid has not changed significantly over time. Most
20 U.S. plants were built between 1960 and 2000. As of 2014, there are 34 active weak nitric acid production plants,
21 including one high-strength nitric acid production plant in U.S. (EPA 2010; EPA 2015).
22 During this reaction, N2O is formed as a byproduct and is released from reactor vents into the atmosphere.
23 Emissions from fuels consumed for energy purposes during the production of nitric acid are accounted for in the
24 Energy chapter.
25 Nitric acid is made from the reaction of ammonia (NH3) with oxygen (O2) in two stages. The overall reaction is:
26 4JV//3 +802 -> 4HN03 + 4H20
27 Currently, the nitric acid industry controls emissions of NO and NO2 (i.e., NOX). As such, the industry in the United
28 States uses a combination of non-selective catalytic reduction (NSCR) and selective catalytic reduction (SCR)
29 technologies. In the process of destroying NOX, NSCR systems are also very effective at destroying N2O. However,
30 NSCR units are generally not preferred in modern plants because of high energy costs and associated high gas
31 temperatures. NSCR systems were installed in nitric plants built between 1971 and 1977 with NSCRs installed at
32 approximately one-third of the weak acid production plants. U.S. facilities are using both tertiary (i.e., NSCR) and
33 secondary controls (i.e., alternate catalysts).
34 N2O emissions from this source were estimated to be 10.9 MMT CO2 Eq. (37 kt of N2O) in 2014 (see Table 4-26).
35 Emissions from nitric acid production have decreased by 10 percent since 1990, with the trend in the time series
36 closely tracking the changes in production. Emissions have decreased by 24 percent since 1997, the highest year of
37 production in the time series.
4-26 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 4-26: NzO Emissions from Nitric Acid Production (MMT COz Eq. and kt NzO)
Year MMT CCh Eg. kt N2O
1990 12.1 41
2010
2011
2012
2013
2014
11.5
10.9
10.5
10.7
10.9
39
37
35
36
37
2 Methodology
3 Emissions of N2O were calculated using the estimation methods provided by the 2006IPCC Guidelines (IPCC
4 2006) and country specific methods from N2O EPA's Greenhouse Gas Reporting Program (GHGRP). The 2006
5 IPCC Guidelines Tier 2 method was used to estimate emissions from nitric acid production for 1990 through 2009,
6 and a country-specific approach similar to the IPCC Tier 3 method was used to estimate N2O emissions for 2010
7 through 2014.
8 2010 through 2014
9 Process N2O emissions and nitric acid production data were obtained directly from EPA's GHGRP for 2010 through
10 2014 by aggregating reported facility-level data (EPA 2015). In the United States, all nitric acid facilities producing
11 weak nitric acid (30-70 percent in strength) are required to report annual greenhouse gas emissions data to EPA as
12 per the requirements of its GHGRP. As of 2014, there are 34 facilities that report to EPA, including the known
13 single high-strength nitric acid production facility in the United States (EPA 2015). All nitric acid (weak acid)
14 facilities are required to calculate process emissions using a site-specific emission factor developed through annual
15 performance testing under typical operating conditions or by directly measuring N2O emissions using monitoring
16 equipment. The high-strength nitric acid facility also reports N2O emissions associated with weak acid production
17 and this may capture all relevant emissions, pending additional further EPA research. More details on the
18 calculation and monitoring methods applicable to Nitric Acid facilities can be found under Subpart V: Nitric Acid
19 Production of the regulation, Part 98 ^5
20 1990 through 2009
21 Using the GHGRP data for 2010,16 country-specific N2O emission factors were calculated for nitric acid production
22 with abatement and without abatement (i.e., controlled and uncontrolled emission factors). The following 2010
23 emission factors were derived for production with abatement and without abatement: 3.3 kg N2O/metric ton HNOs
24 produced at plants using abatement technologies (e.g., tertiary systems such as NSCR systems) and 5.98 kg
25 N2O/metric ton HNOs produced at plants not equipped with abatement technology. Country-specific weighted
26 emission factors were derived by weighting these emission factors by percent production with abatement and
27 without abatement over time periods 1990-2008 and 2009. These weighted emission factors were used to estimate
28 N2O emissions from nitric acid production for years prior to the availability of EPA's GHGRP data (i.e., 1990
29 through 2008 and 2009). A separate weighted factor is included for 2009 due to data availability for that year. At
30 that time, EPA had initiated compilation of a nitric acid database to improve estimation of emissions from this
31 industry and obtained updated information on application of controls via review of permits and outreach with
15 Located at .
16 National N2O process emissions, national production, and national share of nitric acid production with abatement and without
abatement technology was aggregated from the GHGRP facility-level data for 2010-2014 (i.e., percent production with and
without abatement).
Industrial Processes and Product Use 4-27
-------
1 facilities and trade associations. The research indicated recent installation of abatement technologies at additional
2 facilities.
3 Based on the available data, it was assumed that emission factors for 2010 would be more representative of
4 operating conditions in 1990 through 2009 than more recent years. Initial review of historical data indicates that
5 percent production with and without abatement can change over time and also year over year due to changes in
6 application of facility-level abatement technologies, maintenance of abatement technologies, and also due to plant
7 closures and start-ups (EPA 2012, 2013; Desai 2012; CAR 2013). The installation dates of N2O abatement
8 technologies are not known at most facilities, but it is assumed that facilities reporting abatement technology use
9 have had this technology installed and operational for the duration of the time series considered in this report
10 (especially NSCRs).
11 The country-specific weighted N2O emission factors were used in conjunction with annual production to estimate
12 N2O emissions for 1990 through 2009, using the following equations:
13
14 EI = PI X EFweightedji
15 EFweightedii = [(%PCii X EFC} + (%PunC:i X EFunc}\
16 where,
17
18
19
20
21
22
23
24
25
Pi
EFweighted,i
%Pc,i
EFC
%Punc,i
EFmc
i
Annual N2O Emissions for year i (kg/yr)
Annual nitric acid production for year i (metric tons HNOs)
Weighted N2O emission factor for year i (kg N2O/metric ton HNOs)
Percent national production of HNOs with N2O abatement technology
N2O emission factor, with abatement technology (kg N2O/metric ton HNOs)
Percent national production of HNOs without N2O abatement technology (%)
N2O emission factor, without abatement technology (kg N2O/metric ton HNOs)
year from 1990 through 2009
26 • For 2009: Weighted N2O emission factor - 5.45 kg N2O/metric ton HNO3.
27 • For 1990-2008: Weighted N2O emission factor - 5.65 kg N2O/metric ton HNO3.
28
29 Nitric acid production data for the United States for 1990 through 2009 were obtained from the U.S. Census Bureau
30 (U.S. Census Bureau 2008, 2009, 2010a, 2010b) (see Table 4-27). Publicly-available information on plant-level
31 abatement technologies was used to estimate the shares of nitric acid production with and without abatement for
32 2008 and 2009 (EPA 2012, 2013; Desai 2012; CAR 2013). Publicly-available data on use of abatement technologies
33 were not available for 1990-2007. Therefore, the share of national production with and without abatement for 2008
34 was assumed to be constant for 1990 through 2007.
35 Table 4-27: Nitric Acid Production (kt)
Year
kt
1990 7,195
2010
2011
2012
2013
2014
7,444
7,606
7,453
7,572
7,656
4-28 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Uncertainty and Time-Series Consistency
2 Uncertainty associated with the parameters used to estimate N2O emissions includes that of production data, the
3 share of U.S. nitric acid production attributable to each emission abatement technology over the time series
4 (especially prior to 2010), and the associated emission factors applied to each abatement technology type. While
5 some information has been obtained through outreach with industry associations, limited information is available
6 over the time series (especially prior to 2010) for a variety of facility level variables, including plant specific
7 production levels, plant production technology (e.g., low, high pressure, etc.), and abatement technology type,
8 installation date of abatement technology, and accurate destruction and removal efficiency rates.
9 The results of this Approach 2 quantitative uncertainty analysis are summarized in Table 4-28. N2O emissions from
10 nitric acid production were estimated were estimated to be between 10.4 and 11.5 MMT CO2 Eq. at the 95 percent
11 confidence level. This indicates a range of approximately 5 percent below to 5 percent above the 2014 emissions
12 estimate of 10.9 MMT CO2Eq.
13 Table 4-28: Approach 2 Quantitative Uncertainty Estimates for NzO Emissions from Nitric
14 Acid Production (MMT COz Eq. and Percent)
Source Gas 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
(MMT CCh Eq.) (MMT CCh Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
Nitric Acid Production N2O 10.9 10.4 11.5 -5% +5%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
15 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
16 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
17 above.
18
19
4.8 Adipic Acid Production (IPCC Source
Category 2B3)
20 Adipic acid is produced through a two-stage process during which N2O is generated in the second stage. Emissions
21 from fuels consumed for energy purposes during the production of adipic acid are accounted for in the Energy
22 chapter. The first stage of manufacturing usually involves the oxidation of cyclohexane to form a cyclohexanone/
23 cyclohexanol mixture. The second stage involves oxidizing this mixture with nitric acid to produce adipic acid.
24 Nitrous oxide is generated as a byproduct of the nitric acid oxidation stage and is emitted in the waste gas stream
25 (Thiemens and Trogler 1991). The second stage is represented by the following chemical reaction:
26 (CH^CO (cyclohexanone) + (CH2)5CHOH (cyclohexanol) + wHN03
27 -> HOOC(CH2)4COOH(adipic acid) + xN20 + yH20
28 Process emissions from the production of adipic acid vary with the types of technologies and level of emission
29 controls employed by a facility. In 1990, two major adipic acid-producing plants had N2O abatement technologies
30 in place and, as of 1998, three major adipic acid production facilities had control systems in place (Reimer et al.
31 1999). One small plant, which last operated in April 2006 and represented approximately two percent of production,
32 did not control for N2O (VA DEQ 2009; ICIS 2007; VA DEQ 2006). In 2014, catalytic reduction, non-selective
33 catalytic reduction (NSCR) and thermal reduction abatement technologies were applied as N2O abatement measures
34 at adipic acid facilities (EPA 2015).
35 Worldwide, only a few adipic acid plants exist. The United States, Europe, and China are the major producers. In
36 2014, the United States had two companies with a total of three adipic acid production facilities (two in Texas and
37 one in Florida), all of which were operational (EPA 2015). The United States accounts for the largest share of global
Industrial Processes and Product Use 4-29
-------
1 adipic acid production capacity (30 percent), followed by the European Union (29 percent) and China (22 percent)
2 (SEI 2010). Adipic acid is a white crystalline solid used in the manufacture of synthetic fibers, plastics, coatings,
3 urethane foams, elastomers, and synthetic lubricants. Commercially, it is the most important of the aliphatic
4 dicarboxylic acids, which are used to manufacture polyesters. Eighty-four percent of all adipic acid produced in the
5 United States is used in the production of nylon 6,6; 9 percent is used in the production of polyester polyols; 4
6 percent is used in the production of plasticizers; and the remaining 4 percent is accounted for by other uses,
7 including unsaturated polyester resins and food applications (ICIS 2007). Food grade adipic acid is used to provide
8 some foods with a "tangy" flavor (Thiemens and Trogler 1991).
9 Nitrous oxide emissions from adipic acid production were estimated to be 5.4 MMT CO2 Eq. (18 kt N2O) in 2014
10 (see Table 4-29). National adipic acid production has increased by approximately 36 percent over the period of
11 1990 through 2014, to approximately 1,025,000 metric tons (ACC 2015). Over the period 1990 through 2014,
12 emissions have been reduced by 64 percent due to both the widespread installation of pollution control measures in
13 the late 1990s and plant idling in the late 2000s. In April 2006, the smallest of the four facilities ceased production
14 of adipic acid (VA DEQ 2009); furthermore, one of the major adipic acid production facilities was not operational in
15 2009 or 2010 (Desai 2010). All three remaining facilities were in operation in 2014. Very little information on
16 annual trends in the activity data exist for adipic acid.
17 Table 4-29: NzO Emissions from Adipic Acid Production (MMT COz Eq. and kt NzO)
Year MMT CCh Eq. kt N2O
1990 15.2
2010
2011
2012
2013
2014
4.2
10.2
5.5
4.0
5.4
14
34
19
13
18
18 Methodology
19 Emissions are estimated using both Tier 2 and Tier 3 methods consistent with the 2006IPCC Guidelines (IPCC
20 2006). Due to confidential business information, plant names are not provided in this section. Therefore, the four
21 adipic acid-producing facilities will be referred to as Plants 1 through 4. Plant 4 was closed in April 2006. Overall,
22 as noted above, the three plants that are currently operating facilities use abatement equipment. Plants 1 and 2
23 employ catalytic destruction and Plant 3 employs thermal destruction.
24 2010 through 2014
25 All emission estimates for 2010 through 2014 were obtained through analysis of EPA's GHGRP data (EPA 2014
26 through 2015), which is consistent with the 2006 IPCC Guidelines (IPCC 2006) Tier 3 method. Facility-level
27 greenhouse gas emissions data were obtained from the GHGRP for the years 2010 through 2014 (EPA 2014 through
28 2015) and aggregated to national N2O emissions. Consistent with IPCC Tier 3 methods, all adipic acid production
29 facilities are required to calculate emissions using a facility-specific emission factor developed through annual
30 performance testing under typical operating conditions or by directly measuring N2O emissions using monitoring
31 equipment. More information on the monitoring methods for process N2O emissions applicable to adipic acid
32 production facilities under Subpart E can be found in the electronic code of federal regulations.17
17See.
4-30 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
i 1990 through 2009
2 For years prior to EPA's GHGRP reporting, for both Plants 1 and 2, 1990 to 2009 emission estimates were obtained
3 directly from the plant engineer and account for reductions due to control systems in place at these plants during the
4 time series. These prior estimates are considered confidential business information and hence are not published
5 (Desai 2010). These estimates were based on continuous process monitoring equipment installed at the two
6 facilities. In 2009 and 2010, no adipic acid production occurred at Plant 1 per reporting to EPA's GHGRP (EPA
7 2012; Desai 201 Ib).
8 For the Plant 4, 1990 through 2009 N2O emissions were estimated using the following Tier 2 equation from the
9 2006IPCC Guidelines until shutdown of the plant in 2006:
10 Eaa = Qaa X EFaa X (1 - [DF X UF])
11 where,
12 Eaa = N2O emissions from adipic acid production, metric tons
13 Qaa = Quantity of adipic acid produced, metric tons
14 EFaa = Emission factor, metric ton N2O/metric ton adipic acid produced
15 DF N2O destruction factor
16 UF = Abatement system utility factor
17 The adipic acid production is multiplied by an emission factor (i.e., N2O emitted per unit of adipic acid produced),
18 which has been estimated, based on experiments that the reaction stoichiometry for N2O production in the
19 preparation of adipic acid, to be approximately 0.3 metric tons of N2O per metric ton of product (IPCC 2006). The
20 "N2O destruction factor" in the equation represents the percentage of N2O emissions that are destroyed by the
21 installed abatement technology. The "abatement system utility factor" represents the percentage of time that the
22 abatement equipment operates during the annual production period. No abatement equipment was installed at the
23 Inolex/Allied Signal facility, which last operated in April 2006 (VA DEQ 2009). Plant-specific production data for
24 this facility were obtained across the time series from 1990 through 2006 from the Virginia Department of
25 Environmental Quality (VA DEQ 2010). The plant-specific production data were then used for calculating
26 emissions as described above.
27 For Plant 3, 2005 through 2009 emissions were obtained directly from the plant (Desai 201 la). For 1990 through
28 2004, emissions were estimated using plant-specific production data and the IPCC factors as described above for
29 Plant 4. Plant-level adipic acid production for 1990 through 2003 was estimated by allocating national adipic acid
30 production data to the plant level using the ratio of known plant capacity to total national capacity for all U. S. plants
31 (ACC 2015; CMR 2001, 1998; CW 1999; C&EN 1992 through 1995). For 2004, actual plant production data were
32 obtained and used for emission calculations (CW 2005).
33 Plant capacities for 1990 through 1994 were obtained from Chemical & Engineering News, "Facts and Figures" and
34 "Production of Top 50 Chemicals" (C&EN 1992 through 1995). Plant capacities for 1995 and 1996 were kept the
35 same as 1994 data. The 1997 plant capacities were taken from Chemical Market Reporter, "Chemical Profile:
36 Adipic Acid" (CMR 1998). The 1998 plant capacities for all four plants and 1999 plant capacities for three of the
37 plants were obtained from Chemical Week, Product Focus: Adipic Acid/Adiponitrile (CW 1999). Plant capacities
38 for2000for three of the plants were updated using Chemical Market Reporter, " Chemical Profile: Adipic Acid"
39 (CMR 2001). For 2001 through 2003, the plant capacities for three plants were kept the same as the year 2000
40 capacities. Plant capacity for 1999 to 2003 for the one remaining plant was kept the same as 1998.
41 National adipic acid production data (see Table 4-30) from 1990 through 2014 were obtained from the American
42 Chemistry Council (ACC 2015).
43 Table 4-30: Adipic Acid Production (kt)
Year kt
1990
Industrial Processes and Product Use 4-31
-------
2010 720
2011 840
2012 950
2013 980
2014 1,025
i Uncertainty and Time-Series Consistency
2 Uncertainty associated with N2O emission estimates includes the methods used by companies to monitor and
3 estimate emissions. While some information has been obtained through outreach with facilities, limited information
4 is available over the time series on these methods, abatement technology destruction and removal efficiency rates
5 and plant specific production levels.
6 The results of this Approach 2 quantitative uncertainty analysis are summarized in Table 4-31. Nitrous oxide
7 emissions from adipic acid production for 2014 were estimated to be between 5.2 and 5.6 MMT CCh Eq. at the 95
8 percent confidence level. These values indicate a range of approximately 4 percent below to 4 percent above the
9 2014 emission estimate of 5.4 MMT CO2 Eq.
10 Table 4-31: Approach 2 Quantitative Uncertainty Estimates for NzO Emissions from Adipic
11 Acid Production (MMT COz Eq. and Percent)
„ „ 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
6 (MMT CCh Eq.) (MMT CCh Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
Adipic Acid Production N2O 5.4 5.2 5.6 -4% +4%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
12 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
13 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
14 above.
15 4.9 Silicon Carbide Production and
16 Consumption (IPCC Source Category 2B5)
17 Carbon dioxide (CCh) and methane (CH4) are emitted from the production of silicon carbide (SiC), a material used
18 as an industrial abrasive. Silicon carbide is produced for abrasive, metallurgical, and other non-abrasive
19 applications in the United States. Production for metallurgical and other non-abrasive applications is not available
20 and therefore both CCh and CH4 estimates are based solely upon production estimates of silicon carbide for abrasive
21 applications. Emissions from fuels consumed for energy purposes during the production of silicon carbide are
22 accounted for in the Energy chapter.
23 To produce SiC, silica sand or quartz (SiCh) is reacted with carbon in the form of petroleum coke. A portion (about
24 35 percent) of the carbon contained in the petroleum coke is retained in the SiC. The remaining carbon is emitted as
25 CO2, CH4, or CO. The overall reaction is shown below (but in practice it does not proceed according to
26 stoichiometry):
27 Si02 + 3C -> SiC + 2CO (+ 02 -> 2C02)
28 Carbon dioxide is also emitted from the consumption of SiC for metallurgical and other non-abrasive applications.
29 Markets for manufactured abrasives, including SiC, are heavily influenced by activity in the U.S. manufacturing
30 sector, especially in the aerospace, automotive, furniture, housing, and steel manufacturing sectors. The USGS
4-32 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 reports that a portion (approximately 50 percent) of SiC is used in metallurgical and other non-abrasive applications,
2 primarily in iron and steel production (USGS 2006a). As a result of the economic downturn in 2008 and 2009,
3 demand for SiC decreased in those years. Low cost imports, particularly from China, combined with high relative
4 operating costs for domestic producers, continue to put downward pressure on the production of SiC in the United
5 States. However, demand for SiC consumption in the United States has recovered somewhat from its low in 2009
6 (USGS 2012a). Silicon carbide is manufactured at a single facility located in Illinois (USGS 2015a).
7 Carbon dioxide emissions from SiC production and consumption in 2014 were 0.2 MMT CO2 Eq. (173 kt CCh).
8 Approximately 53 percent of these emissions resulted from SiC production while the remainder resulted from SiC
9 consumption. Methane emissions from SiC production in 2014 were 0.01 MMT CO2 Eq. (0.4 kt CH4) (see Table
10 4-32: and Table 4-33). Emissions have fluctuated in recent years, but 2014 emissions are only about 46 percent of
11 emissions in 1990.
12 Table 4-32: COz and CH4 Emissions from Silicon Carbide Production and Consumption (MMT
13 COz Eq.)
15
Year
C02
CH4
Total
1990
0.4
+
0.4
2005
0.2
+
0.2
2010
0.2
+
0.2
2011
0.2
+
0.2
2012
0.2
+
0.2
2013
0.2
+
0.2
2014
0.2
+
0.2
+ Does not exceed 0.05 MMT CO2 Eq.
14 Table 4-33: COz and CH4 Emissions from Silicon Carbide Production and Consumption (kt)
Year 1990 2005 2010 2011 2012 2013 2014
C02 375 219 181 170 158 169 173
CH4 1_H
+ Does not exceed 0.5 kt.
Methodology
16 Emissions of CO2 and CH4 from the production of SiC were calculated using the Tier 1 method provided by the
17 2006 IPCC Guidelines (IPCC 2006). Annual estimates of SiC production were multiplied by the appropriate
18 emission factor, as shown below:
19 ESCjC02 = EFSCjC02 X Qsc
/I metric ton
20 EscrH4 = EFscrH4 x Qsc x
1000 kg
21 where,
22 Esc,co2 = CO2 emissions from production of SiC, metric tons
23 EFsc,co2 = Emission factor for production of SiC, metric ton CCh/metric ton SiC
24 Qsc = Quantity of SiC produced, metric tons
25 Esc,cH4 = CH4 emissions from production of SiC, metric tons
26 EFsc,cH4 = Emission factor for production of SiC, kilogram CH4/metric ton SiC
27
28 Emission factors were taken from the 2006 IPCC Guidelines (IPCC 2006):
29 • 2.62 metric tons CCVmetric ton SiC
30 • 11.6 kg CH4/metric ton SiC
Industrial Processes and Product Use 4-33
-------
1 Emissions of CO2 from silicon carbide consumption for metallurgical uses were calculated by multiplying the
2 annual utilization of SiC for metallurgical uses (reported annually in the USGS Minerals Yearbook: Silicon) by the
3 carbon content of SiC (31.5 percent), which was determined according to the molecular weight ratio of SiC.
4 Emissions of CC>2 from silicon carbide consumption for other non-abrasive uses were calculated by multiplying the
5 annual SiC consumption for non-abrasive uses by the carbon content of SiC (31.5 percent). The annual SiC
6 consumption for non-abrasive uses was calculated by multiplying the annual SiC consumption (production plus net
7 imports) by the percent used in metallurgical and other non-abrasive uses (50 percent) (USGS 2006a) and then
8 subtracting the SiC consumption for metallurgical use.
9 Production data for 1990 through 2013 were obtained from the Minerals Yearbook: Manufactured Abrasives (USGS
10 1991a through 2015b). Production data for 2014 were obtained from the Minerals Industry Surveys: Abrasives
11 (Manufactured) (USGS2015a). Silicon carbide consumption by major end use for 1990 through 2012 were
12 obtained from the Minerals Yearbook: Silicon (USGS 1991b through 2013) (see Table 4-34). Silicon carbide
13 consumption data for 2013 and 2014 are not yet publicly available and so 2012 data were used as proxy. Net imports
14 for the entire time series were obtained from the U.S. Census Bureau (2005 through 2015).
15 Table 4-34: Production and Consumption of Silicon Carbide (Metric Tons)
16
Year Production Consumption
1990 105,000 172,465
2010
2011
2012
2013
2014
35,000
35,000
35,000
35,000
35,000
154,540
136,222
114,265
134,055
140,723
Uncertainty and Time-Series Consistency
17 There is uncertainty associated with the emission factors used because they are based on stoichiometry as opposed to
18 monitoring of actual SiC production plants. An alternative would be to calculate emissions based on the quantity of
19 petroleum coke used during the production process rather than on the amount of silicon carbide produced. However,
20 these data were not available. For CH4, there is also uncertainty associated with the hydrogen-containing volatile
21 compounds in the petroleum coke (IPCC 2006). There is also uncertainty associated with the use or destruction of
22 methane generated from the process in addition to uncertainty associated with levels of production, net imports,
23 consumption levels, and the percent of total consumption that is attributed to metallurgical and other non-abrasive
24 uses.
25 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-35. Silicon carbide
26 production and consumption CCh emissions were estimated to be between 9 percent below and 9 percent above the
27 emission estimate of 0.17 MMT CC>2 Eq. at the 95 percent confidence level. Silicon carbide production CH4
28 emissions were estimated to be between 9 percent below and 10 percent above the emission estimate of 0.01 MMT
29 CO2 Eq. at the 95 percent confidence level.
30 Table 4-35: Approach 2 Quantitative Uncertainty Estimates for CH4 and COz Emissions from
31 Silicon Carbide Production and Consumption (MMT COz Eq. and Percent)
„ „ 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
e (MMT CCh Eq.) (MMT CCh Eq.) (%)
Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
Silicon Carbide Production ^ Q^ ^ Q^ _9% +9%
and Consumption
4-34 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Silicon Carbide Production CH4 + + + -9% +10%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
+ Does not exceed 0.05 MMT CO2 Eq.
1 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
2 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
3 above.
15
16
Planned Improvements
5 Future improvements involve continuing to evaluate and analyze data reported under EPA's GHGRP to improve the
6 emission estimates for the Silicon Carbide Production and Consumption source category. Particular attention will be
7 made to ensure time series consistency of the emission estimates presented in future Inventory reports, consistent
8 with IPCC and UNFCCC guidelines. This is required as the facility-level reporting data from EPA's GHGPJ3, with
9 the program's initial requirements for reporting of emissions in calendar year 2010, are not available for all
10 inventory years (i.e., 1990 through 2009) as required for this Inventory. In implementing improvements and
11 integration of data from EPA's GHGPJ3, the latest guidance from the IPCC on the use of facility-level data in
12 national inventories will be relied upon.18 In addition, improvements will involve continued research to determine if
13 calcium carbide production and consumption data are available for the United States. If these data are available,
14 calcium carbide emission estimates will be included in this source category.
4.10 Titanium Dioxide Production (IPCC
Source Category 2B6)
17 Titanium dioxide (TiCh) is manufactured using one of two processes: the chloride process and the sulfate process.
18 The chloride process uses petroleum coke and chlorine as raw materials and emits process-related CCh. Emissions
19 from fuels consumed for energy purposes during the production of titanium dioxide are accounted for in the Energy
20 chapter. The chloride process is based on the following chemical reactions:
21 2FeTi03 +7C12 + 3C -> 2TiCl4 + 2FeCl3 + 3C02
22 2TiCl4 +202 -> 2TW2 + 4C/2
23 The sulfate process does not use petroleum coke or other forms of carbon as a raw material and does not emit CO2.
24 The carbon in the first chemical reaction is provided by petroleum coke, which is oxidized in the presence of the
25 chlorine and FeTiOs (rutile ore) to form CO2. Since 2004, all TiCh produced in the United States has been produced
26 using the chloride process, and a special grade of "calcined" petroleum coke is manufactured specifically for this
27 purpose.
28 The principal use of TiCh is as a pigment in white paint, lacquers, and varnishes; it is also used as a pigment in the
29 manufacture of plastics, paper, and other products. In 2014, U.S. TiCh production totaled 1,310,000 metric tons
30 (USGS 2015a). There were a total 6 plants producing TiCh in the United States—2 located in Mississippi, and single
31 plants located in Delaware, Louisiana, Ohio, and Tennessee.
32 Emissions of CC>2 from titanium dioxide production in 2014 were estimated to be 1.8 MMT CC>2 Eq. (1,755 kt CCh),
33 which represents an increase of 47 percent since 1990 (see Table 4-36).
18
See.
Industrial Processes and Product Use 4-35
-------
Table 4-36: COz Emissions from Titanium Dioxide (MMT COz Eq. and kt)
Year MMT CCh Eg.
1990
kt
1,195
2010
2011
2012
2013
2014
1.8
1.7
1.5
1.7
1.8
1,769
1,729
1,528
1,715
1,755
J
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Methodology
Emissions of CCh from TiCh production were calculated by multiplying annual national TiCh production by chloride
process-specific emission factors using a Tier 1 approach provided in 2006IPCC Guidelines (IPCC 2006). The Tier
1 equation is as follows:
Etd = EFtd x Q,
td
where,
EFtd
Qtd
CO2 emissions from TiCh production, metric tons
Emission factor (chloride process), metric ton COVmetric
Quantity of TiCh produced
Data were obtained for the total amount of TiCh produced each year. For years prior to 2004, it was assumed that
TiO2 was produced using the chloride process and the sulfate process in the same ratio as the ratio of the total U.S.
production capacity for each process. As of 2004, the last remaining sulfate process plant in the United States
closed; therefore, 100 percent of post-2004 production uses the chloride process (USGS 2005). The percentage of
production from the chloride process is estimated at 100 percent since 2004. An emission factor of 1.34 metric tons
COVmetric ton TiCh was applied to the estimated chloride-process production (IPCC 2006). It was assumed that all
TiO2 produced using the chloride process was produced using petroleum coke, although some TiO2 may have been
produced with graphite or other carbon inputs.
The emission factor for the TiO2 chloride process was taken from the 2006 IPCC Guidelines (IPCC 2006).
Titanium dioxide production data and the percentage of total TiO2 production capacity that is chloride process for
1990 through 2013 (see Table 4-37:) were obtained through the Minerals Yearbook: Titanium Annual Report (USGS
1991 through 2015b). Production data for 2014 was obtained from the Minerals Commodity Summary: Titanium
and Titanium Dioxide (USGS 2015a). Data on the percentage of total TiC>2 production capacity that is chloride
process were not available for 1990 through 1993, so data from the 1994 USGS Minerals Yearbook were used for
these years. Because a sulfate process plant closed in September 2001, the chloride process percentage for 2001 was
estimated based on a discussion with Joseph Gambogi (2002). By 2002, only one sulfate process plant remained
online in the United States and this plant closed in 2004 (USGS 2005).
Table 4-37: Titanium Dioxide Production (kt)
Year
kt
1990
2010
2011
2012
2013
1,320
1,290
1,140
1,280
4-36 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
2014 1,310
i Uncertainty and Time-Series Consistency
2 Each year, USGS collects titanium industry data for titanium mineral and pigment production operations. If TiO2
3 pigment plants do not respond, production from the operations is estimated on the basis of prior year production
4 levels and industry trends. Variability in response rates varies from 67 to 100 percent of TiO2 pigment plants over
5 the time series.
6 Although some TiO2 may be produced using graphite or other carbon inputs, information and data regarding these
7 practices were not available. Titanium dioxide produced using graphite inputs, for example, may generate differing
8 amounts of CO2 per unit of TiO2 produced as compared to that generated through the use of petroleum coke in
9 production. While the most accurate method to estimate emissions would be to base calculations on the amount of
10 reducing agent used in each process rather than on the amount of TiO2 produced, sufficient data were not available
11 to do so.
12 As of 2004, the last remaining sulfate-process plant in the United States closed. Since annual TiO2 production was
13 not reported by USGS by the type of production process used (chloride or sulfate) prior to 2004 and only the
14 percentage of total production capacity by process was reported, the percent of total TiO2 production capacity that
15 was attributed to the chloride process was multiplied by total TiO2 production to estimate the amount of TiO2
16 produced using the chloride process. Finally, the emission factor was applied uniformly to all chloride-process
17 production, and no data were available to account for differences in production efficiency among chloride-process
18 plants. In calculating the amount of petroleum coke consumed in chloride-process TiO2 production, literature data
19 were used for petroleum coke composition. Certain grades of petroleum coke are manufactured specifically for use
20 in the TiO2 chloride process; however, this composition information was not available.
21 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-38. Titanium dioxide
22 consumption CO2 emissions were estimated to be between 1.5 and 2.0 MMT CO2 Eq. at the 95 percent confidence
23 level. This indicates a range of approximately 12 percent below and 13 percent above the emission estimate of 1.8
24 MMT CO2 Eq.
25 Table 4-38: Approach 2 Quantitative Uncertainty Estimates for COz Emissions from Titanium
26 Dioxide Production (MMT COz Eq. and Percent)
^ „ 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
6 (MMT CCh Eq.) (MMT CCh Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
Titanium Dioxide Production CO2 1.8 1.5 2.0 -12% +13%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
27 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
28 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
29 above.
30
Recalculations Discussion
31 Production data for 2013 were updated relative to the previous Inventory based on recently published data in the
32 USGS Minerals Yearbook: Titanium 2013 (USGS 2015b). This resulted in a 7 percent increase in 2013 CO2
33 emissions from TiO2 production relative to the previous report.
34 Planned Improvements
35 Pending resources, a potential improvement to the Inventory estimates for this source category would include the
36 derivation of country-specific emission factors, based on annual data reported under EPA's GHGRP for 2010
37 through 2014 (i.e., aggregated emissions and titanium production). Information on TiO2 production is collected by
Industrial Processes and Product Use 4-37
-------
1 EPA's GHGRP for all facilities for years 2010 through 2014 and would also have to be assessed against criteria
2 EPA has established to publish aggregated confidential business information (CBI) reported under EPA's GHGRP.
3 In order to provide estimates for the entire time series (i.e., 1990 through 2009), the applicability of more recent
4 GHGRP data to previous years' estimates will need to be evaluated, and additional data that could be utilized in the
5 calculations for this source category may need to be researched. In implementing improvements and integration of
6 data from EPA's GHGRP, the latest guidance from the IPCC on the use of facility-level data in national inventories
7 will be relied upon.19
8 In addition, planned improvements include researching the significance of titanium-slag production in electric
9 furnaces and synthetic-rutile production using the Becher process in the United States. Significant use of these
10 production processes will be included in future Inventories.
11
12
4.11 Soda Ash Production and Consumption
(IPCC Source Category 2B7)
13 Carbon dioxide is generated as a byproduct of calcining trona ore to produce soda ash, and is eventually emitted into
14 the atmosphere. In addition, CCh may also be released when soda ash is consumed. Emissions from fuels
15 consumed for energy purposes during the production and consumption of soda ash are accounted for in the Energy
16 sector.
17 Calcining involves placing crushed trona ore into a kiln to convert sodium bicarbonate into crude sodium carbonate
18 that will later be filtered into pure soda ash. The emission of CCh during trona-based production is based on the
19 following reaction:
20 2Na2C03 • NaHC03 • 2H20 (Trona) -> 3Na2C03(Soda Ash) + 5H20 + C02
21 Soda ash (sodium carbonate, Na2COs) is a white crystalline solid that is readily soluble in water and strongly
22 alkaline. Commercial soda ash is used as a raw material in a variety of industrial processes and in many familiar
23 consumer products such as glass, soap and detergents, paper, textiles, and food. (Emissions from soda ash used in
24 glass production are reported under IPCC Source Category 2A3. Glass production is its own sub-category and
25 historical soda ash consumption figures have been adjusted to reflect this change.) After glass manufacturing, soda
26 ash is used primarily to manufacture many sodium-based inorganic chemicals, including sodium bicarbonate,
27 sodium chromates, sodium phosphates, and sodium silicates (USGS 2015b). Internationally, two types of soda ash
28 are produced, natural and synthetic. The United States produces only natural soda ash and is second only to China
29 in total soda ash production. Trona is the principal ore from which natural soda ash is made.
30 The United States represents about one-fourth of total world soda ash output. Only two states produce natural soda
31 ash: Wyoming and California. Of these two states, only net emissions of CO2 from Wyoming were calculated due
32 to specifics regarding the production processes employed in the state.20 Based on preliminary 2014 reported data,
33 the estimated distribution of soda ash by end-use in 2014 (excluding glass production) was chemical production, 56
34 percent; soap and detergent manufacturing, 13 percent; distributors, 10 percent; flue gas desulfurization, 8 percent;
35 other uses, 8 percent; pulp and paper production, 2 percent; and water treatment, 2 percent (USGS 2015a).
19 See.
20 In California, soda ash is manufactured using sodium carbonate-bearing brines instead of trona ore. To extract the sodium
carbonate, the complex brines are first treated with CCh in carbonation towers to convert the sodium carbonate into sodium
bicarbonate, which then precipitates from the brine solution. The precipitated sodium bicarbonate is then calcined back into
sodium carbonate. Although CCh is generated as a byproduct, the CCh is recovered and recycled for use in the carbonation stage
and is not emitted. A third state, Colorado, produced soda ash until the plant was idled in 2004. The lone producer of sodium
bicarbonate no longer mines trona in the state. For a brief time, sodium bicarbonate was produced using soda ash feedstocks
mined in Wyoming and shipped to Colorado. Prior to 2004, because the trona was mined in Wyoming, the production numbers
given by the USGS included the feedstocks mined in Wyoming and shipped to Colorado. In this way, the sodium bicarbonate
production that took place in Colorado was accounted for in the Wyoming numbers.
4-38 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 U.S. natural soda ash is competitive in world markets because the majority of the world output of soda ash is made
2 synthetically. Although the United States continues to be a major supplier of world soda ash, China, which
3 surpassed the United States in soda ash production in 2003, is the world's leading producer.
4 In 2014, CO2 emissions from the production of soda ash from trona were approximately 1.7 MMT CCh Eq. (1,685 kt
5 CO2). Soda ash consumption in the United States generated 1.1 MMT CO2Eq. (1,143 kt CO2) in2014. Total
6 emissions from soda ash production and consumption in 2014 were 2.8 MMT CCh Eq. (2,827 kt CCh) (see Table
7 4-39 and Table 4-40).
8 Total emissions in 2014 increased by approximately 1 percent from emissions in 2013, and have stayed
9 approximately the same as the 1990 levels.
10 Emissions have remained relatively constant over the time series with some fluctuations since 1990. In general,
11 these fluctuations were related to the behavior of the export market and the U.S. economy. The U.S. soda ash
12 industry continued a trend of increased production and value in 2014 since experiencing a decline in domestic and
13 export sales caused by adverse global economic conditions in 2009. The annual average unit value of soda ash set a
14 record high in 2013, and soda ash exports increased as well, accounting for 56 percent of total production (USGS
15 2015b).
16 Table 4-39: COz Emissions from Soda Ash Production and Consumption Not Associated with
17 Glass Manufacturing (MMT COz Eq.)
Year
Production Consumption
Total
1990
1.4
1.4
2010
2011
2012
2013
2014
1.5
1.6
1.7
1.7
1.7
1.1 2.7
1.1 2.7
1.1 2.8
1.1 2.8
1.1 2.8
Note: Totals may not sum due to independent rounding.
18 Table 4-40: COz Emissions from Soda Ash Production and Consumption Not Associated with
19 Glass Manufacturing (kt)
Year
Production Consumption
Total
1990
1,431
1,390
2,822
2010
2011
2012
2013
2014
1,548
1,607
1,665
1,694
1,685
1,148
1,105
1,097
1,109
1,143
2,697
2,712
2,763
2,804
2,827
Note: Totals may not sum due to independent rounding.
20
21
22
23
Methodology
During the production process, trona ore is calcined in a rotary kiln and chemically transformed into a crude soda
ash that requires further processing. Carbon dioxide and water are generated as byproducts of the calcination
process. Carbon dioxide emissions from the calcination of trona can be estimated based on the chemical reaction
Industrial Processes and Product Use 4-39
-------
1 shown above. Based on this formula, which is consistent with an IPCC Tier 1 approach, approximately 10.27 metric
2 tons of trona are required to generate one metric ton of CO2, or an emission factor of 0.0974 metric tons CO2 per
3 metric ton trona (IPCC 2006). Thus, the 17.3 million metric tons of trona mined in 2014 for soda ash production
4 (USGS 2015a) resulted in CO2 emissions of approximately 1.7 MMT CO2 Eq. (1,685 kt).
5 Once produced, most soda ash is consumed in chemical and soap production, with minor amounts in pulp and paper,
6 flue gas desulfurization, and water treatment (excluding soda ash consumption for glass manufacturing). As soda
7 ash is consumed for these purposes, additional CO2 is usually emitted. In these applications, it is assumed that one
8 mole of carbon is released for every mole of soda ash used. Thus, approximately 0.113 metric tons of carbon (or
9 0.415 metric tons of CO2) are released for every metric ton of soda ash consumed.
10 The activity data for trona production and soda ash consumption (see Table 4-41) for 1990 to 2014 were obtained
11 from USGS Minerals Yearbook for Soda Ash (1994 through 2015b) and USGS Mineral Industry Surveys/or Soda
12 Ash (USGS 2015a). Soda ash production and consumption data were collected by the USGS from voluntary surveys
13 of the U. S. soda ash industry.
14 Table 4-41: Soda Ash Production and Consumption Not Associated with Glass Manufacturing
15 (kt)
16
Year Production3 Consumption1*
1990 14,700 3,351
2010
2011
2012
2013
2014
15,900
16,500
17,100
17,400
17,300
2,768
2,663
2,645
2,674
2,754
a Soda ash produced from trona ore only.
b Soda ash consumption is sales reported by
producers which exclude imports. Historically,
imported soda ash is less than 1 percent of the
total U.S. consumption (Kostick 2012).
Uncertainty and Time-Series Consistency
17 Emission estimates from soda ash production have relatively low associated uncertainty levels in that reliable and
18 accurate data sources are available for the emission factor and activity data. Soda ash production data was collected
19 by the USGS from voluntary surveys. A survey request was sent to each of the five soda ash producers, all of which
20 responded, representing 100 percent of the total production data (USGS 2014a). One source of uncertainty is the
21 purity of the trona ore used for manufacturing soda ash. The emission factor used for this estimate assumes the ore
22 is 100 percent pure, and likely overestimates the emissions from soda ash manufacture. The average water-soluble
23 sodium carbonate-bicarbonate content for ore mined in Wyoming ranges from 85.5 to 93.8 percent (USGS
24 1995).The primary source of uncertainty, however, results from the fact that emissions from soda ash consumption
25 are dependent upon the type of processing employed by each end-use. Specific emission factors for each end-use
26 are not available, so a Tier 1 default emission factor is used for all end uses. Therefore, there is uncertainty
27 surrounding the emission factors from the consumption of soda ash.
28 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-42. Soda Ash Production
29 and Consumption CO2 emissions were estimated to be between 2.5 and 2.9 MMT CO2 Eq. at the 95 percent
30 confidence level. This indicates a range of approximately 7 percent below and 6 percent above the emission
31 estimate of 2.8 MMT CO2 Eq.
4-40 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 4-42: Approach 2 Quantitative Uncertainty Estimates for COz Emissions from Soda Ash
2 Production and Consumption (MMT COz Eq. and Percent)
s „ 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
6 (MMTCChEq.) (MMT CCh Eq.) (%)
11
23
24
Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
SodaAshProduction ^ 2g ^ ^ _J% +6%
and Consumption
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
3 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
4 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
5 above.
6 Recalculation Discussion
7 During the development of the current Inventory, an error in the transcription of the 2006IPCC Guidelines default
8 trona production emission factor was identified. This error was corrected in the current Inventory and resulted in a
9 slight change of emissions over the entire time series (approximately 3 percent), compared with the previous
10 Inventory.
Planned Improvements
12 In future Inventory reports, soda ash consumed for other uses of carbonates will be extracted from the current soda
13 ash consumption emission estimates and included under those sources or other process uses of carbonates (IPCC
14 Category 2A4).
15 EPA will continue to analyze and assess uses of facility-level data from EPA's GHGRP to improve the emission
16 estimates for Soda Ash Production source category. Particular attention will be made to ensure time series
17 consistency of the emissions estimates presented in future Inventory reports, consistent with IPCC and UNFCCC
18 guidelines. This is required as the facility-level reporting data from EPA's GHGRP, with the program's initial
19 requirements for reporting of emissions in calendar year 2010, are not available for all inventory years (i.e., 1990
20 through 2009) as required for this Inventory. In implementing improvements and integration of data from EPA's
21 GHGRP, the latest guidance from the IPCC on the use of facility-level data in national inventories will be relied
22 upon.21
4.12 Petrochemical Production (IPCC Source
Category 2B8)
25 The production of some petrochemicals results in the release of small amounts of CC>2 and CH4 emissions.
26 Petrochemicals are chemicals isolated or derived from petroleum or natural gas. Carbon dioxide emissions from the
27 production of acrylonitrile, carbon black, ethylene, ethylene dichloride, ethylene oxide and methanol; and CH4
28 emissions from the production of methanol and acrylonitrile are presented here and reported under IPCC Source
29 Category 2B5. The petrochemical industry uses primary fossil fuels (i.e., natural gas, coal, petroleum, etc.) for non-
30 fuel purposes in the production of carbon black and other petrochemicals. Emissions from fuels and feedstocks
21 See.
Industrial Processes and Product Use 4-41
-------
1 transferred out of the system for use in energy purposes (e.g., such as indirect or direct process heat or steam
2 production) are currently accounted for in the Energy sector.
3 Worldwide more than 90 percent of acrylonitrile (vinyl cyanide, CsH3N) is made by way of direct ammoxidation of
4 propylene with ammonia (NH3) and oxygen over a catalyst. This process is referred to as the SOHIO process,
5 after the Standard Oil Company of Ohio (SOHIO) (IPCC 2006). The primary use of acrylonitrile is as the raw
6 material for the manufacture of acrylic and modacrylic fibers. Other major uses include the production of plastics
7 (acrylonitrile-butadiene-styrene [ABS] and styrene-acrylonitrile [SAN]), nitrile rubbers, nitrile barrier resins,
8 adiponitrile and acrylamide. All U.S. acrylonitrile facilities use the SOHIO process (AN 2014). The SOHIO
9 process involves a fluidized bed reaction of chemical-grade propylene, ammonia, and oxygen over a catalyst. The
10 process produces acrylonitrile as its primary product and the process yield depends on the type of catalyst used and
11 the process configuration. The ammoxidation process also produces byproduct COa, CO, and water from the direct
12 oxidation of the propylene feedstock, and produces other hydrocarbons from side reactions in the ammoxidation
13 process.
14 Carbon black is a black powder generated by the incomplete combustion of an aromatic petroleum- or coal-based
15 feedstock at a high temperature. Most carbon black produced in the United States is added to rubber to impart
16 strength and abrasion resistance, and the tire industry is by far the largest consumer. The other major use of carbon
17 black is as a pigment. The predominant process used in the United States is the furnace black (or oil furnace)
18 process. In the furnace black process, carbon black oil (a heavy aromatic liquid) is continuously injected into the
19 combustion zone of a natural gas-fired furnace. Furnace heat is provided by the natural gas and a portion of the
20 carbon black feedstock; the remaining portion of the carbon black feedstock is pyrolyzed to carbon black. The
21 resultant CO2 and uncombusted CH4 emissions are released from thermal incinerators used as control devices,
22 process dryers, and equipment leaks. Carbon black is also produced in the United States by the thermal cracking of
23 acetylene-containing feedstocks (i.e., acetylene black process), by the thermal cracking of other hydrocarbons (i.e.,
24 thermal black process), and by the open burning of carbon black feedstock (i.e., lamp black process); each of these
25 process are used at only one U.S. plant each (EPA 2000).
26 Ethylene (C2H4) is consumed in the production processes of the plastics industry including polymers such as high,
27 low, and linear low density polyethylene (HDPE, LDPE, LLDPE); polyvinyl chloride (PVC); ethylene dichloride;
28 ethylene oxide; and ethylbenzene. Virtually all ethylene is produced from steam cracking of ethane, propane, butane,
29 naphtha, gas oil, and other feedstocks. The representative chemical equation for steam cracking of ethane to ethylene
30 is shown below:
31 C2H6 -> C2H4+ H2
32 Small amounts of CH4 are also generated from the steam cracking process. In addition, CO2 and CH4 emissions are
3 3 also generated from combustion units..
34 Ethylene dichloride (C2H4Cl2) is used to produce vinyl chloride monomer, which is the precursor to polyvinyl
35 chloride (PVC). Ethylene dichloride was used as a fuel additive until 1996 when leaded gasoline was phased out.
36 Ethylene dichloride is produced from ethylene by either direct chlorination, oxychlorination, or a combination of the
37 two processes (i.e., the "balanced process"); most U.S. facilities use the balanced process. The direct chlorination
3 8 and oxychlorination reactions are shown below:
39 C2H4 + C12 -> C2H4C12 (direct chlorination)
40 C2H4 + \02 + 2HCI -> C2H4C12 + 2H20 (oxychlorination)
41 C2//4 + 302 -> 2C02 + 2H20 (direct oxidation of ethylene during oxychlorination)
42 In addition to the byproduct CO2 produced from the direction oxidation of the ethylene feedstock, CO2 and CH4
43 emissions are also generated from combustion units.
44 Ethylene oxide (C2H4O) is used in the manufacture of glycols, glycol ethers, alcohols, and amines. Approximately
45 70 percent of ethylene oxide produced worldwide is used in the manufacture of glycols, including monoethylene
46 glycol. Ethylene oxide is produced by reacting ethylene with oxygen over a catalyst. The oxygen may be supplied to
47 the process through either an air (air process) or a pure oxygen stream (oxygen process). The byproduct CO2 from
48 the direct oxidation of the ethylene feedstock is removed from the process vent stream using a recycled carbonate
49 solution, and the recovered CO2 may be vented to the atmosphere or recovered for further utilization in other
50 sectors, such as food production (IPCC 2006). The combined ethylene oxide reaction and byproduct CO2 reaction is
4-42 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 exothermic and generates heat, which is recovered to produce steam for the process. The ethylene oxide process also
2 produces other liquid and off-gas byproducts (e.g., ethane, etc.) that may be burned for energy recovery within the
3 process. Almost all facilities, except one in Texas, use the oxygen process to manufacture ethylene oxide (EPA
4 2008).
5 Methanol (CH3OH) is a chemical feedstock most often converted into formaldehyde, acetic acid and olefins. It is
6 also an alternative transportation fuel, as well as an additive used by municipal wastewater treatment facilities in the
7 denitrification of wastewater. Methanol is most commonly synthesized from a synthesis gas (i.e., "syngas" - a
8 mixture containing H2, CO, and CO2) using a heterogeneous catalyst. There are a number of process techniques that
9 can be used to produce syngas. Worldwide, steam reforming of natural gas is the most common method; however, in
10 the United States only two facilities use steam reforming of natural gas. Other syngas production processes in the
11 United States include partial oxidation of natural gas and coal gasification.
12 Emissions of CO2 and CH4 from petrochemical production in 2014 were 26.5 MMT CO2 Eq. (26,509 kt CO2) and
13 0.1 MMT CO2 Eq. (5 kt CH4), respectively (see Table 4-43 and Table 4-44). Since 1990, total CO2 emissions from
14 petrochemical production increased by approximately 23 percent. Methane emissions from petrochemical (methanol
15 and acrylonitrile) production have decreased by approximately 43 percent since 1990, given declining production.
16 Table 4-43: COz and CH4 Emissions from Petrochemical Production (MMT COz Eq.)
Year
C02
CH4
Total
1990
21.6
0.2 1
21.8
+ Does not exceed 0.05
17
18
Table 4-44:
Year
CO2
CH4
COz and
1990
21,609
9
2005
27.4
• 0.1
27.5
MMT CO2 Eq.
• 2010
27.2
H +
| 27.3
2011
26.3
+
26.4
2012
26.5
0.1
26.5
2013
26.4
0.1
26.5
2014
26.5
0.1
26.6
CH4 Emissions from Petrochemical Production (kt)
2005
27,380
1 3
2010
27,246
2
2011
26,326
2
2012
26,464
3
2013
2014
26,437 26,509
3
5
19 Methodology
20 Emissions of CO2 and CH4 were calculated using the estimation methods provided by the 2006IPCC Guidelines
21 (IPCC 2006) and country-specific methods from EPA's Greenhouse Gas Reporting Program (GHGRP). The 2006
22 IPCC Guidelines Tier 1 method was used to estimate CO2 and CH4 emissions from production of acrylonitrile and
23 methanol, and a country-specific approach similar to the IPCC Tier 2 method was used to estimate CO2 emissions
24 from carbon black, ethylene, ethylene oxide, and ethylene dichloride. The Tier 2 method for petrochemicals is a total
25 feedstock carbon mass balance method used to estimate total CO2 emissions, but is not applicable for estimating
26 CH4 emissions. As noted in the 2006 IPCC Guidelines, the total feedstock carbon mass balance method (Tier 2) is
27 based on the assumption that all of the carbon input to the process is converted either into primary and secondary
28 products or into CO2. Further, the guideline states that while the total carbon mass balance method estimates total
29 carbon emissions from the process but does not directly provide an estimate of the amount of the total carbon
30 emissions emitted as CO2, CH4, or NMVOCs. This method accounts for all the carbon as CO2, including CH4.
31 Carbon Black, Ethylene, Ethylene Dichloride and Ethylene Oxide
32 2010-2014
33 Carbon dioxide emissions and national production were aggregated directly from EPA's GHGRP dataset for 2010
34 through 2014 (EPA GHGRP 2015). In 2014, GHGRP data reported CO2 emissions of 3,272,934 metric tons from
35 carbon black production; 18,805,943 metric tons of CO2 from ethylene production; 591,127 metric tons of CO2from
36 ethylene dichloride production; and 1,333,768 metric tons of CO2 from ethylene oxide production. These emissions
37 reflect application of a country-specific approach similar to the IPCC Tier 2 method and were used to estimate CO2
38 emissions from the production of carbon black, ethylene, ethylene dichloride, and ethylene oxide. Since 2010,
Industrial Processes and Product Use 4-43
-------
1 EPA's GHGRP, under Subpart X, requires all domestic producers of petrochemicals to report annual emissions and
2 supplemental emissions information (e.g., production data, etc.) to facilitate verification of reported emissions.
3 Under EPA's GHGRP, petrochemical production facilities are required to use either a mass balance approach or
4 CEMS to measure and report emissions for each petrochemical process unit to estimate facility-level process CCh
5 emissions. The mass balance method is used by most facilities22 and assumes that all the carbon input is converted
6 into primary and secondary products, byproducts, or is emitted to the atmosphere as €62. To apply the mass
7 balance, facilities must measure the volume or mass of each gaseous and liquid feedstock and product, mass rate of
8 each solid feedstock and product, and carbon content of each feedstock and product for each process unit and sum
9 for their facility.23 More details on the greenhouse gas calculation and monitoring methods applicable to
10 petrochemical facilities can be found under Subpart X (Petrochemical Production) of the regulation (40 CFR Part
11 98).24
12 1990-2009
13 For prior years, for these petrochemical types, an average national CCh emission factor was calculated based on the
14 2010 through 2013 GHGRP data and applied to production for earlier years in the time series (i.e., 1990 through
15 2009) to estimate CCh emissions from carbon black, ethylene, ethylene dichloride, and ethylene oxide. Carbon
16 dioxide emission factors were derived from EPA's GHGRP data by dividing annual CC>2 emissions for
17 petrochemical type "i" with annual production for petrochemical type "i" and then averaging the derived emission
18 factors obtained for each calendar year 2010 through 2014 (EPA GHGRP 2015). The average emission factors for
19 each petrochemical type were applied across all prior years because petrochemical production processes in the
20 United States have not changed significantly since 1990, though some operational efficiencies have been
21 implemented at facilities over the time series.
22 The average country-specific CCh emission factors that were calculated from the 2010 through 2014 GHGRP data
23 are as follows:
24 • 2.59 metric tons CO^metric ton carbon black produced
25 • 0.79 metric tons CO^metric ton ethylene produced
26 • 0.040 metric tons CCVmetric ton ethylene dichloride produced
27 • 0.46 metric tons CO^metric ton ethylene oxide produced
28 Annual production data for carbon black for 1990 through 2009 were obtained from the International Carbon Black
29 Association (Johnson 2003 and 2005 through 2010). Annual production data for ethylene and ethylene dichloride for
30 1990 through 2009 were obtained from the American Chemistry Council's (ACC's) Guide to the Business of
31 Chemistry (ACC 2002, 2003, 2005 through 2011). Annual production data for ethylene oxide were obtained from
32 ACC's U.S. Chemical Industry Statistical Handbook for 2003 through 2009 (ACC 2014a) and from ACC's Business
33 of Chemistry for 1990 through 2002 (ACC 2014b). As noted above, annual 2010 through 2014 production data for
34 carbon black, ethylene, ethylene dichloride, and ethylene oxide, were obtained from EPA's GHGRP (EPA GHGRP
35 2015).
36 Acrylonitrile
37 Carbon dioxide and methane emissions from acrylonitrile production were estimated using the Tier 1 method in the
38 2006IPCC Guidelines (IPCC 2006). Annual acrylonitrile production data were used with IPCC default Tier 1 CO2
39 and CH4 emission factors to estimate emissions for 1990 through 2014. Emission factors used to estimate
40 acrylonitrile production emissions are as follows:
41 • 0.18 kg CH4/metric ton acrylonitrile produced
99
A few facilities producing ethylene dichloride used CO2 CEMS, which has been included in the aggregated GHGRP emissions.
9^
For ethylene processes only, because nearly all process emissions are from the combustion of process off-gas. Under GHGRP, Subpart X,
ethylene facilities can report emissions from burning of process gases using the optional combustion methodology for ethylene production
processes, which is requires estimating emissions based on fuel quantity and carbon contents of the fuel. This is consistent with the 2006 IPCC
Guidelines (p. 3.57) which recommends including combustion emissions from fuels obtained from feedstocks (e.g., off-gases) in petrochemical
production under in the IPPU sector.
Available online at: http://www.ecfr.gov/cgi-bin/text-idx?tpl=/ecfrbrowse/Title40/40cfr98_main_02.tpl>
4-44 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 • 1.00 metric tons CCVmetric ton acrylonitrile produced
2 Annual acrylonitrile production data for 1990 through 2014 were obtained from ACC's Business of Chemistry (ACC
3 2015).
4 Methanol
5 Carbon dioxide and methane emissions from methanol production were estimated using Tier 1 method in the 2006
6 IPCC Guidelines (IPCC 2006). Annual methanol production data were used with IPCC default Tier 1 CO2 and CH4
7 emission factors to estimate emissions for 1990 through 2014. Emission factors used to estimate methanol
8 production emissions are as follows:
9 • 2.3 kg CH4/metric ton methanol produced
10 • 0.67 metric tons CCVmetric ton methanol produced
11 Annual methanol production data for 1990 through 2014 were obtained from the ACC's Business of Chemistry
12 (ACC 2015).
13 Table 4-45: Production of Selected Petrochemicals (kt)
Chemical
Carbon Black
Ethylene
Ethylene Bichloride
Ethylene Oxide
Acrylonitrile
Methanol
1990
1,307
16,542
6,283
2,429
1,214
3,750
2005
1,651
23,975
11,260
3,220
1,325
1,225
2010
1,309
24,355
8,149
2,925
1,160
730
2011
1,338
25,143
8,621
3,014
1,055
700
2012
1,283
24,763
11,309
3,106
1,220
995
2013
1,228
25,341
11,462
3,148
1,075
1,235
2014
1,207
25,509
11,288
3,138
1,095
2,105
14 Uncertainty and Time-Series Consistency
15 The CH4 and CCh emission factors used for acrylonitrile and methanol production are based on a limited number of
16 studies. Using plant-specific factors instead of default or average factors could increase the accuracy of the
17 emission estimates; however, such data were not available for the current Inventory report.
18 The results of the quantitative uncertainty analysis for the CC>2 emissions from carbon black production, ethylene,
19 ethylene dichloride, and ethylene oxide are based on reported GHGRP data. Refer to the methodology section for
20 more details on how these emissions were calculated and reported to EPA's GHGPJ3. There is some uncertainty in
21 the applicability of the average emission factors for each petrochemical type across all prior years. While
22 petrochemical production processes in the United States have not changed significantly since 1990, some
23 operational efficiencies have been implemented at facilities over the time series.
24 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-46. Petrochemical
25 production CC>2 emissions were estimated to be between 25.3 and 27.8 MMT CCh Eq. at the 95 percent confidence
26 level. This indicates a range of approximately 5 percent below to 5 percent above the emission estimate of 26.5
27 MMT CO2 Eq. Petrochemical production CH4 emissions were estimated to be between 0.05 and 0.15 MMT CO2
28 Eq. at the 95 percent confidence level. This indicates a range of approximately 55 percent below to 45 percent
29 above the emission estimate of 0.1 MMT CO2 Eq.
30 Table 4-46: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from
31 Petrochemical Production and COz Emissions from Carbon Black Production (MMT COz Eq.
32 and Percent)
2014 Emission
Source Gas Estimate
(MMT CO2 Eq.)
Petrochemical „_ ,. , .
„ , . C(J2 26.5
Production
Uncertainty Range Relative to Emission Estimate3
(MMT CO2 Eq.) (%)
Lower
Bound
25.3
Upper
Bound
27.8
Lower
Bound
-5%
Upper
Bound
+5%
Industrial Processes and Product Use 4-45
-------
Petrochemical _55% +45%
Production
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
1 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
2 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
3 above.
4 Recalculation Discussion
5 Acrylonitrile production data were obtained from ACC (2015). The ACC data included annual production quantities
6 for the 1990 through 2014 time series. These data included revised acrylonitrile production quantities for several
7 years of the time series compared to the production data used in the previous Inventory. This update in the
8 production data caused a change in acrylonitrile emissions compared to the previous Inventory report. As a result of
9 this update, emissions for some years increased and emissions for other years decreased. The change in annual
10 emissions from the previous Inventory ranged from -9 percent (in 2010) to 11 percent (in 2009).
11 Methanol production data for 1990 through 2014 were also obtained from ACC (2015). In the previous Inventory,
12 methanol production data for 1990 through 2013 were obtained from ACC and Argus Media Inc. (ARGUS JJ&A
13 2014). As a result of this update, emissions for some years increased slightly and emissions for other years
14 decreased slightly.
15 Planned Improvements
16 Pending resources, a potential improvement for this source category would focus on analyzing the fuel and
17 feedstock data from EPA's GHGRP to better disaggregate energy-related emissions and allocate them more
18 accurately between the Energy and IPPU sectors of the Inventory. Some degree of double counting may occur
19 between CC>2 estimates of non-energy use of fuels in the energy sector and CCh process emissions from
20 petrochemical production in this sector. Data integration is not feasible at this time as feedstock data from EIA used
21 to estimate non-energy uses of fuels are aggregated by fuel type, rather than disaggregated by both fuel type and
22 particular industries (e.g., petrochemical production). EPA, through its GHGPJ3, currently does not collect complete
23 data on quantities of fuel consumed as feedstocks by petrochemical producers, only feedstock type. Updates to
24 reporting requirements may address this issue future reporting years for the GHGPJ3 data allowing for easier data
25 integration between the non-energy uses of fuels category and the petrochemicals category presented in this chapter.
26
27
4.13 HCFC-22 Production (IPCC Source
Category 2B9a) (TO BE UPDATED)
28 Trifluoromethane (HFC-23 or CHF3) is generated as a byproduct during the manufacture of chlorodifluoromethane
29 (HCFC-22), which is primarily employed in refrigeration and air conditioning systems and as a chemical feedstock
30 for manufacturing synthetic polymers. Between 1990 and 2000, U.S. production of HCFC-22 increased
31 significantly as HCFC-22 replaced chlorofluorocarbons (CFCs) in many applications. Between 2000 and 2007, U.S.
32 production fluctuated but generally remained above 1990 levels. In 2008 and 2009, U.S. production declined
33 markedly and has remained near 2009 levels since. Because HCFC-22 depletes stratospheric ozone, its production
34 for non-feedstock uses is scheduled to be phased out by 2020 under the U.S. Clean Air Act.25 Feedstock production,
35 however, is permitted to continue indefinitely.
36 HCFC-22 is produced by the reaction of chloroform (CHCls) and hydrogen fluoride (HF) in the presence of a
37 catalyst, SbCls. The reaction of the catalyst and HF produces SbClxFy, (where x + y = 5), which reacts with
As construed, interpreted, and applied in the terms and conditions of the Montreal Protocol on Substances that Deplete the
Ozone Layer. [42 U.S.C. §7671m(b), CAA §614]
4-46 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 chlorinated hydrocarbons to replace chlorine atoms with fluorine. The HF and chloroform are introduced by
2 submerged piping into a continuous-flow reactor that contains the catalyst in a hydrocarbon mixture of chloroform
3 and partially fluorinated intermediates. The vapors leaving the reactor contain HCFC-21 (CHC^F), HCFC-22
4 (CHC1F2), HFC-23 (CHF3), HC1, chloroform, and HF. The under-fluorinated intermediates (HCFC-21) and
5 chloroform are then condensed and returned to the reactor, along with residual catalyst, to undergo further
6 fluorination. The final vapors leaving the condenser are primarily HCFC-22, HFC-23, HC1 and residual HF. The
7 HC1 is recovered as a useful byproduct, and the HF is removed. Once separated from HCFC-22, the HFC-23 may
8 be released to the atmosphere, recaptured for use in a limited number of applications, or destroyed.
9 Two facilities produced HCFC-22 in the U.S. in 2013. Emissions of HFC-23 from this activity in 2013 were
10 estimated to be 4.1 MMT CCh Eq. (0.3 kt) (see Table 4-47). This quantity represents a 25 percent decrease from
11 2012 emissions and a 91 percent decline from 1990 emissions. The decrease from 2012 emissions and the decrease
12 from 1990 emissions were caused by a decrease in HCFC-22 production and a decrease in the HFC-23 emission rate
13 (kg HFC-23 emitted/kg HCFC-22 produced). The decrease in the emission rate is primarily attributable to six
14 factors: (a) five plants that did not capture and destroy the HFC-23 generated have ceased production of HCFC-22
15 since 1990, (b) one plant that captures and destroys the HFC-23 generated began to produce HCFC-22, (c) one plant
16 implemented and documented a process change that reduced the amount of HFC-23 generated, and (d) the same
17 plant began recovering HFC-23, primarily for destruction and secondarily for sale, (e) another plant began
18 destroying HFC-23, and (f) the same plant, whose emission factor was higher than that of the other two plants,
19 ceased production of HCFC-22 in 2013.
20 Table 4-47: HFC-23 Emissions from HCFC-22 Production (MMT COz Eq. and kt HFC-23)
21
Year MMTCChEq. kt HFC-23
1990 46.1 3
2005
2009
2010
2011
2012
2013
20.0
6.8
8.0
8.8
5.5
4.1
1
0.5
0.5
0.6
0.4
0.3
Methodology
22 To estimate HFC-23 emissions for five of the eight HCFC-22 plants that have operated in the United States since
23 1990, methods comparable to the Tier 3 methods in the 2006IPCC Guidelines (IPCC 2006) were used. Emissions
24 for 2010 through 2013 were obtained through reports submitted by U.S. HCFC-22 production facilities to EPA's
25 GHGRP. EPA's GHGRP mandates that all HCFC-22 production facilities report their annual emissions of HFC-23
26 from HCFC-22 production processes and HFC-23 destruction processes. Previously, data were obtained by EPA
27 through collaboration with an industry association that received voluntarily reported HCFC-22 production and HFC-
28 23 emissions annually from all U.S. HCFC-22 producers from 1990 through 2009. These emissions were aggregated
29 and reported to EPA on an annual basis.
30 For the other three plants, the last of which closed in 1993, methods comparable to the Tier 1 method in the 2006
31 IPCC Guidelines were used. Emissions from these three plants have been calculated using the recommended
32 emission factor for unoptimized plants operating before 1995 (0.04 kg HCFC-23/kg HCFC-22 produced).
33 The five plants that have operated since 1994 measure (or, for the plants that have since closed, measured)
34 concentrations of HFC-23 to estimate their emissions of HFC-23. Plants using thermal oxidation to abate their
35 HFC-23 emissions monitor the performance of their oxidizers to verify that the HFC-23 is almost completely
36 destroyed. Plants that release (or historically have released) some of their byproduct HFC-23 periodically measure
37 HFC-23 concentrations in the output stream using gas chromatography. This information is combined with
38 information on quantities of products (e.g., HCFC-22) to estimate HFC-23 emissions.
Industrial Processes and Product Use 4-47
-------
1 To estimate 1990 through 2009 emissions, reports from an industry association were used that aggregated HCFC-22
2 production and HFC-23 emissions from all U.S. HCFC-22 producers and reported them to EPA (ARAP 1997, 1999,
3 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010). To estimate 2010 through 2013 emissions,
4 facility-level data (including both HCFC-22 production and HFC-23 emissions) reported through the EPA's
5 GHGRP were analyzed. In 1997 and 2008, comprehensive reviews of plant-level estimates of HFC-23 emissions
6 and HCFC-22 production were performed (RTI 1997; RTI2008). The 1997 and 2008 reviews enabled U.S. totals to
7 be reviewed, updated, and where necessary, corrected, and also for plant-level uncertainty analyses (Monte-Carlo
8 simulations) to be performed for 1990, 1995, 2000, 2005, and 2006. Estimates of annual U.S. HCFC-22 production
9 are presented in Table 4-48.
10 Table 4-48: HCFC-22 Production (kt)
Year
1990
2005
1 •
2009
2010
2011
2012
2013
kt
139
156
1
91
101
110
96
C
Note: HCFC-22 production in 2013 is
considered Confidential Business Information
(CBI) as there were only two producers of
HCFC-22 in 2013.
11 Uncertainty and Time-Series Consistency
12 The uncertainty analysis presented in this section was based on a plant-level Monte Carlo Stochastic Simulation for
13 2006. The Monte Carlo analysis used estimates of the uncertainties in the individual variables in each plant's
14 estimating procedure. This analysis was based on the generation of 10,000 random samples of model inputs from
15 the probability density functions for each input. A normal probability density function was assumed for all
16 measurements and biases except the equipment leak estimates for one plant; a log-normal probability density
17 function was used for this plant's equipment leak estimates. The simulation for 2006 yielded a 95-percent
18 confidence interval for U.S. emissions of 6.8 percent below to 9.6 percent above the reported total.
19 The relative errors yielded by the Monte Carlo Stochastic Simulation for 2006 were applied to the U.S. emission
20 estimate for 2013. The resulting estimates of absolute uncertainty are likely to be reasonably accurate because (1)
21 the methods used by the three plants to estimate their emissions are not believed to have changed significantly since
22 2006, and (2) although the distribution of emissions among the plants may have changed between 2006 and 2013
23 (because both HCFC-22 production and the HFC-23 emission rate declined significantly), the two plants that
24 contribute significantly to emissions were estimated to have similar relative uncertainties in their 2006 (as well as
25 2005) emission estimates. Thus, changes in the relative contributions of these two plants to total emissions are not
26 likely to have a large impact on the uncertainty of the national emission estimate.
27 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-49. HFC-23 emissions
28 from HCFC-22 production were estimated to be between 3.8 and 4.5 MMT CCh Eq. at the 95 percent confidence
29 level. This indicates a range of approximately 7 percent below and 10 percent above the emission estimate of 4.1
30 MMT CO2 Eq.
31 Table 4-49: Approach 2 Quantitative Uncertainty Estimates for HFC-23 Emissions from
32 HCFC-22 Production (MMT CO2 Eq. and Percent)
„ „ 2013 Emission Estimate Uncertainty Range Relative to Emission Estimate3
(MMT CCh Eq.) (MMT CCh Eq.) (%)
1 Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
HCFC-22 Production HFC-23 4.1 3.8 4.5 -7% +10%
4-48 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Range of emissions reflects a 95 percent confidence interval.
1 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
2 through 2013. Details on the emission trends through time are described in more detail in the Methodology section,
3 above.
4 Recalculations Discussion
5 For the current Inventory, emission estimates have been revised to reflect the GWPs provided in the IPCC Fourth
6 Assessment Report (AR4) (IPCC 2007). AR4 GWP values differ slightly from those presented in the IPCC Second
1 Assessment Report (SAR) (IPCC 1996) (used in the previous inventories), which results in time-series recalculations
8 for most inventory sources. Under the most recent reporting guidelines (UNFCCC 2014), countries are required to
9 report using the AR4 GWPs, which reflect an updated understanding of the atmospheric properties of each
10 greenhouse gas. The GWP of HFC-23 has increased, leading to an overall increase in emissions. For more
11 information please see the Recalculations and Improvements Chapter.
12
13
4.14 Carbon Dioxide Consumption (IPCC
Source Category 2B10)
14 Carbon dioxide is used for a variety of commercial applications, including food processing, chemical production,
15 carbonated beverage production, and refrigeration, and is also used in petroleum production for enhanced oil
16 recovery (EOR). Carbon dioxide used for EOR is injected into the underground reservoirs to increase the reservoir
17 pressure to enable additional petroleum to be produced. For the most part, CO2 used in non-EOR applications will
18 eventually be released to the atmosphere, and for the purposes of this analysis CO2 used in commercial applications
19 other than EOR is assumed to be emitted to the atmosphere. Carbon dioxide used in EOR applications is discussed
20 in the Energy chapter under "Carbon Capture and Storage, including Enhanced Oil Recovery" and is not discussed
21 in this section.
22 Carbon dioxide is produced from naturally-occurring CO2 reservoirs, as a byproduct from the energy and industrial
23 production processes (e.g., ammonia production, fossil fuel combustion, ethanol production), and as a byproduct
24 from the production of crude oil and natural gas, which contain naturally occurring CO2 as a component. Only CO2
25 produced from naturally occurring CO2 reservoirs, and as a byproduct from energy and industrial processes, and
26 used in industrial applications other than EOR is included in this analysis. Carbon dioxide captured from biogenic
27 sources (e.g., ethanol production plants) is not included in the Inventory. Carbon dioxide captured from crude oil
28 and gas production is used in EOR applications and is therefore reported in the Energy chapter.
29 Carbon dioxide is produced as a byproduct of crude oil and natural gas production. This CO2 is separated from the
30 crude oil and natural gas using gas processing equipment, and may be emitted directly to the atmosphere, or
31 captured and reinjected into underground formations, used for EOR, or sold for other commercial uses. A further
32 discussion of CO2 used in EOR is described in the Energy chapter under the text box titled "Carbon Dioxide
33 Transport, Injection, and Geological Storage."
34 In 2014, the amount of CO2 produced and captured for commercial applications and subsequently emitted to the
35 atmosphere was 4.5 MMT CO2Eq. (4,471 kt) (see Table 4-50). This is an increase of approximately 7 percent from
36 the previous year and an increase of approximately 204 percent since 1990.
37 Table 4-50: COz Emissions from COz Consumption (MMT COz Eq. and kt)
Year MMT CCh Eq. kt
1990 1.5 1,472
Industrial Processes and Product Use 4-49
-------
Year MMT CCh Eq. kt
2010
2011
2012
2013
2014
4.4
4.1
4.0
4.2
4.5
4,425
4,083
4,019
4,188
4,471
i Methodology
2 Carbon dioxide emission estimates for 1990 through 2014 were based on the quantity of €62 extracted and
3 transferred for industrial applications (i.e., non-EOR end-uses). Some of the CCh produced by these facilities is used
4 for EOR and some is used in other commercial applications (e.g., chemical manufacturing, food production). It is
5 assumed that 100 percent of the CCh production used in commercial applications other than EOR is eventually
6 released into the atmosphere.
7 2010 through 2014
8 For 2010 through 2014, data from U.S. EPA's GHGRP (Subpart PP) were aggregated from facility-level reports to
9 develop a national-level estimate for use in the Inventory (EPA GHGRP 2015). Facilities report CCh extracted or
10 produced from natural reservoirs and industrial sites, and €62 captured from energy and industrial processes and
11 transferred to various end-use applications to EPA's GHGRP. This analysis includes only reported €62 transferred
12 to food and beverage end-uses. EPA is continuing to analyze and assess integration of €62 transferred to other end-
13 uses to enhance the completeness of estimates under this source category. Other end-uses include industrial
14 applications, such as metal fabrication. EPA is analyzing the information reported to ensure that other end-use data
15 excludes non-emissive applications and publication will not reveal confidential business information. Reporters
16 subject to EPA's GHGRP Subpart PP are also required to report the quantity of €62 that is imported and/or
17 exported. Currently, these data are not publicly available through the GHGRP due to data confidentiality issues and
18 hence are excluded from this analysis.
19 Facilities subject to Subpart PP of EPA's GHGRP are required to measure €62 extracted or produced. More details
20 on the calculation and monitoring methods applicable to extraction and production facilities can be found under
21 Subpart PP: Suppliers of Carbon Dioxide of the regulation, Part 98.26 The number of facilities that reported data to
22 EPA's GHGRP Subpart PP (Suppliers of Carbon Dioxide) for 2010 through 2014 is much higher (ranging from 44
23 to 48) than the number of facilities included in the Inventory for the 1990 to 2009 time period prior to the
24 availability of GHGRP data (4 facilities). The difference is largely due to the fact the 1990 to 2009 data includes
25 only CO2 transferred to end-use applications from naturally occurring CO2 reservoirs and excludes industrial sites.
26 1990 through 2009
27 For 1990 through 2009, data from EPA's GHGRP are not available. For this time period, CO2 production data from
28 four naturally-occurring CO2 reservoirs were used to estimate annual CO2 emissions. These facilities were Jackson
29 Dome in Mississippi, Brave and West Bravo Domes in New Mexico, and McCallum Dome in Colorado. The
30 facilities in Mississippi and New Mexico produced CO2 for use in both EOR and in other commercial applications
31 (e.g., chemical manufacturing, food production). The fourth facility in Colorado (McCallum Dome) produced CO2
32 for commercial applications only (New Mexico Bureau of Geology and Mineral Resources 2006).
3 3 Carbon dioxide production data and the percentage of production that was used for non-EOR applications for the
34 Jackson Dome, Mississippi facility were obtained from Advanced Resources International (ARI 2006, 2007) for
35 1990 to 2000, and from the Annual Reports of Denbury Resources (Denbury Resources 2002 through 2010) for
36 2001 to 2009 (see Table 4-51). Denbury Resources reported the average CO2 production in units of MMCF CO2 per
37 day for 2001 through 2009 and reported the percentage of the total average annual production that was used for
38 EOR. Production from 1990 to 1999 was set equal to 2000 production, due to lack of publicly available production
26
4-50 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1
2
o
6
4
5
6
7
8
9
10
data for 1990 through 1999. Carbon dioxide production data for the Bravo Dome and West Bravo Dome were
obtained from ARI for 1990 through 2009 (ARI1990-2010). Data for the West Bravo Dome facility were only
available for 2009. The percentage of total production that was used for non-EOR applications for the Bravo Dome
and West Bravo Dome facilities for 1990 through 2009 were obtained from New Mexico Bureau of Geology and
Mineral Resources (Broadhead 2003 and New Mexico Bureau of Geology and Mineral Resources 2006).
Production data for the McCallum Dome (Jackson County), Colorado facility were obtained from the Colorado Oil
and Gas Conservation Commission (COGCC) for 1999 through 2009 (COGCC 2014). Production data for 1990 to
1998 and percentage of production used for EOR were assumed to be the same as for 1999, due to lack of publicly-
available data.
Table 4-51: COz Production (kt COz) and the Percent Used for Non-EOR Applications
Year
Jackson Dome,
MS
CO2 Production
(kt)(%Non-
EOR)
Bravo Dome,
NM
CO2 Production
(kt)(%Non-
EOR)
West Bravo
Dome, NM CO2
Production
(kt) (%Non-
EOR)
McCallum
Dome, CO
CO2 Production
(kt)(%Non-
EOR)
Total CO2
Production
from Extraction
and Capture
Facilities (kt)
Non-
EOR*
11
12
13
14
15
16
17
18
19
20
21
22
23
1990 1,344(100%)
63 (1%)
65 (100%)
NA
NA
2010
2011
2012
2013
2014
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
64,724
66,241
66,326
68,435
71,431
7%
6%
6%
6%
6%
* Includes only food & beverage applications
+ Does not exceed 0%.
NA - Not available. For 2010-2014, the publicly available GHGRP data were aggregated
at the national level. Facility-level data are not publicly available from EPA's GHGRP.
Uncertainty and Time-Series Consistency
There is uncertainty associated with the data reported through EPA's GHGRP. As noted, reporting of end-uses is not
required, so there is uncertainty associated with the amount of CCh consumed for food and beverage applications, in
addition to the exclusion of the amount of CCh transferred to all other end-use categories. This latter category might
include CC>2 quantities that are being used for non-EOR industrial applications such as firefighting. Second,
uncertainty is associated with the exclusion of imports/exports data for CO2 suppliers. Currently these data are not
publicly available through EPA's GHGRP and hence are excluded from this analysis.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-52. Carbon dioxide
consumption CO2 emissions for 2014 were estimated to be between 3.9 and 5.1 MMT CO2 Eq. at the 95 percent
confidence level. This indicates a range of approximately 12 percent below to 13 percent above the emission
estimate of 4.5 MMT CO2 Eq.
Table 4-52: Approach 2 Quantitative Uncertainty Estimates for COz Emissions from COz
Consumption (MMT COz Eq. and Percent)
Source
Gas
2014 Emission Estimate
(MMT C02 Eq.)
Uncertainty Range Relative to Emission Estimate3
(MMT C02 Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
CO2 Consumption
C02
4.5
3.9 5.1 -12% +13%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
24 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
25 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
26 above.
Industrial Processes and Product Use 4-51
-------
i Planned Improvements
2 EPA will continue to evaluate the potential to include additional GHGRP data on other emissive end-uses to
3 improve accuracy and completeness of estimates for this source category.
4 4.15 Phosphoric Acid Production (IPCC
5 Source Category 2B10)
6 Phosphoric acid (HsPCM) is a basic raw material used in the production of phosphate-based fertilizers. Phosphoric
7 acid production from natural phosphate rock is a source of CCh emissions, due to the chemical reaction of the
8 inorganic carbon (calcium carbonate) component of the phosphate rock.
9 Phosphate rock is mined in Florida, North Carolina, Idaho, and Utah and is used primarily as a raw material for wet-
10 process phosphoric acid production. The composition of natural phosphate rock varies depending upon the location
11 where it is mined. Natural phosphate rock mined in the United States generally contains inorganic carbon in the
12 form of calcium carbonate (limestone) and also may contain organic carbon. The calcium carbonate component of
13 the phosphate rock is integral to the phosphate rock chemistry. Phosphate rock can also contain organic carbon that
14 is physically incorporated into the mined rock but is not an integral component of the phosphate rock chemistry.
15 The phosphoric acid production process involves chemical reaction of the calcium phosphate (C
16 component of the phosphate rock with sulfuric acid (H2SO4) and recirculated phosphoric acid (H3PO4) (EFMA
17 2000). However, the generation of CCh is due to the associated limestone-sulfuric acid reaction, as shown below:
18 CaC03 + H2S04 + H20 -> CaS04 • 2H20 + C02
19 Total U.S. phosphate rock production sold or used in 2014 was 28.1 million metric tons (USGS 2015a).
20 Approximately 80 percent of domestic phosphate rock production was mined in Florida and North Carolina, while
21 the remaining 20 percent of production was mined in Idaho and Utah. Total imports of phosphate rock in 2014 were
22 approximately 2.6 million metric tons (USGS 2015a). Most of the imported phosphate rock (74 percent) is from
23 Morocco, with the remaining 26 percent being from Peru (USGS 2015a). All phosphate rock mining companies are
24 vertically integrated with fertilizer plants that produce phosphoric acid located near the mines. Some additional
25 phosphoric acid production facilities are located in Texas, Louisiana, and Mississippi that used imported phosphate
26 rock.
27 Over the 1990 to 2014 period, domestic production has decreased by nearly 44 percent. Total CO2 emissions from
28 phosphoric acid production were 1.1 MMT CO2 Eq. (1,095 kt CO2) in 2014 (see Table 4-53). Domestic
29 consumption of phosphate rock in 2014 was estimated to have decreased by approximately 2 percent over 2013
30 levels, owing to producers drawing from higher than average inventories and the closure of a mine in Florida.
31 Domestic consumption also decreased because of lower phosphoric acid and fertilizer production (USGS 2015a).
32 Table 4-53: COz Emissions from Phosphoric Acid Production (MMT COz Eq. and kt)
Year MMT CCh Eq. kt
1990 1.5 1,529
2010
2011
2012
2013
2014
1.1
1.2
1.1
1.1
1.1
1,087
1,151
1,093
1,119
1,095
4-52 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1
Methodology
2 Carbon dioxide emissions from production of phosphoric acid from phosphate rock are estimated by multiplying the
3 average amount of inorganic carbon (expressed as CCh) contained in the natural phosphate rock as calcium
4 carbonate by the amount of phosphate rock that is used annually to produce phosphoric acid, accounting for
5 domestic production and net imports for consumption. The estimation methodology is as follows:
6 Epa = Cpr x Qpr
7 where,
8 Epa = CO2 emissions from phosphoric acid production, metric tons
9 Cpr = Average amount of carbon (expressed as CCh) in natural phosphate rock, metric ton CCV
10 metric ton phosphate rock
11 Qpr Quantity of phosphate rock used to produce phosphoric acid
12
13 The CO2 emissions calculation methodology is based on the assumption that all of the inorganic carbon (calcium
14 carbonate) content of the phosphate rock reacts to produce CO2 in the phosphoric acid production process and is
15 emitted with the stack gas. The methodology also assumes that none of the organic carbon content of the phosphate
16 rock is converted to CO2 and that all of the organic carbon content remains in the phosphoric acid product.
17 From 1993 to 2004, the USGSMineral Yearbook: Phosphate Rock disaggregated phosphate rock mined annually in
18 Florida and North Carolina from phosphate rock mined annually in Idaho and Utah, and reported the annual
19 amounts of phosphate rock exported and imported for consumption (see Table 4-54). For the years 1990 through
20 1992, and 2005 through 2014, only nationally aggregated mining data was reported by USGS. For the years 1990,
21 1991, and 1992, the breakdown of phosphate rock mined in Florida and North Carolina, and the amount mined in
22 Idaho and Utah, are approximated using average share of U.S. production in those states from 1993 to 2004 data.
23 For the years 2005 through 2014, the same approximation method is used, but the share of U.S. production in those
24 states data were obtained from the USGS commodity specialist for phosphate rock (USGS 2012). Data for domestic
25 sales or consumption of phosphate rock, exports of phosphate rock (primarily from Florida and North Carolina), and
26 imports of phosphate rock for consumption for 1990 through 2014 were obtained from USGS Minerals Yearbook:
27 Phosphate Rock (USGS 1994 through 2015b), and from USGS Minerals Commodity Summaries: Phosphate Rock in
28 2015 (USGS 2015a). From 2004 through 2014, the USGS reported no exports of phosphate rock from U.S.
29 producers (USGS 2005 through 2015b).
30 The carbonate content of phosphate rock varies depending upon where the material is mined. Composition data for
31 domestically mined and imported phosphate rock were provided by the Florida Institute of Phosphate Research
32 (FIPR 2003a). Phosphate rock mined in Florida contains approximately 1 percent inorganic carbon, and phosphate
33 rock imported from Morocco contains approximately 1.46 percent inorganic carbon. Calcined phosphate rock
34 mined in North Carolina and Idaho contains approximately 0.41 percent and 0.27 percent inorganic carbon,
35 respectively (see Table 4-55).
36 Carbonate content data for phosphate rock mined in Florida are used to calculate the CO2 emissions from
37 consumption of phosphate rock mined in Florida and North Carolina (80 percent of domestic production) and
38 carbonate content data for phosphate rock mined in Morocco are used to calculate CO2 emissions from consumption
39 of imported phosphate rock. The CO2 emissions calculation is based on the assumption that all of the domestic
40 production of phosphate rock is used in uncalcined form. As of 2006, the USGS noted that one phosphate rock
41 producer in Idaho produces calcined phosphate rock; however, no production data were available for this single
42 producer (USGS 2006). The USGS confirmed that no significant quantity of domestic production of phosphate rock
43 is in the calcined form (USGS 2012b).
44 Table 4-54: Phosphate Rock Domestic Consumption, Exports, and Imports (kt)
Location/Year
U.S. Domestic
Consumption
FLandNC
ID and UT
Exports— FL and NC
1990
49,800
42,494
7,306
6,240
2005
35,200
28,160
7,040
"
2010
28,100
22,480
5,620
2011
28,600
22,880
5,720
2012
27,300
21,840
5,460
2013
28,800
23,040
5,760
2014
28,100
22,480
5,620
Industrial Processes and Product Use 4-53
-------
Imports 451 2,630 2,400 3,350 3,080 2,560 2,570
Total U.S.
Consumption 44,011 37,830 30,500 31,950 30,380 31,360 30,670
1 Table 4-55: Chemical Composition of Phosphate Rock (Percent by weight)
Composition
Total Carbon (as C)
Inorganic Carbon (as C)
Organic Carbon (as C)
Inorganic Carbon (as CCh)
Central
Florida
1.60
1.00
0.60
3.67
North
Florida
1.76
0.93
0.83
3.43
North Carolina
(calcined)
0.76
0.41
0.35
1.50
Idaho
(calcined)
0.60
0.27
0.00
1.00
Morocco
1.56
1.46
0.10
5.00
Source: FIPR 2003a
Uncertainty and Time-Series Consistency
3 Phosphate rock production data used in the emission calculations were developed by the USGS through monthly and
4 semiannual voluntary surveys of the active phosphate rock mines during 2014. For previous years in the time series,
5 USGS provided the data disaggregated regionally; however, beginning in 2006, only total U.S. phosphate rock
6 production was reported. Regional production for 2014 was estimated based on regional production data from
7 previous years and multiplied by regionally-specific emission factors. There is uncertainty associated with the
8 degree to which the estimated 2014 regional production data represents actual production in those regions. Total
9 U.S. phosphate rock production data are not considered to be a significant source of uncertainty because all the
10 domestic phosphate rock producers report their annual production to the USGS. Data for exports of phosphate rock
11 used in the emission calculation are reported by phosphate rock producers and are not considered to be a significant
12 source of uncertainty. Data for imports for consumption are based on international trade data collected by the U.S.
13 Census Bureau. These U.S. government economic data are not considered to be a significant source of uncertainty.
14 An additional source of uncertainty in the calculation of CC>2 emissions from phosphoric acid production is the
15 carbonate composition of phosphate rock; the composition of phosphate rock varies depending upon where the
16 material is mined, and may also vary over time. The Inventory relies on one study (FIPR 2003a) of chemical
17 composition of the phosphate rock; limited data are available beyond this study. Another source of uncertainty is
18 the disposition of the organic carbon content of the phosphate rock. A representative of the FIPR indicated that in
19 the phosphoric acid production process, the organic C content of the mined phosphate rock generally remains in the
20 phosphoric acid product, which is what produces the color of the phosphoric acid product (FIPR 2003b). Organic
21 carbon is therefore not included in the calculation of CC>2 emissions from phosphoric acid production.
22 A third source of uncertainty is the assumption that all domestically-produced phosphate rock is used in phosphoric
23 acid production and used without first being calcined. Calcination of the phosphate rock would result in conversion
24 of some of the organic C in the phosphate rock into CO2. However, according to air permit information available to
25 the public, at least one facility has calcining units permitted for operation (NCDENR 2013).
26 Finally, USGS indicated that approximately 7 percent of domestically-produced phosphate rock is used to
27 manufacture elemental phosphorus and other phosphorus-based chemicals, rather than phosphoric acid (USGS
28 2006). According to USGS, there is only one domestic producer of elemental phosphorus, in Idaho, and no data
29 were available concerning the annual production of this single producer. Elemental phosphorus is produced by
30 reducing phosphate rock with coal coke, and it is therefore assumed that 100 percent of the carbonate content of the
31 phosphate rock will be converted to CC>2 in the elemental phosphorus production process. The calculation for CC>2
32 emissions is based on the assumption that phosphate rock consumption, for purposes other than phosphoric acid
33 production, results in CCh emissions from 100 percent of the inorganic carbon content in phosphate rock, but none
34 from the organic carbon content.
35 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-56. Phosphoric acid
36 production CC>2 emissions were estimated to be between 0.9 and 1.4 MMT CCh Eq. at the 95 percent confidence
37 level. This indicates a range of approximately 19 percent below and 20 percent above the emission estimate of 1.1
38 MMT CO2 Eq.
4-54 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 4-56: Approach 2 Quantitative Uncertainty Estimates for COz Emissions from
2 Phosphoric Acid Production (MMT COz Eq. and Percent)
„ „ 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
e (MMT CCh Eq.) (MMT CCh Eq.) (%)
Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
Phosphoric Acid Production CCh U 0.9 1_4 -19% +20%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
3 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
4 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
5 above.
6 Recalculations Discussion
7 Relative to the previous Inventory, the phosphate rock consumption data (sold or used and imports for consumption)
8 for 2013 were revised based on updated data publicly available from USGS (USGS 2015). This revision caused a
9 decrease in the 2013 emission estimate by approximately 2 percent.
10 Additionally, during the development of the current Inventory emission estimates, it was discovered that the
11 phosphate rock CC>2 content had been incorrectly transcribed in the previous Inventory. This error was corrected in
12 the current Inventory and resulted in a slight change of emissions over the entire time series.
13 Planned Improvements
14 EPA continues to evaluate potential improvements to the Inventory estimates for this source category, which include
15 direct integration of GHGRP data for 2010 through 2014 and the use of reported GHGRP data to update the
16 inorganic C content of phosphate rock for prior years. In order to provide estimates for the entire time series (i.e.,
17 1990 through 2009), the applicability of EPA's GHGRP data for the averaged inorganic C content data (by region)
18 from 2010 through 2014 to previous years' estimates will need to be evaluated. In implementing improvements and
19 integration of data from EPA's GHGRP, the latest guidance from the IPCC on the use of facility-level data in
20 national inventories will be relied upon.27
21 4.16 Iron and Steel Production (IPCC Source
22 Category 2C1) and Metallurgical Coke
23 Production
24 Iron and steel production is a multi-step process that generates process-related emissions of CCh and CH4 as raw
25 materials are refined into iron and then transformed into crude steel. Emissions from conventional fuels (e.g., natural
26 gas, fuel oil, etc.) consumed for energy purposes during the production of iron and steel are accounted for in the
27 Energy chapter.
28 Iron and steel production includes six distinct production processes: coke production, sinter production, direct
29 reduced iron (DRI) production, pig iron production, electric arc furnace (EAF) steel production, and basic oxygen
30 furnace (EOF) steel production. The number of production processes at a particular plant is dependent upon the
31 specific plant configuration. In addition to the production processes mentioned above, CCh is also generated at iron
27 See.
Industrial Processes and Product Use 4-55
-------
1 and steel mills through the consumption of process byproducts (e.g., blast furnace gas, coke oven gas, etc.) used for
2 various purposes including heating, annealing, and electricity generation. Process byproducts sold for use as
3 synthetic natural gas are deducted and reported in the Energy chapter. In general, CO2 emissions are generated in
4 these production processes through the reduction and consumption of various carbon-containing inputs (e.g., ore,
5 scrap, flux, coke byproducts, etc.). In addition, fugitive CH4 emissions are also generated by the sinter production
6 process.
7 Currently, there are between 15 and 20 integrated iron and steel steelmaking facilities that utilize BOFs to refine and
8 produce steel from iron and more than 100 steelmaking facilities that utilize EAFs to produce steel primarily from
9 recycled ferrous scrap. In addition, there are 18 cokemaking facilities, of which 7 facilities are co-located with
10 integrated iron and steel facilities. Slightly more than 62 percent of the raw steel produced in the United States is
11 produced in one of seven states: Alabama, Arkansas, Indiana, Kentucky, Mississippi, Ohio, and Tennessee (AISI
12 2015a).
13 Total production of crude steel in the United States between 2000 and 2008 ranged from a low of 99,320,000 tons to
14 a high of 109,880,000 tons (2001 and 2004, respectively). Due to the decrease in demand caused by the global
15 economic downturn (particularly from the automotive industry), crude steel production in the United States sharply
16 decreased to 65,459,000 tons in 2009. In 2010, crude steel production rebounded to 88,731,000 tons as economic
17 conditions improved and then continued to increase to 95,237,000 tons in 2011 and 97,770,000 tons in 2012; crude
18 steel production slightly decreased to 95,766,000 tons in 2013 and then slightly increased to 97,195,000 tons in 2014
19 (AISI2015a). The United States was the third largest producer of raw steel in the world, behind China and Japan,
20 accounting for approximately 5.3 percent of world production in 2013 (AISI 2015a).
21 The maj ority of CO2 emissions from the iron and steel production process come from the use of coke in the
22 production of pig iron and from the consumption of other process byproducts, with lesser amounts emitted from the
23 use of flux and from the removal of carbon from pig iron used to produce steel.
24 According to the 2006IPCC Guidelines (IPCC 2006), the production of metallurgical coke from coking coal is
25 considered to be an energy use of fossil fuel and the use of coke in iron and steel production is considered to be an
26 industrial process source. Therefore, the 2006 IPCC Guidelines suggest that emissions from the production of
27 metallurgical coke should be reported separately in the Energy sector, while emissions from coke consumption in
28 iron and steel production should be reported in the IPPU sector. However, the approaches and emission estimates for
29 both metallurgical coke production and iron and steel production are both presented here because much of the
30 relevant activity data is used to estimate emissions from both metallurgical coke production and iron and steel
31 production. For example, some byproducts (e.g., coke oven gas, etc.) of the metallurgical coke production process
32 are consumed during iron and steel production, and some byproducts of the iron and steel production process (e.g.,
33 blast furnace gas, etc.) are consumed during metallurgical coke production. Emissions associated with the
34 consumption of these byproducts are attributed at the point of consumption. Emissions associated with the use of
35 conventional fuels (e.g., natural gas, fuel oil, etc.) for electricity generation, heating and annealing, or other
36 miscellaneous purposes downstream of the iron and steelmaking furnaces are reported in the Energy chapter.
37 Metallurgical Coke Production
38 Emissions of CO2 from metallurgical coke production in 2014 were 1.9 MMT CO2 Eq. (1,938 kt CO2) (see Table
39 4-57 and Table 4-58). Emissions increased in 2014 from 2013 levels, but have decreased overall since 1990.
40 Domestic coke production data for 2014 are not yet published and so 2013 data were used as proxy for 2014. Coke
41 production in 2014 was 26 percent lower than in 2000 and 45 percent below 1990. Overall, emissions from
42 metallurgical coke production have declined by 23 percent (0.6 MMT CO2 Eq.) from 1990 to 2014.
43 Table 4-57: COz Emissions from Metallurgical Coke Production (MMT COz Eq.)
Gas 1990 2005 2010 2011 2012 2013 2014
C02 2.5 2.0 2.1 1.4 0.5 1.8 1.9
Total 2.5 2.0 2.1 1.4 0.5 1.8 1.9
4-56 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Table 4-58: COz Emissions from Metallurgical Coke Production (kt)
Gas
1990
2005
2010
2011
2012
2013
2014
C02
2,503
2,044
2,085 1,426
543 1,824 1,938
3 Iron and Steel Production
4 Emissions of CO2 and CH4 from iron and steel production in 2014 were 53.4 MMT CO2 Eq. (53,417 kt) and 0.0094
5 MMT CO2 Eq. (0.4 kt), respectively (see Table 4-59 through Table 4-62), totaling approximately 53.4 MMT CO2
6 Eq. Emissions decreased in 2014 and have decreased overall since 1990 due to restructuring of the industry,
7 technological improvements, and increased scrap steel utilization. Carbon dioxide emission estimates include
8 emissions from the consumption of carbonaceous materials in the blast furnace, EAF, and EOF, as well as blast
9 furnace gas and coke oven gas consumption for other activities at the steel mill.
10 In 2014, domestic production of pig iron decreased by 3 percent from 2013 levels. Overall, domestic pig iron
11 production has declined since the 1990s. Pig iron production in 2014 was 3 9 percent lower than in 2000 and 41
12 percent below 1990. Carbon dioxide emissions from steel production have decreased by 4 percent (0.3 MMT CO2
13 Eq.) since 1990, while overall CO2 emissions from iron and steel production have declined by 45 percent (43.7
14 MMT CO2Eq.) from 1990 to 2014.
15 Table 4-59: COz Emissions from Iron and Steel Production (MMT COz Eq.)
Source/Activity Data
Sinter Production
Iron Production
Steel Production
Other Activities*
Total
1990
45.6
7.9
41.2
97.2
2005
9.4
35. 9H
64.5
2010
1.0
17.8
9.2
25.5
53.6
2011
1.2
18.4
9.3
29.7
58.5
2012
1.2
10.9
9.9
31.7
53.7
2013
1.1
11.9
8.6
28.7
50.4
2014
1.1
16.8
7.6
27.9
53.4
1 Includes emissions from blast furnace gas and coke oven gas combustion for activities at the steel
mill other than consumption in blast furnace, EAFs, or BOFs.
Note: Totals may not sum due to independent rounding.
16
17
Table 4-60: COz Emissions from Iron and Steel Production (kt)
Source/Activity Data
Sinter Production
Iron Production
Steel Production
Other Activities a
Total
1990
2,448
45,5921
7,933i
41,193
97,166
2005
1,663
17,545
9,356
35,934
64,499
2010
1,045
17,802
9,235
25,504
53,586
2011
1,188
18,375
9,255
29,683
58,501
2012
1,159
10,917
9,860
31,750
53,686
2013
1,117
11,934
8,617
28,709
50,378
2014
1,104
16,754
7,648
27,911
53,417
a Includes emissions from blast furnace gas and coke oven gas combustion for activities at the steel mill
other than consumption in blast furnace, EAFs, or BOFs.
Note: Totals may not sum due to independent rounding.
Table 4-61: CH4 Emissions from Iron and Steel Production (MMT COz Eq.)
Source/Activity Data
1990
2005
2010
2011 2012 2013 2014
Sinter Production
Total
+ Does not exceed 0.05 MMT CO2 Eq.
Industrial Processes and Product Use 4-57
-------
Table 4-62: ChU Emissions from Iron and Steel Production (kt)
Source/Activity Data 1990 2005 2010 2011 2012 2013 2014
Sinter Production 0.9 0.6 0.4 0.4 0.4 0.4 0.4
Total 0.9 0.6 0.4 0.4 0.4 0.4 0.4
Methodology
3 Emission estimates presented in this chapter are largely based on Tier 2 methodologies provided by the 2006IPCC
4 Guidelines (IPCC 2006). These Tier 2 methodologies call for a mass balance accounting of the carbonaceous inputs
5 and outputs during the iron and steel production process and the metallurgical coke production process. Tier 1
6 methods are used for certain iron and steel production processes (i.e., sinter production and DRI production) for
7 which available data are insufficient for utilizing a Tier 2 method.
8 The Tier 2 methodology equation is as follows:
44
Ern —
U2
X •
12
10 where,
11 ECo2 = Emissions from coke, pig iron, EAF steel, or EOF steel production, metric tons
12 a = Input material a
13 b = Output material b
14 Qa = Quantity of input material a, metric tons
15 Ca = Carbon content of input material a, metric tons C/metric ton material
16 Qb Quantity of output material b, metric tons
17 Cb = Carbon content of output material b, metric tons C/metric ton material
18 44/12 = StoichiometricratioofCO2toC
19
20 The Tier 1 methodology equations are as follows:
21 ESjp = Qsx EFSjp
22 Edf02 = Qd x EFdf02
23
24 where,
25 ES)P = Emissions from sinter production process for pollutant p (CCh or CH4), metric ton
26 Qs = Quantity of sinter produced, metric tons
27 EFS)P = Emission factor for pollutant p (CCh or CH4), metric ton/>/metric ton sinter
28 Ed,co2 = Emissions from DRI production process for CC>2, metric ton
29 Qd = Quantity of DRI produced, metric tons
30 EFd,co2 = Emission factor for CO2, metric ton CO2/metric ton DRI
31
32 Metallurgical Coke Production
33 Coking coal is used to manufacture metallurgical coke that is used primarily as a reducing agent in the production of
34 iron and steel, but is also used in the production of other metals including zinc and lead (see Zinc Production and
35 Lead Production sections of this chapter). Emissions associated with producing metallurgical coke from coking coal
36 are estimated and reported separately from emissions that result from the iron and steel production process. To
37 estimate emissions from metallurgical coke production, a Tier 2 method provided by the 2006 IPCC Guidelines
38 (IPCC 2006) was utilized. The amount of carbon contained in materials produced during the metallurgical coke
39 production process (i.e., coke, coke breeze, coke oven gas, and coal tar) is deducted from the amount of carbon
40 contained in materials consumed during the metallurgical coke production process (i.e., natural gas, blast furnace
4-58 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 gas, and coking coal). Light oil, which is produced during the metallurgical coke production process, is excluded
2 from the deductions due to data limitations. The amount of carbon contained in these materials is calculated by
3 multiplying the material-specific carbon content by the amount of material consumed or produced (see Table 4-63).
4 The amount of coal tar produced was approximated using a production factor of 0.03 tons of coal tar per ton of
5 coking coal consumed. The amount of coke breeze produced was approximated using a production factor of 0.075
6 tons of coke breeze per ton of coking coal consumed (AISI 2008c; DOE 2000). Data on the consumption of
7 carbonaceous materials (other than coking coal) as well as coke oven gas production were available for integrated
8 steel mills only (i.e., steel mills with co-located coke plants). Therefore, carbonaceous material (other than coking
9 coal) consumption and coke oven gas production were excluded from emission estimates for merchant coke plants.
10 Carbon contained in coke oven gas used for coke-oven underfiring was not included in the deductions to avoid
11 double-counting.
12 Table 4-63: Material Carbon Contents for Metallurgical Coke Production
Material kg C/kg
Coal Tar 0.62
Coke 0.83
Coke Breeze 0.83
Coking Coal 0.73
Material kg C/GJ
Coke Oven Gas 12.1
Blast Furnace Gas 70.8
Source: IPCC 2006, Table 4.3. Coke Oven Gas and
Blast Furnace Gas, Table 1.3.
13 Although the 2006 IPCC Guidelines provide a Tier 1 CH4 emission factor for metallurgical coke production (i.e.,
14 0.1 g CH4 per metric ton of coke production), it is not appropriate to use because CCh emissions were estimated
15 using the Tier 2 mass balance methodology. The mass balance methodology makes a basic assumption that all
16 carbon that enters the metallurgical coke production process either exits the process as part of a carbon-containing
17 output or as CC>2 emissions. A preliminary assessment of facility-level greenhouse gas emissions reported by coke
18 production facilities under EPA's GHGRP also indicates that CH4 emissions from coke production are low.
19 Estimation of CH4 emissions using a Tier 1 approach to estimate these emissions is likely to significantly
20 overestimate these emissions. EPA is currently finalizing compilation of this information to include in the Inventory.
21 Data relating to the mass of coking coal consumed at metallurgical coke plants and the mass of metallurgical coke
22 produced at coke plants were taken from the Energy Information Administration (EIA), Quarterly Coal Report:
23 October through December (EIA 1998 through 2015a) (see Table 4-64). Data on the volume of natural gas
24 consumption, blast furnace gas consumption, and coke oven gas production for metallurgical coke production at
25 integrated steel mills were obtained from the American Iron and Steel Institute (AISI), Annual Statistical Report
26 (AISI 2004 through 2015a) and through personal communications with AISI (2008c) (see Table 4-65). The factor
27 for the quantity of coal tar produced per ton of coking coal consumed was provided by AISI (2008c). The factor for
28 the quantity of coke breeze produced per ton of coking coal consumed was obtained through Table 2-1 of the report
29 Energy and Environmental Profile of the U.S. Iron and Steel Industry (DOE 2000). Data on natural gas
30 consumption and coke oven gas production at merchant coke plants were not available and were excluded from the
31 emission estimate. Carbon contents for coking coal, metallurgical coke, coal tar, coke oven gas, and blast furnace
32 gas were provided by the 2006IPCC Guidelines (IPCC 2006). The carbon content for coke breeze was assumed to
3 3 equal the carbon content of coke.
34 Table 4-64: Production and Consumption Data for the Calculation of COz and CH4 Emissions
35 from Metallurgical Coke Production (Thousand Metric Tons)
Source/Activity Data
Metallurgical Coke Production
Coking Coal Consumption at Coke Plants
Coke Production at Coke Plants
Coal Breeze Production
Coal Tar Production
a 2013 data were used as a proxy because 2014
1990
3 5, 269 1
25,054 •
2,645 •
1,058
2005
21,259
15,167
1,594
| 638
2010
19,135
13,628
1,435
574
2011
19,445
13,989
1,458
583
2012
18,825
13,764
1,412
565
2013
19,481
13,898
1,461
584
2014"
19,481
13,898
1,461
584
data are not yet published.
Industrial Processes and Product Use 4-59
-------
1 Table 4-65: Production and Consumption Data for the Calculation of COz Emissions from
2 Metallurgical Coke Production (Million ft3)
Source/Activity Data
Metallurgical Coke Production
Coke Oven Gas Production
Natural Gas Consumption
Blast Furnace Gas Consumption
1990
250,767
599
24,602
2005
114,2131
2,996
4,460
2010
95,405
3,108
3,181
2011
109,044
3,175
3,853
2012
113,772
3,267
4,351
2013
108,162
3,247
4,255
2014
102,899
3,039
4,346
3 Iron and Steel Production
4 Emissions of CC>2 from sinter production and direct reduced iron production were estimated by multiplying total
5 national sinter production and the total national direct reduced iron production by Tier 1 CC>2 emission factors (see
6 Table 4-66). Because estimates of sinter production and direct reduced iron production were not available,
7 production was assumed to equal consumption.
8 Table 4-66: COz Emission Factors for Sinter Production and Direct Reduced Iron Production
Metric Ton
Material Produced CCh/Metric Ton
Sinter 0.2
Direct Reduced Iron 0/7
Source: IPCC 2006, Table 4.1.
9 To estimate emissions from pig iron production in the blast furnace, the amount of carbon contained in the produced
10 pig iron and blast furnace gas were deducted from the amount of carbon contained in inputs (i.e., metallurgical coke,
11 sinter, natural ore, pellets, natural gas, fuel oil, coke oven gas, and direct coal injection). The carbon contained in
12 the pig iron, blast furnace gas, and blast furnace inputs was estimated by multiplying the material-specific carbon
13 content by each material type (see Table 4-67). Carbon in blast furnace gas used to pre-heat the blast furnace air is
14 combusted to form CCh during this process. Carbon contained in blast furnace gas used as a blast furnace input was
15 not included in the deductions to avoid double-counting.
16 Emissions from steel production in EAFs were estimated by deducting the carbon contained in the steel produced
17 from the carbon contained in the EAF anode, charge carbon, and scrap steel added to the EAF. Small amounts of
18 carbon from direct reduced iron, pig iron, and flux additions to the EAFs were also included in the EAF calculation.
19 For BOFs, estimates of carbon contained in EOF steel were deducted from carbon contained in inputs such as
20 natural gas, coke oven gas, fluxes, and pig iron. In each case, the carbon was calculated by multiplying material -
21 specific carbon contents by each material type (see Table 4-67). For EAFs, the amount of EAF anode consumed
22 was approximated by multiplying total EAF steel production by the amount of EAF anode consumed per metric ton
23 of steel produced (0.002 metric tons EAF anode per metric ton steel produced [AISI 2008c]). The amount of flux
24 (e.g., limestone and dolomite) used during steel manufacture was deducted from the Other Process Uses of
25 Carbonates source category to avoid double-counting.
26 Carbon dioxide emissions from the consumption of blast furnace gas and coke oven gas for other activities occurring
27 at the steel mill were estimated by multiplying the amount of these materials consumed for these purposes by the
28 material-specific carbon content (see Table 4-67).
29 Carbon dioxide emissions associated with the sinter production, direct reduced iron production, pig iron production,
30 steel production, and other steel mill activities were summed to calculate the total CO2 emissions from iron and steel
31 production (see Table 4-59 and Table 4-60).
32 Table 4-67: Material Carbon Contents for Iron and Steel Production
Material kg C/kg
Coke 0.83
Direct Reduced Iron 0.02
4-60 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Dolomite
EAF Carbon Electrodes
EAF Charge Carbon
Limestone
Pig Iron
Steel
Material
Coke Oven Gas
Blast Furnace Gas
0.13
0.82
0.83
0.12
0.04
0.01
kgC/GJ
12.1
70.8
Source: IPCC 2006, Table 4.3. Coke Oven Gas and
Blast Furnace Gas, Table 1.3.
1 The production process for sinter results in fugitive emissions of CH4, which are emitted via leaks in the production
2 equipment, rather than through the emission stacks or vents of the production plants. The fugitive emissions were
3 calculated by applying Tier 1 emission factors taken from the 2006 IPCC Guidelines (IPCC 2006) for sinter
4 production (see Table 4-68). Although the 1995 IPCC Guidelines (IPCC/UNEP/OECD/IEA 1995) provide a Tier 1
5 CH4 emission factor for pig iron production, it is not appropriate to use because CO2 emissions were estimated using
6 the Tier 2 mass balance methodology. The mass balance methodology makes a basic assumption that all carbon that
7 enters the pig iron production process either exits the process as part of a carbon-containing output or as CCh
8 emissions; the estimation of CH4 emissions is precluded. A preliminary analysis of facility-level emissions reported
9 by sinter facilities further supports this assumption and indicates that CH4 emissions are very low to negligible.
10 Estimation of CH4 emissions using a Tier 1 approach will significantly overestimate these emissions. The
11 production of direct reduced iron also results in emissions of CH4 through the consumption of fossil fuels (e.g.,
12 natural gas, etc.); however, these emission estimates are excluded due to data limitations.
13 Table 4-68: CH4 Emission Factors for Sinter and Pig Iron Production
Material Produced Factor Unit
Sinter 0.07 kg CHVmetric ton
Source: Sinter (IPCC 2006, Table 4.2)
14 Sinter consumption data for 1990 through 2014 were obtained from AiSTs Annual Statistical Report (AISI2004
15 through 2015a) and through personal communications with AISI (2008c) (see Table 4-69). In general, direct
16 reduced iron (DRI) consumption data were obtained from the USGS Minerals Yearbook - Iron and Steel Scrap
17 (USGS 1991 through 2014) and personal communication with the USGS Iron and Steel Commodity Specialist
18 (Fenton 2015). However, data for DRI consumed in EAFs were not available for the years 1990 and 1991. EAF
19 DRI consumption in 1990 and 1991 was calculated by multiplying the total DRI consumption for all furnaces by the
20 EAF share of total DRI consumption in 1992. Also, data for DRI consumed in BOFs were not available for the years
21 1990 through 1993. EOF DRI consumption in 1990 through 1993 was calculated by multiplying the total DRI
22 consumption for all furnaces (excluding EAFs and cupola) by the EOF share of total DRI consumption (excluding
23 EAFs and cupola) in 1994.
24 The Tier 1 CC>2 emission factors for sinter production and direct reduced iron production were obtained through the
25 2006 IPCC Guidelines (IPCC 2006). Time series data for pig iron production, coke, natural gas, fuel oil, sinter, and
26 pellets consumed in the blast furnace; pig iron production; and blast furnace gas produced at the iron and steel mill
27 and used in the metallurgical coke ovens and other steel mill activities were obtained from AISI's Annual Statistical
28 Report (AISI 2004 through 2015a) and through personal communications with AISI (2008c) (see Table 4-69 and
29 Table 4-70).
30 Data for EAF steel production, flux, EAF charge carbon, and natural gas consumption were obtained from AISI's
31 Annual Statistical Report (AISI 2004 through 2015a) and through personal communications with AISI (2006
32 through 2015b and 2008c). The factor for the quantity of EAF anode consumed per ton of EAF steel produced was
33 provided by AISI (2008c). Data for EOF steel production, flux, natural gas, natural ore, pellet sinter consumption as
34 well as EOF steel production were obtained from AISI's Annual Statistical Report (AISI 2004 through 2015a) and
35 through personal communications with AISI (2008c). Data for EAF and EOF scrap steel, pig iron, and DRI
36 consumption were obtained from the USGSMinerals Yearbook-Iron and Steel Scrap (USGS 1991 through 2014).
37 Data on coke oven gas and blast furnace gas consumed at the iron and steel mill (other than in the EAF, EOF, or
Industrial Processes and Product Use 4-61
-------
1 blastfurnace) were obtained from AISI's Annual Statistical Report (AISI 2004 through 2015a) and through personal
2 communications with AISI (2008c).
3 Data on blast furnace gas and coke oven gas sold for use as synthetic natural gas were obtained from EIA's Natural
4 Gas Annual (EIA 2015a). Carbon contents for direct reduced iron, EAF carbon electrodes, EAF charge carbon,
5 limestone, dolomite, pig iron, and steel were provided by the 2006IPCC Guidelines (IPCC 2006). The carbon
6 contents for natural gas, fuel oil, and direct injection coal were obtained from EIA (2015b) and EPA (2010). Heat
7 contents for fuel oil and direct injection coal were obtained from EIA (1992, 2011); natural gas heat content was
8 obtained from Table 37 of AISI's Annual Statistical Report (AISI 2004 through 2015a). Heat contents for coke oven
9 gas and blast furnace gas were provided in Table 37 of AISI's Annual Statistical Report (AISI 2004 through 2015a)
10 and confirmed by AISI staff (Carroll 2015).
11 Table 4-69: Production and Consumption Data for the Calculation of COz and CH4 Emissions
12 from Iron and Steel Production (Thousand Metric Tons)
Source/Activity Data
1990
2005
2010
2011 2012
2013
Sinter Production
Sinter Production 12,239
Direct Reduced Iron
Production
Direct Reduced Iron
Production 516
Pig Iron Production
Coke Consumption 24,946
Pig Iron Production 49,669
Direct Injection Coal
Consumption 1,485
EAF Steel Production
EAF Anode and Charge
Carbon Consumption 67J
Scrap Steel
Consumption 42,691
Flux Consumption 319
EAF Steel Production 33,511
EOF Steel Production
Pig Iron Consumption
Scrap Steel
Consumption 14,713
Flux Consumption 576
EOF Steel Production 43,973J
I 1
l,303l
13,832
37,222
2,573l
5,225
1,441
10,883
26,844
5,941 5,795 5,583
1,582
3,530 3,350
11,962 9,571 9,308
30,228 32,063 30,309
2,279 2,604 2,802
2,675
1,127
146,600
52,194
47,307 34,400
11,400
582
42,705
1,189 1,257 1,318 1,122
47,500 50,500 50,900 47,300
640 726 748 771
49,339 52,108 52,415 52,641
31,200 31,300 31,500 29,600
9,860 8,800 8,350 7,890
431 454 476 454
31,158 34,291 36,282 34,238
2014
5,521
2,113
11,136
29,375
2,425
1,127
48,873
771
55,174
23,755
5,917
454
33,000
13 Table 4-70: Production and Consumption Data for the Calculation of COz Emissions from Iron
14 and Steel Production (Million ft3 unless otherwise specified)
Source/Activity Data 1990 2005 2010 2011 2012 2013 2014
Pig Iron Production
Natural Gas
Consumption
Fuel Oil Consumption
(thousand gallons)
Coke Oven Gas
Consumption
Blast Furnace Gas
Production
EAF Steel Production
Natural Gas
56,27sB 59,844
163,397 16,170
22,03sB 16,557
1,439,380 1,299,980
47,814 59,132 62,469 48,812 47,734
27,505 21,378 19,240 17,468 16,674
14,233 17,772 18,608 17,710 16,896
911,180 1,063,326 1,139,578 1,026,973 1,000,536
Consumption
EOF Steel Production
Coke Oven Gas
Consumption
15,905|
19,985
524
10,403
546
6,263 11,145
554 568
10,514
568
9,622
524
4-62 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Other Activities
Coke Oven Gas
Consumption
Blast Furnace Gas
Consumption
224,883
1,414,778
97,132
1,295,520
80,626 90,718
907,9991,059,473
94,596
1,135,227
89,884
1,022,718
85,479
996,190
Uncertainty and Time-Series Consistency
2 The estimates of CCh emissions from metallurgical coke production are based on material production and
3 consumption data and average carbon contents. Uncertainty is associated with the total U.S. coking coal
4 consumption, total U.S. coke production and materials consumed during this process. Data for coking coal
5 consumption and metallurgical coke production are from different data sources (EIA) than data for other
6 carbonaceous materials consumed at coke plants (AISI), which does not include data for merchant coke plants.
7 There is uncertainty associated with the fact that coal tar and coke breeze production were estimated based on coke
8 production because coal tar and coke breeze production data were not available. Since merchant coke plant data is
9 not included in the estimate of other carbonaceous materials consumed at coke plants, the mass balance equation for
10 CO2 from metallurgical coke production cannot be reasonably completed. Therefore, for the purpose of this
11 analysis, uncertainty parameters are applied to primary data inputs to the calculation (i.e., coking coal consumption
12 and metallurgical coke production) only.
13 The estimates of €62 emissions from iron and steel production are based on material production and consumption
14 data and average carbon contents. There is uncertainty associated with the assumption that direct reduced iron and
15 sinter consumption are equal to production. There is uncertainty associated with the assumption that all coal used
16 for purposes other than coking coal is for direct injection coal; some of this coal may be used for electricity
17 generation. There is also uncertainty associated with the carbon contents for pellets, sinter, and natural ore, which
18 are assumed to equal the carbon contents of direct reduced iron. For EAF steel production, there is uncertainty
19 associated with the amount of EAF anode and charge carbon consumed due to inconsistent data throughout the time
20 series. Also for EAF steel production, there is uncertainty associated with the assumption that 100 percent of the
21 natural gas attributed to "steelmaking furnaces" by AISI is process-related and nothing is combusted for energy
22 purposes. Uncertainty is also associated with the use of process gases such as blast furnace gas and coke oven gas.
23 Data are not available to differentiate between the use of these gases for processes at the steel mill versus for energy
24 generation (i.e., electricity and steam generation); therefore, all consumption is attributed to iron and steel
25 production. These data and carbon contents produce a relatively accurate estimate of €62 emissions. However,
26 there are uncertainties associated with each.
27 For the purposes of the CH4 calculation from iron and steel production it is assumed that all of the CH4 escapes as
28 fugitive emissions and that none of the CH4 is captured in stacks or vents. Additionally, the €62 emissions
29 calculation is not corrected by subtracting the carbon content of the CH4, which means there may be a slight double-
30 counting of carbon as both €62 and CH4.
31 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-71 for metallurgical coke
32 production and iron and steel production. Total €62 emissions from metallurgical coke production and iron and
33 steel production were estimated to be between 47.2 and 63.6 MMT €62 Eq. at the 95 percent confidence level. This
34 indicates a range of approximately 15 percent below and 15 percent above the emission estimate of 55.4 MMT €62
35 Eq. Total CH4 emissions from metallurgical coke production and iron and steel production were estimated to be
36 between 0.008 and 0.01 MMT €62 Eq. at the 95 percent confidence level. This indicates a range of approximately
37 19 percent below and 19 percent above the emission estimate of 0.009 MMT CO2 Eq.
38 Table 4-71: Approach 2 Quantitative Uncertainty Estimates for COz and CH4 Emissions from
39 Iron and Steel Production and Metallurgical Coke Production (MMT COz Eq. and Percent)
„ „ 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
6 (MMT CCh Eq.) (MMT CCh Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
Metallurgical Coke & Iron
and Steel Production
Industrial Processes and Product Use 4-63
-------
Metallurgical Coke & Iron + + + +
and Steel Production
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
+ Does not exceed 0.05 MMT CO2 Eq.
1 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
2 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
3 above.
4 Recalculations Discussion
5 Several adjustments were incorporated into the emission calculations for the Iron and Steel Production and
6 Metallurgical Coke Production source categories. These adjustments applied to the entire time series from 1990 to
7 2014 and are briefly described below.
8 Previous Inventory reports included CH4 emissions calculated using a Tier 1 CH4 emission factor for three different
9 production processes: metallurgical coke, sinter, and pig iron. However, the use of a Tier 1 CH4 emission factor was
10 not appropriate and may significantly overestimate emissions for the metallurgical coke and pig iron production
11 processes in the U.S., because the CCh emissions for these production processes were estimated using the Tier 2
12 mass balance methodology. The Tier 2 mass balance methodology makes a basic assumption that all carbon that
13 enters the specific production process either exits the process as part of a carbon-containing output or as CCh
14 emissions; the estimation of CH4 emissions is necessarily precluded by definition. Because CC>2 emissions for the
15 sinter production process were estimated using a Tier 1 CC>2 emission factor, it is still appropriate to use a Tier 1
16 CH4 emission factor for the sinter production process. Due to exclusion of CH4 emissions from the metallurgical
17 coke and pig iron production processes, CH4 emissions reported in the Inventory were significantly reduced. This
18 assumption and the revisions are further supported by a preliminary analysis of facility-level greenhouse gas
19 emissions reported to EPA's GHGRP.
20 Previous Inventory reports have also relied significantly on activity data (i.e., production and input statistics) from
21 AISI's Annual Statistical Report (AISI2004 through 2015a); three key fuels used in the Tier 2 mass balance
22 methodology were natural gas, coke oven gas, and blast furnace gas. For all three of these fuels, volumetric
23 consumption was multiplied by a heat content to obtain the quantity of energy, which was then multiplied by carbon
24 content to obtain the quantity of carbon. The heat content of natural gas was obtained from EIA's Natural Gas
25 Annual (EIA 2015a) and varied from year to year with values ranging from 1,022 to 1,031 BTU/ft3, while the heat
26 contents of coke oven gas (500 BTU/ft3) and blast furnace gas (90 BTU/ft3) were obtained from the report, Energy
27 and Environmental Profile of the U.S. Iron and Steel Industry (DOE 2000). However, close examination of Table 37
28 of the AISI's Annual Statistical Report (AISI 2004 through 2015a) indicates that the reported quantities of natural
29 gas and blast furnace gas have different reporting bases based on heat contents (i.e., 1,000 BTU/ft3 for natural gas
30 and 95 BTU/ft3 for blast furnace gas); the reporting basis for coke oven gas is identically 500 BTU/ft3. AISI staff
31 confirmed that the reporting bases included in Table 37 of the AISI's Annual Statistical Report (AISI 2004 through
32 2015a) have been used dating back to at least 1990. Therefore, the use of other heat contents with AISI's data is not
33 appropriate. The heat content of natural gas was changed to 1,000 BTU/ft3 for all years in the time series and the
34 heat content of blast furnace gas was changed to 95 BTU/ft3. Because blast furnace gas is used as both an input and
35 an output in the Tier 2 mass balance methodology, the use of revised heat contents for natural gas and blast furnace
36 gas only resulted in a slight decrease in estimated CCh emissions; however, the CC>2 emissions for individual
37 production processes did change noticeably. For instance, across the entire time series, an increase in CC>2 emissions
38 from heating, annealing, and other processes was essentially offset by a decrease in €62 emissions from the iron
39 production process.
40 Planned Improvements
41 Future improvements involve evaluating and analyzing data reported under EPA's GHGRP to improve the emission
42 estimates for the Iron and Steel Production source category. Particular attention will be made to ensure time series
43 consistency of the emissions estimates presented in future Inventory reports, consistent with IPCC and UNFCCC
44 guidelines. This is required as the facility-level reporting data from EPA's GHGRP, with the program's initial
45 requirements for reporting of emissions in calendar year 2010, are not available for all inventory years (i.e., 1990
46 through 2009) as required for this Inventory. In implementing improvements and integration of data from EPA's
4-64 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 GHGRP, the latest guidance from the IPCC on the use of facility-level data in national inventories will be relied
2 upon.28
3 Additional improvements include accounting for emission estimates for the production of metallurgical coke to the
4 Energy chapter as well as identifying the amount of carbonaceous materials, other than coking coal, consumed at
5 merchant coke plants. Other potential improvements include identifying the amount of coal used for direct injection
6 and the amount of coke breeze, coal tar, and light oil produced during coke production. Efforts will also be made to
7 identify information to better characterize emissions from the use of process gases and fuels within the Energy and
8 Industrial Processes and Product Use chapters.
9 4.17 Ferroalloy Production (IPCC Source
Category 2C2)
10
11 Carbon dioxide and CH4 are emitted from the production of several ferroalloys. Ferroalloys are composites of iron
12 (Fe) and other elements such as silicon (Si), manganese (Mn), and chromium (Cr). Emissions from fuels consumed
13 for energy purposes during the production of ferroalloys are accounted for in the Energy chapter. Emissions from
14 the production of two types of ferrosilicon (25 to 55 percent and 56 to 95 percent silicon), silicon metal (96 to 99
15 percent silicon), and miscellaneous alloys (32 to 65 percent silicon) have been calculated. Emissions from the
16 production of ferrochromium and ferromanganese are not included here because of the small number of
17 manufacturers of these materials in the United States, and therefore, government information disclosure rules
18 prevent the publication of production data for these production facilities.
19 Similar to emissions from the production of iron and steel, CO2 is emitted when metallurgical coke is oxidized
20 during a high-temperature reaction with iron and the selected alloying element. Due to the strong reducing
21 environment, CO is initially produced, and eventually oxidized to CO2. A representative reaction equation for the
22 production of 50 percent ferrosilicon (FeSi) is given below:
23 Fe203 + 2Si02 + 7C -> 2FeSi + 7CO
24 While most of the carbon contained in the process materials is released to the atmosphere as CO2, a percentage is
25 also released as CH4 and other volatiles. The amount of CH4 that is released is dependent on furnace efficiency,
26 operation technique, and control technology.
27 When incorporated in alloy steels, ferroalloys are used to alter the material properties of the steel. Ferroalloys are
28 used primarily by the iron and steel industry, and production trends closely follow that of the iron and steel industry.
29 Twelve companies in the United States produce ferroalloys (USGS 2015a).
30 Emissions of CO2 from ferroalloy production in 2014 were 1.9 MMT CO2 Eq. (1,914 kt CO2) (see Table 4-72 and
31 Table 4-73), which is an 11 percent reduction since 1990. Emissions of CH4 from ferroalloy production in 2014
32 were 0.01 MMT CO2 Eq. (0.5 kt CH4), which is a 21 percent decrease since 1990.
33
34 Table 4-72: COz and CH4 Emissions from Ferroalloy Production (MMT COz Eq.)
Gas 1990 2005 2010 2011 2012 2013 2014
C02 2.2 1.4 1.7 1.7 1.9 1.8 1.9
CH4 + B +B+ + + + +
Total 2.2 1.4 1.7 1.7 1.9 1.8 1.9
+ Does not exceed 0.05 MMT CO2 Eq.
28 See.
Industrial Processes and Product Use 4-65
-------
Table 4-73: COz and ChU Emissions from Ferroalloy Production (kt)
Gas
1990
2005
2010 2011 2012 2013 2014
C02
CH4
+ Does not exceed 0.5 kt.
1,735
1,903
1
1,785
1,914
1
Methodology
Emissions of CC>2 and CH4 from ferroalloy production were calculated using a Tier 1 method from the 2006IPCC
Guidelines (IPCC 2006) by multiplying annual ferroalloy production by material-specific default emission factors
provided by IPCC (2006). The Tier 1 equations for CO2 and CH4 emissions are as follows:
4
5
6
7
8 Where,
9
10
11
12
13
14
15
16 Where,
17
18
19
20
Ec02
MPi
EFi
EcH4
MPi
EF,
EC02 =
CO2 emissions, metric tons
Production of ferroalloy type /', metric tons
Generic emission factor for ferroalloy type /', metric tons CO2/metric ton specific
ferroalloy product
X
CH4 emissions, kg
Production of ferroalloy type /', metric tons
Generic emission factor for ferroalloy type /', kg CHVmetric ton specific ferroalloy
product
21 Default emission factors were used because country-specific emission factors are not currently available. The
22 following emission factors were used to develop annual CO2 and CH4 estimates:
23 • Ferrosilicon, 25 to 55 percent Si and Miscellaneous Alloys, 32 to 65 percent Si - 2.5 metric tons
24 COVmetric ton of alloy produced; 1.0 kg CHVmetric ton of alloy produced.
25 • Ferrosilicon, 56 to 95 percent Si - 4.0 metric tons CCh/metric ton alloy produced; 1.0 kg CH4/metric ton of
26 alloy produced.
27 • Silicon Metal - 5.0 metric tons CCh/metric ton metal produced; 1.2 kg CHVmetric ton metal produced.
28 It was assumed that 100 percent of the ferroalloy production was produced using petroleum coke in an electric arc
29 furnace process (IPCC 2006), although some ferroalloys may have been produced with coking coal, wood, other
30 biomass, or graphite carbon inputs. The amount of petroleum coke consumed in ferroalloy production was
31 calculated assuming that the petroleum coke used is 90 percent C and 10 percent inert material (Onder and
32 Bagdoyan 1993).
33 Ferroalloy production data for 1990 through 2014 (see Table 4-74) were obtained from the USGS through the
34 Minerals Yearbook: Silicon (USGS 1996 through 2013) and the Mineral Industry Surveys: Silicon (USGS 2014,
35 2015b). The following data were available from the USGS publications for the time-series:
36 • Ferrosilicon, 25 to 55 percent Si: Annual production data were available from 1990 through 2010.
37 • Ferrosilicon, 56 to 95 percent Si: Annual production data were available from 1990 through 2010.
38 • Silicon Metal: Annual production data were available from 1990 through 2005. The production data for
39 2005 were used as proxy for 2006 through 2010.
4-66 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 • Miscellaneous Alloys, 32 to 65 percent Si: Annual production data were available from 1990-1999.
2 Starting 2000, USGS reported miscellaneous alloys and ferrosilicon containing 25 to 55 percent silicon as a
3 single category.
4 Starting with the 2011 publication, USGS reported all the ferroalloy production data as a single category (i.e., Total
5 Silicon Materials Production). This is due to the small number of ferroalloy manufacturers in the United States and
6 government information disclosure rules. Ferroalloy product shares developed from the 2010 production data (i.e.,
7 ferroalloy product production/total ferroalloy production) were used with the total silicon materials production
8 quantity to estimate the production quantity by ferroalloy product type for 2011 through 2014 (USGS 2013, 2014,
9 2015b).
10 Table 4-74: Production of Ferroalloys (Metric Tons)
Year
1990
2005
Ferrosilicon
25%-55%
Ferrosilicon
56%-95%
Silicon Metal
321,385
123,000
109,566
86,100
145,744
148,000
Misc. Alloys
32-65%
72,442
NA
2010
2011
2012
2013
2014
153,000
159,667
175,108
164,229
176,161
135,000
140,883
154,507
144,908
155,436
148,000
154,450
169,385
158,862
170,404
NA
NA
NA
NA
NA
NA (Not Available for product type, aggregated along with ferrosilicon, 25%-
55% Si)
11 Uncertainty and Time-Series Consistency
12 Annual ferroalloy production was reported by the USGS in three broad categories until the 2010 publication:
13 ferroalloys containing 25 to 55 percent silicon (including miscellaneous alloys), ferroalloys containing 56 to 95
14 percent silicon, and silicon metal (through 2005 only, 2005 value used as proxy for 2005 through 2010). Starting
15 with the 2011 Minerals Yearbook, USGS started reporting all the ferroalloy production under a single category:
16 Total silicon materials production. The total silicon materials quantity was allocated across the three categories
17 based on the 2010 production shares for the three categories. Refer to the Methodology section for further details.
18 Additionally, production data for silvery pig iron (alloys containing less than 25 percent silicon) are not reported by
19 the USGS to avoid disclosing proprietary company data. Emissions from this production category, therefore, were
20 not estimated.
21 Also, some ferroalloys may be produced using wood or other biomass as a primary or secondary carbon source
22 (carbonaceous reductants), however information and data regarding these practices were not available. Emissions
23 from ferroalloys produced with wood or other biomass would not be counted under this source because wood-based
24 carbon is of biogenic origin.29 Even though emissions from ferroalloys produced with coking coal or graphite inputs
25 would be counted in national trends, they may be generated with varying amounts of €62 per unit of ferroalloy
26 produced. The most accurate method for these estimates would be to base calculations on the amount of reducing
27 agent used in the process, rather than the amount of ferroalloys produced. These data, however, were not available,
28 and are also often considered confidential business information.
29 Emissions of CH4 from ferroalloy production will vary depending on furnace specifics, such as type, operation
30 technique, and control technology. Higher heating temperatures and techniques such as sprinkle charging will
31 reduce CH4 emissions; however, specific furnace information was not available or included in the CH4 emission
32 estimates.
33 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-75. Ferroalloy
34 production CC>2 emissions were estimated to be between 1.7 and 2.1 MMT CCh Eq. at the 95 percent confidence
35 level. This indicates a range of approximately 12 percent below and 12 percent above the emission estimate of 1.9
29 Emissions and sinks of biogenic carbon are accounted for in the Land Use, Land-Use Change, and Forestry chapter.
Industrial Processes and Product Use 4-67
-------
1 MMT CO2 Eq. Ferroalloy production CH4 emissions were estimated to be between a range of approximately 12
2 percent below and 12 percent above the emission estimate of 0.01 MMT CCh Eq.
3 Table 4-75: Approach 2 Quantitative Uncertainty Estimates for COz Emissions from
4 Ferroalloy Production (MMT COz Eq. and Percent)
^ „ 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
6 (MMT CCh Eq.) (MMT CCh Eq.) (%)
17
18
Ferroalloy Production
Ferroalloy Production
C02 1.9
CH4 +
Lower Upper Lower
Bound Bound Bound
1.7 2.1 -12%
+ + -12%
Upper
Bound
+12%
+12%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
+ Does not exceed 0.05 MMT CO2 Eq.
5 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
6 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
7 above.
Planned Improvements
9 Future improvements involve evaluating and analyzing data reported under EPA's GHGRP that would be useful to
10 improve the emission estimates for the Ferroalloy Production source category. Particular attention will be made to
11 ensure time series consistency of the emissions estimates presented in future Inventory reports, consistent with IPCC
12 and UNFCCC guidelines. This is required as the facility-level reporting data from EPA's GHGRP, with the
13 program's initial requirements for reporting of emissions in calendar year 2010, are not available for all inventory
14 years (i.e., 1990 through 2009) as required for this Inventory. In implementing improvements and integration of data
15 from EPA's GHGRP, the latest guidance from the IPCC on the use of facility-level data in national inventories will
16 be relied upon.30
4.18 Aluminum Production (IPCC Source
Category 2C3) (TO BE UPDATED)
19 Aluminum is a light-weight, malleable, and corrosion-resistant metal that is used in many manufactured products,
20 including aircraft, automobiles, bicycles, and kitchen utensils. As of recent reporting, the United States was the
2 1 fourth largest producer of primary aluminum, with approximately 4 percent of the world total production (USGS
22 2014). The United States was also a major importer of primary aluminum. The production of primary aluminum
23 in addition to consuming large quantities of electricity — results in process-related emissions of CCh and two
24 perfluorocarbons (PFCs): Perfluoromethane (CF4) and perfluoroethane (CJe).
25 CO2 is emitted during the aluminum smelting process when alumina (aluminum oxide, A^Os) is reduced to
26 aluminum using the Hall-Heroult reduction process. The reduction of the alumina occurs through electrolysis in a
27 molten bath of natural or synthetic cryolite (NasAlFe). The reduction cells contain a carbon lining that serves as the
28 cathode. Carbon is also contained in the anode, which can be a C mass of paste, coke briquettes, or prebaked C
29 blocks from petroleum coke. During reduction, most of this C is oxidized and released to the atmosphere as CO2.
30 Process emissions of CO2 from aluminum production were estimated to be 3.3 MMT CO2 Eq. (3,255 kt) in 2013
3 1 (see Table 4-76). The C anodes consumed during aluminum production consist of petroleum coke and, to a minor
32 extent, coal tar pitch. The petroleum coke portion of the total CO2 process emissions from aluminum production is
33 considered to be a non-energy use of petroleum coke, and is accounted for here and not under the CO2 from Fossil
30
See.
4-68 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Fuel Combustion source category of the Energy sector. Similarly, the coal tar pitch portion of these CO2 process
2 emissions is accounted for here.
3 Table 4-76: COz Emissions from Aluminum Production (MMT COz Eq. and kt)
Year
1990
2005
1
2009
2010
2011
2012
2013
MMT CO2 Eq.
6.8
4.1
3.0
2.7
3.3
3.4
3.3
kt
6,831
4,142
3,009
2,722
3,292
3,439
3,255
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
In addition to CC>2 emissions, the aluminum production industry is also a source of PFC emissions. During the
smelting process, when the alumina ore content of the electrolytic bath falls below critical levels required for
electrolysis, rapid voltage increases occur, which are termed "anode effects." These anode effects cause C from the
anode and fluorine from the dissociated molten cryolite bath to combine, thereby producing fugitive emissions of
CF4 and C2p6. In general, the magnitude of emissions for a given smelter and level of production depends on the
frequency and duration of these anode effects. As the frequency and duration of the anode effects increase,
emissions increase.
Since 1990, emissions of CF4 and C2p6 have declined by 87 percent and 81 percent, respectively, to 2.3 MMT CCh
Eq. of CF4 (0.31 kt) and 0.7 MMT CO2 Eq. of C2F6 (0.05 kt) in 2013, as shown in Table 4-77 and Table 4-78. This
decline is due both to reductions in domestic aluminum production and to actions taken by aluminum smelting
companies to reduce the frequency and duration of anode effects. These actions include technology and operational
changes such as employee training, use of computer monitoring, and changes in alumina feeding techniques. Since
1990, aluminum production has declined by 52 percent, while the combined CF4 and C2Pe emission rate (per metric
ton of aluminum produced) has been reduced by 71 percent. Emissions increased by approximately 1 percent
between 2012 and 2013 due to a slight increase in both CF4 and C2Pe emissions per metric ton of aluminum
produced.
Table 4-77: PFC Emissions from Aluminum Production (MMT COz Eq.)
Year
1990
2005
2009
2010
2011
2012
2013
CF4
17.9
2.9
1.5
1.4
2.7
2.3
2.3
C2F6
3.5
• ••
0.6
|
0.4
0.5
0.8
0.7
0.7
Total
21.5
3.4
1.9
1.9
3.5
2.9
3.0
22
Note: Totals may not sum due to
independent rounding.
Industrial Processes and Product Use 4-69
-------
Table 4-78: PFC Emissions from Aluminum Production (kt)
10
Year
1990
2005
2009
2010
2011
2012
2013
CF4
2.4
0.4
|
0.2
0.2
0.4
0.3
0.3
C2F6
0.3
+
+
+
0.1
0.1
0.1
+ Does not exceed 0.05 kt.
2 In 2013, U.S. primary aluminum production totaled approximately 1.9 million metric tons, a 6 percent decrease from
3 2012 production levels (USAA 2014). In 2013, five companies managed production at ten operational primary
4 aluminum smelters. Three smelters were closed temporarily for the entire year in 2013 (USGS 2014). During 2013,
5 monthly U.S. primary aluminum production was lower for every month in 2013, when compared to the
6 corresponding months in 2012 (USAA 2014).
7 For 2014, total production was approximately 1.7 million metric tons compared to 1.9 million metric tons in 2013, a
8 12 percent decrease (USAA 2014). Based on the decrease in production, process CC>2 and PFC emissions are likely
9 to be lower in 2014 compared to 2013 if there are no significant changes in process controls at operational facilities.
Methodology
11 Process CC>2 and perfluorocarbon (PFC)—i.e., perfluoromethane (CF4) and perfluoroethane (C2F6)—emission
12 estimates from primary aluminum production for 2010 through 2013 are available from EPA's GHGRP—Subpart F
13 (Aluminum Production) (EPA 2014). Under EPA's GHGRP, facilities began reporting primary aluminum
14 production process emissions (for 2010) in 2011; as a result, GHGRP data (for 2010 through 2013) are available to
15 be incorporated into the Inventory. EPA's GHGRP mandates that all facilities that contain an aluminum production
16 process must report: CF4 and C2p6 emissions from anode effects in all prebake and Soderberg electrolysis cells,
17 carbon dioxide (CCh) emissions from anode consumption during electrolysis in all prebake and Soderberg cells, and
18 all CO2 emissions from onsite anode baking. To estimate the process emissions, EPA's GHGRP uses the process-
19 specific equations (and certain technology-specific defaults) detailed in subpart F (aluminum production).31 These
20 equations are based on the Tier 2/Tier 3 IPCC (2006) methods for primary aluminum production, and Tier 1
21 methods when estimating missing data elements. It should be noted that the same methods (i.e., 2006 IPCC
22 Guidelines) were used for estimating the emissions prior to the availability of the reported GHGRP data in the
23 Inventory.
24 Process COz Emissions from Anode Consumption and Anode Baking
25 CO2 emission estimates for the years prior to the introduction of EPA's GHGRP in 2010 were estimated with IPCC
26 (2006) methods, but individual facility reported data were combined with process-specific emissions modeling.
27 These estimates were based on information previously gathered from EPA's Voluntary Aluminum Industrial
28 Partnership (VAIP) program, U.S. Geological Survey (USGS) Mineral Commodity reviews, and The Aluminum
29 Association (USAA) statistics, among other sources. Since pre- and post-GHGRP estimates use the same
30 methodology, emission estimates are comparable across the time series.
31 Most of the CO2 emissions released during aluminum production occur during the electrolysis reaction of the C
32 anode, as described by the following reaction:
31 See Code of Federal Regulations, Title 40: Protection of Environment, Part 98: Mandatory Greenhouse Gas Reporting,
Subpart F—Aluminum Production. Available online at:
.
4-70 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 2A1203 + 3C -» 4A1 + 3C02
2 For prebake smelter technologies, CO2 is also emitted during the anode baking process. These emissions can
3 account for approximately 10 percent of total process CO2 emissions from prebake smelters.
4 Depending on the availability of smelter-specific data, the CO2 emitted from electrolysis at each smelter was
5 estimated from: (1) The smelter's annual anode consumption, (2) the smelter's annual aluminum production and
6 rate of anode consumption (per ton of aluminum produced) for previous and/or following years, or, (3) the smelter's
7 annual aluminum production and IPCC default CO2 emission factors. The first approach tracks the consumption and
8 carbon content of the anode, assuming that all C in the anode is converted to CO2. Sulfur, ash, and other impurities
9 in the anode are subtracted from the anode consumption to arrive at a C consumption figure. This approach
10 corresponds to either the IPCC Tier 2 or Tier 3 method, depending on whether smelter-specific data on anode
11 impurities are used. The second approach interpolates smelter-specific anode consumption rates to estimate
12 emissions during years for which anode consumption data are not available. This approach avoids substantial errors
13 and discontinuities that could be introduced by reverting to Tier 1 methods for those years. The last approach
14 corresponds to the IPCC Tier 1 method (2006), and is used in the absence of present or historic anode consumption
15 data.
16 The equations used to estimate CO2 emissions in the Tier 2 and 3 methods vary depending on smelter type (IPCC
17 2006). For Prebake cells, the process formula accounts for various parameters, including net anode consumption,
18 and the sulfur, ash, and impurity content of the baked anode. For anode baking emissions, the formula accounts for
19 packing coke consumption, the sulfur and ash content of the packing coke, as well as the pitch content and weight of
20 baked anodes produced. For Soderberg cells, the process formula accounts for the weight of paste consumed per
21 metric ton of aluminum produced, and pitch properties, including sulfur, hydrogen, and ash content.
22 Through the VAIP, anode consumption (and some anode impurity) data have been reported for 1990, 2000, 2003,
23 2004, 2005, 2006, 2007, 2008, and 2009. Where available, smelter-specific process data reported under the VAIP
24 were used; however, if the data were incomplete or unavailable, information was supplemented using industry
25 average values recommended by IPCC (2006). Smelter-specific CO2 process data were provided by 18 of the 23
26 operating smelters in 1990 and 2000, by 14 out of 16 operating smelters in 2003 and 2004, 14 out of 15 operating
27 smelters in 2005, 13 out of 14 operating smelters in 2006, 5 out of 14 operating smelters in 2007 and 2008, and 3 out
28 of 13 operating smelters in 2009. For years where CO2 emissions data or CO2 process data were not reported by
29 these companies, estimates were developed through linear interpolation, and/or assuming representative (e.g.,
30 previously reported or industry default) values.
31 In the absence of any previous historical smelter specific process data (i.e., 1 out of 13 smelters in 2009, 1 out of 14
32 smelters in 2006, 2007, and 2008, 1 out of 15 smelters in 2005, and 5 out of 23 smelters between 1990 and 2003),
33 CO2 emission estimates were estimated using Tier 1 Soderberg and/or Prebake emission factors (metric ton of CO2
34 per metric ton of aluminum produced) from IPCC (2006).
35 Process PFC Emissions from Anode Effects
36 Smelter-specific PFC emissions from aluminum production for 2010 through 2013 were reported to EPA under its
37 GHGRP. To estimate their PFC emissions and report them under EPA's GHGRP, smelters use an approach
38 identical to the Tier 3 approach in the 2006IPCC Guidelines (IPCC 2006). Specifically, they use a smelter-specific
39 slope coefficient as well as smelter-specific operating data to estimate an emission factor using the following
40 equation:
41 PFC (CF4 or C2F6) kg/metric ton Al = S x (Anode Effect Minutes/Cell-Day)
42 where,
43 S = Slope coefficient ((kg PFC/metric ton Al)/( Anode Effect Minutes/Cell-Day))
44 Anode Effect Minutes/Cell-Day = (Anode Effect Frequency/Cell-Day) x Anode Effect Duration (minutes)
45 They then multiply this emission factor by aluminum production to estimate PFC emissions. All U. S. aluminum
46 smelters are required to report their emissions under EPA's GHGRP.
47 PFC emissions for the years prior to 2010 were estimated using the same equation, but the slope-factor used for
48 some smelters was technology-specific rather than smelter-specific, making the method a Tier 2 rather than a Tier 3
Industrial Processes and Product Use 4-71
-------
1 approach for those smelters. Emissions and background data were reported to EPA under the VAIP. For 1990
2 through 2009, smelter-specific slope coefficients were available and were used for smelters representing between 30
3 and 94 percent of U.S. primary aluminum production. The percentage changed from year to year as some smelters
4 closed or changed hands and as the production at remaining smelters fluctuated. For smelters that did not report
5 smelter-specific slope coefficients, IPCC technology-specific slope coefficients were applied (IPCC 2006). The
6 slope coefficients were combined with smelter-specific anode effect data collected by aluminum companies and
7 reported under the VAIP to estimate emission factors over time. For 1990 through 2009, smelter-specific anode
8 effect data were available for smelters representing between 80 and 100 percent of U.S. primary aluminum
9 production. Where smelter-specific anode effect data were not available, representative values (e.g., previously
10 reported or industry averages) were used.
11 For all smelters, emission factors were multiplied by annual production to estimate annual emissions at the smelter
12 level. For 1990 through 2009, smelter-specific production data were available for smelters representing between 30
13 and 100 percent of U.S. primary aluminum production. (For the years after 2000, this percentage was near the high
14 end of the range.) Production at non-reporting smelters was estimated by calculating the difference between the
15 production reported under VAIP and the total U.S. production supplied by USGS or USAA, and then allocating this
16 difference to non-reporting smelters in proportion to their production capacity. Emissions were then aggregated
17 across smelters to estimate national emissions.
18 Between 1990 and 2009, production data were provided under the VAIP by 21 of the 23 U.S. smelters that operated
19 during at least part of that period. For the non-reporting smelters, production was estimated based on the difference
20 between reporting smelters and national aluminum production levels (from USGS and USAA), with allocation to
21 specific smelters based on reported production capacities (from USGS).
22 National primary aluminum production data for 2013 were obtained via The Aluminum Association (USAA 2014).
23 For 1990 through 2001, and 2006 (see Table 4-79) data were obtained from USGS Mineral Industry Surveys:
24 Aluminum Annual Report (USGS 1995, 1998, 2000, 2001, 2002, 2007). For 2002 through 2005, and 2007 through
25 2011, national aluminum production data were obtained from the USAA's Primary Aluminum Statistics (USAA
26 2004-2006,2008-2013).
27 Table 4-79: Production of Primary Aluminum (kt)
Year kt
1990 4,048
2005 2,478
2009 1,727
2010 1,727
2011 1,986
2012 2,070
2013 1,948
28 Uncertainty and Time Series Consistency
29 Uncertainty was assigned to the CCh, CF4, and C2p6 emission values reported by each individual facility to EPA's
30 GHGRP. As previously mentioned, the methods for estimating emissions for EPA's GHGRP and this report are the
31 same, and follow the IPCC (2006) methodology. As a result, it was possible to assign uncertainty bounds (and
32 distributions) based on an analysis of the uncertainty associated with the facility-specific emissions estimated for
33 previous Inventory years. Uncertainty surrounding the reported CC>2, CF4, and C2p6 emission values were
34 determined to have a normal distribution with uncertainty ranges of ±6, ±16, and ±20 percent, respectively. A
35 Monte Carlo analysis was applied to estimate the overall uncertainty of the CCh, CF4, and C2F6 emission estimates
36 for the U.S. aluminum industry as a whole, and the results are provided below.
37 The results of this Approach 2 quantitative uncertainty analysis are summarized in Table 4-80. Aluminum
38 production-related CO2 emissions were estimated to be between 3.2 and 3.3 MMT CO2 Eq. at the 95 percent
39 confidence level. This indicates a range of approximately 2 percent below to 2 percent above the emission estimate
40 of 3.3 MMT CO2 Eq. Also, production-related CF4 emissions were estimated to be between 2.2 and 2.4 MMT CO2
4-72 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Eq. at the 95 percent confidence level. This indicates a range of approximately 6 percent below to 7 percent above
2 the emission estimate of 2.3 MMT CC>2 Eq. Finally, aluminum production-related C2p6 emissions were estimated to
3 be between 0.6 and 0.7 MMT CCh Eq. at the 95 percent confidence level. This indicates a range of approximately
4 11 percent below to 11 percent above the emission estimate of 0.7 MMT CCh Eq.
5 Table 4-80: Approach 2 Quantitative Uncertainty Estimates for COz and PFC Emissions from
6 Aluminum Production (MMT COz Eq. and Percent)
„ „ 2013 Emission Estimate Uncertainty Range Relative to Emission Estimate3
6 (MMT CCh Eq.) (MMT CCh Eq.) (%)
11
16
Aluminum Production
Aluminum Production
Aluminum Production
CO2
CF4
C2F6
3.3
2.3
0.7
Lower
Bound
3.2
2.2
0.6
Upper
Bound
3.3
2.4
0.7
Lower
Bound
-2%
-6%
-11%
Upper
Bound
+2%
+7%
+11%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
7
8 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
9 through 2013. Details on the emission trends through time are described in more detail in the Methodology section,
10 above.
QA/QC and Verification
12 Tier 1 and Tier 2 QA/QC activities were conducted consistent with the U.S. QA/QC plan. Source-specific quality
13 control measures for Aluminum Production included checking input data, documentation, and calculations to ensure
14 data were properly handled through the inventory process. Errors that were found during this process were
15 corrected as necessary.
Recalculations Discussion
17 For the current Inventory, emission estimates have been revised to reflect the GWPs provided in the IPCC Fourth
18 Assessment Report (AR4) (IPCC 2007). AR4 GWP values differ slightly from those presented in the IPCC Second
19 Assessment Report (SAR) (IPCC 1996) (used in the previous Inventory reports) which results in time-series
20 recalculations for most Inventory sources. Under the most recent reporting guidelines (UNFCCC 2014), countries
21 are required to report using the AR4 GWPs, which reflect an updated understanding of the atmospheric properties of
22 each greenhouse gas. The GWPs of CH4 and most fluorinated greenhouse gases have increased, leading to an
23 overall increase in CCh-equivalent emissions from PFCs. The AR4 GWPs have been applied across the entire time
24 series for consistency. For more information please see the Recalculations and Improvements Chapter.
25 As a result, emission estimates for each year from 1990 to 2012 increased by 14 percent for CF4, and increased by
26 33 percent for C2p6, relative to the emission estimates in the previous Inventory report.
27 Planned Improvements
28 Future improvements involve plans to replace proxy (e.g., interpolated) data with additional historical VAIP data
29 that recently became available in order to calculate more accurate PFC emission estimates for the historical time
30 series.
Industrial Processes and Product Use 4-73
-------
4.19 Magnesium Production and Processing
(IPCC Source Category 2C4) (TO BE
UPDATED)
4 The magnesium metal production and casting industry uses sulfur hexafluoride (SF6) as a cover gas to prevent the
5 rapid oxidation of molten magnesium in the presence of air. Sulfur hexafluoride has been used in this application
6 around the world for more than thirty years. A dilute gaseous mixture of SF6 with dry air and/or CC>2 is blown over
7 molten magnesium metal to induce and stabilize the formation of a protective crust. A small portion of the SF6
8 reacts with the magnesium to form a thin molecular film of mostly magnesium oxide and magnesium fluoride. The
9 amount of SF6 reacting in magnesium production and processing is considered to be negligible and thus all SF6 used
10 is assumed to be emitted into the atmosphere. , Alternative cover gases, such as AM-cover™ (containing HFC-
11 134a), Novec™ 612 (FK-5-1-12) and dilute SCh systems can, and are being used by some facilities in the United
12 States. However, many facilities in the United States are still using traditional SF6 cover gas systems.
13 The magnesium industry emitted 1.4 MMT CO2 Eq. (0.06 kt) of SF6, 0.08 MMT CO2 Eq. (0.06 kt) of HFC-134a,
14 and 0.002 MMT €62 Eq. (2.1 kt) of CC>2, in 2013. This represents a decrease of approximately 8 percent from total
15 2012 emissions (see Table 4-81). The decrease can be attributed to reduction in primary, secondary, and die casting
16 SF6 emissions between 2012 and 2013 as reported through EPA's GHGRP, with the largest absolute reduction being
17 seen for secondary emissions. The reduction in SF6 emissions is likely due in part to decreased production from
18 reporting facilities in 2013. The decrease in SF6 emissions can also be attributed by continuing industry efforts to
19 utilize SF6 alternatives, such as HFC-134a, Novec™612 and SCh, to reduce greenhouse gas emissions. In 2013, total
20 HFC-134a emissions increased from 0.01 MMT CO2 Eq. to 0.08 MMT CO2 Eq., while the FK 5-1-12 emissions
21 were constant. The emissions of carrier gas, CC>2, also decreased from 2.3 kt in 2012 to 2.1 kt in 2013.
22 Table 4-81: SFe, HFC-134a, FK 5-1-12 and COz Emissions from Magnesium Production and
23 Processing (MMT COz Eq.)
24
25
Year
SFe
HFC-134a
CO2
FK5-1-12
Total3
1990
5.2
0.0
+
0.0
5.2
2005
2.7
0.0
+
0.0
2.8
2009
1.6
+
+
+
1.7
2010
2.1
+
+
+
2.1
2011
2.8
+
+
+
2.8
2012
1.6
+
+
+
1.7
2013
1.4
0.1
+
+
1.5
Note: Totals may not sum due to independent rounding.
+ Does not exceed 0.05 MMT CO2 Eq.
a Total does not include FK 5-1-12. Values shown for informational purposes only.
Table 4-82: SFe, HFC-134a, FK 5-1-12 and COz Emissions from Magnesium Production and
Processing (kt)
Year
SFe
HFC-134a
C02
FK 5-1-1 2
1990
0.2
0.0
1.4
0.0
2005
0.1
0.0
2.9
0.0
2009
0.1
+
1.2
+
2010
0.1
+
1.3
+
2011
0.1
+
3.1
+
2012
0.1
+
2.3
+
2013
0.1
0.1
2.1
+
+ Does not exceed 0.5 kt.
26 Methodology
27 Emission estimates for the magnesium industry incorporate information provided by some industry participants in
28 EPA's SF6 Emission Reduction Partnership for the Magnesium Industry as well as emissions data reported through
29 subpart T (Magnesium Production and Processing) of the EPA's GHGRP. The Partnership started in 1999 and, in
4-74 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 2010, participating companies represented 100 percent of U.S. primary and secondary production and 16 percent of
2 the casting sector production (i.e., die, sand, permanent mold, wrought, and anode casting). SF6 Emissions for 1999
3 through 2010 from primary production, secondary production (i.e., recycling), and die casting were generally
4 reported by Partnership participants. Partners reported their SF6 consumption, which was assumed to be equivalent
5 to emissions. Along with SF6, some Partners also reported their HFC-134a and FK 5-1-12 usage, which is assumed
6 to be equal to emissions. 2010 was the last reporting year under the Partnership. Emissions data for 2011 through
7 2013 were obtained through EPA's GHGRP. Under the program, owners or operators of facilities that have a
8 magnesium production or casting process must report emissions from use of cover or carrier gases, which include
9 SF6, HFC-134a, FK 5-1-12 and CC>2. Consequently, cover and carrier gas emissions from magnesium production
10 and processing were estimated for three time periods, depending on the source of the emissions data: 1990 through
11 1998, 1999 through 2010, and 2011 through 2013. The methodologies described below also make use of
12 magnesium production data published by the U.S. Geological Survey (USGS).
13 1990 through 1998
14 To estimate emissions for 1990 through 1998, industry SF6 emission factors were multiplied by the corresponding
15 metal production and consumption (casting) statistics from USGS. For this period, it was assumed that there was no
16 use of HFC-134a or FK 5-1-12 cover gases and hence emissions were not estimated for these alternatives.
17 SFe emission factors from 1990 through 1998 were based on a number of sources and assumptions. Emission
18 factors for primary production were available from U.S. primary producers for 1994 and 1995. The primary
19 production emission factors were 1.2 kg SF6 per metric ton for 1990 through 1993, and 1.1 kg SF6 per metric ton for
20 1994 through 1997. The emission factor for secondary production from 1990 through 1998 was assumed to be
21 constant at the 1999 average Partner value. Emission factor for die casting of 4.1 kg SF6 per metric ton was
22 available for the mid-1990s from an international survey (Gjestland & Magers 1996) that was used for years 1990
23 through 1996. For 1996 through 1998, the emission factor for die casting was assumed to decline linearly to the
24 level estimated based on Partner reports in 1999. This assumption is consistent with the trend in SF6 sales to the
25 magnesium sector that is reported in the RAND survey of major SF6 manufacturers, which shows a decline of 70
26 percent from 1996 to 1999 (RAND 2002). Sand casting emission factors for 1990 through 2001 were assumed to be
27 the same as the 2002 emission factor. The emission factors for the other processes (i.e., permanent mold, wrought,
28 and anode casting), about which less is known, were assumed to remain constant at levels defined in Table 4-81.
29 These emission factors for the other processes (i.e., permanent mold, wrought, and anode casting) were based on
30 discussions with industry representatives.
31 The quantities of CCh carrier gas used for each production type have been estimated using the 1999 estimated €62
32 emissions data and the annual calculated rate of change of SF6 use in the 1990 through 1999 time period. For each
33 year and production type, the rate of change of SF6 use between the current year and the subsequent year was first
34 estimated. This rate of change is then applied to the CCh emissions of the subsequent year to determine the CCh
3 5 emission of the current year. The emissions of carrier gases for permanent mold, wrought and anode processes are
36 not estimated in this Inventory.
37 1999 through 2010
38 The 1999 through 2010 emissions from primary and secondary production are based on information provided by
39 EPA's industry Partners. In some instances, there were years of missing Partner data, including SF6 consumption
40 and metal processed. For these situations, emissions were estimated through interpolation where possible, or by
41 holding company-reported emissions (as well as production) constant from the previous year. For alternative cover
42 gases, including HFC-134a and FK 5-1-12, mainly reported data was relied upon. That is, unless a Partner reported
43 using an alternative cover gas, it was not assumed it was used. Emissions of alternate gases were also estimated
44 through linear interpolation where possible.
45 The die casting emission estimates for 1999 through 2010 are also based on information supplied by industry
46 Partners. When a Partner was determined to be no longer in production, its metal production and usage rates were
47 set to zero. Missing data on emissions or metal input was either interpolated or held constant at the last available
48 reported value. In 1999 and from 2008 through 2010, Partners did not account for all die casting tracked by USGS,
49 and, therefore, it was necessary to estimate the emissions of die casters who were not Partners. For 1999, die casters
50 who were not Partners were assumed to be similar to Partners who cast small parts. Due to process requirements,
Industrial Processes and Product Use 4-75
-------
1 these casters consume larger quantities of SF6 per metric ton of processed magnesium than casters that process large
2 parts. Consequently, emission estimates from this group of die casters were developed using an average emission
3 factor of 5.2 kg SF6 per metric ton of magnesium. This emission factor was developed using magnesium production
4 and SF6 usage data for the year 1999. For 2008 through 2010, the characteristics of the die casters who were not
5 Partners were not well known, and therefore the emission factor for these die casters was set equal to 3.0 kg SF6 per
6 metric ton of magnesium, the average of the emission factors reported over the same period by the die casters who
7 were Partners.
8 The emissions from other casting operations were estimated by multiplying emission factors (kg SF6 per metric ton
9 of metal produced or processed) by the amount of metal produced or consumed from USGS, with the exception of
10 some years for which Partner sand casting emissions data are available. The emission factors for sand casting
11 activities were acquired through the data reported by the Partnership for 2002 to 2006. For 1999-2001, the sand
12 casting emission factor was held constant at the 2002 Partner-reported level. For 2007 through 2010, the sand
13 casting Partner did not report and the reported emission factor from 2005 was applied to the Partner and to all other
14 sand casters.
15 The emission factors for primary production, secondary production and sand casting for the 1999 to 2010 are not
16 published to protect company-specific production information. However, the emission factor for primary production
17 has not risen above the average 1995 Partner value of 1.1 kg SF6 per metric ton. The emission factors for the other
18 industry sectors (i.e., permanent mold, wrought, and anode casting) were based on discussions with industry
19 representatives. The emission factors for casting activities are provided below in Table 4-83.
20 The emissions of HFC-134a and FK-5-1-12 were included in the estimates for only instances where Partners
21 reported that information to the Partnership. Emissions of these alternative cover gases were not estimated for
22 instances where emissions were not reported.
23 CO2 carrier gas emissions were estimated using the emission factors developed based on GHGRP-reported carrier
24 gas and cover gas data, by production type. It was assumed that the use of carrier gas, by production type, is
25 proportional to the use of cover gases. Therefore, an emission factor, in kg CC>2 per kg cover gas and weighted by
26 the cover gases used, was developed for each of the production types. GHGRP data on which these emissions
27 factors are based was available for primary, secondary, die casting and sand casting. The emission factors were
28 applied to the total quantity of all cover gases used (SF6, HFC-134a, and FK5-1-12) by production type in this time
29 period. Carrier gas emissions for the 1999 through 2010 time period were only estimated for those Partner
30 companies that reported using CCh as a carrier gas through the GHGRP. Using this approach helped ensure time
31 series consistency. The emissions of carrier gases for permanent mold, wrought and anode processes are not
32 estimated in this Inventory.
33 Table 4-83: SFe Emission Factors (kg SFe per metric ton of magnesium)
Year Die Casting3 Permanent Mold Wrought Anodes
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2.14b
0.72
0.72
0.71
0.81
0.79
0.77
0.88
0.64
0.10
2.30
2.94
2
2
2
2
2
2
2
2
2
2
2
2
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
a Weighted average includes all die casters, Partners and non-Partners. For
the majority of the time series (2000-2007), Partners made up 100 percent
of die casters in the U.S.
b Weighted average that includes an estimated emission factor of 5.2 kg SFe
per metric ton of magnesium for die casters that do not participate in the
Partnership.
4-76 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
i 2011 through 2013
2 For 2011 through 2013, for the primary and secondary producers, GHGRP-reported cover and carrier gases
3 emissions data were used. For die and sand casting, some emissions data was obtained through EPA's GHGRP. The
4 balance of the emissions for these industry segments were estimated based on previous Partner reporting (i.e., for
5 Partners that did not report emissions through EPA's GHGRP) or were estimated by multiplying emission factors by
6 the amount of metal produced or consumed. Partners who did not report through EPA's GHGRP were assumed to
7 have continued to emit SF6 at the last reported level, which was from 2010 in most cases. All Partners were
8 assumed to have continued to consume magnesium at the last reported level. Where the total metal consumption
9 estimated for the Partners fell below the U.S. total reported by USGS, the difference was multiplied by the emission
10 factors discussed in the section above. For the other types of production and processing (i.e., permanent mold,
11 wrought, and anode casting), emissions were estimated by multiplying the industry emission factors with the metal
12 production or consumption statistics obtained from USGS. For 2013, pre-published USGS consumption statistics
13 were obtained via communications with USGS (USGS 2013)
14 Uncertainty and Time Series Consistency
15 Uncertainty surrounding the total estimated emissions in 2013 is attributed to the uncertainties around SF6, HFC-
16 134a and CCh emission estimates. To estimate the uncertainty surrounding the estimated 2013 SF6 emissions from
17 magnesium production and processing, the uncertainties associated with three variables were estimated: (1)
18 emissions reported by magnesium producers and processors for 2013 through EPA's GHGRP, (2) emissions
19 estimated for magnesium producers and processors that reported via the Partnership in prior years but did not report
20 2013 emissions through EPA's GHGRP, and (3) emissions estimated for magnesium producers and processors that
21 did not participate in the Partnership or report through EPA's GHGRP. An uncertainty of 5 percent was assigned to
22 the emissions (usage) data reported by each GHGRP reporter for all the cover and carrier gases (per the 2006IPCC
23 Guidelines). If facilities did not report emissions data during the current reporting year through EPA's GHGRP
24 reporting program, SF6 emissions data were held constant at the most recent available value reported through the
25 Partnership. The uncertainty associated with these values was estimated to be 30 percent for each year of
26 extrapolation. Alternate cover gas and carrier gases data was set equal to zero if the facilities did not report via the
27 GHGRP program. One known sand caster (the lone Partner) has not reported since 2007 and its activity and
28 emission factor were held constant at 2005 levels due to a reporting anomaly in 2006 because of malfunctions at the
29 facility. The uncertainty associated with the SF6 usage for the sand casting Partner was 85 percent. For those
30 industry processes that are not represented in the Partnership, such as permanent mold and wrought casting, SF6
31 emissions were estimated using production and consumption statistics reported by USGS and estimated process-
32 specific emission factors (see Table 4-84). The uncertainties associated with the emission factors and USGS-
33 reported statistics were assumed to be 75 percent and 25 percent, respectively. Emissions associated with die
34 casting and sand casting activities utilized emission factors based on Partner reported data with an uncertainties of
35 75 percent. In general, where precise quantitative information was not available on the uncertainty of a parameter, a
36 conservative (upper-bound) value was used.
37 Additional uncertainties exist in these estimates that are not addressed in this methodology, such as the basic
38 assumption that SF6 neither reacts nor decomposes during use. The melt surface reactions and high temperatures
39 associated with molten magnesium could potentially cause some gas degradation. Previous measurement studies
40 have identified SF6 cover gas degradation in die casting applications on the order of 20 percent (Bartos et al. 2007).
41 Sulfur hexafluoride may also be used as a cover gas for the casting of molten aluminum with high magnesium
42 content; however, the extent to which this technique is used in the United States is unknown.
43 The results of this Approach 2 quantitative uncertainty analysis are summarized in Table 4-84. Total emissions
44 associated with magnesium production and processing were estimated to be between 1.3 and 1.7 MMT CCh Eq. at
45 the 95 percent confidence level. This indicates a range of approximately 11 percent below to 12 percent above the
46 2013 emission estimate of 1.5 MMT CCh Eq. The uncertainty estimates for 2013 are similar relative to the
47 uncertainty reported for 2012 in the previous Inventory report.
Industrial Processes and Product Use 4-77
-------
1 Table 4-84: Approach 2 Quantitative Uncertainty Estimates for SF6, HFC-134a and COz
2 Emissions from Magnesium Production and Processing (MMT COz Eq. and Percent)
„ „ 2013 Emission Estimate Uncertainty Range Relative to Emission Estimate3
(MMT CCh Eq.) (MMT CCh Eq.) (%)
33
Lower Upper Lower Upper
Bound Bound Bound Bound
Magnesium
Production
SF6,HFC-
134a,CO2
1.3 1.7 -11% +12%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence
interval.
3 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
4 through 2013. Details on the emission trends through time are described in more detail in the Methodology section,
5 above.
6 Recalculations Discussion
7 In the current Inventory, emission estimates for alternate cover gases and carrier gas has been incorporated as the
8 information is now available from EPA's GHGRP. The alternative cover gases have lower GWPs than SF6, and tend
9 to quickly degrade during their exposure to the molten metal. Magnesium producers and processors began using
10 these cover gases starting in around 2006, as based on Partnership reported data. The amounts being used by
11 companies on the whole are low and have a minor effect on the overall emissions from the industry. This is also
12 attributable to their relatively lower GWPs. SF6 has a GWP of 22,800, whereas HFC-134a has a GWP of 1,430.
13 Similarly, EPA's GHGRP now provides access to data on carrier gases, allowing for this information to be
14 integrated in the Inventory. Emissions of CC>2 have also been included in the total emissions from the industry. This
15 has led to slight increase in overall emissions for each year compared to the previous Inventory. CCh carrier gas
16 emissions have been included across the entire time series to ensure time series consistency.
17 For the current Inventory, emission estimates have been revised to reflect the GWPs provided in the IPCC Fourth
18 Assessment Report (AR4) (IPCC 2007). AR4 GWP values differ slightly from those presented in the IPCC Second
19 Assessment Report (SAR) (IPCC 1996) (used in the previous inventories) which results in time-series recalculations
20 for most inventory sources. Under the most recent reporting guidelines (UNFCCC 2014), countries are required to
21 report using the AR4 GWPs, which reflect an updated understanding of the atmospheric properties of each
22 greenhouse gas. The GWPs of CH4 and most fluorinated greenhouse gases have increased, leading to an overall
23 increase in CCh-equivalent emissions from CH4, HFCs, and PFCs. The GWPs of N2O and SF6 have decreased,
24 leading to a decrease in CCh-equivalent emissions for SF6. The AR4 GWPs have been applied across the entire time
25 series for consistency. For more information please see the Recalculations and Improvements Chapter.
26 As a net result, emission estimates for each year from 1990 to 2013 have slightly decreased, relative to the previous
27 Inventory report.
28 For one facility, a recalculation for 2011 SF6 emissions was performed to ensure methodological consistency. The
29 emissions for this facility and year were previously estimated using a company-specific growth rate based on data
30 reported through the Partnership. This estimate has been revised by interpolating the reported emissions between
31 2010 and 2012, reported via the Partnership and EPA's GHGRP respectively. This has caused a slight increase in
32 the SF6 emissions for 2011 compared to the previous Inventory.
Planned Improvements
34 Cover gas research conducted over the last decade has found that SF6 used for magnesium melt protection can have
35 degradation rates on the order of 20 percent in die casting applications (Bartos et al. 2007). Current emission
36 estimates assume (per the 2006IPCC Guidelines) that all SF6 utilized is emitted to the atmosphere. Additional
37 research may lead to a revision of the 2006 IPCC Guidelines to reflect this phenomenon and until such time,
38 developments in this sector will be monitored for possible application to the inventory methodology.
4-78 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Usage and emission details of carrier gases in permanent mold, wrought and anode processes will be researched as
2 part of a future inventory. Based on this research, it will be determined if CC>2 carrier gas emissions are to be
3 estimated.
4 4.20 Lead Production (IPCC Source Category
5 2C5)
6 Lead production in the United States consists of both primary and secondary processes — both of which emit
7 (Sjardin 2003). Emissions from fuels consumed for energy purposes during the production of lead are accounted for
8 in the Energy chapter.
9 Primary production of lead through the direct smelting of lead concentrate produces CCh emissions as the lead
10 concentrates are reduced in a furnace using metallurgical coke (Sjardin 2003). Primary lead production, in the form
11 of direct smelting, previously occurred at a single smelter in Missouri. This primary lead smelter was closed at the
12 end of 2013. In 2014, the smelter processed a small amount of residual lead during demolition of the site (USGS
13 2015).
14 Similar to primary lead production, CCh emissions from secondary lead production result when a reducing agent,
15 usually metallurgical coke, is added to the smelter to aid in the reduction process. Carbon dioxide emissions from
16 secondary production also occur through the treatment of secondary raw materials (Sjardin 2003). Secondary
17 production primarily involves the recycling of lead acid batteries and post-consumer scrap at secondary smelters. Of
18 all the domestic secondary smelters operational in 2014, 12 smelters had capacities of 30,000 tons or more and were
19 collectively responsible for more than 95 percent of secondary lead production in 2014 (USGS 2015). Secondary
20 lead production has increased in the United States over the past decade while primary lead production has decreased.
21 In 20 14, secondary lead production accounted for nearly 100 percent of total lead production. The lead -acid battery
22 industry accounted for about 90 percent of the reported U.S. lead consumption in 2014 (USGS 2015).
23 In 2014, total secondary lead production in the United States was slightly greater than that in 2013. Increased
24 production at a couple of smelters was offset by temporary closure of one smelter. In March 20 14, a producer
25 temporarily shut down operations of a lead smelter in Vernon, CA (90,000 metric ton capacity smelter) due to
26 environmental concerns from state regulators. The company intends to restart operations in 2015, after making
27 improvements to the plant. Increases in exports of spent lead -acid batteries in recent years have decreased the
28 amount of scrap available to secondary smelters (USGS 2015).
29 U.S. primary lead production decreased by approximately 99 percent from 2013 to 2014, and has decreased by
30 almost 100 percent since 1990. This is due to the closure of the only domestic primary lead smelter in 2013 (year
3 1 end). In 2014, U.S. secondary lead production was unchanged from 2013 levels, and has increased by 25 percent
32 since 1990 (USGS 1995 through 2013, USGS 2014, 2015).
33 In 2014, U.S. primary and secondary lead production totaled 1,151,000 metric tons (USGS 2015). The resulting
34 emissions of CO2 from 20 14 lead production were estimated to be 0.5 MMT €62 Eq. (518kt) (see Table 4-85). The
35 majority of 2014 lead production is from secondary processes, which accounted for almost 100 percent of total 2014
36 CO2 emissions from lead production. At last reporting, the United States was the third largest mine producer of lead
37 in the world, behind China and Australia, accounting for approximately 7 percent of world production in 20 14
38 (USGS 2015).
39 Table 4-85: COz Emissions from Lead Production (MMT COz Eq. and kt)
Year MMT CCh Eq. kt
1990
2010
2011
2012
0.5
0.5
0.5
542
538
527
Industrial Processes and Product Use 4-79
-------
25
2013 0.5 546
2014 0.5 518
1 After a steady increase in total emissions from 1995 to 2000, total emissions have gradually decreased since 2000
2 and are currently at the 1990 levels.
Methodology
4 The methods used to estimate emissions for lead production are based on Sjardin's work (Sjardin 2003) for lead
5 production emissions and Tier 1 methods from the 2006IPCC Guidelines (IPCC 2006). The Tier 1 equation is as
6 follows:
7 C02 Emissions = (DS x EFDS) + (S x ŁFS)
8 Where,
9 DS Lead produced by direct smelting, metric ton
10 S = Lead produced from secondary materials
11 EFDs = Emission factor for direct Smelting, metric tons CCh/metric ton lead product
12 EFS = Emission factor for secondary materials, metric tons CCh/metric ton lead product
13 For primary lead production using direct smelting, Sjardin (2003) and the IPCC (2006) provide an emission factor of
14 0.25 metric tons CO^metric ton lead. For secondary lead production, Sjardin (2003) and IPCC (2006) provide an
15 emission factor of 0.25 metric tons CCh/metric ton lead for direct smelting, as well as an emission factor of 0.2
16 metric tons CCh/metric ton lead produced for the treatment of secondary raw materials (i.e., pretreatment of lead
17 acid batteries). Since the secondary production of lead involves both the use of the direct smelting process and the
18 treatment of secondary raw materials, Sjardin recommends an additive emission factor to be used in conjunction
19 with the secondary lead production quantity. The direct smelting factor (0.25) and the sum of the direct smelting and
20 pretreatment emission factors (0.45) are multiplied by total U.S. primary and secondary lead production,
21 respectively, to estimate CCh emissions.
22 The 1990 through 2014 activity data for primary and secondary lead production (see Table 4-86) were obtained from
23 the USGS (USGS 1995 through 2013, 2014, 2015).
24 Table 4-86: Lead Production (Metric Tons)
Year Primary Secondary
1990 404,000 922,000
2010
2011
2012
2013
2014
115,000
118,000
111,000
114,000
1,000
1,140,000
1,130,000
1,110,000
1,150,000
1,150,000
Uncertainty and Time-Series Consistency
26 Uncertainty associated with lead production relates to the emission factors and activity data used. The direct
27 smelting emission factor used in primary production is taken from Sjardin (2003) who averaged the values provided
28 by three other studies (Dutrizac et al. 2000, Morris et al. 1983, Ullman 1997). For secondary production, Sjardin
29 (2003) added a CC>2 emission factor associated with battery treatment. The applicability of these emission factors to
30 plants in the United States is uncertain. There is also a smaller level of uncertainty associated with the accuracy of
31 primary and secondary production data provided by the USGS.
32 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-87. Lead production
33 emissions were estimated to be between 0.4 and 0.6 MMT €62 Eq. at the 95 percent confidence level. This
4-80 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 indicates a range of approximately 15 percent below and 16 percent above the emission estimate of 0.5 MMT CO2
2 Eq.
3 Table 4-87: Approach 2 Quantitative Uncertainty Estimates for COz Emissions from Lead
4 Production (MMT COz Eq. and Percent)
„ ~ 2014 Emission Estimate Uncertainty Ranee Relative to Emission Estimate3
xoii t*f*p I-w-nc
(MMT CCh Eq.) (MMT CCh Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
Lead Production CCh 0.5 0.4 0.6 -15% +16%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
5 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
6 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
7 above.
8 Recalculations Discussion
9 For the current Inventory, primary and secondary lead production quantities were revised to reflect the most recent
10 USGS publication (USGS 2015). In the previous Inventory report, the 2013 primary and secondary lead production
11 quantities were based on preliminary USGS estimates that were available at the time. This change resulted in a 4
12 percent increase in the 2013 emission estimate compared to the previous Inventory report.
13 Planned Improvements
14 Future improvements will involve evaluating and analyzing data reported under EPA's GHGRP to improve the
15 emission estimates for the Lead Production source category. Particular attention will be made to ensure time series
16 consistency of the emission estimates presented in future Inventory reports, consistent with IPCC and UNFCCC
17 guidelines. This is required as the facility-level reporting data from EPA's GHGRP, with the program's initial
18 requirements for reporting of emissions in calendar year 2010, are not available for all inventory years (i.e., 1990
19 through 2009) as required for this Inventory. In implementing improvements and integration of data from EPA's
20 GHGRP, the latest guidance from the IPCC on the use of facility-level data in national inventories will be relied
21 upon.32
22
23
4.21 Zinc Production (IPCC Source Category
2C6)
24 Zinc production in the United States consists of both primary and secondary processes. Of the primary and
25 secondary processes used in the United States, only the electrothermic and Waelz kiln secondary processes result in
26 non-energy CC>2 emissions (Viklund-White 2000). Emissions from fuels consumed for energy purposes during the
27 production of zinc are accounted for in the Energy chapter.
28 The majority of zinc produced in the United States is used for galvanizing. Galvanizing is a process where zinc
29 coating is applied to steel in order to prevent corrosion. Zinc is used extensively for galvanizing operations in the
30 automotive and construction industry. Zinc is also used in the production of zinc alloys and brass and bronze alloys
31 (e.g., brass mills, copper foundries, copper ingot manufacturing, etc.). Zinc compounds and dust are also used, to a
32 lesser extent, by the agriculture, chemicals, paint, and rubber industries.
32 See.
Industrial Processes and Product Use 4-81
-------
1 Primary production in the United States is conducted through the electrolytic process, while secondary techniques
2 include the electrothermic and Waelz kiln processes, as well as a range of other metallurgical, hydrometallurgical,
3 and pyrometallurgical processes. Worldwide primary zinc production also employs a pyrometallurgical process
4 using the Imperial Smelting Furnace process; however, this process is not used in the United States (Sjardin 2003).
5 In the electrothermic process, roasted zinc concentrate and secondary zinc products enter a sinter feed where they
6 are burned to remove impurities before entering an electric retort furnace. Metallurgical coke is added to the electric
7 retort furnace as a carbon-containing reductant. This concentration step, using metallurgical coke and high
8 temperatures, reduces the zinc oxides and produces vaporized zinc, which is then captured in a vacuum condenser.
9 This reduction process also generates non-energy COa emissions.
10 ZnO + C -> Zn(gas) + C02 (Reaction 1)
11 ZnO +CO -^Zn(gas) + C02 (Reaction 2)
12 In the Waelz kiln process, electric arc furnace (EAF) dust, which is captured during the recycling of galvanized
13 steel, enters a kiln along with a reducing agent (typically carbon-containing metallurgical coke). When kiln
14 temperatures reach approximately 1 100-1200°C, zinc fumes are produced, which are combusted with air entering
15 the kiln. This combustion forms zinc oxide, which is collected in a baghouse or electrostatic precipitator, and is then
16 leached to remove chloride and fluoride. The use of carbon-containing metallurgical coke in a high-temperature
17 fuming process results in non-energy CC>2 emissions. Through this process, approximately 0.33 metric tons of zinc is
18 produced for every metric ton of EAF dust treated (Viklund- White 2000).
19 The only companies in the United States that use emissive technology to produce secondary zinc products are
20 Horsehead, PIZO, and Steel Dust Recycling. For Horsehead, EAF dust is recycled in Waelz kilns at their
21 Beaumont, TX; Calumet, IL; Palmerton, PA; Rockwood, TN; and Barnwell, SC facilities. These Waelz kiln
22 facilities produce intermediate zinc products (crude zinc oxide or calcine), most of which was transported to their
23 Monaca, PA facility where the products were smelted into refined zinc using electrothermic technology. In April
24 20 14, Horsehead permanently shut down their Monaca smelter. This was replaced by their new facility in
25 Mooresboro, NC. The new Mooresboro facility uses a hydrometallurgical process (i.e., solvent extraction with
26 electrowinning technology) to produce zinc products. The current capacity of the new facility is 155,000 short tons,
27 with plans to expand to 170,000 short tons per year. Direct consumption of coal, coke, and natural gas have been
28 replaced with electricity consumption at the new Mooresboro facility. The new facility is reported to have a
29 significantly lower greenhouse gas and other air emissions than the Monaca smelter (Horsehead 2012b).
30 The Mooresboro facility uses leaching and solvent extraction (SX) technology combined with electrowinning,
3 1 melting, and casting technology. In this process, Waelz Oxide (WOX) is first washed in water to remove soluble
32 elements such as chlorine, potassium, and sodium, and then is leached in a sulfuric acid solution to dissolve the
33 contained zinc creating a pregnant liquor solution (PLS). The PLS is then processed in a solvent extraction step in
34 which zinc is selectively extracted from the PLS using an organic solvent creating a purified zinc -loaded electrolyte
35 solution. The loaded electrolyte solution is then fed into the electrowinning process in which electrical energy is
36 applied across a series of anodes and cathodes submerged in the electrolyte solution causing the zinc to deposit on
37 the surfaces of the cathodes. As the zinc metal builds up on these surfaces, the cathodes are periodically harvested in
38 order to strip the zinc from their surfaces (Horsehead 2015). Hydrometallurgical production processes are assumed
39 to be non-emissive since no carbon is used in these processes (Sjardin 2003).
40 PIZO and Steel Dust Recycling recycle EAF dust into intermediate zinc products using Waelz kilns, and then sell
41 the intermediate products to companies who smelt it into refined products.
42 In 2014, U.S. primary and secondary refined zinc production were estimated to total 185,000 metric tons (USGS
43 2015) (see Table 4-88). Domestic zinc mine production increased slightly in 2014 compared to 2013 levels,
44 primarily owing to an increase in zinc production at the Red Dog mine in Alaska. Zinc metal production decreased
45 by 20 percent owing to a decline in secondary production; Horsehead closed its smelter in Monaca, PA, while
46 starting up its new recycling facility in Mooresboro, NC. However, the new facility experienced delayed ramp -up
47 efforts due to technical issues and did not reach optimum production levels until the end of 2014 (USGS 2015).
48 Primary zinc production (primary slab zinc) increased slightly in 2014, while, secondary zinc production in 2014
49 decreased relative to 20 1 3 .
50 Emissions of COa from zinc production in 2014 were estimated to be 1.0 MMT CO2 Eq. (956 kt CO2) (see Table
51 4-89). All 2014 CO2 emissions resulted from secondary zinc production processes. Emissions from zinc production
4-82 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
in the U.S. have increased overall since 1990 due to a gradual shift from non-emissive primary production to
emissive secondary production. In 2014, emissions were estimated to be 51 percent higher than they were in 1990.
Table 4-88: Zinc Production (Metric Tons)
Year Primary
Secondary
1990 262,704
95,708
2010
2011
2012
2013
2014
120,000
110,000
114,000
106,000
115,000
129,000
138,000
147,000
127,000
70,000
4 Table 4-89: COz Emissions from Zinc Production (MMT COz Eq. and kt)
Year MMT CCh Eq.
kt
1990
2010
2011
2012
2013
2014
1.2
1.3
1.5
1.4
1.0
1,182
1,286
1,486
1,429
956
Methodology
6 The methods used to estimate non-energy CCh emissions from zinc production using the electrothermic primary
7 production and Waelz kiln secondary production processes are based on Tier 1 methods from the 2006IPCC
8 Guidelines (IPCC 2006). The Tier 1 equation used to estimate emissions from zinc production is as follows:
-C02
= Zn x EF,
default
10 where,
11
12
13
ECo2
Zn
= CO2 emissions from zinc production, metric tons
= Quantity of zinc produced, metric tons
= Default emission factor, metric tons CCh/metric ton zinc produced
14
15 The Tier 1 emission factors provided by IPCC for Waelz kiln-based secondary production were derived from coke
16 consumption factors and other data presented in Vikland-White (2000). These coke consumption factors as well as
17 other inputs used to develop the Waelz kiln emission factors are shown below. IPCC does not provide an emission
18 factor for electrothermic processes due to limited information; therefore, the Waelz kiln-specific emission factors
19 were also applied to zinc produced from electrothermic processes. Starting in 2014, refined zinc produced in the
20 U.S. used hydrometallurgical processes and is assumed to be non-emissive.
21 For Waelz kiln-based production, IPCC recommends the use of emission factors based on EAF dust consumption, if
22 possible, rather than the amount of zinc produced since the amount of reduction materials used is more directly
23 dependent on the amount of EAF dust consumed. Since only a portion of emissive zinc production facilities
24 consume EAF dust, the emission factor based on zinc production is applied to the non-EAF dust consuming
25 facilities while the emission factor based on EAF dust consumption is applied to EAF dust consuming facilities.
Industrial Processes and Product Use 4-83
-------
1 The Waelz kiln emission factor based on the amount of zinc produced was developed based on the amount of
2 metallurgical coke consumed for non-energy purposes per ton of zinc produced (i.e., 1.19 metric tons coke/metric
3 ton zinc produced) (Viklund- White 2000), and the following equation:
1.19 metric tons coke 0.85 metric tons C 3.67 metric tons C02 3.70 metric tons C02
4 Lji ' Waelz Kiln ~ - ^ - ^ - ~ -
metric tons zinc metric tons coke metric tons C metric tons zinc
5 The Waelz kiln emission factor based on the amount of EAF dust consumed was developed based on the amount of
6 metallurgical coke consumed per ton of EAF dust consumed (i.e., 0.4 metric tons coke/metric ton EAF dust
7 consumed) (Viklund-White 2000), and the following equation:
0.4 metric tons coke O.»b metric tons L 3. b I metric tons LU2 1.Z4 metric ton? CO,
o pp _ _ ^, _ ^, _ Ł_ _ __ Ł_
EAF Dust metric tons ^AF Dust metric tons coke metric tons C metric tons EAl Dust
9 The total amount of EAF dust consumed by Horsehead at their Waelz kilns was available from Horsehead financial
10 reports foryears 2006 through 2014 (Horsehead 2007, 2008, 2010a, 2011, 2012a, 2013, 2014, and 2015).
1 1 Consumption levels for 1990 through 2005 were extrapolated using the percentage change in annual refined zinc
12 production at secondary smelters in the United States as provided by USGS Minerals Yearbook: Zinc (USGS 1995
13 through 2006). The EAF dust consumption values for each year were then multiplied by the 1.24 metric tons
14 CCVmetric ton EAF dust consumed emission factor to develop CC>2 emission estimates for Horsehead's Waelz kiln
15 facilities.
16 The amount of EAF dust consumed by Steel Dust Recycling (SDR) and their total production capacity were
17 obtained from SDR's facility in Alabama for the years 201 1 through 2014 (SDR 2012, 2014, and 2015). SDR's
18 facility in Alabama underwent expansion in 201 1 to include a second unit (operational since early- to mid-2012).
19 SDR's facility has been operational since 2008. Annual consumption data for SDR was not publicly available for the
20 years 2008, 2009, and 2010. These data were estimated using data for Horsehead's Waelz kilns for 2008-2010
21 (Horsehead 2007, 2008, 2010a, 2010b, and 201 1). Annual capacity utilization ratios were calculated using
22 Horsehead's annual consumption and total capacity for the years 2008 through 2010. Horsehead's annual capacity
23 utilization ratios were multiplied with SDR's total capacity to estimate SDR's consumption for each of the years,
24 2008 through 2010 (SDR 2013).
25 PIZO Technologies Worldwide LLC's facility in Arkansas has been operational since 2009. The amount of EAF
26 dust consumed by PIZO's facility for 2009 through 2014 was not publicly available. EAF dust consumption for
27 PIZO's facility for 2009 and 2010 were estimated by calculating annual capacity utilization of Horsehead's Waelz
28 kilns and multiplying this utilization ratio by PIZO's total capacity (PIZO 2012). EAF dust consumption for PIZO's
29 facility for 20 1 1 through 20 14 were estimated by applying the average annual capacity utilization rates for
30 Horsehead and SDR (Grupo PROMAX) to PIZO's annual capacity (Horsehead 2012, 2013, 2014, and 2015; SDR
31 2012 and 2014; PIZO 2012 and 2014). The 1.24 metric tons COz/metric ton EAF dust consumed emission factor
32 was then applied to PIZO's and Steel Dust Recycling's estimated EAF dust consumption to develop CO2 emission
33 estimates for those Waelz kiln facilities.
34 Refined zinc production levels for Horsehead's Monaca, PA facility (utilizing electrothermic technology) were
35 available from the company foryears 2005 through2013 (Horsehead 2008, 2011, 2012, 2013, and 2014). The
36 Monaca facility was permanently shut down in April 20 14 and was replaced by Horsehead's new facility in
37 Mooresboro, NC. The new facility uses hydrometallurgical process to produce refined zinc products. This process is
38 assumed to be non-emissive. Production levels for 1990 through 2004 were extrapolated using the percentage
39 changes in annual refined zinc production at secondary smelters in the United States as provided by USGS Minerals
40 Yearbook: Zinc (USGS 1995 through 2005). The 3.70 metric tons CO2/metric ton zinc emission factor was then
41 applied to the Monaca facility's production levels to estimate CO2 emissions for the facility. The Waelz kiln
42 production emission factor was applied in this case rather than the EAF dust consumption emission factor since
43 Horsehead's Monaca facility did not consume EAF dust.
4-84 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
i Uncertainty and Time-Series Consistency
2 The uncertainty associated with these estimates is two-fold, relating to activity data and emission factors used.
3 First, there is uncertainty associated with the amount of EAF dust consumed in the United States to produce
4 secondary zinc using emission-intensive Waelz kilns. The estimate for the total amount of EAF dust consumed in
5 Waelz kilns is based on (1) an EAF dust consumption value reported annually by Horsehead Corporation as part of
6 its financial reporting to the Securities and Exchange Commission (SEC), and (2) an EAF dust consumption value
7 obtained from the Waelz kiln facility operated in Alabama by Steel Dust Recycling LLC. Since actual EAF dust
8 consumption information is not available for PIZO's facility (2009 through 2010) and SDK's facility (2008 through
9 2010), the amount is estimated by multiplying the EAF dust recycling capacity of the facility (available from the
10 company's website) by the capacity utilization factor for Horsehead Corporation (which is available from
11 Horsehead's financial reports). Also, the EAF dust consumption for PIZO's facility for 2011 through 2013 was
12 estimated by multiplying the average capacity utilization factor developed from Horsehead Corp. and SDK's annual
13 capacity utilization rates by PIZO's EAF dust recycling capacity. Therefore, there is uncertainty associated with the
14 assumption used to estimate PIZO and SDK's annual EAF dust consumption values (except SDK's EAF dust
15 consumption for 2011 through 2013, which were obtained from SDK's recycling facility in Alabama).
16 Second, there is uncertainty associated with the emission factors used to estimate CO2 emissions from secondary
17 zinc production processes. The Waelz kiln emission factors are based on materials balances for metallurgical coke
18 and EAF dust consumed as provided by Viklund-White (2000). Therefore, the accuracy of these emission factors
19 depend upon the accuracy of these materials balances. Data limitations prevented the development of emission
20 factors for the electrothermic process. Therefore, emission factors for the Waelz kiln process were applied to both
21 electrothermic and Waelz kiln production processes. The results of the Approach 2 quantitative uncertainty analysis
22 are summarized in Table 4-90. Zinc production CO2 emissions were estimated to be between 0.8 and 1.2 MMT CO2
23 Eq. at the 95 percent confidence level. This indicates a range of approximately 19 percent below and 21 percent
24 above the emission estimate of 1.0 MMT CO2 Eq.
25 Table 4-90: Approach 2 Quantitative Uncertainty Estimates for COz Emissions from Zinc
26 Production (MMT COz Eq. and Percent)
Source Gas 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
(MMT CCh Eq.) (MMT CCh Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
"Zinc Production CCh 1.0 0.8 1.2 -19% +21% ~
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
27 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
28 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
29 above.
30
Planned Improvements
31 EPA continues to evaluate use of facility-level data reported under EPA's GHGRP to improve the emission
32 estimates for the Zinc Production source category, provided aggregated data meet criteria for safe disclosure and
33 will not reveal confidential business information. In particular, EPA is assessing feasibility to derive an average
34 emission factor based on reported production and emissions. Particular attention will be made to ensure time series
35 consistency of the emissions estimates presented in future Inventory reports, consistent with IPCC and UNFCCC
36 guidelines. This is required as the facility-level reporting data from EPA's GHGRP, with the program's initial
37 requirements for reporting of emissions in calendar year 2010, are not available for all inventory years (i.e., 1990
38 through 2009) as required for this Inventory. In implementing improvements and integration of data from EPA's
Industrial Processes and Product Use 4-85
-------
1 GHGRP, the latest guidance from the IPCC on the use of facility-level data in national inventories will be relied
2 upon.33
3 4.22 Semiconductor Manufacture (IPCC
4 Source Category 2E1) (TO BE UPDATED)
5 The semiconductor industry uses multiple long-lived fluorinated greenhouse gases in plasma etching and plasma
6 enhanced chemical vapor deposition (PECVD) processes to produce semiconductor products. The gases most
7 commonly employed are trifluoromethane (HFC-23 or CHF3), perfluoromethane (CF4), perfluoroethane (CJe),
8 nitrogen trifluoride (NF3), sulfur hexafluoride (SF6), and nitrous oxide (N2O), although other compounds such as
9 perfluoropropane (CsFg) and perfluorocyclobutane (c-C4P8) are also used. The exact combination of compounds is
10 specific to the process employed.
11 A single 300 mm silicon wafer that yields between 400 to 600 semiconductor products (devices or chips) may
12 require more than 100 distinct fluorinated-gas-using process steps, principally to deposit and pattern dielectric films.
13 Plasma etching (or patterning) of dielectric films, such as silicon dioxide and silicon nitride, is performed to provide
14 pathways for conducting material to connect individual circuit components in each device. The patterning process
15 uses plasma-generated fluorine atoms, which chemically react with exposed dielectric film to selectively remove the
16 desired portions of the film. The material removed as well as undissociated fluorinated gases flow into waste
17 streams and, unless emission abatement systems are employed, into the atmosphere. PECVD chambers, used for
18 depositing dielectric films, are cleaned periodically using fluorinated and other gases. During the cleaning cycle the
19 gas is converted to fluorine atoms in plasma, which etches away residual material from chamber walls, electrodes,
20 and chamber hardware. Undissociated fluorinated gases and other products pass from the chamber to waste streams
21 and, unless abatement systems are employed, into the atmosphere.
22 In addition to emissions of unreacted gases, some fluorinated compounds can also be transformed in the plasma
23 processes into different fluorinated compounds which are then exhausted, unless abated, into the atmosphere. For
24 example, when C2p6 is used in cleaning or etching, CF4 is generated and emitted as a process byproduct. Besides
25 dielectric film etching and PECVD chamber cleaning, much smaller quantities of fluorinated gases are used to etch
26 polysilicon films and refractory metal films like tungsten.
27 Nitrous oxide is used in manufacturing semiconductor devices to produce thin films by CVD and nitridation
28 processes as well as for N-doping of compound semiconductors and reaction chamber conditioning (Doering 2000).
29 For 2013, total CCh weighted emissions of all fluorinated greenhouse gases and nitrous oxide by the U.S.
30 semiconductor industry were estimated to be 4.2 MMT CCh Eq. Combined emissions of all greenhouse gases are
31 presented in Table 4-91 and Table 4-92 below for years 1990, 2005 and the period 2009 to 2013. The rapid growth
32 of this industry and the increasing complexity (growing number of layers34) of semiconductor products led to an
33 increase in emissions of 153 percent between 1990 and 1999, when emissions peaked at 9.1 MMT CO2 Eq. The
34 emissions growth rate began to slow after 1999, and emissions declined by 54 percent between 1999 and 2013.
35 Together, industrial growth, adoption of emissions reduction technologies, including but not limited to abatement
36 technologies, and shift in gas usages resulted in a net increase in emissions of 16 percent between 1990 and 2013.
37 There was a sizable dip seen in emissions between 2008 and 2009, a 28 percent decrease, due to the slowed
38 economic growth, and hence production, during this time. The industry recovered and emissions rose between 2009
39 and 2010 by more than 25 percent and between 2010 and 2011 by 29 percent; reductions in emissions were observed
40 between 2011 and 2012, and 2012 and 2013 at 9 percent and 7 percent, respectively.
See.
Complexity is a term denoting the circuit required to connect the active circuit elements (transistors) on a chip. Increasing
miniaturization, for the same chip size, leads to increasing transistor density, which, in turn, requires more complex
interconnections between those transistors. This increasing complexity is manifested by increasing the levels (i.e., layers) of
wiring, with each wiring layer requiring fluorinated gas usage for its manufacture.
4-86 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 4-91: RFC, HFC, SFe, NFs, and NzO Emissions from Semiconductor Manufacture (MMT
2 COz Eq.)
Year
CF4
C2F6
CsFs
C4F8
HFC-23
SFe
NFs
Total F-
GHGs
1990 2005
0.8
2.0
+
+
0.2
0.5
+
1.1
1.9
0.1
0.1
0.2
H""
-.-
3.6 4.6
2009
0.8
1.1
0.1
+
0.2
0.3
0.4
2.9
2010
1.1
1.4
0.1
+
0.2
0.4
0.5
3.7
2011
1.4
1.8
0.2
0.1
0.2
0.4
0.7
4.7
2012
1.3
1.6
0.1
0.1
0.2
0.4
0.6
4.3
2013
1.2
1.5
0.1
0.1
0.2
0.4
0.6
4.0
N20 + 0.1 0.1
Total 3.6 4.7 3.1
Note: Totals may not sum due to independent rounding.
+ Does not exceed 0.05 MMT CO2 Eq.
0.1
3.8
Table 4-92: PFC, HFC, SF6, NFs, and NzO Emissions from
Year 1990 2005 2009
CF4 0.11 0.14 0.11
C2F6 0.16 0.16 0.09
CsF8 + + +
C4F8 + + +
HFC-23 + + +
SFe + + +
NFs + + +
N20 0.12 0.41 0.45
2010
0.14
0.11
+
+
+
+
+
0.49
0.2
4.9
0.2
4.5
Semiconductor
2011
0.19
0.14
+
+
+
+
+
0.79
2012
0.17
0.13
+
+
+
+
+
0.65
0.2
4.2
Manufacture (kt)
2013
0.16
0.12
+
+
+
+
+
0.61
+ Does not exceed 0.05 kt
5
6
7
8
9
10
1 1
Methodology
Emissions are based on data reported through Subpart I, Electronics Manufacture, of EPA's GHGRP, Partner
reported emissions data received through the EPA's PFC35 Reduction/Climate Partnership, EPA's PFC Emissions
Vintage Model (PEVM) — a model that estimates industry emissions in the absence of emission control strategies
(Burton and Beizaie 2001)36, and estimates of industry activity (i.e., total manufactured layer area). The availability
and applicability of reported data from the EPA Partnership and EPA's GHGPJ3 differs across the 1990 through
2013 time series. Consequently, F-GHG emissions from semiconductor manufacturing were estimated using five
distinct methods, one each for the periods 1990 through 1994, 1995 through 1999, 2000 through 2006, 2007 through
35 In the context of the EPA Partnership and PEVM, PFC refers to perfluorocompounds, not perfluorocarbons.
A Partner refers to a participant in the U.S. EPA PFC Reduction/Climate Partnership for the Semiconductor Industry.
Through a Memorandum of Understanding (MoU) with the EPA, Partners voluntarily reported their PFC emissions to the EPA
by way of a third party, which aggregated the emissions through 2010. For 2011, while no MOU existed, it was assumed that the
same companies that were Partners in 2010 were "Partners" in 2011 for purposes of estimating inventory emissions.
Industrial Processes and Product Use 4-87
-------
1 2010, and 2011 through 2013. N2O emissions were estimated using three distinct methods, one each for the period
2 1990 through 1994, 1995 through 2010, and 2011 through 2013.
3 1990 through 1994
4 From 1990 through 1994, Partnership data were unavailable and emissions were modeled using the PEVM (Burton
5 and Beizaie 2001).37 The 1990 to 1994 emissions are assumed to be uncontrolled, since reduction strategies such as
6 chemical substitution and abatement were yet to be developed.
7 PEVM is based on the recognition that fluorinated greenhouse gas emissions from semiconductor manufacturing
8 vary with: (1) the number of layers that comprise different kinds of semiconductor devices, including both silicon
9 wafer and metal interconnect layers, and (2) silicon consumption (i.e., the area of semiconductors produced) for
10 each kind of device. The product of these two quantities, Total Manufactured Layer Area (TMLA), constitutes the
11 activity data for semiconductor manufacturing. PEVM also incorporates an emission factor that expresses emissions
12 per unit of layer-area. Emissions are estimated by multiplying TMLA by this emission factor.
13 PEVM incorporates information on the two attributes of semiconductor devices that affect the number of layers: (1)
14 linewidth technology (the smallest manufactured feature size),38 and (2) product type (discrete, memory or logic).39
15 For each linewidth technology, a weighted average number of layers is estimated using VLSI product-specific
16 worldwide silicon demand data in conjunction with complexity factors (i.e., the number of layers per Integrated
17 Circuit (1C)) specific to product type (Burton and Beizaie 2001, ITRS 2007). PEVM derives historical consumption
18 of silicon (i.e., square inches) by linewidth technology from published data on annual wafer starts and average wafer
19 size (VLSI Research, Inc. 2012).
20 The emission factor in PEVM is the average of four historical emission factors, each derived by dividing the total
21 annual emissions reported by the Partners for each of the four years between 1996 and 1999 by the total TMLA
22 estimated for the Partners in each of those years. Over this period, the emission factors varied relatively little (i.e.,
23 the relative standard deviation for the average was 5 percent). Since Partners are believed not to have applied
24 significant emission reduction measures before 2000, the resulting average emission factor reflects uncontrolled
25 emissions. The emission factor is used to estimate world uncontrolled emissions using publicly-available data on
26 world silicon consumption.
27 As it was assumed for this time period that there was no consequential adoption of fluorinated-gas-reducing
28 measures, a fixed distribution of fluorinated-gas use was assumed to apply to the entire U.S. industry to estimate
29 gas-specific emissions. This distribution was based upon the average fluorinated-gas purchases made by
30 semiconductor manufacturers during this period and the application of IPCC default emission factors for each gas
31 (Burton and Beizaie 2001).
32 To estimate N2O emissions, it is assumed the proportion of N2O emissions estimated for 1995 (discussed below)
33 remained constant for the period of 1990 through!994.
37 Various versions of the PEVM exist to reflect changing industrial practices. From 1990 to 1994 emissions estimates are from
PEVM vl.O, completed in September 1998. The emission factor used to estimate 1990 to 1994 emissions is an average of the
1995 and 1996 emissions factors, which were derived from Partner reported data for those years.
38 By decreasing features of Integrated Circuit components, more components can be manufactured per device, which increases
its functionality. However, as those individual components shrink it requires more layers to interconnect them to achieve the
functionality. For example, a microprocessor manufactured with 65 nm feature sizes might contain as many as 1 billion
transistors and require as many as 11 layers of component interconnects to achieve functionality, while a device manufactured
with 130 nm feature size might contain a few hundred million transistors and require 8 layers of component interconnects (ITRS
2007).
Memory devices manufactured with the same feature sizes as microprocessors (a logic device) require approximately one-half
the number of interconnect layers, whereas discrete devices require only a silicon base layer and no interconnect layers (ITRS
2007). Since discrete devices did not start using PFCs appreciably until 2004, they are only accounted for in the PEVM
emissions estimates from 2004 onwards.
4-88 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
i 1995 through 1999
2 For 1995 through 1999, total U.S. emissions were extrapolated from the total annual emissions reported by the
3 Partners (1995 through 1999). Partner-reported emissions are considered more representative (e.g., in terms of
4 capacity utilization in a given year) than PEVM estimated emissions, and are used to generate total U.S. emissions
5 when applicable. The emissions reported by the Partners were divided by the ratio of the total capacity of the plants
6 operated by the Partners and the total capacity of all of the semiconductor plants in the United States; this ratio
7 represents the share of capacity attributable to the Partnership. This method assumes that Partners and non-Partners
8 have identical capacity utilizations and distributions of manufacturing technologies. Plant capacity data is contained
9 in the World Fab Forecast (WFF) database and its predecessors, which is updated quarterly (Semiconductor
10 Equipment and Materials Industry 2012 and 2013). Gas-specific emissions were estimated using the same method as
11 for 1990 through 1994.
12 For this time period, the N2O emissions were estimated using an emission factor that is applied to the annual, total
13 U.S. TMLA manufactured. The emission factor was developed using a regression-through-the-origin (RTO) model:
14 GHGRP reported N2O emissions were regressed against the corresponding TMLA of facilities that reported no use
15 of abatement systems. Details on the GHGRP reported emissions and development of emission factor using the RTO
16 model are presented in the 2011 through2013 section. ThetotalU.S. TMLA manufactured were estimated using
17 PEVM.
is 2000 through 2006
19 Emissions for the years 2000 through 2006—the period during which Partners began the consequential application
20 of fluorinated greenhouse gas-reduction measures—were estimated using a combination of Partner-reported
21 emissions and adjusted PEVM modeled emissions. The emissions reported by Partners for each year were accepted
22 as the quantity emitted from the share of the industry represented by those Partners. Remaining emissions, those
23 from non-Partners, were estimated using PEVM, with one change. To ensure time series consistency and to reflect
24 the increasing use of remote clean technology (which increases the efficiency of the production process while
25 lowering emissions of fluorinated greenhouse gases), the average non-Partner emission factor was assumed to begin
26 declining gradually during this period. Specifically, the non-Partner emission factor for each year was determined
27 by linear interpolation, using the end points of 1999 (the original PEVM emission factor) and 2011 (a new emission
28 factor determined for the non-Partner population based on GHGRP-reported data, described below).
29 The portion of the U.S. total attributed to non-Partners is obtained by multiplying PEVM's total U.S. emissions
30 figure by the non-Partner share of U.S. total silicon capacity for each year as described above.40 Gas-specific
31 emissions from non-Partners were estimated using linear interpolation of gas-specific emission distribution of 1999
32 (assumed same as total U.S. Industry in 1994) and 2011 (calculated from a subset of non-Partner facilities from
33 GHGRP reported emissions data). Annual updates to PEVM reflect published figures for actual silicon consumption
34 from VLSI Research, Inc., revisions and additions to the world population of semiconductor manufacturing plants,
35 and changes in 1C fabrication practices within the semiconductor industry (see ITRS 2008 and Semiconductor
This approach assumes that the distribution of linewidth technologies is the same between Partners and non-Partners. As
discussed in the description of the method used to estimate 2007 emissions, this is not always the case.
Industrial Processes and Product Use 4-89
-------
1 Equipment and Materials Industry 2011).4 * '4Z43
2 The N2O emissions were estimated using the same methodology as 1995-1999 methodology.
3 2007 through 2010
4 For the years 2007 through 2010, emissions were also estimated using a combination of Partner reported emissions
5 and adjusted PEVM modeled emissions to provide estimates for non-Partners; however, two improvements were
6 made to the estimation method employed for the previous years in the time series. First, the 2007 through 2010
7 emission estimates account for the fact that Partners and non-Partners employ different distributions of
8 manufacturing technologies, with the Partners using manufacturing technologies with greater transistor densities and
9 therefore greater numbers of layers.44 Second, the scope of the 2007 through 2010 estimates was expanded relative
10 to the estimates for the years 2000 through 2006 to include emissions from research and development (R&D) fabs.
11 This additional enhancement was feasible through the use of more detailed data published in the WFF. PEVM
12 databases were updated annually as described above. The published world average capacity utilization for 2007
13 through 2010 was used for production fabs, while for R&D fabs a 20 percent figure was assumed (SIA 2009).
14 In addition, publicly-available actual utilization data was used to account for differences in fab utilization for
15 manufacturers of discrete and 1C products for 2010 emissions for non-Partners. PEVM estimates were adjusted
16 using technology-weighted capacity shares that reflect the relative influence of different utilization. Gas-specific
17 emissions for non-Partners were estimated using the same method as for 2000 through 2006.
18 The N2O emissions were estimated using the same methodology as 1995 through 1999 methodology.
19 2011 through 2013
20 The fifth and final method for estimating emissions from semiconductor manufacturing covers the period 2011
21 through 2013, the years after EPA's Partnership with the semiconductor industry ended (in 2010) and reporting
22 under the GHGRP began. Manufacturers whose estimated uncontrolled emissions equal or exceed 25,000 mt CCh
23 Eq. per year (based on default emission factors and total capacity in terms of substrate area) are required to report
24 their emissions to the EPA. This population of reporters to EPA's GHGRP included both historical Partners of
41 Special attention was given to the manufacturing capacity of plants that use wafers with 300 mm diameters because the actual
capacity of these plants is ramped up to design capacity, typically over a 2-3 year period. To prevent overstating estimates of
partner-capacity shares from plants using 300 mm wafers, design capacities contained in WFW were replaced with estimates of
actual installed capacities for 2004 published by Citigroup Smith Barney (2005). Without this correction, the partner share of
capacity would be overstated, by approximately 5 percent. For perspective, approximately 95 percent of all new capacity
additions in 2004 used 300 mm wafers, and by year-end those plants, on average, could operate at approximately 70 percent of
the design capacity. For 2005, actual installed capacities were estimated using an entry in the World Fab Watch database (April
2006 Edition) called "wafers/month, 8-inch equivalent," which denoted the actual installed capacity instead of the fully-ramped
capacity. For 2006, actual installed capacities of new fabs were estimated using an average monthly ramp rate of 1100 wafer
starts per month (wspm) derived from various sources such as semiconductor fabtech, industry analysts, and articles in the trade
press. The monthly ramp rate was applied from the first-quarter of silicon volume (FQS V) to determine the average design
capacity over the 2006 period.
4^ In 2006, the industry trend in co-ownership of manufacturing facilities continued. Several manufacturers, who are Partners,
now operate fabs with other manufacturers, who in some cases are also Partners and in other cases are not Partners. Special
attention was given to this occurrence when estimating the Partner and non-Partner shares of U.S. manufacturing capacity.
43 Two versions of PEVM are used to model non-Partner emissions during this period. For the years 2000 to 2003 PEVM
v3.2.0506.0507 was used to estimate non-Partner emissions. During this time, discrete devices did not use PFCs during
manufacturing and therefore only memory and logic devices were modeled in the PEVM v3.2.0506.0507. From 2004 onwards,
discrete device fabrication started to use PFCs, hence PEVM v4.0.0701.0701, the first version of PEVM to account for PFC
emissions from discrete devices, was used to estimate non-Partner emissions for this time period.
44 EPA considered applying this change to years before 2007, but found that it would be difficult due to the large amount of data
(i.e., technology-specific global and non-Partner TMLA) that would have to be examined and manipulated for each year. This
effort did not appear to be justified given the relatively small impact of the improvement on the total estimate for 2007 and the
fact that the impact of the improvement would likely be lower for earlier years because the estimated share of emissions
accounted for by non-Partners is growing as Partners continue to implement emission-reduction efforts.
4-90 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 EPA's PFC Reduction/Climate Partnership as well as non-Partners. In EPA's GHGRP, the population of non-
2 Partner facilities also included manufacturers that use GaAs technology in addition to Si technology45. Emissions
3 from the population of manufacturers that were below the reporting threshold were also estimated for this time
4 period using EPA-developed emission factors and estimates of facility-specific production obtained from WFF.
5 Inventory totals reflect the emissions from both populations.
6 Under EPA's GHGRP, semiconductor manufacturing facilities report emissions of fluorinated GHGs used in etch
7 and clean processes and as heat transfer fluids. They also report N2O emissions from CVD and other processes.
8 The fluorinated GHGs, and N2O were aggregated, by gas, across all semiconductor manufacturing GHGRP reporters
9 to calculate gas-specific emissions for the GHGRP-reporting segment of the U.S. industry.
10 For the segment of the semiconductor industry, which is below EPA's GHGRP reporting threshold, and for R&D
11 facilities, which are not covered by EPA's GHGRP, emission estimates are based on EPA-developed emission
12 factors for the fluorinated GHGs and N2O. The new emission factors (in units of mass of CO2 Eq. / TMLA [MSI])
13 are based on the emissions reported by facilities under EPA's GHGRP and TMLA estimates for these facilities from
14 the WFF (SEMI 2012 and SEMI 2013). In a refinement of the method used in prior years to estimate emissions for
15 the non-Partner population, different emission factors were developed for different subpopulations of fabs, one for
16 facilities that manufacture devices on Si wafers and one for facilities that manufacture on GaAs wafers. An analysis
17 of the emission factors of reporting fabs showed that the characteristics that had the largest impacts on emission
18 factors were the substrate (i.e., Si or GaAs) used at the fab, whether the fab contained R&D activities, and whether
19 the fab reported using point-of-use fluorinated greenhouse gas abatement46. For each of these groups, a
20 subpopulation-specific emission factor was obtained using a regression-through-the-origin (RTO) model: facility -
21 reported aggregate emissions of seven fluorinated GHGs (CF4, C2F6, CsF8, C4p8, CHF3, SF6 and NF3)47 were
22 regressed against the corresponding TMLA to estimate an aggregate F-GHG emissions factor (CO2 Eq./MSI TMLA)
23 and facility-reported N2O emissions were regressed against the corresponding TMLA to estimate a N2O emissions
24 factor (CO2 Eq./MSI TMLA). For each subpopulation, the slope of the RTO model is the emission factor for that
25 subpopulation. To estimate emissions from fabs that are solely doing research and development (R&D) or are Pilot
26 fabs (i.e., fabs that are excluded from subpart I reporting requirements), emission factors were estimated based on
27 GHGRP reporting fabs containing R&D activities. EPA applied a scaling factor of 1.15 to the slope of the RTO
28 model to estimate the emission factor applicable to the non-reporting fabs that are only R&D or Pilot fabs. This was
29 done as R&D activities lead to use of more F-GHGs and N2O for development of chips that are not counted towards
30 the final estimated TMLA. Hence, it is assumed that the fabs with only R&D activities use 15 percent more F-GHGs
31 and N2O per TMLA. However, as was assumed for 2007 through 2010, fabs with only R&D activities were assumed
32 to utilize only 20 percent of their manufacturing capacity. Other fabs were assumed to utilize 89 percent of their
33 manufacturing capacity, held constant at 2012 levels which is slightly lower than 2011 levels. Fabs that produce
34 discrete products are assumed to utilize 84 percent of their manufacturing capacity, held constant at 2011 levels.
35 These utilizations at 2011 levels are based on the Semiconductor Industry Association report (SICAS, 2011).
36 Non-reporting fabs were then broken out into similar subpopulations. Information on the technology and R&D
37 activities of non-reporting fabs was available through the WFF. Information on the use of point-of-use abatement
38 by non-reporting fabs was not available; thus, EPA conservatively assumed that non-reporting facilities did not use
39 point-of-use abatement. The appropriate emission factor was applied to the total TMLA of each subpopulation of
40 non-reporting facilities to estimate the GWP-weighted emissions of that subpopulation.
41 Gas-specific, GWP-weighted emissions for each subpopulation of non-reporting facilities were estimated using the
42 corresponding reported distribution of gas-specific, GWP-weighted emissions from which the aggregate emission
43 factors were developed. Estimated in this manner, the non-reporting population accounted for 9, 10 and 10 percent
45 GaAs and Si technologies refer to the wafer on which devices are manufactured, which use the same PFCs but in different
ways.
For the non-reporting segment of the industry using GaAs technology, emissions were estimated only for those fabs that
manufactured the same products as manufactured by reporters. The products manufactured were categorized as discrete
(emissions did not scale up with decreasing feature size).
Only seven gases were aggregated because inclusion of fluorinated GHGs that are not reported in the inventory results in
overestimation of emission factor that is applied to the various non-reporting subpopulations.
Industrial Processes and Product Use 4-91
-------
1 of U.S. emissions in 2011, 2012 and 2013, respectively. The GHGRP-reported emissions and the calculated non-
2 reporting population emissions are summed to estimate the total emissions from semiconductor manufacturing.
3 The methodology used for this time period included, for the first time, emissions from facilities employing Si- and
4 GaAs-using technologies. The use of GaAs technology became evident via analysis of GHGRP emissions and WFF
5 data. However, no adjustment of pre-2011 emissions was made because (1) the use of these technologies appears
6 relatively new, (2) in the aggregate make a relatively small contribution to total industry emissions (i.e., 4 percent in
7 2013), and (3) would require a large effort to retroactively adjust pre-2011 emissions.
8 Data Sources
9 GHGRP reporters estimated their emissions using a default emission factor method established by EPA. This
10 method is very similar to the Tier 2b Method in the 2006IPCC Guidelines, but it goes beyond that method by
11 establishing different default emission and byproduct generation factors for different wafer sizes (i.e., 300mm vs.
12 150 and 200mm) and CVD clean subtypes (in situ thermal, in situ thermal, and remote plasma). Partners estimated
13 their emissions using a range of methods. It is assumed that most Partners used a method at least as accurate as the
14 IPCC's Tier 2a Methodology, recommended in the 2006 IPCC Guidelines. Estimates of operating plant capacities
15 and characteristics for Partners and non-Partners were derived from the Semiconductor Equipment and Materials
16 Industry (SEMI) WFF (formerly World Fab Watch) database (1996 through 2013) (e.g., Semiconductor Materials
17 and Equipment Industry, 2013). Actual worldwide capacity utilizations for 2011 were obtained from Semiconductor
18 International Capacity Statistics (SICAS) (SIA, 2011). Estimates of the number of layers for each linewidth was
19 obtained from International Technology Roadmap for Semiconductors: 2013 Edition (Burton and Beizaie 2001,
20 ITRS 2007, ITRS 2008, ITRS 2011, ITRS 2013). PEVM utilized the WFF, SICAS, and ITRS, as well as historical
21 silicon consumption estimates published by VLSI.
22
Uncertainty and Time-Series Consistency
23 A quantitative uncertainty analysis of this source category was performed using the IPCC-recommended Approach 2
24 uncertainty estimation methodology, the Monte Carlo Stochastic Simulation technique. The equation used to
25 estimate uncertainty is:
26 Total Emissions (ET) = GHGRP Reported F-GHG Emissions (ER,F.GHG) + Non-Reporters' Estimated F-GHG
27 E
28
29 where ER and ENR denote totals for the indicated subcategories of emissions for F-GHG and N2O, respectively.
30 The uncertainty in ET presented in Table 4-93 below results from the convolution of four distributions of emissions,
31 each reflecting separate estimates of possible values of ERJ-GHG, ER)N20, ENR.F-GHG, and EMU^D. The approach and
32 methods for estimating each distribution and combining them to arrive at the reported 95 percent CI are described in
33 the remainder of this section.
34 The uncertainty estimate of ER, F-GHG, or GHGRP reported F-GHG emissions, is developed based on gas-specific
35 uncertainty estimates of emissions for two industry segments, one processing 200 mm wafers and one processing
36 300 mm wafers. Uncertainties in emissions for each gas and industry segment were developed during the assessment
37 of emission estimation methods for the subpart I GHGRP rulemaking in 2012 (see Technical Support for
38 Modifications to the Fluorinated Greenhouse Gas Emission Estimation Method Option for Semiconductor Facilities
39 under Subpart I, docket EPA-HQ-OAR-2011-0028).48 The 2012 analysis did not take into account the use of
48 On November 13, 2013, EPA published a final rule revising subpart I (Electronics Manufacturing) of the GHGRP (78 FR
68162). The revised rule includes updated default emission factors and updated default destruction and removal efficiencies that
are slightly different from those that semiconductor manufacturers were required to use to report their 2012 emissions. The
uncertainty analyses that were performed during the development of the revised rule focused on these updated defaults, but are
expected to be reasonably representative of the uncertainties associated with the older defaults, particularly for estimates at the
country level. (They may somewhat underestimate the uncertainties associated with the older defaults at the facility level.) For
simplicity, the 2012 estimates are assumed to be unbiased although in some cases, the updated (and therefore more
4-92 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 abatement. For the industry segment that processed 200 mm wafers, estimates of uncertainties at a 95 percent CI
2 ranged from ±29 percent for CsF8 to ±10 percent for CF4. For the corresponding 300 mm industry segment,
3 estimates of the 95 percent CI ranged from ±36 percent for C^s to ±16 percent for CF4. These gas and wafer-
4 specific uncertainty estimates are applied to the total emissions of the facilities that did not abate emissions as
5 reported under EPA's GHGRP.
6 For those facilities reporting abatement of emissions under EPA's GHGRP, estimates of uncertainties for the no
7 abatement industry segments are modified to reflect the use of full abatement (abatement of all gases from all
8 cleaning and etching equipment) and partial abatement. These assumptions used to develop uncertainties for the
9 partial and full abatement facilities are identical for 200 mm and 300 mm wafer processing facilities. For all
10 facilities reporting gas abatement, a triangular distribution of destruction or removal efficiency is assumed for each
11 gas. The triangular distributions range from an asymmetric and highly uncertain distribution of 0 percent minimum
12 to 90 percent maximum with 70 percent most likely value for CF4 to a symmetric and less uncertain distribution of
13 85 percent minimum to 95 percent maximum with 90 percent most likely value for C4p8, NFs and SFe. For facilities
14 reporting partial abatement, the distribution of fraction of the gas fed through the abatement device, for each gas, is
15 assumed to be triangularly distributed as well. It is assumed that no more than 50 percent of the gases area abated
16 (i.e., the maximum value) and that 50 percent is the most likely value and the minimum is 0 percent. Consideration
17 of abatement then resulted in four additional industry segments, two 200 mm wafer-processing segments (one fully
18 and one partially abating each gas) and two 300 mm wafer-processing segment (one fully and the other partially
19 abating each gas). Gas-specific emission uncertainties were estimated by convolving the distributions of unabated
20 emissions with the appropriate distribution of abatement efficiency for fully and partially abated facilities using a
21 Montel Carlo simulation.
22 The uncertainty in ER,F-GHG is obtained by allocating the estimates of uncertainties to the total GHGRP-reported
23 emissions from each of the six industry segments, and then running a Monte Carlo simulation which results in the 95
24 percent CI for emissions from GHGRP reporting facilities (ER,F-GHG).
25 The uncertainty in ER)N2o is obtained by assuming that the uncertainty in the emissions reported by each of the
26 GHGRP reporting facilities results from the uncertainty in quantity of N2O consumed and the N2O emission factor
27 (or utilization). Similar to analyses completed for subpart I (see Technical Support for Modifications to the
28 Fluorinated Greenhouse Gas Emission Estimation Method Option for Semiconductor Facilities under Subpart I,
29 docket EPA-HQ-OAR-2011-0028), the uncertainty of N2O consumed was assumed to be 20 percent. Consumption
30 of N2O for GHGRP reporting facilities was estimated by back- calculating from emissions reported and assuming no
31 abatement. The quantity of N2O utilized (the complement of the emission factor) was assumed to have a triangular
32 distribution with a minimum value of 0 percent, mode of 20 percent and maximum value of 84 percent. The
33 minimum was selected based on physical limitations, the mode was set equivalent to the subpart I default N2O
34 utilization rate for chemical vapor deposition, and the maximum was set equal to the maximum utilization rate found
35 in ISMI Analysis of Nitrous Oxide Survey Data (ISMI, 2009). The inputs were used to simulate emissions for each
36 of the GHGRP reporting, N2O-emitting facilities. The uncertainty for the total reported N2O emissions was then
37 estimated by combining the uncertainties of each of the facilities reported emissions using Monte Carlo simulation.
38 The estimate of uncertainty in ENR.F-GHG and ENR.NM entailed developing estimates of uncertainties for the emissions
39 factors for each non-reporting sub-category and the corresponding estimates of TMLA.
40 The uncertainty in TMLA depends on the uncertainty of two variables—an estimate of the uncertainty in the average
41 annual capacity utilization for each level of production of fabs (e.g., full scale or R&D production) and a
42 corresponding estimate of the uncertainty in the number of layers manufactured. For both variables, the distributions
43 of capacity utilizations and number of manufactured layers are assumed triangular for all categories of non-reporting
44 fabs. For production fabs the most probable utilization is assumed to be 89 percent, with the highest and lowest
45 utilization assumed to be 100 percent and 63 percent, respectively. The corresponding values for facilities that
46 manufacture discrete devices are, 84 percent, 100 percent, and 66 percent, respectively, while the values for
47 utilization for R&D facilities, are assumed to be 20 percent, 33 percent, and 9 percent, respectively. The most
48 probable utilizations are unchanged compared to 2012 Inventory year. To address the uncertainty in the capacity
representative) defaults are higher or lower than the older defaults. Multiple models and sensitivity scenarios were run for the
subpart I analysis. The uncertainty analysis presented here made use of the Input gas and wafer size model (Model 1) under the
following conditions: Year = 2010, f = 20, n = SIA3.
Industrial Processes and Product Use 4-93
-------
1 utilization for Inventory year 2013, the lower bound has been decreased by 10 percent, and the upper bound has
2 been increased by 10 percent (or 100 percent if greater than 100 percent) compared to the bounds used in the 2012
3 Inventory year. For the triangular distributions that govern the number of possible layers manufactured, it is
4 assumed the most probable value is one layer less than reported in the ITRS; the smallest number varied by
5 technology generation between one and two layers less than given in the ITRS and largest number of layers
6 corresponded to the figure given in the ITRS.
7 The uncertainty bounds for the average capacity utilization and the number of layers manufactured are used as
8 inputs in a separate Monte Carlo simulation to estimate the uncertainty around the TMLA of both individual
9 facilities as well as the total non-reporting TMLA of each sub-population.
10 The uncertainty around the emission factors for each non-reporting category of facilities is dependent on the
11 uncertainty of the total emissions (MMT CC>2 Eq. units) and the TMLA of each reporting facility in that category.
12 For each subpopulation of reporting facilities, total emissions were regressed on TMLA (with an intercept forced to
13 zero) for 10,000 emissions and 10,000 TMLA values in a Monte Carlo simulation, which results in 10,000 total
14 regression coefficients (emission factors). The 2.5th and the 97.5th percentile of these emission factors are
15 determined and the bounds are assigned as the percent difference from the estimated emission factor.
16 For simplicity, the results of the Monte Carlo simulations on the bounds of the gas- and wafer size-specific
17 emissions as well as the TMLA and emission factors are assumed to be normally distributed and the uncertainty
18 bounds are assigned at 1.96 standard deviations around the estimated mean. The departures from normality were
19 observed to be small.
20 The final step in estimating the uncertainty in emissions of non-reporting facilities is convolving the distribution of
21 emission factors with the distribution of TMLA using Monte Carlo simulation.
22 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-93, which is also obtained
23 by convolving—using Monte Carlo simulation—the distributions of emissions for each reporting and non-reporting
24 facility. The emissions estimate for total U.S. F-GHG and N2O emissions from semiconductor manufacturing were
25 estimated to be between 4.0 and 4.4 MMT CCh Eq. at a 95 percent confidence level. This range represents 5 percent
26 below to 5 percent above the 2013 emission estimate of 4.2 MMT CCh Eq. This range and the associated
27 percentages apply to the estimate of total emissions rather than those of individual gases. Uncertainties associated
28 with individual gases will be somewhat higher than the aggregate, but were not explicitly modeled.
29 Table 4-93: Approach 2 Quantitative Uncertainty Estimates for HFC, RFC, SF6, NF3 and N2O
30 Emissions from Semiconductor Manufacture (MMT COz Eq. and Percent)
Source
Semiconductor
Manufacture
2013 Emission
Gas Estimate Uncertainty Range Relative to Emission Estimate3
(MMT C02 Eq.) (MMT CCh Eq.) (%)
HFC,
PFC
' 42
SF6, NF3,
andN2O
Lower
Boundb
4.0
Upper
Boundb
4.4
Lower
Bound
-5%
Upper
Bound
5%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence
interval.
b Absolute lower and upper bounds were calculated using the corresponding lower and upper bounds in percentages.
31 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
32 through 2013. Details on the emission trends through time are described in more detail in the Methodology section,
33 above.
34 Recalculations Discussion
35 For the current Inventory, emission estimates have been revised to reflect the GWPs provided in the IPCC Fourth
36 Assessment Report (AR4) (IPCC 2007). AR4 GWP values differ slightly from those presented in the IPCC Second
37 Assessment Report (SAR) (IPCC 1996) (used in the previous inventories) which results in time-series recalculations
4-94 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 for most inventory sources. Under the most recent reporting guidelines (UNFCCC 2014), countries are required to
2 report using the AR4 GWPs, which reflect an updated understanding of the atmospheric properties of each
3 greenhouse gas. The GWPs of CH4 and most fluorinated greenhouse gases have increased, leading to an overall
4 increase in emissions from CH4, HFCs, PFCs, SF6, and NF3. The AR4 GWPs have been applied across the entire
5 time series for consistency. For more information please see the Recalculations and Improvements Chapter.
6 The decrease in the GWP of SF6 and increase in the GWP of all other gases had several impacts on Inventory
7 estimates. In the 1990 through 1994 time period, an overall increase in total annual GWP-weighted emissions is
8 seen. In the 1995 through 2010 time period, the Inventory methodology relies on various gas distributions based on
9 Partner reported emissions and PEVM estimated emissions. The changes in GWP carry through to changes in the
10 estimated gas distributions, and hence changes in gas-by-gas emission estimates, in CCh Eq., and total annual
11 fluorinated greenhouse gas emission estimates, in CC>2 Eq..
12 For the first time, NF3 and N2O have been included in total annual GWP-weighted emission estimates for the United
13 States. This, along with an increased weighted GWP from SAR to AR4 led to increase in total emissions for all
14 years as compared to previous Inventories. The emissions of each gas were impacted by the increase in overall
15 emissions as well as the percent distribution of each gas as a result of changes in their GWPs.
16 Emissions in years 2011 and 2012 were updated to reflect updated emissions reporting in EPA's GHGRP. For the
17 non-reporting population, the methodology to determine the non-reporting population for GaAs using facilities has
18 been updated. In the updated methodology, revised assumptions were made about the GaAs using facilities that use
19 fluorinated greenhouse gases (e.g., only the non-reporters that use wafers greater than or equal to four inches have
20 been assumed to use fluorinated greenhouse gases, facilities that use wafers less than 4 inches are assumed to use
21 wet etching and hence do not consume or emit any fluorinated greenhouse gases). Further, EPA has drawn an
22 analogy between GaAs-using GHGRP reporters and non-reporters provided the non-reporters use wafers greater
23 than 4 inches and manufacture the many versions of high electron mobility transistors (HEMT, PHEMT, MHEMT,
24 HET, MOFETs), which are discrete devices and may be made to specific order by certain foundries. By virtue of
25 this analogy, EPA has estimated emissions only from the non-reporters that use GaAs technology and manufacture
26 HEMT and their variations. While other devices may be made using GaAs technology, EPA has no reporters under
27 the GHGRP that manufacture them and hence has no basis for estimating an emission factor. EPA has thus assumed
28 that they do not use or emit F-GHGs. This has decreased the non-reporting facilities subpopulation, and
29 subsequently total emissions for the years 2011 and 2012.
30
Planned Improvements
31 This Inventory contains estimates of seven fluorinated gases for semiconductor manufacturing. However, other
32 fluorinated gases (e.g., CsF8) are used in relatively smaller, but significant amounts. Previously, emissions data for
33 these other fluorinated gases was not reported through the EPA Partnership. Through EPA's GHGRP, these data, as
34 well as heat transfer fluid emission data, are available. Therefore, a point of consideration for future Inventory
35 reports is the inclusion of other fluorinated gases, and emissions from heat transfer fluid (RTF) loss to the
36 atmosphere.
37 Fluorinated heat transfer fluids, of which some are liquid perfluorinated compounds, are used for temperature
38 control, device testing, cleaning substrate surfaces and other parts, and soldering in certain types of semiconductor
39 manufacturing production processes. Evaporation of these fluids is a source of fluorinated emissions (EPA 2006).
40 The GHGRP-reported HTF emissions along with WFF database could be used to develop emission factors for
41 identified subpopulations. Further research needs to be done to determine if the same subpopulations identified in
42 developing new emission factors for F-GHGs are applicable or new subpopulations have to be studied as HTFs are
43 used primarily by manufacturers of wafer size 300 mm and above.
44 Along with more emissions information for semiconductor manufacturing, EPA's GHGRP requires the reporting of
45 emissions from other types of electronics manufacturing, including micro-electro-mechanical systems, flat panel
46 displays, and photovoltaic cells. There currently are no flat panel displays, and photovoltaic cell manufacturing
47 facilities that are reporting to EPA's GHGRP, and five reporting MEMs manufacturers. The MEMs manufacturers
48 also report emissions from semiconductor manufacturing and do not distinguish between these two types of
49 manufacturing in their report; thus, emissions from MEMs manufacturers are included in the totals here. Emissions
50 from manufacturing of flat panel displays and photovoltaic cells may be included in future Inventory reports;
51 however, estimation methodologies would need to be developed.
Industrial Processes and Product Use 4-95
-------
i 4.23 Substitution of Ozone Depleting
2 Substances (IPCC Source Category 2F)
3 Hydrofluorocarbons (HFCs) and perfluorocarbons (PFCs) are used as alternatives to several classes of ozone-
4 depleting substances (OD Ss) that are being phased out under the terms of the Montreal Protocol and the Clean Air
5 Act Amendments of 1990.49 Ozone depleting substances—chlorofluorocarbons (CFCs), halons, carbon
6 tetrachloride, methyl chloroform, and hydrochlorofluorocarbons (HCFCs)—are used in a variety of industrial
7 applications including refrigeration and air conditioning equipment, solvent cleaning, foam production, sterilization,
8 fire extinguishing, and aerosols. Although HFCs and PFCs are not harmful to the stratospheric ozone layer, they are
9 potent greenhouse gases. Emission estimates for HFCs and PFCs used as substitutes for ODSs are provided in Table
10 4-94 and Table 4-95.
11 Table 4-94: Emissions of HFCs and PFCs from ODS Substitutes (MMT COz Eq.)
Gas
HFC-23
HFC-32
HFC-125
HFC-134a
HFC-143a
HFC-236fa
CF4
Others3
Total
1990
+
+ 1
+ 1
+ 1
+ 1
:
0.3
0.3
2005
+
0.3 1
11.9
82.0 1
11.7 1
4
5.9
113.0
2010
+
2.6
32.5
87.0
22.1
1.4
+
7.8
153.5
2011
+
3.4
38.1
82.0
23.9
1.4
+
8.2
157.1
2012
+
4.4
44.2
77.2
25.5
1.5
+
8.6
161.4
2013
+
5.4
50.3
72.2
26.8
1.5
+
9.0
165.3
2014
+
6.4
56.1
70.2
27.7
1.4
+
9.4
171.4
12 + Does not exceed 0.05 MMT CO2 Eq.
13 a Others include HFC-152a, HFC-227ea, HFC-245fa, HFC-43-10mee, HFO-1234yf, C4Fio, and PFC/PFPEs, the latter being a
14 proxy for a diverse collection of PFCs and perfluoropolyethers (PFPEs) employed for solvent applications. For estimating
15 purposes, the GWP value used for PFC/PFPEs was based upon CeFi4.
16 Note: Totals may not sum due to independent rounding.
17
18 Table 4-95: Emissions of HFCs and PFCs from ODS Substitution (Metric Tons)
Gas
HFC-23
HFC-32
HFC-125
HFC-134a
HFC-143a
HFC-236fa
CF4
Others3
1990
+
-r -
M
2005
1 1
508
3,390
57,335
2,613
125
2
M |
2010
2
3,906
9,294
60,848
4,937
146
3
M
2011
2
5,022
10,874
57,356
5,354
147
4
M
2012
2
6,467
12,627
54,013
5,699
148
4
M
2013
2
7,972
14,378
50,516
5,991
151
4
M
2014
3
9,460
16,036
49,101
6,204
148
4
M
19 M (Mixture of Gases)
20 + Does not exceed 0.5 MT
21 a Others include HFC-152a, HFC-227ea, HFC-245fa, HFC-43-10mee, HFO-1234yf, C4Fio, and PFC/PFPEs, the latter being a
22 proxy for a diverse collection of PFCs and perfluoropolyethers (PFPEs) employed for solvent applications.
23
24 In 1990 and 1991, the only significant emissions of HFCs and PFCs as substitutes to ODSs were relatively small
25 amounts of HFC-152a—used as an aerosol propellant and also a component of the refrigerant blend R-500 used in
26 chillers—and HFC-134a in refrigeration end-uses. Beginning in 1992, HFC-134a was used in growing amounts as a
27 refrigerant in motor vehicle air-conditioners and in refrigerant blends such as R-404A.50 In 1993, the use of HFCs
49 [42 U.S.C § 7671, CAA Title VI]
50 R-404A contains HFC-125, HFC-143a, andHFC-134a.
4-96 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 in foam production began, and in 1994 ODS substitutes for halons entered widespread use in the United States as
2 halon production was phased-out. In 1995, these compounds also found applications as solvents.
3 The use and subsequent emissions of HFCs and PFCs as ODS substitutes has been increasing from small amounts in
4 1990 to 171.4 MMT €62 Eq. in 2014. This increase was in large part the result of efforts to phase out CFCs and
5 other ODSs in the United States. In the short term, this trend is expected to continue, and will likely continue over
6 the next decade as HCFCs, which are interim substitutes in many applications, are themselves phased-out under the
7 provisions of the Copenhagen Amendments to the Montreal Protocol. Improvements in the technologies associated
8 with the use of these gases and the introduction of alternative gases and technologies, however, may help to offset
9 this anticipated increase in emissions.
10 Table 4-96 presents emissions of HFCs and PFCs as ODS substitutes by end-use sector for 1990 through 2014. The
11 end-use sectors that contributed the most toward emissions of HFCs and PFCs as ODS substitutes in 2014 include
12 refrigeration and air-conditioning (149.4 MMT CO2 Eq., or approximately 87 percent), aerosols (10.8 MMT CO2
13 Eq., or approximately 6 percent), and foams (8.0 MMT CO2 Eq., or approximately 4 percent). Within the
14 refrigeration and air-conditioning end-use sector, motor vehicle air-conditioning was the highest emitting end-use
15 (40.9 MMT CO2 Eq.), followed by refrigerated retail food and refrigerated transport. Each of the end-use sectors is
16 described in more detail below.
17 Table 4-96: Emissions of HFCs and PFCs from ODS Substitutes (MMT COz Eq.) by Sector
Sector
Refrigeration/Air
Conditioning
Aerosols
Foams
Solvents
Fire Protection
Total
1990
I+
.
+
0.3 H
2005
101.1
17.6
2.1
1.7
0.7
113.0
2010
135.1
19.7
5.9
1.7
1.1
153.5
2011
137.7
10.1
6.4
1.7
1.2
157.1
2012
141.2
10.3
6.9
1.7
1.3
161.4
2013
144.2
10.5
7.4
1.8
1.3
165.3
2014
149.4
10.8
8.0
1.8
1.4
171.4
18 + Does not exceed 0.05 MMT CO2 Eq.
19 Refrigeration/Air Conditioning
20 The refrigeration and air-conditioning sector includes a wide variety of equipment types that have historically used
21 CFCs or HCFCs. End-uses within this sector include motor vehicle air-conditioning, retail food refrigeration,
22 refrigerated transport (e.g., ship holds, truck trailers, railway freight cars), household refrigeration, residential and
23 small commercial air-conditioning and heat pumps, chillers (large comfort cooling), cold storage facilities, and
24 industrial process refrigeration (e.g., systems used in food processing, chemical, petrochemical, pharmaceutical, oil
25 and gas, and metallurgical industries). As the ODS phaseout is taking effect, most equipment is being or will
26 eventually be retrofitted or replaced to use HFC-based substitutes. Common HFCs in use today in refrigeration/air-
27 conditioning equipment are HFC-134a, R-410A,51 R-404A, and R-507A.52 Low-GWP options such as HFO-
28 1234yf in motor vehicle air-conditioning, R-717 (ammonia) in cold storage and industrial applications, and R-744
29 (carbon dioxide) in retail food refrigeration, are also being used. These refrigerants are emitted to the atmosphere
30 during equipment manufacture and operation (as a result of component failure, leaks, and purges), as well as at
31 servicing and disposal events.
32 Aerosols
33 Aerosol propellants are used in metered dose inhalers (MDIs) and a variety of personal care products and
34 technical/specialty products (e.g., duster sprays and safety horns). Many pharmaceutical companies that produce
35 MDIs—a type of inhaled therapy used to treat asthma and chronic obstructive pulmonary disease—have replaced
36 the use of CFCs with HFC-propellant alternatives. The earliest ozone-friendly MDIs were produced with HFC-134a,
37 but the industry has started to use HFC-227ea as well. Conversely, since the use of CFC propellants was banned in
38 1978, most non-medical consumer aerosol products have not transitioned to HFCs, but to "not-in-kind"
51 R-410A contains HFC-32 andHFC-125.
52 R-507A, also called R-507, contains HFC-125 and HFC-143a.
Industrial Processes and Product Use 4-97
-------
1 technologies, such as solid roll-on deodorants and finger-pump sprays. The transition away from ODS in specialty
2 aerosol products has also led to the introduction of non-fluorocarbon alternatives (e.g., hydrocarbon propellants) in
3 certain applications, in addition to HFC-134a or HFC-152a. Other low-GWP options such as HFO-1234ze(E) are
4 being used as well. These propellants are released into the atmosphere as the aerosol products are used.
5 Foams
6 CFCs and HCFCs have traditionally been used as foam blowing agents to produce polyurethane (PU), polystyrene,
7 polyolefm, and phenolic foams, which are used in a wide variety of products and applications. Since the Montreal
8 Protocol, flexible PU foams as well as other types of foam, such as polystyrene sheet, polyolefin, and phenolic
9 foam, have transitioned almost completely away from fluorocompounds, into alternatives such as CC>2, methylene
10 chloride, and hydrocarbons. The majority of rigid PU foams have transitioned to HFCs—primarily HFC-134a and
11 HFC-245fa. Today, these HFCs are used to produce polyurethane appliance, PU commercial refrigeration, PU spray,
12 and PU panel foams—used in refrigerators, vending machines, roofing, wall insulation, garage doors, and cold
13 storage applications. In addition, HFC-152a, HFC-134a and CCh are used to produce polystyrene sheet/board foam,
14 which is used in food packaging and building insulation. Low-GWP fluorinated foam blowing agents in use include
15 HFO-1234ze(E) and -1233zd(E). Emissions of blowing agents occur when the foam is manufactured as well as
16 during the foam lifetime and at foam disposal, depending on the particular foam type.
17 Solvents
18 CFCs, methyl chloroform (1,1,1 -trichloroethane or TCA), and to a lesser extent carbon tetrachloride (CCU) were
19 historically used as solvents in a wide range of cleaning applications, including precision, electronics, and metal
20 cleaning. Since their phaseout, metal cleaning end-use applications have primarily transitioned to non-fluorocarbon
21 solvents and not-in-kind processes. The precision and electronics cleaning end-uses have transitioned in part to high-
22 GWP gases, due to their high reliability, excellent compatibility, good stability, low toxicity, and selective solvency.
23 These applications rely on HFC-43-10mee, HFC-365mfc, HFC-245fa, and to a lesser extent, PFCs. Electronics
24 cleaning involves removing flux residue that remains after a soldering operation for printed circuit boards and other
25 contamination-sensitive electronics applications. Precision cleaning may apply to either electronic components or to
26 metal surfaces, and is characterized by products, such as disk drives, gyroscopes, and optical components, that
27 require a high level of cleanliness and generally have complex shapes, small clearances, and other cleaning
28 challenges. The use of solvents yields fugitive emissions of these HFCs and PFCs.
29 Fire Protection
30 Fire protection applications include portable fire extinguishers ("streaming" applications) that originally used halon
31 1211, and total flooding applications that originally used halon 1301, as well as some halon 2402. Since the
32 production and import of virgin halons were banned in the United States in 1994, the halon replacement agent of
33 choice in the streaming sector has been dry chemical, although HFC-236fa is also used to a limited extent. In the
34 total flooding sector, HFC-227ea has emerged as the primary replacement for halon 1301 in applications that require
35 clean agents. Other HFCs, such as HFC-23 and HFC-125, are used in smaller amounts. The majority of HFC-227ea
36 in total flooding systems is used to protect essential electronics, as well as in civil aviation, military mobile weapons
37 systems, oil/gas/other process industries, and merchant shipping. Fluoroketone FK-5-1-12 is also used as a low-
38 GWP option. As fire protection equipment is tested or deployed, emissions of these HFCs occur.
39 Methodology
40 A detailed Vintaging Model of ODS-containing equipment and products was used to estimate the actual—versus
41 potential—emissions of various ODS substitutes, including HFCs and PFCs. The name of the model refers to the
42 fact that it tracks the use and emissions of various compounds for the annual "vintages" of new equipment that enter
43 service in each end-use. The Vintaging Model predicts ODS and ODS substitute use in the United States based on
44 modeled estimates of the quantity of equipment or products sold each year containing these chemicals and the
45 amount of the chemical required to manufacture and/or maintain equipment and products over time. Emissions for
46 each end-use were estimated by applying annual leak rates and release profiles, which account for the lag in
47 emissions from equipment as they leak over time. By aggregating the data for more than 60 different end-uses, the
4-98 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 model produces estimates of annual use and emissions of each compound. Further information on the Vintaging
2 Model is contained in Annex 3.9.
Uncertainty and Time-Series Consistency
4 Given that emissions of ODS substitutes occur from thousands of different kinds of equipment and from millions of
5 point and mobile sources throughout the United States, emission estimates must be made using analytical tools such
6 as the Vintaging Model or the methods outlined in IPCC (2006). Though the model is more comprehensive than the
7 IPCC default methodology, significant uncertainties still exist with regard to the levels of equipment sales,
8 equipment characteristics, and end-use emissions profiles that were used to estimate annual emissions for the
9 various compounds.
10 The Vintaging Model estimates emissions from 60 end-uses. The uncertainty analysis, however, quantifies the level
11 of uncertainty associated with the aggregate emissions resulting from the top 21 end-uses, comprising over 95
12 percent of the total emissions, and 6 other end-uses. These 27 end-uses comprise 97 percent of the total emissions,
13 equivalent to 166.4 MMT CCh Eq. In an effort to improve the uncertainty analysis, additional end-uses are added
14 annually, with the intention that over time uncertainty for all emissions from the Vintaging Model will be fully
15 characterized. Any end-uses included in previous years' uncertainty analysis were included in the current
16 uncertainty analysis, whether or not those end-uses were included in the top 95 percent of emissions from ODS
17 Substitutes.
18 In order to calculate uncertainty, functional forms were developed to simplify some of the complex "vintaging"
19 aspects of some end-use sectors, especially with respect to refrigeration and air-conditioning, and to a lesser degree,
20 fire extinguishing. These sectors calculate emissions based on the entire lifetime of equipment, not just equipment
21 put into commission in the current year, thereby necessitating simplifying equations. The functional forms used
22 variables that included growth rates, emission factors, transition from ODSs, change in charge size as a result of the
23 transition, disposal quantities, disposal emission rates, and either stock for the current year or original ODS
24 consumption. Uncertainty was estimated around each variable within the functional forms based on expert
25 judgment, and a Monte Carlo analysis was performed. The most significant sources of uncertainty for this source
26 category include the emission factors for residential unitary AC, as well as the percent of non-MDI aerosol
27 propellantthatisHFC-152a.
28 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-97. Substitution of ozone
29 depleting substances HFC and PFC emissions were estimated to be between 153.0 and 172.3 MMT CO2 Eq. at the
30 95 percent confidence level. This indicates a range of approximately 0.22 percent below to 12.4 percent above the
31 emission estimate of 158.6 MMT CO2Eq.
32 Table 4-97: Approach 2 Quantitative Uncertainty Estimates for HFC and PFC Emissions from
33 ODS Substitutes (MMT COz Eq. and Percent)
2014 Emission
Source Gases Estimate
(MMT CO2 Eq.)a
Uncertainty Range Relative to Emission Estimate1"
(MMT CO2 Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
Substitution of Ozone HFCs and 1586
Depleting Substances PFCs
153.0 172.3 -0.22% +12.4%
34 a 2014 emission estimates and the uncertainty range presented in this table correspond to selected end-uses within the aerosols,
35 foams, solvents, fire extinguishing agents, and refrigerants sectors that comprise 97 percent of total emissions, but not for other
36 remaining categories. Therefore, because the uncertainty associated with emissions from "other" ODS substitutes was not
37 estimated, they were excluded in the uncertainty estimates reported in this table.
38 b Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
39 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
40 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
41 above.
Industrial Processes and Product Use 4-99
-------
i Comparison of Reported Consumption to Modeled Consumption of MFCs
2 Data from EPA's Greenhouse Gas Reporting Program (GHGRP) was also used to perform quality control on the
3 modeled emissions from this source category. To do so, reported consumption patterns demonstrated through
4 GHGRP Subpart OO: Suppliers of Industrial Greenhouse Gases reported data were compared to the modeled
5 demand for new saturated HFCs (excluding HFC-23) used as ODS substitutes from the Vintaging Model. The
6 collection of data from suppliers of HFCs enables EPA to calculate the reporters' aggregated net supply-the sum of
7 the quantities of chemical produced or imported into the United States less the sum of the quantities of chemical
8 transformed (used as a feedstock in the production of other chemicals), destroyed, or exported from the United
9 States.53 This allows for a quality control check on emissions from this source because the Vintaging Model uses
10 modeled demand for new chemical as a proxy for total amount supplied, which is similar to net supply, as an input
11 to the emission calculations in the model.
12 Reported Net Supply (GHGRP Top-Down Estimate)
13 Under EPA's GHGRP, suppliers (i.e., producers, importers and exporters) of HFCs began annually reporting their
14 production, transformation, destruction, imports, and exports to EPA in 2011 (for supply that occurred in 2010). For
15 the first time in 2015, bulk consumption data for aggregated HFCs were made publicly available under EPA's
16 GHGRP. Data include all saturated HFCs (except HFC-23) reported to EPA across the GHGRP-reporting time
17 series (2010 through 2014). The data include all 19 such saturated HFCs listed in Table A-1 of 40 CFR Part 98,
18 where regulations for the GHGRP are promulgated, though not all species were reported in each reporting year.
19 Modeled Consumption (Vintaging Model Bottom- Up Estimate)
20 The Vintaging Model, used to estimate emissions from this source category, calculates chemical demand based on
21 the quantity of equipment and products sold, serviced and retired each year, and the amount of the chemical required
54
22 to manufacture and/or maintain the equipment and products. It is assumed that the total demand equals the amount
23 supplied by either new production, chemical import, or quantities recovered (usually reclaimed) and placed back on
24 the market. In the Vintaging Model, demand for new chemical, as a proxy for consumption, is calculated as any
25 chemical demand (either for new equipment or for servicing existing equipment) that cannot be met through
26 recycled or recovered material. No distinction is made in the Vintaging Model between whether that need is met
27 through domestic production or imports. To calculate emissions, the Vintaging Model estimates the quantity
28 released from equipment over time. Thus, verifying the Vintaging Model's calculated consumption against GHGRP
29 reported data is one way to check the Vintaging Model's emission estimates.
30 There are ten saturated HFC species modeled in the Vintaging Model: HFC-23, HFC-32, HFC-125, HFC-134a,
31 HFC-143a, HFC-152a, HFC-227ea, HFC-236fa, HFC-245fa, and HFC-43-10mee. For the purposes of this
32 comparison, only nine HFC species are included (HFC-23 is excluded), to more closely align with the aggregated
33 total reported under the GHGRP. While some amounts of less-used saturated HFCs, including isomers of those
34 included in the Vintaging Model, are reportable under the GHGRP, the data are believed to represent an amount
35 comparable to the modeled estimates as a quality control check.
36 Comparison Results and Discussion
37 Comparing the estimates of consumption from these two approaches ultimately supports and improves estimates of
38 emissions, as noted in the 2006IPCC Guidelines for National Greenhouse Gas Inventories (which refer to
39 fluorinated greenhouse gas consumption based on supplies as "potential emissions"):
40 [W]hen considered along with estimates of actual emissions, the potential emissions approach can assist in
41 validation of completeness of sources covered and as a QC check by comparing total domestic
Chemical that is exported, transformed, or destroyed—unless otherwise imported back to the United States—will never be
emitted in the United States.
54 The model builds an inventory of the in-use stock of equipment and products and ODSs and HFCs in each of the sub-
applications. Emissions are subsequently estimated by applying annual and disposal emission rates to each population of
equipment and products.
4-100 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 consumption as calculated in this 'potential emissions approach' per compound with the sum of all activity
2 data of the various uses (IPCC 2006).
3 Table 4-98 compares the net supply of saturated HFCs (excluding HFC-23) in million metric tons of CCh Eq. as
4 determined from SubpartOO of EPA's GHGRP for the years 2010 through 2014 and the chemical demand as
5 calculated by the Vintaging Model for the same time series.
6 Table 4-98: U.S. HFC Consumption (MMT COz Eq.)
2010 2011 2012 2013 2014
Reported Net Supply (GHGRP) 235 241 227 278 254
Modeled Supply (Vintaging Model) 268 267 284 289 292
Percent Difference 14% 11% 25% 4% 15%
7
8 As shown, the estimates from the Vintaging Model are higher than the GHGRP estimates by an average of 13
9 percent across the time series (i.e., 2010-2014). Potential reasons for these differences include:
10 • The Vintaging Model includes fewer HFCs than are reported to EPA's GHGRP. However, the additional
11 reported HFCs represent a small fraction of total HFC use for this source category, both in GWP-weighted
12 and unweighted terms, and as such, it is not expected that the additional HFCs reported to EPA are a major
13 driver for the difference between the two sets of estimates. To the extent lower-GWP isomers were used in
14 lieu of the modeled chemicals (e.g., HFC-134 instead of HFC-134a), lower CCh Eq. amounts in the
15 GHGRP data compared to the modeled estimates would be expected.
16 • Because the top-down data are reported at the time of actual production or import and the bottom-up data
17 are calculated at the time of actual placement on the market, there could be a temporal discrepancy when
18 comparing data. Because the GHGRP data generally increases over time (although some year-to-year
19 variations exist) and the Vintaging Model estimates also increase, EPA would expect the modeled estimates
20 to be slightly higher than the corresponding GHGRP data due to this temporal effect.
21 • Under EPA's GHGRP, all facilities that produce HFCs are required to report their quantities, whereas
22 importers or exporters of HFCs are only required to report if either their total imports or their total exports
23 of greenhouse gases are greater than or equal to 25,000 metric tons of CCh Eq. per year. Thus, some
24 imports may not be accounted for in the GHGRP data. On the other hand, some exports might also not be
25 accounted for in this data.
26 • In some years, imports and exports may be greater than consumption because the excess is being used to
27 increase chemical stockpiles; in other years, the opposite may hold true. Averaging imports and exports
28 over multiple years can minimize the impact of such fluctuations. For example, when the 2012 and 2013
29 net additions to the supply are averaged, the difference between the consumption estimates decreases
30 compared to the 2012-only estimates.
31 Table 4-99: Averaged U.S. HFC Demand (MMT COz Eq.)
Reported Net Supply (GHGRP)
Modeled Demand (Vintaging Model)
Percent Difference
2010-2011 Avg.
238
268
12%
2011-2012 Avg.
234
276
18%
2012-2013 Avg.
253
286
13%
2013-2014 Avg.
266
290
9%
32 • The Vintaging Model does not reflect the dynamic nature of reported HFC consumption, with significant
33 differences seen in each year. Whereas the Vintaging Model projects a slowly increasing overall demand,
34 actual consumption for specific chemicals may vary over time and could even switch from positive to
35 negative (indicating more chemical exported, transformed, or destroyed than produced or imported in a
36 given year). Furthermore, consumption as calculated in the Vintaging Model is a function of demand not
37 met by disposal recovery. If, in any given year, a significant number of units are disposed, there will be a
38 large amount of additional recovery in that year that can cause an unexpected and not modeled decrease in
39 demand and thus a decrease in consumption. On the other hand, if market, economic or other factors cause
40 less than expected disposal and recovery, actual supply would decrease, and hence consumption would
Industrial Processes and Product Use 4-101
-------
1 increase to meet that demand not satisfied by recovered quantities, increasing the GHGRP data and
2 bringing those totals closer to the Vintaging Model estimates.
3 • The Vintaging Model is used to estimate the emissions that occur in the United States. As such, equipment
4 or products that contain ODS or alternatives, including saturated HFCs, are assumed to consume and emit
5 chemicals equally as like equipment or products originally produced in the United States. Therefore, the
6 GHGRP data may include HFCs produced or imported and used to fill or manufacture products that are
7 then exported from the United States. The Vintaging Model estimates of demand and supply are not meant
8 to incorporate such chemical. Likewise, chemicals may be used outside the United States to create products
9 or charge equipment that is then imported to and used in the United States. The Vintaging Model estimates
10 of demand and supply are meant to capture this chemical, as it will lead to emissions inside the United
11 States. Depending on whether the U. S. is a net importer or net exporter of such chemical, this factor may
12 account for some of the difference shown above or might lead to a further discrepancy. Reporting under the
13 GHGRP Subpart QQ: Importers and Exporters of Fluorinated Greenhouse Gases Contained in Pre-Charged
14 Equipment or Closed-Cell Foams could be analyzed in the future to investigate this issue further.
15 One factor however would only lead to modeled estimates to be even higher than the estimates shown and hence
16 higher than EPA's GHGRP data:
17 • Saturated HFCs are also known to be used as a cover gas in the production of magnesium. The Vintaging
18 Model estimates here do not include the amount of HFCs for this use, but rather only the amount for uses
19 that traditionally were served by ODS. Nonetheless, EPA expects that this supply not included in the
20 Vintaging Model estimates to be very small compared to the ODS substitute use for the years analyzed. An
21 indication of the different magnitudes of these categories is seen in the fact that the 2014 emissions from
22 that non-modeled sources (0.2 MMT CO2 Eq.) are much smaller than those for the ODS substitute sector
23 (171.4 MMT CO2Eq).
24
25 Using a Tier 2 bottom-up modeling methodology to estimate emissions requires assumptions and expert judgement.
26 Comparing the Vintaging Model's estimates to GHGRP reported estimates, particularly for more widely used
27 chemicals, can help validate the model but it is expected that the model will have limitations. This comparison
28 shows that Vintaging Model consumption estimates are well within the same order of magnitude as the actual
29 consumption data as reported to EPA's GHGRP although the differences in reported net supply and modeled
30 demand are still significant in some of the years. Although it can be difficult to capture the observed market
31 variability, the Vintaging Model is periodically reviewed and updated to ensure that the model reflects the current
32 and future trajectory of ODS and ODS substitutes across all end-uses and the Vintaging Model will continue to be
33 compared to available top-down estimates in order to ensure the model accurately estimates HFC consumption and
34 emissions.
35
Recalculations Discussion
36 For the current Inventory, a review of the large retail food end-use resulted in revisions to the Vintaging Model since
37 the previous Inventory report. In addition, a vending machine end-use was added to the Vintaging Model since the
38 previous Inventory. Methodological recalculations were applied to the entire time-series to ensure time-series
39 consistency from 1990 through 2014.
40 For the large retail food end-use, assumptions regarding new installations by system type and refrigerant transitions
41 were revised based on a review of data collected by EPA's GreenChill Partnership and the California Air Resources
42 Board's Refrigerant Management Program. The vending machine end-use was added based on a review of technical
43 reports and sales data. Combined, these assumption changes and additions decreased greenhouse gas emissions on
44 average by 3 percent between 1990 and 2002 and increased greenhouse gas emissions on average by 4 percent
45 between 2003 and 2014.
4-102 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
i 4.24 Electrical Transmission and Distribution
(IPCC Source Category 2G1) (TO BE
3 UPDATED)
4 The largest use of sulfur hexafluoride (SF6), both in the United States and internationally, is as an electrical insulator
5 and interrupter in equipment that transmits and distributes electricity (RAND 2004). The gas has been employed by
6 the electric power industry in the United States since the 1950s because of its dielectric strength and arc-quenching
7 characteristics. It is used in gas-insulated substations, circuit breakers, and other switchgear. SF6 has replaced
8 flammable insulating oils in many applications and allows for more compact substations in dense urban areas.
9 Fugitive emissions of SF6 can escape from gas-insulated substations and switchgear through seals, especially from
10 older equipment. The gas can also be released during equipment manufacturing, installation, servicing, and
11 disposal. Emissions of SF6 from equipment manufacturing and from electrical transmission and distribution systems
12 were estimated to be 5.1 MMT CCh Eq. (0.2 kt) in 2013. This quantity represents an 80 percent decrease from the
13 estimate for 1990 (see Table 4-100 and Table 4-101). There are two potential causes for this decrease: a sharp
14 increase in the price of SF6 during the 1990s and a growing awareness of the magnitude and environmental impact
15 of SF6 emissions through programs such as EPA's voluntary SF6 Emission Reduction Partnership for Electric Power
16 Systems (Partnership) and EPA's GHGRP. Utilities participating in the Partnership have lowered their emission
17 factor (kg SF6 emitted per kg of nameplate capacity) by more than 75 percent since the Partnership began in 1999. A
18 recent examination of the SF6 emissions reported by electric power systems to EPA's GHGRP revealed that SF6
19 emissions from reporters has decreased by 25 percent from 2011 to 2013, with much of the reduction seen from
20 utilities that are not participants in the Partnership. These utilities may be making relatively large reductions in
21 emissions as they take advantage of relatively large and/or inexpensive emission reduction opportunities (i.e., "low
22 hanging fruit," such as replacing major leaking circuit breakers) that Partners have already taken advantage of under
23 the voluntary program (Ottinger et al. 2014).
24 Table 4-100: SFe Emissions from Electric Power Systems and Electrical Equipment
25 Manufacturers (MMT COz Eq.)
Year
1990
2005
2009
2010
2011
2012
2013
Electric Power
Systems
25.1
9.8
6.7
6.2
5.7
4.6
4.2
Electrical Equipment
Manufacturers
0.3
0.8
0.6
0.9
1.1
1.1
0.9
Total
25.4
10.6
7.3
7.0
6.8
5.7
5.1
Note: Totals may not sum due to independent rounding.
26 Table 4-101: SFe Emissions from Electric Power Systems and Electrical Equipment
27 Ma n ufactu rers ( kt)
Year Emissions
1990 1.1
2005 0.5
I
2009 0.3
2010 0.3
2011 0.3
Industrial Processes and Product Use 4-103
-------
2012 0.2
2013 02
i Methodology
2 The estimates of emissions from Electrical Transmission and Distribution are comprised of emissions from electric
3 power systems and emissions from the manufacture of electrical equipment. The methodologies for estimating both
4 sets of emissions are described below.
5 1990 through 1998 Emissions from Electric Power Systems
6 Emissions from electric power systems from 1990 through 1998 were estimated based on (1) the emissions
7 estimated for this source category in!999, which, as discussed in the next section, were based on the emissions
8 reported during the first year of EPA's SF6 Emission Reduction Partnership for Electric Power Systems
9 (Partnership), and (2) the RAND survey of global SF6 emissions. Because most utilities participating in the
10 Partnership reported emissions only for 1999 through 2011, modeling was used to estimate SF6 emissions from
11 electric power systems for the years 1990 through 1998. To perform this modeling, U.S. emissions were assumed to
12 follow the same trajectory as global emissions from this source during the 1990 to 1999 period. To estimate global
13 emissions, the RAND survey of global SF6 sales were used, together with the following equation for estimating
14 emissions, which is derived from the mass-balance equation for chemical emissions (Volume 3, Equation 7.3) in the
15 2006IPCC Guidelines (IPCC 2006)55 (Although Equation 7.3 of the 2006IPCC Guidelines appears in the
16 discussion of substitutes for ozone-depleting substances, it is applicable to emissions from any long-lived
17 pressurized equipment that is periodically serviced during its lifetime.)
18 Emissions (kilograms SF6) = SF6 purchased to refill existing equipment (kilograms) + nameplate capacity of retiring
19 equipment (kilograms)56
20 Note that the above equation holds whether the gas from retiring equipment is released or recaptured; if the gas is
21 recaptured, it is used to refill existing equipment, thereby lowering the amount of SF6 purchased by utilities for this
22 purpose.
23 Gas purchases by utilities and equipment manufacturers from 1961 through 2003 are available from the RAND
24 (2004) survey. To estimate the quantity of SF6 released or recovered from retiring equipment, the nameplate
25 capacity of retiring equipment in a given year was assumed to equal 81.2 percent of the amount of gas purchased by
26 electrical equipment manufacturers 40 years previous (e.g., in 2000, the nameplate capacity of retiring equipment
27 was assumed to equal 81.2 percent of the gas purchased in 1960). The remaining 18.8 percent was assumed to have
28 been emitted at the time of manufacture. The 18.8 percent emission factor is an average of IPCC default SF6
29 emission rates for Europe and Japan for 1995 (IPCC 2006). The 40-year lifetime for electrical equipment is also
30 based on IPCC (2006). The results of the two components of the above equation were then summed to yield
31 estimates of global SF6 emissions from 1990 through 1999.
32 U.S. emissions between 1990 and 1999 are assumed to follow the same trajectory as global emissions during this
33 period. To estimate U.S. emissions, global emissions for each year from 1990 through 1998 were divided by the
34 estimated global emissions from 1999. The result was a time series of factors that express each year's global
35 emissions as a multiple of 1999 global emissions. Historical U.S. emissions were estimated by multiplying the
36 factor for each respective year by the estimated U.S. emissions of SF6 from electric power systems in 1999
37 (estimated to be 14.3 MMT CO2 Eq.).
38 Two factors may affect the relationship between the RAND sales trends and actual global emission trends. One is
39 utilities' inventories of SF6 in storage containers. When SF6 prices rise, utilities are likely to deplete internal
40 inventories before purchasing new SF6 at the higher price, in which case SF6 sales will fall more quickly than
41 emissions. On the other hand, when SF6 prices fall, utilities are likely to purchase more SF6 to rebuild inventories,
55 Ideally, sales to utilities in the U.S. between 1990 and 1999 would be used as a model. However, this information was not
available. There were only two U.S. manufacturers of SFe during this time period, so it would not have been possible to conceal
sensitive sales information by aggregation.
56 Nameplate capacity is defined as the amount of SFe within fully charged electrical equipment.
4-104 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 in which case sales will rise more quickly than emissions. This effect was accounted for by applying 3 -year
2 smoothing to utility SF6 sales data. The other factor that may affect the relationship between the RAND sales trends
3 and actual global emissions is the level of imports from and exports to Russia and China. SF6 production in these
4 countries is not included in the RAND survey and is not accounted for in any another manner by RAND. However,
5 atmospheric studies confirm that the downward trend in estimated global emissions between 1995 and 1998 was real
6 (see the Uncertainty discussion below).
7 1999 through 2013 Emissions from Electric Power Systems
8 Emissions from electric power systems from 1999 to 2013 were estimated based on: (1) reporting from utilities
9 participating in EPA's SF6 Emission Reduction Partnership for Electric Power Systems (Partners), which began in
10 1999; (2) reporting from utilities covered by the EPA's GHGRP, which began in 2012 for emissions occurring in
11 2011 (GHGRP-Only Reporters); and (3) the relationship between utilities' reported emissions and their
12 transmission miles as reported in the 2001, 2004, 2007, 2010, and 2013 Utility Data Institute (UDI) Directories of
13 Electric Power Producers and Distributors (UDI 2001, 2004, 2007, 2010, 2013), which was applied to the electric
14 power systems that do not report to EPA (Non-Reporters). (Transmission miles are defined as the miles of lines
15 carrying voltages above 34.5 kV).
16 Partners
17 Over the period from 1999 to 2013, Partner utilities, which for inventory purposes are defined as utilities that either
18 currently are or previously have been part of the Partnership, represented between 42 percent and 48 percent of total
19 U.S. transmission miles. Partner utilities estimated their emissions using a Tier 3 utility-level mass balance
20 approach (IPCC 2006). If a Partner utility did not provide data for a particular year, emissions were interpolated
21 between years for which data were available or extrapolated based on Partner-specific transmission mile growth
22 rates. In 2012, many Partners began reporting their emissions (for 2011 and later years) through EPA's GHGRP
23 (discussed further below) rather than through the Partnership. In 2013, approximately 0.3 percent of the total
24 emissions attributed to Partner utilities were reported through Partnership reports. Approximately 91 percent of the
25 total emissions attributed to Partner utilities were reported and verified through EPA's GHGRP. Partners without
26 verified 2013 data accounted for approximately 9 percent of the total emissions attributed to Partner utilities.57
27 GHGRP-Only Reporters
28 EPA's GHGRP requires users of SF6 in electric power systems to report emissions if the facility has a total SF6
29 nameplate capacity that exceeds 17,820 pounds. (This quantity is the nameplate capacity that would result in annual
30 SF6 emissions equal to 25,000 metric tons of CO2 equivalent at the historical emission rate reported under the
31 Partnership.) As under the Partnership, electric power systems that report their SF6 emissions under EPA's GHGRP
32 are required to use the Tier 3 utility-level mass-balance approach. Many Partners began reporting their emissions
33 through EPA's GHGRP in 2012 (reporting emissions for 2011 and later years) because their nameplate capacity
34 exceeded the reporting threshold. Partners who did not report through EPA's GHGRP continued to report through
35 the Partnership.
36 In addition, many non-Partners began reporting to EPA for the first time through its GHGRP in 2012. Non-Partner
37 emissions reported and verified under EPA's GHGRP were compiled to form a new category of reported data
57 It should be noted that data reported through EPA's GHGRP must go through a verification process; only data verified as of
September 1, 2014 could be used in the emission estimates for 2013. For Partners whose GHGRP data was not yet verified,
emissions were extrapolated based upon historical Partner-specific transmission mile growth rates, and those Partners are
included in the 'non-reporting Partners' category.
For electric power systems, verification involved a series of electronic range, completeness, and algorithm checks for each report
submitted. In addition, EPA manually reviewed the reported data and compared each facility's reported transmission miles with
the corresponding quantity in the UDI 2013 database (UDI 2013). In the first year of GHGRP reporting, EPA followed up with
reporters where the discrepancy between the reported miles and the miles published by UDI was greater than 10 percent, with a
goal to improve data quality. Only GHGRP data verified as of September 1, 2014 was included in the emission estimates for
2011,2012, and2013.
Industrial Processes and Product Use 4-105
-------
1 (GHGRP-Only Reporters). GHGRP-Only Reporters accounted for 24 percent of U.S. transmission miles and 26
2 percent of estimated U.S. emissions from electric power system in 2013.58
3 Non-Reporters
4 Emissions from Non-Reporters (i.e., utilities other than Partners and GHGRP-Only Reporters) in every year since
5 1999 were estimated using the results of a regression analysis that correlated emissions from reporting utilities
6 (using verified data from both Partners and GHGRP-Only Reporters) with their transmission miles.59 Two equations
7 were developed, one for "non-large" and one for "large" utilities (i.e., with fewer or greater than 10,000
8 transmission miles, respectively). The distinction between utility sizes was made because the regression analysis
9 showed that the relationship between emissions and transmission miles differed for non-large and large transmission
10 networks. As noted above, non-Partner emissions were reported to the EPA for the first time through its GHGRP in
11 2012 (representing 2011 emissions). This set of reported data was of particular interest because it provided insight
12 into the emission rate of non-Partners, which previously was assumed to be equal to the historical (1999) emission
13 rate of Partners for both large and non-large utilities.60 The availability of non-Partner emissions estimates allowed
14 the regression analysis to be modified for both large and non-large groups. Specifically, emissions were estimated
15 for Non-Reporters as follows:
16 • Non-Reporters. 1999 to 2011: First, the 2011 emission rates (per kg nameplate capacity and per
17 transmission mile) reported by Partners and GHGRP-Only Reporters were reviewed to determine whether
18 there was a statistically significant difference between these two groups. Transmission mileage data for
19 2011 was reported through GHGRP, with the exception of transmission mileage data for Partners that did
20 not report through GHGRP, which was obtained from UDI. It was determined that there is no statistically
21 significant difference between the emission rates of Partners and GHGRP-Only reporters; therefore, Partner
22 and GHGRP-Only reported data for 2011 were combined to develop regression equations to estimate the
23 emissions of Non-Reporters for both "non-large" and "large" utilities. Historical emissions from Non-
24 Reporters for both "non-large" and "large" utilities were estimated by linearly interpolating between the
25 1999 regression coefficients (based on 1999 Partner data) and the 2011 regression coefficients.
26
27 • Non-Reporters. 2012 - Present: It was determined that there continued to be no statistically significant
28 difference between the emission rates reported by Partners and by GHGRP-Only Reporters. Therefore, the
29 emissions data from both groups were combined to develop regression equations for 2012. This was
30 repeated for 2013 using Partner and GHGRP-Only Reporter data for 2013.
31
32 o "Non-large" utilities (less than 10,000 transmission miles): The 2013 regression equation for "non-
33 large" utilities was developed based on the emissions reported by a subset of 89 Partner utilities and
34 GHGRP-Only utilities (representing approximately 47 percent of total U.S. transmission miles for
35 utilities with fewer than 10,000 transmission miles). The regression equation for 2013 is:
36 Emissions (kg) = 0.217 x Transmission Miles
37 o "Large" utilities (more than 10,000 transmission miles): The 2013 regression equation was developed
38 based on the emissions reported by a subset of 17 Partners and GHGRP-only utilities (representing
39 approximately 83 percent of total U.S. transmission miles for utilities with greater than 10,000
40 transmission miles). The regression equation for 2013 is:
Also, GHGRP-reported emissions from 17 facilities that had one or fewer transmission miles were included in the emission
estimates for 2011. Emissions from these facilities comprise approximately 1.2 percent of total reported and verified emissions.
In 2012,16 facilities had one or fewer transmission miles, comprising 1.4 percent of verified emissions and in 2013,16 facilities
had one or fewer transmission miles, comprising 3.2 percent of verified emissions. These facilities were not included in the
development of the regression equations (discussed further below). EPA is continuing to investigate whether or not these
emissions are already implicitly accounted for in the relationship between transmission miles and emissions, and whether to
update the regression analysis to better capture emissions from non-reporters that may have zero transmission miles.
59 In the United States, SFe is contained primarily in transmission equipment rated above 34.5 kV.
Partners in EPA's SFe Emission Reduction Partnership reduced their emissions by approximately 77 percent from 1999 to
2013.
4-106 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Emissions (kg) = 0.225 x Transmission Miles
2 Table 4-102 below shows the percentage of transmission miles covered by reporters (i.e., associated with reported
3 data) and the regression coefficient for both large and non-large reporters for 1999 (the first year data was reported),
4 and for 2011 through 2013 (the first three years with GHGRP reported data). The coefficients for non-large utilities
5 and large utilities both decreased slightly between 2012 and 2013.
6 Table 4-102: Transmission Mile Coverage and Regression Coefficients for Large and Non-
7 Large Utilities, Percent
36
Non-large
Percentage of Miles
Covered by Reporters
Regression Coefficient3
1999
31
0.89
2011
45
0.33
2012
44
0.23
2013
47
0.22
1999
86
0.58
Large
2011
97
0.27
2012
88
0.24
2013
83
0.22
a Regression coefficient is defined as emissions (in kg) divided by transmission miles.
Note: "Non-large" represents reporters with fewer than 10,000 transmission miles.
9 Data on transmission miles for each Non-Reporter for the years 2000, 2003, 2006, and 2009, and 2012 were
10 obtained from the 2001, 2004, 2007, 2010, and 2013 UDI Directories of Electric Power Producers and Distributors,
11 respectively (UDI 2001, 2004, 2007, 2010, 2013). The U.S. transmission system grew by over 25,000 miles
12 between 2000 and 2003 yet declined by almost 4,000 miles between 2003 and 2006. Given these fluctuations,
13 periodic increases are assumed to occur gradually. Therefore, transmission mileage was assumed to increase at an
14 annual rate of 1.2 percent between 2000 and 2003 and decrease by -0.20 percent between 2003 and 2006. This
15 growth rate grew to 3 percent from 2006 to 2009 as transmission miles increased by more than 59,000 miles. The
16 annual growth rate for 2009 through 2012 was calculated to be 2.0 percent as transmission miles grew by
17 approximately 43,000 during this time period.
18 Total Industry Emissions
19 As a final step, total electric power system emissions from 1999 through 2013 were determined for each year by
20 summing the Partner reported and estimated emissions (reported data was available through the EPA's SF6 Emission
21 Reduction Partnership for Electric Power Systems), the GHGRP-Only reported emissions, and the non-reporting
22 utilities' emissions (determined using the regression equations).
23 1990 through 2013 Emissions from Manufacture of Electrical Equipment
24 The 1990 to 2013 emission estimates for original equipment manufacturers (OEMs) were derived by assuming that
25 manufacturing emissions equal 10 percent of the quantity of SF6 provided with new equipment. The quantity of SF6
26 provided with new equipment was estimated based on statistics compiled by the National Electrical Manufacturers
27 Association (NEMA). These statistics were provided for 1990 to 2000; the quantities of SF6 provided with new
28 equipment for 2001 to 2013 were estimated using Partner reported data and the total industry SF6 nameplate
29 capacity estimate (198.2 MMT CO2 Eq. in 2013). Specifically, the ratio of new nameplate capacity to total
30 nameplate capacity of a subset of Partners for which new nameplate capacity data was available from 1999 to 2013
31 was calculated. These ratios were then multiplied by the total industry nameplate capacity estimate for each year to
32 derive the amount of SF6 provided with new equipment for the entire industry. The 10 percent emission rate is the
33 average of the "ideal" and "realistic" manufacturing emission rates (4 percent and 17 percent, respectively)
34 identified in a paper prepared under the auspices of the International Council on Large Electric Systems (CIGRE) in
35 February 2002 (O'Connell et al. 2002).
Uncertainty and Time-Series Consistency
37 To estimate the uncertainty associated with emissions of SF6 from Electrical Transmission and Distribution,
38 uncertainties associated with four quantities were estimated: (1) emissions from Partners, (2) emissions from
39 GHGRP-Only Reporters, (3) emissions from Non-Reporters, and (4) emissions from manufacturers of electrical
40 equipment. A Monte Carlo analysis was then applied to estimate the overall uncertainty of the emissions estimate.
41 Total emissions from the SF6 Emission Reduction Partnership include emissions from both reporting (through the
42 Partnership or GHGRP) and non-reporting Partners. For reporting Partners, individual Partner-reported SF6 data
Industrial Processes and Product Use 4-107
-------
1 was assumed to have an uncertainty of 10 percent. Based on a Monte Carlo analysis, the cumulative uncertainty of
2 all Partner-reported data was estimated to be 2.5 percent. The uncertainty associated with extrapolated or
3 interpolated emissions from non-reporting Partners was assumed to be 20 percent.
4 For GHGRP-Only Reporters, reported SF6 data was assumed to have an uncertainty of 20 percent.61 Based on a
5 Monte Carlo analysis, the cumulative uncertainty of all GHGRP-Only reported data was estimated to be 5.8 percent.
6 There are two sources of uncertainty associated with the regression equations used to estimate emissions in 2013
7 from Non-Reporters: (1) uncertainty in the coefficients (as defined by the regression standard error estimate), and
8 (2) the uncertainty in total transmission miles for Non-Reporters. Uncertainties were also estimated regarding (1)
9 the quantity of SF6 supplied with equipment by equipment manufacturers, which is projected from Partner provided
10 nameplate capacity data and industry SF6 nameplate capacity estimates, and (2) the manufacturers' SF6 emissions
11 rate.
12 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-103. Electrical
13 Transmission and Distribution SF6 emissions were estimated to be between 4.0 and 6.0 MMT CCh Eq. at the 95
14 percent confidence level. This indicates a range of approximately 20 percent below and 19 percent above the
15 emission estimate of 5.1 MMT CO2 Eq.
16 Table 4-103: Approach 2 Quantitative Uncertainty Estimates for SFe Emissions from
17 Electrical Transmission and Distribution (MMT COz Eq. and Percent)
33
2013 Emission
Source Gas Estimate Uncertainty Range Relative to 2013 Emission Estimate3
(MMT C02 Eq.) (MMT CCh Eq.) (%)
1 Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
Electrical Transmission ^ 4Q 6Q _2Q%
and Distribution
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
18 In addition to the uncertainty quantified above, there is uncertainty associated with using global SF6 sales data to
19 estimate U.S. emission trends from 1990 through 1999. However, the trend in global emissions implied by sales of
20 SF6 appears to reflect the trend in global emissions implied by changing SF6 concentrations in the atmosphere. That
21 is, emissions based on global sales declined by 29 percent between 1995 and 1998 (RAND 2004), and emissions
22 based on atmospheric measurements declined by 17 percent over the same period (Levin et al. 2010).
23 Several pieces of evidence indicate that U.S. SF6 emissions were reduced as global emissions were reduced. First,
24 the decreases in sales and emissions coincided with a sharp increase in the price of SF6 that occurred in the mid-
25 1990s and that affected the United States as well as the rest of the world. A representative from DILO, a major
26 manufacturer of SF6 recycling equipment, stated that most U.S. utilities began recycling rather than venting SF6
27 within two years of the price rise. Finally, the emissions reported by the one U.S. utility that reported its emissions
28 for all the years from 1990 through 1999 under the Partnership showed a downward trend beginning in the mid-
29 1990s.
30 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
31 through 2013. Details on the emission trends through time are described in more detail in the Methodology section,
32 above.
Recalculations Discussion
34 For the current Inventory, emission estimates have been revised to reflect the GWPs provided in the IPCC Fourth
35 Assessment Report (AR4) (IPCC 2007). AR4 GWP values differ slightly from those presented in the IPCC Second
36 Assessment Report (SAR) (IPCC 1996) (used in the previous inventories) which results in time-series recalculations
Uncertainty is assumed to be higher for the GHGRP-Only category, because 2011 is the first year that those utilities have
reported to EPA.
4-108 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 for most inventory sources. Under the most recent reporting guidelines (UNFCCC 2014), countries are required to
2 report using the AR4 GWPs, which reflect an updated understanding of the atmospheric properties of each
3 greenhouse gas. The GWPs of CH4 and most fluorinated greenhouse gases have increased, leading to an overall
4 increase in CCh Eq. emissions from CH4, HFCs, and PFCs. The GWPs of N2O and SF6 have decreased, leading to a
5 decrease in CCh Eq. emissions for SF6. The AR4 GWPs have been applied across the entire time series for
6 consistency. For more information please see the Recalculations and Improvements Chapter.
7 Only taking this change into consideration, emissions estimates for each year from 1990 to 2012 would have slightly
8 decreased, relative to the emissions estimates in the previous Inventory report. However, other changes to the
9 historical calculations, as noted below, resulted in emission estimates fluctuating slightly (increasing for some years
10 and decreasing for other years) across the time series.
11 The historical emissions estimated for this source category have undergone several minor revisions. SF6 emission
12 estimates for the period 1990 through 2012 were updated relative to the previous report based on revisions to
13 interpolated and extrapolated non-reported Partner data as well as resubmissions of estimates through the GHGRP
14 for 2011 and2012.62 The previously-described interpolation between 1999 and 2012 regression coefficients to
15 estimate emissions from non-reporting utilities were updated using revised GHGRP reports, which impacted
16 historical estimates for the period 2000 through 2012. Additionally, updated leak rates were calculated from
17 resubmitted Partner data through the GHGRP. These leak rates are used to estimate the nameplate capacity of non-
18 reporters during these years, and are interpolated back through 1999 to calculate Non-Reporter nameplate capacity
19 over the entire time series.63 Finally, revisions were made regarding the incorporation of transmission mile data
20 from the UDI database to remove instances of double counting transmission miles between parent and subsidiary
21 companies. Reductions in the total transmission miles reduced the total number of non-reporter transmission miles,
22 which reduced non-reporter emissions, and therefore total emissions.
23 As a result of the recalculations, SF6 emissions from electrical transmission and distribution decreased by 6 percent
24 for 2012 relative to the previous report. On average, the change in SF6 emission estimates for the entire time series is
25 approximately 0.5 percent per year.
26
35
Planned Improvements
27 EPA is exploring the use of OEM data that is reported under EPA's GHGRP to use for future Inventory reports
28 instead of estimating those emissions based on elements reported by utilities to the GHGRP and Partner data.
29 Specifically, using the GHGRP-reported OEM emissions and the estimated nameplate capacity increase estimated
30 for users of electrical equipment (available in the existing methodology), a leak rate would be calculated. This
31 approach would require estimating the portion of industry not reporting to the GHGRP program, which would
32 require market research. Once a new leak rate is established, leak rates could be interpolated for years between 2000
33 (at 10 percent) and 2011. In implementing improvements and integration of data from EPA's GHGRP, the latest
34 guidance from the IPCC on the use of facility-level data in national inventories will be relied upon.64
36 Emissions of HFCs, PFCs, SF6, and NF3 from industrial processes can be estimated in two ways, either as potential
37 emissions or as actual emissions. Emission estimates in this chapter are "actual emissions," which are defined by
38 the 2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2006) as estimates that take into account
39 the time lag between consumption and emissions. In contrast, "potential emissions" are defined to be equal to the
40 amount of a chemical consumed in a country, minus the amount of a chemical recovered for destruction or export in
41 the year of consideration. Potential emissions will generally be greater for a given year than actual emissions, since
62 The earlier year estimates within the time series (i.e., 1990-1998) were updated based on revisions to the 1999 U.S. emission
estimate because emissions for 1990-1998 are estimated by multiplying a series of annual factors by the estimated U.S. emissions
of SFe from electric power systems in 1999 (see Methodology section).
63 Nameplate capacity estimates affect sector emissions because OEM emission estimation is calculated using total industry
nameplate capacity.
64 See.
Industrial Processes and Product Use 4-109
-------
1 some amount of chemical consumed will be stored in products or equipment and will not be emitted to the
2 atmosphere until a later date, if ever. Although actual emissions are considered to be the more accurate estimation
3 approach for a single year, estimates of potential emissions are provided for informational purposes.
4 Separate estimates of potential emissions were not made for industrial processes that fall into the following
5 categories:
6 • Byproduct emissions. Some emissions do not result from the consumption or use of a chemical, but are the
7 unintended byproducts of a process. For such emissions, which include emissions of CF4 and C2p6 from
8 aluminum production and of HFC-23 from HCFC-22 production, the distinction between potential and
9 actual emissions is not relevant.
10 • Potential emissions that equal actual emissions. For some sources, such as magnesium production and
11 processing, no delay between consumption and emission is assumed and, consequently, no destruction of
12 the chemical takes place. In this case, actual emissions equal potential emissions.
13 Table 4-104 presents potential emission estimates for HFCs and PFCs from the substitution of ozone depleting
14 substances, HFCs, PFCs, SF6, and NF3 from semiconductor manufacture, and SF6 from magnesium production and
15 processing and electrical transmission and distribution.65 Potential emissions associated with the substitution for
16 ozone depleting substances were calculated using the EPA's Vintaging Model. Estimates of HFCs, PFCs, and SF6
17 consumed by semiconductor manufacture were developed by dividing chemical-by-chemical emissions by the
18 appropriate chemical-specific emission factors from the 2006IPCC Guidelines (Tier 2c). Estimates of CF4
19 consumption were adjusted to account for the conversion of other chemicals into CF4 during the semiconductor
20 manufacturing process, again using the default factors from the 2006 IPCC Guidelines. Potential SF6 emissions
21 estimates for electrical transmission and distribution were developed using U.S. utility purchases of SF6 for
22 electrical equipment. From 1999 through 2013, estimates were obtained from reports submitted by participants in
23 EPA's SF6 Emission Reduction Partnership for Electric Power Systems as well as EPA's Greenhouse Gas Reporting
24 Program (GHGRP). U.S. utility purchases of SF6 for electrical equipment from 1990 through 1998 were backcasted
25 based on world sales of SF6 to utilities. Purchases of SF6 by utilities were added to SF6 purchases by electrical
26 equipment manufacturers to obtain total SF6 purchases by the electrical equipment sector.
27 Table 4-104: 2013 Potential and Actual Emissions of HFCs, PFCs, SFe, and NFs from Selected
28 Sources (MMT COz Eq.)
Source
Substitution of Ozone Depleting Substances
Aluminum Production
HCFC-22 Production
Semiconductor Manufacture
Magnesium Production and Processing
Electrical Transmission and Distribution
NA - Not applicable.
Potential
306.9
NA
NA
43.7
1.5
33.3
Actual
158.6
3.0
4.1
4.0
1.5
5.1
29
30 Under EPA's GHGRP, producers and larger importers and exporters66 of fluorinated greenhouse gases (F-GHG) in
31 bulk began annually reporting their production, destruction, imports, and exports in 2011 (for 2010 supplies), and
32 larger importers and exporters of F-GHGs inside of pre-charged equipment began reporting their imports and
33 exports in2012 (for 2011 supplies). The collection of data from both emitters and suppliers of F-GHGs enables the
34 comparison of consumption that is implied by emissions (downstream estimation method) to the consumption that is
35 implied by balancing of production, destruction, imports, and exports (upstream estimation method). This type of
36 comparison ultimately supports and improves estimates of emissions, as noted in the 2006 IPCC Guidelines:
37 "[W]hen considered along with estimates of actual emissions, the potential emissions approach can assist
38 in validation of completeness of sources covered and as a QC check by comparing total domestic
65 See Annex 5 for a discussion of sources of SFe emissions excluded from the actual emissions estimates in this report.
Importers and exporters report only if either their total imports or their total exports of F-GHGs are greater than or equal to
25,000 metric tons of CCh Eq. per year
4-110 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 consumption as calculated in this 'potential emissions approach' per compound with the sum of all
2 activity data of the various uses (IPCC 2006)."
13
14
15
3 A comparison of upstream and downstream consumption estimates of SF6 was performed to help evaluate the
4 accuracy and completeness of the emissions inventory. This analysis revealed that the two potential emissions
5 estimates for 2012 (the upstream estimation and downstream estimation methods) differed with the supply-based,
6 upstream consumption estimate significantly larger than emitter-based, downstream consumption estimate (Ottinger
7 et al. 2014). This finding indicates that methods for determining national SF6 actual emission estimates by industry
8 sector are generating results that, when summed, do not fall within a close proximity to the overall total U.S. supply
9 ofSFegas.
10 While multiple sources of uncertainty affect both data sets, Ottinger et al. (2014) conclude that current SF6 emission
11 estimates likely do not account for all significant sources of SF6 in the United States. Additional research is
12 necessary to identify the other significant applications that consume and emit SF6
4.25 Nitrous Oxide from Product Uses (IPCC
Source Category 2G3)
16 Nitrous oxide (N2O) is a clear, colorless, oxidizing liquefied gas, with a slightly sweet odor which is used in a wide
17 variety of specialized product uses and applications. The amount of N2O that is actually emitted depends upon the
18 specific product use or application.
19 There are a total of three N2O production facilities currently operating in the United States (Ottinger 2014). N2O is
20 primarily used in carrier gases with oxygen to administer more potent inhalation anesthetics for general anesthesia,
21 and as an anesthetic in various dental and veterinary applications. The second main use of N2O is as a propellant in
22 pressure and aerosol products, the largest application being pressure-packaged whipped cream. Small quantities of
23 N2O also are used in the following applications:
24 • Oxidizing agent and etchant used in semiconductor manufacturing;
25 • Oxidizing agent used, with acetylene, in atomic absorption spectrometry;
26 • Production of sodium azide, which is used to inflate airbags;
27 • Fuel oxidant in auto racing; and
28 • Oxidizing agent in blowtorches used by jewelers and others (Heydorn 1997).
29 Production of N2O in 2014 was approximately 15 kt (Table 4-105).
30 Table 4-105: N2O Production (kt)
Year kt
1990
2010 15
2011 15
2012 15
2013 15
2014 15
Industrial Processes and Product Use 4-111
-------
1 Nitrous oxide emissions were 4.2 MMT CO2 Eq. (14 kt N2O) in 2014 (Table 4-106). Production of N2O stabilized
2 during the 1990s because medical markets had found other substitutes for anesthetics, and more medical procedures
3 were being performed on an outpatient basis using local anesthetics that do not require N2O. The use of N2O as a
4 propellant for whipped cream has also stabilized due to the increased popularity of cream products packaged in
5 reusable plastic tubs (Heydorn 1997).
6 Table 4-106: NzO Emissions from NzO Product Usage (MMT COz Eq. and kt)
Year MMT CCh Eq. kt
1990 4.2
2010
2011
2012
2013
2014
4.2
4.2
4.2
4.2
4.2
14
14
14
14
14
7 Methodology
8 Emissions from N2O product uses were estimated using the following equation:
10 where,
11 Epu = N2O emissions from product uses, metric tons
12 P Total U.S. production of N2O, metric tons
13 a = specific application
14 Sa Share of N2O usage by application a
15 ERa = Emission rate for application a, percent
16 The share of total quantity of N2O usage by end use represents the share of national N2O produced that is used by
17 the specific subcategory (i.e., anesthesia, food processing, etc.). In 2014, the medical/dental industry used an
18 estimated 86.5 percent of total N2O produced, followed by food processing propellants at 6.5 percent. All other
19 categories combined used the remainder of the N2O produced. This subcategory breakdown has changed only
20 slightly over the past decade. For instance, the small share of N2O usage in the production of sodium azide has
21 declined significantly during the 1990s. Due to the lack of information on the specific time period of the phase -out
22 in this market subcategory, most of the N2O usage for sodium azide production is assumed to have ceased after
23 1996, with the majority of its small share of the market assigned to the larger medical/dental consumption
24 subcategory (Heydorn 1997). The N2O was allocated across the following categories: medical applications, food
25 processing propellant, and sodium azide production (pre-1996). A usage emissions rate was then applied for each
26 sector to estimate the amount of N2O emitted.
27 Only the medical/dental and food propellant subcategories were estimated to release emissions into the atmosphere,
28 and therefore these subcategories were the only usage subcategories with emission rates. For the medical/dental
29 subcategory, due to the poor solubility of N2O in blood and other tissues, none of the N2O is assumed to be
30 metabolized during anesthesia and quickly leaves the body in exhaled breath. Therefore, an emission factor of 100
31 percent was used for this subcategory (IPCC 2006). For N2O used as a propellant in pressurized and aerosol food
32 products, none of the N2O is reacted during the process and all of the N2O is emitted to the atmosphere, resulting in
33 an emission factor of 100 percent for this subcategory (IPCC 2006). For the remaining subcategories, all of the N2O
34 is consumed/reacted during the process, and therefore the emission rate was considered to be zero percent (Tupman
35 2002).
36 The 1990 through 1992 N2O production data were obtained from SRI Consulting's Nitrous Oxide, North America
37 report (Heydorn 1997). Nitrous oxide production data for 1993 through 1995 were not available. Production data
4-112 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 for 1996 was specified as a range in two data sources (Heydorn 1997, Tupman 2002). In particular, for 1996,
2 Heydorn (1997) estimates N2O production to range between 13.6 and 18.1 thousand metric tons. Tupman (2003)
3 provided a narrower range (15.9 to 18.1 thousand metric tons) for 1996 that falls within the production bounds
4 described by Heydorn (1997). Tupman (2003) data are considered more industry-specific and current. Therefore,
5 the midpoint of the narrower production range was used to estimate N2O emissions for years 1993 through 2001
6 (Tupman 2003). The 2002 and 2003 N2O production data were obtained from the Compressed Gas Association
7 Nitrous Oxide Fact Sheet and Nitrous Oxide Abuse Hotline (CGA 2002, 2003). These data were also provided as a
8 range. For example, in 2003, CGA (2003) estimates N2O production to range between 13.6 and 15.9 thousand
9 metric tons. Due to the unavailability of data, production estimates for years 2004 through 2014 were held constant
10 at the 2003 value.
11 The 1996 share of the total quantity of N2O used by each subcategory was obtained from SRI Consulting's Nitrous
12 Oxide, North America report (Heydorn 1997). The 1990 through 1995 share of total quantity of N2O used by each
13 subcategory was kept the same as the 1996 number provided by SRI Consulting. The 1997 through 2001 share of
14 total quantity of N2O usage by sector was obtained from communication with a N2O industry expert (Tupman 2002).
15 The 2002 and 2003 share of total quantity of N2O usage by sector was obtained from CGA (2002, 2003). Due to the
16 unavailability of data, the share of total quantity of N2O usage data for years 2004 through 2014 was assumed to
17 equal the 2003 value. The emissions rate for the food processing propellant industry was obtained from SRI
18 Consulting's Nitrous Oxide, North America report (Heydorn 1997), and confirmed by a N2O industry expert
19 (Tupman 2002). The emissions rate for all other subcategories was obtained from communication with a N2O
20 industry expert (Tupman 2002). The emissions rate for the medical/dental subcategory was obtained from the 2006
21 IPCC Guidelines.
22 Uncertainty and Time-Series Consistency
23 The overall uncertainty associated with the 2014 N2O emission estimate from N2O product usage was calculated
24 using the 2006 IPCC Guidelines (2006) Approach 2 methodology. Uncertainty associated with the parameters used
25 to estimate N2O emissions include production data, total market share of each end use, and the emission factors
26 applied to each end use, respectively.
27 The results of this Approach 2 quantitative uncertainty analysis are summarized in Table 4-107. N2O emissions
28 from N2O product usage were estimated to be between 3.2 and 5.2 MMT CO2 Eq. at the 95 percent confidence level.
29 This indicates a range of approximately 24 percent below to 24 percent above the emission estimate of 4.2 MMT
30 CO2 Eq.
31 Table 4-107: Approach 2 Quantitative Uncertainty Estimates for NzO Emissions from NzO
32 Product Usage (MMT COz Eq. and Percent)
Source Gas 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
(MMT CCh Eq.) (MMT CCh Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
N2Q Product Use N2O 4.2 3.2 5.2 -24% +24%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
33 Methodological recalculations were applied to the entire time series to ensure consistency in emissions from 1990
34 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
35 above
36
Planned Improvements
37 Pending resources, planned improvements include a continued evaluation of alternative production statistics for
38 cross verification, a reassessment of N2O product use subcategories to accurately represent trends, investigation of
39 production and use cycles, and the potential need to incorporate a time lag between production and ultimate product
40 use and resulting release of N2O. Additionally, planned improvements include considering imports and exports of
41 N2O for product uses. Finally, for future Inventories EPA will examine data from EPA's GHGRP to improve the
Industrial Processes and Product Use 4-113
-------
1 emission estimates for the N2O product use subcategory. Particular attention will be made to ensure aggregated
2 information can be published without disclosing confidential business information and time series consistency, as
3 the facility-level reporting data from EPA's GHGRP are not available for all inventory years as required in this
4 Inventory.
5 4.26 Industrial Processes and Product Use
6 Sources of Indirect Greenhouse Gases
7 In addition to the main greenhouse gases addressed above, many industrial processes can result in emissions of
8 various ozone precursors (i.e., indirect greenhouse gases). As some of industrial applications also employ thermal
9 incineration as a control technology, combustion byproducts, such as carbon monoxide (CO) and nitrogen oxides
10 (NOX), are also reported with this source category. Non-CH4 volatile organic compounds (NMVOCs), commonly
11 referred to as "hydrocarbons," are the primary gases emitted from most processes employing organic or petroleum
12 based products, and can also result from the product storage and handling. Accidental releases of greenhouse gases
13 associated with product use and handling can constitute major emissions in this category. In the United States,
14 emissions from product use are primarily the result of solvent evaporation, whereby the lighter hydrocarbon
15 molecules in the solvents escape into the atmosphere. The major categories of product uses include: degreasing,
16 graphic arts, surface coating, other industrial uses of solvents (e.g., electronics), dry cleaning, and non-industrial
17 uses (e.g., uses of paint thinner). Product usage in the United States also results in the emission of small amounts of
18 hydrofluorocarbons (HFCs) and hydrofluoroethers (HFEs), which are included under Substitution of Ozone
19 Depleting Substances in this chapter.
20 Total emissions of nitrogen oxides (NOX), carbon monoxide (CO), and non-CH4 volatile organic compounds
21 (NMVOCs) from non-energy industrial processes and product use from 1990 to 2014 are reported in Table 4-108.
22 Table 4-108: NOX, CO, and NMVOC Emissions from Industrial Processes and Product Use (kt)
Gas/Source
NOx
Industrial Processes
Other Industrial
Processes
Metals Processing
Chemical and Allied
Product Manufacturing
Storage and Transport
Miscellaneous*
Product Use
Surface Coating
Graphic Arts
Degreasing
Dry Cleaning
Other Industrial
Processes'5
Non-Industrial Processes0
Other
CO
Industrial Processes
Metals Processing
Other Industrial
Processes
Chemical and Allied
Product Manufacturing
Miscellaneous*
Storage and Transport
1990
592
343 1
88
152 1
3
5
1
+ 1
+ 1
+ 1
+ 1
+ 1
NA 1
4,129
2,395
487 1
1,073
101 1
•
2005 2010
572 472
437
60
55
15
2
3
0
0
0
0
0
0
1,557
752
484
189
32
97
339
67
48
15
2
2
0
0
0
0
0
0
1,280
717
333
157
48
22
2011
452
320
64
47
18
3
1
0
0
0
0
0
0
1,229
695
306
152
51
25
2012
452
320
64
47
18
3
1
0
0
0
0
0
0
1,229
695
306
152
51
25
2013
452
320
64
47
18
3
1
0
0
0
0
0
0
1,229
695
306
152
51
25
2014
452
320
64
47
18
3
1
0
0
0
0
0
0
1,229
695
306
152
51
24
4-114 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
NA
7,638
1,352
364
575
111
20
0
0
0
0
0
0
5,849
1,308
414
213
45
17
0
0
0
0
0
0
4,133
992
308
77
32
26
0
0
0
0
0
0
3,929
947
298
76
31
27
0
0
0
0
0
0
3,929
947
298
76
31
27
0
0
0
0
0
0
3,929
947
298
76
31
27
0
0
0
0
0
0
3,928
946
298
75
31
27
Product Use
Surface Coating
Other Industrial
Processes'5
Dry Cleaning
Degreasing
Graphic Arts
Non-Industrial Processes0
Other
NMVOCs
Industrial Processes
Storage and Transport
Other Industrial
Processes
Chemical and Allied
Product Manufacturing
Metals Processing
Miscellaneous*
Product Use
Surface Coating
Non-Industrial Processes0
Degreasing
Dry Cleaning
Graphic Arts
Other Industrial
Processes'5
Other
a Miscellaneous includes the following categories: catastrophic/accidental release, other combustion, health
services, cooling towers, and fugitive dust. It does not include agricultural fires or slash/prescribed
burning, which are accounted for under the Field Burning of Agricultural Residues source.
b Includes rubber and plastics manufacturing, and other miscellaneous applications.
0 Includes cutback asphalt, pesticide application adhesives, consumer solvents, and other miscellaneous
applications.
NA (Not Available)
+ Does not exceed 0.5 kt
Note: Totals may not sum due to independent rounding.
Methodology
2 Emission estimates for 1990 through 2014 were obtained from data published on the National Emission Inventory
3 (NEI) Air Pollutant Emission Trends web site (EPA 2015), and disaggregated based on EPA (2003). Data were
4 collected for emissions of carbon monoxide (CO), nitrogen oxides (NOX), volatile organic compounds (VOCs), and
5 sulfur dioxide (SCh) from metals processing, chemical manufacturing, other industrial processes, transport and
6 storage, and miscellaneous sources. Emission estimates for 2013 for non-EGU and non-mobile sources are held
7 constant from 2011 in EPA (2015). Emissions were calculated either for individual source categories or for many
8 categories combined, using basic activity data (e.g., the amount of raw material processed or the amount of solvent
9 purchased) as an indicator of emissions. National activity data were collected for individual categories from various
10 agencies. Depending on the category, these basic activity data may include data on production, fuel deliveries, raw
11 material processed, etc.
12 Emissions for product use were calculated by aggregating product use data based on information relating to product
13 uses from different applications such as degreasing, graphic arts, etc. Emission factors for each consumption
14 category were then applied to the data to estimate emissions. For example, emissions from surface coatings were
15 mostly due to solvent evaporation as the coatings solidify. By applying the appropriate product-specific emission
16 factors to the amount of products used for surface coatings, an estimate of NMVOC emissions was obtained.
17 Emissions of CO and NOX under product use result primarily from thermal and catalytic incineration of solvent-
18 laden gas streams from painting booths, printing operations, and oven exhaust.
Industrial Processes and Product Use 4-115
-------
1 Activity data were used in conjunction with emission factors, which together relate the quantity of emissions to the
2 activity. Emission factors are generally available from the EPA's Compilation of Air Pollutant Emission Factors,
3 AP-42 (EPA 1997). The EPA currently derives the overall emission control efficiency of a source category from a
4 variety of information sources, including published reports, the 1985 National Acid Precipitation and Assessment
5 Program emissions inventory, and other EPA databases.
6 Uncertainty and Time-Series Consistency
7 Uncertainties in these estimates are partly due to the accuracy of the emission factors and activity data used. A
8 quantitative uncertainty analysis was not performed.
9 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
10 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
11 above.
12
4-116 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
5. Agriculture
2 Agricultural activities contribute directly to emissions of greenhouse gases through a variety of processes. This
3 chapter provides an assessment of non-carbon-dioxide emissions from the following source categories: enteric
4 fermentation in domestic livestock, livestock manure management, rice cultivation, agricultural soil management,
5 and field burning of agricultural residues (see Figure 5-1). Carbon dioxide (CO2) emissions and removals from
6 agriculture-related land-use activities, such as liming and conversion of grassland to cultivated land, are presented in
7 the Land Use, Land-Use Change, and Forestry chapter. Carbon dioxide emissions from on-farm energy use are
8 accounted for in the Energy chapter.
10
11
12
13
14
15
16
17
18
19
20
21
22
Figure 5-1: 2014 Agriculture Chapter Greenhouse Gas Emission Sources (MMT COz Eq.)
Agricultural Soil Management
Enteric Fermentation
Manure Management
Rice Cultivation
Field Burning of Agricultural Residues
Agriculture as a Portion
of all Emissions
25 50
75 100 125 150 175 200 225 250 275 300
MMT CO2 Eq.
In 2014, the Agriculture sector was responsible for emissions of 574.1 MMT CO2 Eq.,1 or 8.4 percent of total U.S.
greenhouse gas emissions. Methane (CH4) and nitrous oxide (N2O) were the primary greenhouse gases emitted by
agricultural activities. Methane emissions from enteric fermentation and manure management represent 23.2 percent
and 8.6 percent of total CH4 emissions from anthropogenic activities, respectively. Of all domestic animal types,
beef and dairy cattle were by far the largest emitters of CH4. Rice cultivation and field burning of agricultural
residues were minor sources of CH4. Agricultural soil management activities such as fertilizer application and other
cropping practices were the largest source of U.S. N2O emissions, accounting for 77.4 percent. Manure management
and field burning of agricultural residues were also small sources of N2O emissions.
Table 5-1 and Table 5-2 present emission estimates for the Agriculture sector. Between 1990 and 2014, CH4
emissions from agricultural activities increased by 11.8 percent, while N2O emissions fluctuated from year to year,
but overall increased by 6.0 percent.
1 Following the revised reporting requirements under the UNFCCC, this Inventory report presents CCh equivalent values based
on the IPCC Fourth Assessment Report (AR4) GWP values. See the Introduction chapter for more information.
Agriculture 5-1
-------
Table 5-1: Emissions from Agriculture (MMT COz Eq.)
Gas/Source
CH4
Enteric Fermentation
Manure Management
Rice Cultivation
Field Burning of Agricultural Residues
N2O
Agricultural Soil Management
Manure Management
Field Burning of Agricultural Residues
Total
1990
212.8
164.2
37.2 1
11.3
14.0 1
0.1 •
529.8
2005
239.6
168.9
56.3 1
14.2 1
0.2
313.3
296.7
16.5 1
0.1 •
552.9
2010
244.7
171.3
60.9
12.2
0.3
337.7
320.4
17.2
0.1
582.3
2011
242.9
168.9
61.5
12.2
0.3
340.4
322.9
17.4
0.1
583.3
2012
242.9
166.7
63.7
12.2
0.3
340.5
322.9
17.5
0.1
583.4
2013
239.3
165.5
61.4
12.2
0.3
336.0
318.4
17.5
0.1
575.4
2014
238.0
164.3
61.2
12.2
0.3
336.1
318.5
17.5
0.1
574.1
Note: Totals may not sum due to independent rounding.
Table 5-2: Emissions from Agriculture (kt)
Gas/Source
CH4
Enteric Fermentation
Manure Management
Rice Cultivation
Field Burning of Agricultural Residues
N2O
Agricultural Soil Management
Manure Management
Field Burning of Agricultural Residues
1990
8,513
6,566
1,486
451
10
1,064
1,017
47
+
2005
9,584
6,755
2,254
567
1
1,051
996
55 1
+
2010
9,788
6,853
2,437
486
11
1,133
1,075
58
+
2011
9,715
6,757
2,460
487
11
1,142
1,084
58
+
2012
9,717
6,670
2,548
488
11
1,143
1,083
59
+
2013
9,573
6,619
2,455
488
11
1,128
1,069
59
+
2014
9,518
6,572
2,447
488
11
1,128
1,069
59
+
+ Does not exceed 0.5 kt.
Note: Totals may not sum due to independent rounding.
9
10
11
12
13
14
15
16
17
18
19
20
5.1 Enteric Fermentation (IPCC Source
Category 3A)
Methane is produced as part of normal digestive processes in animals. During digestion, microbes resident in an
animal's digestive system ferment food consumed by the animal. This microbial fermentation process, referred to as
enteric fermentation, produces CH4 as a byproduct, which can be exhaled or eructated by the animal. The amount of
CH4 produced and emitted by an individual animal depends primarily upon 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 system. Ruminants possess a rumen, or large "fore-stomach," in which microbial fermentation
breaks down the feed they consume into products that can be absorbed and metabolized. The microbial
fermentation that occurs in the rumen enables them to digest coarse plant material that non-ruminant animals cannot.
Ruminant animals, consequently, have the highest CH4 emissions per unit of body mass among all animal types.
Non-ruminant animals (e.g., swine, horses, and mules and asses) also produce CH4 emissions through enteric
fermentation, although this microbial fermentation occurs in the large intestine. These non-ruminants emit
significantly less CH4 on a per-animal-mass basis than ruminants because the capacity of the large intestine to
produce CH4 is lower.
5-2 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 In addition to the type of digestive system, an animal's feed quality and feed intake also affect CH4 emissions. In
2 general, lower feed quality and/or higher feed intake leads to higher CH4 emissions. Feed intake is positively
3 correlated to animal size, growth rate, level of activity and production (e.g., milk production, wool growth,
4 pregnancy, or work). Therefore, feed intake varies among animal types as well as among different management
5 practices for individual animal types (e.g., animals in feedlots or grazing on pasture).
6 Methane emission estimates from enteric fermentation are provided in Table 5-3 and Table 5-4. Total livestock CH4
7 emissions in 2014 were 164.3 MMT CChEq. (6,572 kt). Beef cattle remain the largest contributor of CH4 emissions
8 from enteric fermentation, accounting for 71 percent in 2014. Emissions from dairy cattle in 2014 accounted for 26
9 percent, and the remaining emissions were from horses, sheep, swine, goats, American bison, mules and asses.
10 From 1990 to 2014, emissions from enteric fermentation have increased by 0.08 percent. While emissions generally
11 follow trends in cattle populations, over the long term there are exceptions as population decreases have been
12 coupled with production increases or minor decreases. For example, beef cattle emissions decreased 2.0 percent
13 from 1990 to 2014, while beef cattle populations actually declined by 7 percent and beef production increased
14 (USD A 2015), and while dairy emissions increased 6.5 percent over the entire time series, the population has
15 declined by 5 percent and milk production increased 40 percent (USDA 2015). This trend indicates that while
16 emission factors per head are increasing, emission factors per unit of product are going down. Generally, from 1990
17 to 1995 emissions from beef increased and then decreased from 1996 to 2004. These trends were mainly due to
18 fluctuations in beef cattle populations and increased digestibility of feed for feedlot cattle. Beef cattle emissions
19 generally increased from 2004 to 2007, as beef populations underwent increases and an extensive literature review
20 indicated a trend toward a decrease in feed digestibility for those years. Beef cattle emissions decreased again from
21 2008 to 2014 as populations again decreased. Emissions from dairy cattle generally trended downward from 1990 to
22 2004, along with an overall dairy population decline during the same period. Similar to beef cattle, dairy cattle
23 emissions rose from 2004 to 2007 due to population increases and a decrease in feed digestibility (based on an
24 analysis of more than 350 dairy cow diets). Dairy cattle emissions have continued to trend upward since 2007, in
25 line with dairy population increases. Regarding trends in other animals populations of sheep have steadily declined,
26 with an overall decrease of 54 percent since 1990. Horse populations are 56 percent greater than they were in 1990,
27 but their numbers have been declining by about 2 percent annually since 2007. Goat populations increased by about
28 20 percent through 2007 but have since dropped below 1990 numbers, while swine populations have increased 19
29 percent since 1990. The population of American bison more than tripled over the 1990 through 2014 time period,
30 while mules and asses have more than quadrupled.
31 Table 5-3: ChU Emissions from Enteric Fermentation (MMT COz Eq.)
32
Livestock Type
Beef Cattle
Dairy Cattle
Horses
Sheep
Swine
Goats
American Bison
Mules and Asses
Total
Note: Totals ma>
1990
119.1
39.4
2.0
1.0
2.3
0.3
0.1
+
164.2
' not sum due to
2005
125.2
37.6
2.3
1.7
1.2
0.4
0.4
0.1
168.9
2010
124.6
40.7
2.4
1.7
1.1
0.4
0.4
0.1
171.3
2011
121.8
41.1
2.5
1.7
1.1
0.3
0.3
0.1
168.9
2012
119.1
41.7
2.5
1.6
1.1
0.3
0.3
0.1
166.7
2013
118.0
41.6
2.5
1.6
1.1
0.3
0.3
0.1
165.5
2014
116.7
41.9
2.4
1.6
1.0
0.3
0.3
0.1
164.3
independent rounding.
+ Does not exceed 0.05 MMT CO2 Eq.
Table 5-4: CH4
Livestock Type
Beef Cattle
Dairy Cattle
Swine
Horses
Sheep
Goats
Emissions from
1990
4,763
1,574
81
40
91
13
Enteric Fermentation (kt)
2005
5,007
1,503
92
70
49
14
2010
4,984
1,627
97
68
45
14
2011
4,873
1,645
98
67
44
14
2012
4,763
1,670
100
65
43
13
2013
4,722
1,664
98
64
43
13
2014
4,667
1,677
96
62
42
12
Agriculture 5-3
-------
American Bison 4 I 17 15 14 13 13 12
Mules and Asses 1 2 3 3 3 3 *3_
Total 6,566 6,755 6,853 6,757 6,670 6,619 6,572
Note: Totals may not sum due to independent rounding.
2 Methodology
3 Livestock enteric fermentation emission estimate methodologies fall into two categories: cattle and other
4 domesticated animals. Cattle, due to their large population, large size, and particular digestive characteristics,
5 account for the majority of enteric fermentation CH4 emissions from livestock in the United States. A more detailed
6 methodology (i.e., IPCC Tier 2) was therefore applied to estimate emissions for all cattle. Emission estimates for
7 other domesticated animals (horses, sheep, swine, goats, American bison, and mules and asses) were handled using a
8 less detailed approach (i.e., IPCC Tier 1).
9 While the large diversity of animal management practices cannot be precisely characterized and evaluated,
10 significant scientific literature exists that provides the necessary data to estimate cattle emissions using the IPCC
11 Tier 2 approach. The Cattle Enteric Fermentation Model (CEFM), developed by EPA and used to estimate cattle
12 CH4 emissions from enteric fermentation, incorporates this information and other analyses of livestock population,
13 feeding practices, and production characteristics.
14 National cattle population statistics were disaggregated into the following cattle sub-populations:
15 • Dairy Cattle
16 o Calves
17 o Heifer Replacements
18 o Cows
19 • Beef Cattle
20 o Calves
21 o Heifer Replacements
22 o Heifer and Steer Stackers
23 o Animals in Feedlots (Heifers and Steer)
24 o Cows
25 o Bulls
26
27 Calf birth rates, end-of-year population statistics, detailed feedlot placement information, and slaughter weight data
28 were used to create a transition matrix that models cohorts of individual animal types and their specific emission
29 profiles. The key variables tracked for each of the cattle population categories are described in Annex 3.10. These
30 variables include performance factors such as pregnancy and lactation as well as average weights and weight gain.
31 Annual cattle population data were obtained from the U.S. Department of Agriculture's (USDA) National
32 Agricultural Statistics Service (NASS) QuickStats database (USDA 2015).
33 Diet characteristics were estimated by region for dairy, foraging beef, and feedlot beef cattle. These diet
34 characteristics were used to calculate digestible energy (DE) values (expressed as the percent of gross energy intake
35 digested by the animal) and CH4 conversion rates (Ym) (expressed as the fraction of gross energy converted to CH4)
36 for each regional population category. The IPCC recommends Ym ranges of 3.0±1.0 percent for feedlot cattle and
37 6.5±1.0 percent for other well-fed cattle consuming temperate-climate feed types (IPCC 2006). Given the
38 availability of detailed diet information for different regions and animal types in the United States, DE and Ym
39 values unique to the United States were developed. The diet characterizations and estimation of DE and Ym values
40 were based on information from state agricultural extension specialists, a review of published forage quality studies
41 and scientific literature, expert opinion, and modeling of animal physiology.
42 The diet characteristics for dairy cattle were based on Donovan (1999) and an extensive review of nearly 20 years of
43 literature from 1990 through 2009. Estimates of DE were national averages based on the feed components of the
44 diets observed in the literature for the following year groupings: 1990 through 1993, 1994 through 1998, 1999
5-4 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 through 2003, 2004 through 2006, 2007, and 2008 onward.2 Base year Ym values by region were estimated using
2 Donovan (1999). A ruminant digestion model (COWPOLL, as selected in Kebreab et al. 2008) was used to evaluate
3 Ym for each diet evaluated from the literature, and a function was developed to adjust regional values over time
4 based on the national trend. Dairy replacement heifer diet assumptions were based on the observed relationship in
5 the literature between dairy cow and dairy heifer diet characteristics.
6 For feedlot animals, the DE and Ym values used for 1990 were recommended by Johnson (1999). Values for DE
7 and Ym for 1991 through 1999 were linearly extrapolated based on the 1990 and 2000 data. DE and Ym values for
8 2000 onwards were based on survey data in Galyean and Gleghorn (2001) and Vasconcelos and Galyean (2007).
9 For grazing beef cattle, Ym values were based on Johnson (2002), DE values for 1990 through 2006 were based on
10 specific diet components estimated from Donovan (1999), and DE values from 2007 onwards were developed from
11 an analysis by Archibeque (2011), based on diet information in Preston (2010) and USDA:APHIS:VS (2010).
12 Weight and weight gains for cattle were estimated from Holstein (2010), Doren et al. (1989), Enns (2008), Lippke et
13 al. (2000), Pinchack et al. (2004), Platter et al. (2003), Skogerboe et al. (2000), and expert opinion. See Annex 3.10
14 for more details on the method used to characterize cattle diets and weights in the United States.
15 Calves younger than 4 months are not included in emission estimates because calves consume mainly milk and the
16 IPCC recommends the use of a Ym of zero for all juveniles consuming only milk. Diets for calves aged 4 to 6
17 months are assumed to go through a gradual weaning from milk decreasing to 75 percent at 4 months, 50 percent at
18 age 5 months, and 25 percent at age 6 months. The portion of the diet made up with milk still results in zero
19 emissions. For the remainder of the diet, beef calf DE and Ym are set equivalent to those of beef replacement heifers,
20 while dairy calf DE is set equal to that of dairy replacement heifers and dairy calf Ym is provided at 4 and 7 months
21 of age by Soliva (2006). Estimates of Ym for 5 and 6 month old dairy calves are linearly interpolated from the values
22 provided for 4 and 7 months.
23 To estimate CH4 emissions, the population was divided into state, age, sub-type (i.e., dairy cows and replacements,
24 beef cows and replacements, heifer and steer stackers, heifers and steers in feedlots, bulls, beef calves 4 to 6 months,
25 and dairy calves 4 to 6 months), and production (i.e., pregnant, lactating) groupings to more fully capture differences
26 in CH4 emissions from these animal types. The transition matrix was used to simulate the age and weight structure
27 of each sub-type on a monthly basis in order to more accurately reflect the fluctuations that occur throughout the
28 year. Cattle diet characteristics were then used in conjunction with Tier 2 equations from IPCC (2006) to produce
29 CH4 emission factors for the following cattle types: dairy cows, beef cows, dairy replacements, beef replacements,
30 steer stackers, heifer stackers, steer feedlot animals, heifer feedlot animals, bulls, and calves. To estimate emissions
31 from cattle, monthly population data from the transition matrix were multiplied by the calculated emission factor for
32 each cattle type. More details are provided in Annex 3.10.
33 Emission estimates for other animal types were based on average emission factors representative of entire
34 populations of each animal type. Methane emissions from these animals accounted for a minor portion of total CH4
35 emissions from livestock in the United States from 1990 through 2014. Additionally, the variability in emission
36 factors for each of these other animal types (e.g., variability by age, production system, and feeding practice within
37 each animal type) is less than that for cattle. Annual livestock population data for sheep; swine; goats; horses; mules
38 and asses; and American bison were obtained for available years from USDA NASS (USDA 2015). Horse, goat and
39 mule and ass population data were available for 1987, 1992, 1997, 2002, 2007, and 2012 (USDA 1992, 1997, 2015);
40 the remaining years between 1990 and 2014 were interpolated and extrapolated from the available estimates (with
41 the exception of goat populations being held constant between 1990 and 1992). American bison population
42 estimates were available from USDA for 2002, 2007, and 2012 (USDA 2014) and from the National Bison
43 Association (1999) for 1997 through 1999. Additional years were based on observed trends from the National Bison
44 Association (1999), interpolation between known data points, and extrapolation beyond 2012, as described in more
45 detail in Annex 3.10. Methane emissions from sheep, goats, swine, horses, American bison, and mules and asses
46 were estimated by using emission factors utilized in Crutzen et al. (1986, cited in IPCC 2006). These emission
47 factors are representative of typical animal sizes, feed intakes, and feed characteristics in developed countries. For
48 American bison the emission factor for buffalo was used and adjusted based on the ratio of live weights to the 0.75
49 power. The methodology is the same as that recommended by IPCC (2006).
Due to inconsistencies in the 2003 literature values, the 2002 values were used for 2003, as well.
Agriculture 5-5
-------
1 See Annex 3.10 for more detailed information on the methodology and data used to calculate CH4 emissions from
2 enteric fermentation.
Uncertainty and Time-Series Consistency
4 A quantitative uncertainty analysis for this source category was performed using the IPCC-recommended Approach
5 2 uncertainty estimation methodology based on a Monte Carlo Stochastic Simulation technique as described in ICF
6 (2003). These uncertainty estimates were developed for the 1990 through 2001 Inventory report (i.e., 2003
7 submission to the UNFCCC). There have been no significant changes to the methodology since that time;
8 consequently, these uncertainty estimates were directly applied to the 2014 emission estimates in this Inventory
9 report.
10 A total of 185 primary input variables (177 for cattle and 8 for non-cattle) were identified as key input variables for
11 the uncertainty analysis. A normal distribution was assumed for almost all activity- and emission factor-related
12 input variables. Triangular distributions were assigned to three input variables (specifically, cow-birth ratios for the
13 three most recent years included in the 2001 model run) to ensure only positive values would be simulated. For
14 some key input variables, the uncertainty ranges around their estimates (used for inventory estimation) were
15 collected from published documents and other public sources; others were based on expert opinion and best
16 estimates. In addition, both endogenous and exogenous correlations between selected primary input variables were
17 modeled. The exogenous correlation coefficients between the probability distributions of selected activity-related
18 variables were developed through expert judgment.
19 The uncertainty ranges associated with the activity data-related input variables were plus or minus 10 percent or
20 lower. However, for many emission factor-related input variables, the lower- and/or the upper-bound uncertainty
21 estimates were over 20 percent. The results of the quantitative uncertainty analysis are summarized in Table 5-5.
22 Based on this analysis, enteric fermentation CH4 emissions in 2014 were estimated to be between 146.2 and 193.9
23 MMT CO2 Eq. at a 95 percent confidence level, which indicates a range of 11 percent below to 18 percent above the
24 2014 emission estimate of 164.3 MMT CCh Eq. Among the individual cattle sub-source categories, beef cattle
25 account for the largest amount of CH4 emissions, as well as the largest degree of uncertainty in the emission
26 estimates—due mainly to the difficulty in estimating the diet characteristics for grazing members of this animal
27 group. Among non-cattle, horses represent the largest percent of uncertainty in the previous uncertainty analysis
28 because the Food and Agricultural Organization of the United Nations (FAO) population estimates used for horses
29 at that time had a higher degree of uncertainty than for the USD A population estimates used for swine, goats, and
30 sheep. The horse populations are now from the same USDA source as the other animal types, and therefore the
31 uncertainty range around horses is likely overestimated. Cattle calves, American bison, mules and asses were
32 excluded from the initial uncertainty estimate because they were not included in emission estimates at that time.
33 Table 5-5: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Enteric
34 Fermentation (MMT COz Eq. and Percent)
Source
Enteric Fermentation
Gas
CH4
2014 Emission
Estimate
(MMT CO2 Eq.)
164.3
Uncertainty Range Relative to Emission Estimate3' b> c
(MMT CO2 Eq.) (%)
Lower Upper
Bound Bound
146.2 193.9
Lower Upper
Bound Bound
-11% 18%
1 Range of emissions estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
b Note that the relative uncertainty range was estimated with respect to the 2001 emission estimates from the 2003
submission and applied to the 2014 estimates.
0 The overall uncertainty calculated in 2003, and applied to the 2014 emission estimate, did not include uncertainty
estimates for calves, American bison, and mules and asses. Additionally, for bulls the emissions estimate was based
on the Tier 1 methodology. Since bull emissions are now estimated using the Tier 2 method, the uncertainty
surrounding their estimates is likely lower than indicated by the previous uncertainty analysis.
35 Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
36 through 2014. Details on the emission trends through time are described in more detail in the Methodology section.
5-6 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
QA/QC and Verification
2 In order to ensure the quality of the emission estimates from enteric fermentation, the IPCC Tier 1 and Tier 2
3 Quality Assurance/Quality Control (QA/QC) procedures were implemented consistent with the U.S. QA/QC plan.
4 Tier 2 QA procedures included independent peer review of emission estimates. Over the past few years, particular
5 importance has been placed on harmonizing the data exchange between the enteric fermentation and manure
6 management source categories. The current Inventory now utilizes the transition matrix from the CEFM for
7 estimating cattle populations and weights for both source categories, and the CEFM is used to output volatile solids
8 and nitrogen excretion estimates using the diet assumptions in the model in conjunction with the energy balance
9 equations from the IPCC (2006). This approach facilitates the QA/QC process for both of these source categories.
10
Recalculations Discussion
11 For the current Inventory, differences can be seen in emission estimates for years prior to 2014 when compared
12 against the same years in the previous Inventory—from 2008 through 2013 in particular. These recalculations were
13 due to changes made to historical data and corrections made to erroneous formulas in the CEFM. No modifications
14 were made to the methodology.
15 Revisions to input data include the following:
16 • The USDA published minor revisions in several categories that affected historical emissions estimated for cattle
17 for 2008 and subsequent years, including the following:
18 o Cattle populations for all animal types were revised for many states for 2009 and subsequent
19 years;
20 o Dairy cow milk production values were revised for several states for 2008 and subsequent years;
21 o Beef cattle feedlot placement data were revised for 2008 and subsequent years;
22 o Slaughter values were revised for 2008 and subsequent years;
23 o Calf birth data were revised for 2010 and subsequent years; and
24 o Cattle on feed data were revised for many states for 2009 and subsequent years.
25 • The USDA also revised population estimates for some categories of non-cattle animals, which affected
26 historical emissions estimated for "other" livestock. Changes included:
27 o Revised 2008 through 2012 populations for market and breeding swine in some states; and
28 o Revised 2011 and 2012 populations of sheep for some states.
29 In addition to these changes in input data, there were transcription and formula cell reference errors in the CEFM
30 calculations for the state-by-state estimates of cattle on feed. These errors, when corrected, affected emission
31 estimates for 2009 and subsequent years for all stackers and feedlot cattle.
32 These recalculations had an insignificant impact on the overall emission estimates.
33 Planned Improvements
34 Continued research and regular updates are necessary to maintain an emissions inventory that reflects the current
35 base of knowledge. Future improvements for enteric fermentation could include some of the following options:
36 • Further research to improve the estimation of dry matter intake (as gross energy intake) using data from
37 appropriate production systems;
38 • Updating input variables that are from older data sources, such as beef births by month and beef cow lactation
39 rates;
40 • Investigation of the availability of annual data for the DE, Ym, and crude protein values of specific diet and feed
41 components for foraging and feedlot animals;
42 • Further investigation on additional sources or methodologies for estimating DE for dairy, given the many
43 challenges in characterizing dairy diets;
Agriculture 5-7
-------
1 • Further evaluation of the assumptions about weights and weight gains for beef cows, such that trends beyond
2 2007 are updated, rather than held constant;
3 • Further evaluation of the estimated weight for dairy cows (i.e., 1,500 Ibs) that is based solely on Holstein cows
4 as mature dairy cow weight is likely slightly overestimated, based on knowledge of the breeds of dairy cows in
5 the United States;
6 • Potentially updating to a Tier 2 methodology for other animal types (i.e., sheep, swine, goats, horses);
7 • Investigation of methodologies and emission factors for including enteric fermentation emission estimates from
8 poultry;
9 • Comparison of the current CEFM processing of animal population data to estimates developed using annual
10 average populations to determine if the model could be simplified to use annual population data; and
11 • Recent changes that have been implemented to the CEFM warrant an assessment of the current uncertainty
12 analysis; therefore, a revision of the quantitative uncertainty surrounding emission estimates from this source
13 category will be initiated.
14
15
5.2 Manure Management (IPCC Source
Category 3B)
16 The treatment, storage, and transportation of livestock manure can produce anthropogenic CH4 and N2O emissions.
17 Methane is produced by the anaerobic decomposition of manure. Nitrous oxide emissions are produced through
18 both direct and indirect pathways. Direct N2O emissions are produced as part of the N cycle through the
19 nitrification and denitrification of the organic N in livestock dung and urine.3 There are two pathways for indirect
20 N2O emissions. The first is the result of the volatilization of N in manure (as NH3 and NOX) and the subsequent
21 deposition of these gases and their products (NH4+ and NOs") onto soils and the surface of lakes and other waters.
22 The second pathway is the runoff and leaching of N from manure to the groundwater below, in riparian zones
23 receiving drain or runoff water, or in the ditches, streams, rivers, and estuaries into which the land drainage water
24 eventually flows.
25 When livestock or poultry manure are stored or treated in systems that promote anaerobic conditions (e.g., as a
26 liquid/slurry in lagoons, ponds, tanks, or pits), the decomposition of the volatile solids component in the manure
27 tends to produce CH4. When manure is handled as a solid (e.g., in stacks or drylots) or deposited on pasture, range,
28 or paddock lands, it tends to decompose aerobically and produce little or no CH4. Ambient temperature, moisture,
29 and manure storage or residency time affect the amount of CH4 produced because they influence the growth of the
30 bacteria responsible for CH4 formation. For non-liquid-based manure systems, moist conditions (which are a
31 function of rainfall and humidity) can promote CH4 production. Manure composition, which varies by animal diet,
32 growth rate, and type, including the animal's digestive system, also affects the amount of CH4 produced. In general,
33 the greater the energy content of the feed, the greater the potential for CH4 emissions. However, some higher-energy
34 feeds also are more digestible than lower quality forages, which can result in less overall waste excreted from the
35 animal.
36 The production of direct N2O emissions from livestock manure depends on the composition of the manure and urine,
37 the type of bacteria involved in the process, and the amount of oxygen and liquid in the manure system. For direct
38 N2O emissions to occur, the manure must first be handled aerobically where ammonia (NH3) or organic N is
39 converted to nitrates and nitrites (nitrification), and then handled anaerobically where the nitrates and nitrites are
40 reduced to dinitrogen gas (N2), with intermediate production of N2O and nitric oxide (NO) (denitrification)
3 Direct and indirect N2O emissions from dung and urine spread onto fields either directly as daily spread or after it is removed
from manure management systems (e.g., lagoon, pit, etc.) and from livestock dung and urine deposited on pasture, range, or
paddock lands are accounted for and discussed in the Agricultural Soil Management source category within the Agriculture
sector.
5-8 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 (Groffman et al. 2000). These emissions are most likely to occur in dry manure handling systems that have aerobic
2 conditions, but that also contain pockets of anaerobic conditions due to saturation. A very small portion of the total
3 N excreted is expected to convert to N2O in the waste management system (WMS). Indirect N2O emissions are
4 produced when nitrogen is lost from the system through volatilization (as NH3 or NOX) or through runoff and
5 leaching. The vast majority of volatilization losses from these operations are NH3. Although there are also some
6 small losses of NOX, there are no quantified estimates available for use, so losses due to volatilization are only based
7 on NH3 loss factors. Runoff losses would be expected from operations that house animals or store manure in a
8 manner that is exposed to weather. Runoff losses are also specific to the type of animal housed on the operation due
9 to differences in manure characteristics. Little information is known about leaching from manure management
10 systems as most research focuses on leaching from land application systems. Since leaching losses are expected to
11 be minimal, leaching losses are coupled with runoff losses and the runoff/leaching estimate provided in this chapter
12 does not account for any leaching losses.
13 Estimates of CH4 emissions in 2014 were 61.2 MMT CO2 Eq. (2,447 kt); in 1990, emissions were 37.2 MMT CO2
14 Eq. (1,486 kt). This represents a 65 percent increase in emissions from 1990. The majority of this increase is due to
15 swine and dairy cow manure, where emissions increased 44 and 118 percent, respectively. Emissions increased on
16 average by 1.0 MMT CO2 Eq. (2.2 percent) annually over this period. From 2013 to 2014, there was a 0.3 percent
17 decrease in total CH4 emissions, mainly due to minor shifts in the animal populations and the resultant effects on
18 manure management system allocations.
19 Although the majority of managed manure in the United States is handled as a solid, producing little CH4, the
20 general trend in manure management, particularly for dairy and swine (which are both shifting towards larger
21 facilities), is one of increasing use of liquid systems. Also, new regulations controlling the application of manure
22 nutrients to land have shifted manure management practices at smaller dairies from daily spread systems to storage
23 and management of the manure on site. Although national dairy animal populations have generally been decreasing
24 since 1990, some states have seen increases in their dairy populations as the industry becomes more concentrated in
25 certain areas of the country and the number of animals contained on each facility increases. These areas of
26 concentration, such as California, New Mexico, and Idaho, tend to utilize more liquid-based systems to manage
27 (flush or scrape) and store manure. Thus the shift toward larger dairy and swine facilities has translated into an
28 increasing use of liquid manure management systems, which have higher potential CH4 emissions than dry systems.
29 This significant shift in both the dairy and swine industries was accounted for by incorporating state and WMS-
30 specific CH4 conversion factor (MCF) values in combination with the 1992, 1997, 2002, and 2007 farm-size
31 distribution data reported in the Census of Agriculture (USDA 2014a).
32 In 2014, total N2O emissions were estimated to be 17.5 MMT CO2 Eq. (59 kt); in 1990, emissions were 14.0 MMT
33 CO2 Eq. (47 kt). These values include both direct and indirect N2O emissions from manure management. Nitrous
34 oxide emissions have remained fairly steady since 1990. Small changes in N2O emissions from individual animal
35 groups exhibit the same trends as the animal group populations, with the overall net effect that N2O emissions
36 showed a 25 percent increase from 1990 to 2014 and a 0.1 percent decrease from 2013 through 2014. Overall shifts
37 toward liquid systems have driven down the emissions per unit of nitrogen excreted.
38 Table 5-6 and Table 5-7 provide estimates of CH4 and N2O emissions from manure management by animal
39 category.
40 Table 5-6: CH4 and NzO Emissions from Manure Management (MMT COz Eq.)
"Gas/Animal Type 1990 2005 2010 2011 2012 2013 2014
CH4a 37.2 56.3 60.9 61.5 63.7 61.4 61.2
Dairy Cattle 14.7 26.4 30.4 31.1 32.6 31.8 32.2
Swine 15.6 22.9 23.6 23.6 24.3 23.0 22.4
Poultry 3.3l 3.2B 3.2 3.2 3.2 3.2 3.2
Beef Cattle 3.11 3.sB 3.3 3.3 3.2 3.0 3.0
Horses 0.21 O.sB 0.2 0.2 0.2 0.2 0.2
Sheep 0.2B 0.11 0.1 0.1 0.1 0.1 0.1
Goats + B + B + + + + +
Bison + B + B + + + + +
Mules and Asses + B + B + + + + +
N2Ob 14.01 16.5 17.2 17.4 17.5 17.5 17.5
Agriculture 5-9
-------
Beef Cattle
Dairy Cattle
Dairy Heifers
Swine
Poultry
Sheep
Horses
Goats
Mules and Asses
Bison
,,_
72
32
23
1.7
NA
7.6
3.3
2.5
1.9
1.5
0.3
0.1
NA
7.7
3.3
2.5
1.9
1.5
0.3
0.1
NA
7.7
3.4
2.5
1.9
1.6
0.3
0.1
NA
7.7
3.4
2.5
1.9
1.6
0.3
0.1
NA
7.8
3.4
2.5
1.8
1.6
0.3
0.1
NA
Total
51.1
72.9
78.1
78.9
81.2
78.9
78.7
+ Does not exceed 0.05 MMT CO2 Eq.
Note: Totals may not sum due to independent rounding. American bison are maintained entirely on
unmanaged WMS; there are no American bison N2O emissions from managed systems.
NA: Not available
a Accounts for CH4 reductions due to capture and destruction of CH4 at facilities using anaerobic
digesters.
b Includes both direct and indirect N2O emissions.
1 Table 5-7: CH4 and NzO Emissions from Manure Management (kt)
Gas/Animal Type
1990
2005
2010 2011 2012
2013
2014
CH4a
Dairy Cattle
Swine
Poultry
Beef Cattle
Horses
Sheep
Goats
Bison
Mules and Asses
N2Ob
Beef Cattle
Dairy Cattle
Dairy Heifers
Swine
Poultry
Sheep
Horses
Goats
Mules and Asses
Bison
1,486 2,254 2,437
590
622
131
126
9
7
1
+
+
47
20
11
7
4
5
+
+
+
+
1,057 1,217
916
129
133
12
3
1
+
+
55
24
11
8
6
5
1
+
+
+
945
129
132
10
3
1
+
+
58
25
11
8
6
5
1
+
+
+
NA NA NA
2,460
1,245
942
127
131
10
3
1
+
+
58
26
11
8
6
5
1
+
+
+
NA
2,548
1,306
972
128
128
10
3
1
+
+
59
26
11
8
6
5
1
+
+
+
NA
2,455
1,271
920
128
121
9
3
1
+
+
59
26
11
8
6
5
1
+
+
+
NA
2,4^7
1,289
896
130
120
9
3
1
+
+
59
26
11
8
6
5
1
+
+
+
NA
+ Does not exceed 0.5 kt.
Note: Totals may not sum due to independent rounding. American bison are maintained entirely on
unmanaged WMS; there are no American bison N2O emissions from managed systems.
NA: Not available
a Accounts for CH4 reductions due to capture and destruction of CH4 at facilities using anaerobic
digesters.
b Includes both direct and indirect N2O emissions.
Methodology
4 The methodologies presented in IPCC (2006) form the basis of the CH4 and N2O emission estimates for each animal
5 type. This section presents a summary of the methodologies used to estimate CH4 and N2O emissions from manure
5-10 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 management. See Annex 3.11 for more detailed information on the methodology and data used to calculate CH4 and
2 N2O emissions from manure management.
3 Methane Calculation Methods
4 The following inputs were used in the calculation of CH4 emissions:
5 • Animal population data (by animal type and state);
6 • Typical animal mass (TAM) data (by animal type);
7 • Portion of manure managed in each WMS, by state and animal type;
8 • Volatile solids (VS) production rate (by animal type and state or United States);
9 • Methane producing potential (B0) of the volatile solids (by animal type); and
10 • Methane conversion factors (MCF), the extent to which the CH4 producing potential is realized for each
11 type of WMS (by state and manure management system, including the impacts of any biogas collection
12 efforts).
13 Methane emissions were estimated by first determining activity data, including animal population, TAM, WMS
14 usage, and waste characteristics. The activity data sources are described below:
15 • Annual animal population data for 1990 through 2014 for all livestock types, except goats, horses, mules
16 and asses, and American bison were obtained from USDA National Agriculture Statistics Service (NASS).
17 For cattle, the USDA populations were utilized in conjunction with birth rates, detailed feedlot placement
18 information, and slaughter weight data to create the transition matrix in the Cattle Enteric Fermentation
19 Model (CEFM) that models cohorts of individual animal types and their specific emission profiles. The
20 key variables tracked for each of the cattle population categories are described in Section 5.1 and in more
21 detail in Annex 3.10. Goat population data for 1992, 1997, 2002, 2007, and 2012; horse and mule and ass
22 population data for 1987, 1992, 1997, 2002 2007, and 2012; and American bison population for 2002, 2007
23 and 2012 were obtained from the Census of Agriculture (USDA 2014a). American bison population data
24 for 1990 through 1999 were obtained from the National Bison Association (1999).
25 • The TAM is an annual average weight that was obtained for animal types other than cattle from
26 information in USD A's Agricultural Waste Management Field Handbook (USDA 1996), the American
27 Society of Agricultural Engineers, Standard D384.1 (ASAE 1998) and others (Meagher 1986, EPA 1992,
28 Safley 2000, ERG 2003b, IPCC 2006, and ERG 2010a). For a description of the TAM used for cattle,
29 please see Section 5.1.
30 • WMS usage was estimated for swine and dairy cattle for different farm size categories using data from
31 USDA (USDA, APHIS 1996, Bush 1998, Ott 2000, USDA 2014a) and EPA (ERG 2000a, EPA 2002a,
32 2002b). For beef cattle and poultry, manure management system usage data were not tied to farm size but
33 were based on other data sources (ERG 2000a, USDA, APHIS 2000, UEP 1999). For other animal types,
34 manure management system usage was based on previous estimates (EPA 1992). American bison WMS
35 usage was assumed to be the same as not on feed (NOF) cattle, while mules and asses were assumed to be
36 the same as horses.
37 • VS production rates for all cattle except for calves were calculated by head for each state and animal type
38 in the CEFM. VS production rates by animal mass for all other animals were determined using data from
39 USDA's Agricultural Waste Management Field Handbook (USDA 1996, 2008 and ERG 2010b and 2010c)
40 and data that was not available in the most recent Handbook were obtained from the American Society of
41 Agricultural Engineers, Standard D 3 84.1 (ASAE 1998) orthe 2 006 IPCC Guidelines. American bison VS
42 production was assumed to be the same as NOF bulls.
43 • The maximum CH4-producing capacity of the VS (B0) was determined for each animal type based on
44 literature values (Morris 1976, Bryant et al. 1976, Hashimoto 1981, Hashimoto 1984, EPA 1992, Hill 1982,
45 Hill 1984).
46 • MCFs for dry systems were set equal to default IPCC factors based on state climate for each year (IPCC
47 2006). MCFs for liquid/slurry, anaerobic lagoon, and deep pit systems were calculated based on the
48 forecast performance of biological systems relative to temperature changes as predicted in the van't Hoff-
49 Arrhenius equation which is consistent with IPCC (2006) Tier 2 methodology.
Agriculture 5-11
-------
1 • Data from anaerobic digestion systems with CH4 capture and combustion were obtained from the EPA
2 AgSTAR Program, including information presented in the AgSTAR Digest (EPA 2000, 2003 , 2006) and the
3 AgSTAR project database (EPA 2012). Anaerobic digester emissions were calculated based on estimated
4 methane production and collection and destruction efficiency assumptions (ERG 2008).
5 • For all cattle except for calves, the estimated amount of VS (kg per animal-year) managed in each WMS
6 for each animal type, state, and year were taken from the CEFM, assuming American bison VS production
7 to be the same as NOF bulls. For animals other than cattle, the annual amount of VS (kg per year) from
8 manure excreted in each WMS was calculated for each animal type, state, and year. This calculation
9 multiplied the animal population (head) by the VS excretion rate (kg VS per 1,000 kg animal mass per
10 day), the TAM (kg animal mass per head) divided by 1,000, the WMS distribution (percent), and the
11 number of days per year (365.25).
12 The estimated amount of VS managed in each WMS was used to estimate the CH4 emissions (kg CH4 per year)
13 from each WMS . The amount of VS (kg per year) were multiplied by the maximum CH4 producing capacity of the
14 VS (B0) (m3 CH4 per kg VS), the MCF for that WMS (percent), and the density of CH4 (kg CH4 per m3 CH4). The
15 CH4 emissions for each WMS, state, and animal type were summed to determine the total U.S. CH4 emissions.
16 Nitrous Oxide Calculation Methods
17 The following inputs were used in the calculation of direct and indirect N2O emissions:
18 • Animal population data (by animal type and state);
19 • TAM data (by animal type);
20 • Portion of manure managed in each WMS (by state and animal type);
21 • Total Kjeldahl N excretion rate (Nex);
22 • Direct N2O emission factor (EFwMs);
23 • Indirect N2O emission factor for volatilization (EFvoiatuization);
24 • Indirect N2O emission factor for runoff and leaching (EFnmoff/ieach);
25 • Fraction of N loss from volatilization of NH3 and NOX (Fracgas); and
26 • Fraction of N loss from runoff and leaching
27 N2O emissions were estimated by first determining activity data, including animal population, TAM, WMS usage,
28 and waste characteristics. The activity data sources (except for population, TAM, and WMS, which were described
29 above) are described below:
30 • Nex rates for all cattle except for calves were calculated by head for each state and animal type in the
3 1 CEFM. Nex rates by animal mass for all other animals were determined using data from USDA's
32 Agricultural Waste Management Field Handbook (USDA 1996, 2008 and ERG 2010b and 2010c) and data
33 from the American Society of Agricultural Engineers, Standard D384. 1 (ASAE 1998) and IPCC (2006).
34 American bison Nex rates were assumed to be the same as NOF bulls.
35 • All N2O emission factors (direct and indirect) were taken from IPCC (2006). These data are appropriate
36 because they were developed using U.S. data.
37 • Country -specific estimates for the fraction of N loss from volatilization (Fracgas) and runoff and leaching
38 (Fracmnoff/ieach) were developed. Fracgas values were based on WMS-specific volatilization values as
3 9 estimated from EP A' s National Emission Inventory - Ammonia Emissions from Animal Agriculture
40 Operations (EPA 2005). FraCnmoff/ieachmg values were based on regional cattle runoff data from EPA's
4 1 Office of Water (EPA 2002b; see Annex 3.11).
42 To estimate N2O emissions for cattle (except for calves) and American bison, the estimated amount of N excreted
43 (kg per animal-year) managed in each WMS for each animal type, state, and year were taken from the CEFM. For
44 calves and other animals, the amount of N excreted (kg per year) in manure in each WMS for each animal type,
45 state, and year was calculated. The population (head) for each state and animal was multiplied by TAM (kg animal
46 mass per head) divided by 1,000, the nitrogen excretion rate (Nex, in kg N per 1,000 kg animal mass per day), WMS
47 distribution (percent), and the number of days per year.
48 Direct N2O emissions were calculated by multiplying the amount of N excreted (kg per year) in each WMS by the
49 N2O direct emission factor for that WMS (EFwMs, in kg N2O-N per kg N) and the conversion factor of N2O-N to
5-12 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 N2O. These emissions were summed over state, animal, and WMS to determine the total direct N2O emissions (kg
2 of N2O per year).
3 Next, indirect N2O emissions from volatilization (kg N2O per year) were calculated by multiplying the amount of N
4 excreted (kg per year) in each WMS by the fraction of N lost through volatilization (Fractas) divided by 100, and the
5 emission factor for volatilization (EFvoiatiiization, in kg N2O per kg N), and the conversion factor of N2O-N to N2O.
6 Indirect N2O emissions from runoff and leaching (kg N2O per year) were then calculated by multiplying the amount
7 of N excreted (kg per year) in each WMS by the fraction of N lost through runoff and leaching (FraCmnoff/ieach)
8 divided by 100, and the emission factor for runoff and leaching (EFrunoff/ieach, in kg N2O per kg N), and the
9 conversion factor of N2O-N to N2O. The indirect N2O emissions from volatilization and runoff and leaching were
10 summed to determine the total indirect N2O emissions.
11 The direct and indirect N2O emissions were summed to determine total N2O emissions (kg N2O per year).
12 Uncertainty and Time-Series Consistency
13 An analysis (ERG 2003a) was conducted for the manure management emission estimates presented in the 1990
14 through 2001 Inventory report (i.e., 2003 submission to the UNFCCC) to determine the uncertainty associated with
15 estimating CH4 and N2O emissions from livestock manure management. The quantitative uncertainty analysis for
16 this source category was performed in 2002 through the IPCC-recommended Approach 2 uncertainty estimation
17 methodology, the Monte Carlo Stochastic Simulation technique. The uncertainty analysis was developed based on
18 the methods used to estimate CH4 and N2O emissions from manure management systems. A normal probability
19 distribution was assumed for each source data category. The series of equations used were condensed into a single
20 equation for each animal type and state. The equations for each animal group contained four to five variables
21 around which the uncertainty analysis was performed for each state. These uncertainty estimates were directly
22 applied to the 2014 emission estimates as there have not been significant changes in the methodology since that
23 time.
24 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 5-8. Manure management
25 CH4 emissions in 2014 were estimated to be between 50.2 and 73.4 MMT CO2 Eq. at a 95 percent confidence level,
26 which indicates a range of 18 percent below to 20 percent above the actual 2014 emission estimate of 61.2 MMT
27 CO2 Eq. At the 95 percent confidence level, N2O emissions were estimated to be between 14.7 and 21.7 MMT CO2
28 Eq. (or approximately 16 percent below and 24 percent above the actual 2014 emission estimate of 17.5 MMT CO2
29 Eq.).
30 Table 5-8: Approach 2 Quantitative Uncertainty Estimates for CH4 and NzO (Direct and
31 Indirect) Emissions from Manure Management (MMT COz Eq. and Percent)
2014 Emission
Source Gas Estimate
(MMT CO2 Eq.)
Uncertainty Range Relative to Emission Estimate3
(MMT CO2 Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
Manure Management CH4 61.2
Manure Management N2O 17.5
50.2 73.4 -18% 20%
14.7 21.7 -16% 24%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
32 Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
33 through 2014. Details on the emission trends through time are described in more detail in the Methodology section.
Agriculture 5-13
-------
QA/QC and Verification
2
o
J
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Tier 1 and Tier 2 QA/QC activities were conducted consistent with the U.S. QA/QC plan. Tier 2 activities focused
on comparing estimates for the previous and current Inventories for N2O emissions from managed systems and CH4
emissions from livestock manure. All errors identified were corrected. Order of magnitude checks were also
conducted, and corrections made where needed. Manure N data were checked by comparing state-level data with
bottom up estimates derived at the county level and summed to the state level. Similarly, a comparison was made
by animal and WMS type for the full time series, between national level estimates for N excreted and the sum of
county estimates for the full time series.
Any updated data, including population, are validated by experts to ensure the changes are representative of the best
available U.S.-specific data. The U.S.-specific values for TAM, Nex, VS, B0, and MCF were also compared to the
IPCC default values and validated by experts. Although significant differences exist in some instances, these
differences are due to the use of U.S.-specific data and the differences in U.S. agriculture as compared to other
countries. The U.S. manure management emission estimates use the most reliable country-specific data, which are
more representative of U.S. animals and systems than the 2006 IPCC default values.
For additional verification, the implied CH4 emission factors for manure management (kg of CH4 per head per year)
were compared against the default 2006 IPCC values. Table 5-9 presents the implied emission factors of kg of CH4
per head per year used for the manure management emission estimates as well as the IPCC default emission factors.
The U.S. implied emission factors fall within the range of the 2006 IPCC default values, except in the case of sheep,
goats, and some years for horses and dairy cattle. The U.S. implied emission factors are greater than the 2006 IPCC
default value for those animals due to the use of U.S.-specific data for typical animal mass and VS excretion. There
is an increase in implied emission factors for dairy and swine across the time series. This increase reflects the dairy
and swine industry trend towards larger farm sizes; large farms are more likely to manage manure as a liquid and
therefore produce more CH4 emissions.
Table 5-9: 2006 IPCC Implied Emission Factor Default Values Compared with Calculated
Values for CtLi from Manure Management (kg/head/year)
Animal Type
IPCC Default
CH4 Emission
Factors
(kg/head/vear)
Implied CH4 Emission Factors (kg/head/year)
1990
2005
2010 2011 2012 2013 2014
Dairy Cattle
Beef Cattle
Swine
Sheep
Goats
Poultry
Horses
Mules and Asses
American Bison
48-112
1-2
10-45
0.19-0.37
0.13-0.26
0.02-1.4
1.56-3.13
0.76-1.14
NA
67.5
1.7
14.4
0.5
0.3
0.1
2.6
1.0
2.1
70.3
1.7
14.6
0.5
0.3
0.1
2.7
1.0
2.1
68.7
1.6
14.1
0.5
0.3
0.1
2.5
0.9
2.0
69.7
1.6
14.0
0.5
0.3
0.1
2.5
0.9
2.0
NA - Not applicable
In addition, 2006 default IPCC emission factors for N2O were compared to the U.S. Inventory implied N2O emission
factors. Default N2O emission factors from the 2006 IPCC Guidelines were used to estimate N2O emission from
each WMS in conjunction with U.S.-specific Nex values. The implied emission factors differed from the U.S.
Inventory values due to the use of U. S.-specific Nex values and differences in populations present in each WMS
throughout the time series.
32 Recalculations Discussion
33 The CEFM produces population, VS and Nex data for cattle, excepting calves, that are used in the manure
34 management inventory. As a result, all changes to the CEFM described in Section 5.1 contributed to changes in the
35 population, VS and Nex data used for calculating CH4 and N2O cattle emissions from Manure Management. In
5-14 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 addition, the Manure Management emission estimates included the following recalculations relative to the previous
2 Inventory:
3 • State animal populations were updated to reflect updated USDA NASS datasets, which resulted in
4 population changes for poultry in 2013, both beef and dairy calves from 2009 through 2013, sheep in 2011
5 and 2012, and swine from 2008 through 2013.
6 • Indirect N2O emissions for daily spread were added, as they are not accounted for in the Agricultural Soil
7 Management category. This inclusion increased indirect and total N2O emissions for dairy cows and dairy
8 heifers. Indirect N2O emissions increased between 0.9 and 5.2 percent per year, while total N2O emissions
9 increased between 0.6 to 1.4 percent per year.
10 Planned Improvements
11 The uncertainty analysis for Manure Management will be updated in future Inventories to more accurately assess
12 uncertainty of emission calculations. This update is necessary due to the extensive changes in emission calculation
13 methodology, including estimation of emissions at the WMS level and the use of new calculations and variables for
14 indirect N2O emissions.
15 In the next Inventory report, updated AgSTAR anaerobic digester data will be incorporated. In addition, potential
16 data sources (such as the USDA Agricultural Resource Management Survey) for updated WMS distribution
17 estimates will be reviewed and discussed with USDA. Further, future Inventories may present emissions on a
18 monthly basis to show seasonal emission changes for each WMS; this update would help compare these Inventory
19 data to other data and models.
20
21
5.3 Rice Cultivation (IPCC Source Category 3C)
(TO BE UPDATED)
22 Most of the world's rice, and all rice in the United States, is grown on flooded fields (Baicich 2013). When fields
23 are flooded, aerobic decomposition of organic material gradually depletes most of the oxygen in the soil. Once
24 depleted, soil conditions become anaerobic, and CH4 is produced by the decomposition of soil organic matter by
25 anaerobic methanogenic bacteria. Most of the CH4 produced does not reach the atmosphere. Up to 60 to 90 percent
26 is oxidized by aerobic methanotrophic bacteria in the soil (some oxygen remains at the interfaces of soil and water,
27 and soil and root systems) (Holzapfel-Pschorn et al. 1985, Sass et al. 1990) and some is leached away as dissolved
28 CH4 in floodwater that percolates from the field. The remaining un-oxidized CH4 is transported from the submerged
29 soil to the atmosphere primarily by diffusive transport through the rice plants. Minor amounts of CH4 also escape
30 from the soil via diffusion and bubbling through floodwaters.
31 The water management systems used to cultivate rice are one of the most important factors affecting CH4 emissions.
32 Upland rice fields are not flooded, and therefore are not believed to produce much CH4. In deepwater rice fields
33 (i.e., fields with flooding depths greater than one meter), the lower stems and roots of the rice plants die, thus
34 blocking the primary CH4 transport pathway to the atmosphere. The quantities of CH4 released from deepwater
35 fields are therefore believed to be significantly less than rice fields with shallower flooding depths (Sass 2001).
36 Some flooded rice fields are drained periodically during the growing season, either intentionally or accidentally. If
37 water is drained and soils are allowed to dry sufficiently, CH4 emissions decrease or stop entirely due to soil
38 aeration. Aeration not only causes existing soil CH4 to oxidize, but also inhibits further CH4 production in soils. In
39 the United States, rice is grown under continuously flooded, shallow water conditions (USDA 2012) and mid-season
40 drainage does not occur except by accident (e.g., due to levee breach).
41 Other factors that influence CH4 emissions from flooded rice fields include fertilization practices (i.e., the use of
42 urea and organic fertilizers), soil temperature, soil type, rice variety, and cultivation practices (e.g., tillage, seeding,
43 and weeding practices). Factors that influence the amount of organic material available for anaerobic decomposition
Agriculture 5-15
-------
1 (i.e., fertilizer use, soil type, rice variety,4 and cultivation practices) are the most important variables influencing the
2 amount of CH4 emitted over the growing season. Soil temperature is an important factor regulating the activity of
3 methanogenic bacteria which in turn affects the rate of CH4 production. However, although temperature influences
4 the time required to convert organic material to CH4, the impact of soil temperature on CH4 emissions is minor over
5 the length of the growing season. The application of synthetic fertilizers also influences CH4 emissions; in
6 particular, both nitrate and sulfate fertilizers (e.g., ammonium nitrate and ammonium sulfate) appear to inhibit CH4
7 formation. Nitrate and sulfate fertilizers are not commonly used in rice cultivation in the United States.
8 Rice is currently cultivated in seven states: Arkansas, California, Florida, Louisiana, Mississippi, Missouri, and
9 Texas.5 Soil types, rice varieties, and cultivation practices for rice vary from state to state, and even from farm to
10 farm. Most rice farmers recycle crop residues from the previous rice or rotational crop, either by leaving them
11 standing, disking them, or rolling them into fields. Most farmers also apply synthetic fertilizer (usually urea) to their
12 fields. In addition, the climatic conditions of Arkansas, Florida, southwest Louisiana, and Texas often allow for a
13 second, or ratoon, rice crop. Ratoon crops are produced from regrowth of the stubble remaining after the harvest of
14 the first rice crop. Ratoon crops are infrequent to non-existent in California, Mississippi, and Missouri. In 2012,
15 Arkansas reported a larger-than-usual ratoon crop (10 percent) due to an early rice harvest followed by warm
16 weather and heavy rains (ideal conditions for secondary growth and ratoon crops) (Hardke 2014). CH4 emissions
17 from ratoon crops are considerably higher than those from the primary crops due to the lack of a delay between
18 cropping seasons (which would allow the stubble to decay aerobically) (Wang et al. 2013). Specifically, the amount
19 of organic material available for anaerobic decomposition during ratoon crop production is considerably higher than
20 the amount available with the first (i.e., primary) crop production.
21 Rice cultivation is a minor source of CH4 emissions in the United States (see Table 5-10 and Table 5-11). In 2013,
22 CH4 emissions from rice cultivation were 8.3 MMT CCh Eq. (332 kt). Annual emissions have fluctuated unevenly
23 between 1990 and 2013, ranging from an annual decrease of 24 percent from 2010 and 2011 to an annual increase of
24 18 percent from 2009 to 2010. There was an overall decrease of 16 percent between 1990 and 2006, due to an
25 overall decrease in primary crop area. However, between 2006 and 2013 emission levels have increased by
26 8 percent due to increased ratooning and changes in production areas. California, Louisiana and Texas reported an
27 increase in rice crop area from 2012 to 2013. All other states reported a decrease in rice crop area from 2012 to
28 2013. The factors that affect the rice acreage in any year vary from state to state and are typically the result of
29 weather phenomena (Baldwin et al. 2010).
30 Table 5-10: CH4 Emissions from Rice Cultivation (MMT COz Eq.)
State 1990 2005 2010 2011 2012 2013 2014
Arkansas 4.0 5.7 6.2 6.2 6.2 6.2 6.2
California 1.0 2.3 1.6 1.6 1.6 1.6 1.6
Florida
Illinois
Louisiana 3.0 3.1 2.6 2.6 2.6 2.6 2.6
Minnesota
Mississippi 0.7 0.8 0.4 0.4 0.4 0.4 0.4
Missouri 0.3 0.5 0.8 0.8 0.8 0.8 0.8
New York + + + + + + +
South Carolina + +1 + + + + +
Tennessee + +1 + + + + +
Texas 2.4 1.7 0.6 0.6 0.6 0.6 0.6
I
Total 11.3 14.2 12.2 12.2 12.2 12.2 12.2
+ Less than 0.05 MMT CO2 Eq.
Note: Totals may not sum due to independent rounding.
The roots of rice plants shed organic material, which is referred to as "root exudate." The amount of root exudate produced by
a rice plant over a growing season varies among rice varieties.
5 Oklahoma has also historically produced rice. 2007 was the most recent production year reported (77 hectares).
5-16 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 5-11: ChU Emissions from Rice Cultivation (kt)
State 1990 2005 2010 2011 2012 2013 2014
Arkansas 160.0 229.4 246.5 246.8 247.1 247.1 246.4
California 38.1 90.3 64.2 64.4 64.2 64.3 64.5
Florida + I 0.8 + + + + +
Illinois + I 0.3 + + + + +
Louisiana 118.1 124.9 103.8 103.5 104.6 104.5 105.7
Minnesota 0.8 0.8 0.2 0.2 0.2 0.2 0.2
Mississippi 27.5 32.5 16.5 16.7 16.7 16.6 16.5
Missouri 11.3 20.6 31.4 31.5 31.5 31.5 31.4
New York + I + I + + + + +
South
Carolina + I + I + + + + +
Tennessee + I 0.7 + + + + +
Texas 95.5 66.8 23.6 23.7 23.6 23.7 23.4
Total 451 567 486 487 488 488 488
+ Lessthan0.5kt
Note: Totals may not sum due to independent rounding.
2 Methodology
3 IPCC (2006) recommends using harvested rice areas, and seasonally integrated emission factors (i.e., country
4 specific emission factors that have been developed from standardized field measurements (representing the mix of
5 different conditions that influence CH4 emissions in the area) for each commonly occurring rice production system).
6 To that end, the recommended GPG methodology and Tier 2 U.S.-specific seasonally integrated emission factors
7 derived from U.S. based rice field measurements are used.
8 Regional emission factors were derived based on a literature review of recent research on CH4 emissions from U.S.
9 rice production. In California, some rice fields are flooded during the winter to prepare the fields for the next
10 growing season, and to create waterfowl habitat (Young 2013). Winter flooded rice crops generate CH4 year round
11 due to the anaerobic conditions the winter flooding creates (Environmental Defense Fund 2011), and up to 50
12 percent of the CH4 emissions occur in the winter (Fitzgerald et al. 2000). Thus for winter flooded rice crops in
13 California, an annual CH4 emission factor is used. For non-winter flooded California rice crops, a seasonal emission
14 factor is applied as almost all of the CH4 emissions occur during the growing season (Fitzgerald et al. 2000).
15 California-specific winter flooded and non-winter flooded emission factors were applied to rice area harvested in
16 California. Average U.S. seasonal emission factors were applied to Arkansas, Florida, Louisiana, Missouri,
17 Mississippi, and Texas as there was not sufficient data to develop state-specific, or daily emission factors, or both.
18 As described above, seasonal emissions are much higher for ratooned crops than for primary crops. Therefore,
19 emissions from ratooned and primary areas are estimated separately using the appropriate representative emission
20 factors. This approach is consistent with IPCC (2006).
21 To determine what CH4 emission factors should be used for the primary and ratoon crops, CH4 flux information
22 from rice field measurements in the United States was collected. Experiments that involved atypical or non-
23 representative management practices (e.g., the application of nitrate or sulfate fertilizers, or other substances
24 believed to suppress CH4 formation, or floodwaters were drained mid-season), as well as experiments in which
25 measurements were not made over an entire flooding season were excluded from the analysis. The remaining
26 experimental results were then sorted by state, season (i.e., primary and ratoon), flooding practices, and type of
27 fertilizer amendment (i.e., no fertilizer added, organic fertilizer added, and synthetic and organic fertilizer added).
28 Eleven California-specific primary crop experimental results were added for California rice emissions starting with
29 the 1990-2012 Inventory. These California-specific studies were selected because they met the criteria of
30 experiments on primary crops with added synthetic and organic fertilizer, without residue burning, and without
31 winter flooding (Bossio et al. 1999; Fitzgerald et al. 2000). The seasonal emission rates estimated in these studies
32 were averaged to derive a seasonal emission factor for California's primary, non-winter flooded rice crop.
33 Similarly, separate California-specific studies meeting the same criteria, (i.e., primary crops with added synthetic
34 and organic fertilizer, without residue burning) but with winter flooding (Bossio et al. 1999; Fitzgerald et al. 2000;
35 McMillan et al. 2007) were averaged to derive an annual emission factor for California's primary, winter-flooded
Agriculture 5-17
-------
1 rice crop. Approximately 60 percent of California's rice crop is winter-flooded (Environmental Defense Fund
2 2011), therefore the California-specific, winter flooded emission factor was applied to 60 percent of the California
3 rice area harvested and the California-specific, non-winter flooded emission factor was applied to the 40 percent of
4 the California rice area harvested. The resultant seasonal emission factor for the California, non-winter flooded crop
5 is 133 kg CHVhectare/season, and the annual emission factor for the California, winter-flooded crop is 266 kg
6 CHVhectare/season.
7 For the remaining states, a non-California U.S. seasonal emission factor was derived by averaging seasonal
8 emissions rates from primary crops with added synthetic and organic fertilizer (Byrd 2000; Kongchum 2005; Rogers
9 et al. 2011; Sass et al. 1991a, 1991b, 2002a, 2002b; Yao 2001). The seasonal emissions rates from ratoon crops
10 with added synthetic fertilizer (Lindau and Bollich 1993; Lindau et al. 1995) were averaged to derive a seasonal
11 emission factor for the ratoon crop. The resultant seasonal emission factor for the primary crop is
12 237 kg CHVhectare/season, and the resultant emission factor for the ratoon crop is 780 kg CH4/hectare/season.
13 The harvested rice areas for the primary and ratoon crops in each state are presented in Table 5-12, and the ratooned
14 crop area as a percent of primary crop area is shown in Table 5-13. Primary crop areas for 1990 through 2013 for all
15 states except Florida and Oklahoma were taken from U.S. Department of Agriculture's Field Crops Final Estimates
16 1987-1992 (USDA 1994), Field Crops Final Estimates 1992-1997 (USDA 1998), Field Crops Final Estimates
17 1997-2002 (USDA 2003), and Crop Production Summary (USDA 2005 through 2014). Source data for non-USDA
18 sources of primary and ratoon harvest areas are shown in Table 5-14. California, Mississippi, Missouri, and
19 Oklahoma have not ratooned rice over the period 1990 through 2013 (Anderson 2008 through 2014; Beighley 2011
20 through 2012; Buehring 2009 through 2011; Guethle 1999 through 2010; Lee 2003 through 2007; Mutters 2001
21 through 2005; Street 1999 through 2003; Walker 2005, 2007 through 2008).
22 Table 5-12: Rice Area Harvested (Hectare)
State/Crop
Arkansas
Primary
Ratoona
California
Florida
Primary
Ratoon
Louisiana
Primary
Ratoon
Mississippi
Missouri
Oklahoma
Lexas
Primary
Ratoon
Total Primary
Total Ratoon
Total
1990
485,633
-
159,854
4,978
2,489
220,558
66,168
101,174
32,376
617
142,857
57,143
1,148,047
125,799
1,273,847
2005
661,675
662
212,869
4,565
-
212,465
27,620
106,435
86,605
271
81,344
21,963
1,366,228
50,245
1,416,473
• 2009
594,901
6
225,010
5,664
2,266
187,778
65,722
98,341
80,939
^B
68,798
39,903
1,261,431
107,897
1,369,328
2010
722,380
7
223,796
5,330
2,275
216,512
86,605
122,622
101,578
^H
76,083
41,085
1,468,300
129,971
1,598,271
2011
467,017
5
234,723
8,212
2,311
169,162
59,207
63,537
51,801
^M
72,845
56,091
1,067,298
117,613
1,184,911
2012
520,032
52,003
225,415
6,244
2,748
160,664
64,265
52,206
71,631
^B
54,229
33,080
1,090,421
152,096
1,242,517
2013
433,023
21,651
227,034
6,739
2,159
167,139
63,513
50,182
63,132
^1
58,276
39,628
1,005,525
126,951
1,132,476
1 Arkansas ratooning occurred only in 1998,1999, and 2005 through 2013, with particularly
ratoon rates in 2012 and 2013.
"-" No reported value
Note: Lotals may not sum due to independent rounding.
high
23
Table 5-13: Ratooned Area as Percent of Primary Growth Area
State
Arkansas
Florida
Louisiana
1990
+
50%
30%
2005
0.1%
+
13%
2009
+
40%
35%
2010
+
43%
40%
2011
+
28%
35%
2012
10%
44%
40%
2013
5%
32%
38%
5-18 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Texas
40%
27%
58% 54% 77% 61% 68%
+ Indicates ratooning less than 0.05 percent of primary growth area.
Table 5-14: Non-USDA Data Sources for Rice Harvest Information
State/Crop
1990
2005
2009 2010 2011 2012 2013
Arkansas - Ratoona
Florida - Primaryb
Florida - Ratoonc
Louisiana - Ratoond
Oklahoma - Primary6
Texas - Ratoonf
Slaton
Schueneman
Schueneman
Linscombe
Lee
Klosterboer
Wilson
Gonzalez
Gonzalez
Linscombe
Lee
• Hardke
Wilson (2009-2011) (2012-2013)
Gonzalez (2009-2013)
Gonzalez (2009-2013)
Linscombe (2009-2013)
Anderson (2009-2013)
Texas Agricultural Experiment Station (TAES)
(2009-2013)
a Arkansas: 1990 - 2000 (Slaton 1999 through 2001); 2001 - 2011 (Wilson 2002 through 2007, 2009
through 2012). 2012 - 2013 (Hardke 2013, 2014).
bFlorida - Primary: 1990 - 2000 (Schueneman 1997,1999 through 2001); 2001 (Deren 2002); 2002 - 2004
(Kirstein 2003 through 2004,2006); 2005 - 2013 (Gonzalez 2007 through 2014)
cFlorida - Ratoon: 1990 - 2000 (Schueneman 1997, 1999 through 2001); 2001 (Deren 2002); 2002 - 2003
(Kirstein 2003 through 2004,2006); 2004 (Cantens 2004- 2005); 2005 - 2013 (Gonzalez 2007 through
2014)
Louisiana: 1990 - 2013 (Linscombe 1999, 2001 through 2014).
e Oklahoma: 1990 - 2006 (Lee 2003 through 2007); 2007 - 2013 (Anderson 2008 through 2014).
•Texas: 1990 - 2002 (Klosterboer 1997, 1999 through 2003); 2003 - 2004 (Stansel 2004 through 2005);
2005 (Texas Agricultural Experiment Station 2006); 2006-2013 (Texas Agricultural Experiment Station
2007-2014).
3 Emissions from rice production were estimated using a Tier 2 methodology consistent with IPCC (2006).
4 Representative emission factors using experimentally determined seasonal CH4 emissions from U.S. rice fields for
5 both primary and ratoon crops were derived from a literature review. Emissions are compared with the 1996 IPCC
6 Guidelines default U.S. seasonal emission factor, and not the more recent 2006IPCC Guidelines global daily
7 emission factor. The rationale for this comparison is that the evaluated studies were specific to the U.S., were
8 regional specific seasonal emission factors, and did not include daily emission factors or season length. As explained
9 above, four different emission factors were calculated: (1) a seasonal, California-specific factor without winter
10 flooding (133 kg CH4/hectare/season), (2) an annual, California specific-factor with winter flooding (266 kg
11 CHVhectare/year), (3) a seasonal, non-California primary crop factor (237 kg CHVhectare/season), and (4) a
12 seasonal, non-California ratoon crop factor (780 kg CH4/hectare/season). These emission factors represent averages
13 across rice field measurements representing typical water management practices and synthetic and organic
14 amendment application practices in the United States according to regional experts (Anderson 2013; Beighly 2012;
15 Fife 2011; Gonzalez 2013; Linscombe 2013; Vayssieres 2013; Wilson 2012). The IPCC (1996) default factor for
16 U.S. (i.e., Texas) rice production for both primary and ratoon crops is 250 kg CH4/hectare/season. This default
17 value is based on a study by Sass and Fisher (1995) which reflects a growing season in Texas of approximately 275
18 days. Data results in the evaluated studies were provided as seasonal emission factors; therefore, neither daily
19 emission factors nor growing season length was estimated. Some variability within season lengths in the evaluated
20 studies is assumed. The Tier 2 emission factors used here represent rice cultivation practices specific to the United
21 States. For comparison, the 2013 U.S. emissions from rice cultivation are 8.3 MMT CCh Eq. using the four U.S.-
22 specific emission factors for both primary and ratoon crops and 7.2 MMT CCh Eq. using the IPCC (1996) emission
23 factor.
24
25 Table 5-15: Non-California Seasonal Emission Factors (kg ChU/hectare/season)
Primary
Minimum
61
Ratoon
Minimum
481
Agriculture 5-19
-------
Maximum 500
Mean 237
Maximum 1490
Mean 780
1
2 Table 5-16: California Emission Factors (kg CH4/hectare/year or season)
Winter Flooded
(Annual)b
Minimum 131
Maximum 369
Mean 266
Non- Winter Flooded
(Seasonal)0
Minimum
Maximum
Mean
62
221
133
Note: See methodology text for why the emission factor is annual for winter flooded and
seasonal for non-winter flooded California rice production.
b Percentage of California rice crop winter flooded: 60 percent.
0 Percentage of California rice crop not winter flooded: 40 percent.
3 Uncertainty and Time-Series Consistency
4 The largest uncertainty in the calculation of CH4 emissions from rice cultivation is associated with the emission
5 factors. Seasonal emissions, derived from field measurements in the United States, vary by more than one order of
6 magnitude. This inherent variability is due to differences in cultivation practices, particularly fertilizer type,
7 amount, and mode of application; differences in cultivar type; and differences in soil and climatic conditions. A
8 portion of this variability is accounted for by separating primary from ratooned areas. However, even within a
9 cropping season or a given management regime, measured emissions may vary significantly. Of the experiments
10 used to derive the emission factors applied here, primary emissions ranged from 61 to 500 kg CHVhectare/season
11 and ratoon emissions ranged from 481 to 1,490 kg CH4/hectare/season. The uncertainty distributions around the
12 California winter flooding, California non-winter flooding, non-California primary, and ratoon emission factors
13 were derived using the distributions of the relevant emission factors available in the literature and described above.
14 Variability around the rice emission factor means was not normally distributed for any crop system, but rather
15 skewed, with a tail trailing to the right of the mean. A lognormal statistical distribution was, therefore, applied in
16 the uncertainty analysis.
17 Other sources of uncertainty include the primary rice-cropped area for each state, percent of rice-cropped area that is
18 ratooned, the length of the growing season, and the extent to which flooding outside of the normal rice season is
19 practiced. Expert judgment was used to estimate the uncertainty associated with primary rice-cropped area for each
20 state at 1 to 5 percent, and a normal distribution was assumed. Uncertainties were applied to ratooned area by state,
21 based on the level of reporting performed by the state. Within California, the uncertainty associated with the
22 percentage of rice fields that are winter flooded was estimated at plus and minus 20 percent. No uncertainty
23 estimates were calculated for the practice of flooding outside of the normal rice season outside of California because
24 CH4 flux measurements have not been undertaken over a sufficient geographic range or under a broad enough range
25 of representative conditions to account for this source in the emission estimates or its associated uncertainty.
26 To quantify the uncertainties for emissions from rice cultivation, a Monte Carlo (Approach 2) uncertainty analysis
27 was performed using the information provided above. The results of the Approach 2 quantitative uncertainty
28 analysis are summarized in Table 5-17. Rice cultivation CH4 emissions in 2013 were estimated to be between 4.2
29 and 15.9 MMT CO2 Eq. at a 95 percent confidence level, which indicates a range of 50 percent below to 91 percent
30 above the actual 2013 emission estimate of 8.3 MMT CO2 Eq.
5-20 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 5-17: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Rice
2 Cultivation (MMT COz Eq. and Percent)
10
26
Source
2013 Emission
Gas Estimate
(MMT CO2 Eq.)
Uncertainty Range Relative to Emission Estimate3
(MMT C02 Eq.) (%)
1 | 1 Lower Upper Lower Upper
Bound Bound Bound Bound
Rice Cultivation CH4 8J 42 15.9 -50% 91%
a Range of emissions estimates predicted by Monte Carlo Stochastic Simulation for a 95% confidence interval
3 Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
4 through 2013. Details on the emission trends through time are described in more detail in the Methodology section,
5 above.
QA/QC and Verification
7 A source-specific QA/QC plan for rice cultivation was developed and implemented. This effort included a Tier 1
8 analysis, as well as portions of a Tier 2 analysis. The Tier 2 procedures focused on comparing trends across years,
9 states, and cropping seasons to attempt to identify any outliers or inconsistencies. No problems were found.
Recalculations Discussion
11 For the current Inventory, emission estimates have been revised to reflect the GWPs provided in the IPCC Fourth
12 Assessment Report (AR4) (IPCC 2007). AR4 GWP values differ slightly from those presented in the IPCC Second
13 Assessment Report (SAR) (IPCC 1996) (used in the previous inventories) which results in time-series recalculations
14 for most Inventory sources. Under the most recent reporting guidelines (UNFCCC 2014), countries are required to
15 report using the AR4 GWPs, which reflect an updated understanding of the atmospheric properties of each
16 greenhouse gas. The GWPs of CH4 and most fluorinated greenhouse gases have increased, leading to an overall
17 increase in CCh-equivalent emissions from CH4. The GWPs of N2O and SF6 have decreased, leading to a decrease in
18 CCh-equivalent emissions for these greenhouse gases. The AR4 GWPs have been applied across the entire time
19 series for consistency. For more information please see the Recalculations Chapter. As a result of the updated GWP
20 value for CH4, emissions estimates for each year from 1990 to 2012 increased by 19 percent relative to the
21 emissions estimates in previous Inventory reports.
22 Additionally, the 2012 emission estimates were updated to reflect an increase in previously-reported ratooning in
23 Arkansas. Rice was harvested early in 2012, after which a high percentage of "secondary growth" occurred.
24 Estimated percent ratooning of secondary growth in 2012 increased from 5 to 10 percent (Hardke 2014), resulting in
25 a 0.4 MMT CO2 eq. (21 kt C) increase in emissions.
Planned Improvements
27 A planned improvement for the 1990 through 2014 Inventory will be the expansion of the California-specific rice
28 emission factors to include an emission factor for the period prior to the passage of the Air Resources Board (ARE)
29 Mandate phasing out rice residue burning. This non-flooded residue burned emission factor will take into account
30 the phase down of rice straw burning that occurred in California from 1990 to 2002. During this time period, the
31 percentage of acres burned annually decreased from 75 percent in 1992 to 13 percent in 2002 (California Air
32 Resources Board 2003). California studies that include rice burning on non-flooded lands will be used to develop
33 the pre-2002 rice burning emission factor, and further research will be conducted to determine the percentage of
34 winter flooded area to which the current California winter flooded emission factor will be applied. This new time
35 series dependent emission factor will be applied to non-flooded burned area during the 1990 through 2002 time
36 period to capture the significant change in the percentage of rice area burned due to the California ARE Mandate.
37 Following 2002, the current methodology and emission factors will be applied.
Agriculture 5-21
-------
1 Another possible future improvement is to create additional state- or region-specific emission factors for rice
2 cultivation. This prospective improvement would likely not take place for another two to three years, because the
3 analyses needed for it are currently taking place.
4 5.4 Agricultural Soil Management (IPCC Source
5 Category 3D)
6 Nitrous oxide (N2O) is naturally produced in soils through the microbial processes of nitrification and denitrification
7 that is driven by the availability of mineral N (Firestone and Davidson 1989).6 Mineral N is made available in soils
8 through decomposition of soil organic matter and plant litter, as well as asymbiotic fixation of N from the
9 atmosphere.7 A number of agricultural activities increase mineral nitrogen (N) availability in soils that lead to direct
10 N2O emissions from nitrification and denitrification at the site of a management activity (see Figure 5-2) (Mosier et
11 al. 1998), including fertilization; application of managed livestock manure and other organic materials such as
12 sewage sludge; deposition of manure on soils by domesticated animals in pastures, rangelands, and paddocks (PRP)
13 (i.e., by grazing animals and other animals whose manure is not managed); production of N-fixing crops and
14 forages; retention of crop residues; and drainage of organic soils (i.e., soils with a high organic matter content,
15 otherwise known as Histosols8) in croplands and grasslands (IPCC 2006). Additionally, agricultural soil
16 management activities, including irrigation, drainage, tillage practices, and fallowing of land, can influence N
17 mineralization by impacting moisture and temperature regimes in soils. Indirect emissions of N2O occur when N is
18 transported from a site and is subsequently converted to N2O; there are two pathways for indirect emissions: (1)
19 volatilization and subsequent atmospheric deposition of applied/mineralized N, and (2) surface runoff and leaching
20 of applied/mineralized N into groundwater and surface water.9
21 Direct and indirect emissions from agricultural lands are included in this section (i.e., cropland and grassland as
22 defined in Chapter 6.1 Representation of the U.S. Land Base). The U.S. Inventory includes all greenhouse gas
23 emissions from managed land based on guidance in IPCC (2006), and consequently N mineralization from
24 decomposition of soil organic matter and asymbiotic N fixation are also included in this section to fully address
25 emissions from the managed land base (see Methodology section for more information).
Nitrification and denitrification are driven by the activity of microorganisms in soils. Nitrification is the aerobic microbial
oxidation of ammonium (NH4+) to nitrate (NOs"), and denitrification is the anaerobic microbial reduction of nitrate to N2. Nitrous
oxide is a gaseous intermediate product in the reaction sequence of denitrification, which leaks from microbial cells into the soil
and then into the atmosphere. Nitrous oxide is also produced during nitrification, although by a less well-understood mechanism
(Nevison 2000).
Asymbiotic N fixation is the fixation of atmospheric N2 by bacteria living in soils that do not have a direct relationship with
plants.
8 Drainage of organic soils in former wetlands enhances mineralization of N-rich organic matter, thereby increasing N2O
emissions from these soils.
These processes entail volatilization of applied or mineralized N as NHs and NOX, transformation of these gases within the
atmosphere (or upon deposition), and deposition of the N primarily in the form of particulate NH4+, nitric acid (HNOs), and NOX.
5-22 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Figure 5-2: Sources and Pathways of N that Result in NzO Emissions from Agricultural Soil
2 Management
Sources and Pathways of N that Result in ^0 Emissions from Agricultural Soil Management
o
Synthetic N Fertilizers
Synthetic N fertilizer applied to soil
Organic
Amendments
Includes both commercial and
non-co,m merdsl fertilizers {i.e.,
sewage sludge, tankage, etc)
Urine and Dung from
Grazing Animals
Manure deposited on pasture,
and paddock
Crop Residues
Indudesabove-and belowground
residues for all crops(non-N and N-
fixing (and from perennial forage
crops and pastures followinq tencvja
Mineralization of
Soil Organic Matter
IndudesN convertedto mineral form
upon decomposition of soil organic
matter
Asymbiotic Fixation
Fixation of atmospheric N; by bacteria
living in soils that do not have a direct
relations hip with plants
This graphic illustrates the sources and pathways of nitrogen that result
in direct and indirect N20 emissions from soils using the methodologies
described in this Inventory. Emission pathways are shown with arrows.
On the lower right-hand side is a cut-away view of a representative
section of a managed soil; histosol cultivation is represented here.
N Volatilization
and Deposition
Agriculture 5-23
-------
1 Agricultural soils produce the majority of N2O emissions in the United States. Estimated emissions from this source
2 in 2014 are 318.5 MMT CO2 Eq. (1,069 kt) (see Table 5-18 and Table 5-19) Annual N2O emissions from
3 agricultural soils fluctuated between 1990 and 2014, although overall emissions are 5 percent higher in 2014 than in
4 1990. Year-to-year fluctuations are largely a reflection of annual variation in weather patterns, synthetic fertilizer
5 use, and crop production. From 1990 to 2014, on average cropland accounted for approximately 70 percent of total
6 direct emissions, while grassland accounted for approximately 30 percent. The percentages for indirect emissions
7 on average are approximately 66 percent for croplands, 34 percent for grasslands. Estimated direct and indirect N2O
8 emissions by sub-source category are shown in Table 5-20 and Table 5-21.
9 Table 5-18: NzO Emissions from Agricultural Soils (MMT COz Eq.)
Activity
Direct
Cropland
Grassland
Indirect
Cropland
Grassland
Total
1990
245.0
171.9
73.2
57.9
36.2
21.7
302.9
2005
248.3
174.4
73.9
48.4
34.0
14.5
296.7
2010
263.8
185.7
78.1
56.6
39.7
16.9
320.4
2011
264.5
186.9
77.6
58.4
40.6
17.7
322.9
2012
264.5
187.9
76.6
58.3
41.1
17.3
322.9
2013
261.2
185.2
76.0
57.2
40.3
17.0
318.4
2014
261.3
185.3
76.0
57.2
40.3
17.0
318.5
10 Table 5-19: NzO Emissions from Agricultural Soils (kt)
Activity
Direct
Cropland
Grassland
Indirect
Cropland
Grassland
Total
1990
2005
2010 2011 2012 2013 2014
822
577
246
194
121
73
833
II
885
623
262
190
133
57
888
627
260
196
136
60
888
630
257
196
138
58
877
621
255
192
135
57
1,017
996
877
622
255
192
135
57
1,075 1,084 1,083 1,069 1,069
11 Table 5-20: Direct NzO Emissions from Agricultural Soils by Land Use Type and N Input Type
12 (MMT COz Eq.)
Activity
Cropland
Mineral Soils
Synthetic Fertilizer
Organic Amendment*
Residue Nb
Mineralization and
Asymbiotic Fixation
Organic Soils
Grassland
Mineral Soils
Synthetic Fertilizer
PRP Manure
Managed Manure0
Sewage Sludge
Residue Nd
Mineralization and
Asymbiotic Fixation
Organic Soils
Total
1990
171.9
168.6
59.2
11.9
25.9
71.6
3.2 1
73.2 1
70.3
1.1 1
13.4 1
0.1
0.2 1
19.7 1
35.8 1
2.9
245.0 I
2005
174.4
171.2
61
12
26
70
.4
.9
.6
.3
3.2
73.9
71.0
1
12
0
0
21
35
2
.3
.3
.2
.5
.0
.8
.9
248.3
2010
185.7
182.6
59.3
13.4
27.8
82.2
3.0
78.1
75.5
1.3
12.5
0.2
0.5
21.8
39.1
2.7
263.8
2011
186.9
183.9
61.0
13.5
27.6
81.8
3.0
77.6
74.9
1.2
11.9
0.2
0.5
21.7
39.4
2.7
264.5
2012
187.9
184.9
61
13,
27,
82,
.8
.6
.5
.0
3.0
76.6
74.0
1.
11
0.
0.
21,
,2
.0
,2
,6
.7
39.4
2.
,7
264.5
2013
185.2
182.2
59,
13,
27,
81
.5
.5
.5
.6
3.0
76.0
73.3
1.
10,
0.
0.
21,
,2
.3
,2
,6
.7
39.4
2.
,7
261.2
2014
185.3
182.2
59.5
13.5
27.6
81.6
3.0
76.0
73.3
1.2
10.3
0.2
0.6
21.7
39.4
2.7
261.3
a Organic amendment inputs include managed manure, daily spread manure, and commercial organic
fertilizers (i.e., dried blood, dried manure, tankage, compost, and other).
b Cropland residue N inputs include N in unharvested legumes as well as crop residue N.
5-24 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
0 Managed manure inputs include managed manure and daily spread manure amendments that are
applied to grassland soils.
d Grassland residue N inputs include N in ungrazed legumes as well as ungrazed grass residue N
1 Table 5-21: Indirect NzO Emissions from Agricultural Soils (MMT COz Eq.)
Activity 1990 2005 2010 2011 2012 2013 2014
Cropland 36.2 34.0 39.7 40.6 41.1 40.3 40.3
Volatilization & Atm.
Deposition 13.0 13.8 13.9 14.3 14.5 14.2 14.2
Surface Leaching & Run-Off 23.2 20.2 25.8 26.4 26.6 26.0 26.0
Grassland 21.7 14.5 16.9 17.7 17.3 17.0 17.0
Volatilization & Atm.
Deposition 4.4 4.7 4.8 4.7 4.6 4.5 4.5
Surface Leaching & Run-Off 17.4 9.8 12.1 13.0 12.7 12.4 12.4
Total 57.9 48.4 56.6 58.4 58.3 57.2 57.2
2 Figure 5-3 and Figure 5-4 show regional patterns for direct N2O emissions for croplands and grasslands, and Figure
3 5-5 and Figure 5-6 show N losses from volatilization, leaching, and runoff that lead to indirect N2O emissions.
4 Annual emissions and N losses in 2014 are shown for the Tier 3 Approach only.
5 Direct N2O emissions from croplands tend to be high in the Corn Belt (Illinois, Iowa, Indiana, Ohio, southern and
6 western Minnesota, and eastern Nebraska), where a large portion of the land is used for growing highly fertilized
7 corn and N-fixing soybean crops (Figure 5-3). Kansas has high direct emissions associated with N management in
8 wheat production systems. Hay production in Missouri and irrigated cropping systems in California also contribute
9 relatively large amounts of direct N2O emissions, along with a combination of irrigated cropping in the west Texas
10 and hay production in east Texas. Direct emissions are low in many parts of the eastern United States because only
11 a small portion of land is cultivated and in many western states where rainfall and access to irrigation water are
12 limited.
13 Direct emissions from grasslands are highest in the central and western United States (Figure 5-4) where a high
14 proportion of the land is used for cattle grazing. In contrast, most areas in the Great Lake states, the Northeast, and
15 Southeast have moderate to low emissions due to less land dedicated to livestock grazing. However, emissions from
16 the Northeast and Great Lake states tend to be higher on a per unit area basis compared to other areas in the country.
17 This effect is likely due to a larger impact of freeze-thaw cycles in these regions, and possibly greater water-filled
18 pore space in the soil, which is key driver of N2O emissions (Kessavalou et al. 1998, Bateman and Baggs 2005).
19 Indirect emissions from croplands and grasslands (Figure 5-5 and Figure 5-6) show similar emission patterns to
20 those of direct emissions because the same driving variables (N inputs, weather patterns, soil characteristics) are
21 controlling both types of emissions. There are some exceptions to the similarity in patterns, however, because the
22 processes that contribute to indirect emissions (NOs" leaching, N volatilization) do not respond in exactly the same
23 manner to the driving variables as the processes that contribute to direct emissions (nitrification and denitrification).
24 Figure 5-3: Crops, Annual Direct NzO Emissions Estimated Using the Tier 3 DAYCENT Model,
25 1990-2014 (MMT COz Eq./year) (TO BE UPDATED)
26
27 Figure 5-4: Grasslands, Annual Direct NzO Emissions Estimated Using the Tier 3 DAYCENT
28 Model, 1990-2014 (MMT COz Eq./year) (TO BE UPDATED)
29
30 Figure 5-5: Crops, Average Annual N Losses Leading to Indirect NzO Emissions Estimated
31 Using the Tier 3 DAYCENT Model, 1990-2014 (kt N/year) (TO BE UPDATED)
32
Agriculture 5-25
-------
1 Figure 5-6: Grasslands, Average Annual N Losses Leading to Indirect NzO Emissions
2 Estimated Using the Tier 3 DAYCENT Model, 1990-2014 (kt N/year) (TO BE UPDATED)
3
4 Methodology
5 The 2006 IPCC Guidelines (IPCC 2006) divide emissions from the Agricultural Soil Management source category
6 into five components, including (1) direct emissions from N additions to cropland and grassland mineral soils from
7 synthetic fertilizers, sewage sludge applications, crop residues, organic amendments, and biological N fixation
8 associated with planting of legumes on cropland and grassland soils; (2) direct emissions from soil organic matter
9 mineralization due to land use and management change, (3) direct emissions from drainage of organic soils in
10 croplands and grasslands; (4) direct emissions from soils due to manure deposited by livestock on PRP grasslands;
1 1 and (5) indirect emissions from soils and water from N additions and manure deposition to soils that lead to
12 volatilization, leaching, or runoff of N and subsequent conversion to N2O.
13 The United States has adopted methods in the IPCC (2006) for agricultural soil management. These methods
14 include (1) estimating the contribution of N in crop residues to indirect soil N2O emissions; (2) adopting the revised
15 emission factor for direct N2O emissions for Tier 1 methods used in the Inventory (described later in this section);
16 (3) removing double counting of emissions from N-fixing crops associated with biological N fixation and crop
17 residue N input categories; (4) using revised crop residue statistics to compute N inputs to soils from harvest yield
18 data; and (5) estimating emissions associated with land use and management change (which can significantly change
19 the N mineralization rates from soil organic matter). 10 The Inventory also reports on total emissions from all
20 managed land, which is a proxy for anthropogenic impacts on greenhouse gas emissions (IPCC 2006), including
2 1 direct and indirect N2O emissions from asymbiotic fixation and mineralization of soil organic matter and litter. One
22 recommendation from IPCC (2006) that has not been completely adopted is the estimation of emissions from
23 grassland pasture renewal, which involves occasional plowing to improve forage production in pastures. Currently
24 no data are available to address pasture renewal.
25 Direct NzO Emissions
26 The methodology used to estimate direct N2O emissions from agricultural soil management in the United States is
27 based on a combination of IPCC Tier 1 and 3 approaches (IPCC 2006, Del Grosso et al. 2010). A Tier 3 process-
28 based model (DAYCENT) is used to estimate direct emissions from a variety of crops that are grown on mineral
29 (i.e., non-organic) soils, including alfalfa hay, barley, corn, cotton, dry beans, grass hay, grass-clover hay, oats,
30 onions, peanuts, potatoes, rice, sorghum, soybeans, sugar beets, sunflowers, tomatoes, and wheat; as well as the
3 1 direct emissions from non-federal grasslands with the exception of sewage sludge amendments (Del Grosso et al.
32 2010). The Tier 3 approach has been specifically designed and tested to estimate N2O emissions in the United
33 States, accounting for more of the environmental and management influences on soil N2O emissions than the IPCC
34 Tier 1 method (see Box 5-2 for further elaboration). Moreover, the Tier 3 approach allows for the Inventory to
35 address direct N2O emissions and soil C stock changes from mineral cropland soils in a single analysis. Carbon and
36 N dynamics are linked in plant-soil systems through biogeochemical processes of microbial decomposition and plant
37 production (McGill and Cole 1981). Coupling the two source categories (i.e., agricultural soil C and N2O) in a
38 single inventory analysis ensures that there is consistent activity data and treatment of the processes, and interactions
39 are taken into account between C and N cycling in soils.
40 The Tier 3 approach is based on the cropping and land use histories recorded in the USDA National Resources
41 Inventory (NRI) survey (USDA-NRCS 2013). The NRI is a statistically-based sample of all non-federal land, and
42 includes 380,956 points in agricultural land for the conterminous United States that are included in the Tier 3
43 method. The Tier 1 approach is used to estimate the emissions from the remaining 92,013 in the NRI survey that are
44 designated as cropland or grassland (discussed later in this section). Each point is associated with an "expansion
45 factor" that allows scaling of N2O emissions from NRI points to the entire country (i.e., each expansion factor
N inputs from asymbiotic N fixation are not directly addressed in 2006 IPCC Guidelines, but are a component of the total
emissions from managed lands and are included in the Tier 3 approach developed for this source.
5-26 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 represents the amount of area with the same land-use/management history as the sample point). Land-use and some
2 management information (e.g., crop type, soil attributes, and irrigation) were originally collected for each NRI point
3 on a 5-year cycle beginning in 1982. For cropland, data were collected in 4 out of 5 years in the cycle (i.e., 1979
4 through 1982, 1984 through 1987, 1989 through 1992, and 1994 through 1997). In 1998, the NRI program began
5 collecting annual data, the annual data are currently available through 2012 (USDA-NRCS 2015) although this
6 Inventory only uses NRI data through 2010 because newer data were not made available in time to incorporate the
7 additional years into this Inventory.
Box 5-2: Tier 1 vs. Tier 3 Approach for Estimating NzO Emissions
9 The IPCC (2006) Tier 1 approach is based on multiplying activity data on different N inputs (e.g., synthetic
10 fertilizer, manure, N fixation, etc.) by the appropriate default IPCC emission factors to estimate N2O emissions on
11 an input-by-input basis. The Tier 1 approach requires a minimal amount of activity data, readily available in most
12 countries (e.g., total N applied to crops); calculations are simple; and the methodology is highly transparent. In
13 contrast, the Tier 3 approach developed for this Inventory employs a process-based model (i.e., DAYCENT) that
14 represents the interaction of N inputs and the environmental conditions at specific locations. Consequently, the Tier
15 3 approach produces more accurate estimates; it accounts more comprehensively for land-use and management
16 impacts and their interaction with environmental factors (i.e., weather patterns and soil characteristics), which will
17 enhance or dampen anthropogenic influences. However, the Tier 3 approach requires more detailed activity data
18 (e.g., crop-specific N amendment rates), additional data inputs (e.g., daily weather, soil types, etc.), and considerable
19 computational resources and programming expertise. The Tier 3 methodology is less transparent, and thus it is
20 critical to evaluate the output of Tier 3 methods against measured data in order to demonstrate the adequacy of the
21 method for estimating emissions (IPCC 2006). Another important difference between the Tier 1 and Tier 3
22 approaches relates to assumptions regarding N cycling. Tier 1 assumes that N added to a system is subject to N2O
23 emissions only during that year and cannot be stored in soils and contribute to N2O emissions in subsequent years.
24 This is a simplifying assumption that is likely to create bias in estimated N2O emissions for a specific year. In
25 contrast, the process-based model used in the Tier 3 approach includes the legacy effect of N added to soils in
26 previous years that is re-mineralized from soil organic matter and emitted as N2O during subsequent years.
27
28 DAYCENT is used to estimate N2O emissions associated with production of alfalfa hay, barley, corn, cotton, dry
29 beans, grass hay, grass-clover hay, oats, onions, peanuts, potatoes, rice, sorghum, soybeans, sugar beets, sunflowers,
30 tomatoes, and wheat, but is not applied to estimate N2O emissions from other crops or rotations with other crops,"
31 such as sugarcane, some vegetables, tobacco, and perennial/horticultural crops. Areas that are converted between
32 agriculture (i.e., cropland and grassland) and other land uses, such as forest land, wetland and settlements, is not
33 simulated with DAYCENT. DAYCENT is also not used to estimate emissions from land areas with very gravelly,
34 cobbly, or shaley soils (greater than 35 percent by volume), or to estimate emissions from organic soils (Histosols).
35 The Tier 3 method has not been fully tested for estimating N2O emissions associated with these crops and rotations,
36 land uses, as well as organic soils or cobbly, gravelly, and shaley mineral soils. In addition, federal grassland areas
37 are not simulated with DAYCENT due to limited activity on land use histories. For areas that are not included in the
38 DAYCENT simulations, the Tier 1 IPCC (2006) methodology is used to estimate (1) direct emissions from crops on
39 mineral soils that are not simulated by DAYCENT; (2) direct emissions from Pasture/Range/Paddock (PRP) on
40 federal grasslands; and (3) direct emissions from drainage of organic soils in croplands and grasslands.
41 Tier 3 Approach for Mineral Cropland Soils
42 The DAYCENT biogeochemical model (Parton et al. 1998, and Del Grosso et al. 2001, 2011) is used to estimate
43 direct N2O emissions from mineral cropland soils that are managed for production of a wide variety of crops based
44 on the cropping histories in the 2010 NRI (USDA-NRCS 2013). The crops include alfalfa hay, barley, corn, cotton,
45 dry beans, grass hay, grass-clover hay, oats, onions, peanuts, potatoes, rice, sorghum, soybeans, sugar beets,
1! A small proportion of the major commodity crop production, such as corn and wheat, is included in the Tier 1 analysis because
these crops are rotated with other crops or land uses (e.g., forest lands) that are not simulated by DAYCENT.
Agriculture 5-27
-------
1 sunflowers, tomatoes, and wheat. Crops simulated by D AYCENT are grown on approximately 91 percent of total
2 cropland area in the United States. For agricultural systems in the central region of the United States, crop
3 production for key crops (i.e., corn, soybeans, sorghum, cotton and wheat) is simulated in DAYCENT with a
4 NASA-CASA production algorithm (Potter et al. 1993, Potter et al. 2007) using the MODIS Enhanced Vegetation
5 Index (EVI) products, MOD13Q1 andMYD13Ql, withapixel resolution of 250m.12
6 DAYCENT is used to estimate direct N2O emissions due to mineral N available from the following sources: (1) the
7 application of synthetic fertilizers; (2) the application of livestock manure; (3) the retention of crop residues and
8 subsequent mineralization of N during microbial decomposition (i.e., leaving residues in the field after harvest
9 instead of burning or collecting residues); and (4) mineralization of soil organic matter, in addition to asymbiotic
10 fixation. Note that commercial organic fertilizers (TVA 1991 through 1994 and AAPFCO 1995 through 2011) are
11 addressed with the Tier 1 method because county-level application data would be needed to simulate applications in
12 DAYCENT, and currently data are only available at the national scale. The third and fourth sources are generated
13 internally by the DAYCENT model.
14 Synthetic fertilizer data are based on fertilizer use and rates by crop type for different regions of the United States
15 that are obtained primarily from the USDA Economic Research Service Cropping Practices Survey (USDA-ERS
16 1997, 2011) with additional data from other sources, including the National Agricultural Statistics Service (NASS
17 1992, 1999, 2004). Frequency and rates of livestock manure application to cropland during 1997 are estimated from
18 data compiled by the USDA Natural Resources Conservation Service (Edmonds et al. 2003), and then adjusted
19 using county-level estimates of manure available for application in other years. The adjustments are based on
20 county-scale ratios of manure available for application to soils inotheryears relative to 1997 (see Annex 3.12 for
21 further details). Greater availability of managed manure N relative to 1997 is assumed to increase the area amended
22 with manure, while reduced availability of manure N relative to 1997 is assumed to reduce the amended area. Data
23 on the county-level N available for application is estimated for managed systems based on the total amount of N
24 excreted in manure minus N losses during storage and transport, and including the addition of N from bedding
25 materials. N losses include direct N2O emissions, volatilization of ammonia and NOX, runoff and leaching, and
26 poultry manure used as a feed supplement. For unmanaged systems, it is assumed that no N losses or additions
27 occur prior to the application of manure to the soil. More information on livestock manure production is available in
28 the Manure Management Section 5.2 and Annex 3.11.
29 The IPCC approach considers crop residue N and N mineralized from soil organic matter as activity data. However,
30 they are not treated as activity data in DAYCENT simulations because residue production, symbiotic N fixation
31 (e.g., legumes), mineralization of N from soil organic matter, and asymbiotic N fixation are internally generated by
32 the model as part of the simulation. In other words, DAYCENT accounts for the influence of symbiotic N fixation,
33 mineralization of N from soil organic matter and crop residue retained in the field, and asymbiotic N fixation on
34 N2O emissions, but these are not model inputs. The N2O emissions from crop residues are reduced by approximately
35 3 percent to avoid double-counting associated with non-CO2 greenhouse gas emissions from agricultural residue
36 burning. The estimate of residue burning is based on state inventory data (ILENR 1993, Oregon Department of
37 Energy 1995, Noller 1996, Wisconsin Department of Natural Resources 1993, Cibrowski 1996).
38 Additional sources of data are used to supplement the mineral N (USDA ERS 1997, 2011), livestock manure
39 (Edmonds et al. 2003), and land-use information (USDA-NRCS 2013). The Conservation Technology Information
40 Center (CTIC 2004) provided annual data on tillage activity with adjustments for long-term adoption of no-till
41 agriculture (Towery 2001). Tillage data has an influence on soil organic matter decomposition and subsequent soil
42 N2O emissions. The time series of tillage data began in 1989 and ended in 2004, so further changes in tillage
43 practices since 2004 are not currently captured in the Inventory. Daily weather data are used as an input in the model
44 simulations, based on gridded weather data at a 32 km scale from the North America Regional Reanalysis Product
45 (NARR) (Mesinger et al. 2006). Soil attributes are obtained from the Soil Survey Geographic Database (SSURGO)
46 (Soil Survey Staff 2011).
47 Each 2010 NRI point is run 100 times as part of the uncertainty assessment, yielding a total of over 18 million
48 simulations for the analysis. Soil N2O emission estimates from DAYCENT are adjusted using a structural
49 uncertainty estimator accounting for uncertainty in model algorithms and parameter values (Del Grosso et al. 2010).
50 SoilN2O emissions and 95 percent confidence intervals are estimated for each year between 1990 and 2010, but
12 See .
5-28 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 emissions from 2011 to 2014 are assumed to be similar to 2010. Annual data are currently available through 2012
2 (USDA-NRCS 2015). However, this Inventory only uses NRI data through 2010 because newer data were not
3 available in time to incorporate the additional years into this Inventory.
4 Nitrous oxide emissions from managed agricultural lands are the result of interactions among anthropogenic
5 activities (e.g., N fertilization, manure application, tillage) and other driving variables, such as weather and soil
6 characteristics. These factors influence key processes associated with N dynamics in the soil profile, including
7 immobilization of N by soil microbial organisms, decomposition of organic matter, plant uptake, leaching, runoff,
8 and volatilization, as well as the processes leading to N2O production (nitrification and denitrification). It is not
9 possible to partition N2O emissions into each anthropogenic activity directly from model outputs due to the
10 complexity of the interactions (e.g., N2O emissions from synthetic fertilizer applications cannot be distinguished
11 from those resulting from manure applications). To approximate emissions by activity, the amount of mineral N
12 added to the soil for each of these sources is determined and then divided by the total amount of mineral N that is
13 made available in the soil according to the DAYCENT model. The percentages are then multiplied by the total of
14 direct N2O emissions in order to approximate the portion attributed to key practices. This approach is only an
15 approximation because it assumes that all N made available in soil has an equal probability of being released as
16 N2O, regardless of its source, which is unlikely to be the case (Delgado et al. 2009). However, this approach allows
17 for further disaggregation of emissions by source of N, which is valuable for reporting purposes and is analogous to
18 the reporting associated with the IPCC (2006) Tier 1 method, in that it associates portions of the total soil N2O
19 emissions with individual sources of N.
20 Tier 1 Approach for Mineral Cropland Soils
21 The IPCC (2006) Tier 1 methodology is used to estimate direct N2O emissions for mineral cropland soils that are not
22 simulated by DAYCENT. For the Tier 1 Approach, estimates of direct N2O emissions from N applications are
23 based on mineral soil N that is made available from the following practices: (1) the application of synthetic
24 commercial fertilizers; (2) application of managed manure and non-manure commercial organic fertilizers; and (3)
25 the retention of above- and below-ground crop residues in agricultural fields (i.e., crop biomass that is not
26 harvested). Non-manure, commercial organic amendments is not included in the DAYCENT simulations because
27 county-level data are not available." Consequently, commercial organic fertilizer, as well as additional manure that
28 is not added to crops in the DAYCENT simulations, are included in the Tier 1 analysis. The following sources are
29 used to derive activity data:
30 • A process-of-elimination approach is used to estimate synthetic N fertilizer additions for crop areas not
31 simulated by DAYCENT. The total amount of fertilizer used on farms has been estimated at the county -
32 level by the USGS from sales records (Ruddy et al. 2006), and these data are aggregated to obtain state-
33 level N additions to farms. For 2002 through 2014, state-level fertilizer for on-farm use is adjusted based on
34 annual fluctuations in total U.S. fertilizer sales (AAPFCO 1995 through 2007, AAPFCO 2008 through
35 2014).M After subtracting the portion of fertilizer applied to crops and grasslands simulated by DAYCENT
36 (see Tier 3 Approach for Cropland Mineral Soils Section and Grasslands Section for information on data
37 sources), the remainder of the total fertilizer used on farms is assumed to be applied to crops that are not
3 8 simulated by DAYCENT.
39 • Similarly, a process-of-elimination approach is used to estimate manure N additions for crops that are not
40 simulated by DAYCENT. The amount of manure N applied in the Tier 3 approach to crops and grasslands
41 is subtracted from total manure N available for land application (see Tier 3 Approach for Cropland Mineral
42 Soils Section and Grasslands Section for information on data sources), and this difference is assumed to be
43 applied to crops that are not simulated by DAYCENT.
13 Commercial organic fertilizers include dried blood, tankage, compost, and other, but the dried manure and sewage sludge is
removed from the dataset in order to avoid double counting with other datasets that are used for manure N and sewage sludge.
14 Values are not available for 2013 so a "least squares line" statistical extrapolation using the previous 5 years of data is used to
arrive at an approximate value for 2014.
Agriculture 5-29
-------
1 • Commercial organic fertilizer additions are based on organic fertilizer consumption statistics, which are
2 converted to units of N using average organic fertilizer N content (TVA 1991 through 1994, AAPFCO
3 1995 through 2011). Commercial fertilizers do include some manure and sewage sludge, but the amounts
4 are removed from the commercial fertilizer data to avoid double counting with the manure N dataset
5 described above and the sewage sludge amendment data discussed later in this section.
6 • Crop residue N is derived by combining amounts of above- and below-ground biomass, which are
7 determined based on NRI crop area data (USDA-NRCS 2013), crop production yield statistics (USDA-
8 NASS 2014), dry matter fractions (IPCC 2006), linear equations to estimate above-ground biomass given
9 dry matter crop yields from harvest (IPCC 2006), ratios of below-to-above-ground biomass (IPCC 2006),
10 and N contents of the residues (IPCC 2006).
11 The total increase in soil mineral N from applied fertilizers and crop residues is multiplied by the IPCC (2006)
12 default emission factor to derive an estimate of direct N2O emissions using the Tier 1 Approach.
13 Drainage of Organic Soils in Croplands and Grasslands
14 The IPCC (2006) Tier 1 methods are used to estimate direct N2O emissions due to drainage of organic soils in
15 croplands or grasslands at a state scale. State-scale estimates of the total area of drained organic soils are obtained
16 from the 2010 NRI (USDA-NRCS 2013) using soils data from the Soil Survey Geographic Database (SSURGO)
17 (Soil Survey Staff 2011). Temperature data from Daly etal. (1994, 1998) are used to subdivide areas into temperate
18 and tropical climates using the climate classification from IPCC (2006). Annual data are available between 1990
19 and 2010. Emissions are assumed to be similar to 2010 from 2011 to 2014 because no additional activity data are
20 currently available from the NRI for the latter years. To estimate annual emissions, the total temperate area is
21 multiplied by the IPCC default emission factor for temperate regions, and the total tropical area is multiplied by the
22 IPCC default emission factor for tropical regions (IPCC 2006).
23 Direct N2O Emissions from Grassland Soils
24 As with N2O from croplands, the Tier 3 process-based DAYCENT model and Tier 1 method described in IPCC
25 (2006) are combined to estimate emissions from non-federal grasslands and PRP manure N additions for federal
26 grasslands, respectively. Grassland includes pasture and rangeland that produce grass forage primarily for livestock
27 grazing. Rangelands are typically extensive areas of native grassland that are not intensively managed, while
28 pastures are typically seeded grassland (possibly following tree removal) that may also have addition management,
29 such as irrigation or interseeding legumes. DAYCENT is used to simulate N2O emissions from NRI survey locations
30 (USDA-NRCS 2013) on non-federal grasslands resulting from manure deposited by livestock directly onto pastures
31 and rangelands (i.e., PRP manure), N fixation from legume seeding, managed manure amendments (i.e., manure
32 other than PRP manure such as Daily Spread), and synthetic fertilizer application. Other N inputs are simulated
33 within the DAYCENT framework, including N input from mineralization due to decomposition of soil organic
34 matter and N inputs from senesced grass litter, as well as asymbiotic fixation of N from the atmosphere. The
35 simulations used the same weather, soil, and synthetic N fertilizer data as discussed under the Tier 3 Approach for
36 Mineral Cropland Soils section. Managed manure N amendments to grasslands are estimated from Edmonds et al.
37 (2003) and adjusted for annual variation using data on the availability of managed manure N for application to soils,
38 according to methods described in the Manure Management section (5.2 Manure Management (IPCC Source
39 Category 3B)) and Annex 3.11. Biological N fixation is simulated within DAYCENT, and therefore is not an input
40 to the model.
41 Manure N deposition from grazing animals in PRP systems (i.e., PRP manure) is another key input of N to
42 grasslands. The amounts of PRP manure N applied on non-federal grasslands for each NRI point are based on
43 amount of N excreted by livestock in PRP systems. The total amount of N excreted in each county is divided by the
44 grassland area to estimate the N input rate associated with PRP manure. The resulting input rates are used in the
45 DAYCENT simulations. DAYCENT simulations of non-federal grasslands accounted for approximately 72 percent
46 of total PRP manure N in aggregate across the country. The remainder of the PRP manure N in each state is assumed
47 to be excreted on federal grasslands, and the N2O emissions are estimated using the IPCC (2006) Tier 1 method with
48 IPCC default emission factors. Sewage sludge is assumed to be applied on grasslands because of the heavy metal
49 content and other pollutants in human waste that limit its use as an amendment to croplands. Sewage sludge
50 application is estimated from data compiled by EPA (1993, 1999, 2003), McFarland (2001), and NEBRA (2007).
51 Sewage sludge data on soil amendments to agricultural lands are only available at the national scale, and it is not
5-30 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 possible to associate application with specific soil conditions and weather at the county scale. Therefore,
2 DAYCENT could not be used to simulate the influence of sewage sludge amendments on N2O emissions from
3 grassland soils, and consequently, emissions from sewage sludge are estimated using the IPCC (2006) Tier 1
4 method.
5 Grassland area data are consistent with the Land Representation reported in the LULUCF chapter. Data are
6 obtained from the U.S. Department of Agriculture NRI (Nusserand Goebel 1998) and the U.S. Geological Survey
7 (USGS) National Land Cover Dataset (Vogelman et al. 2001), which are reconciled with the Forest Inventory and
8 Analysis Data. The area data for pastures and rangeland are aggregated to the county level to estimate non-federal
9 and federal grassland areas.
10 N2O emissions for the PRP manure N deposited on federal grasslands and applied sewage sludge N are estimated
11 using the Tier 1 method by multiplying the N input by the appropriate emission factor. Emissions from manure N
12 are estimated at the state level and aggregated to the entire country, but emissions from sewage sludge N are
13 calculated exclusively at the national scale.
14 As previously mentioned, each NRI point is simulated 100 times as part of the uncertainty assessment, yielding a
15 total of over 18 million simulation runs for the analysis. Soil N2O emission estimates from DAYCENT are adjusted
16 using a structural uncertainty estimator accounting for uncertainty in model algorithms and parameter values (Del
17 Grosso et al. 2010). Soil N2O emissions and 95 percent confidence intervals are estimated for each year between
18 1990 and 2010, but emissions from 2011 to 2014 are assumed to be similar to 2010. The annual data are currently
19 available through 2012 (USDA-NRCS 2015). However, this Inventory only uses NRI data through 2010 because
20 newer data are not made available in time to incorporate the additional years into this Inventory.
21 Total Direct IVhO Emissions from Cropland and Grassland Soils
22 Annual direct emissions from the Tier 1 and 3 approaches for cropland mineral soils, from drainage and cultivation
23 of organic cropland soils, and from grassland soils are summed to obtain the total direct N2O emissions from
24 agricultural soil management (see Table 5-18 and Table 5-19).
25 Indirect IVhO Emissions
26 This section describes the methods used for estimating indirect soil N2O emissions from croplands and grasslands.
27 Indirect N2O emissions occur when mineral N made available through anthropogenic activity is transported from the
28 soil either in gaseous or aqueous forms and later converted into N2O. There are two pathways leading to indirect
29 emissions. The first pathway results from volatilization of N as NOX and NH3 following application of synthetic
30 fertilizer, organic amendments (e.g., manure, sewage sludge), and deposition of PRP manure. N made available
31 from mineralization of soil organic matter and residue, including N incorporated into crops and forage from
32 symbiotic N fixation, and input of N from asymbiotic fixation also contributes to volatilized N emissions.
3 3 Volatilized N can be returned to soils through atmospheric deposition, and a portion of the deposited N is emitted to
34 the atmosphere as N2O. The second pathway occurs via leaching and runoff of soil N (primarily in the form of NOs)
35 that is made available through anthropogenic activity on managed lands, mineralization of soil organic matter and
36 residue, including N incorporated into crops and forage from symbiotic N fixation, and inputs of N into the soil from
37 asymbiotic fixation. The NOs" is subject to denitrification in water bodies, which leads to N2O emissions.
38 Regardless of the eventual location of the indirect N2O emissions, the emissions are assigned to the original source
39 of the N for reporting purposes, which here includes croplands and grasslands.
40 Indirect N2O Emissions from Atmospheric Deposition of Volatilized N
41 The Tier 3 DAYCENT model and IPCC (2006) Tier 1 methods are combined to estimate the amount of N that is
42 volatilized and eventually emitted as N2O. DAYCENT is used to estimate N volatilization for land areas whose
43 direct emissions are simulated with DAYCENT (i.e., most commodity and some specialty crops and most
44 grasslands). The N inputs included are the same as described for direct N2O emissions in the Tier 3 Approach for
45 Cropland Mineral Soils Section and Grasslands Section. N volatilization for all other areas is estimated using the
46 Tier 1 method and default IPCC fractions for N subject to volatilization (i.e., N inputs on croplands not simulated by
47 DAYCENT, PRP manure N excreted on federal grasslands, sewage sludge application on grasslands). For the
Agriculture 5-31
-------
1 volatilization data generated from both the D AYCENT and Tier 1 approaches, the IPCC (2006) default emission
2 factor is used to estimate indirect N2O emissions occurring due to re-deposition of the volatilized N (Table 5-21).
3 Indirect N2O Emissions from Leaching/Runoff
4 As with the calculations of indirect emissions from volatilized N, the Tier 3 DAYCENT model and IPCC (2006)
5 Tier 1 method are combined to estimate the amount of N that is subject to leaching and surface runoff into water
6 bodies, and eventually emitted as N2O. DAYCENT is used to simulate the amount of N transported from lands in
7 the Tier 3 Approach. N transport from all other areas is estimated using the Tier 1 method and the IPCC (2006)
8 default factor for the proportion of N subject to leaching and runoff. This N transport estimate includes N
9 applications on croplands that are not simulated by DAYCENT, sewage sludge amendments on grasslands, and PPJ3
10 manure N excreted on federal grasslands. For both the DAYCENT Tier 3 and IPCC (2006) Tier 1 methods, nitrate
1 1 leaching is assumed to be an insignificant source of indirect N2O in cropland and grassland systems in arid regions
12 as discussed in IPCC (2006). In the United States, the threshold for significant nitrate leaching is based on the
13 potential evapotranspiration (PET) and rainfall amount, similar to IPCC (2006), and is assumed to be negligible in
14 regions where the amount of precipitation plus irrigation does not exceed 80 percent of PET. For leaching and
15 runoff data estimated by the Tier 3 and Tier 1 approaches, the IPCC (2006) default emission factor is used to
16 estimate indirect N2O emissions that occur in groundwater and waterways (Table 5-21).
Uncertainty and Time-Series Consistency
1 8 Uncertainty is estimated for each of the following five components of N2O emissions from agricultural soil
19 management: (1) direct emissions simulated by DAYCENT; (2) the components of indirect emissions (N volatilized
20 and leached or runoff) simulated by DAYCENT; (3) direct emissions approximated with the IPCC (2006) Tier 1
21 method; (4) the components of indirect emissions (N volatilized and leached or runoff) approximated with the IPCC
22 (2006) Tier 1 method; and (5) indirect emissions estimated with the IPCC (2006) Tier 1 method. Uncertainty in
23 direct emissions, which account for the majority of N2O emissions from agricultural management, as well as the
24 components of indirect emissions calculated by DAYCENT are estimated with a Monte Carlo Analysis, addressing
25 uncertainties in model inputs and structure (i.e., algorithms and parameterization) (Del Grosso et al. 2010).
26 Uncertainties in direct emissions calculated with the IPCC (2006) Approach 1 method, the proportion of
27 volatilization and leaching or runoff estimated with the IPCC (2006) Approach 1 method, and indirect N2O
28 emissions are estimated with a simple error propagation approach (IPCC 2006). Uncertainties from the Approach 1
29 and Approach 3 (i.e., DAYCENT) estimates are combined using simple error propagation (IPCC 2006). Additional
30 details on the uncertainty methods are provided in Annex 3.12. The combined uncertainty for direct soil N2O
31 emissions ranged from 16 percent below to 24 percent above the 2014 emissions estimate of 261.3 MMT CO2Eq.,
32 and the combined uncertainty for indirect soil N2O emissions range from 46 percent below to 139 percent above the
33 2014 estimate of 57.2 MMT CO2 Eq.
34 Table 5-22: Quantitative Uncertainty Estimates of NzO Emissions from Agricultural Soil
35 Management in 2014 (MMT COz Eq. and Percent)
2014 Emission
Source Gas Estimate Uncertainty Range Relative to Emission Estimate
(MMT C02 Eq.) (MMT CCh Eq.) (%)
Direct Soil N2O Emissions N2O
Indirect Soil N2O Emissions N2O
Lower
Bound
261.3 220.6
57.2 30.8
Upper
Bound
324.9
136.6
Lower
Bound
-16%
-46%
Upper
Bound
24%
139%
Notes: Due to lack of data, uncertainties in managed manure N production, PRP manure N production, other organic
fertilizer amendments, and sewage sludge amendments to soils are currently treated as certain; these sources of
uncertainty will be included in future Inventories.
36 Additional uncertainty is associated with the lack of an estimation of N2O emissions for croplands and grasslands in
37 Hawaii and Alaska, with the exception of drainage for organic soils in Hawaii. Agriculture is not extensive in either
38 state, so the emissions are likely to be small compared to the conterminous United States.
5-32 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Methodological recalculations are applied to the entire time series to ensure time-series consistency from 1990
2 through 2014. Details on the emission trends are described in more detail in the Methodology section above.
QA/QC and Verification
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
DAYCENT results for N2O emissions and NOs" leaching are compared with field data representing various cropland
and grassland systems, soil types, and climate patterns (Del Grosso et al. 2005, Del Grosso et al. 2008), and further
evaluated by comparing the model results to emission estimates produced using the IPCC (2006) Tier 1 method for
the same sites. N2O measurement data are available for 26 sites in the United States, 4 in Europe, and one in
Australia, representing over 75 different combinations of fertilizer treatments and cultivation practices. DAYCENT
estimates of N2O emissions are closer to measured values at most sites compared to the IPCC Tier 1 estimate
(Figure 5-7). In general, IPCC Tier 1 methodology tends to over-estimate emissions when observed values are low
and under-estimate emissions when observed values are high, while DAYCENT estimates have less bias.
DAYCENT accounts for key site-level factors (weather, soil characteristics, and management) that are not addressed
in the IPCC Tier 1 Method, and thus the model is better able to represent the variability in N2O emissions. Nitrate
leaching data are available for four sites in the United States, representing 12 different combinations of fertilizer
amendments/tillage practices. DAYCENT does have a tendency to under-estimate very high N2O emission rates;
and estimates are adjusted using the statistical model derived from the comparison of model estimates to
measurements (See Annex 3.12 for more information). Regardless, the comparison demonstrates that DAYCENT
provides relatively high predictive capability for N2O emissions and NOs" leaching, and is an improvement over the
IPCC Tier 1 method.
Figure 5-7: Comparison of Measured Emissions at Field Sites and Modeled Emissions Using
the DAYCENT Simulation Model and IPCC Tier 1 Approach.
9 n
8
t. 7 "
jp- 6 -
to 5 -
5,3 -
z 2 -
1 -
D Measured
1
_
EDayCent
• IPCC
M^^^l kfl
-n
:• r
ki
EEraifll/!1S1Sn'1SŁi/!JS
oon> MM-E-Et-g u w> —
H ^ >: s z ^ >• ll "1 - i "I-
PD O Z ^ O O (D
DO (fl (D (A ±j OJ
'E "> Ł">•?; -Q
L C ^ c o >»
o 8 3 8 « ^
u Q ? ra =
ui S Ł ^
| Z
5
-
1
r
i i;
11
i i i i i i i i
320.0055™
-Q M'~~-(J ° nbobP
8 1 J o K < 2
"=" 5 5
o S I-
z ^
S
Spreadsheets containing input data and probability distribution functions required for DAYCENT simulations of
croplands and grasslands and unit conversion factors have been checked, in addition to the program scripts that are
used to run the Monte Carlo uncertainty analysis. Links between spreadsheets have been checked, updated, and
corrected when necessary. Spreadsheets containing input data, emission factors, and calculations required for the
Tier 1 approach have been checked and updated as needed. Quality control is still underway and may lead to further
changes in the next Inventory report.
Agriculture 5-33
-------
i Recalculations Discussion
2 Methodological recalculations in the current Inventory are associated with the following improvements: (1) Driving
3 the D AYCENT simulations with updated input data for land management from the National Resources Inventory
4 extending the time series through 2010; (2) accounting for N inputs from residues associated with additional crops
5 not simulated by DAYCENT including most vegetable crops; (3) modifying the number of experimental study sites
6 used to quantify model uncertainty for direct N2O emissions; and (4) using DAYCENT for direct N2O emissions
7 from most flooded rice lands, instead of using the Tier 1 approach for all rice lands. These changes resulted in an
8 increase in emissions of approximately 24 percent on average relative to the previous Inventory and a decrease in
9 the upper bound of the 95 percent confidence interval for direct N2O emissions from 26 to 24 percent. The
10 differences are mainly due to increasing the number of study sites used to quantify model uncertainty and correct
11 bias.
12 Planned Improvements
13 Several planned improvements are underway:
14 (1) Improvements are underway to update the time series of land use and management data from the 2012
15 USDA NRI so that the time series of activity data are extended through 2012. Fertilization and tillage
16 activity data will also be updated as part of this improvement. In addition, the remote-sensing based
17 data on the Enhanced Vegetation Index will be extended through 2012 in order to use the EVI data to
18 drive crop production in DAYCENT.
19 (2) Improvements in the DAYCENT biogeochemical model are underway. Model structure will be
20 improved with a better representation of plant phenology, particularly senescence events following
21 grain filling in crops, such as wheat. In addition, crop parameters associated with temperature effects on
22 plant production will be further improved in DAYCENT with additional model calibration. An
23 improved representation of drainage is also under development. Experimental study sites will continue
24 to be added for quantifying model structural uncertainty, and studies that have continuous (daily)
25 measurements of N2O (e.g., Scheer et al. 2013) will be given priority because they provide more robust
26 estimates of annual emissions compared to studies that sample trace gas emissions weekly or less
27 frequently.
28 (3) Improvements are also underway to account for the influence of nitrification inhibitors and slow-release
29 fertilizers (e.g., polymer-coated fertilizers). Field data suggests that nitrification inhibitors and slow-
30 release fertilizers reduce N2O emissions significantly. The DAYCENT model can represent nitrification
31 inhibitors and slow-release fertilizers, but accounting for these in national simulations is contingent on
32 testing the model with a sufficient number of field studies and collection of activity data about the use
33 of these fertilizers;
34 (4) Improvements are underway to simulate crop residue burning in the DAYCENT model based on the
35 amount of crop residues burned according to the data that is used in the Field Burning of Agricultural
36 Residues source category (Section 5.5). See the Planned Improvement section in the Field Burning of
37 Agricultural Residues for more information.
38 (5) Alaska and Hawaii are not included in the current Inventory for agricultural soil management, with the
39 exception of N2O emissions from drained organic soils in croplands and grasslands for Hawaii. A
40 planned improvement over the next two years is to add these states into the Inventory analysis.
5-34 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
J
4
5
6
7
10
11
12
13
14
15
16
17
18
19
5.5 Field Burning of Agricultural Residues (IPCC
Source Category 3F)
Crop production creates large quantities of agricultural crop residues, which farmers manage in a variety of ways.
For example, crop residues can be left in the field and possibly incorporated into the soil with tillage; collected and
used as fuel, animal bedding material, supplemental animal feed, or construction material; composted and applied to
soils; transported to landfills; or burned in the field. Field burning of crop residues is not considered a net source of
CO2 emissions because the C released to the atmosphere as CCh during burning is reabsorbed during the next
growing season for the crop. However, crop residue burning is a net source of CH4, N2O, CO, and NOX, which are
released during combustion.
In the United States, field burning of agricultural residues commonly occurs in southeastern states, the Great Plains,
and the Pacific Northwest (McCarty 2011). The primary crops that are managed with residue burning include corn,
cotton, lentils, rice, soybeans, sugarcane, and wheat (McCarty 2009). Rice, sugarcane, and wheat residues account
for approximately 70 percent of all crop residue burning and emissions (McCarty 2011). In 2014, CH4 and N2O
emissions from Field Burning of Agricultural Residues were 0.3 MMT CO2 Eq. (11 kt) and 0.1 MMT. CO2 Eq. (0.3
kt), respectively. Furthermore, annual emissions from this source from 1990 to 2014 have remained relatively
constant, averaging approximately 0.3 MMT CO2 Eq. (10 kt) of CH4 and 0.1 MMT CO2 Eq. (0.3 kt) of N2O (see
Table 5-23 and Table 5-24).
Table 5-23: CH4 and NzO Emissions from Field Burning of Agricultural Residues (MMT COz
Eq.)
Gas/Crop Type
1990
2005
2010 2011
2012 2013
2014
CH4
Wheat
Rice
Sugarcane
Com
Cotton
Soybeans
Lentil
N2O
Wheat
Rice
Sugarcane
Com
Cotton
Soybeans
Lentil
Total
0.3
0.1
0.1
0.3
0.1
0.1
0.3
0.1
0.1
0.3
0.1
0.1
0.1
0.1
0.1
0.1
0.4
0.4
0.4
0.4
0.4
+ Does not exceed 0.05 MMT CO2 Eq.
Note: Totals may not sum due to independent rounding.
20 Table 5-24: ChU, NzO, CO, and NOX Emissions from Field Burning of Agricultural Residues
21 (kt}
Gas/Crop Type 1990
CH4
Wheat
Rice
Sugarcane
Corn
Soybeans
Lentil
Cotton
2005
2010 2011
2012 2013
11
5
2
1
1
1
11
5
2
1
1
1
11
5
2
1
1
1
11
5
2
1
1
1
2014
11
5
2
2
2
1
Agriculture 5-35
-------
N20
Wheat
Rice
Sugarcane
Corn
Cotton
Soybeans
Lentil
CO
NOx
229
7
232
7
233
8
237
8
238
8
+ Does not exceed 0.5 kt.
Note: Totals may not sum due to independent rounding
Methodology
2 A U.S.-specific Tier 2 method was used to estimate greenhouse gas emissions from Field Burning of Agricultural
3 Residues (for more details, see Box 5-3). In order to estimate the amounts of C and N released during burning, the
4 following equation was used:
6 C or N released = Ł for all crop types and states f
AB
8 where,
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
CAH x CP x RCR x DMF x BE x CE x (FC orFN)
Total area of crop burned, by state
Total area of crop harvested, by state
Annual production of crop in kt, by state
Amount of residue produced per unit of crop production
Amount of dry matter per unit of bio mass for a crop
Amount of C or N per unit of dry matter for a crop
The proportion of prefire fuel biomass consumed15
The proportion of C or N released with respect to the total amount of C or N
available in the burned material, respectively
Crop Production and Crop Area Harvested were available by state and year from USDA (2014) for all crops (except
rice in Florida and Oklahoma, as detailed below). The amount C or N released was used in the following equation
to determine the CH4, CO, N2O and NOX emissions from the field burning of agricultural residues:
CH4 and CO, or N2O and NOX Emissions from Field Burning of Agricultural Residues =
C or N Released x ER x CF
Area Burned (AB)
Crop Area Harvested (CAH)
Crop Production (CP)
Residue: Crop Ratio (RCR)
Dry Matter Fraction (DMF)
Fraction of C or N (FC or FN)
Burning Efficiency (BE)
Combustion Efficiency (CE)
where,
Emissions Ratio (ER)
Conversion Factor (CF)
= g CH4-C or CO-C/g C released, or g N2O-N or NOx-N/g N released
= conversion, by molecular weight ratio, of CH4-C to C (16/12), or CO-C to C
(28/12), or N2O-N to N (44/28), or NOX-N to N (30/14)
Box 5-3: Comparison of Tier 2 U.S. Inventory Approach and IPCC (2006) Default Approach
Emissions from Field Burning of Agricultural Residues were calculated using a Tier 2 methodology that is based on
method developed by the IPCC/UNEP/OECD/IEA (1997) and incorporates crop- and country-specific emission
factors and variables. The rationale for using the IPCC/UNEP/OECD/IEA (1997) approach rather than the method
In IPCC/UNEP/OECD/IEA (1997), the equation for C or N released contains the variable 'fraction oxidized in burning'. This
variable is equivalent to (burning efficiency x combustion efficiency).
5-36 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 provided in the 2006IPCC Guidelines is as follows: (1) the equations from both guidelines rely on the same
2 underlying variables (though the formats differ); (2) the IPCC (2006) equation was developed to be broadly
3 applicable to all types of biomass burning, and, thus, is not specific to agricultural residues; and (3) the IPCC (2006)
4 default factors are provided only for four crops (corn, rice, sugarcane, and wheat) while this Inventory includes
5 emissions from seven crops (corn, cotton, lentils, rice, soybeans, sugarcane, and wheat).
6 A comparison of the methods and factors used in: (1) The current Inventory and (2) the default IPCC (2006)
7 approach was undertaken in the 1990 through 2014 Inventory report to determine the difference in overall estimates
8 between the two approaches. To estimate greenhouse gas emissions from Field Burning of Agricultural Residue
9 using the IPCC (2006) methodology, the following equation—cf. IPCC (2006) Equation 2.27—was used:
10 Emissions (kt) = AB x (MBx Cf) x Gef x lO"6
11 where,
12 Area Burned (AB) = Total area of crop burned (ha)
13 Mass Burned (MB x Cf) = IPCC (2006) default fuel biomass consumption (metric tons dry matter burnt
14 ha"1)
15 Emission Factor (Gef) = IPCC (2006) emission factor (g kg"1 dry matter burnt)
16 The IPCC (2006) default approach resulted in 5 percent higher emissions of CH4 and 21 percent higher emissions of
17 N2O compared to this Inventory (and are within the uncertainty ranges estimated for this source category). The
18 IPCC/UNEP/OECD/IEA (1997) is considered a more appropriate method for U.S. conditions because it is more
19 flexible for incorporating country-specific data compared to IPCC (2006) approach for Tier 1 and 2 methods.
20
21 Crop yield data (except rice in Florida) were based on USDA's QuickStats (USDA 2015), and crop area data were
22 based on the 2010 National Resources Inventory (NRI) (USDA 2013). In order to estimate total crop production,
23 the crop yield data from USDA Quick Stats crop yields was multiplied by the NRI crop areas. Rice yield data for
24 Florida was estimated separately because yield data were not collected by USDA. Total rice production for Florida
25 was determined using NRI crop areas and total yields were based on average primary and ratoon rice yields from
26 Schueneman and Deren (2002). Relative proportions of ratoon crops were derived from information in several
27 publications (Schueneman 1999, 2000, 2001; Deren 2002; Kirstein 2003, 2004; Cantens 2004, 2005; Gonzalez 2007
28 through 2014). The production data for the crop types whose residues are burned are presented in Table 5-25. Crop
29 weight by bushel was obtained from Murphy (1993).
30 The fraction of crop area burned was calculated using data on area burned by crop type and state16 from McCarty
31 (2010) for corn, cotton, lentils, rice, soybeans, sugarcane, and wheat.17 McCarty (2010) used remote sensing data
32 from Moderate Resolution Imaging Spectroradiometer (MODIS) to estimate area burned by crop. State-level area
33 burned data were divided by state-level crop area harvested data to estimate the percent of crop area burned by crop
34 type for each state. The average percentage of crop area burned at the national scale is shown in Table 5-26. Data
35 on fraction of crop area burned were only available from McCarty (2010) for the years 2003 through 2007. For
36 other years in the time series, the percent area burned was set equal to the average over the five-year period from
37 2003 to 2007. Table 5-26 shows the resulting percentage of crop residue burned at the national scale by crop type.
38 State-level estimates are also available upon request.
39 All residue: crop product mass ratios except sugarcane and cotton were obtained from Strehler and Stutzle (1987).
40 The ratio for sugarcane is from Kinoshita (1988) and the ratio for cotton is from Huang et al. (2007). The residue:
41 crop ratio for lentils was assumed to be equal to the average of the values for peas and beans. Residue dry matter
42 fractions for all crops except soybeans, lentils, and cotton were obtained from Turn et al. (1997). Soybean and lentil
43 dry matter fractions were obtained from Strehler and Stutzle (1987); the value for lentil residue was assumed to
16 Alaska and Hawaii were excluded.
^ McCarty (2009) also examined emissions from burning of Kentucky bluegrass and a general "other crops/fallow" category,
but USDA crop area and production data were insufficient to estimate emissions from these crops using the methodology
employed in the Inventory. McCarty (2009) estimates that approximately 18 percent of crop residue emissions result from
burning of the Kentucky bluegrass and "other crops" categories.
Agriculture 5-37
-------
1 equal the value for bean straw. The cotton dry matter fraction was taken from Huang et al. (2007). The residue C
2 contents and N contents for all crops except soybeans and cotton are from Turn et al. (1997). The residue C content
3 for soybeans is the IPCC default (IPCC/UNEP/OECD/IEA 1997), and the N content of soybeans is from Barnard
4 and Kristoferson (1985). The C and N contents of lentils were assumed to equal those of soybeans. The C and N
5 contents of cotton are from Lachnicht et al. (2004). The burning efficiency was assumed to be 93 percent, and the
6 combustion efficiency was assumed to be 88 percent, for all crop types, except sugarcane (EPA 1994). For
7 sugarcane, the burning efficiency was assumed to be 81 percent (Kinoshita 1988) and the combustion efficiency was
8 assumed to be 68 percent (Turn et al. 1997). See Table 5-27 for a summary of the crop-specific conversion factors.
9 Emission ratios and mole ratio conversion factors for all gases were based on the Revised 1996 IPCC Guidelines
10 (IPCC/UNEP/OECD/IEA 1997) (see Table 5-28).
11 Table 5-25: Agricultural Crop Production (kt of Product)
Crop
Corna
Cotton
Lentils
Rice
Soybeans
Sugarcane
Wheat
1990
229,152
4,446
38
8,903
55,129
31,827
79,011
"
2010
2011
2012
2013
2014
500,826
6,811
248
12,577
86,903
32,496
70,074
335,526
4,814
406
11,372
94,445
30,333
71,017
321,791
4,369
234
11,791
90,746
32,469
62,131
270,180
5,156
251
12,543
86,910
34,925
71,094
350,338
4,841
271
12,928
95,454
34,186
68,772
378,432
5,104
156
12,869
103,570
34,160
64,748
a Com for grain (i.e., excludes com for silage).
12
13
14
15
Table 5-26: U.S. Average Percent Crop Area Burned by Crop (Percent)
State
Com
Cotton
Lentils
Rice
Soybeans
Sugarcane
Wheat
1990
+
1%
2%
9%
10%
2%
2005
+
1 %
+
5%
14%
2%
2010
+
Io/
/o
+
70/
/o
23%
20/
/o
2011
+
1 %
1%
7%
25%
3%
2012
+
Io/
/O
1 %
70/
/o
23%
20/
/o
2013
+
1 %
1%
7%
22%
2%
2014
+
1 %
1%
7%
24%
2%
+ Does not exceed 0.5 percent
Table 5-27: Key Assumptions for Estimating Emissions from Field Burning of Agricultural
Residues
Crop
Com
Cotton
Lentils
Rice
Soybeans
Sugarcane
Wheat
Residue: Crop
Ratio
1.0
1.6
2.0
1.4
2.1
0.2
1.3
Dry Matter
Fraction
0.91
0.90
0.85
0.91
0.87
0.62
0.93
C Fraction
0.448
0.445
0.450
0.381
0.450
0.424
0.443
N Fraction
0.006
0.012
0.023
0.007
0.023
0.004
0.006
Burning
Efficiency
(Fraction)
0.93
0.93
0.93
0.93
0.93
0.81
0.93
Combustion
Efficiency
(Fraction)
0.88
0.88
0.88
0.88
0.88
0.68
0.88
Table 5-28: Greenhouse Gas Emission Ratios and Conversion Factors
Gas
CH4:C
CO:C
N20:N
NOX:N
Emission Ratio
0.005a
0.060a
0.007b
0.121b
Conversion Factor
16/12
28/12
44/28
30/14
a Mass of C compound released (units of C) relative to
mass of total C released from burning (units of C).
5-38 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
b Mass of N compound released (units of N) relative to
mass of total N released from burning (units of N).
i Uncertainty and Time-Series Consistency
2 The results of the Approach 2 Monte Carlo uncertainty analysis are summarized in Table 5-29. CH4 emissions from
3 Field Burning of Agricultural Residues in 2014 were estimated to be between 0.18 and 0.44 MMT CChEq. at a 95
4 percent confidence level. This indicates a range of 41 percent below and 42 percent above the 2014 emission
5 estimate of 0.3 MMT CCh Eq. Also at the 95 percent confidence level, N2O emissions were estimated to be between
6 0.07 and 0.14 MMT €62 Eq., or approximately 30 percent below and 31 percent above the 2014 emission estimate
7 of0.1MMTCO2Eq.
8 Table 5-29: Approach 2 Quantitative Uncertainty Estimates for CH4 and NzO Emissions from
9 Field Burning of Agricultural Residues (MMT COz Eq. and Percent)
2014 Emission
Source Gas Estimate
(MMT CO2 Eq.)
Field Burning of Agricultural „„ _ ,
Residues
Field Burning of Agricultural N _ „.
Residues
Uncertainty Range Relative to Emission Estimate
(MMT C02 Eq.) (%)
Lower
Bound
0.18
0.07
Upper
Bound
0.44
0.14
Lower
Bound
-41%
-30%
Upper
Bound
42%
31%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
10 Due to data limitations, there are additional uncertainties in agricultural residue burning due to the omission of
11 burning associated with Kentucky bluegrass and "other crop" residues.
12 Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
13 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
14 above
15
16
17
18
QA/QC and Verification
A source-specific QA/QC plan for Field Burning of Agricultural Residues was implemented with Tier 1 and 2
analyses. The Tier 1 analysis conducted this year uncovered a data transcription error in the corn production data
for 1990. No other errors were found.
19 Recalculations Discussion
20 The source data for crop areas was changed from USD A NASS QuickStats to the 2010 National Resources
21 Inventory (NRI). This change ensures greater consistency across cropland source categories, including direct and
22 indirect soil nitrous oxide emissions in Section 5.4 Agricultural Soil Management, and soil carbon stock changes in
23 the Cropland Remaining Cropland and Land Converted to Cropland sections, which also rely on the NRI data as the
24 basis for crop areas. The NRI data were used to recalculate percent crop area burned and total crop production. This
25 change resulted in higher crop production estimates (ranging from 4 to 40 percent) and lower burned area
26 percentages (ranging from -2 to -42 percent) relative to the previous method. However, the overall impact on the
27 recalculated emissions was relatively small, with CH4 and N2O emissions decreasing by 13 and 8 percent
28 respectively. In addition, correcting a transcription error in crop production for corn in 1990 (Table 5-25) led to a
29 larger recalculation in emissions for 1990 relative to the other years.
Agriculture 5-39
-------
Planned Improvements
2 A new method is in development that will directly link agricultural residue burning with the Tier 3 methods that are
3 used in the Agricultural Soil Management, Cropland Remaining Cropland, and Land Converted to Cropland
4 chapters of the Inventory. The method is based on the DAYCENT model, and burning events will be simulated
5 directly within the process-based model framework using information derived from remote sensing fire products.
6 This improvement will lead to greater consistency in the methods for these sources, and better ensure mass balance
7 of C and N in the inventory analysis.
5-40 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
i 6. Land Use, Land-Use Change, and
Forestry
3 This chapter provides an assessment of the net greenhouse gas flux resulting from the use and conversion of land-
4 use categories in the United States.1 The Intergovernmental Panel on Climate Change 2006 Guidelines for National
5 Greenhouse Gas Inventories (IPCC 2006) recommends reporting fluxes according to changes within and
6 conversions between certain land-use types termed: Forest Land, Cropland, Grassland, Settlements, Wetlands (as
7 well as Other Land). The greenhouse gas flux from Forest Land Remaining Forest Land is reported using estimates
8 of changes in forest ecosystem carbon (C) stocks, harvested wood pools, non-carbon dioxide (non-CCh) emissions
9 from forest fires, and the application of synthetic fertilizers to forest soils. Fluxes are also included for forest
10 ecosystem pools for Land Converted to Forest Land. 2 The greenhouse gas flux from agricultural lands (i.e.,
11 Cropland and Grassland) that is reported in this chapter includes changes in organic C stocks in mineral and organic
12 soils due to land use and management, and emissions of CCh due to the application of crushed limestone and
13 dolomite to managed land (i.e., soil liming) and urea fertilization.3 Fluxes are reported for four agricultural land
14 use/land-use change categories: Cropland Remaining Cropland, Land Converted to Cropland, Grassland Remaining
15 Grassland, and Land Converted to Grassland. Fluxes from Wetlands Remaining Wetlands include CO2, CEL and
16 N2O emissions from managed peatlands; estimates for Land Converted to Wetlands are currently not available.
17 Fluxes resulting from Settlements Remaining Settlements include those from urban trees and soil fertilization; fluxes
18 from Land Converted to Settlement are currently not available. Landfilled yard trimmings and food scraps are
19 accounted for separately under Other.
20 Land use, land-use change, and forestry activities in 2014 resulted in a C sequestration (i.e., total sinks) of 685.8
21 MMT CO2 Eq.4 (187.0 MMT C).5 This represents an offset of approximately 10.0 percent of total (i.e., gross)
22 greenhouse gas emissions in 2014. Emissions from land use, land-use change, and forestry activities in 2014
23 represent 0.4 percent of total greenhouse gas emissions.6
24 Total land use, land-use change, and forestry C sequestration decreased by approximately 2.6 percent between 1990
25 and 2014. This increase was primarily due to a decrease in the rate of net C accumulation in agricultural soil carbon
The term "flux" is used to describe the net emissions of greenhouse gases to the atmosphere accounting for both the emissions
of CO2 to and the removals of CCh from the atmosphere. Removal of CCh from the atmosphere is also referred to as "carbon
sequestration".
Estimates from Land Converted to Forest Land are currently under development.
3 Direct and indirect emissions of N2O from inputs of N to cropland and grassland soils are included in the Agriculture Chapter.
4 Following the revised reporting requirements under the UNFCCC, this Inventory report presents CCh equivalent values based
on the IPCC Fourth Assessment Report (AR4) GWP values. See the Introduction chapter for more information.
^ Net flux from LULUCF includes the positive C sequestration reported for Forest Land Remaining Forest Land, Land
Converted to Forest Land, Cropland Remaining Cropland, Land Converted to Grassland, Settlements Remaining Settlements,
and Other Land plus the loss in C sequestration reported for Land Converted to Cropland and Grassland Remaining Grassland.
LULUCF emissions include the CCh, CELi, andN2O emissions reported for Forest Fires, Forest Soils, Liming, Urea
Fertilization, Settlement Soils, and Peatlands Remaining Peatlands.
Land Use, Land-Use Change, and Forestry 6-1
-------
1
2
o
J
4
5
6
7
stocks.7 Net C accumulation in Forest Land Remaining Forest Land, Land Converted to Grassland, and Settlements
Remaining Settlements increased, while net C accumulation in Cropland Remaining Cropland, Grassland
Remaining Grassland, and Landfilled Yard Trimmings and Food Scraps slowed over this period. Emissions from
Land Converted to Cropland and Wetlands Remaining Wetlands decreased. Emissions and removals for Land Use,
Land-Use Change, and Forestry are summarized in Table 6-1 by land-use and category.
Table 6-1: Emissions and Removals (Flux) from Land Use, Land-Use Change, and Forestry by
Land Use and Land-Use Change Category (MMT COz Eq.)
Land-Use/Source Category
Forest Land Remaining Forest Land
Changes in Forest Carbon Stocka
Forest Fires
Forest Soilsb
Land Converted to Forest Land
Changes in Forest Carbon Stock
Cropland Remaining Cropland
Changes in Agricultural Soil Carbon Stock
Liming
Urea Fertilization
Land Converted to Cropland
Changes in Agricultural Soil Carbon Stock
Grassland Remaining Grassland
Changes in Agricultural Soil Carbon Stock
Land Converted to Grassland
Changes in Agricultural Soil Carbon Stock
Settlements Remaining Settlements
Changes in Urban Tree Carbon Stock0
Settlement Soilsd
Wetlands Remaining Wetlands
Peatlands Remaining Peatlands
Other
Landfilled Yard Trimmings and Food
Scraps
LULUCF Emissions6
LULUCF Total Net Fluxf
LULUCF Sector Total?
1990
(570.5)
(576.0)
5.4
0.1
(36.1)
(43.2)
4.7
2.4 1
22.8
22.8 1
(12.9)
(12.9)
(8.5)
(8.5) 1
(59.0)
(60.4)
1.4
1.1 1
1.1
(26.0)
(26.0)
15.0
(704.2)
(689.1)
2005
(515.5)
(532.4) 1
16.5
0.5
(8.7)
(16.5) 1
4.3
3.5
14.6
14.6
12.9
2.9 1
(12.8)
(12.8)
(78.2)
(80.5)
2.3
1.1
(1 1.4) 1
(11.4)
28.2
(636.1) •
(607.9)
2010
(579.1)
(585.0)
5.4
0.5
3.9
(4.7)
4.8
3.8
15.6
15.6
2.6
2.6
(12.3)
(12.3)
(83.8)
(86.1)
2.4
1.0
1.0
(13.2)
(13.2)
17.8
(683.2)
(665.3)
2011
(566.6)
(578.1)
11.0
0.5
(12.1)
(20.0)
3.9
4.1
14.2
14.2
11.3
11.3
(11.0)
(11.0)
(84.8)
(87.3)
2.5
0.9
0.9
(12.7)
(12.7)
22.9
(683.6)
(660.7)
2012
(558.0)
(576.7)
18.3
0.5
(8.5)
(18.7)
6.0
4.2
14.5
14.5
11.7
11.7
(10.9)
(10.9)
(85.8)
(88.4)
2.5
0.8
0.8
(12.2)
(12.2)
32.3
(680.8)
(648.5)
2013
(567.5)
(580.1)
12.2
0.5
(8.6)
(16.8)
3.9
4.3
14.8
14.8
11.9
11.9
(10.9)
(10.9)
(87.1)
(89.5)
2.4
0.8
0.8
(11.7)
(11.7)
24.1
(682.4)
(658.3)
2014
(570.7)
(583.4)
12.2
0.5
(7.3)
(16.0)
4.1
4.5
14.7
14.7
11.9
11.9
(10.9)
(10.9)
(88.2)
(90.6)
2.4
0.8
0.8
(11.6)
(11.6)
24.6
(685.8)
(661.3)
9
10
Note: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
a Includes the effects of net additions to stocks of carbon stored in forest ecosystem pools and harvested wood products.
b Estimates include emissions from N fertilizer additions on both Forest Land Remaining Forest Land, and Land Converted to
Forest Land, but not from land-use conversion.
0 Estimates include C stock changes on both Settlements Remaining Settlements and Land Converted to Settlements.
d Estimates include emissions from N fertilizer additions on both Settlements Remaining Settlements, and Land Converted to
Settlements, but not from land-use conversion.
e LULUCF emissions include the CCh, CELi, andN2O emissions reported for Forest Fires, Forest Soils, Liming, Urea
Fertilization, Settlement Soils, and Peatlands Remaining Peatlands.
f Total net flux from LULUCF are only included in the Net Emissions total. Net flux from LULUCF includes the positive C
sequestration reported for Forest Land Remaining Forest Land, Land Converted to Forest Land, Cropland Remaining
Cropland, Land Converted to Grassland, Settlements Remaining Settlements, and Other Land plus the loss in C sequestration
reported for Land Converted to Cropland and Grassland Remaining Grassland.
g The LULUCF Sector Total is sum of positive emissions (i.e., sources) of greenhouse gases to the atmosphere plus removals
of CO2 (i.e., sinks or negative emissions) from the atmosphere.
Carbon dioxide removals are presented in Table 6-2 along with CCh, CH4, and N2O emissions from Land use, Land-
Use Change, and Forestry source categories. Liming and urea fertilization in 2014 resulted in CC>2 emissions of 8.7
MMT CO2 Eq. (8,653 kt of CCh). Lands undergoing peat extraction (i.e., Peatlands Remaining Peatlands) resulted
7 Carbon sequestration estimates are net figures. The C stock in a given pool fluctuates due to both gains and losses. When
losses exceed gains, the C stock decreases, and the pool acts as a source. When gains exceed losses, the C stock increases, and
the pool acts as a sink; also referred to as net C sequestration or removal.
6-2 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 in CO2 emissions of 0.8 MMT CO2 Eq. (842 kt of CO2), methane (CH4) emissions of less than 0.05 MMT CO2 Eq.,
2 and nitrous oxide (N2O) emissions of less than 0.05 MMT CO2 Eq. The application of synthetic fertilizers to forest
3 soils in 2014 resulted in N2O emissions of 0.5 MMT CO2 Eq. (2 kt of N2O). Nitrous oxide emissions from fertilizer
4 application to forest soils have increased by 455 percent since 1990, but still account for a relatively small portion of
5 overall emissions. Additionally, N2O emissions from fertilizer application to settlement soils in 2014 accounted for
6 2.4 MMT CO2 Eq. (8 kt of N2O). This represents an increase of 78 percent since 1990. Forest fires in 2014 resulted
7 in CH4 emissions of 7.3 MMT CO2 Eq. (294 kt of N2O), and inN2O emissions of 4.8 MMT CO2 Eq. (16 kt of N2O).
8 Emissions and removals for Land Use, Land-Use Change, and Forestry are shown in Table 6-2 and Table 6-3.
9 Table 6-2: Emissions and Removals (Flux) from Land Use, Land-Use Change, and Forestry by
10 Gas(MMTCOzEq.)
Gas/Land-Use Category
Net CO2 Flux3
Forest Land Remaining Forest Landb
Land Converted to Forest Land
Cropland Remaining Cropland
Land Converted to Cropland*
Grassland Remaining Grassland*
Land Converted to Grassland
Settlements Remaining Settlements0
Other: Landfilled Yard Lrimmings and
Food Scraps
C02
Cropland Remaining Cropland: Liming
Cropland Remaining Cropland: Urea
Fertilization
Wetlands Remaining Wetlands:
Peatlands Remaining Peatlands
CH4
Forest Land Remaining Forest Land:
Forest Fires
Wetlands Remaining Wetlands:
Peatlands Remaining Peatlands
N20
Forest Land Remaining Forest Land:
Forest Fires
Forest Land Remaining Forest Land:
Forest Soilsd
Settlements Remaining Settlements:
Settlement Soils6
Wetlands Remaining Wetlands:
Peatlands Remaining Peatlands
LULUCF Emissions'
LULUCF Total Net Flux3
LULUCF Sector Total?
1990
(704.2)
(576.0)
(43.2)
22.8
(12.9)
(8.5)
(60.4)
(26.0)
8.1
4.7 1
2.4 1
1.1 1
3.3 1
3.3 1
+ 1
3.6 1
2.2 1
0.1 1
1.4 1
15.0
(704.2)
(689.1)
2005 2010
(636.1) (683.2)
(532.4) 1 (585.0)
(16.5)
14.6
2.9
(12.8)
(80.5)
(11.4)
9.0
4.3
3.5
1.1
9.9
9.9
+
9.3
6.5
0.5
2.3
(4.7)
15.6
2.6
(12.3)
(86.1)
(13.2)
9.6
4.8
3.8
1.0
3.3
3.3
+
5.0
2.2
0.5
2.4
| + +
28.2 17.8
(636.1) (683.2)
(607.9) (665.3)
2011
(683.6)
(578.1)
(20.0)
14.2
11.3
(11.0)
(87.3)
(12.7)
8.9
3.9
4.1
0.9
6.6
6.6
+
7.3
4.4
0.5
2.5
+
22.9
(683.6)
(660.7)
2012
(680.8)
(576.7)
(18.7)
14.5
11.7
(10.9)
(88.4)
(12.2)
11.0
6.0
4.2
0.8
11.1
11.1
+
10.3
7.3
0.5
2.5
+
32.3
(680.8)
(648.5)
2013
(682.4)
(580.1)
(16.8)
14.8
11.9
(10.9)
(89.5)
(11.7)
9.0
3.9
4.3
0.8
7.3
7.3
+
7.7
4.8
0.5
2.4
+
24.1
(682.4)
(658.3)
2014
(685.8)
(583.4)
(16.0)
14.7
11.9
(10.9)
(90.6)
(11.6)
9.5
4.1
4.5
0.8
7.4
7.3
+
7.7
4.8
0.5
2.4
+
24.6
(685.8)
(661.3)
Note: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
+ Does not exceed 0.05 MMT CO2 Eq.
a Total net flux from LULUCF is only included in the Net Emissions total. Net flux from LULUCF includes the positive C
sequestration reported for Forest Land Remaining Forest Land, Land Converted to Forest Land, Cropland Remaining
Cropland, Land Converted to Grassland, Settlements Remaining Settlements, and Other Land plus the loss in C
sequestration reported for Land Converted to Cropland and Grassland Remaining Grassland.
b Includes the effects of net additions to stocks of carbon stored in forest ecosystem pools and harvested wood products.
0 Estimates include C stock changes on both Settlements Remaining Settlements and Land Converted to Settlements.
d Estimates include emissions from N fertilizer additions on both Forest Land Remaining Forest Land, and Land Converted
to Forest Land, but not from land-use conversion.
e Estimates include emissions from N fertilizer additions on both Settlements Remaining Settlements, and Land Converted to
Settlements, but not from land-use conversion.
f LULUCF emissions include the CCh, CELi, andN2O emissions reported for Forest Fires, Forest Soils, Liming, Urea
Fertilization, Settlement Soils, and Peatlands Remaining Peatlands.
Land Use, Land-Use Change, and Forestry 6-3
-------
g The LULUCF Sector Total is the sum of positive emissions (i.e., sources) of greenhouse gases to the atmosphere plus
removals of CCh (i.e., sinks or negative emissions) from the atmosphere.
1 Table 6-3: Emissions and Removals (Flux) from Land Use, Land-Use Change, and Forestry by
2 Gas (kt)
Gas/Land-Use Category
1990
2005
2010
2011
2012
2013
2014
(636,143)
(683,157) (683,557) (680,812) (682,365)
(532,408) I I (584,984) (578,060) (576,738) (580,109)
(4,699)
15,621
2,580
(12,346)
(20,042)
14,160
11,259
(10,964)
(18,739)
14,505
11,706
(10,932)
(86,129) (87,250)
(13,200)
9,584
4,784
3,778
1,022
131
131
17
(12,659)
8,898
3,873
4,099
926
265
265
25
15
(12,242)
11,015
5,978
4,225
812
443
443
34
24
(16,833)
14,754
11,924
(10,909)
3,372) (89,493) (90,614)
Net CO2 Flux3
Forest Land Remaining Forest
Landb
Land Converted to Forest Land
Cropland Remaining Cropland
Land Converted to Cropland*
Grassland Remaining Grassland*
Land Converted to Grassland
Settlements Remaining
Settlements0
Other: Landfilled Yard Trimmings
and Food Scraps
C02
Cropland Remaining Cropland:
Liming
Cropland Remaining Cropland:
Urea Fertilization
Wetlands Remaining Wetlands:
Peatlands Remaining Peatlands
CH4
Forest Land Remaining Forest
Land: Forest Fires
Wetlands Remaining Wetlands:
Peatlands Remaining Peatlands
N2O
Forest Land Remaining Forest
Land: Forest Fires
Forest Land Remaining Forest
Land: Forest Soils'1
Settlements Remaining Settlements:
Settlement Soils6
Wetlands Remaining Wetlands:
Peatlands Remaining Peatlands
+ Does not exceed 0.5 kt
a Net flux from LULUCF includes the positive C sequestration reported for Forest Land Remaining Forest Land, Land
Converted to Forest Land Cropland Remaining Cropland, Land Converted to Grassland, Settlements Remaining Settlements,
and Other Land plus the loss in C sequestration reported for Land Converted to Cropland and Grassland Remaining Grassland.
b Includes the effects of net additions to stocks of carbon stored in forest ecosystem pools and harvested wood products.
c Estimates include C stock changes on both Settlements Remaining Settlements and Land Converted to Settlements.
d Estimates include emissions from N fertilizer additions on both Forest Land Remaining Forest Land, and Land Converted to
Forest Land, but not from land-use conversion.
e Estimates include emissions from N fertilizer additions on both Settlements Remaining Settlements, and Land Converted to
Settlements, but not from land-use conversion.
(704,191)
(576,023)
(43,199)
22,812
(12,906)
(8,491)
(60,408)
(25,975)
8,139
4,667
2,417
1,055
131
131
12
(16,547)
14,646
2,888
(12,838)
(80,523)
(11,360)
8,955
4,349
3,504
1,101
397
397
31
22
(11,698)
9,021
3,909
4,342
770
294
294
26
16
(685,827)
(583,385)
(15,988)
14,726
11,926
(10,907)
(11,585)
9,495
4,139
4,514
842
294
294
+
26
16
2
Box 6-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Sinks
4 In following the UNFCCC requirement under Article 4.1 to develop and submit national greenhouse gas emissions
5 inventories, the emissions and sinks presented in this report are organized by source and sink categories and
6 calculated using internationally-accepted methods provided by the Intergovernmental Panel on Climate Change
6-4 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 (IPCC).8 Additionally, the calculated emissions and sinks in a given year for the United States are presented in a
2 common manner in line with the UNFCCC reporting guidelines for the reporting of inventories under this
3 international agreement.9 The use of consistent methods to calculate emissions and sinks by all nations providing
4 their inventories to the UNFCCC ensures that these reports are comparable. In this regard, U.S. emissions and sinks
5 reported in this Inventory report are comparable to emissions and sinks reported by other countries. The manner that
6 emissions and sinks are provided in this Inventory is one of many ways U.S. emissions and sinks could be
7 examined; this Inventory report presents emissions and sinks in a common format consistent with how countries are
8 to report inventories under the UNFCCC. The report itself follows this standardized format, and provides an
9 explanation of the IPCC methods used to calculate emissions and sinks, and the manner in which those calculations
10 are conducted.
11
12
6.1 Representation of the U.S. Land Base
13 A national land-use categorization system that is consistent and complete, both temporally and spatially, is needed in
14 order to assess land use and land-use change status and the associated greenhouse gas fluxes over the Inventory time
15 series. This system should be consistent with IPCC (2006), such that all countries reporting on national greenhouse
16 gas fluxes to the UNFCCC should: (1) describe the methods and definitions used to determine areas of managed
17 and unmanaged lands in the country (Table 6-4), (2) describe and apply a consistent set of definitions for land-use
18 categories over the entire national land base and time series (i.e., such that increases in the land areas within
19 particular land-use categories are balanced by decreases in the land areas of other categories unless the national land
20 base is changing) (Table 6-5), and (3) account for greenhouse gas fluxes on all managed lands. The IPCC (2006,
21 Vol. IV, Chapter 1) considers all anthropogenic greenhouse gas emissions and removals associated with land use
22 and management to occur on managed land, and all emissions and removals on managed land should be reported
23 based on this guidance (see IPCC 2010 for further discussion). Consequently, managed land serves as a proxy for
24 anthropogenic emissions and removals. This proxy is intended to provide a practical framework for conducting an
25 inventory, even though some of the greenhouse gas emissions and removals on managed land are influenced by
26 natural processes that may or may not be interacting with the anthropogenic drivers. Guidelines for factoring out
27 natural emissions and removals may be developed in the future, but currently the managed land proxy is considered
28 the most practical approach for conducting an inventory in this sector (IPCC 2010). The implementation of such a
29 system helps to ensure that estimates of greenhouse gas fluxes are as accurate as possible, and does allow for
30 potentially subjective decisions in regards to subdividing natural and anthropogenic driven emissions. This section
31 of the Inventory has been developed in order to comply with this guidance.
32 Three databases are used to track land management in the United States and are used as the basis to classify U.S.
33 land area into the thirty-six IPCC land-use and land-use change categories (Table 6-5) (IPCC 2006). The primary
34 databases are the U.S. Department of Agriculture (USD A) National Resources Inventory (NRI)10 and the USD A
35 Forest Service (USFS) Forest Inventory and Analysis (FIA)11 Database. The Multi-Resolution Land Characteristics
36 Consortium (MRLC) National Land Cover Dataset (NLCD)12 is also used to identify land uses in regions that were
37 not included in the NRI or FIA.
38 The total land area included in the U.S. Inventory is 936 million hectares across the 50 states.13 Approximately 890
39 million hectares of this land base is considered managed and 46 million hectares is unmanaged, which has not
8 See .
9 See .
10 NRI data is available at .
1! FIA data is available at .
12 NLCD data is available at and MRLC is a consortium of several U.S. government agencies.
The current land representation does not include areas from U.S. Lerritories, but there are planned improvements to include
these regions in future reports.
Land Use, Land-Use Change, and Forestry 6-5
-------
1
2
o
6
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
changed by much over the time series of the Inventory (Table 6-5). In 2014, the United States had a total of 295
million hectares of managed Forest Land (3.2 percent increase since 1990), 164 million hectares of Cropland (6.3
percent decrease since 1990), 321 million hectares of managed Grassland (1.7 percent decrease since 1990), 42
million hectares of managed Wetlands (7.2 percent decrease since 1990), 43 million hectares of Settlements (28
percent increase since 1990), and 25 million hectares of managed Other Land (Table 6-5). Wetlands are not
differentiated between managed and unmanaged, and are reported solely as managed. In addition, C stock changes
are not currently estimated for the entire land base, which leads to discrepancies between the managed land area data
presented here and in the subsequent sections of the Inventory (e.g., Grassland Remaining Grassland, interior
Alaska).14 Planned improvements are under development to account for C stock changes on all managed land (e.g.,
grasslands and forest lands in Alaska) and ensure consistency between the total area of managed land in the land-
representation description and the remainder of the Inventory.
Dominant land uses vary by region, largely due to climate patterns, soil types, geology, proximity to coastal regions,
and historical settlement patterns, although all land uses occur within each of the 50 states (Table 6-4). Forest Land
tends to be more common in the eastern states, mountainous regions of the western United States, and Alaska.
Cropland is concentrated in the mid-continent region of the United States, and Grassland is more common in the
western United States and Alaska. Wetlands are fairly ubiquitous throughout the United States, though they are
more common in the upper Midwest and eastern portions of the country. Settlements are more concentrated along
the coastal margins and in the eastern states.
Table 6-4: Managed and Unmanaged Land Area by Land-Use Categories for All 50 States
(Thousands of Hectares)
Land-Use Categories
Managed Lands
Forest Land
Croplands
Grasslands
Settlements
Wetlands
Other Land
Unmanaged Lands
Forest Land
Croplands
Grasslands
Settlements
Wetlands
Other Land
Total Land Areas
Forest Land
Croplands
Grasslands
Settlements
Wetlands
Other Land
Table 6-5: Land Use
1990
890,019
285,837
174,678
326,526
33,420
45,361
24,197
46,211
9,634 •
0
25,782
"1
10,795
936,230
295,471
174,678
352,308
33,420
45,361
34,992
2005
890,016
292,106
166,064
323,239
40,450
43,004
25,154
46,214
9,634 •
0
25,782
"1
10,797
936,230
301,740
166,064
349,021
40,450
43,004
35,951
and Land-Use Change
2010
890,017
294,175
163,745
321,717
42,645
42,336
25,398
46,213
9,634
0
25,782
0
0
10,797
936,230
303,810
163,745
347,499
42,645
42,336
36,195
for the
2011
890,017
294,585
163,745
321,421
42,645
42,223
25,398
46,213
9,634
0
25,782
0
0
10,797
936,230
304,219
163,745
347,203
42,645
42,223
36,195
2012
890,017
294,988
163,752
321,118
42,648
42,113
25,399
46,213
9,634
0
25,782
0
0
10,797
936,230
304,622
163,752
346,900
42,648
42,113
36,196
2013
890,017
294,988
163,752
321,118
42,648
42,112
25,399
46,213
9,634
0
25,782
0
0
10,797
936,230
304,622
163,752
346,900
42,648
42,112
36,196
U.S. Managed Land Base
2014
890,017
294,988
163,752
321,118
42,648
42,113
25,399
46,213
9,634
0
25,782
0
0
10,797
936,230
304,622
163,752
346,900
42,648
42,113
36,196
for All 50 States
(Thousands of Hectares)
Land-Use & Land-
Use Change
Categories3
Total Forest Land
FF
LJJV
285,837
284,642
*„„.,
292,106
291,098
2010
294,175
293,234
2011
294,585
293,644
2012
294,988
294,051
2013
294,988
294,051
2014
294,988
294,051
14 These "managed area" discrepancies also occur in the Common Reporting Format (CRF) tables submitted to the UNFCCC.
6-6 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
CF
GF
WF
SF
OF
Total Cropland
CC
FC
GC
we
SC
oc
Total Grassland
GG
FG
CG
WG
SG
OG
Total Wetlands
WW
FW
CW
GW
SW
OW
Total Settlements
SS
FS
CS
GS
WS
OS
Total Other Land
00
FO
CO
GO
WO
SO
Grand Total
233
841
20
15
86 1
174,678
161,960
252^
12,066
141 1
^
182
326,526
316,489
8991
8,396 1
283
53
406
45,361
44,649
214
396
2
63 1
33,420
30,632
2321
l,227l
l,268l
«
55 •
24,197
23,162
37 B
328
531
135
4
890,019
215
635
23
15
120 •
166,064
151,903
91 1
13,581
166 1
78
245
323,239
303,987
1,538
16,335
437
115
827
43,004
41,785
41 1
362
770
1 1
45 1
40,450
32,188
339l
3,530 1
4,164 1
26
201
25,154
23,312
54 1
706
966
109
7
890,016
189
637
23
16
77
163,745
152,079
48
11,215
114
72
217
321,717
303,284
1,481
15,776
250
119
806
42,336
41,280
35
321
661
2
38
42,645
34,870
362
3,205
3,981
24
204
25,398
23,475
61
812
969
70
12
890,017
189
637
23
16
77
163,745
152,079
48
11,215
114
72
217
321,421
302,989
1,481
15,776
250
119
806
42,223
41,167
35
321
661
2
38
42,645
34,870
362
3,205
3,981
24
204
25,398
23,475
61
812
969
70
12
890,017
183
638
23
15
77
163,752
152,084
49
11,215
114
72
217
321,118
302,687
1,479
15,776
250
119
806
42,113
41,056
35
321
661
2
38
42,648
34,870
365
3,205
3,981
24
204
25,399
23,476
61
812
969
70
12
890,017
183
638
23
15
77
163,752
152,084
49
11,215
114
72
217
321,118
302,688
1,479
15,776
250
119
806
42,112
41,056
35
321
661
2
38
42,648
34,870
365
3,205
3,981
24
204
25,399
23,476
61
812
969
70
12
890,017
183
638
23
15
77
163,752
152,084
49
11,215
114
72
217
321,118
302,687
1,479
15,776
250
119
806
42,113
41,056
35
321
661
2
38
42,648
34,870
365
3,205
3,981
24
204
25,399
23,476
61
812
969
70
12
890,017
a The abbreviations are "F" for Forest Land, "C" for Cropland, "G" for Grassland, "W" for Wetlands, "S" for
Settlements, and "O" for Other Lands. Lands remaining in the same land-use category are identified with the land-use
abbreviation given twice (e.g., "FF" is Forest Land Remaining Forest Land), and land-use change categories are
identified with the previous land use abbreviation followed by the new land-use abbreviation (e.g., "CF" is Cropland
Converted to Forest Land).
Note: All land areas reported in this table are considered managed. A planned improvement is underway to deal with an
exception for wetlands, which based on the definitions for the current U.S. Land Representation Assessment includes
both managed and unmanaged lands. U.S. Territories have not been classified into land uses and are not included in the
U.S. Land Representation Assessment. See the Planned Improvements section for discussion on plans to include
territories in future inventories. In addition, C stock changes are not currently estimated for the entire land base, which
leads to discrepancies between the managed land area data presented here and in the subsequent sections of the Inventory.
Figure 6-1: Percent of Total Land Area for Each State in the General Land-Use Categories for
2014(TO BE UPDATED)
Land Use, Land-Use Change, and Forestry 6-7
-------
i Methodology
2 IPCC Approaches for Representing Land Areas
3 IPCC (2006) describes three approaches for representing land areas. Approach 1 provides data on the total area for
4 each individual land-use category, but does not provide detailed information on changes of area between categories
5 and is not spatially explicit other than at the national or regional level. With Approach 1, total net conversions
6 between categories can be detected, but not the individual changes (i.e., additions and/or losses) between the land-
7 use categories that led to those net changes. Approach 2 introduces tracking of individual land-use changes between
8 the categories (e.g., Forest Land to Cropland, Cropland to Forest Land, and Grassland to Cropland), using survey
9 samples or other forms of data, but does not provide location data on all parcels of land. Approach 3 extends
10 Approach 2 by providing location data on all parcels of land, such as maps, along with the land-use history. The
11 three approaches are not presented as hierarchical tiers and are not mutually exclusive.
12 According to IPCC (2006), the approach or mix of approaches selected by an inventory agency should reflect
13 calculation needs and national circumstances. For this analysis, the NRI, FIA, and the NLCD have been combined
14 to provide a complete representation of land use for managed lands. These data sources are described in more detail
15 later in this section. NPJ and FIA are Approach 2 data sources that do not provide spatially-explicit representations
16 of land use and land-use conversions, even though land use and land-use conversions are tracked explicitly at the
17 survey locations. NPJ and FIA data are aggregated and used to develop a land-use conversion matrix for a political
18 or ecologically-defined region. NLCD is a spatially-explicit time series of land-cover data that is used to inform the
19 classification of land use, and is therefore Approach 3 data. Lands are treated as remaining in the same category
20 (e.g., Cropland Remaining Cropland) if a land-use change has not occurred in the last 20 years. Otherwise, the land
21 is classified in a land-use change category based on the current use and most recent use before conversion to the
22 current use (e.g., Cropland Converted to Forest Land).
23 Definitions of Land Use in the United States
24 Managed and Unmanaged Land
25 The United States definition of managed land is similar to the basic IPCC (2006) definition of managed land, but
26 with some additional elaboration to reflect national circumstances. Based on the following definitions, most lands in
27 the United States are classified as managed:
28 • Managed Land: Land is considered managed if direct human intervention has influenced its condition.
29 Direct intervention occurs mostly in areas accessible to human activity and includes altering or maintaining
30 the condition of the land to produce commercial or non-commercial products or services; to serve as
31 transportation corridors or locations for buildings, landfills, or other developed areas for commercial or
32 non-commercial purposes; to extract resources or facilitate acquisition of resources; or to provide social
33 functions for personal, community, or societal objectives where these areas are readily accessible to
34 society.15
35 • Unmanaged Land: All other land is considered unmanaged. Unmanaged land is largely comprised of areas
36 inaccessible to society due to the remoteness of the locations. Though these lands may be influenced
15 Wetlands are an exception to this general definition, because these lands, as specified by IPCC (2006), are only considered
managed if they are created through human activity, such as dam construction, or the water level is artificially altered by human
activity. Distinguishing between managed and unmanaged wetlands in the United States is difficult due to limited data
availability. Wetlands are not characterized by use within the NRI. Therefore, unless wetlands are managed for cropland or
grassland, it is not possible to know if they are artificially created or if the water table is managed based on the use of NRI data.
As a result, all wetlands are reported as managed. See the Planned Improvements section of the Inventory for work being done to
refine the Wetland area estimates.
6-8 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 indirectly by human actions such as atmospheric deposition of chemical species produced in industry or
2 CO2 fertilization, they are not influenced by a direct human intervention.16
3 In addition, land that is previously managed remains in the managed land base for 20 years before re-classifying the
4 land as unmanaged in order to account for legacy effects of management on C stocks.
5 Land-Use Categories
6 As with the definition of managed lands, IPCC (2006) provides general non-prescriptive definitions for the six main
7 land-use categories: Forest Land, Cropland, Grassland, Wetlands, Settlements and Other Land. In order to reflect
8 national circumstances, country-specific definitions have been developed, based predominantly on criteria used in
9 the land-use surveys for the United States. Specifically, the definition of Forest Land is based on the FIA definition
10 of forest,17 while definitions of Cropland, Grassland, and Settlements are based on the NRI.18 The definitions for
11 Other Land and Wetlands are based on the IPCC (2006) definitions for these categories.
12 • Forest Land: A land-use category that includes areas at least 120 feet (36.6 meters) wide and at least one
13 acre (0.4 hectare) in size with at least 10 percent cover (or equivalent stocking) by live trees including land
14 that formerly had such tree cover and that will be naturally or artificially regenerated. Trees are woody
15 plants having a more or less erect perennial stem(s) capable of achieving at least 3 inches (7.6 cm) in
16 diameter at breast height, or 5 inches (12.7 cm) diameter at root collar, and a height of 16.4 feet (5 meters)
17 at maturity in situ. Forest Land includes all areas recently having such conditions and currently
18 regenerating or capable of attaining such condition in the near future. Forest Land also includes transition
19 zones, such as areas between forest and non-forest lands that have at least 10 percent cover (or equivalent
20 stocking) with live trees and forest areas adjacent to urban and built-up lands. Unimproved roads and trails,
21 streams, and clearings in forest areas are classified as forest if they are less than 120 feet (36.6 meters) wide
22 or an acre (0.4 hectare) in size. However, land is not classified as Forest Land if completely surrounded by
23 urban or developed lands, even if the criteria are consistent with the tree area and cover requirements for
24 Forest Land. These areas are classified as Settlements. In addition, Forest Land does not include land that
25 is predominantly under an agricultural land use (Oswalt et al. 2014).
26 • Cropland: A land-use category that includes areas used for the production of adapted crops for harvest;
27 this category includes both cultivated and non-cultivated lands. Cultivated crops include row crops or
28 close-grown crops and also hay or pasture in rotation with cultivated crops. Non-cultivated cropland
29 includes continuous hay, perennial crops (e.g., orchards) and horticultural cropland. Cropland also includes
30 land with agroforestry, such as alley cropping and windbreaks,19 if the dominant use is crop production,
31 assuming the stand or woodlot does not meet the criteria for Forest Land. Lands in temporary fallow or
32 enrolled in conservation reserve programs (i.e., set-asides20) are also classified as Cropland, as long as
33 these areas do not meet the Forest Land criteria. Roads through Cropland, including interstate highways,
34 state highways, other paved roads, gravel roads, dirt roads, and railroads are excluded from Cropland area
35 estimates and are, instead, classified as Settlements.
36 • Grassland: Aland-use category on which the plant cover is composed principally of grasses, grass-like
37 plants (i.e., sedges and rushes), forbs, or shrubs suitable for grazing and browsing, and includes both
38 pastures and native rangelands. This includes areas where practices such as clearing, burning, chaining,
39 and/or chemicals are applied to maintain the grass vegetation. Grassland may have three or fewer years of
16 There are some areas, such as Forest Land and Grassland in Alaska that are classified as unmanaged land due to the
remoteness of their location.
See , page 22.
18 See .
19 Currently, there is no data source to account for biomass C stock change associated with woody plant growth and losses in
alley cropping systems and windbreaks in cropping systems, although these areas are included in the cropland land base.
A set-aside is cropland that has been taken out of active cropping and converted to some type of vegetative cover, including,
for example, native grasses or trees.
Land Use, Land-Use Change, and Forestry 6-9
-------
1 hay production21 that is otherwise pasture or rangelands. Savannas, deserts, and tundra are considered
2 Grassland.22 Drained wetlands are considered Grassland if the dominant vegetation meets the plant cover
3 criteria for Grassland. Woody plant communities of low forbs and shrubs, such as mesquite, chaparral,
4 mountain shrub, and pinyon-juniper, are also classified as Grassland if they do not meet the criteria for
5 Forest Land. Grassland includes land managed with agroforestry practices, such as silvipasture and
6 windbreaks, if the land is principally grasses, grass-like plants, forbs, and shrubs suitable for grazing and
7 browsing, and assuming the stand or woodlot does not meet the criteria for Forest Land. Roads through
8 Grassland, including interstate highways, state highways, other paved roads, gravel roads, dirt roads, and
9 railroads are excluded from Grassland and are, instead, classified as Settlements.
10 • Wetlands: A land-use category that includes land covered or saturated by water for all or part of the year,
11 in addition to the areas of lakes, reservoirs, and rivers. Managed Wetlands are those where the water level
12 is artificially changed, or were created by human activity. Certain areas that fall under the managed
13 Wetlands definition are included in other land uses based on the IPCC guidance, including Cropland
14 (drained wetlands for crop production and also systems that are flooded for most or just part of the year,
15 such as rice cultivation and cranberry production), Grassland (drained wetlands dominated by grass cover),
16 and Forest Land (including drained or un-drained forested wetlands).
17 • Settlements: A land-use category representing developed areas consisting of units of 0.25 acres (0.1 ha) or
18 more that includes residential, industrial, commercial, and institutional land; construction sites; public
19 administrative sites; railroad yards; cemeteries; airports; golf courses; sanitary landfills; sewage treatment
20 plants; water control structures and spillways; parks within urban and built-up areas; and highways,
21 railroads, and other transportation facilities. Also included are tracts of less than 10 acres (4.05 ha) that
22 may meet the definitions for Forest Land, Cropland, Grassland, or Other Land but are completely
23 surrounded by urban or built-up land, and so are included in the Settlements category. Rural transportation
24 corridors located within other land uses (e.g., Forest Land, Cropland, and Grassland) are also included in
25 Settlements.
26 • Other Land: A land-use category that includes bare soil, rock, ice, and all land areas that do not fall into
27 any of the other five land-use categories, which allows the total of identified land areas to match the
28 managed land base. Following the guidance provided by the IPCC (2006), C stock changes and non-CCh
29 emissions are not estimated for Other Lands because these areas are largely devoid of biomass, litter and
30 soil C pools. However, C stock changes and non-CO2 emissions are estimated for Land Converted to Other
31 Land during the first 20 years following conversion to account for legacy effects.
32 Land-Use Data Sources: Description and Application to U.S.
33 Land Area Classification
34 U.S. Land-Use Data Sources
35 The three main sources for land-use data in the United States are the NRI, FIA, and the NLCD (Table 6-6). These
36 data sources are combined to account for land use in all 50 states. FIA and NRI data are used when available for an
37 area because the surveys contain additional information on management, site conditions, crop types, biometric
38 measurements, and other data from which to estimate C stock changes on those lands. If NRI and FIA data are not
39 available for an area, however, then the NLCD product is used to represent the land use.
21 Areas with four or more years of continuous hay production are Cropland because the land is typically more intensively
managed with cultivation, greater amounts of inputs, and other practices.
22 2006 IPCC Guidelines do not include provisions to separate desert and tundra as land use categories.
6-10 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 6-6: Data Sources Used to Determine Land Use and Land Area for the Conterminous
2 United States, Hawaii, and Alaska
NRI FIA NLCD
Forest Land
Conterminous United
States
Non-Federal
Federal
Hawaii
Non-Federal
Federal
Alaska
Non-Federal
Federal
Croplands, Grasslands, Other Lands, Settlements, and Wetlands
Conterminous United
States
Non-Federal •
Federal •
Hawaii
Non-Federal •
Federal •
Alaska
Non-Federal •
Federal •
3 National Resources Inventory
4 For the Inventory, the NRI is the official source of data on all land uses on non-federal lands in the conterminous
5 United States and Hawaii (except Forest Land), and is also used as the resource to determine the total land base for
6 the conterminous United States and Hawaii. The NRI is a statistically-based survey conducted by the USDA
7 Natural Resources Conservation Service and is designed to assess soil, water, and related environmental resources
8 on non-federal lands. The NRI has a stratified multi-stage sampling design, where primary sample units are
9 stratified on the basis of county and township boundaries defined by the United States Public Land Survey (Nusser
10 and Goebel 1997). Within a primary sample unit (typically a 160 acre [64.75 hectare] square quarter-section), three
11 sample points are selected according to a restricted randomization procedure. Each point in the survey is assigned
12 an area weight (expansion factor) based on other known areas and land-use information (Nusser and Goebel 1997).
13 The NRI survey utilizes data derived from remote sensing imagery and site visits in order to provide detailed
14 information on land use and management, particularly for croplands and grasslands, and is used as the basis to
15 account for C stock changes in agricultural lands (except federal Grasslands). The NRI survey was conducted every
16 5 years between 1982 and 1997, but shifted to annualized data collection in 1998. The land use between five-year
17 periods from 1982 and 1997 are assumed to be the same for a five-year time period if the land use is the same at the
18 beginning and end of the five-year period. (Note: most of the data has the same land use at the beginning and end of
19 the five-year periods.) If the land use had changed during a five-year period, then the change is assigned at random
20 to one of the five years. For crop histories, years with missing data are estimated based on the sequence of crops
21 grown during years preceding and succeeding a missing year in the NRI history. This gap -filling approach allows
22 for development of a full time series of land-use data for non-federal lands in the conterminous United States and
23 Hawaii. This Inventory incorporates data through 2010 from the NRI. The land use patterns are assumed to remain
24 the same from 2010 through 2014 for this Inventory, but the time series will be updated when new data are released.
25 Forest Inventory and Analysis
26 The FIA program, conducted by the USFS, is another statistically-based survey for the conterminous United States,
27 and the official source of data on Forest Land area and management data for the Inventory in this region of the
28 country. FIA engages in a hierarchical system of sampling, with sampling categorized as Phases 1 through 3, in
29 which sample points for phases are subsets of the previous phase. Phase 1 refers to collection of remotely-sensed
30 data (either aerial photographs or satellite imagery) primarily to classify land into forest or non-forest and to identify
Land Use, Land-Use Change, and Forestry 6-11
-------
1 landscape patterns like fragmentation and urbanization. Phase 2 is the collection of field data on a network of
2 ground plots that enable classification and summarization of area, tree, and other attributes associated with forest -
3 land uses. Phase 3 plots are a subset of Phase 2 plots where data on indicators of forest health are measured. Data
4 from all three phases are also used to estimate C stock changes for Forest Land. Historically, FIA inventory surveys
5 have been conducted periodically, with all plots in a state being measured at a frequency of every five to 14 years.
6 A new national plot design and annual sampling design was introduced by FIA about ten years ago. Most states,
7 though, have only recently been brought into this system. Annualized sampling means that a portion of plots
8 throughout each state is sampled each year, with the goal of measuring all plots once every five years. See Annex
9 3.13 to see the specific survey data available by state. The most recent year of available data varies state by state
10 (range of most recent data is from 2012 through 2014; see Table A-246).
11 National Land Cover Dataset
12 While the NRI survey sample covers the conterminous United States and Hawaii, land use data are only collected on
13 non-federal lands. In addition, FIA only records data for forest land across the land base in the conterminous United
14 States and a portion of Alaska.23 Consequently, major gaps exist in the land use classification when the datasets are
15 combined, such as federal grassland operated by Bureau of Land Management (BLM), USD A, and National Park
16 Service, as well as Alaska.24 The NLCD is used as a supplementary database to account for land use on federal
17 lands in the conterminous United States and Hawaii, in addition to federal and non-federal lands in Alaska.
18 NLCD products provide land-cover for 1992, 2001, 2006, and 2011 in the conterminous United States (Homer et al.
19 2007), and also for Alaska and Hawaii in 2001. For the conterminous United States, the NLCD data have been
20 further processed to derive Land Cover Change Products for 2001, 2006, and 2011 (Fry et al. 2011, Homer et al.
21 2007, Jin et al. 2013). A change product is not available for Alaska and Hawaii because the data are only available
22 for one year, i.e., 2001). The NLCD products are based primarily onLandsat Thematic Mapper imagery at a 30
23 meter resolution, and contain 21 categories of land-cover information, which have been aggregated into the thirty -
24 six IPCC land-use categories for the conterminous United States and into the six IPCC land use categories for
25 Hawaii and Alaska.
26 The aggregated maps of IPCC land use categories were used in combination with the NRI database to represent land
27 use and land-use change for federal lands, as well as federal and non-federal lands in Alaska. Specifically, NRI
28 survey locations designated as federal lands were assigned a land use/land use change category based on the NLCD
29 maps that had been aggregated into the IPCC categories. This analysis addressed shifts in land ownership across
30 years between federal or non-federal classes as represented in the NRI survey (i.e., the ownership is classified for
31 each survey location is the NRI). NLCD is strictly a source of land-cover information, however, and does not
32 provide the necessary site conditions, crop types, and management information from which to estimate C stock
33 changes on those lands. The sources of these additional data are discussed in subsequent sections of the NIR.
34 Managed Land Designation
35 Lands are designated as managed in the United States based on the definition provided earlier in this section. In
36 order to apply the definition in an analysis of managed land, the following criteria are used:
37 • All Croplands and Settlements are designated as managed so only Grassland, Forest Land or Other
38 Lands may be designated as unmanaged land;
39 • All Forest Lands with active fire protection are considered managed;
40 • All Grassland is considered managed at a county scale if there are livestock in the county;25
41 • Other areas are considered managed if accessible based on the proximity to roads and other
42 transportation corridors, and/or infrastructure;
23 FIA does collect some data on non-forest land use, but these are held in regional databases versus the national database. The
status of these data is being investigated.
The FIA and NRI survey programs also do not include U.S. Territories with the exception of non-federal lands in Puerto Rico,
which are included in the NRI survey. Furthermore, NLCD does not include coverage for all U.S. Territories.
25 Assuming all grasslands are grazed in a county with even very small livestock populations is a conservative assumption about
human impacts on grasslands. Currently, detailed information on grazing at sub-county scales is not available for the United
States to make a finer delineation of managed land.
6-12 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 • Protected lands maintained for recreational and conservation purposes are considered managed (i.e.,
2 managed by public and private organizations);
3 • Lands with active and/or past resource extraction are considered managed; and
4 • Lands that were previously managed but subsequently classified as unmanaged remain in the managed
5 land base for 20 years following the conversion to account for legacy effects of management on C
6 stocks.
7 The analysis of managed lands is conducted using a geographic information system. Lands that are used for crop
8 production or settlements are determined from the NLCD (Fry et al. 2011, Homer et al. 2007, Jin et al. 2013).
9 Forest Lands with active fire management are determined from maps of federal and state management plans from
10 the National Atlas (U.S. Department of Interior 2005) and Alaska Interagency Fire Management Council (1998). It
11 is noteworthy that all forest lands in the conterminous United States have active fire protection, and are therefore
12 designated as managed regardless of accessibility or other criteria. The designation of grasslands as managed is
13 based on livestock population data at the county scale from the USDA National Agricultural Statistics Service (U.S.
14 Department of Agriculture 2014). Accessibility is evaluated based on a 10-km buffer surrounding road and train
15 transportation networks using the ESRI Data and Maps product (ESRI2008), and a 10-km buffer surrounding
16 settlements using NLCD. Lands maintained for recreational purposes are determined from analysis of the Protected
17 Areas Database (U.S. Geological Survey 2012). The Protected Areas Database includes lands protected from
18 conversion of natural habitats to anthropogenic uses and describes the protection status of these lands. Lands are
19 considered managed that are protected from development if the regulations permit extractive or recreational uses or
20 suppression of natural disturbance. Lands that are protected from development and not accessible to human
21 intervention, including no suppression of disturbances or extraction of resources, are not included in the managed
22 land base. Multiple data sources are used to determine lands with active resource extraction: Alaska Oil and Gas
23 Information System (Alaska Oil and Gas Conservation Commission 2009), Alaska Resource Data File (U.S.
24 Geological Survey 2012), Active Mines and Mineral Processing Plants (U.S. Geological Survey 2005), and Coal
25 Production and Preparation Report (U.S. Energy Information Administration 2011). A buffer of 3,300 and 4,000
26 meters is established around petroleum extraction and mine locations, respectively, to account for the footprint of
27 operation and impacts of activities on the surrounding landscape. The buffer size is based on visual analysis of
28 approximately 130 petroleum extraction sites and 223 mines. The resulting managed land area is overlaid on the
29 NLCD to estimate the area of managed land by land use for both federal and non-federal lands. The remaining land
30 represents the unmanaged land base. The resulting spatial product is used to identify NRI survey locations that are
31 considered managed and unmanaged for the conterminous United States and Hawaii, in addition to determining
32 which areas in the NLCD for Alaska are included in the managed land base.
33 Approach for Combining Data Sources
34 The managed land base in the United States has been classified into the thirty-six IPCC land-use/land-use
35 conversion categories using definitions developed to meet national circumstances, while adhering to IPCC (2006).26
36 In practice, the land was initially classified into a variety of land-use categories within the NRI, FIA, and NLCD
37 datasets, and then aggregated into the thirty-six broad land use and land-use change categories identified in IPCC
38 (2006). All three datasets provide information on forest land areas in the conterminous United States, but the area
39 data from FIA serve as the official dataset for estimating Forest Land in the conterminous United States.
40 Therefore, another step in the analysis is to address the inconsistencies in the representation of the forest land among
41 the three databases. NRI and FIA have different criteria for classifying Forest Land in addition to different sampling
42 designs, leading to discrepancies in the resulting estimates of Forest Land area on non-federal land in the
43 conterminous United States. Similarly, there are discrepancies between the NLCD and FIA data for defining and
44 classifying Forest Land on federal lands. Any change in Forest Land Area in the NRI and NLCD also requires a
45 corresponding change in other land use areas because of the dependence between the Forest Land area and the
46 amount of land designated as other land uses, such as the amount of Grassland, Cropland, and Wetlands (i.e., areas
47 for the individual land uses must sum to the total area of the country).
48 FIA is the main database for forest statistics, and consequently, the NRI and NLCD are adjusted to achieve
49 consistency with FIA estimates of Forest Land in the conterminous United States. Adjustments are made in the
26 Definitions are provided in the previous section.
Land Use, Land-Use Change, and Forestry 6-13
-------
1 Forest Land Remaining Forest Land, Land Converted to Forest Land, and Forest Land converted to other uses (i.e.,
2 grassland, cropland and wetlands). All adjustments are made at the state scale to address the differences in forest
3 land definitions and the resulting discrepancies in areas among the land use and land-use change categories. There
4 are three steps in this process. The first step involves adjustments for Land Converted to Forest Land (Grassland,
5 Cropland and Wetlands), followed by adjustments in Forest Land converted to another land use (i.e., Grassland,
6 Cropland and Wetlands), and finally adjustments to Forest Land Remaining Forest Land.
7 In the first step, Land Converted to Forest Land in the NRI and NLCD are adjusted to match the state-level
8 estimates in the FIA data for non-federal and federal Land Converted to Forest Land, respectively. FIA data do not
9 provide specific land use categories that are converted to Forest Land, but rather a sum of all Land Converted to
10 Forest Land. The NRI and NLCD provide information on specific land use conversions, however, such as
11 Grassland Converted to Forest Land. Therefore, adjustments at the state level to NRI and NLCD are made
12 proportional to the amount of land use change into Forest Land for the state, prior to any adjustments. For example,
13 if 50 percent of land use change to Forest Land is associated with Grassland Converted to Forest Land in a state
14 according to NRI or NLCD, then half of the discrepancy with FIA data in the area of Land Converted to Forest
15 Land is addressed by increasing or decreasing the area in Grassland Converted to Forest Land. Moreover, any
16 increase or decrease in Grassland Converted to Forest Land in NRI or NLCD is addressed by a corresponding
17 change in the area of Grassland Remaining Grassland, so that the total amount of managed area is not changed
18 within an individual state.
19 In the second step, state-level areas are adjusted in the NRI and NLCD to address discrepancies with FIA data for
20 Forest Land converted to other uses. Similar to Land Converted to Forest Land, FIA does not provide information
21 on the specific land use changes, and so areas associated with Forest Land conversion to other land uses in NRI and
22 NLCD are adjusted proportional to the amount area in each conversion class in these datasets.
23 In the final step, the area of Forest Land Remaining Forest Land in a given state according to the NRI and NLCD is
24 adjusted to match the FIA estimates for non-federal and federal land, respectively. It is assumed that the majority of
25 the discrepancy in Forest Land Remaining Forest Land is associated with an under- or over-prediction of Grassland
26 Remaining Grassland and Wetland Remaining Wetland in the NRI and NLCD. This step also assumes that there are
27 no changes in the land use conversion categories. Therefore, corresponding increases or decreases are made in the
28 area estimates of Grasslands Remaining Grasslands and Wetlands Remaining Wetlands from the NRI and NLCD, in
29 order to balance the change in Forest Land Remaining Forest Land area, and ensure no change in the overall amount
30 of managed land within an individual state. The adjustments are based on the proportion of land within each of
31 these land-use categories at the state level, (i.e., a higher proportion of Grassland led to a larger adjustment in
32 Grassland area).
33 The modified NRI data are then aggregated to provide the land-use and land-use change data for non-federal lands
34 in the conterminous United States, and the modified NLCD data are aggregated to provide the land use and land-use
35 change data for federal lands. Data for all land uses in Hawaii are based on NRI for non-federal lands and on NLCD
36 for federal lands. Land use data in Alaska are based solely on the NLCD data (Table 6-6). The result is land use
37 and land-use change data for the conterminous United States, Hawaii, and Alaska.27
38 A summary of the details on the approach used to combine data sources for each land use are described below.
39 • Forest Land: Both non-federal and federal forest lands in the conterminous United States and coastal
40 Alaska are covered by FIA. FIA is used as the basis for both Forest Land area data as well as to estimate C
41 stocks and fluxes on Forest Land in the conterminous United States. FIA does have survey plots in coastal
42 Alaska that are used to determine the C stock changes, but the area data for this region are based on the
43 2001 NLCD. In addition, interior Alaska is not currently surveyed by FIA so forest land in this region are
44 also based on the 2001 NLCD. NRI is being used in the current report to provide Forest Land areas on
45 non-federal lands in Hawaii and NLCD is used for federal lands. FIA data will be collected in Hawaii in the
46 future.
47 • Cropland: Cropland is classified using the NRI, which covers all non-federal lands within 49 states
48 (excluding Alaska), including state and local government-owned land as well as tribal lands. NRI is used
49 as the basis for both Cropland area data as well as to estimate soil C stocks and fluxes on Cropland. NLCD
27 Only one year of data are currently available for Alaska so there is no information on land-use change for this state.
6-14 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 is used to determine Cropland area and soil C stock changes on federal lands in the conterminous United
2 States and Hawaii. NLCD is also used to determine croplands in Alaska, but C stock changes are not
3 estimated for this region in the current Inventory.
4 • Grassland: Grassland on non-federal lands is classified using the NRI within 49 states (excluding Alaska),
5 including state and local government-owned land as well as tribal lands. NRI is used as the basis for both
6 Grassland area data as well as to estimate soil C stocks and fluxes on Grassland. Grassland area and soil C
7 stock changes are determined using the classification provided in the NLCD for federal land within the
8 conterminous United States. NLCD is also used to estimate the areas of federal and non-federal grasslands
9 in Alaska, and the federal lands in Hawaii, but the current Inventory does not include C stock changes in
10 these areas.
11 • Wetlands: NRI captures wetlands on non-federal lands within 49 states (excluding Alaska), while federal
12 wetlands and wetlands in Alaska are covered by the NLCD.28
13 • Settlements: NRI captures non-federal settlement area in 49 states (excluding Alaska). If areas of Forest
14 Land or Grassland under 10 acres (4.05 ha) are contained within settlements or urban areas, they are
15 classified as Settlements (urban) in the NRI database. If these parcels exceed the 10 acre (4.05 ha)
16 threshold and are Grassland, they will be classified as such by NRI. Regardless of size, a forested area is
17 classified as non-forest by FIA if it is located within an urban area. Settlements on federal lands and in
18 Alaska are covered by NLCD.
19 • Other Land: Any land that is not classified into one of the previous five land-use categories, is categorized
20 as Other Land using the NRI for non-federal areas in the conterminous United States and Hawaii and using
21 the NLCD for the federal lands in all regions of the United States and for non-federal lands in Alaska.
22 Some lands can be classified into one or more categories due to multiple uses that meet the criteria of more than one
23 definition. However, a ranking has been developed for assignment priority in these cases. The ranking process is
24 from highest to lowest priority, in the following manner:
25 Settlements > Cropland > Forest Land > Grassland > Wetlands > Other Land
26 Settlements are given the highest assignment priority because they are extremely heterogeneous with a mosaic of
27 patches that include buildings, infrastructure, and travel corridors, but also open grass areas, forest patches, riparian
28 areas, and gardens. The latter examples could be classified as Grassland, Forest Land, Wetlands, and Cropland,
29 respectively, but when located in close proximity to settlement areas they tend to be managed in a unique manner
30 compared to non-settlement areas. Consequently, these areas are assigned to the Settlements land-use category.
31 Cropland is given the second assignment priority, because cropping practices tend to dominate management
32 activities on areas used to produce food, forage, or fiber. The consequence of this ranking is that crops in rotation
33 with pasture will be classified as Cropland, and land with woody plant cover that is used to produce crops (e.g.,
34 orchards) is classified as Cropland, even though these areas may meet the definitions of Grassland or Forest Land,
35 respectively. Similarly, Wetlands are considered Croplands if they are used for crop production, such as rice or
36 cranberries. Forest Land occurs next in the priority assignment because traditional forestry practices tend to be the
37 focus of the management activity in areas with woody plant cover that are not croplands (e.g., orchards) or
38 settlements (e.g., housing subdivisions with significant tree cover). Grassland occurs next in the ranking, while
39 Wetlands then Other Land complete the list.
40 The assignment priority does not reflect the level of importance for reporting greenhouse gas emissions and
41 removals on managed land, but is intended to classify all areas into a discrete land use. Currently, the IPCC does
42 not make provisions in the guidelines for assigning land to multiple uses. For example, a wetland is classified as
43 Forest Land if the area has sufficient tree cover to meet the stocking and stand size requirements. Similarly,
44 wetlands are classified as Cropland if they are used for crop production, such as rice or cranberries, or as Grassland
45 if they are composed principally of grasses, grass-like plants (i.e., sedges and rushes), forbs, or shrubs suitable for
46 grazing and browsing. Regardless of the classification, emissions from these areas are included in the Inventory if
This analysis does not distinguish between managed and unmanaged wetlands, which is a planned improvement for the
Inventory.
Land Use, Land-Use Change, and Forestry 6-15
-------
1 the land is considered managed and presumably impacted by anthropogenic activity in accordance with the guidance
2 provided in IPCC (2006).
3
21
QA/QC and Verification
4 The land base derived from the NRI, FIA, and NLCD was compared to the Topologically Integrated Geographic
5 Encoding and Referencing (TIGER) survey (U.S. Census Bureau 2010). The U.S. Census Bureau gathers data on
6 the U.S. population and economy, and has a database of land areas for the country. The land area estimates from the
7 U.S. Census Bureau differ from those provided by the land-use surveys used in the Inventory because of
8 discrepancies in the reporting approach for the Census and the methods used in the NRI, FIA, and NLCD. The area
9 estimates of land-use categories, based on NRI, FIA, and NLCD, are derived from remote sensing data instead of the
10 land survey approach used by the U.S. Census Survey. More importantly, the U.S. Census Survey does not provide
11 a time series of land-use change data or land management information. Consequently, the U.S. Census Survey was
12 not adopted as the official land area estimate for the Inventory. Rather, the NRI, FIA, and NLCD datasets were
13 adopted because these databases provide full coverage of land area and land use for the conterminous United States,
14 Alaska, and Hawaii, in addition to management and other data relevant for the Inventory. Regardless, the total
15 difference between the U.S. Census Survey and the combined NRI, FIA, and NLCD data is about 46 million
16 hectares for the total U.S. land base of about 936 million hectares currently included in the Inventory, or a 5 percent
17 difference. Much of this difference is associated with open waters in coastal regions and the Great Lakes, which is
18 included in the TIGER Survey of the U.S. Census, but not included in the land representation using the NRI, FIA
19 and NLCD. There is only a 0.4 percent difference when open water in coastal regions is removed from the TIGER
20 data.
Recalculations Discussion
22 In previous years, FIA did not separate Forest Land into land use and land use change categories, reporting all areas
23 as Forest Land Remaining Forest Land for the purpose of estimating forest carbon stock changes. In this Inventory,
24 forest carbon stock changes are reported for Land Converted to Forest, Forest Converted to other Land Uses, in
25 addition to Forest Land Remaining Forest Land. As such, adjustments to NRI and NLCD accounted for land use
26 changes associated with Forest Land, which led to minor adjustments to the time series. Other small changes
27 occurred in the areas of Grassland, Wetland, and Cropland due to the modifications to the Forest Land data in FIA
28 and the process of combining the NRI, NLCD and FIA products into a harmonized dataset.
29 In addition to the changes in the FIA data, a new NRI dataset was incorporated into the current Inventory extending
30 the time series from 2007 to 2010. The NRI program also recalculated the previous time series based on changes to
31 the classification and imputation procedures for filling gaps.
32 The definition of Grassland also changed so that a land use history that includes three or fewer years within a
33 sequence of grass pasture or rangeland is considered Grassland, rather than converting these areas into Cropland.
34 Land use remains virtually unchanged in these cases with harvesting of the existing grass vegetation, with no impact
35 on carbon stocks. In contrast, longer term adoption of continuous hay tends to change the management to a more
36 intensive set of practices that influences the carbon stocks. This exception is only applied to hay. Any change in
37 land management that involves cultivation of other crops, such as corn, wheat, or soybeans, is still considered a land
38 use change.
39 The revisions in land representation led to the following changes in land use areas for the managed land base: on
40 average over the time series, Forest Land area decreased by 0.2 percent, Cropland area increased by 3.1 percent,
41 Grassland area increased by 0.7 percent, Wetland area decreased by 0.8 percent, Settlements decreased by 16.6
42 percent, and Other Lands increased by 5.8 percent.
43 Planned Improvements
44 A key planned improvement is to fully incorporate area data by land-use type for U.S. Territories into the Inventory.
45 Fortunately, most of the managed land in the United States is included in the current land-use statistics, but a
46 complete accounting is a key goal for the near future. Preliminary land-use area data for U.S. Territories by land-
47 use category are provided in Box 6-2: Preliminary Estimates of Land Use in U.S. Territories.
48
6-16 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Box 6-2: Preliminary Estimates of Land Use in U.S. Territories
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Several programs have developed land cover maps for U.S. Territories using remote sensing imagery, including the
Gap Analysis program, Caribbean Land Cover project, National Land Cover Dataset, USFS Pacific Islands Imagery
Project, and the National Oceanic and Atmospheric Administration (NOAA) Coastal Change Analysis Program.
Land-cover data can be used to inform a land-use classification if there is a time series to evaluate the dominate
practices. For example, land that is principally used for timber production with tree cover over most of the time
series is classified as forest land even if there are a few years of grass dominance following timber harvest. These
products were reviewed and evaluated for use in the national Inventory as a step towards implementing a planned
improvement to include U.S. Territories in the land representation for the Inventory. Recommendations are to use
the NOAA Coastal Change Analysis Program (C-CAP) Regional Land Cover Database for the smaller island
Territories (U.S. Virgin Islands, Guam, Northern Marianas Islands, and American Samoa) because this program is
ongoing and therefore will be continually updated. The C-CAP product does not cover the entire territory of Puerto
Rico so the NLCD was used for this area. The final selection of a land-cover product for these territories is still
under discussion. Results are presented below (in hectares). The total land area of all U.S. Territories is 1.05
million hectares, representing 0.1 percent of the total land base for the United States.
Table 6-7: Total Land Area (Hectares) by Land-Use Category for U.S. Territories
Cropland
Forest Land
Grasslands
Other Land
Settlements
Wetlands
Total
Puerto Rico
19,712
404,004
299,714
5,502
130,330
24,525
883,788
U.S. Virgin
Islands
138
13,107
12,148
1,006
7,650
4,748
38,796
Guam
236
24,650
15,449
1,141
11,146
1,633
54,255
Northern
Marianas
Islands
289
25,761
13,636
5,186
3,637
260
48,769
American
Samoa
389
15,440
1,830
298
1,734
87
19,777
Total
20,764
482,962
342,777
13,133
154,496
31,252
1,045,385
Additional work will be conducted to reconcile differences in Forest Land estimates between the NRI and FIA. This
improvement will include an analysis designed to develop finer scale adjustments at the survey locations.
Harmonization is planned at the survey location scale using ancillary data, such as the NLCD, which is expected to
better capture the differences in Forest Land classification between the two surveys, as well as the conversions of
land to other uses that involve Forest Land.
NLCD data for Alaska were recently released for 2011, and will be used to analyze land use change for this state in
the next inventory. There are also other databases that may need to be reconciled with the NRI and NLCD datasets,
particularly for Settlements. Urban area estimates, used to produce C stock and flux estimates from urban trees, are
currently based on population data (1990, 2000, and 2010 U.S. Census data). Using the population statistics, "urban
clusters" are defined as areas with more than 500 people per square mile. The USFS is currently moving ahead with
an urban forest inventory program so that urban forest area estimates will be consistent with FIA forest area
estimates outside of urban areas, which would be expected to reduce omissions and overlap of forest area estimates
along urban boundary areas.
As adopted by the UNFCCC, new guidance in the 2013 Supplement to the 2006 Guidelines for National Greenhouse
Gas Inventories: Wetlands will be implemented in the Inventory. This will likely have implications for the
classification of managed and unmanaged wetlands in the Inventory report. More detailed wetlands datasets will
also be evaluated and integrated into the analysis in order to implement the new guidance.
Land Use, Land-Use Change, and Forestry 6-17
-------
i 6.2 Forest Land Remaining Forest Land
2 Changes in Forest Carbon Stocks (IPCC Source Category 4A1)
3 Delineation of Carbon Pools
4 For estimating carbon (C) stocks or stock change (flux), C in forest ecosystems can be divided into the following
5 five storage pools (IPCC 2006):
6 • Aboveground biomass, which includes all living biomass above the soil including stem, stump, branches,
7 bark, seeds, and foliage. This category includes live understory.
8 • Belowground biomass, which includes all living biomass of coarse living roots greater than 2 mm diameter.
9 • Dead wood, which includes all non-living woody biomass either standing, lying on the ground (but not
10 including litter), or in the soil.
11 • Litter, which includes the litter, fumic, and humic layers, and all non-living biomass with a diameter less
12 than 7.5 cm at transect intersection, lying on the ground.
13 • Soil organic C (SOC), including all organic material in soil to a depth of 1 meter but excluding the coarse
14 roots of the belowground pools.
15 In addition, there are two harvested wood pools to account for when estimating C flux:
16 • Harvested wood products (HWP) in use.
17 • HWP in solid waste disposal sites (SWDS).
is Forest Carbon Cycle
19 Carbon is continuously cycled among the previously defined C storage pools and the atmosphere as a result of
20 biological processes in forests (e.g., photosynthesis, respiration, decomposition, and disturbances such as fires or
21 pest outbreaks) and anthropogenic activities (e.g., harvesting, thinning, and replanting). As trees photosynthesize
22 and grow, C is removed from the atmosphere and stored in living tree biomass. As trees die and otherwise deposit
23 litter and debris on the forest floor, C is released to the atmosphere and is also transferred to the soil pool by
24 organisms that facilitate decomposition.
25 The net change in forest C is not equivalent to the net flux between forests and the atmosphere because timber
26 harvests do not cause an immediate flux of all harvested biomass C to the atmosphere. Instead, harvesting transfers a
27 portion of the C stored in wood to a "product pool." Once in a product pool, the C is emitted over time as CCh when
28 the wood product combusts or decays. The rate of emission varies considerably among different product pools. For
29 example, if timber is harvested to produce energy, combustion releases C immediately, and these emissions are
30 reported for information purposes in the Energy sector with the harvest (i.e., the associated reduction in forest C
31 stocks) and subsequent combustion implicitly accounted for under the Land Use, Land-Use Change, and Forestry
32 (LULUCF) sector (i.e., the harvested timber does not enter the HWP pools). Conversely, if timber is harvested and
33 used as lumber in a house, it may be many decades or even centuries before the lumber decays and C is released to
34 the atmosphere. If wood products are disposed of in SWDS, the C contained in the wood may be released many
35 years or decades later, or may be stored almost permanently in the SWDS. These latter fluxes are also accounted for
36 under the LULUCF sector.
37 Quantification of the U.S. Forest Carbon Cycle
38 This section describes the general method for quantifying the net changes in C stocks in the five forest C pools and
39 two harvested wood pools. The basic methodology for determining C stock and stock-change relies on data from the
40 annual forest inventory system that is implemented across all U.S. forest lands (except for interior Alaska) with
41 continual improvements in this monitoring system and associated forest C estimation techniques reflected in the C
6-18 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 estimates (Woodall et al. 2015a). The net change in stocks for each pool is estimated, and then the changes in stocks
2 are summed for all pools to estimate total net flux. The focus on C implies that all C-based greenhouse gases are
3 included, and the focus on stock change suggests that specific ecosystem fluxes do not need to be separately
4 itemized in this report. Changes in C stocks from disturbances, such as forest fires or harvesting, are included in the
5 net changes. For instance, an inventory conducted after fire counts only the trees that are left. Therefore, changes in
6 C stocks from natural disturbances, such as wildfires, pest outbreaks, and storms, are accounted for in the forest
7 inventory approach; however, they are highly variable from year to year. The IPCC (2006) recommends estimating
8 changes in C stocks from forest lands according to several land-use types and conversions, specifically Forest Land
9 Remaining Forest Land and Land Converted to Forest Land. This is the first report to delineate forest C stock
10 changes by the two land-use types as specifically requested by a UNFCCC in country review (UNFCCC 2012). In
11 order to facilitate the delineation between the two land use classes, a new approach to forest C accounting was
12 developed in the United States: The Forest Carbon Accounting Framework (FCAF; Woodall et al. 2015a).
is Forest Area Status in the United States
14 Approximately 34 percent of the U.S. land area is estimated to be forested (Oswalt et al. 2014). The most recent
15 forest inventories from each of the conterminous 48 states (USDA Forest Service 2014a, 2014b) include an
16 estimated 266 million hectares of forest land that are considered managed and are included in this inventory. An
17 additional 6 million hectares of forest land in southeast and south central Alaska are inventoried and are included
18 here. Some differences exist in forest land defined in Oswalt et al. (2014) and the forest land included in this report,
19 which is based on the USDA Forest Service (2015b) forest inventory. Survey data are not yet available for Hawaii
20 and interior Alaska, but estimates of these areas are included in Oswalt et al. (2014). Updated survey data for central
21 and western forest land in both Oklahoma and Texas have only recently become available, and these forests
22 contribute to overall C stocks reported below. While Hawaii and U.S. Territories have relatively small areas of
23 forest land and thus may not substantially influence the overall C budget, these regions will be added to the C budget
24 as sufficient data become available. Agroforestry systems that meet the definition of forest land are also not
25 currently accounted for in the Inventory, since they are not explicitly inventoried by either the Forest Inventory and
26 Analysis (FIA) program of the USDA Forest Service or the Natural Resources Inventory (NRI) of the USDA
27 Natural Resources Conservation Service (Perry et al. 2005).
28 An estimated 77 percent (211 million hectares) of U.S. forests in Alaska and the conterminous U.S. are classified as
29 timberland, meaning they meet minimum levels of productivity and have not been removed from production. Ten
30 percent of Alaska forest land and 80 percent of forest land in the conterminous U.S. are classified as timberland. Of
31 the remaining non-timberland, 30 million hectares are reserved forest lands (withdrawn by law from management
32 for production of wood products) and 69 million hectares are lower productivity forest lands (Oswalt et al. 2014).
33 Historically, the timberlands in the conterminous 48 states have been more frequently or intensively surveyed than
34 other forest land.
35 Since the late 1980s, forest land area has increased by about 14 million hectares (Oswalt et al. 2014) with the
36 southern region of the United States containing the most forest land (Figure 6-2). A substantial portion of this
37 accrued forest land is from the conversion of abandoned croplands to forest (e.g., Woodall etal. 2015b). Current
38 trends in the forest land area in the conterminous United States and coastal Alaska represented here show an average
39 annual rate of increase of 0.1 percent. In addition to the increase in forest area, the major influences to the current
40 net C flux from forest land are management activities and the ongoing impacts of previous land-use changes. These
41 activities affect the net flux of C by altering the amount of C stored in forest ecosystems. For example, intensified
42 management of forests that leads to an increased rate of growth may increase the eventual biomass density of the
43 forest, thereby increasing the uptake and storage of C.29 Though harvesting forests removes much of the
44 aboveground C, on average the estimated volume of annual net growth in the conterminous U.S. states is about
45 double the volume of annual removals on timberlands (Oswalt et al. 2014). The reversion of cropland or grassland to
46 forest land increases C storage in all pools. In concert with this trend, conversion of croplands and grasslands to
47 forest lands continues to drive net increases in forest C stocks over time especially in northern and southern regions.
48 The net effects of forest management and the effects of land-use change involving forest land are captured in the
The term "biomass density" refers to the mass of live vegetation per unit area. It is usually measured on a dry-weight basis.
Dry biomass is 50 percent C by weight.
Land Use, Land-Use Change, and Forestry 6-19
-------
1 estimates of C stocks and fluxes presented in this section [outside of the "hold-out" period (i.e., the time period
2 lands remain in a conversion category) of the estimates contained within the lands converted to forest section].
3 Figure 6-2: Changes in Forest Area by Region for ForestLandRemaining Forest Land'in the
4 conterminous United States and coastal Alaska (1990-2014)
300-.
0)
_03
| 200 H
o
| 150H
03
<6 iooH
o
V
o
50H
0
South
North
Rocky
Mountain
Pacific
Coast
I : I :
1990 1995 2000 2005 2010 2015
Year
South
7 Forest Carbon Stocks and Stock Change
8 In the United States, improved forest management practices, the regeneration of previously cleared forest areas, and
9 timber harvesting and use have resulted in net uptake (i.e., net sequestration) of C each year from 1990 through
10 2014. The rate of forest clearing in the 17th century following European settlement had slowed by the late 19th
11 century. Through the later part of the 20th century many areas of previously forested land in the United States were
12 allowed to revert to forests or were actively reforested. The impacts of these land-use changes still influence C
13 fluxes from these forest lands. More recently, the 1970s and 1980s saw a resurgence of federally-sponsored forest
14 management programs (e.g., the Forestry Incentive Program) and soil conservation programs (e.g., the Conservation
15 Reserve Program), which have focused on tree planting, improving timber management activities, combating soil
16 erosion, and converting marginal cropland to forests. In addition to forest regeneration and management, forest
17 harvests have also affected net C fluxes. Because most of the timber harvested from U.S. forest land is used in wood
6-20 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 products, and many discarded wood products are disposed in solid-waste disposal sites (SWDS) rather than by
2 incineration, significant quantities of C in harvested wood are transferred to long-term storage pools rather than
3 being released rapidly to the atmosphere (Skog 2008). The size of the stocks in these long-term C storage pools has
4 increased during the last century with the question arising as to how long U.S. forest land can remain a net C sink
5 (Woodall et al. 2013; Coulston et al. 2015; Wear and Coulston 2015). Changes in C stocks in U.S. forests and
6 harvested wood associated with Forest Land Remaining Forest Land were estimated to account for net sequestration
7 of 583.4 MMT CO2Eq. (159.1 MMT C) in 2014 (Table 6-8 and Table 6-9). In addition to the net accumulation of C
8 in harvested wood pools, sequestration is a reflection of net forest growth. Overall, estimates of average C density in
9 forest ecosystems (including all pools) remained stable at approximately 0.0003 T C/ha from 1990 to 2014 (Table
10 6-9 and Table 6-10). The stable forest ecosystem C density when combined with increasing forest area results in net
11 C accumulation over time. Management practices that increase C stocks on forest land, as well as legacy effects of
12 afforestation and reforestation efforts, influence the trends of increased C densities in forests and increased forest
13 land in the United States (see Section 6.3 Land Converted to Forest Land for details) (Woodall et al. 2015b). By
14 region, the southern and northern regions of the United States were the largest contributors to net forest ecosystem C
15 flux (Figure 6-3). As per prior submissions, aboveground live biomass accounted for the majority of net
16 sequestration among all forest ecosystem pools (Figure 6-4).
17 Estimated annual net additions to HWP C stock increased slightly between 2013 and 2014. Estimated net additions
18 to solidwood products in use slightly increased due to a further recovery of the housing market. Estimated net
19 additions to products in use for 2014 are about 20 percent of the level of net additions to products in use in 2007,
20 i.e., prior to the recession. The decline in net additions to HWP C stocks continued through 2009 from the recent
21 high point in 2006. This is due to sharp declines in U. S. production of solidwood and paper products in 2009
22 primarily due to the decline in housing construction. The low level of gross additions to solidwood and paper
23 products in use in 2009 was exceeded by discards from uses. The result is a net reduction in the amount of HWP C
24 that is held in products in use during 2009. For 2009, additions to landfills still exceeded emissions from landfills
25 and the net additions to landfills have remained relatively stable. Overall, there were net C additions to HWP in use
26 and in landfills combined due, in large part, to updated data on Products in Use from 2010 to the present.
27 Table 6-8: Net COz Flux from Forest Pools in Forest Land Remaining Forest Land and
28 Harvested Wood Pools. (MMT COz Eq.)
Carbon Pool
Forest
Aboveground
Belowground
Dead Wood
Litter
Soil Organic C
Harvested Wood
Products in Use
SWDS
Total Net Flux
1990
(444.3)
(311.8)
(66.5)
(34.7)
(35.8)
4.5
(131.7)
(64.7)
(67.0) •
(576.0)
2005
(430.0)
(309.7)
(65.5)
(44.0)
(28.4)
17.7
(102.4)
(42.7)
(59.7)
(532.4)
2010
(492.8)
(330.6)
(69.5)
(50.2)
(34.4)
(8.2)
(92.2)
(29.2)
(63.0)
(585.0)
2011
(483.4)
(328.8)
(69.1)
(52.8)
(33.9)
1.2
(94.7)
(30.3)
(64.3)
(578.1)
2012
(478.0)
(324.0)
(68.1)
(53.6)
(33.1)
0.7
(98.7)
(33.1)
(65.6)
(576.7)
2013
(476.9)
(323.0)
(67.8)
(53.8)
(32.8)
0.6
(103.2)
(36.3)
(66.9)
(580.1)
2014
(475.7)
(321.9)
(67.5)
(54.1)
(32.6)
0.4
(107.7)
(39.5)
(68.2)
(583.4)
Note: Forest C stocks do not include forest stocks in U.S. Territories, Hawaii, a portion of managed forests in
Alaska, or trees on non-forest land (e.g., urban trees, agroforestry systems). Parentheses indicate net C
sequestration (i.e., a net removal of C from the atmosphere). Total net flux is an estimate of the actual net flux
between the total forest C pool and the atmosphere. Harvested wood estimates are based on results from annual
surveys and models. Totals may not sum due to independent rounding.
29 Table 6-9: Net C Flux from Forest Pools in Forest Land Remaining Forest Land and Harvested
30 Wood Pools (MMT C)
Carbon Pool
Forest
Aboveground Biomass
Belowground Biomass
Dead Wood
Litter
1990
(121.2)
(85.0)
(18.1)
(9.5)
(9.8)
2005
(117.3)
(84.5)
(17.9)
(12.0)
(7.8)
2010
(134.4)
(90.2)
(19.0)
(13.7)
(9.4)
2011
(131.8)
(89.7)
(18.9)
(14.4)
(9.2)
2012
(130.4)
(88.4)
(18.6)
(14.6)
(9.0)
2013
(130.1)
(88.1)
(18.5)
(14.7)
(9.0)
2014
(129.7)
(87.8)
(18.4)
(14.7)
(8.9)
Land Use, Land-Use Change, and Forestry 6-21
-------
1
2
3
4
5
6
7
10
11
Soil Organic C
Harvested Wood
Products in Use
SWDS
Total Net Flux
1.2 1
(35.9)
(17.7)
(18.3)
(157.1)
4.8
(27.9) 1
(11.7) 1
1 (16.3)
(145.2)
(2.2)
(25.1)
(8.0)
(17.2)
(159.5)
0.3
(25.8)
(8.3)
(17.5)
(157.7)
0.2
(26.9)
(9.0)
(17.9)
(157.3)
0.2
(28.2)
(9.9)
(18.3)
(158.2)
0.1
(29.4)
(10.8)
(18.6)
(159.1)
Note: Forest C stocks do not include forest stocks in U.S. Territories, Hawaii, a portion of managed lands in
Alaska, or trees on non-forest land (e.g., urban trees, agroforestry systems). Parentheses indicate net C
sequestration (i.e., a net removal of C from the atmosphere). Total net flux is an estimate of the actual net flux
between the total forest C pool and the atmosphere. Harvested wood estimates are based on results from annual
surveys and models. Totals may not sum due to independent rounding.
Stock estimates for forest and harvested wood C storage pools are presented in Table 6-10. Together, the estimated
aboveground live and forest soil pools account for a large proportion of total forest C stocks. The estimated C
stocks summed for non-soil pools increased over time. Therefore, the estimated C sequestration was greater than C
emissions from forests, as discussed above. Note that the forest land area estimates in Table 6-10 do not precisely
match those in the Representation of the U.S. Land Base section for Forest Land Remaining Forest Land. This is
because the forest land area estimates in Table 6-10 only include managed forest land in the conterminous 48 states
and southeast/southcentral coastal Alaska (which is the current area encompassed by FIA survey data,
approximately 6.2 million ha) while the estimates in the Representation of the U.S. Land Base section include all
managed forest land in Alaska (approximately 28.0 million ha).
Table 6-10: Forest area (1,000 ha) and C Stocks in Forest Land Remaining Forest Land and
Harvested Wood Pools (MMT C)
Forest Area (1000 ha)
Carbon Pools (MMT C)
Forest
Aboveground Biomass
Belowground Biomass
Dead Wood
Litter
Soil Organic C
Harvested Wood
Products in Use
SWDS
Total C Stock
1990
261,796
84,891
11,896
2,442
2,404
5,833
62,316
1,897
1,250
647
86,788
2005
268,029
87,271
13,076
2,691
2,574
5,958
62,972
2,356
1,449
906
89,627
2010
270,065
88,444
13,688
2,820
2,665
6,016
63,255
2,527
1,499
1,028
90,971
2011
270,462
88,270
13,598
2,801
2,651
6,006
63,214
2,500
1,490
1,010
90,770
2012
270,871
88,444
13,688
2,820
2,665
6,016
63,255
2,527
1,499
1,028
90,971
2013
271,871
88,618
13,777
2,839
2,680
6,025
63,297
2,555
1,509
1,046
91,173
2014
271,719
88,790
13,865
2,857
2,695
6,034
63,339
2,584
1,520
1,065
91,374
2015
272,158
88
13
2
2
6
63
2
1
1
91
,962
,953
,876
,710
,042
,381
,615
,531
,084
,577
Note: Forest area and C stock estimates include all forest land in the conterminous 48 states plus managed forests in
southeast/southcentral coastal Alaska, which is the current area encompassed by FIA survey data. Forest C stocks do not include
forest stocks in U.S. Territories, Hawaii, a large portion of interior Alaska, or trees on non-forest land (e.g., urban trees,
agroforestry systems). Wood product stocks include exports, even if the logs are processed in other countries, and exclude imports.
Harvested wood estimates are based on results from annual surveys and models. Totals may not sum due to independent rounding.
Inventories are assumed to represent stocks as of January 1 of the Inventory year. Flux is the net annual change in stock. Thus, an
estimate of flux for 2014 requires estimates of C stocks for 2014 and 2015.
12
6-22 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Figure 6-3: Forest Ecosystem C Stock Change by Region in Forest Land Remaining Forest
Land'm the conterminous U.S. and coastal Alaska (1990-2014)
0
Year
1990 1995 2000 2005 2010 2015
i I i i i i I i i I i i i i I i i i I i
3
4
5
6
Rocky
Mountain
Pacific
Coast
-10-
-20-
l -3°
0)
O -40-
| -50-
-60-
-70-
-80-
Figure 6-4: Estimated Net Annual Changes in C Stocks for Major C Pools in Forest Land
Remaining Forest Land'm the Conterminous U.S. and Coastal Alaska (MMT C/year)
North
South
it;
s o
20-
0--
-20-
-40-
-60-
Ł -80-
o
| -100-
| -120-
-140-
-160-
i i i i | i i i i | i i i i | i i i i | i i i i | i
1995 2000 2005 2010 2015
Year
Soil organic carbon Harvested Wood Products
^^ Litter ^^ Aboveground biomass
^^ Dead wood ^^ Total net change
Belowground biomass
Land Use, Land-Use Change, and Forestry 6-23
-------
Box 6-3: COz Emissions from Forest Fires
2 As stated previously, the forest inventory approach implicitly accounts for all C losses due to disturbances such as
3 forest fires, because only C remaining in the forest is estimated. Net C stock change is estimated by subtracting
4 consecutive C stock estimates. A forest fire disturbance removes C from the forest. The inventory data on which
5 net C stock estimates are based already reflect this C loss. Therefore, estimates of net annual changes in C stocks
6 for U.S. forest land already account for CC>2 emissions from forest fires occurring in the conterminous states as well
7 as the portion of managed forest lands in Alaska that are captured in this Inventory. Because it is of interest to
8 quantify the magnitude of CCh emissions from fire disturbance, these separate estimates are highlighted here. Non-
9 CO2 greenhouse gas emissions from forest fires are also quantified in a separate section below.
10 The IPCC (2006) methodology and a combination of U. S. -specific data on annual area burned and potential fuel
11 availability together with default combustion factors were employed to estimate CCh emissions from forest fires.
12 CO2 emissions for wildfires in the conterminous 48 states and in Alaska as well as prescribed fires in 2014 were
13 estimated to be 92.3 MMT CCh/year (Table 6-11). Most of this quantity is an embedded component of the net
14 annual forest C stock change estimates provided previously (e.g., Table 6-9), but this separate approach to estimate
15 emissions is necessary in order to associate a portion of emissions, including estimates of CH4 and N2O with fire.
16 See the discussion in Annex 3.13 for more details on this methodology. Note that the estimates for Alaska provided
17 in Table 6-11 include all managed forest land in the state and are not limited to the subset with permanent inventory
18 plots on managed lands as specified elsewhere in this chapter (e.g., Table 6-9).
19 Table 6-11: Estimates of COz (MMT/yr) Emissions from Forest Fires in the Conterminous 48
20 States and Alaska3
21
Year
CCh emitted from
Wildfires in the
Conterminous 48
States (MMT/yr)
CCh emitted from
Wildfires in Alaska
(MMT/yr)
CCh emitted from
Prescribed Fires
(MMT/yr)
Total CO2
emitted
(MMT/yr)
1990
21.3
19.5
40.9
2010
2011
2012
2013
2014b
12.2
73.9
133.7
64.7
64.7
11.2
3.5
2.7
22.3
22.3
18.4
5.9
2.9
5.3
5.3
41.7
83.3
139.3
92.3
92.3
a These emissions have already been accounted for in the estimates of net annual changes in C stocks, which
accounts for the amount sequestered minus any emissions, including the assumption that combusted wood may
continue to decay through time.
b The data for 2014 were incomplete when these estimates were summarized; therefore 2013, the most recent
available estimate, is applied to 2014.
22 Methodology and Data Sources
23 The methodology described herein is consistent with IPCC (2006). Forest ecosystem C stocks and net annual C
24 stock change were determined according to stock-difference methods, which involved applying C estimation factors
25 to annual forest inventories across time to derive stock change. Harvested wood C estimates were based on factors
26 such as the allocation of wood to various primary and end-use products as well as half-life (the time at which half of
27 the amount placed in use will have been discarded from use) and expected disposition (e.g., product pool, SWDS,
28 combustion). An overview of the different methodologies and data sources used to estimate the C in forest
29 ecosystems or harvested wood products is provided here. See Annex 3.13 for details and additional information
30 related to the methods and data.
6-24 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Forest Ecosystem Carbon from Forest Inventory
1 The United States adopted Forest Carbon Accounting Framework (FCAF) (Woodall et al. 2015a), a new approach
3 that removes the older inventory information from the accounting procedures and enables the delineation of forest C
4 accumulation by forest growth, land use change, and natural disturbances such as fire as suggested in the report
5 provided by the UNFCCC's Expert Review Team (UNFCCC 2012). This framework adheres to accounting
6 guidelines set forth by the Intergovernmental Panel on Climate Change (IPCC 2006) while charting a path forward
7 for the incorporation of emerging research, data, and the needs of stakeholders (e.g., reporting at small scales and
8 boreal forest C). Preliminary results from the new FCAF demonstrate the ability of the new framework to both
9 backcast the annual inventory system while attributing changes in forest C to disturbances and delineating Land
10 Converted to Forest Land from Forest Land Remaining Forest Land. Numerous improvements are planned, such as
11 refining the estimation of individual C pools and land use change identification, which can be incorporated into the
12 framework with future iterations (see the Planned Improvements section below).
13 The FCAF system is comprised of a forest dynamics module and a land use dynamics module (Coulston et al. 2015,
14 Woodall et al. 2015a). The forest dynamics module assesses forest C sequestration, forest aging, and disturbance
15 effects (e.g., disturbances such as wind, fire, and floods identified by foresters on inventory plots). The land use
16 dynamics module assesses C stock transfers associated with afforestation and deforestation. Both modules are
17 developed from land use area statistics and C stock change or C stock transfer by age class. The required inputs are
18 estimated from more than 625,000 forest and non-forest observations in the FIA national database (U.S. Forest
19 Service 2015a, b, c). Model predictions for before or after the annual inventory period are constructed from the
20 FCAF system using the annual observations. This modeling framework includes opportunities for user-defined
21 scenarios to evaluate the impacts of land use change and disturbance rates on future C stocks and stock changes.
22 The accounting system is flexible and can incorporate emerging inventory data (e.g., remeasured western plots and
23 Alaskan lichen biomass), future image-based change estimation information (see Planned Improvements sub-
24 section), information from the Monitoring Trends in Burn Severity database (Eidenshink et al. 2007), and process
25 model output (e.g., inform future forest C densities or land use dynamics). The future accounting system will be
26 increasingly transparent and verifiable through open-source, publicly available R software that links with the FIA
27 database and associated distillations. The accounting system is scalable to allow other users to parameterize models
28 at scales relevant to them, but inherently the framework is built for application at the strategic scale, using FIA data
29 to parameterize the matrices. This introduction to the FCAF is just the first step in a process to engage the public and
30 policy makers in interpreting forest C status and trends.
31 The FCAF is fundamentally driven by the annual forest inventory system conducted by the FIA program of the
32 USDA Forest Service (Prayer and Furnival 1999, Bechtold and Patterson 2005, USDA Forest Service 2015d,
33 2015a). The FIA program relies on a rotating panel statistical design with a sampling intensity of one 674.5 m2
34 ground plot per 2,403 ha of land and water area. A five-panel design, with 20 percent of the field plots typically
35 measured each year, is used in the eastern United States and a ten-panel design, with 10 percent of the field plots
36 measured each year, is used in the western United States. The interpenetrating hexagonal design across the U.S.
37 landscape enables the sampling of plots at various intensities in a spatially and temporally unbiased manner.
38 Typically, tree and site attributes are measured with higher sample intensity while other ecosystem attributes such as
39 downed dead wood are sampled during summer months at lower intensities. The first step in incorporating FIA data
40 into FCAF was to identify annual inventory datasets by U.S. state. Inventories include data collected on permanent
41 inventory plots on forest lands and were organized as separate datasets, each representing a complete inventory, or
42 survey, of an individual state at a specified time. Many of the annual inventories reported for states are represented
43 as "moving window" averages, which mean that a portion—but not all—of the previous year's inventory is updated
44 each year (USDA Forest Service 2015d). Forest C calculations are organized according to these state surveys, and
45 the frequency of surveys varies by state.
46 Separate estimates were prepared for the five IPCC C storage pools described above. All estimates were based on
47 data collected from the extensive array of permanent, annual forest inventory plots and associated models (e.g., live
48 tree belowground biomass) in the United States (USDA Forest Service 2015b, 2015c). Carbon conversion factors
49 were applied at the disaggregated level of each inventory plot and then appropriately expanded to population
50 estimates. Tier 3 methods, as outlined by IPCC (2006), were used for the five IPCC (2006) reporting pools.
Land Use, Land-Use Change, and Forestry 6-25
-------
1 Carbon in Biomass
1 Live tree C pools include aboveground and belowground (coarse root) biomass of live trees with diameter at breast
3 height (dbh) of at least 2.54 cm at 1.37 m above the forest floor. Separate estimates were made for above- and
4 below-ground biomass components. If inventory plots included data on individual trees, tree C was based on
5 Woodall et al. (201 la), which is also known as the component ratio method (CRM), and is a function of volume,
6 species, and diameter. An additional component of foliage, which was not explicitly included in Woodall et al.
7 (201 la), was added to each tree following the same CRM method.
8 Understory vegetation is a minor component of biomass, which is defined as all biomass of undergrowth plants in a
9 forest, including woody shrubs and trees less than 2.54 cm dbh. For this Inventory, it was assumed that 10 percent of
10 total understory C mass is belowground. Estimates of C density were based on information in Birdsey (1996) and
11 biomass estimates from Jenkins et al. (2003). Understory frequently represented over 1 percent of C in biomass, but
12 its contribution rarely exceeded 2 percent of the total.
13 Carbon in Dead Organic Matter
14 Dead organic matter was initially calculated as three separate pools—standing dead trees, downed dead wood, and
15 litter—with C stocks estimated from sample data or from models. The standing dead tree C pools include
16 aboveground and belowground (coarse root) mass and include trees of at least 12.7 cm dbh. Calculations followed
17 the basic method applied to live trees (Woodall et al. 201 la) with additional modifications to account for decay and
18 structural loss (Domke et al. 2011, Harmon et al. 2011). Downed dead wood estimates are based on measurement of
19 a subset of FIA plots for downed dead wood (Domke et al. 2013, Woodall and Monleon 2008, Woodall et al. 2013).
20 Downed dead wood is defined as pieces of dead wood greater than 7.5 cm diameter, at transect intersection, that are
21 not attached to live or standing dead trees. This includes stumps and roots of harvested trees. To facilitate the
22 downscaling of downed dead wood C estimates from the state-wide population estimates to individual plots, downed
23 dead wood models specific to regions and forest types within each region are used. Litter C is the pool of organic C
24 (also known as duff, humus, and fine woody debris) above the mineral soil and includes woody fragments with
25 diameters of up to 7.5 cm. A subset of FIA plots are measured for litter C. A novel modeling approach, using litter C
26 measurements from FIA plots (Domke et al. In Review), was used to estimate litter C for every FIA plot used in
27 FCAF.
28 Carbon in Forest Soil
29 Soil organic carbon (SOC) is the largest terrestrial C sink, and management of this pool is a critical component of
30 efforts to mitigate atmospheric C concentrations. SOC also affects essential biological, chemical, and physical soil
31 functions such as nutrient cycling, water retention, and soil structure (Jandl et al. 2014). Much of the SOC on earth
32 is found in forest ecosystems and is thought to be relatively stable. However, there is growing evidence that SOC is
33 sensitive to global change effects, particularly land use histories, resource management, and climate. In the United
34 States, SOC in forests is monitored by the national forest inventory conducted by the FIA program (O'Neill et al.
35 2005). In previous C inventory submissions, SOC predictions were based, in part, on a model using the State Soil
36 Geographic (STATSGO) database compiled by the Natural Resources Conservation Service (NRCS) (Amichev and
37 Glabraith 2004). Estimates of forest SOC found in the STATSGO database may be based on expert opinion rather
38 than actual measurements, but these country-specific model predictions have been used in past submissions. The
39 FIA program has been consistently measuring soil attributes as part of the inventory since 2001 and has amassed an
40 extensive inventory of SOC in forest land in the conterminous United States and coastal Alaska (O'Neill et al.
41 2005). More than 5,000 profile observations of SOC on forest land from FIA and the International Soil Carbon
42 Monitoring Network were used to develop and implement an approach that enabled the prediction of soil C to a
43 depth of 100 cm from empirical measurements to a depth of 20 cm and included site-, stand-, and climate-specific
44 variables that yield predictions of SOC stocks and stock changes specific to forest land in the United States (Domke
45 etal. In Prep).
46 Harvested Wood Carbon
47 Estimates of the HWP contribution to forest C sinks and emissions (hereafter called "HWP Contribution") were
48 based on methods described in Skog (2008) using the WOODCARB II model and the U.S. forest products module
49 (Ince et al. 2011). These methods are based on IPCC (2006) guidance for estimating HWP C. IPCC (2006) provides
6-26 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 methods that allow for reporting of HWP contribution using one of several different accounting approaches:
2 Production, stock change and atmospheric flow, as well as a default method that assumes there is no change in HWP
3 C stocks (see Annex 3.13 for more details about each approach). The United States used the production accounting
4 approach to report HWP Contribution. Under the production approach, C in exported wood was estimated as if it
5 remains in the United States, and C in imported wood was not included in Inventory estimates. Though reported
6 U.S. HWP estimates are based on the production approach, estimates resulting from use of the two alternative
7 approaches, the stock change and atmospheric flow approaches, are also presented for comparison (see Annex 3.13).
8 Annual estimates of change were calculated by tracking the additions to and removals from the pool of products
9 held in end uses (i.e., products in use such as housing or publications) and the pool of products held in solid waste
10 disposal sites (SWDS). Emissions from HWP associated with wood biomass energy are not included in this
11 accounting—a net of zero sequestration and emissions as they are a part of energy accounting (see Chapter 3).
12 Solidwood products added to pools include lumber and panels. End-use categories for solidwood include single and
13 multifamily housing, alteration and repair of housing, and other end-uses. There is one product category and one
14 end-use category for paper. Additions to and removals from pools were tracked beginning in 1900, with the
15 exception that additions of softwood lumber to housing began in 1800. Solidwood and paper product production and
16 trade data were taken from USDA Forest Service and other sources (Hair and Ulrich 1963; Hair 1958; USDC
17 Bureau of Census 1976; Ulrich 1985, 1989; Steer 1948; AF&PA 2006a, 2006b; Howard 2003, 2007, forthcoming).
18 Estimates for disposal of products reflected the change over time in the fraction of products discarded to SWDS (as
19 opposed to burning or recycling) and the fraction of SWDS that were in sanitary landfills versus dumps.
20 There are five annual HWP variables that were used in varying combinations to estimate HWP contribution using
21 any one of the three main approaches listed above. These are:
22 (1 A) annual change of C in wood and paper products in use in the United States,
23 (IB) annual change of C in wood and paper products in SWDS in the United States,
24 (2 A) annual change of C in wood and paper products in use in the United States and other countries where
25 the wood came from trees harvested in the United States,
26 (2B) annual change of C in wood and paper products in SWDS in the United States and other countries
27 where the wood came from trees harvested in the United States,
28 (3) C in imports of wood, pulp, and paper to the United States,
29 (4) C in exports of wood, pulp and paper from the United States, and
30 (5) C in annual harvest of wood from forests in the United States.
31 The sum of variables 2 A and 2B yielded the estimate for HWP contribution under the production accounting
32 approach. A key assumption for estimating these variables was that products exported from the United States and
33 held in pools in other countries have the same half-lives for products in use, the same percentage of discarded
34 products going to SWDS, and the same decay rates in SWDS as they would in the United States.
35 Uncertainty and Time Series Consistency
36 A quantitative uncertainty analysis placed bounds on current flux for forest ecosystems through a combination of
37 sample based and model based approaches to uncertainty for forest ecosystem CO2 flux (Approach 1). A Monte
38 Carlo Stochastic Simulation of the Methods described above and probabilistic sampling of C conversion factors was
39 used to determine the harvested wood product uncertainty (Approach 2). See Annex 3.13 for additional information.
40 The 2014 net annual change for forest C stocks was estimated to be between-823.0 and-351.4MMT CChEq.
41 around a central estimate of -583.4 MMT CC>2 Eq. at a 95 percent confidence level. This includes a range of -686.2
42 to -269.3 MMT CO2 Eq. around a central estimate of -475.7 MMT CC>2 Eq. for forest ecosystems and -136.8 to -
43 82.2 MMT CO2 Eq. around a central estimate of -107.7 for HWP.
Land Use, Land-Use Change, and Forestry 6-27
-------
1 Table 6-12: Quantitative Uncertainty Estimates for Net COz Flux from Forest Land
1 Remaining Forest Land: Changes in Forest C Stocks (MMT COz Eq. and Percent)
„ 2014 Flux Estimate Uncertainty Range Relative to Flux Estimate
°UrCe aS (MMT CCh Eq.) (MMT CCh Eq.) (%)
Forest C Poolsa
Harvested Wood Products'5
Total Forest
C02
C02
CO2
(475.7)
(107.7)
(583.4)
Lower
Bound
(686.2)
(136.8)
(823.0)
Upper
Bound
(269.3)
(82.2)
(351.4)
Lower
Bound
-43.6%
-27.0%
-40.6%
Upper
Bound
+43.6%
+23.7%
+40.0%
Note: Parentheses indicate negative values or net sequestration.
a Range of flux estimates predicted through a combination of sample based and model based uncertainty, Approach 1.
bRange of flux estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval, Approach 2.
3 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
4 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
5 above.
6 QA/QC and Verification
7 As discussed above, the FIA program has conducted consistent forest surveys based on extensive statistically-based
8 sampling of most of the forest land in the conterminous United States, dating back to 1952. The FIA program
9 includes numerous quality assurance and quality control (QA/QC) procedures, including calibration among field
10 crews, duplicate surveys of some plots, and systematic checking of recorded data. Because of the statistically-based
11 sampling, the large number of survey plots, and the quality of the data, the survey databases developed by the FIA
12 program form a strong foundation for C stock estimates. Field sampling protocols, summary data, and detailed
13 inventory databases are archived and are publicly available on the Internet (USDA Forest Service 2015d).
14 General quality control procedures were used in performing calculations to estimate C stocks based on survey data.
15 For example, the derived C datasets, which include inventory variables such as areas and volumes, were compared
16 to standard inventory summaries such as the forest resource statistics of Oswalt et al. (2014) or selected population
17 estimates generated from FIADB 6.0, which are available at an FIA internet site (USDA Forest Service 2015b).
18 Agreement between the C datasets and the original inventories is important to verify accuracy of the data used.
19 Finally, C stock estimates were compared with previous Inventory report estimates to ensure that any differences
20 could be explained by either new data or revised calculation methods (see the "Recalculations" discussion, below).
21 Estimates of the HWP variables and the HWP contribution under the production accounting approach use data from
22 U.S. Census and USDA Forest Service surveys of production and trade. Factors to convert wood and paper to units
23 of C are based on estimates by industry and Forest Service published sources. The WOODCARB II model uses
24 estimation methods suggested by IPCC (2006). Estimates of annual C change in solidwood and paper products in
25 use were calibrated to meet two independent criteria. The first criterion is that the WOODCARB II model estimate
26 of C in houses standing in 2001 needs to match an independent estimate of C in housing based on U.S. Census and
27 USDA Forest Service survey data. Meeting the first criterion resulted in an estimated half-life of about 80 years for
28 single family housing built in the 1920s, which is confirmed by other U.S. Census data on housing. The second
29 criterion is that the WOODCARB II model estimate of wood and paper being discarded to S WDS needs to match
30 EPA estimates of discards each year over the period 1990 to 2000 (EPA 2006). These criteria help reduce
31 uncertainty in estimates of annual change in C in products in use in the U.S. and, to a lesser degree, reduce
32 uncertainty in estimates of annual change in C in products made from wood harvested in the United States. In
33 addition, WOODCARB II landfill decay rates have been validated by ensuring that estimates of CH4 emissions from
34 landfills based on EPA (2006) data are reasonable in comparison to CH4 estimates based on WOODCARB II
35 landfill decay rates.
36 Recalculations Discussion
37 Forest ecosystem stock and stock-change estimates differ from the previous Inventory report principally due to the
38 adoption of a new accounting approach (FCAF; Woodalletal. 2015a). The major differences between FCAF and
39 past accounting approaches is the sole use of annual forest inventory data and the back casting of forest C stocks
6-28 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 across the 1990s based on forest C stock density and land use change information derived from the nationally
2 consistent forest inventory (coupled with in situ observations of non-tree C pools such as soils, dead wood, and
3 litter). The adoption of FCAF has enabled the creation of the two land use sections for forest C stocks: Forest Land
4 Remaining Forest Land and Land Converted to Forest Land. In prior submissions (e.g., the 1990-2013 Inventory
5 submission), these two land use sections were combined. A second major change was the adoption of a new
6 approach to estimating forest soil C, the largest stock in the United States. For detailed discussion of these new
7 approaches please refer to the Methodology section, Annex 3.13, Domke et al. (in prep), and Woodall et al. (2015a).
8 In addition to these major changes, the model of Ogle et al. (in preparation) identifies some of the forest land in
9 south central and southeastern coastal Alaska as unmanaged; this is in contrast to past assumptions of "managed" for
10 these forest lands included in the FIADB. Therefore, the estimates for coastal Alaska as included here reflect that
11 adjustment, which effectively reduces the forest area included here by about 5 percent.
12 In addition to the creation of explicit estimates of removals and emissions by Forest Land Remaining Forest Land
13 versus Land Converted to Forest Land, the FCAF eliminated the use of inconsistent periodic data which contributed
14 to a data artefact in prior estimates of emissions/removals from 1990 to the present. In the previous Inventory report,
15 there was a reduction in net sequestration from 1995 to 2000 followed by an increase in net sequestration from 2000
16 to 2004. This artefact of comparing inconsistent inventories of the 1980s through 1990s to the nationally consistent
17 inventories of the 2000s has been removed in this Inventory submission. This has resulted in a fairly consistent
18 annual net sequestration estimate of approximately -160.0 MMT C. Overall, there were net C additions to HWP in
19 use and in landfills combined due, in large part, to updated data on Products in Use from 2010 through present.
20 Planned Improvements
21 Reliable estimates of forest C across the diverse ecosystems/industries of the United States require a high level of
22 investment in both annual monitoring and associated analytical techniques. Development of improved
23 monitoring/reporting techniques is a continuous process that occurs simultaneously with annual Inventory
24 submissions. Planned improvements can be broadly assigned to the following categories: continued development of
25 the FCAF, individual C pool estimation, and annual inventory data incorporation.
26 As this is the first report to explicitly delineate C change by Forest Land Remaining Forest Land and Land
27 Converted to Forest Land via a new technical approach (FCAF), there are many improvements envisioned. First, the
28 length of time that land remains in a conversion category after a change in land use is currently based on the
29 remeasurement period from the FIA inventory, not the 20 year default (IPCC 2006). Research into the length of time
30 that land remains in a conversion category will be undertaken and a mechanism to account for this in the FCAF will
31 be developed. Second, as FCAF currently operates at the regional scale for the United States research will occur to
32 enable FCAF to operate at finer scales. Third, as the FCAF system was not fully developed in time for this year's
33 inventory report, only Land Converted to Forest Land (i.e., LCFL, reforestation or afforestation) was delineated
34 from Forest Land Remaining Forest Land (i.e., FLRFL). The explicit accounting of C stock changes associated with
3 5 land conversion and disturbance will be developed in the 1990-2015 Inventory submission. As in past submissions,
36 deforestation is implicitly included in the report given the annual forest inventory system but not explicitly
37 estimated. Fourth, the transparency and repeatability of FCAF will be increased through the dissemination of open
38 source code (R programming language) which was used in development of this report in concert with the public
39 availability of the annual forest inventory data (USDA 2015b). Fifth, several FIA database processes will be
40 institutionalize to increase efficiency in reporting and further improve transparency, consistency, and availability of
41 data used in reporting. Finally, a Tier 1 approach was used to estimate uncertainty associated with C stock changes
42 in the FLRFL and LCFL categories in this report. There is research underway investigating more robust approaches
43 to total uncertainty (Woodall et al. 2015a) which will be considered in future Inventory reports.
44 In an effort to reduce the uncertainty associated with the estimation of stocks and stock changes in individual forest
45 C pools, the empirical data and associated models for each pool are annually evaluated for potential improvement
46 (Woodall 2012). In the 1990 through 2010 Inventory report, the approach to tree volume/biomass estimation was
47 evaluated and refined (Domke et al. 2012). In the 1990 through 2011 Inventory report, the standing dead tree C
48 model was replaced with a nationwide inventory and associated empirical estimation techniques (Woodall et al.
49 2012, Domke et al. 2011, Harmon et al. 2011). In the 1990 through 2012 Inventory report, the downed dead tree C
50 estimates were refined by incorporation of a national field inventory of downed dead wood (Woodall et al. 2013,
51 Domke etal. 2013). In the 1990 through 2013 Inventory report, the litter C model was refined with a nearly
52 nationwide field inventory (Domke et al. in review). In the current Inventory, the approach to estimating the soil C
Land Use, Land-Use Change, and Forestry 6-29
-------
1 pool was greatly improved (Domke et al. in prep) by incorporating a national inventory of SOC (O'Neil et al. 2005,
2 Woodall et al. 201 Ib) in combination with auxiliary soil, site, and climate information (Domke et al. in prep). The
3 model used to estimate downed dead wood within the dead wood pool will be refined using nationwide inventory
4 data and auxiliary information similar to the litter and soil C pools (Domke et al. in review, Domke et al. in prep).
5 Finally, components of other pools, such as C in belowground biomass (Russell et al. 2015) and understory
6 vegetation (Russell et al. 2014), are being explored but may require additional investment in field inventories before
7 improvements can be realized with Inventory submissions.
8 The foundation of forest C accounting is the annual forest inventory system. The ongoing annual surveys by the FIA
9 Program are expected to improve the accuracy and precision of forest C estimates as new state surveys become
10 available (USDA Forest Service 2015b), particularly in western states. Hawaii and U.S. Territories will be included
11 when appropriate forest C data are available (as of July 21,2015, Hawaii is not yet reporting any data from the
12 annualized sampling design). Forest lands in interior Alaska (AK) are not yet included in this report as an annual
13 inventory has never been conducted in this remote region. A pilot study of an efficient method for inventorying
14 forest C stocks in interior AK (Woodall et al. 2015) has been conducted with results still being evaluated. Although
15 an annual forest inventory of interior AK may be implemented in the 2016 field season, alternative methods of
16 estimating C stock change will need to be explored as it may take over a decade to remeasure newly established
17 plots in the 2016 field season. Leveraging auxiliary information, particularly remotely sensed data (e.g., LiDar and
18 Landsat) and climate information will aid not only the interior AK effort but the entire inventory system. In addition,
19 to fully inventorying all managed forest land in the United States, the more intensive sampling of fine woody debris,
20 litter, and SOC on a subset of FIA plots continues and will substantially improve resolution of C pools (i.e., greater
21 sample intensity; Westfall et al. 2013) as this information becomes available (Woodall et al. 20 lib). Increased
22 sample intensity of some C pools and using annualized sampling data as it becomes available for those states
23 currently not reporting are planned for future submissions. The USDA Forest Service FIA Program's forest and
24 wooded land inventories extend beyond the forest land-use (e.g., woodlands and urban areas) with inventory -
25 relevant information for these lands which will likely become increasingly available in coming years.
26
Non-COz Emissions from Forest Fires
27 Emissions of non-CO2 gases from forest fires were estimated using U.S.-specific data for annual area of forest
28 burned and potential fuel availability as well as the default IPCC (2006) emissions and combustion factors applied to
29 the IPCC methodology. Emissions from this source in 2014 were estimated to be 7.3 MMT CO2 Eq. of CH4 and 4.8
30 MMT CO2 Eq. of N2O (Table 6-13, kt units available in Table 6-14). The estimates of non-CO2 emissions from
31 forest fires account for wildfires in the conterminous 48 states and Alaska as well as prescribed fires.
32 Table 6-13: Estimated Non-COz Emissions from Forest Fires (MMT COz Eq.) for U.S. Forests
Gas
CH4
N2O
Total
1990 •
3.3
2.2 •
5.4
2005
9.9
6.5 •
16.5
2010
3.3
2.2
5.4
2011
6.6
4.4
11.0
2012
11.1
7.3
18.3
2013
7.3
4.8
12.2
2014a
7.3
4.8
12.2
a The data for 2014 were incomplete when these estimates were summarized; therefore 2013, the most recent
available estimate, is applied to 2014.
33 Table 6-14: Estimated Non-COz Emissions from Forest Fires (kt) for U.S. Forests
~Gas 1990 2005 2010 2011 2012 2013 2014a
CH4 131 397 131 265 443 294 294
N2Q 7 22 7 15 24 16 16_
a The data for 2014 were incomplete when these estimates were summarized; therefore 2013, the most
recent available estimate, is applied to 2014.
34 Methodology
35 Non-CO2 emissions from forest fires - specifically for CH4 and N2O emissions - were calculated following IPCC
36 (2006) methodology, which included a combination of U.S. specific data on area burned and potential fuel available
6-30 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 for combustion along with IPCC default combustion and emission factors. The estimates were calculated according
2 to equation 2.27 of IPCC (2006, Volume 4, Chapter 2), which in general terms is:
3 Emissions = Area burned x Fuel available x Combustion factor x Emission factor x 10~3
4 where area burned is based on Monitoring Trends in Burn Severity (MTBS) data summaries (MTBS 2015), fuel
5 estimates are based on current inventory to carbon estimates as applied in EPA (2015), and combustion and
6 emission factors are from IPCC (2006, Volume 4, Chapter 2). See Annex 3.13 for further details.
7 Uncertainty and Time-Series Consistency
8 In order to quantify the uncertainties for emissions from forest fires calculated as described above, a Monte Carlo
9 ("Approach 2") sampling approach was employed to propagate uncertainty in the equation as it was applied for U.S.
10 forest land. See IPCC (2006) and Annex 3.13 for the quantities and assumptions employed to define and propagate
11 uncertainty. The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 6-15.
12 Table 6-15: Quantitative Uncertainty Estimates of Non-COz Emissions from Forest Fires in
13 Forest Land Remaining Forest Land (MWf COz Eq. and Percent)
-------
1 relating potential fuel to more localized forest structure is the best example of this. An additional improvement
2 would be combustion factors more locally appropriate for the type, location, and intensity of fire. All planned
3 improvements depend onfuture availability of appropriate U.S.-specific data.
4 N2O Fluxes from Forest Soils (IPCC Source Category 4A1)
5 Of the synthetic nitrogen (N) fertilizers applied to soils in the United States, no more than one percent is applied to
6 forest soils. Application rates are similar to those occurring on cropland soils, but in any given year, only a small
7 proportion of total forested land receives N fertilizer. This is because forests are typically fertilized only twice
8 during their approximately 40-year growth cycle (once at planting and once midway through their life cycle). While
9 the rate of N fertilizer application for the area of forests that receives N fertilizer in any given year is relatively high,
10 the annual application rate is quite low over the entire forestland area.
11 N additions to soils result in direct and indirect N2O emissions. Direct emissions occur on-site due to the N
12 additions. Indirect emissions result from fertilizer N that is transformed and transported to another location in a form
13 other than N2O (NH3 and NOX volatilization, NOs leaching and runoff), and later converted into N2O at the off-site
14 location. The indirect emissions are assigned to forest land because the management activity leading to the
15 emissions occurred in forest land.
16 Direct N2O emissions from forest soils in 2014 were 0.3 MMT CO2 Eq. (1 kt), and the indirect emissions were 0.1
17 MMT CO2 Eq. (0.4 kt). Total emissions for 2014 were 0.5 MMT CO2 Eq. (2 kt) and have increased by 455 percent
18 from 1990 to 2014. Increasing emissions over the time series is a result of greater area of N fertilized pine
19 plantations in the southeastern United States and Douglas-fir timberland in western Washington and Oregon. Total
20 forest soil N2O emissions are summarized in Table 6-16.
21 Table 6-16: NzO Fluxes from Soils in Forest Land Remaining Forest Land (MWT COz Eq. and
22 kt NzO)
1990 2005 2010 2011 2012 2013 2014
Direct N2O Fluxes from Soils
MMTC02Eq. 0.11 0.3M 0.3 0.3 0.3 0.3 0.3
ktN2O +• 11 1 1 1 1 1
Indirect N2O Fluxes from Soils
MMTCO2Eq. 0.0 O.ll 0.1 0.1 0.1 0.1 0.1
ktN2O + + + + + + +
Total
MMTC02
ktN2O
Eq.
0.5
2
0.5
2
0
.5
2
0
.5
2
0.5
2
0.5
2
+ Does not exceed 0.05 MMT CO2 Eq. or 0.5 kt.
23 Methodology
24 The IPCC Tier 1 approach is used to estimate N2O from soils within Forest Land Remaining Forest Land.
25 According to U.S. Forest Service statistics for 1996 (USDA Forest Service 2001), approximately 75 percent of trees
26 planted are for timber, and about 60 percent of national total harvested forest area is in the southeastern United
27 States. Although southeastern pine plantations represent the majority of fertilized forests in the United States, this
28 Inventory also accounted for N fertilizer application to commercial Douglas-fir stands in western Oregon and
29 Washington. For the Southeast, estimates of direct N2O emissions from fertilizer applications to forests are based on
30 the area of pine plantations receiving fertilizer in the southeastern United States and estimated application rates
31 (Albaugh et al. 2007; Fox et al. 2007). Not accounting for fertilizer applied to non-pine plantations is justified
32 because fertilization is routine for pine forests but rare for hardwoods (Binkley et al. 1995). For each year, the area
33 of pine receiving N fertilizer is multiplied by the weighted average of the reported range of N fertilization rates (121
34 Ibs. N per acre). Area data for pine plantations receiving fertilizer in the Southeast are not available for 2005-2014,
35 so data from 2004 are used for these years. For commercial forests in Oregon and Washington, only fertilizer
36 applied to Douglas-fir is addressed in the inventory because the vast majority (approximately 95 percent) of the total
37 fertilizer applied to forests in this region is applied to Douglas-fir (Briggs 2007). Estimates of total Douglas-fir area
3 8 and the portion of fertilized area are multiplied to obtain annual area estimates of fertilized Douglas-fir stands.
39 Similar to the Southeast, data are not available for 2005 through 2014, so data from 2004 are used for these years.
6-32 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 The annual area estimates are multiplied by the typical rate used in this region (200 Ibs. N per acre) to estimate total
2 N applied (Briggs 2007), and the total N applied to forests is multiplied by the IPCC (2006) default emission factor
3 of 1 percent to estimate direct N2O emissions.
4 For indirect emissions, the volatilization and leaching/runoff N fractions for forest land are calculated using the
5 IPCC default factors of 10 percent and 30 percent, respectively. The amount of N volatilized is multiplied by the
6 IPCC default factor of 1 percent for the portion of volatilized N that is converted to N2O off-site. The amount of N
7 leached/runoff is multiplied by the IPCC default factor of 0.075 percent for the portion of leached/runoff N that is
8 converted to N2O off-site The resulting estimates are summed to obtain total indirect emissions.
9 Uncertainty and Time-Series Consistency
10 The amount of N2O emitted from forests depends not only on N inputs and fertilized area, but also on a large
11 number of variables, including organic C availability, oxygen gas partial pressure, soil moisture content, pH,
12 temperature, and tree planting/harvesting cycles. The effect of the combined interaction of these variables on N2O
13 flux is complex and highly uncertain. IPCC (2006) does not incorporate any of these variables into the default
14 methodology, except variation in estimated fertilizer application rates and estimated areas of forested land receiving
15 N fertilizer. All forest soils are treated equivalently under this methodology. Furthermore, only synthetic N
16 fertilizers are captured, so applications of organic N fertilizers are not estimated. However, the total quantity of
17 organic N inputs to soils is included in the Agricultural Soil Management and Settlements Remaining Settlements
18 sections.
19 Uncertainties exist in the fertilization rates, annual area of forest lands receiving fertilizer, and the emission factors.
20 Fertilization rates are assigned a default level30 of uncertainty at ±50 percent, and area receiving fertilizer is
21 assigned a ±20 percent according to expert knowledge (Binkley 2004). The uncertainty ranges around the 2005
22 activity data and emission factor input variables are directly applied to the 2014 emission estimates. IPCC (2006)
23 provided estimates for the uncertainty associated with direct and indirect N2O emission factor for synthetic N
24 fertilizer application to soils.
25 Uncertainty is quantified using simple error propagation methods (IPCC 2006). The results of the quantitative
26 uncertainty analysis are summarized in Table 6-17. Direct N2O fluxes from soils in 2014 are estimated to be
27 between 0.1 and 1.1 MMT CO2 Eq. at a 95 percent confidence level. This indicates a range of 59 percent below and
28 211 percent above the 2014 emission estimate of 0.3 MMT CO2 Eq. Indirect N2O emissions in 2014 are between
29 0.02 and 0.4 MMT CO2Eq., ranging from 86 percent below to 238 percent above the 2014 emission estimate of 0.1
30 MMT CO2 Eq.
31 Table 6-17: Quantitative Uncertainty Estimates of NzO Fluxes from Soils in Forest Land
32 Remaining Forest Land (MWf COz Eq. and Percent)
-------
i QA/QC and Verification
2 The spreadsheet tab containing fertilizer applied to forests and calculations for N2O and uncertainty ranges are
3 checked and verified.
4 Planned Improvements
5 Additional data will be compiled to update estimates of forest areas receiving N fertilizer as new reports are made
6 available. Another improvement is to further disaggregate emissions by state for southeastern pine plantations and
7 northwestern Douglas-fir forests to estimate soil N2O emission. This improvement is contingent on the availability
8 of state-level N fertilization data for forest land.
9 6.3 Land Converted to Forest Land (IPCC
Source Category 4A2) (TO BE UPDATED)
10
12
13
14
11 Estimates for the Land Converted to Forest Land source category are currently under development.
6.4 Cropland Remaining Cropland (IPCC Source
Category 4B1)
Mineral and Organic Soil Carbon Stock Changes
15 Carbon (C) in cropland ecosystems occurs in bio mass, dead biomass, and soils. However, C storage in biomass and
16 dead organic matter is relatively ephemeral, with the exception of C stored in perennial woody crop biomass, such
17 as citrus groves and apple orchards. Within soils, C is found in organic and inorganic forms of C, but soil organic C
18 (SOC) is the main source and sink for atmospheric CC>2 in most soils. IPCC (2006) recommends reporting changes
19 in SOC stocks due to agricultural land-use and management activities on both mineral and organic soils.31
20 Well-drained mineral soils typically contain from 1 to 6 percent organic C by weight, whereas mineral soils with
21 high water tables for substantial periods during the year may contain significantly more C (NRCS 1999).
22 Conversion of mineral soils from their native state to agricultural land uses can cause up to half of the SOC to be
23 lost to the atmosphere due to enhanced microbial decomposition. The rate and ultimate magnitude of C loss
24 depends on subsequent management practices, climate and soil type (Ogle et al. 2005). Agricultural practices, such
25 as clearing, drainage, tillage, planting, grazing, crop residue management, fertilization, and flooding, can modify
26 both organic matter inputs and decomposition, and thereby result in a net C stock change (Parton et al. 1987,
27 Paustian et al. 1997a, Conant et al. 2001, Ogle et al. 2005). Eventually, the soil can reach a new equilibrium that
28 reflects a balance between C inputs (e.g., decayed plant matter, roots, and organic amendments such as manure and
29 crop residues) and C loss through microbial decomposition of organic matter (Paustian et al. 1997b).
30 Organic soils, also referred to as histosols, include all soils with more than 12 to 20 percent organic C by weight,
31 depending on clay content (NRCS 1999, Brady and Weil 1999). The organic layer of these soils can be very deep
32 (i.e., several meters), and form under inundated conditions that results in minimal decomposition of plant residues.
33 When organic soils are prepared for crop production, they are drained and tilled, leading to aeration of the soil that
34 accelerates both the decomposition rate and CO2 emissions. Due to the depth and richness of the organic layers, C
35 loss from drained organic soils can continue over long periods of time, which varies depending on climate and
31 CO2 emissions associated with liming are also estimated but are included in a separate section of the report.
6-34 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 composition (i.e., decomposability) of the organic matter (Armentano and Menges 1986). Due to deeper drainage
2 and more intensive management practices, the use of organic soils for annual crop production leads to higher C loss
3 rates than drainage of organic soils in grassland or forests (IPCC 2006).
4 Cropland Remaining Cropland includes all cropland in an Inventory year that has been used as cropland for the
5 previous 20 years according to the 2010 USDA National Resources Inventory (NRI) land-use survey (USDA-NRCS
6 2013).32 The inventory includes croplands on privately-owned lands, as well as a small amount of cropland on
7 public federal lands in the conterminous United States and Hawaii. Cropland in Alaska is not included in the
8 inventory, but is a relatively small amount of United States cropland area (Approximately 28,700 hectares). Some
9 miscellaneous croplands are also not included in the Inventory due to limited understanding of greenhouse gas
10 emissions from these management systems (e.g., aquaculture). This leads to a small discrepancy between the total
11 amount of managed area in Cropland Remaining Cropland (see Section6.1 -Representation of the U.S. LandBase)
12 and the cropland area included in the Inventory analysis (0.5 to 0.7 million hectares between 1990 and 2014).
13 Improvements are underway to include croplands in Alaska and other miscellaneous cropland areas as part of future
14 C inventories.
15 Carbon dioxide emissions and removals33 due to changes in mineral soil C stocks are estimated using a Tier 3
16 approach for the majority of annual crops (Ogle et al. 2010). A Tier 2 IPCC method is used for the remaining crops
17 not included in the Tier 3 method (see methodology section for a list of crops in the Tier 2 and 3 methods) (Ogle et
18 al. 2003, 2006). In addition, a Tier 2 method is used for very gravelly, cobbly, or shaley soils (i.e., classified as soils
19 that have greater than 3 5 percent of soil volume comprised of gravel, cobbles, or shale) regardless of crop, and for
20 additional changes in mineral soil C stocks that are not addressed with the Tier 3 approach (i.e., change in C stocks
21 after 2010 due to Conservation Reserve Program enrollment). Emissions from organic soils are estimated using a
22 Tier 2 IPCC method.
23 Land-use and land management of mineral soils are the largest contributor to total net C stock change, especially in
24 the early part of the time series (see Table 6-18 and Table 6-19). (Note: Estimates after 2010 are based on NRI data
25 from 2010 and therefore do not fully reflect changes occurring in the latter part of the time series). In 2014, mineral
26 soils are estimated to remove 43.8 MMT CO2 Eq. (11.9 MMT C). This rate of C storage in mineral soils represents
27 about a 39 percent decrease in the rate since the initial reporting year of 1990. Emissions from organic soils are 27.8
28 MMT CO2 Eq. (7.6 MMT C) in 2014, which is a 0.8 percent decrease compared to 1990. In total, United States
29 agricultural soils in Cropland Remaining Cropland sequestered approximately 16.0 MMT CO2 Eq. (4.4 MMT C) in
30 2014.
31 Table 6-18: Net COz Flux from Soil C Stock Changes in Cropland Remaining Crop/and (MMJ
32 COz Eq.)
33
34
35
Soil Type
Mineral Soils
Organic Soils
Total Net Flux
1990
(71.2)
28.0
(43.2)
2005
(45.2)
28.7
(16.5)
2010
(32.5)
27.8
(4.7)
2011
(47.9)
27.8
(20.0)
2012
(46.5)
27.8
(18.7)
2013
(44.6)
27.8
(16.8)
2014
(43.8)
27.8
(16.0)
Notes: Totals may not sum due to independent rounding. Parentheses indicate net
sequestration.
Estimates after 2010 are based on NRI data from 2010 and therefore may not fully reflect
changes occurring in the latter part of the time series.
NRI survey locations are classified according to land-use histories starting in 1979, and consequently the classifications are
based on less than 20 years from 1990 to 2001. This may have led to an overestimation of Cropland Remaining Cropland in the
early part of the time series to the extent that some areas are converted to cropland prior to 1979.
Note that removals occur through uptake of CCh into crop and forage biomass that is later incorporated into soil C pools.
Land Use, Land-Use Change, and Forestry 6-35
-------
1 Table 6-19: Net COz Flux from Soil C Stock Changes in Cropland Remaining Crop/and (MMT
2 C)
Soil Type
Mineral Soils
Organic Soils
Total Net Flux
1990
(19.4)
7.6
(11.8)
2005
(12.3) 1
7.8
(4.5)
2010
(8.9)
1 7.6
(1.3)
2011
(13.1)
7.6
(5.5)
2012
(12.7)
7.6
(5.1)
2013
(12.2)
7.6
(4.6)
2014
(11.9)
7.6
(4.4)
Notes: Totals may not sum due to independent rounding. Parentheses indicate net
sequestration.
Estimates after 2010 are based on NRI data from 2010 and therefore may not fully reflect
changes occurring in the latter part of the time series.
3 The major cause of the reduction in soil C accumulation over the time series (i.e., 2014 is 63 percent less than 1990)
4 is the decline in annual cropland enrolled in the Conservation Reserve Program (CRP)34 which was initiated in 1985
5 (Jones et al, in prep). For example, over 2 million hectares of land in the CRP were returned to agricultural
6 production during the last 5 years resulting in a loss of soil C. However, positive increases in C stocks continue on
7 the nearly 10 million hectares of land currently enrolled in the CRP, as well as from intensification of crop
8 production by limiting the use of bare-summer fallow in semi-arid regions, increased hay production, and adoption
9 of conservation tillage (i.e., reduced- and no-till practices).
10 The spatial variability in the 2014 annual CC>2 stock changes are displayed in Figure 6-5 and Figure 6-6 for C stock
11 changes in mineral and organic soils, respectively. The highest rates of net C accumulation in mineral soils occurred
12 in the Midwest, which is the region with the largest amounts of conservation tillage, with the next highest rates of
13 accumulation in the South-central and Northwest regions of the United States. The regions with the highest rates of
14 emissions from organic soils occur in the Southeastern Coastal Region (particularly Florida), upper Midwest and
15 Northeast surrounding the Great Lakes, and the Pacific Coast (particularly California), which coincides with largest
16 concentrations of organic soils in the United States that are used for agricultural production.
17 Figure 6-5: Total Net Annual COz Flux for Mineral Soils under Agricultural Management
18 within States, 2014, Cropland Remaining CroplandtfQ BE UPDATED)
19
20 Figure 6-6: Total Net Annual COz Flux for Organic Soils under Agricultural Management
21 within States, 2014, Cropland Remaining CroplandtfQ BE UPDATED)
22
23 Methodology
24 The following section includes a description of the methodology used to estimate changes in soil C stocks for
25 Cropland Remaining Cropland, including (1) agricultural land-use and management activities on mineral soils; and
26 (2) agricultural land-use and management activities on organic soils.
27 Soil C stock changes are estimated for Cropland Remaining Cropland (as well as agricultural land falling into the
28 IPCC categories Land Converted to Cropland, Grassland Remaining Grassland, and Land Converted to Grassland)
29 according to land-use histories recorded in the USDA NRI survey (USDA-NRCS 2013). The NRI is a statistically-
30 based sample of all non-federal land, and includes approximately 596,787 survey locations in agricultural land for
31 the conterminous United States and Hawaii.35 Each survey location is associated with an "expansion factor" that
32 allows scaling of C stock changes from NRI survey locations to the entire country (i.e., each expansion factor
33 represents the amount of area with the same land-use/management history as the sample point). Land-use and some
34 The Conservation Reserve Program (CRP) is a land conservation program administered by the Farm Service Agency (FSA).
In exchange for a yearly rental payment, farmers enrolled in the program agree to remove environmentally sensitive land from
agricultural production and plant species that will improve environmental health and quality. Contracts for land enrolled in CRP
are 10-15 years in length. The long-term goal of the program is to re-establish valuable land cover to help improve water quality,
prevent soil erosion, and reduce loss of wildlife habitat.
35 NRI survey locations are classified as agricultural if under grassland or cropland management between 1990 and 2010.
6-36 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 management information (e.g., crop type, soil attributes, and irrigation) were collected for each NRI point on a 5-
2 year cycle beginning from 1982 through 1997. For cropland, data had been collected for 4 out of 5 years during
3 each survey cycle (i.e., 1979 through 1982, 1984 through 1987, 1989 through 1992, and 1994 through!997). In
4 1998, the NRI program began collecting annual data, and the annual data are currently available through 2012
5 (USDA-NRCS 2015). However, this Inventory only uses NRI data through 2010 because newer data were not
6 available in time to incorporate the additional years. NRI survey locations are classified as Cropland Remaining
1 Cropland in a given year between 1990 and 2010 if the land use had been cropland for a continuous time period of
8 at least 20 years.36 Cropland includes all land used to produce food and fiber, in addition to forage that is harvested
9 and used as feed (e.g., hay and silage), and cropland that has been enrolled in the Conservation Reserve Program
10 (i.e., considered reserve cropland).
11 Mineral Soil Carbon Stock Changes
12 An IPCC Tier 3 model-based approach (Ogle et al. 2010) is applied to estimate C stock changes for mineral soils on
13 the majority of land that is used to produce annual crops in the United States. These crops include alfalfa hay,
14 barley, corn, cotton, dry beans, grass hay, grass-clover hay, oats, onions, peanuts, potatoes, rice, sorghum, soybeans,
15 sugar beets, sunflowers, tomatoes, and wheat. The model-based approach uses the DAYCENT biogeochemical
16 model (Parton et al. 1998; Del Grosso et al. 2001, 2011) to estimate soil C stock changes and soil nitrous oxide
17 emissions from agricultural soil management. Carbon and N dynamics are linked in plant-soil systems through the
18 biogeochemical processes of microbial decomposition and plant production (McGill and Cole 1981). Coupling the
19 two source categories (i.e., agricultural soil C and N2O) in a single inventory analysis ensures that there is a
20 consistent treatment of the processes and interactions between C and N cycling in soils.
21 The remaining crops on mineral soils are estimated using an IPCC Tier 2 method (Ogle et al. 2003), including some
22 vegetables, tobacco, perennial/horticultural crops, and crops that are rotated with these crops. The Tier 2 method is
23 also used for very gravelly, cobbly, or shaley soils (greater than 35 percent by volume), and stock changes on federal
24 croplands are estimated with the Tier 2 method. Mineral SOC stocks are estimated using a Tier 2 method for these
25 areas because the DAYCENT model, which is used for the Tier 3 method, has not been fully tested for estimating C
26 stock changes associated with these crops and rotations, as well as cobbly, gravelly, or shaley soils. In addition,
27 there is insufficient information to simulate croplands on federal lands. The Tier 2 methods is also used to estimate
28 additional stock changes on lands enrolled in CRP after 2010, which is the last year of data in the NRI time series,
29 using aggregated data on CRP enrollment compiled by the USD A Farm Services Agency.
30 Further elaboration on the methodology and data used to estimate stock changes from mineral soils are described
31 below and in Annex 3.12.
32 Tier 3 Approach
33 Mineral SOC stocks and stock changes are estimated using the DAYCENT biogeochemical37 model (Parton et al.
34 1998; Del Grosso et al. 2001, 2011), which simulates cycling of C, N and other nutrients in cropland, grassland,
35 forest, and savanna ecosystems. The DAYCENT model utilizes the soil C modeling framework developed in the
36 Century model (Parton et al. 1987, 1988, 1994; Metherell et al. 1993), but has been refined to simulate dynamics at a
37 daily time-step. The modeling approach uses daily weather data as an input, along with information about soil
38 physical properties. Input data on land use and management are specified at a daily resolution and include land-use
39 type, crop/forage type, and management activities (e.g., planting, harvesting, fertilization, manure amendments,
40 tillage, irrigation, residue removal, grazing, and fire). The model simulates net primary productivity using the
41 NASA-CASA production algorithm for most croplands38 (Potter et al. 1993, Potter et al. 2007) and subsequent C
NRI survey locations are classified according to land-use history records starting in 1982 when the NRI survey began.
Therefore, the classification prior to 2002 was based on less than 20 years of recorded land-use history for the time series.
37 Biogeochemical cycles are the flow of chemical elements and compounds between living organisms and the physical
environment
NPP is estimated with the NASA-CASA algorithm for most of the cropland that is used to produce major commodity crops in
the central United States from 2000-2010. Other regions and years prior to 2000 are simulated with a method that incorporates
water, temperature and moisture stress on crop production (see Metherell et al. 1993), but does not incorporate the additional
information about crop condition provided with remote sensing data.
Land Use, Land-Use Change, and Forestry 6-37
-------
1 additions to soil. The model also simulates soil temperature, and water dynamics, in addition to turnover,
2 stabilization, and mineralization of soil organic matter C and nutrients (N, P, K, S). This method is more accurate
3 than the Tier 1 and 2 approaches provided by the IPCC (2006) because the simulation model treats changes as
4 continuous over time as opposed to the simplified discrete changes represented in the default method (see Box 6-4
5 for additional information).
Box 6-4: Tier 3 Approach for Soil C Stocks Compared to Tier 1 or 2 Approaches
8 A Tier 3 model-based approach is used to estimate soil C stock changes on the majority of agricultural land on
9 mineral soils. This approach results in a more complete accounting of soil C stock changes and entails several
10 fundamental differences from the IPCC Tier 1 or 2 methods, as described below.
11 (1) The IPCC Tier 1 and 2 methods are simplified and classify land areas into discrete categories based on
12 highly aggregated information about climate (six regions), soil (seven types), and management (eleven
13 management systems) in the United States. In contrast, the Tier 3 model incorporates the same variables
14 (i.e., climate, soils, and management systems) with considerably more detail both temporally and spatially,
15 and captures multi-dimensional interactions through the more complex model structure.
16 (2) The IPCC Tier 1 and 2 methods have a simplified spatial resolution in which data are aggregated to soil
17 types in climate regions, and there about 30 of combinations in the United States. In contrast, the Tier 3
18 model simulates soil C dynamics at more than 300,000 individual NPJ survey locations in individual fields.
19 (3) The IPCC Tier 1 and 2 methods use simplified equilibrium step changes for changes in carbon emissions.
20 In contrast, the Tier 3 approach simulates a continuous time period. More specifically, the DAYCENT
21 model (i.e., daily time-step version of the Century model) simulates soil C dynamics (and CC>2 emissions
22 and uptake) on a daily time step based on C emissions and removals from plant production and
23 decomposition processes. These changes in soil C stocks are influenced by multiple sources that affect
24 primary production and decomposition, including changes in land use and management, weather variability
25 and secondary feedbacks between management activities, climate, and soils.
26
27 Historical land-use patterns and irrigation histories are simulated with DAYCENT based on the 2010 USD A NPJ
28 survey (USDA-NRCS 2013). Additional sources of activity data are used to supplement the land-use information
29 from NPJ. The Conservation Technology Information Center (CTIC 2004) provided annual data on tillage activity
30 at the county level for the conterminous United States between 1989 and 2004, and these data are adjusted for long-
31 term adoption of no-till agriculture (Towery 2001). Information on fertilizer use and rates by crop type for different
32 regions of the United States are obtained primarily from the USDA Economic Research Service Cropping Practices
33 Survey (USDA-ERS 1997, 2011) with additional data from other sources, including the National Agricultural
34 Statistics Service (NASS 1992, 1999, 2004). Frequency and rates of manure application to cropland during 1997 are
35 estimated from data compiled by the USDA Natural Resources Conservation Service (Edmonds et al. 2003), and
36 then adjusted using county-level estimates of manure available for application in other years. Specifically, county -
37 scale ratios of manure available for application to soils in other years relative to 1997 are used to adjust the area
38 amended with manure (see Annex 3.12 for further details). Greater availability of managed manure N relative to
39 1997 is assumed to increase the area amended with manure, while reduced availability of manure N relative to 1997
40 is assumed to reduce the amended area. Data on the county-level N available for application are estimated for
41 managed systems based on the total amount of N excreted in manure minus N losses during storage and transport,
42 and including the addition of N from bedding materials. Nitrogen losses include direct N2O emissions, volatilization
43 of ammonia and NOX, runoff and leaching, and poultry manure used as a feed supplement. More information on
44 livestock manure production is available in Section 5.2- Manure Management and Annex 3.11.
45 Daily weather data are another input to the model simulations, and these data are based on a 4 km gridded product
46 from the PRISM Climate Group (2015). Soil attributes are obtained from the Soil Survey Geographic Database
47 (SSURGO) (Soil Survey Staff 2015). The C dynamics at each NRI point are simulated 100 times as part of the
48 uncertainty analysis, yielding a total of over 18 million simulation runs for the analysis. Uncertainty in the C stock
49 estimates from DAYCENT associated with parameterization and model algorithms are adjusted using a structural
50 uncertainty estimator accounting for uncertainty in model algorithms and parameter values (Ogle et al. 2007, 2010).
51 Carbon stocks and 95 percent confidence intervals are estimated for each year between 1990 and 2010. C stock
6-38 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 changes from 2011 to 2014 are assumed to be similar to 2010 for this Inventory. Future Inventories will be updated
2 with new activity data and the time series will be recalculated (see Planned Improvements section).
3 Tier 2 Approach
4 In the IPCC Tier 2 method, data on climate, soil types, land-use, and land management activity are used to classify
5 land area and apply appropriate stock change factors (Ogle et al. 2003, 2006). Reference C stocks are estimated
6 using the National Soil Survey Characterization Database (NRCS 1997) with cultivated cropland as the reference
7 condition, rather than native vegetation as used in IPCC (2006). Soil measurements under agricultural management
8 are much more common and easily identified in the National Soil Survey Characterization Database (NRCS 1997)
9 than are soils under a native condition, and therefore cultivated cropland provided a more robust sample for
10 estimating the reference condition. U.S.-specific stock change factors are derived from published literature to
11 determine the impact of management practices on SOC storage (Ogle et al. 2003, Ogle et al. 2006). The factors
12 include changes in tillage, cropping rotations, intensification, and land-use change between cultivated and
13 uncultivated conditions. U.S. factors associated with organic matter amendments are not estimated due to an
14 insufficient number of studies in the United States to analyze the impacts. Instead, factors from IPCC (2006) are
15 used to estimate the effect of those activities.
16 Climate zones in the United States are classified using mean precipitation and temperature (1950-2000) variables
17 from the WorldClim data set (Hijmans et al. 2005) and potential evapotranspiration data from the Consortium for
18 Spatial Information (CGIAR-CSI) (Zomer et al. 2008, Zomer et al. 2007) (Figure A-14). IPCC climate zones are
19 then assigned to NRI point locations.
20 Activity data are primarily based on the historical land-use/management patterns recorded in the 2010 NRI (USDA-
21 NRCS 2013). Each NRI point is classified by land use, soil type, climate region, and management condition.
22 Survey locations on federal lands are included in the NRI, but land use and cropping history are not compiled at
23 these locations in the survey program (i.e., NRI is restricted to data collection on non-federal lands). Land use
24 patterns at the NRI survey locations on federal lands are based on the National Land Cover Database (NLCD) (Fry et
25 al. 2011; Homer et al. 2007; Homer et al. 2015). Classification of cropland area by tillage practice is based on data
26 from the Conservation Technology Information Center (CTIC 2004, Towery 2001) as described above. Activity
27 data on wetland restoration of Conservation Reserve Program land are obtained from Euliss and Gleason (2002).
28 Manure N amendments over the inventory time period are based on application rates and areas amended with
29 manure N from Edmonds et al. (2003), in addition to the managed manure production data discussed in the
30 methodology subsection for the Tier 3 analysis.
31 Combining information from these data sources, SOC stocks for mineral soils are estimated 50,000 times for 1985,
32 1990, 1995, 2000, 2005, and 2010, using a Monte Carlo stochastic simulation approach and probability distribution
33 functions for U.S.-specific stock change factors, reference C stocks, and land-use activity data (Ogle et al. 2002,
34 Ogle et al. 2003, Ogle et al. 2006). The annual C stock changes for 1990 through 2014 for the Tier 2 method are
35 determined by calculating the average annual change in stocks between each of the five-year blocks, and assuming
36 that C stock changes from 2011-2014 are similar to 2006-2010.
37 Additional Mineral C Stock Change
38 Annual C stock change estimates for mineral soils between 2011 and 2014 are adjusted to account for additional C
39 stock changes associated with gains or losses in soil C after 2010 due to changes in CRP enrollment (USDA-FSA
40 2014). The change in enrollment relative to 2010 is based on data from USDA-FSA (2014) for 2011 through 2014.
41 The differences in mineral soil areas are multiplied by 0.5 metric tons C per hectare per year to estimate the net
42 effect on soil C stocks. The stock change rate is based on country-specific factors and the IPCC default method (see
43 Annex 3.12 for further discussion).
44 Organic Soil Carbon Stock Changes
45 Annual C emissions from drained organic soils in Cropland Remaining Cropland are estimated using the Tier 2
46 method provided in IPCC (2006), with U.S.-specific C loss rates (Ogle et al. 2003) rather than default IPCC rates.
47 The final estimates included a measure of uncertainty as determined from the Monte Carlo Stochastic Simulation
48 with 50,000 iterations. Emissions are based on the annual data for drained organic soils from 1990 to 2010 for
Land Use, Land-Use Change, and Forestry 6-39
-------
1 Cropland Remaining Cropland areas in the 2010 NRI (USDA-NRCS 2013). The annual emissions estimated for
2 2010 are applied to 2011 through 2014.39
3 Uncertainty and Time-Series Consistency
4 Uncertainty associated with the Cropland Remaining Cropland land-use category is addressed for changes in
5 agricultural soil C stocks (including both mineral and organic soils). Uncertainty estimates are presented in Table
6 6-20 for each subsource (mineral soil C stocks and organic soil C stocks) and the method that is used in the
7 inventory analysis (i.e., Tier 2 and Tier 3). Uncertainty for the Tier 2 and 3 approaches is derived using a Monte
8 Carlo approach (see Annex 3.12 for further discussion), but the C stock changes from the individual Tier 2 and 3
9 approaches are combined using the simple error propagation method provided by the IPCC (2006). The combined
10 uncertainty is calculated by taking the square root of the sum of the squares of the standard deviations of the
11 uncertain quantities. The combined uncertainty for soil C stocks in Cropland Remaining Cropland ranged from 211
12 percent below to 217 percent above the 2014 stock change estimate of -16.0 MMT CCh Eq.
13 Table 6-20: Approach 2 Quantitative Uncertainty Estimates for Soil C Stock Changes
14 occurring within Cropland Remaining Crop/and (MMJ COz Eq. and Percent)
2014 Flux Estimate Uncertainty Range Relative to Flux Estimate3
CC (MMT CCh Eq.) (MMT CCh Eq.) (%)
Mineral Soil C Stocks: Cropland Remaining
Cropland, Tier 3 Inventory Methodology
Mineral Soil C Stocks: Cropland Remaining
Cropland, Tier 2 Inventory Methodology
Mineral Soil C Stocks: Cropland Remaining
Cropland (Change in CRP enrollment relative
to 2003)
Organic Soil C Stocks: Cropland Remaining
Cropland, Tier 2 Inventory Methodology
Combined Uncertainty for Flux associated
with Agricultural Soil Carbon Stock
Change in Cropland Remaining Cropland
a Range of C stock change estimates predicted by
Notes: Parentheses indicate net sequestration.
(44.3)
(3.2)
3.7
27.8
(16.0)
Monte Carlo Stochastic
Lower
Bound
(76.4)
(5.2)
1.9
17.8
(49.7)
Simulation
Upper
Bound
(12.2)
(1.5)
5.6
41.0
18.8
Lower
Bound
-72%
-64%
-50%
-36%
-211%
for a 95 percent confidence
Upper
Bound
+72%
+54%
+50%
+48%
+217%
interval.
15 Uncertainty is also associated with lack of reporting of agricultural biomass and litter C stock changes. Biomass C
16 stock changes are likely minor in perennial crops, such as orchards and nut plantations, given the small amount of
17 change in land that is used to produce these commodities in the United States. In contrast, agroforestry practices,
18 such as shelterbelts, riparian forests and intercropping with trees, may be significantly changing biomass C stocks
19 over the Inventory times series, at least in some regions of the United States, but there are currently no datasets to
20 evaluate the trends. Changes in litter C stocks are also assumed to be negligible in croplands over annual time
21 frames, although there are certainly significant changes at sub-annual time scales across seasons. However, this
22 trend may change in the future, particularly if crop residue becomes a viable feedstock for bioenergy production.
23 Methodological recalculations are applied to the entire time series to ensure time-series consistency from 1990
24 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
25 above.
Future inventories will be updated with new activity data and the time series will be recalculated; see Planned Improvements
section.
6-40 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
i QA/QC and Verification
2 Quality control measures included checking input data, model scripts, and results to ensure data are properly
3 handled throughout the inventory process. Inventory reporting forms and text are reviewed and revised as needed to
4 correct transcription errors. Results from the D AYCENT model are compared to field measurements, and a
5 statistical relationship has been developed to assess uncertainties in the predictive capability of the model. The
6 comparisons included over 45 long-term experiments, representing about 800 combinations of management
7 treatments across all of the sites (Ogle et al. 2007) (See Annex 3.12 for more information). Quality control
8 identified problems with simulation of hydric soils in the equilibrium and base histories, which proceed the
9 simulation of the NRI histories from 1979 to 2010. Hydric soils were draining more quickly than expected in the
10 simulations, and so parameters were adjusted to reduce the drainage rate on these soils.
11 Recalculations Discussion
12 Methodological recalculations in the current Inventory are associated with the following improvements: 1)
13 incorporation of updated NRI data for 1990 through 2010; and 2) inclusion of federal croplands; and 3) improving
14 the simulation of hydric soil. As a result of these changes, the change in SOC stocks declined by an average of 8.4
15 MMT CO2 Eq. over the time series.
16 Planned Improvements
17 Two major planned improvements are underway. The first is to update the time series of land use and management
18 data from the USD A NRI so that it is extended from 2010 through 2012 for both the Tier 2 and 3 methods (USDA-
19 NRCS 2015). Fertilization and tillage activity data will also be updated as part of this improvement. The remote-
20 sensing based data on the Enhanced Vegetation Index will be extended through 2012 in order to use the EVI data to
21 drive crop production in D AYCENT. Overall, this improvement will extend the time series of activity data for the
22 Tier 2 and 3 analyses through 2012.
23 The second major planned improvement is to analyze C stock changes in Alaska for cropland and managed
24 grassland, using the Tier 2 method for mineral and organic soils that is described earlier in this section. This
25 analysis will initially focus on land use change, which typically has a larger impact on soil C stock changes, but will
26 be further refined over time to incorporate more of the management data.
27 An improvement is also underway to simulate crop residue burning in the D AYCENT based on the amount of crop
28 residues burned according to the data that is used in the Field Burning of Agricultural Residues source category
29 (Section 5.5). This improvement will more accurately represent the C inputs to the soil that are associated with
30 residue burning. Other improvements are underway to refine the production part of the D AYCENT biogeochemical
31 model. For example, senescence events following grain filling in crops, such as wheat, have been refined based on
32 recent model algorithm development, and will be incorporated into next year's Inventory.
33 All of these improvements are expected to be completed for the 1990 through 2015 Inventory. However, the time
34 line may be extended if there are insufficient resources to fund all or part of these planned improvements.
35 CO2 Emissions from Liming
36 IPCC (2006) recommends reporting CO2 emissions from lime additions (in the form of crushed limestone (CaCOs)
37 and dolomite (CaMg(CO3)2) to soils. Limestone and dolomite are added by land managers to increase soil pH (i.e.,
38 to reduce acidification). Carbon dioxide emissions occur as these compounds react with hydrogen ions in soils. The
39 rate and ultimate magnitude of degradation of applied limestone and dolomite depends on the soil conditions, soil
40 type, climate regime, and whether limestone or dolomite is applied. Emissions from liming of soils have fluctuated
41 over the past 24 years, ranging from 3.7 MMT CO2 Eq. to 6.0 MMT CO2 Eq. In 2014, liming of soils in the United
42 States resulted in emissions of 4.1 MMT CO2 Eq. (1.1 MMT C), representing an 11 percent decrease in emissions
43 since 1990 (see Table 6-21 and Table 6-22). The trend is driven by the amount of limestone and dolomite applied to
44 soils over the time period.
Land Use, Land-Use Change, and Forestry 6-41
-------
Table 6-21: Emissions from Liming (MMT COz Eq.)
Source
Limestone
Dolomite
Total3
1990 2005
4.1 3.9
0.6 0.4
4.7 4.3
2010 2011
4.3 3
B 0.5 0
4.8 3
4
4
9
2012
4.5
1.5
6.0
2013
3.6
0.3
3.9
2014
3.8
0.3
4.1
a Also includes emissions from liming on Land Converted to Cropland, Grassland
Remaining Grassland, Land Converted to Grassland, Settlements Remaining Settlements,
Forest Land Remaining Forest Land and Land Converted to Forest Land as it is not
currently possible to apportion the data by land-use category.
Note: Totals may not sum due to independent rounding.
Table 6-22: Emissions from Liming (MMT C)
Source
Limestone
Dolomite
Total3
1990
1.1
0.2
1.3
2005
1.1
0.1
1.2
2010
1.2
0.1
1.3
2011
0.9
0.1
1.1
2012
1.2
0.4
1.6
2013
1.0
0.1
1.1
2014
1.0
0.1
1.1
a Also includes emissions from liming on Land Converted to Cropland, Grassland
Remaining Grassland, Land Converted to Grassland, and Settlements Remaining
Settlements, Forest Land Remaining Forest Land and Land Converted to Forest Land as it is
not currently possible to apportion the data by land-use category.
Note: Totals may not sum due to independent rounding.
3 Methodology
4 Carbon dioxide emissions from application of limestone and dolomite to soils were estimated using a Tier 2
5 methodology consistent with IPCC (2006). The annual amounts of limestone and dolomite applied (see Table 6-23)
6 were multiplied by CCh emission factors from West and McBride (2005). These emission factors (0.059 metric ton
7 C/metric ton limestone, 0.064 metric ton C/metric ton dolomite) are lower than the IPCC default emission factors
8 because they account for the portion of carbonates that are transported from soils through hydrological processes
9 and eventually deposited in ocean basins (West and McBride 2005). This analysis of lime dissolution is based on
10 studies in the Mississippi River basin, where the vast majority of lime application occurs in the United States (West
11 2008). Moreover, much of the remaining lime application is occurring under similar precipitation regimes, and so
12 the emission factors are considered a reasonable approximation for all lime application in the United States (West
13 2008).
14 The annual application rates of limestone and dolomite were derived from estimates and industry statistics provided
15 in the Minerals Yearbook and Mineral Industry Surveys (Tepordei 1993 through 2006; Willett 2007a, 2007b, 2009,
16 2010, 2011a, 20lib, 2013a, 2014 and 2015; USGS 2008 through 2015). The U.S. Geological Survey (USGS; U.S.
17 Bureau of Mines prior to 1997) compiled production and use information through surveys of crushed stone
18 manufacturers. However, manufacturers provided different levels of detail in survey responses so the estimates of
19 total crushed limestone and dolomite production and use were divided into three components: (1) production by end-
20 use, as reported by manufacturers (i.e., "specified" production); (2) production reported by manufacturers without
21 end-uses specified (i.e., "unspecified" production); and (3) estimated additional production by manufacturers who
22 did not respond to the survey (i.e., "estimated" production).
Box 6-5: Comparison of the Tier 2 U.S. Inventory Approach and IPCC (2006) Default Approach
24 Emissions from liming of soils were estimated using a Tier 2 methodology based on emission factors specific to the
25 United States that are lower than the IPCC (2006) emission default factors. Most lime application in the United
26 States occurs in the Mississippi River basin, or in areas that have similar soil and rainfall regimes as the Mississippi
27 River basin. Under these conditions, a significant portion of dissolved agricultural lime leaches through the soil into
28 groundwater. Groundwater moves into channels and is transported to larger rives and eventually the ocean where
6-42 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 CaCO3 precipitates to the ocean floor (West and McBride 2005). The U.S. specific emission factors (0.059 metric
2 ton C/metric ton limestone and 0.064 metric ton C/metric ton dolomite) are about half of the IPCC (2006) emission
3 factors (0.12 metric ton C/metric ton limestone and 0.13 metric ton C/metric ton dolomite). For comparison, the
4 2014 U.S. emissions estimate from liming of soils is 4.1 MMT CCh Eq. using the U.S.-specific factors. In contrast,
5 emissions would be estimated at 8.4 MMT CCh Eq. using the IPCC (2006) default emission factors.
7 Data on "specified" limestone and dolomite amounts were used directly in the emission calculation because the end
8 use is provided by the manufactures and can be used to directly determine the amount applied to soils. However, it
9 is not possible to determine directly how much of the limestone and dolomite is applied to soils for manufacturer
10 surveys in the "unspecified" and "estimated" categories. For these categories, the amounts of crushed limestone and
11 dolomite applied to soils were determined by multiplying the percentage of total "specified" limestone and dolomite
12 production applied to soils by the total amounts of "unspecified" and "estimated" limestone and dolomite
13 production. In other words, the proportion of total "unspecified" and "estimated" crushed limestone and dolomite
14 that was applied to soils is proportional to the amount of total "specified" crushed limestone and dolomite that was
15 applied to soils.
16 In addition, data were not available for 1990, 1992, and 2013 on the fractions of total crushed stone production that
17 were limestone and dolomite, and on the fractions of limestone and dolomite production that were applied to soils.
18 To estimate the 1990 and 1992 data, a set of average fractions were calculated using the 1991 and 1993 data. These
19 average fractions were applied to the quantity of "total crushed stone produced or used" reported for 1990 and 1992
20 in the 1994 Minerals Yearbook (Tepordei 1996). To estimate 2014 data, 2013 fractions were applied to a 2014
21 estimate of total crushed stone presented in the USGS Mineral Industry Surveys: Crushed Stone and Sand and
22 Gravel in the First Quarter of 2015 (USGS 2015).
23 The primary source for limestone and dolomite activity data is the Minerals Yearbook, published by the Bureau of
24 Mines through 1994 and by the USGS from 1995 to the present. In 1994, the "Crushed Stone" chapter in the
25 Minerals Yearbook began rounding (to the nearest thousand metric tons) quantities for total crushed stone produced
26 or used. It then reported revised (rounded) quantities for each of the years from 1990 to 1993. In order to minimize
27 the inconsistencies in the activity data, these revised production numbers have been used in all of the subsequent
28 calculations.
29 Emissions from limestone and dolomite are estimated at the state level and summed to obtain the national estimate.
30 The state-level estimates are not reported here, but are available upon request. Also, it is important to note that all
31 emissions from liming are reported in Cropland Remaining Cropland because it is not possible to subdivide the data
32 to each land-use category (i.e., Cropland Remaining Cropland, Land Converted to Cropland, Grassland Remaining
33 Grassland, Land Converted to Grassland, Settlements Remaining Settlements, Forest Land Remaining Forest Land
34 and Land Converted to Forest Land).
35 Table 6-23: Applied Minerals (MMT)
Mineral 1990 2005 2010 2011 2012 2013 2014
Limestone3 19.0 18.1 20.0 15.9 20.8 16.6 17.5
Dolomite8 2.4_^^ 1.9 63 1.4 1.5
a Data represent amounts applied to Cropland Remaining Cropland, Land Converted to Cropland, Grassland
Remaining Grassland, Land Converted to Grassland, Settlements Remaining Settlements, Forest Land
Remaining Forest Land and Land Converted to Forest Land as it is not currently possible to apportion the data
by land-use category.
36 Uncertainty and Time-Series Consistency
37 Uncertainty regarding the amount of limestone and dolomite applied to soils was estimated at ±15 percent with
38 normal densities (Tepordei 2003, Willett 2013b). Analysis of the uncertainty associated with the emission factors
39 included the fraction of lime dissolved by nitric acid versus the fraction that reacts with carbonic acid, and the
40 portion of bicarbonate that leaches through the soil and is transported to the ocean. Uncertainty regarding the time
41 associated with leaching and transport was not addressed in this analysis, but is assumed to be a relatively small
42 contributor to the overall uncertainty (West 2005). The probability distribution functions for the fraction of lime
43 dissolved by nitric acid and the portion of bicarbonate that leaches through the soil were represented as smoothed
Land Use, Land-Use Change, and Forestry 6-43
-------
1 triangular distributions between ranges of zero and 100 percent of the estimates. The uncertainty surrounding these
2 two components largely drives the overall uncertainty. More information on the uncertainty estimates for CC>2
3 Emissions from Liming is contained within the Uncertainty Annex 7.
4 A Monte Carlo (Approach 2) uncertainty analysis was applied to estimate the uncertainty in €62 emissions from
5 liming. The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 6-24. CCh
6 emissions from Liming in 2014 were estimated to be between -0.5 and 7.8 MMT CO2 Eq. at the 95 percent
7 confidence level. This confidence interval represents a range of -111 percent below to 88 percent above the 2014
8 emission estimate of 4.1 MMT CO2 Eq.
9 Table 6-24: Approach 2 Quantitative Uncertainty Estimates for COz Emissions from Liming
10 (MMT COz Eq. and Percent)
s r 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
6 (MMT CCh Eq.) (MMT CCh Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
Limingb CO2 4.1 (0.5) 7.8 -111% +88%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
b Includes emissions from liming on Land Converted to Cropland, Grassland Remaining Grassland, Land Converted to
Grassland, Settlements Remaining Settlements, Forest Land Remaining Forest Land and Land Converted to Forest Land as it
is not possible to subdivide the data by land-use category.
11 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
12 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
13 above.
14 QA/QC and Verification
15 A source-specific QA/QC plan for liming has been developed and implemented, and the quality control effort
16 focused on the Tier 1 procedures for this Inventory. Quality control procedures did uncover a transcription error in
17 the spreadsheets that was corrected.
18 Recalculations Discussion
19 Adjustments were made in the current Inventory to improve the results. First, limestone and dolomite application
20 data for 2013 were approximated in the previous Inventory using a ratio of total crushed stone for 2013 relative to
21 2012 (similar to 2014 in the current Inventory). The estimates for 2013 were updated with the recently published
22 data from USGS (2015). Second, quality control measures uncovered a transcription error in the 2012 activity data
23 that increased the emission estimate by 0.2 MMT €62 Eq. related to the previous Inventory. With these revisions in
24 the activity data, the emissions increased by 3.5 percent in 2012 and decreased by 34 percent in 2013 relative to the
25 previous Inventory.
26
CO2 Emissions from Urea Fertilization
27 The use of urea (CO(NH2)2) as a fertilizer leads to €62 emissions through the release of €62 that was fixed during
28 the industrial production process. In the presence of water and urease enzymes, urea is converted into ammonium
29 (NH4+), hydroxyl ion (OH), and bicarbonate (HCOs")- The bicarbonate then evolves into €62 and water. Emissions
30 from urea fertilization in the United States totaled 4.5 MMT CO2 Eq. (1.2 MMT C) in 2014 (Table 6-25 and Table
31 6-26). Due to an increase in application of urea fertilizers between 1990 and 2014, €62 emissions have increased by
32 87 percent from this management activity.
33 Table 6-25: COz Emissions from Urea Fertilization (MMT COz Eq.)
^Source 1990 2005 2010 2011 2012 2013 2014
Urea Fertilization3 2.4 3.5 3.8 4.1 4.2 4.3 4.5
6-44 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
a Also includes emissions from urea fertilization on Land Converted to Cropland, Grassland
Remaining Grassland, Land Converted to Grassland, Settlements Remaining Settlements, Forest
Land Remaining Forest Land and Land Converted to Forest Land because it is not currently
possible to apportion the data by land-use category.
1 Table 6-26: COz Emissions from Urea Fertilization (MMT C)
Ifource 1990 2005 2010 2011 2012 2013 2014
Urea Fertilization8 0/7 1_0 1.0 1.1 1.2 1.2 1.2
a Also includes emissions from urea fertilization on Land Converted to Cropland, Grassland
Remaining Grassland, Land Converted to Grassland, Settlements Remaining Settlements, Forest
Land Remaining Forest Land and Land Converted to Forest Land because it is not currently possible
to apportion the data by land-use category.
2 Methodology
3 Carbon dioxide emissions from the application of urea to agricultural soils were estimated using the IPCC (2006)
4 Tier 1 methodology. The method assumes that all CC>2 fixed during the industrial production process of urea are
5 released after application. The annual amounts of urea applied to croplands (see Table 6-27) were derived from the
6 state-level fertilizer sales data provided in Commercial Fertilizers (TVA 1991, 1992, 1993, 1994; AAPFCO 1995
7 through 2014). These amounts were multiplied by the default IPCC (2006) emission factor (0.20 metric tons of C
8 per metric ton of urea), which is equal to the C content of urea on an atomic weight basis. Because fertilizer sales
9 data are reported in fertilizer years (July previous year through June current year), a calculation was performed to
10 convert the data to calendar years (January through December). According to monthly fertilizer use data (TVA
11 1992b), 35 percent of total fertilizer used in any fertilizer year is applied between July and December of the previous
12 calendar year, and 65 percent is applied between January and June of the current calendar year. For example, for the
13 2000 fertilizer year, 35 percent of the fertilizer was applied in July through December 1999, and 65 percent was
14 applied in January through June 2000.
15 Fertilizer sales data for the 2013 and 2014 fertilizer years (i.e., July 2012 through June 2013 and July 2013 through
16 June 2014) were not available for this Inventory. Therefore, urea application in the 2013 and 2014 fertilizer years
17 were estimated using a linear, least squares trend of consumption over the data from the previous five years (2008
18 through 2012) at the state level. A trend of five years was chosen as opposed to a longer trend as it best captures the
19 current inter-state and inter-annual variability in consumption. State-level estimates of CC>2 emissions from the
20 application of urea to agricultural soils were summed to estimate total emissions for the entire United States. The
21 fertilizer year data is then converted into calendar year data using the method described above.
22 Emissions are estimated at the state level and summed to obtain the national estimate. The state-level estimates are
23 not reported here, but are available upon request. Also, it is important to note that all emissions from urea
24 fertilization are reported in Cropland Remaining Cropland because it is not currently possible to apportion the
25 emissions to each land-use category (i.e., Cropland Remaining Cropland, Land Converted to Cropland, Grassland
26 Remaining Grassland, Land Converted to Grassland, Settlements Remaining Settlements, Forest Land Remaining
27 Forest Land and Land Converted to Forest Land), however, the majority of urea fertilization is likely to have
28 occurred on Cropland Remaining Cropland.
Land Use, Land-Use Change, and Forestry 6-45
-------
Table 6-27: Applied Urea (MMT)
1990 2005 2010 2011 2012 2013 2014
Urea Fertilizer3 3.3 4.8 5.2 5.6 5.8 5.9 6.2
a These numbers represent amounts applied to all agricultural land, including Cropland
Remaining Cropland, Land Converted to Cropland, Grassland Remaining Grassland, Land
Converted to Grassland, Settlements Remaining Settlements, Forest Land Remaining Forest
Land and Land Converted to Forest Land because it is not currently possible to apportion the
data by land-use category.
2 Uncertainty and Time-Series Consistency
3 Uncertainty estimates are presented in Table 6-28 for Urea Fertilization. An Approach 2 Monte Carlo analysis was
4 completed. The largest source of uncertainty was the default emission factor, which assumes that 100 percent of the
5 C in CO(NH2)2 applied to soils is ultimately emitted into the environment as €62. This factor does not incorporate
6 the possibility that some of the C may be retained in the soil, and therefore the uncertainty range was set from 0
7 percent emissions to the maximum emission value of 100 percent using a triangular distribution. In addition, each
8 urea consumption data point has an associated uncertainty. €62 emissions from urea fertilization of agricultural
9 soils in 2014 were estimated to be between 2.6 and 4.5 MMT €62 Eq. at the 95 percent confidence level. This
10 indicates a range of 42 percent below to 0 percent above the 2014 emission estimate of 4.5 MMT €62 Eq.
11 Table 6-28: Quantitative Uncertainty Estimates for COz Emissions from Urea Fertilization
12 (MMT COz Eq. and Percent)
13
Source Gas 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate3
(MMT CCh Eq.) (MMT CCh Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
Urea Fertilization CCh 45 2.6 45 -42% 0%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
14 There are additional uncertainties that are not quantified in this analysis. Urea for non-fertilizer use, such as aircraft
15 deicing, may be included in consumption totals, but the amount is likely very small. For example, research on
16 aircraft deicing practices is consistent with this assumption based on a 1992 survey that found a known annual usage
17 of approximately 2,000 tons of urea for deicing; this would constitute 0.06 percent of the 1992 consumption of urea
18 (EPA 2000). Similarly, surveys conducted from 2002 to 2005 indicate that total urea use for deicing at U.S. airports
19 is estimated to be 3,740 metric tons per year, or less than 0.07 percent of the fertilizer total for 2007 (Itle 2009). In
20 addition, there is uncertainty surrounding the underlying assumptions behind the calculation that converts fertilizer
21 years to calendar years. These uncertainties are negligible over multiple years, however, because an over- or under-
22 estimated value in one calendar year is addressed with corresponding increase or decrease in the value for the
23 subsequent year.
24 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
25 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
26 above.
27 QA/QC and Verification
28 A source-specific QA/QC plan for Urea has been developed and implemented. For this year, the Tier 1 analysis was
29 performed and an error was found in a formula reference to an incorrect cell in the spreadsheets.
30 Recalculations Discussion
31 In the current Inventory, the 2013 emission estimate was updated to reflect a correction to the calculations made in
32 the previous Inventory report. Quality control checks uncovered an incorrect spreadsheet cell reference influencing
6-46 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 the state-level emission calculations. The 2013 emission estimate increased by 8.3 percent, relative to the previous
2 report, due to this correction.
3 Planned Improvements
4 No improvements are planned for this source.
5 6.5 Land Converted to Cropland (IPCC Source
6 Category 4B2)
7 Land Converted to Cropland includes all cropland in an Inventory year that had been in another land use(s) during
8 the previous 20 years40 (USDA-NRCS 2013). For example, grassland or forestland converted to cropland during the
9 past 20 years would be reported in this category. Recently-converted lands are retained in this category for 20 years
10 as recommended in the IPCC guidelines (IPCC 2006). This Inventory includes all croplands in the conterminous
11 United States and Hawaii, but does not include a minor amount of Land Converted to Cropland in Alaska. Some
12 miscellaneous croplands are also not included in the inventory due to limited understanding of greenhouse gas
13 dynamics in management systems (e.g., aquaculture) or climate zones (e.g., boreal climates). Consequently there is
14 a discrepancy between the total amount of managed area in Land Converted to Cropland (see Section 6.1-
15 Representation of the U.S. Land Base) and the cropland area included in the Inventory. Improvements are underway
16 to include croplands in Alaska and miscellaneous crops in future C inventories.
17 Background on agricultural carbon (C) stock changes is provided in Section 0 Cropland Remaining Cropland and
18 therefore will only be briefly summarized here. Soils are the largest pool of C in agricultural land, and also have the
19 greatest potential for long-term storage or release of C, because bio mass and dead organic matter C pools are
20 relatively small and ephemeral compared with soils, with the exception of C stored in perennial woody crop
21 biomass. The 2006 IPCC Guidelines recommend reporting changes in soil organic carbon (SOC) stocks due to (1)
22 agricultural land-use and management activities on mineral soils, and (2) agricultural land-use and management
23 activities on organic soils.41
24 Land use and management of mineral soils in Land Converted to Cropland is the largest contributor to C loss
25 throughout the time series, accounting for approximately 71 percent of the emissions in the category (Table 6-29 and
26 Table 6-30). The conversion of grassland to cropland is the largest source of soil C loss (accounting for
27 approximately 84 percent of the average emissions in the category), though losses declined over the time series. The
28 net change in soil C stocks for 2014 is 14.7 MMT CO2 Eq. (4.0 MMT C), including 10.4 MMT CO2 Eq. (2.8 MMT
29 C) from mineral soils and 4.3 MMT CC>2 Eq. (1.2 MMT C) from drainage and cultivation of organic soils.
30 Table 6-29: Net COz Flux from Soil C Stock Changes in Land Converted to Cropland^ Land
31 Use Change Category (MMT COz Eq.)
Soil Type 1990 2005 2010 2011 2012 2013 2014
Grassland Converted to Cropland
Mineral 17.3 7.6 8.8 7.3 7.6 7.8 7.8
Organic 3.2 4.2 3.8 3.8 3.8 3.8 3.8
Forest Converted to Cropland
Mineral (+) (0.1) (0.1) (0.1) (0.1) (0.1) (0.1)
Organic (+) + I + + + + +
Other Lands Converted Cropland
Mineral 1.3 1.2 1.3 1.3 1.3 1.3 1.3
40 NRI survey locations are classified according to land-use histories starting in 1979, and consequently the classifications are
based on less than 20 years from 1990 to 2001, which may have led to an underestimation of Land Converted to Cropland in the
early part of the time series to the extent that some areas are converted to cropland prior to 1979.
41 CO2 emissions associated with liming urea fertilization are also estimated but included in 6.4 Cropland Remaining Cropland.
Land Use, Land-Use Change, and Forestry 6-47
-------
0.0 0.0 0.0 0.0 0.0
1.2 1.3 1.3 1.3
0.1 0.1 0.1 0.1
0.1 0.1 0.1 0.1
0.4 0.4 0.4 0.4
9.8 10.2 10.4 10.4
4.3 4.3 4.3 4.3
15.6 14.2 14.5 14.8 14.7
Notes: Estimates after 2010 are based on NRI data from 2010 and therefore may not fully reflect changes occurring
in the latter part of the time series.
Parentheses indicate net sequestration.
+ Does not exceed 0.05 MMT CO2 Eq.
Table 6-30: Net COz Flux from Soil C Stock Changes in Land Converted to Cropland (MWf C)
Organic
Settlements Converted Cropland
Mineral
Organic
Wetlands Converted Cropland
Mineral
Organic
Total Mineral Soil Flux
Total Organic Soil Flux
Total Net Flux
(+) 1
1.3 1
0.0 1
(0.2)
19.8
3.0
22.8
0.0
1.1
0.1
0.1
0.5
9.9
4.8
14.6
Soil Type 1990 2005 2010 2011 2012 2013 2014
Grassland Converted to Cropland
Mineral 4.7 2.1 2.4 2.0 2.1 2.1 2.1
Organic 0.9 1.1 1.0 1.0 1.0 1.0 1.0
Forest Converted to Cropland
Mineral (+) (+) (+) (+) (+) (+) (+)
Organic (+) + I + + + + +
Other Lands Converted Cropland
Mineral 0.4 0.3 0.4 0.4 0.4 0.4 0.4
Organic (+) 0.0 0.0 0.0 0.0 0.0 0.0
Settlements Converted Cropland
Mineral 0.4 0.3 0.3 0.3 0.3 0.3 0.3
Organic (+) + I + + + + +
Wetlands Converted Cropland
Mineral 0.0 + I + + + + +
Organic (+) 0.1 0.1 0.1 0.1 0.1 0.1
Total Mineral Soil Flux 5.4 2.7 3.1 2.7 2.8 2.8 2.8
Total Organic Soil Flux 0.8 1.3 1.2 1.2 1.2 1.2 1.2
Total Net Flux 6.2 40 4.3 3.9 4.0 4.0 4.0
Notes: Estimates after 2010 are based on NRI data from 2010 and therefore may not fully reflect changes
occurring in the latter part of the time series.
Parentheses indicate net sequestration.
+ Does not exceed 0.05 MMT C
3 The spatial variability in the 2014 annual C stock changes for mineral soils is displayed in Figure 6-7 and from
4 organic soils in Figure 6-8. Losses occurred in most regions of the United States. In particular, conversion of
5 grassland and forestland to cropland led to enhanced decomposition of soil organic matter and a net loss of C from
6 the soil pool. The regions with the highest rates of emissions from organic soils coincide with the largest
7 concentrations of organic soils used for agricultural production, including Southeastern Coastal Region (particularly
8 Florida), upper Midwest and Northeast surrounding the Great Lakes, and the Pacific Coast (particularly California).
9
10 Figure 6-7: Total Net Annual COz Flux for Mineral Soils under Agricultural Management
11 within States, 2014, Land Converted to Cropland'(TO BE UPDATED)
12
13 Figure 6-8: Total Net Annual COz Flux for Organic Soils under Agricultural Management
14 within States, 2014, Land Converted to Cropland'(TO BE UPDATED)
15
6-48 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Methodology
2 The following section includes a description of the methodology used to estimate changes in soil C stocks for Land
3 Converted to Cropland, including (1) agricultural land-use and management activities on mineral soils; and (2)
4 agricultural land-use and management activities on organic soils Biomass and litter C stock changes associated with
5 conversion of forest to cropland are not explicitly included in this category, but are included in the Forest Land
6 Remaining Forest Land section. Further elaboration on the methodologies and data used to estimate stock changes
7 for mineral and organic soils are provided in the Cropland Remaining Cropland section and Annex 3.12.
8 Soil C stock changes are estimated for Land Converted to Cropland according to land-use histories recorded in the
9 2010 USDA NRI survey (USDA-NRCS 2013). Land-use and some management information (e.g., crop type, soil
10 attributes, and irrigation) had been collected for each NRI point on a 5-year cycle beginning in 1982. In 1998, the
11 NRI program began collecting annual data, and data are currently available through 2012 (USDA-NRCS 2015).
12 However, this Inventory only uses NRI data through 2010 because newer data were not available in time to
13 incorporate the additional years. NRI survey locations are classified as Land Converted to Cropland in a given year
14 between 1990 and 2010 if the land use is cropland but had been another use during the previous 20 years. Cropland
15 includes all land used to produce food or fiber, or forage that is harvested and used as feed (e.g., hay and silage).
16 Mineral Soil Carbon Stock Changes
17 An IPCC Tier 3 model-based approach (Ogle et al. 2010) is applied to estimate C stock changes for mineral soils on
18 the majority of land that is used to produce annual crops in the United States. These crops include alfalfa hay,
19 barley, corn, cotton, dry beans, grass hay, grass-clover hay, oats, onions, peanuts, potatoes, rice, sorghum, soybeans,
20 sugar beets, sunflowers, tomatoes, and wheat. Soil C stock changes on the remaining soils are estimated with the
21 IPCC Tier 2 method (Ogle et al. 2003), including land used to produce some vegetables, tobacco,
22 perennial/horticultural crops and crops rotated with these crops; land on very gravelly, cobbly, or shaley soils
23 (greater than 35 percent by volume); and land converted from another land use or federal ownership.42
24 Tier 3 Approach
25 For the Tier 3 method, mineral SOC stocks and stock changes are estimated using the DAYCENT biogeochemical43
26 model (Parton et al. 1998; Del Grosso et al. 2001, 2011). The DAYCENT model utilizes the soil C modeling
27 framework developed in the Century model (Parton et al. 1987, 1988, 1994; Metherell et al. 1993), but has been
28 refined to simulate dynamics at a daily time-step. National estimates are obtained by using the model to simulate
29 historical land-use change patterns as recorded in the USDA NRI (USDA-NRCS 2013). C stocks and 95 percent
30 confidence intervals are estimated for each year between 1990 and 2010, but C stock changes from 2010 to 2014 are
31 assumed to be similar to 2010. (Future inventories will be updated with new activity data and the time series will be
32 recalculated; See Planned Improvements section in Cropland Remaining Cropland). The methods used for Land
33 Converted to Cropland are the same as those described in the Tier 3 portion of Cropland Remaining Cropland
34 section for mineral soils.
35 Tier 2 Approach
36 For the mineral soils not included in the Tier 3 analysis, SOC stock changes are estimated using a Tier 2 Approach
37 for Land Converted to Cropland as described in the Tier 2 Approach for mineral soils in the Grassland Remaining
38 Grassland section.
42Federal land is not a land use, but rather an ownership designation that is treated as grassland for purposes of these
calculations. The specific land use on federal lands is not identified in the NRI survey (USDA-NRCS 2013).
Biogeochemical cycles are the flow of chemical elements and compounds between living organisms and the physical
environment.
Land Use, Land-Use Change, and Forestry 6-49
-------
i Organic Soil Carbon Stock Changes
2 Annual C emissions from drained organic soils in Land Converted to Cropland are estimated using the Tier 2
3 method provided in IPCC (2006), with U. S. -specific C loss rates (Ogle et al. 2003) as described in the Cropland
4 Remaining Cropland section for organic soils.
5 Uncertainty and Time-Series Consistency
Uncertainty analysis for mineral soil C stock changes using the Tier 3 and Tier 2 methodologies are based on a
Monte Carlo approach that is described for Cropland Remaining Cropland. The uncertainty for annual C emission
estimates from drained organic soils in Land Converted to Cropland is estimated using a Monte Carlo approach,
which is also described in the Cropland Remaining Cropland section.
Uncertainty estimates are presented in Table 6-31 for each subsource (i.e., mineral soil C stocks and organic soil C
stocks) and the method applied in the inventory analysis (i.e., Tier 2 and Tier 3). Uncertainty estimates from the
Tier 2 and 3 approaches are combined using the simple error propagation methods provided by the IPCC (2006),
i.e., by taking the square root of the sum of the squares of the standard deviations of the uncertain quantities. The
combined uncertainty for soil C stocks in Land Converted to Cropland ranged from 94 percent below to 95 percent
above the 2014 stock change estimate of 14.7 MMT CC>2 Eq. The uncertainties are large for some of the estimates
due to small change in soil C stocks, even though the absolute amount of uncertainty is relatively small.
Table 6-31: Approach 2 Quantitative Uncertainty Estimates for Soil C Stock Changes
occurring within Land Converted to Crop/and (MMT COz Eq. and Percent)
6
7
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Source
Grassland Converted to Cropland
Mineral Soil C Stocks: Tier 3
Mineral Soil C Stocks: Tier 2
Organic Soil C Stocks: Tier 2
Forests Converted to Cropland
Mineral Soil C Stocks: Tier 2
Organic Soil C Stocks: Tier 2
Other Lands Converted to Cropland
Mineral Soil C Stocks: Tier 2
Organic Soil C Stocks: Tier 2
Settlements Converted to Cropland
Mineral Soil C Stocks: Tier 2
Organic Soil C Stocks: Tier 2
Wetlands Converted to Croplands
Mineral Soil C Stocks: Tier 2
Organic Soil C Stocks: Tier 2
Total: Land Converted to Cropland
Mineral Soil C Stocks: Tier 3
Mineral Soil C Stocks: Tier 2
Organic Soil C Stocks: Tier 2
2014 Flux Estimate
(MMT CO2 Eq.)
11.6
10.9
(3.1)
3.8
(+)
(0.1)
+
1.3
1.3
0.0
1.3
1.3
0.1
0.5
0.1
0.4
14.7
10.9
(0.5)
4.3
Uncertainty Range Relative to Flux Estimate3
(MMT
Lower
Bound
(2.1)
(0.8)
(0.1)
10.2
(0.1)
(+)
0.1
(+)
(+)
0.0
0.1
(+)
0.1
0.2
+
0.7
0.9
(0.8)
(4.1)
(2.1)
CO2 Eq.)
Upper
Bound
25.0
22.6
2.2
+
0.1
0.1
0.0
2.3
0.3
0.1
2.6
0.1
0.4
4.4
0.2
4.3
28.8
22.6
5.0
9.8
('
Lower
Bound
-118%
-107%
-97%
-168%
-172%
-92%
-154%
-102%
-102%
0%
-95%
-101%
-91%
-58%
-43%
-65%
-94%
-107%
-677%
-148%
'/»)
Upper
Bound
+115%
+107%
+172%
+99%
+372%
+232%
+100%
+76%
+76%
0%
+90%
+92%
+454%
+801%
+246%
+908%
+95%
+107%
+1,065%
+126%
Note: Parentheses indicate negative values or net sequestration.
a Range of C stock change estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
+ Does not exceed 0.05 MMT CO2 Eq.
Uncertainty is also associated with lack of reporting of agricultural biomass and litter C stock changes other than the
loss of forest biomass and litter, which is reported in the Forest Land Remaining Forest Land section of this report.
Biomass C stock changes are likely minor in perennial crops, such as orchards and nut plantations, given the small
amount of change in land used to produce these commodities in the United States. In contrast, agroforestry
practices, such as shelterbelts, riparian forests and intercropping with trees, may have led to significant changes in
biomass C stocks, at least in some regions of the United States, but there are currently no datasets to evaluate the
trends. Changes in litter C stocks are also assumed to be negligible in croplands over annual time frames, although
6-50 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 there are certainly significant changes at sub-annual time scales across seasons. However, this trend may change in
2 the future, particularly if crop residue becomes a viable feedstock for bioenergy production.
3 Methodological recalculations are applied to the entire time series to ensure time-series consistency from 1990
4 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
5 above.
6 QA/QC and Verification
7 See the QA/QC and Verification section in Cropland Remaining Cropland.
8 Recalculations Discussion
9 Methodological recalculations in the current Inventory are associated with the following improvements: 1)
10 incorporation of updated NRI data for 1990 through 2010; 2) inclusion of federal croplands; and 3) improving the
11 simulation of hydric soils in DAYCENT. As a result of these improvements to the Inventory, SOC stock changes
12 declined on average, leading to less carbon loss from Land Converted to Cropland by an average of 4.4 MMT CC>2
13 Eq. over the time series.
14 Planned Improvements
15 Soil C stock changes with land use conversion from forest land to cropland are undergoing further evaluation to
16 ensure consistency in the time series. Different methods are used to estimate soil C stock changes in forest land and
17 croplands, and while the areas have been reconciled between these land uses, there has been limited evaluation of
18 the consistency in C stock changes with conversion from forest land to cropland. This planned improvement may
19 not be fully implemented for another year, depending on resource availability. Additional planned improvements
20 are discussed in the Cropland Remaining Cropland section.
21
22
6.6 Grassland Remaining Grassland (IPCC
Source Category 4C1)
23 Grassland Remaining Grassland includes all grassland in an Inventory year that had been classified as grassland for
24 the previous 20 years44 (USDA-NRCS 2013). Grassland includes pasture and rangeland that are primarily but not
25 exclusively used for livestock grazing. Rangelands are typically extensive areas of native grassland that are not
26 intensively managed, while pastures are typically seeded grassland (possibly following tree removal) that may also
27 have additional management, such as irrigation or interseeding of legumes. This Inventory includes all privately -
28 owned and federal grasslands in the conterminous United States and Hawaii, but does not include approximately 50
29 million hectares of Grassland Remaining Grassland in Alaska. This leads to a discrepancy with the total amount of
30 managed area in Grassland Remaining Grassland (see Section 6.1 -Representation of the U.S. LandBase) and the
31 grassland area included in the Inventory analysis (IPCC Source Category 4C1—Section 6.6).
32 Background on agricultural carbon (C) stock changes is provided in Section 0, Cropland Remaining Cropland, and
33 will only be summarized here. Soils are the largest pool of C in agricultural land, and also have the greatest
34 potential for longer-term storage or release of C. Biomass and dead organic matter C pools are relatively small and
35 ephemeral compared to the soil C pool, with the exception of C stored in tree and shrub biomass that occurs in
44NRI survey locations are classified according to land-use histories starting in 1979, and consequently the classifications are
based on less than 20 years from 1990 to 2001. This may have led to an overestimation of Grassland Remaining Grassland in
the early part of the time series to the extent that some areas are converted to grassland prior to 1979.
Land Use, Land-Use Change, and Forestry 6-51
-------
1 grasslands. The 2006IPCC Guidelines (IPCC 2006) recommend reporting changes in soil organic C (SOC) stocks
2 due to (1) agricultural land-use and management activities on mineral soils, and (2) agricultural land-use and
3 management activities on organic soils.45
4 In Grassland Remaining Grassland, there has been considerable variation in soil C stocks between 1990 and 2014.
5 These changes are driven by variability in weather patterns and associated interaction with land management
6 activity. Moreover, changes remain small on a per hectare rate across the time series even in the years with a larger
7 total change in stocks. Land use and management generally increased soil C in mineral soils for Grassland
8 Remaining Grassland between 1990 and 2010, after which the trend is reversed to a small decline in soil C. In
9 contrast, organic soils lost relatively small amounts of C annually from 1990 through 2014. Overall, the average
10 change in soil C stocks is an increase by 24.8 MMT CO2 Eq., from 1990 to 2014 (6.8 MMT C) (Table 6-32 and
11 Table 6-33). Soil C stocks decreased in 2014 by 11.9 MMT CO2 Eq. (3.3 MMT C), with 7.6 MMT CO2 Eq. (2.1
12 MMT C) from mineral soils and 4.3 MMT CO2 Eq. (1.2 MMT C) from organic soils.
13 Table 6-32: Net COi Flux from Soil C Stock Changes in Grass/and Remaining Grass/and (MMJ
14 COz Eq.)
Soil Type
Mineral Soils
Organic Soils
Total Net Flux
1990
(19.0)
6.lH
(12.9)
2005
(1.6)
4.5
2.9
2010
(1.8)
4.4
2.6
2011
6.9
4.4
11.3
2012
7.4
4.3
11.7
2013
7.6
4.3
11.9
2014
7.6
4.3
11.9
Note: Totals may not sum due to independent rounding. Estimates after 2010 are based on NRI data
from 2010 and therefore may not fully reflect changes occurring in the latter part of the time series.
Parentheses indicate net sequestration.
15 Table 6-33: Net COz Flux from Soil C Stock Changes in Grassland Remaining Grass/and (MMJ
16 C)
Soil Type
Mineral Soils
Organic Soils
Total Net Flux
1990
(5.2)
1.7
(3.5)
2005
(0.4)
1.2
0.8
2010
(0.5)
1.2
0.7
2011
1.9
1.2
3.1
2012
2.0
1.2
3.2
2013
2.1
1.2
3.3
2014
2.1
1.2
3.3
Notes: Totals may not sum due to independent rounding. Estimates after 2010 are based on NRI data
from 2010 and therefore may not fully reflect changes occurring in the latter part of the time series.
Parentheses indicate net sequestration.
17 The spatial variability in the 2014 annual flux in CO2 from mineral soils is displayed in Figure 6-9 and organic soils
18 in Figure 6-10. Although relatively small on a per-hectare basis, grassland soils gained C in several regions during
19 2014, including the Northeast, Southeast, portions of the Midwest, and Pacific Coastal Region. The regions with the
20 highest rates of emissions from organic soils coincide with the largest concentrations of organic soils used for
21 managed grassland, including the Southeastern Coastal Region (particularly Florida), upper Midwest and Northeast
22 surrounding the Great Lakes, and the Pacific Coast (particularly California).
23 Figure 6-9: Total Net Annual COz Flux for Mineral Soils under Agricultural Management
24 within States, 2014, Grassland Remaining Grassland'(TO BE UPDATED)
25
26
27 Figure 6-10: Total Net Annual COz Flux for Organic Soils under Agricultural Management
28 within States, 2014, Grassland Remaining Grassland'(TO BE UPDATED)
29
CO2 emissions associated with liming and urea fertilization are also estimated but included in 6.4 Cropland Remaining
Cropland.
6-52 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Methodology
2 The following section includes a brief description of the methodology used to estimate changes in soil C stocks for
3 Grassland Remaining Grassland, including (1) agricultural land-use and management activities on mineral soils;
4 and (2) agricultural land-use and management activities on organic soils. Further elaboration on the methodologies
5 and data used to estimate stock changes from mineral and organic soils are provided in the Cropland Remaining
6 Cropland section and Annex 3.12.
7 Soil C stock changes are estimated for Grassland Remaining Grassland according to land use histories recorded in
8 the 2010 USDA NRI survey (USDA-NRCS 2013). Land-use and some management information (e.g., crop type,
9 soil attributes, and irrigation) were originally collected for each NRI survey location on a 5-year cycle beginning in
10 1982. In 1998, the NRI program began collecting annual data, and the annual data are currently available through
11 2012 (USDA-NRCS 2015). However, this Inventory only uses NRI data through 2010 because newer data were not
12 available in time to incorporate the additional years. NRI survey locations are classified as Grassland Remaining
13 Grassland in a given year between 1990 and 2010 if the land use had been grassland for 20 years.
14 Mineral Soil Carbon Stock Changes
15 An IPCC Tier 3 model-based approach (Ogle et al. 2010) is applied to estimate C stock changes for most mineral
16 soils in Grassland Remaining Grassland. The C stock changes for the remaining soils are estimated with an IPCC
17 Tier 2 method (Ogle et al. 2003), including gravelly, cobbly, or shaley soils (greater than 35 percent by volume) and
18 additional stock changes associated with sewage sludge amendments.
19 Tier 3 Approach
20 Mineral SOC stocks and stock changes for Grassland Remaining Grassland are estimated using the DAYCENT
21 biogeochemical46 model (Parton et al. 1998; Del Grosso et al. 2001, 2011), as described in Cropland Remaining
22 Cropland. The DAYCENT model utilizes the soil C modeling framework developed in the Century model (Parton
23 et al. 1987, 1988, 1994; Metherell et al. 1993), but has been refined to simulate dynamics at a daily time-step.
24 Historical land-use patterns and irrigation histories are simulated with DAYCENT based on the 2010 USDA NRI
25 survey (USDA-NRCS 2013), with supplemental information on fertilizer use and rates from the USDA Economic
26 Research Service Cropping Practices Survey (USDA-ERS 1997, 2011) and National Agricultural Statistics Service
27 (NASS 1992, 1999, 2004). Frequency and rates of manure application to grassland during 1997 are estimated from
28 data compiled by the USDA Natural Resources Conservation Service (Edmonds, et al. 2003), and then adjusted
29 using county-level estimates of manure available for application in other years. Specifically, county-scale ratios of
30 manure available for application to soils in other years relative to 1997 are used to adjust the area amended with
31 manure (see Cropland Remaining Cropland for further details). Greater availability of managed manure nitrogen
32 (N) relative to 1997 is, thus, assumed to increase the area amended with manure, while reduced availability of
33 manure N relative to 1997 is assumed to reduce the amended area.
34 The amount of manure produced by each livestock type is calculated for managed and unmanaged waste
35 management systems based on methods described in Section 5.2 - Manure Management and Annex 3.11. Manure N
36 deposition from grazing animals (i.e., PRP manure) is an input to the DAYCENT model (see Annex 3.11), and
37 included approximately percent of total PRP manure (the remainder is deposited on federal lands, which are not
38 included in this Inventory). C stocks and 95 percent confidence intervals are estimated for each year between 1990
39 and 2010, but C stock changes from 2011 to 2014 are assumed to be similar to 2010 (Future inventories will be
40 updated with new activity data and the time series will be recalculated; See Planned Improvements section in
41 Cropland Remaining Cropland). The methods used for Grassland Remaining Grassland are the same as those
42 described in the Tier 3 portion of Cropland Remaining Cropland section for mineral soils.
Biogeochemical cycles are the flow of chemical elements and compounds between living organisms and the physical
environment.
Land Use, Land-Use Change, and Forestry 6-53
-------
1 Tier 2 Approach
1 The Tier 2 approach is based on the same methods described in the Tier 2 portion of Cropland Remaining Cropland
3 section for mineral soils, with the exception of the land use and management data that are used in the Inventory for
4 federal grasslands. The NRI (USDA-NRCS 2013) provides land use and management histories for all non-federal
5 lands, and is the basis for the Tier 2 analysis for these areas. However, NRI does not provide land use information
6 on federal lands. These data are based on the National Land Cover Database (NLCD) (Fry et al. 2011; Homer et al.
7 2007; Homer et al. 2015). In addition, the Bureau of Land Management (BLM) manages some of the federal grassland,
8 and has compiled information on grassland condition through the BLM Rangeland Inventory (BLM 2014). Rangeland
9 conditions in the BLM data are aligned with IPCC grassland categories of nominal, moderately degraded, and severely
10 degraded. Further elaboration on the Tier 2 methodology and data used to estimate stock changes from mineral soils
11 are described in Annex 3.12.
12 Additional Mineral C Stock Change Calculations
13 A Tier 2 method is used to adjust annual C stock change estimates for mineral soils between 1990 and 2014 to
14 account for additional C stock changes associated with sewage sludge amendments. Estimates of the amounts of
15 sewage sludge N applied to agricultural land are derived from national data on sewage sludge generation,
16 disposition, and N content. Total sewage sludge generation data for 1988, 1996, and 1998, in dry mass units, are
17 obtained from EPA (1999) and estimates for 2004 are obtained from an independent national biosolids survey
18 (NEBRA 2007). These values are linearly interpolated to estimate values for the intervening years, and linearly
19 extrapolated to estimate values for years since 2004. N application rates from Kellogg et al. (2000) are used to
20 determine the amount of area receiving sludge amendments. Although sewage sludge can be added to land managed
21 for other land uses, it is assumed that agricultural amendments occur in grassland. Cropland is not likely to be
22 amended with sewage sludge due to the high metal content and other pollutants in human waste. The soil C storage
23 rate is estimated at 0.38 metric tons C per hectare per year for sewage sludge amendments to grassland. The stock
24 change rate is based on country-specific factors and the IPCC default method (see Annex 3.12 for further
25 discussion).
26 Organic Soil Carbon Stock Changes
27 Annual C emissions from drained organic soils in Grassland Remaining Grassland are estimated using the Tier 2
28 method provided in IPCC (2006), which utilizes U.S.-specific C loss rates (Ogle et al. 2003) rather than default
29 IPCC rates. For more information, see the Cropland Remaining Cropland section for organic soils.
30 Uncertainty and Time-Series Consistency
31 Uncertainty analysis for mineral soil C stock changes using the Tier 3 and Tier 2 methodologies are based on a
32 Monte Carlo approach that is described in the Cropland Remaining Cropland section. The uncertainty for annual C
33 emission estimates from drained organic soils in Grassland Remaining Grassland is estimated using a Monte Carlo
34 approach, which is also described in the Cropland Remaining Cropland section.
35 Uncertainty estimates are presented in Table 6-34 for each subsource (i.e., mineral soil C stocks and organic soil C
36 stocks) and the method applied in the inventory analysis (i.e., Tier 2 and Tier 3). Uncertainty estimates from the
37 Tier 2 and 3 approaches are combined using the simple error propagation methods provided by the IPCC (2006),
38 i.e., by taking the square root of the sum of the squares of the standard deviations of the uncertain quantities. The
39 combined uncertainty for soil C stocks in Grassland Remaining Grassland ranges from 344 percent below to 346
40 percent above the 2014 stock change estimate of 11.9 MMT CC>2 Eq. The large relative uncertainty is due to the
41 small net C stock change estimate in 2014, particularly on federal grasslands, even though the absolute amount of
42 uncertainty is relatively small.
6-54 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 6-34: Approach 2 Quantitative Uncertainty Estimates for C Stock Changes Occurring
2 Within Grassland Remaining Grass/and (MWT COz Eq. and Percent)
2014 Flux Estimate Uncertainty Range Relative to Flux Estimate3
Source (MMT CCh Eq.) (MMT CCh Eq.) (%)
Mineral Soil C Stocks Grassland Remaining
Grassland, Tier 3 Methodology
Mineral Soil C Stocks: Grassland Remaining
Grassland, Tier 2 Methodology
Mineral Soil C Stocks: Grassland Remaining
Grassland, Tier 2 Methodology (Change in
Soil C due to Sewage Sludge Amendments)
Organic Soil C Stocks: Grassland Remaining
Grassland, Tier 2 Methodology
Combined Uncertainty for Flux Associated
with Agricultural Soil Carbon Stock
Change in Grassland Remaining Grassland
Note: Parentheses indicate negative values.
a Range of C stock change estimates predicted by
9.2
(0.3)
(1.4)
4 3
11.9
Monte Carlo Stochastic
Lower
Bound
(30.8)
(8.8)
(2.1)
2 2
(29.0)
Simulation
Upper
Bound
49.2
9.3
(0.7)
7 2
53.2
Lower
Bound
-433%
-3,308%
-50%
-49%
-344%
Upper
Bound
+433%
+3,680%
+50%
+66%
+346%
for a 95 percent confidence interval.
4 Uncertainty is also associated with a lack of reporting on agricultural biomass and litter C stock changes and non-
5 CO2 greenhouse gas emissions from burning. Biomass C stock changes may be significant for managed grasslands
6 with woody encroachment that has not attained enough tree cover to be considered forest lands. This Inventory does
7 not currently include the non-CCh greenhouse gas emissions that occur with biomass burning. Grassland burning is
8 not as common in the United States as in other regions of the world, but fires do occur through both natural ignition
9 sources and prescribed burning. Changes in litter C stocks are assumed to be negligible in grasslands over annual
10 time frames, although there are certainly significant changes at sub-annual time scales across seasons.
11 Methodological recalculations are applied to the entire time series to ensure time-series consistency from 1990
12 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
13 above.
14 QA/QC and Verification
15 See the QA/QC and Verification section in Cropland Remaining Cropland.
16 Recalculations Discussion
17 Methodological recalculations in the current Inventory are associated with the following improvements, including 1)
18 incorporation of updated NRI data for 1990 through 2010; 2) inclusion of federal grasslands in the Tier 2 analysis;
19 and 3) improving the simulation of hydric soils in DAY CENT. As a result of these improvements to the Inventory,
20 changes in SOC stocks increased by an average of 4.3 MMT CC>2 eq. annually over the time series.
21 Planned Improvements
22 Grasslands in Alaska are not currently included in the Inventory. This is a significant planned improvement and
23 estimates are expected to be available for the 1990 through 2015 Inventory. Another key planned improvement is to
24 estimate non-CCh greenhouse gas emissions from burning of grasslands. For information about other
25 improvements, see the Planned Improvements section in Cropland Remaining Cropland.
Land Use, Land-Use Change, and Forestry 6-55
-------
i 6.7 Land Converted to Grassland (IPCC Source
2 Category 4C2)
3 Land Converted to Grassland includes all grassland in an Inventory year that had been in another land use(s) during
4 the previous 20 years47 (USDA-NRCS 2013). For example, cropland or forestland converted to grassland during
5 the past 20 years would be reported in this category. Recently-converted lands are retained in this category for 20
6 years as recommended by IPCC (2006). Grassland includes pasture and rangeland that are used primarily but not
7 exclusively for livestock grazing. Rangelands are typically extensive areas of native grassland that are not
8 intensively managed, while pastures are typically seeded grassland (possibly following tree removal) that may also
9 have additional management, such as irrigation or interseeding of legumes. This Inventory includes all grasslands in
10 the conterminous United States and Hawaii, but does not include Land Converted to Grassland in Alaska.
11 Consequently there is a discrepancy between the total amount of managed area for Land Converted to Grassland
12 (see Section 6.1—Representation of the U.S. Land Base) and the grassland area included in in the inventory analysis
13 (IPCC Source Category 4C2—Section 6.7).
14 Background on agricultural carbon (C) stock changes is provided in Cropland Remaining Cropland and therefore
15 will only be briefly summarized here. Soils are the largest pool of C in agricultural land, and also have the greatest
16 potential for long-term storage or release of C. Biomass and dead organic matter C pools are relatively small and
17 ephemeral compared with soils, with the exception of C stored in tree and shrub biomass that occurs in grasslands.
18 IPCC (2006) recommend reporting changes in soil organic C (SOC) stocks due to (1) agricultural land-use and
19 management activities on mineral soils, and (2) agricultural land-use and management activities on organic soils.48
20 Land use and management of mineral soils in Land Converted to Grassland led to an increase in soil C stocks
21 between 1990 and 2014 (see Table 6-35 and Table 6-36). The average soil C stock change for mineral soils between
22 1990 and 2014 is 12.4 MMT CC>2 Eq. (3.4 MMT C). In contrast, over the same period, drainage of organic soils for
23 grassland management led to an increase in C emissions to the atmosphere of 1.5 MMT CO2 Eq. (0.4 MMT C). The
24 total soil C stock change in 2014 for Land Converted to Grassland is estimated at 10.9 MMT CO2 Eq. (3.0 MMT
25 C).
26 Table 6-35: Net COz Flux from Soil C Stock Changes for Land Converted to Grass/and (MMJ
27 COz Eq.)
Soil Type
Cropland Converted to Grassland
Mineral
Organic
Forest Converted to Grassland
Mineral
Organic
Other Lands Converted Grassland
Mineral
Organic
Settlements Converted Grassland
Mineral
Organic
Wetlands Converted Grassland
Mineral
Organic
Total Mineral Soil Flux
1990
(7.
0
(0.
5)
.5
5)
(0.6)
2005
(11.
1
(1.
(1.
3)
.0
0)
1)
=
2010
(11.8)
1.1
(0.8)
+
(0.8)
2011
(10.4)
1.2
(0.8)
+
(0.8)
2012
(10.4)
1.2
(0.8)
+
(0.8)
2013
(10.4)
1.2
(0.8)
+
(0.8)
2014
(10.4)
1.2
(0.8)
+
(0.8)
+ 0.0 + + + + +
(0.
1)
(0.
2)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
+ | 0.0 + + + + +
(0.
0
5)
.1
(0.
6)
H 0.2
(9.2)
(14.1)
(0.2)
0.3
(13.8)
(0.2)
0.3
(12.5)
(0.2)
0.3
(12.4)
(0.2)
0.3
(12.4)
(0.2)
0.3
(12.4)
47 NRI survey locations are classified according to land-use histories starting in 1979, and consequently the classifications are
based on less than 20 years from 1990 to 2001. This may have led to an underestimation of Land Converted to Grassland in the
early part of the time series to the extent that some areas are converted to grassland prior to 1979.
48 CO2 emissions associated with liming are also estimated but included in 6.4 Cropland Remaining Cropland.
6-56 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Total Organic Soil Flux 0.7 L2 1.5 1.5 1.5 1.5 1.5
Total Net Flux (8.5) (12.8) (12.3) (11.0) (10.9) (10.9) (10.9)
Notes: Estimates after 2010 are based on NRI data from 2010 and therefore may not fully reflect changes
occurring in the latter part of the time series. Parentheses indicate net sequestration.
+ Does not exceed 0.05 MMT CO2 Eq.
1
2 Table 6-36: Net COz Flux from Soil C Stock Changes for Land Converted to Grass/and (MMT
3 C)
17
18
Soil Type
Cropland Converted to Grassland
Mineral
Organic
Forest Converted to Grassland
Mineral
Organic
Other Lands Converted Grassland
Mineral
Organic
Settlements Converted Grassland
Mineral
Organic
Wetlands Converted Grassland
Mineral
Organic
Total Mineral Soil Flux
Total Organic Soil Flux
Total Net Flux
1990
(2.0)
0.1
(0.1)
+
(0.2)
+
(+) 1
+ 1
(0.1)
+
(2.5)
0.2
(2.3)
2005
(3.1)
0.3 |
(+) 1
+ 1
(0.2)
0.1
(3.8)
0.3 •
(3.5)
2010
(3.2)
0.3
(0.2)
+
(0.2)
+
(+)
+
(0.1)
0.1
(3.8)
0.4
(3.4)
2011
(2.8)
0.3
(0.2)
+
(0.2)
+
(+)
+
(0.1)
0.1
(3.4)
0.4
(3.0)
2012
(2.8)
0.3
(0.2)
+
(0.2)
+
(+)
+
(0.1)
0.1
(3.4)
0.4
(3.0)
2013
(2.8)
0.3
(0.2)
+
(0.2)
+
(+)
+
(0.1)
0.1
(3.4)
0.4
(3.0)
2014
(2.8)
0.3
(0.2)
+
(0.2)
+
(+)
+
(0.1)
0.1
(3.4)
0.4
(3.0)
Notes: Estimates after 2010 are based on NRI data from 2010 and therefore may not fully reflect changes
occurring in the latter part of the time series. Parentheses indicate net sequestration.
+ Does not exceed 0.05 MMT CO2 Eq.
4
5 The spatial variability in the 2014 annual flux in CC>2 from mineral soils is displayed in Figure 6-11 and from
6 organic soils in Figure 6-12. Soil C stocks increased in most states for Land Converted to Grassland, which is
7 driven by conversion of annual cropland into continuous pasture. The largest gains are in the Southeastern region,
8 Northeast, South-Central, Midwest, and northern Great Plains. The regions with the highest rates of emissions from
9 organic soils coincide with the largest concentrations of organic soils used for managed grasslands, including
10 Southeastern Coastal Region (particularly Florida), upper Midwest and Northeast surrounding the Great Lakes, and
11 the Pacific Coast (particularly California).
12 Figure 6-11: Total Net Annual COz Flux for Mineral Soils under Agricultural Management
13 within States, 2014, Land Converted to Grassland'(TO BE UPDATED)
14
15 Figure 6-12: Total Net Annual COz Flux for Organic Soils under Agricultural Management
16 within States, 2014, Land Converted to Grassland'(TO BE UPDATED)
Methodology
19 The following section includes a description of the methodology used to estimate changes in soil C stocks for Land
20 Converted to Grassland, including (1) agricultural land-use and management activities on mineral soils; and (2)
21 agricultural land-use and management activities on organic soils. Biomass and litter C stock changes associated
22 with conversion of forest to grassland are not explicitly included in this category, but are included in the Forest
23 Land Remaining Forest Land section. Further elaboration on the methodologies and data used to estimate stock
24 changes for mineral and organic soils are provided in the Cropland Remaining Cropland section and Annex 3.12.
Land Use, Land-Use Change, and Forestry 6-57
-------
1 Soil C stock changes are estimated for Land Converted to Grassland according to land-use histories recorded in the
2 2010 USDA NRI survey (USDA-NRCS 2013). Land use and some management information (e.g., crop type, soil
3 attributes, and irrigation) were originally collected for each NRI survey locations on a 5-year cycle beginning in
4 1982 In 1998, the NRI program began collecting annual data, and the annual data are currently available through
5 2012 (USDA-NRCS 2015). However, this Inventory only uses NRI data through 2010 because newer data were not
6 available in time to incorporate the additional years. NRI survey locations are classified as Land Converted to
1 Grassland in a given year between 1990 and 2010 if the land use is grassland but had been classified as another use
8 during the previous 20 years.
9 Mineral Soil Carbon Stock Changes
10 An IPCC Tier 3 model-based approach (Ogle et al. 2010) is applied to estimate C stock changes for Land Converted
11 to Grassland on most mineral soils. C stock changes on the remaining soils are estimated with an IPCC Tier 2
12 approach (Ogle et al. 2003), including prior cropland used to produce vegetables, tobacco, and
13 perennial/horticultural crops; land areas with very gravelly, cobbly, or shaley soils (greater than 35 percent by
14 volume); and land converted to grassland from another land use other than cropland.
15 Tier 3 Approach
16 Mineral SOC stocks and stock changes are estimated using the DAYCENT biogeochemical49 model (Parton et al.
17 1998; Del Grosso et al. 2001, 2011). The DAYCENT model utilizes the soil C modeling framework developed in
18 the Century model (Parton et al. 1987, 1988, 1994; Metherell et al. 1993), but has been refined to simulate dynamics
19 at a daily time-step. Historical land-use patterns and irrigation histories are simulated with DAYCENT based on the
20 2010 USDA NRI survey (USDA-NRCS 2013), with supplemental information on fertilizer use and rates from the
21 USDA Economic Research Service Cropping Practices Survey (USDA-ERS 1997, 2011) and the National
22 Agricultural Statistics Service (NASS 1992, 1999, 2004). See the Cropland Remaining Cropland section for
23 additional discussion of the Tier 3 methodology for mineral soils.
24 Tier 2 Approach
25 For the mineral soils not included in the Tier 3 analysis, SOC stock changes are estimated using a Tier 2 Approach
26 for Land Converted to Grassland as described in the Tier 2 Approach for mineral soils in the Grassland Remaining
27 Grassland section.
28 Organic Soil Carbon Stock Changes
29 Annual C emissions from drained organic soils in Land Converted to Grassland are estimated using the Tier 2
30 method provided in IPCC (2006), with U. S. -specific C loss rates (Ogle et al. 2003) as described in the Cropland
31 Remaining Cropland section for organic soils.
32
Uncertainty and Time-Series Consistency
33 Uncertainty analysis for mineral soil C stock changes using the Tier 3 and Tier 2 methodologies are based on a
34 Monte Carlo approach that is described in the Cropland Remaining Cropland section. The uncertainty for annual C
35 emission estimates from drained organic soils in Land Converted to Grassland is estimated using a Monte Carlo
36 approach, which is also described in the Cropland Remaining Cropland section.
37 Uncertainty estimates are presented in Table 6-37 for each subsource (i.e., mineral soil C stocks and organic soil C
38 stocks) and the method applied in the inventory analysis (i.e., Tier 2 and Tier 3). Uncertainty estimates from the
39 Tier 2 and 3 approaches are combined using the simple error propagation methods provided by the IPCC (2006),
40 i.e., by taking the square root of the sum of the squares of the standard deviations of the uncertain quantities. The
Biogeochemical cycles are the flow of chemical elements and compounds between living organisms and the physical
environment.
6-58 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 combined uncertainty for soil C stocks in Land Converted to Grassland ranges from 82 percent below to 82 percent
2 above the 2014 stock change estimate of -10.9 MMT CO2 Eq.
3 Table 6-37: Approach 2 Quantitative Uncertainty Estimates for Soil C Stock Changes
4 occurring within Land Converted to Grass/and (MMJ COz Eq. and Percent)
Source
Cropland Converted to Grassland
Mineral Soil C Stocks: Tier 3
Mineral Soil C Stocks: Tier 2
Organic Soil C Stocks: Tier 2
Forests Converted to Grassland
Mineral Soil C Stocks: Tier 2
Organic Soil C Stocks: Tier 2
Other Lands Converted to Grassland
Mineral Soil C Stocks: Tier 2
Organic Soil C Stocks: Tier 2
Settlements Converted to Grassland
Mineral Soil C Stocks: Tier 2
Organic Soil C Stocks: Tier 2
Wetlands Converted to Grasslands
Mineral Soil C Stocks: Tier 2
Organic Soil C Stocks: Tier 2
Total: Land Converted to Grassland
Mineral Soil C Stocks: Tier 3
Mineral Soil C Stocks: Tier 2
Organic Soil C Stocks: Tier 2
2014 Flux Estimate
(MMT CO2 Eq.)
(9.2)
(9.0)
(1.4)
1.2
(0.8)
(0.8)
+
(0.8)
(0.8)
+
(0.1)
(0.1)
+
0.1
(0.2)
0.3
(10.9)
(9.0)
(3.4)
1.5
Uncertainty Range Relative to Flux Estimate3
(MMT
Lower
Bound
(18.1)
(17.8)
(2.2)
0.4
(4.3)
(1.8)
0.0
(1.3)
(1.3)
0.0
(0.2)
(0.2)
+
(0.1)
(0.4)
0.1
(19.8)
(17.8)
(4.7)
0.7
C02 Eq.)
Upper
Bound
(0.3)
(0.2)
(0.8)
2.3
0.1
0.1
+
(0.4)
(0.4)
+
(0.0)
(0.1)
+
0.3
(0.1)
0.5
(2.0)
(0.2)
(2.2)
2.6
("
Lower
Bound
-96%
-98%
-55%
-63%
-433%
-120%
-100%
-55%
-54%
-100%
-63%
-55%
-79%
-314%
-52%
-51%
-82%
-98%
-39%
-50%
/o)
Upper
Bound
+96%
+98%
+47%
+96%
+112%
+112%
+300%
+47%
+46%
+179%
+56%
+47%
+125%
+382%
+44%
+71%
+82%
+98%
+35%
+77%
Note: Parentheses indicate negative values.
a Range of C stock change estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
5 Uncertainty is also associated with lack of reporting of agricultural biomass and litter C stock changes, other than
6 the loss of forest biomass and litter, which is reported in the Forest Land Remaining Forest Land section of the
7 report. Biomass C stock changes may be significant for managed grasslands with woody encroachment that has not
8 attained enough tree cover to be considered forest lands. This Inventory does not currently include the non-CCh
9 greenhouse gas emissions that occur with biomass burning. Grassland burning is not as common in the United States
10 as in other regions of the world, but fires do occur through both natural ignition sources and prescribed burning.
11 Changes in litter C stocks are assumed to be negligible in grasslands over annual time frames, although there are
12 likely significant changes at sub-annual time scales across seasons.
13 Methodological recalculations are applied to the entire time series to ensure time-series consistency from 1990
14 through 2014. Details on the emission trends through time are described in more detail in the above Methodology
15 section.
16
17
QA/QC and Verification
See the QA/QC and Verification section in Cropland Remaining Cropland.
18 Recalculations Discussion
19 Methodological recalculations in the current Inventory are associated with the following improvements, including 1)
20 incorporation of updated NRI data for 1990 through 2010; 2) inclusion of federal grasslands in the Tier 2 analysis;
21 and 3) improving the simulation of hydric soils in D AYCENT. As a result of these improvements to the Inventory,
22 changes in SOC stocks increased by an average of 0.2 MMT CC>2 eq. annually over the time series.
Land Use, Land-Use Change, and Forestry 6-59
-------
Planned Improvements
2 Soil C stock changes with land use conversion from forest land to grassland are undergoing further evaluation to
3 ensure consistency in the time series. Different methods are used to estimate soil C stock changes in forest land and
4 grasslands, and while the areas have been reconciled between these land uses, there has been limited evaluation of
5 the consistency in C stock changes with conversion from forest land to grassland. This planned improvement may
6 not be fully implemented for another year, depending on resource availability. Another planned improvement for
7 the Land Converted to Grassland category is to develop an inventory of carbon stock changes for grasslands in
8 Alaska. For information about other improvements, see the Planned Improvements section in Cropland Remaining
9 Cropland and Grassland Remaining Grassland.
10
11
29
6.8 Wetlands Remaining Wetlands (IPCC
Source Category 4D1)
12 Peatlands Remaining Peatlands
13 Emissions from Managed Peatlands
14 Managed peatlands are peatlands which have been cleared and drained for the production of peat. The production
15 cycle of a managed peatland has three phases: land conversion in preparation for peat extraction (e.g., clearing
16 surface biomass, draining), extraction, and abandonment, restoration, or conversion of the land to another use.
17 Carbon dioxide emissions from the removal of biomass and the decay of harvested peat constitute the major
18 greenhouse gas flux from managed peatlands. Managed peatlands may also emit CH4 and N2O, however, this is a
19 very small component of total emissions from this source category in the United States. The natural production of
20 CH4 is largely reduced but not entirely shut down when peatlands are drained in preparation for peat extraction
21 (Stack et al. 2004 as cited in the 2006 IPCC Guidelines). Drained land surface and ditch networks contribute to the
22 CH4 flux in peatlands managed for peat extraction. Methane emissions were considered insignificant under IPCC
23 Tier 1 methodology (IPCC 2006), but are included in the emissions estimates for Peatlands Remaining Peatlands
24 consistent with the 2013 Supplement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories:
25 Wetlands (IPCC 2013). Nitrous Oxide emissions from managed peatlands depend on site fertility. In addition,
26 abandoned and restored peatlands continue to release greenhouse gas emissions. This Inventory estimates CO2,
27 N2O, and CH4 emissions from peatlands managed for peat extraction in accordance with IPCC (2006 and 2013)
28 guidelines.
IVhO, and ChU Emissions from Peatlands Remaining Peatlands
30 IPCC (2013) recommends reporting CC>2, N2O, and CH4 emissions from lands undergoing active peat extraction
31 (i.e., Peatlands Remaining Peatlands) as part of the estimate for emissions from managed wetlands. Peatlands occur
32 where plant biomass has sunk to the bottom of water bodies and water-logged areas and exhausted the oxygen
33 supply below the water surface during the course of decay. Due to these anaerobic conditions, much of the plant
34 matter does not decompose but instead forms layers of peat over decades and centuries. In the United States, peat is
35 extracted for horticulture and landscaping growing media, and for a wide variety of industrial, personal care, and
36 other products. It has not been used for fuel in the United States for many decades. Peat is harvested from two
37 types of peat deposits in the United States: sphagnum bogs in northern states (e.g., Minnesota) and wetlands in states
38 further south (e.g., Florida). The peat from sphagnum bogs in northern states, which is nutrient poor, is generally
39 corrected for acidity and mixed with fertilizer. Production from more southerly states is relatively coarse (i.e.,
40 fibrous) but nutrient rich.
41 IPCC (2006 and 2013) recommend considering both on-site and off-site emissions when estimating CCh emissions
42 from Peatlands Remaining Peatlands using the Tier 1 approach. Current methodologies estimate only on-site N2O
6-60 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 and CH4 emissions, since off-site N2O estimates are complicated by the risk of double-counting emissions from
2 nitrogen fertilizers added to horticultural peat, and off-site CH4 emissions are not relevant given the non-energy uses
3 of peat, so methodologies are not provided in IPCC (2013) guidelines. On-site emissions from managed peatlands
4 occur as the land is cleared of vegetation and the underlying peat is exposed to sun and weather. As this occurs,
5 some peat deposit is lost and CO2 is emitted from the oxidation of the peat. Since N2O emissions from saturated
6 ecosystems tend to be low unless there is an exogenous source of nitrogen, N2O emissions from drained peatlands
7 are dependent on nitrogen mineralization and therefore on soil fertility. Peatlands located on highly fertile soils
8 contain significant amounts of organic nitrogen in inactive form. Draining land in preparation for peat extraction
9 allows bacteria to convert the nitrogen into nitrates which leach to the surface where they are reduced to N2O, and
10 contributes to the activity of methanogens, which produce CH4, and methanotrophs which oxidize CH4 into CO2
11 (Blodau 2002; Treat et al. 2007 as cited in IPCC 2013). Ditch networks, which are constructed in order to drain the
12 water off in preparation for peat extraction, also contribute to the flux of CH4 through in situ production and lateral
13 transfer of CH4 from the organic soil matrix (IPCC 2013).
14 The two sources of off-site CO2 emissions from managed peatlands are waterborne carbon losses and the
15 horticultural and landscaping use of peat. Drainage waters in peatlands accumulate dissolved organic carbon which
16 then reacts within aquatic ecosystems and is converted to CO2 where it is then emitted to the atmosphere (Billet et
17 al. 2004 as cited in IPCC 2013). Most (nearly 98 percent) of the CO2 emissions from peat occur off-site, as the peat
18 is processed and sold to firms which, in the United States, use it predominantly forhorticultural and landscaping
19 purposes. Nutrient-poor (but fertilizer-enriched) peat tends to be used in bedding plants and in greenhouse and plant
20 nursery production, whereas nutrient-rich (but relatively coarse) peat is used directly in landscaping, athletic fields,
21 golf courses, and plant nurseries.
22 Total emissions from Peatlands Remaining Peatlands were estimated to be 0.8 MMT CO2 Eq. in 2014 (see Table
23 6-38) comprising 0.8 MMT CO2 Eq. (842 kt) of CO2, 0.001 MMT CO2 Eq. (0.002 kt) of N2O, and 0.004 MMT CO2
24 Eq. (0.17 kt) of CH4. Total emissions in 2014 were about 9 percent larger than total emissions in 2013. Peat
25 production in Alaska in 2014 was not reported inAlaska 's Mineral Industry 2013 report. However, peat production
26 reported in the lower 48 states in 2014 was 10 percent more than in 2013, and as a result, the emissions from
27 Peatlands Remaining Peatlands in the lower 48 states and Alaska were greater in 2014 compared to 2013.
28 Total emissions from Peatlands Remaining Peatlands have fluctuated between 0.8 and 1.3 MMT CO2 Eq. across the
29 time series with a decreasing trend from 1990 until 1993 followed by an increasing trend through 2000. After 2000,
30 emissions generally decreased until 2006 and then increased until 2009, when the trend reversed until a slight
31 increase from2013 to 2014. Carbon dioxide emissions from Peatlands Remaining Peatlands have fluctuated
32 between 0.8 and 1.3 MMT CO2 across the time series, and these emissions drive the trends in total emissions. CH4
33 and N2O emissions remained close to zero across the time series.
34 Table 6-38: Emissions from Peatlands Remaining Peatlands (MMT COz Eq.)
Gas
C02
Off-site
On-site
N2O (On-site)
CH4 (On-site)
Total
1990
1.1
1.0 1
0.1 1
+
+
1.1
2005
1.1
1.0
0.1
+ 1
+
1.1
2010
1.0
1.0
0.1
+
+
1.0
2011
0.9
0.9
0.1
+
+
0.9
2012
0.8
0.8
0.1
+
+
0.8
2013
0.8
0.7
+
+
+
0.8
2014
0.8
0.8
0.1
+
+
0.8
+ Does not exceed 0.05 MMT CO2 Eq.
Note: These numbers are based on U.S. production data in accordance with Tier 1 guidelines, which does
not take into account imports, exports, and stockpiles (i.e., apparent consumption). Off-site N2O emissions
are not estimated to avoid double-counting N2O emitted from the fertilizer that the peat is mixed with prior
to horticultural use (see IPCC 2006). Totals may not sum due to independent rounding.
35
36 Table 6-39: Emissions f rom Peatlands Remaining Peatlands (\Ł)
Gas
C02
Off-site
1990
1,055
985 |
2005
1,101
1,030
2010
1,022
956
2011
926
866
2012
812
760
2013
770
720
2014
842
787
Land Use, Land-Use Change, and Forestry 6-61
-------
On-site 70 | 71 66 60 53 50 55
N2O (On-site)
CH4 (On-site)
+ Does not exceed 0.5 kt.
Note: These numbers are based on U.S. production data in accordance with Tier 1 guidelines, which does not
take into account imports, exports, and stockpiles (i.e., apparent consumption). Off-site N2O emissions are not
estimated to avoid double-counting N2O emitted from the fertilizer that the peat is mixed with prior to
horticultural use (see IPCC 2006). Totals may not sum due to independent rounding.
i Methodology
2 Off-site CO2 Emissions
3 Off-site CO2 emissions from domestic peat production were estimated using a Tier 1 methodology consistent with
4 IPCC (2006). The emissions were calculated by apportioning the annual weight of peat produced in the United
5 States (Table 6-40) into peat extracted from nutrient-rich deposits and peat extracted from nutrient-poor deposits
6 using annual percentage-by-weight figures. These nutrient-rich and nutrient-poor production values were then
7 multiplied by the appropriate default C fraction conversion factor taken from IPCC (2006) in order to obtain off-site
8 CO2 emission estimates. For the lower 48 states, both annual percentages of peat type by weight and domestic peat
9 production data were sourced from estimates and industry statistics provided in the Minerals Yearbook and Mineral
10 Commodity Summaries from the U.S. Geological Survey (USGS 1995-2015a; USGS 2015b). To develop these
11 data, the U.S. Geological Survey (USGS; U.S. Bureau of Mines prior to 1997) obtained production and use
12 information by surveying domestic peat producers. On average, about 75 percent of the peat operations respond to
13 the survey; and USGS estimates data for non-respondents on the basis of prior-year production levels (Apodaca
14 2011).
15 The Alaska estimates rely on reported peat production from the annual Alaska's Mineral Industry reports (DGGS
16 1997-2014). Similar to the U.S. Geological Survey, the Alaska Department of Natural Resources, Division of
17 Geological & Geophysical Surveys (DGGS) solicits voluntary reporting of peat production from producers for the
18 Alaska's Mineral Industry report. However, the report does not estimate production for the non-reporting producers,
19 resulting in larger inter-annual variation in reported peat production from Alaska depending on the number of
20 producers who report in a given year (Szumigala 2011). In addition, in both the lower 48 states and Alaska, large
21 variations in peat production can also result from variations in precipitation and the subsequent changes in moisture
22 conditions, since unusually wet years can hamper peat production. The methodology estimates Alaska emissions
23 separately from lower 48 emissions because the state conducts its own mineral survey and reports peat production
24 by volume, rather than by weight (Table 6-41). However, volume production data were used to calculate off-site
25 CO2 emissions from Alaska applying the same methodology but with volume-specific C fraction conversion factors
26 from IPCC (2006).50 Peat production was not reported for 2014 in Alaska's Mineral Industry 2013 report (DGGS
27 2014); therefore Alaska's peat production in 2014 (reported in cubic yards) was assumed to be equal to its peat
28 production in 2013.
29 Consistent with IPCC (2013) guidelines, off-site CO2 emissions from dissolved organic carbon transported off-site
30 were estimated based on the total area of peatlands managed for peat extraction, which is calculated from production
31 data using the methodology described in the On-Site CO2 Emissions section below. Carbon dioxide emissions from
32 dissolved organic C were estimated by multiplying the area of peatlands by the default emission factor for dissolved
33 organic C provided in IPCC (2013).
34 The apparent consumption of peat, which includes production plus imports minus exports plus the decrease in
35 stockpiles, in the United States is over two-and-a-half times the amount of domestic peat production. However,
36 consistent with the Tier 1 method whereby only domestic peat production is accounted for when estimating off-site
37 emissions, off-site CO2 emissions from the use of peat not produced within the United States are not included in the
38 Inventory. The United States has largely imported peat from Canada for horticultural purposes; from 2010 to 2013,
Peat produced from Alaska was assumed to be nutrient poor; as is the case in Canada, "where deposits of high-quality [but
nutrient poor] sphagnum moss are extensive" (USGS 2008).
6-62 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 imports of sphagnum moss (nutrient-poor) peat from Canada represented 63 percent of total U.S. peat imports
2 (USGS 2015c). Most peat produced in the United States is reed-sedge peat, generally from southern states, which is
3 classified as nutrient rich by IPCC (2006). Higher-tier calculations of CCh emissions from apparent consumption
4 would involve consideration of the percentages of peat types stockpiled (nutrient rich versus nutrient poor) as well
5 as the percentages of peat types imported and exported.
6 Table 6-40: Peat Production of Lower 48 States (kt)
Type of Deposit
Nutrient-Rich
Nutrient-Poor
Total Production
1990
595
55
692
2005
658
27
685
2010
559
69
628
2011
511
57
568
2012
410
78
488
2013
419
47
465
2014
459
51
510
Sources: United States Geological Survey (USGS) (\99\-2Q\5a) Minerals Yearbook: Peat (1994-2014);
United States Geological Survey (USGS) (2015b) Mineral Commodity Summaries: Peat (2014).
Table 6-41: Peat Production of Alaska (Thousand Cubic Meters)
1990 2005 2010 2011 2012 2013 2014
Total Production 49.7 47.8 59.8 61.5 93.1 93.1 93.1
Sources: Division of Geological & Geophysical Surveys (DGGS), Alaska Department of Natural Resources
(1997-2014) Alaska's Mineral Industry Report (1997-2013).
10 On-site CO2 Emissions
11 IPCC (2006) suggests basing the calculation of on-site emission estimates on the area of peatlands managed for peat
12 extraction differentiated by the nutrient type of the deposit (rich versus poor). Information on the area of land
13 managed for peat extraction is currently not available for the United States, but in accordance with IPCC (2006), an
14 average production rate per area for the industry was applied to derive an area estimate. In a mature industrialized
15 peat industry, such as exists in the United States and Canada, the vacuum method can extract up to 100 metric tons
16 per hectare per year (Cleary et al. 2005 as cited in IPCC 2006).51 In the lower 48 states, the area of land managed
17 for peat extraction was estimated using nutrient-rich and nutrient-poor production data and the assumption that 100
18 metric tons of peat are extracted from a single hectare in a single year. The nutrient-rich and nutrient-poor annual
19 land area estimates were then multiplied by the IPCC (2013) default emission factor in order to calculate on-site
20 CO2 emission estimates. Production data are not available by weight for Alaska. In order to calculate on-site
21 emissions resulting from Peatlands Remaining Peatlands in Alaska, the production data by volume were converted
22 to weight using annual average bulk peat density values, and then converted to land area estimates using the same
23 assumption that a single hectare yields 100 metric tons. The IPCC (2006) on-site emissions equation also includes a
24 term which accounts for emissions resulting from the change in C stocks that occurs during the clearing of
25 vegetation prior to peat extraction. Area data on land undergoing conversion to peatlands for peat extraction is also
26 unavailable for the United States. However, USGS records show that the number of active operations in the United
27 States has been declining since 1990; therefore, it seems reasonable to assume that no new areas are being cleared of
28 vegetation for managed peat extraction. Other changes in C stocks in living biomass on managed peatlands are also
29 assumed to be zero under the Tier 1 methodologies (IPCC 2006 and 2013).
30 On-site N2O Emissions
31 IPCC (2006) suggests basing the calculation of on-site N2O emission estimates on the area of nutrient-rich peatlands
32 managed for peat extraction. These area data are not available directly for the United States, but the on-site CCh
33 emissions methodology above details the calculation of area data from production data. In order to estimate N2O
51 The vacuum method is one type of extraction that annually "mills" or breaks up the surface of the peat into particles, which
then dry during the summer months. The air-dried peat particles are then collected by vacuum harvesters and transported from
the area to stockpiles (IPCC 2006).
Land Use, Land-Use Change, and Forestry 6-63
-------
1 emissions, the area of nutrient rich Peatlands Remaining Peatlands was multiplied by the appropriate default
2 emission factor taken from IPCC (2013).
3 On-site CH4 Emissions
4 IPCC (2013) also suggests basing the calculation of on-site CH4 emission estimates on the total area of peatlands
5 managed for peat extraction. Area data is derived using the calculation from production data described in the On-
6 site CO2 Emissions section above. In order to estimate CH4 emissions from drained land surface, the area of
7 Peatlands Remaining Peatlands was multiplied by the emission factor for direct CH4 emissions taken from IPCC
8 (2013). In order to estimate CH4 emissions from drainage ditches, the total area of peatland was multiplied by the
9 default fraction of peatland area that contains drainage ditches, and the appropriate emission factor taken from IPCC
10 (2013).
11 Uncertainty and Time-Series Consistency
12 A Monte Carlo (Approach 2) uncertainty analysis was applied to estimate the uncertainty of CO2, CH4, and N2O
13 emissions from Peatlands Remaining Peatlands, using the following assumptions:
14 • The uncertainty associated with peat production data was estimated to be ± 25 percent (Apodaca 2008) and
15 assumed to be normally distributed.
16 • The uncertainty associated with peat production data stems from the fact that the USGS receives data from
17 the smaller peat producers but estimates production from some larger peat distributors. The peat type
18 production percentages were assumed to have the same uncertainty values and distribution as the peat
19 production data (i.e., ± 25 percent with a normal distribution).
20 • The uncertainty associated with the reported production data for Alaska was assumed to be the same as for
21 the lower 48 states, or ± 25 percent with a normal distribution. It should be noted that the DGGS estimates
22 that around half of producers do not respond to their survey with peat production data; therefore, the
23 production numbers reported are likely to underestimate Alaska peat production (Szumigala 2008).
24 • The uncertainty associated with the average bulk density values was estimated to be ± 25 percent with a
25 normal distribution (Apodaca 2008).
26 • IPCC (2006 and 2013) gives uncertainty values for the emissions factors for the area of peat deposits
27 managed for peat extraction based on the range of underlying data used to determine the emission factors.
28 The uncertainty associated with the emission factors was assumed to be triangularly distributed.
29 • The uncertainty values surrounding the C fractions were based on IPCC (2006) and the uncertainty was
30 assumed to be uniformly distributed.
31 • The uncertainty values associated with the fraction of peatland covered by ditches was assumed to be ± 100
32 percent with a normal distribution based on the assumption that greater than 10 percent coverage, the upper
33 uncertainty bound, is not typical of drained organic soils outside of The Netherlands (IPCC 2013).
34 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 6-42. Carbon dioxide
35 emissions from Peatlands Remaining Peatlands in 2014 were estimated to be between 0.7 and 1.0 MMT CO2 Eq. at
36 the 95 percent confidence level. This indicates a range of 14 percent below to 19 percent above the 2014 emission
37 estimate of 0.8 MMT CO2 Eq. Methane emissions from Peatlands Remaining Peatlands in 2014 were estimated to
38 be between 0.002 and 0.008 MMT CChEq. This indicates a range of 62 percent below to 61 percent above the 2014
39 emission estimate of 0.005 MMT CChEq. Nitrous Oxide emissions from Peatlands Remaining Peatlands in 2014
40 were estimated to be between 0.0003 and 0.0010 MMT CO2 Eq. at the 95 percent confidence level. This indicates a
41 range of 51 percent below to 61 percent above the 2014 emission estimate of 0.0006 MMT CO2 Eq.
42 Table 6-42: Approach 2 Quantitative Uncertainty Estimates for COz, CH4, and NzO Emissions
43 from Peatlands Remaining Peatlands (MMT COz Eq. and Percent)
2014 Emission
Source Gas Estimate
(MMT C02 Eq.)
Uncertainty Range Relative to Emission Estimate3
(MMT C02 Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
Peatlands Remaining Peatlands CO2 0.8
Peatlands Remaining Peatlands CH4 +
0.7 1.0 -14% +19%
+ + -62% +61%
6-64 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Peatlands Remaining Peatlands N2O + + + -51% +61%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
+ Does not exceed 0.05 MMT CO2 Eq.
1
2 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
3 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
4 above.
5 QA/QC and Verification
6 A QA/QC analysis was performed for data gathering and input, documentation, and calculation and no issues were
7 identified.
8 Recalculations Discussion
9 The emission estimates for Peatlands Remaining Peatlands were updated for 2014 using the Peat section of the
10 Mineral Commodity Summaries 2015. The new edition provided 2014 data for the lower 48 states, but data for
11 Alaska were still unavailable. Because no peat production has been reported since Alaska's Mineral Industry 2012
12 report, the 2013 and 2014 values were assumed to be equal to the 2012 value. If updated data are available for the
13 next inventory cycle, this will result in a recalculation in the next Inventory report.
14 Planned Improvements
15 In order to further improve estimates of CO2, N2O, and CEL emissions from Peatlands Remaining Peatlands, future
16 efforts will investigate if data sources exist for determining the quantity of peat harvested per hectare and the total
17 area undergoing peat extraction.
18 The 2013 Supplement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands describes
19 inventory methodologies for various wetland source categories. In the 1990 through 2013 Inventory, updated
20 methods for Peatlands Remaining Peatlands to align them with the IPCC Supplement were begun to be incorporated.
21 For future inventories, the need for additional updates will be evaluated, in order to further address the IPCC
22 Supplement for Peatlands Remaining Peatlands.
23 The 2006 IPCC Guidelines do not cover all wetland types; they are restricted to peatlands drained and managed for
24 peat extraction, conversion to flooded lands, and some guidance for drained organic soils. They also do not cover all
25 of the significant activities occurring on wetlands (e.g., rewetting of peatlands). Since this Inventory only includes
26 Peatlands Remaining Peatlands, additional wetland types and activities found in the 2013 IPCC Supplement (IPCC
27 2013) will be reviewed to determine if they apply to the United States. For those that do, available data will be
28 investigated to allow for the estimation of greenhouse gas fluxes in future Inventory reports.
29
Box 6-6: Progress on Inclusion of Managed Coastal Wetlands in the U.S. Greenhouse Gas Inventory
30 In 2014, the IPCC released the 2013 Supplement to the 2006 IPCC Guidelines for National Greenhouse Gas
31 Inventories: Wetlands (Wetlands Supplement). The Wetlands Supplement provides methods for estimating
32 anthropogenic emissions and removals of greenhouse gases from wetlands and drained soils. Specific consideration
33 is given here to the inclusion of coastal wetlands as part of LULUCF reporting for anthropogenic emissions and
34 removals of CC>2 and CH4 and N2O emissions.
35 In preparation for the next submission of the U.S. Inventory, the United States is exploring methodological
36 approaches based on guidance in the Wetlands Supplement. The goal is to assemble all necessary activity data and
37 emission factors, implement the methods described in the Wetlands Supplement and generate estimates at the Tier 1
38 or 2 level for managed coastal wetlands in the conterminous United States.
39 Fundamental considerations for inclusion of coastal wetlands as part of LULUCF reporting are: (1) how to apply the
40 guidance in the Wetlands Supplement to specify what coastal wetlands are managed, (2) understanding what land
41 use categories coastal wetlands are in (i.e., Forest Land, Cropland, Grassland, Wetlands, Settlements and Other
42 Land) and ensuring there is no overlap or missing lands within the U.S. land use matrix, and (3) understanding how
43 the guidance can be applied when significant greenhouse gas emissions and removals occur in managed coastal
Land Use, Land-Use Change, and Forestry 6-65
-------
1 wetlands outside of the U.S. land use matrix (i.e., seagrass meadows). These issues are under consideration and
2 review by an interagency (U.S. Government) and academic team in anticipation of the next submission of the U.S.
3 Inventory.
4 The availability of data and resources will be primary drivers in determining how the approaches in the Wetlands
5 Supplement are applied. Specifically, the United States will work toward developing its inventory reporting of
6 greenhouse gas emissions and removals from coastal wetlands by: (1) obtaining, collating and refining land use and
7 land-use change data including (a) creating the coastal wetland boundary, (b) recognizing management activities and
8 coastal wetland change resulting in land-use conversion (c) creating seamless integration where coastal wetlands
9 may overlap with other land-use categories, (d) distinguishing salinity levels and soil types to apply appropriate C
10 stocks and emission factors; and (2) developing the sector-specific inventory report for each new category and sub-
11 category by: (a) increasing efforts toward reconciling land cover and land cover change spatial databases (i.e.,
12 Coastal Change Analysis Program) with vegetation, soil C stock and stock change data, and other levels of
13 disaggregation that improve estimation accuracy, (b) developing Tier 1 (or Tier 2, if activity data and emission
14 factors are available) emissions estimates for new source/sink categories under Forest Land, Cropland, Grassland,
15 Wetlands, Settlements and Other Land, and (c) developing Tier 1 (or Tier 2, if activity data and emission factors are
16 available) estimates of new source/sink categories that fall under new subcategories under Wetlands (Other
17 Wetlands Remaining Other Wetlands and Land Converted to Other Wetlands) from the following activities: i) forest
18 management in mangroves, ii) extraction in mangroves, tidal marshes and seagrass meadows (including excavation,
19 aquaculture and salt production), iii) rewetting, revegetation and creation in mangroves, tidal marshes and seagrass
20 meadows, iv) soil drainage in mangroves and tidal marshes (CO2) and v) new categories of CH4 emissions from
21 rewetting of mangroves and tidal marshes and N2O emissions from aquaculture, and (d) developing QA/QC
22 procedures and protocols to be used in generating the estimates, and (e) refining uncertainty estimates.
23
24
25
28
6.9 Land Converted to Wetlands (IPCC Source
Category 4D1) (TO BE UPDATED)
26 Estimates for the Land Converted to Wetlands source category are currently under development.
27
6.10 Settlements Remaining Settlements
29 Changes in Carbon Stocks in Urban Trees (IPCC Source
30 Category 4E1)
31 Urban forests constitute a significant portion of the total U.S. tree canopy cover (Dwyer et al. 2000). Urban areas
32 (cities, towns, and villages) are estimated to cover over 3 percent of the United States (U.S. Census Bureau 2012).
33 With an average tree canopy cover of 35 percent, urban areas account for approximately 5 percent of total tree cover
34 in the continental United States (Nowak and Greenfield 2012). Trees in urban areas of the United States were
35 estimated to account for an average annual net sequestration of 76.4 MMT CO2 Eq. (20.8 MMT C) over the period
36 from 1990 through 2014. Net C flux from urban trees in 2014 was estimated to be -90.6 MMT CO2 Eq. (-24.7
37 MMT C). Annual estimates of CO2 flux (Table 6-43) were developed based on periodic (1990, 2000, and 2010)
38 U.S. Census data on urbanized area. The estimate of urbanized area is smaller than the area categorized as
39 Settlements in the Representation of the U.S. Land Base developed for this report: over the 1990 through 2014 time
40 series the Census urban area totaled, on average, about 63 percent of the Settlements area.
6-66 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 In 2014, Census urban area totaled about 68 percent of the total area defined as Settlements. Census area data are
2 preferentially used to develop C flux estimates for this source category since these data are more applicable for use
3 with the available peer-reviewed data on urban tree canopy cover and urban tree C sequestration. Annual
4 sequestration increased by 50 percent between 1990 and 2014 due to increases in urban land area. Data on C storage
5 and urban tree coverage were collected since the early 1990s and have been applied to the entire time series in this
6 report. As a result, the estimates presented in this chapter are not truly representative of changes in C stocks in
7 urban trees for Settlements areas, but are representative of changes in C stocks in urban trees for Census urban area.
8 The method used in this report does not attempt to scale these estimates to the Settlements area. Therefore, the
9 estimates presented in this chapter are likely an underestimate of the true changes in C stocks in urban trees in all
10 Settlements areas—i.e., the changes in C stocks in urban trees presented in this chapter are a subset of the changes in
11 C stocks in urban trees in all Settlements areas.
12 Urban trees often grow faster than forest trees because of the relatively open structure of the urban forest (Nowak
13 and Crane 2002). Because tree density in urban areas is typically much lower than in forested areas, the C storage
14 per hectare of land is in fact smaller for urban areas than for forest areas. To quantify the C stored in urban trees, the
15 methodology used here requires analysis per unit area of tree cover, rather than per unit of total land area (as is done
16 for forests). Expressed in this way per unit of tree cover, areas covered by urban trees actually have a greater C
17 density than do forested areas (Nowak and Crane 2002). Expressed per unit of land area, however, the situation is
18 the opposite: because tree density is so much lower in urban areas, these areas have a smaller C density per unit land
19 area than forest areas.
20 Table 6-43: Net C Flux from Urban Trees (MMT COz Eq. and MMT C)
Year MMT CCh Eq. MMT C
1990 (60.4) (16.5)
2010
2011
2012
2013
2014
(86.1)
(87.3)
(88.4)
(89.5)
(90.6)
(23.5)
(23.8)
(24.1)
(24.4)
(24.7)
Note: Parentheses indicate net
sequestration.
21 Methodology
22 Methods for quantifying urban tree biomass, C sequestration, and C emissions from tree mortality and
23 decomposition were taken directly from Nowak et al. (2013), Nowak and Crane (2002), and Nowak (1994). In
24 general, the methodology used by Nowak et al. (2013) to estimate net C sequestration in urban trees followed three
25 steps, each of which is explained further in the paragraphs below. First, field data from cities and states were used
26 to develop allometric equations that are then used to estimate C in urban tree biomass from data on measured tree
27 dimensions. Second, estimates of annual tree growth and biomass increment were generated from published
28 literature and adjusted for tree condition, land-use class, and growing season to generate estimates of gross C
29 sequestration in urban trees for all 50 states and the District of Columbia. Third, estimates of C emissions due to
30 mortality and decomposition were subtracted from gross C sequestration values to derive estimates of net C
31 sequestration.
32 For this Inventory report, net C sequestration estimates for all 50 states and the District of Columbia, that were
33 generated using the Nowak et al. (2013) methodology and expressed in units of C sequestered per unit area of tree
34 cover, were then used to estimate urban tree C sequestration in the United States. To accomplish this, we used urban
35 area estimates from U.S. Census data together with urban tree cover percentage estimates for each state and the
36 District of Columbia from remote sensing data, an approach consistent with Nowak et al. (2013).
37 This approach is also consistent with the default IPCC Gain-Loss methodology in IPCC (2006), although sufficient
38 field data are not yet available to separately determine interannual gains and losses in C stocks in the living biomass
Land Use, Land-Use Change, and Forestry 6-67
-------
1 of urban trees. Instead, the methodology applied here uses estimates of net C sequestration based on modeled
2 estimates of decomposition, as given by Nowak et al. (2013).
3 The first step in the methodology is to develop allometric equations that can be used to estimate C in urban tree
4 biomass. In order to generate these allometric relationships between tree dimensions and tree biomass for cities and
5 states, Nowak et al. (2013) and previously published research (Nowak and Crane 2002; Nowak 1994, 2007b, 2009)
6 collected field measurements in a number of U.S. cities between 1989 and 2012. For a sample of trees in
7 representative cities, data including tree measurements of stem diameter, tree height, crown height and crown width,
8 and information on location, species, and canopy condition were collected. The data for each tree were converted
9 into C storage by applying allometric equations to estimate aboveground biomass, a root-to-shoot ratio to convert
10 aboveground biomass estimates to whole tree biomass, moisture content, a C content of 50 percent (dry weight
11 basis), and an adjustment factor of 0.8 to account for urban trees having less aboveground biomass for a given stem
12 diameter than predicted by allometric equations based on forest trees (Nowak 1994). Carbon storage estimates for
13 deciduous trees include only C stored in wood. These calculations were then used to develop an allometric equation
14 relating tree dimensions to C storage for each species of tree, encompassing a range of diameters.
15 The second step in the methodology is to estimate rates of tree growth for urban trees in the United States. Tree
16 growth was estimated using annual height growth and diameter growth rates for specific land uses and diameter
17 classes. In the Nowak et al. (2013) methodology that is applied here, growth calculations were adjusted by a factor
18 to account for tree condition (fair to excellent, poor, critical, dying, or dead). For each tree, the difference in C
19 storage estimates between year 1 and year (x + 1) represents the gross amount of C sequestered. These annual gross
20 C sequestration rates for each species (or genus), diameter class, and land-use condition (e.g., parks, transportation,
21 vacant, golf courses) were then scaled up to city estimates using tree population information. The area of
22 assessment for each city or state was defined by its political boundaries; parks and other forested urban areas were
23 thus included in sequestration estimates (Nowak 2011).
24 Most of the field data used to develop the methodology of Nowak et al. (2013) were analyzed using the U.S. Forest
25 Service's Urban Forest Effects (UFORE) model. UFORE is a computer model that uses standardized field data
26 from random plots in each city and local air pollution and meteorological data to quantify urban forest structure,
27 values of the urban forest, and environmental effects, including total C stored and annual C sequestration. UFORE
28 was used with field data from a stratified random sample of plots in each city to quantify the characteristics of the
29 urban forest (Nowak et al. 2007).
30 Where gross C sequestration accounts for all carbon sequestered, net C sequestration for urban trees takes into
31 account C emissions associated with tree death and removals. In the third step in the methodology developed by
32 Nowak etal. (2013), estimates of net C emissions from urban trees were derived by applying estimates of annual
33 mortality and condition, and assumptions about whether dead trees were removed from the site to the total C stock
34 estimate for each city. Estimates of annual mortality rates by diameter class and condition class were derived from a
35 study of street-tree mortality (Nowak 1986). Different decomposition rates were applied to dead trees left standing
36 compared with those removed from the site. For removed trees, different rates were applied to the
37 removed/aboveground biomass in contrast to the belowground biomass. The estimated annual gross C emission
38 rates for each species (or genus), diameter class, and condition class were then scaled up to city estimates using tree
39 population information.
40 The data for all 50 states and the District of Columbia are described in Nowak et al. (2013) and reproduced in Table
41 6-44, which builds upon previous research, including: Nowak and Crane (2002), Nowak et al. (2007), Nowak and
42 Greenfield (2012), and references cited therein. The full methodology development is described in the underlying
43 literature, and key details and assumptions were made as follows. The allometric equations applied to the field data
44 for the Nowak methodology for each tree were taken from the scientific literature (see Nowak 1994, Nowak et al.
45 2002), but if no allometric equation could be found for the particular species, the average result for the genus was
46 used. The adjustment (0.8) to account for less live tree biomass in urban trees was based on information in Nowak
47 (1994). Measured tree growth rates for street (Frelich 1992; Fleming 1988; Nowak 1994), park (deVries 1987), and
48 forest (Smith and Shifley 1984) trees were standardized to an average length of growing season (153 frost free days)
49 and adjusted for site competition and tree condition. Standardized growth rates of trees of the same species or genus
50 were then compared to determine the average difference between standardized street tree growth and standardized
51 park and forest growth rates. Crown light exposure (CLE) measurements (number of sides and/or top of tree
52 exposed to sunlight) were used to represent forest, park, and open (street) tree growth conditions. Local tree base
53 growth rates (BG) were then calculated as the average standardized growth rate for open-grown trees multiplied by
6-68 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 the number of frost free days divided by 153. Growth rates were then adjusted for CLE. The CLE adjusted growth
2 rate was then adjusted based on tree health and tree condition to determine the final growth rate. Assumptions for
3 which dead trees would be removed versus left standing were developed specific to each land use and were based on
4 expert judgment of the authors. Decomposition rates were based on literature estimates (Nowak et al. 2013).
5 Estimates of gross and net sequestration rates for each of the 50 states and the District of Columbia (Table 6-44)
6 were compiled in units of C sequestration per unit area of tree canopy cover. These rates were used in conjunction
7 with estimates of state urban area and urban tree cover data to calculate each state's annual net C sequestration by
8 urban trees. This method was described in Nowak et al. (2013) and has been modified here to incorporate U.S.
9 Census data.
10 Specifically, urban area estimates were based on 1990, 2000, and 2010 U.S. Census data. The 1990 U.S. Census
11 defined urban land as "urbanized areas," which included land with a population density greater than 1,000 people
12 per square mile, and adjacent "urban places," which had predefined political boundaries and a population total
13 greater than 2,500. In 2000, the U.S. Census replaced the "urban places" category with a new category of urban
14 land called an "urban cluster," which included areas with more than 500 people per square mile. In 2010, the
15 Census updated its definitions to have "urban areas" encompassing Census tract delineated cities with 50,000 or
16 more people, and "urban clusters" containing Census tract delineated locations with between 2,500 and 50,000
17 people. Urban land area increased by approximately 23 percent from 1990 to 2000 and 14 percent from 2000 to
18 2010; Nowak et al. (2005) estimate that the changes in the definition of urban land are responsible for approximately
19 20 percent of the total reported increase in urban land area from 1990 to 2000. Under all Census (i.e., 1990, 2000,
20 and 2010) definitions, the urban category encompasses most cities, towns, and villages (i.e., it includes both urban
21 and suburban areas). Settlements area, as assessed in the Representation of the U.S. Land Base developed for this
22 report, encompassed all developed parcels greater than 0.1 hectares in size, including rural transportation corridors,
23 and as previously mentioned represents a larger area than the Census-derived urban area estimates. However, the
24 smaller, Census-derived urban area estimates were deemed to be more suitable for estimating national urban tree
25 cover given the data available in the peer-reviewed literature (i.e., the data set available is consistent with Census
26 urban rather than Settlements areas), and the recognized overlap in the changes in C stocks between urban forest and
27 non-urban forest (see Planned Improvements below). U.S. Census urban area data is reported as a series of
28 continuous blocks of urban area in each state. The blocks or urban area were summed to create each state's urban
29 area estimate.
30 Net annual C sequestration estimates were derived for all 50 states and the District of Columbia by multiplying the
31 gross annual emission estimates by 0.74, the standard ratio for net/gross sequestration set out in Table 3 of Nowak et
32 al. (2013) (unless data existed for both gross and net sequestration for the state in Table 2 of Nowak et. al. (2013), in
33 which case they were divided to get a state-specific ratio). The gross and net annual C sequestration values for each
34 state were multiplied by each state's area of tree cover, which was the product of the state's urban/community area
35 as defined in the U.S. Census (2012) and the state's urban/community tree cover percentage. The urban/community
36 tree cover percentage estimates for all 50 states were obtained from Nowak and Greenfield (2012), which compiled
37 ten years of research including Dwyer et al. (2000), Nowak et al. (2002), Nowak (2007a), and Nowak (2009). The
38 urban/community tree cover percentage estimate for the District of Columbia was obtained from Nowak et al.
39 (2013). The urban area estimates were taken from the 2010 U.S. Census (2012). The equation, used to calculate the
40 summed carbon sequestration amounts, can be written as follows:
41 Net annual C sequestration = Gross sequestration rate x Net to Gross sequestration ratio x Urban Area x
42 % Tree Cover
43 Table 6-44: Annual C Sequestration (Metric Tons C/yr), Tree Cover (Percent), and Annual C
44 Sequestration per Area of Tree Cover (kg C/m2-yr) for 50 states plus the District of Columbia
45 (2014)
State
Alabama
Alaska
Arizona
Arkansas
Gross Annual
Sequestration
1,165,574
44,744
385,644
424,922
Net Annual
Sequestration
862,524
33,111
285,376
314,443
Tree
Cover
55.2
39.8
17.6
42.3
Gross Annual
Sequestration
per Area of
Tree Cover
0.343
0.168
0.354
0.331
Net Annual
Sequestration
per Area of
Tree Cover
0.254
0.124
0.262
0.245
Net: Gross
Annual
Sequestration
Ratio
0.74
0.74
0.74
0.74
Land Use, Land-Use Change, and Forestry 6-69
-------
California
Colorado
Connecticut
Delaware
DC
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Total
2,106,024
153,806
771,006
136,070
14,559
3,429,742
2,580,659
246,168
25,533
760,263
406,015
119,006
186,077
243,641
749,632
108,092
597,897
1,309,649
740,048
354,139
494,558
498,925
53,940
50,920
44,096
250,531
1,201,070
70,002
1,096,654
2,076,636
14,946
927,316
366,160
261,067
1,264,702
137,147
1,107,882
21,348
1,063,362
2,808,539
91,713
46,571
839,610
571,062
255,369
364,611
19,203
33,056,852
1,558,458
113,817
570,544
100,692
11,569
2,538,009
1,909,688
182,164
18,894
562,594
375,425
88,064
144,799
180,295
554,727
79,988
442,444
969,140
547,635
262,063
365,973
369,205
39,916
42,970
32,631
185,393
888,792
51,801
811,524
1,536,711
7,102
686,214
270,959
193,190
935,879
101,489
819,832
18,513
950,771
2,078,319
67,868
34,462
621,311
422,586
188,973
269,812
14,210
24,712,872
25.1
18.5
67.4
35.0
35.0
35.5
54.1
39.9
10.0
25.4
23.7
19.0
25.0
22.1
34.9
52.3
34.3
65.1
35.0
34.0
47.3
31.5
36.3
15.0
9.6
66.0
53.3
12.0
42.6
51.1
13.0
31.5
31.2
36.6
41.0
51.0
48.9
14.0
43.8
31.4
16.4
53.0
39.8
34.6
61.0
31.8
19.9
0.389
0.197
0.239
0.335
0.263
0.475
0.353
0.581
0.184
0.283
0.250
0.240
0.283
0.286
0.397
0.221
0.323
0.254
0.220
0.229
0.344
0.285
0.184
0.238
0.207
0.217
0.294
0.263
0.240
0.312
0.223
0.248
0.332
0.242
0.244
0.258
0.338
0.236
0.303
0.368
0.215
0.213
0.293
0.258
0.241
0.225
0.182
0.288
0.146
0.177
0.248
0.209
0.352
0.261
0.430
0.136
0.209
0.231
0.178
0.220
0.212
0.294
0.164
0.239
0.188
0.163
0.169
0.255
0.211
0.136
0.201
0.153
0.161
0.218
0.195
0.178
0.231
0.106
0.184
0.246
0.179
0.181
0.191
0.250
0.205
0.271
0.272
0.159
0.158
0.217
0.191
0.178
0.167
0.135
0.74
0.74
0.74
0.74
0.79
0.74
0.74
0.74
0.74
0.74
0.92
0.74
0.78
0.74
0.74
0.74
0.74
0.74
0.74
0.74
0.74
0.74
0.74
0.84
0.74
0.74
0.74
0.74
0.74
0.74
0.48
0.74
0.74
0.74
0.74
0.74
0.74
0.87
0.89
0.74
0.74
0.74
0.74
0.74
0.74
0.74
0.74
i Uncertainty and Time-Series Consistency
2 Uncertainty associated with changes in C stocks in urban trees includes the uncertainty associated with urban area,
3 percent urban tree coverage, and estimates of gross and net C sequestration for each of the 50 states and the District
4 of Columbia. A 10 percent uncertainty was associated with urban area estimates based on expert judgment.
5 Uncertainty associated with estimates of percent urban tree coverage for each of the 50 states was based on standard
6 error estimates reported by Nowak and Greenfield (2012). Uncertainty associated with estimate of percent urban
7 tree coverage for the District of Columbia was based on the standard error estimate reported by Nowak et al. (2013).
8 Uncertainty associated with estimates of gross and net C sequestration for each of the 50 states and the District of
9 Columbia was based on standard error estimates for each of the state-level sequestration estimates reported by
10 Nowak et al. (2013). These estimates are based on field data collected in each of the 50 states and the District of
11 Columbia, and uncertainty in these estimates increases as they are scaled up to the national level.
6-70 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Additional uncertainty is associated with the biomass equations, conversion factors, and decomposition assumptions
2 used to calculate C sequestration and emission estimates (Nowak et al. 2002). These results also exclude changes in
3 soil C stocks, and there may be some overlap between the urban tree C estimates and the forest tree C estimates.
4 Due to data limitations, urban soil flux is not quantified as part of this analysis, while reconciliation of urban tree
5 and forest tree estimates will be addressed through the land-representation effort described in the Planned
6 Improvements section of this chapter.
7 A Monte Carlo (Approach 2) uncertainty analysis was applied to estimate the overall uncertainty of the
8 sequestration estimate. The results of the Approach 2 quantitative uncertainty analysis are summarized in Table
9 6-45. The net C flux from changes in C stocks in urban trees in 2014 was estimated to be between -134.0 and -47.4
10 MMT CO2 Eq. at a 95 percent confidence level. This indicates a range of 51 percent more sequestration to 46
11 percent less sequestration than the 2014 flux estimate of-90.6 MMT CO2 Eq.
12 Table 6-45: Approach 2 Quantitative Uncertainty Estimates for Net C Flux from Changes in C
13 Stocks in Urban Trees (MMT COz Eq. and Percent)
2014 Flux Estimate Uncertainty Range Relative to Flux Estimate3
Source Gas (MMT CCh Eq.) (MMT CCh Eq.) (%)
Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
Changes in C Stocks in _51% +46%
Urban Trees
Note: Parentheses indicate negative values or net sequestration.
a Range of flux estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
14 Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
15 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
16 above.
17 QA/QC and Verification
18 Tier 1 and Tier 2 QA/QC activities were conducted consistent with the U.S. QA/QC plan. Source-specific quality
19 control measures for urban trees included checking input data, documentation, and calculations to ensure data were
20 properly handled through the Inventory process. Errors that were found during this process were corrected as
21 necessary. One key edit in the current Inventory report is that Table 6-44 has been updated. For this Table, the
22 values in the 2014 (1990-2012) and 2015 (1990-2013) Inventory reports were the same. The updated values forthe
23 2016 (1990-2014) Inventory were inserted here, noting that they represent a two-year increment in urban tree C
24 sequestration from what was presented in the previous Inventory.
25 Planned Improvements
26 A consistent representation of the managed land base in the United States is discussed in the Representation of the
27 U.S. Land Base chapter, and discusses a planned improvement by the USD A Forest Service to reconcile the overlap
28 between urban forest and non-urban forest greenhouse gas inventories. Because some plots defined as "forest" in
29 the Forest Inventory and Analysis (FIA) program of the USD A Forest Service actually fall within the boundaries of
30 the areas also defined as Census urban, there may be "double-counting" of these land areas in estimates of C stocks
31 and fluxes for this report. Specifically, Nowak et al. (2013) estimates that 1.5 percent of forest plots measured by
32 the FIA program fall within land designated as Census urban, suggesting that approximately 1.5 percent of the C
33 reported in the Forest source category might also be counted in the Urban Trees source category.
34 Future research may also enable more complete coverage of changes in the C stock in urban trees for all Settlements
35 land. To provide estimates for all Settlements, research would need to establish the extent of overlap between the
36 areas of land included in the Settlements land use category and Census-defined urban areas, and would have to
37 separately characterize sequestration on non-urban Settlements land.
Land Use, Land-Use Change, and Forestry 6-71
-------
1
N2O Fluxes from Settlement Soils (IPCC Source Category 4E1)
2 Of the synthetic N fertilizers applied to soils in the United States, approximately 3.1 percent are currently applied to
3 lawns, golf courses, and other landscaping occurring within settlement areas. Application rates are lower than those
4 occurring on cropped soils, and, therefore, account for a smaller proportion of total U.S. soil N2O emissions per unit
5 area. In addition to synthetic N fertilizers, a portion of surface applied sewage sludge is applied to settlement areas.
6 N additions to soils result in direct and indirect N2O emissions. Direct emissions occur on-site due to the N
7 additions. Indirect emissions result from fertilizer and sludge N that is transformed and transported to another
8 location in a form other than N2O (NH3 and NOX volatilization, NOs leaching and runoff), and later converted into
9 N2O at the off-site location. The indirect emissions are assigned to settlements because the management activity
10 leading to the emissions occurred in settlements.
11 Total N2O emissions from settlement soils were 2.4 MMT CO2 Eq. (8 kt of N2O) in 2014. There was an overall
12 increase of 78 percent from 1990 to 2014 due to an expanding settlement area requiring more synthetic N fertilizer.
13 Interannual variability in these emissions is directly attributable to interannual variability in total synthetic fertilizer
14 consumption and sewage sludge applications in the United States. Emissions from this source are summarized in
15 Table 6-46.
16 Table 6-46: NzO Fluxes from Soils in Settlements Remaining Settlements(MMT COz Eq. and
17 kt N2O)
Direct N2O Fluxes from Soils
MMT CO2 Eq.
ktN2O
Indirect N2O Fluxes from Soils
MMT CO2 Eq.
ktN20
Total
MMT CO2 Eq.
ktN20
1990
1.0
3
0.4
!_•
1.4
5
2005
1.8
6
0.6
2
2.3
8_M
2010
1.8
6
0.6
2
2.4
8
2011
1.9
6
0.6
2
2.5
8
2012
1.9
6
0.6
2
2.5
9
2013
1.8
6
0.6
2
2.4
8
2014
1.8
6
0.6
2
2.4
8
18 Methodology
19 For soils within Settlements Remaining Settlements, the IPCC Tier 1 approach is used to estimate soil N2O emissions
20 from synthetic N fertilizer and sewage sludge additions. Estimates of direct N2O emissions from soils in settlements
21 are based on the amount of N in synthetic commercial fertilizers applied to settlement soils, and the amount of N in
22 sewage sludge applied to non-agricultural land and surface disposal (see Annex 3.12 for a detailed discussion of the
23 methodology for estimating sewage sludge application).
24 Nitrogen applications to settlement soils are estimated using data compiled by the USGS (Ruddy et al. 2006). The
25 USGS estimated on-farm and non-farm fertilizer use is based on sales records at the county level from 1982 through
26 2001 (Ruddy et al. 2006). Non-farm N fertilizer is assumed to be applied to settlements and forest lands; values for
27 2002 through 2014 are based on 2001 values adjusted for annual total N fertilizer sales in the United States because
28 there is no new activity data on application after 2001. Settlement application is calculated by subtracting forest
29 application from total non-farm fertilizer use. Sewage sludge applications are derived from national data on sewage
30 sludge generation, disposition, and N content (see Annex 3.12 for further detail). The total amount of N resulting
31 from these sources is multiplied by the IPCC default emission factor for applied N (1 percent) to estimate direct N2O
32 emissions (IPCC 2006).
33 For indirect emissions, the total N applied from fertilizer and sludge is multiplied by the IPCC default factors of 10
34 percent for volatilization and 30 percent for leaching/runoff to calculate the amount of N volatilized and the amount
35 of N leached/runoff. The amount of N volatilized is multiplied by the IPCC default factor of 1 percent for the
36 portion of volatilized N that is converted to N2O off-site and the amount of N leached/runoff is multiplied by the
37 IPCC default factor of 0.075 percent for the portion of leached/runoff N that is converted to N2O off-site. The
38 resulting estimates are summed to obtain total indirect emissions.
6-72 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
i Uncertainty and Time-Series Consistency
2 The amount of N2O emitted from settlements depends not only on N inputs and fertilized area, but also on a large
3 number of variables, including organic C availability, oxygen gas partial pressure, soil moisture content, pH,
4 temperature, and irrigation/watering practices. The effect of the combined interaction of these variables on N2O flux
5 is complex and highly uncertain. The IPCC default methodology does not explicitly incorporate any of these
6 variables, except variations in fertilizer N and sewage sludge application rates. All settlement soils are treated
7 equivalently under this methodology.
8 Uncertainties exist in both the fertilizer N and sewage sludge application rates in addition to the emission factors.
9 Uncertainty in fertilizer N application is assigned a default level of ±50 percent.52 Uncertainty in the amounts of
10 sewage sludge applied to non-agricultural lands and used in surface disposal is derived from variability in several
11 factors, including: (1) N content of sewage sludge; (2) total sludge applied in 2000; (3) wastewater existing flow in
12 1996 and 2000; and (4) the sewage sludge disposal practice distributions to non-agricultural land application and
13 surface disposal. In addition, the uncertainty ranges around 2005 activity data and emission factor input variables
14 are directly applied to the 2014 emission estimates. Uncertainty in the direct and indirect emission factors is
15 provided by IPCC (2006).
16 Uncertainty is quantified using simple error propagation methods (IPCC 2006), and the results are summarized in
17 Table 6-47. Direct N2O emissions from soils in Settlements Remaining Settlements in 2014 are estimated to be
18 between 0.9 and 4.8 MMT CO2 Eq. at a 95 percent confidence level. This indicates a range of 49 percent below to
19 163 percent above the 2014 emission estimate of 1.8 MMT CO2 Eq. Indirect N2O emissions in 2014 are between
20 0.1 and 1.9 MMT CO2 Eq., ranging from a -85 percent to 212 percent around the estimate of 0.6 MMT CO2 Eq.
21 Table 6-47: Quantitative Uncertainty Estimates of NzO Emissions from Soils in Settlements
22 Remaining Settlements (MMT COz Eq. and Percent)
Source
Settlements Remaining
Settlements:
Direct N2O Fluxes from Soils
Indirect N2O Fluxes from Soils
_ 2014 Emissions Uncertainty Range Relative to Emission Estimate
(MMTCChEq.) (MMT CCh Eq.) (%)
N20
N2O
1.8
0.6
Lower
Bound
0.9
0.1
Upper
Bound
4.8
1.9
Lower
Bound
-49%
-85%
Upper
Bound
+163%
+212%
Note: These estimates include direct and indirect N2O emissions from N fertilizer additions to both Settlements Remaining
Settlements and from Land Converted to Settlements.
23 Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
24 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
25 above.
26 QA/QC and Verification
27 The spreadsheet containing fertilizer and sewage sludge applied to settlements and calculations for N2O and
28 uncertainty ranges have been checked and verified.
29 Planned Improvements
30 A minor improvement is planned to update the uncertainty analysis for direct emissions from settlements to be
31 consistent with the most recent activity data for this source.
No uncertainty is provided with the USGS fertilizer consumption data (Ruddy et al. 2006) so a conservative ±50 percent is
used in the analysis.
Land Use, Land-Use Change, and Forestry 6-73
-------
i 6.11 Land Converted to Settlements (IPCC
2 Source Category 4E2)
3 Land-use change is constantly occurring, and land under a number of uses undergoes urbanization in the United
4 States each year. However, data on the amount of land converted to settlements is currently lacking. Given the lack
5 of available information relevant to this particular IPCC source category, it is not possible to separate CCh or N2O
6 fluxes on Land Converted to Settlements from fluxes on Settlements Remaining Settlements at this time.
7 6.12 Other (IPCC Source Category 4H)
s Changes in Yard Trimming and Food Scrap Carbon Stocks in
9 Landfills
10 In the United States, yard trimmings (i.e., grass clippings, leaves, and branches) and food scraps account for a
11 significant portion of the municipal waste stream, and a large fraction of the collected yard trimmings and food
12 scraps are put in landfills. Carbon (C) contained in landfilled yard trimmings and food scraps can be stored for very
13 long periods.
14 Carbon-storage estimates are associated with particular land uses. For example, harvested wood products are
15 accounted for under Forest Land Remaining Forest Land because these wood products are considered a component
16 of the forest ecosystem. The wood products serve as reservoirs to which C resulting from photosynthesis in trees is
17 transferred, but the removals in this case occur in the forest. Carbon stock changes in yard trimmings and food
18 scraps are associated with settlements, but removals in this case do not occur within settlements. To address this
19 complexity, yard trimming and food scrap C storage is reported under the "Other" source category.
20 Both the amount of yard trimmings collected annually and the fraction that is landfilled have declined over the last
21 decade. In 1990, over 53 million metric tons (wet weight) of yard trimmings and food scraps were generated (i.e.,
22 put at the curb for collection to be taken to disposal sites or to composting facilities) (EPA 2015a). Since then,
23 programs banning or discouraging yard trimmings disposal have led to an increase in backyard composting and the
24 use of mulching mowers, and a consequent 2.3 percent decrease in the tonnage of yard trimmings generated (i.e.,
25 collected for composting or disposal). At the same time, an increase in the number of municipal composting
26 facilities has reduced the proportion of collected yard trimmings that are discarded in landfills—from 72 percent in
27 1990 to 32 percent in 2014. The net effect of the reduction in generation and the increase in composting is a 57
28 percent decrease in the quantity of yard trimmings disposed of in landfills since 1990.
29 Food scrap generation has grown by 55 percent since 1990, and though the proportion of food scraps discarded in
30 landfills has decreased slightly from 82 percent in 1990 to 76 percent in 2014, the tonnage disposed of in landfills
31 has increased considerably (by 45 percent). Although the total tonnage of food scraps disposed in landfills has
32 increased from 1990 to 2014, the annual carbon stock net changes from food scraps have decreased (as shown in
33 Table 6-48 and Table 6-49), due to smaller annual differences in the amount of food waste disposed in landfills.
34 Overall, the decrease in the landfill disposal rate of yard trimmings has more than compensated for the increase in
35 food scrap disposal in landfills, and the net result is a decrease in annual landfill C storage from 26.0 MMT CO2 Eq.
36 (7.1 MMT C) in 1990 to 11.6 MMT CO2 Eq. (3.2 MMT C) in 2014 (Table 6-48 and Table 6-49).
6-74 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 6-48: Net Changes in Yard Trimming and Food Scrap Carbon Stocks in Landfills
2 (MMT COz Eq.)
Carbon Pool
Yard Trimmings
Grass
Leaves
Branches
Food Scraps
Total Net Flux
1990
(21.0)
(1.8)
(9.0)
(10.2)
(5.0) •
(26.0)
2005 2010
(7.4) (9.3)
(0.6) (0.9)
(3.4) (4.2)
(3.4) (4.1)
(4.0) (3.9)
(11.4) (13.2)
2011
(9.2)
(0.9)
(4.2)
(4.1)
(3.5)
(12.7)
2012
(9.1)
(0.9)
(4.2)
(4.1)
(3.1)
(12.2)
2013
(8.5)
(0.8)
(3.9)
(3.8)
(3.2)
(11.7)
2014
(8.5)
(0.8)
(3.9)
(3.8)
(3.1)
(11.6)
Note: Parentheses indicate net sequestration.
3 Table 6-49: Net Changes in Yard Trimming and Food Scrap Carbon Stocks in Landfills
4 (MMT C)
"Carbon Pool 1990 2005 2010 2011 2012 2013 2014
Yard Trimmings
Grass
Leaves
Branches
Food Scraps
Total Net Flux (7.1) (3.1) (3.6) (3.5) (3.3) (3.2) (3.2)
Note: Parentheses indicate net sequestration.
Methodology
6 When wastes of biogenic origin (such as yard trimmings and food scraps) are landfilled and do not completely
7 decompose, the C that remains is effectively removed from the terrestrial C cycle. Empirical evidence indicates that
8 yard trimmings and food scraps do not completely decompose in landfills (Barlaz 1998, 2005, 2008; De la Cruz and
9 Barlaz 2010), and thus the stock of C in landfills can increase, with the net effect being a net atmospheric removal of
10 C. Estimates of net C flux resulting from landfilled yard trimmings and food scraps were developed by estimating
11 the change in landfilled C stocks between inventory years, based on methodologies presented for the Land Use,
12 Land-Use Change, and Forestry sector in IPCC (2003). Carbon stock estimates were calculated by determining the
13 mass of landfilled C resulting from yard trimmings or food scraps discarded in a given year; adding the accumulated
14 landfilled C from previous years; and subtracting the mass of C that was landfilled in previous years that
15 decomposed.
16 To determine the total landfilled C stocks for a given year, the following were estimated: (1) The composition of the
17 yard trimmings; (2) the mass of yard trimmings and food scraps discarded in landfills; (3) the C storage factor of the
18 landfilled yard trimmings and food scraps; and (4) the rate of decomposition of the degradable C. The composition
19 of yard trimmings was assumed to be 30 percent grass clippings, 40 percent leaves, and 30 percent branches on a
20 wet weight basis (Oshins and Block 2000). The yard trimmings were subdivided, because each component has its
21 own unique adjusted C storage factor (i.e., moisture content and C content) and rate of decomposition. The mass of
22 yard trimmings and food scraps disposed of in landfills was estimated by multiplying the quantity of yard trimmings
23 and food scraps discarded by the proportion of discards managed in landfills. Data on discards (i.e., the amount
24 generated minus the amount diverted to centralized composting facilities) for both yard trimmings and food scraps
25 were taken primarily from Advancing Sustainable Materials Management: Facts and Figures 2013 (EPA 2015a),
26 which provides data for 1960, 1970, 1980, 1990, 2000, 2005, 2009 and 2011 through 2013. To provide data for
27 some of the missing years, detailed backup data were obtained from historical data tables that EPA developed for
28 1960 through 2013 (EPA 2015b). Remaining years in the time series for which data were not provided were
29 estimated using linear interpolation. Data for 2014 are not yet available, so they were set equal to 2013 values. The
30 EPA (2015a) report and historical data tables (EPA 2015b) do not subdivide the discards (i.e., total generated minus
31 composted) of individual materials into masses landfilled and combusted, although it provides a mass of overall
Land Use, Land-Use Change, and Forestry 6-75
-------
1 waste stream discards managed in landfills53 and combustors with energy recovery (i.e., ranging from 67 percent
2 and 33 percent, respectively, in 1960 to 92 percent and 8 percent, respectively, in 1985); it is assumed that the
3 proportion of each individual material (food scraps, grass, leaves, branches) that is landfilled is the same as the
4 proportion across the overall waste stream.
5 The amount of C disposed of in landfills each year, starting in 1960, was estimated by converting the discarded
6 landfilled yard trimmings and food scraps from a wet weight to a dry weight basis, and then multiplying by the
7 initial (i.e., pre-decomposition) C content (as a fraction of dry weight). The dry weight of landfilled material was
8 calculated using dry weight to wet weight ratios (Tchobanoglous et al. 1993, cited by Barlaz 1998) and the initial C
9 contents and the C storage factors were determined by Barlaz (1998, 2005, 2008) (Table 6-50).
10 The amount of C remaining in the landfill for each subsequent year was tracked based on a simple model of C fate.
11 As demonstrated by Barlaz (1998, 2005, 2008), a portion of the initial C resists decomposition and is essentially
12 persistent in the landfill environment. Barlaz (1998, 2005, 2008) conducted a series of experiments designed to
13 measure biodegradation of yard trimmings, food scraps, and other materials, in conditions designed to promote
14 decomposition (i.e., by providing ample moisture and nutrients). After measuring the initial C content, the materials
15 were placed in sealed containers along with methanogenie microbes from a landfill. Once decomposition was
16 complete, the yard trimmings and food scraps were re-analyzed for C content; the C remaining in the solid sample
17 can be expressed as a proportion of the initial C (shown in the row labeled "C Storage Factor, Proportion of Initial C
18 Stored (Percent)" in Table 6-50).
19 The modeling approach applied to simulate U.S. landfill C flows builds on the findings of Barlaz (1998, 2005,
20 2008). The proportion of C stored is assumed to persist in landfills. The remaining portion is assumed to degrade
21 over time, resulting in emissions of CH4 and CC>2. (The CH4 emissions resulting from decomposition of yard
22 trimmings and food scraps are accounted for in the Waste chapter.) The degradable portion of the C is assumed to
23 decay according to first-order kinetics. The decay rates for each of the materials are shown in Table 6-50.
24 The first-order decay rates, k, for each refuse type were derived from De la Cruz and Barlaz (2010). De la Cruz and
25 Barlaz (2010) calculate first-order decay rates using laboratory data published inEleazer et al. (1997), and a
26 correction factor,/ is found so that the weighted average decay rate for all components is equal to the EPA AP-42
27 default decay rate (0.04) for mixed MSW for regions that receive more than 25 inches of rain annually (EPA 1995).
28 Because AP-42 values were developed using landfill data from approximately 1990, 1990 waste composition for the
29 United States from EPA's Characterization of Municipal Solid Waste in the United States: 1990 Update was used to
30 calculate/ This correction factor is then multiplied by the Eleazer et al. (1997) decay rates of each waste component
31 to develop field-scale first-order decay rates.
32 De la Cruz and Barlaz (2010) also use other assumed initial decay rates for mixed MSW in place of the AP-42
33 default value based on different types of environments in which landfills in the United States are found, including
34 dry conditions (less than 25 inches of rain annually, &=0.02) and bioreactor landfill conditions (moisture is
35 controlled for rapid decomposition, &=0.12). As in the Landfills section of the Inventory (which estimates CEU
36 emissions), the overall MSW decay rate is estimated by partitioning the U.S. landfill population into three
37 categories, based on annual precipitation ranges of: (1) Less than 20 inches of rain per year, (2) 20 to 40 inches of
38 rain per year, and (3) greater than 40 inches of rain per year. These correspond to overall MSW decay rates of
39 0.020, 0.038, and 0.057 year1, respectively.
40 De la Cruz and Barlaz (2010) calculate component-specific decay rates corresponding to the first value (0.020
41 year1), but not for the other two overall MSW decay rates. To maintain consistency between landfill methodologies
42 across the Inventory, the correction factors (/) were developed for decay rates of 0.038 and 0.057 year1 through
43 linear interpolation. A weighted national average component-specific decay rate was calculated by assuming that
44 waste generation is proportional to population (the same assumption used in the landfill methane emission estimate),
45 based on population data from the 2000 U.S. Census. The component-specific decay rates are shown in Table 6-50.
53 EPA (2015a and 2015b) reports discards in two categories: "combustion with energy recovery" and "landfill, other disposal,"
which includes combustion without energy recovery. For years in which there is data from previous EPA reports on combustion
without energy recovery, EPA assumes these estimates are still applicable. For 2000 to present, EPA assumes that any
combustion of MSW that occurs includes energy recovery, so all discards to "landfill, other disposal" are assumed to go to
landfills.
6-76 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 For each of the four materials (grass, leaves, branches, food scraps), the stock of C in landfills for any given year is
2 calculated according to Equation 1:
3
4 LFCi,t= I Wi,n x (1 - MG) x ICQ*. {[CSiX. ICG\ + [(1 - (C&x ICG)) x e-^-")]}
5
6 where,
7 t = Year for which C stocks are being estimated (year),
8 /' = Waste type for which C stocks are being estimated (grass, leaves, branches, food scraps),
9 LFd,t = Stock of C in landfills in year /, for waste /' (metric tons),
10 Wi:n = Mass of waste /' disposed of in landfills in year n (metric tons, wet weight),
11 n Year in which the waste was disposed of (year, where 1960
-------
1 Table 6-51: C Stocks in Yard Trimmings and Food Scraps in Landfills (MMT C)
20
23
Carbon Pool
Yard Trimmings
Branches
Leaves
Grass
Food Scraps
Total Carbon Stocks
1990
155.8
14.5 1
66.7 1
74.6
17.6
173.5
2005
202.9
18.1
87.3
97.5
32.8
235.6
2010
213.6
19.0
92.2
102.3
38.0
251.6
2011
216.1
19.3
93.4
103.4
38.9
255.0
2012
218.6
19.5
94.5
104.5
39.8
258.4
2013
220.9
19.7
95.6
105.6
40.7
261.5
2014
223.2
20.0
96.6
106.6
41.5
264.7
Uncertainty and Time-Series Consistency
3 The uncertainty analysis for landfilled yard trimmings and food scraps includes an evaluation of the effects of
4 uncertainty for the following data and factors: disposal in landfills per year (tons of C), initial C content, moisture
5 content, decay rate, and proportion of C stored. The C storage landfill estimates are also a function of the
6 composition of the yard trimmings (i.e., the proportions of grass, leaves and branches in the yard trimmings
7 mixture). There are respective uncertainties associated with each of these factors.
8 A Monte Carlo (Approach 2) uncertainty analysis was applied to estimate the overall uncertainty of the
9 sequestration estimate. The results of the Approach 2 quantitative uncertainty analysis are summarized in Table
10 6-52. Total yard trimmings and food scraps CC>2 flux in 2014 was estimated to be between -18.0 and -4.5 MMT
11 CO2 Eq. at a 95 percent confidence level (or 19 of 20 Monte Carlo stochastic simulations). This indicates a range of
12 44 percent below to 64 percent above the 2014 flux estimate of -11.6 MMT CCh Eq. More information on the
13 uncertainty estimates for Yard Trimmings and Food Scraps in Landfills is contained within the Uncertainty Annex.
14 Table 6-52: Approach 2 Quantitative Uncertainty Estimates for COz Flux from Yard
15 Trimmings and Food Scraps in Landfills (MMT COz Eq. and Percent)
2014 Flux
Estimate
Source Gas (MMT CCh Eq.)
Yard Trimmings and Food __ ., , -.
Scraps C°2 (1L6)
Uncertainty Range Relative to Flux Estimate3
(MMT C02 Eq.) (%)
Lower
Bound
(18.0)
Upper
Bound
(4.5)
Lower
Bound
-44%
Upper
Bound
+64%
a Range of flux estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
Note: Parentheses indicate negative values or net C sequestration.
16
17 Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
18 through 2014. Details on the emission trends through time are described in more detail in the Methodology section,
19 above.
QA/QC and Verification
21 A QA/QC analysis was performed for data gathering and input, documentation, and calculation. The QA/QC
22 analysis did not reveal any inaccuracies or incorrect input values.
Recalculations Discussion
24 The current Inventory has been revised relative to the previous report. Generation and recovery data for yard
25 trimmings and food scraps was not previously provided for every year from 1960 in the Advancing Sustainable
26 Materials Management: Facts and Figures 2013 report. EPA has since released historical data, which included data
27 for each year from 1960 through 2013. The recalculations based on these historical data resulted in changes ranging
6-78 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 from a 1 percent increase in sequestration in 2001 to a 7 percent decrease in sequestration in 2013, and an average
2 0.66 percent decrease in sequestration across the 1990 through 2013 time series compared to the previous Inventory.
Planned Improvements
4 Future work is planned to evaluate the consistency between the estimates of C storage described in this chapter and
5 the estimates of landfill CH4 emissions described in the Waste chapter. For example, the Waste chapter does not
6 distinguish landfill CH4 emissions from yard trimmings and food scraps separately from landfill CH4 emissions from
7 total bulk (i.e., municipal solid) waste, which includes yard trimmings and food scraps.
8 In addition, additional data will be evaluated from recent peer-reviewed literature that may modify the default C
9 storage factors, initial C contents, and decay rates for yard trimmings and food scraps in landfills. Based upon this
10 evaluation, changes may be made to the default values. Whether to update the weighted national average
11 component-specific decay rate using new U.S. Census data, if any are available, will also be investigated.
12 The yard waste composition will also be evaluated to determine if changes need to be made based on changes in
13 residential practices, research will be conducted to determine if there are changes in the allocation of yard
14 trimmings. For example, leaving grass clippings in place is becoming a more common practice, thus reducing the
15 percentage of grass clippings in yard trimmings disposed in landfills.
Land Use, Land-Use Change, and Forestry 6-79
-------
-------
7. Waste
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Waste management and treatment activities are sources of greenhouse gas emissions (see Figure 7-1). Landfills
accounted for approximately 25.7 percent of total U.S. anthropogenic methane (CH4) emissions in 2014, the largest
contribution of any CH4 source in the United States. Additionally, wastewater treatment and composting of organic
waste accounted for approximately 2.1 percent and less than 1 percent of U.S. CH4 emissions, respectively. Nitrous
oxide (N2O) emissions from the discharge of wastewater treatment effluents into aquatic environments were
estimated, as were N2O emissions from the treatment process itself. N2O emissions from composting were also
estimated. Together, these waste activities account for 1.6 percent of total U.S. N2O emissions. Nitrogen oxides
(NOX), carbon monoxide (CO), and non-CH4 volatile organic compounds (NMVOCs) are emitted by waste
activities, and are addressed separately at the end of this chapter. A summary of greenhouse gas emissions from the
Waste chapter is presented in Table 7-1 and Table 7-2.
Figure 7-1: 2014 Waste Chapter Greenhouse Gas Sources (MMT COz Eq.)
Landfills
Wastewater Treatment
Composting
125
Waste as a Portion of
all Emissions
50 75
MMT CO2 Eq.
100
Box 7-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Sinks
In following the UNFCCC requirement under Article 4.1 to develop and submit national greenhouse gas emission
inventories, the emissions and sinks presented in this report and this chapter, are organized by source and sink
categories and calculated using internationally-accepted methods provided by the Intergovernmental Panel on
Climate Change (IPCC 2006).l Additionally, the calculated emissions and sinks in a given year for the United
States are presented in a common manner in line with the UNFCCC reporting guidelines for the reporting of
inventories under this international agreement.2 The use of consistent methods to calculate emissions and sinks by
1 See .
2See.
Waste 7-1
-------
1
2
o
6
4
5
6
7
8
9
10
11
12
13
14
15
16
17
all nations providing their inventories to the UNFCCC ensures that these reports are comparable. In this regard, U.S.
emissions and sinks reported in this Inventory report are comparable to emissions and sinks reported by other
countries. The manner that emissions and sinks are provided in this Inventory is one of many ways U.S. emissions
and sinks could be examined; this Inventory report presents emissions and sinks in a common format consistent with
how countries are to report inventories under the UNFCCC. Emissions and sinks provided in the current Inventory
do not preclude alternative examinations,3 but rather presents emissions and sinks in a common format consistent
with how countries are to report inventories under the UNFCCC. The report itself, and this chapter, follows this
standardized format, and provides an explanation of the IPCC methods used to calculate emissions and sinks, and
the manner in which those calculations are conducted. The UNFCCC incorporated the 2006 IPCC Guidelines for
National Greenhouse Gas Inventories as the standard for Annex I countries at the Nineteenth Conference of the
Parties (Warsaw, November 11-23, 2013). This chapter presents emission estimates calculated in accordance with
the methodological guidance provided in these guidelines.
Overall, in 2014, waste activities generated emissions of 205.6 MMT CC>2 Eq., or just under 3 percent of total U.S.
greenhouse gas emissions.
Table 7-1: Emissions from Waste (MMT COz Eq.)
Gas/Source
CH4
Landfills
Wastewater Treatment
Composting
N20
Wastewater Treatment
Composting
Total
Note: Totals may not sum due to
Table 7-2: Emissions from
Gas/Source
CH4
Landfills
Wastewater Treatment
Composting
N2O
Wastewater Treatment
Composting
1990
200.4
184.4 1
15.7
0.4 1
3.7
3.4
0.3
204.1
2005
205.1
187.3
15.9 1
1.9 1
6.0
4.3
1.7
211.1
2010
193.6
176.3
15.5
1.8
6.4
4.7
1.6
200.0
2011
194.0
176.9
15.3
1.9
6.5
4.8
1.7
200.5
2012
190.6
173.5
15.2
1.9
6.6
4.9
1.7
197.2
2013
193.7
176.7
15.0
2.0
6.7
4.9
1.8
200.5
2014
198.9
181.8
15.0
2.1
6.8
4.9
1.8
205.6
independent rounding.
Waste (kt)
1990
8,017
7,376 I
626
ii
1 •
2005
8,203
7,493 1
635
ii
6
2010
7,744
7,052
619
73
21
16
5
2011
7,759
7,074
610
75
22
16
6
2012
7,625
6,942
606
77
22
16
6
2013
7,749
7,066
601
81
23
17
6
2014
7,954
7,271
601
82
23
17
6
Note: Totals may not sum due to independent rounding.
18 Carbon dioxide, CH4, and N2O emissions from the incineration of waste are accounted for in the Energy sector
19 rather than in the Waste sector because almost all incineration of municipal solid waste (MSW) in the United States
20 occurs at waste-to-energy facilities where useful energy is recovered. Similarly, the Energy sector also includes an
21 estimate of emissions from burning waste tires and hazardous industrial waste, because virtually all of the
22 combustion occurs in industrial and utility boilers that recover energy. The incineration of waste in the United States
23 in 2014 resulted in 9.7 MMT CO2 Eq. emissions, more than half of which is attributable to the combustion of
24 plastics. For more details on emissions from the incineration of waste, see Section 7.4.
1 For example, see < http://www.epa.gov/ghgreporting/ghgrp-methodology-and-verification >.
7-2 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Box 7-2: Waste Data from the Greenhouse Gas Reporting Program
On October 30, 2009, the U.S. EPA published a rule for the mandatory reporting of greenhouse gases from large
greenhouse gas emissions sources in the United States. Implementation of 40 CFR Part 98 is referred to as EPA's
Greenhouse Gas Reporting Program (GHGRP). 40 CFR part 98 applies to direct greenhouse gas emitters, fossil
fuel suppliers, industrial gas suppliers, and facilities that inject CO2 underground for sequestration or other
reasons and requires reporting by 41 industrial categories. Reporting is at the facility level, except for certain
suppliers of fossil fuels and industrial greenhouse gases. In general, the threshold for reporting is 25,000 metric
tons or more of CO2 Eq. per year.
EPA's GHGRP dataset and the data presented in this Inventory report are complementary and, as indicated in the
respective planned improvements sections for source categories in this chapter, EPA is analyzing how to use
facility-level GHGRP data to improve the national estimates presented in this Inventory. Most methodologies
used in EPA's GHGRP are consistent with IPCC, though for EPA's GHGRP, facilities collect detailed
information specific to their operations according to detailed measurement standards. This may differ with the
more aggregated data collected for the Inventory to estimate total, national U.S. emissions. It should be noted that
the definitions for source categories in the GHGRP may differ from those used in this Inventory in meeting the
UNFCCC reporting guidelines. In line with the UNFCCC reporting guidelines, the Inventory report is a
comprehensive accounting of all emissions from source categories identified in the IPCC guidelines. Further
information on the reporting categorizations in EPA's GHGRP and specific data caveats associated with
monitoring methods in EPA's GHGRP has been provided on the EPA's GHGRP website.4
EPA presents the data collected by EPA's GHGRP through a data publication tool5 that allows data to be viewed
in several formats including maps, tables, charts and graphs for individual facilities or groups of facilities.
2 7.1 Landfills (IPCC Source Category 5A1)
3 In the United States, solid waste is managed by landfilling, recovery through recycling or composting, and
4 combustion through waste-to-energy facilities. Disposing of solid waste in modern, managed landfills is the most
5 commonly used waste management technique in the United States. More information on how solid waste data are
6 collected and solid waste is managed in the United States is provided in Box 7-1 and Box 7-2. The municipal solid
7 waste (MSW) and industrial waste landfills referred to in this section are all modern landfills that must comply with
8 a variety of regulations as discussed in Box 7-3. Disposing of waste in illegal dumping sites is not considered to
9 have occurred in years later than 1980 and these sites are not considered to contribute to net emissions in this section
10 for the time frame of 1990 to the current inventory year. MSW landfills, or sanitary landfills, are sites where MSW
11 is managed to prevent or minimize health, safety, and environmental impacts. Waste is deposited in different cells
12 and covered daily with soil; many have environmental monitoring systems to track performance, collect leachate,
13 and collect landfill gas. Industrial waste landfills are constructed in a similar way as MSW landfills, but accept
14 waste produced by industrial activity, such as factories, mills, and mines.
15 After being placed in a landfill, organic waste (such as paper, food scraps, and yard trimmings) is initially
16 decomposed by aerobic bacteria. After the oxygen has been depleted, the remaining waste is available for
17 consumption by anaerobic bacteria, which break down organic matter into substances such as cellulose, amino acids,
18 and sugars. These substances are further broken down through fermentation into gases and short-chain organic
19 compounds that form the substrates for the growth of methanogenic bacteria. These methane (CH4) producing
20 anaerobic bacteria convert the fermentation products into stabilized organic materials and biogas consisting of
21 approximately 50 percent biogenic carbon dioxide (CO2) and 50 percent CH4, by volume. Landfill biogas also
4 See
.
5 See .
Waste T
-------
1 contains trace amounts of non-methane organic compounds (NMOC) and volatile organic compounds (VOC) that
2 either result from decomposition by-products or volatilization of biodegradable wastes (EPA 2008).
3 Methane and CO2 are the primary constituents of landfill gas generation and emissions. However, the 2006
4 Intergovernmental Panel on Climate Change (IPCC) Guidelines set an international convention to not report
5 biogenic CO2 released due to landfill decomposition in the Waste sector (IPCC 2006). Carbon dioxide emissions
6 from landfills are estimated and reported under the Land Use/Land Use Change and Forestry (LULUCF) sector (see
7 Box 7-4). Additionally, emissions of NMOC and VOC are not estimated because they are considered to be emitted
8 in trace amounts. Nitrous oxide (N2O) emissions from the disposal and application of sewage sludge on landfills are
9 also not explicitly modeled as part of greenhouse gas emissions from landfills. N2O emissions from sewage sludge
10 applied to landfills as a daily cover or for disposal are expected to be relatively small because the microbial
11 environment in an anaerobic landfill is not very conducive to the nitrification and denitrification processes that result
12 in N2O emissions. Furthermore, the 2006 IPCC Guidelines (IPCC 2006) did not include a methodology for
13 estimating N2O emissions from solid waste disposal sites "because they are not significant." Therefore, only CH4
14 generation and emissions are estimated for landfills under the Waste sector.
15 Methane generation and emissions from landfills are a function of several factors, including: (1) the total amount of
16 waste-in-place, which is the total waste landfilled annually over the operational lifetime of a landfill; (2) the
17 characteristics of the landfill receiving waste (e.g., composition of waste-in-place, size, climate, cover material); (3)
18 the amount of CH4 that is recovered and either flared or used for energy purposes; and (4) the amount of CH4
19 oxidized as the landfill gas passes through the cover material into the atmosphere. Each landfill has unique
20 characteristics, but all managed landfills practice similar operating practices, including the application of a daily and
21 intermediate cover material over the waste being disposed of in the landfill to prevent odor and reduce risks to
22 public health. Based on recent literature, the specific type of cover material used can affect the rate of oxidation of
23 landfill gas (RTI2011). The most commonly used cover materials are soil, clay, and sand. Some states also permit
24 the use of green waste, tarps, waste derived materials, sewage sludge or biosolids, and contaminated soil as a daily
25 cover. Methane production typically begins within the first year after the waste is disposed of in a landfill and will
26 continue for 10 to 60 years or longer as the degradable waste decomposes over time.
27 In 2014, landfill CH4 emissions were approximately 181.8 MMT CO2 Eq. (7,271 kt), representing the largest source
28 of CH4 emissions in the United States, followed by enteric fermentation and natural gas systems. Emissions from
29 MSW landfills accounted for approximately 95 percent of total landfill emissions, while industrial landfills
30 accounted for the remainder. Approximately 1,900 to 2,000 operational MSW landfills exist in the United States,
31 with the largest landfills receiving most of the waste and generating the majority of the CH4 emitted (EPA 2015b;
32 EPA 2015d). Conversely, there are approximately 3,200 MSW landfills in the United States that have been closed
33 since 1980 (for which a closure data is known, EPA 2015b; WBJ 2010). While the number of active MSW landfills
34 has decreased significantly over the past 20 years, from approximately 6,326 in 1990 to approximately 2,000 in the
35 2010s, the average landfill size has increased (EPA 2015c; EPA 2015d; BioCycle 2010; WBJ 2010). The exact
36 number of active and closed dedicated industrial waste landfills is not known at this time, but the Waste Business
37 Journal total for landfills accepting industrial and construction and demolition debris for 2010 is 1,305 (WBJ 2010).
38 Only 176 facilities with industrial waste landfills reported under subpart TT (Industrial Waste Landfills) of EPA's
39 Greenhouse Gas Reporting Program (GHGRP) since reporting began in 2011, indicating that there may be several
40 hundreds of industrial waste landfills that are not required to report under EPA's GHGRP, or that the actual number
41 of industrial waste landfills in the United States is relatively low compared to MSW landfills.
42 The estimated annual quantity of waste placed in MSW landfills increased 27 percent from approximately 234
43 MMT in 1990 to 299 MMT in 2000 and then decreased by 11 percent to 266 MMT in 2014 (see Annex 3.14). The
44 annual amount of waste generated and subsequently disposed in MSW landfills varies annually and depends on
45 several factors (e.g., the economy, consumer patterns, recycling and composting programs, inclusion in a garbage
46 collection service). The total amount of MSW generated is expected to increase as the U.S. population continues to
47 grow, but the percentage of waste landfilled may decline due to increased recycling and composting practices. The
48 estimated quantity of waste placed in industrial waste landfills (from the pulp and paper, and food processing
49 sectors) has remained relatively steady since 1990, ranging from 9.7 MMT in 1990 to 11.3 MMT in 2014.
50 Net CH4 emissions have fluctuated over the time-series, with peak emissions in the late 2000's and a slowly
51 decreasing trend since. For example, from 1990 to 2000, net CH4 emissions from landfills decreased by
52 approximately 4.6 percent, from 184.4 MMT to 176.0 MMT, but then increased by 8.7 percent from 2000 to 2009.
53 From 2009 to 2014, however, net CH4 emissions decreased by nearly 5 percent, from 191.3 MMT to 181.8 MMT
7-4 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 (see Table 7-3). This decreasing trend can be mostly attributed to an approximately 21 percent reduction in the
2 amount of decomposable materials (i.e., paper and paperboard, food scraps, and yard trimmings) discarded in MSW
3 landfills over the time-series (EPA 2015d) and an increase in the amount of landfill gas collected and combusted
4 (i.e., used for energy or flared) at MSW landfills, resulting in lower net CH4 emissions from MSW landfills. For
5 instance, in 1990, approximately 0.8 MMT of CH4 were recovered and combusted from landfills, while in 2014,
6 approximately 7.9 MMT of CH4 were recovered and combusted, representing an average annual increase in the
7 quantity of CH4 recovered and combusted at MSW landfills from 1990 to 2014 of 11 percent (see Annex 3.14).
8 Landfill gas collection and control is not accounted for at industrial waste landfills in this chapter (see the
9 Methodology discussion for more information).
10 The quantity of recovered CH4 that is either flared or used for energy purposes at MSW landfills has continually
11 increased as a result of 1996 federal regulations that require large MSW landfills to collect and combust landfill gas
12 (see 40 CFR Part 60, Subpart Cc 2005 and 40 CFR Part 60, Subpart WWW 2005). Voluntary programs that
13 encourage CH4 recovery and beneficial reuse, such as EPA's Landfill Methane Outreach Program (LMOP) and
14 federal and state incentives that promote renewable energy (e.g., tax credits, low interest loans, and Renewable
15 Portfolio Standards), have also contributed to increased interest in landfill gas collection and control. In 2014, an
16 estimated 10 new landfill gas-to-energy (LFGTE) projects (EPA 2015a; EPA 2015b) and 3 new flares began
17 operation. While the amount of landfill gas collected and combusted continues to increase every year, the rate of
18 increase in collection and combustion no longer exceeds the rate of additional CH4 generation from the amount of
19 organic MSW landfilled as the U.S. population grows.
20 Table 7-3: CH4 Emissions from Landfills (MMT COz Eq.)
23
Activity
MSW Landfills
Industrial Landfills
Recovered
Gas-to-Energy
Flared
Oxidized*
Total
1990
212.7 1
12.1
(7.1)
(12.8)
(20.5) •
184.4
2005
329.6
15.9
(86.8) 1
(50.6) 1
(20.8) •
187.3
2010
366.1
16.4
(125.9)
(60.7)
(19.6)
176.3
2011
370.7
16.4
(129.3)
(61.3)
(19.7)
176.9
2012
375.0
16.5
(135.7)
(62.9)
(19.3)
173.5
2013
379.1
16.5
(136.0)
(63.4)
(19.6)
176.7
2014
383.5
16.6
(134.7)
(63.4)
(20.2)
181.8
Note: Lotals may not sum due to independent rounding. Parentheses indicate negative values.
a Includes oxidation at municipal and industrial landfills.
21
22 Table 7-4: CH4 Emissions from Landfills (kt)
Activity
MSW Landfills
Industrial Landfills
Recovered
Gas-to-Energy
Flared
Oxidized*
Total
1990
8,508
484 1
(285)
(512)
(820)
7,376
2005
13,185
636
(3,471)
(2,024)
1 (833)
7,493
r2010
4,642
656
1,035)
U,428)
(784)
7,052
2011
14,828
657
(5,172)
(2,428)
(786)
7,074
2012
15,001
659
(5,430)
(2,517)
(771)
6,942
2013
15,163
661
(5,438)
(2,535)
(785)
7,066
2014
15,338
665
(5,387)
(2,537)
(808)
7,271
Note: Lotals may not sum due to independent rounding. Parentheses indicate negative values.
a Includes oxidation at municipal and industrial landfills.
Methodology
24 CH4 emissions from landfills were estimated as the CH4 produced from MSW landfills, plus the CH4 produced by
25 industrial waste landfills, minus the CH4 recovered and combusted from MSW landfills, minus the CH4 oxidized
26 before being released into the atmosphere:
27 CH4,Solid Waste = [CH4.MSW + CH4,Ind — R] — Ox
Waste 7-5
-------
1 where,
2 CH4, solid waste = CH4 emissions from solid waste
3 CH4,Msw = CH4 generation from MSW landfills,
4 CH4,ind = CH4 generation from industrial landfills,
5 R = CH4 recovered and combusted (only for MSW landfills), and
6 Ox CH4 oxidized from MSW and industrial waste landfills before release to the atmosphere.
7 The methodology for estimating CH4 emissions from landfills is based on the first order decay (FOD) model
8 described by the IPCC (IPCC 2006). Methane generation is based partly on nationwide waste disposal data (for
9 years prior to 1983) and partly on facility-specific waste acceptance data (for 1983 and later). The amount of CH4
10 recovered is landfill-specific for all years in the time-series, but only for MSW landfills due to a lack of data specific
11 to industrial waste landfills. Values for the CH4 generation potential (L0) and the decay rate constant (k) used in the
12 first order decay model were obtained from an analysis of CH4 recovery rates for a database of 52 landfills and from
13 published studies of other landfills (RTI 2004; EPA 1998; SWANA 1998; Peer, Thorneloe, and Epperson 1993).
14 The decay rate constant was found to increase with average annual rainfall; consequently, values of k were
15 developed for three ranges of rainfall, or climate types (wet, arid, and temperate). The annual quantity of waste
16 placed in landfills was apportioned to the three ranges of rainfall based on the percent of the U.S. population in each
17 of the three ranges. Historical census data were used to account for the shift in population to more arid areas over
18 time (U.S. Census Bureau, 2015). An overview of the data sources and methodology used to calculate CH4
19 generation and recovery is provided below, while a more detailed description of the methodology used to estimate
20 CH4 emissions from landfills can be found in Annex 3.14.
21 States and local municipalities across the United States do not consistently track and report quantities of MSW
22 generated or collected for management, nor are end-of-life disposal methods reported to a centralized system. The
23 GHGRP, however, requires landfills meeting or exceeding a threshold of 25,000 metric tons of CH4 generation per
24 year to report a variety of facility-specific information, including historical and current waste disposal quantities by
25 year, CH4 generation, gas collection system details, CH4 recovery, and CH4 emissions. The landfills reporting to the
26 GHGRP are considered the largest emitters, but not all landfills are required to report. However, when this dataset is
27 supplemented with others, such as the EPA LMOP data (incorporated into the Inventory through the Landfill Gas-
28 to-energy [LFGTE] database), or the Waste Business Journal data, a complete data set of the annual quantity of
29 waste landfilled is represented. A bottom-up approach has been taken with the current inventory whereby the
30 GHGRP data, supplemented with the LFGTE database, provides the annual waste disposal data needed for the FOD
31 model. In previous inventories, a top-down approach was used to estimate national MSW landfill waste generation
32 (i.e., the State of Garbage [SOG] surveys). The SOG survey is a nationwide survey of waste disposed in landfills
3 3 and relies on the principles of mass balance, where all MSW generated is equal to the amount of MSW landfilled,
34 combusted in waste-to-energy plants, composted, and/or recycled (BioCycle 2010; Shin 2014). The SOG survey data
35 was used to estimate nationwide MSW generation by state, and then a waste disposal factor was applied to estimate
36 the quantity of waste landfilled. Switching from the SOG survey data to the GHGRP and LFGTE data provides a
37 higher tier and quality of data, and improves the accuracy of landfill CH4 emissions in the U.S.
38 Estimates of the annual quantity of waste landfilled for 1960 through 1982 were obtained fromEPA's
39 Anthropogenic Methane Emissions in the United States, Estimates for 1990: Report to Congress (EPA 1993) and an
40 extensive landfill survey by the EPA's Office of Solid Waste in 1986 (EPA 1988). Although waste placed in
41 landfills in the 1940s and 1950s contributes very little to current CH4 generation, estimates for those years were
42 included in the FOD model for completeness in accounting for CH4 generation rates and are based on the population
43 in those years and the per capita rate for land disposal for the 1960s. For calculations in the current Inventory,
44 wastes landfilled prior to 1980 were broken into two groups: wastes disposed in landfills (Methane Conversion
45 Factor, MCF, of 1) and those disposed in dumps (MCF of 0.6). All calculations after 1980 assume waste is disposed
46 in managed, modern landfills. Please see Annex 3.14 for more details.
47 Methane recovery is currently only accounted for at MSW landfills. Data collected through EPA's GHGRP for
48 industrial waste landfills (subpart TT) show that only 2 of the 176 facilities, or 1 percent of facilities, reporting have
49 active gas collection systems (EPA 2015b). EPA's GHGRP is not a national database and no comprehensive data
50 regarding gas collection systems have been published for industrial waste landfills. Assumptions regarding a
51 percentage of landfill gas collection systems, or a total annual amount of landfill gas collected for the non-reporting
52 industrial waste landfills, have not been made for the Inventory methodology.
7-6 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 The estimated landfill gas recovered per year (R) at MS W landfills was based on a combination of four databases
2 and grouped into recovery from flares and recovery from landfill gas-to-energy projects:
3 • the flare vendor database (contains updated sales data collected from vendors of flaring equipment)
4 • a database of LFGTE projects that is primarily based on information compiled by EPA LMOP (EPA
5 2015a)
6 • a database developed by the Energy Information Administration (EIA) for the voluntary reporting of
7 greenhouse gases (EIA 2007), and
8 • EPA's GHGRP dataset for MSW landfills (EPA 2015b).
9 EPA's GHGRP MSW landfills database was first introduced as a data source for the previous Inventory. EPA's
10 GHGRP MSW landfills database contains facility-reported data that undergoes rigorous verification, thus it is
11 considered to contain the least uncertain data of the four databases. However, as mentioned earlier, this database is
12 unique in that it only contains a portion of the landfills in the U.S. (although, presumably the highest emitters since
13 only those landfills that meet a certain CH4 generation threshold must report) and only contains data for 2010 and
14 later. Methane generation can be estimated for prior years based on reported annual waste disposal. Methane
15 recovery prior to 2010 can also be estimated, but it is more uncertain than the CH4 generation because the GHGRP
16 does not require facilities to report the year the flare(s) or gas collection system(s) were installed and fully
17 operational.
18 The total amount of CH4 recovered and destroyed was estimated using the four databases listed above. To avoid
19 double- or triple-counting CH4 recovery, the landfills across each database were compared and duplicates identified.
20 A hierarchy of recovery data is used based on the certainty of the data in each database as described below.
21 Forthe years 2010 to 2014, if a landfill in EPA's GHGRP MSW landfills database was also in the EIA, LFGTE,
22 and/or flare vendor database, the avoided emissions were based on EPA's GHGRP MSW landfills database only to
23 avoid double or triple counting the recovery amounts. Directly reported values for CH4 recovery to the GHGRP
24 database were used for years 2010 through 2014. Because these data are not reported for years prior to 2010, CH4
25 recovery had to be calculated using available data (e.g., landfill gas project information) from LFGTE and the EIA
26 databases. Methane recovery from years 1990 to 2009 was calculated for landfills in EPA's GHGRP database using
27 the known values of CH4 recovery from the years 2010 to 2014. If a landfill was also in the LFGTE or EIA
28 databases, the landfill gas project information, specifically the project start year, was used as the cutoff years for the
29 estimated CH4 recovery in the EPA's GHGRP database. For example, if a landfill reporting under EPA's GHGRP
30 was also included in the LFGTE database under a project that started in 2002 and is still operational, the CH4
31 recovery in EPA's GHGRP database was back-calculated to the year 2002. This method, although somewhat
32 uncertain, can be refined in future Inventory years after further investigating the landfill gas project start years for
33 landfills in EPA's GHGRP database.
34 If a landfill in the EIA database was also in the LFGTE and/or the flare vendor database, the CH4 recovery was
35 based on the EIA data because landfill owners or operators directly reported the amount of CH4 recovered using gas
36 flow concentration and measurements, and because the reporting accounted for changes over time. However, as the
37 EIA database only includes data through 2006, the amount of CH4 recovered for years 2007 and later were assumed
38 to be the same as in 2006 for landfills that are in the EIA database, but not in EPA's GHGRP or the LFGTE
39 databases. This quantity likely underestimates flaring because the EIA database does not have information on all
40 flares in operation for the years after 2006. If both the flare data and LFGTE recovery data were available for any of
41 the remaining landfills (i.e., not in the EIA or EPA's GHGRP databases), then the avoided emissions were based on
42 the LFGTE data, which provides reported landfill-specific data on gas flow for direct use projects and project
43 capacity (i.e., megawatts) for electricity projects. The flare vendor database, on the other hand, estimates CH4
44 combusted by flares using the midpoint of a flare's reported capacity.
45 Given that each LFGTE project is likely to also have a flare, double counting reductions from flares and LFGTE
46 projects in the LFGTE database was avoided by subtracting emission reductions associated with LFGTE projects for
47 which a flare had not been identified from the emission reductions associated with flares (referred to as the flare
48 correction factor). A further explanation of the methodology used to estimate the landfill gas recovered can be found
49 in Annex 3.14.
50 The amount of landfill gas recovered and combusted is also presented in terms of avoided emissions by flaring and
51 avoided emissions by LFGTE. The amount combusted by flaring was directly determined using information
52 provided by the EIA and flare vendor databases and indirectly determined using information in EPA's GHGRP
Waste 7-7
-------
1 dataset for MSW landfills. Information provided by the EIA and LFGTE databases were used to directly estimate
2 methane combusted in LFGTE projects over the time-series. EPA's GHGRP MSW landfills database provides a
3 total amount of CH4 recovered at the facility-level and was indirectly used to estimate methane combusted in
4 LFGTE projects. Unlike the three other databases, EPA's GHGRP dataset does not identify whether the amount of
5 CH4 recovered is combusted by a flare versus an LFGTE project. Therefore, a mapping exercise was performed
6 between EPA's GHGRP MSW landfills database and the three other databases to make a distinction between
7 landfills contained in both EPA's GHGRP MSW landfills database and one or more of the other databases. The CH4
8 recovered by landfills matched to the EIA (and marked as LFGTE) and LFGTE databases was allocated as CH4
9 recovered and combusted by LFGTE projects. The remaining CH4 recovered from EPA's GHGRP dataset was
10 allocated as CH4 recovered and combusted by flares.
11 The destruction efficiencies reported through EPA's GHGRP were applied to the landfills in EPA's GHGRP MSW
12 landfills database. The median value of the reported destruction efficiencies was 99 percent for all reporting years
13 (2010 through 2013). A destruction efficiency of 99 percent was applied to CH4 recovered to estimate CH4
14 emissions avoided due to the combusting of CH4 in destruction devices (i.e., flares) in the EIA, LFGTE, and flare
15 vendor databases. The 99 percent destruction efficiency value was selected based on the range of efficiencies (86 to
16 99+ percent) recommended for flares in EPA's AP-42 Compilation of Air Pollutant Emission Factors, Draft Chapter
17 2.4, Table 2.4-3 (EPA 2008). A typical value of 97.7 percent was presented for the non-CH4 components (i.e.,
18 volatile organic compounds and non-methane organic compounds) in test results (EPA 2008). An arithmetic
19 average of 98.3 percent and a median value of 99 percent are derived from the test results presented in EPA (2008).
20 Thus, a value of 99 percent for the destruction efficiency of flares has been used in Inventory methodology. Other
21 data sources supporting a 99 percent destruction efficiency include those used to establish New Source Performance
22 Standards (NSPS) for landfills and in recommendations for shutdown flares used by the EPA LMOP.
23 Emissions from industrial waste landfills were estimated from industrial production data (ERG 2014), waste
24 disposal factors, and the FOD model. As over 99 percent of the organic waste placed in industrial waste landfills
25 originated from the food processing (meat, vegetables, fruits) and pulp and paper sectors, estimates of industrial
26 landfill emissions focused on these two sectors (EPA 1993). There are currently no data sources that track and report
27 the amount and type of waste disposed of in industrial waste landfills in the United States. Therefore, the amount of
28 waste landfilled is assumed to be a fraction of production that is held constant over the time-series as explained in
29 Annex 3.14. The composition of waste disposed of in industrial waste landfills is expected to be more consistent in
30 terms of composition and quantity than that disposed of in MSW landfills.
31 The amount of CH4 oxidized by the landfill cover at both municipal and industrial waste landfills was assumed to be
32 10 percent of the CH4 generated that is not recovered (IPCC 2006, Mancinelli and McKay 1985, Czepiel et al.
33 1996). To calculate net CH4 emissions, both CH4 recovered and CH4 oxidized were subtracted from CH4 generated
34 at municipal and industrial waste landfills.
35
Uncertainty and Time-Series Consistency
36 Several types of uncertainty are associated with the estimates of CH4 emissions from MSW and industrial waste
37 landfills. The primary uncertainty concerns the characterization of landfills. Information is not available on two
38 fundamental factors affecting CH4 production: the amount and composition of waste placed in every MSW and
39 industrial waste landfill for each year of a landfill's operation. EPA's GHGRP allows facilities to report annual
40 quantities of waste disposed by composition, but most MSW landfills report annual waste disposed as bulk MSW
41 versus the detailed waste composition data. Some MSW landfills have conducted detailed waste composition
42 studies, but the data are scarce over the time-series and across the country. EPA is currently compiling the waste
43 composition studies and data that have been performed in the past decade and may revise the default waste
44 composition applied to MSW landfilled in the FOD model in future inventory estimates.
45 The approach used here assumes that the CH4 generation potential (L0) and the rate of decay that produces CH4 from
46 MSW, as determined from several studies of CH4 recovery at MSW landfills, are representative of conditions at U. S.
47 MSW landfills. When this top-down approach is applied at the nationwide level, the uncertainties are assumed to be
48 less than when applying this approach to individual landfills and then aggregating the results to the national level. In
49 other words, this approach may over- and under-estimate CH4 generation at some landfills if used at the facility -
50 level, but the end result is expected to balance out because it is being applied nationwide. There is also a high degree
7-8 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 of uncertainty and variability associated with the first order decay model, particularly when a homogeneous waste
2 composition and hypothetical decomposition rates are applied to heterogeneous landfills (IPCC 2006).
3 Additionally, there is a lack of landfill-specific information regarding the number and type of industrial waste
4 landfills in the United States. The approach used here assumes that the majority (99 percent) of industrial waste
5 disposed of in industrial waste landfills consists of waste from the pulp and paper and food processing sectors.
6 However, because waste generation and disposal data are not available in an existing data source for all U.S.
7 industrial waste landfills, we apply a straight disposal factor over the entire time-series to the amount of waste
8 generated to determine the amounts disposed. Industrial waste facilities reporting under EPA's GHGRP do report
9 detailed waste stream information, and these data have been used to improve, for example, the DOC value used in
10 the inventory methodology for the pulp and paper sector. Additional improvements to reduce the uncertainty in CH4
11 emissions estimates are being investigated for both MSW and industrial waste landfills.
12 Aside from the uncertainty in estimating landfill CH4 generation, uncertainty also exists in the estimates of the
13 landfill gas oxidized. A constant oxidation factor of 10 percent as recommended by the Intergovernmental Panel on
14 Climate Change (IPCC) for managed landfills is used for both MSW and industrial waste landfills regardless of
15 climate, the type of cover material, and/or presence of a gas collection system. The number of field studies
16 measuring the rate of oxidation has increased substantially since the 2006 IPCC Guidelines were published and, as
17 discussed in the Potential Improvements section, efforts are being made to review the literature and revise this value
18 based on recent, peer-reviewed studies.
19 Another significant source of uncertainty lies with the estimates of CH4 that are recovered by flaring and gas-to-
20 energy projects at MSW landfills. The EPA's GHGRP MSW landfills database was added as a fourth recovery
21 database in the previous Inventory. Relying on multiple databases for a complete picture introduces uncertainty
22 because the coverage and characteristics of each database differs, which increases the chance of double counting
23 avoided emissions. Additionally, the methodology and assumptions that go into each database differ. For example,
24 the flare database assumes the midpoint of each flare capacity at the time it is sold and installed at a landfill; in
25 reality, the flare may be achieving a higher capacity, in which case the flare database would underestimate the
26 amount of CH4 recovered.
27 The LFGTE database and the flare vendor databases are updated annually. The EIA database has not been updated
28 since 2005 and has, for the most part, been replaced by EPA's GHGRP MSW landfills database. To avoid double
29 counting and to use the most relevant estimate of CH4 recovery for a given landfill, a hierarchical approach is used
30 among the four databases. EPA's GHGRP data are given precedence because CH4 recovery is directly reported by
31 landfills and undergoes a rigorous verification process; the EIA data are given second priority because facility data
32 were directly reported; the LFGTE data are given third priority because CH4 recovery is estimated from facility -
33 reported LFGTE system characteristics; and the flare data are given fourth priority because this database contains
34 minimal information about the flare and no site-specific operating characteristics (Bronstein et al. 2012). The
35 coverage provided across the databases most likely represents the complete universe of landfill CH4 gas recovery,
36 however the number of unique landfills between the four databases does differ.
37 The IPCC default value of 10 percent for uncertainty in recovery estimates was used for 2 of the 4 recovery
3 8 databases in the uncertainty analysis where metering of landfill gas was in place (for about 64 percent of the CH4
39 estimated to be recovered). This 10 percent uncertainty factor applies to the LFGTE database; 12 percent to the EIA
40 database; and 1 percent for the GHGRP MSW landfills dataset because of the supporting information provided and
41 rigorous verification process. For flaring without metered recovery data (the flare database), a much higher
42 uncertainty value of 50 percent is used. The compounding uncertainties associated with the 4 databases in addition
43 to the uncertainties associated with the FOD model and annual waste disposal quantities leads to the large upper and
44 lower bounds for MSW landfills presented in Table 7-5. Industrial waste landfills are shown with a lower range of
45 uncertainty due to the smaller number of data sources and associated uncertainty involved. For example, 3 data
46 sources are used to generate the annual quantities of MSW waste disposed over the 1940 to current year, while
47 industrial waste landfills rely on 2 data sources.
48 The results of the 2006 IPCC Guidelines Approach 2 quantitative uncertainty analysis are summarized in Table 7-5.
49 In 2014, landfill CH4 emissions were estimated to be between 119.4 and 278.7 MMT CO2 Eq., which corresponds to
50 a range of 34 percent below to 53 percent above the 2014 emission estimate of 181.8 MMT CCh Eq.
Waste 7-9
-------
1 Table 7-5: Approach 2 Quantitative Uncertainty Estimates for ChU Emissions from Landfills
2 (MMT COz Eq. and Percent)
Source
Landfills
MSW
Industrial
2014 Emission
Gas Estimate
(MMT CO2 Eq.)
CH4
CH4
CH4
181.8
166.8
15.0
Uncertainty Range Relative to Emission Estimate3
(MMT C02 Eq.) (%)
Lower
Bound
119.4
106.3
10.4
Upper
Bound
278.7
261.5
18.8
Lower
Bound
-34%
-36%
-30%
Upper
Bound
+53%
+57%
+25%
1 Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval.
3 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
4 through 2014. Details on the emission trends through time-series are described in more detail in the Methodology
5 section, above.
9
10
11
12
13
14
QA/QC and Verification
A QA/QC analysis was performed for data gathering and input, documentation, and calculation. QA/QC checks are
performed for the transcription of the published data set used to populate the Inventory data set, including the
published GHGRP and LFGTE databases, but are not performed on the data itself against primary data used. A
primary focus of the QA/QC checks was to ensure that CH4 recovery estimates were not double-counted and that all
LFGTE projects and flares were included in the respective project databases. Both manual and electronic checks
were used to ensure that emission avoidance from each landfill was calculated only once. The primary calculation
spreadsheet is tailored from the IPCC waste model and has been verified previously using the original, peer-
reviewed IPCC waste model. All model input values were verified by secondary QA/QC review.
15
Recalculations Discussion
16 Five major methodological recalculations were performed for the current Inventory.
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
First, the SOG survey data previously used to estimate annual waste generation and disposal was replaced
by EPA's GHGRP and LFGTE data (EPA 2015c) for the inventory time-series (1990 to 2014) and years
from 1983 to 1989. The waste acceptance data included in the databases in EPA (2015c) were used directly
and no modifications to this data set were made. This improvement increased the annual quantities of waste
disposed for each year in the time-series, and consequently, led to an increase in net CH4 emissions
compared to the previous Inventory.
Second, a review of the flare and LFGTE projects across the 4 recovery databases was made. Several
corrections were made to avoid double counting. Additionally, several facilities in the LFGTE database
were removed because they were not in the past three LMOP databases. The LFGTE is an enhanced
version of the LMOP database and if a landfill is no longer in the LMOP database, it was assumed to be
added erroneously. These revisions resulted in larger than expected changes to the annual quantities of the
annual CH4 recovery estimates used in the net CH4 emissions compared to the previous Inventory and are
the second primary driver for the increase in net CH4 emissions after the change to EPA's GHGRP waste
acceptance data.
Third, the GHGRP CH4 recovery data were back-calculated for landfills in EPA's GHGRP database for
years prior to the first GHGRP reporting year (typically 2010 for most landfills). In the previous Inventory,
there was a significant change in the total recovery between years 2009 and 2010. This methodological
change was made to smooth the recovery data for years prior to 2009.
Fourth, the flare correction factor was revised. This effort included reviewing the 27 flare projects included
in the flare correction factor to identify them with landfills in EPA's GHGRP, the LFGTE, or EIA
databases, or match them to existing operational or closed landfills through and Internet search (RTI
2015a). The number of flares included in the flare correction factor decreased from 27 to 19. The impact on
CH4 recovery varies by year and is a modest amount.
7-10 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 • Fifth, the DOC value for landfilled pulp and paper waste was revised from 0.20 to 0.15 based a literature
2 review of pulp and paper waste characterization studies (RTI 2015b) and data reported under the GHGRP.
3 This conservative approach allows for a revision to the DOC value and will be re-assessed in future
4 Inventories as more information becomes available.
5 The overall impact from these changes resulted in an average increase in emissions of nearly 14 percent across the
6 time-series. A significant increase in net CH4 emissions for the years 2010 through 2013 ranging from 20 to 52
7 percent higher in the current Inventory compared to the previous inventory. The drivers for this large increase are
8 the increased quantities of annual waste disposal, which results in higher CH4 generation, and corrections to the
9 landfill matching across the recovery databases.
10 Planned Improvements
11 Improvements being examined for future Inventory estimates include (1) incorporating additional data from recent
12 peer-reviewed literature to modify the default oxidation factor applied to MSW and industrial waste landfills
13 (currently 10 percent); (2) either modifying the bulk waste DOC value or estimating emissions using a waste-
14 specific approach in the first order decay model using data from EPA's GHGRP and peer-reviewed literature; and
15 (3) reviewing waste-stream specific DOC and decay rate constant (k) value data reported for industrial waste
16 landfills as reported under EPA's GHGRP.
17 A standard CH4 oxidation factor of 10 percent has been used for both industrial and MSW landfills in prior
18 Inventory reports and is currently recommended as the default for well-managed landfills in the latest IPCC
19 guidelines (2006). Recent comments received on the Inventory methodology indicated that a default oxidation factor
20 of 10 percent may be less than oxidation rates achieved at well-managed landfills with gas collection and control. As
21 a first step toward revising this oxidation factor, a literature review was conducted in 2011 (RTI 2011). In addition,
22 facilities reporting under EPA's GHGRP have the option to use an oxidation factor other than 10 percent (e.g., 0, 25,
23 or 35 percent) if the calculated result of methane flux calculations warrants it. Various options are being investigated
24 to incorporate this facility-specific data for landfills reporting under EPA's GHGRP and or the remaining facilities.
25 The standard oxidation factor (10 percent) is applied to the total amount of waste generated nationwide. Changing
26 the oxidation factor and calculating the amount of CH4 oxidized from landfills with gas collection and control
27 requires the estimation of waste disposed in these types of landfills over the entire time-series. Although EPA's
28 GHGRP does not capture every landfill in the United States, larger landfills are expected to meet the reporting
29 thresholds and are reporting waste disposal information by year. At this time, data are available to calculate the
30 amount of waste disposed of at landfills with and without gas collection systems in the United States for landfills
31 reporting under EPA's GHGRP. After investigating the landfills not reporting under EPA's GHGRP to determine
32 the presence of a landfill gas collection and control system and waste disposal data, a modification to the Inventory
33 waste model to apply different oxidation factors depending on the presence of a gas collection system may be
34 possible.
35 Other potential improvements to the methodology may be made in the future using other portions of EPA's GHGRP
36 dataset, specifically for inputs to the first order decay equation. The approach used in the Inventory to estimate CH4
37 generation assumes a bulk waste-specific DOC value that may not accurately capture the changing waste
38 composition over the time-series (e.g., the reduction of organics entering the landfill environment due to increased
39 composting, see Box 7-2). Using data obtained from EPA's GHGRP and any publicly available landfill-specific
40 waste characterization studies in the United States, the methodology may be modified to incorporate a waste
41 composition approach, or revisions may be made to the bulk waste DOC value currently used. Additionally, EPA's
42 GHGRP data could be analyzed and a weighted average for the CH4 correction factor (MCF), fraction of CH4 (F) in
43 the landfill gas, the destruction efficiency of flares, and the decay rate constant (k) could replace the values currently
44 used in the Inventory. At this time, the majority of landfills reporting under EPA's GHGRP select bulk MSW for
45 their waste composition.
46 In addition to MSW landfills, industrial waste landfills at facilities emitting CH4 in amounts equivalent to 25,000
47 metric tons or more of CO2 Eq. were required to report their greenhouse gas emissions beginning in September 2012
48 through EPA's GHGRP. Similar data for industrial waste landfills as is required for the MSW landfills are being
49 reported. Any additions or improvements to the Inventory using reported GHGRP data will be made for the
50 industrial waste landfill source category. One potential improvement includes a revision to the waste disposal factor
51 currently used in the Inventory for the pulp and paper sector using production data from pulp and paper facilities that
Waste 7^1?
-------
1 reported annual production and annual disposal data under EPA's GHGRP. Another possible improvement is the
2 addition of emission estimates from landfills at industrial sectors other than pulp and paper, and food and beverage
3 (e.g., metal foundries, petroleum refineries, and chemical manufacturing facilities) to the Inventory.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
Box 7-3: Nationwide Municipal Solid Waste Data Sources
Municipal solid waste generated in the United States can be managed through landfilling, recycling, composting,
and combustion with or without energy recovery. There are three main sources for nationwide solid waste
management data in the United States,
• The BioCycle and Earth Engineering Center of Columbia University's State of Garbage (SOG) in
America surveys
• The EPA's Municipal Solid Waste in The United States: Facts and Figures reports, and
• The EPA's Greenhouse Gas Reporting Program (GHGRP).
The SOG surveys collect state-reported data on the amount of waste generated and the amount of waste managed via
different management options: landfilling, recycling, composting, and combustion. The survey asks for actual
tonnages instead of percentages in each waste category (e.g., residential, commercial, industrial, construction and
demolition, organics, tires) for each waste management option. If such a breakdown is not available, the survey asks
for total tons landfilled. The data are adjusted for imports and exports across state lines so that the principles of mass
balance are adhered to, whereby the amount of waste managed does not exceed the amount of waste generated. The
SOG reports present survey data aggregated to the state level.
The EPA Facts and Figures reports use a materials flow methodology, which relies heavily on a mass balance
approach. Data are gathered from industry associations, key businesses, similar industry sources, and government
agencies (e.g., the Department of Commerce and the U.S. Census Bureau) and are used to estimate tons of materials
and products generated, recycled, or discarded nationwide. The amount of MSW generated is estimated by adjusting
the imports and exports of produced materials to other countries. MSW that is not recycled, composted, or
combusted is assumed to be landfilled. The data presented in the report are nationwide totals.
The EPA GHGRP is a mandatory reporting program to which facilities meeting specific definitions and methane
generation and other greenhouse gas emission thresholds are required to report. If eligible, facilities that operate a
MSW landfill must report a variety of information, including calculated annual methane generation and emissions
using a prescribed methodology, annual waste disposal since the first year of waste acceptance, disposed waste
characteristics (where available), landfill gas collection and control system details (if applicable), including landfill
gas characteristics (e.g., methane concentration, annual volumetric flow).
EPA's GHGRP data are the preferred data source for municipal solid waste disposal amounts in the Inventory
because they are directly reported by facilities using a similar methodology and undergo a rigorous quality assurance
and verification review by the EPA. Supplementary data on the quantity of waste disposed by landfills not subject to
EPA's GHGRP was sourced from the EPA's Landfill Methane Outreach Program (LMOP) and direct information
from state agencies for select landfills that began accepting waste in, or after 2013 for the current Inventory year
(EPA 2015c). In future Inventories, supplementary waste disposal data may be sourced from the Waste Business
Journal, or SOG surveys. The EPA Facts and Figures reports are useful when investigating waste management
trends at the nationwide level and for typical waste composition data, which the State of Garbage surveys do not
request and the GHGRP does not require. The EPA Facts and Figures reports have never been used as the source
for annual waste disposal data in the Inventory.
In this Inventory, emissions from solid waste management are presented separately by waste management option,
except for recycling of waste materials. Emissions from recycling are attributed to the stationary combustion of
fossil fuels that may be used to power on-site recycling machinery, and are presented in the stationary combustion
chapter in the Energy sector, although the emissions estimates are not called out separately. Emissions from solid
waste disposal in landfills and the composting of solid waste materials are presented in the Landfills and
Composting chapters in the Waste sector of this report. In the United States, almost all incineration of MSW occurs
at waste-to-energy (WTE) facilities or industrial facilities where useful energy is recovered, and thus emissions from
waste incineration are accounted for in the Incineration of Waste chapter of the Energy sector of this report.
7-12 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Box 7-4: Overview of the Waste Sector
2 As shown in Figure 7-2 and Figure 7-3, landfilling of MSW is currently and has been the most common waste
3 management practice. A large portion of materials in the waste stream are recovered for recycling and composting,
4 which is becoming an increasingly prevalent trend throughout the country. Materials that are composted and
5 recycled would have normally been disposed of in a landfill.
6 Figure 7-2: Management of Municipal Solid Waste in the United States, 2013
Management of MSW in the United States
Composted
9%
MSWtoWTE
13%
Source: EPA (2015d)
Figure 7-3: MSW Management Trends from 1990 to 2013 (Million Tons)
160
140
120
100
Ifl
i
I 80
i
60
40
20
— — . Landfilling
Recycling
Combustion
with Energy
Recovery
Composting
LOlDr^OOCTlOiHfMrO
ooooooooo
10
11 Source: EPA (2015d)
Waste 7-13
-------
1
2
o
6
4
5
6
7
8
9
10
11
12
Table 7-6 presents a typical composition of waste disposed of at a typical MSW landfill in the United States over
time. It is important to note that the actual composition of waste entering each landfill will vary from that presented
in Table 7-6. Understanding how the waste composition changes over time, specifically for the degradable waste
types, is important for estimating greenhouse gas emissions. For certain degradable waste types (i.e., paper and
paperboard), the amounts discarded have decreased over time due to an increase in waste recovery, including
recycling and composting (see Table 7-6 and Figure 7-4). Landfill ban legislation affecting yard trimmings resulted
in an increase of composting from 1990 to 2008. Table 7-6 and Figure 7-4 do not reflect the impact of backyard
composting on yard trimming generation and recovery estimates. The recovery of food trimmings has been
consistently low. Increased recovery of degradable materials reduces the CH4 generation potential and CH4
emissions from landfills.
Table 7-6: Materials Discarded in the Municipal Waste Stream by Waste Type from 1990 to
2013 (Percent)
Waste Type
Paper and Paperboard
Glass
Metals
Plastics
Rubber and Leather
Lextiles
Wood
Other3
Food Scrapsb
Yard Lrimmingsc
Miscellaneous Inorganic
Wastes
1990
30.0% 1
6.0%
7.2%
9.6%
3.1%
2.9%
6.9%
1.4%
13.6%
17.6%
1.7% H
2005
24.5%
5.7%
7.7%
15.7%
3.5%
5.5%
7.4%
1.8%
17.9%
7.0%
2.1% H
2009
14.8%
5.0%
8.0%
15.8%
3.7%
6.3%
7.7%
1.9%
19.1%
7.6%
2.2%
2010
16.2%
5.1%
8.8%
17.4%
3.7%
6.7%
8.1%
2.0%
21.0%
8.6%
2.3%
2011
14.8%
5.1%
8.9%
17.8%
3.8%
6.8%
8.2%
2.0%
21.4%
8.8%
2.4%
2012
14.8%
5.1%
9.0%
17.6%
3.8%
7.4%
8.2%
2.0%
21.1%
8.7%
2.4%
2013
15.1%
5.0%
9.1%
17.7%
3.9%
7.7%
8.0%
2.0%
21.1%
8.1%
2.4%
a Includes electrolytes in batteries and fluff pulp, feces, and urine in disposable diapers. Details may not add to totals due to
rounding. Source: EPA (2015d).
b Data for food scraps were estimated using sampling studies in various parts of the country in combination with
demographic data on population, grocery store sales, restaurant sales, number of employees, and number of prisoners,
students, and patients in institutions. Source: EPA (2015d).
c Data for yard trimmings were estimated using sampling studies, population data, and published sources documenting
legislation affecting yard trimmings disposal in landfills. Source: EPA (2015d).
13
7-14 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Figure 7-4: Percent of Recovered Degradable Materials from 1990 to 2013 (Percent)
2
3
4
5
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
80% -
70% -
60% -
50% -
40% -
30% -
20% -
10% -
0%
• Paper and Paperboard
• Food Scraps
Yard Trimmings
r-tr-tr-tr-lr-tr-lr-tr-tr-tr-t
OOOOOOOOOOiHiH
oooooooooooo
iHTH
oo
Source: EPA (2015d)
Box 7-5: Description of a Modern, Managed Landfill
Modern, managed landfills are well-engineered facilities that are located, designed, operated, and monitored to
ensure compliance with federal, state, and tribal regulations. Municipal solid waste (MSW) landfills must be
designed to protect the environment from contaminants which may be present in the solid waste stream.
Additionally, many new landfills collect and destroy landfill gas through flares or landfill gas-to-energy projects.
Requirements for affected MSW landfills may include:
• Siting requirements to protect sensitive areas (e.g., airports, floodplains, wetlands, fault areas,
seismic impact zones, and unstable areas)
• Design requirements for new landfills to ensure that Maximum Contaminant Levels (MCLs) will
not be exceeded in the uppermost aquifer (e.g., composite liners and leachate collection systems)
• Leachate collection and removal systems
• Operating practices (e.g., daily and intermediate cover, receipt of regulated hazardous wastes, use
of landfill cover material, access options to prevent illegal dumping, use of a collection system to
prevent stormwater run-on/run-off, record-keeping)
Air monitoring requirements (explosive gases)
Groundwater monitoring requirements
Closure and post-closure care requirements (e.g., final cover construction), and
Corrective action provisions.
Waste 7-15
-------
1 Specific federal regulations that affected MSW landfills must comply with include the 40 CFR Part 258 (Subtitle D
2 of RCRA), or equivalent state regulations and the New Source Performance Standards (NSPS) 40 CFR Part 60
3 Subpart WWW. Additionally, state and tribal requirements may exist.6
Box 7-6: Biogenic Wastes in Landfills
6 Regarding the depositing of wastes of biogenic origin in landfills (i.e., all degradable waste), empirical evidence
7 shows that some of these wastes degrade very slowly in landfills, and the C they contain is effectively sequestered in
8 landfills over a period of time (Barlaz 1998, 2006). Estimates of C removals from landfilling of forest products, yard
9 trimmings, and food scraps are further described in the Land Use, Land-Use Change, and Forestry chapter, based on
10 methods presented in IPCC (2003) and IPCC (2006).
11
12
13
7.2 Wastewater Treatment (IPCC Source
Category 5D) (TO BE UPDATED)
14 Due to circumstances, the wastewater treatment section was not updated for the current Inventory. The numbers are
15 currently being updated using the same methodology as the previous Inventory, with no significant improvements.
16 Minimal impacts on the emission estimates are expected as a result of the update.
17 Wastewater treatment processes can produce anthropogenic CH4 and N2O emissions. Wastewater from domestic7
18 and industrial sources is treated to remove soluble organic matter, suspended solids, pathogenic organisms, and
19 chemical contaminants. Treatment may either occur on site, most commonly through septic systems or package
20 plants, or off site at centralized treatment systems. Centralized wastewater treatment systems may include a variety
21 of processes, ranging from lagooning to advanced tertiary treatment technology for removing nutrients. In the
22 United States, approximately 20 percent of domestic wastewater is treated in septic systems or other on-site systems,
23 while the rest is collected and treated centrally (U.S. Census Bureau 2011).
24 Soluble organic matter is generally removed using biological processes in which microorganisms consume the
25 organic matter for maintenance and growth. The resulting biomass (sludge) is removed from the effluent prior to
26 discharge to the receiving stream. Microorganisms can biodegrade soluble organic material in wastewater under
27 aerobic or anaerobic conditions, where the latter condition produces CH4. During collection and treatment,
28 wastewater may be accidentally or deliberately managed under anaerobic conditions. In addition, the sludge may be
29 further biodegraded under aerobic or anaerobic conditions. The generation of N2O may also result from the
30 treatment of domestic wastewater during both nitrification and denitrification of the N present, usually in the form of
31 urea, ammonia, and proteins. These compounds are converted to nitrate (NOs) through the aerobic process of
32 nitrification. Denitrification occurs under anoxic conditions (without free oxygen), and involves the biological
33 conversion of nitrate into dinitrogen gas (N2). N2O can be an intermediate product of both processes, but has
34 typically been associated with denitrification. Recent research suggests that higher emissions of N2O may in fact
35 originate from nitrification (Ahn et al. 2010). Other more recent research suggests that N2O may also result from
3 6 other types of wastewater treatment operations (Chandran 2012).
37 The principal factor in determining the CH4 generation potential of wastewater is the amount of degradable organic
38 material in the wastewater. Common parameters used to measure the organic component of the wastewater are the
39 Biochemical Oxygen Demand (BOD) and Chemical Oxygen Demand (COD). Under the same conditions,
6 For more information regarding federal MSW landfill regulations, see
.
Throughout the inventory, emissions from domestic wastewater also include any commercial and industrial wastewater
collected and co-treated with domestic wastewater.
7-16 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 wastewater with higher COD (or BOD) concentrations will generally yield more CH4 than wastewater with lower
2 COD (or BOD) concentrations. BOD represents the amount of oxygen that would be required to completely
3 consume the organic matter contained in the wastewater through aerobic decomposition processes, while COD
4 measures the total material available for chemical oxidation (both biodegradable and non-biodegradable). Because
5 BOD is an aerobic parameter, it is preferable to use COD to estimate CH4 production. The principal factor in
6 determining the N2O generation potential of wastewater is the amount of N in the wastewater. The variability of N
7 in the influent to the treatment system, as well as the operating conditions of the treatment system itself, also impact
8 the N2O generation potential.
9 In 2013, CH4 emissions from domestic wastewater treatment were 9.2 MMT CO2 Eq. (368 kt CH4). Emissions
10 remained fairly steady from 1990 through 1997, but have decreased since that time due to decreasing percentages of
11 wastewater being treated in anaerobic systems, including reduced use of on-site septic systems and central anaerobic
12 treatment systems (EPA 1992, 1996, 2000, and 2004, U.S. Census 2011). In 2013, CH4 emissions from industrial
13 wastewater treatment were estimated to be 5.8 MMT CO2 Eq. (233 kt CH4). Industrial emission sources have
14 generally increased across the time series through 1999 and then fluctuated up and down with production changes
15 associated with the treatment of wastewater from the pulp and paper manufacturing, meat and poultry processing,
16 fruit and vegetable processing, starch-based ethanol production, and petroleum refining industries. Table 7-7 and
17 Table 7-8 provide CH4 and N2O emission estimates from domestic and industrial wastewater treatment.
18 With respect to N2O, the United States identifies two distinct sources for N2O emissions from domestic wastewater:
19 emissions from centralized wastewater treatment processes, and emissions from effluent from centralized treatment
20 systems that has been discharged into aquatic environments. The 2013 emissions of N2O from centralized
21 wastewater treatment processes and from effluent were estimated to be 0.3 MMT CO2 Eq. (1 kt N2O) and 4.6 MMT
22 CO2 Eq. (15 kt N2O), respectively. Total N2O emissions from domestic wastewater were estimated to be 4.9 MMT
23 CO2 Eq. (17 kt N2O). N2O emissions from wastewater treatment processes gradually increased across the time
24 series as a result of increasing U.S. population and protein consumption.
25 Table 7-7: CH4 and NzO Emissions from Domestic and Industrial Wastewater Treatment
26 (MMT COz Eq.)
Activity
CH4
Domestic
Industrial*
N20
Domestic
Total
1990
15.7
10.5
5.1
3.4[
3.4
19.1
2005
15.9
10.0
5,
4.3
4.3
20.2
2009
15.6
9.8
5.8
4.6
4.6
20.2
2010
15.5
9.6
5.9
4.7
4.7
20.2
2011
15.3
9.4
5.9
4.8
4.8
20.1
2012
15.2
9.3
5.8
4.9
4.9
20.1
2013
15.0
9.2
5.8
4.9
4.9
19.9
a Industrial activity includes the pulp and paper manufacturing, meat and poultry processing, fruit and
vegetable processing, starch-based ethanol production, and petroleum refining industries.
Note: Totals may not sum due to independent rounding.
27 Table 7-8: CH4 and NzO Emissions from Domestic and Industrial Wastewater Treatment (kt)
Activity
CH4
Domestic
Industrial*
N2O
Domestic
1990
626
421
206
11
11
2005
635
401
234
15
15 | |
2009
623
392
231
16
16
2010
619
384
235
16
16
2011
610
375
235
16
16
2012
606
373
233
16
16
2013
601
368
233
17
17
a Industrial activity includes the pulp and paper manufacturing, meat and poultry processing, fruit and
vegetable processing, starch-based ethanol production, and petroleum refining industries.
Note: Totals may not sum due to independent rounding.
Waste 7-17
-------
i Methodology
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
Domestic Wastewater CH4 Emission Estimates
Domestic wastewater CH4 emissions originate from both septic systems and from centralized treatment systems,
such as publicly owned treatment works (POTWs). Within these centralized systems, CH4 emissions can arise from
aerobic systems that are not well managed or that are designed to have periods of anaerobic activity (e.g.,
constructed wetlands), anaerobic systems (anaerobic lagoons and facultative lagoons), and from anaerobic digesters
when the captured biogas is not completely combusted. CH4 emissions from septic systems were estimated by
multiplying the United States population by the percent of wastewater treated in septic systems (about 20 percent)
and an emission factor (10.7 g CH4/capita/day), and then converting the result to kt/year. CH4emissions from
POTWs were estimated by multiplying the total BOD5 produced in the United States by the percent of wastewater
treated centrally (about 80 percent), the relative percentage of wastewater treated by aerobic and anaerobic systems,
the relative percentage of wastewater facilities with primary treatment, the percentage of BOD5 treated after primary
treatment (67.5 percent), the maximum CH4-producing capacity of domestic wastewater (0.6), and the relative
MCFs for well-managed aerobic (zero), not well managed aerobic (0.3), and anaerobic (0.8) systems with all aerobic
systems assumed to be well-managed. CH4emissions from anaerobic digesters were estimated by multiplying the
amount of biogas generated by wastewater sludge treated in anaerobic digesters by the proportion of CH4 in digester
biogas (0.65), the density of CH4 (662 g CH4/m3 CH4), and the destruction efficiency associated with burning the
biogas in an energy/thermal device (0.99). The methodological equations are:
= USP
Emissions from Septic Systems = A
,P x (% onsite) x (EFSEpTic) x 1/10A9
Days
Emissions from Centrally Treated Aerobic Systems = B
= [(% collected) x (total BOD5 produced) x (% aerobic) x (% aerobic w/out primary) + (% collected) x (total BOD5
produced) x (% aerobic) x (% aerobic w/primary) x (1-% BOD removed in prim, treat.)] x (% operations not well
managed) x (B0) x (MCF-aerobic_not_well_man)
Emissions from Centrally Treated Anaerobic Systems = C
= [(% collected) x (total BOD5 produced) x (% anaerobic) x (% anaerobic w/out primary) + (% collected) x (total
BOD 5 produced) x (% anaerobic) x (% anaerobic w/primary) x (1-%BOD removed in prim, treat.)] x (B0) x (MCF-
anaerobic)
= [(POTW_flow_AD)
Emissions from Anaerobic Digesters = D
(digester gas)/ (per capita flow)] x conversion to m3 x (FRAC_CH4)
ofCH4) x(l-DE) x 1/10A9
Total CH4 Emissions (kt) = A + B + C + D
(365.25) x (density
where,
USpop
% onsite
% collected
% aerobic
% anaerobic
% aerobic w/out primary
% aerobic w/primary
% BOD removed in prim, treat.
% operations not well managed
% anaerobic w/out primary
% anaerobic w/primary
EFsEPTIC
Days
= U.S. population
= Flow to septic systems / total flow
= Flow to POTWs / total flow
= Flow to aerobic systems / total flow to POTWs
= Flow to anaerobic systems / total flow to POTWs
= Percent of aerobic systems that do not employ primary treatment
= Percent of aerobic systems that employ primary treatment
= 32.5%
= Percent of aerobic systems that are not well managed and in which
some anaerobic degradation occurs
= Percent of anaerobic systems that do not employ primary treatment
= Percent of anaerobic systems that employ primary treatment
= Methane emission factor (10.7 g CH4/capita/day) - septic systems
= days per year (365.25)
7-18 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Total BOD5produced = kg BOD/capita/day x U.S. population x 365.25 days/yr
2 Bo = Maximum CH4-producing capacity for domestic wastewater (0.60 kg
3 CHVkgBOD)
4 1/10A6 = Conversion factor, kg to kt
5 MCF-aerobic_not_well_man. = CH4 correction factor for aerobic systems that are not well managed
6 (0.3)
7 MCF-anaerobic = CH4 correction factor for anaerobic systems (0.8)
8 DE = CH4 destruction efficiency from flaring or burning in engine (0.99 for
9 enclosed flares)
10 POTW_flow_AD = Wastewater influent flow to POTWs that have anaerobic digesters
11 (MOD)
12 digester gas = Cubic feet of digester gas produced per person per day (1.0
13 ft3/person/day)
14 per capita flow = Wastewater flow to POTW per person per day (100 gal/person/day)
15 conversion to m3 = Conversion factor, ft3 to m3 (0.0283)
16 FRAC_CH4 = Proportion CH4 in biogas (0.65)
17 density of CH4 = 662 (g CH4/m3 CH4)
18 1/10A9 = Conversion factor, g to kt
19 U.S. population data were taken from the U.S. Census Bureau International Database (U.S. Census 2014) and
20 include the populations of the United States, American Samoa, Guam, Northern Mariana Islands, Puerto Rico, and
21 the Virgin Islands. Table 7-9 presents U.S. population and total BOD5 produced for 1990 through 2013, while Table
22 7-10 presents domestic wastewater CH4 emissions for both septic and centralized systems in 2013. The proportions
23 of domestic wastewater treated onsite versus at centralized treatment plants were based on data from the 1989, 1991,
24 1993, 1995, 1997, 1999, 2001, 2003, 2005, 2007, 2009, and 2011 American Housing Surveys conducted by the U.S.
25 Census Bureau (U.S. Census 2011), with data for intervening years obtained by linear interpolation and data for
26 2013 forecasted using 1990-2012 data. The percent of wastewater flow to aerobic and anaerobic systems, the
27 percent of aerobic and anaerobic systems that do and do not employ primary treatment, and the wastewater flow to
28 POTWs that have anaerobic digesters were obtained from the 1992, 1996, 2000, and 2004 Clean Watershed Needs
29 Survey (EPA 1992, 1996, 2000, and 2004). Data for intervening years were obtained by linear interpolation and the
30 years 2004 through 2013 were forecasted from the rest of the time series. The BOD5 production rate (0.09
31 kg/capita/day) and the percent BOD5 removed by primary treatment for domestic wastewater were obtained from
32 Metcalf and Eddy (2003). The maximum CH4-producing capacity (0.6 kg CHVkg BOD5) and both MCFs used for
33 centralized treatment systems were taken from IPCC (2006), while the CH4 emission factor (10.7 g CHVcapita/day)
34 used for septic systems were taken from Leverenz et al. (2010). The CH4 destruction efficiency for methane
35 recovered from sludge digestion operations, 99 percent, was selected based on the range of efficiencies (98 to 100
36 percent) recommended for flares in AP-42 Compilation of Air Pollutant Emission Factors, Chapter 2.4 (EPA 1998),
37 efficiencies used to establish New Source Performance Standards (NSPS) for landfills, along with data from CAR
38 (2011), Sullivan (2007), Sullivan (2010), and UNFCCC (2012). The cubic feet of digester gas produced per person
39 per day (1.0 ft3/person/day) and the proportion of CH4 in biogas (0.65) come from Metcalf and Eddy (2003). The
40 wastewater flow to a POTW (100 gal/person/day) was taken from the Great Lakes-Upper Mississippi River Board
41 of State and Provincial Public Health and Environmental Managers, "Recommended Standards for Wastewater
42 Facilities (Ten-State Standards)" (2004).
43 Table 7-9: U.S. Population (Millions) and Domestic Wastewater BODs Produced (kt)
1
Year
1990
2005
2009
2010
2011
2012
2013
Population
253
300
r
311
313
316
318
320
BODs
8,333
9,853
10,220
10,303
10,377
10,452
10,534
1
Waste 7-19
-------
Sources: U.S. Census Bureau (2014);
Metcalf& Eddy (2003).
1 Table 7-10: Domestic Wastewater CH4 Emissions from Septic and Centralized Systems
2 (2013)
CH4 Emissions (MMT CCh Eg.) % of Domestic Wastewater CH4
Septic Systems
Centralized Systems (including anaerobic
sludge digestion)
6.0
3.2
65.5%
34.5%
Total
9.2
100%
Note: Totals may not sum due to independent rounding.
4
5
6
7
8
9
10
11
12
13
14
15
Industrial Wastewater ChU Emission Estimates
Methane emission estimates from industrial wastewater were developed according to the methodology described in
IPCC (2006). Industry categories that are likely to produce significant CH4 emissions from wastewater treatment
were identified and included in the Inventory. The main criteria used to identify these industries are whether they
generate high volumes of wastewater, whether there is a high organic wastewater load, and whether the wastewater
is treated using methods that result in CH4 emissions. The top five industries that meet these criteria are pulp and
paper manufacturing; meat and poultry processing; vegetables, fruits, and juices processing; starch-based ethanol
production; and petroleum refining. Wastewater treatment emissions for these sectors for 2013 are displayed in
Table 7-11 below. Table 7-12 contains production data for these industries.
Table 7-11: Industrial Wastewater CH4 Emissions by Sector (2013)
Meat & Poultry
Pulp & Paper
Fruit & Vegetables
Petroleum Refineries
Ethanol Refineries
Total
CH4 Emissions (MMT CCh Eq.)
4.4
1.1
0.1
0.1
0.1
5.8
% of Industrial Wastewater CH4
75%
18%
2%
2%
2%
100%
Note: Totals may not sum due to independent rounding.
Table 7-12: U.S. Pulp and Paper, Meat, Poultry/ Vegetables, Fruits and Juices, Ethanol, and
Petroleum Refining Production (MMT)
Meat Poultry Vegetables,
| Pulp and (Live Weight (Live Weight Fruits and
Year
1990
2005
2009
2010
2011
2012
2013
aPulp
Petroleum
Paper3 Killed) Killed) Juices Ethanol
128.9 27.3 14.6
138.5 31.4 25.1
120.4 33.8 25.2
128.6 33.7 25.9
127.5 33.8 26.2
127.0 33.8 26.1
131.5 33.6 26.5
38.7
42.9
46.5
43.2
44.3
45.3
43.9
2.5
11.7
32.7
39.7
41.6
39.5
39.8
Refining
702.4
818.6
822.4
848.6
858.8
856.1
875.9
and paper production is the sum of woodpulp production plus paper and paperboard production.
Sources: Lockwood-Post (2002); FAO (2014); USDA (2014a); RFA (2014);
EIA(2014).
7-20 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 CH4 emissions from these categories were estimated by multiplying the annual product output by the average
2 outflow, the organics loading (in COD) in the outflow, the maximum CH4 producing potential of industrial
3 wastewater (B0), and the percentage of organic loading assumed to degrade anaerobically in a given treatment
4 system (MCF). Ratios of BOD:COD in various industrial wastewaters were obtained from EPA (1997a) and used to
5 estimate COD loadings. The B0 value used for all industries is the IPCC default value of 0.25 kg CHVkg COD
6 (IPCC 2006).
7 For each industry, the percent of plants in the industry that treat wastewater on site, the percent of plants that have a
8 primary treatment step prior to biological treatment, and the percent of plants that treat wastewater anaerobically
9 were defined. The percent of wastewater treated anaerobically onsite (TA) was estimated for both primary treatment
10 (%TAP) and secondary treatment (%TAS). For plants that have primary treatment in place, an estimate of COD that
11 is removed prior to wastewater treatment in the anaerobic treatment units was incorporated. The values used in the
12 %TA calculations are presented in Table 7-13 below.
13 The methodological equations are:
14
15
16
17 where,
18 CH4 (industrial wastewater) = Total CH4 emissions from industrial wastewater (kg/year)
19 P = Industry output (metric tons/year)
20 W = Wastewater generated (m3/metric ton of product)
21 COD = Organics loading in wastewater (kg/m3)
22 %TAP = Percent of wastewater treated anaerobically on site in primary treatment
23 %TAs = Percent of wastewater treated anaerobically on site in secondary treatment
24 %Plants0 = Percent of plants with onsite treatment
25 %WWa,p = Percent of wastewater treated anaerobically in primary treatment
26 %CODP = Percent of COD entering primary treatment
27 %Plantsa = Percent of plants with anaerobic secondary treatment
28 %Plantst = Percent of plants with other secondary treatment
29 %WWa,s = Percent of wastewater treated anaerobically in anaerobic secondary treatment
30 %WWa,t = Percent of wastewater treated anaerobically in other secondary treatment
31 %CODS = Percent of COD entering secondary treatment
32 Bo = Maximum CH4 producing potential of industrial wastewater (default value of
33 0.25 kg CHVkg COD)
34 MCF = CH4 correction factor, indicating the extent to which the organic content
35 (measured as COD) degrades anaerobically
36 Alternate methodological equations for calculating %TA were used for secondary treatment in the pulp and paper
37 industry to account for aerobic systems with anaerobic portions. These equations are:
38 %TAa = [%Plantsa * %WWas * %CODs]+[%Plantst * %WWat x CODS]
39 %TAat = [%Plantsat x %WWas x %CODS]
40 where,
41 %TAa = Percent of wastewater treated anaerobically on site in secondary treatment
42 %TAat = Percent of wastewater treated in aerobic systems with anaerobic portions on
43 site in secondary treatment
44 %PlantSa = Percent of plants with anaerobic secondary treatment
45 %Plantsa,t = Percent of plants with partially anaerobic secondary treatment
46 %WWa,s = Percent of wastewater treated anaerobically in anaerobic secondary treatment
47 %WWa,t = Percent of wastewater treated anaerobically in other secondary treatment
Waste 7^2?
-------
%CODS
= Percent of COD entering secondary treatment
2 As described below, the values presented in Table 7-13 were used in the emission calculations and are described in
3 detail inERG (2008), ERG (2013a), and ERG (2013b).
5 Table 7-13: Variables Used to Calculate Percent Wastewater Treated Anaerobically by
6 Industry (percent)
Variable
%TAP
%TAS
%TAa
%TAa,t
%Plants0
%PlantSa
%Plantsa,t
%Plantst
%WWa,p
%WWa,s
%WWa,t
%CODP
%CODS
Pulp
and
Paper
0
0
2.2
11.8
0
5
28
35
0
100
0
100
42
Meat
Processing
0
33
0
0
100
33
0
67
0
100
0
100
^^^^H
Poultry
Processing
0
25
0
0
100
25
0
75
0
100
0
100
^^^^B
Industry
Fruit/
Vegetable
Processing
0
4.2
0
0
11
5.5
0
5.5
0
100
0
100
77
Ethanol
Production
-Wet Mill
0
33.3
0
0
100
33.3
0
66.7
0
100
0
100
100
Ethanol
Production
- Dry Mill
0
75
0
0
100
75
0
25
0
100
0
100
100
Petroleum
Refining
0
23.6
0
0
100
23.6
0
0
0
100
0
100
100
Sources: ERG (2008); ERG (2013a); and ERG (2013b).
7 Pulp and Paper. Wastewater treatment for the pulp and paper industry typically includes neutralization, screening,
8 sedimentation, and flotation/hydrocycloning to remove solids (World Bank 1999, Nemerow and Dasgupta 1991).
9 Secondary treatment (storage, settling, and biological treatment) mainly consists of lagooning. In determining the
10 percent that degrades anaerobically, both primary and secondary treatment were considered. In the United States,
11 primary treatment is focused on solids removal, equalization, neutralization, and color reduction (EPA 1993). The
12 vast majority of pulp and paper mills with on-site treatment systems use mechanical clarifiers to remove suspended
13 solids from the wastewater. About 10 percent of pulp and paper mills with treatment systems use settling ponds for
14 primary treatment and these are more likely to be located at mills that do not perform secondary treatment (EPA
15 1993). However, because the vast majority of primary treatment operations at U.S. pulp and paper mills use
16 mechanical clarifiers, and less than 10 percent of pulp and paper wastewater is managed in primary settling ponds
17 that are not expected to have anaerobic conditions, negligible emissions are assumed to occur during primary
18 treatment.
19 Approximately 42 percent of the BOD passes on to secondary treatment, which consists of activated sludge, aerated
20 stabilization basins, or non-aerated stabilization basins. BasedonEPA'sOAQPS Pulp and Paper Sector Survey, 5.3
21 percent of pulp and paper mills reported using anaerobic secondary treatment for wastewater and/or pulp
22 condensates (ERG 2013a). Twenty-eight percent (28 percent) of mills also reported the use of quiescent settling
23 ponds. Using engineering judgment, these systems were determined to be aerobic with possible anaerobic portions.
24 For the truly anaerobic systems, an MCF of 0.8 is used, as these are typically deep stabilization basins. For the
25 partially anaerobic systems, an MCF of 0.2 is used, which is the IPCC suggested MCF for shallow lagoons.
26 A time series of CH4 emissions for 1990 through 2001 was developed based on production figures reported in the
27 Lockwood-Post Directory (Lockwood-Post 2002). Data from the Food and Agricultural Organization of the United
28 Nations (FAO) database FAOSTAT were used for 2002 through 2013 (FAO 2014). The overall wastewater outflow
29 varies based on a time series outlined in ERG (2013a) to reflect historical and current industry wastewater flow, and
30 the average BOD concentrations in raw wastewater was estimated to be 0.4 gram BOD/liter (EPA 1997b, EPA
31 1993, World Bank 1999). The COD:BOD ratio used to convert the organic loading to COD for pulp and paper mills
32 was 2 (EPA 1997a).
7-22 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Meat and Poultry Processing. The meat and poultry processing industry makes extensive use of anaerobic lagoons
2 in sequence with screening, fat traps, and dissolved air flotation when treating wastewater on site. About 33 percent
3 of meat processing operations (EPA 2002) and 25 percent of poultry processing operations (U.S. Poultry 2006)
4 perform on-site treatment in anaerobic lagoons. The IPCC default B0 of 0.25 kg CHVkg COD and default MCF of
5 0.8 for anaerobic lagoons were used to estimate the CH4 produced from these on-site treatment systems. Production
6 data, in carcass weight and live weight killed for the meat and poultry industry, were obtained from the USD A
7 Agricultural Statistics Database and the Agricultural Statistics Annual Reports (USD A 2014a). Data collected by
8 EPA's Office of Water provided estimates for wastewater flows into anaerobic lagoons: 5.3 and 12.5 m3/metric ton
9 for meat and poultry production (live weight killed), respectively (EPA 2002). The loadings are 2.8 and 1.5 g
10 BOD/liter for meat and poultry, respectively. The COD:BOD ratio used to convert the organic loading to COD for
11 both meat and poultry facilities was 3 (EPA 1997a).
12 Vegetables, Fruits, and Juices Processing. Treatment of wastewater from fruits, vegetables, and juices processing
13 includes screening, coagulation/settling, and biological treatment (lagooning). The flows are frequently seasonal,
14 and robust treatment systems are preferred for on-site treatment. Effluent is suitable for discharge to the sewer.
15 This industry is likely to use lagoons intended for aerobic operation, but the large seasonal loadings may develop
16 limited anaerobic zones. In addition, some anaerobic lagoons may also be used (Nemerow and Dasgupta 1991).
17 Consequently, 4.2 percent of these wastewater organics are assumed to degrade anaerobically. The IPCC default B0
18 of 0.25 kg CH4/kg COD and default MCF of 0.8 for anaerobic treatment were used to estimate the CH4 produced
19 from these on-site treatment systems. The USDA National Agricultural Statistics Service (USDA 2014a) provided
20 production data for potatoes, other vegetables, citrus fruit, non-citrus fruit, and grapes processed for wine. Outflow
21 and BOD data, presented in Table 7-14, were obtained from EPA (1974) for potato, citrus fruit, and apple
22 processing, and from EPA (1975) for all other sectors. The COD:BOD ratio used to convert the organic loading to
23 COD for all fruit, vegetable, and juice facilities was 1.5 (EPA 1997a).
24 Table 7-14: Wastewater Flow (m3/ton) and BOD Production (g/L) for U.S. Vegetables, Fruits,
25 and Juices Production
Commodity Wastewater Outflow (mVton) BOD (g/L)
Vegetables
Potatoes
Other Vegetables
Fruit
Apples
Citrus
Non-citrus
Grapes (for wine)
10.27
8.67
3.66
10.11
12.42
2.78
1.765
0.791
1.371
0.317
1.204
1.831
Sources: EPA 1974, EPA 1975.
26 Ethanol Production. Ethanol, or ethyl alcohol, is produced primarily for use as a fuel component, but is also used in
27 industrial applications and in the manufacture of beverage alcohol. Ethanol can be produced from the fermentation
28 of sugar-based feedstocks (e.g., molasses and beets), starch- or grain-based feedstocks (e.g., corn, sorghum, and
29 beverage waste), and cellulosic biomass feedstocks (e.g., agricultural wastes, wood, and bagasse). Ethanol can also
30 be produced synthetically from ethylene or hydrogen and carbon monoxide. However, synthetic ethanol comprises
31 only about 2 percent of ethanol production, and although the Department of Energy predicts cellulosic ethanol to
32 greatly increase in the coming years, currently it is only in an experimental stage in the United States. Currently,
33 ethanol is mostly made from sugar and starch crops, but with advances in technology, cellulosic biomass is
3 4 increasingly used as ethanol feedstock (DOE 2013).
35 Ethanol is produced from corn (or other starch-based feedstocks) primarily by two methods: wet milling and dry
36 milling. Historically, the majority of ethanol was produced by the wet milling process, but now the majority is
37 produced by the dry milling process. The dry milling process is cheaper to implement, and has become more
38 efficient in recent years (Rendleman and Shapouri 2007). The wastewater generated at ethanol production facilities
39 is handled in a variety of ways. Dry milling facilities often combine the resulting evaporator condensate with other
40 process wastewaters, such as equipment wash water, scrubber water, and boiler blowdown and anaerobically treat
41 this wastewater using various types of digesters. Wet milling facilities often treat their steepwater condensate in
42 anaerobic systems followed by aerobic polishing systems. Wet milling facilities may treat the stillage (or processed
Waste 7^23"
-------
1 stillage) from the ethanol fermentation/distillation process separately or together with steepwater and/or wash water.
2 CH4 generated in anaerobic digesters is commonly collected and either flared or used as fuel in the ethanol
3 production process (ERG 2006).
4 Available information was compiled from the industry on wastewater generation rates, which ranged from 1.25
5 gallons per gallon ethanol produced (for dry milling) to 10 gallons per gallon ethanol produced (for wet milling)
6 (Ruocco 2006a,b; Merrick 1998; Donovan 1996; and NRBP 2001). COD concentrations were also found to be
7 about 3 g/L (Ruocco 2006a; Merrick 1998; White and Johnson 2003). The amount of wastewater treated
8 anaerobically was estimated, along with how much of the CH4 is recovered through the use of biomethanators.
9 Biomethanators are anaerobic reactors that use microorganisms under anaerobic conditions to reduce COD and
10 organic acids and recover biogas from wastewater (ERG 2006). Methane emissions were then estimated as follows:
11
12 Methane = [Production x Flow x COD x 3.785 x ([%Plants0 x %WWa,P x %CODP] + [%Plantsa x %WWa,s x %CODS] +
13 [%Plantst x %WWa,t x %CODS]) x B0 x MCF x % Not Recovered] + [Production x Flow x 3.785 x COD x ([%Plants0 x
14 %WWa,P x %CODP] + [%PlantSa x %WWa,s x %CODS] + [%Plantst x %WWa,t x %CODS]) x B0 x MCF x (% Recovered) x (1-
15 DE)] x 1/10A9
16 where,
17 Production = gallons ethanol produced (wet milling or dry milling)
18 Flow = gallons wastewater generated per gallon ethanol produced (1.25 dry milling, 10 wet milling)
19 COD = COD concentration in influent (3 g/1)
20 3.785 = conversion, gallons to liters
21 %PlantSo = percent of plants with onsite treatment (100%)
22 %WWa,p = percent of wastewater treated anaerobically in primary treatment (0%)
23 %CODP = percent of COD entering primary treatment (100%)
24 %PlantSa = percent of plants with anaerobic secondary treatment (33.3% wet, 75% dry)
25 %Plantst = percent of plants with other secondary treatment (66.7% wet, 25% dry)
26 %WWa,s = percent of wastewater treated anaerobically in anaerobic secondary treatment (100%)
27 %WWa,t = percent of wastewater treated anaerobically in other secondary treatment (0%)
28 %CODS = percent of COD entering secondary treatment (100%)
29 Bo = maximum methane producing capacity (0.25 g CEU/g COD)
30 MCF = methane conversion factor (0.8 for anaerobic systems)
31 % Recovered = percent of wastewater treated in system with emission recovery
32 % Not Recovered = 1 - percent of wastewater treated in system with emission recovery
33 DE = destruction efficiency of recovery system (99%)
34 1/10A9 = conversion factor, g to kt
35 A time series of CH4 emissions for 1990 through 2013 was developed based on production data from the Renewable
3 6 Fuels Association (RFA 2014).
37 Petroleum Refining. Petroleum refining wastewater treatment operations have the potential to produce CH4
38 emissions from anaerobic wastewater treatment. EPA's Office of Air and Radiation performed an Information
39 Collection Request (ICR) for petroleum refineries in 2011.8 Of the responding facilities, 23.6 percent reported using
40 non-aerated surface impoundments or other biological treatment units, both of which have the potential to lead to
41 anaerobic conditions (ERG 2013b). In addition, the wastewater generation rate was determined to be 26.4 gallons
42 per barrel of finished product (ERG 2013b). An average COD value in the wastewater was estimated at 0.45 kg/m3
43 (Benyahia et al. 2006).
44 The equation used to calculate CH4 generation at petroleum refining wastewater treatment systems is presented
45 below:
46 Methane = Flow x COD x TA x B0 x MCF
47 where,
48 Flow = Annual flow treated through anaerobic treatment system (m3/year)
49 COD = COD loading in wastewater entering anaerobic treatment system (kg/m3)
50 TA = Percent of wastewater treated anaerobically on site
' Available online at
-------
1 BO = maximum methane producing potential of industrial wastewater (default value of 0.25
2 kg CH4 /kg COD)
3 MCF = methane conversion factor (0.3)
4 A time series of CH4 emissions for 1990 through 2013 was developed based on production data from the Energy
5 Information Association (El A 20 14).
6 Domestic Wastewater NzO Emission Estimates
7 N2O emissions from domestic wastewater (wastewater treatment) were estimated using the IPCC (2006)
8 methodology, including calculations that take into account N removal with sewage sludge, non-consumption and
9 industrial/commercial wastewater N, and emissions from advanced centralized wastewater treatment plants:
10 • In the United States, a certain amount of N is removed with sewage sludge, which is applied to land, incinerated,
11 or landfilled (NSLUDGE). The N disposal into aquatic environments is reduced to account for the sewage sludge
12 application.
13 • The IPCC methodology uses annual, per capita protein consumption (kg protein/person-year). For this
14 inventory, the amount of protein available to be consumed is estimated based on per capita annual food
15 availability data and its protein content, and then adjusts that data using a factor to account for the fraction of
16 protein actually consumed.
17 • Small amounts of gaseous nitrogen oxides are formed as byproducts in the conversion of nitrate to N gas in
1 8 anoxic biological treatment systems. Approximately 7 g N2O is generated per capita per year if wastewater
19 treatment includes intentional nitrification and denitrification (Scheehle and Doom 2001). Analysis of the 2004
20 CWNS shows that plants with denitrification as one of their unit operations serve a population of 2.4 million
2 1 people. Based on an emission factor of 7 g per capita per year, approximately 21.2 metric tons of additional N2O
22 may have been emitted via denitrification in 2004. Similar analyses were completed for each year in the
23 Inventory using data from CWNS on the amount of wastewater in centralized systems treated in denitrification
24 units. Plants without intentional nitrification/denitrification are assumed to generate 3.2 g N2O per capita per
25 year.
26 N2O emissions from domestic wastewater were estimated using the following methodology:
27 N2OlOTAL = N2OpLANT + N2OEFFLUENT
28 N2OpLANT = N2ONIT/DENIT + N2OwOUT MT/DENIT
29 N2ONIT/DENIT= [(USPOPND) X EF2 X FlND-COM] X 1/10A9
30 N2OwOUT NIT/DENIT = { [(USpOP x WWTP) - USPOPND] X FrND-COM x EFl } X 1/10A9
3 1 NZOEFFLUENT = {[(((USpop x WWTP) - (0.9 x USPOPND)) x Protein x FNPR x FNON-CON x FIND-COM) - NSLUDGE] x EF3 x
32 44/28} x 1/10A6
33 where,
34 N2OioTAL = Annual emissions of N2O (kt)
35 N2OpLANi = N2O emissions from centralized wastewater treatment plants (kt)
36 N2ONiT/DENii = N2O emissions from centralized wastewater treatment plants with
37 nitrification/denitrification (kt)
3 8 N2Owour NIT/DENIT = N2O emissions from centralized wastewater treatment plants without
39 nitrification/denitrification (kt)
40 N2OEFFLUENT = N2O emissions from wastewater effluent discharged to aquatic environments (kt)
41 USpop = U.S. population
42 USPOPND = U. S. population that is served by biological denitrification (from CWNS)
43 WWTP = Fraction of population using WWTP (as opposed to septic systems)
44 EFi = Emission factor (3 .2 g N2O/person-year) - plant with no intentional denitrification
45 EF2 = Emission factor (7 g N2O/person-year) - plant with intentional denitrification
46 Protein = Annual per capita protein consumption (kg/person/year)
47 FNPR = Fraction of N in protein, default = 0.16 (kg N/kg protein)
Waste 7-25
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
2 1
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
FNON-CON
FIND-COM
NSLUDGE
EF3
0.9
44/28
= Factor for non-consumed protein added to wastewater (1.4)
= Factor for industrial and commercial co -discharged protein into the sewer system
(1.25)
= N removed with sludge, kg N/yr
= Emission factor (0.005 kg N2O -N/kg sewage-N produced) - from effluent
= Amount of nitrogen removed by denitrification systems
= Molecular weight ratio of N2O to N2
U.S. population data were taken from the U.S. Census Bureau International Database (U.S. Census 2014) and
include the populations of the United States, American Samoa, Guam, Northern Mariana Islands, Puerto Rico, and
the Virgin Islands. The fraction of the U.S. population using wastewater treatment plants is based on data from the
1989, 1991, 1993, 1995, 1997, 1999, 2001, 2003, 2005, 2007, 2009, and 2011 American Housing Survey (U.S.
Census 201 1). Data for intervening years were obtained by linear interpolation and data from 2012 and 2013 were
forecasted using 1990-201 1 data. The emission factor (EFi) used to estimate emissions from wastewater treatment
for plants without intentional denitrification was taken from IPCC (2006), while the emission factor (EF2) used to
estimate emissions from wastewater treatment for plants with intentional denitrification was taken from Scheehle
and Doom (200 1). Data on annual per capita protein intake were provided by the U.S. Department of Agriculture
Economic Research Service (USDA 2014b). Protein consumption data for 2010 through 2013 were extrapolated
from data for 1990 through 2006. An emission factor to estimate emissions from effluent (EF3) has not been
specifically estimated for the United States, thus the default IPCC value (0.005 kg N2O-N/kg sewage-N produced)
was applied (IPCC 2006). The fraction of N in protein (0. 16 kg N/kg protein) was also obtained from IPCC (2006).
The factor for non-consumed protein and the factor for industrial and commercial co-discharged protein were
obtained from IPCC (2006). Sludge generation was obtained from EPA (1999) for 1988, 1996, and 1998 and from
Beecher et al. (2007) for 2004. Intervening years were interpolated, and estimates for 2005 through 2012 were
forecasted from the rest of the time series. The amount of nitrogen removed by denitrification systems was taken
from EPA (2008). An estimate for the N removed as sludge (NSLUDGE) was obtained by determining the amount of
sludge disposed by incineration, by land application (agriculture or other), through surface disposal, in landfills, or
through ocean dumping (US EPA 1993b, Beecher et al. 2007, McFarland 2001, US EPA 1999). In 2013, 286 kt N
was removed with sludge. Table 7-15 presents the data for U.S. population, population served by biological
denitrification, population served by wastewater treatment plants, available protein, protein consumed, and nitrogen
removed with sludge.
Table 7-15: U.S. Population (Millions), Population Served by Biological Denitrification
(Millions), Fraction of Population Served by Wastewater Treatment (percent), Available
Protein (kg/person-year), Protein Consumed (kg/person-year), and Nitrogen Removed with
Sludge (kt-N/year)
Year Population Populations WWTP Population Available Protein Protein Consumed N Removed
1990
1
2005
1 •
2009
2010
2011
2012
2013
253
300
311
313
316
318
320
2.0
2.7
2.9
3.0
3.0
3.0
3.1
75.6
78.8
79.3
80.0
80.6
80.4
80.7
38.4
39.8
|
40.9
41.0
41.1
41.2
41.3
29.5
30.7
31.5
31.6
31.7
31.8
31.9
214.1
1
261.1
1
273.4
276.4
279.5
282.6
285.6
Sources: Beecher et al. 2007, McFarland 2001, U.S. Census 201 1, U.S. Census 2014, USDA 2014b, US EPA 1992, US EPA
1993b, US EPA 1996, US EPA 1999, US EPA 2000, US EPA 2004.
Uncertainty and Time-Series Consistency
The overall uncertainty associated with both the 2013 CH4 and N2O emission estimates from wastewater treatment
and discharge was calculated using the 2006 IPCC Guidelines Approach 2 methodology (2006). Uncertainty
associated with the parameters used to estimate CH4 emissions include that of numerous input variables used to
model emissions from domestic wastewater, and wastewater from pulp and paper manufacture, meat and poultry
processing, fruits and vegetable processing, ethanol production, and petroleum refining. Uncertainty associated with
7-26 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 the parameters used to estimate N2O emissions include that of sewage sludge disposal, total U.S. population,
2 average protein consumed per person, fraction of N in protein, non-consumption nitrogen factor, emission factors
3 per capita and per mass of sewage-N, and for the percentage of total population using centralized wastewater
4 treatment plants.
5 The results of this Approach 2 quantitative uncertainty analysis are summarized in Table 7-16. Methane emissions
6 from wastewater treatment were estimated to be between 9.2 and 15.3 MMT CC>2 Eq. at the 95 percent confidence
7 level (or in 19 out of 20 Monte Carlo Stochastic Simulations). This indicates a range of approximately 39 percent
8 below to 2 percent above the 2013 emissions estimate of 15.0 MMT CO2 Eq. N2O emissions from wastewater
9 treatment were estimated to be between 1.2 and 10.2 MMT CC>2 Eq., which indicates a range of approximately 76
10 percent below to 107 percent above the 2013 emissions estimate of 4.9 MMT CO2 Eq.
11 Table 7-16: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from
12 Wastewater Treatment (MMT COz Eq. and Percent)
16
30
Source
Wastewater Treatment
Domestic
Industrial
Wastewater Treatment
2013 Emission Estimate Uncertainty Range Relative to Emission Estimate3
(MMTCChEq.) (MMT CCh Eq.) (%)
CH4
CH4
CH4
N20
15.0
9.2
5.8
4.9
Lower
Bound
9.2
5.7
2.4
1.2
Upper
Bound
15.3
9.9
6.9
10.2
Lower
Bound
-39%
-38%
-59%
-76%
Upper
Bound
+2%
+7%
+18%
+107%
a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent
confidence interval.
13 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
14 through 2013. Details on the emission trends through time are described in more detail in the Methodology section,
15 above.
QA/QC and Verification
17 A QA/QC analysis was performed on activity data, documentation, and emission calculations. This effort included a
18 Tier 1 analysis, including the following checks:
19 • Checked for transcription errors in data input;
20 • Ensured references were specified for all activity data used in the calculations;
21 • Checked a sample of each emission calculation used for the source category;
22 • Checked that parameter and emission units were correctly recorded and that appropriate conversion factors
23 were used;
24 • Checked for temporal consistency in time series input data for each portion of the source category;
25 • Confirmed that estimates were calculated and reported for all portions of the source category and for all years;
26 • Investigated data gaps that affected emissions estimates trends; and
27 • Compared estimates to previous estimates to identify significant changes.
28 All transcription errors identified were corrected. The QA/QC analysis did not reveal any systemic inaccuracies or
29 incorrect input values.
Recalculations Discussion
31 Production data were updated to reflect revised USDA NASS datasets. In addition, the most recent USDA ERS data
32 were used to update percent protein values from 1990 through 2010. The updated ERS data also resulted in small
33 changes in forecasted values from 2011. The factor for sewage sludge production change per year was updated to
34 include all available data. This change resulted in updated 1990 through 1995 values for total N in sludge along with
35 a change in forecasted values from 2005 through 2012.
Waste 7-27
-------
1 Workbooks were also updated to show emissions in kilotons and MMT CO2 Eq. In addition, global warming
2 potentials for N2O and CH4 were updated with the AR4 100-year values (IPCC 2007).
3 For the current Inventory, emission estimates have been revised to reflect the GWPs provided in the IPCC Fourth
4 Assessment Report (AR4) (IPCC 2007). AR4 GWP values differ slightly from those presented in the IPCC Second
5 Assessment Report (SAR) (IPCC 1996) (used in the previous inventories) which results in time-series recalculations
6 for most inventory sources. Under the most recent reporting guidelines (UNFCCC 2014), countries are required to
7 report using the AR4 GWPs, which reflect an updated understanding of the atmospheric properties of each
8 greenhouse gas. The GWPs of CH4 and most fluorinated greenhouse gases have increased, leading to an overall
9 increase in CCh-equivalent emissions from CH4. The GWPs of N2O and SF6 have decreased, leading to a decrease in
10 CO2-equivalent emissions for N2O. The AR4 GWPs have been applied across the entire time series for consistency.
11 For more information please see the Recalculations and Improvements Chapter.
12
Planned Improvements
13 The methodology to estimate CH4 emissions from domestic wastewater treatment currently utilizes estimates for the
14 percentage of centrally treated wastewater that is treated by aerobic systems and anaerobic systems. These data
15 come from the 1992, 1996, 2000, and 2004 CWNS. The question of whether activity data for wastewater treatment
16 systems are sufficient across the time series to further differentiate aerobic systems with the potential to generate
17 small amounts of CH4 (aerobic lagoons) versus other types of aerobic systems, and to differentiate between
18 anaerobic systems to allow for the use of different MCFs for different types of anaerobic treatment systems,
19 continues to be explored. The CWNS data for 2008 were evaluated for incorporation into the Inventory, but due to
20 significant changes in format, this dataset is not sufficiently detailed for inventory calculations. However, additional
21 information and other data continue to be evaluated to update future years of the Inventory, including anaerobic
22 digester data compiled by the North East Biosolids and Residuals Association (NEBRA) in collaboration with
23 several other entities. While NEBRA is no longer involved in the project, the Water Environment Federation (WEF)
24 now hosts and manages the dataset which has been relocated to www.wef.org/biosolids. WEF will complete the
25 second phase of their data collection and by late fall. They are currently collecting additional data on a Region by
26 Region basis which should add to the quality of the database by decreasing uncertainty and data gaps (ERG 2014a).
27 EPA will continue to monitor the status of these data as a potential source of digester, sludge, and biogas data from
28 POTWs.
29 Data collected under the EPA's Greenhouse Gas Reporting Program Subpart II, Industrial Wastewater Treatment
30 (GHGRP) is being investigated for use in improving the emission estimates for the industrial wastewater category.
31 Ensuring time series consistency has been the focus, as the reporting data from EPA's GHGRP are not available for
32 all inventory years. Whether EPA's GHGRP reporters sufficiently represent U.S. emissions is being investigated to
33 determine if moving to a facility-level implementation of GHGRP data is warranted, or whether the GHGRP data
34 will allow update of activity data for certain industry sectors, such as use of biogas recovery systems or update of
35 waste characterization data. Since EPA's GHGRP only includes reporters that have met a certain threshold and
36 because EPA is unable to review whether the reporters represent the majority of U.S. production, GHGRP data are
37 not believed to be sufficiently representative to move toward facility-level estimates in the Inventory. However, the
38 GHGRP data continues to be evaluated for improvements to activity data, and in verifying methodologies currently
39 in use in the Inventory to estimate emissions (ERG 2014b). In implementing any improvements and integration of
40 data from EPA's GHGRP, EPA will follow the latest guidance from the IPCC on the use of facility-level data in
41 national inventories.9
42 For industrial wastewater emissions, EPA is also working with the National Council of Air and Stream Improvement
43 (NCASI) to determine if there are sufficient data available to update the estimates of organic loading in pulp and
44 paper wastewaters treated on site. These data include the estimates of wastewater generated per unit of production,
45 the BOD and/or COD concentration of these wastewaters, and the industry-level production basis used in the
46 Inventory. EPA has received data on the industry-level production basis to date and intends to incorporate these data
47 once a full evaluation of the production basis is made in relation to the wastewater generation rate and the organic
9 See:
7-28 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 content of the wastewater. In this way, EPA plans to make a coordinated update to the three values that are used to
2 estimate the total organic industry load to wastewater treatment plants, rather than multiple changes over time.
3 In addition to this investigation, any reports based on international research will be investigated to inform potential
4 updates to the Inventory. The Global Water Research Coalition report has been evaluated, regarding wastewater
5 collection and treatment systems (GWRC 2011). The report included results of studies from Australia, France, the
6 Netherlands, and the United States. Since each dataset was taken from a variety of wastewater treatment plant types
7 using different methodologies and protocols, it was determined that it was not representative enough to include in
8 the Inventory at this time (ERG 2014a). In addition, wastewater inventory submissions from other countries have
9 been investigated to determine if there are any emission factors, specific methodologies, or additional industries that
10 could be used to inform the U.S. inventory calculations. Although no comparable data have been found, other
11 countries' submissions will continue to be investigated for potential improvements to the inventory.
12 IPCC's 2013 wetlands supplement has also been investigated regarding the inclusion of constructed and semi-
13 natural treatment wetlands in Inventory calculations (IPCC 2014). Methodologies are presented for estimating both
14 CH4 and N2O. Next, the use of CWNS treatment system data will be investigated to determine if these data can be
15 used to estimate the amount of wastewater treated in constructed wetlands for potential implementation in future
16 Inventory reports.
17 Currently, for domestic wastewater, it is assumed that all aerobic wastewater treatment systems are well managed
18 and produce no CH4 and that all anaerobic systems have an MCF of 0.8. Efforts to obtain better data reflecting
19 emissions from various types of municipal treatment systems are currently being pursued by researchers, including
20 the Water Environment Research Federation (WERF). This research includes data on emissions from partially
21 anaerobic treatment systems which was reviewed (Willis et al. 2013). It was determined that the emissions were too
22 variable and the sample size too small to include in the Inventory at this time. In addition, information on flare
23 efficiencies was reviewed and it was determined that they were not suitable for use in updating the Inventory
24 because the flares used in the study are likely not comparable to those used at wastewater treatment plants (ERG
25 2014a). The status of this and similar research will continue to be monitored for potential inclusion in the Inventory
26 in the future.
27 With respect to estimating N2O emissions, the default emission factors for indirect N2O from wastewater effluent
28 and direct N2O from centralized wastewater treatment facilities have a high uncertainty. Research is being
29 conducted by WERF to measure N2O emissions from municipal treatment systems and is periodically reviewed for
30 its utility for the Inventory. The Phase I report from WERF on N2O emissions was recently reviewed and EPA
31 concluded, along with the author, that there were not enough data to create an emission factor for N2O (Chandran
32 2012). While the authors suggested a facility-level approach, there are not enough data available to estimate N2O
33 emissions on a facility-level for the more than 16,000 POTWs in the United States (ERG 2014a). In addition, a
34 literature review has been conducted focused on N2O emissions from wastewater treatment to determine the state of
35 such research and identify data to develop a country-specific N2O emission factor or alternate emission factor or
36 method (ERG 2011). Such data will continue to be reviewed as they are available to determine if a country-specific
37 N2O emission factor can or should be developed, or if alternate emission factors should be used. EPA will also
3 8 follow up with the authors of any relevant studies, including those from WERF, to determine if there is additional
39 information available on potential methodological revisions.
40 There is the potential for N2O emissions associated with on-site industrial wastewater treatment operations;
41 however, the methodology provided in IPCC (2006) only addresses N2O emissions associated with domestic
42 wastewater treatment. A literature review was initiated to assess other Annex I countries' wastewater inventory
43 submissions for additional data and methodologies that could be used to inform the U.S. wastewater inventory
44 calculations, in particular to determine if any countries have developed industrial wastewater N2O emission
45 estimates (ERG 2014a). Currently, there are insufficient data to develop a country-specific methodology; however,
46 available data will continue to be reviewed, and will consider if indirect N2O emissions associated with on-site
47 industrial wastewater treatment using the IPCC default factor for domestic wastewater (0.005 kg N2O-N/kg N)
48 would be appropriate.
49 Previously, a new measurement data from WERF was used to develop a U.S.-specific emission factor for CH4
50 emissions from septic systems, and these were incorporated into the inventory emissions calculation. Due to the high
51 uncertainty of the measurements for N2O from septic systems, estimates of N2O emissions were not included.
52 Appropriate emission factors for septic system N2O emissions will continue to be investigated as the data collected
53 by WERF indicate that septic soil systems are a source of N2O emissions.
Waste 7^29
-------
1 In addition, the estimate of N entering municipal treatment systems is under review. The factor that accounts for
2 non-sewage N in wastewater (bath, laundry, kitchen, industrial components) also has a high uncertainty. Obtaining
3 data on the changes in average influent N concentrations to centralized treatment systems over the time series would
4 improve the estimate of total N entering the system, which would reduce or eliminate the need for other factors for
5 non-consumed protein or industrial flow. The dataset previously provided by the National Association of Clean
6 Water Agencies (NACWA) was reviewed to determine if it was representative of the larger population of
7 centralized treatment plants for potential inclusion into the Inventory. However, this limited dataset was not
8 representative of the number of systems by state or the service populations served in the United States, and therefore
9 could not be incorporated into the inventory methodology. Additional data sources will continue to be researched
10 with the goal of improving the uncertainty of the estimate of N entering municipal treatment systems. Unfortunately,
11 NACWA's suggestion of using National Pollution Discharge Elimination System (NPDES) permit data to estimate
12 nitrogen loading rates is not feasible as influent concentration are not available. EPA is also evaluating whether
13 available effluent nitrogen concentrations reported under POTW NPDES permits would support a more robust
14 analysis of nitrogen contributing to indirect nitrous oxide emissions. Not every POTW is required to measure for
15 effluent nitrogen so the database is not a complete source. Often, only those POTWs that are required to reduce
16 nutrients are monitoring effluent nitrogen, so the database may reflect lower N effluent loadings than that typical
17 throughout the United States. However, EPA is continuing to evaluate the utility of these data in future inventories.
18 The value used for N content of sludge continues to be investigated. This value is driving the N2O emissions for
19 wastewater treatment and is static over the time series. To date, new data have not been identified that would be able
20 to establish a time series for this value. The amount of sludge produced and sludge disposal practices will also be
21 investigated. In addition, based on UNFCCC review comments, the transparency of the fate of sludge produced in
22 wastewater treatment will continue to be improved.
23 A review of other industrial wastewater treatment sources for those industries believed to discharge significant loads
24 of BOD and COD has been ongoing. Food processing industries have the highest potential for CH4 generation due
25 to the waste characteristics generated, and the greater likelihood to treat the wastes anaerobically. However, in all
26 cases there is dated information available on U.S. wastewater treatment operations for these industries. Previously,
27 organic chemicals, the seafood processing industry, and coffee processing were investigated to estimate their
28 potential to generate CH4. Due to the insignificant amount of CH4 estimated to be emitted and the lack of reliable,
29 up-to-date activity data, these industries were not selected for inclusion in the Inventory. Analyses of breweries and
30 dairy products processing facilities have been performed. While the amount of COD present in brewery wastewater
31 is substantial, it is likely that the majority of the industry utilizes aerobic treatment or anaerobic treatment with
32 biogas recovery. As a result, breweries will not be included in the Inventory at this time. There are currently limited
33 data available on the wastewater characteristics and treatment of dairy processing wastewater, but EPA will continue
34 to investigate this and other industries as necessary for inclusion in future years of the Inventory.
35
7.3 Composting (IPCC Source Category 5B1)
36 Composting of organic waste, such as food waste, garden (yard) and park waste, and wastewater treatment sludge
37 and/or biosolids, is common in the United States. Advantages of composting include reduced volume of the waste,
3 8 stabilization of the waste, and destruction of pathogens in the waste. The end products of composting, depending on
39 its quality, can be recycled as a fertilizer and soil amendment, or be disposed of in a landfill.
40 Composting is an aerobic process and a large fraction of the degradable organic carbon in the waste material is
41 converted into carbon dioxide (CO2). Methane (CH4) is formed in anaerobic sections of the compost, which are
42 created when there is excessive moisture or inadequate aeration (or mixing) of the compost pile. This CH4 is then
43 oxidized to a large extent in the aerobic sections of the compost. The estimated CH4 released into the atmosphere
44 ranges from less than 1 percent to a few percent of the initial C content in the material (IPCC 2006). Depending on
45 how well the compost pile is managed, nitrous oxide (N2O) emissions can be produced. The formation of N2O
46 depends on the initial nitrogen content of the material and is mostly due to nitrogen oxide (NOX) denitrification
47 during the later composting stages. Emissions vary and range from less than 0.5 percent to 5 percent of the initial
48 nitrogen content of the material (IPCC 2006). Animal manures are typically expected to generate more N2O than, for
49 example, yard waste, however data are limited.
7-30 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 From 1990 to 2014, the amount of waste composted in the United States has increased from 3,810 kt to 20,533 kt, an
2 increase of approximately 439 percent. The amount of material composted in the United States in the last decade has
3 increased by approximately 11 percent. Emissions of CH4 and N2O from composting have increased by the same
4 percentage. In 2014, CH4 emissions from composting (see Table 7-17 and Table 7-18) were 2.1 MMT CO2 Eq. (82
5 kt), and N2O emissions from composting were 1.8 MMT CO2 Eq. (6 kt). The wastes composted primarily include
6 yard trimmings (grass, leaves, and tree and brush trimmings) and food scraps from the residential and commercial
7 sectors (such as grocery stores; restaurants; and school, business, and factory cafeterias). The composted waste
8 quantities reported here do not include backyard composting or agricultural composing.
9 The growth in composting since the 1990s and specifically over the past decade is attributable to primarily two
10 factors: (1) the enactment of legislation by state and local governments that discouraged the disposal of yard
11 trimmings in landfills, (2) yard trimming collection and yard trimming drop off sites provided by local solid waste
12 management districts/divisions, and (3) an increased awareness of the environmental benefits of composting. Most
13 bans on the disposal of yard trimmings were initiated in the early 1990s by state or local governments (U.S.
14 Composting Council 2010). By 2010, 25 states, representing about 50 percent of the nation's population, had
15 enacted such legislation (BioCycle 2010). An additional 16 states are known to have commercial-scale composting
16 facilities (Shin 2014). Despite these factors, the total amount of waste composted exhibited a downward trend after
17 peaking in 2008 (see Table 7-17 and Figure 7-5), but has been increasing since 2010 and the annual quantity
18 composted is now on par with the 2008 quantity composted. While there is no definitive reason for the decreasing
19 trend in the amount of waste composted, it is most likely a result of the recession and the fact that the quantities
20 composted are estimated using a mass balance approach on the municipal waste stream across the entire United
21 States. As presented in Figure 7-5, the quantity of CH4 and N2O emitted from composting operations across the
22 time-series parallels the trends for the quantities composted, although the trend in emissions has a much lower slope
23 compared to the quantities composted.
24 Table 7-17: CH4 and NzO Emissions from Composting (MMT COz Eq.)
26
Activity
CH4
N2O
Total
1990
0.4
0.3 1
0.7
2005
1.9
1.7
3.5
2010
1.8
1.6
3.5
2011
1.9
1.7
3.5
2012
1.9
1.7
3.7
2013
2.0
1.8
3.9
2014
2.1
1.8
3.9
Note: Totals may not sum due to independent rounding.
25 Table 7-18: CH4 and NzO Emissions from Composting (kt)
Activity
CH4
N2O
1990 1
15
1
2005
75
6
2010
73
| 5
2011
75
6
2012
77
6
2013
81
6
2014
82
6
Waste 7-31
-------
1 Figure 7-5: CH4 and N2O Emitted from Composting Operations Between 1990 and 2014 (kt)
90.0
Year
•Millions of tons composted
•kt of methane emitted
•kt of nitrous oxide emitted
4
5
6
9
10
Methodology
Methane and N2O emissions from composting depend on factors such as the type of waste composted, the amount
and type of supporting material (such as wood chips and peat) used, temperature, moisture content (e.g., wet and
fluid versus dry and crumbly), and aeration during the composting process.
The emissions shown in Table 7-17 and Table 7-18 were estimated using the IPCC default (Tier 1) methodology
(IPCC 2006), which is the product of an emission factor and the mass of organic waste composted (note: no CH4
recovery is expected to occur at composting operations in the emission estimates presented):
11 where,
12
13
14
15
16
M
EFi
E. = MxEK
CH4 or N2O emissions from composting, kt CH4 or N2O,
mass of organic waste composted in kt,
emission factor for composting, 4 t CH4/kt of waste treated (wet basis) and 0.3 t N2O/kt
of waste treated (wet basis) (IPCC 2006), and
designates either CH4 or N2O.
17 Estimates of the quantity of waste composted (M) are presented in Table 7-19. Estimates of the quantity composted
18 for 1990, 2005 and 2007 through 2009 were taken from Municipal Solid Waste in the United States: 2010 Facts and
19 Figures (EPA 2011); estimates of the quantity composted for 2006 were taken from EPA's Municipal Solid Waste
20 In The United States: 2006 Facts and Figures (EPA 2007); estimates of the quantity composted for 2011 through
21 2012 were taken from EPA's Municipal Solid Waste In The United States: 2012 Facts and Figures (EPA 2014);
22 estimates of the quantity composted for 2013 was taken from EPA's Advancing Sustainable Materials Management:
23 Facts and Figures 2013 (EPA 2015); estimates of the quantity composted for 2014 were extrapolated using the 2013
24 quantity composted and a ratio of the U.S. population in 2013 and 2014 (U.S. Census Bureau 2015).
7-32 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 7-19: U.S. Waste Composted (kt)
Activity 1990 2005 2010 2011 2012 2013 2014
Waste Composted 3,810 18,643 18,298 18,661 19,351 20,358 20,533
2 Uncertainty and Time-Series Consistency
3 The estimated uncertainty from the 2006IPCC Guidelines is ±50 percent for the Approach 1 methodology.
4 Emissions from composting in 2014 were estimated to be between 1.9 and 5.8 MMT CO2 Eq., which indicates a
5 range of 50 percent below to 50 percent above the actual 2014 emission estimate of 3.9 MMT €62 Eq. (see Table
6 7-20).
7 Table 7-20: Approach 1 Quantitative Uncertainty Estimates for Emissions from Composting
8 (MMT COz Eq. and Percent)
Source Gas 2014 Emission Estimate Uncertainty Range Relative to Emission Estimate
(MMT CCh Eq.) (MMT CCh Eq.) (%)
12
16
Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
Composting CH4, N2O 3.9 1.9 5.8 -50% +50%
9 Methodological recalculations were applied to the entire time-series to ensure time-series consistency from 1990
10 through 2014. Details on the emission trends through time-series are described in more detail in the Methodology
11 section, above.
QA/QC and Verification
13 A QA/QC analysis was performed for data gathering and input, documentation, and calculation. A primary focus of
14 the QA/QC checks was to ensure that the amount of waste composted annually was correct according to the latest
15 EPAAdvancing Sustainable Materials Management: Facts and Figures report.
Recalculations Discussion
17 The estimated amount of waste composted in 2013 was updated based on new data contained in EPA's Advancing
18 Sustainable Materials Management: Facts and Figures 2013 report (EPA 2015). The amounts of CH4 and N2O
19 emissions estimates presented in Table 7-17 and Table 7-18 were revised accordingly.
20 Planned Improvements
21 For future Inventories, additional efforts will be made to improve the estimates of CH4 and N2O emissions from
22 composting. For example, a literature search on emission factors and their drivers (e.g., the type of composting
23 system, material composition, management technique, impact of varying climatic regions) is underway. The purpose
24 of this literature review is to compile all published emission factors to determine whether the emission factors used
25 in the current methodology should be revised, or expanded to account for various composting system, material
26 composition, management techniques, and/or geographical/climatic differences. For example, composting systems
27 that primarily compost food waste may generate CH4 at different rates than composting yard trimmings because the
28 food waste may have a higher moisture content and more readily degradable material. Further cooperation with
29 estimating emissions in the LULUCF Other section will also be investigated.
Waste 7-33
-------
i 7.4 Waste Incineration (IPCC Source Category
2 5C1)
3 As stated earlier in this chapter, CO2, N2O, and CH4 emissions from the incineration of waste are accounted for in
4 the Energy sector rather than in the Waste sector because almost all incineration of municipal solid waste (MSW) in
5 the United States occurs at waste-to-energy facilities where useful energy is recovered. Similarly, the Energy sector
6 also includes an estimate of emissions from burning waste tires and hazardous industrial waste, because virtually all
7 of the combustion occurs in industrial and utility boilers that recover energy. The incineration of waste in the United
8 States in2014 resulted in 9.7 MMT CO2 Eq. emissions, over half of which (4.9 MMT CO2 Eq.) is attributable to the
9 combustion of plastics. For more details on emissions from the incineration of waste, see Section 3.3 of the Energy
10 chapter.
11 Additional sources of emissions from waste incineration include non-hazardous industrial waste incineration and
12 medical waste incineration. As described in Annex 5 of this report, data are not readily available for these sources
13 and emission estimates are not provided. An analysis of the likely level of emissions was conducted based on a 2009
14 study of hospital/ medical/ infectious waste incinerator (HMIWI) facilities in the United States (RTI 2009). Based
15 on that study' s information of waste throughput and an analysis of the fossil-based composition of the waste, it was
16 determined that annual greenhouse gas emissions for medical waste incineration would be below 500 kt CO2 Eq. per
17 year and considered insignificant for the purposes of Inventory reporting under the UNFCCC. More information on
18 this analysis is provided in Annex 5.
19 7.5 Waste Sources of Indirect Greenhouse
20 Gases
21 In addition to the main greenhouse gases addressed above, waste generating and handling processes are also sources
22 of indirect greenhouse gas emissions. Total emissions of NOX, CO, and NMVOCs from waste sources for the years
23 1990 through 2014 are provided in Table 7-21.
7-34 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Table 7-21: Emissions of NOX, CO, and NMVOC from Waste (kt)
11
Gas/Source 1990 2005 2010 2011 2012 2013 2014
NOx + 21 11111
Landfills + I 2 I 1 1 1 1 1
Wastewater Treatment + I 0 I 0 0000
Miscellaneous3 + I 0 I 0 0000
CO l|7J55555
Landfills I I 6 I 5 4444
Wastewater Treatment + I + I + + + + +
Miscellaneous3 + I 0 I 0 0 0 0 0
NMVOCs 673 114 44 38 38 38 39
Wastewater Treatment 57 49 19 17 17 17 17
Miscellaneous3 557 43 17 15 15 15 15
Landfills 58 22_^__8 7777
a Miscellaneous includes TSDFs (Treatment, Storage, and Disposal Facilities under the Resource Conservation
and Recovery Act [42 U.S.C. § 6924, SWDA § 3004]) and other waste categories.
Note: Totals may not sum due to independent rounding.
+ Does not exceed 0.5 kt.
2 Methodology
3 Emission estimates for 1990 through 2014 were obtained from data published on the National Emission Inventory
4 (NEI) Air Pollutant Emission Trends web site (EPA 2015), and disaggregated based on EPA (2003). Emission
5 estimates for 2014 for non-EGU and non-mobile sources are held constant from 2011 in EPA (2015). Emission
6 estimates of these gases were provided by sector, using a "top down" estimating procedure—emissions were
7 calculated either for individual sources or for many sources combined, using basic activity data (e.g., the amount of
8 raw material processed) as an indicator of emissions. National activity data were collected for individual categories
9 from various agencies. Depending on the category, these basic activity data may include data on production, fuel
10 deliveries, raw material processed, etc.
Uncertainty and Time-Series Consistency
12 No quantitative estimates of uncertainty were calculated for this source category. Methodological recalculations
13 were applied to the entire time-series to ensure time-series consistency from 1990 through 2014. Details on the
14 emission trends through time are described in more detail in the Methodology section, above.
Waste 7-35
-------
-------
i
8. Other
2 The United States does not report any greenhouse gas emissions under the Intergovernmental Panel on Climate
3 Change (IPCC) "Other" sector.
Other 8-1
-------
-------
i 9. Recalculations and Improvements
2 Each year, emission and sink estimates are recalculated and revised for all years in the Inventory of U.S. Greenhouse
3 Gas Emissions and Sinks, as attempts are made to improve both the analyses themselves, through the use of better
4 methods or data, and the overall usefulness of the report. In this effort, the United States follows the 2006IPCC
5 Guidelines (IPCC 2006), which states, "Both methodological changes and refinements over time are an essential
6 part of improving inventory quality. It is good practice to change or refine methods when available data have
7 changed; the previously used method is not consistent with the IPCC guidelines for that category; a category has
8 become key; the previously used method is insufficient to reflect mitigation activities in a transparent manner; the
9 capacity for inventory preparation has increased; new inventory methods become available; and for correction of
10 errors."
11 The results of all methodological changes and historical data updates made in the current Inventory report are
12 presented in this section; detailed descriptions of each recalculation are contained within each source's description
13 found in this report, if applicable. Table 9-1 summarizes the quantitative effect of these changes on U.S. greenhouse
14 gas emissions and sinks and Table 9-2 summarizes the quantitative effect on annual net CC>2 fluxes, both relative to
15 the previously published U.S. Inventory (i.e., the 1990 through 2013 report). These tables present the magnitude of
16 these changes in units of million metric tons of carbon dioxide equivalent (MMT CC>2 Eq.).
17 The Recalculations Discussion section of each source's description in the respective chapter of this Inventory
18 presents the details of each recalculation. In general, when methodological changes have been implemented, the
19 entire time series (i.e., 1990 through 2013) has been recalculated to reflect the change, per IPCC (2006). Changes in
20 historical data are generally the result of changes in statistical data supplied by other agencies.
21 The following ten emission sources and sinks underwent some of the most significant methodological and historical
22 data changes. These emission sources consider only methodological and historical data changes. A brief summary of
23 the recalculations and/or improvements undertaken is provided for each of the ten sources.
24 • Natural Gas Systems and Petroleum Systems (CB.4). Substantial new data are available on natural gas and
25 petroleum systems from subpart W of the EPA's greenhouse gas reporting program (GHGRP) and a number of
26 new studies. The EPA is evaluating approaches for incorporating this new data into its emission estimates for
27 the Inventory of U.S. GHG Emissions and Sinks: 1990-2014. The details of the revisions under consideration
28 for this year's Inventory, and key questions for stakeholder feedback are available in segment-level memoranda
29 at http://www3.epa.gov/climatechange/ghgemissions/usinventoryreport/natural-gas-systems.html. See the
30 Energy chapter, sections 3.7 and 3.6, for further information.
31 • Landfills (CH4). Five major methodological recalculations were performed for the current Inventory. First, the
32 SOG survey data previously used to estimate annual waste generation and disposal was replaced by the GHGRP
33 and LFGTE data (EPA 2015) for the inventory time series (1990 to 2014) and years from 1983 to 1989. Second,
34 a review of the flare and LFGTE projects across the 4 recovery databases was made. Several corrections were
35 made to avoid double counting. Additionally, several facilities in the LFGTE database were removed because
36 they were not in the past three LMOP databases. Third, the GHGRP CH4 recovery data were back-calculated
37 for landfills in EPA's GHGRP database for years prior to the first GHGRP reporting year (typically 2010 for
38 most landfills). Fourth, the flare correction factor was revised. Fifth, the DOC value for landfilled pulp and
39 paper waste was revised from 0.20 to 0.15 based a literature review of pulp and paper waste characterization
40 studies (RTI 2015) and data reported under EPA's GHGRP. Overall there is a significant increase in net CH4
41 emissions for the years 2010 through 2013 ranging from 20 to 52 percent higher in the current Inventory
Recalculations and Improvements 9-1
-------
1 compared to the 1990 through 2013 Inventory. The drivers for this large increase are the increased quantities of
2 annual waste disposal, which results in higher CH4 generation, and corrections to the landfill matching across
3 the recovery databases. These changes resulted in an average annual increase in emissions of 19.4 MMT CO2
4 Eq. relative to the previous Inventory.
5 • Forest Land Remaining Forest Land (CO2 sink). Forest ecosystem stock and stock-change estimates differ from
6 the previous inventory report principally due to the adoption of a new accounting approach (FCAF; Woodall et
7 al. 2015). The major differences between FCAF and past accounting approaches is the sole use of annual forest
8 inventory data and the back casting of forest C stocks across the -1990s based on forest C stock density and
9 land use change information derived from the nationally consistent forest inventory (coupled with in situ
10 observations of non-tree C pools such as soils, dead wood, and litter). The adoption of FCAF has enabled the
11 creation of the two land use sections for forest C stocks: Forest Land Remaining Forest Land and Land
12 Converted to Forest Land. In prior submissions (e.g., the 2015 Inventory submission), these two land use
13 sections were combined. A second major change was the adoption of a new approach to estimating forest soil C,
14 the largest stock in the U.S. For detailed discussion of these new approaches please refer to the Methodolgy
15 section, Annex 3.12, Domke etal. (in preparation), and Woodall et al. 2015. In addition to these major changes,
16 the model of Ogle et al. (in preparation) identifies some of the forest land in south central and southeastern
17 coastal Alaska as unmanaged; this is in contrast to past assumptions of "managed" for these forest lands
18 included in the FIADB. Therefore, the estimates for coastal Alaska as included here reflect that adjustment,
19 which effectively reduces the forest area included here by about 5 percent.
20 In addition to the creation of explicit estimates of removals and emissions by Forest Land Remaining Forest
11 Land versus Land Converted to Forest Land, the FCAF eliminated the use of inconsistent periodic data which
22 contributed to a data artefact in prior estimates of emissions/removals from 1990 to the present. In the 2015
23 submission, there was a reduction in net sequestration from 1995 to 2000 followed by an increase in net
24 sequestration from 2000 to 2004. This artefact of comparing inconsistent inventories of the 1980s-1990sto the
25 nationally consistent inventories of the 2000s has been removed in this 2016 submission. This has resulted in a
26 fairly consistent estimated annual net sequestration of-160 MMT C. Overall, there were net C additions to
27 HWP in use and in landfills combined due, in large part, to updated data on Products in Use from 2010-present.
28 All these changes resulted in an average annual decrease in sequestration of 144.7 MMT CO2 Eq. relative to the
29 previous Inventory.
30 • Agricultural Soil Management (N2O). Methodological recalculations in the current Inventory are associated
31 with the following improvements: 1) driving the DAYCENT simulations with updated input data for land
32 management from the National Resources Inventory extending the time series through 2010; 2) accounting for
33 N inputs from residues associated with additional crops not simulated by DAYCENT including most vegetable
34 crops; 3) modifying the number of experimental study sites used to quantify model uncertainty for direct N2O
35 emissions; and 4) using DAYCENT for direct N2O emissions from most flooded rice lands, instead of using the
36 Tier 1 approach for all rice lands. These changes resulted in an increase in emissions of approximately 24
37 percent on average relative to the previous Inventory and a decrease in the upper bound of the 95 percent
38 confidence interval for direct N2O emissions from 26 to 24 percent. The differences are mainly due to
39 increasing the number of study sites used to quantify model uncertainty and correct bias. These changes resulted
40 in an average annual increase in emissions of 59.8 MMT CO2 Eq. relative to the previous Inventory.
41 • Cropland Remaining Cropland (CO2 sink). Methodological recalculations in the current Inventory are
42 associated with the following improvements: 1) incorporation of updated NRI data for 1990 through 2010; 2)
43 inclusion of federal croplands; and 3) improving the simulation of hydric soil. These changes in SOC stocks
44 resulted in an average annual increase in sequestration of 8.4 MMT CO2 Eq. relative to the previous Inventory.
45 • Land Converted to Cropland (CO2 sink). Methodological recalculations in the current Inventory are associated
46 with the following improvements: 1) incorporation of updated NRI data for 1990 through 2010; 2) inclusion of
47 federal croplands; and 3) improving the simulation of hydric soils in DAYCENT. These changes in SOC stocks
48 resulted in an average annual decrease in emissions of 4.3 MMT CO2 Eq. relative to the previous Inventory.
49 • Grassland Remaining Grassland (CO2 sink). Methodological recalculations in the current Inventory are
50 associated with the following improvements, including 1) incorporation of updated NRI data for 1990 through
51 2010; 2) inclusion of federal grasslands in the Tier 2 analysis; and 3) improving the simulation of hydric soils in
9-2 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 DAYCENT. These changes in SOC stocks resulted in an average annual decrease in emissions of 4.3 MMT
2 CO2 Eq. relative to the previous Inventory.
3 • Land Converted to Grassland (CO2 sink). Methodological recalculations in the current Inventory are associated
4 with the following improvements, including 1) incorporation of updated NRI data for 1990-2010; 2) inclusion
5 of federal grasslands in the Tier 2 analysis; and 3) improving the simulation of hydric soils in DAYCENT. As a
6 result of these improvements to the Inventory, changes in SOC stocks increased by an average of 0.2 MMT CO2
7 eq. annually over the time series. These changes resulted in an average annual increase in sequestration of 3.4
8 MMT CO2 Eq. relative to the previous Inventory.
9 • Substitution of Ozone Depleting Substances (HFCs). For the current Inventory, a review of the large retail food
10 end-use resulted in revisions to the Vintaging Model since the previous Inventory. In addition, a vending
11 machine end-use was added to the Vintaging Model since the previous Inventory. Methodological
12 recalculations were applied to the entire time-series to ensure time-series consistency from 1990 through 2014.
13 For the large retail food end-use, assumptions regarding new installations by system type and refrigerant
14 transitions were revised based on a review of data collected by EPA's GreenChill Partnership and the California
15 Air Resources Board's Refrigerant Management Program. The vending machine end-use was added based on a
16 review of technical reports and sales data. Combined, these assumption changes and additions decreased GHG
17 emissions on average by 3 percent between 1990 and 2002 and increased GHG emissions on average by 4
18 percent between 2003 and 2014. These changes resulted in an average annual increase in emissions of 2.1 MMT
19 CO2 Eq. relative to the previous Inventory.
20 • Forest Fires (CB.4). The current non-CO2 emissions estimates are based on the calculation described in Forest
21 Land Remaining Forest Land (see Section 6.2) and inlPCC (2006), which is a very similar form to the basic
22 calculation of previous Inventory reports (i.e., this report, last year and previous). However, some of the data
23 summarized and applied to the calculation are now very different. The use of the MTBS data summaries is the
24 most prominent example (see Planned Improvements discussion in previous Inventories). Annual burned areas
25 on managed forest lands were identified according to Ruefenacht et al. (2008) and Ogle et al. (in preparation).
26 The other change with this year's estimates is in the use of the underlying plot level carbon densities based on
27 forest inventory plots. Although the base data are similar to past years, the current uncertainty estimates are
28 based on an assumption that plot-to-plot variability is a greater influence on uncertainty than the uncertainty in
29 the forest-inventory to carbon conversions factors (as employed for uncertainty in the past). See Annex 3.13 for
30 additional details. These changes resulted in an average annual decrease in emissions of 1.5 MMT CO2 Eq.
31 relative to the previous Inventory.
32
Recalculations and Improvements 9-3
-------
Table 9-1: Revisions to U.S. Greenhouse Gas Emissions (MMT COz Eq.)
Gas/Source
CO2
Fossil Fuel Combustion
Electricity Generation
Transportation
Industrial
Residential
Commercial
U.S. Territories
Non-Energy Use of Fuels
1990
0.3
+
NCU
NCM
+
NCM
NCM
NCM
0.5
2005
(1.4)
(0.5)
NM
(0.8)
0.2M
+
+
NCM
+1
2010
(6.3)
(8.9)
NC
(3.7)
(0.1)
(0.1)
(0.1)
(4.8)
(0.5)
2011
(0.3)
(3.7)
NC
(3.9)
(0.8)
(0.4)
(0.3)
1.7
0.2
2012
2.7
(1.3)
NC
(4.0)
(1.3)
(0.6)
(0.4)
5.0
0.7
Average
Annual
2013 Change
8.1
(0.1)
(1.6)
(5.4)
(5.0)
0.1
0.3
11.6
1.8
(0.6)
(1.1)
(0.1)
(1.5)
(0.2)
+
+
0.6
0.1
Natural Gas Systems*
Cement Production
Lime Production
Other Process Uses of Carbonates
Glass Production
Soda Ash Production and Consumption
Carbon Dioxide Consumption
Incineration of Waste
Titanium Dioxide Production
Aluminum Production
Iron and Steel Production & Metallurgical Coke
Production
Ferroalloy Production
Ammonia Production
Urea Consumption for Non-Agricultural Purposes
Phosphoric Acid Production
Petrochemical Production
Silicon Carbide Production and Consumption
Lead Production
Zinc Production
Liming
Peatlands Remaining Peatlands
Petroleum Systems'5
Magnesium Production and Processing
Urea Fertilization
LULUCF Total NetFluof
Biomass - Woodc
International Bunker Fuelsc
Biomass - Ethanolc
CH4
Stationary Combustion
Mobile Combustion
Coal Mining
Abandoned Underground Coal Mines
Natural Gas Systems*
Petroleum Systems'5
Petrochemical Production
Silicon Carbide Production and Consumption
Iron and Steel Production & Metallurgical Coke
Production
Ferroalloy Production
Enteric Fermentation
(0.1)
NC
NC
NC
(0.1)
+
NC
NC
(0.1)
NC
NC
NC
(0.1)
(0.7)
NC
NC
NC NC
NC NC
NC NC
NC
NC
NC
0.1
3.2
NC
NC
NC
(0.1)
NC
NC
NC
(0.1)
NC
NC
NC
NC
NC
NC
NC
NC
NC
0.1
3.3
NC
NC
NC
(0.1)
NC
NC
NC
+
(0.1)
NC
NC
NC
+
NC
NC
NC
197.4
NC
NC
NC
50.8
NC
NC
0.1
3.2
+
NC
NC
(0.1)
NC
NC
NC
+
+
NC
NC
NC
0.2
NC
NC
NC
199.6
NC
NC
NC
56.2
6.0
0.2
0.1
3.3
(0.7)
0.1
NC
(0.1)
NC
(0.2)
(0.5)
(0.1)
(0.1)
+
+
NC
(2.0)
NC
NC
0.3
199.4
3.0
NC
NC
67.7
NC
NC
+
NC
NC
NC
+
NC
NC
NC
+
NC
(0.7) (0.7) (0.7)
NC NC NC
0.2 0.4 0.9
0.2
+
0.1
0.5
NC
(0.1)
NC
(0.1)
(0.2)
NC
(0.1)
NC
NC
+
NC
0.1
NC
NC
19.4
+
(0.1)
NC
NC
NC
(0.9)
NC
0.1
9-4 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
Manure Management
Rice Cultivation
Field Burning of Agricultural Residues
Forest Fires
Peatlands Remaining Peatlands
Landfills
Wastewater Treatment
Composting
Incineration of Waste
International Bunker Fuels'
N20
Stationary Combustion
Mobile Combustion
Adipic Acid Production
Nitric Acid Production
Manure Management
Agricultural
Soil Management
NC
2.1
(0.1)
0.8
NC
(1.9)1
NC
NC
NC
NC
79.6
NC
0.2
78.9
NCB
5.2!
+
i.eB
NCB
21.8B
NcB
NcB
E
50.5
1
53.1
+
1.1
+
(1.5)
NC
54.5
NC
NC
NC
NC
55.2
(0.1)
NC
0.1
56.1
0.1
3.7
+
(8.0)
NC
55.6
NC
NC
NC
NC
51.9
(0.1)
NC
0.1
57.1
+
2.9
+
(4.6)
NC
58.3
NC
NC
NC
NC
53.8
(0.2)
NC
0.2
56.9
+
3.9
+
1.5
NC
62.0
NC
0.1
NC
NC
55.8
(0.2)
NC
NC
0.2
54.7
+
2.5
+
(1.5)
NC
19.4
NC
+
NC
NC
57.5
(1.4)
NC
0.1
59.8
Field Burning of Agricultural Residues + + + + + + +
Wastewater
Treatment
N2O from Product Uses
Incineration
of Waste
Settlement Soils
Forest Fires
Forest Soils
Composting
Peatlands Remaining Peatlands
Semiconductor Manufacture
International Bunker Fuels0
HFCs
Substitution of Ozone Depleting Substances
HCFC-22 Production
Semiconductor Manufacture
Magnesium Production and Processing
PFCs
Aluminum Production
Semiconductor Manufacture
SF6
Electrical Transmission and Distribution
Semiconductor Manufacture
Magnesium Production and Processing
NF3
Semiconductor Manufacture
Net Change in
Total Emissions'1
Percent Change
NC
NC
NC
NC
0.5
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
71.6
9.2 %H
NCB
NcB
NcB
+
i.iB
NcB
i
M:B
1.9!
L9I
NcB
NcB
NcB
NcB
NcB
NcB
NcB
NcB
NcB
NcB
NcB
NC
1275.8
30.2%
NC
NC
NC
+
(1.0)
NC
NC
NC
NC
NC
9.1
9.1
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
188.5
21.6%
NC
NC
NC
+
(5.2)
NC
NC
NC
NC
NC
8.7
8.7
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
197.4
22.4%
NC
NC
NC
+
(3.1)
NC
NC
NC
NC
NC
7.8
7.8
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
199.6
22.7%
NC
NC
NC
+
1.0
NC
0.1
NC
NC
NC
6.6
6.6
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
199.4
22.6%
NC
NC
NC
+
(1.0)
NC
NC
NC
NC
2.1
2.1
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
Note: Totals may not sum due to independent rounding. Parentheses indicate negative values.
NC (No Change)
+ Absolute value does not exceed 0.05 MMT CCh Eq. or 0.05 percent
a The values presented in this table for Natural Gas Systems are from the previous Inventory and do not reflect updates to
emission estimates for this category. Please see the Energy Chapter, Section 3.7, Natural Gas Systems.
b The values presented in this table for Petroleum Systems are from the previous Inventory and do not reflect updates to the
emission estimates for this category. Please see the Energy Chapter, Section 3.6, Petroleum Systems.
c Not included in emissions total.
d Excludes net CCh flux from Land Use, Land-Use Change, and Forestry, and emissions from International Bunker Fuels.
Recalculations and Improvements 9-5
-------
1 Table 9-2: Revisions to Total Net Flux from Land Use, Land-Use Change, and Forestry (MMT
2 CO2 Eq.)
Component: Total Net Flux from
Land Use, Land-Use Change, and
Forestry3
Forest Land Remaining Forest Land
Land Converted to Forest Land
Cropland Remaining Cropland
Land Converted to Cropland
Grassland Remaining Grassland
Land Converted to Grassland
Settlements Remaining Settlements
Other (Landfilled Yard Trimmings and
Food Scraps)
Net Change in LULUCF Total Net
Flux3
Percent Change
1990
63.4
NC*
22.0
(1.7)
(n.o*
111
NC
NC
71.6
9.2%B
2005
274.7
NC*
11.5
(5.2)
(1.3)
(3.8)
NC
NC
275.sl
30.2 %|
2010
180.4
NC*
21.2
(0.6)
(9.1)
(3.5)
NC
+
188.5
21.6%
2011
195.8
NC*
5.7
(2.0)
(0.4)
(2.1)
NC
0.5
197.4
22.4%
2012
196.4
NC*
6.3
(1.6)
0.2
(2.1)
NC
0.5
199.6
22.7%
2013
195.6
NC*
6.6
(1.4)
(0.2)
(2.2)
NC
0.9
199.4
22.6%
Average
Annual
Change
144.7
NC*
8.4
(4.3)
(4.3)
(3.4)
NC
0.1
NC (No Change)
Note: Numbers in parentheses indicate an increase in C sequestration.
+ Absolute value does not exceed 0.05 MMT CCh Eq. or 0.05 percent
* Indicates a new source for the current Inventory year
a Total net flux from LULUCF includes the positive C sequestration reported for Forest Land Remaining Forest
Land, Land Converted to Forest Land, Cropland Remaining Cropland, Land Converted to Grassland, Settlements
Remaining Settlements, and Other Land plus the loss in C sequestration reported for Land Converted to Cropland
and Grassland Remaining Grassland.
Note: Totals may not sum due to independent rounding.
9-6 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
10. References
Executive Summary
3 BEA (2015) 2014 Comprehensive Revision of the National Income and Product Accounts: Current-dollar and
4 "real" GDP, 1929-2014. Bureau of Economic Analysis (BEA), U.S. Department of Commerce, Washington, D.C.
5 Available online at .
6 Carbon Dioxide Information Analysis Center (CDIAC) (2014) Recent Greenhouse Gas Concentrations. February
7 2014. Available online at: .
8 EIA (2015a) Electricity Generation. Monthly Energy Review, December 2015. Energy Information Administration,
9 U.S. Department of Energy, Washington, D.C. DOE/EIA-0035(2015/12).
10 EIA (2015b) Electricity in the United States. Electricity Explained. Energy Information Administration, U.S.
11 Department of Energy, Washington, D.C. Available online at
12 .
13 EIA (2013) International Energy Statistics 2013. Energy Information Administration (EIA), U.S. Department of
14 Energy. Washington, D.C. Available online at
15 . 30 November 2014.
16 EPA (2015a) 1970 - 2014 Average annual emissions, all criteria pollutants in MS Excel. National Emissions
17 Inventory (NEI) Air Pollutant Emissions Trends Data. Office of Air Quality Planning and Standards. Last Modified
18 March 2015. Available online at .
19 EPA (2015b) Advancing Sustainable Materials Management: Facts and Figures 2013. June 2015. Available online
20 at.
21 IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
22 Assessment Report of the Intergovernmental Panel on Climate Change. [Stacker, T.F., D. Qin, G.-K., Plattner, M.
23 Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
24 Cambridge, United Kingdom and New York, NY, US A, 153 5 pp.
25 IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
26 Assessment Report of the Intergovernmental Panel on Climate Change. S. Solomon, D. Qin, M. Manning, Z. Chen,
27 M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.). Cambridge University Press. Cambridge, United
28 Kingdom 996 pp.
29 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
30 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
31 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
32 IPCC (2001) Climate Change 2001: The Scientific Basis. Intergovernmental Panel on Climate Change. J.T.
33 Houghton, Y. Ding, D.J. Griggs, M. Noguer, P.J. van der Linden, X. Dai, C.A. Johnson, and K. Maskell (eds.).
34 Cambridge University Press. Cambridge, United Kingdom.
References 10-1
-------
1 IPCC (1996) Climate Change 1995: The Science of Climate Change. Intergovernmental Panel on Climate Change.
1 J.T. Houghton, L.G. Meira Filho, B.A. Callander, N. Harris, A. Kattenberg, and K. Maskell (eds.). Cambridge
3 University Press. Cambridge, United Kingdom.
4 NOAA/ESRL (2016) Trends in Atmospheric Carbon Dioxide. Available online at
5 . 5 February 2016.
6 UNFCCC (2014) Report of the Conference of the Parties on its nineteenth session, held in Warsaw from 11 to 23
7 November 2013. (FCCC/CP/2013/10/Add.3). January 31, 2014. Available online at
8 .
9 U.S. Census Bureau (2015) U.S. Census Bureau International Database (IDE). Available online at
10 .
11
Introduction
12 CDIAC (2014) Recent Greenhouse Gas Concentrations. TJ. Biasing; DOI: 10.3334/CDIAC/atg.032. Available
13 online at . 11 November 2014.
14 EPA (2009) Technical Support Document for the Endangerment and Cause or Contribute Findings for Greenhouse
15 Gases Under Section 202(a) of the Clean Air Act. U.S. Environmental Protection Agency. December 2009.
16 IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
17 Assessment Report of the Intergovernmental Panel on Climate Change [Stacker, T.F., D. Qin, G.-K. Plattner, M.
18 Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
19 Cambridge, United Kingdom and New York, NY, US A, 153 5 pp.
20 IPCC (2007; Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
21 Assessment Report of the Intergovernmental Panel on Climate Change. S. Solomon, D. Qin, M. Manning, Z. Chen,
22 M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.). Cambridge University Press. Cambridge, United
23 Kingdom 996 pp.
24 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
25 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
26 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
27 IPCC (2001) Climate Change 2001: The Scientific Basis. Intergovernmental Panel on Climate Change. J.T.
28 Houghton, Y. Ding, D.J. Griggs, M. Noguer, P.J. van der Linden, X. Dai, C.A. Johnson, and K. Maskell (eds.).
29 Cambridge University Press. Cambridge, United Kingdom.
30 IPCC (1999) Aviation and the Global Atmosphere. Intergovernmental Panel on Climate Change. J.E. Penner, et al.
31 (eds.). Cambridge University Press. Cambridge, United Kingdom.
32 IPCC/TEAP (2005) Special Report: Safeguarding the Ozone Layer and the Global Climate System, Chapter 4:
33 Refrigeration. 2005. Available online .
35 Jacobson, M.Z. (2001) "Strong Radiative Heating Due to the Mixing State of Black Carbon in Atmospheric
36 Aerosols." Nature, 409:695-697.
37 NOAA (2014) Vital Signs of the Planet. Available online at . 12 December 2014.
38 NOAA/ESRL (2015) Trends in Atmospheric Carbon Dioxide. Available online at
39 . 6 February 2015.
40 UNEP/WMO (1999) Information Unit on Climate Change. Framework Convention on Climate Change. Available
41 online at .
42 UNFCCC (2014) Report of the Conference of the Parties on its nineteenth session, held in Warsaw from 11 to 23
43 November 2013. United Nations Framework Convention on Climate Change, Warsaw. (FCCC/CP/2013/10/Add.3).
44 January 31, 2014. Available online at < http://unfccc.int/resource/docs/2013/copl9/eng/10a03.pdf>
10-2 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
i Trends in Greenhouse Gas Emissions
2 BEA (2015) 2014 Comprehensive Revision of the National Income and Product Accounts: Current-dollar and
3 "real" GDP, 1929-2014. Bureau of Economic Analysis (BEA), U.S. Department of Commerce, Washington, D.C.
4 Available online at .
5 Duffield, J. (2006) Personal communication. Jim Duffield, Office of Energy Policy and New Uses, USDA and
6 Lauren Flinn, ICF International. December 2006.
7 EIA (2015) Monthly Energy Review, December 2015. Energy Information Administration, U.S. Department of
8 Energy, Washington, D.C. DOE/EIA-0035(2015/12).
9 EPA (2015a) Light-Duty Automotive Technology, Carbon Dioxide Emissions, and Fuel Economy Trends: 1975 -
10 2014. Office of Transportation and Air Quality, U.S. Environmental Protection Agency. Available online at <
11 http://www.epa.gov/otaq/fetrends-complete.htm >.
12 EPA (2015b) "1970 - 2014 Average annual emissions, all criteria pollutants in MS Excel." National Emissions
13 Inventory (NEI) Air Pollutant Emissions Trends Data. Office of Air Quality Planning and Standards, March 2015.
14 Available online at .
15 FRB (2015) Industrial Production and Capacity Utilization. Federal Reserve Statistical Release, G.17, Federal
16 Reserve Board. Available online at . July 21, 2015.
17 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
18 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
19 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
20 IPCC (2001) Climate Change 2001: The Scientific Basis. Intergovernmental Panel on Climate Change. J.T.
21 Houghton, Y. Ding, D.J. Griggs, M. Noguer, P.J. van der Linden, X. Dai, C.A. Johnson, and K. Maskell (eds.).
22 Cambridge University Press. Cambridge, United Kingdom.
23 U.S. Census Bureau (2015) U.S. Census Bureau International Database (IDE). Available online at
24 .
25
Energy
29
26 EIA (2013) Indicators: CO2 Emissions. International Energy Statistics 2013. Energy Information Administration,
27 U.S. Department of Energy. Washington, D.C. Available at
28 .
Carbon Dioxide Emissions from Fossil Fuel Combustion
30 AAR (2008 through 2015) Railroad Facts. Policy and Economics Department, Association of American Railroads,
31 Washington, D.C. Obtained from Clyde Crimmel at AAR.
32 AISI (2004 through 2013) Annual Statistical Report, American Iron and Steel Institute, Washington, D.C.
3 3 APTA (2007 through 2015) Public Transportation Fact Book. American Public Transportation Association,
34 Washington, D.C. Available online at .
35 APTA (2006) Commuter Rail National Totals. American Public Transportation Association, Washington, D.C.
36 Available online at .
37 BEA (2015) Table 1.1.6. Real Gross Domestic Product, Chained 2009 Dollars. Bureau of Economic Analysis
38 (BEA), U.S. Department of Commerce, Washington, D.C. December, 2015. Available online at
39 .
References 10-3
-------
1 Benson, D. (2002 through 2004) Unpublished data. Upper Great Plains Transportation Institute, North Dakota State
2 University and American Short Line & Regional Railroad Association.
3 Coffeyville Resources Nitrogen Fertilizers (2014) Nitrogen Fertilizer Operations. Available online at
4 .
5 Dakota Gasification Company (2006) CO2 Pipeline Route and Designation Information. Bismarck, ND. Available
6 online at .
7 DHS (2008) Email Communication. Elissa Kay, Department of Homeland Security and Joe Aamidor, ICF
8 International. January 11,2008.
9 DLA Energy (2015) Unpublished data from the Fuels Automated System (FAS). Defense Logistics Agency Energy,
10 U.S. Department of Defense. Washington, D.C.
11 DOC (1991 through 2015) Unpublished Report of Bunker Fuel Oil Laden on Vessels Cleared for Foreign Countries.
12 Form-563. Foreign Trade Division, Bureau of the Census, U.S. Department of Commerce. Washington, D.C.
13 DOE (1993 through 2015) Transportation Energy Data Book. Office of Transportation Technologies, Center for
14 Transportation Analysis, Energy Division, Oak Ridge National Laboratory. ORNL-6978.
15 DOE (2012) 2010 Worldwide Gasification Database. National Energy Technology Laboratory and Gasification
16 Technologies Council. Available online at
17 . Accessed on 15 March
18 2012.
19 DOT (1991 through 2015) Airline Fuel Cost and Consumption. U.S. Department of Transportation, Bureau of
20 Transportation Statistics, Washington, D.C. DAI-10. http://www.transtats.bts.gov/fuel.asp.
21 Eastman Gasification Services Company (2011) Project Data on Eastman Chemical Company's Chemicals -fro m-
22 Coal Complex in Kingsport, TN. Available online at
23 .
24 EIA (2015a) Monthly Energy Review, December 2015, Energy Information Administration, U.S. Department of
25 Energy, Washington, DC. DOE/EIA-0035(2015/12).
26 EIA (2015b) Natural Gas Annual 2014. Energy Information Administration, U.S. Department of Energy.
27 Washington, D.C. DOE/EIA-0131(06).
28 EIA (2015c) Quarterly Coal Report: January - March 2015. Energy Information Administration, U.S. Department
29 of Energy. Washington, D.C. DOE/EIA-0121.
30 EIA (2015d) U.S. Energy-Related Carbon Dioxide Emissions, 2013. Energy Information Administration, U.S.
31 Department of Energy. Washington, D.C. October 2014. Available online at
32 .
33 EIA (1991 through 2015) Fuel Oil and Kerosene Sales. Energy Information Administration, U.S. Department of
34 Energy. Washington, D.C. Available online at: .
35 EIA (2014) Indicators: CO2 Emissions. International Energy Statistics 2014. Energy Information Administration,
36 U.S. Department of Energy. Washington, D.C. Available at:
37 .
38 EIA (2013) Alternative Fuels Data Tables. Energy Information Administration, U.S. Department of Energy.
39 Washington, D.C. Available online at .
40 EIA (2009a) Emissions of Greenhouse Gases in the United States 2008, Draft Report. Office of Integrated Analysis
41 and Forecasting, Energy Information Administration, U.S. Department of Energy. Washington, D.C. DOE-EIA-
42 0573(2009).
43 EIA (2009b) Manufacturing Consumption of Energy 2006. Energy Information Administration, U.S. Department of
44 Energy. Washington, D.C. Released July, 2009.
45 EIA (2008) Historical Natural Gas Annual, 1930 - 2008. Energy Information Administration, U.S. Department of
46 Energy. Washington, D.C.
10-4 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 EIA (2007) Personal Communication. Joel Lou, Energy Information Administration, and Aaron Beaudette, ICF
2 International. Residual and Distillate Fuel Oil Consumption for Vessel Bunkering (Both International and Domestic)
3 for American Samoa, U.S. Pacific Islands, and Wake Island. October 24, 2007.
4 EIA (2001) U.S. Coal, Domestic and International Issues. Energy Information Administration, U.S. Department of
5 Energy. Washington, D.C. March 2001.
6 EPA (2015a) Acid Rain Program Dataset 1996-2014. Office of Air and Radiation, Office of Atmospheric Programs,
7 U.S. Environmental Protection Agency, Washington, D.C.
8 EPA (2015a) Light-Duty Automotive Technology, Carbon Dioxide Emissions, and Fuel Economy Trends: 1975 -
9 2015. Office of Transportation and Air Quality, U.S. Environmental Protection Agency. Available online at <
10 http://www.epa.gov/otaq/fetrends-complete.htm >.
11 EPA (2015b/ Motor Vehicle Emissions Simulator (Moves) 2014. Office of Transportation and Air Quality, U.S.
12 Environmental Protection Agency. Available online at < http://www.epa.gov/otaq/models/moves/index.htm>.
13 EPA (2010a) Carbon Content Coefficients Developed for EPA's Mandatory Reporting Rule. Office of Air and
14 Radiation, Office of Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
15 Erickson, T. (2003) Plains CO2 Reduction (PCOR) Partnership. Presented at the Regional Carbon Sequestration
16 Partnership Meeting Pittsburgh, Pennsylvania, Energy and Environmental Research Center, University of North
17 Dakota. November 3, 2003. Available online at .
19 FAA (2016) Personal Communication between FAA and Leif Hockstad for aviation emissions estimates from the
20 Aviation Environmental Design Tool (AEDT). February 2016.
21 FHWA (1996 through 2015) Highway Statistics. Federal Highway Administration, U.S. Department of
22 Transportation, Washington, D.C. Report FHWA-PL-96-023-annual. Available online at
23 .
24 Fitzpatrick, E. (2002) The Weyburn Project: A Model for International Collaboration. Available online at
25 .
26 FRB (2015) Industrial Production and Capacity Utilization. Federal Reserve Statistical Release, G.17, Federal
27 Reserve Board. Available online at . March 28, 2014.
28 Gaffney, J. (2007) Email Communication. John Gaffney, American Public Transportation Association and Joe
29 Aamidor, ICF International. December 17, 2007.
30 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
31 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
32 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
33 Jacobs, G. (2010) Personal communication. Gwendolyn Jacobs, Energy Information Administration and Rubaab
34 Bhangu, ICF International. U.S. Territories Fossil Fuel Consumption, 1990-2013. Unpublished. U.S. Energy
35 Information Administration. Washington, D.C.
36 Marland, G. and A. Pippin (1990) "United States Emissions of Carbon Dioxide to the Earth's Atmosphere by
37 Economic Activity." Energy Systems and Policy, 14(4):323.
38 SAIC/EIA (2001) Monte Carlo Simulations of Uncertainty in U.S. Greenhouse Gas Emission Estimates. Final
39 Report. Prepared by Science Applications International Corporation (SAIC) for Office of Integrated Analysis and
40 Forecasting, Energy Information Administration, U.S. Department of Energy. Washington, D.C. June 22, 2001.
41 U.S. Census Bureau (2011) Current Industrial Reports Fertilizer Materials and Related Products: 2010 Summary.
42 Available online at .
43 USAA (2014) U.S. Primary Aluminum Production 2013. U.S. Aluminum Association, Washington, D.C. January,
44 2014.
45 USAF (1998) Fuel Logistics Planning. U.S. Air Force: AFPAM23-221. May 1, 1998.
References 10-5
-------
1 United States Geological Survey (USGS) (1994 through 2011) Minerals Yearbook: Lead Annual Report. U.S.
2 Geological Survey, Reston, VA.
3 USGS (1991 through 20 ll)Minerals Yearbook: Manufactured Abrasives Annual Report. U.S. Geological Survey,
4 Reston, VA.
5 USGS (2011) 2010Mineral Yearbook; Aluminum [AdvancedRelease]. U.S. Geological Survey, Reston, VA.
6 USGS (1991 through 2010a) Minerals Yearbook: Silicon Annual Report. U.S. Geological Survey, Reston, VA.
7 USGS (1991 through 201 Ob) Mineral Yearbook: Titanium Annual Report. U.S. Geological Survey, Reston, VA.
8 USGS (2010) 2009Mineral Commodity Summaries: Aluminum. U.S. Geological Survey, Reston, VA.
9 USGS (2009) 2008Mineral Yearbook: Aluminum. U.S. Geological Survey, Reston, VA.
10 USGS (2007) 2006Mineral Yearbook: Aluminum. U.S. Geological Survey, Reston, VA.
11 USGS (1995, 1998, 2000 through 2002) Mineral Yearbook: Aluminum Annual Report. U.S. Geological Survey,
12 Reston, VA.
13 Whorton, D. (2006 through 2014) Personal communication, Class II and III Rail energy consumption, American
14 Short Line and Regional Railroad Association.
15
35
Stationary Combustion (excluding CO2)
16 EIA (2015a) Supplemental Tables on Petroleum Product detail. Monthly Energy Review, December 2015, Energy
17 Information Administration, U.S. Department of Energy, Washington, D.C. DOE/EIA-0035(2015/12).
18 EIA (2015b) Electricity in the United States. Electricity Explained. Energy Information Administration, U.S.
19 Department of Energy, Washington, D.C. Available online at
20 .
21 EPA (2015) "1970 - 2014 Average annual emissions, all criteria pollutants in MS Excel." National Emissions
22 Inventory (NEI) Air Pollutant Emissions Trends Data. Office of Air Quality Planning and Standards. Available
23 online at .
24 EPA (2003) E-mail correspondance. Air pollutant data. Office of Air Pollution to the Office of Air Quality Planning
25 and Standards, U.S. Environmental Protection Agency (EPA). December 22, 2003.
26 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
27 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
28 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
29 Jacobs, G. (2010) Personal communication. Gwendolyn Jacobs, Energy Information Administration and Rubaab
30 Bhangu, ICF International. U.S. Territories Fossil Fuel Consumption, 1990-2009. Unpublished. U.S. Energy
31 Information Administration. Washington, D.C.
32 SAIC/EIA (2001) Monte Carlo Simulations of Uncertainty in U.S. Greenhouse Gas Emission Estimates. Final
33 Report. Prepared by Science Applications International Corporation (SAIC) for Office of Integrated Analysis and
34 Forecasting, Energy Information Administration, U.S. Department of Energy. Washington, D.C. June 22, 2001.
Mobile Combustion (excluding CO2)
36 AAR (2008 through 2015) Railroad Facts. Policy and Economics Department, Association of American Railroads,
37 Washington, D.C. Obtained from Clyde Crimmel at AAR.
38 ANL (2006) Argonne National Laboratory (2006) GREET model Version 1.7. June 2006.
39 ANL (2015) The Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation Model
40 (GREET_1_2015). Argonne National Laboratory. October 2015. Available at .
41 APTA (2007 through 2015) Public Transportation Fact Book. American Public Transportation Association,
42 Washington, D.C. Available online at .
10-6 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 APTA (2006) Commuter Rail National Totals. American Public Transportation Association, Washington, D.C.
2 Available online at .
3 Benson, D. (2002 through 2004) Personal communication. Unpublished data developed by the Upper Great Plains
4 Transportation Institute, North Dakota State University and American Short Line & Regional Railroad Association.
5 BEA (1991 through 2015) Unpublished BE-36 survey data. Bureau of Economic Analysis, U.S. Department of
6 Commerce. Washington, D.C.
7 Browning, L. (2015) "Methodology for Highway Vehicle Alternative Fuel GHG Estimates". Technical Memo,
8 December 2015.
9 Browning, L. (2009) Personal communication with Lou Browning, "Suggested New Emission Factors for Marine
10 Vessels.", ICF International.
11 Browning, L. (2005) Personal communication with Lou Browning, Emission control technologies for diesel
12 highway vehicles specialist, ICF International.
13 DHS (2008) Email Communication. Elissa Kay, Department of Homeland Security and Joe Aamidor, ICF
14 International. January 11, 2008.
15 DLA Energy (2015) Unpublished data from the Defense Fuels Automated Management System (DFAMS). Defense
16 Energy Support Center, Defense Logistics Agency, U.S. Department of Defense. Washington, D.C.
17 DOC (1991 through 2015) Unpublished Report of Bunker Fuel Oil Laden on Vessels Cleared for Foreign Countries.
18 Form-563. Foreign Trade Division, Bureau of the Census, U.S. Department of Commerce. Washington, D.C.
19 DOE (1993 through 2015) Transportation Energy Data Book. Office of Transportation Technologies, Center for
20 Transportation Analysis, Energy Division, Oak Ridge National Laboratory. ORNL-6978.
21 DOT (1991 through 2015) Airline Fuel Cost and Consumption. U.S. Department of Transportation, Bureau of
22 Transportation Statistics, Washington, D.C. DAI-10. Available online at: .
23 EDTA (2015) Electric Drive Sales Dashboard. Electric Drive Transportation Association, Washington, D.C.
24 Available at: .
25 EIA (2015). Monthly Energy Review, December 2015, Energy Information Administration, U.S. Department of
26 Energy, Washington, D.C. DOE/EIA-0035(2015/12).
27 EIA (1991 through 2015) Fuel Oil and Kerosene Sales. Energy Information Administration, U.S. Department of
28 Energy. Washington, D.C. Available at: http://www.eia.gov/petroleum/fueloilkerosene/
29 EIA (2007 through 2015) Natural Gas Annual. Energy Information Administration, U.S. Department of Energy,
30 Washington, D.C. DOE/EIA-0131(11).
31 EIA (2011) Annual Energy Review 2010. Energy Information Administration, U.S. Department of Energy,
32 Washington, D.C. DOE/EIA-0384(2011). October 19, 2011.
3 3 EIA (2015a) "Table 3.1: World Petroleum Supply and Disposition." International Energy Annual. Energy
34 Information Administration, U.S. Department of Energy. Washington, D.C. Available online at
35 .
36 EIA (2007) Personal Communication. Joel Lou, Energy Information Administration and Aaron Beaudette, ICF
37 International. Residual and Distillate Fuel Oil Consumption for Vessel Bunkering (Both International and Domestic)
38 for American Samoa, U.S. Pacific Islands, and Wake Island. October 24, 2007.
39 EIA (2002) Alternative Fuels Data Tables. Energy Information Administration, U.S. Department of Energy,
40 Washington, D.C. Available online at .
41 EPA (2015b/ Motor Vehicle Emissions Simulator (Moves) 2014. Office of Transportation and Air Quality, U.S.
42 Environmental Protection Agency. Available online at .
43 EPA (2015c) Annual Certification Test Results Report. Office of Transportation and Air Quality, U.S.
44 Environmental Protection Agency. Available online at .
References 10-7
-------
1 EPA (2015d) Confidential Engine Family Sales Data Submitted To EPA By Manufacturers. Office of
2 Transportation and Air Quality, U.S. Environmental Protection Agency.
3 EPA (2015) "1970 - 2014 Average annual emissions, all criteria pollutants in MS Excel." National Emissions
4 Inventory (NEI) Air Pollutant Emissions Trends Data. Office of Air Quality Planning and Standards. Available
5 online at .
6 EPA (2000) Mobile6 Vehicle Emission Modeling Software. Office of Mobile Sources, U.S. Environmental
7 Protection Agency, Ann Arbor, Michigan.
8 EPA (1999a) Emission Facts: The History of Reducing Tailpipe Emissions. Office of Mobile Sources. May 1999.
9 EPA 420-F-99-017. Available online at .
10 EPA (1999b) Regulatory Announcement: EPA's Program for Cleaner Vehicles and Cleaner Gasoline. Office of
11 Mobile Sources. December 1999. EPA420-F-99-051. Available online at .
13 EPA (1998) Emissions of Nitrous Oxide from Highway Mobile Sources: Comments on the Draft Inventory of U.S.
14 Greenhouse Gas Emissions and Sinks, 1990-1996. Office of Mobile Sources, Assessment and Modeling Division,
15 U.S. Environmental Protection Agency. August 1998. EPA420-R-98-009.
16 EPA (1997) Mobile Source Emission Factor Model (MOBILESa). Office of Mobile Sources, U.S. Environmental
17 Protection Agency, Ann Arbor, Michigan.
18 EPA (1994a) Automobile Emissions: An Overview. Office of Mobile Sources. August 1994. EPA 400-F-92-007.
19 Available online at .
20 EPA (1994b) Milestones in Auto Emissions Control. Office of Mobile Sources. August 1994. EPA 400-F-92-014.
21 Available online at .
22 EPA (1993) Automobiles and Carbon Monoxide. Office of Mobile Sources. January 1993. EPA 400-F-92-005.
23 Available online at .
24 Esser, C. (2003 through 2004) Personal Communication with Charles Esser, Residual and Distillate Fuel Oil
25 Consumption for Vessel Bunkering (Both International and Domestic) for American Samoa, U.S. Pacific Islands,
26 and Wake Island.
27 FAA (2016) Personal Communication between FAA and Leif Hockstad for aviation emissions estimates from the
28 Aviation Environmental Design Tool (AEDT). February 2016.
29 FHWA (1996 through 2015) Highway Statistics. Federal Highway Administration, U.S. Department of
30 Transportation, Washington, D.C. Report FHWA-PL-96-023-annual. Available online at
31 .
32 Gaffney, J. (2007) Email Communication. John Gaffney, American Public Transportation Association and Joe
33 Aamidor, ICF International. December 17, 2007.
34 ICF (2006a) Revised Gasoline Vehicle EFsfor LEV and Tier 2 Emission Levels. Memorandum from ICF
35 International to John Davies, Office of Transportation and Air Quality, U.S. Environmental Protection Agency.
36 November 2006.
37 ICF (2006b) Revisions to Alternative Fuel Vehicle (AFV) Emission Factors for the U.S. Greenhouse Gas Inventory.
38 Memorandum from ICF International to John Davies, Office of Transportation and Air Quality, U.S. Environmental
39 Protection Agency. November 2006.
40 ICF (2004) Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles. Final Report to U.S.
41 Environmental Protection Agency. February 2004.
42 Lipman, T. and M. Delucchi (2002) "Emissions of Nitrous Oxide and Methane from Conventional and Alternative
43 Fuel Motor Vehicles." Climate Change, 53:477-516.
44 Santoni, G., B. Lee, E. Wood, S. Herndon, R. Miake-Lye, S Wofsy, J. McManus, D. Nelson, M. Zahniser (2011)
45 Aircraft emissions of methane and nitrous oxide during the alternative aviation fuel experiment. Environ Sci
46 Technol. 2011 Aug 15; 45(16):7075-82.
10-8 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 U.S. Census Bureau (2000) Vehicle Inventory and Use Survey. U.S. Census Bureau, Washington, D.C. Database
2 CD-EC97-VIUS.
3 Whorton, D. (2006 through 2014) Personal communication, Class II and III Rail energy consumption, American
4 Short Line and Regional Railroad Association.
5 Carbon Emitted from Non-Energy Uses of Fossil Fuels
6 ACC (2015) "PIPS Year-End Resin Statistics for 2014 vs. 2013: Production, Sales and Captive Use." Available
7 online at .
9 ACC (2014a) "U.S. Resin Production & Sales: 2013 vs. 2012," American Chemistry Council. Available online at:
10 .
19 Bank of Canada (2013) Financial Markets Department Year Average of Exchange Rates. Available online at
20 .
21 Bank of Canada (2012) Financial Markets Department Year Average of Exchange Rates. Available online at
22 .
23 EIA (2015a) Supplemental Tables on Petroleum Product detail. Monthly Energy Review, December 2015. Energy
24 Information Administration, U.S. Department of Energy, Washington, D.C. DOE/EIA-0035(2015/12).
25 EIA (2015b) Supplemental Tables on Petroleum Product detail. Monthly Energy Review, December 2015. Energy
26 Information Administration, U.S. Department of Energy, Washington, D.C. DOE/EIA-0035(2015/12).
27 EIA (2013) EIA Manufacturing Consumption of Energy (MECS) 2010. U.S. Department of Energy, Energy
28 Information Administration, Washington, D.C.
29 EIA (2010) EIA Manufacturing Consumption of Energy (MECS) 2006. U.S. Department of Energy, Energy
30 Information Administration, Washington, D.C.
31 EIA (2005) EM Manufacturing Consumption of Energy (MECS) 2002. U.S. Department of Energy, Energy
32 Information Administration, Washington, D.C.
33 EIA (2001) EM Manufacturing Consumption of Energy (MECS) 1998. U.S. Department of Energy, Energy
34 Information Administration, Washington, D.C.
35 EIA (1997) EM Manufacturing Consumption of Energy (MECS) 1994. U.S. Department of Energy, Energy
36 Information Administration, Washington, D.C.
37 EIA (1994) EM Manufacturing Consumption of Energy (MECS) 1991. U.S. Department of Energy, Energy
38 Information Administration, Washington, D.C.
39 EPA (2015a) "1970 - 2014 Average annual emissions, all criteria pollutants in MS Excel." National Emissions
40 Inventory (NEI) Air Pollutant Emissions Trends Data. Office of Air Quality Planning and Standards, March 2015.
41 Available online at .
42 EPA (2015b). Resource Conservation and Recovery Act (RCRA) Info, Biennial Report, GM Form (Section 2-
43 Onsite Management) and WR Form.
References 10-9
-------
1 EPA (2014a) Municipal Solid Waste in the United States: 2012 Facts and Figures. Office of Solid Waste and
2 Emergency Response, U.S. Environmental Protection Agency, Washington, D.C. Available online at
3 .
4 EPA (2014b) Chemical Data Access Tool (CDAT). U.S. Environmental Protection Agency, June 2014. Available
5 online at . Accessed January 2015.
6 EPA (2013a) Municipal Solid Waste in the United States: 2011 Facts and Figures. Office of Solid Waste and
7 Emergency Response, U.S. Environmental Protection Agency, Washington, D.C. Available online at
8 .
9 EPA (2013b). Resource Conservation and Recovery Act (RCRA) Info, Biennial Report, GM Form (Section 2-
10 Onsite Management) and WR Form.
11 EPA (2011) EPA's Pesticides Industry Sales and Usage, 2006 and 2007 Market Estimates. Available online at
12 .
13 EPA (2009) Biennial Reporting System (BRS) Database. U.S. Environmental Protection Agency, Envirofacts
14 Warehouse. Washington, D.C. Available online at . Data for 2001-2007 are
15 current as of Sept. 9, 2009.
16 EPA (2004) EPA's Pesticides Industry Sales and Usage, 2000 and 2001 Market Estimates. Available online at
17 .
18 EPA (2002) EPA's Pesticides Industry Sales and Usage, 1998 and 1999 Market Estimates, table 3.6. Available
19 online at . Accessed July 2003.
20 EPA (2001) AP 42, Volume I, Fifth Edition. Chapter 11: Mineral Products Industry. Available online at
21 .
22 EPA (2000a) Biennial Reporting System (BRS). U.S. Environmental Protection Agency, Envirofacts Warehouse.
23 Washington, D.C. Available online at .
24 EPA (2000b) Toxics Release Inventory, 1998. U.S. Environmental Protection Agency, Office of Environmental
25 Information, Office of Information Analysis and Access, Washington, D.C. Available online at
26 .
27 EPA (1999) EPA's Pesticides Industry Sales and Usage, 1996-1997 Market Estimates. Available online at
28 .
29 EPA (1998) EPA's Pesticides Industry Sales and Usage, 1994-1995 Market Estimates. Available online at
30 .
31 FEE (2013) Fiber Economics Bureau, as cited in C&EN (2013) Lackluster Year for Chemical Output: Production
32 stayed flat or dipped in most world regions in 2012. Chemical &Engineering News, American Chemical Society, 1
33 July. Available online at .
34 FEE (2012) Fiber Economics Bureau, as cited in C&EN (2012) Too Quiet After the Storm: After a rebound in 2010,
35 chemical production hardly grew in 2011. Chemical & Engineering News, American Chemical Society, 2 July.
36 Available online at .
37 FEB (2011) Fiber Economics Bureau, as cited in C&EN (2011) Output Ramps up in allRegions. Chemical
38 Engineering News, American Chemical Society, 4 July. Available online at .
39 FEB (2010) Fiber Economics Bureau, as cited in C&EN (2010) Output Declines in U.S., Europe. Chemical &
40 Engineering News, American Chemical Society, 6 July. Available online at .
41 FEB (2009) Fiber Economics Bureau, as cited in C&EN (2009) Chemical Output Slipped In Most Regions Chemical
42 & Engineering News, American Chemical Society, 6 July. Available online at .
43 FEB (2007) Fiber Economics Bureau, as cited in C&EN (2007) Gains in Chemical Output Continue. Chemical &
44 Engineering News, American Chemical Society. July 2, 2007. Available online at .
45 FEB (2005) Fiber Economics Bureau, as cited in C&EN (2005) Production: Growth in Most Regions Chemical &
46 Engineering News, American Chemical Society, 11 July. Available online at .
10-10 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 FEB (2003) Fiber Economics Bureau, as cited in C&EN (2003) Production Inches Up in Most Countries, Chemical
2 & Engineering News, American Chemical Society, 7 July. Available online at .
3 FEB (2001) Fiber Economics Bureau, as cited in ACS (2001) Production: slow gains in output of chemicals and
4 products lagged behind U.S. economy as a whole Chemical & Engineering News, American Chemical Society, 25
5 June. Available online at .
6 Financial Planning Association (2006) Canada/US Cross-Border Tools: US/Canada Exchange Rates. Available
7 online at . Accessed August 16, 2006.
8 Gosselin, Smith, and Hodge (1984) "Clinical Toxicology of Commercial Products." Fifth Edition, Williams &
9 Wilkins, Baltimore.
10 Huurman, J. W.F. (2006) 'Recalculation of Dutch Stationary Greenhouse Gas Emissions Based on Sectoral Energy
11 Statistics 1990-2002. Statistics Netherlands, Voorburg, The Netherlands.
12 IISRP (2003) "IISRP Forecasts Moderate Growth in North America to 2007" International Institute of Synthetic
13 Rubber Producers, Inc. New Release. Available online at .
15 IISRP (2000) "Synthetic Rubber Use Growth to Continue Through 2004, Says IISRP and RMA" International
16 Institute of Synthetic Rubber Producers press release.
17 INEGI (2006) Production bruta total de las unidades economicas manufactureras por Subsector, Rama, Subrama y
18 Clase de actividad. Available online at
19 . Accessed August
20 15.
21 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
22 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
23 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
24 Marland, G., and R.M. Rotty (1984) "Carbon dioxide emissions from fossil fuels: A procedure for estimation and
25 results for 1950-1982", Tellus 36b:232-261.
26 NPRA (2002) North American Wax - A Report Card. Available online at
27 .
28 RMA (2014) 2013 U.S. Scrap Tire Management Summary. Rubber Manufacturers Association, Washington, D.C.
29 November 2014.
30 RMA (2011) U.S. Scrap Tire Management Summary: 2005-2009. Rubber Manufacturers Association, Washington,
31 D.C. October 2011, updated September 2013.
32 RMA (2009) "Scrap Tire Markets: Facts and Figures - Scrap Tire Characteristics." Available online at:
33 http://www.rma.org/scrap_tires/scrap_tire_markets/scrap_tire_characteristics/ Accessed 17 September 2009.
34 Schneider, S. (2007) E-mail between Shelly Schneider of Franklin Associates (a division of ERG) and Sarah
35 Shapiro of ICF International, January 10, 2007.
36 U.S. Bureau of the Census (2015) U.S International Trade Commission (USITC) Trade Dataweb. Available online
37 at: < http://dataweb.usitc.gov>.
38 U.S. Census Bureau (2014) 2012 Economic Census. Available online at:
39 . Accessed November 2014.
40 U.S. Census Bureau (2009) Soap and Other Detergent Manufacturing: 2007. Available online at
41 .
43 U.S. Census Bureau (2004) Soap and Other Detergent Manufacturing: 2002, Issued December 2004, EC02-3II-
44 325611 (RV). Available online at .
45 U.S. Census Bureau (1999) Soap and Other Detergent Manufacturing: 1997, Available online at
46 .
References 10-11
-------
1 U.S. International Trade Commission (1990-2015) "Interactive Tariff and Trade DataWeb: Quick Query." Available
2 online at . Accessed December 2015.
3 Incineration of Waste
4 ArSova, Ljupka, Rob van Haaren, Nora Goldstein, Scott M. Kaufman, and Nickolas J. Themelis (2008) "16th
5 Annual BioCycle Nationwide Survey: The State of Garbage in America" Biocycle, JG Press, Emmaus, PA.
6 December.
7 Bahor, B (2009) Covanta Energy's public review comments re: Draft Inventory of U.S. Greenhouse Gas Emissions
8 and Sinks: 1990-2007. Submitted via email on April 9, 2009 to Leif Hockstad, U.S. EPA.
9 Berenyi, E. B. (2014) "A Compatibility Study: Recycling and Waste-to-Energy Work in Concert". Governmental
10 Advisory Associates, Inc. Available online at http://www.energyrecoverycouncil.org/userfiles/files/2014-Berenyi-
11 recycling-study.pdf
12 De Soete, G.G. (1993) "Nitrous Oxide from Combustion and Industry: Chemistry, Emissions and Control." In A. R.
13 Van Amstel, (ed.) Proc. of the International Workshop Methane and Nitrous Oxide: Methods in National Emission
14 Inventories and Options for Control, Amersfoort, NL. February 3-5, 1993.
15 Energy Recovery Council (2009) "2007 Directory of Waste-to-Energy Plants in the United States." Accessed
16 September 29, 2009.
17 EPA (2015) Advancing Sustainable Materials Management: Facts and Figures 2013 - Assessing Trends in Material
18 Generation, Recycling and Disposal in the United States. Office of Solid Waste and Emergency Response, U.S.
19 Environmental Protection Agency. Washington, D.C. Available online at
20 http://www3.epa.gov/epawaste/nonhaz/municipal/pubs/2013_advncng_smm_rpt.pdf.
21 EPA (2007, 2008, 2011, 2013, 2014) Municipal Solid Waste in the United States: Facts and Figures. Office of Solid
22 Waste and Emergency Response, U.S. Environmental Protection Agency. Washington, D.C. Available online at
23 .
24 EPA (2006) Solid Waste Management and Greenhouse Gases: A Life-Cycle Assessment of Emissions and Sinks.
25 Office of Solid Waste and Emergency Response, U.S. Environmental Protection Agency. Washington, D.C.
26 EPA (2000) Characterization of Municipal Solid Waste in the United States: Source Data on the 1999 Update.
27 Office of Solid Waste, U.S. Environmental Protection Agency. Washington, D.C. EPA530-F-00-024.
28 Goldstein, N. and C. Madtes (2001) "13th Annual BioCycle Nationwide Survey: The State of Garbage in America."
29 BioCycles, JG Press, Emmaus, PA. December 2001.
30 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
31 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
32 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
33 Kaufman, et al. (2004) "14th Annual BioCycle Nationwide Survey: The State of Garbage in America 2004"
34 Biocycle, JG Press, Emmaus, PA. January, 2004.
35 RMA (2014) "2013 U.S. Scrap Tire Management Summary." Rubber Manufacturers Association. November 2014.
36 Available online at: .
37 RMA (2012a) "Rubber FAQs." Rubber Manufacturers Association. Available online at . Accessed 19 November 2014.
39 RMA (2012b) "Scrap Tire Markets: Facts and Figures - Scrap Tire Characteristics." Available online at
40 . Accessed 18 January 2012.
41 RMA (2011) "U.S. Scrap Tire Management Summary 2005-2009." Rubber Manufacturers Association. October
42 2011. Available online at: .
43 RTI (2009) Updated Hospital/Medical/Infectious Waste Incinerator (HMIWI) Inventory Database. Memo dated July
44 6, 2009. Available online at: .
10-12 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Schneider, S. (2007) E-mail between Shelly Schneider of Franklin Associates (a division of ERG) and Sarah
2 Shapiro of ICF International, January 10, 2007.
3 Shin, D. (2014) Generation and Disposition of Municipal Solid Waste (MSW) in the United States-A National
4 Survey. Thesis. Columbia University, Department of Earth and Environmental Engineering, January 3, 2014.
5 Shin and Themelis. (2014) "U.S. Survey of Generation and Disposition of Municipal Solid Waste". January 2016.
6 Available online at:
8 Simmons, et al. (2006) "15th Nationwide Survey of Municipal Solid Waste Management in the United States: The
9 State of Garbage in America." BioCycle, JG Press, Emmaus, PA. April 2006.
10 van Haaren, Rob, Themelis, N., and Goldstein, N. (2010) "The State of Garbage in America." BioCycle, October
11 2010. Volume 51, Number 10, pg. 16-23.
12 Coal Mining (TO BE UPDATED)
13 AAPG (1984) Coalbed Methane Resources of the United States. AAPG Studies in Geology Series #17.
14 CAR (2014) Project Database. Climate Action Reserve. Available at .
15 Consol (2014) Ruby Canyon Summary 2013. CONSOL Energy Inc. excel spreadsheet
16 Creedy, D.P. (1993) Chemosphere. Vol. 26, pp. 419-440.
17 EIA (2014) Annual Coal Report 1991-2013 (Formerly called Coal Industry Annual). Table 1. Energy Information
18 Administration, U.S. Department of Energy, Washington, D.C.
19 EPA (2014) Greenhouse Gas Reporting Program (GHGRP): Underground Coal Mines. Retrieved from
20 http://www.epa.gov/ghgreporting/ghgdata/reported/coalmines.html
21 EPA (2005) Surface Mines Emissions Assessment. U.S. Environmental Protection Agency Draft Report.
22 EPA (1996) Evaluation and Analysis of Gas Content and Coal Properties of Major Coal Bearing Regions of the
23 United States. U.S. Environmental Protection Agency. EPA/600/R-96-065.
24 GSA (2014) Well Records Database. Geological Survey of Alabama State Oil and Gas Board. Retrieved from
25 .
26 IEA (2014) Key World Energy Statistics. Coal Production, International Energy Agency.
27 IPCC (2011) Use of Models and Facility-Level Data in Greenhouse Gas Inventories (Report of IPCC Expert
28 Meeting on Use of Models and Measurements in Greenhouse Gas Inventories 9-11 August 2010, Sydney, Australia)
29 eds: EgglestonH.S., Srivastava N., Tanabe K., Baasansuren J., Fukuda M., Pub. IGES, Japan 2011.
3 0 JWR (2014) Wells Intercepted 2013. Jim Walter Resources excel spreadsheet.
31 JWR (2010) No. 4 & 7 Mines General Area Maps. Walter Energy: Jim Walter Resources.
32 King, Brian (1994) Management of Methane Emissions from Coal Mines: Environmental, Engineering, Economic
33 and Institutional Implication of Options, Neil and Gunter Ltd., Halifax, March 1994.
34 MSHA (2014) Data Transparency atMSHA. Mine Safety and Health Administration. Retrieved from
35 .
36 Mutmansky, Jan M. and Yanbei Wang (2000) "Analysis of Potential Errors in Determination of Coal Mine Annual
37 Methane Emissions." Mineral Resources Engineering, 9(4). December 2000.
38 Saghafi, Abouna (2013) Estimation of fugitive emissions from open cut coal mining and measurable gas content,
39 13th Coal Operators' Conference, University of Wollongong, The Australian Institute of Mining and Metallurgy &
40 Mine Managers Association of Australia, 2013, 306-313.
41 USBM (1986) Results of the Direct Method Determination of the Gas Contents of U.S. Coal Basins. Circular 9067,
42 U.S. Bureau of Mines.
References 10-13
-------
1 West Virginia Geological & Economic Survey (WVGES) (2014) Oil & Gas Production Data. Retrieved from
2 .
3 Abandoned Underground Coal Mines (TO BE UPDATED)
4 EPA (2004) Methane Emissions Estimates & Methodology for Abandoned Coal Mines in the U.S. Draft Final
5 Report. Washington, D.C. April 2004.
6 Mutmansky, Jan M., and Yanbei Wang (2000) Analysis of Potential Errors in Determination of Coal Mine Annual
1 Methane Emissions. Department of Energy and Geo-Environmental Engineering, Pennsylvania State University.
8 University Park, PA.
9 U.S. Department of Labor, Mine Health & Safety Administration (2014) Data Retrieval System. Available online at
10 .
11
Petroleum Systems (TO BE UPDATED)
12 Allen et al. (2014) Methane Emissions from Process Equipment at Natural Gas Production Sites in the United
13 States: Pneumatic Controllers. ES&T. December 9, 2014. Available online at:
14 .
15 API (2009) Compendium of Greenhouse gas Emissions Methodologies for the Oil and Gas Industry. American
16 Petroleum Institute. Austin, TX, August 2009.
17 BOEM (2011a) OCS Platform Activity. Bureau of Ocean Energy Management, U.S. Department of Interior.
18 Available online at
19 .
23 BOEM (2011c) Pacific OCS Region. Bureau of Ocean Energy Management, U.S. Department of Interior. Available
24 online at .
25 BOEM (2014) Year 2011 Gulfwide Emission Inventory Study. Bureau of Ocean Energy Management, U.S.
26 Department of Interior. OCS Study BOEM 2014-666. Available online at
27
28 Drillinglnfo (2014) December 2014 Download. DI Desktop® Drillinglnfo, Inc.
29 EIA (1990 through 2014) Refinery Capacity Report. Energy Information Administration, U.S. Department of
30 Energy. Washington, DC. Available online at < http://www.eia.gov/petroleum/refinerycapacity/ >.
31 EIA (1995 through 2014a) Annual Energy Review. Energy Information Administration, U.S. Department of Energy.
32 Washington, DC. Available online at < http://www.eia.gov/totalenergy/data/annual/index.cfm >.
33 EIA (1995 through 2014b) Monthly Energy Review. Energy Information Administration, U.S. Department of
34 Energy. Washington, DC. Available online at < http://www.eia.gov/totalenergy/data/monthly/index.cfm >.
35 EIA (1995 through 2014c) Petroleum Supply Annual. Volume 1. U.S Department of Energy Washington, DC.
36 Available online at: < http://www.eia.gov/petroleum/supply/annual/volumel/>.
37 EPA (2015a) Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2013: Update to Data Source for Well
38 Counts. Available at http://www.epa.gov/climatechange/ghgemissions/usinventoryreport/natural-gas-systems.html.
39 EPA (2015b) Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2013: Update to Offshore Oil and Gas
40 Platforms Emissions Estimate. Available at
41 http://www.epa.gov/climatechange/ghgemissions/usinventoryreport/natural-gas-systems.html.
10-14 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 EPA (2015c) Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2013: Update to Refineries Emissions
2 Estimate. Available at .
4 EPA (2015d) Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2013: Potential Updates to Pneumatic
5 Controller Emissions Estimate. Available at
6 .
7 EPA (2014) Greenhouse Gas Reporting Program. Environmental Protection Agency. Data reported as of August 18,
8 2014.
9 EPA (2005) Incorporating the Mineral Management Service Gulfwide Offshore Activities Data System (GOADS)
10 2000 data into the methane emissions inventories. Prepared by ICF International. U.S. Environmental Protection
11 Agency. 2005.
12 EPA (1999a) Estimates of Methane Emissions from the U.S. Oil Industry (Draft Report). Prepared by ICF
13 International. Office of Air and Radiation, U.S. Environmental Protection Agency. October 1999.
14 EPA (1999b) Methane Emissions from the U.S. Petroleum Industry. Prepared by Radian International. U.S.
15 Environmental Protection Agency. February 1999.
16 EPA/GRI (1996a) Methane Emissions from the Natural Gas Industry, V7: Blow and Purge Activities. Prepared by
17 Radian. U.S. Environmental Protection Agency. April 1996.
18 EPA/GRI (1996b) Methane Emissions from the Natural Gas Industry, VI1: Compressor Driver Exhaust. Prepared
19 by Radian. U.S. Environmental Protection Agency. April 1996.
20 EPA/GRI (1996c) Methane Emissions from the Natural Gas Industry, V12: Pneumatic Devices. Prepared by Radian.
21 U.S. Environmental Protection Agency. April 1996.
22 EPA/GRI (1996d) Methane Emissions from the Natural Gas Industry, VI3: Chemical Injection Pumps. Prepared by
23 Radian. U.S. Environmental Protection Agency. April 1996.
24 HPDI (2011) Production and Permit Data, October 2009.
25 IOGCC (2011) Marginal Wells: fuel for economic growth 2010 Report. Interstate Oil & Gas Compact Commission.
26 Available online at .
27 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
28 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
29 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
30 IPCC (2007) Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II and HI to the Fourth
31 Assessment Report of the Intergovernmental Panel on Climate Change. Pachauri, R.K and Reisinger, A. eds.; IPCC,
32 Geneva, Switzerland.
33 OGJ (2014a) Oil and Gas Journal 1990-2013. Pipeline Economics Issue, September 2014.
34 OGJ (2013b) Oil and Gas Journal 1990-2013. Worldwide Refining Issue, January 2013.
35 United States Army Corps of Engineers (1995 through 2012) Waterborne Commerce of the United States, Part 5:
36 National Summaries. U.S. Army Corps of Engineers. Washington, DC.
37 Natural Gas Systems (TO BE UPDATED)
38 AGA (1991 through 1998) Gas Facts. American Gas Association. Washington, DC.
39 Alabama (2014) Alabama State Oil and Gas Board. Available online at .
40 Allen et al. (2014a) Methane Emissions from Process Equipment at Natural Gas Production Sites in the United
41 States: Liquids Unloading. ES&T. December 9, 2014. Available online at:
42 .
References 10-15
-------
1 Allen et al. (2014b) Methane Emissions from Process Equipment at Natural Gas Production Sites in the United
2 States: Pneumatic Controllers. ES&T. December 9, 2014. Available online at:
3 .
4 Allen etal. (2013) Measurements of methane emissions at natural gas production sites in the United States, doi:
5 10.1073/pnas.l304880110 PNAS September 16, 2013. Available online at
6 .
7 API/ANGA (2012) Characterizing Pivotal Sources of Methane Emissions from Natural Gas Production - Summary
8 and Analysis of API andANGA Survey Responses. Final Report. American Petroleum Institute and America's
9 Natural Gas Alliance. September 21.
10 BOEMRE (201 la) Gulf of Mexico Region Offshore Information. Bureau of Ocean Energy Management, Regulation
11 and Enforcement, U.S. Department of Interior.
12 BOEMRE (201 Ib) Pacific OCS Region Offshore Information. Bureau of Ocean Energy Management, Regulation
13 and Enforcement, U.S. Department of Interior.
14 BOEMRE (20 lie) GOM and Pacific OCS Platform Activity. Bureau of Ocean Energy Management, Regulation and
15 Enforcement, U.S. Department of Interior.
16 BOEMRE (20 lid) Pacific OCS Region. Bureau of Ocean Energy Management, Regulation and Enforcement, U.S.
17 Department of Interior.
18 Drillinglnfo (2014) December 2014 Download. DI Desktop® Drillinglnfo, Inc.
19 EIA (2014a) "Table 1— Summary of natural gas supply and disposition in the United States, 2009-2014." Natural
20 Gas Monthly, Energy Information Administration, U.S. Department of Energy, Washington, DC. Available online at
21 .
22 EIA (2014b) "Table 2—Natural Gas Consumption in the United States, 2009-2014." Natural Gas Monthly, Energy
23 Information Administration, U.S. Department of Energy, Washington, DC. Available online at
24 .
25 EIA (2014c) "Table 7 - Marketed production of natural gas in selected states and the Federal Gulf of Mexico, 2009-
26 2014." Natural Gas Monthly, Energy Information Administration, U.S. Department of Energy, Washington, DC.
27 Available online at .
28 EIA (2014d) U.S. Natural Gas Imports by Country. Energy Information Administration, U.S. Department of Energy,
29 Washington, DC. Available online at .
30 EIA (2014e) Natural Gas Gross Withdrawals and Production. Energy Information Administration, U.S. Department
31 of Energy, Washington, DC. Available online at .
32 EIA (2012a) Formation crosswalk. Energy Information Administration, U.S. Department of Energy, Washington,
33 DC. Provided July 7.
34 EIA (2012b) Lease Condensate Production, 1979-2012, Natural Gas Navigator. Energy Information Administration,
35 U.S. Department of Energy, Washington, DC. Available online at
36 .
37 EIA (2005) "Table 5—U.S. Crude Oil, Natural Gas, and Natural Gas Liquids Reserves, 1977-2003." Energy
38 Information Administration, Department of Energy, Washington, DC.
39 EIA (2004) US LNG Markets and Uses. Energy Information Administration, U.S. Department of Energy,
40 Washington, DC. June 2004.
41 EIA (2001) "Documentation of the Oil and Gas Supply Module (OGSM)." Energy Information Administration, U.S.
42 Department of Energy, Washington, DC.
43 EPA (2015a) Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2013: Update to Data Source for Well
44 Counts. Available at .
10-16 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 EPA (2015b) Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2013: Update to Offshore Oil and Gas
2 Platform Emission Estimates. Available at
3 .
4 EPA (2015c) Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2013: Update to Hydraulically Fractured
5 Gas Well Completions and Workover Estimate. Available at
6 .
7 EPA (2015d) Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2013: Potential Updates to Pneumatic
8 Controller Emissions Estimate. Available at
9 .
10 EPA (2015e) Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2013: Potential Updates to Liquids
11 Unloading Emissions Estimate. Available at
12 .
13 EPA (2014) Greenhouse Gas Reporting Program- Subpart W-Petroleum and Natural Gas Systems. Environmental
14 Protection Agency. Data reported as of August 18, 2014.
15 EPA (2013 a) Oil and Natural Gas Sector: Standards of Performance for Crude Oil and Natural Gas Production,
16 Transmission, and Distribution. Background Supplemental Technical Support Document for the Final New Source
17 Performance Standards. Environmental Protection Agency. September 2013.
18 EPA (2013b) Oil and Natural Gas Sector: New Source Performance Standards and National Emission Standards
19 for Hazardous Air Pollutants Reviews. Environmental Protection Agency, 40 CFR Parts 60 and 63, [EPA-HQ-OAR-
20 2010-0505; FRL-9665-1], RIN 2060-AP76.
21 EPA (2013c) Natural Gas STAR Reductions 1990-2012. Natural Gas STAR Program. September 2013.
22 EPA (2013d) Updating GHG Inventory Estimate for Hydraulically Fractured Gas Well Completions and
23 Workovers. Available online at .
25 EPA (1999) Estimates of Methane Emissions from the U.S. Oil Industry (Draft Report). Prepared by ICF-Kaiser,
26 Office of Air and Radiation, U.S. Environmental Protection Agency. October 1999.
27 EPA/GRI (1996) Methane Emissions from the Natural Gas Industry. Prepared by Harrison, M, T. Shires, J.
28 Wessels, and R. Cowgill, eds., Radian International LLC for National Risk Management Research Laboratory, Air
29 Pollution Prevention and Control Division, Research Triangle Park, NC. EPA-600/R-96-080a.
30 FERC (2014) North American LNG Terminals. Federal Energy Regulatory Commission, Washington, D.C.
31 GTI (2001) Gas Resource Database: Unconventional Natural Gas and Gas Composition Databases. Second Edition.
32 GRI-01/0136.
33 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
34 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
35 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
36 Jackson etal., (2014) Natural Gas Pipeline Leaks Across Washington, D.C., 48 Environ. Science Technology 2051-
37 2058, January 16, 2014. Available online at . March 24, 2014.
38 McGeehan et al., (2014) Beneath Cities, a Decaying Tangle of Gas Pipes, N.Y. Times, March 24, 2014. Available
39 online at .
41 Miller et al. (2013) Anthropogenic emissions of methane in the United States. November 25, 2013, doi:
42 10.1073/pnas.l314392110. Available online at
43 .
44 OGJ (1997-2013) "Worldwide Gas Processing." Oil & Gas Journal, PennWell Corporation, Tulsa, OK. Available
45 online at .
References 10-17
-------
1 Payne, B & Ackley, R., (2013a) "Extended Report and Preliminary Investigation of Ground-Level Ambient
2 Methane Levels in Manhattan, New York, NY" (11 March 2013).
3 Payne, B. & Ackley, R., (2013b) "Report on a Survey of Ground-Level Ambient Methane Levels in the Vicinity of
4 Wyalusing, Bradford County, PA," (Nov. 2013).
5
6 Payne, B. & Ackley, R., (2012) "Report to the Clean Air Council on 8 June, 2012 Field Inspection and Methane
7 Sampling Survey of Parts of Leroy, Granville and Franklin Townships, Bradford County, PA," (2012).
8
9 Peischl, J. et al., (2013) "Quantifying Sources of Methane Using Light Alkenes in the Los Angeles Basin, CA," J.
10 Geophys. Res. Atmos. 118, 4974-4990, doi: 10.1002/jgrd.50413
11 Petron, Gabrielle, et al. (2012) Hydrocarbon Emissions Characterization in the Colorado Front Range: A Pilot
12 Study, Journal of Geophysical Research doi: 10.1029/2011JDO16360.
13 Phillips, N.G., et al., (2012) "Mapping Urban Pipeline Leaks: Methane Levels Across Boston," Environmental
14 Pollution Available online at .
15 PHMSA (2013a) Transmission Annuals Data. Pipeline and Hazardous Materials Safety Administration, U.S.
16 Department of Transportation, Washington, DC. Available online at .
18 PHMSA (2013b) Gas Distribution Annual Data. Pipeline and Hazardous Materials Safety Administration, U.S.
19 Department of Transportation, Washington, DC. Available online at .
21 Wyoming (2013) Wyoming Oil and Gas Conservation Commission. Available online at
22 .
31
23 Energy Sources of Indirect Greenhouse Gases
24 EPA (2015) "1970 - 2014 Average annual emissions, all criteria pollutants in MS Excel." National Emissions
25 Inventory (NEI) Air Pollutant Emissions Trends Data. Office of Air Quality Planning and Standards, March 2015.
26 Available online at .
27 EPA (2003) E-mail correspondence containing preliminary ambient air pollutant data. Office of Air Pollution and
28 the Office of Air Quality Planning and Standards, U.S. Environmental Protection Agency. December 22, 2003.
29 EPA (1997) Compilation of Air Pollutant Emission Factors, AP-42. Office of Air Quality Planning and Standards,
30 U.S. Environmental Protection Agency. Research Triangle Park, NC. October 1997.
International Bunker Fuels
32 Anderson, B.E., et al., Alternative Aviation Fuel Experiment (AAFEX), NASA Technical Memorandum, in press,
33 2011.
34 ASTM (1989) Military Specification for Turbine Fuels, Aviation, Kerosene Types, NATO F-34 (JP-8) and NATO F-
35 35. February 10, 1989. Available online at .
36 Chevron (2000) Aviation Fuels Technical Review (FTR-3). Chevron Products Company, Chapter 2. Available
37 online at .
38 DHS (2008) Personal Communication with Elissa Kay, Residual and Distillate Fuel Oil Consumption (International
39 Bunker Fuels). Department of Homeland Security, Bunker Report. January 11, 2008.
40 DLA Energy (2015) Unpublished data from the Defense Fuels Automated Management System (DFAMS). Defense
41 Energy Support Center, Defense Logistics Agency, U.S. Department of Defense. Washington, D.C.
42 DOC (2015) Unpublished Report of Bunker Fuel Oil Laden on Vessels Cleared for Foreign Countries. Form-563.
43 Foreign Trade Division, Bureau of the Census, U.S. Department of Commerce. Washington, D.C.
10-18 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 DOT (1991 through 2013) Fuel Cost and Consumption. Federal Aviation Administration, Bureau of Transportation
2 Statistics, U.S. Department of Transportation. Washington, D.C. DAI-10.
3 E1A (2015) Monthly Energy Review, December 2015, Energy Information Administration, U.S. Department of
4 Energy, Washington, D.C. DOE/EIA-0035(2015/12).
5 FAA (2016) Personal Communication between FAA and Leif Hockstad for aviation emissions estimates from the
6 Aviation Environmental Design Tool (AEDT). February 2016.
7 FAA (2006) System for assessing Aviation's Global Emission (SAGE) Model. Federal Aviation Administration's
8 Office of Aviation Policy, Planning, and Transportation Topics, 2006.
9 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
10 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
11 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
12 USAF (1998) Fuel Logistics Planning. U.S. Air Force pamphlet AFPAM23-221, May 1, 1998.
13
22
28
Wood Biomass and Ethanol Consumption
14 EIA (2015) Monthly Energy Review, December 2015. Energy Information Administration, U.S. Department of
15 Energy. Washington, D.C. DOE/EIA-0035(2015/12).
16 EPA (2015) Acid Rain Program Dataset 1996-2014. Office of Air and Radiation, Office of Atmospheric Programs,
17 U.S. Environmental Protection Agency, Washington, D.C.
18 EPA(2010) Carbon Content Coefficients Developed for EPA's Mandatory Reporting Rule. Office of Air and
19 Radiation, Office of Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
20 Lindstrom, P. (2006) Personal Communication. Perry Lindstrom, Energy Information Administration and Jean Kim,
21 ICF International.
Industrial Processes and Product Use
23 IPCC (2011) Use of Models and Facility-Level Data in Greenhouse Gas Inventories (Report of IPCC Expert
24 Meeting on Use of Models and Measurements in Greenhouse Gas Inventories 9-11 August 2010, Sydney, Australia)
25 eds: EgglestonH.S., Srivastava N., Tanabe K., Baasansuren J., Fukuda M., Pub. IGES, Japan 2011.
26 EPA (2014) Greenhouse Gas Reporting Program. Developments on Publication of Aggregated Greenhouse Gas
27 Data, November 25, 2014. See http://www.epa.gov/ghgreporting/confidential-business-information-ghg-reporting.
Cement Production
29 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
30 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
31 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
32 U.S. Bureau of Mines (1990 through 1993) Minerals Yearbook: Cement Annual Report. U.S. Department of the
33 Interior, Washington, D.C.
34 United States Geological Survey (USGS) (2015a) Mineral Industry Survey: Cement in June 2015. U.S. Geological
35 Survey, Reston, VA. August, 2015.
36 USGS (2015b) Mineral Commodity Summaries: Cement 2015. U.S. Geological Survey, Reston, VA. January, 2015.
37 USGS (1995 through 2013) Minerals Yearbook - Cement. U.S. Geological Survey, Reston, VA.
38 Van Oss (2013a) 1990-2012 Clinker Production Data Provided by Hendrik van Oss (USGS) via email on November
39 8,2013.
References 10-19
-------
1 Van Oss (2013b) Personal communication. Hendrik van Oss, Commodity Specialist of the U.S. Geological Survey
2 and Gopi Manne, Eastern Research Group, Inc. October 28, 2013.
3 Lime Production
4 Corathers (2015) Personal communication, Lisa Corathers, U.S. Geological Survey and Gopi Manne, Eastern
5 Research Group, Inc. September 22, 2015.
6 EPA (2015) Greenhouse Gas Reporting Program (GHGRP). Aggregation of reported facility level data under
7 Subpart S -National Lime production for calendar years 2010-2014. Office of Air and Radiation, Office of
8 Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
9 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
10 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
11 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
12 Males, E. (2003) Memorandum from Eric Males, National Lime Association to Mr. William N. Irving & Mr. Leif
13 Hockstad, Environmental Protection Agency. March 6, 2003.
14 Miner, R. and B. Upton (2002) Methods for estimating greenhouse gas emissions from lime kilns at kraft pulp mills.
15 Energy. Vol. 27 (2002), p. 729-738.
16 Seeger (2013) Memorandum from Arline M. Seeger, National Lime Association to Mr. Leif Hockstad,
17 Environmental Protection Agency. March 15, 2013.
18 United States Geological Survey (USGS) (1992 through 2014) Minerals Yearbook: Lime. U.S. Geological Survey,
19 Reston, VA.
20
39
Glass Production
21 EPA (2009) Technical Support Document for the Glass Manufacturing Sector: Proposed Rule for Mandatory
22 Reporting of Greenhouse Gases. U.S. Environmental Protection Agency, Washington, D.C.
23 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
24 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
25 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
26 OIT (2002) Glass Industry of the Future: Energy and Environmental Profile of the U.S. Glass Industry. Office of
27 Industrial Technologies, U.S. Department of Energy. Washington, D.C.
28 U.S. Bureau of Mines (1991 and 1993a) Minerals Yearbook: Crushed Stone Annual Report. U.S. Department of the
29 Interior. Washington, D.C.
30 United States Geological Survey (USGS) (2015a) Minerals Industry Surveys; Soda Ash in January 2015. U.S.
31 Geological Survey, Reston, VA. March, 2015.
32 USGS (1995 through 20 \5\3~) Minerals Yearbook: Crushed Stone Annual Report. U.S. Geological Survey, Reston,
33 VA.
34 USGS (1995 through 2014) Minerals Yearbook: Soda Ash Annual Report. U.S. Geological Survey, Reston, VA.
35 Willett (2015) Personal communication, Jason Christopher Willett, U.S. Geological Survey and Gopi Manne,
36 Eastern Research Group, Inc. September 9, 2015.
37 Willett (2014) Personal communication., Jason Christopher Willett, U.S. Geological Survey and Gopi Manne,
38 Eastern Research Group, Inc. September 25, 2014.
Other Process Uses of Carbonates
40 U.S. Bureau of Mines (1991 and 1993a) Minerals Yearbook: Crushed Stone Annual Report. U.S. Department of the
41 Interior. Washington, D.C.
10-20 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 U.S. Bureau of Mines (1990 through 1993b) Minerals Yearbook: Magnesium and Magnesium Compounds Annual
1 Report. U.S. Department of the Interior. Washington, D.C.
3 United States Geological Survey (USGS) (2Ql3a) Magnesium Metal Mineral Commodity Summary for 2013. U.S.
4 Geological Survey, Reston, VA.
5 USGS (1995a through 2015) Miner ah Yearbook: Crushed Stone Annual Report. U.S. Geological Survey, Reston,
6 VA.
7 USGS (1995b through 2012) Minerals Yearbook: Magnesium Annual Report. U.S. Geological Survey, Reston, VA.
8 Willett (2015) Personal communication, Jason Christopher Willett, U.S. Geological Survey and Gopi Manne,
9 Eastern Research Group, Inc. September 9, 2015.
10
Ammonia Production
11 ACC (2015) Business of Chemistry (Annual Data). American Chemistry Council, Arlington, VA.
12 Bark (2004) CoffeyvilleNitrogen Plant Available online at
13 . December 15, 2004.
14 Coffeyville Resources Nitrogen Fertilizers (2012) Nitrogen Fertilizer Operations. Available online at
15 .
16 Coffeyville Resources Nitrogen Fertilizers (2011) Nitrogen Fertilizer Operations. Available online at
17 .
18 Coffeyville Resources Nitrogen Fertilizers (2010) Nitrogen Fertilizer Operations. Available online at
19 .
20 Coffeyville Resources Nitrogen Fertilizers (2009) Nitrogen Fertilizer Operations. Available online at
21 .
22 Coffeyville Resources Nitrogen Fertilizers, LLC (2005 through 2007a) Business Data. Available online at
23 .
24 Coffeyville Resources Nitrogen Fertilizers (2007b) Nitrogen Fertilizer Operations. Available online at
25 .
26 Coffeyville Resources Energy, Inc. (CVR) (2015) CVR Energy, Inc. 2014 Annual Report. Available online at
27 .
28 CVR (2014) CVR Energy, Inc. 2013 Annual Report. Available online at .
29 CVR (2012) CVR Energy, Inc. 2012 Annual Report. Available online at .
30 EFMA (2000a) Best Available Techniques for Pollution Prevention and Control in the European Fertilizer Industry.
31 Booklet No. 1 of 8: Production of Ammonium. Available online at
32 .
33 EFMA (2000b) Best Available Techniques for Pollution Prevention and Control in the European Fertilizer Industry.
34 Booklet No. 5 of 8: Production of Urea and Urea Ammonium Nitrate. Available online at
35 .
36 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
37 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
38 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
39 U.S. Census Bureau (2011) Current Industrial Reports Fertilizer Materials and Related Products: 2010Summary.
40 Available online at .
41 U.S. Census Bureau (2010) Current Industrial Reports Fertilizer Materials and Related Products: 2009 Summary.
42 Available online at .
References 10-21
-------
1 U.S. Census Bureau (2009) Current Industrial Reports Fertilizer Materials and Related Products: 2008 Summary.
1 Available online at .
3 U.S. Census Bureau (2008) Current Industrial Reports Fertilizer Materials and Related Products: 2007 Summary.
4 Available online at .
5 U.S. Census Bureau (2007) Current Industrial Reports Fertilizer Materials and Related Products: 2006Summary.
6 Available online at .
7 U.S. Census Bureau (2006) Current Industrial Reports Fertilizer Materials and Related Products: 2005 Summary.
8 Available online at .
9 U.S. Census Bureau (2004, 2005) Current Industrial Reports Fertilizer Materials and Related Products: Fourth
10 Quarter Report Summary. Available online at .
11 U.S. Census Bureau (1998 through 2003) Current Industrial Reports Fertilizer Materials and Related Products:
12 Annual Reports Summary. Available online at .
13 U.S. Census Bureau (1991 through 1994) Current Industrial Reports Fertilizer Materials Annual Report. Report No.
14 MQ28B. U.S. Census Bureau, Washington, D.C.
15 United State Goelogical Survey (USGS) (2015) 2013 Minerals Yearbook: Nitrogen [Advance Release]. August
16 2015. Available online at < http://minerals.usgs.gov/minerals/pubs/commodity/nitrogen/mybl-2013-nitro.pdf>.
17 USGS (2014) 2012 Minerals Yearbook: Nitrogen [Advance Release]. September 2014. Available online at
18 .
19 USGS (1994 through 2009) Minerals Yearbook: Nitrogen. Available online at
20 .
21 Urea Consumption for Non-Agricultural Purposes
22 EFMA (2000) Best Available Techniques for Pollution Prevention and Control in the European Fertilizer Industry.
23 Booklet No. 5 of 8: Production of Urea and Urea Ammonium Nitrate.
24 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
25 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
26 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
27 TFI (2002) U.S. Nitrogen Imports/Exports Table. The Fertilizer Institute. Available online at
28 . August 2002.
29 U.S. Census Bureau (2001 through 2011) Current Industrial Reports Fertilizer Materials and Related Products:
30 Annual Summary. Available online at < http://www.census.gov/manufacturing/cir/historical_data/index.html >.
31 U.S. Department of Agriculture (2012) Economic Research Service Data Sets, Data Sets, U.S. Fertilizer
32 Imports/Exports: Standard Tables. Available online at .
34 U.S. ITC (2002) United States International Trade Commission Interactive Tariff and Trade DataWeb, Version
35 2.5.0. Available online at . August 2002.
36 United States Geological Survey (USGS) (2014 through 2015) Minerals Yearbook: Nitrogen [Advance Release].
37 Available online at .
38 USGS (1994 through 2009) Minerals Yearbook: Nitrogen. Available online at
39 .
40
Nitric Acid Production
41 Climate Action Reserve (CAR) (2013), Project Report,
42 . Accessed on January 18, 2013.
10-22 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Desai (2012) Personal communication. Mausami Desai, U.S. Environmental Protection Agency, January 25, 2012.
2 EPA (2015) Greenhouse Gas Reporting Program (GHGRP). Aggregation of reported facility level data under
3 Subpart V-National Nitric Acid production for calendar years 2010-2014. Office of Air and Radiation, Office of
4 Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
5 EPA (2013) Draft Nitric Acid Database. U.S. Environmental Protection Agency, Office of Air and Radiation.
6 September, 2010.
7 EPA (2012) Memorandum from Mausami Desai, U.S. EPA to Mr. Bill Herz, The Fertilizer Institute. November 26,
8 2012.
9 EPA (2010) Available and Emerging Technologies for Reducing Greenhouse Gas Emissions from the Nitric Acid
10 Production Industry. Office of Air Quality Planning and Standards, U.S. Environmental Protection Agency.
11 Research Triangle Park, NC. December 2010. Available online at:
12 .
13 EPA (1998) Compilation of Air Pollutant Emission Factors, AP-42. Office of Air Quality Planning and Standards,
14 U.S. Environmental Protection Agency. Research Triangle Park, NC. February 1998.
15 IPCC (2007) Forster, P., V. Ramaswamy, P. Artaxo, T. Berntsen, R. Betts, D.W. Fahey, J. Haywood, J. Lean, D.C.
16 Lowe, G. Myhre, J. Nganga, R. Prinn, G. Raga, M. Schulz and R. Van Dorland, 2007: Changes in Atmospheric
17 Constituents and in Radiative Forcing. In: Climate Change 2007: The Physical Science Basis. Contribution of
18 Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S.,
19 D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M.Tignor and H.L. Miller (eds.)]. Cambridge University
20 Press, Cambridge, United Kingdom and New York, NY, USA.
21
22 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
23 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
24 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
25 U.S. Census Bureau (2010a) Current Industrial Reports. Fertilizers and Related Chemicals: 2009. "Table 1:
26 Summary of Production of Principle Fertilizers and Related Chemicals: 2009 and 2008." June, 2010. MQ325B(08)-
27 5. Available online at .
28 U.S. Census Bureau (20lOb) Personal communication between Hilda Ward (of U.S. Census Bureau) and Caroline
29 Cochran (of ICF International). October 26, 2010 and November 5, 2010.
30 U.S. Census Bureau (2009) Current Industrial Reports. Fertilizers and Related Chemicals: 2008. "Table 1:
31 Shipments and Production of Principal Fertilizers and Related Chemicals: 2004 to 2008." June, 2009. MQ325B(08)-
32 5. Available online at .
33 U.S. Census Bureau (2008) Current Industrial Reports. Fertilizers and Related Chemicals: 2007. "Table 1:
34 Shipments and Production of Principal Fertilizers and Related Chemicals: 2003 to 2007." June, 2008. MQ325B(07)-
35 5. Available online at .
36 Adipic Acid Production
37 ACC (2015) Business of Chemistry (Annual Data). American Chemistry Council, Arlington, VA.
38 C&EN (1995) "Production of Top 50 Chemicals Increased Substantially in 1994." Chemical & Engineering News,
39 73(15): 17. April 10, 1995.
40 C&EN (1994) "Top 50 Chemicals Production Rose Modestly Last Year." Chemical & Engineering News,
41 72(15): 13. April 11, 1994.
42 C&EN (1993) "Top 50 Chemicals Production Recovered Last Year." Chemical & Engineering News, 71(15): 11.
43 April 12, 1993.
44 C&EN (1992) "Production of Top 50 Chemicals Stagnates in 1991." Chemical & Engineering News, 70( 15): 17.
45 April 13, 1992.
46 CMR (2001) "Chemical Profile: Adipic Acid." Chemical Market Reporter. July 16, 2001.
References 10-23
-------
1 CMR (1998) "Chemical Profile: Adipic Acid." Chemical Market Reporter. June 15, 1998.
2 CW (2005) "Product Focus: Adipic Acid." Chemical Week. May 4, 2005.
3 CW (1999) "Product Focus: Adipic Acid/Adiponitrile." Chemical Week, p. 31. March 10, 1999.
4 Desai (201 la) Personal communication. Mausami Desai, U.S. Environmental Protection Agency and Roy Nobel,
5 Ascend Performance Materials, October 18, 2011.
6 Desai (201 Ib) Personal communication. Mausami Desai, U.S. Environmental Protection Agency with Steve Zuiss of
7 Invista, November 18, 2011.
8 Desai (2010) Personal communication. Mausami Desai, U.S. Environmental Protection Agency with Steve Zuiss of
9 Invista, October 15, 2010.
10 EPA (2014 through 2015) Greenhouse Gas Reporting Program. Annual Detailed Data for Additional Industry Types
11 (Adipic Acid Tab). Office of Air and Radiation, Office of Atmospheric Programs, U.S. Environmental Protection
12 Agency, Washington, D.C. Accessed 10/07/2015, Available online at: < http://www2.epa.gov/ghgreporting/ghg-
13 reporting-program-data-sets>.
14 EPA (2012) Analysis of Greenhouse Gas Reporting Program data - Subpart E (Adipic Acid), Office of Air and
15 Radiation, Office of Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
16 ICIS (2007) "Adipic Acid." ICIS Chemical Business Americas. July 9, 2007.
17 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
18 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
19 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
20 Reimer, R.A., Slaten, C.S., Seapan, M., Koch, T.A. and Triner, V.G. (1999) "Implementation of Technologies for
21 Abatement of N2O Emissions Associated with Adipic Acid Manufacture." Proceedings of the 2nd Symposium on
22 Non-CO2 Greenhouse Gases (NCGG-2), Noordwijkerhout, The Netherlands, 8-10 Sept. 1999, Ed. J. van Ham et al.,
23 Kluwer Academic Publishers, Dordrecht, pp. 347-358.
24 SEI (2010) Industrial N2O Projects Under the COM: Adipic Acid-A Case for Carbon Leakage? Stockholm
25 Environment Institute Working Paper WP-US-1006. October 9, 2010.
26 Thiemens, M.H., and W.C. Trogler (1991) "Nylon production; an unknown source of atmospheric nitrous oxide."
27 Science 251:932-934.
28 VA DEQ (2010) Personal communication. Stanley Faggert, Virginia Department of Environmental Quality and
29 Joseph Herr, ICF International. March 12, 2010.
30 VA DEQ (2009) Personal communication. Stanley Faggert, Virginia Department of Environmental Quality and
31 Joseph Herr, ICF International. October 26, 2009.
32 VA DEQ (2006) Virginia Title V Operating Permit. Honeywell International Inc. Hopewell Plant. Virginia
33 Department of Environmental Quality. Permit No. PRO50232. Effective January 1, 2007.
34
Silicon Carbide Production
35 IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
36 Assessment Report of the Intergovernmental Panel on Climate Change. S. Solomon, D. Qin, M. Manning, Z. Chen,
37 M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.). Cambridge University Press. Cambridge, United
38 Kingdom 996 pp.
39 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
40 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
41 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
42 U.S. Census Bureau (2005 through 2015) U.S. International Trade Commission (USITC) Trade DataWeb.
43 Available online at .
10-24 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 United States Geological Survey (USGS) (2015a) Minerals Industry Surveys: Abrasives (Manufactured) in First
1 Quarter of 2015. U.S. Geological Survey, Reston, VA. September 2015. Available online at <
3 http://minerals.usgs.gov/minerals/pubs/commodity/abrasives/index.html>.
4 USGS (1991a through 2015b)M/'wera/5 Yearbook: Manufactured Abrasives Annual Report. U.S. Geological
5 Survey, Reston, VA. Available online at .
6 USGS (1991b through 2013) Miner ah Yearbook: Silicon Annual Report. U.S. Geological Survey, Reston, VA.
7 Available online at .
8 Titanium Dioxide Production
9 Gambogi, J. (2002) Telephone communication. Joseph Gambogi, Commodity Specialist, U.S. Geological Survey
10 and Philip Groth, ICF International. November 2002.
11 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
12 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
13 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
14 United States Geological Survey (USGS) (1991 through 2015b) Minerals Yearbook: Titanium. U.S. Geological
15 Survey, Reston, VA.
16 USGS (2015a) 2015 Mineral Commodity Summary: Titanium and Titanium Dioxide . U.S. Geological Survey,
17 Reston, VA. January, 2015.
18
30
Soda Ash Production and Consumption
19 Kostick, D. S. (2012) Personal communication. Dennis S. Kostick of U.S. Department of the Interior - U.S.
20 Geological Survey, Soda Ash Commodity Specialist with Gopi Manne and Bryan Lange of Eastern Research Group,
21 Inc. October 2012.
22 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
23 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
24 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
25 United States Geological Survey (USGS) (2015a) Mineral Industry Surveys: Soda Ash in July 2015. U.S.
26 Geological Survey, Reston, VA. September, 2015.
27 USGS (1994 through 20 \5b)Miner als Yearbook: Soda Ash Annual Report. U.S. Geological Survey, Reston, VA.
28 USGS (1995a) Trona Resources in the Green River Basin, Southwest Wyoming. U.S. Department of the Interior,
29 U.S. Geological Survey. Open-File Report 95-476. Wiig, Stephen, Grundy, W.D., Dyni, John R.
Petrochemical Production
31 ACC (2015) Business of Chemistry (Annual Data). American Chemistry Council, Arlington, VA.
32 ACC (2014a) U.S. Chemical Industry Statistical Handbook. American Chemistry Council, Arlington, VA.
33 ACC (2014b) Business of Chemistry (Annual Data). American Chemistry Council, Arlington, VA.
34 ACC (2002, 2003, 2005 through 2011) Guide to the Business of Chemistry. American Chemistry Council,
35 Arlington, VA.
36 AN (2014) About Acrylonitrile: Production. AN Group, Washington, D.C. Available online at:
37
3 8 EPA Greenhouse Gas Reporting Program (2015). Aggregation of reported facility level data under Subpart X -
39 National Petrochemical production for calendar years 2010-2014. Office of Air and Radiation, Office of
40 Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
References 10-25
-------
1 EPA Greenhouse Gas Reporting Program (2014). Aggregation of reported facility level data under Subpart X -
2 National Petrochemical production for calendar years 2010-2013. Office of Air and Radiation, Office of
3 Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
4 EPA (2008) Technical Support Document for the Petrochemical Production Sector: Proposed Rule for Mandatory
5 Reporting of Greenhouse Gases. U.S. Environmental Protection Agency. September 2008.
6 EPA (2000) Economic Impact Analysis for the Proposed Carbon Black Manufacturing NESHAP, U.S.
7 Environmental Protection Agency. Research Triangle Park, NC. EPA-452/D-00-003. May 2000.
8 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
9 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
10 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
11 Jordan, J. (2011) Personal communication, Jim Jordan of Jordan Associates on behalf of the Methanol Institute and
12 Pier LaFarge, ICF International. October 18, 2011
13 Johnson, G. L. (2005 through 2010) Personal communication. Greg Johnson of Liskow & Lewis, onbehalf of the
14 International Carbon Black Association (ICBA) and Caroline Cochran, ICF International. September 2010.
15 Johnson, G. L. (2003) Personal communication. Greg Johnson of Liskow & Lewis, onbehalf of the International
16 Carbon Black Association (ICBA) and Caren Mintz, ICF International November 2003.
17
HCFC-22 Production (TO BE UPDATED)
18 ARAP (2010) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
19 Atmospheric Policy to Deborah Ottinger of the U.S. Environmental Protection Agency. September 10, 2010.
20 ARAP (2009) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
21 Atmospheric Policy to Deborah Ottinger of the U.S. Environmental Protection Agency. September 21, 2009.
22 ARAP (2008) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
23 Atmospheric Policy to Deborah Ottinger of the U.S. Environmental Protection Agency. October 17, 2008.
24 ARAP (2007) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
25 Atmospheric Policy to Deborah Ottinger of the U.S. Environmental Protection Agency. October 2, 2007.
26 ARAP (2006) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
27 Atmospheric Policy to Sally Rand of the U.S. Environmental Protection Agency. July 11, 2006.
28 ARAP (2005) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
29 Atmospheric Policy to Deborah Ottinger of the U.S. Environmental Protection Agency. August 9, 2005.
30 ARAP (2004) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
31 Atmospheric Policy to Deborah Ottinger of the U.S. Environmental Protection Agency. June 3, 2004.
32 ARAP (2003) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
33 Atmospheric Policy to Sally Rand of the U.S. Environmental Protection Agency. August 18, 2003.
34 ARAP (2002) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
35 Atmospheric Policy to Deborah Ottinger of the U.S. Environmental Protection Agency. August 7, 2002.
36 ARAP (2001) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
37 Atmospheric Policy to Deborah Ottinger of the U.S. Environmental Protection Agency. August 6, 2001.
38 ARAP (2000) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
39 Atmospheric Policy to Sally Rand of the U.S. Environmental Protection Agency. August 13, 2000.
40 ARAP (1999) Facsimile from Dave Stirpe, Executive Director, Alliance for Responsible Atmospheric Policy to
41 Deborah Ottinger Schaefer of the U.S. Environmental Protection Agency. September 23, 1999.
42 ARAP (1997) Letter from Dave Stirpe, Director, Alliance for Responsible Atmospheric Policy to Elizabeth Dutrow
43 of the U.S. Environmental Protection Agency. December 23, 1997.
10-26 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
1 Assessment Report of the Intergovernmental Panel on Climate Change. S. Solomon, D. Qin, M. Manning, Z. Chen,
3 M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.). Cambridge University Press. Cambridge, United
4 Kingdom 996 pp.
5 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
6 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
7 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
8 IPCC (1996) Climate Change 1995: The Science of Climate Change. Intergovernmental Panel on Climate Change,
9 J.T. Houghton, L.G. Meira Filho, B.A. Callander, N. Harris, A. Kattenberg, and K. Maskell (eds.). Cambridge
10 University Press. Cambridge, United Kingdom.
11 RTI (2008) "Verification of Emission Estimates of HFC-23 from the Production of HCFC-22 Emissions from 1990
12 through 2006." Report prepared by RTI International for the Climate Change Division. March, 2008.
13 RTI (1997) "Verification of Emission Estimates of HFC-23 from the Production of HCFC-22: Emissions from
14 1990 through 1996." Report prepared by Research Triangle Institute for the Cadmus Group. November 25, 1997;
15 revised February 16, 1998.
16 UNFCCC (2014) Report of the Conference of the Parties on its nineteenth session, held in Warsaw from 11 to 23
17 November 2013. United Nations Framework Convention on Climate Change, Warsaw. (FCCC/CP/2013/10/Add.3).
18 January 31, 2014. Available online at .
19 Carbon Dioxide Consumption
20 Allis, R. et al. (2000) Natural CO2 Reservoirs on the Colorado Plateau and Southern Rocky Mountains: Candidates
21 for CO2 Sequestration. Utah Geological Survey and Utah Energy and Geoscience Institute. Salt Lake City, Utah.
22 ARI (1990 through 2010) CO2 Use in Enhanced Oil Recovery. Deliverable to ICF International under Task Order
23 102, July 15, 2011.
24 ARI (2007) CO2-EOR: An Enabling Bridge for the Oil Transition. Presented at "Modeling the Oil Transition—a
25 DOE/EPA Workshop on the Economic and Environmental Implications of Global Energy Transitions."
26 Washington, D.C. April 20-21, 2007.
27 ARI (2006) CO2-EOR: An Enabling Bridge for the Oil Transition. Presented at "Modeling the Oil Transition—a
28 DOE/EPA Workshop on the Economic and Environmental Implications of Global Energy Transitions."
29 Washington, D.C. April 20-21, 2006.
30 Broadhead (2003) Personal communication. Ron Broadhead, Principal Senior Petroleum Geologist and Adjunct
31 faculty, Earth and Environmental Sciences Department, New Mexico Bureau of Geology and Mineral Resources,
32 and Robin Pestrusak, ICF International. September 5, 2003.
33 COGCC (1999 through 2009) Monthly CO2 Produced by County. Available online at
34 . Accessed October
35 2014.
36 Denbury Resources Inc. (2002 through 2010) Annual Report: 2001 through 2009, Form 10-K. Available online at
37 .
38 Accessed September 2014.
3 9 EPA Greenhouse Gas Reporting Program (2015). Aggregation of reported facility level data under Subpart PP -
40 National level CO2 transferred for food & beverage applications for calendar years 2010 -2014. Office of Air and
41 Radiation, Office of Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
42 New Mexico Bureau of Geology and Mineral Resources (2006) Natural Accumulations of Carbon Dioxide in New
43 Mexico and Adjacent Parts of Colorado and Arizona: Commercial Accumulation of CO2. Available online at
44 .
References 10-27
-------
i Phosphoric Acid Production
2 EFMA (2000) "Production of Phosphoric Acid." Best Available Techniques for Pollution Prevention and Control in
3 the European Fertilizer Industry. Booklet 4 of 8. European Fertilizer Manufacturers Association. Available online at
4 .
5 FIPR (2003a) "Analyses of Some Phosphate Rocks." Facsimile Gary Albarelli, the Florida Institute of Phosphate
6 Research, Bartow, Florida, to Robert Lanza, ICF International. July 29, 2003.
7 FIPR (2003b) Florida Institute of Phosphate Research. Personal communication. Mr. Michael Lloyd, Laboratory
8 Manager, FIPR, Bartow, Florida, to Mr. Robert Lanza, ICF International. August 2003.
9 NCDENR (2013) North Carolina Department of Environment and Natural Resources, Title V Air Permit Review for
10 PCS Phosphate Company, Inc. - Aurora. Available online at
11 . Accessed on January 25, 2013.
12 United States Geological Survey (USGS) (1994 through 2015b) Minerals Yearbook. Phosphate Rock Annual
13 Report. U.S. Geological Survey, Reston, VA.
14 USGS (2015a) Mineral Commodity Summaries: Phosphate Rock 2015. January 2015. U.S. Geological Survey,
15 Reston, VA. Available online at: < http://minerals.usgs.gov/minerals/pubs/commodity/phosphate_rock/mcs-2015-
16 phosp.pdfX
17 USGS (2012b) Personal communication between Stephen Jasinski (USGS) and Mausami Desai (EPA) on October
18 12,2012.
19 Iron and Steel Production and Metallurgical Coke Production
20 AISI (2004 through 2015a) Annual Statistical Report, American Iron and Steel Institute, Washington, D.C.
21 AISI (2006 through 2014b) Personal communication, Mausami Desai, U.S. EPA, and American Iron and Steel
22 Institute, December 8, 2014.
23 AISI (2008c) Personal communication, Mausami Desai, U.S. EPA, and Bruce Steiner, Technical Consultant with
24 the American Iron and Steel Institute, October 2008.
25 Carroll (2015) Personal communication, Mausami Desai, U.S. EPA, and Colin P. Carroll, Director of Environment,
26 Health and Safety, American Iron and Steel Institute, September 2015.
27 DOE (2000) Energy and Environmental Profile of the U. S. Iron and Steel Industry. Office of Industrial
28 Technologies, U.S. Department of Energy. August 2000. DOE/EE-0229.EIA
29 EIA (1998 through 2014) Quarterly Coal Report: October-December, Energy Information Administration, U.S.
30 Department of Energy. Washington, D.C. DOE/EIA-0121.
31 EIA (2015a) Natural Gas Annual 2014, Energy Information Administration, U.S. Department of Energy.
32 Washington, D.C. DOE/EIA-0131(06).
33 EIA (2015b) Monthly Energy Review, December 2015, Energy Information Administration, U.S. Department of
34 Energy, Washington, D.C. DOE/EIA-0035(2015/12).
35 EIA (1992) Coal and lignite production. EIA State Energy Data Report 1992, Energy Information Administration,
36 U.S. Department of Energy, Washington, D.C.
37 EPA (2010) Carbon Content Coefficients Developed for EPA's Mandatory Reporting Rule. Office of Air and
38 Radiation, Office of Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
39 Fenton (2015) Personal communication. Michael Fenton, Commodity Specialist, U.S. Geological Survey and Marty
40 Wolf, Eastern Research Group. September 16, 2015.
41 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
42 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
43 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
10-28 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 IPCC/UNEP/OECD/IEA (1995) "Volume 3: Greenhouse Gas Inventory Reference Manual. Table 2-2". IPCC
2 Guidelines for National Greenhouse Gas Inventories. Intergovernmental Panel on Climate Change, United Nations
3 Environment Programme, Organization for Economic Co-Operation and Development, International Energy
4 Agency. IPCC WG1 Technical Support Unit, United Kingdom.
5 United States Geological Survey (USGS) (1991 through 2014) USGSMinerals Yearbook - Iron and Steel Scrap.
6 U.S. Geological Survey, Reston, VA.
7
21
Ferroalloy Production
9 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
10 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
11 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
12 Onder, H., and E.A. Bagdoyan (1993) Everything You've Always Wanted to Know about Petroleum Coke. Allis
13 Mineral Systems.
14 United States Geological Survey (USGS) (2015a) 2012Minerals Yearbook: Ferroalloys. U.S. Geological Survey,
15 Reston, VA. April 2015.
16 USGS (2015b) Mineral Industry Surveys: Silicon in June 2015. U.S. Geological Survey, Reston, VA. September
17 2015.
18 USGS (2014) Mineral Industry Surveys: Silicon in September 2014. U.S. Geological Survey, Reston, VA.
19 December 2014.
20 USGS (1996 through 2013) Minerals Yearbook: Silicon. U.S. Geological Survey, Reston, VA.
Aluminum Production (TO BE UPDATED)
22 EPA (2014) Greenhouse Gas Reporting Program (GHGRP). Envirofacts, Subpart: F Aluminum Production.
23 Available online at . Accessed on: November 13, 2014.
24 IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
25 Assessment Report of the Intergovernmental Panel on Climate Change. S. Solomon, D. Qin, M. Manning, Z. Chen,
26 M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.). Cambridge University Press. Cambridge, United
27 Kingdom 996 pp.
28 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
29 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
30 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
31 IPCC (1996) Climate Change 1995: The Science of Climate Change. Intergovernmental Panel on Climate Change,
32 J.T. Houghton, L.G. Meira Filho, B.A. Callander, N. Harris, A. Kattenberg, and K. Maskell (eds.). Cambridge
33 University Press. Cambridge, United Kingdom.
34 USAA (2014) U.S. Primary Aluminum Production 2013. U.S. Aluminum Association, Washington, D.C. October,
35 2014.
36 USAA (2013) U.S. Primary Aluminum Production 2012. U.S. Aluminum Association, Washington, D.C. January,
37 2013.
38 USAA (2012) U.S. Primary Aluminum Production 2011. U.S. Aluminum Association, Washington, D.C. January,
39 2012.
40 USAA (2011) U.S. Primary Aluminum Production 2010. U.S. Aluminum Association, Washington, D.C.
41 USAA (2010) U.S. Primary Aluminum Production 2009. U.S. Aluminum Association, Washington, D.C.
42 USAA (2008, 2009) U.S. Primary Aluminum Production. U.S. Aluminum Association, Washington, D.C.
References 10-29
-------
1 USAA (2004, 2005, 2006) Primary Aluminum Statistics. U.S. Aluminum Association, Washington, D.C.
2 USGS (2014) 2014Mineral Commodity Summaries: Aluminum. U.S. Geological Survey, Reston, VA.
3 USGS (2007) 2006Mineral Yearbook: Aluminum. U.S. Geological Survey, Reston, VA.
4 USGS (1995, 1998, 2000, 2001, 2002) Minerals Yearbook: Aluminum Annual Report. U.S. Geological Survey,
5 Reston, VA.
6 Magnesium Production and Processing (TO BE UPDATED)
7 Bartos S., C. Laush, J. Scharfenberg, and R. Kantamaneni (2007) "Reducing greenhouse gas emissions from
8 magnesium die casting." Journal of Cleaner Production, 15: 979-987, March.
9 EPA (2014) Envirofacts. Greenhouse Gas Reporting Program (GHGRP), Subpart T: Magnesium Production and
10 Processing. Available online at . Accessed on: November,
11 2014.
12 Gjestland, H. and D. Magers (1996) "Practical Usage of Sulphur [Sulfur] Hexafluoride for Melt Protection in the
13 Magnesium Die Casting Industry." #13,1996 Annual Conference Proceedings, International Magnesium
14 Association. Ube City, Japan.
15 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
16 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
17 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
18 RAND (2002) RAND Environmental Science and Policy Center, "Production and Distribution of SFe by End-Use
19 Applications" Katie D. Smythe. International Conference on SFe and the Environment: Emission Reduction
20 Strategies. San Diego, CA. November 21 -22, 2002.
21 United States Geological Survey (USGS) (2002, 2003, 2005 through 2008, 201 Ib, 2012, and 2013) Minerals
22 Yearbook: Magnesium Annual Report. U.S. Geological Survey, Reston, VA. Available online at
23 .
24 USGS (2010a) Mineral Commodity Summaries: Magnesium Metal. U.S. Geological Survey, Reston, VA. Available
25 online at .
26
Lead Production
27 Dutrizac, J.E., V. Ramachandran, and J.A. Gonzalez (2000) Lead-Zinc 2000. The Minerals, Metals, and Materials
28 Society.
29 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
30 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
31 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
32 Morris, D., F.R. Steward, and P. Evans (1983) Energy Efficiency of a Lead Smelter. Energy 8(5):337-349.
33 Sjardin, M. (2003) CC>2 Emission Factors for Non-Energy Use in the Non-Ferrous Metal, Ferroalloys and
34 Inorganics Industry. Copernicus Institute. Utrecht, the Netherlands.
35 Ullman (1997) Ullman 's Encyclopedia of Industrial Chemistry: Fifth Edition. Volume A5. John Wiley and Sons.
36 United States Geological Survey (USGS) (2015) 2015Mineral Commodity Summary, Lead. U.S. Geological
37 Survey, Reston, VA. January 2015.
38 USGS (2014) Mineral Commodity Summary, Lead. U.S. Geological Survey, Reston, VA. February 2014.
39 USGS (1995 through 2013) Minerals Yearbook: Lead Annual Report. U.S. Geological Survey, Reston, VA.
10-30 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
i Zinc Production
2 Horsehead Corp. (2015) Form 10-k, Annual Report for the Fiscal Year Ended December 31, 2014. Available at: <
3 http://www.sec.gov/Archives/edgar/data/1385544/000138554415000005/zinc-2014123110k.htm>. Submitted
4 March 2, 2015.
5 Horsehead Corp. (2014) Form 10-k, Annual Report for the Fiscal Year Ended December 31, 2013. Available at:
6 . Submitted
7 March 13, 2014.
8 Horsehead Corp. (2013) Form 10-k, Annual Report for the Fiscal Year Ended December 31, 2012. Available at:
9 .
10 Submitted March 18, 2013.
11 Horsehead Corp. (2012a) Form 10-k, Annual Report for the Fiscal Year Ended December, 31, 2011. Available at:
12 . Submitted March 9,
13 2012.
14 Horsehead Corp. (2012b) Horsehead's New Zinc Plant and its Impact on the Zinc Oxide Business. February 22,
15 2012. Available online at: . Accessed
16 September 10, 2015.
17 Horsehead Corp. (2011) 10-k Annual Report for the Fiscal Year Ended December, 31 2010. Available at:
18 . Submitted March 16, 2011.
19 Horsehead Corp. (2010a) 10-k Annual Report for the Fiscal Year Ended December, 31 2009. Available at:
20 . Submitted March 16, 2010.
21 Horsehead Corp. (2010b) Horsehead Holding Corp. Provides Update on Operations at its Monaco, PA Plant. July
22 28, 2010. Available at: .
23 Horsehead Corp (2008) 10-k Annual Report for the Fiscal Year Ended December, 31 2007. Available at:
24 . Submitted March 31, 2008.
25 Horsehead Corp (2007) Registration Statement (General Form) S-l. Available at . Submitted April 13, 2007.
27 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
28 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
29 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
30 PIZO (2014) Available at . Accessed December 9, 2014.
31 PIZO (2012) Available at . Accessed October 10, 2012.
32 Sjardin (2003) CO2 Emission Factors for Non-Energy Use in the Non-Ferrous Metal, Ferroalloys and Inorganics
33 Industry. Copernicus Institute. Utrecht, the Netherlands.
34 Steel Dust Recycling (SDR) (2015) Personal communication. Jeremy Whitten, EHS Manager, Steel Dust Recycling
35 LLC and Gopi Manne, Eastern Research Group, Inc. September 22, 2015.
36 SDR (2014) Personal communication. Art Rowland, Plant Manager, Steel Dust Recycling LLC and Gopi Manne,
37 Eastern Research Group, Inc. December 9, 2014.
38 SDR (2013) Available at . Accessed October 29, 2013.
39 SDR (2012) Personal communication. Art Rowland, Plant Manager, Steel Dust Recycling LLC and Gopi Manne,
40 Eastern Research Group, Inc. October 5, 2012.
41 United States Geological Survey (USGS) (2015) 2015Mineral Commodity Summary: Zinc. U.S. Geological Survey,
42 Reston, VA. Jqanuary 2015.
43 USGS (1995 through 2014) Minerals Yearbook: Zinc Annual Report. U.S. Geological Survey, Reston, VA.
References 10-31
-------
1 Viklund-White C. (2000) "The Use of LCA for the Environmental Evaluation of the Recycling of Galvanized
2 Steel." ISIJ International. Volume 40 No. 3: 292-299.
34
38
3 Semiconductor Manufacture (TO BE UPDATED)
4 Burton, C.S., and R. Beizaie (2001) "EPA's PFC Emissions Model (PEVM) v. 2.14: Description and
5 Documentation" prepared for Office of Global Programs, U. S. Environmental Protection Agency, Washington, DC.
6 November 2001.
7 Citigroup Smith Barney (2005) Global Supply/Demand Model for Semiconductors. March 2005.
8 Doering, R. and Nishi, Y (2000) "Handbook of Semiconductor Manufacturing Technology", Marcel Dekker, New
9 York, USA, 2000.
10 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
11 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
12 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
13 ISMI (2009) Analysis of Nitrous Oxide Survey Data. Walter Worth. June 8, 2009. Available online at
14
15 ITRS (2007, 2008, 2011, 2013) International Technology Roadmap for Semiconductors: 2006 Update, January
16 2007; International Technology Roadmap for Semiconductors: 2007 Edition, January 2008; International
17 Technology Roadmap for Semiconductors: 2011, January 2012; Update, International Technology Roadmap for
18 Semiconductors: 2013 Edition, Available online at . These
19 and earlier editions and updates are available at . Information about the number of
20 interconnect layers for years 1990-2010 is contained in Burton and Beizaie, 2001. PEVM is updated using new
21 editions and updates of the ITRS, which are published annually.
22 SEMI - Semiconductor Equipment and Materials Industry (2013) World Fab Forecast, May 2013 Edition.
23 SEMI - Semiconductor Equipment and Materials Industry (2012) World Fab Forecast, August 2012 Edition.
24 Semiconductor Industry Association (SIA) (2011) SICAS Capacity and Utilization Rates Q4 2011. Available online
25 at.
26 Semiconductor Industry Association (SIA) (2009) STATS: SICAS Capacity and Utilization Rates Q1-Q4 2008, Ql-
27 Q4 2009, Q1-Q4 2010, Q1-Q4 2011. Available online at <
28 http://www.semiconductors.org/industry_statistics/semiconductor_capacity_utilization_sicas_reports/>.
29 U.S. EPA (2006) Uses and Emissions of Liquid PFC Heat Transfer Fluids from the Electronics Sector. U.S.
30 Environmental Protection Agency, Washington, DC. EPA-430-R-06-901.
31 U.S. EPA Greenhouse Gas Reporting Program (GHGRP) Envirofacts. Subpart I: Electronics Manufacture.
32 Available online at
33 VLSI Research, Inc. (2012) Worldwide Silicon Demand. August 2012.
Substitution of Ozone Depleting Substances
35 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
36 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
37 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
Electrical Transmission and Distribution (TO BE UPDATED)
39 IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
40 Assessment Report of the Intergovernmental Panel on Climate Change. S. Solomon, D. Qin, M. Manning, Z. Chen,
41 M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.). Cambridge University Press. Cambridge, United
42 Kingdom 996 pp.
10-32 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
2 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
3 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
4 IPCC (1996) Climate Change 1995: The Science of Climate Change. Intergovernmental Panel on Climate Change,
5 J.T. Houghton, L.G. Meira Filho, B.A. Callander, N. Harris, A. Kattenberg, and K. Maskell (eds.). Cambridge
6 University Press. Cambridge, United Kingdom.
7 Levin et al. (2010) "The Global SF6 Source Inferred from Long-term High Precision Atmospheric Measurements
8 and its Comparison with Emission Imentones."Atmospheric Chemistry and Physics, 10: 2655-2662.
9 O'Connell, P., F. Heil, J. Henriot, G. Mauthe, H. Morrison, L. Neimeyer, M. Pittroff, R. Probst, J.P. Tailebois
10 (2002) SF6 in the Electric Industry, Status 2000, CIGRE. February 2002.
11 RAND (2004) "Trends in SF6 Sales and End-Use Applications: 1961-2003," Katie D. Smythe. International
12 Conference on SFg and the Environment: Emission Reduction Strategies. RAND Environmental Science and Policy
13 Center, Scottsdale, AZ. December 1-3, 2004.
14 UDI (2013). 2013 UDI Directory of Electric Power Producers and Distributors, 121st Edition, Platts.
15 UDI (2010) 2010 UDI Directory of Electric Power Producers and Distributors, 118th Edition, Platts.
16 UDI (2007) 2007 UDI Directory of Electric Power Producers and Distributors, 115th Edition, Platts.
17 UDI (2004) 2004 UDI Directory of Electric Power Producers and Distributors, 112th Edition, Platts.
18 UDI (2001) 2007 UDI Directory of Electric Power Producers and Distributors, 109th Edition, Platts.
19 UNFCCC (2014) Report of the Conference of the Parties on its nineteenth session, held in Warsaw from 11 to 23
20 November 2013. United Nations Framework Convention on Climate Change, Warsaw. (FCCC/CP/2013/10/Add.3).
21 January 31, 2014. Available online at .
22
Nitrous Oxide from Product Use
23 CGA (2003) "CGA Nitrous Oxide Abuse Hotline: CGA/NWSA Nitrous Oxide Fact Sheet." Compressed Gas
24 Association. November 3, 2003.
25 CGA (2002) "CGA/NWSA Nitrous Oxide Fact Sheet." Compressed Gas Association. March 25, 2002.
26 Heydorn, B. (1997) "Nitrous Oxide—North America." Chemical Economics Handbook, SRI Consulting. May 1997.
27 IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
28 Assessment Report of the Intergovernmental Panel on Climate Change. S. Solomon, D. Qin, M. Manning, Z. Chen,
29 M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.). Cambridge University Press. Cambridge, United
30 Kingdom 996 pp.
31 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
32 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
33 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
34 Ottinger (2014) Personal communication. Deborah Ottinger (CCD, U.S. EPA) and Mausami Desai (U.S. EPA).
35 Email received on January 29, 2014.
36 Tupman, M. (2003) Personal communication .Martin Tupman, Airgas Nitrous Oxide and Daniel Lieberman, ICF
37 International. August 8, 2003.
References 10-33
-------
i Industrial Processes and Product Use Sources of Indirect
2 Greenhouse Gases
3 EPA (2015) "1970 - 2014 Average annual emissions, all criteria pollutants in MS Excel." National Emissions
4 Inventory (NEI) Air Pollutant Emissions Trends Data. Office of Air Quality Planning and Standards, March 2015.
5 Available online at .
6 EPA (2003) E-mail correspondence containing preliminary ambient air pollutant data. Office of Air Pollution and
7 the Office of Air Quality Planning and Standards, U.S. Environmental Protection Agency. December 22, 2003.
8 EPA (1997) Compilation of Air Pollutant Emission Factors, AP-42. Office of Air Quality Planning and Standards,
9 U.S. Environmental Protection Agency. Research Triangle Park, NC. October 1997.
10
11
Agriculture
Enteric Fermentation
12 Archibeque, S. (2011) Personal Communication. Shawn Archibeque, Colorado State University, Fort Collins,
13 Colorado and staff at ICF International.
14 Crutzen, P.J., I. Aselmann, and W. Seiler (1986) Methane Production by Domestic Animals, Wild Ruminants, Other
15 Herbivores, Fauna, and Humans. Tellus, 38B:271-284.
16 Donovan, K. (1999) Personal Communication. Kacey Donovan, University of California at Davis and staff at ICF
17 International.
18 Doren, P.E., J. F. Baker, C. R. Long and T. C. Cartwright (1989) Estimating Parameters of Growth Curves of Bulls,
19 J Animal Science 67:1432-1445.
20 Enns, M. (2008) Personal Communication. Dr. Mark Enns, Colorado State University and staff at ICF International.
21 Galyean and Gleghorn (2001) Summary of the 2000 Texas Tech University Consulting Nutritionist Survey. Texas
22 Tech University. Available online at . June
23 2009.
24 Holstein Association (2010) History of the Holstein Breed (website). Available online at
25 . Accessed September 2010.
26 ICF (2006) Cattle Enteric Fermentation Model: Model Documentation. Prepared by ICF International for the
27 Environmental Protection Agency. June 2006.
28 ICF (2003) Uncertainty Analysis of 2001 Inventory Estimates of Methane Emissions from Livestock Enteric
29 Fermentation in the U.S. Memorandum from ICF International to the Environmental Protection Agency. May 2003.
30 IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
31 Assessment Report of the Intergovernmental Panel on Climate Change. S. Solomon, D. Qin, M. Manning, Z. Chen,
32 M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.). Cambridge University Press. Cambridge, United
33 Kingdom 996 pp.
34 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
35 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
36 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
37 Johnson, D. (2002) Personal Communication. Don Johnson, Colorado State University, Fort Collins, and ICF
38 International.
39 Johnson, D. (1999) Personal Communication. Don Johnson, Colorado State University, Fort Collins, and David
40 Conneely, ICF International.
10-34 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Johnson, K. (2010) Personal Communication. Kris Johnson, Washington State University, Pullman, and ICF
2 International.
3 Kebreab E., K. A. Johnson, S. L. Archibeque, D. Pape, and T. Wirth (2008) Model for estimating enteric methane
4 emissions from United States dairy and feedlot cattle. J. Anim. Sci. 86: 2738-2748.
5 Lippke, H., T. D. Forbes, and W. C. Ellis. (2000) Effect of supplements on growth and forage intake by stacker
6 steers grazing wheat pasture. J. Anim. Sci. 78:1625-1635
7 National Bison Association (2011) Handling & Carcass Info (on website). Available online at:
8 . Accessed August 16, 2011.
9 National Bison Association (1999) Total Bison Population—1999. Report provided during personal email
10 communication with Dave Carter, Executive Director, National Bison Association July 19, 2011.
11 NRC (1999) 1996 BeefNRC: Appendix Table 22. National Research Council.
12 NRC (1984) Nutrient requirements for beef cattle (6th Ed.). National Academy Press, Washington, DC.
13 Pinchak, W.E., D. R. Tolleson, M. McCloy, L. J. Hunt, R. J. Gill, R. J. Ansley, and S. J. Bevers (2004) Morbidity
14 effects on productivity and profitability of stacker cattle grazing in the southern plains. J. Anim. Sci. 82:2773 -2779.
15 Platter, W. J., J. D. Tatum, K. E. Belk, J. A. Scanga, and G. C. Smith (2003) Effects of repetitive use of hormonal
16 implants on beef carcass quality, tenderness, and consumer ratings of beef palatability. J. Anim. Sci. 81:984-996.
17 Preston, R.L. (2010) What's The Feed Composition Value of That Cattle Feed? Beef Magazine, March 1, 2010.
18 Available at: .
19 Skogerboe, T. L., L. Thompson, J. M. Cunningham, A. C. Brake, V. K. Karle (2000) The effectiveness of a single
20 dose of doramectin pour-on in the control of gastrointestinal nematodes in yearling stacker cattle. Vet. Parasitology
21 87:173-181.
22 Soliva, C.R. (2006) Report to the attention of IPCC about the data set and calculation method used to estimate
23 methane formation from enteric fermentation of agricultural livestock population and manure management in Swiss
24 agriculture. On behalf of the Federal Office for the Environment (FOEN), Berne, Switzerland,
25 USDA (2015) Quick Stats: Agricultural Statistics Database. National Agriculture Statistics Service, U.S.
26 Department of Agriculture. Washington, D.C. Available online at . Accessed
27 August 3, 2015.
28 USDA (2007) Census of Agriculture: 2007 Census Report. United States Department of Agriculture. Available
29 online at: .
30 USDA (2002) Census of Agriculture: 2002 Census Report. United States Department of Agriculture. Available
31 online at: .
32 USDA (1997) Census of Agriculture: 1997 Census Report. United States Department of Agriculture. Available
33 online at: . Accessed July 18, 2011.
34 USDA (1996) Beef Cow/Calf Health and Productivity Audit (CHAPA): Forage Analyses from Cow/Calf Herds in 18
35 States. National Agriculture Statistics Service, U.S. Department of Agriculture. Washington, D.C. Available online
36 at . March 1996.
37 USDA (1992) Census of Agriculture: 1992 Census Report. United States Department of Agriculture. Available
38 online at: . Accessed July 18, 2011.
39 USDA:APHIS:VS (2010) Beef 2007-08, Part V: Reference of Beef Cow-calfManagement Practices in the United
40 States, 2007-08. USDA-APHIS-VS, CEAH. Fort Collins, CO.
41 USDA:APHIS:VS (2002) Reference of 2002 Dairy Management Practices. USDA-APHIS-VS, CEAH. Fort
42 Collins, CO. Available online at .
43 USD A: APHIS: VS (1998) Beef '97, Parts I-IV. USDA-APHIS-VS, CEAH. Fort Collins, CO. Available online at
44
References 10-35
-------
1 USDA:APHIS:VS (1996) Reference of 1996 Dairy Management Practices. USDA-APHIS-VS, CEAH. Fort
2 Collins, CO. Available online at .
3 USD A: APHIS: VS (1994) Beef Cow/Calf Health and Productivity Audit. USDA-APHIS-VS, CEAH. Fort Collins,
4 CO. Available online at .
5 USD A: APHIS: VS (1993) Beef Cow/Calf Health and Productivity Audit. USDA-APHIS-VS, CEAH. Fort Collins,
6 CO. August 1993. Available online at .
7 Vasconcelos and Galyean (2007) Nutritional recommendations of feedlot consulting nutritionists: The 2007 Texas
8 Tech University Study. J. Anim. Sci. 85:2772-2781.
9 Manure Management
10 Anderson, S. (2000) Personal Communication. Steve Anderson, Agricultural Statistician, National Agriculture
11 Statistics Service, U.S. Department of Agriculture and Lee-Ann Tracy, ERG. Washington, D.C. May 31, 2000.
12 ASAE (1998) ASAE Standards 1998, 45th Edition. American Society of Agricultural Engineers. St. Joseph,
13 MI.Bryant, M.P., V.H. Varel, R.A. Frobish, and H.R. Isaacson (1976) In H.G. Schlegel (ed.); Seminar on Microbial
14 Energy Conversion. E. Goltz KG. Gottingen, Germany.
15 Bush, E. (1998) Personal communication with Eric Bush, Centers for Epidemiology and Animal Health, U.S.
16 Department of Agriculture regarding National Animal Health Monitoring System's (NAHMS) Swine '95 Study.
17 Deal, P. (2000) Personal Communication. Peter B. Deal, Rangeland Management Specialist, Florida Natural
18 Resource Conservation Service and Lee-Ann Tracy, ERG. June 21, 2000.
19 EPA (2012) AgSTAR Anaerobic Digester Database. Available online at:
20 .
21 EPA (2008) Climate Leaders Greenhouse Gas Inventory Protocol Offset Project Methodology for Project Type
22 Managing Manure with Biogas Recovery Systems. Available online at
23 .
24 EPA (2006) AgSTAR Digest. Office of Air and Radiation, U.S. Environmental Protection Agency. Washington, D.C.
25 Winter 2006. Available online at . Retrieved July 2006.
26 EPA (2005) National Emission Inventory—Ammonia Emissions from Animal Agricultural Operations, Revised
27 Draft Report. U.S. Environmental Protection Agency. Washington, D.C. April 22, 2005. Available online at
28 . Retrieved
29 August 2007.
30 EPA (2003) AgSTAR Digest. Office of Air and Radiation, U.S. Environmental Protection Agency. Washington, D.C.
31 Winter 2003. Available online at . Retrieved July 2006.
32 EPA (2002a) Development Document for the Final Revisions to the National Pollutant Discharge Elimination
33 System (NPDES) Regulation and the Effluent Guidelines for Concentrated Animal Feeding Operations (CAFOS).
34 U.S. Environmental Protection Agency. EPA-821-R-03-001. December 2002.
35 EPA (2002b) Cost Methodology for the Final Revisions to the National Pollutant Discharge Elimination System
36 Regulation and the Effluent Guidelines for Concentrated Animal Feeding Operations. U.S. Environmental
37 Protection Agency. EPA-821-R-03-004. December 2002.
38 EPA (2000) AgSTAR Digest. Office of Air and Radiation, U.S. Environmental Protection Agency. Washington, D.C.
39 Spring 2000. Available online at: .
40 EPA (1992) Global Methane Emissions from Livestock and Poultry Manure, Office of Air and Radiation, U.S.
41 Environmental Protection Agency. February 1992.
42 ERG (2010a) "Typical Animal Mass Values for Inventory Swine Categories." Memorandum to EPA from ERG.
43 July 19, 2010.
44 ERG (20 lOb) Telecon with William Boyd of USD A NRCS and Cortney Me of ERG Concerning Updated VS and
45 Nex Rates. August 8, 2010.
10-36 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 ERG (2010c) "Updating Current Inventory Manure Characteristics new USD A Agricultural Waste Management
2 Field Handbook Values." Memorandum to EPA from ERG. August 13, 2010.
3 ERG (2008) "Methodology for Improving Methane Emissions Estimates and Emission Reductions from Anaerobic
4 Digestion System for the 1990-2007 Greenhouse Gas Inventory for Manure Management." Memorandum to EPA
5 from ERG. August 18, 2008.
6 ERG (2003a) "Methodology for Estimating Uncertainty for Manure Management Greenhouse Gas Inventory."
7 Contract No. GS-10F-0036, Task Order 005. Memorandum to EPA from ERG, Lexington, MA. September 26,
8 2003.
9 ERG (2003b) "Changes to Beef Calves and Beef Cows Typical Animal Mass in the Manure Management
10 Greenhouse Gas Inventory." Memorandum to EPA from ERG, October 7, 2003.
11 ERG (2001) Summary of development ofMDP Factor for methane conversion factor calculations. ERG, Lexington,
12 MA. September 2001.
13 ERG (2000a) Calculations: Percent Distribution of Manure for Waste Management Systems. ERG, Lexington, MA.
14 August 2000.
15 ERG (2000b) Discussion of Methodology for Estimating Animal Waste Characteristics (Summary of B0 Literature
16 Review). ERG, Lexington, MA. June 2000.
17 Garrett, W.N. and D.E. Johnson (1983) "Nutritional energetics of ruminants." Journal of Animal Science,
18 57(suppl.2):478-497.
19 Groffman, P.M., R. Brumme, K. Butterbach-Bahl, K.E. Dobbie, A.R. Mosier, D. Ojima, H. Papen, W.J. Parton,
20 K. A. Smith, and C. Wagner-Riddle (2000) "Evaluating annual nitrous oxide fluxes at the ecosystem scale." Global
21 Biogeochemcial Cycles, 14(4):1061-1070.
22 Hashimoto, A.G. (1984) "Methane from Swine Manure: Effect of Temperature and Influent Substrate Composition
23 on Kinetic Parameter (k)." Agricultural Wastes, 9:299-308.
24 Hashimoto, A.G., V.H. Varel, and Y.R. Chen (1981) "Ultimate Methane Yield from Beef Cattle Manure; Effect of
25 Temperature, Ration Constituents, Antibiotics and Manure Age." Agricultural Wastes, 3:241-256.
26 Hill, D.T. (1984) "Methane Productivity of the Major Animal Types." Transactions oftheASAE, 27(2):530-540.
27 Hill, D.T. (1982) "Design of Digestion Systems for Maximum Methane Production." Transactions oftheASAE,
28 25(1):226-230.
29 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
30 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
31 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
32 Johnson, D. (2000) Personal Communication. Dan Johnson, State Water Management Engineer, California Natural
33 Resource Conservation Service and Lee-Ann Tracy, ERG. June 23, 2000.
34 Lange, J. (2000) Personal Communication. John Lange, Agricultural Statistician, U.S. Department of Agriculture,
35 National Agriculture Statistics Service and Lee-Ann Tracy, ERG. Washington, D.C. May 8, 2000.
36 Meagher, M. (1986). Bison. Mammalian Species. 266: 1-8.
37 Miller, P. (2000) Personal Communication. Paul Miller, Iowa Natural Resource Conservation Service and Lee-Ann
38 Tracy, ERG. June 12, 2000.
39 Milton, B. (2000) Personal Communication. Bob Milton, Chief of Livestock Branch, U.S. Department of
40 Agriculture, National Agriculture Statistics Service and Lee-Ann Tracy, ERG. May 1, 2000.
41 Moffroid, K. and D. Pape. (2014) 1990-2013 Volatile Solids and Nitrogen Excretion Rates. Dataset to EPA from
42 ICF International. August 2014.
43 Morris, G.R. (1976) Anaerobic Fermentation of Animal Wastes: A Kinetic and Empirical Design Fermentation.
44 M.S. Thesis. Cornell University.
References 10-37
-------
1 National Bison Association (1999) Total Bison Population—1999. Report provided during personal email
2 communication with Dave Carter, Executive Director, National Bison Association July 19, 2011.
3 NOAA (2014) National Climate Data Center (NCDC). Available online at
4 (for all states except Alaska and Hawaii) and
5 . (for Alaska and Hawaii). September 2014.
6 Oft, S.L. (2000) Dairy '96 Study. Stephen L. Ott, Animal and Plant Health Inspection Service, U.S. Department of
7 Agriculture. June 19, 2000.
8 Poe, G., N. Bills, B. Bellows, P. Crosscombe, R. Koelsch, M. Kreher, and P. Wright (1999) Staff Paper
9 Documenting the Status of Dairy Manure Management in New York: Current Practices and Willingness to
10 Participate in Voluntary Programs. Department of Agricultural, Resource, and Managerial Economics; Cornell
11 University, Ithaca, New York, September.
12 Safley, L.M., Jr. (2000) Personal Communication. Deb Bartram, ERG and L.M. Safley, President, Agri-Waste
13 Technology. June and October 2000.
14 Safley, L.M., Jr. and P.W. Westerman (1990) "Psychrophilic anaerobic digestion of animal manure: proposed design
15 methodology." Biological Wastes, 34:133-148.
16 Stettler, D. (2000) Personal Communication. Don Stettler, Environmental Engineer, National Climate Center,
17 Oregon Natural Resource Conservation Service and Lee-Ann Tracy, ERG. June 27, 2000.
18 Sweeten, J. (2000) Personal Communication. John Sweeten, Texas A&M University and Indra Mitra, ERG. June
19 2000.
20 UEP (1999) Voluntary Survey Results—Estimated Percentage Participation/Activity. Caged Layer Environmental
21 Management Practices, Industry data submissions for EPA profile development, United Egg Producers and National
22 Chicken Council. Received from John Thorne, Capitolink. June 2000.
23 USDA (2015a) Quick Stats: Agricultural Statistics Database. National Agriculture Statistics Service, U.S.
24 Department of Agriculture. Washington, D.C. Available online at .
25 USDA (2015b) Chicken and Eggs 2014 Summary. National Agriculture Statistics Service, U.S. Department of
26 Agriculture. Washington, D.C. February 2015. Available online at
27 .
28 USDA (2015c) Poultry - Production and Value 2014 Summary. National Agriculture Statistics Service, U.S.
29 Department of Agriculture. Washington, D.C. April 2015. Available online at
30 .
31 USDA (2014a) 1987, 1992, 1997, 2002, 2007, and 2012 Census of Agriculture. National Agriculture Statistics
32 Service, U.S. Department of Agriculture. Washington, D.C. Available online at
33 . May 2014.
34 USDA (2014b) Chicken and Eggs 2013 Summary. National Agriculture Statistics Service, U.S. Department of
35 Agriculture. Washington, D.C. February 2014. Available online at
36 .
37 USDA (2014c) Poultry - Production and Value 2013 Summary. National Agriculture Statistics Service, U.S.
38 Department of Agriculture. Washington, D.C. April 2014. Available online at
39 .
40 USDA (2013a) Chicken and Eggs 2012 Summary. National Agriculture Statistics Service, U.S. Department of
41 Agriculture. Washington, D.C. February 2013. Available online at
42 .
43 USDA (2013b) Poultry - Production and Value 2012 Summary. National Agriculture Statistics Service, U.S.
44 Department of Agriculture. Washington, D.C. April 2013. Available online at
45 .
10-38 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 USD A (2012a) Chicken and Eggs 2011 Summary. National Agriculture Statistics Service, U.S. Department of
2 Agriculture. Washington, D.C. February 2012. Available online at
3 .
4 USDA (2012b) Poultry - Production and Value 2011 Summary. National Agriculture Statistics Service, U.S.
5 Department of Agriculture. Washington, D.C. April 2012. Available online at
6 .
7 USDA (201 la) Chicken and Eggs 2010 Summary. National Agriculture Statistics Service, U.S. Department of
8 Agriculture. Washington, D.C. February 2011. Available online at
9 .
10 USDA (201 Ib) Poultry - Production and Value 2010 Summary. National Agriculture Statistics Service, U.S.
11 Department of Agriculture. Washington, D.C. April 2011. Available online at
12 .
13 USDA (2010a) Chicken and Eggs 2009 Summary. National Agriculture Statistics Service, U.S. Department of
14 Agriculture. Washington, D.C. February 2010. Available online at
15 .
16 USDA (2010b) Poultry - Production and Value 2009 Summary. National Agriculture Statistics Service, U.S.
17 Department of Agriculture. Washington, D.C. April 2010. Available online at
18 .
19 USDA (2009a) Chicken and Eggs 2008 Summary. National Agriculture Statistics Service, U.S. Department of
20 Agriculture. Washington, D.C. February 2009. Available online at
21 .
22 USDA (2009b) Poultry - Production and Value 2008 Summary. National Agriculture Statistics Service, U.S.
23 Department of Agriculture. Washington, D.C. April 2009. Available online at
24 .
25 USDA (2009c) Chicken and Eggs - Final Estimates 2003-2007. National Agriculture Statistics Service, U.S.
26 Department of Agriculture. Washington, D.C. March 2009. Available online at
27 .
28 USDA (2009d) Poultry Production and Value—Final Estimates 2003-2007. National Agriculture Statistics Service,
29 U.S. Department of Agriculture. Washington, D.C. May 2009. Available online at
30 .
31 USDA (2008) Agricultural Waste Management Field Handbook, National Engineering Handbook (NEH), Part 651.
32 Natural Resources Conservation Service, U.S. Department of Agriculture.
33 USDA (2004a) Chicken and Eggs—Final Estimates 1998-2003. National Agriculture Statistics Service, U.S.
34 Department of Agriculture. Washington, D.C. April 2004. Available online at
35 .
36 USDA (2004b) Poultry Production and Value—Final Estimates 1998-2002. National Agriculture Statistics Service,
37 U.S. Department of Agriculture. Washington, D.C. April 2004. Available online at
38 .
39 USDA (1999) Poultry Production and Value—Final Estimates 1994-97. National Agriculture Statistics Service,
40 U.S. Department of Agriculture. Washington, D.C. March 1999. Available online at
41 .
42 USDA (1998) Chicken and Eggs—Final Estimates 1994-97. National Agriculture Statistics Service, U.S.
43 Department of Agriculture. Washington, D.C. December 1998. Available online at
44 .
45 USDA (1996) Agricultural Waste Management Field Handbook, National Engineering Handbook (NEH), Part 651.
46 Natural Resources Conservation Service, U.S. Department of Agriculture. July 1996.
References 10-39
-------
1 USDA (1995a) Poultry Production and Value—Final Estimates 1988-1993. National Agriculture Statistics Service,
2 U.S. Department of Agriculture. Washington, D.C. March 1995. Available online at
3 .
4 USDA (1995b) Chicken and Eggs—Final Estimates 1988-1993. National Agriculture Statistics Service, U.S.
5 Department of Agriculture. Washington, D.C. December 1995. Available online at
6 .
7 USDA (1994) Sheep and Goats—Final Estimates 1989-1993. National Agriculture Statistics Service, U.S.
8 Department of Agriculture. Washington, D.C. January 31, 1994. Available online at
9 .
10 USDA, APHIS (2003) Sheep 2001, Part I: Reference of Sheep Management in the United States, 2001 and Part
11 IV:Baseline Reference of 2001 Sheep Feedlot Health and Management. USDA-APHIS-VS. Fort Collins, CO.
12 #N356.0702. .
13 USDA, APHIS (2000) Layers '99—Part II: References of 1999 Table Egg Layer Management in the U.S. USDA-
14 APHIS-VS. Fort Collins, CO.
15 .
16 USDA, APHIS (1996) Swine '95: Grower/Finisher Part II: Reference of 1995 U.S. Grower/Finisher Health &
17 Management Practices. USDA-APHIS-VS. Fort Collins, CO.
18 .
19 Wright, P. (2000) Personal Communication. Lee-Ann Tracy, ERG and Peter Wright, Cornell University, College of
20 Agriculture and Life Sciences. June 23, 2000.
21
Rice Cultivation (TO BE UPDATED)
22 Anderson, M. (2008 through 2014) Email correspondence. Monte Anderson, Oklahoma Farm Service Agency and
23 ICF International.
24 Baldwin, K., E. Dohlman, N. Childs and L. Forman (2010). Consolidation and Structural Change in the U.S. Rice
25 Sector. Economic Research Service: U.S. Department of Agriculture, Washington D.C. Available online at
26 . September2013.
27 Baicich, P. (2013). The Birds and Rice Connection. Bird Watcher's Digest. Available online at
28 .
29 Beighley, D. (2011 through 2012) Email correspondence. Donn Beighley, Southeast Missouri State University,
30 Department of Agriculture and ICF International.
31 Bossio, D. A., W. Horwath, R.G. Mutters, and C. van Kessel (1999) "Methane pool and flux dynamics in a rice field
32 following straw incorporation." Soil Biology and Biochemistry, 31:1313-1322.
33 Buehring, N. (2009 through 2011) Email correspondence. Nathan Buehring, Assistant Professor and Extension Rice
34 Specialist, Mississippi State University Delta Branch Exp. Station and ICF International.
35 Byrd, G. T., F. M. Fisher, & R. L. Sass. (2000) Relationships between methane production and emission to lacunal
36 methane concentrations in rice. Global biogeochemical cycles, 14(1), 73-83.
37 California Air Resources Board (2003) 2003 Progress Report on the Phase-down of Rice Straw Burning in the
38 Sacremento Valley Air Basin. Available online at .
39 Cantens, G. (2004 through 2005) Personal Communication. Janet Lewis, Assistant to Gaston Cantens, Vice
40 President of Corporate Relations, Florida Crystals Company and ICF International.
41 Deren, C. (2002) Personal Communication and Dr. Chris Deren, Everglades Research and Education Centre at the
42 University of Florida and Caren Mintz, ICF International. August 15, 2002.
43 Environmental Defense Fund (2011) Creating and Quantifying Carbon Credits from Voluntary Practices on Rice.
44 Available online at .
10-40 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Fife, L. (2011) Email correspondence. Les Fife, Sacramento Valley Agricultural Burning Coordinator and ICF
2 International.
3 Fitzgerald, G.J., K. M. Scow & J. E. Hill (2000) "Fallow Season Straw and Rice Management Effects on Methane
4 Emissions in California Rice." Global bigeochemical cycles, 14 (3), 767-776.
5 Gonzalez, R. (2007 through 2014) Email correspondence. Rene Gonzalez, Plant Manager, Sem-Chi Rice Company
6 and ICF International.
7 Hardke, J. (2014) Personal Communication. Dr. Jarrod Hardke, Rice Extension Agronomist at the University of
8 Arkansas Rice Research and Extension Center and Kirsten Jaglo, ICF International. September 11, 2014.
9 Hardke, J. (2013) Email correspondence. Dr. Jarrod Hardke, Rice Extension Agronomist at the University of
10 Arkansas Rice Research and Extension Center and Cassandra Snow, ICF International. July 15, 2013.
11 Holzapfel-Pschorn, A., R. Conrad, and W. Seiler (1985) "Production, Oxidation, and Emissions of Methane in Rice
12 Paddies." FEMSMicrobiology Ecology, 31:343-351.
13 IPCC (2007) Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II and III to the Fourth
14 Assessment Report (AR4) of the IPCC. The Intergovernmental Panel on Climate Change, R.K. Pachauri, A. Resinger
15 (eds.). Geneva, Switzerland.
16 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
17 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
18 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
19 IPCC/UNEP/OECD/IEA (1997) Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories.
20 Intergovernmental Panel on Climate Change, United Nations Environment Programme, Organization for Economic
21 Co-Operation and Development, International Energy Agency, Paris, France.
22 Kirstein, A. (2003 through 2004, 2006) Personal Communication. Arthur Kirstein, Coordinator, Agricultural
23 Economic Development Program, Palm Beach County Cooperative Extension Service, FL and ICF International.
24 Klosterboer, A. (1997, 1999 through 2003) Personal Communication. Arlen Klosterboer, retired Extension
25 Agronomist, Texas A&M University and ICF International. July 7, 2003.
26 Kongchum, M. (2005). Effect of Plant Residue and Water Management Practices on Soil Redox Chemistry,
27 Methane Emission, and Rice Productivity. LSU PhD Thesis.
28 Lee, D. (2003 through 2007) Email correspondence. Danny Lee, OK Farm Service Agency and ICF International.
29 Lindau, C.W. and P.K. Bollich (1993) "Methane Emissions from Louisiana First and Ratoon Crop Rice." Soil
30 Science, 156:42-48.
31 Lindau, C.W., P.K Bollich, and R.D. DeLaune (1995) "Effect of Rice Variety on Methane Emission from Louisiana
32 ~Rice." Agriculture, Ecosystems and Environment, 54:109-114.
33 Linscombe, S. (1999, 2001 through 2014) Email correspondence. Steve Linscombe, Professor with the Rice
34 Research Station at Louisiana State University Agriculture Center and ICF International.
35 McMillan, A., M. L.Goulden, and S.C. Tyler. (2007) Stoichiometry of CH4 and CO2 flux in a California rice paddy.
36 Journal of Geophysical Research: Biogeosciences (2005-2012), 112(G1).
37 Mutters, C. (2001 through 2005) Personal Communication. Mr. Cass Mutters, Rice Farm Advisor for Butte, Glen,
38 and Tehama Counties University of California, Cooperative Extension Service and ICF International.
39 Rogers, C.W., K. R. Brye, R.J. Norman, T. Gasnier, D. Frizzell, and J. Branson. (2011). Methane Emissions from a
40 Silt-Loam Soil Under Direct-Seeded, Delayed-Flood Rice Management. B.R. Wells Rice Research Studies 2011.
41 306-315.
42 Sass, R. L. (2001). CH4 Emissions from Rice Agriculture. Good Practice Guidance and Uncertainty Management in
43 National Greenhouse Gas Inventories. 399-417. Available at .
References 10-41
-------
1 Sass, R. L., J.A. Andrews, A. Ding and P.M. Fisher Jr. (2002a). Spatial and temporal variability in methane
2 emissions from rice paddies: Implications for assessing regional methane budgets. Nutrient Cycling in
3 Agroecosystems, 64(1-2), 3-7.
4 Sass, R. L., P.M. Fisher, and J. A. Andrews. (2002b). Spatial variability in methane emissions from a Texas rice
5 field with some general implications. Global biogeochemical cycles, 16(1), 15-1.
6 Sass, R.L., F.M Fisher, P.A. Harcombe, and F.T. Turner (1991a) "Mitigation of Methane Emissions from Rice
7 Fields: Possible Adverse Effects of Incorporated Rice Straw." Global Biogeochemical Cycles, 5:275-287.
8 Sass, R.L., F.M. Fisher, F.T. Turner, and M.F. Jund (1991b) "Methane Emissions from Rice Fields as Influenced by
9 Solar Radiation, Temperature, and Straw Incorporation." Global Biogeochemical Cycles, 5:335-350.
10 Sass, R.L., F.M. Fisher, P.A. Harcombe, and F.T. Turner (1990) "Methane Production and Emissions in a Texas
11 Rice Field." Global Biogeochemical Cycles, 4:47-68.
12 Schueneman, T. (1997, 1999 through 2001) Personal Communication. Tom Schueneman, Agricultural Extension
13 Agent, Palm Beach County, FL and ICF International.
14 Slaton, N. (1999 through 2001) Personal Communication. Nathan Slaton, Extension Agronomist—Rice, University
15 of Arkansas Division of Agriculture Cooperative Extension Service and ICF International.
16 Stansel, J. (2004 through 2005) Email correspondence. Dr. Jim Stansel, Resident Director and Professor Emeritus,
17 Texas A&M University Agricultural Research and Extension Center and ICF International.
18 Street, J. (1999 through 2003) Personal Communication. Joe Street, Rice Specialist, Mississippi State University,
19 Delta Research Center and ICF International.
20 Texas Agricultural Experiment Station (2007 through 2014) Texas Rice Acreage by Variety. Agricultural Research
21 and Extension Center, Texas Agricultural Experiment Station, Texas A&M University System. Available online at
22 .
23 Texas Agricultural Experiment Station (2006) 2005 - Texas Rice Crop Statistics Report. Agricultural Research and
24 Extension Center, Texas Agricultural Experiment Station, Texas A&M University System, p. 8. Available online at
25 .
26 USD A (2005 through 2014) Crop Production Summary. National Agricultural Statistics Service, Agricultural
27 Statistics Board, U.S. Department of Agriculture, Washington, D.C. Available online at
28 .
29 USDA (2012) Summary of USDA-ARS Research on the Interrelationship of Genetic and Cultural Management
30 Factors That Impact Grain Arsenic Accumulation in Rice. News and Events. Agricultural Research Service, U.S.
31 Department of Agriculture, Washington, D.C. Available online at
32 . September 2013.
33 USDA (2003) Field Crops, Final Estimates 1997-2002. Statistical Bulletin No. 982. National Agricultural
34 Statistics Service, Agricultural Statistics Board, U.S. Department of Agriculture, Washington, D.C. Available
35 online at . September 2005.
36 USDA (1998) Field Crops Final Estimates 1992-1997. Statistical Bulletin Number 947 a. National Agricultural
37 Statistics Service, Agricultural Statistics Board, U.S. Department of Agriculture, Washington, D.C. Available
38 online at . July 2001.
39 USDA (1994) Field Crops Final Estimates 1987-1992. Statistical Bulletin Number 896. National Agricultural
40 Statistics Service, Agricultural Statistics Board, U.S. Department of Agriculture, Washington, D.C. Available
41 online at . July 2001.
42 Vayssieres, M. (2013). Email correspondance. Marc Vayssieres, Ph.D. Cal/EPA Air Resources Board Air Quality
43 Planning & Science Division and Rachel Steele, ICF International. January 2014.
44 Walker, T. (2005, 2007 through 2008) Email correspondence. Tim Walker, Assistant Research Professor,
45 Mississippi State University Delta Branch Exp. Station and ICF International.
10-42 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Wang, J.J., S.K. Dodla, S. Viator, M. Kongchum, S. Harrison, S. D. Mudi, S. Liu, Z. Tian (2013). Agriculture Field
2 Management Practices and Greenhouse Gas Emissions from Louisiana Soils. Louisiana Agriculture, Spring 2013: 8-
3 9. Available online at .
5 Wilson, C. (2002 through 2007, 2009 through 2012) Personal Communication. Dr. Chuck Wilson, Rice Specialist at
6 the University of Arkansas Cooperative Extension Service and ICF International.
7 Yao, H., J. Jingyan, Z. Lianggang, R. L. Sass, and F. M. Fisher. (2001). Comparison of field measurements of CH4
8 emission from rice cultivation in Nanjing, China and in Texas, USA. Advances in Atmospheric Sciences, 18(6),
9 1121-1130.
10 Young, M. (2013) Rice and Ducks. Ducks Unlimited, Memphis, TN. Available online at
11 .
12
Agricultural Soil Management
13 AAPFCO (2008 through 2014) Commercial Fertilizers. Association of American Plant Food Control Officials.
14 University of Missouri. Columbia, MO.
15 AAPFCO (1995 through 2000a, 2002 through 2007) Commercial Fertilizers. Association of American Plant Food
16 Control Officials. University of Kentucky. Lexington, KY.
17 Bateman, E. J. and E. M. Baggs (2005) "Contributions of nitrification and denitrification to N2O emissions from
18 soils at different water-filled pore space." Biology and Fertility of Soils 41(6): 379-388.
19 Cibrowski, P. (1996) Personal Communication. Peter Cibrowski, Minnesota Pollution Control Agency and Heike
20 Mainhardt, ICF Incorporated. July 29, 1996.
21 CTIC (2004) 2004 Crop Residue Management Survey. Conservation Technology Information Center. Available at
22 .
23 Del Grosso, S.J., A.R. Mosier, W.J. Parton, and D.S. Ojima (2005) "DAYCENT Model Analysis of Past and
24 Contemporary Soil N2O and Net Greenhouse Gas Flux for Major Crops in the USA." Soil Tillage and Research, 83:
25 9-24. doi: 10.1016/j.still.2005.02.007.
26 Del Grosso, S.J., S.M. Ogle, W.J. Parton, and F.J. Breidt (2010) "Estimating Uncertainty in N2O Emissions from
27 U.S. Cropland Soils." Global Biogeochemical Cycles, 24, GB1009, doi:10.1029/2009GB003544.
28 Del Grosso, S.J., W.J. Parton, C.A. Keough, and M. Reyes-Fox. (2011) Special features of the DayCent modeling
29 package and additional procedures for parameterization, calibration, validation, and applications, in Methods of
30 Introducing System Models into Agricultural Research, L.R. Ahuja and Liwang Ma, editors, p. 155-176, American
31 Society of Agronomy, Crop Science Society of America, Soil Science Society of America, Madison, WI. USA.
32 Del Grosso, S.J., W.J. Parton, A.R. Mosier, M.D. Hartman, J. Brenner, D.S. Ojima, and D.S. Schimel (2001)
33 "Simulated Interaction of Carbon Dynamics and Nitrogen Trace Gas Fluxes Using the DAYCENT Model." In
34 Schaffer, M., L. Ma, S. Hansen, (eds.). Modeling Carbon and Nitrogen Dynamics for Soil Management. CRC Press.
35 Boca Raton, Florida. 303-332.
36 Del Grosso, S.J., T. Wirth, S.M. Ogle, W.J. Parton (2008) Estimating agricultural nitrous oxide emissions. EOS 89,
37 529-530.
38 Delgado, J.A., S.J. Del Grosso, and S.M. Ogle (2009) "15N isotopic crop residue cycling studies and modeling
39 suggest that IPCC methodologies to assess residue contributions to N2O-N emissions should be reevaluated."
40 Nutrient Cycling in Agroecosystems, DOI 10.1007/sl0705-009-9300-9.
41 Edmonds, L., N. Gollehon, R.L. Kellogg, B. Kintzer, L. Knight, C. Lander, J. Lemunyon, D. Meyer, D.C. Moffitt,
42 and J. Schaeffer (2003) "Costs Associated with Development and Implementation of Comprehensive Nutrient
43 Management Plans." Part 1. Nutrient Management, Land Treatment, Manure and Wastewater Handling and
44 Storage, and Recordkeeping. Natural Resource Conservation Service, U.S. Department of Agriculture.
45 EPA (2003) Clean Watersheds Needs Survey 2000—Report to Congress, U.S. Environmental Protection Agency.
46 Washington, D.C. Available online at .
References 10-43
-------
1 EPA (1999) Biosolids Generation, Use and Disposal in the United States. Office of Solid Waste, U.S.
2 Environmental Protection Agency. Available online at .
3 EPA (1993) Federal Register. Part II. Standards for the Use and Disposal of Sewage Sludge; Final Rules. U.S.
4 Environmental Protection Agency, 40 CFR Parts 257, 403, and 503.
5 Firestone, M. K., and E.A. Davidson, Ed. (1989). Microbiological basis of NO and N2O production and
6 consumption in soil. Exchange of trace gases between terrestrial ecosystems and the atmosphere. New York, John
7 Wiley & Sons.
8 H. Berbery, M. B. Ek, Y. Fan, R. Grumbine, W. Higgins, H. Li, Y. Lin, G. Manikin, D. Parrish, and W. Shi (2006)
9 North American regional reanalysis. Bulletin of the American Meteorological Society 87:343-360.
10 ILENR (1993) Illinois Inventory of Greenhouse Gas Emissions and Sinks: 1990. Office of Research and Planning,
11 Illinois Department of Energy and Natural Resources. Springfield, IL.
12 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
13 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
14 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
15 Kessavalou, A., A.R. Mosier, J.W. Doran, R.A. Drijber, D.J. Lyon, and O. Heinemeyer (1998). "Fluxes of carbon
16 dioxide, nitrogen oxide, and methane in grass sod and winter wheat -fallow tillage management." Journal of
17 Environmental Quality 27: 1094-1104.
18 McFarland, M.J. (2001) Biosolids Engineering, New York: McGraw-Hill, p. 2.12.
19 McGill, W.B., and C.V. Cole (1981) Comparative aspects of cycling of organic C, N, S and P through soil organic
20 matter. Geoderma 26:267-286.
21 Mesinger, F., G. DiMego, E. Kalnay, K. Mitchell, P. C. Shafran, W. Ebisuzaki, D. Jovic, J. Woollen, E. Rogers, E.
22 Mosier, A. R., J.M. Duxbury, J.R. Freney, O. Heinemeyer, K. Minami (1998) "Assessing and mitigating N2O
23 emissions from agricultural soils." Climatic Change 40: 7-38.
24 NASS (2004) Agricultural Chemical Usage: 2003 Field Crops Summary. Report AgChl(04)a, National Agricultural
25 Statistics Service, U.S. Department of Agriculture. Available online at
26 .
27 NASS (1999) Agricultural Chemical Usage: 1998 Field Crops Summary. Report AgChl(99). National Agricultural
28 Statistics Service, U.S. Department of Agriculture. Available online at
29 .
30 NASS (1992) Agricultural Chemical Usage: 1991 Field Crops Summary. Report AgChl(92). National Agricultural
31 Statistics Service, U.S. Department of Agriculture. Available online at
32 .
33 NEBRA (2007) A National Biosolids Regulation, Quality, End Use & Disposal Survey. North East Biosolids and
34 Residuals Association, July 21, 2007
35 Noller, J. (1996) Personal Communication. John Noller, Missouri Department of Natural Resources and Heike
36 Mainhardt, ICF Incorporated. July 30, 1996.
37 Nusser, S.M., J.J. Goebel (1997) The national resources inventory: a long term monitoring programme.
38 Environmental and Ecological Statistics, 4, 181-204.
39 Oregon Department of Energy (1995) Report on Reducing Oregon's Greenhouse Gas Emissions: Appendix D
40 Inventory and Technical Discussion. Oregon Department of Energy. Salem, OR.
41 Parton, W.J., M.D. Hartman, D.S. Ojima, and D.S. Schimel (1998) "DAYCENT: Its Land Surface Submodel:
42 Description and Testing". Glob. Planet. Chang. 19: 35-48.
43 Potter, C., S. Klooster, A. Huete, and V. Genovese (2007) Terrestrial carbon sinks for the United States predicted
44 from MODIS satellite data and ecosystem modeling. Earth Interactions 11, Article No. 13, DOI 10.1175/EI228.1.
10-44 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Potter, C. S., J.T. Randerson, C.B. Fields, P.A. Matson, P.M. Vitousek, H.A. Mooney, and S.A. Klooster (1993)
2 "Terrestrial ecosystem production: a process model based on global satellite and surface data." Global
3 Biogeochemical Cycles 7:811-841.
4 Ruddy B.C., D.L. Lorenz, and D.K. Mueller (2006) County-level estimates of nutrient inputs to the land surface of
5 the conterminous United States, 1982-2001. Scientific Investigations Report 2006-5012. U.S Department of the
6 Interior.
7 Scheer, C., S.J. Del Grosso, W.J. Parton, D.W. Rowlings, P.R. Grace (2013) Modeling Nitrous Oxide Emissions
8 from Irrigated Agriculture: Testing DAYCENT with High Frequency Measurements, Ecological Applications, in
9 press, .
10 Soil Survey Staff (2011) State Soil Geographic (STATSGO) Database for State. Natural Resources Conservation
11 Service, United States Department of Agriculture. Available online at
12 .
13 Towery, D. (2001) Personal Communication. Dan Towery regarding adjustments to the CTIC (1998) tillage data to
14 reflect long-term trends, Conservation Technology Information Center, West Lafayette, IN, and Marlen Eve,
15 National Resource Ecology Laboratory, Fort Collins, CO. February 2001.
16 TVA (1991 through 1992a, 1993 through 1994) Commercial Fertilizers. Tennessee Valley Authority, Muscle
17 Shoals, AL.
18 USDA-ERS (2011) Agricultural Resource Management Survey (ARMS) Farm Financial and Crop Production
19 Practices: Tailored Reports. Online at: .
21 USDA-ERS (1997) Cropping Practices Survey Data—1995. Economic Research Service, United States Department
22 of Agriculture. Available online at .
23 USDA-NASS (2014) Quick Stats. National Agricultural Statistics Service, United States Department of
24 Agriculture, Washington, D.C. .
25 USDA-NRCS (2015) Summary Report: 2012 National Resources Inventory, Natural Resources Conservation
26 Service, Washington, D.C., and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
27 .
28 USDA-NRCS (2013) Summary Report: 2010 National Resources Inventory, Natural Resources Conservation
29 Service, Washington, D.C, and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
30
31 Vogelman, J.E., S.M. Howard, L. Yang, C. R. Larson, B. K. Wylie, and J. N. Van Driel (2001) "Completion of the
32 1990's National Land Cover Data Set for the conterminous United States." Photogrammetric Engineering and
33 Remote Sensing, 67:650-662.
34 Wisconsin Department of Natural Resources (1993) Wisconsin Greenhouse Gas Emissions: Estimates for 1990.
3 5 Bureau of Air Management, Wisconsin Department of Natural Resources, Madison, WI.
36 Field Burning of Agricultural Residues
37 Barnard, G., andL. Kristoferson (1985) Agricultural Residues as Fuel in the Third World. Earthscan Energy
38 Information Programme and the Beijer Institute of the Royal Swedish Academy of Sciences. London, England.
39 Cantens, G. (2004 through 2005) Personal Communication. Janet Lewis, Assistant to Gaston Cantens, Vice
40 President of Corporate Relations, Florida Crystals Company and ICF International.
41 Deren, C. (2002) Personal communication. Dr. Chris Deren, Everglades Research and Education Centre at the
42 University of Florida and Caren Mintz, ICF International. August 15, 2002.
43 EPA (1994) International Anthropogenic Methane Emissions: Estimates for 1990, Report to Congress. EPA 230-R-
44 93-010. Office of Policy Planning and Evaluation, U.S. Environmental Protection Agency, Washington, D.C.
References 10-45
-------
1 Gonzalez, R. (2007 through 2014) Email correspondence. Rene Gonzalez, Plant Manager, Sem-Chi Rice Company
2 and ICF International.
3 Huang, Y., W. Zhang, W. Sun, and X. Zheng (2007) "Net Primary Production of Chinese Croplands from 1950 to
4 1999." Ecological Applications, 17(3):692-701.
5 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
6 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
7 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
8 IPCC/UNEP/OECD/IEA (1997) Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories.
9 Intergovernmental Panel on Climate Change, United Nations Environment Programme, Organization for Economic
10 Co-Operation and Development, International Energy Agency, Paris, France.
11 Kinoshita, C.M. (1988)" Composition and processing of burned and unburned cane in Hawaii." Intl. Sugar Jnl.
12 90:1070,34-37.
13 Kirstein, A. (2003 through 2004) Personal Communication. Arthur Kirstein, Coordinator, Agricultural Economic
14 Development Program, Palm Beach County Cooperative Extension Service, Florida and ICF International.
15 Lachnicht, S.L., P.P. Hendrix, R.L. Potter, D.C. Coleman, and D.A. Crossley Jr. (2004) "Winter decomposition of
16 transgenic cotton residue in conventional-till and no-till systems." Applied Soil Ecology, 27:135-142.
17 Lee, D. (2003 through 2007) Email correspondence. Danny Lee, OK Farm Service Agency and ICF International.
18 McCarty, J.L. (2011) "Remote Sensing-Based Estimates of Annual and Seasonal Emissions from Crop Residue
19 Burning in the Contiguous United States." Journal of the Air & Waste Management Association, 61:1,22-34, DOI:
20 10.3155/1047-3289.61.1.22.
21 McCarty, J.L. (2010) Agricultural Residue Burning in the Contiguous United States by Crop Type and State.
22 Geographic Information Systems (GIS) Data provided to the EPA Climate Change Division by George Pouliot,
23 Atmospheric Modeling and Analysis Division, EPA. Dr. McCarty's research was supported by the NRI Air Quality
24 Program of the Cooperative State Research, Education, and Extension Service, USD A, under Agreement No.
25 20063511216669 and the NASA Earth System Science Fellowship.
26 McCarty, J.L. (2009) Seasonal and Inter annual Variability of Emissions from Crop Residue Burning in the
27 Contiguous United States. Dissertation. University of Maryland, College Park.
28 Murphy, W.J. (1993). "Tables for weights and measurement: crops". Extension publications. (University of Missouri
29 Extension) .
30 Schueneman, T. (1999 through 2001) Personal Communication. Tom Schueneman, Agricultural Extension Agent,
31 Palm Beach County, FL and ICF International. July 30, 2001.Schueneman, T.J. and C.W. Deren (2002) "An
32 Overview of the Florida Rice Industry." SS-AGR-77, Agronomy Department, Florida Cooperative Extension
33 Service, Institute of Food and Agricultural Sciences, University of Florida. Revised November 2002.
34 Strehler, A., and W. Stiitzle (1987) "Biomass Residues." In Hall, D.O. and Overend, R.P. (eds.). Biomass. John
35 Wiley and Sons, Ltd. Chichester, UK.
36 Turn, S.Q., B.M. Jenkins, J.C. Chow, L.C. Pritchett, D. Campbell, T. Cahill, and S.A. Whalen (1997) "Elemental
37 characterization of paniculate matter emitted from biomass burning: Wind tunnel derived source profiles for
38 herbaceous and wood fuels." Journal of Geophysical Research 102(D3):3683-3699.
39 USDA (2013) 2010 National Resources Inventory, Natural Resources Conservation Service, Washington DC and
40 Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
41 USDA (2015) Quick Stats: U.S. & All States Data; Crops; Production and Area Harvested; 1990 - 2014. National
42 Agricultural Statistics Service, U.S. Department of Agriculture. Washington, D.C. U.S. Department of Agriculture,
43 National Agricultural Statistics Service. Washington, D.C., Available online at .
10-46 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
i Land Use, Land-Use Change, and Forestry
2 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
3 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
4 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
5 UNFCCC (2013) Report of the Conference of the Parties on its nineteenth session, held in Warsaw from 11 to 23
6 November 2013. Available at
7 Representation of the U.S. Land Base
8 Alaska Department of Natural Resources (2006) Alaska Infrastructure 1:63,360. Available online at
9 .
10 Alaska Interagency Fire Management Council (1998) Alaska Interagency Wildland Fire Management Plan.
11 Available online at .
12 Alaska Oil and Gas Conservation Commission (2009) Oil and Gas Information System. Available online at
13 .
14 EIA (2011) Coal Production and Preparation Report Shapefile. Available online at .
16 ESRI (2008) ESRI Data & Maps. Redlands, CA: Environmental Systems Research Institute. [CD-ROM]
17 Fry, I, Xian, G., Jin, S., Dewitz, J., Homer, C., Yang, L., Barnes, C., Herold, N., and J. Wickham. (2011)
18 Completion of the 2006 National Land Cover Database for the Conterminous United States, PE&RS, Vol.
19 77(9):858-864.
20 Homer, C., J. Dewitz, J. Fry, M. Coan, N. Hossain, C. Larson, N. Herold, A. McKerrow, J.N. VanDriel and J.
21 Wickham. (2007) Completion of the 2001 National Land Cover Database for the Conterminous United States,
22 Photogrammetric Engineering and Remote Sensing, Vol. 73, No. 4, pp 337-341.
23 IPCC (2010) Revisiting the use of managed land as a proxy for estimating national anthropogenic emissions and
24 removals. Eggleston HS, Srivastava N, Tanabe K, Baasansuren J, (eds.).Institute for Global Environmental Studies,
25 Intergovernmental Panel on Climate Change, Hayama, Kanagawa, Japan.
26 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
27 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
28 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
29 Jin, S., L. Yang, P. Danielson, C. Homer, J. Fry, and G. Xian. (2013) A comprehensive change detection method for
30 updating the National Land Cover Database to circa 2011. Remote Sensing of Environment, 132: 159-175.
31 NOAA Coastal Change Analysis Program (C-CAP) Regional Land Cover Database. Data collected 1995-present.
32 Charleston, SC: National Oceanic and Atmospheric Administration (NOAA) Coastal Services Center. Data accessed
33 at .
34 Nusser, S.M. and J.J. Goebel (1997) "The national resources inventory: a long-term multi-resource monitoring
35 programme." Environmental and Ecological Statistics 4:181-204.
36 Smith, W.B., P.O. Miles, C.H. Perry, and S.A. Pugh (2009) Forest Resources of the United States, 2007. Gen.
37 Tech. Rep. WO-78. U.S. Department of Agriculture Forest Service, Washington, D.C.
38 U.S. Census Bureau (2010) Topologically Integrated Geographic Encoding and Referencing (TIGER) system
39 shapefiles. U.S. Census Bureau, Washington, D.C. Available online at .
40 U.S. Department of Agriculture (2014) County Data - Livestock, 1990-2014. U.S. Department of Agriculture,
41 National Agriculture Statistics Service, Washington, D.C.
References 10-47
-------
1 U.S. Department of Interior (2005) Federal Lands of the United States. National Atlas of the United States, U.S.
2 Department of the Interior, Washington D.C. Available online at
3 .
4 United States Geological Survey (USGS), Gap Analysis Program (2012) Protected Areas Database of the United
5 States (PADUS), version 1.3 Combined Feature Class. November 2012.
6 USGS (2012) Alaska Resource Data File. Available online at .
7 USGS (2005) Active Mines and Mineral Processing Plants in the United States in 2003. U.S. Geological Survey,
8 Reston, VA.
9 Forest Land Remaining Forest Land: Changes in Forest Carbon
10 Stocks
11 AF&PA (2006a and earlier) Statistical roundup. (Monthly). Washington, D.C. American Forest & Paper
12 Association.
13 AF&PA (2006b and earlier) Statistics of paper, paperboard and wood pulp. Washington, D.C. American Forest &
14 Paper Association.
15 Amichev, B.Y. and J.M. Galbraith (2004) "A Revised Methodology for Estimation of Forest Soil Carbon from
16 Spatial Soils and Forest Inventory Data Sets." Environmental Management 33(Suppl. 1):S74-S86.
17 Bechtold, W.A.; Patterson, P.L. (2005) The enhanced forest inventory and analysis program—national sampling
18 design and estimation procedures. Gen. Tech. Rep. SRS-80. Asheville, NC: U.S. Department of Agriculture Forest
19 Service, Southern Research Station. 85 p.
20 Birdsey, R. (1996) "Carbon Storage for Major Forest Types and Regions in the Conterminous United States." In
21 R.N. Sampson and D. Hair, (eds.). Forest and Global Change, Volume 2: Forest Management Opportunities for
22 Mitigating Carbon Emissions. American Forests. Washington, D.C., 1-26 and 261-379 (appendices 262 and 263)..
23 Coulston, J.W., Wear, D.N., and Vose, J.M. 2015. Complex forest dynamics indicate potential for slowing carbon
24 accumulation in the southeastern United States. Scientific Reports. 5:8002.
25 Domke, G.M., J.E. Smith, and C.W. Woodall. (2011) Accounting for density reduction and structural loss in
26 standing dead trees: Implications for forest biomass and carbon stock estimates in the United States. Carbon
27 Balance and Management. 6:14.
28 Domke, G.M., Woodall, C.W., Smith, J.E., Westfall, J.A., McRoberts, R.E. (2012) Consequences of alternative
29 tree-level biomass estimation procedures on U. S. forest carbon stock estimates. Forest Ecology and Management.
30 270: 108-116
31 Domke, G.M., Perry, C.H., Walters, B.F., Woodall, C. W., and Smith, J.E. (in review). A framework for estimating
32 litter carbon stocks in forests of the United States. Intended outlet: Science of the Total Environment
33 Domke, G.M., Woodall, C.W., Walters, B.F., Smith, J.E. (2013). From models to measurements: comparing down
34 dead wood carbon stock estimates in the U.S. forest inventory. PLoS ONE 8(3): e59949.
35 Domke, G.M., Perry, C.H., Walters, B.F., Nave, L.E., Woodall, C.W., Swanston, C.W. In Prep. Towards field -
36 based estimates of soil organic carbon in forests of the United States.
37 EPA (2006) Municipal solid waste in the United States: 2005 Facts and figures. Office of Solid Waste, U.S.
38 Environmental Protection Agency. Washington, D.C. (5306P) EPA530-R-06-011. Available online at
3 9 .
40 Prayer, W.E., and G.M. Furnival (1999) "Forest Survey Sampling Designs: A History." Journal of Forestry 97(12):
41 4-10.
42 Freed, R. (2004) Open-dump and Landfill timeline spreadsheet (unpublished). ICF International. Washington, D.C.
43 Hair, D. (1958) "Historical forestry statistics of the United States." Statistical Bull. 228. U.S. Department of
44 Agriculture Forest Service, Washington, D.C.
10-48 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Hair. D. and A.H. Ulrich (1963) The Demand and price situation for forest products - 1963. U.S. Department of
2 Agriculture Forest Service, Misc Publication No. 953. Washington, D.C.
3 Harmon, M.E., C.W. Woodall, B. Fasth, J. Sexton, M. Yatkov. (2011) Differences between standing and downed
4 dead tree wood density reduction factors: A comparison across decay classes and tree species. Res. Paper. NRS-15.
5 Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 40 p.
6 Howard, J. L. (forthcoming) U.S. timber production, trade, consumption, and price statistics 1965 to 2013. Res.
7 Pap. FPL-RP-XXX. Madison, WI: USD A, Forest Service, Forest Products Laboratory.
8 Howard, J. L. (2007) U.S. timber production, trade, consumption, and price statistics 1965 to 2005. Res. Pap. FPL-
9 RP-637. Madison, WI: USD A, Forest Service, Forest Products Laboratory.
10 Howard, J. L. (2003) U.S. timber production, trade, consumption, and price statistics 1965 to 2002. Res. Pap. FPL-
11 RP-615. Madison, WI: USD A, Forest Service, Forest Products Laboratory. Available online at
12 .
13 Ince, P.J., Kramp, A.D., Skog, K.E., Spelter, H.N. and Wear, D.N., 2011. US Forest Products Module: a technical
14 document supporting the forest service 2010 RPA assessment. Research Paper-Forest Products Laboratory, USDA
15 Forest Service, (FPL-RP-662).
16 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
17 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
18 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
19 IPCC 2007. Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
20 Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen,
21 M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.)]. Cambridge University Press, Cambridge, United
22 Kingdom and New York, NY, USA, 996 pp.
23 Jenkins, J.C., D.C. Chojnacky, L.S. Heath, and R.A. Birdsey (2003) "National-scale biomass estimators for United
24 States tree species." Forest Science 49(1): 12-35.
25 Jandl, R., Rodeghiero, M., Martinez, C., Cotrufo, M. F., Bampa, F., van Wesemael, B., Harrison, R.B., Guerrini,
26 I.A., deB Richter Jr., D., Rustad, L., Lorenz, K., Chabbi, A., Miglietta, F. 2014. Current status, uncertainty and
27 future needs in soil organic carbon monitoring. Science of the Total Environment, 468, 376-383.
28 Ogle, S.M., Woodall, C.W., Swan, A., Smith, J.E., Wirth. T. In preparation. Determining the Managed Land Base
29 for Delineating Carbon Sources and Sinks in the United States. Environmental Science and Policy.
30 O'Neill, K.P., Amacher, M.C., Perry, C.H. 2005. Soils as an indicator of forest health: a guide to the collection,
31 analysis, and interpretation of soil indicator data in the Forest Inventory and Analysis program. Gen. Tech. Rep. NC-
32 258. St. Paul, MN: US Department of Agriculture, Forest Service, North Central Research Station. 53 p.
33 Oswalt, S.N.; Smith. W.B.; Miles, P.O.; Pugh, S.A. 2014. Forest Resources of the United States, 2012. Gen. Tech.
34 Rep. WO-91. Washington, D.C. U.S. Department of Agriculture, Forest Service, Washington Office. 218 p.
35 Perry, C.H., C.W. Woodall, and M. Schoeneberger (2005) Inventorying trees in agricultural landscapes: towards an
36 accounting of "working trees". In: "Moving Agroforestry into the Mainstream." Proc. 9th N. Am. Agroforestry
37 Cow/, Brooks, K.N. and Folliott, P.P. (eds.). 12-15 June 2005, Rochester, MN [CD-ROM]. Dept. of Forest
38 Resources, Univ. Minnesota, St. Paul, MN, 12 p. Available online at . (verified
39 23 Sept 2006).
40 Russell, M.B.; D'Amato, A.W.; Schulz, B.K.; Woodall, C.W.; Domke, G.M.; Bradford, J.B. 2014 Quantifying
41 understory vegetation in the U.S. Lake States: a proposed framework to inform regional forest carbon stocks.
42 Forestry. 87:629-638.
43 Russell, M.B.; Domke, G.M.; Woodall, C.W.; D'Amato, A.W. 2015. Comparisons of allometric and climate-
44 derived estimates of tree coarse root carbon in forests of the United States. Carbon Balance and Management. 10:
45 20.
46
47 Skog, K.E. (2008) Sequestration of carbon in harvested wood products for the United States. Forest Products
48 Journal 58:56-72.
References 10-49
-------
1 Smith, W. B., P. D. Miles, C. H. Perry, and S. A. Pugh (2009) Forest Resources of the United States, 2007. General
2 Technical Report WO-78, U.S. Department of Agriculture Forest Service, Washington Office.
3 Steer, Henry B. (1948) Lumber production in the United States. Misc. Pub. 669, U.S. Department of Agriculture
4 Forest Service. Washington, D.C.
5 Ulrich, Alice (1985) U.S. Timber Production, Trade, Consumption, and Price Statistics 1950-1985. Misc. Pub.
6 1453, U.S. Department of Agriculture Forest Service. Washington, D.C.
7 Ulrich, A.H. (1989) U.S. Timber Production, Trade, Consumption, and Price Statistics, 1950-1987. USDA
8 Miscellaneous Publication No. 1471, U.S. Department of Agriculture Forest Service. Washington, D.C, 77.
9 United Nations Framework Convention on Climate Change. 2013. Report on the individual review of the inventory
10 submission of the United States of America submitted in 2012. FCCC/ARR/2012/US A. 42 p.
11 USDA Forest Service (2015a) Forest Inventory and Analysis National Program: Program Features. U.S. Department
12 of Agriculture Forest Service. Washington, D.C. Available online at .
13 Accessed 17 September 2015.
14 USDA Forest Service. (2015b) Forest Inventory and Analysis National Program: FIA Data Mart. U.S. Department
15 of Agriculture Forest Service. Washington, D.C. Available online at . Accessed 17 September 2015.
17 USDA Forest Service. (2015c) Forest Inventory and Analysis National Program, FIA library: Field Guides, Methods
18 and Procedures. U.S. Department of Agriculture Forest Service. Washington, D.C. Available online at
19 . Accessed 17 September 2015.
20 USDA Forest Service (2015d) Forest Inventory and Analysis National Program, FIA library: Database
21 Documentation. U.S. Department of Agriculture, Forest Service, Washington Office. Available online at
22 . Accessed 17 September 2015.
23 U.S. Census Bureau (1976) Historical Statistics of the United States, Colonial Times to 1970, Vol. 1. Washington,
24 D.C.
25 Wear, D.N., Coulston, J.W. 2015. From sink to source: Regional variation in U.S. forest carbon futures. Scientific
26 Reports. 5: 16518.
27 Westfall, J.A., Woodall, C.W., Hatfield, M.A. (2013) A statistical power analysis of woody carbon flux from forest
28 inventory data. Climatic Change. 118:919-931.
29 Woodall, C.W., Coulston, J.W., Domke, G.M., Walters, B.F., Wear, D.N., Smith, J.E., Anderson, H.-E., Clough,
30 B.J., Cohen, W.B., Griffith, D.M., Hagan, S.C., Hanou, I.S.; Nichols, M.C., Perry, C.H., Russell, M.B., Westfall,
31 J.A., Wilson, B.T. 2015a. The US Forest Carbon Accounting Framework: Stocks and Stock change 1990-2016.
32 Gen. Tech. Rep. NRS-154. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern
33 Research Station. 49 pp.
34 Woodall, C.W. (2012) Where did the U.S. forest biomass/carbon go? Journal of Forestry. 110:113-114.
35 Woodall, C.W., Amacher, M.C., Bechtold, W.A., Coulston, J.W., Jovan, S., Perry, C.H., Randolph, K.C., Schulz,
36 B.K., Smith, G.C., Tkacz, B., Will-Wolf, S. (201 Ib) "Status and future of the forest health indicators program of the
37 United States." Environmental Monitoring and Assessment. 177:419-436.
38 Woodall, C.W., Domke, G.M., MacFarlane, D.W., Oswalt, C.M. (2012) Comparing Field- and Model-Based
39 Standing Dead Tree Carbon Stock Estimates Across Forests of the United States. Forestry 85(1): 125-133.
40 Woodall, C.W., Domke, G.M., Riley, K, Oswalt, C.M., Crocker, S.J. Yohe, G.W. (2013) Developing a framework
41 for assessing global change risks to forest carbon stocks. PLOS One. 8: e73222.
42 Woodall, C. W., L.S. Heath, G.M. Domke, and M.C. Nichols (201 la) Methods and equations for estimating
43 aboveground volume, biomass, and carbon for trees in the U.S. forest inventory, 2010. Gen. Tech. Rep. NRS-88.
44 Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 30 p.
10-50 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Woodall, C. W., and V.J. Monleon (2008) Sampling protocol, estimation, and analysis procedures for the down
2 woody materials indicator of the FIA program. Gen. Tech. Rep. NRS-22. Newtown Square, PA: U.S. Department of
3 Agriculture, Forest Service, Northern Research Station. 68 p.
4 Woodall, C.W., Walters, B.F., Oswalt, S.N., Domke, G.M., Toney, C., Gray, A.N. (2013) Biomass and carbon
5 attributes of downed woody materials in forests of the United States. Forest Ecology and Management 305: 48-59.
6 Woodall, C.W., Walters, B.F., Coulston, J.W., D'Amato, A.W., Domke, G.M., Russell, M.B., Sowers, P.A. 2015b.
7 Monitoring network confirms land use change is a substantial component of the forest carbon sink in the eastern
8 United States. Scientific Reports. 5: 17028.
9 Forest Land Remaining Forest Land: Non-CO2 Emissions from
10 Forest Fires
11 deVries,R.E. (1987) A Preliminary Investigation of the Growth and Longevity of Trees in Central Park. M.S.
12 thesis, Rutgers University, New Brunswick, NJ.
13 Dwyer, J.F., D. J. Nowak, M.H. Noble, and S.M. Sisinni (2000) Connecting People with Ecosystems in the 21st
14 Century: An Assessment of Our Nation's Urban Forests. General Technical Report PNW-GTR-490, U.S.
15 Department of Agriculture, Forest Service, Pacific Northwest Research Station, Portland, OR.
16 Fleming, L.E. (1988) Growth Estimation of Street Trees in Central New Jersey. M.S. thesis, Rutgers University,
17 New Brunswick, NJ.
18 Frelich, L.E. (1992) Predicting Dimensional Relationships for Twin Cities Shade Trees. University of Minnesota,
19 Department of Forest Resources, St. Paul, MN, p. 33.
20 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
21 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
22 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
23 Nowak, D.J. (2011) Phone conference regarding Changes in Carbon Stocks in Urban Trees estimation methodology.
24 David Nowak, USD A, Jennifer Jenkins, EPA, and Mark Flugge and Nikhil Nadkarni, ICF International. January 4,
25 2011.
26 Nowak, D.J. (2009) E-mail correspondence regarding new data for Chicago's urban forest. David Nowak, USDA
27 Forest Service to Nikhil Nadkarni, ICF International. October 7, 2009.
28 Nowak, D.J. (2007a) "New York City's Urban Forest." USDA Forest Service. Newtown Square, PA, February
29 2007.
30 Nowak, D.J. (2007b) E-mail correspondence regarding revised sequestration values and standard errors for
31 sequestration values. David Nowak, USDA Forest Service to Susan Asam, ICF International. October 31, 2007.
32 Nowak, D.J. (1994) "Atmospheric Carbon Dioxide Reduction by Chicago's Urban Forest." In: Chicago's Urban
33 Forest Ecosystem: Results of the Chicago Urban Forest Climate Project. E.G. McPherson, D.J. Nowak, and R.A.
34 Rowntree (eds.). General Technical Report NE-186. U.S. Department of Agriculture Forest Service, Radnor, PA.
35 pp. 83-94.
36 Nowak, D.J. (1986) "Silvics of an Urban Tree Species: Norway Maple (Acer platanoides L.)." M.S. thesis, College
37 of Environmental Science and Forestry, State University of New York, Syracuse, NY.
38 Nowak, D.J., Buckelew-Cumming, A., Twardus, D., Hoehn, R., and Mielke, M. (2007). National Forest Health
39 Monitoring Program, Monitoring Urban Forests in Indiana: Pilot Study 2002, Part 2: Statewide Estimates Using the
40 UFORE Model. Northeastern Area Report. NA-FR-01e07, p. 13.
41 Nowak, D.J. and D.E. Crane (2002) "Carbon Storage and Sequestration by Urban Trees in the United States."
42 Environmental Pollution 116(3):381-389.
43 Nowak, D.J., D.E. Crane, J.C. Stevens, and M. Ibarra (2002) Brooklyn's Urban Forest. General Technical Report
44 NE-290. U.S. Department of Agriculture Forest Service, Newtown Square, PA.
References 10-51
-------
1 Nowak, D.J., and E. J. Greenfield (2012) Tree and impervious cover in the United States. Journal of Landscape and
1 Urban Planning (107) pp. 21-30.
3 Nowak, D.J., EJ. Greenfield, R.E. Hoehn, and E. Lapoint (2013) Carbon Storage and Sequestration by Trees in
4 Urban and Community Areas of the United States. Environmental Pollution 178: 229-236. March 12, 2013.
5 Nowak, D.J., J.T. Walton, L.G. Kaya, and J.F. Dwyer (2005) "The Increasing Influence of Urban Environments on
6 U.S. Forest Management." Journal of Forestry 103(8):377-382.
7 Smith, W.B. and S.R. Shifley (1984) Diameter Growth, Survival, and Volume Estimates for Trees in Indiana and
8 Illinois. Research Paper NC-257. North Central Forest Experiment Station, U.S. Department of Agriculture Forest
9 Service, St. Paul, MN.
10 U.S. Census Bureau (2012) "A national 2010 urban area file containing a list of all urbanized areas and urban
11 clusters (including Puerto Rico and the Island Areas) sorted by UACE code." U.S. Census Bureau, Geography
12 Division.
13 Forest Land Remaining Forest Land: N2O Fluxes from Soils
14 Albaugh, T.J., Allen, H.L., Fox, T.R. (2007) Historical Patterns of Forest Fertilization in the Southeastern United
15 States from 1969 to 2004. Southern Journal of Applied Forestry, 31, 129-137(9).
16 Binkley, D. (2004) Email correspondence regarding the 95% CI for area estimates of southern pine plantations
17 receiving N fertilizer (±20%) and the rate applied for areas receiving N fertilizer (100 to 200 pounds/acre). Dan
18 Binkley, Department of Forest, Rangeland, and Watershed Stewardship, Colorado State University and Stephen Del
19 Grosso, Natural Resource Ecology Laboratory, Colorado State University. September 19, 2004.
20 Binkley, D., R. Carter, and H.L. Allen (1995) Nitrogen Fertilization Practices in Forestry. In: Nitrogen Fertilization
21 in the Environment, P.E. Bacon (ed.), Marcel Decker, Inc., New York.
22 Briggs, D. (2007) Management Practices on Pacific Northwest West-Side Industrial Forest Lands, 1991-2005: With
23 Projections to 2010. Stand Management Cooperative, SMC Working Paper Number 6, College of Forest Resources,
24 University of Washington, Seattle.
25 Fox, T.R., H. L.Allen, T.J. Albaugh, R. Rubilar, and C. A. Carlson (2007) Tree Nutrition and Forest Fertilization of
26 Pine Plantations in the Southern United States. Southern Journal of Applied Forestry. 31, 5-11.
27 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
28 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
29 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
30 USD A Forest Service (2001) U.S. Forest Facts and Historical Trends. FS-696. U.S. Department of Agriculture
31 Forest Service, Washington, D.C. Available online at .
32 Cropland Remaining Cropland: Mineral and Organic Soil
33 Carbon Stock Changes
34 Armentano, T. V., and E.S. Menges (1986). Patterns of change in the carbon balance of organic soil-wetlands of the
35 temperate zone. Journal of Ecology 74: 755-774.
36 Brady, N.C. and R.R. Weil (1999) The Nature and Properties of Soils. Prentice Hall. Upper Saddle River, NJ, 881.
37 Conant, R. T., K. Paustian, and E.T. Elliott (2001). "Grassland management and conversion into grassland: effects
38 on soil carbon." Ecological Applications 11: 343-355.
39 CTIC (2004) National Crop Residue Management Survey: 1989-2004. Conservation Technology Information
40 Center, Purdue University, Available at
41 Daly, C., R.P. Neilson, and D.L. Phillips (1994) "A Statistical-Topographic Model for Mapping Climatological
42 Precipitation Over Mountainous Terrain." Journal of Applied Meteorology 33:140-158.
10-52 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Del Grosso, S.J., W.J. Parton, A.R. Mosier, M.D. Hartman, J. Brenner, D.S. Ojima, and D.S. Schimel (2001)
2 "Simulated Interaction of Carbon Dynamics and Nitrogen Trace Gas Fluxes Using the DAYCENT Model." In
3 Modeling Carbon and Nitrogen Dynamics for Soil Management, Schaffer, M., L. Ma, S. Hansen, (eds.). CRC Press,
4 Boca Raton, Florida, pp. 303-332.
5 Del Grosso, S.J., S.M. Ogle, W.J. Parton (2011) Soil organic matter cycling and greenhouse gas accounting
6 methodologies, Chapter 1, pp 3-13 DOI: 10.1021/bk-201 l-1072.ch001. In: Understanding Greenhouse Gas
7 Emissions from Agricultural Management, L. Guo, A. Gunasekara, L. McConnell (eds.). American Chemical
8 Society, Washington, D.C.
9 Edmonds, L., R. L. Kellogg, B. Kintzer, L. Knight, C. Lander, J. Lemunyon, D. Meyer, D.C. Moffitt, and J.
10 Schaefer (2003) "Costs associated with development and implementation of Comprehensive Nutrient Management
11 Plans." Part I—Nutrient management, land treatment, manure and wastewater handling and storage, and
12 recordkeeping. Natural Resources Conservation Service, U.S. Department of Agriculture. Available online at
13 .
14 Euliss, N., and R. Gleason (2002) Personal communication regarding wetland restoration factor estimates and
15 restoration activity data. Ned Euliss and Robert Gleason of the U.S. Geological Survey, Jamestown, ND, to Stephen
16 Ogle of the National Resource Ecology Laboratory, Fort Collins, CO. August 2002.
17 Fry, J., Xian, G., Jin, S., Dewitz, J., Homer, C., Yang, L., Barnes, C., Herold, N., and Wickham, J., 2011. Completion of the
18 2006 National Land Cover Database for the Conterminous United States, PE&RS, Vol. 77(9):858-864.
19 Gurung, R.B., F.J. Breidt, A. Dutin, and S.M. Ogle (2009) Predicting Enhanced Vegetation Index (EVI) for
20 ecosystem modeling applications. Remote Sensing of Environment 113:2186-2193.
21 Homer, C., Dewitz, J., Fry, J., Coan, M., Hossain, N., Larson, C., Herold, N., McKerrow, A., VanDriel, J.N., and Wickham,
22 J. 2007. Completion of the 2001 National Land Cover Database for the Conterminous United States. Photogrammetric
23 Engineering and Remote Sensing, Vol. 73, No. 4, pp 337-341.
24 Homer, C.G., Dewitz, J.A., Yang, L., Jin, S., Danielson, P., Xian, G., Coulston, J., Herold, N.D., Wickham, J.D., and
25 Megown, K., 2015, Completion of the 2011 National Land Cover Database for the conterminous United States-Representing
26 a decade of land cover change information. Photogrammetric Engineering and Remote Sensing, v. 81, no. 5, p. 345-354
27 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
28 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
29 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
30 IPCC (2003) Good Practice Guidance for Land Use, Land-Use Change, and Forestry. The Intergovernmental
31 Panel on Climate Change, National Greenhouse Gas Inventories Programme, J. Penman, et al., eds. August 13,
32 2004. Available online at .
33 Jones, C., R. Claassen, E. Marx and S.M. Ogle (forthcoming) GHG mitigation and the CRP: the role of post-contract
34 land use change.
35 McGill, W.B., and C.V. Cole (1981) Comparative aspects of cycling of organic C, N, S and P through soil organic
36 matter. Geoderma 26:267-286.
37 Metherell, A.K., L.A. Harding, C.V. Cole, and W.J. Parton (1993) "CENTURY Soil Organic Matter Model
38 Environment." Agroecosystem version 4.0. Technical documentation, GPSR Tech. Report No. 4, USDA/ARS, Ft.
39 Collins, CO.
40 Mesinger, F., G. DiMego, E. Kalnay, K. Mitchell, P. C. Shafran, W. Ebisuzaki, D. Jovic, J. Woollen, E. Rogers, E.
41 H. Berbery, M. B. Ek, Y. Fan, R. Grumbine, W. Higgins, H. Li, Y. Lin, G. Manikin, D. Parrish, and W. Shi (2006)
42 North American regional reanalysis. Bulletin of the American Meteorological Society 87:343-360.
43 NASS (2004) Agricultural Chemical Usage: 2003 Field Crops Summary. Report AgChl(04)a. National Agricultural
44 Statistics Service, U.S. Department of Agriculture, Washington, D.C. Available online at
45 .
46 NASS (1999) Agricultural Chemical Usage: 1998 Field Crops Summary. Report AgCHl(99). National Agricultural
47 Statistics Service, U.S. Department of Agriculture, Washington, DC. Available online at
48 .
References 10-53
-------
1 NASS (1992) Agricultural Chemical Usage: 1991 Field Crops Summary. Report AgChl(92). National Agricultural
2 Statistics Service, U.S. Department of Agriculture, Washington, D.C. Available online at
3 .
4 NRCS (1999) Soil Taxonomy: A basic system of soil classification for making and interpreting soil surveys, 2nd
5 Edition. Agricultural Handbook Number 436, Natural Resources Conservation Service, U.S. Department of
6 Agriculture, Washington, D.C.
7 NRCS (1997) "National Soil Survey Laboratory Characterization Data," Digital Data, Natural Resources
8 Conservation Service, U.S. Department of Agriculture. Lincoln, NE.
9 NRCS (1981) Land Resource Regions and Major Land Resource Areas of the United States, USDA Agriculture
10 Handbook 296, United States Department of Agriculture, Natural Resources Conservation Service, National Soil
11 Survey Cente., Lincoln, NE, pp. 156.
12 Ogle, S.M., F.J. Breidt, M. Easter, S. Williams, K. Killian, and K. Paustian (2010) "Scale and uncertainty in
13 modeled soil organic carbon stock changes for U.S. croplands using a process-based model." Global Change
14 Biology 16:810-820.
15 Ogle, S.M., F.J. Breidt, M. Easter, S. Williams and K. Paustian. (2007) "Empirically-Based Uncertainty Associated
16 with Modeling Carbon Sequestration Rates in Soils." Ecological Modeling 205:453 -463.
17 Ogle, S.M., F.J. Breidt, and K. Paustian. (2006) "Bias and variance in model results due to spatial scaling of
18 measurements for parameterization in regional assessments." Global Change Biology 12:516-523.
19 Ogle, S. M., et al. (2005) "Agricultural management impacts on soil organic carbon storage under moist and dry
20 climatic conditions of temperate and tropical regions." Biogeochemistry 72: 87-121.
21 Ogle, S.M., M.D. Eve, F.J. Breidt, and K. Paustian (2003) "Uncertainty in estimating land use and management
22 impacts on soil organic carbon storage for U.S. agroecosystems between 1982 and 1997." Global Change Biology
23 9:1521-1542.
24 Ogle, S., M. Eve, M. Sperrow, F.J. Breidt, and K. Paustian (2002) Agricultural Soil C Emissions, 1990-2001:
25 Documentation to Accompany EPA Inventory Tables. Natural Resources Ecology Laboratory, Fort Collins, CO.
26 Provided in an e-mail from Stephen Ogle, NREL to Barbara Braatz, ICF International. September 23, 2002
27 Parton, W.J., M.D. Hartman, D.S. Ojima, and D.S. Schimel (1998) "DAYCENT: Its Land Surface Submodel:
28 Description and Testing". Glob. Planet. Chang. 19: 35-48.
29 Parton, W.J., D.S. Ojima, C.V. Cole, and D.S. Schimel (1994) "A General Model for Soil Organic Matter
30 Dynamics: Sensitivity to litter chemistry, texture and management," in Quantitative Modeling of Soil Forming
31 Processes. Special Publication 39, Soil Science Society of America, Madison, WI, 147-167.
32 Parton, W.J., D.S. Schimel, C.V. Cole, D.S. Ojima (1987) "Analysis of factors controlling soil organic matter levels
33 in Great Plains grasslands." Soil Science Society of America Journal 51:1173-1179.
34 Parton, W.J., J. W.B. Stewart, C.V. Cole. (1988) "Dynamics of C, N, P, and S in grassland soils: a model."
35 Biogeochemistry 5:109-131.
36 Paustian, K., et al. (1997a). "Agricultural soils as a sink to mitigate CO2 emissions." Soil Use and Management 13:
37 230-244.
38 Paustian, K., et al. (1997b). Management controls on soil carbon. In Soil organic matter in temperate
39 agroecosystems: long-term experiments in North America (Paul E.A., K. Paustian, and C.V. Cole, eds). Boca Raton,
40 CRC Press, pp. 15-49.
41 Potter, C. S., J.T. Randerson, C.B. Fields, P.A. Matson, P.M. Vitousek, H.A. Mooney, and S.A. Klooster. (1993)
42 "Terrestrial ecosystem production: a process model based on global satellite and surface data." Global
43 Biogeochemical Cycles 7:811-841.
44 Potter, C., S. Klooster, A. Huete, and V. Genovese (2007) Terrestrial carbon sinks for the United States predicted
45 from MODIS satellite data and ecosystem modeling. Earth Interactions 11, Article No. 13, DOI 10.1175/EI228.1.
46 PRISM Climate Group (2015) PRISM Climate Data. Oregon State University. July 24, 2015. Available online at:
47 .
10-54 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Soil Survey Staff (2015) State Soil Geographic (STATSGO) Database for State. Natural Resources Conservation
2 Service, United States Department of Agriculture. Available online at
3 .
4 Towery, D. (2001) Personal Communication. Dan Towery regarding adjustments to the CTIC (1998) tillage data to
5 reflect long-term trends, Conservation Technology Information Center, West Lafayette, IN, and Marlen Eve,
6 National Resource Ecology Laboratory, Fort Collins, CO. February 2001.
7 USDA-ERS (2011) Agricultural Resource Management Survey (ARMS) Farm Financial and Crop Production
8 Practices: Tailored Reports. Online at: .
9 USDA-ERS (1997) Cropping Practices Survey Data—1995. Economic Research Service, United States Department
10 of Agriculture. Available online at .
11 USD A-FS A (2014) Conservation Reserve Program Monthly Summary - September 2014. U. S. Department of
12 Agriculture, Farm Service Agency, Washington, D.C. Available online at .
14 USDA-NRCS (2000) Digital Data and Summary Report: 1997 National Resources Inventory. Revised December
15 2000. Resources Inventory Division, Natural Resources Conservation Service, United States Department of
16 Agriculture, Beltsville, MD.
17 USDA-NRCS (2015) Summary Report: 2012 National Resources Inventory, Natural Resources Conservation
18 Service, Washington, D.C., and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
19 .
20 USDA-NRCS (2013) Summary Report: 2010 National Resources Inventory, Natural Resources Conservation
21 Service, Washington, D.C., and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
22 .
23 Cropland Remaining Cropland: Liming
24 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
25 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
26 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
27 Tepordei, V.V. (1997 through 2006) "Crushed Stone," InMinerals Yearbook. U.S. Department of the Interior/U.S.
28 Geological Survey. Washington, D.C. Available online at
29 .
30 Tepordei, V.V. (2003b) Personal communication. Valentin Tepordei, U.S. Geological Survey and ICF Consulting,
31 August 18, 2003.
32 Tepordei, V.V. (1996) "Crushed Stone," InMinerals Yearbook 1994. U.S. Department of the Interior/Bureau of
33 Mines, Washington, D.C. Available online at
34 . Accessed August 2000.
35 Tepordei, V.V. (1995) "Crushed Stone," InMinerals Yearbook 1993. U.S. Department of the Interior/Bureau of
36 Mines, Washington, D.C. pp. 1107-1147.
37 Tepordei, V V (1994) "Crushed Stone," InMinerals Yearbook 1992. U.S. Department of the Interior/Bureau of
38 Mines, Washington, D.C. pp. 1279-1303.
39 Tepordei, V.V. (1993) "Crushed Stone," InMinerals Yearbook 1991. U.S. Department of the Interior/Bureau of
40 Mines, Washington, D.C. pp. 1469-1511.
41 USGS (2015) Mineral Industry Surveys: Crushed Stone and Sand and Gravel in the First Quarter of 2015, U.S.
42 Geological Survey, Reston, VA. Available online at
43 .
44 USGS (2014) Mineral Industry Surveys: Crushed Stone and Sand and Gravel in the First Quarter of 2014, U.S.
45 Geological Survey, Reston, VA. Available online at
46 .
References 10-55
-------
1 USGS (2013) Mineral Industry Surveys: Crushed Stone and Sand and Gravel in the First Quarter of 2013, U.S.
2 Geological Survey, Reston, VA. Available online at
3 .
4 USGS (2012) Mineral Industry Surveys: Crushed Stone and Sand and Gravel in the First Quarter of 2012, U.S.
5 Geological Survey, Reston, VA. Available online at
6 .
7 USGS (2011) Mineral Industry Surveys: Crushed Stone and Sand and Gravel in the First Quarter of 2011, U.S.
8 Geological Survey, Reston, VA. Available online at
9 .
10 USGS (2010) Mineral Industry Surveys: Crushed Stone and Sand and Gravel in the First Quarter of 2010, U.S.
11 Geological Survey, Reston, VA. Available online at
12 .
13 USGS (2009) Mineral Industry Surveys: Crushed Stone and Sand and Gravel in the First Quarter of 2009, U.S.
14 Geological Survey, Reston, VA. Available online at
15 .
16 USGS (2008) Mineral Industry Surveys: Crushed Stone and Sand and Gravel in the First Quarter of 2008, U.S.
17 Geological Survey, Reston, VA. Available online at
18 .
19 USGS (2007) Mineral Industry Surveys: Crushed Stone and Sand and Gravel in the First Quarter of 2007. U.S.
20 Geological Survey, Reston, VA. Available online at
21 .
22 West, T.O. (2008) Email correspondence. Tristram West, Environmental Sciences Division, Oak Ridge National
23 Laboratory, U.S. Department of Energy and Nikhil Nadkarni, ICF International on suitability of liming emission
24 factor for the entire United States. June 9, 2008.
25 West, T.O., and A.C. McBride (2005) "The contribution of agricultural lime to carbon dioxide emissions in the
26 United States: dissolution, transport, and net emissions," Agricultural Ecosystems & Environment 108:145-154.
27 Willett, J.C. (2015) "Crushed Stone," InMinerals Yearbook 2013. U.S. Department of the Interior/U.S. Geological
28 Survey, Washington, D.C. Available online at
29 . Accessed September 2015.
30 Willett, J.C. (2014) "Crushed Stone," In Minerals Yearbook 2012. U.S. Department of the Interior/U.S. Geological
31 Survey, Washington, D.C. Available online at
32 . Accessed September 2014.
33 Willett, J.C. (2013a) "Crushed Stone," InMinerals Yearbook 2011. U.S. Department of the Interior/U.S. Geological
34 Survey, Washington, D.C. Available online at
35 . Accessed May 2013.
36 Willett, J.C. (2013b) Personal Communication. Jason Willet, U.S. Geological Survey and ICF International.
37 September 9, 2013.
38 Willett, J.C. (201 la) "Crushed Stone," InMinerals Yearbook 2009. U.S. Department of the Interior/U.S. Geological
39 Survey, Washington, D.C. Available online at
40 . Accessed August 2011.
41 Willett, J.C. (201 Ib) "Crushed Stone," InMinerals Yearbook 2010. U.S. Department of the Interior/U.S. Geological
42 Survey, Washington, D.C. Available online at
43 . Accessed September 2012.
44 Willett, J.C. (2010) "Crushed Stone," In Minerals Yearbook 2008. U.S. Department of the Interior/U.S. Geological
45 Survey, Washington, D.C. Available online at
46 . Accessed August 2010.
10-56 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 Willett, J.C. (2009) "Crashed Stone," InMinerals Yearbook 2007. U.S. Department of the Interior/U.S. Geological
2 Survey, Washington, D.C. Available online at
3 . Accessed August 2009.
4 Willett, J.C. (2007a) "Crashed Stone," InMinerals Yearbook 2005. U.S. Department of the Interior/U.S. Geological
5 Survey, Washington, D.C. Available online at
6 . Accessed August 2007.
7 Willett, J.C. (2007b) "Crashed Stone," InMinerals Yearbook 2006. U.S. Department of the Interior/U.S. Geological
8 Survey, Washington, D.C. Available online at
9 . Accessed August 2008.
10 Cropland Remaining Cropland: Urea Fertilization
11 AAPFCO (2008 through 2014) Commercial Fertilizers. Association of American Plant Food Control Officials.
12 University of Missouri. Columbia, MO.
13 AAPFCO (1995 through 2000a, 2002 through 2007) Commercial Fertilizers. Association of American Plant Food
14 Control Officials. University of Kentucky. Lexington, KY.
15 AAPFCO (2000b) 1999-2000 Commercial Fertilizers Data, ASCII files. Available from David Terry, Secretary,
16 AAPFCO.
17 EPA (2000) Preliminary Data Summary: Airport Deicing Operations (Revised). EPA-821-R-00-016. August 2000.
18 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
19 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
20 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
21 Itle, C. (2009) Email correspondence. Cortney Itle, ERG and Tom Wirth, U.S. Environmental Protection Agency on
22 the amount of urea used in aircraft deicing. January 7, 2009.
23 Terry, D. (2007) Email correspondence. David Terry, Fertilizer Regulatory program, University of Kentucky and
24 David Berv, ICF International. September 7, 2007.
25 TVA (1991 through 1994) Commercial Fertilizers. Tennessee Valley Authority, Muscle Shoals, AL.
26 TVA (1992b) Fertilizer Summary Data 1992. Tennessee Valley Authority, Muscle Shoals, AL.
27
Land Converted to Cropland
28 Del Grosso, S.J., W.J. Parton, A.R. Mosier, M.D. Hartman, J. Brenner, D.S. Ojima, and D.S. Schimel (2001)
29 "Simulated Interaction of Carbon Dynamics and Nitrogen Trace Gas Fluxes Using the DAYCENT Model." In
30 Modeling Carbon and Nitrogen Dynamics for Soil Management, Schaffer, M., L. Ma, S. Hansen, (eds.). CRC Press,
31 Boca Raton, Florida, pp. 303-332.
32 Del Grosso, S.J., S.M. Ogle, W.J. Parton (2011) Soil organic matter cycling and greenhouse gas accounting
33 methodologies, Chapter 1, pp 3-13 DOI: 10.1021/bk-201 l-1072.ch001. In: Understanding Greenhouse Gas
34 Emissions from Agricultural Management (L. Guo, A. Gunasekara, L. McConnell. Eds.), American Chemical
35 Society, Washington, D.C.
36 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
37 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
38 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
39 Metherell, A.K., L. A. Harding, C.V. Cole, and W.J. Parton (1993) "CENTURY Soil Organic Matter Model
40 Environment." Agroecosystem version 4.0. Technical documentation, GPSR Tech. Report No. 4, USDA/ARS, Ft.
41 Collins, CO.
42 Ogle, S.M., F.J. Breidt, M. Easter, S. Williams, K. Killian, and K. Paustian (2010) "Scale and uncertainty in
43 modeled soil organic carbon stock changes for U.S. croplands using a process-based model." Global Change
44 Biology 16:810-820.
References 10-57
-------
1 Ogle, S.M., M.D. Eve, F. J. Breidt, and K. Paustian (2003) "Uncertainty in estimating land use and management
2 impacts on soil organic carbon storage for U.S. agroecosystems between 1982 and 1997." Global Change Biology
3 9:1521-1542.
4 Parton, W.J., M.D. Hartman, D.S. Ojima, and D.S. Schimel (1998) "DAYCENT: Its Land Surface Submodel:
5 Description and Testing". Glob. Planet. Chang. 19: 35-48.
6 Parton, W.J., D.S. Ojima, C.V. Cole, and D.S. Schimel (1994) "A General Model for Soil Organic Matter
7 Dynamics: Sensitivity to litter chemistry, texture and management," in Quantitative Modeling of Soil Forming
8 Processes. Special Publication 39, Soil Science Society of America, Madison, WI, 147-167.
9 USDA-NRCS (2015) Summary Report: 2012 National Resources Inventory, Natural Resources Conservation
10 Service, Washington, D.C., and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
11 .
12 USDA-NRCS (2013) Summary Report: 2010 National Resources Inventory, Natural Resources Conservation
13 Service, Washington, D.C. and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
14 .
15 Grassland Remaining Grassland
16 Del Grosso, S.J., S.M. Ogle, WJ. Parton (2011) Soil organic matter cycling and greenhouse gas accounting
17 methodologies, Chapter 1, pp 3-13 DOI: 10.1021/bk-201 l-1072.ch001. In: Understanding Greenhouse Gas
18 Emissions from Agricultural Management (L. Guo, A. Gunasekara, L. McConnell. Eds.), American Chemical
19 Society, Washington, D.C.
20 Del Grosso, S.J., WJ. Parton, A.R. Mosier, M.D. Hartman, J. Brenner, D.S. Ojima, and D.S. Schimel (2001)
21 "Simulated Interaction of Carbon Dynamics and Nitrogen Trace Gas Fluxes Using the DAYCENT Model." In
22 Modeling Carbon and Nitrogen Dynamics for Soil Management, Schaffer, M., L. Ma, S. Hansen, (eds.). CRC Press,
23 Boca Raton, Florida, pp. 303-332.
24 Edmonds, L., R. L. Kellogg, B. Kintzer, L. Knight, C. Lander, J. Lemunyon, D. Meyer, D.C. Moffitt, and J.
25 Schaefer (2003) "Costs associated with development and implementation of Comprehensive Nutrient Management
26 Plans." Part I—Nutrient management, land treatment, manure and wastewater handling and storage, and
27 recordkeeping. Natural Resources Conservation Service, U.S. Department of Agriculture. Available online at
28 .
29 EPA (1999) Biosolids Generation, Use and Disposal in the United States. Office of Solid Waste, U.S.
30 Environmental Protection Agency. Available online at .
31 Fry, J., Xian, G., Jin, S., Dewitz, J., Homer, C., Yang, L., Barnes, C., Herold, N., and Wickham, J., 2011. Completion of
32 the 2006 National Land Cover Database for the Conterminous United States, PE&RS, Vol. 77(9):858-864.
33 Homer, C., Dewitz, J., Fry, J., Coan, M., Hossain, N., Larson, C., Herold, N., McKerrow, A., VanDriel, J.N., and Wickham,
34 J. 2007. Completion of the 2001 National Land Cover Database for the Conterminous United States. Photogrammetric
35 Engineering and Remote Sensing, Vol. 73, No. 4, pp 337-341.
36 Homer, C.G., Dewitz, J.A., Yang, L., Jin, S., Danielson, P., Xian, G., Coulston, J., Herold, N.D., Wickham, J.D., and
37 Megown, K., 2015, Completion of the 2011 National Land Cover Database for the conterminous United States-Representing
38 a decade of land cover change information. Photogrammetric Engineering and Remote Sensing, v. 81, no. 5, p. 345-354
39 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
40 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
41 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
42 Kellogg, R.L., C.H. Lander, D.C. Moffitt, and N. Gollehon (2000) Manure Nutrients Relative to the Capacity of
43 Cropland and Pastureland to Assimilate Nutrients: Spatial and Temporal Trends for the United States. U.S.
44 Department of Agriculture, Washington, D.C. Publication number npsOO-0579.
45 Metherell, A.K., L.A. Harding, C.V. Cole, and WJ. Parton (1993) "CENTURY Soil Organic Matter Model
46 Environment." Agroecosystem version 4.0. Technical documentation, GPSR Tech. Report No. 4, USDA/ARS, Ft.
47 Collins, CO.
10-58 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 NASS (2004) Agricultural Chemical Usage: 2003 Field Crops Summary. Report AgChl(04)a. National Agricultural
2 Statistics Service, U.S. Department of Agriculture, Washington, D.C. Available online at
3 .
4 NASS (1999) Agricultural Chemical Usage: 1998 Field Crops Summary. Report AgCHl(99). National Agricultural
5 Statistics Service, U.S. Department of Agriculture, Washington, D.C. Available online at
6 .
7 NASS (1992) Agricultural Chemical Usage: 1991 Field Crops Summary. Report AgChl(92). National Agricultural
8 Statistics Service, U.S. Department of Agriculture, Washington, D.C. Available online at
9 .
10 NEBRA (2007) A National Biosolids Regulation, Quality, End Use & Disposal Survey. North East Biosolids and
11 Residuals Association. July 21, 2007.
12 Ogle, S.M., F.J. Breidt, M. Easter, S. Williams, K. Killian, and K. Paustian (2010) "Scale and uncertainty in
13 modeled soil organic carbon stock changes for U.S. croplands using a process-based model." Global Change
14 Biology 16:810-820.
15 Ogle, S.M., M.D. Eve, F.J. Breidt, and K. Paustian (2003) "Uncertainty in estimating land use and management
16 impacts on soil organic carbon storage for U.S. agroecosystems between 1982 and 1997." Global Change Biology
17 9:1521-1542.
18 Parton, W.J., D.S. Ojima, C.V. Cole, and D.S. Schimel (1994) "A General Model for Soil Organic Matter
19 Dynamics: Sensitivity to litter chemistry, texture and management," in Quantitative Modeling of Soil Forming
20 Processes. Special Publication 39, Soil Science Society of America, Madison, WI, 147-167.
21 Parton, W.J., D.S. Schimel, C.V. Cole, D.S. Ojima (1987) "Analysis of factors controlling soil organic matter levels
22 in Great Plains grasslands." Soil Science Society of America Journal 51:1173-1179.
23 Parton, W.J., J.W.B. Stewart, C.V. Cole. (1988) "Dynamics of C, N, P, and S in grassland soils: a model."
24 Biogeochemistry 5:109-131.
25 Parton, W.J., M.D. Hartman, D.S. Ojima, and D.S. Schimel (1998) "DAYCENT: Its Land Surface Submodel:
26 Description and Testing". Glob. Planet. Chang. 19: 35-48.
27 United States Bureau of Land Management (BLM) (2014) Rangeland Inventory, Monitoring, and Evaluation
28 Reports. Bureau of Land Management. U.S. Department of the Interior. Available online at:
29 .
30 USDA-ERS (2011) Agricultural Resource Management Survey (ARMS) Farm Financial and Crop Production
31 Practices: Tailored Reports. Online at: .
32 USDA-ERS (1997) Cropping Practices Survey Data—1995. Economic Research Service, United States Department
33 of Agriculture. Available online at .
34 USDA-NRCS (2015) Summary Report: 2012 National Resources Inventory, Natural Resources Conservation
35 Service, Washington, D.C., and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
36 .
37 USDA-NRCS (2013) Summary Report: 2010 National Resources Inventory, Natural Resources Conservation
38 Service, Washington, D.C., and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
39 .
40
Land Converted to Grassland
41 Del Grosso, S.J., S.M. Ogle, W.J. Parton. (2011). Soil organic matter cycling and greenhouse gas accounting
42 methodologies, Chapter 1, pp 3-13 DOI: 10.1021/bk-201 l-1072.ch001. In: Understanding Greenhouse Gas
43 Emissions from Agricultural Management (L. Guo, A. Gunasekara, L. McConnell. Eds.), American Chemical
44 Society, Washington, D.C.
45 Del Grosso, S.J., W.J. Parton, A.R. Mosier, M.D. Hartman, J. Brenner, D.S. Ojima, and D.S. Schimel (2001)
46 "Simulated Interaction of Carbon Dynamics and Nitrogen Trace Gas Fluxes Using the DAYCENT Model." In
References 10-59
-------
1 Modeling Carbon and Nitrogen Dynamics for Soil Management (Schaffer, M., L. Ma, S. Hansen, (eds.). CRC
2 Press, Boca Raton, Florida, pp. 303-332.
3 Fry, J., Xian, G., Jin, S., Dewitz, J., Homer, C., Yang, L., Barnes, C., Herold, N., and Wickham, J., 2011. Completion of
4 the 2006 National Land Cover Database for the Conterminous United States, PE&RS, Vol. 77(9):858-864.
5 Homer, C., Dewitz, J., Fry, J., Coan, M., Hossain, N., Larson, C., Herold, N., McKerrow, A., VanDriel, J.N., and Wickham,
6 J. 2007. Completion of the 2001 National Land Cover Database for the Conterminous United States. Photogrammetric
7 Engineering and Remote Sensing, Vol. 73, No. 4, pp 337-341.
8 Homer, C.G., Dewitz, J.A., Yang, L., Jin, S., Danielson, P., Xian, G., Coulston, J., Herold, N.D., Wickham, J.D., and
9 Megown, K., 2015, Completion of the 2011 National Land Cover Database for the conterminous United States-Representing
10 a decade of land cover change information. Photogrammetric Engineering and Remote Sensing, v. 81, no. 5, p. 345-354
11 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
12 Inventories Programme, The Intergovernmental Panel on Climate Change, H.S. Eggleston, L. Buendia, K. Miwa, T
13 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
14 Metherell, A.K., L.A. Harding, C.V. Cole, and W.J. Parton (1993) "CENTURY Soil Organic Matter Model
15 Environment." Agroecosystem version 4.0. Technical documentation, GPSR Tech. Report No. 4, USDA/ARS, Ft.
16 Collins, CO.
17 NASS (2004) Agricultural Chemical Usage: 2003 Field Crops Summary. Report AgChl(04)a. National Agricultural
18 Statistics Service, U.S. Department of Agriculture, Washington, D.C. Available online at
19 .
20 NASS (1999) Agricultural Chemical Usage: 1998 Field Crops Summary. Report AgCHl(99). National Agricultural
21 Statistics Service, U.S. Department of Agriculture, Washington, D.C. Available online at
22 .
23 NASS (1992) Agricultural Chemical Usage: 1991 Field Crops Summary. Report AgChl(92). National Agricultural
24 Statistics Service, U.S. Department of Agriculture, Washington, D.C. Available online at
25 .
26 Ogle, S.M., F.J. Breidt, M. Easter, S. Williams, K. Killian, and K. Paustian (2010) "Scale and uncertainty in
27 modeled soil organic carbon stock changes for U.S. croplands using a process-based model." Global Change
28 Biology 16:810-820.
29 Ogle, S.M., M.D. Eve, F.J. Breidt, and K. Paustian (2003) "Uncertainty in estimating land use and management
30 impacts on soil organic carbon storage for U.S. agroecosystems between 1982 and 1997." Global Change Biology
31 9:1521-1542.
32 Parton, W.J., D.S. Ojima, C.V. Cole, and D.S. Schimel (1994) "A General Model for Soil Organic Matter
33 Dynamics: Sensitivity to litter chemistry, texture and management," in Quantitative Modeling of Soil Forming
34 Processes. Special Publication 39, Soil Science Society of America, Madison, WI, 147-167.
35 Parton, W.J., D.S. Schimel, C.V. Cole, D.S. Ojima (1987) "Analysis of factors controlling soil organic matter levels
36 in Great Plains grasslands." Soil Science Society of America Journal 51:1173-1179.
37 Parton, W.J., J.W.B. Stewart, C.V. Cole. (1988) "Dynamics of C, N, P, and S in grassland soils: a model."
38 Biogeochemistry 5:109-131.
39 Parton, W.J., M.D. Hartman, D.S. Ojima, and D.S. Schimel (1998) "DAYCENT: Its Land Surface Submodel:
40 Description and Testing". Glob. Planet. Chang. 19: 35-48.
41 United States Bureau of Land Management (BLM) (2014) Rangeland Inventory, Monitoring, and Evaluation
42 Reports. Bureau of Land Management. U.S. Department of the Interior. Available online at:
43 .
44 USDA-ERS (2011) Agricultural Resource Management Survey (ARMS) Farm Financial and Crop Production
45 Practices: Tailored Reports. Online at: .
46 USDA-ERS (1997) Cropping Practices Survey Data—1995. Economic Research Service, United States Department
47 of Agriculture. Available online at .
10-60 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 USDA-NRCS (2015) Summary Report: 2012 National Resources Inventory, Natural Resources Conservation
2 Service, Washington, D.C., and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
3 .
4 USDA-NRCS (2013) Summary Report: 2010 National Resources Inventory, Natural Resources Conservation
5 Service, Washington, D.C. and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
6 .
7 Wetlands Remaining Wetlands: CO2, CH4/ and N2O Emissions
s from Peatlands Remaining Peatlands
9 Apodaca, L. (2011) Email correspondence. Lori Apodaca, Peat Commodity Specialist, USGS and Emily Rowan,
10 ICF International. November.
11 Apodaca, L. (2008) E-mail correspondence. Lori Apodaca, Peat Commodity Specialist, USGS and Emily Rowan,
12 ICF International. October and November.
13 Cleary, J., N. Roulet and T.R. Moore (2005) "Greenhouse gas emissions from Canadian peat extraction, 1990-2000:
14 A life-cycle analysis." Ambio 34:456-461.
15 Division of Geological & Geophysical Surveys (DGGS), Alaska Department of Natural Resources (1997-2014)
16 Alaska's Mineral Industry Report (1997-2013). Alaska Department of Natural Resources, Fairbanks, AK.
17 Available online at .
18 IPCC (2013) 2013 Supplement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands.
19 Hiraishi, T., Krug, T., Tanabe, K., Srivastava, N., Baasansuren, J., Fukuda, M. and Troxler, T.G. (eds.). Published:
20 IPCC, Switzerland.
21 IPCC (2007) Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II and III to the Fourth
22 Assessment Report (AR4) of the IPCC. The Intergovernmental Panel on Climate Change, R.K. Pachauri, A. Resinger
23 (eds.). Geneva, Switzerland.
24 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
25 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
26 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
27 Szumigala, D.J. (2011) Phone conversation. Dr. David Szumigala, Division of Geological and Geophysical Surveys,
28 Alaska Department of Natural Resources and Emily Rowan, ICF International. January 18, 2011.
29 Szumigala, D.J. (2008) Phone conversation. Dr. David Szumigala, Division of Geological and Geophysical Surveys,
30 Alaska Department of Natural Resources and Emily Rowan, ICF International. October 17, 2008.
31 USGS (1991-2015a) Minerals Yearbook: Peat (1994-2014). United States Geological Survey, Reston, VA.
32 Available online at < http://minerals.usgs.gov/minerals/pubs/commodity/peat/index.html#myb >.
33 United States Geological Survey (USGS) (2015b) Mineral Commodity Summaries: Peat (2014). United States
34 Geological Survey, Reston, VA. Available online at <
35 http://minerals.usgs.gov/minerals/pubs/mcs/2015/mcs2015.pdf>.
36 USGS (2015c) Mineral Commodity Summaries: Peat (2014). United States Geological Survey, Reston, VA.
37 Available online at .
38 Settlements Remaining Settlements: Changes in Carbon Stocks
39 in Urban Trees
40 deVries,R.E. (1987) A Preliminary Investigation of the Growth and Longevity of Trees in Central Park. M.S.
41 thesis, Rutgers University, New Brunswick, NJ.
References 10-61
-------
1 Dwyer, J.F., D.J. Nowak, M.H. Noble, and S.M. Sisinni (2000) Connecting People with Ecosystems in the 21st
2 Century: An Assessment of Our Nation's Urban Forests. General Technical Report PNW-GTR-490, U.S.
3 Department of Agriculture, Forest Service, Pacific Northwest Research Station, Portland, OR.
4 Fleming, L.E. (1988) Growth Estimation of Street Trees in Central New Jersey. M.S. thesis, Rutgers University,
5 New Brunswick, NJ.
6 Frelich, L.E. (1992) Predicting Dimensional Relationships for Twin Cities Shade Trees. University of Minnesota,
7 Department of Forest Resources, St. Paul, MN, p. 33.
8 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
9 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
10 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
11 Nowak, D.J. (2011) Phone conference regarding Changes in Carbon Stocks in Urban Trees estimation methodology.
12 David Nowak, USD A, Jennifer Jenkins, EPA, and Mark Flugge and Nikhil Nadkarni, ICF International. January 4,
13 2011.
14 Nowak, D.J. (2009) E-mail correspondence regarding new data for Chicago's urban forest. David Nowak, USDA
15 Forest Service to Nikhil Nadkarni, ICF International. October 7, 2009.
16 Nowak, D.J. (2007a) "New York City's Urban Forest." USDA Forest Service. Newtown Square, PA, February
17 2007.
18 Nowak, D.J. (2007b) E-mail correspondence regarding revised sequestration values and standard errors for
19 sequestration values. David Nowak, USDA Forest Service to Susan Asam, ICF International. October 31, 2007.
20 Nowak, D.J. (1994) "Atmospheric Carbon Dioxide Reduction by Chicago's Urban Forest." In: Chicago's Urban
21 Forest Ecosystem: Results of the Chicago Urban Forest Climate Project. E.G. McPherson, D.J. Nowak, and R.A.
22 Rowntree (eds.). General Technical Report NE-186. U.S. Department of Agriculture Forest Service, Radnor, PA.
23 pp. 83-94.
24 Nowak, D.J. (1986) "Silvics of an Urban Tree Species: Norway Maple (Acer platanoides L.)." M.S. thesis, College
25 of Environmental Science and Forestry, State University of New York, Syracuse, NY.
26 Nowak, D.J., Buckelew-Cumming, A., Twardus, D., Hoehn, R., and Mielke, M. (2007). National Forest Health
27 Monitoring Program, Monitoring Urban Forests in Indiana: Pilot Study 2002, Part 2: Statewide Estimates Using the
28 UFORE Model. Northeastern Area Report. NA-FR-01e07, p. 13.
29 Nowak, D.J. and D.E. Crane (2002) "Carbon Storage and Sequestration by Urban Trees in the United States."
30 Environmental Pollution 116(3):381-389.
31 Nowak, D.J., D.E. Crane, J.C. Stevens, and M. Ibarra (2002) Brooklyn's Urban Forest. General Technical Report
32 NE-290. U.S. Department of Agriculture Forest Service, Newtown Square, PA.
33 Nowak, D.J., and E. J. Greenfield (2012) Tree and impervious cover in the United States. Journal of Landscape and
34 Urban Planning (107) pp. 21-30.
35 Nowak, D.J., E.J. Greenfield, R.E. Hoehn, and E. Lapoint (2013) Carbon Storage and Sequestration by Trees in
36 Urban and Community Areas of the United States. Environmental Pollution 178: 229-236. March 12, 2013.
37 Nowak, D.J., J.T. Walton, L.G. Kaya, and J.F. Dwyer (2005) "The Increasing Influence of Urban Environments on
38 U.S. Forest Management." Journal of Forestry 103(8):377-382.
39 Smith, W.B. and S.R. Shifley (1984) Diameter Growth, Survival, and Volume Estimates for Trees in Indiana and
40 Illinois. Research Paper NC-257. North Central Forest Experiment Station, U.S. Department of Agriculture Forest
41 Service, St. Paul, MN.
42 U.S. Census Bureau (2012) "A national 2010 urban area file containing a list of all urbanized areas and urban
43 clusters (including Puerto Rico and the Island Areas) sorted by UACE code." U.S. Census Bureau, Geography
44 Division.
10-62 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1
Settlements Remaining Settlements: N2O Fluxes from Soils
2 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
3 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
4 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
5 Ruddy B.C., D.L. Lorenz, and D.K. Mueller (2006) County-level estimates of nutrient inputs to the land surface of
6 the conterminous United States, 1982-2001. Scientific Investigations Report 2006-5012. U.S. Department of the
7 Interior.
8 Other: Changes in Yard Trimming and Food Scrap Carbon
9 Stocks in Landfills
10 Barlaz, M.A. (2008) "Re: Corrections to Previously Published Carbon Storage Factors." Memorandum to Randall
11 Freed, ICF International. February 28, 2008.
12 Barlaz, M.A. (2005) "Decomposition of Leaves in Simulated Landfill." Letter report to Randall Freed, ICF
13 Consulting. June 29, 2005.
14 Barlaz, M.A. (1998) "Carbon Storage during Biodegradation of Municipal Solid Waste Components in Laboratory -
15 Scale Landfills." Global Biogeochemical Cycles 12:373-380.
16 De la Cruz, F.B. and M.A. Barlaz (2010) "Estimation of Waste Component Specific Landfill Decay Rates Using
17 Laboratory-Scale Decomposition Data" Environmental Science & Technology 44:4722- 4728.
18 Eleazer, W.E., W.S. Odle, Y. Wang, and M.A. Barlaz (1997) "Biodegradability of Municipal Solid Waste
19 Components in Laboratory-Scale Landfills." Environmental Science & Technology 31:911-917.
20 EPA (2015a) Advancing Sustainable Materials Management: Facts andFigures 2013. U.S. Environmental
21 Protection Agency, Office of Solid Waste and Emergency Response, Washington, D.C. Available online at
22 .
23 EPA (2015b) Advancing Sustainable Materials Management: 2013 Historical (summary) Data Tables. U.S.
24 Environmental Protection Agency, Office of Solid Waste and Emergency Response, Washington, D.C. Available
25 online at .
26 EPA (1995) Compilation of Air Pollutant Emission Factors. U.S. Environmental Protection Agency, Office of Air
27 Quality Planning and Standards, Research Triangle Park, NC. AP-42 Fifth Edition. Available online at <
28 http://www3.epa.gov/ttnchiel/ap42/>.
29 IPCC (2003) Good Practice Guidance for Land Use, Land-Use Change, and Forestry. The Intergovernmental Panel
30 on Climate Change, National Greenhouse Gas Inventories Programme, J. Penman et al. (eds.). Available online at
31 .
32 Oshins, C. and D. Block (2000) "Feedstock Composition at Composting Sites." Biocycle 41(9):31-34.
33 Tchobanoglous, G., H. Theisen, and S.A. Vigil (1993) Integrated Solid Waste Management, 1st edition. McGraw-
34 Hill, NY. Cited by Barlaz (1998) "Carbon Storage during Biodegradation of Municipal Solid Waste Components in
35 Laboratory-Scale Landfills." Global Biogeochemical Cycles 12:373-380.
36
37
Waste
Landfills
38 40 CFR Part 60, Subpart CC (2005) Emission Guidelines and Compliance Times for Municipal Solid Waste
39 Landfills, 60.30c~60.36c, Code of Federal Regulations, Title 40. Available online at
40 .
References 10-63
-------
1 40 CFR Part 60, Subpart WWW (2005) Standards of Performance for Municipal Solid Waste Landfills, 60.750-
2 60.759, Code of Federal Regulations, Title 40. Available online at
3 .
4 BioCycle (2010) "The State of Garbage in America" By L. Arsova, R. Van Haaren, N. Goldstein, S. Kaufman, and
5 N. Themelis. BioCycle. December 2010. Available online at .
6 Bronstein, K., Coburn, J., and R. Schmeltz (2012) "Understanding the EPA's Inventory of U.S. Greenhouse Gas
7 Emissions and Sinks and Mandatory GHG Reporting Program for Landfills: Methodologies, Uncertainties,
8 Improvements and Deferrals." Prepared for the U.S. EPA International Emissions Inventory Conference, August
9 2012, Tampa, Florida. Available online at .
10 Czepiel, P., B. Mosher, P. Crill, and R. Harriss (1996) "Quantifying the Effect of Oxidation on Landfill Methane
11 Emissions." Journal of Geophysical Research, 101(D11):16721-16730.
12 EIA (2007) Voluntary Greenhouse Gas Reports for El A Form 1605B (Reporting Year 2006). Available online at
13 .
14 EPA (2015a) Landfill Gas-to-Energy Project Database. Landfill Methane and Outreach Program. August 2015.
15 EPA (2015b) Greenhouse Gas Reporting Program (GHGRP). 2015 Envirofacts. Subpart HH: Municipal Solid Waste
16 Landfills. Available online at .
17 EPA (2015c) Rule and Implementation Information for Standards of Performance for Municipal Solid Waste
18 Landfills, Docket #A-88-09. Available at < http://www3.epa.gov/ttn/atw/landfill/landflpg.html>.
19 EPA (2015d) Advancing Sustainable Materials Management: Facts and Figures 2013. June 2015. Available online
20 at.
21 EPA (2008) Compilation of Air Pollution Emission Factors, Publication AP-42, Draft Section 2.4 Municipal Solid
22 Waste Landfills. October 2008.
23 EPA (1998) Compilation of Air Pollution Emission Factors, Publication AP-42, Section 2.4 Municipal Solid Waste
24 Landfills. November 1998.
25 EPA (1993) Anthropogenic Methane Emissions in the United States, Estimates for 1990: Report to Congress, U.S.
26 Environmental Protection Agency, Office of Air and Radiation. Washington, D.C. EPA/430-R-93-003. April 1993.
27 EPA (1988) National Survey of Solid Waste (Municipal) Landfill Facilities, U.S. Environmental Protection Agency.
28 Washington, D.C. EPA/530-SW-88-011. September 1988.
29 ERG (2014) Draft Production Data Supplied by ERG for 1990-2013 for Pulp and Paper, Fruits and Vegetables, and
30 Meat. August.
31 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
32 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
33 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
34 Mancinelli, R. and C. McKay (1985) "Methane-Oxidizing Bacteria in Sanitary Landfills." Proc. First Symposium on
35 Biotechnical Advances in Processing Municipal Wastes for Fuels and Chemicals, Minneapolis, MN, 437-450.
36 August.
37 Peer, R., S. Thorneloe, and D. Epperson (1993) "A Comparison of Methods for Estimating Global Methane
38 Emissions from Landfills." Chemosphere, 26(l-4):387-400.
39 RTI (2015a) Update the flare correction factor for the 1990-2014 Inventory. Memorandum prepared by K. Bronstein
40 and M. McGrath for R. Schmeltz (EPA), November 25, 2015.
41 RTI (2015b) Investigate the potential to update DOC and k values for the Pulp and Paper industry in the US Solid
42 Waste Inventory. Memorandum prepared by K. Bronstein and M. McGrath for R. Schmeltz (EPA), December 4,
43 2015.RTI (2011) Updated Research on Methane Oxidation in Landfills. Memorandum prepared by K. Weitz (RTI)
44 for R. Schmeltz (EPA), January 14, 2011.
45 RTI (2011) Updated Research on Methane Oxidation in Landfills. Memorandum prepared by K. Weitz (RTI) for R.
46 Schmeltz (EPA), January 14, 2011.
10-64 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 RTI (2004) Documentation for Changes to the Methodology for the Inventory of Methane Emissions from Landfills.
2 Memorandum prepared by M. Branscome and J. Coburn (RTI) to E. Scheehle (EPA), August 26, 2004.
3 Shin, D. (2014). Generation and Disposition of Municipal Solid Waste (MSW) in the United States - A National
4 Survey. Master of Science thesis submitted to the Department of Earth and Environmental Engineering Fu
5 Foundation School of Engineering and Applied Science, Columbia University. January 3, 2014. Available online at
6 .
7 Solid Waste Association of North America (SWANA) (1998) Comparison of Models for Predicting Landfill
8 Methane Recovery. Publication No. GR-LG 0075. March 1998.
9 U.S. Census Bureau (2015). Annual Population Estimates, Vintage 2014 April 1, 2010 to July 1, 2014. Available
10 online at < http://www.census.gov/popest/data/national/asrh/20147 >.
11 Waste Business Journal (WBJ) (2010). Directory of Waste Processing & Disposal Sites 2010.
12
Wastewater Treatment (TO BE UPDATED)
13 Ahn et al. (2010) N2O Emissions from Activated Sludge Processes, 2008-2009: Results of a National Monitoring
14 Survey in the United States. Environ. Sci. Technol. 44: 4505-4511.
15 Beecher et al. (2007) "A National Biosolids Regulation, Quality, End Use & Disposal Survey, Preliminary Report."
16 Northeast Biosolids and Residuals Association, April 14, 2007.
17 Benyahia, F., M. Abdulkarim, A. Embaby, and M. Rao. (2006) Refinery Wastewater Treatment: A true
18 Technological Challenge. Presented at the Seventh Annual U.A.E. University Research Conference.
19 Climate Action Reserve (CAR) (2011) Landfill Project Protocol V4.0, June 2011. Available online at
20 .
21 Chandran, K. (2012) Greenhouse Nitrogen Emissions from Wastewater Treatment Operation Phase I: Molecular
22 Level Through Whole Reactor Level Characterization. WERF Report U4R07.
23 Donovan (1996) Siting an Ethanol Plant in the Northeast. C.T. Donovan Associates, Inc. Report presented to
24 Northeast Regional Biomass Program (NRBP). (April). Available online at .
25 Accessed October 2006.
26 EIA (2014) Energy Information Administration. U.S. Refinery and Blender Net Production of Crude Oil and
27 Petroleum Products (Thousand Barrels). Available online at: .
28 Accessed June 2014.
29 EPA (2008) US Environmental Protection Agency. Municipal Nutrient Removal Technologies Reference
30 Document: Volume 2 - Appendices. U.S. Environmental Protection Agency, Office of Wastewater Management.
31 Washington, D.C.
32 EPA (2004) U.S. Environmental Protection Agency. Clean Watersheds Needs Survey 2004 - Report to Congress.
33 U.S. Environmental Protection Agency, Office of Wastewater Management. Washington, D.C.
34 EPA (2002) U.S. Environmental Protection Agency. Development Document for the Proposed Effluent Limitations
35 Guidelines and Standards for the Meat and Poultry Products Industry Point Source Category (40 CFR 432). EPA-
36 821-B-01-007. Office of Water, U.S. Environmental Protection Agency. Washington, D.C. January 2002.
37 EPA (2000) U.S. Environmental Protection Agency. Clean Watersheds Needs Survey 2000 - Report to Congress.
38 Office of Wastewater Management, U.S. Environmental Protection Agency. Washington, D.C. Available online at
39 . Accessed July 2007.
40 EPA (1999) U.S. Environmental Protection Agency. Biosolids Generation, Use and Disposal in the United States.
41 Office of Solid Waste and Emergency Response, U.S. Environmental Protection Agency. Washington, D.C.
42 EPA530-R-99-009. September 1999.
43 EPA (1998) U.S. Environmental Protection Agency. "AP-42 Compilation of Air Pollutant Emission Factors."
44 Chapter 2.4, Table 2.4-3, page 2.4-13. Available online at
45 .
References 10-65
-------
1 EPA (1997a) U.S. Environmental Protection Agency. Estimates of Global Greenhouse Gas Emissions from
2 Industrial and Domestic Wastewater Treatment. EPA-600/R-97-091. Office of Policy, Planning, and Evaluation,
3 U.S. Environmental Protection Agency. Washington, D.C. September 1997.
4 EPA (1997b) U.S. Environmental Protection Agency. Supplemental Technical Development Document for Effluent
5 Guidelines and Standards (Subparts B & E). EPA-821-R-97-011. Office of Water, U.S. Environmental Protection
6 Agency. Washington, D.C. October 1997.
7 EPA (1996) U.S. Environmental Protection Agency. 1996 Clean Water Needs Survey Report to Congress.
8 Assessment of Needs for Publicly Owned Wastewater Treatment Facilities, Correction of Combined Sewer
9 Overflows, and Management of Storm Water and Nonpoint Source Pollution in the United States. Office of
10 Wastewater Management, U.S. Environmental Protection Agency. Washington, D.C. Available online at
11 . Accessed July 2007.
12 EPA (1993) U.S. Environmental Protection Agency. Development Document for the Proposed Effluent Limitations
13 Guidelines and Standards for the Pulp, Paper and Paperboard Point Source Category. EPA-821-R-93-019. Office of
14 Water, U.S. Environmental Protection Agency. Washington, D.C. October 1993.
15 EPA (1992) U.S. Environmental Protection Agency. Clean Watersheds Needs Survey 1992 - Report to Congress.
16 Office of Wastewater Management, U.S. Environmental Protection Agency. Washington, D.C.
17 EPA (1975) U.S. Environmental Protection Agency. Development Document for Interim Final and Proposed
18 Effluent Limitations Guidelines and New Source Performance Standards for the Fruits, Vegetables, and Specialties
19 Segment of the Canned and Preserved Fruits and Vegetables Point Source Category. United States Environmental
20 Protection Agency, Office of Water. EPA-440/1-75-046. Washington D.C. October 1975.
21 EPA (1974) U.S. Environmental Protection Agency. Development Document for Effluent Limitations Guidelines
22 and New Source Performance Standards for the Apple, Citrus, and Potato Processing Segment of the Canned and
23 Preserved Fruits and Vegetables Point Source Category. Office of Water, U.S. Environmental Protection Agency,
24 Washington, D.C. EPA-440/l-74-027-a. March 1974.
25 ERG (2014a) Recommended Improvements to the 1990-2013 Wastewater Greenhouse Gas Inventory. September
26 2014.
27 ERG (2014b) Recommended Improvements to the 1990-2013 Wastewater Greenhouse Gas Inventory Using the
28 GHGRP Data. September 2014.
29 ERG (2013a) Revisions to Pulp and Paper Wastewater Inventory. October 2013.
30 ERG (2013b) Revisions to the Petroleum Wastewater Inventory. October 2013.
31 ERG (2011) Review of Current Research on Nitrous Oxide Emissions from Wastewater Treatment. April 2011.
32 ERG (2008) Planned Revisions of the Industrial Wastewater Inventory Emission Estimates for the 1990-2007
33 Inventory. August 10, 2008.
34 ERG (2006) Memorandum: Assessment of Greenhouse Gas Emissions from Wastewater Treatment of U.S. Ethanol
35 Production Wastewaters. Prepared for Melissa Weitz, EPA. 10 October 2006.
36 FAO (2014) FAOSTAT-Forestry Database. Available online at
37 Accessed July 2013.
38 Global Water Research Coalition (GWRC) (2011) N2O and CH4 Emission from Wastewater Collection and
3 9 Treatment Systems - technical Report. GWRC Report 2011-30.
40 Great Lakes-Upper Mississippi River Board of State and Provincial Public Health and Environmental Managers.
41 (2004) Recommended Standards for Wastewater Facilities (Ten-State Standards).
42 IPCC (2014) 2013 Supplement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands.
43 Hiraishi, T., Krug, T., Tanabe, K., Srivastava, N., Baasansuren, J., Fukuda, M. and Troxler, T.G. (eds.). Published:
44 IPCC, Switzerland.
45 IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
46 Assessment Report of the Intergovernmental Panel on Climate Change. S. Solomon, D. Qin, M. Manning, Z. Chen,
10-66 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
1 M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.). Cambridge University Press. Cambridge, United
2 Kingdom 996 pp.
3 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
4 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
5 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
6 Leverenz, H.L., G. Tchobanoglous, and J.L. Darby (2010) "Evaluation of Greenhouse Gas Emissions from Septic
7 Systems". Water Environment Research Foundation. Alexandria, VA.
8 Lockwood-Post (2002) Lockwood-Post's Directory of Pulp, Paper and Allied Trades, Miller-Freeman Publications.
9 San Francisco, CA.
10 McFarland (2001) Biosolids Engineering, New York: McGraw-Hill, p. 2.12.
11 Merrick (1998) Wastewater Treatment Options for the Biomass-to-Ethanol Process. Report presented to National
12 Renewable Energy Laboratory (NREL). Merrick & Company. Subcontract No. AXE-8-18020-01. October 22,
13 1998.
14 Metcarf & Eddy, Inc. (2003) Wastewater Engineering: Treatment, Disposal and Reuse, 4th ed. McGrawHill
15 Publishing.
16 Nemerow, N.L. and A. Dasgupta (1991) Industrial and Hazardous Waste Treatment. Van Nostrand Reinhold. NY.
17 ISBN 0-442-31934-7.
18 NRBP (2001) Northeast Regional Biomass Program. An Ethanol Production Guidebook for Northeast States.
19 Washington, D.C. (May 3). Available online at . Accessed October 2006.
20 Rendleman, C.M. and Shapouri, H. (2007) New Technologies in Ethanol Production. USDA Agricultural Economic
21 Report Number 842.
22 RFA (2014) Renewable Fuels Association. Historic U.S. Fuel Ethanol Production. Available online at
23 . Accessed October 2012.
24 Ruocco (2006a) Email correspondence. Dr. Joe Ruocco, Phoenix Bio-Systems to Sarah Holman, ERG. "Capacity of
25 Bio-Methanators (Dry Milling)." October 6, 2006.
26 Ruocco (2006b) Email correspondence. Dr. Joe Ruocco, Phoenix Bio-Systems to Sarah Holman, ERG. "Capacity
27 of Bio-Methanators (Wet Milling)." October 16, 2006.
28 Scheehle, E.A., and Doom, M.R. (2001) "Improvements to the U.S. Wastewater Methane and Nitrous Oxide
29 Emissions Estimate." July 2001.
30 Sullivan (SCS Engineers) (2010) The Importance of Landfill Gas Capture and Utilization in the U.S. Presented to
31 SWICS, April 6, 2010. Available online at
32 .
33 Sullivan (SCS Engineers) (2007) Current MSW Industry Position and State of the Practice on Methane Destruction
34 Efficiency in Flares, Turbines, and Engines. Presented to Solid Waste Industry for Climate Solutions (SWICS). July
35 2007. Available online at
36 .
37 UNFCCC (2012) COM Methodological tool, Project emissions from flaring (Version 02.0.0). EB 68 Report. Annex
38 15. Available online at .
40 U.S. Census Bureau (2014) International Database. Available online at and
41 . Accessed August 2014.
42 U.S. Census Bureau (2011) "American Housing Survey." Table 1A-4: Selected Equipment and Plumbing-All
43 Housing Units. From 1989, 1991, 1993, 1995, 1997, 1999, 2001, 2003, 2005, 2007, 2009 and 2011 reports.
44 Available online at . Accessed October 2012.
45 U.S. DOE (2013) U.S. Department of Energy Bioenergy Technologies Office. Biofuels Basics. Available online at
46 . Accessed September 2013.
References 10-67
-------
1 USD A (2014a) U.S. Department of Agriculture. National Agricultural Statistics Service. Washington, D.C.
2 Available online at and
3 . Accessed June 2014.
4 USD A (2014b) U.S. Department of Agriculture. Economic Research Service. Nutrient Availability. Washington
5 D.C. Available online at
6 .
7 Accessed August 2014.
8 USPoultry (2006) Email correspondence. John Starkey, USPOULTRY to D. Bartram, ERG. 30 August 2006.
9 White and Johnson (2003) White, P.J. and Johnson, L.A. Editors. Corn: Chemistry and Technology. 2nd ed.
10 AACC Monograph Series. American Association of Cereal Chemists. St. Paul, MN.
11 Willis et al. (2013) Methane Evolution from Lagoons and Ponds. Prepared for the Water Environment Research
12 Foundation under contract U2R08c.
13 World Bank (1999) Pollution Prevention and Abatement Handbook 1998, Toward Cleaner Production. The
14 International Bank for Reconstruction and Development, The World Bank, Washington, D.C. ISBN 0-8213-3638-
15 X.
16 Composting
17 BioCycle (2010) The State of Garbage in America. Prepared by Rob van Haaren, Nickolas Themelis and Nora
18 Goldstein. Available online at .
19 EPA (2015) Advancing Sustainable Materials Management: Facts and Figures 2013. Office of Solid Waste and
20 Emergency Response, U.S. Environmental Protection Agency, Washington, D.C. Available online at
21 .
22 EPA (2014) Municipal Solid Waste in the United States: 2012 Facts and Figures. Office of Solid Waste and
23 Emergency Response, U.S. Environmental Protection Agency, Washington, D.C. Available online at
24 .
25 EPA (2011) Municipal Solid Waste in the United States: 2010 Facts and Figures. Office of Solid Waste and
26 Emergency Response, U.S. Environmental Protection Agency, Washington, D.C. Available online at
27 .
28 EPA (2007) Municipal Solid Waste in the United States: 2006 Facts and Figures. Office of Solid Waste and
29 Emergency Response, U.S. Environmental Protection Agency, Washington, D.C. Available online at
30 .
31 IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. Volume 5: Waste, Chapter 4:
32 Biological Treatment of Solid Waste, Table 4.1. The National Greenhouse Gas Inventories Programme, The
33 Intergovernmental Panel on Climate Change, H.S. Eggleston, L. Buendia, K. Miwa, T. Ngara, and K. Tanabe (eds.).
34 Hayama, Kanagawa, Japan. Available online at .
36 Shin, D. (2014). Generation and Disposition of Municipal Solid Waste (MSW) in the United States - A National
37 Survey. Table 3. Master of Science thesis, Department of Earth and Environmental Engineering, Fu Foundation
38 School of Engineering and Applied Science, Columbia University. Available online at
39 .
40 U.S. Census Bureau (2015). Population Estimates: Vintage 2014 Annual Estimates Available online at
41 .
42 U.S. Composting Council (2010). Yard Trimmings Bans: Impact and Support. Prepared by Stuart Buckner,
43 Executive Director, U.S, Composting Council. Available online at
44 .
10-68 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2014
-------
i
Waste Sources of Indirect Greenhouse Gas Emissions
2 EPA (2015) "1970 - 2014 Average annual emissions, all criteria pollutants in MS Excel." National Emissions
3 Inventory (NEI) Air Pollutant Emissions Trends Data. Office of Air Quality Planning and Standards, March 20 1 5 .
4 Available online at .
5 EPA (2003) E-mail correspondence containing preliminary ambient air pollutant data. Office of Air Pollution and
6 the Office of Air Quality Planning and Standards, U.S. Environmental Protection Agency. December 22, 2003.
7 EPA (1997) Compilation of Air Pollutant Emission Factors, AP-42. Office of Air Quality Planning and Standards,
8 U.S. Environmental Protection Agency. Research Triangle Park, NC. October 1997.
9 Recalculations and Improvements
10 Domke, G.M., Perry, C.H., Walters, B.F., Nave, L.E., Woodall, C. W., Swanston, C. W. In Prep. Towards field-
1 1 based estimates of soil organic carbon in forests of the United States.
12 EPA (2015) Rule and Implementation Information for Standards of Performance for Municipal Solid Waste
13 Landfills, Docket #A-88-09. Available at < http://www3.epa.gov/ttn/atw/landfill/landflpg.html>.
14 IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
15 Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
16 Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
17 Ogle, S.M., Woodall, C.W., Swan, A., Smith, J.E., Wirth. T. In preparation. Determining the Managed Land Base
18 for Delineating Carbon Sources and Sinks in the United States. Environmental Science and Policy.
19 RTI (2015) Investigate the potential to update DOC and k values for the Pulp and Paper industry in the US Solid
20 Waste Inventory. Memorandum prepared by K. Bronstein and M. McGrath for R. Schmeltz (EPA), December 4,
21 2015.RTI (201 1) Updated Research on Methane Oxidation in Landfills. Memorandum prepared by K. Weitz (RTI)
22 for R. Schmeltz (EPA), January 14, 20 1 1 .
23 Woodall, C.W., Coulston, J.W., Domke, G.M., Walters, B.F., Wear, D.N., Smith, J.E., Anderson, H.-E., Clough,
24 B.J., Cohen, W.B., Griffith, D.M., Hagan, S.C., Hanou, I.S.; Nichols, M.C., Perry, C.H., Russell, M.B., Westfall,
25 J.A., Wilson, B.T. (2015) The US Forest Carbon Accounting Framework: Stocks and Stock change 1990-2016.
26 Gen. Tech. Rep. NRS-154. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern
27 Research Station. 49 pp.
28
29
30
References 10-69
------- |